From 2cbfdc70baf3fa40f09a218f8eb94e0b5a166efa Mon Sep 17 00:00:00 2001 From: Michael Wyatt Date: Fri, 3 Nov 2023 13:32:29 -0700 Subject: [PATCH 01/13] Update version.txt after 0.11.2 release (#4609) Co-authored-by: jeffra --- version.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/version.txt b/version.txt index bc859cbd6d99..1a96df19c09a 100644 --- a/version.txt +++ b/version.txt @@ -1 +1 @@ -0.11.2 +0.11.3 From 737ef296cd745d72cd613473a56869c9f4a0ed41 Mon Sep 17 00:00:00 2001 From: Logan Adams <114770087+loadams@users.noreply.github.com> Date: Fri, 3 Nov 2023 15:06:35 -0700 Subject: [PATCH 02/13] Pin transformers in nv-inference (#4606) nv-inference is failing due to changes on the transformers side. Sample passing build [here](https://github.com/microsoft/DeepSpeed/actions/runs/6619077509/job/17978847686#step:5:19) and sample failing build [here](https://github.com/microsoft/DeepSpeed/actions/runs/6619077509/job/18349776000#step:5:19) --- .github/workflows/nv-inference.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/nv-inference.yml b/.github/workflows/nv-inference.yml index 065f8b93f1e0..b5cf46f79011 100644 --- a/.github/workflows/nv-inference.yml +++ b/.github/workflows/nv-inference.yml @@ -34,6 +34,7 @@ jobs: run: | git clone https://github.com/huggingface/transformers cd transformers + git checkout f370bebdc git rev-parse --short HEAD pip install . From 38b41dffa1d593ad8a8ab3ee2795dc203800f08c Mon Sep 17 00:00:00 2001 From: Connor Holmes Date: Fri, 3 Nov 2023 15:07:35 -0700 Subject: [PATCH 03/13] DeepSpeed-FastGen (#4604) Co-authored-by: Jeff Rasley Co-authored-by: Michael Wyatt Co-authored-by: Ammar Ahmad Awan Co-authored-by: Masahiro Tanaka Co-authored-by: Logan Adams <114770087+loadams@users.noreply.github.com> --- .github/workflows/nv-a6000.yml | 56 ++ .github/workflows/nv-pre-compile-ops.yml | 2 +- .gitmodules | 0 .pre-commit-config.yaml | 1 + MANIFEST.in | 2 + accelerator/cuda_accelerator.py | 19 +- csrc/includes/activation_type.h | 17 + csrc/includes/ds_kernel_utils.h | 5 + csrc/includes/reduction_utils.h | 228 ++++++- csrc/transformer/inference/csrc/gelu.cu | 4 +- deepspeed/inference/__init__.py | 3 + deepspeed/inference/v2/__init__.py | 7 + deepspeed/inference/v2/allocator.py | 32 + deepspeed/inference/v2/checkpoint/__init__.py | 8 + .../inference/v2/checkpoint/base_engine.py | 41 ++ .../v2/checkpoint/huggingface_engine.py | 99 +++ .../v2/checkpoint/in_memory_engine.py | 40 ++ deepspeed/inference/v2/config_v2.py | 31 + deepspeed/inference/v2/engine_factory.py | 46 ++ deepspeed/inference/v2/engine_v2.py | 217 ++++++ deepspeed/inference/v2/inference_utils.py | 105 +++ deepspeed/inference/v2/kernels/__init__.py | 6 + .../inference/v2/kernels/core_ops/__init__.py | 10 + .../core_ops/bias_activations/__init__.py | 6 + .../bias_activations/bias_activation.cpp | 68 ++ .../bias_activations/bias_activation.cu | 140 ++++ .../bias_activations/bias_activation.h | 22 + .../bias_activations/bias_activation.py | 62 ++ .../kernels/core_ops/blas_kernels/__init__.py | 6 + .../v2/kernels/core_ops/blas_kernels/blas.h | 138 ++++ .../core_ops/blas_kernels/blas_linear.py | 55 ++ .../core_ops/blas_kernels/blas_utils.h | 275 ++++++++ .../v2/kernels/core_ops/core_ops.cpp | 36 + .../core_ops/cuda_layer_norm/__init__.py | 8 + .../cuda_layer_norm/cuda_fp_ln_base.py | 37 ++ .../core_ops/cuda_layer_norm/cuda_ln.py | 30 + .../core_ops/cuda_layer_norm/cuda_post_ln.py | 34 + .../core_ops/cuda_layer_norm/cuda_pre_ln.py | 39 ++ .../core_ops/cuda_layer_norm/layer_norm.cpp | 102 +++ .../core_ops/cuda_layer_norm/layer_norm.cu | 490 ++++++++++++++ .../core_ops/cuda_layer_norm/layer_norm.h | 67 ++ .../core_ops/cuda_rms_norm/__init__.py | 7 + .../core_ops/cuda_rms_norm/rms_norm.cpp | 123 ++++ .../core_ops/cuda_rms_norm/rms_norm.cu | 262 ++++++++ .../kernels/core_ops/cuda_rms_norm/rms_norm.h | 33 + .../core_ops/cuda_rms_norm/rms_norm.py | 28 + .../core_ops/cuda_rms_norm/rms_norm_base.py | 37 ++ .../core_ops/cuda_rms_norm/rms_pre_norm.py | 39 ++ .../core_ops/gated_activations/__init__.py | 6 + .../gated_activations/gated_activation.py | 65 ++ .../gated_activation_kernels.cpp | 72 ++ .../gated_activation_kernels.cu | 169 +++++ .../gated_activation_kernels.h | 25 + .../inference/v2/kernels/cutlass_ops/LICENSE | 202 ++++++ .../v2/kernels/cutlass_ops/__init__.py | 7 + .../v2/kernels/cutlass_ops/cutlass_ops.cpp | 19 + .../cutlass_ops/mixed_gemm/__init__.py | 6 + .../cutlass_ops/mixed_gemm/mixed_gemm.cu | 93 +++ .../cutlass_ops/mixed_gemm/mixed_gemm.h | 16 + .../cutlass_ops/mixed_gemm/mixed_gemm.py | 64 ++ .../cutlass_ops/mixed_gemm/mixed_gemm_api.h | 57 ++ .../kernels/cutlass_ops/moe_gemm/__init__.py | 7 + .../cutlass_ops/moe_gemm/mixed_moe_gemm.py | 67 ++ .../kernels/cutlass_ops/moe_gemm/moe_gemm.cu | 175 +++++ .../kernels/cutlass_ops/moe_gemm/moe_gemm.h | 24 + .../kernels/cutlass_ops/moe_gemm/moe_gemm.py | 60 ++ .../cutlass_ops/moe_gemm/moe_gemm_api.h | 64 ++ .../shared_resources/weight_variant.h | 11 + deepspeed/inference/v2/kernels/ds_kernel.py | 32 + .../v2/kernels/ragged_ops/__init__.py | 13 + .../ragged_ops/atom_builder/__init__.py | 6 + .../ragged_ops/atom_builder/atom_builder.cpp | 53 ++ .../ragged_ops/atom_builder/atom_builder.h | 21 + .../ragged_ops/atom_builder/atom_builder.py | 50 ++ .../ragged_ops/blocked_flash/__init__.py | 6 + .../ragged_ops/blocked_flash/attention_atom.h | 39 ++ .../blocked_flash/blocked_flash.cpp | 101 +++ .../ragged_ops/blocked_flash/blocked_flash.h | 16 + .../ragged_ops/blocked_flash/blocked_flash.py | 107 +++ .../kernels/ragged_ops/blocked_flash/flash.h | 74 +++ .../v2/kernels/ragged_ops/embed/__init__.py | 6 + .../v2/kernels/ragged_ops/embed/embed.cpp | 101 +++ .../v2/kernels/ragged_ops/embed/embed.cu | 137 ++++ .../v2/kernels/ragged_ops/embed/embed.cuh | 26 + .../v2/kernels/ragged_ops/embed/embed.h | 23 + .../v2/kernels/ragged_ops/embed/embed.py | 67 ++ .../linear_blocked_kv_rotary/__init__.py | 8 + .../blocked_kv_rotary.cpp | 188 ++++++ .../blocked_kv_rotary.cu | 314 +++++++++ .../blocked_kv_rotary.cuh | 45 ++ .../blocked_kv_rotary.h | 63 ++ .../blocked_kv_rotary.py | 70 ++ .../blocked_trained_kv_rotary.py | 76 +++ .../linear_blocked_kv_copy.py | 74 +++ .../ragged_ops/logits_gather/__init__.py | 6 + .../logits_gather/logits_gather.cpp | 45 ++ .../ragged_ops/logits_gather/logits_gather.cu | 86 +++ .../logits_gather/logits_gather.cuh | 22 + .../ragged_ops/logits_gather/logits_gather.h | 20 + .../ragged_ops/logits_gather/logits_gather.py | 52 ++ .../kernels/ragged_ops/moe_gather/__init__.py | 6 + .../ragged_ops/moe_gather/moe_gather.cpp | 53 ++ .../ragged_ops/moe_gather/moe_gather.cu | 122 ++++ .../ragged_ops/moe_gather/moe_gather.cuh | 20 + .../ragged_ops/moe_gather/moe_gather.h | 19 + .../ragged_ops/moe_gather/moe_gather.py | 52 ++ .../ragged_ops/moe_scatter/__init__.py | 6 + .../ragged_ops/moe_scatter/moe_scatter.cpp | 62 ++ .../ragged_ops/moe_scatter/moe_scatter.cu | 208 ++++++ .../ragged_ops/moe_scatter/moe_scatter.cuh | 22 + .../ragged_ops/moe_scatter/moe_scatter.h | 23 + .../ragged_ops/moe_scatter/moe_scatter.py | 55 ++ .../ragged_ops/ragged_helpers/ragged_dtypes.h | 48 ++ .../ragged_helpers/ragged_kernel_helpers.cpp | 28 + .../ragged_helpers/ragged_kernel_helpers.h | 16 + .../v2/kernels/ragged_ops/ragged_ops.cpp | 48 ++ .../ragged_ops/top_1_gating/__init__.py | 6 + .../ragged_ops/top_1_gating/top_1_gating.cpp | 55 ++ .../ragged_ops/top_1_gating/top_1_gating.cu | 106 +++ .../ragged_ops/top_1_gating/top_1_gating.cuh | 24 + .../ragged_ops/top_1_gating/top_1_gating.h | 21 + .../ragged_ops/top_1_gating/top_1_gating.py | 59 ++ deepspeed/inference/v2/logging.py | 26 + .../v2/model_implementations/AddingAModel.md | 84 +++ .../v2/model_implementations/__init__.py | 9 + .../common_parameters/__init__.py | 13 + .../attn_output_parameters.py | 29 + .../common_parameters/embedding_parameters.py | 29 + .../common_parameters/invfreq_parameters.py | 21 + .../common_parameters/mlp_parameters.py | 81 +++ .../common_parameters/moe_parameters.py | 71 ++ .../common_parameters/norm_parameters.py | 22 + .../common_parameters/qkv_parameters.py | 116 ++++ .../common_parameters/unembed_parameters.py | 26 + .../inference_model_base.py | 196 ++++++ .../inference_policy_base.py | 155 +++++ .../inference_transformer_base.py | 616 ++++++++++++++++++ .../layer_container_base.py | 289 ++++++++ .../llama_v2/__init__.py | 4 + .../llama_v2/llama_v2_containers.py | 80 +++ .../llama_v2/llama_v2_model.py | 204 ++++++ .../llama_v2/llama_v2_policy.py | 37 ++ .../model_implementations/mistral/__init__.py | 4 + .../mistral/container.py | 77 +++ .../v2/model_implementations/mistral/model.py | 202 ++++++ .../model_implementations/mistral/policy.py | 36 + .../v2/model_implementations/opt/__init__.py | 4 + .../v2/model_implementations/opt/container.py | 95 +++ .../v2/model_implementations/opt/model.py | 195 ++++++ .../v2/model_implementations/opt/policy.py | 36 + .../model_implementations/parameter_base.py | 257 ++++++++ .../sharding/__init__.py | 12 + .../v2/model_implementations/sharding/attn.py | 63 ++ .../sharding/attn_out.py | 111 ++++ .../sharding/embedding.py | 34 + .../v2/model_implementations/sharding/mlp.py | 75 +++ .../v2/model_implementations/sharding/qkv.py | 166 +++++ .../model_implementations/sharding/types.py | 18 + .../model_implementations/sharding/unembed.py | 41 ++ .../model_implementations/sharding/utils.py | 104 +++ deepspeed/inference/v2/modules/__init__.py | 8 + .../inference/v2/modules/configs/__init__.py | 11 + .../v2/modules/configs/attention_configs.py | 82 +++ .../v2/modules/configs/embedding_config.py | 70 ++ .../v2/modules/configs/linear_config.py | 43 ++ .../v2/modules/configs/moe_config.py | 50 ++ .../v2/modules/configs/norm_config.py | 32 + .../v2/modules/configs/unembed_config.py | 39 ++ deepspeed/inference/v2/modules/ds_module.py | 62 ++ deepspeed/inference/v2/modules/heuristics.py | 179 +++++ .../v2/modules/implementations/__init__.py | 13 + .../implementations/attention/__init__.py | 6 + .../attention/dense_blocked_attention.py | 167 +++++ .../implementations/embedding/__init__.py | 6 + .../embedding/ragged_embedding.py | 77 +++ .../implementations/linear/__init__.py | 7 + .../implementations/linear/blas_fp_linear.py | 101 +++ .../linear/cutlass_fp_linear.py | 81 +++ .../modules/implementations/moe/__init__.py | 6 + .../implementations/moe/cutlass_multi_gemm.py | 225 +++++++ .../v2/modules/implementations/moe/gate_fn.py | 62 ++ .../v2/modules/implementations/moe/test.py | 38 ++ .../implementations/post_norm/__init__.py | 6 + .../implementations/post_norm/cuda_post_ln.py | 54 ++ .../implementations/pre_norm/__init__.py | 7 + .../implementations/pre_norm/cuda_pre_ln.py | 67 ++ .../implementations/pre_norm/cuda_pre_rms.py | 77 +++ .../implementations/unembed/__init__.py | 6 + .../implementations/unembed/ragged_unembed.py | 115 ++++ .../v2/modules/interfaces/__init__.py | 12 + .../v2/modules/interfaces/attention_base.py | 97 +++ .../v2/modules/interfaces/embedding_base.py | 84 +++ .../v2/modules/interfaces/linear_base.py | 71 ++ .../v2/modules/interfaces/moe_base.py | 90 +++ .../v2/modules/interfaces/post_norm_base.py | 68 ++ .../v2/modules/interfaces/pre_norm_base.py | 72 ++ .../v2/modules/interfaces/unembed_base.py | 61 ++ .../inference/v2/modules/module_registry.py | 58 ++ deepspeed/inference/v2/ragged/__init__.py | 15 + .../inference/v2/ragged/blocked_allocator.py | 105 +++ .../v2/ragged/csrc/fast_host_buffer.cu | 18 + .../inference/v2/ragged/csrc/ragged_ops.cpp | 45 ++ .../v2/ragged/includes/fast_host_buffer.h | 14 + deepspeed/inference/v2/ragged/kv_cache.py | 165 +++++ .../inference/v2/ragged/manager_configs.py | 164 +++++ .../inference/v2/ragged/ragged_manager.py | 204 ++++++ .../inference/v2/ragged/ragged_wrapper.py | 267 ++++++++ .../v2/ragged/sequence_descriptor.py | 214 ++++++ deepspeed/inference/v2/scheduling_utils.py | 54 ++ op_builder/builder.py | 12 +- op_builder/inference_core_ops.py | 92 +++ op_builder/inference_cutlass_builder.py | 92 +++ op_builder/ragged_ops.py | 116 ++++ op_builder/ragged_utils.py | 77 +++ requirements/requirements-dev.txt | 1 + scripts/check-torchcuda.py | 2 + tests/pytest.ini | 4 +- tests/unit/inference/__init__.py | 4 + tests/unit/inference/inference_test_utils.py | 46 ++ tests/unit/inference/kernels/__init__.py | 4 + .../inference/kernels/core_ops/__init__.py | 4 + .../kernels/core_ops/test_bias_activation.py | 101 +++ .../kernels/core_ops/test_blas_linear.py | 73 +++ .../kernels/core_ops/test_gated_activation.py | 133 ++++ .../kernels/core_ops/test_post_ln.py | 47 ++ .../inference/kernels/core_ops/test_pre_ln.py | 51 ++ .../kernels/core_ops/test_rms_norm.py | 77 +++ .../inference/kernels/cutlass_ops/__init__.py | 4 + .../kernels/cutlass_ops/test_moe_gemm.py | 113 ++++ .../inference/kernels/ragged_ops/__init__.py | 4 + .../ragged_ops/ragged_testing_utils.py | 300 +++++++++ .../kernels/ragged_ops/test_atom_builder.py | 45 ++ .../kernels/ragged_ops/test_blocked_flash.py | 197 ++++++ .../ragged_ops/test_blocked_kv_copy.py | 112 ++++ .../ragged_ops/test_blocked_rotary_emb.py | 203 ++++++ .../kernels/ragged_ops/test_logits_gather.py | 96 +++ .../kernels/ragged_ops/test_moe_gather.py | 83 +++ .../kernels/ragged_ops/test_moe_scatter.py | 74 +++ .../kernels/ragged_ops/test_ragged_embed.py | 177 +++++ .../kernels/ragged_ops/test_top_1_gating.py | 120 ++++ .../model_implementations/__init__.py | 4 + .../parameters/__init__.py | 4 + .../parameters/test_layer_inheritance.py | 50 ++ .../parameters/test_mapping.py | 165 +++++ .../parameters/test_multi_parameter_layer.py | 111 ++++ .../parameters/test_parameter_list.py | 104 +++ .../model_implementations/parameters/utils.py | 58 ++ .../sharding/__init__.py | 4 + .../sharding/test_attn_out_sharding.py | 129 ++++ .../sharding/test_mlp_sharding.py | 116 ++++ .../sharding/test_qkv_sharding.py | 251 +++++++ tests/unit/inference/modules/__init__.py | 4 + .../modules/test_blas_linear_module.py | 111 ++++ .../inference/modules/test_blocked_attn.py | 210 ++++++ .../modules/test_cuda_pre_ln_module.py | 88 +++ .../inference/modules/test_custom_module.py | 76 +++ .../inference/modules/test_cutlass_moe.py | 214 ++++++ .../inference/modules/test_post_ln_module.py | 58 ++ .../inference/modules/test_pre_rms_module.py | 88 +++ tests/unit/inference/ragged/__init__.py | 4 + .../ragged/test_blocked_allocator.py | 166 +++++ .../inference/ragged/test_manager_configs.py | 58 ++ .../inference/ragged/test_ragged_wrapper.py | 112 ++++ 263 files changed, 19167 insertions(+), 33 deletions(-) create mode 100644 .github/workflows/nv-a6000.yml create mode 100644 .gitmodules create mode 100644 csrc/includes/activation_type.h create mode 100644 deepspeed/inference/v2/__init__.py create mode 100644 deepspeed/inference/v2/allocator.py create mode 100644 deepspeed/inference/v2/checkpoint/__init__.py create mode 100644 deepspeed/inference/v2/checkpoint/base_engine.py create mode 100644 deepspeed/inference/v2/checkpoint/huggingface_engine.py create mode 100644 deepspeed/inference/v2/checkpoint/in_memory_engine.py create mode 100644 deepspeed/inference/v2/config_v2.py create mode 100644 deepspeed/inference/v2/engine_factory.py create mode 100644 deepspeed/inference/v2/engine_v2.py create mode 100644 deepspeed/inference/v2/inference_utils.py create mode 100644 deepspeed/inference/v2/kernels/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/bias_activations/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cpp create mode 100644 deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cu create mode 100644 deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.h create mode 100644 deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/blas_kernels/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas.h create mode 100644 deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_linear.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_utils.h create mode 100644 deepspeed/inference/v2/kernels/core_ops/core_ops.cpp create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_fp_ln_base.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_ln.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_post_ln.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_pre_ln.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cpp create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cu create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.h create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cpp create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cu create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.h create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm_base.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_pre_norm.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/gated_activations/__init__.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation.py create mode 100644 deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cpp create mode 100644 deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cu create mode 100644 deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.h create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/LICENSE create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/__init__.py create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/cutlass_ops.cpp create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/__init__.py create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.cu create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.h create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.py create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm_api.h create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/__init__.py create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/mixed_moe_gemm.py create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.cu create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.h create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.py create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm_api.h create mode 100644 deepspeed/inference/v2/kernels/cutlass_ops/shared_resources/weight_variant.h create mode 100644 deepspeed/inference/v2/kernels/ds_kernel.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/atom_builder/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/attention_atom.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/flash.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/embed/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cu create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cuh create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/embed/embed.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/embed/embed.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cu create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cuh create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_trained_kv_rotary.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/linear_blocked_kv_copy.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/logits_gather/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cu create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cuh create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_gather/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cu create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cuh create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cu create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cuh create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_dtypes.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/ragged_ops.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/__init__.py create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cpp create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cu create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cuh create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.h create mode 100644 deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.py create mode 100644 deepspeed/inference/v2/logging.py create mode 100644 deepspeed/inference/v2/model_implementations/AddingAModel.md create mode 100644 deepspeed/inference/v2/model_implementations/__init__.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/__init__.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/attn_output_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/embedding_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/invfreq_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/mlp_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/moe_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/norm_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/qkv_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/common_parameters/unembed_parameters.py create mode 100644 deepspeed/inference/v2/model_implementations/inference_model_base.py create mode 100644 deepspeed/inference/v2/model_implementations/inference_policy_base.py create mode 100644 deepspeed/inference/v2/model_implementations/inference_transformer_base.py create mode 100644 deepspeed/inference/v2/model_implementations/layer_container_base.py create mode 100644 deepspeed/inference/v2/model_implementations/llama_v2/__init__.py create mode 100644 deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_containers.py create mode 100644 deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_model.py create mode 100644 deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_policy.py create mode 100644 deepspeed/inference/v2/model_implementations/mistral/__init__.py create mode 100644 deepspeed/inference/v2/model_implementations/mistral/container.py create mode 100644 deepspeed/inference/v2/model_implementations/mistral/model.py create mode 100644 deepspeed/inference/v2/model_implementations/mistral/policy.py create mode 100644 deepspeed/inference/v2/model_implementations/opt/__init__.py create mode 100644 deepspeed/inference/v2/model_implementations/opt/container.py create mode 100644 deepspeed/inference/v2/model_implementations/opt/model.py create mode 100644 deepspeed/inference/v2/model_implementations/opt/policy.py create mode 100644 deepspeed/inference/v2/model_implementations/parameter_base.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/__init__.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/attn.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/attn_out.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/embedding.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/mlp.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/qkv.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/types.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/unembed.py create mode 100644 deepspeed/inference/v2/model_implementations/sharding/utils.py create mode 100644 deepspeed/inference/v2/modules/__init__.py create mode 100644 deepspeed/inference/v2/modules/configs/__init__.py create mode 100644 deepspeed/inference/v2/modules/configs/attention_configs.py create mode 100644 deepspeed/inference/v2/modules/configs/embedding_config.py create mode 100644 deepspeed/inference/v2/modules/configs/linear_config.py create mode 100644 deepspeed/inference/v2/modules/configs/moe_config.py create mode 100644 deepspeed/inference/v2/modules/configs/norm_config.py create mode 100644 deepspeed/inference/v2/modules/configs/unembed_config.py create mode 100644 deepspeed/inference/v2/modules/ds_module.py create mode 100644 deepspeed/inference/v2/modules/heuristics.py create mode 100644 deepspeed/inference/v2/modules/implementations/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/attention/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/attention/dense_blocked_attention.py create mode 100644 deepspeed/inference/v2/modules/implementations/embedding/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/embedding/ragged_embedding.py create mode 100644 deepspeed/inference/v2/modules/implementations/linear/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/linear/blas_fp_linear.py create mode 100644 deepspeed/inference/v2/modules/implementations/linear/cutlass_fp_linear.py create mode 100644 deepspeed/inference/v2/modules/implementations/moe/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/moe/cutlass_multi_gemm.py create mode 100644 deepspeed/inference/v2/modules/implementations/moe/gate_fn.py create mode 100644 deepspeed/inference/v2/modules/implementations/moe/test.py create mode 100644 deepspeed/inference/v2/modules/implementations/post_norm/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/post_norm/cuda_post_ln.py create mode 100644 deepspeed/inference/v2/modules/implementations/pre_norm/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_ln.py create mode 100644 deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_rms.py create mode 100644 deepspeed/inference/v2/modules/implementations/unembed/__init__.py create mode 100644 deepspeed/inference/v2/modules/implementations/unembed/ragged_unembed.py create mode 100644 deepspeed/inference/v2/modules/interfaces/__init__.py create mode 100644 deepspeed/inference/v2/modules/interfaces/attention_base.py create mode 100644 deepspeed/inference/v2/modules/interfaces/embedding_base.py create mode 100644 deepspeed/inference/v2/modules/interfaces/linear_base.py create mode 100644 deepspeed/inference/v2/modules/interfaces/moe_base.py create mode 100644 deepspeed/inference/v2/modules/interfaces/post_norm_base.py create mode 100644 deepspeed/inference/v2/modules/interfaces/pre_norm_base.py create mode 100644 deepspeed/inference/v2/modules/interfaces/unembed_base.py create mode 100644 deepspeed/inference/v2/modules/module_registry.py create mode 100644 deepspeed/inference/v2/ragged/__init__.py create mode 100644 deepspeed/inference/v2/ragged/blocked_allocator.py create mode 100644 deepspeed/inference/v2/ragged/csrc/fast_host_buffer.cu create mode 100644 deepspeed/inference/v2/ragged/csrc/ragged_ops.cpp create mode 100644 deepspeed/inference/v2/ragged/includes/fast_host_buffer.h create mode 100644 deepspeed/inference/v2/ragged/kv_cache.py create mode 100644 deepspeed/inference/v2/ragged/manager_configs.py create mode 100644 deepspeed/inference/v2/ragged/ragged_manager.py create mode 100644 deepspeed/inference/v2/ragged/ragged_wrapper.py create mode 100644 deepspeed/inference/v2/ragged/sequence_descriptor.py create mode 100644 deepspeed/inference/v2/scheduling_utils.py create mode 100755 op_builder/inference_core_ops.py create mode 100644 op_builder/inference_cutlass_builder.py create mode 100644 op_builder/ragged_ops.py create mode 100755 op_builder/ragged_utils.py create mode 100644 tests/unit/inference/__init__.py create mode 100644 tests/unit/inference/inference_test_utils.py create mode 100644 tests/unit/inference/kernels/__init__.py create mode 100644 tests/unit/inference/kernels/core_ops/__init__.py create mode 100644 tests/unit/inference/kernels/core_ops/test_bias_activation.py create mode 100644 tests/unit/inference/kernels/core_ops/test_blas_linear.py create mode 100644 tests/unit/inference/kernels/core_ops/test_gated_activation.py create mode 100644 tests/unit/inference/kernels/core_ops/test_post_ln.py create mode 100644 tests/unit/inference/kernels/core_ops/test_pre_ln.py create mode 100644 tests/unit/inference/kernels/core_ops/test_rms_norm.py create mode 100644 tests/unit/inference/kernels/cutlass_ops/__init__.py create mode 100644 tests/unit/inference/kernels/cutlass_ops/test_moe_gemm.py create mode 100644 tests/unit/inference/kernels/ragged_ops/__init__.py create mode 100644 tests/unit/inference/kernels/ragged_ops/ragged_testing_utils.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_atom_builder.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_blocked_flash.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_blocked_kv_copy.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_blocked_rotary_emb.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_logits_gather.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_moe_gather.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_moe_scatter.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_ragged_embed.py create mode 100644 tests/unit/inference/kernels/ragged_ops/test_top_1_gating.py create mode 100644 tests/unit/inference/model_implementations/__init__.py create mode 100644 tests/unit/inference/model_implementations/parameters/__init__.py create mode 100644 tests/unit/inference/model_implementations/parameters/test_layer_inheritance.py create mode 100644 tests/unit/inference/model_implementations/parameters/test_mapping.py create mode 100644 tests/unit/inference/model_implementations/parameters/test_multi_parameter_layer.py create mode 100644 tests/unit/inference/model_implementations/parameters/test_parameter_list.py create mode 100644 tests/unit/inference/model_implementations/parameters/utils.py create mode 100644 tests/unit/inference/model_implementations/sharding/__init__.py create mode 100644 tests/unit/inference/model_implementations/sharding/test_attn_out_sharding.py create mode 100644 tests/unit/inference/model_implementations/sharding/test_mlp_sharding.py create mode 100644 tests/unit/inference/model_implementations/sharding/test_qkv_sharding.py create mode 100644 tests/unit/inference/modules/__init__.py create mode 100644 tests/unit/inference/modules/test_blas_linear_module.py create mode 100644 tests/unit/inference/modules/test_blocked_attn.py create mode 100644 tests/unit/inference/modules/test_cuda_pre_ln_module.py create mode 100644 tests/unit/inference/modules/test_custom_module.py create mode 100644 tests/unit/inference/modules/test_cutlass_moe.py create mode 100644 tests/unit/inference/modules/test_post_ln_module.py create mode 100644 tests/unit/inference/modules/test_pre_rms_module.py create mode 100644 tests/unit/inference/ragged/__init__.py create mode 100644 tests/unit/inference/ragged/test_blocked_allocator.py create mode 100644 tests/unit/inference/ragged/test_manager_configs.py create mode 100644 tests/unit/inference/ragged/test_ragged_wrapper.py diff --git a/.github/workflows/nv-a6000.yml b/.github/workflows/nv-a6000.yml new file mode 100644 index 000000000000..38d4ffd1828b --- /dev/null +++ b/.github/workflows/nv-a6000.yml @@ -0,0 +1,56 @@ +name: nv-a6000 + +on: + pull_request: + paths-ignore: + - 'docs/**' + - 'blogs/**' + workflow_dispatch: + +concurrency: + group: ${{ github.workflow }}-${{ github.ref }} + cancel-in-progress: true + +permissions: + contents: read + issues: write + +jobs: + unit-tests: + runs-on: [self-hosted, nvidia, a6000] + container: + image: nvcr.io/nvidia/pytorch:23.03-py3 + ports: + - 80 + options: --gpus all --shm-size "8G" + + steps: + - uses: actions/checkout@v3 + + - name: Check container state + run: | + ldd --version + nvcc --version + nvidia-smi + python -c "import torch; print('torch:', torch.__version__, torch)" + python -c "import torch; print('CUDA available:', torch.cuda.is_available())" + - name: Install transformers + run: | + git clone https://github.com/huggingface/transformers + cd transformers + git rev-parse --short HEAD + python -m pip install . + - name: Install deepspeed + run: | + python -m pip install docutils==0.18.1 jinja2==3.0 urllib3==1.26.11 ninja + python -m pip install .[dev,1bit,autotuning] + ds_report + - name: Python environment + run: | + python -m pip list + - name: Unit tests + run: | + unset TORCH_CUDA_ARCH_LIST # only jit compile for current arch + cd tests + python -m pytest --color=yes --durations=0 --verbose -rF -m 'inference_v2' unit/ --torch_ver="2.0" --cuda_ver="12" + python -m pytest --color=yes --durations=0 --verbose -rF -m 'inference_v2_ops' unit/ --torch_ver="2.0" --cuda_ver="12" diff --git a/.github/workflows/nv-pre-compile-ops.yml b/.github/workflows/nv-pre-compile-ops.yml index ccb6c25e14f7..f253340c6966 100644 --- a/.github/workflows/nv-pre-compile-ops.yml +++ b/.github/workflows/nv-pre-compile-ops.yml @@ -33,7 +33,7 @@ jobs: #python -c "import torch; print('CUDA available:', torch.cuda.is_available())" - name: Compile DeepSpeed Ops run: | - TORCH_CUDA_ARCH_LIST="7.0;7.5;8.0" DS_BUILD_OPS=1 DS_BUILD_SPARSE_ATTN=0 DS_BUILD_EVOFORMER_ATTN=0 pip3 install . + TORCH_CUDA_ARCH_LIST="7.0;7.5;8.0" DS_BUILD_OPS=1 DS_BUILD_SPARSE_ATTN=0 DS_BUILD_CUTLASS_OPS=0 DS_BUILD_RAGGED_DEVICE_OPS=0 DS_BUILD_EVOFORMER_ATTN=0 pip3 install . - name: DS Report run: | ds_report diff --git a/.gitmodules b/.gitmodules new file mode 100644 index 000000000000..e69de29bb2d1 diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index 6b11b3acba51..2432a7a24124 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -49,6 +49,7 @@ repos: entry: ./scripts/check-license.py language: python files: \.(py|c|cpp|cu|cc|h|hpp|cuh|hip|tr)$ + exclude: ^(deepspeed/inference/v2/kernels/ragged_ops/blocked_flash|deepspeed/inference/v2/kernels/cutlass_ops/grouped_gemm) - repo: https://github.com/codespell-project/codespell rev: v2.1.0 diff --git a/MANIFEST.in b/MANIFEST.in index 2fec750c6644..ab79573ef96c 100644 --- a/MANIFEST.in +++ b/MANIFEST.in @@ -1,4 +1,6 @@ include *.txt README.md +include deepspeed/inference/v2/kernels/ragged_ops/libs/*.so +include deepspeed/inference/v2/kernels/cutlass_ops/libs/*.so recursive-include requirements *.txt recursive-include deepspeed *.cpp *.h *.cu *.hip *.tr *.cuh *.cc *.json recursive-include csrc *.cpp *.h *.cu *.tr *.cuh *.cc diff --git a/accelerator/cuda_accelerator.py b/accelerator/cuda_accelerator.py index 4715e28b192c..2786b425ca7f 100644 --- a/accelerator/cuda_accelerator.py +++ b/accelerator/cuda_accelerator.py @@ -153,9 +153,26 @@ def max_memory_reserved(self, device_index=None): def total_memory(self, device_index=None): return torch.cuda.get_device_properties(device_index).total_memory + def _get_nvml_gpu_id(self, torch_gpu_id): + """ + credit: https://discuss.pytorch.org/t/making-pynvml-match-torch-device-ids-cuda-visible-devices/103020 + + Remap torch device id to nvml device id, respecting CUDA_VISIBLE_DEVICES. + + If the latter isn't set return the same id + """ + # if CUDA_VISIBLE_DEVICES is used automagically remap the id since pynvml ignores this env var + if "CUDA_VISIBLE_DEVICES" in os.environ: + ids = list(map(int, os.environ.get("CUDA_VISIBLE_DEVICES", "").split(","))) + return ids[torch_gpu_id] # remap + else: + return torch_gpu_id + def available_memory(self, device_index=None): if pynvml: - handle = pynvml.nvmlDeviceGetHandleByIndex(device_index) + if device_index is None: + device_index = self.current_device() + handle = pynvml.nvmlDeviceGetHandleByIndex(self._get_nvml_gpu_id(device_index)) info = pynvml.nvmlDeviceGetMemoryInfo(handle) return info.free else: diff --git a/csrc/includes/activation_type.h b/csrc/includes/activation_type.h new file mode 100644 index 000000000000..a44921d5d650 --- /dev/null +++ b/csrc/includes/activation_type.h @@ -0,0 +1,17 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +enum ActivationType { + GELU = 0, + RELU = 1, + SILU = 2, + GEGLU = 3, + ReGLU = 4, + SiGLU = 5, + IDENTITY = 6, + InvalidType = -1 +}; diff --git a/csrc/includes/ds_kernel_utils.h b/csrc/includes/ds_kernel_utils.h index a3af561adfe5..8e4888109fcd 100644 --- a/csrc/includes/ds_kernel_utils.h +++ b/csrc/includes/ds_kernel_utils.h @@ -11,6 +11,11 @@ used throughout the codebase. #pragma once #include +#include + +#ifdef BF16_AVAILABLE +#include +#endif #define DS_HD_INLINE __host__ __device__ __forceinline__ #define DS_D_INLINE __device__ __forceinline__ diff --git a/csrc/includes/reduction_utils.h b/csrc/includes/reduction_utils.h index 8d0c2d6986a1..eb8efab77ac1 100644 --- a/csrc/includes/reduction_utils.h +++ b/csrc/includes/reduction_utils.h @@ -233,6 +233,60 @@ DS_D_INLINE __half2 element(const __half2 lhs, const __half2 rhs) #endif } +template <> +DS_D_INLINE int32_t element(const int32_t lhs, const int32_t rhs) +{ + return lhs + rhs; +} + +template <> +DS_D_INLINE int32_t element(const int32_t lhs, const int32_t rhs) +{ + return (lhs > rhs) ? lhs : rhs; +} + +template <> +DS_D_INLINE int32_t element(const int32_t lhs, const int32_t rhs) +{ + return (lhs < rhs) ? lhs : rhs; +} + +template <> +DS_D_INLINE uint32_t element(const uint32_t lhs, const uint32_t rhs) +{ + return lhs + rhs; +} + +template <> +DS_D_INLINE uint32_t element(const uint32_t lhs, const uint32_t rhs) +{ + return (lhs > rhs) ? lhs : rhs; +} + +template <> +DS_D_INLINE uint32_t element(const uint32_t lhs, const uint32_t rhs) +{ + return (lhs < rhs) ? lhs : rhs; +} + +template <> +DS_D_INLINE int64_t element(const int64_t lhs, const int64_t rhs) +{ + return lhs + rhs; +} + +template <> +DS_D_INLINE int64_t element(const int64_t lhs, const int64_t rhs) +{ + return (lhs > rhs) ? lhs : rhs; +} + +template <> +DS_D_INLINE int64_t element(const int64_t lhs, const int64_t rhs) +{ + return (lhs < rhs) ? lhs : rhs; +} + /* Reduction initialization primitives */ @@ -310,6 +364,78 @@ DS_D_INLINE __half2 init() #endif } +template <> +DS_D_INLINE int32_t init() +{ + return 0; +} + +template <> +DS_D_INLINE int32_t init() +{ + return 0x7FFFFFFF; +} + +template <> +DS_D_INLINE int32_t init() +{ + return 0x80000000; +} + +template <> +DS_D_INLINE uint32_t init() +{ + return 0; +} + +template <> +DS_D_INLINE uint32_t init() +{ + return 0xFFFFFFFF; +} + +template <> +DS_D_INLINE uint32_t init() +{ + return 0; +} + +template <> +DS_D_INLINE int64_t init() +{ + return 0; +} + +template <> +DS_D_INLINE int64_t init() +{ + return 0x7FFFFFFFFFFFFFFF; +} + +template <> +DS_D_INLINE int64_t init() +{ + return 0x8000000000000000; +} + +template <> +DS_D_INLINE uint64_t init() +{ + return 0; +} + +template <> +DS_D_INLINE uint64_t init() +{ + return 0xFFFFFFFFFFFFFFFF; +} + +template <> +DS_D_INLINE uint64_t init() +{ + return 0; +} + template DS_D_INLINE void init(T* data) { @@ -352,8 +478,8 @@ here (fold is C++17 only and I don't think helps and recursion feels like huge overkill that harms readability) that would be wonderful. */ -template -DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) +template +DS_D_INLINE void _warp(cg::thread_block_tile& warp, T* data) { #pragma unroll for (int i = 1; i < reduce_width; i *= 2) { @@ -361,8 +487,8 @@ DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) } } -template -DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) +template +DS_D_INLINE void _warp(cg::thread_block_tile& warp, T* data) { #pragma unroll for (int i = 1; i < reduce_width; i *= 2) { @@ -371,8 +497,8 @@ DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) } } -template -DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) +template +DS_D_INLINE void _warp(cg::thread_block_tile& warp, T* data) { #pragma unroll for (int i = 1; i < reduce_width; i *= 2) { @@ -382,8 +508,13 @@ DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) } } -template -DS_D_INLINE void _warp(cg::thread_block_tile& warp, float* data) +template +DS_D_INLINE void _warp(cg::thread_block_tile& warp, T* data) { #pragma unroll for (int i = 1; i < reduce_width; i *= 2) { @@ -403,16 +534,15 @@ the number of warps in the block (which may exceed that if the block is partitioned or if we do a conservative bound at compile time). */ -template +template DS_D_INLINE void _block(cg::thread_block& tb, cg::thread_block_tile& warp_arg, - float* data) + T* data) { constexpr int elems = sizeof...(Ops); - // Separated for now in case this no longer is true - constexpr int bytes = sizeof(float); + constexpr int bytes = sizeof(T); // Unused when `partition_size == 1` or total_warps == 1 - __shared__ float reduce_buffer[max_warps * elems]; + __shared__ T reduce_buffer[max_warps * elems]; #ifdef __HIP_PLATFORM_AMD__ const int total_threads = blockDim.x * blockDim.y * blockDim.z; @@ -422,7 +552,7 @@ DS_D_INLINE void _block(cg::thread_block& tb, #endif // Always perform warp-scope reduction - _warp(warp_arg, data); + _warp(warp_arg, data); // If max_warps == 1 let's skip the runtime check if (total_warps != 1) { @@ -447,7 +577,7 @@ DS_D_INLINE void _block(cg::thread_block& tb, init(data); } - _warp(warp_arg, data); + _warp(warp_arg, data); #pragma unroll for (int i = 0; i < elems; i++) { @@ -476,7 +606,7 @@ us to obfuscate the details of the partitioned implementation. template DS_D_INLINE void block(cg::thread_block& tb, cg::thread_block_tile& warp, float& val) { - _block(tb, warp, &val); + _block(tb, warp, &val); } template @@ -486,7 +616,7 @@ DS_D_INLINE void block(cg::thread_block& tb, float& val2) { float data[2] = {val1, val2}; - _block(tb, warp, data); + _block(tb, warp, data); val1 = data[0]; val2 = data[1]; } @@ -499,7 +629,7 @@ DS_D_INLINE void block(cg::thread_block& tb, float& val3) { float data[3] = {val1, val2, val3}; - _block(tb, warp, data); + _block(tb, warp, data); val1 = data[0]; val2 = data[1]; val3 = data[2]; @@ -514,7 +644,7 @@ DS_D_INLINE void block(cg::thread_block& tb, float& val4) { float data[4] = {val1, val2, val3, val4}; - _block(tb, warp, data); + _block(tb, warp, data); val1 = data[0]; val2 = data[1]; val3 = data[2]; @@ -531,10 +661,10 @@ DS_D_INLINE void partitioned_block(cg::thread_block& tb, float& val) { if (num_threads <= hw_warp_size) { - _warp(warp, &val); + _warp(warp, &val); } else { constexpr int num_warps = num_threads / hw_warp_size; - _block(tb, warp, &val); + _block(tb, warp, &val); } } @@ -547,10 +677,10 @@ DS_D_INLINE void partitioned_block(cg::thread_block& tb, float data[2] = {val1, val2}; if (num_threads <= hw_warp_size) { - _warp(warp, data); + _warp(warp, data); } else { constexpr int num_warps = num_threads / hw_warp_size; - _block(tb, warp, data); + _block(tb, warp, data); } val1 = data[0]; @@ -567,10 +697,10 @@ DS_D_INLINE void partitioned_block(cg::thread_block& tb, float data[3] = {val1, val2, val3}; if (num_threads <= hw_warp_size) { - _warp(warp, data); + _warp(warp, data); } else { constexpr int num_warps = num_threads / hw_warp_size; - _block(tb, warp, data); + _block(tb, warp, data); } val1 = data[0]; @@ -589,10 +719,10 @@ DS_D_INLINE void partitioned_block(cg::thread_block& tb, float data[4] = {val1, val2, val3, val4}; if (num_threads <= hw_warp_size) { - _warp(warp, data); + _warp(warp, data); } else { constexpr int num_warps = num_threads / hw_warp_size; - _block(tb, warp, data); + _block(tb, warp, data); } val1 = data[0]; @@ -601,4 +731,48 @@ DS_D_INLINE void partitioned_block(cg::thread_block& tb, val4 = data[3]; } +/* +Arg-reduce is a specialization of the above. We only support this with a single reduction +parameter. This only works for max/min reductions. +*/ + +__align__(8) struct IdxReduceResult { + /* + NOTE: ORDERING MATTERS HERE! The idx is the least significant set of bits + and the val is the most significant. Changing the order of this declaration + will break the code. + */ + int idx; + float val; +}; + +template +DS_D_INLINE IdxReduceResult +idx_reduce(cg::thread_block& tb, cg::thread_block_tile& warp, float val, int idx) +{ + IdxReduceResult res = {idx, val}; + + // Clear out the nan. This shouldn't be an issue for our initial applications + if (isnan(val)) res.val = init(); + + // Can do float compares as integers. By packing the index into the lower bits + // we can just do a single int64 rather than a branch, compare, and select. + // One side benefit of this is that it is by nature a stable algorithm and + // will always bias ties to the higher index. + int64_t* res_as_int = reinterpret_cast(&res); + + // The way floating point compare works is normally to perform a sign comparison + // and if they match, then do a comparison of the rest of the bits as unsigned + // integers. Since we are bundling these, that means for negative values we need + // to reverse the sort order, which we can do with an XOR. + if (val < 0) { *res_as_int ^= 0x7fffffff00000000; } + + _block(tb, warp, res_as_int); + + // Sign bit is preserved, so we can check if we need to invert the mantissa back + if (res.val < 0) { *res_as_int ^= 0x7fffffff00000000; } + + return res; +} + } // namespace reduce diff --git a/csrc/transformer/inference/csrc/gelu.cu b/csrc/transformer/inference/csrc/gelu.cu index 3e6701d81e64..dc7ff4d1e7c0 100644 --- a/csrc/transformer/inference/csrc/gelu.cu +++ b/csrc/transformer/inference/csrc/gelu.cu @@ -18,8 +18,8 @@ using __nv_bfloat162 = __half2; inline __device__ float gelu(const float x) { - const float sqrt_param = 0.79788456080286535587989211986876f; - const float mul_param = 0.044715; + constexpr float sqrt_param = 0.79788456080286535587989211986876f; + constexpr float mul_param = 0.044715; return x * 0.5f * (1.0f + tanhf(sqrt_param * (x + mul_param * x * x * x))); } diff --git a/deepspeed/inference/__init__.py b/deepspeed/inference/__init__.py index 208299fb8c50..7fed50cbe177 100644 --- a/deepspeed/inference/__init__.py +++ b/deepspeed/inference/__init__.py @@ -2,3 +2,6 @@ # SPDX-License-Identifier: Apache-2.0 # DeepSpeed Team +from .v2 import RaggedInferenceEngineConfig, DeepSpeedTPConfig +from .v2.engine_v2 import InferenceEngineV2 +from .v2 import build_hf_engine diff --git a/deepspeed/inference/v2/__init__.py b/deepspeed/inference/v2/__init__.py new file mode 100644 index 000000000000..bba5d1a82081 --- /dev/null +++ b/deepspeed/inference/v2/__init__.py @@ -0,0 +1,7 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team +from .config_v2 import RaggedInferenceEngineConfig, DeepSpeedTPConfig +from .engine_v2 import InferenceEngineV2 +from .engine_factory import build_hf_engine diff --git a/deepspeed/inference/v2/allocator.py b/deepspeed/inference/v2/allocator.py new file mode 100644 index 000000000000..fa2c5368604e --- /dev/null +++ b/deepspeed/inference/v2/allocator.py @@ -0,0 +1,32 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from functools import reduce +from typing import Iterable + +import torch + +from deepspeed.accelerator import get_accelerator + + +def empty_from(tensor: torch.Tensor, shape: Iterable[int]) -> torch.Tensor: + shape_size = reduce(lambda x, y: x * y, shape) + if shape_size == 0: + raise ValueError("Cannot create empty tensor with size 0") + return tensor.flatten()[:shape_size].view(shape) + + +def on_device(method) -> torch.Tensor: + """ + Wraps a method to ensure the returned tensor is on the current device. + """ + + def wrapped(self, *args, **kwargs): + tensor = method(self, *args, **kwargs) + if isinstance(tensor, torch.Tensor): + return tensor.to(get_accelerator().current_device()).contiguous() + return tensor + + return wrapped diff --git a/deepspeed/inference/v2/checkpoint/__init__.py b/deepspeed/inference/v2/checkpoint/__init__.py new file mode 100644 index 000000000000..45e523ab62b9 --- /dev/null +++ b/deepspeed/inference/v2/checkpoint/__init__.py @@ -0,0 +1,8 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .base_engine import CheckpointEngineBase +from .in_memory_engine import InMemoryModelEngine +from .huggingface_engine import HuggingFaceCheckpointEngine diff --git a/deepspeed/inference/v2/checkpoint/base_engine.py b/deepspeed/inference/v2/checkpoint/base_engine.py new file mode 100644 index 000000000000..26fc467d4d86 --- /dev/null +++ b/deepspeed/inference/v2/checkpoint/base_engine.py @@ -0,0 +1,41 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import ABC, abstractmethod +from typing import Iterable, Tuple + +import torch + +#from .huggingface_engine import HuggingFaceCheckpointEngine + +MEGATRON = 'megatron' +HUGGINGFACE = 'huggingface' + + +class CheckpointEngineBase(ABC): + """ + Abstract interface for checkpoint engines to implement. + + There is no ``__init__`` method here by design, since the creation of the checkpoint + engine will happen outside the policy/engine code. The tradeoff being made here is + that we will write different frontends for different checkpoint engines, but these + frontends can be tailored to the specific checkpoint engine/model source needs. + """ + + @abstractmethod + def parameters(self) -> Iterable[Tuple[str, torch.Tensor]]: + """ + This method should create a generator of tuples of the form (name, parameter) for + all parameters in the model. The name should be the fully qualified name of the + parameter, and the parameter should be a torch.Tensor. + + The expected use of a checkpoint engine is the following: + ```python + for name, parameter in checkpoint_engine.parameters(): + container_map.map_param(name, parameter) + ``` + For a concrete use example, see ``InferenceV2Policy``. + """ + ... diff --git a/deepspeed/inference/v2/checkpoint/huggingface_engine.py b/deepspeed/inference/v2/checkpoint/huggingface_engine.py new file mode 100644 index 000000000000..515378d31d02 --- /dev/null +++ b/deepspeed/inference/v2/checkpoint/huggingface_engine.py @@ -0,0 +1,99 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import os +import json +import torch +from .base_engine import CheckpointEngineBase +from typing import Iterable, Tuple + +from ..logging import inference_logger + + +class HuggingFaceCheckpointEngine(CheckpointEngineBase): + + def __init__(self, model_name_or_path: str, auth_token: str = None) -> None: + super().__init__() + from transformers import AutoConfig, GenerationConfig + + self.model_name_or_path = model_name_or_path + self.auth_token = auth_token + self.model_config = AutoConfig.from_pretrained(self.model_name_or_path) + self.generation_config = GenerationConfig.from_pretrained(self.model_name_or_path) + # Define this property here so we can use it in the model implementation + if not hasattr(self.model_config, "max_seq_length"): + self.model_config.max_seq_length = self.model_config.max_position_embeddings + else: + self.model_config.max_seq_length = self.generation_config.max_length + + self._all_ckpt_paths = self._fetch_checkpoint_files() + + def _fetch_checkpoint_files(self): + """ + Fetch the checkpoint files from the HuggingFace Hub. + """ + # TODO(jeff): for models like llama-2 the user will have to provide an auth `token`, + # currently coming from the ckpt engine init but maybe a catch all kwargs for other + # snapshot download parameters would be more flexible. + + # NOTE(jeff): allow_patterns here are explicitly not using safetensors or other + # checkpoint files that may be present. Example of all files in the llama-2-7b + # repo here: https://huggingface.co/meta-llama/Llama-2-7b-hf/tree/main + from huggingface_hub import snapshot_download + + if os.path.isdir(self.model_name_or_path): + self._local_checkpoint_dir = self.model_name_or_path + else: + self._local_checkpoint_dir = snapshot_download(self.model_name_or_path, + allow_patterns=[ + "*.bin", + "*.json", + "*.pt", + ], + revision=None, + token=self.auth_token) + + assert os.path.isdir( + self._local_checkpoint_dir + ), f"Checkpoint dir {self._local_checkpoint_dir} is not a directory, cannot load checkpoint." + + model_param_json = os.path.join(self._local_checkpoint_dir, "pytorch_model.bin.index.json") + + if not os.path.isfile(model_param_json): + # We don't need any json as all such HF models will have pytorch_model.bin + all_checkpoint_files = [os.path.join(self._local_checkpoint_dir, 'pytorch_model.bin')] + else: + param_map = json.load(open(model_param_json, "r")) + + # weight_map -> { "lm_head.weight": "pytorch_model-00002-of-00002.bin", ... } + weight_map = param_map["weight_map"] + + # unique set of all checkpoint files + all_checkpoint_files = set(weight_map.values()) + + # get absolute path of all unique checkpoint files + all_checkpoint_files = [os.path.join(self._local_checkpoint_dir, f) for f in all_checkpoint_files] + + return all_checkpoint_files + + def parameters(self) -> Iterable[Tuple[str, torch.Tensor]]: + """ + Generator of model parameters (satisfies the CheckpointEngineBase interface). + """ + for checkpoint in self._all_ckpt_paths: + inference_logger().info(f"Loading checkpoint: {checkpoint}") + checkpoint_sd = torch.load(checkpoint, map_location='cpu') + param_keys = list(checkpoint_sd.keys()) + for param_name in param_keys: + param = checkpoint_sd[param_name] + yield param_name, param + + +if __name__ == "__main__": + # To test, add your auth_token here and run `python huggingface_engine.py` + engine = HuggingFaceCheckpointEngine(model_name_or_path="meta-llama/Llama-2-7b-hf", + auth_token="hf_xxxxxxxxxxxxxxxxx") + for name, param in engine.parameters(): + print(name, param.shape) diff --git a/deepspeed/inference/v2/checkpoint/in_memory_engine.py b/deepspeed/inference/v2/checkpoint/in_memory_engine.py new file mode 100644 index 000000000000..13ec7b288f5f --- /dev/null +++ b/deepspeed/inference/v2/checkpoint/in_memory_engine.py @@ -0,0 +1,40 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Iterable, Tuple +import torch + +from .base_engine import CheckpointEngineBase + + +class InMemoryModelEngine(CheckpointEngineBase): + """ + This "checkpoint" engine uses the existing interface to enable loading parameters into an + inference model from a model already instantiated in memory. In general, this is not the + recommended way to use the inference engine, and should only be used when absolutely necessary. + + The primary limitation of this approach is that the model must be fully instantiated in memory. + In a tensor parallel scenario, this means that the model is either replicated many times in host + memory. Currently, it is also recommended to only use this approach for models held in host memory. + + In order to free the memory held by this copy of the model, we delete the model in the first call + to `parameters`, so it is not safe to make this call twice. + """ + + def __init__(self, model: torch.nn.Module) -> None: + """ + Create virtual checkpoint engine for the provided module. + + Args: + model (torch.nn.Module): Model to load parameters from. + """ + super().__init__() + self.model = model + + def parameters(self) -> Iterable[Tuple[str, torch.Tensor]]: + for name, parameter in self.model.named_parameters(): + yield name, parameter + + del self.model diff --git a/deepspeed/inference/v2/config_v2.py b/deepspeed/inference/v2/config_v2.py new file mode 100644 index 000000000000..64e7e29b1844 --- /dev/null +++ b/deepspeed/inference/v2/config_v2.py @@ -0,0 +1,31 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from deepspeed.pydantic_v1 import Field + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from .ragged import DSStateManagerConfig + + +class DeepSpeedTPConfig(DeepSpeedConfigModel): + """ Configure tensor parallelism settings """ + + tp_size: int = 1 + """ Number of devices to split the model across using tensor parallelism. """ + + +class RaggedInferenceEngineConfig(DeepSpeedConfigModel): + """ Sets parameters for DeepSpeed Inference Engine. """ + + tensor_parallel: DeepSpeedTPConfig = Field({}, alias="tp") + """ + Configuration for tensor parallelism used to split the model across several + GPUs. Expects a dictionary containing values for :any:`DeepSpeedTPConfig`. + """ + + state_manager: DSStateManagerConfig = Field({}, alias="manager") + """ + Configuration for managing persistent state + """ diff --git a/deepspeed/inference/v2/engine_factory.py b/deepspeed/inference/v2/engine_factory.py new file mode 100644 index 000000000000..48274d6c3d53 --- /dev/null +++ b/deepspeed/inference/v2/engine_factory.py @@ -0,0 +1,46 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import logging +from typing import Any + +from .engine_v2 import InferenceEngineV2 +from .config_v2 import RaggedInferenceEngineConfig +from .checkpoint import HuggingFaceCheckpointEngine +from .logging import inference_logger + + +def build_hf_engine(path: str, + engine_config: RaggedInferenceEngineConfig, + debug_level: int = logging.INFO, + random_weights_config: Any = None, + fill_random: bool = False) -> InferenceEngineV2: + """ + Build an InferenceV2 engine for HuggingFace models. + """ + # Set up logging + inference_logger(level=debug_level) + + # get HF checkpoint engine + checkpoint_engine = HuggingFaceCheckpointEngine(path) + + # get model config from HF AutoConfig + model_config = checkpoint_engine.model_config + + # get the policy + # TODO: generalize this to other models + if model_config.model_type == "opt": + from .model_implementations.opt.policy import OPTPolicy + policy = OPTPolicy(checkpoint_engine, model_config) + elif model_config.model_type == "llama": + from .model_implementations.llama_v2.llama_v2_policy import Llama2Policy + policy = Llama2Policy(checkpoint_engine, model_config) + elif model_config.model_type == "mistral": + from .model_implementations.mistral.policy import MistralPolicy + policy = MistralPolicy(checkpoint_engine, model_config) + else: + raise ValueError(f"Unsupported model type {model_config.model_type}") + + return InferenceEngineV2(policy, engine_config) diff --git a/deepspeed/inference/v2/engine_v2.py b/deepspeed/inference/v2/engine_v2.py new file mode 100644 index 000000000000..c670ddd09417 --- /dev/null +++ b/deepspeed/inference/v2/engine_v2.py @@ -0,0 +1,217 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import os +from typing import Iterable, Tuple + +import torch + +import deepspeed.comm as dist + +from deepspeed.accelerator import get_accelerator +from deepspeed.comm.comm import init_distributed + +from .model_implementations import InferenceV2Policy +from .logging import inference_logger +from .ragged import DSStateManager, RaggedBatchWrapper, PlaceholderSequenceDescriptor +from .scheduling_utils import SchedulingError, SchedulingResult + +from .config_v2 import RaggedInferenceEngineConfig + +INFERENCE_MODEL_TIMER = "model-forward-inference" + + +class InferenceEngineV2: + + _config: RaggedInferenceEngineConfig + """ + Configuration of the inference engine. + """ + + #_model: DSInferenceModelBase + """ + Inference model supporting ragged inference. + """ + + _state_manager: DSStateManager + """ + Persistent state manager for sequences and KV-cache. + """ + + @property + def free_blocks(self) -> int: + """ + Number of free KV blocks. + """ + return self._state_manager.free_blocks + + def __init__(self, policy: InferenceV2Policy, engine_config: RaggedInferenceEngineConfig) -> None: + """ + Create the Inference V2 engine. + + Arguments: + policy (InferenceV2Policy): Policy for the model implementation. This policy object + will be used to build the model and load the checkpoint associated with it. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + """ + self._config = engine_config + self._policy = policy + self._base_mp_group = self._initialize_tp_group() + + # Build model from policy + inference_logger().info("Building model...") + self._model = self._policy.build_model(self._config, self._base_mp_group) + inference_logger().info("Model built.") + + # Create state manager + self._batch = RaggedBatchWrapper(self._config.state_manager) + self._state_manager = DSStateManager(self._config.state_manager, + self._model.kv_cache_config(), + base_mp_group=self._base_mp_group) + self._model.set_state_manager(self._state_manager) + + def _initialize_tp_group(self): + """ + Implementation of our TP group initialization. + """ + init_distributed() + local_rank = int(os.getenv("LOCAL_RANK", 0)) + get_accelerator().set_device(local_rank) + + if local_rank >= self._config.tensor_parallel.tp_size: + raise RuntimeError("Local rank is greater than TP size, ensure that the TP config is correct.") + + ranks = list(range(self._config.tensor_parallel.tp_size)) + return dist.new_group(ranks=ranks) + + def put(self, batch_uids: Iterable[int], batch_tokens: Iterable[torch.Tensor]) -> torch.Tensor: + """ + Put a ragged batch onto the inference engine. This will perform one forward and return + a Tensor of the shape [len(batch_uids), *output_shape]. Logits for the non-final tokens + are not calculated. + + Arguments: + batch_uids: Iterable of uids for the batch on the host + batch_tokens: Iterable of token tensors for the batch on the host + """ + + token_lens = [len(tokens) for tokens in batch_tokens] + schedule_check = self.can_schedule(batch_uids, token_lens) + if schedule_check != SchedulingResult.Success: + raise SchedulingError(schedule_check) + + self._batch.clear() + for uid, tokens in zip(batch_uids, batch_tokens): + + host_seq_desc = self._state_manager.get_or_create_sequence(uid) + self._model.maybe_allocate_kv(host_seq_desc, tokens.numel()) + host_seq_desc.pre_forward(tokens.numel()) + + # We can disable checks since we already validated schedulability. + self._batch.insert_sequence(host_seq_desc, tokens, do_checks=False) + + # Send all metadata to the device + self._batch.finalize() + + # Prep all data structures for the actual forward (in anticipation of CG in the future) + # and also to amortize some of the costs in a more straightforward way. + self._model.prepare_batch(self._batch) + + # Model implementation will pick up in the forward. + logits = self._model.forward(self._batch) + + # We return one set of logits per sequence in the batch (saves cost on unembedding) + assert logits.shape[0] == self._batch.current_sequences + + for uid in batch_uids: + host_seq_desc = self._state_manager.get_sequence(uid) + host_seq_desc.post_forward() # Updates sequence metadata. + self._model.maybe_free_kv(host_seq_desc) + + return logits + + def query(self, uid: int, max_request_tokens: int, max_request_blocks) -> Tuple[int, int]: + """ + Determine the number of tokens and KV blocks to reserve for a given request. Given a UID + (this UID may not be recognized by the model yet), this will return the number of tokens + and blocks to reserve for the request. + + Arguments: + uid (int): The UID of the sequence (as tracked by the scheduling entity). If + this is a new sequence (with a UID unknown to the inference engine), then + an empty placeholder is created to pass to the occupancy logic. + n_tokens (int): The number of tokens to hypothetically send. + + Returns: + Tuple[int, Optional[int]]: Tuple of free kv blocks and the number of blocks + required to schedule the sequence. + """ + seq_desc = self._state_manager.get_sequence(uid) + if seq_desc is None: + if (self._state_manager.n_tracked_sequences == self._config.state_manager.max_tracked_sequences): + return (0, 0) + seq_desc = PlaceholderSequenceDescriptor() + + req_tokens, req_blocks = self._model.get_kv_requirements(seq_desc, max_request_tokens, max_request_blocks) + + return (req_tokens, req_blocks) + + def can_schedule(self, uids: Iterable[int], lengths: Iterable[int]) -> SchedulingResult: + """ + Dry run a batch to determine if it can be scheduled. Placeholder sequences will be + created for any UIDs that are unknown to the inference engine. + + Arguments: + uids (Iterable[int]): Iterable of UIDs for the batch + lengths (Iterable[int]): Iterable of lengths for each sequence of the batch. This lengths + corresponds to the number of tokens to send in the hypothetical forward; history + tokens will be determined via UID lookup and future tokens are disregarded. + + Returns: + bool: True if the batch can be scheduled, False otherwise. + """ + + cur_seqs = self._state_manager.n_tracked_sequences + free_blocks = self._state_manager.free_blocks + req_blocks = 0 + batch_len = 0 + + if len(uids) > self._config.state_manager.max_ragged_sequence_count: + # Can only compose a batch from a limited number of sequences + return SchedulingResult.BatchSequenceLimitExceeded + + for uid, length in zip(uids, lengths): + seq_desc = self._state_manager.get_sequence(uid) + if seq_desc is None: + cur_seqs += 1 + seq_desc = PlaceholderSequenceDescriptor() + + sched_len, sched_blocks = self._model.get_kv_requirements(seq_desc, length, free_blocks) + + if sched_len != length: + # We ran out of KV cache + return SchedulingResult.KVCacheLimitExceeded + + batch_len += length + free_blocks -= sched_blocks + + if cur_seqs > self._config.state_manager.max_tracked_sequences: + # Would run out of tracking metadata + return SchedulingResult.EngineSequenceLimitExceeded + + if batch_len > self._config.state_manager.max_ragged_batch_size: + # Would exceed the maximum batch size + return SchedulingResult.BatchTokenLimitExceeded + + return SchedulingResult.Success + + def flush(self, uid: int) -> None: + """ + Remove all state associated with a sequence from the inference engine. + + Arguments: + uid (int): The UID of the sequence to flush. + """ + self._state_manager.flush_sequence(uid) diff --git a/deepspeed/inference/v2/inference_utils.py b/deepspeed/inference/v2/inference_utils.py new file mode 100644 index 000000000000..7b2dd4237353 --- /dev/null +++ b/deepspeed/inference/v2/inference_utils.py @@ -0,0 +1,105 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Dict + +import torch + +from enum import Enum, IntEnum + + +class NormTypeEnum(Enum): + LayerNorm: str = "layer_norm" + RMSNorm: str = "rms_norm" + + +class DtypeEnum(Enum): + # The torch dtype must always be the first value (so we return torch.dtype) + fp16 = torch.float16, "torch.float16", "fp16", "float16", "half" + fp32 = torch.float32, "torch.float32", "fp32", "float32", "float" + bf16 = torch.bfloat16, "torch.bfloat16", "bf16", "bfloat16", "bfloat" + int8 = torch.int8, "torch.int8", "int8" + + # Copied from https://stackoverflow.com/a/43210118 + # Allows us to use multiple values for each Enum index and returns first + # listed value when Enum is called + def __new__(cls, *values): + obj = object.__new__(cls) + # first value is canonical value + obj._value_ = values[0] + for other_value in values[1:]: + cls._value2member_map_[other_value] = obj + obj._all_values = values + return obj + + def __repr__(self): + return "<%s.%s: %s>" % ( + self.__class__.__name__, + self._name_, + ", ".join([repr(v) for v in self._all_values]), + ) + + +ELEM_SIZES: Dict[torch.dtype, int] = { + torch.float16: 2, + torch.bfloat16: 2, + torch.float32: 4, + torch.float64: 8, + torch.int8: 1, + torch.uint8: 1, + torch.int16: 2, + torch.int32: 4, + torch.int64: 8, + torch.bool: 1, +} + + +class ActivationType(IntEnum): + """ + Types of activations supported by DS-Inference + """ + + GELU = 0 + + RELU = 1 + + SILU = 2 + + GEGLU = 3 + + ReGLU = 4 + + SiGLU = 5 + + IDENTITY = 6 + + InvalidType = -1 + + +def is_gated(act_fn: ActivationType) -> bool: + """ + Return True if the given activation function is gated. + """ + if not isinstance(act_fn, ActivationType): + act_fn = ActivationType(act_fn) + + return act_fn in [ActivationType.GEGLU, ActivationType.ReGLU, ActivationType.SiGLU] + + +def elem_size(dtype: torch.dtype) -> int: + """ + Return size in bytes of the given dtype. + """ + try: + return ELEM_SIZES[dtype] + except KeyError: + raise ValueError("Unknown dtype size for {}".format(dtype)) + + +def ceil_div(a: int, b: int) -> int: + """ + Return ceil(a / b). + """ + return -(-a // b) diff --git a/deepspeed/inference/v2/kernels/__init__.py b/deepspeed/inference/v2/kernels/__init__.py new file mode 100644 index 000000000000..01b7b0580073 --- /dev/null +++ b/deepspeed/inference/v2/kernels/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .ds_kernel import DSKernelBase diff --git a/deepspeed/inference/v2/kernels/core_ops/__init__.py b/deepspeed/inference/v2/kernels/core_ops/__init__.py new file mode 100644 index 000000000000..bbb53e5b58a2 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/__init__.py @@ -0,0 +1,10 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .bias_activations import * +from .blas_kernels import * +from .cuda_layer_norm import * +from .cuda_rms_norm import * +from .gated_activations import * diff --git a/deepspeed/inference/v2/kernels/core_ops/bias_activations/__init__.py b/deepspeed/inference/v2/kernels/core_ops/bias_activations/__init__.py new file mode 100644 index 000000000000..ea7f8a7d1996 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/bias_activations/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .bias_activation import * diff --git a/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cpp b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cpp new file mode 100644 index 000000000000..4f0cc9cbd77c --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cpp @@ -0,0 +1,68 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "bias_activation.h" +#include +#include "ds_kernel_utils.h" + +#ifdef BF16_AVAILABLE +#define DTYPE_SWITCH(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat16) { \ + using scalar_t = __half; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kBFloat16) { \ + using scalar_t = __nv_bfloat16; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dtype for BiasActivation"); \ + } \ + }() +#else +#define DTYPE_SWITCH(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat16) { \ + using scalar_t = __half; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dtype for BiasActivation"); \ + } \ + }() +#endif + +/* +In-place bias and activation fusion kernel. +*/ +void bias_activation(torch::Tensor& activation, + c10::optional& bias, + const int32_t act_type) +{ + const ActivationType atype = static_cast(act_type); + const int32_t rows = activation.size(0); + const int32_t cols = activation.size(1); + + TORCH_CHECK(atype == ActivationType::GELU || atype == ActivationType::RELU || + atype == ActivationType::SILU || atype == ActivationType::IDENTITY, + "Unsupported activation type for BiasActivation"); + TORCH_CHECK(activation.dim() == 2, "BiasActivation only supports 2D activation tensors"); + + DTYPE_SWITCH(activation.scalar_type(), [&] { + scalar_t* activation_ptr = reinterpret_cast(activation.data_ptr()); + + const scalar_t* bias_ptr; + if (bias.has_value()) { + TORCH_CHECK(activation.scalar_type() == bias.value().scalar_type(), + "BiasActivation activation and bias must have same dtype"); + bias_ptr = reinterpret_cast(bias.value().data_ptr()); + } else { + bias_ptr = nullptr; + } + + if (atype == ActivationType::IDENTITY && bias_ptr == nullptr) { return; } + + launch_bias_activation( + activation_ptr, bias_ptr, rows, cols, atype, c10::cuda::getCurrentCUDAStream()); + }); +} diff --git a/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cu b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cu new file mode 100644 index 000000000000..66bca0c175c3 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.cu @@ -0,0 +1,140 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include +#include "activation_type.h" +#include "conversion_utils.h" +#include "ds_kernel_utils.h" +#include "memory_access_utils.h" + +// Default activation function will error out +template +DS_D_INLINE float act_fn(float val); + +template <> +DS_D_INLINE float act_fn(float val) +{ + return val; +} + +template <> +DS_D_INLINE float act_fn(float val) +{ + return val > 0.0f ? val : 0.0f; +} + +template <> +DS_D_INLINE float act_fn(float val) +{ + constexpr float sqrt_param = 0.79788456080286535587989211986876f; + constexpr float mul_param = 0.044715f; + return val * 0.5f * (1.0f + tanhf(sqrt_param * (val + mul_param * val * val * val))); +} + +template <> +DS_D_INLINE float act_fn(float val) +{ + return val / (1.0f + expf(-val)); +} + +namespace bias_act { + +constexpr int access_size = 16; +constexpr int threads = 512; +constexpr int unroll = 4; + +} // namespace bias_act + +template +__global__ void bias_activation_kernel(T* activation, + const T* bias, + const int32_t rows, + const int32_t cols) +{ + constexpr int vector_T = bias_act::access_size / sizeof(T); + + const int32_t thread_offset = threadIdx.x * vector_T; + const int32_t block_offset = blockIdx.x * vector_T * bias_act::unroll * bias_act::threads; + const int32_t base_offset = block_offset + thread_offset; + + const int32_t thread_stride = bias_act::threads * vector_T; + +#pragma unroll + for (int i = 0; i < bias_act::unroll; i++) { + const int32_t iter_offset = base_offset + i * thread_stride; + + const int32_t row = iter_offset / cols; + + T buffer[vector_T]; + T bias_buffer[vector_T]; + + if (row < rows) { + const int32_t col = iter_offset % cols; + + mem_access::load_global(buffer, activation + iter_offset); + mem_access::load_global( + bias_buffer, bias + col, bias != nullptr); + +#pragma unroll + for (int j = 0; j < vector_T; j++) { + float val = + conversion::to(buffer[j]) + conversion::to(bias_buffer[j]); + buffer[j] = conversion::to(act_fn(val)); + } + + mem_access::store_global(activation + iter_offset, buffer); + } + } +} + +#define ACT_TYPE_SWITCH(ACT_TYPE, ...) \ + if (ACT_TYPE == ActivationType::IDENTITY) { \ + constexpr ActivationType act_fn_t = ActivationType::IDENTITY; \ + return __VA_ARGS__(); \ + } else if (ACT_TYPE == ActivationType::RELU) { \ + constexpr ActivationType act_fn_t = ActivationType::RELU; \ + return __VA_ARGS__(); \ + } else if (ACT_TYPE == ActivationType::GELU) { \ + constexpr ActivationType act_fn_t = ActivationType::GELU; \ + return __VA_ARGS__(); \ + } else if (ACT_TYPE == ActivationType::SILU) { \ + constexpr ActivationType act_fn_t = ActivationType::SILU; \ + return __VA_ARGS__(); \ + } else { \ + assert(false); \ + } + +template +void launch_bias_activation(T* activation, + const T* bias, + const int32_t n_rows, + const int32_t n_cols, + const ActivationType activation_type, + cudaStream_t stream) +{ + constexpr int32_t elems_per_block = + bias_act::threads * bias_act::unroll * bias_act::access_size / sizeof(T); + const int32_t total_elems = n_rows * n_cols; + + const int32_t blocks = (total_elems + elems_per_block - 1) / elems_per_block; + + const dim3 grid(blocks); + const dim3 block(bias_act::threads); + + ACT_TYPE_SWITCH(activation_type, [&] { + bias_activation_kernel + <<>>(activation, bias, n_rows, n_cols); + }); +} + +#define INSTANTIATE_FOR_T(T) \ + template void launch_bias_activation( \ + T*, const T*, const int32_t, const int32_t, const ActivationType, cudaStream_t); + +INSTANTIATE_FOR_T(__half); + +#ifdef BF16_AVAILABLE +INSTANTIATE_FOR_T(__nv_bfloat16); +#endif diff --git a/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.h b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.h new file mode 100644 index 000000000000..db6174633a09 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.h @@ -0,0 +1,22 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "activation_type.h" + +template +void launch_bias_activation(T* activation, + const T* bias, + const int32_t n_rows, + const int32_t n_cols, + const ActivationType activation_type, + cudaStream_t stream); + +void bias_activation(torch::Tensor& activation, + c10::optional& bias, + const int32_t activation_type); diff --git a/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.py b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.py new file mode 100644 index 000000000000..436d7f8805d5 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/bias_activations/bias_activation.py @@ -0,0 +1,62 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from ....inference_utils import ActivationType, DtypeEnum +from deepspeed.ops.op_builder import InferenceCoreBuilder +from ... import DSKernelBase + + +class CUDABiasActivation(DSKernelBase): + """ + CUDA implementation of bias activation kernel. This kernel should be deprecated once + we are fusing the bias activation into the linear kernel in all scenarios. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_act_fns = [ActivationType.IDENTITY, ActivationType.GELU, ActivationType.RELU, ActivationType.SILU] + + def __init__(self, channels: int, dtype: DtypeEnum, act_fn: ActivationType) -> None: + """ + Compile and validate for the fused bias-activation kernel. + + Parameters: + channels (int): Number of channels to expect in the activation. + dtype (torch.dtype): Data type for the input/output. Supported values + are DtypeEnum.fp16 and DtypeEnum.bf16. + act_fn (ActivationType): Activation function to use. Only IDENTITY, GELU, RELU, and SILU are supported. + """ + + if channels % 8 != 0: + raise ValueError("channels must be divisible by 8") + + if DtypeEnum(dtype) not in CUDABiasActivation.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + dtype, CUDABiasActivation.supported_dtypes)) + + act_fn = ActivationType(act_fn) + if act_fn not in CUDABiasActivation.supported_act_fns: + raise ValueError("Unsupported activation function: {}, supported_act_fns are {}".format( + act_fn, CUDABiasActivation.supported_act_fns)) + + inf_module = InferenceCoreBuilder().load() + self.kernel = inf_module.bias_activation + self.act_fn = act_fn + + def __call__(self, activation: torch.Tensor, bias: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Add an optional bias and perform the non-linear activation function. + + Parameters: + activation (torch.Tensor): Input tensor of shape [tokens, channels] + bias (torch.Tensor): Optional bias tensor of shape [channels] + + Returns: + activation that has been updated in-place + """ + self.kernel(activation, bias, self.act_fn.value) diff --git a/deepspeed/inference/v2/kernels/core_ops/blas_kernels/__init__.py b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/__init__.py new file mode 100644 index 000000000000..4af5a579ca1b --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .blas_linear import * diff --git a/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas.h b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas.h new file mode 100644 index 000000000000..1854e40a227d --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas.h @@ -0,0 +1,138 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include +#include "blas_utils.h" + +#define DISPATCH_BLAS_MATMUL(T_TYPE, C_TYPE) \ + if (output.options().dtype() == torch::T_TYPE) { \ + blas_gemm_ex(output.data_ptr(), \ + (const void*)weights.data_ptr(), \ + (const void*)hidden_states.data_ptr(), \ + m, \ + n, \ + k, \ + lda, \ + ldb, \ + ldc, \ + trans_a, \ + trans_b, \ + &alpha, \ + &beta, \ + C_TYPE); \ + } + +void blas_linear(at::Tensor& output, at::Tensor& hidden_states, at::Tensor& weights) +{ + /* + Expected shape: output([total_tokens_across_dims], out_neurons) + hidden_states([total_tokens_across_dims], in_neurons) + weights(out_neurons, in_neurons) + + We are going to assume contiguous for the above shapes. + + The shapes are going to get messed with a little internally to handle column-major + GEMMs. + */ + + // Number of tokens is N (since the GEMM output is column-major but our Tensor + // is row-major, we need to transpose the shapes) + const int n = output.numel() / output.size(-1); + const int k = weights.size(1); + const int m = weights.size(0); + + // A strides + const bool trans_a = weights.stride(1) == 1; + const int lda = (trans_a) ? weights.stride(0) : weights.stride(1); + + // B strides + const bool trans_b = hidden_states.stride(-1) != 1; + const int ldb = (trans_b) ? hidden_states.stride(-1) : hidden_states.stride(-2); + + // C strides + const int ldc = output.stride(-2); + + const float alpha = 1.0f; + const float beta = 0.0f; + + TORCH_CHECK(output.scalar_type() == hidden_states.scalar_type(), + "Output and hidden states must have the same scalar type"); + TORCH_CHECK(output.scalar_type() == weights.scalar_type(), + "Output and weights must have the same scalar type"); + + // Dispatch the datatypes + DISPATCH_BLAS_MATMUL(kFloat, BlasType::FP32); + DISPATCH_BLAS_MATMUL(kHalf, BlasType::FP16); +#ifdef BF16_AVAILABLE + DISPATCH_BLAS_MATMUL(kBFloat16, BlasType::BF16); +#endif +} + +#define DISPATCH_4D_BLAS(T_TYPE, C_TYPE) \ + if (C.options().dtype() == torch::T_TYPE) { \ + blas_strided_batched_gemm(C.data_ptr(), \ + (const void*)A.data_ptr(), \ + (const void*)B.data_ptr(), \ + m, \ + n, \ + k, \ + lda, \ + ldb, \ + ldc, \ + trans_a, \ + trans_b, \ + &alpha, \ + &beta, \ + stride_a, \ + stride_b, \ + stride_c, \ + batch, \ + C_TYPE); \ + } + +void blas_4d_matmul(at::Tensor& C, at::Tensor& B, at::Tensor& A) +{ + /* + C shape: (batch_size, N, M) + A shape: (batch_size, N, K) + B shape: (batch_size, K, M) + */ + + const int n = C.size(-2); + const int k = C.size(-1); + const int m = B.size(-1); + + // A strides + const bool trans_a = A.stride(-1) == 1; + const int lda = (trans_a) ? A.stride(-2) : A.stride(-1); + const int stride_a = A.stride(-3); + + // B strides + const bool trans_b = B.stride(-1) != 1; + const int ldb = (trans_b) ? B.stride(-1) : B.stride(-2); + const int stride_b = B.stride(-3); + + // C strides + const int ldc = C.stride(-2); + const int stride_c = C.stride(-3); + + const float alpha = 1.0f; + const float beta = 0.0f; + + const int batch = C.numel() / (n * m); + + // Dispatch the datatypes + DISPATCH_4D_BLAS(kFloat, BlasType::FP32); + DISPATCH_4D_BLAS(kHalf, BlasType::FP16); +#ifdef BF16_AVAILABLE + DISPATCH_4D_BLAS(kBFloat16, BlasType::BF16); +#endif +} + +void create_handle() { BlasContext::getInstance().get_handle(); } diff --git a/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_linear.py b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_linear.py new file mode 100644 index 000000000000..9a151ce36dc4 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_linear.py @@ -0,0 +1,55 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ....inference_utils import DtypeEnum +from deepspeed.ops.op_builder import InferenceCoreBuilder +from ... import DSKernelBase + + +class BlasLibLinear(DSKernelBase): + """ + Wrapper around the BLAS matmul kernel for FP16/BF16/FP32 for CUDA/RoCM. + + Performs z = x @ y + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16, DtypeEnum.fp32] + + def __init__(self, fp_dtype: DtypeEnum): + """ + Parameters: + fp_dtype (torch.dtype): Data type for the input/output. Supported values + are torch.float16, torch.bfloat16, and torch.float32. + """ + fp_dtype = DtypeEnum(fp_dtype) + if fp_dtype not in BlasLibLinear.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, BlasLibLinear.supported_dtypes)) + + self.inf_module = InferenceCoreBuilder().load() + self.inf_module.create_handle() + self.kernel = self.inf_module.blas_linear + + def __call__(self, output: torch.Tensor, hidden_states: torch.Tensor, weights: torch.Tensor) -> torch.Tensor: + """ + Matmul kernel as implemented by platform BLAS library. The input must be 2D or larger. If + n-dimensional, the leading dimensions are folded into each other: + 2D: m = x.size(0) + 3D: m = x.size(0) * x.size(1) + 4D: m = x.size(0) * x.size(1) * x.size(2) (etc...) + All inputs should be contiguous. + + Parameters: + output (torch.Tensor): Output tensor. Shape is of [*, out_features] + hidden_states (torch.Tensor): Input tensor. Shape is of [*, in_features] + weights (torch.Tensor): Input tensor. Shape is of [out_features, in_features] + + Returns: + z (torch.Tensor): Output tensor. Shape is of [m, n] + """ + self.kernel(output, hidden_states, weights) + return output diff --git a/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_utils.h b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_utils.h new file mode 100644 index 000000000000..450991b3c387 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/blas_kernels/blas_utils.h @@ -0,0 +1,275 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include +#ifdef BF16_AVAILABLE +#include +#endif +#include +#include +#ifndef __HIP_PLATFORM_HCC__ +#include +#endif +#include +#include +#include + +class BlasContext { + /* + Slim wrapper for managing the lifetime of the platform's BLAS handle. This should + be hipified for ROCm. + */ +public: + BlasContext() + { + if (cublasCreate(&_handle) != CUBLAS_STATUS_SUCCESS) { + auto message = std::string("Fail to create cublas handle."); + std::cerr << message << std::endl; + throw std::runtime_error(message); + } +#ifndef __HIP_PLATFORM_HCC__ + cublasSetMathMode(_handle, CUBLAS_TENSOR_OP_MATH); +#endif + } + + virtual ~BlasContext() { cublasDestroy(_handle); } + + static BlasContext& getInstance() + { + // Should always access the singleton through this function. + static BlasContext _instance; + return _instance; + } + + cublasHandle_t get_handle() const { return _handle; } + +private: + cublasHandle_t _handle; +}; + +enum class BlasType { FP32, FP16, BF16 }; + +#ifdef __HIP_PLATFORM_HCC__ +rocblas_operation get_trans_op(bool do_trans) +{ + return (do_trans) ? rocblas_operation_transpose : rocblas_operation_none; +} + +rocblas_datatype get_datatype(BlasType type) +{ + switch (type) { + case BlasType::FP32: return rocblas_datatype_f32_r; + case BlasType::FP16: return rocblas_datatype_f16_r; + case BlasType::BF16: return rocblas_datatype_bf16_r; + default: throw std::runtime_error("Unsupported BlasType"); + } +} +#else +cublasOperation_t get_trans_op(bool do_trans) { return (do_trans) ? CUBLAS_OP_T : CUBLAS_OP_N; } + +cublasDataType_t get_datatype(BlasType type) +{ + switch (type) { + case BlasType::FP32: return CUDA_R_32F; + case BlasType::FP16: return CUDA_R_16F; + case BlasType::BF16: return CUDA_R_16BF; + default: throw std::runtime_error("Unsupported BlasType"); + } +} +#endif + +int blas_gemm_ex(void* C, + const void* A, + const void* B, + int m, + int n, + int k, + int lda, + int ldb, + int ldc, + bool transa, + bool transb, + const float* alpha, + const float* beta, + BlasType type) +{ +#ifdef __HIP_PLATFORM_HCC__ + rocblas_operation_t transa_op = get_trans_op(transa); + rocblas_operation_t transb_op = get_trans_op(transb); + + rocblas_datatype_t abc_type = get_datatype(type); + + rocblas_status status = rocblas_gemm_ex(BlasContext::getInstance().get_handle(), + transa_op, + transb_op, + m, + n, + k, + (const void*)alpha, + A, + abc_type, + lda, + B, + abc_type, + ldb, + (const void*)beta, + C, + abc_type, + ldc, + C, + abc_type, + ldc, + rocblas_datatype_f32_r, + rocblas_gemm_algo_standard, + 0, + 0); +#else + cublasOperation_t transa_op = get_trans_op(transa); + cublasOperation_t transb_op = get_trans_op(transb); + + cublasDataType_t abc_type = get_datatype(type); + cublasStatus_t status = cublasGemmEx(BlasContext::getInstance().get_handle(), + transa_op, + transb_op, + m, + n, + k, + (const void*)alpha, + A, + abc_type, + lda, + B, + abc_type, + ldb, + (const void*)beta, + C, + abc_type, + ldc, + CUDA_R_32F, + CUBLAS_GEMM_DEFAULT_TENSOR_OP); +#endif + +#ifdef __HIP_PLATFORM_HCC__ + if (status != rocblas_status_success) { +#else + if (status != CUBLAS_STATUS_SUCCESS) { +#endif + fprintf(stderr, + "!!!! kernel execution error. (m: %d, n: %d, k: %d, error: %d) \n", + m, + n, + k, + (int)status); + return EXIT_FAILURE; + } + return 0; +} + +int blas_strided_batched_gemm(void* C, + const void* A, + const void* B, + int m, + int n, + int k, + int lda, + int ldb, + int ldc, + bool transa, + bool transb, + const float* alpha, + const float* beta, + int stride_A, + int stride_B, + int stride_C, + int batch, + BlasType type) +{ +#ifdef __HIP_PLATFORM_HCC__ + rocblas_operation_t transa_op = get_trans_op(transa); + rocblas_operation_t transb_op = get_trans_op(transb); + + rocblas_datatype_t abc_type = get_datatype(type); + + rocblas_status status = + rocblas_gemm_strided_batched_ex(BlasContext::getInstance()::get_handle(), + transa_op, + transb_op, + m, + n, + k, + (const void*)alpha, + A, + abc_type, + lda, + stride_A, + B, + abc_type, + ldb, + stride_B, + (const void*)beta, + C, + abc_type, + ldc, + stride_C, + C, + abc_type, + ldc, + stride_C, + batch, + rocblas_datatype_f32_r, + rocblas_gemm_algo_standard, + 0, + 0); +#else + cublasOperation_t transa_op = get_trans_op(transa); + cublasOperation_t transb_op = get_trans_op(transb); + + cublasDataType_t abc_type = get_datatype(type); + + cublasStatus_t status = cublasGemmStridedBatchedEx(BlasContext::getInstance().get_handle(), + transa_op, + transb_op, + m, + n, + k, + (const void*)alpha, + A, + abc_type, + lda, + stride_A, + B, + abc_type, + ldb, + stride_B, + (const void*)beta, + C, + abc_type, + ldc, + stride_C, + batch, + CUDA_R_32F, + CUBLAS_GEMM_DEFAULT_TENSOR_OP); +#endif + +#ifdef __HIP_PLATFORM_HCC__ + if (status != rocblas_status_success) { +#else + if (status != CUBLAS_STATUS_SUCCESS) { +#endif + fprintf(stderr, + "!!!! kernel execution error. (batch: %d, m: %d, n: %d, k: %d, error: %d) \n", + batch, + m, + n, + k, + (int)status); + return EXIT_FAILURE; + } + return 0; +} diff --git a/deepspeed/inference/v2/kernels/core_ops/core_ops.cpp b/deepspeed/inference/v2/kernels/core_ops/core_ops.cpp new file mode 100644 index 000000000000..58df88e56136 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/core_ops.cpp @@ -0,0 +1,36 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include +#include + +#include "bias_activation.h" +#include "blas.h" +#include "gated_activation_kernels.h" +#include "layer_norm.h" +#include "rms_norm.h" + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) +{ + // bias_activation.h + m.def("bias_activation", &bias_activation, "DeepSpeed bias activation in CUDA"); + + // layer_norm.h + m.def("layer_norm", &ds_layer_norm, "DeepSpeed layer norm in CUDA"); + m.def("pre_layer_norm", &ds_pre_layer_norm, "DeepSpeed pre layer norm in CUDA"); + m.def("post_layer_norm", &ds_post_layer_norm, "DeepSpeed pre layer norm in CUDA"); + + // blas.h + m.def("blas_linear", &blas_linear, "Linear implemented by vendor BLAS"); + m.def("blas_4d_matmul", &blas_4d_matmul, "4D matmul implemented by vendor BLAS"); + m.def("create_handle", &create_handle, "Create a handle for vendor BLAS"); + + // gated_activation_kernels.h + m.def("gated_activation", &ds_gated_activation, "DeepSpeed gated activation in CUDA"); + + // rms_norm.h + m.def("rms_norm", &rms_norm, "DeepSpeed rms norm in CUDA"); + m.def("rms_pre_norm", &rms_pre_norm, "DeepSpeed rms pre norm in CUDA"); +} diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/__init__.py b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/__init__.py new file mode 100644 index 000000000000..bed7688b15d2 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/__init__.py @@ -0,0 +1,8 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .cuda_ln import * +from .cuda_post_ln import * +from .cuda_pre_ln import * diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_fp_ln_base.py b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_fp_ln_base.py new file mode 100644 index 000000000000..3c2aa5cb5eb4 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_fp_ln_base.py @@ -0,0 +1,37 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from ....inference_utils import elem_size +from deepspeed.ops.op_builder import InferenceCoreBuilder + + +class CUDAFPLNBase(DSKernelBase): + """ + Base class for CUDA LN kernels. They all same the same validation logic, + so we can share it here. + """ + + supported_dtypes = [torch.float16, torch.bfloat16, torch.float32] + + def __init__(self, channels: int, fp_dtype: torch.dtype, epsilon: float = 1e-5): + """ + Parameters: + channels (int): Number of channels in the input tensor. Must be divisible to align + to 16 bytes. + fp_dtype (torch.dtype): Data type for the input/output/gamma. Supported values + are torch.float16, torch.bfloat16, and torch.float32. + """ + if fp_dtype not in CUDAFPLNBase.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, CUDAFPLNBase.supported_dtypes)) + + if elem_size(fp_dtype) * channels % 16 != 0: + raise ValueError("channels must be divisible by 16 bytes") + + self.inf_module = InferenceCoreBuilder().load() + self.epsilon = epsilon diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_ln.py b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_ln.py new file mode 100644 index 000000000000..583736fb8bbc --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_ln.py @@ -0,0 +1,30 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from .cuda_fp_ln_base import CUDAFPLNBase + + +class CUDAFPLN(CUDAFPLNBase): + """ + Floating point layer norm kernel for CUDA/RoCM. + + Performs: z = ln(x) + """ + + def __call__(self, output_z: torch.Tensor, input_x: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor) -> torch.Tensor: + """ + output_z may alias input_x directly. All Tensors should have the same shape. + + Parameters: + output_z (torch.Tensor): Output tensor. + input_x (torch.Tensor): Input tensor. + gamma (torch.Tensor): Gamma tensor. + beta (torch.Tensor): Beta tensor. + """ + self.inf_module.layer_norm(output_z, input_x, gamma, beta, self.epsilon) + return output_z diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_post_ln.py b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_post_ln.py new file mode 100644 index 000000000000..0ced1ecf207e --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_post_ln.py @@ -0,0 +1,34 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from .cuda_fp_ln_base import CUDAFPLNBase + + +class CUDAFPPostLN(CUDAFPLNBase): + """ + Floating point post-LayerNorm kernel for CUDA/RoCM. + + Performs: z = ln(x + y) + """ + + def __call__(self, output_z: torch.Tensor, input_x: torch.Tensor, input_y: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor) -> torch.Tensor: + """ + Either input_x or input_y can alias output_z. + + Parameters: + output_z (torch.Tensor): Output tensor. + input_x (torch.Tensor): Input tensor. + input_y (torch.Tensor): Input tensor. + gamma (torch.Tensor): Gamma tensor. + beta (torch.Tensor): Beta tensor. + + Returns: + output (torch.Tensor): Output tensor. + """ + self.inf_module.post_layer_norm(output_z, input_x, input_y, gamma, beta, self.epsilon) + return output_z diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_pre_ln.py b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_pre_ln.py new file mode 100644 index 000000000000..74b2d9cf5880 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/cuda_pre_ln.py @@ -0,0 +1,39 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Tuple + +import torch + +from .cuda_fp_ln_base import CUDAFPLNBase + + +class CUDAFPPreLN(CUDAFPLNBase): + """ + Floating point pre-LayerNorm kernel for CUDA/RoCM. + + Performs: z_res = x_res + y_hid + z_hid = ln(z_hid) + """ + + def __call__(self, z_res: torch.Tensor, z_hid: torch.Tensor, x_res: torch.Tensor, y_hid: torch.Tensor, + gamma: torch.Tensor, beta: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]: + """ + z_res can alias x_res. All non-parameter input/output tensors + must have the same shape. z_hid can alias y_hid. + + Parameters: + z_res (torch.Tensor): Output residual. + z_hid (torch.Tensor): Output hidden states. + x_res (torch.Tensor): Input residual. + y_hid (torch.Tensor): Input hidden states. + gamma (torch.Tensor): Gamma tensor. + beta (torch.Tensor): Beta tensor. + + Returns: + output (torch.Tensor): Output tensor. + """ + self.inf_module.pre_layer_norm(z_res, z_hid, x_res, y_hid, gamma, beta, self.epsilon) + return z_res, z_hid diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cpp b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cpp new file mode 100644 index 000000000000..b2c95d410a1f --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cpp @@ -0,0 +1,102 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "layer_norm.h" + +#define DISPATCH_LAYER_NORM(T_TYPE, C_TYPE) \ + if (input.options().dtype() == torch::T_TYPE) { \ + launch_fused_ln((C_TYPE*)output.data_ptr(), \ + (const C_TYPE*)input.data_ptr(), \ + (const C_TYPE*)gamma.data_ptr(), \ + (const C_TYPE*)beta.data_ptr(), \ + epsilon, \ + rows, \ + elems_per_row, \ + at::cuda::getCurrentCUDAStream()); \ + } + +void ds_layer_norm(at::Tensor& output, + at::Tensor& input, + at::Tensor& gamma, + at::Tensor& beta, + float epsilon) +{ + bool ragged_input = input.dim() == 2; + + const int rows = ragged_input ? input.size(0) : input.size(0) * input.size(1); + const int elems_per_row = ragged_input ? input.size(1) : input.size(2); + + DISPATCH_LAYER_NORM(kFloat, float); + DISPATCH_LAYER_NORM(kHalf, __half); +#ifdef BF16_AVAILABLE + DISPATCH_LAYER_NORM(kBFloat16, __nv_bfloat16); +#endif +} + +#define DISPATCH_LAYER_NORM_RESIDUAL(T_TYPE, C_TYPE) \ + if (input.options().dtype() == torch::T_TYPE) { \ + launch_fused_post_ln((C_TYPE*)output.data_ptr(), \ + (const C_TYPE*)input.data_ptr(), \ + (const C_TYPE*)residual.data_ptr(), \ + (const C_TYPE*)gamma.data_ptr(), \ + (const C_TYPE*)beta.data_ptr(), \ + epsilon, \ + rows, \ + elems_per_row, \ + at::cuda::getCurrentCUDAStream()); \ + } + +void ds_post_layer_norm(at::Tensor& output, + at::Tensor& input, + at::Tensor& residual, + at::Tensor& gamma, + at::Tensor& beta, + float epsilon) +{ + bool ragged_input = input.dim() == 2; + + const int rows = ragged_input ? input.size(0) : input.size(0) * input.size(1); + const int elems_per_row = ragged_input ? input.size(1) : input.size(2); + + DISPATCH_LAYER_NORM_RESIDUAL(kFloat, float); + DISPATCH_LAYER_NORM_RESIDUAL(kHalf, __half); +#ifdef BF16_AVAILABLE + DISPATCH_LAYER_NORM_RESIDUAL(kBFloat16, __nv_bfloat16); +#endif +} + +#define DISPATCH_PRE_LAYER_NORM_RESIDUAL(T_TYPE, C_TYPE) \ + if (input.options().dtype() == torch::T_TYPE) { \ + launch_fused_pre_ln((C_TYPE*)norm_output.data_ptr(), \ + (C_TYPE*)res_output.data_ptr(), \ + (const C_TYPE*)input.data_ptr(), \ + (const C_TYPE*)residual.data_ptr(), \ + (const C_TYPE*)gamma.data_ptr(), \ + (const C_TYPE*)beta.data_ptr(), \ + epsilon, \ + rows, \ + elems_per_row, \ + at::cuda::getCurrentCUDAStream()); \ + } + +void ds_pre_layer_norm(at::Tensor& res_output, + at::Tensor& norm_output, + at::Tensor& input, + at::Tensor& residual, + at::Tensor& gamma, + at::Tensor& beta, + float epsilon) +{ + bool ragged_input = input.dim() == 2; + + const int rows = ragged_input ? input.size(0) : input.size(0) * input.size(1); + const int elems_per_row = ragged_input ? input.size(1) : input.size(2); + + DISPATCH_PRE_LAYER_NORM_RESIDUAL(kFloat, float); + DISPATCH_PRE_LAYER_NORM_RESIDUAL(kHalf, __half); +#ifdef BF16_AVAILABLE + DISPATCH_PRE_LAYER_NORM_RESIDUAL(kBFloat16, __nv_bfloat16); +#endif +} diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cu b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cu new file mode 100644 index 000000000000..15f52c46622b --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cu @@ -0,0 +1,490 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "conversion_utils.h" +#include "ds_kernel_utils.h" +#include "memory_access_utils.h" +#include "reduction_utils.h" + +namespace cg = cooperative_groups; +using rop = reduce::ROpType; + +namespace ln { +constexpr int granularity = 16; +} // namespace ln + +/* +Regular layer norm implementation. Assumes elems_per_row % 8 +is equal to 0. + +Args: + output: buffer for output data + vals: buffer for input data + gamma: gain for normalization + beta: bias for normalization + epsilon: numeric stability + elems_per_row: number of elements each block will normalize +*/ +template +__global__ void fused_ln(T* output, + const T* vals, + const T* gamma, + const T* beta, + float epsilon, + int elems_per_row) +{ + constexpr int T_per_load = ln::granularity / sizeof(T); + + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + + // X-dimension of the block + const int block_offset = (tb.group_index().x * (maxThreads / threadsPerGroup) * elems_per_row) + + (tb.thread_index().y * elems_per_row); + const int thread_offset = tb.thread_index().x * T_per_load; + const int base_offset = block_offset + thread_offset; + const int stride = blockDim.x * T_per_load; + + float sum = reduce::init(); + + const T* input_base = vals + base_offset; + + T local_buffer[unRoll * T_per_load]; + +#pragma unRoll + for (int i = 0; i < unRoll; i++) { + T* iteration_buffer = local_buffer + i * T_per_load; + + mem_access::load_global( + iteration_buffer, input_base + i * stride, thread_offset + i * stride < elems_per_row); + +#pragma unRoll + for (int j = 0; j < T_per_load; j++) { + float vals_up_cast = conversion::to(iteration_buffer[j]); + sum = reduce::element(sum, vals_up_cast); + } + } + + reduce::partitioned_block(tb, warp, sum); + const float mean = sum / elems_per_row; + + float mean_diff = reduce::init(); + +#pragma unRoll + for (int i = 0; i < unRoll; i++) { +#pragma unRoll + for (int j = 0; j < T_per_load; j++) { + // Using a 0 value here skews the variance, have to if-guard + if (thread_offset + i * stride < elems_per_row) { + float diff = (conversion::to(local_buffer[i * T_per_load + j]) - mean); + mean_diff = reduce::element(mean_diff, diff * diff); + } + } + } + + reduce::partitioned_block(tb, warp, mean_diff); + const float variance = mean_diff / elems_per_row; + const float denom = __frsqrt_rn(variance + epsilon); + + T* block_output = output + block_offset; + +#pragma unRoll + for (int i = 0; i < unRoll; i++) { + T* iteration_buffer = local_buffer + i * T_per_load; + const int iter_idx = i * stride + thread_offset; + const bool do_loads = iter_idx < elems_per_row; + + T gamma_local[T_per_load], beta_local[T_per_load]; + + mem_access::load_global(gamma_local, gamma + iter_idx, do_loads); + mem_access::load_global(beta_local, beta + iter_idx, do_loads); + +#pragma unRoll + for (int j = 0; j < T_per_load; j++) { + float val = conversion::to(iteration_buffer[j]); + val = (val - mean) * denom; + val = + val * conversion::to(gamma_local[j]) + conversion::to(beta_local[j]); + iteration_buffer[j] = conversion::to(val); + } + + if (do_loads) { + mem_access::store_global(block_output + iter_idx, iteration_buffer); + } + } +} + +#define LAUNCH_FUSED_LN(unRollFactor, threadsPerGroup, maxThreads) \ + fused_ln \ + <<>>(output, vals, gamma, beta, epsilon, elems_per_row); + +template +void launch_fused_ln(T* output, + const T* vals, + const T* gamma, + const T* beta, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream) +{ + // 8 for __half, 4 for float + constexpr int T_per_load = ln::granularity / sizeof(T); + + constexpr int maxThreads = 256; + + // For Flaoat, unRoll 4, for __half, unRoll 2 + constexpr int internal_unRoll = sizeof(T) == 4 ? 4 : 2; + + const bool is_subblock_schedule = (elems_per_row <= 128) ? true : false; + const int h_per_step = is_subblock_schedule ? T_per_load : T_per_load * internal_unRoll; + + // Scheduling concern: may be slightly faster for some inputs to assign multiple stages of + // warp-sized blocks rather than stepping up to 64/96 threads + const int one_step_threads = next_pow2((elems_per_row + h_per_step - 1) / h_per_step); + const int threadsPerGroup = (one_step_threads < maxThreads) ? one_step_threads : maxThreads; + + const int groups_per_block_max = + is_subblock_schedule ? (maxThreads + threadsPerGroup - 1) / threadsPerGroup : 1; + const int groups_per_block = (rows < groups_per_block_max) ? rows : groups_per_block_max; + const int groups_launch = (groups_per_block + rows - 1) / groups_per_block; + + dim3 block(threadsPerGroup, groups_per_block); + dim3 grid(groups_launch); + + const int elems_per_step = threadsPerGroup * h_per_step; + const int external_unRoll = (elems_per_row + elems_per_step - 1) / elems_per_step; + + if (is_subblock_schedule) { + // <=128 + if (threadsPerGroup == 1) { + LAUNCH_FUSED_LN(1, 1, maxThreads); + } else if (threadsPerGroup == 2) { + LAUNCH_FUSED_LN(1, 2, maxThreads); + } else if (threadsPerGroup == 4) { + LAUNCH_FUSED_LN(1, 4, maxThreads); + } else if (threadsPerGroup == 8) { + LAUNCH_FUSED_LN(1, 8, maxThreads); + } else if (threadsPerGroup == 16) { + LAUNCH_FUSED_LN(1, 16, maxThreads); + } + } else if (external_unRoll == 1) { + // 129 - 4096 elems + // (this can launch with 1-7 warps as well) + LAUNCH_FUSED_LN(1 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 2) { + // 4097 - 8192 elems + LAUNCH_FUSED_LN(2 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 3) { + // 8193 - 12288 elems + LAUNCH_FUSED_LN(3 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 4) { + // 12289 - 16384 elems + LAUNCH_FUSED_LN(4 * internal_unRoll, maxThreads, maxThreads); + } +} + +#define INSTANTIATE_FUSED_LN(T) \ + template void launch_fused_ln(T*, const T*, const T*, const T*, float, int, int, cudaStream_t); + +INSTANTIATE_FUSED_LN(__half); +#ifdef BF16_AVAILABLE +INSTANTIATE_FUSED_LN(__nv_bfloat16); +#endif +INSTANTIATE_FUSED_LN(float); + +/* +Fused resiual + bias + layer norm implementation. Assumes elems_per_row % 8 +is equal to 0. + +TODO(cmikeh2): Goal is to deprecate this implementation. The bias + residual +need to be fused into compute-bound producer operations. + +Args: + output: buffer for output data + res_output: output of residual addition + vals: buffer for input data + residual: residual data + bias: bias of of input data + gamma: gain for normalization + beta: bias for normalization + epsilon: numeric stability + elems_per_row: number of elements each block will normalize +Template arg: + StoreResidual: controls whether the residual calculation is stored + or not. When set to false, the input `res_output` is unused. +*/ +template +__global__ void fused_residual_ln(T* output, + T* res_output, + const T* vals, + const T* residual, + const T* gamma, + const T* beta, + float epsilon, + int elems_per_row) +{ + constexpr int T_per_load = ln::granularity / sizeof(T); + + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + + // X-dimension of the block + const int block_offset = (tb.group_index().x * (maxThreads / threadsPerGroup) * elems_per_row) + + (tb.thread_index().y * elems_per_row); + const int thread_offset = tb.thread_index().x * T_per_load; + const int base_offset = block_offset + thread_offset; + const int stride = tb.size() * T_per_load; + + float sum = reduce::init(); + + const T* input_base = vals + base_offset; + const T* residual_base = residual + base_offset; + + T local_buffer[unRoll * T_per_load]; + + // Unlike a vanilla layernorm, since we're fusing the two adds as well + // an inner unRoll seems to be less valuable. If anything, a double unRoll + // makes the most sense if we find we are having performance issues. +#pragma unRoll + for (int i = 0; i < unRoll; i++) { + T* iteration_buffer = local_buffer + i * T_per_load; + T residual_buffer[T_per_load]; + T bias_buffer[T_per_load]; + + mem_access::load_global( + iteration_buffer, input_base + i * stride, thread_offset + i * stride < elems_per_row); + mem_access::load_global(residual_buffer, + residual_base + i * stride, + thread_offset + i * stride < elems_per_row); + +#pragma unRoll + for (int j = 0; j < T_per_load; j++) { + float vals_up_cast = conversion::to(iteration_buffer[j]); + float res_up_cast = conversion::to(residual_buffer[j]); + vals_up_cast += res_up_cast; + sum = reduce::element(sum, vals_up_cast); + iteration_buffer[j] = conversion::to(vals_up_cast); + } + + if (preLnResidual && (thread_offset + i * stride < elems_per_row)) { + mem_access::store_global(res_output + base_offset + i * stride, + iteration_buffer); + } + } + + reduce::partitioned_block(tb, warp, sum); + const float mean = sum / elems_per_row; + + float mean_diff = reduce::init(); +#pragma unRoll + for (int i = 0; i < unRoll; i++) { +#pragma unRoll + for (int j = 0; j < T_per_load; j++) { + // Using a 0 value here skews the variance, have to if-guard + if (thread_offset + i * stride < elems_per_row) { + float diff = (conversion::to(local_buffer[i * T_per_load + j]) - mean); + mean_diff = reduce::element(mean_diff, diff * diff); + } + } + } + + reduce::partitioned_block(tb, warp, mean_diff); + const float variance = mean_diff / elems_per_row; + const float denom = __frsqrt_rn(variance + epsilon); + + T* block_output = output + block_offset; + +#pragma unRoll + for (int i = 0; i < unRoll; i++) { + T* iteration_buffer = local_buffer + i * T_per_load; + const int iter_idx = i * stride + thread_offset; + const bool do_loads = iter_idx < elems_per_row; + + T gamma_local[T_per_load], beta_local[T_per_load]; + + mem_access::load_global(gamma_local, gamma + iter_idx, do_loads); + mem_access::load_global(beta_local, beta + iter_idx, do_loads); + +#pragma unRoll + for (int j = 0; j < T_per_load; j++) { + float val = conversion::to(iteration_buffer[j]); + val = (val - mean) * denom; + val = + val * conversion::to(gamma_local[j]) + conversion::to(beta_local[j]); + iteration_buffer[j] = conversion::to(val); + } + + if (do_loads) { + mem_access::store_global(block_output + iter_idx, iteration_buffer); + } + } +} + +// TODO(cmikeh2): There's a bunch of redundancy here that needs to be removed/simplified. +#define LAUNCH_FUSED_RES_LN(unRollFactor, threadsPerGroup, maxThreads) \ + fused_residual_ln \ + <<>>( \ + output, nullptr, vals, residual, gamma, beta, epsilon, elems_per_row); + +template +void launch_fused_post_ln(T* output, + const T* vals, + const T* residual, + const T* gamma, + const T* beta, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream) +{ + // 8 for __half, 4 for float + constexpr int T_per_load = ln::granularity / sizeof(T); + + constexpr int maxThreads = 256; + + // For Flaoat, unRoll 4, for __half, unRoll 2 + constexpr int internal_unRoll = sizeof(T) == 4 ? 4 : 2; + + const bool is_subblock_schedule = (elems_per_row <= 128) ? true : false; + const int h_per_step = is_subblock_schedule ? T_per_load : T_per_load * internal_unRoll; + + // Scheduling concern: may be slightly faster for some inputs to assign multiple stages of + // warp-sized blocks rather than stepping up to 64/96 threads + const int one_step_threads = next_pow2((elems_per_row + h_per_step - 1) / h_per_step); + const int threadsPerGroup = (one_step_threads < maxThreads) ? one_step_threads : maxThreads; + + const int groups_per_block_max = + is_subblock_schedule ? (maxThreads + threadsPerGroup - 1) / threadsPerGroup : 1; + const int groups_per_block = (rows < groups_per_block_max) ? rows : groups_per_block_max; + const int groups_launch = (groups_per_block + rows - 1) / groups_per_block; + + dim3 block(threadsPerGroup, groups_per_block); + dim3 grid(groups_launch); + + const int elems_per_step = threadsPerGroup * h_per_step; + const int external_unRoll = (elems_per_row + elems_per_step - 1) / elems_per_step; + + if (is_subblock_schedule) { + // <=128 + if (threadsPerGroup == 1) { + LAUNCH_FUSED_RES_LN(1, 1, maxThreads); + } else if (threadsPerGroup == 2) { + LAUNCH_FUSED_RES_LN(1, 2, maxThreads); + } else if (threadsPerGroup == 4) { + LAUNCH_FUSED_RES_LN(1, 4, maxThreads); + } else if (threadsPerGroup == 8) { + LAUNCH_FUSED_RES_LN(1, 8, maxThreads); + } else if (threadsPerGroup == 16) { + LAUNCH_FUSED_RES_LN(1, 16, maxThreads); + } + } else if (external_unRoll == 1) { + // 129 - 4096 elems + // (this can launch with 1-7 warps as well) + LAUNCH_FUSED_RES_LN(1 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 2) { + // 4097 - 8192 elems + LAUNCH_FUSED_RES_LN(2 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 3) { + // 8193 - 12288 elems + LAUNCH_FUSED_RES_LN(3 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 4) { + // 12289 - 16384 elems + LAUNCH_FUSED_RES_LN(4 * internal_unRoll, maxThreads, maxThreads); + } +} + +#define LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(unRollFactor, threadsPerGroup, maxThreads) \ + fused_residual_ln \ + <<>>( \ + norm_output, res_output, vals, residual, gamma, beta, epsilon, elems_per_row); + +template +void launch_fused_pre_ln(T* norm_output, + T* res_output, + const T* vals, + const T* residual, + const T* gamma, + const T* beta, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream) +{ + // 8 for __half, 4 for float + constexpr int T_per_load = ln::granularity / sizeof(T); + + constexpr int maxThreads = 256; + + // For Flaoat, unRoll 4, for __half, unRoll 2 + constexpr int internal_unRoll = sizeof(T) == 4 ? 4 : 2; + + const bool is_subblock_schedule = (elems_per_row <= 128) ? true : false; + const int h_per_step = is_subblock_schedule ? T_per_load : T_per_load * internal_unRoll; + + // Scheduling concern: may be slightly faster for some inputs to assign multiple stages of + // warp-sized blocks rather than stepping up to 64/96 threads + const int one_step_threads = next_pow2((elems_per_row + h_per_step - 1) / h_per_step); + const int threadsPerGroup = (one_step_threads < maxThreads) ? one_step_threads : maxThreads; + + const int groups_per_block_max = + is_subblock_schedule ? (maxThreads + threadsPerGroup - 1) / threadsPerGroup : 1; + const int groups_per_block = (rows < groups_per_block_max) ? rows : groups_per_block_max; + const int groups_launch = (groups_per_block + rows - 1) / groups_per_block; + + dim3 block(threadsPerGroup, groups_per_block); + dim3 grid(groups_launch); + + const int elems_per_step = threadsPerGroup * h_per_step; + const int external_unRoll = (elems_per_row + elems_per_step - 1) / elems_per_step; + + if (is_subblock_schedule) { + // <=128 + if (threadsPerGroup == 1) { + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(1, 1, maxThreads); + } else if (threadsPerGroup == 2) { + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(1, 2, maxThreads); + } else if (threadsPerGroup == 4) { + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(1, 4, maxThreads); + } else if (threadsPerGroup == 8) { + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(1, 8, maxThreads); + } else if (threadsPerGroup == 16) { + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(1, 16, maxThreads); + } + } else if (external_unRoll == 1) { + // 129 - 4096 elems + // (this can launch with 1-7 warps as well) + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(1 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 2) { + // 4097 - 8192 elems + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(2 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 3) { + // 8193 - 12288 elems + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(3 * internal_unRoll, maxThreads, maxThreads); + } else if (external_unRoll == 4) { + // 12289 - 16384 elems + LAUNCH_FUSED_RES_LN_STORE_PRE_LN_RES(4 * internal_unRoll, maxThreads, maxThreads); + } +} + +#define INSTANTIATE_RES_LN(T) \ + template void launch_fused_post_ln( \ + T*, const T*, const T*, const T*, const T*, float, int, int, cudaStream_t); + +#define INSTANTIATE_PRE_LN_RES(T) \ + template void launch_fused_pre_ln( \ + T*, T*, const T*, const T*, const T*, const T*, float, int, int, cudaStream_t); + +INSTANTIATE_RES_LN(__half); +INSTANTIATE_RES_LN(float); +#ifdef BF16_AVAILABLE +INSTANTIATE_RES_LN(__nv_bfloat16); +#endif + +INSTANTIATE_PRE_LN_RES(__half); +INSTANTIATE_PRE_LN_RES(float); +#ifdef BF16_AVAILABLE +INSTANTIATE_PRE_LN_RES(__nv_bfloat16); +#endif diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.h b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.h new file mode 100644 index 000000000000..9ea3a8c42524 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.h @@ -0,0 +1,67 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "ds_kernel_utils.h" + +/* +Kernel launch methods for layer norm variants. +*/ + +template +void launch_fused_ln(T* output, + const T* vals, + const T* gamma, + const T* beta, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream); + +template +void launch_fused_post_ln(T* output, + const T* vals, + const T* residual, + const T* gamma, + const T* beta, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream); +template +void launch_fused_pre_ln(T* norm_output, + T* res_output, + const T* vals, + const T* residual, + const T* gamma, + const T* beta, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream); + +void ds_layer_norm(at::Tensor& output, + at::Tensor& input, + at::Tensor& gamma, + at::Tensor& beta, + float epsilon); + +void ds_post_layer_norm(at::Tensor& output, + at::Tensor& input, + at::Tensor& residual, + at::Tensor& gamma, + at::Tensor& beta, + float epsilon); + +void ds_pre_layer_norm(at::Tensor& res_output, + at::Tensor& norm_output, + at::Tensor& input, + at::Tensor& residual, + at::Tensor& gamma, + at::Tensor& beta, + float epsilon); diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/__init__.py b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/__init__.py new file mode 100644 index 000000000000..640a72307650 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/__init__.py @@ -0,0 +1,7 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .rms_norm import CUDARMSNorm +from .rms_pre_norm import CUDARMSPreNorm diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cpp b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cpp new file mode 100644 index 000000000000..c67712df438a --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cpp @@ -0,0 +1,123 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "rms_norm.h" + +#ifdef BF16_AVAILABLE +#define DISPATCH_FOR_FLOAT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat32) { \ + using scalar_t = float; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kFloat16) { \ + using scalar_t = __half; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kBFloat16) { \ + using scalar_t = __nv_bfloat16; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dtype for BiasActivation"); \ + } \ + }() +#else +#define DISPATCH_FOR_FLOAT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat32) { \ + using scalar_t = float; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kFloat16) { \ + using scalar_t = __half; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dtype for BiasActivation"); \ + } \ + }() +#endif + +void rms_norm(torch::Tensor& norm_output, + torch::Tensor& norm_input, + torch::Tensor& gamma, + float epsilon) +{ + TORCH_CHECK(norm_output.scalar_type() == norm_input.scalar_type(), + "norm_output and norm_input should have the same data type"); + TORCH_CHECK(norm_output.scalar_type() == gamma.scalar_type(), + "norm_output and gamma should have the same data type"); + + const int32_t rows = norm_input.size(0); + const int32_t cols = norm_input.size(1); + + TORCH_CHECK(norm_output.size(0) == rows, + "norm_output and norm_input should have the same first dimension"); + TORCH_CHECK(norm_output.size(1) == cols, + "norm_output and norm_input should have the same second dimension"); + + DISPATCH_FOR_FLOAT(norm_output.scalar_type(), [&] { + scalar_t* norm_output_ptr = reinterpret_cast(norm_output.data_ptr()); + scalar_t* norm_input_ptr = reinterpret_cast(norm_input.data_ptr()); + scalar_t* gamma_ptr = reinterpret_cast(gamma.data_ptr()); + scalar_t* null_t = nullptr; + + launch_rms_norm(norm_output_ptr, + null_t, + norm_input_ptr, + null_t, + gamma_ptr, + epsilon, + rows, + cols, + at::cuda::getCurrentCUDAStream()); + }); +} + +void rms_pre_norm(torch::Tensor& norm_output, + torch::Tensor& residual_output, + torch::Tensor& norm_input, + torch::Tensor& residual_input, + torch::Tensor& gamma, + float epsilon) +{ + TORCH_CHECK(norm_output.scalar_type() == norm_input.scalar_type(), + "norm_output and norm_input should have the same data type"); + TORCH_CHECK(norm_output.scalar_type() == gamma.scalar_type(), + "norm_output and gamma should have the same data type"); + + const int32_t rows = norm_input.size(0); + const int32_t cols = norm_input.size(1); + + TORCH_CHECK(norm_output.size(0) == rows, + "norm_output and norm_input should have the same first dimension"); + TORCH_CHECK(norm_output.size(1) == cols, + "norm_output and norm_input should have the same second dimension"); + + TORCH_CHECK(residual_output.size(0) == rows, + "residual_output and norm_input should have the same first dimension"); + TORCH_CHECK(residual_output.size(1) == cols, + "residual_output and norm_input should have the same second dimension"); + + TORCH_CHECK(residual_input.size(0) == rows, + "residual_input and norm_input should have the same first dimension"); + TORCH_CHECK(residual_input.size(1) == cols, + "residual_input and norm_input should have the same second dimension"); + + DISPATCH_FOR_FLOAT(norm_output.scalar_type(), [&] { + scalar_t* norm_output_ptr = reinterpret_cast(norm_output.data_ptr()); + scalar_t* residual_output_ptr = reinterpret_cast(residual_output.data_ptr()); + const scalar_t* norm_input_ptr = reinterpret_cast(norm_input.data_ptr()); + const scalar_t* residual_input_ptr = + reinterpret_cast(residual_input.data_ptr()); + const scalar_t* gamma_ptr = reinterpret_cast(gamma.data_ptr()); + + launch_rms_norm(norm_output_ptr, + residual_output_ptr, + norm_input_ptr, + residual_input_ptr, + gamma_ptr, + epsilon, + rows, + cols, + at::cuda::getCurrentCUDAStream()); + }); +} diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cu b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cu new file mode 100644 index 000000000000..e69d3c36cc00 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cu @@ -0,0 +1,262 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "conversion_utils.h" +#include "ds_kernel_utils.h" +#include "memory_access_utils.h" +#include "reduction_utils.h" + +namespace cg = cooperative_groups; +using rop = reduce::ROpType; + +namespace rms { +constexpr int granularity = 16; +} // namespace rms + +template +__global__ void rms_norm(T* output, const T* vals, const T* gamma, float epsilon, int elems_per_row) +{ + constexpr int T_per_load = rms::granularity / sizeof(T); + + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + + // X-dimension of the block + const int block_offset = (tb.group_index().x * (maxThreads / threadsPerGroup) * elems_per_row) + + (tb.thread_index().y * elems_per_row); + const int thread_offset = tb.thread_index().x * T_per_load; + const int base_offset = block_offset + thread_offset; + const int stride = blockDim.x * T_per_load; + + float var_sum = reduce::init(); + + const T* input_base = vals + base_offset; + + T local_buffer[UNROLL * T_per_load]; + +#pragma unroll + for (int i = 0; i < UNROLL; i++) { + T* iteration_buffer = local_buffer + (i * T_per_load); + + mem_access::load_global(iteration_buffer, + input_base + (i * stride), + thread_offset + (i * stride) < elems_per_row); + +#pragma unroll + for (int j = 0; j < T_per_load; j++) { + float up_cast = conversion::to(iteration_buffer[j]); + float sq_val = up_cast * up_cast; + var_sum = reduce::element(var_sum, sq_val); + } + } + + reduce::partitioned_block(tb, warp, var_sum); + const float var = var_sum / elems_per_row; + const T denom = conversion::to(__frsqrt_rn(var + epsilon)); + + T* block_output = output + block_offset; + +#pragma unroll + for (int i = 0; i < UNROLL; i++) { + T* iteration_buffer = local_buffer + (i * T_per_load); + const int iter_idx = i * stride + thread_offset; + const bool do_loads = (iter_idx < elems_per_row); + + T gamma_local[T_per_load]; + + mem_access::load_global(gamma_local, gamma + iter_idx, do_loads); + +#pragma unroll + for (int j = 0; j < T_per_load; j++) { + iteration_buffer[j] *= denom; + iteration_buffer[j] *= gamma_local[j]; + } + + if (do_loads) { + mem_access::store_global(block_output + iter_idx, iteration_buffer); + } + } +} + +template +__global__ void pre_rms_norm(T* output, + T* res_out, + const T* vals, + const T* residual, + const T* gamma, + float epsilon, + int elems_per_row) +{ + constexpr int T_per_load = rms::granularity / sizeof(T); + + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + + // X-dimension of the block + const int block_offset = (tb.group_index().x * (maxThreads / threadsPerGroup) * elems_per_row) + + (tb.thread_index().y * elems_per_row); + const int thread_offset = tb.thread_index().x * T_per_load; + const int base_offset = block_offset + thread_offset; + const int stride = blockDim.x * T_per_load; + + float var_sum = reduce::init(); + + const T* input_base = vals + base_offset; + const T* residual_base = residual + base_offset; + T* res_output = res_out + base_offset; + + T local_buffer[UNROLL * T_per_load]; + +#pragma unroll + for (int i = 0; i < UNROLL; i++) { + T* iteration_buffer = local_buffer + (i * T_per_load); + T residual_buffer[T_per_load]; + + const int iter_offset = i * stride + thread_offset; + const bool do_loads = (iter_offset < elems_per_row); + + mem_access::load_global( + iteration_buffer, input_base + (i * stride), do_loads); + mem_access::load_global( + residual_buffer, residual_base + (i * stride), do_loads); + +#pragma unroll + for (int j = 0; j < T_per_load; j++) { + iteration_buffer[j] += residual_buffer[j]; + float vals_up_cast = conversion::to(iteration_buffer[j]); + + var_sum = reduce::element(var_sum, vals_up_cast * vals_up_cast); + } + + if (do_loads) { + mem_access::store_global(res_output + i * stride, iteration_buffer); + } + } + + reduce::partitioned_block(tb, warp, var_sum); + const float var = var_sum / elems_per_row; + const T denom = conversion::to(__frsqrt_rn(var + epsilon)); + + T* block_output = output + block_offset; + +#pragma unroll + for (int i = 0; i < UNROLL; i++) { + T* iteration_buffer = local_buffer + (i * T_per_load); + const int iter_idx = i * stride + thread_offset; + const bool do_loads = (iter_idx < elems_per_row); + + T gamma_local[T_per_load]; + + mem_access::load_global(gamma_local, gamma + iter_idx, do_loads); + +#pragma unroll + for (int j = 0; j < T_per_load; j++) { + iteration_buffer[j] *= denom; + iteration_buffer[j] *= gamma_local[j]; + } + + if (do_loads) { + mem_access::store_global(block_output + iter_idx, iteration_buffer); + } + } +} + +#define LAUNCH_RMS_NORM(UNROLL, threadsPerGroup, maxThreads) \ + rms_norm \ + <<>>(norm_output, vals, gamma, epsilon, elems_per_row); + +#define LAUNCH_PRE_RMS_NORM(UNROLL, threadsPerGroup, maxThreads) \ + pre_rms_norm<<>>( \ + norm_output, res_output, vals, residual, gamma, epsilon, elems_per_row); + +#define LAUNCH_ALL_RMS_NORM(UNROLL, threadsPerGroup, maxThreads) \ + if (pre_norm) { \ + LAUNCH_PRE_RMS_NORM(UNROLL, threadsPerGroup, maxThreads) \ + } else { \ + LAUNCH_RMS_NORM(UNROLL, threadsPerGroup, maxThreads) \ + } + +template +void launch_rms_norm(T* norm_output, + T* res_output, + const T* vals, + const T* residual, + const T* gamma, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream) +{ + // 8 for __half, 4 for float + constexpr int T_per_load = rms::granularity / sizeof(T); + constexpr int maxThreads = 256; + constexpr int internalUnroll = sizeof(T) == 4 ? 4 : 2; + + const bool is_subblock_schedule = (elems_per_row <= 128) ? true : false; + const int h_per_step = is_subblock_schedule ? T_per_load : T_per_load * internalUnroll; + + // Scheduling concern: may be slightly faster for some inputs to assign multiple stages of + // warp-sized blocks rather than stepping up to 64/96 threads + const int one_step_threads = next_pow2((elems_per_row + h_per_step - 1) / h_per_step); + const int threads_per_group = (one_step_threads < maxThreads) ? one_step_threads : maxThreads; + + const int groups_per_block_max = + is_subblock_schedule ? (maxThreads + threads_per_group - 1) / threads_per_group : 1; + const int groups_per_block = (rows < groups_per_block_max) ? rows : groups_per_block_max; + const int groups_launch = (groups_per_block + rows - 1) / groups_per_block; + + dim3 block(threads_per_group, groups_per_block); + dim3 grid(groups_launch); + + const int elems_per_step = threads_per_group * h_per_step; + const int external_unRoll = (elems_per_row + elems_per_step - 1) / elems_per_step; + + bool pre_norm = (residual == nullptr) ? false : true; + + if (is_subblock_schedule) { + // <=128 + if (threads_per_group == 1) { + LAUNCH_ALL_RMS_NORM(1, 1, maxThreads); + } else if (threads_per_group == 2) { + LAUNCH_ALL_RMS_NORM(1, 2, maxThreads); + } else if (threads_per_group == 4) { + LAUNCH_ALL_RMS_NORM(1, 4, maxThreads); + } else if (threads_per_group == 8) { + LAUNCH_ALL_RMS_NORM(1, 8, maxThreads); + } else if (threads_per_group == 16) { + LAUNCH_ALL_RMS_NORM(1, 16, maxThreads); + } + } else if (external_unRoll == 1) { + // 129 - 4096 elems + // (this can launch with 1-7 warps as well) + LAUNCH_ALL_RMS_NORM(1 * internalUnroll, maxThreads, maxThreads); + } else if (external_unRoll == 2) { + // 4097 - 8192 elems + LAUNCH_ALL_RMS_NORM(2 * internalUnroll, maxThreads, maxThreads); + } else if (external_unRoll == 3) { + // 8193 - 12288 elems + LAUNCH_ALL_RMS_NORM(3 * internalUnroll, maxThreads, maxThreads); + } else if (external_unRoll == 4) { + // 12289 - 16384 elems + LAUNCH_ALL_RMS_NORM(4 * internalUnroll, maxThreads, maxThreads); + } +} + +#define INSTANTIATE_LAUNCH_RMS_NORM(T) \ + template void launch_rms_norm(T * norm_output, \ + T * res_output, \ + const T* vals, \ + const T* residual, \ + const T* gamma, \ + float epsilon, \ + int rows, \ + int elems_per_row, \ + cudaStream_t stream); + +INSTANTIATE_LAUNCH_RMS_NORM(float) +INSTANTIATE_LAUNCH_RMS_NORM(__half) +#ifdef BF16_AVAILABLE +INSTANTIATE_LAUNCH_RMS_NORM(__nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.h b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.h new file mode 100644 index 000000000000..7867fb65964f --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.h @@ -0,0 +1,33 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "ds_kernel_utils.h" + +template +void launch_rms_norm(T* norm_output, + T* res_output, + const T* vals, + const T* residual, + const T* gamma, + float epsilon, + int rows, + int elems_per_row, + cudaStream_t stream); + +void rms_norm(torch::Tensor& norm_output, + torch::Tensor& norm_input, + torch::Tensor& gamma, + float epsilon); + +void rms_pre_norm(torch::Tensor& norm_output, + torch::Tensor& residual_output, + torch::Tensor& norm_input, + torch::Tensor& residual_input, + torch::Tensor& gamma, + float epsilon); diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.py b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.py new file mode 100644 index 000000000000..deb5d33111a9 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.py @@ -0,0 +1,28 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from .rms_norm_base import CUDARMSNormBase + + +class CUDARMSNorm(CUDARMSNormBase): + """ + Floating point layer norm kernel for CUDA/RoCM. + + Performs: z = ln(x) + """ + + def __call__(self, output_z: torch.Tensor, input_x: torch.Tensor, gamma: torch.Tensor) -> torch.Tensor: + """ + output_z may alias input_x directly. All Tensors should have the same shape. + + Parameters: + output_z (torch.Tensor): Output tensor. + input_x (torch.Tensor): Input tensor. + gamma (torch.Tensor): Gamma tensor. + """ + self.inf_module.rms_norm(output_z, input_x, gamma, self.epsilon) + return output_z diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm_base.py b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm_base.py new file mode 100644 index 000000000000..62bc9d056ade --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm_base.py @@ -0,0 +1,37 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from ....inference_utils import elem_size +from deepspeed.ops.op_builder import InferenceCoreBuilder + + +class CUDARMSNormBase(DSKernelBase): + """ + Base class for CUDA LN kernels. They all same the same validation logic, + so we can share it here. + """ + + supported_dtypes = [torch.float16, torch.bfloat16, torch.float32] + + def __init__(self, channels: int, fp_dtype: torch.dtype, epsilon: float = 1e-5): + """ + Parameters: + channels (int): Number of channels in the input tensor. Must be divisible to align + to 16 bytes. + fp_dtype (torch.dtype): Data type for the input/output/gamma. Supported values + are torch.float16, torch.bfloat16, and torch.float32. + """ + if fp_dtype not in CUDARMSNormBase.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, CUDARMSNormBase.supported_dtypes)) + + if elem_size(fp_dtype) * channels % 16 != 0: + raise ValueError("channels must be divisible by 16 bytes") + + self.inf_module = InferenceCoreBuilder().load() + self.epsilon = epsilon diff --git a/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_pre_norm.py b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_pre_norm.py new file mode 100644 index 000000000000..3b040d88b50f --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/cuda_rms_norm/rms_pre_norm.py @@ -0,0 +1,39 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Tuple + +import torch + +from .rms_norm_base import CUDARMSNormBase + + +class CUDARMSPreNorm(CUDARMSNormBase): + """ + Floating point pre-LayerNorm kernel for CUDA/RoCM. + + Performs: z_res = x_res + y_hid + z_hid = ln(z_hid) + """ + + def __call__(self, z_res: torch.Tensor, z_hid: torch.Tensor, x_res: torch.Tensor, y_hid: torch.Tensor, + gamma: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]: + """ + z_res can alias x_res. All non-parameter input/output tensors + must have the same shape. z_hid can alias y_hid. + + Parameters: + z_res (torch.Tensor): Output residual. + z_hid (torch.Tensor): Output hidden states. + x_res (torch.Tensor): Input residual. + y_hid (torch.Tensor): Input hidden states. + gamma (torch.Tensor): Gamma tensor. + beta (torch.Tensor): Beta tensor. + + Returns: + output (torch.Tensor): Output tensor. + """ + self.inf_module.rms_pre_norm(z_hid, z_res, y_hid, x_res, gamma, self.epsilon) + return z_res, z_hid diff --git a/deepspeed/inference/v2/kernels/core_ops/gated_activations/__init__.py b/deepspeed/inference/v2/kernels/core_ops/gated_activations/__init__.py new file mode 100644 index 000000000000..05479d86c906 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/gated_activations/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .gated_activation import * diff --git a/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation.py b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation.py new file mode 100644 index 000000000000..ca1b62ba5c36 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation.py @@ -0,0 +1,65 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from ... import DSKernelBase +from ....inference_utils import ActivationType, elem_size +from deepspeed.ops.op_builder import InferenceCoreBuilder + + +class CUDAGatedActivation(DSKernelBase): + """ + CUDA implementation of gated activation kernel. This kernel assumes that the input + tensor has gate and activation values in adjacent channels. The output tensor should + have half the dimensionality of the input tensor. + """ + + supported_dtypes = [torch.float16, torch.bfloat16, torch.float32] + supported_act_fns = [ActivationType.GEGLU, ActivationType.ReGLU, ActivationType.SiGLU] + + def __init__(self, channels: int, fp_dtype: torch.dtype, act_fn: ActivationType) -> None: + """ + Compile and validate for the gated activation function. + + Args: + channels (int): Number of columns in the output tensor. Must be divisible to align + to 8 bytes. + fp_dtype (torch.dtype): Data type for the input/output/gamma. Supported values + are torch.float16, torch.bfloat16, and torch.float32. + act_fn (ActivationType): Activation function to use. Only GEGLU is supported. + """ + if fp_dtype not in CUDAGatedActivation.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, CUDAGatedActivation.supported_dtypes)) + + act_fn = ActivationType(act_fn) + if act_fn not in CUDAGatedActivation.supported_act_fns: + raise ValueError("Unsupported activation function: {}, supported_act_fns are {}".format( + act_fn, CUDAGatedActivation.supported_act_fns)) + + if elem_size(fp_dtype) * channels % 8 != 0: + raise ValueError("Channels must be divisible by 16 bytes") + + if elem_size(fp_dtype) * channels > 98304: + raise ValueError( + "Kernel only compiled to support 98304 bytes per row, please file an issue if your model requires more." + ) + + self.inf_module = InferenceCoreBuilder().load() + self.act_fn = act_fn + self.kernel = self.inf_module.gated_activation + + def __call__(self, output: torch.Tensor, input: torch.Tensor, bias: Optional[torch.Tensor] = None) -> None: + """ + Performs gated activation on the input tensor, writing the result to the output tensor. + + Args: + output (torch.Tensor): Output tensor. Can be of [T, C // 2] or [B, S, C // 2] + input (torch.Tensor): Input tensor. Can be of [T, C] or [B, S, C] + """ + self.kernel(output, input, bias, self.act_fn.value) diff --git a/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cpp b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cpp new file mode 100644 index 000000000000..05463c75138c --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cpp @@ -0,0 +1,72 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "gated_activation_kernels.h" + +#ifdef BF16_AVAILABLE +#define DISPATCH_FOR_FLOAT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat32) { \ + using scalar_t = float; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kFloat16) { \ + using scalar_t = __half; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kBFloat16) { \ + using scalar_t = __nv_bfloat16; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dtype for BiasActivation"); \ + } \ + }() +#else +#define DISPATCH_FOR_FLOAT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat32) { \ + using scalar_t = float; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kFloat16) { \ + using scalar_t = __half; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dtype for BiasActivation"); \ + } \ + }() +#endif + +void ds_gated_activation(at::Tensor& output, + at::Tensor& input, + c10::optional& bias, + int activation_type_raw) +{ + bool ragged_input = input.dim() == 2; + + const ActivationType activation_type = static_cast(activation_type_raw); + + const int rows = ragged_input ? input.size(0) : input.size(0) * input.size(1); + const int cols = ragged_input ? input.size(1) : input.size(2); + + DISPATCH_FOR_FLOAT(input.scalar_type(), [&] { + scalar_t* bias_ptr = nullptr; + if (bias.has_value()) { + TORCH_CHECK(bias.value().scalar_type() == input.scalar_type(), + "Bias type must match input type"); + TORCH_CHECK(bias.value().numel() == cols, + "Bias must have the same number of elements as the input channels"); + bias_ptr = reinterpret_cast(bias.value().data_ptr()); + } + + scalar_t* output_ptr = reinterpret_cast(output.data_ptr()); + const scalar_t* input_ptr = reinterpret_cast(input.data_ptr()); + + launch_gated_activation(output_ptr, + input_ptr, + bias_ptr, + rows, + cols, + activation_type, + c10::cuda::getCurrentCUDAStream()); + }); +} diff --git a/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cu b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cu new file mode 100644 index 000000000000..84a9906cf037 --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cu @@ -0,0 +1,169 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include +#include "activation_type.h" +#include "conversion_utils.h" +#include "ds_kernel_utils.h" +#include "memory_access_utils.h" + +namespace cg = cooperative_groups; + +namespace gated_act { + +constexpr int access_size = 16; +constexpr int threads = 1024; + +template +float gated_act_fn(float x, float y); + +template <> +DS_D_INLINE float gated_act_fn(float x, float y) +{ + constexpr float sqrt_param = 0.79788456080286535587989211986876f; + constexpr float mul_param = 0.044715; + return y * x * 0.5f * (1.0f + tanhf(sqrt_param * (x + mul_param * x * x * x))); +} + +template <> +DS_D_INLINE float gated_act_fn(float x, float y) +{ + return y * (x > 0.0f ? x : 0.0f); +} + +template <> +DS_D_INLINE float gated_act_fn(float x, float y) +{ + return y * (x / (1.0f + expf(-x))); +} + +} // namespace gated_act + +template +__global__ void gated_activation_kernel(T* output, + const T* input, + const T* bias, + int rows, + int cols) +{ + constexpr int read_vector = gated_act::access_size / sizeof(T); + constexpr int write_vector = read_vector / 2; + + const int row = blockIdx.x; + const int col = threadIdx.x * read_vector; + + const T* input_row = input + row * cols; + T* output_row = output + row * cols / 2; + +#pragma unroll + for (int i = 0; i < loopUnroll; i++) { + T read[read_vector]; + T bias_read[read_vector]; + T store[write_vector]; + + const int read_offset = col + gated_act::threads * read_vector * i; + const int write_offset = col / 2 + gated_act::threads * write_vector * i; + + if (i != loopUnroll - 1 || read_offset < cols) { + mem_access::load_global(read, input_row + read_offset); + mem_access::load_global( + bias_read, bias + read_offset, bias != nullptr); + + for (int j = 0; j < write_vector; j++) { + float g_val = + conversion::to(read[j * 2]) + conversion::to(bias_read[j * 2]); + float a_val = conversion::to(read[j * 2 + 1]) + + conversion::to(bias_read[j * 2 + 1]); + + float act_val = gated_act::gated_act_fn(g_val, a_val); + store[j] = conversion::to(act_val); + } + + mem_access::store_global(output_row + write_offset, store); + } + } +} + +#define DISPATCH_UNROLL(unroll_val) \ + gated_activation_kernel \ + <<>>(output, input, bias, rows, cols); + +template +void launch_gated_activation_impl(T* output, + const T* input, + const T* bias, + int rows, + int cols, + cudaStream_t stream) +{ + constexpr int read_vector = gated_act::access_size / sizeof(T); + constexpr int cols_per_unroll = gated_act::threads * read_vector; + const int req_threads = (cols + read_vector - 1) / read_vector; + const int threads = std::min(req_threads, gated_act::threads); + + const dim3 grid(rows); + const dim3 block(threads); + const int unroll = (cols + cols_per_unroll - 1) / cols_per_unroll; + + if (unroll == 1) { + DISPATCH_UNROLL(1); + } else if (unroll == 2) { + DISPATCH_UNROLL(2); + } else if (unroll == 3) { + DISPATCH_UNROLL(3); + } else if (unroll == 4) { + DISPATCH_UNROLL(4); + } else if (unroll == 5) { + DISPATCH_UNROLL(5); + } else if (unroll == 6) { + DISPATCH_UNROLL(6); + } else { + throw std::runtime_error( + "Called with more columns than supported, please report this bug and this limit will " + "be increased."); + } +} + +template +void launch_gated_activation(T* output, + const T* input, + const T* bias, + int rows, + int cols, + ActivationType act_type, + cudaStream_t stream) +{ + switch (act_type) { + case ActivationType::GEGLU: + launch_gated_activation_impl( + output, input, bias, rows, cols, stream); + break; + case ActivationType::ReGLU: + launch_gated_activation_impl( + output, input, bias, rows, cols, stream); + break; + case ActivationType::SiGLU: + launch_gated_activation_impl( + output, input, bias, rows, cols, stream); + break; + default: throw std::runtime_error("Unsupported activation type"); + } +} + +#define INSTANTIATE_FOR_TYPE(T) \ + template void launch_gated_activation(T * output, \ + const T* input, \ + const T* bias, \ + int rows, \ + int cols, \ + ActivationType act_type, \ + cudaStream_t stream); + +INSTANTIATE_FOR_TYPE(float) +INSTANTIATE_FOR_TYPE(__half) + +#ifdef BF16_AVAILABLE +INSTANTIATE_FOR_TYPE(__nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.h b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.h new file mode 100644 index 000000000000..6ae01e99679a --- /dev/null +++ b/deepspeed/inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.h @@ -0,0 +1,25 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "activation_type.h" +#include "ds_kernel_utils.h" + +template +void launch_gated_activation(T* output, + const T* vals, + const T* bias, + int rows, + int cols, + ActivationType activation_type, + cudaStream_t stream); + +void ds_gated_activation(at::Tensor& output, + at::Tensor& input, + c10::optional& bias, + int activation_type_raw); diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/LICENSE b/deepspeed/inference/v2/kernels/cutlass_ops/LICENSE new file mode 100644 index 000000000000..d64569567334 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/LICENSE @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/__init__.py b/deepspeed/inference/v2/kernels/cutlass_ops/__init__.py new file mode 100644 index 000000000000..44b9adbae794 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/__init__.py @@ -0,0 +1,7 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .mixed_gemm import * +from .moe_gemm import * diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/cutlass_ops.cpp b/deepspeed/inference/v2/kernels/cutlass_ops/cutlass_ops.cpp new file mode 100644 index 000000000000..18e834f3e60a --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/cutlass_ops.cpp @@ -0,0 +1,19 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include + +#include "mixed_gemm.h" +#include "moe_gemm.h" + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) +{ + // mixed_gemm.h + m.def("mixed_gemm", &mixed_gemm, "Mixed-precision GEMM"); + + // moe_gemm.h + m.def("moe_gemm", &moe_gemm, "MultiGEMM for MoE (16-bit weights)"); + m.def("mixed_moe_gemm", &mixed_moe_gemm, "MultiGEMM for MoE (4-bit/8-bit weights)"); +} diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/__init__.py b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/__init__.py new file mode 100644 index 000000000000..14ccf2ce5354 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .mixed_gemm import * diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.cu b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.cu new file mode 100644 index 000000000000..7c522203bb48 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.cu @@ -0,0 +1,93 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include +#include "mixed_gemm.h" +#include "mixed_gemm_api.h" +#include "weight_variant.h" + +// Switch helpers inspired by +// https://github.com/NVIDIA/DALI/blob/main/include/dali/core/static_switch.h +// and https://github.com/pytorch/pytorch/blob/master/aten/src/ATen/Dispatch.h + +#define ACT_DTYPE_SWITCH(COND, ...) \ + [&] { \ + if (COND) { \ + using ActivationDtype = __half; \ + return __VA_ARGS__(); \ + } else { \ + using ActivationDtype = __nv_bfloat16; \ + return __VA_ARGS__(); \ + } \ + }() + +#define WEIGHT_VARIANT_SWITCH(COND, ...) \ + [&] { \ + if (COND) { \ + constexpr WeightVariant WVariant = WeightVariant::kFP8; \ + return __VA_ARGS__(); \ + } else { \ + constexpr WeightVariant WVariant = WeightVariant::kFP4; \ + return __VA_ARGS__(); \ + } \ + }() + +void mixed_gemm(at::Tensor& output, + at::Tensor& hidden_states, + at::Tensor& weight, + at::Tensor& scales, + c10::optional& bias, + int num_bits, + int activation_raw) +{ + TORCH_CHECK(output.dtype() == hidden_states.dtype(), + "Output and hidden states must have the same dtype"); + TORCH_CHECK(num_bits == 4 || num_bits == 8, "Data width must be 4 or 8"); + TORCH_CHECK(output.size(0) == hidden_states.size(0), "Token dimension mismatch"); + + int32_t m = output.size(0); + int32_t k = hidden_states.size(1); + int32_t n = weight.size(1); + + TORCH_CHECK(weight.size(0) == k, "Weight dimension mismatch"); + + ACT_DTYPE_SWITCH(hidden_states.dtype() == torch::kFloat16, [&] { + WEIGHT_VARIANT_SWITCH(num_bits == 8, [&] { + fastertransformer::CutlassFpAIntBGemmRunner runner = + *MixedGemmContext::Instance().GeMM_Runner(); + + ActivationType activation_type = (ActivationType)activation_raw; + if (!bias.has_value() && activation_type == ActivationType::IDENTITY) { + runner.gemm((ActivationDtype*)hidden_states.data_ptr(), + (const char*)weight.data_ptr(), + (ActivationDtype*)scales.data_ptr(), + (ActivationDtype*)output.data_ptr(), + m, + n, + k, + nullptr, + 0, + at::cuda::getCurrentCUDAStream()); + return; + } else { + ActivationDtype* bias_ptr = nullptr; + if (bias.has_value()) { bias_ptr = (ActivationDtype*)bias.value().data_ptr(); } + runner.gemm_bias_act((ActivationDtype*)hidden_states.data_ptr(), + (char*)weight.data_ptr(), + (ActivationDtype*)scales.data_ptr(), + bias_ptr, + (ActivationDtype*)output.data_ptr(), + m, + n, + k, + activation_type, + nullptr, + 0, + at::cuda::getCurrentCUDAStream()); + return; + } + }); + }); +} diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.h b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.h new file mode 100644 index 000000000000..1fc3831e9084 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.h @@ -0,0 +1,16 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include + +void mixed_gemm(at::Tensor& output, + at::Tensor& hidden_states, + at::Tensor& weight, + at::Tensor& scales, + c10::optional& bias, + int num_bits, + int activation_raw); diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.py b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.py new file mode 100644 index 000000000000..dddb555e267a --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.py @@ -0,0 +1,64 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from ....inference_utils import ActivationType, DtypeEnum +from deepspeed.ops.op_builder import InferenceCutlassBuilder + +from typing import Optional + + +class MixedGEMM(DSKernelBase): + """ + CUTLASS implementation of MoE GEMM. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_act_fns = [ActivationType.GELU, ActivationType.SILU, ActivationType.RELU, ActivationType.IDENTITY] + + def __init__(self, fp_dtype: DtypeEnum, act_fn: ActivationType, num_bits: int) -> None: + + if not isinstance(fp_dtype, DtypeEnum): + fp_dtype = DtypeEnum(fp_dtype) + + if fp_dtype not in MixedGEMM.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, MixedGEMM.supported_dtypes)) + + if act_fn not in MixedGEMM.supported_act_fns: + raise ValueError("Unsupported activation function: {}, supported_act_fns are {}".format( + act_fn, MixedGEMM.supported_act_fns)) + + if num_bits != 4 and num_bits != 8: + raise ValueError("Unsupported num_bits: {}, supported num_bits are 4 and 8".format(num_bits)) + + inf_module = InferenceCutlassBuilder().load() + self.num_bits = num_bits + self.kernel = inf_module.moe_gemm + self.act_fn = act_fn + + def __call__(self, + output: torch.Tensor, + hidden_states: torch.Tensor, + weights: torch.Tensor, + scales: torch.Tensor, + biases: Optional[torch.Tensor] = None) -> None: + """ + Performs a MoE GEMM. Note that the stride between token inputs must be even (the distance between byte 1 of token 0 and token 1 must be the same as the distance between byte 1 of token 1 and token 2). + + Arguments: + output (torch.Tensor): The output of the MoE GEMM of shape [n_tokens, out_neurons]. + hidden_states (torch.Tensor): The direct input for the MoE GEMM of shape [n_tokens, in_neurons]. + weights (torch.Tensor): The weights of shape [in_neurons, out_neurons]. These weights must be contiguous. + scales (torch.Tensor): The scales of shape [out_neurons]. These scales must be contiguous. + biases (torch.Tensor): The biases of shape [out_neurons]. These biases must be contiguous. + + Returns: + output + """ + self.kernel(output, hidden_states, weights, biases, self.num_bits, self.act_fn) + return output diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm_api.h b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm_api.h new file mode 100644 index 000000000000..74fc07ffc4a2 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm_api.h @@ -0,0 +1,57 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "activation_type.h" +#include "weight_variant.h" + +namespace fastertransformer { + +template +class CutlassFpAIntBGemmRunner { +public: + void gemm(const T* A, + const char* B, + const T* weight_scales, + T* C, + int m, + int n, + int k, + char* workspace_ptr, + const size_t workspace_bytes, + cudaStream_t stream); + + void gemm_bias_act(const T* A, + const char* B, + const T* weight_scales, + const T* biases, + T* C, + int m, + int n, + int k, + ActivationType activation_type, + char* workspace_ptr, + const size_t workspace_bytes, + cudaStream_t stream); +}; + +} // namespace fastertransformer + +template +class MixedGemmContext { +public: + MixedGemmContext() { _runner = new fastertransformer::CutlassFpAIntBGemmRunner(); } + + virtual ~MixedGemmContext() { delete _runner; } + + static MixedGemmContext& Instance() + { + static MixedGemmContext _ctx; + return _ctx; + } + + fastertransformer::CutlassFpAIntBGemmRunner* GeMM_Runner() const { return _runner; } + + fastertransformer::CutlassFpAIntBGemmRunner* _runner; +}; diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/__init__.py b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/__init__.py new file mode 100644 index 000000000000..aff4e77bba98 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/__init__.py @@ -0,0 +1,7 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .mixed_moe_gemm import * +from .moe_gemm import * diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/mixed_moe_gemm.py b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/mixed_moe_gemm.py new file mode 100644 index 000000000000..9c55ce341532 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/mixed_moe_gemm.py @@ -0,0 +1,67 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from ....inference_utils import ActivationType, DtypeEnum +from deepspeed.ops.op_builder import InferenceCutlassBuilder + +from typing import Optional + + +class MixedMoEGEMM(DSKernelBase): + """ + CUTLASS implementation of MoE GEMM. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_act_fns = [ActivationType.GELU, ActivationType.SILU, ActivationType.RELU, ActivationType.IDENTITY] + + def __init__(self, fp_dtype: DtypeEnum, act_fn: ActivationType, num_bits: int) -> None: + + if not isinstance(fp_dtype, DtypeEnum): + fp_dtype = DtypeEnum(fp_dtype) + + if fp_dtype not in MixedMoEGEMM.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, MixedMoEGEMM.supported_dtypes)) + + if act_fn not in MixedMoEGEMM.supported_act_fns: + raise ValueError("Unsupported activation function: {}, supported_act_fns are {}".format( + act_fn, MixedMoEGEMM.supported_act_fns)) + + if num_bits != 4 and num_bits != 8: + raise ValueError("Unsupported num_bits: {}, supported num_bits are 4 and 8".format(num_bits)) + + inf_module = InferenceCutlassBuilder().load() + self.num_bits = num_bits + self.kernel = inf_module.moe_gemm + self.act_fn = act_fn + + def __call__(self, + ordered_output: torch.Tensor, + ordered_input: torch.Tensor, + weights: torch.Tensor, + scales: torch.Tensor, + total_rows_before_expert: torch.Tensor, + biases: Optional[torch.Tensor] = None) -> None: + """ + Performs a MoE GEMM. Note that the stride between token inputs must be even (the distance between byte 1 of token 0 and token 1 must be the same as the distance between byte 1 of token 1 and token 2). + + Arguments: + ordered_output (torch.Tensor): The output of the MoE GEMM of shape [n_tokens, out_neurons]. + ordered_input (torch.Tensor): The direct input for the MoE GEMM of shape [n_tokens, in_neurons]. + weights (torch.Tensor): The weights of shape [n_experts, in_neurons, out_neurons]. These weights must be contiguous. + scales (torch.Tensor): The scales of shape [n_experts, out_neurons]. These scales must be contiguous. + total_rows_before_expert (torch.Tensor): The total number of rows before each expert of shape [n_experts]. + biases (torch.Tensor): The biases of shape [n_experts, out_neurons]. These biases must be contiguous. + + Returns: + ordered_output + """ + self.kernel(ordered_output, ordered_input, weights, scales, biases, total_rows_before_expert, self.num_bits, + self.act_fn) + return ordered_output diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.cu b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.cu new file mode 100644 index 000000000000..d1cafc9fff4c --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.cu @@ -0,0 +1,175 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include +#include "moe_gemm.h" +#include "moe_gemm_api.h" +#include "weight_variant.h" + +// Switch helpers inspired by +// https://github.com/NVIDIA/DALI/blob/main/include/dali/core/static_switch.h +// and https://github.com/pytorch/pytorch/blob/master/aten/src/ATen/Dispatch.h + +#define HIDDEN_DTYPE_SWITCH(COND, ...) \ + [&] { \ + if (COND) { \ + using ActivationDtype = __half; \ + constexpr WeightVariant WVariant = WeightVariant::kFP16; \ + return __VA_ARGS__(); \ + } else { \ + using ActivationDtype = __nv_bfloat16; \ + constexpr WeightVariant WVariant = WeightVariant::kBF16; \ + return __VA_ARGS__(); \ + } \ + }() + +void moe_gemm(at::Tensor& output, + at::Tensor& hidden_states, + at::Tensor& weight, + c10::optional& bias, + at::Tensor& total_rows_before_expert, + int activation_raw) +{ + TORCH_CHECK(output.dtype() == hidden_states.dtype(), + "Output and hidden states must have the same dtype"); + TORCH_CHECK(output.dtype() == weight.dtype(), "Output and weight must have the same dtype"); + + int64_t total_rows = hidden_states.size(0); + int64_t gemm_k = hidden_states.size(1); + int64_t gemm_n = weight.size(2); + int num_experts = weight.size(0); + + TORCH_CHECK(total_rows == output.size(0), "Total rows dimension mismatch"); + TORCH_CHECK(gemm_k == weight.size(1), "GEMM K dimension mismatch"); + TORCH_CHECK(gemm_n == output.size(1), "GEMM N dimension mismatch"); + TORCH_CHECK(num_experts == total_rows_before_expert.size(0), "Number of experts mismatch"); + + HIDDEN_DTYPE_SWITCH(hidden_states.dtype() == torch::kFloat16, [&] { + fastertransformer::MoeGemmRunner runner = + *MoeGemmContext::Instance().GeMM_Runner(); + + ActivationType activation_type = (ActivationType)activation_raw; + if (!bias.has_value() && activation_type == ActivationType::IDENTITY) { + runner.moe_gemm((ActivationDtype*)hidden_states.data_ptr(), + (char*)weight.data_ptr(), + nullptr, + (ActivationDtype*)output.data_ptr(), + (int64_t*)total_rows_before_expert.data_ptr(), + total_rows, + gemm_n, + gemm_k, + num_experts, + at::cuda::getCurrentCUDAStream()); + return; + } else { + ActivationDtype* bias_ptr = nullptr; + if (bias.has_value()) { + bias_ptr = (ActivationDtype*)bias.value().data_ptr(); + TORCH_CHECK(num_experts == bias.value().size(0), "Number of experts mismatch"); + TORCH_CHECK(gemm_n == bias.value().size(1), "GEMM N dimension mismatch"); + } + runner.moe_gemm_bias_act((ActivationDtype*)hidden_states.data_ptr(), + (char*)weight.data_ptr(), + nullptr, + bias_ptr, + (ActivationDtype*)output.data_ptr(), + (int64_t*)total_rows_before_expert.data_ptr(), + total_rows, + gemm_n, + gemm_k, + num_experts, + activation_type, + at::cuda::getCurrentCUDAStream()); + return; + } + }); +} + +#define ACT_DTYPE_SWITCH(COND, ...) \ + [&] { \ + if (COND) { \ + using ActivationDtype = __half; \ + return __VA_ARGS__(); \ + } else { \ + using ActivationDtype = __nv_bfloat16; \ + return __VA_ARGS__(); \ + } \ + }() + +#define WEIGHT_VARIANT_SWITCH(COND, ...) \ + [&] { \ + if (COND) { \ + constexpr WeightVariant WVariant = WeightVariant::kFP8; \ + return __VA_ARGS__(); \ + } else { \ + constexpr WeightVariant WVariant = WeightVariant::kFP4; \ + return __VA_ARGS__(); \ + } \ + }() + +void mixed_moe_gemm(at::Tensor& output, + at::Tensor& hidden_states, + at::Tensor& weight, + at::Tensor& scales, + c10::optional& bias, + at::Tensor& total_rows_before_expert, + int num_bits, + int activation_raw) +{ + TORCH_CHECK(output.dtype() == hidden_states.dtype(), + "Output and hidden states must have the same dtype"); + + int64_t total_rows = hidden_states.size(0); + int64_t gemm_k = hidden_states.size(1); + int64_t gemm_n = weight.size(2); + int num_experts = weight.size(0); + + TORCH_CHECK(total_rows == output.size(0), "Total rows dimension mismatch"); + TORCH_CHECK(gemm_k == weight.size(1), "GEMM K dimension mismatch"); + TORCH_CHECK(gemm_n == output.size(1), "GEMM N dimension mismatch"); + TORCH_CHECK(num_experts == total_rows_before_expert.size(0), "Number of experts mismatch"); + + ACT_DTYPE_SWITCH(hidden_states.dtype() == torch::kFloat16, [&] { + WEIGHT_VARIANT_SWITCH(num_bits == 8, [&] { + fastertransformer::MoeGemmRunner runner = + *MoeGemmContext::Instance().GeMM_Runner(); + + ActivationType activation_type = (ActivationType)activation_raw; + if (!bias.has_value() && activation_type == ActivationType::IDENTITY) { + runner.moe_gemm((ActivationDtype*)hidden_states.data_ptr(), + (char*)weight.data_ptr(), + (ActivationDtype*)scales.data_ptr(), + (ActivationDtype*)output.data_ptr(), + (int64_t*)total_rows_before_expert.data_ptr(), + total_rows, + gemm_n, + gemm_k, + num_experts, + at::cuda::getCurrentCUDAStream()); + return; + } else { + ActivationDtype* bias_ptr = nullptr; + if (bias.has_value()) { + bias_ptr = (ActivationDtype*)bias.value().data_ptr(); + TORCH_CHECK(num_experts == bias.value().size(0), "Number of experts mismatch"); + TORCH_CHECK(gemm_n == bias.value().size(1), "GEMM N dimension mismatch"); + } + runner.moe_gemm_bias_act((ActivationDtype*)hidden_states.data_ptr(), + (char*)weight.data_ptr(), + (ActivationDtype*)scales.data_ptr(), + bias_ptr, + (ActivationDtype*)output.data_ptr(), + (int64_t*)total_rows_before_expert.data_ptr(), + total_rows, + gemm_n, + gemm_k, + num_experts, + activation_type, + at::cuda::getCurrentCUDAStream()); + return; + } + }); + }); +} diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.h b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.h new file mode 100644 index 000000000000..dfd3d4561567 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.h @@ -0,0 +1,24 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include + +void moe_gemm(at::Tensor& output, + at::Tensor& hidden_states, + at::Tensor& weight, + c10::optional& bias, + at::Tensor& total_rows_before_expert, + int activation_raw); + +void mixed_moe_gemm(at::Tensor& output, + at::Tensor& hidden_states, + at::Tensor& weight, + at::Tensor& scales, + c10::optional& bias, + at::Tensor& total_rows_before_expert, + int num_bits, + int activation_raw); diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.py b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.py new file mode 100644 index 000000000000..0cc233e8d87a --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.py @@ -0,0 +1,60 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from ....inference_utils import ActivationType, DtypeEnum +from deepspeed.ops.op_builder import InferenceCutlassBuilder + +from typing import Optional + + +class MoEGEMM(DSKernelBase): + """ + CUTLASS implementation of MoE GEMM. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_act_fns = [ActivationType.GELU, ActivationType.SILU, ActivationType.RELU, ActivationType.IDENTITY] + + def __init__(self, fp_dtype: DtypeEnum, act_fn: ActivationType) -> None: + + if not isinstance(fp_dtype, DtypeEnum): + fp_dtype = DtypeEnum(fp_dtype) + + if fp_dtype not in MoEGEMM.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, MoEGEMM.supported_dtypes)) + + if act_fn not in MoEGEMM.supported_act_fns: + raise ValueError("Unsupported activation function: {}, supported_act_fns are {}".format( + act_fn, MoEGEMM.supported_act_fns)) + + inf_module = InferenceCutlassBuilder().load() + self.kernel = inf_module.moe_gemm + self.act_fn = act_fn + + def __call__(self, + ordered_output: torch.Tensor, + ordered_input: torch.Tensor, + weights: torch.Tensor, + total_rows_before_expert: torch.Tensor, + biases: Optional[torch.Tensor] = None) -> None: + """ + Performs a MoE GEMM. Note that the stride between token inputs must be even (the distance between byte 1 of token 0 and token 1 must be the same as the distance between byte 1 of token 1 and token 2). + + Arguments: + ordered_output (torch.Tensor): The output of the MoE GEMM of shape [n_tokens, out_neurons]. + ordered_input (torch.Tensor): The direct input for the MoE GEMM of shape [n_tokens, in_neurons]. + weights (torch.Tensor): The weights of shape [n_experts, in_neurons, out_neurons]. These weights must be contiguous. + total_rows_before_expert (torch.Tensor): The total number of rows before each expert of shape [n_experts]. + biases (torch.Tensor): The biases of shape [n_experts, out_neurons]. These biases must be contiguous. + + Returns: + ordered_output + """ + self.kernel(ordered_output, ordered_input, weights, biases, total_rows_before_expert, self.act_fn) + return ordered_output diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm_api.h b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm_api.h new file mode 100644 index 000000000000..7ad92070b35f --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm_api.h @@ -0,0 +1,64 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "activation_type.h" +#include "weight_variant.h" + +namespace fastertransformer { + +template +class MoeGemmRunner { +public: + MoeGemmRunner(); + + void moe_gemm_bias_act(const T* A, + const char* B, + const T* weight_scales, + const T* biases, + T* C, + int64_t* total_rows_before_expert, + int64_t total_rows, + int64_t gemm_n, + int64_t gemm_k, + int num_experts, + ActivationType activation_type, + cudaStream_t stream); + + void moe_gemm(const T* A, + const char* B, + const T* weight_scales, + T* C, + int64_t* total_rows_before_expert, + int64_t total_rows, + int64_t gemm_n, + int64_t gemm_k, + int num_experts, + cudaStream_t stream); + +private: + int sm_; + int multi_processor_count_; +}; + +} // namespace fastertransformer + +template +class MoeGemmContext { +public: + MoeGemmContext() { _runner = new fastertransformer::MoeGemmRunner(); } + + virtual ~MoeGemmContext() { delete _runner; } + + static MoeGemmContext& Instance() + { + static MoeGemmContext _ctx; + return _ctx; + } + + fastertransformer::MoeGemmRunner* GeMM_Runner() const { return _runner; } + + fastertransformer::MoeGemmRunner* _runner; +}; diff --git a/deepspeed/inference/v2/kernels/cutlass_ops/shared_resources/weight_variant.h b/deepspeed/inference/v2/kernels/cutlass_ops/shared_resources/weight_variant.h new file mode 100644 index 000000000000..4d17c799f726 --- /dev/null +++ b/deepspeed/inference/v2/kernels/cutlass_ops/shared_resources/weight_variant.h @@ -0,0 +1,11 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +// Data structure that allows us to abstract internal CUTLASS datatypes/mappings +// to the DeepSpeed-Kernels repo. + +#pragma once + +enum WeightVariant { kFP16, kBF16, kFP8, kFP4 }; diff --git a/deepspeed/inference/v2/kernels/ds_kernel.py b/deepspeed/inference/v2/kernels/ds_kernel.py new file mode 100644 index 000000000000..8dbfa1de86a6 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ds_kernel.py @@ -0,0 +1,32 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import ABC, abstractmethod + + +class DSKernelBase(ABC): + + @abstractmethod + def __init__(self, *args, **kwargs): + """ + If necessary trigger compilation and warmup + Autotuning of the kernel would happen at this stage to + eliminate any potential hangs that might occur mid-deployment + Validate that the desired run configuration is compatible. + + It is not necessary to call super on this method. + """ + raise NotImplementedError() + + @abstractmethod + def __call__(self, *args, **kwargs): + """ + However the kernel needs to be called, it can be called here. Auto-tuning + should never be performed here. + + All inputs/outputs should be passed as arguments to this function. No allocations + should be performed here. + """ + raise NotImplementedError() diff --git a/deepspeed/inference/v2/kernels/ragged_ops/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/__init__.py new file mode 100644 index 000000000000..988152b2e7c0 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/__init__.py @@ -0,0 +1,13 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .atom_builder import * +from .blocked_flash import * +from .embed import * +from .linear_blocked_kv_rotary import * +from .logits_gather import * +from .moe_gather import * +from .moe_scatter import * +from .top_1_gating import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/__init__.py new file mode 100644 index 000000000000..c79201cdf165 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .atom_builder import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.cpp b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.cpp new file mode 100644 index 000000000000..7ad4dc5faa20 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.cpp @@ -0,0 +1,53 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "atom_builder.h" +#include "attention_atom.h" +#include "ragged_dtypes.h" + +int32_t build_atoms(torch::Tensor& atoms_ten, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& kv_ptrs, + const int32_t q_block_size, + const int32_t kv_block_size) +{ + const RaggedBatchDescriptor* batch_desc = + reinterpret_cast(batch_metadata.data_ptr()); + + const InflightSeqDescriptor* seq_desc = + reinterpret_cast(seq_metadata.data_ptr()); + + int32_t** kv_ptr_list = reinterpret_cast(kv_ptrs.data_ptr()); + + AttentionAtom* atoms = reinterpret_cast(atoms_ten.data_ptr()); + + int32_t n_atoms = 0; + for (int i = 0; i < batch_desc->n_sequences; i++) { + const int seq_atoms = (seq_desc[i].n_tokens + q_block_size - 1) / q_block_size; + int32_t cur_start_idx = seq_desc[i].start_idx; + int32_t global_start_idx = seq_desc[i].seen_tokens; + int32_t remaining_toks = seq_desc[i].n_tokens; + + for (int j = 0; j < seq_atoms; j++) { + atoms[n_atoms].block_idx_list = kv_ptr_list[i]; + atoms[n_atoms].q_start_idx = cur_start_idx; + atoms[n_atoms].q_len = std::min(remaining_toks, q_block_size); + atoms[n_atoms].global_q_idx = global_start_idx; + + const int32_t end_toks = global_start_idx + atoms[n_atoms].q_len; + // TODO(cmikeh2): This logic needs to be changed for sparse implementations + atoms[n_atoms].kv_blocks = (end_toks + kv_block_size - 1) / kv_block_size; + atoms[n_atoms].total_extent = end_toks; + + cur_start_idx += atoms[n_atoms].q_len; + global_start_idx += atoms[n_atoms].q_len; + remaining_toks -= atoms[n_atoms].q_len; + n_atoms++; + } + } + + return n_atoms; +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.h b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.h new file mode 100644 index 000000000000..a3342d0e6695 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.h @@ -0,0 +1,21 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include + +/* +Construct the attention atoms given the ragged metadata for the current batch. +This could largely be done at the Python level, but since we pack the KV ptr +alongside the int32_t metadata, it gets very ugly to handle the mixed-width +data structures (since we're packing them in a single tensor). +*/ +int32_t build_atoms(torch::Tensor& atoms_ten, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& kv_ptrs, + const int32_t q_block_size, + const int32_t kv_block_size); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.py b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.py new file mode 100644 index 000000000000..3355ca76c6a4 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/atom_builder/atom_builder.py @@ -0,0 +1,50 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Tuple + +import torch + +from ... import DSKernelBase +from deepspeed.ops.op_builder import RaggedOpsBuilder +from ....ragged import RaggedBatchWrapper + + +class AtomBuilder(DSKernelBase): + """ + C++ implementation to populate the attention atoms for the blocked attention + kernel. + """ + + def __init__(self) -> None: + """ + Triggers compilation of the C++ implementation. + """ + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.build_atoms + + def __call__(self, atoms: torch.Tensor, ragged_batch: RaggedBatchWrapper, q_block_size: int, + kv_block_size: int) -> Tuple[torch.Tensor, int]: + """ + Populates the attention atoms for the blocked attention kernel. + + Args: + atoms (torch.Tensor): Pre-allocated int32 tensor of shape [max_atoms, 8] + ragged_batch (torch.Tensor): Wrapper for the ragged batch. + q_block_size (int): The block size for the queries (as determined by the + attention implementation) + kv_block_size (int): The block size for the keys/values (as determined by the + attention implementation) + + Returns: + + """ + if atoms.device != torch.device("cpu"): + raise RuntimeError("AtomBuilder must be called on tensors") + + n_atoms = self.kernel(atoms, ragged_batch.batch_metadata_buffer(on_device=False), + ragged_batch.inflight_seq_descriptors(on_device=False), + ragged_batch.kv_ptrs(on_device=False), q_block_size, kv_block_size) + return atoms, n_atoms diff --git a/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/__init__.py new file mode 100644 index 000000000000..87b2b3d68777 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .blocked_flash import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/attention_atom.h b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/attention_atom.h new file mode 100644 index 000000000000..ed8eb9e19b3d --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/attention_atom.h @@ -0,0 +1,39 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include "cuda.h" + +struct AttentionAtom { + /* + The attention atom describes the workload of a particular query. The attention + kernel will execute each ``AttentionAtom`` for each head of the model. + */ + + // Pointer to a list of KV block indices. + int32_t* block_idx_list; + + // Index of first token in the ragged batch associated with this atom. + int32_t q_start_idx; + + // Number of tokens in the ragged batch associated with this atom. + int32_t q_len; + + // Number of key/value blocks associated with this atom. All but the last are + // assumed to be fully dense. + int32_t kv_blocks; + + // Number of tokens in the last key/value block. + int32_t total_extent; + + // Global index of the first token in the atom. For example, in a prompt continuation + // in which we have already processed 768 tokens, this would be 768. + int32_t global_q_idx; + + // Unused + int32_t unused; +}; diff --git a/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.cpp b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.cpp new file mode 100644 index 000000000000..6cb60ad00f9b --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.cpp @@ -0,0 +1,101 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +/****************************************************************************** + * Copyright (c) 2023, Tri Dao. + ******************************************************************************/ + +#include +#include +#include + +#include "blocked_flash.h" +#include "flash.h" + +#define CHECK_SHAPE(x, ...) \ + TORCH_CHECK(x.sizes() == torch::IntArrayRef({__VA_ARGS__}), \ + #x " must have shape (" #__VA_ARGS__ ")") + +void flash_attn_by_atoms(at::Tensor& out, + at::Tensor& q, + at::Tensor& k, + at::Tensor& v, + at::Tensor& attention_atoms, + const float softmax_scale, + const bool is_causal) +{ + auto dprops = at::cuda::getCurrentDeviceProperties(); + + bool is_sm8x = dprops->major == 8 && dprops->minor >= 0; + bool is_sm90 = dprops->major == 9 && dprops->minor == 0; + TORCH_CHECK(is_sm90 || is_sm8x, "FlashAttention only supports Ampere GPUs or newer."); + + auto q_dtype = q.dtype(); + TORCH_CHECK(q_dtype == torch::kFloat16 || q_dtype == torch::kBFloat16, + "FlashAttention only support fp16 and bf16 data type"); + if (q_dtype == torch::kBFloat16) { + TORCH_CHECK(is_sm90 || is_sm8x, "bfloat16 is only supported on Ampere GPUs or newer"); + } + TORCH_CHECK(k.dtype() == q_dtype, "query and key must have the same dtype"); + TORCH_CHECK(v.dtype() == q_dtype, "query and value must have the same dtype"); + + TORCH_CHECK(q.is_cuda(), "Input tensor must be on CUDA device"); + TORCH_CHECK(k.is_cuda(), "Input tensor must be on CUDA device"); + TORCH_CHECK(v.is_cuda(), "Input tensor must be on CUDA device"); + + TORCH_CHECK(q.stride(-1) == 1, "Input tensor must have contiguous last dimension"); + TORCH_CHECK(k.stride(-1) == 1, "Input tensor must have contiguous last dimension"); + TORCH_CHECK(v.stride(-1) == 1, "Input tensor must have contiguous last dimension"); + + const int total_q = q.size(0); + const int head_size = k.size(-1); + const int num_heads_kv = k.size(-2); + const int num_heads_q = q.size(-1) / head_size; + + TORCH_CHECK(head_size <= 256, "head_size must be <= 256"); + TORCH_CHECK(head_size % 8 == 0, "head_size must be divisible by 8"); + TORCH_CHECK(num_heads_q % num_heads_kv == 0, "num_heads_q must be divisible by num_heads_kv"); + + Flash_fwd_params params; + + params.is_bf16 = q.dtype() == torch::kBFloat16; + + // Set the pointers and strides. + params.q_ptr = q.data_ptr(); + params.k_ptr = k.data_ptr(); + params.v_ptr = v.data_ptr(); + params.o_ptr = out.data_ptr(); + params.atoms = reinterpret_cast(attention_atoms.data_ptr()); + + // All stride are in elements, not bytes. + params.q_row_stride = q.stride(0); + params.k_row_stride = k.stride(1); + params.v_row_stride = v.stride(1); + params.o_row_stride = out.stride(0); + + // Assume heads are contiguous. + params.q_head_stride = head_size; + params.k_head_stride = head_size; + params.v_head_stride = head_size; + params.o_head_stride = head_size; + + // Head params + params.h = num_heads_q; + params.h_k = num_heads_kv; + params.h_h_k_ratio = num_heads_q / num_heads_kv; + params.d = head_size; + auto round_multiple = [](int x, int m) { return (x + m - 1) / m * m; }; + params.d_rounded = round_multiple(head_size, 32); + params.num_atoms = attention_atoms.size(0); + + // Set the different scale values. + params.scale_softmax = softmax_scale; + params.scale_softmax_log2 = softmax_scale * M_LOG2E; + + params.is_causal = is_causal; + + auto stream = at::cuda::getCurrentCUDAStream().stream(); + run_mha_fwd(params, stream); +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.h b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.h new file mode 100644 index 000000000000..68037b425113 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.h @@ -0,0 +1,16 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include + +void flash_attn_by_atoms(at::Tensor& out, + at::Tensor& q, + at::Tensor& k, + at::Tensor& v, + at::Tensor& attention_atoms, + const float softmax_scale, + const bool is_causal); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.py b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.py new file mode 100644 index 000000000000..54d465698b4e --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.py @@ -0,0 +1,107 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from deepspeed.accelerator import get_accelerator +from ....inference_utils import DtypeEnum +from deepspeed.ops.op_builder import RaggedOpsBuilder + +from ... import DSKernelBase + + +def get_q_block_size(head_size: int) -> int: + """ + Returns the query block size required by the kernel given a head size. + """ + cc_major, cc_minor = torch.cuda.get_device_capability(get_accelerator().current_device()) #ignore-cuda + + if cc_major < 8: + raise RuntimeError("Blocked attention requires CUDA compute capability >= 8.0") + + if head_size <= 64: + return 128 + elif head_size <= 160: + if cc_minor != 0: + return 64 + else: + return 128 + elif head_size == 192: + return 128 + elif head_size == 224: + if cc_minor != 0: + return 64 + else: + return 128 + else: + if cc_major == 8 and cc_minor == 0: + return 128 + else: + return 64 + + +def get_kv_block_size(head_size: int) -> int: + """ + Return preferred granulatity for blocked KV-cache implementation. + """ + cc_major, cc_minor = torch.cuda.get_device_capability(get_accelerator().current_device()) #ignore-cuda + + if cc_major < 8: + raise RuntimeError("Blocked attention requires CUDA compute capability >= 8.0") + + if head_size <= 64: + return 128 + elif head_size != 160 or cc_minor != 0: + return 64 + else: + return 32 + + +class BlockedFlashAttn(DSKernelBase): + """ + Modified implementation of flash-attn-2 tuned for inference on blocked KV-cache and wider + range of input sequence lengths. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + + def __init__(self, head_size: int, dtype: DtypeEnum) -> None: + """ + Triggers any compilation of the kernels. + """ + if not isinstance(dtype, DtypeEnum): + dtype = DtypeEnum(dtype) + + if dtype not in BlockedFlashAttn.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported data types are {}".format( + dtype, BlockedFlashAttn.supported_dtypes)) + + # For testing, need to revert to 32 + if head_size % 16 != 0: + raise ValueError("Head size must be divisible by 32 (configured with {})".format(head_size)) + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.flash_attn_by_atoms + + def __call__(self, out: torch.Tensor, q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, atoms: torch.Tensor, + softmax_scale: float) -> torch.Tensor: + """ + Flash attention implementation atop a blocked KV-cache. Atoms should be pre-populated. + See attention_atom.h for further details on the structure of the information. + + Arguments: + out (torch.Tensor): Output tensor of shape [tokens, hidden_size] + q (torch.Tensor): Query tensor of shape [tokens, hidden_size] + k (torch.Tensor): Key cache tensor of shape [n_blocks, block_size, n_heads_kv, head_size]. This Tensor only needs to be contiguous on the final dimension. + v (torch.Tensor): Value cache tensor of shape [n_blocks, block_size, n_heads_kv, head_size]. This Tensor only needs to be contiguous on the final dimension. + atoms (torch.Tensor): Atom information tensor of shape [num_atoms, 8] and type int32. + Not all data is readable in this format. See attention_atom.h for further details. + softmax_scale (float): Softmax scale factor. + + Returns: + out (torch.Tensor): Output tensor of shape [tokens, hidden_size] + """ + self.kernel(out, q, k, v, atoms, softmax_scale, True) + return out diff --git a/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/flash.h b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/flash.h new file mode 100644 index 000000000000..b4a53e6d7f52 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/blocked_flash/flash.h @@ -0,0 +1,74 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +/****************************************************************************** +Copyright (c) 2023, Tri Dao. + ******************************************************************************/ + +#pragma once + +#include +#include + +#include "attention_atom.h" + +constexpr int TOTAL_DIM = 0; +constexpr int H_DIM = 1; +constexpr int D_DIM = 2; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +struct Qkv_params { + using index_t = uint32_t; + // The QKV matrices. + void* __restrict__ q_ptr; + void* __restrict__ k_ptr; + void* __restrict__ v_ptr; + + // The stride between rows of the Q, K and V matrices. + index_t q_row_stride; + index_t k_row_stride; + index_t v_row_stride; + index_t q_head_stride; + index_t k_head_stride; + index_t v_head_stride; + + // The number of heads. + int h, h_k; + // In the case of multi-query and grouped-query attention (MQA/GQA), nheads_k could be + // different from nheads (query). + int h_h_k_ratio; // precompute h / h_k, +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +struct Flash_fwd_params : public Qkv_params { + // The O matrix (output). + void* __restrict__ o_ptr; + + // The attention metadata + AttentionAtom* __restrict__ atoms; + + // Total attention atoms + int num_atoms; + + // The stride between rows of O. + index_t o_row_stride; + index_t o_head_stride; + + // The dimensions + int d, d_rounded; + + // The scaling factors for the kernel. + float scale_softmax; + float scale_softmax_log2; + + bool is_bf16; + bool is_causal; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +void run_mha_fwd(Flash_fwd_params& params, cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/embed/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/embed/__init__.py new file mode 100644 index 000000000000..d6b8e6047d74 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/embed/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .embed import RaggedEmbeddingKernel diff --git a/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cpp b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cpp new file mode 100644 index 000000000000..04b72bf948db --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cpp @@ -0,0 +1,101 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "embed.h" +#include "ragged_kernel_helpers.h" + +#ifdef BF16_AVAILABLE +#define DISPATCH_FOR_FLOAT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat32) { \ + using float_t = float; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kFloat16) { \ + using float_t = __half; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kBFloat16) { \ + using float_t = __nv_bfloat16; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dispatch type"); \ + } \ + }() +#else +#define DISPATCH_FOR_FLOAT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kFloat32) { \ + using float_t = float; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kFloat16) { \ + using float_t = __half; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dispatch type"); \ + } \ + }() +#endif + +#define DISPATCH_FOR_INT(DTYPE, ...) \ + [&] { \ + if (DTYPE == torch::kInt32) { \ + using int_t = int32_t; \ + return __VA_ARGS__(); \ + } else if (DTYPE == torch::kInt64) { \ + using int_t = int64_t; \ + return __VA_ARGS__(); \ + } else { \ + TORCH_CHECK(false, "Unsupported dispatch type"); \ + } \ + }() + +/* +Embeddings kernel aware of ragged batch structure. +*/ +void ragged_embed(torch::Tensor& embedded_tokens, + torch::Tensor& input_ids, + torch::Tensor& embedding_weight, + c10::optional& position_embedding_weight, + int32_t pos_embed_offset, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs) +{ + // We don't care about KV cache here, so just hardcoding 0s for block_size/num_blocks + BatchWrapperCPP batch_wrapper = + make_cpp_batch_wrapper(batch_metadata, seq_metadata, tokens_to_seq, kv_ptrs, 0, 0); + + const int32_t n_tokens = input_ids.numel(); + const int32_t embed_dim = embedding_weight.size(1); + const int32_t vocab_size = embedding_weight.size(0); + + DISPATCH_FOR_INT(input_ids.scalar_type(), [&] { + DISPATCH_FOR_FLOAT(embedding_weight.scalar_type(), [&] { + float_t* pos_embed_ptr = nullptr; + int32_t max_position_embed_idx = 0; + if (position_embedding_weight.has_value()) { + TORCH_CHECK( + position_embedding_weight.value().options().dtype() == + embedding_weight.options().dtype(), + "position_embedding_weight and embedding_weight must have the same dtype"); + pos_embed_ptr = + reinterpret_cast(position_embedding_weight.value().data_ptr()); + max_position_embed_idx = position_embedding_weight.value().size(0) - 1; + } + + launch_ragged_embed_kernel((float_t*)embedded_tokens.data_ptr(), + (const int_t*)input_ids.data_ptr(), + (const float_t*)embedding_weight.data_ptr(), + pos_embed_ptr, + batch_wrapper, + n_tokens, + embed_dim, + vocab_size, + max_position_embed_idx, + pos_embed_offset, + at::cuda::getCurrentCUDAStream()); + }); + }); +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cu b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cu new file mode 100644 index 000000000000..81d6d534ddf5 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cu @@ -0,0 +1,137 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "ds_kernel_utils.h" +#include "embed.cuh" +#include "memory_access_utils.h" +#include "ragged_dtypes.h" + +namespace embed { + +constexpr int granularity = 16; +constexpr int threads = 512; + +} // namespace embed + +template +__global__ void ragged_embed_kernel(EmbedType* embedded_tokens, + const TokenType* input_ids, + const EmbedType* embedding_weight, + const EmbedType* position_weight, + const BatchWrapperCPP batch_desc, + const int32_t embed_dim, + const int32_t vocab_size, + const int32_t max_position_embed_idx, + const int32_t position_embed_offset) +{ + constexpr int T_vector = embed::granularity / sizeof(EmbedType); + + const int32_t token_idx = blockIdx.y; + + // It's possible our batch is padded (under CG conditions typically) + if (token_idx >= batch_desc.batch_metadata->n_tokens) return; + + TokenType token_value = input_ids[token_idx]; + + if (token_value >= vocab_size || token_value < 0) { + // TODO(cmikeh2): This is invalid, but not sure how we want to handle it being invalid + // yet. + return; + } + + const EmbedType* embedding_row = embedding_weight + token_value * embed_dim; + EmbedType* dest_row = embedded_tokens + token_idx * embed_dim; + + const int channel_offset = (threadIdx.x + embed::threads * blockIdx.x) * T_vector; + + if (channel_offset < embed_dim) { + EmbedType reg_buf[T_vector]; + + mem_access::load_global(reg_buf, embedding_row + channel_offset); + + if (position_weight != nullptr) { + // Map the token to its global idx (indirect memory accesses aren't great but whatever) + const int32_t seq_idx = batch_desc.tokens_to_seq[token_idx]; + const InflightSeqDescriptor seq_desc = batch_desc.seq_metadata[seq_idx]; + int32_t pos_emb_idx = seq_desc.seen_tokens + (token_idx - seq_desc.start_idx); + + // Position embed offset is an OPT-specific feature I think? + pos_emb_idx = pos_emb_idx + position_embed_offset; + + // This clamping is technically + pos_emb_idx = (pos_emb_idx < 0) ? 0 : pos_emb_idx; + pos_emb_idx = (pos_emb_idx >= max_position_embed_idx) ? max_position_embed_idx + : pos_emb_idx; + + const EmbedType* position_embedding_row = position_weight + pos_emb_idx * embed_dim; + + EmbedType pos_buf[T_vector]; + mem_access::load_global(pos_buf, + position_embedding_row + channel_offset); + +#pragma unroll + for (int i = 0; i < T_vector; i++) { reg_buf[i] += pos_buf[i]; } + } + + mem_access::store_global(dest_row + channel_offset, reg_buf); + } +} + +template +void launch_ragged_embed_kernel(EmbedType* embedded_tokens, + const TokenType* input_ids, + const EmbedType* embedding_weight, + const EmbedType* position_weight, + const BatchWrapperCPP batch_desc, + const int32_t n_tokens, + const int32_t embed_dim, + const int32_t vocab_size, + const int32_t max_position_embed_idx, + const int32_t position_embed_offset, + cudaStream_t stream) +{ + constexpr int T_vector = embed::granularity / sizeof(EmbedType); + constexpr int elems_per_block = embed::threads * T_vector; + const int parallel_blocks = (embed_dim + elems_per_block - 1) / elems_per_block; + + const dim3 grid_dim(parallel_blocks, n_tokens, 1); + const dim3 block_dim(embed::threads, 1, 1); + + ragged_embed_kernel + <<>>(embedded_tokens, + input_ids, + embedding_weight, + position_weight, + batch_desc, + embed_dim, + vocab_size, + max_position_embed_idx, + position_embed_offset); +} + +#define INSTANTIATE_EMBED_FOR_TYPES(TOKEN_TYPE, EMBED_TYPE) \ + template void launch_ragged_embed_kernel( \ + EMBED_TYPE * embedded_tokens, \ + const TOKEN_TYPE* input_ids, \ + const EMBED_TYPE* embedding_weight, \ + const EMBED_TYPE* position_weight, \ + const BatchWrapperCPP batch_descriptor, \ + const int32_t n_tokens, \ + const int32_t embed_dim, \ + const int32_t vocab_size, \ + const int32_t max_position_embed_idx, \ + const int32_t position_embed_offset, \ + cudaStream_t stream); + +INSTANTIATE_EMBED_FOR_TYPES(int32_t, float) +INSTANTIATE_EMBED_FOR_TYPES(int64_t, float) + +INSTANTIATE_EMBED_FOR_TYPES(int32_t, __half) +INSTANTIATE_EMBED_FOR_TYPES(int64_t, __half) + +#ifdef BF16_AVAILABLE +INSTANTIATE_EMBED_FOR_TYPES(int32_t, __nv_bfloat16) +INSTANTIATE_EMBED_FOR_TYPES(int64_t, __nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cuh b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cuh new file mode 100644 index 000000000000..94c397439b80 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.cuh @@ -0,0 +1,26 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" +#include "ragged_dtypes.h" + +#ifdef BF16_AVAILABLE +#include +#endif + +template +void launch_ragged_embed_kernel(EmbedType* embedded_tokens, + const TokenType* input_ids, + const EmbedType* embedding_weight, + const EmbedType* position_weight, + const BatchWrapperCPP batch_desc, + const int32_t n_tokens, + const int32_t embed_dim, + const int32_t vocab_size, + const int32_t max_position_embed_idx, + const int32_t position_embed_offset, + cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.h b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.h new file mode 100644 index 000000000000..7897c1362669 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.h @@ -0,0 +1,23 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "embed.cuh" + +/* +Embeddings kernel aware of ragged batch structure. +*/ +void ragged_embed(torch::Tensor& embedded_tokens, + torch::Tensor& input_ids, + torch::Tensor& embedding_weight, + c10::optional& position_weight, + int32_t position_embed_offset, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.py b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.py new file mode 100644 index 000000000000..0443ce3fdd8e --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/embed/embed.py @@ -0,0 +1,67 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from ... import DSKernelBase +from deepspeed.ops.op_builder import RaggedOpsBuilder +from ....inference_utils import elem_size +from ....ragged import RaggedBatchWrapper + + +class RaggedEmbeddingKernel(DSKernelBase): + """ + Ragged-aware CUDA kernel implementation for an embedding lookup. This will only lookup + the necessary tokens for a padded batch (i.e. if we are CGed and running with a slightly + larger batch size than the actual tokens). + """ + + supported_dtypes = [torch.float16, torch.bfloat16, torch.float32] + supported_token_dtypes = [torch.int32, torch.int64] + + def __init__(self, embed_dtype: torch.dtype, token_dtype: torch.dtype, embed_dim: int) -> None: + """ + Args: + fp_dtype (torch.dtype): Data type of the embedding table and output dtype. + Supported values are torch.float16, torch.bfloat16, and torch.float32. + token_dtype (torch.dtype): Data type of the token ids. Supported values are + torch.int32 and torch.int64. + embed_dim (int): Embedding dimension. Must be aligned to 16 bytes. + """ + if embed_dtype not in RaggedEmbeddingKernel.supported_dtypes: + raise ValueError("Unsupported embedding data type: {}, supported_dtypes are {}".format( + embed_dtype, RaggedEmbeddingKernel.supported_dtypes)) + + if token_dtype not in RaggedEmbeddingKernel.supported_token_dtypes: + raise ValueError("Unsupported token data type: {}, supported_dtypes are {}".format( + token_dtype, RaggedEmbeddingKernel.supported_token_dtypes)) + + if elem_size(embed_dtype) * embed_dim % 16 != 0: + raise ValueError("Embedding dimension must be aligned to 16 bytes, got {}".format(embed_dim)) + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.ragged_embed + + def __call__(self, + embedded_tokens: torch.Tensor, + ragged_wrapper: RaggedBatchWrapper, + embedding_weight: torch.Tensor, + position_embed_weight: Optional[torch.Tensor] = None, + position_embed_offset: int = 0) -> torch.Tensor: + """ + Ragged aware embedding lookup. + + Args: + embedded_tokens (torch.Tensor): Output tensor of shape [num_tokens, embed_dim] + ragged_wrapper (RaggedBatchWrapper): Wrapper for the ragged batch. + embedding_weight (torch.Tensor): Embedding table of shape [vocab_size, embed_dim] + """ + self.kernel(embedded_tokens, ragged_wrapper.input_ids(), + embedding_weight, position_embed_weight, position_embed_offset, + ragged_wrapper.batch_metadata_buffer(), ragged_wrapper.inflight_seq_descriptors(), + ragged_wrapper.tokens_to_seq(), ragged_wrapper.kv_ptrs()) + return embedded_tokens diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/__init__.py new file mode 100644 index 000000000000..0e239dd6b4c7 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/__init__.py @@ -0,0 +1,8 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .blocked_kv_rotary import * +from .blocked_trained_kv_rotary import * +from .linear_blocked_kv_copy import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cpp b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cpp new file mode 100644 index 000000000000..8493bbf4b9af --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cpp @@ -0,0 +1,188 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "blocked_kv_rotary.h" +#include "ragged_kernel_helpers.h" + +#define DISPATCH_KV_ROTARY(T_TYPE, C_TYPE) \ + if (q.options().dtype() == torch::T_TYPE) { \ + launch_kv_rotary_kernel((C_TYPE*)kv_cache.data_ptr(), \ + (C_TYPE*)q.data_ptr(), \ + (C_TYPE*)k.data_ptr(), \ + (C_TYPE*)v.data_ptr(), \ + (C_TYPE*)inv_freq_ptr, \ + batch_wrapper, \ + qkv_stride, \ + kv_cache_stride, \ + v_offset, \ + inv_freq_stride, \ + q_ratio, \ + head_size, \ + n_tokens, \ + n_q_heads, \ + at::cuda::getCurrentCUDAStream()); \ + } + +/* +Rotary position embeddings + copy into KV cache. This implementation assumes +that the inverse frequencies should be ready from global memory rather than +synthesized in the kernel. + +Arguments: + kv_cache: [n_blocks, block_size, 2, n_kv_heads, head_size] + q: [n_tokens, n_q_heads * head_size] + k: [n_tokens, n_kv_heads * head_size] + v: [n_tokens, n_kv_heads * head_size] + inv_freq: [max_seq_len, head_size // 2] +*/ +void kv_trained_rotary_embeddings(torch::Tensor& kv_cache, + torch::Tensor& q, + torch::Tensor& k, + torch::Tensor& v, + torch::Tensor& inv_freq, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs) +{ + const int32_t n_tokens = q.size(0); + TORCH_CHECK(n_tokens == k.size(0)); + TORCH_CHECK(n_tokens == v.size(0)); + + // Dimensions + const int32_t block_size = kv_cache.size(1); + const int32_t n_kv_heads = kv_cache.size(3); + const int32_t head_size = kv_cache.size(4); + + // Strides + const int32_t qkv_stride = q.stride(0); // Per token + const int32_t kv_cache_stride = kv_cache.stride(1); // Per token + const int32_t v_offset = kv_cache.stride(2); // From k_cache to v_cache + const int32_t inv_freq_stride = inv_freq.stride(0); // Per token idx + + const int n_q_heads = q.size(1) / head_size; + const int q_ratio = n_q_heads / n_kv_heads; + + void* inv_freq_ptr = (void*)inv_freq.data_ptr(); + + BatchWrapperCPP batch_wrapper = make_cpp_batch_wrapper( + batch_metadata, seq_metadata, tokens_to_seq, kv_ptrs, block_size, kv_cache.size(0)); + + DISPATCH_KV_ROTARY(kHalf, __half); + +#ifdef BF16_AVAILABLE + DISPATCH_KV_ROTARY(kBFloat16, __nv_bfloat16); +#endif +} + +/* +Rotary position embeddings + copy into KV cache. This implementation assumes +that the inverse frequencies should be synthesized in the kernel. + +Arguments: + kv_cache: [n_blocks, block_size, 2, n_kv_heads, head_size] + q: [n_tokens, n_q_heads * head_size] + k: [n_tokens, n_kv_heads * head_size] + v: [n_tokens, n_kv_heads * head_size] +*/ +void kv_rotary_embeddings(torch::Tensor& kv_cache, + torch::Tensor& q, + torch::Tensor& k, + torch::Tensor& v, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs) +{ + const int32_t n_tokens = q.size(0); + TORCH_CHECK(n_tokens == k.size(0)); + TORCH_CHECK(n_tokens == v.size(0)); + + // Dimensions + const int32_t block_size = kv_cache.size(1); + const int32_t n_kv_heads = kv_cache.size(3); + const int32_t head_size = kv_cache.size(4); + + // Strides + const int32_t qkv_stride = q.stride(0); // Per token + const int32_t kv_cache_stride = kv_cache.stride(1); // Per token + const int32_t v_offset = kv_cache.stride(2); // From k_cache to v_cache + const int32_t inv_freq_stride = 0; // Per token idx + + const int n_q_heads = q.size(1) / head_size; + const int q_ratio = n_q_heads / n_kv_heads; + + void* inv_freq_ptr = nullptr; + + BatchWrapperCPP batch_wrapper = make_cpp_batch_wrapper( + batch_metadata, seq_metadata, tokens_to_seq, kv_ptrs, block_size, kv_cache.size(0)); + + DISPATCH_KV_ROTARY(kHalf, __half); + +#ifdef BF16_AVAILABLE + DISPATCH_KV_ROTARY(kBFloat16, __nv_bfloat16); +#endif +} + +#define DISPATCH_KV_COPY(T_TYPE, C_TYPE) \ + if (q.options().dtype() == torch::T_TYPE) { \ + launch_kv_copy_kernel((C_TYPE*)kv_cache.data_ptr(), \ + (C_TYPE*)q.data_ptr(), \ + (C_TYPE*)k.data_ptr(), \ + (C_TYPE*)v.data_ptr(), \ + batch_wrapper, \ + qkv_stride, \ + kv_cache_stride, \ + v_offset, \ + q_ratio, \ + head_size, \ + n_tokens, \ + n_q_heads, \ + at::cuda::getCurrentCUDAStream()); \ + } + +/* +Copy into linear KV cache. +*/ +void linear_kv_copy(torch::Tensor& kv_cache, + torch::Tensor& q, + torch::Tensor& k, + torch::Tensor& v, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs) +{ + const int32_t n_tokens = q.size(0); + TORCH_CHECK(n_tokens == k.size(0)); + TORCH_CHECK(n_tokens == v.size(0)); + + // Dimensions + const int32_t block_size = kv_cache.size(1); + const int32_t n_kv_heads = kv_cache.size(3); + const int32_t head_size = kv_cache.size(4); + + // Strides + const int32_t qkv_stride = q.stride(0); // Per token + TORCH_CHECK(qkv_stride == k.stride(0)); + TORCH_CHECK(qkv_stride == v.stride(0)); + + const int32_t kv_cache_stride = kv_cache.stride(1); // Per token + const int32_t v_offset = kv_cache.stride(2); // From k_cache to v_cache + + const int n_q_heads = q.size(1) / head_size; + + TORCH_CHECK(n_q_heads % n_kv_heads == 0); + const int q_ratio = n_q_heads / n_kv_heads; + + BatchWrapperCPP batch_wrapper = make_cpp_batch_wrapper( + batch_metadata, seq_metadata, tokens_to_seq, kv_ptrs, block_size, kv_cache.size(0)); + + DISPATCH_KV_COPY(kHalf, __half); + +#ifdef BF16_AVAILABLE + DISPATCH_KV_COPY(kBFloat16, __nv_bfloat16); +#endif +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cu b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cu new file mode 100644 index 000000000000..63ea5bc88bab --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cu @@ -0,0 +1,314 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "blocked_kv_rotary.cuh" +#include "conversion_utils.h" +#include "ds_kernel_utils.h" +#include "memory_access_utils.h" + +namespace cg = cooperative_groups; + +namespace kv_rot { + +constexpr int granularity = 16; +constexpr int threads = 256; + +} // namespace kv_rot + +/* +Supports head size 32, 64, 128, 256 +*/ + +template +__global__ void kv_rotary_pos_kernel(T* kv_cache, + T* q, + T* k, + T* v, + const T* inv_freq, + const BatchWrapperCPP batch_desc, + const int qkv_stride, + const int kv_cache_stride, + const int v_offset, + const int inv_freq_stride) +{ + // Derived constexpr + constexpr int vector_T = kv_rot::granularity / sizeof(T); + constexpr int threads_per_head = headSize / vector_T; + constexpr int half_head_size = headSize >> 1; + constexpr int tokens_per_block = kv_rot::threads / threads_per_head; + + // CG helpers + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + cg::thread_block_tile head_group = + cg::tiled_partition(warp); + + // Parallelize on the head dimension for X blocks + const int head_idx = blockIdx.x; + + const int block_seq_idx = threadIdx.x / threads_per_head; + const int base_neuron_idx = (threadIdx.x * vector_T) % headSize; + const int half_idx = base_neuron_idx % half_head_size; + const int half_head_lanes = threads_per_head / 2; + + // Multiple tokens processed by the same threadblock + const int token_idx = blockIdx.y * tokens_per_block + block_seq_idx; + const bool valid_token = token_idx < batch_desc.batch_metadata->n_tokens; + const bool load_inv_freq = (inv_freq != nullptr) && valid_token; + + // If we have GQA, then only one of the Q heads needs to do rotary + copy + // for each of the heads in the group. + bool need_kv = head_idx % qRatio == 0; + // Make sure the following code is warp uniform + need_kv = warp.shfl(need_kv, 0); + + const int kv_head_idx = head_idx / qRatio; + + // Ensure we don't access invalid portions of the seq_metadata + const int32_t seq_id = (valid_token) ? batch_desc.tokens_to_seq[token_idx] : 0; + const InflightSeqDescriptor seq_desc = batch_desc.seq_metadata[seq_id]; + // This will give an invalid index if valid_token is false, but should never affect memory. + const int32_t global_token_idx = seq_desc.seen_tokens + (token_idx - seq_desc.start_idx); + + T* q_row = q + token_idx * qkv_stride + head_idx * headSize; + T q_reg[vector_T]; + + if (need_kv) { + // The following logic assumes a linearly blocked KV cache. This means that no sparsity has + // been introduced into cache history. + const KVCacheDescriptor kv_desc = batch_desc.kv_desc; + const int32_t seq_kv_block_idx = global_token_idx / kv_desc.block_size; + const int32_t mapped_kv_block_idx = + (valid_token) ? kv_desc.block_lists[seq_id][seq_kv_block_idx] : 0; + + const int32_t kv_block_offset = global_token_idx % kv_desc.block_size; + const int32_t kv_offset = + (mapped_kv_block_idx * kv_desc.block_size + kv_block_offset) * kv_cache_stride + + kv_head_idx * headSize; + + // Load indices from QKV output + T* k_row = k + token_idx * qkv_stride + kv_head_idx * headSize; + T* v_row = v + token_idx * qkv_stride + kv_head_idx * headSize; + + T k_reg[vector_T], v_reg[vector_T], inv_freq_reg[vector_T]; + + mem_access::load_global(q_reg, q_row + base_neuron_idx, valid_token); + mem_access::load_global(k_reg, k_row + base_neuron_idx, valid_token); + mem_access::load_global(v_reg, v_row + base_neuron_idx, valid_token); + mem_access::load_global( + inv_freq_reg, inv_freq + half_idx, load_inv_freq); + + if constexpr (doRotary) { +#pragma unroll + for (int i = 0; i < vector_T; i++) { + const int head_neuron_idx = base_neuron_idx + i; + + float inv_freq_flt; + if (inv_freq != nullptr) { + inv_freq_flt = conversion::to(inv_freq_reg[i]) * (float)global_token_idx; + } else { + inv_freq_flt = + (float)((head_neuron_idx % half_head_size) * 2) / (float)headSize; + // Conversion to T and back means that both branches of this if statement + // will produce the same results if using the same algo for producing the + // freqs. + T trunc_freq = conversion::to(1.0 / powf(10000.0, inv_freq_flt)); + inv_freq_flt = conversion::to(trunc_freq) * (float)global_token_idx; + } + + float rotary_sign = (head_neuron_idx >= half_head_size) ? -1.0f : 1.0f; + float q_f = conversion::to(q_reg[i]); + float k_f = conversion::to(k_reg[i]); + float q_rot = q_f * rotary_sign; + float k_rot = k_f * rotary_sign; + + const float q_rot_temp = head_group.shfl_xor(q_rot, half_head_lanes); + const float k_rot_temp = head_group.shfl_xor(k_rot, half_head_lanes); + + q_reg[i] = + conversion::to(q_f * cosf(inv_freq_flt) + q_rot_temp * sinf(inv_freq_flt)); + k_reg[i] = + conversion::to(k_f * cosf(inv_freq_flt) + k_rot_temp * sinf(inv_freq_flt)); + } + } + + if (valid_token) { + mem_access::store_global(kv_cache + kv_offset + base_neuron_idx, + k_reg); + mem_access::store_global( + kv_cache + kv_offset + base_neuron_idx + v_offset, v_reg); + } + } else { + T inv_freq_reg[vector_T]; + + mem_access::load_global(q_reg, q_row + base_neuron_idx, valid_token); + mem_access::load_global( + inv_freq_reg, inv_freq + half_idx, load_inv_freq); + + if constexpr (doRotary) { +#pragma unroll + for (int i = 0; i < vector_T; i++) { + const int head_neuron_idx = base_neuron_idx + i; + + float inv_freq_flt; + if (inv_freq != nullptr) { + inv_freq_flt = conversion::to(inv_freq_reg[i]) * (float)global_token_idx; + } else { + inv_freq_flt = + (float)((head_neuron_idx % half_head_size) * 2) / (float)headSize; + inv_freq_flt = 1.0 / powf(10000.0, inv_freq_flt) * (float)global_token_idx; + } + + float rotary_sign = (head_neuron_idx >= half_head_size) ? -1.0f : 1.0f; + float q_f = conversion::to(q_reg[i]); + float q_rot = q_f * rotary_sign; + + const float q_rot_temp = head_group.shfl_xor(q_rot, half_head_lanes); + + q_reg[i] = + conversion::to(q_f * cosf(inv_freq_flt) + q_rot_temp * sinf(inv_freq_flt)); + } + } + } + + if (valid_token && doRotary) { + mem_access::store_global(q_row + base_neuron_idx, q_reg); + } +} + +#define DISPATCH_KV_ROTARY_IMPL(Q_RATIO, HEAD_SIZE) \ + if (q_ratio == Q_RATIO && head_size == HEAD_SIZE) \ + kv_rotary_pos_kernel \ + <<>>(kv_cache, \ + q, \ + k, \ + v, \ + inv_freq, \ + batch_desc, \ + qkv_stride, \ + kv_cache_stride, \ + v_offset, \ + inv_freq_stride); + +template +void launch_kv_rotary_kernel(T* kv_cache, + T* q, + T* k, + T* v, + T* inv_freq, + const BatchWrapperCPP batch_desc, + const int qkv_stride, + const int kv_cache_stride, + const int v_offset, + const int inv_freq_stride, + const int q_ratio, + const int head_size, + const int n_tokens, + const int n_q_heads, + cudaStream_t stream) +{ + constexpr int vector_T = kv_rot::granularity / sizeof(T); + const int threads_per_head = head_size / vector_T; + const int tokens_per_block = kv_rot::threads / threads_per_head; + + const dim3 block(kv_rot::threads); + const int token_blocks = (n_tokens + tokens_per_block - 1) / tokens_per_block; + const dim3 grid(n_q_heads, token_blocks); + + DISPATCH_KV_ROTARY_IMPL(1, 64) + DISPATCH_KV_ROTARY_IMPL(1, 128) + DISPATCH_KV_ROTARY_IMPL(2, 64) + DISPATCH_KV_ROTARY_IMPL(2, 128) + DISPATCH_KV_ROTARY_IMPL(4, 64) + DISPATCH_KV_ROTARY_IMPL(4, 128) + DISPATCH_KV_ROTARY_IMPL(5, 64) + DISPATCH_KV_ROTARY_IMPL(5, 128) + DISPATCH_KV_ROTARY_IMPL(8, 64) + DISPATCH_KV_ROTARY_IMPL(8, 128) +} + +#define INSTANTIATE_KV_ROTARY_KERNEL(TYPE) \ + template void launch_kv_rotary_kernel(TYPE * kv_cache, \ + TYPE * q, \ + TYPE * k, \ + TYPE * v, \ + TYPE * inv_freq, \ + const BatchWrapperCPP batch_desc, \ + const int qkv_stride, \ + const int kv_cache_stride, \ + const int v_offset, \ + const int inv_freq_stride, \ + const int q_ratio, \ + const int head_size, \ + const int n_tokens, \ + const int n_q_heads, \ + cudaStream_t stream); + +INSTANTIATE_KV_ROTARY_KERNEL(__half) + +#ifdef BF16_AVAILABLE +INSTANTIATE_KV_ROTARY_KERNEL(__nv_bfloat16) +#endif + +#define DISPATCH_KV_COPY_IMPL(Q_RATIO, HEAD_SIZE) \ + if (q_ratio == Q_RATIO && head_size == HEAD_SIZE) \ + kv_rotary_pos_kernel<<>>( \ + kv_cache, q, k, v, nullptr, batch_desc, qkv_stride, kv_cache_stride, v_offset, 0); + +template +void launch_kv_copy_kernel(T* kv_cache, + T* q, + T* k, + T* v, + const BatchWrapperCPP batch_desc, + const int qkv_stride, + const int kv_cache_stride, + const int v_offset, + const int q_ratio, + const int head_size, + const int n_tokens, + const int n_q_heads, + cudaStream_t stream) +{ + constexpr int vector_T = kv_rot::granularity / sizeof(T); + const int threads_per_head = head_size / vector_T; + const int tokens_per_block = kv_rot::threads / threads_per_head; + + const dim3 block(kv_rot::threads); + const int token_blocks = (n_tokens + tokens_per_block - 1) / tokens_per_block; + const dim3 grid(n_q_heads, token_blocks); + + DISPATCH_KV_COPY_IMPL(1, 64) + DISPATCH_KV_COPY_IMPL(1, 128) + DISPATCH_KV_COPY_IMPL(2, 64) + DISPATCH_KV_COPY_IMPL(2, 128) + DISPATCH_KV_COPY_IMPL(4, 64) + DISPATCH_KV_COPY_IMPL(4, 128) + DISPATCH_KV_COPY_IMPL(5, 64) + DISPATCH_KV_COPY_IMPL(5, 128) + DISPATCH_KV_COPY_IMPL(8, 64) + DISPATCH_KV_COPY_IMPL(8, 128) +} + +#define INSTANTIATE_KV_COPY_KERNEL(TYPE) \ + template void launch_kv_copy_kernel(TYPE * kv_cache, \ + TYPE * q, \ + TYPE * k, \ + TYPE * v, \ + const BatchWrapperCPP batch_desc, \ + const int qkv_stride, \ + const int kv_cache_stride, \ + const int v_offset, \ + const int q_ratio, \ + const int head_size, \ + const int n_tokens, \ + const int n_q_heads, \ + cudaStream_t stream); + +INSTANTIATE_KV_COPY_KERNEL(__half) + +#ifdef BF16_AVAILABLE +INSTANTIATE_KV_COPY_KERNEL(__nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cuh b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cuh new file mode 100644 index 000000000000..be38ff30c46c --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cuh @@ -0,0 +1,45 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" +#include "ragged_dtypes.h" + +#ifdef BF16_AVAILABLE +#include +#endif + +template +void launch_kv_rotary_kernel(T* kv_cache, + T* q, + T* k, + T* v, + T* inv_freq, + const BatchWrapperCPP batch_desc, + const int qkv_stride, + const int kv_cache_stride, + const int v_offset, + const int inv_freq_stride, + const int q_ratio, + const int head_size, + const int n_tokens, + const int n_q_heads, + cudaStream_t stream); + +template +void launch_kv_copy_kernel(T* kv_cache, + T* q, + T* k, + T* v, + const BatchWrapperCPP batch_desc, + const int qkv_stride, + const int kv_cache_stride, + const int v_offset, + const int q_ratio, + const int head_size, + const int n_tokens, + const int n_q_heads, + cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.h b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.h new file mode 100644 index 000000000000..0615825c0a21 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.h @@ -0,0 +1,63 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "blocked_kv_rotary.cuh" + +/* +Rotary position embeddings + copy into KV cache. This implementation assumes +that the inverse frequencies should be ready from global memory rather than +synthesized in the kernel. + +Arguments: + kv_cache: [n_blocks, block_size, 2, n_kv_heads, head_size] + q: [n_tokens, n_q_heads * head_size] + k: [n_tokens, n_kv_heads * head_size] + v: [n_tokens, n_kv_heads * head_size] + inv_freq: [max_seq_len, head_size // 2] +*/ +void kv_trained_rotary_embeddings(torch::Tensor& kv_cache, + torch::Tensor& q, + torch::Tensor& k, + torch::Tensor& v, + torch::Tensor& inv_freq, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs); + +/* +Rotary position embeddings + copy into KV cache. This implementation assumes +that the inverse frequencies should be synthesized in the kernel. + +Arguments: + kv_cache: [n_blocks, block_size, 2, n_kv_heads, head_size] + q: [n_tokens, n_q_heads * head_size] + k: [n_tokens, n_kv_heads * head_size] + v: [n_tokens, n_kv_heads * head_size] +*/ +void kv_rotary_embeddings(torch::Tensor& kv_cache, + torch::Tensor& q, + torch::Tensor& k, + torch::Tensor& v, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs); + +/* +Copy into linear KV cache. +*/ +void linear_kv_copy(torch::Tensor& kv_cache, + torch::Tensor& q, + torch::Tensor& k, + torch::Tensor& v, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_ptrs); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.py b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.py new file mode 100644 index 000000000000..630d58d90a23 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.py @@ -0,0 +1,70 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ....inference_utils import DtypeEnum +from deepspeed.ops.op_builder import RaggedOpsBuilder +from ....ragged import RaggedBatchWrapper +from ... import DSKernelBase + + +class BlockedRotaryEmbeddings(DSKernelBase): + """ + CUDA Kernel implementation that will perform rotary position embeddings on the queries and keys + before copying into a blocked KV cache. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_head_sizes = [64, 128] + supported_q_ratios = [1, 2, 4, 5, 8] + + def __init__(self, head_size: int, n_q_heads: int, n_kv_heads: int, dtype: torch.dtype) -> None: + """ + Args: + head_size: The size of the attention head. + q_ratio: Ratio of q heads to kv heads (for GQA) + dtype: Data type for the input/output. Supported values are torch.float16 and torch.bfloat16. + """ + + q_ratio = n_q_heads // n_kv_heads + + if head_size not in BlockedRotaryEmbeddings.supported_head_sizes: + raise ValueError("Unsupported head size: {}, supported_head_sizes are {}".format( + head_size, BlockedRotaryEmbeddings.supported_head_sizes)) + + if q_ratio not in BlockedRotaryEmbeddings.supported_q_ratios: + raise ValueError("Unsupported q_ratio: {}, supported_q_ratios are {}".format( + q_ratio, BlockedRotaryEmbeddings.supported_q_ratios)) + + if not isinstance(dtype, DtypeEnum): + dtype = DtypeEnum(dtype) + + if dtype not in BlockedRotaryEmbeddings.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + dtype, BlockedRotaryEmbeddings.supported_dtypes)) + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.kv_rotary_embeddings + self.head_size = head_size + self.n_q_heads = n_q_heads + self.n_kv_heads = n_kv_heads + + def __call__(self, kv_cache: torch.Tensor, qkv: torch.Tensor, ragged_batch: RaggedBatchWrapper) -> None: + """ + Perform rotary embeddings on the queries and keys before copying into a blocked KV cache. + + Args: + kv_cache (torch.Tensor): Pre-allocated KV cache of [num_blocks, block_size, 2, n_kv_heads, head_size] + qkv: Input tensor of shape [num_tokens, head_size * (n_q_heads + 2 * n_kv_heads)] + ragged_batch: Wrapper for the ragged batch. + """ + + q = qkv[:, :self.head_size * self.n_q_heads] + k = qkv[:, self.head_size * self.n_q_heads:self.head_size * (self.n_q_heads + self.n_kv_heads)] + v = qkv[:, self.head_size * (self.n_q_heads + self.n_kv_heads):] + + self.kernel(kv_cache, q, k, v, ragged_batch.batch_metadata_buffer(), ragged_batch.inflight_seq_descriptors(), + ragged_batch.tokens_to_seq(), ragged_batch.kv_ptrs()) diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_trained_kv_rotary.py b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_trained_kv_rotary.py new file mode 100644 index 000000000000..59da1db0f5d6 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_trained_kv_rotary.py @@ -0,0 +1,76 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ....inference_utils import DtypeEnum +from deepspeed.ops.op_builder import RaggedOpsBuilder +from ....ragged import RaggedBatchWrapper +from ... import DSKernelBase + + +class BlockedTrainedRotaryEmbeddings(DSKernelBase): + """ + CUDA Kernel implementation that will perform rotary position embeddings on the queries and keys + before copying into a blocked KV cache. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_head_sizes = [64, 128] + supported_q_ratios = [1, 2, 4, 5, 8] + + def __init__(self, head_size: int, n_q_heads: int, n_kv_heads: int, dtype: torch.dtype) -> None: + """ + Args: + head_size: The size of the attention head. + dtype: Data type for the input/output. Supported values are torch.float16 and torch.bfloat16. + """ + + q_ratio = n_q_heads // n_kv_heads + + if head_size not in BlockedTrainedRotaryEmbeddings.supported_head_sizes: + raise ValueError("Unsupported head size: {}, supported_head_sizes are {}".format( + head_size, BlockedTrainedRotaryEmbeddings.supported_head_sizes)) + + if q_ratio not in BlockedTrainedRotaryEmbeddings.supported_q_ratios: + raise ValueError("Unsupported q_ratio: {}, supported_q_ratios are {}".format( + q_ratio, BlockedTrainedRotaryEmbeddings.supported_q_ratios)) + + if not isinstance(dtype, DtypeEnum): + dtype = DtypeEnum(dtype) + + if dtype not in BlockedTrainedRotaryEmbeddings.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + dtype, BlockedTrainedRotaryEmbeddings.supported_dtypes)) + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.kv_trained_rotary_embeddings + self.head_size = head_size + self.n_q_heads = n_q_heads + self.n_kv_heads = n_kv_heads + + def __call__(self, kv_cache: torch.Tensor, qkv: torch.Tensor, ragged_batch: RaggedBatchWrapper, + inverse_freqs: torch.Tensor) -> None: + """ + Perform rotary embeddings on the queries and keys before copying into a blocked KV cache. + + Args: + kv_cache (torch.Tensor): Pre-allocated KV cache of [num_blocks, block_size, 2, n_kv_heads, head_size] + qkv: Input tensor of shape [num_tokens, head_size * (n_q_heads + 2 * n_kv_heads)] + ragged_batch: Wrapper for the ragged batch. + inverse_freqs: Inverse frequencies for the rotary embeddings. Shape [max_seq_len, head_size // 2] + """ + + q = qkv[:, :self.head_size * self.n_q_heads] + k = qkv[:, self.head_size * self.n_q_heads:self.head_size * (self.n_q_heads + self.n_kv_heads)] + v = qkv[:, self.head_size * (self.n_q_heads + self.n_kv_heads):] + + self.kernel(kv_cache, q, k, v, inverse_freqs, ragged_batch.batch_metadata_buffer(), + ragged_batch.inflight_seq_descriptors(), ragged_batch.tokens_to_seq(), ragged_batch.kv_ptrs()) diff --git a/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/linear_blocked_kv_copy.py b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/linear_blocked_kv_copy.py new file mode 100644 index 000000000000..c9f6ffd37b3e --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/linear_blocked_kv_copy.py @@ -0,0 +1,74 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ....inference_utils import DtypeEnum +from ....ragged import RaggedBatchWrapper +from deepspeed.ops.op_builder import RaggedOpsBuilder +from ... import DSKernelBase + + +class LinearBlockedKVCopy(DSKernelBase): + """ + CUDA Kernel implementation that will perform rotary position embeddings on the queries and keys + before copying into a blocked KV cache. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + supported_head_sizes = [64, 128] + supported_q_ratios = [1, 2, 4, 5, 8] + + def __init__(self, head_size: int, n_q_heads: int, n_kv_heads: int, dtype: torch.dtype) -> None: + """ + Args: + head_size: The size of the attention head. + dtype: Data type for the input/output. Supported values are torch.float16 and torch.bfloat16. + """ + + q_ratio = n_q_heads // n_kv_heads + + if head_size not in LinearBlockedKVCopy.supported_head_sizes: + raise ValueError("Unsupported head size: {}, supported_head_sizes are {}".format( + head_size, LinearBlockedKVCopy.supported_head_sizes)) + + if q_ratio not in LinearBlockedKVCopy.supported_q_ratios: + raise ValueError("Unsupported q_ratio: {}, supported_q_ratios are {}".format( + q_ratio, LinearBlockedKVCopy.supported_q_ratios)) + + if not isinstance(dtype, DtypeEnum): + dtype = DtypeEnum(dtype) + + if dtype not in LinearBlockedKVCopy.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + dtype, LinearBlockedKVCopy.supported_dtypes)) + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.linear_kv_copy + self.head_size = head_size + self.n_q_heads = n_q_heads + self.n_kv_heads = n_kv_heads + + def __call__(self, kv_cache: torch.Tensor, qkv: torch.Tensor, ragged_batch: RaggedBatchWrapper) -> None: + """ + Perform rotary embeddings on the queries and keys before copying into a blocked KV cache. + + Args: + kv_cache (torch.Tensor): Pre-allocated KV cache of [num_blocks, block_size, 2, n_kv_heads, head_size] + qkv: Input tensor of shape [num_tokens, head_size * (n_q_heads + 2 * n_kv_heads)] + ragged_batch: Wrapper for the ragged batch. + """ + + q = qkv[:, :self.head_size * self.n_q_heads] + k = qkv[:, self.head_size * self.n_q_heads:self.head_size * (self.n_q_heads + self.n_kv_heads)] + v = qkv[:, self.head_size * (self.n_q_heads + self.n_kv_heads):] + + self.kernel(kv_cache, q, k, v, ragged_batch.batch_metadata_buffer(), ragged_batch.inflight_seq_descriptors(), + ragged_batch.tokens_to_seq(), ragged_batch.kv_ptrs()) diff --git a/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/__init__.py new file mode 100644 index 000000000000..72103a0d82a1 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .logits_gather import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cpp b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cpp new file mode 100644 index 000000000000..1a7e7c0a2167 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cpp @@ -0,0 +1,45 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "logits_gather.h" + +#define DISPATCH_TO_LOGITS_GATHER(T_TYPE, C_TYPE) \ + if (all_acts.options().dtype() == torch::T_TYPE) { \ + launch_logits_gather((C_TYPE*)final_token_acts.data_ptr(), \ + (const C_TYPE*)all_acts.data_ptr(), \ + batch_metadata_raw, \ + seq_metadata_raw, \ + n_seqs, \ + embed_dim, \ + at::cuda::getCurrentCUDAStream()); \ + } + +/* +Logits gather will parse the ragged batch data structure and gather only the logits that +will be used for token sampling. +*/ +void gather_for_logits(torch::Tensor& final_token_acts, + torch::Tensor& all_acts, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata) +{ + const RaggedBatchDescriptor* batch_metadata_raw = + reinterpret_cast(batch_metadata.data_ptr()); + + const InflightSeqDescriptor* seq_metadata_raw = + reinterpret_cast(seq_metadata.data_ptr()); + + const int n_seqs = final_token_acts.size(0); + const int embed_dim = final_token_acts.size(1); + + TORCH_CHECK(all_acts.scalar_type() == final_token_acts.scalar_type(), + "all_acts and final_token_acts must have the same scalar type"); + + DISPATCH_TO_LOGITS_GATHER(kFloat, float) + DISPATCH_TO_LOGITS_GATHER(kHalf, half) +#ifdef BF16_AVAILABLE + DISPATCH_TO_LOGITS_GATHER(kBFloat16, __nv_bfloat16) +#endif +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cu b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cu new file mode 100644 index 000000000000..a539888ff904 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cu @@ -0,0 +1,86 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "ds_kernel_utils.h" +#include "logits_gather.cuh" +#include "memory_access_utils.h" +#include "ragged_dtypes.h" + +namespace logits_gather { + +constexpr int granularity = 16; +constexpr int threads = 512; + +} // namespace logits_gather + +template +__global__ void logits_gather_kernel(T* final_token_acts, + const T* token_acts, + const RaggedBatchDescriptor* ragged_batch, + const InflightSeqDescriptor* inflight_batch, + const int32_t embed_dim) +{ + constexpr int T_vector = logits_gather::granularity / sizeof(T); + + const int32_t seq_id = blockIdx.y; + + // It's possible we've padded the output Tensor (under CG conditions) + if (seq_id >= ragged_batch->n_sequences) return; + + const InflightSeqDescriptor seq = inflight_batch[seq_id]; + const int final_token_idx = seq.start_idx + seq.n_tokens - 1; + + const int token_offset = final_token_idx * embed_dim; + const int thread_offset = + threadIdx.x * T_vector + blockIdx.x * logits_gather::threads * T_vector; + + const int final_token_offset = seq_id * embed_dim; + + T reg_buf[T_vector]; + + if (thread_offset < embed_dim) { + mem_access::load_global( + reg_buf, token_acts + token_offset + thread_offset); + + mem_access::store_global( + final_token_acts + final_token_offset + thread_offset, reg_buf); + } +} + +template +void launch_logits_gather(T* final_token_acts, + const T* all_acts, + const RaggedBatchDescriptor* ragged_batch, + const InflightSeqDescriptor* inflight_batch, + const int32_t n_seqs, + const int32_t embed_dim, + cudaStream_t stream) +{ + constexpr int T_vector = logits_gather::granularity / sizeof(T); + constexpr int elems_per_block = logits_gather::threads * T_vector; + const int parallel_blocks = (embed_dim + elems_per_block - 1) / elems_per_block; + + const dim3 grid(parallel_blocks, n_seqs, 1); + const dim3 block(logits_gather::threads, 1, 1); + + logits_gather_kernel<<>>( + final_token_acts, all_acts, ragged_batch, inflight_batch, embed_dim); +} + +#define INSTANTIATE_FOR_TYPE(T) \ + template void launch_logits_gather(T * final_token_acts, \ + const T* all_acts, \ + const RaggedBatchDescriptor* ragged_batch, \ + const InflightSeqDescriptor* inflight_batch, \ + const int32_t n_seqs, \ + const int32_t embed_dim, \ + cudaStream_t stream); + +INSTANTIATE_FOR_TYPE(float) +INSTANTIATE_FOR_TYPE(__half) + +#ifdef BF16_AVAILABLE +INSTANTIATE_FOR_TYPE(__nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cuh b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cuh new file mode 100644 index 000000000000..c4e84c05e6d8 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cuh @@ -0,0 +1,22 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" +#include "ragged_dtypes.h" + +#ifdef BF16_AVAILABLE +#include +#endif + +template +void launch_logits_gather(T* final_token_acts, + const T* all_acts, + const RaggedBatchDescriptor* batch_metadata, + const InflightSeqDescriptor* seq_metadata, + const int32_t n_seqs, + const int32_t embed_dim, + cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.h b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.h new file mode 100644 index 000000000000..73a855984daa --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.h @@ -0,0 +1,20 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "logits_gather.cuh" +#include "ragged_dtypes.h" + +/* +Logits gather will parse the ragged batch data structure and gather only the logits that +will be used for token sampling. +*/ +void gather_for_logits(torch::Tensor& final_token_acts, + torch::Tensor& all_acts, + torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.py b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.py new file mode 100644 index 000000000000..64b453e9e9e3 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/logits_gather/logits_gather.py @@ -0,0 +1,52 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from deepspeed.ops.op_builder import RaggedOpsBuilder +from ....inference_utils import elem_size +from ....ragged import RaggedBatchWrapper + + +class RaggedLogitsGather(DSKernelBase): + """ + CUDA Kernel implementation for gather the hidden states of the final token + of each sequence. This is used to reduce the cost of the performing the unembedding. + """ + + supported_dtypes = [torch.float16, torch.bfloat16, torch.float32] + + def __init__(self, model_dim: int, fp_dtype: torch.dtype): + """ + Parameters: + fp_dtype (torch.dtype): Data type for the input/output. Supported values + are torch.float16, torch.bfloat16, and torch.float32. + """ + if fp_dtype not in RaggedLogitsGather.supported_dtypes: + raise ValueError("Unsupported data type: {}, supported_dtypes are {}".format( + fp_dtype, RaggedLogitsGather.supported_dtypes)) + + if elem_size(fp_dtype) * model_dim % 16 != 0: + raise ValueError("Embedding dimension must be aligned to 16 bytes, got {}".format(model_dim)) + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.gather_for_logits + + def __call__(self, final_token_activations: torch.Tensor, all_activations: torch.Tensor, + ragged_wrapper: RaggedBatchWrapper) -> torch.Tensor: + """ + Gather the hidden states of the final token of each sequence from `all_activations` into + `final_token_activations`. + + Args: + final_token_activations (torch.Tensor): Output tensor of shape [num_seqs, model_dim] + all_activations (torch.Tensor): Input tensor of shape [num_tokens, model_dim] + ragged_wrapper (RaggedBatchWrapper): Wrapper for the ragged batch. + """ + + self.kernel(final_token_activations, all_activations, ragged_wrapper.batch_metadata_buffer(), + ragged_wrapper.inflight_seq_descriptors()) + return final_token_activations diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/__init__.py new file mode 100644 index 000000000000..096c0d984a5a --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .moe_gather import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cpp b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cpp new file mode 100644 index 000000000000..e55e1f48c125 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cpp @@ -0,0 +1,53 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "moe_gather.h" +#include + +#define DISPATCH_MOE_GATHER(T_TYPE, C_TYPE) \ + if (layer_output.options().dtype() == torch::T_TYPE) { \ + launch_moe_gather((C_TYPE*)layer_output.data_ptr(), \ + (const C_TYPE*)moe_output.data_ptr(), \ + (const float*)scores.data_ptr(), \ + (const int32_t*)mapped_slots.data_ptr(), \ + (int32_t*)expert_count.data_ptr(), \ + n_channels, \ + n_experts, \ + n_tokens, \ + at::cuda::getCurrentCUDAStream()); \ + return; \ + } + +/* +Re-gather the outputs of MoE and scale them by the gating score. +*/ +void moe_gather(torch::Tensor& layer_output, + const torch::Tensor& moe_output, + const torch::Tensor& scores, + const torch::Tensor& mapped_slots, + const torch::Tensor& expert_count) +{ + const int32_t n_channels = layer_output.size(1); + const int32_t n_experts = expert_count.size(0); + const int32_t n_tokens = layer_output.size(0); + + TORCH_CHECK(moe_output.size(0) == n_tokens); + TORCH_CHECK(moe_output.size(1) == n_channels); + TORCH_CHECK(scores.size(0) == n_tokens); + TORCH_CHECK(mapped_slots.size(0) == n_tokens); + + TORCH_CHECK(layer_output.scalar_type() == moe_output.scalar_type()); + TORCH_CHECK(scores.scalar_type() == torch::kFloat32); + TORCH_CHECK(mapped_slots.scalar_type() == torch::kInt32); + TORCH_CHECK(expert_count.scalar_type() == torch::kInt32); + + DISPATCH_MOE_GATHER(kHalf, __half); + +#ifdef BF16_AVAILABLE + DISPATCH_MOE_GATHER(kBFloat16, __nv_bfloat16); +#endif + + TORCH_CHECK(false, "Unsupported data type for MoE gather"); +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cu b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cu new file mode 100644 index 000000000000..c2fae24f5080 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cu @@ -0,0 +1,122 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "conversion_utils.h" +#include "ds_kernel_utils.h" +#include "moe_gather.cuh" +#include "reduction_utils.h" +#include "top_1_gating.cuh" + +namespace gather { + +constexpr int access_granularity = 16; +constexpr int threads = 256; + +} // namespace gather + +template +__global__ void moe_gather_kernel(T* layer_output, + const T* moe_output, + const float* scores, + const int32_t* mapped_slots, + int32_t* expert_counts, + const int32_t n_channels, + const int32_t n_experts) +{ + constexpr int32_t vector_size = gather::access_granularity / sizeof(T); + constexpr int32_t stride = vector_size * gather::threads; + + const int32_t token_idx = blockIdx.x; + const int32_t mapped_slot = mapped_slots[token_idx]; + + if (token_idx == 0) { + // Reset expert counts for its next use. + if (threadIdx.x < n_experts) { expert_counts[threadIdx.x] = 0; } + } + + if (mapped_slot == gating::unassigned) { + // This token was not assigned. + // TODO(cmikeh2): It's possible we want different behavior here moving forward. + return; + } + + const float score = scores[token_idx]; + const int32_t channel_offset = threadIdx.x * vector_size; + + const T* moe_output_base = moe_output + mapped_slot * n_channels + channel_offset; + T* layer_output_base = layer_output + token_idx * n_channels + channel_offset; + +#pragma unroll + for (int i = 0; i < copyUnroll; i++) { + T reg_buffer[vector_size]; + + if (i * stride + channel_offset < n_channels) { + mem_access::load_global(reg_buffer, + moe_output_base + i * stride); + +#pragma unroll + for (int j = 0; j < vector_size; j++) { + // There are accuracy implications of downcasting the score to a 16-bit + // data type, so we up-convert the input to 32-bit, multiply, and then + // down-convert back to 16-bit. + float up_cast = conversion::to(reg_buffer[j]); + reg_buffer[j] = conversion::to(up_cast * score); + } + + mem_access::store_global(layer_output_base + i * stride, + reg_buffer); + } + } +} + +#define LAUNCH_FOR_UNROLL(COUNT) \ + case COUNT: \ + moe_gather_kernel<<>>( \ + layer_output, moe_output, scores, mapped_slots, expert_counts, n_channels, n_experts); \ + break; + +template +void launch_moe_gather(T* layer_output, + const T* moe_output, + const float* scores, + const int32_t* mapped_slots, + int32_t* expert_counts, + const int32_t n_channels, + const int32_t n_experts, + const int32_t n_tokens, + cudaStream_t stream) +{ + constexpr int vals_per_unroll = gather::threads * gather::access_granularity / sizeof(T); + const int copy_unroll = (n_channels + vals_per_unroll - 1) / vals_per_unroll; + + const dim3 block(gather::threads); + const dim3 grid(n_tokens); + + switch (copy_unroll) { + LAUNCH_FOR_UNROLL(1) + LAUNCH_FOR_UNROLL(2) + LAUNCH_FOR_UNROLL(3) + LAUNCH_FOR_UNROLL(4) + LAUNCH_FOR_UNROLL(5) + LAUNCH_FOR_UNROLL(6) + } +} + +#define INSTANTIATE_GATHER_FOR_TYPE(TYPE) \ + template void launch_moe_gather(TYPE * layer_output, \ + const TYPE* moe_output, \ + const float* scores, \ + const int32_t* mapped_slots, \ + int32_t* expert_counts, \ + const int32_t n_channels, \ + const int32_t n_experts, \ + const int32_t n_tokens, \ + cudaStream_t stream); + +INSTANTIATE_GATHER_FOR_TYPE(__half) + +#ifdef BF16_AVAILABLE +INSTANTIATE_GATHER_FOR_TYPE(__nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cuh b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cuh new file mode 100644 index 000000000000..f98a727ead58 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cuh @@ -0,0 +1,20 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" +#include "ragged_dtypes.h" + +template +void launch_moe_gather(T* layer_output, + const T* moe_output, + const float* scores, + const int32_t* mapped_slots, + int32_t* expert_counts, + const int32_t n_channels, + const int32_t n_experts, + const int32_t n_tokens, + cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.h b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.h new file mode 100644 index 000000000000..7ffe9f8b4dc6 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.h @@ -0,0 +1,19 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "moe_gather.cuh" + +/* +Re-gather the outputs of MoE and scale them by the gating score. +*/ +void moe_gather(torch::Tensor& layer_output, + const torch::Tensor& moe_output, + const torch::Tensor& scores, + const torch::Tensor& mapped_slots, + const torch::Tensor& expert_counts); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.py b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.py new file mode 100644 index 000000000000..c37683d03fbe --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_gather/moe_gather.py @@ -0,0 +1,52 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ... import DSKernelBase +from ....inference_utils import DtypeEnum +from deepspeed.ops.op_builder import RaggedOpsBuilder + + +class MoEGather(DSKernelBase): + """ + CUDA implementation of MoE gather. This will bring the tokens back + to their original indices and perform the output scaling. + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + + def __init__(self, dtype: DtypeEnum, channels: int) -> None: + + if not isinstance(dtype, DtypeEnum): + dtype = DtypeEnum(dtype) + + if dtype not in MoEGather.supported_dtypes: + raise RuntimeError(f"Unsupported dtype {dtype}") + + if channels % 8 != 0: + raise RuntimeError(f"Channels {channels} must be divisible by 8") + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.moe_gather + + def __call__(self, layer_output: torch.Tensor, moe_output: torch.Tensor, scores: torch.Tensor, + mapped_slots: torch.Tensor, expert_counts: torch.Tensor) -> torch.Tensor: + """ + Reorders the moe_output tokens into their original order and scales them by their + gating scale. This will be a no-op for padded tokens. + + Arguments: + layer_output (torch.Tensor): The output of the layer of shape [n_tokens, hidden_size]. This has been scaled appropriately. + moe_output (torch.Tensor): The output of the MoE of shape [n_tokens, hidden_size]. + scores (torch.Tensor): The gating scores of shape [n_tokens]. + mapped_slots (torch.Tensor): The index of the token in the expert's input of shape [n_tokens]. The index of token ``i`` in layer_output is ``mapped_slots[i]``. + expert_counts (torch.Tensor): The number of tokens assigned to each expert of shape [n_experts]. This is passed to fuse the clearing of this data structure into the gather. + + Returns: + layer_output + """ + self.kernel(layer_output, moe_output, scores, mapped_slots, expert_counts) + return layer_output diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/__init__.py new file mode 100644 index 000000000000..a7ca91fe5363 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .moe_scatter import * diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cpp b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cpp new file mode 100644 index 000000000000..902f1cc0ea15 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cpp @@ -0,0 +1,62 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "moe_scatter.h" +#include + +#define DISPATCH_MOE_SCATTER(T_TYPE, C_TYPE) \ + if (activations.options().dtype() == torch::T_TYPE) { \ + launch_moe_scatter((C_TYPE*)moe_input.data_ptr(), \ + (int64_t*)expert_count_cumsums.data_ptr(), \ + (int32_t*)mapped_slots.data_ptr(), \ + (const C_TYPE*)activations.data_ptr(), \ + (const int32_t*)expert_counts.data_ptr(), \ + (const int32_t*)assignments.data_ptr(), \ + (const int32_t*)offsets.data_ptr(), \ + n_channels, \ + n_tokens, \ + n_experts, \ + at::cuda::getCurrentCUDAStream()); \ + return; \ + } + +/* +Performs a cumsum on the expert counts and copies the hidden states to the +appropriate spot to ensure that each experts inputs are contiguous. +*/ +void moe_scatter(torch::Tensor& moe_input, + torch::Tensor& expert_count_cumsums, + torch::Tensor& mapped_slots, + torch::Tensor& activations, + torch::Tensor& expert_counts, + torch::Tensor& assignments, + torch::Tensor& offsets) +{ + const int32_t n_tokens = activations.size(0); + const int32_t n_channels = activations.size(1); + + // Should have a lot of matching buffer sizes here. + TORCH_CHECK(n_tokens == moe_input.size(0)); + TORCH_CHECK(n_tokens == assignments.size(0)); + TORCH_CHECK(n_tokens == offsets.size(0)); + TORCH_CHECK(n_channels == moe_input.size(1)); + + const int32_t n_experts = expert_count_cumsums.size(0); + + TORCH_CHECK(moe_input.scalar_type() == activations.scalar_type()); + TORCH_CHECK(expert_count_cumsums.scalar_type() == torch::kInt64); + TORCH_CHECK(mapped_slots.scalar_type() == torch::kInt32); + TORCH_CHECK(expert_counts.scalar_type() == torch::kInt32); + TORCH_CHECK(assignments.scalar_type() == torch::kInt32); + TORCH_CHECK(offsets.scalar_type() == torch::kInt32); + + DISPATCH_MOE_SCATTER(kHalf, __half); + +#ifdef BF16_AVAILABLE + DISPATCH_MOE_SCATTER(kBFloat16, __nv_bfloat16); +#endif + + TORCH_CHECK(false, "Unsupported dtype for moe_scatter") +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cu b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cu new file mode 100644 index 000000000000..0746cd7be645 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cu @@ -0,0 +1,208 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "ds_kernel_utils.h" +#include "moe_scatter.cuh" +#include "reduction_utils.h" +#include "top_1_gating.cuh" + +using ROp = reduce::ROpType; + +namespace scatter { + +constexpr int access_granularity = 16; +constexpr int threads = 256; +constexpr int warps = threads / hw_warp_size; + +} // namespace scatter + +template +__global__ void moe_scatter_kernel(T* moe_input, + int64_t* expert_count_cumsums, + int32_t* mapped_slots, + const T* activations, + const int32_t* assignments, + const int32_t* expert_counts, + const int32_t* offsets, + const int32_t n_channels, + const int32_t n_experts) +{ + constexpr int32_t vector_size = scatter::access_granularity / sizeof(T); + constexpr int32_t load_stride = vector_size * scatter::threads; + + const int32_t token_idx = blockIdx.x; + const int32_t tidx = threadIdx.x; + const int32_t warp_rank = tidx / hw_warp_size; + + // Bank aligned and sufficient + __shared__ int32_t red_buffer[32]; + __shared__ int32_t token_0_row; + + // CG helpers + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + + const int assigned_expert = assignments[token_idx]; + + // For the different codepaths, we'll converge on this variable for doing + // the token copy. + int32_t token_base_row; + + if (token_idx == 0) { + // Token 0 will perform a cumsum on the data + int32_t expert_vals; + if (tidx < n_experts) { + expert_vals = expert_counts[tidx]; + } else { + expert_vals = 0; + } + +#pragma unroll + for (int i = 1; i < hw_warp_size; i *= 2) { + int32_t maybe_add = warp.shfl_up(expert_vals, i); + expert_vals = (warp.thread_rank() < i) ? expert_vals : expert_vals + maybe_add; + } + + if (warp.thread_rank() == hw_warp_size - 1) { + mem_access::store_shared<4>(red_buffer + warp_rank, &expert_vals); + } + + tb.sync(); + + int32_t phase_2_val = 0; + if (warp.thread_rank() < scatter::warps) { + mem_access::load_shared<4>(&phase_2_val, red_buffer + warp.thread_rank()); + } + +#pragma unroll + for (int i = 1; i < hw_warp_size; i *= 2) { + int32_t maybe_add = warp.shfl_up(phase_2_val, i); + phase_2_val = (warp.thread_rank() < i) ? phase_2_val : phase_2_val + maybe_add; + } + + int warp_offset = 0; + if (warp_rank > 0) { warp_offset = warp.shfl(phase_2_val, warp_rank - 1); } + const int32_t expert_cumsum = warp_offset + expert_vals; + + if (tidx < n_experts) { + int64_t expert_cumsum_64 = (int64_t)expert_cumsum; + expert_count_cumsums[tidx] = expert_cumsum_64; + } + + if (assigned_expert == gating::unassigned) return; + if (assigned_expert - 1 == tidx) token_0_row = expert_cumsum; + + tb.sync(); + + if (assigned_expert != 0) { + token_base_row = token_0_row; + } else { + token_base_row = 0; + } + + } else if (assigned_expert == gating::unassigned) { + // For whatever reason, don't need to perform the copy, so we'll early return + // and signal this wasn't mapped with a negative 1. + if (tidx == 0) mapped_slots[token_idx] = gating::unassigned; + return; + } else { + // For all other valid tokens, we can just do a block-scoped sum. + if (tidx < assigned_expert) { + token_base_row = expert_counts[tidx]; + } else { + token_base_row = 0; + } + + warp.sync(); + + // TODO(cmikeh2): Shouldn't use the internal api. + reduce::_block(tb, warp, &token_base_row); + } + + // Data copy to appropriate location + const int32_t thread_offset = tidx * vector_size; + + const int32_t base_load_offset = token_idx * n_channels + thread_offset; + const T* load_base_ptr = activations + base_load_offset; + + const int32_t store_row = token_base_row + offsets[token_idx]; + const int32_t base_store_offset = store_row * n_channels + thread_offset; + T* store_base_ptr = moe_input + base_store_offset; + +#pragma unroll + for (int i = 0; i < copyUnroll; i++) { + T tmp_buf[vector_size]; + + if (i * load_stride + thread_offset < n_channels) { + mem_access::load_global(tmp_buf, + load_base_ptr + i * load_stride); + mem_access::store_global(store_base_ptr + i * load_stride, + tmp_buf); + } + } + + if (threadIdx.x == 0) { mapped_slots[token_idx] = store_row; } +} + +#define LAUNCH_FOR_UNROLL(COUNT) \ + case COUNT: \ + moe_scatter_kernel<<>>(moe_input, \ + expert_count_cumsums, \ + mapped_slots, \ + activations, \ + assignments, \ + expert_counts, \ + offsets, \ + n_channels, \ + n_experts); \ + break; + +template +void launch_moe_scatter(T* moe_input, + int64_t* expert_count_cumsums, + int32_t* mapped_slots, + const T* activations, + const int32_t* expert_counts, + const int32_t* assignments, + const int32_t* offsets, + const int32_t n_channels, + const int32_t n_tokens, + const int32_t n_experts, + cudaStream_t stream) +{ + constexpr int vals_per_unroll = scatter::threads * scatter::access_granularity / sizeof(T); + const int copy_unroll = (n_channels + vals_per_unroll - 1) / vals_per_unroll; + + const dim3 block(scatter::threads); + const dim3 grid(n_tokens); + + switch (copy_unroll) { + LAUNCH_FOR_UNROLL(1); + LAUNCH_FOR_UNROLL(2); + LAUNCH_FOR_UNROLL(3); + LAUNCH_FOR_UNROLL(4); + LAUNCH_FOR_UNROLL(5); + LAUNCH_FOR_UNROLL(6); + } +} + +#define INSTANTIATE_SCATTER_FOR_TYPE(TYPE) \ + template void launch_moe_scatter(TYPE*, \ + int64_t*, \ + int32_t*, \ + const TYPE*, \ + const int32_t*, \ + const int32_t*, \ + const int32_t*, \ + const int32_t, \ + const int32_t, \ + const int32_t, \ + cudaStream_t); + +INSTANTIATE_SCATTER_FOR_TYPE(__half); + +#ifdef BF16_AVAILABLE +INSTANTIATE_SCATTER_FOR_TYPE(__nv_bfloat16); +#endif diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cuh b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cuh new file mode 100644 index 000000000000..5c94cb0ef734 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cuh @@ -0,0 +1,22 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" +#include "ragged_dtypes.h" + +template +void launch_moe_scatter(T* moe_input, + int64_t* expert_count_cumsums, + int32_t* mapped_slots, + const T* activations, + const int32_t* expert_counts, + const int32_t* assignments, + const int32_t* offsets, + const int32_t n_channels, + const int32_t n_tokens, + const int32_t n_experts, + cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.h b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.h new file mode 100644 index 000000000000..59597f63d123 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.h @@ -0,0 +1,23 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "moe_scatter.cuh" +#include "ragged_dtypes.h" + +/* +Performs a cumsum on the expert counts and copies the hidden states to the +appropriate spot to ensure that each experts inputs are contiguous. +*/ +void moe_scatter(torch::Tensor& moe_input, + torch::Tensor& expert_count_cumsums, + torch::Tensor& mapped_slots, + torch::Tensor& activations, + torch::Tensor& expert_counts, + torch::Tensor& assignments, + torch::Tensor& offsets); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.py b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.py new file mode 100644 index 000000000000..5cd6ae5f0fe2 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.py @@ -0,0 +1,55 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from typing import Tuple + +from ... import DSKernelBase +from ....inference_utils import DtypeEnum +from deepspeed.ops.op_builder import RaggedOpsBuilder + + +class MoEScatter(DSKernelBase): + """ + CUDA implementation of MoE scatter + """ + + supported_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16] + + def __init__(self, dtype: DtypeEnum, channels: int) -> None: + + if not isinstance(dtype, DtypeEnum): + dtype = DtypeEnum(dtype) + + if dtype not in MoEScatter.supported_dtypes: + raise RuntimeError(f"Unsupported dtype {dtype}") + + if channels % 8 != 0: + raise RuntimeError(f"Channels {channels} must be divisible by 8") + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.moe_scatter + + def __call__(self, moe_input: torch.Tensor, expert_cumsum: torch.Tensor, mapped_slots: torch.Tensor, + activations: torch.Tensor, expert_counts: torch.Tensor, assignments: torch.Tensor, + offsets: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]: + """ + Scatters the hidden states such that the token stride for each expert's input is contiguous. + + Arguments: + moe_input (torch.Tensor): The direct input for the MoE GEMM of shape [n_tokens, hidden_size]. + expert_cumsum (torch.Tensor): The cumulative sum of the expert counts of shape [n_experts]. + mapped_slots (torch.Tensor): The index of the token in the expert's input of shape [n_tokens]. + hidden_states (torch.Tensor): The hidden states of shape [n_tokens, hidden_size]. + expert_counts (torch.Tensor): The number of tokens assigned to each expert of shape [n_experts]. + assignments (torch.Tensor): The expert assignments of shape [n_tokens]. + offsets (torch.Tensor): The offsets into the expert for a given token of shape [n_tokens]. + + Returns: + Tuple[torch.Tensor, torch.Tensor, torch.Tensor]: The MoE input (with scattered values), the cumsum of the offsets (for the MoE kernels themselves), and the assignments Tensor modified in place to show which row that token was mapped to in the input. + """ + self.kernel(moe_input, expert_cumsum, mapped_slots, activations, expert_counts, assignments, offsets) + return moe_input, expert_cumsum, mapped_slots diff --git a/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_dtypes.h b/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_dtypes.h new file mode 100644 index 000000000000..7876b354af0d --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_dtypes.h @@ -0,0 +1,48 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include + +struct +#ifdef __CUDA_CC__ + __align__(8) +#endif +{ + int32_t n_tokens; + int32_t n_sequences; +} +typedef RaggedBatchDescriptor; + +struct +#ifdef __CUDA_CC__ + __align__(16) +#endif +{ + int32_t start_idx; + int32_t n_tokens; + int32_t seen_tokens; + int32_t UNUSED; // Explicit padding to match the Python code pattern. +} +typedef InflightSeqDescriptor; + +struct +#ifdef __CUDA_CC__ + __align__(8) +#endif +{ + int32_t** block_lists; + int32_t block_size; + int32_t n_blocks; +} +typedef KVCacheDescriptor; + +struct { + const RaggedBatchDescriptor* batch_metadata; // Offset 0 + const InflightSeqDescriptor* seq_metadata; // Offset 8 + const int32_t* tokens_to_seq; // Offset 16 + const KVCacheDescriptor kv_desc; // Offset 24 +} typedef BatchWrapperCPP; diff --git a/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.cpp b/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.cpp new file mode 100644 index 000000000000..a6cb7f275366 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.cpp @@ -0,0 +1,28 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "ragged_kernel_helpers.h" + +BatchWrapperCPP make_cpp_batch_wrapper(torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_cache_desc, + int32_t block_size, + int32_t n_blocks) +{ + const RaggedBatchDescriptor* batch_metadata_raw = + reinterpret_cast(batch_metadata.data_ptr()); + + const InflightSeqDescriptor* seq_metadata_raw = + reinterpret_cast(seq_metadata.data_ptr()); + + const int32_t* tokens_to_seq_raw = tokens_to_seq.data_ptr(); + + int32_t** kv_ptrs_raw = reinterpret_cast(kv_cache_desc.data_ptr()); + KVCacheDescriptor kv_desc = {kv_ptrs_raw, block_size, n_blocks}; + + BatchWrapperCPP wrapper = {batch_metadata_raw, seq_metadata_raw, tokens_to_seq_raw, kv_desc}; + return wrapper; +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.h b/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.h new file mode 100644 index 000000000000..7ce082d31853 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.h @@ -0,0 +1,16 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include "ragged_dtypes.h" + +BatchWrapperCPP make_cpp_batch_wrapper(torch::Tensor& batch_metadata, + torch::Tensor& seq_metadata, + torch::Tensor& tokens_to_seq, + torch::Tensor& kv_cache_desc, + int32_t block_size, + int32_t n_blocks); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/ragged_ops.cpp b/deepspeed/inference/v2/kernels/ragged_ops/ragged_ops.cpp new file mode 100644 index 000000000000..1c09fc52bbb1 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/ragged_ops.cpp @@ -0,0 +1,48 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include + +#include "atom_builder.h" +#include "blocked_flash.h" +#include "blocked_kv_rotary.h" +#include "embed.h" +#include "logits_gather.h" +#include "moe_gather.h" +#include "moe_scatter.h" +#include "top_1_gating.h" + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) +{ + // atom_builder.h + m.def("build_atoms", &build_atoms, "Host kernel for building the atoms."); + + // blocked_flash.h + m.def("flash_attn_by_atoms", + &flash_attn_by_atoms, + "Blocked flash attention scheduled with atoms"); + + // blocked_kv_rotary.h + m.def("kv_rotary_embeddings", &kv_rotary_embeddings, "KV rotary embedding for blocked KV"); + m.def("kv_trained_rotary_embeddings", + &kv_trained_rotary_embeddings, + "KV rotary embeddings for blocked KV"); + m.def("linear_kv_copy", &linear_kv_copy, "Linear copy for blocked KV"); + + // embed.h + m.def("ragged_embed", &ragged_embed, "Embedding lookup for ragged batch"); + + // logits_gather.h + m.def("gather_for_logits", &gather_for_logits, "Sparse gather from ragged batch"); + + // moe_gather.h + m.def("moe_gather", &moe_gather, "MoE gather for top-1-gating."); + + // moe_scatter.h + m.def("moe_scatter", &moe_scatter, "MoE scatter for top-1-gating."); + + // top_1_gating.h + m.def("top_1_gating", &top_1_gating, "Top-1 gating for MoE with ragged batch awareness."); +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/__init__.py b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/__init__.py new file mode 100644 index 000000000000..b50a0838d9f8 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .top_1_gating import RaggedTop1Gating diff --git a/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cpp b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cpp new file mode 100644 index 000000000000..55c68454b228 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cpp @@ -0,0 +1,55 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "top_1_gating.h" +#include + +#define DISPATCH_TOP_1_GATING(T_TYPE, C_TYPE) \ + if (logits.options().dtype() == torch::T_TYPE) { \ + launch_top_1_gating((int32_t*)expert_counts.data_ptr(), \ + (float*)scores.data_ptr(), \ + (int32_t*)assignments.data_ptr(), \ + (int32_t*)offsets.data_ptr(), \ + (const C_TYPE*)logits.data_ptr(), \ + batch_metadata_ptr, \ + n_tokens, \ + n_experts, \ + at::cuda::getCurrentCUDAStream()); \ + return; \ + } + +/* +Perform softmax plus atomics in order to do first pass of top_1_gating. +*/ +void top_1_gating(torch::Tensor& expert_counts, + torch::Tensor& scores, + torch::Tensor& assignments, + torch::Tensor& offsets, + torch::Tensor& logits, + torch::Tensor& batch_metadata) +{ + const int32_t n_tokens = scores.size(0); + + // Should have the same buffer size for scores and offsets + TORCH_CHECK(n_tokens == offsets.size(0)); + TORCH_CHECK(n_tokens == logits.size(0)); + + TORCH_CHECK(expert_counts.scalar_type() == torch::kInt32); + TORCH_CHECK(scores.scalar_type() == torch::kFloat); + TORCH_CHECK(assignments.scalar_type() == torch::kInt32); + TORCH_CHECK(offsets.scalar_type() == torch::kInt32); + + const int32_t n_experts = logits.size(1); + const RaggedBatchDescriptor* batch_metadata_ptr = + reinterpret_cast(batch_metadata.data_ptr()); + + DISPATCH_TOP_1_GATING(kFloat, float) + DISPATCH_TOP_1_GATING(kHalf, __half) +#ifdef BF16_AVAILABLE + DISPATCH_TOP_1_GATING(kBFloat16, __nv_bfloat16) +#endif + + TORCH_CHECK(false, "Unsupported dtype for logits in top_1_gating"); +} diff --git a/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cu b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cu new file mode 100644 index 000000000000..02daee9f692e --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cu @@ -0,0 +1,106 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "conversion_utils.h" +#include "memory_access_utils.h" +#include "reduction_utils.h" +#include "top_1_gating.cuh" + +using ROp = reduce::ROpType; + +template +__global__ void top_1_gating_kernel(int32_t* expert_counts, + float* scores, + int32_t* assignments, + int32_t* offsets, + const T* logits, + const RaggedBatchDescriptor* batch_metadata, + const int32_t n_experts) +{ + const int32_t token_idx = blockIdx.x; + const int32_t expert_idx = threadIdx.x; + const int32_t max_warps = 1024 / hw_warp_size; + + // CG helpers + cg::thread_block tb = cg::this_thread_block(); + cg::thread_block_tile warp = cg::tiled_partition(tb); + + // Padding tokens do not require + if (token_idx >= batch_metadata->n_tokens) { + if (threadIdx.x == 0) { + offsets[token_idx] = gating::unassigned; + assignments[token_idx] = gating::unassigned; + } + return; + } + + const T* token_logits = logits + token_idx * n_experts; + + float logit_val; + if (expert_idx < n_experts) { + logit_val = conversion::to(token_logits[expert_idx]); + } else { + reduce::init(&logit_val); + } + + // Training code tends to use ``torch.argmax`` to select the expert, which + // which has ties broken by the lower index. Since our fused comparison algorithm + // breaks ties by the higher index (since it's the lower 32-bits of the 64-bit + // comparison), we invert the expert index to break ties by the lower index. + int32_t inverted_expert = n_experts - expert_idx - 1; + // Perform softmax + const reduce::IdxReduceResult res = + reduce::idx_reduce(tb, warp, logit_val, inverted_expert); + // Recover the original expert index + const int32_t assigned_expert = n_experts - res.idx - 1; + const float max_logit = res.val; + + float softmax_sum = __expf(logit_val - max_logit); + reduce::block(tb, warp, softmax_sum); + + // Compute the score + const float score = __expf(max_logit - max_logit) / softmax_sum; + + if (threadIdx.x == 0) { + scores[token_idx] = score; + assignments[token_idx] = assigned_expert; + offsets[token_idx] = atomicAdd(expert_counts + assigned_expert, 1); + } +} + +template +void launch_top_1_gating(int32_t* expert_counts, + float* scores, + int32_t* assignments, + int32_t* offsets, + const T* logits, + const RaggedBatchDescriptor* batch_metadata, + const int32_t n_tokens, + const int32_t n_experts, + cudaStream_t stream) +{ + const dim3 grid(n_tokens); + const dim3 block(((n_experts + hw_warp_size - 1) / hw_warp_size) * hw_warp_size); + + top_1_gating_kernel<<>>( + expert_counts, scores, assignments, offsets, logits, batch_metadata, n_experts); +} + +#define INSTANTIATE_TOP_1_KERNEL(T) \ + template void launch_top_1_gating(int32_t * expert_counts, \ + float* scores, \ + int32_t* assignments, \ + int32_t* offsets, \ + const T* logits, \ + const RaggedBatchDescriptor* batch_metadata, \ + const int32_t n_tokens, \ + const int32_t n_experts, \ + cudaStream_t stream); + +INSTANTIATE_TOP_1_KERNEL(float) +INSTANTIATE_TOP_1_KERNEL(__half) +#ifdef BF16_AVAILABLE +INSTANTIATE_TOP_1_KERNEL(__nv_bfloat16) +#endif diff --git a/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cuh b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cuh new file mode 100644 index 000000000000..c83ad56ff2f1 --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cuh @@ -0,0 +1,24 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" +#include "ragged_dtypes.h" + +namespace gating { +constexpr int unassigned = -1; +} // namespace gating + +template +void launch_top_1_gating(int32_t* expert_counts, + float* scores, + int32_t* assignments, + int32_t* offsets, + const T* logits, + const RaggedBatchDescriptor* batch_metadata, + const int32_t n_tokens, + const int32_t n_experts, + cudaStream_t stream); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.h b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.h new file mode 100644 index 000000000000..b431f4cad30c --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.h @@ -0,0 +1,21 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include +#include +#include "ragged_dtypes.h" +#include "top_1_gating.cuh" + +/* +Perform softmax plus atomics to get token mapping. +*/ +void top_1_gating(torch::Tensor& expert_counts, + torch::Tensor& scores, + torch::Tensor& assignments, + torch::Tensor& offsets, + torch::Tensor& logits, + torch::Tensor& batch_metadata); diff --git a/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.py b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.py new file mode 100644 index 000000000000..1df97c2e9f8d --- /dev/null +++ b/deepspeed/inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.py @@ -0,0 +1,59 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from typing import Tuple + +from ... import DSKernelBase +from ....inference_utils import DtypeEnum +from ....ragged import RaggedBatchWrapper +from deepspeed.ops.op_builder import RaggedOpsBuilder + + +class RaggedTop1Gating(DSKernelBase): + """ + CUDA implementation of top-1 gating. This will perform a softmax on the logits, + and return the scale as well as its idx within that expert's allocation. + """ + + supported_logit_dtypes = [DtypeEnum.fp16, DtypeEnum.bf16, DtypeEnum.fp32] + + def __init__(self, logit_dtype: DtypeEnum) -> None: + + if not isinstance(logit_dtype, DtypeEnum): + logit_dtype = DtypeEnum(logit_dtype) + + if logit_dtype not in RaggedTop1Gating.supported_logit_dtypes: + raise RuntimeError(f"Unsupported logit dtype {logit_dtype}") + + inf_module = RaggedOpsBuilder().load() + self.kernel = inf_module.top_1_gating + + def __call__(self, expert_counts: torch.Tensor, scores: torch.Tensor, assignments: torch.Tensor, + offsets: torch.Tensor, logits: torch.Tensor, + batch: RaggedBatchWrapper) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]: + """ + Perform the ragged top_1_gating. + + Arguments: + expert_counts (torch.Tensor): Tensor of 0s of shape [n_experts] to be filled with + number of tokens assigned to each expert. This must be filled with 0s else + the copy kernel will buffer overflow. In order to minimize the zero-fill cost, + it is recommended to write to 0 during the MoE output remapping. + scores (torch.Tensor): Preallocated output of shape [n_tokens] to place expert scaling + value. + expert_assignment (torch.Tensor): Preallocated output of shape [n_tokens] to place + which expert a token has been assigned to. + expert_offset (torch.Tensor): Preallocated output of shape [n_tokens] to place which + offset within an experts group a token is. + logits (torch.Tensor): Raw logits of gating function. + batch (RaggedBatchWrapper): Batch information for ragged tensor. + + Returns: + tuple of (expert_counts, scores, expert_assignment, expert_offset) + """ + self.kernel(expert_counts, scores, assignments, offsets, logits, batch.batch_metadata_buffer()) + return expert_counts, scores, assignments, offsets diff --git a/deepspeed/inference/v2/logging.py b/deepspeed/inference/v2/logging.py new file mode 100644 index 000000000000..77afe351cbea --- /dev/null +++ b/deepspeed/inference/v2/logging.py @@ -0,0 +1,26 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import logging + +from deepspeed.utils.logging import LoggerFactory + +inf_logger = None + + +def inference_logger(level: int = logging.INFO) -> logging.Logger: + """ + Create the inference logger. NOTE: Logging is not cost free. On a 3960X, + there is a cost of about 6 us per call to a no-op logger, so this should + be used during setup only and not during the inference loop. + + Args: + level (int, optional): The logging level. Defaults to logging.INFO. + """ + global inf_logger + if inf_logger is None: + inf_logger = LoggerFactory.create_logger(name="DS-Inference", level=level) + inf_logger.debug("Inference logger created.") + return inf_logger diff --git a/deepspeed/inference/v2/model_implementations/AddingAModel.md b/deepspeed/inference/v2/model_implementations/AddingAModel.md new file mode 100644 index 000000000000..8fe27297080b --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/AddingAModel.md @@ -0,0 +1,84 @@ +# Adding Support for a New Model in DeepSpeed Inference V2 + +Adding supoprt for a new model in DeepSpeed Inference requires developing three related components: +- Containers: These describe the parameters contained in the model +- Model implementation: How should the model be computed. +- Policy: The map for adding parameters to your containers and creating the model implementation. + +In this tutorial, we will assume that you'd like to use a relatively traditionally styled Transformer model and will be able to inherit from `DSTransformerModelBase` and can take advantage of the utilities that provides. + +## Defining Your Containers + +A container is the bridge between the original model's parameters and how to transform them to serve them for inference. For a model implementation, there are two primary kinds of containers: transformer containers and non-transformer containers. A transformer container consists of the parameters for a single Transformer layer in the model. So this includes your traditional parameters like the projections for the fully connected network, or query-key-value projections. The non-transformer container will contain basically everything else! However, before defining these containers, we need to understand how to define an individual parameter. + +In DeepSpeed inference, the original model parameters are populated into the model and mapped as dependencies to a parameter. A `Parameter` has two primary components: its dependencies and its `finalize` method. Let's do an example. In Llama models, the native format is for the `query`, `key`, and `value` projections to be performed independently. However, we can achieve higher throughput by fusing them into a single larger projection. We can define this fusion with a parameter: + +```python +from deepspeed.inference.module_implementations.parameter_base import ParameterBase + +class UnfusedQKVParameter(ParameterBase): + query: torch.Tensor + key: torch.Tensor + value: torch.Tensor + + def finalize(self) -> torch.Tensor: + fused_param = torch.cat([self.query, self.key, self.value], dim=0) + return self.inference_model.transform_qkv_param(fused_param) +``` + +Let's walk through each part of this implementation. First, parameters should inherit from `ParameterBase`. This will allow it to automatically determine when its dependencies are met and set the appropriate components of a parent `LayerContainer`. The second key component is the type annotations on the class itself. Each type annotation represents a dependency of the parameter. Since the original Llama mode has separate query, key, and value dependencies, our fused parameter will declare dependencies for each. Finally, we have the `finalize` method. This method is automatically called once all dependencies on the layer are met and should return the final parameter. + +In this `finalize` method, we are doing two things: the first is the act of fusing the parameters together through the concatenate method. Note that each of the dependencies can be accessed via `self.{name}`. The second is calling `self.inference_model.transform_qkv_param`. A parameter's finalize method always has access to the inference model. In this case we are using that to use a feature provided by `DSTransformerBase`. This method will automatically shard the parameter for tensor parallelism and then pass it to the linear module implementation to perform additional optimizations or shape transformations, like quantization. + +Since many patterns are very common in Transformer models, `model_implementations.common_parameters` provides implementations for many of the patterns (all compatible with `DSTransformerBase`) to help accelerate development. + +Once all parameters are created, we need to compose them into a layer container. In our simplified Llama model, let's assume there's only QKV and attention output projection matrices. A layer container would appear as the following: + +```python +from deepspeed.inference.module_implementations.layer_container_base import LayerContainer + +class ExampleContainer(LayerContainer): + qkvw: UnfusedQKVParameter + + attn_o: AttentionOutputParameter + + PARAM_MAPPING: { + "self_attn.q_proj.weight": "qkvw.query", + "self_attn.k_proj.weight": "qkvw.key", + "self_attn.v_proj.weight": "qkvw.value", + "self_attn.o_proj.weight": "attn_o.params", + } +``` + +Once again, we have a couple of key components. The first are parameter type annotations. Each annotation corresponds to a parameter that can be used in the model implementation. In the model implementation, I can simply write `container.qkvw` to access my fused and transformed QKV parameter. The second key component is the `PARAM_MAPPING` dictionary. This is our explicit mapping of the names of parameters in the source model to a parameter dependency. This mapping dictionary will be used by the policy to automatically populate dependencies. + +Once you have written `LayerContainer`s for both the transformer and non-transformer parameters, it's time to work on the model implementation! + +## Building a Model Implementation that Inherits from `DSTransformerBase` + +By inheriting from `DSTransformerBase`, most of the implementation work for sharding and transforming parameters will be automatically handled for you. However, there are four key tasks that still need to be completed. + +1. Defining the abstract properties based on your model configuration. +2. Configuring embedding and unembedding modules and the forward implementations for them. +3. Configuring the attention configuration and desired KV cache behaviors. +4. Writing the forward implementation for your layer. + +## Writing a Policy + +The `InferenceV2Policy` is the level of composition. This is the object that will be passed directly to the inference engine and will compose the model implementation and your containers to create an end-to-end solution. There are two main components to be implemented: the first is to create the model that you defined earlier. This is done by implementing the `instantiate_model` method of the policy. In general, this can just be implemented by calling the constructor for your model and passing the engine config, tensor-parallel communication object, and your custom model config. + +The second component is to define how the parameters from the checkpoint will map to each container. From the section on `LayerContainer`s above, you may remember that the `LayerContainer` can handle the internal routing of a checkpoint parameter to its dependency. In order to find the correct `LayerContainer` though, we need a second abstraction: the `ContainerMap`. + +A `ContainerMap` performs this mapping by categorizing checkpoint prefix strings to the type of container they map to. Typically, the easiest way to do this is through iterating over a model checkpoint's state dict or by iterating over the `named_parameters` of a PyTorch model. There are three types of mappings to define: the transformer mappings, the non-transformer mappings, and the what we'll call the rest. Let's work through an example: + +```python +from deepspeed.inference.module_implementations.inference_policy_base import ContainerMap + +def build_container_map(self) -> ContainerMap: + map = ContainerMap() + + transformer_containers = [MyTransformerContainer(self.model) for _ in range(self.model.num_layers)] + map.set_transformer_params("model.layers", transformer_containers) + + non_transformer_container = MyNonTransformerContainer(self.model) +``` diff --git a/deepspeed/inference/v2/model_implementations/__init__.py b/deepspeed/inference/v2/model_implementations/__init__.py new file mode 100644 index 000000000000..a3481023a8fd --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/__init__.py @@ -0,0 +1,9 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .inference_model_base import DSInferenceModelBase +from .inference_transformer_base import DSTransformerModelBase, DSMoETransformerModelBase +from .inference_policy_base import InferenceV2Policy, ContainerMap +from .sharding import * diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/__init__.py b/deepspeed/inference/v2/model_implementations/common_parameters/__init__.py new file mode 100644 index 000000000000..60963011cd66 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/__init__.py @@ -0,0 +1,13 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .attn_output_parameters import * +from .embedding_parameters import * +from .mlp_parameters import * +from .moe_parameters import * +from .norm_parameters import * +from .qkv_parameters import * +from .unembed_parameters import * +from .invfreq_parameters import * diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/attn_output_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/attn_output_parameters.py new file mode 100644 index 000000000000..f220cf7a7125 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/attn_output_parameters.py @@ -0,0 +1,29 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +""" +Common Attention Output Parameter Patterns +""" + + +class AttentionOutputParameter(ParameterBase): + """ + Attention output parameter container. + + Note: The differentiation for something like GQA for this matrix is primarily + encompassed in the sharding logic, which is currently expected to be performed by + the model implementation. + """ + + params: torch.Tensor + """ + Unsharded attention output parameter of shape [model_dim, model_dim] + """ + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform_attn_out_param(self.params) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/embedding_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/embedding_parameters.py new file mode 100644 index 000000000000..4babc0ee0127 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/embedding_parameters.py @@ -0,0 +1,29 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +from ...allocator import on_device +""" +Embedding containers. +""" + + +class EmbeddingParameter(ParameterBase): + """ + Embedding container. This should be safe to use for all types of embeddings (i.e. word, position, + and token type). + """ + + params: torch.Tensor + """ + Vocabulary parameter of shape [vocab_size, model_dim]. + """ + + @on_device + def finalize(self) -> torch.Tensor: + return self.params + #return self.inference_model.transform_embed_param(self.params) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/invfreq_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/invfreq_parameters.py new file mode 100644 index 000000000000..3a5a7fb04b9a --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/invfreq_parameters.py @@ -0,0 +1,21 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +from ...allocator import on_device +""" +Common InvFreq Parameter Patterns +""" + + +class InvFreqParameter(ParameterBase): + + params: torch.Tensor + + @on_device + def finalize(self) -> torch.Tensor: + return self.params.to(self.inference_model.activation_dtype.value) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/mlp_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/mlp_parameters.py new file mode 100644 index 000000000000..ddb8996e03a3 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/mlp_parameters.py @@ -0,0 +1,81 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +""" +MLP Parameter Containers +""" + + +class MLP1Parameter(ParameterBase): + """ + First MLP projection weight container. This performs a straight pass-through to the + model implementation for transformation. + """ + params: torch.Tensor + + def finalize(self) -> torch.Tensor: + # NOTE(cmikeh2): If we are gated but not in the format specified below, we should trigger a permutation here. + # I am not currently aware of any models that use this format (or how we should even detect it; probably should + # just be a different param entirely, but until then we'll just assume the format is correct). + return self.inference_model.transform_mlp_1_param(self.params) + + +class GatedMLPParameter(ParameterBase): + """ + Gated MLP projection container. + """ + + gate_params: torch.Tensor + """ + Weight parameter for the gating matrix. + """ + + up_params: torch.Tensor + """ + For lack of a better name, the non-gating weight parameters. + """ + + def finalize(self) -> torch.Tensor: + """ + Our gated format (this is different from InferenceV1!) is to have the gate and activated neurons + interleaved. So if we have 4 output neurons (two effective neurons) with 4 input neurons, the finalized + parameter will look like: + [g0_0, g0_1, g0_2, g0_3] + [a0_0, a0_1, a0_2, a0_3] + [g1_0, g1_1, g1_2, g1_3] + [a1_0, a1_1, a1_2, a1_3] + + As a reference, in inference v1, the format is: + [g0_0, g0_1, g0_2, g0_3] + [g1_0, g1_1, g1_2, g1_3] + [a0_0, a0_1, a0_2, a0_3] + [a1_0, a1_1, a1_2, a1_3] + """ + assert self.gate_params.shape[0] == self.up_params.shape[ + 0], "Gated MLP parameters must have the same number of neurons." + total_neurons = self.gate_params.shape[0] + self.up_params.shape[0] + + # flip the order if even with the correct tokenizer we get wrong output + #fused_param = torch.cat([self.up_params, self.gate_params], dim=-1).reshape(total_neurons, -1) + fused_param = torch.cat([self.gate_params, self.up_params], dim=-1).reshape(total_neurons, -1) + return self.inference_model.transform_mlp_1_param(fused_param) + + +class MLP2Parameter(ParameterBase): + """ + Second MLP projection weight container. This performs a straight pass-through to the + model implementation for transformation. + """ + + params: torch.Tensor + """ + Full weight parameter. + """ + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform_mlp_2_param(self.params) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/moe_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/moe_parameters.py new file mode 100644 index 000000000000..ae95e628b779 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/moe_parameters.py @@ -0,0 +1,71 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...allocator import on_device +from ...model_implementations.parameter_base import ParameterBase, ParamList +""" +Moe Parameters + +These parameters are compatible with any model inheriting from ``DSMoETransformerModelBase``. +""" + + +class MoEGatingWeightParameter(ParameterBase): + """ + Gating weight matrix. + """ + + params: torch.Tensor + """ + Projection matrix from the input activations to the gate logits. + """ + + @on_device + def finalize(self) -> torch.Tensor: + return self.inference_model.transform_moe_gate_param(self.params) + + +class UnfusedMoEMLP1Parameter(ParameterBase): + """ + This container should be used when the experts are held in separate parameters + and need to be joined into a single group. + """ + + experts: ParamList("num_experts") # noqa: F821 + + def finalize(self) -> torch.Tensor: + stacked_experts = torch.stack([p for p in self.experts], dim=0) + return self.inference_model.transform_moe_mlp_1_param(stacked_experts) + + +class UnfusedMoEMLP2Parameter(ParameterBase): + """ + This container should be used when the experts are held in separate parameters + and need to be joined into a single group. + """ + + experts: ParamList("num_experts") # noqa: F821 + + def finalize(self) -> torch.Tensor: + stacked_experts = torch.stack([p for p in self.experts], dim=0) + return self.inference_model.transform_moe_mlp_2_param(stacked_experts) + + +class UnfusedMoEGatedMLPParameter(ParameterBase): + """ + MoE Parameter for a gated activation function in which the gating matrix is not + fused in the same parameter as the non-gating matrix. + """ + + gating_experts: ParamList("num_experts") # noqa: F821 + + up_experts: ParamList("num_experts") # noqa: F821 + + def finalize(self) -> torch.Tensor: + fused_params = [torch.cat([gate, weight], dim=0) for gate, weight in zip(self.gating_experts, self.up_experts)] + stacked_params = torch.stack(fused_params, dim=0) + return self.inference_model.transform_moe_mlp_2_param(stacked_params) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/norm_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/norm_parameters.py new file mode 100644 index 000000000000..81ffcc3221df --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/norm_parameters.py @@ -0,0 +1,22 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +""" +Common Attention Output Parameter Patterns +""" + + +class NormParameter(ParameterBase): + """ + Simple normalization container. + """ + + params: torch.Tensor + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform_norm_param(self.params) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/qkv_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/qkv_parameters.py new file mode 100644 index 000000000000..2ed8a8654f5b --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/qkv_parameters.py @@ -0,0 +1,116 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +""" +Common QKV Parameter Patterns +""" + + +class FusedQKVParameter(ParameterBase): + """ + Traditional fused QKV parameters for QKV projection. This is functionally + a direct copy. + + src_qkv_w shape: [3 * out_features, in_features] + qkv_w shape: [3 * out_features, in_features] + """ + + params: torch.Tensor + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform_qkv_param(self.params) + + +class UnfusedQKVParameter(ParameterBase): + """ + QKV parameter container for unfused QKV projection. + + src_param shapes: 3 x [out_features, in_features] + dst_param shape: [3 x out_features, in_features] + """ + + q_params: torch.Tensor + + k_params: torch.Tensor + + v_params: torch.Tensor + + def finalize(self): + fused_param = torch.cat([self.q_params, self.k_params, self.v_params], dim=0) + return self.inference_model.transform_qkv_param(fused_param) + + +def megatron_qkv_reshape(param: torch.Tensor, head_size: int, n_heads: int) -> torch.Tensor: + assert param.shape[0] == 3 * n_heads * head_size + + all_heads = torch.chunk(param, chunks=3 * n_heads, dim=0) + q_heads = all_heads[::3] + k_heads = all_heads[1::3] + v_heads = all_heads[2::3] + return torch.cat([q_heads, k_heads, v_heads], dim=0) + + +class MegatronQKVParameter(ParameterBase): + """ + QKV parameter container for Megatron-style QKV projection. Megatron stores the parameter + as [n_heads, 3, head_size, in_features] whereas our inference system is built around + [3, n_heads, head_size, in_features]. This container handles the conversion. + + Note: this container expects the model implementation to implement properties for + `head_size` and `n_heads`. + + src_qkv_w shape: [3 * out_features, in_features] + qkv_w shape: [3 * out_features, in_features] + """ + + params: torch.Tensor + + def finalize(self) -> torch.Tensor: + head_size = self.inference_model.head_size + n_heads = self.inference_model.n_heads + + transposed_param = megatron_qkv_reshape(self.params, head_size, n_heads) + return self.inference_model.transform_qkv_param(transposed_param) + + +def transform_gqa_megatron(src_param: torch.Tensor, head_size: int, n_q_heads: int, n_kv_heads: int) -> torch.Tensor: + assert src_param.shape[0] == (2 * n_kv_heads + n_q_heads) * head_size + + head_ratio = n_q_heads // n_kv_heads + + # Reshape to get the groups as the leading dimension + groups_leading_view = src_param.reshape(n_kv_heads, 2 + head_ratio, head_size, -1) + q_heads = groups_leading_view[:, :head_ratio, :, :].reshape(-1, groups_leading_view.shape[-1]) + k_heads = groups_leading_view[:, head_ratio, :, :].reshape(-1, groups_leading_view.shape[-1]) + v_heads = groups_leading_view[:, head_ratio + 1, :, :].reshape(-1, groups_leading_view.shape[-1]) + # Squeeze will remove extra dimension for bias + return torch.cat([q_heads, k_heads, v_heads], dim=0).squeeze() + + +class GQAMegatronQKVParameter(ParameterBase): + """ + QKV parameter for Megatron-style QKV projection with GQA-style QKV projection. In this + storage format each of the groups is stored consecutively, so there will be multiple q_heads, + then one k head, and one v head. + + Note: this container expects the model implementation to implement properties for + `head_size`, `n_q_heads`, and `n_kv_heads`. + + src_qkv_w shape: [(2 * n_kv_heads + n_q_heads) * head_size, in_features] + qkv_w shape: [(2 * n_kv_heads + n_q_heads) * head_size, in_features] + """ + + params: torch.Tensor + + def finalize(self) -> torch.Tensor: + head_size = self.inference_model.head_size + n_q_heads = self.inference_model.n_heads_q + n_kv_heads = self.inference_model.n_heads_kv + + transposed_param = transform_gqa_megatron(self.params, head_size, n_q_heads, n_kv_heads) + return self.inference_model.transform_qkv_param(transposed_param) diff --git a/deepspeed/inference/v2/model_implementations/common_parameters/unembed_parameters.py b/deepspeed/inference/v2/model_implementations/common_parameters/unembed_parameters.py new file mode 100644 index 000000000000..9f67c0ce3c27 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/common_parameters/unembed_parameters.py @@ -0,0 +1,26 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from ...model_implementations.parameter_base import ParameterBase +""" +Unembedding containers. +""" + + +class UnembedParameter(ParameterBase): + """ + Unembedding parameter. This will likely be mapped to the same original weight in the model as the + embedding, but we have a different preferred sharding approach. + """ + + params: torch.Tensor + """ + Unembedding parameter of shape [vocab_size, model_dim]. + """ + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform_unembed_param(self.params) diff --git a/deepspeed/inference/v2/model_implementations/inference_model_base.py b/deepspeed/inference/v2/model_implementations/inference_model_base.py new file mode 100644 index 000000000000..bc6b28f8ce52 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/inference_model_base.py @@ -0,0 +1,196 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import ABC, abstractmethod +from typing import Iterable, Optional, Tuple, Type + +import torch + +from ..ragged import DSStateManager, RaggedBatchWrapper +from ..ragged.manager_configs import KVCacheConfig +from ..ragged import DSSequenceDescriptor +from ..model_implementations.layer_container_base import LayerContainer +from ..config_v2 import RaggedInferenceEngineConfig +""" +This abstract class defines the interfaces that a model implementation should implement +in order to include anything that may be called by the engine. Most models should be able +to inherit from `DSInferenceTransformerModelBase` to reduce implementation work so it is recommended +to begin there. +""" +""" +Placeholder for typing the model config, which can vary based on model implementation/ +""" +DSModelImplementationConfig = Type['DSModelImplementationConfig'] +""" +Placeholder for typing the distributed comm object. + +TODO(cmikeh2): Replace when we have a more defined API for the inference communication system. +""" +MPType = Type["MPType"] + + +class DSInferenceModelBase(torch.nn.Module, ABC): + """ + Implementation of a model for inference composable with ragged batching. + """ + + _config: DSModelImplementationConfig + """ + Model-specific configuration. No abstraction surrounds this yet. + """ + + _engine_config: RaggedInferenceEngineConfig + """ + Engine configuration. + """ + + _base_mp_group: MPType + """ + Base communication group for Tensor-parallel inference. + """ + + _non_transformer: Optional[LayerContainer] + """ + Abstract container for storing both embedding (pre-transformer) and unembedding (post-transformer) + parameters. This attribute should be None at model instantiation until the Policy sets + the model parameters. These parameters are grouped together since many model implementations + will tie the embedding and unembedding parameters together. + """ + + _transformer: Optional[Iterable[LayerContainer]] + """ + List of abstract containers (1 per layer) for storing transformer (transformer) + parameters. This attribute should be None at model instantiation until the Policy + sets the model parameters. + """ + + state_manager: Optional[DSStateManager] + """ + Since the state manager is lazy initialized, by the engine, it is not guaranteed to be present + until full initialization. + """ + + def __init__(self, config: DSModelImplementationConfig, engine_config: RaggedInferenceEngineConfig, + base_mp_group: MPType) -> None: + """ + Minimal initialization of the model. + + Arguments: + config (DSModelImplementationConfig): Model-specific configuration. No assumptions + should be made about this config that are not closely tied to the specific + model implementation. + engine_config (RaggedInferenceEngineConfig): Engine configuration. + base_mp_group (MPType): Base communication group for Tensor-parallel inference. + """ + super().__init__() + self._config = config + self._engine_config = engine_config + self._base_mp_group = base_mp_group + + # Set to None until the Policy sets the model parameters + self._non_transformer = None + self._transformer = None + + def set_parameters(self, transformer: Iterable[LayerContainer], non_transformer: LayerContainer): + """ + Set the model parameters for the embedding, transformer, and unembedding containers. + """ + self._transformer = transformer + self._non_transformer = non_transformer + + def set_state_manager(self, state_manager: DSStateManager): + """ + Sets the state manager attribute. This is called by the inference engine after + the model is fully initialized. + """ + self.state_manager = state_manager + + @abstractmethod + def get_kv_requirements(self, sequence: DSSequenceDescriptor, max_new_tokens: int, + max_new_blocks: int) -> Tuple[int, int]: + """ + Given a sequence and the number of new tokens in the sequence, determine the + number of new KV blocks needed to support the sequence. This method is + used to help the engine provide schedulability APIs and can be used as a helper + for ``maybe_allocate_kv``. + + Args: + sequence (DSSequenceDescriptor): The sequence for which to allocate KV-storage. + max_new_tokens (int): Maximum number of tokens to hypothetically schedule. + max_new_blocks (int): Maximum number of blocks to hypothetically allocate. + + Returns: + Tuple[int, int]: The tuple of number of tokens scheduled and number + of blocks allocated. In general, only one of these numbers will match the + corresponding input argument, but this is not guaranteed. + """ + raise NotImplementedError() + + @abstractmethod + def maybe_allocate_kv(self, sequence: DSSequenceDescriptor, n_new_tokens: int) -> None: + """ + Given a sequence and the number of new tokens in the sequence, determine + whether or not additional KV-storage is needed and allocate it if so. + + Args: + sequence (DSSequenceDescriptor): The sequence for which to allocate KV-storage. + n_new_tokens (int): The number of new tokens in the sequence. + """ + raise NotImplementedError() + + @abstractmethod + def kv_cache_config(self) -> KVCacheConfig: + """ + Return the KV-cache configuration for this model. + """ + raise NotImplementedError() + + @property + @abstractmethod + def max_sequence_length(self) -> int: + """ + The maximum sequence length supported by the model. + """ + ... + + def maybe_free_kv(self, sequence: DSSequenceDescriptor): + """ + After completing a forward pass, determine whether or not the there are any KV blocks + that maybe freed since they are no longer in use. + + Consider the following example: + + We have a block size of 4 and a local window size of 8. At the beginning of the forward + pass there 10 tokens had been seen and the new forward has a size of 4. This would lend + itself to the following cache structure prior to the forward: + [[0, 1, 2*, 3*] [4*, 5*, 6*, 7*] [8*, 9*, x, x] [x x x x]] + Where x's denote empty cache locations and * denote values that are needed for attention + of the next open slot. After the forward, the cache would look like the following: + [[0, 1, 2, 3] [4, 5, 6*, 7*] [8*, 9*, 10*, 11*] [12* 13* x x]] + In this case, the first block is no longer needed since it is not needed for any future + local attention windows. This function would be responsible for freeing that block. + + Default behavior assumes no local patterns that require freeing and in general should + be sufficient. + """ + pass + + @abstractmethod + def prepare_batch(self, wrapped_batch: RaggedBatchWrapper) -> None: + """ + This will be called before each forward with the intent of building forward-specific metadata + about a batch. The intent here is to build data structures like attention atoms without necessarily + needing to implement graphable kernels to do so. + + Abstract so as to force model implementations to opt out of doing anything here explicitly. + """ + raise NotImplementedError() + + def forward(wrapped_batch: RaggedBatchWrapper) -> torch.Tensor: + """ + Complete a forward pass of the model. This interface should be graphable, so it + should not rely on the ability to use python control flow. + """ + raise NotImplementedError() diff --git a/deepspeed/inference/v2/model_implementations/inference_policy_base.py b/deepspeed/inference/v2/model_implementations/inference_policy_base.py new file mode 100644 index 000000000000..f87c4f0c7cfc --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/inference_policy_base.py @@ -0,0 +1,155 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import ABC, abstractmethod +from typing import Any, Iterable, List, Union + +from ..config_v2 import RaggedInferenceEngineConfig +from ..checkpoint import CheckpointEngineBase +from ..logging import inference_logger +from .layer_container_base import LayerContainer +from .inference_model_base import DSInferenceModelBase + + +class ContainerMap: + + def __init__(self) -> None: + self._prefix_map = {} + self._transformer_params = None + self._non_transformer_params = None + + @property + def transformer_params(self) -> Iterable[LayerContainer]: + return self._transformer_params + + @property + def non_transformer_params(self) -> LayerContainer: + return self._non_transformer_params + + def set_transformer_params(self, prefixes: Union[str, Iterable[str]], containers: List[LayerContainer]) -> None: + if not isinstance(containers, list): + raise ValueError( + f"The transformer containers should be a list, of one container per layer, but got {type(containers)} instead." + ) + + self._transformer_prefixes = prefixes if isinstance(prefixes, list) else [prefixes] + self._transformer_params = containers + + def set_non_transformer_params(self, container: LayerContainer) -> None: + self._non_transformer_params = container + + def set_unmapped_params(self, prefixes: Union[str, Iterable[str]]) -> None: + self._unmapped_prefixes = prefixes + + def map_param(self, name, parameter) -> None: + for unmapped_prefix in self._unmapped_prefixes: + if name.startswith(unmapped_prefix): + inference_logger().debug(f"Ignoring: {name} for {unmapped_prefix}") + return + + for transformer_prefix in self._transformer_prefixes: + if name.startswith(transformer_prefix): + popped_name = name[len(transformer_prefix) + 1:] + layer_idx = popped_name.split(".")[0] + assert layer_idx.isdigit( + ), f"expected name to start w. list index but got {layer_idx} instead, name={name}" + layer_idx = int(layer_idx) + inference_logger().debug( + f"Setting: {'.'.join(popped_name.split('.')[1:])} for layer-idx={layer_idx} to {parameter.shape}") + self._transformer_params[layer_idx].set_dependency(".".join(popped_name.split(".")[1:]), parameter) + return + + try: + inference_logger().debug(f"Setting: {name} to {parameter.shape}") + self._non_transformer_params.set_dependency(name, parameter) + except ValueError: + # Catch the ValueError here from the non_transformer_params because we are knowingly + # calling it with something that may not match. This should allow us to raise a slightly more + # informative error message. + raise ValueError(f"Cannot find container for {name}, please double check the Containers/ContainerMap") + + def validate(self) -> None: + if not self._non_transformer_params.is_initialized: + raise RuntimeError("Non-transformer parameters not fully initialized after checkpoint load.") + + for layer_idx, container in enumerate(self._transformer_params): + if not container.is_initialized: + raise RuntimeError( + f"Transformer container at index {layer_idx} not fully initialized after checkpoint load.") + + +class InferenceV2Policy(ABC): + """ + The InferenceV2Policy is the base class for all inference policies. An inference policy + is responsible for instantiating the inference model and mapping the parameters from the + checkpoint engine to the model itself. + """ + + def __init__(self, checkpoint_engine: CheckpointEngineBase, model_config: Any) -> None: + self._checkpoint_engine = checkpoint_engine + self._model_config = model_config + + def build_model(self, engine_config: RaggedInferenceEngineConfig, mp_group: Any) -> DSInferenceModelBase: + """ + Completely instantiate the inference model. This will both create the ops needed to run the + model, as well as load the model parameters via the checkpoint engine. For more context + on each of these components please see ``instantiate_model`` and ``populate_model_parameters``. + + Arguments: + engine_config: The config that has been used to instantiate the engine. This is used + to communicate to the model implementation the limits on batches (sequences/tokens) + and bound the size of intermediate buffers. + mp_group: Object to enable communication between tensor parallel ranks. + + Returns: + DSInferenceModelBase: An implementation of the inference model abstraction that will be + run by the engine. + """ + self.model = self.instantiate_model(engine_config, mp_group) + self.populate_model_parameters() + return self.model + + @abstractmethod + def instantiate_model(self, engine_config: RaggedInferenceEngineConfig) -> DSInferenceModelBase: + """ + Instantiate the inference model. Depending on the engine/model config, this could be where + different model implementations could be selected. + + Arguments: + engine_config: The config that has been used to instantiate the engine. This is used + to communicate to the model implementation the limits on batches (sequences/tokens) + and bound the size of intermediate buffers. + + Returns: + DSInferenceModelBase: An implementation of the inference model abstraction that will be + run by the engine. + """ + ... + + @abstractmethod + def build_container_map(self) -> ContainerMap: + """ + Build a dictionary representing the structure of the string prefixes leading + to the parameters to be mapped to the container. + + Returns: + ContainerMap: An instantiated mapping describing how checkpoint prefixes map + to ``LayerContainer`` instances. + """ + raise NotImplementedError() + + def populate_model_parameters(self) -> None: + """ + This model will iterate over the parameters (as provided by the checkpoint engine) and + use the container map built by ``build_container_map`` to populate the model + """ + + container_map = self.build_container_map() + for name, parameter in self._checkpoint_engine.parameters(): + container_map.map_param(name, parameter) + container_map.validate() + + self.model.set_parameters(transformer=container_map.transformer_params, + non_transformer=container_map.non_transformer_params) diff --git a/deepspeed/inference/v2/model_implementations/inference_transformer_base.py b/deepspeed/inference/v2/model_implementations/inference_transformer_base.py new file mode 100644 index 000000000000..ce3a486373bc --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/inference_transformer_base.py @@ -0,0 +1,616 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import abstractmethod +from typing import Optional + +import torch + +import deepspeed.comm as dist +from deepspeed.accelerator import get_accelerator +from ..allocator import on_device +from ..config_v2 import RaggedInferenceEngineConfig +from ..inference_utils import ActivationType, ceil_div, is_gated +from ..model_implementations import * +from ..model_implementations.sharding import * +from ..modules.configs import ( + DSEmbeddingsConfig, + DSLinearConfig, + DSMoEConfig, + DSNormConfig, + DSSelfAttentionConfig, + DSUnembedConfig, + NormTypeEnum, + PositionalEmbeddingType, +) +from ..modules import heuristics +from ..ragged import ( + DSSequenceDescriptor, + KVCacheConfig, + RaggedBatchWrapper, +) +from .inference_model_base import ( + DSInferenceModelBase, + DSModelImplementationConfig, + MPType, +) + +try: + from functools import cached_property +except ImportError: + + def cached_property(func): + return property(func) + + +class DSTransformerModelBase(DSInferenceModelBase): + """ + Dimensioning properties + """ + + @property + @abstractmethod + def num_layers(self) -> int: + """ + Number of the layers in the model + """ + ... + + @property + @abstractmethod + def model_dim(self) -> int: + """ + Size of embedding projection and residuals. + """ + ... + + @property + @abstractmethod + def vocab_size(self) -> int: + """ + Size of the vocabulary (including padding). + """ + ... + + @property + @abstractmethod + def head_size(self) -> int: + """ + Size of each attention head. + """ + ... + + @property + @abstractmethod + def n_heads(self) -> int: + """ + The number of query heads on the model. This should not take into account + any dimension reductions from model sharding. + """ + ... + + @property + def n_heads_q(self) -> int: + """ + Alias to n_heads. + """ + return self.n_heads + + @property + def n_heads_kv(self) -> int: + """ + The number of key and value heads on the model. For GQA or MQA, overload this attribute. + Otherwise it adopts MHA formulations and uses n_heads. This should not take into account + any dimension reductions from model sharding. + """ + return self.n_heads + + @property + @abstractmethod + def intermediate_dim(self) -> int: + """ + The size of the (unsharded) intermediate projection dim. For a gated activation function + this is the size of the input to the second MLP layer. This should not take into account + any dimension reductions from model sharding. + """ + ... + + @property + @abstractmethod + def positional_embedding_type(self) -> PositionalEmbeddingType: + """ + The type of positional embedding used by the model. + """ + ... + + """ + Architectural properties + """ + + @property + @abstractmethod + def activation_dtype(self) -> torch.dtype: + """ + The activation dtype of the model. + """ + ... + + @property + @abstractmethod + def mlp_activation_fn(self) -> ActivationType: + """ + The activation function used in the MLP. + """ + ... + + @property + @abstractmethod + def norm_type(self) -> NormTypeEnum: + """ + The type of normalization used in the model. + """ + ... + + """ + Derived helpers + """ + + @cached_property + def tp_rank(self) -> int: + """ + The rank of the current process. + + # TODO(cmikeh2): Kind of a hack right now, but this is too verbose to use at + the frequency we need. + """ + return dist.get_rank(group=self._base_mp_group) + + @cached_property + def tp_size(self) -> int: + """ + The total number of processes. + + # TODO(cmikeh2): Kind of a hack right now, but this is too verbose to use at + the frequency we need. + """ + return dist.get_world_size(group=self._base_mp_group) + + @cached_property + def n_heads_q_local(self) -> int: + """ + Number of local heads post sharding. + """ + return get_local_heads(self.tp_rank, self.tp_size, self.n_heads_q, self.n_heads_kv)[0] + + @cached_property + def n_heads_kv_local(self) -> int: + """ + Number of local heads post sharding. + """ + return get_local_heads(self.tp_rank, self.tp_size, self.n_heads_q, self.n_heads_kv)[1] + + @property + def gated_mlp(self) -> bool: + """ + Return a boolean to determine whether the model uses a gated activation function. + """ + return is_gated(self.mlp_activation_fn) + + """ + Method implementations + """ + + def __init__(self, config: DSModelImplementationConfig, engine_config: RaggedInferenceEngineConfig, + base_mp_group: MPType) -> None: + """ + Base implementation for initialization. By default, this will initialize + the traditional components of a transformer model: + - Embedding + - QKV projection + - Self attention + - Attention output projection + - Feed forward network + - Normalization + - Unembedding + + Arguments: + config (DSModelImplementationConfig): Model-specific configuration. No assumptions + should be made about this config that are not closely tied to the specific + model implementation. + engine_config (RaggedInferenceEngineConfig): Engine configuration. + base_mp_group (MPType): Base communication group for Tensor-parallel inference. + """ + super().__init__(config, engine_config, base_mp_group) + + self.make_norm_layer() + self.make_qkv_layer() + self.make_attn_layer() + self.make_attn_out_layer() + self.make_mlp_1_layer() + self.make_mlp_2_layer() + self.make_embedding_layer() + self.make_unembedding_layer() + self._kv_cache_config = None + + ######### Embedding ######### + def make_embedding_layer(self) -> None: + """ + Performs setup and creates embedding DSModule. This will set the `self.embed` attribute. + """ + + embed_config = DSEmbeddingsConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + residual_dtype=self.activation_dtype, + embedding_dim=self.model_dim, + ) + + self.embed = heuristics.instantiate_embed(embed_config, self._engine_config) + + @on_device + def transform_embedding_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Performs embedding sharding along the channels dimension. + """ + # Until we can do non-contiguous all-gather, we won't shard the embedding parameters. + return param.to(self.activation_dtype.value) + + ######### Unembedding ######### + def make_unembedding_layer(self) -> None: + """ + Performs setup and creates an unembedding layer. This implementation assumes + normalization prior to the LM head projection. If this does not match the model's + implementation, override this method. This will set the ``self.unembed`` attribute. + """ + unembed_dim = sharded_unembed_dim(self.vocab_size, self.tp_rank, self.tp_size) + + unembed_config = DSUnembedConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + max_sequences=self._engine_config.state_manager.max_ragged_sequence_count, + dtype=self.activation_dtype, + model_dim=self.model_dim, + vocab_size=unembed_dim, + norm_type=self.norm_type, + ) + + self.unembed = heuristics.instantiate_unembed(unembed_config, self._engine_config) + + if self.tp_size > 1: + self._comm_logits = torch.empty(self.tp_size, + self._engine_config.state_manager.max_ragged_sequence_count, + unembed_dim, + device=get_accelerator().current_device(), + dtype=self.activation_dtype.value) + self._return_logits = torch.empty(self._engine_config.state_manager.max_ragged_sequence_count, + self.vocab_size, + device=get_accelerator().current_device(), + dtype=self.activation_dtype.value) + + @on_device + def transform_unembed_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Performs sharding along the vocab dimension. + """ + return shard_unembed_param(param, self.tp_rank, self.tp_size).to(self.activation_dtype.value) + + ######### QKV ######### + def make_qkv_layer(self) -> None: + """ + Instantiates the linear projection layer for the QKV linear layer. This sets the + `self.qkv` attribute. + """ + out_features = qkv_out_features(self.model_dim, self.tp_rank, self.tp_size, self.head_size, self.n_heads_q, + self.n_heads_kv) + + linear_config = DSLinearConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + in_channels=self.model_dim, + out_channels=out_features, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + ) + + self.qkv = heuristics.instantiate_linear(linear_config, self._engine_config) + + @on_device + def transform_qkv_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Passes a QKV parameter to the underlying implementation for any necessary + transformations. + + Args: + param (torch.Tensor): The parameter to transform. This may be either a bias or weight and should have + the shape (out_neurons, in_neurons) + """ + param = shard_qkv_param(param, self.tp_rank, self.tp_size, self.head_size, self.n_heads_q, self.n_heads_kv) + return self.qkv.transform_param(param) + + ######### Attention ######### + def make_attn_layer(self) -> None: + """ + Builds the attention layer for the model. This sets the `self.attn` attribute. + """ + softmax_scale = 1.0 / (self.head_size**0.5) + + attn_config = DSSelfAttentionConfig(max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + n_heads_q=self.n_heads_q_local, + n_heads_kv=self.n_heads_kv_local, + head_size=self.head_size, + max_sequences=self._engine_config.state_manager.max_ragged_sequence_count, + scale_factor=softmax_scale, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + positional_embedding_type=self.positional_embedding_type) + + self.attn = heuristics.instantiate_attention(attn_config, self._engine_config) + + def get_kv_requirements(self, sequence: DSSequenceDescriptor, max_new_tokens: int, + max_new_blocks: int) -> Tuple[int, int]: + """ + See ``DSInferenceModelBase.get_kv_requirements`` for documentation. + + This method assumes an autoregressive dense attention pattern. Override this method + if this does not match the model's attention pattern. + """ + total_tokens = sequence.seen_tokens + max_new_tokens + req_blocks = ceil_div(total_tokens, self.attn.kv_block_size) + block_lim = req_blocks - sequence.cur_allocated_blocks + + if block_lim <= max_new_blocks: + return max_new_tokens, block_lim + + token_capacity = (max_new_blocks + + sequence.cur_allocated_blocks) * self.attn.kv_block_size - sequence.seen_tokens + + return token_capacity, max_new_blocks + + def maybe_allocate_kv(self, sequence: DSSequenceDescriptor, n_new_tokens: int) -> None: + """ + See ``DSInferenceModelBase.maybe_allocate_kv`` for documentation. + + This method assumes an autoregressive dense attention pattern. Override this method + if this does not match the model's attention pattern. + """ + _, n_needed_blocks = self.get_kv_requirements(sequence, n_new_tokens, self.state_manager.free_blocks) + + if n_needed_blocks > 0: + new_blocks = self.state_manager.allocate_blocks(n_needed_blocks) + sequence.extend_kv_cache(new_blocks) + + def kv_cache_config(self) -> KVCacheConfig: + """ + See ``DSInferenceModelBase.kv_cache_config`` for documentation. + + This method assumes an autoregressive dense attention pattern. Override this method + if this does not match the model's attention pattern. + """ + if self._kv_cache_config is None: + cache_shape = (self.num_layers, self.n_heads_kv_local, self.head_size) + max_blocks = ceil_div(self.max_sequence_length, self.attn.kv_block_size) + self._kv_cache_config = KVCacheConfig(block_size=self.attn.kv_block_size, + cache_shape=cache_shape, + cache_dtype=self.activation_dtype, + max_blocks_per_allocation_group=max_blocks) + return self._kv_cache_config + + def prepare_batch(self, wrapped_batch: RaggedBatchWrapper) -> None: + """ + See ``DSInferenceModelBase.prepare_batch`` for documentation. + + This method assumes an autoregressive dense attention pattern. Override this method + if this does not match the model's attention pattern. + """ + self.attn.build_atoms(wrapped_batch) + + ######### Attention output ######### + def make_attn_out_layer(self) -> None: + """ + Instantiates the linear projection layer for the attention output linear layer. This sets the + `self.attn_out` attribute. + """ + in_features = attn_out_in_features(self.model_dim, self.tp_rank, self.tp_size, self.head_size, self.n_heads_q, + self.n_heads_kv) + + linear_config = DSLinearConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + in_channels=in_features, + out_channels=self.model_dim, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + ) + + self.attn_out = heuristics.instantiate_linear(linear_config, self._engine_config) + + @on_device + def transform_attn_out_param(self, param: torch.Tensor) -> Optional[torch.Tensor]: + """ + Shards an attention output projection parameter and passes it to the underlying + implementation for any necessary transformations. This will return `None` for bias parameters + if they are not on TP rank 0. + + Args: + param (torch.Tensor): The parameter to transform. This may be either a bias or weight and should have + the shape (out_neurons, in_neurons). + """ + param = shard_attn_out_param(param, self.tp_rank, self.tp_size, self.head_size, self.n_heads_q, + self.n_heads_kv) + + if param is not None: + param = self.attn_out.transform_param(param) + + return param + + ######### MLP ######### + def make_mlp_1_layer(self) -> None: + """ + Instantiates the linear projection layer for the first MLP in the feedforward network. + This sets the `self.mlp_1` attribute. + """ + shard_size = sharded_intermediate_dim(self.intermediate_dim, self.tp_size, self.tp_rank) + + linear_config = DSLinearConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + in_channels=self.model_dim, + out_channels=shard_size, + activation=self.mlp_activation_fn, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + ) + + self.mlp_1 = heuristics.instantiate_linear(linear_config, self._engine_config) + + @on_device + def transform_mlp_1_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Shards the first MLP parameter and passes it to the underlying implementation + for any necessary transformations. + + Args: + param (torch.Tensor): The parameter to transform. This may be either a bias or weight and should have + the shape (out_neurons, in_neurons). + """ + param = shard_mlp_1_param(param, self.tp_rank, self.tp_size, gated=self.gated_mlp) + + return self.mlp_1.transform_param(param) + + def make_mlp_2_layer(self) -> None: + """ + Instantiates the linear projection layer for the second MLP in the feedforward network. + This sets the `self.mlp_2` attribute. + """ + shard_size = sharded_intermediate_dim(self.intermediate_dim, self.tp_size, self.tp_rank) + + linear_config = DSLinearConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + in_channels=shard_size, + out_channels=self.model_dim, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + ) + + self.mlp_2 = heuristics.instantiate_linear(linear_config, self._engine_config) + + @on_device + def transform_mlp_2_param(self, param: torch.Tensor) -> Optional[torch.Tensor]: + """ + Shards the second MLP parameter and passes it to the underlying implementation + for any necessary transformations. This will return `None` for bias parameters + if they are not on TP rank 0. + + Args: + param (torch.Tensor): The parameter to transform. This may be either a bias or weight and should have + the shape (out_neurons, in_neurons). + """ + param = shard_mlp_2_param(param, self.tp_rank, self.tp_size) + + if param is not None: + param = self.mlp_2.transform_param(param) + + return param + + ######### Norm ######### + def make_norm_layer(self) -> None: + """ + Instantiates the normalization layer for the model. This sets the `self.norm` attribute. + + TODO(cmikeh2): In the future we'll distinguish between the different norm objects, + but for now we'll just use the same one for all of them. + """ + norm_config = DSNormConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + type=self.norm_type, + channels=self.model_dim, + residual_dtype=self.activation_dtype, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + ) + + self.norm = heuristics.instantiate_pre_norm(norm_config, self._engine_config) + + @on_device + def transform_norm_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Passes a normalization parameter to the underlying implementation for any + necessary transformations. + + TODO(cmikeh2): In the future we'll distinguish between the different norm objects, + but for now we'll just use the same one for all of them. + + Args: + param (torch.Tensor): The parameter to transform. This may be either a bias or weight and should have + shape (model_dim,) + """ + return self.norm.transform_param(param) + + +class DSMoETransformerModelBase(DSTransformerModelBase): + + @property + def num_experts(self) -> int: + """ + Return the number of experts in the model. + """ + raise NotImplementedError("Attempted to access an unimplemented number of experts") + + def make_moe_layer(self) -> None: + """ + Instantiates the MoE layer for the model. This sets the `self.moe` attribute. + """ + sharded_dim = sharded_intermediate_dim(self.intermediate_dim, self.tp_size, self.tp_rank) + + moe_config = DSMoEConfig( + max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + model_dim=self.model_dim, + intermediate_features=sharded_dim, + activation=self.mlp_activation_fn, + n_experts=self.num_experts, + input_dtype=self.activation_dtype, + output_dtype=self.activation_dtype, + ) + + self.moe = heuristics.instantiate_moe(moe_config, self._engine_config) + + @on_device + def transform_moe_gate_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Passes a MoE gate parameter to the underlying implementation for any necessary transformations. + + TODO(cmikeh2): This will need to be updated/overridden for expert parallelism. + """ + return self.moe.transform_gate_param(param) + + @on_device + def transform_moe_mlp_1_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Shards the first MoE param and passes it to the underlying implementation. Since it's possible for an architecture + to have both MoE and non-MoE layers, this can't be overloaded on the MLP1 transform. Furthermore, since both + the MoE DSModule owns both MLP1 and MLP2, under certain sharding conditions it's not possible for the model implementation + to infer from the shape whether to perform a different transformation based on MLP1 or MLP2. This (and the below) + separations are intended to solve both these issues. + + Args: + param (torch.Tensor): The parameter to transform. This should have shape (n_experts, out_neurons, in_neurons). + """ + param = shard_mlp_1_param(param, self.tp_rank, self.tp_size, gated=self.gated_mlp, is_moe=True) + + return self.moe.transform_moe_mlp_1_param(param) + + @on_device + def transform_moe_mlp_2_param(self, param: torch.Tensor) -> Optional[torch.Tensor]: + """ + Shards the second MoE param and passes it to the underlying implementation. See the above for context on why this API + exists. + + This will return `None` for expert bias params not on TP rank 0. NOTE(cmikeh2): Does it make sense to round-robin assign? + My intuition is that this will make debugging much more difficult for minimal memory reduction. + + Args: + param (torch.Tensor): The parameter to transform. This should have shape (n_experts, out_neurons, in_neurons). + """ + param = shard_mlp_2_param(param, self.tp_rank, self.tp_size, is_moe=True) + + if param is not None: + param = self.moe.transform_moe_mlp_2_param(param) + + return param diff --git a/deepspeed/inference/v2/model_implementations/layer_container_base.py b/deepspeed/inference/v2/model_implementations/layer_container_base.py new file mode 100644 index 000000000000..e0ec19372569 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/layer_container_base.py @@ -0,0 +1,289 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Type + +import torch + +from deepspeed.accelerator import get_accelerator +from .parameter_base import ParameterBase, ParametrizedList + +# Currently have dependency loops for the type hints. +InferenceModel = Type["InferenceModel"] +LayerContainer = Type["LayerContainer"] + +MAPPING_KEY = "PARAM_MAPPING" +PLIST_HELPERS = "_ds_plist_strip_vals" + + +def make_finalization_callback(all_names: str): + """ + Helper method for building the finalization callback for a LayerContainer. This + is not client code and should not be used or called directly. + """ + + def finalization_callback(self, param: ParameterBase, finalized_param: torch.Tensor) -> None: + """ + Callback for when a parameter is finalized. + """ + self._finalized_params += 1 + + for name in all_names: + if getattr(self, name) is param: + setattr(self, name, finalized_param) + + return finalization_callback + + +class LayerMetaclass(type): + """ + MetaClass for the LayerContainer base class. This class will parse the annotations + of the class that correspond to `ParameterBase` and create None initializers for each + as well as a finalization callback that for when each `ParameterBase` is finalized + and should be replaced with a Tensor. + """ + + def __new__(cls, clsname, bases, attrs): + + annotations = attrs.get("__annotations__", {}) + + for base in bases: + # We'll pick up all annotations on any base classes. This will allow us to + # to use inheritance to share common parameter groups in base classes. + if hasattr(base, "__annotations__"): + annotations.update(base.__annotations__) + + if hasattr(base, MAPPING_KEY): + if MAPPING_KEY not in attrs: + # This is likely a fail state. If a parent has MAPPING KEY but the child does + # not, then we're guaranteed only a subset of the parameters will be mapped. + attrs[MAPPING_KEY] = {} + attrs[MAPPING_KEY].update(getattr(base, MAPPING_KEY)) + + all_names = [name for name, annotation in annotations.items() if issubclass(annotation, ParameterBase)] + + if MAPPING_KEY in attrs: + # If we have a mapping key at all, then we will enter the validation mode for building + # helpers for mapping and ensuring we have complete mapping. + + # First we'll build a flat list of every dependency for this layer. + all_deps = set() + for name in all_names: + parameter_deps = [ + name for name, annotation in annotations[name].__annotations__.items() + if issubclass(annotation, (torch.Tensor, ParametrizedList)) + ] + + all_deps.update([f"{name}.{dep}" for dep in parameter_deps]) + + # Create static helper for doing the string processing only once. + attrs[PLIST_HELPERS] = [] + + # Iterate over all the mappings + for src_name, target_or_targets in attrs[MAPPING_KEY].items(): + if isinstance(target_or_targets, str): + target_or_targets = [target_or_targets] + + actual_targets = [] + for target_name in target_or_targets: + base_dependency, dependency_attr = target_name.split(".") + + # Check for invalid mappings + if base_dependency not in all_names: + raise ValueError( + "Target parameter \"{}\" not found in this layer. Valid targets are {}".format( + base_dependency, all_names)) + if dependency_attr not in annotations[base_dependency].__annotations__: + # This check is not universal (see below) if a single dependency is being + # mapped to by a single row. + raise ValueError( + "Target dependency \"{}\" not found on parameter \"{}\". Valid targets are {}".format( + dependency_attr, base_dependency, annotations[base_dependency].__annotations__.keys())) + if target_name not in all_deps: + raise ValueError( + "Target dependency \"{}\" was targeted with multiple mapping rules.".format(target_name)) + + # If we've made it this far, the dependency definitely exists. + actual_targets.append(annotations[base_dependency].__annotations__[dependency_attr]) + + all_deps.remove(target_name) + + are_plists = [issubclass(target, ParametrizedList) for target in actual_targets] + if all(are_plists): + # We can do direct sets on everything but ParametrizedLists, so we'll only explicitly + # handle these here. + # TODO(cmikeh2): SPLIT, error if more than 1 + glob_count = src_name.count("*") + if glob_count > 1: + raise ValueError( + "ParametrizedList index inference can only work with a single glob: {}".format(src_name)) + elif glob_count == 0: + raise ValueError( + "Must have wildcard (*) in source name for ParametrizedList mapping: {}".format(src_name)) + + wildcard_idx = src_name.find("*") + prefix = src_name[:wildcard_idx] + suffix = src_name[wildcard_idx + 1:] + attrs[PLIST_HELPERS].append((prefix, suffix, target_or_targets)) + elif any(are_plists): + raise ValueError("Cannot mix ParametrizedLists and Tensors in a single mapping rule.") + + if len(all_deps) > 0: + raise ValueError( + "A parameter mapping was provided for {}, but the following dependencies were not mapped: {}". + format(clsname, all_deps)) + + attrs["finalization_callback"] = make_finalization_callback(all_names) + + new_obj = super().__new__(cls, clsname, bases, attrs) + + setattr(new_obj, "_n_params", len(all_names)) + setattr(new_obj, "_annotation_attrs", all_names) + + return new_obj + + def __call__(cls, *args, **kwargs): + instance = cls.__new__(cls, *args, **kwargs) + instance.__init__(*args, **kwargs) + + for name, annotation in instance.__annotations__.items(): + if issubclass(annotation, ParameterBase): + # TODO(cmikeh2): Do we want to make this a property + # It might also make sense to do this in the base class __init__ + # but since it is tied with the changes made in __new__ it feels + # to me like it should be here. + setattr(instance, name, annotation(instance.inference_model, instance)) + + return instance + + +class LayerContainer(metaclass=LayerMetaclass): + """ + Abstract base class for containing model parameters. + + This is primarily a guidance abstraction since we do not put any restrictions + on how the parameters are stored. + + To use this class, annotate the class with `ParameterBase` subclasses and give them + names. As a checkpoint is loaded into this container, the `ParameterBase` instances + will be replaced with realized Tensors as soon as each of their dependencies are met. + + To enable automatic mapping, add a static attribute `PARAM_MAPPING` to the class + definition. This should be a dictionary mapping from a source string to one or + more dependencies. + + ```python + class MyLayer(LayerContainer): + PARAM_MAPPING = { + "path.to.param.dependency", "container_param_1.dependency", + "path.to.param2.dependency", "container_param_2.dependency", + "path.to.param3.*.dependency", "container_param_3.list_dependency" + } + + ... + ``` + """ + + def __init__(self, model: InferenceModel) -> None: + """ + Initialization of the LayerContainer. This method does not need to be overridden + for any children classes. + + Args: + model (InferenceModel): Inference model that will be used to shard and transform + parameters correctly, as well as provide specific information about the model + for `ParameterizedList`s that may be part of one of the member `ParameterBase`s. + """ + self.inference_model = model + self._finalized_params = 0 + + @property + def is_initialized(self) -> bool: + """ + Returns whether or not all parameters have been initialized and transformed by + the model. Once this returns True, all the `ParameterBase` instances will be + torch.Tensors. + """ + if self._finalized_params != self.n_params: + return False + + for name in self._annotation_attrs: + tensor = getattr(self, name) + if tensor is None: + continue + elif not isinstance(tensor, torch.Tensor): + raise ValueError("Layer should be finalized, but {} is neither Tensor or None".format(name)) + elif tensor.device != torch.device(get_accelerator().current_device()): + raise RuntimeError("Layer should be finalized, but {} is not on device {}".format( + name, + get_accelerator().current_device())) + return True + + @property + def n_params(self) -> int: + """ + The number of parameters this container holds. This is a read-only value + that is set by the metaclass. + """ + return self._n_params + + @property + def mapping_params(self) -> dict: + return getattr(self.__class__, MAPPING_KEY, {}) + + @property + def plist_helpers(self) -> list: + return getattr(self.__class__, PLIST_HELPERS, []) + + def set_dependency(self, dep_name: str, dep_value: torch.Tensor) -> None: + """ + Set dependency can be used for managing dependencies when a mapping is provided + in the class definition for the layer. The dep_name here should have any prefix + for transformer layers removed (such as model.layers.*.attn.qkv.weight -> attn.qkv.weight). + + Args: + dep_name (str): The name of the dependency to set. + dep_value (torch.Tensor): The value to set the dependency to. + """ + if dep_name in self.mapping_params: + # If we have an exact match, it's a direct mapping and we can immediately set + # the value. + target = self.mapping_params[dep_name] + + # Convert single targets to a list for consistency + if isinstance(target, str): + target = [target] + + for target_name in target: + # Double setting doesn't set the attribute correctly, so we do a getattr then setattr + target_param_name, target_dependency_name = target_name.split(".") + target_param = getattr(self, target_param_name) + setattr(target_param, target_dependency_name, dep_value) + return + + # Otherwise we need to map to one of the parameter lists. + for prefix, suffix, dests in self.plist_helpers: + if dep_name.startswith(prefix) and dep_name.endswith(suffix): + # We have a match, so we can set the value. + target_idx = int(dep_name[len(prefix):-len(suffix)]) + + # Convert single targets to a list for consistency + if isinstance(dests, str): + dests = [dests] + + for dest in dests: + target_param_name, target_dependency_name = dest.split(".") + target_param = getattr(self, target_param_name) + target_dependency = getattr(target_param, target_dependency_name) + target_dependency[target_idx] = dep_value + return + + raise ValueError( + "Could not find a mapping for dependency \"{}\". Check that it is included in the ``MAPPING_PARAMS``. See docstring for more on ``MAPPING_PARAMS``" + .format(dep_name)) + + +class ContainerMap: + pass diff --git a/deepspeed/inference/v2/model_implementations/llama_v2/__init__.py b/deepspeed/inference/v2/model_implementations/llama_v2/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/llama_v2/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_containers.py b/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_containers.py new file mode 100644 index 000000000000..ec39866d0d8d --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_containers.py @@ -0,0 +1,80 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# Create a container object to save model-specific tensors using the policy file above. + +from ...model_implementations.common_parameters import * +from ...model_implementations.layer_container_base import LayerContainer +''' + # HF Llama model looks like this: + +LlamaForCausalLM( + (model): LlamaModel( + (embed_tokens): Embedding(32000, 4096, padding_idx=0) + (layers): ModuleList( + (0-31): 32 x LlamaDecoderLayer( + (self_attn): LlamaAttention( + (q_proj): Linear(in_features=4096, out_features=4096, bias=False) + (k_proj): Linear(in_features=4096, out_features=4096, bias=False) + (v_proj): Linear(in_features=4096, out_features=4096, bias=False) + (o_proj): Linear(in_features=4096, out_features=4096, bias=False) + (rotary_emb): LlamaRotaryEmbedding() + ) + (mlp): LlamaMLP( + (gate_proj): Linear(in_features=4096, out_features=11008, bias=False) + (up_proj): Linear(in_features=4096, out_features=11008, bias=False) + (down_proj): Linear(in_features=11008, out_features=4096, bias=False) + (act_fn): SiLUActivation() + ) + (input_layernorm): LlamaRMSNorm() + (post_attention_layernorm): LlamaRMSNorm() + ) + ) + (norm): LlamaRMSNorm() + ) + (lm_head): Linear(in_features=4096, out_features=32000, bias=False) +) +''' + + +class Llama2TransformerContainer(LayerContainer): + """ + Transformer layer container for the Llama-2 model. + """ + qkv_w: UnfusedQKVParameter + attn_out_w: AttentionOutputParameter + mlp_1_w: GatedMLPParameter + mlp_2_w: MLP2Parameter + attn_norm_gamma: NormParameter + mlp_norm_gamma: NormParameter + #rotary_emb: InvFreqParameter + + PARAM_MAPPING = { + "self_attn.q_proj.weight": "qkv_w.q_params", + "self_attn.k_proj.weight": "qkv_w.k_params", + "self_attn.v_proj.weight": "qkv_w.v_params", + "self_attn.o_proj.weight": "attn_out_w.params", + "mlp.gate_proj.weight": "mlp_1_w.gate_params", + "mlp.up_proj.weight": "mlp_1_w.up_params", + "mlp.down_proj.weight": "mlp_2_w.params", + "input_layernorm.weight": "attn_norm_gamma.params", + "post_attention_layernorm.weight": "mlp_norm_gamma.params", + #"self_attn.rotary_emb.inv_freq": "rotary_emb.params", + } + + +class Llama2NonTransformerContainer(LayerContainer): + """ + Non-Transformer layer container for the Llama-2 model. + """ + word_emb: EmbeddingParameter + word_unembed: UnembedParameter + final_norm: NormParameter + + PARAM_MAPPING = { + "model.embed_tokens.weight": "word_emb.params", + "model.norm.weight": "final_norm.params", + "lm_head.weight": "word_unembed.params", + } diff --git a/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_model.py b/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_model.py new file mode 100644 index 000000000000..9b628f77de01 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_model.py @@ -0,0 +1,204 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Iterable, Optional, Tuple + +import torch + +import deepspeed.comm as dist + +from ...allocator import empty_from +from ...inference_utils import ActivationType, DtypeEnum +from ...model_implementations import * +from ...modules.configs import * +from ...modules.interfaces import * +from ...ragged import RaggedBatchWrapper + +from .llama_v2_containers import Llama2NonTransformerContainer, Llama2TransformerContainer + + +class Llama2InferenceModel(DSTransformerModelBase): + """ + Inference model implementation for ragged batching for Llama-2 models. + """ + + _non_transformer: Optional[Llama2NonTransformerContainer] + """ + Embed + unembed container. Specializing the type annotation. + """ + + _transformer: Optional[Iterable[Llama2TransformerContainer]] + """ + Per-layer transformer container. Specializing the type annotation. + """ + """ + Properties ineherited from `DSInferenceModelBase` + """ + + @property + def max_sequence_length(self) -> int: + return self._config.max_seq_length + + """ + Properties ineherited from `DSTransformerModelBase` + """ + + @property + def num_layers(self) -> int: + return self._config.num_hidden_layers + + @property + def model_dim(self) -> int: + return self._config.hidden_size + + @property + def vocab_size(self) -> int: + return self._config.vocab_size + + @property + def head_size(self) -> int: + return self.model_dim // self.n_heads + + @property + def n_heads(self) -> int: + return self._config.num_attention_heads + + @property + def intermediate_dim(self) -> int: + return self._config.intermediate_size + + @property + def n_heads_kv(self) -> int: + return self._config.num_key_value_heads + + @property + def activation_dtype(self) -> DtypeEnum: + if self._config.torch_dtype == torch.float16: + return DtypeEnum.fp16 + elif self._config.torch_dtype == torch.bfloat16: + return DtypeEnum.bf16 + else: + raise NotImplementedError("Only fp16 and bf16 are supported") + + @property + def mlp_activation_fn(self) -> ActivationType: + activation = self._config.hidden_act.lower() + # llama model family is special and is always gated so force gated versions of relu, gelu, silu + if activation == "gelu": + return ActivationType.GEGLU + elif activation == "relu": + return ActivationType.ReGLU + elif activation == "gegelu": + return ActivationType.GEGLU + elif activation == "silu": + return ActivationType.SiGLU + else: + raise NotImplementedError(f"Activation {activation} not supported") + + @property + def norm_type(self) -> NormTypeEnum: + return NormTypeEnum.RMSNorm + + @property + def positional_embedding_type(self) -> PositionalEmbeddingType: + return PositionalEmbeddingType.rotate_half + + """ + Forward implementations + """ + + def _forward_embed(self, ragged_batch: RaggedBatchWrapper) -> torch.Tensor: + """ + Performs the embedding lookup prior to running the transformer of the model. + + Arguments: + ragged_batch (RaggedBatchWrapper): The batch to embed. + + Returns: + torch.Tensor: The embedded batch. + """ + embed = self.embed(ragged_batch, self._non_transformer.word_emb) + + if embed.shape[-1] != self.model_dim: + raise ValueError(f"Embedding output shape {embed.shape} does not match model_dim {self.model_dim}") + + return embed + + def _forward_transformer_layer(self, layer_idx: int, residual: torch.Tensor, hidden_states: torch.Tensor, + ragged_batch_info: RaggedBatchWrapper) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Executes one (slightly offset) layer of the transformer. This implementation does a peak-ahead + optimization to fuse the layer norm of the next layer into the current layer. + + Arguments: + layer_idx (int): The index of the layer to execute. + residual (torch.Tensor): The residual tensor from the previous layer. + hidden_states (torch.Tensor): The hidden states from the previous layer. This is the + hidden states after pre normalization. + ragged_batch_info (RaggedBatchWrapper): The batch metadata. + """ + # TODO(cmikeh2): Distribute ragged_batch_info to all modules + + cur_params = self._transformer[layer_idx] + kv_cache = self.state_manager.get_cache(layer_idx) + + hidden_states = self.qkv(hidden_states, cur_params.qkv_w, b=None) + hidden_states = self.attn(hidden_states, kv_cache, + ragged_batch_info) #, inv_freqs=None) #cur_params.rotary_emb) + hidden_states = self.attn_out(hidden_states, cur_params.attn_out_w, b=None) + + if self.tp_size > 1: + dist.all_reduce(hidden_states, group=self._base_mp_group) + + residual, hidden_states = self.norm(residual, hidden_states, cur_params.mlp_norm_gamma, beta=None) + + # Should be configurable in the future + hidden_states = self.mlp_1(hidden_states, cur_params.mlp_1_w, b=None) + hidden_states = self.mlp_2(hidden_states, cur_params.mlp_2_w, b=None) + + if self.tp_size > 1: + dist.all_reduce(hidden_states, group=self._base_mp_group) + + if layer_idx != self.num_layers - 1: + next_params = self._transformer[layer_idx + 1] + residual, hidden_states = self.norm(residual, hidden_states, next_params.attn_norm_gamma, beta=None) + else: + # On last layer, we just need to perform the residual add. Adding into the residual + # here is safe. + residual.add_(hidden_states) + + return residual, hidden_states + + def _forward_unembed(self, hidden_states: torch.Tensor, ragged_batch_info: RaggedBatchWrapper) -> torch.Tensor: + """ + Performs unembedding of the hidden states to logits. This will only sample the final + token of each sequence. + """ + logits = self.unembed(hidden_states, self._non_transformer.word_unembed, ragged_batch_info, + self._non_transformer.final_norm) + + if self.tp_size > 1: + comm_buffer = empty_from(self._comm_logits, (self.tp_size, logits.shape[0], logits.shape[1])) + full_logits = empty_from(self._return_logits, (logits.shape[0], self.vocab_size)) + + dist.all_gather_into_tensor(comm_buffer, logits, group=self._base_mp_group) + + full_logits.copy_(comm_buffer.permute(1, 0, 2).reshape(logits.shape[0], self.vocab_size)) + + return full_logits + else: + return logits + + def forward(self, wrapped_batch: RaggedBatchWrapper) -> torch.Tensor: + + residual = self._forward_embed(wrapped_batch) + + residual, hidden_states = self.norm(residual, None, self._transformer[0].attn_norm_gamma, beta=None) + + for layer_idx in range(self.num_layers): + residual, hidden_states = self._forward_transformer_layer(layer_idx, residual, hidden_states, + wrapped_batch) + + return self._forward_unembed(residual, wrapped_batch) diff --git a/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_policy.py b/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_policy.py new file mode 100644 index 000000000000..65fe7b705e53 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/llama_v2/llama_v2_policy.py @@ -0,0 +1,37 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import argparse + +from typing import Any + +from ...checkpoint import CheckpointEngineBase +from ...config_v2 import RaggedInferenceEngineConfig +from ...model_implementations.inference_policy_base import ContainerMap, InferenceV2Policy +from ...model_implementations.llama_v2.llama_v2_containers import Llama2NonTransformerContainer, Llama2TransformerContainer +from ...model_implementations.llama_v2.llama_v2_model import Llama2InferenceModel + + +class Llama2Policy(InferenceV2Policy): + + def __init__(self, checkpoint_engine: CheckpointEngineBase, model_config: argparse.Namespace) -> None: + super().__init__(checkpoint_engine, model_config) + + def instantiate_model(self, engine_config: RaggedInferenceEngineConfig, mp_group: Any) -> Llama2InferenceModel: + return Llama2InferenceModel(config=self._model_config, engine_config=engine_config, base_mp_group=mp_group) + + def build_container_map(self) -> ContainerMap: + map = ContainerMap() + + transformer_containers = [Llama2TransformerContainer(self.model) for _ in range(self.model.num_layers)] + + map.set_transformer_params(['model.layers'], transformer_containers) + + map.set_non_transformer_params(Llama2NonTransformerContainer(self.model)) + + map.set_unmapped_params( + [f'model.layers.{i}.self_attn.rotary_emb.inv_freq' for i in range(self.model.num_layers)]) + + return map diff --git a/deepspeed/inference/v2/model_implementations/mistral/__init__.py b/deepspeed/inference/v2/model_implementations/mistral/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/mistral/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/deepspeed/inference/v2/model_implementations/mistral/container.py b/deepspeed/inference/v2/model_implementations/mistral/container.py new file mode 100644 index 000000000000..b4c0956f4049 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/mistral/container.py @@ -0,0 +1,77 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# Create a container object to save model-specific tensors using the policy file above. + +from deepspeed.inference.v2.model_implementations.common_parameters import * +from deepspeed.inference.v2.model_implementations.layer_container_base import LayerContainer +''' + # HF Mistral model (mistralai/Mistral-7B-v0.1) looks like this: +MistralForCausalLM( + (model): MistralModel( + (embed_tokens): Embedding(32000, 4096) + (layers): ModuleList( + (0-31): 32 x MistralDecoderLayer( + (self_attn): MistralAttention( + (q_proj): Linear(in_features=4096, out_features=4096, bias=False) + (k_proj): Linear(in_features=4096, out_features=1024, bias=False) + (v_proj): Linear(in_features=4096, out_features=1024, bias=False) + (o_proj): Linear(in_features=4096, out_features=4096, bias=False) + (rotary_emb): MistralRotaryEmbedding() + ) + (mlp): MistralMLP( + (gate_proj): Linear(in_features=4096, out_features=14336, bias=False) + (up_proj): Linear(in_features=4096, out_features=14336, bias=False) + (down_proj): Linear(in_features=14336, out_features=4096, bias=False) + (act_fn): SiLUActivation() + ) + (input_layernorm): MistralRMSNorm() + (post_attention_layernorm): MistralRMSNorm() + ) + ) + (norm): MistralRMSNorm() + ) + (lm_head): Linear(in_features=4096, out_features=32000, bias=False) +) +''' + + +class MistralTransformerContainer(LayerContainer): + """ + Transformer layer container for the Mistral model. + """ + qkv_w: UnfusedQKVParameter + attn_out_w: AttentionOutputParameter + mlp_1_w: GatedMLPParameter + mlp_2_w: MLP2Parameter + attn_norm_gamma: NormParameter + mlp_norm_gamma: NormParameter + + PARAM_MAPPING = { + "self_attn.q_proj.weight": "qkv_w.q_params", + "self_attn.k_proj.weight": "qkv_w.k_params", + "self_attn.v_proj.weight": "qkv_w.v_params", + "self_attn.o_proj.weight": "attn_out_w.params", + "mlp.gate_proj.weight": "mlp_1_w.gate_params", + "mlp.up_proj.weight": "mlp_1_w.up_params", + "mlp.down_proj.weight": "mlp_2_w.params", + "input_layernorm.weight": "attn_norm_gamma.params", + "post_attention_layernorm.weight": "mlp_norm_gamma.params", + } + + +class MistralNonTransformerContainer(LayerContainer): + """ + Non-Transformer layer container for the Mistral model. + """ + word_emb: EmbeddingParameter + word_unembed: UnembedParameter + final_norm: NormParameter + + PARAM_MAPPING = { + "model.embed_tokens.weight": "word_emb.params", + "model.norm.weight": "final_norm.params", + "lm_head.weight": "word_unembed.params", + } diff --git a/deepspeed/inference/v2/model_implementations/mistral/model.py b/deepspeed/inference/v2/model_implementations/mistral/model.py new file mode 100644 index 000000000000..d9b06b91e308 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/mistral/model.py @@ -0,0 +1,202 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Iterable, Optional, Tuple + +import torch + +import deepspeed.comm as dist + +from ...allocator import empty_from +from ...inference_utils import ActivationType, DtypeEnum +from ...model_implementations import * +from ...modules.configs import * +from ...modules.interfaces import * +from ...ragged import RaggedBatchWrapper + +from .container import MistralNonTransformerContainer, MistralTransformerContainer + + +class MistralInferenceModel(DSTransformerModelBase): + """ + Inference model implementation for ragged batching for Mistral models. + """ + + _non_transformer: Optional[MistralNonTransformerContainer] + """ + Embed + unembed container. Specializing the type annotation. + """ + + _transformer: Optional[Iterable[MistralTransformerContainer]] + """ + Per-layer transformer container. Specializing the type annotation. + """ + """ + Properties ineherited from `DSInferenceModelBase` + """ + + @property + def max_sequence_length(self) -> int: + return self._config.max_seq_length + + """ + Properties ineherited from `DSTransformerModelBase` + """ + + @property + def num_layers(self) -> int: + return self._config.num_hidden_layers + + @property + def model_dim(self) -> int: + return self._config.hidden_size + + @property + def vocab_size(self) -> int: + return self._config.vocab_size + + @property + def head_size(self) -> int: + return self.model_dim // self.n_heads + + @property + def n_heads(self) -> int: + return self._config.num_attention_heads + + @property + def intermediate_dim(self) -> int: + return self._config.intermediate_size + + @property + def n_heads_kv(self) -> int: + return self._config.num_key_value_heads + + @property + def activation_dtype(self) -> DtypeEnum: + if self._config.torch_dtype == torch.float16: + return DtypeEnum.fp16 + elif self._config.torch_dtype == torch.bfloat16: + return DtypeEnum.bf16 + else: + raise NotImplementedError("Only fp16 and bf16 are supported") + + @property + def mlp_activation_fn(self) -> ActivationType: + activation = self._config.hidden_act.lower() + if activation == "gelu": + return ActivationType.GEGLU + elif activation == "relu": + return ActivationType.ReGLU + elif activation == "gegelu": + return ActivationType.GEGLU + elif activation == "silu": + return ActivationType.SiGLU + else: + raise NotImplementedError(f"Activation {activation} not supported") + + @property + def norm_type(self) -> NormTypeEnum: + return NormTypeEnum.RMSNorm + + @property + def positional_embedding_type(self) -> PositionalEmbeddingType: + return PositionalEmbeddingType.rotate_half + + """ + Forward implementations + """ + + def _forward_embed(self, ragged_batch: RaggedBatchWrapper) -> torch.Tensor: + """ + Performs the embedding lookup prior to running the transformer of the model. + + Arguments: + ragged_batch (RaggedBatchWrapper): The batch to embed. + + Returns: + torch.Tensor: The embedded batch. + """ + embed = self.embed(ragged_batch, self._non_transformer.word_emb) + + if embed.shape[-1] != self.model_dim: + raise ValueError(f"Embedding output shape {embed.shape} does not match model_dim {self.model_dim}") + + return embed + + def _forward_transformer(self, layer_idx: int, residual: torch.Tensor, hidden_states: torch.Tensor, + ragged_batch_info: RaggedBatchWrapper) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Executes one (slightly offset) layer of the transformer. This implementation does a peak-ahead + optimization to fuse the layer norm of the next layer into the current layer. + + Arguments: + layer_idx (int): The index of the layer to execute. + residual (torch.Tensor): The residual tensor from the previous layer. + hidden_states (torch.Tensor): The hidden states from the previous layer. This is the + hidden states after pre normalization. + ragged_batch_info (RaggedBatchWrapper): The batch metadata. + """ + # TODO(cmikeh2): Distribute ragged_batch_info to all modules + + cur_params = self._transformer[layer_idx] + kv_cache = self.state_manager.get_cache(layer_idx) + + hidden_states = self.qkv(hidden_states, cur_params.qkv_w, b=None) + hidden_states = self.attn(hidden_states, kv_cache, + ragged_batch_info) #, inv_freqs=None) #cur_params.rotary_emb) + hidden_states = self.attn_out(hidden_states, cur_params.attn_out_w, b=None) + + if self.tp_size > 1: + dist.all_reduce(hidden_states, group=self._base_mp_group) + + residual, hidden_states = self.norm(residual, hidden_states, cur_params.mlp_norm_gamma, beta=None) + + # Should be configurable in the future + hidden_states = self.mlp_1(hidden_states, cur_params.mlp_1_w, b=None) + hidden_states = self.mlp_2(hidden_states, cur_params.mlp_2_w, b=None) + + if self.tp_size > 1: + dist.all_reduce(hidden_states, group=self._base_mp_group) + + if layer_idx != self.num_layers - 1: + next_params = self._transformer[layer_idx + 1] + residual, hidden_states = self.norm(residual, hidden_states, next_params.attn_norm_gamma, beta=None) + else: + # On last layer, we just need to perform the residual add. Adding into the residual + # here is safe. + residual.add_(hidden_states) + + return residual, hidden_states + + def _forward_unembed(self, hidden_states: torch.Tensor, ragged_batch_info: RaggedBatchWrapper) -> torch.Tensor: + """ + Performs unembedding of the hidden states to logits. This will only sample the final + token of each sequence. + """ + logits = self.unembed(hidden_states, self._non_transformer.word_unembed, ragged_batch_info, + self._non_transformer.final_norm) + + if self.tp_size > 1: + comm_buffer = empty_from(self._comm_logits, (self.tp_size, logits.shape[0], logits.shape[1])) + full_logits = empty_from(self._return_logits, (logits.shape[0], self.vocab_size)) + + dist.all_gather_into_tensor(comm_buffer, logits, group=self._base_mp_group) + + full_logits.copy_(comm_buffer.permute(1, 0, 2).reshape(logits.shape[0], self.vocab_size)) + + return full_logits + else: + return logits + + def forward(self, wrapped_batch: RaggedBatchWrapper) -> torch.Tensor: + + residual = self._forward_embed(wrapped_batch) + + residual, hidden_states = self.norm(residual, None, self._transformer[0].attn_norm_gamma, beta=None) + + for layer_idx in range(self.num_layers): + residual, hidden_states = self._forward_transformer(layer_idx, residual, hidden_states, wrapped_batch) + + return self._forward_unembed(residual, wrapped_batch) diff --git a/deepspeed/inference/v2/model_implementations/mistral/policy.py b/deepspeed/inference/v2/model_implementations/mistral/policy.py new file mode 100644 index 000000000000..93458827aa96 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/mistral/policy.py @@ -0,0 +1,36 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import argparse + +from typing import Any + +from deepspeed.inference.v2.checkpoint import CheckpointEngineBase +from deepspeed.inference.v2.config_v2 import RaggedInferenceEngineConfig +from deepspeed.inference.v2.model_implementations.inference_policy_base import ContainerMap, InferenceV2Policy +from deepspeed.inference.v2.model_implementations.mistral.container import MistralNonTransformerContainer, MistralTransformerContainer +from deepspeed.inference.v2.model_implementations.mistral.model import MistralInferenceModel + + +class MistralPolicy(InferenceV2Policy): + + def __init__(self, checkpoint_engine: CheckpointEngineBase, model_config: argparse.Namespace) -> None: + super().__init__(checkpoint_engine, model_config) + + def instantiate_model(self, engine_config: RaggedInferenceEngineConfig, mp_group: Any) -> MistralInferenceModel: + return MistralInferenceModel(config=self._model_config, engine_config=engine_config, base_mp_group=mp_group) + + def build_container_map(self) -> ContainerMap: + map = ContainerMap() + + transformer_containers = [MistralTransformerContainer(self.model) for _ in range(self.model.num_layers)] + + map.set_transformer_params(['model.layers'], transformer_containers) + + map.set_non_transformer_params(MistralNonTransformerContainer(self.model)) + + map.set_unmapped_params([]) + + return map diff --git a/deepspeed/inference/v2/model_implementations/opt/__init__.py b/deepspeed/inference/v2/model_implementations/opt/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/opt/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/deepspeed/inference/v2/model_implementations/opt/container.py b/deepspeed/inference/v2/model_implementations/opt/container.py new file mode 100644 index 000000000000..5b1c9ce4c8a3 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/opt/container.py @@ -0,0 +1,95 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# Create a container object to save model-specific tensors using the policy file above. + +from ...model_implementations.common_parameters import * +from ...model_implementations.layer_container_base import LayerContainer +''' + # HF OPT model looks like this: + +OPTForCausalLM( + (model): OPTModel( + (decoder): OPTDecoder( + (embed_tokens): Embedding(50272, 768, padding_idx=1) + (embed_positions): OPTLearnedPositionalEmbedding(2050, 768) + (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True) + (layers): ModuleList( + (0-11): 12 x OPTDecoderLayer( + (self_attn): OPTAttention( + (k_proj): Linear(in_features=768, out_features=768, bias=True) + (v_proj): Linear(in_features=768, out_features=768, bias=True) + (q_proj): Linear(in_features=768, out_features=768, bias=True) + (out_proj): Linear(in_features=768, out_features=768, bias=True) + ) + (activation_fn): ReLU() + (self_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True) + (fc1): Linear(in_features=768, out_features=3072, bias=True) + (fc2): Linear(in_features=3072, out_features=768, bias=True) + (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True) + ) + ) + ) + ) + (lm_head): Linear(in_features=768, out_features=50272, bias=False) +) + +''' + + +class OPTTransformerContainer(LayerContainer): + """ + Transformer layer container for the OPT model. + """ + qkv_w: UnfusedQKVParameter + qkv_b: UnfusedQKVParameter + attn_out_w: AttentionOutputParameter + attn_out_b: AttentionOutputParameter + mlp_1_w: MLP1Parameter + mlp_1_b: MLP1Parameter + mlp_2_w: MLP2Parameter + mlp_2_b: MLP2Parameter + attn_norm_beta: NormParameter + attn_norm_gamma: NormParameter + mlp_norm_beta: NormParameter + mlp_norm_gamma: NormParameter + + PARAM_MAPPING = { + "self_attn.q_proj.weight": "qkv_w.q_params", + "self_attn.q_proj.bias": "qkv_b.q_params", + "self_attn.k_proj.weight": "qkv_w.k_params", + "self_attn.k_proj.bias": "qkv_b.k_params", + "self_attn.v_proj.weight": "qkv_w.v_params", + "self_attn.v_proj.bias": "qkv_b.v_params", + "self_attn.out_proj.weight": "attn_out_w.params", + "self_attn.out_proj.bias": "attn_out_b.params", + "fc1.weight": "mlp_1_w.params", + "fc1.bias": "mlp_1_b.params", + "fc2.weight": "mlp_2_w.params", + "fc2.bias": "mlp_2_b.params", + "self_attn_layer_norm.weight": "attn_norm_gamma.params", + "self_attn_layer_norm.bias": "attn_norm_beta.params", + "final_layer_norm.weight": "mlp_norm_gamma.params", + "final_layer_norm.bias": "mlp_norm_beta.params", + } + + +class OPTNonTransformerContainer(LayerContainer): + """ + Non-Transformer layer container for the OPT model. + """ + word_emb: EmbeddingParameter + word_emb_pos: EmbeddingParameter + word_unembed: UnembedParameter + final_norm_w: NormParameter + final_norm_b: NormParameter + + PARAM_MAPPING = { + "model.decoder.embed_tokens.weight": "word_emb.params", + "model.decoder.embed_positions.weight": "word_emb_pos.params", + "model.decoder.final_layer_norm.weight": "final_norm_w.params", + "model.decoder.final_layer_norm.bias": "final_norm_b.params", + "lm_head.weight": "word_unembed.params", + } diff --git a/deepspeed/inference/v2/model_implementations/opt/model.py b/deepspeed/inference/v2/model_implementations/opt/model.py new file mode 100644 index 000000000000..fa221e15a0b7 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/opt/model.py @@ -0,0 +1,195 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Iterable, Optional, Tuple + +import torch + +import deepspeed.comm as dist + +from ...allocator import empty_from +from ...inference_utils import ActivationType, DtypeEnum +from ...model_implementations import * +from ...modules.configs import ( + DSEmbeddingsConfig, + NormTypeEnum, + PositionalEmbeddingType, +) +from ...ragged import RaggedBatchWrapper +from .container import OPTNonTransformerContainer, OPTTransformerContainer + +from ...modules.heuristics import instantiate_embed + + +class OPTInferenceModel(DSTransformerModelBase): + """ + Inference model implementation for ragged batching for OPT models. + """ + + _non_transformer: Optional[OPTNonTransformerContainer] + """ + Embed + unembed container. Specializing the type annotation. + """ + + _transformer: Optional[Iterable[OPTTransformerContainer]] + """ + Per-layer transformer container. Specializing the type annotation. + """ + """ + Properties ineherited from `DSInferenceModelBase` + """ + + @property + def max_sequence_length(self) -> int: + return self._config.max_seq_length + + """ + Properties ineherited from `DSTransformerModelBase` + """ + + @property + def num_layers(self) -> int: + return self._config.num_hidden_layers + + @property + def model_dim(self) -> int: + return self._config.hidden_size + + @property + def vocab_size(self) -> int: + return self._config.vocab_size + + @property + def head_size(self) -> int: + return self.model_dim // self.n_heads + + @property + def n_heads(self) -> int: + return self._config.num_attention_heads + + @property + def intermediate_dim(self) -> int: + return self._config.ffn_dim + + @property + def activation_dtype(self) -> DtypeEnum: + if self._config.torch_dtype == torch.float16: + return DtypeEnum.fp16 + elif self._config.torch_dtype == torch.bfloat16: + return DtypeEnum.bf16 + else: + raise NotImplementedError("Only fp16 and bf16 are supported") + + @property + def mlp_activation_fn(self) -> ActivationType: + return ActivationType.RELU + + @property + def norm_type(self) -> NormTypeEnum: + return NormTypeEnum.LayerNorm + + @property + def positional_embedding_type(self) -> PositionalEmbeddingType: + return PositionalEmbeddingType.none + + """ + Overrides of ``DSTransformerModelBase`` methods + """ + + def make_embedding_layer(self) -> None: + """ + Performs setup and creates embedding DSModule. Since OPT includes trained + positional embeddings, we will override the base model implementation. + """ + + embed_config = DSEmbeddingsConfig(max_tokens=self._engine_config.state_manager.max_ragged_batch_size, + residual_dtype=self.activation_dtype, + embedding_dim=self.model_dim, + positional_embedding=True, + positional_offset=2) + + self.embed = instantiate_embed(embed_config, self._engine_config) + + """ + Forward implementations + """ + + def _forward_embed(self, ragged_batch: RaggedBatchWrapper) -> torch.Tensor: + embed = self.embed(ragged_batch, self._non_transformer.word_emb, self._non_transformer.word_emb_pos) + if embed.shape[-1] != self.model_dim: + raise ValueError(f"Embedding output shape {embed.shape} does not match model_dim {self.model_dim}") + + return embed + + def _forward_transformer_layer(self, layer_idx: int, residual: torch.Tensor, hidden_states: torch.Tensor, + ragged_batch_info: RaggedBatchWrapper) -> Tuple[torch.Tensor, torch.Tensor]: + # TODO(cmikeh2): Distribute ragged_batch_info to all modules + + cur_params = self._transformer[layer_idx] + kv_cache = self.state_manager.get_cache(layer_idx) + + hidden_states = self.qkv(hidden_states, cur_params.qkv_w, b=cur_params.qkv_b) + hidden_states = self.attn(hidden_states, kv_cache, + ragged_batch_info) #, inv_freqs=None) #cur_params.rotary_emb) + hidden_states = self.attn_out(hidden_states, cur_params.attn_out_w, b=cur_params.attn_out_b) + + if self.tp_size > 1: + dist.all_reduce(hidden_states, group=self._base_mp_group) + + residual, hidden_states = self.norm(residual, + hidden_states, + cur_params.mlp_norm_gamma, + beta=cur_params.mlp_norm_beta) + + # Should be configurable in the future + hidden_states = self.mlp_1(hidden_states, cur_params.mlp_1_w, b=cur_params.mlp_1_b) + hidden_states = self.mlp_2(hidden_states, cur_params.mlp_2_w, b=cur_params.mlp_2_b) + + if self.tp_size > 1: + dist.all_reduce(hidden_states, group=self._base_mp_group) + + if layer_idx != self.num_layers - 1: + next_params = self._transformer[layer_idx + 1] + residual, hidden_states = self.norm(residual, + hidden_states, + next_params.attn_norm_gamma, + beta=next_params.attn_norm_beta) + else: + # On last layer, we just need to perform the residual add. Adding into the residual + # here is safe. + residual.add_(hidden_states) + + return residual, hidden_states + + def _forward_unembed(self, hidden_states: torch.Tensor, ragged_batch_info: RaggedBatchWrapper) -> torch.Tensor: + logits = self.unembed(hidden_states, self._non_transformer.word_unembed, ragged_batch_info, + self._non_transformer.final_norm_w, self._non_transformer.final_norm_b) + + if self.tp_size > 1: + comm_buffer = empty_from(self._comm_logits, (self.tp_size, logits.shape[0], logits.shape[1])) + full_logits = empty_from(self._return_logits, (logits.shape[0], self.vocab_size)) + + dist.all_gather_into_tensor(comm_buffer, logits, group=self._base_mp_group) + + full_logits.copy_(comm_buffer.permute(1, 0, 2).reshape(logits.shape[0], self.vocab_size)) + + return full_logits + else: + return logits + + def forward(self, wrapped_batch: RaggedBatchWrapper) -> torch.Tensor: + + residual = self._forward_embed(wrapped_batch) + + residual, hidden_states = self.norm(residual, + None, + self._transformer[0].attn_norm_gamma, + beta=self._transformer[0].attn_norm_beta) + + for layer_idx in range(self.num_layers): + residual, hidden_states = self._forward_transformer_layer(layer_idx, residual, hidden_states, + wrapped_batch) + + return self._forward_unembed(residual, wrapped_batch) diff --git a/deepspeed/inference/v2/model_implementations/opt/policy.py b/deepspeed/inference/v2/model_implementations/opt/policy.py new file mode 100644 index 000000000000..0f5002cdaa54 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/opt/policy.py @@ -0,0 +1,36 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import argparse + +from typing import Any + +from ...checkpoint import CheckpointEngineBase +from ...config_v2 import RaggedInferenceEngineConfig +from ...model_implementations.inference_policy_base import ContainerMap, InferenceV2Policy +from ...model_implementations.opt.container import OPTNonTransformerContainer, OPTTransformerContainer +from ...model_implementations.opt.model import OPTInferenceModel + + +class OPTPolicy(InferenceV2Policy): + + def __init__(self, checkpoint_engine: CheckpointEngineBase, model_config: argparse.Namespace) -> None: + super().__init__(checkpoint_engine, model_config) + + def instantiate_model(self, engine_config: RaggedInferenceEngineConfig, mp_group: Any) -> OPTInferenceModel: + return OPTInferenceModel(config=self._model_config, engine_config=engine_config, base_mp_group=mp_group) + + def build_container_map(self) -> ContainerMap: + map = ContainerMap() + + transformer_containers = [OPTTransformerContainer(self.model) for _ in range(self.model.num_layers)] + + map.set_transformer_params(['model.decoder.layers'], transformer_containers) + + map.set_non_transformer_params(OPTNonTransformerContainer(self.model)) + + map.set_unmapped_params([]) + + return map diff --git a/deepspeed/inference/v2/model_implementations/parameter_base.py b/deepspeed/inference/v2/model_implementations/parameter_base.py new file mode 100644 index 000000000000..a413c6c4027a --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/parameter_base.py @@ -0,0 +1,257 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import weakref +from abc import abstractmethod +from typing import Type + +import torch + +# Currently have dependency loops for the type hints. +InferenceModel = Type["InferenceModel"] +LayerContainer = Type["LayerContainer"] +ParametrizedList = Type["ParametrizedList"] + +MAPPING_KEY = "PARAM_MAPPING" +PLIST_HELPERS = "_ds_plist_strip_vals" + + +def make_param_getter(clsname, param): + """ + Normal getter implementation for a property. + """ + + def param_getter(self): + return getattr(self, f"__{clsname}__{param}") + + return param_getter + + +def make_param_setter(clsname, param): + """ + Setter implementation that will call complete component to potentially + finalize the parameter. + """ + + def param_setter(self, value): + setattr(self, f"__{clsname}__{param}", value) + self.complete_component() + + return param_setter + + +def make_readonly_setter(): + """ + Setter implementation that will raise an error if called. + """ + + def paramlist_setter(self, value): + raise ValueError("Cannot set a ParametrizedList directly.") + + return paramlist_setter + + +class ParameterMetaclass(type): + """ + MetaClass for the ParameterBase base class. This class will parse the `src_params` + attribute and create properties for each of the dependencies. A dependency can either + be represented as a string, which is interpreted as a named Tensor, or a `ParametrizedList` + subclass. + """ + + def __new__(cls, clsname, bases, attrs): + + annotations = attrs.get("__annotations__", {}) + dependencies = { + name: annotation + for name, annotation in annotations.items() if issubclass(annotation, (torch.Tensor, ParametrizedList)) + } + n_dependencies = len(dependencies) + + # Create properties for each of our dependencies + for d_name, d_type in dependencies.items(): + if issubclass(d_type, ParametrizedList): + assert hasattr( + d_type, "count_attr" + ), "ParametrizedList must have a count_attr attribute to access on the inference module." + attrs[d_name] = property(make_param_getter(clsname, d_name), make_readonly_setter()) + else: # torch.Tensor + attrs[d_name] = property(make_param_getter(clsname, d_name), make_param_setter(clsname, d_name)) + + new_cls = super().__new__(cls, clsname, bases, attrs) + new_cls.n_dependencies = n_dependencies + + return new_cls + + def __call__(cls, *args, **kwargs): + new_obj = super().__call__(*args, **kwargs) + new_obj.__init__(*args, **kwargs) + + setattr(new_obj, "dest_param", None) + + # Initialize our dependences to None/empty `ParametrizedList`s + for name, annotation in new_obj.__annotations__.items(): + if issubclass(annotation, ParametrizedList): + #TODO(jeff): update assert with this, model implementation attribute does not align or missing wrt the ParametrizedList attributes + assert hasattr( + new_obj.inference_model, annotation.count_attr + ), f"new_obj={new_obj.__class__.__name__}, name={name}, annotation.count_attr={annotation.count_attr}" + param_list = annotation(new_obj, getattr(new_obj.inference_model, annotation.count_attr)) + setattr(new_obj, f"__{new_obj.__class__.__name__}__{name}", param_list) + else: # torch.Tensor + setattr(new_obj, f"__{new_obj.__class__.__name__}__{name}", None) + + return new_obj + + +class ParameterBase(metaclass=ParameterMetaclass): + """ + A ParameterBase allows us to consolidate tracking the dependencies of loading a parameter from + a checkpoint into a single object. This class should not be used directly, but rather subclassed + and the `src_params` attribute set to a list of strings and/or `ParametrizedList`s. + """ + + # inference_model: InferenceModel + """ + Inference model that will provide context on how to shard and transform the parameter. + """ + + #completed_components: int + """ + How many of the layer dependencies have been met. This is used to determine when the parameter + is ready to be finalized. A ParametrizedList counts as a single dependency for the purposes + of this counter. + """ + + def __init__(self, model: InferenceModel, parent_container: LayerContainer) -> None: + """ + Direct constructor. This should not be called from client code. + + Args: + model (InferenceModel): Inference model that will be used to shard and transform the + parameter in `finalize`. + parent_container (LayerContainer): The parent container that this parameter is a member + of. We will build a weakref to this container to call the finalization callback. + """ + self.inference_model = model + self.completed_components = 0 + self.parent_container = weakref.ref(parent_container) + + @abstractmethod + def finalize(self) -> torch.Tensor: + """ + Finalize the parameter after all of its source parameters have been set. This method + will be automatically called when all inputs have been set. It should return the Tensor + with all transformations performed on it. + """ + pass + + def complete_component(self) -> None: + """ + Mark a component as completed. This should be called by the relevant setter of a direct + property or a ParametrizedList. This method will automatically call `finalize` when all + dependencies have been met and then call the finalization callback on the parent container. + + Once the finalization callback has been called, the parameter will be replaced with the + `dst_param` attribute on the parent container, and this instance will be destroyed. + """ + self.completed_components += 1 + + if self.completed_components != self.n_dependencies: + return + + finalized_param = self.finalize() + self.parent_container().finalization_callback(self, finalized_param) + + +class ParametrizedList: + """ + A ParametrizedList is a list of parameters that are dependencies + of a `ParameterBase` but may vary in length depending on the model + configuration (rather than architecture). For example, a MoE layer + may have different number of experts depending on the size of the model. + + This class is used to manage these lists and provide integer indexing + of a single component rather than accessing names directly. For example, + it tends to be more natural to access the 8th expert with `experts[8]` + rather than a name like `expert_8`, especially as an attribute. + + To inherit from this class, set static variables `name` and `count_attr`. + + ```python + class MyParametrizedList(ParametrizedList): + count_attr: str = "my_list_count" + ``` + + In the above example, `my_list_count` should be an accessible attribute + of the inference model (i.e. via `self.inference_model.my_list_count`). + + NOTE: There are some APIs in which this type cannot be used as if it is + just a list of Tensors. For example, `torch.cat(param_list)` will not work. + However, you can make it compatible with a tuple wrapper: + `torch.cat(tuple(param_list))` + """ + + n_params: int + """ + Number of params this list contains. + """ + + param: ParameterBase + """ + WeakRef to the owning parameter. + """ + + def __init__(self, param: ParameterBase, n_params: int) -> None: + """ + Constructor. Should not be called from client code. + + Args: + param (ParameterBase): The owning parameter. + n_params (int): The number of parameters this list contains. This should be + """ + self.n_params = n_params + self.set_params = 0 + self.param = weakref.ref(param) + self._params = [None] * n_params + + def __getitem__(self, index): + return self._params[index] + + def __setitem__(self, index, value): + if self._params[index] is not None: + raise ValueError("Cannot set a parameter twice.") + + self._params[index] = value + self.set_params += 1 + + if self.set_params != self.n_params: + return + + self.param().complete_component() + + def __iter__(self): + return iter(self._params) + + +def ParamList(attr: str): + """ + Helper to create a subclass of ParametrizedList with the desired `count_attr`. + + In this manner, we can annotate the type of a Parameter dependency with the + following: + + ```python + class CustomParameter(ParameterBase): + dependency_list: ParamList("dependencies_count_name") + ``` + + where "dependencies_count_name" is the name of the attribute on the inference model. + """ + + class ParametrizedListInstance(ParametrizedList): + count_attr: str = attr + + return ParametrizedListInstance diff --git a/deepspeed/inference/v2/model_implementations/sharding/__init__.py b/deepspeed/inference/v2/model_implementations/sharding/__init__.py new file mode 100644 index 000000000000..63421bc1c622 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/__init__.py @@ -0,0 +1,12 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .attn import * +from .attn_out import * +from .embedding import * +from .mlp import * +from .qkv import * +from .types import * +from .unembed import * diff --git a/deepspeed/inference/v2/model_implementations/sharding/attn.py b/deepspeed/inference/v2/model_implementations/sharding/attn.py new file mode 100644 index 000000000000..de8d6f6ac4c5 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/attn.py @@ -0,0 +1,63 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional, Tuple + + +def get_local_heads(shard_rank: int, + num_shards: int, + n_heads_q: int, + n_heads_kv: Optional[int] = None) -> Tuple[int, int]: + """ + Helper to determine the number of local heads of a given shard. + + Args: + shard_rank (int): The rank of the shard. + num_shards (int): The total number of shards that attention is distributed over. + n_heads_q (int): The number of query heads. + n_heads_kv (int): The number of key/value heads. If not passed, it is assumed that + the number of query and key/value heads are the same. + """ + if n_heads_q < num_shards: + raise ValueError("There must be at least as many attention heads as there are shards.") + + if n_heads_kv is None or n_heads_kv == n_heads_q: + # MHA attention + base_heads = n_heads_q // num_shards + extra_heads = n_heads_q % num_shards + + if shard_rank < extra_heads: + return (base_heads + 1), (base_heads + 1) + else: + return base_heads, base_heads + else: + # GQA attention + if n_heads_q % n_heads_kv != 0: + raise ValueError("Must be an even ratio between query and key/value heads.") + + if n_heads_kv < num_shards and num_shards % n_heads_kv != 0: + raise ValueError( + "If splitting a group across multiple shards, we must be able to distribute the groups evenly.") + + if n_heads_kv >= num_shards and n_heads_kv % num_shards != 0: + raise ValueError("If parallelizing groups, must be able to evenly distribute them.") + + q_ratio = n_heads_q // n_heads_kv + + if n_heads_kv >= num_shards: + local_kv_heads = n_heads_kv // num_shards + local_q_heads = local_kv_heads * q_ratio + return local_q_heads, local_kv_heads + else: + group_sharding_size = num_shards // n_heads_kv + group_rank_idx = shard_rank % group_sharding_size + + base_heads = q_ratio // group_sharding_size + extra_heads = q_ratio % group_sharding_size + + if group_rank_idx < extra_heads: + return (base_heads + 1), 1 + else: + return base_heads, 1 diff --git a/deepspeed/inference/v2/model_implementations/sharding/attn_out.py b/deepspeed/inference/v2/model_implementations/sharding/attn_out.py new file mode 100644 index 000000000000..ce7c105531ea --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/attn_out.py @@ -0,0 +1,111 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from .types import ShardingType +from .utils import shard_param, get_shard_endpoints + + +def shard_attn_out_param(param: torch.Tensor, + shard_rank: int, + num_shards: int, + head_size: int, + n_heads_q: Optional[int] = None, + n_heads_kv: Optional[int] = None) -> Optional[torch.Tensor]: + """ + Utility method for sharding an attention output parameter. + """ + if len(param.shape) == 1: + # We will do the bias addition on the 0th rank only rather than scale the parameter and + # implicitly reconstruct this in the distributed reduce. + return param if shard_rank == 0 else None + + assert n_heads_kv is None or (n_heads_q is not None + and n_heads_kv is not None), "n_heads_kv should not be passed without n_heads_q" + + mha_sharding = n_heads_kv is None or n_heads_q == n_heads_kv + + if mha_sharding: + return shard_param(param, ShardingType.INNER_DIMENSION, shard_rank, num_shards, granularity=head_size) + else: + assert param.shape[0] == head_size * n_heads_q, "GQA param shape is not correct" + + # 32 KV heads, 16 shards for example + even_kv_sharding = n_heads_kv % num_shards == 0 + + # 8 KV heads, 16 shards for example + even_kv_distribution = num_shards % n_heads_kv == 0 + + assert even_kv_sharding or even_kv_distribution, "No partitioning algorithm for this yet." + + if even_kv_sharding: + # Same as original sharding scenario + return shard_param(param, ShardingType.INNER_DIMENSION, shard_rank, num_shards, granularity=head_size) + else: + # We will first do a sharding on the KV and Q to map to the one KV shard per group of Q. + q_sharding_degree = num_shards // n_heads_kv + + kv_head = shard_rank // q_sharding_degree + + q_sharding_rank = shard_rank % q_sharding_degree + q_factor = n_heads_q // n_heads_kv + + q_chunk = param[..., q_factor * kv_head * head_size:q_factor * (kv_head + 1) * head_size] + + return shard_param(q_chunk, + ShardingType.INNER_DIMENSION, + q_sharding_rank, + q_sharding_degree, + granularity=head_size) + + +def attn_out_in_features(out_features: int, + shard_rank: int, + num_shards: int, + head_size: int, + n_heads_q: Optional[int] = None, + n_heads_kv: Optional[int] = None) -> int: + """ + Helper to calculate the expected output projection dimension of a QKV projection matrix. + + Args: + in_features (int): The model dimension. + shard_rank (int): Which rank to return the corresponding size for. + num_shards (int): The total number of shards the parameter is distributed across. + head_size (int): The size of each attention head. + n_heads_q (int): The number of query heads on the model. This only needs to be passed if the number + of query and key/value heads are different. If passed without n_heads_kv, default + MHA partitioning will be used. + n_heads_kv (int): The number of key and value heads on the model. This only needs to be passed + if the number of query and key/value heads are different. This argument cannot be passed without + also passing n_heads_q (we want to explicitly opt into GQA sharding). + """ + assert n_heads_kv is None or (n_heads_q is not None + and n_heads_kv is not None), "n_heads_kv should not be passed without n_heads_q" + + mha_sharding = n_heads_kv is None or n_heads_q == n_heads_kv + + if mha_sharding: + endpoints = get_shard_endpoints(out_features, shard_rank, num_shards, granularity=head_size) + return endpoints[1] - endpoints[0] + else: + if n_heads_kv >= num_shards: + assert n_heads_kv % num_shards == 0, "No partitioning algorithm for this yet." + n_local_groups = n_heads_kv // num_shards + group_size = n_heads_q // n_heads_kv + + return n_local_groups * head_size * group_size + else: + assert num_shards % n_heads_kv == 0, "No partitioning algorithm for this yet." + q_split_degree = num_shards // n_heads_kv + q_split_rank = shard_rank % q_split_degree + split_granularity = (n_heads_q // n_heads_kv) * head_size + + q_endpoints = get_shard_endpoints(split_granularity, q_split_rank, q_split_degree, granularity=head_size) + + return q_endpoints[1] - q_endpoints[0] diff --git a/deepspeed/inference/v2/model_implementations/sharding/embedding.py b/deepspeed/inference/v2/model_implementations/sharding/embedding.py new file mode 100644 index 000000000000..00d335768ae6 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/embedding.py @@ -0,0 +1,34 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from .types import ShardingType +from .utils import shard_param, get_shard_endpoints + + +def shard_embedding_param(param: torch.Tensor, shard_rank: int, num_shards: int) -> torch.Tensor: + """ + Utility method for sharding an embedding parameter. + + Args: + param (torch.Tensor): The parameter to shard. Should be of shape [vocab_size, model_dim] + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + """ + return shard_param(param, ShardingType.INNER_DIMENSION, shard_rank, num_shards) + + +def sharded_embedding_dim(embedding_size: int, shard_rank: int, num_shards: int) -> int: + """ + Utility method for getting the size of the embedding dimension of a sharded embedding. + + Args: + embedding_size (int): The size of the embedding. + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + """ + start_idx, end_idx = get_shard_endpoints(embedding_size, shard_rank, num_shards) + return end_idx - start_idx diff --git a/deepspeed/inference/v2/model_implementations/sharding/mlp.py b/deepspeed/inference/v2/model_implementations/sharding/mlp.py new file mode 100644 index 000000000000..8abd0ff8622d --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/mlp.py @@ -0,0 +1,75 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from .types import ShardingType, DEFAULT_SHARD_GRANULARITY +from .utils import shard_param, get_shard_endpoints + + +def shard_mlp_1_param(param: torch.Tensor, + shard_rank: int, + num_shards: int, + gated: bool = False, + is_moe: bool = False) -> torch.Tensor: + """ + Utility method for sharding an MLP 1 parameter. Both biases and weights are supported, as well + as for fused weights for MoE. + + Args: + param (torch.Tensor): The parameter to shard. + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + gated (bool): Whether or not the parameter is from a gated MLP. + """ + bias_dims = 2 if is_moe else 1 + + if gated: + return shard_param(param, + ShardingType.OUTER_DIMENSION, + shard_rank, + num_shards, + granularity=DEFAULT_SHARD_GRANULARITY * 2, + bias_dims=bias_dims) + else: + return shard_param(param, ShardingType.OUTER_DIMENSION, shard_rank, num_shards, bias_dims=bias_dims) + + +def shard_mlp_2_param(param: torch.Tensor, + shard_rank: int, + num_shards: int, + is_moe: bool = False) -> Optional[torch.Tensor]: + """ + Utility method for sharding an MLP 2 parameter. + + Args: + param (torch.Tensor): The parameter to shard. + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + is_moe (bool): Whether or not the parameter is from a MoE model. + """ + bias_dim_size = 2 if is_moe else 1 + + if len(param.shape) == bias_dim_size: + # We will do the bias addition on the 0th rank only rather than scale the parameter and + # implicitly reconstruct this in the distributed reduce. + return param if shard_rank == 0 else None + + return shard_param(param, ShardingType.INNER_DIMENSION, shard_rank, num_shards) + + +def sharded_intermediate_dim(intermediate_size: int, num_shards: int, shard_rank: int) -> int: + """ + Utility method for getting the size of the intermediate dimension of a sharded MLP. + + Args: + intermediate_size (int): The size of the intermediate dimension. + num_shards (int): The total number of shards the parameter is distributed across. + shard_rank (int): Which shard of the partitioned tensor to return. + """ + endpoints = get_shard_endpoints(intermediate_size, shard_rank, num_shards) + return endpoints[1] - endpoints[0] diff --git a/deepspeed/inference/v2/model_implementations/sharding/qkv.py b/deepspeed/inference/v2/model_implementations/sharding/qkv.py new file mode 100644 index 000000000000..2b6d7f40836e --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/qkv.py @@ -0,0 +1,166 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from .types import ShardingType +from .utils import shard_param, get_shard_endpoints + + +def shard_qkv_param(param: torch.Tensor, + shard_rank: int, + num_shards: int, + head_size: int, + n_heads_q: Optional[int] = None, + n_heads_kv: Optional[int] = None) -> Optional[torch.Tensor]: + """ + Utility method for sharding a QKV parameter. Both biases and weights are supported. It is assumed + that the layout of the parameter is such that all Q heads, all K heads, and all V heads + are contiguous with respect to each other. + + Args: + param (torch.Tensor): The parameter to shard. + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + head_size (int): The size of each head. + n_heads_q (int): The number of query heads. This only needs to be passed if the number + of query and key/value heads are different. If passed without n_heads_kv, default + MHA partitioning will be used. + n_heads_kv (int): The number of key/value heads. This only needs to be passed if the number + of query and key/value heads are different. This argument should not be passed without + n_heads_q (we want to explicitly opt into GQA sharding). + """ + if n_heads_kv is not None and n_heads_q is None: + raise ValueError("n_heads_kv should not be passed without n_heads_q") + + if n_heads_q is None: + # Guaranteed to be in MHA + if param.shape[0] // 3 % head_size != 0: + raise ValueError("MHA param shape is not correct") + n_heads_q = param.shape[0] // head_size // 3 + mha_sharding = True + else: + mha_sharding = n_heads_q == n_heads_kv + + if n_heads_q < num_shards: + raise ValueError("There must be at least as many query heads as there are shards.") + + if mha_sharding: + return shard_param(param, + ShardingType.OUTER_DIMENSION, + shard_rank, + num_shards, + num_concatenated_matrices=3, + granularity=head_size) + else: + if n_heads_q % n_heads_kv != 0: + raise ValueError("Must be an even ratio between query and key/value heads.") + + if param.shape[0] != head_size * (n_heads_q + 2 * n_heads_kv): + raise ValueError("GQA param shape is not correct") + + # 32 KV heads, 16 shards for example + if n_heads_kv >= num_shards and n_heads_kv % num_shards != 0: + raise ValueError("Currently do not support uneven partitioning of KV heads for GQA.") + + # 8 KV heads, 16 shards for example + if n_heads_kv < num_shards and num_shards % n_heads_kv != 0: + raise ValueError("Currently do not support distributing KV heads across different numbers of shards.") + else: + even_kv_sharding = n_heads_kv >= num_shards + + if param is None: + return None + + q_param = param[:head_size * n_heads_q] + kv_param = param[head_size * n_heads_q:] + + if even_kv_sharding: + # This is equivalent to the original sharding algorithm since n_heads_q = C * n_heads_kv. + # If n_heads_kv % num_shards == 0, then n_heads_q % num_shards == 0. + q_param = shard_param(q_param, ShardingType.OUTER_DIMENSION, shard_rank, num_shards, granularity=head_size) + kv_param = shard_param(kv_param, + ShardingType.OUTER_DIMENSION, + shard_rank, + num_shards, + num_concatenated_matrices=2, + granularity=head_size) + return torch.cat([q_param, kv_param], dim=0) + else: + # We will first do a sharding on the KV and Q to map to the one KV shard per group of Q. + q_sharding_degree = num_shards // n_heads_kv + + kv_head = shard_rank // q_sharding_degree + k_param = kv_param[kv_head * head_size:(kv_head + 1) * head_size] + v_param = kv_param[(n_heads_kv + kv_head) * head_size:(n_heads_kv + kv_head + 1) * head_size] + + q_sharding_rank = shard_rank % q_sharding_degree + q_factor = n_heads_q // n_heads_kv + + q_chunk = q_param[q_factor * kv_head * head_size:q_factor * (kv_head + 1) * head_size] + + q_param = shard_param(q_chunk, + ShardingType.OUTER_DIMENSION, + q_sharding_rank, + q_sharding_degree, + granularity=head_size) + + return torch.cat([q_param, k_param, v_param], dim=0) + + +def qkv_out_features(in_features: int, + shard_rank: int, + num_shards: int, + head_size: int, + n_heads_q: Optional[int] = None, + n_heads_kv: Optional[int] = None) -> int: + """ + Helper to calculate the expected output projection dimension of a QKV projection matrix. + + Args: + in_features (int): The model dimension. + shard_rank (int): Which rank to return the corresponding size for. + num_shards (int): The total number of shards the parameter is distributed across. + head_size (int): The size of each head. + n_heads_q (int): The number of query heads. This only needs to be passed if the number + of query and key/value heads are different. If passed without n_heads_kv, default + MHA partitioning will be used. + n_heads_kv (int): The number of key/value heads. This only needs to be passed if the number + of query and key/value heads are different. This argument cannot be passed without also + passing n_heads_q (we want to explicitly opt into GQA sharding). + """ + if n_heads_kv is not None and n_heads_q is None: + raise ValueError("n_heads_kv should not be passed without n_heads_q") + + mha_sharding = n_heads_kv is None or n_heads_q == n_heads_kv + + if n_heads_q is not None and in_features != head_size * n_heads_q: + raise ValueError("in_features is not consistent with n_heads_q and head_size") + + if mha_sharding: + endpoints = get_shard_endpoints(in_features, shard_rank, num_shards, granularity=head_size) + return (endpoints[1] - endpoints[0]) * 3 + else: + if n_heads_kv >= num_shards: + if n_heads_kv % num_shards != 0: + raise ValueError("The KV heads must be evenly distributed across the shards.") + + n_local_groups = n_heads_kv // num_shards + group_size = n_heads_q // n_heads_kv + + return n_local_groups * head_size * (2 + group_size) + else: + if num_shards % n_heads_kv != 0: + raise ValueError("A shared KV head must always partition across the same number of shards.") + + q_split_degree = num_shards // n_heads_kv + q_split_rank = shard_rank % q_split_degree + split_granularity = (n_heads_q // n_heads_kv) * head_size + + q_endpoints = get_shard_endpoints(split_granularity, q_split_rank, q_split_degree, granularity=head_size) + + return (q_endpoints[1] - q_endpoints[0]) + 2 * head_size diff --git a/deepspeed/inference/v2/model_implementations/sharding/types.py b/deepspeed/inference/v2/model_implementations/sharding/types.py new file mode 100644 index 000000000000..01dce0db523a --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/types.py @@ -0,0 +1,18 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from enum import Enum + +DEFAULT_SHARD_GRANULARITY = 32 + + +class ShardingType(Enum): + # Inner dimension sharding corresponds to splitting the Tensor along the K-dimension + # of a matrix multiplication. This would be used for attention_output or MLP2. + INNER_DIMENSION = 1 + + # Outer dimension sharding corresponds to splitting the Tensor along the N-dimension + # of a matrix multiplication. This would be used for the QKV and MLP1 projections. + OUTER_DIMENSION = 0 diff --git a/deepspeed/inference/v2/model_implementations/sharding/unembed.py b/deepspeed/inference/v2/model_implementations/sharding/unembed.py new file mode 100644 index 000000000000..6cc771969ad9 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/unembed.py @@ -0,0 +1,41 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from .types import ShardingType +from .utils import shard_param, get_shard_endpoints + + +def shard_unembed_param(param: torch.Tensor, shard_rank: int, num_shards: int) -> torch.Tensor: + """ + Utility method for sharding an unembed parameter. We shard unembeddings on the vocab dimension + with the expectation of an all-gather to produce the full results. + + TODO(cmikeh2): Really ideal would be if MII could have access to the comm and we would do + an A2A and sharded sampling. + + Args: + param (torch.Tensor): The parameter to shard. Should be of shape [vocab_size, model_dim] + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + + Returns: + torch.Tensor: The sharded parameter of shape [sharded_vocab_size, model_dim] + """ + return shard_param(param, ShardingType.OUTER_DIMENSION, shard_rank, num_shards, granularity=1) + + +def sharded_unembed_dim(vocab_size: int, shard_rank: int, num_shards: int) -> int: + """ + Utility method for determining the sharded vocab size of a sharded unembed parameter. + + Args: + vocab_size (int): The size of the vocabulary. + shard_rank (int): Which shard of the partitioned tensor to return. + num_shards (int): The total number of shards the parameter is distributed across. + """ + start_idx, end_idx = get_shard_endpoints(vocab_size, shard_rank, num_shards, granularity=1) + return end_idx - start_idx diff --git a/deepspeed/inference/v2/model_implementations/sharding/utils.py b/deepspeed/inference/v2/model_implementations/sharding/utils.py new file mode 100644 index 000000000000..fd0eb51873f8 --- /dev/null +++ b/deepspeed/inference/v2/model_implementations/sharding/utils.py @@ -0,0 +1,104 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional, Tuple + +import torch + +from .types import ShardingType, DEFAULT_SHARD_GRANULARITY + + +def get_shard_endpoints(dim_size: int, + shard_rank: int, + num_shards: int, + granularity: int = DEFAULT_SHARD_GRANULARITY) -> Tuple[int, int]: + """ + Given a dimension to shard with size dim_size, return the start and end indices of the slice + that belong to the given rank. + + The typical use of this is as an internal helper function, so see if there is a higher level + API that better suits the application. + + Args: + dim_size (int): The size of the dimension to shard. + shard_rank (int): The rank of the shard to return. + num_shards (int): Total number of shards the dimension will be distributed across. + granularity (int): The minimum alignment of the shard endpoints. This is used to support + non-even head counts as well as align dimensions to cleaner GEMM boundaries. + """ + assert dim_size % granularity == 0, "Dimension size must be divisible by granularity" + + total_chunks = dim_size // granularity + base_chunks_per_rank = total_chunks // num_shards + remainder_chunks = total_chunks % num_shards + + start_chunk_id = shard_rank * base_chunks_per_rank + min(shard_rank, remainder_chunks) + end_chunk_id = start_chunk_id + base_chunks_per_rank + (1 if shard_rank < remainder_chunks else 0) + + return start_chunk_id * granularity, end_chunk_id * granularity + + +def shard_param(param: Optional[torch.Tensor], + shard_mode: ShardingType, + shard_rank: int, + num_shards: int, + num_concatenated_matrices: int = 1, + granularity: int = 32, + bias_dims: int = 1) -> torch.Tensor: + """ + Utility for sharding a parameter. This will return the slice of the parameter that should + exist on the given shard_rank given the sharding configuration. The workflow here is + to find the minimum bounded Tensor to shard, get the slicing endpoints, and then concatenate + as needed. + + The typical use of this is as an internal helper function, so see if there is a higher level + API that better suits the application. + + Args: + param (torch.Tensor): The parameter to shard. + shard_mode (ShardingType): The type of sharding to apply. See ShardingType for more context. + shard_rank (int): The rank of the shard to return. + num_shards (int): Total number of shards the parameter will be distrbuted across. + num_concatenated_matrices (int): The number of matrices that have been concatenated together in the original + parameter. An example of this is a fused QKV projection matrix, where the `num_concatenated_matrices` + argument would be 3. + granularity (int): The minimum alignment of the shard endpoints. For attention projection matrices, this + should be set to the head size to support non-even sharding. + bias_dims (int): The number of dimensions that are considered bias dimensions. This is used to support + sharding of MoE and non-MoE biases on the same codepath. + """ + assert shard_rank < num_shards, "Shard rank must be less than num_shards" + + # Easier to hide this inside of the sharding logic than to add checks in every model + # implementation. + if param is None: + return None + + if num_shards == 1: + # Trivial case of no sharding. + return param + + if shard_mode == ShardingType.OUTER_DIMENSION: + + def get_matrices(dim_idx: int) -> torch.Tensor: + dim_size = param.size(dim_idx) // num_concatenated_matrices + start_channel_id, end_channel_id = get_shard_endpoints(dim_size, shard_rank, num_shards, granularity) + return torch.chunk(param, num_concatenated_matrices, dim=dim_idx), start_channel_id, end_channel_id + + if param.ndim == bias_dims: + # Special case for bias parameters. + matrices, start_channel_id, end_channel_id = get_matrices(dim_idx=-1) + return torch.cat([mat[..., start_channel_id:end_channel_id] for mat in matrices], dim=-1) + else: + # General case for weight parameters. This assumes MoE parameters are stored in the format of + # [num_experts, out_features, in_features] + matrices, start_channel_id, end_channel_id = get_matrices(dim_idx=-2) + return torch.cat([mat[..., start_channel_id:end_channel_id, :] for mat in matrices], dim=-2) + + elif shard_mode == ShardingType.INNER_DIMENSION: + dim_size = param.size(-1) // num_concatenated_matrices + start_channel_id, end_channel_id = get_shard_endpoints(dim_size, shard_rank, num_shards, granularity) + matrices = torch.chunk(param, num_concatenated_matrices, dim=-1) + return torch.cat([mat[..., start_channel_id:end_channel_id] for mat in matrices], dim=-1) diff --git a/deepspeed/inference/v2/modules/__init__.py b/deepspeed/inference/v2/modules/__init__.py new file mode 100644 index 000000000000..917c1599de2e --- /dev/null +++ b/deepspeed/inference/v2/modules/__init__.py @@ -0,0 +1,8 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from . import implementations +from . import interfaces +from .module_registry import ConfigBundle diff --git a/deepspeed/inference/v2/modules/configs/__init__.py b/deepspeed/inference/v2/modules/configs/__init__.py new file mode 100644 index 000000000000..19b9fb99ddea --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/__init__.py @@ -0,0 +1,11 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .attention_configs import (DSSelfAttentionConfig, PositionalEmbeddingType, MaskingType) +from .embedding_config import DSEmbeddingsConfig +from .linear_config import DSLinearConfig +from .moe_config import DSMoEConfig +from .norm_config import DSNormConfig, NormTypeEnum +from .unembed_config import DSUnembedConfig diff --git a/deepspeed/inference/v2/modules/configs/attention_configs.py b/deepspeed/inference/v2/modules/configs/attention_configs.py new file mode 100644 index 000000000000..bcdc3d2613d5 --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/attention_configs.py @@ -0,0 +1,82 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from enum import Enum +from typing import Dict + +from ...inference_utils import DtypeEnum +from ...modules.ds_module import DSModuleConfig + + +class PositionalEmbeddingType(Enum): + + # No positional embeddings + none = "none" + + # Rotary positional embeddings - every half + rotate_half = "rotate_half" + + # Rotary positional embeddings - every other + rotate_every_other = "rotate_every_other" + + # Alibi + alibi = "alibi" + + +class MaskingType(Enum): + + # No masking + none = "none" + + # Causal masking + causal = "causal" + + # Local masking + local = "local" + + # Symmetric masking (this is a 1D tensor mask) + symmetric = "symmetric" + + # Arbitrary masking (this would correspond to a 2D tensor mask) + asymmetric = "asymmetric" + + +class DSSelfAttentionConfig(DSModuleConfig): + """ + Config class for attention. + """ + + # Number of query attention heads on this shard + n_heads_q: int + + # Number of KV attention heads on this shard + n_heads_kv: int + + # Size of each attention head + head_size: int + + # Max number of sequences that may compose a ragged batch + max_sequences: int + + # Scale factor for attention scores + scale_factor: float = 1.0 + + # Input data type + input_dtype: DtypeEnum = DtypeEnum.fp16 + + # Output data type + output_dtype: DtypeEnum = DtypeEnum.fp16 + + # Masking type + masking_type: MaskingType = MaskingType.causal + + # Masking args + masking_args: Dict = {} + + # Positional embedding type + positional_embedding_type: PositionalEmbeddingType = PositionalEmbeddingType.none + + # Positional embedding args + positional_embedding_args: Dict = {} diff --git a/deepspeed/inference/v2/modules/configs/embedding_config.py b/deepspeed/inference/v2/modules/configs/embedding_config.py new file mode 100644 index 000000000000..2486c5986e95 --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/embedding_config.py @@ -0,0 +1,70 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +from ...inference_utils import DtypeEnum, NormTypeEnum +from ...modules.ds_module import DSModuleConfig +""" +Trying to define the space we need to support here right now: + +Types of embeddings I've found so far: + 1. Token embedding + 2. Position embedding + 3. Token type embedding + 4. LN + +GPTNeo: 1, 2, 3 (shared with 1) +GPTNeoX: 1 +GPTJ: 1, 3 +LLaMA: 1 +BERT: 1, 2, 3, 4 +GPT2: 1, 2, 3 (shared with 1) + +Sidebar for OPT: +OPT: 1, 2 +1 may not actually project to the actual hidden dimension according to the raw +code, but for the model configs we care about it does. +2 has a weird offset associated with it that the others do not. +""" + + +class DSEmbeddingsConfig(DSModuleConfig): + """ + Config class for DSEmbeddings. + """ + + residual_dtype: DtypeEnum = DtypeEnum.fp16 + """ + Data type the module should use for its output. + """ + + embedding_dim: int + """ + Dimensionality of the embedding projections. + """ + + positional_embedding: bool = False + """ + Whether the module should expect a positional embedding matrix. The shape of this + matrix should be of shape [max_seq_len + positional_offset, embedding_dim] + """ + + positional_offset: int = 0 + """ + Whether the linearized token IDs should be offset by a certain amount. For an example + of this, see the OPT model implementation. + """ + + use_token_type: bool = False + """ + Whether the module should expect a token type embedding matrix. + """ + + output_normalization: Optional[NormTypeEnum] = None + """ + If a the output of the embedding module should be normalized, specify here. See + ``inference.inference_utils.NormTypeEnum`` for supported values. + """ diff --git a/deepspeed/inference/v2/modules/configs/linear_config.py b/deepspeed/inference/v2/modules/configs/linear_config.py new file mode 100644 index 000000000000..40fe0773aeee --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/linear_config.py @@ -0,0 +1,43 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from ...inference_utils import ActivationType, DtypeEnum +from ...modules.ds_module import DSModuleConfig + + +class DSLinearConfig(DSModuleConfig): + """ + Config class for DSLinearBase. + """ + + in_channels: int + """ + Number of input channels + """ + + out_channels: int + """ + Number of output channels. NOTE: If this linear layer is using a gated activation function, + the value for ``out_channels`` passed here should refer to the number of channels after + gating (i.e., the expected weight shape before transformations will be ``[out_channels * 2, in_channels]``). + """ + + activation: ActivationType = ActivationType.IDENTITY + """ + The activation function for this layer. See :class:`deepspeed.inference.inference_utils.ActivationType` for + supported activation functions. + """ + + input_dtype: DtypeEnum = DtypeEnum.fp16 + """ + The data type of the input tensor. See :class:`deepspeed.inference.inference_utils.DtypeEnum` for supported + data types. + """ + + output_dtype: DtypeEnum = DtypeEnum.fp16 + """ + The data type of the output tensor. See :class:`deepspeed.inference.inference_utils.DtypeEnum` for supported + data types. + """ diff --git a/deepspeed/inference/v2/modules/configs/moe_config.py b/deepspeed/inference/v2/modules/configs/moe_config.py new file mode 100644 index 000000000000..1a88d54af19f --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/moe_config.py @@ -0,0 +1,50 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from ...inference_utils import ActivationType, DtypeEnum +from ...modules.ds_module import DSModuleConfig + + +class DSMoEConfig(DSModuleConfig): + """ + Config class for DSMoEBase + """ + + model_dim: int + """ + Size of input activation. + """ + + intermediate_features: int + """ + Size of intermediate activation. Specifically, this is the number of input features + in the second linear layer. Depending on the activation function, the output of the first + linear layer may have increased dimensionality. + """ + + n_experts: int + """ + Number of experts. + """ + + top_k: int = 1 + """ + top-k gating function (like top-1 or top-2) + """ + + input_dtype: DtypeEnum = DtypeEnum.fp16 + """ + Data type for the input activations. + """ + + output_dtype: DtypeEnum = DtypeEnum.fp16 + """ + Data type for the output activations. + """ + + activation: ActivationType = ActivationType.IDENTITY + """ + Activation function of the first MLP1 + """ diff --git a/deepspeed/inference/v2/modules/configs/norm_config.py b/deepspeed/inference/v2/modules/configs/norm_config.py new file mode 100644 index 000000000000..358982253756 --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/norm_config.py @@ -0,0 +1,32 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from ...inference_utils import DtypeEnum, NormTypeEnum +from ...modules.ds_module import DSModuleConfig + + +class DSNormConfig(DSModuleConfig): + """ + Config class for both DSPreLN and DSPostLN. + """ + + # Type of normalization + type: NormTypeEnum + + # Number of channels in the model embedding + channels: int + + # Data type of the residual input/outputs (we assume the residual must + # be the same data type for the entire model). + residual_dtype: DtypeEnum = DtypeEnum.fp16 + + # Data type of the hidden states input + input_dtype: DtypeEnum = DtypeEnum.fp16 + + # Data type of the hidden states output + output_dtype: DtypeEnum = DtypeEnum.fp16 + + # Epsilon value for numerical stability + eps: float = 1e-5 diff --git a/deepspeed/inference/v2/modules/configs/unembed_config.py b/deepspeed/inference/v2/modules/configs/unembed_config.py new file mode 100644 index 000000000000..ea4cc3cc99c1 --- /dev/null +++ b/deepspeed/inference/v2/modules/configs/unembed_config.py @@ -0,0 +1,39 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from ...inference_utils import DtypeEnum, NormTypeEnum +from ...modules.ds_module import DSModuleConfig +from typing import Optional + + +class DSUnembedConfig(DSModuleConfig): + """ + Config class for DSUnembed + """ + + dtype: DtypeEnum = DtypeEnum.fp16 + """ + Expected data type. + """ + + norm_type: Optional[NormTypeEnum] = None + """ + Whether the input to the unembed is normalized prior to the unembedding projection. + """ + + model_dim: int + """ + Model embedding size. + """ + + max_sequences: int + """ + Max sequences composing the ragged batch. + """ + + vocab_size: int + """ + Local vocab size (the full vocab size may have been sharded across model parallel ranks) + """ diff --git a/deepspeed/inference/v2/modules/ds_module.py b/deepspeed/inference/v2/modules/ds_module.py new file mode 100644 index 000000000000..2a6d294f3266 --- /dev/null +++ b/deepspeed/inference/v2/modules/ds_module.py @@ -0,0 +1,62 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import ABC, abstractstaticmethod +from typing import Any, Dict, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel + + +class DSModuleConfig(DeepSpeedConfigModel): + + max_tokens: int + + +class DSModuleBase(torch.nn.Module, ABC): + """ + Base class for all DeepSpeed Inference modules. This class establishes + the basic attributes of a DSModule. Only abstract functionality modules should inherit + directly from this class, not specific implementations. + """ + + @abstractstaticmethod + def name() -> str: + """ + Return a memorable, human-readable name for this module. + + This will be used as a key in custom inference configurations and should only + be implemented by the children of functionality modules. + """ + ... + + @abstractstaticmethod + def config_class() -> Type[DSModuleConfig]: + """ + Return the associated config class for this module. + + This should be implemented (along with the config class) by an abstract functionality + module. + """ + ... + + @abstractstaticmethod + def supports_config(config: DSModuleConfig) -> bool: + """ + Return whether or not this module supports the given config. + + This should be implemented by the children of functionality modules and should report + whether it would be feasible to instantiate this module with the given config. + """ + ... + + def __init__(self, config: DSModuleConfig, implementation_config: Dict[str, Any] = {}) -> None: + """ + Initialize the module with the given config. + """ + super().__init__() + self._config = config + self._implementation_config = implementation_config diff --git a/deepspeed/inference/v2/modules/heuristics.py b/deepspeed/inference/v2/modules/heuristics.py new file mode 100644 index 000000000000..b89e95c0d834 --- /dev/null +++ b/deepspeed/inference/v2/modules/heuristics.py @@ -0,0 +1,179 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from ..config_v2 import RaggedInferenceEngineConfig +from ..inference_utils import NormTypeEnum + +from .module_registry import ConfigBundle +from ..modules.configs import ( + DSEmbeddingsConfig, + DSLinearConfig, + DSMoEConfig, + DSNormConfig, + DSSelfAttentionConfig, + DSUnembedConfig, +) +from ..modules.interfaces import ( + DSEmbeddingBase, + DSEmbeddingRegistry, + DSLinearBase, + DSLinearRegistry, + DSMoEBase, + DSMoERegistry, + DSPostNormBase, + DSPostNormRegistry, + DSPreNormBase, + DSPreNormRegistry, + DSSelfAttentionBase, + DSSelfAttentionRegistry, + DSUnembedBase, + DSUnembedRegistry, +) + + +def instantiate_attention(attention_config: DSSelfAttentionConfig, + engine_config: RaggedInferenceEngineConfig) -> DSSelfAttentionBase: + """ + Choose an appropriate attention implementation based on the given configurations. This + method is currently a stub, but as more implementations may be developed we can centralize + the logic for choosing between them here. + + Arguments: + attention_config (DSSelfAttentionConfig): Configuration for the attention module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + An attention module implementing the given configuration. + """ + + # Currently, we only have one implementation, so we just return it. + config = ConfigBundle(name="dense_blocked_attention", config=attention_config) + return DSSelfAttentionRegistry.instantiate_config(config) + + +def instantiate_embed(embed_config: DSEmbeddingsConfig, engine_config: RaggedInferenceEngineConfig) -> DSEmbeddingBase: + """ + Choose an appropriate embedding implementation based on the given configurations. This + method is currently a stub, but as more implementations may be developed we can centralize + the logic for choosing between them here. + + Arguments: + embed_config (DSEmbeddingsConfig): Configuration for the embedding module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + An embedding module implementing the given configuration. + """ + + # Currently, we only have one implementation, so we just return it. + config = ConfigBundle(name="ragged_embedding", config=embed_config) + return DSEmbeddingRegistry.instantiate_config(config) + + +def instantiate_linear(linear_config: DSLinearConfig, engine_config: RaggedInferenceEngineConfig) -> DSLinearBase: + """ + Choose an appropriate linear implementation based on the given configurations. This + method is currently a stub, but as more implementations may be developed we can centralize + the logic for choosing between them here. + + Arguments: + linear_config (DSLinearConfig): Configuration for the linear module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + A linear module implementing the given configuration. + """ + + # Currently, we only have one implementation, so we just return it. + config = ConfigBundle(name="blas_fp_linear", config=linear_config) + return DSLinearRegistry.instantiate_config(config) + + +def instantiate_moe(moe_config: DSMoEConfig, engine_config: RaggedInferenceEngineConfig) -> DSMoEBase: + """ + Choose an appropriate MoE implementation based on the given configurations. This + method is currently a stub, but as more implementations may be developed we can centralize + the logic for choosing between them here. + + Arguments: + moe_config (DSMoEConfig): Configuration for the MoE module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + A MoE module implementing the given configuration. + """ + + moe_type = "cutlass_multi_gemm_moe" + + if moe_type == "cutlass_multi_gemm_moe": + # TODO: Get this off an engine config + implementation_config = { + "weight_dtype": moe_config.input_dtype, + } + + # Currently, we only have one implementation, so we just return it. + config = ConfigBundle(name="cutlass_multi_gemm_moe", + config=moe_config, + implementation_config=implementation_config) + return DSMoERegistry.instantiate_config(config) + + +def instantiate_post_norm(norm_config: DSNormConfig, engine_config: RaggedInferenceEngineConfig) -> DSPostNormBase: + """ + Choose an appropriate post-norm implementation based on the given configurations. This + method is currently a stub, but as more implementations may be developed we can centralize + the logic for choosing between them here. + + Arguments: + norm_config (DSNormConfig): Configuration for the post-norm module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + A post-norm module implementing the given configuration. + """ + + # Currently, we only have one implementation, so we just return it. + config = ConfigBundle(name="cuda_post_ln", config=norm_config) + return DSPostNormRegistry.instantiate_config(config) + + +def instantiate_pre_norm(norm_config: DSNormConfig, engine_config: RaggedInferenceEngineConfig) -> DSPreNormBase: + """ + Choose an appropriate pre-norm implementation based on the given configurations. Currently, + this will select between two CUDA implementations, one for LayerNorm and one for RMSNorm. + + Arguments: + norm_config (DSNormConfig): Configuration for the pre-norm module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + A pre-norm module implementing the given configuration. + """ + if NormTypeEnum(norm_config.type) == NormTypeEnum.LayerNorm: + module_name = "cuda_pre_ln" + elif NormTypeEnum(norm_config.type) == NormTypeEnum.RMSNorm: + module_name = "cuda_pre_rms" + + config = ConfigBundle(name=module_name, config=norm_config) + return DSPreNormRegistry.instantiate_config(config) + + +def instantiate_unembed(unembed_config: DSUnembedConfig, engine_config: RaggedInferenceEngineConfig) -> DSUnembedBase: + """ + Choose an appropriate unembedding implementation based on the given configurations. This + method is currently a stub, but as more implementations may be developed we can centralize + the logic for choosing between them here. + + Arguments: + unembed_config (DSUnembedConfig): Configuration for the unembed module. + engine_config (RaggedInferenceEngineConfig): Configuration for the inference engine. + + Returns: + An unembed module implementing the given configuration. + """ + + # Currently, we only have one implementation, so we just return it. + config = ConfigBundle(name="ragged_unembed", config=unembed_config) + return DSUnembedRegistry.instantiate_config(config) diff --git a/deepspeed/inference/v2/modules/implementations/__init__.py b/deepspeed/inference/v2/modules/implementations/__init__.py new file mode 100644 index 000000000000..1b500a9a0b5a --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/__init__.py @@ -0,0 +1,13 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# Imports for registering ops +from .attention import * +from .linear import * +from .post_norm import * +from .pre_norm import * +from .embedding import * +from .unembed import * +from .moe import * diff --git a/deepspeed/inference/v2/modules/implementations/attention/__init__.py b/deepspeed/inference/v2/modules/implementations/attention/__init__.py new file mode 100644 index 000000000000..1ff593ac1368 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/attention/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .dense_blocked_attention import DSDenseBlockedAttention diff --git a/deepspeed/inference/v2/modules/implementations/attention/dense_blocked_attention.py b/deepspeed/inference/v2/modules/implementations/attention/dense_blocked_attention.py new file mode 100644 index 000000000000..bb482f0c58d6 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/attention/dense_blocked_attention.py @@ -0,0 +1,167 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional + +import torch + +from deepspeed.accelerator import get_accelerator +from ....allocator import empty_from +from ....inference_utils import DtypeEnum +from ....kernels.ragged_ops import ( + AtomBuilder, + BlockedFlashAttn, + BlockedRotaryEmbeddings, + BlockedTrainedRotaryEmbeddings, + get_q_block_size, + get_kv_block_size, + LinearBlockedKVCopy, +) +from ....ragged import RaggedBatchWrapper, split_kv +from deepspeed.ops.op_builder import RaggedUtilsBuilder + +from ...interfaces import DSSelfAttentionBase, DSSelfAttentionRegistry +from ...configs import DSSelfAttentionConfig, PositionalEmbeddingType, MaskingType + +try: + from functools import cached_property +except ImportError: + + def cached_property(func): + return property(func) + + +@DSSelfAttentionRegistry.register_module +class DSDenseBlockedAttention(DSSelfAttentionBase): + """ + Self attention implementation for dense, blocked self attention. + """ + + @staticmethod + def name() -> str: + return 'dense_blocked_attention' + + @staticmethod + def supports_config(config: DSSelfAttentionConfig) -> bool: + + if config.input_dtype != config.output_dtype: + return False + + if DtypeEnum(config.input_dtype) not in (DtypeEnum.fp16, DtypeEnum.bf16): + return False + + if PositionalEmbeddingType(config.positional_embedding_type) not in [ + PositionalEmbeddingType.none, PositionalEmbeddingType.rotate_half + ]: + return False + + if MaskingType(config.masking_type) != MaskingType.causal: + return False + + return True + + def __init__(self, config: DSSelfAttentionConfig, implementation_config: Dict[str, Any]) -> None: + """ + Create the Attention DSModule. + + Args: + config (DSSelfAttentionConfig): The self attention config for all attention DSModules. + implementation_config (Dict[str, Any]): The implementation config for this DSModule may + contain a `trained_freqs` key. If passed, the implementation will expect a `trained_freqs` + tensor in the `forward` method and will not synthesize the frequencies internally. + """ + super().__init__(config, implementation_config) + + embed_type = PositionalEmbeddingType(config.positional_embedding_type) + if embed_type == PositionalEmbeddingType.none: + self._kv_copy = LinearBlockedKVCopy(self._config.head_size, self._config.n_heads_q, + self._config.n_heads_kv, self._config.input_dtype) + elif embed_type == PositionalEmbeddingType.rotate_half: + use_trained_freqs = "trained_freqs" in self._config.positional_embedding_args and self._config.positional_embedding_args[ + "trained_freqs"] + if use_trained_freqs: + self._kv_copy = BlockedTrainedRotaryEmbeddings(self._config.head_size, self._config.n_heads_q, + self._config.n_heads_kv, self._config.input_dtype) + else: + self._kv_copy = BlockedRotaryEmbeddings(self._config.head_size, self._config.n_heads_q, + self._config.n_heads_kv, self._config.input_dtype) + + self._softmax_scale = self._config.scale_factor + + # TODO(cmikeh2): Attention kernel gets created here. + self._attn_kernel = BlockedFlashAttn(self._config.head_size, self._config.input_dtype) + self._atom_builder = AtomBuilder() + + self.model_dim = self._config.head_size * self._config.n_heads_q + self._output = torch.empty((self._config.max_tokens, self._config.head_size * self._config.n_heads_q), + dtype=self._config.output_dtype, + device=get_accelerator().current_device()) + + # TODO(cmikeh2): Pre-allocate storage buffer for the attention atoms. + self._max_atoms = self._config.max_sequences + self._atoms = torch.empty((self._max_atoms, 8), dtype=torch.int32, device=get_accelerator().current_device()) + + alloc_func = RaggedUtilsBuilder().load().allocate_fast_host_buffer + self._atoms_shadow = alloc_func(self._atoms) + self._cur_atoms = 0 + + @cached_property + def kv_block_size(self) -> int: + """ + Return preferred granulatity for blocked KV-cache implementation. + """ + return get_kv_block_size(self._config.head_size) + + @cached_property + def q_block_size(self) -> int: + """ + Property to calculate blocking granularity for the query dimension. + This has no impact on the KV-cache structure, but will affect the + number of attention atoms associated with a batch. + """ + return get_q_block_size(self._config.head_size) + + def build_atoms(self, ragged_batch: RaggedBatchWrapper) -> None: + """ + Build the atoms for the attention kernel. + + Args: + ragged_batch (RaggedBatchWrapper): The input ids and associated ragged batch metadata. + """ + host_atoms, n_atoms = self._atom_builder(self._atoms_shadow, ragged_batch, self.q_block_size, + self.kv_block_size) + + self._cur_atoms = n_atoms + self._atoms[:n_atoms].copy_(host_atoms[:n_atoms], non_blocking=True) + + def forward(self, + q_k_v: torch.Tensor, + kv_cache: torch.Tensor, + batch: RaggedBatchWrapper, + inv_freqs: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Forward implementation. + + Args: + q_k_v (torch.Tensor): Query/Key/Value projection Tensor of shape + [n_heads, (n_heads_q + 2 * n_heads_kv) * head_size]. + kv_cache (torch.Tensor): Blocked persistent cache of shape + [2, batch, block_size, n_heads_kv, head_size]. + batch (RaggedBatchWrapper): The input ids and associated ragged batch metadata. + inv_freqs (Optional[torch.Tensor]): The inverse frequencies for the rotary embeddings if they + have been modified from synthesizable values. + """ + if inv_freqs is not None: + self._kv_copy(kv_cache, q_k_v, batch, inv_freqs) + else: + self._kv_copy(kv_cache, q_k_v, batch) + + q = q_k_v[:, :self._config.head_size * self._config.n_heads_q] + output = empty_from(self._output, q.shape) + k_cache, v_cache = split_kv(kv_cache) + + self._attn_kernel(output, q, k_cache, v_cache, self._atoms[:self._cur_atoms], self._softmax_scale) + + return output diff --git a/deepspeed/inference/v2/modules/implementations/embedding/__init__.py b/deepspeed/inference/v2/modules/implementations/embedding/__init__.py new file mode 100644 index 000000000000..5458a855abf4 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/embedding/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .ragged_embedding import DSRaggedEmbedding diff --git a/deepspeed/inference/v2/modules/implementations/embedding/ragged_embedding.py b/deepspeed/inference/v2/modules/implementations/embedding/ragged_embedding.py new file mode 100644 index 000000000000..6782bcae81c8 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/embedding/ragged_embedding.py @@ -0,0 +1,77 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional + +import torch + +from deepspeed.accelerator import get_accelerator +from ....allocator import empty_from +from ....inference_utils import DtypeEnum +from ....kernels.ragged_ops import RaggedEmbeddingKernel +from ....ragged import RaggedBatchWrapper +from ...interfaces import DSEmbeddingBase, DSEmbeddingRegistry +from ...configs import DSEmbeddingsConfig + + +@DSEmbeddingRegistry.register_module +class DSRaggedEmbedding(DSEmbeddingBase): + + @staticmethod + def name(): + return 'ragged_embedding' + + @staticmethod + def supports_config(config: DSEmbeddingsConfig) -> bool: + + if DtypeEnum(config.residual_dtype) not in [DtypeEnum.fp16, DtypeEnum.bf16, DtypeEnum.fp32]: + return False + + if config.use_token_type: + return False + + if config.output_normalization != None: + return False + + try: + _ = RaggedEmbeddingKernel(config.residual_dtype, torch.int32, config.embedding_dim) + except ValueError: + return False + + return True + + def __init__(self, config: DSEmbeddingsConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + self.embed_offset = self._config.positional_offset + + # TODO(cmikeh2): How do we want to avoid the int32 vs int64 issue? + self._ragged_embed = RaggedEmbeddingKernel(self._config.residual_dtype, torch.int32, + self._config.embedding_dim) + + self._output = torch.empty((self._config.max_tokens, self._config.embedding_dim), + dtype=self._config.residual_dtype, + device=get_accelerator().current_device()) + + @property + def output(self) -> torch.Tensor: + return self._output + + def forward(self, + ragged_batch: RaggedBatchWrapper, + word_embeddings: torch.Tensor, + position_embeddings: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Parameters: + ragged_batch (RaggedBatchWrapper): The input ids and associated ragged batch metadata. + word_embeddings (torch.Tensor): The word embedding table + """ + output = empty_from(self._output, (ragged_batch.tensor_toks, self._config.embedding_dim)) + self._ragged_embed(output, + ragged_batch, + word_embeddings, + position_embed_weight=position_embeddings, + position_embed_offset=self.embed_offset) + return output diff --git a/deepspeed/inference/v2/modules/implementations/linear/__init__.py b/deepspeed/inference/v2/modules/implementations/linear/__init__.py new file mode 100644 index 000000000000..5acdc69dba7c --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/linear/__init__.py @@ -0,0 +1,7 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .blas_fp_linear import BlasFPLinear +from .cutlass_fp_linear import DSCutlassFPLinear diff --git a/deepspeed/inference/v2/modules/implementations/linear/blas_fp_linear.py b/deepspeed/inference/v2/modules/implementations/linear/blas_fp_linear.py new file mode 100644 index 000000000000..cbb8801e7193 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/linear/blas_fp_linear.py @@ -0,0 +1,101 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional + +import torch + +from deepspeed.accelerator import get_accelerator +from ....allocator import empty_from +from ....inference_utils import is_gated +from ....kernels.core_ops import ( + BlasLibLinear, + CUDABiasActivation, + CUDAGatedActivation, +) + +from ...interfaces import DSLinearBase, DSLinearRegistry +from ...configs import DSLinearConfig + + +@DSLinearRegistry.register_module +class BlasFPLinear(DSLinearBase): + """ + Linear DSModule based on BLAS library and standalone bias + activation kernel implementation. + """ + + @staticmethod + def name(): + return 'blas_fp_linear' + + @staticmethod + def supports_config(config: DSLinearConfig) -> bool: + if config.input_dtype != config.output_dtype: + return False + + if config.input_dtype != torch.float16 and config.input_dtype != torch.bfloat16: + return False + + if is_gated(config.activation): + try: + _ = CUDAGatedActivation(config.out_channels, config.output_dtype, config.activation) + except ValueError: + return False + else: + try: + _ = CUDABiasActivation(config.out_channels, config.output_dtype, config.activation) + except ValueError: + return False + + return True + + def __init__(self, config: DSLinearConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + self._linear_impl = BlasLibLinear(self._config.input_dtype) + + if is_gated(config.activation): + self._is_gated = True + self._act_fn = CUDAGatedActivation(config.out_channels, config.output_dtype, config.activation) + self._double_buffer = torch.empty((config.max_tokens, config.out_channels * 2), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + else: + self._is_gated = False + self._act_fn = CUDABiasActivation(config.out_channels, config.output_dtype, config.activation) + + self._output = torch.empty((config.max_tokens, config.out_channels), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Converts param to same data type as input and output. + + Parameters: + param (torch.Tensor): Weight or bias tensor. + """ + return param.to(self._config.input_dtype) + + def forward(self, hidden_states: torch.Tensor, w: torch.Tensor, b: Optional[torch.Tensor] = None) -> torch.Tensor: + + output = empty_from(self._output, (hidden_states.shape[0], self._config.out_channels)) + + if self._is_gated: + staging_output = empty_from(self._double_buffer, (hidden_states.shape[0], self._config.out_channels * 2)) + self._linear_impl(staging_output, hidden_states, w) + self._act_fn(output, staging_output, b) + else: + self._linear_impl(output, hidden_states, w) + self._act_fn(output, b) + + return output + + @property + def output(self) -> torch.Tensor: + """ + Return the padded, pre-allocated output Tensor. + """ + return self._output diff --git a/deepspeed/inference/v2/modules/implementations/linear/cutlass_fp_linear.py b/deepspeed/inference/v2/modules/implementations/linear/cutlass_fp_linear.py new file mode 100644 index 000000000000..a3704eaa82a2 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/linear/cutlass_fp_linear.py @@ -0,0 +1,81 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional + +import torch + +from deepspeed.accelerator import get_accelerator +from ....allocator import empty_from +from ....inference_utils import ActivationType +from ....kernels.core_ops import CUDAGatedActivation + +from ...interfaces import DSLinearBase, DSLinearRegistry +from ...configs import DSLinearConfig + + +@DSLinearRegistry.register_module +class DSCutlassFPLinear(DSLinearBase): + """ + Linear DSModule based on CUTLASS floating point kernel implementation. + """ + + @staticmethod + def name(): + return 'cutlass_fp_linear' + + @staticmethod + def supports_config(config: DSLinearConfig) -> bool: + if config.input_dtype != config.output_dtype: + return False + + if config.input_dtype != torch.float16 and config.input_dtype != torch.bfloat16: + return False + + return True + + def __init__(self, config: DSLinearConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + # TODO: Load kernel + + if config.activation == ActivationType.GEGLU: + self._geglu = CUDAGatedActivation(config.out_channels, config.output_dtype, ActivationType.GEGLU) + self._activation_int = torch.empty((config.max_tokens, config.out_channels * 2), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + + self._output = torch.empty((config.max_tokens, config.out_channels), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Converts param to same data type as input and output. + + Parameters: + param (torch.Tensor): Weight or bias tensor. + """ + return param.to(self._config.input_dtype) + + def forward(self, hidden_states: torch.Tensor, w: torch.Tensor, b: Optional[torch.Tensor] = None) -> torch.Tensor: + + output = empty_from(self._output, (hidden_states.shape[0], self._config.out_channels)) + + if self._config.activation == ActivationType.GEGLU: + intermediate = empty_from(self._activation_int, (hidden_states.shape[0], self._config.out_channels * 2)) + self._linear_impl(intermediate, hidden_states, w, b) + self._geglu(output, intermediate) + else: + self._linear_impl(output, hidden_states, w, b) + + return output + + @property + def output(self) -> torch.Tensor: + """ + Return the padded, pre-allocated output Tensor. + """ + return self._output diff --git a/deepspeed/inference/v2/modules/implementations/moe/__init__.py b/deepspeed/inference/v2/modules/implementations/moe/__init__.py new file mode 100644 index 000000000000..053ad5da7746 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/moe/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .cutlass_multi_gemm import DSMultiGemmMoE diff --git a/deepspeed/inference/v2/modules/implementations/moe/cutlass_multi_gemm.py b/deepspeed/inference/v2/modules/implementations/moe/cutlass_multi_gemm.py new file mode 100644 index 000000000000..fb2388c450f0 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/moe/cutlass_multi_gemm.py @@ -0,0 +1,225 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional, Tuple + +import torch + +from deepspeed.accelerator import get_accelerator +from ....allocator import empty_from +from ....inference_utils import ActivationType +from ....kernels.core_ops import BlasLibLinear +from ....kernels.ragged_ops import ( + MoEGather, + MoEScatter, + RaggedTop1Gating, +) +from ....ragged import RaggedBatchWrapper + +from ...interfaces import DSMoEBase, DSMoERegistry +from ...configs import DSMoEConfig +from ....kernels.cutlass_ops import MoEGEMM + + +@DSMoERegistry.register_module +class DSMultiGemmMoE(DSMoEBase): + """ + MoE implementation based on the CUTLASS multi-GEMM. + """ + + @staticmethod + def name(): + return 'cutlass_multi_gemm_moe' + + @staticmethod + def supports_config(config: DSMoEConfig) -> bool: + if config.input_dtype != config.output_dtype: + return False + + if config.input_dtype != torch.float16 and config.input_dtype != torch.bfloat16: + return False + + if config.top_k != 1: + return False + + if config.activation in [ActivationType.GEGLU, ActivationType.ReGLU, ActivationType.SiGLU]: + # Currently not supporting gated activations in MoE + return False + + return True + + def __init__(self, config: DSMoEConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + # Convenience variables for frequently accessed items. + self.max_tokens = self._config.max_tokens + self.n_experts = self._config.n_experts + self.intermediate_dim = self._config.intermediate_features + + self._mlp_1 = MoEGEMM(fp_dtype=implementation_config['weight_dtype'], act_fn=config.activation) + self._mlp_2 = MoEGEMM(fp_dtype=implementation_config['weight_dtype'], act_fn=ActivationType.IDENTITY) + + self._gate_proj = BlasLibLinear(self._config.input_dtype) + self._top_1_gate = RaggedTop1Gating(config.input_dtype) + self._moe_scatter = MoEScatter(config.input_dtype, config.model_dim) + self._moe_gather = MoEGather(config.input_dtype, config.model_dim) + + self._create_buffers() + + def _create_buffers(self): + + # Gating buffers + self._logits = torch.empty((self._config.max_tokens, self.n_experts), + dtype=self._config.input_dtype, + device=get_accelerator().current_device()) + self._expert_counts = torch.empty((self.n_experts, ), + dtype=torch.int32, + device=get_accelerator().current_device()) + self._scores = torch.empty((self._config.max_tokens, ), + dtype=torch.float32, + device=get_accelerator().current_device()) + self._assignments = torch.empty((self._config.max_tokens, ), + dtype=torch.int32, + device=get_accelerator().current_device()) + self._offsets = torch.empty((self._config.max_tokens, ), + dtype=torch.int32, + device=get_accelerator().current_device()) + + # Scatter buffers + self._moe_input = torch.empty((self._config.max_tokens, self._config.model_dim), + dtype=self._config.input_dtype, + device=get_accelerator().current_device()) + self._expert_cumsum = torch.empty((self._config.n_experts, ), + dtype=torch.int64, + device=get_accelerator().current_device()) + self._mapped_slots = torch.empty((self._config.max_tokens, ), + dtype=torch.int32, + device=get_accelerator().current_device()) + + # GEMM Buffers + self._intermediate = torch.empty((self._config.max_tokens, self._config.intermediate_features), + dtype=self._config.output_dtype, + device=get_accelerator().current_device()) + self._output_unordered = torch.empty((self._config.max_tokens, self._config.model_dim), + dtype=self._config.output_dtype, + device=get_accelerator().current_device()) + + # Gather buffer + self._output = torch.empty((self._config.max_tokens, self._config.model_dim), + dtype=self._config.output_dtype, + device=get_accelerator().current_device()) + + def transform_gate_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Ensures gate param is going to match the activation data type. + """ + return param.to(self._config.input_dtype) + + def transform_moe_mlp_1_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Converts param to same data type as input and output. + + Parameters: + param (torch.Tensor): Weight or bias tensor. + """ + param = param.to(self._config.input_dtype) + + if len(param.shape) == 3: + return param.permute(0, 2, 1).contiguous() + else: + return param + + def transform_moe_mlp_2_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Converts param to same data type as input and output. + + Parameters: + param (torch.Tensor): Weight or bias tensor. + """ + param = param.to(self._config.input_dtype) + + if len(param.shape) == 3: + return param.permute(0, 2, 1).contiguous() + else: + return param + + @property + def output(self) -> torch.Tensor: + return self._output + + def _gate(self, hidden_states: torch.Tensor, batch_metadata: RaggedBatchWrapper, + gate_w: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]: + """ + Helper function to isolate the logit for gating. This will take the hidden states + and produce the metadata + tensors for the CUTLASS ragged GEMMs. If the input has + been padded for CG, this will strip the padding for MoE. + + Parameters: + hidden_states (torch.Tensor): Hidden states tensor. Expected shape is [n_tokens, model_dim]. + batch_metadata (RaggedBatchWrapper): Batch metadata for the hidden states. + gate_w (torch.Tensor): Gate weight tensor. Expected shape is [num_experts, model_dim]. + + Returns: + Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]: The MoE input, the cumsum of the offsets (for the MoE kernels themselves), the scores, and the mapped slots (to recover the original order of the tokens) + """ + + # Get views on the buffers for gating + logits = empty_from(self._logits, (hidden_states.shape[0], self._logits.shape[-1])) + scores = empty_from(self._scores, (hidden_states.shape[0], )) + assignments = empty_from(self._assignments, (hidden_states.shape[0], )) + offsets = empty_from(self._offsets, (hidden_states.shape[0], )) + mapped_slots = empty_from(self._mapped_slots, (hidden_states.shape[0], )) + moe_input = empty_from(self._moe_input, (hidden_states.shape[0], self._moe_input.shape[-1])) + + self._gate_proj(logits, hidden_states, gate_w) + self._expert_counts.zero_() + self._top_1_gate(self._expert_counts, scores, assignments, offsets, logits, batch_metadata) + self._moe_scatter(moe_input, self._expert_cumsum, mapped_slots, hidden_states, self._expert_counts, + assignments, offsets) + + return moe_input, self._expert_cumsum, scores, mapped_slots + + def forward(self, + hidden_states: torch.Tensor, + batch_metadata: RaggedBatchWrapper, + gate_w: torch.Tensor, + mlp_1_w: torch.Tensor, + mlp_2_w: torch.Tensor, + mlp_1_b: Optional[torch.Tensor] = None, + mlp_2_b: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + MoE forward pass built on top of CUTLASS multi-GEMM. + + Parameters: + hidden_states (torch.Tensor): Hidden states tensor. Expected shape is [batch, seq_len, model_dim]. + gate_w (torch.Tensor): Gate weight tensor. Expected shape is [num_experts, model_dim]. + """ + + moe_input, expert_cumsum, scores, mapped_slots = self._gate(hidden_states, batch_metadata, gate_w) + + # Get views on the buffers for GEMM + intermediate = empty_from(self._intermediate, (hidden_states.shape[0], self._intermediate.shape[-1])) + output_unordered = empty_from(self._output_unordered, + (hidden_states.shape[0], self._output_unordered.shape[-1])) + output = empty_from(self._output, (hidden_states.shape[0], self._output.shape[-1])) + + self._mlp_1( + intermediate, + moe_input, + mlp_1_w, + expert_cumsum, + mlp_1_b, + ) + + self._mlp_2( + output_unordered, + intermediate, + mlp_2_w, + expert_cumsum, + mlp_2_b, + ) + + self._moe_gather(output, output_unordered, scores, mapped_slots, self._expert_counts) + return output diff --git a/deepspeed/inference/v2/modules/implementations/moe/gate_fn.py b/deepspeed/inference/v2/modules/implementations/moe/gate_fn.py new file mode 100644 index 000000000000..9eceaab156e4 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/moe/gate_fn.py @@ -0,0 +1,62 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch +import torch.nn.functional as F +from torch import Tensor + +from typing import Tuple + +#TODO(cmikeh2): DELETE + + +@torch.jit.script +def _capacity(gates: Tensor, capacity_factor: Tensor, min_capacity: Tensor) -> Tensor: + # gates has shape of SE + num_tokens = gates.shape[0] + num_experts = gates.shape[-1] + # to(torch.int64) works around a bug in torch.onnx.export: + # it should cast k to int64 when converting torch.topk but it doesn't. + capacity = torch.ceil((num_tokens / num_experts) * capacity_factor).to(torch.int64) + if capacity < min_capacity: + capacity = min_capacity.to(torch.int64) + return capacity + + +@torch.jit.script +def _top_idx(source, k): + return torch.topk(source, k=k, dim=0)[1] + + +def top1gating(logits: Tensor, + capacity_factor: float, + min_capacity: int, + drop_tokens: bool = False) -> Tuple[Tensor, Tensor, Tensor, Tensor]: + # everything is in fp32 in this function + gates = F.softmax(logits, dim=1) + + capacity = _capacity(gates, torch.tensor(capacity_factor), torch.tensor(min_capacity)) + + # Create a mask for 1st's expert per token + indices1_s = torch.argmax(gates, dim=1) + num_experts = int(gates.shape[1]) + mask1 = F.one_hot(indices1_s, num_classes=num_experts) + + # gating decisions + exp_counts = torch.sum(mask1, dim=0).detach().to('cpu') + + assert logits.shape[ + 0] >= min_capacity, "No. of tokens (batch-size) should be greater than min_capacity. Either set min_capacity to 0 or increase your batch size." + + top_idx = _top_idx(mask1, capacity) + + mask1 = mask1 * torch.zeros_like(mask1).scatter_(0, top_idx, 1) + + indices_mask = mask1.sum(dim=1) * num_experts - 1 + indices1_s = torch.min(indices1_s, indices_mask) + + gates1_s = (gates * mask1).sum(dim=1) + + return indices1_s, gates1_s diff --git a/deepspeed/inference/v2/modules/implementations/moe/test.py b/deepspeed/inference/v2/modules/implementations/moe/test.py new file mode 100644 index 000000000000..b714366d32ec --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/moe/test.py @@ -0,0 +1,38 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.modules.interfaces import DSMoERegistry +from deepspeed.inference.modules.configs import DSMoEConfig +from deepspeed.inference.modules.module_registry import ConfigBundle + + +def run_multi_gemm_with_gating(inputs, gate_weight, moe_weight1, moe_bias1, moe_weight2): + config = DSMoEConfig(model_dim=4096, intermediate_features=4096, n_experts=64, max_tokens=128) + moe = DSMoERegistry.instantiate_config( + ConfigBundle(name='cutlass_multi_gemm_moe', + config=config, + implementation_config={ + "weight_dtype": torch.bfloat16, + "transpose_weight": True, + "min_capacity": 8, + "capacity_factor": 1.0 + })) + out = moe(inputs, gate_weight, moe_weight1, moe_weight2, moe_bias1) + return out + + +a = torch.randn( + 128, + 4096, +).bfloat16().to(get_accelerator().current_device()) +weight1 = torch.randn(64, 4096, 4096).bfloat16().to(get_accelerator().current_device()) +bias1 = torch.randn(64, 4096).bfloat16().to(get_accelerator().current_device()) +weight2 = torch.randn(64, 4096, 4096).bfloat16().to(get_accelerator().current_device()) +gate_weight = torch.randn(64, 4096).bfloat16().to(get_accelerator().current_device()) + +out = run_multi_gemm_with_gating(a, gate_weight, weight1, bias1, weight2) +print(out) diff --git a/deepspeed/inference/v2/modules/implementations/post_norm/__init__.py b/deepspeed/inference/v2/modules/implementations/post_norm/__init__.py new file mode 100644 index 000000000000..653a2fe4fb5b --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/post_norm/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .cuda_post_ln import DSPostLNCUDAModule diff --git a/deepspeed/inference/v2/modules/implementations/post_norm/cuda_post_ln.py b/deepspeed/inference/v2/modules/implementations/post_norm/cuda_post_ln.py new file mode 100644 index 000000000000..b30c5b937ed2 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/post_norm/cuda_post_ln.py @@ -0,0 +1,54 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Tuple + +import torch + +from deepspeed.accelerator import get_accelerator +from ...interfaces import DSPostNormBase, DSPostNormRegistry +from ...configs import DSNormConfig +from ....kernels.core_ops.cuda_layer_norm.cuda_post_ln import CUDAFPPostLN +from ....allocator import empty_from + + +@DSPostNormRegistry.register_module +class DSPostLNCUDAModule(DSPostNormBase): + + @staticmethod + def name(): + return 'cuda_post_ln' + + @staticmethod + def supports_config(config: DSNormConfig): + if len(set([config.residual_dtype, config.input_dtype, config.output_dtype])) != 1: + return False + + try: + _ = CUDAFPPostLN(config.channels, config.residual_dtype) + except ValueError: + return False + return True + + def __init__(self, config: DSNormConfig, implementation_config: Dict[str, Any]): + super().__init__(config, implementation_config) + self._fp_post_ln = CUDAFPPostLN(self._config.channels, self._config.residual_dtype, epsilon=self._config.eps) + + self._output = torch.empty((config.max_tokens, config.channels), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + return param.to(self._config.input_dtype) + + def forward(self, residual: torch.Tensor, hidden_in: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Since the CUDA FP only supports all data types being the same, we will alias the residual + with our output. + """ + self._residual_output = empty_from(self._output, residual.shape) + self._fp_post_ln(residual, residual, hidden_in, gamma, beta) + return residual, residual diff --git a/deepspeed/inference/v2/modules/implementations/pre_norm/__init__.py b/deepspeed/inference/v2/modules/implementations/pre_norm/__init__.py new file mode 100644 index 000000000000..12605f13f955 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/pre_norm/__init__.py @@ -0,0 +1,7 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .cuda_pre_ln import DSPreLNCUDAModule +from .cuda_pre_rms import DSPreRMSCUDAModule diff --git a/deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_ln.py b/deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_ln.py new file mode 100644 index 000000000000..f50f4a3d2db6 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_ln.py @@ -0,0 +1,67 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional, Tuple + +import torch + +from deepspeed.accelerator import get_accelerator +from ...interfaces import DSPreNormBase, DSPreNormRegistry +from ...configs import DSNormConfig, NormTypeEnum +from ....kernels.core_ops.cuda_layer_norm.cuda_pre_ln import CUDAFPPreLN +from ....kernels.core_ops.cuda_layer_norm.cuda_ln import CUDAFPLN +from ....allocator import empty_from + + +@DSPreNormRegistry.register_module +class DSPreLNCUDAModule(DSPreNormBase): + + @staticmethod + def name(): + return 'cuda_pre_ln' + + @staticmethod + def supports_config(config: DSNormConfig): + type = NormTypeEnum(config.type) + if type != NormTypeEnum.LayerNorm: + return False + + if len(set([config.residual_dtype, config.input_dtype, config.output_dtype])) != 1: + return False + + try: + _ = CUDAFPPreLN(config.channels, config.residual_dtype) + except ValueError: + return False + return True + + def __init__(self, config: DSNormConfig, implementation_config: Dict[str, Any]): + super().__init__(config, implementation_config) + self._fp_pre_ln = CUDAFPPreLN(self._config.channels, self._config.residual_dtype, epsilon=self._config.eps) + self._fp_ln = CUDAFPLN(self._config.channels, self._config.residual_dtype, epsilon=self._config.eps) + + # Buffers for the hidden output (residual is updated in-place) + self._hidden_output = torch.empty((config.max_tokens, config.channels), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + return param.to(self._config.input_dtype) + + def forward(self, residual: torch.Tensor, hidden_in: Optional[torch.Tensor], gamma: torch.Tensor, + beta: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Since the CUDA FP only supports all data types being the same, we will alias the residual + with our output. + + If hidden_in is None, that means we do not need to perform the residual add and will + only return the hidden output modified. + """ + hidden_out = empty_from(self._hidden_output, residual.shape) + if hidden_in is None: + self._fp_ln(hidden_out, residual, gamma, beta) + else: + self._fp_pre_ln(residual, hidden_out, residual, hidden_in, gamma, beta) + return residual, hidden_out diff --git a/deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_rms.py b/deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_rms.py new file mode 100644 index 000000000000..7aeea4b2d386 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/pre_norm/cuda_pre_rms.py @@ -0,0 +1,77 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional, Tuple + +import torch + +from deepspeed.accelerator import get_accelerator +from ...interfaces import DSPreNormBase, DSPreNormRegistry +from ...configs import DSNormConfig, NormTypeEnum +from ....kernels.core_ops import CUDARMSNorm, CUDARMSPreNorm +from ....allocator import empty_from + + +@DSPreNormRegistry.register_module +class DSPreRMSCUDAModule(DSPreNormBase): + + @staticmethod + def name(): + return 'cuda_pre_rms' + + @staticmethod + def supports_config(config: DSNormConfig): + type = NormTypeEnum(config.type) + if type != NormTypeEnum.RMSNorm: + return False + + if len(set([config.residual_dtype, config.input_dtype, config.output_dtype])) != 1: + return False + + try: + # Only need to check one since the support matrix for the two rms kernels is the same + _ = CUDARMSPreNorm(config.channels, config.residual_dtype) + except ValueError: + return False + return True + + def __init__(self, config: DSNormConfig, implementation_config: Dict[str, Any]): + super().__init__(config, implementation_config) + self._fp_rms = CUDARMSNorm(self._config.channels, self._config.residual_dtype, epsilon=self._config.eps) + self._fp_rms_pre = CUDARMSPreNorm(self._config.channels, self._config.residual_dtype, epsilon=self._config.eps) + + # Buffers for both the hidden and residual outputs + self._hidden_output = torch.empty((config.max_tokens, config.channels), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + self._residual_output = torch.empty((config.max_tokens, config.channels), + dtype=config.output_dtype, + device=get_accelerator().current_device()) + + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + return param.to(self._config.input_dtype) + + def forward(self, + residual: torch.Tensor, + hidden_in: Optional[torch.Tensor], + gamma: torch.Tensor, + beta: Optional[torch.Tensor] = None) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Since the CUDA FP only supports all data types being the same, we will alias the residual + with our output. + + If hidden_in is None, that means we do not need to perform the residual add and will + only return the hidden output modified. + """ + assert beta is None, "Beta is not supported for RMSNorm" + + hidden_out = empty_from(self._hidden_output, residual.shape) + if hidden_in is None: + self._fp_rms(hidden_out, residual, gamma) + residual_out = residual + else: + residual_out = empty_from(self._residual_output, residual.shape) + self._fp_rms_pre(residual_out, hidden_out, residual, hidden_in, gamma) + return residual_out, hidden_out diff --git a/deepspeed/inference/v2/modules/implementations/unembed/__init__.py b/deepspeed/inference/v2/modules/implementations/unembed/__init__.py new file mode 100644 index 000000000000..4a5fd24d518b --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/unembed/__init__.py @@ -0,0 +1,6 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .ragged_unembed import DSRaggedUnembed diff --git a/deepspeed/inference/v2/modules/implementations/unembed/ragged_unembed.py b/deepspeed/inference/v2/modules/implementations/unembed/ragged_unembed.py new file mode 100644 index 000000000000..40d70cbd4df7 --- /dev/null +++ b/deepspeed/inference/v2/modules/implementations/unembed/ragged_unembed.py @@ -0,0 +1,115 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional + +import torch + +from deepspeed.accelerator import get_accelerator +from ....allocator import empty_from +from ....inference_utils import DtypeEnum +from ....kernels.core_ops import CUDAFPLN, BlasLibLinear, CUDARMSNorm +from ....kernels.ragged_ops import RaggedLogitsGather +from ....ragged import RaggedBatchWrapper +from ...interfaces import DSUnembedBase, DSUnembedRegistry +from ...configs import DSUnembedConfig + + +@DSUnembedRegistry.register_module +class DSRaggedUnembed(DSUnembedBase): + """ + Ragged unembedding implementation. This implementation will gather only the last token + of each sequence in the ragged inflight batch and calculate the logits only for those rows. + """ + + @staticmethod + def name(): + return 'ragged_unembed' + + @staticmethod + def supports_config(config: DSUnembedConfig): + + if DtypeEnum(config.dtype) not in [DtypeEnum.fp16, DtypeEnum.bf16, DtypeEnum.fp32]: + return False + + try: + _ = RaggedLogitsGather(config.model_dim, config.dtype) + except ValueError: + return False + + if config.norm_type == 'rms_norm': + try: + _ = CUDARMSNorm(config.model_dim, config.dtype) + except ValueError: + return False + elif config.norm_type == 'layer_norm': + try: + _ = CUDAFPLN(config.model_dim, config.dtype) + except ValueError: + return False + + return True + + def __init__(self, config: DSUnembedConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + self._logits_gather = RaggedLogitsGather(config.model_dim, self._config.dtype) + + if self._config.norm_type == 'layer_norm': + self._norm = CUDAFPLN(self._config.model_dim, self._config.dtype) + elif self._config.norm_type == 'rms_norm': + self._norm = CUDARMSNorm(self._config.model_dim, self._config.dtype) + else: + self._norm = None + + self._linear = BlasLibLinear(self._config.dtype) + + self._intermediate = torch.empty((self._config.max_sequences, self._config.model_dim), + dtype=self._config.dtype, + device=get_accelerator().current_device()) + + self._output = torch.empty((self._config.max_sequences, self._config.vocab_size), + dtype=self._config.dtype, + device=get_accelerator().current_device()) + + @property + def output(self) -> torch.Tensor: + return self._output + + def forward(self, + hidden_states: torch.Tensor, + vocab_embedding: torch.Tensor, + ragged_metadata: RaggedBatchWrapper, + gamma: Optional[torch.Tensor] = None, + beta: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Return final model logits. + + Args: + hidden_states (torch.Tensor): The hidden states from the model. This is the output of the + final layer of the model. + vocab_embedding (torch.Tensor): The vocab embedding table. + raged_metadata (RaggedBatchWrapper): The ragged batch metadata. + gamma (Optional[torch.Tensor]): The gamma tensor for normalization. + beta (Optional[torch.Tensor]): The beta tensor for normalization. + """ + + cut_down_hidden_states = empty_from(self._intermediate, + (ragged_metadata.current_sequences, self._config.model_dim)) + self._logits_gather(cut_down_hidden_states, hidden_states, ragged_metadata) + + if self._config.norm_type == 'rms_norm': + if gamma is None: + raise ValueError('RMS Normalization enabled but gamma not provided.') + self._norm(cut_down_hidden_states, cut_down_hidden_states, gamma) + elif self._config.norm_type == 'layer_norm': + if gamma is None or beta is None: + raise ValueError('Normalization enabled but gamma and/or beta not provided.') + self._norm(cut_down_hidden_states, cut_down_hidden_states, gamma, beta) + + output = empty_from(self._output, (ragged_metadata.current_sequences, self._config.vocab_size)) + self._linear(output, cut_down_hidden_states, vocab_embedding) + + return output diff --git a/deepspeed/inference/v2/modules/interfaces/__init__.py b/deepspeed/inference/v2/modules/interfaces/__init__.py new file mode 100644 index 000000000000..13b556789e4e --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/__init__.py @@ -0,0 +1,12 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .attention_base import DSSelfAttentionRegistry, DSSelfAttentionBase +from .embedding_base import DSEmbeddingRegistry, DSEmbeddingBase +from .linear_base import DSLinearRegistry, DSLinearBase +from .moe_base import DSMoERegistry, DSMoEBase +from .post_norm_base import DSPostNormRegistry, DSPostNormBase +from .pre_norm_base import DSPreNormRegistry, DSPreNormBase +from .unembed_base import DSUnembedRegistry, DSUnembedBase diff --git a/deepspeed/inference/v2/modules/interfaces/attention_base.py b/deepspeed/inference/v2/modules/interfaces/attention_base.py new file mode 100644 index 000000000000..c67dc033f92a --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/attention_base.py @@ -0,0 +1,97 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional, Tuple, Type + +import torch + +from ...ragged import RaggedBatchWrapper +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ..ds_module import DSModuleBase +from ..module_registry import DSModuleRegistryBase +from ..configs import DSSelfAttentionConfig + + +class DSSelfAttentionBase(DSModuleBase): + """ + Base mixin for all attention modules. The interface represented by this module + is broadly: + + output = attention(query_key_value, + Optional[kv_cache], + Optional[attention_mask], + Optional[attention_bias]) + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSSelfAttentionConfig + + def __init__(self, config: DSSelfAttentionConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + @property + def kv_block_size(self) -> int: + """ + Return preferred granulatity for blocked KV-cache implementation. + """ + raise NotImplementedError() + + @property + def q_block_size(self) -> int: + """ + Property to calculate blocking granularity for the query dimension. + This has no impact on the KV-cache structure, but will affect the + number of attention atoms associated with a batch. + """ + raise NotImplementedError() + + def build_atoms(self, ragged_batch: RaggedBatchWrapper) -> None: + """ + Build the atoms for this module. This is not a strict requirement for the class, + so this method is a no-op by default rather than abstract. + """ + pass + + def forward(self, + q_k_v: torch.Tensor, + kv_cache: torch.Tensor, + batch: RaggedBatchWrapper, + attention_mask: Optional[torch.Tensor] = None, + attention_bias: Optional[torch.Tensor] = None, + inv_freqs: Optional[torch.Tensor] = None) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Parameters: + q_k_v (torch.Tensor): Query, key, and value tensors. Expected shape is: + [ + batch, + seq_len, + 2 * self._config.n_heads_kv + self._config.n_heads_q, + self._config.head_size + ]. + kv_cache (Optional[torch.Tensor]): Key and value cache tensor. Expected shape is + [ + 2, + batch, + kv_cache_len, + self._config.n_heads_kv, + self._config.head_size + ]. If None, cache is disabled. The `kv_cache_len` dimension does not need to + be contiguous (it should expand stride by `max_out_tokens`). + batch (RaggedBatchWrapper): Ragged batch metadata. + attention_mask (Optional[torch.Tensor]): Attention mask tensor. If None, masking is + disabled. This will defer to the config in the case of conflicting information. + This means if the config class is implying causal attention, the mask will be ignored. + attention_bias (Optional[torch.Tensor]): Attention bias tensor. If None, bias is disabled. + """ + raise NotImplementedError() + + +class DSSelfAttentionRegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSSelfAttentionBase diff --git a/deepspeed/inference/v2/modules/interfaces/embedding_base.py b/deepspeed/inference/v2/modules/interfaces/embedding_base.py new file mode 100644 index 000000000000..8078013e36b6 --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/embedding_base.py @@ -0,0 +1,84 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import abstractmethod +from typing import Any, Dict, Optional, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ...ragged import RaggedBatchWrapper +from ..ds_module import DSModuleBase +from ..module_registry import DSModuleRegistryBase +from ..configs import DSEmbeddingsConfig + + +class DSEmbeddingBase(DSModuleBase): + """ + Base mixin for embedding modules. The interface represented by this module is: + + hidden_out = embedding(input_ids) + + position_embedding(position_ids) + + token_type_embedding(token_type_ids) + with optional normalization. + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSEmbeddingsConfig + + def __init__(self, config: DSEmbeddingsConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + def transform_param(self, embed_param: torch.Tensor) -> torch.Tensor: + """ + Perform any necessary transformations on an embedding parameter. This module assumes + that all embedding parameters would require the same set of transformations. + + Parameters: + embed_param (torch.Tensor): Embedding parameter. Shape is of [vocab_size, hidden_size] + """ + raise NotImplementedError() + + @property + @abstractmethod + def output(self) -> torch.Tensor: + """ + Pre-allocated output Tensor. This currently needs to be exposed for gather operations + on the output. + + TODO(cmikeh2): This is not ideal. We need a better abstraction for this, such as giving + access to the inference comm object to the DSModule. + """ + raise NotImplementedError() + + def forward(self, + ragged_batch: RaggedBatchWrapper, + word_embeddings: torch.Tensor, + position_embeddings: Optional[torch.Tensor] = None, + token_type_ids: Optional[torch.Tensor] = None, + token_type_embeddings: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Parameters: + ragged_batch (torch.Tensor): Ragged batch of token ids + associated metadata. + word_embeddings (torch.Tensor): Word embeddings. + position_embeddings (torch.Tensor): Position embeddings. If passed, IDs will be + inferred from the ragged batch itself. + token_type_ids (torch.Tensor): Token type ids. + token_type_embeddings (torch.Tensor): Token type embeddings. + + Returns: + torch.Tensor: Hidden states. This should be the sum of the relevant + encodings for the model. + """ + raise NotImplementedError() + + +class DSEmbeddingRegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSEmbeddingBase diff --git a/deepspeed/inference/v2/modules/interfaces/linear_base.py b/deepspeed/inference/v2/modules/interfaces/linear_base.py new file mode 100644 index 000000000000..bcaad6fe269a --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/linear_base.py @@ -0,0 +1,71 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import abstractmethod +from typing import Any, Dict, Optional, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ..ds_module import DSModuleBase +from ..module_registry import DSModuleRegistryBase +from ..configs import DSLinearConfig + + +class DSLinearBase(DSModuleBase): + """ + Base mixin for all Linear modules. The interface represented by this module + is: + + hidden_out = activation(hidden_in * weight + bias) + + The format and dtype of the weight and bias tensors are not defined and implementations + may compress as necessary. Must support a bias. + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSLinearConfig + + def __init__(self, config: DSLinearConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + @abstractmethod + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Perform any necessary transformations of the parameters of this module. + + Parameters: + param (torch.Tensor): Weight or bias tensor. + """ + ... + + def forward(self, hidden_states: torch.Tensor, w: torch.Tensor, b: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Parameters: + hidden_states (torch.Tensor): Hidden states tensor. Expected shape is either + [batch, seq_len, in_channels] or [batch, in_channels]. + + Returns: + torch.Tensor: Output tensor. Tensor should have same number of dimensions as + input tensor. + """ + raise NotImplementedError() + + @property + @abstractmethod + def output(self) -> torch.Tensor: + """ + Return the padded, pre-allocated output Tensor. + """ + ... + + +class DSLinearRegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSLinearBase diff --git a/deepspeed/inference/v2/modules/interfaces/moe_base.py b/deepspeed/inference/v2/modules/interfaces/moe_base.py new file mode 100644 index 000000000000..cc80ca55f60a --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/moe_base.py @@ -0,0 +1,90 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import abstractmethod +from typing import Any, Dict, Optional, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ..ds_module import DSModuleBase +from ..module_registry import DSModuleRegistryBase +from ..configs import DSMoEConfig + + +class DSMoEBase(DSModuleBase): + """ + Base mixing for MoE modules. The interface represented by this module is: + + expert_assignments = gate(hidden_states) + intermediate = ragged_linear(hidden_states, expert_assignments) + output = ragged_linear(intermediate, expert_assignments) + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSMoEConfig + + def __init__(self, config: DSMoEConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + @abstractmethod + def transform_gate_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Perform any necessary transformations of the gate parameter. + + Args: + param (torch.Tensor): gate_w (shape: [num_experts, model_dim]) + """ + ... + + @abstractmethod + def transform_moe_mlp_1_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Perform any necessary transformations of the parameter. The specific component + being transformed should be inferred from the shape of the parameter. + + Args: + param (torch.Tensor): One of either mlp_1_w, mlp_1_b + """ + ... + + @abstractmethod + def transform_moe_mlp_2_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Perform any necessary transformations of the parameter. The specified component being + transformed should be inferred from the shape of the parameter. This interface is + separate from transform_moe_1_param because the two components may have identical + shapes. + + Args: + param (torch.Tensor): One of either mlp_2_w or mlp_2_b + """ + ... + + def forward(self, + hidden_states: torch.Tensor, + gate_w: torch.Tensor, + mlp_1_w: torch.Tensor, + mlp_2_w: torch.Tensor, + mlp_1_b: Optional[torch.Tensor] = None, + mlp_2_b: Optional[torch.Tensor] = None) -> torch.Tensor: + raise NotImplementedError() + + @property + @abstractmethod + def output(self) -> torch.Tensor: + """ + Returns the pre-allocated, padded output Tensor. + """ + ... + + +class DSMoERegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSMoEBase diff --git a/deepspeed/inference/v2/modules/interfaces/post_norm_base.py b/deepspeed/inference/v2/modules/interfaces/post_norm_base.py new file mode 100644 index 000000000000..c2a6bf69de8a --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/post_norm_base.py @@ -0,0 +1,68 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import abstractmethod +from typing import Any, Dict, Optional, Tuple, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ..ds_module import DSModuleBase +from ..configs.norm_config import DSNormConfig +from ..module_registry import DSModuleRegistryBase + + +class DSPostNormBase(DSModuleBase): + """ + Base MixIn for all Post-Normalization modules. The interface represented by this + module is: + + residual, hidden_out = norm(residual + hidden_in) + + If residual and hidden_out are the same data type, then they may alias each other. + Furthermore, residual should be updated in-place. + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSNormConfig + + def __init__(self, config: DSNormConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + @abstractmethod + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Transform a gamma/beta parameter. It is assumed that both transformations are + the same. + + Parameters: + param (torch.Tensor): Gamma or beta parameter. + """ + ... + + def forward(self, + residual: torch.Tensor, + hidden_states: torch.Tensor, + gamma: torch.Tensor, + beta: Optional[torch.Tensor] = None) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Parameters: + residual (torch.Tensor): Residual tensor. + hidden_states (torch.Tensor): Hidden states tensor. + + Returns: + (torch.Tensor, torch.Tensor): Tuple of residual and hidden states. + Hidden states may alias with residual. + """ + raise NotImplementedError() + + +class DSPostNormRegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSPostNormBase diff --git a/deepspeed/inference/v2/modules/interfaces/pre_norm_base.py b/deepspeed/inference/v2/modules/interfaces/pre_norm_base.py new file mode 100644 index 000000000000..7d8b4ebf1587 --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/pre_norm_base.py @@ -0,0 +1,72 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import abstractmethod +from typing import Any, Dict, Optional, Tuple, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ..ds_module import DSModuleBase +from ..configs.norm_config import DSNormConfig +from ..module_registry import DSModuleRegistryBase + + +class DSPreNormBase(DSModuleBase): + """ + Base mixin for all Pre-Normalization modules. The interface represented by this module + is: + + if hidden_in is not None: + residual_out = residual + hidden_in + else: + residual_out = residual + + hidden_out = normalize(residual_out) + return residual_out, hidden_out + + Residual should be updated in-place. + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSNormConfig + + def __init__(self, config: DSNormConfig, implementation_config: Dict[str, Any]): + super().__init__(config, implementation_config) + + @abstractmethod + def transform_param(self, param: torch.Tensor) -> torch.Tensor: + """ + Transform a gamma/beta parameter. It is assumed that both transformations are + the same. + + Parameters: + param (torch.Tensor): Gamma or beta parameter. + """ + ... + + def forward(self, + residual: torch.Tensor, + hidden_states: Optional[torch.Tensor], + gamma: torch.Tensor, + beta: Optional[torch.Tensor] = None) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Parameters: + residual (torch.Tensor): Residual tensor. + hidden_states (torch.Tensor): Hidden states tensor. + + Returns: + (torch.Tensor, torch.Tensor): Tuple of residual and hidden states. + """ + raise NotImplementedError() + + +class DSPreNormRegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSPreNormBase diff --git a/deepspeed/inference/v2/modules/interfaces/unembed_base.py b/deepspeed/inference/v2/modules/interfaces/unembed_base.py new file mode 100644 index 000000000000..9eca6fcde768 --- /dev/null +++ b/deepspeed/inference/v2/modules/interfaces/unembed_base.py @@ -0,0 +1,61 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Any, Dict, Optional, Type + +import torch + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ...ragged import RaggedBatchWrapper +from ..ds_module import DSModuleBase +from ..module_registry import DSModuleRegistryBase +from ..configs import DSUnembedConfig + + +class DSUnembedBase(DSModuleBase): + """ + Base mixin for unmebedding modules. The interface represented by this module is: + + if config.do_normalization + hidden = layer_norm(hidden) + logits = hidden @ projection + """ + + @staticmethod + def config_class() -> Type[DeepSpeedConfigModel]: + return DSUnembedConfig + + def __init__(self, config: DSUnembedConfig, implementation_config: Dict[str, Any]) -> None: + super().__init__(config, implementation_config) + + def forward(self, + hidden_states: torch.Tensor, + vocab_embedding: torch.Tensor, + ragged_metadata: RaggedBatchWrapper, + gamma: Optional[torch.Tensor] = None, + beta: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Forward interface. Gamma and beta are optional parameters passed depending on + `self.config.do_normalization`. + + Args: + hidden_states (torch.Tensor): Hidden states of shape [tokens, model_dim] + vocab_embedding (torch.Tensor): Embedding matrix of shape [vocab_size, model_dim] + ragged_metadata (RaggedBatchWrapper): Metadata for the ragged batch. + gamma (Optional[torch.Tensor]): Gamma parameter for layer norm. + beta (Optional[torch.Tensor]): Beta parameter for layer norm. + + Returns: + torch.Tensor: Unembedded hidden states of shape [n_seqs, model_dim] + """ + raise NotImplementedError() + + +class DSUnembedRegistry(DSModuleRegistryBase): + registry: Dict = {} + + @staticmethod + def associated_class() -> Type[DSModuleBase]: + return DSUnembedBase diff --git a/deepspeed/inference/v2/modules/module_registry.py b/deepspeed/inference/v2/modules/module_registry.py new file mode 100644 index 000000000000..e04b8d734518 --- /dev/null +++ b/deepspeed/inference/v2/modules/module_registry.py @@ -0,0 +1,58 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from abc import ABC, abstractstaticmethod +from typing import Any, Dict, Type + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from .ds_module import DSModuleBase + + +class ConfigBundle(DeepSpeedConfigModel): + """ + A config bundle is a collection of configs that are used to instantiate a model implementation. + """ + name: str + config: DeepSpeedConfigModel + implementation_config: Dict[str, Any] = {} + + +class DSModuleRegistryBase(ABC): + """ + Class holding logic for tracking the DSModule implementations of a given interface. + """ + + @classmethod + def instantiate_config(cls, config_bundle: ConfigBundle) -> DSModuleBase: + """ + Given a DSModule key, attempt to instantiate + """ + if config_bundle.name not in cls.registry: + raise KeyError(f"Unknown DSModule: {config_bundle.name}, cls.registry={cls.registry}") + + target_implementation = cls.registry[config_bundle.name] + if not target_implementation.supports_config(config_bundle.config): + raise ValueError(f"Config {config_bundle.config} is not supported by {target_implementation}") + + return cls.registry[config_bundle.name](config_bundle.config, config_bundle.implementation_config) + + @abstractstaticmethod + def associated_class() -> Type[DSModuleBase]: + """ + Return the class associated with this registry. + """ + raise NotImplementedError("Must associated a DSModule class with its registry.") + + @classmethod + def register_module(cls, child_class: DSModuleBase) -> None: + """ + Register a module with this registry. + """ + if not issubclass(child_class, cls.associated_class()): + raise TypeError( + f"Can only register subclasses of {cls.associated_class()}, {child_class} does not inherit from {cls.associated_class()}" + ) + cls.registry[child_class.name()] = child_class + return child_class diff --git a/deepspeed/inference/v2/ragged/__init__.py b/deepspeed/inference/v2/ragged/__init__.py new file mode 100644 index 000000000000..3af09cff4be5 --- /dev/null +++ b/deepspeed/inference/v2/ragged/__init__.py @@ -0,0 +1,15 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from .kv_cache import split_kv +from .manager_configs import ( + AllocationMode, + DSStateManagerConfig, + KVCacheConfig, + MemoryConfig, +) +from .ragged_manager import DSStateManager +from .ragged_wrapper import RaggedBatchWrapper +from .sequence_descriptor import DSSequenceDescriptor, PlaceholderSequenceDescriptor diff --git a/deepspeed/inference/v2/ragged/blocked_allocator.py b/deepspeed/inference/v2/ragged/blocked_allocator.py new file mode 100644 index 000000000000..7884d8cccb47 --- /dev/null +++ b/deepspeed/inference/v2/ragged/blocked_allocator.py @@ -0,0 +1,105 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Iterable, Union + +import torch + + +class BlockedAllocator: + """ + Allocator class for managing which blocks are free/used in the + blocked KV-cache. This is a simple allocator that uses a linked list + to keep track of which blocks are free/used. The cost of allocation/deallocation + is O(blocks), where blocks is the number of blocks to allocate/deallocate. + + TODO(cmikeh2): Evaluate performance of this allocator and migrate + to C++ if necessary. + """ + # Number of blocks in the KV-cache(s). + _num_blocks: int + + # Array of blocks, where each element is the next block in the linked list. + _blocks: torch.Tensor + + # Index of the head of the linked list. + _head: int + + # Number of free blocks in the KV-cache. + _free_blocks: int + + def __init__(self, num_blocks: int) -> None: + """ + Initialize an allocator with `num_blocks` blocks. This requires at least + `num_blocks` * 4 bytes of host memory. + + Parameters: + num_blocks (int): The number of blocks to allocate. + """ + + if num_blocks < 1: + raise ValueError(f'Blocked KV-cache must have at least 1 block, provided {num_blocks}') + + self._num_blocks = num_blocks + self._blocks = torch.arange(1, num_blocks + 1, dtype=torch.int32, device='cpu', pin_memory=True) + self._head = 0 + self._free_blocks = num_blocks + + def allocate(self, num_blocks: int) -> torch.Tensor: + """ + Allocate a list of blocks from the associated KV-caches. This will + return `num_blocks` blocks from the KV-cache if they are available, + or raise an exception if there are not enough free blocks. + + Parameters: + num_blocks (int): The number of blocks to allocate. + + Returns: + List[int]: The list of blocks allocated. + """ + if num_blocks > self._free_blocks: + raise ValueError(f'Not enough free blocks in the KV-cache to allocate {num_blocks} blocks') + + allocated_blocks = torch.zeros(num_blocks, dtype=torch.int32) + for i in range(num_blocks): + allocated_blocks[i] = self._head + self._head = self._blocks[self._head].item() + self._blocks[allocated_blocks[i]] = -1 # Mark as used + self._free_blocks -= 1 + + return allocated_blocks + + def free(self, blocks: Union[Iterable[int], int]) -> None: + """ + Return a list of blocks to the free pool. If a single invalid block is provided (i.e., + one that is out of range of the allocator or is already free), then an exception is raised + and no blocks are freed. + + Parameters: + blocks (Union[Iterable[int], int]): The list of blocks to free. If only one block + is to be freed, this can be alone as an integer. + """ + if isinstance(blocks, int): + blocks = [blocks] + + for block in blocks: + # Parse all blocks for validity before mutating the list. + if block < 0 or block >= self._num_blocks: + raise ValueError(f'Invalid block {block} provided to free') + + if self._blocks[block] != -1: + raise ValueError(f'Block {block} is already free') + + for block in blocks: + self._blocks[block] = self._head + self._head = block + self._free_blocks += 1 + + @property + def free_blocks(self) -> int: + """ + Return the number of free blocks in the KV-cache. + """ + return self._free_blocks diff --git a/deepspeed/inference/v2/ragged/csrc/fast_host_buffer.cu b/deepspeed/inference/v2/ragged/csrc/fast_host_buffer.cu new file mode 100644 index 000000000000..31347636b50c --- /dev/null +++ b/deepspeed/inference/v2/ragged/csrc/fast_host_buffer.cu @@ -0,0 +1,18 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include "ds_kernel_utils.h" +#include "fast_host_buffer.h" + +void* get_cuda_fast_buffer(int64_t size) +{ + void* buffer_ptr; + // Host allocation flags that should minimize the host -> accelerator copy latency + unsigned int alloc_flags = + cudaHostAllocPortable | cudaHostAllocMapped | cudaHostAllocWriteCombined; + + cudaHostAlloc(&buffer_ptr, size, alloc_flags); + return buffer_ptr; +} diff --git a/deepspeed/inference/v2/ragged/csrc/ragged_ops.cpp b/deepspeed/inference/v2/ragged/csrc/ragged_ops.cpp new file mode 100644 index 000000000000..0c5e8812c84c --- /dev/null +++ b/deepspeed/inference/v2/ragged/csrc/ragged_ops.cpp @@ -0,0 +1,45 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#include +#include + +#include "fast_host_buffer.h" + +/* +Similar to doing an empty_like to replicate a Tensor on the host, but will +attempt to optimize for faster host -> accelerator copies. Since this is on the critical +path for the forward pass, this should directly improve performance. +Allocates the shadow buffers for the input_ids, batch, seq and kv_ids tensors. + +Arguments: + device_mirror: A tensor on the accelerator that should be mirrored by the host. + +Returns: + A tensor on the host of the same size and datatype optimized for fast host -> accelerator +copies. +*/ +torch::Tensor allocate_fast_host_buffer(torch::Tensor device_mirror) +{ +#ifdef __HIP_PLATFORM_HCC__ + auto options = + torch::TensorOptions().device(torch::kCPU).pinned_memory(true).dtype(device_mirror.dtype()); + auto buffer = torch::empty(device_mirror.sizes(), options); +#else + + void* buffer_ptr = get_cuda_fast_buffer(device_mirror.numel() * device_mirror.element_size()); + + auto options = torch::TensorOptions().device(torch::kCPU).dtype(device_mirror.dtype()); + auto buffer = torch::from_blob(buffer_ptr, device_mirror.sizes(), options); +#endif + return buffer; +} + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) +{ + m.def("allocate_fast_host_buffer", + &allocate_fast_host_buffer, + "Allocate a host mirror of an accelerator Tensor."); +} diff --git a/deepspeed/inference/v2/ragged/includes/fast_host_buffer.h b/deepspeed/inference/v2/ragged/includes/fast_host_buffer.h new file mode 100644 index 000000000000..81f24ed8fdaa --- /dev/null +++ b/deepspeed/inference/v2/ragged/includes/fast_host_buffer.h @@ -0,0 +1,14 @@ +// Copyright (c) Microsoft Corporation. +// SPDX-License-Identifier: Apache-2.0 + +// DeepSpeed Team + +#pragma once + +#include "ds_kernel_utils.h" + +/* +Wrapper around cudaHostAlloc with some specific flags. Returns a pointer to the +memory region of `size` bytes. +*/ +void* get_cuda_fast_buffer(int64_t size); diff --git a/deepspeed/inference/v2/ragged/kv_cache.py b/deepspeed/inference/v2/ragged/kv_cache.py new file mode 100644 index 000000000000..d743d6e92530 --- /dev/null +++ b/deepspeed/inference/v2/ragged/kv_cache.py @@ -0,0 +1,165 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import operator +from functools import reduce +from typing import Any, Iterable, Optional, Tuple + +import torch + +import deepspeed.comm as dist +from deepspeed.comm.reduce_op import ReduceOp + +from deepspeed.accelerator import get_accelerator +from ..inference_utils import elem_size +from ..logging import inference_logger +from .blocked_allocator import BlockedAllocator +from .manager_configs import AllocationMode, KVCacheConfig, MemoryConfig + + +def split_kv(kv_cache: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Split a KV cache instance into its key and value components. + + Parameters: + kv_cache (torch.Tensor): The KV-cache to split. This should be a 5D tensor with the + following shape: [num_blocks, block_size, 2, num_heads, head_size] + + Returns: + Tuple[torch.Tensor, torch.Tensor]: The key and value components of the KV-cache. Both + tensors will have the shape [num_blocks, block_size, num_heads, head_size]. + """ + if kv_cache.ndim != 5: + raise ValueError(f"KV-cache must have 5 dimensions, got {kv_cache.ndim}.") + + return kv_cache[:, :, 0, :, :], kv_cache[:, :, 1, :, :] + + +class BlockedKVCache: + + _caches: torch.Tensor + """ + Backing storage for all KV caches. This is a 6D tensor with the following shape: + (num_caches, num_blocks, block_size, 2, num_heads, head_size) + """ + + _allocator: BlockedAllocator + """ + Block allocator for tracking cache usage. This manages the GPU cache. + """ + + _config: KVCacheConfig + """ + Configuration of the KV cache. See ``KVCacheConfig`` for more details. + """ + + def __init__(self, + config: KVCacheConfig, + memory_config: MemoryConfig, + mp_group: Optional[Any] = None, + offload: bool = False) -> None: + """ + Create a container that will maintain the storage and allocations for a set of + blocked KV-caches. + + Parameters: + config (KVCacheConfig): The configuration of the KV-cache. + slack (int): The amount of slack space to reserve in GPU memory for the cache. + enable_offload (bool): Whether to enable offloading of the cache to the host. + blocks (int): The number of blocks to pre-allocate for the cache. If this is set, + slack will be ignored. + """ + self._config = config + self._memory_config = memory_config + self._enable_offload = offload + + if self._enable_offload: + raise NotImplementedError("Offloading of KV-caches is not yet supported.") + + if AllocationMode(self._memory_config.mode) is AllocationMode.RESERVE: + per_block_footprint = reduce(operator.mul, self._config.cache_shape, self._config.block_size) + per_block_footprint *= 2 # for key and value + per_block_footprint *= elem_size(self._config.cache_dtype) + + # Perform a dummy nccl call before calculating available memory, on some systems (H100) we've observed higher memory allocations from NCCL + if dist.get_world_size(group=mp_group) > 1: + dummy_tensor = torch.tensor(0, dtype=torch.int32, device=get_accelerator().current_device()) + dist.all_reduce(dummy_tensor, op=ReduceOp.MIN, group=mp_group) + + get_accelerator().empty_cache() + available_kv_memory = get_accelerator().available_memory() - self._memory_config.size + total_memory = get_accelerator().total_memory() + + inference_logger().debug( + f"Memory usage before KV-cache allocation: total_memory={total_memory}, available_kv_memory={available_kv_memory}, per_block_footprint={per_block_footprint}" + ) + + if available_kv_memory < per_block_footprint: + raise ValueError( + f"Insufficient memory to allocate KV-caches. Required: {per_block_footprint}, Available: {available_kv_memory}" + ) + + num_blocks = available_kv_memory // per_block_footprint + + # In a multi-process setting, we need to ensure that all processes have the same + # KV cache capacity to ensure scheduling guarantees are equivalent on all ranks. + if dist.get_world_size(group=mp_group) > 1: + reduce_tensor = torch.tensor(num_blocks, dtype=torch.int32, device=get_accelerator().current_device()) + dist.all_reduce(reduce_tensor, op=ReduceOp.MIN, group=mp_group) + num_blocks = reduce_tensor.item() + + # This is ugly but don't want the fragmentation of the 8 byte Tensor maybe + # hanging around. + del reduce_tensor + get_accelerator().empty_cache() + else: # AllocationMode.ALLOCATE + num_blocks = self._memory_config.size + + num_caches = self._config.cache_shape[0] + num_heads = self._config.cache_shape[1] + head_size = self._config.cache_shape[2] + + alloc_shape = (num_caches, num_blocks, self._config.block_size, 2, num_heads, head_size) + inference_logger().info(f"Allocating KV-cache with shape: {alloc_shape} consisting of {num_blocks} blocks.") + self._caches = torch.empty(alloc_shape, + dtype=self._config.cache_dtype, + device=get_accelerator().current_device()) + self._allocator = BlockedAllocator(num_blocks) + + def reserve(self, num_blocks: int) -> torch.Tensor: + """ + Reserve a number of blocks from the cache. This will return a 1D tensor of + block_ids that have been marked as reserved. + """ + return self._allocator.allocate(num_blocks) + + def free(self, blocks: Iterable[int]) -> None: + """ + Free a set of blocks from the cache. This will mark the blocks as free in the + allocator. + """ + self._allocator.free(blocks) + + def offload(self, blocks: Iterable[int]) -> torch.Tensor: + """ + Offload KV-cache blocks from accelerator memory to the host. + """ + raise NotImplementedError("Offloading is not yet supported.") + + def restore(self, blocks: Iterable[int]) -> torch.Tensor: + """ + Restore KV-cache blocks from the host to accelerator memory. + """ + raise NotImplementedError("Offloading is not yet supported.") + + def get_cache(self, cache_id: int) -> torch.Tensor: + """ + Get the tensor associated with the given cache ID. + """ + return self._caches[cache_id] + + @property + def free_blocks(self): + return self._allocator.free_blocks diff --git a/deepspeed/inference/v2/ragged/manager_configs.py b/deepspeed/inference/v2/ragged/manager_configs.py new file mode 100644 index 000000000000..a35c1cfd2369 --- /dev/null +++ b/deepspeed/inference/v2/ragged/manager_configs.py @@ -0,0 +1,164 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from enum import Enum +from typing import Tuple + +from deepspeed.pydantic_v1 import PositiveInt, validator + +from deepspeed.runtime.config_utils import DeepSpeedConfigModel +from ..inference_utils import DtypeEnum + + +class KVCacheConfig(DeepSpeedConfigModel): + + block_size: int = 128 + """ + Number of tokens that may be contained in each cache block. + """ + + num_allocation_groups: PositiveInt = 1 + """ + Allocation groups are assumed to be able to use the same allocation block size because + the allocation granularity is the same but the number of blocks required in each group + may differ. + + As a concrete example, consider a model with alternating layers of local and global + attention (such as GPTNeo). The local attention layers do not require the same number + of cache blocks as the global layer. However, a static partitioning scheme is sub-optimal since the ratio of local to global KV-cache blocks is not constant across + the range of sequence lengths that may be encountered. + + NOTE: In theory, this functionality could be used to do per-head and per-layer + KV-cache allocation, but it is likely the allocator will struggle with managing that + many blocks. + + NOTE: This will need to be primarily understood and handled by the model implementation + itself, rather than the KV cache manager. However, I'd like to make this explicit. + """ + + cache_shape: Tuple[PositiveInt, PositiveInt, PositiveInt] + """ + The shape of the cache per token. The first dimension is the number of individual + caches, the second is the number of heads, and the third is the head size. The number + of caches argument here is per allocation group. + """ + + cache_dtype: DtypeEnum = DtypeEnum.fp16 + """ + Data type of the KV-cache. + """ + + max_blocks_per_allocation_group: PositiveInt = 64 + """ + Maximum number of blocks that can be associated with an allocation group. + """ + + +""" +The config above is a little confusing so let's use a couple of concrete examples of +usage: + +Model 1: Llama-13B with a block size of 256 + +Llama is uniform attention so we have a single allocation group. The cache shape is +(40 layers, 40 heads, 128 head size) + +```python +llama_kv_config = KVCacheConfig(block_size=256, + num_allocation_groups=1, + cache_shape=(40, 40, 128)) +``` + +Model 2: GPTNeo-2.7B with a block size of 128 + +GPTNeo has alternating local and global attention layers. We have two allocation groups. +There are 16 layers of each type with 20 heads apiece at 128 head size. + +```python +gptneo_kv_config = KVCacheConfig(num_allocation_groups=2, cache_shape=(16, 20, 128)) +``` +""" + + +class AllocationMode(Enum): + """ + Helper class to describe memory allocation strategies for the KV-cache. + """ + + RESERVE = "reserve" + """ + Reserve a small amount of memory for non-KV cache allocations. + """ + + ALLOCATE = "allocate" + """ + Allocate an explicit number of KV blocks. + """ + + +class MemoryConfig(DeepSpeedConfigModel): + + mode: AllocationMode = AllocationMode.RESERVE + + size: PositiveInt = 1_000_000_000 + """ + Parameter for each of the modes. + + If mode is RESERVE, this is the amount of memory in bytes to reserve after allocating the + KV-cache. If in a tensor-parallel regime, this amount is guaranteed to be reserved on + all devices. + + If mode is ALLOCATE, this is the number of blocks to allocate for the KV-cache. This may + require tuning for model/GPU setups. + """ + + +class DSStateManagerConfig(DeepSpeedConfigModel): + + max_tracked_sequences: PositiveInt = 2048 + """ + How many sequences this engine will track simultaneously. This limit should be greater + than the ``max_ragged_sequence_count``. + """ + + max_ragged_batch_size: PositiveInt = 768 + """ + The maximum number of tokens that can be contained in a single ragged batch. Passing + a larger value than this will raise an exception that must be handled by the runtime. + """ + + max_ragged_sequence_count: PositiveInt = 512 + """ + The maximum number of sequences that can compose a batch. This limitation is only + relevant under CUDA graphing scenarios currently, where the maximum number of blocks + is largely bound by the total number of sequences in the ragged batch. This number cannot + be larger than ``max_tracked_sequences`` or ``max_ragged_batch_size``. + """ + + max_context: PositiveInt = 8192 + """ + The maximum number of tokens (inclusive of generation) that can be contained in a single + sequence. Currently used to bound the size of the KV cache metadata. + """ + + memory_config: MemoryConfig = MemoryConfig() + """ + Directive for how to manage the creation of the KV-cache. See MemoryConfig for more + details. + """ + + offload: bool = False + """ + Enable tracking for offloading KV-cache to host memory. Currently unsupported. + """ + + @validator("max_ragged_sequence_count") + def max_ragged_sequence_count_validator(cls, v: int, values: dict): + # If the attributes below failed their validation they won't appear in the values dict. + if "max_tracked_sequences" in values and v > values["max_tracked_sequences"]: + raise ValueError("max_ragged_sequence_count must be less than max_tracked_sequences") + if "max_ragged_batch_size" in values and v > values["max_ragged_batch_size"]: + raise ValueError("max_ragged_sequence_count must be less than max_ragged_batch_size") + return v diff --git a/deepspeed/inference/v2/ragged/ragged_manager.py b/deepspeed/inference/v2/ragged/ragged_manager.py new file mode 100644 index 000000000000..202f007707c5 --- /dev/null +++ b/deepspeed/inference/v2/ragged/ragged_manager.py @@ -0,0 +1,204 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch +from typing import Any, Dict, Optional, Tuple + +from deepspeed.accelerator import get_accelerator +from deepspeed.ops.op_builder import RaggedUtilsBuilder +from deepspeed.utils.logging import logger + +from .blocked_allocator import BlockedAllocator +from .kv_cache import BlockedKVCache +from .manager_configs import DSStateManagerConfig, KVCacheConfig +from .sequence_descriptor import DSSequenceDescriptor + + +class DSStateManager: + """ + Base abstract class for managing blocked KV caches. Will probably have a single + implementation for now. + """ + + _config: DSStateManagerConfig + """ + Config for state management. See DSStateManagerConfig for more details. The arguments here + should come from the engine config. + """ + + _kv_config: KVCacheConfig + """ + Config for the KV cache. See KVCacheConfig for more details. These arguments should derive + from the model implementation. + """ + + _kv_cache: BlockedKVCache + """ + Persistent KV cache store. + """ + + # Container for tracking all sequences in the system. + _seqs: Dict[int, DSSequenceDescriptor] + """ + Container for tracking all sequences in the system. + + TODO(cmikeh2): Evaluate if this has any performance implications. + """ + + # Allocator for tracking sequences. + _tracking_allocator: BlockedAllocator + _all_block_ids: torch.Tensor + _all_block_ids_shadow: torch.Tensor + + # TODO(cmikeh2): This interface both needs to be more flexible and more concrete. + def __init__(self, + config: DSStateManagerConfig, + kv_config: KVCacheConfig, + base_mp_group: Optional[Any] = None) -> None: + """ + The key + + Parameters: + block_size (int): The number of tokens to allocate in each block. + """ + self._config = config + self._kv_config = kv_config + + # Load our helpers for host allocation. + self._ragged_utils = RaggedUtilsBuilder().load() + + # Initialize the allocator for tracking sequences (so this doesn't need to be ad-hoc). + self._tracking_allocator = BlockedAllocator(self._config.max_tracked_sequences) + + # Storage to back tracking the KV cache allocation. + ids_shape = ( + self._config.max_tracked_sequences, + self._kv_config.num_allocation_groups, + self._kv_config.max_blocks_per_allocation_group, + ) + self._all_block_ids = torch.zeros(ids_shape, dtype=torch.int32, device=get_accelerator().current_device()) + self._all_block_ids_shadow = self._ragged_utils.allocate_fast_host_buffer(self._all_block_ids) + + # Initialize the sequence container. + self._seqs = {} + + # Finally initialize the KV cache. + self._kv_cache = BlockedKVCache(self._kv_config, + self._config.memory_config, + mp_group=base_mp_group, + offload=self._config.offload) + + def get_cache(self, cache_id: int) -> torch.Tensor: + """ + Return the Tensor associated with the given cache id. + """ + return self._kv_cache.get_cache(cache_id) + + def query(self, uid: Optional[int] = None) -> Tuple[int, int, Optional[int]]: + """ + Query the state of the KV cache for occupancy. + + Parameters: + seq_id (Optional[int]): The sequence id to query. If None, the last + return value will be None. + + Returns: + Tuple[int, int, Optional[Tuple[int, int]]: A tuple of the block size, the number of + free blocks, and the number of cached tokens for the given sequence. + """ + if uid is not None: + cached_toks = self._seqs[uid].cached_tokens + free_toks = cached_toks % self._block_size + return (self._block_size, self._kv_cache.free_blocks, free_toks) + else: + return (self._block_size, self._kv_cache.free_blocks, None) + + def flush_sequence(self, uid: int) -> None: + """ + Free all resources associated with the given sequence id. + """ + if uid not in self._seqs: + logger.warning(f"Attempting to flush sequence {uid} which does not exist.") + return + + seq = self._seqs[uid] + self._kv_cache.free(seq.all_block_ids) + self._tracking_allocator.free(seq.tracking_id) + del self._seqs[uid] + + def get_sequence(self, uid: int) -> Optional[DSSequenceDescriptor]: + """ + Get the sequence descriptor for the given sequence id. If the sequence does not exist, + then None is returned. + """ + if uid not in self._seqs: + return None + + return self._seqs[uid] + + def get_or_create_sequence(self, uid: int) -> DSSequenceDescriptor: + """ + Get the existing sequence descriptor for a given uid or initialize one if + it does not exist. NOTE: This will always return a valid sequence descriptor + if one may be allocated and should not be used from APIs that are attempting + to test the schedulability of a hypothetical batch. + """ + if uid in self._seqs: + return self._seqs[uid] + else: + return self._create_sequence(uid) + + def _create_sequence(self, uid: int) -> DSSequenceDescriptor: + """ + Create a new sequence descriptor for the given sequence id. + """ + if uid in self._seqs: + raise ValueError(f"Sequence {uid} already exists.") + + try: + tracking_slot = self._tracking_allocator.allocate(1).item() + except ValueError: + raise RuntimeError( + f"Unable to create tracking slot for sequence {uid} since the metadata buffers are full.") + + seq_block_ids = self._all_block_ids[tracking_slot] + seq_block_ids_shadow = self._all_block_ids_shadow[tracking_slot] + self._seqs[uid] = DSSequenceDescriptor(tracking_slot, + seq_block_ids, + seq_block_ids_shadow, + max_context=self._config.max_context) + logger.debug(f"Created sequence {uid} with tracking slot {tracking_slot}.") + return self._seqs[uid] + + @property + def tracked_sequences(self) -> Dict[int, DSSequenceDescriptor]: + """ + Return the tracked sequences. + """ + return self._seqs + + @property + def n_tracked_sequences(self) -> int: + """ + Return the number of sequences currently tracked. + """ + return len(self._seqs) + + @property + def kv_block_size(self) -> int: + """ + Return the block size of the KV cache. + """ + return self._kv_config.block_size + + @property + def free_blocks(self) -> int: + """ + Return the number of free blocks in the KV cache. + """ + return self._kv_cache.free_blocks + + def allocate_blocks(self, n_blocks: int) -> torch.Tensor: + return self._kv_cache.reserve(n_blocks) diff --git a/deepspeed/inference/v2/ragged/ragged_wrapper.py b/deepspeed/inference/v2/ragged/ragged_wrapper.py new file mode 100644 index 000000000000..15abd1987aa1 --- /dev/null +++ b/deepspeed/inference/v2/ragged/ragged_wrapper.py @@ -0,0 +1,267 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.ops.op_builder import RaggedUtilsBuilder + +from .sequence_descriptor import DSSequenceDescriptor +from .manager_configs import DSStateManagerConfig + + +def to_padded(original_size: int) -> int: + """ + Pad to a backend friendly granularity. + """ + + def _pad_to_mul_of_pow2(val: int, pow_2_val: int) -> int: + return val + (pow_2_val - 1) & ~(pow_2_val - 1) + + # TODO(cmikeh2): Tune this approach. This is mainly a placeholder right now. + granularity = 64 if original_size <= 512 else 128 + + return _pad_to_mul_of_pow2(original_size, granularity) + + +class RaggedBatchWrapper: + """ + Container for all the auxiliary Tensors used in the management of a ragged batch. + + For each Tensor, we maintain a shadow Tensor on the host. This Tensor is what is + directly populated when constructing the ragged batch. The shadow Tensors, when possible, + should be allocated so as to support fast host-to-accelerator copies. + """ + + # Tensors to populate the ragged batch into. + _input_ids_shadow: torch.Tensor + _input_ids: torch.Tensor + """ + Forward pass input buffer. + """ + + _batch_metadata_storage: torch.Tensor + _batch_metadata_storage_shadow: torch.Tensor + """ + Holds the number of inflight sequences and tokens for the ragged batch. + """ + + _token_to_seq_storage: torch.Tensor + _token_to_seq_storage_shadow: torch.Tensor + """ + Linear mapping for each of the tokens. Let's say we have 8 tokens in the batch, + with the sequence breakdown being [4, 1, 3]. Then, the mapping would be: + [0, 0, 0, 0, 1, 2, 2, 2] + """ + + _inflight_seq_descriptors: torch.Tensor + _inflight_seq_descriptors_shadow: torch.Tensor + """ + For each sequence in the batch, we store the start token in the batch, the number of tokens + the number of tokens in the history of this sequence, and an unused 4th reserved for alignment. + For the above example this would give: + [[0, 4, H0, X], [4, 1, H1, X], [5, 3, H2, X]] + """ + + # Holds the block ids for each sequence in the ragged batch. + _kv_ptrs: torch.Tensor + _kv_ptrs_shadow: torch.Tensor + """ + List of ptrs pointing to the GPU buffer that holds the KV-block ids for each sequence. + If there are multiple allocation groups associated with each of the sequences, then + then accessing the Nth cache will require accessing the Nth block id + """ + + def __init__(self, config: DSStateManagerConfig) -> None: + """ + Convenience wrapper around the data structures used to represent a ragged + batch for inference. Only a single `RaggedBatchWrapper` should be used per + ragged inference engine. + + The underlying data structures are implemented in `ragged_batch_descriptor.h`. + """ + self._config = config + self._input_ids = torch.zeros((self._config.max_ragged_batch_size), + dtype=torch.int64, + device=get_accelerator().current_device()) + + self._batch_metadata_storage = torch.zeros(2, dtype=torch.int32, device=get_accelerator().current_device()) + + self._token_to_seq_storage = torch.zeros((self._config.max_ragged_batch_size), + dtype=torch.int32, + device=get_accelerator().current_device()) + self._inflight_seq_descriptors = torch.zeros((self._config.max_ragged_sequence_count, 4), + dtype=torch.int32, + device=get_accelerator().current_device()) + self._kv_ptrs = torch.zeros((self._config.max_ragged_sequence_count), + dtype=torch.int64, + device=get_accelerator().current_device()) + + self._utils_module = RaggedUtilsBuilder().load() + host_alloc = self._utils_module.allocate_fast_host_buffer + + self._input_ids_shadow = host_alloc(self._input_ids) + self._batch_metadata_storage_shadow = host_alloc(self._batch_metadata_storage) + self._token_to_seq_storage_shadow = host_alloc(self._token_to_seq_storage) + self._inflight_seq_descriptors_shadow = host_alloc(self._inflight_seq_descriptors) + self._kv_ptrs_shadow = host_alloc(self._kv_ptrs) + + # Default behavior should be no padding + self._is_padded = False + + def clear(self) -> None: + """ + Clear the ragged batch. This will reset the number of tokens and sequences to 0. + """ + self._batch_metadata_storage_shadow[0] = 0 + self._batch_metadata_storage_shadow[1] = 0 + + def insert_sequence(self, seq_descriptor: DSSequenceDescriptor, tokens: torch.Tensor, do_checks=True) -> None: + """ + Incrementally insert a sequence into the ragged batch. This will update the + metadata for the ragged batch and the sequence. + + Arguments: + seq_descriptor () + """ + if tokens.device != torch.device("cpu"): + # This doesn't really fall under schedulability, so we'll unconditionally check for it. + raise RuntimeError(f"Expected tokens to be on host but found device '{tokens.device}'") + + if do_checks and self.current_sequences == self._config.max_ragged_sequence_count: + raise RuntimeError(f"Ragged batch is full due to sequence limit: {self._config.max_ragged_sequence_count}") + + seq_tokens = tokens.numel() + + if do_checks and self.current_tokens + seq_tokens > self._config.max_ragged_batch_size: + raise RuntimeError(f"Ragged batch is full due to capacity limit: {self._config.max_ragged_batch_size})") + + self._input_ids_shadow[self.current_tokens:self.current_tokens + seq_tokens].copy_(tokens) + self._token_to_seq_storage_shadow[self.current_tokens:self.current_tokens + seq_tokens].fill_( + self.current_sequences) + + self._inflight_seq_descriptors_shadow[self.current_sequences][0] = self.current_tokens + self._inflight_seq_descriptors_shadow[self.current_sequences][1] = seq_tokens + self._inflight_seq_descriptors_shadow[self.current_sequences][2] = seq_descriptor.seen_tokens + + self._kv_ptrs_shadow[self.current_sequences] = seq_descriptor.kv_blocks_ptr + + self._batch_metadata_storage_shadow[0] += seq_tokens + self._batch_metadata_storage_shadow[1] += 1 + + @property + def tensor_toks(self) -> torch.Tensor: + """ + The number of tokens in the in-flight ragged batch. This will not trigger + synchronization with the device. + """ + cur_toks = self.current_tokens + if self._is_padded: + return to_padded(cur_toks) + else: + return cur_toks + + def finalize(self, padding: Optional[bool] = False) -> None: + """ + Completes construction of the ragged batch by flushing the host buffers to the device. + """ + cur_toks = self.current_tokens + + if padding: + padded_toks = to_padded(cur_toks) + self._input_ids_shadow[cur_toks:padded_toks].fill_(-1) + self._token_to_seq_storage_shadow[cur_toks:padded_toks].fill_(-1) + self._is_padded = True + else: + padded_toks = cur_toks + self._is_padded = False + + current_sequences = self.current_sequences + + def _noblock_copy(dst: torch.Tensor, src: torch.Tensor) -> None: + dst.copy_(src, non_blocking=True) + + _noblock_copy(self._input_ids[:padded_toks], self._input_ids_shadow[:padded_toks]) + _noblock_copy(self._batch_metadata_storage, self._batch_metadata_storage_shadow) + _noblock_copy(self._token_to_seq_storage[:padded_toks], self._token_to_seq_storage_shadow[:padded_toks]) + _noblock_copy(self._inflight_seq_descriptors[:current_sequences], + self._inflight_seq_descriptors_shadow[:current_sequences]) + _noblock_copy(self._kv_ptrs[:current_sequences], self._kv_ptrs_shadow[:current_sequences]) + + def input_ids(self, on_device: bool = True) -> torch.Tensor: + """ + The input ids tensor for the ragged batch. If the device Tensor is requested, the Tensor + is truncated to the number of tokens in the batch. + """ + if on_device: + return self._input_ids[:self.tensor_toks] + else: + return self._input_ids_shadow + + def batch_metadata_buffer(self, on_device: bool = True) -> torch.Tensor: + """ + Buffer associated with the batch metadata tensor that can + be populated in preparation for passing a new input to the device. + """ + if on_device: + return self._batch_metadata_storage + else: + return self._batch_metadata_storage_shadow + + def tokens_to_seq(self, on_device: bool = True) -> torch.Tensor: + """ + Mapping of token to which sequence it belongs to in the ragged batch. If the device Tensor + is requested, the Tensor is truncated to the number of tokens in the batch. + """ + if on_device: + return self._token_to_seq_storage[:self.tensor_toks] + else: + return self._token_to_seq_storage_shadow + + def inflight_seq_descriptors(self, on_device: bool = True) -> torch.Tensor: + """ + Buffer associated with the metadata of each sequence in the ragged batch. If the device Tensor + is requested, the Tensor is truncated to the number of sequences in the batch. + """ + if on_device: + return self._inflight_seq_descriptors[:self.current_sequences] + else: + return self._inflight_seq_descriptors_shadow + + def kv_ptrs(self, on_device: bool = True) -> torch.Tensor: + """ + Pointer to where the list of KV ids associated with a sequence are. If the device Tensor + is requested, the Tensor is truncated to the number of sequences in the batch. + """ + if on_device: + return self._kv_ptrs[:self.current_sequences] + else: + return self._kv_ptrs_shadow + + def masks(self, on_device: bool = True) -> Optional[torch.Tensor]: + """ + Placeholder for supporting complex masks. Currently not supported. + + Models that will need this will be BERT-like, not generative. + """ + return None + + @property + def current_tokens(self) -> int: + """ + The number of tokens in the in-flight ragged batch. This will not trigger + synchronization with the device. + """ + return self._batch_metadata_storage_shadow[0].item() + + @property + def current_sequences(self) -> int: + """ + The number of sequences in the in-flight ragged batch. This will not trigger + synchronization with the device. + """ + return self._batch_metadata_storage_shadow[1].item() diff --git a/deepspeed/inference/v2/ragged/sequence_descriptor.py b/deepspeed/inference/v2/ragged/sequence_descriptor.py new file mode 100644 index 000000000000..b5cad9d2cd4d --- /dev/null +++ b/deepspeed/inference/v2/ragged/sequence_descriptor.py @@ -0,0 +1,214 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import List, Union + +import torch + + +class BaseSequenceDescriptor: + + @property + def seen_tokens(self) -> int: + """ + The number of tokens for this sequence that have completed a forward pass. + """ + raise NotImplementedError() + + @property + def cur_allocated_blocks(self) -> int: + """ + The number of KV blocks currently allocated for this sequence. + """ + raise NotImplementedError() + + @property + def kv_blocks_ptr(self) -> int: + """ + The pointer to the KV blocks for this sequence. + """ + raise NotImplementedError() + + +class PlaceholderSequenceDescriptor(BaseSequenceDescriptor): + """ + The DummySequenceDescriptor is an empty object that allows us to perform schedulability + checks before formally tracking a sequence. + """ + + def __init__(self, seen_tokens=0, cur_allocated_blocks=0, kv_blocks_ptr=0) -> None: + self._seen_tokens = seen_tokens + self._cur_allocated_blocks = cur_allocated_blocks + self._kv_blocks_ptr = kv_blocks_ptr + + @property + def seen_tokens(self) -> int: + return self._seen_tokens + + @property + def cur_allocated_blocks(self) -> int: + return self._cur_allocated_blocks + + @property + def kv_blocks_ptr(self) -> int: + return self._kv_blocks_ptr + + +class DSSequenceDescriptor(BaseSequenceDescriptor): + + _seen_tokens: int + """ + Number of tokens in the sequence that have completed a forward pass. + """ + + _in_flight_tokens: int + """ + Number of tokens that have begun a forward pass but not yet completed it. + """ + + _max_context: int + """ + Maximum number of tokens this sequence may eventually include. Currently unused but + may be used in future implementations for speculative caching. + """ + + _num_allocation_groups: int + """ + Number of unique allocation groups associated with the sequence. + """ + + _blocks_per_allocation_group: torch.IntTensor + """ + Number of blocks allocated for each allocation group. + """ + + # Padded list of KV-cache IDs for the sequence. + _kv_cache_ids: torch.Tensor + _kv_cache_ids_shadow: torch.Tensor + """ + Padded list of KV-cache IDs for the sequence. The padded shape is [num_allocation_groups, max_blocks_per_allocation_group]. + """ + + # The location in the broader ID tensor where the KV-cache IDs for the sequence + # are stored. Used on flush. + _tracking_id: int + + def __init__(self, + tracking_id: int, + kv_cache_ids: torch.Tensor, + kv_cache_ids_shadow: torch.Tensor, + max_context: int = -1) -> None: + self._tracking_id = tracking_id + self._kv_cache_ids = kv_cache_ids + self._kv_cache_ids_shadow = kv_cache_ids_shadow + self._max_context = max_context + + self._seen_tokens = 0 + self._in_flight_tokens = 0 + + self._num_allocation_groups = kv_cache_ids_shadow.shape[0] + self._blocks_per_allocation_group = torch.zeros(self._num_allocation_groups, dtype=torch.int32, device="cpu") + + assert kv_cache_ids.shape[0] == self._num_allocation_groups + assert len(kv_cache_ids.shape) == 2 + + @property + def seen_tokens(self) -> int: + return self._seen_tokens + + @property + def in_flight_tokens(self) -> int: + return self._in_flight_tokens + + @property + def max_context(self) -> int: + return self._max_context + + @property + def cur_allocated_blocks(self) -> int: + return self._blocks_per_allocation_group.sum() + + @property + def tracking_id(self) -> int: + return self._tracking_id + + def kv_cache_ids(self, on_device: bool = False) -> torch.Tensor: + """ + Returns the Tensor containing the block IDs for this sequence on the appropriate device. + """ + if on_device: + return self._kv_cache_ids + else: + return self._kv_cache_ids_shadow + + @property + def kv_blocks_ptr(self) -> int: + return self._kv_cache_ids.data_ptr() + + @property + def all_block_ids(self) -> torch.Tensor: + block_ids = [] + for allocation_group, num_blocks in zip(self._kv_cache_ids, self._blocks_per_allocation_group): + block_ids.append(allocation_group[:num_blocks]) + return torch.cat(block_ids) + + def pre_forward(self, num_tokens: int) -> None: + """ + Update the state of the sequence before a forward pass. + """ + self._in_flight_tokens = num_tokens + + def post_forward(self) -> None: + """ + Update the state of the sequence after a forward pass. + """ + self._seen_tokens += self._in_flight_tokens + self._in_flight_tokens = 0 + + def extend_kv_cache(self, new_ids: Union[List[torch.IntTensor], torch.IntTensor]) -> None: + """ + Extend the KV-cache for the sequence. + + Args: + new_ids (Union[List[torch.IntTensor], torch.IntTensor]): For each allocation group, the IDs + to add to the KV-cache. If there is only one allocation group, a single tensor can be + provided. Otherwise, a list of tensors should be provided. The tensors do not need + to have the same shape. + """ + if isinstance(new_ids, torch.Tensor): + new_ids = [new_ids] + + if len(new_ids) != self._num_allocation_groups: + raise ValueError(f"Only {len(new_ids)} allocation groups provided, expected {self._num_allocation_groups}") + + for group_id, new_group_ids in enumerate(new_ids): + new_blocks = new_group_ids.numel() + + if new_blocks == 0: + # If we have multiple groups, it's possible to have an empty group. + continue + + shadow_alloc_group = self._kv_cache_ids_shadow[group_id] + alloc_group = self._kv_cache_ids[group_id] + cur_blocks = self._blocks_per_allocation_group[group_id] + + shadow_alloc_group[cur_blocks:cur_blocks + new_blocks].copy_(new_group_ids) + alloc_group[cur_blocks:cur_blocks + new_blocks].copy_(shadow_alloc_group[cur_blocks:cur_blocks + + new_blocks], + non_blocking=True) + + self._blocks_per_allocation_group[group_id] += new_blocks + + def free_kv_cache(self, free_ids: Union[List[torch.IntTensor], torch.IntTensor]) -> None: + """ + Free blocks from the KV-cache for the sequence. + + Args: + free_ids (Union[List[torch.IntTensor], torch.IntTensor]): The ids of blocks to free + from the KV-cache. If there is only one allocation group, a single tensor can be + provided. Otherwise, a list of tensors should be provided. The tensors do not need + to have the same shape. + """ + raise NotImplementedError("Partial KV-cache freeing is not yet supported.") diff --git a/deepspeed/inference/v2/scheduling_utils.py b/deepspeed/inference/v2/scheduling_utils.py new file mode 100644 index 000000000000..6d3818d46675 --- /dev/null +++ b/deepspeed/inference/v2/scheduling_utils.py @@ -0,0 +1,54 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from enum import Enum + + +class SchedulingResult(Enum): + + Success = 0 + """ + The proposed batch is valid and can be scheduled. + """ + + EngineSequenceLimitExceeded = 1 + """ + The proposed batch would would overflow the number of concurrent sequences the engine may support. + """ + + BatchSequenceLimitExceeded = 2 + """ + The proposed batch contains more sequences than the engine was configured + to support in a single forwardp + """ + + BatchTokenLimitExceeded = 3 + """ + The proposed batch contains more tokens than the engine was configured + to support in a single forward. + """ + + KVCacheLimitExceeded = 4 + """ + The proposed batch would require more KV cache to be allocated than the engine + currently has available. + """ + + SequenceTokenLimitExceeded = 5 + """ + The proposed batch contains a sequence that is longer than the engine/model can support. + """ + + +class SchedulingError(RuntimeError): + + result: SchedulingResult + """ + The failed result of the scheduling check. Guaranteed to not be SchedulingResult.Success. + """ + + def __init__(self, result: SchedulingResult) -> None: + self.result = result + super().__init__(f"Batch scheduling failed with result {result}") diff --git a/op_builder/builder.py b/op_builder/builder.py index 62683774f20a..163463c28c47 100644 --- a/op_builder/builder.py +++ b/op_builder/builder.py @@ -102,6 +102,7 @@ def assert_no_cuda_mismatch(name=""): class OpBuilder(ABC): _rocm_version = None _is_rocm_pytorch = None + _loaded_ops = {} def __init__(self, name): self.name = name @@ -433,6 +434,9 @@ def builder(self): extra_link_args=self.strip_empty_entries(self.extra_ldflags())) def load(self, verbose=True): + if self.name in __class__._loaded_ops: + return __class__._loaded_ops[self.name] + from deepspeed.git_version_info import installed_ops, torch_info if installed_ops.get(self.name, False): # Ensure the op we're about to load was compiled with the same @@ -441,7 +445,9 @@ def load(self, verbose=True): if torch.cuda.is_available() and isinstance(self, CUDAOpBuilder): self.validate_torch_op_version(torch_info) - return importlib.import_module(self.absolute_name()) + op_module = importlib.import_module(self.absolute_name()) + __class__._loaded_ops[self.name] = op_module + return op_module else: return self.jit_load(verbose) @@ -485,6 +491,8 @@ def jit_load(self, verbose=True): if not self.build_for_cpu and self.enable_bf16: cxx_args.append("-DBF16_AVAILABLE") nvcc_args.append("-DBF16_AVAILABLE") + nvcc_args.append("-U__CUDA_NO_BFLOAT16_OPERATORS__") + nvcc_args.append("-U__CUDA_NO_BFLOAT162_OPERATORS__") if self.is_rocm_pytorch(): cxx_args.append("-D__HIP_PLATFORM_AMD__=1") @@ -505,6 +513,8 @@ def jit_load(self, verbose=True): if torch_arch_list: os.environ["TORCH_CUDA_ARCH_LIST"] = torch_arch_list + __class__._loaded_ops[self.name] = op_module + return op_module diff --git a/op_builder/inference_core_ops.py b/op_builder/inference_core_ops.py new file mode 100755 index 000000000000..b8ea54cd0b3f --- /dev/null +++ b/op_builder/inference_core_ops.py @@ -0,0 +1,92 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import os + +from .builder import CUDAOpBuilder, installed_cuda_version + + +class InferenceCoreBuilder(CUDAOpBuilder): + BUILD_VAR = "DS_BUILD_INFERENCE_CORE_OPS" + NAME = "inference_core_ops" + + def __init__(self, name=None): + name = self.NAME if name is None else name + super().__init__(name=name) + + def absolute_name(self): + return f'deepspeed.inference.v2.kernels{self.NAME}' + + def is_compatible(self, verbose=True): + try: + import torch + except ImportError: + self.warning("Please install torch if trying to pre-compile inference kernels") + return False + + cuda_okay = True + if not self.is_rocm_pytorch() and torch.cuda.is_available(): #ignore-cuda + sys_cuda_major, _ = installed_cuda_version() + torch_cuda_major = int(torch.version.cuda.split('.')[0]) + cuda_capability = torch.cuda.get_device_properties(0).major #ignore-cuda + if cuda_capability < 6: + self.warning("NVIDIA Inference is only supported on Pascal and newer architectures") + cuda_okay = False + if cuda_capability >= 8: + if torch_cuda_major < 11 or sys_cuda_major < 11: + self.warning("On Ampere and higher architectures please use CUDA 11+") + cuda_okay = False + return super().is_compatible(verbose) and cuda_okay + + def filter_ccs(self, ccs): + ccs_retained = [] + ccs_pruned = [] + for cc in ccs: + if int(cc[0]) >= 6: + ccs_retained.append(cc) + else: + ccs_pruned.append(cc) + if len(ccs_pruned) > 0: + self.warning(f"Filtered compute capabilities {ccs_pruned}") + return ccs_retained + + def get_prefix(self): + ds_path = self.deepspeed_src_path("deepspeed") + return "deepspeed" if os.path.isdir(ds_path) else ".." + + def sources(self): + sources = [ + "inference/v2/kernels/core_ops/core_ops.cpp", + "inference/v2/kernels/core_ops/bias_activations/bias_activation.cpp", + "inference/v2/kernels/core_ops/bias_activations/bias_activation.cu", + "inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cpp", + "inference/v2/kernels/core_ops/cuda_layer_norm/layer_norm.cu", + "inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cpp", + "inference/v2/kernels/core_ops/cuda_rms_norm/rms_norm.cu", + "inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cpp", + "inference/v2/kernels/core_ops/gated_activations/gated_activation_kernels.cu", + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + return sources + + def extra_ldflags(self): + return [] + + def include_paths(self): + sources = [ + 'inference/v2/kernels/core_ops/bias_activations', + 'inference/v2/kernels/core_ops/blas_kernels', + 'inference/v2/kernels/core_ops/cuda_layer_norm', + 'inference/v2/kernels/core_ops/cuda_rms_norm', + 'inference/v2/kernels/core_ops/gated_activations', + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + + sources.append('csrc/includes') + return sources diff --git a/op_builder/inference_cutlass_builder.py b/op_builder/inference_cutlass_builder.py new file mode 100644 index 000000000000..73400b2ce10e --- /dev/null +++ b/op_builder/inference_cutlass_builder.py @@ -0,0 +1,92 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team +import os + +from .builder import CUDAOpBuilder, installed_cuda_version + + +class InferenceCutlassBuilder(CUDAOpBuilder): + BUILD_VAR = "DS_BUILD_CUTLASS_OPS" + NAME = "cutlass_ops" + + def __init__(self, name=None): + name = self.NAME if name is None else name + super().__init__(name=name) + + def absolute_name(self): + return f'deepspeed.inference.v2.kernels.cutlass_ops.{self.NAME}' + + def is_compatible(self, verbose=True): + try: + import torch + except ImportError: + self.warning("Please install torch if trying to pre-compile inference kernels") + return False + + cuda_okay = True + if not self.is_rocm_pytorch() and torch.cuda.is_available(): #ignore-cuda + sys_cuda_major, _ = installed_cuda_version() + torch_cuda_major = int(torch.version.cuda.split('.')[0]) + cuda_capability = torch.cuda.get_device_properties(0).major #ignore-cuda + if cuda_capability < 6: + self.warning("NVIDIA Inference is only supported on Pascal and newer architectures") + cuda_okay = False + if cuda_capability >= 8: + if torch_cuda_major < 11 or sys_cuda_major < 11: + self.warning("On Ampere and higher architectures please use CUDA 11+") + cuda_okay = False + return super().is_compatible(verbose) and cuda_okay + + def filter_ccs(self, ccs): + ccs_retained = [] + ccs_pruned = [] + for cc in ccs: + if int(cc[0]) >= 8: + # Only support Ampere and newer + ccs_retained.append(cc) + else: + ccs_pruned.append(cc) + if len(ccs_pruned) > 0: + self.warning(f"Filtered compute capabilities {ccs_pruned}") + return ccs_retained + + def get_prefix(self): + ds_path = self.deepspeed_src_path("deepspeed") + return "deepspeed" if os.path.isdir(ds_path) else ".." + + def sources(self): + sources = [ + "inference/v2/kernels/cutlass_ops/cutlass_ops.cpp", + "inference/v2/kernels/cutlass_ops/mixed_gemm/mixed_gemm.cu", + "inference/v2/kernels/cutlass_ops/moe_gemm/moe_gemm.cu", + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + return sources + + def extra_ldflags(self): + import dskernels + lib_path = dskernels.library_path() + prefix = self.get_prefix() + lib_path = os.path.join(prefix, lib_path) + lib_path = self.deepspeed_src_path(lib_path) + + args = [f'-L{lib_path}', '-ldeepspeedft'] + if self.jit_load: + args.append(f'-Wl,-rpath,{lib_path}') + return args + + def include_paths(self): + sources = [ + 'inference/v2/kernels/cutlass_ops/mixed_gemm', + 'inference/v2/kernels/cutlass_ops/moe_gemm', + 'inference/v2/kernels/cutlass_ops/shared_resources/', + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + sources.append('csrc/includes') + return sources diff --git a/op_builder/ragged_ops.py b/op_builder/ragged_ops.py new file mode 100644 index 000000000000..505aaa9708cf --- /dev/null +++ b/op_builder/ragged_ops.py @@ -0,0 +1,116 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import os + +from .builder import CUDAOpBuilder, installed_cuda_version + + +class RaggedOpsBuilder(CUDAOpBuilder): + BUILD_VAR = "DS_BUILD_RAGGED_DEVICE_OPS" + NAME = "ragged_device_ops" + + def __init__(self, name=None): + name = self.NAME if name is None else name + super().__init__(name=name) + + def absolute_name(self): + return f'deepspeed.inference.v2.kernels.ragged_ops.{self.NAME}' + + def is_compatible(self, verbose=True): + try: + import torch + except ImportError: + self.warning("Please install torch if trying to pre-compile inference kernels") + return False + + cuda_okay = True + if not self.is_rocm_pytorch() and torch.cuda.is_available(): #ignore-cuda + sys_cuda_major, _ = installed_cuda_version() + torch_cuda_major = int(torch.version.cuda.split('.')[0]) + cuda_capability = torch.cuda.get_device_properties(0).major #ignore-cuda + if cuda_capability < 6: + self.warning("NVIDIA Inference is only supported on Pascal and newer architectures") + cuda_okay = False + if cuda_capability >= 8: + if torch_cuda_major < 11 or sys_cuda_major < 11: + self.warning("On Ampere and higher architectures please use CUDA 11+") + cuda_okay = False + return super().is_compatible(verbose) and cuda_okay + + def filter_ccs(self, ccs): + ccs_retained = [] + ccs_pruned = [] + for cc in ccs: + if int(cc[0]) >= 8: + # Blocked flash has a dependency on Ampere + newer + ccs_retained.append(cc) + else: + ccs_pruned.append(cc) + if len(ccs_pruned) > 0: + self.warning(f"Filtered compute capabilities {ccs_pruned}") + return ccs_retained + + def get_prefix(self): + ds_path = self.deepspeed_src_path("deepspeed") + return "deepspeed" if os.path.isdir(ds_path) else ".." + + def sources(self): + sources = [ + "inference/v2/kernels/ragged_ops/ragged_ops.cpp", + "inference/v2/kernels/ragged_ops/atom_builder/atom_builder.cpp", + "inference/v2/kernels/ragged_ops/blocked_flash/blocked_flash.cpp", + "inference/v2/kernels/ragged_ops/embed/embed.cpp", + "inference/v2/kernels/ragged_ops/embed/embed.cu", + "inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cpp", + "inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary/blocked_kv_rotary.cu", + "inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cpp", + "inference/v2/kernels/ragged_ops/logits_gather/logits_gather.cu", + "inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cpp", + "inference/v2/kernels/ragged_ops/moe_scatter/moe_scatter.cu", + "inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cpp", + "inference/v2/kernels/ragged_ops/moe_gather/moe_gather.cu", + "inference/v2/kernels/ragged_ops/ragged_helpers/ragged_kernel_helpers.cpp", + "inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cpp", + "inference/v2/kernels/ragged_ops/top_1_gating/top_1_gating.cu", + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + return sources + + def extra_ldflags(self): + import dskernels + lib_path = dskernels.library_path() + + prefix = self.get_prefix() + lib_path = os.path.join(prefix, lib_path) + lib_path = self.deepspeed_src_path(lib_path) + + args = [f'-L{lib_path}', '-lblockedflash'] + if self.jit_load: + args.append(f'-Wl,-rpath,{lib_path}') + return args + + def include_paths(self): + sources = [ + 'inference/v2/kernels/ragged_ops', + 'inference/v2/kernels/ragged_ops/atom_builder', + 'inference/v2/kernels/ragged_ops/blocked_flash', + 'inference/v2/kernels/ragged_ops/embed', + 'inference/v2/kernels/ragged_ops/linear_blocked_kv_rotary', + 'inference/v2/kernels/ragged_ops/logits_gather', + 'inference/v2/kernels/ragged_ops/moe_gather', + 'inference/v2/kernels/ragged_ops/moe_scatter', + 'inference/v2/kernels/ragged_ops/ragged_helpers', + 'inference/v2/kernels/ragged_ops/top_1_gating', + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + + sources.append('csrc/includes') + + return sources diff --git a/op_builder/ragged_utils.py b/op_builder/ragged_utils.py new file mode 100755 index 000000000000..c1d38eef1981 --- /dev/null +++ b/op_builder/ragged_utils.py @@ -0,0 +1,77 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import os + +from .builder import CUDAOpBuilder, installed_cuda_version + + +class RaggedUtilsBuilder(CUDAOpBuilder): + BUILD_VAR = "DS_BUILD_RAGGED_OPS" + NAME = "ragged_ops" + + def __init__(self, name=None): + name = self.NAME if name is None else name + super().__init__(name=name) + + def absolute_name(self): + return f'deepspeed.inference.v2.{self.NAME}' + + def is_compatible(self, verbose=True): + try: + import torch + except ImportError: + self.warning("Please install torch if trying to pre-compile inference kernels") + return False + + cuda_okay = True + if not self.is_rocm_pytorch() and torch.cuda.is_available(): #ignore-cuda + sys_cuda_major, _ = installed_cuda_version() + torch_cuda_major = int(torch.version.cuda.split('.')[0]) + cuda_capability = torch.cuda.get_device_properties(0).major #ignore-cuda + if cuda_capability < 6: + self.warning("NVIDIA Inference is only supported on Pascal and newer architectures") + cuda_okay = False + if cuda_capability >= 8: + if torch_cuda_major < 11 or sys_cuda_major < 11: + self.warning("On Ampere and higher architectures please use CUDA 11+") + cuda_okay = False + return super().is_compatible(verbose) and cuda_okay + + def filter_ccs(self, ccs): + ccs_retained = [] + ccs_pruned = [] + for cc in ccs: + if int(cc[0]) >= 6: + ccs_retained.append(cc) + else: + ccs_pruned.append(cc) + if len(ccs_pruned) > 0: + self.warning(f"Filtered compute capabilities {ccs_pruned}") + return ccs_retained + + def get_prefix(self): + ds_path = self.deepspeed_src_path("deepspeed") + return "deepspeed" if os.path.isdir(ds_path) else ".." + + def sources(self): + sources = [ + "inference/v2/ragged/csrc/fast_host_buffer.cu", + "inference/v2/ragged/csrc/ragged_ops.cpp", + ] + + prefix = self.get_prefix() + sources = [os.path.join(prefix, src) for src in sources] + return sources + + def extra_ldflags(self): + return [] + + def include_paths(self): + include_dir = "inference/v2/ragged/includes" + prefix = self.get_prefix() + include_dir = os.path.join(prefix, include_dir) + + return ['csrc/includes', include_dir] diff --git a/requirements/requirements-dev.txt b/requirements/requirements-dev.txt index 5d0e963779a3..105dd094f995 100644 --- a/requirements/requirements-dev.txt +++ b/requirements/requirements-dev.txt @@ -1,6 +1,7 @@ accelerate clang-format==16.0.2 coverage +deepspeed-kernels docutils<0.18 future importlib-metadata>=4 diff --git a/scripts/check-torchcuda.py b/scripts/check-torchcuda.py index 48581628930b..0723c9888369 100755 --- a/scripts/check-torchcuda.py +++ b/scripts/check-torchcuda.py @@ -19,6 +19,8 @@ def err(s: str) -> None: print(s, file=sys.stderr) +print(*sys.argv[1:]) + # There are many ways we could search for the string "torch.cuda", but `git # grep --no-index` is nice because # - it's very fast (as compared to iterating over the file in Python) diff --git a/tests/pytest.ini b/tests/pytest.ini index 20d6ca0624f2..cc6b6564daa8 100644 --- a/tests/pytest.ini +++ b/tests/pytest.ini @@ -1,9 +1,11 @@ [pytest] -addopts = -m "not sequential and not nightly and not inference and not seq_inference and not inference_ops" +addopts = -m "not sequential and not nightly and not inference and not seq_inference and not inference_ops and not inference_v2 and not inference_v2_ops" markers = sequential:Tests that need to be run sequentially inference:Inference model tests inference_ops:Individual inference operator tests + inference_v2: Inference tests for the v2 stack + inference_v2_ops: Op tests for the v2 stack seq_inference:Inference model tests to run sequentially nightly:Tests that should be run nightly world_size:Change world size of individual tests in a class diff --git a/tests/unit/inference/__init__.py b/tests/unit/inference/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/inference_test_utils.py b/tests/unit/inference/inference_test_utils.py new file mode 100644 index 000000000000..d63c51267e51 --- /dev/null +++ b/tests/unit/inference/inference_test_utils.py @@ -0,0 +1,46 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Tuple + +import torch +from deepspeed.accelerator import get_accelerator + +TOLERANCES = None + + +def get_tolerances(): + global TOLERANCES + if TOLERANCES is None: + TOLERANCES = {torch.float32: (5e-4, 5e-5), torch.float16: (3e-2, 2e-3)} + if get_accelerator().is_bf16_supported(): + # Note: BF16 tolerance is higher than FP16 because of the lower precision (7 (+1) bits vs + # 10 (+1) bits) + TOLERANCES[torch.bfloat16] = (4.8e-1, 3.2e-2) + return TOLERANCES + + +DTYPES = None + + +def get_dtypes(include_float=True): + global DTYPES + if DTYPES is None: + DTYPES = [torch.float16, torch.float32] if include_float else [torch.float16] + try: + if get_accelerator().is_bf16_supported(): + DTYPES.append(torch.bfloat16) + except (AssertionError, AttributeError): + pass + return DTYPES + + +def allclose(x, y, tolerances: Tuple[int, int] = None): + assert x.dtype == y.dtype + if tolerances is None: + rtol, atol = get_tolerances()[x.dtype] + else: + rtol, atol = tolerances + return torch.allclose(x, y, rtol=rtol, atol=atol) diff --git a/tests/unit/inference/kernels/__init__.py b/tests/unit/inference/kernels/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/kernels/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/kernels/core_ops/__init__.py b/tests/unit/inference/kernels/core_ops/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/kernels/core_ops/test_bias_activation.py b/tests/unit/inference/kernels/core_ops/test_bias_activation.py new file mode 100644 index 000000000000..2c6134991597 --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/test_bias_activation.py @@ -0,0 +1,101 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import ActivationType, DtypeEnum +from deepspeed.inference.v2.kernels.core_ops import CUDABiasActivation +from ...inference_test_utils import get_dtypes, allclose + + +def reference_bias_act_implementation(input: torch.Tensor, bias: Optional[torch.Tensor], + act_type: ActivationType) -> torch.Tensor: + bias_func_map = { + ActivationType.RELU: torch.nn.functional.relu, + ActivationType.GELU: torch.nn.functional.gelu, + ActivationType.SILU: torch.nn.functional.silu, + ActivationType.IDENTITY: lambda x: x, + } + + dtype = input.dtype + input_f = input.to(torch.float32) + if bias is not None: + bias_f = bias.to(torch.float32) + output_f = input_f + bias_f + else: + output_f = input_f + output_f = bias_func_map[act_type](output_f) + + return output_f.to(dtype) + + +def _bias_activation_test_helper(tokens: int, + channels: int, + act_fn: ActivationType, + dtype: DtypeEnum, + use_bias: bool = True) -> None: + """ + Fully parameterized testing entry point. + """ + # Input vals + input_tensor = torch.randn((tokens, channels), dtype=dtype.value, device=get_accelerator().current_device_name()) + if use_bias: + bias = torch.randn((channels), dtype=dtype.value, device=get_accelerator().current_device_name()) + else: + bias = None + + # Reference output + ref_output = reference_bias_act_implementation(input_tensor, bias, act_fn) + + bias_act = CUDABiasActivation(channels, dtype, act_fn) + + # New output + ds_tensor = input_tensor.clone() + bias_act(ds_tensor, bias) + + # Check + assert allclose(ds_tensor, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, channels", [(1, 4096), (37, 2048), (112, 14432), (1024, 6144)]) +@pytest.mark.parametrize("dtype", get_dtypes(include_float=False)) +def test_token_channels_permutations(tokens: int, channels: int, dtype: torch.dtype) -> None: + """ + Validate bias activation kernel with different token and channel permutations when using the RELU + activation function. + """ + act_fn = ActivationType.RELU + dtype = DtypeEnum(dtype) + _bias_activation_test_helper(tokens, channels, act_fn, dtype) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("act_fn", + [ActivationType.RELU, ActivationType.GELU, ActivationType.SILU, ActivationType.IDENTITY]) +def test_act_fns(act_fn: ActivationType) -> None: + """ + Validate bias activation kernel with different activation functions. + """ + tokens = 223 + channels = 4096 + dtype = DtypeEnum.fp16 + _bias_activation_test_helper(tokens, channels, act_fn, dtype) + + +@pytest.mark.inference_v2_ops +def test_no_bias() -> None: + """ + Validate bias activation kernel with no bias. + """ + tokens = 223 + channels = 4096 + dtype = DtypeEnum.fp16 + act_fn = ActivationType.IDENTITY + _bias_activation_test_helper(tokens, channels, act_fn, dtype, use_bias=False) diff --git a/tests/unit/inference/kernels/core_ops/test_blas_linear.py b/tests/unit/inference/kernels/core_ops/test_blas_linear.py new file mode 100644 index 000000000000..0f9f99b4f879 --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/test_blas_linear.py @@ -0,0 +1,73 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.core_ops import BlasLibLinear +from ...inference_test_utils import allclose + +# Note: only testing with FP16 and BF16 because we use TF32 on Ampere and we don't have a good +# set of tolerances. Since this is just on top of BLAS though, the test is more about +# making sure the stride/contiguity is correct and that's data type agnostic. + + +def reference_implementation(hidden_states, weights): + return hidden_states @ weights.t() + + +problem_shapes = [ + (1, 1, 1024, 1024), + (1, 1024, 1024, 1024), + (2, 1024, 1024, 1024), + (1, 128, 768, 3072), + (1, 128, 3072, 768), + (1, 1024, 8192, 8192), + (1, 733, 8192, 32768), + (1, 13, 32768, 8192), +] + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("fp_dtype", [torch.float16, torch.bfloat16]) +@pytest.mark.parametrize("problem_shape", problem_shapes) +def test_blas_linear(fp_dtype: torch.dtype, problem_shape: Tuple[int, int, int, int]): + batch, seq_len, in_features, out_features = problem_shape + hidden_states = torch.randn(batch, seq_len, in_features, dtype=fp_dtype, + device=get_accelerator().current_device()) * 0.1 + weights = torch.randn(out_features, in_features, dtype=fp_dtype, device=get_accelerator().current_device()) * 0.01 + ds_output = torch.empty(batch, seq_len, out_features, dtype=fp_dtype, device=get_accelerator().current_device()) + + ds_kernel = BlasLibLinear(fp_dtype) + + ds_output = ds_kernel(ds_output, hidden_states, weights) + ref_output = reference_implementation(hidden_states, weights) + + assert allclose(ds_output, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("fp_dtype", [torch.float16, torch.bfloat16]) +@pytest.mark.parametrize("problem_shape", problem_shapes) +def test_blas_linear_t(fp_dtype: torch.dtype, problem_shape: Tuple[int, int, int, int]): + batch, seq_len, in_features, out_features = problem_shape + hidden_states = torch.randn(batch, seq_len, in_features, dtype=fp_dtype, + device=get_accelerator().current_device()) * 0.1 + weights = torch.randn(out_features, in_features, dtype=fp_dtype, device=get_accelerator().current_device()) * 0.01 + ds_output = torch.empty(batch, seq_len, out_features, dtype=fp_dtype, device=get_accelerator().current_device()) + + ds_kernel = BlasLibLinear(fp_dtype) + + # Transpose the weights then revert to the format we expect. + weights = weights.t().contiguous() + weights = weights.t() + ds_output = ds_kernel(ds_output, hidden_states, weights) + + ref_output = reference_implementation(hidden_states, weights) + + assert allclose(ds_output, ref_output) diff --git a/tests/unit/inference/kernels/core_ops/test_gated_activation.py b/tests/unit/inference/kernels/core_ops/test_gated_activation.py new file mode 100644 index 000000000000..ebfca4801eea --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/test_gated_activation.py @@ -0,0 +1,133 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Iterable, Optional + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.core_ops import CUDAGatedActivation +from deepspeed.inference.v2.inference_utils import ActivationType +from ...inference_test_utils import get_dtypes, allclose + + +def reference_geglu_implementation(input: torch.Tensor, + bias: Optional[torch.Tensor] = None, + act_fn: Optional[ActivationType] = ActivationType.GEGLU) -> torch.Tensor: + act_func_map = { + ActivationType.ReGLU: torch.nn.functional.relu, + ActivationType.GEGLU: lambda x: torch.nn.functional.gelu(x, approximate="tanh"), + ActivationType.SiGLU: torch.nn.functional.silu, + } + + dtype = input.dtype + input = input.to(torch.float32) + + if bias is not None: + bias = bias.to(torch.float32) + input = input + bias + + act_act = input[..., ::2] + act_linear = input[..., 1::2] + + act_act = act_func_map[act_fn](act_act) + + return (act_act * act_linear).to(dtype) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("shape", [(1372, 16384), (2, 743, 22016)]) +@pytest.mark.parametrize("dtype", get_dtypes()) +def test_dtypes(shape: Iterable[int], dtype: torch.dtype) -> None: + input_tensor = torch.randn(shape, dtype=dtype, device=get_accelerator().current_device_name()) + + # Reference output + ref_output = reference_geglu_implementation(input_tensor, act_fn=ActivationType.GEGLU) + + # Build kernel + geglu = CUDAGatedActivation(input_tensor.size(-1), input_tensor.dtype, ActivationType.GEGLU) + + # New output + output_shape = list(input_tensor.shape) + output_shape[-1] //= 2 + output_tensor = torch.empty(output_shape, dtype=input_tensor.dtype, device=get_accelerator().current_device_name()) + geglu(output_tensor, input_tensor) + + # Check + assert allclose(output_tensor, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("act_fn", [ActivationType.GEGLU, ActivationType.ReGLU, ActivationType.SiGLU]) +def test_act_fn(act_fn: ActivationType) -> None: + input_tensor = torch.randn(832, 4096, dtype=torch.float16, device=get_accelerator().current_device()) + + # Reference output + ref_output = reference_geglu_implementation(input_tensor, act_fn=act_fn) + + cuda_act = CUDAGatedActivation(4096, torch.float16, act_fn) + + # New output + output_tensor = torch.empty(832, 2048, dtype=torch.float16, device=get_accelerator().current_device()) + cuda_act(output_tensor, input_tensor) + + assert allclose(output_tensor, ref_output) + + +@pytest.mark.inference_v2_ops +def test_act_with_bias(): + input_tensor = torch.randn(832, 4096, dtype=torch.float16, device=get_accelerator().current_device()) + bias = torch.randn(4096, dtype=torch.float16, device=get_accelerator().current_device()) + + # Reference output + ref_output = reference_geglu_implementation(input_tensor, bias=bias, act_fn=ActivationType.GEGLU) + + cuda_act = CUDAGatedActivation(4096, torch.float16, ActivationType.GEGLU) + + # New output + output_tensor = torch.empty(832, 2048, dtype=torch.float16, device=get_accelerator().current_device()) + + cuda_act(output_tensor, input_tensor, bias) + + assert allclose(output_tensor, ref_output) + + +@pytest.mark.inference_v2_ops +def test_max_channels(): + input_tensor = torch.randn(832, 48152, dtype=torch.float16, device=get_accelerator().current_device()) + + ref_output = reference_geglu_implementation(input_tensor, act_fn=ActivationType.GEGLU) + + cuda_act = CUDAGatedActivation(48152, torch.float16, ActivationType.GEGLU) + + output_tensor = torch.empty(832, 24076, dtype=torch.float16, device=get_accelerator().current_device()) + cuda_act(output_tensor, input_tensor) + + assert allclose(output_tensor, ref_output) + + +@pytest.mark.inference_v2_ops +def test_bad_dtype() -> None: + with pytest.raises(ValueError): + CUDAGatedActivation(128, torch.int8, ActivationType.GEGLU) + + +@pytest.mark.inference_v2_ops +def test_bad_act_fn() -> None: + with pytest.raises(ValueError): + CUDAGatedActivation(128, torch.float16, ActivationType.RELU) + + +@pytest.mark.inference_v2_ops +def test_bad_alignment() -> None: + with pytest.raises(ValueError): + CUDAGatedActivation(127, torch.float16, ActivationType.GEGLU) + + +@pytest.mark.inference_v2_ops +def test_too_many_channels() -> None: + with pytest.raises(ValueError): + CUDAGatedActivation(49160, torch.float16, ActivationType.GEGLU) diff --git a/tests/unit/inference/kernels/core_ops/test_post_ln.py b/tests/unit/inference/kernels/core_ops/test_post_ln.py new file mode 100644 index 000000000000..8b54e5651acb --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/test_post_ln.py @@ -0,0 +1,47 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.core_ops import CUDAFPPostLN +from ...inference_test_utils import get_dtypes, allclose + + +def reference_implementation(residual: torch.Tensor, hidden_states: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor, epsilon: float) -> torch.Tensor: + residual_f = residual.to(torch.float32) + hidden_states_f = hidden_states.to(torch.float32) + gamma_f = gamma.to(torch.float32) + beta_f = beta.to(torch.float32) + return torch.nn.functional.layer_norm(residual_f + hidden_states_f, (hidden_states_f.size(-1), ), + weight=gamma_f, + bias=beta_f, + eps=epsilon).to(hidden_states.dtype) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, channels", [(1, 4096), (37, 2048), (112, 14432), (1024, 6144)]) +@pytest.mark.parametrize("dtype", get_dtypes()) +def test_cuda_post_ln(tokens: int, channels: int, dtype: torch.dtype) -> None: + + # Input vals + hidden_states = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + residual = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + gamma = torch.randn((channels), dtype=dtype, device=get_accelerator().current_device_name()) + beta = torch.rand((channels), dtype=dtype, device=get_accelerator().current_device_name()) + epsilon = 1e-5 + + # Reference output + ref_output = reference_implementation(residual, hidden_states, gamma, beta, epsilon) + + # New output + post_ln_kernel = CUDAFPPostLN(hidden_states.size(-1), residual.dtype) + ds_output = torch.empty_like(residual) + post_ln_kernel(ds_output, residual, hidden_states, gamma, beta) + + # Check + assert allclose(ds_output, ref_output) diff --git a/tests/unit/inference/kernels/core_ops/test_pre_ln.py b/tests/unit/inference/kernels/core_ops/test_pre_ln.py new file mode 100644 index 000000000000..e5ac3ae1428f --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/test_pre_ln.py @@ -0,0 +1,51 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.core_ops import CUDAFPPreLN +from ...inference_test_utils import get_dtypes, allclose + + +def reference_implementation(residual: torch.Tensor, hidden_states: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor, epsilon: float) -> torch.Tensor: + residual_f = residual.to(torch.float32) + hidden_states_f = hidden_states.to(torch.float32) + gamma_f = gamma.to(torch.float32) + beta_f = beta.to(torch.float32) + residual_out = residual_f + hidden_states_f + hidden_out = torch.nn.functional.layer_norm(residual_out, (hidden_states_f.size(-1), ), + weight=gamma_f, + bias=beta_f, + eps=epsilon) + return residual_out.to(hidden_states.dtype), hidden_out.to(hidden_states.dtype) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, channels", [(1, 4096), (37, 2048), (112, 14432), (1024, 6144)]) +@pytest.mark.parametrize("dtype", get_dtypes()) +def test_cuda_pre_ln(tokens: int, channels: int, dtype: torch.dtype) -> None: + + # Input vals + hidden_states = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + residual = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + gamma = torch.randn((channels), dtype=dtype, device=get_accelerator().current_device_name()) + beta = torch.rand((channels), dtype=dtype, device=get_accelerator().current_device_name()) + epsilon = 1e-5 + + # Reference output + ref_output_res, ref_output_hid = reference_implementation(residual, hidden_states, gamma, beta, epsilon) + + # New output + pre_ln_kernel = CUDAFPPreLN(hidden_states.size(-1), residual.dtype) + ds_output_res = torch.empty_like(residual) + ds_output_hid = torch.empty_like(hidden_states) + pre_ln_kernel(ds_output_res, ds_output_hid, residual, hidden_states, gamma, beta) + + # Check + assert allclose(ds_output_res, ref_output_res) + assert allclose(ds_output_hid, ref_output_hid) diff --git a/tests/unit/inference/kernels/core_ops/test_rms_norm.py b/tests/unit/inference/kernels/core_ops/test_rms_norm.py new file mode 100644 index 000000000000..d2893a2115b7 --- /dev/null +++ b/tests/unit/inference/kernels/core_ops/test_rms_norm.py @@ -0,0 +1,77 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import DtypeEnum +from deepspeed.inference.v2.kernels.core_ops import CUDARMSNorm, CUDARMSPreNorm +from ...inference_test_utils import get_dtypes, allclose + + +def reference_rms_norm(vals: torch.Tensor, gamma: torch.Tensor, epsilon: float = 1e-5) -> torch.Tensor: + variance = vals.to(torch.float32).pow(2).mean(-1, keepdim=True) + vals = vals * torch.rsqrt(variance + epsilon) + + if gamma.dtype in [torch.float16, torch.bfloat16]: + vals = vals.to(gamma.dtype) + + return gamma * vals + + +def reference_rms_pre_norm(vals: torch.Tensor, + residual: torch.Tensor, + gamma: torch.Tensor, + epsilon: float = 1e-5) -> torch.Tensor: + residual = residual + vals + return residual, reference_rms_norm(residual, gamma, epsilon) + + +def _rms_norm_testing_helper(rows: int, channels: int, do_residual: bool, dtype: DtypeEnum) -> None: + device = get_accelerator().current_device_name() + t_dtype = dtype.value + + vals = torch.randn((rows, channels), dtype=t_dtype, device=device) + gamma = torch.randn((channels), dtype=t_dtype, device=device) + epsilon = 1e-5 + + if do_residual: + residual_in = torch.randn((rows, channels), dtype=t_dtype, device=device) + ds_residual = residual_in.clone() + + ref_residual, ref_output = reference_rms_pre_norm(vals, residual_in, gamma, epsilon) + + kernel = CUDARMSPreNorm(channels, t_dtype, epsilon=epsilon) + ds_out = torch.empty_like(ds_residual) + + kernel(ds_residual, ds_out, residual_in, vals, gamma) + + assert allclose(ds_out, ref_output) + assert allclose(ds_residual, ref_residual) + else: + + ref_output = reference_rms_norm(vals, gamma, epsilon) + + kernel = CUDARMSNorm(channels, t_dtype, epsilon=epsilon) + ds_out = torch.empty_like(vals) + + kernel(ds_out, vals, gamma) + + assert allclose(ds_out, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("dtype", get_dtypes()) +@pytest.mark.parametrize("do_residual", [True, False]) +def test_rms_dtypes(dtype: DtypeEnum, do_residual: bool) -> None: + _rms_norm_testing_helper(883, 1024, do_residual, DtypeEnum(dtype)) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("rows, cols", [(1, 4096), (37, 2048), (112, 14432), (1024, 6144)]) +@pytest.mark.parametrize("do_residual", [True, False]) +def test_rms_shapes(rows: int, cols: int, do_residual: bool) -> None: + _rms_norm_testing_helper(rows, cols, do_residual, DtypeEnum.fp16) diff --git a/tests/unit/inference/kernels/cutlass_ops/__init__.py b/tests/unit/inference/kernels/cutlass_ops/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/kernels/cutlass_ops/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/kernels/cutlass_ops/test_moe_gemm.py b/tests/unit/inference/kernels/cutlass_ops/test_moe_gemm.py new file mode 100644 index 000000000000..4dd6d286fe00 --- /dev/null +++ b/tests/unit/inference/kernels/cutlass_ops/test_moe_gemm.py @@ -0,0 +1,113 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import ActivationType, DtypeEnum +from deepspeed.inference.v2.kernels.cutlass_ops import MoEGEMM +from ...inference_test_utils import allclose + +SINGLE_EXPERT_CASES = [(13, 2048, 2048), (256, 1024, 4096), (278, 5120, 2048), (893, 5120, 2560)] + +PYTORCH_ACT_FN_MAP = { + ActivationType.GELU: torch.nn.functional.gelu, + ActivationType.SILU: torch.nn.functional.silu, + ActivationType.RELU: torch.nn.functional.relu +} + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, in_neurons, out_neurons", SINGLE_EXPERT_CASES) +def test_single_expert(n_tokens: int, in_neurons: int, out_neurons: int) -> None: + """ + Validate that the GEMM kernel produces identical results for a single GEMM instance. + """ + device = get_accelerator().current_device() + + activations = torch.rand((n_tokens, in_neurons), device=device, dtype=torch.float16) - 0.5 + weights = torch.rand((1, in_neurons, out_neurons), device=device, dtype=torch.float16) - 0.5 + biases = torch.randn((1, out_neurons), device=device, dtype=torch.float16) + + weights_ref = weights.reshape(in_neurons, out_neurons) + biases_ref = biases.reshape(out_neurons) + ref_output = torch.matmul(activations, weights_ref) + biases_ref + + moe_gemm = MoEGEMM(DtypeEnum.fp16, ActivationType.IDENTITY) + output = torch.empty((n_tokens, out_neurons), device=device, dtype=torch.float16) + cumsum_rows = torch.tensor([n_tokens], dtype=torch.int64, device=device) + + moe_gemm(output, activations, weights, cumsum_rows, biases) + assert allclose(output, ref_output, tolerances=(1e-2, 1e-2)) + get_accelerator().synchronize() + + +def moe_test_helper(in_neurons: int, out_neurons: int, n_experts: int, max_tokens_per_expert: int, + act_fn: ActivationType, dtype: DtypeEnum) -> None: + """ + Helper function for validating the GEMM kernel for a single expert. + """ + device = get_accelerator().current_device() + + expert_allocations = torch.randint(0, max_tokens_per_expert, (n_experts, ), device=device, dtype=torch.int32) + cumsum_rows = expert_allocations.cumsum(dim=0) + print(cumsum_rows.dtype) + + activations = torch.rand((cumsum_rows[-1], in_neurons), device=device, dtype=dtype.value) - 0.5 + weights = torch.rand((n_experts, in_neurons, out_neurons), device=device, dtype=dtype.value) - 0.5 + biases = torch.randn((n_experts, out_neurons), device=device, dtype=dtype.value) + + out_ref = torch.empty((cumsum_rows[-1], out_neurons), device=device, dtype=dtype.value) + + for expert_idx in range(n_experts): + start = cumsum_rows[expert_idx - 1] if expert_idx > 0 else 0 + end = cumsum_rows[expert_idx] + activations_slice = activations[start:end] + weights_slice = weights[expert_idx] + biases_slice = biases[expert_idx] + out_ref[start:end] = torch.matmul(activations_slice, weights_slice) + biases_slice + + if act_fn != ActivationType.IDENTITY: + act_fn_fn = PYTORCH_ACT_FN_MAP[act_fn] + out_ref = act_fn_fn(out_ref) + + moe_gemm = MoEGEMM(DtypeEnum.fp16, act_fn) + output = torch.empty((cumsum_rows[-1], out_neurons), device=device, dtype=dtype.value) + + moe_gemm(output, activations, weights, cumsum_rows, biases) + + if dtype == DtypeEnum.bf16: + assert allclose(output, out_ref, tolerances=(1e-1, 1e-1)) + else: + assert allclose(output, out_ref, tolerances=(1e-2, 1e-2)) + get_accelerator().synchronize() + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("max_tokens_per_expert", [1, 4, 16, 64, 128]) +def test_multi_expert(max_tokens_per_expert: int) -> None: + """ + Validate for multi-expert GEMM instances that the output is identical to the reference. + """ + moe_test_helper(5120, 2048, 64, max_tokens_per_expert, ActivationType.IDENTITY, DtypeEnum.fp16) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("act_fn", [ActivationType.GELU, ActivationType.SILU, ActivationType.RELU]) +def test_act_fns(act_fn: ActivationType) -> None: + """ + Validate activation function behavior. + """ + moe_test_helper(5120, 2048, 64, 32, act_fn, DtypeEnum.fp16) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("dtype", [DtypeEnum.fp16, DtypeEnum.bf16]) +def test_dtypes(dtype: DtypeEnum) -> None: + """ + Validate data type behavior. + """ + moe_test_helper(5120, 2048, 64, 32, ActivationType.IDENTITY, dtype) diff --git a/tests/unit/inference/kernels/ragged_ops/__init__.py b/tests/unit/inference/kernels/ragged_ops/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/kernels/ragged_ops/ragged_testing_utils.py b/tests/unit/inference/kernels/ragged_ops/ragged_testing_utils.py new file mode 100644 index 000000000000..445c6c38b87f --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/ragged_testing_utils.py @@ -0,0 +1,300 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import random +from typing import List, Optional, Tuple + +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.ragged import ( + AllocationMode, + DSSequenceDescriptor, + DSStateManager, + DSStateManagerConfig, + KVCacheConfig, + MemoryConfig, + PlaceholderSequenceDescriptor, + RaggedBatchWrapper, +) +from ...inference_test_utils import allclose + + +def build_simple_batch(seq_lens: List[int], + vocab_range: Optional[int] = 100, + padding: Optional[bool] = False) -> RaggedBatchWrapper: + """ + Construct a simple batch with the given sequence lengths. This method should not + be used for for testing scenarios that require information about KV or sequence + history. + """ + total_tokens = max(sum(seq_lens), 1024) + n_seqs = max(len(seq_lens), 128) + + config = DSStateManagerConfig(max_tracked_sequences=n_seqs, + max_ragged_sequence_count=n_seqs, + max_ragged_batch_size=total_tokens) + batch = RaggedBatchWrapper(config) + + batch.clear() + + for seq_len in seq_lens: + seq_desc = PlaceholderSequenceDescriptor() + tokens = torch.randint(0, vocab_range, (seq_len, )) + batch.insert_sequence(seq_desc, tokens) + + batch.finalize(padding=padding) + + return batch + + +def build_complex_batch(seq_params: List[Tuple[int, int, int]], + kv_block_size: int, + vocab_range: Optional[int] = 100, + padding: Optional[bool] = False) -> Tuple[RaggedBatchWrapper, int]: + """ + Construct a fully paramtrized batch with the given sequence lengths. This method + can be used to construct more realistic inputs for testing scenarios that will interact + with all the members of the RaggedBatchWrapper. + """ + seq_lens = [seq_param[0] for seq_param in seq_params] + total_tokens = max(sum(seq_lens), 1024) + n_seqs = max(len(seq_lens), 128) + + config = DSStateManagerConfig(max_tracked_sequences=n_seqs, + max_ragged_sequence_count=n_seqs, + max_ragged_batch_size=total_tokens) + batch = RaggedBatchWrapper(config) + + batch.clear() + + total_kv_blocks = 0 + + for seq_len, n_seen_tokens, kv_ptr in seq_params: + n_kv_blocks = (seq_len + n_seen_tokens + kv_block_size - 1) // kv_block_size + seq_desc = PlaceholderSequenceDescriptor(seen_tokens=n_seen_tokens, + cur_allocated_blocks=n_kv_blocks, + kv_blocks_ptr=kv_ptr) + tokens = torch.randint(0, vocab_range, (seq_len, )) + batch.insert_sequence(seq_desc, tokens) + total_kv_blocks += n_kv_blocks + + batch.finalize(padding=padding) + + return batch, total_kv_blocks + + +def build_batch_and_manager( + seq_params: List[Tuple[int, int]], + head_size: int, + n_heads_kv: int, + kv_block_size: int, + vocab_range: Optional[int] = 100, + padding: Optional[bool] = False, + kv_fill: Optional[List[torch.Tensor]] = None +) -> Tuple[RaggedBatchWrapper, DSStateManager, List[DSSequenceDescriptor]]: + """ + Will construct and populate a batch and KVCache with the given sequence parameters. + + Arguments: + seq_params (List[Tuple[int, int]]): A list of tuples containing the sequence length and + the number of tokens that have already been seen for that sequence. + head_size (int): The size of each attention head. + n_heads_kv (int): The number of attention heads for the KV-cache. + kv_block_size (int): The size of each block in the KV-cache. + vocab_range (Optional[int]): The range of the vocabulary. Defaults to 100. + padding (Optional[bool]): Whether to pad the batch. Defaults to False. + kv_fill (Optional[List[torch.Tensor]]): A list of tensors to use to populate the KV-cache. + If this is not provided, the KV-cache will be treated as empty and the contents should + not be relied upon. NOTE(cmikeh2): This functionality relies on the functionality + of LinearBlockedKVCopy. If tests relying on this feature are failing, make sure that + LinearBlockedKVCopy is working correctly. + """ + seq_lens = [seq_param[0] for seq_param in seq_params] + fill_lens = [seq_param[1] for seq_param in seq_params] + max_created_batch_len = max(sum(seq_lens), sum(fill_lens)) + total_tokens = max(max_created_batch_len, 1024) + n_seqs = max(len(seq_lens), 128) + + req_kv_blocks = [None] * n_seqs + total_kv_blocks = 0 + for i, (seq_len, n_seen_tokens) in enumerate(seq_params): + req_kv_blocks[i] = (seq_len + n_seen_tokens + kv_block_size - 1) // kv_block_size + total_kv_blocks += req_kv_blocks[i] + + kv_config = KVCacheConfig(block_size=kv_block_size, + num_allocation_groups=1, + cache_shape=(1, n_heads_kv, head_size)) + memory_config = MemoryConfig(mode=AllocationMode.ALLOCATE, size=total_kv_blocks) + + config = DSStateManagerConfig(max_tracked_sequences=n_seqs, + max_ragged_sequence_count=n_seqs, + max_ragged_batch_size=total_tokens, + memory_config=memory_config) + + batch = RaggedBatchWrapper(config) + state_manager = DSStateManager(config, kv_config) + + # At the beginning of operation, the design of the allocator is such that it will return + # linear blocks of memory. The following will "warm up" the allocator so that we can be + # more certain that code is not dependent on this behavior. + all_allocs = [] + for _ in range(20): + decision = random.randint(0, 1) + + if decision == 0: + blocks_to_allocate = random.randint(0, total_kv_blocks) + if blocks_to_allocate <= state_manager.free_blocks and blocks_to_allocate > 0: + all_allocs.append(state_manager.allocate_blocks(blocks_to_allocate)) + else: + if len(all_allocs) > 0: + idx = random.randint(0, len(all_allocs) - 1) + state_manager._kv_cache.free(all_allocs[idx]) + + del all_allocs[idx] + + for alloc in all_allocs: + state_manager._kv_cache.free(alloc) + + assert state_manager.free_blocks == total_kv_blocks + + batch.clear() + seq_descs = [] + + if kv_fill is None or sum(fill_lens) == 0: + for i, (seq_len, n_seen_tokens) in enumerate(seq_params): + # Create empty descriptor + seq_desc = state_manager.get_or_create_sequence(i) + + # Update `seen_tokens` in the descriptor + seq_desc.pre_forward(n_seen_tokens) + seq_desc.post_forward() + + # Ensure there's enough KV-cache for the sequence + kv_block_ids = state_manager.allocate_blocks(req_kv_blocks[i]) + print(f"Allocated {req_kv_blocks[i]} blocks for sequence {i}: {kv_block_ids}") + seq_desc.extend_kv_cache(kv_block_ids) + + # Insert sequence into batch + tokens = torch.randint(0, vocab_range, (seq_len, )) + batch.insert_sequence(seq_desc, tokens) + seq_desc.pre_forward(seq_len) + seq_descs.append(seq_desc) + else: + qkv = torch.empty((total_tokens, (n_heads_kv * 3) * head_size), + dtype=torch.float16, + device=get_accelerator().current_device()) + fills_as_tensor = torch.tensor(fill_lens, dtype=torch.int32) + fill_cumsum = torch.cat((torch.tensor([0], dtype=torch.int32), torch.cumsum(fills_as_tensor, dim=0))) + + for i, (_, n_seen_tokens) in enumerate(seq_params): + # Create empty descriptor + seq_desc = state_manager.get_or_create_sequence(i) + + # Update `seen_tokens` in the descriptor + if n_seen_tokens > 0: + dummy_fill_toks = torch.randint(0, vocab_range, (n_seen_tokens, )) + batch.insert_sequence(seq_desc, dummy_fill_toks) + seq_desc.pre_forward(n_seen_tokens) + + # Ensure there's enough KV-cache for the sequence + kv_block_ids = state_manager.allocate_blocks(req_kv_blocks[i]) + print(f"Allocated {req_kv_blocks[i]} blocks for sequence {i}: {kv_block_ids}") + seq_desc.extend_kv_cache(kv_block_ids) + seq_descs.append(seq_desc) + + if n_seen_tokens == 0: + continue + + assert kv_fill[i].shape[0] == n_seen_tokens + assert kv_fill[i].shape[1] == n_heads_kv * head_size * 2 + + local_q = torch.randn((n_seen_tokens, n_heads_kv * head_size), dtype=torch.float16, device=qkv.device) + local_qkv = torch.cat((local_q, kv_fill[i]), dim=1) + qkv[fill_cumsum[i]:fill_cumsum[i + 1]] = local_qkv + + batch.finalize(padding=padding) + + from deepspeed.inference.v2.kernels.ragged_ops import LinearBlockedKVCopy + kv_copy = LinearBlockedKVCopy(head_size, n_heads_kv, n_heads_kv, torch.float16) + kv_cache = state_manager.get_cache(0) + kv_copy(kv_cache, qkv, batch) + + for seq_desc in seq_descs: + if seq_desc.in_flight_tokens > 0: + seq_desc.post_forward() + + batch.clear() + + for i, (seq_len, _) in enumerate(seq_params): + seq_desc = state_manager.get_or_create_sequence(i) + tokens = torch.randint(0, vocab_range, (seq_len, )) + batch.insert_sequence(seq_desc, tokens) + seq_desc.pre_forward(seq_len) + + # We will skip KV cache allocation here because we did a lump allocation above + # for both the fill and the sequence itself. + + batch.finalize(padding=padding) + + return batch, state_manager, seq_descs + + +def validate_kv_cache(kv_cache: torch.Tensor, + k: torch.Tensor, + v: torch.Tensor, + seq_descs: List[DSSequenceDescriptor], + batch: RaggedBatchWrapper, + exact: bool = True) -> None: + """ + Given a QKV tensor and a KV cache, validate that the cache contains the correct values. + """ + block_size = kv_cache.shape[1] + n_kv_heads = kv_cache.shape[3] + head_size = kv_cache.shape[4] + + inflight_descs = batch.inflight_seq_descriptors(on_device=False)[:batch.current_sequences] + + if inflight_descs.shape[0] != len(seq_descs): + raise ValueError("The number of sequence descriptors does not match the number of sequences in the batch.") + + for seq_desc, inflight_seq in zip(seq_descs, inflight_descs): + start_idx = inflight_seq[0] + assigned_kv_blocks = seq_desc.kv_cache_ids(on_device=False) + + real_k_values = k[start_idx:start_idx + seq_desc.in_flight_tokens] + real_v_values = v[start_idx:start_idx + seq_desc.in_flight_tokens] + + start_block_idx = seq_desc.seen_tokens // block_size + local_start_idx = 0 + cur_start_idx = seq_desc.seen_tokens + + for block_idx in range(start_block_idx, seq_desc.cur_allocated_blocks): + block = kv_cache[assigned_kv_blocks[0, block_idx].item()] + block_start_idx = cur_start_idx % block_size + n_tokens_to_check = min(block_size - block_start_idx, seq_desc.in_flight_tokens - local_start_idx) + block_end_idx = block_start_idx + n_tokens_to_check + + if exact: + assert torch.equal( + block[block_start_idx:block_end_idx, 0, :, :], + real_k_values[local_start_idx:local_start_idx + n_tokens_to_check].reshape( + n_tokens_to_check, n_kv_heads, head_size)) + assert torch.equal( + block[block_start_idx:block_end_idx, 1, :, :], + real_v_values[local_start_idx:local_start_idx + n_tokens_to_check].reshape( + n_tokens_to_check, n_kv_heads, head_size)) + else: + assert allclose( + block[block_start_idx:block_end_idx, 0, :, :], + real_k_values[local_start_idx:local_start_idx + n_tokens_to_check].reshape( + n_tokens_to_check, n_kv_heads, head_size)) + assert allclose( + block[block_start_idx:block_end_idx, 1, :, :], + real_v_values[local_start_idx:local_start_idx + n_tokens_to_check].reshape( + n_tokens_to_check, n_kv_heads, head_size)) + + local_start_idx += n_tokens_to_check + cur_start_idx += n_tokens_to_check diff --git a/tests/unit/inference/kernels/ragged_ops/test_atom_builder.py b/tests/unit/inference/kernels/ragged_ops/test_atom_builder.py new file mode 100644 index 000000000000..a33c938a0608 --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_atom_builder.py @@ -0,0 +1,45 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.inference.v2.kernels.ragged_ops import AtomBuilder +from .ragged_testing_utils import build_complex_batch + +Q_BLOCK_SIZE = 128 +KV_BLOCK_SIZE = 128 + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('seq_params', [(1, 0, 0), (1, 228, 0), (383, 0, 0), (1, 494, 0)]) +def test_single_sequence(seq_params) -> None: + seq_len, n_seen_tokens, _ = seq_params + + batch, _ = build_complex_batch([seq_params], kv_block_size=KV_BLOCK_SIZE, padding=False) + atom_builder = AtomBuilder() + + atoms = torch.empty((8, 8), dtype=torch.int32, device=torch.device("cpu")) + atoms, n_atoms = atom_builder(atoms, batch, Q_BLOCK_SIZE, KV_BLOCK_SIZE) + + calc_n_atoms = (seq_len + 127) // 128 + + assert n_atoms == calc_n_atoms + + for i, atom in enumerate(atoms[:n_atoms]): + # Since the ptr was 0, first 2 elements should be 0 + assert atom[0] == 0 + assert atom[1] == 0 + + # Since we have a single sequence, the q_start_idx should always be + # whichever atom we're on multiplied by the block size + assert atom[2] == i * Q_BLOCK_SIZE + assert atom[3] == min(Q_BLOCK_SIZE, seq_len - i * Q_BLOCK_SIZE) + total_toks = i * Q_BLOCK_SIZE + min(Q_BLOCK_SIZE, seq_len - i * Q_BLOCK_SIZE) + + assert atom[4] == (total_toks + n_seen_tokens + KV_BLOCK_SIZE - 1) // KV_BLOCK_SIZE + assert atom[5] == (total_toks + n_seen_tokens) + + assert atom[6] == n_seen_tokens + i * Q_BLOCK_SIZE diff --git a/tests/unit/inference/kernels/ragged_ops/test_blocked_flash.py b/tests/unit/inference/kernels/ragged_ops/test_blocked_flash.py new file mode 100644 index 000000000000..a16a7775e964 --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_blocked_flash.py @@ -0,0 +1,197 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import itertools + +from typing import List, Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import DtypeEnum +from deepspeed.inference.v2.kernels.ragged_ops import ( + AtomBuilder, + BlockedFlashAttn, + get_q_block_size, + get_kv_block_size, + LinearBlockedKVCopy, +) +from deepspeed.inference.v2.ragged import split_kv +from deepspeed.ops.op_builder import RaggedUtilsBuilder + +from .ragged_testing_utils import build_batch_and_manager +from ...inference_test_utils import allclose + +try: + from flash_attn.flash_attn_interface import flash_attn_varlen_func + validate_accuracy = True +except ImportError: + validate_accuracy = False +""" +NOTE(cmikeh2): These tests depend on atom construction and KV-cache copying to behave correctly. +If one or the other of those is not working, then these tests will fail. Before debugging here, +make sure that the atom construction and KV-cache copying tests are passing. +""" + + +def _blocked_flash_testing_helper(head_size: int, n_heads_q: int, n_heads_kv: int, + seq_params: List[Tuple[int, int]]) -> None: + """ + Helper function for testing blocked flash attention. Used to enable parametrize to only set up + a subset of parameters before being passed to the unified test function. + """ + q_block_size = get_q_block_size(head_size) + kv_block_size = get_kv_block_size(head_size) + + kvs = [] + for _, history_len in seq_params: + if history_len > 0: + kvs.append( + torch.randn((history_len, 2 * n_heads_kv * head_size), + device=get_accelerator().current_device(), + dtype=torch.float16)) + else: + kvs.append(None) + + batch, state_manager, _ = build_batch_and_manager(seq_params, head_size, n_heads_kv, kv_block_size, kv_fill=kvs) + + atom_builder = AtomBuilder() + kv_copy = LinearBlockedKVCopy(head_size, n_heads_q, n_heads_kv, DtypeEnum.fp16) + atom_flash = BlockedFlashAttn(head_size, DtypeEnum.fp16) + + total_atoms = sum((seq[0] + q_block_size - 1) // q_block_size for seq in seq_params) + atoms = torch.empty((total_atoms, 8), dtype=torch.int32, device=get_accelerator().current_device()) + alloc_func = RaggedUtilsBuilder().load().allocate_fast_host_buffer + atoms_host = alloc_func(atoms) + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=get_accelerator().current_device(), + dtype=torch.float16) + + atoms_host, n_atoms = atom_builder(atoms_host, batch, q_block_size, kv_block_size) + atoms.copy_(atoms_host[:n_atoms]) + + kv_cache = state_manager.get_cache(0) + kv_copy(kv_cache, qkv, batch) + + out = torch.empty((batch.current_tokens, head_size * n_heads_q), + device=get_accelerator().current_device(), + dtype=torch.float16) + k_cache, v_cache = split_kv(kv_cache) + q = qkv[:, :head_size * n_heads_q] + + atom_flash(out, q, k_cache, v_cache, atoms, 1.0) + + if validate_accuracy: + cu_seqlens_q = torch.tensor([0] + list(itertools.accumulate([seq[0] for seq in seq_params])), + dtype=torch.int32, + device=get_accelerator().current_device()) + cu_seqlens_kv = torch.tensor([0] + list(itertools.accumulate([seq[1] + seq[0] for seq in seq_params])), + dtype=torch.int32, + device=get_accelerator().current_device()) + + inflight_kv = qkv[:, head_size * n_heads_q:] + full_kvs = [] + for i, kv in enumerate(kvs): + if kv is not None: + full_kvs.append(torch.cat([kv, inflight_kv[cu_seqlens_q[i]:cu_seqlens_q[i + 1]]], dim=0)) + else: + full_kvs.append(inflight_kv[cu_seqlens_q[i]:cu_seqlens_q[i + 1]]) + run_kvs = torch.cat(full_kvs, dim=0) + k = run_kvs[:, :head_size * n_heads_kv] + v = run_kvs[:, head_size * n_heads_kv:] + + q_ref = q.reshape((batch.current_tokens, n_heads_q, head_size)) + k_ref = k.reshape((k.shape[0], n_heads_kv, head_size)) + v_ref = v.reshape((v.shape[0], n_heads_kv, head_size)) + + max_seqlen_q = max([seq[0] for seq in seq_params]) + max_seqlen_kv = max([seq[1] + seq[0] for seq in seq_params]) + + ref_o = flash_attn_varlen_func(q_ref, + k_ref, + v_ref, + cu_seqlens_q, + cu_seqlens_kv, + max_seqlen_q, + max_seqlen_kv, + softmax_scale=1.0, + causal=True) + + ref_o = ref_o.reshape(batch.current_tokens, head_size * n_heads_q) + + assert allclose(out, ref_o) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens", [2, 33, 65, 128, 256, 2037]) +def test_single_prompt(n_tokens: int) -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(n_tokens, 0)] + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("prompt_lengths", [(128, 128), (192, 38), (514, 713), (83, 312, 610)]) +def test_multiple_prompts(prompt_lengths: Tuple[int, int]) -> None: + """ + Test multiple prompts in a single batch. + """ + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(prompt_lengths[i], 0) for i in range(len(prompt_lengths))] + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("seq_params", [(1, 34), (43, 40), (1, 144), (64, 128), (332, 628)]) +def test_continuation(seq_params: Tuple[int, int]) -> None: + """ + Test continued generation/prompt processing. + """ + head_size = 64 + n_heads_q = 32 + n_heads_kv = 32 + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, [seq_params]) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("head_size", [64, 128]) +def test_head_size(head_size: int) -> None: + n_heads_q = 16 + n_heads_kv = 16 + seq_params = [(128, 128), (192, 38), (1, 814)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("head_config", [(32, 8), (64, 16), (40, 8)]) +def test_gqa(head_config: Tuple[int, int]) -> None: + head_size = 128 + n_heads_q = head_config[0] + n_heads_kv = head_config[1] + + seq_params = [(128, 128), (192, 38), (1, 814)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +def test_fully_composed() -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(332, 628), (1, 718), (1, 323), (180, 5), (224, 0)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) diff --git a/tests/unit/inference/kernels/ragged_ops/test_blocked_kv_copy.py b/tests/unit/inference/kernels/ragged_ops/test_blocked_kv_copy.py new file mode 100644 index 000000000000..90fe26eb4490 --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_blocked_kv_copy.py @@ -0,0 +1,112 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.ragged_ops import LinearBlockedKVCopy +from .ragged_testing_utils import build_batch_and_manager, validate_kv_cache + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, history_size", [(1, 0), (17, 0), (33, 8), (63, 1)]) +def test_single_sequence_single_block(n_tokens: int, history_size: int): + """ + Validate that the copy works correctly + """ + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + kv_block_size = 64 + device = get_accelerator().current_device() + + batch, state_manager, seq_descs = build_batch_and_manager([(n_tokens, history_size)], head_size, n_heads_kv, + kv_block_size) + + assert batch.current_sequences == 1 + assert batch.current_tokens == n_tokens + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=device, + dtype=torch.float16) + + kv_cache = state_manager.get_cache(0) + + copy_impl = LinearBlockedKVCopy(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch) + + k = qkv[:, head_size * n_heads_q:head_size * (n_heads_q + n_heads_kv)] + v = qkv[:, head_size * (n_heads_q + n_heads_kv):] + + validate_kv_cache(kv_cache, k, v, seq_descs, batch) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, history_size", [(128, 0), (177, 0), (169, 8), (117, 88)]) +def test_single_sequence_multiple_blocks(n_tokens: int, history_size: int): + """ + Validate that the copy works correctly + """ + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + kv_block_size = 64 + device = get_accelerator().current_device() + + batch, state_manager, seq_descs = build_batch_and_manager([(n_tokens, history_size)], head_size, n_heads_kv, + kv_block_size) + + assert batch.current_sequences == 1 + assert batch.current_tokens == n_tokens + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=device, + dtype=torch.float16) + + kv_cache = state_manager.get_cache(0) + + copy_impl = LinearBlockedKVCopy(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch) + + k = qkv[:, head_size * n_heads_q:head_size * (n_heads_q + n_heads_kv)] + v = qkv[:, head_size * (n_heads_q + n_heads_kv):] + + validate_kv_cache(kv_cache, k, v, seq_descs, batch) + + +@pytest.mark.inference_v2_ops +def test_multi_sequence() -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + kv_block_size = 64 + device = get_accelerator().current_device() + + batch_config = [ + (128, 0), + (177, 0), + (169, 8), + (117, 88), + (1, 293), + (1, 733), + (1, 33), + ] + + batch, state_manager, seq_descs = build_batch_and_manager(batch_config, head_size, n_heads_kv, kv_block_size) + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=device, + dtype=torch.float16) + + kv_cache = state_manager.get_cache(0) + + copy_impl = LinearBlockedKVCopy(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch) + + k = qkv[:, head_size * n_heads_q:head_size * (n_heads_q + n_heads_kv)] + v = qkv[:, head_size * (n_heads_q + n_heads_kv):] + + validate_kv_cache(kv_cache, k, v, seq_descs, batch) diff --git a/tests/unit/inference/kernels/ragged_ops/test_blocked_rotary_emb.py b/tests/unit/inference/kernels/ragged_ops/test_blocked_rotary_emb.py new file mode 100644 index 000000000000..35b92ef86305 --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_blocked_rotary_emb.py @@ -0,0 +1,203 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import List, Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.ragged_ops import BlockedRotaryEmbeddings, BlockedTrainedRotaryEmbeddings +from deepspeed.inference.v2.ragged import RaggedBatchWrapper, DSSequenceDescriptor +from .ragged_testing_utils import build_batch_and_manager, validate_kv_cache +from ...inference_test_utils import allclose +""" +NOTE(cmikeh2): It is very possible to see unit test failures (even on FP16) depending on when +certain values are casted up to or down from float32. If we are seeing accuracy issues, we should +make sure we are aligning on the training implementation's cast pattern here, given these tolerances +tend to be sufficient elsewhere. +""" + + +def rotary_pos_embs(q: torch.Tensor, k: torch.Tensor, seq_descs: List[DSSequenceDescriptor], batch: RaggedBatchWrapper, + head_size: int): + + def make_cos_sin_emb(seq_len: int) -> Tuple[torch.Tensor, torch.Tensor]: + t = torch.arange(seq_len, dtype=torch.float32, device=get_accelerator().current_device()) + inv_freq = (1.0 / (10000.0**(torch.arange( + 0, head_size, 2, dtype=torch.float32, device=get_accelerator().current_device()) / head_size))).half() + + freqs = torch.einsum("i,j->ij", t, inv_freq) + emb = torch.cat((freqs, freqs), dim=-1) + + return torch.cos(emb)[:, None, :], torch.sin(emb)[:, None, :], inv_freq + + def rotate_half(x: torch.Tensor) -> torch.Tensor: + return torch.cat((-x[..., x.shape[-1] // 2:], x[..., :x.shape[-1] // 2]), dim=-1) + + cos, sin, freqs = make_cos_sin_emb(1024) + + q_out = torch.empty_like(q) + k_out = torch.empty_like(k) + n_heads_q = q.shape[1] // head_size + n_heads_kv = k.shape[1] // head_size + + inflight_descs = batch.inflight_seq_descriptors(on_device=False)[:batch.current_sequences] + + if inflight_descs.shape[0] != len(seq_descs): + raise ValueError("The number of sequence descriptors does not match the number of sequences in the batch.") + + for seq_desc, inflight_seq in zip(seq_descs, inflight_descs): + start_idx = inflight_seq[0] + n_tokens = seq_desc.in_flight_tokens + + q_src = q[start_idx:start_idx + n_tokens].reshape(n_tokens, n_heads_q, head_size).float() + k_src = k[start_idx:start_idx + n_tokens].reshape(n_tokens, n_heads_kv, head_size).float() + freq_start_offset = seq_desc.seen_tokens + + cos_chunk = cos[range(freq_start_offset, freq_start_offset + n_tokens)] + sin_chunk = sin[range(freq_start_offset, freq_start_offset + n_tokens)] + + q_emb = q_src * cos_chunk + rotate_half(q_src) * sin_chunk + k_emb = k_src * cos_chunk + rotate_half(k_src) * sin_chunk + + q_out[start_idx:start_idx + n_tokens] = q_emb.reshape(n_tokens, n_heads_q * head_size).to(q_out.dtype) + k_out[start_idx:start_idx + n_tokens] = k_emb.reshape(n_tokens, n_heads_kv * head_size).to(k_out.dtype) + + return q_out, k_out, freqs + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, history_size", [(1, 0), (17, 0), (33, 15), (1, 63)]) +@pytest.mark.parametrize("trained_emb", [False, True]) +def test_single_sequence_single_block(n_tokens: int, history_size: int, trained_emb: bool): + """ + Validate that the copy works correctly + """ + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + kv_block_size = 64 + device = get_accelerator().current_device() + + batch, state_manager, seq_descs = build_batch_and_manager([(n_tokens, history_size)], head_size, n_heads_kv, + kv_block_size) + + assert batch.current_sequences == 1 + assert batch.current_tokens == n_tokens + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=device, + dtype=torch.float16) + qkv_ref = qkv.clone() + + q = qkv_ref[:, :head_size * n_heads_q] + k = qkv_ref[:, head_size * n_heads_q:head_size * (n_heads_q + n_heads_kv)] + v = qkv_ref[:, head_size * (n_heads_q + n_heads_kv):] + + q_ref, k, freqs = rotary_pos_embs(q, k, seq_descs, batch, head_size) + freqs = freqs.half() + + kv_cache = state_manager.get_cache(0) + + if trained_emb: + copy_impl = BlockedTrainedRotaryEmbeddings(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch, freqs) + else: + copy_impl = BlockedRotaryEmbeddings(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch) + + assert allclose(qkv[:, :head_size * n_heads_q], q_ref) + validate_kv_cache(kv_cache, k, v, seq_descs, batch, exact=False) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, history_size", [(128, 0), (177, 0), (169, 8), (117, 88)]) +@pytest.mark.parametrize("trained_emb", [False, True]) +def test_single_sequence_multiple_blocks(n_tokens: int, history_size: int, trained_emb: bool): + """ + Validate that the copy works correctly + """ + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + kv_block_size = 64 + device = get_accelerator().current_device() + + batch, state_manager, seq_descs = build_batch_and_manager([(n_tokens, history_size)], head_size, n_heads_kv, + kv_block_size) + + assert batch.current_sequences == 1 + assert batch.current_tokens == n_tokens + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=device, + dtype=torch.float16) + qkv_ref = qkv.clone() + + q = qkv_ref[:, :head_size * n_heads_q] + k = qkv_ref[:, head_size * n_heads_q:head_size * (n_heads_q + n_heads_kv)] + v = qkv_ref[:, head_size * (n_heads_q + n_heads_kv):] + + q_ref, k, freqs = rotary_pos_embs(q, k, seq_descs, batch, head_size) + freqs = freqs.half() + + kv_cache = state_manager.get_cache(0) + + if trained_emb: + copy_impl = BlockedTrainedRotaryEmbeddings(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch, freqs) + else: + copy_impl = BlockedRotaryEmbeddings(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch) + + assert allclose(qkv[:, :head_size * n_heads_q], q_ref) + validate_kv_cache(kv_cache, k, v, seq_descs, batch, exact=False) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("trained_emb", [False, True]) +def test_multi_sequences(trained_emb: bool) -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + kv_block_size = 64 + device = get_accelerator().current_device() + + batch_config = [ + (128, 0), + (177, 0), + (169, 8), + (117, 88), + (1, 293), + (1, 733), + (1, 33), + ] + + batch, state_manager, seq_descs = build_batch_and_manager(batch_config, head_size, n_heads_kv, kv_block_size) + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=device, + dtype=torch.float16) + qkv_ref = qkv.clone() + + q = qkv_ref[:, :head_size * n_heads_q] + k = qkv_ref[:, head_size * n_heads_q:head_size * (n_heads_q + n_heads_kv)] + v = qkv_ref[:, head_size * (n_heads_q + n_heads_kv):] + + q_ref, k, freqs = rotary_pos_embs(q, k, seq_descs, batch, head_size) + freqs = freqs.half() + + kv_cache = state_manager.get_cache(0) + + if trained_emb: + copy_impl = BlockedTrainedRotaryEmbeddings(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch, freqs) + else: + copy_impl = BlockedRotaryEmbeddings(head_size, n_heads_q, n_heads_kv, torch.float16) + copy_impl(kv_cache, qkv, batch) + + assert allclose(qkv[:, :head_size * n_heads_q], q_ref) + validate_kv_cache(kv_cache, k, v, seq_descs, batch, exact=False) diff --git a/tests/unit/inference/kernels/ragged_ops/test_logits_gather.py b/tests/unit/inference/kernels/ragged_ops/test_logits_gather.py new file mode 100644 index 000000000000..0208b733ab5b --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_logits_gather.py @@ -0,0 +1,96 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import List + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.ragged_ops import RaggedLogitsGather +from ...inference_test_utils import allclose, get_dtypes +from .ragged_testing_utils import build_simple_batch + + +def baseline_implementation(hidden_states: torch.Tensor, seq_lens: List[int]) -> torch.Tensor: + output = torch.empty((len(seq_lens), hidden_states.shape[1]), + dtype=hidden_states.dtype, + device=hidden_states.device) + + offset = 0 + for i, seq_len in enumerate(seq_lens): + output[i] = hidden_states[offset + seq_len - 1] + offset += seq_len + + return output + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('dtype', get_dtypes()) +def test_supported_dtypes(dtype: torch.dtype) -> None: + """ + Validate support on nominally supported data types. + """ + model_dim = 4096 + + batch = build_simple_batch([256], padding=False) + hidden_states = torch.randn((batch.current_tokens, model_dim), + dtype=dtype, + device=get_accelerator().current_device()) + + reference_result = baseline_implementation(hidden_states, [256]) + + kernel = RaggedLogitsGather(model_dim, dtype) + output = torch.empty_like(reference_result) + kernel(output, hidden_states, batch) + + assert allclose(output, reference_result) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('seq_lens', [[128, 64, 192, 32], [57, 112, 63, 89, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], + [63, 27, 74, 83, 32, 17, 1, 1, 1, 1, 1]]) +def test_multiple_sequences(seq_lens: List[int]) -> None: + """ + Validate support on more multi-sequence inputs. + """ + model_dim = 4096 + dtype = torch.float16 + + batch = build_simple_batch(seq_lens, padding=False) + hidden_states = torch.randn((batch.current_tokens, model_dim), + dtype=dtype, + device=get_accelerator().current_device()) + + reference_result = baseline_implementation(hidden_states, seq_lens) + + kernel = RaggedLogitsGather(model_dim, dtype) + output = torch.empty_like(reference_result) + kernel(output, hidden_states, batch) + + assert allclose(output, reference_result) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("model_dim", [1024, 6144, 6784]) +def test_problem_size_permutations(model_dim: int) -> None: + """ + Validate for different embedding sizes. + """ + dtype = torch.float16 + seq_lens = [128, 64, 192, 32] + + batch = build_simple_batch(seq_lens, padding=False) + hidden_states = torch.randn((batch.current_tokens, model_dim), + dtype=dtype, + device=get_accelerator().current_device()) + + reference_result = baseline_implementation(hidden_states, seq_lens) + + kernel = RaggedLogitsGather(model_dim, dtype) + output = torch.empty_like(reference_result) + kernel(output, hidden_states, batch) + + assert allclose(output, reference_result) diff --git a/tests/unit/inference/kernels/ragged_ops/test_moe_gather.py b/tests/unit/inference/kernels/ragged_ops/test_moe_gather.py new file mode 100644 index 000000000000..5fa375b49c19 --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_moe_gather.py @@ -0,0 +1,83 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import DtypeEnum +from deepspeed.inference.v2.kernels.ragged_ops import ( + MoEGather, + MoEScatter, + RaggedTop1Gating, +) +from .ragged_testing_utils import build_simple_batch +""" +For simplicity's sake, these tests do rely on ``RaggedTop1Gating`` and +``MoEScatter`` to produce correct inputs. If either of these kernels is broken +these tests will fail, so double check the unit test results there before +debugging here. +""" + + +def build_inputs(n_tokens, n_experts, do_padding): + + assert n_tokens <= 2048, "This test will break if n_tokens > 2048" + + # Sequence composition shouldn't matter here + batch = build_simple_batch([n_tokens], padding=do_padding) + + logits = torch.randn((batch.tensor_toks, n_experts), + dtype=torch.float16, + device=get_accelerator().current_device()) + + # This will make each token's value equal to its index. NOTE: This will break for + # tokens with index > 2048. + hidden_states = torch.arange(batch.tensor_toks, dtype=torch.float16, + device=get_accelerator().current_device()).repeat_interleave(4096, dim=0).reshape( + batch.tensor_toks, 4096).contiguous() + + gate = RaggedTop1Gating(DtypeEnum.fp16) + + # Gating outputs + expert_counts = torch.zeros((n_experts, ), dtype=torch.int32, device=get_accelerator().current_device()) + scores = torch.empty((batch.tensor_toks, ), dtype=torch.float32, device=get_accelerator().current_device()) + expert_assignment = torch.empty((batch.tensor_toks, ), + dtype=torch.int32, + device=get_accelerator().current_device()) + expert_offset = torch.empty((batch.tensor_toks, ), dtype=torch.int32, device=get_accelerator().current_device()) + + gate(expert_counts, scores, expert_assignment, expert_offset, logits, batch) + + # Scatter outputs + moe_input = torch.empty((batch.tensor_toks, 4096), dtype=torch.float16, device=get_accelerator().current_device()) + expert_cumsum = torch.empty((n_experts, ), dtype=torch.int64, device=get_accelerator().current_device()) + mapped_slots = torch.empty((batch.tensor_toks, ), dtype=torch.int32, device=get_accelerator().current_device()) + + scatter = MoEScatter(DtypeEnum.fp16, 4096) + scatter(moe_input, expert_cumsum, mapped_slots, hidden_states, expert_counts, expert_assignment, expert_offset) + + return batch, moe_input, scores, mapped_slots, expert_counts + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, n_experts", [(13, 64), (278, 64), (1977, 64)]) +@pytest.mark.parametrize("do_padding", [True, False]) +def test_moe_gather(n_tokens, n_experts, do_padding): + + batch, moe_input, scores, mapped_slots, expert_counts = build_inputs(n_tokens, n_experts, do_padding) + + output = torch.randn((batch.tensor_toks, 4096), dtype=torch.float16, device=get_accelerator().current_device()) + + gather = MoEGather(DtypeEnum.fp16, 4096) + gather(output, moe_input, scores, mapped_slots, expert_counts) + + for token_idx in range(n_tokens): + assert torch.equal( + output[token_idx], + torch.full((4096, ), + token_idx * scores[token_idx], + dtype=torch.float16, + device=get_accelerator().current_device())) diff --git a/tests/unit/inference/kernels/ragged_ops/test_moe_scatter.py b/tests/unit/inference/kernels/ragged_ops/test_moe_scatter.py new file mode 100644 index 000000000000..4ca051410c1c --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_moe_scatter.py @@ -0,0 +1,74 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import DtypeEnum +from deepspeed.inference.v2.kernels.ragged_ops import MoEScatter, RaggedTop1Gating +from .ragged_testing_utils import build_simple_batch +""" +For simplicity's sake, these tests do rely on ``RaggedTop1Gating`` to produce correct +inputs. If ``RaggedTop1Gating`` is broken, these tests will fail, so double check +the unit test results there before debugging here. +""" + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens, n_experts", [(13, 64), (278, 64), (1977, 64)]) +@pytest.mark.parametrize("do_padding", [True, False]) +def test_moe_scatter(n_tokens, n_experts, do_padding): + + # Sequence composition shouldn't matter here + batch = build_simple_batch([n_tokens], padding=do_padding) + + logits = torch.randn((batch.tensor_toks, n_experts), + dtype=torch.float16, + device=get_accelerator().current_device()) + + # This will make each token's value equal to its index. NOTE: This will break for + # tokens with index > 2048. + hidden_states = torch.arange(batch.tensor_toks, dtype=torch.float16, + device=get_accelerator().current_device()).repeat_interleave(4096, dim=0).reshape( + batch.tensor_toks, 4096).contiguous() + + gate = RaggedTop1Gating(DtypeEnum.fp16) + + # Gating outputs + expert_counts = torch.zeros((n_experts, ), dtype=torch.int32, device=get_accelerator().current_device()) + scores = torch.empty((batch.tensor_toks, ), dtype=torch.float32, device=get_accelerator().current_device()) + expert_assignment = torch.empty((batch.tensor_toks, ), + dtype=torch.int32, + device=get_accelerator().current_device()) + expert_offset = torch.empty((batch.tensor_toks, ), dtype=torch.int32, device=get_accelerator().current_device()) + + gate(expert_counts, scores, expert_assignment, expert_offset, logits, batch) + + # Scatter outputs + moe_input = torch.empty((batch.tensor_toks, 4096), dtype=torch.float16, device=get_accelerator().current_device()) + expert_cumsum = torch.empty((n_experts, ), dtype=torch.int64, device=get_accelerator().current_device()) + mapped_slots = torch.empty((batch.tensor_toks, ), dtype=torch.int32, device=get_accelerator().current_device()) + + scatter = MoEScatter(DtypeEnum.fp16, 4096) + scatter(moe_input, expert_cumsum, mapped_slots, hidden_states, expert_counts, expert_assignment, expert_offset) + assert torch.equal(expert_cumsum, torch.cumsum(expert_counts, dim=0).to(torch.int64)) + + for token_idx in range(batch.tensor_toks): + if token_idx < n_tokens: + expert_idx = expert_assignment[token_idx].item() + if expert_idx == 0: + expert_cumsum_val = 0 + else: + expert_cumsum_val = expert_cumsum[expert_idx - 1] + offset = expert_offset[token_idx] + total_offset = offset + expert_cumsum_val + + assert total_offset == mapped_slots[token_idx].item() + assert torch.equal(moe_input[total_offset], hidden_states[token_idx]) + else: + assert mapped_slots[token_idx].item() == -1 + + assert expert_cumsum[-1] == n_tokens diff --git a/tests/unit/inference/kernels/ragged_ops/test_ragged_embed.py b/tests/unit/inference/kernels/ragged_ops/test_ragged_embed.py new file mode 100644 index 000000000000..94f3f143274e --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_ragged_embed.py @@ -0,0 +1,177 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import List, Optional, Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.kernels.ragged_ops import RaggedEmbeddingKernel +from ...inference_test_utils import allclose, get_dtypes +from .ragged_testing_utils import build_batch_and_manager + + +def baseline_implementation(token_ids: torch.Tensor, + embedding_table: torch.Tensor, + unpadded_size: int, + positional_embedding_table: Optional[torch.Tensor] = None, + positional_ids: Optional[torch.Tensor] = None) -> torch.Tensor: + """ + Baseline implementation for our ragged embedding kernel. + """ + if unpadded_size == token_ids.shape[0]: + token_embed = torch.nn.functional.embedding(token_ids, embedding_table) + + if positional_embedding_table is not None: + pos_embed = torch.nn.functional.embedding(positional_ids, positional_embedding_table) + token_embed += pos_embed + return token_embed + else: + real_token_ids = token_ids[:unpadded_size] + output = torch.empty((token_ids.shape[0], embedding_table.shape[1]), + dtype=embedding_table.dtype, + device=get_accelerator().current_device()) + unpadded_output = torch.nn.functional.embedding(real_token_ids, embedding_table) + + # Positional embeddings aren't padded because it's simulated + if positional_embedding_table is not None: + pos_embed = torch.nn.functional.embedding(positional_ids, positional_embedding_table) + unpadded_output += pos_embed + + output[:unpadded_size] = unpadded_output + return output + + +def _ragged_embed_test_helper(sequence_config: List[Tuple[int, int]], + embed_dtype: torch.dtype, + token_dtype: torch.dtype, + embed_dim: int, + vocab_size: int, + do_padding: bool = False, + pos_embed_size: int = -1, + pos_embed_offset: int = 0) -> None: + """ + Helper for embedding test to limit the number of tests to run. + + Params: + embed_dim (int): Model dimension + vocab_size (int): Leading dimension on embedding weight + pos_embed_size (int): Size of positional embedding. If negative, no positional embedding + is used. + pos_embed_offset (int): Offset for positional embedding. Effectively, the raw offsets + of a token into a sequence are offset by this amount into the embedding matrix. ( + i.e. the shape of the positional embeddings is (pos_embed_size + pos_embed_offset + embed_dim) + """ + device = get_accelerator().current_device() + + # Heads/Block size are irrelevant here but need something. + batch, _, _, = build_batch_and_manager(sequence_config, 64, 16, 64, vocab_range=vocab_size, padding=do_padding) + + embedding_table = torch.randn((vocab_size, embed_dim), dtype=embed_dtype, device=device) + + if pos_embed_size > 0: + pos_embedding_table = torch.randn((pos_embed_size + pos_embed_offset, embed_dim), + dtype=embed_dtype, + device=device) + positional_ids = torch.cat([ + torch.arange(start_idx, start_idx + seq_len, dtype=token_dtype, device=device) + for seq_len, start_idx in sequence_config + ]) + pos_embed_offset + else: + pos_embedding_table = None + positional_ids = None + + baseline_output = baseline_implementation(batch.input_ids().to(token_dtype), embedding_table, batch.current_tokens, + pos_embedding_table, positional_ids) + + kernel = RaggedEmbeddingKernel(embed_dtype, token_dtype, embed_dim) + output = torch.empty_like(baseline_output) + + kernel(output, + batch, + embedding_table, + position_embed_weight=pos_embedding_table, + position_embed_offset=pos_embed_offset) + + if do_padding: + assert output.shape[0] != batch.current_tokens + + assert allclose(output[:batch.current_tokens], baseline_output[:batch.current_tokens]) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('token_dtype', [torch.int32, torch.int64]) +@pytest.mark.parametrize('embed_dtype', get_dtypes()) +def test_dtype_permutations(token_dtype: torch.dtype, embed_dtype: torch.dtype) -> None: + """ + Validate (on a single problem size) that the kernel support for different data types + is correct. + """ + embed_dim = 4096 + vocab_size = 50304 + + _ragged_embed_test_helper([(256, 0)], embed_dtype, token_dtype, embed_dim, vocab_size) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('vocab_size, embed_dim', [(1024, 1024), (32000, 5120), (50304, 6144)]) +def test_problem_size_permutations(vocab_size: int, embed_dim: int) -> None: + """ + Validate on wider range of problem sizes. + """ + + _ragged_embed_test_helper([(256, 0)], torch.float16, torch.int32, embed_dim, vocab_size) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('seq_lens', [[128, 64, 192, 32], [57, 112, 63, 89, 1, 1, 1, 1]]) +@pytest.mark.parametrize('do_padding', [True, False]) +def test_complex_sequences(seq_lens: List[int], do_padding: bool) -> None: + """ + Validate on different ragged batch construction scenarios. + """ + embed_dim = 4096 + vocab_size = 50304 + + _ragged_embed_test_helper([(seq_len, 0) for seq_len in seq_lens], + torch.float16, + torch.int32, + embed_dim, + vocab_size, + do_padding=do_padding) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("seq_lens", [[(256, 0)], [(256, 0), + (128, 0)], [(256, 0), (128, 0), + (64, 0)], [(1, 877), (619, 0), (213, 372), (1, 45)]]) +def test_positional_embedding(seq_lens: List[Tuple[int, int]]) -> None: + """ + Validate that positional embedding works correctly. + """ + embed_dim = 4096 + vocab_size = 50304 + + _ragged_embed_test_helper(seq_lens, torch.float16, torch.int32, embed_dim, vocab_size, pos_embed_size=2048) + + +@pytest.mark.inference_v2_ops +def test_positional_embedding_offset() -> None: + """ + Validate that positional embedding works correctly with an offset. + """ + embed_dim = 4096 + vocab_size = 50304 + seq_config = [(1, 877), (619, 0), (213, 372), (1, 45)] + + _ragged_embed_test_helper(seq_config, + torch.float16, + torch.int32, + embed_dim, + vocab_size, + pos_embed_size=2048, + pos_embed_offset=2) diff --git a/tests/unit/inference/kernels/ragged_ops/test_top_1_gating.py b/tests/unit/inference/kernels/ragged_ops/test_top_1_gating.py new file mode 100644 index 000000000000..96bf28eea7ad --- /dev/null +++ b/tests/unit/inference/kernels/ragged_ops/test_top_1_gating.py @@ -0,0 +1,120 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch +import torch.nn.functional as F + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import DtypeEnum +from deepspeed.inference.v2.kernels.ragged_ops import RaggedTop1Gating +from .ragged_testing_utils import build_simple_batch +from ...inference_test_utils import allclose + + +def _test_single_mapping_helper(n_tokens: int, + n_experts: int, + assigned_expert: int, + logit_fill: float = 0.0, + match_fill: float = 1.0) -> None: + logits = torch.full((n_tokens, n_experts), + logit_fill, + dtype=torch.float16, + device=get_accelerator().current_device()) + + logits[:, assigned_expert] = match_fill + + gate = RaggedTop1Gating(DtypeEnum.fp16) + + expert_counts = torch.zeros((n_experts, ), dtype=torch.int32, device=get_accelerator().current_device()) + scores = torch.empty((n_tokens, ), dtype=torch.float32, device=get_accelerator().current_device()) + expert_assignment = torch.empty((n_tokens, ), dtype=torch.int32, device=get_accelerator().current_device()) + expert_offset = torch.empty((n_tokens, ), dtype=torch.int32, device=get_accelerator().current_device()) + batch = build_simple_batch([n_tokens], padding=False) + + gate(expert_counts, scores, expert_assignment, expert_offset, logits, batch) + + assert expert_counts[assigned_expert] == n_tokens + assert torch.all(expert_assignment == assigned_expert) + assert torch.unique(expert_offset).shape[0] == n_tokens + assert allclose(scores, F.softmax(logits.float(), dim=1)[:, assigned_expert]) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('n_tokens, n_experts', [(1, 16), (17, 16), (32, 128), (89, 128), (433, 128)]) +def test_single_mapping_gating(n_tokens: int, n_experts: int) -> None: + """ + Evaluate our expert stacking behavior in complete isolation. This ensures all tokens + mapped to the same expert are getting unique offsets and identical scores. + """ + assigned_expert = 13 + _test_single_mapping_helper(n_tokens, n_experts, assigned_expert) + + +@pytest.mark.inference_v2_ops +def test_negative_logits(): + """ + Ensure that scores/values are propagated correctly when all the logits are negative. An + earlier implementation of the scoring would return NaN for this case. + """ + _test_single_mapping_helper(128, 32, 13, logit_fill=-2.0, match_fill=-1.0) + + +@pytest.mark.inference_v2_ops +def test_determinism(): + """ + Ensure that ties between two logits are broken deterministically. This is essential when + the gating is distributed across multiple devices that need to map the same token to + the same expert. + """ + + n_tokens = 512 + n_experts = 64 + + logits = torch.zeros((n_tokens, n_experts), dtype=torch.float16, device=get_accelerator().current_device()) + batch = build_simple_batch([n_tokens], padding=False) + + logits[:, 19] = 1.0 + logits[:, 26] = 1.0 + + gate = RaggedTop1Gating(DtypeEnum.fp16) + + for _ in range(1024): + expert_counts = torch.zeros((n_experts, ), dtype=torch.int32, device=get_accelerator().current_device()) + scores = torch.empty((n_tokens, ), dtype=torch.float32, device=get_accelerator().current_device()) + expert_assignment = torch.empty((n_tokens, ), dtype=torch.int32, device=get_accelerator().current_device()) + expert_offset = torch.empty((n_tokens, ), dtype=torch.int32, device=get_accelerator().current_device()) + batch = build_simple_batch([n_tokens], padding=False) + + gate(expert_counts, scores, expert_assignment, expert_offset, logits, batch) + + assert expert_counts[19] == n_tokens + assert expert_counts[26] == 0 + assert torch.all(expert_assignment == 19) + assert torch.unique(expert_offset).shape[0] == n_tokens + assert allclose(scores, F.softmax(logits.float(), dim=1)[:, 19]) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize('n_tokens, n_experts', [(1, 16), (17, 16), (32, 128), (89, 128), (433, 2)]) +def test_score_accuracy(n_tokens: int, n_experts: int) -> None: + """ + Validate expert scores are correct. + """ + logits = torch.randn((n_tokens, n_experts), dtype=torch.float16, device=get_accelerator().current_device()) + batch = build_simple_batch([n_tokens], padding=False) + + gate = RaggedTop1Gating(DtypeEnum.fp16) + + expert_counts = torch.zeros((n_experts, ), dtype=torch.int32, device=get_accelerator().current_device()) + scores = torch.empty((n_tokens, ), dtype=torch.float32, device=get_accelerator().current_device()) + expert_assignment = torch.empty((n_tokens, ), dtype=torch.int32, device=get_accelerator().current_device()) + expert_offset = torch.empty((n_tokens, ), dtype=torch.int32, device=get_accelerator().current_device()) + + ref_scores = F.softmax(logits.float(), dim=1).max(dim=1).values + + gate(expert_counts, scores, expert_assignment, expert_offset, logits, batch) + assert allclose(scores, ref_scores) + assert expert_counts.sum() == n_tokens diff --git a/tests/unit/inference/model_implementations/__init__.py b/tests/unit/inference/model_implementations/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/model_implementations/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/model_implementations/parameters/__init__.py b/tests/unit/inference/model_implementations/parameters/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/model_implementations/parameters/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/model_implementations/parameters/test_layer_inheritance.py b/tests/unit/inference/model_implementations/parameters/test_layer_inheritance.py new file mode 100644 index 000000000000..20803e53a320 --- /dev/null +++ b/tests/unit/inference/model_implementations/parameters/test_layer_inheritance.py @@ -0,0 +1,50 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.inference.v2.model_implementations.layer_container_base import LayerContainer + +from .utils import validate_device, SimpleParam, DummyInferenceModel + + +class ParentLayer(LayerContainer): + """ + A layer that has a dependency on a simple parameter. + """ + + param_1: SimpleParam + + +class ChildLayer(ParentLayer): + """ + A layer that inherits from another layer. + """ + + param_2: SimpleParam + + +@pytest.mark.inference_v2 +def test_layer_inheritance(): + inference_model = DummyInferenceModel() + + multi_param_layer = ChildLayer(inference_model) + + assert multi_param_layer.n_params == 2 + assert multi_param_layer.is_initialized is False + + multi_param_layer.param_1.param = torch.ones(16, 16) + + assert multi_param_layer.is_initialized is False + + multi_param_layer.param_2.param = torch.full((16, 16), 2.0) + + assert multi_param_layer.is_initialized is True + assert isinstance(multi_param_layer.param_1, torch.Tensor) + assert isinstance(multi_param_layer.param_2, torch.Tensor) + + validate_device(multi_param_layer.param_1) + validate_device(multi_param_layer.param_2) diff --git a/tests/unit/inference/model_implementations/parameters/test_mapping.py b/tests/unit/inference/model_implementations/parameters/test_mapping.py new file mode 100644 index 000000000000..3c74d7a0479a --- /dev/null +++ b/tests/unit/inference/model_implementations/parameters/test_mapping.py @@ -0,0 +1,165 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.inference.v2.allocator import on_device +from deepspeed.inference.v2.model_implementations.parameter_base import ParameterBase, ParamList +from deepspeed.inference.v2.model_implementations.layer_container_base import LayerContainer + + +class MultiDependencyContainer(ParameterBase): + + dependency_1: torch.Tensor + + dependency_2: torch.Tensor + + @on_device + def finalize(self) -> torch.Tensor: + return torch.cat([self.dependency_1, self.dependency_2]) + + +class ListDependencyContainer(ParameterBase): + + dependencies: ParamList("list_items") # noqa: F821 + + @on_device + def finalize(self) -> torch.Tensor: + return torch.cat(tuple(self.dependencies)) + + +class MappingLayer(LayerContainer): + PARAM_MAPPING = { + "model.val.item.d_1": "multi_depend.dependency_1", + "model.val.item.d_2": "multi_depend.dependency_2", + "model.list_vals.*.d": "list_depend.dependencies" + } + + multi_depend: MultiDependencyContainer + + list_depend: ListDependencyContainer + + +class SubMappingLayer(MappingLayer): + PARAM_MAPPING = { + "model.val.item2.d_1": "multi_depend2.dependency_1", + "model.val.item2.d_2": "multi_depend2.dependency_2", + } + + multi_depend2: MultiDependencyContainer + + +class DoubleMappingLayer(LayerContainer): + PARAM_MAPPING = { + "model.val.item.d_1": ["multi_depend.dependency_1", "multi_depend.dependency_2"], + } + + multi_depend: MultiDependencyContainer + + +class InferenceModel: + + @property + def list_items(self) -> int: + return 16 + + +@pytest.mark.inference_v2 +def test_mapping_syntax(): + model = InferenceModel() + + mapping_layer = MappingLayer(model) + + mapping_layer.set_dependency("model.val.item.d_1", torch.ones(1)) + mapping_layer.set_dependency("model.val.item.d_2", torch.ones(1) * 2) + + assert isinstance(mapping_layer.multi_depend, torch.Tensor) + + for i in range(16): + mapping_layer.set_dependency(f"model.list_vals.{i}.d", torch.ones(1) * i) + if i != 16 - 1: + assert mapping_layer.is_initialized == False + + assert isinstance(mapping_layer.list_depend, torch.Tensor) + assert mapping_layer.is_initialized == True + + +@pytest.mark.inference_v2 +def test_sub_mapping_syntax(): + model = InferenceModel() + + mapping_layer = SubMappingLayer(model) + + mapping_layer.set_dependency("model.val.item.d_1", torch.ones(1)) + mapping_layer.set_dependency("model.val.item.d_2", torch.ones(1) * 2) + + assert isinstance(mapping_layer.multi_depend, torch.Tensor) + + mapping_layer.set_dependency("model.val.item2.d_1", torch.ones(1)) + mapping_layer.set_dependency("model.val.item2.d_2", torch.ones(1) * 2) + + assert isinstance(mapping_layer.multi_depend2, torch.Tensor) + + # We want to check into double digits to make sure that this isn't specific + # to single difit indexing. + for i in range(16): + mapping_layer.set_dependency(f"model.list_vals.{i}.d", torch.ones(1) * i) + if i != 16 - 1: + assert mapping_layer.is_initialized == False + + assert isinstance(mapping_layer.list_depend, torch.Tensor) + assert mapping_layer.is_initialized == True + + +@pytest.mark.inference_v2 +def test_double_mapping_syntax(): + model = InferenceModel() + + mapping_layer = DoubleMappingLayer(model) + mapping_layer.set_dependency("model.val.item.d_1", torch.ones(1)) + + # The single parameter setting should immediately make the parameter finalized + # and the whole layer initialized. + assert isinstance(mapping_layer.multi_depend, torch.Tensor) + assert mapping_layer.is_initialized == True + + +@pytest.mark.inference_v2 +def test_insufficient_mapping_syntax(): + """ + In the above example, we don't have a mapping for `multi_depend2.dependency_2`. + """ + + with pytest.raises(ValueError): + + class InsuffienctMappingLayer(LayerContainer): + PARAM_MAPPING = { + "model.val.item.d_1": "multi_depend1.dependency_1", + "model.val.item.d_2": "multi_depend1.dependency_2", + "model.val.item2.d_1": "multi_depend2.dependency_1", + } + + multi_depend1: MultiDependencyContainer + + multi_depend2: MultiDependencyContainer + + +@pytest.mark.inference_v2 +def test_unknown_target_mapping_syntax(): + """ + In the above example, `multi_depend_unknown` does not exist + """ + + with pytest.raises(ValueError): + + class UnknownTargetMappingLayer(LayerContainer): + PARAM_MAPPING = { + "model.val.item.d_1": "multi_depend1.dependency_1", + "model.val.item.d_2": "multi_depend1.dependency_2", + "model.val.item2.d_1": "multi_depend_unknown.dependency_1", + } + + multi_depend: MultiDependencyContainer diff --git a/tests/unit/inference/model_implementations/parameters/test_multi_parameter_layer.py b/tests/unit/inference/model_implementations/parameters/test_multi_parameter_layer.py new file mode 100644 index 000000000000..6bfc04e97c30 --- /dev/null +++ b/tests/unit/inference/model_implementations/parameters/test_multi_parameter_layer.py @@ -0,0 +1,111 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.inference.v2.model_implementations.layer_container_base import LayerContainer + +from .utils import validate_device, SimpleParam, ListParam, DummyInferenceModel + + +class MultiParameterLayer(LayerContainer): + """ + Two dependencies, both of which are simple parameters. + """ + + param_1: SimpleParam + + param_2: SimpleParam + + +class MixedMultiParameterLayer(LayerContainer): + """ + Two dependencies, one of which is a simple parameter, the other is a list parameter. + """ + + param_1: SimpleParam + + param_2: ListParam + + +@pytest.mark.inference_v2 +def test_multi_parameter_layer(): + inference_model = DummyInferenceModel() + + multi_param_layer = MultiParameterLayer(inference_model) + + assert multi_param_layer.n_params == 2 + assert multi_param_layer.is_initialized is False + + multi_param_layer.param_1.param = torch.ones(16, 16) + + assert multi_param_layer.is_initialized is False + + multi_param_layer.param_2.param = torch.full((16, 16), 2.0) + + assert multi_param_layer.is_initialized is True + assert isinstance(multi_param_layer.param_1, torch.Tensor) + assert isinstance(multi_param_layer.param_2, torch.Tensor) + + validate_device(multi_param_layer.param_1) + validate_device(multi_param_layer.param_2) + + +@pytest.mark.inference_v2 +def test_mixed_multi_parameter_layer(): + inference_model = DummyInferenceModel() + + mixed_multi_param_layer = MixedMultiParameterLayer(inference_model) + + assert mixed_multi_param_layer.n_params == 2 + assert mixed_multi_param_layer.is_initialized is False + + mixed_multi_param_layer.param_2.params[1] = torch.full((16, 16), 2.0) + assert mixed_multi_param_layer.is_initialized is False + assert not isinstance(mixed_multi_param_layer.param_2, torch.Tensor) + + mixed_multi_param_layer.param_1.param = torch.ones(16, 16) + assert mixed_multi_param_layer.is_initialized is False + assert isinstance(mixed_multi_param_layer.param_1, torch.Tensor) + + validate_device(mixed_multi_param_layer.param_1) + + mixed_multi_param_layer.param_2.params[0] = torch.full((16, 16), 2.0) + + assert mixed_multi_param_layer.is_initialized is True + assert isinstance(mixed_multi_param_layer.param_2, torch.Tensor) + + validate_device(mixed_multi_param_layer.param_2) + + +class NoCopyInferenceModel: + + @property + def num_dependencies(self) -> int: + return 2 + + def transform(self, param: torch.Tensor) -> torch.Tensor: + return param + + +@pytest.mark.inference_v2 +def test_device_validation(): + inference_model = NoCopyInferenceModel() + + multi_param_layer = MultiParameterLayer(inference_model) + + assert multi_param_layer.n_params == 2 + assert multi_param_layer.is_initialized is False + + multi_param_layer.param_1.param = torch.ones(16, 16) + + assert multi_param_layer.is_initialized is False + + multi_param_layer.param_2.param = torch.full((16, 16), 2.0) + + with pytest.raises(RuntimeError): + # NoCopyInference model did not copy the parameters, so the device validation should fail. + assert multi_param_layer.is_initialized is True diff --git a/tests/unit/inference/model_implementations/parameters/test_parameter_list.py b/tests/unit/inference/model_implementations/parameters/test_parameter_list.py new file mode 100644 index 000000000000..42edd90595fa --- /dev/null +++ b/tests/unit/inference/model_implementations/parameters/test_parameter_list.py @@ -0,0 +1,104 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.inference.v2.model_implementations.parameter_base import ParameterBase, ParamList +from deepspeed.inference.v2.model_implementations.layer_container_base import LayerContainer +from deepspeed.inference.v2.model_implementations.common_parameters import * +from deepspeed.inference.v2.allocator import on_device + +from .utils import validate_device + + +class SimpleMoELayer(LayerContainer): + + moe_mlp_1: UnfusedMoEMLP1Parameter + + +class DummyInferenceModel: + + def __init__(self, experts_per_rank: int) -> None: + self._num_experts = experts_per_rank + + @property + def num_experts(self) -> int: + return self._num_experts + + @on_device + def transform_moe_mlp_1_param(self, param: torch.Tensor) -> torch.Tensor: + return param + + +@pytest.mark.inference_v2 +def test_simple_moe_layer(): + + inference_model = DummyInferenceModel(experts_per_rank=2) + + simple_moe_layer = SimpleMoELayer(inference_model) + + assert simple_moe_layer.moe_mlp_1.experts[0] is None + assert simple_moe_layer.moe_mlp_1.experts[1] is None + + # Set the first expert + simple_moe_layer.moe_mlp_1.experts[0] = torch.zeros(16, 16) + + assert simple_moe_layer.moe_mlp_1.experts[0] is not None + assert simple_moe_layer.moe_mlp_1.experts[1] is None + + assert not simple_moe_layer.is_initialized + + # Set the second expert + simple_moe_layer.moe_mlp_1.experts[1] = torch.ones(16, 16) + + # We have all the experts, so the layer should be initialized + assert simple_moe_layer.is_initialized + assert isinstance(simple_moe_layer.moe_mlp_1, torch.Tensor) + + validate_device(simple_moe_layer.moe_mlp_1) + + +""" +Check that we can mix the number of elements in lists in the same context and have that +be tracked correctly. +""" + + +class CustomListParam1(ParameterBase): + + deps: ParamList("attr_1") + + +class CustomListParam2(ParameterBase): + + deps: ParamList("attr_2") + + +class MixedLayer(LayerContainer): + + list_1: CustomListParam1 + list_2: CustomListParam2 + + +class MixedInferenceModel: + + @property + def attr_1(self) -> int: + return 1 + + @property + def attr_2(self) -> int: + return 2 + + +@pytest.mark.inference_v2 +def test_mixed_param_lists(): + model = MixedInferenceModel() + + layer = MixedLayer(model) + + assert layer.list_1.deps.n_params == 1 + assert layer.list_2.deps.n_params == 2 diff --git a/tests/unit/inference/model_implementations/parameters/utils.py b/tests/unit/inference/model_implementations/parameters/utils.py new file mode 100644 index 000000000000..0d2cbb27d40e --- /dev/null +++ b/tests/unit/inference/model_implementations/parameters/utils.py @@ -0,0 +1,58 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.allocator import on_device +from deepspeed.inference.v2.model_implementations.parameter_base import ParameterBase, ParametrizedList + + +class SimpleParam(ParameterBase): + """ + Parameter with single dependency. + """ + + param: torch.Tensor + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform(self.param) + + +class SimpleParametrizedList(ParametrizedList): + """ + Parameter list based on `num_dependencies` attribute. + """ + + count_attr: str = "num_dependencies" + + +class ListParam(ParameterBase): + """ + Parameter with list dependency. + + NOTE: This uses the tuple workaround for the `ParametrizedList` class + as described in the docstring of `ParametrizedList`. + """ + + params: SimpleParametrizedList + + def finalize(self) -> torch.Tensor: + return self.inference_model.transform(torch.cat(tuple(self.params))) + + +class DummyInferenceModel: + + @property + def num_dependencies(self) -> int: + return 2 + + @on_device + def transform(self, param: torch.Tensor) -> torch.Tensor: + return param + + +def validate_device(tensor: torch.Tensor): + assert tensor.device == torch.device(get_accelerator().current_device()) diff --git a/tests/unit/inference/model_implementations/sharding/__init__.py b/tests/unit/inference/model_implementations/sharding/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/model_implementations/sharding/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/model_implementations/sharding/test_attn_out_sharding.py b/tests/unit/inference/model_implementations/sharding/test_attn_out_sharding.py new file mode 100644 index 000000000000..850c4c24fde6 --- /dev/null +++ b/tests/unit/inference/model_implementations/sharding/test_attn_out_sharding.py @@ -0,0 +1,129 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.model_implementations.sharding import * + +# None of the logic should be dependent on head size. +HEAD_SIZE = 64 + + +def fill_with_head_ids(head_size: int, n_heads: int) -> torch.Tensor: + """ + Fills a tensor with the associated head ids. All columns should have the same value. + """ + head_ids = torch.arange(n_heads, dtype=torch.half, device=get_accelerator().current_device()) + + head_ids = head_ids.repeat_interleave(head_size).repeat(head_size * n_heads).reshape(n_heads * head_size, -1) + return head_ids + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("n_heads, n_shards", [(1, 1), (8, 4), (32, 8)]) +def test_mha_even_sharding(n_heads: int, n_shards: int): + """ + Even head sharding for MHA. + + Args: + n_heads (int): The number QKV heads. + n_shards (int): The number of shards to test for. + """ + param = fill_with_head_ids(HEAD_SIZE, n_heads) + + n_local_heads = n_heads // n_shards + sharded_shape = (HEAD_SIZE * n_heads, HEAD_SIZE * n_local_heads) + + for shard_rank in range(n_shards): + sharded_param = shard_attn_out_param(param, shard_rank, n_shards, HEAD_SIZE) + n_heads_local_q, _ = get_local_heads(shard_rank, n_shards, n_heads) + + assert sharded_param.shape[-1] == HEAD_SIZE * n_heads_local_q + assert sharded_param.shape == sharded_shape + + heads = torch.chunk(sharded_param, n_local_heads, dim=1) + + for i, head in enumerate(heads): + assert torch.all(head == i + shard_rank * n_local_heads) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("n_heads, n_shards", [(3, 2), (20, 8)]) +def test_mha_unbalanced_sharding(n_heads: int, n_shards: int): + """ + Unbalanced head sharding for MHA. + + Args: + n_heads (int): The number QKV heads. + n_shards (int): The number of shards to test for. + """ + param = fill_with_head_ids(HEAD_SIZE, n_heads) + + max_heads = 0 + min_heads = n_heads + seen_heads = set() + total_heads = 0 + + for shard_rank in range(n_shards): + sharded_param = shard_attn_out_param(param, shard_rank, n_shards, HEAD_SIZE) + n_heads_local_q, _ = get_local_heads(shard_rank, n_shards, n_heads) + + assert sharded_param.shape[-1] == HEAD_SIZE * n_heads_local_q + + n_local_heads = sharded_param.shape[1] // HEAD_SIZE + total_heads += n_local_heads + max_heads = max(max_heads, n_local_heads) + min_heads = min(min_heads, n_local_heads) + + for i in range(n_local_heads): + head_ids = torch.unique_consecutive(sharded_param[:, i * HEAD_SIZE:(i + 1) * HEAD_SIZE]) + assert len(head_ids) == 1 + seen_heads.add(head_ids.item()) + + assert max_heads == min_heads + 1 + assert total_heads == n_heads + assert len(seen_heads) == n_heads + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("n_heads_q, n_heads_kv, n_shards", [(20, 4, 8)]) +def test_gqa_uneven_sharding(n_heads_q: int, n_heads_kv: int, n_shards: int): + """ + We only test the uneven GQA test case because even GQA shards the attention output + in the exact same manner as MHA. + + Args: + n_heads_q (int): The number of query heads. + n_heads_kv (int): The number of key/value heads. + n_shards (int): The number of shards to test for. + """ + param = fill_with_head_ids(HEAD_SIZE, n_heads_q) + + min_heads = n_heads_q + max_heads = 0 + seen_heads = set() + total_heads = 0 + + for shard_rank in range(n_shards): + sharded_param = shard_attn_out_param(param, shard_rank, n_shards, HEAD_SIZE, n_heads_q, n_heads_kv) + n_heads_local_q, _ = get_local_heads(shard_rank, n_shards, n_heads_q, n_heads_kv) + + assert sharded_param.shape[-1] == HEAD_SIZE * n_heads_local_q + + n_local_heads = sharded_param.shape[1] // HEAD_SIZE + total_heads += n_local_heads + max_heads = max(max_heads, n_local_heads) + min_heads = min(min_heads, n_local_heads) + + for i in range(n_local_heads): + head_id = torch.unique_consecutive(sharded_param[:, i * HEAD_SIZE:(i + 1) * HEAD_SIZE]) + assert len(head_id) == 1 + seen_heads.add(head_id.item()) + + assert max_heads == min_heads + 1 + assert total_heads == n_heads_q + assert len(seen_heads) == n_heads_q diff --git a/tests/unit/inference/model_implementations/sharding/test_mlp_sharding.py b/tests/unit/inference/model_implementations/sharding/test_mlp_sharding.py new file mode 100644 index 000000000000..aac7e5391d8f --- /dev/null +++ b/tests/unit/inference/model_implementations/sharding/test_mlp_sharding.py @@ -0,0 +1,116 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.model_implementations.sharding import * + + +def round_up_to_256(x: int) -> int: + """ + Round up to the nearest multiple of 256. + """ + return x + (256 - x % 256) + + +def make_params(model_dim: int, ffn_multiplier: int, n_experts: int, gated: bool = False) -> torch.Tensor: + """ + + """ + if gated: + mlp_1_intermediate = round_up_to_256(int(model_dim * ffn_multiplier * 4 / 3)) + mlp_2_intermediate = mlp_1_intermediate // 2 + else: + mlp_1_intermediate = ffn_multiplier * model_dim + mlp_2_intermediate = ffn_multiplier * model_dim + + mlp_1_shared_dim = torch.arange(mlp_1_intermediate, dtype=torch.float32, device=get_accelerator().current_device()) + + mlp_1_w = mlp_1_shared_dim.repeat_interleave(model_dim).reshape(mlp_1_intermediate, model_dim) + mlp_1_b = mlp_1_shared_dim + + mlp_2_shared_dim = torch.arange(mlp_2_intermediate, dtype=torch.float32, device=get_accelerator().current_device()) + mlp_2_w = mlp_2_shared_dim.repeat(model_dim).reshape(model_dim, mlp_2_intermediate) + mlp_2_b = torch.ones(model_dim, dtype=torch.float32, device=get_accelerator().current_device()) + + if n_experts > 1: + mlp_1_w = mlp_1_w.expand(n_experts, -1, -1) + mlp_1_b = mlp_1_b.expand(n_experts, -1) + mlp_2_w = mlp_2_w.expand(n_experts, -1, -1) + mlp_2_b = mlp_2_b.expand(n_experts, -1) + + return (mlp_1_w, mlp_1_b, mlp_2_w, mlp_2_b) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("model_dim, ffn_multiplier, n_shards", [(1024, 4, 1), (1024, 4, 8), (1024, 4, 6)]) +@pytest.mark.parametrize("n_experts", [1, 16]) +def test_even_ffn_sharding(model_dim: int, ffn_multiplier: int, n_shards: int, n_experts: int): + """ + FFN sharding tends to be much simpler than attention sharding since it works on larger granularities. + While the test case of (1024, 4, 6) is not a use case we're likely to see, this does ensure that + the sharding logic will round correctly for the alignments we care about. + """ + mlp_1_w, mlp_1_b, mlp_2_w, mlp_2_b = make_params(model_dim, ffn_multiplier, n_experts) + + total_ffn_dim = model_dim * ffn_multiplier + mapped_neurons = 0 + + is_moe = n_experts > 1 + + for shard_rank in range(n_shards): + shard_1_w = shard_mlp_1_param(mlp_1_w, shard_rank, n_shards, is_moe=is_moe) + shard_1_b = shard_mlp_1_param(mlp_1_b, shard_rank, n_shards, is_moe=is_moe) + shard_2_w = shard_mlp_2_param(mlp_2_w, shard_rank, n_shards, is_moe=is_moe) + shard_2_b = shard_mlp_2_param(mlp_2_b, shard_rank, n_shards, is_moe=is_moe) + + assert shard_1_w.shape[-2] == shard_2_w.shape[-1] + assert shard_1_w.shape[-2] % DEFAULT_SHARD_GRANULARITY == 0 + assert shard_1_w.shape[-2] == shard_1_b.shape[-1] + + mapped_neurons += shard_1_w.shape[-2] + + if shard_rank != 0: + assert shard_2_b is None + else: + assert shard_2_b.shape[-1] == model_dim + + assert mapped_neurons == total_ffn_dim + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("model_dim, ffn_multiplier, n_shards", [(1024, 4, 1), (1024, 4, 8), (1024, 4, 6)]) +@pytest.mark.parametrize("n_experts", [1, 16]) +def test_gated_ffn_sharding(model_dim: int, ffn_multiplier: int, n_shards: int, n_experts: int): + """ + Test the same cases assuming a gated regime. + """ + mlp_1_w, mlp_1_b, mlp_2_w, mlp_2_b = make_params(model_dim, ffn_multiplier, n_experts, gated=True) + + total_ffn_dim = round_up_to_256(int(model_dim * ffn_multiplier * 4 / 3)) + mapped_neurons = 0 + + is_moe = n_experts > 1 + + for shard_rank in range(n_shards): + shard_1_w = shard_mlp_1_param(mlp_1_w, shard_rank, n_shards, gated=True, is_moe=is_moe) + shard_1_b = shard_mlp_1_param(mlp_1_b, shard_rank, n_shards, gated=True, is_moe=is_moe) + shard_2_w = shard_mlp_2_param(mlp_2_w, shard_rank, n_shards, is_moe=is_moe) + shard_2_b = shard_mlp_2_param(mlp_2_b, shard_rank, n_shards, is_moe=is_moe) + + assert shard_1_w.shape[-2] == shard_2_w.shape[-1] * 2 + assert shard_1_w.shape[-2] % DEFAULT_SHARD_GRANULARITY == 0 + assert shard_1_w.shape[-2] == shard_1_b.shape[-1] + + mapped_neurons += shard_1_w.shape[-2] + + if shard_rank != 0: + assert shard_2_b is None + else: + assert shard_2_b.shape[-1] == model_dim + + assert mapped_neurons == total_ffn_dim diff --git a/tests/unit/inference/model_implementations/sharding/test_qkv_sharding.py b/tests/unit/inference/model_implementations/sharding/test_qkv_sharding.py new file mode 100644 index 000000000000..9a1cb9c09c64 --- /dev/null +++ b/tests/unit/inference/model_implementations/sharding/test_qkv_sharding.py @@ -0,0 +1,251 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.model_implementations.sharding import * + + +def fill_with_head_ids(head_size: int, n_heads_q: int, n_heads_kv: Optional[int] = None) -> torch.Tensor: + """ + + """ + head_ids_q = torch.arange(n_heads_q, dtype=torch.half, device=get_accelerator().current_device()) + head_vals_q = head_ids_q.repeat_interleave(head_size * head_size * n_heads_q).reshape(n_heads_q * head_size, -1) + + if n_heads_kv is None: + return torch.cat([head_vals_q, head_vals_q, head_vals_q], dim=0) + + head_ids_k = torch.arange(n_heads_kv, dtype=torch.half, device=get_accelerator().current_device()) + head_vals_k = head_ids_k.repeat_interleave(head_size * head_size * n_heads_q).reshape(n_heads_kv * head_size, -1) + + return torch.cat([head_vals_q, head_vals_k, head_vals_k], dim=0) + + +def validate_inferred_shape(shard: torch.Tensor, head_size: int, n_local_q_heads: int, n_local_kv_heads: int): + """ + Validate that the leading dim of the shard is of the expected size and aligns with the sharding + logic for the attention computation itself. + """ + inferred_leading_dim = head_size * (n_local_q_heads + 2 * n_local_kv_heads) + assert shard.shape[0] == inferred_leading_dim + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("head_size", [64]) +@pytest.mark.parametrize("n_heads,n_shards", [(1, 1), (32, 1), (32, 8)]) +def test_even_mha_sharding(head_size: int, n_heads: int, n_shards: int): + """ + Test for MHA sharding. In these scenarios, we expect that each of the shards + should be the same size. + """ + param = fill_with_head_ids(head_size, n_heads) + + heads_per_shard = n_heads // n_shards + + for shard_rank in range(n_shards): + + shard = shard_qkv_param(param, shard_rank, n_shards, head_size, n_heads, n_heads) + n_local_q_heads, n_local_kv_heads = get_local_heads(shard_rank, n_shards, n_heads, n_heads) + validate_inferred_shape(shard, head_size, n_local_q_heads, n_local_kv_heads) + + assert shard.shape == (3 * head_size * heads_per_shard, head_size * n_heads) + + heads = shard.chunk(heads_per_shard * 3, dim=0) + for i in range(heads_per_shard): + assert torch.all(heads[i] == i + shard_rank * heads_per_shard) + assert torch.all(heads[i + heads_per_shard] == i + shard_rank * heads_per_shard) + assert torch.all(heads[i + heads_per_shard * 2] == i + shard_rank * heads_per_shard) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("head_size", [64]) +@pytest.mark.parametrize("n_heads, n_shards", [(3, 2), (20, 8)]) +def test_unbalanced_mha_sharding(head_size: int, n_heads: int, n_shards: int): + """ + Test MHA sharding when the distribution of heads will not be equal across all ranks. + """ + param = fill_with_head_ids(head_size, n_heads) + + max_heads = 0 + min_heads = n_heads + total_heads = 0 + seen_heads = set() + + for shard_rank in range(n_shards): + shard = shard_qkv_param(param, shard_rank, n_shards, head_size, n_heads, n_heads) + n_local_q_heads, n_local_kv_heads = get_local_heads(shard_rank, n_shards, n_heads, n_heads) + validate_inferred_shape(shard, head_size, n_local_q_heads, n_local_kv_heads) + + n_heads_in_shard = shard.shape[0] // head_size // 3 + + max_heads = max(max_heads, n_heads_in_shard) + min_heads = min(min_heads, n_heads_in_shard) + total_heads += n_heads_in_shard + + heads = shard.chunk(n_heads_in_shard * 3, dim=0) + + for local_head_id in range(n_heads_in_shard): + head_qkv = torch.cat([ + heads[local_head_id], heads[local_head_id + n_heads_in_shard], + heads[local_head_id + 2 * n_heads_in_shard] + ], + dim=0) + assert head_qkv.shape == (3 * head_size, head_size * n_heads) + + global_head_id = torch.unique_consecutive(head_qkv) + assert len(global_head_id) == 1 + + seen_heads.add(global_head_id.item()) + + assert max_heads - min_heads <= 1 + assert total_heads == n_heads + assert len(seen_heads) == n_heads + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("head_size", [64]) +@pytest.mark.parametrize("n_heads_q, n_heads_kv, n_shards", [(4, 2, 1), (8, 2, 1), (64, 16, 8)]) +def test_gqa_even_sharding(head_size: int, n_heads_q: int, n_heads_kv: int, n_shards: int): + """ + Test GQA sharding when the KV heads are evenly divisible by the number of shards. + """ + param = fill_with_head_ids(head_size, n_heads_q, n_heads_kv) + + n_kv_heads_in_shard = n_heads_kv // n_shards + n_q_heads_in_shard = n_heads_q // n_shards + + for shard_rank in range(n_shards): + shard = shard_qkv_param(param, shard_rank, n_shards, head_size, n_heads_q, n_heads_kv) + n_local_q_heads, n_local_kv_heads = get_local_heads(shard_rank, n_shards, n_heads_q, n_heads_kv) + validate_inferred_shape(shard, head_size, n_local_q_heads, n_local_kv_heads) + + assert shard.shape[0] == (n_q_heads_in_shard + n_kv_heads_in_shard * 2) * head_size + + q = shard[:n_q_heads_in_shard * head_size] + k = shard[n_q_heads_in_shard * head_size:(n_q_heads_in_shard + n_kv_heads_in_shard) * head_size] + v = shard[(n_q_heads_in_shard + n_kv_heads_in_shard) * head_size:] + + for local_head_id in range(n_q_heads_in_shard): + assert torch.all(q[local_head_id * head_size:(local_head_id + 1) * head_size] == local_head_id + + shard_rank * n_q_heads_in_shard) + + for local_head_id in range(n_kv_heads_in_shard): + assert torch.all(k[local_head_id * head_size:(local_head_id + 1) * head_size] == local_head_id + + shard_rank * n_kv_heads_in_shard) + assert torch.all(v[local_head_id * head_size:(local_head_id + 1) * head_size] == local_head_id + + shard_rank * n_kv_heads_in_shard) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("head_size", [64]) +@pytest.mark.parametrize("n_heads_q, n_heads_kv, n_shards", [(4, 2, 4), (20, 4, 8)]) +def test_gqa_uneven_sharding(head_size: int, n_heads_q: int, n_heads_kv: int, n_shards: int): + """ + Test GQA sharding when there are more shards than KV heads. + """ + param = fill_with_head_ids(head_size, n_heads_q, n_heads_kv) + + n_kv_heads_in_shard = 1 + n_shards_per_kv_head = n_shards // n_heads_kv + + max_heads = 0 + min_heads = n_heads_q + total_heads = 0 + seen_heads = set() + + for shard_rank in range(n_shards): + shard = shard_qkv_param(param, shard_rank, n_shards, head_size, n_heads_q, n_heads_kv) + n_local_q_heads, n_local_kv_heads = get_local_heads(shard_rank, n_shards, n_heads_q, n_heads_kv) + validate_inferred_shape(shard, head_size, n_local_q_heads, n_local_kv_heads) + + local_n_heads_q = (shard.shape[0] - 2 * n_kv_heads_in_shard * head_size) // head_size + + max_heads = max(max_heads, local_n_heads_q) + min_heads = min(min_heads, local_n_heads_q) + total_heads += local_n_heads_q + + q = shard[:local_n_heads_q * head_size] + kv = shard[local_n_heads_q * head_size:] + + for local_head_id in range(local_n_heads_q): + q_head_id = torch.unique_consecutive(q[local_head_id * head_size:(local_head_id + 1) * head_size]) + assert len(q_head_id) == 1 + + seen_heads.add(q_head_id.item()) + + kv_id_calc = shard_rank // n_shards_per_kv_head + kv_id = torch.unique_consecutive(kv) + assert len(kv_id) == 1 + assert kv_id.item() == kv_id_calc + + assert max_heads - min_heads <= 1 + assert total_heads == n_heads_q + assert len(seen_heads) == n_heads_q + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("head_size", [64]) +@pytest.mark.parametrize("n_heads, n_shards", [(6, 8)]) +def test_unsupported_mha_configs(head_size: int, n_heads: int, n_shards: int): + """ + Sharding should fail if there are fewer heads than shards. + + TODO(cmikeh2): Look to support this configuration. + """ + param = fill_with_head_ids(head_size, n_heads) + + for shard_rank in range(n_shards): + with pytest.raises(ValueError): + shard_qkv_param(param, shard_rank, n_shards, head_size, n_heads, n_heads) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize("head_size", [64]) +@pytest.mark.parametrize("n_heads_q, n_heads_kv, n_shards", [(5, 2, 1), (40, 10, 8), (30, 5, 8)]) +def test_unsupported_gqa_configs(head_size: int, n_heads_q: int, n_heads_kv: int, n_shards: int): + """ + GQA has stricter requirements. We must be able to evenly shard or distribute the KV heads. + + Test cases are to test the following preconditions specifically: + 1. n_heads_q % n_heads_kv == 0 + 2. We must be able to evenly distribute KV heads + 3. We must be able to evely split KV heads + """ + param = fill_with_head_ids(head_size, n_heads_q, n_heads_kv) + + for shard_rank in range(n_shards): + with pytest.raises(ValueError): + shard_qkv_param(param, shard_rank, n_shards, head_size, n_heads_q, n_heads_kv) + + +@pytest.mark.inference_v2 +def test_mha_input_shape_error(): + + param = torch.empty(256, 128) + + n_heads = 2 + head_size = 64 + + with pytest.raises(ValueError): + shard_qkv_param(param, 0, 1, 64) + + +@pytest.mark.inference_v2 +def test_gqa_input_shape_error(): + + head_size = 64 + n_heads_q = 16 + n_heads_kv = 4 + + # Correct shape is 1536 (=16 * 64 + 2 * 4 * 64), 1024 + param = torch.empty(2048, 1024) + + with pytest.raises(ValueError): + shard_qkv_param(param, 0, 1, head_size, n_heads_q, n_heads_kv) diff --git a/tests/unit/inference/modules/__init__.py b/tests/unit/inference/modules/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/modules/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/modules/test_blas_linear_module.py b/tests/unit/inference/modules/test_blas_linear_module.py new file mode 100644 index 000000000000..18b546bab6bd --- /dev/null +++ b/tests/unit/inference/modules/test_blas_linear_module.py @@ -0,0 +1,111 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import ActivationType, DtypeEnum, is_gated +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.configs import DSLinearConfig +from deepspeed.inference.v2.modules.interfaces import DSLinearRegistry +from ..inference_test_utils import allclose + + +def reference_implementation(hidden_states: torch.Tensor, weight: torch.Tensor, bias: Optional[torch.Tensor], + act_type: ActivationType) -> torch.Tensor: + dtype = hidden_states.dtype + out_states = torch.nn.functional.linear(hidden_states, weight, bias) + out_states.float() + + if is_gated(act_type): + act_func_map = { + ActivationType.ReGLU: torch.nn.functional.relu, + ActivationType.GEGLU: lambda x: torch.nn.functional.gelu(x, approximate="tanh"), + ActivationType.SiGLU: torch.nn.functional.silu, + } + + act_act = out_states[..., ::2] + act_linear = out_states[..., 1::2] + + act_act = act_func_map[act_type](act_act) + out_states = act_act * act_linear + else: + act_func_map = { + ActivationType.RELU: torch.nn.functional.relu, + ActivationType.GELU: torch.nn.functional.gelu, + ActivationType.SILU: torch.nn.functional.silu, + ActivationType.IDENTITY: lambda x: x, + } + + out_states = act_func_map[act_type](out_states) + return out_states.to(dtype) + + +def _blas_linear_helper(tokens: int, + in_channels: int, + out_channels: int, + dtype: DtypeEnum, + act_fn: ActivationType, + use_bias: bool = True) -> None: + linear_config = DSLinearConfig(max_tokens=2048, + in_channels=in_channels, + out_channels=out_channels, + activation=act_fn, + input_dtype=dtype, + output_dtype=dtype) + + bundle = ConfigBundle(name='blas_fp_linear', config=linear_config) + + module = DSLinearRegistry.instantiate_config(bundle) + + # Input vals + hidden_states = torch.randn( + (tokens, in_channels), dtype=dtype.value, device=get_accelerator().current_device_name()) * .01 + + weight_out_channels = 2 * out_channels if is_gated(act_fn) else out_channels + weight = torch.randn( + (weight_out_channels, in_channels), dtype=dtype.value, device=get_accelerator().current_device_name()) * .01 + if use_bias: + bias = torch.randn( + (weight_out_channels), dtype=dtype.value, device=get_accelerator().current_device_name()) * .01 + else: + bias = None + + # Reference output + ref_output = reference_implementation(hidden_states, weight, bias, act_fn) + + # New output + ds_output = module(hidden_states, weight, bias) + + # Check + assert allclose(ds_output, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, in_channels, out_channels", [(1, 4608, 1728), (37, 8192, 4096), (1280, 3072, 6144)]) +def test_blas_linear_shapes(tokens: int, in_channels: int, out_channels: int) -> None: + + _blas_linear_helper(tokens, in_channels, out_channels, DtypeEnum.fp16, ActivationType.IDENTITY) + + +all_acts = [ + ActivationType.RELU, + ActivationType.GELU, + ActivationType.SILU, + ActivationType.GEGLU, + ActivationType.ReGLU, + ActivationType.SiGLU, +] + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("act_fn", all_acts) +@pytest.mark.parametrize("use_bias", [True, False]) +def test_blas_linear_act_fn(act_fn: ActivationType, use_bias: bool) -> None: + + _blas_linear_helper(283, 512, 4096, DtypeEnum.fp16, act_fn, use_bias=use_bias) diff --git a/tests/unit/inference/modules/test_blocked_attn.py b/tests/unit/inference/modules/test_blocked_attn.py new file mode 100644 index 000000000000..1f03b46bd002 --- /dev/null +++ b/tests/unit/inference/modules/test_blocked_attn.py @@ -0,0 +1,210 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import itertools + +from typing import List, Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.configs import DSSelfAttentionConfig, PositionalEmbeddingType +from deepspeed.inference.v2.modules.interfaces import DSSelfAttentionRegistry, DSSelfAttentionBase + +from ..kernels.ragged_ops.ragged_testing_utils import build_batch_and_manager +from ..inference_test_utils import allclose + +try: + from flash_attn.flash_attn_interface import flash_attn_varlen_func + validate_accuracy = True +except ImportError: + validate_accuracy = False + + +def _blocked_flash_testing_helper(head_size: int, + n_heads_q: int, + n_heads_kv: int, + seq_params: List[Tuple[int, int]], + trained_freqs: bool = None) -> None: + """ + Helper function for testing blocked flash attention. This implementation is based on + the implemnentation in ``unit.inference.kernels.ragged_ops.test_blocked_flash`` but + integrates functionality to validate the composability. + """ + if trained_freqs is None: + embed_type = PositionalEmbeddingType.none + embed_args = {} + else: + embed_type = PositionalEmbeddingType.rotate_half + if trained_freqs: + embed_args = {'trained_freqs': True} + else: + embed_args = {'trained_freqs': False} + + attn_config = DSSelfAttentionConfig(max_tokens=2048, + n_heads_q=n_heads_q, + n_heads_kv=n_heads_kv, + head_size=head_size, + max_sequences=32, + positional_embedding_type=embed_type, + positional_embedding_args=embed_args) + + config = ConfigBundle(name='dense_blocked_attention', config=attn_config) + attn_module: DSSelfAttentionBase = DSSelfAttentionRegistry.instantiate_config(config) + + kv_block_size = attn_module.kv_block_size + + kvs = [] + for _, history_len in seq_params: + if history_len > 0: + kvs.append( + torch.randn((history_len, 2 * n_heads_kv * head_size), + device=get_accelerator().current_device(), + dtype=torch.float16)) + else: + kvs.append(None) + + batch, state_manager, _ = build_batch_and_manager(seq_params, head_size, n_heads_kv, kv_block_size, kv_fill=kvs) + + qkv = torch.randn((batch.current_tokens, (n_heads_q + 2 * n_heads_kv) * head_size), + device=get_accelerator().current_device(), + dtype=torch.float16) + + kv_cache = state_manager.get_cache(0) + + attn_module.build_atoms(batch) + if not trained_freqs: + out = attn_module(qkv, kv_cache, batch) + else: + inv_freqs = torch.randn((head_size // 2, ), device=get_accelerator().current_device(), dtype=torch.float16) + out = attn_module(qkv, kv_cache, batch, inv_freqs) + + if validate_accuracy and trained_freqs is None: + cu_seqlens_q = torch.tensor([0] + list(itertools.accumulate([seq[0] for seq in seq_params])), + dtype=torch.int32, + device=get_accelerator().current_device()) + cu_seqlens_kv = torch.tensor([0] + list(itertools.accumulate([seq[1] + seq[0] for seq in seq_params])), + dtype=torch.int32, + device=get_accelerator().current_device()) + + inflight_kv = qkv[:, head_size * n_heads_q:] + full_kvs = [] + for i, kv in enumerate(kvs): + if kv is not None: + full_kvs.append(torch.cat([kv, inflight_kv[cu_seqlens_q[i]:cu_seqlens_q[i + 1]]], dim=0)) + else: + full_kvs.append(inflight_kv[cu_seqlens_q[i]:cu_seqlens_q[i + 1]]) + run_kvs = torch.cat(full_kvs, dim=0) + k = run_kvs[:, :head_size * n_heads_kv] + v = run_kvs[:, head_size * n_heads_kv:] + + q = qkv[:, :head_size * n_heads_q] + q_ref = q.reshape((batch.current_tokens, n_heads_q, head_size)) + k_ref = k.reshape((k.shape[0], n_heads_kv, head_size)) + v_ref = v.reshape((v.shape[0], n_heads_kv, head_size)) + + max_seqlen_q = max([seq[0] for seq in seq_params]) + max_seqlen_kv = max([seq[1] + seq[0] for seq in seq_params]) + + ref_o = flash_attn_varlen_func(q_ref, + k_ref, + v_ref, + cu_seqlens_q, + cu_seqlens_kv, + max_seqlen_q, + max_seqlen_kv, + softmax_scale=1.0, + causal=True) + + ref_o = ref_o.reshape(batch.current_tokens, head_size * n_heads_q) + + assert allclose(out, ref_o) + + get_accelerator().synchronize() + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("n_tokens", [2, 33, 65, 128, 256, 2037]) +def test_single_prompt(n_tokens: int) -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(n_tokens, 0)] + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("prompt_lengths", [(128, 128), (192, 38), (514, 713), (83, 312, 610)]) +def test_multiple_prompts(prompt_lengths: Tuple[int, int]) -> None: + """ + Test multiple prompts in a single batch. + """ + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(prompt_lengths[i], 0) for i in range(len(prompt_lengths))] + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("seq_params", [(1, 34), (43, 40), (1, 144), (64, 128), (332, 628)]) +def test_continuation(seq_params: Tuple[int, int]) -> None: + """ + Test continued generation/prompt processing. + """ + head_size = 64 + n_heads_q = 32 + n_heads_kv = 32 + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, [seq_params]) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("head_size", [64, 128]) +def test_head_size(head_size: int) -> None: + n_heads_q = 16 + n_heads_kv = 16 + seq_params = [(128, 128), (192, 38), (1, 814)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("head_config", [(32, 8), (64, 16), (40, 8)]) +def test_gqa(head_config: Tuple[int, int]) -> None: + head_size = 128 + n_heads_q = head_config[0] + n_heads_kv = head_config[1] + + seq_params = [(128, 128), (192, 38), (1, 814)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +def test_fully_composed() -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(332, 628), (1, 718), (1, 323), (180, 5), (224, 0)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("trained_freqs", [True, False]) +def test_rotary_emb(trained_freqs: bool) -> None: + head_size = 64 + n_heads_q = 16 + n_heads_kv = 16 + + seq_params = [(332, 628), (1, 718), (1, 323), (180, 5), (224, 0)] + + _blocked_flash_testing_helper(head_size, n_heads_q, n_heads_kv, seq_params, trained_freqs=trained_freqs) diff --git a/tests/unit/inference/modules/test_cuda_pre_ln_module.py b/tests/unit/inference/modules/test_cuda_pre_ln_module.py new file mode 100644 index 000000000000..d6c42a3e1336 --- /dev/null +++ b/tests/unit/inference/modules/test_cuda_pre_ln_module.py @@ -0,0 +1,88 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional, Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.configs import DSNormConfig +from deepspeed.inference.v2.modules.interfaces import DSPreNormRegistry +from ..inference_test_utils import get_dtypes, allclose + + +def reference_implementation(residual: torch.Tensor, hidden_states: Optional[torch.Tensor], gamma: torch.Tensor, + beta: torch.Tensor, epsilon: float) -> Tuple[torch.Tensor, torch.Tensor]: + dtype = residual.dtype + + residual = residual.to(torch.float32) + gamma = gamma.to(torch.float32) + beta = beta.to(torch.float32) + + if hidden_states is not None: + hidden_states = hidden_states.to(torch.float32) + residual = residual + hidden_states + hidden_states = torch.nn.functional.layer_norm(residual, (residual.size(-1), ), + weight=gamma, + bias=beta, + eps=epsilon) + return residual.to(dtype), hidden_states.to(dtype) + + +def _pre_ln_test_helper(n_tokens: int, n_channels: int, dtype: torch.dtype, res_add: bool = False): + config = DSNormConfig(max_tokens=2048, + type="layer_norm", + channels=n_channels, + residual_dtype=dtype, + input_dtype=dtype, + output_dtype=dtype, + eps=1e-5) + bundle = ConfigBundle(name='cuda_pre_ln', config=config) + + # Input vals + if res_add: + hidden_states = torch.randn((n_tokens, n_channels), + dtype=dtype, + device=get_accelerator().current_device_name()) + else: + hidden_states = None + + residual = torch.randn((n_tokens, n_channels), dtype=dtype, device=get_accelerator().current_device_name()) + gamma = torch.randn((n_channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + beta = torch.rand((n_channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + epsilon = 1e-5 + + # Reference output + ref_residual, ref_output = reference_implementation(residual, hidden_states, gamma, beta, epsilon) + + # New output + pre_ln_module = DSPreNormRegistry.instantiate_config(bundle) + gamma = pre_ln_module.transform_param(gamma) + beta = pre_ln_module.transform_param(beta) + + ds_residual, ds_output = pre_ln_module(residual, hidden_states, gamma, beta) + + # Check + assert allclose(ds_residual, ref_residual) + assert allclose(ds_output, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, channels", [(1, 2048), (37, 8192), (1280, 768), (2048, 5120)]) +def test_token_channels(tokens: int, channels: int) -> None: + _pre_ln_test_helper(tokens, channels, torch.float16) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("dtype", get_dtypes(include_float=False)) +def test_dtype(dtype: torch.dtype) -> None: + _pre_ln_test_helper(733, 2560, dtype) + + +@pytest.mark.inference_v2_ops +def test_no_res_add(): + _pre_ln_test_helper(733, 2560, torch.float16, res_add=False) diff --git a/tests/unit/inference/modules/test_custom_module.py b/tests/unit/inference/modules/test_custom_module.py new file mode 100644 index 000000000000..e14ccd3f2244 --- /dev/null +++ b/tests/unit/inference/modules/test_custom_module.py @@ -0,0 +1,76 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.interfaces import DSPostNormRegistry +from deepspeed.inference.v2.modules.configs import DSNormConfig +from deepspeed.inference.v2.modules.implementations import cuda_post_ln +from ..inference_test_utils import allclose + + +def reference_implementation(residual: torch.Tensor, hidden_states: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor, epsilon: float) -> torch.Tensor: + residual_f = residual.to(torch.float32) + hidden_states_f = hidden_states.to(torch.float32) + gamma_f = gamma.to(torch.float32) + beta_f = beta.to(torch.float32) + return torch.nn.functional.layer_norm(residual_f + hidden_states_f, (hidden_states_f.size(-1), ), + weight=gamma_f, + bias=beta_f, + eps=epsilon).to(hidden_states.dtype) + + +@DSPostNormRegistry.register_module +class CustomPostLNModule(cuda_post_ln.DSPostLNCUDAModule): + + @staticmethod + def name(): + return 'custom_post_ln' + + +""" +Here, we explicitly register an LN implementation outside the core deepspeed repo. This should +validate that the registry is working as expected and we can implement modules outside the core +repo. +""" + + +@pytest.mark.inference_v2_ops +def test_custom_registration(): + channels = 4096 + dtype = torch.float16 + tokens = 1024 + + config = DSNormConfig(max_tokens=2048, + type="layer_norm", + channels=channels, + residual_dtype=dtype, + input_dtype=dtype, + output_dtype=dtype, + eps=1e-5) + bundle = ConfigBundle(name='custom_post_ln', config=config) + + # Input vals + hidden_states = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + residual = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + gamma = torch.randn((channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + beta = torch.rand((channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + epsilon = 1e-5 + + # Reference output + ref_output = reference_implementation(residual, hidden_states, gamma, beta, epsilon) + + # New output + post_ln_module = DSPostNormRegistry.instantiate_config(bundle) + gamma = post_ln_module.transform_param(gamma) + beta = post_ln_module.transform_param(beta) + ds_output, _ = post_ln_module(residual, hidden_states, gamma, beta) + + # Check + assert allclose(ds_output, ref_output) diff --git a/tests/unit/inference/modules/test_cutlass_moe.py b/tests/unit/inference/modules/test_cutlass_moe.py new file mode 100644 index 000000000000..98a48b5b149d --- /dev/null +++ b/tests/unit/inference/modules/test_cutlass_moe.py @@ -0,0 +1,214 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.inference_utils import ActivationType, DtypeEnum +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.configs import DSMoEConfig +from deepspeed.inference.v2.modules.interfaces import DSMoERegistry + +from ..kernels.ragged_ops.ragged_testing_utils import build_simple_batch +from ..inference_test_utils import allclose, get_dtypes + + +def _gating_reference(logits: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]: + """ + Reference gating code. + """ + logits = logits.float() + probs = torch.nn.functional.softmax(logits, dim=1) + + indices1_s = torch.argmax(probs, dim=-1) + mask1 = torch.nn.functional.one_hot(indices1_s, num_classes=logits.shape[-1]) + indices_mask = mask1.sum(dim=1) * logits.shape[-1] - 1 + indices1_s = torch.min(indices1_s, indices_mask) + + gates1_s = (probs * mask1).sum(dim=1) + + sorted_indices = indices1_s.sort()[1] + original_indices = sorted_indices.sort()[1] + + exp_count = torch.bincount(indices1_s, minlength=logits.shape[-1]).long() + exp_count_cumsum = exp_count.cumsum(dim=0) + + return sorted_indices, original_indices, exp_count_cumsum, gates1_s + + +def _reference_impl(hidden_states: torch.Tensor, gate_weight: torch.Tensor, mlp_1_w: torch.Tensor, + mlp_2_w: torch.Tensor, mlp_1_b: torch.Tensor, mlp_2_b: torch.Tensor, + act_fn: ActivationType) -> torch.Tensor: + """ + Reference implementation of the MoE module. + """ + + act_fn_dict = { + ActivationType.GELU: torch.nn.functional.gelu, + ActivationType.RELU: torch.nn.functional.relu, + ActivationType.SILU: torch.nn.functional.silu, + ActivationType.IDENTITY: lambda x: x, + } + + logits = torch.matmul(hidden_states, gate_weight.t()) + sorted_indices, original_indices, exp_count_cumsum, gate_scales = _gating_reference(logits) + + moe_input = hidden_states[sorted_indices] + + output_unordered = torch.empty_like(hidden_states) + + for expert_idx in range(mlp_1_w.shape[0]): + min_bound = 0 if expert_idx == 0 else exp_count_cumsum[expert_idx - 1] + max_bound = exp_count_cumsum[expert_idx] + + input_slice = moe_input[min_bound:max_bound] + intermediate = torch.nn.functional.linear(input_slice, mlp_1_w[expert_idx], mlp_1_b[expert_idx]) + + intermediate = act_fn_dict[act_fn](intermediate) + output_slice = torch.nn.functional.linear(intermediate, mlp_2_w[expert_idx], mlp_2_b[expert_idx]) + + output_unordered[min_bound:max_bound] = output_slice + + output = output_unordered[original_indices] + + output.mul_(gate_scales.unsqueeze(-1)).reshape(hidden_states.shape) + return output + + +def _cutlass_moe_testing_helper(tokens: int, + in_channels: int, + intermediate_dim: int, + experts: int, + dtype: int, + activation_type: ActivationType = ActivationType.GELU, + use_bias: bool = True, + iters: int = 1) -> None: + + config = DSMoEConfig(max_tokens=4096, + model_dim=in_channels, + intermediate_features=intermediate_dim, + n_experts=experts, + activation=activation_type, + input_dtype=dtype, + output_dtype=dtype) + + implementation_config = {"weight_dtype": DtypeEnum(dtype)} + + bundle = ConfigBundle(name='cutlass_multi_gemm_moe', config=config, implementation_config=implementation_config) + moe_module = DSMoERegistry.instantiate_config(bundle) + + batch = build_simple_batch([tokens]) + + # Parameters + gate_weight = torch.randn( + (experts, in_channels), dtype=dtype.value, device=get_accelerator().current_device()) * .1 + + mlp_1_w = torch.randn( + (experts, intermediate_dim, in_channels), dtype=dtype.value, device=get_accelerator().current_device()) * .1 + mlp_2_w = torch.randn( + (experts, in_channels, intermediate_dim), dtype=dtype.value, device=get_accelerator().current_device()) * .1 + + if use_bias: + mlp_1_b = torch.randn( + (experts, intermediate_dim), dtype=dtype.value, device=get_accelerator().current_device()) * .1 + mlp_2_b = torch.randn( + (experts, in_channels), dtype=dtype.value, device=get_accelerator().current_device()) * .1 + else: + mlp_1_b = None + mlp_2_b = None + + gate_ds = moe_module.transform_gate_param(gate_weight) + mlp_1_w_ds = moe_module.transform_moe_mlp_1_param(mlp_1_w) + mlp_1_b_ds = moe_module.transform_moe_mlp_1_param(mlp_1_b) + mlp_2_w_ds = moe_module.transform_moe_mlp_2_param(mlp_2_w) + mlp_2_b_ds = moe_module.transform_moe_mlp_2_param(mlp_2_b) + + for _ in range(iters): + # Input vals + hidden_states = torch.randn( + (tokens, in_channels), dtype=dtype.value, device=get_accelerator().current_device()) * .1 + + # Reference implementation + ref_output = _reference_impl(hidden_states, gate_weight, mlp_1_w, mlp_2_w, mlp_1_b, mlp_2_b, activation_type) + + output = moe_module(hidden_states, + batch, + gate_ds, + mlp_1_w_ds, + mlp_2_w_ds, + mlp_1_b=mlp_1_b_ds, + mlp_2_b=mlp_2_b_ds) + + # Increase the tolerance for larger meta ops since the error is additive + assert allclose(output, ref_output, tolerances=(1e-2, 1e-2)) + + get_accelerator().synchronize() + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("experts", [2, 32, 64]) +def test_expert_variance(experts: int) -> None: + _cutlass_moe_testing_helper(tokens=876, + in_channels=4096, + intermediate_dim=2048, + experts=experts, + dtype=DtypeEnum.fp16, + activation_type=ActivationType.IDENTITY, + use_bias=True) + + +@pytest.mark.inference_v2_ops +def test_successive_inputs(): + """ + The CUTLASS MoE uses persistent state (expert counts) that is assumed to be cleared + on each forward pass. This ensures that the module is clearing that metadata. + """ + _cutlass_moe_testing_helper(tokens=876, + in_channels=4096, + intermediate_dim=2048, + experts=64, + dtype=DtypeEnum.fp16, + activation_type=ActivationType.IDENTITY, + use_bias=True, + iters=10) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("dtype", get_dtypes(include_float=False)) +def test_dtypes(dtype: torch.dtype) -> None: + _cutlass_moe_testing_helper(tokens=876, + in_channels=4096, + intermediate_dim=2048, + experts=64, + dtype=DtypeEnum(dtype), + activation_type=ActivationType.IDENTITY, + use_bias=True) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("activation_type", [ActivationType.GELU, ActivationType.RELU, ActivationType.SILU]) +def test_activation_types(activation_type: ActivationType) -> None: + _cutlass_moe_testing_helper(tokens=876, + in_channels=4096, + intermediate_dim=2048, + experts=64, + dtype=DtypeEnum.fp16, + activation_type=activation_type, + use_bias=True) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("in_channels, out_channels", [(4096, 2048), (2048, 8192), (6144, 3072)]) +def test_in_out_channels(in_channels: int, out_channels: int) -> None: + _cutlass_moe_testing_helper(tokens=876, + in_channels=in_channels, + intermediate_dim=out_channels, + experts=64, + dtype=DtypeEnum.fp16, + activation_type=ActivationType.IDENTITY, + use_bias=True) diff --git a/tests/unit/inference/modules/test_post_ln_module.py b/tests/unit/inference/modules/test_post_ln_module.py new file mode 100644 index 000000000000..238d8fa4d1b1 --- /dev/null +++ b/tests/unit/inference/modules/test_post_ln_module.py @@ -0,0 +1,58 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.configs import DSNormConfig +from deepspeed.inference.v2.modules.interfaces import DSPostNormRegistry +from ..inference_test_utils import get_dtypes, allclose + + +def reference_implementation(residual: torch.Tensor, hidden_states: torch.Tensor, gamma: torch.Tensor, + beta: torch.Tensor, epsilon: float) -> torch.Tensor: + residual_f = residual.to(torch.float32) + hidden_states_f = hidden_states.to(torch.float32) + gamma_f = gamma.to(torch.float32) + beta_f = beta.to(torch.float32) + return torch.nn.functional.layer_norm(residual_f + hidden_states_f, (hidden_states_f.size(-1), ), + weight=gamma_f, + bias=beta_f, + eps=epsilon).to(hidden_states.dtype) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, channels", [(1, 2048), (37, 8192), (1280, 768), (2048, 5120)]) +@pytest.mark.parametrize("dtype", get_dtypes()) +def test_cuda_post_ln_module(tokens: int, channels: int, dtype: torch.dtype) -> None: + config = DSNormConfig(max_tokens=2048, + type="layer_norm", + channels=channels, + residual_dtype=dtype, + input_dtype=dtype, + output_dtype=dtype, + eps=1e-5) + bundle = ConfigBundle(name='cuda_post_ln', config=config) + + # Input vals + hidden_states = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + residual = torch.randn((tokens, channels), dtype=dtype, device=get_accelerator().current_device_name()) + gamma = torch.randn((channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + beta = torch.rand((channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + epsilon = 1e-5 + + # Reference output + ref_output = reference_implementation(residual, hidden_states, gamma, beta, epsilon) + + # New output + post_ln_module = DSPostNormRegistry.instantiate_config(bundle) + gamma = post_ln_module.transform_param(gamma) + beta = post_ln_module.transform_param(beta) + ds_output, _ = post_ln_module(residual, hidden_states, gamma, beta) + + # Check + assert allclose(ds_output, ref_output) diff --git a/tests/unit/inference/modules/test_pre_rms_module.py b/tests/unit/inference/modules/test_pre_rms_module.py new file mode 100644 index 000000000000..bbbec2d15709 --- /dev/null +++ b/tests/unit/inference/modules/test_pre_rms_module.py @@ -0,0 +1,88 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import Optional, Tuple + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.modules import ConfigBundle +from deepspeed.inference.v2.modules.configs import DSNormConfig +from deepspeed.inference.v2.modules.interfaces import DSPreNormRegistry +from ..inference_test_utils import get_dtypes, allclose + + +def reference_implementation(residual: torch.Tensor, hidden_states: Optional[torch.Tensor], gamma: torch.Tensor, + epsilon: float) -> Tuple[torch.Tensor, torch.Tensor]: + dtype = residual.dtype + + if hidden_states is not None: + hidden_states = hidden_states + residual = residual + hidden_states + + rms_vals = residual.to(torch.float32) + variance = rms_vals.pow(2).mean(-1, keepdim=True) + rms_vals = rms_vals * torch.rsqrt(variance + epsilon) + + if gamma.dtype in [torch.float16, torch.bfloat16]: + rms_vals = rms_vals.to(gamma.dtype) + + hidden_states = gamma * rms_vals + + return residual.to(dtype), hidden_states.to(dtype) + + +def _pre_rms_test_helper(n_tokens: int, n_channels: int, dtype: torch.dtype, res_add: bool = False): + config = DSNormConfig(max_tokens=2048, + type="rms_norm", + channels=n_channels, + residual_dtype=dtype, + input_dtype=dtype, + output_dtype=dtype, + eps=1e-5) + bundle = ConfigBundle(name='cuda_pre_rms', config=config) + + # Input vals + if res_add: + hidden_states = torch.randn((n_tokens, n_channels), + dtype=dtype, + device=get_accelerator().current_device_name()) + else: + hidden_states = None + + residual = torch.randn((n_tokens, n_channels), dtype=dtype, device=get_accelerator().current_device_name()) + gamma = torch.randn((n_channels), dtype=torch.float32, device=get_accelerator().current_device_name()) + epsilon = 1e-5 + + # Reference output + ref_residual, ref_output = reference_implementation(residual, hidden_states, gamma, epsilon) + + # New output + pre_ln_module = DSPreNormRegistry.instantiate_config(bundle) + gamma = pre_ln_module.transform_param(gamma) + + ds_residual, ds_output = pre_ln_module(residual, hidden_states, gamma) + + # Check + assert allclose(ds_residual, ref_residual) + assert allclose(ds_output, ref_output) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("tokens, channels", [(1, 2048), (37, 8192), (1280, 768), (2048, 5120)]) +def test_token_channels(tokens: int, channels: int) -> None: + _pre_rms_test_helper(tokens, channels, torch.float16) + + +@pytest.mark.inference_v2_ops +@pytest.mark.parametrize("dtype", get_dtypes(include_float=False)) +def test_dtype(dtype: torch.dtype) -> None: + _pre_rms_test_helper(733, 2560, dtype) + + +@pytest.mark.inference_v2_ops +def test_no_res_add(): + _pre_rms_test_helper(733, 2560, torch.float16, res_add=False) diff --git a/tests/unit/inference/ragged/__init__.py b/tests/unit/inference/ragged/__init__.py new file mode 100644 index 000000000000..208299fb8c50 --- /dev/null +++ b/tests/unit/inference/ragged/__init__.py @@ -0,0 +1,4 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team diff --git a/tests/unit/inference/ragged/test_blocked_allocator.py b/tests/unit/inference/ragged/test_blocked_allocator.py new file mode 100644 index 000000000000..4596e81c5652 --- /dev/null +++ b/tests/unit/inference/ragged/test_blocked_allocator.py @@ -0,0 +1,166 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import random +from typing import List + +import pytest +import torch + +from deepspeed.inference.v2.ragged.blocked_allocator import BlockedAllocator + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('bad_size', [0, -1]) +def test_bad_initialization(bad_size: int) -> None: + with pytest.raises(ValueError): + BlockedAllocator(bad_size) + + +@pytest.mark.inference_v2 +def test_allocation() -> None: + + allocator = BlockedAllocator(16) + + a1 = allocator.allocate(4) + assert a1.numel() == 4 + assert allocator.free_blocks == 12 + + a2_allocs = [] + for i in range(3): + a2_allocs.append(allocator.allocate(2)) + assert allocator.free_blocks == 12 - (i + 1) * 2 + + a3 = allocator.allocate(6) + assert a3.numel() == 6 + + assert allocator.free_blocks == 0 + + # Test that we can't allocate more blocks than we have. + with pytest.raises(ValueError): + allocator.allocate(1) + + all_vals = torch.cat([a1, *a2_allocs, a3], dim=0) + unique_vals = torch.unique(all_vals, sorted=False) + assert unique_vals.numel() == all_vals.numel() + + +@pytest.mark.inference_v2 +def test_too_large_allocation(): + allocator = BlockedAllocator(16) + + with pytest.raises(ValueError): + allocator.allocate(17) + + +@pytest.mark.inference_v2 +def test_deallocation() -> None: + allocator = BlockedAllocator(16) + + # Allocate + all_blocks = allocator.allocate(16) + assert allocator.free_blocks == 0 + + # Deallocate all blocks + allocator.free(all_blocks) + assert allocator.free_blocks == 16 + + # Get all the blocks again + all_blocks = allocator.allocate(16) + + # Deallocate in chunks + c1 = all_blocks[:4] + c2 = all_blocks[4:8] + + allocator.free(c1) + assert allocator.free_blocks == 4 + + allocator.free(c2) + assert allocator.free_blocks == 8 + + with pytest.raises(ValueError): + allocator.free(c1) + + with pytest.raises(ValueError): + allocator.free(c2) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('index', [-1, 2]) +def test_invalid_dealloc_indices(index: int): + allocator = BlockedAllocator(1) + + with pytest.raises(ValueError): + allocator.free(torch.tensor([index])) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('index', [-1, 2]) +def test_invalid_alloc_indices(index: int): + allocator = BlockedAllocator(1) + allocator.allocate(1) + + to_free = [0, index] + + with pytest.raises(ValueError): + allocator.free(torch.tensor(to_free)) + + # Block 0 should not be freed if passed with an invalid index. + assert allocator.free_blocks == 0 + + allocator.free(torch.tensor([0])) + assert allocator.free_blocks == 1 + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('test_iters', [8192]) +def test_long_running_allocation(test_iters: int) -> None: + """ + Evaluate the stability of the allocator over a longer sequence of allocations/deallocations. + """ + TOTAL_BLOCKS = 128 + + allocator = BlockedAllocator(TOTAL_BLOCKS) + + def validate_uniqueness(all_blocks: List[torch.Tensor]) -> None: + all_vals = torch.cat(all_blocks, dim=0) + assert all_vals.numel() <= TOTAL_BLOCKS + + unique_vals = torch.unique(all_vals, sorted=False) + assert unique_vals.numel() == all_vals.numel() + + all_allocs: List[torch.Tensor] = [] + num_allocs = 0 + num_frees = 0 + num_blocks_allocated = 0 + num_blocks_freed = 0 + + for _ in range(test_iters): + decision = random.randint(0, 1) + + if decision == 0: + blocks_to_allocate = random.randint(1, 24) + if blocks_to_allocate > allocator.free_blocks: + with pytest.raises(ValueError): + allocator.allocate(blocks_to_allocate) + else: + all_allocs.append(allocator.allocate(blocks_to_allocate)) + num_allocs += 1 + num_blocks_allocated += blocks_to_allocate + else: + if len(all_allocs) > 0: + idx = random.randint(0, len(all_allocs) - 1) + allocator.free(all_allocs[idx]) + + num_frees += 1 + num_blocks_freed += all_allocs[idx].numel() + + del all_allocs[idx] + + if len(all_allocs) > 0: + validate_uniqueness(all_allocs) + + assert num_allocs == num_frees + len(all_allocs) + assert num_blocks_allocated == num_blocks_freed + (TOTAL_BLOCKS - allocator.free_blocks) diff --git a/tests/unit/inference/ragged/test_manager_configs.py b/tests/unit/inference/ragged/test_manager_configs.py new file mode 100644 index 000000000000..bdd513445ddb --- /dev/null +++ b/tests/unit/inference/ragged/test_manager_configs.py @@ -0,0 +1,58 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +import pytest + +from pydantic import ValidationError + +from deepspeed.inference.v2.ragged import DSStateManagerConfig + + +@pytest.mark.inference_v2 +def test_negative_max_tracked_sequences() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_tracked_sequences=-1) + + +@pytest.mark.inference_v2 +def test_zero_max_tracked_sequences() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_tracked_sequences=0) + + +@pytest.mark.inference_v2 +def test_negative_max_ragged_batch_size() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_ragged_batch_size=-1) + + +@pytest.mark.inference_v2 +def test_zero_max_ragged_batch_size() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_ragged_batch_size=0) + + +@pytest.mark.inference_v2 +def test_negative_max_ragged_sequence_count() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_ragged_sequence_count=-1) + + +@pytest.mark.inference_v2 +def test_zero_max_ragged_sequence_count() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_ragged_sequence_count=0) + + +@pytest.mark.inference_v2 +def test_too_small_max_ragged_batch_size() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_ragged_batch_size=512, max_ragged_sequence_count=1024) + + +@pytest.mark.inference_v2 +def test_too_small_max_tracked_sequences() -> None: + with pytest.raises(ValidationError): + DSStateManagerConfig(max_tracked_sequences=512, max_ragged_sequence_count=1024) diff --git a/tests/unit/inference/ragged/test_ragged_wrapper.py b/tests/unit/inference/ragged/test_ragged_wrapper.py new file mode 100644 index 000000000000..3cb74f4c49d2 --- /dev/null +++ b/tests/unit/inference/ragged/test_ragged_wrapper.py @@ -0,0 +1,112 @@ +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +from typing import List + +import pytest +import torch + +from deepspeed.accelerator import get_accelerator +from deepspeed.inference.v2.ragged import ( + PlaceholderSequenceDescriptor, + RaggedBatchWrapper, + DSStateManagerConfig, +) + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('max_ragged_sequence_count, max_ragged_batch_size', [(128, 512), (128, 1024)]) +def test_wrapper_initialization(max_ragged_sequence_count: int, max_ragged_batch_size: int) -> None: + config = DSStateManagerConfig(max_tracked_sequences=max_ragged_sequence_count, + max_ragged_batch_size=max_ragged_batch_size, + max_ragged_sequence_count=max_ragged_sequence_count) + + batch = RaggedBatchWrapper(config) + + assert batch.current_tokens == 0 + assert batch.current_sequences == 0 + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('seq_len', [1, 37, 128, 512]) +def test_single_sequence_batch(seq_len: int) -> None: + """ + Test we successfully construct single sequence batches and the on device metadata is accurate. + """ + + config = DSStateManagerConfig() + batch = RaggedBatchWrapper(config) + + batch.clear() + + assert batch.current_tokens == 0 + assert batch.current_sequences == 0 + + seq_desc = PlaceholderSequenceDescriptor() + tokens = torch.randint(0, 100, (seq_len, )) + batch.insert_sequence(seq_desc, tokens) + + batch.finalize() + + assert batch.current_tokens == seq_len + assert batch.current_sequences == 1 + assert torch.equal(batch.input_ids(), tokens.to(get_accelerator().current_device())) + assert torch.equal(batch.tokens_to_seq(), torch.zeros_like(tokens, device=get_accelerator().current_device())) + assert torch.equal(batch.batch_metadata_buffer(), + torch.tensor([seq_len, 1], device=get_accelerator().current_device())) + + batch.clear() + + assert batch.current_tokens == 0 + assert batch.current_sequences == 0 + + +@pytest.mark.inference_v2 +@pytest.mark.parametrize('seq_lens', [[128, 128], [1, 32, 243], [64, 1, 1, 1, 1, 393, 27, 2]]) +def test_multi_sequence_batch(seq_lens: List[int]) -> None: + """ + Test sequentially adding new tokens to a batch and validate device data structures hold + the appropriate data. + """ + config = DSStateManagerConfig() + batch = RaggedBatchWrapper(config) + + batch.clear() + + assert batch.current_tokens == 0 + assert batch.current_sequences == 0 + + all_toks = [torch.randint(0, 100, (seq_len, )) for seq_len in seq_lens] + + for i, toks in enumerate(all_toks): + seq_desc = PlaceholderSequenceDescriptor() + batch.insert_sequence(seq_desc, toks) + + assert batch.current_tokens == sum(seq_lens[:i + 1]) + assert batch.current_sequences == i + 1 + + batch.finalize() + + assert batch.current_tokens == sum(seq_lens) + assert batch.current_sequences == len(seq_lens) + + assert torch.equal(batch.input_ids(), torch.cat(all_toks, dim=0).to(get_accelerator().current_device())) + assert torch.equal( + batch.tokens_to_seq(), + torch.cat([torch.full((seq_len, ), i, dtype=torch.int32) for i, seq_len in enumerate(seq_lens)], + dim=0).to(get_accelerator().current_device())) + + for i, seq_len in enumerate(seq_lens): + assert batch.inflight_seq_descriptors()[i][0] == sum(seq_lens[:i]) + assert batch.inflight_seq_descriptors()[i][1] == seq_len + assert batch.inflight_seq_descriptors()[i][2] == 0 + + assert torch.equal(batch.batch_metadata_buffer(), + torch.tensor([sum(seq_lens), len(seq_lens)], device=get_accelerator().current_device())) + + batch.clear() + + assert batch.current_tokens == 0 + assert batch.current_sequences == 0 From 1d9e256c03ef1a04b5a9fbfd8ebf48b189849904 Mon Sep 17 00:00:00 2001 From: Jeff Rasley Date: Fri, 3 Nov 2023 15:32:40 -0700 Subject: [PATCH 04/13] DeepSpeed-FastGen blog (#4607) Co-authored-by: Connor Holmes Co-authored-by: Masahiro Tanaka Co-authored-by: Heyang Qin Co-authored-by: Michael Wyatt Co-authored-by: Ammar Ahmad Awan Co-authored-by: Lev Kurilenko --- README.md | 1 + blogs/deepspeed-fastgen/README.md | 300 ++++++++++++++++++ .../assets/images/A6000_benchmark.png | Bin 0 -> 203812 bytes .../assets/images/H100_benchmark.png | Bin 0 -> 176743 bytes .../assets/images/effective_throughput.png | Bin 0 -> 222683 bytes .../images/effective_throughput_main.png | Bin 0 -> 182665 bytes .../assets/images/fast-gen-overview.jpg | Bin 0 -> 277661 bytes .../assets/images/fastgen-arch-dark.png | Bin 0 -> 179085 bytes .../assets/images/fastgen-arch-light.png | Bin 0 -> 176270 bytes .../assets/images/fastgen-hero-dark.png | Bin 0 -> 146609 bytes .../assets/images/fastgen-hero-light.png | Bin 0 -> 131148 bytes .../assets/images/fastgen-overview-dark.png | Bin 0 -> 232115 bytes .../assets/images/fastgen-overview-light.png | Bin 0 -> 222753 bytes .../images/observation-prompt-v-flops.png | Bin 0 -> 48387 bytes .../images/observation-prompt-v-latency.png | Bin 0 -> 91276 bytes .../repl_scale_llama70b_tp4_p2600g60.png | Bin 0 -> 28961 bytes .../th_lat_curve_llama70b_tp4_p1200g128.png | Bin 0 -> 37807 bytes .../th_lat_curve_llama70b_tp4_p2600g128.png | Bin 0 -> 36645 bytes .../assets/images/throughput_latency.png | Bin 0 -> 340539 bytes .../assets/images/throughput_latency_13B.png | Bin 0 -> 339033 bytes .../assets/images/token_latency.png | Bin 0 -> 60148 bytes docs/index.md | 1 + 22 files changed, 302 insertions(+) create mode 100644 blogs/deepspeed-fastgen/README.md create mode 100644 blogs/deepspeed-fastgen/assets/images/A6000_benchmark.png create mode 100644 blogs/deepspeed-fastgen/assets/images/H100_benchmark.png create mode 100644 blogs/deepspeed-fastgen/assets/images/effective_throughput.png create mode 100644 blogs/deepspeed-fastgen/assets/images/effective_throughput_main.png create mode 100644 blogs/deepspeed-fastgen/assets/images/fast-gen-overview.jpg create mode 100644 blogs/deepspeed-fastgen/assets/images/fastgen-arch-dark.png create mode 100644 blogs/deepspeed-fastgen/assets/images/fastgen-arch-light.png create mode 100755 blogs/deepspeed-fastgen/assets/images/fastgen-hero-dark.png create mode 100755 blogs/deepspeed-fastgen/assets/images/fastgen-hero-light.png create mode 100755 blogs/deepspeed-fastgen/assets/images/fastgen-overview-dark.png create mode 100755 blogs/deepspeed-fastgen/assets/images/fastgen-overview-light.png create mode 100644 blogs/deepspeed-fastgen/assets/images/observation-prompt-v-flops.png create mode 100644 blogs/deepspeed-fastgen/assets/images/observation-prompt-v-latency.png create mode 100644 blogs/deepspeed-fastgen/assets/images/repl_scale_llama70b_tp4_p2600g60.png create mode 100644 blogs/deepspeed-fastgen/assets/images/th_lat_curve_llama70b_tp4_p1200g128.png create mode 100644 blogs/deepspeed-fastgen/assets/images/th_lat_curve_llama70b_tp4_p2600g128.png create mode 100644 blogs/deepspeed-fastgen/assets/images/throughput_latency.png create mode 100644 blogs/deepspeed-fastgen/assets/images/throughput_latency_13B.png create mode 100644 blogs/deepspeed-fastgen/assets/images/token_latency.png diff --git a/README.md b/README.md index bea26ea1828a..721ba62cee37 100755 --- a/README.md +++ b/README.md @@ -15,6 +15,7 @@ ## Latest News DeepSpeed empowers ChatGPT-like model training with a single click, offering 15x speedup over SOTA RLHF systems with unprecedented cost reduction at all scales; [learn how](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-chat). +* [2023/11] [DeepSpeed-FastGen: High-throughput Text Generation for LLMs via MII and DeepSpeed-Inference](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-fastgen) * [2023/10] [DeepSpeed-VisualChat: Improve Your Chat Experience with Multi-Round Multi-Image Inputs](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-visualchat/10-03-2023/README.md) [[English](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-visualchat/10-03-2023/README.md)] [[中文](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed-visualchat/10-03-2023/README-Chinese.md)] [[日本語](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed-visualchat/10-03-2023/README-Japanese.md)] * [2023/09] Announcing the DeepSpeed4Science Initiative: Enabling large-scale scientific discovery through sophisticated AI system technologies [[DeepSpeed4Science website](https://deepspeed4science.ai/)] [[Tutorials](https://www.deepspeed.ai/deepspeed4science/)] [[White paper](https://arxiv.org/abs/2310.04610)] [[Blog](https://www.microsoft.com/en-us/research/blog/announcing-the-deepspeed4science-initiative-enabling-large-scale-scientific-discovery-through-sophisticated-ai-system-technologies/)] [[中文](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed4science/chinese/README.md)] [[日本語](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed4science/japanese/README.md)] * [2023/08] [DeepSpeed ZeRO-Inference: 20X faster inference through weight quantization and KV cache offloading](https://github.com/microsoft/DeepSpeedExamples/blob/master/inference/huggingface/zero_inference/README.md) diff --git a/blogs/deepspeed-fastgen/README.md b/blogs/deepspeed-fastgen/README.md new file mode 100644 index 000000000000..ad6bfda16626 --- /dev/null +++ b/blogs/deepspeed-fastgen/README.md @@ -0,0 +1,300 @@ +
+ +# DeepSpeed-FastGen: High-throughput Text Generation for LLMs via MII and DeepSpeed-Inference + +
+ +
+ + +
+ +## Table of Contents +1. [Introduction](#introduction) +2. [Key LLM Serving Techniques](#background) +3. [Dynamic SplitFuse: A Novel Prompt and Generation Composition Strategy](#technical-approach) +4. [Performance Evaluation](#performance-evaluation) +5. [DeepSpeed-FastGen: Implementation and Usage](#using-deepspeed-fastgen) +6. [Try out DeepSpeed-FastGen](#try) +7. [Acknowledgements](#acknowledgements) + + +## 1. Introduction + +Large language models (LLMs) like GPT-4 and LLaMA have emerged as a dominant workload in serving a wide range of applications infused with AI at every level. From general chat models to document summarization, and from autonomous driving to copilots at every layer of the software stack, the demand to deploy and serve these models at scale has skyrocketed. While frameworks like DeepSpeed, PyTorch, and several others can regularly achieve good hardware utilization during LLM training, the interactive nature of these applications and the poor arithmetic intensity of tasks like open-ended text generation have become the bottleneck for inference throughput in existing systems. + +To this end, frameworks like [vLLM](https://arxiv.org/pdf/2309.06180.pdf) powered by PagedAttention and research systems like [Orca](https://www.usenix.org/system/files/osdi22-yu.pdf) have significantly improved the performance of inference for LLMs. However, these systems still struggle to provide consistent quality of service, particularly for workloads with longer prompts. These long prompt workloads are becoming increasingly important as more and more models, like [MPT-StoryWriter](https://www.mosaicml.com/blog/mpt-7b), and systems, such as [DeepSpeed Ulysses](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-ulysses), support context windows stretching to tens of thousands of tokens. To better understand the problem space, we provide detailed examples of how text generation works for LLMs in two distinct phases called prompt processing and generation. When systems treat them as distinct phases, generation will be preempted by prompt processing that risks breaking the service level agreements (SLAs). + +Today, we are glad to present DeepSpeed-FastGen, a system that overcomes these limitations by leveraging the proposed Dynamic SplitFuse technique and offers up to 2.3x higher effective throughput compared to state-of-the-art systems like vLLM. DeepSpeed-FastGen leverages the combination of DeepSpeed-MII and DeepSpeed-Inference to provide an easy-to-use serving system. + +**Quick Start:** Trying DeepSpeed-FastGen is as simple as installing the latest [DeepSpeed-MII](https://github.com/microsoft/DeepSpeed-MII) release: + +```bash +pip install deepspeed-mii +``` + +To generate text using a simple non-persistent pipeline deployment, run the following code. For more details, please see [Section 5](#using-deepspeed-fastgen). + +```python +from mii import pipeline +pipe = pipeline("mistralai/Mistral-7B-v0.1") +output = pipe(["Hello, my name is", "DeepSpeed is"], max_new_tokens=128) +print(output) +``` + +## 2. Existing LLM Serving Techniques in Literature + +A text generation workload for a single sequence consists of two phases: 1) prompt processing, in which the user-provided text is efficiently processed as a batch of tokens to build a key-value (KV) cache for attention, and 2) token generation, which will add a single token to that cache and generate a new token. Over the course of generating a sequence of text, the model will make many forward calls to the model to generate the full sequence of text. Two major techniques have been proposed in the literature and deployed in systems that address various limitations and bottlenecks that may arise during these phases. + +_ Blocked KV Caching: _ + +vLLM identified that memory fragmentation due to large monolithic KV-caches significantly reduced the concurrency of LLM serving systems and proposed [Paged Attention](https://arxiv.org/pdf/2309.06180.pdf) to enable non-contiguous caches and increase total system throughput. Rather than assign individual variable-sized contiguous chunks of memory, the underlying storage in the KV cache is fixed-sized blocks (also known as pages). The blocked KV-cache increases system throughput by increasing the amount of potential sequence concurrency by eliminating KV-cache induced memory fragmentation. Non-contiguous KV cache implementations are also included in [HuggingFace TGI](https://github.com/huggingface/text-generation-inference) and [NVIDIA TensorRT-LLM](https://github.com/NVIDIA/TensorRT-LLM). + +_ Continuous Batching: _ + +In the past, dynamic batching, in which a server would wait for multiple requests to process in phase with each other, was used to improve GPU utilization. However, this approach has drawbacks, as it typically requires padding inputs to identical lengths or stalling the system to wait to construct a larger batch. + +Recent advancement in large language model (LLM) inference and serving has been focusing on fine granularity scheduling and optimizing memory efficiency. For instance, Orca proposes _iteration-level scheduling_ (also known as continuous batching) which makes distinct scheduling decisions at each forward pass of the model. This allows requests to join/leave the batch as needed, eliminating the need for padding requests thus improving the overall throughput. In addition to Orca, continuous batching has been implemented in NVIDIA TRT-LLM, HuggingFace TGI, and vLLM. + +In current systems, there are two primary approaches to implement continuous batching. In TGI and vLLM, the generation phase is preempted to perform prompt processing (called infill in TGI) before continuing with generation. In Orca, these phases are not distinguished; instead, Orca will add a prompt into the running batch so long as the total number of sequences doesn't reach a fixed bound. Both of these approaches to varying degrees need to stall generation to process long prompts (see [Section 3B](#splitfuse)). + +To address these shortcomings, we propose a novel prompt and generation composition strategy, Dynamic SplitFuse. + +## 3. Dynamic SplitFuse: A Novel Prompt and Generation Composition Strategy + +DeepSpeed-FastGen is built to leverage continuous batching and non-contiguous KV caches to enable increased occupancy and higher responsivity for serving LLMs in the data center, similar to existing frameworks such as TRT-LLM, TGI, and vLLM. In order to achieve a new level of performance, DeepSpeed-FastGen introduces SplitFuse which leverages dynamic prompt and generation decomposition and unification to further improve continuous batching and system throughput. + +### A. Three Performance Insights +Before describing Dynamic SplitFuse, we answer three key performance questions that together motivate its design. + +*__1. What factors impact the forward pass of a single LLM?__* In order to effectively schedule, it is necessary to understand what are the relevant independent variables the scheduling loop should control. We observe below that the composition of sequences in a forward pass (the batch size in sequences) has a negligible impact on performance compared to the raw number of tokens in the forward pass. This means an effective scheduler can be built around a single signal, the number of tokens in the forward pass. + +
+
+
+ +*__2. How does a model's throughput respond to changing the number of tokens in the forward pass?__* An LLM has two key operating regions with a relatively steep transition. With a small number of tokens, the GPU bottleneck is reading the model from memory and so throughput scales with the number of tokens, whereas with many tokens the model is throughput bound by compute and sees near-constant throughput. The model should run highly efficiently if all forward passes are in the throughput-saturating region. + +
+
+
+ +*__3. How should a pool of tokens be scheduled across multiple forward passes?__* We observe above that for well-aligned inputs the token-throughput curve is concave, which means the second derivative is bound to be less than or equal to 0. As an example, let $f(x)$ be a concave function of latency to throughput for a given model. For a concave function $f(x)$, the following holds: + + $$0 \geq \lim_{h \to 0} \frac{f(x + h) - 2f(x) + f(x - h)}{h^2}$$ + + $$0 \geq f(x + h) - 2f(x) + f(x - h)$$ + + $$2f(x) \geq f(x + h) + f(x - h)$$ + +This states that for a given pool of `2x` tokens to process, the manner that maximizes throughput is that which evenly splits them between two batches. More generally, in a system that must consume and process P tokens over F forward passes, the ideal partitioning scheme will divide them equally. + +### B. Dynamic SplitFuse + +Dynamic SplitFuse is a novel token composition strategy for prompt processing and token generation. DeepSpeed-FastGen utilizes Dynamic SplitFuse to run at a consistent forward size by leveraging the capability to take partial tokens from prompts and compose this with generation. In particular, Dynamic SplitFuse performs two key behaviors: + +1. Long prompts are decomposed into much smaller chunks and scheduled across multiple forward passes (iterations) with only the final pass performing any generation. +2. Short prompts will be composed to exactly fill a target token budget. Even short prompts may be decomposed to ensure the budget is precisely met and the forward sizes are well-aligned. + +Together, these two techniques provide concrete benefits on all user metrics: + +1. *__Better Responsiveness__:* Since long prompts no longer require extremely long forward passes to process, the model will provide lower client latency. More forward passes are performed within the same window of time. +2. *__Higher Efficiency:__* Fusion of short prompts to larger token budgets enables the model to consistently operate in the high throughput regime. +3. *__Lower variance and better consistency:__* Since forward passes are of consistent size and forward pass size is the primary determinant of performance, the latency of each forward pass is much more consistent than competing systems as is the perceived generation frequency. There are no pre-emption or long-running prompts to increase the latency as in other prior work. + +Consequently, DeepSpeed-FastGen will consume tokens from incoming prompts at a rate that permits fast ongoing generation while adding tokens to the system that increase system utilization, providing lower latency and higher throughput streaming generation to all clients as compared to other state-of-the-art serving systems. + +
+ +
+ + *Figure 1: Illustration of continuous batching strategies. Each block shows the execution of a forward pass. An arrow indicates that the forward pass has sequences with one or more tokens generated. vLLM performs either token generations or prompt processing in a forward pass; token generation preempts prompt processing. Orca runs prompts at their complete length alongside generation. Dynamic SplitFuse performs dynamic composition of fixed-sized batches composed of both generation and prompt tokens.* + +
+ +## 4. Performance Evaluation + +DeepSpeed-FastGen provides state-of-the-art LLM serving performance leveraging its blocked KV cache and Dynamic SplitFuse continuous batching. We evaluate DeepSpeed-FastGen against vLLM on a range of models and hardware configurations following the benchmarking methodology discussed below. + +### A. Benchmarking Methodology + +We use two primary quantitative schemes for measuring performance. + +**Throughput-Latency Curves:** Two key metrics for production readiness are throughput (measured in requests per second) and latency (the responsiveness of each request). To measure this, we instantiate multiple clients (ranging from 1 to 32) concurrently and send requests (512 in total) to the server. The resulting latency of each request is measured at the endpoint and throughput is measured by the end-to-end time to complete the experiment. + +**Effective Throughput:** Interactive applications, such as chat applications, can have more stringent and complex requirements than can be captured by top-level metrics like end-to-end latency. In particular, we focus on the increasingly popular chat user scenario: + + 1. A user initiates a task by sending a prompt. + 2. The system processes the prompt and returns the first token. + 3. Subsequent tokens are streamed to the user as they are produced. + +At each point in this process there is an opportunity for a system to provide an adverse user experience; for example, if the first token arrives too slowly or the generation appears to stop for some time. We propose an SLA framework that considers both of these dimensions. + +As the lengths of prompts and generated texts vary significantly, affecting computational costs, it is impractical to set rigid SLA values for throughput and latency. Therefore, we define the SLA for prompt latency as |tokens in prompt| / 512 seconds (= 512 tokens/s). Additionally, considering humans' reading speed, we set the SLA for generation latency on the Exponential Moving Average (EMA) to 2, 4, or 6 tokens/sec. Requests that adhere to these SLAs are deemed successful, and the throughput of these successful requests is referred to as **effective throughput**. + +We evaluate vLLM and DeepSpeed-FastGen on both Llama-2 7B, Llama-2 13B, and Llama-2 70B on NVIDIA A100, H100, and A6000. + +### B. Throughput-Latency Analysis + +In this experiment, DeepSpeed-FastGen outperforms vLLM in both throughput and latency, providing equivalent latency with greater throughput or more responsive latency and the same throughput. On Llama-2 70B with 4 A100x80GB, DeepSpeed-FastGen demonstrates up to 2x higher throughput (1.36 rps vs. 0.67 rps) at identical latency (9 seconds) or up to 50% latency reduction (7 seconds vs. 14 seconds) while achieving the same throughput (1.2 rps), as shown in Figure 2. These trends hold when evaluating Llama-2 13B as shown in Figure 3. + +
+
+ + *Figure 2: Throughput and latency of text generation using Llama 2 70B (Tensor parallelism across 4 A100-80GB GPUs). A normal distribution was applied to prompt and generation lengths with averages of 1200/2600 and 128/60, respectively, and a 30% variance* +

+ +
+
+ + *Figure 3: Throughput and latency of text generation using Llama 2 13B (A100-80GB GPU, no tensor parallelism). A normal distribution was applied to prompt and generation lengths with averages of 1200/2600 and 60/128, respectively, and a 30% variance* +
+ +### C. Effective Throughput Analysis + +Under the effective throughput analysis that considers both first token latency and the rate at which generation occurs, DeepSpeed-FastGen provides up to 2.3x higher throughput than vLLM. Figure 4 presents a comparative analysis of the effective throughputs of DeepSpeed-FastGen and vLLM. Each plotted point denotes the effective throughput derived from a specific number of clients. As we scaled the number of clients, we initially observed an increase in effective throughput. However, the latency also significantly increases as the number of clients approaches the system's capacity, causing many requests to fail in meeting the SLA. Consequently, the effective throughput will either saturate or decrease at some point. From a usability perspective, it's not particularly relevant how many clients are required to achieve the max effective throughput; the maximum point of the line is the optimal serving point. + +
+ + + *Figure 4: Effective throughput of DeepSpeed-FastGen and vLLM (Llama 2 70B/A100-80GB using tensor parallelism across 4 A100-80GB GPUs. A normal distribution was applied to prompt and generation lengths with averages of 2600 and 60, respectively, and a 30% variance)* +

+ +When vLLM preempts the ongoing generation of previous requests, the generation latency experiences a notable increase. This leads to vLLM's effective throughput appearing lower than its directly measured throughput. At vLLM's peak, the effective throughput was 0.63 queries/sec and around 28% of requests failed to meet the 4 tokens/s SLA. At the same SLA, DeepSpeed-FastGen achieved 1.42 queries/sec (less than 1% of requests failed to meet the SLA), which is 2.3x higher than vLLM. + +### D. Token Level Timing Analysis + +Figure 5 displays the P50, P90, and P95 latencies of the generation processes. Both vLLM and DeepSpeed-FlexGen exhibit similar P50 latencies, but vLLM demonstrates significantly higher latencies for P90 and P95. + +This discrepancy is due to a noticeable spike in vLLM's generation latency when it preempts the ongoing generation to process new prompts. +In contrast, DeepSpeed-FastGen typically processes the prompt and generation for previous requests concurrently, leading to much more consistent generation latency. + +
+
+ + *Figure 5: Per-Token generation Latency of Llama 2 70B/A100-80GB using tensor parallelism across 4 A100-80GB GPUs, 16 clients. A normal distribution was applied to prompt and generation lengths with averages of 2600 and 128, respectively, and a 30% variance.* +

+ + +### E. Scalability using Load Balancing + +DeepSpeed-FastGen offers replica-level load balancing that evenly distributes requests across multiple servers, allowing you to effortlessly scale up your application. + +Figure 6 illustrates the scalability of DeepSpeed-FastGen when employing the load balancer and up to 16 replicas. Note that we utilized 4 A100 GPUs to compute the Llama 2 70B model. In total, we employed 8 nodes to run the 16 replicas. The results demonstrate nearly perfect scalability with DeepSpeed-FastGen. +Given that the throughput of a single replica is 1.46 queries/sec, the throughput with 16 replicas reaches 23.7 queries/sec, marking a linear 16x increase compared to a single replica. + +
+
+ + *Figure 6: Scalability using the load balancing feature. A normal distribution was applied to prompt and generation lengths with averages of 2600 and 60, respectively, and a 30% variance*
+
+ +### F. Other Hardware Platforms + +In addition to the deep analysis on A100, we provide additional benchmarking results for H100 and A6000. The same performance trends were observed on both A6000 and H100 as A100. + +
+
+ + *Figure 7: Throughput-latency curve and effective throughput of Llama 2 70b using 8 H100 GPUs. A normal distribution was applied to prompt and generation lengths with averages of 2600 and 60, respectively, and a 30% variance*
+
+ +
+
+ + *Figure 8: Throughput-latency curve and effective throughput of Llama 2 7b using A6000. A normal distribution was applied to prompt and generation lengths with averages of 2600 and 60, respectively, and a 30% variance*
+
+ +## 5. DeepSpeed-FastGen: Implementation and Usage + +DeepSpeed-FastGen is the synergistic composition of [DeepSpeed-MII](https://github.com/microsoft/DeepSpeed-MII) and [DeepSpeed-Inference](https://github.com/microsoft/DeepSpeed) as illustrated in the figure below. Together, both of these software packages provide various components of the system including the frontend APIs, the host and device infrastructure to schedule batches using Dynamic SplitFuse, optimized kernel implementations, and the tools to construct new model implementations. + +
+ + +
+ + +The fastest way to get started with our alpha release of DeepSpeed-FastGen is: `pip install deepspeed-mii`. + +Please follow our [Getting Started](https://github.com/microsoft/deepspeed-mii#getting-started-with-mii) guide for more details. For usage and reporting issues, please use the [DeepSpeed-MII Github repository](https://github.com/microsoft/DeepSpeed-MII). + +### A. Supported Models + +We currently support the following model architectures in this alpha release of DeepSpeed-FastGen: + +* [LLaMA](https://huggingface.co/models?other=llama) and [LLaMA-2](https://huggingface.co/models?other=llama-2) +* [Mistral](https://huggingface.co/models?other=mistral) +* [OPT](https://huggingface.co/models?other=opt) + +All current models leverage [HuggingFace](https://github.com/huggingface) APIs in our backend to provide both the model weights and the model's corresponding tokenizer. + +We plan to add additional models in the coming weeks and months after the initial release. If there are specific model architectures you would like supported, please [file an issue](https://github.com/microsoft/DeepSpeed-MII/issues) and let us know. + +### B. Deployment options +All of the examples below are runnable in [DeepSpeedExamples](https://github.com/microsoft/DeepSpeedExamples/tree/master/inference/mii). Once installed you have two options for deployment: an interactive non-persistent pipeline or a persistent serving deployment: + +#### Non-persistent pipeline + +The non-persistent pipeline deployment is a great and fast way to get started and can be done with only a few lines of code. Non-persistent models are only around for the duration of the python script you are running but are useful for temporary interactive sessions. + +```python +from mii import pipeline +pipe = pipeline("mistralai/Mistral-7B-v0.1") +output = pipe(["Hello, my name is", "DeepSpeed is"], max_new_tokens=128) +print(output) +``` + +#### Persistent deployment + +A persistent deployment is ideal for use with long-running and production applications. The persistent deployment uses a lightweight GRPC server that can be created using the following 2 lines: + + +```python +import mii +mii.serve("mistralai/Mistral-7B-v0.1") +``` + +The above server can be queried by multiple clients at once thanks to the built-in load balancer from DeepSpeed-MII. Creating a client also just takes 2 lines of code: + +```python +client = mii.client("mistralai/Mistral-7B-v0.1") +output = client.generate("Deepspeed is", max_new_tokens=128) +print(output) +``` + +A persistent deployment can be terminated when it is no longer needed: + +```python +client.terminate_server() +``` + +### C. Advanced Installation Information + +For ease of use and a significant reduction in lengthy compile times that many projects require in this space, we distribute a pre-compiled Python wheel covering the majority of our custom kernels through a new library called [DeepSpeed-Kernels](https://github.com/microsoft/DeepSpeed-Kernels). We have found this library to be very portable across environments with NVIDIA GPUs with compute capabilities 8.0+ (Ampere+), CUDA 11.6+, and Ubuntu 20+. In most cases, you shouldn't even need to know this library exists as it is a dependency of DeepSpeed-MII and will be installed with it. However, if for whatever reason you need to compile our kernels manually please see our [advanced installation docs](https://github.com/microsoft/DeepSpeed-Kernels#source). + + +# 6. Try Out DeepSpeed-FastGen +We are very excited to share this DeepSpeed-FastGen alpha release. + +* To get started, please visit our GitHub page for DeepSpeed-MII: [GitHub Landing Page](https://github.com/microsoft/DeepSpeed-MII) + +DeepSpeed-FastGen is part of the bigger DeepSpeed ecosystem comprising a multitude of Deep Learning systems and modeling technologies. To learn more, + +* Please visit our [website](https://www.deepspeed.ai/) for detailed blog posts, tutorials, and helpful documentation. +* You can also follow us on our [English Twitter](https://twitter.com/MSFTDeepSpeed), [Japanese Twitter](https://twitter.com/MSFTDeepSpeedJP), and [Chinese Zhihu](https://www.zhihu.com/people/deepspeed) for latest news on DeepSpeed. + +DeepSpeed welcomes your contributions! We encourage you to report issues, contribute PRs, and join discussions on the [DeepSpeed GitHub](https://github.com/microsoft/DeepSpeed/) page. Please see our [contributing guide](https://github.com/microsoft/DeepSpeed/blob/master/CONTRIBUTING.md) for more details. We are open to collaborations with universities, research labs, and companies, such as those working together on deep learning research, applying DeepSpeed to empower real-world AI models and applications, and so on. For such requests (and other requests unsuitable for GitHub), please directly email to deepspeed-info@microsoft.com. + +The following items are on our roadmap and we plan to engage with our community on these through our GitHub issues and PRs: + +- Performance improvements +- Broader model support +- New hardware backends through collaboration with partners +- Release performance benchmarks (used to generate plots in this blog) + +**"Star" our [DeepSpeed GitHub](https://github.com/microsoft/DeepSpeed/) and [DeepSpeedMII GitHub](https://github.com/microsoft/DeepSpeed-MII/) repositories if you like our work!** + +# 7. Acknowledgements + +We would like to thank various open-source community projects including HuggingFace, vLLM, and HuggingFace TGI. We have leveraged HF APIs to support models and tokenizers in our alpha release and will continue to add more models. We especially acknowledge and thank the developers of [Flash Attention](https://github.com/Dao-AILab/flash-attention) for their great work. We have extensively leveraged FlashAttention kernels in our system with modifications that have been acknowledged in our code repositories at appropriate file headers. Finally, we want to thank the developers of [FasterTransformer](https://github.com/NVIDIA/FasterTransformer) kernels that we have used in our MoE kernels (released as part of DeepSpeed-Kernels repository). diff --git a/blogs/deepspeed-fastgen/assets/images/A6000_benchmark.png b/blogs/deepspeed-fastgen/assets/images/A6000_benchmark.png new file mode 100644 index 0000000000000000000000000000000000000000..9d4ab55f5f7a32f16d7ced398d88c0f98a41d40c GIT binary patch literal 203812 zcmd42XIGP5_%BFDq^T%Hs-gs?2c(xEO{FR-QUg)~4<+;-s)$Nxh9VtAK%^-IkQxvT zp%Z#90wMGkS|DWNnYCv9&&>G%^MaL?+~Lm7-uwEsYv;X*ksj;io0qAms8}CA(tbuo z#c)AIMWcF&f%1w{_RJf~1GUdHy@ynl{d{YbCv*@^Lrp5GABoK8cJ!3zOrDR-eWh8C$8){Qc;nM9&2kp5472wW&C+p2g|U)piOt3_L`24K}1B9gUOXHhC8K?ydwqa zG^6Oc-ci#%zWwy?HSLcMN#aVYLYh}JHKS-g-pL7{Yb{B)Q&gCrZ3(DvT~!L%?$1~= zd0}mBJvu7;xa{F&I+{q3(3MO7?+4+Vh&mMQ|M5~}+a+DPi~sY#MKV!cJz@TT-|YXs zjR=+Yg4F-_lm1=&M0F(+^?w>hB*>Nq^#9Wj8w4ONDvrQA+Qz7OF_u_)y18GER`p7b zV&U{O*6)5?;pV;6aGCBc1LAq~JzdlPZH{FE7h|}F3KP{cb3?Wp2q-d1s^z$4$MB$6 z);KeCziWPPdc@Q?N1c2+usV-^U=vO{B=pJ-e9;pKx&zyd7BtAaAe2m_Sd3Lq7NXZy z(!oZuz0z8l{!=c)4hX}s3&&q=lzZ;Ni?jDA%Cm zYnKeH0*@vk!^6fJIi0`UYWmXr!iA(?4OSVepE?Auq?!#uG|ndFMxUG3hBFc9@kayU z7qY`%l?}KjrbbGRXX#8_RV_yr+dBsX8hLZat;aYBLLuR>DqKyXVfo|Q**rPS|9ix|{Cwy^J?SWqd=4PXR~8s@S$S#_`vY1LM^mm< zc6s-{Z&WrFAC6f{<-m?6*Zy#U-M~4Xo}(|!igNs??p2^Jh)Uclw-)+Ub_^K#JVC@2 zsqhRk8=c5YtAJTQm-<Q^IaSGe=Z}3s3P`g#%F-g2Mv>z z__+|}a&fw9%q;#LbP>822ksrG>+4tEC_m!3XH8kaHMng5qeQi%SHyPqu%rYf+^5aw zAD;#c8!D{Kr!--8c~|DcE)McaT#7S_Nrb_rxsV-0*Qyr(DJY6GG#fBK+d4(h1BjET zdo>lZ(^4ZF{PD2}Wl^_paqNH8mwZf_ArIncRoG}N(I4{$Az@!nd~=ECQ&A!fW{H=k zoqW=%F=XLdk?KDew3_{5Xx`sCWUF?v{pb09Pst=CuLV}iOqnY)0dQGWYcO#?9hL3q z_2-6}0k7M4P385kg+ucfNA(0WFBINjrb2ghr(rcWv_dM$E3ogu__EixT!fMGy#OR=4ZTY}QYh5}R*Vqfu(TvMb9!$J z8m^)trnqeZum&CHdUxM*>5-`KZOn3$Xdw)p@{BuL9@8gkTm+6`eGdgMc4Lo5Bkh02 z?_5yj^nsgpT0^Gn6V?7Ic6q`117>6@YWkIfmXcU%m0NdXq`nx+Pu|(0EUd7kkA8l< zM=>+qpEH`eUq$IQnJ-`2#SkbvuodRFAk7)CpY>jM>L(rV&t349s;aH@AY;H#49@u= zG{cn^dW$Cfe`29Q{=)UEn5O-1v0>E3VU^sh@9+xM9HaVfq|AJj|4!I()3ouI>{356 z>9`~@bXizwK_uzVsp{Gn1Wr7WQ6Nx=jBekmhr%{2Pak#uSb4N^ha<3It8RKO`L4pB z>!!=Xcg-x_YmxZRaco~%wMIQwRQG;!tTi9>>nqHM`kbv#FDgEFmz?Lg$!BdrJgN+6 zDTCF@zRfht3FEkgdvr%5JNm*n;f!WQjWDRSCKZ0Fe~}dmJLwSgAWz*mSGqV^8a`dk z%i~W^F!Q-nNu)ccE2sDTAiGaljwh?xK1TkbDxb@%k_^SlUM%cVir$B9AtnDASo(^x z@gtB%q@sPbg9JOL?;V>3S)~9sD7G{qN0y;mgpN>Bzq^y5vW-!W+6>Hv`+C0|{=sDv zXg5|Lb;m(JLtg#T^`xd&I=YT@uOiKK}sncV80W}K>vl~{=BFE z-}%axhCVjQ_6`4`#%Y&v_0m>i___TBW=4fX7#@a(9b_SUfj!G-j^3V5KKZ?hOK>IA zn2r?tKM&ZT;b(mTr}4u2*4Yy@e%V`d^mp6Zu@Hq|03?^?+Hr_N_8IJa%ObM+2}oyG zJE&EEETYoqA)P(7>&p~_q^VHdBIZxhG(S_ znYxWIHYuUXV^i7uo((%JV26piQ!w=Or;O181%Ck*4Y}VU(k#8Rh?jje+2s!6i**vLw9blW3K~M_(lJ5;4fQiS0fLUn)Q+du1;&FI zx_6t-NOR@$q`7egbEfJX2aP*OzKjbZj;}`_K#njRQV@B$+GMdD5@K2w1(< zi0uqWZU+D9*>)4uf$O^X-nP?Y+zn`gB8 zh>cPR&Fz{(Q$NsALx#!<1$_3Pg)0?c{H@UtgPFfm#v$tE_$Jt`xNxaL!QWn|hKnn!#h$Z&aJ zL*FIK-$e@xZp}I$-T{uSyO$n3$x1176I{V-1u>{KBHZN&25VK2Ztkk#9$1Xz`Y);D z%ND!{;+S|D0lQNKjtQH3w2JNMw$pPiX&fyJU*FI^-nU%K+j)ox&LUNKY+>{E>mi{V zTF|8;)X$T@aXvzGfq6l=jI(nOwJkj{xqHO#6Y_ZLd_g~+`!HxiOm6=C`oj$%Zt#g( zpG;!EpE<2xLnqWvZEY9I20Mlmm+`tr}4swik5rjJ<%=6fplUfj69Uyn$js3 z!bzfGecsFTKP8eMUSyU9IS$y+bPS&fd@<(e>wfAOMZC;r`VKv| zG&PXXaqYfW;s#sO2mj~Zyoau!=d$|#U<*0ix>YX~y2xB%9WYyacz!Uj3KR(4Z6B+f z_L#P?jik4b9Z>bzpzKTs*O(hGok@%A;m)LE6uIyCdh;nRt3qY_XJPHIkxFeVg}UpU zGF^(t{;mARq({@1MOE|ulZ}(5L}(v{!5jL!0C~-at;a1@hmq`|LS(gI0^X{h0(Mgt zZa@B5{Ht{D3}X$Fuj{PexVuCn{&PjFK(HvdZ0Tqk3jN0(6)~I^U;Vaf$Lp%C+&eS5V4m$6c<8xL#(N zjju7i>P9=~O@ZAE2 zfB#ftg?)mO$nnmEed5*&kquCQx9`YPJ;zMRnqHtidX+u=q+={zVeZp3>r|Xnk7SdZ zYwC-aF4h!)G51hSR9H4ajiAPDLL6_V{Ls(cYqLIQN zeF{)DBEGO9oP0JP)Xdux));yJjx+;ya(ida_s); z)@=)Dz94Q?tmcMD_FPe`!DYHSSuDn=Kd_q_y61qW3<6wS01O?D_wBKZQh;;IcQo~SO=@Z^4abgWzty3?(kV+~gDaRHd_fU_%ORI{z2YB?VZ1#z zD(Wda-y4Nw9>9}|#FYiVQOgb0F zTcAi=P}!}xToPKLHF?E`H&a@Ex0Wz2D^}I)EwJRs#)zDA%5tyf%ipUCJ1pyE+%I&q zzqZxOeDb)ebb3a&>gGVv1*NJL2ma`=#j7cQlF&pYg*Rf)5B|B80aL%hDoeK?IvYND zXI8mfcyVCh=}J1vaOY!tPQaYfd;mrI*M}X|gCzqBw%ZYS2KCpz# zk7n!k%{L2)zR29b{)**~5kSkC>+VG9OE!F|ZD-(h-~md)5Pex5jrXSY=r=1V_T$VP z_2YECO{LN4aMTP+q70*N93BreD}?;e+->&YC>W)%a+(3q1FpqC{nPFm)X>30o9K zR)QUtf&274`;=C*m9+ls_xuSgx7^*dZ?1gil*J3wG7V&E>l?nx%8o}$f5$E{+AT~kT! zD|q>)ZOA9Pw@8ZmUNzutLe&-`n_4-*oaq5Kzjdhi6NV*;6x5eR=^K*%ZKvWFWux7l z8!pk-E9yByTI~seJdXc>I@52={Gh2BX`1I~wZ25LKskwgNLlMFz85Pt1W$hV7O_YE z-1*79)p~+3xt;QHy`2n%-7YpUOm5r{b1QZk(gHSU;-CG-pCDkWE`X19WA**TobhC( z1s>K>NC8@zVegTzPa6b0RfPm*!_KxE_zsmNU!GPE&Llz|sQE@$)c#B0x+502C@QJy znVlY%f6D^nUtLJ1AXML0fOs*>r3~-&>KbylSXJ+NTgbdn&1JGfX&6p6kH?3f@BLbf z%hdbu^`ymBjCSbJMGHOWE1~*H6-l*uemKvhW#A&e5fS7>CB>$Gve2vGt=!4yv%Ik! z5n7hMxo+HLvu9(#Qcq*UUWy8wqsEi^yT49^h;Ee=4v-k1Yvc>BVi1#4|n>A}BHeF&~zaehsZt>=|>)=ghh%Ump5CE?VY8U2QKa8H0n z0)wXf%SKsy@7=IGCx!z}W+e5>hD5%PDDQF!B0_`sR{Q-o{PAiIUoPn zo_%m$HFoZ#NF&gM@NUmU(9Ge_Syta_^;8$LB}=%PDbD+F=sgis85jQr>UQ(+=J=Ei zZVeRkA^(3GbvhBy1hvG{K)(x*m;e4WRwBC1C^&rDi$sB>+C1l#^6aY66|o@3hVmF? zHpPXA^Q`hW6iHMCt93nY=3n%w4|rR)dN>Xqn3Mo2Ik5FoK+oR)fcy`O&F%faR@F`Y z7X)eQ<;kF%a#CoqUdzw%hfiBe3Da~}!Uixx{|jUKPpaeQ`@J(h_;-#^GfgQ_#9Tea zNzNnx&F6~iIx^1sDB5}zSoe?LXC!&J!c3ivlOk_VceMb6FF&!=#Enq4lX`LUKZdI) z$ql{_q3bsI6n=4fkN`tQQ}PZm$P)v#!;uCu{(~;@y#hsF`1Sh|aSEsmS?S%qG)s@J zaX^N{$81eN(;w3DTnMyZ^pj(nxi?NL@<{Dy!oGrHUZ<4`KW}EHE4{CyBvRJ>tv_l6 zHENCPKC{q!+?`aR2hupWH&bH7E@a9JA%r%=rJ#?ckbqWV8^vZ+C;yXe+3klEd9DUg z>6&T};0bW}3aZ0hH#JC>HJ&kx-DKFh-BFZCxl1Abtb<)cRivj3^CV!f&N%chhf<+I z)_Umkx8)q(U%)^E2i>FxFtKV&M&g=rIPij!C;ZGVy(MPMNk{Av&w=Cyc^}z;Z>g|U zIZR`6u`j*v)I1qu2urqJiM|-($xV(7z-2&in|DZPVMF}I;TbfyYQX_J7E*w7^RK#IEe<}BU2pg0dj zh*WaGabc-ER&&|3m=B|l&E0|Rgq%X#dVYnQYXPGMM?Q%U9MbW46CXOMDg49QO*8vT ztoW$y<&9J`n=a)Tfzi%%6^C;_MxqgChKx)3Xbn<%v7J${SDDIig#NIVcWFvfu#4@B zDPCRK^Pg`uA3$418p!%;O98d(F~KQH#l{ZJ5muN5VH6>&Ze688+8=H@ZY>Z8!=pJrPg=FJcSgOleD}!(|_o^B+O<53cF8|Rva&q9obJl%;6qS+WRJ6I8&rm zz@(8*6?bCwNgp#_pOY{i`a1U$#T)n(H?For2`PK+Y(iY<;#1=^iJ)H`?sQ9iR*r1u z6dhgD?>#WgTrAP6fMzb&%=)0j$)R?1kX3ype?A?z6nyK?6%7Cm}-F_dXPfm zh^kU8`CA(I7P_LQIn|!3_|9js%^|6=HKKJ}YrLLp-$yvK}a zo&eCPw{y9HY5?<2#w{#BN_nAW!#A_M52)9};>7hy7!D(s$gWu)rOJq^RJBdyxK=h) zpAQ0v+8%h9ftzLY#B%p;>3P~*EH#d{MCmItcp|}Te#0Okl7?R=ikyqACijsG z%P_Rs)1eKc{{3u9rPP6A!f%WGFRYzgf!6=*Ei?ne9_)yxlSD$Gh4J#;JI&QimbfVX zx$4K89*lme#-%lrl_Hy$(+%yo&P|QgQma}hHadu_S^Rc}5-@D@_Hj6mBJeB4CGV`% zb$eD9S~b87Vjgus3Y1|zsj*rUy zsJ(=e2#5{eh)b`C50uJsFNx)mpWvuaz`fDZd4;A%u$`I%7U}4$l z7@51%v3&nUC&E)HMob*ybX{qUHIN3aI>#fpj5#?_e>kZ;XAI%vN|Ku7pm-je{{3W4 zh`~aMK|Td`MBJ;pM9kB=zHeqx7(rwBVUq*2Ync`Z9{^fnKK9;P$}(8XGSFB`fG=qT z<(FaJgsEGG1R%Uq6sFy3{#T=-r|HZw&8=?IDFZy`Xt7+qTY;gvo*^2(p`nz%@@%;f zz+F!?3J~z=mC0E3jorhMLVz~^8bn94TC}hl&s68hM9Vb=}uK z(rcr7OymVIC%4Q%Qb{}iqVt33Ft5B#7jr9ijq_dZ3xv#)=JcA^)@KF$T^@twrnbO) zJZ}Rbs!8DBCwXgfeqC3dSC#7o2MbA$cVj{*fyEHT)TTPx%y)^koX0~$g?ozC2ZeFB zcRmzC2NSKZ=|thUb%4g{nvf%h~9h z0{xk1H#lPSyP2ZQY}`Jh8S666Y3Y`HX7HBK-uyvi^9>~y5T6IsUdtxR$ zztqjxL}RwH51r(n%xj<6|9PId*S-4s8u)-QP$0ZlZIj<`u~a=v4D<7cMJ%~&SoLYW z1Wl?N@?Rg8HU;rMGr8-^>+2`sPk%>K>o{CX9Lo&VCX*xHY zR~dURY~I%4^&$HtJBo%i`$NL%l4xD)HIpk5H3HWn7BG>)lYX7H4Qz*~5SkK>&XhuA zuU1>evZ5rLj?tGPh>?8cBf)-;trgpSEW74|gA93<*X1RLme8#Zb~ar-Y-*d6rdL}j zqoJm?dbX{)->X%xa%G#urE<0edA-<0rbaDe_YFeUe#I>%_AI?8-sj#JpQCltNR(2f z()o;OEZOJp8rZ9OJqwbO2=KcE@;vv#rKuC>`&viHJ3F~}lZXqxjMlw3_CL(oh#ob4 z@=Ch^llHYOn|`1E5H_=fwW09~PzzCfuzj&dQ{N*(_LEN<$MjVMOfwlk&8Al|+O8xy z(H>+OwVXfe@X@}RFXPgtk~hH97QSf8-@R`&dnD%VRzA*;in_6Vm}RSXpdi@#DL4Yr zoOE|7J)6suZR!0yK}(2Uv6xwTFSq$`p3vv~r=*K*lPE1N8xXZ8d6^t|vv$zHWwwT@!I}CCCSZD9k{VGvqMXi@By#ED5#542 zx2JDGyZ}L`riunkBah?tDG0lyy-)~xXRA016^>~rS0@v=)`tKv=Y$8Ez~E-Q0qO2E z`U5N5zwoR$F^HpLu#e$Vq~`t`@QS&&zJkI^+b7YbVmg$5+w)W0DiFmqrLsMRo$yUP zO97|Tx^G{%m9L}6g;6pe2r;$Vf6_6124J7GDfcwJ%SaTJddiAgucr6CW%Kdx?n3obBT`>>RtA~_A| zTV?L%i?3RQ2C6VrcF{z>r((GkXWJBK;H>q-C*qe}LFScvFVLMJP8w2VO~F0b<>Bq` zZM#wL))!PYBH3ZAN7ue@(h zKJB+#H{YpSRQxN)`G$r~(&$BkaKE@HY%t96()z@nP1p`wOYK0g2~T!XUx?BbEPSaO zM!)S7K79HiY}>DeE@pk<(SciupBrG+oP#g!Zd$T-e#Jr6f9~sa`pE%$aO?^gOVNpLfM`lBOZj&$!$sv zJ~BxZwi(oUIr&tNKNUr&W8k* zzjGuGB6oG>k)}y3uLFj*_-uF4L)+}g!=ia44+Eyu%b)His46HLPde4M{J|R}H!<$C<5%h-A2^9hks!k?smw2WOvhDdMf%m8^KszX-%l#&CYz(rYW#0kpSP+Gaq$!6<1ijK~_ju_vXhxq@3GX zVo8ZhzdL=1rgql5qV==#OIH3SABS)^G`+~W)Uh7-lWN88VPRWsP2~au?S~m{yOKiG z*|W+JOqCT8U1?Q=H~#Ark+%(VYdShT4AIE8O_1FcslK#3+c!B2xU5y^bd(@YBDr^*UdKInVnixWDOWQ+IUI{de`=g!THk z2U+1_bLRCU*>AMa$b|3FT~|df1bBr2{tqYL>oH*1GxGJekjr#J8nDQmmxS0^urE`_ zGQmf#2aheR$t~r5=0|dXaCeMV8Vx9UcPLg1!zDME2sfe!5` z9Y5>wvy{}n+6qzeSyqWPcxZgw%E$2EQ3o_!rzy3iEf)d(8X<+|0VE5%tRFm^%<%9k zx*$S1-Ecxi*X5IR($z;-MUR8FV>Qn0{?~z+&R1UhEO(zUBwKeMHcTtD{As0h^;)9) z%(IjD%J?EyOtTfDV;amN1~ONEa~Sv5*`IP(ZF1{1d-Z-zZU^&}3GK+uA zYV}_1Mdd+D%3}TB6YgzLTVWm$ZfB4p#B(+nI#*uC#%8LAReH8fIiF-c6?)L;lcn^A@ijOYjl0)x zuc_`+(R=SmVfux%WnBs--Sk(?FE}a$5J#S7*G-kp`voSzM^9spw#8N+xKxNc4oMuY zDi=s{7^ZCGbhknFpM{UjPc11_PjBi22xF~A_%)tJ9cf*a#$ngPRnn|!kh1|BUdUB4 zt8=Zu$;4i$O>f_w&&3rnS6SfLz>Z{ek7td%t;yr653m^Rj0IwfY(KCZ;6n2?LfI33 zU6&IFmT!uvA_zD&~%>N-? z8{DvB-|y5@L&oZywjLIlo)QKqA$0n90PdN&&)9HDw)c~|fjFyC$ImEHd+wG@<>y(w z(4u&lW=R-)eIz3hyg?e{k|29sNdU_7Nko!uaGZB!_5&`va6LfWOdG z;iYp^Ur;G$WSlKD&{P9@Y>spmd-+gW!cKF_srd>cF6&Ui$&Pag^iL&ys==wWs^=!w zO_$v;f5TrZq_qjo;y%5Me5e^3{Fd((qC+qMv19fyw`cNXO&^;qD{CSUwDQ*y3o(SOJW-dAz&)}G-IHR(3A zP||T@d<>4(_#^s!-h}&|Fc06Lh@Bi#rL#$u&VBHO$sfOcbjED9?HNteZ%l2z|3M*} z?w5VTuQWO$Hny)GuR@j^I*nwHvqtzf+9v7?yM3Zm`F=Q#&zeL~+bgi$^nC6bZ5^|2 zrl4FfUJxsg$x2J&`_`zZ+|6cne$$fp{;K&&;)TEJVw|!P2lu3}t97w*0Ue6F7)@<{ z7urk9(T)>3VU~044WU^+0x!yX11T6O(pwd+cYUSw7R2?^dPKkv4y*Z{PZ@?X`Y&Im% zdqXWLt`>KWl?}{wNT4J^T{e^6QhwGhE{GE@xuy=+# z*bbg?t|DCtTq+pJv*76-y{r%m<5d3yu@pBdHZ-m=(;Cf_sDNj2mg4TPDTj`L%#IR; z5GdyIXvY_y^fI%I`W&;faqoz-B>K$Le3;LLbZlwI&fO+jOnTj#)~ttwE3Vk^*{V6% z5d%y3qy=WvA*}U}ZP6Vi$_%>D{-_Qq%m_;|H+ROF9lZ@|3eAe6zdsl&T+1Wp{&@t@ z^egixTipk5q*-_f?G&%&x&!lrGYH(=^73yuD>+&QZ9X$2T9mjhQs&lw`*_OiToSfO zvdycm(}7;L+c**azl>AYe$~VrcE<-?-&6Vg#l_R-8a88jngT? z6^`%qe3D}$RJjGq%F|tyM4j)SXP;p-V8$0+ykR4LrKeQkcpGkmh({3*ztiTFuQoT0 z7Qbre`M#zl*rHO$v4uJR94DDs_62^hith)AF!jpYkQ_0!1GTwbc{4tpmm9y8CC?r%wj>7sPBW#s^ zHFFFWsN}o>169_H?8}s-)&Wyzl6qaltX5FC{j-I>lRBPJ)n_JdvVMhPlU#S)x@k^* z25iwq-sR*qrbz9q&OVK(2h2^2HS^{;5u6`SxNb?sP5|t3 zGI-?B(~#&Kb-PUF7N8#*o3rbfIBZB(c9-XCV6%NPx1#)!% z>B9(GHq1azwWr)ih>}@t^01#!d)GE77q0I&aq2Xl70f-z!oW}qXGRh!Cqai-+Ya8a z#Yv+DX&v&L8iS*Suk7-ZB&xUW9PRr_In;F|VBdiKsr@aGkugfG$#?rhJF_R^b^AAw zzISLRmqc9{*i)ggoUtl^H*k@|t{HpeRQ!s<@o0{eZSviBwvcL_^qj$>Er%!h+D8Rc zT@zHT_!?yiU7ob5oy=C=4bDOI$A$beIvU!R=SE(q6hh#~&uo)$$H0e(y=I!6aBt)*kX_($eR6 zgbuH&m|0^!i$J*`lkv8YKf9OUNm6ItK z3xl(SsFZ$ec2KEmQ*v{kNKifmZb<4WUdgvxs;-}G#lhHgl@V&Th)cZ9Zr^_} zXOrA3>lPzh?A41^)*jo>9e~oYvV5hplgc?D<4>!T4f2-GB5a~pG%mS=cR~~Q z+SQG>flhm`#LeNyMP-$-;u*Gzul5g90swcvU%S)5dvoL17L+)A$hYODwZY;GIa}w- z*%57|femuWJ9$xEKAL!^h3>7$huY?v#`S1B5 zQW}eSUU%z%pE+8}H5JqJ5(RQj-+<9d4Q*Oiu1n9Yo4f?CQM$aE$YQ|m%*{D6z0_nj zy0SH&HOXZwKLi8fK*L#e1J?+7bJ?@qi=9B5-AA+N>o-5g0w+*?LicZCp$>l(EW2v= z!^5L576H7LFZuLZKs2qq%%sRxS4OghX~6xsInIURDtERO(0Dk^RDq{pTz>%rJt>LL zT{6DPqRGLF#HexS%}Izz+vPk$@BXPFy7XS6`Wid8QwA-=gY9ExGBkE0#Q!QN9fvz+ zDzwc+qoYD}M9z(>y4HmiK*odoj&4Idey~@1#y~d~B^qwyE?V<;SJfg&yxG%^N_=7Nc!WQfD*tzb@^5MP?M~@e=!@-y$)+^FiTB2Id#pR!h&67iDU>qCK0fHD;oFIzF0x&Ye&mEq$$~KWgkk!SM5PJ(}D3|_qv86 zxIO{IZ2!h%aJAa5YaU$(S>L*LU4&!#4};_}!`v^2(bus0qxs0TS2{3ng~g}vl*>ih zPt%y1Bd`vg&F|k_yZ)mf_2uMHAPu%l zQs?a}@M6$_%n{v#(DMp6#HX&fr+wdRd}0i;kOr?s#5k&$%%(SuoZk6#-|zUGh;_eR zXCkWv^yW5+){zla+N+J`7-ufxiG^!dL$A>@n{U;|C92Em~5Sa|c&-Pn*3^vKSgr0@qQoPwlmMGs(OU*SPSEdUu ztVckEjj=XF{tB$Y)OZes`Bul7M#QKqY+E1PnvddLFNxb53gi_>7jGWDX&u{8=f-nO z3w~ra5$zZX5sEqv8zU zQ$o{cp`Y6F*%MkI@v_X&I04E*lJCsi#E5=L9|W{L-00a6CxE6izVDv@v$T*Hir7~JdYku)m@L#=(^7vBK~fpvnaX-s2H@Sq5Q4}G1tyWlIA)Zzsh;Zz+fSC z@%JKM0b=b=UFF_-2Q?-k-*U2A$W~i}`v4u%z~-AWuV{{!IVy*&-S-z)8J#f#d@rk@ zt^m`~DOOCR_%d`#H@%7fb>UFuDdc_G8}Cwhs?$KY4$!dvVinweQ;+w&Qq7G`IYnLa z;LXKoKE$Z9rCzA8UY`FuOb_j_6gCHa3}ws#V2;ff?Bh-RF1?7fMOAN`voAZ(aHS@XG9;QCHff087e6RS$|f4 zn4mnUYXWM(n?>2;``-#P;~wk%POJXl=0e*P5&Cn!#laATWaRI<1W?Hgd>j8xlGxXO zS5U1$@y4hf?B%AFy91-wC$WLdd`z5w5*Kh~!mUo~LX25zuXJ#Zv^=h0gJ2AmUd>$X zZPKk8$_>L| z7fj>Wv6HJN@^4R-CgvP2B@&W15vjmRYU{mT?gx^S#`L~i=7ov7c{y)9&|Ey9p{hI6 zSxQ2&PItzs1!zPadDX>W;acS?i8 zr_r}WGv8vy+71}LL~Yq5!`H~%M{IhkPq7$)r}h3iyG%DItQ=0N;Kq|D(CaRV_hm~-w~KU9l!NZjQC1v zIcMgh>)}S#Drzqtx=aLFuCWo5XCDi6l?Vwdr+MXDk$&qn4tYhQ>-*f!`Q09E4e80< zvfS1)4t~lXh|k1!d%@7FH@nL3&{;O<|mKQ zqf&S`tC2_nXQS1%tP8y0>6j$BUj2K&Bu~&(xm7dmk|pn+c2;enI+VM_xv*vG>6F`f zVXXYFe);!bmm5^&mb6|bjE{2b2R>a;@Gc2G&^XoR2>!qZ+{S(@`B7!4!+v&xVsA~> zY1Wj&Kh_|SPf zU(mMPUkkGzs_A~oW!d>t?l{R%If~UjnjU<|@)zGT;#)9HT8^p8uA=YWq$WK)&d-vusbp-F{^@QW z*+yk6WGcrH;Kg8iZWtTSgn6UcVDj*~t?`Fpx&?J3@h6DaU%nXJ<}*$K0Y33@3+6ts zi(KWFg8TR#J(qGc%#fi{g#uqUvUrbKAx$Qu{ zo#WfKx2rFSq8<4ibn^B-{%C9a{%9E8`Wet!c!Y3ec*|7PM0K@ci3O|6T%^q1J~U(K zfv@TfBR;?+vuS!U($`{yW#nH zaBKTwrKbo+g!<_oZ^n?(IHr6an`2kVLN=2!7wf*N#o-?O1pK!IxDK4;(mN4!oV`m{4e+8sG2mvc)emRX&NH+3*}#9M2WpQQuGs z=Rw{-h}g@wc1R7=k*Z!ieO3wGYKN{g&C`q;Jwys^I5wYZ!0coo;@DEZmJ0C!v4KE| zm1JgOEZ+v0KUXI4yC59eNbkNzM*;~BYzR00r_bT2N_;7xHYzw6!1(}o--x}xM`V3V z@{sDn^5-Bm=Mvig)G$j8qRPY0ck?i&50vi~XObs^;j?gI4qh@W393aJYkY{0t_!rl zF##Wu&G8wugkfDOB)@peDCB%B2Qp=3#z+%6_tcFo;aBxs@4j~slk z{96NuW99Z}wwH+rV1Colw^&KXfluj^RkM@NtB2PD{5dSX)2|sTC(?!j>17HC;gOYa}NK!z3ETQ`yK&S zJB%o6>|&bML3L5*ezYU}K`C3+P{+f)DzS!L+oJO6(5x=S7xb73 zpxc$LXLa(IKP@B@snrf=E&5A1niDhK(Fvxy0rJZEOd=L~6SgCu>$;4Tea#=(L?U& zZ+jI|c$+pOxuXfd7$YH8{c^cE7j-5=%ikF~D0PI~6sZF=*cL9RSgbeg;DQH=z4t9* zYtte))!{cE+^-j$P>eFk7sz|a=w4zvE9I=Y>dV-e!?xfC0eiy{rt?H2#_8)%`r)hq z^L+ZO5hqi(^WM1MMXq1yAT&w|uW@Ao)t}PE4nV#_ujFHe~2s_NX z25oD#_E)}*a8n;$if4)zb_?B!)&f+=uIea6aCjlF3Z7RQuFF8u49^`6G>wr_MCiJV z&D!I~m*gcaEu(_^xUXK3t&D&lK5q?b=QE$1$G>|)iFw^BjMp6YDe1qcAPKJz+<`V% zo41{~j6qZnxVXVM(*xOXhPeYxW!Ucnld2M>?R7kXcXF6B*imR#oPV6ZqLDTzU+mN~ zJ3(8ImuzHwZ|7;J0?`nBDhR*fy`X5ZRFm)!HN;l^H_#B-4(0vcHy*l|K>O#Ia^z1b z-O;Fxc4_YCA2DwW6RCP8xe9V|d??$wLns$Wgc7Pdm?pIDa-Uc6SFHeh*o6-8-M-@$MKx2yZmUgKlZqSo#Mf5>p7 zBl0TNS}30h@vY9}jmT>RK-yOE3lC&!;EH_mx1Kv~Ccs)Wb*Fb|*bh|s_|n+F9EdI21R zpmq#!@_wF-75>4i6yZf~cg~{hPxNpe-3`hQ0Mhuc*}U#~=fYs!r7S!fHKPrMj{sUC zqmo@?*@Kqd#$LwJa`;kD%G-<>kMRt{!H5RorNQ{cs64(W{#TOe6K<~XFn8Nd$ar9~ zTmR;Ta4R|7Y{_Ea36>Gy`9)J7R_`+q^{&9N_wS8d8O3;+YF+i($lOIwI|Etqb%w|= zUbCq?_}_|Vbur11uIxVv!Y2f46U<206|<5*_JCIr=J3-&>OeO;YR>=AuZ(qF!tG)u zt(}^&6@FTS53ho|A6jL>?cImCbZxw}Tz2?V^qU5M0N7X;a?IY!kFKA z>JHK;!0(d%q=)d_%Tkk+ZyPWWr z>aO<*Tgx=kFZ-}KxqR@Xi<@X9y>eN8$$Z<*Xd64d46#YRF{d%+O*!Rs*bCn;A#Pv4 zKh-|#_3HMO3VAY(oP>@^5;a!`8{FiMSb2X#-7Y135kXFN`tY^6S^4MyMD;_;!Adl1 zz6bAuo=%^=Hpa(+;7Uwt$Q}kIqHzR`amrm2lX~9rkU_1p^KAvBJb8)Ldfa8!dHbbc zX4!xm&RrhZVp*@W7tQ)B4XYPAsIWEcf%>sRXxP0pJIHQ5aVfiF9oY)u&4{IE2Zf|A z@%V#amAfp&Z{rr944mtgqT$~aCUkP^ZAKV;2X{nJDFb!4*oox={myko%7l`B$oS}r zy^XhZL-zZYc4CD6bp4SzZ=rDTMpbnez%pMFg?|p#Yf2=8<}oi};`aw?HU-1Q#xpI* z8|v?r2SMSOR=;P%Hs3me`-0mQN*e0^GV&bo%|9;HYx=(kI}5%h-~W9}gCM9VUCKu~ zM~)5wl~g1KQZl-`K|nyIyGsP5VRVDYXc!G6l^Ua4M*Z%6AHe_r0N8-n9oPH1&f_={ zNe+BFUh@?eKwvmo#>QjgglgzfHbm+d)ezf_uB_7OcZh`>*RRp*BBolO& z&j0nsR!zS#5n`?$08i?e+9v;J&PvzvU6!sO)NfqDq@HptYSPkV=dkV8I@R?kVM)cf zo>_VaCj*yWWd4tJ=^YwDes)D~^v3xcxdm(J@sjp>?}}33cBh7ih#9#6Uw@w9E+l~6 z{1YEinQclBnnqr-<85au)lKjEc156Mg8FEjp z`LHUs_nrAhnF^7Yq-Wpmg&|DuBY0Iw~4>uNOyMZ(5 z#oWTYBrE1Kw3wI@fMgW=@mu$=BEo(R8YVk3{u~a^slB6>bIx3%um9ks1ABeP#@_2b zZq@;#6IL=LEs+d(OJyd5MM5_wl7|y5lTr&wf@k0P8Kayq#o6#5=RPUd9JT$#0r|$T zSAqS3Bm$X~b*kzA40FW>opN--U1C(A+MPq!x~M%FbHh%6Wdc)lIvB9{t!v5J8@Mws z{aC~Z1qAz!XUW$8i^1UD%<2Z5NP%bhV1~HWsH%}LxVJf3o={Gn2(SU+Kh|ixl`7Hy zB(NY-CIoi$dg)X;9j@rm)u_M%*lWE1N&J*Jit>G;{|-QSTkB9- zmw_w|l3lFWw=nBlIYH_om{&e@n8$*40kuO;i1n=R$qy>jE9x z`3eBA`fo)~OP#J_@RNfMp=Q~;mld`(2m?TNV8~rj_pS_-VVL~R?tl0j5T!5E;T$~I zHfq|14H;k)pk)~kU0ZBWvW+x1)pg*vpldnSHxES|1gNMpEiy(wlm{kJ@z9+(ROU{zN{u&%V%$PR!|9UYqf zgjWFgQn3!La)s^V5#e&Oo#9km2l-YVT}aqQ`oaGgIGd_FEc;ewAXi^Hz?SL-@Q(GO z%ie3ll$|=HJu_JT#K%CLqau`C!kX-klxyv0{oKb$3*=g;t8_61%0Qi`Vp`X*!UM>U z0w3_>W(r{nF#97n?`{DEbFo9hIgi?SThZbrG~+#`(bX*9drA)`%4yst9d&5MAhCaD zn+~CCO2Nv_z89ouE6qG+CYXq6&KKb17QD#Fr(dp=QhkJ+A2E)VzOK8mqwe3<(KAZ6 z_@3c@(j;8Ty;gdL#UXXntDE=!DY5zQl&$P{TMdwUIuf-s&)L7VdhL*d$T7fAskbO4! z(Jb&gQ~-3T#E>H+)_!4!#U0IIAZ1wA1OZYr)w=kcOL2j#heuYAb|hT`E!M%s`roV} zuZ$tRL)}}uizP27BlmHo=9N6APCa-)8R!lC12fn4VT7zbaH%nwC!5W5AbS^X&KLMk zB4SFZcp&^XD)lTw5>Xw}fQ7+uI;8n#%20r5QIGa*1wewf1)TB5MFQ3j(TfMsSctTd zYfa!zmv<2+^B@0dq07=si@ADTlYlcd*}5vS-gACDqU9_AZuKdTi?lTp;gAjvPc`IV$x%DgqjhycJH$orADo4{L53%aOvueD6=&q&j zl*^d(MIBbevKVXoHg!w=j84-1hgGuAuwWl>;VE8~@F51=(_JhUTUNXec`|@(IPay* zaq#W>@_~DW&!^oTr~03^j4r<)RSf>hT2B4^ZnS|;FyEOI?` z)BT&ws*!eaochV*fM4O(6P4_%16cM;Z2^17RJe_T*@H4C&mOgJUkQzhv+nG(p*V#n zjnC!S5xhu^gK-;kQub{3cQc*t%sC!J=4gPxwjTlPEa{d2$-69RZM(?qwdm69v>v_m zemB!LKl^fca{*W{9@lr@>+HYZuD+rhIdjer9AJIDMTt2F6iq62x4W7?riBiSMwDhCL^iDtiyF1DF%uI` z#^j?y8w?{2{0w_MB06S7o;aXCFAvr`Hi*{2fxD3xsTU!OQLMV1@U%br&W|eD@2<1s z7_6tNNvThf*htbw>RnNY?z3guC)_Ar<6KL&chmJF8=D(y9rJd|UCM6=Lo{yqXMHga z0yJv0cs`yS$VGa%5+h!(v*UtkcM8CQkisYmiou;?G=|s4{+_LS4Te{?d0+EYMTqe+ zIO9I8_`AooZoTBBtvGVRD>YP}FTSiVcszrdaCTh#F;0KZ^tzt~7>mI4!=o=-AMCwK zHF{n!=!grkhV|`2IjRe~1uZ-|Y*pv$0ZGr!mJyLB(|}s7Vt%uES1ppmZ6M|`<4HZaR0&lAnDUnpsujR& z*UMjwt(8U5N|m2@R67f&mort<_iChe<1Y1=7)4=T)H*cGFZ&k1VzjojBAtiPPr9hY z$R<492~F|T7_hIEBx4MxgR|excu8#Iwe;uJuJxzY85-P7e@M7=qK?f@@vMvI^iQz) z9ONnew`R>_x2{f-RCa(ar?UMV>C^aKB1%WnA1z0nM?F>S2$6!f_(qBCN$fYR9ruy5 zr|S&rbd)o|_C56`-{eX^-zSz5vLS11+MP6K-q2>J_ zPNG=J801YG4gSPu+j=S`LSS0VnxVmgb0Ss|g@V$6&85RZk2&WbA{C{=Jf4nuoeFH= zK%Mj3my$f<`fUFsY*Z$;#Zz{cXft}IH+SAPZtXtYIhFeNfMQYhiF>}^_-{u+e|G** z97gk9+CJnPlLVvVh!to&UviBy98rEx^Hkk%So^fd-oRXuc9W}hpMjROhgSB!RDskr zNRRG9iBHsX1Jul%3*hw1EDQ1649 zXC|FOyG;ja`fC-(H$3ZVQB@YhRs0a$_ih^75`KU9dyesY4S+nLp$PDIwXrfo1?JU-fTY!y(H)5Q9Pm(OHIEPfYXvgVuAUFx>T0-c z`4{3FU^ndob>T5aC!QMbNxx+Uu(Dcz<1Ffz>_tO-@&Y_h)nqQ$!G0AvM;Abxa9wmz zxmqy1<6ZAS?(aXcGm=R-wnMeigkpuk63=nS137Sm7 z`UIMTtYdRwVQ4wlbWW5W=nfh%?XuszwaG(Ciw^{I^{&u{9Kt*Jw0NMnI;T9z!VDHW zn5oQ3`WW=~eK@4gKKk~TBR}m6)Xx!sMz^|3ZPWtYLSyqe1E*yxFV(;sN$DvtX!v6c z-`ZG=(e*D1_vEjfLgB;vfN%-Sts>b!0XR8Qb4AFpAWzoDvo)vUY|p`)lAeF5d&T@P zcT|YPp+R`0+T(yEABFJyP=Z81kA)F!W;d7y^m~8NaqDohWq4GBZ7N0#Sc&i~D0TD~ zl08cofh8Ft* z(g=@NGq3xZs!sh>+D0ltv9-2LNoCn4Y{vZ=ay55rz&@BU0hY1dlIhgZ46Gzj^uAa% zTC@LkekO{_+&{->bHJfsZD)fDF$u~;SOD&~pa?!D;=`oP91Pw(UY4YOzFOun)28!G8$l>>}4sAk%ZNdmA~wN6{1(d5?bnX$Wc5gvKW zw}UOUCoS_b9B#6-S{eb%3SP9XNULIFp`j+|46O-l4T_*b+EL$B1+CMLWco%hDb4~G ziRY6=VmC zzuZ3I?gKkoHyOkEtZvIXN3+R{F(N#PEZ$P7^C{=gK7GDl_a;;8ublB)`x*AuDNZh4 zh1D}Jq#HRJMFGnNc73-Rv-$DRqaFPqDXbd=AjvivcbfK1N6gi$0mGLi^ZoXL7)vih zZD~N}r})?IKE5(t)H4~Dj{NT7VtC}t9RZK=4Y3~Jlgt(ub7~jWK@GGoa>pxLn%At(8S^XtkeDCgrlv1H!%I(qGTmB z<6F-d5ev8KYHo7rY;Y_gjMC!3N3h*?r0U0agHT|)4wY5Z)@MQQl4sBrg)LfZAYYLg90$54h?=<4NUF-t(CO#7k*l7 zra0_BSbvj}q4wH$^y(5TkSj0irw)ATt!okugxwM$d9?b;igzY1&&^6nME(1-SUn~6{0n)5<2Jiz~Wol7}gCc;Mx^<80bsuB;&a~RGwOR z0Htu-Gc|4Bt0QBc5AAtmv20o9-WbTVZ(2~T*K%SK^<))KvWG-LL*E8D9V~}4e^@Dg zZs)a=w`{uCZMl0-cnfvn(L8RY z;hF_^GzhuW6FgY5_<3yvaRE%num7jG8N;%ivH%6`==sBTR!jUPK^O(#F1wyEeNj~>M# zmM!Tqq&Q)$G2+jMQt!o?lV_afH@=a^Xja8nfn-!ot+3Oa#yxp7CO%&2;0%8wRpPWv zO8b_tGu1DHWA^Huo=INTJ6*3cTl(zx&@BHraGS^Vn?HqJNzwHCUOYcc-;SScoJ5*j zKaW+p@Xd*O^3%|8YU3YV6=tJX^1 zTa2mN>G~?C+jqI2sWX{G!#LP6XP@79Dl~!MOmKBF@mlQJ!{T=LP8VMB`2kbvCnBAB zIKEGesahtAQOK;GZiN+Zt<(nu1UVx7HL~_kI5KH4p%W%B^PZ3{T^Ak%HEudQ>GDC) z8$_Lq%o`I4;1Ko#BeX`3{K0Fj5lb&?f3I6pWh{f${zE=LsZ=PRyR0w8j64HYvg1N} zUhv>u*6fU#EFkVWA*9a487T}62LD2;-0M?TJeM;?1&{6q_zO|jLX^46$?_Ha>4?S5Jv|`GVEst6Jy^M--JlT&(yv!lr1;1|LaSKe}%us ze^FaRLasrssxhfSu9zrn#od<&0^coG-TpOws%CBZ_6bMov+Y2OIPD|Kl{nn`pGOkzBID*tdc6nOLoZp~zOmL{eh3OC^jZn<;?pdm zBof#v?0w0u;u`eq({stBN#rm6)v99GahfH3b9^QOEAd`}D&C2<9YpJsGsEfzXgt}| zByPDEb6@sB!cbX~`^>KW&%oL}xMfTqxYlsRy`q~jBCCrRF734UDixQY12lFLXx`(f z0S(Eu{6`-;g?8?Cd(1B@)7%ycL}gSDb!JCz77KUW!o_rfhRi1Fs8|;5d=*O68%GJ8F~p2_mD9tLNoy84 zD`p3n^Ff_Ad(ex#SDLA6Y5~28Ahwo;rtc$4nzW%7WU)@xiGc$KY^$Y)6{5cB9d?^WeB~rNytIbth{@#7}hFObp`BBM!T`NHHq)v zhgOOI)s!Rb-SGn=F7%xWJC#S z1mcj8Utz`rL&J{2g98obBo+}Rn#vKY@|pj$CvfP{Iw4>$4 zckUOzC0%U-ChTMUG4#gFA2Hah;iu9ngr*y5wrXR#Nrt@Y8FBRjL}PUPZILd?dn#1! zY7|R!`7Bn$x0Bb>!7pKQ?>C&J@0aK}b-MjvvhOzOymm;S3ZuuAB%0S_a>;W@OelHV z$sP4XONo-61LZ3(6D$&N+NdTwyK}SsyR$ng>{e!B`*s=LYDG8+*ANJQLs;0^x-k67 zYoAr}h2cBHXc33!7ullE{ypG-lQ-E!!*?E9lRvg;;TBqBUU_)8k6VqE>(F%4pnU-H z1E1}Q`$TNM5MFTC=Md}@hg)1kb(KP#rF~(jCgiNk#B22j6?D~ofArbF{ASPg0C?*b zvyzk8K1njTL9_8wTUYSjn{zt$RMQx;a$xsYgn4h{#&l!tF2?S3U!ImUhgL)RbJVlG zcyz4S;e7X)OZkXQjA1wPDShgmXI5qb4`cj5NfWR~y-~n7uHV|`7ZdfwYDaEiQ~E>J zLP#X;y^zS5ni8uWGAbzYikn~pp9(`a^>fIOv2t5%o;7p1nO#fhy_O7kqgt0}e#?2E zFKza-pVSI|MuO1BZ0vJ;1%X7?#W^Ygy4gGm3GmFZ8@VLBcLdXAw@ z0KFKbzb^_5hW_sC7|#&%$3u@2r7dW-GrMxS^s*yD%>F&|)(xTdAjbbUQd7HbxXFUy z2wl9QAG475@MRjg4&M*aMKkl9&N77P!3LtcHO$BAAAjLb*H>r47%`t}7f~9`Pn~V7 z-H0uF&Dh+cUyDjadt4N>VxnaaT537-TU(}8P44^7XA$#E%~^YmB}TsD;Z zsi`CGN&^ywu~i#Cu%24q{`WGS6}5$75PY&GioNrJK!qPJ%C;C!Ji7~1nkX$=Qy_`- zh`3!|p$~6qi;v1vGPN(@Ge0eueE=pH+A*;ocH>1ze~co;WhKJGTzbyOF5v@~90vqz zTE`s>w94(lLKAPlux()^OYVy=;WsiQ!UZ?JNRkVatF%tujLf5QifT>LV?M1P@{_lM zy63wO{%MT9A%CsWHUSz%HpBAgCw1Cio=vrsXn4&?J_O{1PQSl) zfB8iiJHuD3)X~#+i09$UE~h@Vzof++WS_~iN|z})YTrt4F!~NM-(RgYC1RcM<6xZR z5*=_Lt#}ei&C1c5lKZ4PZ|s%R~9VJ9UPT-ad}; za49&HX>X~3Ioq0Ied@qlhoZWlZ`!2uk0vFwH@<6m?G?q0P!KSl8ssj8cBucHZC&oK zVWo_CzH0ZOpW)(dF{{LOOAaDw=~zr+G7_nxOn{W!zC6xXlprv>`gKyoNBL5O{Eco^ zpht$)2W|e2GxmFmqu_f-y6{>rt+}6noib&B7wgNL{bdb_627~e@!mmE_Qa>462@%M z8w-mo6g_xfld)YsbD#?FZ9DW+1x~G)Ao1r#Be3E9O`m{bQt1`^CA)o=Ry3_fuq?~a zO*X5Pke08Pf;M^hJ3ip`VEUu0%_e(OvGe=~DPa6OZd*Q*S;?2;h5Gkge`KYk(`SY( zT>V?{ePo2etux=MJjWyj^d8zP=Hs`1uM2LIJEv#ca9-zLYbElq- z+yw8P-eH7ptE3$0%I!;g#JSHnOy&co`+%q3XD?r8+~A{oF}?M(#=xcF0`=3>nO9$- z6#q)_Kph%GoZPGxu@fQ*BL-6wG>E?jr%>yg2JgOH@0ZyruU-B_w)Le!;IB8gd@C;b z5LRrenMG~l@3JgEl+idFG}2C75CyERiBY@Zyg!PY0CmIQ1)8Fx9bc|HyO zPU3Er<&vSmZso4oMTsMO(5Iln4*^7|L5utmK2CBc^o~u>tgK2Dm32uRF98YbmD9PBeVkj1p?FQ84O?NdPMj>Y&VYRDn3#@*y-%D(_9sXRnzbwMzbt-iFx8X zb2jNv1LFwBe$8z!_RD12iWP(nUtGL3xuBK%1Y}8`B}xt1{|vSD(`m0NG^P_c9e`yK ze;R)_5{8kJcN{qg`8INdpvOOsqRjL~{R4ja%9|lW7B_bh8byx)OFjhkwJ=jACS3Ys zRy-^F^CWX6GXe9D{=XFb@=o{3q20xEWY|?u%^zT2e#kBj>lAa+-y^YThgLF`HoJ>| zTC#W$OpZNvT(;j{L%FK_sr_^H4eO&_DA=?q7E7WqoY~?wxqKtzQ7~t2&qsea$6>$f z+NoZzVS9Ls(hybfa#jd?fbeH1`C1auQ zY8NM*@kX=rMT$7`0u5~u8D#EA{75JE z9Pf`ZH^13t{*Z$20zVJ+<37F=yvD#-2**xVUkde5=yIuE4)1XF5ZJLgw|lt zP?3KEdGMss9)DH5F0|)!p*^{WhBsyQ-9pV;_Nyq_?%#mIR7es%)yFy91u}CX6sqG1 z5vNgqm34>S7^eCx94wfI|DZm)Hlcgnc+`I0iXoT5@OCVHYsnF-eeTd@*kZHk6yn*% zSnY0WI+1ISajKYr@udqG$t309^qHqnSd|14d)ij9jAQEx)Jzg{dP%%v9<9^H7{)k| zgD*(aY*~>_jv%Y|tHBN4%5m^^vY(bkpEe{Pg{Q(N=@|s^J{ogqqltg0UtGM`odhrf zGX{DTjbYot1EhM4cnfVjm40ut77N~e{(;Y;f2A)y)U|p-GzDh7q5nD7jVsT;;yILL zZ|T75yAA~U>+kwmV~3B25FY^F+e(T@7!jGAR88JR>rq}d29O)-Z5^%p_%}X$?zaAk zjM;Cu26EotyyiS$7u=BrDC91|u#X=m!7}itG6}33*oFy6)vB2EEN9#VNe=eB{DjD4 z`+>WbC?v1=+afx2=j;LLt!^qS<2KQSb*AZeB_5w=hA6Y6!H4{(LZYbm*Sx;E=gC$B z|M{FgVxb3|D=9<1orjPBxLZC+V7!n-Mq*e8207c20)zxAJ~+9&x|BJa(muh>$Pv&y z72(!v$IU_i=Z1MrIHd%jT#*DFD=2?|;TdeN#Ck#4wgvO4TlT0|6=)oPELwA8;zCrF zgW_gr^Xb^X2kqhn{l=8=Fu~6nF{C2rr$(P%GeS-7Mi|%M-5@pi+RCQKjo`%dJyi;K zSk=r98pJjnD<>gv-6XPNpOC2Zc zD%+Al8pRwrek(^GCA|kH+JnT}Y#lzq#GYHLRQCUu^M^a2$FUhW z9X?i=!)=06hxiVJx|lDfdtFJ~dEj=GX}5y&(ZZ#8%;SsK-!(2KdOV~jDW%FxP!*lo zaCuD@O=uW7i2PbTXq7R$i|N0{7qyM3Hehy0i!+mk@|oS$Jn%yb_XV2mK(3r&8Ic&g zhB~mdcz(IboHJ(Y2fpb0nwm?BVeqbQ=NX|}e)HlclW1&g^V|o=TeAak!t5op=+A7e zyD-(9)PEe`Y8S3iUMC{UNt1NGh6l9)_*4~Mb5Wbn0JB<5w6w+O!4Tp0t?5s1NrWr@ z*jx>Y-#c;{kKpQlZnc%nSVMwpL(#5AT&;RdjjtvR5D&Zyb8&ie58bvo9Z1%DO9(wAjc7 zu8DL|F{g;co;_*i=)B^Sh&D!#4ZJ6qAw+%*u*pJXo-)%Jt{}`pLN@95PBJh1=BnAM zMQ5mF9s~$z%w51W<3~Cwh0u9D(XRtVFnQ%lmqF13EwP*r(`bNtr7rQ%*TIyde}At( zrwpy>1(T8;yFww&x?TSsvwXzAKkiWI~L|^NbEiOi-0l730LT=koEM~xW3H(pD1N^;kB-(|4xE5e3W_8o}9_w)f?rG zYdL>nr5U@9u_Pl2mUWn+s(Z|1_5lh0t^bo=E6W}Em5|K^r?hA?UEUdR-&^(-yiU`m zPAVYZn_v>Q$U~P&oM}S#6)7cs-sJj*dJ3JZ-9|k-Gl1$l1B^6yf|u#&>zXeMp?J$Y z?-xR{46Md)NZV^wSUG6aSed4PK@*|ZnxvRRD-iWU{^h|Rm|NBU;(uParVPK}kDDDP z+9`RI#kHoZR&ax&zIWi?)@71`j;9)dOqgYkSnx7*ltwYAO^Z5xqpJ@u5)_Qn-P4k9 zwWUXLOZqqU9^hJatD7XCT7vtx@2$ViX3ug~Fm7{^5R$8>lpc5rVbz8%HHkX%e8Uf| ziw}hY?&$eHMq7oSzQA}n5OF6t*LjgH6(ipN$(e2y!g$Wbi4U9kSx!LjBgY4u*skFC z^DK=r=ohl?8Uf=`RLJ1vLFe49n+kh!%5cFB$^DvbEjen;Cn>HK?;rHrZ(K@0*XI2;AV35_AM;9Hdqq`Aih%=PLH(UR-#WxnZOr4}g5HPZgs4JmNz6ac4i;+; zg?#o$$cF@>m98G2c9Ftfnk-DK<=%JvV>C{JYq*YVdSBICPA>c|a9qUtyT5A)Ux-JD zCA`pX^I_lTr!-#ZvnXGD6cA{yPalqF6c&I-+Z=HIrmmaT=$uSN>k(#Ps903QT)BR- zAF7N|=U7t&yrSB6akHm{XXR0KOeDK2;`qz~p(@wb{ysoad*jt_tcjqGw1oW_L&oZ^ zt9EH#jnA;Lw`5#DLDk;vgkDJxQE|e)o4~8KhMm-<|9E8*FRIS)q!Haew##AjbEk)F zyNS1wmqo5O$Ddj0qmb-g&^F4=c>0^_uI`=A6?hZ4cWsk$~VdA4sONxYd41VHa*W{M&jG&Z!l()i5L5Z zRj{&T0PG@y%^r=m(3{MGcinCjog9R74fQhc#^bZsB@8J7=B-e+w|u9)Vq-PY{Rw`D zzlCOA4{~2?F?C?Fqr-_p33!{H_RnCs#ChRRAh9@PI}G0uIkg9qb+my?ZE)l za#CGv#7^HlF1v_4_}625ZJAq={!3Z#^QxHw>lYk~mE}iQwuT~RQiEvNa<5$`&HNIw zFnZ?MO!qo&6@hJC)jj_D#P$diRR8T?uV0a{jxMW<8nYX9!e|rMBICPRqKm6`@@w4& z2`R;WRKzd4!zdSZ!YJ@qljfy00sM1=h)1etJowlEtxUaxV`n^eNb)_{YDqP4o8NWs z4yiy$DrJH!=W`fW!IUo+^Ibf>Ax3^wD=4^~>Kl~HU$VczHUz{{w-xf2_gqrh+p`tJ zgYvh47LS}26KkGHZc+V2o3+_+8R7f`-$_mNpP@lX9rZ5B9q(h#zIFih5watJ1g@`Q z=|X3|L&uPcEL!mI>M=7U?SM~sD*tUm&HP9)S<3Od}Giat>j*vc3FnL0k%+V2OEYn*;5cHsVg0*uRrLx zYDoFuIxvH+F448<0@9xyI<)U92CeOkfKITz)|Ib7Ahu9P{Z)nScN-l_ud<9NLu_>& zUzGvN#ODvL(&{m1|3-Y~|NcqNNL7_0Ii!*pilI-5`-i^<1H~ zI<)t(r5`HFNeNr9B=YE?3>JdaYn{5@GbSV8RcfPR3SnCif z-d5b}1Qgf@GuV@|p)#`DbAG2kfJxO>NBIq(Iq=loD>M2@R;X)H*HGH4y=i0dslJrJ zw#;Cv0qg9B2BKoEb^_23Fmj^zvsC zkFwJtbmN;8T5krWod|inCk>M-ngu+eQo5)|bw9}tX0TNmf@2?mkWB@J|0>?4B3mhe zT*or?d(t7EJ0&Jx%zM&F-TG>K*-NU^RPsweXugC2T5Qs;40(UJ`B52r|CzxdVutPN$lhz7Rt=RErWAKV6LtA(<+-@CuC0+O6oc8Thu18cBW zUu=6bBgJ|Z92TqU5%zcR-OF;tlUG{GI^mj2E1QR(wc#GK*Q;el{f1P zW*RI+*c=f&NnQrw4_s7E*c40!uoZFP;HprBRZN$YVq@9o(~WD`fklc%cD;@SbBgVn zVG>(^yoYcGjt@s9f)|;*SA%`{$_os3q3L(PH_ibQSOijUg~Z$&P-hQY=tOp0=QjeK zJVWKAKo&K3U7{?^=;|vUjovhFMLqCMmOi<@=n|o-C>;lENkFZm;065w{HJGH<4)I| zEl#G-8pQRrmu}>Q)SX`uC-doxjQnnFH4ISVmv;57jR`o=0 zI-s6r8#@6k!74e5(Lrr&Rh!CdZQAqGzoaMkQ#-FMm|}YaS@5QRiN{`QUeaC>U3JIi9(-QfC0H|)g^=QD0Vs{f0vkv}~1)z{;t^?^6httxf z4*-YEYc4=Fh{F=Ft+2-qh-giJ_1Y9?r|eNT+14FDm z_{0Md0i28X`avUb`p&xW45J@ZDft1&X3smCdrj}zT0ms~R*^P6+E9@#YF#_>5>$EYecvTB(a7(vqTzvrjUI%-e0zWb|cLO zp?TOYxbsh@UoMb#Rb#IXoN}`B^purM2vAYB1g6(T-*!DX!0Be{w77<(;=y6= zihc7&3VrRUkxgMTY^>6T9rmIk@n2kj$Z@?FFn5n#%x({b&yvGb!U2omh`os&1%$w1 zL-ie@Pn{@hBlhIEul(!boWxQSi=ND`!-{+AdF^z?Z2t*g>u!QY`==;d($<{l(eus5 z>V2RJs4+b@g#R~90co;DRKfL=axOCVmiqt6zoADVBby?k@G#(d}Pm?|i$^@Q=>HPLnD zX7la82YA6G1d0EO-}wSj4wXXh6LMMz=QK3ai$N#6JIzQ}x|>`{61jU_JybC{N+Ic9 zN}#aZY)%xjBxX_k%7V;2vl%*@4wH78Rkd*5w9gsL8)^LoefOTyY8f)SZQTK6C0nz1 zabbqaveOP?v9~&j0n0{zodni*8)_((r9BOF2tBGK=cV*$HJdA8sulhj)s(N&VdhiO z`yjDhxw5Fjw#3T3j3*}=zbAt3QVTy>#t6Oi>&yq!<=Je9ZY5PP0bvoPPvo8_O+>MK zhi@Gj(Ua7z#Ry%n$DCjxnPku|QeJNlK_qpQ#rPTa0nyCg#R_pyQsxR3OVt5Xm`4Sm z30T{}+zv&O)kS$ZMVnmXVBD??_0yJu2z$kw7ZHU-AfX1*2hZ zfKYQr+w(2=SejT&-AI!)?p{ATggq9{c{xkr@T>~?(JRk|Vm&y<2A#N{_d)XMKqBU} zv0sG|6KdsI?J2Q8;D*wOrRh0kZdTCT+ui=caaeB8SNG#;uojv%Dp=HlXFz3SNA#oN zdwh1$zv&-gH6=2A|NYH0YVs_bInxU=-A&ETfES+^q&F@wLd*{W54gRSa1HB{zx;Pl}4 zPF^Q@f!6HlVmZkbA3(t>C?&mu_^1$xMLVfPPLLx{Uc`}wUZG&wl6&U4Zsb3&UC$R0 zr%ciccQms5)BNTdSDDVaQT%lRLGBW7(yx>JjTBwU(IH)_P;C!gRq0ffBoxmoG*z6@0lsO z6P;I(v5HpNyLyU=bADbj+zmYcb~z`Ze4Z8c?3`5mSr0cQ^lLSD7F(C&N6MP%@}4Cp zjr`(yT>E?TjP4DS;G_+)lfss5k8_gNp6jt?flf&hWh&Jh@0;7MtUUGQ6FA?=ggGB|Fo6iz7TT&z=v!}`c^ILHYD3q?T|rImia9m<|(rjSp4AaFrpC)0?@JruY`{o z5O^B1`{*Vbn64c;qGJyMreiDUkYs&d`!4UEV6H^8vZ%($uF(c4zUt#r*oD`mNKN3x|&K?Avi~ z{a>h91lULCAGkk3IJnapePrxvD5#x^_CzNx{R}|!wnSFEE`s_@n9Lm5YZm9NN*hm- z*xOrhMlAnew&;mhWtNf9B4Wm&&s~I2*JRhv*BWc*FPScFB+=F7lP0MWp+k>UbLRG7 zug+4*>twj9jt^}kmv?P9`hN)J1oXQ=%*S!u)dSRD7NPhZ1o$q!(RRH+6`pu5%?r1< z(e3q!ZGMJXJEK>G(Pnm;62rVeE2v@$V_jvel|z2!FcBOlK;=SW8=s;r;d3Bjr9&Yd zI1E=$X-(**Csup^Gs&MtVCKluq3?Cj`1X8vl9+zfSPM}|zN3@LAU#St)MiJkzKoJ^ zvQi?vG%JV2uNL(ESfhGy)z*a3kA^w`{a)q9z{>$;x9|p@n_cbsJNf}xQFOqb3MF2E zV^f~Nrv?omzuo3G$v1L!(Ajx2)GyD@6s4X+?S66Xob)xgu4GbZ#7=r?(kOe$h?x%| zBN=9Ks|ox_ML~=Nnh}z7Bd;k;_lB%t_|Ir6HrrrA^DRmiqM-5Bry<}CaW;3WRY_sW zVEQcxb|2b=6(3k({MYDL`qKq=>x**(C9Jm7GqciQy;O$lmQ7mjL;C znIYWe!C#yZZ|_!Yw;`rfF;i|PZMX>A=Vqlm;QJg8TcnvH0&&j09JX@RH806**UW*2 zfS=`k1=fRNxpb>Ph>By$p4fBHLSaW)w)yHTJwtQ`tyz4v%f34slPsrtMR97<^EziP zYk{ZJvzFneU}-q=9ow_G6KpAGJ%D=w=-Hpk|CMye=MTv_1MBL#4xs30E6Rx0$wS;l zB@-11NRJylfIQEp+Dd{>Bn8ae;^lT{)RNx9{JV9z%<*oL=j+5!RD3(|X?(E(=6KTC zfAbFlwNnK@mR~U*SOg8WaBalBE*_;(`KMVa=Lgl+0NRo&faR;$&i&3m`3)tJm2xJt ztH2rVP{4l+GqDGBvbYgGVjDevTvxNh;V%yC5( znm{6!-wew=q<_)qLs$kRL&O2-W;05!4c8(Y=o|t^9Sug8Is_^0B~vGA$>y5 z8+;8{&CHu-X|=vaU9ax-y#&?qV@zAC&w5-(LG#w65SDCBo0yZP~ENa60o= zO+-lyrhXqOz=xdC18yw5y*_939k9h%uvFvu64`FuxBg1;|8+xE3+a327|nai@&kTT zpRF6YamwiM$DzD#O(!pamyt3Ede~2sDF3W=#8*K%>_TKfE53WOG+W?kGtz3cz+zN6!5o^=0A{E;q5 zZhjT?b#CJ}_pr z!OMOmE%D(m5(e059V6o#&tx_2(>NdYwJ95NZ0>Y4PC*u8fjy=T^fC$dL*!nIX*w#1 z-RM+Ov>htUa*73V(-siq%(A(6Uj^U0sn?#dIQdLb%!d*)?nj;&+tIpc4B;H_wcA1~ za{fd{KnJ~()VT^O4%4r30MSU)g8;VN7X^bQRLIZgu+QUN-f>4?39Y{{T~p2Y%>;m` zNe@-dxh{*i5wARRqeMJ7Bd%L1mioydBFozuzL1;X*fP6>bDl{GaJkL#~sr1!2va1^`x>#3WN;H{i3(?V}U6) zFZ>RZv^54G3&_yo_4lpO1x4j8MV;Q--V&8^DbW8)^=@D=i&VPE^`B5F>Woa4e)C!i ztRM9E7i`-w34Rm{Vf)%U2F{6G0!tQwMN>1Yw}UQ!QSqD4!4K2sy5d*+pQ}6qAKT`Q zRsj!gV*L9IhFc;yG4Cl^Tz}-`WfzcvJHUIS;`QJqg2^{=C0#8&mQPOJ3mtYf4_}V_ zF~TKU{pX=_=yJU<{A-?@c%Wq8yCYcQJFrWz2y_=oZul~qu?q-3B)%p&7ax5qmrsdM zk(0~ocL@d>*lcCqA!_37^;Bx^-@pI;H2m=w*Z)e_@P@eA8GQQYow6fTK2giq2X9(x zcH{W>vIaLx%ss}Xor%U}eSs+A2__2Q>BMV|>Dv87>EFnC5^9!d1(M!0Fv@dZxV(#44_iCf(ap7 z|DaOeL*?@2U|l~=jTmF>>>U+)+;I3WFmb#pxA)`5MFu^REICaO$VVw5Klbqg1el`Q zTcoH^WqJ>)BN(?|J8{JN6Nk|9!{ymemCf7a=p6*Z36MLkWQ8k*BZX?1_%_klvqH)M z{1*Jba(DQBjGg{nI7CsjchdQ&m#%7}NAUPujAn!+*WCCZZh{taI(sqM&Re)LFBc=f z-2a?WX&EJ@^7Fn-6)@tPyHgt^sH*fLY2L|}0tFkVF+k!F7ue^-4bNmWr{bY93+@Qc zAF^&_sgh*HU8A7lWOzju*viV=HsIw%S}$j+K&bLLyBq#FyN9YnkPVw#MVKVXMcOmO zcpseo{bJ#jZsIS#NnneP33Q9GQcZcqxGUU1zvW5t^NBZHg!};Hu zyMMYOg6B3sYuGtI>p@Qm4|fGW`I*MUpK&iN1A66GNJ2th~rWy1r~R-LEq5Mj4N{_A#R0>AqgUr>%2^V9hPchoPeU4 z7IYm?r%wYd^ciWk=$=jN3Zf7a#NJdX9Wo-5bHLwLp#EvzS2GVk_Ow_o_tek`cNW@- z?~AXGUv#P?fV?8iN0i_auc)2@`RV%W1LOvBt4?$zrak}$$yhH8>{ z-j5>Lr|t-Lg6N{z`rvtLGW7TGlc0yuz?!Esvl`I}obIA4@f^;8c1`-FcZ1ZcZ!3?J zjnHwNJJJy#^==^~2TFxP#BB1@rxun@3H*Vb7E*NZ$gSMfG=JLC`s?jRKZMTGJ~v2C z1M%C7(E>o$U8_JzyxvY9W-b1Yt@n`jdLk^&I8OB3cco zAk@ckGuw#ufppLJKAaRm>lw3aA{$aLfuL*ERj!X?$JQ5`CVGOgUyy& z%g_&e{*kqJyVMx->&1d^9^bfm-zWOzO@DY6ajmRLjij|#d@UAt$JE4FsE@p<@{=)E z^NR)+g3!){NIdfS?JT=2s%&^WN4M@0>qv_2g$;sLxO>8!)dB;glUV?(2`vU1y3o!l*5wVSI$)0Bw7QZS*Yrs{@7nFd(>-AA70>|Fm19l4H$Qq(i*Oi~C0v%j9N+%lL1D5s_O zg(Uv`90jVm_I(DWwuKJEf$IFY1{#87iV9Q^o5jPuEtuTT)yF3R;q@6-d^O#g5R=+?eOAX$M$(Qarh?)`$YWg4?V<>61~bRg0oN}~H3_#CIdsNu zP7Rte%Add9crFt-V&zWR<7Cdaf0pkegK&DHTtLX^?!+dQhA8=l|6*3idz*(=e+7phTay6$bXc<%c>(~g(UAC)h3@e3$0svTO7kdKRn_GzgHRhB6V1;o+p#UN}7 zA%az3t%5(bR4pcNm^40nya5?|jY>GoeuR`zLi-kfO9(tdz+1EA=Tfh}+$yJAYgTu0 zk1x$J=!iIOcgrrBuFKIuv_2GWmpqTT|1Dxe2{OEA8N#Njfwibk3-(m1wSfIyB-@dS zR+U7K+GQxrvbihcX4L1joX8&d6927TdNiLz1WbJFUZ%|C9?> zbbSkX$GB9p{&BHx6fK)}Lcxg&U{t}v@N4kX51C19?qJlDOD9>w&(}Mu((3xfP{BcQ zsS*##4}^3;bb3I7y=SLr!94#x_}ade>ewX;ZBmMT&+4iU`m-?gNe8!QpaY`f3xp4H z1~9SI=lLU+x~_T!k!yJ&P#JpE53bvreEAVT)u5nb2s7-e8@KhNrT$zq`*uS(jCaJ0 z!nM4u{@qyww8Qr5?0cIo9k-9yPdl>E=B(oqamM>|-~*m$99M{Nr9H!;y*&& zo4zidPB||JYjD^Y^PAlM*r=TUBIefljsu??KFyc=)V8{e!?&e&yX!7N%en#<+R1o* zzkMJxr$J?Rfq>|=2_-P&3p^<$ODC?yVVKh0@GAvNFUI%^lT?(-P5W>=T~fVcL9O-W zkdJ+k?ueueym^)i4kwQ1bZ$*^CrLH}|4I#ozTPbGfpqoN3X2rCi|`>~ela$LW{)k{ zEdv#?=5dz8HL*T8lX?EE6ym5$4Jyd9B>LyB0s*j1f1k+pbQD{cZEw_=BI_x%>Q_%m zi5Mh(i@48Wv6U7l-J@6s8M)H_)C>Ve+rH$RIX~8>LfCI@cGER>$Y#(~COIR~KSMga zE7-1(mJRX^(^pFIf4>9NN*0ni$vpbyIy+LG!T*%K9&4FKl&^jf?eRC@TjIeLn)(;D zwURnF&(H{0!q^oUlYsb`2D zpAE(3tWbB12Uu$>zCVQudxJK$bSxq$lpM>NP~50Z=FY#%Ems@|3PJ%k5asEko) zQ&r0OmV>9QymcnoJB9|=C%L^Si0pD!2N z)eSK+Y+x3Rs!1AwokWOOL;X&w68i&qEZm=xfP zBS2+f`oJmg(fM|xL3Q^vPMaU*H^hPt^zV>s$&@j0I+Ryo|1_Xd@E-lQ*cUMGZRAJ6 zq*XBW@GJOU{eTG75JkFn=EK!6`&m(VecGv7xZ+h~J-i75wQqjO%%O zuVXP$Y5%+)bhEzBD9Z@GR=zq0x_f!5OF@v6K*&R^5f4SqcwQi%`=>xp$=c>a(XDF9 zvFxm1nNW2Qrs#OHHC_VyDPxBSliqKKTKStw8z^_A9;}J#VCV@!eUz&Gv60)#oI84p zWwonCe$k9)skEQq@K^xq|kPJM94=PkQic%D)at4+(!J*5lGM6Dx5^x@ASGrs4;! z*n9|QxN4hp!Ix8y3t~Eok)vW5-M!STzNAK6-c@oI3}&g&`+dcu6pC{)5%C=~pPwU~ zQ}GZ{Tl2m61R$_-WUy6wrB__Q7VJ3pNCdpcv@3W7++o156+Nw3jvE!sqMW+OGf zQm|22YVdcV3P%#nL%8G|^=7n|2A50pi^YdzVJgdFYOCxiu++`027JdDJTi_=IvQ%4 zM5A%G4UyO=^3YHx%hWB5K8BLe90jP?J1!cEOv^VBm8^1XIN4yx5QNb86>=Wx**8%K zfyTufm_nZSbttbw0@ujxHR#9<#=JL~#aGW#*e-{%3-aeSjfaIVXJnbzs_7Y33sZ<|p1C)rf z42Ym34Et<+w1Gywcd@DAy*$ER`}+?+)grmf<~S+jnEPc}7Z34!q!tePeVuXnrhk0T zeXWLPFiQNSFMsi#UoLBWUbpS&G1oE6p==UI=yIsfA}yXaeffO(t*MB9&^EPQPWU8 z`4D}hI@z^nl!Y5*6zQ+%xz;Sx>t-aDIykQ2_|4(E?QpkZSws1Q>(3R-)>5A773wtE z-u?uSekI$mIcro(OpIun!uV}4i+dteoeOC8-2T>OYBmn7xmIdXlE*Xx`{EECKkhx$ zg>KyxCGBd?2iZQ^PCVsO(2IQ;gO9i%)$vi)d;FmR5MeAE<69fc%)T2i9s6?`N!E!w zmThniDm$^3{)Dts3VRtZZ0v6l8(7_#yK>HFxK1`}?xvX8(a)yrJNXpQ8yw`Ye;<`1++>|A_iI9z z>1L5I_mCes%CG17Y|;#>v13;~lh-(6 z$0o7-;YPSp*?Y5{Twa1%;!X#H68qG7HODgNVAmh2^QJ-g0^Xe&&SpUsF%2zPzyTMl zt!@KFBtA>k_FKfLl2p7w&tz!ZD_%XtedQ0u;XNg5Fv31pZNF2-sJ&erUnTpwlHD9( zM|IAkt{9oUHXl~sPg|P@;*IAg`|IH$M%5Yo%l>O6Nr{%;WBXD;p;^+ChK_yQHp{ua z%de->#~+4Q#|ckUpl!1ils>`LXy=h!&SGarh+O)tzqd()R+y9kP1ChDq ziTEzxH+@?std@}=>KZsQ$pty$-V^51T}(J|KkHz{=wNZPj-;g^wNr|(AE2JhajEt{ z6ImYn6w#bGF}>7gs!m&b>>{%2^0CRTQ_d1t40$9LfBsHvw{-RL!Uq)||IOS9tKEcI zU%y0wV3l6CcM}hhb0LQ*^9acs&f@icZ}ia6^PTgfZ=`38|9Zpq5E90?eus`nK9I@A z?ob2vtBr%E!u!zr37;6=)u-Gh$$QFgslH;o>R+8JWI}zy6MQ(Mi2e3krfWXuip?e$ z2g$-N@Z|ys67-ON#4la_xMdt%eQJH3CotRV-f{y~vH%zJVJi9s3H6@5aIV^;P*|w# zDy8v-Isxj=Z5reZx%)D`r%LP-zB){s6gHdtbgl+E2L2rBUis>?i`K)RKm1N82s}Zd z!jVSXG!ckDf9I)rZF0HOSvi`dO}XLog@PV$b0&p^8mWC_b$V(^_KALN!38#d{+&;!JRH1VcBgeBI9c}C1E^a0(iFDr4y{Vpe@%fq z2~*$^GR|$18exdw@l~fJECg{){bJ%cCB-BTPWnuzFR_y6qp=~@>j1J(+9=DIijm8{ z(Knc1Cz$j|h6fbW8^!;c3f&rPu&j5AV&C!bBA8*2+}f;jHEQ@#JxFx>?c6cl%gN?* zb!m$zT~_1#xjQh$eKB9zzz}xa%PllkqyO;&E5|8yFunS@;Gcmq_0nIX65dozlm6%A z>yITw>w*B(bBCmiVZh(Y&;Sk>%ov;hnettu7_3>E)2%KbX-g-A8i3DSEkYG&k zk2%zrZNp!=oF|?9=tE5iQHVu%UAg2*!5EezINre8m)}=L<=)a=ti3P)q4WupjqTwj zUzYz}C)8+xD9!)0P5=L2Xvl!5QWyJra~R6)XX%#tA&P9;qaT2n41J^ez7icwDJZgP24; z_Q#R3N&g{bhRX4J#mZ`NnLe<~bjJT(Qd4jYzRCZ>5eulbQfps66=Wz$09ujLz+IcB zp2JqzUX{4LQIOO)hD+%U%n!OIt+tH`Wy}Y;=E4%??U$T@=btnDY=o-{;18_;0+1eH zYWaKsMr`9|!n`EQ91Ec&P!fBdY`1+YJ3}E_WyT{iC=S5V`$?IS@NcCg=rN~%ec36y0RmiXt)VyPhQAO zyut&|&c!#?)Yv;v&8WD2)h#~*|I_1X^^eCg7d2pRK%Wj4u+I+~I+9yxEf_}*QqYM~`_;foBEiJm`*hJhb!}V!0=8;S>UB3Q0zX1=U0vCkL zDMPA1;lE=aEgQ@6{iRPXBW^&7AfjwQF~|2xik>`>Ce-?*CAp1NKd(8G&5p*L25<39 z4-9gf=g7gn-NFdoS}@p!us0#ZOIUVa;Xe;ZiaRS0l0m0a(Wm@Q?-vDviW4iPb(`8z zs{7`EA=nr{1^h$316-{Sk=^?s(>=H8WI@UaTvV6*3#_YeOR|_s?MxV#dmqnMnk_|h zT}&u&JyV^y=vh!D0yzzndRqqjGBQ4Td;VxCQynfgCnFtr?@-w9uH@WCzKXDl@&uohp!W z1EF^Kg)~b-^q!eeF8UD4*}{Ip@4)<<_oHZe$#-Z|bdVo+?A}{*>%X7NZD2GS*h5hVWDBj=@7GdJ?gLW1YFA8^uC;CR z<`!wNtK2txE+Xa?31y@rVkOtNV!~MtA3k{_=s#S6oS`Htd)TOA**Pp~@TdRCUG z=F1YuX_c8h`^I{cWWx7M#9Vq1?Al<{~sjhy~ajZQXt=``?b% zwWnEX`khX*B2YNC;vd*bDCl{I>B1htoQRxh^!aJ&;*KQqT4lQ%&^brKVl^h70L|{ z+3LAc3U|f9Ro!c~!j>U8dT?(+>!-Y+b6dY7GL?z>t@B|P_L5Ni%$sb?8%!pe@8|Pu z(mj(AJ#@5v2r79K-wal==ZySkVzjdxM~GkFkHU3VJ{23+9s0KdME>oQ*g4xvn z>Lq{v7cTm*OmCr~2rMHC=b2>`Ho~=6z787KdacD<`4$FgLQ~~ z-N^$wr*~K0sORC(&7!IvHriz>=>za6-IbX+WBkFSXv5KMZCQzgfi_bEr=0`Qtm3qP zhthu^WUM@-ua)bPISa=dCToAf-sm9M|C7Kjfyi6RM5v+k|1F2_$+qS3`Ih{7y@bno zDQ3D(t4_xLdrhLMZjRar)hl^zy#^~&4MzT}AL6LoCoF4BzW|vgiA(Wa+ z`qZ63XRHt0`TWL#oy!36cq(?p($0nFrwErZrU;nmI@dZi)E`#w#@#GvhG<3m8P-{`lYECxz_!%&X zJ{kweUfgqV!tS$b((UX6Lza9}txR>LtP#J8YvppSL~Pu!tOk~JDYzx|Q$-T!ewXCz zi0_XIj$d#GYIAKsv$s|Pc94T0k>V&)hVJ>M7e#}7W8$Z-c^#7$_y1X|tg_&Zb1X3>`5rBRHJQB@aQ&Ak5%@Oc ztDnp;Dq;iz)}s(7Hbm;=Ba*Z13b6lfM@$2cAjLm&ERpen;3*6$E~Ulkt8D+o9Z#PI z>mu|4vw{Q!=aNV*RL!kaVQ+)M)%yAO2290G9W)-~iRLGN45|vCJ^^!k7bmd)x7W8$ z+(-i4_-nfaKzY7HIuefGWVexAm*oJM%6(F|Kkhu38`o$lnDgG7){%C7CN2si*XT3{ zEW;9dKv?hZX|R*3Oe-5w%ROzm6TJfNGPw|&!D)m?tBC5!-@m44aw;|W%Y;v7h2UX& ziZDCA@xPw~@(mBk5X0kj&Lc#Sb`pGQj}V2ZL~xv8!X^XK_#k*6N4U7mlgkc|JIYzi zV>Z44_r(Iz#=26VPH^h^uzOtd6IMKDK`NZ~<0xC|rr3JsNPC#6V}k{! z&Qh-DQmJUePMPFdx+3PdA8bHt?4k|V_vJtR?~jp*!y-BNTJ)V8)<0plN1qm#Bu243 z0!C$kV$Z;N3#AF1oz>l$o`6rHcg$Gx4JcCcCZ6l8py-ThPlMae=i~5e<&{PsyvO5# z+h=?tTn*H37w@|`s>?x`e*1$e%pH-R=H6*VsR@cHopIMDg8Sn1x5mAaKyIO|7uX5}ZS2Y6LZz$w#!WXL}FxuR#haZTa` z@Vk1)9sq8a4IUhx>6AAB?c^Z|xEmy8@rK@jvRiuI>$&p~h@L%KXTX=4M#O`FplP+^ zvEFyM%_I3+A90quwVYL$ zhy9W15|0Cd9L5Fu3Lw9~q^=Bj=>-X<4@852YQFnmbf(6HkklMzwR-fYO)c>}K(Dwf z@n|YF>97p^ia`5f@J3($zcLWuzrS%T@t-v_`|nQJT7EnRMC??jb?T9dUNz-!E+o3Y zdOv5G8HLF43ldO=d=xlTE!58BR|6K;V8g*wD#;WJ2P#`TrNB*(OdV38wFy_X7DD?2 zGwsFT7-SMWSkd8P@L>yIPvvosPh;HlU0=XfScsZ-0M@3u2*O~dp2DPD1E>{yJ*UCK zb)Oh|g#fLB2T?D5)90sfMmB-HNaLnAZ6%*&)#?6(Yp+F^vEH@&iOf4y?U&xoyZ;Rr zLN)&ztqRR59Br%kfyg9GfPvDFO4RLg#twDfeY+Fr6wOJV$8rdi{SlVOA{+UFv2D`| z^#2yq7qpPujQ%4phH+ygwNRrH60-;PtTkq^&%cb%Xe?bD2kE|cN2)u=@m59jMW3!{ zK21z8#t+?2xOWXPs@g)fDCAM+SEeSVC+;>*kwSv(E6$#$CVS|7^t8W)*QV=yK!fs^ zj`jf-9dYpGqT5kN9rzg?INqqu>$bJ;j|Yn+9};J|`r@)(3IWQIFY8RT6$+|c9BLTN zFU+ZBc1Y-X>eFUwO7%=H8J^B4MbMm7A!s7bVXM@@dtd2-6W#ZR&*0CfDLXe#6ZgLz z?cHmkiQ8z+;ZC-GoIOLWiDYVisV}1MPiIJ(8tNy)Ryn6q?BxQ?rsE@3WCVq^i5spc zOHtJmH*G5kpDqdRzoL4wdfFAgFRc|4u>9Gs6nF6ghW)8C{BG`G5!gMT9-=WtE+i=gkYnXZ6I?!1xqgT3xUa`z9-{B(ytLemQ)w=Z9o>V zyaw^#fdFj@_Eo9!mn_Np?s)<6V=C;MQUK~00(H)ng*PXniX9a=`O6n`FQypa5-S&fCd zI*bE2C6tjvt1nnw(z26a;!u=Ux5)XDE^sJaZarzk6YBGI`}utH8B=fSNlxTM45i2A z%=6O^KL=^?9%g@zYpxXs9UmoMZuin+V|E0c?lPugj?xO~1_w`e88hG4Fy%-~n{E&2 zSho4`y1}$_w0xlC%FRK;9CtxHgcri+;JsM#mq4ZFdBHy6n#+Bcb)u0Js3*Wpo~g3y zA=xeOEZsVTz*$yBYvG)tcmk-DgsFx|Ml;W%i9mVQ-uBj=+K{81EDFe;>eoKOm|l@z zHW$g!9?3?h@TXhnxmvi=TVm4A-mb^b*8@ zUX02;Wm#Zi6STP9F43HY2*+??!ek1Hk3u}%dSJJX$ALtT?qRt}wQ)Y40IHsh=_a!b zv!)PUhfMTW{ydwTs!P7+ZSAP(cn2eXF&PFO(~LP~=yRw6bP6)oS701wCwz`_3>AK| z0me;=6F**5Ef?#_Q>1}QOc6%4L4PT}bG0S3sS{tQnR=2`YOLxOzrK+wu!vjrM^Ck* zPDS$`BGLvbeDZB4uXQ^Ze^_!54LVV_L?9mbu5Mj3@;@*cCX^N6Fpoti3A^!DejjTU z8F2rnegi$y6VOaj*s%4gr4DQbCk>ocsFd;yWw8PFsHLCLOBrkUQ@2+(bQ`>^WO1rc zlqQ7;>JkYy@WAM}Eg5IxO&D}(@YIXcRx~K&`i&Ix$U)C@#WqS%wk1#-2M&F-5t)?Y zD$RlRGS6uOPU)U_9pVwh4{ejcktg9pJn!NlP{{5`C&rl_bG}S zntPlM=w3Prs2z#uq)?Rc$MqcZY*VT7Y=^g@O3Qh-?}RWwOm*zjavlfbXn08dzOMQX zC90gbALFH!zy2)l*5$(@?y-pZd_R@w2u-A2{k{DCqlbN8(bKlgt^kfZkZpj3^K5_b zQCiC-dE*TlhcH9v-v!2KScJcmYV_Td!XGxJcTqG`r2T7ezCoIajpydTiLpLp&U$x`% zSDJsrcUA>bg&S7Allr=iw~o3!)`97SicwG zUX)Zrn%eZ+gykGafb``T@LttH5>ILa_k(3u_+QA;Q;Lb z7!cyG6(YcgD_OX}k?07yT->iACO4oSZ zUO%8ur_U&woiPE93J6jy>ZnubFKU0l;~4t9^ZvK}_TbMBb2K;UK73>yA6?LaV}GhT zKFCr#n5zRFocdAhq=>J$;oGKp5mM)4L*q$2R!bkyGAj|9N~o1NR(lX+Fm;VR3h6@# zdQ!5=dr*=%K(EuVn?ZURphDiJ4|rd^;8pHg^We6bv!O8FSM~j(i04j= zf(d+NSMO=)<*S&q(cjvzXZ_gd_V2n4N$B$ zf!$Aq_I}vvz#_JqSPNI3F$U3H4L~7lulB(hL$As$>`$sYlPa#5J>+@wDP2a^&a{&@ z{^&Jg{J!s%xq<$4$IU;+Dc%6nHjp$!vz^lQm0{a`WOrF6w~ds$f}oBex5q_2JV>MI z(KWlQR9G>lbp2;-pdUPo-q0?oqE3ek+g`^3C0ldLLT3Owy(6WXR zHk^lHBqq%|UE#x@!=Mg9QALjSnQKcKsj=Te-DG|9(OfzZXs?QFUPaF;KE5f)Oy}?_ zV%l%1u4r#{ei&cfRd!DT?%}t3zhEo%xpqf}g@zXSEpiQ{#=(j~kkh8COFulQ==47` z3AcZ{`Pm3H;Ij4p(p9eBsyUGs7K+Q|KLY7cG(GJMtnu!;bAYg)0nApvoIJ1EVgAYL zxJ^{jkZis-_B?|PI}R%N0aqZ9zj_eBHvw{qyr5USP~!B8KRhH{h7)SsrDeU*uodLW zkT#Qw<{76=o&m(Gr>Qp5{dXq%C~doDnhs8Bp5VI!XrS=DgmoO0qHgt^@3S+qmQox0 zvzG1!TYKnC8^yimuNhN(WASAqElra_^ofQOBBo48!t&q+T=otp=_Oul%^f~D-aIIf z8~#1)GkR>M{G&h9xF>!fAiqp#PW07%vK5K5q-WEH@;CP)_WRq6YxT^K>$` zsLD;7-ExSH{+ZO(icBq-VOZw6+E{RBVPb8^!^_%>>0rI3`+v~U?=qH*-c%blx~87f zb5}UXc}_g(6>Q!nDsLWr$%P5WN`^CKL*CYCnLceSW(3xLMq^7w;;Ccl_I9DQ^3JXS zy`KN}m6r2_gJ>#>AV^ME0f#?&jbvpKhBbrC{}_h`C%rvTJA}TZKB64?6qP z^r&n7q|z~oIox(3pWuEnELBaAz#(IqvDN6iLVX1i&`0FGY`u@+5ZZf_0kN`U>S|W|r$Gp!!cGS>DnNdwT>Ci|?l2+;i zB_fheLSqCqVBq3t4?{UevhQWw-P&Ed3q>%Oge5Cg#UD}tVSc2HK`p=FG z^I`I>;CzTJZiJkxiSrU}7uM;$nkuuz-7iWlQn@QDRxyWhCu97&L!BDmvALPwpl*wLtZN5S zrCE#oMnGoq0wRn^hg-?Cx!urt`xHvst5&ymItrg=J9mzIXsNYa^*@IT>RUhxCGsLf z;O*7ni;q8R)nr~>KH7d|AJTH#b#1F^SMAy~%7M;gq=4+VmfB7XH_lzIAVN|@vH5a} z1?oP^a?c%2km;)D`T|++BnUDI@*6XD##_f6wE49?;yL{!et)ZF_v9IXM1-U-i!2H_ zf3D1~)Fd+?Gqz7wqElJp-HP&Zn4{-6{?TCEvTW1? zWO&*1oa{ELk0BO_o6E0SBRPh(nx#dWa`-*~NoaJby5Vgfq?8<+<4=cX6)nF+AEM&y z;Jt#?&r}~g5OW0ChjP9<*QO)$Q&Y*lJ3!t3|L5>(^iFR8^Q zJDWrwksIlfwt8HT#8jxr)itbe1B}a;;oVE5GxEQ1OXez{`pJ}qZm8=MUnvrg$y@=+pHKiN%i_2RAnF1VGARK7VM zV#AF;IF}Nsu=UkF9^SU^4TdPiWd1^${?R?_bNC~5#m5|{mV_K<)1TAbH!Xlo-P7Vp z*ka~qHIG{n)a-i^1+q%)?Jk>{y1ybZKAk`#tIz((O+!t8Css$YV{36Y0y8$8z_v2! zA=5((d4~Z&YP;Q}p+z#vfrLZ{Pp3NAws**|lGTBuV31jhJ0{A9BzW}5DzaUl1i1E* zLR%pCR-e<*)4=>pL)w*(x%)qzW-td1C?JJ6LuGiIr)4nl3&~x(J<3v4OGAR1JGGq` z_A^wrdMSH9F-~$TehMmI`zz&prKP)WEnc9VR%TNTUUCUk&7c2DZjBBe0NY>jbO5Iw zol-c0G3p+qSuYdJwHhCgO(upuqX?c#TAmDYL7c(B-HN|qL2dV@BY4TFUO>%=~ z&E9r~bM)s`QnV7{bv?veP@q=Z%vwAH81+n<5z4l@V-_LiV=#C zd1m$g_kU;JS8HMaQK0{`jD((<%RKNOFI)Sg!|D8pVYvKidZ4?RP-<2peKr1~CIcJG z--_l5;`ErqR_7m9-MBmeSz4y}07uOMnJ!Q0MCU*nS7=oGjMgyAHHRr0hTwSt%Z2Sz zEcBC2(WlzYt=^rXhSd*G-U2Bj!D93%&-Oy_qO2uRNTU%zw;c~}6)kx8a0BxPkP$Ds zVHzpF0s;gr95(Ql535)u>}cG#sgr55Mr4V7jOljtv=BXCB9q~5#pF_cwXLTTSwDMlT70{2~amh>TBSag&a}9uKg7l{{%@EW1Zq|+jvA> zO`T{7EN2w0jU%b{i1HeIT2?@u(oTJjw)Hz)dI`jFb|7KRU)@y-`QM`lK<}R}+X?wF zbWwVq`mV)`$fWr6Gx%(E3Ns2DK&AJPgd=cWu2YO!x4HXQ{@iK4Z1H514Kt*Lyj98k z@?8AGkLBw;e4@qlxXF8&s381D$)I}%+3C|mg1*A$VZ!*3e8&yWAm*W)QUnbFq%N}j zhO>*NO;ZkYZ6ZAkskxAJK;cbAF0;X^vZt>`{;1yZM^J0H8eJl%@p#xKg*U#)faKJ| zVEZ(@F_vqgjy+up&=k@X%_Gfz8QXPvnv|k5)w0V@5x_p?x0Y8d>jIryC(BJSLR;?- ztv#B=Sg}BHt(=eS7o0)-Y;(h{Myv_hJ#WVjCIY~qGbbbpB|=%@bShH5di$V4YZP~$ z)R@NJSQPla+yw@BtSGEZz=}PN;X)*Wx`r#fJnDTn`jBeUu~ZJ8Vnf+Nxkx7eT2Fkz z9Q%f5(JuA*GwQf)K5DY#*AjL?v*yR;*=5QkT82epJM+UU0RIVo5~{qCJNWekIYXD~ zR$_xK4KDAx6TU8M5jY&H+%%!@4X>i_MT%;9_;+d9t=h-KQr5qf>*j|gH=(>y){n?3 zo=^mC%1+Y7G9PT;m0yDc5NRwmW8)=0wrd)ngFpofy0xp$)fnkKk$a5#*vswVX(?uB z@>Hd1uXV_|c_R70{Z+3^a1RxQ%k z-!PU-?8uiAqTRRx7G+DgV)1Y@icT^)>iU9;R@5tYtLp$9$qUIuXVyZ8aJ>YN`-xtP zAC1tuZbs3#-r=Wt5UPF}OS^6-O0_*mkkkCW+F1eBKhF#(hT6S1fCp z%OxQ+lFPOJ#p2lq$Ti0ei<35=jijK$_w&63&rs{?YG+D|YmaHvMnbbet}(g3403u> z15D|;RQ_L_6jabTWYZg98_)HfqXVhlX4HA5Ldw+olaX=)ahX;Ud3J1gM~*7z%KiAq z9c`b}?Gj6^Gte0gx!}3E_gh)n0|({Hz0ImZNedOH#|=YQx!d;+=R)aDu4TjVg-v)D z*Uu5^ayGMvV4P7(CURlqfVJz)IrBYxcCA;mU+~@yM$9Q%r8MocRQ^%YFeC zchED(L;>OqeTyPi;qEnQfv~K0KT+ zfGGIl2H#U_IYrZ%iRn$HxS+zKB{_EL0Y38hDS9f}9*S;RRCZXnL!d|l3&*VjeT|P4 zbe?>kRo2*gd1-5ihVA!ZVU2f)X1-=KMCkP7^yl8~SQ4NJhA|E9Zm(raxS3aYBORMR z7^z)wGZCBk7CiG^{whl1hTWiFzKC~oC98qTRle;N>K>B$r9+KVs>9!|l}RV^wTuA- zA}70g$Z1J2tY)pVXiz7BuHkRERfZ^`-}@jjM{=?_;s?AEHa*3?GoHs?N4|H<-qMGa z>qEz;rsm1edAUKgy-#M-@iRW75ciHaVhGXE%&!!^g*Vzfs3-a-SqE`{%)`HnhjgKbt z^$QORKRhU*tBXv;bxZI^Uk1Z=$>p|3ds9lyN(W{-4Itgb5~)tLYAz3a>(#6gbnfqM z|M+$>WGe>@R@iVI&TtqutCKfr6U>|!B!iDy>vjOnTTnUv)Uw$GUG`Klxk;ky&Fss1 z?s?5GoM}-%?B}q25#rck3*o_R`39dQI2}Fwjuj>q_n;`&kZGAG(0e41H1-`NP4D)H zXF$4-9)^Pfsb2Wj;NWlEK!TY!kF&KFYiJzrr=l*07BhcA1mnU7Q@5L6ZXo>WR5)k_ zn!g+ZLPdHtn0{B1h7WVM13GIjh(L(?*JoaGA1cpBYr3my(#Qx?Lv2gTJhvF5ZqurV z({YUX4x}4@Y^sbxPxtZvzgSk_vwZ>5Uj$f3WCR4U4tFJt*T3$s-G`)b^N1GyoWAiZ z83iUJr8!r!SFL)e&l$4N&!>LVt@tgH8522k?vOR`7vr>Ag8723x(2%FjiT(J7N)n3 zaV;EDbFi3N=X<5X@Ql#u*vq%nC-0vDLLZ52N!jAYaPo@ksw+NY3s$y;S#`!0InQX5B2Y^Q|e*^I+1l^7)p{CY$3^xT?}z=x^<^ z>gH4^`#NoB@Y+MslA0%@b(CvN*t)e+NurgHB?$!hwd-7ZcZGOdNAa4KrJeuiW`&zN zz!=aN=0ShHtBvP6Q}=aR!Y&t z_{S1+-I5QsV2vPRaHlS^>l3qTbC}>5Z@D*#a^gwS)d8N30^%fyL>(Zdq&u9yz6ihF z;FaW)>T6Ij-PN#QvGr%4Ve=0?8#SbrO`O0pJ!TS6niLYvA5Rt&63ahbzw3Qu0ZDk) zxZl`#hl7`+=ArGq&RS3)#7C;i=jBA{uO!y_)#k9}=!D?>F zg;Dr(*S8ysNIX114GvlXbkk0{4pN}H6R6C)+gSXlOq&56FEeWOMG5z}u)u+&AnuEp z>jI)HjR7cj^#>-}IaPnoIevYO+Yq7I5`O5A?lr*e@w-3xi8Y6$R#2oAZmF<029^4} zR#aIu*GlT9`J1L zS#5wvph{8|0QXMyN`XjxnlBT$1zqzl`tTezo?j~*21vo8A6KKHks+T=9_4Y>QD`Py z>1u_tl|jm*eg3om9p)FZc1 zo7XVZHXrT4{m8v8# z#4Z)!ImL#@_o#kbyJKKsi3)^2XmTY;wMj9pi0oP_mK zp?iXG+}c2Ac}3lQ3nYwZPR!^3a~Nyt+HUo^F1v02 z0^{RZ4IeE(TAV7uQ`;eJrn1E63>6jKi6?amVG31?3SqEuk1s}V#p>i&wqLRwZP!q& z(W$FnyO%i2X66Z%g!YCFZl^eEw=ybT&dZ5hjU`MxUyJo|Igq?X?uJy?`b?bT(=-4S z9j4^ItM^i&-hSiVM(IM_*%*%S#zpOU#Thr5||k@4qp znIY%qcYD;38NGUx<&Id~kjJr(cewR6p;STw4I;nM95%67_Wf_t61ze(1%wwP}7wp%?7`U<3z&gCl z$SR!UqSn=$)yK+E^CWCzSy||*akF0)#d8$G+qH*8n_zO#_`~PB^ueb~yt40xqV=d* z{6X)eQFrTo z-E-$2IOpN^%6y=H^=sm&pr97(5kpj|%tF0!nfl>qQ?1O1Rp8g4$J~!sKh>drTG^S` zcIT}e8cdfa(AVzv=m64TE$V}9Z-4n5omE}{w%L-r9W{H#rB)Mq70^*i$R1lC3Y>5V za5Xp4P(7BH>zfXl0LqZ~P;Ubh+5Q0?DN9bLgoF{Qc%EYKIRAnyAG7P)o&XTl(c<;9 z4F=9Lf9@-pkhlG(_8{FdhdvXQb!hiNILUGM2$D>p&rlYEbq{WP7koT1tuF?4W2!TP{|xcyi>9Yc%K;$-`V(K6$$mSU2#J= z%>0bN`@ZvYwWXNu6z=n(PbiU=m%b}@N!7Il=oT`fJQOh@lsm_K{z}!xeOn{}35jUh z^H&)}A?+OV<7Hr*z(Bb^BoLf$yT^h;M0d1{pfjvV_C3xV0P+Q>4x7~#`YpXk3$V&sMH*vH+t@kuA}~! zX6(<(S!&zb-vK5K%58^qIKJiO$+FgtA6D~B99xGVpfau$sD`g<1yi-EbGP{bO2^Xg zw8fYpW9!!tMAf4TDw^FlSB$6#esPOui=Fd3<+Qj)O9qdX5j)J z!@Qls-(jCZ8>%Lr*DYL2I`~;!Fz6%;V!{Do4$)f{kF9;z z{+yiBLVUnc83&%N{ZzuiJ8!CI606p+cWYZWQ(g6qNeeVStL z-%!qXnzVIrw1p7v(30*Pm$`5HRJdWyvSfAYt~=RV|H34`Wv;`qXXTkdH@=InE@QH0 zLwnp@>~NjM`p z_d%C#qE)b8`6}KvAwus%e)@hluG_USR*kr=P;|;F=l^zYde`<|%DfNOec4Y&0_Hzr zA)4;d^DyOYDe&uCb4vHx)H6_5wY!Db8rmpglF-YMvF34o+sKpOg9gcJewl^t(6#v~ zPX}w?W^Oynjd?nc%o^cxB^^&U_4L{e#51^eiDwm`|FqQBbHfj%V9I03xU?jC-InH2jEPCt9& z1bBRZG!wxZm2HwukyZ1RAtI#~c$^}KebQ$C%+n3GpX{y34ZAVCxNyWf0qpM<>%mrY z+2I{YM&yKB?%#k@QvNc85SKSe1aOifaV1>+NBJG#9W<*!fo`^#BX3 zM^W#y)1|lScNe>JP=2kqmrH>2GxGc5Lelj8F`+FzK$mWOswZc6jY=i=Wvt)rs; zzNlfkQ$V@}20>c7L`tPmx244K89|hgZk6s%=@_~tC6w+O;Jx$xJ=(CHprQ-SZH>eWs|3e?LFxgUF5LcAZgMH}m4H z*zF3o=?i@B!!n`0>ay1k@Y0%stQxz&QU}x*g`bdiPRh^pFaDEgBb7_FZ8ZZ(ypCpD zeFY{seUz3qzkD^~HIB@KhYmaBn2!8FkTf#M$~Qkr)=YbdEISNQqLxT!@*908j25O! z@*02UKglalzz=q`&87gW7%X&&`L~M5l9?+*6t4gHoeE5@)MGn2zHMQ`9*@Cle^elR*@%VNy)bd4Y)+!r208e=6Juo>KbP*q zXS--6=N!rsc^Zp$W`~v->KclEJl<>ddwCFae<&W85c>r-)G}k5S#7_ZC$@cYJSgyL zSw(Q<(w;5X**fL4VJ-Rv7mZqib;`!@X}@~ykAuX5l!Bj`Y-AoxvbyjAX#iuEs`p<` z)N&U3Klw=Gq3(O!&pBaN7|?h-X-FLeME5UW@yvZO{+>p~+l9}z4)?Y!taxv^M5|x^ zLQRXLNg)lCI|yC?a#Brd-C6EaO+1lhH>qCFomns59UcM{T2qJQK5vLGnGPBqj-wY; zvl0ZtO5W|qO6~fi-dmrEeTeCW1DVO*TTZ>(5$27slejZinLS;Vk4BczTe3;$yklGa zArLarO<4$P`-~4ht|mV28kbVrFoxD7Zf4w+crAJdIuaTSi!LlmG_$Td+nRCgAyHZ3 zMVGE>WQ&5MB~v;e>-HIDXKDgX3N~23-QXz&t`)XtYQOfyN1TNN1)H!$9w@gPFd01b zjZ$I^5MorN=`Das_W8DgTEubS?7q=FLKI6kYXpUcd$r6h^R_LXEeUnR+ShMd0;IWs zE9Wpr64_~T%?6X5z`Zzpj zzkcAnn-uNFEk+9S%hgxdYH>j(ci9)}q~&KXBGZf13$Ptz#MC@sQjO9c%{#`<(HeZA}C4d+9!p(>SgiSo0 z!$(3`Zd$YdZD>eH8#=KhANF0d?sqb(Lu5O-r<{A4UQlEk>jOTVcu%J2{b@sIRnSqv zNqSnhj(XjGDOod2=#FBPnE*6Z-lhMo(5#; zKnpK+Iqhx>Q5TyAMJV4mzr2DMgXegaY_rq46nDK#G++>Twc&;`hBdsjw4jR?M8{W& z0nkU!k!4V9H&s3-*g6W4S0`Gt2o}?Ocpj;Es$?t%NmP0)-m4KbJ?0$>Gmv_1qCPoOyc5= zyJ`831?HLZsm%Fo;^w!q0DivRb_UCviEPK(*=ot-_9dQ>wvgEdGOo+_#aEG*E4+&4 zm@mKkC$(Y5at~!0{RBPD_WOFW(sR;+(#caJid?B;1&$ELD+jJ6lZO8)>esnHLATQ( zVW@@t?*E>nZWsmfq9)_~IwiE_DA!CQ{5>}8A=Kk`yh43{0Gg4%s(4s`8hiW=p6GhJ zFTK}(k}<&GG&>R)&XMyKAy|tUu2x=UM4SZ z_ILDt?8ByisO-4^BIOi&ce44d=KF1zp8WQ12kPx?UEE9e$^2n?s~7D_1m8o zX3}}in*ZgHxq5Mpu5N*C1ONfCu)zI>tr6IDy2P^E`_DSoi$>Tw<{<8H~gQS`2Qd1Npl4p~N zcbC1Tv9c+C=-p`ec|d37OsqvosWb8A>4XX=wAJS&u+qPgGXu>e=kG%*U@ObvVIRZ~ zl|DT=<%?`@?=bGCcd6xECPf0Po9@%((VsAn^8Fv@`-o^}#EX8H| z@hVuUC!N)nZkV%E8u*gsgLJIz_E-Hgg8jP?^mgjjxOa_xsE1PRn#@Q|);Qp1Q_g|G z@X&gOS<*bRqGzMMLQ!PV<+*3D(6c}~eJiEmms}-@$cioQrvEYHXk-HWEpA>)Qj&FY}aWT@QDyYx!trBm|KdO*8I(n%hV}wO8EyBFKV|Ik~5Kf#%s}3A)o_ zG<+6YlddZy>N_m~mzf+E^UL_0yp&wy3FfzXj^+3mXz1uq zEs|%@#n2dMjBV2I&V7^w_Bmuy#z8TojGqlEqwbjJ(PwtM1a@d@W0Qtgp&}-yR#yG< z?0Sczhcm==l@Yv~Yoj8AFn845*O= z??SOK@vObxw##c`w#X}?6F}OotXHz%2N!0zvP^A{@m$CLrMcKUq2oTrgtHEINZ&W# zK+Bql>QA*+{{8&zeE07s&pa&NB_Gq|q}kI)^3*_k~Dnm|4@ zFo}d#$ffN`nW?h{j-rlh+}G3roBfXx9%xRDxT1SP9L4y|;eAqX4K8z!`nJUj{IfM* zge80z)sw#k9mYF>3sWSxW{B2%K31(UX7edKYdT+YxkZYs;dql@-pL>i(XEG5-%2UA zjB1|&+^!d7I3c}^L5qRsC)UCpmRBvdLjo#^k#C)nW%koyq{`B3*hw#Wi{GpNAibt11bpK}#%kjJy@5{P&N7qwwSQV_I!^l51=&cc7 zPM>jxA7uN<<1o8ME^Tq?He%}CrPJ1<47_%<{=w@{7Z;SqRaN?UYgVP{#rs#*2CLHI zt^5m^RknN|4Hs&m&_+|m6MD{pv04jNov-eqd({(h1_L3~M#*xCy~izA4gCE4I*>m* zKy9jXYH=Ayk~2w0)fdaA7B$ZQ9#XCs6T*|GKP5`=$2j z&9%eIMv{ITIh6=evgBM1V-SXkPxl9{HpH5JZGT=LBRP~B)oWUx=Ls9;8@xBM(oH97 zE1meL)_1&dA0vO)@L1PP0C;$NrFOPMP4&!268!p9zv4Vx48 z{&@3qUbzLB9$*2J5vhkfM&5eL*{jN9kytiLf)DsvV4PwwtYugYQ%+9KP`jO?8nHiN zW|ggwOp@IC<{La&-T^vFwKxG%GuP_&*P$}-`wZ6f^+-3ey66}NumZ;UJZ zE^aJI=-BPST>YnZhmaiJ33~w#4oYHgX_&51YT4U#L54vZ`rZ@rO*HJ7_L&mH83v$4 z3!2+`DiOPD`RMyaRSM15nKV~BuqsJ%gsSX5JNQK4*T<4es;yyacDe+rW;=I=;|fQ^ zYN_d0x5;~zWu=2aYg(u^!AJ=;X5SD^F{<~Y>IS`}bYaQ~cZQOEK8Fv$ZdXZq@M1xAw4#7>5=S3RebyuQ4un096z( zwXS1~MVZ6oG8c1p4nY%8 z>!6i|?>v!a(5w8VkXcETB|>5AWS(fcHNxY;lVo~zr*t$yDu)yQYM}aX>tp*5-Lec- z+KC(m{BH~eb`u3U@^*8T`6bIrq;Z&woCSwZPakhM&stxU|p;C8!ZOPOhWCFQKhM72VDoN36^J@zjKP3ZNgz1hdP))Nu8 zMY##>Y|{Uk)AA zq*Qqkli%s_Caf`)t*+-2Qy^y!LRjwfU_yQaoBgGRD+ZX!xg*6(&2qg8mc`lMq_d4r z+vBL87>ZYv`FT-(lt0RSlyx#Np08%sPw4JyZa+TtQ;Iem!ju+N!9tpg(lYqfsL{f} zNc`>IY{w0@UU|E*-{9AV#$XkEU&+UBTo3yF)N(ROw%cq{|N0>#W|U zYA(diQhYxnjq@rL*&XPxh96oPH9<2)Y7v1f2ld_Bx2qpMmGoM9&@=UV$QxwD6#OGJX8yZcX z59g~ewrj>7=`VOfomp(u!dw#O@$+9DgR&Tyv*2#2uj-JT^8aY)}<& zb6%uVlB(LTpLpoDGgAiptR{{-n8aJR#%LZ5Z}GdBSJo7c{<`!!6FFRxd}>8Q(AIN_ z(oCNtlJm!(7d+u4ED~S1& zBJX6D$S~oi#l*-T$_87M{J7w=H=Id~I1GD*K2T(kZ%8Q@we}2*&H(Ls@XO&~>lsv4 zrIam-;9pX*r||YlrC|C^{&&12Zie@KUO2f@kr!u7cSOf$feCAMBbaX8K zJtCA!j&{|DBY_8v?cw@|BcXUS_2bQ;B8x!dr|;7GJIG`ATx>@{;I!~zILeVzkY9O21z6EpEwKq*(afr>dB$O!vOaKryu)@>cuY80vYZd@e=6vy{j zLHBn$OnEqZliv-gHI>a_ngPXJKHH6V?DDyKFwCWt&=OBSVK{KfXORTgSZ z%@@+78%gwd@lPFmGyXh?{;Kj$^_gPS8k4{S+cC85IkN3}3fq|x<1q3}A7TdR-VE2?B8bcnlw{^$faxl zcN6r-@W2$}Bz9Bi)-z7C(`vexkL z8{)Rv-9gFxDcCb#YgAxHShE=Qi0&ZSwE}8*LBkwPW^J`n<251{o0lKf@~!U8_cgKh z=N2sVd-pz!c;J0SkVyUt3+g>4JENrP*rKAFc%zE06Kp}jM$2)0)^Pxr2{8@w zh!0c|RL-3H`Xodx-ROEtGWnQ|sJHoNY0mAzoC|?ooy#O7>jfD&S6-qNpl&=cNaERV zJt?p)WV|gkZqIws4wFH9m4w#cukY!kJDE}g-kY0gRh^ue{Gv8+gY>QzebpMvAeU96 z19%b(F39lTvZ~0PZ`!ZrKf}=qUNe>7Tmu*5z{{UWbA0!B6VPS{#6Q- zgM-(fF!v-OAz`R<#k$0Y$|XKh=^rVY*nKM4FjVt4rb#4;7yBVMPW!-@FXd1} z%wOI86~;Ksc+uf9n!l1e#5YjN_n(+;OvwJYl9-LXEIymrcyr>N;KKlF`iP6Y(e&<1 z#Wu72jB&M>^)SfqJa=b909H3#Fm1Khw%eIW8SPVyS>HWmFrBQAzwmGB=id5!jRVwJ zIy!KvIvtysp*oc1+r=e#>~mWZUi;MtDN>=MTzmpA4}ocMFNjEwevzqXas9pp%N$Hf zL=8gR-Q{ndo@vkWLwB%W$>|CR^JYcfvGDl`lQ16qi+k#@X@8mSikEJmz8=-~3^k&2 z1DzOUD}75#Pp}deALFQ!+s78;Q2H+ZK!uQINPtW3lFDe)B2_&)q`9=t5w|^o?M=Ch z`J?IL{>n3yFL&BF%*QY=7w#_PGWBSBu5B^ruY_|w8xnWl%>Rbq8fY)P#uaD zYIoeadt#crK1^^v`m!~|FY`f8bPB)M{vtIqGqdmm7^e(1bSCttG6U6zq_i|gf)C!3 zpqO72$O@p37#4s{ff-|2{TiM>C{cx&wl-7TLWl%H4`~F+WI#_DB1RdDUpTRJVh@~~ zkFd^n=fc|CW$8p++FQNOi#qVEbzbKx62AWNx_|oSB*CnhmH~bMs&Bvl?+gz~$*vRu zyXh^l;(3V);qC&qCA_KVU_uQY9O+YZCtl34>$I%DGkx(uO5?rrWaR_L;>pF5+rob1 zAa1(c zcL>lXNz$5AQoOyRX8>cz`;4EMQ`1NBJh)@VezQ2l~#;D-?3MsAd*$QatIfwl=O{BQe1* z8Y8y7&m+63)Gomu?Qkp`o1xe3r-RR2p~M#X0Ra%7c_#)K2*-WYIV%7Ze*qlWn)Uh) zaiWd@6hpF*T>=#GZK3BQ?R&s4Pco>79D;w>eSh8n_v2ouwwpNYB9;L~=M3dJj50f* z(QpUg6=Q4NUr7=HgBJ>E#NqQaU?x`NS!{KW+&=$JLFRL2vl~yd9i3pdQz;4o5(Xmo z{atyPf^MrPGSMOV6nLrR>Qr-m;`VZvWW}ysY{jLMHr2!f18HBPL$(#P37`Ehf>!xR$!Kl+c*b^&l@=Hg4 zqP#wp#9BnMu+agp1Qx*%>hPgU8}xNHFACSlZ#1s_|f-$u- zC^Y$mE5Tdv4G0&K`Macge%0@R554;kF7v+!>>tX_#LA#NhyI8W^j5WWt`;z(|1XW8}z@mb%}v0tAU~6EUIBs zPIhx%<`N&Ea+}iJD`|>R?1w+gPozAH9sUNL#*|3jLx6W?(gb|`-;fvtz&dP*4hMPL z7aN?$+T<{^?6gC83+kr|#UBf5hrx)u={#bfjub*%YV&zc1E zIIUOuQF4!Xc!-)j_vGB&h0gaD-p9rgz!|`bFskoU&vx#@5sdE{G4cBL~KN8HqjD*V^siQT&r~=m$7-7Gr z1Ig$WMb=AQTAZlxC+%x~As-phs5xadAfu-SS+{yj#RYyJNg`BqbRJq*Sd@G284RTe zVk059*2^_~e;|pwyj+JJ0h)OglrHsvU0?40SBndaonKN3T-3Ncni!arBssh7cfs+2v6pXUMyFIh&+(3;nr*3>O>>*w`2)~!ONa?v2h_MhYo>`h zN9s8R2P;yP{*yjOQh2^aRW22NuJv(Xv&VvRyI3 zQ9T9gQv-z%e*r$1f6&*r9sm^YfVdRkxbR+}(r{T0xB$vDhE8xaeE_}%R9%sT zRSO_KfN~pDE3dbbuq<00NTZ<-ljz}Z&yrFJU{FcZ*)E&ZAi_yx6$I(`Cvin@9A|i6 z|J=NWz4;l8OBS3W_~t?#N3#~3J9ibznHd>ziHYRjD&J%I+1ZA}jG9GVGY$HM(qqOK zaNy1c2J@?vree4u;6b=PlV#FqkRlnIloC;go|>9LH1hE^vNL{j%bN>2FG)&*PTr#6 zK1RBsQkt8a>#Xa5};jiDUOARxv2g3L#)$Qab;GFSUeZKTju7o>k|=(-=e^eX-#( z=9GWJhKi`DsG`EdF${mQN=q-{1sgSATqyE?eo`1<&onj~$wZxn@)@!?_j}W-b^3uvn8}>KQt4uP)wrR(l-%En>U4FA0V(V{4DGIAq@75KI zOXK3shc3P+-8tcessQIxwlkT=SsF7KWguB6+pr-1s~> zt*~Qx4N3&bn%=0F%Y~EKk%ew<3`}LhX6Qp#ctit4CI#GhtmbVb)PK{^#l5f^J=|c$ zsQ>l^XW^r~bL&63!4eA%=5~&XgnaP_RdrPS78pxG_W?*cC8Oj%l2rUX5q)L%b%UNd z=_mYi$GcdOW979l=MJa$TM-!8*gMUaC`*137t)5$MsSy7Zj}LGp!r>kO(DDrIzYU$t81*q&+9f$8pXkQI8AJV$Fa_Kth4G(;voSwyRW;-; zX5X)RTksuku#{&B!xhm8c~RCxP@3+9!oG_Y7!YTTvB{DN_TrzHG~4){?{*aHSB;h$ zO97I9@wHEPoQMS8UD-G;HP7J%P&|Hp=r~O8&SD`YYl05$ae154 zR$Tlb+UVTsh60t8(9uy1ct)CjtA9hw*XxUe%&ilqmke?0KXppvmRkcfQH7riMNjIt;be3t*x^$`TGo=9FICBXJzsb#M*;gC1EA#7f@ zE%0{NH=~cGs5an2)B_+0e>f^Z^7{qo}iKY`Akm{n9egva;!bHEGhPf=N_@ zR_l-Ih#>w{Pm5EAJ1O;h>L#~Ow}g^}ieNVdy*LPa_KNG1RUBE)Cn_OZ5@Zin{|~BZ z=vzvupJQh^n^KmrhLiZPJ^HiE8Iqd| zat`#?yZNcRg*tfEa~$TE{Lz!OB99>JTYmW^+l;*h;4?vdbmVM-s_|GZ}NK>AUx(;nPe_vE<7X)m-@`(gCKd2 z*Wh&;zcLRr8(Hp#swMep^p2Uz_Y9e&P;m%QQ)4maQ=!>4L(J6eODXN03&H3ysvJ(& znA6t|H#no)T3*e>Sy3|gahHSmDq9U6J34;%?-)osDMaE~!Phl}tnu0;JHhmWQ%2nO z49t3oD6eMfD_%{_RC8)n75x1D=D}!&=C914T<{)w>Kk!%C?mQ(A(nu^aK*dwn!ACN89Yr4zE@r2`L9YoJsTUqr_^n`yLLcI`jA%` zzu&oIlOD(w#LR~XX??0(vSS+O8mICX*Vp`D45h-lP{GCiSdh~_bmuk`;!nr7pK|4t znZtgo&ad(wz^4$XtvFj8UbjGyQ+OwOH(SC4KR6VXJ)cDdA=l>7(*qyKke>p$bER?P zP<~|MOv+#vc?pkRx-$J}u_?r6(7xx+T2lI)4^g z1Dbw+{q!jq@QV*~m0&%@GKgSsv#s6V9o+Ll=Y@7wz$vfPsod5b6LNJD^2aL6{!NyL zgW07^7SvE-3EahuJTm+Vmjr@u_ooI`zU_!*J1+>Y%$SDse>VQ7u4)W|J;D;AC@q*g z&TR2V<=udrNwpkBip_d(x~_jV$?~Q1-c8Y(Z4=fzRqw(X1q**HCljF zr^vLeRduWX^`y#1$<#C=`l zV7=2@RHh+4Fm-{?4|{pS8U5+kFZo|RJvq<9u%cpPtIuR0h&Ml8<35ih=*W5&<|y@U zI@|JET>-oij^KSkzE1{~oeR1?A+!cd8&DaFe$K5bFQJ zMY*Wl2c)DSNSw{K6`g4QTpIw)gfY8Wair*t4h zlg8LVO#Q^Jwp$+$w1&wowJe<3EY$f4gD^9bc~u?!L5}8^rupk-Kuy3_R?LPnq-Y&x zO1lqNyT-n2b2Q1YvZ6P69VoW=o{fTQGLefzP&=v;tGneRt@$;MSqIumJ8G69A?33! z@xWtj61htj^83fSktyql-|GBTSO2Ekhrn4D(OwIEiT*hJP{u0$f9nJ}MwK%$t3DEc zK+GXi)sG0jtBs_wYCG+w%&{PParbs)qydVHFhW8?JV8k6*gl@2vUN3~-acQ$bIdC+ zo)4!L!m*tu_#_id(#dG1Iv$1{;<01&=Qx3_{qjf+XUL=spn(4LsxALZT}k8+=!$3j z1I|Fwh~&jc>FJ>p-T-qNi}nW?O9rcXgmeM=+kf!uvmHv&DH}OCG(dqy0cw;~*EHa% zn#`?x8gK*f2BB(&RFYU75xsagE*bj@&=wi@3ppIRjihjQ5Mk;uUB-yjxaAj$+!8^bv!XAAQ za{D!99{`P%!_97mJL~MKYlu#_EHJ?UHT{3X2^$(x0b9zVJDNi0(ho$JhX4dITdaSl zS8j${3th0*_J4pE5Q|4ZOkDR@7XPJ%1&c6r-b&^@3>n)1OVdc7lTH0+ydrsdd1l~* z07orjxzCApcx2@KL61(WzYlGGvT@eXz<`{;fAj5yrrBb-Il7j%c0~}d&drdd25q5& z98+j65yE98dK!>U%_2*yB~Szv3?$*drwQ}Sk`{z=$j>0hM@L9(qm#K!a3M43&Q4(n zV26Hjo*sBML&+h)z>qG6lDEfW#`rz(CdL4j^$&p2btLGAhXQPT?~5V*2Oxib&v}r% z_8EVL9K~T0!<+ZC^#@0w=oCDPCanBfF>?hz8%D4qqhd6 z%2-=akP*llgn!%KobQz!Ch3~TqtP2~416iE?JJV6efj0_+_rpcA!X|YkW|3_{IR-m zw)WMW=CiUN1=6sYyB7Z|S1y{(i{)ggP7R!(_jRy_H)AE=$Zpgm$7<}SB(ZTc<%-JA zg8cu|Rc2cLTQg}%3qwz5(7S=z$=D2AM54k&*7zE+Z)wxp#>4*K+4^Nj2*fMOF&B;` zIo?@g>J_m$)DAaKotNl1(%m1706^W45ms}dhh3&yW};*C4TRVk0q6QgLAPhZTA%*% zTUuC19()x6jw~UiUDyE7SZYBmnB*~zNx<2gq12~?c>Z%QUcIXN zu0YK;TV^Wrda%Gm$Z2tQTG;BJ`ZU9kvlZ*Tu}KAg)3&!{HOhLXH{8%pEXZiry5#3> z%;M1Q;C3sIu~D^YcXS4Qpvw8l*QA@*mki=TcM%uZiPb}tO=s4oU+B%PPoU17Yir>H$tS{Vsi1lq^3RQ#R3E90WP4&`dSAd%K6RHOg) zSospmNYDH++qBNAg{G6X@P`~^isji6hi?cK{Y)W^I#F=c?9x#MAbM%3YzUVNso&;M zitSt_n?beh>fGa(52o)78eH?~)d5YtKW7l#r;Md#Ic&+IA&rNFCo-+@$)@n2(gx1J zRmx{k{NgivMo9#54Q&}#7pCpB*%7AReca5W+*K-S-P8|paXpI-ZqQq)>2^ygsptsc zsGIQ-$8lKB6f@ky5S}3g#aIEN?ocw& z44b%ip?|5xZ!G^6r35jutZuJnCq4Jr19Z}(V^>WLSNuq``cY&_R!=YacM=&L7dS)B zmp^+`YKmeb^`^yH*FbeUcabHQ@%hEbI+FZd4kLvvSw;l4pY-}!f@}EPz#@@z$NHMq z=*Due4`OqD#^h$ozv1-Ie(I&jeF1@+s@AAy@8N;J{hBo9K@}3=zo~&P)J|JZwQQNX zI4R$J4X{yGjvC2iT%l)85zCT`z?+^vek$6l@^bBr!faD&W2iput>dW*kW*}CH4n{F zKDaaf$|I2>Bd##UzJA-&K)$8Xc!7v{>M%Wfdb=y}7pL&V%hCI{o0`nq0dfLHX*!^T zVFx4KnA@RM2Kj*U<>$k#80E_osYi$__6K+V#JEANO5Z8(MBk_{5k68tr?!*z4#<)- zhN~^-i-yZg>+9Dd`GfBKCE%8(`cx;;2%Mm_ z1(0OV<0X*q=)s-u?`~_S85Kl$Y|A4}XG)FtmUmwzn^Y^o7(Ggd;w;z&6wYN?|J3Y09$-|3#Kcd+h3{$-1iEnaRtt3cvVtxb^M-Ob|o9$t5fG z3+uyJkQ>rUj2HyE8Fc`hKn2OQq~!i_3sAMtb5SWODz2Tt zWn>Btdr#mOmerI;t;9iO1{*aR z(B{{Q@4M5*xb4vt&t@xfn;Rm-G^(uGug_H#tE}xdm;Z>>lXskeGlJc3S;S#lg)}Iz zZ@R<~%(2alnxf@4;!1i~P)xDZxGCFe)D#8f{`7_Xv)9vcGe${=3ZlzaxqVw15+Se! z&jag6W+?VEBLXLQ)96D2w^84kzG8Z3j`6Uj>Duje-kVJmjaEfznSsHkIJ*_;B)%J3xt-pLc=;mOWfXCJz!W z!;{9>dGO5{#ojKXCL*Zgnv2qufPis72jx{Lc#cO5Hvn_ol?7rk1^N3Lu)=mE(s~PXq|8!2ny&k-2j7uof>p!2}0_J4HFP zawqz7Z~pRkxiJ>2CBq9aYX96EN$jt3$+gKl&n3N5gKijOTUW$s(?KspINoxIrGWP4 zJJm0o4Y#=B*(uBm;ssMhtJi%xB`E6O*91^!dKyN0d&!2a7Io1q!8$)xiHn>fvP9t%wKKrT*?? zFb9H(!Y0cem=|eC&dR8@FF?pjILc?P;^pRmv@%xAZF4q?U0h)N?;(Ir%RNk@qdWDK zbDk4oN-Bnr_D8y8$1m80RY+%DEWfxE=~k5yeBugM=*ZgZ+{RTz0QFo7 z+V&SX0UIzRLk0aSs2yfT)l$Vh?6dKMmZE6!oEko~iYFCjP{c;k8b`=M~WeZt+`G z|Hl_lvz{@zxbTw*GDa=0{6oC`Vtg<+tG6ZA;#p$T^)73j4F=pKo%uEXFp4XT<0kyY zvm@hw?6lPEQv@hdRRz@!*mqG*3@r_=?NC%xbq-Z$tcL{$TfzmDZxb10mTABD(geB} zAwZF!<0AY zzXqEF*cB<9FAu(!2F#aTx6!_E4adaegZT~n!t-rZ_kS2dw(U>J;Q(*f48Qz#D4Y4H z;Bt4C)LQz_I18hYwoi0ml@0o z$UZ0wG%lytdGS6%qL-t#r!=YI3&-M+G0Vix43QQNDWw#X$et}^f(c@NgY|@-AOBlU zmU;Ww7P{!z57CckLryWODjXl{4F$p-uVW?qxxNXHt&Ti#k3x_=cwnK>vbRuYnEeq( zKB_keH$0g9wpqY8{(F>Rp~fM}BQXz0$sD!-j{NsPNlOptn6MJou0E+~U!<^h$^S$Z z^mt*+@LFkDwlakW$lRWi^?+c^H1Mi>n$4)v>cj0ms2crez6!Ho7!(=h$^FdUE$#eo z9AqlLqBE+e^y;(#eOXdo$`}A^kR6_X4P8KM@kKrR*AM{(M}=6|ENOI5EMjM#!a8u3 zZfslGynwsb+TMt6=93gJg#FS7VN}5MmQYTxh2M=fNg<5%LmG#6-lUO!)cE@XC!cVW z!dFQ=-9X!I-z^vI3G$1!CnT9kiNZkHHjQ6XlJY_%YsX0VhfYZ}V9V!1Q-z%HLq+U? zHe7-E-&)p&N(M6^3-f+D4&{p^;YZ(SaHxoSwlwcHxDW~h52f>7*5>iGy^4c z`9N0iHQFyezq%u;HIk6lfOKy{U8WnEvC&gEe4Kl^jq&0p@|*;k5*lY3yPZq+2bj8A zYk!bF?ql`5`P2UyI()`eT8{>ozTy7*#L$}c45B#hnNzaq(U+(3!u$|KqTNU`38`cS zRY5f)N!9Cm!EHcjb1vFh%!{bwBg2>Z9Lh>BpdT7?hR^7?%a82HaIVMvlB?Kgys6Gl z^6#UhrO#*}`Hos7W!TTj8-CqYfbup^h_xu_j#{r_TuhIgLuDDU;M~5EguazYyci(D zXH5X;tM>e)&LLPDPTv6}9Ad0|Ifx#Mm8Y!4t4`Avd!5VwV*LoT&FBJt9+ui2f3RBP zu%1Y%Q2$M$pH4C>k)8fzcH0)JG2Nos5<_GldGW+s+TF2!4@DW87`f;ML>57 z+TU3>gFbon;D`ABd{M;pk1}HU{+1Cb74QcuS_k9LJMN=)SA23pSa@6tIUXm$u^TYc zXEp5llwx2%UHn(1h9nQE5` z8gmlDF7VPxH%HPPCQROCt8C?39W8Ou@p_wp$^>PXB|&i!Htd5h$;ng;K~JIclhcT) zD~P^#8DBc1gU36`Y)c*UJZf%!s=VW41hi@x1x%5EP&1Me=^r zP*-5E1CEj9YC~?xV+6iptQ$jT>xuVm@gaS^nvlo8*AiuB@`0Lhz1_IljsofI5yD`H zlAFCd0{v5Hol$~rH-LC=z7E@|nUqQ_;n|~%k3PX}-6?`L^U~E`2TO0INxB(V#j$~u zFw$Hjp^tu|P`wvu2#P0G&V(cbaBzc^T)0D>+!$>@=3esQjM0dOJP? zr+iPh*^|%_4Oi0Scl9{!imSXgTN;-DhPf!7;f~oj=@fgIHhFlWzY{6~Y5V3C{Ko=R zKwC|8ruQ;!rd}uGM0EuZ8KrZf>WSl!qSYgpvnix@ zRAFPW{VraULA$VVYm;lfC~+-T{^*V9{nx{-WsnxCzx;WqB+9!%_m#cHEA!PExv@iq}=m3WgUg#1x-YG5)z;q_#mJmTY&g*}oby zJ6)-K%Xh)w{*q1TvxNJKf$~1fk5>}AZ~$wFJh(%A9_L1ojy!L8#aK%2SIK&%~wG~m-D+>$|AK>qfWNh;z!0kuW$Ssb?ddVFDm1s*FZvZKtzyZheh zO316iZ=W%EFpjaTo+wPSld=(mBaB@Q$5x&{y(IQ~Lc~^he21X*P`QWCJfH3-$h-gZ zgu10^WKt!QxFtCNU|JO?1~?R105uIx*8ht+Oe;358{sJi-)+L$=y1W!;_iv^hi9@w z{#Nuxp~VE26}pZ@lQglxqd&TH4JaN+@&aK`C4OOXQqN>PrWh~64rk&~z!lr!_$xj) z$-3LVx-%z1#t-iq(A1>oJ_L22og6KzC3l3yFrZ@-aZ`Mk#)WgW`jMKwmhdY0J)>v*h**VR z-K+GWTFdu-A5+HlD$1XQS>i{&#f;Mh(+wW<-3vR)yUH$~-EiEI?6AR+bP4nl1Ju>k z40Jr4^bK}(PdD@F@Zl#O|4`u4_~oG#z3-cuR^j*YhesEnS5+qdmH#ZHD_8>Dz_h@28!A z|6+8KLCGtw#J1eL74oOrw8V<0KhVUzEYbfYBZif6>Sd`9ydxeKICuGbf-L0n3`aRe zq2X8)qFcc^fT`g1qFC0K-O2gb=s;h*6e+=JC2slz^yw_P@w7mvE9hRh`DIF};NjQdxORK||dDbzY;` zm9^7`Ewo%Ebgw1cv><6dM=F==zaYYFR8CrO##R%=!Cy)TJg@tXhIvj0Jn zbmCtrzf+gL8~>(G=pF$DyPg?|Xq8rZ?c+j~-KM{P_${$MC2xxrq43cMp8ezb9bjU% zs~l(o5?ob~+$yo2W&JhQM~7J~{Q5vv@~PK=U&!fi#Q>k1;zmUQv3ALSqBPL*qZp;H zXJ+*BtsX~&WI@V+$txsc;tga|&;eDv_a^fxO99)X*?=`zTLu}3Gvk#BIoeLjoCx8L zwb5?~ib<|vPu(G#yVNy$#gdA$N*K)du<$1yMxu{QOHjM49Q&`V zK2U5Mgdgm+JpB5)J(*t(CvjW1es{59qJTT^w)51m7Zh@{CkB=EcjsZ@xgQkWYgE8>Z9B>b4NFDcn(}A>MS6EORpo?lv^gXG> z`jll9Px&iH@wEdokSG!oe@7AXJ+-~bP|55Hz2t2^3?2T?^&e+AfI_a>X`lLHO%^B_ zWY45C=+|sVX&XEovS3%=eb3#;f^%%M=$~&XPouF{ziVDI@kAY62q$SpXMN$h?{2rD z*9JJ@4}l;*AJ^FZY09`gzHuOZ)s9M8)8dl-^S(DR>ncJsB8}Crz^OOCfdBdRh1WV* zKJFR?5i$k;A60J|Rb|_Cf72-?Eh(UYDBUemN=PXn%_gP0Q$VCqlx_r6K)R%)Rl204 zQ$SKW-nD(6`~E-ghwF>u8jkIG?tL6<%{Aw5<$chr;pD7ay3Wy129y{VBs>RMKj{@+ z0w19~qqwN%=huUFfWKq^Y1mig7%%(ff3{T%i(3HMMwtt~IUIca{e~~OZ<3!#0e+_z zR$AKq?%NzGRBxkbL!uCi0=wN|I6Ea1cNb50%ZeOY3V1KF>&!s6yoUK*0quUM*01J- zHZ6HFl!@SAaT>YoHnI#75|WoTr*yg=@0Ll{q#t%Y5*7CDkiIErS)!WjfuoD0rQ$pRNJLc7Q7avvGf{MOdwzotf+t+3m#{K5A7^qC3UK(ROh!gh&I%RC>8-%lPj&4ZO#nntbmE@4p@=axA+WRyjIRY)40zwx_AbDMu#t<~kqRld%ZjpI z*Etw$s!X$pbLL79hg*W`^p1-S>8K>rK6nP@vGmKM=VpI;>h@Qi16*|AI-S=n4ajf9 z9(LjV%c?nEVZ^mI#M>b?)+snFFzwC3ZHfVhCg+Wv(+vZ^%}<}*t+N^j17yO7g6ysT zeaxGWA`WDYuVjo@F^rFlq+10U`9>==ab~mJw*MV za5bL?gZeO8zkb)PMJr1tJh<)jy|(_m?3dj2b$v364w8tF7g(qq+GzL`wJH7+@kZq^ zA{JC0p04_|(NFEy#h*3PB*u)c_|Pd}6*Gp8j6@*@qca9mJ{IdwJ-+80D(mSlXZ(St zMx|6PkP^C?{#nI|bMh3$FxpGp7L2Av5u>c+cMD>P2Y?-UEPGL@}{^XN_3>j|Gl#m(D&Spi6 z&o=7OuhpRno=1xx3IdO2JxG{d5~5-g*2>L4s|w6l=^e-+XSf-%BR1|~89QWHPX=wp z|Jah}=I-~$h_@b@VK8b+@2exgE=BWe;oo&ClYsHyu({*NRDB|3VL7qkbURlS(e1+QFZiPWkwjQ0`s_E)mL?|u_s5DYMU37Bd z52tbO`t;CxNR*yl^_+DtP=+cY2P|-*n0;s1zNlk}klw>8at2(zDrcm_AffE|F zH%kcH^`tSEuV_D$PNM}X&1iZ}{9E{p9kg%KAAAZ_VMs;z zb8lH1AF=F9-r2i}M>MtymS`%Zh$bs5J8lQV$ja84i=)}t!CVD~s%Y>&KCrs|3yx+0^2s%7eI`8ET*qkXR@6R=ZDrWFbW&E!ce^R;{(nCm z3T|rP2)eYIT$ZT|BX|2|H;T-iFET1T`$6WhXg>MLz2BtYl`XwFLQTEGFY@ig4gPA9 z|AhC9idv^yj^erUvrRTyG-tMJhw#x5J9A%?#JzYmGIfdpLONIcl1#tqzc0X!e(Wdxw)C4e5?Y&Am_+{`x9_ z7QMAKN_PN*&7lXL;7-nZ_J6)$Y8b?$n1z$*g=R&Ed!42Qj@nJUYA%cewq%e%iQJoN zCL2FsFI_QD#Ny|?fg{i*P={ybn(UQ%D4kVy_=7F#0FjdWcO$6IB}l|2u+c|COfFj; z{CF_*#e$ApPPvC<-7=I|?vZmjHI`uVys=c=Jp09?-B^K)-y==|V0>cJla)iI6RS@N z@HMNqln;SFqWn)9hKzMN`n{nIquQ!g)+1Iv=6;L!iTG$Yj4X&$O{7yOT@i49jLGnUC%6=J3rZ#r`vtcpDL(<0;2RO+=zo7zOVc@p@6-Z(rm3!@RgDO>u@`Gj;hl%AWbP zoE%E7jmwT6bY2Zlt>iE%)(X`>6N^KBVIkbS1W+%!1hVnJe)Gl&V%}OM`V;bpcH^aR zIhRzLGERa2*iNs(lLv-3mNuw(H}{B`FQ__tDCoArBJ15oo58>+EJ${igo+rwW1PW5 zEohGns3nN}_a`_~YG|})$Y@OzM}cjY@02^0NcT|}cPe`H@?Pm5g8dsVzM(i-8g^N} z+@%hj02DwL8qX*CPADw><;NODU2$-(VcgMn|67vuTak=2K0bbJx{ePPFWKtDwsvF+ z0{n?Dp=`$ZUdF8y!qZ%{J|LH@_2>cmKPSLfD@3Y6kl81=<;*4>qGb9JJ8; zXlW^^L4?AQi8a=+i6Jv~Wy6iOws(VgVvLUT3x$qrqSt4KbTu9M>^TL8m1EB=Z?`Nj zmskFBauCnphWcLoVBmo#Y=htJW{#bMMXPPdvC`FHHLU8q!K@G?%lKC3rfd^D|Kq>f z#@mK`7zr-%{%v6*a96^Urm%~y<920J`|z5NH0gT4xu;M*1o0{U`!#%?RLR;n>w+JDvfVmg2N3D( zfAnfL$Pi)|%Y)0iM=G`I;yP&%nBm{uGpK*8UTbD?TmBLM z`n9vC4-8A}vNeoCyD9av49YQ*BxuGh!%*D+f&_DMDzb{Tz7F5nn3pvxf3xm$l47lX zQ>Kf@J>*iWcN3x9f@+0FxCWNT1=H%m>-|;1DwgB%e-*wm5YVq~ZE%DO@I8lia768kO!)@UO|A3}F?LRx2FMBoH&v$>|TKNUH%-~e@ylYIT| zU6|D6F*{L2-97(#^3W{EODR_~pSRpCF&JrUNI&3`EoP zzHCD$^A*Wih>IiUI5xMiU=kLl@;f~!>*9P?*=82!WJ0l6jen!l8Bx_Ym;~zJYuU>Ob{`JaP+^=o1@t6Y>GTvI&^;Ow-9^)pMB!~ zbNkSG=@%}6`9DzvPQ3(Iq#bX$OxLZ{ehQj=VXXP`9Z&F!pgT_jJEILEBlb3lity&l%Sa= zBg?H=i?-Gm7@_n|Lz<4UY**pS*O{ z9z;a{z{iYqCMm~k-_biqY!M9khvKgT15r4RVQk<3rvODKMV&tvikwZ<+bYhG%z|45 zl!|biYp&LNVjk}5D-$MR?kWx;Y<~ZJ_-u-inAfY(*w1>3zKE-pu@f`Rb2W3v+5Vsy zi$i~R7uEd_Tm*8eNbhdeeRQakG0%>+Yx>!bLqkI+4l>TqK#c1ETczFi$C6w)Q8)CP zd_|B>v}veSYhkmU6EIvCesDGOuS$5ZO5VQR=|77MlK}Zjg0o>qCN=*6(1x{QffebGN$W#@=!y%TVGFxUr8@A*~NGWT5_| z?ysWnuhf-U|KtSaLiWRd+lo*XX>Q_Y_=OV;g#tK*(_@-bkCzgQ_0olSDZ6R-QGSo? zBHtiH#~eYN8Mmqd-#D9HR31D#s%>P4`2(<}XiD$y-iJPK zm0a4r#NilYlTVs_<7UBr`98nC%=$ao`(er){{O@~!dt`EQC_qKJ-QckC%qTJk@!b* ziJDW{k;L@Kp7LJXT2g0pkoRA4>~N-+!A8URD#(UzTq_B_c-1N|zS(U}R#hxagGtzI z*qTD?fLUE#eQm0StE6t_MruZd%WB`DUvdmz&B5={A9!KPq?3?=;D0`X17Vvom=bzu z$eLj++8gRj%ry&$i_`wy_^BW(ivgH93z}T&2h6Z$#A*&kX@4hFKN%~;~%<}@BZPFw3_2PGc-rpUuvCOn6U`x?#%!jMs33im%u zZ`-9zq{qStAYWJ_h)em+=I7H8Wv8#*lC4(4`RC%Go;r>`(Vk#yR?n562U>iE=_Cjc zAPP|JRIpu+e$ed7Cnz>k^)`IIu`9O!m+TDm9dmHY;gy$hnn3=gD9#hg#nkchVL~67D^gVwEEPq*S3g~dl9r5u=P_>;w9s?zW z5kn(SpIs@emU$Q5_V!b{PSx=1YnD8bJ{M+ zCKuk6>s_uLBqRZmkp4mjIAp@FGu^!gk&dpOD2Ko0l!P+msXM?C49vgAso!I?eZZ8#U?;P9LP?1|MWc32fh1PBLZ)I9aCD zn~mLqFM4OK)8eM2WM78F)Y#holViUEps@50mjWHO9eKG*qDLA?Pf=IG- zfx(8te!7+f{xx5UNs?$l2Jf5=>OD(CyUpkJ}RT)x*MRm?~^ogzG$ z@5jVx-b{A{q=tLk68ASo9yNJw>%P9Ye6s&pzFg|#i97Q|m)tL(jwu5@bxs4lkn|OLI>-Rj@Z7_xC{d3(HSz)@vpbh|X)N^{@VFuZeIzdIz_k>FL3` z)pRY-BqX&U5pW})VbRN^I2&qw%16Ig)VU$qZ>v&>)E5&xa+Y+P{Tq3dwI_G;L?ds% z_QPh`#=Ikl`SH;4gX9mL&IEv$7>^a{R$tO=d|alC@c-9MiF`&_ytWB-j?xdx@oKx&#@pD}r~<)8)l)GVklo?|pI@)+?wL+|D5?WL{jfFf z!e=ngAS=qvve1s!Q2fCP%g4xsN-aq5j%fI~%Oqpi`e5q@&@H`9mgd0X*^@%GuYXpn zLAt*wATXML!<>b&M*Uz(#}|_Q*+$L&)txuH0ujAP*PczNTAv5C-`Yj*v5yJ_lZ&5K zl@Ep#QG4EXfmhHcuf}V~zS`T`toZ0r%S_{|@rF7I_w5+&RWW=zvCe0MTr45+l~JmH zY47@C&Ad@Y?~#0Ql(8~^zVbY90gnmF{vOS^$M5qK%9*m*eec8c8)weg zTXcPAy4z`oL`sau*RNkb61jc}({rZ(4F&wiCOxm&YGm;dq=em1KG!}G@C(n59&sSZ z-tE-*iYQCr^ki;pEc1NlVr~<7#zS_R#JQcI8K}f%&Jz6zU1umka6Z2C1r&W3QP+mjhqZhsT%A~5I{m5FJ$ekaDCaoSO&k%<+4M&xpTLGSDs-k(NS5Icw z8Fs`6gThyPV25r5kdN*2l7mM=O~cdU`exvK}7q%fZ8>hlfcDp7WRBId3ZnX-=RrzCs(O z0QLpv!skEM(kFK^(7zLg-a1cSM{7DGOyWizZ%D?+%b-(GtKoDUde{;zRDa`E)FsEm zN2%Z;S)*<+?Z86@i~tDKCP!%y0``8&-$_r@=40diku3#OELZdE1)0^YnL-_w_H#^o zw-WVDnMv4ehYBqvT5`Y2k@k61zwM|6HOz)~s_z1=k8X{0v_}(KoS%?NI_j+RV^k2FhK}Stx&}IDfCt>~q0b=DAJz z;rcK(a3e$m{875ZW8(E$P22t8XF(L3o^!l&xujgy9MJs5;f4?)_dfZmtH)j8Rg)#s z?HYOsZ>#kF{y8tbN~`$jq?Pe5uaeU~M4M#9PZ1@YG_z}iLNlJ|P*XHoes;(rmqp&2 zP)NV1d?0;}#VA37Q;e!5ii(ra<$Qmx6+QRzX_0XKnnF%4>TjD_fBYi7)d{u;Oi|}S z>fm5nP0c3Q|MwE}b89OHlQ+TU;=DoO9J5?A^g835OTha^JoWq$xu$S@g7i+vpir{J z)=+)h?zWPQaI+;l9#5+8p~3y8(8LwH?BHp(=>BKb(kqj}D=KBX>iC#ZIr+L-NuspN zpXsXaTg%@uQ(ThvZAN@Qd2OrA_?+t)%U$^ry{hbIIdda`xwYBr*BSgm|2=>ZHnwDH z={-q+`kaX2;(S2UvbDFEQi#R7-u3eDg++?P`n6L30<{Z#|P(rfE>b>~3(6(VCq;&i`AW z*hQ8Bsma&mR}DySjr|pb^oxzV5+;x&HtRw*)xK1re)>gn@{{EM;Am3A-YlaXpZLbY z_Pw8D5vJdJgK+|$sj)XE>7o93_f_9WIhzOwMyK2as4`?yziloU(jUyWDu1`>8%B_pjzZGtofG6h{hlb{9 z=|OA%1so`wtFR)t7OQRj@16dFRtlJo9JqPusq**d_{i^ZW z(9Aw9b|&F__z-Dtj29{^duGd}Om)ix#GeK5K6ij~+P!x7r15CRCUSx=a=kuxjrzA8 zA*+Ib%qtR)D)AJDt4ChxYpmb9I!s~;tb^Z}{0{bE!4bb-!Y}n$Dn4F1S6;fpLi%LS zK1R}#u+*!fUFxPUu!rCrga2bG5yp=85+FKs^S^UY!(^@>dbxV@KEZp$_@;vMQk6;CcDR(%9F2ZR+IE znSljm?o+T^@_C0ieSZM+a1d|G4k+7Incw;seZY-O`S#9}Jzue#v{5UFfR1i#`5UQv zZFv4Gys~faWW2zOUSto~phBuG1Y~Dy=T5#hL+y5m))3%E6zeT7F@=1Fs9l%3p~1f& zJ42Yj1+PY63pZosZQv1=#pH^M$r;WuiH0`WP3bRPlL+4@74FO`!XjXM>LnPC#%DXu zohtgqZMT74vN5`9C+X2XeAIGv_ge(amwPGCE3-a4l$!|2VW1;xsET=3m$@xQ$ZsSEc7*kG~t7U;;)XfrnT^osbVbc!xSMDKYbs+ zOI)QOWal1S=EA%(|GXH!NiAxOtvb!|c7cU4((lLYQ@i=Btpk4kML0fL?E|;o%4#&X zyb@;-1$`U%TBb)nvCe-kWxJ~&E5aWErmyQd8{zTal=r_iy^PG>n)=dIFfezoVoSmJ z&)=5w)W8e;=|g6d%9Hw<9?JUbKvyAd?mz-vfXVfYMiZG`a<>!uY4LG`T{CT;KHa!7 z>R!k;m{Xd80)-4x;^n_I*}}>X$QH60B9W7OTB2{gq0POFANh4(N0Rqf58S2E zX19;~ZvlGsD{M}-&>T4cQ=xa)UKz+|)obU_6xuYu_$M&#n4m><(CCvCPGc2LVit~T zwHiszYk6-yqulCB@X}XoUSmo+asI<(>DhxxiE;tzUARn8OgN|pW$}I$7QN-_{!yok zvjCP5Z;0;w(mQp0!@eM$<)ys(v$^*@@8af8={Wt}Ava^IlZ2n4yV9!}_(CaYjX{|P z+t!n4T%x05_Q7{ao$H`&D1J8>_UceD}L&0dh$wkPZmHc&ysLZgiNHeUDuS-y04CI8m=&)&(4+Rkq*F7oR&Tu#<(>c50Xmu+t87r?NC1#(C_o+1^V$>WQ@kngJcgoie3G&*D2E=u zy@s2j6;_yrNL`%|M+7c*#dg7xHmm5B!ya;`f0c_bDQD4S$CLNmciDUnYpt3S$bcUV z6Rpv2ra;56Xxa>s!1F!eUuhLfO)j6pJm%{|k&ksB^zXWkBX_B&s92h;aa%Pl9J0+= z;9+^=W~yIlMS`?aR%8x@%?pYO?CA)Ks%r3y4`pirG{Eoi_Dsuq!IuQf%|yIJlvTx{ z1JSI-ed$n9`@)G`4(=<9#6y?1vzFJ`fq4e@<^_G+SsSVOxwc%`WN<3T;bBUSRV`B~ z7JSis+*{6+mG3KaOla;cNJUQ{!)Z`ImlMg^O0KOR1ku!N=jHDrH}c6)9}88SBE!0t zX--&xHV%bpmtOV`EXJ$+AVQ6+*e1yJmm@ehSgq094Z()NeCG}x(z+cz`(g?&Lc2lh zIo-U{n)iW(tEw+)I^|7Ya=xERkI@M4q8;{Pygo`%S(lU z!?%GAXz1c+cXyT(TlV|K;Yf0DQ&s+^J?qZH+{zRmLaOoz#x(okEt%0gSSA zy1qd~>$@a4IH}xzk02Q~jsWUkgv+poFy?Nq^qD+PfaR6{YMQjL9708XjM;qYLA=TY z*wCs}(IzFaw5*BpZl^d|Qj{K$7`Fuk9`~-a{XkVI5u{r!mR`b@HV$9yfbW~^Ac<)h zP5#O%ZG%N`bGq?b8fyrf7cxYckiu_hou%RWNvAA_d@APlG$_yb#dN$2Z%caf&fI`Aqd#5TZvW(fd3r+{(d|)~9-Sw=oR+!Y=*<~w zcqoS+5n%B>`psi^w{y9R5`MDS7QYKMyLzWYE>wgp zH1ugpJ-CntRz)0jCJZWcWeOd^r0O?rr-uKV7K#Y!LcmdXd@%gDxMl+>@B#3yvto6v zl#_A80EsHBHZ+!%4IyWOJHWcR4X7mc!+~LgemlZ>R7xS+U&+MDt~KHlkg=XzLi?3l z+;ty4z3UzMvDl7>$@iO=B~e3UO{j9)nD`XqHccSon;T|vT6=yjJ?w*|f;up5r;fs3 z1I5ly@QgbKlTOS%;aIzMCv9`KdG^rA^%y4lA1KO|s4__N$@M`DqIq^s`Z5_O+yK`skBNC&AAj_W`IC$vhFQLM%-t_L4VwaElEcyv4FcO_%xWk;Mp{bA zGjijYdoO!a(bejlDX^|}0}(FS!l-Y7`ismy2}P6L&66MGc;_|4@o~x0;3TlhowZ^; z7IRfI3t_N1{JTASWh97a7l&l{vq??L!yF*-UOM9?u1$L;I2nQBQc<eGAzUamgd)Ovzv4pV4i5_rWSO zQ!y$Ox8s2Uv2;&U>ubX(;M?yijt>ym**As!fV+Bo>f{>%3|8f&z*|c1#$82WLeM+v zE?EjByDEfd|aI4^3)&R-+=*U+d%dMJ1tY);6$OTUa};96NvH|Hbf$FmIfOa zQg<2#P=*>8#0IkFMdtll&*T!sqhuFMMYrP@h`TD#g4J%i5IujdqHuEHrzlHyh*|yM zk)wneR#+-JueicHJ7<^AJFf2wu#Kmxxr=l)anBAzvmU;AU^OF-Dqxd8>8G$@fW9X= z^m_M}P1uFNRWBqsYr$$^w=@N9``W}ta|3X71pQd;2=TNT~g_wjMY2 zT&aOW6PQ`|%c~Rrzw8e(T#{^&mLlO3ll+A{Qh|$w6+26vIm^XT8(r-f0>OJA^hhyr zwK8DeqNok{JE~m0@z-M+1va0Gi@Bbz($_(%x{6Qq2wTS~%was7tyCO8x_& zZi#>XPVG(hI8~4UiQzIeU6_BxPN{3<=`xG#Lq4XLQ|v0-d_gLC*`&OB=MjAdCFhX$ z=YWXml!3Q*&o3r~b_`7poU%PxBGDEbwS{k&8CkNt`SbR5XtR{2cg595tq-s1ngyco z=z6M}A`r!kq+O(9h!2Jv!hzcCgG;DO&*pc!fag zI^){5FFb^?ct@~c2{y9Y-R`5t=_rEZeE|%174FyUC`DPGf1N`HtGJRr+IC(uUe`#B zs0Fi!0#TOTRPzQEU8D@@rG{x>@hr-QA-3r|^yOYRO$1|HErGCAMj}4O4TF`Tps~$W zulQ~*+Rw-r1j!QQ3nxgwwCK+;@07GSShHIROCv?^O>H}PA5ef!-W2)2!L07$ovzGi z>s@-k19hkrf61>PH=Nf!>h%(bY9b0)21H1*TX#IO-M#2@8T7Qy6R62zVFWe_EKKR# zlvYdF#W0xt0K5OG`jDfHP@B#me-6x|*%9Bd=W;}_Kf#zfH+s6v7YnakoJ)UmW2W&7 zQ;#po*vk7{f?>*P=(>=QPJJDNfAE}`)w@sF+G%XuuVl>Vo>>?D#FN%yg>?`drmvdk zeN!gxO;+343@^~rq%a%4X+HyNI?MG*ZP!!h#?2zW{2dzVhSKw-uv%FGN;n-j0p zXo+b9vP;J4TFBw6&B61YQ{?l(mOo0)?!XHW^JtCY;ue&$oh;vR?7!cYtrpU(MREUv zK*yCbt`wckq2$HK#z4ZyDG$Hd8uxx$`C-3+hB+`tn{`3On3GJMN#>La{M8ew|9#}q zq?Z;+O>bg&8+oiXvWjDWtSfzS=QPb&$lNhtIbS9+qPWz$%a$2MBh;6_VD??rd%t`| z&P0W6ja=nHiaKUmQTTb=tKi%CP0@X^4q5wish<6jr@XTdR@;u)s$E{!l7>$Oq9CTR zLWE=?QnEjh;kUp&arSw!4Po`Jbl2hP>b{a5!+mT~`EvqD{zU{<2CU^NG9N|p^|g>b z77v0*>>TJNoo}@bKcrN)et!74z>`Qh6>IYs{zwsSr2|pvHFtZt=HNHxO^Rrxmj=eW zjnc2FeLVF>sj*!B&dsL^82jpNq}DqFW5|5<&=nhO!x&#V?}}xZU_9;`L$ZEHY;nIgOO?qK@w(OMyJ; zMPg7urSbP^3|(G3a?J+vcOJUddcMS$(IpB{fnL5$ZhlXX_By&Fg~Aexk1x7JtzRJC z^7{3_i?ig*3>QSd{Cvd#Cq$Ezdl~lth>id_+|#))kZ>*@@4CxYgxalqYkqz{WdpZ4 z^kNWrcGGnw0j4qxF6+6PtaDUPq$9!t8Tn?CoLH@~3D~lB$)zLp0=wTy*BLX8`({6T z5$GO^re54j_4XQzd}@GfK20u^@!_WJVg&`qT}#4I2cPtQCOx+BG@Y~^`(b{=?k21< zS*W=0@Vy6oIupS=%08_-a|^+lkAtFj;`=A4O#c+WT)mK-?!4g(ofhX&xsYM9dP@Bh+!R_7ZKHo^($IixK2!TEPZoHXj(nMD9uE9Jo7 zHJ`P#f2nDx()$_* zhSm9u^wt9F*`~(mL!ZsEuA=d-BG#f?{xBGt#E~AGBx^M?nQedOPj)$0#PiG7OFK1O zcPH{{NO1Amu-nr=I3r@%g+5pVwk&G7pDmkT#(m`6`wU#o+BZR>1*q8|-ysK%3oj(^ z&?-$Nk;N_;Dey%-5qLnpdiEpnOLz0QPCq_$7LQ)x2at`f349kN;irX?Lwj?w>PJP5 z)qL=s4hetyjylk@CJs`_=5!Lxz`Tt#Z~50V(;(saHcJ}k$kAz00Y{?|`ycgEv&A!W zO3q}4BXA^^3GnFTe#N)RXn;Y#yA&p%O|Z=f0+DL&7^T_8OAy?%U<{U_{XZk<>&$!| zIv+miw;MTPsC~dxlN_oTj!P!Y>p=f-%|t9L;%&xS%r`~BuD88fw))rRa$wNM`F18p z?$O0!EcfC=m_^{f@2n$kLZKrDVi4xT+cO_DFgtI1kObV4Z{!R5lBxF<=X}L&} zz-Lr_@^|HjKT0J7eu&Txm=2+|W$62h`(sMuI2T0``$j}S+L=qy+u=-Nx!A``0a>2% z*;~j`A4$}kcAv6)igA88Y9J)49|{bw~~9}IE_`SYr|N78bLue`rw)!L`_ z$V<)mOV8OE{ohPwkctIhQ|#BW1}2?kM4Mx8w~EsEZS=TKb7II^46Dg~dypEUtGY&( z%&@j}47Jy|@@<)8eEvonL5wvY8)c_!^w&o2tj~$aU+{eVEMu^L#3SqOqfE!gnY{bZ zS`xly$2iJweig)iI_gTPbIs32M8EqI%I5vZSHAn1Hxf1;jXAF?6dODX1Rbl~CCRf* zJ^*&l;8~K*HmvjFqXSB&Jm&8K+W^n+63~7Ghd80L$^Vt7(^Z)kwS@TH3rfWl;?9$g zAz;naHF+H$Y#T+Km(_K2paJVP(m*@48qSYiA6M>Q$$nSH_^jI2x*1(6RXyOGz9=0e zRvg#>Okma5zyvQE{18jHcZsxkT`IoY^&*rQi(-d!HG#sIhs7tB!k(H+RB<`zz=pDSlG-r zMxWcpv_&_mFvE*$c%78A_+EwQuIV{^7d*k0v;!*k|&cktbMphY&^fer+cbg>`vkHZ(37=K{n;p}p#9hiRn@rsny6!t<@|6$Oyay9~h zp-R^P^$RS5^5|9?%(m7*YZ3O5U7iA_gXzwG)1~yIaRdn&nOW>nQ{s`p^orfg)`Sh- z^=+GD-HF*;Mm5>@amUbOPnn*Z6GT);)tZr+gT5oKC-<<;N{^Fq*MHqy-e0czoj91v z+QCRO2)gw~A)_z4X!=eAa!pWu`3#S20<9}f3M_zfo3S3u35BuzL*W4Ze}zOuQ&UoM z1&;)4f&Fp=BGKzL6~Wzl%JI`GujeZfQ5d$~_XbvOQu^H|yo*UOjklRMqgHeU9Zyop3T^f{m&A*l!1<)@I7kz++07BA#T8bDdu*^uXPs4oSOn7#w{ou zX(u$Bt86nDFgg)TXSF%o)pyR2?&<`|ZPJT2oi}+aq%XB6uT@d%viHWC&@b0F#b*jH zAI@hPcz$Ou7{5-UnXg>dm){lK_qhRy26y;F@?lX9)opV`OOz$mKH0$OX6vps$O7Qm z`D`#ue2Mzg65xY~GocKzL$}su9T*y>zbyU#bxwUewSV-eY;tmr=R}28V>cgwM>zeH znNq%q)Qv0@S2_Fz_FO6r=_($@-q6-JW<~K@_!Uo@}Bd%f$yx{AB@@Je?a4J zCb9Can#yz?coq1A9&=#~U5quUnF0N}CM#fOVrt!wkvXQ8V;hh;*RF-j6ZuZLUg<}FN}y6eKO-4*w7)C-Rj zGE@gYf&FMq>>b)1q&$lX*qBGx?tRxvhAXAY?LhK)4jsw69wT*w#ilwH7SSpd>q@jL z4Q~qnMTX{y%uubJ@~Lh3nFs6MugDr>W_7-je)B2N+Y;U6ZP0jc%gSR|9fCq5g;9NF zJk!;npoO0`-yxl=6co|h+g|GWrN-jwTeZD_)TN4_TNq+73VQ#%f(r;Go9Jey#RfIhL!fR$l#%`C16K z4F$3|xZhwA4YL`fb8Ebb?s$W1)wAvOz}X>^&xA*iS>2vQv4&jj0GG#0Bz_g7*&-Fa zhYwyu1hN@c!VZS1?CrjWRQ=3?2N9MSBOl z!rUBnR|Vflm3nRS?lZ$7DSRVpFV~=@dOYdbr>JioIFK-%pQq$ZIv_JAXjxmc(pwCy zt4}^l=C)e_n`9BUFhqE1=#JsbHo>(2?{Ua1(Qs;YmS7zAp&a(2UiV0{>nc_J;4H_F z_H34jUznkrf2x)~#r~ON-lvnF1GYqjt6%fq`cv(6xOzt&OquvM9ec8#!>k;+U{vv# zyhuR%i@V*(XCsCu)aC3U4CS`e+&W1N z?K5W;pcv=F96$M@^rR6%OMJmsDxo)Z_-4ZXaLZ_|^ZRCdvB2j{i%wNOd*`cca*HWZ zMw3q`4Zl>ilppg-Nn$^Qw`v{_OaI)r0v!<)JX?ZY=Auh13|;*iIPpHdPmh&URA7}l zjUAm7MLTk3>n>>J6KJb5-hpi|iMn8Z83sqj^jLP5F!I8mLq^;oWWMDBe=j_vy=%;* zucOGncZZGmNqf`UUPx9i#?up-vD7s;1`+7bE&0$Y3EvG>!#8J9d3_>6^;cmrgIyBP zA?Wy zN5bcKfHby<961R5)PEuB_S+cD3f0HwGmRG4A?O27P1bNehn-S98LQa$31fpK@zbfE zj2GhAOMxl`^*40>GL>%m1l}+?F&^$=AiH?lh+=z!ra0knyV;z1C}^WPYG;4!z-DGz zZcGii!fGJ||CPR)nby~8J84B6nAqrfF6HDJnWR5l==Hpgp8E97V!z3E)K@DEit1bz zP1gAOOTU8T^6&91lI^FKK;5gn{7oxu0+)O7X*l(fj7^HuXF|1XRN}$$%G(*8%_{gp zQx86+F2+((qCZgSau7XnUctWY>S;3hSagWRo(HjUV=hYscbBI3kU6dD#9=~c%I~O= zulLAsv-3$rtZS!9r7^Wj771h`A5}eM*o&bL_OUeTzeen*iq|2&c!o(X1`CHRdJ z$iTNp0K~Pt=sARm=h#6i4R6&w||^l zHgPlWQjv6xq>*j0g{yZ*3dBxyh=uWJ5Sw|cj4dE~8v~so3?Rh$g!^hIJS<;8Al=D; zu`p*fls+k@{@LTKu=Prt^1niMjA|Qswsvt7vy|bscA{RfEvxTM`Swp|R$gwQ$-1H4 zsS)T(N73S)N+uMW_H803>t$}%bZ6U~VC(qC$rP{3FXr#opX_{wg4sr(DqD$`m4gP* zZ=-p~j`x0(SkW|MmiJ?Zl@e33Taq4CGKUbK6tss*$)8?875B8qXZfx6&Wz9Y#47C6 zwi^BTmVEC|b*)W{RzoKuTl=sQf4C^5axo3gOl>hq; z{`1W_GEe{RG8zkRF=mJm@7TI}h?Km+CKXXp0MWPri6T!wM8|PaCYrKrL zOvGOZ#t-SBbe_v9d#+@+Wc03hTe~DpFTwoTTn|ZV}~~X`32#RwBs7*TRW?n(yb(@>#$}{eF2P#7WpcR zC&=03sju?>!`(l~W(^T5VR7^en3LdJ!udG-{KfJYnAq4phyS`>oOpb; z_A+-8%?O@H{ql+TuRzWm@&)dPct6 zZ^fK{wMV%@=iuuw)f4M>?6AF0`z|5rdg?;9-O8DIJL31_Gp`kw9}3)Gn`LKOV8lUa~vQ$8z(==Fn@p`Gao;*-i7nO@097@>^YEmPCxC zj2x(_pK>pq@mr(;Cm#7k>q1+Nrk~Xq>y$Jo?S31vQs~h67(}j&?}4AD-(!2mHgJ9# zz-jf=7R}X{zEG<5UW79VKX>c7o#C6@thHG)vxK(G3lp8HJ2)F-sMZrd9^*C#3sR)> z0f9XlxbXdBX%i3wfw#z5rgltAt}J9w!#U6-0e<99tV?|CFWGf?hQ!5msY@tknv$h^ zXeoC)5yI{R*&&0T>U0K(dX(r1S$N988u^z-^x-7`qtNY2P$H}8Mw0EfBM0aQit<23v?yiyH%-HYoN?hR>3l`s` z51IuFkhtMq9Zc)al`h+ctW78UXW!SHU^uVG?WXhN9gN1&Sv8bCu1u&8<#nQ!ZRj^P zZO#vpRetB=iuU!vuci^6Jz2}FzoLNlQ)-_53>`wj~ciAKL8ln5)&2_ zW-Q*JG_@`{?R+@|%WPSew9dA2Rxdhy60+98LXj3)p&P>U?&AkG?-Ockr2&utnxe7o zEa4`f)Ti$bXI$|0?SSRUaS3LB;&Y{GY++OiUbr%*%aF${%fXu$gu~upNqYk-%<_l6!Pa~qVJnqWg-^F!{syab#>42#Tlc3;XVK99*$u}}SSZLA-=lg#<-oHQb6nd8fBzN1v8`M2tm(nFG% z2PZPbWmk!u27?+3)48x;tA1&`=6J|VC#|?A@jbn2aoY-9_3MuDju+nfg_yTh(4H3O zFb7*&^0ZE6=|4>&eWPO&GiWgtfOXj8xIkL2Z=jQ3W+pMg+(i@fus(4QOHLgx&viO` zPTP`A?dV$K^{VHrM`GHAme0^BZ&~rI2CcS5l<)V$OGGWH})io-aU3Cus6Ta34Ccij?l_R}cqY&}4-p%1g%{2^gF+~S{l<0Mk zKr0L=iR?RSjbt`?L25}IN`1livE#99^oR&26W_f)-8~Z011Ly?AfeLT3`j{M(j5}g(*2zA`+M`O^?#wu7cO1gbI&=Ozz5T4P>z;ds_$GSox@$pWbeT=7$>@Z7z$*^C7=YN+6gJ z>22OuUDPXU`1V}KBk$XZvP=x?Gx8v7YBl0j9VYHai5)!I0RuX`QYVB#uWi*lg6&%@ zUYZjmS5cbTWfE3~a8{o9c*v+B;+463h1O(Sfa_RZrOm%mJ^WKqfTU-C;j#6cMs$cQ z6>V3jE!k&7i3`Gair|9`b zk0h?`Xcdbr?8D!KHhJ1*D01>uOPM6qcS-(de^#TJb70t9t$>=;KC5saPa{DAT+a)u z=KzmB*9SCWOJE@f?31GIB>a=i1=xD8T){k17XYHwghYV`!g0QSc#32QTRa$8_v%|w zCX(TzYe^fUuB^Q0&-N$eV@b?_0pZ+Zr(|G=|0~Dg{LzPUBW!kq&3q66gs$J<4ejg% zno~++qMc~zpU!*Qe6nPWylFK>Y4Wu%r^mH1mR{Ap=x&_tCLySIHddTPYs-CclJm6 z=3Lr?&rw0{*60ulrP_@kQ0!y2PA5LtmeK$6ve2+N0uvh>Sk0wyc2+g`KoEGxF1r6FuGGQ`IjmqA>JmjM7 z5A4q251p(0&#<;9&~2@oR~v;*C!T(^3x9$J9^^Vvjm8u-FR_oGvq@kBep|ijc>PU# zzmoYsN<4udg9uubP%0y-quvd3J|+R#+AOO(cFoxYc7r-~!_3m>7bLQ>yz!sP`7H_| zR)eW}gnK|D)(eh4|Hd}BpdP+bfius*(6M&3?IWq~`jaUxMX}J|#cbS^hM5j_i5g{|vKjicRZ%m`w>vPGy4oDC zea<#4Y~GH**!!#jonaXLqXjYhjik#%@orB=|9~vo+e8{N&|LFx)pBUo#i~r|e<@67 zlk>&nXFBwrMu_N0o8=6UHm_BA&?nZk-Pm%vo2xwpX~l>w_nz}@M>P$>*pBzV7Ko*q zlzxQ|=VSv`TkaU>B1{Y>o=hy6PFDOLH2N&{Du{G!OEv1{3sBMWFTht~@ofG{BWXPA zC1)fU@$5w=X7;Io+g0l@J<_iIze7T19E$C!`tGqV4aPTIIetMtOUw{I&{HioOZO#k z*3m?Cxc{ymT3Gaumduj@MylASpKU5wM8M%4!TPFbMiA^KzsK)z0S5~{@Zob^g3qRO zTk%s_UYD+=^M&g*bd#s8>xlf|Sfb4)oZCnu$(WPPizMeeHMwWGt`2M>G%|@mJ6sV z&nQiQPjD`>OU1IjV3p+z7z(64kw6L?@~JROz-3=J1Tl}V#`HV$J-hyxmYz-5v1wN2 z$DXfWj{1fExx+km0rFjown(tQG6%-QsaGG+3WWBVifCC7F>{SU6{?vQaC|nTdW{4n{B6nF2TKS14Moz{qOr;#$Are^Dek?M_#fh$qz0_EPnE;Y%G(XhldadscP&@{&4sH zPn58boP@>Xi2A6Z3AHI^c(E&+h1iy>`nOlvd#`5WVDiRohL!VqNFPmD6>ezD0TPH< zB$J@ftqvd8%OM3EfP7MF&P;y0w_pzcP~eAJOaDAlcO2Ts^TyhHy!5kaJPbIN^TBuL z;Tdjclx%9;O7L6?h(aF`R6qr_#)(NRO(3oFJV=Xd2h>kCD>?dO$xSp47**y#3IvZj zv^{8%!IUO=Zb4=%@h_qC8y^0jfu<^XBbs2BwdtY&kxzd=C>s1AuCa6iY}O_qMW(M! z!zo6`V|I_syEHVlc?^C$U8FPB8>+5Agygg?A2UyE=92~<%##p?)9u>xI zdkj%5dW}Z{LD!xwQPYv1D7u)HUc9K~qMuK8bFc;ik~Q{l*ZxnKb6nJFkW` z!nf;~+5PUQyZlcaJ!q~<4@u39bwKZ)JYr|8r0kyL_hoY{F3Odr(b#1kls7ZKZdEIk zI=6kfCGh1tG^9}5AC?~Pj$$gDXW6MwKGtd6mYq8JE1&i#d78;7^`G&(&a(!%HsC;# zsd5V}O@)!iBybG0o;eHWlKUWswj*#fft(F6uaz+Lp0%pC>cF$pxjzhfs9C=FjxWVO zJ@HuZ8oCv`wGYllsVw6Ut_Rm3l5zzuSrmN6NX1G(6A#j8=c1l3p?9}LbzL+85tHM_ z3LN*b9JYBmuIfHjw-sm>e$pLOQr2G`<8KcW+b z--!u%suNA=7i!oTTP>OUMHYjq3erP{H&J7A{a=3#1lm@Luus&6gwTx&RRHT$!`z+% zaSt#LBIhtGobzI(Z5_^Ri^|*$s+`}`!sjVM3aS5@c{_*(Y2lrrb(JM7gKHHmf>x23<5^}}R`(A0zmp5ff_xbAedtBY2w5D<*J9JG~xV{Rcq2}LYGtZ4lye|5=VrFB`*@75y1)5>f& zm{NNLSiYmHz%U3MW_mu)bK})doClEBPY#-QId=mG?hN-fhKNB)jyo!sHv-_0j_i&R^ zeQZD}v;ImhcODf|VszyY+3OKQn|t-X16cXy!;+HZ2_o@TltgYw?koCxNR#@2RbjsD zl=16AV@Wuw3-OkME{q($T|Sb0a!WX+I|C{SC;qf1GPPr9 zY)xyb}+&LJ0U08?!4Ga4NtF%6(=6 z)jQiGcVk8Rcm~ow_H!uvur&Tm7lTw6?V$V^xbQq#6cPZ)`r}6+gIfg>unG?5;~&hq zE5O}Z%PJ21+4dE)aI5CXyH=0a_mp~X*T85)8rUJr6c4MyrDF1kACd3(Ecee}N@_@) zeTb*n*rMfHa{*xeDb7tFn-6&?vi>JFX0z4OL%Fn3QBkU%e?HjtuCvb{Ze$~zG4z@;&PPi&QfHn^_7Pez)sqL&^y4zwVi;F z5cs>9kHmS*<1W}njg}lIJ1UZ-KQjokA%Y=|(kPa!zBCFiQrFNMoAT^9O8~w_DL}4y zCa9?0NY4EHYt!z4LL$ws{;Jm?_L8{s9mUM?H@^-P50A87C)HhYLrh!Gr1WJRUY8VW ziscIMEj6fj1jyl1YBu7!H7WS&6m&9zUD>;j_7o!?U{O+YR;|k!LN3OR4y3na6Gj%D zNT8^o1LJl+k1&3Ihg7SYB-dPFNpY7fT0V@}6ZP`x7`XG^0>|qeSip{oVm@2r!so6- zdagR_H%6KK-YQAx|HK;jL8aE+@)kTG;&3RxHnC1N>!!_Cp!M4z^aP%Wns>3!u}gza z0dv-X@eY_@8qUBAv1=pW1di`{=M&lY>AAzFAt2AC4_MSJBN}g|>65M4Z40P0sEpbJ zo7z%+sj2mfD@$d!LPb1GuY+{vU5-qF!hdw+HjxmWX`erG-JegeR3PLKaUcRf+94(0 z&T%Y0o=F$Fn&NAh8*Sylu2J&al27<};qi}KF}K)ngFgL+{e)`SL@#wEnD0>){qRK5 z(pjzP?x}TeBGX4zxwAjaYELd;s9yNF)={QW8l&*H_2MMoEjkPkfOy~f5h{2d!)s<@ zty;~?OfY1GtMQ!KSDxNtjifZf!~8VF$43ZbvFA{&^{w`k?{3UCqenRH?3;Cx0(6Tr zJzkG^J-Pi&Q;zS}4xrH$>)_Xg;~8$CKWY*(KW-iHt0)7)F1*x@Kz3M`t3!WQhxY8( z`|Xd4+ae$?qlZJs0_t@^A;>Rf^MrZeH|o^dnyln9TfyVt)ia zfv3A`K|u2Vjs7e7473nbLZe2NM-6wWVEN<-|u$-)ksUG?T1n=t!;$7f8D z5nBy9#`6}Q>#!r;>H$OHxM#LNmW{H3cZ8V^oTz);83`gKQIIbjMPePdjV9+k z)Ud7Cx}~-?+R-1p^w&LiGkT-Vn%j!w1z%LJ6PmXwQazL}dl-qYWZ&})2}<%N!_2tB z2d~n%w#l104I8T*D=HOce_}r95H*h(F`&{MIrC+ zT2s1392aJ08M{@FOT!4LAIl%C`ke>@b%gh7an@4}EQ9S07a>WW+K}dw*Sowuh;J5ih#~^ZquAAvL`c;lQKMS=H zPU?Yu#}s7WT>^2Fqy6xEb~RxWRX3X7sE{x!vd~gqYEpPvkKV{>^(vFKaUen zAlhYF2>o^$>Fu-CC||554nIkk;B!b^#vLiHC!j3=SY!+w8wDV5{4^jQEi?7EZG7({ zsP`P(1X6DF{_FFwI5*KE5j!8pTa}#p`9We2z*0_lcudXFfz{D?o*N=8R*zZRyo>uuQ6vUf)!W$ z0H-Ers@s5`PLLu|0D)O8!*b%8KPg)pPox(jc|^bHUwB>W43dj(j8)bA!F=6xEKla$ zE0kA978x`#kkWNUo}}FGhp=B3#Hd#k)}NF1^;$JImTP~=U-ot*2-qVSH6zs7cTS!E zAS#_Cbr7vWRxz^InDWOJaM7)6@moifE=-kHdOT6Z(v!j>i!{9QSR1`$fu_Xok>;J; zM`J+=Or-@pDt)}YOw7IXwS43@*ZX9Dgs5c<%M6F616CNOPey0s^zsSC6o<1{?(Ebg zm4kCm7o&*WIxEjJMMQQw->5JCqOW% z+X+P@i7kUti=CImudRT?%glk}-}A?Cd%eArlY?MDsu+uT6$#N9{3vFTkxx5#A^$s{ zYJ5B|_+^Xt{m(O=Iu=IU=dyr^k(6k(k#E{tn+|G7Y668DEzbgp-jY_v3D-?h(Foqj zrx6lU>lPnOiq*4Al}M?noeB=&v#ta!)+Av*y0OVOs!AgkZiFvDDBnd-Vt}VN*N#sROeg zy@4qQ<`20vHu|kZvR#yjZ$!mbnW`TXycS-o0}B{yI_@x&-}NA{Vw1wsU(W+KG9ALf z_&CZ4nMxn^GtFr$4CYX84qC`B(q)kHNRvB{g#bUy1cZ$A{di7+zEc6Gl5eyiS+0=k zIyc+(Z6-QY>c>MshO%PR>rL|r3hIJ5*6S*sW@vGl zz+la$PA(fMj+_@F!#q;TaONcCLH-@tZ)Uj{GF9d7=y~@|()*vJ3rVb9WBKQYdJfc3 z|3nZ5EiNJX-}X_7b=UBqQ<&hKUSn+k4cW{P3abMkWW;Sc%c2s&-^^Qt==jYg9r2VB zAZkAH2y@j1g#S)e=A-X`sg6D4ow5JU6UyPf4m-oY$Njx`7QJy*b~gW8=Mxt7n15xf zPrK?xGBkVosofxap7hgd>(XtY8a1s zcXd57>r4>1@O4zLyYP+kkuuqmNI1+T63?-dstFO4+O_SA%@6bP+Z=Mc!#fPHwgTZf z_+EFH3{z@imEH1z0jo{5TH4G=W_#Tm5_EwG=QgT{6ys!Y)TB=;QAv9HgvvlL-$y<4 zZWkokAzZ@1IPv)f+Hn+&GVAhVQ#gK$N#XM`T0OZ_x2jXaaDA*W28ZHshK` zR3bs125nw|&lVM|f5ouy@-wUthAV z3%A}*r54SLme3&_3q);%XTOkTzmUDS9Aqo)13MDM@pu7#toKnE4gDb?T2$;1B)TC~ zTF&<;!Hj$mU0M;vk5RZiOJYTddldEcGve1QaL@k?HKZXg^nJ+vG4zE=j$WoSSKzMM z+e>!E&CGsu@s*+4G@uUP4zx&im}8&lPIx2|1S?64O?%-K$+2G}6G73d1_?rXFU?W<@yrMdaJ>So9A z#2U)k2Ioc5Jy3$xoaM5cv{L~%+V;;vaQFEzJsYNTN3Q9n5k_FLWc$%~AIT~O#oVp$ zEk_2@Xgm%d)|{O^!pTlgin~zPO^nH}xknzl?~|wb9T2n+al`Uc_aysrvd2DpxEX9_G+fXghu&a zwVzL*6pOqj)QDP{7bE@KTUzM@*s0+f2120^4G6kam8ql?f4P4a+r0udXa>zohxvLA zICldKMo>Y*NCStcD215!9xPuz7BDAF7@j=oq3PQv#%OYxE;mxr(0HfJ7qu=!YwEFM zdgX3<=TFDOBNPM`q06S8R3N>Q8xOO?fBqGGZ1zGYGjnl>1ME!7n7@a`i`)b9{Au%) zN%PG4YRbN*@9ufkJ)Eohe35qgqXBvpW8I3tMN!icD$|{xA)@4%S`_~mLZ`zgOFg^o#HwJWJ>bqSon`Ort1a`f(thjKNvJZ_% zik;2?J}^^b13e?S0~g~l0Aj6z8=?!8wc(gVPglTCK3WCL@Gjm`@hX0a|M2Ef-%?MR{#KuX8?q`hy8UmzdDR???BiMSCU2Fp&M{E+K;{7)xfG9h6!+Zj=@8GACk~Ugm_2{JU+bglw zBm6GIl{eYKu_(JDZT#36qRmwABLq3pT|b&0drEl!t5 zt5e=BfH&&~M3HbzVmAze>;_~cvv+_FDv$bAIRDr84|?o~O*T-XLwD>tG5@8B&d@`x zZzN--=>g)-xkO;OSGF164!BX1oYIzTbp4*XvL7D5Wy_Fh#?&nQX16I>X*Vp`waw%mkVo(!o{?J$p&J?zHQmBgii@-=Uin%5wUazIaDKR z9La|aDNA0WZjT{Gy@pw)7lIGh_s^H&BQWM%??k7qg+e9H)4Wy!PsbIH;q@A@7z^TV z1LYH513Ux)U;HSvDRM}jN%_fXvW=)ZRxqCp(t2OZDQu#sDT4oCu>XiE3pr4t?Lemj zP0+;~+l^WbP?BAV$Ds+-j>d`!?whED@kaKeb8)A9T*UdMkEqNxrM!k{m*RBeqTvznAVWUa_ND9psr&sQn`3=sKhV620D~l+9tVAS+4w+hGa?@W}`P z2rh{_#tVcTz$(oEPzLE@8DIR84p=AQ>P)COQmOuL5n=SkF@>bW!{-{e*F`PP<6WMM zlj7)9Kkg%7kRDF2RkLHRnMomRERWF$0z8GA!abn|ZWK4?yJmT;Cg`@9G!x2 zrrKAwcqH&9GS7j;DuqqwoGM%D(uEQvfGmYP1Z8Fvmv*we%$WyLsyt+)$p{T;t?W7z zvc-9jQHJu&MK}6;zr)lT$bjaV^xkK-v(<$> zWN|%o%2cA3Wrj^Z6PPEz>0CHIMl~B}FzrMlEqh*7;h+QDilZZjT_Pu?dNq5k&6q~^ zgJh*px3fSk7Jzktj_bxX6{>n^=Vj+tg^?cKM9;w@f8{Ck4NN=AZt>PN`k40Ha>#EV^ou1Cd4#`EH5S(Y1 zsJVI&HqGim;0h8Gc+2Sn3QA|&q<-R_rba2&ihxX z-UObYNZT?X`Z{{jGTxZTmJ#W%nPX0|OpC0*^hWaT$e7(w~k5G>g zsRJuEBCxzHsH>6NiL?t1xkT)Zl=#Pbg+Qgy+VAJiIA&+-hegP_XL0YGb{NN+&$Q$^ zX6j%|8b*~Dpx zu>gw4RO%gJFwjaS1#yw|7VA8cK)wn#NT^Nt8+EZvr$PSzF~QuQ*WWN%L93=x)1o&x z=@<`LX9h)8oBFmXIe&Tl8LoCb^)Guf2r`0dG*1KIfJ?AI8BI_FK(ff;r|yeh)3d5i zmVj7R$O2qzG3RW^WPH>W9S10|^-kEB&ob)f=!GNq*pJX4+~s#qZszMzo4vxyI!^F; zhepzxU;k@qqHD~XhBQpW*7^TLy^IzBeF(3gg9bRlj%yE3PfxEZLL z1x>AqMkF$XeDI*h$Ij{ua(1f^vtrn=lJe8oTrzrRCS6ybVpQc%zl<98r;O?kI1QHn zbkuwK_>h(ov*<4@%0{(395(qxsw`rkm5Ni96z#N!k(23wl!_M*_4_XBK?0r|7exj} zm4IqX1ZpGet0#a;|CgP8{W)Rc1)S!M+GQlO3T0RYo4hCxfBXn&dTS4hfo=m3wuzjU zW$z_Y5z4UbZ>TY=gvJiFfc7@q%uC9{h?Or#e?;7LVsz8HJ(F&UbyI{L@}@4<8gIe* z0`t~E$BZ5T#s|ld-Q0k8tGpxEoB)Wb*-A~^fe&~788aA{1D~HbIB8tL1?+63+qMBN z+&6ZA5Z*lzjTdv^yR@Vz2wb1@4=>es@-)7}Q%yAsAbGGVt-*Sc9%DMJC%uhv!pwnQ)@jtgv zj1|#4~E;JQD!Spt@tWhl_5zhk(`uy z$cqtJX>m4L|J48$*7 ziMD7@&!4{EW|rH>z(dU_LcqKW{VcQD%G=;3G3+-vlBKg09JxtV`oi@Curr9TpV$rQ z>|Rv<{J}ZLg}thSLSFofx%K6FjYS8B^_3QiBit1>52Pg)hofmc-$j!gNt6*2p71aO z&JY~erI5<|yqwc?RxQRI?$&eK7%Hm1%_QD&ZadQkEMMkJkT1p!ko+BT)h}x|ys<(& zqbl0Y^+!|pJ%IUa&KOeBe*M)eXO?dadoWWHEWQL86GOI=6lsnfXC;ZzSwpK41yElK z<`&fqU-WI>0EcuBT!@4E2J2%y=DEYhVMKZs`}ng!WH?ZK{<&{PCjfEfI_rUL5(Nz7do3B|e zN9v-A3icUB$nW}TZKqb$xnpr1U$tl}qIdzE=!L9- zG_D%lhl2ckaf9vyX{nFx3_-mhCT=!{M=T7n;+CW8G0GjeiP?SN^!(wtq6C|Ll*kj; zSu0rkjlcF@)K68NVAnxQA%uvuL)Ln+u{An6_PLt(hcy5xINud^f^}RMpHFaxSKKe$ z(6vhjrS3ZoE#jZQwTKJl_9#2QrbsF5-4@s~OB8vK9iI8_RNnp81g30MD5 zwgle+;1O2hZNvK#ILhX3_w7DrNS5#KfY$W{vC{1e;5WYAKLpWW#PD#ND!@K9Xa|9} z&L@snY`4uKcD};`)9@(FdhCm><>RjQU}Dq|5cm$X#=q`ez0u^y;n3?;`QMm-Bt-m@ zEIm1Kr{H{HVBmEv00B5b6TT~`ifsyk*fgY%dXFf^SGEn0jeHGTiGJj-jyJ7Ypr96f zZN)IN${ee9M`K8es7{n~bmqm0=wvM^UE`UX`c=rRgYe;VW()E0_ISF3mTye43Gs!{ zt}tQm51SUK9t#SuY>rhm63$^OmW#+Q;>QuQG$E_22l_^oE|3u^`Gj8Ya^F6OYZM(r zw+P%;5bQ85Zg7X3;4!ZbI@)W{%#2)mB*>LHj4!y^;_B#jZ@r``20Mea%TADY@|Dt$fyIk(H1 z7i<0RPXYg>Qtdo9;6P+P_WL|r6%XtXvit*{(vvrIAKfLCCnE%+iXAX;$iVTrg^M%# z*m!|lE-@n`hSRZ{fr5&LG+~zNyD-m8g^83^pCGsR)AaOo?^|;dQ%+n*1p)uBZ!_qE zCiy>eT_R<_dKM>PwO3z)SE3|B_Qvdm^*ej=K;y3i@j@4|$D0ReF;khylvr{;NZ0q2 zO0&6QWXO|yHGerQRwPgjHs7ZQM~6Y~@ez28hCS+;S+2QSaVUv?jx-P*mi2vJaPlp$ zB)tI2?(sSY9RsB5sw}qKJsf*|u)IIRwB4Fmx)9oi@g5};71d5`I~?&oWJ!3(zE&lP?nP70d)b2ceT&=YbILr{c5642 zW&nxC43KD<;6G>$iW}A-{MAlY+FiX7Pj4gBWwM`m)Ih}YH5*K91zXCXRrjooRfp8@ z&t8OFRAC+~D|H`*Z|wWozJ_y>Oi6^M3v3{q9W25xgmd;~7<4wQ0LaIP@(W zUPXQ8Q#5n2F$9bEdY?}XziiHUKM<3%o*(m_9Et;-QVvJUC_H}`7-OgIse?!3abtS< zsq`D1^^QV=!6si|@^U_lS^1>oZ-Y^v$MCeLe5<(W^=>NO~3e zyROA+*P6LQ{`>p(4d%~xS0tjfF^aDT@1=io($Rpi%5;|p_%l(iUxI{j*XV8#v_Z8Ay+W%q|IwG~t zbsps4SjuPetj|dlQhK!NQaOo`ty7`9nti$HRyZt8nIJLRP__Vbx?MP3=Qlg@0pDUq zFl`_;+F)3VZ>oM}`v)ZcfwBm8$(tb7M}-V>ju+ zFkdcS;wLOaBui|?A3H|MyW0Vo+~eVb z)*RNu?h-A$X$!iFQpdf)M#mSU!*qGz&cIWoatk+_iGd8-6=z3+Gh{~lk(ui!pBA6h zW+!AOdDAYE8yzFmTKnI6RYWWYxU>h#R^vUJ97daE%{+r#I)_!t85z`%&@aojnZr$p z#*u7;n8wzm>FQ%e+`^17PltO2^)H?zwxE=-obLhN3V_ZfKa=#bH6{0``RW%++j4sB5V@g`fzl5jq zXE;}ic6YksqSQ;%$cCIpp~>@LVYa5;oB#X;>1yWo1%Sq-0caej>|R#hf_yKwfirXD zH?TfTDuX0aDCE84M{PE;Xl>4fV2mhG*7$z+fyI4`=c$J|3PIQX5$|Zw=CZc^UYaJQ zq7CKj%Solyq>DFgTHVN*ewSR2(*^b?k9~Krz@M21LZs!GV8hn=uUy2aP?1ftHy*Yq zuDp9=&mp_z$dR}wr3-j*AEfsA^UBEcqut!fFw#u1mtG@9?1r8#i^@Rsf{lfr&Q}k} zMxjCS#RdbO9^%&uli8Z9>&FD4It*Q1Q&#TiScGx#;i*JyVaj3R3+buX6p`Z+&*|DfrKkTi_2uR^b12G!RyfryZ>6^{5Pzs^ zaA2T%clJ57l9)An1+J$(9|2@J))(iu7KeCPX?vZo+debNccIhxBG3jNA%ftSKYG|L z_=7RkP+AiQfN2s8NRp_BM*lYTBOMiR!koSz>Pd88&qY1H=qLo?xk)D61ia>U&R#JFF=`V`GU*uXE=cB0d%Hgs)!B7F3LyLYx~rTfq1p23et*Ur2- z5l@+JqxYk|aS}e`_vZCyu;XM|YJMJgjV|f zuq?6oq^2%kFKg;DizR0ohpA;$%V94D3k0o~NJCke+)p`pA1SGMX_c{e7Rzx|NszIK zSoRZ8xxx^Ik6_3=_Pd> zN|4xkp%6Hz+tY_*Xm))!8hhjfD2%bBErOUhwl5tskI*R$FD>MDR^QP&*DTM}9Kv@m zYC?;a5klu3>!LA^#Mj%dk{}_u5nLGDfJBtJ?*re882i;hf^Kx{H{twfAR`NhlNj%D zIpM-;(4xM|g!q7UJVVkM6vxfY?cC*WP%C)ymw+SZv`H&3)=t}Q?Zu*6jl&WlZ;hk< zPU7~miIuy$mDyw$9@GMUIY!8(O8>Xtn`i9zwHVeUS;F({{8EDP)yzeTE5!=-^8pE) z*6Yc)XKbM)=m>W-DCT0wR==gsb75KYkTVm%oQ;P{aa?T2$|2^P*JGco7AOV-v8otn zwz396Xpoav^4^}~yfZ{%j(od+;OYCPYF+T~GO2Dsw9_QrNlP$c6~uUn@(jn>3$ffY zVyCMbLEXPOlg`4MhnCNN>HLaq*-}v5yRGw?$}sfX<1t5Cz5{P$S%>Ma=NMWVnk6>% zCUI5o%|*!>y99SVKUh?bPrr;0r zH!Jr;k1Q?O1FoMudBYbYUKxTWFL|%lPQqw)b+N4rM<>cgWfR5?B(i)!l`pXO@gJ&W z9{zeU^!I6^;oBMBuLwP4Kr9naWN>!ZgopENC_E_m58oQ^n zqT`+dWD%v5%`qUwRf~VDYPb@jbBVAi3~O`!`P1-L&eQg(zF>B2`xg&FQb)90j+0>o z{=w*AtuN({rm`V3?PjtV4|Kvq9JRTv#AWEbMH%lyvdAf|#SFgXch`WC-Yt>!HP3Ph zNL?`y+7ffSHbKP-ml|8?$|c3JV|Vd`6AV>_VgiOPEua_*NxaC1U=dW0w*hGBZi5^R zE6FOHS1N}cZNul+MtgCKi@=}{9@icMn(3QVpd z8k({(!^u5_%hAV9p5}cgm$r1s>%)~U#ttn>9r0P{hNV>YJj7 zY?MA6{Ds;I^6KV!bB6u|fKjIt@s6HNwiTn&sMw+St|l+h5qy z=*1NqBwh-EvK`T#G|j80Maqnd*q+^VF0mJq z**r9axm$cx1Gz;vA=2t1m7T}LqWy-9_}>52b)=L)%mOIF>uGPmFmgQeECI)C^WFD| z&IHGIuxHFTGaJFg2(u5+w!x1a2G#~bQ3!7l37%v^_L{8Ojd4qZ@zZ>`H0yh6kjI!x zi6x%j<0CWD$lUZtV+>~_Zm^>&dkppZ3T^hLFF#U2$D+T@)wR&%r^S^k%HZiDqrS3s z?9V#zXC+O$0mji{6v|f-g+uG{S6!hmq`c7O8M}*19Yqs7@n(tdf4$P^foCDZ9kdEb z?yD9G>D5CvKz;ZMW+&tWtbPu)NlW3e1TZAdyWyOd!d8@ z6-dtOm`jbaMXZ&4wC&3|cJ?xF_=WWQj*m_k%q&M$(tH+S~Jlj^itfP;7S;|e8Nw^&$07V&ISlly9h<^ ziGEq)Or!T(MKan?z{qj6_2^L=X*_Ry&`>AgSHcd(P%83sw^kf4m1K_xz( z$qWQSzceA293w^A()_NJ*+QB zU7|37?IY~h8M`Pd!OoX7{X7;($W4IG1T_<*%zZUhTa|s?U@(vXP=e%Oaer3&-=&DC*r`wVG2bLQQ_gvgDr(36 zKnOM9c3)DaJj>>Nqmim)=G~rT)xqJl$S-42PlWSj+U=h`fe}18xUzdlsTbl0jq68X z4!I8O#C>p*GO%1=gQ6eeG!DxNhLc<$U#d&pc>>Qj+<8;kh6tcE2~X9AuAU-6G5AO> zZiEKJ3?aC))W^Gf^q1RZXK9z?_$;Fue9ueBpX&+wtq@tmc*5}czRW5Lq{&dqMuk2R z|B-T{CNYwt3NRDL)^tJngGj`n-i{Pe0r9U6@v+L}7F{T&fc>mM9~gy3dY`r`1Y+Pl zl2u`UCcio9rDY_hr$>u;Y|iw*6ifs>9q>g##0vPnL4Nr4=S*~x6C_Ptx};d)H%&@_ zO;423oA{YY?*|6)g|iZn!G#b}f=MmvSy@Jw02(#6wRn)17ZXg31eX%B8@JA!KUlb& zhvA9#Ti=4fv`326Uc{V$7r?}H=~7=hgE@2?yOVI*&&SPoDC^j77h9@E&Z24(VVY`g z{bF>J{a6HsBdj)bZ!L}+N{2qk^2L~ik(GZhl}54)*q|18lw zvNrjV>2Kv}#kR;MRoP#ZxCaLlte}%fjhW3}!e40c_ap!B7%;8GSAFqRvyp_w`do>{ zv4}R{G@2qQ1=Z5BQIgopPLPssLA1H2N7`;@`D|L^r$vNDr}Qv@(d_|o1Dt8=3R27q zKs8%0B^JaR;}3%Eau<8s_c-)xD@S)HzkS>v9Z~HBbk7@6h9efC^??`;BW|g6QxsJ6 z0;hqewN5z)$=7MCeQ`Tj0v&|;TCy!1^_heB>&fC>@z}$Qsh{YFB`>29H6+w7PP@8#R_G|eKm29Pkh%nw)#2*4q=szLI}URt+o*x zhZ3$4vX-k-F(>zMnyD^+egR|7MOg1eG^qcTP1j(J3$wo=VnzzN|AH|sUDRCB!`>cp zf0A3@x&`;J1{>tAfZ81fmwm{Dq0wi3`78$e7$g{-TxZj{@RB`U5WmnNL|#{u)4$d@ z&iISU@JD+==Udd7({H=u`kBF{vQe-+J6?k2K2&EvzI)+=R7)~s^CScCH67iyjjT2M zK15&>S@-5!=+y~?f4yG$mZ)8}4LgMWcazm1i9rsZl+>`s7@1Y)b;1eQ+hVQ{LG&0 zX4YzRKl;3DuhS$1^`L#8178Y9{2tyVz{(ns04TlE^^=G9mxm*l7aEv2EOCBpW!y)+ z0Pky!=pGG-#`k1lB25}b+Lq};E;InGc|RHx-$xU`4yTUH8*qz>&DZoLn5jeoO&b9i z{-EGVoN7`9U#=REwoi#{Wrcx2G9nQAUX{-jrVL-r{~J4gX1x&GtNgy_aYgZPJ>9{b zfYm&s>}*X>qMjA=m~V@&)*U<2>kgT^r3djxTlPpF#CE*_w#ICDUQdwk*7Y-2hW=m* zR6_X(N)OaC;)wle_Rh={&#k5{TEKD*D>_=b!5NTOxa)Z;gC&ky$ft-2SOX*cN$+Pe zfyuC=R{b&Q|6}VbqoUm2ckwj95Uv_tZ$!wtU`@R)xB z^w^-tCz}HA^XUK723UN9xZvVU#CVmAx;`N1KscApZ<(T9m?24-D?>uF_eiJvBDYS@ zl!4Eql=Kp08+q+>uVpv~9<^i2JkWXl6<-7GT9+%X6Xjf}suN=$6IyzlIvD)Nq)zb= z>-;J0)sb29&TuE-`*%*j`MC%?+q!rKFBoqGaympT|I9-{PCu;u;lBBhyWc_aCP}iO zxVRr9Q{Y|gO_%v3qyJSW^rde7*{oyy7UD}bs z!){!eLZl@Ra8hz{&iM2ebU&XIrtca?OfCVry+P2mNYwZG$R#VtdhLY2EBMA<3k4E& z$vw`?@3+3&sz4L?mSLSt-6AU14EpHdfrTu&X0o%VV@CfH?(U_wi-_=5!E+4i7;1D9 zQ%pZR;c5C!f)Lsn)V)m_xi6^n+%IcbIM)Yo^=dL_c_Rv_7$`er1jnO{!1;VSW0p-2A9FwBTb_^lyCy{jqCGil(xQ5 zY|S~y{5ZXxMD%;>i`D&#W(G=M#)BX*P$2^Cz&=oQKUXiafGDl1Sx+^A!Q!O;MT}%G zl7gbXP|yq47Zq7mRpkZtn-j3CF19BD-HO_$(W2^BQUdYcL@gkA;R@uzBYMQ`>%b9V9gu<{*pZt6lIDqxx=qc5Q!Kk;C|fzS>PQ&@+RdY;t^HleJ!AV z4OjW?I3Mi6UBg#-3uW=xCG4r^57Cg#9W#+TMGXYU4)a_$5P`WB3*h3fJ}c{FVxm{Z<*W% za*2$JF~>jz3Tdh?TZ<&-vD!fRmd72S`X1_!#erFikU;8KZr@-9bR>xV>wo-N$=l?= zZ3B(qV3Trj#5$%hS}6OKjXs;|0f|I9SCiqZ--DyIPKTKj3wkiKCK9hdtjQ*4rE|z4 zUhODX>zLq9I<(QM#tTCb&se9lUIRFG6)pd8OhNan&rn$Y5*XK`w?{G-|D^(5P7OEa zPj)_oGuP9{`4tG|;G51epo*;j#m-ps8Vy|>`sgp#1#NU(TtjF68Kog>cpN47n^2(A zT7)6kIKS0tY<&E>Ll)4MXRVg@@7lOAwree$h@jS9*HnW0n-F$|uQ%s-@AW-gH41D*8u9v@%|7W3;!Fb$RumJKU zfAp>R2V`YZE`DScN#SwdVDzg8iP0cDqW}3AL0#ay57!^Vsiz;80w!5SwpW z-}-ShOLFmFPAOeYbLA!DkjKBe5}RA%HfJcACMXpLm41wf`ivH!T}>5pvCK-m{%x~n z!@RPXzGhF#x2~Z`^m&fvvtxLHO}m%dK11-=0c<%pTBf`T>YC;Z!o^?c&TeA(yWRzc z<*u9FkGp)9go3usUdcK&yJknd=?Dz``a~|DS~UIWh+0x1C|?-hD{|Yfa?Ok8C#cBk zA8pbn2=)BQYiRqUW|) zf{yX2ilpr(S!i{nK|`!0MMOBOq$DlKCZd&a+YK~r{osutyY8I5jZ{r=3i`QflYbG97a z)Q`OJ)@dw?X0l`!3i`dQq8XN9Vr}%&_OyidBb2aJ7F1f$7vea!&;bQOQeHqbTUOjB zoa$Hqq*An zQ*HF_Dz$-bbMASNB_-m4XZ9^fM$xZhbGy~rY!FB|^8+64FFb$6!?X@) zj;c1r2RcwDnT9`VPmHRPdZ*zA_oX~)g4zE@2s7)dY{s&UX7}f=MOHR@ynT}`c1|pX zWsjuHAmaX+Kk}A3quA*@uId~(9j`X@b?&4JxW~NR7P3rl4R%&9PW+u)`as1L0hrDc z55q>$M9uP7!rBi{05G$FrFyXkd6ruY`1coAr^DfaACIf zI7`Z23C+|8aN8KM8`-0x_W)hQKr)5&b}GanQj<6Hm1VnhY%ke_!F>(eak>PM&@aw9 zgGaL)(Ub~mL>*e|t9hA~WpNzgK4)sUI5?$(4ovSXUteI!*YKsk|Q|M^u_2Od3hKmMP<=y?BP90Emce8C5| zzt_AgO2k<{_mQ!Tu$+f9sTHI0Q%pCw2O~})opBp>G>s7M@U2*}xjU3Va+YKB3nnR$;ul`)_78n4+;@{u~L{oh`R>8bM2q1`S zr)v(tb!2mkuNz-*47(2&4ziYC1w>Q^_wy=5zxSns%dzX%qN4eBMq*+M!2_=xY9lVf z!O0vl{3lc*_A?IOR8Oq%?}GZMx#N^{uOcgdpa^jr*)w1k2(b{e%10IpKyXw@1oP9G!(ctp zg0B}V$m;#?K0t7{=BuVM>K^$rX1j9gm4d)`jR}B6n=CMNHRMWfa(9lBoW%J34c^Y; zU*`*uiN;5&{Mph_Ce(1fFqSbZbHl4Xyk|@I+_}DaPP}$ZyxniX?%O(OO)9A}(>1$N z>N9hHd+OoFKgUmU4_=B#;H3b}cN?I`R~5TfO$)2>=A;;&)7s!EY~r)Vy;DmNt8szV zXNStF=zQKk6dm>TYZ=P}N`o}J-}8CNrG}fK?Ew#qHTmJiziiB{;VzG2XCE&(O#AV! zd^1rA+5-c5&AbB;Fkv}0AJ$8!JNN;CDuSaiWZVWEf}(;#@S?eS9Tgxr2DG45GrwRK zt8HNKm}*nG$dCmGrnHdyY-;iZS6SnZ5Zm49~!s&qyD1FIwfDrApo zzl>xoe@V{_OKs!}723M}jNdMp%xmUUp#|RMcJA{h@Q;YedSDE=MF#a9nSMM~9Z_bR zgPS>)mT_UIHl)UEOZ@6FrgXR-YZnAVE$2^0z{uc}2r<6}vE(P4utjr5Q`fM7&THS% z)05;Ra++IXzZa4AtsQTWDKnrcKiiOQ(jyx;$JoIc=oNym~o!} zc_|pdRrC)?rq>azU3d2eGh0CF*(;ApZ$&Bb!gqKJgoD*%oPm*(4|E)jasD^YVYP=S zn2q`}1qio3OF5y;MhmTKG28T@vBxG0QQkHer#^I{YR58Uz6V|ax{ z(5=2Q2797^~h@D((=BZ!eM6baHA+y5fKR({O@| z^C3e#0Qf7Uqqr;wMZB`}Z$?(GPo>gd05-YY*PLVF5&-ZwAEsj8o<_q7yZyVrlE5q3 zkiX5h4+kkhgbV?27paRmNPwZiO2R$u?R?c4bbqGoQfyEXWSM>c{(VNpYOHmSk0h;C zXfC9psR264V-DrJz2%quOx(2GoWTzj<>2C~eZ<19z@aKq&*2f}Z=bxKXE^5@N|9dm z;ncu<$QgRTb9?|$5Fcv2aE^uD!7fS~AO1T}^UXLpyJQ_ZYn+%AClu@eO+C4Q`RLytzLs< zRBOBL&H5jbw66b(3^{CnNjDicR2lW;E+`^hV^l$&Vssp;u7&{_?8x3~BZ03E=C#mJ25(4%xwVQCXQ?Z1dv-b}h zG=zI3!iY->R8U2i(dhhYZMHmeU-|&UkN6If^aqYj-gxt4YgN^#^N~6G*fdu%*xV0h zQPEk9_mi?EoE$>cho}!2sEUXvH=~m+gbv`f*&HP44g!)W!V!TKy(oz@#F(^!3}mb$Esm=a^=h{GEria!HCUhKth( zYdvG!*vkESG|R-A;=;ma0OWcp{KAxBhEK>ts8c}kpaG_HPXsZQkZt_P6{PzV5JHHu z7trm+t0{G&WTX%tOZiy#kCiSTfPKM3>5su~i!+mIUBVcttq(d?O@Z{ayIiHC5yR{%qBB=wuwS$@<`wH|?Dbn59F*;RO zEb*hSfrlNBczFp|eNST!|9DO)Q?7&TuezWN2k4kX%g;URu6w`IgRYgMGfZza*?1H> zzdl4qdQ?DY@Mbc%(LuGNtXx};@V1f#n^~l|e5}|8R?oX_Yf0^4h~6r=q#!jkaT0qR zK0{2eF|q}a&KV!BKH400eEj%+^O^a2CQ46=*0Y^NW!n`A5{nmEBgg`qZc1un*}xd{52*M<5G?xM%Zp{`!g%{#YzaoB3 zF4tDI{!bqziB--Pjc9sQ+0~Uq<$wq}!M_-1$;e5|^9IdfZf&sdKm$|Y3APyg-P-y` zV1z(&YU;JVTJ~l~s658*0dQ4Tyf@q?&I6{tx^J-Ykd(;aN^^9&>uE}^IMSm(+_W|(ZZV4WpM&5Q!3}sYv6hn7>^zW z{(5zxglf=P*%n$4E^LeLK(GxXqb{vd1mU9c7-A$?i~ebd0sK1xmemwnN6S#2ZT56t zq&)&+vIbUx{immK(x||P;wXrV$ScnDSg_Lzgf!_WF!GCtP<&+gz(F8+z~^=h&O%+eey$iC@Q`V zO8rK8SpKxox2MiFc6_@vkjnL+y@CnW#CYg4S^N3v$HaUsaLFEL}Z}Qlp*B-eL7w-r&P>qghuJO@ma<9lK!{ zZNhoIcS{#l$IhvTm*6&e^Eu911gK{Hjo)>mi-9bOcR=>N0qoHYTe4|~u=kmz3%ZOB z`;Vifzm?J%w4rdQ8il9+lO?83j-x13AWm5ze@?aC=LCLrbT4eMq0|COuD+t`Nk*E> zTfQHUUIzX6xi1O2Cwy=qGgMpy*}G&9bBbBp@9BF7v8zlF;WmVr$;KAmS``WX&(%fj zZR%G68KpD)L4MjiAo379_`j2^i_kpn>y?Zg$dLUiNR;#zujAphKhdJi6dN!oGD$8b zMcJ(^oySDPfIGa$G_ZAA)M3uNW-oQQfg5i_3aB;89{NP^+;#R|3K4COyO8{7DW$Vu z?zPRvZp zdeGCu06B*R#CI?t4|f5*1RHrD8LEq{of_9Me1YO%=DXDxCeFwI{TyR!J?hUJGYWNh zld}}Kt=APmNWpBtr!C9qGzZ#V!Bee>))xv2C;*QP1DNx|+E8`1KyxKGD&>VSqofDd zztdAh@%L-%6G-W?&C>pg;+tHUe>d^$ITM=7w{&it7nF!&Hc8G-BwiYGPHKjWG!UM| zg9^xQng~Zhcr>-6Uvn@@(jA$5oR$?F3}ziXhydUqmS`4w;kRT6Fn6)?4l#D15=xoh zzCetd^$n}Qry2r6zOc*oo4-f?2dvLYiy^;Ec9zP}kaPKY;y~qI_C8QJY;-MCydzy$ z&P!~(Jd5-`@sNEl1oVY01P%clf8fRnA;Lm%B?#_mZ3N)<_<&SyCcmGS^YAoOY% zSpU28Wz@ZyL9)v6ZQ0Bfv)&7Ih)-!5hyp+t{(HFDLIgDD{fn+F%WFol@+&~lFZQ~$vH{fF1oR8?<270flc+Hn56+mU_fO~GbU}A6j&xQCxJL%?((cqcaP`FfShAgAK z%|q-NtDv?{;t01BVnq-YD~Kgx;$nz6L>8jO={Xmo6ojZJIe2WCZL0&N6!l)zvMcbp zX0L1a{>x@%?b8@Vq2PMAkwfGjHWq3b9u{VZ9N&8X12@1|ojqZqD4Izcalis%r2vG& z*O%hepcz^|``;nXQPd=N|CZ9kGmY2L{aFTy8Z-h)HU^{gVifOOA}( zY8%hEEAW0zWhvwmxd*L|);lcJ4nS|p460DV!ut4$|2quYUi^K;1oIN$qMj|JCYol5 z_+!PTQ&f6LKuj!mx-D0tw_TC|eY*D<)S0h?3+LIhXBZ0Nf)){gP7TT!z9PNH9(u9m zv=Fi3{!&{fJ}4q|`}x5fr9~C@@Vm0jr=eI(3Xg++6Zgf_pexbRCI;Dh519z#LX|b6o9ZTL$9!~dJW)^mUTC7z9T5q`=DZ>nGfG6ItxX6UJ_gARvIa@C4*A9p?W?1+ zVuTEZ&*tsZ@iDv?{C#kNul3y514aY%mRcu^oV2&kDp@$3rxpKqo??FIK&g(|u~hS= z87s1@KCzWRB_5HMjG+Jg85w^4&HiGvU+XRR4<%NbC_vJg0~HUVaEbeN0BIPUtUk?$ z`MK&3xm|cqsX4cq?YE@RK87?b@KAnbSse`e(u35%q`Hdxd}S3ou!V7%0LpZB zU|wwXBPN=MS?Nx$(|Mdpe>x0Ws}C=b<><-Iji>nUWvlY!(8R2UHkP&@L*;xKT{A%G zp9k8lM5!shdKwD+7K~+e8sa>RAg|vgZJCfIP%9J$tw(D<+l%h@xSsu*$@IDtyQow5sg|v z-{pDfskTLQB?=@osiY{LDF5F4yVk|+{@lcC|1^;i^njOJ8O=&cO*ui%|fsfd0mOuCajp7=xO4iK>cB$}hI=S?{wJN_N?KRvdrXUU} z76~C<%;!!I{pjrh0)&v*`>9?Z2-6!oE^)5lS*fFoXd%_BBS(pR5W!;ZSqbwu1w{O8 zIk;8g6CmM_1X)h>A{O%hp#Jwo>v$$Yk@wq>^;$3-;1fx=(GI9Rlb>F^C*?)}h{xQO zaPpdm*}vW~bZjpi=TW#!-nM;lM-*%pV2*!JXy#Q?)G;=8@VUijkfK^{Zf3}u2WI{w zv-2=%Btb#JLyz?gvh5Vh)1MB%zdFn7xbSstEaq8{-A9;bJz);D9ZuW={(_*TUb$^N zCi{BQhaY5xY@u_vN=itFi;@pS z?~E6pY3iSpnvR207QX!tuD4X$l`}O}CtVSv*QX=1^~B~~WNK|^M}%&>YxD9E7#j$5 zZcAGB`Zpeow+oU<`$coMx0rvoYCE&w#qCU*ksuPiQfvu5KVdB#=}SDID+f z75hc>2lD&9>RCkq{}Stt<&D*OKlt3FO*4X57_7>uhZN7$D4mg1PXoK^I<_O_yg?5t zcLs|i3g0OW(; zp#6~}8oI^FVR^!eCyh~*zm;1s?$Xow{4lT0UjfS^66Ewx66N-GZu{ffv{RJ3l0TiR z`B~jlQvlSh^ak64GAni7wv#A#E z{z`wG+j~Urk_Vt*V@;r=nURR`X$NbSMMwzi#S>!NDQR-Y<~k+IQ;sok_}KnNqekVD znxCs|4T9bx{EW(#OLNsF!@bVooDYUyEM};0LH)-CicrJ+(HJM@cp;Av&*?&4ZRfHC z^IO$;m%+Vh&J1ZEk@Ch39L`*RduPuRiOd6LL?qLq*gbG#b@0RJAhSn~a}B{7`-tj~ z#FUhb1pc`;cK?{9q@;_xp1!_~VZpH;Pz4^wV^Ys3-{f^-*#hh(ddDTj#d1nYp?>&0 zzWDy42-jh(%#H1s;9Jb0lT5eC7hb?v?l5)NhY?^y%C4zYqE|K!te=1*9X}S80W2$! z{aFGv%6WDx=RH9PG2$9`1MxwvAI|@klH&shd89Otj*bqMgj;dzc0_m>=?YN0NP=GI z@bZHe?jIKFgFjRofcyrR5jnZtRw@bK$#qCEh$jU_Eyum}q4=M2J%)rC%2^`=7PjZ@ zAsY~(bO)p39-OnxYfWt$%$$D92{f!(x|`Xyx2Rw32+F316D{lk)6BCR8Ncx=h|1VP zU-z5MtW!~O&|TRALb0;z?s5GRTUw;{yS0yvfc6Rp7q{#|u|&R9(RhFq zLdN7N?zAMfkNP(n)OO6TPJ)(mhnb>dUN65ljqjKdI#jRlBi#1DO?uRkPf<|^RRv>D z;hzlqa`WlCs%k_|V<|=xx4F?L*-@13ph#n=U6w(>auM1lM>@S)VlSrE^JN_r#R90K zUvClYzO~qY*j@4HWqcbCIwbOQI~sotnFQLteLC!UE2~pkp7uXucqQ;g6huzq+vDvV z?4C-45D~;VFJ+;DbYyADoS-~I{X(#$Ole|;snSM~QU!r6ls+R! z{NI42MXqg%W;KFb4|igqT`c{?|G{S!U&wLPP{{mLY*WJ+=(QRL5+jo3asp7Zcz@LG zQeu%&qN*h$+0QgG1Z0|7naM(LvDZ1l^o0$yP0WZyE!v&|9FDV(2;+430%Ldr z2A%_A34l{*mcrc7-Y}x}czEVBa_Sdqq2SCk`Gpc$@^e1U5FJ5Fz|0gZ>O;z3W1M6coG{a~! z$!g3+`c5|v`KJkX4~BB=PV2b|0RJVj^kKj3Vy1ZsB&604MhI9T@DOQv<~}BS3>HFE z5=>P7zU-KN>`=}{Ei4^bC1r8h&)A7TLx97ttehRAa1ac1a0bs}{SFYh#!|qBQ)fDd z8rQ&;H1J*kAO9A=+LQTZy5VH8u)WYU5|p#tsA2R%#ux~=TE|ub-IIEYe|aHp!~=I# zfvmZdb`&5bU+KrlW0)JiA%I?G<e>B8x$(jhZKo40GM{HQd@aYqy&&-_=kYhoRHSewWL$80^)$k;j5!yRQe#5yhw#NV6x916F9PBt$v zbh!YNtyld4RWgQ2vqzcui<-}wKpHWUo|l`81d@PYV0bXE!JK~_A}!=Gy${s#n6#o! zwHlX0LqiRkLfXb~paMjo25TU;t@fxS^s&<8AW{vu<_-`Nw0CPdWIjeDA3%SO=E$x> zkw8sk_exZF2#kf7cPyLpV0o&|&u5HcBL28BF*$i*vnw|q;1OzYdvh;!uaT{*>8~2J z%|fPvQSH4rp-JB!3)w#nu7>6baNt?Hpz1g!=dOM2_?NP*^gQXrN#Rl8HYCvF8_;TE zh{U;WF#3Ee%tmr+L&Ay?sCek&}mkNVkO}1vvT5Te!*iUm8)T%sT>4 z3`eA!=z?c43P4z7bLiA@>NI4<^l5&2+dpG7C?eU($erV=?Cw{RU95nI`6pD99K9+P zE~EP4*bphR4eGDFa1_&uubNp=`T3804t~}AdPo(vPi~EMIdJ&dFv#)Fm-~fhz0=Bj z8<@c3cxj}91@T4w1By)cbUusVvcDq07dcPtk5RIKC|~!$vsO@tr0r~KcYMia48UtS zB^Y*|tFfR3lQUq#d46I6V9om9A&=`$*IXP=U}BLhCjNVCx&6J8Y!S!wtRwT+3JVYa z%Uhw6{!m-C>ovfd;x%)FL2|zrXp`eNleu(c5E7jBL5Rl#z1@Q#s%5Kp*hhY|@Gl^n z0Y(ZU6c#3PV2sXmwOMinXFw_1a1^9zb>2B`#B&BV0vI6g55j7z*6$*RY>&C!jC~1@ zyjXm3vpCy6#zwjv5CxC_u&gc43@VOt@5|*SzpE4+S5(-Fwpy9!1YGyvjEoN!WPXA?}(Uw z62OZ~S6@L|%8yScDQvV?5HMSb51-ulsyRBz4U4HolqDM~1wyFCY5q&tlK@xcQlYm30(;g{~mvxcvOBiDIl+LEu|zP0y3)iQdY)b`7G| zZx$FHcpQ|+i^5MY&HL!>r%>64&{-akpPKgpiB;ay63}rJ@TwU({WwRSH2Q>@JIOfSg zE#~v95_*z*6e=8hsp8duht^i>1RsfJpDqwYi%^lwAGCl>R2qE~_avL4i}fCq8bVNX zYo4=7Z)}%aZOl;+4JHta#dkT2-X{)tAG2~25(uLkV`F0KCbsj}*Im{NVIZunbbRDX z1AOEy$6GNkb)1;GE=X6K^zhBpiLIBH@h&s-8>IYzV;$A5@MQ`Ry7OSzlW?78x~XRq#ivYRU- z9#vGmmTfjQz%^XhpiCSN`#Yc+)Ep(jLJeQo+1Xvttt$obGLRm9AMyeUIR$>mGg3|@ z5$Dl1{~{{BTp|mL*B6}m)H};Lar5k5tNu_a9x@RFiZ8QddiAzkJ2=c76D zli1)sN<3oQ#m%U&*EDh~ZY?7ZJ>RCu0?zsKb>E~j;XSOI%D%TbvAyv ziv9W3sg_skm*bt)Um3?Li{X}L0AvHWYI|3iO^!WIz(BV&Ik%Dq2vk`W(pz_)V$)wq zXobK1V&9$b&xQJ-8$X-88CxyN|GtU;Fj=y5%=qg#EY(5Bo$uHG{r$;I6O*|Ng9GmN z%{(lVi)1kIX=>Flo%AacN4Xo8EDPSo0d6*L%;JP_4bqpmdL!XTCRGf(INj=J+CYCcO1vRB`wD=k%z*V39q_+K;HThn7KM0*L z{f;8-?gR32g$RQh=2%~P&61ZiUf{-gC`v+w!Y7~ew>xB`8JLpmqo(-AZ}e_IL`4No z>f&&hqYbZQRxD(IY9FJW_P7(JOCgR?MvE8f7d-`}{aoS!Aqgvj(BaCWuYY)Shkv6X zvAf&CT-NUnKLfZO&>_VW>Q@H@=Q$_(C{`LMGwG{qKSX0Jbb>OYWUe5mXhh$t6xk*VUFxQo6xq{LA~7JrFrZIL z$m1vJovyBN#g()&2*WwHUe~HQ^g&`DMC44-+N~6+QGx^_^EhSMu7v2|NAwO*vIFdv zcVGH`9DR`WgxtZ~aPU#+X;-@>y}xa)8@-qAr(ss}?eGIRC| zy9?nzuDq2;v~oZl!@YFlXtyxv&^~f-`E&kZ^{@Nepmt-e==q2j3C3=0JGmSeI+Eo) zH&s@jP8yDWW4Lr!**Pj%wGFn+My0IA9dHh;K@gD8_-#QrH|_L^z{R9&=hIj7G_bEkK@riSQ| zw)w^KWD(ChSI@bYv`+bQn&H0+q|F)BPSM|;eBgCXCZhQ{Sn42M3VE0f@_-xV7Y&^= zWg#B&kO4gJ73jFLPjcmT-5EthblN)CR99=idGpr!`ojTre{~kv+^au#`2;@#D~PZ> zB+NN~^N}w_@^1%ks(6P3vgD8gTbPVrXpvk-PxHU5Xpa@0?_&3DO^%#cy5cWP`OX%>( zvP!907Q(0)^%bW6ta+w%wUgt@?iMM?7AJ%JTw_Uxm_pFBR-y8E{h*<6@Onu%r z^TF>Uu!{6ZHbA?FPjM%(qvrU5kN>Dc#HCNEG${li+FkF%$j%HT9970CEw2X&Wv=A9 zatHzB*s*wo262DpBXu=>c;&W$E!QY+n%CDCfkffB8=E2Pg*Bxg*U=c6=e1=m`k585 zJrtq6)Q}3o@tfx<=>4VvAki?lMBR%S1wVNqI&%Il)D*gV;_W_v9^ucnsA`Mz;+0C5 z8TWjnOToxCBnRUnc2Vcft{n26mS?&_P%YkCY(C@&WJLFX!7>WuJp8&SrWSFaLU_vv z!Teo0_!|LFcB$l2R>iOrPYhb`M;R z!L{wMVlyFa(@P~XOXnKLHEbUfz}Q$e-^ZqObtx>y#RkRU7~6SyVToZ2IZ9N{M9a#{ zzkVL6rea)Z_(IAFI#p>|8?hK#9HIN$N0r1rlUN}_I&j0xfso0rcptLdBXxn%jBwx^ z1_*u+Bf3|y%r9B1ZZ>I87sUjnuGW&5WPJgAj%FN{cjIPJ)2sCEwE#FQ7g+p4K;8UvzHoWzX@othqfzIlMm~gYgg<8NT=#2NQMTZ#79iFBLK~|&lCoJ3If60!`O9IwcaYV^ zptXW7qyS1xht>ckhdKc8H=KB{lb~XnsMR}ZizC}r)th0*HmOas0I_dvZZ;$3pt#uE zf_MJxzvATe3|{Stbl}|nKp=3ehV^K@B02Qq$HyRzT%-aN0xWxE89n>S#=>Ioxx({k zWxG(bYQM3kw>QtI!Kwb^D)@374Yk^45T<-Eecs;u2Rq5g->Wh~XyV1Emdq@qq7}w4 zsswk3Cz>D+VaeP<{&(gF&l+M9_Q5&LIihd$l7p!l%{K4)rZqU1$Ord=&gM1;??)_#_n)TUtnI_QZ~HCGF#=8X#MbY$C{ zcjfJ_&?a07m^gh}Gz#8^*&3Dy9nW_4uAX-rDH_=##MD7dH9kZ`rf96&^>HB5WJX511 z^Z^{#%Ju9R@3S76BcP^r(Zp>OaooUR>c7cdjA~@S{nvV{Jl#EC401C2+IA3Gjz03FeIsi7cX4D*4`(<96CkG21=7VBdmu+EP7 zFFzymy=oRVB9*wV72S|{D;Srj1CeWU`)y}y%ll%Ee^}M(brNWL#sDLa2#^V@BYbck z>;EN4I*4r6x6QM;TKWS5_`hHs1kLoLq#RT-wglprF7B(Ky?)IsRY>#=--l6@?i}ml z5RRFylxPWu-?=>OG^~FV|G55#vAm;Xb5sx}f_7+B(brU93d2{CiQeYl&l7M14<8II zEInX&?M*U1St*LZJs_aapGGZt00}~;Kn~Zj=Sgpe|E;&46DIVE+Li19&OToA&BmHu zPB7V@O)Yj34n3x*xu2nZsfmUCKe>wKZQ?ePkw0dX6zNLdgq%A2=6bDL9550`YF)fD zHV;wv55n6D{rd zCTU}rS$6YSW;Jb$GL%{kW8)lqXRUSM8*?)sazgUKM@s)Ds?|{$y6Yw#syQcydvUm* z+fjTZ@SA}rRH*=}uS3*K%u1Y%rbJu7_w_N6Re6H<&~Wctfq21H(pTRiaGBxwIL1X}-In{xC_U|I!S0+pKJC{lR~36;bz>p#=1E2bf3M$MPqD<@owqxGcu zE|1rNgA{;#@T@43Tyx?0W-5*WK8C0h4iZeUiMRyv4QD>wI9927{k$GuKU0IwtTg{= z(od=QjT-qnF>9_IY&ca7k&%&AWmsUcVnhUG*E@^X+!)+XHG`)3!Ik$pz~ff%f_}ux ztL7X?BJ36~GdJja+lF|WhoLL{r*TEkJu!M!WKwcM#=ed3kl%xVObbEm$5dYrgNoFz zMLBNxrx0ldFoc`Rrw#^)5r7}Sz#z0czjz6zdaUIS&ldQ7G@wtDzq$9&GcJE>@Ryvb z`3TwmA>X(+!p?6o@w{|QRRL<}N^KUR09$S796BcGAw@IdNbEh z(b6HkLTyncrx1|4^;ynJlBtI(qA0B+{3tN`&*%z;!WHXCI`X$n6bL6uMGFdiN47)y zvSV({X&DlSj2g{JPqN{yK(9CvhXQ$(Sn^cyB+f7UHS1TG0z+EFH;i@Si8F)L^l6C& zEv&7$UT(vjiL0+BWR4=rpR4-40c+}s$w}QzAW%UV&?mskME7AgPo&`U;y3lQU?AeY zXMH)If{lZ(Xt+Oh@^bXF@%S3X=h7)tTnxAWEoc3jPp_z}pL*E(}q`yvz3Rr6K zem>7_!RpT>D2x&AE97Unz6>Ms$A+xkoCQ@q4(cXwb!40ii2GI51gE4Lnwe?b2lW#t zC#Al~>QDujSR2d{VJt!X>peU0-XG=e%{LyeUWfiISo?D|1qsojHmXm^nvrWJae%_y{x4VL!92- zh4QQa;gNdUL<^UGOe<$Qm;16W*}sc|M}=fDr%bMz3^+Y9&k z00+VyGulKQvQ1Rn#@N=f`ew>EdDW7W|YLzQm(?HBEu@~$e5QuiP)h~%5_7{WY*F%_P=LSL7gGeoMBkzK%D4U%dgzF=XZrw^U#>R zC6hMN@OW;u0+3sJz@E_#c31{P0yF|70PcNs1Xf0V``plwT0l^+j0hmV)@JPzzPw~a z*vbPpB=;?(ygm}h-Bu$L1%WAM7>EdjZR)kQ+dvp2=88Y<)!9o?3LP>aWn>$N<1$qI zl(-C)C9&cWij*5O3NrN+lds6fGq``sclw6Om4ffV-@G$~r!8b&&JR8Bla zpY!~hn;i*9dD6$BKF?_Dl>GLt_8Rw$CrI=BRAVRmLC5ca3}0<$v;%~;Kjm-rX{I1; zR)F#FVGCgDsfYvu5scP55IbkMb0HP}PDDV=j?ZO7X>q8)tldAmEX#)0eyf%GIMCeCeky?B11XL%89Yak#Xa13l>i^Fl{wS}lL z?~|$12_4|XK$f~O>@~jk96)qQ zn~qO8^ZWR2|01OEV5fSk=E~syODBMM0~`n+frf$_5#Qc^f6i;{d(xZD9Z&?w4smhu zi#Ifsls4!}HyEFW2L@*W4KDE)sGTJR@rHxtelrss5Q_Q)Z{Doc4(x8D(B z*uFd1nKNu|!f+inj1_cl;k%XX3LSO;utO66=Mk1&1yI-#@iAy9rzCeqc~5~t&f2GD zd&nl>n;K4bkAo8!wp}9d6~n+~{Hl>k2#dJP z^-LEr8D(<|?;x@$5WA<(r3t9X1p5;G`f1sx_(X4|`8I$v&``T^WK&@);bU z7WFVDLCl5`rBVXr0e~GKA>=oH_~3^w7gY-*5DGz4ixuE>vH2_Zz*9riWM(ptH-{4)>->*Sw9@R2otGkBh6Z$eH;$>@5%8 zw=t@%X>9$=+-!nD`dqcHD5dZqq~LpY9>>Wd42RN3en%IFou_dqaE7?77%ZFs z0DA%piiyn~E_Oh^DY(Da$0$D^@Av@VSp#A1M&^YZM%(qTJmRwN;n-$t&NCNA!e@m~ zxTMSGq)C_eu-C=i-vyx(yvsLp^ebXjGtQBd$x&b&3cq;_es}#V2nCHSqS~g^{F!8P zIXhlDj3}ckI(o<+m)ok-JZA zBzl}~0g`M!pjA9f8i!5>el61;aOk$(3<%}(`^NKN+TCr>Q1Q8zzM!Ch-@GGGqJbMI z0cXV)Y^7GkJoCI|R)fBgXv7k@b^kn?(HUem$^#V-G5k6yTX-PkCQ&98W`jf!J)}ah0H9t!02r%^vU))aH`3+ z{|{Sl9TnC4hIRf!64D?gjnXCE4MW5G zY`^~Awa!`RPuJ4r%$~jXbH{akE>gu?rHZ2*XFTMi5X=i&XRWGgPPal)JV2d2gzR;g zM0-A26~MLK$3g(YWPh{Fvh(Ago^hzQ5lEw8X1c2MV%+4g>L{{Ki1+GaPSjU3*XZ%oNA8m`W)CdFdLm3wl>z!Y%0?-4yJ z`{UR-q$Rpxlj#2oQPUraeCby!As;UO8NVe7*h@gw9A znsi$BN>}R$mP6aS?8W>^mE8Qa*NbPgx<}Xcp5y4Ov<%|2-^YdJ+7e=kAkFc3-#t@S zm@P3~>%?65Y|3Y{zuf>0dya^7(H@Ks2^)9nU3dabpx8Qj{p&oOGLhM6I5|N$8=C)ho0mc z-4MVHT~k#UDZ9nm)&XEI6R1r@4l%cPKOk5YQQRJP_5%oM!HD$py@cZCO|b^7tfUTl zKk^5s$0u(0*nB|>&)=E86nFWR4qaBI`sN2UFzKw1aVsdiOtMPXHnpDrmH~I#3FSPZ zrpK=k#-Z9R;XPBqx*cz#yfAqqvltda+w*I^`T}WrCw73rIK00)y?r1O?aeQL&SwF= zWd66EGIEsQBnI48)kx=FGlSDwVM+C7ucSP}0i| zzG5#?lh}9f%tC(9+SLs<-0b$H_c%}2Sfq`RvUtl*L4ITT_8Vq{<%nGBo)hJ#E`#x6FIicFXIuBv zb#-+uXB+A%IxAT>)ulh;bka+>pz-e{9l3$vusH5*K#bC|tNbrs_P9f_JS!O|qrEJKk zSB|?qU{;wr*tm7@VeSzyIh*?(NQ$B_VQ$+YuU}f9Eu~)ulg#hPDYT63N@Mc}%gfhj zWc>O9uc=9=lWV&D@2)9rCTJ`eFVa`DCtEU%4)O^iMuZXK$h$fdZIAf2vo_HzM<>89$@xF#gB}nAls>4 zAY#9w1ND}k7aC}zaqMMo2Yh0Z&X?_V6dpJ-lAdSmS0E-wtJTN)L~nr-fhkchmF=Ne zynPyo8>%c3wdL51?C#IWJzRmze5=)6M61AL;MH4dzMEVaBgL~GU8-}}N+*9|QI;Hl z5U>>CNC_W52zZQG46~*a7Ow&-#{fOH%3_Z>Wtm%`zgbZp$@n4c;MI z^+_4uA!|(~h&|JB-oJ-L z(-=JQpdnT5fK)QdMv@>gt+TtkJZ1-Qh*N=i3G(s|f?aQl`pQ-cGjAWT)73;IVE|J_JRWY|6^E(#~|VqwM@{`Bo1P zPv5&_eij{IOa5*SqR><@7JIPw6VP&Wr7ZWt zObC%CCJ&;0P>|6a(pLj0cLziU2FC>rFbEe|*g-SF-wc}V?gqdBWskc0oZ&wupMK?s zOd*k>(s}pMTsiS{w3Ipa9R`~rxBXz&UpTBK!SjTgbYIP!BJD>%EujITx^;J%c~iB6 zv@I72h5R6#O7vZQpST##pi-_Wj{SqAaW^2JSb%1x&r~HsfXf&M+7JMLnw*_*7X~lps7TLG%Myxvb1gnybk^k?%6I4I)s-c zx}h{14Kdv(&x{(k%AD`<*v(vfp7PQ-{*h)B87~hai3-ciWb8>5A%i9sd!Q|10-Rz1 z-(B`rs$7x;hA|RK$}f;Dfk`fi6hq*VfS|8m<~R@Iom*_Z_CEDOG1&Xs)E_IS%8S9} zC!@i;^I8@wY5;?ae#yG(QZpj&ikF>-5x`CScoOSh=Kz~gxF zJY>8aQQ~M`r^~#=t@iJ7kXowkb_+&JD4#EKqW#2|pd}CtR&BpPSxy}K5%!ID1ItyW zj~C(AUrn`nlM4HSy8v`NGO>aN>_l$^5<9eK*+9BX4W-(h7Nh< zv*~ed(eI$JKg3{})_$O9NA(*Kne}448Hy}cn?nKbB@n`^#ikR1_E*n9pshXX?XtuB zvr9~zf1W9HT;v%=5jI~GB15uE_{0z097r}yEq=7s#HAa2hh3flZXz(cFwyXn?|~f| zWZDH114&_k5pn>P#Uw!!Bo>1LobByPDA1sQEcPo+`av{I$8r&j4_=#9Wzext}=j{wX$?75uFj(ZlDR8Mm}EE#t}1lq-`aZ}cma z@+T2cGZ5eMWaBXnt}Uj=&MqaZNQ=(|?gC*zO|n^@g<{>HKw>aE*19+c#>l&WZAuN& z_EaRJa3#S|&hqOx6x!_<&w;N=yb$f(S8h2(Fw)nNFUT+-2tV)c@B4#CrPQO`0()-s z-|2!w4(zEnB|7MC12tU+-WWk%2n}|EHTHAP20-8g$uMz7uG{^*%<(-(fi@S z-(cgf{m`BejF{v#G;RQL2$CQ6cP+^*SLvt=-Od3UH44vm+4-+z9*L?I9nv$=u$!0^ zO+0;dU)fu9&u-5Y;xiMq*u7N*LTQjlgHw`f(I;NsgMI&vKTEIw7`?Q&AELLz-Svjy z&+Y)te5%1tUo2?_EM*#y^1*baU}C}qy6#ZO2?QY1WJTZC??b>&ohniutS;V7#p;qb~N2~My}p{Yv6eo9Gl5@*|;rWXXszX zXZ+XZ-LEiL_3s+(x>4n%fLsy-(^Ql-Vvc%6@@2`!TJ!llx7s1%G%lGxpcD}RDBKk6 znTA+0oBrRdM&|*lrX2!(v^5Y*U67Pp^ecd@7b4+5u#_V=$aD^Vz77z1ksk@Y zer?DgiHc!dz)u9#`Hc!&gOF8A<5nyXV$Uy5?&V7A`|p*FlGGVQjJr9d_<477DmM@Oady{!eZ_ha8H?n(!_o7t>P}2%|dSQRt!Xyj19hofF zG{!ci%r8$__XqwyX&3L`++g&hh_&F@&m>7)O5}lsS+CJY6qfigD0`@65Isp*mTL&c z$V@r?8A|M?3krjDcE60ZjmfGVt!7AsproQ@8~|OB(UMsPpHSdT0LfH&^;Qhhdoi3C zWh>Q=nc#PND33`(PvpF*V}5f69e8z3khyBy-q&Dq0sbl2vfLQ@5$@BQtM7s}@0mQQ zf(+6at-LElNhQ=L&zbN}_zQAX&3M=N;K5^N3R8BF(9XuirQqv(rA1m&S9i<9!=qN; zu{T@`?13`=7Fkl6FZmg}sM_v*?76&&4B2A_tCSR9jQs>*UNs!j-GNODdydP+nR>9) zf&wNYy2Xw+!Q;8!LIJVk+wX&0`iDg)_|()*bskNDn9|74>;^IxpB0G{ z`1dyN1dsg$K09D76b1}!>(ZFcVijsJVf zXd~|$YMLaxl1HuODpsJ+d%;Z{(%)x`_UwcFMX=MsK%AqRa@GCwDDX;s<`)(Q!Sa2w zF-A^KE)y&$Er;`lRT6*yey)f0PB#d;=)24^k#BqaTY#(Xk*xk~W3DT}hk$)-nDpxw zhQtB2D)*#UUWwVM3ViZVbKvh7e(V9eyIbi6&y3*e;!FNc0eB2lwY_AoD^{f|GKXl%~0lj-$X zTxPRqGxU>O-xWs#*?I!JRm1mYaL5Z>(KAVI|1oi$edYACT>FaB2-bLsWd`}9G8E~P z|9%41XPD35fz7c$XM7q=$1K$rHK{f->uQ2DI=XS^-aW8cfnn3HdQ#vpxe|rcuIqY% z88KW%Y)ha3WU3d<4(@}B9!djH)J+Zq=}pjQ=)6!BYkIa4BdIv|VjfoRt=JLzTMcSo3XxWYd~@g$^n zx07B6Eh!~%@*C-Xw~&93ovmMt&h%ru)P7C|N(Wl5o9GMIi~{Dgc~E7@<$Ul1cLPX` zJC8`DM7!Hg)M+%A7h~23!yCOvq8Mkp6{SE`1`x9Vx@XfwA0!_E{Ah;HDC|Fn{K2 z-rg5vrIgvEITM)akwC)AHZ5vIve?%55+v&We62_zWoNE02WtP5P;`YX-1R?j-uNj{Xw)JdLS$p2Lk%|wT=?*Njr#= z+#r(shW23#Uyin0Pp1w^Dc?xJ?*H{!a@yi;(kKt}ZeKhm0P*xN^4`O2I9;X%#-k2A zUtse50V?`XkkaJ6(c+V|bo?iX*c8BUEvLV`L|ribz3p)c1kD;7m(SsxL;}YMYNkcc zXuFfXsXieEx3?A+OmdlnWp@V(X-Vj|R4xf=0T2dF^Z3{XD4ei391ei3exe-(=Gn5> zSC?Rv3_IMKqzu9)ZdX^N3jx+echrmsco{VVg+>$zKEZ|--!|!fK)W;##vSG0{sv!_ zhL?jrmUo)N50;axf)Tl9dTT2v;B%KPg9$FKL-M0?3n1GSW)0?j{zP>B_oEz%kX%lX zdDE`UT^|hEB?8EmJ?@j?dv*`Xv#zb%ZNYb)y9EE^bi(Q8P!7I|=-Qo0FvtPl77e5P zs%754I$e5S3BYu8U#2DmYPMS>FXz1@w{tzmV?rlF_3r$_2CVM-tL0uNpX=bA+jDk8 z(CF8Pvs!KQe6;Z@1k6YoK%_TXpTx2L#R9`Mph*;pd)YClJfR>b*ZFb^db!fU5^OqW z63Bl0qoZR1sxLr6SlfL?4gIt@YCrKo{tTC6!qdZXA?6vKv2eJ)N7^5r{^-DW?}AN+ z^DH^^sb1Npt&WznMiycr5=-T>EI>(MyTG~fh>06Q|6h7^dueeWcWnXLdK)flsy~zk znt%m$f^x7MG=z5uc^kT9?g_ zc!SDrC04PLR_#-SK;u3@Ip}2VKuAPr5h4zju!1*GBYJl7Ld*5r7^0nES;Y!!Au&C^ zl`!V$zX=RLB1LPF=5iC-G_whVxG<3N;*kzyso<=Y%_+@Ru7a71aFq%24GWJ3MIolg z-;a~(Wp!4W$dYmB-HbWfkhRh6e}3RXC}^Y&Py})qTSEPPGGWt!GW6{4O6tI|H4_d} z9>xh8fz1OA_(LTAqH}-kCft=W0=cTRILYtcebUH)zI$f;0I{X7|ArL(O~9&jw83-F zPxq6-xD~m97-2ZSEInlPu|o}O&&kZnmn(krkG7_&)zu2xpj!LR&RX^969$w$wUyjF zlnHd6e4y2__A*IzXr~sG5fl}j*ltY^fPO$rGl6g+qC;uM+Y}&!<<>f{v>7eJIv2TA z$)N-5Nv8`vJY;*9fc4SG%Xm>K6QP*-@>#zM_(GwYkCVQ2WGXs>3Ths}TFbWxTUuLt zPltco9&q(wEsCBg8Ty44L(gRxj3qkrF7PwS}Lz)YsPhY$mSX6!N znC#yImMm2%z^7&k1ZI`DRscLd8i=O|z$7Lk%6X~MpgB?sc}6PQ2bvI6L*h)m*JPy3 zcB`bn*PGG3nfHwMkbteg7MJfku@7 zR={Sv>C`ypSZrMzXEcKR@xNR<@a)?K z(LsmIpb}05JbM!Uwh%~aDRQI-kzia@GrbD3j&Lxjti28gfdWhVNu)*2uE;06Ei4d- zC#bvwi4A=&4ZZi|Cs$%NsHQ!Ew0MG^N(=yCflH-6>dl+8NsQawi)zMZevKc@NNxy@ zuQZS|*DI^5zcuJT4-RC`BI`S^BU}v0c$<-Nd}`pe<5lIcWA$}5WLAPwg=nwZe$Hp- zY)eA1d&U4LUiyhOE_GIu1g11Dh21-N(%1=WcfDo@Ycw`;0*fsSnVpF43tqi>)8oO$ z{Zqjsr`FJF{7OSp)7Co2vBE|#deu^a>L>X0ok`_l$fRD`DImAi(`r-O6JZF^&Ps zot0&u=-B}DAX-XbIE3O1{@~Nb>V_?7dtK$O%%r_t@9mC+ebLqJaez#j0 zuP-om+k+C$1Yowyefa90q=`}heI*MtH?6za?}pfSfFw-Ny#`dY+M ziNWu1Lir8X4DG7|$ z!J&MUaicDVvQsvXa3cwCnnpjuy>SEYZR)L;MaNmob$jBE1lVSrLPp6^zE&LesYr8T zt&?hSV%c18Lnl`Gu%F4zS)kqx15?%;K&x^+oA+j*@?-B)sj!w=P{Uu(FJRobDEL-S z|7%7R0vpsc6Lc9ED4rMqUGz=h;EeNGehr+N1o)I(KR~!7DFeffF1t5f`a)`=3D1KV z1GXrVIqjYg54y1m>^H5S2!{2)tpY6Q!6`1Zz!aGJRzS5Oa*Zgs&C~Uap@NR}Ns{qO zSSSJY?b*@(KT~bs1KD;A%(cGO^U^t?1mk;I6_x03M}z7lqby#lvNAKe045#VPU(PlLtbjp1fC5E&g$Kq{+sGWJ%Veq}B~DGGHI5 zajK~?LLPH-kGd`6cUP`FF6^h2GRX70k`D02O$7QjC;FRSBUi*{f}58eHZCHM`bMw` z%09WvkXm=e-BnUno?KtpB)}3!nwjZbTZ4W`P$#aq984Y04eQ5~4c&XFK6Pb2x+<)3 zTCp`aabFwCapjXEdkZ~2ggB~L^XD*if>#7s``WM+R+z{^ue|6leRqmNRyjuFDL8^1 za;M|pOnN5XQT|Xs_~%JJ^VP+HYW_Mf1>a&#bH33hyg<;O&?v)tkoz&Z?B^*ar9RXv z&T$W>^gyJ7ea@(YWCeu9xl+SS$nUboQg!99{R|CJKTqvU5%(zTRY39t0e5AQJ^K5OkWt)( z_cMa6w|O$D=NaO2FRn!H_3Y&X@6swD32(6+pQaY9;rvV`dT0Gm&HP^8?a53&bAdn* z;hfdZ70X;l@qz2vXV8naWIogs42*4_Y`-X7!DU?&`P#pMyTvmhpKUT%Q}*%~3+)4l z`Ru$o#``C@gsyq^&K51I8DSm+ClYQ#p&z>ex)PaQ;Ub7(?*MUyM?$q9Zcl*qq=^@d zuRL6`b6XD2jySy=+Ui35CGc9|uH47#--EKVDUG~KFAcVOLiJ0G(V!ya0*r=SBRl*0 zDzDVPsHEA0#nb|D_{+U|l_!A=My-LW3^^;-Cb{zN(a00sZ@i82=b|Y9xeixo-Fh_n@fUSiJHMXSa z*SbvJb?3?#Bh!a0OLR$yhWFVT@pg_9(}W#tWKP zw>s50bV%OZRhHS^D9-Ba{8MT_S>F+`du#A$Mkqm4MEk<|>y;P5CtP;CY68PgFn93V z%z+tiSv;J;_Y>POkk>3gT*!=!fT)?BtjdzM7DyN?hK4a9sodIrO0hG3F5KCuq(Tsh&^eH?@q;6pL0A*^yEo-~7;t49X&2!_7 zV+aF{0$FY#}vJ1F~lMXogAK6ugVExLKioIx_7pjXT}>6=@I zOBdiEXaj&wYv+5EfXB=rHIq<0Ct3L7l6ya=k_~WC?i=#@XlH-VHF6gBPV^TQjJz;w ze7#hRoR66K28jPQ>MQs?4YN6J;X8kJ6*Z?Mfq)eGk5KuL}8vuMa9X<6ZAsiL@gQDGl5TIh6FYOOIMR2~$P zO<^Mt{}N2RznRkWqu8cOBU!qghHA`2`9q{c<0FI3gW0M|8!@tEd-l^fcxAtUt*lff zI^#!mUHaEV9Q>C$=%2ZQrJjc38p$;(&dPTD5W*c>01)*8ce*4OIL|RX_+1Nv1$7(sEl4IZ|=OTWLCc8dwqFaX-(AaIOJx!#+_hu4q%nxfF?o9^Jvai zakqwiXVvSQ-Rw-luvv{fUVwgwv`WH|SHV%B$neezav?EfZxLbs53lKp)O%-6=h5kbjmGk8lvedsZP;bH61fVbvf|vD+*_% z?)T6>&TdQy;1(bcfJDYzc_}?{xWFZ}JcU-^{mXXZ@f8j0N#QB5p1|LXGz$h4AGVh8 zqBhuH;FEvoe-URwpIEDK+#A(PGOm$5D>(4*h)O*N*(3zhJ;lg}I-9o3#c4+gU4>%tuouOyJe zM3bDH?ArK2O+wEO(mo}{Us2DBmBVmZMRXiOlA{g}bWGpnJOy?hj4~6c{Dw_j0`#MV zhtp|}xgVqtRO>pPcJw#q`c9!LtV=C*K&}uu>N4Ub@f{ktRAu?bxdjd$(L-TJ60OQ+ zOl7>r?6YqwqY~G_9lf=`LzAfdHTNbfO!mon{2HvrNmozyzp-=*Kg`qBfS)|rT5Hmz zEH14rL6d5GjG7Y$LI!awTH;i1;|zJxy-Q&$Hg9xf6Q4{f=?KJIk7vv_F%iiJY>i$O z=nU;H*Afu8d-pD?+CI(+7;pks>iAOZwi^Sc0DmKpsPp|LuTa*_0Sn@}A^`c4Hbuitw@U5@b}gQ(J? z(64o!b+bmu=ggTsUEN*m`r^Q*?Zz9xDN^|+NH1RloL8e0J6ZN8V~7|is-zk0;KIR? zdd5A^(Ct( z->#2+LCtzzzCDy4c?|N8vO<}7T|sjKE1=v}uqcv{~E z=Igs%2Xw?NG(l_*hR0dDm&cdzbPMZ^mV@^l(USRn64$1na(2B~8l$GCFIrfMDA@`) zig&lp>#1jamw3~pw}6Tfj0^_gg>L-CH;xkYE3;_B$!^BXBJzPR#As)@z|`NMeS@0dlu4{ehvN#R}ro~6kH z#WVH-3=fDL^Zyc!-}w{LSSHT8f9b(r(l&*aQtGekKp|f=E35P@bK|}AN<_eFW&5j_ zCp0k+Zd~10_SZn{g|?;iliSWzSx6*dKC`{~xe#-#bV_#YB<1pAmA&i({mR^H0*AYj z$&uyd-N7KHN3O-l7}qZqmH}vikPPm8e09}P<9!RY9@VaA+c~)SG9V|#9su_*3xknQ zvxH^lc9v7@DkTQil(dV}!Jj~yUxlEg7v9M86T~?tJ$}MnbNK|Z?dsMEiQ(cxavoeT z#}pF0A;iP96AxSShl1m-k5wlcScGo{{RlkKX^oMvhe_r=_G#j`2W|L=Pf|cnG+j z3As32$jIC;MLw4c)UW z^6ePsxSRe@EjYU22{Uk$tse(a6rtk;$4Bmm?L7KKNsS()!jR)0Hn#1y0b*h7Wc|tB#zeOl#ALR-31>~UZ!zk# zX`od2yzDL5fLz#bn~!^V&s{tjSE)I= zklK~SdVhx(5k~Fx!|U?%>M=lQGu^w__2$DSZ|89hb7JZx*1ZzGAPrsiD@;GHP5ds7 zXVC#On<{hXyTXNr^%G|W;;#Fez>VQ)KZO)opYDA@N{CiGyiE0;g3;^rIE=&K&AYMt z>{suDKV^z%Hgfged?#{7Js3Mv~TsOkl^Pa zpM+pzR7f6)3@c6R9?aAHTI$g)?u9{AapWE~mpTn$;lz)poT^XeBP8^E;JAKh(ZN!{ zSrPU(Xez^Qh%H!3)E0Q3+f>n&4PAX1$QDdrz~Bt<7xd`CLglq*=wcLCNnewy2s%HQ zcPDCvvM@ES*;{ikATc>dsw|LtGUz=ACIqkiA90Yjzn%`t6m1M1VV(4MI4Qmw*Din! z4e5V2a<8^n;sVr}6r~e<4xl05-Uga9_OjCFcVoEAN0KgGn=vVMQlf7UgPns_8`s1v zdEP60kQ=pExBS8CuW}U^hr5nz23QYJ6P>v)kQm#MK=G_&vBS&g_Kr+*=_Qe`r3 zlpwVH*Iv;~;F9o2672p^Y&i12xtEqGl0@g&e>^WzoJD0x@3EAR2oqQz$MR5xZNq}^ zeoqVOf9Mf#;9X~KKb&qB6&@O}bn)_IS)`v0yBo>bS-Ue(R#2Fl>Ja@ZXOf<~Albsi z;yQcrRaKYg@0412?7+4;sA(bw$ouT?ykNb%)-+WcaAZ{L(x9R%87l%G6DMWUF_o0X z;d4anp6NYxxY*;tVzlp|Yw|fp8#TaAb;N7BPttgSj?niq{79;Ho|h+}t=dpwdd_+y6A>+$2qc;DT2Y~{)9MpBP=)NnpB zDm)Yc_<$mW8Z{*<@hcIlFO=p6NkO#Cyg=(*g$4ICDi-bOT(GiFaT6Qeli1~d=4~}z zW?mQqYe=Z-pWHt5nw;^XXoxwkUOPFZ9FCdXPW{s2wo5NdCv}~jmv=)xOwpmDMn#AL zE?{G#6^iDu5#TZlTl*EcQ$1K-wfi8?Xal^jSj9tK9JW3#(^BIKSfR8Bizig?WbQ7_ z*8Zns6Clv_6CJes)!!c8E!bxuK5zsq7HkL!O<0X`z!EpSI~^NHr&RYJ;t9d1pY^UJ z!^w+f%{|ygkgjbD#l+1{cHcT$LVCyrB(3OK`=`qWPYM>=(PG5Wy)VCcyh>R4MuSFv z!J>-A^3+r?!V0z>LhVm}NzQmb*lwO;Bu~-7OP&W0L0-uQ+rHPci(r`rBGEm7&z%72 zAs(mTwLTV?R00Xs( zr;B15X`lC2ac}06=8}?<0i+=C_#6XDH2GHQUV3VPP_7v#T)R63d@0p{7qH2h(b4Y$#Cb1xHGtp>Xl<)_h}2_Gg{s5%@?Qwg6^}QxYg-e) zTcYvf=@q7EPdto>15aijn)L+6f^2n2p@iHe(|@|@qcDe|H<7J^z)0Cu-sgZ%h&_=` zaOejK*DJ&*R*y+YCZ~1piyZhTkAlQom+UO=Cd=O~-?k-qo!&-UhMnHx#5Nu|dwbG< zP#^o=p#1yE6E%4RR9!Y;QvVQ%{OmKksd)loGZ&x{O|nnV9WxZ;TmuL1&&G!V@sl3i z{T@(!Lso2#3baQlr8E)vq&*s8i7P;hb%)SV;%aOsW+RY1m^oOgX`W_njq6Xcrr%9& z*Y~mL6076d|LGXRJnwYsFqlXum!cjY1PoYao)1fZ!H;Af{b?=u`H>RY!W1Fe*%YP# z1L1Qjt5ShUo!1mdFSw9j`O;R8SC4@u$JisT?&2x$gwNs9KiL*<68s3%EG$7jjeX!~ zCG!Zk#0@g6bcA36Z{6yB=%RZy)EunDA6Mdbz6U~0Yq)Kcw0#K{Y?SJ-(Bfw~ zHTaokTdloKz`8hP1(eqDhnt%JdyH|T&@>d#ZK55B^j|xr@@uE@l&en0`=hP}&Tnmd zip#m7aO?=fS9bWHl?M7?N8O_soX{y-zGNl(@x4b&r?C_X01zgH=S$rNhkBV!W%Qdi zopI?LTV1P?cqxT1)a77UL;|meyA4BY2J+*h++uh)^Yl<(@O2yk%`0fBJk54Li=`fP?-+MV)ZV`papus*FSTJhOA zsKe(+ug9k;PkQ+LBBK6v9p&HI6E_gBgHk@LAAU&>GZl3gw0f$>$F z-#VU1BY*Ht7B!U5pUhNtM33x*O-~y^1DhiP%*OT6jmHbZrDugsaO5l7}e0}y$wY1Lg#&5HqWVl_2nM)Q?8_Zhcp}OHT?^L>n41yEhcR51lb_SBT+3fF%gD@f7 z4+5xKH*6va^3@@EpUdD~mvkSLF8ikXE;Hy*S^z6JY^ghOlA{i+r%Zq6stdN-fvr&G z6wuv+y+*xy!otKujMpFZ5XELx9YN!Mr&jtdfU~aJR6~Sy>AL#o{uR%v2g=%b>qS42 zY4N|8C85&OG7cIxvLHOIuA7WrT%eL+<54lYo2FErj&5upY{kPsvxX8Hh+QhH{n3t2 zCnmbbO_J%L8k8=>!hB9O*D;h%e#r*3%HFgp&72@Sa zxw4nCiU`=+8d?aOLhbwyXZ`1U)4}ey0k(qi-Ni{WT4z5Vq@eh28~lqPuvXT*tM&oY ztWU6S!L4a|PkQDCB4=6jp*?nqtxpWS%4Xh6vD^W*%xA2b7OL01C;3oWBQJZUpGI!S<*%M(QdIs9bO%moEi+l>ubody+0r z{r>sYU(pi^yu{2jlY|$X@6di>o|)o1FSeA3H3wdYXHbShNMHhp!=l=psw~_%?LUPl zoCDd<8@zs^?7n}X40Sru*COrcIe$8yfabP5PWQDyiffvWQu9V2zJC%`{~H4kl?p(B z@TW9+aclG~VG;`KrF9k_H5NGm^A?T1wppNp*=0COCaVH4BFGQ22VQ+I=D3B0si1Ho zL-#q%Isou?2T8xzm5^s0>Mx_<+{Axj2}jbFQ^PcWIjz$BQTOvN%aef@%dzO%!(bU}#P(W@Rf+D|d@wo2ICm0^mX zXHASH0mzDABajcSI3F#x-2_SYa&y_pDJZh?^6;RUk0gMWwQO=QICa zG6P-JgqUz16@NwLEM z%P7l{o&z(fKx$v|OQ@FoP#jscF$%RQshdkLeAd^&1R(5Zv2_7Iz z>~cG^QA=_k``>Zlfbcr)#TfA=TIAq~DG$gbLA*t|1Tu8D4}{+lQ#m}Rovwvz8(@0x zHcYvE(SaG+{}h}wLgqEn%`X`3YMjui zNng>mp-DTljtZGt*)@I;3yai85T2tgn@3Ad9PNoMk!zzL#>(omFe>X~oWFh0Zs-f@ zQ2_pqpoG&3eUHm2_|;#Q!!Zn|!G4fgKNW=WjV*qi?RLxn8Z8jk^7qXmNeGX<9fcU0 zhh@Huv&g|lU390z--wiU@$s28=mJ~r0Wr=&|5L-h5K1df3 z`~9m}bAcqDPvk4atdyT}Xckk0IXA+LoB`i%S$<-1BsDWk4O7`4ls@}>dY2+`IY4pw?-Thd68SR;ghOHL62e%myMJ=G#i#3vrY;Mp=q$4k-`*C zuTiSp0*v-(%+IuqN!dmbH>EqGxxC(1aU6}u{uOH6NBuAGqG#vF zg&GYtdDlsb8FwA{7?b5Ae(qq~Q@C0M0>r&81Z}6n0F-0=$`|NK%g?R*qd3^uKs=U@ z16YqTMUtY=bA#-Vg_2exSwlm52qh7gCZMpicm+7(r6B!?;Fj>fHv8zbpK&Ee%rwY1 z+!CgurOjVhVRi49KuR8r4O$$@3%5JI<-T-tug|9Ad-a=5z-Hoixz>s3UvxLRsi`Rs zV0Vi}k8mOhirCOVxSy^QI3)Y7@p=26JYmfoZzwubR;nH!?f+SR0=p%)#(MBV$os@T z>ZxyM#mir@@$rbQOO{PN4dTTQ&vvV3fx2`}vHJd_4xXEnG52vD`l2Og!8xwJ2W)J<9FpPM%rhk4l>FuiW^&3)IWJ14l3Tguz5XAdgWd} z=a!L@VkOT!4!gqM66UH{0K^R9)Z2)!2o=+VvElj?H6i;a#*A1UW=1|SgkX4hZMK8& z_v0iq6^na({zqehZo5#A^SqIcKrBxa$E81uSf$Gk6rdwq3k2_4_zHIZL~ppb(LQQl zC0RWT1i zFI>?_ZTh1=mAtnUM>PjwTix~)`z?0@Vq{7tJ?c%`Nilx|V}}*@f!O==ZK;8kIbIO1 zkx00rxO)pwJ@n|sjA*-0d{1gk)nM|Rcz!_PmR=YLM!n2BhQAVKrxZ^38;Zpa#AT*c z?w9?hISa+i510x;>b|;_slh0Y8@6qnnzp?yD7VG%>J7MTUzmcWTAod%o+a_Zmv?Fl z8s9PpsHBODSqD+yFR@KaNy)r7U1r{@#T2*8N(11;{|CcmmR`-apy^@>$=j!&Rf-oe zLK{&wB;w?}XsWwjPC@eB>!v9qzF$uJsKG1mnL}rPTnT~de0=YvrM}LYM}d!n82=+y zT(=mxxN~NoK&~H}l?nIJ2mZF7e>65~kj=#eviCC-1D~uKm*Jf13HHUMUCQ2(afnku^GEyI&zAJX=QLXVPBllfstgo>x6m<#N&yZM#4?` zuc|>@pH43ALCN~QOuwW?bdWLoLWbPNIoa>t$EjKTh`miVXe538B`a8} z%p}N)vn3>vpWhqt1IKCIYbNy6gNdI^$3#VY;S90#*lB&awk90xlhY;A)H!YG; zWljiRA^>fkc4fiQcv`#tH9H^sOT7VC9(>H2dCsPNZGVUuH*0-S>*0-@Zh0be7E%~} z_&n$(Zg3l1-Nv6Jyj|uuW#NmOCPbf*y{~C64c5sJgTv&nA%E@$7)Xy1zV`l;?geqJ zwi_9$#|TIP!rFTmEr@&is^wop5%4gll%$*zM(YO?^j^sp%S+iFf(W_lO_+a9ujDVq5v0)Ts^W(SZJ1lVMN#bpdq?{cqc<{#G zFezXk>{RAKVnEgW`!OH3)P1iz9K5-b;lP&C+=LfROwN{%{Ss%6vg8x83z=zrSIx{e zQ5=)Qz4Mz{CJ80^s}VbNB|$ooDPPPxEk5 zbI$nkW%4m+@R(JFC0dCN1KGFa!Hyou4{hX^Gd`5n8(dYklP9u&YvfH=x6oQw@*-4abkm6%z0*KN#J`PP#i zN?wwmtTG?FE8=eJ_n$`fOTs0EWMAu({xiFsFuXVCdQsu|o0YD)M3^$GD~@x|C*aP) z^Zm+~Z%RGaYsPAcWrC8Dy?ZuOBvM>1(pI%~a4Kx@57*!XG>Zv8L~XM`b;qqEX$y{A z!SZ6!h?|^c^&Df$UV0~~rk@GH8h*@5>5$BqkpB$0Up=Rhw#0nCFDl4>-;`WO7cO6+ z!d|*g#i1G|P?Dfu_4%^5=6TSy{Q$*6&QkxJdL=wLbn`(H3cthyJ+_g6t@|{e;ePNK zLZ)}1DK6vd*X#Zn4_K$>dz+K#^y$MHFXm6A=${5;DIb2v$@PC+<@TIJt{FXB3jKZ< z`CG$JZb~vZxL-dV&$z|+qOLMlrfN+1Hj!?>T#?6Ki=ruFZnh#T>8SdZ^XT7S$>RTf z(#%Jn+gXTV#vTr&{p`NipG}FGv}-{XYC@%88o-Y+X9Uuk9%}mn^2?o3erp%Y5-KlB zp|DN9X2P3Zn6uRM7?K!`LbC)dUZ(x5@381L$)--uKgEAylGVo^^b89fT*M{pE^Rj# zAH?C+P8&=vak36I5#Y?O_akiOS(iJQ8CtE)F{B(?!)bY0_C!{`ly+f@KztX=Pn*WK z{M`)S{>LDK_@}ylJeb7Nz@%ms^Z-|=exp4*KM+CCI%dasMdzC2e6cqaM(T(l`ud?J z)XR+QZJ-5qWz6U3sGkI_!Y7|TedW8?YUhjh8hj-KW5UM32SdwM26o#|Uv^^;p^`N` z9xlR!s1W{2O8o`&MTrP2=C;5NhVo?{j@pXn>l2!yD4jKYqu8_K#1|D|w=Ahb=WC%@7t|HLE}=jo^WYGjaZO4^6GJ-&Bp>_)#_LbP##ZXD+lmosAU zQ2;jMwIBhaQMx=L3gomkNd~MK`UHjBUxX5BsH>YPlC^vChMHgubhYGew^GOv_N}^; z{;G42&wpUht6R;LwDi1W!iGKPf$HmN6x8CL}>roNMjINr4 z;cV;V4~gGrH;re@I(MdP9#__!heaoEgDr)Y`!TrFAl(wu zAxn2REX~sVEGkcR#X zh}1{TK=w=*2x$ON%>gBYUWdkVb9>zvA`yJUs?p3$BBpqs^ByEmKo@;x>jcWQUhdt| zd$*}604ZV?fCx%~8grG!&+j{szC{29rwp3EL=&iGAyze=P4xPUdH_Yz2}l#Z0#iR1 zqDOsyIK^WC^wbP6PeeUu{hUt;8#O@rQ_4pr@s`4r`f@A zpkVW?Fp^Euuf8rQF_J~<+yo}Vq`*W3!+R93f^u@qqUAz$-S8KI{~IaAZhS104C=yD zc7@xTy~}S@b*9#6l6r=;=fz4NW@dKRz(19mE9nfLhR0(v7Be)5I|+fOLzIE&jliAg zVo)2y(I5KCNKe0f!v%*F@(gd-z=i(VIkb;K@82NQO_iOGH1M9-_OH6y=ORLXhG##s zpvHTbzHKvv&>5hEJTVzupNHKkxwzB;Kq$}mDr5ZsNbSyh>}T!g{M(Ygn7l~oei8%5 zFe8YvG=Sj^ub7s(D(%Pf?nZxw(JMXYDY1uM+^&r)j+LQ*}T_kB{TR875*=x_;b`JouYue?xYK2d(IDfen zW<{)gRpY`NcnP=mOhOs9LwtD{(tH$Ae#v+{%8 zuPbq%m-X}a%x)#spk?}gH&5k$6uSz#2q%?P?c9~qU0eIX|KykI3Z(A&Mpe{O_aChI zRQH-L^L41yKhPAh3Z1X?8JB$I4y8lUMWP!qPdus48Doi9y;))VFy;Vbna*-gPZXd6 zRekwn^=#!2YK2$)on?lb1Z{@CaCH)gA!>g@+S-hkU-d{v$fxe5$PS#+69Ik@h}_SW zIJp4T%jY}VTT?{}xQAF~+uYe3_Oi!CrR6?d0h0^b0>iCmQFV@&Ge5{9Bl3J(A?m5Q z;X&kDr|zdV<8!+{=Ex(ESoc3Kp$+Ks=$!fo?KQ#{Gy!r5uue&ByPtNMI)@Z-F@J|) zI64){>!!FLsaSL!AI-a=CBeVl!pOrJj3KQ7^d$wL*Drv?^ML&Mwz>PpXak5YJwM*{ zk-jLfnYB6eyfL*CCFw5)0)>ky_#H#wmlvSU3CwTLn36Vlstc?l2qC%J8@+7E+#z7eSxY3 zX-mJ3aOU$%MZ-q)9p-Bb*IewGYY3rA)@Kvf1Ug-Ti8~IxQSOAb)@}j+)mTYb2&M$M zs?Uy?XXM1hqqE`kEG*;*V(=s-(ocN%g^Jm7v9U}DNvq>l-Yx{ z{zJSt%b)J>o-aQYSjNNw*i!nt9c92kYvpen_Y)hZEOJ;{I+1g-ux|Hf{Ukp1GRI|W z)tzdG&`S0D=$rB}{j=cqop@P)X0_yk_uq*gI3 zrY$9|YGpWSRpd5Q#iuu_n-Z!QY`G*XDvGJDvU^!DF=$VJJ?7+?ZgYXcj_DgMM2a}= zB&R|$8JB81+vm*O=A>Xgs`V2`Yks-$zjW5VyDE4{nK5XrYLCvl&J0wgEgjg}j1D;N zS`R6|6jzOm%BuP5wY|HHSCQ)e<01oz*Zouhae+(2+{~j?r5Bw6e*p=_bjeAA_HTY>i{V2KAS0JS=8-y3zNMmSt-8<{ zW06x(TWYeYkjvr~yCpzFLsY*E)||y7WA5cd9y!0%hK`(`d^x9|bn5Jy>w!D3!a#0P z3oMMht*$gBn)#)nqO#M+?_%FMe0#cjvyIU4U{9cN7#~5Swoq`_zM`rL@?N21TmLVE zU)vb-8MNN$a;$c9vZy?s!4-4DL69C005^}Ju^-jZz#xl;`T)Z5dAPn1hl&RMx&eA` zyr$snvVio<9Y2*FFw*R^Zg{mf?kltm2uROPqyVt~aGBC=4QIeX#?ODdVC)+i01hI< zMC<+UP)kFj#ma}}b_!!b-%w>#r#bm{^JcyXS1)$WQ#zehEL~6l{O01rr;hiAd4%Oz zU%rUzK<8_6x*k{0jsTTv8i??h1Br%G^%_0Bj>QIx1l+tb{SXXN=(?mqCr>|qFIyLo zL|mJW=u)AFeQ_@l6Ir@M#~O$OXL?h7gcRFH)k5=M#U5V)u;XzXv$lnEenSfsr$g#M z`ym;$eGhDB-<$QcW7O0E@pW@x=k|pMn6BMxCKn00TfdaskR0%-1fo>CUz2fDK@PmD zCg{5rfG_*f85Zo74!|rzw?N1OHu+m{!~nq1C?tFkpHrEyX>x)nicp=bo%u$bt^|FL z^iU_puO_kumc3X*Zr&IQmTg-eW9{zmKQ?Tw3ht#EJI|`B-(u zDL&F2j$PRcWPiHlO)TSa+PiD}8MQ6R)awGULU;RVLVjJ)PMDq_(G5Oq>|W9?8P#JAuzkEjn1{V2a{`}0n^|`N7~#XM1r3aC2;wYaTAy<1 zZ*i**PNB0Hf_0b`0&6ZXq1Oz6H+pE;#XqL_ad}cmKfm$3w`#>|9ITb=yP=zpRxJs9 zWxzo+10M7*3{{>pE(G`B{eJ$_?bwG@QBWt1bx^PU0P}V8m0v0!V&%2uO4l} z>I(qPGVL(=BRCtnQxpXWtvgtMU{f-9r_&1-9pTGg172Ryq#9XolGUVLZg7y`s#@Uv zqV4LcB(}SM>_*M4Z}0CsfR9u|DlK%i{`%4!V{YG%DIt10fJFVo;V5%w^|@ESa2jlj zO;!DM!htww+jLxrfgq~G34~MQwMqM#)IV9^zTe%UKOs3a@q3oVpgQL2`fq(`}6 zvqv^QDzPi-<9sl6B(8JCiWjA0{}G0>8ZNb63vKgolGaTpnH{G#6Zs@X}D#cP_cQ!dIhmg~{8MjU=jRWvxmOHo2+Ves!8(bdyw3$p6)0 zQ@&KEqxvVf{4;~PiDXy#PuDi~F|~O8cKgm_Yv7~yCNDjskMin$dLu@gnezfz4f*R& zZjDX3kv%if)-&)Mo`HSLoDJVJgFr^Cm#u^Wrl;WD!A(+zOK*^0uoYa5kgG}CL;I{j zSznwI-+VHMQ=lNC2HKYnWIrhOh4>1Cwod3m)V|=0hStuJ91*A%pJf?sGscMconI6h za6v4rm-qU2eX&t-*|#M?k9HT6tS>qsSQDOLGA!VB8WKsYa37TxjBO~RU64JouLzi2X|LCP zkQaDqmqftR{CuNFPvvXoze>rD6tZ4SBUED}QL?2|b8=LN|7!Rz5#pk^G2dxgq5Fq9 zg#>Ix6Y~LIG&6V1@L&EKeFSbG7ZA@5mI>L3gddP%(+i@gv1{E;))i;x6ct$dj8Jp_ zw8EE~<(DTFjHKGW;YW&SsJl6aEM z$#$+*geVKxw8UlNiGD#lBnbe8l7$=$dwTQJZ_x!{8A`RDwY~G>N6II>>8~3kA-p$r z1ABX*b7rHEW05n-xLpQa7s!uwd&oi2Vifw>s{Knt+@-YQNI{`HZ#;SggGnm;0dDUzYpsG!cQ4`xTTImQCgj} z`uJ5|<&Wf8iK6{w@o2U>Ej-}rrS>|5(XSWAR)+i|+>L1!FwB>9F^=@0|NhK7=mb2D~4 z;1PkiIqC9qqzS>0_G$MLG%1jWPS) z+Jl%)Dyh}p2GkC>HY_ZOj5c3SADsrOr1FYtu8NnAe}U;W5T5cX7BmzXl~T){6*KT? z+Wr@Jf;NJUO-olcq@-sdar*hR^63{kGAb~VJc89pu_=kqH4);ZJRM0!I6nT&|G}2;* zL*ZJGo`V$m+qZt#EKjbnslp~~x421WTuPkQYYnbaJ%qNJ5W30J+A{z6q^}zMuH3VN zyETPQS^H63L^8<6{xIV?PTJ3{H{V66=aS5q&kFS+HqX|tr~Jzo(ZnL<6av&OthfIj z7vfA$b=05o`SNsAdGOJN&s$8$-T3XbF%o!5wh`&7Gq*SAe;?ofNjO(mqt>6q%(q!$ z=gs^YX2RR~VVlJ`pelwi)JWaBvWe{fj|wk~lZCpq(Gh3-za%GmBsN zZWW0=$CUOy{>w3Vy@}jLp1wP^^IPst2~#fQ+gi0K$=X|&<*xvD`2w$`?v!ZR!2ohr z<+qOeSz|-&V|3r0nBLzPq01pkUvJs_0m>7gEb9s;kxgDmzO_j%-bKMou0hVL7!4kS zC5C3glY^%0;zntpXCw4A*J}Zh+lw(=1omhfg5WgY%Y+>FnAUUre@*{>j9u_Sib3TH zL81!A)?>V`{P4m8MkJHJ+i`w6hAH6vT9{n&*4YQnIyZ-Sp0A37K;z~8tS62*);qt3 z#jv=b@k(?Enw{ij0&DNUdLeQEiAOl>)BR*!H*gB;(1=iQ0|<^I!u7ilKi__y2a9X8 zp*~=J%=1jTYg_kjkE{DICwx7w`QOYy_X~jYZT=A;>pW6QA;-J)g2X<&ln6bS!90uQ z1MI^Mq{6h8gdzUU+nu>lJ9ul#)uUdKxia_rYZgMcLCAg_#=ZKPB0)tfZ~Zs=5spN< zn9%MOnd5-nOq1^|b#`YxQb*its?iy{DVf7ZhjjH$g~`tWPW@PwHz~cSqotRFALnH4 zT6-2hbD1yWFN{IlV7x-LE#f)D4afRnD zr90@_b2bN7XAW;Fm?GgFQ^*YP!zO+ZvTw=OytOvy`7@UX%_~Y?9l%!9|b?-8)*jDxsCw$^(2A@~^d}<0C zp}a{1%(d?x`XVE7AeJQHq*0tPHk<@lM4Hc z<16y3d(h4q8z&*z>A9+5!4Be(h%CN@1_3VF<*OxgV2`WWWoE8Jj;{Qu($uECOuHK@ zWjWh7Xi+nAQlKgQXtf53*6L;vvC_#g*=}RqtsGU|OunHTp;wapnl=(~sjAtbD@nB| zRm}D$1c_Q(Jee;=dV0q`L?zfN&v7-XRy>c7oUQcWE^a<7$Q=(taerQk7K-e5Az1*d zlseaDi>&=J_?^SlU=`K0zCUVWZ?R3LlcrKB= z;uN#806uFRny}1eFuWu^edno-_AvIQh!dEj!3|%X>566?Crv}`*agpAd} zg-c=t5(BA*S-4j~N`zxQfX9VEB(0k(2sh#wII2I`0+ zxJX^BQmNfPY|)CZKWnc)BnxGmZ9H@cxzBHkh2S!hE7hT`0Fc8=W~Js}{l$|z2}Ps> z;|L=#H0%wmV0QjDvJMIg%Wr73c5fkPyzB_`Z@dWym$vBlYW|9AayQvp3}p^kkm5DA zj=Zfz>Es#4LXCeyzex7Ve|{a+f7!~xCdbJ@niT0AZRHMm%+HMTPB2w zY;s5EM;vDB^?pCXRtYbOAu6!O5M%_!{`w)1Wd|SYP^laCy>xwu>D9CSDkDF_9*lEU;pMA5&6u5o=Q_{)FDk4nXSbRpY9h;x; zoL1H9^h`RFc0QX@mJEW7*=uKPD$TPM-d4*L`UtK#1a6P&$}Ec9QCfcBq)4tZ=%tS< zOkEklgZV;JsuCu7q4cLl(sK_Tf*arky~aT3TG@L%>}t+z3UJFl9OZbcRUwe8MhS zmrIg!dKJ9+ZqXYK*(@kzgc3|`DE9iIBB*q24iu_%i@gX!Ti&7Z6n?Q;#zCA5237m* z`+4o9edRahD!zhk49vZxe*&&(pZVB%Oi1FJ>eXZ$M;fBiBGpy1V!+Al-hPqra4oCU zOmb+sT$GgFe3qM-tg9M7r!8gyue$W4Fz?~~$X?WH3RBv8-<0bRBwo@2RHEFKu@&iptVQ6`}3$m&0R9X3FMM>Y_cTJS&Ew|&Z=R=(=m~xR|wwK?hS{-rzUkIX33~zicwt{RHbTbZ0l?as6ZD z7_;=Lk4Sx6R7Q8a0PT^0W!qsJp8U*a{<>NdYf%xUf$6NP-7Kz_{Ok4a2($u03Su}y zM2R5fNHur(t?=L!+~d^?IVYkuL%dJ>;~D#6ANxtW1~X7E5WgbEa7H5oCvozpPx%D- z6maD8;?DhEQBmvU4Gwl7Rgeit5^wVSvKJoX;s96{x>*Zcq1+N>U*UtBGD*?hLo4|K zx=yO`{u80<-bV(8qo9}!H3&`}dcGXvhtDS!SI>?+Yx$IoPyWvTeGlj&VH=a0C!3uD z4RenS^*Ee7Lb4xZ4oVKw2vc1lJ^V0-o0)NSky%L^rVw6=!;l{*gone!mB+l;LNZ96 z=!h27tl&-+1xCHN-sX3Q53vw`pzLzc3o@EJ*KV{-!o3GSgA~VD`1!i4&>;aFGMJNK zWlT&S{c9)k0f+7tjI|hf!TUm7LI2v?%H2VpJG3+MGt9#;f)2j@f@^%e-XRnp&JZX* zo`Ld-4a})@J=7pW%mf@&NbNnN^F{vZ?cJf`uHqvW&K{=+h@9{>2@g7e*co*a>4f*X zr72-9OI^phnPINDT#V>SHuuYn^~4|XqWYK?@?XlpY8O~WX)>GkpIU#fao&2-`9r<*VsB-F=!(DJj3UEPF7TD8vNX9Ls^LN|XysPKG(ImqxqsL=zl&%G$-KwcYOAi8gK~>jTt&~sXO)=0Ok6d#>>Q6+9P$daY`XC- zxzLH1%EAPBUv@YjAgyJ9B3D%7Bir`ms7|Z5NB4E!*Y85+FfpOXcx5w`h~kgj?~;Iz z=&S=q$uA;deBcD<^8sUO2b10GD5=N@je?(En|ge%s2!S~tz8L(+q9-WzJA>j#Jky7 z5s;Lwk6eGz34-XW?GHB|5>xajvd$oH3^D!d?aqcUC=+pz*fLh#CrjQ=>Z!$5p9W_& zeqY8+uk#I{uHsj=V`%$=HSet&A>Cd5lLSg3u!`I+XCKa`HjQ&LA5qhI8InZH__{aF zF!g}Vm)Kme;PtLNjN`5DcG5oL4CqJ3S{&d@@#@7OMQF3t2fs}c(Ri)w<@m8H#@M6N zi!Uy`dBkUGz1#9;z*Bv`%3}bHan14U^*v0ky$5}Zq@G(F)jo%X4LNcWQveZqG|=gO zqbS5W@37y|JWS*Q1Mf3jhs3(jVL#A%{LkXe-Nr4lwI*W;u?~>XC^Aebs^EZ#-x1n?A14e=Zl=`^R~b>-#L&vdN44~Y+sN2Q!f@7Vtf<7 z#P#zVqmRPaus)t)lpsN2o$NzI?aW7D0`@P^XaPOr`mg~$Iye1TXwpJnh*|7(cfKj; zH`E8f-ixGzJlv!tQI$CoTJ>-AVnO-*gycfV5f8u#emIm>tTJMvomoEr3eBAn8YK^n z*+)ZOuUUV&ZPpv36hP+tiR+b0!cRy+w-&mBlFw`**+zlrB-CLQ*1PEgpB+C%l0&}3 zw#9UBfF}#rZR^G1npa!N+LK}Hy*A&hj@^}FjYu!1r({$0d7!S)?O7JepX+bhR}21a@1%lC@6o#T7G5&B%!_FRCr_9F~7(a){D&{riSSr>OjW%EeUefI%AJidj4QmiYYFl}lS@ zW$Si6o^0g~K4+ql)ic3;T5~)!0Od$k9NGx+AD^hLcFHmmZUFx(x%EP+|y7Yzh~>Ok(ntKh+3ae5~rKa8v7zWmrq zYZa$)4vt07gAo7-x;^onCSW+e9`@zGtiL|z;P!x}Tldd(+-WEX-iKf+#>sbFxeFq2 z&JFcRV_u+6MKNc9yg=zX5sz^E*F$#I!#LHd?*FXpsdIo3zsxsC<(lF`); zbZ5<_#e0GdJ6vkx)=;A)r(GfQd9boL!iH>h&wW0^e7H$C`!-#TPAw|2luC_pm?R*! z#49F59PkZ#@lZ5OBGhPgqikRYkWaSC6R&LVq^Tch+64aX%lhRXY26=H?v;Mb#-=FkT;>td16Z+)TC z1rn{d9?mFeM62?Y1Flnd+uD{-Q?vw|UB}?5OS@ z&Rey#Md=Jz=b4|XnrlPx^?hgu%}J%^qlig1nRz2IZJkp8m;b8`Hvt&Q>&^8u+8g$&>3is!vuL*iq39Yto6Si>Qro%erYWpa1lCk%bCoj}&0yVOUe8z+u z-XP{=U8r;#wT~IKfcIc;Hkax&?%WNHr#cKfQyBcLQIAd@5*r?L(l7n)4Kfppu!)}I zFx47L#AJ=nsg+((?>0L4H#k>A1l^yOpOc@2CCcgC2V?}!)(#Vyf(6$Gg@v|=-A#dS zDWSFGtXsRmf5jGL`1{VGy=-W8bXo`$DpYHxQ)CFiYd=@Gf+l~^1gjIlh@XKID4`(} zW;%B=doN#HqXZL>zR?}L;U|ThadvvJ5MkI{aNvZW{t~__yH5Bv?)S zPLqiAx$Rju9eG2F&RA#J!jloV<6InwPmj0~nN zHce0OnL84LhKe#3Oj&huEY%;Qs@g~RJPc6EvT3{$H(55BCXo3O#f`q;WqCM7TtoS= zJ@@br>jn$=b+ZJWwBnlVR&wJ5vdYUV6@q^y>dwdpTVTfE>KwKEXfhz?R5_TDlwC2| zZuhCME?$09Uq|EVC*B(1vC(ROXUl|m>VR}}Yi8Wv65>Fc9f-Gp%<{A92mjZ^KY`#n z?7QE~;3=?O&&4);AV0aP*NvQ7MN?Y2lvNF}NoQl7eQA>lNl^d}mdfn2b=f49V3&`@ ze#G@uK%VDy_-BF3u#>={Nx|OdL3p4vRNqUfe0gbl!9akHZMoHlgcOQHN$DW248aL-PdS65YIC^T60y6-VGMw)0jVtUfKbNxY8QL--G{5Zz z2q~JI9{B>T;0tg7c)rDy8@f*r7FIcrs~4yc2QmWL1ciP_2Ln18p)>#i)9Be*JL46=|8Q=GthXZgD`0V#;z1m#b?MlX9< z@ZWsvr|8)9%7SkXv79?R*>O5dAGQ|cc^%h@LI+7HmVwfX<4%m>66Viz**;I`n+ay9 z(}dt_FwSrcpjeOBOb`0R`gYiDK2tngYG1Oh45r6~jeAR^Z@d{|ODOh^SF4o7qu(J2 znd&#_>}1|T%SW_&SiiTW#bV z#b<@Vc&A&7AgjsT+Fy#)K(#|UL3CI>o!G9D?q+3>(}y0K+lFNq+buczdKk^*Ht6bf zzs7mjUiMB^kpC$9%YUn^!!FL8Z=1;Zzw;{(Y{he!8HFc&q?x}^q2^{b}S*|tHlMpF0GHguBw zXvP?}C1*4OuX)UGu7QD(ro&Rw@|SR=+aEaxICbdTgkpzOl2clHsI6oQ?_y~;H%{4f z{>X`dKjk6zHtcg2G%q|EkRraPexUf^dh{VIFi7`wl}qyZ;Zxb`kMAyLNkDEvd8SDF zr&hD9s-sdlT>@JJWp}XYq)jn4z4A(~3Z4D-^-9+8Qj>r*U78`p$ ze-!Qg>7De!dOqaF>M0$Ym~WG7{hMAB0#YBG3mzhPDbRddD)4t?ztQ|5#S3pB03DAZ*%i_PhIJCEC0%W`6~`$FDk`vl#wyN2RF;ED8sb;P}Y zjmeTF^Y z=^J?izW7%q$`r2%BSR!#tJG>7siDI41VKWSnKm26&BY;^&(<>l$-a4iGeBr}*A#3# zj8!e!8PbAPJtZy_<#GZ_2-o=2=5u0J8MA5tbU^CqP+|;ASKk9YP7lY9IqP==Tu+`} zJ50S(KqbY=142ZnK`;{WQtzSjH9HypQ*yzqi~U=s+fwR(Akz;1wg( zPy4{I!1#@_pWz`nOqXbqZJRtabK;s1NbiT1rJF1$)P?-$Q z@rcIE9A`Ld!5twA4%mS2S|dK2=^Z}_>-R7}GvRNP9mK;q0FQJG)F4H9D+}n*e@MbK zZ=FXZrSC`Xz)Cc^%ywpZ55EQ+Oh+_yB7HCr!U!Y7FD#8B#~g8ev-&^_IBqQtu5A^{ zY2Vs~7S7YzA47XQ>gpRw8HjI@xE7p^y#_8pzM=_IF7V^75~2f#H=&%-ZSTvQuQ-Ii zRiLJ*6GtaYaU!{LbZZ(`g6$#8Sd+5q8`f;Go4QUyiLQdxWb~ERlQXi4Mv*5rnc?hmgz9A zD8Sl}b8nAUI0s#ZDrU-k2pdoSN{wU4J!1(e%z}UZGS3>3in!SXf@C!ZWe&#Cciw^V z%{oi1GzadrU{=Hb2r)aPJCq~8{@p*IKR?Snb%g(Fq1sOtNT#4h5;o|~>Q7acuc2Dx zaWpNUABr|QAM0N^8&^`^tCNItGLs%qkeI=~f;N|sxVcsrS$;Y)lv92Ev?xXUDN-Rm z9$7wpPj9S2t(xr+!rhs4x#}fg_TuP5$Svm+O0v!PfkpJ5IS-2XWc|^tfPF}>f#*lo zf5x~CY9X@86pqpez!Y@JLNT+AaDS8@?zgt|9)>+5wiO{%)oLRFb)_3b4e~nLw@dhzbz~7h;D0Mc516fE#+-#Ve zw1A3vTrIFXn423(7^PQ)+dpty7@krbV*sP0d*&YR74sc10hQ^KojA-LWB&sr5e5ol zoMI+H6``S+VV>v*O9>hH;Tb?X)!SZu52xiR`ELKiWUTH~Yuoxc6dgEZ=X*scq$Yk) z@Y;8kfd2c_q39Y*EB5Bc4_e^ye>L<>O+kkEl3bkq8UNy?5vF(zLv{!_{rsro-jjIz zGG++u{kilyaAd5TouLbXzx&(!9>c-{5BgFQB!7pwuV9fuI{>uzMXwg$UVw9g)H8UK zILwUz4xI2-&-dPB)%WqV>s3NSDLd!c&|R_D7dudneZFbZ(RI@ z$Xt>+!LZEB{*T7lS#B9!RD<;nf+KUlM8fglWSEJJv$&3k=%#m2(?=W30kFzF+WxlY zEyC{6zvMM85iR_VgNwM2HDhJ#H~bCW9VZ%diW3T*tt3F=+Eed-m}Rt+m>9f@H6`)` z_TjwpK0KmYywvABkD)jhVou8GN8g}o!Z5l0?p|V>i9)&YJkQ=(4B52^b|W2mT|NM_ z?B_11yXo7lxNGVNGHd^)Y973(1b{9E3qQXdH;~0O5R8~(KzXy?vZZ@`-WxNXt+Oj>$ypX4kFGt z*#GOsW1y1Qa-F`j2J?V+ibe`j)KZcM4gSm%Wz2AmYIxwG6%aXc$o2|d?dFq^C8JO| zP?{g6nz3}K_Uev_8)$_QI6Ra67>{Nl#hR{e9az;SG9>Z4`H^Od3U_EjKnXJWtCf6T6dXm zd6vpaXn6XPry{Z2wRlNYR`qd*r`msQgYMQ_-;k=Ls6sUbO2FMv6-|(_*(ZQb5?@{2 zsa58)KW90`D{^QdTPqUHcl(~pUGq*t)z(wq3Zx|0t;^qtRWg@|LfuqOeqBcQ@#pl;g7i z7g)Aj$uUXViZ_sISw*u}+=K^Z_Lr_3$C-OreEeZyYX;>QLcPgCShubw0sENVi35fK zkpgqELg>^X??F8Yei$22`iG2uy>m&Xfw9GA;(&nP*j-|rMbTlYcDd}i&4)cjtQM& z2ji6CnQY{}5sd}h3@ykVWFQk?wDu`#NTeTQ4F!_sR7-vr>nro+3o{|7^+NTvsE%02 z<_d_Wfl4O@_H$d454{JKJk%I$ojN?8{$mB}4C#>vDLWK2vdK663D(aS%p4XL`6Ki$ zLSdM|uksDFC?B#GU+WkfV;Cxz3-tz&o^MU_X1qa|daV zg}ks>vb6y>k?k_znA|kJW-NbDP}rBE_X@Rx)UfT6=t-7$1NrBVVT6O)4yB1!*M;W2 zl@rRGR?-chFDq21->{%>;AuR*dB(b~APJNEh1%<_(I0`LKUKAST9byvIK@|o6({;U zFc3-WUFw1?`p@om_<~{tdc15&{AT;w#k=C<3nSw2&gHl~W4g`VyAI4gYcX|aN^dUx zeo>M;kfrWy`AR{eAX8l>pKfrh`GjlnY}NCne%jq)!$pZ{zGYiBj7IpHrTcSayK8p4 zY4D?LGgHg+4e7qpJ_iqQ#dDR>6X@kc5!GR#+R~56I=UNVxbE`y}alA z)zw!Lzv{l>LdVya%e8hl_;$=Qb_$v`k6F(zZR>{8j5?=~nIf~PxYlNN;Y$WV5%8} z5uVvnrX0^O{*>LY#gg>N#M1E+BXJ346vgNLLkU75PH1GnbYIc%fNb#yHxF(RH(00v zs6AKWgklcm7}(#a`wj={r|_83VvD6C;R z4d29a(Yl0n`dwn2R%OG?dy~QsAD@!(*_Y}Zj-omdxBqj&6K6n0eC)i6beCwQ zUnc2b56VBUf26n%1y62GFXpuPwR}aom)U>!HG5HGhohJD3<#*-cVT)x_N=)-DTdJI zED+N)S3hn+n(fqxqTaX3CMQw zz7Nwm54UH_>n}~PHi{Y2ZN>rppcRpsQsf}~0pZB;-tY{NkNbyB^59(2Az@Ke)Mvcq zG-eNgu=jiu{P$-h>A8Y%{~zP?_Vwp#7{QV@!O}K{d5oT~=0nA?Ba`k32c2gMHIhQM0-uABhYx(#?yasja8_9pMiXI)Vm&W=e-EK;!dIl=Y{NH7S4g82_<@T{g=ExOa?6{KApXoD(kZB8q0BhBXUN<@sO4nT@Pv3yV1^SO1&h_?vcf z-fzs4%F~V7k&9PP>QOTjcOa+eLXgnr#^?YKsaS2qeH$=*AT)o*i+NDPZV@9yW=++U za5&!*4bLgtjB)CE5Cb=b;lLctLq_7qjgg3rU=awjY_LgzVeloo@h`w*Ly0O@+$|@V zboMQk6&QKtV?$bDUWd=sPLDeFzH3+HLTHes(0hJvmv8)DPk+oXsL}JbjGDms!?|pk zihstKOxsw<(2y*h-yg&67aCHc4$YJ)`+WB5c*kPELbF23-inXIGSgh307dN5Zj!vx zedb@Coz1D0O~GN1gG^+?xyWQy^$?pK{uQ+1FffQ;az4dm|9y%9tE&Y9n0eQ7l(QYW z5+M;n|lxA(7%Iu(UCuPCOR*p=wKAL(Cqd3hP(S+!bzK8en^@_ej~ zCH;7L2{%c|6#|*7g3jqIPl3;e8tNHzST|_=@xGLHzj=m_`Tg>|Qh|&4{6oHX5S<9G zavRaLd0-MANypcrR^UnuQ)m+UD$Cj$p?UsVg+qw{}bG^qwZ` zgQR(SKK}xMhB()7sZYt=m(HRWG!0^I$NhR3-Ul+|8=FNEdf?&&%^(kH-RyA(51wV8 ziFbmUuTwA?N-`*f$7*W1GYtBtpxRjWzwTDIP99w-`H7b)p#u|aiOeh%T%XfnD4RrQu~FR%{Sr<^YvGeuL_6pks@0L)&_k7 zFTjRZ=lMNWNRy1cj9aRI3dIMsLGDEASfXl%KCaXjIzStqFAk$0<$n|iMzIt|(%gK+ zf!M<2+iY{-K1vh}O8(G)DbYJ)yn2?Y9}eql3d>$SFr#gx7_tN#*?uxc>Ig$Z30w^$fL1I zMg2vRd&F@1%W+CEOmEku(sbdykmFKn+pXPIm+nkA(85xc2tlOqU9?g6CG1Br&=pb0 z&$8(|E@@ZPZcT#6T<3$Bl{>$6N*DIUUn%VhTGyXaB&-7bLW`eon55`+kx<^yzge)m z51prtMmWw+;B0lnonj<78`eVdPiGzqd7nMi8TA}q8Q)RY#yr`8 z%ag+oktDAp+&<5k5id?l#}Hb#Wd8;V@;8$%mQ1lR*T{7`G8PNQpvJkwCpf0mn{#qB zVQr}I!N{#hy@AXHvE7R;W{4fA>G&(}83X)32QC%Yl2?Y1FQ%GE4#ggv(bKsK3DTA+ z@wuv49&Snb@HZ;aC#!p&(V0&{8rz9~vX_qXyBzCOqOUHO`K-UOoLl9{Bo?pu^~XT9 z*hQrX;>K5EqnH6vS@cXShwxRfn{~}m#V9&nAiAtQxREw@I4*xG&wQ0o#j#a1kqV(( zZgD2|4W;8FDj6yN;1>OJGRH~CR#la;Tbi!%yJXScq}y2!M9RcWz3q)xkydl{3)e-1 z!)sbv($Q3bQXrDL<;L%gvxzMF*)E?Xt>$tSCzy|xWs!cxFXI$;z$BJVn_<%gnw=X8 zioMgz-j23eUF0`u)#fUp;8I)Nlq>B*X%u}G+kTFumlg8sVkgan;uXiM?p_v^XSyJ{ zm6w~>(@nRR@>k^0(d+1nbYa`6&(6iDrU^(VwMM9FHcZ(-TegVi_lH$YH2vvDrB^Qa zCCI{5VBWG01?UBgpBnIO-sA-=sIS(|+7TU($T~F9q0Cj#*)@&^QkAa=%B%9!t2WxS zmAkfdU#XXs+ZU8ex!&o@8?^9WHETv=9^^mnH87VYXoVO$( z+>n$jn_pF`)s?eSC&LiuS=-2yL2x#o~l@cMp&+*fj2{|+pb0<&`9K+E5K?&EMJ`jST0#UYyeg zcA$Nj`8VK|EvS%U656tjPD+d5&w|=BbaB!LA6NHw5a3Jmyf_lFjoQ=9^ zL4>)H3H!j6^js0W0S~(LjY34L8s_|2|^^%#Ue*cR0Fi2lK zQ5coXQ5AZ!1Nn(rMUVf&P&X{zo>jFa*)}1MjzW~-wM06o1iq)~*}g2{SR<*`+t{OY z%jIhc-Dr_R`e-*d0+ZuIDYxLMa(6=bz6|F&kNn6pJ4H~NPo+m{UCJ@xd&75e1e0Ax z6XQXy%}F@zZu3PJ>!Ug)x7s~?-m#mV%?deDsPPPy{Rf#?jEIv0!fyWnZr|5zGJdpt z(gb_ENg_Q%t0*lM58Y~)EXqFD^aHW3YWA$y z=@NFi>@jPjZCpuoSwzy-!*r5`P0|j*2A%qa3Khp=rD1>lKuxNZH;KC5w~~09@)7Og z-VgI}POlS6P4T52X?3FfSFCeVl^|2TMuPWaGu&Ser7%ssgUrzKqE?Q#5as<#q6vlW z&omh6*nL}WC(+wC49pe2=`H`pD;-8#wa1gVr zMpGK+l4TiI96n#?6h>YnMo_llXFCZc z(%p|YFe`BUwhlcT?W9j~yWdp}s_Mt0w?>IV7$%bfwL(ia{b+9Ui)ca>QO3f(Zo@tR zR&YD`mD*ANyH=RrI=s2nCCbml6JD4m)I`+!rrY5$LACfp5?xWbMp-qM7(tSH(eSJW z)FNOzUWcx+CbeV_+bU6JKOw3kzoWHVS9(w6EV-<8Bwo9mz43oc{RLE%Z?ygoD<~lX zgNSsO(%lWx-Q696z|bL$(k(3|-QCU5NcYfPLpS^%zUO_<@2mwZ77X*uJlwJOz4vu} zir=N8s+l>mUd#2Q`hm!E(F3CO0$xo-ve^_k+i)?lRUe}S8*;>EweU^X(=OIqG|_k0 z_FAYsVMa%o@-BWKeGR={>GT3PRcyz6#lzp)OWp++8&MBYMG_qwxSr z#0uv$TB-RW^qhJfjSLs)VU%u%;4l7o0~i350d^AVzPCdLEf8X3nSKBvx_-^t-v)p& z9(eYqzrRQ;Uew_R3hAe~{(4AVFT~@+Rs^m3BjBX_GHVT~+Z`am>;e8ES6Knwr$pkQ z8t;wf`w%j3Lt?{Eksh+g%@L+uEXQ3u^b}UKgpmWk2=shq3V84!7Vt z`j4-#sW0$zj(BQBZTpTD|0bt}VSl!%dW3}tUO)>DsmF@QhwY(+XXvL{e;+pZWZzT4 z{nK+@-YWkN$iMDSt~4k}a(qa7VhuJ4-5cA7plu*1dKZgP%pN@R0CyS*Na1=xktAE? zFH^?^dL`RW-!2}^W*R$kFx`C$kWk{@CmpFOAHz`|rIqKdM=;3g@}^HKs2xSBV)txZsH7roiX)w~huP}fuO`$C4O4Q0&e^0Z0VLGp#cWP>O+R@(N=(nlba1zH;+0}@ z(&z^U#}yN5@Hfu>4ufxomVcCp3}AhwDIZo$W(jx~{cgFZ&102?s+XNe3U?4#tfYd9 zr;Ds3Vrky_JDVO=pqb>E>n&9>$PMYe_4z?#8%}^^RI+okmp}aOeUIQBmDkRqy(>!i z7V{6LZS$XGRKA!#aVG1HSMLXfF7dC4-uZR3hy9|du~w_Rx82ETY%QjZh0m&{7(k^` zS)zWHSC+DhJwVE}6ngebcRT!4Yp?Rh-0Ex5XL^+@bJs{YyIT74+JllLnpUkUZSzCb z`l%r6QJVIr3--#QwOloo(58k++GV}ubhuOzb4M$Vd zrgzuJs=-~E_e+#S;yit7!)eB(`+DvhEK$8byt$M`8;K?(apIlbXiSA5tVC~;m zj%IMn{P29o*dGMRGLI3wmpf0_wtaS3zZ7(He~N1DT8fY^qf=I|R-Lsx7ip>~q;Yc9 z{Tok{l&IFrkSD!lQAjhvQ&iB4Q6w#H2)1y5)YGSAZ|4nhrOGIC!YKcGR4pp&@+YPh zzN>H49!2?)r=~?yW{riRO8#^Agg=%elhQVhl#bJOV4(8{8#T2?a~~(Z%8i?X;RY14NF5$X0?bU^1e`NUrUy9{wO9S;=i8t6Q^HDfQ?7_x*AG?K%ZUk z07$*&H#|q?!aX4}AdwC1OUz5GgYHK_`yVb#aw~hY;gOb9Ns1{dI!@Yxo?}Gvh;f6{ z9%mstQ92YDyn9tp$Jb-{>qkImgj2oE?C3%@0!5Q-ANdmT3B~%SxK=CsTh9B^HY6&g z=ucCvi9|m|Z}j6V=2+=Z{Mn?sQcJQmwG!}`>$G@P5DVDyfCF#V*M-uBU2V#H5ZIQ; zt_xU$>*qB|SL;Nga9n>oNfzP2vEVHAXGRB|jG^KW&$ZZj6|M1GP2$ikQjWyOGCN_28+YsIdIVuR(q`gyCo!>^3=-QBz%IV%#k?lQz9__hKpY+yH6Rvf zKaoVB*xx^(bWKKxxlHUug((^=DK5xe!u zkjS+B_^QhynlH{+E-A6PluFab4r*Y2a;8X?fg3gy+V`{^o}!wKCVa^}PL;|t4-zIc zPun~+y=C8>QDv(+gbz=@8N{PrAHM?NeQ_|ltO6l;9M=THaOWA`3{KL&-qjioZVZ;p6?A&@e#xtvdmNrt3FH*&KX0({ z=e1B9dR?>+uJlY^c7Fgb$)vE7OTw2)Nk(6|VEEMp@Q#9`$_#+Nb7@z-8^jAMAdeKkg9C7&xD;zzdk-ujJd&+=>{4>=*!giWjncy4(6sftnm8 zSDbRt*DavK;G&Bl4qMBB-xS*7?ql11xNuO`EDz|+z?-SdowVp{=qEy>euxH0JF81> zZ`N-mxRiQ$S`bgsOuzfQSpwaaA%Vp#aG8w*zqLQQFb+B7IphgJ-O+$+QkO$LLITkY^3CZ)U#~G<>^4_fYi2yO zmc7p*U<@~OjuzS_{A?_qzb~FHe3!e)conx_S*Xf!P*zj|{Z?{v#|JH2TcELz;)4APF6L9Jd8zxvPF#f+m^(sP#UX|3 zhZp%2bzF1-@4<299<+PCRGLORZz`L8uw-cyI@u+3W6Q2-HgZ=$Q5)uIuXAT}gS-cZ z%RxSAkRd+=bF=1w#}Gwm?zK^4o6^77-FD<1*}A$l%Gk_up16pwD($69HmJL0z{6io z$A3||s4e#%IS^rwkimP0`LyrnD@=}xUM+@8A@F!H$pdf^AMjA$!(E5avi8v;G!0*M zhrtj9Z-h5f@6G{_^n0!1Vbj+2U$kVNsR(80CAa5cm)ogfdeB<6`>R&f)%}dB3F%td z#gf;A+r=6ofQFn|M~|n0vFM%d@*2(0?Xnn(C-%yaR1Edq4mFTm!r3~XHe3K7kh(npn>$9N1CQ*tl1Lz}*zwsuI`O&Y71R&A zy6JX8UpoO9EY~*3Ra6kbLH%SFZc~%*3D$xF@j>EuBVj=>mFobUOM89WIYH|2KC^(` zRfUY?yH3kKQ@TO)ECFWCYV@&r%W`3IpS5)4BLtkWIT+}DFXB-94dpPe5Zk)_puS~G zWKs*t3FaEQU)WutL=;hGu|f*V)T!I<1VRYr=kS}5pC4)>WUsPH+4uWrekzZ$&1wzA zRC!3zFXJRQ{1Th$!u*9iiuigz51^g?^CfZ`W^TSyj3Bb*M|mP?Pb94Slf~IoV^4g` z(o;*U19gQwj;pPg$?bu&rmk(i`FiDH#La{%N-k@aOJTZ)RK{4byy;xZ9Q?clf zU8ihdqG%r4hThjB9-LAM$&seUYWA4meZwx<+ywOEcjyd^+NM(c^6^a#&O6|d-gsFAF_+^061Ub!N`7HYQL^J%ip_MN`a)H8gtAVZ#)mLD3uiU6 z^9=3goOW|Y%N;y9t20xpK3S`${5|cqZxu6JM1^){8wp7+HG|rOJKuWr-uRU*6j5%=D;U8xW$}DfQBsUyT&ijRGG%&#Y7x!qw(#zsvoB!Dz$GNSIlT=`p#JCJ6 z)s00gEP3KRjqu_hZ_HVO_I|&vn{2h^)s={zJ8ncFOKqmQKTkw#7hR&G%-yp{7K$O! zmC&;GRJ)abSQR3$KgnngklBokakdm>n1ftb5`_t*1rG7?81$sFGyzAhgKEMXkf|6A|?G{e6b3Y4v<&s#cg^j&q;yv8wA&fm+Q0sOA^V|dNH_jWn>zB@V=F>#j< z<(@~*fA|t*t3Zn?(kAalzfx=_$V^i<{kM|R>b`OuURJE`E zutAQx+qUVg5i{J%%4KhgsT=4qwYPzO;`4wJgNJo&9^g2WY~>h@kQu@CJ>C58pIFBh zE}sI_1->Y@P#AS~)uzLk-m5p8NUb1tB{60gbb0$hew9YuaFX($ZZA~$Imm2T&$cOh z->L3~VsNK8%-IsMnx@=c45}1g(Z??oM4DQjyDMZEuA{`PH8;llncO7?kmDuF$c;q;>DMPT?8p;_Nw>6DmE~GylkDcPlxv>=vK6O<2snMg4{P{Jk2OcL+>18r{@{(s-sb zf2^rydT|QwD1!*U8QQ0+gnW9O&AvQ2OBQRwfZ-zzx~@zA*S z|BNt?L(`sO=3S>IL6j-|uOq<)MexvY z>V_UCWZ4^kqT}151-7u<^&NOl9m&ZJcCoZNyK4V9nbG4qcF= z%cJjCNFWQDEEBXKG(bp`dFBI!GttV|Fwcv)-wy+Gk(5MN>NmwGm3>|>Dia?~{*)U{ zq8W^d!eyh(c?3|_A5$HsdAIK3iKe97qiO-pfw?&aLq=))E&j!xo~x7&hCtL`CO5p1 z^Cq{&B@Q~to=Kr1SbjTQF$m2b*oXUbF4rM?gUb>)GzpZB6{JxOa{Hdi@hMmYT@&nL zBY2nj9e;zr$9P^qebl1}8ZczY1AFn1dC!5Q`0gKn$&t{K@wy`UYEb1<4(jfEHv85~ z_4ZFB%m$+7qiK8Cchq92Yq%hUTZ)hW9fBb?h5%N=u2CG;!OVO%kr?JovMzmV@Q6~6 zh|+|-)K>pi2>>ame7dVok$B`$R$cmH)2~i7P*3ZzW7X6-pX=(l-)o;tb21$E4VHWg z32ox&gYyV?$&`>WiP}Gp8}d>@x8XfI2Wqudu=i z{$5ex!FF27SAr;Z*{S|xIlEi49s}7@ktuq_04vmIo;9Ey{a=pzC|734TQg&xc9m)F`Vp)IaEq85jZ@4HkaFIg{Q=hheX_}Dpo7BgPY6p_KZ#Be<4 zQr_*BF@=5m?A(V1jN8S6pd@{dfiUl39!lN!q=rzB1Ns+A${nCg*b&q!f348rK=^LP z&(QAOdf4^Rq6YcT*V{s%9vjeF)FlPf=icsl;7@Hug(7fxkZquu@%o^x!~g8XQ`7)I zz|Y>~2U0;vy%-M145rBlWe>27!>#@Bs7bDx<^i`Zn3pK4S=Zg6+m^O@^U%=qf0cSa zl}!21+TkCUc@28cYs^P95(X8`4wn&?iA1R)*};F>5lW5_=KXo!z}Uim6&BsC%(N`RElNV^!S&4G*$nfs zS;QW}L8`3}=${icgoFkh=Lwsq;2@3!Sz^ekbHzz5J0CMt2{QMq+Pf0G-|&k8p1rvk zCGIG(wGK*E1||-VcO3x^8+{aZ%ZY!}2yq98?Ehx3o70q9{w<1^&9g9V z!fcMK79i4Z+bc5x(H4AvVsn*kY9=AY-&JID;@lHl-fq3Vsh45ph%{MS1Nn?`b8~_t zRzhn}o<`W+S03+AB8_qwevxUedZjB*R!`+v*t_!Vf616Hpmp0ja!^A?BM)_VyCkI} zX9_Vo_}*`@RYbt|s0oR6lxKDrsY74nHkDi*hp?EHYGct*Th!jSwJy}Vy4$mH1wQ21 z10nXzI`cX_UkS(p*6>R|v$XI>f6v1A7bDICJN(Ol+M)^Foi%{;c_BjZZ2ou-um$Dq zTTX)w3{HSDLmL7r-sVd6iyHe5uy}lYa%k!G4ABg=l*q+eZ8cr$<3>6dX!3r%zh2qu zgs|i0%cb*7Ux(Z*=SsyrEF9MigpbL8C+uv!6eHMOsNEy{1pM=hJ>Q4|$#@^4I(L&duIf=^W8J)}>Z?53i z2KwC>-=nr5^M*SGytzixu)y4Sw7~ttkS}!<<#_G->NpRfB5!@EH)Ty?^%5kZ`K1KW z`N{3Humf}hdobbn+_eAzWj3BO01wczOX64XA0gFqjyU27NaWpjprItjYgT!HRc6+1 zPBP22mm7Bhi2Mh~xa{8lv*?3q@AfTvsjD(M{i&PhZ6@wEg$E8kf({TZN7I0JNXW#; zI;Wv57NC}$yIyLoUn$h8qcD!qnqT0wXeTWd$!7k~SZGyER8z{F5(g2H2ir{mOhK1b z9Ec&g8!A>V(Z@Xe&oE{YEd&u#w@a+LdardTq_~h8?S-W7R>@8^&Pen}A#@J%@TTQ8 zh6c54bfkbb9;-OdQY%__2qM{VRtN%AWBgI_t|XO1I_&#LFt{sMYC^|es=S?{^**wg zmP(q~mty2WEJaX5yPc{)Bm2u2)WEN{0cGVIy?@TmDAd#&WvNxEMbcMvTQR;NdFX%K zoU}7)@Uw*BiTx7=Efpgps2v}0;51XUzh#F;h@tjkCKFzNzHK0wu?RFjIPdYn(mY1hXO$d!DpmPcfjKN z;KeqK>EWe7HKk<%SX=7~9o`=pJn!GVDN|V$66;pOH9s~O4Q=u(fAV*m~<6dX9OI?RP?K`%zFcf%<{jL+Kr zLzt>sCsuEiIisNoU0@L{Gr+vS9IbpDF!!S@k_xA=IW?dfZO_$mM zhlGFXu8M{D7aCXv4GqeZGj&Rk*AT0ISdPCFKB~m2dJ^Y64eghy5T_Yh_UeGX2>$)^ zwO9vSMn}TvsZW{iN`$YFuy3~*heCo&XmccF#+o_M*9w`1mxCe983XtGRxRy+w0l{@ z=&$Vcaa#~Y=M;RZ1n>`-*0CfBh1o?CMoxi~pL#djDm0^uVh9CXOJ_lwPJ;&Hk^_mz zZiMOqBGv-ZS&+*X0O4Jje!^)MeVQBvt)X^Cp)NAoA)dTmA%EYr>3mc%)$U8ew4sOB zRoTMnH5T{^idr{EVd02ww88Yr#ES!us4J)X>XXqNugf9MybT*$pag~N25hbSk;hsN z)tv8KKEt648QMKG$14eG{PwCN#k1z8TkSsuBM$lvN})sY2kMP4Gb{MQl0T3RyUzm# zlBclNxVeJqJ&etf^m}yUJl>AUl?BWqWl1DAG%~I}2K^bhiMC{1n_~+6PO4-EGp2g; zWkZ;cWEat^tlicN?X0V?3C(~&P0h($X`fnQ?&jmwUrSGFQW7NN2k;SuURi+YT z8Yf`fx=3uM0tm3j?AJicu+TMlz!%Oj&($FGzVgYj=2OD1$agRC2TtH z#0Zar%>0NUr}Hx{?Q{sS{3*O)xbd_rQb7uuE%(}T%enmCDTH}j!sYn2$fKT5<3;Pq z+cj!@%+t>g(}>zERBuVHm`GoZF>>?}aI|!EMO5*Z(MCBq;m23|_qEnFb)0uvBY z4ridC00vy;Jq7~H0$K+c^qZ@!EhCqGd@4xrjB;3p<1C7;PhMJcZudK1R_2N~XMYIO z3C3eT>tU1I{^`$pM*Zec-_onDy~V)x8^GT0rYBGk^9c>%-cuL+2VzC_a3m;+EoLCn zIw4q8`ut1d^P?Ws?$z<~y_1o8PN!!)%P>1g73p3(-K~QshHkHuS5&hJt<33zcN#L# zc6#{(_!ts_P3e1nXfz(7#IpfcT1t(Aq2{WO5N|k!vzsM>(TD}(;h`E?*wDsaeOr= zxrkYaP&Qj6S4&xt!cqh9x%P*|NGDo0++47&o`^N@yqP1GGs)((v&@CYLIcO~Fl*lV zm_5!IBhX^#I)xf&VroU)70~oO<9Z*Dli`Lls}7x)?xroOm@NBz@lx(leQlpyj686) zwk)dJQmNuN)dtd90I0e{7rIw$7saz~EoNFZ!>RFiZAAU+?@ji98i2c+<9j6Ni6DMwjMwHt@t58U{R9 zFvfMoA=<%fva8BXw1bAq;6=b6f?Yuxe6nD%j;oWr~3D2XuSQ)3lt4LDNTIo?Lq z6MJegR*fJPE@I~%TOeGyU_O3882Vz;?!OC3(XNNm$fe=T?@7uRDTdSqeWN`ymdX62 zGR?W;seLv6aU>I)JCwa-ru8%6gjsBuNaJaNPp>ztrc+1d9>4^}!?|o<-A{2ErTYh@ zTT@M|SXPhC${Ec}Gn?jwd38CuW*3x&8{eg^xA9mENE=VWZ(c_YbM5jm4W#@_7j!IN zQqy>Qq6r4goaYOoMjCPBrdFhKk?bn(tfoWo{7P`Sd|Co?)JvlK#7<2jk!LUgN$;}G z{$UZQrRVo&={9nPwXdIO?_RG@*_6M{2vBHhV?iUbmGyDk%QGAi-tWa_{Z0CG>$IBn zDVQ0}N;aEVjV+tHj)P-@B?Wu{oXkY)yOF}Wyf4N{j`Y7z^VyusrG7&42`zFvh}9oo z-Tmq=K4;?H&)8_bIw2F_&}NZFN(fe6-TPU;{dCsP_7WhNJS(O<)KfE~GZM9+zzR%| zBNjU1D0HoJIATCbF{q5K0E@=sB+Ihl7(m2(A!$c$thhOy0qLU=&UbW|iA4CkvV3X& zu_T^wFscppYzKg0Ud(IB^FwO5(y}@TIUBmdiO66vD^#V2=L|O|%)K(nCE7takua|h z^ltd5axxk~%{@55(uNSZbZ(>aB<;zGk^5nZ3gx>~9(`%LFlDTG8 z`FEhE|0)<7fKDnTmvydR%1bAVEU{m`AyT7zdtPcUb-hfz>{ac2IV~(Xc2E}gtzGYK z$N=v2)p&;_-RGxJ^G((+y@p=#^i{4IPf-G~n{FA9!oNxRxKRNeo*ja{l4>|%{v{28ztqmqKS@)zUk9k{CG|D%W+Z9Ak*ZQ zuziOwEy`kT=`gHSU@VLiI#oE4E9K`(VN7kj`}~BBowu)htEpzdN=<2#-DTJv=m<-C>YSGpP8jJdjuM|7JA=`C6xKdp$MeqVW~is zkBb!G4{RZ)*1U(dgpR|S5q$o&rT!rA<>tma1ceA`{bh4yj}Q_BaB9r4=`o2hszrl$ zvpI?3yub~-26<%%?V;jqedCSs-rYf-O*v|hU|>7`H>5yfzP@3Zf>ZzW;Ojqjiyfk% z@lZYqAu>AC&j^uVZ{q!I{Z;|S2U2lTo=Q@Zb@our;YLs%mtecFL-D8ysPo(1g%B5c z&1djl{1>Sd!S-5#$C|tqh2KgdQR>tWr2V7?r~Kg4!}(6}&|!yEUQQ6s>F=;dei8gG z3Vg`dX9hm(ovoEXeJWimHYDG*#}JzqpBp0>o$W*F$!WOQlBi{PFsKO$oGZ9aWESfz z_M2J`dHQFCgg%ULjW0(SsYf)fW}BC`{J-V-ZF&OPtfooNV&fcG(wGO16syc{2S>!~ z7WWC!p02max&-y^@x<2vpoaUJo*eT5J@rkOGdhpjS3yBR`u&S5tF5z5U>icu!%BUa z3BA^gG7A4)o!#PJpdhaXkBD}0fK%Dd!6@=Sxd}_NMrM+8^gf z1cg{0FMVu%I8k8xt}RhILio8BmoB13PRg4>xtL>OS1Dm18X-=BW9dGzgiRO~!$Zgj z7GSpg9jc&5KeLJh&KC_4vJ|ov*R_Q5odp?jf#N~=FoCd)sfa(Wn=unX*a*<9z~2vp z8gL52U6`59vCd21g>j^zjDg*n<)Vk`E%r!(Uh9RQ-UU%oN)3)FENsO{^bc#^e}D#M z{R?>&+JAKl@mmdsiT_q=f!*Mr(yP1a)Y9B!)V^Qw73325c*upmv(b{4S$l2lJJx>s z@jd@L|l0@@I#k*Y=t5bvJSEFDav=44$>H9oz z^m-CDy}z)ut4x1luZe)}L{6Gnb$4Fbx9^@@W2X_L3-`2>>mj1sx79;Eh75@7zAf*! zqG3GsHH_*7;lEG)kFH0-hrFjkGz#+9t+1vSHG@UTn-g?S;ts~?{1aK@)DXCYah6Q!2E8E^Cq9TD7b%s{9N>Lh!-*;A>BjDUYpk;4P6P@+ysgYqUgbTv+3qs78 zepuuz#@k{ps&ujB6u?xJ#)wxY%oAioduD%3E~`>kAv+U~_%d^iri8RIYP|pA*Nnt+ zw3E{}kTb!3tv)iTf4)AxN6#%~%_Vg5pMO9^7BO>iCCtwf4e=1R2X+o0IRTRQpKXrp z=Bs*%SzQh_iLEimkA!|iCmr$-LkL~d4ytR$0ebZMEb^uY3O2*V3OXK}r1$+5dz2tB zbaGCkHxcf>;0m+VZnx%6l`JE!UQok;LVUPG%mrngY%E7xR^r#{sTN}}taaQ06kO{p z0b27DSpHj|ceuyUp4i}}!>HEOhGlaF=25rio!H{6E*jX7H0Wz$iwNHA&o?4abqv$C zfi4(nx0cpFyFQN&u>+tcu;tf7%h`TK;WdtoBZR;2`SfNGw?y2m`)H zJ;b4Ep>}|qt5e{IuP~ADFsyQzTX(A6?mv>W&!qI|yd~1`h}%}MRk&w+G4?<~5u?6$ zz9mYbv!rWG^IRMc;ZQD4Qpv`0OHDz>E6qBX*IFY|qZHDVW)yFr;3y%{(4)9LQbq{D z2VMLq{=i;KeB7}BYLcA@Z;qEa#dP&{c9p7`Ssj1&r{x~rT5S46m9Hc%F-o5%7Ttvm z-El}B)^jXCvs3rZn<-NvGruF?Ljh&D1da_mru)0Dc+aBv0I2AV=ZGZYE<=ggQ=#7| z-S)@!T#WeF6Qw$w-T-ssi3rAJn%;>~CK|j8@g@T@!xwZV8&8o_m88uY!f_d+AGM*! zJ;{Xy)R@vPy(F=>db-F~nCZLa4q0Id8kDr}32_m~MC;`eINDEmlrklIk5>vAP97Y! zy7R6iB-5kN?>w>^nu{4tyi@x8k$wAoUT<&^QMp-8i90Gc-nnZ3WfgDxD?q4Gq>v9z zVOlpy7x1V;$P=XS_NcD8dTX_jI+2L(KTm<=r>f0f`V|m zJ97)Q-Atdjnc(yTON1r4lxE2J?siG|sGwj>sy9&A{Pa%`5klCsOP43af%BU<0^AuU zxBNS#@2NAqjBs?1>GAmBPrq$GHOwlmimh&F6?OAnzXY}$hO2nGBCZh)a(K?IHfcIT zV+aNwnQSNOR`&!}$BcPc-{=rV7Gy+yNKo$+?ibnj5xKV~$ICD$N5L+CWs|d8_&7f~ z11R)^)@$hb8asLp`G$~fw+u`OZk&GK z?3EwiJ^{9r&aF{Dvtb8JXY#nPWUM2- z8cAk_gvN9lz4*x_NXe{dj3-`Rv zd17)$K^N#M`Q}B7w4+r)b(R;vj=Q&0yGtB+e(0l4GV@_34p|QxH~C&x?lr(8czj2? z3S%i86zXF8Y-2Nn1MmW$=(w*ynW4d^yjN(`}(mXyHn! zLdZ4|-%FW$O_?gDUC9Zlla>!ds%4Qcviu42RHnp@05 zhkX1B)c04{O-wM5Es3&YbcTaoD84Hjo;&9ZYu^`8GB4H(uiEl zQtt~r{bkq6PwI>U03e|Qz;n;8*?-J!MZ(wi{dpNIJ^Wv^l*K(~>vuQ0f9~X?J1%Aw zbyb%x zYWJu!$dQTikxdGJj_N~ukW3z#OjthU9kc=$s_GM{irGoL&9Xo;+M%s+FMGX5$$s!L zt>=Gk(aT-({2dh4Akn}ei6SuN@f*J;{orYtDZ^*{;$T+61QIztoXmRpf*@l`4fV-d z{nT1V3}^VcBgp1FNQaF8()LfrUI7y9V4E+W?tEAY@feTxe5pUg3)4CMw}$dbeX*jE zw$ASw&MMk>bR^;Vy68BB5|Ecx@l5t=dpMS(m9o^6IrOFu{Hta|<%_)g8Q>n-+6HEn!q-oz5=3jE;zk%7(ClPmpS(OkLq2>+_e_KIklIy{j$R%cTPSC^jAIqP zAB$(MpU5!qU;LjBHl8yLDJT2Hs(ge`ud~h1q;5*x+rMvA$q`KdA$f=(7Ssp0a+5E` z82%$F+24QA=;(;y4-QKW+VxX8?XNNqW=hZTeRsCE?@VB4qD!=NbW?xjQoS1F+t8yd zR)8huv2q+5kVV2}JIBO-e^~7Sz@41u&618>!vA{(ig^@`F4avAUk7mgEDQl}K=lyG zJ-dL%P<6NatLKN5YAI`VTDOq%@w!RU#>uCVL~yqir(6o^3tXYJwEQ#$@|n}LHTQ+3O0(zI<1pczBdj%<*0G#xlY+Mla5*$nN#M<2KE3&(A` zTClksvG-AVAuT-L?-!TP0A-MJL?Sm5$C?_Tw$=*dC3l_ZK7Z`JHt?S`Iec6rb)2b5-rvU0`3WvA8xOLyNDq@L^dxkP%h?=S&8Gb+9c6BxbM!slXa{8Mr>f=ga7?-eZ`2jAC` z0!UnQ!>=bx4K2b?fp>=hn_9s4GpeiL;jt`)~v5goFV2_c@=_v z3$HbDMSVXEf3ly}p@EXp-#6Esjdzf!;Rg2FA|V~T)Tr*q$fNj_b0nhFfpzP_)eS}p z_W{lM-ydP^Cg;LWjQWRy-4Wo|oZpsq-!{GwbJogUIEG?90Ze?~24X3-fu?OB``hjn z8x5&9I)S#@pzh5B!3!bKY6ul&21tulvoDKD__ENH%y+x}GWORYh+SA+nFg4!&<+OK zxzj{=xELngsIn?HB=jb`qME&Qq0ftVMxbLBPjM2sAuyeKpo}H8Pa`tr^SugebSMT(Fb-uqTc!^Wr zONUB4V{ch#DLj}dy%ZfadxI1fBTfiucIFkR$RxjEKUn4YYBcQ9J(r$aw2M_4A&yx6 ziLqf7j{jygpy_OomfRDaA#cieD(6@(XH;IHSos$^BLQ;MI_;38S!z69VTBt#`N|3W z?d@#}n^i1vH1j{!b$EoS7jCaXgD&^IN^Z&eh!;F_muTuxK&O&sD(n{tT@fR(_6iLh7-Q5Eaya`f&tdw8AI$HF`k>-Q$(voWf0L%KP zDtw(vpS$oA1lBEPMI$$SpW>;LcVB9g5B^|F%A6!@KaW19)8;jJ^N+=eKPn9#?vGZU zm{i|=C>wwThY(tU2#w=Tq^kLo*>-tPn_+zHghTFtz4XpzeeGo{7d0dVBECWl1=65| z&Ab?#=tfOcB_+(@726qai9k+VmTn0pI2qUVRJ2VfKk zkcRUz4~xssdW23Ri->ctk!7c3dOxS2>ox(xQvzF5>P{(b99o%gkga*g-H zZBJbqRDRzjcq-eTgw64I-L~G&k#SzWe(bVUc|s47)MX=0m-T^{kNSFZv{{DYJ*Vxw zS=o8;IKC3IAQ$Iso+-U9J?@&%3g69Oxs#-a zgTb*^^Pt#Gqi?c+cN2Py&(e3u7yjkE043^*dFJhApqIgm;-?##fG3l9j(?Lsvz>b4 zy))QA`W=?4D&^N+X-b*5jW&s0euX{%J7rToeY5Rd;DK^&)d8K4{VP)ZV+qO_#LL%K z^tn)M8S-xlEIg81y`7celX#Z4pWc19va-_W3S);mx4$~~mea9x6Kk2Dtn|Vr8lrwK zlY=4m3j2)N4@=s;ta6_4NBN5_-Z&wk`k-uLlE7L*B3Vh}gK)^=n(~le=O)D+(~leV z#2ud~2W(2dL|aroxYG|y{vper6Mgmm1BxY?^+13>!a7kzox9}j7mcl4r#G0q7wN99 z{b*XT4F&Kq;A75%C40^$CtD9**j)EH=sWXAM$zB8YWt(zkEEM9j(Rvz)uQ`!C3>ohmteO~_|<8cw+y5{-7Y zAh@`StRwWKiN#}v2X}P-mMiJpGrmH$_1BV`{~orKp~hAazWNdHJtrb0m;~;Fxicic z^Ez;!!qft)Txb}yxWo zJB5Bi10KOkD&`ihBPFYD-E(6vknR6}R0GhqMJ(2Clb}3RIXtm&O3@Ws1or>^Hs&)) z-O+Z;$rE5zXL-Ov3W7Fee!1k5*aw=`cVZ=T1MH;(%Fy_bwSLjO8LxO$QhuggVm`iq z(m;T)a{pTSkc}x^evh4tW@6lOwN$GzvDsZzUG~R_z*P2z^?&C zi?F10f&p*TE@B})TzOlKx*xTHcB)|)CY*ssv(QD@C$wAoxGd1c+{5>nm75hOB3FUY zJzxAS`rS3%nQ|69g#WWE{_i#3=U^CiW&Z!7xc~RYfm9Ce|Gt!$z3R=MqN2h75*#BO zcBB9MM{G<|Qc}|L@>d9tXJ?MAhn#TzD&*gP&OU6kW%mktj{SGF&|ZMV@V}$M(h&5h ztgQTCAtm*5CA^2?qiJ?8E;+MgU`ugH$;GAU*RMh4jB~ON|J@C8y#M>s$#q*r({DJrzF*+X?D2*L50BQ|Vw^Yim5 zfHf)tI#G#m0K!+`Sq7@ihH4Bu;Z`!759gSxW-Tr2Z0ED11nvOCdvuL>z+XeIOVy_M6=%@g#%*h(X$(*gn7y zF8BPzoUSsHA#~r2+3rm-RAgk9|9eaz=+knk^C5a~GXGx6;^ipJ{yDtK4>12*;lEzW zu_vsG82v1{267hNUQtjm3|K^%_Enred4&i$m-e4qRjslADd;k*`u$~eGe(=Nu z{GjLMcVjnv;M1Eh^QVVf?P~J~AQs-}_FeeL|Gh)S>~eovyV5uaP4K~?_S^Ep8Kba} z!}GLOcK8S(hVagn-{Y z$H|mYSs9GoXS4Xhrh6AR_^#yGb8+1@XUV8%+y2en+zGxT$_b+I#Ddm;L)yA4u$@Xo z

l!W=FVwc$2>?nl*6I0rJN@Sq0y_*@*Abx-XMqYRdN{<2>Rn#U z>mj>dm9KrgKy|cUc&Ea18JZbA*hY`js^`7?cKvj0tQMJX`)tbF{IL;bsfZ_CbCp8^y z_mAONVh>G<(F5gS!jQ(am{hTUmfmUzMffNM9eZ|Xd;y_Z$(9~d~j)W{U@en|NZ zJPySy=(0SK(cpOY%xobFe3yC~_Ak=LR>_${;|M|WN<-9zAQbHVIg=>0R1P~JuyYG(TL!F>OR}@c zGF{g+rzHt&BiegliYzxL`KeCRk#QGNHflKQYEk%6LmL31fu7Pr?3s0khP>yu;r z<^dP|3JddwCZF=08pi5>ey9{Lioy9OKEY9EJw1rxD{d_8EBNv0Md+D&k@y41`ly=& z1;Y^3T?HOsqASpstVp$A+gw`^ps>@>(Oo2bQ(@uyCu~v2y}t8DIN*U~QU5wPAd=Ya0=WMxI{3=mOj0r`)<`xchq#x>C6 zWgUCl#hM8yQ=gXEx^rGCquDdJK4uUEkmRy?@W^{%-0)HRHrCpY{3V_<#n9UD~$Y)9W$hr|*o86y~D^ zz!lZFNJim*sWngEM^`aAos}jyzbevL0Ww{b0V;8|sK{g^H|XUotGB(Px)^;{^MFbK ziftkVSA8|L&@Hqqh`=t5$M4wM<}V0Q58kdjBdmL#Y<_tpe8XV8F|bZH^APlBNHTV8 z-18wI^mcJrfY;kj`QYM`8no2$Kw;L5WGVwnXcyn4j%ZjHto1RFW}d2@i{t?DnFSH| z{>%3jd-N>z-EYk$NVa^jKG*tMzT1Hkif6z>V7Y^;Udus#G+IUEL@&g&+CFIuMPkH* zc*L6k8X8OMuK1gKeiik)oXEL^TB*I%8e*(D@;7-+uTHX%?_7dTs%`QeXlp?0S)Xaf zCmH9-uTNT&UN^M!T|P^n<(znS)QSw$ZC1D#x_ z(Z4yHMYZa!g7-g|uckE-ZLG_O5=s7%gl@y20@-OFb>SS3zrY^{#VeO-&5yzc^0oq& z6_OhCW8Y0u$^Juy>H*5dxUoYHDO z0^Xl$BE$@!XPN}&19t3L+pH>ZELr)dEazICrlSrquz@1m0gg)$7=&I3Jo!Bnzi|um zj<(I2(CpLlmpKEYx8WeGfS#Z(D;$rZe!=oo-mMv4?f6S_tMaz{!X2KPDZH4(o6=AH z#vm#Aa>b%qrpjzybtiLM?sOpN*=l`%bw$@Jf>3{ccJlA#i9{cD zdaQ^CSlifWHcGJ5u=`L5wN{><9X1`(r44P*?r0a&&TTEGnaj+H(Wfi=;>(Qkd`{Wu z26DWo-}9WVXe&ETcMIonk`R)58`Kd^fDu}k)DUU8G5Kx8KU9LOOnxO?LPOi}2)t#v zHc#M>sD7*O?wA6ZfCQITW$EL?C>3JMAv0iCti-)Z4c^gD_f8oA^oxPm8lPn_!JOcz zzOgl*Rq@CAua3VTJXt*9`;nvMe=xb&Vbw0Yz;%orPwVcA;`S>!13>jZlDUtjJ7gDl zIOdY5?F@WrHv<-O7R>{gY64wX-=_Kf=9NhR3rRQ_#Jdat6N+6hI&XC0adB`5!%^wy z$T?U-Qxk|@q&pE97|*c>gF{d;e;fE(XuVv)&e+T8fs|l{qDpHO*s41KGiWqPA+&Dv zNNP?jNF0uTIr@TbCz^7@ePP*r|*F-9$jnGO#yH5?>lrMcE~L+_A4DeCRyCE2&fpWOn$3s zJlik-jeaE;X4aBs9`vo&dyD%Qs-1kH6PP~d^QlLIUleIrFdq136daC&pmYpmv^A;hmVp zSj!C#A897Owo!zx9wihpE1W+ee>>)q(-`Qg27V39D#}df^cuGeVvEIKFkRdN47`$; z*FCHH^Y;0eVEmUQEHi8ZMIRblAt}TQ^o?2KuxdT!TuDv;;s^dy%|w%nKddrXnr#XH z4oMD4v;#ND556eQ`L-_{G3vIq?+#73#>(^9DwyCs91f}ijlYNaV>CfqF~FLcb}la+ zlns-IW`2i%oaUV_BB$7D0%dmIx+UJXQU3bhI`scONZKC%aRm%n*|JnXC1b2$@&GHG zD)xV$x0DtFnerVuuZ{71o8HuCEn$~>zJK56yD#tkrxcv*A3&pNpQfYo`Oj-N9#-E0 zefN8omigy59K@kK%9Bw)EB6s=^s8Uo7w>-0k~Jjd@Xn`P@*acH3!@<*@2OF`W$&YqzG{Sdt7R9aj~bnd*~(ebz+_V?W?ku8ylo+oKztzoY z(;U(OVd>tBrM{iTctyUA%lRB4!cXeFKEvrf-+cW#o<=+D3f`_~VBi5L{$Qu=1f`VM z6T?ZGsZ~`~pZ)fqe>AJYE=wHWg+zkZDj`t>T~+BBvSwgrmSK3WhsG1OCI77F&{ z?;Q1D%hlnxjT=c#3s-%6j|qFLD4TaIcsxGf*t9X=TX5bpDhT-tpg?;fE)IH*7)~>V zYVDLSASWf9aqjajhPfL2h)QApZ}JVU5*>*8=&-TNI@0@~I`tW%Z*%M(Q4&Ktk&kX^1U%5?<5rtS?9gYo^%P62T3Chq?*638%Nj1;_*{kL1VGi zKpSC%C#pn-0}lSVPIXm2D8x3 z3Kq!z+$Gk>v%*I|>=@JjZ&8b=OU65BFSxWtJ8Y+CfWzSo5LWCbs-KZ6C;*#$;2*vQ zHsao}u9&aDA28-L{XoW*kU`RBm==$1$=>C?mJ@jJYx+V$VsyIYw9V&mJ{W54^l+y9 z99-?E;tfD3&SNRV&aeCeCII0A3~?8f*ay4Sp`l=0S|@u1l*MN77e$ej(?;cugnwi& z2-sWjN*7vupOpx=WGTG%^m1`tP_6Ai?k4MdmHi+08f2NC`g?{-;=_4}&C62!Gjy_( zvkh{$RjjCU(|zn+oeOP|Fg@j+SC;B29XN;^^;~|9H~So{?Ie5oJ;E`9kHzIMac&E+?kZL2@x$G=CiSZmM$z02yk4 z+=KZJASi^OX6U4LNP%d#k|XlocW-D3I423K=ZjuxHf^oTpgyyhwEpD04DK+kNICU z;=`a+di@kp+-I4cqZ~*mOY~jpuD07&oS)LRmLseb24>DBbqj{05|l<`**;TyW{2Y= zRW+v~#;z1V_W7DRdy^FlH~5Yons!g{>8mH`3^S2kbP+t@E14&IYCf0X9tzP7)T~3A z6>j7J*E>(fJ5XOs>Y>-4)ViGNcNywB6L_|E|B;zX2Tr^K_i|yuQ;5N z4u6a^E81$Vk)3zG7bZdtQU|=JO7e&8mv6trBARL;6MF;O3q{j~jaGPOFO&W6=^;)SdS`4HjjFd>++-6kC;i6dZ zuc*)J9;%mHqdPG{NOrsn6n%@WL86D>R@^^l`wjXSd0((kP|LG^SllXO<$>(>&6AP5 zvxN~DOpt;fAF`*O*xXBetcibmKQp!?w15x zAhTH{>37XYgyuK4@>mL<6ULed1e$U?d`QLckAxp_;LFe#@U+Tp$>)CXIY}0-Ka5`T zWH{T7fBUghJynP13m^`ZQPV=gAeAXR{VoOOn?}p`mJ@{m=I0RKD|VtI=gE8q;fT|g zbfQf5OZQ+l7~~f_;ub?HQ}f(?Pey@+t@i zRhny{a{&Z>+&Pn?wm~n*OQ{ylI{3lFn`9P7uCBl>dRvK8OPSx?#u zvW)6VFh+_Ca?mN}`u5RN&$+ZKV@f^d#T!Q{m=Vi-8?4N1`R`13p@tXgf0{xOm~U0dAy%B5(zB>D3#x+TK<<7<8K92i{t4nAUQP1njzk~7L%-@u_Hx)sjQbC z!UJ&Pc5Mu$C4-oOJVP{I>(s)vwke-n-?wPjDQBf2JRrvT(HeQ!CFJcK6#9B=#jsG4 z72+<^twTWlBoQkv$w(;FJfLhb7J}EQ5wE zZj)xA!EM{(BCAX`I4mCl!uerf5d80O9V?CB(UktJ=chC0$Eb8U1G6H>PnCVN4l^jC z3_bn0Pde{$avrh$iyheRtE^2&5nE|e-3ly1>kNU>cx{HXFmx=_yPI$t=QjIhrfXPl&sape zG{=yQ2)uktNve3ipBJjF6vkgkxcXw@a3JVF%ADnWcwBocT1CK;S3`K72U}=r$b#{P zXd`0b{NkcNDdygljHg%jZ!lde11ZB(0KM7QDhA?4gHZ)9H|5xWs%Iue_TgKY1iVHC zr1&zvt0Tnln=lL!;g*ORo(yLe%09(T2f6u^PN-LgoHSQMry3j0SQ(;pv@&Uodhp_9i zdmxo5zxUDXI>=N`Kjzt-gAn+Gqg%Yu$V7Z{@DIn%B~O9m%Nfnh6T5o$&89EavOEVf zte;COVxGYJSFZzk*1P1#tO`1eu(!pzem$|;KWfl8YC99o^*L6dgLm^%VU^yC2rP+hjtuLn9{)?N}5Smc$;cg<;-QUU2c#&)U1OsH{`Rg_&Gi>!e^GTZ@#p z>I1wPp)BM!_^?43n&zW!x-9Bk)#q{#E#qN2ZB!aa(ekVl33+8(W4t|R)gzDY3mD8%>b3hp+Ko6J{hYhUsBb)3 z*1i^&P~Asb=IA|@_0p<I+`WCICIO$Zpnl-GKi^-;9b)IaXNLs)Rbad{>+3OW> zq?rD*UahU(c|l#2OSQfiv|q78*JQ4#KtnzF4?MU8c;y=$?us3}n9RjT{`x+*!AOC8 z+{<8*5!czbG^mslY3qV)V%rz)E1&53os0h|S7E_fq$j=`uO^<1aOC7S#-gcu-wGYhHiTU zrB{NESMQS>~o%BE8n)6oM17YX(A} zB9mWcOd?l;uwRMs%1Wr~Jv;?sUP|pWyR}Lc;*oU1O5(uPt?lOD ztfzXWb3_iRL%;OIV z7{iSz@90@uDABITQkIftH?s2SWZTWi292+`Cj2O`Ar6p7@n#8#*e-=uqY-a$deOc2 z+uJA^@u$HEnpW}k=n5anBs;yM7G67qJ{!~qVNy~oKCBU*PpJ;|Ond$MDC{7@4OXy8 zs$}9Oecr)8F}GEJzO|#HkGo5FzW2G3cF-kyf84rYs=~6Zp6_=m}&B@R#a=v+n?1Gkf?epB3%hZK?k#mFS#z)0DpbS7^ZNE-&syg$ZyK? z3>%(YZ~IG>Qx?XtcK7{am*E% zHMJuFeTFk})pq*l8H+vachM^g-+Eh~lJuo}{Z<0A9wPn_{(y1!*{KJ&VHUqx)TRHkwiCymMZpy?2 zY3*w)Yd*n~r)%~Qco8qO61X?*s0H?n?UEj#p#@OBS11i}%sXhli-gxOES7Fce!#Bv-1Zd%L+;86gH6ciPIww7+%VlGMGUyKOshWKUmfW@ym9rx zc+wFEhlOKs@|l3urR1X()WrDOY&h!;hj#+_B=5cDXSAV;D0iOC*?+|>I<~X1>d7&c zdXb;~c4rw4Od5d>D6vQvStBVCGnl7BoP#x>pkB0&YjPLv+u;Pv8O|!Utj=#Sl9M`3 zWTByy5k2-`YFc;e5AnNlBkJ8nY*FSLBnlomr|6Z+`kGV9k?p?JxXg1ZOJ{&)REB~6 z!Ra^35Vo?St7VJ+ckjtlJ0dJl11(JqxkrscEu-V-;&t>m#^bx43heFP2dI*`gLe{Z z*hZ*9NgZVC7-X*fX;+cBp-P#7-pEk-7*8BTw=c8EN0gxvzO?`Z)qg)-u)tr;%|K@t zJ>KlTJ~i$wPMNr*_t@euqYaG4TDhy!MGU2h*Fi4fhYi9>j$2KD{a^@-BDUON1tO`1 zSCO|NCoL&areZ-oR3}%p5pI~Hkbr_IJ}c^rt>Y)`1SIpGEx$gpmg{aX3wc8 z5;bHaVdC>fOq*x0$BG!Hze1vB8&is;LEw6+f@n#mG=^-*m1mp*B8Dj(ql^N33Q1q! zHtL}{TibUEzJGXTb$y<~NvM;JiWQKMIx2MFE#$Dn75P`h_O>lhp3VGyP*#t>yLGJH z9!5$w4T-Y!H>;mv;@T=YF;S~cXyZu{T|JD2x#wbhASR3LONNxzyoG=RxZkj3RgCr` z+`&Y*Ev`tT&JEtt2L&To37-aD6LAM3$f`cTaPS67%5X8;EZ1Bxo$c@3>?9Az#$JlZ zIPK6Au!`#R+KwAgUmUl z2mMhl@eO%s?`=9*91XWL;=9B&I-7RJaSF|}F?%)# z%vMh!bK?-<4!t`X7?M@*r%zTLH2;=1OOl;NCKx3q_PY0WT~x|#fpZuuyr6WnVXmU@ zgWJ@XXo9rAK3b~=VC3e3Yjy{dpvPCQ)>@)sX~qXDaY6+FS?0s@(%GWa68++9hbK}q`3w;oBk|(tFuQO`qnMl57dSeH zm*9=7Jm*K9Wz;m#JK}rd43MM*cMXMB-(y)PW+x__sBTqkE_K|s6d3QRj=yp@55cOM zu6ZVUwpw^7|C>aG@lAU$H#A#LATIb%_^(Ohqt95mFGmjDZ`o#iIg_GPV=v-gZOn-} zojVGxf0I)@((BY zjG-90PickpB*|dL+%Q%`=AIJ5oQ;ZyUk$JUl!8ioi>T50i4}2$*&Q|E@F!=bHx!R5 zGzLHeXlaDU(xnwmo#b+{qVrSyHJsH1Qs|+i$r%{G&Zq6PlxbLJ4<1-@`c(Xm2iWa| zZzS@`xo-#a1E1g5Yv|V0dYB8$^%rk$r2VzjmzyN_AK5YSdG>3(@($5(pCn7`0(Yb$ zGGx}D@&VfYO<_oghK-hXSUfkkU9%(QHKqR0qfSREXlQ90x{`}8DZcZ4aqe?$vAk-8 zVWap9wB}O@idXME3Z5B>Ut~1QV=M3L;XfCL;Je?XlS!iEsQxUY9o`5SaJzBiFIlF= zlS}O#Rg;-~AEaK&5eky$;QHX2DzQk>cHLKl&1zf9VN`VMy#_l|5M9f~R=G{lESQr)KbSLmnM{e6vaxpQn+<3Tf2}&U# zj}>b?DqzhO5ssdXzMGvU{>l+O@P=3sY?wQsb)kUVa|{IcrM7qYr}u12zW6k?eL`~c z{V~>pRQ9+<-?gk`q&H}z6LEklslv-@`y z3^?>N3LXzW700D`MMz`E+ZHI-ko}#)HW>F-K^{u!(uVfFk@4*(*gw2sBL=)HL%E4T z)pPZ@YPtpLydov>@LYF|prN#RIT;O?H4%bMcJe39rcgp+5T_^0oNan@MpA(_=F^3z4}RtI$;rX4+1P4H3GY=?^8t9HjiUOxT5h3P8;m;*t4icZ#jA^ z#_@kEf5`(d2*Adw4LNZo?;~(lhOMdXK631YqXmDFKamHtxMuhg7fw3+X8Qf^qt$qY zKif9tWi`C`^fMK!5ztfIbTVAXcbnxquefIc>l9MU(f(!$je9}q$$x@K8|u`!jQRe6 zASp*v?EsMhZ zmy!KX#_upZohA99IM;2;xLhs@JDC`3TuP959<`^XNB~W@KO^+{@y8S%vn!=f0(Lrr z7-j}5*!#I-zO%BkVJa5Huhl3pol`$bbWD+e{;!r0gCJf6HkFy(Bi@Ei3bUTVtmv@v=5*~r6TYf$MG@g2*+Na} zJpAgA(66wymm?&%`dUt$4>*0g6Mh$9S7_|;2@rcAu_nz`h8(^C76}p+8 zU+g6UgYMn&U*m-46e+Bm7^e*x+mO zc!w^M*p)Y-;@)7eb^YD*SKV`^yLx7^ zkzR>9X(1w^a0a^bcX91Pmc8+uK3xh0W$^Feo~P02VW%)K^Put0PvcqapJ)%KUDdC> z_dwFNaFbuGE#?at>mMeYisRP=zIv?n=a;(;!i_S=@fxRwylwKm zXA-E;o+vfVHKgMgGu-%6Bsj zXOJ0urM$Y&;fYApu0LGh@O>HGdyK3UaKs*SgvNXdGjfoscpfF)&XnrW@#fw=vwS*B zEq?O*7?mOwr~X__*#fOiWWd7XK8nuuU*l3g)a!A57vBA$-W9AZ9bCG1ZuHdtMai=L zi5ZpL=*Vk4RyC4nW2FAAuw1pyI9o9Y%-&b>5fNCjzfN1 zODirIBFw_QqWzKxaWZsU=d@z=gNNklgJJp7oaO9v~ab6}`v^?0$m@@Qq6Q~Cl( zMN;v?jZl z?-Y{A@K3e^Hfcd6ZWeFqEy?-og_FBh+n8DjVYD}+IV%-X^yBRcm8>V5O+$91aCIaY z%9$dX=BszVTGwNft#!q=VX?hh(q#2opy^yUk7`p|vs;oOlw(>G5W;FMI zC4edG{-R`c7GeizKWP=B$K*{gCC+)3!6gn854i;?Qw?{t?L$gnsR>lMgm;yT-JzYs za%UBa5*R|LaU6j6h%))X)uC1te9Ajz9fNT#&~jU*#CdX1)V}Ut_~3@a`U*$`zH6-t zn9wS3i5FNd)EHsuXtkc_%9K7ra3q64|3!G;X6X%)1=P;2T*Bj0e{dY)3Q%Mh0i zi8SV5AcqO{IO$)EtPJL}qOj!emMXYYfu>CsSrz`ocgN@e?y)M%KR8VP$(ow!CPP*_ zrn_keA$UIC=}vjTGOYCi?@sewNb+9bOSXVe=}@q)>Eq+w zV-t0K1?JgxCEOZae?f80NDpSTY3q_?_p3$$rt_=1!9*O-sm6Cw8KiuTc3)t9S&AX& zY?)MZpS%TbcsYs1dzz$J047lXg}7tnE|9Z@U39RjDigx8Lq#?0RZUw_&;h}E>$szk zdbRPkEWT|0gq_U#w&qB2bP~r-i@oXg1=OYmCS5c<1g>GtRFKZQR$1BJ4&X3zX=lw$ z7hzXM9fe%F4e=b7iHe%i|E$-#V;h+w(Zc7^jO|T38hk}^p8N$TN!r}2O41BM+lkl@ z5%QNugclj-5^<&v;;&eD>EfEh?HFvR;Q`T08vfdiiNboWYV|gF3I@ULaX}XbHVEwH zpzhR5*97~7S?x2k@iTH-8>ow}5Eq%lHAaT00D*J+WjF>tYbF#Kj5Ch)q4<%pz`?=K z&p&d!oVTS}GJfXwaqqhoG?-dE$ZHWZ>@k3vJ|!*A9afqp(gQE z24#5oRbLqrTxUp2J&Z5;bBpQq^!nD5&_o*_ZcAv#bUOREh^N+qd}|a49&M6uHGmE1AS@M!ghEl+`7^*SA^b#_os$gw6k2sQ&}_U?-~)3He`x>Hh=D z`G1%c|Jj)sEuawhe_r#ym={A35|r(~yZ65j*vUZa`~Q8tmH=qv{GYGz-xovM>?!_# zOqT!tW88)R&87J7KNd11-d~I+0*choLjGO7yLVp_9cHeIhztVk&i-~A>;r(1$R95x z-wGo7*x1@?#fz`q_42AAL4ox%BsHgMJx72fzz2FR`HeB6{P>T(#?664*JJ#m-|lzW zTb^t8p9JpzWS6qV0MZBl47)f0odbmd|A${)^KaI!@aJhcO;pDOa0K5!4BF^bxMyPW z4q)%@A5OkV066q~fStKVP6@rU^2z3ZRw^#^)TxlPN|}`ZE=zmGvk{>irWMWSIoZSv zl*jm-t@-#)SxL#!W9{rfK;4iB@$zpov-czP>UIa`IcIsn7)a9v7iyP2E-lG3@ERY20v1F$xO z1rC7)&^Jo>1GEhR9YgoaqL$JRxB!G7c_yj140KKFI5LZS6xk^9Fp44^8?nlY5 zgT(h|1nbk1rwx9qnGR(;`}ZF{e7FGlpQ{orMng;c*>l}QCDN*tOUcVny5j0(X6CE1 zvJb9YxpFpRW@ZMs!8m0V)?^}yOR~4;@!9G3=;Zko5fcN!{yE$7vl|I2O7Jq=z6`p@+tnuI$U=~Z_ z;~4a{w6HfNau2UFOn=no_SLY>H^0=VdM0VM9cxZ3 zXatDiX8_R^q!M;@NEtz{m#g?|i*x|6v;Jpu48GN6{a2eQ`t}@}$;%1ihrXb3**=J(5(BhmM-6 zaeB1y=sy>>P@=Wy2LR`hV5}b!{M^Q#cpo=fNc%($9oBEw{Qg&k_*{V-qI8yCk-CEW z15-XyC25%-Hlrk`4TA~yW@I~cUCghFiIMmPL_}CqC2E#0W@_F}7a;Dz{X?E>qV z-~Q~BcA*K7q4IdB(Gxv>JDe)6rCsK zxZJz=o*EIsfVBAMhX>k?xBjyM*spWQI1a{&HI{&O{v9xV%sYKWGc?Z~EWmjRP;&d7 z8*!#Dc(G*YQftRJIqXb~mPckE&}akKf43K~C4lXvv688OfO&*UnHsx9V|QQQ9?tGy zXN$>DaftSt|8?R|S;KEKxp6-B${@Z>o4kS~nShUGWd2|YKrL1lao26f0KnQ;9z5&q zRGj0L37d{~le=(v0@re(&%vqWKIL;d$s9d8orM;)5 zVJYuB`;Geq|-CwsYcMvfU`GrSHEELu1EkxjcE`OLi4wOg$2w7>%SKg`}!sF#- zxN(*4=qBN0?N3zTQ0D_}{Vg~B?bF;p?}ZA2S;1$OP4pWxRW8RUn?LC{8`)a-qX%+7 z{`@`0taBd`qdM^1V8gMJRU+a7{7jCgDqjD8vL<^XZ>EITTUd1SANSU-}*;j&zd&)OEC^r{%B-p&TWXJli&1j z+5UvLHS6eS7kygEk`vV{#SVi0DXev<_}e?qv~}ih@0_)Mj=(LrQVnfF!kPZY4v}Ln z!8^`P+e2Q9hc)-QqM_%qB)B%ug}eXGHhBR=q<=7sqSPm|0N2=DY!k4)@lwmr!zu(2 zgP>k~F4i0XQ2(POQR%VsHV)c0`J5@Q&57e9LUbGg|A8EqP0AeC02&wADwMAQ*7ztu z4U_?`$LEVZucvjRZjr&c>%6!A06GfABo1I21G`vS{Rf{dJ%upbx${?jfV{`bY`I!8 zoSq(Xj4Zihi{tZipZOR5dXv3Ofy9SmiA;Q@NIqw{!}~IlUZDyx$3rh3d^X?4l;5| zji6IDl3Mv;v%GDoXY<3j<_q2ch&ZhtA9~~IeG=#)*Fy1ubP<#92{@OAidE+LKwoU0 z-A>p2hV6j63hBISlASf9{?OK(xjWaQ$&TCa^lu5|op6Ii_j^Wo_<}QUp9J=p`3(ew zs~$Zx=?uF>8{ngjryPRUdTp+aMBNhlnnr@#D-)ruJ|bbsETS6c&E&8ZQo*ye2oA9o zrR9H6But9F`t{~C)_hy~%C^#%Uh(>3nD?b}1*m|cLT*ZZ zXA#pP+O_3pi)xyG!DDc$cBN^f&4*K~1sed6)paZ9Hi?#5?iJ+^>+JzP5`n%e0Q=kR;XcADog8Bd_3yZ-~w(Ky3DmebNEaRHzca%%8l9IB^TYrcDaWiK!!SXQ0({h*G35#NOFUGULgX z5IbE^+5tAgg#A%AvEk+7t?$FUMa(!PHUQ)H%^C5O*%-4cWkM}{d8>Qf1u(OM9PyxW zbrdazSqL=JQhG)?uGqEYY&BwF;kWQ9j#XNo-G#*-^-wTqmKQ&Vw3&L_`p= z?`x!Yf^-XIs{&iRE{L$037zB50{AZ=RW_!L<_5smB3KlX>h)lrN1wyRov?nz95?wo z^tp!k8&tn)`0n72O^9Yx_N{3I#s21RS@*~R-i02JZ5htwYlm2fM~HOthBi^3M)8KZ zM(5{-a)jTI-+*?>#h=)zbPmU8cUgN0hQlcphommHg>sImidc)c^YXD!QbJ$C7va0` zvaeAqWHj@TaQSuwrP~}Y?GU^r-asiQi1IHeCEG0D8iq$j**#KK!^Wsg>m&| zbeOotL;f|soA}2?vi!VO4?Q3Zxj8cu=@u%7TJn72JN|#74DsP8w&q0T%JmXAm1tkl zYXi4T&y$EB=_x-#q!+lzdD^)wjMUkasi4U22W8q5iH6ZKP)W8f6XDuXidMWbJofLl2)+p|-0sp!1z+o13ZN&8Wf65;FfI*dad3sG5XA(FVAuL*Sk9_sT zh)MuJiwBq|=|o^YBibaEETGyvnLCdlIBv8UmK9E*H2 zc;c*b!D0~HABWsX>UpwUnZ9d9$y~HtN88CQKS37gFNSdCuE!xlBG$hSS!x@#s?+XJ zw5gDPA!!H&ovNXEDaj!gecS37)$y3fM6rBX=Db@UvzmEB`ej@B$m!&lHew?)f}s^Y z%9Iy8CVahYHzhv2I#JhXV-mZ}hf<JxI418;;)o_IwiVP;kApYkLnX!--xo~r7tt<@bo`mnyW&uGKIOe z=^0acG^gI^)bTJX0UXEW!gc{~+T ziSz@S+!8104wFC}ycTf%PAzUx2#}H6)bsm_u_5kJw$7NlbugbjnLRs-Cw_<@B-O9ny!CVWSdH80 zu9?$6-a&f4fI4wfDaj!9m7D1xoe;bRK9JtrB`DuN&_D#R z|3lQ1Gh0lnT1G3cPG_)6;8X)31ZS z;=6*qqiJk?$>wsVnMXdR%u3^1@PNZ{fK#1a-tOZiUm-hUz2%GC+jd)@N6*SWdj=lb z1vd5axyQplxV&p|CUVP&28ZPwB?d?Ymb)CWtQiYc8UzN3{b(^Dx}4sSijmpQ-&a!D zmWcTz8NYq33ByNNH_b(@2L~m+m{iE?Ig130^yjTY{{D8Y58}-kk&sy7R&xknsie?| z(1r=z@3;x%p-bZFeN2InOY~xAId)c$<5lzAziKFET-MF8WX|8q%jSIlK#o^vTxdgR zv8R%KtBvQ(H8#YZ+p_mo_BAHw>+~@qCQs-)KE6hxlmCvKl`hFD(JXzj^%@z0P-~v` zsQ8U=)L609ZRM*(fp**XRL346v@|`tv5N>%?HBz;(awHPewGjVX#W0}_P6OvUYD9A zyR$rC>=zI&=Kp7pSq%ue8ai3s?9|CK(B+e-`{1&FJz8hvuUikl-PMpWh`#qHtBwRcW2Czgw6NUeJg=yHbuKa3#<`ER zOq3Ti0@QkY<{zKQRF4Gi4rz4k=%>7=-^`BectN4H`Kj6P8_VL(XHi?x-~W7Y9a!UU zeQhM~V0rZ%9enL>yrGp_QAFs(-8})piI)~vcR|dY$sF$v7EyU!p)0b;v#WF@f8`xv zFXw*#u%CRw*Y`>k);6nBn-Tuk>uTdoOlp z=n5vj=4c_Q)h||YZ8!ygxT|V_-CvHce~Z6-pKw2xlb6YX2^Hm_-xCA@?$ZRaBjk`);T{-U}D)?#>!l`8S zF=&K4{!;^TyldQsBgvCfZ72sTEfE*u5I`r9`drUcVqH^9NsRD$L%x^qfb(vkAYtUz z(`zMJZO%}KHlLv7Ri zsBzy|TWDnP(3L9ofio|*rv`|Pv0^AP>DcG^YrxYe5!Kkk zXjhhp@}d~2;4kTApIk%+TB1rCE`4uUPmjFRRB%>A{_%3=PnpAyAA`!jog|vswAJUy zbuWvY=eJ9Z5we=z^uL7!_k4!@@~0@^KiPH>)%)!8s{lFln$UbeVWdnp5SZFzy1DG) zqte0+0>a3xk8e&-tZsM9YW%8*80z+F`l#TOKB!j(%uMtbpnd{&}deu(4(Xg?#eS z!>O--ePu;n8fq)bGwS@shSVEQow#40{Z&FvkQ2$zTxlJB!tdzXpWS6DyqnwG> zZx)Aq6ArKWJ0IL`S2?U2Hm!VUl{w&U?M*RMFyRnPX&# zd;POD{WE_GSJ7Kh?To4Y{P5;ZTMAMBEIVC7ujLQ(sVI+hlluNoZK9>)S6(`Ve}Y6i zu9wIh>t|8> z6%?6i(NeE=WB*JrF;95QMbAeHA$7t_@kA5asU!8JT%yK{@n^f8`62f($IOjY(f24O zUFk~$V!4mY+8n+co|bW)u4mijDm~sPQCeCGxV~$)ypw#onW))z5&6O^!$tJ|-(u(G zH3CKn?xF=*uVr4``Q9{DcpS{5Ol>|I2o3yU)#n_TVjheBImD_Hh1u_wo)qKJ{6&iqh~9UOx*|b$-z~o zO;^G9M>oNd<5$z)WF|5h)xY7HSD>dA?(Ow5Xg0d=kB;;0yCbuZ^b{ug+Xg0=>AC1# zv-Kt&Y`EvzX*yq{)eCR*EMndpwvWIK$fEvfT6@91(HdHTch4j@;it{2ykb{qc$=+V zZBb7>7;ZZLGTii;K-#?OshCLUtB;-7NBQQ_I-zha9~GZm%Qk0-#=eqa_*DLFDYIrL;Iqz2 z^@z90WrZ=H_kmJV2`i631U0x;P<4Xl!&F%k@^3C$%X&X8%~OV&sMfs3IkbQO#?-Ud zHJ6*tQV+cJh*^;mI8E53HrV-aq>qo$i@+E9wC8;eZ@TYQ(f2gs|0c`IO8^`fT2qyBS zEKkTUzD1~PmL_YrKs)E?zYpVW9_j_j7dX1e#|GCkA=lT)EU%6PV z-h#HXqO|oe+n^?!rK5xG8dZYSeR>^Z8>U*@m=Z$xi!OtnNH!%9RJ2P}n-|oNr6=te)VZ=7couA=$AkY6-*_Fpbz4iZl)xA;37L~1$ zs3=#-mZdCFcA@Mv*t5|ig zJpO~XDfxPzRMgzi6Lj4IiHiL4z?xL-ThXC@$o_P2KISq)zP+XwEyHNI4^$*$1_m;d zOC)B%DSz(F1*PTUe!fGkB`Sv2NkU}>ZgK|Kk}~~z>-bi)5Rw#TDM4>ij`MOt7Ix}q zE)@C5^d+c`60PS`D2&gi;Z8f*i!se9HN^`uc-J$o#x?v&ZWS?3?mE8oZi2D}B0c}e z%vWoR&VF~c=E{tw->;C&Zbet5@{;7n4n3k;J<0jD!Z`A!G~QcR}+M zpZvWg|MBHQ?u9ol+|=H;@8E^p$^(Hgo_+DTWvOaI-FYEL3n5q5B{YNAQa%MpM3r=) zGS1LpH+!F+_9chm7ul*bJkb-v`jFELxm}wKIY5Ffy#4yV1yt;^G?=?in-E~8twT2R z7aIB+&bFvHuH)Iwf2v=3mlve<=~E4EW8pvtB(3sG9Wq7LTyziX>osSb($A25XP40R zsSiR|4Vku4?>BkcRqG?#E{-aPO)I#Pa1pUI>sf&8mA{uWu6TOt(^itkE3n-FZ(>Ujd06f>v{xAkfRbH(~v_lSw2`dV#swsJao$0v1bUi!squ;i)3BXSk zp3)w7k_|&ERg0P05$-kZuizlJSrr`%3ULH?O*HG?6A<4Jru)2F)Q-}1*F~$kTf{3? zNRk6evUEIhvaIoDXI3lJZ>D^;{a`X@3u9`#5`ejC4ol`rp-m8!8=@tFSrXIZzL+F= z$n4!?`{~!lH}-c6ozBik4R2a}i$C6$bh@dfrB3(MB2=jGafzzn^E@Z2pr)tUw}L-q z?Q(q~C7!SNL`$u#xO!DyWb~x+BBpr-pII{Q_Nq`sE>bS%4l2=MuO=OEhc9-=F$O`6 zH!QRlO5*ZU?qY0)@y)ing88c2rzmp9ge>*FP z1aY-eP7qNp=H&-8aJ+Iv*7S+l80Ja=0khfEa|dI4?B-yOLkl+fw2L|Bm-_XIcvhr&@f_V9 zUj(WXdt}9@+lqE3V5h5SfdKQcNQ%U_sKk@Fy*Ln`g9BRc7&{}eHV>acu!J45XLY$ZH?JoUI^%%^&%O5XUT&XDWLVI>6jhB8a+ z_AE>oav@v4j%XylA$mU&zjTAyK&)NJm8Zlou3o>>OJIK&=U}*IfwF$0wrH!h)%i*N zYehcTU^I_zGyzBRYe@_|*m>5;ktWs8E@ce~)sKB1fRE(#*k zqSDkVn|fVYZvQF$)#vwOP2^Sw_R8_!oL-QIOV|l9r;fc>&u+S6-BxXcm>zspX7YZy zs)V`Pj?U zWxZL3tcw_*H%NXmeAse|>$+~HvAe&CZ1_;!=+`Z6(g_!Uh^qaX zka!p9cdBZ9*hk@}aED+%Pzt+;x}-*B6^!Xn-(Ve6nfYGYAG1bou*0ZnL%nsgxOlQt zb>HZ93+;-m#HmK(I(BJtL%uT#=az}I@Lh%DG)Ye` ze<2voRX_~0FO_RzNaex8jmAyec>ILO%aeAm>*1mrmSndqZ=M9y;CK}lSQXL6^sT(J ztndIbrF*k%1sq$yF9_7Txjqo;xa5?(EuDW(QkbZ&?w2X?cqDo8nhJd9R$7C-0GdX4 zr-f7ed6(DRr$FVU_2+9N%IcwZP@$<->$O>CY4qK5!(F&zs#jqT?|nD`jb6&3;SO-x ztK(K>A1JHwak`~YMZz&Rw4BiOi-Nhv`?5Y!g1$!e+G2^=@9E0DnlTRAQHDNY>hl*w ziG%95Cjs6}%9P0|)mYKzPG~)A_0v!(+zLs0=WM<9*j)lSnl?F_kogrYrO(+0n%Vpl zfF?382jE?AE(hrqbGVJqB+Ul}atgZb)bqM8?ISH^VcK63zOU5|H5A7B6&|OH7)Y(9 z-fcYsb0Nq4$$W6|zS}aO;fB}c3wO$TYjaHfE=6f=Eq!vdc{}!gP=3@$`oT3xCGdVq zPc=5-E+hT1!Qbo)&QiOi^cFOM*QY#iJZator{!qVd}tEWO16Q|ez#2_Mjtyp`N#rM zxfBq(Xkg9E*|Cq`+VqyE++pt1U2n3BVY%^U$IOab2wn-GT zo~jK?ytkI;>dINj{p~F{6Y0TO&`Y=);Luik{E2#Gb6=0y)4IlOTEI%<;}xL2&Eojk z)zw@LR8xCE2i|Z*y!gE_6f>A}OPz-NyA=5Q+6vzx1joUS8qt6S4n9@k2ep>|q65te zmW==tpPEhwB8DGNw3gcuJE<$*P?O?gui0yjTfmZJ(Om0i3!-;ewvGZ!L74zayt^}0 z$Rb4S3^*fc6$RRGb3--i`)>BKD?rRf{r#Iq(KmXnj3mE01I(*{?=weQP6uLC7z>3( zF?`ml$8>-hif-N7{7{HRc>i;982pJ#2=J~Uu)}p5x2Bcq2Fx!OAXhV`M|w2EMkGUq zOPn?OJ$W?!0V$)L}{|>>ITq7^pjU#w`pak zz-?+ZsL(om3LIT)1j3ere2tu3Qa*u$wfgsJ%rPhdjsCqIKhk$rb3$$abZx7M38$uo zAmF{03T>!uKqOCRdu$AYonr)WV+QGq!1m?`;G9=(xJq>c#>IetZ|pn+K<{*bm)uvQ z@~X1d+{GBfKwaPvwofw1avnaJLjYVevid!KM?PAIDt;`2y-m@uKrcV9i$7cIy)(qx zSi!P~8^BmC&PepD+Rdac(Luwo>|i7quZSSs%`b`MY7WIgn++R6fh}f%UwgIBj*wcW z&ZY(7m!g$|fHu^r3XqP1ILswV8Ig({W4SP28!i~d=v^f7(5xa7F0PZrGUaabwGtu$pUn!3TNps zexi9Al#MzCXts0vHfQt8df&1cO@|fWy@;}CwhkvkTWQ?g5TZsBl)-_i4ic}CLnvSI zdRaE8Y4@z8_Qpr;kM={iy6nwSWn}I7DoOVfonRe^IlvA{pB4rSf@rk&;V`!07m41RqyMFtmn;3~Q*GTRR=o1ak zy05RJXO$Ir4Ov&kPTt9rVHi+0s+=}}wUS|1#^QYiB>*;2aLD~1e;TNg98C>U+LdEw zm6Gl;1c>E6*C_+t4YA8Y1zFDY5*#t#z=l61*5Mv(dWq>H9SNkEd0EmTRx_Gk-0a{| zfZ$TVTigIw_$~fW936f*LH3!|?Kc?ZB&gjx2g#Gq_LG(7Ov^Lp*P=hP$Jq?%AP*D8 zo6)?TPv1_L519L+MqS7A%sSz&ezKcZM`XlkMRGj}W5dZiU;P!NSsza)2qFZ++ONob zdeQW)(6$jNQ5c>UjqGQ0rmr;SB%HS-&w2`1GL28L9XcXUkKt3QTPUp*i$;r7ba-wO zJICUdAF|3a@a>lofejzdLHllQtmf#BsA9d&Gc2oiXhny=GW)%2*^U7X6*Oib1y)tz zh2!@)tdJfJ;OlzrM7zmrzV_Y^HAmW3(wX&$?&b}@chUx*?Y~f}s=Rff*4wSvm5MIy z+SW!xlTs$AVNGHX`$9HJsJN3O)psYS09Ou%ynR?29a8-*gi%_&nC|uYaM&mD(LOuY zoEY1=PYieu2C)Ln6Q-Nj-NZh6&}uTe$qJv={Bq4e84I)dgCA?NE|gvi$55Rk6!i0s zK^w;iV1{r~xbCwi?*EU;wHv7M2q<#Q3tObPFvBZlQMxkxC4W7l>^iO2NU zkCAJ1ePpxjWaESqH*wM!PyF%hL6jijg~bO*ZS_I*nAh(Y7Y6LS8}+w0+rm7cFi3ih z#gC(6@!=RRgJaOX3WH3@Liei{xyi(*h9W2Ug`W^FH6FEyx=Dj4=t%94*c=*{7*a3_ zGq~yFgc8~W?o0{;jigLV)p5c~hW1`xn%eqmr z)3V>x2NB{l;ZasIw$d*A=tRC(uHTdctbZ)sind%&6zgjc+`-gwJip$)%ePo7IyZ;r zew2QO8DJ_jTrS@F%xpC_Cf&Aq>M$7}wld=>3^k?@o7NTxQmu!7C9a%3W1(9S;JzSi zt8LW3Y;$qb$y~MzAIBt=9Jp`eh(-VnqMj{XyK~!s$F4v=WDWxSI zFa7aE?#XDaj4Li%K7tX^vkl}bg2|SUBF2ib=OS=_qQs+^fuJ&k>KAuz_nfF}&H9EJ zH4!q{+2dRijskBmCxR0X{oF^5YcGvSsjZ0qqe*qI#;tg^Bxzs)7@P&%;&I1CB>E}! z#}|*Md-6o|LIi5Bv?4^V_&a$jp{VbSY~VjZ@9MIV+r5k$#e z*hk7gO>_lO6*tC;6On)Lf9*j>S=@TvZk&WV;l_;CjhJ^uO=vPRBv1uxQooc~Yf<}o zs_^l)s1w^BkHy(!)4xu5YP`rv!f3=$ullat3Q7&+^%4>L$Q5`sYgAv@oh{dm_3d%Zv#QN%LTZOi+a!!XW4R?yX z|03L!EM%^0YYz5!J-U{RT^>R~uP~@H_&!bU_N*jg@^+bV7PsQsVSzZ@(!9`=VQOcPz z9o#}_$L<4R>{xj{+oL8lrsR0!rArGCT#t~_Ws&QTDg_T=c>4u}!ZAlW3)=r+e%OZU zy)izSgeFfNop^NbeV@onuj2^ukHR(F{;CM2R98J7aZIR0w$@(3&l(02w@wQh#r4Ep z60tWlF5%fYm6O_!Q_YGSJj=9a=I~bv3KA1R%0uVY8%@$p-P-hDTJe4chh>IKHLT)M zkBZNTi#rr=Ei5a`B4CpDTkjZ2AqGtGxX-)M8ExlB|1x`g*1`X;y9|SEu}l zP7_S=mMM$WoTCFjxcS=BYdFuS)9%(e3gdbZKg6Y^ASlOAZ|QovZ$)fEY;IH!zRxsB zwmsKmOz5lU_k8S)Z+9KpliZA<%U>|ElNPT2F(2;Qx5|bXHGK1 zrAW`y?$qZ_3S-w>rhs=~YCR1W;H!7h{Vp%9FJ#UVk2u;jXP1t@tsv9jg}}_mj?NDi z1rSQTtfn#;e{b}=NQiUHZc@v5i&}~c+%#pOMdKT?d&MtCWSo07Q26q_tL_7+DffcD zUp27RcW|(Y9Q%=j9mP!+KV#~xmoZ1>bvHUX01H@4WAgENhZhs~Zu-|Du8bAed^i`G zG}}1@bRs?3P~4=R!1*lm;D(c(!SmnD4(;p=JtiT%h{&TC&s}*>mE9t-7N41%^&$Lr z`R;{dgLz##QG4~0B7kOcDUtx&&GYVFLcL2+E>)o6nWo6F#?<7^|I^sUS-B+ReuFqg zp%X^0mwBw5Injd&;bg!0=??Vi=CAc)TwV94D8W zv?CjUUROV*EtMRJG5>KnK`<#*=73jL72ocrxqCV2EVC62R?J`n{#eFy&*vCO!uOe; z6bghk!wLP;VZPt;eORy0l*?0|9P51A2f@V&$SntkT&can)z1ln8P-m{+odn(7sPX< zNkzB|pXM&DmkoUvG?Gvb>xHKe$he2dY6gpLTC0qIbXm)j6Ojgd-hao0e@~T!e?2Yo z5vRm}nGKN>6Qgede?!(3uoij?R$-~q3H>K R)i2=Zs-d|-p`Kgh{{TV4trGwM literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/H100_benchmark.png b/blogs/deepspeed-fastgen/assets/images/H100_benchmark.png new file mode 100644 index 0000000000000000000000000000000000000000..89fb9ca3e1cead55b7634c1fdcf57059c4b1f83a GIT binary patch literal 176743 zcmd?Q_g7O*)c8wBrRjsxJ4%%<9YSx?ML?uC={=NCMFdnJ^d^KNz4uNiqV(Q7DghE& zB!mE=-FWX>cfISq-~ZtLkagBMnKd(K&z@aAd(LMAJq?Nn3=i<|@F=u2RbS!Zk$~{< z2p-%c!978~=jvAGcz--#msb_ z>GC9kNJX}X_lW-2h4TT@EsW%UJ^X)Ksz3YyyZ^u2|NDzB@?Hw}|F^41MH?ax&Ak89 zD3Omov4Q^2gRuSoZxAKbq8aBK@nexRr00&F7k?06DwaPI=9zld^kTwLREtbBr(csn_x=rRHqX)%0Krk(;_8c-Dl{<>9|;1I6Z^6w2p8;QO(^e z8D#hAEOX#`a)xEVN;CXhTlUewg!Hba{N20Fkbjk)KWAzq!BRkt~9 zA!wS!X3+MF6^GQ8aQa(1=ML}`Tra~5dS-H*U)OcqiO!x1<*I32b*-^hCmBr7gVUlH zsO0`CViEcgU%B!_QHw%Fgm>3_3Ug;m*7+=D{{=Komh#Nu7rzmQO;eNS^5^S*E0+pm&zqVzUw9|Qa_5JN>WYp$MG&%H zC?2e(G3Sm(hF*-C_N0_%zx%oMPu!&A5XP?sXrvuFd9! z9A|Hy_nenR)4YYtW0mtCy*Zwtfy;_&GS6Lp=FXGq-NQ8*IZ|WT?~+6=0z7+fFu?5u z$=z#x@rBo%Z)r&w94>xP>pQG)rVptDW>W@P?=FUP>MO_W_O9=c5qGk>B0nFSdHqtj zyVUm_@HZbuRP}6_l3E5GHVmPl`$oY`1BV=4FA&A40aaAt_ST4pVOPT%fE2D=@9&Fs zVFS978)<8QaPL;ueB60J8IlqkKWw5PvK=qh+Cvd90DsKq9rS}Umt1&jVotMJR66-+o>g%z8RSpqLvN8CUp!1YKk^Wgo8vRSW)t2N-P7Iz8j z?cdme9)Siknq_Bgv56N(2Wz#FhjUJR+w(^H5E^s$hD9MxriA}#k_PP$>jx+k=$&$B z1ZfkrxUyjox9|74-F0wsNvI1yt4fGv4VT)wYU{nJX}HB6NbcrG+z#S4*-FqsO&fz( z?@hq12NHVj`MeYUEdjR+>rQ61g~V?!ry?Y7PxJ3!TO8<_U2pQkaFaNA{Q z4ASazz7g2g;}L$>M@aRebt7PH&4GmXH_#rj}UeH+)A2a|UnenLsc z+e(px>ZaY03=7|pLFv8Rxz24%In*#`g7wY5X@XGbZ?XiOygNkB=8Yr)sN}>K8CJoF zK|>AZg{siIlf(%3YMgCI_@^vu6SseG()XzD*!?QKCDi>_`C#pL9L@|?HH`9A!CD#V z-dz8o?H$I3V-aP0;q8R(eDQy{aMOaevrUTP>pqGjL)cK&q;#&+)$~2R;*>7 zV}@Bwn7QaA==x-+sm8(y;*X*$m-J>eZ|Emsd~w{dm$%r)$>_tS|3Cccpxc{E{ziw> z3CSt`s+zpECpqZ|6GB$tv=S&<`ptcJVhlIn7>@Gu;`k`VWY4!RQt!ku}tG@%bHt{=GI&vuL$;p-ydG(9Sa*=9;H!Arv%Q#Zq{zL z>M}C7O<2r$)dz=;3AE|ZS_RpHRb_EvMe!lFaMlbVyLG7t#>|jB_|kJ$0`TDG5rvs= zVNGyK8;5#uxM{<7U`aIN9mA7Vy9%;4gq2!(m^zjy$PR6KbRD(2w644x>y3glwGITB zC}3w2eScV*K8keDyVtmrX$@e;X&Rz&Oi=ruwqoHL3SV1sYY+VMnH%%QiO_5k*So*7 zdx7p}j(W@2`gDRI7Cv;#HL${y`e52R?(L~`?oKeyj+9dCNwEvG%)f$yCxO|)y;nO^1%W=+V8|A&$XnIw1`r(a^SI(MkdUYJL+)bVbXJs7<*6)xdX-%paQ z*x0gkGQ^r++qqwPNO$v#R3WSF%Sp&pfM=^qBP&3oJw#f`DL)KzXcO2oH5HC&0^-ci zyFHSjnDdp=ooN-SiQasiYF%DJ&o^bgR(w}o%62^B-E~n7iDA`e8C+7>67i$vxm{rY zB6kGNI-H)jHmWd;<2(rHs_mb#xy;iuP@g*NJg7w=%JL_t*t#BCZ(R&Z&SDSh@(<}E zaLQInA-OK>mHo$^$T1h^NE!zsK0nJCZWKRn+?mu5N2NCJesy~nyMWs<*=@;KOj|Ex z@vU1;Xg*TP*`5ApLkXbgY6H3J>-##Dj2nG-vvubfdq&%P@zcaJXOHr*3)zFfEvX)E z(ameAPvy7PIQG(v^hS2rLi|U3YAD-Q-sR}TFTeX@e|3F2HML{3S(FHnfu0O9W)q<| z_5kM{i)*ho2dP}(^l7Bz&c_oKrfcbS98-7 zezkQM{?>QoSyQ}w)R4437-q-M;o{jmqh3>XOM8XvjgZ;P^+TD-<=&ym=RcogCtFR2 zb$ku2ne0%WjKzO00^bqRCn04Hs!#P~Ec;J^pO3?lAzNde1r&S=mWcbUeCuK78>bS= zldq=(TIl!mD?;=P*BEBlk7ib0qpo&@BBYmi2{P@&pl6F?{yw;1#tEVt#KXNYvk3nW z%8wW>0-C;cDe2bDuVK2Wd%k&)OL{~;gAe?8J%&*N&LGjH^W2aNTriYQW8vLzF=c)6 zofdVzuLkk8O8NX^_z33%Lb13}vi)0xo(f+5*u(`5fXnW$$NQS{b0JUZUWGJGNQ|A& zN&3Ggl{Y4x)0MPbB=s-=KoYmDj$(8dSY#qunkqi2?iBXsoip78deSbt>X~QOUgZ`x z3^KUEnxGGcZ*VSb?CG;BQ=pc-%huuZuKq{WdMjG~LU%t0XEb`z4L|2iT87gZvl{IcA;!G- z{rH*Q(O3H%!3c!5^~dcofn$!y@Q-q8ZZNXarhUTW>n8(eZ$pqWk4zh#i_73>nQ?WR^FL%C|kDV{u zWbYoa09-u~Z82u(8jygGjo7Cfo49P8SOrb-*L9~{5n%c9^G>0m$&Wx$BKNMo3BubO zMkRYMCv@g%x(cjc0k-jEELUe$sN~aVq4l;37yVR?+edBv39Zj9yZ&1W~JcQ{~k6y}!pVqb2>?n7}YRZzb}LTdl7MHrj#UpfIyYn88R#DDk5 z1*NvWq?Yr0(@39w^y}a8nq;MX2o{|rcWF{lYFk|u2~-GDb1t~Wdh1FlIMY* zfZhgyUu6+khEFy)O;;M0JP=KtvfV~4^>POvp_+%9ystqsS}=b6o9#bcHr<`y54R6@ z-tVIE&qw*49uM(x3!$`DpixEXa?-g{iLj<5qu$%IICae z#8wA3wrob@o#bMTgqY%bIF~S&T-@?bPcJ)JMHDw-j$3yGjz)D%A9w3!pV3~vQg@*f z!ju0U4-6+FHDz}A`B@-Bk+fK47Z?ZlG=5?kn8fPee`%d~%P!{l%^v->)1jz#{+W}% zX)?2X-l2<5r(p^6?<1C6A(xlVqLjT!bdK`8;AyA3rLlMDQh5VSx(Y-(*X-kEF8% z=ZUev4KH<9=F3`Pko}jCX>%E_UMTV22}{Uf@>u`K%XcW_*kC99d9fkq4O`3DyPeeA z;haz!+0py_ce(SmwFNcpiR<#FqZN=V%7?ZNDoE&5$gtjzq}J8@KR+J;WcnBDwP9rm zKD%+E%J&+KPzB75&{Zy;dl{Gcg;`28!;54mHC|fops0a_>Iefvpl{h%#2@n0vqm7)tkjdwdhFq zqV5^uAJfSiEZJ8jA5e^4zq};k;!Acc^OC(#yh*LsNs+C`(wBywj>=E*q&bMsxO;J~ zsxi!zt8*-z^W8{`BFpp8B#D)C67&a8?M!5{U|LUw?6i2J-O#n=s4f*kvYhaSYVFKE*k_7SbJk5qlnB6ew3H5 zUA>*KonqnmM!y(?6wTU$2GM%wOhZ{ZapvR&3vx|;!o3*+lB}7z-;AoFoxqsGFCq@G zD}u&Fb~-wQFuC(Vrry#ydZq6Tgr=26&P$P9pJY$wIdhuQi+iG`Uciy})mC1F_yIEoUjR=w zO8EE9x$BM?<$$D7@~MQK+utOC`gxHQE>n`SMHc9};vZrtC9jrD(Mvedr1`($G=kq$k_KNU|goU5BIkz!p(moV40u6NfN-)1pOqHIhLepA`QCBUTGXl3}t=A79jX zo+&szA#DPs9D&!}#6=i?vcp5FE~T}{wTih3h8~zq#@jiqNF|Hqax$3@&gYu!zBI9j z66q<&bK#F|PO$u!q2+>y$upnyISjP=zJI&2K0?`OJnJjOs7^v7^+sAR;E>Ub@$P_D z4d&j7C2mQ$b43Y~vaH48avX##r_?VB&)YqXtkaAy-1CbKa)5$C$2?4QVOG<^M9JF7iu zD0;KH)Pn(!%PMW;B&KE>pE4EfSlx)SD)|Ve#@)^YWreq$Bach0(K+i~jN2K$$(iK@ z$xbl+)}LBS)brf2Y|JeL7R+W9wfV)=qwLJv9{xY1^(C0Tn1sgFKEk)_*4?|%gQo{P zmB)@%2qmIta;;^UQfEcLMpGQ-FT2ACf#%Mey zr|0VpH?w}tCg3v7>x-a7)j%7xut03&@H_1^{rtuIz}8_)=x|)%7nAv=3oG;4GOa`% zMD}%+QR{@o&=vaVSk2Wfa9>d&a(HGJP95UVZCkB0HE!B063odFtIMH(6m#rGxcYn} zRUzO96dePeiSWn-^O1hg$So|JXB|IZIWc4YFs2Hw8TV_VVr<^FEc@ z;|kgz;t{EV?*p`ZTl|uP2j4&re|=JN3?(cgPgin(O-PBh@2R}6R3xfw$?QMxJZ zUFXV^P-%}iQ!=)Xa($Rbb+zW0X~~c-;uW^0@H*;ysCSv@YlwQ0D-4GM)E$!;7y3#K z=|8JG{$)tVAKyCr=!%y_#Ru$rCobg`CW|esUT~-|q;O_f=#yvdF{2PX{hdQ;{5p}+ z$%r;AQpGK@@AE``@IXnvJE zDr$&ZZ3n}GO{t-Ob%N`E>%p#vhW_uQ6VC!on0>;X7VviiUiKL8tpwX$D~q>}k)mCD zjrX>&0X7D@!wd^-xS70VT)-1$z4w;lDZ|2h;iW+F5jP9qjM@gvwK>HPKZXUoGD8YY zTpRg=HT0bRRym3EWzvdk!B*#$T{;o~wzhM=2w!pZcT3~wPtzXWyoZ$NXk|ktqu0+?AxvGXM}@qaj4ewU@g$L`eY_Cytx6F zaYLQ1fkVuJuuHhan)~Tw=<#lj*W~tkZzhixi%Jj3w;L^`sB#zr9%JnAq$_)E17@JOP7yV1FBKGJ{ zmF>swu6pmrww^FWT%VYM(Z$0CRulKL~XCJw4YWehA<5GNdQ&r|l~{aN2P2xW(%QuD_>>~K6}Vq0aK$uR;~ zrg$4D!Hzr@=dN2)2V*7v6%^!3rE#pK@v!ga zUyi|uD{ts$kWB}MC%Kenmp1HVnC~JV%&!IFV$9t9Z*gUL!LEx>&I3^L&V+8&4eE_gwtx2z5XCM`NE4 z5qSKh)0P#)G=Tp>l7L$r76IJpJB_oBbO1=jC~t9iKXXk?mh!N+p*}FMY$#D4^9FDT z*~kn$X92;ISKcvFQ4ztPnK_&KM=*9IlZKVVVM&k*$L-s`jDklo@;nle9`-rf)+$@@ zk}Cu#vU%AxprrXxuG$39_UKYpXcY9i7Wlg5GN^?9{vm$OU{cf~L zVJpj!n(}ioplclzGa8^D!t(|PZBB!_pi?q9;OoaavG*EQ&I(gV(YpkDHma=(L7(ms z9KF4oVLc-{)jWy;3`<^S+@`S>aQ_{t!7n~oC6G-^|ZKzv)I`0X!kmk8!DP$r5pJ%xq9$S zTtx|)Yhub40W!(oXBF(dMy)G2@0EB2kRBJ*gkhUdxY>9wqwdkfBA9SG)pa_>)cZ|`!_G%zIuq(QhLM&--@EFB z>U|Xz3-`vJag3Ecd1S)$8scYlKhmd81zxs5YEqS+j)(CZ9K5n}ZhYeCpZ0{(1o7sQ z{l!+-Xx^kro*Cr*(fPGve@IPxqo^uN&5Qcr;YxN;8^QD_1RKD-O@q_f}-8jHat%RR85CA6->NwM1jELsd=dD1S7PPro^D4>wNc;r7B_U?2)L0Eh23AP7Qp7zHbk1x zy=^I(NT8?|ZApn2bkjEZ+i(Ijjj$u8m9+@BkE?N=8>|{pvvM4>GEZ<`c+;@73Vjk` zQpsDjfH0|}cmbzmZv47bDLq~HQ|j46It%9zv2kIQLwgatK%w`N<^ zu(d3!ryUZywFW}q{k%c7b#C|5_2a?X>;KNA-&q&1bZa;eT)i$JU)5}|hiObY zs0+Hwt(n65chgV3(G>0^<V+wr<7{zeStsY55X<(UcEa`m|4 zoQkb?_jquRL!$V1qw%+`d5?-XnNC-Yoe)Q(EtFuQKRj?5u*2hsqUm`aUA zxeXETKH{j{H!OzOfTy^Xge=i<&&Sz!T52Q0h~@wDAEc>>vI`-bF99WWde_MmWnxuO zCS11?9-Wkp)cC4ih=>_R%qfnhtAQ_O)n76iOrg60J%)nY54ZDYt-s{X_DpoBhR-M+ zDFODUrw%gXOJJWzJW#!huJw%`&m=p1c+k;2jE8#VWoXh$iYjVCqTeG& z4lMB#A;9JDut(26Y5hLU?LdPMb!{jTA7i$)!pX#7RlBSYeZ;X zDEVX96L!Y~GGvOi$Cq##vlW7$iX%u83#iW_Id^;lC{o`f5>y_X9C}u zAKp)urKR{i?6-}i06A3x&eTi_z-VxrEVr~ki7O85I~lrq=;uJV zIhp$0KkEdx84w8CzJB9Ul8(Y?34Uz$c=KGYQg)|HB_!WIbO2O9RezJvz}4ks&U3e> z_t8THv*W%E7k*bQJ2gmyeWu5>Q(rJ_M?uMNKKYn~bhtlj9+%9%x&1Pr6`o)pPb#$- zoI={QP;qw#x%RecQTaM(SW9r;G_qet>5BsYap3<8315_JmwHMTY_BaHv9WY)z;A!4 z=R!xiud?s0Fb`=S{R7zE=yZ>NR4mXtYCNqKPV8Ui9C+?eY(FuJoaqW?m^D_&&B6k2 zS{kZ**4i2R2k19V>4x6Q{SlM!~>7T2^YH{5RU6=gJAJrH6!D+GEcmheR2f$(m#4XGKY zyhPi2tkgWs^30#CcXt8?qPq?e0(^XA)|Y132Uu%tOF+&uuN~R^(Y{Aho_tZ!-9L}l zUQOG*A-j7}cpQ%9gaq86WtlPZ<|dlL=8g1;OLO8vl8JvK8g3kF1e(XpbA z)7=mqinnX(UgsO4@T$5uZX8yTL^rxDXbtjn(LO?<6t4QLS`)P&WIN^--;`_T4@wV` zT?>VR{F&3#yKXo&ki0hmX3cxb)29%guAd#NDh+~}K%{3A-y0vMd&hP<7U%~@LYYvK3aApTP3uvnOH1a+ z`*fDsqshX$-`T2k)u5PxCMnhKsxA@-QJ59-Mr7IftxjyvF6*8px6^q0JNBLqmKWzFC`37fsfPwaGT{*L{g}Y7 z&T-SfY|!D3g8sG4<`cHw<&aq_4+G2DN9XDNgqF&yvrtsmq+VAYmAc}F`A;PY;9OG) zWR{;fAE}Fht;oVFkR?ZRH;x1rimka#Xq_9lcMTCQ2vy;iA(++-9aBx7#8UPvIWdz| zFw@0fSYb#1WMf4ANuwOFkxxZ_)mMJ3)_sc*)u;<*GxsQ>1kG+XGx|FudDywtD{fW0 z+c`aoIg3mY<I!hQxw4?O+^}=k$1L&0`@GGD`$6ZK%K{D&Hd4-9!I6TZf4K?TC@xk*zW8EBRp7&@2n|k}* zVX^BV?2E({CgXF_NCpNzCm+sFcK?gqvRkhos$HW?@?-(fe7~_J1xkrJtV*aHI0dE& z9^~kt3wsg71d%-|YowU(TRWlwheFHG5DmIaEWOx;I$xqfl{+F!0j``Qenmma5o3M0qD6+O_ZarDv ziS=yY2>Dwvy;l^#5Ar}-qf2(H&zcivirtt@Nf|AfGLGOL%)@0qL@TtGggY4lU8mZ& z(5bEiVM?J>JRYQwC{j0kJbJ$Anl9FVVaa=SpEN-&p+o@A#i>8u+sJ(GON6JBT2RaS1{u9)`PHG8n5AHXG32L52v6jiJ1?zloz$W%y4O2m<;GQ6Z!T~CcRg7n*9nDyI`o)<`Vr$M0i&kON504W^KGCOZg2E-@JKU z#f*NMgGHk8B1L#?5**Jty*%T*CI@;-~mwpM=3M`-j<`!@Ih&cyf<4wGtbI5v5DFr7@! zRCv3XoUCS>LzXrWU@ODREEzR&FC~+b1ysms-#1TD#-Jg-mj$lB-UQ&|gSGijYs$F- zLCM^QEelk+f5RdtJF1kvZjr<2f|v^?ZQX4ja(+-AvP)M-B)vcOCSqWl@uD0IjyqdQ92v==4feE<@XW|se4;o04%%Eq`OPbMQ}o{R zN}jo6GKOm;bOz=lB@{v6iTcJ0fKL%WWm$2mjI*N>L!*{SILP#KEiWt$`7VnGa2NfiXozUFu_uV%Jk$H7yFuAF1bv`p706eo-iV8cA*aoo}3jYBmo zpgz3*omCq(7Yj}5QL^gNMeq-&31(UN%{HAJ6|Nm6d4_hsza_RM5BeZTjQz-F~3^Oi{2OTQe?(c8i(f4L{#Q<7QzFPU6l z{%qde{)_IJAc%D8^DK{H+|Rr#^EI}*PEN(ERQ(^1rOHiL(#QOseKQE7cXsVqrOKp$PzWPxlC?P(AwXw!zeTv z8ZdLW10BrEed4Y~?{+y+!dj8YS<+8(79jV_lIhuDO9xGen}vPDgI8VaQ9w!cxvBc- zpZL3Bx90%+_qBt=Ot#Ow&U&h-gByQt;G0ePICh)lN6=Z*_g+FDI(-T~XQ)b(7LgSY zeY<7C%FS_>-fz(%Xg<|x9d@?#=&7Z>&$jCJ2tfqrI;;)na;2(v+R!@=m37IPormk@ z_X>lz=9!0!=DRv;6;ve9s;t)NrtYb-OIAqv!1=7VLeStHfJ{noLLlBWK000{VP(`2 z?8a`MO&So*5#g^&Lp*vc>&PGG&wPX(=Q=QRRdlpC9ebX+l41X?IKLp}T!s1+4N*j}!H;(5TYTd)RHJ5I##YZcS z_@l6V2rP7yq%$f6maz5tPNT~)c6@8a+oLs40SnuoWXT9N4iB(?izXiRaww|__A&!n zXm|mO9`ck>c79N!+b!4y?oRFw4Y7o{=P`~Jfyz6+b?!gdPl!)9TOpel-`LX^ZsqA3 zdc0D_iskJbKkdo1mT>@lxp2r=Ep?Inx%>A%+&Y2#LZiD38s3!>VVV@P2eW2=`^nC# zN;fCsZO{k?aj-yPnMq~93S2|F6PRTWG-=Hm=8wL$p)}W&YaJBhq*p@4zj~}P(BLwL zPcr>7M^vg9o%>qm<+(5yxe0es5%nKMYMDyTo!o}6mV`6Xwp-G`HPb%x{&CQdrme2{ zbFdQ$czYZvQy&0df2f4XbPUx|y`C^EZ{XBPqSZ|OO#o@YS1`%8-R9YIXnv5-E^c!c zShk=?&U25|p>KhLa8Q~@V9V#dwm!H1yWEp3UuNQzphwKY?V zdUq^^JfdFx@&qB{$2_K}q+!Y$`Z%b!|NcBF6?#BXhuJUuAg$)GP~U`sCo^3M$1ZQY zUeyxr!o)RaNGu)9>fAsErY5uLe(#e2_vmHDRy-GyFbQ%TR5BG&d$~JodvYdJ{*TQ* zhZfG|ZZR#O$^@nRZ#PMb1T!&bnL9bW0$%}*#(z~f}2KdbU)r} zmIpe0&?VLWtA_+k?3(P2CNv&3loI3n4}-@s0UWx+y;2YNkEV$}r)I{F6fvWm3UoAJ zabYZs^_N@$=qQN93j9UT5e1qGJ-f&V`wX=0^1x@$-=Brmbe5yFdc7oa-zdZH;iXHfL&*nR_C7 zIl%xW7TwyF7liP}*~&hL_N1??iQQDB%}dnfLw0otuNmG{=e2umytB=71!ZkpIt za&M0&V4^>ktow5<_$y6Z?}d=4V1N?$8nJtQs5b*cs#%(ih;f*Bhuqh|k=S$kD{tY^ zDiQIkK^HS_W1PKDk-VJgZ9rhjmlVG+e;~ckoBHA@p*=N*9QK6xh?2Ug!#d7#sXD3B=YM|DZC}hU<($*UJt58WX=(Tz z0yEnxx|6s~KL5iec)6I#5Wx_e+HsDa?>(5B;DsA(JkkiKNzS6WNTBQ-cY^q7aiQK8 zSnOw7qz{$EV}bdr`k!se7}3*m3vaS-JK}9G=;9xxwRJUwXJ47YzFEO$vR40y5S2EN zC=fip1b&dq;+%8yV?_D;++*1G3!3GG3qpF7&0kQ7Es2)yn2H3xS{n3r1bjakPOb?Z zTy?O>Bx6>7+FV!b73(zif$u1Lqu_iHJnCv?w{934^3rm)>d$M(I9LZvWtB!Yn~*N{ zRCZXbYeX-om0SMx-l6(TCMrHai+j9lS(EY|IQ;w~2k`cBQC`8av%i|jF3Ij7FpBD@ zA24=<-@Gc@w9>D0?ASFHs04A|vs{^zSyy2J>R(NV4<}YHyMxrx(>|aRSyoc%$jdJf zVpX23oRK&ughaNR*s6>ITEJv@J z{^&*9aR}l;i6E|GUMOVW1#br;(n0CcuClAimzh1~aIi>jh+Nh6YT_Q^o?TpC zl3V19c;gHA;ToY!Ch)x!C%O0d;Qkt(B)c?Tz`)O6eh!Ri*Wn{-+Fri^m)ysBwsS6= z>~!IiCg*AW){*LVe}0)EcQr*Xgl&H}&w5vDx_T*+5OB%M{B4Bb+rz_hT?Jz-N#Q-e z2^u*E{%;`q@fvGxwv(MZ9BVscOUn}>qE4d0#e2Nx{nUEWi&C69PH%;6oh|xw>SG*! z28&r+U9iM*GA*?h$ftrdRWyo-BqcXw|1g*^>e-q=B=(s1zXEP}(kMokX_dsr-*RR~ zc%EcPraOQRFRT0O#k5)sG|Wb8iToQmM#^L))_!IG-jC?Jo5dV_9cx3@?xY9=UKpP~ zbnf@zjza+{7aOXQnf-WJU7q3EWeuetF)zJQq#`47c390y=*OnLlhTS`a;=%V>HGA` z9MCM18G>K}X>Q5ql}v3~Yw$9LnNl8^=Iz8D3FA#8UM{_#!HXtb@7pNQ>f{iSn%{|? z;H7>O6&7{B^LQn4+EQ35&6?23qo~%6L>?FoRR%IT&{_C6&nZyzfr&iob%4vNGETy( zw~%25o9*E-sN?(ES_&Jvxz9x0LZGMIpvtw>+X8t0Iin`S3k>ODrdAPt5!N9eUf3!5 z_ubsXV6JaRoLMsZs`=dWJk!`)}z=_A(Q2r5%ePzY+3y$)fWu>ATIJPY=&>_r`uuk6sX{50xbwRjLQ9Rziaka8uVS`ABqGaC#}H`ms}7 zzmFGlW}dy*G9^gj(W3ezOEgr9U)pKi=$BU}w)6^O^G8WvN*d}^Zs_iaPwxZZVKgsZ z_YbjjPt%5^gjw6X`e8*VUzEQg-A)kM*QeZDbMyB4JNSS)UP+oRp-L?*Bl^LipsZv3 zG=ODB@}lRVATbTzn;&-F+srhNF7&LUq>?Cej{jV5Kb}P%kV;i05S|g{MiXQC{2uCw zf}Y?P=ZnYKq{g$Zot8zs0H4U_)F5kQhn4{Wt0B~XtAD>kTtdDF{1 z_PJ7;H;Xx%{Z27_@M=)Uf4E96)pn#!A2Tnj9br}Ai3YW1Ju^bi1u~3fKI#2Q|4#5g z_9mV<`mp-;CBvX1RtwQJ?ik;atXWb!A1m?|1VN>T&vBda>33~74Is3j`bAFBu<0$c zG!7veUGLY)4L@xUJ6 zaefL^CqhiGnUgwoH7*-}_F$g3!(7MtH^qja%(0Qor^nlr_NAgdOH&zS4=n~)kHxnR z9;hiaYi}FabCW$ygKT0qgF5GV%_?FsC*V#=lT?ju^*H0v4^OtGJjO4IyMSLPr)Ppg z{Ccq>yVVDsR|^d%g_v;!-_#&JFw#A4vokhoo$ib0DQ!8`CEm6QfL0VvxmtC?G9<`c z#JVcZVahzC5>QLoSrkU6&v4#I9C|-b+*DopeM1E>eeS|yO(d@*QS;@tQ&0J^A`5zf zX|p?-Wm*U`zHImuj2S*L=h!K!&~?snq^t}}@9?YOEN%&j|9zoyLEp^#k&w~WVvj}Q zi@ZqNVxBg%N(GVb6A{^ofsA!6d!9L1)lrs|9bXGgZQ5EH4O5Hq9VyAqkR-|yb<3}EOM~!)B_s2ceqcim0>-S^%!8m$FC^Si`H`6E+X?ym( zXxN7CvG6pPVEi-tLs8+GAb-zrr_k!~#d;}l{V+frFe!~r%z(>s`=|{WcwIUh`bU?J-kMSXQ<4dHQjOkb zUqO-`dU1>E;2vQVSw6||{+yz92rpq>Lv(9#EqRcRRI`@9jd*OoxJI+yyyB0pg>7}Ucv|x)hq!~OfPam|dtJft*_DZr3tdp}L{=s=sUMn8E0Ty~Xle07H& zRFhoRlvrg;*(jPC<`kV1O1~!-uyuuYLDdL74*=)Na|>*_v2vYZ6;S~d3YRG{*Y#Vs zwQGJJaUO+8_c*ymdi3l*+YyN&Z!fGUZ&Y>T{oJqlvhXhnBvE6cAW^)9yX1U1v+}~4 z12Xv69u0F*;nKiWg&aLkE@T>GVf$GGYZtI)j-kyvJ z306_P7d>ZT4X4M(ZLp{M4i344VE#;iPI&j2Sl$`*GRHT|Z#ea&BmA4(o)svib#JFk zadV-`H&xslO&?LI9roxlLj)qXG}H#eovW5(7uXTA{(JmQwxnmn39LtR{LF7>b|{e8 zzE+fb7nVt5#@n`4yq)S?67J|2$MH?43BX4}d;ThWSWran{I8!4*_ zpO->5&*_bwQrJIp8RNdpZkwL_gC<0=+xrTEUH?M&hO>J>ZP%^`x$CQe^7;EXOuwnN zx%*($;FnpW5ly2o$qQ03mK4)q4=cD?u$#M5jG7mMv?HT?hR~mlYJa$CKQp`{zqTM3 z{SuIF8nT!t7yF}7106G&5y%-Ge^8v5^j9w;BoU);4V#i5Gz(u+FV;<5H)S=t>fui^ zqW7xu+j$x*dY(=roq}(c^bGf5#0VwraBqNtL-F|VD{N^Gyh$`DZ@fe+u$Z@Q^e4@n zsYfR~sXIAY?#cf})q6*?`TybnRl7(jH50X3o0b-__i7tOQ8ioCD7AvdR+QSasuEl6 zS)%r8iD>P;TQecFvG?!x{`|h@e82y2;vDNfulvsHbv>`^alK_pZ_t}d^&4wDWoZ%> zJ+cLMPcE@PMUYR$pG+f9KiaP(W>(W^WZ~lP33VbZKP7)W#*@_u5V{tKdEnjSLnj!| z0=`;gmzU9l%YL9ptH$P{mfEp5ug0%ULZ@_6JIX-`7I;G2JzX=tFPhp7^4=QhZz&+$5#=z4fj;nQk(b%Dr-_&l_5@y@G;lv-0jrf z09%Vq<+=GG9T$@_ke_r|z2ocu(19$@u zgh(ff*aK|REW8~HMjeoVQ8l)NM5?5F_S>H*;pYpj(I%FN0(Srpu(nX(1tbQ6Zy1+Z z&kkizWpH{5QCS+&y5;f|6}a&MurbG62H3ypKd90i`$jw|39&mT?9aHmKJAuh+X-WY zVQ@--GxDee*zi~PaK1*mceBM}Bc+8xhrR(8N%k>iIVCeMc*9Snlc@prgN2}=^9s4A zkk=m{+O2aKg}RypoPwjK%qL#mu~_T>!s(h1{yGMnSPJk$n-Bj)o1y_af?l4)d(&>f zz-qoxp$U)tu)! zC$4n6L%_HywGVjwH0{bbjI>7Hp6yNSY@dbro&m^&(IVM;&9RXCLQ9`F{G8o?hqFkP zfC}{KUU>x6hjKoJxQyT1o4Z*;64*ZhcrnxZ=xi;$8e>bA9G!t=bL>V%Q-wZV?aqpY z9HG{*qO?S^*Xs;O=L;)Iqh){$HM^14c#$m9Jd426Vgebz{bm9bW)P@SvpQMIN5Z^W+`i=&f+aa7WxPiL280N@{?N>U*V3 zy?4l3J_T43X9_Nd1!ry zC!Y>S7m4U89@%cpbcCZDE>q|wA+J6T0r~Z*qC2|fRvFe4qt}mO*(6sL_DCd_m+$;Ev_d2g$Vy5CnMC+zca13v^&JB7Ue~i> zCHaffomeO9v>kvLF>}c#z#E?wQ)q4vO*jQFOajT;h3a%Ih}YAP`VXjkJ-B9GHcg41 z4PV-Qwu>pAG+e)eZe)M*%2cOv?n~0`rp{I0h+51dU_Kjgfddz+Z`(6V%JTS{tcZe3 z!{__=0o!gYa^9tW?&F3!n>Uc)7ITqdYT3sje76;*Ly4vJBLMf}e?$Xm^sGl8a{tng zdogr-@RkW<`s~_ewzk}rBerFKT6pH4)8gm7Du~aVmx&F4V8A#Ev-oYb)X)3A&SQ^} zE}?$91OO47n@@>-<@}$IzW6Pdwx9y`ryBGi9NLCfhP%oa-`|h_0|*GK`%fRIc+A~g z&r3D4ocas<20%f+DjFwmbOU&WalojV4XSKAKh}IO)faFj!wUZm!eRl}M{AAot)Xhh zE1f(>pV`Dd@Gt{YRiW@TpW*jQsGT^|*wn7bwSU3e#~K3X%E3IvJ3{N&Oj_pSj~vxo*3zs=)5tMlRG{g!n>J%9n?%1N^cObSUqUu+HIaSE${ zfn2WdQZOlK-D7qdLh+Te@f{-LeCV*M^l0R zvb}y18Y1wp@P4CiecLw2j6Zw4YfbN+*ermka;Dn{?mP4aEaCz2R>i`fO6z9sdEVZU z*s1KguGGBevOUC-9#`Ic9NpUNQ?(;Ah&xD`xBh5fWjDt1*Rc}t_)e`&WcfRC9J}vS z0?W5{kAUfEg4%`I8_scjQbU1l8i{Pvp%RYpahoK40Ah%E$q-L+0;8B<-tZx zIaw!WK|u85LJPzXb?9qhv*pG&G#v_J{U=l`vGHejDrH9@`|M>*4wa-o8&m0RUA&Yb z_)aIA8E4ab${&eKt}k|tl@MPHTMj5PzqTZ$F75UTQF!IG_E4a0Ylbc7uWc-Vh@C%2 zvbOLG+u=XNWsR+GzDuQD)`$Pp^?C8P|7q2+vRSoiSxf(MsKEeUe>A40i(n=a)tT9r zYlW#P;}d*Bu_vZPu%o}pSMmYmV7O<*LT)_ptXL(}7^{s*wL=T^WXmuKFsgG0J6;$Ko{z;iA+RBM1+coex#4 zM+$%l{gN7M>-HwHPPYY8e1c_n+ZJT&9xtRWEY~Tt1%%6&FyiWF=>IlpvkXD^PsgQp zDR(O-U9IcOwV8@-ZqhcHOy4|8KoKCLO@6(_?M+h>HK}Q`m^odJZ6d6ODTL>S-c+5w zzC1ps)-E92NSoNV<=Zc(f`~Ft?ZReOaPjnvgEkw8&ve!x#?2pICE44F;bQ49YJTHV zWgbqH9E;K}L;#%8R~nZ>j_DQ>CBB}ZB#gSkzhGRQ7g1gTbuZ%{;S)?s4OR+v{T87| z2SL4la+u-mX%&7ysS4#UM9llk#~Z{?^y+gah#%nsc4ioq{c~PLubioNxD1p2^BO*g z@@hzqS|NHe)q+E0cQ0V?q{qwBFB*1)7px35WoGzkOdb7)$w(w>?!vq8ZUxS%epky|@qV)C_uB?f?VVd?6FKMb)9Nr7>Ei^m2aA8stz3_O z)MIRUreQTNH;DkR@A~Pw-tCi|tijXF0~rypo!vzG`EvAj=*1LszIQvjXynu%fryU0 zkn2QcZ4MxzIpG}NMQsLV+QN}VRr3CZULVrQ z=L-gb>xi=;yDz-(xJ@~gsZ-d%tl9E2uB;#Fx)B!m$oZ+Xr=|(Hkxj7v?N-&6xBP{| z61`IT$PqLnJ$v+Et~kM^+Y_#KKAq!_EsjA2QiA4*8PxO&#S@7KbsNS*!cVTxTLpD5 zw5A!GBr|Sr>e`5vdhVE=ECy}YbPX2vZdeL~ZbA7*(4@?@DE3CsHtQV|0>azEdoxZke#K@3 zED2Wb`4ij$WYEH<@-BL4f%#^G%eD2;W44)IgbMEUn0*tivLY^g0rPyOxL!vBUxM=0 zY$&i24+_$+e}K}KsJRnF)DUrW<+=!W!(-VD@#AvSGt=KFj*746&r@(XC-nBctYio} z-h&;Fm=x(Hdzsazi1Oq2o=sGT4A)=Mjdp3H6JtoRGINu2f5`n#HyB0o@z*BGh-OjPin7 zx*3<5ek}NB6Dq@&B%>iNyf>s>I}D-Z@qVC6>F&aE)*}BJBfIFKm7c2gXvj#|RnMnx z^ryvM&t!Y#3ypjh+ZU_uu@bNiPbi9zmSX+X{cQb!+KouLBx zY3wO4vr0!Iw>2Na85$CE`mhXfxGj94+~h-JvM`eOXNDSopCwArD|4Dc`Gbrh?d~EdiETnc+p3K<7=sPU8j`P`0sq}se!sYxVDpja(M!t#o!TaFRJzl6%eH z)P|!pAQmbzJj}n#!Ym8`z$%VT6$6glg73Yn4sZ3zK56K9hjCeY;Aye6mUz|c%}PV) zEJ}fawXL1f8a%gvsQOnIuDM!a@yiZxT`J=t_1ZxCzVZoB~WJk6_<`hZR?s!4(#uXY-T zUeZfE%tXV~_y^tFUW?n}zJl!FjuV?E>8yVJ!6BOv-vyA@;-E#NvQo6gBALB-LJA7Q zP0*|5c&E}jD)F@~VzQV4TjB%= zArvN=8D z-9DkgvFpi%L3D=F{^=}QB;-N&FGJ!i5g?YDGU>Y}cPp44*fp1kgtS`+ze%@>T1F0Ft+^KI_l zLq_!1nsT~APz0-eL4W5IIT#Cu8mmn}tj@jTIsu&8KczSy5Q0GyT-A>ly4TQF`kHGI z0`?{jXJ*Z-vV%P-TyFDZFEcQBUnT{49I}bC5M9gB4`W-mZ+vMsvzTUo&)87ONMkpJ zSVMU-#W${sXuYCSZ;%lZ&VJABg0{<}_D5Hgyy5lS5*ZEX&BtG4 zp{Y$mO7Gq!8+-;Q-zFS??IO4Hur`C{JINtk!~WhV*GHCzn#S=6tLGh{Nvk_TU={Ep zWs8)MS0;QZ+fXS62)tQ*#jS0On5!f66slYP3g1gnqlIQy` zWO<`3wn2DZi&4}9G4^TkgkN=Sa`~Uj3WlgTe`wFAbMRn*3<=>yFLNcsM3O>N%n+WA!KWqBb*eD~XCtDW&z+ zS~C18EP{Hjc!4GItlf7l6jG6Khnb-Ic>aX60x?nDb;04c&V5R(MKc0nZmz`0Do=DM zA!stcyng-|)o1$bb$Oj}>-C)I!{4?dW)zorRBN}vG{t6d)-rgHbj%&-dl(r7O^#Ve z`(XqZqFa*7yXpWfP;*?szn{sNC>fEr0Yt3_?4M0ui!qGw{Pj_5KXDDNy7?`BiSpiZ z5DerOIL-NCDQagx*Y|t;*eL-fA$&q`7l9l=Wilk0OutX{LVr^8MSV|aCnb?s1Runi zy=k~D_gQ8dyMmgxF#hF@7lF=WMs(jsczF1`FDBh&16H2<)KsV&n{yA-k!NNycX}3} z?4(n1T3#9?Z7Pgf0L(Ayi}2sxBEOjE#aHD>AC+Fx;mi9HXSOe1s~%9i-}N8*-g z|C5H1YhRxZ!nFLx-jliL*ZWd<)3$ znyy|V9lV#BQ_7LG8ZMYKjFia-e-Cs7aqz`K%C_ITlQRGM0@4Pz`!mVUW&mY0C=C$?0jempBKQJgfP#6}BP zFQtw&)d};JZBOZg8Hotf?xJ=%q};`nB8fkChHw9(aozb!C{5%r2@w3k5(oVc?V zASN+9;7|(UAkYkTSw=e71VCz@-D*bP9{bucad_z{ub=ez8mfIut&;$U6-I>>rhz>1U-Y1!@wRvsMqpGXyj?WgDe+KM+_Z(ge%S*E@v&50ACI`Cty8=YJ zDQXj9XG8QeZmr|&*gXonR9OW>lgOdW>;SjyVevTq{tIu1K710Ww~f$|aMSkPX|B!m z7neJj5W_59(?FRy9Jn((@GwY6@5FGR|A`rw{+aRYruw?sp2%Zg7W)8&zf9m zMO?1$Kkz$1yOz5$P{S}?Eo-0@&Xu4g^{9c~4LkxzZu;dlO*5>?uYEn~O>D^#CYEkN z|0dprM?r+i-g~@G#@yNuR{N}wB<|a(SVTIb5s48@1TU*d70?N{-LKX1u-B?JVbCa; zV=kgONUD_j@x@R4<&5WUe@W1GX_^&V;K|Iws~cebsM8TDx61iw(?Q8~@1G01 z9ZiXJ(oqmpUuNxTn7+H?DwXASMXs_^ z=RUG)y6>yr)@!10Xf{`*cU&3EtlIlE^F_v-xL)M701dfWe#r4+ZW~5y-_X}NenBH- z=taO6`_XW;VXkHBf<|F4W%_%*@rfvr9i-@kkj=3s*@;)p17ET4PL-Xj*zl;^zVK}Kr)$i;x!#@Pv4;*~I`CHF& z4{f4+PPB3F!3tNN_V2?LR1_KHotY#aLDJmyO|k*5mi%tJeS?2xyez&buSWDEOSHpe zG3*G~k%ass>5j3OQf5eGMyyZh!o9Nn;QRvsCr+n=dqdgk$pp(%pr>9dwIcV@&qgIz zf6dxb<_2EXy{KBXz z%4npf8pe4h@apd}Y27J`|Fb9p(%tsTq6+m+(rSQOmM^ZzRH_BI1!Dc&pOr|*@tCo> zlL>|%8;JEwdHpcu%UTw8^(wSBNuY#za7n|A_Wu0&CeG4>`jDz!F|TUDRgxf2j6)VP zUa?R)Yf4E6Lvx_6xH3_fvbV+GX~(7}mI ztm%e|YrE-B$D(r6^v9YCB)wfP^b|1<zL=zaA;|9eMNjG*6GPn|fco zPL+2zNq@uhMW47=jT(WN3ZjTWXw9IYdUt)>89{{iSnnL!AIS}~^NQNwW>BL!QQvi$6x*E`Dr6whJ&~mDTbcf+{gVTuZ@Ew|iK;QIEno&!gxT zS%d6fT13qPmiHDjYpq)Pq(O}hFZ~T{u6T09C7Si%0o4`THu)kommxV41or?|^F3C> zQkx~9++Ly3g2hd<*eY4ZYaNQvzzn7VgMfX72pX0otuSP(8DcWk7v*>BYWO)6UY8(W zaIupCM-KJ8I(WZKA~8IHx7P)%%7+I;pq(XkKw+x~1PchhU zmvH-Tmi4|}mTpF8gbz*cu99KWNkyoyZX`)?=t^kx%(!U5s&qc6X|VItQ5LC{zYz8!Jjp= zT3IUqt9Qn>m&5rhIaa0Mv9ySjQ&y=%@-F7>vqt!jXHPRYsQz@w#3+6$OgqZ<85F-e z|7KvsLD>flv-rI5X*)gMp{Yxsf~YNyHP5j43&~PxigVgtv9w#Y&v8{a5}U~>A4>W3R4k@Uj2L~V+23{{+L>0}eTvxlvF z(!EXOZQ3S3=p*Q!jI)d$y^@y)X_@WX?BaKccCz;$Z4B8f{>bNa6JPzP-QBdXvQn6( zrck%DYf(w=&oT7m;y}?A-q5Ng(d(`6FL`s?J7;I~b$7*xveUgRnFL{IIT*rPB=O)Fjjj*1likO>QUPG4WzUhVE3Rb6vY>WLSaX0|2r7`{syZ zz$2Lbes9|Vk>f33Y3hB3+{KqmFMhvoeOuZV)woW*^?;RR1qb^A*EyzGpGJ91tURf- zA(#iQE24xGCCCE{Z^8S%BHO&6fB{{2*8qBro4<|1gy$sTVI;*slSw9K{A69LP$k*B z!ez)Ts4wD#{`zhdXU4a${I(%XgxD>o_-PV|KIrJ_w%4K6bK;j+v|IwNel4{@og2%j z;1z`8D^+X6hl%ug4)22%w(oNbE903sw3(NCK00)vz_S#^m%yS{6S8L~#-=SH9zwil zgPNlm+W^b!RlD#u0f?|N-mMunagbaV+P1ZQQo10JZ0;3RYiLa9blp3TI>p4jDRtd` zGXY_A@LdWZ3R`UrDWx2wo@eh06AK9?O%>S#p5jmpWqOlUOS=3ZZ_w8!4d$)VgH(KH zz%;!CcEOoSA>m_VS=Tm}hYoSu4>2LW;i?O5cb!)7TU9vr+cR)w99DBEW+Y=nBvkt2 zt#;E3z{HV>-dx{gD_ODYcII3ys&BonPB|nW-}HVKFyw+596*OMYKPG_xGC*tfffu4?1lOFBPL7U7LMaxgPIVr{~oz{j!ka zV7#X^Af)!_`SOp$apa)6>hUT=cNcs}9w~An%M9Kc1c`D!wX*4s=_NSusmyVIU*Ei(4rSvYbn@c8b$3}VV;QN5h zp>4cgA#Vg2<@9|^u;Sn4w_0g(>KvB=c2xkctwjGZC;QlsbnD{NwjY36=eXptEdejz_V)lM#f*_4z+%F2DZ$vFyv7|L z0b)Tcpa2y71NN=U_Odw&s3a55h{sY>`S<4FBM+FZ{Qn{C1&C#RW6f%^`sr~ms&p)>snA1yN!pyo!ugfQFoC1Ht&jaQub`ul;sA>@VuRH+g z0GRv+m@R5R*b`mh^4-kFd^Nz<;9>L8^pztRK)IQk#3`L0cB|5ejy`kiAm;$;Jd8H_ zTh*qA(><_)92_pS`u%A&PK-dU$wM1}bLAAnCi$frP#ga%bUV=}%d!lQ z`9c5@-+gJnD6`pQn(+WgXsucioWAQh-D$161!bNW2SFEUUg@zpN}JLFcwqp7RyPZq z#Y}FNjDKZ zMek3XE~j)v}AZMmU}nxLf9S_bP)FD&<)q^Ni}}+7J1P+;be9g$Ww)$zhYHVgL3n zar@0iHS3o8azxR0JwiTaKcc`4o$j+E!dvebhwzqh6f6PL!~2;$Jf_5!N5(AwhQV|0z&TX(XJg`{`)ik3P{z z0*X_V`WgE1u;nT$X9*uoXFc66p`aB_ea`Dfq_ele8R<8seG2tqI+C5ZAl<U|oyQuy^U88t zP&!457^KMh;}DFb*$nekkqO}$PO>Y)PkQnhYXuX`!RjWYIg@ap$ zS1%*c&3@64m%0HCZPO$ig_d=6US%;neH({B<-*4u75>Tknvp9re>O+o<#u_&k1Iek zQzLSTGSiHR$U&A=twLRR8obtJru2jHbD(y^1LB}X>`@%^YN$_8lS>#&nO$Ie|-uOjTR!e{s{knMPYi8gn-4dpf zPEy#>67rDiF0u+RHm$Xko_r-31F!yVSY5~hki$U4Fni;FtwF8rAML@-{aIgKn0vMW z{}Bo#|F`|oC#D{;LP2|xwLDQb=o{1rtWGcYPeG?XZje#GJP=kX){6d|+g;v!`wKxA@;yt=KKDBoZq^7xG`VA!8f ztIg!N7~o9rL5tMdguodS*8{~A4O!XfE8~>f!rGbb z*1WH<$O%t=W&v8?!zA5FzTQ#&Qdvc%a5k0zBc1|%oS%_daN-K`IbT7dPRQi%Z{Dy z_uzl55%pgm(5oTy!I?yEI(^AA5=xPX7VU)AOvV?F`f=OlyfzwqdL&v8w4gmuEl&vY z5WHp^=J0zK&g~}S1i1ryPbD(M0><+cV2TvVVRzG4}6FaCb3 z_5AScZ`bvK@97c*patgW2KA5XxH2`_;DZ$S=)DL9Vpj`R0HV)_hpBgf;>83ryNzYiM^DyAfIjJ=V~6BkCI7XFbc)8@{qk-Ko=(0)1+wye4yo z6Kh~`8%)kyz=1ukQ)dUt>@cK<&n~x%?QNH+!LJ9n8w=->4Hd?KiCB|}ZJC<7(tWC= z`C!8ZPiBL0uZS`RL#)b#`l6$!T@Td|H1eBACe0sproQ(aZ$6=RG2q7Mk)pDpPYlJp zY5+s%s>6+yBv^Wz$8ef9*V-@Lk?^8qB?QA6#dH|FKqaj?gP^?09X_kKkiZGo;M9ob zQeTOmcO1dk3_ZxrTW35IVUV6CwF>h%gmagV-cB;e$osmPgwFAw<{z$BCX1HJ7pr5T zA*;K+e4{DMS;UqbrEI_@i8C}2u#K#F9f1!;<+~DB9|`CiV6Xl7nDCHWfCi)7x{|{$ z(ab-3cHiR2+20B?e3hs|_HLg`k61xh&48U|Oz^gKRS9-b{GNZDYu3UYc{QZEgjH(gHXFYElQtJ*^x>twC2XaU)}o7=`?p7Rs+yFyhi zHzR}%l+{6$LiSmrKV-rZ6~T+GT48QfEQJkzWeIET%V4;>c@$p7((Tj1P4GW(KXQgg zMMnGNDojiG5L)JPU^`QBb>YMFs_b_=fUgMPmquLWBY)S4^@h35EVux$&-AbxI@Yi6 z!hU78cj%tu$*00TRlT%SryRc8+u86-P%jhx`NnT9J>}Kq+&~Ta1W9|p5331lHDj!l zZ63^PijV}YL|!U|IA<@uVbvpg=G%X$9SxO(Wcpm@Twe3z(sw4iA5K9DavC-8&(cZB zqC@mdEr})&@NaOh?G^&Zln^aH;@a2Ew`AjleY4hPNuq=?_dBT3k=m%q^oY(3U(F_b5nUpfWzrJ=X~Tw%}@8 zN*CQp^h*r-V5>Hp;s9-YDJP?D_@PKOj!S4g6a#AHi=S2?zMtJ-u+-+a<&=nk>7@P2 z_JQl%r%WI6lDJXLJ4{2Ou}$Nr{?aOk`DNl8lEMI>HVx z(U?9}Z(helv)p2JbYh~f@Y$6HI3@Ah;yW4%l4lrMfZo6Eaz=6W>_;R&f>j7YXY>6p zJKy2xGnyu=Y?Y(_maS5&n8^O%4!X;wv+yF@HiJyKtSsua)$g1Cr(!SWA(wIBLDPbr zR)qk`_YgvzbHn#y6EN1!sNQHITwP$X>_Nqz83Z}EwQ?%jN~uV0CgsjE6BpTK@;$z~ z+GvrW7&YIH_fI56R`=)QHdKYS)rl#R60$j^ZM=}+kj1$jQ(txB$z%`KNTt?ISafUf@PHFa9%ZjQSo|7tDj{#h24nz1xk2(u)_W z<1#Egow8(+5=iwIiY1Oug66mIbX*y|S#Fen{Gsxf+T>7+oHpN4mZBZiWkKGI zoB#*8+$~_Q)|73IeqYA8OyHvQRk517-)PuAuU%hh`-W@KB<5|B*4q+NY}x0k&+qPi zRT8t|d0G`fe|vVk!;#`Sdyz_}URj$0*T#w4|5F>v3KamUnvrt9EfZQRTH#%6{_2uc zj+VI-ZTR&{dOMp*#08Xp29nb1bF)xV=Kd_RCAb}lzqH@ad-vkK4?d<6TS87hJa-dl zbhrufi<@axo>!-dl&_Mw6=FcCRp>D_U)C}kJoqo*tYx^iF-z~qZ6S&e-2Em@F;q4iB$tM_3{_9!7!g;oMXSCEv;%fX6gXB6liEq=r^<@LfFP3tJpv1S8+!b z&Y?yT+X@>N{y9vnX%*yWM%O?vyIK>s;t7wk_gSD1v|W^V2=kTew1a z0Pu#G;_yjURzuDGc!u~>3W4Q5ek_|?sy@mDAxCV`ET<|)&KLL5`fKs`DLjEs0KeIH zzUB6HjjYt(1&fZa;nuc{^e%h5iMqQ_;2brNeZ`0bGT)+#=@0uIR~n*!sve$BLlvQP^?l;yP?Wm&avVA){fVtMg>=f<3^`YQ$X7zvGX}Ipb8QQ{ZUIWKl-)sxd z0>gP6)hFDdJQv*+JskGCtEa~l_O_uHGAI|D_fM+Jnbv+^5BwXt>8kw7jU9bC#Zjl? zucHFPRB`b7sAgd|iJ4)Y^8^y((g_uizu=azJ}tTNd^Fi&bDaPh(`ST4DfStG|M}m~ z+@MC588?@*oU_ijf2c`<3)er1rTF_hyS8LJpP!^Vv(QKFb@3vDntE8$Ig`k z_>xMweJ&Yww?RFePM`fzhC*>xC|N}?$~^A(_TF}97D9e=Z-EWmpU$A{u@ z#-tV9IU_ky3!>k7M~N&TM`7nsA}R4L^|!67%cs*Ca|T~|J#ZGefmwZ zgMdN*nu@_TDv)pN{O&JPML4_9evnX6R}KPaqPQS&Z-hso$1Zt4LNNdMZy_g)iziT9 z79=VjKsBR%l2)q=_xH7-Ta&{-8y@i3$dt`+$D>t30Xb|LLR8IUcmy5{T>DfemnYYY z{K|Guxu3v5hvJG=rLWUsfvcXOh+NxfKDCm>%7w9dS+_eB2wwEHdxH?);}&I0;BhOb z=EPKp9y#VtDcyp zq@NDUC4yF{-X{g+U-{F2fqyu)LMgZ?lAm-71aC_&ZRUO}+gmIbd8+0tlm;&_0q@x6 zp^B0-x!SND6jQ1dlUy@YvWr%`T61@}Nd2|!eN`Qsa<$Yq##Vq<2bL9hVg+G6wt|-m zC*6{8v9+1MZ>Mz|H=L8?b1~v-_G(zC)eNoAk>g`)hEZ+In-9S5eB(j{TQcQVF{YR) zb#Y$fdm?Sos|ql{(T=6kRmIiDf$C@F$ z^z2c0jWmOYtgYcCvQ+x=v{fxH0t!ha@9xLgnbX0mFka{exur2iR>>GBZ@YlRd!XD5 zK;HlF!-LOarJtZM8D4z<2jIcL>hQ@-?u0yqOE=i_80cOcV4wh(Q!(^EZoSff|6t}$ z1Zj#||Gpz~j_27SDbntbn+r&@mGlyCw-)1N{J}y=J$V!rGyFGj0lJVMoV*VdB@QT( zM{6H}WgejB7Gr2wF0Moz0zWwGjJ)5fy2an8>dM$Y%sALA1)sb#a+|LI`+Nmi{RVZ7 z_U@H7D(G0pMF(4{2y-tn+3h|4fSN$U8%DJA!MO4P5dD!CQ$<8EY>^E{Ij^0l6GrF0nli_=FR79+y3}%rB|pf z?KaALdKigJH{kCCGY^DRfU%@oUkNt?5~aOhw=Nx7=H7|y@G#^`Y^$JfK#I}=POkS1 z@G)u6IDdWYec0IOaeu=HI-n8nrhTIR_ifulw!_*eXG{usSxa&3<^@wv8-WIQP*Pg) zc~4hyU2uWz{4aB19smOUK#nyj<^?nCvniwqV}-8aL_O(R8iKOkm>C zC3ktI&Q~}r+55H<&(7>a{uVP1=9{?ziLw4p2h;K%jBIO)M3V#bl80D7Q1&q4>uK2P z9AmGZ>PU*solkJWX1~E`uw&%I$Qr7JGwZ`SE_hVS=%HNOMCI%B*2E&(%N*c&KkGG; z;44JVbKQkL$#(|RSZ(j!xQghjJbs8>j;4r7!96`K7n)T>zABWz{!oiS0w4&@Y8$L@ zvt+?l())q7M=-UKc1K0F`7}fBQqFTVt z=g||k*?Cb@))#jV1w&aWQgZH5xmjmu-(ob3?N?58hyOM8E1YMc&+Eu_;CbJ9!CUY< zDf%I$tnS#0@5+VBuIo>%9rTT_jOi@44{}d;|5d4t{T(8aD8QK2K4=4kqOjU)k0Q$r z%H*zH9>TIa1Rc+}%9uU<;}#|CSH7*vtTDm29=rSBx-55?x`vDakLUXAW~`ndk;rr8 zX09?)c*$q^nnyeS^$dYYrYj;Dg!nBVdo@l;JyM%$ZN;s^6TSr`PN${>+mNfnYqOpS zN0-^hfxY!>$Wo9uOC1UN;+~1gt0n6Iy0*~rwSWZ(y)?1~yvSp=oX3bs1kY|Ll94D~ zOZtAK>VF%YIzK>tF}EEkoh<7wNRBMRHez_9^F`5ravdYfeCp{_?M>k$jlU93l)7q)Ax$G1O6#K;__hV~D0l%C`6N_=5lW-wiG7F?8B z?Hgv{eZcV8RwsyN8P~ki-^UkDWS1V+i_S?@&lQYn(`DBdBBL_syU4Z4*Lk1od~Fm( zb!+YA2pd7$5);X2ejHAJm2%v#dq13DKUu5xovvkAEU^&7N3d$TE$%TkZVccEPg^Cn zh%vrEC2VO8=&Fky3UoAlgfp*(UblfHGc)>`?9Qpt zxO9aj8Rjeb4aM`hm3%6CzSleSLxlx6UWSsx9(ukl{RO=rJ3=#Q*>4UqglUt-ok6E5AecdzOAISz00fL{f8A>U@rVDzJFiqUe8)B%SsJU(bL*< zG!Lq=ozQs1g-wu;WY@pm9k6)D+!uBKE`^{M_pL1zYo;6JP0gurnK!+l7h^XMoiGdX zM;QyJuX-F<`rqzZpWks=;?bu?;_P5^MzY9e+t_dPKb|aTeh`t#gGNRtbNr6T_K*)= z7tIIH3EFZJ^OY@6zbN)bo8w3&lr07sU^ckHUYH%JdPE>gNQX&wg$%YsE?;8{yy%rs zy@vTANz}eruBMMD$fSqZ7X4f`$3B|SaIZbNeF~${CU;4yNT~a#+(J9N_N}lKyoa6Q zNGSF%5>`)_=Siwn+Q!wAr%Ul)WO8%3=r{oZPmLnL8fyq{a6ZWOzTp zZdhP17gTGp^aPsWHa+;hcO6c8PFZ9Mb?8}l9yDx43^$GALviUjNk zhGO=!qdFPy3}B7?S=jY60?`u|hf%fVK-~uss9UNzc)Dt%eS?Om@dMs(r%P)m(^}ur z2`ElfVD^=*x$fe55O5D!31M5-$G<*J(q+*sD;}x{c_+V-t(C4DBd~r*cg}h~U-pEG76jkld?$NX z8UH5f2CMYlL1}?8scC-p5O>PtJ8?Hcy&2Qn2ZJ9=Q2hQ@{ch@Jz`}G(UOg z#r+N^m(nEMGpm4$Oq}4s-cS_ptsPoaB)_64t z8&^qg+j_ny>aM|t{?Xy5gd%h!X~ocuY7=GauVRHM&U#?#OTUp;9kR@EPuz_ z4a|E#Gv$?72L}1QF+Ort^E{DvcZMN$fBw^XEwZSaMw2W08)}|Tf&3=a;9+J<`LZx z@c`Vz?J0gVzfAihrZ|N+gxk|NH?f>z3?z-d-Ip)EMTC%yQ4g+yU&|9h|`%1^%+mB3Si;x!{v>Vq1pz14}A=TmF{ci|mq%S#6|@klPD z|E|~r*Sm^;glQHuvz4-yb}?5^E6|d(WLLLeswd0W^bhHy9X#KD*5>egdJfc`Q2Y#2zH*&wsWNQw>M0%PL>5?xWa;7pp<~)` zV}=Z*&J`&0;Nuf>B}YMf84o8Y$ocO5e86g}cQH!AH3AePG+ci-&iC@dx5sg$@=@*r z@bLvNM&AY&IB@MFXV5N~21ms#RRP-w$;d|Pj}cdC#7>st1=eDjlFVuQ3YAezbsG>q z-lwNtEH6prP-E(N^xx$1m6KCTm5Z-`79)tJ)Xammlk>fFc^N<#0J9($fCtc7ZF27O zZPNlUfo)50S%>BFE< z*Cq)*sIbz(EmGeTa~r08_j!iK%=!MXv|ELT%GLhm4;^hY$S-?8c6E^?Zbz~orUY<;L z=F6R9^wzzYz|ytO66}b)ye#;e7_gXakhp6|u!sEk2y7h`80BvrA={|cl0^f4@$V1|zYi(5V3kjPF@w+F-$+N@_kh_S}Ez=-Gl1cx;+ zP9z@u(}wkVFeBYvy3q1up;D%8iL+llV_>%&i~Sx!K&1cY50(nJwEGvpNpPbZECCA#v-#l@AZ{6%v0yBtE528U;yA7x zOAcD$tdhQ@T@yaBZ@nu6RA|RjLUwg2F6laA8dfLhti8+g&+a+9!0-P*2vZLgr(Bx7 zL}3$Ga(%dYE{gT!Ys{XQr!wZ}>MKDi>d(}=>xB`evHR3hNd9nlMyWSs_JJOEwZ*2M z;V+q|L^t-57Ug-o))pQTsj|YCWmnH%3}f?;rUXkeEJ@OM1Pwg;Z)cSNiH-OiIK5qG zUaf*Mgj0FLHv=u(B0B(Ej4U+u43pseHcl{lp$NNd+e{JgzwWG8?|n6qU4~mLchM;i znd5is)y2Pc)kVfGf!S<>LICbhStKa9AL}c0W_>|-a6jm^-1iFiG78qnIfjr}U$iJH z(#p|4MiE!JXmF~^qTi2>IAND;vgkvfD^Ha|2;7=*Zq5Wfj{6w8Fr8|Kq z!EoGzN3r?iao)mtWysZk>!vpQ#$o8%yqf9#Rgrj=uX;p!w{7<5F?=Ekd;?N$IqiTj zE9Fd<5pqjSoV~jIc=W5yn7uq$GUy0S?28xAa&7p9w$P*z;J3wjQ`ZM&N1o}}MuPXp3H9;OHj~$G3)j2qAcJ` z`odaWJ%5r#0~3`~k=2=o-Y?K{9B6(l|Lxa8lVQM8H+()o5(!~afPwfx zrnpC-O1*)_?M9Ew00+idFf`^kN|v_%7X@)`b$M(ZMu=we`jXlwzNQE+pDO}*B$@dg zQm)oZ-%bxwDJY*vAWuS!*Jc*=37yPfi~VDn&e z9ziq<$!2&ANe zFRyP5Sllpyc&K8f!4gwLNo;S63vBz~;1;j?P;rg?0v~D*{&MaLRU!7rA`~n&fzMx< zXr>2CASL@s8oKzQIuaV#{Ibi3=N&ql0_R3Vh9?^@p?o0^HPacjljdVl|K6r&k21cz zjsN;&T~f1>sY%lx2=(+pX7kz_KI)dm8O?xPmzQ&V-G0STPc{QY?~eWR-B5Yy+kxF6 z?-W#boNRi}`An`hr`=Dh&FaKV;|On@dCu%`FA*9pZ-4e3!Ow9snm8djF)?vhN+p4? zm#=}^W#+@o=JW0Qg5LV9zNgk%lA2SdxN8NuI;i;k=vJ!H16JkajfEK21U|5IZ#1Zy z(^G0%gWB7u(e0vjU}7V`5v)ILYbS~+LgnJ#5r)_5STk!zt6n)2SSyWzMV=|7?7=-aBc%}V z8)Z0WfplRI%e;KUeTawaY?36o^Ij~)Ne(UL*1jhu4pV2Lwx)=vjSA5YRMGtWDDXws zpYK86UIeCTJSLOCQ@-Y+V>YTwdUdpLDUG9KA?!Iht{3rHC6ZgvY zk99MSlz8{?LuGw^HEU$E(JN!6A|uHL0c!*B{12bh-}+mQw3#7Auy?CsRK@?zQfUu* zUpHC_3w!{dvX^&~?qcarkU;s zc3tWHleU?r+)d|i(hDO_AbIYD;Jj0eymYLL;DnbdbaMdJVB$>3XMZ`W8LrB&*sjD>b!49<9oy zLy%LU%{H?r?bzH`W{=DFg5-i2a%k|0ICN+)5zd`0gcQ%$!!$CR5sbi~I_LeHN0*Fl zB>6amS)GnIege8qRtZGLlGpks6i_UqlEGp96@+f+_aCZwJ|uHyr!Ooq6zmz&$Tlmq z&;;xsr6q}UUE1VWn&W$!lIjtIwKFX|)gFEsVr9B#tUf{wCUdUG`=Mp0OAO*Bj+CG6 zKyFH|m?cYgJrxAA=v(jF!S3KEkHH4mF<>mVHPMv!8_Wdp z?NWDu#gc2mZLK}PN-EQ&;~rq*C6~&_BF7GI;?$^ihLb??8KnNoC=plW&=I}3ghPi; zi{WeYCC#FYKbFc)V47sxdFU9`(0v%LJf_*y^?e0yve4AiR*_|_@4g81R+P zm>JwpF$cvL;_C+cSykCW%+<;zudl@}989qBIf5HdEHK4dao?l03aSOdWr-_?xVntp zgCZ7}Bjvq)fbrI&6f6K@o+WT-xFAqGG#0!5M?lW$ zb8JbxJzZMdNw+cn_sRb@HV&YAEWTPJn~OHgD%)TA3~fI!*cuc&JAJ>IuT>FPzBNv4 zW|hSF>erOoAsARuKbrHi+L}4)zN@%00J3F{@}bVPR;iWuU7E^A-zN&o@OLXYazQDC z?j{q+#abu-mcps1y@8`165!xcWv7Ld!FP;V0I}WW{xNjElKemvIu~lKsb>zqTU>x{ zz?O=vdPC2#weTHjHe_gSCVES+k>zj2?8X*;0Z5b~vOHIy)uW3pCd-KjVoNy zz^5!j@NP0r6pA8eI*kF7yvLv1uKU8J>C2}a05bueqJMn{Pnn8M|Xz!v#kJGMn~CTjV> zKsTUCzpTGIa*4AJlj}NqQ=BxwOEPFog-_`3uK(vHZ}lslN?fgB>G){xDd93;Y*KM$ zLC`<%PjLmv{18dxU3kd)GsPWFl#ZhNLCKy_80FSm-mB@T2@ys;jcoiGmI|K~fypyJ(>&7qF zYE!mw+Win$Q0uDWR%m07Gl`kK_2-HFft}va(f<<|d{j=6;bcgJUtj|?-DI-S@~pa; z_r!)RE~jvo6s?hH3NMnujAOQVHkj2h409-A!q#FNM+iW z$Mos2;<_nLdR)JX4msV<(v|uB$$P@oew8GjGf9G@Akfj=Qk{XU5cd1*r@g!u5Ss>H zxB`blEucVmkpk7q7`=UNY7H(=Z3Fi#uwl|TV7&u;gnzFbo8SkK5MegqN1Buc-)393S?x=`FWJFIs_Z4Y|D~dIe>9_Fo!e1Pp29k#KqFta zIj@xgXC0xhHS+ObBeWgAqS}UL2x-kzI+cZ{?hZ>+Dd1M| z{)JJx+`9Wx3IK2yta z$w}z&TLyIQJO|DqEsYF3)W*cc8zLOi;H6eAEp9nZVod>F<5-Dk(uKLMd*G3Ic5Zzo&L)tw;zn7HYfBJYCB6-q``8TP{)*drj%%)DJecpL zL2>_^;45kNFC`66P%ojuf4rcp6^WFZA^t#G*(O`j11s7@ma+{*YEvnfG!uRMs!}Ak=7R>|PqW zXnc1pxbdLtL(Coe^||jf*PHPV4>~}f;R&ZqvlrxasFd*bpPcm?$s=WX00T>mb%&GN zKe^nZHG2iy#nLwPk!X$u2mMNL^QcyH^|dRjmK^+Q4>4 zvG_@tD|YY5UD=dUh6!sdIBQEsLO?_e*7MM43qH{_47rCS^wq!=Tj`MMg$e#XUR*Ui z`Br0qaGEdgDLD#QlR4qNNXvC$+q>Zf7ou&pkBB{g?zHASgSSu;;Njz*g$zVTujgX+zh!(>G}GJUkx;uzIUWs+ z>B6pMuw^>B^KvS{+TqtCWvB!nm{r0IOYdrzVtNKenri)z ztKqJdA)zNP*Ups*My1sAfsgC15s$a(c=l|%F%Ub59DFlu+Ic}M(7tbAyHW7vD08c@ zT-?9pmYY;zeau~8y1{td!Bd`(oISE0$I6%=NR7yyI(L-m{Qb4@eaeMD;&1w-H(^z`JK{u#Am;rxz_dnEkBDtdjnuYL6WOtc=QLMqHK@AOm1j^F2g&kh^OPge*_lyg4 z&Ql9}-g#zwBm~1_Hv*n7$T2;)8zvVQ=es$R{XF6VT;v>Nqz~133zFsMD`V=1ozl!= zOXSbkk?{h$7_^pQ%hF4Db!SJjtoiv(qEOE8vyK^8XjQALX6tS~8u;L1IR_j-?Iv~c zhuaa52`8WzQkA+eJiwzgn)r3dY#4w^Dm}>aKY-Iz-a8{N(8Hv+mWa$u>i~M*7P(tl zxrras$Pv&ht0Oo5&=7e$^AU%R$SCaCIbV&JcjsO%W$=dbvEKtpHzV1J=SFV6p5r~0 zZF>WbiJ}wgI20HSoR$yHj&W@=fT(qh)}>5!d$Q%gmJOSZ$=&L4V>73VYh7ZhzX?R!-J$-V#P9lA$sk^x%tv4;u-34;!l zD=+W7>bz_}P!U&1VN*N7i%fb$x9S&?#P=hG2Xzy}?goUic2fl*EeHC9TqzrjCBqpe zrQWk1LmRpIY<0xVJ>yaON?q>kzAc-cYf4_7s6t>Atojo!90w6D3WQy04%$7&+p11| zmg9;*zb$L*aE`EVoZ&}uCX*NOKYPIbnoKnMa+Oce?WVlT3 z;BCWvz!p-4#*|A30F%cS7K+cP!X0o^h`Xf!zZgP7P7W)b4LJ?uRBge3idP%v?pQU$ z{P0D=*;G;Gc`&v1SCk1e3FEUwEjeJ9Y_X)e67P_mgMabdf{KSs8h0)DfbuE@i z>HBvIo&ALEamyEl8WOkK7BJ+7A_6Oe*pZ{8v%9H^*x?+sNhbnai21gT;~{v*@5HtD zYY1pnLmIXrhf2$uIuE>U%%IhpfR+Gy5VEbemITUJ!BNEE{3{5y&pmLh@Zt8<#6pjK|)ToAeiP_)PP5*S_AH8wp=MvCaz8kL!HA@z11 znl23NP5fRI0Ekfa5Z`R+@k5X{+v-|4OrqDw8O+LDfbB8{R3hNR=?F~PfkN`?0o;s~ zJ^ap@2N)A0`xll{fFZ!Hx-CpuZUZN5Zv?KN0rbWLAA-J%k@nCx0nzEYQqXh2jC@(& zxo2{qgd68hI>a=Hnuo;}fP^d284AqD^wK59?)-a;|8u7W+1|<0cbDo~?h{0Y+4n~E zu=aBflm7H&=A=ZE{4+Al~C1EHfIP| z{g*WKtXJ87zn;`nHl#ohJCcxA`|bN=vN3og$- z#3O4v#kBgbX@I-4aN9p9;8KZ9l$3eWUT&-{MJ8V0f`3tXt5WC2H57$LRORM%*l)IO z6e?Sxv4}c2l1nMmBG5%M^!&LYF^q&Ax3;Gh6vid^dcrf0GgM@2>)ynaiTK!NNH0bh z3Dj;VLiv0QG`*7EI(1OCcfE)_nUeiK=?r%iVT5&sc?uZP{I>;1LSiYI6|YP9oO+|w z6{NoHhFzZ7hi@KfjBV9i?p6z+NbS)w$IpbHsy3iMQI|c-yT8|KpOX6GLH&bojnx}Aes_B->-_f%l1I-Y>~U4!(w-X@8#fNfZa;s*`Ir%SrrhO zAM}xMY;4K{)-&t+dq5MaB*N}z-XSpZ(zOn0X0@5{6SfO~!?_1bkvt*}TO(&P`xf`= z2H5vQj{svg>*CxhjP#{AgH>HYk7A^rW+sK)3?BuiD0~|1aMW= z|9St*_$f5GOWRz1ABc*UfP|D$G3=Sbo&DZas{xY%GyVq(MKLM(>hv)3KSQR%jbvV8 z3H8sN^CnnCzd2~90GQ|;GbH}pS%+*)8|x~P@P;NbaeK%ZAieJ=oK)x8?5{!iup{6vq(Xh;zCa_6u}KP&r-OL(gP2!+uo$wr8n0gz(`=k2PMiO<+sMMmCASLHHF; zEh@Ffm%MfEEU8sLt2^b8UL!85r?J28u4JFw^Rd{JhtOF82go-{l(H83)3}GCtKD#n zWa0et@AJKuj!k%%_Z^H$%TU_C>3WjUKaK?Fvuct*?D%;SCAkK5%LoJ;gLg;xo z1~Y`=y9&?|RaV5X=9uLwKWUq#xqnSdWo+ zR{atCCDmkoA=aJhx6tv*ZW|MNdzN=?>ek&6zI;MEO+gk!?zFN$Cyg1r5jI zM1zxwCPdx)_qjsxbbHX2H!1#`78ZbF6vo>Cqv)i>c*|D-x$rkk3I_e&b;>-f=`##i zy@1b$AP;OA8OCET-p3dQ9PC2dkHZBJ3&8#OKM%Yt2~(38S>ch_aE#|OI5{3Moc8F$ zm7UzsCuH|#Y!aWxPCW9Pv)k9ydncM*HJBna{Gz!N!NBGS^OX43=}g)?B@jRmw|bA3 zD>)D%*Lc;-HOb|>#HGuialSIJ(*8h3)WpI3$UUWF~h_OegTE8R@T zMyEVkSvR#OiE}RE*|dL)Gyquz-f?jNUwgGYr_TLp7N1f(rC-aedDOopcvy0%Gq3hr zj_~V^^gH}!hmO*?_T*m+vuJ_*w{{MbH3ev&@q1esr_u>hxAs?PD?fD%;=MsjV6s!= zL=3h0|L*^TYwPoD(uecX$lyZ_pejSeOaWIzujh3O=lr0ibrDJIxl8d`rIn5eC~QJShiiKopjGv{ z!kT5&@NFAbEet44Fj`hl!IDY;M4CQ{;-(*`9yo?sxGcu}0dwqrTQ?S!d9E~<C<%H^l~0KmJ*Je|Z6A4szEtF&~{B5P;C_0HX76 z5a{|9Ns<>T2S*E!f7x$#f`zz=qC-|1CX)Hu9uIBqoY{(|SJ=HBnSVI6Rz7;%8DkB7 z`fv0Q{L=L{5haQJvgT`?XDwZSe)v31G2-g3N_X3D)pYe6Iv@M_m!g%e9`;T@7mp)( z_L8r|oWmh^rhe8_$Ns|Va@zY_eBC=f+8bphro)zt8-6toDN(epX_CXewL_Bwyq#Ki zs~R7t|9Y#I{bC|8Rg;hepGrT+B@?6>@%}fY<&1y?C-uC5C;==#TMJI>i1Q(fR=tf?n68t@KpX{T<`U^?WK+cajs0lDB7I+Ngg`jHZ0FEZ5& z&T;sE=P0Nm`naGHp`p_R)1>a`WJokA-aVz%#=IS@NNt%~TrdpSh}>+AD5tj+9X$Qv zJpQE;F{EQ%zxBP2@4wMlFl0H>=}pk6N>ymfM*vNF>x-7~KkHXex*6#*27~2>hG>@M zrlG;Qh+-}Z$r6mUn(EoYCDQxL{UMTT{*vz)4&(Ky7Q+%02_zy}=_6m824L1EoMIDi zq~ddULZlk~sOdSDdVK9rl${5+Vp+^)<2cf+U@7%a;VK01qkV1zTfzIw2|7i6Vm_Ne z3&A;_G6&yEgYxVn_UR`T+Ssx~jtAkr2^c2i@)wRzFBhncZO9I=ae5;CJqHt7n4AUY z&6h{oQjXZ8BI3qNPF4bXyf@U&*FTIK>$G1l;t|ZH`j3tUFKjW6wA}O~0Hqq#MU{_J zsFf<73<{6`++5>B6dW1XX9XUrV>;c8I1e2SPF@_GhqoGq&geY}w`2G>b9KESO1(Z@ zP-~HKCp!NJ%doSV=>b8i1hBF|_fU!#I{`6<-OaDf>tdux;59M#J-pt{IcwFKsZ(+b z#`PIHB3|s@mV=$6G0Xj3;%Mt=qD@sq&FG(5TZ?!UI3bTbe9XbbEK|P|Zpl*^p-;TY6K(*VLQye4RE) zlE>TK&BNue065yqSljD1+0nt24wYxG#*J70I&j*T3)UnvD#l7qRH&$mb;+b4SR$mV)SOwJ>X)}e4VkocU3U>Qykw&ZtId~XW)8O zDPH5;Ql(l0TSN-|j$)m9s5v-;7zKS6u5LiQ!ndb4sdSY|k0DgHyrPkF(TrkEs`WYx zprGTjAgZ7kr!FFkgZ06};+gEYte3gn)_&p{M=vYECCWiqo$+5#s^b2~6ccr8?}Wn_d!t#c1bx z(WW1alXq5f@8pA2Hz&1yJ(m#4On}!z^}m-%|EKHcxTqS|EyAm!eO}(@Id2pDEB1do zq254uni&+giIy05rS$oiMJpb6-t?&RH%? z^L9e}qjWv7CTHcth5~txmnvL|^)3*h;ep!gS;_}0tKTUGZwnPuP~C34?j=UC-XQ{8 z+`sJxW`353#0>+s-`h!V@FONv_Ak%f7@;y8O1j8H}BwY7KU{|>v>t7yIwnN)fxxr)OG?pEmC!25zsW-wH4xui3 z34n*lP^#U$u7olb2l_6LgX2^%r`j$rYjHhpOJQ%;U0G9nVK(7r-2t5MTW*|9g(!x@ zMam^&f`VYkE(Zztks-cs5b^pi=#yGPU5}aCxGz-W%emCLZVQ~nkOxIx(R=?~D!gt| zcq)Z`E3i)E$m2CKcXrBp>7a^7h}T01t`-sVC7@d>t`3c3fC5IZ^JKgREfDfMz5$R3 zNGqUC!{F3rQmb%7G;Yz(g)QgglcY(jP}w=Kd_&^z-ktDGc>kSqv z0q88Sn&E3gF4d5kTYCm>v|xW-{<4_?V$9j?`p+Z(4hDaYh;Id27A?L22HgjuXYM?} zrs2pfI(>TetF-&PV z)WEz~{PPO9e^lYBngSQ1$iSlt)lvJ0iDy$zM3`?ABPIKzPDE0EqSSe-TRKJS+0PAI zE`PvTWciRyGrDpA%2o-8=o-b9=hr_zUx*|jBt9JQ-t}gC3>r1%>CwI z7$9G&$`8D}bS@woru867B4MD|Af!=uto>{Nx_#~kV3o}f-wJHqJ1W~=*Ky9r@UdIF z9YONtV5U=8mqTwKcK;&EbT%JSQ8OBDs1Qh_F2nlEU zNgXGQ*-{LQxC+A~J|6+LG!0D6n8(whm*>L$Jy(+@J^;`jJt&sMT&h69rko^q zveB>*a*D+1?Xy0Xc`^aFAPiO6yTscNx?M)KoH5-;i6lHZ&UO1ADMJK94Rdz~)k zmylToCm=2y(Hoi%KLvI#pK1x2%4&~vq#TP8HJ#a|xpk@YR@eQuAP?T{qoKhUSZMI? z_#td40P2u#A;L(2>p8Aw7=j+*;m~5v%xCt{6mixQeXJ6B-2RknD_m@f3rQLJA;XbaLn{ms%^{(QLz5X#=KIXXz%S8qmY&&$MOxG4dq<0@YbsuWCXxazN;=0SN6H zd*tLbNN?}86rOyCbHT_Lp`r|ck?=O4oV?73umZ5OZd7 z3cxPN#Zl?8Sp{b+JQ>b4T40%&uGIPdnDG6BQz!Ug59@`X4lv0XXLJ5Z_ufO}mx9&8 z>O~wk;thjHh_FQf1_{&wt^-*c*}&&XPtxGI-Czh2-M;-xXenxvi}|Ue$}cw05TIE5 zI|v^PmOG&cW^ZSir||FhchNKSK-0A4gPSVe2pg%^x7bfRHMVRNj%)j*0%=8gOAVq~S48DNxnbeIEp*rS z9uZbdcu;NLwCqud_d@W{rKC?F*{~IX-YGXXDg#iKj@8ZciJyR^8QyVb3meM8U<*QJ zs@_lJNN&EScpzSItgT7L|NX2L^n5#SdnTNQzk2wNK+8(@XBJ8ub~~VX?n|SQky}av!Q2C)G)FKU`6cNh9Qd*cxc~_tj@vCK5)n3z$l>*b6R%&`gAII6koU%- zu_rIM#YzD9x4u%pN3!4dKcW-)tKKqA;&Nh!OpKank2F**$<`b@6>>c}0d?wPQ~kwGC@ zE@;%#k4}+CK*dv~-Q|d5PR{hecz9K@E;X6&drNrREXW$;)?V3v z-MFZw1|}#Dg2DE%)$J`7VDauGc;-}%zpIqwFMuvp*iR>6Dk<0-toNSl1lD-ZBpuAu zMLQkAcl$L)=y&5Fe`^UYZDl`W5Vq5HHNoNbo^v`t8hPBlnVdhRx%<}XrrhsO)F#`n zok;lD)|{8QgFSCjtxSQ0Be4}df(Oft<%le)+JYPA4UB*~Aa^U))t6A{=G5VzC0o)SKXG{=+I+w8{DOh!?7JIt4F%)$XKiR3@(I zLcqVY{#FYy(-8dnBp6XX*Wu>9MQZk$dkE;HFgC2GkG5vI^ut_CSOF{E6L|7!^corO zz1iBR?J{y9{4m9T?r(y(z)*IGA&d!B$y;I*g zzVdx(et&DbM9z(|3X*Clz>ODOJV7sgQC*rge%9fiNx`Z$<&?;XcRxb#ZRuEh_`%B} zConC4l<#NAaYf?QvAcPx%-#X`*N}{^AGm?yLCXmH4f9xUpLO7AUsgG0^=$vCa|2b3~ka=`i2Z+kZT5s7(DXKc%un09@NOudQWm!+gJ-v*OUHA6vvp1MpC zsk5z6j{lndv)S?vVkUiG@2fH1E}?wceg4ZPj~#_4vr*&(MAwPvWAo0B#4)bb!-99H z_IRJYb20(-1)?+l7iLg1{UZU7_*2q2#kX0Ho!)AfvUvyL)e%6Sw2CJMLHfhqDYQtC z_4)G==412hEz2u~h^2%y2x7KdAUd&PKiH@b!W7&swLIE58Ds!cbG5JZ$bHFugVU@e zQ6Ys@H;<6!(GTVuz2_*hXX(IcVWp4F6fxihR3x@HN))CCLe75ot?y;Lq0h_Mpynd+ zEd2pgPI@>`#3bVfG$qvGDyH<;p4o&gP)QEoxa-I%3%fn%-g7;>WV32{7N#xPvMGT? z{^TbZC`s2vEbLk|Hje!UUg$mwKz-G+;$)EgLUA~yrUIoLaJyVBn@Su-lTJk?vsKd> zf&k);I_ehKGFzzkJ+PJV`=tc1nXi^C-^e-IN`X9$Eg)uj!L|i~OR0R9wAXB2(Xh<; zc)mIhidBVIL2FJWZJwg>maQOF@vywRai7n>&vr@?sgzH&9AjEz;iAir6{6Fj`Z_5v@A`&wcBz+Gq}eL6atv9#gh z1X^XPBT!|3mw@t-R}u5n=JiyYL0GDvx)r}P7uCX+`(7teFC{LI{%AW}PS(SqICH3E z@~qjG@@f-7_ukh8^QGU+#o!b$7ZxI< z#L|ye-EMvN=wr?zB{A!aEAd#P1Ov%E*r*SSxwJ-J_M%Uff3gv$KygLvyeHW4K%oVS z+hA@h2odTI0sh{nDH)D1m=~MqHIIk?JW`-hx0>d#Zk%?i$n|@mfMSb`G3xw>yt#@& zmVGg`|e*~z6azb8A&YJ zP5zt$E&sGm|nJE);(0%vC~0Mu8u+#?$a z`}kp9Y(L~{OoV?qm0nnJu2KZ|6S$lj{M&V^-jdK(`R-2pNX6}A#jQUbfmJT?C7~Df z8WX@1Qxh=Dvjg+{r$D>V@~ay@EZ^y$Nz|<-MacAlsRK}NZBNMFI}POJHRra+-$S=K zr`m#Ay4wrJuFU^lX6R3vs}C=as|a$V*^-|y2956!Dl+{8Wq^=o(O3C zQzx1g%QVw9buwt|vO}vFsqii@`!q$rk)-~Bkpmz4?72jwk~A!UL=UDN%cjW0Cg5zv zQ>V5ctLLvs!=~Q|$jpY5&5HG^S!YH&vUdeQ+|V3^B3EEwf9>2KeQ%ED zkaxLqu@4Q5sSKZ~C6S~if_66j4ys}v_L{8*h3C&rZk*i)F6_E)1wb~r-=OsAQ$369 zAr?FJ2M8A|tQYRUQbOI#{uQ=(XIErQ+weMqw*LTmapmgmyD>p2_IvCviBkd_2RLgE zH#klVSJ)}L-(temnO{pQsk?Nl_I;DgGvHpocWm|P6$$H|fa8pspxbVyF|S~E{b@9~ z7&PW$6|Mo%NNiJu*r~&K*!l9llW}OE=u8z~6A;GyE!h(u|FU>hj(9rtaitAwAhnzT z=Wlt_91*u$-63V<^<@H|k1ZS`Jq{P3mTfe=F=}%+7WKAtBpc=ExAb_@L?>fRlNafv zn}eQyUgpFy*H*b%Q#v+P*t$_PbgHrzbksW3iJVE=V>Uot-3?|3F{TUk4<8Mc%<%RT zdFRQs1`;m)t7g#+*TI?{ATP2Og!qb5OF6G>)E|N4<9pDv_Vmkcy@98?7f)cw=L68AJv1cUcf z1x}_WSA^e2n?lW@Hk+()pIRNS6w0gbc~ zz^GL`opU$~dVD34yhzD1KzT2P-az$29BjXJN;z3(DBTtrr)O}J;4*=JVt=wV!M z^=uWQQ&R765sZS>KB)z?sgaMUJvf<5C~o?n+zX{eiRV3#gXuGK)TM;%K`A%2SiX^= zVzxJ(>3KrEqmoj2&T(ybeI*(6A;&)(gKB1j?^i|xQw{fwIS}BmuY5d~c9ShYcji}0 z)e;*76xT)UZdt56`7fkz{j3=OHeQ?I^@;#n7`_0yo$(-DRfpP z>?4@1;G`oWBa#`B89OP;dYALfot*LJxjqGE{hiuQ?AOJgN9``n5dPc^6!TH6CskZT ziNc=ybbO1#$%ln@IZ2GD)geDLa@@<%HNg3Q9zK@b_DUfy>M`bFZ(W$Vd{N4Z!BG^)*kew2AAHs ze?Vb}jN0&uPKcb!cI=hCHot%D$8SqEy1B)ac)#tO9{>vMeZ0%JKR<(g*^>?`yA1_> ze+NWC6GxZ&j<+$Y;*6TZ12*^Q(_?F?7ZavXptZ$1=Q=&kKBsR~S|a|P%Qsp1zG1#b z?z-1nLdJe>6^C0n#$(>D3bAB_J>gD`AgFt9@Vq5|(G8ZMg*aGf!MpL7n|jw{aPnEe z01RY397;A6rPRU1avDrOiNO^QzX{cCQp0_+s1~-t5;gDb_e1qvkNMjbtQDy8)2PTh zoI3VS*u&bxP6@dijt63xa<}+rDm=ZDg z*jab6M81K6^H}^WtmQQ2{5sri!6`9{vJFQ=7JNyVaKRK0F5n?Foo z>X2e8BIO2QB#d{a}Flv{@+a;wITMR)kO<8x6e z$phg?F#7UvXt~udb}vp{uf=v;#=SwxP#Km~?^SsK59~^) zJj)Q*)+-R*Wl=OvuJ}N0+N;O*dg^Y@wghoPJM~l#KM@{2m2x%*vlyUw&Qs=aWV zj*!r1a1whqT4Gd!KC!>t8()g76!XLqrP1!HFpLi#<7Si9Go=PHZ5#hd=Xqt3>qj`M z#i0*t{)Vk;L)|2ZkL7rleCOtZzu&oKRvWD+EMRX{CRYlg(%4F^^DXYn=C!D2n*^iM z>bGXLBo|N!3P}u?A51G9QBqJp}#SEr| zYM9Z#<^wuhOw@*PZCO%pnw}>=?H+2^2VKq6ur2^J#kVicN1XO&IQ+S@8(XA)M*J)O zqVsfr_IRYvW2&)VNz~6Jzi{~;v;8K}Zl-@){1x=;@ z_~fpTL%u&^-HR;*WY{ZH_YiKS*$7{X{vkUswctes%(@{!wL%afF!iI*M?}5mcWQEp z$mQA4*X2TjC8A>wAdg;CXSSb_a9=1qRA(vqXS*MAvq%|c#HIj&7QsL@k&nY|*RXa$ zeNzH=R|LlIj7~WE&HLMTBx5^M)G0;**Que7oN=&nR1Du@is1T!OzeO-N`zf_RBw;%GFFDo>QZa)wwpSgzbnrQ z#F?ZBXCok4HRn6SH#hPdlYb$H>d3ep1iA<${riEX6eE0B)$>|+b%9J5VfN-0 znyCD(f4~U_&ZH5ausEU+`^)~9-b2Hd2G9uc)oWwLXX@F(rj{{Sq`oL2u4sBf$OHS% z?^*R85i$!#3+M}9d?bVd6i+8j8bW8L{Xr{gVj-+HvS&wfGS7%Z=`}UXDi#4r{8%{0 zGWwO4?$5*im9Vt1hi(bs)>JXKrgJ8+U8MY0B+<3~dL}9{5I**WopL;wG-p zuunv3Yq)7UZZ1uc$U6*N#7cc($3;)if^k??Oouma8u^~va7dQ78nLrxFdYO|yl_Mz#HFZNRY- z_WpOa3d$|RlQbAjaqa-N=qTrJSMholyc=_E8h&+?t;fFF`G<{DcvRBrY>9S(-FA~r z&6AsFB~~bfd2gr{#_UhKM}Ev%vzK6Qs9N`1{%N0>(`_e>XOzcp$+*bt22RJ3LN8zL zS%L*XS!ZWeNpM4!TbPa6EfbPn60VTkzCZeN}#@RjAe%wZlgIXT~V z-vjnEBYE-ZCroy24U5r?Nwlli$*(8qQ36}>w(&iHz#R>rSvy;^r+F-a5!c_t+3{h8 zF||=AbiZm+K_SBuRfYA(;RFdAuXYw ziCs1<&>Uyt#Ss#!Z9zY6hWN(4p&@D%8d~2dL)F}@r5kQ%X#<=#cV_y;^JZ7Lp_iFg zptLnXG|ifF5N(if(sdGY%7n;^JVyNHi0>0{tKj#J3ru5)zn%!&)7_R%$KpTdStD(q zVs;F|?<~Dvy{tN&e3PCrvdsDBk9aF>$1ktunU|Ulr*WhukoQn&qab@|Ix9c{|L zxW4;HG;c=>3PA40%H)Lr2Y@-qY2v+ODp8p72 z?I*n9fJOL8xT#KaD9}~m)RR?p19zv6-!2mu6s4xf5BvX4I7OZU!vq<%*H`GXzun<^ zN=j-oSjT$4p@sqAzAbljTSP?bkC}xF-|kUOD=sSkOkwvFSR)H>l00)BHVCG{4h7@e z1uH@B6x9|`!sh*6j~$jhWovKIQA@Bt)DscnXPFwr9@KI8zQ(o=7M$UCY<82%(;G~D z>4KSEe>r@hG!W+?Pmw@dI21++nJVj<7j@3Yze&9SUQT&9$gt87Hw~3ZO#hNY>6f}y z{y=6hQ9s8&Fk?(m?(GJkTDEQfEgnws#E>HyVxMo+x4p+fU6Oe9vw{RWsJkeJ=bLxC z;I5B45#zfL6Mc7eFb7TzrxWoyb?fKOe#Hz|t@6=UVmLhk7%_dhR{t;Kc~z+xu_dw> zj@DJ!Ve=`+_{y0qMKdDID2>Kef^6hsa)<7*;qXWtIB}YT{&__v+gMjg#K6 zjENXRvyP5zACDK{<~FU(!-GA8h!`DC8v(xt{4fyD^qgmCKtTMuv1y&CQ1J1=7mR>X zxAEg`HYX|)rmwk!!LnP$(lS8C51i?6+Hqpz?RT_4DZywk;(6)*@8qTq=Z|+KXAtZ{ zM0e$V*rW!W*j5GZ_G#<{=dQ)#o0#m<26rzeLEb%{sL#dRPAlGd)JNA!EBv@|Dvh5f z$e*%X{Zn*+k|3yASYogIs5}Z~7hcKEg)6e0C7SahWEkbe zE*T7aCX199c_*Ho8E>+ha%~?5wbI__4n$8>nGnb=%6VFLKR|6-*2l%8o4GYL6fLdCs=x_|n7+ics!#onc8mNcT+ zY!U{@GhaKN0rI1jlGN zzC~sZ$i3|P8eY)l%Fj{pv+f2R3til~^yC{p=#=57k#I6GP6jpnKRWNJv5|Fw4}Mt0 zdJ6TvXPcAodkwUsSDxO{V0rK0X2$FI(_+i*;{W67tHYvPzi5G>Te?$1>5wi7K^j3o zx)hLZ7+M+$5tJ^GP+Gc&?(R+r=?1C$&N=6I@AKS$iJ9+vW5-%+@9h@EfsE%qLt;0l zt&bq>UWv!t%V(g$;w4)h?oQ51^x3kT+G8M2V2JP!hMc6b$*$~>l+;{PGTG%t^D&}@ zhC=>{zs|3nk3K;gJG21Lq9GIeoZr@Bth95C#kTHa0qRK&+_Wb>#<{+X@qf(YcY5~c zZklnBHA9D{M5N~Zio|s2lfh=y^P9FwtGr@!N<8q0@csgUFc|1Ja180{24&UCvbH8*bG7+bzl5CxJZw^NR}8k`G_*Zi!8XzqzPqoY2r+l@b09-=`U? zoP-#OIkL)gIa2Po`c9ny(tGf~qh_QAAISwETQQ!`H`^SY`>C2&b6#5s)yopO#WnVR2bG$N%h#$d2m*7cV&*|>j?*u1 zsI2JF75_HL^~PJG5j3R}p|y?&Lsz8a{Qgqb$E);!nFk2U{BnvF7=nxJx#SZ1&x#^| zB}z%iXlBpVx&Jz5zLj@2*A9xlR22D2GsYSv?K4LCXbAo{sTLBVV`Ob#wRuY!+$X%H z9B1F+TmPp4W2|C0@9b)Q@RR|wNTcF+3M-|4tEhfhscGL6eL-xxDMP_rVr{FtaU3cu zP>j8L2qaUCcnnOllWW>*9|&5n;Dj(BI6(!@5?wt^BfOW{hQ?@nqt*^N|ORDgADJXkRd*GKC+*bxB3 zxO`Y&_!21s850?_NP&5NgjZV{x=p?Y1!pzXM@P>A+ShV>vICkKseo!QrQLb>jP?y8u6 z8O#$y-%;gNG+BD$I(tSSaKJ>v09=hHF>q}4(UzV)VT12x6U=`}VJElY74+u!=EGaG zet&ULO-5^d_Yag(-U)}}V!)hMpO_7JdreZw>jm>z@iF*Nrc13u+@XF^P;W8L1&Z0= z9va+@B@uX0M>02bB@j)|0~B65;ZPM;xHwjUem;DPTR#ial;-QRQS_UKCH+rby2w)> ziBt?;4F74c94A!cALAS5kI2s6uvM8%=Nd@%S_yn|2gIIdg~rd)(?AqIme?XDq~8gz z=d2Co^L&@`gFJbAMoEC2+(A@DBhfaK(cq`4p2+PH5G}FhX5O*qo=>ao)UKzDCIudo z0{KvvQq(S=*4n^Vu(8{~YS8C)b1Cc`EPp@iTP}>-Cz>3%%Ng{Qq7miF^~O`!xlVoc zcNmtL3m>bqPMMeH+*jF#$@U%az=DR(y-YiQ3QLEAahxrI7^Odq?4jW!;qWrh8NoY} zyfw4@`6EwJp2o`p4o;3gEiC#BDve5&Y3Kvtj~}vbNb6f!hcbWYpE<_U%|Gl>_KdPE zzxnYXd30 zt>V=&c}vn*{-wfB5UoR3k))fGqM%*MSii@!%zxzgE%5nd4Dcpv)*#2~_4mJBr&WVD zrO^Y46etU6`Bx$Wu#M^iKQ0>3&(#$S{mKK<2qs@JD0ygO2>dF;@2Z_PPtEs_=1%pz zy;7o~`#F+JGDU_5;;Vpg) znt8;CnbzI(r_nA!<0i{20+pqXYxa%7zF|6@r zpOjNVw3LxMk+Af6TBT*n5FV--A_H{`f*BSU7jbr@A8r~C>x4GD{@6FnXB>tlx*Q5` zEv0*}6>pK=`*9NZXG&jT#F3qf$sR(!;v#C}p@DhuZn^r1Uxv_r8lYEFAZl%6cXUE- zNQLOF`UO?V2&BWsaK&xh50|118{Bh+byW4TSny=9Njdb0T3cIL{yrHU9jzwfGNQ`P z&UQC^<;gyJu^y_j@(q0WVGVG9Eb>(oXv>Z&=m`zcaT$DMhc zQnU z+HtkX8&!*<5+(qNUO<&+4wHd%M@NTlq2SU{1e%sGqQg@oKim~Ed-~%Q)f%~rxGNud zS=pJ#RdSI;<(J_lau;i&%j{{&t2b>SSo3e+m7+)v+jyUCYlyyVPg8&WQ|mdpjCzDt zGM{BkIy=zMRk#j0!IWoLcbCiMUZ?LGfl5XQ3`Vwy%?>>EmlM<#Km(+09Y2VCI7A zH1L`HDt#~L_>kXjZsJf_Sgvd3uEyuud0wzYr!=ER`!N&ic5PD_A;E`%nZ;e;ezL~AINQ}#=K4Yw0_#fBcqx+yjn4gkolF8 zmNxicQ82aM7w<)^{kwbZ@H;=hE^Nc2KZV@qF{e}nZ}3X5vt`0;?-mL)7%3?za??yz_OQK);MG4|Qu^RmpJTN}ngnN{|SBBC|tU~J+VXz6oSZ9x+u=J%L`0Z+7i zzQ&=3u%|Ne^iB8g?{fnY7+;T$j;ILJf=~BWR=!9hl**|t5nfXpVTH1qS_V=hbMj%e z1W`+?jvR?p1P%3)#|b5=`K{~r z2qI{&^<%Jbol-WxX1PsbI9BpL9CT6Wo#F`I-v<+2KEgYG$*E;T5VEWR_tAB<+P?qG zcJ|j2iQb)m7zid>Y1V@=2vxn5i9DBlPE|nPXHo5=mkVf{hu=Q|LSdRgKwzq5{OI)t zsZVV5;n#cw!bgujL{om8xBJ~MpL8s{5wWjdP1HV=E~lQ;WqK<1;?1M*d7n-+$^J_s zd!NU?vc5NuyDl?gt(@mW@TDzqiX3K9sYFTMC*~@q&mA#Yr*k4fF2aNlRUsvyWVoF| z8myG^{w?R*pH9`>--_+b1Dajnihlv)=jvYZ~O24&|)3Xa}wbc&J$F~lf z{y^y7h?+KYG+niF`D^6QSCX1aT_S541~rpUxeRy_;z)a`ID5&6l%d61d|AqrSqcn% z-)GvTI<)D&v#OUxy&*Xk*x&-&L_**dESWVUeV7PP>z?L!;{7)fUe4Q0?e&SiZq;G> z)wbo@0ZNISRnviXK;*p~1zgjR9%(*WW?9GK+OlD#)ua2Xb-cah7TWc$Xc=(xK2tNQ z2Y;(Q1*hs(e&hQ7=I|Kki00SdEoEq)=U&|&9abFaUe(k}wvT}ZF>fzC!~sj<-ZB6? z{9aBi{$;Mx+Ptai8PSlJ=O8J&t_e<|>a)6Kk9FRD^yN!p+vy+)JPqxPTwc$KT=@S8iWeSIbYHP!9U z#OOIX!pZxap}(2gLJ#+po+mF?R-R{0zx`Ca9&u>t+$#{2AxUbq{pY=lcis%H1ydsp z|JF?Y6Mf(SrKMn+Eq{91mp(i`NMnIy3{jNjz>~vAJHDP|7A+Gk!$BfGO=vY3V^OJYiT23*D*bHY^QT0Kf36wl z3I68%?1lIzd@QZK&6AU0l2d`dCQV5#nY91iDZDEKNzgh>Aegs3+VLbDVjjuzQYO}s zsukZx)P1jP^`+2DRYrp9c=8{azIU|7R&SuoS7NTLzqlW(yaI_i8fD%-XeCh7-$By{ zT5i;x5B+#u_?5VrcWf(rMKdkudJAc?#ZI4apGkq`5s#!5&Zf$9U7_^z-iR(VSwIrj}1V>+)laCs!*wmtWEXzB*&a zl?_Z#?D3B@(6Pj_NVW&Eaj;l-&iv41VMD?<{J0l19k7*3NX2J4R5=~+mF}Lv{7U@f zYd;x*7g7*AsaFgOLPH*+sJ{S?*7K}6;-OUeTeSZzIoZ&cHVm2X8VfJ+_-D+v!^}D( zXQA{Jj|a1Z0(%!|F-gb@kj-7M|tA3+s1oTw!Bg454;lNtM6quuKkuwBP;+Ns8k zH9UU*veGz|ArBx7g`gqP`fd~-)gIInZ%^6r9W++3efxHgGp$?C-)We199O|jIt$rC zyZ^h(rJu=skwhd6!Kq5FQ3aMHE^+=c96jkT+0`Gml^Gs~x0n;9cw!ex{k`$=+;SM1 zgfWDl`H1=s9ceOzx$Zbltg3m!j#d7S8M3ipJ~TVPHoPZ>=1jhl4&YuhYug-3gWF=V6^9sA`EQFbws=fP0i!MHr1^q;m|WVSg4zV z%%MqkW4wfNYMz}%KO6gqoi6M5wj0-EQ`w{on-m(hhRJd;R|Vm$N8g^pDbV3^LBh+{ zM0bOpCb1~3$tON%PI(5g>VFz2k52y!Z_a<&jdY}Zz4zStmC)iuQ^V;A(KX?qP$Z!h z9aAtqL%&H3_2&3XA-T^z;nAGSs_NYMM&#%imjdYYviCv0E|FYkN5>d!2Hhj#XIIyN z2PwDcC1{YyiK3OH0j&fd%0Mu)w@^o@Ut?b->i+X!OjASWzB5&fl~- zJ&e|m%)C3MGfyRTuO0+o^X0gwD~`7=hfSArPB`Y7mdJjzoeY~3I}3GuoQ+4)l<%XB z4nePtfCKJ!}!IrwqIP~RbIGgerd*gVXX{=cDb=)sM1UNK1 z8Hq({bql=fOYq-yjPr*tWbw3BVAm~qC6##rf?m8+*~w}c?cA+U#41AgDrCw3!50js z=F+Wt|0P;jB|98K(B()ZjZPm-n{8Eb;j1$h3%R(LL7AMl$_g7M8RuKP5HH(uSlNcs z6@A~y?J>}TFQio^h;`udnZ0%|utD>;HlXARH-Rm;O-n&014ywg@+t0ssrz(}I~rj{vD3!>>_;EugnkIr_d z)vi;56_7a!C9hz=70Gy!J&)y9uD1dcjok@diF|yJ(H$_A}-Q&0QMNM+@g$%i%^G-(jA%S2N=<9_3x*mLVB}%RZW4Z{X|LA~b|AlV^y-@g7Q@H($QvV$!P#i>n&) zW}zpU5r^nHN;;bG;8URne?gG0^)I`@r4Q~2<6a&*SZvrA%9RemoN4xLz=_aqaJO3k zthFa#hzDTz?^9fcy-Bvvvh3VRP?SjWLLc z_Wi*0B9ur{F26C)i0c4WbWjxIA|Qg)Nyi+I^QE^1|LgfKqNVm;M~k-EAug>rl?0Gy;s+mnWa$%SyOBgVCe6Ryc zn-BD@2*9Bg3EY}_S5Z2vokKeE9$CHPOm`pDy7aU-ViRs84{08mP9Q>;M=HV`8(prB zR_W~kjnh4*dN;&oK>2*GeRnnQ)BH-GS5vEL;Q8{vl@)HK~`nhUp zB`n77#!`#EYp>^AZY zgWXl!`AK%CvRVb<$dyLcw>Mocki6d^C}pB(x1dnzSwkOZ=dsL|q+FUVg@vH| zXbDCX1(;JpkJt@rM>ac5mr}tU$_B6jt4>K8v3m@Sc;wNMBQjreJU+upk_5HqI?mzX zlbH%MGLZzclA3>376FF>LB|lqz>{A6XF|=Z4J_xXp7^nVlx!R@yDzY(M}ZCDb&>Aq zL4WILiDAQ@RQT!w<_MSYBP!X^r}p47un5 z`SRtZMsG&jO~@tnAwd|h$T^RlopY#QZr?dJ#!>FZU8PRD^#15M+gQgkl zd|z>tV}w8PM>KoC7?G6K7LLIYp0uLg5vna^@PiNE-xbU3l2cZ><9UT-rnuqmM&uzm zU9-E`4)`#g&k~egbunuWjLo?{dVPspv7itFaeuz1rptoJ^PurmRm*b95OOHG*E)st7#3fuH7-z#bNn9 zZ-hK9K;LftppP%2DZFgtRC6xG-%gddWy{pr&fsq^2=C654$My<}Mf95w%~sUColO2E+@#ilq^E1dX(FjGCbCly91S>WH10m6HMhI5YprF2d{n zdaJm{1yn9&`#;=IK{=nC%_Q4r;F93}^ocj1ave+;o%Z9B%R(f>WDw`!pVf)^s;1{%#@jGZ>+AUTqR2Q{+;3J{GwwN zp_(p|4bW48IjL7y3l&H8S6wn*7!hLEkfnC7=NmkWj5bUvreOkRI>v*&CM#9b0X54P z?uw8c`G(=I0*l|%q(L%4AIDPmsfRGBz*mj@FJls?=(_>O)p@;Y;~3@dEs@?|9xS>`_1WzyRR1orI|>y@_q~`E*a+$oPSQ<`pG_C_&<+zpTeYALF^J5-j$?49 ziQG=xGZoL>7!yX5(NIYjnGAjYRyCYoW;nfxFHkNv)o9a)t`cAULOH9~eDjXVxX``y*pi6&u05W<8s*E^F4n7N6dpe7 zVEub#;zylff%}a@&_1oG%HaJ`#p9Ztcg^dAj8Nl}7t?gy3O>}XDoBt?&A{iOCGAe* z0tdI$Nt4U`)u$@M?XKcaUD4*Ed2n1TPwOewb6N29`b4fwPCZSQeDf@8BWO1%;;^A4 zNLZ5^ZJ8r8TL2d7cs6Yilp*e|rxAvVTk&qNE&LIqtq~1H$*53te0dZNU2u=lc5eHK zp#SZvY{oMi1JZmimn$?Uy zM}5onCZ!MXwE}9aSWWEKWnCj<(&0rx<5*$1JVcFHzI-msO+x`sUmKvO#Za4Oyw1{k zQX`p8eRJ;Oh5yPqGTZk_NVe!f&n&kc2qaFJ`K1fqj900>F05a0lHk(fqOR(JY(V3} zAQaF8?gYg4@1VuLyysCX8jsZo9neiIx>LFwUp3mM@LPWX%AmUc9*Z9Q z-4|11XMNkuot}Tl>1*^I`u_57jXp!iCNM@bB&5HUI!2XhY6w8RM69|&$W_h%fZ0JK z(A4L<>4vWUi(Gw|IcroV4-E>z-yDnq!M;SkdbYHH9%8$+Kd2(t-Y~u54SjE+D+s+f zJ6NR5c@h1~oPl0;>Up81Y7|cq1xq&>aX8^SQbXjk?NuX|JYAle{VzHOWCEdG7(9zK zm=f$nZY51$uLIM+R3)D10JnN1%Aq%M-5Nri8;CUjOWeCLnqfMASB{36W)qeV43|J= zFQ|(O0R1)n1?bt00aZ9GkVffZiyvycJ6V{hVx>rSn&Rd2XSy`j@9yucffzhAJ7VnI zZ4eX9mqbet9oM%GpL5WwqVp z&(~AX=xoedKuqPIPrXdx=4NNd_Px2Naw^g(F|d{twOGv&Gs?i>^X zw&6-#O%s@s&j6Pne8iT9eky#a*#p9i5S^SYT$xH@u8d!>^&n}y)3Y=$X2;)Oq)Lro zOo>8H`4+tiGId~)@b827A2oT+#JFuEOcboPg+?x@m5_Lmy@>drGWt!1}98yMm`@ASKkJV%0`t zokd13wSY4HF$8H29`v~JdNWl(I?!RE0>C+qmrLFPEmp;PRb{R_uNsCg_UHHA>4W`E z7~bdA{yoP^*aC*|lyR@5xdc?(5gVxh8?a>YTMzh(p$__W&bjBC*&~d>J!C|lLOqtYz z32@&Jz@-jY>ujP1t}I+kVe+Tc?v#D6fYT_$)?!YXT^-=Y;P*g*wtIchA4973m6QT& znN7+Lh6Ltb=3(2faw=T}cApO0)IS#(d^fLX6~kRt=gHPM4hr6wYKI1&>VwaH{J}^J zf`%)5yw=I%v?g0**f7;x`lcdpvDs<;1;`;#_vflAzQy7*nA$i<0`~2OWQy%`=j&gBV;Gm55%dg2(%ZKgRJ?B=WOu0YCi1ctkoqhfdh*tZ2Yb!Gh=@y(`zey>SQw(KKy$REn;|!(?g7%|F?56m75DU$?EoUrXrdMBJT19M3~+xkD||T_sS)0-LaAtSUN)A&x!YrJl#& zK^F)2b@<8V)0wc7t-#tG1P-$|Yq5y^M*Q>lWevJ!xJhfwERP5dPL7Gx!|B2CkD(TI zUxDWa;ALYl2$@Yf!XKgI;K<>-q>uxmIOr>fzS+)LZg_Zj*={g?FCi@+yB@M~^4bYD zcN=|`1}6Np1q&s=hz_tRL4;l)Kh-+YjtqDKZc5(V|EWnHFl>u2fiKdq$R~ttm-{yL^7D(#th(8J&}*4=>*C8XtBN^+Zp$Rs+nQ} zYoPP*zS$$Y@(98xbb8ef6x##IoTad8JE`8!Qia;EWJ@HuNKbg^0PI<3BqDLLwBX*x zY74r-e@FGMgBsEO9OO|i4Pd)^HN(@KFwR*_0*<>YW6bqkqBLe*{_saU^( zn;+@}jBogZv?)uu@96yh&jxNUi?aEQG&?tYUCTNVejary%za zA$Gweerh38`rRvvHHp*@b$8P zxGQ)5Ou{C9l&FZi56sgcVg5f;lZDSR;)oieE%s#yqSu6*V71~--?@v(5eE6fl_gxS zuk&3*{XjWO8ssA>*D>lfM(Fe%{;Y3)JqGQF43sm($N?&j#=yv^QsupXT@-5yD3vSF zz~Q(ZPjUrBhc!2t^!Fi#M4?tWJuXLCgl_fjD07uOh6!s+s{gFam-etgmDGSQY=O^g zfKS^tJ+LD}4EoQXL4Jzl1?!;xS;R=Ypi{pT$l4=k@!mXgo2`~g*I$%4KSYe2M3erA zWXfc~D#%(}3S6LU2qy8-N)Xl(6h`YuX1N02i_dUZrXqglnpkLgX>YRg%H+_=oI;ekaCKHMyMV)d2Y$(uukdz_dN#1e;G{gcW6Uh0H zK_JDY2hg2;rr??8#oyQ~CnNMA_sw!L*O4ZsheK&S+^efni}g}>e-FgvXsYY5W1k{H z{yiW)dfzUQ8uAb*CDCx%oin7sj;VgQc8{SF{xGA<63X)U@yuly9&NRkCs}i`^?07u zbP4rOol<+3Ge^qH)n@4pd7(PjGuARc@uOZA*!CzaKND%$7mfYR+R1T8x3?(VkTJ9; zaA3J^BP6gQ?#)7q^iXV<_Uwp+WBwP?xCWy{#-e_6M|z|mXr6~g<}u)7u!sjXNzv7n zAAnqZ|9D@^`*Z*vkb&`Mk0fr7WTGeqap7bbm_N5?Q?f2_;IaZkydkGQp(JwZiuItM z1uH|wgG44qR{NU-B*OqXH~3}+-(icG6_YGQ;w}1q9}h}FB>y`U&x@A~VWaFH;f9SM zj>Lc0^~ys25Rt*7?vuUTPoxlK2X*+J$g=#$i2yqB#t+2LWD-6?pl92&2IUgRdC=&{ z^pcpNxs{UEJzg`p7Qx-+(tSyx8K`ypFiLRedN1-IBssoWs0K>#0^>2aXWfHvwZ5 z&#F~`Z{M&l!2f=tr>oS~XA4T4L9f zXEyJqiP?2WurowgOBxQ^V1W>jQ)fLSfW9j*do~b=-r>@pV1?m9B=5IR4!(EMj~ga) zBZ70R58tYAWsMezk*dP}QO%!3>Qb08F*!#jzLZ98Q|D5_ItEjpbL5vvv=`jbC)MA8<)jPeG$$e~ZZh1B;@A~Tua z*S?>8>bg6HL&%~ktEHt#^Y!8Lf(LpWsp;OsumpsJ{Yd*e;^x8XeUZ#R>m-axiW*6D zt+TK?5x5uaO!ChqlJyH)oyH7CKeacPp2#4 z)n?E;eZ_0I3?gnmaHf4zv^}8gGv*|-EUIagO;_VP=kX_+0TG0O4lO8oDlQ-JwjPqh zq&GM)lQj4`(?>?sgM*Vq^w3Q^_jktA(eRqQ#xPkjE)Gz>WdX1K%tQzWjOa$Mbv5u%%?!Unf`_n&C`UZI0A$tN@Zyq>#eX z_6xRqh+IZ^@lZU=cXWKCL9bKb=LdmXsTf8Qha&h6nhSe!AHg87Xu*27ftT6bvR-(b zvv4?{$7YJXNWW%CKb04An)w8hoyVq;{XsvErI(KuPmdZcmGhsV;TAG)w%K*FF9;GJ zlUd&B&B#6Zf5V9YFx@O!Kx#ZcKyfjk*2?68%H%j6mZAY0jn11^AKBM06I1_He&E%o zfPnCD_EPq_mX|-~=%#1d^#GI(Du{0@q4PauR&rpTHCO=)`(>pW%26L1Y^Hou!(g3# zl-xaBnAttG+veEJX}eW3PnrA{pb%uj&fjt7FTe=L)aPdoNxW|( z4V@_yLAZi5<7h{a87!Gyt650m*oq<*gu2wnk56-Of=0x)Sq!%rvU#SL@gX!~XF#(w z#W8yqFRym-ZJYysOwr_0eDhQ-S2UuW1Ux9dMGHxbBu-zIt8?ab*xs9c9v=DXfvX{|2SirDF8!FnDR3(G9xP+t z%hXu-;ew@Ji}~pj6v+`e>>t^&5o&_Z_cPOIt`eyM5WHrp5hGroH zZo7LZT~w18YG!U;Eci%r92Ru|SJGk`3%;K~2Se@>#k^$mRyb;{3Cp}r{aua>Tpx|( z$6hj%`}}2K$v9}CW!36wB8A^1f9qwM0?qDdeF1y+0K2-y%G+lIPk?UYU~$nh3Y^=zXIg$T>vkQ0ODW$a>ekdwz7nl2V56pL_SL- z`4~@1?DIc}HH)ZXS9I#XzSr@0MJ(Teo`7}k2Q+*p0R!94yIhbnYzKe z91QQq4oL?6l5haS)dhx}gaDBH8{mMfM!%sjt7X{Alyvf_PCa2&&kEo;LlyxNAIIz8 ziSuoTHuKfRH9CLegdUSE`*&gMlwLP_eul2|zehgXMQgm}b-Y}J%OrFR5EontL7_7& zCl*++o{#^+g+Qv5BADO$2LECotGg|PK9cuSv;s&O(dm)#t0-swHl41_HXyL}P>GQu z;vAxZ*fhazH<&uqAV+YxWI3>w=-uZ!3FmymU;?wIFVf*>Qcf%xJEdb7lJPXA3dqb7 zfs9KABTPzPO#mb)a5)4J@M*bOn+jcI~wYmu&wyOBH}d`<3s5)0+c6N;F6T>FS0&#G~E zv_kO20svQ^%~K(=Ng!Z5V?xuGn+NeBQ5`u>?+4S;v?=U=DLu1IZjyU<~=l5LgLJVLug)JcM+RMmEDkzY9o_mc9INJ*#mwd}l%LNA}*OxR3; z0qd_18&7fV!Jq*Lo6R^N7JYmYKEM#i!VQQ^Plb(M)a_2bEmgYrqU!lvBP$Z)5E~N# z&rAP3(~kX&AZs>4b)Q2)D?=#8$SPl!!slL5=+Ry(WG@phSO@v3J6h!Au>ht&oVn`( zw6Ah9l7TyUX-y3SvDM)HZfO}|_L~UVj6G8T=V)8a*1NGC)UM;+ENC8GPa0T6ikRclRqwv&?9!(%uDu20#r6ynIVemGA=#eLJLwEDEe=G?$`ea?GK^L9C7 zHLg3FG7Ly7TIWA|JwCa56d}e?0Z2`3AZ?}r)M+5&Zh)O6!l-vZ-xkoWKq>E;I0aD0 z?^B}Jc-ThjOLnqfHz1*_5I}y@@pV2xWNjfJz3%fsiqyLN-&#M7vm9qms-q$C;Yk^618Wz9Qbqc#AJ6#xQ2xY*#VFSLa>D(2Rz)+;{ z;6qVY97m?7vlB<)Oe*zhLS1r_d05bB!iwdN6CpXd^^FRV%az#3w+q0XbORY$apEwj z;`(%in?ZG8hc0*b4Zhhkncr7j<_0kqAG;V1YH;OwwqWRXX z=jln-_odMEz&q6KNz`J4j7A-=<7To`!E3g*vj?HTZnXmAj4wyRv*qS!7l+FZnS|#J zE1^+BQXkbsBM?A|Le9;947Sw^|bS`xA9~7c*+k1dG_m7r_JoNt?pmkpg{Gom-G`) zEj$yhqyGP{1~ZuY9?svyr}=<<>PBQNPA7yJlV0K?k|f{?*;`{>?yLoxr&QS14y z3~E!;)@v3?#x{Tmz7Q7v(!=%(4by(WZ3-?7XAe&caY5k^&)WLgB4#DGsd1Cxl0=Ws z*_^bl=@62w^+O?w`UdMolvSYQz0mB+X>MXb812Qh3 zZE%hQjdQ1PG?)k!|EQsDd)ul-U*V=Xmfld-WTm9nX*ah$v1rT6Lt*gOTct^B;lPY| zSepm6t2h#5+3m>V<}q*BpW-yrLnD*k&+#ZU1QaO&wY3{eVUvZXe^ZxIeH67PY5&8u zy<)l{ar$-#G(_kS%&$moiYYiJ!GsKajuI_$Sfru*sSW`;#wY;M@Z95p#zBnzaU3O@ z5jCwM#8Bi2peMIy<7xtFZn##Sba4CKUum9aSsgcjITuVrF z*6V!`C2If@masdGhClGD|4MF*;3B$9|9azTK!!M?2=*yJ7traE;(HWU05_&W5fN@u zV00LL#P+|3Wj&#q-;>Sw^i68@X5 zqszNz{rCi}wY6I#E1;q|Zwq&~@^dBOKlmuD?M!~>Sx;7XNtDnp)c>F_F^qCpIfv?o zDQ-sFq+t(Wc8ymJ{FRI>xf z8DOYOkrpeJQozD=#~j>^j4x_|D}$ZxCtpZA7e z^cE8`aB#2k?iRo8*Y!ZjwqJx4=p3SB5|5q1(S^8FlbG}p!$8ya001%XGB7TV2}=Ah zruE!^w$gZNFeLusE_XC#i=Fc}7^(r!YkR1k=ZUt8i7}-DKmtjT{|ppd@M0A&T?+JafYauw>!JN1)tsM(7uDi~fQheo zt3U(Sp{XYRRwfrCj?zS24$BC0+R3Sz$~=zs%`3%!h0xj39&yp3(WC{6}ja zVQIqlV_F+HKEjT0Sy;na)4@nKM6cS``f|3>i|_qJL3X1c!*rQ*E|y@O11)Sq)ozYT zzBBR|9aHh>0&dx4?Q^U&lJ1{4_H#&ON+4*ya6iPwCR^!9cm5p!=;fcU7w`3J>c5rQ z-OK!&1@o#l?tJN_=IbW|;;A-c?u1r-2jRlYgd?xRk@>JCaSGPoI7x?jKgm&A95|nz zE;{b8!I9~!<8@F84G(Hi$9&a)YCSoOmqunTD#WF6^#Y8xOkh}V3Z;N+w$a)e(fvPH z1xxG4+R0G@7v5vlVTIK1DbE7usshJy$y3O}qt68*l^OrNG2|>zJV!@zM-vJBa%zQ7 zGmOD4fWEx?omh$EC*;-VTVs2y9d4kacd`fvNkFb}dIWl~n}a}kNsk_h5X$^2>OziW z<1wC012S4e(yik0V{n|nlE$HY?yu9#LA@#C^wCch7=kEVfdZ|6PL5Ew0HSvuWZoYM zmjls^SXT$#zs!XN6ymgCD2=ye%WyV7OJq3I{3eS*^yF1XsO`tESc&M4(lU!3xaMbk z21@hQL*EkDVooC9MBX)I+^x#XOV?_f{xrppva^5wW9y-JC^YM1Q^8F1KkLGG%o!C|E^TqQa61M+jz6 z1GM&Z3Amt~3I8DaC#3ubS+()HE7{$#ZPjl<-7fr%6q47KgVQ=Xrn_&L1Lf9m4GS5MS+F592R!$Ir}Kp+ z^_aQjGX~)Zp%_UYi6MGb&SrSx(0^{I=!FF^7`@jn2s`Iw!oclW z7ip*_X8Mrii29`m$w_E@#p_zq=u1I_4-M9*qT~L1k#2u33z&gww=)mi3Fy`8M;#_F zk}hqVNZp>a-(a^7n`-Ru_?t~L%A81DJz-MVFsB$V(FlF{ef<83DU*>p=xhby)B```SBqfmnv5n~LQ{dmoiY*1*Jzfpd zqX@i*XIai&&A#mjVgPZA&CR2zgs1@0ua$99+O_w!E9hMh?(qe;y2sl5*CN1Z2}IfG z&`~VpGn|AyocaLbvH*stMF&5#*pTNNr|7kD(U__t{8x*_kO%vLp=6zmqq1u;Tp_K2 zp?;^$Z~I4U*nsl69hmCL}-To+OK zWe;U-yJ|k2GOS1W6;M8+LMt3jcf5_o**g4m$@vIJAnJhaK4LdO(SLHRCGOTDiv>^c zB!j2R&`IbIgkxU)cVzZc8epM#)_GDqE`lxw?1WyHEuV);WTbEi&PyZGXDyREWf|5? zH3TF90HD3t6JO5p4xu*1LJ(tuv*f-e7MO%1M9F+(XJ+((l$jjtN&5O_n35lKhz||x z2k%r|0j7HQ^%VfWXL#kOuD1$wk~;msjR-Oj!H@IJpzWX_CEb0b!6^Duar)n*+SMnFf~2!aqt zfz%4c9NuZ&Xd>io9^KPv@$*EPA)G%MHt>YeZfoI5Q#xb@X*Tn{`If-2s7mT=86}+8 zu2X~{#CRBZNAEJz zH*hlWF2DEYB26g+elU2b-{9OrS;hpUH<+elsy*>=w_gPco*Be_oSe9GowG3%%RmbLlT=P zm7rYIGdw~}O>yGJBUOONZ_cueM?S_xT`7zfKw&x9gI@_}F7J}v#e_KEW#3SIgGj(B z6vNz_droDP-R}Lyt2(4KTM2I;BfdsC@f1o58b~P9Jt$6_wnJpYL)i1Mpgw3@6S$4% zCe+Yq^VbTCMGDarFyI>D2bRD%z>b5~xU1T-*HORP0vi?X@HHN#U%Vzp7RNE}|IV}? zXnw150UUNWSE0WqLYq88ffA_c6@cYXSN{&34xzneo}8@w|CDchfgj;plI<1JLq~Yp z^`hF9(g|tOl#1b`%=EaSuQfuN{)Vgyg%pX0{Fn&2zzdoQj3i^$(QX7TQ{eZazVj`*K36)R>3H?{^yRH+oGzQ>NUvQ z0IHv^JYekW0TU-VCFPpazhSW=tYG$=7uL5&*uNKSuygX?8*`9(cfG;Hdq9Tq$d~@T z7n^odG^bPO=u=>ABg=^J<&nCfxmdX8AA80XLV zej?bhL-g5U`mvr~I_jzydIEu*54g0UbLskUfolbQIi5#Sjh3vu zaDp4No!Wz^-IdH*R#vPiuodM=MQQwOhBB=qP`*aFbpcgj?727$u^^Jk9b)=Kp_X!D znT03xq+H=^5e{etiXjNhP{CfzAyceQYZV@No7r9tsr1R>yUrJ;pexvZxTvrgzA}>= zeB+D44i+dG=Z-*;pZ8q2ub^Ay%up5jL~jDG1^<_Qqiy0pGaeZ;CUDa$Kp*quzc_3S zT*Lu46j~9VY8^*CL^B4IlU>9j3jnkTqyjcj?}~$+LpA{vOMN8NfgQ^HJd~m|qo)Nz zMjeqKrGXbCZnK!ybIANXG~X8TGWr8ute4YXm%izYuG_l4$M3M)Vk96w zf6jJe0bK?qrhC-BF_d)(F!&YJIcB}ymkjDojRxeKtMkphZzs%`-X)qjFS+NWZZ0eX zz?2SBj-J^6{c{T3rFn;AN{r@!)~S!96~qB%)!Mw-8f(I^H$Q+;7kstDV!b=^+Jj8T z0?=N0wI)~RBy&KqArj|%;x)HaNa@isDtOiv^8C%8!GQ}LpHPdOG1l_!dJk9Z9q!Jq zXi6tmY^H13u|gT2#`F8}5Uw7+@I{mcS1yXEy~6qONOKFH#LK))#OX*Fyw+d)CMr#( zxiV2vK8+xjWvn$21j-hmdEEXzU(5C1>i|G1>LlEm`%FkqP97B>KU9S~1ia~o;LDjq zA48!pl94E~Q8CYwXf+LwKmzn~HS+);_X6>83}-&R*}TmMot(JOz>`yGWwVhctNz}s zEc1|nMXC{t>}g+f+P~jtY+>-RSfqoao53qkaHjYUIvUc*1>_R(wq6u6Oa}miSh{qu z6v&iv>Q$3!TmmfL&dkpN-nK53V_sTbzUkm(_#0s|#^;ylLJoc|<7nkhcZDzjLXQ$J zfe$xnyg1ZmC}dRI?AT^=sK{;j!=ZnpaOVN~(BRwrd=zmrdohL*UZKcfAaGs*|D_vb zzr{HJXj%aV@*ePu5Jho<$><#S-12oQO(a43FD6(BM!555d3pI(@5>YA2d1qOD8U=o z#ULn005|J8`883_2ey0J`ud~l-6|G8Zb%gOnf0?!fhhb3Jiba%55s;;KP?`2k=5Gv zB|#8?uq6{K+(+r#L2bg*DA`(<&jt?;DJMusubNhgYQ!dMpwgB9_L9#|h z^e89A+#_E|C#mbxVr#YKx`}40(vUGs(KjW1DRD&^|3=oPWxe;B4Ea_nA)!t6F5U$n z6u()`^6^cc5X4gRCV-Ej08E^XU^Xd_m__l8Exy2=6bFjsTor~qf=KrE<}d%NE$wTN zE(7y9$Uo=)fZX4o{E+hZ2IYL;FWyQqqQQ)jz#hVD1S-U>xzC@d3OOV|=F!|M6D|2@ z3btlR6#nWK2Bx&G`x8k$ECTNz7KW<=+LSrgjJn=jKL6$S;HA2KvtE0zM3sLIr3lPgY%6s{ z>RO*HeEFGb=-SVgmJeXTskKPC5kL@HBFSB=jsW!oit-iX#&FJ6o~uK@EYpXIU(f69 z#yVR|wB@%O*7FbKNAf(kr)E1Sk+s?Q1!w7-=8y_mv!n~LtxT*t zP0*WrpG}?!%qP~yKQcyOV3RXEBdAuku2+f*KuB-j529*uH)z12d!#%sgD`(P*Wi80 z_e^7h7*=miQ=b>~ryt=6R2i8<36^+{h!z_XC=Xa^N1N@|L(6fGnkN~X?_>_J*OhI1 zzl+vk=SItKUzqj$`E@Syzg7^Rt}-X6Db~HpHgQ1OZ#}Bv$#tY7EwQ$YIlX<1Fj4$t zmmHL{lFW(dlP)+lC+r=~&i};d)*86&>J4qr>*XBkHV$)Yv<_Ua)Z z(~%)+c{y~c8HG)dimNJ&JnR&5l*;viH~x(vOD%h!1w^6Ft)f3!H(K|2HqZRead?_X zTc~|8>5yYwz)>&oUZ<18g___<*KT@c58=`sq?^P~ey+XsWtCw0hj-}J zo(I!{-qy--2X1?nxy`Z=nbbDOFC9ByM|4^Rn3rsj(6e&8-i} zBuBu9lik~c!#GcrfvUAOFQA&Dn?%AHu@46@&ExLEK_s|C| zeT-613KA917jMX+-An?+>IpBt&~WHY?fh}Ua0UJ?5ZyrO0t?N2s!4?K^#8@ij>Ac% zwuo4Z115L~kI@W)jGPs_^|!`saqk86g+LL2#%Zp9IeeT35&i5692B4)(MB-Mi+Q>y zlc>MYdnYq`_>fQXUGMN~Hnp-OBwLG@Zly^)%1;c>2U4LrCOnP}kkJ9KWi9o0&P%Nf zl=s!E`W0!vIRVXT<3^^#PJNvWfAp6?0J(tqL`lWQ?q~+kWk5ru&R!Qqy!#maCGb!I ziipuMy%ET3SHPMsnDs-48(}}@YLG!UwV<( zVz2Oi4y2{AnUSDGn%gap_l2VoqPTg6Q!kovtf`D7?#(o~mV?j)1AvL! z$BS1=^0e_*R;o*Q;vP*)4%a=A6}El+_Q7z>;=&z$lB?Fdz_W}YOk@CmsURYZOurp; zi@QFDU|DD<2vGC4XildU1|IBQ$CH@VbZsjlgRwtcIowx2h?M%aJB+?CZ-VwZB2f1t z+b-zd<}65S(-4%ww2s&CfSz@TGAgsU58KAH%k27_tz?OB!a_$>ZOoUsmuNV7!Q72A z-&?%V5meE3@Tm`}ZHG!0!)!$%*TZxXWKw1jcRh43fZ$8Lg2+A%lj{d^k|?*eo|WB( zrTSGh1Hi?lP|&{t4)FzqY(6`=X(?7WnIAL|pbdI$K`jh}lq6!{f4AqlCTDp(Q$--b z7z*OCpdPPY)qN(KT(=(vdh>JQ$_NQ%cM_?F!KH2onNRwFbe&yheSnG%Ad(|zXM)0- zaH6EloL^1o5b!fB7kUm}1#YnJ^ND8&b2aj5zbiSOi5CAog&mPj;g}nbb>&!9e(*V% z*>?F@21%vactxbCFx=r4J{=CdgRK)&=|wQ}ceOpD05|>h>s6BJYW!7&m#%zQ8H!c5 zT8+P6^yV9UE-I-cv;E^LTN3nYl|<*MjnZk~)$XUt)=q#aTV7oc+4b-$+_{&`9^C#% z=l``*|7ZTiy>oNN_lWL`+`4%JLktW^#%uRQ(G^zKI|+HafMqaiLv|TdqH=apevMl|qksPSyZa-tx&{|{hI95V4FZtp4h~~X#4rMUM z5#*uMR2pa_${4K1{o&9r%(iAZU7P0My?+0q5(<1mpX$XUqobp%r^6Q!pfx(3Dq%2@ ziV;)iXtKg*f0b>k8BZr;d7OSD-s`on=dg$2!8k)TOR{ZtHbVPbFIR$@3)roXwSA;~ z$YyG~BO(V0u1M-lL7>4QC{^z5VO-%UGWt^UH+c!T?NAVji__eJ=Z(GfAlXa%P-Z~CTGZWRE|`m)!@mIzBhP z3_g;txZPV1=(cA2963XhM)}}xoZ(4RmG|C_XW)rKqmzo%$ZD2(;EtDAOovzGPpdfO zGD&5Xe{BD-O?ar#x0gp=bIS|wa+H66S&@IwoBQ>x_=DLeGrGt6_X5i6#W!R3{i-Td zo&8^0|NTQIajcHV(@YZ@DgfR#WnEGg>7xgD&uhF|-CD_t1g<595S#CrC zfFbBHUa*{+zMxT0iAMY051G1O{~OTPv54$jz_9u9ZER5cP4oqITeJlIuO zqE{^VEd33iBIF3WsS%G=V@6`6K6em!^p^>~@acr(3roR}#!s5f`yTmWXk-@q>< zI{~~PN+(+deT;IoxBfjdM1gQ;9)n1vwnv6aYSuiv(IfLW$6It*IMJ2i6)I*0+birb zLVU#q{=3`w@V7Xs4d`;QnUoB;3cK#2z61~qg69|nsmETELcq{OTb@<#gxPw563(5A ziC%mM!b{5@w6@#5#n|f(;IMK9ZO7cG3TO4VICoqiwE!4J-p&GkndyF)HA2Ds%~T9k zOZ>m{e3l5)j$c5ybkBRo^)KX*H;B@gqsKL506rMU-Pmw()j*brZDTcm)5{;k;Q2^BwWkXETFA@ z*cQ=8u=PRk<(mQjWICd^3j7?xQ-(}O60Pqs2mLcC7V2GBa^NciqF$;nW77xJ@1wuva4aCh1@BPjmJv z8I4!%0?_j8^ZsApb8tMt*if~=!eY8c-)0d*I4!D&s{*-UzNf?H(Eosm6bKK|)~us-E8x{?32zmlv);q+KHkFyiwqo zyX#dBPMx#A`nEaN9=m+956>NzS|yua{jxaUpZm-;SML-n=J`8YnH2>ftxG0|ljb~U z2^m@owe$~J^)zrJ}Xw%?hwz+SU#~UgssP58VHxTg&*fy96XFe0oXmSfN z@L{ugobvs8-}f~DZ0;3iMrR=&%dYqW&I4{p|H2v}LD%Y`cAD?4d&t>+APM)Od zqF>p3b#)r-)>)o>d>06Ck^u9Nd z(PaSqF#(vEW|gr}*8eqLzRUK%N%JaF7|$aaj4uVGkYEb;uOq zYTj}A=7}}ej0tfe!on5Z&#~|fP3l}EkptzgH}*AKi--Q6Vc^m*3W z8#z*1ses1#H|+u*8|sEM_$a-du)-<8o)SCbjF}Yfi@KtvOD}bQv1NE(groi8b;A?X zG{7vApXT=cB)!np{aTUUa#4Dw&+;{)eGQI2BsVr9_9K>QSV}tDPBg;=96KC`BwFC@iVhC}NB-DQUCj+ch)!NVQu0 z7I9DyBnqw2KqxIM(!4AqODR_(=#DR}9at3K4b|4hw?4HG0iKc;R~+}s$+kiySeXMo+kcza(Xq{6sGTq#2WXG zcv|_%JreJLp_70EK;!E5+E^9^9Y&AMI2tdHH;VwwBAmwQkv49(5Bh#2=+0f?8RuH) z@h9oAp1@pt0rHc}WV-h`4C*+*217~^KR!;>`c~mVgZAhvOb(g6K;hBC2+Gt5GSS#d zjnM#ov;6IQ3T85imI=j8o|s~ki6ZKh5a;ZWRRb=`?`om$C8b@X&ay$f4}g75mg^E7Gepd@nX13tYBS&w5d3dV z^cr#=VV{KgqKpR@!Z94Jod-@QnZ9fBA_t&)>L+G^&%rlOl@Vf%(l!Bzh8h4T&D=n+ z^>yeJXev1F3hzDZ>gkzVkdeMVz>I-#hAA-yLV(xgpMkDPa~7?Y**R~Eoeh;?bb?BUVBubFMPG%*2QOMV}4DDwqf z``lGwV$mK2TODvmn|(d*CWp!qBR?Ahs|d_>8EvY<*#BFV)5*O*z)@PR*EIstB6=R% znKjmfNn>te`Lw$5Q% zlV=tktwoAf0C5Tt|ITci&nC=WFB&?X+3`m)p+n^bLf>L97HBQ}KnPezG_?t-<|~gH zi5AzOpi6gtkg5B@a&74iQ~bDEM57d29tY8OkBQ+qnYyW)qC zx6D!JrKcO@SIBJ3_a>`61R!9TAXqJX7X3P$oOZ@V5`@zK>zV=sQM$%6lY%sI?2kI} zg0D$bF7ICc&=|yY1uyO@7}aXG1q5IH+!O(7klfmCcDZjq&-;12+PzI5?Ew=kO;QyDYrXRik%4k3((tNWosMdFAs`Erpz*5nH#pN_(`$Q;r^5; ztA$neUuMchxNGe_ojP4vS*ho?S*gu^p?w9_JMeQY`vaQFbSXu_EFFPQl}g&9Ze2Z4 z5l^*mwsKSwN2cmc~)-lz30sa z>kkmpwriFSl~CvK1C!B=wKMR;^*sdR6|m=qMk?e!ygz(`0M`DWgK5CGcM<6uPkqM%NHQRkKg2n01eWjG$e~65y0!2*2sRhOYq%;@T7@WR!kP@(Yw)5 z)oj@D=i{Bk>c4Hf_46pL-`uEc25r+(!1(Ko;Bd>_1xKSTQXhXeW)PX>suTQHz%ogD z@`DEvgZQ4%n)hPxtSzw`d1%4BJ>&t15fcpnFvvlstI@a3*On|iLc$cx-W~NRP7m@N z>WigT_y|kFJ+MCShvIf?i+NTNo6jddKhLb>^*At-Ei*M8k4u&x^4UrMmAD61Udnv! z_N}aGLk47tEVyF!BrrIn-7u>uKszyQF3=2|o;4bbo1ymzk9PGwhd) zrTiKD(KoCYkIlS)_a$Sb3w9sFa%y%;MOSG=+LmrY=iti_12WViROR+V+2MJ=1AG(nVwI4ZxyF8CZEG}C zC9bw%Y2X(gDwKV)0gnj$8uL7=v#+9QNH=?cj_)eu^aah_+O#De+oCFlW0tSvg^_EJ zph$-l_Pi4`7{J$MqbdIiMir?N3u6N1^-uet2ditL_>>DXALX!@YX;E%q+op{8SwS- z>%0;-GEDuzMNd}6VhWmmah16uB~q{#3EQjx+-PLCM&TbwkhF$h(%aTc^|bK;neKSY z9k~kz0tvO$#{TE6Ht*{jVP0pUpB+Mfd~IP+Y>iCeBMSZtn7mo*RXMj!(0SE_BY|pe zrfMc5G>rUs5fupt;NiMO3Hxg84ZQ^2+H6>1sHw$8u6bUgBnt*{ytBeowRhzJw%1O& z{TVW3?sIsJ@iKWRVYi4>c=qUqm6u5-(Xi9SPEBo}l%Z9~f*<$A!z(#k>nU;FD48&0 z-IppvoY1F0PwoBb|US)Fp9NDXc_go{BWG`z1?FVvQrtYx}6K~VB`wWOoM(NE{lAM4JATJQ7`Bv&+Oi5jj6pYPBXHnOL$LS_O(=H65ON%Bv zUK;g}!~m{02}_*MHqWwTX$p5Es7=}dtRIgOtGdpqaG&X9;uVr9?B@U-w>b-?DqEOt zjQXOcnxcLlE_CDdY*UnlN6LMf?;k<8DG@Ltsc4`mOvZd(w5hjl{qlREh(P?iknf+^ zjxMMS0!GW<{N|I_d#Br&hk7hE$!muoAz5Y+{aziGHe70mpqoiS*KIxOcxq++!9baQ z`}-&D@}%e^QiMd{c^#kuwO;%22g%ncMp$&x5x0VIt8tB3-v)g)r5 zXcXw?;;KMYmIrA<4byjyJtMF`Ot{YLDUAQs!Ln=s$cIbkGY^Yr)4cHJZ8>>*H#3WQ z%)ldj;*%|SN2lha4i5nVYvGH&PWIL7OEife!MJ5_@6BH{AyI^V5%GPN_%z;qATNq? zye}$`;hO=>-k8X`v5n{qUf4z`T#|8WX>hZ=mtdadBkH=00FIYoTUKDcjbD% zDS>mf->V>KN-pXlJXb<#T2u?_cL=G(#d2WQj4HF}B?0MHq1=Wc8G^>5?ev=}`9EiS z88Njav9FI4CH#(9vKI>bxtFUF&`fZXrBZa-nAmG=V) zR>DMRxS0J^O=q>ESN{TgX3?;rG2&xap%*mlN0170Kb*`cN@gyz3WkZ@?z7$YFzn%) z+=LxY+K)Zyy+;59JYO6L#)P=JD^opIGQ%+R8+Toe1){rzc>&B`3y>-1MStE!skIE& z)pv!UW4eMu2QBpmH{m&pm5KCw}AA_pGI{LG_eg~&j6E) z;Z?)K+xZSDkANl0c<3Me*#<;fyRkQVi{FzOkLusF!|&8-0c?@(@H7Y{ZidkFv)2LUN1M<)rc|20tX7__6UA zZSF4(9S^qk?{BswQDjfQbA#{C==m52`2H4{cLuRbh`*UQ#50LrYdp~s86j_}6mx*G zXS(Tn;(Hr#-fvBmCVJj#eOLgC-h`*>7A^p+5ar0chv)iAl8g}@2m}dIu{&e3h{65r zTyoWn!XJK7ZXw`jtGF+!lufaBd^*h|o!Hklm@P^P&et~L0o_tf215XoI83_@zNJOA zs6ZJpK({`&83XA0yQrRd>`}M9owCw z#h|OMAx?P!@1s+F>;9K6MuJF1EqN1?>6nYG1S$KtGY8xvMS3K1Tfe12yxI2H@kW-5 z1(;#`@5IQd(_}vCDE(L5qV%-ZRiZEnUSP*OI|eB2H9j`fU{H|f5xcJFjB+f9G1P=) zlO~C1-Gozm4+bN%B~W}yCrCN0b}j4ngy=Sso)6)j)Nq=%Hs?TqLU=_iVFWh&n9v4M z$_eK8zv|-$4hy>Pl*zg94n3^Dy%9asPnLA0KoXVn``0%x)6&4guL3WN2S}$rQjt-t zIzgCmm+2+se~-F_4`ja*E9k|R z>lL{?|2vwiZ{e4ol=JZH$| zPGAsoA`qHD7UEd8poo<1JbPy<|Q+)UCSkGQ`qpM7>y7MteMmmuzPUF~T1dCn7 zvegi&MFkka)VES2+}k!hzHun!VtJ{SqWg3#4jIUJsGW|Vj_e&?Tjtjg5}5@VrPUy| zRD5`opAAts_J&hur`p3bNIKZT#|IX3hx=tEcO~HnRWN1*oeAYlDquwnwRGvKkLjyh zz#aw>a56}294r8PUf%i*B~ zl?Bw;BhmjZ2AmINk%D&EyvG@Ar^ zO(WOr?oP4vaJe&ml!pst(;P7dT%>*P^}{S=>0-yA#<$YK;kEm1{6CTgW*D$P)U2rpmcRV6MT% zWz(OdjPIeI&M}%>+!3g{$?g?-_T8e_2^__04RviOm%7ofb&U&~#1Pp%{SO6TFqHO&>_AYs>j*_-z@1xd29 z@Fle;RzR6})@r#(7bGH|Q+)sEqJmnHGKRCiW2t9bo=J`Y2L}VN*754wY4$$r&hZn+m8DHlX|T!g04SyO(940+00oonNMK+yFUn`X zFq;nqHPetK{v8HsC2yNQ$l5+5NI!eRRdC zocqx!rH`d1HXGvWqvkP4)z{#q3-;#aa9aKxj)|kfX5{+^w{z^A!qbyIlIg&MNAGqZ z@62OmZLAVB`jh{&nRb3YRLP(~h)$ZC!$gAFf2A=d1)j~JO2D2PV987SZrz20j!`1L z6+a%tj4_!PMB|*Mr)9$&$D-gusjOn_d-CH$6X-#^!LSD|gq;+9~?17>O!IX60u2p|!wdWVq{-#R) z0aDwW|GfN$QM3=;ID27Y4ih++($23M#t-)M(SB1vja@Xu$4Dgkh^BrUP?ik^O&kH* zkb+T}CvYe$Bxnm;jrX1(uAB76NWqJ5z^K~NwkiP!>C};^;AOwh2mTq`5i2vWeyziZ zMS3u+RACTF$~VTZe7ndkkm9UkWe+)Mh+y0{F5-8h3+STDTp9MZPxtt5LA1i+gX5!ncqnTgH-=biUqNfyd6v#^@zgqy7 zG5yl>8K$X-52GBR9AW6D1==VdzPP(BmcM-ENjV9;;LP-j+z5GaWEBZjEPpXKVF_Zy zrXhIfn1-(i^MfAw{8CgLi%0Akjr37QuSm*O7JyWWS&zUWq3LC{9?nOj$M7)XwlvWp zPUR=+R`a)@wV&TFLF-;p0b3Cw+G4+4I=7taa-AGpaFiVy9Q)xj(L=mAGy=|O4}N(6 ziY-hsJp9xr{|QYPjoz+m%)h(t+bc{BQu_~sD8eq8T>b7{e+aA!Z$_FiYSC7g&0SDg zu!e@3S;%hT&)@Ux!DXxKHl{Z;#+5Fn23TLuOT$v?8AeT-EVcH>wat54zY@cr=o$n)TRcl5Z& z^t&g>o<>}`-O79%hK-H*@@j7J0s{|Xp}4_ z2Wac1_gb%<4LM*{izN*>$TyeEa7p>{Cw9wA>sZp=Y%vhX9U6*Zc5Usd0;NIJzEVZ3 z8%YA(6S0-0RX2*fUiP-EFyaZWQA>o$JqQ!WYi%?OJb_NWm=6qsiUi27M>)H`R zEhz!7L}jzc>k=6=WpM&yi1P)*_1|VjyyM^6Xhmjy4nANx)$DQ4vBpZ$)gi2rw36!f zegvzDt?3QJjV8fM_!W5Yk8EaJ=0CI1$L{{Ps$`Ni0rmk>E+M~zDhu-nl^625`Eg1r zBBLWxR}W{80Ew>uLRt7}G?yys+r9vZ23UN#974;#Z;D25>@ro%e|3egKGHX>TRNxq zdIiyE%GZR4taX~}f}K^Q^HdjyOmEvvmUC5Nfn}FJ?<(VKwww5KkA6y{M1Xy6mMofa zm+k_55=7jo`ZT(TZ5U^t+)AV+@BIjytyG>J&{_L`+;Y!VdSRkj5%2T z%2bT4#P)SOrN^(jyTea8!dw?)Q*O)rMGuGB6fymD1=0w*y{9@(?wCtaXr5Nz@#dk?tN^qZ zyM*Q!>4@TYzGa+!2LWzbq)@_58qSx8;+#nb4U{)Q1PrJ#kIE)JWeZ?$sJI|=N%)1N zY4Bem@J(B%N$^0W6)A^bJheOe{Pv`?Fp^l(wvka$%x|hn!@7dWgfu8`gd<07$)WSO zf6ewnB)6eUJ7lH$@DGq{B2A>1T-Fyj6$Li>R+fsrAtQGc#bTUJ9u#_0B*vcSXsxgO zO`{=@lz@KzD|b!ECGL8%~2o7thxV2sDdx-FRlBs-F6R#b#wkn&+Ay z$DiGNQ8(K1`u47{Yz9^1lNDLJFqEdy$PGibBuhrK*>Z9vWNkudJjh_p1Tem(W-rY0 z+x9k3*XU!qnB|8JfO0!8xamJ|thE();2$pFdx;B@#jD#QFiHG27Tj>v+h!sxEHPef zESPh?AhDaoD+wkdqU-7p4VLCQqRtKxb6>|ZXmEC-6!j>Kq?CV{t!J6;=)5MCt+%Q` z9y`)xxXufUcH3{^JBRz{z20TrP62MKVaaJjCIw;HQhZ)))y$Qf+*(DM_63n_WmH;s z;K-kJ4+HhNk@cHLCMEAXry=WC!0y1lp|7RWp3jbtjq7*o%9x5G2>mPs9%YdRA&wo& zCMefFZHj)#o0Q4yl_Mo3kjgeKDiA^@JmJgdOVOiI{?h{A=71c0#lMwG%(!%aDrSSg zUkkdX@6HAf{r;?nZ1pcygGQPFixghQ1kxCqR#Dm3`zNC-pcf%R_VSyO$bTfy+mdyQ zHqn%D=JBOBQVNZEym|82qEDDE@A#`7S%;B?4bUJAe<3HGoEjb;tx zg}bIJP0)R>RyDsEc_SCGK?ST&WV(7~@7`OaaT`wQrW!pp-xSkzHM-AZl-5{?S|Xru zhs@=1f3GlFb!14Gbtpx=1~RaQP46QXHIA1SuFVgU!RG!J&4ehY_G{b_+Ua!Y_q#Gi zXz`;FeHb%uCH*FS5BbLNyP*aH-aA{J*sKw`Vh}%oA_9{XNTd!ekPqdzTxjYD&#`U= z>(AU1qTlGFGF|WfZ9mTiP*-pVX6QhY^z& zdK!MN+N>7L!G6Ck&Q5pt2?-y=lC`SasXK>6OzQz5xE;}{5}c>^n&O1#(04>T>M&Xk zf_1?l;min8zi8k%;d@29V(U9dG@m-oe@t`XZ3jm=!nqa>yf5mnQm@PQ&kv1Uhm$r1 z`Ad3cH!hOa4Mc028So+nrqxpxWNAng~D*X2fe&sbTm zTOg_F6H!i_5sWdsgR#Bax#OaxAw2=g*TjyIIuynfrkHZPj4?`s_+7>O-uIT2@#$hIKUA>yT{>d+PSu1SI&KW~ZktSaD6e<#xtxx=rbW zl@@4^R(Hp{HChGU`9C-oVNOK$vm!RkS6DOmJy$(B%e9=3RM!K2X6NhBbNvV|E-dr4 z81v8ww&;kyzN53|FN`v<~1Gq@*`^hsc;=4^F*3>H7nlX0+W}B$FQG3SFEIwX~|$5o=-B%rnXeph!KK3-+6NgynX+I#%N zf-M0)IEC!nbeji)1}>7rh4JvHE7#yfZ%7C1g(~kusQ=5tT(Z?Rc%(Qc^@Nb{DY|-2 z7hg2btM|Tj>N(5J0Y?NKT6deLGs51oa~XfUsk6&x9zt}zxR5ID{Cg=Hl)IEN6(c^H zFV1ePe1(b_z9PCYiQ;53N%blWp+*7dv7X%enmTKT3uSpH#_UJxJsCoE-CPU(t{$ zqg_7`M}9(!OvP|m!WF(AIY}Qht8w(m|IH?oF;y}xJ0LEpbrUjzi92wTx^Yi;{UO-P zlNEZYS!QgwK=fxozax^eoZC>07wUI0>OM%F0sCM}%JP16Dw^$Ay$qa^Bzw{VW@C^M zQx=h;88d*i z>HQLOiHd<~TwImB0Y(*DbK#o=#I~aLzy5X9IbqbBdG0P+n##^zJg%70x7K^}IL@Wiar@%FJJ%+DaP0t)=ABcn!&&HRN&vXr!XyR$ zFekCAl)+8;!x-(+Wsi(h%Mv&A&DV{*5d=`e5$yBgB&28pP!!DUuYXwv{p+c(5x-#<5Z7BtG8=WJhmE zqP0ZxTq>!%YSD}_ZA9Se%T1>Xvz96qCKvgGm1ntM)HMP}hNy_R{6YSnAV8V^%ip_h z=b!jAD>CnU0&&5610%+}avA@wW3HGRzpY)RF8|DV=Eu+V=tXRLeHTB?*ih!rA)oit z$F7w2FuMn@dJ@r3|71XM=cy&tZcws%a)unUKOst0o(@k0VI1~xXLM1@^5)9olJMdo2Pd=-n49jec` z&IW`bB=@O>W+;~a~DA?a}88?#sTH650>nXslx zfd0s}BuXJPBskmxeT4UT5RiuBnh1A_+%M%|A@R$v2^d)0zV-OiMjpfF5%w5+N#$^q63JA&2d@fKc;tY4QHl2FM@$a>&n$V9m+3JSsJ!VJVKRN zty8MGXn-4xd3$FYvWHw-xipvb0IY!lDcj=ab&G_QpK3L<5QdZ@8KDfm_1(;|m29MO zPB?&ZemhuP7Fzk+F>-8lKws;0_>;)uYRIAYSZyq6Gs=nf81vBk&+?ainm&gO%AyN> z6BoMfJYXTMuk5}`T0)+pmQ<^3Moz;wGcB#1)xb_b=C63H$-N6_FdwpX3EFhnShV=Y zliF31d1eBs61ti-L60k|*iqD8x+LGe+DBdlRPMs5Wv+U(WAQ1!y-eUL%TS+XbVE+T zMO)`kQs`Z%FjVyK!B1~H^sZqXg;vK3>bU6 z(3S!)1O9MUx#Z0!x-RtN^W2x(pQW^s*_+zDpAJty+_hhR1+YbbhlsWnt2MxY+ie?b zDv!N}*;g=V?zyc-Q9k$8STFc4AkXBsntVxqJ;u1!_$I|U+itDP!6cF;(4azqTRDFY z(^V|&99D_T9;P^q#CO*8(ZO1lAd#ej_3!esDAY%0Kl49*E;8G5HjX>5jNoQ?)W?0i zmXy9S3e(r(k5oD5v~ZP68d0{{z|$)gxSj`*0&)NLD*)=dLGc}K{a5<5r^>okpCUuU`&u@YKJ1RQw!#J(Sxev1A>?;Y=Wo`Os_OmzHr**84bmyyEg^_BNOvR6p&RK2QRxx{l$^he*UBKzOM0))fbyw&pOlU1K_`{Yp?QUB#H6YUR{qHRP0~z+wFVol% zk9VTw>MG=hgFUd6*hL#ER_KXT+^p4>dIW_N*u`dJ(zy3)Q|FKc@>A}_)FYJnnQ?Yn zy!V!Aq$$FaihjJOls#GqpC~gx`1Jvjq)`wZZNJ!(`8h}sRqU=u=bI}23u>lI)`D4z z?)8bJa1$rS9a=Vw*?}Kj4``ICs+TQSD@2UaJ{LWb2pFz`#y*1sYOUUaCz@<`SqG5R z;1HwSGw#wgVzps@**?16aW44Q9U`P8hzrItJr`1Wym%QPTO8yTGD=mJYM+&-K$yR= zd)x!b(Wv5WZFB!n9EBi;-rM!CvH(}%CIHg0A9;r+ET$#FyozDd_XeGmr0Bi>siwOA zU3Ij0DannMCgD>Dbh@`uA7yM3E|uBUr}X zG649d$M2UNk9Unr5uiyN#RK#liLpP2=Ufyv`V#CFXt&-QX4M3lQgcSViMf}z{jYg^ zn|jJDamQwf;D@LJvP@qFPvI^n1V4IJtl+qK9(F8^c{LY z>ku&m><-S)O~jBM+47iGZ`cR1xBGuQ4+I zpF@Km+5m$)ivQL9-vl}dI<1p&a*|V5BTvS<1^H&b}4JSX}2unYO zCSpTl+EBVD2-^38g}yK_Zzc9v@nV36M8U!qHfWmPN80OOgA@RpjXA?PJAS0ckxN#{ zCF|Cf^y9z(+oAH6pDq9}F;7RepnG1Mjn=`?+YdUqvRK{KlPthNGSvn0#P($S=a#N4 z9;(m>hH?)X0v0QRyCUy4_S%51qZEB)Jj_rQYK=eof9|>BeMJNxQqj3A!eV$HO0%Zyd?I z520&He2$;WCz1*wyJPODo}%x8gZQY5(F_vlJG|cA5k;Wn%VxOrM@D~#QpsYV%l@49 z<$psYfc)@S-tI|Hzyau-hBjw2;|`em00(HhS_*@Q&nh%h5+F=tA4lYDbF{EU-S+mA z1io|tStO^&vHrmzq=|TW3LadYAvpf8yxE;R32nnNFxb&PK18Un;YS<|5NNkdMlR>Y;<8f?m6U+g;3>a z8@yUF3bF`ZimCHUmFK6fJ(Apd!FUQ{k^lpIHX+&efhFhT2S;S=odYIyBmnr_`@P>D z9MQv3#fePDznamEj7NQ-U2Ta8_?eM)_KO=a++#qdSbF3G#exBW+=2_38*V~5IcBVh z?E>x?b5)1}t^=**>I{^q{iI^Bmiq7`jf_yb&4!0hvd zRDh3^8!3ec72u{kuXWa9`JvmPD*=GDe^eD>WM$MfB3B3AV(6Az1=!Ok<@W|CL(Rjp zP)8^ugFMjb5enM}L?P{<-Kt@>>TD?C3#>skb?Ve2rMlOu2=3aecVvsc)ai&4Wl+x;LL`KZ^ymUMOd8A(DANgI4tEUY^9o4pTT5&K>E_-cQ3; zA*g5J$NOYzDymMB{z@>sa@GSWN@svC@CDQZPk>mgaM_mcQBTFc`{Z}8zbu{DsLPvj zPm?V$K9#dQ@xpkQ5-=b-K2z;H$~)#!b&9|jufom5!n!|hO3wB*$k9le)5RDkeW@vX zk9K3Ck+IPmJ;wzb}DGio<4#w)wHLpQ0L7dLI!=@I(=JawP|q!NmsC3J%JsR?uDO zDT~hyT);EtgkdB#7G9di9?W$7mAf2l6hFRm+tWfQGVo)0Zs7tePQAmLL~^@q#1}@j z{D`ZWePf3P+fnOF@Brz7a)Qu}Zn$XFz3(Qf!xsVi-zJY1JWdGtqCO2!I<3^}3urWZ~ z27ChgOxh#EIQ4hvs%Ycn0k_EMVw68U=qRrNTA)C4s@9WmAj=~Of_#&pUeWiv`eT`M zj{+U}{4NyChkiV{2AyB+B%uF!=`p8hiC=8>EeerN`fYBqqR9ng;Y9j zL~hpa;Uc|fVonqsvnfOaNgZGu8hgJUlp{B}I>Vz2Rz|BkgN;T~#Pv_;Y#pF#aFEGl z+X0rpKnNqpa7|9l2J?`J* z74|5=k)NmlEo`VHT1($Kl=q8a*Ou-wR&pTiWV<=^bXE@uZGfA zNdkV_g7fAv3_D2BLm)UA2O>#=1(znaiD48X2Ors8M2GqF<%<`lQZ`(lMt>Sd@E-Sn z&LE7(b^t~dSrB%ua99>%Ib_%YEEDZIyHtY5cYu6;=mEWvz8gHBfkH~5J+NA#2EZY5 z27~swt(QIKz-4X# z3e2&Zh?)(Ye+>HIn5N3Hxc?v&O?!@)Az{Hru_}V@=oMyehW2jtXk2m9=16@v z&K+2nfBe^2VJH5C0&sHI1k^MWRm~vw4HZmsf)KYtwir1&AsduB>Hq~uxKsdE5Z~cA zsB=^SuX0;>bQ565D_~Ph(oTA#gMyJ;G4BYxPo z4`6hzOAOLW?9WIhpQI@cjLl9vN38J2sJP!c#fexjn|U6$dmcEd_IrV| zTTPohKy%Kfs@jpp+i^G_)B0?b8)h^$qfh9qm^#R@3B>kcjd?opPXK?v@{+`- z#{`5N`Dp1O*&q%&T7ID^#;)LvRBOLDvr|#vd4dN4K&DWGX7}~O>hZd(W=UtTg)N5) zxP2k;m@kdYzLMDG&&dr@gWEo!f93WDRviU>)PkW)_hCt*>a1x*M1u<*N(b(}Gl=n> z%G!kdE+)PY!kpMVMj0hyoC91}RGQy#YGTnHS$7L+V)%R|{O!FX!^{L{<_Vq@HYJ*H zc`H=AJHt6CxQmAu?y|Xc00`f(au6+Sx1x$`ho-`F;nJ%3HZQAmkXIe!@(1>@!b zuNw~+_1pcvgpYu;_Y2p8@Zrf7}8dBeHLR z8-O0pW+v1fDzCEg!_C2ot~A_26XkUVUfzIuy@SC6;Co*`ws>8Xgj<4>k=>Xrg`T9O zWZLcmxNx`F!MxYW>CNm1U8;QEmpq76d>U1J&?)Z&LMZ9CFN`>F8jUE{_gTqBb0v#c z!W{Q}#?K;FcTKmP$>rgz-Ka_5%3;^P>X$I1ic-f*+IsH%Uf*I8bq6U+yrRh5ynukp%I@X(6Dm3g+&GaXEqKRA7i;g91s594RH6x{f?RaMYh zB*#uKXC5LEs^lQi3e@MPxW>!(9h0SO4|;jXflYovr$d$rx!sIrtuMU=py-Zi4xJY{M7`?N4fyen>PpgF z@q-4lM9Y-ceI1V$lw!sn5!OAur8=vWUo>`KYVpCD^al3D!}0p0K1B&q9kzer_e(EO zPR0A(snJ)EE$wJwoButbT2Nm59m6FLxhct9pG1$L@dZLMgG;o|Ik=?%OHLw)+I&n8 zkNSz6f*w8*9sV61vN4fFU)=RWwS!5!BIofwmXyCQ1QqWTxTY7{K^ATC!z_5f3J|2K zeIVlc&r%qHOLGbq40NNb0H$zWpa9Sr(4je);l0VcXZ02DgV*=OMlGsMeOxbpK?}1f z)KbcuaYU@w9-owTMeTu#{12N4wNj(|PBmvoo%pCnDJ)v*8@H?C<+>MzUX%%lxdZ7c zfjYHAiWyvFdizrn!{2t!ijXF}M*HK)M;P|$wWg!?N%T%AcXn{8MGqJ2$Kz53W=I5G zw%&yvfRneqo@hYsm)JIt5LVmh%9~FP(h=C;@ z=LOd9#(Dj8|AJ0?1nEO$J_R1kf}j1j-~*d^Z$X0r@mjDc%mtwGTTtC|1U&gdZH?do zEEA4J&PDs{f}7PQsBdcI4gy@}y78mD1s4);vSq;SP%c&>QpA;c`CI4eGwPm^>=Q~V;kNSu;r0_g`B#bFWNrZ@uAJ`FxGVL7^L(;nFRv~^^ zoj55Yj!>(o*~9Hw!E$`n$-dw_U#jVbORp-uH)#Q-#9 zTCyDIX;G8*FUWX`c!7Tn?0`YLbC$3EXl-h*C_qkC5OuNo|U8k0z z?a5lUF;NEHkb&+AVXzyJfs75&2dEg}#=uD%G1I6iO4nAz&PdO-N|;iPx>N&uff?2h z;8cDF00Ehd;ErF_=K^4bjiaj^04J{Fy^BZ0>{4xWtwWgUP9lokf~kXe|I#_JAltd4 zRS64?jnmSn<)?%);yr03!%s78knMP7z1&>nbSeBr^gIcDQ6}%Jy*S+B=>1WW!B|I; z=uUMO?6C8N2Ua;J*w8hYX+yVrM4A&__9w(*gQxarGCRk$Q&Q7*p5~v@my+nb|9Gcj zJYBd$l6uq9q7QB))4?prMZnmFxya}D&(#;;=VqdJK^yP9PKdEdCB#I#9^69~=q|3E z8Y7R}kP4`C7^S+c3mA->;2ds_nlq0*SNz3%!5G}_$uG?IU@|YUvle$51VwyBgX~8`r4&%Vr;8e2D-iP;KF$tS5*;4X0g>6AWAS@nYe0I?bu_?QFF#@U zQA<-$^UZ{+=`~D*elWwjpH}wxuHZ8H!J5~o-0*x+4Ja^D+Bj8O8tQ=qv}@u`%~k;? zS#n#~4LFw}9=F~Y3V%!EG(*Cop#D~DoRPf76RJ_ zL!`)Mm&w8>P|wYah>j$8(+z|5h?oOFSRPJhuZLJDm~~V*@`M#=Z!VsItbIMwIB=ZC z^SWC4+eW<+i>KUKH*v^f{_tzgy`vxMu-F<%!~?B*XU)Z z;F1WqJCFYI2R*U`o=!>pVF?VgoG>UCS1EgKhw%_=nbQhd6sslDnqSjcalHWjHp6U7 z+NRNk<543=iQLhQOJyn7*sQ4qrJ&cDxgvcrHyHhyUfMgiRseJT^UJV?0ynQhrOZR~ zEwESa3>I=v)Tzt6nxd#72QzGh%AsgB#o=5S&ug8%9RT87gnLX$Ddg?ovWHJ^ z=k!BU$0~QF=({I`X(ocdU^gz`7e7ol6Kn3d6s=1?xuP#(TAC2}@sKK@Qp6jF?|`gv zXz8y_s4nEw+rqb~S~SxBNb0?XUxxIBm%w@cTH@?&!HaThrMD^jg=D+|0eDWviGoNM z{5B&l-^We;u~%hJhB?`BP1Z!K&fTWlALjInMI}-#cfu&lNmliO-dg^*Y7R(2R@@`Q z7r)h&=ZkoF49fthuRgSo;*T8<2-E>eZK|g0?$k}1p#;7sQ0eKrGrG zwFQ*>mc+plUnLbk)58~bqm_FWhRusp6P^sPr=ygW4&JBFbbzAZB62-vR!3#Vst(?7 zNpswcfi-hfsaod^>XsKOJh9w^_>Kac)oWC*Owm;;a8$NUbGDswq@{9ZFeRp`Br2N< z3ULtZp5U|}@~bnp^Oe1qh(b_|tzW$#&qQZ)GF6Br84S01uW+=`(Ru?Hx#Rm3e1g0G z8+C7%J(|H(rY&?uxCUkwU^rQuFJMd37wQm!KG=FX3@LdftmM=BB-&R9HCiY!Noa`= z;f!$D`UK+=YXIt;(7cArjxpLdq_`5d^*?pWT#qrUBloIV2nnUU#bLnTOLiTn(dr)tQ6tpX| z2hEj$oTtnk-LJ5Jm)D_9!*buzL-BDxwsg=3L1P~gd<6GHUFFOkm6k>TA^`(HM>7Dw zVLJwu9-zFZ?Xu^LH%tlhKTgEy9-+ z)L?6+4~o=It0Dt6aw3L5p%+*5=RxQdYZFK%A6Mdz^s6QWsJb+*5lQt2d}mk%?^GRO zCN7z*m=1!!Roky1$@6i?R(nyJU$fvGD^?YizVt#4Q!O@HjTE&({n@uPjFI@f1X1dt z!j0kj0we7M2{ke4cwG;JEwV;$1|SkK5(oBSyYI@to-Q8HwfBHpCkmJ%GV!MvXq+&1 zSoZ18tndgEd4YT$4e9wQ)Gst`7PeG2P5f`8z?7}k*=EN|K;V2meDgDcq(Z2y33#6h zUM!)LR>ca!ry>^??Pq_nN7iVED{_v-vm=TcH5KAG3`aoBgV|7gM^~tkXVe{DTl4ZN_IAccu*5U_HZif)A1}&AQb9f zOa|(t@8XqFL$lD?5rAm0z~}&hT&accBf^q5%jwj6uz@l5*AOm#fDz~t>T)ZgW^E}0 z1~!Z!Kd1(`etzK-C*4x0~S35(FJ#=8%9eqcTH>ROI?Cjm&+(ZEVwJjggM zbg8^$#TZnKc2D%0+uBXZd&w|6oEBBGtUc`rf-C)L={tIBl+z3X4Zq3tV^<_tAf@xN zC|m>-_jvfeAD2-ynqM$_AgSax9cY;oQiU22>17kf_ddM5{)+aMmt86_CAee#&seqB zW+7WRcK769HG^&L-mE{_#5x`%+tuSn0bni;gnjLizr{(E?Q1y2CbQwj&bgf29G z@DpWVu3wLEkI&6QL0AT2V3&`G%RIvG86dtE-oW6^K|l8!8e|@LcnSh!9zBZY_&)(G z^1QAPSGSd~1zEtp0?+qzULdF>TVSmph*Vk`a7Hl3H9Ygr0fvx4UeH6-3ai2rQ(<*_ zF#&ey3+$LUWr6sChn-blE z^p|XEv}SmFJR?Brm0pWu754{F)Qo|Bm~Y?nBS#31xOl&a3jTM8#_gBs!Jk6QQ^P-) z1MGfawm)2f5%`fQkfu2~Bzgh(`fQ}a^XU73z^#Xzm6%`eq zKNB(xz*q^aSE@I=+Y7@OZ>SvqhdWc3@T0p<(CHI2DRe?Y$m*G^uzy#=h}T`gT;o|Q zw}ji6{--zywEr3*L14Dzesaxc{#nRPzbyuM1DHRwe3dkR5tnLBB%3Gj4XRy$V#@%$ zt%Q@DhE5->jJvmdowP;)%yik|Z?7i{l_^OLa-$l9f`fzY-(fx}#;rD-fRU|b{1vEh zYAkL~*P(p;`a(JEOTPRH$ZZ9ltF|%#O+8!@$_P%=T1BQ>GWU>k2c$d(aontUG;?ZU z#VXDz=Y%6cHaJ4;SZ%K>|3L!y?TOHMx*bk);1q_E?Ft>BMJvG^mlVfm=;w;3Dekln1 zZM|?$wz--aGn$C1btRz{w0mk@eGD zB>)n`7j_8+&Vp}$!QN-2C>>S$YYtSuDF=U9mOIg{FA`duHY4IZ;zcE`rZD?kx65&9 z4^k=#a*1YImNi15=Rf#oNPX;IWMgZS2lBsb zL<;YL?VF~*%!DsKQ}c%ItEDtnq};*00sww+F4> z_c@f6U$!f>)3)w?oW2ZBl9xzfAM;?`fNf&~+^Xx$nxLXqu+DE%&=T9#)%C!3dE3UW zfmCmZQQ55GnapL8zL^Yz9%)V?r!i+0QKiNrOJJ=D;rGacn2$8j%qc8Dce(5(NHKw^J8W zeXKiw%+5?(d3*7}-mZ)oPyCugrRdXss#SnCz7n-kI^XDy6JVIs)@yDd)FeEE=8Jt< z_y6dDE+ZNHE;t}2x%|xzlWJDpq zpn#f>Z-#Zl%zsAw$JvtiLFta)?Enh){il<&aGohJEy~T$@9{*!jgrY=0qMr=K1k#z zRjGn9vT;stggiOGKAnt^8?|ZJQOW*K1{KFd)P!O%f6Fg}5dN^Nn+@S>7_AU8zl+H3pA2%j2Y!ls&sE7$X0&?>L5y#cS$S1VIEpQHn-B z)oan=EWv)DGEn`yO3`b`$r$?JUX9ZK$bUhvie)|m=dSSiW<&5C`7^_iAerN8_NI_G z>Yx{2?s|$@<))2<+^$e&cVEa+$b~A~nEh&7tIoi1Z!;8zNy2NDZc}`kQT=R-wyyV+ zvW22{)M%Hf;@F9;|DLBGCqupdWp4lU@rgaTA$DEjXgdG!E#25$>Xo;)r^B2hKwjfA zWH`Y=3AvlFtJ*xj2Bd=F4K8hnwtM1>#|DEUa5XDn+ov! zt%17C%;$`lGBOn765u5zLy=31KA^*9k*e?{45Nw+`-YF~Z;&MTt2Z)PkKK%tBaWR; zw&{F2oBQkGdhA&h9J@KK9j6Y@x5!|L+Wp64o3Z>LDxSPWj+hL^CTK)OiO6qOfQdX5 zs0QdV%w?_Dx=G^wx(n#Lh%;8M}#F5KGa>>Ie~z9(7@Hxo%tmV~_7Wx3|%!sKgo*WZXsCTyi| zsc7V3iwqS#0r+pvqNtZCmK8w%mKYUUw_tjM%g%*%3kZ5o>AjTi5y!Q884C zQmPQ-DAQ!%q^=q!S+2eH&s8edQC8zFbkMOVSIKL5w4Xu6LEfbkecMuln$TBkgHNQng>|ck^PGdJ;*BuVTbyBR`nNmjI8~0pje`&9yi?8eZ?h3>yRaI~5YR!X< znVAdnF{%OIXKqaIedc-oeRG$W_79x~sbWcrB-&1CTY<#D0iMVm~){U8p3= z8cVKZy2{QWNMFKkU{{`ZTgGDP)=B;j)vf|^iS0O=4jwyxid zgI80}#Jr~xYQpze5kxG^B8g`+(AZKX2@gdlOha@%X@3?|5;^N>UwXD4xdpQhC(x-k z1-vNw6B{WT*2G<)e(1Ey_6wtu3QIxsyC-=+iPN##wZUaRB_Rcmf5HnkS8$;y0&X!| z6kj7W`7$qMYwJBWspX)sWS2$0wX0d!kH+}99@*3^=NcNGsj^XGhK3@t2?rY%E;k%Y zv@+k{c0Q5~By;TMi%=TQN5h^-Y*v2yMeT{$Gq{RxvNWFRyI&>3;(Z$e_e53o(}88t z>k-`6O%M=f2$g%DOdSQ;5se+FPhwIfoYQ>oOdC^7T+E0=el>6woks`G8cDMW$}VX( zkf-?YCh&bQ)G{5%YxVHmZgFde+w*7^NxqmJsmSCYqeay8UUzQktQ$B|RaJd^s|x=& zDAbQK_5x#e{G@Gne0@x1sN&$rnSkMsiAyLYU5@|X6FU`~C(Ss{Xs{xzf0QaK zkEXMDJ36K&ChxD}fjl030xuDG@QUYSKR?xMJ@R?wS73mf=ER62*g=mb7FaZ}DOfa6 z9k7LOGDT}VN)o7A!niQttNG7$H?XyXj0iY(0jKY&)l(k<@#{w0eZg7`_r_#n^?&yX z0P_Qos1+18eTn|IS#q`#GYtgqp$cI19n$;9y4FX(62+r&LgTWx)^7|Wt=e>J%K zo1dTV4|u&6nspmOcDeLKg`Ih)K4yr#28*s0?1qTT|e zB*8BGtkHGwu$6rJDR2773FyulcbC#3pXI%p>d%r7^R(aK=F=;Gd+ccHsYEQ7jV98Y zKrbv&Vcz@LXfpEkLIc57WZGNnl0zcK5}4TQlb5tc+s$ zo{4a-eJC|{YhGdLZ^pNfHy^Qq^U&9Ol1hThICJGjY8eN?vqP~oM9Bka1W1LJ3 z-Pot!ZxU3Q&ECz*A{M^MI4dIVTHDKu=VPWqECBUgClNw%cbIbSpxYrRbw+93KzOfU zS{+|T6+iw{vG~O-eDBX|K`pPaD}gn*`0q750_{ob6k`JL#_|aZC?e8f#q=R+2G$6R z)!!cwNlZaICutWP_jtmz#{9I2hGOxgqC`Xr+R$s42uC=5qlZ%lc9;5Od_yX`#sBTm zKn-twbMHGtSA!vqUs<%$HU*PSGoGQ{gphps zaWF69;o$JLZ-h6HrDh215aewe5)U)j%$hK-8;9{&%;H3ED9|wH52I5qq8saP^@oF4 z4i{}wcHwNq&QH?frd`(1TBTFok8R_sPjhBz&-tfPhN5pQZhe0zV*}6givgPf@nE*R zzr6kFV^%geMmCE;7{5DFcfEP$`&WRDA1ZKaMJBX!u&!8KmIO1q>wHBb{91_zz_OzO zg0&ary(^P@=MgCX`NY7ocQr}I#|;o8g-$eS>F9`a_neSy%NhO-g6r&R>kQ*}QvZeZ zbe2fWx)M8u8B9VG5ig9K00uplzx~=l>y7gm(T!hrU;1(pL;P+|mPl>5t*7?K^uH#% zO2BRr1Y&q-=$WziR2&ufH>>4Y;7Kf*7z#;;@-vB&1lr}~v;-=Fuyf7Hb=S^EpNRhfpX9Rc@3&O<|)1x z+}+rP^os9lW*GhF&jo{|zoP9qPCXyuO8+E}H1{ap+g|C*0|tNi5p`D`<=p%U2__g(bG);#6iLExqM zwn!hGiNUt1Tg|Rtdd5xfja;uGIzsO}qHyZP7ReWwqd3HWEA+F*Eir>>0W0grgX(@t zztGLFS=ruXPC62`CGnNdum$=EwNOs0*Mhh?m{lrAH&T=?qbrMgiJ)I?&GWzRN(wI) z5x|AIK%BwmRXQweFNGvc576&1;q8zrvpG1v; z;9qrgTelTm&4}-g8)HToNGt2)mlytyJ607>f$|!g1b7?y(dEVp*HuaL3>1pncl{=CJe{q`T@@ zyw)mP>tqqsXpkZo!#?}VIj@aK252%NjbO_+;5cvK1dXug0cYtgyl6XKch#BOR-!CL zr$w?s0s6o-Y-k!dfL7y za9hHoJ689z28?(odcL2tI`ejBr z=~e@Bs?0BOlka<|(eN*(3r)+H^AIM<$^Gx|LC;wz9<#Zt`43{5rJ1Belstg?@;LPP z=ov`XW+uIidea?_&5y^xe8Ru%n&d^<*Sf>ZE2Fx=_qQ?669eDDYv3OBaaJdbizx5%Cc9gfvyBY2t z79r!??IpbKmjB)7oG=+?@Lff8hJ|U!aByINgBtx}>j4l5coT6Oy0SvY_f@+IiZO#& z68BrXC%E^N5h5AtEp1Zn6eY{6a&mL(m1Hn&KItqi@We?wqA^(5#Oj~oz92KG{j|Lne=7~I%2+e>{_d8 zOQ;8I44w1!?9A^2t607KnywIZy~Qj3>X63u_Qn|n3TnYMvXsJTFlSqso)xovnmRY8 zU3-`GYE$%Ju-d;>Z@h9C*whtIu&X1#wc=7AIr-GrRPe-~R#$^2T*LJv zWC_GPfq8#iokgpWmlrvw%)Z^c zKinE6Rlq$=uFELWK<|lnE_vS9odMDTXM4K#L~sN7&KUc)I-UXm_~sejo2mLqzB!9< zR@i5fe#5}Y>6{#nyAE;KbjFZXG15JbV5rWvkFYKOi+A%Z^3YXh16 z!DZxN+Bn6e0k@v{<2Tn$ln2a&1NFp-<|_^)#yO;t08DLf?1c8WS?yRQxBLmUDGY>r z-dGbCloBxAe-9tYn*Y(;Q;xU3_tKTwvkByMX(@VU8%I?w!+i+2Q$o|%AC}UO8__R1zQESnqQpN zdT!c)Bv(P89@?%f(S~%61B2G0;%Bxnf8cMuzZP;fF;xxYG&b4ePp;_x^U-$|5T!6c zpsEyAv47ltXS&(wR=80pfr>a~F=DrOx*cL0I%^7OB=n6U-6&NK`?b~-DU>l@U2|o` z)G{{yDJgM+({swTn7A6pZ$50FGS4BDABLWmu}WBsSZjn`C8Ft~OHpbCP7Nfq<7%HF zs5I!Tl*6Iyqq(*(d7tGg{wOY9>gWn`NQQ}&9q>tBu;=x>sEB7~XXmFV*XQWe$C1D2 zua}X+{TbA?4mbA@JYL>2f4BFLqr@~=UnAB2==J`1J9A5!a}y( z-?vZr?sbmH1qwkiM=1#YH$JDWsc_q;-8K86?D72^j5^^iO{B-der}k35d_%QKFwe*+Ek*IRuo^+EN`|ozba^ASy@Z|YoUB(Q6E`W)W){JZ9Y{qruQ{hAK&h5S z$fo@S78T5Wxo5VLHPZ+e?ow2;3~Or^8MA^IR#z9PU8{_R*XRBa>}btn9_f*7wt=-| z%Zx?s8W5m;%M=zY4B{M$DZ!vlTnNE+QpCTVuU~SBMkVD3UKMVeWm1g$KF{CKC|V=q zaS|&@*j?K*pwYyhMQ1O*aQYB5!^deCxmSBicB$+6(*DX@qx1dKhc0ZyqQLeqFn7&t z_4a~BBdguhnoKveBskV`9Mm}V+6x81$k4o5&RrJFI`qplHb;M?9{*su1;bJ5E5uj158&xVxe7OaG5 z(f8BKWCe|omGhw9-Q7WZHHu?oI}}h0iGr9qnthpfpLhSRK0@b&O=)Bu{rST!O}?Uz zvSjLK`NgWnw!eto6%J1rDQ-WHa<=&5Q`#iT(}UJ%y~N;9A}ag&xJ$ZAb+@QKa3Kw^ zV%R+oqS;o`Yzx(RxO-E8mcz^cd(r>z>mezD*T;8?)Yw38Cn&J}MLos?I^_PS+17+W z;(rZ_7>77Pu+HH`v*~b%h{hiiKic)7#ur;jK;YI3+CWnK4FF?*6g>5!fV@(Iqw+66 z6XU#mxpTJK$SZY{cz9@sK_Q}z{CgDTJ_d}t(w{yVdc~S(sE{Brq=$xv7G!UH+}6aF zA*ydnk;F#KZjw5vDUUrHqa7*EN@{-$ijq8NoIc5xP^ltgI8lV}@m~{72)4q?uu;)S zL%x1TCcv&;(SebsvoO_~ZV>q3o8)12Zf@^V@s{Wxg`>h~kr-(rr zuS@2mSziq2Bs|K!VL(8@&*{U_s0|>mn6nc!%wLA~^h=6o}r-(;)pp zR`LWFh&T73x$XF^$ej<#x14mlntIga zJ{WEfFo_gQk3g78Nbl=T;CBY(F*3)R-ipuoP6bNaAy_Dwp}%E~M~G}0A0Iu8`!GoV z$tfk_hMJn%#YNM^BmBDwzdRWu<$XbBb?!||OIw3F865qy4A43-XJ>jEc3zipQXLTA zn|P$uheuNiAe-#<%}DFvdNB&I>sd-ksijbGzz7`Gq=jzlR<4sb?zW@k4~nkxchFyr z>!9V!_Y=K>w8mM%w{EXct`kS|F2oF|<6VANNd&fF=7d zi>K5gcQu$?xH)x~rgNgcG+{g0a{6$f68ttCgAPPe#z!Mw)5+ z7%N(1*FR)V%kyGj6|E%kh{*`9*Jcd;I`-hH;3)6TW_b3KxFGvDi4^fr#27j&8Jhn& zDu}xRMgtt+yh8SUuLz(leGoZfZ!t!EWi>EkT~m9I?&p~G@wSR{Gw1$rH8~{-_gNO)`IkpX23;Ok6b@}}JWfiD_CgLw4EuQl@cwRvG zK9q!I6K!8)wv~GG)ANvkStZZTH(p13137VBRo2Z)Hy5dN_`@&h>U|=@^wS6b6;Sfz zHWdTi?+pE-(32U}jX)`-jMW+lN)ao~ma|7SH8q7ea{7psv64nOQLhx1T@di;0)YoR z$ycX|&Sj0J0RT6+x6=Alc7SeNOI|H9=?>XyiqtFEONT{1C2OTYA{m*xGV1P8mW>6*fsFS!}^qk z*cDTZr7E1)`3VpGu8BdV3DqG4{b5XGR;t{cw+iyQn9CLrCV|yj&A-9;!MVP{Q}%t= zlozd8zoXA-l9!;==XLUfukkefLFC>GufK8n&!6y(C!ck416PbpS_AWGjviZ=-*p7u z2S$jgRkj3un8J1Nl)5GmEZ0g49a1BXmw|zu)XZOe$|Jug7`jzhtoiTP=9cq@8oSjv z1wMQo)}i^R&YOVV;RlEMDj9l=>Lcg~r05I9=!~A?X26Kvt_4HqM9v^*nmS>>nVw{{}1PfPs33_V0ZR9o-{bd52&U@Pl>Ba>kAwPBC6 zQw-?Q1q|11MRrVB1*WV^vUk4dn*oQ?th z{Wq9#xvQ~-@Qd_eLq0Dq0F}RP{9c1lm2V1Kh9{o)NA+jGez*b50COLR#$w}%UG3F% z$uO829mrKQ%roto&ZT9Ag%3gq*GTC%!w|eMV5yPic0E~fejBc+9N)h=?*sjb#HqVb z_ay8j?6af|yu%hl!P5cWi};FJrZdLIw{iT2J0ix5H|GMYPo>b3JxUFog!bI&&m5=^ zZRuZ*$f{?bgVA>CjfvgeKO7t!W7ElO5U>U}is}C>UI13k)95xPdZWMD+KppZMurt* z=Kbe`R;nL#m;#M_(mbf(eC#i2f&S-HWE$UwW8GL(8bSkLOqK&D%G`}I&ngXtbkI#i zyoIvhE*nT9h)v8|!`)$pvKbc35=m_Y5}Hv5boTdzG0}5>F)G4djB!85y)Rl{$XPSC z#jq&KlZh^wP)`pX_bs2V4!Sz9^=SFdLzqFsiM;@oEEK9F zO?>d)#<_w$Is!(>4bJ-pi3Hz`or6M749Pm+1r9F}N+Q3a5DCiXqY%{R{Yl}icgDXD z&U39M4MvXD(@+JUS1^iaWy`+m3ma0Zz1rmRzC27p_(Cve5B({>Gu}f#fT4Kw{?YOA z<&eneXzoxr#rx|cpf&37_PPiO3)@JL!s>)NZ#3&fpKdql4X>cAn4Ur?O8_LszN7Z{ zaUhwPnDpiCK0lp2k4Q)MyKLm69)c6V7t0?`)e3+4g>r?vN^tOSHV;c^Fh*a_Op|!L zpoeKKga5qUD6G)1bmKlRhASu1gC?Q>Sj5O+_y!^I=k@K6m{yr?FR6?08DC@V)F{ity^V!? z8;|t2b*JK27u#`;6-qyLB7D%E3D?U#2N?uHF<6&XaV+ZqY8MnJkm>6Q>!N}<} zq`MYDKsp8KR6wL9lvui3k?xdkkb2MJ_kHL6&HVn@!5Nuh@4e@q^E}U|V!DWFY2Ro@ zJHDGBkhoMXJplQy#v^&pnDQvmQ$GbcekBFJ+0NK)aA)}|)F%+$bcuFA+r8V!D*rHWShr-~6KVQy}B zL6$Jg79Km$>ToIvuaH!FY?u7l`TKl*6KV1QbYfq0V4pobb!l-0iSAh(hP!TCH+cP# zUC;z*0_;t^qYMwBDia&=Ra|iUp*qTp6-thBCMS&flQ`HMM*-!D z=m4oeY;w;q2UZ3U}f6HKH zAYFk=4u7hsxC^E+vWk7zVxq26dcAK`^mEsp`aUQ9fPgGA3N;=~0$JMC88uy6YRrzv zfbL(!vzVco%Kepk?fSP*jP{t?( zFvyG1oO$=cYM-ZX`7OGj&w~b>O+4pq56tm)8ltN_dAW(9clQ+z7gyMYew0X#EWk6T zo}??(39t6_$9hsDmw~LxwDNBX1HgZ8Xp^{K zy>Z(2xw$$$18}ryi~bF85KjhNe~W=MD_R48AgvYI1^cr{on|S{Vnonclxn8_eR(uePjCq&`NGDv zf+e*p*wb(in*u9-ZdzF@Puoc|F%M0igBTt&EFKq@2Fk5D0jb!vU@B{}YP;#jz<L?|9 z)6O8+%{H`#*Axzr;*y5353Tr62sE9HEew+gY)nro`#(Oy z1cP{?>~ZoBc0CEKpB$*9>*|sKN#vu09Z39fo|WBqAy0{X>t?}3NTd=y6ua_H(DwJ= z8h{AoFZDUPv3vy~KC#ec3-=TnA`v@&*YqR@;5mTrt>BDUuG2_) ztfitqG#dP6awAdNI7D!h%ixd2W!})?X%iHPU12BW$YLvy00LskVCN4Ux)>xK%ACt` zJje|V4YBUrk|$VslCX99 zXIUq?2~|GRxSb|=lPyU|Z4v{woosqw_0{4cy}E9I<&?YR$=%M%=l_;PXc!E6r=7CE zviOG5WsuJxNYWm;=kTK_{pnya2M81mE^~4lLk$)eS*PFZf8@q7R2W|=t)ffF%F61o zRo)5m;iuV8fMwNYKX0K+fEtX}i@ih#N^)<~k4t!BLPZa|21WS!*IMbpsCj&Nrc0W_ zRDf`DvNx+6389$GA1OjGI|B8ot7%;XyMRw7a;L451yDZnB!S*pDp)JQ0L?+rr8|{% zzTguA=IsmI_U0=#a#l`EO)E+fXXkv@JXEz$D8Q{_+MRCiXmpHtt$GUnoEv@C3unB# z-_w{g+mf)I*8fH%_=zeBL3Tj_77&|CD-F!HIs#tIOps&m;GlaoO9rl(hoJ5Jw{H-F z1_GN%9r}u)Cjv(gaQM$c#HZ7MDyxAGN~7ecNIH}$zbay1*?otK$^h9BlIt@l zFxs6zUkk9zKyAFbm&~LAM7jab;)@Z|!)wsOSy4A_M|l<9N4U*T6`ntTfmL(hL^WAi zQ85B@zk0Dt<8=UHfLk)HKP?XP{3bu#Gc(gBXm69DXYHP-lOUf>G^E7+)v;$A5Mk#@ zK2PdMsij+s{##c6GMgo zz5c{eluHTvK@7LA5u$ybfY)6v$Cb&J^#D4)$HlY8X4JebdBSIznd8B9G;iFMlj)B! z>rLQsq%vPQc)~T$p^pVSz}f>DL6Su*Uq9NEHrxZ@6&-WeCqj$|`kpI^KpTdF50)1& zzUWE-wqK@QF$^Ai&6vDkcqPLV*t9rILe{%2G3Z6n(FuM|PN-1z9DH4~kBwkRj(=ww z0xu9@I#khpD1NP)-GNfh2+xwS%r$b7_)~<1yfiEJg05S<028}oWKfqQen)|x9s_H` zNNdT&{wd|hw4^m1NUOjION=rF zdcF1%5>O7;^a#?gqz37{j*4BsT!)QNZVY8ZG4QEF0lG@Lo1T~V;AmrXjeBKCnxGw6 zmWAoPsY7>E+{^LcCvg{X$54V1JzfkISzcmI9N;#MGB}drjvOkHlxgziCRCZcC*CX= zM0a>hhz%n_cfYhfltA_=VIzNrW3!R32Q?PUcODt41`l#>u^a-zk@kr?q57?&K0_gt z@>clb6R?9Coiic^C{7wah-2AaVYBq1J(+^~zWK)c_t#mj2fd<*xOCCZNWP@kcY|L8 zJ}{-DLfbi<^>Ms$Ud?EqlGQbcEfxZ{PQFC{I2aa}D}2d2%AV6sIOUQIL@lK%T zb&@A0qMCV&6Udj%xgU*uO8f$eq>^LzH@la}P=k)0TB?2p>?EOH?O;3+(dAA*wue5a zt7i9sGMeAa1)!e(`D575nEiR6Cs7cSpuzo#G{8Nr`l_G2G|19cS6{x&X8B4kAt4d( z&74&dDYO50e7Ah}FV03KoMH{ybb}|KzGz$gbQQ88c>~=_Hft>}`tW%RC)gH8Aue_6 z&RAtCEljphvS&?mPbFI7_MJ%l@11yYN6h}<_wxy?1PdMRjAY{j@*QgB=PX+fXz_8q zAXpKYwdA8RvwM)gS9C)w5QH3FK-ij!CWc}rzG6CCKm9A~V)Y3kM))0K__+33mWX5k z%Y(Laa6AZ$i;Is~{4AOq0Y|OJjqm)$ZossDQo6yl>3~@9uO0(Dj)p<6Ki^I7xlHdy zFpi-T$*|V6>q$ zZwJIENh-fJu+q^|cofiAC+QjcZJ#_=E@Mct&A}mTG2l#I_oum_)zjMWHZ0Fow{%BGt0Mk;crcXA?3Mw`0r2vdn7_;M7vCvDak}a20Y-fK%;)|(LfJH$sPvNiIz5Eqr71UxqOORAZ@TL4Fr>4D(437U9;uqkw zM`<+NwCvZYA@XbkB|h(v`Xq-sX}(@k4#Jee*dcBiyfFl>1#qAKg&=oVcnB~Vv6Cq zG$=D|$w8$vq4)nz^qp9#Az|MQ|1IB~5|$UZktxrUAe+BMsi-e6ZMkUN8A4`-bDyFL z&7h_ER$u)Psbc1Bn*!piSQgzBcuti78#{Zs=b;6Xe-EblY$sKOv$yXx_Q4obCw#?`Mp=b}>W^tm{4+P% z_jq4a%80Rcv5vmet~*c?CSaY(jwgPcz9&S&!iq9-gQigF_Og$31w~aGt1le~?4lgn zD7LegFk?7|CW+FqJe8S(3XKCZ4GJ%vn=@=7MIN_l`D)&_aEpD&|T@pQG8k ztl$-R#;#z0UMdDe8SD{e=K~gVTIA0@&~I*roVRfB!}K_>}n|X5J|EOUSGU z$OCTA@V_Po_RjN#pW$jRzhy^}qoGveVR{qpKT5!WzTzI*A8?n;Q-sIl;Gwj z0Cs`A7oc406Kr?B?SahYl~Er&*vchXGKosy&RgAdNzNR>$ZdkTxGyNea8kOm{#GgoCjesH_IUji!i z?ekZBY-}-o@U~xYOI8|1=gUA7&5vTcVYEqBmuo>=vgWm}A@ zG|>7TRqD>z4>^^Tm@0mw?z={_FwoB!9hhARHFzd_TlSF~wPCTJ)UM4J;#Iv;BR6rh zod_FYa`gO@2B>z(H0L|Rs~O?U_wPf>%gaY-N0pyGl-g(EdlWz;YTHxqwr#z)Tv&us zeS_8a9gSuDwPr}Uz(fevxcon?WPmItjckBa>9FCLpDQgZmQE5X=4RrV?v3& zf)jf39)99kGu+i%;rUp7_}JQK)irEHKFF>7m}_4?b&Z!EH-J=>P_tahO%vs62kpBd z)>+8Z((P1aU;e)auH!AcqqEjzjGdpc=JG_+X=F|n{d)N!5U`SJp%6&m*8gSO8`DGB zl%j%MaX^1=CB~zw&Rkv$%lU~7@Wcm9OeUPfKd^utu_Ki^#<~b;s@InyfI}ndCWG?m zf}K{@rHZs(E~;wK=0OhK(3)n}uAOAkA-t{CT<_J#{ENw}qoHlD0BXhyo%#Di^GBDm%0%3M8+~UVxmM6&m!A1W3MCRnHRP={9j=dDP=AkQ={9%hUnE|>XlBf zT?rehFdo@EKMVjtbcUk1-I~NUV<>d~xQeMAJZfMlq4MX2nc0IVu@h}8wNWI=&7?IL zuj7X$AXsQO`H6|2FB1cgVp1TMmPugyLW?QpUeL*_IAthMyO$x^GzAYhMWq>Z=)g}; zPsbJpEqQ?9w;(QW8vD}kBMM0?BA_gnkF`#q_u zVfEaVWe?P?yoVG{?-(>|?+Ve62}`Qza*=o%`p>M+FuY-vx^DTvi3sP0qzVdC0S1cD z{{DXL$)pW3<`iila0X($*|*CfARFLz=6`uHcM}{PO@Krn+pK(Kq62&%gg`F-u-`xq zGtg?1_NC+chQ5S&99LYB=Cc|VEC4&<${T!=H?kU%)NEJ8${VC8+&8bkXsRyjI0-2y z=pH`SzAcO_*A=y7iXvzhpZnj5Uva zMbg}U4LS&J1*%$M!`#lDECX{`jyrz;CIeQo2!I4cBONaqyiYrZvZeDmEr1dOY3uu$ z$cMb7e@gGOYevg=*RkAK&z^Dn9gQsjHumI$GN)Oj_KyT2M-;y3!pP)>SC8|~z+UPp z!a@D`J1W+uv6i>(6=kyPotVMBSXULE+dD2yxNa+nyoM9GW+tQV2E?=tT!34T=(7S- zJ=f!BTuSJC7Es{mQGLVVzw6+}rqKJ&_;+$RGjr#JGR|up-ETc!#{iv?D55rpKY97P z#J~|a4pC&U0idX}>H3&^8W^OUP0d|f1krgN{!BVLazRo&J>~;2L14YsBCzo21buWc z&_)W&ZPA4(0bE;ztLf4y4NMVhfr4cp+=%-U5_7^ic2cpH%@Cs%1@?oXFQ1MO4UE;p!bS+|>+56U;{(C^!aiIZ49U!-L%JmaES97bl#Cq_6oH_F zUR(cL32^*A&_Bd8zw7{lw>{vXum~>fscy>jp*3m~W5fG(JAQ!8#|y24@}m<_>@0(5 zgOhBYgC6~7r*@cG7a@sRuE7;L7k;-M)T}R7$*u8{GyN;&Kq54Mk7%PtbAQM$!F@@|-mjDq~%ETmxC}CZyN~A2vN(2%2ft# z3UV7{4{0p&d}BeYFqw&q>yJwA(eyRsEg75Q#KTn=SgY508*WXG*|pbqs!88}r!3lk z!=P&{1&r^nEVM*Aj=p7xQq2O>Mw9qqJ}rX76PeR^Q?5j(x_YKz&vL*QuEq8}wAf5k z(|SouDY&$hE&+BHGH`Czg7BP}loac>4B*z5LTGmbc9SwU&O;OaG;!a*^T4dSSwqX< zoix5MT61&rw|g-F2;T!_e|f#=PKoP%0&36Y*w=H7Vgs;4#We4ox~F-6fuJWXS5M+- zalxFG3uAE+-XNl@@nMSE@10(Nq%e;K%Y1H8?TDhX{?Q$@%a$~}yp;Ja!gg~{KAv$C z>3fgbj804}KO?INmJIm6)=ZVE-Un_dGorN4jG zDuf4zJNNd$fvR2Dd5i5RIk}ju)o*-q9uzF+8Zy=}RZHMU=m<*PC8WMc&%WUu z5`O~98zlJaIP#3y7`SGh0S<1Y1epBEs+(^H)bQ(i?F{99E~)cf*N}OcE)|GD;n;~j z7yVZKqss#@Ri8!MfAy2B9?k)LJ33scq-pvL3-a0ctQpI!X=Gm`N!|U754=E*9jGJ&WcN3#KTcf zp^eQQ;-+EtSPnzdZgk8|VIyy|zn*>0EJm~rN&ERBSvyyMg)z6FtC^q^$nti--FAZ+ zMw8tw=Me!#>2}11 z!~%xiOx{)TaBX@wazY_hO3MP=d#bkUf?h}JiK3XSF9N>!z82lbOISM>3d2)Ciev4L*>P~H$C;&_(0W z!!R?SNV^NoS8)G>rkH|Of5FA${ORzmwaYeknh&4DFAcz#lB7w8=<@fyXxVHo(>Stz zz0IRPF7R7F5QH=>Wd-6pRkLA+wX{6pc-1)VaWvE#;y;)vK?hxb@>>$f7A<7*F=;(e zt}ag9tH=5P_wOwb+wL+=Q=u0O*q-5s=P}m({qrS~jG$8;=->7)>}TsIg8Y|!gM(3N z2mN+No^!W@ROGivVPsWhu%f|vBd%|a-<6k|J*%QXhkCNFfnTGLX30_Opu|OwduiLS zT#`v!C^m6R|025m#ExDSs^3&oLbC;tLQ6($QusyGQB3MM+K`3M?=L zdjwSe5KPfdEdc?+i&qHY!aAlBSw-`Sz1efoOKeTV?=QF-sQU#GneYlYenG+FrAFy1 z)0WNH_;}O5m%u=k(mS#`b~W&KpkWcn34f}upQbaE^&1d`iY|VrN{lP58xNElX?>(; zdhnnNR=FMJJiSnty>2 zHFYiN72JImRnTsc&5vp_l9Kswm)e`dp@Twm%gP26nEa$^Mdd^A*u+HhwlokIMgjH% z#xYIV#aq(Y zlQq(y%wf}H$Z=w1YkW(48nx&z`+MH=u@s(!*58V@~4gXw-37mTaN%V&~CzF~go%jop+*8*fI_ABNRGII
Q_?KjSz6`I;O;tR|hp+JT1L+v*+v!@z+#7VbxiDZX9g)gL7IW?inVI z6z(itR=H^pl5y$32BATPS^V6#%=&nP$`RC?+bQv?KtNLX)A~$(xrNyuy?%tL7m-D7 z&by>cB&JI#Q>KHN&yQbETTd|L9OUcl;!-6SHe+|k>!2@PlG-ZU3v9S+4HrSk8;@aPWM%khjAW z_fF@oFChIeCUb>FhT<-~FrIrhe8JtC7zW9b?6@N<3g4R@ab)Ux+D{|-xO)D0sQjHJ z5}zinh}<^IT-q72rPu4|B9-McT1O2UNz__<5&7cj)8LP(v_5;yPlvQp$Oc0E0K`gM zWaoeSYy~l#0Pg3G!&Z-5OSv(q%hJAvg#%~AlF|Yrq{N}oKteM;;<4S*fkM~xJrJ2uhAvl&joX~XYJMyk|b6-f`KY)=V0paJ~eqE z7}0v{$FIF#xvnBziY>l|t@vgw%=`%>BUy#Y_>%kmh;k+S`>SEe;=bRBGS4SLTFe57 zIN_bM-x@DJzGrzh5!paHeC!u_+<4@r_ z5z7aMfbA6k9FfHTVtJez%r8aT;nQecQKIuFa$I`q5qL+eL>w`}8wG{Vx3CG$BjtYb*KpLl#D%fFv%<{4OWAzJ`Wx0()-T#cBD0@;4Wwr?UQ+& z6huFQ6Fz|%$K!-bFBQTH~>Q6{5MfS&Ol=aOawKt zH$8?0L92)O6b&zaH-LZJ8m3VG-`3D$qKqwl08Jzfqb&jjvRGW8mM%F#JE?Ob*Lxkz zjL=N*mExuiHoLfV24m$4T47mP&0<#WnrM9$aZ?P3wRD7KChF+t-IJZ{$1)e2wT_Zi z>N-hUV6xjU9mU@R`lfqxD&I)SOH<2NI-}|AX>Hg_eEa6q=*XS60pm>|^O-%Z*Wys% z8lda9Ajh%D8ircBG!5LN>1BFzhymI4^))$&Ti9v3@muPb9?S*2Wip_cHP2neIV2f( z7r)|P)<;SGXejqrf0e~lUG$`eq<^8fx)(?YLdhiL<%)AE~Z{epjT9+;R(c~VAr3D?`><5j_}$;vGYdzjZk&*v(eCc5( zq7?me4~R#6z$Dz`UR60CjIhDU`uVdvgtYi+PS`_VQ+)w62!nbxQag1q$SWxD`Hz|g zXAiA2RF<}wdK8biko1#6Q&Rh=!VU zqKJ(r`f7M*RZT{g*!K(13bBZ98672yUB(nU;(-PS4Uj~tGwYigSGrwFioqUtuBZ@i+( z<)`)%8__AMo{jES(Hj;s#=G<}yCj^Xz>07^VNsxJ5hBOh#d|>erj;gXZQ`ApNQox+ zI|)zac1w_hhQ*d3D?My8&V)M!O^S*a3LECqFL(xO0r6?pmAac(UC-8b8!r)HoF1xb zYA+_Ltm3OQxNWDZ@lu8CJ31n%%0Vc!TCM%YNWONN5$gT>_vd%HPdW0tv2fA6PO6@G z+N~4DzuoC(4UO0j)Kd3HyqsZ96vTd)vCg21G1bY_5q9k8_K#el{ofkq8o8;e^H`@{ zpu%p8C}YqBcfeR6zd5#}^J?L)>v#7WWcR$U4{S}yU_h^Z25hbR<|J+|9~pXY3)F+y zk{>-FnK1zDJe03MYIZO##eH8X&R>}MhG%Ce0C$FBY&8u=;tohA<*B=D>cjVfls(>^ z#t31#ov=M8#WUIN|SragF|B(tD-hQmX?=nCq0mx`J9|a8* ztwznJb}?3YzOcBsB|AiOVXqZP@ZreAVA%uNrqlKUKipg1IpBRKR_mF|WB`LvVsPr+ zn*~nmYG*#c6$Kb<)9f@}yaDreky6c+xGyD;r%^7~<2cRYI_tlGU|nIzLZoI;<_Il; zSA($+1jYW<8UbchgU~_F`n&A;?(D0qPOtv~zD*j&1>1 z1GE7*RHGj6-3H5gPo!&e1Uc`MNl#AwzX=$REN~qzwh`UTqEtoA1)iVNrtswq97q;} z)49?sT^bF4;ZB8(i~*d+Nee1!37PUkQT>dPnExN`=}+M8cA(H3_)HH+6<#kODVtqc zPMK)#=4y2PG>gA;LL1OloFb?@^4nHDD==Cfv4jv~PcXv7?R3je>l4 z`?fFzf!Jo3^YYz&yi7RXj2ZBr*iH$caxgNl9Y6=d1lsRT3d^(rV2D71^p`=V0)mu4 zqzmj1?Dlk!rpLg-g4H+E`kU=VcxL8tX#KI~{Pjh^;XtKf9CI3qzDxPhHv&|$SF}O1 z(xUPq982#J4_?Q4$`(eE&2t7O^q6G5rUDy`d?v%vg^y*^?Yr$NE0+e|GzFYyPTR@V zmjQq;Qptr`VP>LDjGhYj{0;F>gBJU0FvWR;7N}k=YR2~T5ll~pjy^9CFh^n0>8j?R zUmP~qdzO!X7wEzg(KooB^+ul0FD39`dnkKbm58KQ5c>vn#bo+ldNiXX?MvC(OK%4| zBdJ1wnb-2#&9Ad=J1U_!ZIj<@RCIN}ciJR=Vr1z$sdJw(M|TmQ^%dG13e>1Q{(y~3 zbHB5e`U`(^eaQnu)5)LA;96cSN~s0gt=!wi1grL*O6`LnD?m|m0Vqi?`EgXFX}eN{N%F2bUJ zsv9jl0aze+-e*jD*z{D8Np8|Wfr@6I@Ui-QaV1<$soW5z_)97`FNcLQoKvqFaCNym zfxWw3P}(1dDL}@JMpC+f-KTRk+l%-8Ar?59V2%?(-mILQC?7S5iM=fH+K)(fLIT&6 z+@FV#-PFEG{>QlW#RGch#%c!SOI?WhuS(>MV52V<1?o}pxg(0Q6LWgsYDMgD_*OpR z@9+HjI|;oHcq^g}Sl4&%EJyhVQ9n{<`k1@r=w}RyE-UJuUO=j$XAfuJO^-az(=KvR z!av-XCk|zmY4S7gK#Uutpc%7dpLt}N%3T!T0CZ;M$;HH zVDk((x--Jc@?szu4`JUloH`P=^W*;V@CCgyL)p2hQSWa_<1ldN5`6Enn(i)}+9sO? z>spLgt6Q0+x&pc^;T32HY4JV%CRi9^OcF0LT8@lfgLe0N2ty znYjaW)(kE)BCuV zpop$hV+o4MccX-kL7>!HW-KeO-~=}Bks=mu?oKd0BBG&*EGui`lANzDC}1HT_*z`t zH&Mj@aH}ba3thAH1B2Z6<8@vVOBCVWfH=jCsyY=?H`zi`alVqZ#jkCQ0SImOWD8D6 z^^L8w!|j=%%KqMkhg1zjfbgivly{Sm*kpv>uL!0V33zcDqS4C9OzW-%jT z83-jiqVu+W7#A0|uvU|#uLk(99zH%kgm>@0{R$dG0cHZ}@93b|Gfn*TE~h~s+jb5= z;lm`%@<;u{a5X!Kr@aJy71cc5_Lee1gKE}?T5%7G&PsWzXXWxb z{AyngApf8&g?!+~6tu@s&HL_1dX^)L2%($2s|YPRKwCvqu zfI97%CljH&Y(0p#8u%1GYPMBAL;2}yS!5`DTxh&eStVg|K^H+PfQGfpIVksBbR_64 zvX)TQ`|oiH0@~Jm)Y#be_b=#7<1kTumCd*3UgQeE83~WCpb8RwXX+D6WG4X0?dy%j zJXywn%2+V%?RO%6>_TkD#_dQP4-ViOUy9ETJLweb-#G8_WJX`S)oipkf0z_wBM*r0 zU)wr6m%`Xfc%BRZVq|a#ogX=%OQy2|o5z@m4qtyT#po|r7;P3$GV!%~CVamYn~=-^ z>s8XB`n6YTU`Y5&@~71mXe_-+a(Ez=*p{g**(FI+m36*qmvRFH_E>Tdv}lV=2cpgrs-8 zMd&Bxrz8o~OGm-7zYm3F$l`!I$2^2H*nWx>N2N;-Gp}dd(D@c(?=b{pD6lB4J}wud zts1b$H@-3-(Sih>q6CXU+oEEcdVz2Oak&{lqh{t*YZwJ&8T}Xl#o-}zN()v~p^QCn zN{vX(5bm2PlUV<#rfekU)_64*?54;1E5%S06%~y6fYJ{Ilg8}!8i_zsArL+hgH*8w zB9HY6XOOvjyPG3HConBVCeQ9U35Bvn;^ebwtBZa$ZTtZOF{F!Nh1s77`eoYge3)WS-n5h=Jt=6vp(fz=Y%WK42 zKRF;#>}b9zBs$VyniSrmXNb_Gv6RnlC;y*H;c++&{ zHVs->B=UvSJ4>$W)0=*vu}J{hp>oMgJwT~N z@gq9G|6`uda?F&|)IUm`@cQPHrT7%}RLX5(^y9yhRp$&1^}1ZrC{FP#E1rX0qha?fcK>_GqJ}V5^r%}Oa?pbdn6AKrm2Ka&2x02?@>kE zAT1#V&?A3Dun#9Unww%p9ZIiL07~5*+{Askq(fBhbZbD6LQto4^GPn6aK(u~pPVTj zSv}YWw+##!syJgUPqrqQ9DzM1T@=7qO~B^*i5iSElOH?Y_xroSye1FUWc#55j_1g= zjL|p_V(12Ce+f0yTGN$I*EARcr9=6CH4`TOOb#`md;m3U*a~I)~s4T?q+g z2n~k57wb@iC}FWx4vDN|R`OvFdudiy79tb|#!kF^e0<6S<1tV4^{GuvO?l0GNS3y9 z`SEZ+KHE>fe|aWtHa`6DX$Pq#U_Mviyi^E>$C*%=w)MW47pFgejNqR3{FcLl7d(5$ zkZull%`>RdVBO}ltRwZ}m%C+AlTRjewf=eo#`W!>k6Bn+GF{XQ2@2X`Yi#86KoYKD znPod*xR#fOfz}~R^+gK$ymQo!MJNoma6rjT2T)#qJiatUcF1JC$VK47%+n||08q*+ z6#c*itz&X|jpzR}{slxzVOWH}}p7ycSLm1l;49|}Q#3E~6f(i7*Cw_51KS`G= zT;}hn+f&eH)Qa+)k^P^sQE3b!?q8)z%GO#%samG&qS8-GlKHax2 z>mP32>rKO4{#u=V7dgJoj~re7KNAAt!&0ElZF5pLb6Jd#W=7rZohg zHs6=?IYNNX=(j!xDk>|> zeJ|W9Ltil7<;N2ih5;woca;G+&op7^9DgM?pUqQYQy*astqFS4LK5 z-Y1q9PnM+wUGzM-vu2(N2BqG<0FIMsU9E|40F6Y5_bWHBJfHShiu3>|&rJniiMV>E z$2Wf6zo+(i>jHbY)1`ri#DgZKzXQKmyI;T2I?TVuSmtnb1EW+Vldc9o8AhR9-V%H) zd?3$$mge`p|A+JF3j_`~%FQ`}!?dGdY|Y4l8px`@vlPuAd9NzskI zA2%~$VkEv{Zjjzgln>(b9^~ro=&B_7QhSg4#KayM!IaCQ$gKJ%Xn}50G<-gebMZe&=E?{NoaeYkbaj#@kWyzJb8?A!u znGSBkT30YO?dwz4$R6=hX&R$}y>!u!>;&WAf0e|&!6P9-!MTUxe~DWB${w9M*zTfJ z0{*2GqvShw?JDKP<2U-PXtdk8u<&X(v8rnt^2NnP;E1e6FD6Dq0NTL}-xHnX<>h%m z-)KWtMqmsw$zk*Z81xZRQEk>MSaGa3w0i&SNj&je*;GRjJmpwx5TnmeDzmw`0F|b;bCnya|fL&Jo!{?l3w~-S>q(Rp~l<8 z`d^hU8>;ci2@ANOiZLyt8U?9Fo*VgIL=yT(+H8WtOPA=?}r-MEoTolw{ zbWwo61CmA|Z!c84-lKfOT`q!rAtciWE&azm@8W_EXB<%|%vOe6J<7m^w>5>$e+ z()s*sfD~cjD5I@Sj)6-)Z~IluRzaVDVJ2-l{_9X{??g6q=p`3Lyk6y$4XXiN6SIH( zm+!jXSq`d z3<*}1&OviZL`=-+TKDQp1MTHk>O)ntS^(0%@JH~Lj#MGRkd9PKkXgA(xhJlztPBTa zq$LnE4zgXQ7pW({clGzn&NccX0Z@wA@TP|BJ!&pMVGp5kOV&ZD6&io}IUn_vKVGo` z=FtwGzJnX8Ptpum`W3gN(cPgH7%LeVbvFNC;l{W(i@p36JuV8>%aBn>9Pdse_xLNM z9l1WiYNl#RCMmmSa&?Pa%k(QwNU%a*GBj@x%|s5WA`Q<%16@Gy6A3#-kWQ!}H3!W~ zw?A0!UI!U&I+7(G7~dM2*fBAp)FzZ3zckWQISEo?{{Q@tSYt!ddGXXsx?^or8-cM@ z8qz3{x&`F;y@?3*4lFO7^H{=LICZliQ(lgU5Q<*q6$xi-CK4cPC7nUaQ2`MBqK?-wli4koEE5g+R3J zw>PK#Uz%PDMm_SN8Rt5how=+nku#Wl`aIAs6Qc%E_^A!p3d9FnIJpcP?Iakm%8sxy|3bI*K?~t_8r1dLyduMAuJwG&S8`QX_ zprBY42#Tf#B*KGk=1-NDYr7i`6>USVR#sLP74pLhR^pEX*4omI(KXhs#968xV-Pk7 zjSAqH^xE*XLxQRXTV%%IxwpQ71l_sNX;QD6Zf?H1iu!{1@9WjTnnFO8{MeXsl^_(! z`Yv@owT7hGoN7VuBW7vL2@8QJ7CNe!aukXuUXuW6&ZpHR|>Z ze+;h=#UZ7GIsZF?H|_$QhTPbUus7?A>t3fW3UPn}51TYll*SPcV{31{^qwD9FTt(HD|Y7tu*O)xfTbvPN5f@2ifw ztx{}AI9}u5o>i@-tM&Q7fNS4nrf>T+^p>QAW5nwYVFL6wE8hDZzd~vJ*pVQ9u51|W zXqJ2WKJlN-;vwn#f5*$h0yy=^|F>sW7B9XKMBInn2Yhu)XPJ{1CQE>g5;E>wzxV?r zQU(who3le`t)J|%^kcSS>>v}Nm=wn!(z9XGx`iWk)(ewh*tueB*UaM+!2asRPkQ>= z@XB0D(a)&)I+DuX^eIn82MYs9#+(li9M;dI=M6quO7ofxptJQ92RyW9DayfY?PVVK zd^bWe6w@y=r}+29?(fN#G>p=E2<#A7k)_g?*Mn!-5IfDkk@PJLH7Wt1$q~4zNdZqK zu*I^o5&NA`F>l?4Kawp0hrkUtc(%bi*ZXv5Kr_E?o;;5DV^C0#AcI#u;6v4dkH)f< z+3PMafu=!9R~WY(BK*1-LV%c6{E?yPMc`7>hbA6#p?=nbt{A9&J*w>l1UGhSrd%$O zp~aiCY};E@GmHYGN@7A-*9z%bMa!wW%s}FGAT%lbVpfDgzQ>&aKq3*C5WP3+g23Av z2YiVC{XBW0_8#`teM1Xx=y7{XK;DmPS?+sEIknbJ3}3i~o0)fZXYM-q&V4v(&|~!C zgh4ZIG+pt4fM-=7r>`D&{Oo;%A`bAzIGS9LdA;w8z^!ZM@FrmyaB}m)IPn_ai|aCA zpL37Dj8a7iK|)1ILMcihgQ!`Jz9jLQ5~tJG=(H6^sARxKE0r0cC+fV-V>SqeyG@rS zvh0(vTCTUVas%5pAXVgVfhK$-HK^Ljpz2xxFD;zsl9EhtXuThu8T4%(wL-wxB0eO% zE_1pUuy^|z#hUfv5u<5cE<_s}n$w@!Um26rYero4m;1}eG?X?+P6J#9?rEU}#$gDB5;G1p+Uv%h%|OH2)}exH%3@BWTmRi zTRBs*HE|<%nb!n`uzQ5=wfkGeMX#HTp$HDr<#SYjRzX8^xE?k7RM*dN|ac1Ge^y zfk9+)7@+$u)4rQJ?@F9iSW;F-$YW}58T0ogzn)lUOqb|XqYgoHuL8B+V36@7%TW-Y zR{%ko^QZLPA)J@UF`h0fxCJ^$EdgjIr#;uwk<2iM8Q0$*9kBEnI-6bs?~@?D69qAq z*gR4xEqix1bCb>oMd+-O*BE;1=P@;7OrCuPoP#U8FKNToz8cS)1T z)EPGO{`)E7xpb8$WXG%-{&G?)8bI^p| zm}BT_7w%8;%62ZdF7`zAph>y z>F8a%1$9fBi;tliE668^=CbF@UjVk}O@l{8xCcfuv_ai;a=mn|@I6DmPhmoxY+;Mp zGDisklg5(ra>-?OvV53qn!NhR(NjYes1v%kDA?2Q)FZ=IKyK#8dS+yd9Vir{04O6W z#|9i8xQb*$~85gQS9pkz9;FBDh12MOjFEC8QU+r|H1c?|6=3<6_ zr*U0iP^=d*B5cQwEB^3EFoHBh6ZAFO)Y&pZ($at9(8P8^jkFV{N%z>50_fJq7PLpN za?_usV&`VyYdxv_NlN(}KcT-uyGZkRx68rtNZdqQkV4#Oh#t=R9bb%O{TB`vfjd4J z;#i@vtb)l9qqsONG9QW-#PCeJjJ{yHX#q~tRxmehTHm8#G3*A9v+{%lpNFp6jq zet+cVe@M3XD2rEhzUwP$>y1)3@A7--9C&@(SDE5c%{60-Zm~L|Af=ttF_zP&A_tdn z|FB7_jkd^|$!Aw0w!oO)5;yrf?cIpb31G}ElYI|Q<@FvX-fj}J^eqk!yhQeg>87Ct z0vQ=NfEt3K_zF3Xb2e8i!FBFiHs={z*&}@!94?)X(@>jYTn-Q^Rbgc!xjcF~SzCze zI6Je4V^P?@p1G_`cNGD$C|t5SHVp9a@KR7UVoaEzew)~2haZ?Nh`fClWs}Yn+FCJF z48&okhE`Bt+0!AaT8|sE*Iqi2Xgm8G3VKKUcgr=J(j(B$L_=Q2c01vJfx*kZ4Dllr6j!wuf0Uzyx?pQzIDLyQo_egyvB|l5#uULc&~Ww}sAk8nX|EgU5Qld$RW+ zU07*ax6{0|BnNBqgBy#xulZIMZ^Z45e=#&d- z_IXU?M8MdyXpx7gfKll+Ch})Ggc?GwczwL_T>A}KOB8BK?otpPzg%Q6(ubH%#l_WWh8+kYT87)ia zV7%=2I-gc-WTYX;R}xf&{|t`ZMMYu?r2aTHIPj!!SRvN(IM{qd&HP5oWkodIyR=N zso5-T+(|CSf!W)T4V`#_GPwx8vd3Uv+9V_PPHkN(Zp^IpHG7O~SJ7 zLcX;mySCsNd`?JuLOr=@q0+m@C3FxZ@BOhdtx1)8^1#L-BZK-dd;(vWQ_9Z*IUMNQ zDMHcFa$2xdJCjYDGR-jp38O_6vbVfvSQ{!E43Sw+$R}cg^$5IqWQJ76mN$)q1h15e z=n~r~MX@Nz`_5bil=qlf-d2lge3~F9p6aBt@DZyNyPAx4?(ngPQ`Yk%{%j4ddLJkpm%(Ro$vH)Z;HG^6by zNi-d2k9~5{sM8SqLGFTQFNmP~U>7c%8&SRl4ZtHjUg5GjUJ$OZ%e=3Au%w~d0;4!cBCvB$%`aYY{&%cRk;lGJp7~m3_?Jc0vM)A7Ou4)l2 zz6$>y^ld{Mi(mo7zhOiOBO#9aO(a4n%-_y3{VY+?S0U2tN%<)ePbK&@e3zp&j!&Hs z*V{(}@E-o0Xs39ntz!(vI8x0O(~)=GL~1Wg$thAy)sw(|B8|Qle&pSIH zTE&y}FVpmi+KK@biTMYv_JeRHlrscABtKtBw{s&MkiC2`=}TzI&?V&fudyPuQb8&2 z@S^E<2H~-UZ7)~mAV1#`fkmMCQjwB<8mjV8+PhCb1KewXSJEKRn4x9h*cpo+~E=@ZE=aUG5t{Ki==xeNo#zjXpGP zlwhG@25UEJhtF0v7qJIa#!M(dBh==ui*xd@n*X1K!iRBV(}4c;&v?A{m3#*!C%;WxYfjp@g1>nYizNZ z(i6a8rOSs)5I_LhY8{8jx*MjJ?E?PsPj|Fk21}K+EJh1n9a(RXXHc#k02WtiGV`!J z>`Ykpinz-<3hJUQteUrrRBw9@YN9tbFqyu>_SLal9;0)W(c2ZZNJ z9PEg4ZSugP7GsVQoqTy-UqBB3Sa6dA#=EP8$7+3LA+z=7RVN=JkX(TG5t#XO^`;ol zX}$wg6`;G{V(<4_(3i{4_V%0K&&eUe4iiI117m9BSU76bf~CSQzrij zg4Re@TbjU;e$ZfWxYPrwuYt(y^C1l)8*#OwErn$ue!weik6N^2vvwEihmS!KAzZhQ zJb9BsT709hy@HgkC~{#iy0kKRG(S$qH|Yfa)uA&-0lLDqsxsxUoT2R#_tMvb`Gm9a zQvflVaJjx#gGcm}^C8v9Fe$dVa_%)7vX2nKt_x4{#;~ ziKXgYHvYcoFsDvoT?X-w49VucO8!B0U3S@8eN1 ztm4QXoedG8%CTR?$#e*zxD*Xc4!$}v?}!Kv&OrpWFM&|iVyP{lCSd>=*ha9K4sHTE zQ$G*|NFcFq*<#rPGJjkNb+d(>{ifkw0WalqtA4j@Er4T8vGsP{C(biDFYlD}S}TLj z3UIW&#F>FES*^`@{!`X<8K%cc;M(sVbGEx>5-6J-2yC~`ok3-JU%M0XtNSR_FU|(O zZ{5Lkqwelvux)pU2_;oA;btACV}zjwa^S>#>X7yoquvyk>QOI-18#I!BJYU_<#eX> zw3148*e8g0hGn|(^6#}iN$*%lFYB2e0J;M9ns;Ep5Iq$f!j8F^LD0XOorz}lE-@%) z$6Bl>9I@lffU8XDk;kU$%Pr==FJzCS)aZLh9jXMAfjm3kfr%Ql|)M~y`)L=6aZA}IZ)}CAt2aP*h&5#a832S zl$+D#`@RP9zOh%O8(9J(kG>$quIIm%E@l&~>4{xk!ptq32vvq%)}*vkr@`3{2tto4;lAt~fW@IA$v=c@ z0#)eMRp*C&T6O})RQUBrl9!;lZYPR`p#_a5>z3()e4sJQFe&v+XET>$Cl0sfC}CS- zitz;f73)IHf9G#jVB;1*CH6yCv%L@xf4%aIo7mP%8^iJqUx~>(5n~Us-iLXY(y>GE zNfJanMIY`aB#+WV0G~UV;^NtIeN~vAfPN(C`a$_=!n@WHyCbP^%^!uG06}tPSEp5= zQin|{Fp>Fl(PCtJ;0@ghd2(y~zP^JvqHVeh`vZv5ry^HgrG)VQt3KYoEV-G}g`ZR> zLaf(9v_Iz-1)(A(dv^W*f;I*Qd4eZhwgGO>GYb(rwOd0bV4S-H3R6LQghAM6ap7(o|T@8EQ0Y%3^K9w=jC#Y@P=XQ?VH)hi`M)6yI*uk06ZOKUm#TTI+QZ zx|3;LRF?bkSx}m7owyjTgOmIZo8Bwi;~DzG``+eq)_+T7klZ1~a2OfsreYTs{%ciW zqt$BHQmVK~?Tlq8Qw@!-kD&;7#CgfM-P`Lav70lhgrfD~7261F1PM*JuQUXdyUgzXn^)5vi{&B9zb z;ud_rdb=+#t9p2g-W!qN$LMK#cN#GSkc!Id;M`UxC<8j@!+aGyKGR}Hq6lzwh3jx> zQK`F~U{{VpsW&&uZ8F&BVP8B3*lpdzGI-ZvIAFpHLE~Ty5;CR`x&MX<#xEgBioGzA zbw0YSzxng;_za)91}!p;oC75jUU?U^_+&xj5aWL9eKWhjS7oA!p9$Q?>hl?hPrDRW zsFvpDRigbqz9ud%ScBKsidUu+1Q11AXW}|J%TUj2F*(T;<-~7*ViMOAKaDEG~MhnPqKrd$;2;sulEa!*bkt^0-=4+bIHvx z0{8Uq>-FJskRo{?DTCm;+?X7@3sWyl#X8kzVdnqwhk(0;I(F z1BhLXtS@eMa#9aAOF(cO%q~m$+Dw9nhX-Y5UKhvy$_OYSP$*QS#-O7Fwz}q(mYP}! zRNYExFL(OEMywDJ5hrZCNW;os%3T1(`?Z1B9v)!UqL@^58)`#aW6?+W{ODSxH?k_W zuN=~)%(gA|b9dF7>TL&^yV(hmnEQ^DE~K<+N3{^YH8MWtHxHKSLU*sEiC)3ry{$M< z&sJ-fbLkY6Bl_1U{2j}f)3ZIf-a=1twKR!TO^9Xj!qEHa<7olUkzM24Kcd#y0H9vr zrPzKvn;Ywi@Tx1U`z+7}u~{EMXf#NPW#VrPw0!-r_hvXYka; zh1ux03*=Ji?lYaY{6Qv}s|0d|L3v{F0V#Mvhh)7$ta10?uGDu6$sN%n zrnqzuk9&V-#l=y!nBH~QDu;}`z<$M2Vo71YZ*nPMFKSuX) z%?jba(^}$OreC`-H9t&#TV>_sGz6_r%Gq2PTU%R%T(%hvOrE6KfFi?cYtQ#tt+YHa8!m>Yq4&EC}H~o|HT73u(u3lbNLR4e1E;1UF1k-=JfeUayhmxiD z-z4>XJlya{Lvh!O{I%j^>@+GWGbSIHK+r|!ILX`nK}&S<*QrmsT7ajbB_LuK6j1&Q z`07>U1$0=Jw*w%WpGu1q9LJnnk4Z70pO!%9Su%XlxTIpd_HX5uDVEnfhd4xpc~*9) zw7c%&#)cn|YcBLW%_rGhvcBG-m756ZT5bddK2f5dApY5j0OoOj)eNxg=q3iIVuh82;&*oElfgj%EUN1&si!F8AtWg@BCDb=rNq+eO@GK8w?4JmUa} zSPueMk;6ZBJPa2c)C?DMINlck?6uh!HTh2Ph;qm@w9fIV#J#!0e36{~0vj`pH-jr- zk1JFcx$pNt&#v=U(^qKo*2?Ucl_x#bJzUD6enioc=IQI)g@wK)@m(U$Pv4ma5`)nl zzIQdmCZh*=BlhK7KM+ixyl(sUt#3tc%ffmiW>4>1on1#!w^A>)H{uLU6{Qrsm~c7X%T+sLI;(IgL;9d)CorBY z7ZEwUgOQ=0Tp%cD_OXOv`gtidrb7__e?nBs;I2rF0K{P(P|Im9&IDH_^8o!3WaxK| zr^J840hFmA9SaXBanzrYPu%Wqfs+ItswVD+&zJyymnL<`TiopI|x9TWglBRaj8JC~MB zt!bYhYhNEKlN*d=Pf42Xt|{g@%fj|hW-q-rBC)W3X8YbK%S>D1+1FZ@88S*agC`qI zjdhUo%yvQe21BjcMQPcnP+CZ9oubZ=*TBuz z(L?Hfm<#p#r8Z=$`{^0*7}(M+JDi|RAwOU+m%L~!<983hz0%``i}Zgf_Z#>QcLtEL zfSc7HZkb<%S#qfzKLQq=zOi8G6`UB~JY!}>S&V?*lF#J? zYh}=FI(|@dLliCmtV%zc?X|s(qq>;RNXxYo-~F>{J*A~Ic}ktWG@gsQn+Dj+0+Y#m zMNCQCNXgpdxhts?qzi#*QU=i9_;mv*5&u~e9DlxSwj5EG=nf+KgOmQLDnh|^NMd;Mq z?^0P)AurVR^>xo4$yWY*0s~|cl~cS>>J3b20y}@t-(OW} zvM?<}KRg0G$DoGqTYyC(vV2$_L-X5(US{D5TzxvC`N#S0Y{&9Bfu`wSp+nIEt@0&2 zBQDADDQq05lfR&{h|T^C2}JI2O5#F&*=!O7=#oO1bG)1Jq;Fs@Id}mj$;o8w@#%h~ z?LRRLP-ff#2%dJ*0oL)z`x1h7_h(F|0PrH4erJPTx&$w(!lP_b#L9O+EC6c6<`?wr zu~L$Ijm@sZl(U(k;_J)2Q0rMX7dB;=&5dBsPZ>)^XufSQg&hlQKUF%&q%I04rdL19 zua+zU%!TaShuuOguIv`;eax%5t$Up=gRDue-#WWoqKB>G4-_5QcR5sxD3~exX;!z4 zJ}=aW{}$Yt!4Tg7ws{Vi9lR`fyJ|{12D|~wiE7%wH9+98?7X%RPQx(i0GQ!$QG=23 z0L$GtnsGJ7^TY!TaGk{hBK?Z3N00TJu8nmgr97bf0UZ$q+>NU&P^Xn->2OBsdOSbj zd7)yvykPkYKis9Tz!j*i&NwqNdl3KwQKNLzz>nhQLaG761S|26=40b2l^(wMxD<3-9~=Fj;+!EWWNyszIQGSmg{ z=a3X+LebV`#aMfR)^=D8)!$ENMoeOW$Rq?EVPTo|+x>@XcFM6yEL(wr-#=%rXy+)GY-B zpzj+Mt@?F4yqj=bwO_R6WEvVRzUc!1q-g5{WTh2=G9l$y&)LKQQK098-DGsGLcK1` ziH*!^7r53|kMN6#W;(K|0T79gg|ahYVpxJsgg&|G@ORu4t@(B5UbcQ6@OR|@gDsra zZ6){4Z&$=TNBuK!$6)6FqxeR0bYd*-tUxb^^%3>)pf~&c0&ig}(I7cT5Fkd{%>D;n zSY45A*u;anMCQ%XW5O@zGF?{x`-AWKir_QZRfem{fzJ8c_O5YmcOO13LS~Z~QlQvN ze6L}73@xe4V=@i?k%RGX%f7H;>@gq5!?g<7n@k`%>l7w;x%C6nJD}!;j|2qK;;?o$ zXDOAOEDyjoNVKGs1F^vvBlBmeo^or(Dd^#8Mj3C2zgUO5iJCI@(axN;J5sSakS7mo z84tqVW(e^Us}tPxIk@nd&K~ER%&-JdnwDLP6}nDDEIy9Czw;jy@f9@X>AZi7zu~JN z@8BQb`~9c)RvT}`X4~7ct+rGGQ~&+e#gDwJm|0=YzaO|A82U}M-j))_Y^pvUt8 z*T?%kmW-#q{EXWb2Bw_Ma7lFeI$p_6VNNmBo0$9ic@j#&wP)IAq37!G)fFP0g`qak zI)@DFQ<8G#Gexh1c+P`h>OI2!(7wdk^w<<_gR?!AFLO$szD4qk+Eu(p+&1{Fj z?;9xV0UI}TV9Nt)hSBhfV|texam*kzA91X`;)FD>ALCOy+qN!1`MGr^s~$;**CEm1 zBR?4J{iiu7ssG)Ul%tcKWnrN6b`%}gFi8NwzPH#rw#@enPY`}yjC`-h`3z9xN1p8H z1_9JTdnloB&b<$m&2@H>`ZTLz=m3;03<)B$(s)$iy6sd_2E*&)-{!i+T*5h`wn zdo~C=yL>lrJtPznG5G_%5WYH<Bg7Md{3#Ph?bQA7RsQVmotxd+*# z!yyHJCmq25_3Z|K!*sd7>ClhA2N8M-+wz=A(Kf!H-ZCRe?dO+4>0oa)uAYCySSEOx zcan#M+t^4f`?hfHThXslhve;vtb1o(+a(jzhA>xeVjVMwPmN1UiOXro4X${9?i7FNA>0%J_ts5dhp_kXs<@MUn>)JY@vU4uM4Fd&&EQ!lmAsTC>9i+pyuhXeHhsH zNUpF+PGoKH35nX9sN~CXf2%xhp^_lWNG=IQ-NcIcW%A?3H zNh%twdSllJHYiWgQhg4QcxzKakTV zNTDJoc72d<`QyG6Dg_uY&-5?Q1P9Ya*c5l)T`zPia(^tWYX)LnB2#*K*}%osQaN7; zLsWDN2tZ|6i;vfc*Nt|loE5^2`470Q=^OC=>GA~M{)~zEp6ktYW`Akb!Zjt<*F+yD z`xUHo*tA2QH=eqhss8=wC-&Y&CpvCxjef zIsXSgUBs-Bwk2V8+=6&tlNU<0qn#~X93+E2(cqwUR)lBf3Fsqxkr*;$RpqJp`AGnqDy>EcTox|CiFj1lmOL`~{9UAuA` zV!2G8^sq|U1F_ditAdCXTPOhoiNfzV8M|ylGhfKsFZM zzWTrxW$q}@9{hRp!&?fKv-Ovv;=$-#D+7|G$EqsD!B@hly9?^POXH)OU*4RWwo z1m!2gdFD@CWPX(t)h3zuVQh@e#T#e#=hxUCuQ5%Bo%06ONvweP0sN9IeDf-YPQNoC zId{62lZ|3nhGUmw2mJ0D#@9!-y#JohZjJ*lrk`J(vV#2UI7)PEI67E8JE31@+RAH- z5Tg;K3wf-da--rC_q(?=jW!&?`rF-oBMRnGdY870`>eYr%{h&K1}=*CO4#*u3luoS z_=-xtlKDT-Ak?()C#3(gnL6S7{aPDUtJ-W5&WCJq0Eh-0+v;zQ8WF25DgPQ$nM*l4 zR;j%^V$`Mhgm~QNbpTHzwRp1?Rr@^47~d6jy)?fg_glKeDTuK?ISx8*GvOcp^m+=h z5~GIYpkQPKjA)kMrp8ta+w#AG2dDorPH!0KzUkKP%=u&Ed$S*LzT_pyQ(^XceIE0& z>!pT@mdaoEt^<`g8p_7T_QkFZ)ZwMQ`n?rq%ELiCcFY)h=I5DE>Va)q_R}Q-%~)CR zT7xK^mRH(|6p!k*8pmRGa%%6u>ABmxd*Sxz*VwDvaTZyuVt{wH@X$HoXvfli7VITV zY)34sVe>8Ui82Fqm9}C|k-&=uRIBFK(mS%L89>33GeV zDXgt54SR1leJ-Bp9a;aP2uEyKD#OwkJjeRKe_ASzWX_Hat%XP+9gM~7aeIbQ(%os> zD*I!;QsR)(eKuHda2GsysE;1#U&MP14L1u+y$fDU~_{lXr&37lvh~I$DaeYSJ&6Qf$-gDTOw`l z`=$F$$#lA+Ro+NAn5!+XLxYvR+n>yQuOS$WUptBFh7G%ny9{jfxk#S(cAN>e28Qkk zO={`q_gr=_BtK1L68EIdurlmz@5+ab>KIWFMbC6c#rb-9|CYoO@e)yf6nc-1CG_eM zP|{>L?{x20tm(Y}ux+;N*e+HtZ;#{>}#PS>OCwyY_de^{XZKzsdZTJfWz#k^Y4F zfDOuo2-8p3PRYnXhhm9-mKeTD{~-)E^fg|2&4PB!v%syQV6MC6l6G_=Z%sOv4EXz% z$xEBmlLP0lRV%niS(V8><~!i4eC6ze0pV8qTO z@DrfXL6d@nFR>!7xm=86z~zkQ!KZ6?excq#5%bZ=PG&hBa3RbScmRUOI&gN&$q^`q zwB#k-xH9M)#_Wf`QSMN9Q5*H4Cezw-@NeG^c&AN|S@a+P4woV@!ER{8Pku90XJ^o! zm*o=7G!%M|*Tc(@3yVl$ZB z_WSLjvEe~|`sMW_Ifsz?ow4w0VaQW<_ z0(asjk^P|I0|r(#K859dy=?SdBypY9UVIeyJ}oR%szNcPxh%_WV2KB29IFi)_}ssZ z;s@&^7fWcb;^Tw9=YJ~`GQf+x(2365>M960P$ZnT=EOW|goONH>e`{A|@YY{$(Rh%)p6m5ToNVYpD;LuAn!nj^WZ@Xcc) zgc!Goh5j5%RC||yZL(VooZhgKsT3^c#IDj@mnNM?jiNgzkrIuP%{M72sN4UZu`(XB z@z8IPkJYKyf-zkZlMwl}PeUvoxh^>6>CJ`q;Q+)#ayRb^mz z%o)b8zpL|OZr*8L{zm+Y!!ad@Ek;s)~dnA0o1ypPRwJmxZ(@m9lUU~@SvmF1w zGURN^XoE`Wk;$FA1-w>`3Au09O~V+wS$i&MJfF{ztBs>x2>4D_*%C)i33ie|6qclj zhUZv!IhA|hjH}4mQ7JEkiig}Kv3CjFXzSkPm>;UNsjZt|?$cllbNsl0Ik=pj3AP~Q zQ-?hq{XIg__U9t_bbieQ5pMC$-v~1u*`%a|mXM0RZP+xd>+vphbfYoUZ6p#xmdD1{ zHh11R^ZN+h*I;=BdyohUr;FiY>^_41@rB>N87C;j33?R+64h3K^x`JtKvpn2MdQ&E zH}~|zjFu=?-r>}p_+b;|$K7Sb$;9eJ?UuhiB;>_fqy8sy8(9II9->7YTA`TZYd?i5 zAbk-jCjBx-S3TW55yP$IEPL1vq#RI!KIvWKUSswen9UUO*6p4H?AMq90mzEOR-L1m zRZ29%xEy+P#`CP{k&P60#E}JbGNL03&m+%A-Cx>08EtzDl?@*-XH6GO`dv}~J>fR` zdnI@b*@!8ev8pYzL_03z9}m#jpEnDlu4t^NH^ocj);;A=-xDW{9>Wl>0ADRek^NBFI=rjxRL*&&=bm|;xIzLE zVeWtc&V9x}uD4Yz;M?KnFt@1d`0L(Z%5?^_=$7$%{GdsQJN~=+>tQ^AlbGt3k5F!- zG_ufO(2%Fjvkx;D7N42iD;@d^3&PX_?mp4=4)a~sVmPqZo;((cPajZhZa>mg1K=IE zXfvTEb(EjFopoUWfVmYq3O_h8_Z8X{$jg)An(O63nDq@hu|zdZNV*C?J~h+GgFo6) z-}Q7If#Ts!&FY)Qn1X|Gq<1oTz#JkFS8#P&Aeu)`!jU&?CFek19a%OlY6`{ZyvlO& zvYCd$*z?p@x8(kbV}?Nx4+CT`VdPyB}b#sNa%(CQc@Pw zkj_ggeVH5jY|@IIy_wqZ%0QB2>qDJx*qIcX*wc4vAv2=aV(QowKX6B8mJXMNu%7M8;V?;)9F zF^?ow`leJdPYoV`2z~XjD+7H%zubvvH{h-ZBM zil$mTXXR*}i7fqmHIvxrGO4X$2U*1!x6#3@`LH%0Nv6Cd6}LHh^dA2o>K3xECRkF| zcb+Md@zRo>6dL#Pdv3#1P3w$Ru%@tsplwmWO6KUMP ziX3hcw7nj02=o(;bh{yM?x4RRWftCdyLU3(MSXw+p!mF6a%+jnE(J$OeNk-x)IG88 z$%}(9>Bgdf_O>f-%HLL$7Dx)zYk^3-Q77<*Lu3&tBcXz(LULOE(TeKp{hu9#`-5Sf zLs`Q9(n~jzXUwahnpr)KI`R(>g6&ce@54Hgx_ELhBpCWb^|8Nh{oJxCUC6p(P04pO}xK)0TKtl#Uh6e&#AG@{$!`p*mfqix#AGS zy$(knKnk$ngVGI~SM1v+Zjjx=4$_M#Y8WL&JPO0fRj>N*uhcg?ZUYxeKVBp~y+#%R z2~9E91Aos+X}2Dja}yUhe4`nIsZ9f=~v0o0!5$u|j{hfQkT;cd16Pcp{TYKHwB( z`CN(x7hr%?<*UTPg_NYlq~YvrN`J^d<@^` zKU8J>sxcb>=+<2nFgoIeTwbv4R^X4VT;TGW2$46Ct@y_hEZwy#=g-f;NPnM5ffX)1 zlcSQ?zJHzQ{h!zf8omI=Hd&jql>_v%Yny-q+e?8rDGI?0^9kQFS}U2)bA}xll~U^! z^A1LuO<>`rI`i*-=pzfS#XI$;lALpZFV|)H!(X&2*tD)5q~c ztVqNG4sHpgH}vn#g=BE5>+sNMbR z#lK_#GmVHt(b=63VvnpTjAi=_k+~J-n_*?Ca$)(^rX=Wb|T;ibf%Cb3l}P) zN*%QIR-4uy#g>X7HNhz1DEQ&CI-Pg~^65{M(%YI!J>xj#_J%gmN_LvXKi*ZDUlczA zJ%GcZw`vst2^%ItX;J>@$2)zmz^mc!x-ZtMY4hh;uXijcoGxu6jtU8J=g*rpW#+8! z3Ba)%+Mu-#?%Vlyobb=j<&Dt&iP0l_nXm=(r-&u=-m`;o5X9o3@4Fd#w7S+j18c? z$mf7^ngz87p7#~{q^N}UmlQoDV z&&d~KtB4J@qDu|ZIWMB+w~*c;U7@MXu*6NdSfX}BX|k*^n)tIpDz{?JK6 zo-Z`sSk?*5Ps>yl;)K1u9XI9K5C}kYa161ffzTYb4nDr~H4v=QvomTlEegOjvCpzI z^$j3+^bM!TrYcJ<8te_Dy2!-I5Z)N`EIqqiqRkA6+(wba@PLxmDZbAsVx+*R;qN%l zunA9Ky7G4>P7pXyu3y|ZnH{R7S0!L0+e3}a88@K+@T{fQq;`c~Dw&ie$zsSGe3w$fG3}MW;3)9Amx;wvET*EZLMEbDD zil813PLz9JwW*TEj9A{uP8HKEHn*?*hoRl%ezJ%CKAH^q4JpG$<~j*`5N=C-F0ZoH zTxn9KlsU1zHPoW8C%tl=z8{-dBwXh1lHUH*?C3j5`jD95*;iNODsT?};A~gtF-H7Qd+4-j zpQQ-W-9;0={25TANJk9QttwA#(;;72_8v_wfOyI^l={mJoTs|?O*`AFlm2m$B-kj{ zK4QJpIbo=;QDHD&*Yx_Pp<-AO$w6I9?R*X0a5)g56kiVV>3X-QhD!iu`6k`d}v# zIi)FqWG}9*0f%79H(H9$RkJsSSW?jp%Sfnx3ZAhuqV^`A!(98@Bz};$bvXC#6&e41 z=3hl)_Pie%G}>zIg|#DGeXF7nq(vd^JMNAd!mLE{Hv->JxBJ#*67R7DrS5>_4n!52 zcGf|!*Wp6{u%qVa+KA9F222dI48Z-yS(&_U&LZFqujdQdtf!A?E88Rb9kU5 z@3YCuuzKJGUot+w^=__yOs3|GoMl@il)_y7+sUbB z32~PWb$9`u&Ja+klmOVa+Z7)TO1+>-@WbdV?STOLWD9_fB|m8Op{DUVUn-@di4e5P z{%x}_=_Hi^BY#dv6&*!tz1Peha>Bv=E;(}dedo7(e9xJJTyakwa!o}1JAn|O&g!S; zqLP5A(z2Klg(M3(GwW5_bE(9>I~d0Riif8QX3oR{E9PL# z!Lmz?vIQe}UDj}+s z$=y$Q;t&Q!tn;5J)gq=NTKAL{w)N3>pQQ(#t=|-9;#%DZ2LkBOuVq1{w>_cwm*@!I zdKh4Dg-$vQtQ#NaUjlKB;I6@vYv7;W{_885f|<03->I9J7+(S&NE~M*Z<6jcGTw*f z@wb<8Yb>67@?f8MQ?cIo87;}d?(2iwtI`HllM6Ml5tJYEGMtx{OG zqG@*&)9u?4#*d6)4Sg9EKkFM!cAPpspsvZeUi;RzyP2EpSG|9gOoBh4wrt{*6m^tTh<&iDz&Zwh7y5k1fK^*ZTfJmFde6hM>b?n9CI~ z_M}^0RUE%`x!-H1Fj>zzy5wKhrk|dfuQAnmF%efF>R}JA#-Er4_NvHpio(v`w8Fc< zXJ(=T7TiC``>}E4Oi!cQjDKqXZTiLr*S#41IdA|ioU^ZVmv(iI&&vP-sef7GVW!XM zB5TeRSuSMj6Cq|Aim)-A_MSNx$9y|U5Qamq=N-}*+^0oBk*Szn{-2k@g#B{I!6Qpuje4e*XczUlc^C2}I;-Dc%Pk$+$noi*i3a)jwdzHp?DJdJ}QupyT zq*TmtdVR~R;3tvDfr~9P!F>oZN?JCI*b-S4WGLX#wj9%riY#CKe3ejZH-S)vuvyY5qJbgh4Fy_`;| zf?=e@cA12^mHYQRdV@w+5{|bfln_@A>9ClmkBNdSy$Vz4-o}4we&2a&b+1&4VOEJ4 zP;tEPI5X6A@=`SQkxj=l(WBkbY&GZfR@(ko* zMJ;Uh2}*@B{kQK6Mf7S?<%S={z<9m47R+z_$lhMOTA7n3z(ZPB2}s7iWOfU&<|kB--Ln05s>^kN&77)coaJdim&Jk6Vs; zkz5k}Y>UW9yaS6gvz8p^xwz=_HTDI&e>`xrYfsj6XWGDyZ2OYXzrr--MWB%Xjald4tKt)4%J0pWdoV#I-Hv@#N4OOowpI)(XEk{`tlsS21B zmWw%stqVYO4(#4p1 zth~E>OG@qqUhQ+9jFr&Bj!(%Mc_zV(U?Rz$ACcnGV-JIAgJ6&H|1tevz9jOYXIh3; z815U~u^=vXaoI0;-T>7HM5x~0%lqGb;wT>YzhbtUvGoPr&4`qO5!5C8*`&EEOwdllpV}w;{}EB0^)6y5gUZ)1*&?JcaPDUV8w>!=8q{J5m(I^*n9|4FLc)M!F%ka)<8 zJ*ItN@KXYaFI9B_=-CGUtSoi}dfI;VdF?hN$9o@6&9u={EGfc3U%{x0g>X5Jsuh#B ze?RX3zeM1xOc8|2HO2fNUuPK>Ww-Wy6$wSUr9(nNK%`5Q5(xpNJBFB{TS^e5K_qS( zrMres>6!tlfguE>8M+1-2HuPP?7g4o)B6E94q~|0thLT{*82az(Di-kD}!r1`mAy* zfKa)$us{_dcIr`SGJ)( z>uy#JSv(3gR%9Myre&FHF=7dvCIv@6Q>dZo6ty9~7a#Mm@;%Q3?#gII1)*p@?#TC( zKV354?oCVtWib4sr-j!6TO>_867wrLBBQ1XCT^27UoywHyoz5I+v>V2cAe8nEAx^h zlf0hAeZ%m;;55XT5lTZ+(EN?Z>3Fw6q+y^dbfoX``L*w}6V*)nH_bd*_@U{j$_8;= zWi@f_%}(v7D{h|-j9dpzkK6vU4|BL5inQXW$}*;4NmwesitSN7@nqdQ(GK!?LB=RNy{RDK}zo-7O{9J4IE1L#~E^xLZ zH*Dxy_jBXe`a#EgO}xAzaLq<{Aq4;Fa+pKDr6730Sv5W3bJDZS6=D|Ea6mRlz^CRk z@!~I`8S=$HqH(f*I5_dNhM(VH;J?KG;rN?@BIz?%3j3j}K}R1t%v$ZL{uN6SPRk-Z z$9(W0ft**;zUuEZSwLlBR*l8zwO-Y)Pp9uW-b<0JubF41;4|Nkm2F=abg%kA4`f^# zG;<$|&Zqp31-dmt4~w|(MQVM%I(>zBZMHY{Yb13<{io3xe=CmOO#uUxaljBa(R$AR zXJ!x}Pkt_l2_RGctA{s)fLA5W7Of-kN&d$1k)5ByEi}1<*V&_A2#gQkVHlW^+uV8)T|$bYn_px zh~CsM@W3|!v=X=y7q}V?TxwlPHvC+eoy<8o{V7SXL0j#Q%)c!)P%WbG8zxbfVcK-Z zd#_Ktb55jp*a%OD8TAh{k54se;(d7%oWK3zbETg@EBly= zYG-2HZtW$K@Kb~B-~ou47XLhSw!>IAp-b^Iq;l{Dyj8ksI4D}Y2&UEz^NCMsYUO(I z8z(!E=F8v<&Q!KFpqp_c3);8G-_di^;KgL7MQKPcGyL6cg!FazL;3t65TBQlA;ur#cuG)>KC0Roc- ztz-+(`xhdQkRT$7G!L<r9u4u%lR#T3ANgmYP#J6z%T{v-Y z^0mB14!`B8)Ubg_=hI%DCfS#m_rM>2;y6mwT@SE4!=}ox^>5q<*Onti;uM8s&wcm) z*~3mOU}Sq`Nek_zZLB^ttUasQxyf1_{CZuR&u47BSW|P=CCZa3F}0I@^DpO+}gGeRGQ0~f)B=^tmpbt+$`pLTFF^Ttb?FSpO~@fChz zriVJyV3y{XM7-YO3>##V@SVa)b2Pt_#N{}VS|1Ve{85aJMj*`{7}?nG%oSL_gO=Av zA{6KZ0GquzIasEdYisT&|D)!f_3liynF!s+ue9f<+r6fwYY+2nDgHg6cqas}Q2I#@ zvgYmBB7t=Dv%>E;*EBa;LBGPFo~hQA`et>O6Xi!caWHD^UabC|!f0R-$KXXstv$41 zT`h~1rw?K3@=k|XA`TxxMZ(fPmOCNN2+-)Qz0-2Gj}}MIKteLnGW{FRrm8~`3yScC zv4RnY)i?ujNtd*@_R57Kjv2@@{X?4Z&)=YCLCBs6kn9cQtd|A^(2E|(vo6s0+9P<$ zeLVjl%Y4zcyO}i{I~?CL86SxFU*@SWqP*q$w-h>4|c=O*W;x)SDg z+FXF;_N)e015TT=wblVs-27@V@k(=WeSd`&b085y`LyWo5`(7F(?3>#sWI;y%wke; zqn;Xiww@kS)P2^gcDFl0x>qSCFv~-@IVXIhsF3xgZfn?J5 z)Czy1iykyo;UZ!?=)+$%Z-+St_=s5UPNdHNjt5U?L?r#>J8}Tbc9^ej2vxU9J8hjR zdn*gwDcQs7!UI|cA)N!@F*W%3h-5?#DAp!weckN)3dsf4V+A z_55`g^pBB+h*db!JS=s$tpVANNn`XaL)J~F5qZ|n$4VyOxKaIQ^5mZ{f$t(lJJI8q zyHFiWn#RP}n|x16X!*4r;&RFMD{s7L(z)vG$Re8jr>#HM>2*864E7ulKf;jBRG@cU zX?*FBdd@5wQv%i=#Ta1fR9)|Dn4Nan%C+ON=Y$2uAT;e8RN{|brHROAnY88DrVEcS z4HDfaM2s#A$(M4}*lH%`#AJuF5Xdq7Q|CF;>?vHT%ir#X`@>~Ld9!J^UU{61 zFWRz<2Ypr%&gN}D6}iv*pE(MTd?1sWj7jn6_yd8ygB$G9lqWI#S{^rh2Q0&s(?&$u zun({g`W9jQ15$#}ryg*7ub1G~O#O*#kcUJs*C`@=pDTpa!cvhW7@e47#1?|#aKi;W^_W-wLZp)z~0Y#4Weoe^PklIQ<1NkmRE*sNt zBaq)exTWdIWG~+0=w`|NV$-sdw6Tm&R{z&(y8T2?rsgFu2MO2T5Q0A~Og(%{oebv} z1qb?FLU?`TYDvjn)y<{}&msNyV?iF0Y1X$Y)>) znQGmXsqrkHbISKM1qUY4a2$Pjx$aL9x8%ZD?WzXloJ*aW5t7>E-Peok6!&iwB8Mz%Z*&Oe4dY8f+LaATkI?sM zPmM#ULX1Pwr-0$9iW(k;A-m#1$L+%D(!AuZ!Ft`&W>ZE`JE!y*%s|z#jJqn6aQnF(mqzxU&!>FX* z&Nk3*$=0&;k8)GFQGZ#d4#=ZL-a_MH_&PH1MtMiB$jpJRCvCwq%ESgvr&S^B@c4nG zWnT(*d5;Hb(@(O*;!odio?X=!_itkXOfNTXii<494Qj0oTMw`nIrV^E$RbUlSD~b9 z_Yk`Mg_0}m&B%EPfsb-JH;!(dG_Yr+Z zSWHJkNE$;(9`2sE5UldF_lWtpJ?@3Pn(ruNe7NG#jqBnKh=VucMU;G2WY_k^p80h? za!CJBpK%wSb`o_|zkUx4`(_ZMoOwJd+Ol23-E0HggQrjz08b7;PHeqMQYp%ty01S* zy4;KWKZD>vPQkYwnDSk}?ss8SFdKUuA;Aa=KIh^TWbG^g* z^#hnS@+&u!U%UL-@99!u7yq^x{V6<9(o>~u*0d@WX=s6DY8~ThefPX)h<-~hTtu4F zToO_C*3>JyZ*1eUC1L^QU*DnMZ8G-BRM?<#rB_?H+?ZiO3Bel?l6QwLZ-7I1Ery%i z_K-P#=}}UNKCEV;pjIhALO$!E9Pc)S)i3r0qUI%1n$dxKu4fshoTVlz44xWVa59p` zE{EBs@9jMyc-5-pKH7<&e!4?<>h|A-Nf0lmjH`4hGJYyLizxeaZ0jH6BdZv@8rq=< zh7ho-MCh$%KM8&OnkYDCKhD&l(Z|I~^|ieE6X;MKyhPZ|zZr@eZG%^Kgx>k`x%;sh zM0LAzL59^$JNe=y=CnCg*b^P})ldNp72RZd@C$O<5RQN-v8ce#R~4Ah#^JJl^@C?) zhv^5#hx?9QkIm9kKNd6Je6yVLgDP|=plJi2JEmhpk-^!#;6d)vG@-?sn7ST$JJ$HZ z*Gm#^L|!V|D^Pd>@KTGDSAfvgVjV#XHSln3j*T{u=D>z4puB^|&# z=_}a?Rjm_a2>(0$@E5NE4*~8O!B8tKDkwXgUF#qu<1VX|Qf{8yhL|MHnPC-X+xkt8 znVz)<9S4MKm8tp9L59GG?VFdYN5%03ADWWq$QB*0+*sN0JL}i}fXYb(=2n+B(I71R zv4f>fhcBhWm&FFHsZFUX8t);Agn++bC(+?4#)W zg(VM=Im*X<23Av#6*Q&gNk=YWN@pL^PUELcC1*>@GM4rv3Fk+e_Y?74r*$bCAF%Ab z+@>h52{nU1_PR;6sUJF>7|DcRHokFaDc7$w(G1Zz-9BWzbt(R5(DKOO=;vZKb4*Ou zKE(#I23G!myg-D35gFC5!~cHaX3(mtuI7y8dPW|w>GrN)>E5#(aFPK?*ib=JSbm^( zcCic@t0_ywGPyKejvVKe45Uj_Vd&u+iesUC^~k=uP(NKxb5bLdSo2|%9_(p(mL-wJ zO8M-gka7)}4qtSjav}%0&r@>u=91H#_bvCnt?Wr6E3zG{6=2PtF*#mH;6)%@>?@jB7mnOx{u2eJahhkj8y{Uj~Ow4qf03H%RLj&fDN$oE5#Tf0E|RU|UR z242w0l+XLl$S{nXk#^>x`Z%qu_f!(hStSt8BG6ZfDi7OKcgoON8qYhZQX~aGPgJsA^%p>KeWtIR$e7T z9R|JHWBM?$U=zMW_Ma6HLb^uzqU~H*q`A-3xyq9uRtGVff^NfLHN*`<30BJ@ZpgW8 z@{u`c?cUcl`Ev#W_v)|l+~njwJ)goi|Id?9;burbAkO}H z3`sWcg4~guc$2 z@{Bn)RdG&!i*$~`5pgB}1xYGx4Tn^AsMPn>@NNZkrJ0q{>$DYLOvVoz8Ebf4^wCHv)}4{(O1 zmK#znYKA`~-kM5Qi(pgXY%>$Nc7fX6#Dx65Vd4Y2?3Q^+d>6NJcSzrgH0 zf@*rDtd+%@aOnoZsX>5s2RCUX$~?eFCrfBa2{vwao{s+9nXW6SPyJxk`-M z{`S4xfE^7Sku0Y?Gb(m!nKfErTL;so&&kAo`jx%;&{nPA2G7N4M70@VJ>B;!UC}vq zAh)kgXBQv0<{k#gGyQk@6RfHXtIPiUgf~sL_alZEn^L2^(iN2!b0we%P3l zws`Cer^Y;1^f5+Tul>>T<~aJd5$aJ?<)wx`d6zr`(8nMEK1;-NM=u@FI4?Iy>PVCZ zBg(3q+%{$b1+8USS%9sUE*schZd{6-eoYWhXR(*vf{c%*>ATI9s*OvQ2m@Pl(*v(5r0XEkj ze7z3KJyB4{0_CpI^44bEcyJ9blcmVUGu^;sSU zcQV^mloN;L8N2Nm5%dC0N{sIeWRRJb@lceJfEFqvOz)#sO9F|;xFnR7Sm%ASHg(kqI{&%fK8s_@kS3k+ zI_S7s()2UcL40Q~!OaQ!3GG0d1AMo_t-dRRCOg;kcicoBbLFuNraDkuf$cxOCZvbg zq;)1M@Y;v|rR}#<;I3Oe5=1Y#m6x!Cs_du-$T2-=6@Z(5;|QPtpr`wbi=WPR>;RRA zsuqCr9wiOf!xwVN`cs!6X~&m9q6gFFnHc<#O8~l6*lvD?-5TrKNGeUMlkF_q?|@4* zPp9pXiW|~H=wy?>e0V38KrwyBLoNvQ7+!GERf?TMzJa$(3X|Wx7tXcV=*o8`JR9b& zF1{y}FT4`a%L`>T_9v3$aIIixDC62w&u?AKwma)(0UY6EgE2O(sb!vc0@49n#^(K3 zs?kmRCu8zAE~h>frPEu7XvT@fs9FDu4-IiGeXcr+U8?3kpX*uLoY4T5GNMwD%1KnK zjMYNN$x+8i2%7AhCOkh2DJX3iL(6-Fa>o-)^$50mdYrGt$159@Bg6YZG^uJ#lG1~MZn^_}h<1AO2JapiWhlYgIp0u!GpQ*d^ zb9oN(VjWsms`z8niZOLgIbX9c&NIY5qh=x|_%gtOM%ZEP)K-R3Mjof15 z00Or%kXXbkGU|p*tiV4l*y-fAU*N7OEjXSXTgPjsoM;;w*&$BvK}$RquBUSAl$?Vp z#whe;cwB4MKk>MJUGaWSt`Ee%ifgl0KadjW3>K|>omL2&vCCHkTCDazbAA1zN^{%} z#vVV3!HftK!jqf#O+}tXe!jREACQX?O5=|6S!_JYB7N{m55OEvg^BvSt*-d8T^ zh8i(m^Zz4}XZ$Xgc)@G(&tw@ICw?Vs(`4v$VXI<9F(n40%R#`Kz2{jDslT8CWsb&H zAcLzwK_3kThtkTf5j9-0GzvA&G324Wvs&RyceN_&#Cyu7{s2)}16;S~=(q3a9sua_$UV;jr1 zF`2@pxtGiVE*~^jm?F15)`E-nd>257v+$yuyb#n22r)Ac`U}dGg6${yvf~KNp24Wi z%g0@`V}`YhHG`TxW6W`Vi_zVSts|J7{_D_4&S9zlo|hKvYk-t)-K|VdnuxZFRCIha~ z0!Y-&i6+2*8X%oYPpK211kx9<5@8X;m)#BtaDMK^H9x3lf7%p;aS<7BEToc@NIUhJ zBBLnji6yLZTIe`;V6au9y-&mZl^L5s5s>6aXR?zaH#@t51$6X6CXLspsXOWBFL04B5;bgh^N*WAr;_P>nLs0Fbn~2_L?JR;X=_Yx1<%HFt{Y~+NSdi__r0Fj z={b(a<|_ybR;Pg#RE~D_XqY#aWL7Nh?US~xcmBZf{OG!%Ujg;^K7ODh=|bnJ(*3^Z zxU%gW<}#C!Y#qBTJ!sONLB5H-!{bb+eEUNZ@y3EJ90Ftcv#O=Qe>y?JX)Q8>JO(ZH zlbL(8ZpOYH79#66^QSUWCBWkX2@r7!MZ21IV&fY)lsIlQkdc~Z(mHB^J-%YAe$PaZ zpWNOd6P%kpzdoC%6Sav|!)t2x?AV~$POF#O9P*q#96uZrh98pQ9|w;neElv`p*06- zZwbU6I&evxgOO_DsMddCBJLNtiZM07bzxtZ0;k1xopJ+V@b4c% zaeZ53xnVI3;^R@m#RO~`!^wQRpK*w*5aouLQLoM4oo5SPrhyrNsGR5`_%=Dq%}VOv z{0VxkkitC*yP4f(+W?s?4V15)m5)75+mhKpDXX$rSPXijU!|Cy6X1_`d$b7{B;eRw zu9Cn(erxT}0)5n%d0(1T;Z3r|CQ7;0&aEi1@gGp+QL3iIQ!P!_BbG~+L^7&ioc4^9@ zMv#*0!0R$pgn4yZwEb9bc5@``Y9018`3Tr1)i4zBZAXjxa}DU2R|@u`QRv(JXli!q zM>p&jWuDoT6>F6{+$xO8FDKu&u(JzRFRxelRkRU=v-Mh*_r-snUJF)seokOCOyCk@ z+$Ori_&~m*9Q*jK+rlaQ(<`6+0`{lz^CGh^t4`l6&9W@{rl9*}+-#uA>%^<#zoUFt^2s-jb)F_*sg5@5|k4f9V+psB$ z;tXetH@Pw(!^Mq&v)Jnq4iu`&6IuGkw$-1b;M?D4DfEC6TdH#6Y0tc@jVNhS&*S%H z^mUuMx*Me{`;yMr>{70b2wS^Sd-vLlq?Wm69ekx=)da?zO_J=(n8*R#kWr@TbPq9P z*(V$?#!*4cp%zqpv1#8K8v<@ew*fiJ9@%eli0Pjg|1T0!LdC>F@=m~oOQhF;&Txnsi|EcLM83GqD4eeX)%F*>T&96^PDk#So5OsjowKZl7yxu zdnM;~1|p80_5*R#|Ab26F_iKQJ&ZYyHPD40;_5m&$$<{NkyQ0ogWu0f>ATXT06HFY zd#coMcb0zS{)SV|C+`jc@KMMpM?mr-Ar!Z3f3VVSc!zqH^=dM?iLX}fuZHrxlHnF4 z=)m^~d@(!YwO}8Gy@>8@mz}eCR$$26Qs3KbY+s$#U|43)9#w8)Tb<1hohfSJtM~M| zg7UO~vHVzKU)^lXi&z(kssmzi6WwF*Ig1~unxIVzvL<4z{o3v7k>U7ZS6{n0idSUC zGK%OXtQW!SvAAMlX(%&gA)Dn??IYXl??xVLn3j%FXV@k%N^kmf6xdk9@@)#`Sk2U+ zSGLZ1SYw^z=Q#I+IIK~^NUMJP+y1IV#tCy&OzU;!bjBc=p+J{~tg)4N)J-h|`me+Z zlElDu8ehc#LZc}c6E|a?(s=>xuf9y`g)~H?)a$~&jy90Sp#)z1bw8l zDb!qi+|jW^C5f+oD}oy34a(zh0x4QX3NxBj^HtR7)%B$ow^SW2&`@GbZL*~EAcGJ4>NRG+ zA&S3!CgAE&Y0mE$zEYX& WSf2kmPat1x$sUN z4ALfB9}q(+2SA30x(Wp>)>ONlX`E&6w8qTMv2G*c{q2}NNoFkT5dVzSKPI1wr2)Aq znexZePE#e0W_z#Zn=pGVyMTyd10d3QiTy73%MOrUVf0yz_goI8G1wT%T-i^-0{Whw zi~i^fVgD)@dJzb1pv@Z)EXBOC^(e9p;6B#-T^yZ``eccl24d?n0UDy`c5&flfX)Y< z3L`tf+Cwi-QBEwj5wIXhi^IRYI>lN~N&)5ItB>(-H^cJudPHBfgu`Bcc;bH_)AC*> zSY>J@p0uNp9`z92Ac;qqN744G`B{$Nd7JvQEtXuCT}D*&XA7q?JmihX>Fd%~Q0&Li zRJ*!qwYn!Cn`e$5FYxN*|FqVRwYDj*Q*q?ZwEPGupQ~wAnsT^4kFH*D37#rx#rbv-?6VNQ|unmX*-et zQjeoG(=a!hYm01Z{^3kQ3Kbn|(dynh408J^A1_yFNlUV8j@zn%`}E0BQYi#E{$%gM zWK?=_0gT(VyRS{%oPCmXvDosWxqfxV!?>epvAb<;#F9wUs7z4Lz#+qn@zdtSNBUIZ zJ=93Iw9-~L>v);`gEhBl+BJZo8X#mSHH28RLUwH%>X%Oi>Yb*5HO8gBZ^wFkbDfvw zs-y(l{d8EUgSy5}>75o%8Hjd1owHBZcKcjjIy)~+hgKe-NV zo%XV%+w(@Y$}F%3u?C0?T5CuhhA?`~yBvB&#>!_|E5ud+>Z~#BxbxweZ!`KlmkBQ;llu=t34cWVz}i= zBBZFIo>eqPD~Z(JMwz_6(6 zBFgndkY%7!?kO@_}xJAlj6_-YNr zg&gd3gbeLY*=~w3l}AUiyx=JaI(@Fx-tCwtp=8n{+XH>^EfTCQ1*a!I-G&>{dBqi~ zy5j50Oi6Rd2pYDrNN{d7=%sh_=f&aF-|^5ibAEKp=$IQVHodSY{*fNps$%Hf@3By~ z;(zklbiZg&9aJ~l$X4!R_%`_&DUeqP9Qnt(ig=}e-8bUEBHowbpp0+|ve9U}d}bik zS7X!WG&QYWZ7)uUl}}J96V&cSKKh8~D4%3~*Sd{|`;&<5N+<0Kl2|pmg~gKueP$&d z&SIMJ4%Xi_1+;&;9fq*Qz;eD{%5p-4I2RvniAI~1NVVH zrzZE2_gM!}5LH0@Ii<9~;QZ?@YNCSd@U^5ymK5?|K@LE;yh26r8a4j*T=39EO%^9) z-o`ZIj4UrUwZ1%?#~At)?Wag#YOvT*|C(D$KSSAfSWY60oY3)e9s^=|BOXP!SpkoTJH-TFD1f% zxN1Q08u_cC(FFh?VBS{q!<$Zmy-A{!LzBe57y=r^VP~~r!l~cml(=ZKyQh_LUKoE$ z_(6X}`>Cw4zNA8SHo6?C0r)zL^6Jw-W5H?rXi{b3ee)rB*$tVlf=7xm9Px6;=8bL^ z#NlaNjiWff0{Bzab5%4LKFdTF!JGf!@5uDtg_KY=v9aWIgc72ltQ%v@n5;#e-J1(< zml`t0?>0`?jk;e{?ij~r)o4+Go!;39Z2WX3J&10EN+l?>bmCG)BxU}j30c`_g7Ps zapVT+@3p|~VWy#*{1~2#*8d08Lp8$fwh{Ds?DXDrobko8Y9>t{feklkrID8f)tfM^nc>+?*X)b{)YLJ&hdZR0Z$Y}?SiGW40gdvO>&7`ZBrRSvUrXnCBzlEy-Y7(lxbFge4JGt$&Abl}>%4y;3XwfO1xvtV62+rO#Vpa(F7 zf5{H%tc9f=$%yW8&r@PP!}?kvs+t3WYgA~`ovktN801Qx)jQl#+>my=i2~$IwdG2s zVlb{XGsP^BNiBC1>HaL z_)3>`_Euv2hC1f&XLYU2djBDbR~A9j>+X%1#Cj(1}B^w1-+hzq+Hh1$U9kU{qwljV* za#tgAXrk($(s$^D4j)E2bh`HZ^`6WcyU!4Fx*!D=>BD(9U1TF;fXpDbFQUMn70D&z8}(R@sBXAS~Y zyJFNdxd3pKSE^mNg2^({bDj9i5JGC+s#~i1UhA|vQ+#)B7IKJ8S^mL#;-?Dp>$i@Ks@b=TdE%eAzNc-w}p!;0t!D|NyZCR|buBHNzm zW_n$8`*his$)V^wJjALwE@)l6&`k zh&1t8HvkILDR}pmDd7nd>_2yR1ehSz(q}W>rYNc0{dSCFcFi|m@-CsRu!!?nw^3>A zd67WyLkg*I^4A}_?6esuHudI$kN$E`g_P9bs6%Gi^Bz*B;u^;J-obK6ZOE^`Qsvrx zf?n@Zm}n7s1Wy@}Io#b5xd{-Q6)5ZM5`6nlt!sVDgHbod3464GdhKBy?|^zFlvHk& z{WJLi3haF2g@z1mE%Mt*J~J@~8#U>Vg4Woec*d~D`%CQsF)cQHwIx#}`fvk}1(_&# zLHpW>OVXT;8}w^_=J;$rZu34`p0siI*KB(HNQb?=>uYtavOZ zF8(1eE;fm`|0D@7)TMn5mR&3sZE!IjxtgrYhOW?~V-Zn@;Jx=uXD-&Jg2RwYNzGKf z{!n43BGh=W>u=A*;Ttty&VaxxBUhs~GI5+I$iBh%Og#ps?M1vO#JC3ag^Q-JO~21U zQYAn0w58x1aof_hgn6&AuZ}C%UHiv@&aw-2NC*v7f-wi}N8d8~yYF3}66%F!JEvsY zlp}qcDeOSi+A^~akhdPb<+w@$QQ*oxHDiQSFToeONwMdDowa#Nr|6@F@r(fGaNdIV zO3WFc!d{sjUpSm|@`JC)>Anaaj2A#N%{gI0m0TVd^Qpu+clW+BBYQcXAayoliNZw9 z=D0ub#|a}NE&y;`Go_X7+l+TwE6%v3UoZZ0mRbGw<_hQ%tsKYunJK^Z%Ky04ESV2% z^ILR81_RlTzFViT_1f1tu8Xey<{E=JRSpmNrD;Nj!P~em#U-wUFdwrDkG#7Zp?p04 zj{o8%IGQZHU3|B80%#{CzLu~_-T_SCUg#PS`OfMXI#AuXzx+atNxJ(C`}1uA=e3Vq z4@p!54do@VW$7(*U=+57=|0uf6T|Bwv)&!gkGg;K;-13+;B=io2m|0z+UZubmvrE> znm^Q^~4arTUy^Zp*dr+64;w1+m-o8E>g zKIUip<_dT>!eGDqybfa(3k4PO6%gLZ*aS(N{JBwwe08nNf;>vpZ%M43(`@EG5&hz> zNQHTi41Z#X>tX6S1~*T66cCGUoh2drwZFphc<7(D+P5X$iInZcQ6r{nxr3=!UvIv0 zXRX82^kbdkzCX{FeC`n1$WZYtgmE#BENxyxzW>9nlDoNxTgq?p4pwy6?H3Oyz#tO7eMx4?d71rZQ4RZGPwJ-RaS1S z#E@bvgPkn)M4Q#7D&p_ww|Ck;ogSG_13(!_H644-lE91&NIEJ$;i--l_5Jr5=CoTR zGUsMjA6cIgo4+zhJII9H>m5{msiC6)9uhvNK4|OrE7bNX^~M zXDCvW>}R%-C(>HRn7!VsiHhp!oq4xZDJl2O_ssC{Y!-Rp615jN4#AsGyqeOr=KBF*^0kfKqq@1CIq zCJbT<3oKWUC={LLIra<7-9>ZyY*D1;N%qVR`3<1~z~a!DRC>{%iT<@8Av?vHI#=X- ziHl{g_OG?}B;398solnt5X)AJQyH2)dp%3)sqAEwagXG-HzN}XVS^U&4dMN~o+7p; zse233#Qwu*OIW9bWUURGl@I@mtCU8{CRB9u;LvN5KZLYb?(O_Lpu{GN`hYq}BuaU} zfcQw78tVCeQW+Uk|9+91zQ`$&$Qu;qp(&62-2U^7GzaqLgdMFO<2yt>L)9FZ%-EIk z0hIbtbnl@Y-`$32G34dwE<;Di1RE56$%YgACgT zeAwpjc7Ke){-3s$%5YWaTC(9RvE)vb%$2Sc-gaO=qu0sqPkl&g=vABVPIXwF)1u6E zU}x5F4bjFt7CtJzkVoF>xC2fYLOi4YHvX+v=87^kEI{t& z+f5nd=Y}sb*zcpD3HQwzYqnz2?Cd>3E_yrc5Jy7S**eHIJIEUXlK95Bm83I3A zJ;b|ubhbqgxZBYkPeZuD0>{njnGyj%1I3w!h@a_{9aw5UShL2-p>!S*+)_&*5h4*< zwhOz)VhpjDrz@3_s;@&{XUq7%M3y}I*ApuN@kvKn+1C#(Tz(;eQujomC`i8(D;4> zp1u)!pADsed^L)$!M#T^>KP)6OOeR*-%8wnntFyQMWlJNig)>xVp zRpL~s?i?T{4gx??-(zS6yE+g}ZZ@~IbVwMZ=YYrZGwRb7=F&r3sXWYq50 zoOKcM>jB#0Sn%KdDzU#n1B5wq!1+&q3~&L?VfLkAvwJZkAYU#5@QdaEjQEQ0 zI&oA#7La&7+suud1#pY&H2f%F3K`vCg7pBJ$N;dlE54EQOOLYz%!B;_2AK$CMj&0v zr>0{BpmJ(R0-=66c)G&p`BMpi8R?5Xo!&LQx%|g8{N*9y?VO40;1kag@eRFvmGKd` zFTJFTzWdUczraQN$s)}F!FGG##0Nlw4%z-^)d0q9z#dUXw*(1l0JqBodL-u z4929zb9ZOPGOo_p6WO~xUCF$tofnK52Y7^Lrv9f%G4!G%4Mc?OrImeCEP1lbq?0zbIGq zf*0B&otV08UH`a;LStqio?Wym*LE(JQF8WKWor&@UnsW%Ip!y~Z;^jb-`5Hs)+5di ztUR`OC#RmrxTeyhCy<8feg;Hr_claG2Pg%?cH=#gVa^|x28KsQ>iOzt*xJ#H`mNs8 z{ugWioDHzYd6WG#?FhK0Ii?BdO_#eH=CZ_)E6AhGXzsFoMzD`jzK8kpkk zm}@y%^BlZU(yY0rh5bU+R0|+e8Wt$?JCM}FJ1Y(d$!}AXK=R0=f7F$X4yRqW~P*c5*Ad;iZVNCfE^!qLIkHAowAnyhs(pvY)rg%-sEH&K7 zWg#hmyw94v!GhM^5kpS&l6VaJCt!FyMd=f!Q(gtP1u-%a>>e+8EnYm}29$%EvLDrY z?ivK{C;A&pnLWD#h>O>kg|Wp>oSa&tQiWPb_z4B)Pl3arvu=drOqF>%mvzLXvzpKr zsbt(`e&CYij}8H{D4AfQ@7}wmd@pchYJpnrbe!N8`;6;dPT#+lzntD=TS!IiPL}}J zN0@TBov}i9iEpgDJ&z^*fg6GChbs(ysj9Jr2N{KNj~FE;(fpTm4#^{qG@)TGV?cw) z-~D$dIF|$z6F|34)iqt^hA$5V1r@OudlV6ql-PQ7@Jx*8A6m>UuZ)6IHGIKy&Oy+o zwHK%>FU~~M30YJe_sUkYLP2nV3jF(xpy<+J_FbDi4AxlbBz+(=>6vcd{1-03A&u)b z0oXt@lKZU`ZX{VVKybqBCfapw(zL<^ss?-m@%8BX8DKRr&`WwkAOISz8iO>$^mYQ?l3x&YTy~9*O^4FFGmPPxg(9emqV+8x?lfS03RMG zEO7>)zsAqFs#z!5r%%6-$k4idNlBr*R{UOT9R(x zY^@yyYOcWvK*{(VwoP=q;1NBhbi?z65s6^I%c(!x<7EN*1G5`mO$W2fz88*SV2mu# zQX?qp9Y;hf@JkkFd}<9icq~Vy(MDxJmhoQ}aB&3?pJMa{_~I6TwQ?x)Kp_Ne{Q8sm zO_^>&28d=qpE-X?npqu8uzZ)yd+0d<$or6@=9}Eg03&>Tfxk3ZCPP>)p zWIn@7R0m+ls^^?K7UtG|)DJMsxJSp>Pgm@su)y!cbWmLQ{qfyzn=$d-t-Efo+bT%! z0{{h^GbFNC-w^7QxGMk%?<9aei#gik)|t5Nk8mQn{MG2PauqN$({Xd%!n|lvI>98h z0l?aAZmvvgZQ~a;C>gT`jaF!k-8CVl<@HDCid#=X!gGhV1@NXUCJNM|_KP=s?Q(fW zs7S}De>yU|^pMFY5JO9hZE9HGd_Di*09!KWyeIU?wcD5(WmEUF;srRULwJJEDJ2ud zBql>e>+ot7U!}gKvll&(4yZCMj+U{F_t3c`00}WS(Kp$9F_t+2LT}_&iKZ*)jnae~ zVbZF8a{?=aboUOWkMRBPk=ro2LO>h^X&0Q=vGm{0m9w-pq^Fc31TZiHzyq5~Pa<2A@d@%Rx*VUPVZa;1aEpQ};y>~6cb#V2 zD)B6olje^Cpe$>^dNpKE`J&1jF7*p<*JPZ1<3Sgpe@{KPNmY6^tFlMLLYV*EnTLb* zccH#F`#0C4vM5_Ft*~q$o!{PeUVO&GRAl^2sefb5Zto|J-B%s|@2Wd28xqW>qLj^6F)P_Nz9Huhm?m&VE8xtQL=w z#1Mn1ixI3ECG%11hgYO7o9`6J4gj|B6+0V>X4cwS*ncju-HqgWaAR^yh8aAxy=?A~R-Dg3k4`K9fc zvNPuHGxK6xs%R0NBCpr^UyfyKab$gBFof7AV*<*!JhP3X%OMWqd2g?#!prMthYEGr z7UbkYrecV;v+IAD9!UNs4xhYQO(X8uz*A?~`$#|}*6CwXMwU&k#|wWh2Mq#U_0 zZwaH-nzFyX)s9#yTG98MwJY@DIs>IUfNU~czB=(Ud=k?*ee_I1m7x{zQyk)?&klHI ze3CPL)>AzvHI;nudFti8730|=Jg(0c0|T9=mz91;3TEVZm?-aW6p-6swy zoD9??lhQeyqmEi_3CU?hfi?$+OZ1j6B>#4^hB5r@S2!sI5-_u95-qlEaZ`6%l4?as z=SmC99CKx*82+#B4tEi|Q^##5Kqc%h>jno21;~nPBh#utGWhPunEQx^<9FDH8N0$v zKP95rhfE(}mjOVcarFl`%cMuiT{TBDy<>q~Mhe{fCfyNA>B^_i8H^@yadFbZ#X#R! zYR5G6WN~uS=pM*_q*`xeyT?weiDwxDW@3LX+BjNW?esVLso5`i6O%nR`-I@`I@JZi z-&SW1CKpkup)>jKX!vEq{08PboM(QE(y~OfE*q?6T>}~QIb*MdH5(9HV)Zo*+ubpp z`7o`{*EeE3gW!zT6j>bBqxTa97coQR@!o^4z6Cu1GmKM7o>A(%W-p z-TU5qynn#!2S4Z-B3xI_GUqYpuX&$9pE4aMX9A{peR&bzyi#dD-ZQJzQ9 zw8PGMcb+i=Ug3lEVM$%ZY>(;~Ven~n|a*F2F5LfRs3W$Q>4Ke`h6^&#KC@;PRX)R7*o*qi# zL<-@Tpao(wNXZ+yYjL${eIGe4X0nU3o2iLh>+bU!lWmYlyzwhW>|sUb$FuKx6X_>_ zlB>CNpLp8!R&BNQdnvq()e%sFcSqyGe!qG0@73SzPW{%Kk-4CvGA_Q0Hs7c^Qrf+4 zpmayH?X+wzZ}T&ht5IgnWYO`#lOo|y=y|Z&Zc1~WQ8HSWT!`q_$w}9BUd!y7zfA63 zx#nQ!h%)w^WLhM+IlM}_S zKJKG_^Rg~L)$EN4W{bVi)g2!FU}NgSvZB+;)^W_$$c1sr zTM;s9q61E;%jT-bSrZnnv;2kk^Et&1o)3zfV(0UxN?bQ;PBH&kB>`$#vl?C+;RE&c zHho2(S5F*f|5tHg@0eKw6~%!jg_C|6}GCGifeIO z1*Fthr*w@UBkLcI{eYAZ(h?KKznO}>|Ma}>#r$?sX&lsei}m*f^|Cj?a<<}|l95bb z<^V$F9d0-HpNG;i`MASgym#Te$TOdfnI8sj#`CLFfvTtkCN19plAze^@`tNBT~%Hg z;_S*T@YD=wT#3c@FAAIG0Vz^Pv`lM23EnRH{fQmq9<8UW-nT7!`21IPYQJpYrR<0+{R^UxpOT^4}BRD>3!^SUf60h z>p}F(KxBP6R$bO=)Djx%@rM-~Pl7*!L|h2c*(LCV7_*cn8i^UtD%~{7Rv?nIK#%*o z=pcI~^-OiBmovp)7wS}SKuIO6MW-6_DrNh7)lVgC_QO0>$Jp%XD$W~o?LlU+5>&_g z+46bM5T<6%OM0`=pg{MB;{&zjKS@SjI|)evdeLP~Fcf~^_|wkM_uJn$_GdePSUu#G zeXDRu+oyLb*FM3$2H>kt%BLJ22m0VoW?g}>b1CFMU7+kHf|;Hl4^A+VZw||GJtK75 zgrp`~ug7j+qex27<;`qo*8~aTdJS>PJZhjlI+kaQu~VSmrSx-%Yn3L`a+Yc2l9Xou1{Rh@f{U{&g!~Gl zvUnunF6+YF9Ene>Go`N9i^|UkLryJcsJ}@K6VLwY4(i+V#nj6xuC$Z(SDGf=mh4m6 zzlc04qHAQ5NVE?wH+DiwWDlJ!bsvCK$#s?MIv)TcVL_Nm4J5QW9n1Z;yq=>z9!I>< zbplHpm6|?1>iPQHfipou+;P7;haQW{<_mpQZnBZ2(KmZ$FH7o#j*Usa$Hx^iY)4V= z?Thi>S9i33-Chz)X7SY4n3lD-a*Otr*p@3)z(kzv5ALwM(|BKGJ@S$^ZlHd&I$A1J znsI#9fP$QYDEGo1i#fmts(mgF2)1h?GI z@ThqBL8?>fHB|!g#PzRL1&qjzXly3O+zWY8dgG^1iCuhS%N~RP^%ZSXIXl!A5x;Nd zC%k@#_&I^zG>UZ@C_n@CzD`B9clS6E@lzVMdNA?njFyTnp|O~KPOY}X!lLHZ&IYqE zA5X^RmZPO?cf&#R5T%omg|FGbkVDxD zotB`TpZRp7{{G1{efkNGNIYtC$Ai$V{cb=MDJ#cc+F0Eb`dG3Rf7dN4w@N&jWxK+s z$*3_szDX}?YPau>I_;T*a|K$4v){vGeX(Y0RVN>lzXpz=>%x9OVqIetEvEv7Hg2O9z1_j4tc)tctBi(pZJg+{q|hz!86s9i zQg%LblH4KIpfJO@%dLBQ4ulidmx;4&;43V&UCg_RU59W*xz;&BFC5!+mZa_oxhcPx zqMO`%qJgAiwoZ)c;vq0=>NTOS$S$T;8Srtwi`f3A-1wRZ;k>;-`2#0^D2KM8%g*3> zxP`wpFD^xaL47+7FX&<3zy1FRwqsr2_^x0LSAdu_Q}84aWWV5xVV>x~n*X|#i-&GX3S6Sph=2)e< zpEDdM&Jp{U!&_y{+Ra@()}Q@&6I~X+&%W8OXch)PZ<4H6D6nMym0l~4zUKsF{CcT3JdniLcS<_a*v;J+M%2PrFyR zh0BF%OypHm%!jRIMx6O8AqA8oOADp3s&c-u@zwR=S^E0=h28C#G$wqy=FTMJ#IE%@ zp+KQdxY_Oe0Q@E2fLboIi!ch%>Anla;TiRB#rhg}Ik|RRI3~$Bl6+^%GAFobKf{!P>FBF-7-j-Ij7Q>&5^)e5Z;(-pghJ!g;m ze5GD=kl*GQ=?{#6BGG}>6N?UXX-FlQ`}Z=AT^R$#!=DvpY3sM$rLrV7kPSd}%2dw3 z$^Lr{$^dglQdcz-Ro7wUdRUv`~~f&wR*e$?J4b9LKp{JB9mAxC3vi zfAX)I3t2$K&<#r1^s5IaHS1fn5na(e{%X(6KfGs`q%UQN6WU|!oC)NYbn$r#9Y1K< zpxYl&Eg#&j679h$Il@I-6vfTp$`HtK>k z@gY4u>fwn_)b<4ruHTi^@n`RwP3Vp=__@ZWf^G=~oNqE+w58F6C{N@j!i{@_>g~K8<&2 zX3<&4$YWa7W*52ZV%O=>Vp#SHY{9<;KVFz9FLRvPjV?$% zPXq^WY=qE&n?btBa)jDcWBjL1g6NX|7e7KYk3QoB$MiYRmd*4c9<_onPM4J(dlD)Bd(A23N41Y|Xt>)7@O zOC;P9Or8cXH8WEiOp<}~77Xv2IFRYng-9#-xzm0#v5cI^H_gzb(3W`hi3k1|*C?m% zGbG28JKEaglBcE~p8Ly$my=W3D^@+EevsKRLH!-sbDWAR4uv8Lw0p|w_IZ}Ih5Xcu zh0Qc~on@(mw%Div0r-&5_o*t0e}n^5F`as%^f8gmFfIzC$mKt*wu}@!%zKQ%iD8Wc zf>&SaBP|e{wtG@8MD$>foYzb*HoZRR7E*B4f{+Rv7uI2|ezrov@U$aB7$&l2 z2b-ym4X%G}k?3*IcNo7XIU>g(Kca9*U$5os`56pb&1lO#%xf_yDk*R`S)#*^0W;zm zVmEF-qAJEc-5z4ugty|vPeyNUc5ghqT^QNkoEcPlxD);eASa`VTkf2fxXy@ zU|?I z!qkkq%TfS92Jd{;)r=`^sppkTG`c)2i$(E-RtcG@EuYh}Ose3L_ zY)*2>$ptz?#a8-O6*CWPNUo{A@s$Nphy1R~rSo~hrhEJnJCn`Xj1?j1EVg+mPaUBF z1NGNCNgLTvhq05EIP@ZYL;?(o-_Bi)7Bfe%=_e+6lruc>%DWhQsNJNUANqjGPY#w?3m<(R8~|SYebxpkh&8$?auapB@CslSo&8UWn0pr zY*wAO(z6T$Z6YnMyKw#!v3s)m+i=38l>blP0l;p_e&ov@0{=DQ{}bJlO8nmx+5hwh zkpBPtBi;wrGW^HqK~VZ;_MBJiTXuF#&zbSonwpyFC{_$x++OCt{V!B%R}M#hcD54; z8UbSB@m0VtJOj|fTo1q9%UgE=nbAXw1pRg8#j6Akp)tUGc>`Z9ZN(Djp4go&x$oo! z{Qjd5TGxT}g~wDH(GIRlbsecfo!k7N=YUNgv?cU#An$elyKd1Ku4k<^MGaYPhxphpNJ8snbln9)XSG?E$J907xveeATRp&d?arZA@Xy6`j zc1AB(8rEj5GVG;SYw z|1)9tP&NX9pIx2r48(%66pkcyT{8D0JjcoWHQ+3~-2DCRD~`v}4~o$MD6oO_#*Jl7w%mzQcsR9T8rJ_90VvC|!ZTiLHiOt?Fu0V?Fc| zwrM&_Z=S;6hUp@j%*`Dr!H%Vupd(}NC;tLzG}R>&5TbotJ7?d_a%=rOE%Wl$KQ#be zxRxz(ehCEPHWj%4EiEGKCahtdih#pCDiXFNS2T)l&drje%g&?9@;MA$dD?APX?t3R zqK2Gsb#ntTyi3zB_)}Z&$!Zm`XEP%nlQkE{-_$NwcZwX%KYr-7&@aZfd*(s~S7bu} zIRFj->;L!iR;*RV+ex*Xj2*fMw%~%XqoxbBjO=Rsf;hebJWQti8YGqUA)Dot2htKb z;nYD{Q%UkYD>m5+eBVP&{2#dmD5KpgkGy}`fj+8m5Vd7~@o(TT9$Y7!0tnr&#iEu^ zzMaBuz~)*@xE(svbbR>Y8#wZ8fu4#n@Dh3hqJrFX0goJy*!ZLz1U^$cF>QHng*Sl= zo!s=!x0Rx@5Z`j?_~NsK_Fh9$*1G!7(<-Z@; zwXsl())WCW@DD0wrNk_Er&Faow%^jVr)T!9xz^$E{iHQDUG`Sa{e$(7vfnx9^Hm`7 z^thn;Z5kirOOOvw>rTii_;Wiav0vfniAJt_VsmS7wo8m}6hLmz?%@qUkP-}Y1QNT1ptJqFj0Rm1^kIMBZF9e9R|k`Zg!8}kQ9 zF)RMYxUF)pdWoHfazWraei+!Y&59vN(4Zq+)?jNuX{P>)- zp#e1tPgrqiq9I7{6sUP*EakS!5hM(;QqX+?bIb2nj?HqxH^pKW8pdn#9EA1k# zc6~{5wn%|Q{}wOTnR>DzbY%oIrv42C!11;&`CKzhN}cEFf7bnLbFibAk`8CVo{HTNMe=qFOC z>of~u#&RVmE-(zwx!Fu_xhzTPk2jl?#z&%NJu@ ze17BIo|D4=v;ofv5kxwQDOyQ)bl@=oJh@e85Dq(ecw61N=9vDlsXzjKq~sXmT7ul^ z>2|=6Ea@{fZp?w%k@So#J6(u&#&&M3KAA$V*}s!O=B?qwRU%p#5Ds8pucf&=DVP}j z6#iY-Os#{}xB{>R*v>jE=YsvAj~$~t?xPoOzQI$R+t5C*V%I2YIoq(&_jPid+rh3~ zraeo@yJ0Xz9jZ@h+O_SwTg^P*6MW${0d48pUINtgFC!+W@LRO}z+!B_SjB(@S)cM@ zC{;!3Hr^GUF2$&%mi->D+(S7u`vTvJ`dN|~F3gf!rv#dmCNwF+@EPLg-tF%oiM=Qm z48KoL|4W2?56ZvB4d_4xEh%5m(5o8C`7xXknFhae5bM>~xeqhL6Sb!i~C zGFnEYjnX9Bu&V1#7XRHgA8ToNj*Uf9!8C-kv8 zjnq^gH~P!)L3G9`baOS4!v`$g`&S)EJr{oX9MDDas2-8u zefRTj)j=3IYHN_%z8bil(%->#P}Z@z)A6zM8*7=7n_!s5M`KJH;XI>#lE z^?mMQLnh_dLlg-!YlpP(Nl{Xln>0_SGu=l;ek;k7Jbih$`E>BUS__!og&FD}p1m{M zxwQP`Ua?IjLD%*A5Lbm3%k0wq@KiS121u^q;tc7yN#VQNnJFiTP%gK0^5dFmX(x)5 z%8`5QOwMcVcI^_<`~?nnD`7g!ZIB<}jMce+tpa-+$D5~m%D6u?I% zns<&6VejPR=D5$(t!F?#bZg@qMSxgou#1H7=)`$aKAFBj7@4WB5u$A5g)A<6*muO3 z=C&(6ZXi6cpHSdm{`PKr&)0j}8iTN^@=@syS z#>nOY*mNL$1MaGu!|7zuVTa0wp6oEoQ6sGTn0BO^{w_VVet35T=!INb5;F?qTlrcp zjvm@Myfd(+y|Q%8u_)!ut{rC(klva`2KER0d5k4@N80dEKHt3E0e229E~t?0D_WDB z5!(_3!sD^}+!J+z#oC?KE8rUO?=IIRg?x>SfC}L~<*Yx)U{c%oow&+`t7H0SRwd(w z%p8?m7ww(1-YJh>?RArb%GDsw;aMCZ2I2}qUDD$(?6ka>=wFE8?5)L~`FK!FP_4ZO zEusCv7G9LEbOu-haYr@-sg4N0VfsG5v`!CcR~Cl>Cz7OU3F0y?2vLkvmDY;mZ^vkU z)E4vo!N($#;6^9M5JUc8Xyn^T$2hOf$i^a{uTa!h**{5=mnei7$o;CuYj$YF^30Vq zpwwnhcO>2Bkh+n-aG#|UpWprrnalf>QewX&TFtj_%%5HC~{KHBx3}@k2i2s%QlXt z=P7zpfMj~KCo+~L7?qRQB0OIRWn^c$5)Nr)K8)89aVfENhp;?{(Fxi}aLU7fwf3sC zqxkj3FENb3BI56qn}V(A<%$AX_IPik=|HYRfbbVxV80yxGW?RXzx4(7wbZwB4T3u- zlT&TfY#`)w;rCju;URk$@B+`37q~Du!N#Ja_HCSxtM>W#@&mJbWul+WRg|A|usl7O z)%eXd?u0azxW_If#DF{`ZHWED6!*%x`qIBw(wqo*Xw?frp3DS1{!evab#1jq3U=LI zORN;}_)7_yM&rkS!Nj9|NFIhV&_E$=q-4oOqipd5yxcI4vIE8e? zP*QT4IyIr*!_4zx0JhoP(xH@!Y=8CEn)ou6zj}Joa59&AL`ifP?cgRz4H`I@eyG1& z6YtAD^P{~x;ukiP1TU?#e5-^ZykQVy(?fNh!OY^Ra>;U4dCc(`$`lybeI=yrOx|Su z2TWR7(dZRq23ZVAMP?wMBIoffN%oi_7BOzIt)<%jS1oanFbJOWrmMshp;IRxIyb9^ zJDHR~bU9cP=@9F{#Kqaqvf$18=u4u>?67~p32v8o()E_57tD`E8kvU@yX*8|#oo?U zqgoEyJTRvYlq$Mo!G&0NE*5l>eYC#_>{O`eRH=PL%5`+bWXaa-g`Ns1KmftT+7bDPnzLruO9!F_Fh*^ne5sm?m?L(3Cb+l zy~#7zifx8ZTe|cHIj!>av^J+fQ7pl*@AbA2z{`KVw>KVjeE6P@+?*JF4>l1mMfhr| zv}adwnq{rLX)>}|1zR_m{!20^Lc9#c{P~}~HCJ&$E9;^`5rs>kky>$z_)Zf8Wy(&oJuytC>Q)h7= zLTTj@I?X!@s0U;eSrBSYR-t?&pLFm@e9witFDK9m4a^Wi6&YR2(Fm32otwc^H^a3^ z>l+Gd2Nz`_*-y!P#BRjxazc|SnV4)0~5vXGl5%9 z`SeTdm3@oAq}Z-pgBQT><}&r{q;z4@EQdF^LE)}8XfrIeLFE7J^>R$$wS?fCeEDZ) z^?e-fndRoCJK;0!wZ)9>yMzM?71Z`X2h0 zJ&rr3wGo;Bfz?2x@)e#7cDYfsh5m`^G)pFLZ$)d!;|ApYN)`#2V@SHqFj=pBwso}c zjYSlVUsqnp3+;%;2&Q$*;mQSZxUl5@NkS92b{)#WQOIK40Gb1dF#ABy$WB0^Hg(SF zp|tb0sclK=+S(Q2kvc?e{M0qa$Jd_bI1g(uU`GTQd7ksMpD9hQh-%jBSF_^y;jZ-@@m(w{fj7}+C7-<& z6F1nv5;IIwa^~FLWOhW;_tA*J7mUvhdlKsP>fj7W>B9;`dwXNMQds!C7(`%qpk7p| z=9|XQ0>%0)2kp+6@fK}?2~|jf4#6bt6J2Tzt~SG8Pn|80a9qChXW6-)6V57lpzv3E zUi$weYI>Ag*33>#F>pVmD9o zI-{e&jA>fiPD6zcs^`Q3ov8?WU<>j-Swtd#@JGhe@0j76*f%fNz*TIa)12bJ>(v+U z3wPq|;M9=4~J7enL%rQ{yiwp{RK6E^nn(uMKrU2(Ys`e=9 z+!X$xJy#+*an+lmg*te*()h0ESO2~TGVR&77UA=p@G%7bqUnTc$ ze6#KzgP*`mujB%G^wI}rm-%pc&CP9OQ?uuJWHUcJ{F*DPkCqsImA zYGcxC$U02c$n9Y5A#-#jLO~o@ZGi{5+YgSR(81(L=PG@^84$#!}?aA^uDBoW;LW zS1RxsjqGYX`-Mrjsp&2EoS(&fBhM7F8Yd~qDM?t)?DB;&yczw6zEMC(FcVGarYqub z((cHI%24Y~t(>(yf|nvCG~D6|A0alUC9d1rbAMoN@ zn^{5>VdOiWZ|3g@yt)Hh@X`LKq(YM%cLZ{+J5TCK+?8|4l0!SaY~SJm4<7TM2j6!E z_Kc=092Y_jWN((i!^tOzX2SQNokYc2(XLR6Zz`s{S2n-B8{U`J6|0l9A!PZu9<7>K zaRPRGT{^q))ZL7jf#_^4K7n^%7@)UQ^;u7p=lh53O4qjQ^LP2f%nnV01t!hMmKDCBGj|MRmgXGZqsG6^&qUc{r<*mSHnF&Z0a2Ou(;Vs?%p)tqd!bha*8g5Migv3IsZ#xmctEacH?A< z_ZyBKJSPr%_dk7kmcI7fh|EdKSL=D+hl-THtFE^JJwcz0FqL7uJ%KyxTr_=Rgb6MU zvMWnQ`|`fPgB`Q2-D)fe$ioy&B6 zMXZTAF*OeF6hvIJE;-n1TwW;XMG`QUK^;(M)=BS3b7ct=mY75trkqToh}ESFmlrat znQxa^i86yj1#+8FGrFC(pK=8+ag~R7J&~%HCm1Ium$7E#YbOM>MkK8_wTBP+0)NY> zmT#R?GS=7lBK`iPx>*n!BF5>IclGEO8p~Xkw}QTO?P;C~J|3Z&>k?l~7^|3O(Bm^y z>`;r#3NTQNHSzmB>x_3>U^K)VxXU*rTAF;bvJ`z!oB-&F1 z^EAEc&!ajNvugN24XNNtG3V<-^FI(PtSFwLj?#TEL4}xF}{BltT%DO zlo_p25!WGi;`w?F|7MIYF2pKA(h+H;5*lgL??(8 zduQu+jw(^ym$p910;LI)%S=73&YB&kp=jp|%^i=P0fL`AkzO~7nUl+IV4w)WG4Pso zH3J8Ev@TaHi*&aU$9WC(D$dB`NkP!MlgDI!+uDYL+$~xfb`k5_2vsN#ZCvj;oyzO~ z>`|L9n(|HWo8^{;-BcK|NntlF=^ljx8D_zS%JviMAjz@gIE$ZkFSo3@bB&3kgfPj^ z1fG2*T)WvPnwfG)EI9vuL-qC}E3NV>iN6?&j1&bg0IfoZ<}~<0%x)3xUqrAr2)JHz zh<@z!N$`Fy^BNai)&LS7OXO(Ibx2@i%W0G}@1lPIC;b?dn-t3fK z;YbLVWY8%G%67%a)!p$0nO+}bW(F%MH{65lDK)B9=I_mE2^(eDi6M^giYONU9O=8w z0>zrhRiFHT&YqNj`SZdoI?43>i`sIGzkxMyy{X!O$B}inByZO+FVFIxdImiqqxG?` z{i=Sb)ajdNkrr#;l3W~|ij(SF?UQR(1&a%5_Da(h-!HDvqWquF^4gj+b(XHW70i{z zBge&K?TM!ddjgi89!covEUt%6a&!_Ogy)<-Nw9R42RLUaIkb{?B&~Q@GJE?@|9$&E zV*q$%O;2Ih)k6JFWXIF28gE&7xzu4WE!*fx;hm@E`>8~2wzD2rCAc%c3+*cq9^xb% zxzX3H#kfO?Sfd*X)g#MIogvydwjc=??GVTFMwen(@a9LsPc;EGBS(pL zNSWhq&u_&$#)*mO$tag86V$4rrYV0phsct#zpy8VRF2!K+m=cQ_}4yZ4Mo=lcaEGp zLr&o*T8FPj%a?O=F*I~)<_k>ReX1EEdn1D6^yHM&+kV~z_}wkxu(XYxJqGhrmnq@ zPO&SDOG#Q9!mQd<&~)-9a5^~tX;S|2!Ms+`$1|@(GNlmMY;b#^^8u>k z2ID)z&yP@z*=zIrS3Nd+75e|QBH|_iGxAlTUwx)5jJTYUhOUsaK4X02-!L_~*B2nU zJFFHUjA3waw+YRBLOR;$WgsXZX}3$&MrjYAd-9F+2ws!zk_)6dPaSuv*SqC#!ane* z7ULh#Da`_pZ^f>Y^ZYEx7Hm&9(_#~z*E~Ma>K;>=p9LFTWFK552xj_G6whT#&aWwB z`7&3_dG$y-ysdq6HnHy`m#RDWq{_kd@5v{F47Q~pkDE(gRz(iV3a`*-g0a+ZgiWUn zr-$4B6X4S%rNu*tnRAli2YD!%$Z~I{bwk0Nb_BZ#}1oEt8sq4>ylB1A1fmYa{ zdU+zdMdU)PkzG=i$eHC9Ik-=>FJtx)Ub(O6RS5X94LfakiWFvy48`_zD{7On!Arq6 z`vLrJ;^+Th{{DCx0viE0C+6Kf^)`MaBs~K6cGr}yO{4ec)ZxXn@W!Hy5Ug!hK@Uo{ z*8_X^lB4aUvvn9?1Z;Kn92Qe&tXN!49>qbqw{%b~C8Dmcd^y7lRf&Axc)Cmiq|d^& zbF1}pjEGzrFqg+iF&oB^o4pm;7w6a+@y~f0C2xH5Ke*^=D4g);4y-u$s=+N5rmR^G zhqzRFXJA}icszHS4R)<*olkSFH4w)U!aE;s#?Y`h=0W~Gv zS12RDyqa5(!&ZuF9g`DhHoMc##^33D`GTsHWGr)SHMibgB?PrI%(IzmyflfT4zXfN zU>Vn)&qt#9=$7@rKE7_4bk%!U3->ww%%#x^9#?;rV5N8wt5ee?e})#bn6H$JD^lY(oBO{qjJ;g5=dOukYf zIkEyO?T}I6pUal`eicS)NMzRJaVR;U(IhEi`1?GV@`6?GQZqpF&FOuWHjyvq+yP$Q zOw?VNjy?;P#?pYJ&W0 zCkvQdjP8CE^o<=x%GeM2yY!FL`o2b&WuvG?)oS533Xq{#}y zrRM71X-c>xrw6K&u)M1Lq6q{LzpthG=MCHB@HiwL`aBj(o18rYnfdo^`9$~^jwvMm z5D^Swr;5jF`+$1j&>4G5SIRcc9%yJRgU!kM07b3lOYX0Yf^_sJuySiPF4CgyDrqXp z`ODDD_>MULg6*!WpGtRUJtR)W#Y@HceDKA)@wdS;weum_XheWFTr6%9b>* z)6TuW9=gCRrHnuxlzTV~_BPnt^|S>V{^3KMBKsv-1fh<~S5T(^Q^Po7gLZ~sl;m04oJ@v{T){{zzu{jXf zUiH-Znkk!q&fj=c;=gMNc|96{J_G}YR0`;I7*e}^RYv1kToxOunM_a6XUc^5o)pY) zfU4jZd~<6m7jhPh>D5vdEn%st{4jqD&I)PB|LGk<(i*Jry0j$V4 zT3--tgV9E_VR9xb<~Q(zzEb`o&{Ozw-=hncQGYPWbe*c5^GJL@LY$}^e<6P&DRK%( ze`Y?7E0b|twxT*r_~uaoL-iCN?hjc@v-3M-zI?7CeS0JRP|f`%1rQK_5u5{=VjGN~ zxO$N7-=dS|FH(J0g*$fdF(n`3$juqr&Xd1GhLiPx`ehrgoZ`gonGId;-_CfO*2TU+ zZ^owQ+?G5O9txShK%X5+^Op?>wf%@t@sH^!+QalVSfn(w&iGt_po*>G*1ysdyK3;wS91wahGu^_jvmX{aFJ7n z6-JZ;o40F=WgN|d*4|R&D@hivUrtTehs(}Q zQU56S706o#@jum{)oC^&y$m5gp02IKV9nPzK@{qLb`Q(TG;hC4imZ-r@Kp!&`skZv zbpi)ki9{5^+F2#3s`JufdXzJ!_^0IB;ZiV!~-*(KYa9 ze9b|uF+-Hnyht2O0%>7@wApn6h4p``&=I9wdIeQMFtbP!WBX|mU!SN$g8w@ZA3EGd zBhkk2%71IzL!nqMO96t=oZ+kv?ZJ{&F|D!g=$<5MG(J(U04izgl(`R5XTFE3=aMGN z-46MBjO&;K$vzCU)Vy7|C`8jzXn3ENV2@dirACtz$Sw1ghD;27=|l=>#rWG!K|;$1 zwUF5_CUeptOk7vZQZgKpi3w-POjw+!S=^=kgNDPxo}Iqj`XbNGHPfR%Y^IR)g|H!m zTla1JlG3HZfJxg>VVK-V$b^tdIx?6*p;qw`<<8<8^^c{5Mx;E=55<0vq<9-4C2_a! z|3qQm4<)JBDeAHwW=2R~DedAR$&Nj!L7QT|EIE7h-qiDxk-8Qt>s?|I4>wf$I0h4M z$H`lAg>$xWVv=vu%3`)g;M3fP?7!b{+sdA#kxEAoh~kX{O>os+qlxL&dGM(Y3eZ7#DC~2+GDn~?Mmhgo{j|L3Fz=)z7J-YB&<&UV|qw;^C?xp zP2Q2YveSx{4WEfP$gXg(i>fTjD)Y9A@fUHRhsbkAcs&f)gmS%My%3FW zyCxm3EIr~g%ySqMttaT4Ay$WGnSA!-EH_5+8u1f|Q{yrB&=P2m@xYNAO3qzy*%PZ_ zjs`bI9Fho9O5n93r@hlAjHta`6?I}V*59g=VPS^-{zFFAoU5v`-blk zQvm>c3)R8$g$qoiAe$Atg8d5fW#CgSTk9idOe{_!0Weplb9z!$HgJfV(5tZgDCj0iT=N}!p zT?u?=k{P-+;haA~XY#;R>_luu3gG#M4P)T+ExNl&m96XkUd(@MwP&_k)GPbe|T_UJ`xe`~4R&5ZukrK3_4_cSuXEH~NDZQTX+ReCm`}41^e* zXkmOyb!gUT@_#3*5kmE7bCPl_;i1+Z$Fye8#VK09S`cXd*hzh9V@)VzoTCVq zNoqZThQms=RXt}NggsMR2WZ(>Af8X^gJHPggN7Ki*<^7e%}Qwez>#_CnO>Xum&cVv zELh#7n`zZa%A|NO3?-6vlhL67Wf|ZOVaf%aCbMtNu2|Z?7y0r+_oo=MgHgnPXFh?S daJR$O@$WJ?lz*c94;}brq;E!eu4^Cr{{S<%vD*Lu literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/effective_throughput.png b/blogs/deepspeed-fastgen/assets/images/effective_throughput.png new file mode 100644 index 0000000000000000000000000000000000000000..11c7f82bc54fd670a08d0b52f85e9163257fbaa6 GIT binary patch literal 222683 zcmeFY_dA>M`!;R_Axdm2Mo?ONwDzb~vsNj!SB;8QMOCa=u~!GRS~{#EXlT`lO^sGt zVy~jDmCzCd-`o5B`QbU9=lOpBg0JI<9J!C&ckW!T>vdk|d0tmiEX)ipFkEDyprE*5 zWT(_;N&GMNis0S4QFQG#H!JEaw8m|HmTCwBU7 ztgEY!3c|q%0_Eh&LP3mBZ4e8!b^FP-25<7lY~;FgKvBVQWa;|DyKBS4D&D1besv4> z)zj|D^za0z4*380clqg`%8vib@$>hlYNI6nj}!iLkpWiv+W+h2ph6}hEv*07>!Brc z7LEVU8#upSgWmuDK?(n#p{vB=bB zLE;tgX8JL0OI_$+&eA%qcZojiEs0p; zQTv&#Yqj&gEhPGZNQ*dK*o?oDjNY$rZ_WQaIn$9P?~Q3S?{u8ae|+$JxmToY7yd;# zY}4-eaJ}@MM}H=E_WtMB^DIwjD2dEk-!Gq;S|+ytjIHqQN}$KCWU5seyP!yas_Q;+X9nlAJo2qKSDNx< zlWCduE{(FPOLe5Nt-Bnf_usJQAPlFs=USWF_NV<61LxHpVov^T5xSY1PYxCq?oHGb z*h?sMoE}x5!AeV&gUhn+5DtsOrUx@s*7Due9{kVJq~1VP-xqYiag18AJ6^#*gK3`p zy>Wg zuIJ9l(Z5E1jYoFnH!={#p%X-#j2JS}pp-l8*URK-N;;S;FQQ)i58Hq71)VE|vr)-@ zx0p(?2`tNc{K~y6`7n_dV@^{+JsgJ=jlMhy$iS#oIr`dM>aje#C1X_ePbiIT`sX2u zx}rxm){0H+o(PeaBblPQ!hnyJAX^nC>G|0YAzLjgnT@9>$6O9>SwES{oJj5*Ms8+g zPsiD*!@EK@sukS(-h8xYY{%$HTI91gSaE@Ehjh9F$+_lS>Ph;v65^oF*yAoYe!|~x z=FLD9tA@m+f+hK*Np7t`#Qn=N6nDF-|4lfI(O$UdIq4K$%w^s7ciiN(%@jkUXODc! zV4J1Bi#~`ULXwyT0@jN{GPT;tbAcTvR;2^{tf92b{EjuHxW>3h=opPgbzWkhXo0L~ zdPd)+2p>eQ!KW-a{#~tH;OGY*L~1hW;Hg7^um248U%v$J`7RY{8S+1m6074SHXt&| zAlpGbRJA7%%q7F1Ct6=LLF-U0>LnXXN{w2&qD|48&po#cQ+MO#NI?3Xw56s}zrXim zJR#sx_do4WJ@J4qNd_wb$`lJq7ORF<9T@pkz6-8&B+PvXeD{9B{&q)nRE7c7qH2Lk z$eTJZ-QJ)P4N+=@e90BR8cY(`*f0A-u9D_S$B>NBo%y*EjP&)!ed`1?gSj7;p0jGG2FQ^})sm-$BLx~uMJEeqr{3!&k+lbJ)1GU2J1vWV?_V1$k>utrnN#PzsyGN2o%N{| zKMx)Kf!wy+AE77ez)^jVTgrHY2M<3kq9DiyU2MN0u$``>s%L?1<%!F10a6K$v)RpF z6fXdxW(ourMqj&2ho~4)oEe^JI+zQ0`SW8qonK@6SjNHsz0DsE%c;d#55M<;3-k9T zYm0&O6SWSmaA>ZbE)qincs-SLfp@KRz1Id0<)F6VaKIvd>x&5SAooU;_J7#A9B$}| z&?QJaiZ!9xOmw@hSKofON$eM>+eJwv^bRL(f840m{FSFEliG3Pp9CT=a{Ggr2joVT z#aztUiSws@0E>?~6FO%PEDiWpcP7Cf&N10Vh91~B&4J$NHsfj?JupqTX}W-vP%;zW zy+TAo%DdL>kI@>Ihq8)AF4k(@-g@b8G3O8q`T!t&vp;s~#-0s53bRzA>Ng>|@#SUw z@(2332+0z;e}8`LjTsK|_3Pk8)vji1Kg5y z`e2jqw;BNWBLrUbeivgCp4qTLRg>WiwW{IU2xfJ{bs+pmAXDaJe&aD7{|Wyhuz`=U zz{Z=I_NzNz0ZzPz0Fn^Ld=KmUP4r(?1tOZXL?>{0HlTG&{`NkU1H#q}z&jy2@_)eH zYy6dwbI^ML&c%ZYp%vLUbuN?yNM_mP;e0N|mye--^tCYS;GXRlR6i2AoLe5>B#o-= zkp9?&tPNv;6dbkXlmz#OY~nHP#fxYbVY1(F=!%%^m~moN*w*xus*jX9wP`Z#__2uZ zmAAzH{fRkT$Y2cAoP zAeYc#oNmldO3Ju+t-)*LIZ(5VDAjU|!p+7T%I2^>FCQ+lM%tM;sEPa0Idn-?lp>VKPm_ z={cM)7k0eWd|ah6%mJ_M8vS`TrgerqW0V>)fB{v$j(_*xq>D_!x6Wk#t3KHoFqLuV zsDCzr)$S4;MPlw`s3V*o&OO>_A7V5Zi+P!-vd+A0KJDd)5Etl6WH4?z@oYDKSCUKZ zpl@Xoi0?sgbDv3{en~zi=!VRm%1J7p?fVVm*JXX+^_JU zQiJu7VYz<%h|tQ7{f zzv{C2Sdq$EYu(~r@+N#|o=3tF-tcHSMQpRcy95K&p`o$Rx=*`9?b94*eSPX%P2Z4` z<+1=@S&{Jk`Y36PYNcydCBz&kQFY7{ZW4DIQx0B%A&;&(Eo^`{@dUYa+<(f8aIQJ;Odl2I9+YVf~+CIFZ6K#D{> ze!D21`pzPJvdZRjGa~0@5uc-kq+Fi02=%CMiA=hQ&BKYtx+bymDy};&=-Oq^2yLKZ z32=u4(7**`EDhTR5Dn)40SOU?+V57fDuUwfB_^oqGa=lM@h{TZ5lUovknL}&2LY`b|e@ZOrhpRr;gxYg zg=2o{5dgI(u)t^1PJd_3ppFW&I~DhTdcL?eDfZ0VW!bOuEHB}95TDeOMj#9@$;?ZuB+-!ZK^OH>?9w{DdJAMTdU#*CcNqUd+^|P2~ ze^+oH9IOOx{8tOIGbGVW36bySUWlibRc&H#XAetfh;%aPmHge`QGYhpPqX~$;n6jz za~YarqNkRW05v}GyArK!f%_T!=Ag-XzROc{J><>7uD|x*_=@BZ%g2msE&!7VEL8M= zu5mCMvcJ{5aC9v2gKy#C$_1JE@^7^66)$y@kM`r5s?*CftBfsv`~eE@zn#51+b>u8 zvzp(QJQ@;P`(kzv~u!f7QGiy z>=!&5woP&9b$YT*@#No8)|mbj?V(Chqk9_7dA3OzK`~L5zh7E(Uk{;?l3x zXD-{-cjOitM^!iQ^={qB-s(&~wQ0Wa*`))(b1?4ErSDDj$&rnFyWEk`Y_rF+yt3M*bCkBt)*vm$Z%sG^-fXj1XLWo+ z{lC|?Pas=mj#)Mq$9I5YYG-UKO5FrNC^Fx%R4K4T>0WT3ve&A<4VUAOcMg*RE2hWN z0PG6lYyyOFHVywn?j(n91@&GVh&q>o^IS%CpCshFgS$>w4o6u}T1X!&)Rj+3Xi_rf zwLyE=VNz-T2^@>i7|n^Wc`Ix%%?sTV_Nlh)F!i_CwbHaH4bYuK?pX_92>J z8#QgLQJCwF{6S`0^~R+;*1Z_F(v=eh2aVcw)5o`iE}1)k;k)Sz zF-nDM&nOb6ntZ)nqKzo#5Pz)s-&8(#Lj1#4#Y9whvZ^X zA$cvPYeF#b&nIi#F5z>br;eSvbG9P}j%0wf6R6Ws(@!1IKq8G!P$h7?wyOwzF zlj-dXw&*Bl`hYPUl!iUJ_6-jCdSib-J{`IP{(kr94C#b7pt>lx6#wF`H0iAo%JtHL z7D`uStmj4GquBH+H$b-S#(Z;G&)>UPy3|PAGiEqf(y6pdG|}L(Nz&%ef~rfcG;YfK z#{b37eCziZZO`6orpt6X*Vj0QE}olyLdk+W{!3lowXDdyX;YnXUV$C{fKB0ba#YZ^ zmpRqC^Ste>eTq&1B*(E_`Bz1%>P03$ALzC~42%9x{dc(Vki=M4UEHC*)m6F}}xh?*7s61!Bn-3V}!$R$Rd5}-~^0Tnw? zX*MX{MU3~K*2ZJ&e4Fuqm6~CIPKu;tO_Jxh*c!Uwb*^`5e?AP46ZO?;tv~7h;@$O( z!d|ieDMqO9>5fOy$MBu)Gok$Q>C2S4h}?QViTL&9 zrJ_e)`uB_4uX?>J-{@~V#(4A}DfZTwr`b$ouzk0VFXPy59Mjy;zm5L`RXi1GoQWJIi0YN1tZ` zWzptKA}sA6ZPXp_Z0!V?qRYm`zE5e1--+4^@cJ|set(lR>I3rDg;hHNJ8sZ!E@Hhn zY^UvD`u(Ky293G(>#}8Qz*c>_DfTY89T19VkI!y^dhP1B5Vv5w_I&aIwtT=jrcT1LH;UXS4^{Oq#Qh$1ZzGrFx3r=u> z`oQ~idN{OWd$EESRZmGPCyJ$jBjucsL~(@=WEa@>@2S(H->*;Y_@e%Pjh<`WX`kuK z;B9`)l7{~ib9P+H7rBXv?of8#DAVVD%pEeJ@%vRCS?<%hI3a2y!vUiaX3;^r9}@dzUD`Hky^{Iw zg?-icdgkO6)QusZa=!23;%8N zRoP_lav96@|FKr59`C&!i@&$iHj=41z4`dou$k%=@)?eg@FOs)u&z#f<iQBW6R!B(?Uxw6?Kc>dA9kxY&Kz?tBFo{4ue$eFub zjb_ED`f5wz#^={YeqrCQJ$H)~e(>edwn`(Nz}3j3-xfJ(w-eCs^?rcdmToH1;1+TW zo)?FgO!>Y|+>2!I3Y(AcmZ4e>0eM3Az5^UG zEf(s`!NrYUy~P3r7(tY?PonFu=YF-sWqy!e=`H@TI6MP}WoBq!Ycs$MoLM2HFN=zL z>296y$ISX!S;O~`ZFg@cpcPoFbD8tt0*2eYsh>Y`I4yBLVx6ybIaT75Yue;gGD0C4 zUt2akB*?!p{l1#37NY?!{53U#7krVbEd??~|5kpAUI%<)#F+YCI@i^&Z*uk%z~er) zc5lu2T;ft!de#AlRZFI+%$^(7>~B%)MUVObk@@tcRH3z{@JgBZX-r+){%${S-0cL6 zmnpQ4=#Wm$fhED#1&#+I1Od%u%~56Ju*s%7o~N5IaJSySgR~5ZTuS8HEQ#DE?(-`u zg{fgButF&iBalP%KsJ1S-Yb#_;g9-j>4JOmD$6nUPm48s^zYZt$fxWyVJe!`(KOLe z4-T&BHig>-Rdmldb8QJK%FKkX4Acv39obQx9ZTw6V?E6&KOz&MWs-uG`#A=~(rs6V zcV3U#D*WjB`?!?`U3>x5C)bJ& zI?nh-?`j_QY<59(d3Im<8u~zz2a-D}m$^h&E5{R{Re&Npw=g4La&+-h2{qNn+x4ba znOocIwEO@}Lzk!&&Ig~&-%Gld!Jr=S`DwRgK1AylcJIjx#n~WTtq&5 zWDxGWjlr;rX^!eNf8cJS-5d_3tL!mRO-Vl^52eJ-LN_Kx0*`5V5sMP`LbwxD+~d@KZun;dfHV&tj=RuMCef7UxL^&RP4kUfNWoP^rFQr&;-T?vA`-K;Gy}RW^5I0=(XZ;=7hSW{(%spx#1W+QXHX2{tOv zl*5&i(Go1xLU)RKvO9L0ChI}AI$+CD*JMFi9oK)S$1iU#F5;FJeq)Fiujz19ZDAEQ zCZw#N=``v6xFT-x-q!=Wl;qX3d+Q=4^O=i5r1*1J!;?lJ_OUp8;e@Ytja zBNvqv9lVPLM^nF7arC%bE0q0!pNT?seSLduyORoXDt;qqRF)!`R*rkvv=O#;+M=7% zqZuU{x@kBOoQQfJk$VqZ3uS|+zGSc|@)X0z#+lkkoKh7GkhPL7QL>44(Cb+INZDCB z`PBkA*#5ul-``smTV1&Y*@iztazr=OYedYX71@*8ry7%|jS50o+)ha*` z1XqSdbH4|%R7^B6aaF~4ps}Q4fZ-kP2%| ziA{;g`!U*%k!Cwa(sf-=q9roY#XVrf{Vv z9F8pC3@|{MSPC6$V9=U=$+?8 zoy)_W)8mn;TWci;OJ@-o6mu(Cg^`MMa9bp7FH*rva=m1-`r}T;HjUF=JJTCzV`|=K zKPf~0kwMI;sQnxUf7pE>D6O}v=7DOd zla!zoqdC{tW$AA7sT34HiYkiCuT{rU>qcHJvb@R)0_A%M`#{$}RQTv0Yuf+)I+}Gm z#=4~}QNX=M1c#Mzb{z4K+-F|<7^-^~z0WMi-4Z$c`clzs# zj{&;#UH6#&H1}n?Y&8JZ(X6oL1VEA2?Ih~5fUa3vE}#6ZVb!zqLu0+n5W5F$X7)&(K>%FeIn!l+nPuWD@i{y;(9XQy9Fnw^knR z|N7c(o9-SNM8ORKnIhN<*qbPSJ|rCGhnj^-f?Poc%PoXR>kA&RxScdy4v7mdRs?@B zWTTxN=kF!}jf9JPIQ2aa4QAj8NpPka6X=T@vLTY;mvz2~GIR zm~slsuK}X3;ZsApv!n#pgxla+LL4Cy`epCe=~A*QwR<~z(>ndMn=@w7fXF4Ku)+E& zB+DdF$-S?|pe(t6FM(d-!w@Z zRhZ~tL9tp)BtaM*t2D!HPRC*f)V5a5++Xh420w&!!{qVYS|G&>Zgd41Svef z_%HXFTM=ADwn6c)2h>PvBqx#;n;gngz2uN{(`*sFgL^BIld#}3-Bii-u~*L32}>7~ zfhAZ^t>SZEXoouPA5>M>?rIwSg-{ASnb=KN@)-VDWAH-sNW*%*?eu74tCO1fZ##S) z)c;Hm8V%{V3AViEI_);wBKUlabF=4bd3=Ug0(99W|6;$(-O6EKr?ryb+t-sJYYXLH zQdJJ(nN$bTJPpx(nK!dCT&wp25!P#8Fk^>GgA~8_);s+kcpJKK@hWR*Qq**RQr=e! zPH53Q8w=MX@Ycr}kpy#|`*&~m$Gns}21xbJ`VV2`)1AfE6b#|n&g$`0CTa$P6#aI} zG3zzZ6jM{^+g=jYZHqqc#n$t$WM8N;*rY3E>VRsqwR1qCsfl3Z^-Puyw&s*GZ(c}% z*74{^gu+JeR#rPtf7;IC!({`421yX1R@%GX<6VB4;@uA;V|Y8xE=ZGj_TWrVG@jX^ z-h@u>`9pQPW$xE(V|1FIcY6I-LujG~hOVD0w#y#*{pAkq1+&(+uc~Z}1zfNBB{5i7 zzbm5=7&|60-er1>y5zg(96{eDZZC-S_rn5;&_KXHY%%WT5eKUA&CRoS_(aa-#O5IIz&&m{bn72IwZIZ1gc8Ep zpyd-x$DL{#aBD7A@?Q)8oj&d|WsqsDK^eyW3x54t`n?2A8n|G$h{mcXy&{22Zqj6% zs*8|bii=&39LJ%1@qOIh8mA+Vnttla>QM-73-HjB3ylt1imxwh*A4d&_36a226eeDHRHYH}IMf5s13p)HCf5OLl_k^mD3~44+fQluaeHe8mc5!-3b8mj(JYHfu0j|u zU#m2VIo|7e;{FQ$SR|_n!ILX#h<%YN^KRf zGb?zrei~tpg+D6h%xOeM>&(HRauf*x4c^{_FGW^tFSLM$PIQT@IV=R)BST9>hEB<4)7k#fg4m* zupY8yP;Y{kO^}h~i&Rc`5zuw;BloZeO&f!*qQO+b%z3|?=KGio>h&$+(+|LwBEE~I^?c~C^Jbe%+dp93?|_~+xC<6pTJ@34KV$YKA++*d+B>r# zN*6kO_gjaou-2-hdY`}N3oA4hFu{2-q5B3 z>WZE#=}NF9$(d}6y%ZHmKMJLQv+)C7s75CJ`o!I5PS~DiH(tCml0*Dz2oirIH)+5# z25_LBiu3{f1l~^cJ6*p$S z%bZK;h?B~UO}fP3W%yn63h1TQAccsd;7_q=Ru78KGhvXs)3UG#O}1m_tVZ-rRq}lF z@z#1&yho-=-%KI?t@YvE60N=qzlLz(xrCrT9&fwmfVocuhSq`w!7Ii4k66UR;d)*3 zmnLGFJHrH-%2u6lDp4;~+2}I- zx(+M}2+(b-kdHzJuBZGn*Y0w8)9kBF-eMJ0?tFE2(XVDHLp9x<(V8)eir%_36{5|^ zIYHy>;r<*!fdbJ?uFAx*yJ|2RrQb;(1^3BkGSbT7(w1GViWdw{1d3T)yNV{_ep`&B zx*E4V`bTcBmr`a0Kg)m7A7QlMCI4D3Uv5+GxyMuOH!HD}{M1sWr^51G1*v9S^-+%{ zykb^1lo??Xnq%RZ3!iSo(7m$z!H>hjlw25t8-@;0`+s*`=~U@l!liVk zh`8cZEeMUuA<8Tu)u4HK9KLs~u}=BAav7*G=%YNJQc*59G^YZXqt=VC_#*0|3Bjn@ zL~D{|@w3!)&!cvYDk!doOaI{0rE%W;NO~xQ0r^ubU*WVb5C6l?3A!aox?UJ?QBz<^ zZhaX2T9a+k{0#PP@|#xwj(^LNkl)v3Z@P0(^an!gggr$2K|P!(Hm+q*zw^yp zmqU6#McmN`1a(_#O{t<@ml5ik2~cSLce_zDs_W1!^lENYK>JkDP3|gZo20Qj`CI)U z)i5a`u}!BcDEz7W_)$&c=&nw*cy&qme8l{y9wS@z*WABxFTjzdFQ_)=fjI|f{fbh;66{3e2{ZV5(AA1Q{WP%X=K`1*x6bOT%ZsUt{XIiw zWLj-{d$dR0o)fyfEGYT<1tF1Wx?q;$UPZQwr30umMyfL`H_n`jwm&X(N{LE@% zgvWYvqqGw%ty3;k=Ac9ZpCLD#B7rYUD|p)Xt(33WSPPW< z1#2MtyCnnS9kQT!0!xh`R7VGl5+{q1I^2X>}_uyX3NWJ%YKAtw!nW z^TF^c+TqZ*Z`PQl{`|a-m2Nl5!amJSqU^9OSaDZCiWp_Y_=O ze^x$1UlTuKi~{YgAer2lvFiK5>Are6Xf&guiYr|sfnN?p2BuAvFwA5MGHs6mxEOka z7>S56T@#FpZ(%HxLbCKYBNTqL=b!bTrAC7-dg+7NFkta@cd=W8&ra6t^t3=OX)DR4 zs2$zccI~`4wuQK$D@Vq?I8uE8E`U3jjP@o(f(r;Ic38YSWT!cMlO<2fE(^qn)MwR% z;+BLhow(ElVlqjLMt?|@=s&=;qqKk)P75BO`zbXaY$Qsi?JrHlpx0T5m&j6mB1aPG zy!{Mk^e;J9;yUzdXckg_z638zkaaT!+UfME454l_giwN$a=3Wf>vDadOgErDTq$WY zZErt)G(gCtvM)*A=HKbn6|vrK3$O56E#bnBs=_N`zc66#|CzE~T{;{J*7eJMH1X(5 zrNzJ06~kKPCr}@3=&kp~sKMMhfI&`jOH zc-L<+!lCnoHy@)h$jr2TFCyJk9vjRlYbA%0yUvot^1?MzX$GMXI4rb@`>Opn@$1kl zWhM|0w=xZgW+30oiP515-2N=8{HH{w5xXD8|5JaDS{%q8x#$BG6zuFgXy7B$vK@Zq zad&Fq82I^uk3LDOA6%XvI(6#vz$4qh;EFWOAba|9$R@-iF3(^?bDz%ub)A}L1`(bR z3oT%&=_k6KO_{?Kqw^LzBjBAlW$5g3tMI4Hcs*HAa)w7*d0fueY0#&`3-UV4enk+F zJh;RnDs2V#2yo``@)KX-X`{J7lN%!`2*U-0WvLvb)6+Bhw63f?J`-zv`ax-!byM*p zR0+YIU`bMTYE4RQ`@PSjeRV!?L9^+`TOpg3KCg{;T=r?g**dSDykqVW76?Hvc%MQA z#QDexx8HGo+$vZ_CUzheqIBIMHVuXkv@3eCSN*U}F*wbyz6BpF$;#f*`0Bc(boX5A zsqlk}+cZhy6lFTzr$-C{1&j6(K!f3U8&a@tgh~lS+)gmU6Y*0`175p@p7i82pVAeB zo&*ZfGK_yEYcEJ^LmoH!O8Jy(KWbPg^G8($O5cxV!r-S~eS8?y+IzJbQ_129x9hrb zBR0Mhof3i)!#ClvzX`k~1=H-j#kj2zMLuApFuD`~GYXn^XdG`t!y)X{jGeq#nhm8! zK(LzqaR@{vCFImzoe7jkA&^4BI5Qk65+AgQw?(!-IHQ{36Np=c#9$`jM^{L4*4IQW zP!H3ugSswJ$suJZnflVkFk2uVg79JO-&6}EmyR+O3#pXk1mneKjlDhQr@?vS)>jS) zOXDZB*g`?(nWTzThIa4|yaAF8`dzDSwYe*T>-e)y6HiI5G|3-!&_g1XN= z@-u{1NtRyy=BSqa(Cofa$UMHGN29*qS2&qDwk@TCu9g2R9uI1#4Vf8k^%%4IfYxK$ zytOu7X_hdO1-^_5gr2b$cDCPmpg&G|N{fLaBE_5C^MqQoDH84I;+U5CCb~<^=pqYg zv=|CY_2}F+5L%}ek3NSb+!8?$_%_GRTZzqwvms;uX0(M=sIGK|WexHn1yJxzns%lQ zR^eqkieQFJnttw?q8;5A`cLeUe3@J`cCecGtxOGb(`0cRdZFU49%Gt6JcpN(z3ei! z_|=n^I#Lv-fm;sR$7;z=to~-MHCHzl+`m))oI{hgot4NUewSGMDXSDMqrI8#-~ZN_ zD;87Q^90R>@qe(_snrusnW@yzCy*(Z*`KwSvJwYn5}-ayvSX0r7=`r2oHs2GObGe} z8v@ZC@lO77^mo9ilZ;!ZnCp>EfI47Y$;ySD8Wdfs>de8a^g7v<8M-N77}l*GGg3*8 z?h|pWnU~_&8YY7`RYrd)aYLp@?z$zTL7hJta4or}4M8W+`f4zARo3lh7!A*?CzA)% zy~&+1k0wqonK+oBXj3BlLhcCagprS_JQv@iLMUvuPK$mO|vPe z`N}+WE#i#=Cft_!%h{7-t8OZ*aG!u!=c5jsO8?<@8b&;$^L#ra$7Wn)a=V`cs$ky*y$(15y1Fm)QkA* zv^sSUi-OI-*DZgft_-8f6S}W+YQ^vCoXDr49wC&69SB5@=A z3CdWvn%fE4v<`XnL7ZeT!iLO+r0N`VlP?a%_}*bIMO7J|5BWAvmTwIPLeo1>SDVLQ zc5&wYL@K#L@=o~J%m`WTG~)7;m_o?(foYFHp>kJ;(Q{WTjnqu2i<`-}Khz6p3>XAk zM;xBMl%gYPrBhMi(1oCE@c|-0u>>Wkkh%6X_!4e{Ta`;T72;rs2v!k72q}t+0<$dJ z(Z9c_=GvTRtxFN-$xYV8W_WaAS@i@xiaZQn#LMB@>5y#5D`f3cR{n|~ z2Md{<^D>G5*pQX3ol{47KWED;3gkknLU-d0a^01^K;!sm^5sm9bpr=M9n3xW zYS?B+vArikvsC@INh^i2@QDbco)r33eF~b%DRUt%FfN8Q(aYLBRq<;5!X`73nsusj zv)8Yt8?)Ggi6Obe#7;}2%DXt*nWI^blk8G@a`Sy~>nV4>Oh8M~9{e->4Rba92Q0S) zIdvt`#x^~d9`f<>XhE+(+8% z^n2;*kUR1>VUit4aH?`)_x`K`=#dZ4c$4qI^S77Wjxb)I>NR97ZiVCv!DT;;H{1nB ztMFFYwY1wQ-Q;$ubsfgI{&EAmSIE1W-KM1#E@0(O<_?S1PTl4Ip|!K3F8D^LMz<3+ zcrV?m(>0Tahx@U$z58xyjsHaN?@n1d(FHkfIsQXukV#WP=jAuAB(zxs!8-e@Gx0A| z@hJ@U)+X-vi-rTS1?kM4mwT`G`D*`)4<1o;ofR@Gla#sZ8b=>0I!hlU$l6u!?bKdq ze4Bd<#4n`G;E8el$%?rCI02kt8WNr5fg^S;3GKzMBsJgoEOLl}Am~D|nrd+*0f@@|(vq5^4YLtoc$~<+0YS41fb7%@C z072756jGr3%KoG}NPH?r&$)>aXSCc3reeuYDaCzVv`OdJ~U9#X)+M>yWtH zxK8T_sw&aSqUZ3G5b<%UF@6Dch8h?u;A>ai{9gG|7x}K9 zXaOw-CyUkeOS~>DZJ)hM6DRT#hvjUQN#gjwrs)bFGPX?n)yeaUZQD>*7 z*DkDsGMIgkGFSt}N>VqSFqp|34AzW)cSrW~^GQ?l!T93 zBQwdCtQgITIm2T)n1)`bn0*t^%wI+K&@729e5PVni^= zq(|H?S(%PVi3k^O6C!Cl_IT$@Wc~0L#!n4zc(`yi`4ybqXwh4ig9gICKK~Gtt5NS% zu-<>rDTX`Eq~B)Q^qMfDQ!C(RT*z?2DGC;KWx2*cL2zdse8_BYzf`4N;(`t`@n2<4 z?>;6F201PD7bM2-+;zehuAn-yrA@m;#h7$I32!S9=`^#ndc&>aIAb!lx=+{k5X`}h zJuW@hDozti4Un`BrFIQ$|g8BLOmhG(#}kGJm@NqqO!Gr{5k_ zBybVir_t~?02s0g3+?xWv6ys~E-H`=d+L-843E}l-FT;LCv;`|i9!LiMUg$`?ln?f z0N8mrtZ3TM%7*~0G;oX7t*;Dk_ouu^Xc@sgi^>_M6L~Ki7oxrB`174^XALl>MkgRf zMT40a%Ea1(zkId*GSoB08-{noMVR+b(Q2v{2Mg!F>NIQ+B39`%4#UzPH{Jvu!Y5yD zp@MPC*9`u0)o7??$%Q+v;4P%8lq>oYb#d#m;|g&(nejlm&P=E0sD{3j8%?JZbnPYy zXd+t#*d{Wj&On!dpM!xdF#LCZgR;OBuXVFiJUWTXt$3{>hJ<^68@(4GI==4n-m`Y4 z!yKRpWH@`N+&20f_~WbiJ)`SsA;PJv=fBBVo=3m)F}c2cDNa?qFYj?|d*n*S2zEhJ zQvDP&7HjZcK?|uvz1Ei+@n*F*z1oR4j%w`R7c8talanA4>KZQ8DPFPurcC;j;Z!iB zdI8CntFN~mm$hShRi@@M=i0;Klr_ZP?+}jbZu?Sf>!gb$dec`5SH|?N=53MKz^MAX zf};MXJxQKA0t9%yKK*@yCf3*S+1L_kz$=Wkmm*pbA(R;7xZd ziyKQcrKZ`ejI;2sY^LKRA9Pfgi`#CID`aW}T#cV-X_Ev|``jD=^j*%P1wOblE3Kar z&Jo8&CGeXtLyAEN&kV^4@+9o&kb8(-$z7UR%R&Bx#02rA9iV~^$)c>98IrFJavP0Ujd+H{pI zd%RXAR|jA76FE=Vc(V=IQciGx`-cQxY zH}zT(hV`;T4_DW=i_fN?1H%~{xNtpL6IOFYSJAw`Ik#U_j%&Z_Mk-@9!<2;Jde?in z&}p)PBRXQtJ?6uE3V|6;>2q1{jF5>3-zM)ek%3t@qZ4C_R>deY{(28gB*p|X-AYmh zhmtO=ouuRu0^L;~{EqdZ^KS4yl$u{}B~n8vm+E?k8m)cBTSm|yC}Z}zB2n)Bvdkv3ONvfcoL4zCxdl3eM& za^UDodWXN}&}V!@cMo_XS;lxR!RliR(Q2^Ix=eCv8#Rls1VXLB4_dsdN*xD4E1g|? zJX<;+qj6J8^U%AWz~VV0JBW)5oIjt%&`g!Fy5%MgQQ12CV*B|EsJYUAUwW3$7w+ie zcgxcCyL*$H-W!=uN!qqwpqZXWj^uD{;r9Aag(#)cyC~lANs}|PQ>}}IG#z=9kkjad z1Hq^hQ_rK*u66)SCME7U?UygA62s;B(oj7s1Kterq2tYBwf_%MXBkyxw03Pl8iY-E zr*ue5x02E!-5nB2cXvyp0@B^x4We{yx=W<%Tby&=@%_RdICSr^o@d=L=QU%Ul=xSX z7W+M!_YUxuVgl|o{|M}@s?RmYH?m2UkFXhnD%Kq_v##%N6MxZQ#yN^&!6heI-bvSU zyHOy+(7dW-t*68uhm(XrFd;CUUy_^igh^Q}_{sdL%Qg@8pK&~jsu!ZS^7p4Sh);tfh^7!B0GJLWMi%*0wS;V6tGh~i(lUXa2MzUlaLqzKv- z=Ia9*$E4musJ7^ARs>uIo42q4X%Mg40xb-&$O~fZD)az?gtuCTkXZO5Ul#W!9&gXYy^0$8Ke19Q#N;FOS~wR(b=f@7VnU)9i=jwMi43@ z)q#4-RoPZGu;EtCM#9$qQVeI&4+(YMLYwy%$M!vmV1GD`Zw%$ps9ZrV(hHN|)9{J! zckE%+kfQh^MR>O6T4*Q`1~6cpwQQ_@l7-#zSPO;z{*FC6Kp)kP(=p~y9eYiBEn-W2 zM->&~=$YF>!pVjqtSvqzM(!<=MKg9)xV%_s9T z2kgE|LotXrrJnDFdQ=^xv~%}+U61EU`m*!;G`-wE)+&a+-$gbsjp{5h`@=i2XmxIU zbRIYS#ffk;>>&m!~O z$2pz5W8#HR8>^eT7sX&It?LyL#x>yOYT^~8Q`b$yA5QAfdgg#@`Y6-^hnpz5<^L2j z$gbdNz`Uz>`lRZzx`mRsJy$G$tMv8F5ULYUNbcxHC4N*$)@GdRuSb<%`OO)JX2_X< z$xOnH5Z+=H$7{IKX@+G_7*e|HD&nuSj z!;(JR^*f@&Lgnp!GEU4sWV_+WpC-nSx^vwqMmk?chdXmtVFvu_M?b?9r>`y>dMpU$ zwUf(L$lz(mCBN&wO!%ST%H|VT>dH9(_X!Ry?K^#^n9VUjZ|sQ-_?LbPvKI9VLiVZ8 zkhg9>52w#Qy~Do}A&&TZP46K}5wS^Fy@dHD{`I(Uf36f-4CJ4YtrevbykUU0TE!Mu z*iez>XcaYAX0aMzDKz=Uf)x;Ne~KZTdM%Raq<%s~3&~4HTaJDbwiqs9$lbUb7pve& zYt1S;c$7PrfW}Kt=ml>3T*Yl$DL0k=7jR+?%jf93x7!#|CA$wpLVV|raPEMhpVxHDre`H)= zFHx>TuEM?)6jr_Esqys+vD@bQbPv^nHW>BD{K7#E$7&_>Mdcun!}*dlyqvc9<7B z8iLNUG)WVf86K$CH&JS|o@t0AMq7QA)wZDpVTv4!TgG4Ejeh>=7k>jPLXgscudfmR>YN zbio=zAe!0RUReeN)~D&yVdb3gtW!qhNZXK63PWW`_3=ELPAhgNC3RN4H{1uyihx*6 zc4dF|C~uRNA@NXPB_1JH6=jwQxm6N**yQ@dAL1nfLmmSkz_n152s7dFfi`lZPx#q? zC485?F?Ky*_-f-FD#pE{PF>Ad*q&An1+IyxVZ_G3BUa;U4_lb(pH^8L{-j`iI#;jJ$7Tn#{%`PCjnW~yb5JL5JPx(ApRjb{0-zm z7(BM|CZk_SB8m2x+sTB{A93V3YF&E~*9j_5)Y`>OrOh}P z{iO_ohRrjTuEo4!R_Sj7JFY`%enS%_Q5JXdW2B^1#_UALgEQtgcXy>+NdkLlUgEpU zy$tip#*R2K<}^~L>Ku82kGI-4r1qzt9kV+;+{7TIORdT_H_q&DfLx)+ugKxRaN?xj zNv}*fOPP&g!glc2t*Qh1V=)NVk&2g2J@b+W(XI^sEF|5=+QGf+-Yd-T^ecE&L0h!L z*THCB^!#afoanr0G7vHoC$AS*yMM^M%1L9;>+)8KT{n_mxXW|?P<-#V)ckUuHev33SfYv(fSMPewYT!$%W109y zJOjpy=pRnn)-}iL2eB0i1iC9W#unuiUu{aKM4DIkFJXh#xQF40knv)kq6pVqt_hJA zWv|1Md$bVGV}~w^U%0VRVcfe_TaKXE6405Sxu2YXF~AN{oqy(e1F^ zb3fQQ1kE@A^V)D!a~+@EB8)ubp??X9@DIEZ)z?m>Zkp`9Q3uZSXxUI?_aH73-fgCI z0*io5Ka_z~*QSoMWfGJ76hO_o4tEZ~^vZTccB~{zJT`4O_g*6ZY+kjjrYkPl>%cA& zS*F0v8l+dHAaITZXGS1X?BMh&kv1CRJ+uHwZlBJAukCbDJ}98BGWzhYjs#TTuJJ*< zjC7}BU$!&*r}I4xyL}KttHK&c_wJQ8Up(_3+TE@R?s8WRkGL_d&&ZV*yVkEqgdw;a zymIoIH-{A^A-3h7F(Iu!GNI!eh};ozVO8-%DDAoFGJv{c>i3y+dQ=+Rvvv%*ab~wp zIJl7HzhLX;*t%-!Cot(xK7%e%iuP+&eI(H=##dHg9_Oz-g3F(y8z(Pz!bMo@ZW`bp zaN~-BU%R{ZDb92{9%55uh2unKbUwy0E0NAfvBpV1{%CA9$r3|k=C&Dc#L7i%8Ge7- zBX+2k0KuN+f>AhjueuE_@`CVlIY^AM|Mvc%MyB?y>wicBLoUj7C_fI<_Y~YY7L{F< zwB-*UP6_2!oC;Jm_*9q5+XXISokLJcY0Q;_QFI}?N#6bFy&>D$VuCKkGyNuAO9Io@ z=`NC5fe&MKm;)Qv8?je8J$M8Cy%-+b=@zX}L{H+yv>lA0H5wb`&$zNwo#t=Bf9}Bo za(MbVflg6ik0vMAk7d{;YE=decagOJ#;4vN4mSbkngn9ez7|5<+95dwk86I$aasd; zFr0<%^gi@RZ*$plC7Ya0@UhMfhSkVMK+}3%=QSk7!+x$e&k3u+)5!%4Yk2Vnuwj{h zxjH`bW|&zcaLk+|uSHfXVL|EN1J>GqE)O(HPz;PD*q3inHm;9{8l9gFye5h7xj)$t}ImBmic3;aS89uQ^w(rq~^m{gzNntkNPp9h-auXN}Ei^vgxDBKJx) zL@z!}%oxFEE!7|d#b~sG#}aZMY*&xWvW*&e@k@xp?;Grg+pTwzGLqUhrI_up9Ph@K zw`rT`8rO;P#g|@U1mXRIU!&(k1ea{7r@Icr_Z0@2x!LN=x@*N=scULletTd`kmk60 z*Y{mM$=owtrtx{t`CX83)l5vTfqL=c{ol49{CHjn?{1sqleVu?gS&kyijF#GlAw&f z04L)aYw0j(?nz}*2Jg@j1(d8udNG{=8b zc^-2BI~Oz(2VS-M6a3yB$Avh5AX50H7Y$$KiQfq%5#J-kg`{ZxE4s^=SByzA>ypI^ zp%|*mq)MqozPLOZ)TQ^GN9-9jrT@f>GnT_pAaj|2B3tawL~Ey@%1w*%Tb2!gxsqSk zypC!OnGEOmO`w{ye|*)Oj+uPc5*8Qs1M_^HH_Vf=`n88$jgqCTu`z58tuS5#zs4lI zPtHTnirEt%-nHSiP_OdMljUOd7zuBE&WVX2zx^T8K!)9GlwNaN{DEMT1S?^>M#n=bx`-^)z9jbjvbnE@S zw+U*{2bdFu!p0HL2=Dp`oL9h96*XtA@jNT+2f^Yee?p+lSjtJ4*m{tM237*V#*eP> zu3)htPILhj$zBD9g>E{*q61~^e7}v}f=`@gODW*1{tpWcp7*F4M~|}E=}KQD z{rI@#7^dr05|3`)7=rroXX(FQ2}-JbZ`0l}H4^bn9{wGX(Cx5woKR~ySL70)W43YQ z>tYO`%0xU)-;YW=ZEi7J1Z)P0)bq}HhT~g|lP#3DN&pB#rpNSOsM7yG*8u#%%&6-a zToysGndKhf7y25*R(pr)b4#7dw<&tMP0kU<#%_9wq9K-q4uC57#(A<98fv14bt;&q zL&VD(^~#6Nr(fUz&LGjxvOWFJTw4=OV?dM37_i^gR=XRV*#0;b@S(Rq9-N6X-80wUO%%*d_~Ok8yLH}f?d9Ll0Les!|1945Sds& z$=|hxRq`Ma- z-*(&R#i*O!e*XaYCQZ-yGG_(HzBZhkZ!#D)sRrmz29Wz+x1&+kMv7}nGt&Tku5l#;*rENLb~6OUn9{(yuDD{9y-l8JcMYhc|uxg)^>iRcJ~Sh z%Qz8^#Fhc*TVxxK%x8txXXpn_#0ok@(kp^nfvp zISBF;tV>OGl3>a4db87lvllHDmN@bIh8DMLchJ^hNrm=svM0(s&D1!#kkAE*!ObE>${!x{X!A~fHG*v?~dIt)lp^Ad) zdN#Lz!*1dX{V34i-9C7+vWA%sqtqii`7V_YQ!JXO{P~i;!gQ^plC#4{VJYF2baaU$ zqwTF!-@7{UuLZY_B|cLPUu8H7%c_WEk5(y!m|j*c&)^ZjC->B?|D{U)()fDvTDCq= zTRV_dQayy(yjsJ_5LuBTu3(eXxa9Uil_}i#_0dCQtdUwlaiBRo_EI!)Y6a49tcduF zNKmzDJluhSG8J=mwu_X9diDfv)ME7;$KWD)(?Hyq7=`LdZ_Th``1WCoob-$1XP(OY zNUy<9QuqFRK|z?s=(_+rz&BIt4E4YtczVd0hj#t+HASZ4OZY;-cCWywyQUgP6FnKZ zugQ_C9e&+1q|W&rV#R$%>;^1TkY?XyYG&Ow|>OBjDYJiMCru z4qAc-Bb^nCFy9lA7*F|4O`B?v0@G3ZcNRO1cd%wnKNE0uS-_NE%^yyNfnT=et32w7 zlq=nVg^tYoKDH=S6wT=O&M6*39IR3wRnt>Mp7eIhhbnskkXO$V04pZ}Q8@mnQp}TDC8&By}A#7y0y-i{TQ_058?I0U1#ih0~DItmXlFVlMGI+i%0a;wQs!868WCX z)9$y}1z?Ied5CzHgVDPs{b_jdWz7J9CET7Vf&gxzYD_MJ6rVxySZGGUVhSb_aCM-W z@~=Q@c$t8wrJ=8F9A^geJ7q)Y%O|&R+~0*6-nTz*iXrpe*KzUkQq^NTsQFCQ$wNBZ zS3m3Dy4e#~Rw51%dS6eN=IFi1s06@C;i2TsKh?D|mI?s>B_V=AZ08K-lTY0& zSH0L0364&m_KZR4oPH#JnBt3qR0tt>y0HNN_~)wWvbz>;2Sug{Wic=Xm9Xg!kQ?&C zM@=rPRI2(WtKoe)AIHFMEa{;HL>@RM^{4xU|oM|1~CX0^Y8HNwCS?8J|nUW7(AN=e-rJ!D&P_|+{|fI+6w-3l5XF2HrhmAN&$xS!tocqdH(Q@rqFZw9aGNi>HF^ji3=!d zNo4)OP%kT-wNEq@Ci|%d-I6@k4*?%h2;Lh%M&ZuiKXXLCk*x9Jv*;Ywmox+Su629q z<^2s)vt6lqNk#kIjLw0(&AfB{bn!)~U}{34rnGEVow!X%b+db)1;Iok_pILUD~ZD4 z@5j-!)MPU8Bm~$+?>4%HO*$R{@p(mq)+ra-H1u!}AZ+!jN_c3A^UNG{klWgJBZw1u~R>56P zi}5#i+ZoEVempF2{RGZI>TnA>wX$W*k1r?Q92)>sN!IjtUxO6oXnz#2w&-r-y)fn= z@$ncGF9?vd%BzOMrGe<4;pZ*0wgjPB+*9>uzhHP&5_@96?;Iy!H}tb>J?o$OE|>VH zWCZ;ZWn%E8{cc>d(V}#udF-Qb;Ml-iY5VQIfIYIm1(4$nzlOGsTTbek${V|%?ekQ2)0V&O_o9e`Qm6zWY${vd zEtG0vh2H_K7h`Vbrw@=%6@Ydvg?RKtr+}i)T}4a$21%Uh5N+q!vkNCxqt3cxuF;8H z_wLeenOl1{ZDot2%QGzYMjoyu_{5AgRq+K<|IaBoR9kx!q*?;~Vkvt7s+P!C^E%E9 zaL9D21aNVSR;7qxlmBt=DBKtr$&D!i4XoW4-AFKrc8+CqOw7nFl7ZxK`R3S2n)xTq zrF~bEwaR60Ri3w3b3N^9copXMi5z3`NmQzQmECrmk%et zJb)&1F!duc3&6u3>v&f@dLB=n9w+wy{M=YX!$ai#%VE-g{Z1RMZ2e^lF^qqLbTrXt z4hQ}ZcUP|Fc3 z{k#5QX1g_b8BW`-_ikmLH&lB%GN1H2=N;8A!k&0J3stj9j080>kS@I>Ol@{O(0Mk{ zw6W@HC0BLIJ#At?Gl^=*1`;~>E{kpj-O*tR;ZYyGJKlz^MjK8%t}cP^>j} z)egX*6x*m#UUuPj`OvKt!JdVCRxUgCywZ2hkRud}6ps3$s1toN3HJ+Z4Ak{4;fwa^M9Yp8%{RnQ`b#+#OM})9i`HrAU|H?J+ zOm!L;ngzc~`YCQc7(YbRSle$B-JjGhaQzcR3~=Jm?8Y&de(4CM7DW6P*Y^xO1vEhb z?`HH(>ey+MQ1h_||16s1~cmir!`n@BKnPz0wt^rS+Jz`b;g7e0GP8#(>UJ-8o4dST-R9ccZlBF z9Y`lQq9T=uqDOE8Q_?@*Wdd^l653I5WNyxB3d`5CTSbOdNn!3S^Lghb73j?Svk#c65*@PAL{7mT`M|>?$ZMCNVTqRd#9~w|g=~j}d6#i9C@M zzcSQK92*ZpAl!i}D)_F@_RVDToq4*G$Y{5?3C`)*Brgxeq<0EELmMBP!u(4;H#!`> zI7wsw&0?AwcA9^gu{gi6ox!C2#)y#3XTXP_-YCKFK0&PDdq-9 z=NgN|AhY(F-~7xn@dQBKcQlRN_yF(-u+VM4c|A@IZ6TIMDk)Lb$7bN&r)F z8#*CYug2(!q2`AzB#w!G;vM_oZxIFHn}tmo@B7E%fuv~Ob2gCi@ypMqyHESUX1zi@ z_X)z|aI85#0&o&mLXCZRT~C9!|91$oUq90slMiGa)vKb2@we%kXoz^wE{%hlK{Xt{|ZA@&D-Z~!+4_T``9sBsQ;7QQqnY`NX zr?l(=!6O_=K`h;5mb3Oc=2>Jy?PtW)Nn4(z8(5+?ESn#%*n%h0{O-h{$(RZHb|1LW zI&|u8HEjXvYAsWOL7GR?Y`H(ZQPfhlfAMrJa@md9ZMMspn4dGOFVmoMl)az%tqu1) zA%&%Xpu&&g?il0>lS~ddanGvm^5n)9fDN|=Z^QP#w27m8s~L|UgUhJj^;a*o{Pp+v zXd&)qHALw_NLo9kR@V3HV74L$Lo~~I__hAhaPN3E)(PhP3ezVgIXt4+`w&KrU5@lZ z17Gi&=rDs=pS&+%i>k47VNyMA`OdYZErUC{0P-lr^Ud`p z&cZb)jnrS<$?(KMp5Q5eUoM8iXV8!)#CTrOh0}+X$(ovZGa7Qx$1H2>8>*h^g6M;p z@{egU@P}U=+L1rLTJfM(Tlp_QPV=ulQ3YS2t-1yLlz=0&)4yE+l@Kow;niP4LQ|bj z0Ib7g4E8OW6^_{AYOH8(52`@}g+ale#QqU{=9;SMFh z?Wl=?nK<2e3uuS^A>(ZkA%1A$ObZ#4K4*EM;+Jzky5f3d?RGg&*;C5=6A<%}l5vU* zUH+)Z{Lmj7WdcHK`m9e09{~Zd{jZr(E9hrt3KWP9T(LR$#LEMe#}Y4%%XI4gvPAky zTG0o|lKj9d{Wb0dkScDVND*i;dztfE8?T$UT_%~?IdylIaW@TZy-w}xnW%^yF+qUQ zo^tvN6%l_f9i#+H)mYxBa8{(~Xy_Vr}H z!(?uppZs5r>4Hi&_hs(6)K^47!ymT1wc|xQPb$5C(U|x+kEZzwsC{&Y#|1VUjFOTg=QxgSuXv3zSUa?$9^2rjA2{)nlEH7@cw zo92t0k8*)iAFF+|P*=wL8fK&uFzC166XMN&IOVdi`+$}8HNeiq~wZR_42!H&jId-fte19F@5iY{FAR}t@krXD4`Ll0d)hwurGK}%9O63 zx`%hF$H5kV_b47j1@xD<{(Z;mW`FuYdHn*AZ0esE7MvBDA6=74oq93bGpiU#)i|W! zG}U!cW?Xq$bWqF}+z(Y3%zpkb+zTrkud#YPnC2+b6`JSD$pXJNZ;^A+JzSuWajW6i zx9q~7O5co#HT}gG0OD!Y>rZ7^TTjIofvx)xAa;kexS3W~L5#(wH%(BEGVbxYTFIBsmTo0t` zkDGn3U6<*X6QZy~j-C!Gq>i-|9eMVfee`4YTg+gtM6zAqZdl@D9q}&XM}{5*?TGNe z_`p5QIK4+uQ|LGEDIC4x2;U*yp&XK$_8ZnB`U)rKW!_nVD{@nC3|IFQTs$Dt;^;4cJOD<;wH|viJx)Kx>7IzKHUU> z)_(y(db67(0syEJ`FH^3x}F8z<%45iG`h)SdCT$N7@{Lrk2#eKe_|$oSVpxYz!-KL z=il+D>Ic`yZkAtn?hlIOiWxl@1p_N%?f>HpOLnP$lYUhu75RixzU0y+9gD6lWP1F@ zPN#YXV%GI;O^r#m!0toA2ifNu~PE2yGxK8BKQoR^e$&_oDH;H-R$HZ>2742OG9f^ ztR}=T*J@na9{~G@zR~+@)L>KdPak_-9{`0v;^=s9c%8HKiz5B7PObm_Xo7VHP z_o;;TEj`|X$d0XiJR}{~)7?#DuH9!$B_{nf(KiBz)>IdEN2N4KcE!!IoaTHOG>U`|o+XcHSUHDMIb42H{IlQ3TYREq5T^k}A! ztB1LrI)v_lX}-L!LlCyYM(pW41!=F-4p%AKu8M!6p|{yv2T- z2t|PeH04#7)xT>~7H97q)^og=psLA{ZbY5sXDMll{4tLAlXvR+4x@(rwk7LRM_qrv zz0B~{+YS9d50pIY7%t=RPtOS!2CqY|(xUQVUrpRTZ;+aRj+*z@%@~QUQgdA7>7Eu_ z-y-{SL4y-J+&n9sX6>6lre=E}X$_3V^JxN@j!VEwrKyI#;4>oc?aQTDt;Q#3^4#fN z0QTKIfk`(Mf+b{zt6mopKxwYDxavhUm<@1oIo+kTmzT+=;uVrIH3jrll`9+*~=ENpD+!m05o~d za1-jTWH`-mkcaV~03k8m0Kf)awVu^q6++^vH|hz<4Yt?C{ngBZ%@@HJ4p6e%Pme$G ze|KuXeeybnryXG)rlo1rA*vn{ew;fTVe64ZZ1^Op*>X$-%VUO|T*r6y<)3U@Tt0gj zZfssC8Rw!y2Zb8*YXUt1eM6@+|MXupep1W8WF+(3kvR*tau{f#e}^W0nvzjJi` zt-#w^MX#kYODGB_)XK?4Pi6td@rm61)~>rBe`Ncd(mu|11l zoew1tNOF=iVpQuNZtsE65qNWy|8 zTk5D>&4(7(QlJ=*BEfR*^dAYzwUgjxok?)ee@mIoO5rhkCwn_*TXU)-B{3zp+D=37 zxT<4Wn7yzHpfx7MBJJTb)UF1)j_1|>S@c(99Y{em@<&2n?yRblGq<2;ler}4|{WYgs^I8EG%-f%IsV1~C? zQe!`mS=5uDC`(%>!5g)Nz4`^JFY-&;u6Rsz_{(HnlEK0HUkg|4{CeD{v)rT$FlD;D?cQ}OVmt#kUkf%B^yP+AIb2mR|Z`s%M2qy(j+pb2(`RQxsP(m;YsiNl( zDnf@G;-8ceJX1f8UM`F*1q!GcPL}+fIg@|A)_~o*5f&jo1H7@?aO*RBV z?Its11aeedCrb<-%SnnSfkiP^p_lQBc;UzSmqa>+cidGy4*S`ZlJm+-b^k)SQ~^xY z%Na?Grg-fGY4{B5(L%6j!YAY{8SSNmQ$ghWvKqz8ox~Kro_fp)MYc>?PnQH?OZqpa zvJeq<91`bZHRaD&F@ganf|irU!uU(R3sH&J z{$Y6xyC|Z%e}8?G@$~$`{>-oM({^aTDZr1#mX{BmSWj&|Mj+DNX5t;BVM9{UL1`%{ zS+C1}rG_B!&*QmBdV+#H!z-BW{BntHUSVepBu3X~E}PrGUovdk5QS{l8ErS9W?Ott zk7}lxXUC7p8d8fV=5#L$nyoT$7Q6BnjPwrvC?BJAFw<3@Cm8gJ2{a_caY?ZvqMUs7 zaz>Xgt6n8YJkQX{RbLN}dz^M5K6vf2G%T&(FUk>`u+n%>!fb{R+iQKuGYI=m*_8~T zs*3yg)>4_9Sg5&PA)`Lx>FUb?yb&t)S~5)HoIC?p(?^_9B>~TFC7#G)7%@Uv8?`!d zkG@SYv0>xmIJmh4q^2}Ynfv5S8Z-;Tit76L0^aoLlTFrtt7GP^COrGLHJ=UA6ukRw z@9U1Amiq!ZR@Px_Ud6EMmK*D^Ot$+mZ~Plp52GoPmBZtmwsyoC{zAovAX;QDrA{0` z4afPM@UsZ`bScI4v_Z1y@^F(=#SW|OGbNgAu-Fw~VRHK!Vs!m1^CCOaIN@tC!sCF{aKEaa0NaD#l0^7sJ4 zPkzjdq9qUiG?b^c3rfAG&5J6QRx1+k?b8dDWtGSEK98QC%**SZ1}e4roRU;zSm&#v zvn-~{WXz9J?y24%&_06aZKU<{c`J)Eh*7w6I|=q*Jq13q>v8gHHmNQFwXMRIL7rOW~iO}RpXm9*lfZq zN)i*(abxxTJ%-hHw`efhY3x}1%5pz_n6YvU+cJMR(lTcLidXkn+3!OfcTpvFT@#b5 zyomJ4!fxE;@9iMMlh!{3#7y3fqUj)DYYiq@sc1|dQ>TdbujLAH{J7$E;T2wpMF#H} zMd3t5)AU4s#RKesjIavJWtVxg)SIEOInKB{8K)}1p_CcBPc*Ru{=^hXMYvuqMs(K^V zfOrEoYP&09NoAHJ;%UQ4clF>6(-zt#O#b0BebB-KA1-I}d!piByKyUrfZd;^o!NO? z|DJ2-MS@>c%tCQubb?1kKf4LBrD!5B^Y7Q%Cp&z74IcdWo7|g-Pr#TM43|$|Y+W4})!t8J zbNzX|+fYC8JBCgYzGhX^V6DcJukxGN4Ysv(0nTdz>irJXb znIfX2X|I=5Jcim;4^bUV_1SJjGfn*b-aL(Oki@)*0Mn4rS4_9(MozdaA2%6;g-OQp z_)bt{tW_Yt$2PztX=}y#eN76>Q&z7m&Z{d|{EN@#HqE5Bp7Ha&^rJr4b4R8MH5Zp} zY`l|OU~JQLnv8RsaqL}&7n>xc-A`>ho*KR8Of#8*j6bi@y7CNXl}0$j^2?pFaL zz`h`th$K#fwvEE^vOs~0SV*Au3&K6Nh4|cY3nHex)B$aJA_+g|+}C^c++u4{+{Lnb zP7pf)@!GoYwfM2De6o_ap#B-@6yoBPt&o%bB@d}lKaIRFgH?_peC8D4n!hVkW5BJ2#)bbFXen@ z@0a()#d`7gzd-BhFyJN5CX3i1aTHizl>jFw= zdr%w1RN8` zOyGoKHGVjq`51HeN)w!$1m8@S=$!bWr&Fltn5#4&;TzmU1u4{ZY(>5A2~p0y;{sJF2y2%IC*0N`khPhPqf})->IH&BPTR+ zdmd{wNZ6mUNW)E;MeCBAJ&Lo&I5u$Yjx@TWh(-^bS)?lte7f z-ii?0yGyathAxD>eAaN0 zPlIH<<%bU{_r~L83A^@1I{pOAgpC%b2FDZ+Zx;>=Im}nvY%hPdcsyn%ep#*R{HXVh z?M=pG%KTxxxy!OY^BaZ_Jcg=q^9rIcInbB(oUe;4xqa?E$f3Do{Q0HTiv67iVzTYC z`kpuS!L^uh3#kvs2w6Nkz|O-`|Dh0)e+bnVbrtkN((w_u%)1DN_k{#$i~p?-1=?VqUfAcC@U9@_5l0q--sDB=m=Pn~o>-9$<)|Rr#A8 zK`10Y=iTWz4>na@KOpY7>sei%A&(>OWqhf6(IHr9^-sz?C-X4VZ&sBpg)J;gmZ$J5 z&d|=@rLcNMk=uwK?Rloz5h^s0T`3}`%HPHO=O8~<@=u8YOi6v3<|$0L7TMVxS;qff zR!4J=RJyeky6#onwvCf1y5)cAFCY#TRnq(9+SRxFNqzDU5^CDbZrossIEfNh?r7>U z;70z#A35dWinnDWX0jV%OhKzCYui!=)ZAZ>8cq1zu}3f-mSLJOaT5qr`FYNls?1F) z91%-S^`TqH0Zfug^5Vi^%Byc#2O?s>ue>r`!uEL1+Qjhdg^&M(A;Pya)FsKF_|nWB zJa)$hI}eSl(#zlownsNFEuymFI{JP zC7(%9eChZaKQD)wchZTv{GjpBoY5U^t1fx6c&2qz_>L~_NzuUek z=R7bO;hS1%lVl=cwgF8&u3Q|$*X1j@CSXZ;}J zdix<29HgB=eA5T!>yL|7-+ZOZrrQUhCef6|z`pz$O0iWF_m zR~U$p%fQ^nKS%dAh}h3n<~`ZXB#Tg`BU%SCy4MI68ukc!{v`+wTLhPxjfH2~e)jnR5&Uiu z-|&YL4eLG>IO5Z@fiVypq6Wu+R-C8r+eVlre_mxtW(+wtsO68GhLbw|VP&Ey^>K9m zp9>CoP_fpC@%ToCs`t96Me#uWSLK@R9Ji;3mmpD<$IZBiu}D0@ZpW)QVZ_ObCf;>A~7*Z3ZOWrG_|8L0+k0}OPkf!bijB2om9=~obkdV&1R z9xQe%vw%I!?e7r%-6}<8HOPsX1Bo+7iDJ+G0TDz;uuJ&iKaxQz}qokblf0 z$yGOnCa_HIMYIi34*&0)Fb4(ESVn}{7Q~O~V?vt@23}{u4e-x-MJ*tOsFfW5GZ00z zMo+Yzfw;)WSPCX=wThOl6x~BG=>B((sx}nS{j)pQj&egUkuR5rRv>9NqG6;+ zd>Y7#9emM9_$DXF@o@~#KC1l&8N5snB>zAvuhkvUDEZE8JoaBdN1&j##qg~oJ>eT{ zKF8=(8F~>idaqAF6XOVcK^63cJQyyyy8K#K9V$R6-Kpi|qY1wlb^i5%rd)v%Zd_zt zeDBJ6#;8j;{~A`9-OZ&Qs07f#)N&F0bYfR|U-EL8h8n175+|kEvpOD80+?_{-!CFvnsZHS?ou<|TjbHT5pe3n@GOcO!I9r7gbN zc|-~F*#)?K5QB8K0Q6`pjC3cAgz#S=h4hTac^Lb=GO&ROuL&2x-k(OZ_v}W73*W{& zknX~zz2A$C>5178<>7~AB?*ydp9hMs2A!+pJVWD+pEYUTdXb z7~v;Ua2HUFzxTh>m>M2~?~2WfsNf<_iIa2ebD+Klg1Q=$*bBB~0^2laY`u7n#|PqZ zF&D*0a*;;z_xv~4q$T!-Gsh}5x5x8ZV$)?*YE5Q09Zz>6&sjF4puirVy_;FIsAZw+ z6oY(w3#WNaAI5=HI{5*P9Bws7&ccg0fC5X8h4?=r4;DB9ezH-yxLkF%vK*3GTAoJ` z^nK5r5e-H0$MIJUu%QWa!gw9?tV?<3p%U}hYV9Zcu90=Np*(RWLf<8!DS|Yia`W@; zGD~Wiax>ot_2sW`(k;d^0D!RGS`Rf-!oWv+fKI90%w@aOmZ{=C)vkHZ7OX_tr;Dlv zvnsEDFY)2-% zMl^1ZUzc27om@`5vsrW~v#;rxS?IGHjCi_b*n)F|$+v}5xF)YXqn7qZI(SC&jv31=|w4AV}|i z%Ry%KL|OZ{i^&<0TGy?N6%VzZr?~FYo8J9p?y?<};)_UXdi++YOc|v>>1kX#hQw^S z`3M^IhX%Dgb6yPL#?Za%>EF+qT}>`s2c;U8rwQ91N$a-#g67}Gx$m~lp~|(bbGC^k znUt6{{KNQ%snwZtERiQn2X@8#{}ExSi~e_cFl;gPU0LeWFiuUA3;trMwt>#hc*z2| zLr(e_lez2sk-+6Nswm749LfI<32n!wGv||)z-vHAEnocYAiqLg&E?V9jq^|5_b}^a z^f_-@hSPU{#$XV_YdS7inOkJ4+^+V7yyoq~jYH#nP^eiu&cF4x_b4lH*==ZEwLLmo z?$bXDNSihAUA9Q!wV4?e0C}g+VS5!6=g-NAS3l(?D*d5wF0sMjX%4FeN`M&MW@rSX z&f7r9|D)=y!>Zb%eqp+m4M-@`-6cpjA|>4^-6#JQjdXo; zpZC4reeQqf@tj9EYt1$07{3}Hs3s+k!`>oK5Ut1*kYQG&fFc7}Iig?*jgNvr1ALGd z&3<~|LA}V(=qQj4)`d+JbnArSv23bV$F^JYxxA+JI{ud%S~Xexhjm z2|FpdO#>U~?kw~juT$%6V6zLH#VTe-osBBbKmKLc8YBjb;;G)JM9B`rG{X2j*#nov z31t!zhQ4(YjfKsP#G@JWO2itp%j)254TM~ z(QbFU3&9-*kZ@aaMdRjD%j~F_SVZ}$`=MfRX-$wgx$c_Mmo#6|hXpOzt7GBkytYXP zRb5ZRwB9m5vEXQQy%YUqKg(qa3s2(}?ze|!jA-TQsqgs?yvpX8FL%qI>-4rHZN|3U zaO?6FK%2q3A1zI&Vj%turzlGh_<(t?2^h+faXnrPPyt7t)5huF%C1GP-!&cw%rguw z4KBV{YvarS4sI3HAeMKR9VG<^2CRz&(HQ}(aZuBXT8i(3A8ZWJE%CFWP2xmf%*Q8t$~-ORHfjbh|$u0IU6FG|Rcg+<4<&GPl$4#s6N1v1^3&k8RyB zpK!-DB{4rkxfh<4u9Rj5eC?igio^$*91wS*mnHSgfJ;I#jOFgrzdwetxS~|1F<`8j zZlGpxmgEd+e-nDGYZM(^3R^OPTN(h4`Gi9H8YsB^w|xKwbPvi>@{|85>zNE1y@gw( z!Ec#IWEw4tB|GkTB}Vd?P27fy`CPze%4@SIm8hnl7EP(A3#kUKk!{9=nugK-CPt|) zTjcBs$Xr5qPeNO2eHHP+uDBA{r)0K#Zk9zQ1AmM^{>q9BE4dk7*ay1|@y}^-P$)GB z%@12jhUw_Z?3qn9gpG?jHBulOvd#ucOKN&Qr$nt{G;jArVU<2V$mCiYw|nbJ9?g(1 ze;PT;OqZw>IV#w-vDMFZu7pG+C)RSA_d}dskB=bVhD?XlK69MNli5pD2w4F&I2CMK z{`K>|2E>YfU%uwprKt)(k%Aa=eFf{Wsy~r%bz+xVXAD<^Sc1NIKuCMbX|k#cHac#P zWgoF*X%%er*D0384o-B3R{3);4HY|w5g|fT9eL(zhLyuUgnxbPqhi=VvdYBtWsg9{ z6&`=ezlyNJjBdeqFZib+Qi46SNDr(SY-^h!+lJ| z65ykUg?6KXoXsb1V5vsjzMFVw4sS}vf2K3ip$MkEvYy^wIirxEK*Za=IZH|2OLOoE znd2M#!Yf8|Clpnl^P{=n*2%PfYsoQ-VdTYOwhxV`Ttn)AYA-*=XA)`{lJQT;i1g34 zvF*>N9u-^ky+7}(s=@MJW#e6=wlc|)XO3wIW(d#O14$W*3?(do2mM9F4|lL0Q;?Q4 zccS+@!_(@CH*1f@ogQmpx&A(;Za6#2uIL-@8~(kzT@Wxi&%~{_HDvuMB6NR)I|_?!oQdV+(ZeY2Q*oannstTN{_J~l&WymAE^%IH>M6J z?gHpV+sL6uO*`Z(g)}!z7$ENB?)BLZZ3CCE1JJG;*`r3hPYbjZp_0Cq3dB?83K~L0 zMq_Py0GsNzSQK3Rf1RQS(Mh~7@q^&7@jteUgw%_dtqeYYjUkY#$qMPo=oxYEzKqtd zWQJL*LT+H6(+pH;z3r>?$KTmESyUOmcva+cwr|itt-BkLB0T7(cr|+Z!Y+?;N~_GK zUX8grg!rm>%TA|GbGqJkibd8w(un9D~F6EO{Ni40tNSe6lb49Y%ciYItJ$>4YrR zZL$^#3MXU!c;S-+bjd=83J60PSoaNP@sU^|c11p(iD4AC*1ErtE#wPoHXDe(jF3w& z4tL6k2!j-Ah??+1qOyC#hWq@N@`eoEa>*KHYl_C{r0xb&T+Hq>6LJ-1hJS zUo!6?>{I=eR!1v?i(43B$wHn?vRJitX$nuAyw1OvYt;)`4a4KtjEagqOm|KwVON{T zXlYN#6^XbBqKlIi5Ky_mA`vj2jTF@@2k#@Iht{XwWnk$IbQ|t!EXio_uF*G~LBTYC z$akd&(;^lM4?u>s5v6@<>9m*UfQjwoi)wSJOhiR})3~NOOKZCmk|g5SEG_!@wvLJ0 zN!Qxj86d~H@a2Y8!6k(nDBsk2$vhX=R;)y}P4NsT@Z;AJ zpV^lUh-#ilC)1l~Ca=y3Ieck4;ra{w$$u-=oiq^SC`j$2o{|M0bQ<{j?K=;D(2E?D zzO38)3^O~~QMOF<`Qq;q-rOXx;9D@YsN{qS6QCLCHF);$pnmctl6P9COFKP}nrSM( zXty5BV2|JD$%BglWi$^qzWJS(c8unl=k5_^jp<}pO9%x&gv{u2R48y`NV=G`G$K5p z1z_{xCJWG@&Ja-BBt?2EQmDrS$T#uewWCG^DNjF3*W^FtKL^FIvQ1_8T29VZ^b)Xi z*Vu`X2vpJmlBYDIG|y{urvjfZA%2@>PG(JqsR%T?Y(V-yH|nusE4~>2vhhoK`ZH} zH_g1uG?e(n^Y&_(;aY*Uw{T9uuPV4($0+XaNT$Fq?T%bJ64TNyT(L#%NqBpgssS5B zIcuq{X1J0Ne}kdg3aOH&bO&2t^?ns9`3sRDq-nAOv0z&9=!M>Gwy471`SJR(#udLg|rOWa=V5+~@{x^rt#454Vot$xRc6BhFY%`<#m7Czkw}90Gy@>^zovz? z(k1t$3zV20;bchbAbqD(3ViYV0(~uy>?@ggvXYZYw5-7<7ZE)|wH$PPG>DTb8;c?9 z=n)mYvNForZBhS_558d}tr~s2V&(&|)?6Dsz&8rQ%{cYf+Zm62*W2ql(EIayR2khp zQU9A+VE@`6Lr#j7jJONUL7gfOct}@N0{oVg1tYVxTn=q_o3lt+s}E!HNapkM*8_r& zcW0CaV6l{{pmRT_b=fld589OE~?K*tjow?d71CoUvFgba>h(V z5~M;SoQ+hN^7I=%qoC5hb$KATVq+Wb+8FE3v37MK)%B*w6zk)exD;u@~QFU#3`h8X?<@?K>tN?sl8!tHy>I^|+J zRk9UAKRnTUdHwF;l&nlvC1mWG{apOhpeL=kue*5-o_( zfp{fgVgGz4c4dTW#C~vu#_jtJ8@#WH# zYgb*d%y97g_>8U?>_8YODgENm?2z_*$wd~H<-uL5&Tw7`F~w9Yfj|~KS2zbG?d0dN zS}(uOp;M@!iPq(TL?hF5LHCjtU|aJ5xX(PCk+aHjPXPdvwB-ol+E>pdZSP#uJD_GvL|cVV&7S`&O zwa?E^4jL?t{oeCgD!xquPh+uNB%?S@XoC#4vXeHo^t-I-Olo>%?3$c+3AtSa<;z#= z`n8FUqcq;%_J>!;O<09e;ZK*B#oOW??I`nYC-TA$S}vD8+7Z!bfdaYlcMU)UynRE* z?5lf-0?3#*3M<_JW_`>2(+~GpU*o8Jj8>KkEF0o{cBVTcQ2r+w8uk{;!d1)FHE~U2 zQSM!O6WEcta=8=3hB6SP2a=>x2Y*5&H$@HBtFdSU9`;Hz-2a%?7sF_*p}@6{6Q_m} zi6W|lk&=s?GeS94!Tf_UTD}%59_nnk^S4fkjwMtpq6kRIEZ{TrSYV#gD7@hTP4Y=O zUvDhgU;Vr@a^3YiGC<+uCgrDA6G?7TSU;^Qhv-o?=3#NT`GVQTmr79X<{!$!%+IDd z8RGOTue|3=L1~rt)2w#|wi9v;=^(A{t!+uy=&gj)FQr8Ivc5t z7y>Gt*UrryVq_6)tBoSU*EG3A;vv_<7+SQq%CgEco^l4P@YV2{-}64_M;E|ib{<81 z{#j>$jkWjxekpH(yxs_u_$GAC@*>MPr9oJ#$AnTw;qtTRYy1;6B)+NAm?ef_`L}f! z=ye1-n$BGK4<&fRT) z=LNeyQ+MJ|8!j-7gXDP?NmpHO7u`vL{j@NtVi)elfE%VdHPpbg7vG3znyBfq&Ikq2 z27*tDqRcvZW5%QY3L3+LfZWF9a?vd=%WFMynZgF_r{qxjgkX1dymkRT${SR%0-!Z& zmFHk{heN)=@0MwN1R5DR{|pEW0vkrrY$n z`^yZl^WpNIu{p~5}zD-1U`cz#^b0y#7I-Qc$mAxck1+)CB9@* zrc5-EHx(hQAyVnSAU%&@=SLR=veB1DhFJ# z)45CDeLIVS==ny^Bx$o;(|IX%nP#P5D5=>?u2-(b0-T~T^y%hq^AYX-&4eN4rJ|HM0eyj8^HsU+DCl&ub+i!6BD zXMN5gxk;z7f4#@zc8Lb7HY=_t+b=y_lEO|ks6di&|LsuEwctK;+~9HWn^ZChsqp0H z5|?oG_-~V?^%_kmMDw{RDcq*y6mpi^ECII;i#`Cw1H~2TGlkp^w(x@|~5;2#E~Jq1)F|-Es&Fb7X`%xz4eZ}R+vU1ye#tt@9$R02#YYg$URgP z@g1GN0aj~j zCH4&K40}Ol6a3i|UL1jpss_L-El=xOkr}s?R9$QhHIwc$mzJ&1%$bBdu?NB-!&3n& z=?97TV!^u6#OPvS(`=x{oX(Lryk$knE>i10$ZGR0DJCRDc|dA0SX%OnyjE9Al z;lYrmH+4PvE~#9=a9jf_)r{!Fm6_E)`&|~@iN2q}DBv-rsePT(CqrzwX*||f^TYhd zcukn%bzTWNr8Jz6@iPezjC+fwVe+l-0i@AnG#q#ghP>~Rlqr4A)7fSd@&Cb@s8?BJJLt$i6TOO+zysO*;In~vMYSF(@ z=HmDSQ}`Im@V=y2TSk}D)f}Vum7{(AvQIH_bkHs1Mm6AIZhZ6KqQTdjw?RSg?xDqV z(M4c^q)BbfaP{YV1Wqtxyn6OtGl~FD6VeqnG8)ctfKW0B=&GO(;C=NY(TRZhQ|2(3$NALClw7>O>kX5t4}n9f2IOn{WOPBQkiX`vX% zZ+rW2c$OGswj`_=?r{BZd)yKc&zk?Q_g=WI?eW3)@iAKWxa zeMDD&`!`4vpr5Y!kj)b~@rGaoJH6KvfbKNslfzolxQ(GEZEA&6Je^iTv>i z>Gn~GhWns)d@}gqqD@9?%#w%Cls@cD4aIFbX*SA(-zQm^xdhM=9NTd+7bX_MyI0kREzRq^hJzMUx5vAm?maiEeuT;EsVs5hOg(<=XrIMiK= zUk~EQ(>(K$r@z>J9DD-!LTdts>3ZV;rpI*DuWKYUdQnFF@3@LO+O8ji|5_Jcg}J=% z=-*z^d~YzjQi4Trr>KK}*F*f?)&6+3W4StS%X_8OzZ`Jd7XazGL5PD;i{*@54{>BD zWxYV+q>aR&l1JC10p5Q`8k-~1S&SvpTj_hT#LBT7tX8=Z2V&Y<$m9Mv^?vv7=HmRPL?j!TX@XuwgDN3WmAm&;w^hUUsdtE`pi!j6w7#nEz@4d`$+8%uPoCj>zL5BW@u zMbV4YFWYBHR$(lK?8U#OE#x1#wZD1S?wbVQH=~ovBF34wU&kzIYL1#+q@F?Gqx`~G z;_VQZif#F@Qt*SP`$XBBHu3+?*OmTUGXQjf0m%(rnFIvr9JT9mvABDt=Wjwn-LoK zg4gz{386x6?w3s0NymlR(=d67E6S)y+K|p(`qr;wRnTFYoO_eyJWwLFnn6?7tf2mf zu!1CqCQf(~9?G#I2+1X+E(IAKS0;YX^d&$#g9DJWbDUw4P)YYl(v`B+b1p!kJhh3Z zh@nL~UbXo9yOVSn4QXt z(gJ{nk!&b_4hEr@*ez=rY299r=D3i+dCBmswMagL*KWI1r&=R<*g&&!&_QFZ#CdxdGZt%( zpK!`U8=P^l2$&8#K_(nG+kxXJ=M=D66O`(@(QyRRQYM6l(Yj%Q@v1TBm2x{U@%!I=AQkbc1?;!1Xk$kk*L0D&j5ztk zl;Gd84bP8}^70rvH2 z7*>{E{UzuiOccmFg=d%oIS~IfeWNMIIL$=?f z*v!{}n?FBm2I#;60F>+373gb_u~Dl-|BbAg(5-yILEGaY3b=BmgG&d)r?WJYZ*T0F zP1tItw4p0MJRG)J7NX6Mr?icmT)u1`7=d3(0_Q2UsLfP~=AI@Z2!j+{kD&3}-+;?N z2FQe?BZ3S9uD7N^0NaWqWK9!vGY23{K*|5VXi~3n)L(C$$8%JcqMH68KP_sc-CpmP z?I3Jis%copO76ipF!=#kt+u20u)7QWqD8rxK;T`{bOi~I-Ws8+SeU47U763J>7;LK z{1cw>8yWEIN<T#x>4u_;iRBvx3Xz5y*%g)KChZP~7eX zD3;0n%{ffl*ncWkaE}rCeog14&u*@0*BumiHCsRgis(?(so`S&slRwwzr}8+6hesW zPmHF4XH;oap>OdYK_Jb&jnKirIJ^~r|BhS2SE0W66M%X@iz|FA?z^3C^|)I!`wWNSXKYUB+zb#Ief4Z!mn*>--GYA12;Z3HQ`& z>9tycKIiiJ=0Dc&_*nA4;6;0TB*SXP!0pqR&K$6t`XhuU_(?`=a+!Su=GWRyE}G{x z638%iLx@hH7DSR3=^yIw0(WY|<+7#k{(JZw^BOD|#Og4|=qwRBZ0v%s)#lmq1cD%; zKlzDezL$r~vXS;6?|-_^PCI+5bTt*ygvq8`L^57s2+^yvF$2a+_+PPw754Y_0;>UZ zH^pj1Z%XvR3zfp;O(uvWOSk4mU)<#*{V1YFTbht(g+uH0c6|?# z{?&~1l<5RD?5GtR`*FkU#xhQXsKodQ2}D}w zXTUEd%P}#E65~G?tf7op+fRyAAw2!OxYn>M+e(OOUVj z0d+Au6jFeopJLA*OkvOvk4YeDnMIEazz;+{ML52Eu@4*518koE))F4go7vyR>i(y) zC}Lr!(^sqjcm+|#))bJ3Y!u#LX4~5iN zvgp->Q|0~Fs*>V8!C-XIt|gZ>;e`h5eL^~ZIG@RAxh`z76FP@u;ueyaXah24>Y_=2 zHynO?v%Og7XkkxS78Zz0>ai)0NRJMbn>J#d`VhlZq>GXKE;H96B&rZ z_?I$&U!M7c!f5_b$N-`~Q({Y~^=YCqSTezT`|lSNISDu=D-A#lY?E{p}A&M{OG5DOw zYa};cV#WJ%PDqI-$c5Pa8eWF3*iX`;iiSe2pOhm3h#T4rP7#1ZmKX z&4KT`4vyV}TU!v0lK;^ zZy=beQvY@BqPt@;uMK|aK%QyG)7-u7&ah(Qs6g7M((514#{G%`gH9zc--J=oD>#HA z81tmoTsMD0?<`uSTK!(b}&Z@=d6044;BB@>zl&)N($F5lqF zUAS&{<*E78DDa%ub4*9kDMC#!vLx_6Q~_&cR}{;G0}z?Tx$i>s{OcXb^s)dbC|U(H zjtg=OS?1t`Rl?zy)*8Yo*T^tofH*H#?oNCQZvk3JKA(TP1$ybxB{D5O+u32Yiv{Ib zMP8YuPQSi?e~3q|g=Pv13}=9VT)+*7y5bqEDXL7AfYhH>&DBs$f_>0=o%;`0wp{4> zg3MxIwtEf7Fp&30pHoISipI|N?^>7y#oVOF&uj;Y~;>uG^PvXW58d02bv@pm(ux%Cv z)SfTE&zdPByQYWeR6X?Viz^}wD!Gf@#kn&7+wnuNboVrfP)nyHy24|@0dC8R-iK3y zNGx~QyC4B>2|#y1YJUl&6QVZ$H~&r?Fl?;eeUF1aT=!KnoKgtBp{(t+q`Owt5Z0(? zmdj(pETy}Tfgb}DP#^&7Bhy;zxUUL|i1Dqhwqdi!c?XV9)=a7Pm=I|}&`~yEE-Dk? z-1AM~0C4f!fh2+k(4~pgX{{YMbAWsmLjcElfUviDuD+LM#0JhmEjN3meA@UPRDNj8 zc`V#6{r#%vc2us}n%cZ}zvv?>()QCugizRR1`?IL9ROM@gJHXxI=+2(l+>Ao^|S5% zB7^w7>UETf9FWKF!@KQZR%SroB_N{0Y&=E1oUJL|X4bfv!5kWR?WS2#V4zANtsp^x z0d8b(M)u!JBZfaA^6tA|3Isq&!3Oq|A)%p56U>#4KkNrS%BEkR=PgmCf^6mE^IVQ=oSGz znCT&8$_u-A>nL?U#w)$^+IAO#Fo)DqC$of;O#o}0^|t806>4863+i-O!zD& zlR~w?B+sokX8fDq!yS}C)P|ubp-*y?xU|x-ek&!H8YK~91xg{UO+X(yIR}cZlS&;{ zHcCeJBYMry2`ccb3JU>))B!x)snV(Ce0m!AGp!!`5%8&*?(Sf78BQOy`c>`gcD`CZ zdBMme9btfULOXMU&5`!xMZSj!oI>})zaM(h(A(kvvd*8^B>@BR1d{SyvoTI*vX-)7JJ|hg2s&;(Jl`TFZc@A+;{a1x=-|}?_>|#6WD0tr=3PwR zGuQR;4gvOW8U-|eSrMb4JQVE9 z`*Ki^O@MnC@6deYSm-SVN~mP>F%a$eh~;aQ>+zZO;;ejp2}R0Y1*RkzIJH+OF$q@$ z@OGuW&*v|&uL@e-UVuu+R-FuK!qB8C;+DvezhHbo5>5Cc^g2&Ir$rc8R;AT8`;dB*T|F zM?)_b*B^fNZ7p*reW_)rx$u)Eoo|4mU0U&@#Qo1?54rC?7l)A#s4Z}oT)9O-Yi|^u z>To-JIvI#y*0hy!HdSK+P`gVgENpCT^;m$}Nk#&_GX8qhlOwu6Ku%Cq50^G|EXO6EBA*sPtF=)e-<1=h5_k~cwsnX7UKK+`H!8^tdVJ*kg@N}$IlXP z=;UK6(#`Suq&7~CPY=@WK#6Qkrc8FPzL&#z{@O9q0o1B`aQWZFxG|r-OXGYmmy=IR zIY|xk@Bd770hdjrP}MSSde#6^9#pDZKph>x{Xp4t{J)w#owD&BWhXU`B{eu}C4060 zj8n1xt;~jBE^d%b`}yDS8RN_A$Jd-Aexb~Um{hKK_RxKg5(XUr*p<*Vy!%R?gYU(KivC{K(Qg%#c25H!JdKrJb!hP$ z_>wuHIG!k`7(H;)YC|0-Mvfl<&@VdBDQ1V$d0HZ(rQ|+G1r7kB#~v8-E)!<3$SsZF z1e+KUpa#Dp1z-lbON&zmH(QWsa)?Nze8{AII!ndF? z^hYJ@CjX(uOwk+0l`aNtZ8NHp<7HoKbdC);=^_445CO7g?Ln>d+*?{i)z9k2y+c9K zf=gh4KU)fZ`Q6OOJ!=BWC6L>Fbbf%Y@WnlLBw{w|h=Rv52JUoXa9&kbk!WCS3q3mo z$EX2lny#j4#P0tkHuzR~gV;QM4m2r}-q|FyJ3?k$&%NIR4N9z7(#B<|$-J>6n}yxXCs83h6%oKh%^u9VT-YpnQIz;gGEEbOA9~g3o!I zitd8pJ3epM#$`McRozZQqsSkJXJK%uC*R2W ztUvf9W#&sJ3YT1m*zN_g;qJ3vtoY$-NQj}@oF5Ph0iIyxswq&%8H_0jm&anqBJz4D(n+`zh3f*L z2b2lOZZU9rvGO<)%9wy0a0~8dI{=u~ed3#EJTHlWSULQ`2?@!1?+nggp%L3x;759| z_^V>$Tb$@AZXVeDxImX`u#QRq(k8ReJ7j|qh6Zb}Dy%6Y5(CY;2V^efToNIfpUo@9 zKs1vmKTK`#MHgQ81f`A(oE@FP3NgWXodqCS+WDX|Th4m^s~CmW4x5V&uv2_E?FFCt zTf@tvJ+0DdNl*U=(@!r(+;(UrzfiWCw0;qo)YN_wD{Ik{*CgZesdw-L^>b^31oWq9 zLC{0PIW0ZK@HoY7#N^5<4uAwDTqPanq@aVRhL&T=1W8PmceWn?(=S&{M zHUh6=0v9nEGFK1Q{U8QZ!Gd7K^Cyi1n$roPB#iu<%cG-2i8R(OT(}zQ(okM7cL4TZ z3TgpH8y9f1gU5FJQw(Uie4nv2=IwD1EGgnpt&1b{)}N5o3Nk1JbUk^^I_00`22buS zV8*HlxEa{b@|4tMxq~E#Cj|cwz-1m^zKR8SL@*AG>wUJ%+SArZqeN6pVgq;^NM^=j zNJY_Z-mi&Pre`?$(k`@4lfVIQYK4GS!0(jsK^&+ienMDQKCkW>FIM@IdQ7vzh}ed& zHJVh@>zFYssJn$3<--w%n%azL`>yaFbHW5PtC(+9nf@2T0`47yG&8WK2aesnio`d- z^D?Qzn1;N{31xrX-0FXO<3XOZu=U8b35JFE8;w7@L>+_+pk1w;2ewT|cm&)M2yz!K zihsf%#QL&x7Xcq<4iK>Sq5D8j(X!d{uWPtXSM+wD=Rf`%&G*&U!T0>qDcvsDK-}XW zhB$A^{S~#3L))c!ZRkbYg&r?>thQi((ayH%w{jf`XVF~;@E|P#1dFNSS)$GO(sQ*V z#w=DmG2^9K?%({=h>hT?a2G0_kr3StHSccU8TkDghAOO%F)3AV)>(x=1if6cxw&;H%qgdG$M*N`k+#RzDQz!MdnZq1-=|0WeggMv*kgFYv``||#5 zBVtYccct+Pv=3lUJNF6eDB6M9-A&=s*%cyB#0<2;g zCG?H1sljK;l3S$YsX~X=DfuL||9e7I@jmGjvHdfZUUh5WR8HV1fh!XO7N z(*V4*Xv*A|ZNc7L4vn@WZK`IQQy^Gh5lpu?du?)pE| za+aaQo*qKs-U6OHx}y(Qz%%j7Gi4_sKuP>%w%_wO1H)JB1NxJF2*RO^CR}5Cm!FG8_xto~!Dg^Ml(% zrc(B~8Tu&(NXdTX{1hgXl<6Rqs z`tiM51fQs3^T$1Wu}Aw0;8H*U(_Ir5+X$}UE~U^;rkh`#01}0{9wH!&T$_K88rz^~ zqf#~2l&P>qd(fg}6#1yk56j(AqyCeg%*~sEjeVK02~Q!{Xin#(sr?gPy7(a2Z_yUjMcRRuB|(6oyUAZu2N%c_W1@vU!`| zH;fldZ`K!EHwUx29X=s?HQ_OwB{TzE)B^Bie}Bs01QQFUdQ(zHi?<3LdGxPL6(j~j zN>D6ZBiK+e`;%o>+2nxpeNO3h`uSe6PcBLSVI%PR0x?<2JBysj!~xorELs)Agu%eP zyH#&)+uw{U%)#{G4xl4%gM?zcAVVx zg5Mq9^0A>}2b=r_hrAA*+^c=bcjj_MmUk(youjiY*Pl8&?l!P8Gz}-?0`5P59NoOB z7EKPnW@Vd}*#g0xC{&}`cN%(roE}@NKeVy-sc)N1B;1>C4{a|d6{eO%ysMij9uM|o zpnji`FVmi;f?ebGkOxCFpI-bf57)(}Q4-veiM7PdA)QU;7;Nl=FKK5sGZjk-(qf;K zdI*m6uOfz#Cq<*JZ22zPL*t2)mbqN z9!PEjvrlOA-Hd6kk=tliK=v>KVTJWNFbP^NI8l`F5~<=+s)AHv1d6VvU#&gYV9w*~ zOBVQO%ampS0*k%Dzj9SIcnuFOo=B=MFaZc5T-3Kx4=Nc6yCP@gvEE@*mw@>Iurd5xSFLb~B|2tN&QzH$kKJNpwG$~fFvEeh2c=~2fJ=@m z>c(KRsyWLH@4g69ghX6>8U{U^Cm;#bHNy|zU?iI&lz`VZjq6d^r*@950fM7gu7YE} zwHVMm7li(0svHE|1W$wrfZIJ;cu>t56-8*tY*;%zCzJ0o@c#1qz2>1JIUORP&>~T^ z!xEi>S?pcsy+3CLJ+_BF!UYNoQD>K6pgzNv_zcE2$8?8N3p|7BayXR-hbZc2O|g%@ za}xCg%|>a_vx1mY9oKgo+b3D@1S^MU*N>eBZC5@CA$pp4y>t|P=GR-q3m_AC_we;O zh|1SodgM6ryD|oJt?uC$-QA0He_U#oTEtnhQM#gM9h!na-X+Mj8Zk7<+w{^Hoei-y z%!0{hqEM;S24s&EmcvG!rfas8M{%!5mXKfd4h2^vZsv07e(_>W+kwZPI{@^s1`uET zRYN!wpRE8d083ModKaviU`!Lo?a^=i7G#mMMDWfA zzLlYBuaa9F-e6Bvq`k7883lsU5qOv|Wfb^-FbBW++GfV8Ezmo9faiDRy9pXrJ2;dA z@9i+3C-8t!J8U>i?f^pajrMEHcBE`jwKcNyUL2zm8>}+?wDElB0p&Z^y-2h*1Y)-4 zW@C<>5U(p5c;X6d!vjFp%_EiCBz|8Jh-2%6p~<*rfgB?5J5i1n)Sl{cOaE2KRX5qUxKMT+g=+`V-zOx9IvF@BTTBTZ6fT)K>ke1p>6_izCp}|_ZyYPQb5}YuGmx}-GvAJ|;-ozYt zGRIuOo6Q+Ubj5VThev=2ULMgnGc(zr2BTv8$=S~ntm{;9Vn&&LOcd9bAfhMINi~tR z=OxL)p5Ed)fgGoz(=CJ|tIUTN0;jW5D_GI_wsB}Yx@_+^m(V7}?PItuSbMRPMVuM7 zUJ($gH+odl`$?Tc1IdSz0nX&hdSg@|IGpCoSps<(;1A^O82mq zh!B4RY|575t5=>fO4CU%X^TV9YShm2NNN$^i+?*kHI_MUi%(tN5VK7eS2gQVMY-R_ z?lM$8iD-ueOo)D_jR%6jo{muXo_hc2&b%B~&IfFuAPo!|^Oi5I^}`M{(HFk2o%{3s zyAdPWdW6RAl2hAn>|~S;)dojHyO2`_TBPmNx4j$QMzi-?8ZWO7F>sC-yn9p7@I~P> z)2_-suiJ$iL_gu5YFmIUoho%6V>4(NzW6F)49oCj*ML6zdcg05W40f|4T1sWx*Ia1 zrP_R8xl@DhVN)M=9EMK3LBL_t&JC^Os(XS+(X@=|RgE~l&_fT1Ao|Hj;fvxztmzjm z6dnPENKr9;eMfWEkr^_Z`d!E@Ce$kwN>{*)U6H2j89x3<_u+Ht8Zu@Dw^VSSRd!gD zis$uBqrA*Ae9sPife)DG5~A0C4AAv@0O5lo_{z3SpW-vNN6_ntYhhu>ilhrElBl~! z;harIlp|P057Ywa@!Es57G={?-Z>&A_++h$BIaq_Ol4)Sbpk=TENc|lTRJS|M~iN z?UAk?Uvd+l)`~K}ZxlVV@Pp|aFV(5+C}fI|NR$ZZwP*mTS9Ta54FSWue-0yN#}V$! zUK>v+#NP{5{@7!N2aiCVblnnd+05_BnR4nP z?3&I`{$qv?xi^SQH|1|=F@G)CeisI$vE~mH~5?TkVPBEPe0^ZzRA+`L9xhM zVR8Z9mWQ-NbRVNOPDijnKKEUUAM5Br6yWj|II8K3hbBQ$vQAx}mw{1D%)Cg1ipl!{ z^ZA}*K5gDVz!;*>9X4(}J))5tS9Ps2OZTM^o&{6s38K1F#LqxSs>6hOCiP)MD9AA> zSpE;BXAE`ywL&GRn$-G>@rk;ukU`tHkQRub=R>-R(T^N$&yp7-qoG39T5hR z`-p(V)?O5hgnEz2EekOB1K67U1`Fg_UJp83Qol#OCb|R>cOPc^vEWMFRb)M5Esh@$ z=%x61fCVh@xa_OxkF*8w@zRN4MT%aKx@;2dt-_1pp#lVLP0H-v5W}@svER8}Bs}Mm zFlrNex(UJ{-bis9Z$1szFN~Ji{Bpi;mMz1zSNK48iuiwK5O8mkWdD@75R%)zB6GR@ zlO~RB%8ap!okZjp0hd4CBE#iXWX|n~tuOFgi00YXp(?tZ^gho*nmD$o;K3KiMkdCnV%*fa~O%$U!5n6La;xGZT?Y z`p?zF?dNL&2R%lsm?Ds&$x@!A*CFw}8eEqkK#DU+u4nZNZGQEabj>;N`!o#4DQAm{{z3}LjjW|>&hwHYMX#?aN`Tw)IF<}P4K_h9sKfJA zyT*E7*lc)n`|*lHqiM~1tg$S!6z1%WkvJ6dWFwCVOQGvpKEqL3@>b=LyHB^uBLQIA$$-7>Va=VXAgkl89<|k6v z?%nFF<{dCcfVthB5oB8fN(W92JsK$C4j#7vJzo_7M{kMt3Qh>G{O7+D#CU_gO1;u@ z+ssP3=0E2IBo5)l6n0U2O6hnx0n}_&T+jMrC@%BysnZUhRWjSC#oYI}IV0Z!qau_> z`scXPnTs?s+~WO%_tEuAyaSZ@$yj4GAUDCQEo(JESKMf4b0D_>_Yyy9%Z(y;DX0JG zeERz%lS|WY5rgO%aYP?L%U)5J2W`_qsy_yD1q-msG*Sdyf^_vGED1ER{GHpw=^OVr z3H0;z4u1*g!D1l3wx!x9fTUR{2f~t8LKB-SKtB3Y3(38#W$%uQP~mL=C2RRjZh~VA z4ncdI@M(kykm?4!OTNZ9M5TrPF4v!-*nrP8PM*({Es!2jp%%+Oy#u~x`gesF(g`{c zo>=jxK5vFcxAjoKrHGf%qHR)w3pc!{Ku^GC1w;WmUloP~pa&E?JRGjA%VNvhBXlIB zcTd(SfnVw0jx>9ml5Xd>$*c=LQ)ze4!}SGix0w^M`h$9ClU?AMaGj|@d!tHxEY?!> zmeuo6r~i+s?+(ZE4g0reR(7_sH`!Yx5mH$XvMGD-O*TnF*^vs5{n$HO6lG+OjO=8T z^*f*M`~HsO{i|~18TWnN*L9ws^E1@w+b5qZ^#--opT2u9<$JW3kUx4ugx6H4KXfxn z`nKhxbtUp*2getfR$W<#eJR(SU^;w&D$2RUOd=u9=C7CsH$2y-r~z>!J?v>uZUc)=BZJNsV0pr+L{`pt`ho_mwKXA(x_eOlsdF>QeZsbQw0fb(M=;1qV z8hCfe=!;6#gw)5=w#)yRiNpmvp3jJ;6Bm8~PErT*mbu>BcTS{WC23~6D!47FUbY0% zZugFjvFGU3Z4RQRvBAyk5=w1Dc=$B7jMQn~MLV{IDy@x@X-0#?@ z-)W{H3iU@c3(^iO#Fk(uzx)FFZLZ! zb&ayn;l#~kiecCN*9sh%xv)=@!&JKF?9bfmxw2eUex;$azh^jUaScqmV_3M|G-omr zbp9oB`I@mi{BKrXa@gvLe&d~Aoz*+6=SMPYVaycbS ze8PsVTtqYWR(nfVCE* zYoj+lNo3wwo1fq06l-Fen#%M$2Es}a_V?dE2c57YzXKg?0+kjkeJBQ~?h$7bF;dDv zINfN|+D2QJFDdWKM=ysw`}?8UqM3C~r_MC4(#mQohBt>cF8Ru(j?LnltBDAaXRy95 z*bT!=Fg)KZ)ZVxdB_Zy$7Je=V2{`B~l~a}wx&CVKXeG;TY$MXC$4@JBci}vC`q>qF z%Xrz*Mw<4?D31<=a|14js&c*9h}vuQE~ZV0|G{LClG@}xH#kh0`k<9hr*oqC9TUy4 zpP7Pve;L&<(&X28k z-{|(8JB&SvgV8{%`rLezY*=0rPyFepjOBotA?oT0cDZDdWs+cLzxh>0Nm-UR36ZR6sjE-dT65AiAOk_uJ6S$PM65QsvkffiDgoT7e;n<=3)_dW zw4pL()Z^H5S5-UAG))^x>U>fNWB%ULg}K4ZEfJ0^Ag zOfNeQA2T!1uut^g<@3u`iRKHoJw4n^|A<=HXyv;>b5fx$?INRf_Pd>sO_Z6F+mMj? zxt6Hk{TBb>tDvShBVN91=F#Zwl^k$XPZL2*D>H!(fG5gW0aapdDM6a>{ysRJ8e!7c zRsWL}K9Qexskqnk{{Gu7l|*v(b76XM_Zz8^g10{ykV(NLx?4ZeBlc)J?0C*4Z;jqtQ`NmPbDyH!v{h9j?^$y`>iayQSEyELb7mO*7@sJ4 z)Ht5SwMORqCidx=(ZMdUeyQs*3y}-20bU%{zPI%Q_t8p18h&mY{s(nP$_cmD6r({r z(K8N0LSq}Y(J;zMog%^J+X-ebS#aMYd+K8{Rt!X=`5PK1bI?Kle4Qr(&#@fe;4 zL476x*ob(V2jpE7HwpJT^A$Yv(}BuV>lyVoSPQScM6OerRa%~=8JKmzygwlw0MJhh z$1}rE(*WT-Q5{txNbyEF<6it5Y%q5ZO6*~xdc=|TynrB{6w|Zv-~4502@>nyu@9Dj zvkvztq)w`+%W<|yzq{vArcwJ&D3XBFlQaIY;c$88r8$BIHQ+E+#Q}2oSV7Z9!(J}u zwcu94GF__fTupU#loop!4taL$MkvSt_})ZVxO4OJ5=Bp8bwmgRuA8+>7*g5enh(|3 ze`Vh!&D6hdc$f0NG*4#lYi49X8vn+4XP|St_g(ar$d8zP$c5Vx3TS84u3@oG;gi}G!9jUW zhzT>`)K+2X`%FMpq#r@4fU^JEX^dG-;%xc0wZXUY^Rs!xD}odz^Q7c$9KuvMg0XnE zW9XQ7X0T4++V^)%H#bC0dxyV~z^t~FF6N5`-e^a-fa69S1NSKO992#MQjU|t)1%4D zaPey*9(%|hkE~!hU&~+k1z9h30;LN7)~{=22x9^yQoDTyA_zdI6=k`;yOE)fnCGp1jg~(?ireJA9 z6c%n)b(@SOdcMNW6V6`0bfe9@BODck&$SE4R>k>{bS7B>ECB2=!)Kh4cc@LYtfvL_ z6U8Q%3m^YgW>eMg+$5)DF8#A)oL%gJUQf|ZV(a?fA<6SJsP^gi(q(E&vfSW*PYNRX z$s5f%zQ+N@$DJGb>kH<4W9s$s#I?M-TLf&PsbtRVNNq66GhbCq|#`3cv@xE7@b#>1-}3io4T*y zdVPgiZgQw4t19T{Kw1V^rFi`#soA&S>)4rXS_bH}tIB536`0D|zVg|XXpvyeJDUb6 z4#%k3LPFQCYO!?#G|~)ln@d8p6({HIn7a(<5!8x5T%#XtA#Js_vlSK*P7@t-KcTZY z(-4rG9LypFV(&n{)BdMeZ9vOTORw3Tk{nS2$*wdKsqgixev?8i&yz#mUDZh^irk1e zl@#>rU67fp{S=cul7KV|Uzpx&%0{U>ZVvcK1Xw1N8&hKq+f@rzYuXES^Pw>WyM(Yb z!)`wPMgVh?3U=;Jy0_ph0fi+kWC)b1_M$nekJI(n+YyC#D9|kRdT@RH0!TJ62ykvLBZvi@)vmY&)0o{_u>i_?DRx&^xZUGi}p~=@ys|UqQ5Rj zpe!D$8;Trk&()zA2b;|pwA0cgmx5!E6d2tao5nq^ZWT=(Ept9%D zE|~B6-5UR@H~UxrVsv#HUWGGoml*XkFP<73BlvS~agIL`+RrJdaPT{bG+Q9%;>GfU zviVnb^I=9-ULyPPE5NBXS|etQWHD-{%aEI@J){#K;827ou$spJndS6W1e`v%ivB$y zNU_j$to0Wcwjr9{!a$&WW#Ru&=8x?>DQ@WabTW`8h_~Mu3@q)&{IQ2B858E3t>A7K zfOx@KIbnCBo>&cBEQ^BnTIIp=o)4MtXTW2uIlJSiZ^6RZhvsh)$Rr{jAOC`6mJOg2PfLEo3JgB|(mOss8^{|H z6B)vcszV=@gxU;dT`z~<#Z8d^f$|a{-oWrAU;p+Y75t(`<}8bqqL`*=iAoJk zl&)6?{x#u4q%`GMbK=gW@iriD@GGTcqR`)|i7?JO!dV}UbGLLnhTU%(aR%-&9c%-w zBwGH>V10Z7G?DFNxK%v*SKi;pw*S^gBFH$}s!zanM#|SZEH}oXIP?vJm=;US?&A*= zf|Ld-9z7h$EmTfsmLD?tNOAWicUWf>aqi8|mub}O-MVt0i9+TkK5sGQ7@`KSJ~GUi=)pC#dr#ms2w;utu7hiQaM9@tc50GYgu zJ^8W7Fwt}Op92=dS<-VfY^v@zoqm113Fkw63|tVVh1cv&_{YC~GXC@4!k`hrTK(*S zG*BU+aGdNpgE4qFwDV!bxxqh3gYVVxH6r#?=FxHUwn;@AugHtTv3|8(dZ%49kBIGm zqAOA*{BKK*+G|vmbX9Pte^I70g%>xNeEJK%~CWI#5(rB5^!dUH@!a9sDEHe5t@rD^Y6jRy`druhl$2=G_ND&lpt>5 zX?$j)pm7ZWn!8PHT4^cQ>g9}n*52!~IohOJv3&|LXTQ$gdD^2iXo=^;2)~uRa2TKV z*ip#d`4WZub)~<3APtEY_k`mvlQkl0?Z09H*H5kM83_e5v$J}&?dLBL&qbLGF#(~x zBT;v9u<=8us>8bDF<%>uMb+J7NDm#|xE{I3@b#;qXBjR0ycLRWtb>?EkK^y*LpXr)AO%5!LP5}Nn%kp>ZO;A0L$nLtTXS!ao6z zfC`^ov!>*<4RHKJCGQifMv`CJtZ8!@o4eiVjzcDzxA@O#vqkRF_a-;_zV&rWzgav; zJz?nCM5|(r1L}|tK(k`OXTq4)H29tmU*kGnmS;n`&+gnDpQ?BGF%j}-brGw$2OQMu zY*Wpf^nRV~Up#xtPifyDKfbiB5ywXF3TlcBof*Zs={zG>plZ z5~fgEbRWMKZ|_TKU(3a230h$eIj1&k`LV5Y#dz+`81{pAQeUZa`S3w%uAS#w!{d~=HcW!ovlATL=ji&LztUIx5 z2W~qy83v569g+cnvp%Qrq?^@5J61+d# z#3q3vx+Sl&c%#TKip`Y`_P*wYX~tD`FJy=e8<`NWjoxk7IEIrrL(Gm;0X$9+pZ1|- z3pJIOq7gDn;LvB;LZ{--0qR9}ua3rivXJ{$5al4yhf3w;r~aR7 zV8m`i-HWb`u#<=d>)TP==5I*#)9Rj#;jsk$Nh$kX9~DQ%=1%U3{CDbN{)Ke`-U%Vz zFUQ=i?=wf3z(lO7cHShu-`WenH#SaAP)yux2VJfuuF+yvB~eO5 zp7=RmImTc**Y;%Rwj=e)4~O<=B+d-9|G2;XKz$a#8!z82-TiMoX$|e;J8dtOheXZThTh^xsY1&jenAT_JDxV{;^zV;bGT~EiP+ilcxo3Oq%!DBH(DFF;HOd-J4996V zxMoX0i0Fc@J_<-!P#!zC0~t0Na4*V42Farw9@SxsDN*wosoggZ_9hp3^$iOLzYFF4P^f0P_tz7(q?*;i$Y*Pe@@!dz&rsauz(iYN z=r$a=_h-Y|4HlMY+-Q;JTkv*7k0C0ZrX{G%b#C5Lay^84G1G67ACW28k=FO5;?dtX zE?B0G7zQSoGd{TvL34FNuB(~t$MakQ5)0=U0;7bxU&Hj*(Z}QB>p{OY%u>uE2SUmX z_yG-fp```>0dwTv=O@BD3uyM1KSI?z0*TDeclV~|;nFo6SbF8Pxl+i3Zslmutz4B& z{Jp1FSQKUyf&l>?w@r_D48)%i@6-8*uL)9aNqKS~rrM*%mTR@3{&)E z-G%SAZ83DeRqjiRtS&OE?mld4N*&?2PWPp!@$d+wHkc(TkudF=+>NId+S#U_!X&a? zr)Y2RcpCRHDv#2*omzl@s2lN%8?EeHA%*X448t=ghCbCEP?Wgfx7j)h{*?1>s8~Su0q-zDsOMstWo#l!=`E2*cQP zVGoNZ^R4peoo0IcEwozpqlK~5rauSM$~af>ySd=Y zuj-M|N~@McyttfuQe@tX{D$P~lN9d}6SebKRbsL*4!?g>)D2X#SGbkJi?}in1LuVv zOh0UAOInL<-iKA&YySIKJ4b3;7Un|Vg z<>n4PNSC|KT<9y{7|y;8mTAn)fqTktCJ7Qi+%iFWBrbHn0Ebt;mU{q$Dl z!}EW0l01>af@Q{-_g>de_?tSz9%0ZCiC-QCWdaUCNjs;Gw?G2?u!V~EAwjySH$AiO z0ww}O)OA4#?NxDf3I#`j!vu0pEUJ&bBC*@0O+tMbtzpbotTp3;N z-i4NO4SYCNZUq~amzBNWJ`7Vpv8RVcENNV09|mV2Ijkso{Zhj)5L8YXqC0;&G21K9 z`SWK11qN>1pegWuT_h`i~%MI`8m&Xrc7_shATocEV@i~&Xfo)nkxs)5u zcNA_{Ldqcu9O6t-o6s?BkNB&O)pvaox2Llcn_wkSqYlh}O5+nNWc7W2nIb6cxtJJuTKX6J7F^95= z`OUH#H;bX@(gXg`pN%nmuo7(_FGG}BIXgP?+m7TUwgen6jcN*YN70(az>0%g792hi zG8{A?Na?$XyUimio-ux)U0PN%jbOqpHm&9KCj(J!K! zbD2`G`RBPN*w6mLd)s7)9js328+9sLq~jO4g^NyJ-a>M!#Jv21azZr2(}*hR1(u?- zn*QbK{AFqU&SrU=GvYPzWrxF=;w6TTJnoDeK4e(7r$27=o_Tv4^R0RldpnbCGc7^N z-KXENK0fkT;zPZ&E=_w_;)BMotV6;__4XcC_SU^>VN?d?r&~DcIZ(k%q-B+*l`RdvQ z&_P)9jYjqDR9t+3gD*pGa=iI2HJATtP{#561M26nYiNgQH*E=;v^G?^c^ZvPN+z?=i68S1pNf#K<(gEs{h zHp|1h0bnrP!-;@o;4fGyd3AMVLlUxr>WgXcqY$m~bo)X9EzY zisF=$o6&y_sb}4&pZNG~t7yDUMMb5w`qAJk0XN=aGqx)1(gNp6a&EJk>E~F!^G>rr z8jK;I{>SMrPxL|adlg!}W<34(3t8M5bT(S7b*5~pdcwydK-NLqI!cUmh`{I-@ZF9i zv~+#v@o=zZr=+SnwouE;r`lrBfoCnJkv|E`zGqzHswQBUOY)L$Z5w?NRFkfAhx6ev zmR<`1hJwyDw4p0C;vC4{@Ab>BAuV;p+1y!6dF^aYLNVJ(Br#|oU&-=*t!bv~*c~PJ zWBh?3Yb5!Sb;cj_&Wumpm<*CKY~OZfoe{U)PlSbi1OYo9*1njs3$x3?%n7{edl?k! z@b+#EL5h5=813`dw!aS{02-dO$2HqS*vqi$djNzE2cs2i=$I86JvS_o@cRQdzT{CMe8oFL`A$zWD{F`b+} zAmwWBN8o_unKZcjp6&NOU{nWqqke;Ptx@xZEVW=Wxzr{<8h7sdS@UF>@sistF$UwE z4nE51zn)#Lv$IpT)KCQAoKHywYVkjI8_t3DATeLzwW+^5)k)?W6dTUhXabuKwto3h zD+PO6#vmrIzUM(4hT~)Y|4ssQsqZ*aY0-tpZ`$~$u+n$mQY%}AnkD2+WcXdO?KNrt zKkumxEA@yfcE>^n($jW)jR>)^hQI`#Jp_qw(>5-|m#F zO@3^uM{ivhJI4tffM)Ve|IL%^Q5w zn}n2#!1atG@EHlLSqv{3y7RE&UsmTwy`sBQs7Q3;MCRG+sd6@sw5`A$ zFfD%uy*Z^_jvr_6$s)cS3?ERxtib?9e~9|TPo*%LnlH5N!X1+d;d#@E=-bDiu+I^= z-33r_Tb}ok`>WlM3&K*PKX< zVb25=ECd9I>ruhN$hMzjtY2eZC@Bw(X2ElA=uUj&f;TLU4|w0KT~}8NEnJZb0evHG zpAOs|8ROE7y6cqYS6@XjYo_BBXE&0wd%QPmX><~h0-h6MPv8twt(I!^9i9wT8t>1b z>#El*240e$J`N^KVB8=&;An&q9-BKU8YC{T1e3fIm?AleD_)v{-Y?XCUN15^5djW0`%mCKhh6dIRyn)dS%uaA_4Fj)Y zL_T#$?wwb%1SzEUWSX=z`W?b~4ilvBAM{T1=q0?!RVmEM@^AgZR1p*a&D2Cr+VI1g zFde>Tg$Fg=`cXV?gG+7FE;%2+sP3>9o2t{L%l%t_i?Ue&@Xw<6Oa85GU#La*fYd)Z zxdeNZ>fDLKrOFD%?=}2iO##}=rfYA*W#wiF@*8Z{rI}_eu@k% z(N}74S85HMBfTa31EX>-wcB5-v-LO8-tWOnQki|E>86QDB>7Zy;A&?_cv4IZ0YfrD zh!Um%@xS%G*Lra59@w&t0)@dO#M&b#TQ;o0!_fB%!+exOG=BFKg>=^ge`x@jFRJG|GR?f3+u?qmmK10GxKp{Xn`|35&6dZ5w# zT>B*3YuMOIkOrTW@s3+y=i2_UNt3t0E1}2EO!bpXdolXvYIu4me%*#Dhx8_ki0Ph_ z$O7ALZs>R9O8aMO`b`zZ{v8d~uYNZO+eBiU`C^{@d9aYG6AGm4%(oc3&C85XH(rDv z`1D|8)#3MmHOV&0X7JS_2v)nHD07@sIl-s7mgSR!K*&(@Yps9lFWVHGaeEA@U~yUU$Oc zLzh&=E}C+=F|<2WVy zSsOphgoDNJnMh>^(LuX+io2zyW$ATO*eSN8=_55?-NA0ofesPDER|t8OYO8+ev&>pX~^)lvhVfg*^}(Ic6jJ%dF2Dv zI^~no-zV-Zm#|S*c+@it5e}KCYDeH6d0t(WdFbx(y*=yIU$WrU>T5I$nO(NQKIU77 zYoZzCNCvm0v%c%?Ht7wWQCWSkK|0;n0|V2o#wG&6=G^LzkqnWY(=i=<5 zcp>d@e~w@27ZCcnh^wz!^gNFMN!M4c#!B0YN55ln(|_)5IsN9cNI%%e=Jpz zy!5Juz(kL)-f~mvVv6eaz7IkBo1Y-YvPUH8f~5G%U|WO+1s*L`N;etC-|9@p2DF@B zfW3~KHXXmR(r?ImlH*A9VhbAj8A6w8E%M#zbQX`bSKX_hyLD^U?|YQb!5K%IXWl_q zt@ty(a=(0J-+3|i#p^Yq6w^hMSEM>wZ|8xgO{WleqH(aYMXrz`%B)R#$BZJOHOxi{}0Jdu=T|943;t1c~XKR_jB$C?-H`3=Pk5N=H9w5HKd#3 z8epz|#-0nJJ^95q@k;<>t;g#Z2ASN#>fqwyToPjpjzK1wY=U;qmdWwdkKd)$p*@K$ z+05_;w7?4k7UzW;P{BX~_WK0HYp1(y%jiNF9CvM}kG?(uN3B0R`b*bKlt(q6;u?UU zr$!?lt>)6t!OuGsnr0R?%v<>A);XmuSHCboa~T)Col$lZ_-fZwEuDWMeW{??o-0u5-qG%)#wt8Geo-3hzpXK zP0If!Dd=B@OdPC+g8o=rJaoqq-2SFE^7Q5$Likhw^>WBz@!wMR(MXFmNnrMUtEl4xxLsKRmP_e3(?WM z7b{J~s%P@)3ECGzasT1n|9F8Z;Uz(*p)PH>FvD`<_suZ6C!m+~e^yMS<)a>w)n!|& zSpBlB11Dq3eG@LxG5kN9Ycqm+Sm{DM!%H%$k~SI?|7X?I2M-xu50>lyD{pnBhgmI- zs*?Rx&Cqf+_LQk>_u<A272+J^ zd-s%1kCdWGwGEk+{#ztzm(Ldl+U~gid<-;KW{9sv3j?oQyGcOFsdeWMySfjz%;R&Y z=Om;wgg#K4n0kIyL!a1Q6Q%|RPI``k7 zB3WqVffGqab-Xp#GOeAZM>$!&dsOoT>7J$E)#8sS{?u1i?e(H1F|891>SHd=WgrEW zl*Ki043g)Rb`p_q0@@7?T6aD7!2MUa83qc+dNg8N-q0dv9I+lRgjL?M`huz7V%5NR zPZ>xm%_JW}1v4~X@pnKH_6MBkWWxscA}@EeGeOOr7Z@@(-7d7#-(|7Z8#j7>NR*`u zaozK+{waNnmFAu+QoWM0&Mlv_q6G%{DBl~k6-B;n!RL{iZFU%V zty-29O#Q%`)S;ihaT^|?v(vG$PF-t$d@|-5;%Is{;QY+jc8Wg06)qo)3M!i|CjI}^ zTP|W`AJP=?3MwNEwK$mXWg|yLy9w`<-@(AR{G*|&{7|yaetIczK?zCX&!D3`jcNz_ zZoWZY=nW3IMT+&q-G6g&3@)Q~N!Aqj!;h*L;to#!oH=gb*kcz46lqFaViP)oZvI#p{<*v~C#K19CORYZbg?>y#` z`%LX=5^?HGIeRn9j3Gx`Q{HX~ zNvJFQQ$)|>Gdxy@Dhuue=Td~o(gx@eUCQ^|rE>lD@zLP)FY{)VUj7}O&m&aQ(Ft<2 zA12<45zSsRsx0>k0P9?N&)oI#&(gS63rP-2gXyyS51%e_{ZK~G^tfqaVo*+HZcIr9;C!8Ewn{k(SNIzBR7S|>Ph+!fVxulgkkM9 zk8a6CVLE~vLfFYv-rcl56htNOZ(CV%v_G-E-L<=c)47J5Gng0Mp)-wg8iN?V1~S1K z0n=YJLw8?d`2R647@Fo8RdXgii@GDwa2U9ldyhoCC$X|CF--k7u)Kf? zwG0Si59pg+Onq4ZsNUtmp^fV>q|f}48HGf6@)eRN3^x~ti;$P}u2&8AwomLsA@;5i zFSoZNH+3>NafnHdCh!VlUHkC|hRRhhxtyJh*SV4IKa<}2?tjWDar@ZeaoshiaNF13 z`cGfRf$H_^>gMuEkMWs3Mi;zl?xro<&bczC@%37q9!WvMkDKG{K!9*LMQoek;=;(H z$KWgZqzd8VpgVVdM~v{8oYco^%#7Um+jh7+BUpKU%}5gKn9{WALbi(dx_VNLQ+wu` zc^#UnN?6|fJ1%Gi=&HLwk@-%!3@Ytz{v)n?9L=qwoIw9kgm%Qv^lip8LDsxR$8l=; zvICvZCzhF>PEy6#Hc`sPE8A1c#2+qSZe@&*f*0Pf#Ri}~}Fg{iK)bUy~ljd2gABK}@!)Ms*kIzneH zzoCPLN(DE<(`3HrzH7dUgf&#w1qzbNUpN13AZ@M+EExIaj7mf_CymiN?3ztnebp(@ ziZT{AaOz3XR7N7|hUa$uOm3_mkvRsmTXik7#~-CGuBB~Px!=`PmO;#d8P0!Z zcZ@AK-(T*V2OOYfVLikNh*8bF?|yfUi9|6wr9v(FE`rh8;-iZ18~W4lmZ89}e)QY3 z=`Inc2{+Ah9>OKhl-!DwrUi{4fdjC4m9~-Ar@g<@EgVkX;j!XGJ9NA6ZtAbd3%JZl z1+XA$KfauLfe(lEkX=G=y7QMEY4*dtvxR*E5f|jn8nE=l&DwmomntcYh<6?2bbMI` zK9p_xG9@M{I4>nz+73VL48%LgznyrU(|OVfSr`5Jn5V?yXKe;$0X+f7n$yc9PxNwVe@)YlzO|;>-H-Wi@)DtHjI@qfg zlWp8XBPen7@41Gz6SG8+Fsq_Q?hs!g=Z>S7^A!dM6_*f+2nPKMv5LihO%d7%hy`sP z3wHf?xET!qtgj;Gkt&OXBc1I~hJ~8=m365~>w*YN{)?;{JTG)Pt#`d%)J+ruW)5do z{Vfkxg?lME-$Pe6n&>=)uBb0l-ryS8@|r6oJ9i}Pgxlq8FqY^w33kMZUhCf*|6C`0 zeY}<9Kb-}~J#9*peKRj`+pyT0tQ-%MWXrOvSPxgb#vtK%$0uB>Hk?l(x}TKh?I)j0 zn}n_#J?K7U`NPT!IxT0h*p3dhgLb>u`y`b7{DOYK)Y_hZO~r4$xSKlDc3qUL=XsWc zh>-!x(J0`tj>hEhmu^Pytw~I+OucWP`u%j9jx0RaU2lFrR&^5%yu|K*z$pZ^(jQr}zV+o0G_shYlv} z9t&CV$5eCk4X?@^CffF7I#=-4wgHT-Jil_927i7X{%d3(3@nGYDp5odugvGd-MS$q z%bd~(hIY^H{csf=4ph#T@p>WfE{05zkDuIctFqVe$~2gp%RVmFq-l!wgroz`@E0<5 zJTvC_S4mgZcprWA?4*z16(ANHmAM?*2`?d}G+AU8ad|6HKj@?>0ryuRVftzbgt7lu zQz}C7Cp5765Bic9y*}Y3CwsZ(xcDwA^!LR-$FBpg$Vps=)`L6A@~03VgVFs?jA%>x z#)p+Q1gTV7j3OUsLb+t4#DR;@iE^6W>U^>}UIq?(vVYUj(i^=F>aysKXPw<5G=%pZ zyXA6zKctWUKTBwJnV|8vM`eQcBU_|>cX=ts$kYFk=lr*9Dr^6u9S(?jGjz>a=`-f{ z%_N+D+}v~?RJ(^_&?;Z}1#MGZ6icah@t1gHlST&)QL0#}oE>@Z(+dobp&O|h3E<)I zW4;6!6LUG=&%@cliiE*}r-k`T zAUI~}xl(70{k!LSxSZ^H%X{-A=(MqS#`cwlvY>1`)UIUpIsHFpK-HM(fe7bQyh?7e zKiOwG!J_;A^GrUv9tDk0=U)!1lu+pxQkgYGNYTQYSp4XZ4ho!j*aSb;I5|<4Y1Dkm z+%NC)P;u?wYJcjUFzvjPt!_nFBH)3u6^i>LIZlN>#E=DD@*RIOn5R`r=7qY4>+t1G&GjN z9eqRxN}~~vmb4Tqq{71}DLFt?rrc=4k|Cve(ig>YHAb0^Ph>T}B3Ae|Iu*3+-@Aoe zoSY^YjBL7hhd-~Ov2}HeP!tNja>HH1^NDp}X8}1X*D_E=ohr6$RA)hM^zVueRL9B- zIjM*w^$&V}TjCO|sHNP`PeLWB?@*MQHXfUyII-W~U#{XQpJLQ%6qVec@z8cnqp74O zWF*Vxx?mH2Oa<-0J@3DzNeq(1G+UxccL>)0@|#(tkVU;G)cEi`;lK|?0k~A7*Vi)r zve%!>EAE6#lU*>P)a)qy_4dxK^ZNmDEK`cF`~n{2+MjqENP?TW9T$aN%3H?u9Byn! z_VN1i-d3&f;;!&6i(j_BgC@jS3yH3b-;3tfHxxFaAlK%gW_3fbR?Jk}C6f?A?urH1 zSqEoPM*=xdn80rY3ddebNely34s(JO@ro8aHG{)n4XU_ID_%u6C-^6juC85xC6Y&tH$< z9u5JC%PC_Ir{i%iuUF%5cf}AEl$W`h;__QQWWCCcZ^O%)8w^vH$GdGC1lv~+Zhn6M zuzwo92K)O8`GuwK9pkKPOxVQ7Ll1FMIWfDQVS+tu^M#D5*sS<@)nL|O8hPhqS7eWh z`|vzv=M4Vz?J5-od0lR5dm*!G+j}l05lehBlwb0t8rXelTq@cl1QEcjuXFnfR$`HW z!r%DYcv4{)xql`5PFwvZJ=T=NODYsiAgV&41s5t1{;;M_gwOleCitD%5vR!gN; zt1hu>UkfS;$m~QCrqd@K+enYv9B`5nhu>V1wz7mBbJe~sxblRj zuRYxP0pwCg*Rt5~p&;eRkoF@q+&{)|pV`?z7+_w#f@9AyctqD7Ppi17^eB94lrhwp z$ks*ghu(8L(q+A#;VrYgf&GQ{_&23o=h4PQ|B`2B^n@|W2R&>5ZE&G7!ZNL-CLlnw zh07WfN)8>c&wAo+42D~L2FMzuzGvc(CZBR@0}*YDHvbZg^p_|qYsAM75B!2WI;{@&MRFC_&bUX( zXbXJhCgY+f?kti))!g)^pV{=g+IT0O$QM1~d|}pNE8ta&Q{x)&56B*ytKoY zU~(dlr{b|(b2WM62ZN{5^grU*& zc-FqU3SDl6nzWT%k9>w&Ar4ABE=$*eJOspUNx+va>XxV!0K?i}gB*{tI>d%L%XWv{hsldjytc>a(U)|*-{ z=M!HiyZjE24O&gCYgGt0vM?B}P>^Cu=N% zf>_^6KqpZ@uwPmqtm*nw2Y{>BS-kaeWT~uK%Oac_LRkl5vHjjro`nT{ary3lEZ{yZfj3nx$FOGpeggSzzibCy`*#@TyPZ zEFrP_AKdj5zm-ZUS5_*kUYhP#0{PsA!kux*Mo%7#34x5{t43tWl(@)dr2YrJZe|7; z>dbWoT$%Xl%QEgY1Y&C4lh0DwYDKUFLMKFf*s~NzaX!2xH1u+040MrNU-w5dV}Cf5 z*V;bd;g}8QYt3p?!oXul<-Av=@i<4T=8a7^PHtCT1eZS4p^%~@yBXdcZ>|lOaO<94 zI3ML3%fqJzPJMK*Zb|=Gxc;AbdLZPE>!DEAR7NtFKGJhrGQ)FKT$ays7QI%`8g>xI z(|s8WN+*o^mgA9tklM|A`>P!2NHMLz zrF#-3!nekR8C`PXp>pxYvwo=!l!f!0zg*L(s4PTMKHZpxeHiiOtlHC+k+jda-1u{_?u|!~B-lSN6H)68f6?JL zYo0_(x@mL3)iBQ*W~AN~@_bEo5$3_p%<$9e9DgDNG|a4_{!DJ7RO7$Jz{=sAOsIJ; zyzcN@;VJUD@ZO@38)SAYL4bL+09rC>$KYFW53fWZ_-tRjgrI35oY=i^^6rXBM!WgX z6CfbIIq8Z<6R_C=af>-WziINmJKKm!&1ai92~32)x`n^tp_-0!Xxw~J_)D{1z9QL@ zuSYP!AVZ2!B3Oqxb_@4l&1APsuRAtjY6l4Rs;epy8G@!+3*r(q8REnlit}&~_oe8X z42&-UYKs!8usS)RIqS4U;AUC^#c=g-f?8uUKW&c#`B-!0_`a>P}Ah z`VJ3Eni*E6Y8 zBkBv~wS>zrBKtxzoEK_q_kI^11`#%1NiX1S%&iDwjlf~6x=w{~%XjQ0rOo*Ihqa`H z3TFK`NV3Z^{^+LUkT(!>GE9D4N|N2Fa4rJyoAq`QaswxRDN&B32 zB08Yc-wZQ{4-di=FY>Czn65N4shotWL}5szyno0;icdDW&6~Vk!+sW>i*#sUVj^bK zBO77JmZjJEf@^R^YKe@kih;_|EzEJ4(ExJ{qzvELxV%9iBb4QigSW`|Vkgq%A*ITS z33o;#SSz`LR-gd3ZHW{mw(tKt3@8BRllK>z6UgrCfJiB8ghnnTFV}+`jiYAEQg?iS z;=PzH%;gs!AB7F1{agp0j}c*cuhuJ+BcWJs9xGOcH05MNDi;2(hDwYhzQ;AsvVSlU zZxOhR+AcCYOR4z$(9^TVYu*OIL=#$wM^u0oJwQ582g=aeAy7sykzNcBow>42JOaTV z)B5Sd#zJjpq7WEufH3cz_X7?z7h@n9lZ-V}=o)ZlXdees0M=hyR^su1n3JgVm4Jt( zp;>w|^P%^O-b=?rmcz~7Fw@MpZx77R|H*kX=L_<|QxJJc*tl>vW{VVCgfjO>92Tmu zxTVB+7~*%1!uk_8#=>|Wg!&zXYOT|7M)t91c-lk6yOW#t*Y9Y#Eh2|j&}0%dGYL#` zCM~JR=;*iL(OM633pXfo^B%VazlU}G5#dgdGF5Rf_>4p_g~eV>Im|8mEC;x_{Grb4 z#o2h@O4gRhdzfv1<)f5TtUaJNwpG&x73mH1Bg|*Ggi>NSz&HzrZrNKg_Za4k%Yb`S zkT!jwt``=R2FAM6{XTFk+8+`dSwL%>fpGq`+YqSNifkDL?>Et)@26Nwoio`t5>8cZ z2SC`2L^JDRxpLaklqQgeFT(U*b$701Z1zZ4R(XQ8kxcLSc9^4K2I#YMbqiFPAC}Tt zF_XRVV3czIl!*yfjl6T-UPC`dl|ZThHrQ@`&Ub44FNSA7-lgUcgF0NcqqZl8z@Xz! zZzrfd&re&jRbm)5Kfe}t0~53*$g1oBKqCwS#TKA-?EtTD2THQDoJR`)=koC0xhGUb zvBiC`mJ5;=jDVv)aW#-L;!PyK4}1T3UfOhDM&LRHTa_flDv7UoYd{1ZmoLB~Xw!?y z$#9Ww(7g*k8NIP-*}cmXxO89!gkZi#Fe^}F;A4)e~r1+o1*gquB}wP}9SL^Ts})MI&fIS#8e#>eJ!>3_^}{o@ zmChmd_(`O#c($xZ*tbW~Wk$=NYCH)jR92WXBHx=f{XeeWIw}h8d*h|M8!15=DUogj zm6VX~?nb1$Q$)IkQl!hFyKx8sNog5S8Wb3*d*=Op?^<`=Km3b#xqOE+XYaH3^L!qt zy5a|J-SHJ}c@du*jL->L#6p>Bd>(R?^j1W3o#MqluNF_tX^R0$-=RD&XBXY?x18enT+FoRV14~dM@LvKI#V3%Eq(?6T{8!-d`QU37KhkIG zTF~e;SK+Bpok3@jhL(7;_uw^9^i;e zDVC0Bm)uZdlXePrk~gTdf%b0?$3=wK(iP!p986oC;3ITo+t426_QtE>tgw}2XZsvI zDlJe<5{i?yY9YzpZuBj<#X*a@bgs50f&^S@k4%iM2>RVjy_Bh0?B$rr89wp#+<5Jm zNCFVEB+~bFEp1w}=b`ke3Y;}MWQ8AMq($APmN3J7-tWg=FMH=3QgHGnzDq>3_iwO7 zk+*Bnf-Mltn~&vzM|g@C^2DbN61T?CSZJw@=X$luWD@ORSkz%!TF18asSj+YDrI zZA#cT+4;12_AigY8ykIVkV^EZ6Xx&$c8RAmO9_0*F3}1Ny8H8T+>*24C!>lLjySAp z*471*o1OlX#hO6FYkBl&y1VX6P7-)y(FI>$DBa)Px>lWe15x2bo@mQtyU%yRG^c** zX*S;xqPOa`CWf_<{Zl_11T7%0QnD@ivM#HYdS}QB)ErkOnUdGraEX+Xa>o_H-`r~} z>&<+FQ)S8twZfN&7F#JiDsx|z2dp9ESlCqRl;!#L!B&su zT}?syM7GV4DMtwQ9jps~VqcBbn+)dXZ&;_p+#<^ygiSrq+3vJ7!(01}IcjX5*SM1d z`Ky$9R7c(5E;753W0#I=a}P z!HSc0g>}90Ex(*Q3e>(zGkgRFewr%oM`F;3k^K}GbUxyY;2TQPA8J#v!9BeH^_hJY zkQuW`NIYn0Xz=x5!tO~)N@~8ooc^kd!z7=q=2A`mUDM|~1XZ(zkED*R`(?_a!^S8d zM}0{o#cHMRlqXsx!*+O0DXW$!Bf16kc$0f@HWTfYh(Xl+PHh7>PVhCrq^5twzW_e@jVr z!bLj|Ow9sBwcXPK&_+|R8=;+K^a<%oi#N+x+M1HuGTc@v%;!=>kFn>8(dUB*Tm({$ zUx+i4N9$|Rg;`&l?91SN*wU$|*KCZA_nW|R$N%sP=fghMYbo*(p8Vtjgd|I&0X0PN z(J?0D&~H@XT7hjZ(Mofy*K`)oLm^J)LVO~!r7iwu=-}s={S!S4KW;kN<-WCR)Kl7p z#NM!l1u4NciG>GD7|EZnr#Y)fol^-c-5lUjqb3$%KyQu|ZL#$SpxjA

#ufv#?{~ z+XuUiY*8naCKmfkvnF;hNB4;Nen0(QF-Y}nZN`%X!mzSm? zyx{7&o-V_kZawWeoVdRlhzGhy1q|wTG*5=ONj-X0Y(wT76$`FXCtZkG;=ydSOqwR5^Xg3N3w2F@X*uh!%0#L@)N+yMD#_7E8&$Z+!yIj{h-Gh}rRsH0+W z-ek?hH?=5PwpJaQKOLt~c{RxRrliz54PIi={juq_fg!|z z0pg#l!KL`G^C3jAcFOFxvK|h3wEGj+4J&~SB>nszo;5y> zki0g6KgY+68Ld<-yM1WyfNs?SBvsiJX8An#{yq2E4~RHx$bpBbJS^uzn^P_8`Pu+- zC@PyPf@$={%UP3@v{v`=Dn7U&Fo=C?9TtdMPZS{5FKX`N4wPz)YSF)L4rl-!QWlbD z6GhU^JbBA_VX>;brzZ=2U^|>6yxVPbsQ)i_IZN%Vy*Lmjo3;W zn+#zI@E^4SxS9wwNCcwYMZalP5$cms5*JUTb2{eCVk)7!!$S}iC0?wzhGzG(ya>!P&IPlMmsK8ug#{zXTaXD$ooBcto56-89$~j2!~evt51O<- zs?i>QSFmAtk#e(wo#zid+ewAEEOENK^X-f>S{w&iEIeHxdl$j8^1y@HL##jl1Q8Yi zzbSvoC4;Fx_h1c+a~75JI~nI138(gKn;NCS6frN{%C zsidl+g439!|4>j+(D%gIZN1S}wK3u`b7O*(|JFNx^L%xph6g=L_-;rSNz`a&efh(> zUuyq!9F)F&FHx$UXajdmz4xxhfy%H`3=2>D!5PR7jjNV75=~ph#q})V8IJO?H}0l4 z$b(>+AtCIVz9(~N&app!s5!0fFy7<>CqZi>a)D?0t&1NjFMmO?vt+V(pplwjkua0 zyPVrzXz2#+sVQku4ylOKLIqv^v1;xcLEibB6CI<9+*__y;XJHn)!ZS2EH!Do^F4qh z9^rrRt_BHKLi?yr^w}&d*Y~zpL%jagzM19 z{i@N)507!1-40D^Mcgoevak(lBvQDF1cw)M(c$PHkh4%Cd-c4*jsIMQ+hCM(0-kfr z!N{m-GiJs4dh6IG_jTxmqt$W?_V2m)XKHGEqjrOel4(Fx)Rf61%!7v6cGnvobDbBWGexR-RS zm7{TV%d`j}OC*%p8I72CIzEyxGh51Havk|YFrU%5ZdxAj4 zA{t=#W&$*WD4>%)9K3YzJdL@;W1hiEC!PI-4pzE)nxDhbs)iB~{%?G%UA}3RVrjN| z5Y+}0N6vJ!^}^S~z7(OZ47ZDfW9y@*?N5+P746)xg@ZEp`!@XS6FS?2Hs1i3h7sRD zC>ul%t>H9KD8$2kNKoji)iufAUG@!AJqn^fOd%s*(xA;v?t8aJm~MG0c^ze?!`ix+ zT)4+EW=!d|3BE#L{vE~ze>!?5iG0L6kEu}B*CigP#sRPz($Pypn5W^AZjMG(>f6oCQBdulpSs-ew+NhKpN zznhFG;i-LA5G*qNIo&w%{e}uoD7k~O2Fs&Ihe&i%@K0YN-@l73?o4(QxzS*6F+^iH z+F)J#_rVYs{N3n|ij6ZT%x(FD9jZo-vOsVB+`e7 zN1K838x7~`u()}Ki`@m~Wuc20y=BO}%K14pYM@f3vnUG2HvPu#UA9Pm zLxcah9Wr>*4;*A^#=AoEC77RbfBA-M1npI>Z2D%;m&hbplTyU-E`>v!rbQJ}Qcg~8lX;n8l z=OO|kye*fd@#B_JC@S%y@^S(UlrDpyEylMp^;eGP&%0gL{_whWc+f8qx;!9y96Z@s zlo-~(6ZShRhxpsJ+=O$O)%P&;{cu~m`{(DVs!PnHlBCRADMu;zz7nS6_N^8 zcPn;(H9t*FN(P48qCPwyy|QyH!04mA#HY@6ymETG&ZWaXKlrsnvzgQNHE8&L`n1Lj z(b$~ZS?p@Kmy-1BgDJ{5l`Hs8Ndp5`jVtx&59~yVR_PPuO3)fiy8-2k9FSS*kH51t ziTh1cx}E=1(Lc8u0`ps&XPm7ibas(ga({m#F@NB)k078nDX4ssKSm7w11#u6O|K_# z(i9xfZ@}~`{S^S`1k#b$+Y_J6sP=v;lIYkd+ZbsE+#sz-`n1!~nwEigt`mfNc1k0q zSG6yE$`g{YN*4J9FhDmQ@3+V(z=ySX#zWFJEY43VVL1_svUrs)Zg`H0NI)6+{B%?q zYW^#X^xs-r?+=^~@ zTt$BQ;@<2?zge{Y-`11IPZ$zv0L69xhQgSV_xB+Q3wFuVFi;)RzStP0t&JfeJJ?=<2cQpH;V-b6dsBAvOYVBA;948<-NG$aD?o<4I5tzD zlw^7X74mJMcBf?3o$vD-H_TjF`-?nc0~w<8ASk&SzW>BNX9~}Eyovxp%~|>$?yT%% z$IHUjFZ!(Ba z+%=L0?8+Z#$LKFo#99X8$us0A``syc^axy+n;T3ek?^-5;5rA!;L_E30No)^b$JBz z`L=+4O95a$xl^{Nrly1lv`ZQ$_9;XY4ti|_ zCV-RuxLkU_EeMY?Ms=Zew3`pj&ba{5z*F8wDyaIE;E4{ZGFft%0J$ZVz`9P*o8>C@ z4YELbLkU0YXIzpG<*q{9!HW!TT^Hk2WMms&6; z)?P3zF5nfD)qO#dcIuJR72I*1yP%6(3tVd&0GZWp$tR609HR=rxN6Yp*Pw2LwAylD zUFQ&k^aUs>J?M_PUsr?ySx!HoEsPBev9mwTj(W*OP^sn;7y(uQ{xa4v*Am8Wlf~`q z=CJj0Wo;U1XY)0GZ*ZT#V8A@nL7}*G=dvjbvZE`=?1Y9#t?Ta-&oSUo)VZ2DuT>wy zljFs)UbpqS^bq2tF4$|`;A84#1fYM%Mp2pw%%ZWRoVdQ0g`D|M0+_&9-a2j_p8lKr zZs|2fim#`OBlH3zKA81igZQoxi{fKa8p10uWgnk9{{(ueC^1SO8pqXK-U46&F2L*p zLKOjW=YE#R>9Y$1fZD;B<`w{qy zd(Ft(h2m8f@N%|sHf6KdA*x3sOrc3&?elXGa{e7`c;W!^7ADh(Sv(payHBCn8vrYa3x>N%O1lUMiS^+~vw&YtrU>~=~20_ji;Wd9CHAlc-90%x% zNT!W{_8g=pz=@1VO?$8oqfC;@_|0@?fpYSJJ{t}8V6N!qJH*Z}t1rRy^{xb<_Q6^cCGE?AC$qh=AiRG3=vr=}W9ASP;7)hN_xZnX)t%}3d@N-i`LOY< z4hJzhp=0)||9sOMWM-FF;rs#7nS|+^SGN>$)@uJ3fP^Xd%wCIRjR;L)fYt=btB=yn8o9aNq?L|0W4z(!$|<{jF2nw4hXb$V@Kn0X$hkiB8VWgD6+jb}d+L zko95F((q-b_1J9h@~g1y*NgGEp(Ius6nM3ZGxft6@Kcj>OLr`aryRtX7O@ldcDsre zQ}lI*Wh7g#PwvR4T~20{PwSr2D*U8ZfJL1k>spI@k@!ft2Ps!1CjD{OZg4um{yZu` zw|qSOwictffhNvC335~0jJar8fm$-}24Na#NEQ4JFeG=r>g-KniL) z+L0rJr4PV(8a3*`2_khHUvs%X!>cf+t}t8GZAxM~HmTuUcA6RdvOnq@!IMz2XE12< za0P>^4*PX??-#7cOK77-345OFRu;qGdutu-s2KM#BwS^{`xZ6jIU3Iz3YE0eMT1ho zXy#9zy}`FTuS7ZeZRNhJhLOPsUZt5?dRv_s7{hjdc|?()GYEgnAz{(l(@;n}I_`N} zE;mgBVl71!tV&ok&{^Ar3h6(vJ z;DbaqFExort}UxEfZS;%Tw|~Ju|LAK_O`{HxY>C(ic8_=(<|{$fq88dhFY5dh)1at z!Grv}&yk?pn{f|zA%{#Th!8goPb~Z?Q!zz+ZNMd# zhU{wagp*uCjo4DP?pMR6YndEUYJB?TpGJK{!#wfCym%k>-Aabf!x~Ks_P9fFcU__= zSaA{FOL}gbk0yU{!Z|f-n5OrbIYu4-CFK+^V6ePJ+wXZY6$ra!WC>JAGwbfxFj3mA zdcpuEwjCGg;hC|WcE`pJC5@-O^WWg!U_4uQWk{-!AGWnh2%j@(!pfFmFzA>E`;!x4 zpQWz<^(xC^tO7?p_aRcJgXC8LE1*@ZfcXRXYvvDC;&l32t*(8K^C;x(71t#Fn3yRn zl@w8CKZ;{l|BNyK-%~Gsy8x>pG#`gg`;^GfJHJx1RI^w?47nap8_5>5?yO;Hf{;WH ztx(iy`<0hqrRi+icWFIZC@I90_Zd{H+yTfwfLDD1@K`h+psvvf*-s9FVGD8vI=TF7 zGEbBk&@6_L=?X|I4+srfEX>VtO53;e@Y%_bTkqBg%Ahd`bNYtXdiosDJ+ppfOD=pl zg{)AvGz9!;8DzjFI!3$TLt*)dD#?MWMR!lxqba8|eiKd~4O5fgkJkHalj{ErdgG*_ z=lxG#hPvWytp78~Y|*;^MqQJql3fQ-|D%rrZ zg*EE*#q7sg1}X-lMLSKLO(v^}D1l)$0&*fZdpDci$%1)Teb=u-)ZMF(+u_s3{9fn) zx4)6_*E)&!SAEb4+qT<}(NY_X%`=Km zfdWcP-6;nWaLChV>#Uyr$5-PZVt;;T@p7rK%@}7M05yf*qetvFWu*i*~ zWI%W%>^?Uk_xU-T&<+3RDNj1;MNhjUgXa= zYlD;546$!4l}CR&WvPz>rTW?R9B0$q2&7gueMINQkv zT+okdIbK%vwur0K?NhLxw*un?q`~dCDsPVC$?~Z1gW839xfE|#&dFcQrxF9 z;Ez`J6wS#Hb?;YxbP$zyb3rFyg)v& zydzAUG-^?64=oJDo1}XXCeJ~r$9@l%uNq1Uw-I?8-P$Vvqdk<~b8B=Y2jpUJK2q>( zldRC*G0#fV@kwu(&S|k#g#i4kE9@GJ=V*z&V3O^7rT;5{-(l89u_l(#*K7r7LkR=5 zH4qq$%r4>>Mj|J?nfSZ@kX;n10<;rK#Q%wr1RqX4@ktpT4}Eb z{M32%eK0Kmpsb+qmlsZAm;i`UPsu@$vfS)wjokxqMX86!{5QDdbT^!*r9*?4O4O+K zMTcZmYBAnSFA!&w+8N|(=ilm;lhqz&@|Y`U(B~fd_JkkluH2)F++rMc@l)j9^rfB}qkaj4tsZ)w%I58pngJ9~$4$EG#fIlUPm)bJVN z-nRW3Hj!sqJ=0J+sE7R35N#7`7(P$I2)SPsX9r{X6p1&#At314VZHlSxwJIrXMsOF zrxh@TTo1GW?BTNc8KJ)v_J5h zGpgr4Z8(%nVY9lf2h1-aR2etZx|snmdIUC`A`5VRb845x-HLKGmu&n7COuiMvG*hi zH9-0o4|oW(cPRk$3aztG{flHT$^t^jEuw$Deu&)ldm^Rxp~TiRj_O*|g32Q06g6h2 zNA)+t7N_aG-9rFecVJHR*#CinU17`ZtMkrib~F=b-|Q*p&$D4(tKl9(l6+QoVbdk@ z6jBA1tg@JF*DUsZQ%2n^?A$!W6z(<}`9$!DHFQpAjS`kG1y>1!49^NH2svpxC@@oM zu~V4dtjsF>@H|c&6jP$oiF+KeU!J`9=F8JvmfjCsy_vj7KB@a?I!9{33V?AgCd#(97SLqypabkG*phY^iKftY-X_NTgb=S5>W-bl_$Hdz=OZOcOK>fO_??bWY*k|IK2XD}9V#I?>L9I%&b>1nU@x&j zuLkV1#0Ndpc~LtULl!~2p)GlC?PKjWT0u%x<`??A$CC?*2ddvn__)@};t>#R)us>c z(BVQ&e=se^gnE~!!XuqK85pq_(Du7MuCt_%2laiLxn0ms-t*=!CIC-EHW>>~@iyg( za;6*l2jL>sg!(hT%3*7*(I$63m66AC20!U7j?o?l(2=Ak%m^0p7(Yjqyc79mp41|U z+F^&17m9fSL*KGtV!c5>=6bU)VLvpji8qFt>+8h)4Cfv>&ZT>lgmn7oOM#mJInh1n zEK(+uI6*r7^!zej|9GgV(3&0m?-z0$xLb65gzK{1o7v=IJ?TsI_6O=Yt#>~U6k_X@ z&tfVQs}UcqjsvKE3S|_-57|`FK%79tvjjgbxGn*UB7C5qXj7JEQm|`qnCK$=J6yJGOB5?(w@_*oU&>H^!!T5 z5^@C!P-oG*C&>AokUt-QB)9fK$$_3(4B%Fz{v57QgS&6&WOD$y06bpeXad~d{g@$W zEYOB4fcE-7`cb>m$~sUV8E$5b%G3u8(TmW%wpH$iN&L+Y6`T87SnY>t@UPRE{AB~q z3N~Nh7;fX03G{+zX|xSHLl(WlM}#C3LLiSLOY$D__vdNpsJBB^TlgT0``iO>`CEp? zEA;w6J`bA~jue$)wy2_eu(p}4MS-v4ADBCZ*hz^hcHd1?dDlu|t920=oK;i$<8E^7 zxq;X-@?SpO#VaNACebHMF4Ute$;dr2GT6hFZS(@}!Ui_`S;m4?)~LgjJ{y!Q<*#kG7s{c6e#JCnHTa0D@xt2vPILF+qfbM^%1TmQ_cx?U2%8RPPOgs>+&i0M7V z(WtDz%G`^tb54L^-0#mXxApaGFXvztD0DV(D1)5NU_ z2|M(hbUFiHE6ckZgfAGHj!2l6f5^_JxfTM94achQ+d8Gs39sFefgLYW=ue@A1VuW- zu=Y={ic&ywMu4$p0rc*P5ZN6}Kw$BQX7F8YwJoC&Hr zcaTLL({V~?0VnL8RPJrAg7II(ljl*?5H?@aSNO+_F*tw~g%gvtVNVV7JXe)CGZm7d%er+p}h$jooWB|Eom<785-nHOM z(@tRX*7MV?F%r&u1;{vO@hsMgTm8el$3?+~6z>bQHL4fLFR}Irz2p~YSJGI0JcklX3_1|0rOsV{kA#WBQ z3YS$M`Uf$q6!+A0b4AEcA+x&iWNg)my4p4J;GrG`o?4UaPmpOz0Nc-i%=!U9%?L0v z+D6LXr=smX_Mf%qc;aJRQSb^Cu%-@ za_nZYD@K0HAb4$@eanw-yNr&=4i>!EDD=5&6sIZR2w~+vqj;Q%?(e+u^2$&y<2+nD zc>*qWp0n<#F^Z5I^s`u(I{%^gLt*iG(-E)Tk8ynV`UR=bzFYXu+rD@xe2wVf3O`PG zS5^1CDUSdZ?@Xd@O3v$q9+x0ZyuQE!$jAiwKL@7NTHYy+usLc=q)zY&Xh2F2fdi79 zQdf(0sI>%0!bhhv#-8@KM)Gx6+qeUZ-M>G-B3YCYHjr^!o*G`dhk&)!|DOn1gp&Jp zJ>?m-nP0JM-W_0%Zi2?J5tFyS=Y+90FkJj6r)KZUUf)He3-e+}i?i_PE^?sZ7_B2L zaX4SKYlNT6h4b-M)}BfnET%wW?|t4>e%yjtATKg<@ti4D`Xc+yjI9Rp@kk@-TYZT~ zsK@t`j!q8^1XA1Vj&cZHt{CHz8>i%>&A|42p-=!0@zR6pv|~k5mH>q{1)%O;2kwP&4(OM~_75 zj!NI&A26Xx zJk)Kr2ICzTY6~V}aYK8BbCTigp-4H%o5AMZfEwU(ahwgS7l4F@zBP;c(*WZA=t2c9 zaDGNA&68#8oXUKtC(j%;Y8}1;F3M7nStcR_JW`&Ll3m|LM6rAgNjRQ0anx?!ABWx# zfbg#kS7~8DFU9L+>ad?;G*+Gzp~1bU#nJ=;A}YEpk7Ou#zPK!rAz9}jA^dat^8G2A zY6H1@2vD>KmykbaH(lI{j3>Z@W(RGmzo$m&4tLo`>PA)|oPt;3HX9>h@%YrI(S|o~ zbE+W@iJ_vA15u%}EvI(9m-6_Gj*m#Zj~CO3y?0}*&*4@$#IklNJB;42M-!(>L#K+c zH}4}yqD9g8Dn9*dJ)zz@l?kC?77WxnTO^|p?N{x)X+TvQ51EpCuWT{NOGAwG_av0=`2+$H z>NP!H1E*ITC`kvGT$`YzeTldQ%{wkIfqZVE)O)GKOjWh^3!=JE6(wMX0Iy9_DykUw zZUlU=gxfj^;1n1^+tR>NJ4hxzmS)sPIhB|KQ;YK%sJ(y#olMz1d-s1-KZXDKJb_*C zTu=v7$UgE-Eo8A01Uotj$ozkJZ2u8=WRd5b)3C9u<*#{m2ZS?m;aU6IKgwFIvYp_I zyp06c*^qx@U$pmUhD?cn`LMl5nn>T1f<$Qx)RPe=lq_+Bjv;RaOWS{dm;s!PNDA8- zKbZ+z>M@3X=c(rOQ-EbysmjL%PcQ0^k#k_0*v0!fl7r{og3lK?w&y|^!ZN!lOavYK zMovyV+^2f#cbIu5e!Y9M9(=i!$zZeI9SkiAorcK!0n3Nygz&=RVSJw=AvU!+y1>|k zMZ%4x-BabRt{_H^S3bXgTa}=!LUdC7@h7f{Z_ZTBRy5w7PsI>!!vvGj7v5>}Kv@U` z#63$4-r;_F-DXzLyU~C?+7!xsc5sEgcvBS*cFv>1SRh}OmZLRz1oNB$FI{uXR!2e}ba zFWxfOD&h}Pijr^CH`?jBlynE6Yk;|AjfKA~RYvzr)?^xQsMR1D1EmP=(+gVqgg4e`J{e>v^4N$?pGPf_#X={U>+ zzqJpq#SW{V>u%781yU;EdSS+uW!ty#sUmkJO{t?BI`q`2OTNPNn?mpIUc{JKv=T$L z-YZ-+0?(BGAqENUhOPjV_CxK1K~@4|BKFKAWYm1JI?cfoqT=3|>EL~&s>iApAf8qw z8q2Y8Ya0Q~+Ok&Aj@$a9#N1dpThBy48y~$2AY}X>o&E1C2i8SzLi zp8P?Mlx92KJoNT^f=ewB**#5!+_1NnoO~$ZIp@d?B++a;OYQXcKyu21#$(cYq8yIf16N?+$D$wW zi^gOBg1G73xV#ZXkEaVMSlsBA>*oL+8Ou8>8Q-=*y+Vv zoYf0*gG=)lzRN4`%xYV znKk3(#PhuiQk+#`#^USUhdyJEaHN~*i#_4~4GnOq<>+bK)JK1MrM%64J;-_AXHA0h z=PG!E2K<#Er*aO17%^RrAKlJu$EWZKLUOR2F-6bBKrNfx`$pBfKK;dw^Gl5qjRFbe zWa0$$kV}hELo^gJo|;jEu0dt^x)#@fu$q8GbYA0rioyS$#FIe9`;SSUavR;v-ng$! zaT|<3zC>_KB$$GbX5X|5u)l>?M0rvaK#^Zrsmz;v?>LzCRH4{ovjgsgg=$ayRhH}0 z^U?LatE-y2kj9iMI*h)HJHo<(atVRE6hh zwSa_4W9?}QQjps89>Q)x#H{-1`b*1s0MK;}dV(F zup7pLFsbEE6^-!c{XYM!5Ah1_-V->dGGK_@;Z*yf$7OH1s>uP5%AtMl_}lQ+rmkeb z7}+~{o{T<~46^C|?iV}+Nj4x7>6IpZEs`)oj!vQX_h4xj<5t%N(8FYhqR|Q7thgs#py&WNk)Cc>1j#_UnbGbG(ACpRZeWkYEWnpV=xz_fYeU8C%W8;2q+(sq(Q3&a%J#LJN!vUON zeMhru8WP6twWM-*YkpznxDKg_L39TgZ3cG#d>XD5uzB<0d@Fy$BV&2|w7EAlQklG+ z>`*#h`b6@t2sfKOe#8U2cy49_vN?-`!4s2hY3=ne|tM!K__QZ1M&1!%&cOs_ASj4$%HmOs0ihO}Y z>~weVhc+ew|2=8J=xJqtu-ZgIa9xM^8>tGJq!?XD)v|atiT!R91=8z4UTi++9H}Ap zv1BI)NMZYdm!bskwdJAm%l;gRkeD>7I|IE&Pup91;mblXA<;#52D?_fabeB0zaYsS zNysAimYtY5k(dFr?fvP0rz1?*!F)xGoE0U0@zJ~a+f64Oit%K66kmc6UMteY-~kBW z_y6K2nJ63@@~{|ObkKH>U1aa(bY3mdY$69-&K?JqeI__lj_ta-wI+xU&Wj*Q2`U@A z|J;Z9r*qD=Mez%xle_`l1o74f**CHl$5N1Xf0r9E&1RABA#SpmhLXeRO>TnZXI*l) zfnguv;>zNdw$_}j)}7q3fGDg*q#vb`?ZV^0r{kMEL+`ZT`*5)H-<_l3xPKxnuCwTf%JezZV9bDbs0A zqwWI-;s}WFv2W2y8agB(yE=Ek30%>nDR?6BHoPR^V3yB}m5_-pOhKz=jtuJu;q(D|=8f zK~L?@B9)2Qy|&0XF)Ya*zOf3#-c5E{9hsU1zDw7Yhu9X*Y@NdLell*e0l{^aUV+HHznh4r}th&3E&PM zl3kP$!s{mHaPmZUMm})aPAn8x7@l+HJ{1qcqa<2p^;0Q5lDiq5;*zbqalxw9k;hq8 z;@&v5_%V6imI59oVy|d;u=2m6E08hd!8kmm{A73h6)df80@Ft6#iGT1Mx&Ni6zaY0 z<>zV{lC{fI@~+3Uz-D=_kR#kbB}H|$?B_PB$C53TsP7!up_CHVesR8&j4uasEUJ zAM7V`ZPS%u1!Zq)gaZNDq+v7n=Nu1gTfy4{_9I+r##IIlFK0b@#BG}xDg0DlfzMyJ zLD&7_j7O#{LC^XMNysh-1%7F3ZfBOW)rwKQjPU;@Q3I^k|47tB7Ln<9V7M`%@v&`@ z?6DYXaqm8mFg3>$==s!BV!wby0)>!+ipIcyq&F(#;r>Eh3n~eZ)ZX|T zFkmivT~PRdwG~e$d~EB_uSu3xS7UqXeA2R)9z(P6cL+?PS`55gNTY)~2UhXKNknY; z06VAi6N-xWFH7E+RAUmOV#0QtKKm0zv$81rSYL2Bl9^yan!e{t0>yWS?q&pKxYsB2 zj48q0XB-`O;j{~=k|TM$j9BA&FY5;uyKgML!F3wGMxGBbUP`mp&3^Ivf7|7!?~9B@-$L`Swii zFRENIYXH>F8`HS7wD=yf?&dquH zVz+1_?cBx`+&A*(+^3LiDOV?*!JUmsvLBYJs#<57j6XGtQr1It;5FyDd~}Ec<{6?dX-e;RG~C^P9duT0H@eHrh0-*jn?3 zB(4E%1^Nw>yC%^Nyr;VfN8LJ4pa0okm|DThyRo1<2l|&ayia6RQuZCP?S`gc?EzJg zMycvEd<_%>EsLzmPN~?tnm%h*s#JKj@2x>l?FAT7y_8N%3nC5=g*8X#zoCK!5)TSV zy%a_jMS)hSO$y7|5Sj1;ax2h)$$~_n1%5o`xqiNV%cPg~FNO~v+fn39O%~J&c*526 z^A|frkCm#xD%~kRBp@|?UQubjPuLnylT!v4p6l-$wSVbSe)zUeH;7c%=;g2drOf;y zQs2#15RCnb?6KgnT$afsEm$w^v*M4?VhgUd0@6t}Io0+fs&OGKmG}*(_O=QEJzO>1Z#ETv3N0gZrwrPU~2xj2~h$l3<59%Nq)~}4Lcsa z6kiL3T}^fGt{L55$WBBB=3GmJDno)^J#><;e_NY##ULD7hA?eQR+=|CZuwqJe|Mwy z5i&5hF2|p=;MmUWQlLlP&%(doz`9KBRBn5ZCvVCPR#< zEFV#T-8A-Xt82Ltnymyn=5|Y*xGqX&6I!zcAZ5uaXKf_V8{pELJjXo=fVDeA+(O6&9G?HjZj!R*vTW7`q0!D^Fs*fwr0El@Q zV^;-2_P53jV$+8~AAWci4y`~FLy%V?KCtZpV~QuKYHnOyKE{29c2cML;_EeN&w zW>O(N2kXRC7C@`u{}Lr629!`UI1q{|&lNj>So15;c)KKx5HSZyw@3p?s-DZh5v#D% z+(}`97zllW%0@>2RdXP}OTOu#m7TUt9OAiIA&(sS5YNMIA|!EPr^mw%7h``#G_p0s z23vniuv^%%3d2nan0op9+#1?MBJoOQV~-EmSI;9$V3+~G2ScoJFDhuyED-{^*B`ANEMB7Z*cf9u|u?}AA69?!|At%vT0S` zM^p_+nxZ^k*(?pr7kV}JQO=V*it?$vs~N+91vp!>$}JneG2(fRvRwtg^};vV(SBf1 zRANxnIz4~$u)1XO3oCN;mi?954z?KKI4iQz_|+&kP#?8Qkisw(Lti<47EI#wPme8h zMOJn|(s3TrrFpTw9h^3n1QKeO{9)kV@3`Xn4k^#q%h9=^*=7lkTOx6R z6K7^gFBh*sp#_gj``#L8_Bq-NP6Jc{`7AnVm5kh^A$r5zu4J5r=#GBU`bvT-L36rq zi)2zbXIA-l7hGmC5{-*hCbf-eSms#_K|_f^NM@VtNj^%D8$|uW@l$~v{&_l?T{+cV z`^_%YwKS31`bG-kdoa$k4SO?%tC=31DD+JdAWBf0u-loaoS|9H*v&zNoJ;**=*sXz ziK4`bL+m)zb8TeNAIpqaSU;K}O2x0^Fxj?sOBg$e+SN7M6dzEXTkC@DHr}$sgxN56 z6Y5U_hBG5ft}P|KL3R!5?(0tKbH>rOXU_}xS%Oc3Hl-q+IXf@CyQ>0|WH%a$>$4hg z7V2`=zc+ZEbZu?~JWiC=UeOkKHWP9{s^M(BJY(iPA#7(ASne9{x#HTgRMyY%_J`%A zx4DIY<|wW*eVV_bVlM`2^@eBHlF;Fr^2$aXyk*RulqZa*fl!7j7G!T#{=df5`984f5`gmA{-sI26>HKhh z>V(?ve@@1WEPctL@q#*(R1qHk4@qYkRaLh}VY*XNX;4y-?hXY61nF*2IK-h*q(Mp$ z326{PrPD)89_g0uM!G}luKnF#4#&`e?pklm`OMa|ul_eD(IwE{uUek0h?09&Dsqov zriQAcC+Sg%tN~6uT=SJuE9u5AIoav{!kx@z9Qam>nG#jP;uPW9IX|SNud|%F`g#w< za3dokT7b+GCf5(Xk+1{@KOW<}p`TG<%n+Y(!0>U*3p~~=K&|Il#JM)UPP;-w_5_u! z#9dsi$k=ssC3Jo;Z8%ChXMPez-W`HRanW$JpmHt5MkwvL?b6k-BX;abdqlsnZ}EO~`up-Zo0t_Fe+G<$ zN4F*)?uh8|V_PbQ4+a^7R{qNvUW=MC4EDa%(2J+XtQiqHl#J@Jbc&vrP09^qEpcQ7GN|4&XT9{36?%b|@=-r!p-sW1x zY^Puo{Rmczu#o>khHaV8GlL7(ejDSMVHjtf4EMGA6-07SGii;+jaw(;<_ zKx4)=`)0djo)S9%V9OduS!JN0c_1X89%n3$u#d)GGLNFftoICw^bO)Mt+iPCFUhkK z2QQ$(EB{CO;WKI2yh&GmGp*RNlpIdH^X9bdvm)RIp@w#DVViy456MkDfg0hh(w1fV zXRryIS|&vh#SB8}xQ+suxcA(RED2gO`T#&V5;Ay-+Yr)APOduw8M`&6!DDK%?c%E7 zX=lNYjd+xOWQtFZ_C1}eI0oWN{O}D9(|ix@+=mI!(`K=CBCaA*?&Kx!89mSZ4`erIaOl2Fiv zXb3Lxfw}#UyJ(vsfi_IHQCjn_{PB-*FL{j2Yh(~VUEmgaMho8`KZds*VrEYNkRKF` z`^ao5xUPve%7F^g`ZtQq_2Nzgh$LiQnF4g#l101aK&M|E2rG$uMlr{Zh!Kn7PJm*2 zYn1`mbb8cETy9JwX!ti%{WZ0zh0uV}pHY(b)d&UxIevTNdvLzsTW%i*WEOMj$Qvg}Bxj zUVgL!#?DA!`{?fw_2HIQJPRjl7IV*-EY#ru%5;SFp%+$=sJjtuTz~_4$%-NaEsTj; zuT{}ygcFa{?IEwQ zeLDYHpgotHL&Chm0rW+5#eeKoGYvY$@3;lvEFU{6gcb90j5dGIJEGAe~Edzp)mQ(Nyx`=uKb@ER(9wmxL3-&J%d4( z;4=byb9clEq>9tfF@q@P)O=Q+KBBVi=c5RV49q$lHj2K?dgj3_3t)Tz4JETXp&2QofJf#a`A)W4XqT#$tcKV>~|h zl)c`lmLwwTD2p+)izG$XVi zim|c;yCdzr_q`^3z-!UwQtqUAy7hke$FOv@6IW{z$$i|F_x!d!i-2=kS@+Qba3G=D zz!(jH^vQuk;3wMSsM5`?D*zNQ(4zPn7=cDPSN_F zJYW)Bv1t1OkUdbSENr`_1(M**4Rl`2hz-cx`DFkpdoF$f`t(687 zNb%~QYf$m^hqisQ@3nqJi-BUdF4N0Spa{2|x#F8>l@lodp<@sWQ4U1kI+?(`AAQc4 z9BjQ<0m|fafB9)k`ZXx?4I&%*9@!Yz4$$JyP+~!71(^RDkXu&Em*l+~FHYt54*)|o zCnC0nw#)M%^u2ibqFkjWy6?=z%tOnMa#p-$cWHu#Fe1Yx*PDP@>eUVFkR`QdQ}fts z+<5J-&x5HC5;;~f&hIZc&lRTVfzPKhv0OJNOxXJ`u#ZkTz_A#1um>nZid%9vw?L9W z<-2|n=h+A_^`iwsUU86336|giJ~DEvUMVT_HN=34W|;mGW@|OE7+JzTF4{;oDg40l ziuH?f-vf&~tHlexeO!4QbJBZQyfRP_!<8NfZw%K+n&KJrucL6Qp2kq)@HryUI9>+`~}Ecva&wTA~6`Y6k%?| z4PP5#2U{GsRU8OtVevUu)E!N?PzM*N68I0_u@f;!`&k8CZ@}Na!Ntc{0N|IB8C^*H zGNTJ3HZ{m&4!kXCJSm%xy{#HrIu?&7jmiNA{}-2um{d|$=j}yJj7Nk2(&(Egef_UWF%qlXn6`6oa9L&VUcls5mg zi}^}cmBlc{&BiGWd~-Ja>~3wWA#P$DD69jX#jc+iw(H6@g<$?^{Jp4yb@%Q09uhDy zUYv;7+hTIbRk|;@##bCzYu5`76_7#8PfTr1&Q9w%l&fSEiB5Dan*RHwQ6XP`R*0Ch zFb1e#>dV_u$>Y`cR7W)de_(K{S8%*zpp<*DJKY%%FX<}kxd~GxHRAZy^dH$g5?@kVDC1@`|f3IyIpjn9U|ZTmmBnG~{6vIKmc2GOGF{r1Z=$jLIY_Ge)!S6$rY2-*vD!eZPQPB^;6$h5E}CnAD7xgwknN`N zc%Vv&4nq7cc_Dp57ThJQKZ02;J+O>JElIdRSr2DyQ=w9t<#tsPSCDytXWMYoTYz1N z&}mIUDB>~(_}d3UK7oj8(!?3|%)-g<*U;`}Y}upy0iP8H7@0nqadw56X}}f+%NMKL z%#DV*u`ssWtNVO;9E3|2Y+~7&a|S4B6M~Fys!|P;J)K3@>d~3Pf>xe2u>mg+Hw@b- zPaG@`AtK;Xp{$&H>E~B_XShpb21?^KWn|`<`JSb~n}{F;_8o3n)yg|#w5_os`+r#s75#uV|Ke)Dx!-nd{< zHzFpen*gsM1w~AiW()(i@N16s>4hYCr1C8y#F#F@iu5WUq?iLLaJ{#;QwnrbchsgT zWoHnj?}p-8T0T8$K_hzr@Zdx<$=UDDrVyjpneJ1SegOM}2)hK19Vb%{kdMJ)TBZT` zQc!9;aCn4$c&_-*V!+~@{#*!ug1jC#-X8&j`_@fW&CSaGcU9QfN3~eFMA>r+&6>o4 z_01CP67dS8P7}%l#FOW;ulz^rP7~dpf2XCo*a?Gjgvm$i*s6klxmAE7PaIR4mJ@pjTWZV6O2ZJ^$mjyLVLSD}Io7}b z;)U;#9?{J?=2o1xEM?APDcs(Z{i3=K;Ged@G?Fp-)s7Y`b_aa0(Y&sA6#?k9|Lhe& ztlRLw^pVI9GIto1JUU9LB=_YY%E*jc)DP1;l2~z~g@*gO} z6|Zp91KCD+?-hjsK-zCR{{(yl|CLM3cL$7qTYTakE@Wlh?#l;6?hL1=y=w+}JOEHO@P^~0~F&Mlx07@BS zC)FwI>r^gDU6x;1;3tndN8;5d&ZT6}`aG?e%Y8A*-6Sm>fXStC94eNUunBd1?vp8ql`{s8?*bMi%^z!P-0|!*krwYX$@Yyl$qd zBf~S>7{neg!M93xhfd{a4o#F-^R%1egs&zBBFi2=rewASwo&VeQq%t^Lu)qd#mm1G zbPj2#KWl#UG!b_D%Y3H%@JNYAam{ZvIGTib!TtuA_zP?B6( z!yMr4F$&1>kz<sUva1W@b(OaYW-jXRv2d9@>Rr03ysnpL>ARSJ(_zUka-0uAc6? zd|oyf1FQhM{BHm$`vB#hfM1c24H`(=zM%Drh8^S#c>=Un5BVaE09}fd@v=U+bYh>k z#)5>MMD>q^fk_m@6dp$_W;F+Wvu^&KBDS#^>{BeVV(UHX9hSL00EnX}sx+!9?z^(U zTW*s7imXElmYRKFtMW6b4@AAp(ZwdtkO3M%@RrS_I!%oh6mVdyJKPDiu>Ed%3&u`aO_z|M^Oi2< zeHtdkTqs_X)vGKTlz_xgR8dK<%(lY7w@N3r58B$P6wi%l}fH&w@#eBvZ@0Z%YNbh9c?wcx0(OUM=Lcc|@!UXwHS0 zn%}_7C0szXz_-FJLPH1GYP>Tvd zIJ10~aT|f5_?{{5cJ1?XQ!{UlU?n@UcfrQr%~sB%hqy%;g713z{*|IcC89Hnpy1Q+5F?6 zw)OYFY5B{FHfJ59#@w&MNSOJK*1abp@tT4Tg!k%Ox0M4$-CM=IhD`Kxn+WAG-bbw? z#)`qZwwlbSKXd!joqWWfD$i}b^vVB9J^qKh7dIdx!Uk!emROFKsiDv8txy3T4QF^A z%rFOTU9kNHgqd`wowKpUl*T@cjljSQ!2Q zNC$&uKdYGorYx{cy@z1WE9$x!xN8e}v@F;AQh(@|WkJtb!Anf= z!4}D^xosLa-0E_+I4f!SmK{Vw2B&br_|rHbb$&p^hA5n*VSm{! zW`@huQP5@2^VFk*%d(X13vw1wYrqJc-KWG8m4bR+3e2^j@ACM&Oi2a=w-oEetK{-| zv7pwr0@+O$Ks*5|zJN?>Ac3gh>o&B{467QHo?(5qIl|@k&V(-gxR$*8!x#>rd_hx1`_5JK_q_NiDcbK;dN zPA?j%6zR$KdMRl(zHwjWo+H5FOYkn88eZdcqjSAZIfpSRI#@pLYh(EhE_Fk4d!afR z_)(wOz>s11>$9(5av|JG8o)aVY&difpH@JuG!!J;%)AZjDk-;W)Z@-E5kusj2xmoo z43_=S+uyg(NA@#q-?~304zBZ8Say`HkLo(h3BwV6?QM#ol~5#DQx$|pn;%lPu$NDR zaZNcI(ETcAT^mI(l2T&*5w78@mpL+1(c&-a1%EOmX*EUxgu(lbSeUdG9z&K|FotOZ zEgIwdA`>Dysqqqb{65&%7DryPLFgJe2W`7K&7kQPy(B?{E$k`kzb6L4UGf=J$}x&%4o?qIzXD%KpIuEAh`% z;NZkXjBogk=b#P^XPBLCH#mWx6{?(;Gt1eV$Poe}e!hZn=lZwVVQGZR0zGgmw*fCz z9Ox(*fZ?5At}$|1sWkX(m6Y_Za9PaotE%)^IGWaW(44^RM8#g~H1@tJN+QfsI@C_s z0B9nrlL6}>;-uXx52+q%SWe*j_}*kQ`J~wWNj`ef#?iN*Nbcj~TZ!pz&8725$$pWa zCt^~JvOM+q<#%~2Am5KsZP6MmtcOcXTtj5wup2L784W@)Qt)S*Xnts-z8d|FYYrfl z5wlpyti(PdbOJ5+{h65U9icgcso6{{%tB$)Yjr#usZ%N7J zqtWwc&>p2X?635#N^67(%I9IP@4PudnXd6(hG0Yg|4Kn9g?GK6cQD+xL(Z>VXF(ZA zt-v^H<>ZyR3i6YpO1{0oULU|b+dUTkmOSSknB;;^;~(n8Vo*wdYeJ!4#OiyGIB6sVdGh-ACJm6u?#=%+_d+X#F^#nwaRi=R&j>!LQ$9Z06{PpcDd zyj*pCq!?ZWUwI15L#V^U!}mjlzHVn7~ct}0d|b zvUQ_D*IW2dNuQQipAEI)91vj*WbS?e7x2f+{HPLY`$0Ene0=<>$QbHm(rA;;AoP|f zik3eGEnV>tgCRDt4UZzKkWvP<_zDjP*E6~G13ema+!0{s3ZuI3yY1F7PkzeQ$nz&suT9`c3cVOlNXFIJ1=# zY`s<+DW)w)>*`9#d*cAS+BDC;=seF^Lbto(%<+)O87Psx&27;l>l&ydGr0MMTbLLR z=JLFk(PFK-r@etzgh77m^<+(ilS25SHq4sb>k*7{OZ=wH%@=3}zh-AZLzp~ecQlnF zhc4(mOASmDQhaGbj^SRr(@S9P2u*YOd=4x^^XfZ)cB%X++n=sU=Gt)jZ+^rF;0XQc z=(fG$@S|KMP!s}i1_mBEy9It9dPvwhtp~zt2M(4^RzNj}@T}`1I6`4MtS{(LM2+7L zd>7q6@xJi`I5^wR+KxBS$yh}XJ@Xz&Lf8Ec8tDS|xdqrsY>s{>2H81Muo(m)2;Hnv ziQ&-xwg^B8E!J^T;UH4jr;J1_Q$0O!tO%9zRMT~=lnxxrH3CX0IUFz8UHkeDVi~~1 z3G+`qYQ+2=h90>M-1zvjNl}?KcA&v=qvYir01z{?bUT(lv$O@^>2${zFW6AHO{QR6 zE@uV%={Q7MT2T96YTDB1+ma>(eOf;06zLIy?;Dcrg=7$@imp2r5de5X2ELyxS^ocC z3p)J4@m>9#T?EoFb+@r{TJOJ?#)58qNN&-ndROn+x<{bRApp(&>G=IS7krzbUWQwl z%20OIXya+KXCBwGo)&`Sbue0y(@aM1Bf2Nrb3dPgzu$!kob13H#tAjU!a309MPt+K0HrY;I0wq7*ZkN7pu$FbV8kms6x9p+fISQ<286sfoipdvXl;wJ*`1Se6*ol}(g+?_!QM zDuat8nhJV4y;74%a9+>o1sAh6uYP*&`uo29;!f2dVo$G_{fSfV2{06ogEPCJ<9B`D zL7{F-Yl!Sx4-oec#7d3Mx9$6-&eBBtV!}7?=~>bU=HCcQ!n$z{ce#q$@*Y3 zx}Z+bSx

NVA11Q|#y|e1Bf5c-uu9pK) zGKjI9t|?m4#eWjV74n!l8gw}(b%Lyd%dr5ys1_(lAzIrLS)wY4rdk8HDn=ofVYg-u z?NSBCSP=#H%-n{taXx2S-eLAKYgiDc;sXu`XKz^VbJXLrU7%|~ksw$0c%(Q&E8*3L zS1f&fpa7wu$9K07GtTWG*CHyOb3 zp|kx37w13g8r*J=P5izbV_k_?sXq(BYI*(LS3>&(ZfNWt{N2QTMm|kWBW(7WIX}~` zx3OT47RoFXbfB>=t~=vrU1}ch7~~X?^5aB*!a*}a@Y6)!>;qv8hx9l~y0Wu9%-es_ zR^DKf7m*MV8~You$(+)Up{5CO0n+j~AmxcQ1%e`vYyW7i_eC`-sTkFIOUlD<=ugVB zVWZRCv!(7fB~LRV+vAi_`MPHr*$Ltzg$I)y^ug0pFl8-26>y8{e~D@5k32PXc@LH* z2<|%|p@gh;ODl0Q|1DXdU1|_mQ!RmPIXS?ebIMY*p(XdF6~|Caea(ZJ(|ykfgM9+I zMsgIEO6824R+4MZZyO=?}gc($k#t43YWOn(>|D+CEVDr7l&+>Cu4!mry9 z8%FCLdONLV#i*L!s&F&#pyCyWAzp->9gyZ1=3?ke=l;f<0%}ytxD41keR>gy7f(Qr zCFrsj+Gnzv;}eKxkCz!n!d3;uWDsi^tStDki0Na1+=U3pfjU6M@Bv6;UD^UB8o+ld zzS6Y{V`jFmaLIPBnQmSX!yGT$r~KefTk$M#%>_lE6kV@r`#)En`>_vgitkr3E&l^y zsKQYpyN4E;NpD$`7_nX$Ov!1t=}ls4X9uF?&iU%qjvVjH|LdO_oqkD9X#c7;7vdL7u&*_n=~ zZ=dD+?e$+dhCE=oWEJHHg$l%_W0mCxK=V;b#e!yD{4CTord2p(8wWk-L5-E7 z^f^?r7^L4={bcdBb<YpR z?6uS7awY?FZZr<_Y5(1zXVB+463}Hd2o{XhXIem=(T?*4lF5PNkreDlE5*i}k8ak8 zc)|RgWf^k7LD`v$07t{C(GU9AOaayR!G&P9pK{EB(*WjJ{lELpfu_GVqAoE?yYYZp zNx!&dxHQu=r9coyLp?EF)0AOu!kkNWbm%#tGZ(es@_@*;4; zAL{NN_qhLv>iww|pL%EagaOZSGRmeuap^~woE%>1q8`+Yv*J>T`y%mOH8Mw*}hII=FwSv>8A3C!tJtf3I zvqpJR6Z*T5h>Qu*Nc_0Lm|n)!iyq@{NZn=9-JnFz_#vkdO7uD3K<8@yd=oN>psEfD zy+$KL@MzsSx#Bqn{|i)5yLNz*$C)Z|*OJ<5tpuR2ZAx3b&J z{_X?Zhnw0aR7bf1f9i7O+U z9zJj(H4O@6zcSGx&*veXz4yw$7{-S3g|6c{#bBZnHbwp?O^LRh%Hsj_aH6013{kn) z=`Ag?(A<2g()JtjvUSL1s_Cr=sPpak|0MlWv10DnK=<0K#$|d>XBSqO6h5^&rM}p^6N9`98sdK+ zj`ptrVv8W0_&3`;;D?amYjF-P1faV#T88Q7>yCWMl=0*|xZHUE%KgC~+6cNwu`RFC zT^H9@Xb*UgMgzI`QH!$3XKs`%&R-mLQjzVg*&8(Y-yFAWsR*YA@3v9oW4)Rg|2H~o z@NQv7!801VUoR7CZN*LMJqik_k;%cgx3@TWc=9?riGR&CNE2%M+QvwSEY5I@v9my$ z<2MJ&nGnEiY1NsSMmW(x`=)o)k`mT6qpbi{eKB@c6AWxAse0LkxMr zu6h$nfr6wHYPE!4s(BLOm2vi$G6ag*qaO31j);g!5C=P5--==y_*FSA>5pFYFwCtz z<@#}iTRA2~^#WyFnsC&Ox157_)AmwR1FKO=a54|Pay}1Fqzqoye60mxi!)II#t0ot zLQSnLMliuhRhji!jemB~!&R1|w+2`~L9aO=KL(VOkg)%o(+|}D16+wc2Xo#naWdjB zdMqY*7RU61$k)Xa!)vlj#0M!uO1Jc2Go6aKafZ=cmHT01M>ic=3(eF99+{Fn<&pG~ zlrl61>{l#a<^B`BWAkz6BmGa^zz24|z*VCVqZ?wfl3q zFSl#;2ye~oK52Fc+4WIlU|_U#c6QoS5L5G_fbk38eb)S>UE5>{bnZS;OS>z;dvXOr z_xpzwGo(l72j_L$Hxtil%PTc@{CFyh&7U9wBv%DIi1b2p`;^q4iMJ?QutC4a`R<_=)yGyo@iHgSE zVma&-Iln%zz^HtHz2kmBP#fp1AIK?Kg3etz61@(ol^6k6uxHX&am|+wu~%r-jiYa= zg(H|PaOWQN5^wAZ9=&?zD& zrUTTRnNoM*-Aihc3*GmMRKvc!MJ6u-oEX<0_q}5YuHGp)J4hqVJs&3y4E2>YGju3P zlfOP}VQmTA2s)!m|E@fP(btSNc`8-@!EkS4+Eam5-J_m<^B99<6$P!8@ld`}oDJv^BBZC$kBe#y$n8iJ9Jvu;{q@<`AiknBy#>$EX zcwHUI+{S8{#h-GtJ68ON!FQGkQjQ1n?legQA`G-sZrtd>$n3d#FyVXps2o|0z{H2S zoc^|2(XR)}J|0Fbf30<*ZMKVY*t0KUIthPj0>9?!u>HIFq0zy`{V|A)pUn=2dGGcU zs%I8OdZn6=Sb>YX>btM3rTFM{TzUX!i@IaPxc=D<=j7_;-I9=9QV|&P|Iuj$Rse2! z1qD)i`dEN`SOGiFjz^zOtj_e;{1+WZM@H0oKZ4mRHE217e}T9IKn}&i%s>4lij&nj z5db>*^-~~)j4f#U304Rz5*$hIrE%Yp=`90~;U{TT_=8iFG`dIHwI(wq+0%Lqx+1yqQaNUZP($>z;M}cQB!H-`kLWM2!C-gk`n~p z6Nb3Z?O}EVD!r+v(kngree!qS3P!PX(2mdRYC$@gR56Rsvn=VevG?A4WZ^*d0$*t! zl|5$jecKPYWzp)7=Bu}|WM#cTX%nB35d(6-Q??8K@X&G@^v7B&f7i%Vz0u$h!fs|G z&$|T(b_m!tWiX{#fEMO^plX6smPm2X6B!~hq2kxBNiIbjwlt^Nk=(Ll* z8zg$rhafL1^mhzNLEo2G2;TJr=D>sOnuh@hlhuc)#tYcVb4(1%$$lj9zt?5`(=PP; zgXGFk)Z<9fSvq`!UOuSy^pygOj~HnkDt3C*gac83I*@^6_fDn2bdv>WGs-r{OVU%D z$p-@+3;`vA@#>qJ5JK!Gfg7 ztf)cKQQzG_%}hNL_)FRt_vlR}I(DZ#s~WGbRckQjH&3w!2c59IyoW2_`BFpt(Qksm z$Tk(@Q>7JNHQQfecu5yU`=Y4q%%`&N0XkDZUA%DXK!sBQ;vD&%bU&+0xGu^b-%p|{ zwRDBW$4xym1*_m&t}!XdA?!m6lvZd}?~5+`Eo$v}v9(l3OHDRFVC zO_8guSf+{mDzTF{(~#+P>|(N)`J=^mNYQCv=v1$vMhYyRXCRqlMVE2n%*t|y-rV|G zoWV-jP+6W%dbOY4k8MAZ*?jwF&Bys&o2@pSK6f|Ahp1W%a)4Sp4$0_9Vow(~-<;8i zd+kJk#f`*U3aZDC&tm!NbzQy_?3uY9Ph>=Ny#RJ91SGj76 z&mDp>vTwT%BZF)7`<0?h+v4l}O}EGzS!pFCeZS;0_$6w0+mbz{03r)iw4+M(#vM)sOD&mpVG^h!^aXk6%U2nwVDvR3b&y z6pQo+T4)-f;@8V zM*v>47&XceQc5not@Bb^$5mI{Q^gheumg0mxK(!j{QhKRk2aX241n`C7HVSHGF;J0 zYnP)GSj`hL_o;pxrOXb9;yIlyE?Ojfl>0)3#8e|AQdrp2JPHM|Mm zjPknP``dEOwRR)s)YEQ+5@H^I-rW(|bFQl2nq>b{A48wuONkNEpt$65=a*F&oJ(N~ z&J%}MDoIRX;>%O55B`&>KNbsNlB%xAKHR93SJ$Ij%gd{+cLXz5t|U+dQKl3H3kys8 zbt_9p@Hf{e3O$qZB&Mo%!;#NZ4+%_Y`?Oz3*>=374gFhPzJ*~aERIgUk?EcFnj6?< ztoo8s?s#~Nh6-|47CDL4o*gW`XApk61@^vI)d!?-%Lf0Yhk`(m-fDI zsUqX$gc_QuX;Iz!n|S^G=%*3=6d~oZ(rj2MnN&%^e)4WClr3);2rz6)gIu&&qk@f? zWzJr$`aJteGQWPo6rd#}_l3S$GXq1A#Q19Hl@%*x2#pDeYx~E19kd5(MIA*}YkaY5 zZ%jGyWke(p6x{|JySg|s_t9_M&Re*ANgMeQX<~vCHMmE?5@=c>nyWOAt(Tj*qTm`! z%lA8<416dRQ!g;}F=bD-IOQ8S-NSLMUG)C=%l}RNlks9)a{E!^U*E$vXGE~u!`HrO zvEa?$r8d*>H}W*Hf^tz{LWxS{HYgi&V-`$n>*sZ9t&-Fb1U8<=SdUUX_lO{!htVei zSGgtf=6$uHi{`s488Vn9rm0>7=WD-jFCg(~8l#YmcT|yyD^U(h--5nYK#(wb7!#P6 z@&D2?)GTi~{%bsj{qkjC7`_vNd*a6nDGI9@3GAZuckVFVqPe54=N){>x+T-*uh4~) zLb*BXTn*r@NdzZiS3ya{k=DnNq63GrKN{mu2hsXtDW3`VBBV^bK2v*4$%al+WtM*0 zec*d1H?Xfs^b}J+D-t|N_~1ce-J>)~!oW0`b=b@W!k@HI{$o_YW*C1K(UUNL{K`U18dF$E%+V3(3YTa^U0rO}Pr}^I0m4_Eyhc zyEAe*!FOX<1gzQARxR#HkSRcGBDDC1qoagKgP_cCMzb zufbe{d8Ig|I4&ypV`KLFgJ-QAx*GKYicAX^dyX|*lQwd}pHRZow_VW+=DJmN-XkwM zIIo&hQB#Y-wW0Nzv>YIXwvxP796TDM@QA2;ESHSz;Z~G*boa32On&2I>H~;kCT`z> zH)d+`=o+SDN+J@FMI$dkj*t~}k+zP0T zrUxYM1@oUzOW6S`rxT!=`#lRBbG^N*z+Mv_7xz@oRmnJAj>Mmeqt>dIc=`@m*3j!x z&@-vh%pxv69Z?c>^+@d&lKrV)sj@~Uk`q`W#6^C6hsyn!`>;#7nfK8PJaFsUT66^8~pyB}U&O3X^!@?6n_#Ic?>tARO>x9*l&Tld<%&!6mCQxs(KCRTh zGNEjwzs@?}GnMc!JThm8*@wwZH{0rx>NsG1j_blaT~>()7p{UrNx%B-935?6chKY{ zqURyLdu3bR`_3pw%%~{p)f|KH=*u3Nq4|8??>soA2FNHlMpVURN=6238CC5dr#$kE zW;eX>B3*5uazu5b6vLzU9i;xLVp@-012cjzzsr zmFM$fvs~k5O&`@SU_XJgqevt!_9Q848n#l*;Av#Rl;~`S5!@h6u?4iaZnKtGP6$dh z6GpNJMeR?~)rqm&8ou!Fwibn{yP!B`A)JWctx9<`Db3r9moGhPaPFE^iIYH{i}w1; zAs3~d%VmT2HM;&`Ei;!s|L_OLCmZv0(^ zf6Txqq1md7>yx-xDC%eJfhKxD;r(gTWW<79g+SMA)8?^Jb(hVOwujR9AdwzW8Gm^4 z>!Qu3GMiJNvG*@7E-v3#>4o&9LS`(g?lrjq-`5Bt{7ekH6!-Y9q)0O@FlZGeD;G^% zj2a*;GJyw|Mn}Dx7h4ussv4xOt&Sx`f_>VQVzgVrXSAIg(_+7h0Y5Y(Ga1_x#MyLe zErvc&JQY;Q*Z#w3eolMsX?6D4gmaU}v_2)$5KdOtExT-WHMyhQ61STtT1MUz;493# zWlJYH!1kQ-$!}8Y&a_L7LxRby564pIK|;E|F~`SgLM?r%!jrAOshLZYQik7!@OP!m zXv{v>mj~)5xWePCg8x0NwK_Wk5fGM?=G1CAGK+Y^Lln!*C?QOJd!>E3_umoXeVmHj zU+ifB4h)gq>&>sc_)QNN-pk$dpC^#jJV9bg4pd#I=Gy5EF7H3tHCx-aBMCnjT8Q3u z$st5iZh$83F{Beb_D;qMsW_Q1iS;s+j$vzq4@Q^xtU0lGMLi^#s$KH~4nU)o&L3E=Y?=EA1p1ve>x7lA> zZ|9jTe%cK%R968#TYr<0^O1X?dKtQkm4f0~O$njtS%~kQTij|sS`q!v0EN-7LF2`e zrnl!4LyP|~JD)|pVGq}FV2IT&?rXu@J|PoY$~jw-R2b&n5WsStYn=U&i0=Plat~;+ zy!E}-Br`rOhxr~~1CDpjq;Z(&?S-VEQyNz%o-J+37OhD@PCrsC>-a4feEg*0i1->- zTRLEa#j5U8ss09Sm*_aRD?XPb_`?#zA{OlK|Vm* zU~0)kS^ZX_DrUi@i@;Q9@W7|~=vvoDq~&h+i&pi?;hufdzn?#u%G+PH*$6~8%?2ZJ zuBvFB;0xn{pq>_LrK5W@X1pTZtUMht7{N2*Y=1~F|F*2xAXVN{C>!903Sx3s4`>>wO_&OOZ6$1O)fEf~V zV!ZBm$0|k~i3xluJE$0V_#Y?jj`r;w!dEgvD1vs;rdZyc3#JR*6UtG&=`4zZj(J`Ax2&ACg(^1Xw(Dsu z#r!s%s0G^{2_HVfwPTpGEd;V_IZf`jQcy*?eOk?MJ#<@ZTrSghh?yoa0Zy|H9beJ} zj6VUI5?_%Fk2z~O>E_Fdf9RNF=m*?helHMJipSUu5Wn_YU9TKXJ?+1YG2BWjQ0&AM z!6JzT6AhMH&a9<=Oc)l+WPM194Ev5aHDWo3G0N~S$VMcB?5_zFgBbT%BX~*L0*UCB zK3Q1da-F+b|7>3lK!3;2nq%TMW&2x(gOKNCe7*D4(V#&!`vc>e@ZLe##-|`Ag|nzK zr11F|b-8s;2u(n~ci;sNf9+BHpqTxJeb=FyuL3NZC%d4muQw^>y|1ZHwjSqX*BRTj z$fdtYa9IFT^XPkS;k63oJwm*?B>sYtQ_b06M@jwWIrpLc$PKN81+szm?Qtc`sY>dK zx~L(96Mwzu^QR!O41Su$9_z2K8lK$?1|BY-ia@XA8F%Yfd;7v*IGIj#7S)>mM0=Bg zcPgWToN8yXbui-W_RBo_AeaHfMh*^n8vwtbq2t>A>VlSDlJalQlv%kYRwP3sYm7Ch z(ygDA>S;83(K<%8yWg+QeU+)gp(ZVMk)hBUd+_&f_2JYQ#Qi>Zabc!2Q(0<@zkWIp&GK#g6|_4s8Xho7U3-{mhVw@D--8s2GK!yQTZC zdrRW1o+vR4TcYdt53bsTlv!8G@4W)D%hPEeoJQf8I&E21b%KQalWBvCDOc+@^iw7% zN`a$K$846l7GMFgaoKqrf}aWrJWg>kxXxbo$8!G#z|{6H)^b{++BiOs%3=p`-(1Mm zR!Tz|74GzzLWRR7Js?}#rz(2@w~F~v?xmBM!yk4?b)Dp#Fa|W#q0t;}9O@Es{>4cH z%wIM+Uo^7u{A-}Rb;Jc#G3EceNZn}OmWERni!p0nU4b`q*wx~g5-TI+K1(JsQp?Jr zp%|#?3#jAjsPb$7PDtAs8&_qUmdzZrU6VecU1k{#Q;?~2Nb+-IQBas9DZL{zfs>|A zG}NklmtVDGKJb^R%P~yeHG7~-VNxN!zuKM9-#i#)P-=mDL#EgH546ySK!%ul_4DDy zt+uvy|5JvJQxV>BHy}u)Nl)Gn8hjpNxJUjKz+9SNX3IIX_x2)9;3d3HX~bST2Ty@L zmw%5U7Y^u+jc*#wE@A9u)6DKWMyb|QBkj-DXS2@HoCcCbn2EH;%`Ap7;8FIye~*Zt zonT;zA6Gm_@x<{V++3{Eeb5|dg0bXTN{uYDejVWFJD`Ol6M&dsoyK~1&-7#AJJC{s zi|2QlC9c3>a8mz=nYLz=g`e-!Ez`s}r?fN27-2IT>}$X3eOS?sbHnHD;KM_+U%)7CxA<(_Gj@%=Rw93APjb8@nb&GPw5!2 zA;yyLT?SB-R)F`nxMvvE`DD{uQx@sab^;e&|Lpa|JwUhgk$KNy*xhw&bNJmu&A#zGTs#n=0!OlvjHhDPDvJ*-O>jtjR-(%HdPJgS%?a{jt+%hoEo zWgnL5w?;e<4>{!D)HkcVF+vnq#1NYChj)I}FS|MNuH9Su7vvZ-x*uczWdi9wEcV-d z*zwrhBpdtmIjpW9F~6g#1wHIm>C)nKeWmmw&L)vC zJ`(g;Yr_qVS{Br;wwT03%nw;v)6Rnb$JALzMcGAbobHC91qA8tkVX_~5QgrQ?vO5# zK}xzqLFu7}QbM{>LAphd4k@X7=DX{zb^mjzi+Ja}=j^lh^ZXtk&24IUFqef%dPh3p za}WKOcwU=yyoo6J=y+JtIRAe||0+Y-Ex@E);Ww|APGZ3EX(4%xk8Szw-y?cSKOV*N zP>wBW;zvBFiq;W27N~+f`@8A1rDB){wm|BSI{&$KX~*XI;c!u8<@a*;NbYqv_IHVf z`o?(mx`$KcFEn<648`=E9a9|ahD|ShFF}VaM^zsZ5hlJ6brkXijC#kH-xViE-;KN> zWyzEAC5EYY-ZKDwE)G6=RuD^ATaY2Y_ulu%o3V;TGf#@fXyiSdgWg6~-G-r8GNAh1 zVu&y>ogQroF@oSD+*&zqtfLMNzV#PQ*rA01GSq3msLKPes&2ZE9fsf9;1?RL%I6Pd_i|`W#3dp6d1b^4X`o@Tk2(pZU!x zus$#Q*f7A0=?$cx6VIg4kratsIxR3_{d8$8R4ZEPMz1=hN81NdfFb76fl(JEw*>u^ zS(D}&tFR5EEb(o~&=Zn=w{3?>e@W1Z84Kbe?0C%^{O9!Om50q}taU9A9(}vHWv4X&WOH|X5klCi5# z8wM(Nzc{ub7Y4m;d-3~GEk*C)Y^|bL{v0Htpq&OzCrev zhZ{9(2&$x7tz)VYxoH^MFMNKb$kV(^#)`qKQH?DmbS-RSYH2(BX>tq{huOYrc!?Z{ zb^p<(40Vg>xOFNspOkV(6B@(knW(nVZ*9o7tr`Q>k1`G6Ix57oFhKtv0rjS7Bu9q@ zA_a*|QvXQukG5*cLnLy=L6huYHU&=Q+nsqqEhauxJUBSR~T z*R-ieLu(Zsyv?3_zgk9}-+L-egojGm=RaDkrhW;yN3kor;lltq+rBwUx}Y%9ki`O3PynY`!cLkthqMLNlHsfNUc zqM+luu-pO453SA*q)%#o_xM|rp!Nx{*5-P1<1hRL47Y9!th}rCEX?t@YoNAZu`T%& z9|`WuYe8P!%Ef@lmgDjqOWEu&sze@<_8#3dyDA`)5o#Vbnjf0mKIDK+SnTT@j@jR0 zzA9D+c_~f3?JGCAQS+`3`r+zWeq|th+~A%CVZHvmLqfZCNQmSx)0Huq_X2H!#C|`; zwb`Xz-C8wesL0=pJaBYtmA7`y`H$T31%=R{?$tFEjIebUQPHcuyqc5DmSkT*pq_4x zP-JfAO?H)!h#6>T8^|&uXc%N(iQHXu-dh3ebboe4;)xE95@V#o z-TGwVs_b1mYn%|^4${=F1^#^w+@20xYv1{cCv<2`^`Fr8-sa`v?SU!cd{#h8-L&MY z=I71M@b7)lUbV#2Jg(Q5KI$jxJM+G4i&|esiWL0rYLi+MYMe_}XeJbD%0!~5Hz z#+=@{w`crn1rYd``C@G@($DWi#p^;TaPL)*T}Wr)x>1e38ZWO-X?p_t!lQFVd5ru{twwAcDV-X zvnf5?-@pG|0gFH{P)q*#T(S|t@aCb*QrooNBxpdD`=~9S{L|Dwu9L=MlgM&D+x z9Jf?ukzO{fbpVKqRsdbJf1N4`J7gFcKAMEkTV`e6t4 zY>h5oz7dlZRh|79R}u8%AeS9$GQud-Q;bBE`pq8q&ny$4Fd(lwqvQJ<-H4bL&Phu3 zY$c&F*U;%z2LB(_*_8YSCY2nN^ONmN`0iefMB;w;HZLY!L zet_SK&I55)soEL}E1QKzXRi}#@Q(NHgSCOY~0lmdIoz(TuCr_x)T; zCs(!fpF#o8=yabsB$LJ{eFqeUVsZ`XYjNxajup`3%V=Ekw}(m!l-Yo!Tz&$AihX-Zwu$3H*Li%Qh=mAWJLdtUZ%!*rQ>z9%!)a8~Xa6lb*{ z*XR^(h4P=TVl*OuTV+7jXqJ?3#h?381hvTWTtyVk@`Uv1XhZ@krW6|J-8`ITFVvRb z+T)10y>^Y^mQAF#(%;>Tq+uLId2WgZwG#X?PIrcZR+3K+e%UnDltzNV#n+~XDv$p8 zD52hLUJ=pt`sxfc9Hi|$WL7L&erkOacQ$~Ro;T5YRLpUf-PkCSN}#AfuB`1c)ZO*u zxY}+cqipk;zDdgI#r?|tEh93DVS`+wK$>R8)O)Nmk{Xi39-wWEGpL>Z9!r?q zL5|&OOI70IxG7AJWw=umD7N1lrKam=CK0eu zt$8dacJtp>&!p|RZp4~8wDrQ6p3+y8#iEZTq8b;VG+GDN2#3$=gx3p5fY%jK+FZ!xu|-xTbIr=ojjuG zL7y^?GeS09fru#m8bhPz%l;2UuzW5sF#+UB3AeUCY`wj`t8KeAJ%A~A3`EkjD;vP^ zB~(o~l=$E=QWp;_PW$rHXn3`NcIje1@q`3A{bi@35^mXzw$VIk@MpfMQqq7E-;>WV zNc=G^PUb`<6ZRklyX}QYUETfj4?qv6D0kbu5{?HCH zxq%qxEfy^?W@L^eHUxb1c&YqRf|U+cAy_J=3BjHHgGX+OiiHA9R;chznZP`UsLDSk z;ZtYw=b>ADs!CDoQAZv{T8z1Nh-bHI+gIMJ4D{o+xi;eC?JlPSlwqjrJeaUf6H}B;I)Z3l z6ziwMVy1lm1ls%FxKqgVx-Bv;j}?q`=a#%IBndQo7BuR@5^i|)fWp3y!M^a>IaZ{} z@Gi`*ck)8@U3Y=2*64Lw!a97^OunGz6;dsJ5c`~w*X|W8p{c}?DsKQ%%Gcw!>O!T7 z`UB}@0iNciH&rbYO*$T+R<#b(M}-bbtwoc+b%dq6|8ao7gP1@PZC5m7LA6N)EFX_G zB{wN9FE6t~C|-OMZ@pZP2knliBqY!3O6ITr=lC!|nQS3)+@-Cp+o;*iw|0}}?o%Vs zB8>&d{oS`-7Sy{$CfNJj683V zQicAyRhB|;uoSALvs$aCWO)vs_dL7l^-^MaxjB52h7O5k=hrtEz%FxrDo;T24ISt3 zk-tS?5GnG@-be1gmZ;(TOr(<0Ezu(QS#ka=e#{YUo9Ge~{c$I1#v;+9q2n)|!>6^s zM1k9@m|1^TL~p(c{nZ5m!qre8j)@<>zmOjV(8IMND~dBs4>gF?(yHFVn%e195wm!TXaj1>6yK zotT{whD@=i9O&GSI~q>~g%`iv+R3`HhH5%6RWMH45|9XwLL0r}YPRn&o(wT9{)E-t z&~@w8o)9H_Q&V$?CP~QA6&Kx%o<{!iKW-YWN=gd@*1g{Y^OWqoTJ;ck@Gt*XOQtu~ znqC+2GE77(HH^Px;zLN-8Lhv2UD3@xlby5wgIpM}+_ki)5h7d1)=TM-Qq2>Pc$Phj z&21=nTmocRY(^!Q{f~Uu^z|TjiD0G7<5U=8W~SEw1{sE&Tsva?lAzB(ebExa4g1uR zlpf6avkg!eJi%BwCt-gAY~SjQ8|(6B;SYq>D-ZryFZ`AQQr~pD<}jk(c(Vhy_6&W8 zY=Q1actlRm}H_}>;pKcps~_5T>30Xpyoo87K0S7+w>umbg#E!3fr{dHo-<}l`s6FiziHKR|u zJ2IvQ4@6;Nc@WO2x7rFiPOta&T!Mi_knOUY=M)qHxjSv@XvlcOBLEf3M*Qw;ZeD(c z5BeCactrWni>8=H?1tNJ{Xn?WDvUe)@gYV{CxgWCllCtsc3eB*~;*ek7%H@?vatzzc z4S7n-#4|n%2(ADf@!|os8kM(=K+`=lBTL=DBFp`Auw+AW3cp z!~2pv?-U7=i+Qfb1GdJd*}DxFAq;YstkvZAkWENmOsCvr1!S@;q2<#$k@u1(xYr|B zn68DV+Sn}cTCA03nk#yF)i-Q)j^>I_NIX>CSGtc+=i|~<$l7`xEul3WnZ4{1_w7@x zt1qc298Uoo{P>vW9r9MF$*GP5vkVs~5BM7Pjd5eDzZ_uC=VMd+n&XPnNKE!@5w~zz z!0DU7I=`{yHsK&78ox4Y8+84FYN^P>-OFPftrY(WQcBbnO^hW2RA2uYbiv(N7&sm3 zjS(){S69F?b31bmoZ~s*P#e&7Vnj%8W<3Ex=*BuE!^cCufrd0u)>X|bWh=NoJGR^x zt(fCYo?$2^av81O{hjF23L4>J4#UAg5fG_{ua7XbmTOe?8AyJe5H_U8WFJ{C@3x+p zC%ua+`tn(;j{W4vru*VYnk(d@OpsrRQdXACJY2-Rx%&8zNp&g4Gl@+jXBD`d{LeY2 ziO4Yx%S8=bPNtebw(l7a)m(kpT2q%_-L$tbeLKSNUGxIUV>tmVN6$!?jfm}Hc@^%h ztP+lB-k4Z|I?#pq=@8=toi{+%ME~9Zp`4_oq#C8J61S zbPG32J#Ee-u=6sGN_#4xKJpxmPCU6%z@EXOAL|<3w#0>gj8Sx}ET8cOf}>^KOd(djB*)^yT+QNV8?jK4M{1aHfC5rJ?&Lo$7v1 z-F6IBxZEAb7qr?+Mg9fcfB(?d*2W`tB58JT!_b4m(9b8Zb!TQPv;+H$wf_JqXl!!B-^!!D%g zwsnzD%PHLX4Y)q6P*uAW3YZP=ftk!T3FL9 z+R*=rT9nYXj=JzO7p;o5dDiypo;uT0m-%J#J~8)){8gyAdHY>2mBe>AH65KDK9+Fs zQG45_bGs?5OL>2?T~GV3AbwUMS^+nfw!^Cyb6do*u&_YJw5w(E^#N)RJmA>6L9pQl zhL-8QT@iR&u2$_K=Q+HLmQ75yhWSI zPtLJkEjx1?#WZ^QQA#?|e8f;0^(dNnFW;OSAx{IRA+e9aht+(YAHfw0f8o++DHJux z*PZd5$r2uAPrcyLl)3Pbtsk1Mlh#InP6O@=ElA5UNG4qc9*bVUKYV?>F^~&oSPTOj zbnY#-I}i-$bf?F5b?u6;2*eG{>0phf@jiQ7c>`a6vGa->0V~9P#79=h_5+r#^!R4x zs<^1UM&TIZ8vCguk!gX{rhRY(Ug9#3eH?~%O@;O+*ES3mJk*Gnhuly^vySGzv_sU| zdw)<^aKhMo&VzyEs|UD+)b~Ib@7VgYX7M}pg{*sIXxut;pLhWy6_tZ|tsYuG=`3yg zk7@a9e3TV=CO0wofu!Y_t>wZq#J>NJHL*GP^pN!~)SvAs%_NoO2ypBrmjpA}Ri8Z+ z>ebOn!_xGCS3e!I)uK^Mn3PgpR0U5~x!hfE0Z-7xBM;11k*MOABPsd$2}mwX+O*92 z8x{e-J;R0vK({0T8q0NFo2ukD+S0M{{8))HKkrQ6F>cTMug7(@Zp{~}eP9}}&-?HO zvy?^}ymAK-Q9_l@X;@TcS%il4(3LkkV4&55kUb&9JP{6WOyFv9Dk6A*2l@#ff4>0} zJ?q3hOxJ*eGZ~0_rGRpa6j;qX&u=o3D(3rA`4R3ZDbrvx1VP!7cA!!FPrTZSC!{84 zY^b>S;Y&YUFjfAExR4HB)t3{ztJXxl;npT=;gf$Hp-{eluBaJ#Q`p=jDC9r5b<6@k zHMvs%J$#yPEpEFop~h{4tg_OhODa1;_TN3MG;^89u%#NA)1%O`1ugObK=&5#-O)qB zsMxZDfhiQ3PpV@1xq9Mb&fv7&&eAdKeI_ui4VZplT z_GN-bnE}aG%ApCg^Y}6K>ug;5R=tc*V^tz99QAJB6gz6x75fL!O>>6QYis^{iP$SZ zkQd4TIBsA7KQsBJileIy6sewof|0ckY(Qd3E{6&ZpdGv{F!u{wEt{^Yo6%;>okxNh z2re|TP!@esMblJStr*~k{B=cl$*!;Wv0tB>iSlO61IcG<@oPdJZZa*b= zQ~21-e$FqFf=+Y?vr!V;G4LHh+Jn4QI+SA9aLR#^m5`E~MWUl~?_S6C4cF^ggvWCu zhZqC!sA0OehJQ+)5f{CDk?j={9w^kOEnm%1K+qnbZF+_@8SVpjv>>DkDLFxQPQq5R zoB`|X0-kKQm{!qKu^+-f+5S0~P*k9z53g_O5F_V-t7f^rA#TP4IlR+fI(xqw&$d-Qoy6-U^h%|S}U#SinJaHBQw|0)R1Sg1TzSl5knf?G6<~c{}yUcZ;KUm~z z+h#r;Nw%h89|b&$s(pOV=epjIc4Y~kZ?zv{m1#Cv^YFpaaH1f8 z7F_HqN6oxmyb*=WPDy?O-ynd;om8kuoPo*4mwawB3!^no{<~1S#0LhCFR{C+ zH6`+&7pFeJ+eQzT{0d8^lxK}6F--{U?d_R)ct+<=-9eA!;lnkHS{)~1IXNbgM(ZEM zsfx1|Mc?7fA1u=HvA=9?!8H*%`g@L7ePecopE<{K332an>PXv+JpFR-QFU&saT2C$ z)sGx=E<%XY(Lrs;2JiFKR-Thfc8{B z;I#l}*~%Iyre!=vn}Cl1n>_1Fm0ms%4F^=ldLfMbg+~*K;l*15X5syg`RERe$~7^W z;eX^V`UM@+(Dvbsiy98Xtn>YdjU;SMxd+>`IvrehMVBbpb{7J<7?@(Js!a(w}> z)bQS`^(T9`4k@x#4fKwdsCeMSYeQ2uxYIdNs`HcziLXBD{;BX>EuU;#RVkeJ^Y0N6 z*&s2veL}`1bcatJBM>;~jwh9fvv;7&%gb57E&7soFyXrOc&oCuV;(~jD}K#HS52|h zzL@c=YTIZqFaQ3$!j1aY^S~jsTYqk|@mQUxaJFY?f-J6d3J7waxD+~O7hx3zvA(%) zN_}&T+(|ul7FD;V9D3Y$`Bq(O$cFYXZdA3Y4bZRSeLQXI`Lf;(+LNA1cemH?wF+hJ z@hPIxhaSjmvVo^@5pmlm&Xy~DrBu$$=^U<4jNTrK;`O7K53AsODxpA4rqynb8# z)p$WbCkkS1^V*0mr=)Vq_k>|CCn(eVU5SWv*8R`Vt**R-(VS+lkl%fW&do^SoT@fk z^NsGsY{vsbc$a-U^{S2#SKqfMI<_WQy4s_7AUhR@ zDwyu9t~&*O4%Pl}qon=969-$Wq_S9kk7MCyAEwDi_=$c$^cY__FaN*oM2y0am=@#j zA+gWC)67FiA2}zdU=%I`WAKqex$97Cq3l4W#j9|VPc>hum4>;H4e8gJedD8KL1tJ9 zFqTKdqL$JUDymfPAUXNo#Ic)||jy#T+XQ?bg$&rjS5Eci(1vO|)t6`oM_q|>*7-lV2ncO zgZ^Z)6r~ZmOct@cCgBk7aKPdD-;Z+u)LmXEc-22^oADXGeG{{hmR^)Jqr^|^FkU6il7z7$`TpE zvQ#TE5l3P_|Ne;RPsq=YWqwHmKXuOK(0znX$9*hh^rW0NMqO)}GbzxT<&p8j1*CDa z(-mTi-037Su&IcxgA_FWX-g9#AUuh+wY7gvbd10EgKp%gGq5w}%c#a2<7N;`nKEn& z6YSy;1d=;4Cq~A1QCqIY%kSRLWnqw`1z!aQ$+KZ{|E_0t8A%PF6tm`b7+KA*za`Nk ziRM)0O=jm~;oiAFr#t$gCB{(x-}#0LG3CU{I5KU(nr;07tHL5`Oy?uQRl8=Zg-75Y zqA=BWN|5evZh2jtAe)=xRrYqtf2+q^RYO?h%T7XX zfmrUB(8mQ}kr1!gRxJ+_xa?+90eD`d%aB=240>>g@@8{K4#RlY7j1)1Ur_Sh1&afZ>aw} z2u{d5<(fQ|WfGW|+(r$EdqF;>cthlApTp=dGt;tMR_dFMg?Ah5H)ZveeIupae*ul8 z>QyQ78T)3_#_`$O*$rXj92csS^a)#%_+ELLd`->#lzz$SGNw~g!{q`|cs3w6oH1Ow zu>p9K|Lwwj#jX)qYw~|dxXbL4bGmmHZMQH=WxtmZPQH(Ym^YNJevR=XeWJ6&Oz<|k z?peO>)@;K7wB0(s_=hGs5OX3(KCk0(cS-qpy^#>BYH18Ba0!cl!G)e3_F?qp34W`m zIX44bhYPzh8dC&|`5LW`qCRe^YeGzX2(QWLn9HiB8dl|sg={km4UQv*7+i!!!a5MN zy3)k`5(t;MND%YD%*u*6G&J!^$(H4f-p5|Dnq-L)5XjbnOBM?-2w3IheB2E}S>9vZf zZ0lu**dR0RwDZ5+({7Qi#rob$>LEz8=s`9kTfMiKW6M?17=?v}Qz|#fq|sW{`{B>k z8eF4aoi{@GhB3K~zGc7krX`UvG#?y{wSNgdj48UCuP&Btm?K5a4cUQK`}t95+z-5d zyxTs*gSOBL^}=e6P`!V!z1jaw52ov$!W2P#<^(RwE|rFYn#yaC5Qu$GM*JcL)ALZ= zN|VjF_dQNf9rWaBkEJI2~z2#BKWR*;=Gaj1`FunBPW?>Qk8XB_b4=+>gyDF4lNcp9VM-?Ptq3 zyboXfiKwRzQNG>#Z^2pWd?#gbX@7`#P_pf`EWK4!5`Y3giPYkN;sLCc~K3&ne zJopSjY~H9cLy!DcZ3&=Llq=@?6%=zY0NAwz!SemMaY;$y`n4+H3TMYW)-#$dJmy{M zXO$5E223Cpq~rEinZ%S?N+GK4eiN^hslfZzDs^)30hhYdQ##W2|l_b zw%qdXTDFMz%j?&v%m3yr#?vIML`f`ak->~>C|F?Z>UUZy@cnV}zG+J#mHWH#ZW!63 zB32EHODQW}7_I)(rcyoHf3I)$ObR#_IL{sC$v!ZV4<{e>w9XEe=$6|3#V6DjC`z9ME7{^VPFv8}%-uflk+T+@ZOUA z5%lLu!vLTmAU7$JW*iXlj0~UGfMRnSB~S~lTWDfEAhVA?k}f@8E46ApfC5m^)B5+L zLIYkrt}G;S3y9aNime%6^Z`X63r_G#A*0$Oy^&&;{S=Myv}s;HBs|iEFPwnd(u4pD zOA4E!l&B%>KJ-Ogb5W7JNl*aOsFNKlKIK!lN1@L@Y9euO-LQ2kzHto;rgH&Lzja;9q|+&b#2IZSev;bcJqFRp=MXv)gS)CRv6o zVQdvOZ*Yqt?$)BWcZ2C#1xNPH{Xu_n){Job2lb4$EmAW(%q!w3Pz#lX+hE&t>61UN zYlFs*HIJTaDf-P;-aF{bML1_ zc&3N;Jo!wI99ZgA$+GW*xH*>_5e%mHB2kn>=E7} z)(>;B7|!Q;j2UswpS8APUxlls?N@{j#X#^Liu9P4tAWccxXf+0@++G8%C@9#_asNnS#56!;ZDwG}=0KtMK+|ah~ zP`?7@0@LlZ=`O@X_*pGVdl*@}0{xGu{nu}+yMFseJg`{UKyDCG;T_nR28)j5Z?a$O zSL*qca;VBg=vyc)pb7T4LJSX4oaux~ABHs4lyWlQ&Y)%TpvvBm1l;-}K_cSID6;p9 zu+r@jgS1)0$1}07=OxX1kQg3Svp(KP?yp@>W#4?n@XB#wcR(nMInQQGqVADxvDxH8 zEIA9*Du9^2-ClI{QK3_r>MFAERoV)#6ZwGNHpn*X_ zj3kDDd-K5gXO)6QO$oOsk~h)2wGx*;PQ669Mv*!R z_vG<*a&+iYXgL6Vd;q}krT|O`FxF4t)NW_C%;y89JsDsN3qjzGU-kEUR)IhOoyzRd z3sr81t>V?-hC?k+mtQHTgxn8bI-5+eJ#mS_fe!r5&?sn#8UF2WTXHo8H64ec|CB|q z3ZBGk#sBq+(zdHn2<(zNv;+j76B!6EuhDsJcW{WoRt`>bD#MiK={WiK2Lk&oFR>3V z=~Y(B5Vh;1kJ!*GLNGyN3nG_^(bEOm0XKJ#lQ|De07{N?@tDLhsKZb`f1kM( z>?z;7j?sdnBH)SgsdWe*sYxJTy6=)?wDp}hI)YkLdL z#9_B*na7V4rFJ-wAk&yVjg83?Wqd&GLy}^E{t`vP?DF3{v`kiZe7s&|1#Qp5QrQ5c z9|{vVPT?QLduVD)lX+J^MNu-xj$C+?$|Y<6?lS*z2gdq6%5Muymv=UEFTG6J3L^O5 zY;bK=9)e53_o5<=uJf;3dFb?h)a1eW=&*Gwcq+JXnrPg-e^jV=OHit}jq=Sujt27y zVWacQj&%ah1PxbQkOpkzGf&?ri;qlkC0YgYTf5+{)3BY1)q`lDWb}ZxM6vb=*$YVN)NR)4|m!GMD z`fP5}aXjuv1@jl7gGFOAPj^T2!Pi~9LATe zJ2~n4E#H0`k-?5~mm^vO-oCWRun%M7&ay;rw(qo7b6q2-nI@JUV!o*6N0!OE@CaCj zn=Vr?SSRE2b8`BTKGl&Xm6MZ;j*iZ<#PP(fYAF;T4*_sgK+`P8Qgn(~H055-6CJT1 zXWR29q8_F;oE1y9uB51t-E52TM=yy*K^2nV^uFWg z5n)`TzC;|pm`-|zo3Lukmr8Iwi5N1M7gZbuDnO1SXgf%v4)%T%giA`0=mdts_a8sT zCaSvH0h0JFjWDp%H=#t8{?%%Mw>c1#xKrAnR!d}BSLBiXqx>}-Pez5>{&B=J(8@8$ zWRB}ot9PXCmSbSWWzImH857XJJz~s#U{lGJdPC64{tIhJ12KWDma%ZTE#NzS1{SDc z_q+#oZQk-aPN-1CKah_g=(DX2uuqIKul@>DqZ=*3h92x4bVH+@D|8Q(b+X_sTkM{B z8K{9$+nl0mU_kJ&#!y*BqdM4^tA^NR7L>_E`4CqvWKW!V6cRd$q%2w@y14X3*>-Z;Es_1;)gL(Qc{pNrqvCl(pcRW9)iAocqZkMch8dw z{FMF?S>C00F|wnY$knq`v3ob?%lQT40ujCW9(FVWcsYEDpr38&uE*Ss>L4LnvTlJB z2!zAa`wqE?{lY=td!w4pGASTc5z(U#U+P@JYog+#cpiUX!Nfbw<4^eObEvA^{Bwl_OfD?)&K18M^Rdf=gNidJTa_ zE*hBd7PdmWhX4Jy+^9U_yZI$5~eE zf$`X5JFIbP5tnoF-@QIcw}D3Xh_PTX^E8OX%Xo0 zvoY0&-oC=X#9S-Ob^8iNJ1>n3ontCrWd2+QoZo>o{e#v`D0yF1P!;pDwZ@ zycAF=H<+p_Dml_4^*7K<>{7cY^f&$QgH1LKbezHwuKS-&%nco)#u9QwJpJ^N9b1tif4;bSY`$5`#FlvQbbKN+5>*7`xgv!7w>Mp-n*pkjtxU)^BjNWe5^qy`;C76dlr_B z{AHpd`c#bUGWDV`T=iwkK2zuyW*{CAy~8od<5n#_ z!X#8}72nlmMY9#vhiFC4- z(W95NYlxkST#W-3+g0@J)vLDh`u3m;w?>fb0eb$-d@ZYJERE|EKt2Y-B%JV|i)O9g z5iFa`T$P9Q(^nCRi3Dtu#4r6BNaiV2QpgPkPhA<~iBE9gr`TO`hQVo`^6qntGV12P zD4U_4bn8m>#?v_;ZI@U0$yI|q4QtbK*Q9d0SN?DX@|k>)s0*a&T3-2;GjjLoPTsO{ zY$xe@3F3SUP#O4J$>KKY@+kqNywpq<3C#T=mAnpe zV79~H39F#`X1Q?2`FFm*`&m(kE5bc_whJA!KZQ^bb^C}5ezf)0(6aOziVLOgY2WoS zL7fU7L&DwPxqTY7>tykra$R{6PWt3z;b6D|;%-=wTR-gM`g!`FqazO_*$~|SC;)mS z7K!U^JiP>&ZpgvBlmn+IY7lz^&sG?y%p?24lTdDGhm?2mL>SpA$LN_n;|zrZ<=-!N z!7suGQO`{~J+235{!?%&{w2^cS@YZQ6)KxtGu^^2c8M==LOKd{<) ze8pv71r3z}WV`ZqFM^YTl9J(5cB{Su2a* z70W~E1@38?+z^*Mci^JXwK@|qc|MMx))sS_e7Bw@WFz+h6EvcWPJPD(V>J!Xzc8e2 z@NP;qqchvqzw~?SGVrw#dVJDUjp?t2{x18xiu(vW48oXIOUqH}+@m^>Iq%hE6N_b& zp0f#}H^t3lw}I}JuRZVvWHTerHV>fY-x%TV>=hiKk3v#S17k|`$TIM)UD%*lvTQ?C zAL4~s(HSed%3S?4btx{o0@nY0MAwpYow-+yjLS-8M}NlX8uLv*LcfM%-kMtpRzpVa ziodpKx`lykS{=(fJ%VJ(jAZMN|BQT5rrr(uoZ+A~(CD@_FZurD=%NGcI}(7vX{#tV zZ(9|1+(*UM@PYa0LLKC2V%Q?Z=^q5pv-#X9$7a(NlNH#A%e0$?^VHpS{_ zLQ$JD6hi{7E~Y6nrtBjJ1W(}3r8(W25QhZJ$tK^SrOJ^7_Hc>fp%=U0cCB(yxdp`q zDSXq9g4nbon52JkpVF8pI| zRl;xJ-iJ&-3qazFSu1PI+DnRxK$403Q!ramFS!EY=C7af5LR>O#~{26v6=j$XS;8E z&&S{p8)3xYgz>(++C*=3#w2MH+!>kkLuomD}D7yz^{t;Qrirk!@Bi~ zZ$aB||JOJt@*R4%F4hgf)Y{Vr{GCF&fenC_?Ew}V9?c_jdf@zM0TmK5Rz%9dfCyHC z#09goSR>jxJLSNA(V?Id{4-?O*kLlJ)+SfUzVihMG;?jip(}OgP?AT__fr5_!uh8T zYZ*c`G;5dF9#6EGS`1egcwRB$H_Did({(-Y`>*-0M~R>&*yT7KCdrPqb+ekriyB?y zNh)X~9n`69#T7(TC9y=uiRGJomyEF3P$;QG!Bs^&QLx#AjhO z?_()*uZbh~F#|uRZixuPvVRjy&QRB;4e=~b|#6+%O083KQi{-C=>3WQ4d zuyhb^OHT7KX_!gE8~L3hGc!~`@GWMvAH1>t4~$cF0@yA!uYDenNTi7#{X8Ih83D!98CSzWw$U_x^)#%)v|MAv0cipjq?PROF3z`$~6KMu$JTQ-IuiiTYuoi^;> zN!%lxHurqUDfES-r9-7jSPlxE@R5~=Rv+(9HW0I|BZ0(Mta9eqVVuNCqVO zO6vL-cRmWuEcfW^dris`&ZdzURjPo^>-jJlL5lOHk(Vw@rwVy+HYs$Bu5fZkj6RV6 zYHA6j2mR0@6=j10)n>QRBgpB)F2t!JhkXpJCT;%12MicUEYJNJ>2iDwU}YuGk{Ar( zmJ#K{L7&3=iU&tVgi#X!E;vgb3!M&(3L$r22toyLzWO~nK0ekDwJ1Poa_Otoe4?EG z5HUP~>`MKVewc*s!o|%^b5@AfD8q;o*ECI>xk$8^&BtiCA zm_Ed4n2Nj^8`(LUJNA>&(Ahx+7MZdHGY6y*mqM!_bh~MfkbHr1SzjQCqZK3<9;_s{ zFRC?d{r;x;pMXpoH9%hpEC7M)@aRuVRvu{I`l0pDAV4duWNm+ssI=iV){FMtg{ixg z*r}O#34fTS{Ts%?C9m_{8r)l@Nw)+(l~=pscjD-XN`Bg}Bne z!1ewnA)yaka&ZoIgBH@0Hobl{_^utT^?Yk^(4q%iS0sClm}!FQ3TPeBWe2Z2pk3dg`Q0;JJpT+XTi}yRy1bm-x7>*^H#_!lJRRum&B8#=LoF9u zJlV>laJQ#1Z7BZ=UIu}d;#M-!#{*1Di<=;hO7uBnY2B+crysu2TXMhx&qnkW&-88F zQJvIEr7m4dus^Jt%yr>xzYz97dWNkI%r=TH$@Ddew23pGAccl$+F!=`JIbrCWFGMB&`I^(3P7|7nCWnT{p$&G#^tGN~67PkVC ziDhRv4Ns_pISKI^%CrjqgLVSLV6@z{2GPw0=+(x9Xo!=pEzhHY(GVWVy!7;#?d@%J z)Pkr9ZvLCWc-ZG(7np-9q!DAex1^*g33s>T-%8aU-rT^{?XNnI-_1a}JDEszvK!WN z+v&2EQ&b4IV?M)|y*?**P8H0{uliasCC+5dD3{BaZr}*_f_Kv4Nv@Tq@X>InR$9eI z8=vXlHD4=zc>_E7FMLf=yEz`RR;&=d%Q|Jg@`jn-t6XaNi{ZtD@Vzk{&e_*RB~K_Y zdy52G269?JIuBzYp)dwi!2bkr8Zx-k$hCb(9o(6eHEQm zCnFz-5QhEI;2|5osO$XHTqXUxv}uHpeZ`_j$H^BfOdd9J(H+iXCUn-)gTMA`Ip z<65ZhcZUlO@Ymqi{cF{2hi^iZ$MOs$iMa|;O_J+Q*ed5KEZVD*Ci5)Xp`+2R$$}BU zWf>luQdpP-Vkn9L#j|prs=u=T#s#3 zbxZ@{!SR?0bz!7);7aH2{UhFMi*S?-+mHf{1t9 zQv6e)dF(nv=Ewh18s~_=x=P@!#c04R03WNpE#z!MiqhdB*M9V#y`)Dr_p*JXhmZYc zB@5I7s^I)e#tA)V4F4EEi`9slm--bibp=RoJ)9?kZFPx0HO~dbBHg|AF}dH~EA+82^i?)KDgB_4RlI&f%X{u2IHgeYT*+bWDtB(~F3a>xGkyO<)mukZwf@oGG}0j5C?VYv5|W~HcWfFd>F!Y6gmg(O zcS@f2D*yX;5*3T@*0i=6a+uAZ zx$v#Ure%iu9ll~F>TmA(r4eaxyw_FwV#`P67bYfV?>#eje%syMU2N=%oj&H)hDBs% z@{^ml<+G>qKK@GfWqWJsdz4-7g#&$+p%;8?Q$GrKhG^p4sjjI(@;4e|z_l{Ftsl@l z!@(SS`3j%@RTQ^}NGP5b#?Ay+8o(kE=u~!bA!E9Y>6U_tGnxU}2D6=7%;_%QxTrab z*;(<nKDozT?5{!?wS3{fB*q_owZ~F~LUJlqhsi zacyN=Iy=}Iqs-(~Y~md=?jc`R|13toCC&Eso96iTPZ;FbH9VbPctg0g7tJdEfG$Pm zNp1JJ8qRD5XF_AeS0nWT?>G&x{v`SHZcTdf9@IbF7_Hs)3S70&!uzKa=NX?o=_Eo~ zxmuIQ*JP%%&LguBt=z*<+$#G0>6y6r@)hlw*x9@n8s}&DX)LCk7}iX!X-d`5qh{hu z$@a{d_ZZ=eiGkg^G;Fg!URh!1wrDnX5$7UzDzadv0$aOP2>-sRaWcVZ5PG#csMJ=m zR_ToZ#E0N0h%BuM0D4b4M>Mn>@s08 zqhC(GE&R{b6F7F#uucF3|MRwA>z+C1eKQ$Co>0_WDrc>aB`10yqe-3hCsOh2M{7=% z(J+EnQ5A?dUAsCb4;L75q-kP{NdV{4!)ONBm26T}9YFnw!_swB3f;cn2EV z?{aZQ>Ew()05-Yu#3Agwpc_}DL>+_nLzm^a3MjOxA$0DP=mT*YJ8=c8+>|X^9*ocs z4(;W9SY+D|&{<(a(#Dcrr05iK$sfvl+x%B#C?tfx4IOi$h_LQ45vc-|caD+BN~$@6 zZIiswXZ8UiIcMb9md6o0PYg(f8;OFkkui;v%5^t>Hxwb-gcr+qNf0hz7Q732%eF&_ zD}#eVG3ZZ>&vvQYD}#P3e(Ylt&NkkqwF*CNM02*IMVm4rE+dOj2Sl7M(N;0Ro-vT)*K>m!2k9etl$}r5BN!`qayJ*x%7C`%n1H`@LpGbqnkkQQ{He%XNK) zFF7B?BVx?6kk^}PKG+=oNd9iOE<91NUTYe{>7dXm@X%42wO3gaJtlT(D0y-Z6VNizi(g1;STMF> zk#N?Ge-eHC%1KLngCwVO)F5~cf02UQjmRj3O{-`|oU=12IqQsFh=ng9I?bh*)Gu+q zTWcu}X~*YtTAz0jH~Q;X3@#CxM@hb)914a~?d&M6u$zoV7}YWzac$tX{y-W}UTJW~*&6P6)@J8E z!mIjqn|VTU+I)ZGIi+Pd^9dJZwN90s-~oIuyKSA8DABg~Hd+(#%EN?@l=UzU7yYXi zC{b6%RA;ZaiBJWS*1a!oc~{O?w#N?1%h4CEcBC8+y7#)GKY9(Pa0w=X^Y&Nc&*}Zo z9i7K4J3@)c8@!6wwGY;=vnIAIQ3=t|_gwQ3-$`?>I{f+rptUWPp#S~NtoTX&R3xIY!!iU->4!3+2m4gIPP%@BcdhQQK%<}}l~moa29(}!`$6C9 z91Ip+hB=tpRLPlBskmVXob9(RfMxn;^5|F2Lr35xXd&hBw8Xpa`e%8vhp1*BMi#%B zymlsT4sV#|2bg*GFXgS@nLBs%rVovegD9ynW#+>}^j`JK{4F#`0#-|kPvY1A&FRn-0cV_b1lDcI z`NYIY`JH8O(gNsjZSHh*%5 z9$T{iLu30sT8PmfhZu+^6(fFaqv2ie^HjUD1S|%eCR!XA@0J=aPjMVj(30mps$1N z@uSMlhZyRFiDu7`D@&pVDmSH_XqpD)7(02=+F+5pmO!J+3eF%+n&pK~>=(pkvNqq1 z8GJn@;@_!&(t<8Z#&WGhw7?W&ym6`aXIUQtR7CM+F2ClMyaQbbW|IX@>;@#!BZ`5Q z`6m^?KoElTdrj!sm~_N>2J^ z(8Pb#y&J^-D&H}STAazd(oH7kY!Z?r$kLEku?Lh@?3KcizTab?ie7ST4l8j5k%^6hUL(T5o>Hnw5SIdD%k=SR=}qRD66b|guI zgt@kwQb)C1$kr=3ELj$4xAFd5PEA*iocWGrls8Y<*kt*ey?^0M!NscQdn4cY`1q08 z&ck$LQmhZzS6+Qijo$S9{Iva43E{ZZn2KE~%`cgZDPuyTyiezC7AN>t7Q2g_sJe9> z)Y`eS5Zk#7LpN$yn+miW&ZxW`Y|42Bq$nqnRtjgw-%}F83STw>?L#r8w)X`L1eXCnIW-#U>LgS92^z7jszk3)+kgbFlzai^Tju&UN7gSS z+#FN?4&X(RFZcGsh=O>kL(Moqb||wms%1!(GKB4^Yoy|J?0AO%1|Gpr3%l?|B4HM! zM+D!-o6l@5aBv`CrvB;>A4tAzf6qbDVT`LHd2P&@U|QxNO}xBG%cITlV**vDu+Wen z{Xc)MZiXV~f9)bhPO_AMKs(k=?YS5XoaCQZ5JClNY89&Y$WU^Y#@Vmo^BZpY4RkSsFlk)szq@CJ1jX^v2A->_m`KgTw6TK~0n zE%N&l_m_Jh<34{tb13?;A|>Ka&q)(jdJ}@W zFnk4vw%qS(^Dy+?nDGW`-K#7eAYDdcCxaSRG<545nh8_sJ9hhW*0c1gdVqq6ns!hPkP$#DBR@3MmW$&CANuB53ZRc8P}4(bV)nD^=6qO$QI z!tZ)+=*v~`_3O|bcekcQ$xv&KCKr)@6)$6GtNxqc?d&GFw^#K2Bm;}M*tEEb)g2~D zxtYV~v}KI8;LXp;9wQ^(CLK$GNZ~PE_GQ^bhpGiX#UWleR6PWg0%1@+WcVqcWOamZ z*hr}x0pa27jf|w`IRD=7O~}_vaie4Goj;hKv#$FXM9hi_1iwEiWXA@zI}_4O9V0>b z4;)sBsz<#)y7_M#Qzg(tB(W}-0>SJ`{<$Vp6-R}KCNp>Pzr9J=(9FocqxVJRYqu#@PBcdY~x=#vC1w+QGs;WXr_Y#|F+~6OKgNHGjyiyi$#QE zw>Tm-%sTVZSmlCzzr8!DHWJ;;DyY|8<#1SiEEu3 z;Epld1M{M-oRoW$m-kY_l%&^QG&`id=m^Y;*TET$r|r3HUA|=Chp7fHBbEUz z+tiFZ0pU1);YINm49^%%ldp26Kj3Nwc8BU#TSXHX@LK)1r}bdjzW8JRq}nFV$bgCu zjqEwN)mx=b0GZX;)j<<`n_Ts$^vlds{25RgCb`~}g+=K3JCeI>PrgO#K6tP4>^qss ze-SH1`cn*sU!S?XZ*SRXIj2DvROsUll#Gt3zs5iOz6zt1m~bGYk89U80Q!bx7n3!TbKa^p;>|2;}>GTxmq-FUvtD z|LL9$b}VOG+{kVktza;?hl5ykV%-) z5kv#S2Fc3zL&zVL>iS_%F+v}Db0e}5V?f(P($&rHp6{LDba)^H^lGt$(eY`95vgz7 zv2&-`7TiKQ1UsHuXV0f3<>5vsjEad?PYphru`vaIbwrBS9%{&ZOPq$9UoU2F({TIU zXD?+M?cuS(;C6Z)&MZfr)UC;h<>J#D+$9f<@(YrhFQRe;spbzI!bXtUfR3G~GSP=AAS9d&s40)P8 zmX+)2_w!R=*|!;*;YCo_O5N=I)h|G0|7>6dW#K)M&gSV~Kqofs(X&KVpY;f99Z$<$ z#eFtNKQN51Vpk^$m!$qd@#v6g*??X+XQ~v^_FpK*j@&5qA(W~hBk5rpZPg||?-+X* z461zEFr(}UcbP%cu4Q}s^^kV^AX{s5TAn5a8l%l*HJ+I1Q`7|=Ct5buq#6im(f92? zdo(eV{#}g{w_f?s-rT?{^L|`xGVYERLG_ke;a{MjYWc0aaP?Cz+Bu&)Fec}l+b)^! zDr&@z(bUA>a>*LrlhH}3K6tF>Y|i*meO>tXLXiLI72bk#vlJY2f=x{BPN$hO>LxTpgJvabG8-DX{7 zaJZyuN~UfYC3C6GwSOGk0G_3~LXU1oDurZrNp0fV7g; zljqDsKEU9m*b;gi7s=T)z|e7ecSMzyU-FW{r_tW)KI_4?zSyUl^-V zxO-ix$iJU4M!#J9F2ZYpI%umOIT{@~;6&8!S!~VEx*tjlC`4~j!{&n!@pvjFTHh$+ zRycg-KrBNA-R;uHkB1RxGYE{S(70yTLZ-afu&U?h)-&(%2o-dH?*bMNZz;Ln9sGct z9Dh|@d()DUP0Hkb05;p@kC$uHRhA0tlfaHh-v0jP%GDCp^f^xZ;l6-h3wE!V&+N$L zgG8J{`W0iEa%1yOoB+9unk}Z)jod}b;M<_Z8blW^NAxwTS(22qB*;4Z0#E5#-E;9% zMWE2o38bJ|CJjl7fHn1HKg3r(LCLPB7M4{kNi>-ExI3SL{5K^v)G8*y99Up!kQk^-zmL#Q0={aAAQQ0w!GY z&`HQHqHc0@fh@Iv@LjrRorcrZNRCdB{Pi!JViA(@T@eX#^33!DJ%_M}!A|U&X|oUz z)fi4jZ1yPJk$2`QX=(;z58j2H3Sv5B9Ie4W;s7b5*e66N=-7N9Kwsel=Gzmsb~=Ri zZ2?&D;u-9mb>`Jo8@cA-zyNn}aueDw-u6=>y<{>Id1J({B{k+A zUZe{o)6XVAy8bzetPW2%ATkV$QeapdB>B1uSNNtow3ma?_fPLx{R%z4&$qdi$8(4`7zLs%*d+2-4xg$|48+>y!Hs7BqA666V({3tC6H zT{d33?M)v9b+B=BPt-7cE$^DCy8ct7Rjk@AO2D)RlF6+};Bn6Hjk0JAUOppzo56Z-b6I&DJ z9uf+Pf_GlUQ}-_p)x#ZuZz?FX*N|YpOeFd|tiP!f`&^pcCg) zm{PWbD5)thNxZ6%kt8;-wV$o%M4@On-Ufl(^lb8)|1e98d+8qWjp=bjf-k&{%baBj z`Ee2!(|_Dw{AMdm$Fkyo|H#LJXv}H$9(9ikUk=QEu2ky#^WaHX>;f6RlId`JxkDtGm0T3+>0LT{| z@iD`5+1m>^ZWK$t)I|3K5k%89Knq4(4Db?jH5A_-x>%deh^10jS0s+Xt@b;u>0|VS zH&7^CV{R83Ln2k8Vv&#`k}c$DNL&oif64)!q+{z7NDR~|Fr7&TSpWU|VfOG{16KBTyZzk;l1@17Tk-oBe?od(pXv1R!x&~DId4eMx z(DMwQgS=SSREf~9T45|3XDWb?7gj|9&8x$Q$oE3zx_`N2Ghvs_E4F@pxUl2Vj{V7C zoO1Y#@d}k@kg$s~@BO$c!+&S<%sE|NN*JdD9?4G}q3n0d*dVl3;aOh&FU?{Lg;N2X84f%dy_ya1 zNJ8@ix>UVRR-F>B{Ga%9$J0($#WAUILD|y)#suaxfJs-Ks*_cPd39q3d+?iUM<1AHDIZQm?v>G#&$s!30x~e-@qbSU_y8<- z3S;4-_2y4A*wYQ3AkihvT&8W z*c%U+!zFyC+yhl#vKlAVi5+zyr}A3DPcdbW*)jdP;`S#Iy`$ykQ0@%LsWfJgn%fc( zUB9^|=Xr_yULb(Pf{MCzoku_P>)Vonj|+IZ;GmT03D{SKNYT5O*NM#TM)Xt*1w7!? zE!y^L4huc7dmVz9r*wIw6Bg|FVzX$1hLYCcdCG!-u#C@G00hckR18oiJ@k7~v1Ux? zOA^@CIcdlAQ%H+Xr*f62F zbgu_X#%z zhuZ%jHL7Ny5z~&9ZQ&rH*XHx z1&{HVJQGnap&DTJ~(Ul}9sg5Y1V%B0es@KDkifi~q>NP;r z75qP+e>Bj>nzXP|c>S6?-=n+h-r&tg#OQ&6*ZZ@T-E&_wlh51$etqSjKA6-5B`c5y zqEnpD1Pccu-Vk~xlj6^#MTlWXP|)ezW!qEz`vjd--KdtSZJeJ@xUu8^AQ~7)5>P^& z9+qkPqX%2C<^_8$$0y__#fok38{u||Le~A*>FQ?_xuL`+sOWYElKDPwmzti>*Srk{ z%n<^v*0aZb(UjJHv7m=82Xxh1t&fBG75^aJJZ9_roc~5&8uqP>2}J6v0eT4i#V3@m z7ef|PvvYc3c`Zczth$l0QU_|eP+IAj(6dbAS&o4q`Bt}m*KUh<-BeCq;)mN?zpn)a zz)sr1kXC)}X=au;w>?ir`bJr~4|t|~$8lbPO3G7P*<)@vd?g&otMaLz3~kQfYTc5E zfdhKqi)SqGZ9=yCy$VwoN;%z1;z;!!iZ9Swm;mw)z57m9&K4FHITaNXb2oT;dU^xX zO)*;UiyST(+YSU))*^!eyAJf;dhN2T>=p;_m9g}wyxe-iIOmOsJQ85${_m+v_X+B@ z91}=P(g%Um6h|L&@YvIxCY|8XY(((@?XQcT@fjjNXf%q5;bjeUE@$8C=CwdwIihUg ztDOYyfl6SPAiNby74yA-0UxbyoxNV++;92kvCp#o9pg`Ba&o2=WF6t?-M&AyrsB!( z2Z!(&S9?wJRhPIqv0~<5}C#S@FWOGSXfa`_jMhU_)<=s+h>%4Uqo%l_WGddTf zG7&t>8e)Ld$~3<(-pf%jp``FPwb=ue|F@1aDb&yTe8w;oi1UxY&Cm7Xz8mza|BDKO z(}iH#uKd($9#d$tBCe_STQx_D*s8~~?H!oqJu2XQD{D&AqCV68jT=Y&_x{UszSv5o zi(w@wEcjDcKZ5c#Z5T0+P5DTgxd9gIQ0{D<17()aE(>JpzHM<{e+3ym3!;VVK`bvd z`e5X@bp{X{g$`;aGj!PC-_~)iP-c56NE-*i^(|=WSL9vsk}c6VBR4)kDKNSIS$1ef zva}F_sMtikqfpAG)z>qM_T{n=6(s!p))5EyOenv>)pCS%fB0~Z2mFUZV5FU?!%WQh z1QFz!f7}rCzv?lEHUHfq5TP6nxFx_4xu=BvBV?lCEufHCAc8` z3BGVdqL65CeBP2NVye%?Chr~_A0JX{H)HnmKQ*Q&8!)J1x@s(i%f#5s(l~$3$=P+9 zUBVFmR?MVryjgo7{EG6mAxh63=XwcTGx5cH6>hkGMp#RS2Q%{tVwDFX$mz!G4BnL_ zHG|)4_vunIvh@1_Zq9pJhZHT|uPHgBI*Eaub(@0AUi1*t{0vYZASnGHHHZ_H&VdWV zi_-yrkbdNYrjKsscnY$GkrhLB(=>GM)Uh>-WuNm)l91CSzIf3X8tg+kFSMg`)U~xB z%VGOffJp;<3PBJ$ueF^TvT%x`5u2~shHdp5c`k(m;Kt9yxoBV;)U(bE|1u~2s*pPh zwba;wuR>m-=N)w`&fNl+bV7u`Zp{FeLW(!N!LT`$4;yM<4f%de8nls*u<)_=1*Xjl zlmX&TewX}dPS56oiqv*%d6wTK~KVdUHEtjI|ixOW@edgUY2cAJog|`Go!%e#CE6gF04uqmZJNAZCWE z3%Q~|m5!I#193d&hldikv=Sa6u6fY6S_kIqDqsu+RW=r-l_ZOR8&N*!pnm}@zTmjA z88iUzD#jpWA|uc~j_yyjSq4{-tG|EW?z!YUc4>#&S5q$z-H#OY-t=7L@4x{tr?>kN zCKB*y=M|T*16n*=3-ksNd9H)+aVQCZH+1)WBvi_0t0Nm38R2($kvi3hsRJzG#Ce8F z?n#)PyV`bdXSPH{aP!{u7`sI*g$=|qFcip{{f!aVZK)!-*mwCy=u3RB3*wcVnNp!_661k~ov_9lK5miY^E7sX5tvQKN8A(h}R+*PMeVn*vx4AGC{giV)!? zUESTFHiE*!`l0^uD993wO#i1w4`5wtWM4O=wz~0<58JRtvOzMlqZC2dS)tN*()gB_ z9>wHt5m)4*jyQXw-#b6=Cv{6pOWXl*@Z!X>8GTem=?B+~j~taURA=o$HKP)7+JhS4 zyrbZx=W_&u&2s1HaOF6gZwKWi2I(AEjdY=s8Un(T*c5|TqKu;>Cx~H>9EX||GrJ>Y zOVH(EO`OQkwGggQBUtRCcm&`ul>a>E>7~zIDY8~+^!G1`O|SgK}k6tuQS{9;Hpz9gVuhpweqSKb)-%Y$TRh8t-= zeiJt_{=9KZlmFslL`CFNY`+o%>t3B2Z=!rwP%Z$wI$AC0{d@L%QpvzoyITJB>kh#3 zQam?;SL&wCe15)@gxL*uTU%R=sHBg!v5PCxK@+$VZHs4bhJzBPT;OHzp!Jn;*5yBUy9nw>*zvhW=a=%b z5QN{w(3v;8 zX61RI8BUvq7xS10B`BLk=^Hjk0QL^}5e*pia>Q2SSwMnyPtD;_U48wBksXOA08Z8|%O)g303$;Ayh&+ttib*I_b){51!I6W zB%CPAiBD>TOeG6IH;k%AYw1| zoiXxUE&P_n8yxNSjh&Dh$I9Ce+}5@l)p_#1IG{!ZAy|J1EmxsI^UU~^dgMQwLK1rJ zD`Fn#6Ge5*+b7Fb84m4yvIKt8L7aj}p(dzJOSm#_gvA3h!pv7!ksrhncFuD$PMaK_ zlTDE_4W?-jtz*A@IYW{1RBR*xjna;pd&WaLJCmPsruX_nW<^kwZ_xX?RWKe%XF%Y zX0~Y@Q-6KU{=8{K@l;c7SKx_5u+@6cckv~(oR*9`=ro^cdFi-vb2#TgF)(vWk&mq_ zoN|vW;zzZ}JJH{d4s5@C2Pl`oV7afA_#(zc2!Bp4qt#U6i?N4^F`U?cK| z;$5amb@7zur7@MzGlBi>gb#H*1{00HeNfOX%oO|#*q0W&sYKQu<-9XFg^sg;HLk@B z*xv_SRpdU5ov!*>%-b1wHSzsBrXJ=Zx%)!kJ_-9}xF8?Fff@PJwH()G-R%%p=9NBA zzu(`~OMswam1lf}iz|><48eM9&R|amfW;#mXtgBJFl~0w|ACM}Z3w+si<(N>_A$YS ztWxcCcylA_y>4qUr3X_TcRyK!1%d_^HQPB-45g%!r5#g@ccnP-+s|M&XvQ|cPyj6l zS~@LT!Vx}QP3Zn9a=cDNzc|k81v+Ej7%`o~$NZoges^g> z-16cM2WXv@gzZQ9wY#UvYLS^(gq?i>v!=|`@l z&!EO_WD$^5QxipDicpieTm8^cq9>u945Ch8n<|RqeHxYYftI zBW%5+3W>^4-HzZm_*H;oAofCpD8H?-XIEBn4$K(a>HdEl7#fR2spe8%y}IAqJS>3THb!&1?09!2)TnNdV~Otok8i! z7zmsGe3t6HZZ1fg*4%TEiLrm%$3$a8?A43>2(Lb5cf$hKj|&?np_4ncgqBcSD0ZVR z3UhXfjIl^tBD;bWG^)Q*7woWGo!Uw8AjxXHle;}i5d7m^!}vbSM|7+;fO+{|h~iMJ znXJ2&Gw2~cvS;>g*gN;%7hH>pd%lj;-j!s+Ez0XC6S3V46Uo}NaA5fNU9o1W)u_F} z21t6V?cdKX18y2IouJ@MMCq9LjXiXasUfuB4$~?f0C(nNFhO1gK#&<$og_NeA?+_% zSYO7s_lYhT1pD{hKbYR!3k-~23f}0>!7%{Vx4O4wE}!}QQ>?ZbEjH>?d*8sXC2?Fm zZa4oJ(W3jzAfAacQFRXDYQRQDrI~1kP0ot~QC;7v+;`)>gst^_gb;zpKQ&ZLr%UvN zKnw;-9j33t0oUHWn4Me*5PJb#-kGnqw@!Tz1%1IiLVK;CGGtR~IbA>OUmVT~WC`Ih zG07Mu#NrRPuJkQ=6>$Giy*aoa&=Z+H`youePaKM?C7nGbyj6dYIlW~eoBX4Z;qzWF z=B3!{Z@u@v zqa1`Ft0vr{GDLp zQQM`(NY4ML;o8;!Bahw8BDLjQn~wL#9O$=7nz)ddtOR+#w1>^exJJrZwgB%%?9nsm z<#f_DN(t7(9AaskZy~=rRVjU+dz+GzLpQkujkY^M^=nUGbXa|TeWA~DpEDbcb6b!; z+7gw0i|-W8{U9O`rbbS|t&}t%AOKoNw_BtN$z&xY7$(lhI-Y5k zpd8=MC`#IYcdL1q5WE%ax>~&{3_BMz8@V3vmV;L==o=T5#{BQtYbr|`n3m9bx9^Ik z4nERAar~E9_l1!J+BMyEWa_L@#doO+_Og<4uipD>CZ5CsD+CbfDSlai!>Nsr>O(UL zhdv#+U4fIFW~Ub8TSmt|SyDtnLlfA1aI#FcxNO7v1E%wjLDnoy3Iz5`5(aanp9p|l zZ-B@rg8SMKd;gw%AU<2m3y5fImE6anAuiWX*n@^JbxyYik1{s_WDhskY&Hgf!9YPy zRAHV7pc*CEKrW$Nbl(lEU2|p}G;JTtACd{(!9jfhRpELeru9c8MnW$yM z3ZETX0JI6{1yw7;0-q{SQX;`{%#dnQzzbL#kHG|uo`VAiSj*9&3oipB<0|+P^j7xM z0FXc-nK6{!@7NCBTJH(ML0j8K3RFSxxJ6=b!J55hT-nC!i9)zmseE6SV!&DdZw27Q zJPlry`8=1B{UK++)+Sw$6xh##sOHKeOOxKy8c88lNQn>m-`F2XtKp&=U;+nDY8ZaJ zmeki0pl{jBZypUUg*ZN;HSKzH2 z;XM<2#wtJB<$YJtZ6DbQNXqVeZpPqCp;m-f&*C(j{;wpUR3yv)LT}M+Qt4ILIoK)I znh)E$8B+|`G72>*v>Pd|kS6|!HpaJ^Jt;xGe?c1Y*|OR;O^y4Enocpw06KxHE#i?T zFvA_)(?u_WQ&jLO+Pxevr<^2NzgzKP7 z*!-8&7mgW3tUt)S9>QFjiD*Hyh>#jfj)qc?a!Wbq|6&>Jl3t=fJn_GQ8$6TtBs+b3 z@;va>-wz=nNE)NNCDmhHU*9Ag0VKrdp{B*`l#~x3dYdjZkiGw=kQoOj<1-)wfTa5pW=jHk!BMcJ9gT?`rn0*NhSK^6Rj;JDI8*&hF3`2oPvts-NRX_K z2=75;8HnYe98g5FS`Pa!@sis@^+^mF*4giL>}<<=#9`xVwV^I{UvY!4Z1U21Y9>HP zT>JJ3T}mHby0Ql(djN(e<}|8)y-RM5U_Fcv#BZq`C8rTccz=~T1b`h`S<``rTy7GQLE^(X!{YsjSN+h(3Z-TYQ z|GlK)wv~>L2?>wE{4-rU*aDz)3YxU6teC*r_B)U+`0b{#L1(pumsf+ZT;Z##i-9c* zc=HRwjh=<5e{BM4wltiae`V>ORnW)#H|KKuBsy^Pl(Z#INfeAMur?Io6A(!B#|Y3z zliMhy=!0`_qHMFrffH1uF@4msB<92NvJeE&3(flMBx@dEb;0u<)QEMrBMrRPvV zyjDu)QMqj?k?|_^eZ0I!6YjZ%=Ec$Wc=j&T@2zX_f59C&Izdi#ovOR1z0S z>kl3rGRS<#%@G4jclvKIgkDRlsL~ovmVo8B9~1S0JFTDTEgbyR{u;vlsp6Ek&jz$N z75|V4yL>&C^w^&`_s&O@rmd{J3LcF?HIqZmJ+Mb*?;=_FOD>U{Or~xd19~u(Gxz7s zX3AVdc6no=7ZgWx*sUun8q5^T=c+6Lh^QA%tlQ?-=)G8Fi6!uE*}e!y{43S83|Edg zS1r8(6bP_atWz34;(;|IOXf=48p1{yV29X7Er9&0mZ<-GA}~p=ft99@ zK#t0^$Ox(vwF(3ibSFTSSc9{baUV?7BE4fY?OmGwh9q`~EMnH2I-Qk`Knku7kuP?2|AjEUIfJ&Es zSgL9zQfyRo2l0)gT%T1w(SlaILe?p8Zi+a7li5@$z(p?;m}E z>2K!(B%4BE^NeqrNEG?t!X_&SW_SeVQ+R-jO?FK(vF6pmh5uR?fxY1eyoz7r-<83Z zii_xDr4};@sqA-NmFgtb$?o8RwyN>L-Qn>%$?01+k#@HHuJBkIv7n0shh8?c7XWXg z4HQQ-0T&mSBk(<=2O%&~>xB+zRRw36KCVW9nc_C>LRoa`#(_3aXV6jenPf+v`1lh_ zQ-sNn`=@jz#IfuqW24!_DSTKFefb-;vL$r{K$9jIIMde8#pd1vw)UX1V-*?N7&!G}76l2o@{ z?#d5LC4@luq2unmz9KT>8U7Vf_R|xrU&s8+74uc7x~&wm4d3eo$JJ*_JG*Qf75}6P zIHY_3+1s=0(Ta+;u(n=ryh{M2%FXF`nDv+5+ z>bF6iVJ=pmcgEDIPAO7+0-$t4Uu?-DDpRN0U8tO?(6M*q7^thnm6er{GeT+F>&*ht z#w_MNnUjKk*jatI_=AGkgW%v9Gwd=ed-RLDx79c5)9^U0OIAbfJ>N+t*k2dQC?|A&2a4xN*XM+U|54-9 z;(46q>PzX#lgnSd`*V9#niN!u zYc}8WP{prSkRcc%C}%Dhr3(xIZoDG^tN|Po8Zy3Dgm^@oFR#MqSdY|@dvG0*H!SD7 zAp8J8b)IT&%g8#@VpqCwIA<;Tlw^&wCja{=$&1(<%v3|uHUOaW35B4eeV7i^9Sb#r zQ+)zq^R$DCZrq!hK36ns{80HqswbtvxH<@#4MrF;cpkVyGH$3ZhALxp29NE6Uy1$i z@#J}76KSFLAjRDUuOv%c7}Gh&5gRk!;#eG%5q7b8C|n%~)+L}Zsi0*AP=I0RO-A_p z#W=?4#%dfdeg{@7RenBrCNk*yPTK_2-yYCsXgZ%72w+X1_hbBkzHw;r{X^ea#Et zGXv~Y>!N+g6%A!Qz&;ubZlZXh=(IIfU^`nuy^=#1jX_GM!VPmLYSu-YKI7;pqYb(e zj-f8m6M^5XJjS2l)9ZBE)BY@)9{gwm>-~|hFzQN6ORKR8 zszt^{bruo$V7^P}XMq$EA6X4w_#*AxJi?#rpZiy90q^ga4gvKETm5r?GhFN6 zamPN%y8lC4dc^9GFljZxCl&xNY`dpu_-nONw-A(Kg0}394m9(?3;&1KTk^MKSVb6U z!Y>B3H)*|A--z)!#v3Q0$d{xBq< z%hW&yqH}|nR>ZLhI~h>BiWgEqfZ3WG%Jl{6Y+Cvh_jFlp=~&V=mWW?s517P$0>XX| z!#mNdl~73HhRW$uF?F*2FUUb?cPpciS-w!l4N;#8k6(o12nP|<#lE&U^Vh6L=)A3g zGhxQ0?>UlwJz7X;>DBJ|%U;w30|>yO#wt|{5bzb4$>f7mMGyt&3T(@<%5N*y#Ah`= zyxQq3{b=a4IxVP4Q=9Z}DB#7v%&Vh8wX_A^E`I!3O%D>p!Q#@0;7FzLtQq|Xv*LH{ zsE(~uRw2~bN=c|#`^U6_B>~^H?Abd)3VC=0R&tpoB_$hfPlh3alo{Bab|!L>!J0n- z03&Ay3#9(nXFMQ#uf6rpL$X+`iLYH^Mx2FHK17y;E^~?8w1`q8Zqx^I6dpG(bM8JaU+E0b=BWdCxNxDks{0a*AWNAL~~ZES5roepY4$j$PT+pzP(eS2%`=Y%ZX zr?F22R_O&v0hvg*c1U|jb@$mB;XnK{xTFs_KRpJuuRkEjmeWr)xinktGfC#AJlZag z(2}-d=Am5tml;WM^uEX(DhOR7A_CmSu;)K!lh?GFo7S4lZZ0I$;?k6rM{p9rFLOCm0s7S z%4q3Ph<>-0fpNO_h&1GGOAQ2|cLuXuzANcb%RAbm9>R_?qYH0$bwm1YR07V}eQhzh z(D1qK#>Pgbg<1-q5dgHX4z2_IrO1FrGWcK*P&gUKN~tn=n)sXn=mN-H%?O}v z`+j`(co;P)R_qY{g$Y&|f68eSQb638q^#O~sL3OIv7dEh_B_kloo=T3F3E$>Ip#0i}(t}rrLbc^7N&Hm%WOq$t$K7K@jj?Y0r0v*yXC;e18qVMj6)J zS}s5vF<1|*7;*$%hDv^6ZF0K0sjkWSHJAXeOB5Q8{Im%<`U3*b7mJ(&yGWb{Scr=8 zbL(RkIH+EAARlSW-21GMJnOCqeO4@7;2dTP&|i+N=i1Ogt}X;z5~L*ox078LmJt_U z2d;Hy$7{X7%=f*8@)|xCmF7wYw{w$KlmCFUSK*=Uw;7_vdHI0z^pgjIfyh-8ahs51*;xeC0E5T>5O?lWPQ}dsHOX+5=K8g+j0f7)$>7Bq> zajj-jSvrN&Xyx$mT?jSkzGQKx6!BmOJ;@ul?}xhO^&#p$g!BW}*~dU+CIs!CHYg^R zlKl}~+?egDaVJ0){XA2qTPa25E;yC(&*cIho;_oatSbrg$d2p=PZ!@LEK z(oIA-GA{QkvG9payU{2koX{01tmfWX)keOx5qSQaf(DHFm=GD}Sejoo2Q0UEGymL4 zxjFiTKyhASR=h?&074iw$1u8)X*Wzi8p)Ov1lI^QND2(Ll70Yef2OiY9f##Jz9iz)5c=t z<10}kvT^qBl+axNDqGrzm);jKPf06A5gW?N%%00KMI}G$)P$CR*R_qoKq(}986aBL zw99JfF=zQoQ7it->O7Njh7DC>Q5V7x6gKQJCN*>l_Vxo~P)lBRUjaiFnoA&**|WMH z-vM%1@77%f$69va{l}6L_U(V)7;v`hS?I4jczSj{DTq+G2lnKSm~yz~)zrQjA)CCG z=7h{CtKf{c`X~z8d{m9cfj!B5iQW{grkW^i9jg6Q(Nr+Jse0oT2>Rdz+@0%yo|Jxj zbe-i*;gXT!Qt=|V)gLyvfvMHY@adavplc1|Izpg4{GPDxz4C?Rfc(v6UbT&GJTU35 z{G;jqqotA9==`^NXpR2BfA=ncWw!?KW#Cdyt+dI?6zsatI(skBI3^%emG6F~7ifQsRqizy5n^@zp6iY)&jcTZ1Mws>8i9L=kElITaya3`DECc=3!`ekykd zp2Tx^OSj%qNBnOa`DAe72yyu}oZbR=_j;1lsN0+gVEOEeufH$Gs$@ktedi6TBkew*!D z2LYN6kOFt81750IyU+Y@9|QT>X8^MioU3X>ROs;Mzt?GDymes1=y||k7^x*j=5ICu zaBu8lec{Zn!xms7+j<$4NMtSdysLv-WiquQIF=RnF0%A=b#4`HwT3HzvLeWFBLrbP;RVQGQ z;-tSf`w$O~vXCWj7CgYTa*f2QINxt{!%T?#@rbD(gNkWTc|C;sQMKFkFGVXh9bn^% zbbsoLFv^h3g8M>)X4Qhg(XO^%JK)oUG))S^Uwu)@!WW2$C036rcqRwpB(}%?S0yQo zGR4%87k3YoVe-|-H)$Wj!_!v!042S`Vle&zg`4a)r=yV759FJe68;qsfMT#rjU-X{4J2 zCm&G02IN@ zGGFh0DEmGrm=vEGvj;g{i!~Kf{zYPF83unrD$PK0^qtks1_nzEr7zTma~j_|&g#=a zL>wJ%l(8P}We;!l4#*aVB}QGUS z0qp|(Z%#Wj-*sI3U~bSK&~uPI@nqee@tDQ|xsu(!x^CCUeC}Cqk+L!;Ce$}M zlc<4i^Ed3z&;L5Q zQ1#Zi`tkiFQX09+ARPx_e53W&qbl=<#N9{_d)z6~>yo@e{KZy$awRJ6C6nsEJ*9@; z?B(gDO%H4S)zZMlGwADtRlYM;H*RtfIRE0sL*~hyJKx4Z&A8BfDg5x*WZ|9I1mfU( z0Lsi=1&$u4Kp-53_%8ge2e77`{LcSqF_9x9vKEZ&3ItN5!wigZ=BZ(&cR+7^Go1J> zueD^e=YAwGa9r|#Hu4#`s5lmz3Uf3&Ps~Dp)XXAf1R~WB%~fEl0`qe|GL{iX&G*>F zMRtZdIx+5nkKvfQC_-jSRcuHBf_-zODi-L+O&7S znCT~_XDNe`k5McPqp8`c*SgVzt$7gkJJ2Ae*jtQ9gl^J{=fUl2h#K#f9#e9QPreW3 zTHwTIZ)w)e@Di zX+VwUlo`_&@)ifx0&W{x@!7;Be&|v)0X&W^K#a9kan^^Fj%)ahKp?aNP~Z+8aNyH4 zcpVxtvasX>9+bPb&U|e=N79+uaX>iuwiLRKgx{AA)XlH*72+qFJsqfVB#`b#fQYD6 zMbWqXb}cB+E=^#=gO9P@=2+2nZ;G>guO>%$mY z4W={W!^&&UVO%GmYfWr@JK6HlY;cS~TQIK@N6;iy6socDQ3_98Ya`|7++#6{f>&%z zbcS=Nakh|QDSHy;>3bzoRNbh@-E3FKC9+W6GG(`m$q4=wP)Jwn{tQ!kT#@6K;pn3> z+!Md;CV&u%h$nu)3+F>9AZSl1DJj7z&|n4_P)299wL}K6jQ6$6nuAbHJDwg}2#qu@ z>a2!{Ha}J_i3#iG-Spj8Apz+mO1Pi-Mo%yP)T!4@b(h{aYj z$IT4eLpyjsAEdE1j1#2{UhjSa#0vKP+>kWwR>Yb!AUv`=x`|_%7LzybrtQjfZP6u* z`n@zvp!I@?*H3UwSl+~LVt`P~upc{`)>gohir&P%#Y-Iho9owD5_*v@@7CH2aLSQ+ zwWa`}1T2$NWk;u%yMn#E^-NidFbr{}cK)OIk%(o3JD0iIYE7P`f}vshAFYo-sc~BD z!XHA;4YFth(763-Yt7hB$uGOkqTtpFz|s`p5H$1rTKKqmpL_GXv0LEDm-XtZS%?eO z3@4aH%#1{Tsb1;Lo4(bcCnuyx;~>l9MFq()TCG&4nzHd$v_V5V6>|Z_$^>G$Q|YUg z;7d}3b&LG;k1MP7-jCde>v>ey)q2Q7Nx_+arIQj_oqVGG(Qoc{yysS%o`P!pTo)q{ z4KiOB1nhrU^c&aad0h5%c4mXbjObLD5@>4fb7JgT6eKhYWjj3rltO6!=MzA~-=%&( z(2iB!cs<~rTazhQ8n7@$QM`x+1?Lx%I+vgrblTwAkb zoB%sy7d&4cNtN|>Ap;>EiUHU<2EbIhFWepoZ&fcS?JCYPCprLr(aJ6QVsbYhblTpC zR(mSr}+!a?a69uR7Dij)b15sAX14uh(oOa_#EuXP%I=0#|z za$*)W)N|jW(v|+0Fvn>z-a>_up1G*KKK}JuV-H_sKx)UMG^T{};)ECY9YUY@1jKTY z*A?s%nzT=Ju$X2Qc=V<0`hMpGYDxw#(XU)*q?7euTE=#l53Ef(IoEryNJvLO3>6iA|8lwEx8+IT1%aSI?B zRSn&R&r|DFz^(%zZV|rf1WTwxuYCpzUV{nQX9aU1mUw_Rf3_%pq;f|aD($922n{el z^R`k`XZ61sI^o->pd%?fHPN7$9pxTO|M}|7iiL33dKHT~)WPAy5`*}O0NwZ65R6)Sl7h4UhT+uDWz>0q!r7_hy;k2|sU(CyQ z@5izMm5%E+7eI=7V_&ldjmC^>IkI4;vl-xCQ=@$`j8*M^#xm`y=ilA0`yrs}wA_9LF=hc}b>mU|;v4p&VD4 z6D;%fP8ozB<>aE<7=9b;fY}y>06}ZToIic0O#=Zm3GS%JV*C|Q%%VU`&dT`J1THTGIYUnNgl*b~I>RTM73jM&l<4Q(Y?^DTTPticeRVHHIN3B{)dQ7hB68sJ`oB~C* z)4|?3l~I2>$x-)tPQK~Bn?Z8uhgPMahcnF$m8V`xmtK^_iIvK|dzYH>H2`Jn3JI42 zVJy?WlORc>pNIp&-Jt0mTMCbv(}KJT))CI3mV!nQfJAs<=-T(qo%`(si>oN0;fc>5 z|4CV1qXrNAO=3tjaOT)Rp3pV-b5z)v8gQFRB}NP4J%{qd%!Dji$*;!7i$)}?0R2G*uJ1T~yO_9sRPTP!0vQbc%M>_0=|Q$=!$GEckSQN&?< zXXjB1tN_B=F^TENeb4=B?UJBBDvsDKTgI6D5XUShpVmaSq)3ZVXzUqNCqEBSkR{~%M4c&P zv5V(CjQKtWe*qQHOS1&y5oW~$oDB?kOeW#7GYEM-6dCs(3TfP!-Psaw2nRsShs@7E zj{wmFZ>Zh4pJ@d2qv8c!n+8e++9*{21GXD{D?0CqPJHm(fy3vfCXba|`%v|{4^A@w zSY&&?6{6 z^xv_1$4abPv1I2NBAqUmUb9z|6Bv8T`B@04WN^wW58F?1i-pt))B%d7m}hUPw2ZgA z`|z+dfr4W>i$8b5*AEC{elyes<&^vBY{krGs)mDdYNoiRuQW-j65vYTE5#n*UCGE0 z0lr`nk&$`rqUTB&mbeY0fjROVx#W!ZTuN*wYyBHw4Sg`_!Z4(%2sJsDHv6oh1H$}X zLFVHVRN0D7%8V@Z_4pIDJ09%ae521ZhS-^WMpmME^Q;7l>7kg*3UX`hOVnHFL5J<# z_hT9#pX|!j4H50d@Ru(Db#&9I)IFb#-W+Ll2*|@^fDr=1cS3Id^i~MWJPX+7*tAS4hv2H#KaZTunmGDO_4l zBX`RUF5QILN0YVn8f%8#l@sPlina-~0jvk1w{8+muSOLZk)*OL5k;iVk= z`c1WqWY!co3Ajm&e}ONJT;&(v{d(WIWE9-WNXJO=+rKT#Aw3{pB#pB0)dv(YW26f| z+t-mL-y>H!UO!&WNQIRC`dOt|Ov(7D?Lo=W3O!_ex-9>#47}gadccd**|1uz{PWrL zk1sJJ+zzp0S+5ELGQ*OCfg0{O@87Q<%bS=0A(X;auEoIiw=KM5EAkhpA{UR0U8Qi* zi!!qKEFcWaC3@foas&=eE)Vb}GmuM4L#4KaqX+y3?<;{Z;alFq(FUU<<{S~d%k4$X;@h?m zeZKOZjXuS5uY%RbPaIN4&cvKX^+axN4!x07Ma25nYk!9ixL%mJ_N{jQ_Tdh0oKV=! z!>Bp9#V)4_x_1LoW{Q#n{-XZ2ISGIj{nP5^ElIT;0qG0Y%Fp=-(xd+m| z4Xi=IB$e}G3D3g_k0r~r+9+sx|3=^yLWI8KYlZCNx2}DIKDX|fwyRlBdD^tbRJ(P3 z`$`e6Ap>o*qAAB^uGP(K5%}#tEqZrHrR6UDlb;nl=V>IYmKDomJd!5g&1wQ*qYjt{ zPjf+hDsfo;NcaQPXaet6-l$0Up+i)+6mnlVjg*g*J3TV4JeE7h#O5v58zhntr)#cu z&Dx?>!9-B?b#OcM6ctOZY3MDC1}yic)vIctlf`Oyc#nuxAEdAc4KWMGQ2F%+gXvcj&rSIb zgZm}b{zzf)rUPzv97+_x`ZkiF+?%h%=u)&zHR`JKz}*h-w{R0WI8 z`S_~t#sN1Y zR=v`G5{stsgn<_p-`s342^MK={R;FC-}H%RE{-y3y!2I46m08o6%9qcd=- zvQ*xe0Nw=19}6_&r%$1rccYJq!&2)jt;GjQ}IB6mI1WaA487;Ah^ zfwq0xI53%IxpJ6Hx5jG%ZwJu~hcV6k>A)$9s_;QV!*Sb4uxMZAvS9m#y!&E~2K1&n z=#;mp>41=e>eGo1lt>UjCE&!EVTICx%PJHr;E)*;RY_i z1qs<6`I4cz;Ay_DyTo%E@3h0g9nQ}zlAu&jU9Eh&>@8WN-@X1yRp9x%;`q5dd3TR} z=75@N>JJT|or%c57D_T>Ew1q)?oZC@!m}cPhJrBS#g7)e=&0-n>8RyDfw^Xvlm9C+%wey#!yuG-f)%X=eih#~$hz=x2_;lr=ApT3&12i!uc zJZmPZEYWeui5MR!a+?2icJsXjc2l`o&v_TuvCvJffKX}%4zzGB|GxNvLALIu-J=r& zA;(GwGK-^_RRnTu$KaS2x0sj#A$t@YTIW7JL8(73C)Y2B38I}WZ&lvg z^q~6?beg$#x5@P`blN35T49AxX73N{##%VgNb#@9jSqHLiwMsEAQa{E#Gu)DB8z!h zF-d=qzt#ugz!X)Bc)Rc02Y{@X-pAyu>Naxap7?J#2r8T*Aj0hRteT7{dt1qb8o#Hk z+Ie3-NBZ1ei%);(OL)uvArB+82AAzR&;&VI9yH$e!Z``ghAvn;k!ySPhyuNQla8V@z}79tEg za$kk?;M??1bwbM&Rpanq(_m)$R|70LO_r&g751k{3qc;UA3!X0&(p)|Q`!NP8&t+S zm$AX}bo{m>Y(hdWxSUJ zng7ef0#@SaErtN0Ce_GsKNkJ{3bfiK`WWV@5f4N+777(3o0Koo*E4jPOGm{X%GEBR`Pt<)rO~GEExQ z*KSXei^Hx?+L9|c6LaJCDUOnbiaTb@kW+Y8PlY{}^KZXR>(oI;({tf@ub<0XG};D7 zgJ^|m8afi2<2rLJWvYG!J>8jUZDwt-6%9}eIi|XUTw#0zYXFDhg~siXEIO*Laqn9H zl|c6<&@bu*#*v|L#G9P=-4|N;{1SMDBq$Nk?Ug}D^IE9m`J{AMyO%7C$s0R?UxlVm ztb%Fc1}x1*DrQ)FLeIoRrd3>>obDd1`2)wnq^`w-@=j+0nO@%Vt7>A-C|8Zw*h zv4CYVHZo`GP>NRvhX+ILHrem4vj&ogeKHtUmuEcCexHjjgHsUz>cf=WGQ412tkF8v zm%O@i0Pal4Ra#gs;>q>#z1`|3FVnbuPnGZmiO=}@O%K=wwqa~lbe7K#@IRmn5z)jflW?DrEmIFv;LAXHUP2lOL zbFF}Y}X3j03rZngu1owWGv7r&(r)KY!`Ser?V@>IYSCE3{6eE~Cut>q244N%)WjsX->+u^kH z#1j|I1oe}Pgu8!Q|g2X@Xsw8_m%ICP81F|c!ToRS*DmmIPWkA?@ zK!pfTyDbW*pA|EnB8d3T#;?VI79cd-=U;8utJbq?A{Y^8|9q(=@ z3-uQ;7{g`#Zh^+eH*oTJ*NIL$$Ej;IK)^xnMA89fil9-X#t@WHh5JTx2D8`I6&QSeuss9NP#Y3Qk6{K z2S@=S9h*{5b8$xD47`PBC6mRTE>tNdqT=sWtVQJi^Mb$jcO=vG>7cS%6c!`1xd-?1 z0JlGfx~y>aviaCrH#f;z7c)TVQmwk|E=HxQT)^h#Kx*^VwaM-s;Ibal9`h3HNX%Ew>zO=a~; z{DhB`W7*F604htVQNvjB)!H2}Qic$d@jEYxXnmK{09j|7Lw@|8k?skRmX$KnWcvPj z8BmkvX>%C+uQOmge2l=}tO^WZ#KO`(F-@BLB_^X&LCaey0m$98N$n={BG~lCz<3Rx z&|?5|a@QuO($P}@!}I5Fc+pp90fl+ZY15r3XMA}G>F{G!%YJr+&f!Cq%b+)AOno=U zb0qh6%H7WTdsN0Zh(Zr2j3buqw`;>=l|KqF@>ITP);=SRO)Cf18>M=0ucHPKcxo1a zde8W(50+M6ODka9#K0Dm#Lf_^BUGvmb*qED>#Gi4QmbM&N#TM4R3_V$oO??4&QV{`1a;@lp-j z)@6jwb)tQ+V4;ca7>Xbli$i@nrw|g;Utl7lcffOID|#zNO<(j-m{c%L!x|_Cgg?bK z4lnC_g=IWRI>W-nA+Xe}XPlQL+=0Bq_ghJ7Le>g!QoUArShm{wmzLjMrAF`$MJy#Z z0BeG{g>TC;tEL9yOI{-Yn?*SxB%WG)My^`GG6j(7MwlCPMMf#I2x`*c_g~ruemIpE zJ^RLl23T{yQz-P*6uf@>0Y5FxT)$mnE}2Hd9uLGkS&o0DVn@{Vi!WM5}+`k~vIKz|{x+&6Y^zN1Q; zI;b6xkd7Fk#Kgn|;1$D$97j51fW~lsVuvv4x+r>)HVsup;m9i9fGdH}ZYpFUx zvJ7H-v`FC$sv<$P-itkuvOk4`fZ$KKNAp$gskGy3I`d%tt&Nww0%!tOTTZC_UyxDl z-ep=UA5F4frTlt35Hp6Zb?bGEcd>_B>(EeB1CY#sXGD>L_tpr*F$=-*mVA!kBWkr8 zyrkhL^92Ii*#M&PYdN_@%zH8^)8;MNU4*_V-Y}9r9pExp$)*d+0}bt zQ_bvU`XXLDCRlNl=m7aJTObyw1YeLs@(TmTM*B#3HEMegG>Ws(G6RHMuh)g%Cx5B? zd+!Vi&GN5aM#JARR4lnK(yYns35;VJ7vT4Lmc)Md0Snkv;F7;&6XF#F+X-}AvEdZ@ z?<_0w#-w!BC0wtPFFs}k59`~L+K;4A!j>@{Z&4M$GV8YLAJc^Xq9uz5Y6xDi^xwkk zf4;0>V$Luc0mH~&NmYKt31WcZl{BHPUleXED#6Oj3m(hTnr!tgH<&h8B1*X$9QDLg zmrwz;1V9-QMaRhri!M#=f{sVw(|!O<5BUHb?qF9}TRR+>+2x2o_Xqwbh}Vwl=Bns9 z^NG?ww6=W3DZ#0A6pCgTAZt4c0&8&?(NwGGM0;Bsr#{4Sl3`aKAQOT(amAb{PUdYo zce9Nam!w8WQ1vYP=ZFG0W!H4Slw}`%Cbc@wu|W5i#%<^ZG&yd}Pca zqQxHhzen_vm)E!ze6NIfQ z`MNb_#EBUfzYCMFXv%4#Q;hsb@=YP!A2jV3+8x|DCcW7u?Z}dG%5MEEL2w@WE!MLL z@EjMigD-sY6-0k8iLP?4SXs}*$T;1~)5=@l0C+Svz+|$~@$|q0PhhDr@!kI1xSe98 zI&{KO-q?Mx`COPrJbm&d@HbGU&a+E&rI0#GOhf!DW?x;tK#{fZE-n93y#o?3W%DG* zDo4ZpG{Ur!LTX&u32D{@6+#k{QDB*N0%Mx4@{?OLWx77SU<~hoV1x3#S#%3)GH}{S zI>xLdE66>H@0Ri6l|FZ3;9xRa=ho2W>To;I!uYq2K-S0Ry=Jy?bz|oZh@$(w*z+i* zPk4&=-H;Z;hjNKWG*}a^{&i*RS58x92KCfEJ0{`|imBgkQxQ1Ljt=0rgyaQfTdD~# z-Gn0}!PLx$(?7An@lyJeI`j?}-;fM|2UNX|G=PYnoGx#ukmhb~(}%Sm%eqpao!J*2 zjzZ-*ts!#zJHNpNq)iQ^0`5!(Ni|r05JiJyU5<4?g^@DrI;!t;P6={`%z;y0oU~t6 zpHYfjF%8#BdDo<~iMy;%n=}_A-`z&h%tc-g@PWfb{uq{+Xh>iS=xNXmH3dOSAMdBI z;%%b@cw1b|a#X%hcNZ)~I;8xh%?*wV(4Q*my#3o*prq?l2Z_gVLcJ8RJ_dR^|Mnkf z^|7mlkU8DNgEL(%O+vQ0fN7OK9jVC8}oH$QeRsD+eS2<$r6-l-IZ>Ft3y zxkP|NthtVamyvk^_L)-yAiG@URN8OI87X{pKrqG!|HWR|&R0NW$wa^*10S9*a5vje zB8gn;udRh0xL;BrHVvF!KG``Kbf%6wDJ%$o?-1Gv=)9A-M%!+L z!95~(t?GIu`$SHQ3qM2SDdt7xQ?PabbWjKr&$usxx~g_ZJBcpekE!32ot5U7kMSP> z6HXTLJ{id5g##K6jibP&Mzq4@FIZY?u(O8nKPERweqJi|A@AH%dS46C{+q4& zS=Hfu^^;!uOSR@}*6$KbR&G~y$6syY6V-*DoPf!`h~nS7x%|5(aIB)oIo{7@SIQd z-1wP<2-ct>8}+pycprt%auUz8OmlxdlL2)aW;~>-7f(52O?1@yja9ksl~HAMM)Ox|39lS=n@!tF6+eaMH{ExPyA5u$|y-QWN3Js!|{=0Ybfp~ZXuPSdaUdek>-wui-ac*!+N%{Y^M$swDC*E;CzoR z&ZsNki_H#N2qG07!2`+G+gta^gUH#$dx&;>qKr;iIYKJ8@>F7Q?8Hd4CD|MQBlOl_ zf2$!>dF=-N+ae8NAXudO?MF_)WUA6{2tFLBjz^WX=!lQh8CekhE!KphNlbk5eBymcK)uODUjb#~vZ`%%0 zKqI7lXD<6@ojXz?TS2{{+2~u2DA1;E^@m(y)raMZ?06sZGH-4#MrJXdf3$2921c}K zDFQ_@6qJae>sHhLt3RE$E@g^_@L~o;MC3@G_l(O?RE8VGw+Y|gHIYwH=MQu<_w79} z2o}41Wmt!a?tGtB9=XrAqx>WlJvUKy>NU%$G&L?Z3rvKV*@TIa$d~wwc#nfoAu$iT zv~+BkMKG2cfh?aq+|n7M>^rnOz6PFrvueK=0KnMGsH*!&&nxGD=j^JvZfpT&=K*D~ z|EWbDQSgT^u`KQt5-yk*k5W0yH)pLh30WAhW*SAp@u*{LEY!S31g;N=hFJ)L6ga&e z2(tUJsM%Ztqew&@)rDx-n%gfQLx1cMn*mvQ(@nzEwDl-kqha0fzmo4%l3qjFmXm;L z{eU2Kncamrp-sVY*H`){Blx_+LT!AWO{7X2uqcrX4Rs>9eTIx8F~6;S(ghWsfxYz6 zl%7t}d3~z`#yItGz-yV06S;sdL%P&ssJ7-av`aG-M$`}XxU>T`*nw}yesBMGb3PDf ze#e7{@g{+b=JySOYp0Hv)cS^FgAt(7O`f&1a-bWiDFu_C3#eWSQ(@EsKd%)MEZ2l!E0l}&oNKL50hCQ(L@ zSk&w^dIVmr?mk%=@cxZ*@wD7IFX==$1RVA!bhT`H{5!cQ!b1JlfjygT>HRF^Ni+-E6)Q5-vI-ml_EB5#-DL zRio0Zef3@&ym{5x%;PvOlOkBU5mS{wkn@HzV_W&TIsD@j!yh=fZL=zsr1>_0`kZ{< zU3kt>FKr|%4E~1?A%W=mXW|}}*mWQiH+nYl#ed6TUO@h*9sj1^4pxp8>Gs`O10uOa z&yI7VB{VZjM_SG@=ae_+Z%O?U*u#M`SwF#iw_zs0g5UJ!9nrd$r*v6}#Yf42QpJ!^ z`xLCN;-um%XL>-#=A1Y`%Pwup)dBX$~E2W{EFZVu(FfWcv%hQeJ!3 zTvhq!1LNp_I|ryRPfifYFz(sM9YG`eZxR*@NKOE8d{;4}$==(QP*_!hvAsCZjALR$ zk~D?KV$U;$c`JN+rXz%_n(v1KFj4j)CAPi8%n~M@3AgdHNV%N&CirhxB>IXV_|o_w zm``YtgfxDm9_9^GXZ$2 zBABUOq6u|B=M5sl`xe&vylfE5wvK1^F~j8$A zDLe}Wu>$Y~>s9j^k(7E%X(UYS8Uyp?JE!x?ieUKv9r{<&M$rP`t>1cAj z8d!r(L-F|^@XsJMi@#i8;s7H@CWwx8+E->!3kuY~l8WE8 zAqyO80!k{biEoMVE4$sFfIX!eeXc=E`m@9@w1d$mKmpq2J8Aw0yR&*XBsro}NR+ul z?5D?dFC^uI1h5Amf=4_XB)TMLj^i~U-Dl%lU)#_Wv(P~hG4bSdv_4t@dT@@X@~+Ix zB}_!GVn$S6e%^?+GEe_>zak_%uzDNWB=u#v4#yvb!n+InblvZ0w=7|322bHk)HdVw z-9J>xi}v1S7yaMG(RZg_OR`948-zHqL6#FpOIO{*B{~?x0%^A>>FVVPxhV8HwPLwk z-6$pour|-_>YeL|8DmMs|9faKigZI7_+ww;q1x$!vILdT#+antKAeZb3j*nu4w3O& zl!!le4cT6C64^^Yi=WO^ZJ9Ql|@P| z_(SJ!m;0M45_vr?qlL8#`B(t-WOc#bLZua@PWq3d1ss{8uCh=j^b8uTT5+R#_Yj4! zkU_{_B1DX-KFKcBq9-c!6`}QqY=2ryIM)A?nduV%Z#XtD_DUL&7+jsVYTDY z;r&6;X=jR5A7-i}sxj>ME`9)Bpc+=y@Zr@rq`CJS8wU)BA?VNyV&|*LZMNF-cz6Xs z(@wxqzjuQ$VT(aPWIqrbfv*zuh|7B=RHZ?RtnoKQw-}*~7STfdH2(78*PB%Tj}wt+ zb(W4SYN>9qLW1(`M%GK~sePO#ROe?}R;5FUmO5oxTG7!6w|{HWIM|f+8AcVmJ{wWI z(MZlRCwGDpKRr@jKX`*br{4c9{NcDt6CGC@GN}w+qfq)tGxOTKpX1U+)Kg*%7%u}m z4*W104JEee+AwgFWiO%zYEWA*#)&f@@x*=%nT9xemg2^f>Koyf%1;+8vNb!;^8|58 z45A>6h83puk`h_sV(*DrzH7l-VcO4)JB#OnWy7k|jE(h8zf_zz?mZLZL$=n&rW1TN z_~q2A=c@kRWrhNN(@DnaM74oj&s_^t9%+HMDXZJ?N*~n^?p;dT#$h@*^(Rqz-qhW0p*wbG&eJXY)d4U*joLu2BF^Kz)eX&l*|%gQo|GK+yQ^yo6@5Mq z%o^}#?MDlrbdR!;lzCJ*4a{12`WO?(4_lFp@u4XW6$dTVFXm#IvheszRnF$NboPLi zM6?%Bk&0++tLpM?>iT;db!TkqDoG}^4`TJtw7LL1DHU3vfDi zgpPvGuV=PC1nM18S<&y-!?nAFQXMH5i_$Sm24QL6^cYtRhXsu&!ef~^rK-R14rglE zQHu?}CY`HoI^O|uis1DobE>EHUSa1}4R|;&c>p+zl+JFaM;s$d_83cPCboEU`l_To zC{wFh$VU#4NNWbwQGi`qlLj`I$o7Cbay_YV#bR`Dll18>I4;|Ces6GEdwm8n`cPR*z=WY7&wunO;%8wH z6YYRv&@J7?Yj}NNvWAVeQP6SpKsI5UI+t5rkxEXyT{VL1C4mty5aoyW!Ghr->+GvOT`)P%3?f6$~PR| z&&A+)^U87E0>$ZifjiKV#)VYjUcl6?dbj|>BZd2HYMbBFxd!5u81^S#uOd<*4c<`J zQ9JVEFkQG{0d82nkNM}J$O6h-ftvloz5z+CDJ@;Tx#?^As)$IW(djtI6!?gtJ!L)? zi#Qz38;vUTz#zjDR8>WOB8^B?GwFPTt^bsDVbC$R3yQDS&XIyC%FF0@aO9Y({V|F} z{%_q*`#kc7(90~j^+-+5av!^w8XP8=8uD>_@Ok>e0X-yX9GAPs@71ZThXY3`^cZRS znA^w&=ya18K~FI-JpF5t$6kYeaZi>d<)q5PZZQh zwJ))O92kQ*Ht12LL~kq;6&lETF@h!q$-Iv6OE(UkO@3DELLU-Z{v;ZH{ag7IDj}eN zwum_Ycbn11S(vD(tOt^n-%kSJO-8o$95<-5oCYXFx(copTS8QbR% zQ^xY^h-qb$A~(wn%nz5`5xP-S49i~6cjfzPDpxHO_)|jfRhGmd+g#e1^{6&-(GI0edQ3c5`w9W=;6{ahmi1g2M9cd9LgBCOvdWQxH z8B;vaS$;AzbQevOB_I!aoWv&8EN^E}8}Nj-;d#wV*vI>w{neaLUNTlguYE)mE-rsXR1n?Om8JqQ&w>9F&|J}QkZ zh-7NF3*9O5{w#-X|7N%B+P)<9HB@tPUE8}nwT(Q^_p-(kpFk@JIHgi8OWTa9QZa9z zzx@l%68f{d93LQgb!g>&9#B0F`!&)ZSfkiabSsE6T(sIv^xOa9%I(^cOrq-`lzD_s zShPqwkZN_na?}PIZ&r^FIDl6DX}!;%#_p;8KR#Vu(17%u7OVt}s<1E~h#^~L9C;l7 zn{axfo3TPe_vxh?t4uDo87cFT(mZ=;6@KkH)nU`dVX3BJGOE?5y%KuBaePEp%>I~~ zV0CeT8b@KRg46#6^V=#{9KZ47cON=$M5BRx#rqoEtrpQCtw@&hRP*_ZnPJa*NfHWBv+E z#!;_KI?l)ii{x|5`?n)m9FBZ;K@g}wI}<|g`f1?RhWOj9f@{=!Li{HWEo8PLi5cN3 z6T4vSQ;^SlnfF(N-aJDMmqCq=66j55i%Cz2Gl4jzfoA@8T7WsJoC(6IIdZLeRBu9=UDx` zk5F=!_6PgGWhjYQp`$B5JRBiQ+^EY~Hk8CP{)^Td850xf|JAHn-Fk+M3a*I~#KnehJZ3_$I(#D53w>u2959g+}15(FUx<=gw)3>2%bTi1z5?aYK%6AIsV8X7x+P z4=ML1v)=Ij?w$~AX^@j0eFd*pRIC3R(R_7?EII4KSB^WEl-Z19f)~C>v;;&>1NwnM z8(6p^=0i#f(9Em60^oe7>BcBceY45a^iirNwi#C2;Pfn^ZU*uToT5LQV9oraU)Mhz z@_G)Y#&_scQn2RA_w7~TM=)mR0>2Yi0Buk#n-gmXPb)n%%u*exm==g34!MZJxg9jL z?9+27R9m{vHvzK^V6HJt4<-urI2)#qMs^X=$hWw@g#WWalH@8_eL_WR(FUNFRpiDx6F@iAUYi;pdRA0(B;-d4Rd7@KLI^ev4U^xkzT%!!l1%IKiy==&Xn?+yGvzSwwQFj^ytxo!hedQz(*K`u&S zp1q(#*QDQH|IGYdoN@7)$wxk0kqE;W}TVcow6wwg2im1>6EX?i%J-2Sv-O!c(K zH2YNJ@#i#683*3(>vyV6fBSUx`;I>|NP;>X&OkS9#D^|x$79LoYuPc0K`vpLa<|x= zlg`RR!@l4Xk&f5x7pFVoDR#)BO@$XKOQLDY;TBDS{3umajCYbLFKuqlFFbaG4dedr zVFv1>6{B*?Iz-V~bl*DlF~Ta5p`|MdBe8WOiY5H?IEG0YueFl{?i)}vxS-V3g-OeN z5>buQ&Q0n8^%|?@3iMg3U}+draJb;eobI_)-PahMoaK+4OegjEY$Q6~Uk|ekesv{Q zKI7Mj6G={v9J~|6(f|(8(=Y}!g+pR;Qy2vy%iniGX(nHLJD3srbc_$yQLJud#mR@` zErAHgmB_@+KPpZjub*<@MeP>p&I4Uz5sY}my}tx4%|Gh0mgNr!e8)HgVBkKYe4OC5 z4ezYb-EL7@kR=}_ra-_8!tAhl1QmTV`7y5NXUn$!j~QK5ctx#Ejm!ae3#xj?6RARF z&YZTuA$&l%kO4Fc(T#CEnvKO^v6--#KtT%5JD#Ljp3eg+R*q*Jqk3BFJ`^hx5M3+s zGvMMt$EQkzk7OY>jQGx!b&G4b7TBLMwzC*Q57YGSZyIbZY%VYSmLuj(bW0Po_VofB z)zlsTW~AzIHWVFDK*fE2OfhRpL{4=iH(zpNsb^+D3wpWq#oS-&j+urzaeT>FnzOjD z6~cGQx5*Jy!Z8@bPL)OpA%Dllo;L^WevZbNysRG|Dw|*WOS;O;L{_$naO6I%RJ~Qb zyAPl6iLvx+T^o7lvl;2jsQFFIZ1?}e(^&?^6>Vu0cc*cOV1eN765JsKcMt9m+@Wy^ z9$b<@aCdhaOK^9ByTe@Go0%$ZbrnVR5Bl~w=kEROwN7&VtY{8+`IRT7t!@l%ien>| zjw1DG&L-#}9lSvIl@Zc~LvF>>5s}r_neY4gIp;#;JftC)fQoMs!N(O1obV#}4&tvx z-s{@!YAjFq5M8>RXp0U?n!@DV3vp$H@B`@>Bg0atQ5DkNWD$`zF0yW^&B|||{j2m~ z`HPtYRt@!uZ+u*L=f9BC?x26DM>-Dw?|BMjp*3l1_5SATO23T&skM*5l4X%&ecC#sX(7n?fj?fu9KnxIn-}cc@i6m75rT zQxyG_hls++ha90UH>g#`AMQ5y25Pg=H6dG#Vg{Vs{cxsGzV+6qV*tH(#<3qx?iyGN z9AcmpANX}9dw=WIDry5J-MjuoF9LM3vDSJ~SP4z!O`rqxc3p^AId=#Ry3|qvd!$fU z`tVzDpE2~Ij2UB#I_>*-qU>01Yp9^OU3xhEJ_fCjv$ePlK;L}Q7(?<-+*dA6kpt{tt=io zFseh+Zqpvz3aH(BI&V|&6MwXbNG}YQR{h?W!?YOJ21z%tw=!9;fD|8xGd%$ohx)DqqqZ-P-v6xQ4&Gzsko!7@~Xj@H?+8 z&>MctUb7Bvw%8&CKQ4LuGi2^$ql7Flv@JeFT#ch{XFptzyZCP@%u));#F^xw#qT}t zA6BC?s=LAtczUBBzkuQSR@MasU4swbTMx30x$f zGDvXh93{FDf=xyyebNdiL-j771dJPQD*2V(PEM(-#a&!>={-b#*z(+PZiyL6QIkSH zFu1^2=+Y8#`Qe89&gO+MR0iKvEF!~eBiPnyIN!0*ImEntX%>ao|BB1Od)HbqYwPw6 zWR%i^O3yPx0LxiXbFd@XbKJMweL5f2i$zqa4YA$*>j5)9TELWw8GQ&~Ap4aArP*u6 zqMrSjF~x$Q70!Sgu{~YR*V6n0){;AHf)LrfJk_@RBe97a-?0feY!jmj6;?mc+LLAS zLc5i(tKW8YZ)8~r|9%~KN=N29;%z?bU}CVd7(Vnn=CdV!`p*8l zUyBf)5*{&Z|J?Bh@wXdx0!v_ddXVgP9Y|&1agP_>{ z2V9ZzJu-5AULU+4Y4<039y~vE5oRmifx`t}3C1!)*fa&Jy(_&*GYb4$W)jSW7gRw0 z5KenAgxhK{%gofW$WZbgo-Ewg`#a2cQ(>Fd(7o_IhJdoCLN_1jdWrxz%P}aIpPoss zc5rDHQYqPbd3Q-V4GWH{W4#?g;u<9WSQIa~#w!F6p^wnP!6Au>hQaro8r9|(ZYd5~ zvEdt|QyeD3U)_|Y_b~V)qwpnbWuM>|yV00k{m{XY`){1Zd{Do9U7{X2VjZ?%Gf+hG zp{gaU;po4C!aSIeBut>|kgZjV3p@bKntw)9b&F4Xl6FhLVar7oE)*1=S|0Ih z$$QkE(RhsC5d9cav5T-`;YJ;=8u#vR`4F7RLwl;d?hUVfhCGk0_GHp&%N7!E8fRqI zJvmMbRTZAfXQywct*qH~C;l^oWKbqMVu~vX6X5~uCE^X)HlnqUDamn+lCi^;a|+>LO;eoRETk5+_SYw|uRPm=X|O?S(9Xx0(uEEY#v13UW3r?{IwTk}>eYPy zu?V_@jPj&8J+jHlx3_CSqL#N^tf=t=Njxeytlwat{_s|l z1;uI2BRNSMbM|WLs3oYgH;GN3VI2^t)T1&24A|fpR00pec9wXhOlX6aF7xI^t#PPY zqNg)hr|rbgRqi|Ay4>fW_lC~c-5AcsI2-ieR8FV|D5EIO)!@cxLyt+l}d)!5qVd^#S#>pPl06a(_~lTX7(~(96a^lAT50Ze!ol)TejS$L_mC37|DtQ!lxOJ1X@@9t~?UPEYMDbEAbyy_jmM zuh#7Q=nGZLN`+O%mgn^y5_qg&>~{>;?+c;q1x2>+|K&5J>tO7-G7y^~up=y}TK6{- zSkCSECI?tb()cD-2T7E-2QK_Zi1gV<2$|d)N{L@OXfXfmih!dmLNb5BS<~Nck~`?i zXH}8S(R`dpGpN}52~w+1k>y`~dP%us5cF0Hc>6ix+!TfjD>O`oyCR*&po;Y|R)UD9 zXL=D@ItU;%J8ypE5thi#JsBE`?S(nltEnw7tY-YppKABdRCEAa(>}!yHw|4}t=gbY zw-p~?F$uk;UH`q$w%+8NBgzn#+ExTS=@$fmS|N0IL zM9K>K#R5D7*tfhm`PY3ao60DL#o9t3WdmSk!Ni-GDUgqG3v?kpH3Yx0vQSSHa_W1IuKt?Ra5ah` z#cRU)(Fbq4asAGXq49Z%Dg62pHHzVf2lUruAk%RocY!6NPD|(U>+>RO8~cV^PFLNv zY^K)iEgtZT8mOhnr~OTGiMCE{$Rb}-FYy=gI)bH8FB6kVLSwTet1bEk+%UhwP5m^T zx5M+{F&n6;#rO;O%y)9WGd8@aL@wTZcnDvONbO(CcOC}Sq1~g$g8ln~#8vX-AU>vn ziZDq#`WMT%2xK#&rLuIMh4nHE{F9i*1A3O$^83(SC$V{i^hkwugj4Jyb^j6q=WisZ zEW4cQjyurUPN%xU*<%h zd3Ip-f2~jQFgD#wd->1 zvgfk*vd^V@19>)L{Rt(k?u38m+$ByUM1Y3ZA89g}s5h>9p_=%rdhi6RPC=!M7lF=ry{8<)x$ZVNat$u7$epvk2T_bK=Tmwhh>}l5~0o!pctMi`0 z4e6=};@MwYPF1!;wPwSBgsl$jHMi<#7V>vpT4p<9dx4OU*5l|}RX|ez;QyaOFTg=o z@H@B15C!c|V#ux_jAbD*>ypQ|QL?0PNtnnHPS5;aOG>1OS5`_hQOrG!$<@_mz|dfE zk}YNQx;P@4>|odYOiPFJj%n1t<89H7Fpoe8@VhUna=&=D{#b9aaDy>ka^7p99yM^r$P7+cvup=<8jE4$MPg zJYbK-O`E6BZ-fM;4lX4bJb8XhiQ(mFV)jSVzGkV~SF0vVh1oD=Wit-e@#y`hCmyjl z=|9wr1O}xY8K;EKNxov#I&G!Xvz6TP^@UpH`)z$|vcAi@J004?CQ6N{qHCXQ(UZA4 z)C-mD!~3_QH^xf8p|Vm8qU~UNs=YI|TgVJFJSNyo-69|SiTE6tKBVvVr2#*U^o6QI z^qa6)1$kHJU+Q4S}V2xea!Dmng{f#f=D?44Ei1`8!jy@QVM9^gF zpK@gUa8JEVd!V55SDdsM;^>(O--y>Zf5f>Me`JHD3JG0luHE1NqHy?v0~QMsmC7lT zX;1e{>Cx&$B+L$2Gcm-K{~*1?!w~_bVXNfa|0i|UKj)mNt={i!vxgMo17PCh8Q^0lUi2_d+u$nHjfse@x-oS7LoRf}ZCt6ew?>JI}T$No`%(wKrI z1`z#A(DH(c$?>1H9nxLnjZpZZ%x4sET6^?nFwQ+BL=|gUR|F_?YW9Hfn9hKNAAqHL zv0m#C29R}MvLm^|H;;mjj_d$t7bf4$yxG)G4SJg+Sm%B`E#s5rX6ZKr+J81Tyda3= zQiIr9hghozkNi`-XrN}<(V@wPSo^~Tp?YYmN)zU*;Db${j1?6>C$v`EHA+XO2TX8C zG<(&vTbtj4)c(9NPY-q9+|SEo=C`-R*1v8-+`nux05M`p(_dVZdoD%QWXgVG%3Ht| z>k;SaS*5ii(`I33aAoh2=Ugv@vNg)r4x+{66HjlBzDgXsOBNbdX4s|W7sM-mKfhA2 z{lcQn5ddRVudprJtOc$)Do~guzTchhU32h)rEZ&FJHh$1N4yKmx)=xW+&Agu;n9lZcA)+3VrqpK} z5!3cHTwry35FVZ!$uk%j?Vl=l7XpkpcL1k#{Gng;^(o}>q!n{`c(@gfTt!~K2N0Eq z08k%d8FKG;z=$Dd8^G9nQ!SDEo{2vJgYiStkvRgNbhtJ!gEvx`3Ra1MzOj$O1PX~@y>Te7Cx#yM6%@T9uGec$^K%r2xp{6=fa)kA zEn#l|BPdn6z-Khf>%4%;`Qt>c^JsPa=0jnwgwqeq1@U9-KaVDN0tVq#Lr!i9CL%*R z{d-C7&0s8b?g-rNA4~Cw5AAhv@MBTJAA~uo_X6)>j)vmZa6mElBoCWue+a7Ctz+6I zSbene5)Oyx{c7WN%JsE#5@?C}=l%eZI=7;03AQ*5d&jOLXv!$jWH7TsZ!4QoWUe1j zCk)*PDg;kH9FT-mS({-#C0%gMN>m8Ozh|_qTmJ$uuLm~n0I=a3`1_4o{KKeIMC4(g zM(BP!xfM@o%7jSNzs*@Oc-ukrq-l%I>&%$hsDrd|^(#{|J>$BxjR0o?C5Hr};WFUs_= z`X*U$o!5pvw}KNz^DHg}c+1{afD}Y479#tmvLj_hB1gl@jcA)EvX{;v&YqvylaFNB zKLhf0&P9n0l%1SNi$2- zU*Uy$a|&8+;2%ZVqRv_jv%iVA73oia-7`_`)CbAU)Pg7hpe~-l2P2Gf{even*7+9A zmhlXaYOSB-+8k%V^t>2gv~~vg2_#dW02`+hpa9LPj~UP6B@x%e(4b1|1tc@)faS_; zdnmE#tOs@cVYn*@*7>idR=(p}$KCVo8heP%!u0+^l{UqN;7P zj9t9&e48iBP&fiA5e`^%Q;uquW;_<3OK#w^)J{H5ZcvU@Y51&fZSP@;Z@P|MtJTaF z$zX$0bwq2s8SH?tJ-05z%aYg;A{hsMjz{EiQ$_v*FA@HCx=*`*@Msj)>18-X5 z1Kltz(ulfeq0NzG231S$Vy3E0gRnE?pUP3x)gV^E4Xz_3-IQ>y81}D+k-Bfzu7kVt z3D)P_)GI22xfNPQvJ$$d7Rm6ykO%wiPx~^8G+pS&99+T%`xd7+e}#es>=@|#I2w(m zpZJy_{+t5XHKheJlSOj4Z?v5zK&lRqZFs5Qy@N9wjO%anzKp5*B?O?hnL}o7wd=p& zz2Wzk0nUP@;MJrgu*wId?4%Nf03HNQ6y2hL-n8y#X{FQjfcztu$c>%NmmIW5;-4Uq1m$Th%e&{rEIZDDKwO2m4R1e6`wt zDdoW)oe87+_MHrex>8kqBx5487jSyP-*_$k-xqKMh@aAln#itQGbAY3p6DCD_#V(5 z>t<0g%C$5piZDr--&IT)X){zN3V}v`cVsMz#=idGVQ9(}@e=^fsg}Qdun^*s$%VhB zKhKmXCG)TtxwX0M$OA+47?R;=;=u2|x3is^t@A6OT%?1$t_e2%a}72ihyR``ZYLYg zd3;(Le|^WPKEM%$Rxdqm`;}cS8%|C{pBbCtEc@^?9iM||So>3-i)ECl=UB3{O(E&N z6|0LrWC=Xo_JILU(AlxvlU0@IHiq>Vqc0f*R>0s(C>;8x#TW)I39a%`0$12 zu}*6Qjaw7&k~?H-vP$(mb;L9M)y!O+@a(M|wQQWS)aMHK8@Z#G&ZePJWVt0p9i_3n(rY^XM$@zCr@fvdbB7kahuauaIZ3huv$`kr=U*hqYmfK&ScoG=JNx}t% zAt1|g*fgEV>a_$fMB%y%HJXZO*1vn>buw6Xw)1BTi&Q5j6k!Pq#~{>b4}Gw3Z8h*d zYNguJfWdOYa))~l7DhJgkt28B|0ow2y-=^Hepo45v+9B*iY;lF3zAv0%U5)Mb}3&I zORSe*Y5M6R`2x-^HZr>`&#g9jv187zM%JLiGue=!n6e|8=m2+{g=p-|K)R0?(Mge_j`8eK*}VJ-F+-W`u2}ThVaMPEk5C`@%_^w4&znTpzdp*Zo5&3Vh#Hf=NbFTkyF<#hG+_3fx6WaO*dT-M<2l^D zsan+=S2G4<%)ShcG>#r%cJ*vTP{i%;TU%A)SbvVb9X-!E#u49%-9%1hG5E3=evCt` zE9|~`5ILOt&0oWJ&B@B@#M1o55B7q0qStlpdjw)f)3CP*B4UGXIA&!#=(jFDb;-nE zB*AsXy>GvQ=4>ks+ZCImaFzMG7s`&%(Pr-eSH)OrVqy>=yv$;_GU*MM%W91^6w!ZY zQZ+PZ6kz}0O1@TUB&zX{jY^Fdgv!L7UiB-Q|4U&9qb($+H3R`V1aYat zegw48@c^Hp5Rf!O5GK>vN~Gicnj2!AT=g!Mh$gE?{(QTxzYj?L9D~OJYr{qxjE3$R;1c3p_^nM>1sM4tOoDHIjj(F(I&hF$iww-~cj#9eT{KYtU_#$czH0R-5-H9nOiLVg)g?;||dnLYDvox^v` znUb)Pc`Vw;U<+Rc@z)o4>v=XA!^-v^e3tJT#^UF9tetlW*~DSS3vy92=)1A7^L^&^P-Y?x!*_PzI6JySX1S;wPAv_2z&fc zYPf1U5{gF0FiRt3iG+B#ealbjy}S#FWNh0zW%_`g z-jQ0b#wS((RI)l`mY!ZvYh=tPlAu|=ABF)w7CdVq) zLRN_nim><8brej0c-oMDoE27$M3!x@9W|ya#m;48!=g;bM-(f6)P``t5jaRXe8$HA zP#KYOw2Kt&2p%gyxVt`O*aq>6*;s=R$k|Bt(ETvQMVdE5qR*Csd(Ri@G!pbEpADT^ z!ZDe8BC!)r4es|6_QFX;OzD`+GrEfJO3&*nAaT1ZH%PoxnSuC=3>st4Vqhx>TNY6iBfZX)F3Cp-<7;| zfNwe+xF!^#COwZmES#rTR>1Q$x;Yl&Ma5?p%N6naJx#fu_!zT7aicpOIEluI;T7#MGa%sL!g5>DO|KN?CU zxT>ckGY1T#{8CQ?L?t)6wyxoB0`N7j!agtHZEx5EPXWU(3cPn|R+M9(`dt;fvBNa< z>qd1j%TsR&EKF^6sgleOm;In52#=NXHO*0*JIGA5l)_P~FHd3dhrBnb9aOwmAKaMU zb5@Qv&tH{g4V~E5|ZW6VpGbEfs$jrbdc^>b{t|M1ddifR?MS>yk>q+z8*zJ3!CYPnSNAK|& z(SlAnH^X;T)=ORl4d_du_Py(3FM`JQJ%^zISwR@^KX3R$h^UE&)0ueSSZU&VL5~~> zK^X$gUF1||vQBl_X`yK)RKK{87pxJMQ$in zN|^@-?+lp+>u>fXW8>z8ZB84T@AT&l5m@YM^#<2qljWE}G~_*C6qX-n2d#fQ6JDw$ zvRZoSJ$B8l{PciN!BM;IIqaT}_1*4f2mj*;e_4E6k1Q8hNkZ{sDmL2rQ^_9pM?V?L zPwHb6dF|5Brbzg#74hgKrsJN*PXULs3;Xlzz;0>Bf9moc1!df}&Fag{-DVCff)>Fx zL80^V=g^4V!YT&PL9?^$u)J9NK3AViN!GJeKh=EpEU4QgKQd;$otQ7EXvxzVBmDu< z;oA(4i~87ckB>MD{$+;9y%A$UqWM6utevE+Go5-;NMfDNGH_<->)$ERImCgJk)mO0 z!sAfL9C0UzwRgcVqv6;QDxvPl?lrcPbEGz~YML*+Ax#<)iWlNt<{MObOuxAWyhlfE zEmmL3*WWWT4*wyxvV=h?e4-KBuj>G84Rj;f5J_(gMB8r~pRNI6QdZ0-2M<_(76OHw zum2d|>Qa1LF`Q#BHEgY;Zzecu*&lFM6S~#D_b#>C-8d0K?RI$=>sC$)mTtopacaLP z8c@l61rirof+B9Q1o<#Vd`2+xKT$ZeR-G1Uiz5qhCCvvoD>~E5r*!ptnCueAF)p&K z2HE#C{B$6R2>?ISvxddQIK88r<(welmj5QgO_9X|ABDb2%8Y^BNfMHf#mg$y{?$$O z_^}K{K&Q@@wf#!n_xi6EA3&u4t_65yh8$S5&lU?X#S{7b&>ui`Mt9jnl_Q6Ugdc#v zh~EM*^nC!v&m@?)$*G6A4dmpU7((RyOu%=8V7T>2-Be)IGffTV1y<9>b@K+4inhU$ zaXmKoMK@BOI?+wA7OV|w;PQtHdE`k`G_198og!>LsuNM{lZjuRLHD;XI<{lW%^3`6 z@dyBK1TE!-^Jz`o*PWzViX#@! zs2tmwRvF~Y*1+5*I1Y)Eru<$k2V`QP&Ny2fYPxPxO2ypc-QLi(Hf-mnkdt)TI4pg2 z>#tZ*?|OWh!rxV=u;ZMQO>*y9OB49w;`>R#+jV{S(b`orV|}E5JZ59rHQ+@pnD$b= z%3hC=m`kaP$O<9fzg_5^iWkDYYp!@Sf?mQmidYg+UlY^38H}_;{{U~uy;Zg8tT#$NU=j%gyk`Z=*O-?5 zo8IQegK9El^((n~>;JamMF-wY82Bma$hKNeIZ@(iS0Codm-#Se_E~@ad`d2k$@u2|Y`p}B4@f~MhaNMBp_aKv|2YyI z`0TV2Lf`$n{aux@(`2K?Nn%A`4IR?khhefvRecW3joahKRX*oH6Jw^`BKGc~sKZ^N z-evd>>9Z{1+|~5ol%$V#ygzd<1y<2iE@JSsbgwzHD-45tHZmu|C%%YduS!*^y;7K`@So<2ZGE>)6}!XFt-x=wl@Myzf=qEq^*4 z<2e;S(cM8QWw|z0gpBI8I<;S0_RT=CG)QB~!R*BW9BSR0?XAnBeBcJeoXnYfB)|#j z)A*xV&zMvE3X_bFl3e9!;YC+)%g_BG)_+bxM5VzNPX(0o_f>w~`jtzpT&SWy_+(BH ztGgMci<(012XJh!#G7U8oQz8DSZfTT1wcCuK3U_r}gfbeBEHY@mSc#ziX`JmqGc$$gr#}?YXnAVwg1toX)4-eGm{fMvb?g(A(Pq>Nc*|8S%E#qGxyQwS+cc@ z&c_~$Tb9B~q=NWaL+YVUMBi+|9IAOh6`3&>rwKQUyFW;adU(W`N8F{Y_@0Jqs}9ZTPMxvFVs7FEj#y z^z{jX-r_s1Rq~;z_BQi$bu)d^_oxI3v;NfIH+F8;vY+g@QTA_1u_ijS_TXRO3I4k9 z0x?X4*#HZC5Oxx9r~_q6PQg1*6(=6$82|mKGY9ZGh3g&)@Ja-_^N0{ zum3^|BO;pDnOC{8)J}4#*Ito7ykG}??8ZvWZIGKRE^rNVTZG;_&KwIUoTz{ zZHJT12QKF}e%ha1kZS~mVebtp8(c=>(A0gjMxY&+fcJ4<^aS$R-Ch@E z&}*D&H7Fyaxxsx>9Pt~|yao&)K8PM)hQU>8D#!_}Q9nvp;03oQB3}7+StfDySWt({ zh0oKB8p5S^Qa&!hAae^9OQyFli z4mHeor0L}G?Yb6@Py7bkkKbSJk-UaLJ3k7g$+p^`kZpE7z3=2ix1J>1zvP=dGk$#Z z&LI%Cvll~%tt#@4h~D9S7KDbzlv9MGQ)#fcF*WTQv!O3F6amNAf7^_yPrqFKZoTn{ zt#oMLn}km0G+GKsAz$Il4uX-kS<<6k1uro)7VQ^ym5~ly7zM~YzqQSPV{efk;_DiT z;JR&7bcPwK-Y4yWsMY5oW<7^`K_-9V&KJxxETTkFDkMxTk`Yk8#m1iu9H!nT1`OyJ z&OPZ{{z6HLjrSRu)u+0sx_C@WY+ToLvuOZ@^>=%%ANriV^j#+8;Kp|&D)wwytpvjA zvGmVgA;>a|;5@VVsz**Fig0tm`3QhSRbjCTnU+y`qPt{44m%5>)`T$Z7BHnxgd+Qb}SJX6XL4yy&XcAX<3FHZ$b3mXmoOXl2r;hr>$i znUYv+j`yp9exyrbxT0hNZcaDvgF+>?b!sg$*o)iX{b{k7WwI5vwuLzIP~@9wi9BQZf6!loXhJ-p5whsMCvj`1$fRdM=La(`YDrF z-drSa0RHaB?Blmr6L>8RXP}r&!^1OhvJHdLh$T%V2pTa~5D#Z1EM_MQ;yY*R#U&k< zVyyCDuG)XK(-Bvv)_VufOb^>m4o982@I2C)pKdYC$zk0d9-Jf_6aN)&1!YyHwI??S zK{#kN9x4_X>rxu2*y!18Q3=xFg=t>!FtVd*#hSB=xe}K;bDoM&JGI=PM`cKgMHOK_ z?M+Hkk{Nb?K0P<>AJH7RA+rcp60UO6!vk4BWevc(Iv{X-9u^+$pL<|~QB2i6av_^M z=b-#LD4|A~ADI?S`j@du;-5LF{Cn{7)hD?AdI`6(4kY>gScy>6Ua(tv&J(A=-g0n> znr%Y10R>U>`TnFi`gARI71zFWs9Q6k0YaUW8~iNuEqo5i;&h#frjy07_+l_9&j5f> zZ*qcX_QkqE&uEJepwE9RUr#ym$I=zl{$k6Jn!I;za+T7TZ2|N$WBTwRQosi#R2yJq z9M1FrY%0H+f2@yW0q!60S0QZGQM4LTGsdPb^nOx*;@szti^gDhVrr>>qz;az;QOY;lt5R$V&~XHROE>s7sA1fnUo2-&a4 zmQ8-& zYA$QdJyT!;yl)+H;d)QFE%YGDYfl0t7_lR|w?;X)get2L+mG6*AG_z@geJ#-BCHY0 ziAfxN81Ih&+cMhzq2qG4VU4;*j$VU@ZRmbPud$}B3h$}f++BwdZscOHUeifR9;Z`s zcVdsaCoRs=iXvX?rHQzcRbbg0k8s}qJ&WL#R5t&(@B?bhF9BH1qF1LGnPetCzDSy!in$Ta8Gb>1DL z;z+(b=g^xb!yK)Zc8xtaEq$y!4SB>EWq2~oG=QT$z(=Y-XwXwX_aHU4e0+ z4fsH&>4yFpCzWWM8o6+m5sMi!^={`K0|PmEpmE@uBKjE%?ouqluO3nhE%>`Jlye+!ee^n!&0}ksU_+ccO1nVN>6peb4ds% zCvs7gc(M?DKW*_mp#a*slmfO^Cc|DO`s5J=ncMJYG-lCD; zlTs}?TGjVe1w(}I1$R81q3yOXUE;Q%ob!`(=EpY6veyQB1LtA4$ypM%U<1h1KlQPP zZTWapXN*WeFg1%74zY4tN!k0$u!&N_dd@yGL>+^#@*wk>3s67xlCg8IX*Ow+u*edl z*Yixbw`Hhk=uw8dX|QJMV0F2e#Kiso=7=sf~ZY7%ktssfDiEOD2XA$ zYn4;iR94j>Dqx?;Iqjqle+HyFd8b3=BY*`6KL=O8c@l$#ekm!)gB2=Uez_9Q zf4RrX7F;HYDF~u^{G!3mc~l@{UnNnd5rjb?zYf0O{9!eMVQn=j~zL@+t6av~N27 z^Eg=$Rp(iYyDgN%$TRt#X_sj%)wX~9Gj05?cdF0`_s4+BYa3G)!7ZZ7y}uNT9-Ks_ z+lwBRBqF5=Bjy!g*v(ecQk<(c4^PdJRi<)pqodu*D%xg|lb3(aW4eE!GF-mOr7!Q= zFT!Y1+;d2K5JjNttgtuk66U|?`?3-DJjF}j#Ls=diJ0hLYx3)&brHqbu6 z1TNgSixsH?NlFV4b!h~;x`h7h9E^eHnLoO4M-I6hWGo1L(?R_GOp{hTet8&FWu*qM#X6T(U;lIlOa5{bi4 zyCD7n%PRMp%LHwJ zD5ox-hFw1g>jlXQ{3nf?V!%G!@CX~aL?`t6SFmJAdlywIg%#bozYI&WYK&RPYFEnC zM%}Xr!Ckz}`8z|JON$oQMJsLp-~-^lkI-qh4|Oc!ZSg)wL2yscxOelk8! z^{{4?g_Cycp8u^$gFLqWfaDhjP9f0*?i(54zUiwCW~vmc#(`OvXyco_~TFTH*-hS288Asa45RQo!s~H@$@H(t0G`1Xl5B5 zPYhqk!%RR8MB*=bT*z%{@fSrc6&RJ??j)z(@pZLi)T#X z*!zLiAJ(lc#ENILL`NmWNBSYNUi$LJWwoGn=-$j{gx~Vu4$pD^FeH6qTW(`dkUrwk z+N_a(Ma7BsIEQ1YfdKw#cQL`ts~rd|D>bTku`u!#j+3Sa?l1*vGVS~sp&g-W_oTRw zx>zEBu=e!*be_)y!6_49Brkn&Og^29bzf0W0d1Q+q%aFzs+0PYQ#kBCj5E0mOONoA z_&QNJwD6!uphZcf%i$dc7$724B2`>m1ky$;oF6wz&DZ5)MdL@99DuQk6FVZsd`^(%k1#R528A0Jc?@KPFzw=jg4Hr0KE&xYN zy|-f~aQXc2YnGRklazXYRDuacG{F*RjQ+bhy&TL=YuWmjwfv!dGjF>xuyKN^c>_`v?;wO zEEbSr5eGz6o`P*=N{C*=y6N2gOh!?w(9`*2Q7KP0|C}tU{QTOqK70dcL1))(QgmS- zM`B5MGns(|FD9?JYwH&R%N7oa=rO#r;_)z6+i%!NROhaDF1&@GPMW9b3IR+B?9_)H z?dG*vU!<{Cg}5a+-1g-yhrF3@7ZJ@2zmFXgFGu>LEhc&!@0_>sA~tV-5KQvF(zr3d zkkiJ7Lq8kU>=*~{wB1Jqfr9ozqQ#hpo3?F;#$4A?W2Fv)Tz)%an>S|*HY>*t{koqL zi{4UQ%)|6$PTVq2NMK5%FwaRtJjG4LHP0{ww=hxvIup`=w*^C|aIM(7EHa&H@{_=_ zeLeU~yicG9!H_F|8hKNIjjLK<&-r&9^)maIG*``$BMhQw#m}96cp2LDEI2ydE0aHz zn66lw&^EkLi&%H=QJo>Ve01$6hyY1rg7Ktm@pR-obs<%ISOSpNuD|J#fP2ERPYXoM z$+;-E$yWoSto@Ar`7#eHTI2+10aW|_4zhih>wAGLa<>kv@i})dCE6B^5AazvIh(mI zcN7GdWdbCkwkVW}8<;q$y5o%QDWSC32HaXgHLvmNH9SbdhX81j&PFqmV}xYx==c*? zB3pnnws5}dALIbKVWYvb2Uig?>$V`sG4c}%Jt2jL&XoK>u`WGYnwKGn%wiA7Zd4EL z_ezny*U2&XYDtpG5%8iHS7(v$dtj}%IU!8oon;~RIQ2m8yK=-LB`!`iTcHz`zAsA@ zo!L#WplX-uTFwVP%hZ01qM61KoJTtD%Rx?^*rRzCO8zMy7oTuGSzV3(BNV(y- zqL)WSu6N*Ft)M4DL4iCr`M}|jw`AZscpY3c@Mfu!%t;aiAjfX>4GW!y^0ah5%KSO==2#wCM7BrF&wx0HqU=VrB0o`je2Q z7D@cp02=#FdxhS$gB_|!%ki$rKm4ZH+i;)UrsG0AW8T{iCkoFwmTvK*H3=-zHDvMM z%7K+D3um-f5%IYH^s^%mgYzsSs2B=z{NhPT{_JwWp?^gePgKGJ;jrkX_@zv|f4q11 zU>cU0d}_t2yOO&D*J6QKq|ukC6k{^hRpDq=Bqsk{Vx?3=xh#&n#8FF&W7jD8?pZ@W@_clX{h>eO0=8J1 zxt8^EOt>(t#3QNv;QISJ>z-T1Q!AnP@lFfLSzC6e7aU#@rS--$lJ3P1fGcw$N&}T8 znDfbYX($&JVjkdmPlq#LBWkuC}8Mx=9)?k?#rK?Oxh1Q|Mop+u#d z0Ys4Q`k(pkcYpW(?^=MYrQ^)|zUS<-_p_hp*@?;?k;OR9!FY^2A}THh^zu`zOtw22 zs%Cld5%R76#4RXtw?qq6 zMUA$ikv0;Cde`6>pE%nvWrgt9?S8Z>p<#@ldNcyy{!1&e>0`g2$-Z$Z2vsrrP5u;q zkKJ)r{F?&Vvs9=vv1+QMhvO zU8Od}T@!2I9xLdSkT0sEz2W;XW@*F+p9Kjw)rVoC~XW&NHd7;;z zTE|y{TU>>8*Rt-EfjT&XP!6Zwx@9M?_RY~^CZWV9pvCdYaQha$5Q9C;9k5$}Mq3PJ z>|*r6#pTL+A!So@Tc#YypHy7xEr)IW{SEU4AfEv`Pro6c72f)>Cga^|6^ZAqF9!Qa zLgYfu7yS!@*9TKY(h&y7x-bf-=OzD94JGJ_Se47-)o$ko2p?(8GQ%HsYqXK8@JB|; zB(XUP3tZC>%|Q-X^t(U zG3&PDzLwTLo<0v$=ih|DT;1Oyt~$}9EH2T#U~pbZoQJL61^84$$UJ37$Tq%7%8%I- zx1POH6t`GLi9tF2V?gE0Pe-4D31zJGTwJ z+L4TGba-n&&+B}ryDpLg{$l)}KRTIxq#>Hr4<89Qvb_2FaAo>Zp#n%D(!hJ;=2K>f$Ji&?kT*fHpjiD9en;3SDxmQ5&$|-0h8O&TS)HNXrG!|;g6nSk zOYQ#Ru|Z=(Bn~6?4(7E59-6>J_tcCtL9Z=MdNj^%7~T$yr%A8o^dbp~(WpHo6q~zc zktTuGXRY7g8(>`#H;Lf=cwm?rHB|0=wBbh`etEaA7>Tg!JrW29T2MH}^zGeS0$0@qaiepMQ!K|D&& zeS;*q-b}R{=C5dbYO|87M~lBcMZ+3-NkP^Z_sQ9|VPQC?hm$4QWukP<(l8LS zKSa{#ZaO=+eSW;mk|2BwDIDn0GfX-F0(N{_xQ;Iv6iK7fhp4Wp75L!I-gD&tq`bV` z1JrG+AnpmVH=8)5F@uR%I!w-h)&HNbS(=uXZi43^|Ey$Z82ULolEXXrx%19BN*`WI zLq-+}IveXiDki}E(cDz-HFh|T#wO0LjI4oAgvBlX_tY8rc=8S;8 zuzQKyxcM#m2K9oIk{e5)lGSG=?W995_n4>uTqp$@Nz+XxP)vKYLcHp#_GQ?P3BSDKbMl?^o`>-i0|n0J#`gj(YZ0-`TFw0LFA}pzl;9g&h&kdcCYkv z8T>_vDH-1#oKFrcUwyq1zLaM7-sLAxvj3dX*+|8rV65AGjD?L?$hPTVg+FfX68P%t zFNq1f>RjJyxzO2Zdq?j`<`nj%iqc}5y)~BD@}@%_yC`1;x2T{P5`*uKQ};S&@jdd{ zf?#mX5kkNYO~cSyRKpElx_UxQ_bjP8;eRhZ-7|ED1$&liP|dD_{+~ck&AZ~)!x+8< z^pX&TSRx@*x`q4giqw2KjECHekVY7u`79KuVl+yJ0L$Dh#k$HOTW$-Cd`Zy5R8+mE zp8w7RYavIVwidYb2`MSjgvP|{jT__X3q;yfpg0cUidy5!lJRH}HZ$fBQTF2f-hlDx zby~QWUM~t~7#~*@DP_En#(XAcQEw&@{m1Lpr5Rr_f?hk;T`-**Qei5I2YWTRpPUts}Sz=#h{Lpu{jMZv3{X6A`TM5jUNA8zxC(#K- z4Y5t;>Mwh9_~lKiQ^VPBKsv%T2?-3lal0HP=qOc94#y;KzU=)G%&pzDRXdgSJGHcE z^6KeqXyGXxButjS@P66Hb-Tf)`VHfS+Fq4Gp{m#yxYaK+v$fx!!ZWm+Z8$$keq;;O z<}TSg5OzC>S1yC%yHdJ$`+11;OE?ukiQA1}$<2vhJ?0hW5zE_RA0X!6e7XTqDoKG@ zihNx1IEnqVTf>mGBuD^h?GP0clfRyO$sr@NEcs1QMP(f>INm39z7Po*s%yxxfy`s; zYikEU$zCx2+Ydy>4_azgYOwyMuVmMl?9FyD@ke(Us(Gjz%%o9n0K5k>Kk9Wk{@ZdF(K7nK*4M6;;Kex-anjW^7KcrMU%R&K;Q z&Ew~Y=vBg$XU|Jdr`ev_eeJE9@G`H@=df`eQl#EXttjl6@1L@KpE}ONc`o>zBiu!4 zl7!*5U(Q;VH#8xj4;uJ^FoXXEHwNEy^vfPL|AFOV=%Z0TNvy3~1 z^%0H*;wb5HjSmu7!cTtfvR!rz8kfoR+igmW4D9OPO@z35=fnJ}XgLS@r#@Yd9Z}uv zaZL`9rkuRHS#8D7IaUf}6gAK?)wwLVzck<|(kyVnl?$Cy*DrcB8j|!hbR^p={=;<{ z+cicYPvslyo%MDTa_gegl@+T-_N8uDGV_@utk)$_vajYB@1f|fWbWmAYnlt5O8;wR zh;pfi?6;rW9Bq8B@&Wqk`xH=TzkcTx#^pIiU^s0N_^EO5N1e3-(>EKFZjkfk=|v!| zKd^{s-@77%%VO^P6Z#akO^ruoG8&iZYqUVU>PuR;V{Ucqeqkh*DjrFUlZ?e!uq(b! zsJ>nA!$et@`|`c8oVAJEH*QCj@L$#bVL}Y3rgtbma%Kon;W}Lc9OSmGeE701*tw)a z;#&q{7@n)O;q4Dc(agmTsFO0i$6ZK=$A&JCm4wsgTX5y&BZoV=0`8!DJD2((cgu#>;(TxvDeUq7f{6TQY<#kffSk}9O&5x}0rP}-R(&7Eke#|$@ak?W?zsqyaqgDb)=_dp+rezZ$uw9r>I?za2 zb>DP<7uvNyLr=Z{pDk1LL?}cJuh!fgW6)LBZ_;mOK8vLd=BV}bzE436jhC9|d zfAwr<5sie@JsARAdD}b9Rjt`6_!~w}!;f9r+@^e1T`1Xq@?I>ug^vH z-eM3nV++b^WK4DfIBI*Qz5;@hf?0||(2U>cU1vVm^wL(?HK(ilY^C|-6w6d77_-YU ztQEEMfWcTa25{vQjvIbt5+ugOO$Z;Nxa2&qF$8KQ!YcFJ0M+^aGv+20N?wa3plm`# zf~M>l?|zy&DV+XWqygH$Zm@uMcDYMF=)cQf5BRu7C?FsJ^!TCqo<)9jq{dEv8<3i% zi5eqbr0z=CeaYpD*MZW`u!|!cp9AwlbLOKF|GFfVGwQa~$rbK_p+`GhnJtxABg%!- zB->1PfLX9T_l0v#Q1P3v!m9a>Ws;lDz@1~lNkZ=I>prv3Tiym~L7q#QD^Y92CZ%Dm9;XB?r(It03TJRoUt!?|) z^P^JT|5eR5VeYNj06?61EF?q`dVecIB)S5K`LU#ON4Ceh;hYUhwZ8Hmyfkqmk$L;bHl&Zi^lR zRE{fPG_?2GKPn2d=K^A4VxoUPHXeZJo(soqi^LD{q5Y%5u_QO&A2<<)V2n+84T}b6 zfmLU}XKYH$8eNk)40rIg+Z@mRz?V?uB-p}`t3R8qdV-*x3KENc&d;#5U2X=vhfo=i zYs}16VC2K$8iJnh0Z9V%WOI{KW_QWmyD*8d!aXLLvLClT*ZeR;2nzL&Ut~iB9xN*f z%hW4gF@j-q$=eQ@A`Gp+ns>F@x{6xt*%w|sZs8by!#$tyA;{WL6rudbaYQIhQfV%z zo@{hnHN*i`Elu;v1+UgG>#`=lY$l<+CY?!Tr{|MOY5OQ98->{|P0CktGvECsb4(+8 zfX)js{I{8tlh4uOWzs(9Ad--RmBOaKSHKaTFS=2^;(T6b&fff@nIKns5wDib6o)dW z%3MPAp1A2{Z-CbpVOAUf1QWqK0$@-Qt;j*^;uDy&ot~aPwiK#h`v=_Br*vb0(&y5U z?xc@FPpvjCea-cd%h-xu=!&<_!-iQxg695mJQTjMg48NvIWyPkyiI%aCIY->AyY;i zbbr2>qY(^d9@g?$pK3Crn-(YM@IIPSv|e zVDyWz}&d1e1ac^h4hw zIw2eivpxER-9Gbyxi8Jhtx?nUOm-U0RAW#&gXh_TWBk?s0qmX6->+0-c>#!mW-Hy^ z)p-rY+_d#-r)E@Eb}~!+EIbxnu)-E_i*3M0m6~Ir4>zH7;#%}y4sHR|!@_p8N7k3T z4#swu;uO!IvEck76j}7iAhDVLY`HvJgOLwRJur|7kZ9p6%?~O{ACf&*+kjC(l@E(j zNz;e8zju>WYu?g&9{usvk$Kt+`$CmhtF2oDq+4bvt7`^8G^!EkRA3e{qE8hJzUSSX#P+dOmtO`q&?ueBhoTMKdrsruzJL$j>LZU=Twjf^wT zxja{zNgUN})_IhGk|tO7NVbdlmP664#+%Q{DCOqNaFoC` z&X{2ik8k5C=7ih`PW}<{k?_a54syR20Ax;cAZ`i-{}p=cdWysWpoJn~r#J_Cm3}fa zFoJe~q)^Z?#i8X0Wp_dJktF~STY!y_ruthak zh)?*ZgRrqOKz%FzfqKap+yZ6pyFp65Dw>TNbXlq*f_0l|Ww-KKy9| z=hxKhy!Ic#wNUK2_Bj@i?hw1}WI(aU7g%qIxa5oqImh$3zu#(+Z>loTkUNY_dLBm~ zCC=x(q7piiE`wIT)oDg^%-G^$_{00w#!tY39Urk!bltWvJ^OoUw`S4r03KZK@=ke) z^B2lEcBIXi*cL79^tUO8@BtraGhP0-wbb@gxtl_rKusi!kOsxs=Do>UgM9nzF;rxP zLMU|qT4?PWmqF*}m;L)&xRHXa#xRFm4(wSaOrthr3`;|zu9;0`Bow_C+J_iimgfMp zvp(CMov*mh?CRD~qfO(59ms=77q_ut2V`yE?RMJ0IamZ1JC%eV*O<(G5co@iY$$B6 zdA`nNwqa3Ap5xaDVAlEJRn&%S zcbN-`q_o=2h3kL8J^d%h+7_TO@&*Q(0h)suT`MWY4d+BYy@Yi-7Df(9=p{P@CN9p5YOo25(vZhSN~>`w5x$CnC&Cz{`fhUnZh z)bPN`h=N~MCq*ONbX|iFZWJXl4uP;G>(+{RC>THXa|w_@h;l4mfAGXG(bK^+3Xjcb z-o4=cw#_IBjF#qE-)DBc$;^yucAZ}X>s$!&BeUxmBo+B=I(Vwkakizfa{48v8l%s` z%w^zU?D#wDSRzzO|J+kqs0e0_0U2vjs>Lgnr9n5=L=jpXgg6m+{RR^#`h(b|1NrxO0K zJ)r9m0hnCTsvf3@2zt$$=KpU~Cr5C;$;RR`M$yRaB3<>f9Lerl>2B^T&G|CU(b6#c z-HuIJi)dqRA-y|cIJHN9J|x^p;lDSg7)&Yp>5coSuucjP-e3J%F{9a@dPl6BOkJ7M zuCj95eI0e4$m0UVP9ULL{EXh)S{HkD=&?V<^J$%0tyGrW|GX@!%>i}nuKIuU987e9K*Dt>hxIYos4({)_19bP9 z!;oM~j+s|H&shM0GI1f0cymM@<5=5Jm<1j7S3L<|qhq6QT}$9hlGQ!Lf=3{E9ElkQ z3JD*lNnW4sL?daY-~|wLY@$9yj-}fG1mK3e!G;wk=tf|lUCX3fh3 zn8A|^1$8{+xYGZ8{j@f2NI^)mXczD9>Y|PIMB`7k2l^uwRxFW~q{g|)Z`E^nSVXsg zgne37HWevlFL?9*cyqL-k@*R%5Ae=4zvE=9{x{;3TF0woliLN9gSpzsjM~e~OM_)Y za9naKGsqw_ALsKTlA#6N`!YRIXWbiW5BGJgTv#yE_p1ZJ#A%{5TueEq0xfY~cy@Ny z7zl^{q^_;6k4n21bgedz0AGakH*ai$h>d(I_eh~L(toEvi!e9M3fFE|3x6Txm2?bwxhsJ}*z`xKobWLPugd2ev-1#ZhehTi``OqpAo>|DbN&p~xoLhaR2L+cmSull zZ|=0l0DT1J;@$c`%6R|HwU?{Sg(l_J?V0u72=cSQHAb6X{(L`q)-W+Ae0Jke5$rPy z1o%(zv?ULtj(3VdUi06Z$)-lP8AId7M}+uKnhfse&aWCzWJ(TvW(L6(gAAl}#ctKa zu_(GD_@kQg zd3%Yx9Wx!JwHVnjn51`V60I2Nw3&BuqmiyNk(pNca#u8)1Io;=BT~vwH_^uAY}1h5K*R^JC z6WC8Jx8>(Kb5&KPu8ochcX!~;`|11_>dkLptp79C8tJ@v2>?ZnC^E`Bus?@pJ=X`U zFAi2&^z?DL3X_n5Y!Vpr=w}-qVkHSU-uuG&2j8dlp>J$$trxg=@H2lg`Db>BmH3cb zz=SMS<_^kDvq~Zweh)IXyd)3uBMvo9JHXsCZge0luc#R9F)(RK#WzT#xP=)Huvv|~*8?VKqiom@zEctkN#+u@r)vY|| zH#@HX^)iIsSnn>f{azt+el?%D_q+W}aM3f?L?MqP2;@mf8=)OV8N+7Ro5LP$$9WIi zhrw5W;Htq>nj(K7>mWTRPjZ`YCIliMqzg$@P3<=|*G7Q@Gd{LfrNoEVPZ?_d$7%V0 zH4Fapv+{>=N08Vzzr6%dD8RYCaV%K_W4`UhHj$9KOGjETeSXY>c>?K91U()VskiCJ zQ&d(K52WDj5~wCOttY3%G*(+Fi;J~|1I}A467rp`fud%U|C_02;z4G;pa>na=*QO~ zn65RK_>UjbB_exz&O)s{B+G=a;BnV^|EObKq*c>h+66Xs!d3i5WY-tYD2Y0%3F6+!2 z8X7d(T5ie5gJ=E9^oj;AmL+I}UHAr%mwrfp@>&rZyN>!Jy?C~AI_?cWd@I>f^0{rW zQF~pq!hlG!b2k)f5fHNIZDvu@7ux2kduFi;eCt#{&@T9unIoF<_p_RsnnD*jT`1}Q z%2Z7VPlc%epQmxga5ep}3X1RlPi=kx8ES>WA=0}y0u z?$6Y_^#K}pMWEZCX!UkPZalDSrSaqv4y8Tb295+o(ic_> zKt$z`ZiXE@9Q`t}B=DbD#=kb;^T^WwWU4{$(b{;9Wk@`rRfN0?jPrE`_6cq~_j*m0 zyE5c2+JV4V^m?~(G`Pp06?Qips#zrdD^6vk$8)pu{`T`T;*c&>?A&i~K&LWn9ptzk z+nsIHyccwULr!*Q$*}36rBovf&=u4vqgFDbC>zeIaYS%8|G>S47|;c4UinpeS*hJ1 z?GR(>U$ebF*$(G1|CSqw0H6yX-wHOXcs#({5M3ew&+j_dYpckg+Pd(6M@F#Lu2aWm zIfj=i&J(}>=fq}NQl1}dbDB5c=|1=RqVqvGm(Bqskw|xxy4R=9g#nv+qN`D4No=d& ziuv<+h-7X&wwx-ft=sJrYI9H5fOAUM%XdW%JGTv*Y+9udo+nTG#9n;BGyL?zelC0~ zV4_s+0#JgIJ;L#qadC0SbelKtUwv*z61wyK#DU8B0qD_%-dzsjXtlJ^Q&1piKkf2L zE`#$yg06Es_gJI`JI}5(<9Txu=b)fxUzxkaS!@4$tsHMr?;^2YXBoj!_{;EwZWl_9=IYl@lu@1RNtTA2Tj2{3!beoHw zU1ihQ1xO~pkIc=10s)=)PdgJm+pkAX3oZ5haZVpW%x|j-zHkrirofaTA3a=Z1?3RD zy?1>!2EI|Trz>_KU6wJ5TTQ~GoZDQ=uQ<61>Y%8Z8Dqh!sOadtb#PvIP6@+1`nZg0 zFC4ogFfYJdqvWo>PP(k%*mbebo_H%W_Y?KD41F$!gdgD4hm)E zi-9w{XL5LfdC$fy``-Vbz&+g2-OA-Kx<%_|XZN2?AdX@_*k%|($K3Hx*W5qy9`{^IJ;EA8@91h$J_ z7SF4xd6JH=w$_BBvB|eHbB5CplRtj^&}gR5*01Ot>puV<$8@W!t1S9?&8CQM)Eq`S zhc9b=IM}O4bJ5DKz{!Xm^*fg_EH<}4RbI5x%R?!i>)GISlnIoLWA(xlm}yX3Gky-v zrh7s~MaA-$#W$V-X!i=LLy3cx zL4YmREvW-3gLF_!s}UA4z00tecW1*ryhMh9Smvhgd*R4+FTiJjIzX)hX>-kt4J@qi-s6QuibR)sGAQbjNcT<~84$ z{q7XGd-oS5tV(^!Y6E(40RusD7CssRRw2>wf5fc50pX3Agp{wcs$+|!u@Qh2k@t+n zA=vsX9b{ZYAaYImy?5|qmRyPA8Y$&3;t|Zaz$g9Cg|HEuPh|Jc8g}%@uO&Y^@YR~| z3eUD7v{h_c5z%5iLE(yUOlI(1cLqs*FmJyoX5tUJ-mPfa3`avSs1#M}ykLJJf@J z2gGL)*Dkmb?2O%S{i2({KeRk5f;(H@UcyU1bB=%#Gw<>M$$VQ!i^heIsZ2gS{}W+( z+TG#2_|4a$aAl24={J|uxImLxP-V$`1MVi%ynn^E*y^Ha~%On9z19sVC;{0$Z`)$oEFVn_HihOC*+X{(yCk7Z_Q3@7I~}-3H=C zW)6o>i#0arL*z+l8=LiJdb$3=X?KX{_lo?N_!-H@yzAkDTiv;*S+o6LU1ou`tN}T& z5Vz2{_663co3w(jZ$M*4;5NOpCAPbWK-6l}91Au{xhmRw9o1bTK5HT7aTgh`c`_PM zJd1s3QO0nzS>T=@7=7FmWm5ydk?$CkdLEkfw1bcF>SQiA$h?#O=ipxhMn}&_Jp>XG zX_9DHqRM0nA!SfG)|WNJS_@upZa%y#7RU%Md}8-=)UhNu=6FnEJT3^Bg+ z#TJLMP&CGaBq80|epRaP>z!G<4MeWD>OXNRKR6#9JG#8bw_6aCw;} z9gP>4v3((#(`pLHlZ!dWrF3lNzpxreqPhS#=xLyn&?=sH>&obECOrEDNytbZZ*{6M z;6546o6VAbH0Y4A;mpTf@OUTubyOI7X4NZ!L(+2mV* zUNUK1t}UmaQ%1EVH5`jPy^i6!^Zr^!-==_hJkR{u2U=slxbk99jx7~#fgL`Nv#|YH za3zM3Lu4gp9p`-Vp>4c(Gwyu4eIlE&U9*w>y;T@?V{$#O4U1^=cVMmPDqy5L(<@6G z3h5d+j0*mZ^x+uDMV481w&fEBv!S966O?wLzDNL8cvW;PFvgFv--G-FV5OEr*I|f1 zvmq{I2`dmt&tcSE-2e0ZbzGDN^D|->ZdN*V=5Z0^N80?maUyDv5+eMhF`v|eUqqUk0YO?^72J@sJ`n0N|ezo&>NJKn!uY3X*6 zYI@0@U!~d(^VTN8r5q9K04YXTc%`vu)fDd~2r22gx!WM;Bqsl&i=kp!)Qw4=zB?9D zNeI7=k;8@W_Fk*()eJ{Zi}j1a`&7~yqyNIw`i}C+w*?B?0+;!E*#=1x0@;pMrIb8* zgOyTbxRgG3$56!H(FgA%F=IOqdu3)k|5!pAhlVEtkYr}v+Yyl?De^i#FAo_SD-niWQK(+6VJMU z`_C$x=sp#X?4>r&QSqvDGeh(fwbP#KDEy{>7`&`xVr64ZmvFa!J*B%0*zjHVVI;mP zc}fsC5v!ikVV2jn#C~pD*9gdH1g(}0iNU;va?8rKx{V2tg}JkgQ;ov=U)FxDxS~gV@NUU#Z*FtrgRM(Kp8=%?3*#Th-P_44{pwv(d&& z5@vC0x$6#WAe`e-nwl6EbWVgju zSzRin)S2o>7A(KSKyGVUVg%jUCOp_K4ao)g=a)Y6^5OX{DK5zaCYjoeHRJl(HZ5~ybA3l)C`zCT)#lCA$W zQ*WzD&NU)?p*_(M3?i>hQTdw&8`5U%LHcuu{r##9iTXI~I4wXUXmjSWSws?&r9^nazF$sDF)}6i4UUMe) zu55G93n7zTkvriX*w^nlYg>9U>1_r`9ky&^@Dr4cuXJ%Gk!i_fj18_UNcIP5Q@Br# z|0r<$!X6C5gX9*ao)bs^eplp|k9%V!=9v9OznbWZ8GiG-(Q$k)1Xqyt(83x;-Xt*L zmTsS-V@&gdS*|bAQ|qZ~_wKXW@xd@+2Q3>}H1%oSg+Snxk@tvxNs!zIO#S-22Qrwk zOi{uE6#{sO_0yg73NCn9&E(-RnP`I)xpe4)(U>D~&wXN8pk{xiEKJnB4x=t0C5# z38(WGnpazS=Y?T8GpSkaIS`GJ^GU3Uzz;_&uf@hi&@mRS!@`Ak8~vC!S6g~7U;8i2 z!UDWo!~dxDueA4g!>=RHQO5-t0`@caK;?K)?6HS(e;w`mL0As4)=W5vXL0>csw*%? z?yG&|%);)V3qJkG`JW)eeju3?>W!a4nmgkc7A45Q)8yV2q}L;|T6C+4d3frRTTuG@ zUsS%H$x82gChx#mX8qK6=VuX?HPNtH(i$TLC3aS|y^*WIL!vR-KC!%7lWPuaY7K3f z+ou))Bq2M!LS_~}wD0&BvrVp`OD|6ZIAibyIe$RQ(?>kxsO~avBeDJhgv_yl1?ddj z@~5hx(CCbS&@AQJEEn1 z&!OhpN1-?+!(MQjb`9-(@AU0;>8Gc_e!uZi;N9g`flGW448S??KE^?+GfmJ4<5rT1 zI>+B9xsUgeemCAzAG02>7hET0Ja<#*J;r}dsJ?J$VZ)+}^(7d*+#%f4{T5<>eo-=r zO9sX;(M0bcf!w6r1dm;h{j4K266ogx>B;ES;pmJFfW>6zy}DVxzjG9bZ=E=oKq^8( z%QG)9`W&Z3&{jabEQLX4Nlfnp0d!BWKpNAC;I6cDP>htP?`Wx%L4LSR=E7o0oz|PY zvZ_}JhSa^C2$64otF#xREXV)&3m+4k;GrGWi!mC#%)f5rQF>uD22unoz-y``X|d01 z(`?+Tlu+<-w!ij73L;nYkVk#07(mtzaKAwkkOg%f04#PDkoyLQ(wQ{p4}$dJ|9JIE zEyv!>!q;A$S=jj9BBlJuj|cGiSde|tgC_dhR%{4gSLqyRRkkmn$c-H~;PnGm%Bul7 z`tdiT@Tb)VM7x6)fqOz3LOG~(yCufb*T`~jnT*3=s60GW0bIjTe!0FKL$ zNXVqx^Ob710EUUn3e*v~Jb4c4Z$ZgP5dI#nkSVc3QO|NJNhsh6zy%$S((IW;%PI|s ze2${!-qAoiW(%tv2Cl)3^p!0iwwX)07KlN))~2MCovn(!OhJP*^z4!{Z5@drRZ zBI{U1BFc~T4f?^}>%Sf{zgc<9UF3&XgQQ;S*M^<{K+;QSeZ^^>mfm%?AESJ&d4G1H z9K8R5SSbM$<1NP9XM$yvWEk@4MC@pc4<46sFyqK2ppw4D#EDnsEsMucWsBDn@*&H| z!NrMVOVv>f-Z_>U8*%?vY`2&-yR|#@CeZu7bH8AJw{c{5Rqfn!RX5k(e^*Z=; z#FvST-Mp?M=$o0C_M};!;=@!YR(#GtI^X>u-9K?3g8rC|0;dD+^KevQaO$~{xzJbh zEvc^eJ7DGok>L3y3$U(U!H$W|q>GFWnmP#SPYRtQ&6EykAx#vAB%z?MFUdv)L(=-Z z6$7oUC=_%g-&koR5At_B_c=TIv{n$qs#NbG=O;h57Q9MVAH&*o-E^5*=>z1R%NzvM z(9rThX@Yluj2#7Nl-U5F{*j>e1;(YOg@9AITgtx_m3ni(S!S7vis1NtHRppgIr$V7 zbnZ#X?OiY!fN*I5GGW!#t404-!$~_A8g@t=%(flH)u|~qA@TvCV~V|8w&*v{O@Oub zC*hg+iEG(AKK6vOD1Y|Tkb|X(X~tl4AAiWtit!?U6!#U-C444qHJ*);YW&} z!q#~d+%xjnmhpsyv=)qe;=^K`BlaS@jTrh!;AIPyE&1e5yglB7Lu8UDAORp-g)Up> zTf-*S@-KX~avf94DYVgU-H^C_&Q+P$4C@BasUpdMqXqx?eL6BZ#lOhXFw(5gj{3?E zAg^Qu{*YAplrQvIPX3}Hqe9ad=%fSkqS*WDG8f&%4ea_=8;%JMvQ@Vr2+=219gL;G zMiLmdp&O0&(ua$-a63S@ARLhWy69(mu%;z8xz(?xU+!C1j`teJmOe>^ziLQyC}69I zl2+{CR=#A5ZmmG+GKpG<117N@H}tBIb7!mK-I}*K7vF(dG%7IzGnN(ALWOR~N;r1J zucw0Wx^GH{5BFkYR;vo!@DO3EI9QSd@E<4>akev$h>i9z(do%Ok{0TW@Auq8X<&)` zfhgg4wiwxQ#;#v}$-m=LS6D&pfHEzY^T}fjTdQ)p?%(046!L#S6OCdCUtHd!fB5ABdUBCPx8?#QP$<>yv5a>2HA-nT=~et{0h&* z5RV{ij)ttRM2(fXdy1`A1buK+2aU3c#Ld-fMON%R@#QCT6iI7P_uR6W!e~!ZGl2)_ zSV>!Kl(lkV8}t3LS#zQF?|TEokJ=9hXs}I2UQ55-bL5MFF}i`>Tj3}vBJ-M8T5`%Vz4eMNjYp%pOkkZeNm zsfk~=tmWf9w1AW?5rWFY+$gtjU7uZrclSBUi*iYh>Rk_$?{Y_QuTUnj@{7L5FR)`CLpLk5b8ttR zm_WP?I>=NF*)R0m#7K|MLwGqS`jw`Z+Q%w3gRF_%+a zlahupiVBIfn&mo3N-{+7ZAI*qW3=^Sb;m?ZYyqy|Cz*tMf{`nt z7^NjfTo4GniueH&<`w2kd>A&fRfM>*6jz0^Eb;1#E1#DPd_T%>6#vRDSIMN~qHcQ+ zDH|>~>NuQyU*~725Zka}K^W&5HDL|B<5(nXyngj46XNn)VY44iXl~8|KNyzi%@i`3;oM} zEcvK{Y52LjkLw)e3Go2MOA#k;s1aY+!2298E7$|E0F#jZqb(?t^V$QU0#{+w!SSo{y#hkNnR;01=prAu{zjkoyGzJaRw60Q8kL+vsRi z3p*Mf2Fea|v6V3N9_h&t(EM*71_+TkXvsNpgx>zObHyf;+`tP&|J>!2Ld=9A!Bf8HfE$Y%Uy zUb!wH8eR^A9M;stdsK)M^!;mL086|~AGnIZ8RGl{^|^$_hT4 zWz8G}4a)2i?)=rzyRA?TKsMEPqP+Nba1T(~4;py_l|SBTmkf*l{ivTO>S=R9PX}!1 zvVfgYn-}P~J--9@IHnm;NBOfp2O&7KLT<03-3uQq=1HB+R5T1wzlsuGLEh@)_N<8* zfY;oW1tWP2NA2Z{6Uv1iJ-At56E!z%yiv!Zdimv-Va+bEy;-GEYJ2oha{Gbh)sLKr zB&7n>AuyAQV9^3Jp5$U02pFVc4Tx-0!!h7$`u(e`9gu(2egp|j7P37&C9(^a0pT*= zm9F%0-nc=>K{T(y?-#E=4VTJAasWQ2Ra%>3pqvPDni+>W_ZI_ufP`8vhu%(yM&Er( z4c@htQ$E{mSx=GckS8MhgdKbfF1F?)7z=4#HO@~4Vqkm|XXr&Q0|!8KjJ7+&LV}V}Q0zOODasj? z6G&2;dC1v=L0_K=!Gn`4Awf*Oi6|c=bsS3g7e1_hVqkFIM$eH^8$YZRcV*@9Xz9v$ zDJCzBi=>}x4y<+&sJ`6*^@vCOS?{h)cAagNnBbU+78?(HpT^VS2>WQoYv}#$d9TX| zXeP&0HkcHL1$%z*h9%ZE40Ck9BCa!|9&yvN;QEjpi{((bZvK;}goPTL7Ze1{Ic5U> ztyYqqIy_VFw21{XQMI$c)Im=+e1{x;zo3zWz9c&U0KsQ#4p!!oTcHnZDQF8O-r@v(CL~-4Q z*ySjHHXGyT!ytHDL@eoW)X^#o4e z=0;3=1pGATJ0L8VJW!2r1PfQq2V*C)k*>>;4*oJW05wAo9-KsW1O$v zlnTf;96?_XHhM{YKmH4U2WF+7yb9hATrO}&FmBkwzO5n51q}bLwu^!Y5b0LmOC%BX zKrYnrIj)`6QXb&!OG2fTKfl$E_QPG&ZHiH=BwdQFNR<}ka2O0hvkUE04(#NM^bIx=81xCf^fll!RcPAlks)`Z5KzCd!3s!HM8S-)YSAKLpVA zlpDn0_! zII8TI4pw-!9O4B0^p%lCS{B({rQT`Gznf$xsb62VpX1TmgiWE8}3fI%uS$T z5`)>RI7cKtO7-Y1Ejz;!g?Ok*N`c&icFs|SHqD}Ui4E$Z zgwzXg1oVn!uxR2n@Gc8eu{RNI)EJNN!;Z&x%M*F>jy#@X{jqMNWIOSWP+}^#|9aXK zI2B*imS!9WGcI3-OfDyC^_(Zfi$fm&h_O0w=$^s5EfS9JPz@d#Zk#F1K|9a*PN_p7sG)1Cm6}XEsPeXW@o|j&Q6B#GeSZVqVbDW`3z-t-d99?rLSK> zYLFr+6_Y6Vest$zulvEyt(wC4(FS&Yzjp>ryYAz^prFbx0$`0mi!&J{(FPV7I{_*k zMo90k2!ZY)>yz!t?#L1+I!J?cX^Ov5mp&<0JP85ip<%MWUD7kYhxG)sqI{6n(ChhS zqUnU+Vdoqyh?P}}^{=bFv5o+`2nlHthp;hnn)%HFkD3j3*q0T4T3cs=`HSo#=vsVt z9pQL9XVNkxEe3ehv8|8w#G`7I8syCnfp2-)q;s4dzb z|KZ2=F8&;!n|of!ijTsUG<%*DKh;SzdJ=WS!;=Y)pVUq#3WASHt;Fi>!ohU z5^`Q|8$2hn#hQF^VBrhiE#_=}LVss?;@ma#w1bq8z=h?o^0>kDl4`}J8uUMb+Uq5WF9&x4#; zJ(E&BuM_k@t!1!0pr#uVlsV00pH=~yGR6Zt=&93K1m`79hpByc#xfu}LYf0C?ZOsA zoJ5?C(?^Y8e?Jw)DVN4k zVe4nC*;F=lw5hm|7yPn`vC)7U!}Ir7jZsLmlL&;iPVoU2KD5E6?s~fO?*C)%yZ@(FXz_n7RNc)^BRxGeO=FYeyw^*!TIw+^)w8iT6~(wPqLOEVhvVCytowh z7;uVgMykjkA@~@iyopRMs5E7hk*11k7lJ0ggL;T5yjVX+clP@O0k&{z0k-yljSO4l zXow|_&;U4S*7yQ<5dVqW6+;tD^xc3Afn13UePmlqfV$>q8t~mtNlXP25AIU{e8k#S zO$H7F67);Z29G{%Xt|(p*Ny#RcK$(=^tO$UaW+7$_xs_KI|H=iLhXLpbNZ;*)$=WS z@XyUoA!{c@om~dI&7oirNXF(h3~Ln+FWbjYhTM>(wQH_&oGt~Lzo1#p{m8x(V^H&% zE~ugPfJR3O>u)Y&`qMl7xBg1f#^fV+P`5U`46pypM_l=|{WYSm{_ipNr=#{!pl#uS zc}!RP5oE&);Ya56MEtQQ9`I2|wtYY_EVJIkP!{EVG)*uk(|9Vj*2T31s;bw2Kv6dg zRKlsmL58A0`Ah|w!YY0u!aoWPU~6SZhnr%HL1!6sswKO9cWqGucD6z1 zNr^3TW43nk7(lJ`Y~zS<_q8}>&s<$_v@d6b`>!20w2_NiU_IV;3NXayWqWe3o%EhKaJKNm z&-SgbyUk}1qLn3SpK~@|Yxl4DEj!_vZur<3soI2f0dXa?P&sz%8?E|Ya6{Aiu*n`D z5=go2BdQD$?lf1q$D8X90tk$&Z3J=zd(8t=TQcLLDP|=$T#4YgWc%~l?H?U9(i1+m zhkq(*=YkKSgM>gub5DUqS>Nm$`B{b^-M<1Ji$&>#I{3=_BKzhgdM(Z)B({UeGy2CM zlVl1y|8p1lyQwr~YV5LHWmcppsVCG24|C{C0nzJgQ<2S;vrG9Bv6DdkKvQ_j(HaGC zQlvz-cJcXB0B%9iWYft z4wpMSZM;!4YN&co5!KsOR4vCl4LpTk!ufMoNS`u9A>=SsBCWqRvTnqd0_9n)pj0+I z9dxayl=}*GbPiEo0CBMZ?mxSc!Q_@OrYjC#whQdZnDd=<1poeRp4viJ`?MBjiGFT#XNgUa?Fy&>ByJkhx(USrxB;v zx=I<=TgFLCQ!-geHRd8me{16GpO4>0yfHc-C$?c z)_;dh7>;pUWZ$p@Z$U+eY6PlY-(t;iuoZLET(Ya%|2+))dk)luSgD!J7qESZ0?n@e z4#0;o%gI-t>LEWFn1x&_)$*8P;J$+TdAb%PX8{;Gm*7><UdU$-W-9hXcNzD=6&G2P& z#G(G$;!H&1+LeHkZvd+;CARzoHUe)VOUsy&vLYa?~U<(+t$MX}6NZFD+XmaC9?bJ@uy-nr?Z~N%ic5<1_FQ z%7*(jF;n{#lyq*>o;Q(U{%qyS-$eF#~TS> zVCywtY`DKv$<3;|iIp})hq;fx7_lShGC!aDmUiN@Hnx2@eu{}>J{2@yQNz6A?7!jV zqzQfSclu^t_v0j5f~)~FdcUHX=^<&YK(>&JO7uv~_N(ae_INKfYb*dsl*Z^@f`F*avVi|i#p^M|=0o`63GfzK%K=sq*To`L1--UZM?&sirtdSS2wwCd!% z@2~#Q=gp!*#y3nn0pE~TRSCcj&w{5xtQZXrA(UUNI6oBSfWc7$n1p7xMV$i26CM&Q z#rjxGV6Ar5Qxpz2)qernp8(%D;?X+-iUP$0ooN271dn`81zNxBI7ZoD){m7F*=C{ek;eEWZ~1 zsl#@4kLMNmyHs%PIyW-OLJvdK^6T~NY z@Opzo;kUyn7&3{b65?5b&I#WI45SHZAc5Uc_c!qTC|3Z)!?Mj_tMH*W+bfs;?jTGeiH1H$VXzR8Bo2=8nCC^+zu(vnl)o}$*Wx5}eX9Td2q3m1e8FAAQER-!o<_|O5!DFS@Y zz-kiDMVa=4o7)MnbcJ!Yg^1AfFTKLkKE7 zPhH4S7NBm!v;YyB2%v-uPq}{e4xb8yv7BGI#j4H_=zt}83k$XeAR&zg+)P=OJ*5m7 z4XkKox9%WK8j&=(Dxmk4jN6_d;FJ#5)Bkyl{$%zaG%4llR3p`_D!76&HMq0}g|r@U zWEezG##Hayz3#g&^v$|teJXViJcEvkF}J03Hm+peA}KffLZvx*DK1env}V)(PUV-> z0x$z6y=4btnOyF;D7tX)Q7|s({r8}gC1Ei<$K9kD=u3sZVE6e?1>t%>xD4s{l|Ko7 zG&@gPT8G6=7Z9+CeBWha#Pe>fzOu}?V`HUcX4#?8tn zCO7{cbiydEM&NpyQeaYvSB{*O@~erAG~?$od5jBhFa7z0g35p;s_DtasAD&C(& zhqtWdhmns`ESLj1NRkAOQ=TH7yRkQZW9&O*rA31H4xK)xb&2R^6 zg%`oc`dK!(s!myzLLJkmi_sZrE)3I?)#{%`u5-zv*;<3c&^f}kC+Eo*rtooyks0Rw zdQ3lpn`4Ie6TM^RARYgq?*^ehT+&%?D%e#i>3L z7TCR%NwY8jvp{{-P2FkojiC>e5w0a@KJGS`53pQBDuzODyNs0$aC9glr4`h?{Ry$R zq<%?a@AJ@{9W;wm_HZeMEac!g-+s`zQhgEoi3j-*KTElg9|gT{sD{p>HqV;>80M*B zlGT(BQ@tVlQH_0uV}XkXwR6k`$=cvk^eP?Y#$CmpG{}c!-(5qiJkURga&LC-sOKWZ zU4WM#&+yBg>kz^}|6xG5K1AH5t05@)pTRFvpZ_@5VuPmL&0$C8m_=W7lP&9+Obf|2 z7BEK#I%+}Xsv<9`tfhZ{6#t0+?^>avnYlA_(_(>IY7{w7g4HOnV#m>c`^ulj&UcGT zO%17NmgK-^GeIJ_()d{jZ)7mW-t$RQ+{CONKxxe3p~wx>VJAM5Vp|l0N@AYrH+k0m zD8dgQZE2JZ#+~EEdVKsu#r2d}K(a|Ym`F!m8Z^jBb?)Qm5$@VeUG?A>PEtPr)3Gw0 zb`-EW`z$um)lt$@!OO%pOI;#Xp$hTo4yo4XnM;8-^CZF~*!~;*T~g}}ia&eFzlzpy zF-U!){X~iiBxgaGU#W=`GnfHATxXO^eeJq>IQ|9(VBq5vv8HQ^w$Itx8E6gW>4jf+ zyQXNN)(+VC@#)hs*7E|wF>Nw&{6@hf=AW^$vA3~MzP94KpSxcEXY&tM`S(^rfW$Kz z5(_y%afo8$&2}ULZ*;j`3WeSuD`OaepEq@}ti9If80kopx4sM^SnB%e3#yNpPTY&E z;2AM$SPlrD7nAS{iheMuK=oE?#Th`#f$r0coizUE{(iSff&_h_>W6}g2hrSbj~+h{ zrZ}`MUK@8a4Lx1g3Dc%yVA1Dedn@Py4Npb(5^zi?h^p4SDloCMkA_@mA}CY1FviPS z@1tQj3QcM0--+g>#ExjvP`ipmJ@QB0W*~|k) zk^Rp?4T&T+);3s0Aqxi=t``0&IMsdwW~%7g3aA8R$%v9y;St61iLs$Cv3r(?ck z+w(O@G1oI%H`>Zo804d?bvf_G+>^1gbgEdGZ+&W#i(9C#wKbK_YeuX47W$Yn#@=2VZIWGc3do?hX)K2Ar%AhF&_6Ptu62~I z2-A0;UJ$Mh?qgYB20hv2ew$E#B+345SLH{(Ap-F0j-oX_8EG-JO-lH+Q2>!W zm6mF#o_{+ph%Ic;w@bFGIH!?d9^ULj8Ed!}+UdF0+aXcHj`~D#00rc?>PDc)MY(Ek z&X)v7rj<&fgM>R-ArfRxU56W$q|H*DbLrj;11^njBeSBs~Sv2!1n$!%iFKT?ne3bD>ABIit-vPT!deG~^4V$fY zR7U6EDW*SAkZYZRtwTarW(EG~{mF5BGvEF~+3j3vE!~@Ood(K%)z}0x76&wk*p0Ka zx8WDLgWgJ*G~sUJ3UPh3Ei$8Ww*e4;#sCX^r!*QRWnD7xi|dHmb6WROqShHDimUPG zD9}`@I)F{mkM0rc40U5dF`}Ryf*{W86X6@f0#}yF8K~AH>I{2A^tf;%f}@hS1TYbBd9isyDZg-0gY9by)o1X~ z65Df>+z^$N)^v(ll}{tuqJL*bQzpp)wDhv-9T2Jk6BiP9s67ZkL{G^1o;x?zNFR7} zxwW^ICa<`wx^3EJ9^#lDYcgaJkr9m(qf^vt{lG9+RxYI~CrrIZ>_QKT@KY+Vl?($X zF&uguInVQK!UGoCYufHZ!m#g8PtR~o7)^D+3OOH>h&X!H!qc13_SA$A6=*GJi@68+ zg+r&a8$3z5blOG^eFR*GDa9V#ZkJ-MeYQ-8W)o!Ryq0%0f}4yFHD}1n7EXaUB7kL8 zI3W0nKR(n5Uc~VC`ODHCyfdJm1{P{srzz0J;t>lz(?qCctvrl9Ir9yA__6`RenQ>z zgULQzdnCEA=2V$^`QJ7?bp>Exoxl~nhBVDKNR;;VQZ?c_l~}1)jv%5l3a2*wqp=}^ z15{i2nIU2TmqY#o9w;X=`iLt8Rgoop1bDj}9Upz-+um?T2B-n?*7LOoE+BBi5}0be z=9?lV(x-(TOwRc2?SH&UzH2VGTbVjz`z`6&?q-J%Dd_CyiQQ>8P{O9e#psGb<4VfS zxmKwBb04q+PnFaTL1l#-F$ga}HihEYvV7fR#!4ya10P-gQ?84Y{z7w6+qy! zs+3?rQ|`aBeh2M%T0BEmc6E(Lr3P;N$aDLOB*w-3 zruM&*Pnp-_JQ8KyUr0;+w!fU(q#-k$#8$Ox`QrNijg(GU!%(O%wN;@mBJom_4`twE z>@m-HI}Oj@f+84yL-Lu~z7!uXF(_e}Sd}6tOibzgCM>Zh?zMb=?2jyj@0O1xuK2JvIS6|-6?GXSuoZsPeutK5m>niW~wINa&=pDO->b&SyqYlIDNo!jKe zymPc$i(AOPyiBGPQ4`|4={wsccz;VUyvfHpWOh+7nQ0>+yNd;_#7}StSGc+*?4sr( z{dbe^lThexdK`4_f@WO0H+$oIfu~OBMGoz2YFDhXY5Cyayq{U#J8_ovoT?Y&6I(-6 z$KDv%tS~|2jKgDm>RuF@!+c7Z`M9tgd}Be*)lv*~=bBPCiv5MV{Goi0Ku@MD4N;qBlL&X@i~5TfU~z4+Pnyt?HvGm| zt9A-oywZ_YT?|bc;%L|A62}Re&th$KG{|@(!97+Et>IsVWm?gY!2r~R`*vZ*O8Raz zQ}RiooRKkuV(<&@f3VzM`Y3vsMCtMJWAsHpHa#<4phms_!$sbr3Fum#wJk|t3ttAr z2X$)@jOCRbkfxee2U4_LYwFRo6k>0^gq}C4l=zWUP_Joil_a4{GQVeSfQm;oJ!MGU zf^&1{BLH-6%2YO+Jbq+_BsH(5t1&aBa{-weFHmUDm9h?!NBxnzrHo9ITAJZoE&KHm&t zSO00jM-+8oaTW!HS~z9{5Rm~MCVbF!u@88Ib$L7k!gYnbEI47q`iCb`M?=#QKJh~4 z3S01b1%r*0INC!2-Z=V2vbhoIe@;Dtvm7%-lw3pGBtUZs~5UMIk z-K58!tgkIZlHQwC_H0!7Q6^$N>_mwi<3^4;7t3uXKS0FoaVZ`nbPp=Yk)Y=t&P$P0=ZRmG$M8TlC7nSgk zPj(g5uNer=;kY_izHABZ31Zjo4AQuTk zbNOZ6l(BeMjGpLz_scqoah>Uu#@eJfPmr5q;d!`2ZBC&{@eZo-cDm#Rk+hoT3g-}; zd?OZ4y!VTF1L7G{LNaCPBxA1jec}iZy#bEd`qzMbi(;Aw?&j+0dkTLZ!6Lzz5=C=? z+qf;wJW_^l1dyH$8#6QK@PU1f!bCcuO{+bm+`eYLB|{T#Z~*?WB#zN{h|4%yZ~(@!F-_zI5NwdzP>Nh zl75z}?dLRm*x=clXIh6_?In*w=JvZC3VEM(M+`Fku|`O6*ru%4EtjpUN^`Dqa9?4w zM?c$r&E3*vuSq_5fS%D`Wci?dL29sd?&9t463Yfm_b#QnrcyXnRZET_-h5YO>f_GI zQIsor3ZjGWxu|==KXo$`s?Fe86ib7Pzd1Vu=0lVPxztNgGzgnRd63tp3{fo$dsb%6 zKu`^zxl^-J8Z|9Ug2Ap$n8c}AOAq$DcF*`ya4ZpF&cw%lzV~N^i@^jtV_gCaG( zOz8ZU)_2-x7|>ENsw+3tjsecY1m#vosG0$0OMSlFx}O5a3+FOzYC3b~6E%*O?pr-6 zS$tC-S?ulL*h_8eqyB3A;gv*!pxBoD;0)dg6vRsoCxA6zjc2jeA%-Nqzb$F{;yMT& z#XdumZ=$o3odEED?y7J>Hy6$BWOd9oOB|Oh7S1;=AkiKef}CNrY3!w}C3#ycJ>eT< zG6b><8uFbse0<-vfxsgAjpQS4qm?N7??g&-u^w}{%+OB(e>uql#|iP(MKgC^YuyyJ zxmb|$Pl4gasi`O5(i$8Z(lO5r62GipfQ1W{VKjL4ovoJ&c^3e&1%6^(AoC~g7U+UB zD^%@+i)qaTEgbngAyC74?okMM=gZ+f!BO??Eb~m7{#Je{hO@}ghP=f%-P`9JO|qOp z2#W<#Y_ASkN18(Rlbjb~fjEiJ9&SHcayF-PVZW)O6u|xjyF3v?tT7-G3Mt*z{n$KI{ zKA%T)5n3ueYVvu4LrpkC6cg4uS9+g4ziGx-{?F^p%ud z&wmm>pN$<Gp%&|J7kX}FOSC*)?U50q7Q^f`V~=V>+tvR8%TVhq{w6!(so&_KSC9>a>a^ z`S$?>J7v;vbr3iw0U2QGfV=HIqcv65;gmH+zQa%?%PbwW`KOd&oxNX!9}N+7*^toX zurKK}vq~pnN%W<+B1p(Lc3MGg^f9z=MWEb7nj`8iYbci-GSJKcM?#J|4M{FXKK3VJ zXt_f_HD}P?6xI45{FO9_5H}x1PtSaf#qB--^%T}FMXYBE19c?I@1vQl>S8g;v@kbh z@p-|^yeFH6qoTK$O(t+j+wfQP2uCrW08@C!x$|yP~E?qvvLn~nC&eG z`m$IOi3~HZ=a&s7e=mn{Zx4jjsrl(kf9DwE;UG-C^4ah}1T{{(q8@f0I=Z2$>*H3YS4}}8;)@$ zH$*p3P$jq)eH9HP`$86(0h=dG79#7vm!#PTc61Zc7SanVjsE1}jfpT{e9Pwaxeska zlt{|PFlK-=-;5E>3bD32dv1Sa8xlOt05>n8hRYB+yExuh33Xl@OT}|g>lwC|=P~*( z=9%E`j#}-wIYn>oNAukC9N3ObG?uelsfVbw3MiexkKztZk6Ox*F%+o+Ie_~78KP?X zU#iTF%f-n&&wq|bu8rVxR^si#Xb$h_3%zOCP?klkMnf^Q=&o>VB!R92-8B2PIzIa( z_*Cw6Q?9X&Lc^%+l*sPY-bK%HDcEy+6)BD4DYQ^D;;&_SE$~Ghm~lO zL(X^BS^Xv!x{a?!T-sY%hVH}#^1$IHIq0$SHU=vK$Ep6I?lDR%xvC$Rs?rUP6M)A`Le@3P=9|T4PfK=MB;#{Z1xpx{%>X z`o%K~Tb9Z-GuL}SFEtV4GrSM?q32&{W?-6gz(~xl%gg^XXz(XU#=}C4&ireB0C!D| zW`yDjoCR}N4Aa??A_=O;D&f;(a)6nJjG3jQJ}9au6{Zu8CW#E0L= z;(5Oxrym}7JE&0QO@~i{V1MXszit&yF+rR$p5+(eics)vwJ~#bRnASr`HP^Hk*|3a zT^!}K@hO5%v7Aao@LvD|I1BqA8IV+i-!9aBeTaxBJTCFh zDcvuwDEbulN471OnlOr=F2mB6W`Mi8xhPG5QcZGKnBII(ex7_krcUx)P zMDEFRwo1z+uuQx_05UmtZ_4DHW0Oa4i>YCUV`0>x57sp>Q7-G!Z#&_`L`-2s58LYO zw=4guwj6ILpt(&a8kHe6O5hZ=fWku8eEbu5KWA~_?F`(aLm&1RSlRk7C0K#sA>*;zLcSglv_f5wX;n>|xrp zM)}c>1~)la61L3lUfAaF@^ucT($dyct)CW(Wg?2`Co)2AP<^ZyV%}XH|Kgq4S?{cp zI8!fS*4Cs<4GTwNr}m}O&{j!x7v>q@>gy?*s-_`LJku8~yyD@N8kF65+3Pd19ZA#G zl;$GwCKqW9j5r_R(3`Hl|%?gGL9sP4RnPdj_blX?fb&3)YQl>QUq5SwWJ~m}I?;URJ&J zG?$;A_1-%SY91Adm15nAn4Qh1|9OncDRg}xnQda>+%$Lmm=bgcyN~ko^_H9uQ9kC) z9&9>Rae5aOws~B=%-iIgR6|fE4R1sqm@YVne=)0154UUpp^jY+7#r0el~ASc6PWd+ z_u0+4wM>T0&lB#BMyyqSITmRO(EbiI4~XzN4{wga>pL7v-+ zvRooFqR+V$Ipv5AHDay3V(mJ{U-mllw)hTeJb$-9XT7ZV5bX|rT^<+uD)yOv+1 z+Ct|>jz>>{Z^n8DAbe0Ie)OeBUi-7dfSLBQ&aZ{s=^OEpbs6qB<^FD5K2h0zXJ69d2nu=XGZCIq98MZ|Mr^yxUi-u{vta%w- zpgX~svS9i-6G$%p0U{2XUDIl1SYdEa?#Qb7uyX&|xNk%e{8T+15i+3CxMFsZ`JEo; zI-CS)%Uis4aoyBo7s#3{L6QTF%@+{(UP_ev%$#%G0`Cm%F&;?O!=eEQ@!)@(GRhC@ ztN5TD#JK|RQ>h&q6 z3yHnxLx6CVfzUaN@x2zQQX>1h^y~A>7X&h9Ql%{Q+V!#{?uF53pZ_#C6DtA&1=X|w ztUEYJ@=-a0GOx&14(a^z_T;VnxxZuN**bOgm z$~al*-;Q<}16mIP(k_VOMO!3lG0zszY~7K7!<>*U$Pc;#I@-YI-yY@dp059`rT)FJ zf?rLDoP%71S*&6FwZmt@#Z#0T-KG;@X)ckj>ctTCNTvp|@j27dgHziJfDoeqEc7LE z;Q(-5ig0t#;Jf)>6p+jD+)eW2kV~*!9-@D2Ezduu7nuK9DgX$@5rH$rB&*Raiw_)4ZYa#yV})0!6NSG!rsK7{icQX95#Ow7x;OM`2jK>zV9_ptaP zx5lY9O;gu)962Web2!}7{pe&ZfZ2#*mj@xuH2F#HXp)w@NG6b662A{5F0|{Vq*H2g z2Yu1G!csTt9H?z_m!;)?7ZoUhn&b9gnEih)!+-C#mtsU^|NZa(`~L+({vTQPzXm+> z;{Ud=iSH8p>VKQVe}C=fU%BLeBxC=6cHrdv|NIXL>m@1GiGa^5e4!h(>C0d#cf}hz zeO*p8kNgU*6iP(UXC8trrpQ9>ctLB^+n@r_+SVYA%IM@D;p5g_&`hN+gN>^D8Gv$Z z-$q<(SB>Hnk5uLbtu##P6mKV}bmT72*PN_Jc^iK0 z6kzVVnl2nji@+1JSchY^FiAv3RL#pU6q@&W-zQ5nV%B5r`1&K+S}GhM zrm=-5jaipj9{xkDgW4)C2Izm^%Li=(w@}jQhi)JxE(O-M?!hGGpi$u0`TmY!1%%+h zQk^4w90vhgk#?HgN<0KIZfXVjlA?CneJYax_;zViSCSTj`ti93=&uvE44bbL&4j;C zQT2OIUf4^=GtaK>T2L z#gJLpQCj^&0=T}JzQTQXB~=!1w$$E-U~38{Bl!f~Xv{Y#+mLOZNZR_eRM_!i)Unm0 z2aT-yM6^MA8$?%kw9gAhbK8Gf$T>15Z@X#PU!8}W908!!HRF>DVlMN&%RzES8R1+% zl?_}+mkI{vk|eBm1PsK!SNcX0FG5Y|4k|@ie*iXsIe*o@5>HT0ja!Ixe7={S3N~6&E5PBicI72nN)nt=)9Vu>`IAE~sr;pMYP0w5CE+uS{Y)^zTaHRwndr0xFx{W7LokO}rl}OR%{->DE)euy_a(#KMR$_Z9$>4W!0WZVt1PA&qJ=6O%n3&Tb(0HkHd~&`p~oa zt=H>B25v11Y=2j*-AetD>Fxl~`j?mIh*jJ;QS0!Dk0_P&fKImYNsT%v^?L8ASgo@I zKUeC{8okEzwJH;{m8rWc?dfHit$X516-|du1HMC6X=JsGq84V8zCmJWUTtqNC)?cn zIfZTeb7|u?PJa55Chs~aIK7x>L+%2GERT-7Q;gJF@CmHm+yo)ybC*qdx8u+GH?Mn+ zOMjEzVoD@;JJTV4U#JX^5z7`*FhFw3%gJqsei+z_b40h?>~pr2ojuNpd=Z46@GM7j_$9;LCt^O73|uEyTN6(*FAGlZ%^+^=!^iaWWa1Y;68r>g=O2dzP&@aRS z@{ZKIw*&W~&B8>4vF_3*h9A=L;@?YUi6e+2dPM5JleFAx<8z75JY5Ux^b@vI`2B27 zvjv!rM8_ASQ=>O|T_(gQdKpYv6zpwPl4^lxl-oFyrsvSMgYx@=R(>DH$DziSFQ{^rk@$TBWxBR}2>>axx z)h6R!hb^#W!&-Pi@N`>LupUDMqj~hRtF(2D19IUV^SqVYal6T=_R8Eb^t4CvoNTj0 z!H3MROp2XK(EEyU79p47+bVw(&KwY>o+f)2aJ%UPH$h(VfK4C;gi}9A4MZ&W=U7K` zm193Rhqnh(v*~^JOEzEkI9v8^+6#d=eP~1N1=!k|haA0C64G}E>Jh+oJ)EcgcEg*a zxLlQ<;+a6Y*9&SzUcX0IEx)DFHmP~NDq7b!cv;$&)!cn;%46}Sa9UN5l80l9X3dWl zf|z#?_DX0fSwN4G{nGniT6s{Loqe_ADVZm;89k84S!n5&{n3P{A7(!{iwY{f83el? z*o;8)fd!9ZUyu55W#3s4hYdf7$$XG{^R<1ggE5=v?u}-Htv~O!YbO##2KQ1*o(&%U z30YX%{(wJPYbgPy4dc4+Nr}7bsUg1T_uE~J;PL*9*bTj)5MOaXoAIbu?CH@d;$BWW%I^U zl>>8$QUi2aZ@w+G;HJiac8F+uoghonQjM4Vg!fI=vfX!lQLTrN!K|&a3y-oRokY8N z5jy*NP(1LPzeqaFceL4e3$fw%wyV4MO-4lJ`Floj>7K{ammK7@J~|cLV>)5w!n9xd zVW%i86v=4%Eyk@?$3~0b6V|l~Xk(%AMa@q4si@tly3sHBbwIU{>8ctI*G?djEae|x zeBlfx&&F?^~CLAJeD+qv@DZ9!OOwIB@!OC8c0+Wcmg; zQs=DW_!$Sm|0E(rdO%pZkTYp%wPQZe#+#l3rUKP0>v^zM{KW`V-`c<}rj=WzEMC&wr>^xtv7MTeJt*Fnqjr^o=h8g4<49Lus>W4=6{Z--H}~u)w#2p z=Y3*`aNAvZw8xpXk*?>uxBJ>-eJ+Ai7A5z9qq}Ty9gjWwtbFi(vPYEio#eq*mf}a$ zY{#2&eUH>jg~a|;?kCCY&7@(hj&JGipxF5#X!t@!6(LE*guZKUYj2vJ5Irwe;q^u|{W z65X$6>bilrQr(vZD7%t&s{jyFT|p^@^(oqR?omzUY< zJEn4$D)N|HuMJ{6_}}0@*>5J5(k-k9?%9PLTV0XWu?c~u27 zv`5kOwN__eTatq4l`4M~7p|24OZi!z^_iM`(ECQ+5vW9aQ8pHj49{)12OsISMKx|R zmz?Tm(0g`GZ%SE?_EHqQ>ROhGXt#>UYHpqU_A0cey-mdllkp~8Pmi-`aH4(b;N#in z1Fe1UQE%FvlV4iO_muQ#a{5F_yrvb@S-3&Q#ylP(-l8IQ=QsYrtm@dXbf^o7G0hQw z;e+RrQ)yo`t8XVFxxct^o}fN#jvV{nJr2bl2bS&(e6WB0jvVZhtc{h{{&5O48;Mdr z8&1}j3st{N8ew+-XxEmuW-7L?rB;-xlTUdV^A+?eC|l~(U(fp){tev3_awFRZQ~u=*n*{tVOtQ75mYd@ACEz zXgs-gos!0J(-!@fIj+WUnyeE}_ghAFEj7wNE=EVTH*sX|!KER$$!I&#JNq;a{;()n zPKeFSgDDL`O^p4Gus?7<5zM%b&H8rdU?Y~3BX{)Mi_n9E)d0T!88UM@in%4_kEJDZ z^ysXfG=_26{)WxNEr&1St`i-NY18Bsx-H@bclSK`uiIRv>M<+YetvDLBe4FEr6YP8 zb7e7g*X)jqoU!$}cQjtcg;~B&d2J8f_qyFzk}eOreUZh`oaUcAoL+MG-x{vcncJV+ z!7&?}Pfv_RywD?d5SAIgZc*=y_@L!Khc4fZ(1RQH1&&DPqi>GbM1AKKCFyx$>3Oz? za~AT`%$rs*?m3Sj%WLc~Vm3N1D938uN$-P2pWaui9V~3S-O93mYAGo}Dy(18fWcch zS6$?|Wr@;A&ChMdk&SuKz_q9k^*{)5E@D6%|0_**%T};=WO*}sCphBUoRWynzP`UG zTR)E9c;ge^D=O1z)Ag3?Fp6z`VQpNr{Ri!{Lj?x}x$i;8p*rin>fx$?4_kLx>y4s- z@)4J4DHY3U*`u7|ebu#tc&WB01MZ(em+^CNCp>z?kg{B)oLdZ}4(d2w{TVcCsr=J= z1mE}LD(g%CIfcM7JIuPO!AFYmXR616#~94X{jd*4HHLRCn(OO`o6yfl%-*co@n}!R zzW>$nx1M+~{HCgQEIKo(Jj}9j8X|czcx(AL5nn6XzmK(fNtvq3Rpm$u|jQLwLWPa}R&`P5VLrSuLA07HB z4^e{zTfy1sOx>7Sg{n)*0Lr{iWWi>=PiW6#9viVT%whlMYB z$HjC_i7CD+9IO%MluJjv=6pUfxb^#IR{C4*MQst|UiUF6Bw-tbd}i zw2hL|ReM`^19RS9cF6v$O>wFHsMgy4ov^!@oC()uPYOX-&e~@D^pcTwwYJdA?=pz>D2PLM~cQ+lg@Xd(b*+{SUS~G#mY=?DBvawwVMG;R8aQ0C2 zVLaSbG|&nzW_G3S>X}Rkg?G5U=UODbGCX@EnS% zwf%>DM)<*s5B_)e-I632-+%4hL9c)jX7}Xk30>Q`dztQKjAKykwVY<74h)s6g=J!9 zy{ug=FMhxkGqhWe&`_w|Jg~mZrNLZF{tlgYkMqQHd*gCO{_^Vhpy~k$`s9}oC`jBM z2{69@E@=ECt9fck*wHm9#VNbf7X40sICqQuG-5W1b$=)B7Ww$tgK-17VV>=--Bn}l zyq#Ngnb(ObvYYbLcT|Tm92K}q$`u#ZQoW4mHmI-Lh`7^E?fB{UeKb1rv1gGCAzH6m z6a6i}cQyc@*)1zKA4{nxpz++$d6eASd&gi|Op(%9&4Xp{)BS5A82jXcAbNj^eod6E z$8uR*0fR{oufC3=7k%BRM{9Wx?!tD~=hu5LW9}50tt|P7#s=Rn*MUBt_#!)?qOs^z zXLO-9w$70Lx9;{E@QXvAI%JOzV;fqh$D`sLMQ&1zKWyjzK0Gza?(W&QE0XPShi^vI zb!CJeOhq#VC2O=D*Pk-`GLQ6vT_;N+SYge|*@+6^fZ8s5rTY4!214YWHAkGYF7dsIPKx{Ikj zs%#Bw@4zb+XNC7th~2VWI`ipOS8V3egA99O%|z|t90t1;tq?8J;g#SyqMny4617)X zDovrqX1dF~Eun|I)>FCq{p_FEptUcM3OYWolB+gh)zEPUfQ6Bp$w zT04F_8Z;I?vAOY;t%B1jLM!TZZr=C_wgy2!ls)&G-qF`Cwp@+M7BGh~Ko&G+-*(|j zXD9j1_fTrZuJ@_OY2{Nh^7jgdw&;vwXqMJjjkLLbG1YOu+YD8F!rqebvEXIf;ZD-f z4(iOd@V&yO#%Y`Ozy_b7q!|A&faO-)E96 zEzaT0iB|-yPQNNDqG-Nuo7M2Oj2xrK$^NvU2gf9fEysF~{80 zzoyz^NUq-ThHG>t{_rEE+#&9c-a_%7(}v~wk~zB2Ma6=ewcX8A`t**LM_R>>=cBy6 zbq3zE)j{|f@6d#8TsS%tdS72eaJB(jcbHXeOP)Zi*{{qCzrFST#|zh0YsP^zOpNu8BFgyG_?!rT;I`h(Om8^|B}Kf}0lCD-mfGeXL-_X_>XxML6Z6-z$+H_j;E7q84oD~{Knh#Rul zUe^$K^?WhFy0WLU!XD>2lKN!s&pKS4ckq1Ls?epqnoA--d~oG^9va!rgJp3i^{#E# zN5AcSJUr0CJXDVzB&!jNC}Q~D-hx+K%d%+NTZv)9kLLy*zWxd^<~(5+WzG$^^hjbf z?TDEGO6{zOIU%aKktNU6h`mzoHLn)&nX0HTx4G84VJmY>!i3!rAJO~$CTR(F-n4Rj zO|rVP4^}H%TPM@LC%i6Nr60c^i|Ub+T1_wb@J|_4&Zg*c2_v2>PQ`t*6yPn^K;7PX zrF!PoGN8=5eoI~%W>?^s`$=Hu7C4+0^-el{iqB2+LN(dAf&`pWz?!)~s{(Ly**!gj zOW?=hfsEvmhB!2GIuYl=I^8#Q2|mJ?om2=^}>IeSoj0Y%hKnx(;IB za3|B-4%%|yI{B}arChtuK;~7(mnJCHC#BJ-+dnz>f$(&rps>uKUeg=H16VNtB;^`;c2tmnw%`HnaU_v+EY+gd!t7f_hdn0WyH@$Z zj?ZXhxAVAH;f!p3g=qOLV%XeC(KXDos@1wf#m~y4Tfc~@eZwG*_B9|3%=TFfUI0q% z_d}cLr;i8czx^@`6Hzo7sdNVFQq{Wea3GuWs{5{>ZOr?p{QGMiLi?j|j+lg)J$9yz z;JOZqQpWHv$$V;q)m71D|Fa_&GMk+R$;vjVS=q zuWlXl{sDQ6U8jt=qmeJ(r?-#kXEs6h--h>QI*saHVGOigqT$*?@$BNEBD&yU2ZR<; zfjJ25%n#prL6R|t1%=Sn4;;KxtnQ*y^qHS|!9jkWE5G@6-4hN^3LD)V&9A3j(*T_9 zseCg>B&7k;$5lgHCox2GB-&0C0i=??;>msr@~;%C%DE{c=uzHb^1zFp$4ECea%;H! zjwmY`_l0bfUoUQ=t=75+&oNcRhYk73WkFOirb5Z-6Bm}}3tRMfRm11+ismU1ak!Hr zl;PZxgLd(%hme8-;juGFEgb$o)t%>8liRk&MXGd^Dg-GaEl5*QS_Gu26oZ=*q#K%G z1Ox&EA)tVCqzO_tNEHyINDClUs`RErdM`l<9qx*IoN?~GXWT#F<^x~yC5*A&cg^|C z=lRX0{w%uHo5%%f9(&<61vem)8|etsveBNBV!-Fawv*vk_i{*6mg#EJIKsm!g2!4* zC0H`mYig;$en3rdj|`I`II;?xYAn!VGG?`sdHw3B4Dyzx{l_>-2Bt96#XC#Qb7i+*2BPYlC&)YYL4JF}sW+tQm$BxV}bH>Z!!UMMaCY|z(T z53}fHV09$|DKU_m>8g@dfC{InU)b4ARITODh$gKSdGcrXd9<7O-fVhq2g1Sm_r%5L zaQWV&;lH!mL?cMN_m^xkC9bEOW&JJd(&UA7c6zc8^d5j7{gPNpul#f(Ee?BTJu{Wn zq$|S?Ow#Ul_LD#sxM8f^nghpuf5O2VCcq==>gz27WJo^z%FoPT-n$nalmNeep9_MX zz0S11lzBRtkbdb&EufVs#M>1Pz@1ZqYz#-rWE_SfI9P{dC2ihs$9S*eBM@L!99oLK zEW1sa>mA>v4sO)thmszT&Up=X`H>g{I?bit1lMf2Jo2<<(n$W){pb7vD}V{O3PY$< zrQ`eJ#1Eo`{wm>Wuh?lwh(ToGd0`v=y{}?Mw1rbypHG7+LZty-Wkq56&~a_rAaP71)}8QpgkKy3hjaP_Wr_#N-crHwx_JxxX>`4+(YFApu5z(3KR94*2YO?zgk z&v&Y3S!n2!2r2T{ZM(NbybWQ|+xdFFVt52PgVWn;8W?@G8E+ah3)Jz|0vp1DsR=Qu z4|4^-lF=cMqF708P{ht?rIP#VNzpg06J@Y6Q03X3w@)-0pHf)#6^I%DA+fJYKI4U_ zdS1GS98DZc+`-M}f*)gy6dBzH=-CGr90pS2&23S)Bly+{^LK^)SSrmq-dDtD=x3k} zlsP$b>bOWdky3H-=C`)*>ax`7iExXQ%#j5$yMZ;x%uaDy0To z!9u%vQ%FXzU^@T1T={pCI-Vt2W(*}9G9QiZ)F1OMb#OA7yp_W8PJ{NFmr}qbLz2p( zciq3H*HZJhZHMgs!CXV%h#(n0wruk*mv7;4+6C|&v8pw&QqdCL-EVk!Hzy?atSLj9 zQ&D&g;I(>QENNtJgkKxyqny2VV*>!4*cjJrIxHP>HM2AvhQ|VTQ(mpDGL+L)geN1K zzTBs?xr0V!rKfqT7~3Wx1C+lneSTEG=WX~nnvq0%1@B<-cK%7SwxcCi7jV#GOUWg7 z;|b8em~egq$P5awDbfQR_L4B(6!|f4?2&@U+SrkW9j02qB6sDwm5LP5d@)aW!^y0- z+Qj6Bd?o@M?RqmA_xC(xzd9W6j51BNO5GGcl;HTg@XyN*Ehxqs`8Q+n z(1%*bv^Ohc*pj3sI(%l@7jY;}v1n83V!4CLGJ+G*Ct4>YA8=OHHRh};ZUd;b`wfix zmy7uOLBcK>lWW?T0VZTW1-R1e7I;l3BvHVr9w-j5rH!=))OSk~qE? zY(nOpCDyD!bmD6jf^JvXV(U#1+)&SaOTr3Nc#djk=3!rK#YJU$)s`y0CA~RoW6(8< zBUWGvIafM?fxbV#S3pZPtaYZVq9ZS)Zil?F0{3L+eAmlOs-}?=duDf#VP5bV7^`V_ z${JF*V6~uC$I+yUllpArod2iG9)?zZcikURB+aTYg|g*6=Diun0d;x1+CwEN-ZEDucMrh&={fCb&J8?# zk24{KKXbi*L}EENoZZh)F1E@A+0DFzQ|#NhPdM_vF#V^qfsu|MYZvl8y>-}JnMP&= zpKce}K_NGKNJnVZ@ZzH=e&mkq@b(k z_Fj<9DKp`@O1eDUn@b(zmi?A!E!W!i?pH`tx5<4|3VnWN(x@42*vX~+@--XhcQ#*8 zJ7vv&0IZkP-^~0q;=ubMKz8+~vnC7Vsd3ccH)E?ylr1Ga_sFf4gx-II=;Sjiq#w~c zQzSTc$?xqX))Ta8FRDw@5FT*~k)0)9o*u8(oCEZajw*I#eSMy z%`^iZ5y;LpU|;Jyb-%5d$413m^Oaouc8uhUoIpKIajo~1a(Q%O63-EnKAF){v+W+m zIa`}#s;q&A5j2^25Y*W&ViciEyVa)dtFB+=9=9ktc*|Up#^2=@C5~avXB?|i4EDXF z-d=aIvV0kN_RzO9(@8VP!aN8ISG?l2*JgsI+PFXs(Gtd!DOjpNpsN+HO$0q6TrDjJ z6YC{M0d4gC;HNjdD>V{ECD9dUZHs=Hx8BC>H+AB{T6IuS7-nvzs=Z(yRjV)E4Z|PN z>_4XV=DW^Ja?zHD#_9TFm>I;aZ>LS6XCoG4$zxCj*wT1TDD{mVwTZAo_6NE z^6nSuz#48o<`KQ>6CAF3WbVOQAzBN+XJi(i&Hk@C={HLpb`=A;O1r+|X4#$7KVR&PUxbei1+`XugD+3E_ht6t{3;lydNSJRx*CK#0&cuN(5 z)NPJctP<&VSQB%jsNE+XepYg0jRJuH5T2XY44bY2faJ14_hD5M`AU4T$!)@I;o1nm zY;QRc>wv*zUNbrHXywnSt&1YfC(=KreSm9Q!CmKpmeMDf@T(HRmS3Q#TF_Lmzbv85 zXd!n*pz&H>B|m)~TUK@G{PUKg+snMUY6>4eKd5?z&*5$F{88R?YEj2wq(!j7BCE7! zf9mFh4Av(B>ka>1PjE>AP0QC`p^U|SXD`{SGjZUT%71&8J@s%`xi+UgU zuofZipQvYl=}86F9F_G0N?xWD^F59>yJvPjOdu)Z+W49UbAua?t8+x63+xgptnP@@Z}7#*b@dT>dza3v3>D-v30?v#>` z$x{cWfB9CczzDkIkG#I&+q*+za(*dNAZU4HPrZYyN?FB^BIenE)Bqj+{UTs=-dBE& zKt@hA=M$E1gclo3Bu!$aF$yxkHB+!&qLgTf^nW;Mq-Scb_ijn8U~^O!(36v8j9R+R zd35_PmbZrpze=n}NE4@#kO`d$f@AiamQN4Zr>p=Gzvw9tCa;| zYky4wxynB}@hjQ^UK8V|_qlgwRASj&0G`Lk z+@$_*!=yV$U`AK`A%1G;7%p>-fW}nA`$GB5xuHk9n|h%d4N8hkn<>u zls?N`n5198W550%B=-}oW!qMYhS;gmyIT=XABT26z2t7~W50ajC`*Oq@^_=y0F`!j zKKmn20R9fH5Q^?Y9!5`4AoRa)jljGi*yc6>Gj;*d}y{s{;d}JdZI`Rm`K4p>a z=|WYRT2H17@o3Xby5axixC$^LEz*2org~ZivYVfP9Q1(wvGwx+?}Uo?47m~(pe))} z>X}yhJ>!yZ$Q?4uPv`8=5A~%hV{rE7GPzXEF`i$jk|b>t$2EbpvUPZnqmwdkY*`OUzUinWHah+|p|xIT>{HS+l?RBS!*ff*EGrp0 zn)SRdGI@Oq$9GLY>*9_{7d=K`CJoS%!r#QW&gkj%{EnZQFF5aH5=xjAEmsgV!-5hj zbNxws=edYyV3WWC@OeBH5ad`^oD<&;riA!c!e%BnNDOiO3cVLk;>|1NHlmt}UI?y) z^6bvx+*V09yh!-#@w(UOsQd~M#6%4{XjC|IrW86SSG!2oYuqs6Gi*N~C8T|>%xfAi zH?UeiqtRE!u2_9tHHKOO58FJP76h0%UEFZf|0<>YKZm6xOP>q!}Z1V=}{oKSneq1HeD@W z+F|cNo*n6Eao)AsF{MJqD{0=!>za6LH=XqPYFm?=n;*)|dHQX=Q@qInO{}?u=%sAY-y_B@kaxU zc;W%ey!gGnUGt*zMCg`Rbim&Ft^As9-vtq#>6>8jZshQxLJZ+s}dAs9*2m5 zxn&=J_Q##B+(6%a&$aAW-IeF&o#J;`cK*4MqfmX;8IbIVDC}khEO-;=%mH^H=#;Qq zbGIKKxew3zi-fh@Hic_E-KsgaF0(DvLL95aQb?|Dr4VmA=rp+#>kK1&pTPhxeJyHU z{bY|AAMIG^_4KqmE?8pnAr1}s#w(beCj@ip8|&j$ODeyguB-%`z?KB7yEi8adD;G= zHo08!hIZ6MD;<`rB}Z8F!qiWjcr9>p>#rwW<0mI)FqlXo$+MSbslqi zlD)NChUeX|ndQSDeS&+g@5I@F5#t4*7ttJsZMplqRz>Sgx#3CP%q-RCdIJ`ZK*I@& zmVf_g5qM>hQ0ufszBt91alcWD5BXtcmWa+GQ%yqVeeUI6eMl#&=74JK59wEvjkXXK z%s^lli5hU0e_!?Lot7=!wAbGxAlfue&@bzL?_f_`mLrm#J_gCe#9>Y8i6~D=({xEQ9RIHShH0Qe&JV2R6v_&B^(pY^_T2-ywcEhy!&$qr4c`T@pTsprcD0n^4 zt3D5ueQ<3c@M%NYH)jO02kZx}Ch1np`)k<l4QwR;)-KjzCkW{8!IDt}s>lhJLWQ4bx8l_8w% zD*9AzCXsgvjknN~7Z_4AwabTrdavWueNNrTq^u!rz8(|-=U*Zx&J^$97f686jpMcL zqU`*nU6YXFfdBNZFY)t$@R$`d=4g-5#CJxGk&b+tls+I|1dIu%q*AsNpcI4hy(Pj$ zRe}+yjzD;zCnMRmk1V7)Aq09wy1vH|;ph$E3$g{<%%4d`io`~~qD0;xajlUH5${E0 z521h9^p_d-BTFfR9!-g~EOwD>xtAQ6vgZNHMI=7KwkyPkC9#CA6z3Ktlwd4?H>KB* z2X;dmhhLj|^>=1@2Igl4Bcz?drtboi;-1FZFTQ{Cvp5ho%_O9#vY-Z=-0rM4H*iW3 z7y~KA@SO&BLCr||-RT!oV|2w%5gFIzwco6>0jEE^d13ud7x_63IEd%}wfn@+a literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/effective_throughput_main.png b/blogs/deepspeed-fastgen/assets/images/effective_throughput_main.png new file mode 100644 index 0000000000000000000000000000000000000000..1b9a38306e8e2c0bf5b6aa453b145aa23c708348 GIT binary patch literal 182665 zcmbrlcQ~8v-#=~yiJ4Huh?#1ugVc!Hi4`qHi&CRjNzJG|YKvH*x@>~pRYj>$T3agB z8gbVuZiL=wQPga$@8x;E-{t;D(7g;D!;Fsd_;CJ8%Q?z?fo~m})b)4xHJ5dk6t% z8^FYbNILmp&iHrZE)&!7b7OssWvJuIKWs7fBh4J^et{3rVj;WGxmaB=DCFXesdLYN z6`RL5(uF+hap!#elVmVhDQQCzO#0`~!v1$@ZTmVAKNo(wMWXWx`j2Vv{X%~B4cO;x zMeb_MTxl%Ekp5RIAF5j}(*Iv=_#>)T?|&T@qa(C z;@1Ao>LUrQ@N&-rX+0utd9-RNo&v9&ATeoFA@&7Xsi`F-C z!9D3>XKva1Cq@xTq#Iw0j<(MH-hLh3Pvsp#^-y^&xc5@ly|DVjTjSoU4Zs3e9J;zB zk;?D6xFnVPJGmsCO)-8X*Zb;qoBZZN{}(S3sr(#4P5l0(&)eRrx6|?n^+388unOgE zioo&?D+7x>tOopg>H{mEW--4;VqzjNb(jiU`SADQmyaUXJ2@`j|Asr6FX;!@$Q z0M~R@;9@Zq_{H?ssGVQ$xfl5*;Xwj0iyxD%5pNHEhP2;oAIwJS^`wcEMxBw7Df!b) z<%L;%p#l?_?*J~R++8?~5tcupz$>pin@mbz%#I+q0aw2>NKYF|90$Tan~6kvV-Gdk)2`|x)3;^;FaQ1m~6k6N}z z?d~sEz5N7i)qjbf{Qsxqm0^QGw9So=`}x1_ec&;^^~F@6g@U{B9- zo*cDYiT^rqjK$#>zYGj#|DW6;GxjH~W9=y3Ga;{{8&g z-Nk_d&tv~fQd(NA3vx%vX9~|2s0ZCx&Pv?vQC*v8xVE$MwR9|{{@xeE%a<>gIlK@4 z`-NFiSw?`2WAtHA*stC*Bigd}LKX``KCubg7Z+*Wu(i#pw5i4VXdJO8Lk0Nx^oZOR z`>`Ls2mbKjPnxd_+jQR+VPj+SEa-Q?Y0R6{O=&xid-H8!p<*%1!oq@*btP3;;i+Hy zp>?Cr3ben^JaPPU`$uTSS+mp}Nhf5zqW(~FmVCX~Y%iNbEvEn!(tpUS<+ zzn^mZ+1U)5Y{c%*C%-orKi|6Z;Pm-h)t2bE{sgSG_0z2%Bi9bed7Ze(*&cOyd;Zvz z)0SSvT8a8yfJk|sh7yh}a$?JPwONyekT5d)>#_5xLKm}ijy5tXE_=;(#LC|b*&eyI zd~28p;^LPm_s&rds`!KOq5k@H|B%xnnV5R%9qzRKb@lJapuXF0eGwu2+)tDM9rvRL zJs{i2x5!+?)5M&idXC>v)-WxTOQFU5y7upHW%3^Dg?r*i_A{orWjkzRVJ(?%7=EtF z^~1I3J3e}v$rnzR|C@08UZUe8IPwi)mGP%zQRe&n{pPJYqj$h6eTzQcFH#qNI}`Q( zYMF!?flGL7_@f%@g-7;{}Z715jtwL|EFS9?%@I>_ZUek==Ug)r9+bc zM0<3!X;3;OIaNPPy*a5dYTnL>KG1XT&pi#aKY~Myh0FX>xFyZ?4%UkMELNu|R?=`@o}yT>jq$g@ zXWAi@0?xMliF;t^!i~-a1iqTz4$jYKoJHUjS1YbCi2_Eh_{P-8HNqF8e@Y1HXZ~R8 z*Jp)S3A2OIM*(5;>1fM1=$a%~wUYQIL_n^5u*1Lb^4+VZH#i9578es-x8_n6J{27w zcm#=~yq#L)&!4yY7hdgyykcIJ;3lT6qoj1x>b_HX1EB0GKq*0;e-`@lULj?&0`Ej0 zHdMIi$p}Hj>fT%w6_4X9gTcY?VPA&C7or7Y|g~oB#RQlvLBTQ|yq8?Bk1CXAPQ{?1czacqnGzWY!CD z;w`(=5kCx@A?PT70mKvargP+?aqBO88}t9(IsWtP>Q5@GPY|+LB7BLGJN*Za<$|yv z1=dAu)^irx7Z zgP+Zd-yWH1TV%|~dR{;VtcQ#qc0xnTFL#*-0&+cL)%DS2A&09%+nhOmEzY(eejluJ z_}y$oRf<`Ni$910X?^Ax{+%@V=28=Hb^B5HcJo5pd^UxJt^1EbJf}Rj2nSAICONU) z=SNM(_#NcUhne*f0hSL>|M6pksqg@&LIndTlbh+@kK6J-Ke^?2E?{wBQ-m!b(e(&jLA4@Z|E5fr9^FYdud)mcl z+onPATPaf0YUAW^$a3AF_Lnt&F=APV+HxpgzN^=|*@*{XBRWs^Tm za2287lz@m@VyLW>@!z<7+e&YgXlEg^@35`nEu(B0a(^}lbrfbIH`A}y^yMfSMQxQz zj$Hj&{5M;73O{Fnk?rnce!zf*pi{XbR`AOQ;X9QT zZdm7W74DabRKB0WHWW^-eMzo1R?1A3jpBjioCEg^5tIwjdbZ#&0_-MEN&Al&jhd)Y} zznJM*wRS-5Dbi#RRvF*ddQW)|h3p7hEV$*h@83ti#y6(xz^ecFgKOMZG2bT2sD~7; z6BT}Zxa$>Oe46XuZ{WQ@pP>nB7sGjS$6(>$Ts?!6 zCTk#&Q(BLk82?4gV!JWQje256MLz16t6^PFSx@`s=U1h6;`NnO1F(hLw6v z40j#h78)0N$09MVEFnh?Bgmemsk8rea2T|mJ;=77b`^pQ&gAX{{fY7EFujiql|j8S zUw=CuUB?sh``>4;c2EoD24Z{yP4fXY61@^D3dJR0|2_`MNa!sUoMzkpPmL!RutAzq zLI!V`o+vV;qA+3-lVrwELK-Z>39Myl%n*0#G6N$%W>HT(994WT+1EtZ&)yQaY@GS`K3jUv z%iD$-5(s6{S&XdimuQ4eDMkEX`OdMZvgmt^Y&oRS`|FzprJC7sh7#b!Cd?LuOpD~J z`s5yc{&UO9N-%Qgt4F+E5-*Rtn;S@>`CC?b(Oaj^xY=$#9dYHj(%IS0_;^Dpp}_b5 zG&D-mxYt5}=Q+}6rSU-EUFu&sBdsc%4X(b^uDwUs<6|?^Iov9SZLx3u`sMk62qTrd zDtnLD{Fh~IaOmxr(b}7h??k!wl4l=XexC zCcGPK*2KIzQRq0Gpfvx;^bVu;fY3cOC_2gOGkAJyV*WD_UB(;MNGh{ro)egjtBsKEytrdc>kO%{k3b_lenv3u*QLSy5a_&MenQI*1=SF4QO z@&6$B=k@&NiPr${IM}zYy3Ho}U~6dhFVno?#iU>Rhn+wO@oKQQzuzkQ%L_9}K_x@L zmC3rjpT_3j`|M7X$c0OnMqgL(tbU6=T)v~SQS#+Gw_`xRiu`o>d1~XJ)^2{v;(8lt zM$RM~>TgGvK&W_H_BYOlJ^*~GKVR3OVvKwsZfIRt8_-y5nSFfj=GJ-7?l5gY*Lpdo z{yQD9W;J##EUY=o$|s2zaRX(Mr|L7j?EmvbuNy=@xOea6yC@D`@$xz81D-i0sz`6o z-ErE9pjk%-PTYIJrt|xoz+s9kLxVg%9H}+84cI_qhg%yR4JYTRabc}ib9?xvymjir zt@b}w{mUB#8{w^ga{lrdEY#{;EQ;1v!_%c2+DUrRJ9R}b_n8(@O$Wa|kJNhO%rrL= z{;U9i!Lo{e;Ds6>&QHeho!w~!yqxe(Q_rIUZYb!cm&s_sNryrrjNk84Z*CA`u6<5J@_8xy- z@7uTFbMVRdP~LsgYH}bi{iji6(xD>d%{L%k7|H0;6XYbnXqaOwm9d$jC9LiN>dfi_ zyV?tQ6pz(~Y*x0->4f;r^i|cQjqch`)aBPn=H>a}3ppN^`R+aCBw5rS=8D*?*ay^Y zQVKlajzvL2MK+FrNGde4R8M?A-PIYX7Zw)w$Krs{nk#7Oo+uC{$B9A^Pg4)9_atN- zYmYQw5b_l=kNjWjAvPxm>sCv_S?V~vA2g#SZpZzB==CA|BI8fzS(tv{fiobQ5R~77 z>`<&Bzi;MSud!-(6DMVDi@zL%@%kmTDc@MK6yZg1^mkLdQ}(|H8`yX#&n=TTq=dw< zy4G8-9q;#uczm?FF?n?OXCX)>A$|9qZnvbXgDl|ii-;b26o$E*I^9jAie%Zi^E1s0_AM7hJx~H6l(c3+%q9Kd5UwCCNda$fx*to z|FY??N*V;CU48KcfhvX6vQIo%=nvMNtgKE(1P*ud+&z!vgeE0o?F7;BLqz=~$@()W z4%vwo{G?2COkUT5;wQYYT~c`p+XiT7XeBLiRd*6$jPSzBh}Y8;6U3#alZUH6KwaTS zW5U(0W~4<=)hBY-YJaAho>-?xzNqBoo%)n%g*fr z#>F4v$`CG=fl{n3Ixb4wj|9)XbnR+z_SGQskS&@*na7mW_e=}a8y3Pbe}ijbE{*R# ze)@D{E%qCfKZ=Czb{LfN<`NRf%G+~fL{eF6gO})wa9)McS}eC;R5WaHJ~vI(+X1Q8 z*0caSFWVgdn+OfaE3oeFZ0-!<>E3kC#&B1WBG_;p$Vc`m?8olT)pGfMkV9?`BpbAkl&tqEgRM zgWNN0jxl0tkGA*0or$=ruNY@6R%^D$eW84y19Z4x>LG5YyhDhe5CspjiDs%5!p{q@ zC4;6$ZryQEYP^4TiwQfzv%^XgOZVO)ccUYrvIC~&Se z$wwEv4&b&oo+!8%Co(}WEDMiNuR?1Mem!fie3F1YeX0?r<9hnS`jt?=EltjUB)xT} z#oA?Od4=)L?ZC-7lKJYvgPaLlM2y&>>buSk15lFJwjp)*v^5hdbloY))TP$i$~k)d zpUYxnj9Il@S5wzXdRqNw77FC=;4k^xZ@{afmNG`+W||qiUgbEN{S%TFy|KqMT43VpLjS#~rt4*O;uXAo*KP4) zrC4@IS*TQX*G>-Iuwl9qEqyPR%3I7TGVzwO$zl+fjjFFoAGqE-m4=q4#D8P={ZVQf z1{m000BwUdJV@S@I6oxFNnVM}Xb5fnlqiTEHhGzGQyJ!i^f+$z~ z2(OE+*AzEtAHc#Q56_0FX=mqgl**kzAf6Z#Aqc*>WSBb}r|7fSQ7uuWL`5#mp7zU!*UUR`)11|71%F7o38v&F&MM;A2I$29}ag zSD9)fBOLO+`NPD0A+7K&^BKfuxc)DpnGkY}ZeZjIHlN&)9--_zqygqxd?;&Fk+;z# z;-fmep>y=`Q?o3ll({!)=3oIWzY$X*q z`lUCvl)IOI<<-|oFpnWJM6nO!z!0B}be6fP%PJv|ZphgNU3c2=Pd@M8GQ_&35dHhw zxGv|Q?C%)qZ*)2dDUI2$GY_clAELvY3BUpkq!= z_M>71Ik5|+pGk7{{mC^T&5v(tT>lCd?<;e-P(Io@!w{{-I;GBwLC89ytD`B1e6N6| zcpJh4V&VC%a;5n1gv_SP0)|Xv49XR z;jxp0;Ws7dWNN?1RJN%*CaZg*mfPpP<&2GnxUKYGx;Hpl_lGdNwMo6+vZc4$zQW|} zWsu?r7aKYBWJ|R;Nyp~NI*+CLG=}5p<12>mlDx2Hz1~PWJx)lI_%wE5|-X~4DA^F6JbWB*a z5f-n*EesT#uk#qRZlCu=&=H0q61=CG9-?wne@=vLqh(#lI@aO ziVYid%_v-l+PqgNQ_D)4_eV%E?*IAz;#E*kB1kZ6s7W%`6G0tcz{%|;dolU9UI9r| z*M-j8$asdIK)*5sy??yNOPu56Q_YvQo!jIj@XE6fPU7;hTq6#1YqCDXcO@^q*sYlJ zL%4thk2*4pe)Fuo8AADwVjnyXTz4Sadtr-=brRr>%cnbc(QopO=#0803Wk4_?sZai z+jzV5@R|3zbE4ROd1co=pl7b&R;0=~IriNT?bC`J;c>+POg&NM7*)QU8mrgaZ zEq49bph@0J|Bb`2#2Hw^b7)XwUGA5{=bU2~EpqM_c(5^ZFH{4F!eyu>MU={kajE|m zw$-G>dPjhTgo_P5yEWkYx_pU{hN>?K^CeV5n`~M2?9uS+G5;+&NreO-J*pwj8d;an z18V9_Jw`Udl)$ zYo?{*;Q1X91XEQ3sSC5>RJDW&;!#H##y{kNzR!UOQN3|x2|X-J%I*jRD}*=F@j(yw z0#0ysB&q#92uQCtG{?%N!6ewIf4nEt2v!rsZFUXzGB3~ud#F%?h6k{ zQ;WTxg8Mo9p6n(=d_fCoZ!4}l2}Es`Jou{7Q}1=>;DiizM0mfxEDSLN?B`YgB}{^I z6+MFUxI4p8Q{Frij!{rhZt73rR8|!*UMYSH4v%;03Q?A#$LL>XGX^Vvq&MjnCoD2Q zQQ3HUz}m>cb3Y*8-BP8PLoADJC=O&fTKv#kE;#cnDv1r8SjLydGinSS+H2kV`4#Swu)_tlyjyCCK$x|=*~+f zYX>Z`fWA|b#m^=Q-xc>R&LYY0>{sHzTb^&8j+ZhgmtC)PW0{`0t(vjhfvVxu zTJswAGqF-cqV8<`psozprkit1fa0$3v32naSpXFr8X}5)q;$h%jbl|E9E7|w!s<88 znF~!m@Xnp%zUe*i?>(P8w7ACIlf2TGA>LBn;`ArD%9z219#OnNlJ$Yp0L&7?&%`L~O)V%K-g4ZCSwBa(2%~w^;K{kDw zEObrcBA|gsRh=g@moRA=YEprn{;KU7I2MNJyLqyPP)J>pwQJ}l!AZ|GF+y! zEte-q{FeQEQ{mi8Q7kn1hd{uEd%9Q%MpuSeInG1;1)T{k(L{#TxJq0Uj{t3fpPE|; zL5POkgcR^c7#Q#u!?M>OklwzytEtF0a{hHo3Gr3OhW>1tv-HC)EX$Bd&0MbPalr}m zK>aphUSyU2u71W9W5>A~Rv(h}y*VRz6JV@(s#j3H+-yPr7;5ZE58QOunIHwNYR0SD zFST4D#rEaxH6^F17Jlu$WJ_w`Z#_r0VWn}R%0&E%9a1fr<`Ty{SKJczz-#ev9zmO> zm?Y`W>FDDJDtCA*2m}F1Su4$7A7$mgy7GP%29^ z3-dKGY;RR?%3Jt5|n$&@oVyDE}>7 z?p#z?`|+XQ2%y=7`7h}lF!-!StTi`WH8Yp6ktzGKO~iGPrsMlV2Erq`A&}Zr14$w` zWr^E^Qj)Xtj&6)qZgc5sPYvelYy+Rps+gVCsg9ssS(o@e7CQf8pgcv`zfiNTgu7oHd7om~*@$<1n z%9RAMI2L87-XLEiL<`gp5_3%YQ5hsz@Bun6~KeQ3ljx{8k!*%n;ou=$Vl+PVG3gSgf5rCKe|F$^+|KTnX)087uD) zqQPRz!nEdclg~6?1wdU(VNygTI=yNYZRw7zD^{PH7}ehwSsip^$X!(L$VPiTEZ0qH zwIOU?Q9kONDM1v0sgD<+F8In}+?Iyg7yy%+lY8n|i(`<3j~q0lTCg-$dyh!@e5 zd8+&2YKzZps@CiFqj$I0Bh*6uyi|mC1ZaXN|E7oReW9Hb&Rpn80Ur;$#qJLPPQ__@ zN~Y62oz)-FGuh0&AtAqv%C~9ZGF;ZxMDPy0xDqJt;*K2adT^(YtzZo*CPl=E$7I9( z1+_UgYjR~XsjM4{)z@^VV!v5C2&BjCv2o36IBW5$MZpU|{-@M;7)r~fUEpIwM()wp z7c;BM1VDLjM9l0@!^Xs1j3S@)%_n$Cj>*9g?I(uMVktJmQFh6gWvMFOQFhc*R9BBf ztD-M&Z3*Ho=kkLko%8XxyjcElU&@APwG8r)Dax%8(^_63s=6Z8&1wyf~@k%sb zv2$AC&5ft>^X@?pb|kveL#Cz-{m{yzut18l*sB+!qT67hp;R!vTz>?db$XN&-hslD z!bilGwq4ns;lFssFlV`iovLQBajc8<38f37`Q5ng*&O;3(98KgKG%Zwa!T>LzIUyF zj;Sn0rBn5nTy`e9n9kjPJ>qFj>-DQ zQ1hs#0gZH3XhLGNf`zKjr*lvvdSm(fi=ykEVM~;xVZ}B%q%d9Nk_i6RoO@ftAraoW z(jvgKb3TyDGUa4ezsc>QzR9lTYn{~dh$1T_MegkXreE<0T9+G8NbGr$tn0#odXf`b zJK!wBLl%_(goknZo%Wq8q<@` zrj=dahEgv2$ypG*327xC+{T0JU(Q#FVmuT~jxq*GA)RdoDFIMF83{J~*Sk{zX5(wA z%~DUpS88d)>n=;)_ku+~Y-eo7hF-9?zxRf>Dt#HYVRJdT%e?6+VX9WANjQeL0E5e${sCVG%YX?*t3z+U?(ujtB`rTjX32uTa zQn}JK0&wS%+k{dH9w!hJo*-k>)m9RF2xRy^y$13nxs+x_d!}$98FNUY?g#E~HFrXz z#Y;m4H1Otjuk;C8bDB1+S|D9;a3vkgGd919*IRdHXO1ZYg|9I7oN$eFwvlUKZN=B8 zvecUv6eA^k!A=)^k)c0iNjRa#A(L^&`*b8(WJ(s^Ie(J)SVgyaG(lpWRHxTGFYKY2 zFmJZQyP@m=^m?a%o4DH1erxYumkN>c4T#A|WV!m*{&5a;hkqTS*1}=JuPN5A>Vj^5 zVx!Mx`lS!%mU5PNX8*dc7Z`)dg8^nflo$GFWS7vq6zX1e!zdr1u=FB>Yx>LUQAsRD z2dSGPJUg&bxnk1{N&|FXN=VP{)Dz+C7RiF*bC60rJChsVUtL_Cn{aE6vlRxgP~@ z_M?qfz9#0a$|*nXb2D-%MkNxfY=hzlYoe|z_hKT*hW!o_Myg(?Ofbj`q0fW=9Bs`zU)o0nIUc zsqiQ4=j%^&lz#*x44C)tr3=p`s95<_v@2Avtb$Y>+fNTLd5&ulaHg1_o@51qS zv*;WCj|F|7&dX^})GzhjmAgf^fIQJyP@J(2y22MT#M8(4mx;zsJA2!OqbND=V03^y zB<0(&V|4pyJxQlLAy1IZmSsmFn4nx@tYr36kx)aJ;-X=f3Mr)=JsqkSWSz>6812gz z<^MfpLok88NaFP%?Q`F92(l_5+Sg-SQbG~_8tT0a&3T*IaB3LeTeGFJzkAOV4Zl0e z0Z7mQw0#50FC=j0x@Eg-p->tO@!kERD5PnKOywxX#1LKa^GX)@v|@knou^noicM>v z%*)R+JzTBc=3mY4Xoc}7eu4z_x#`MsKc7n0ea`DUURo>6saKI>g=&&J@wrtZ;T0Bh zJ~1+xdZi88Vzd~dPS!O!Z$8n!4P6!>^#(a1uFBO*ZGX(HM{%^8LeGHctd>{no|0qkou$P=LURJ|PU zgmT~(Mt6fwlF!4?OcH9bjJdvbh*-^6t_cp2)e@&VFBRYQ$taQ6nEvEV4ZzcK^!_!B zQ|rzj1Z4*5-o>;)E>m^kpDL;eeHO{Sz6$cGP$NUHn=T(O1?pZ-ijDVr-D-UafGtP` zGNz<-MUEAKk`o1IVq8c4<8nNzhbO#vShAJckYbu`) z!%8>bLp-gvt#fyF9mW$51Pb^#6^ARArz?tu9j0>W8J_x!=z$&aOty04J9n;AN0k7;6hvh!AznrDVq(&S34(`Av6xHcl^yS(>cY& zy#z$xsV(&C>N=~A!YZ93<`Mk7$vnyjQNJD)l^d@QW%DF<^l82>FX;X7YQ7V|gwfl287O&`N@m%{Gt-Vy_?41Hp>VsN_fGA7SKpZ9LTAUB zJ}nJJtS!H8x9u)#Y z?CHMx8e^hLlube31v}}UnjYVwxl{su)UeN~Yr{xGZQABz!lA*6i`<)f&Luk*ocbytKdR2;At!798HhSq= zaJi&DI@$psHh7A4Ar-(ZQn5EiEeykc`vLu>~1x@K$lI(jeC zt6LcTR={VxRN77MH2vo3dN&dDP8Cn6I0y2s4B`c$reGoRLa3jz=)(gXdw6h8`W@sc z`d<17(}yFi%|0d1`?u0|)7X`FAhBT47%cmb zWxcZaVl7GlEUR2az*;Q{CqeZbcEu^y#a$2r!NDN^l{p7!rs0p)gn1WUv6wOG@3AS* za9sRy@Zikq^$XjPcXfA zSPw}rLC^#i-YfMruY=@D{|>g2a(~ahVbCI|N-^Jv;k6t6#KB!B?#BVIe~*1t?EeKm zDa@sQEM?eHW4+K;?2vZ#DP2|8Mc8Xm6^PN?M{N}Or($Sc{hKkWfAsjY%7__PxBgO| zlghuT6opyTt=B8XXeREoLr`$T>m>9-rz&xPEkU)l!Bu$fT8qc!4OJJHeAWx@-l=ZV z^6MMTfG}07Ael_y$=wI<-{~I2cun56P2??mNc@fU_A0!YpQNwB0l;@xtRoX1sAgPr zk+;w5je1c<_um4jhqq`HvC;+^g=1AxvEjq4g^`{`Uff7Z}S@zEO`Twk(*#b;$VY&K;rS%2w3Cz)cfzXnG?^lB73!H? zSWei$rL*~fQn<6^BtD6>(2r!M-7d?k@M-YS{WAfJ#y8lf|S3rTqzF8XznjR z&}Bi2f~4SAc@}w}PmD__R}bZJjKJ}Gwn-B20gmQ~Ix^%^%Q`kFUXq6vGV|cp=}?z+ zt9))yI{sTO9c=VQ#(MAyC}UJwwxqKmsFO|jBf$3L55}yE++Gzt=N}W`RP_zR!r6_~ zkC$R*mJJV)>5OM=_BnAfBf=Xf4<(oP6FO+3^|DOV{T_AejV|q?@E$`4ZO$kUV(Mp$|nloMjA|4*Rz{K=Y@l%r#^+rKd)$eBf3{lR(1ruZ}6Hq<(r`CFr zNqcOf(eM|Ce~F51LO|-LEE|F;!?0aNBxxR@zLo(=Msa(2VluE^aGWYeY3xZS0NI8dHj(}U4IPkfO-1pu_-if1^(kqkeJpv2mc=!1F5SKnB|p|%uQ+EodpWadHV z+W4)N?)|m)JFi0cX7^;jIw9Jac}Df+v*_iG?9$+;77c>0b`jMdaC*-Zu^A0_4u9N& z2_xTNGuXnf5Z_B!yz^acNM8NmpNejxZO2Q~`pU$s?}9AwvvxCRac&;G+}L_EIOa#G zhsf;v7+2As;U8)-zn)juyatOu^ww`6=L~nMBzwvs|L$7l^DQra)0;4#J~nDpa*{7U zD_Vc6ZvfqdDd_9>I8W6kxmNZyX z75%IK0p>mLKJ3!Vgy2^y)_RiywFZ?`-U%8|?qj{kOh3{~<2P%8@>vR)yluXN?RK(DtKLHp}}avwKtHs zv$-2*2DA+0bf4J0@fWOS;}#@YTqbb)7oz&K)JUU|nUk@_C(LQ}ed7+FrA^-Ci3Rx*?O0_1 z)#ES9Bl<{9XmmctQHzouKBfQG+M<|}K5e>Tjgw-+-1xxBGpT?6LS~09pHv+seU=M7 zRf^%0Sdrz2Euh*?2y~#|6iB6c`Pr^-&j8=s)zt6I)?+=^;T5a9*f*WIPMA{7hs|a{ z{I?8FMuKc8JYLEJj}YwK`|-sxfAin2&X|v9k7}A_fv74m$NO}7Wh!0+@e4d;slqGv z5p%{n{d+RQwqbOZte7QdB4wzO(9ej-#jT4%ss;MYwoYnh*aDi!7B2$;exiPXB-c#9 z@;bEX9`q-X`6RA+pf}03>YytI(n}vv+x36}0pcy=VwO)#&-Yd>eT9zr|I znMXO33bi6WnWFzGi!x)^=0X*VNK#D6jbY08O}j?|xJD_Ilc|Y*+>XMWHFb8uL*^y_ zz0k{44fh)?$)ZyoZynED<7G?#fFXjGaoM#pE}%4P=sAnnWgc0i)K@30hZ*-H@@{hA zba$u6TsYsaArG4(C! zw$!JU-6*B`TlvxX8E!cW+qsS@g`sN%Y}a*;fSIh_ zGp$5+O-U#d`q>d#avO4*bSUKF>z;k;j}11wc_}P0Z+wA%AFK#9kv6ZvmiC6vi${x@ zu}ToUGi^Gdb}_mTbm`Ezi897Hs#=6Q!S|Nh5+=#%bHaQOz_Y8kS`;c3+DQ*|kKw=x z+0snd^Od~lnID5<$DHApk%CNV6KQrPM++caw|H<3+=0+VHAjt`Ym>_QfmmvySnz9)H|YL_JxXoUFo^Q5~m z-Y1uxk`L;eJ!R#ZZ7w*Y1FC)~6sa0#B2X;`GF{lLl|LyRBxO1KM6*s9CI{0w6Y~}T zI$Ft(AU=S1GX2qveF8h$OCKT{WXZws`_6?1N`{@~oK5%0DWYC4F4cFg!_jYm(wm?W zTUvq|tow(dm<5wRNyLwGn4G}la+>dr9DhhUwLe=2U28ksnN$%J#c#?+k%o==lCRX&$&k4w7y%1? zFQ0%rYr9a`$v;4e#v{qYB_oyYhth_+_b*NnLaj*;UcdjT9<-WMEAA2-eY9ERIg@$7 z`7E?Ubv7b)kmljCuNb&5=GQ`*I~8d}598{Na@DAWpahO2d?GdvVS`fDDXQ3*-_2@% z0g3ikExV?H`)q=iTxCW2^=ePL)YozKSuP# zHNB=)(^GMlAnK44DL%A%L;s{uz*^8bMvQsyk_<^T;)!&*p)T;|7J-!s4e9$K8Tm6C zJbq z7YBrC`2dZZaK3Kdm*XktmuaHQp=}bFG4Gtns0ZqzZXce#Oht?fS>CfriteU|k85-P zliv9Lt1p^yQ0zdy67K&~aTv~jYoVNFk1W-h^xIC>YF6s0p$mJhO7T06YxSEcO+e63 zegybWZ|$QGw>{yi0&_yhI_T!@T^i?p+=I)mS9H0F&37Ng4n&nN>6y67$(@MntJQAv z2nwOHkDNl%(K%S}W>0q}ko$)8Grh4v*DLoPk;r@}rBCd)!I#L0@Sn3Dm+3ajVo`X~X;J#D&FNq0ChJPQ^GxAS6L|$fSj;Eh5y*Dv^eB3R0WPWiEu^$J5&z@sbw1 zFfUNzvP!6!WoR<##7C{rJ2frbT`BFH;QX;t(5n65Uj-}d=?N%Jj*3_c-2+VIEdXh2 z$(2{|K0o((W1jZ;q!y`~Yx5v)kppsZ<+@)>HoM$$8{x*PSvzp!Sl*RoHVLE=1P z$&=HoAg+MSUq9puiSkd+8`J*z+{Orr2QPuWs(xPEI%0yUIGhNe3qEt6T}@OH4=+K* zI%)z%E+el=lAn_L^Y-vtIq9|f&P&Y8HgtUPnaePB(W~VXwW|7bru3kB5p>+pCU3gA z5*!t=lW?ucsbIV33QI=l)!kYIP$p;Ezix#iNkOsL3T;s{e2Yv64wQ87nBl85WKj8EmCv0TljL1Qn( zv3T9Jlrevo@h$6XH}_iVRl%)dE*{Prou=ukPZg@t&%Eoo_55r?NFA^6cv7O^o)$3Xd*r3$=S0d#-w0w@iv`6EkTkDgS9f6$H6 z-FAEs`TE47nlL7!NL*Q~Wo$T0DO)$wmE!)pysb~p*2QzrE-3rP%)37*L01fsvH5Yq z|A)1|46ABu-@wr=qKGs|mozRKq`Rb~*8=GVX<^Z+h;(;10*mej>F(~5?rt~}-*@kK z|IYbvKK%dVgV*JyYtAvp7|-+EPu=4P=(^J!YeD2JP$c%cLPX-hUn#wgmwd~2vy}-d z{pRi$ohDL`&pE1yXt=oivi*53k*lX3>1ikvae>Yr|QkZ}-a`Zt+?wW#UqCLYcji{J6{w&XAD#N`@l>_Z*s>fJmUp ziJHiY$&I5N!fsQ_{`;=D}bQ@ISO32L=B%P&QwH@4nc9tu3!urTd$`(i= z&oLcJ4;aP~G;<{P;x4Vf45XD?7Z}oG5R(F4f?3w^ThVy7&PDbWYFeuhnp445$(>Jx z0iSa-S_t;3J-D{?3<@uV*LX(ug;}4WsFvG=b~0EIC7y10izq$siFxW0+BqHQeCs3d z3tcySxwAB0Oc*FNgMSYyY|Q-l+0K9$2m-BHaU>1@5?$k)NgOf;+P9tsOVE!jYpUg& zhkX$~>Ek(5qm+!&eFeYw9p8!k8YeG2WQfJpnT%PK`TLqKinlD6)hl%LH)QERwWH!U z>_TLcZ*K=V3Hpu65vMq*kvh;Z{3#uW|LU3-C~i0jAXr63@7H%iI4gV28UOC$Mc2?? z_={Y9q{~^!Fd>m4#;64Y!e2x;aKey^x5#s*K(DyvuA8X82=jMD-X*6s6noSX`B%)t zf&D!6@NiV+`Y3G%%Lzh4VaV`q%~u^vVAglxFB&llm*k0P3fNL5{_Z+}|== zM?-#d!`hu;%1_Pyu(zQ!c?KeU-!AxOf8;g^&0(c}cc?@p4hf?dfxN;*1(JKIJIg;D zpju=;BH}?a2Osvl;>dC43%Z08rNFe5#9O}7E$w{0WoP2h+o~ynhlAo7`#FbAbBL=-5~!_x;vm4rN^c|9 zFa8ou!zJj1f|De=M>vS_K&@ioJ^0=4MO%k|>D%oegJ1m=OOk6Iz7(#iv-k#v!f)bd zzIhwNGKAJN)hJJ$CGcbSMb1?(YD z;WfyV*gK##nIls#Vp>ZpW9?OMrwtn}e=swdmS`(ig`{Cx1M%Zpu+OA55DiJthWNJz z8@*TW7sH?qZy$iXFKT7KAs3GRDfl^}Ct~bvJLH+36x^iifC5^8ZF>02}!;VDO7J!z3Z@6CQZlqyXuW(;2|NbB&`9YsVRP$2D{XCk-2dH@u)q(a%adW$-gc>35ANEF5$bWX590NJf#tY>yh>;=g~6w&&1z&C9J7ewF{*DXA+UYq-k&@rRcLwpO-R|W4a|@tt?Y>%?9R{ zblH@I1bn^Vurj3R3IE7v+IpL@^W2iQLc(TjhHrA~lkaO>ux(orc2=^k)ArOl$WBE@(?GW6UHxn6N}{a8~GAU@sguB*eTK-PZ92z z`~pErqa9}4%B<*zFmNqcx^%_)?DFlZEKL8*=ZOpCKE#_E6YNj@8}r8@>jxcX3E7s- z6=)&iguycw1|Rx%H^E8nNWWY?fDq_SN2&B)ns5lF9#ll)E|k;dE^VA=#Q(AooqqYl z`!vOKfIg;||8gG7QS>F}bol316ob=LR7zjw#V-|Q`wE}K$fUj!j(+|WAN^*_{rd;P z2-zBpAo#x`uxr9Y33tCz0w=6SI4H4n#ETSOnyKN)3lv$UQlI9F=~A}xiCnVHBuz22 zG-Sa0IYI5#h);_33MTonGEcUMy17UR3b#8%+kJB@4N|Plzgq z;G7zsFE>aCBEdB+@E*DJ5DybBEK7)qgK|c}0q|^zDBuM@_a_YBA_tqzmvH8bX0&h% ziJDJ=tK69rK`3E}i=`S$Zgr)X%iWFrw!_5Q=xrt?^Pi_F(kCTvgg%>`-j z%~R$#7y}oZ4+<6@ji0%OY zkL2vy?Pg}$cqoOq0noMMOLG=gy2i%FUuMdTZf4r~FQyGz1l(`NEtCsh6z>92oiDvn z?^+@vFz*>hWK;OKodBJNz{bkj3RHk(19Uwt5+u%)H9%wPJ!8*YnUM@LC#NbgQ^O^Y zIYghYFpc!!VBK5Fl8W>KXok>S-M;L}^X|W|{Jp+D0jQ;q4h{~(ud*g84d45n*ZW-V zmtQN0gz%O9aRLS?vCyA+%d^GdoXv<<^EHaiULn~} zX`JLI8nHiJ!Cyj%Gd^@)Y4-vP366fP+xh`)nnokmU^E(@)(RI*`}O3@F8r;(h#G$< zNl~9u58f(rqjibw)(9zLifA?}OjDus>*Dzcb$)g0(W_j^mgB&4{Ef}D*UzJ0Y?Q)~ zkgs?jLtm%xG*Ft`5z+IiZZWq?AT;18{$U;c0khp$klBwU;*_AwhfKc4$@ z3dXlFZH);1#oibYW?P`j1OgCI<%YKm%i<~Zc^`+eFd_SR!sKge1AUap*Va~z8GIlx zdAwevwl+N?3rY*-%U4!#iBD8Y>`?>}T5<15g0G!uuHFM-zmGM3qwZ0bYtVu<=`~|E z>6>G|2(mb~T*fsyjg+)!}x7aGOv!9qUusr3uX67F&4Ya`Piq()}< zTY3K1Cd~|4H**YEk?gNey^`b0OWsZqW?Vn<+6bAW=99RgW`315QO)Bz^9r&7T2 z4DmoPGO{1WTCgWjy+xB`0I<#(uymnPtB4)GfiB|DYjzs0)z;QlVv_XyYM`!Ju^3eJ zdWi|MQt(Gd``8+C`HO4jiV*Cj%Q`BSuPCDTfZk8%{lcF*DBj{>%)3+LNeoc->%a)< zE^wzmmwnhgx&-37@DS0#9m{xY(A8upEs12BUJ5DYUWsSi)cO-EL~Ar*&AU(#T@a+n z^U~o7I-O7fBKZ4a2?wdEcyJ#i*#v<@tFA;HwNbj_DO60FTCSkde47z_V z!`RF*=nC0!bQ8d#Y+U{3?Dl0KbBAg~N+4XQojqjD%%f;VlsTr-e5x1VQ-fjhUTK1Z zpGqig2E^aIhVJS{sQ%($w_enQsCKv}XV`Xr>|V7#N)B}tjQDZ?LBFd(<=7fmfHT~1 zZKR$1iKj3`V6as59HkU-<6I~xB<{s&$EuBcP=!!1!iGf)U-n`1V^aVr^Cbh%b^581 zd*@CRAMpSG%hXpYxu-DZ6@YQ}zO}^Ql|Y2-p2E&_gLAp`#{Kn>r%EW$t7-=c?}RQs z5gmz%*_{D4{Tzk(C%Ojz*{~>#-*G!T7N%D*6JyQ;l%!=!HarpNtdR~ftb*)ljg)eZ z0h+s9n`H$HfE5P=M8TYtz}=1qqEUzzm&LNu)IESbb+1A4(k|b+#HH_qJc*cJgzv@q zuW!%IeS)-WZL9#OpDCh*%7O7*;uBsXj|8tQ z*=Q@S_&VDIa}pC<0lrUc{4Zvb61B?j%g;|=D~@?-8*Q<08TG!29u^&T-mjd3acBIQ zXRe2C-0o9=#_Xwe@?2foUUEklZ`YU9iAGX@6BHsO zEBg3wZ$4-aR85?#lamK{KXI(R{A%+J;Q1Iol-#qJt&qBxlYNe9|HNhYF_{OVUhz2{ zNazngQ&UsB!G4OU{u$@r|MN>J0VotYaTurRd3SY;iTrZHK-yi$<2svOy+V@8z6gj8 zI0LkMR1_4V!O!LX{xLZ9Kc5pHAODAM&>)nb8cpCT4WI|&ZrEND1B|8GKo0bFzxpWz z#RzzDq`uy7|1ADf1G>+jSK9)VP_V?uPfyi5jWJ5ZASiR};1l(ws#vT}0rxy(khr2Zxt&3NLJ%-w!kHovVcIwq!AfDA_vnA3YZ4)2Cihep6+f1Bn)Cg01%4nQ9Gzt zno<2>Mtu6x73#4FDw*-aKhP3LL1HMvfOpdd09*+lfH3uAH$c31MgY9{^`1k0lIO$q zNp_3MiSNKIE1e2aAM##ut8yd%- zlReD_J*W6Xb@li02|;+~c|N2&d?h%KiR`H{Y+SzfxSbu2e6@CoKSwl-7GhY=E53JL zQ`lLmlp4ENsJu44()_`Zi16Lv1w@Hu*F8^@S-W^7Xe|bMJw&-yv3vIV|2#i|3oIHX z>pE~n_4dJQVRfk*0=F)HGc&8HCI3d^)M0gJTxaxUYW@_71{J9Q8HCPcM?#{v=}16| z)TYz}Eb**4O5Pm;ij|EN+9qpcWeM3JZl9?u=bE*{QjPt#y2J5FP;|vfWgzMV-#XCb z9Tlv{ZU5ayL> z*0PQbX)F!(Dy}&@EXSots6 zt8BYg!R(WICQ1c2h2oDvJRS$g22!}L+Jw>c&`SZS`Nj`^TuC<&Ls;BpUX;A2=-qAA z7W>4geBDP@2_1?-RQGcNa=6LA{yuyd?7~g=bJ=F+7J2ID!A{lmkwVAMqH|xVM2UUe z5h}?5Hx4#?<~vOeMh9@OW!a4)NvRf7(rukeuV*l2n{B^2#yuD_SAM<`gcyY#@HkAu z-v4QRRSIOYq9&;01z86s@|s{uC2gx4y0^n81QEngwEWizWfvB<&L*&8_@wr8iV_v^ z_0j=~689lip89bmymUmquGE{Vg=+U~aR0UB9YU4=6~OTG#?XQ{0w+Z_Fw1SjwUdP} zE%mJewh@9OQTUN<)iw;&X@SoTNKpKfmi}8gc6&X8>?>dIcPaXc<}=l_9L|+~0BwX| zEmdJ-o`H8vl#-TcQYOkf>Yb(9uY*G*B#FpNv0k5I&Xk=1XSeLh#Qt~W=9SNkCox?V z7_W9)(+yrU=F^;W;+W(-nAOm^%AMw(Hak@p78Dg6kt~P>Hh(Dwdt7FN8I#g{49tU) z_FX231nrS|FRQO*3>aTO;5@G(Dsg;#?KiL2D4Ul8rz`th@=3X+r4{h_w~=L4uS z+$JBhR(Gq%k4&rj`D`iV#-X{GP*_)S3yi!abMRAf!1eNV#}2bcc31 zb`@B_-UlfMe2?E;v;W z)3S4S&orJKHC7u#(9QU`zBwA!p|j1di#aGz7%-ZbN#^(O}JM9Dgf1A;ML$8z|2vB_mHEdmhZDud){;S+HxiJ(Uiu58rF zyW#-PN0oa+%MgwQs?7j|j)hywM(R2t=TaN(pMjp2FOT;Z^yw2RKNqmg5HYUthga-^ z@C}OOx8`+GsVGm~>ZkI&6QpeaV zzC7DRX0p~%`lef^BPBj(>chrdwK(Js%#dtBl%r&yG(M!>)%|tlRkG? zccreKUm-iq?*!6xgra!E8i5ncB67A@(r+y;G>D$Qcx>12Z0yL5$egoLP0nL{Jk^}KicCi0r66RTA> z1qq668o>-Hv?=V+cF)h%+A?--|1yDFT&oKepCDY|{s$qZLz(OT%kh#=DFcBJc2Ne7 zde{T7Z?Z&){mz)FX$dAB<;GR8DrUV9T*u;YGdWUZrpakC?v=S(IZ*6J6)52~QvN0k z5kWL&Cb4oN)a;aWnh~<(o)>|=0+LE1zLmZiMj9`HemdfXm5t9!;lM_XHwp4Ik7~7* zQT)qc(MysOC#C>888s)=X5LbTNTp9D@KZ==E+7!bYVOTV;{8O0o_gm~mKptq8ANkj z&9`-#{ia&A7?Xu0cSsXpusi%q!iXip+-UFb)ZSfh4Vk?oehf;;Z1?=)>o>$o6x)^u zGD;$Lk(phFf-zfCa34#oSYF(%GR@PiD{EsN#;p5V4-=*a`ron*)hN#M`w&-x((8U* zxJ(Wfsxhm1UhiFI1Coy)x+#MWZ6Ip-iK$|GFA+!~=%(LQe#~p{!Tc!8z>&zM?id~v zSsb+|;|be%(;DI4BTwl*{ln~vlXAI0p>~aFI@?Ey&6m1uUTSebhRScHxAhmZ>&3jv z>fW^cy!(dqyrkyCEnXnNDzJmm@< zNYvGZg#tEpzyevwyb4TRLbh{Bl8#PV;4bpb!=(hf%<){P^ zPmK#Qq$gSOu6}ZNWrW_Y1H3VvNM$*!20GRk$v~N)EgF2@hI)uY%M=x`Z4Gmd4C500 z5$HKYeQ)uKa$Map`b-a3^1=vBvfMQ#Uv2bd3fV=!=6x&AYk0~}iu5VxNFfEN<#YDr z#SSl&7{2MAm$)l}LI8|Jhu_8s;6zlH;Ny_a^CXsymp!$2CJ~NvOH`+4gzXdQ#XMDF zc~5@a--8Fv%9}T#x~)cSLWaMUPaqmeFt&{8QEPXjiJwjf0yfmS@pO?L`8Sw7Vlw#%5hxZzdQEr`YvHPZr&|Cg8 z1+^&SLd-XF%}PE`F=FaFR}1D{#M4gv31a-$g^>8gtc`st5IuS)Lt~9=FtbOXQXy7OK>TCm zYPBZ?73rN~Za()WIr2Aw<6v*>S!67cEd&2pvHWo=pt;sq48XG5xv*r%GX!r}t z(~8~pYJ*CraK6`UoWcCn*@no?l1Y~JyX*k_9LZ7$`vhVZy4e-qn!0VpR_>ITPjgpo z!Oa2En&Lj$=oeYzB)1>aY-Q&OlFT}aKd&4L3*}T+Tu^A1RiT_@?|j@ZaQ6zdQTH() zn+iM6E0isY>ylco)E5+xqxzaCL8;`9Gtrly%No2d4<#FwAYjh4LZ%U zt()P13uaTb%{b`5f7aVhCZ$m6mf&F*S9jfH9Vq?QoAFyKKgJ)!u5qgV_+Sa*UOwQ~ z91x5w;lkwIreMN$TvD8nde*E1XXJavR!NrkR=AlOfO&Ysv?+1j6Yjc;{6lc(qP#7EP6x$uf_2ImRWdj}u2 z6jji#5b^> zhc@BSc2d^b$75P(zkP)JNGR3rd_AS*)-{ZWmWUFa&%7xb0O07JEmp$kA6xfOqunG4 zoqV!T1Za-e+snm*p!0ZdS$@wosRI%^0*mCDJ0A~S{m%W3^@BEOyT-U44MRgc#)Db_ zuV?@;r(lLe*c5s;>c8SQxjqcMI)IyW19WD*&N)lpl({KEqM&I*JqQbMPR*5$*4;vP zv{7hVsyv|Po_dfwv3h{E${YlHzHv1!@~D2rTrK~rn3RY(w?-s7$Z<}(af2D~KTEq5 zkFk0upf%DZdN!BbQc1Z&Y&ggazJ_rO{BvAVjQ2YfWVKkOBEh*B|GIW0*T1Fx4DY8j z%8M9D)@Q{XC=a^(*Yal5d-~-wXW!etj0t>FG^6j3VC;0Dgycn#GZPVR(AnySWW4fP zr&6l4sDtkdXeC8ji<_Ot-`NZgf_{UpUJ^8a;9^?Q4Z+uZzH>$BX_8cbbZFgR-!izf z@;gG3ucHreX$eO6Upe@~CR<9r9xF{HuFpCE8o;^wK{i3~C`O&NOOt84Z4lQnHL_!- zE-)cdVsFLz$L*z61OSkBB*gN*{ZIdAQ2#VLj#auEmCxP^)(J3a6KUkAg4v(8*E{>e zOCy~n%o1Jix5rF;6dZ-~c$az;@P-c>NOZMkR2lQ#vu<=c{KLs6mj#I%D|3+I_!Tf2o@13LM&VXZw}`#e2$ zM_m!F3x|}CT}=5|p_xXw_w;+k&Ss&tB2bFGCjE5rU`q*$ytj<*~PCJHrw8CM(?>sKutj6{& zogQU?wx;QKU1<#M%$RAXP=HK=EML2TIu@WcHuNXwy+>fo&nvi3p@!rAPx&y&ll2(H zNp3H9_P@Hc7%whhP_R|-%I>@{GW5?hGFdwXo!K7;gBepOAD+)m2zZy7tjiF9V)T)- zH`LfvLAXJj93>Z(>jwhkH9tQDWC9)+^-K3gXHJit5cUw<4S(C;m7`3|s5v(D6RURk zff`hxe3gOs7l&aU9eM!h^yluM`M;{sOy7-HCD8g7g!twlT-Ui2K5YEmbRKn)gm{Q! z8(g0Hbui!!Wkd)-@{RNK%I4T;N4(k5ts1>WN65gar^sVk<^}U zH;b!wzBdjfwj1wM-fvN^xJ~MPZ#I}t=rx2jXkUgp?>W%)k1R>arS1^*D}4bpgMPvF z(ruM<5HSMfLeoG)bc6Nv8`3wAivwP@I%bFjkE)9F4-16Br!|>8!g_lt>^*A>b(&PX zpypd8Sd1sbgS!UUYcOR@gqRf2wpUe!=&8pKpktiZZI}3mf$b?K3w-Wc{opC_<>f)Y zDRDl3;?z=x6l}JjOW^h>q8Nb_C8=G0)>R^L3Fe`pL3T{q=|Uf;)BPF4y+alIM&3C) zP7at$8c36qi#nwe8f^Z+Yakl(v4%UL5>Wce?K5O23NB*7D8Bj>f1@{xGV@{8((Ad@*$DDm0(`_al3NVV*EbtHY4fds{k|k!hH8!f)fSMaSV-=aIod zX$8WSQ-DCy=K&m^sTI0&?O=uoGepCD6NmQGZKGaE;2Qf~@{#6%lEA)wUfj7L#Ar>h zq|gO50bR1U))jRuG{l>1u0cZ;tb^N^lP~r*f?_1X{rHzF5aAmJWVCAAm|m*`Ct^7O z7Z&hE%XMlfHsCsy;P$a8%Hy~=!9a=Qn(E!-hE^q0?nk5_ly5gSZS46Ki^-%>{^gM< z8s^sZm{FUyF)Bd!6zqyp+922g&!Z_Yt{jRWqDd<`*qFq490L>U-poD8W&j>)uQ&Av zGu|lm<}|67zjgkW^|8X)Npr#1kFafgeX@vyXEqFss1m&`;mgg7mK$p+y!uX|MsLhC78>j5q`#q&gE~(ilU_{x5(z-VShb^)o5apyo(#0fq|w@)sMG z`QAXoGm@Mqc&@&roH6M8-B};B;STdbD{n0^-KNnYD){ zdw$K77H6H4YO%f7rq2}-iB`#ao#g~iC*7vfuUqufsk-g86h2`1>OMs3ab9#yK*j7a zgQ2?$um2sr^H=0to3krUfl|}S@5MV$aJD5|`SV;NQztp*-&g=yEjz(Y3r3x_4}xckJmAB5I)R2)CoKKA<}MUO;bB==AxJS_%r0n#xMD1U zs84hR^!j-O$Cm<+XNqPv|KDo6+nU0ZCW-N&DF+53oELITb7?=wElN6aKBqH>R_ss~ zG19;iA0$DBDGr5P8jkD_jM9%^pscqGyHuYbyu#@|Hhq6|ha8Nx&71oxGp9n@?=)>? zLXd)HGQyUG{ToZqiXi18)(bk945Vn*+yoKv5Fo~_twFdkfEb073!jv5wiKb;ax9`m z<9_ucf@KG^R@t%ROld%g_lnmyq<-p^4WZ}S2ljwt&fGGQlQx`)BL!%m9X@SU0UrR+ zrO?}|U)4PIS~X1)HA%PQu8~d>$N0g_TBFHOr@a^?^D;ZqvZ^R?zudUb>71p=Z(2KZkgRG>u{{BW)zMdJ{#^bia# zjML7)L2N5of4nkc*VIkk^djd$Wnq{48`1)qyg?vzc7t=>PHlGwz}BQ+{2R#MM2 z91MnZN67XWo=>tXfmJAdgY0=P4@Zj&CbMN}dV2a*dXlOJ=5l`z4r_~MIm@vDuT#>L z{Y*3N6StA)GB2uubewvEbxSK=2<#!s@5cDWLm=2NcgM2bl= zZ7)pQyrD`>LyFWR-7v*I5x9i0=bupC7S~BfgVf28{6itg4#o$OqO$*pe_#hmW_9S^ zwms$M?9QVJ8Xd}t`QGv3PuL1`s#d}v%t+aV=s*#2u0`x!Df^l1BC^zg4Q=0hHL4gKl@8W*?+xSn1;NncK& zC)x5ji?$cTZ0nCjb@Ihjd>Qt&fT-- zH?h&(RX`3;IWc6k$XP=QT6- z#8YacSS@#o$lX!%OCUi6O4koRvr{V_qq}B`O{?_r<-K7E3)A=;;J!@we^g0A8cID! zkD!3l2Z1;ixmA>VVh3O8RxV=H#*GqCu|$JFvB#TK32SGuA_8bFflomxZ9aYE+~dvj z_&TRLZLGurvk>!r@!a<4xUXHOM5Z_m zJa+fM*)fyX3@jYVH__7n`a$f%uBZp$p-tD62U(~U<|8VDM7BQjwg*6O!+qV^L~+2< zO)_CtO~EHYGorsRGdlf$w2}uQ6-`|h0_fpO>vjeZjEPzGYzhR+Jsu# zDn)mj51zf;ZG)rcA&p3$_Oa)T`(=>UI6rlI01kB$Nrh=Ul7gG&{Y>qhU(WJ^yja4 zsm{}xpy1Q(Yg9z4I5UKP^0~xkywtSyg{k-W4WR!tD8Ys&?KQAfz%1*5BmZFzM43(nbgKa`sL(&OOyg zkY}`){+(*g^}pUq3|&`cx|e$`{rW87Na5RPDMaGBPs>vQn^$dH?IylkA^W78F0{9` z@)+fv4=8PJs9F5wIUaDn2!CY%*oph7c@zz3`K`+E^TXMp>4U`e$y2LQ^VT^XJ0TL! z-Jz4?f@8CXluv`i9#wbC5xj=|{yE$1l?-zgnm-v*$ZlU>?y`^EFF^#Aq;kZr?mcEo z%rk>hd0T2eygr}pR&Zu7xK!YF-TKt%lS@*nPqPlpJ$Hrs|A_XL)q?)D-Gvj-CHZmH z^Dx^nFMSkHM!-HO?Azv^h4pJR3KjboL3cw~bG!UU^B@w98`HFmOyMS7_cB!aaNoXa z6O#C75%M|(X*!Kwk5V^RY7?gPZATSZ)_0pgX(Ua>_uA#&E~^4}Xx<=?nVV@;8+^Ap z5Qx2OevnM=`Xn=lV;O8PFIC0{%ypNWIs36g zA#kdf63Zz+qDs45jnY~a%Z_yECRL0sbw=IFQ|a2DA^QsUK_MLT{X<}W)!JA5!|vfcSgHho>5_KO=+MuR3E*fKO^G>nbK&arC(`Cu7sY_A&%8Af=wulM&y~0X zrT$Fm$3IhAi&`+oN_^l~j7z0~q`ABcf9JFBP(QtgxR-i7Dv<2b!B3D{q3H#ne3ks#j+wSExi(+<@pR2xF&heYIQ%oRP=2w||l z{2!$8Www|`TzKi%w=!zJ>f1~_4-HFjrNWMaMD@W6%_;Hxm};R<=I^%$#d)SdCIDq> zSN6=;2o^2JiUwF6C-|ODw>H+|Eb;E|@A#x0L_de@21`mu^F`6_pK2cnkp8EU!v+g^ zd(jvhWjX~P6vx;1b(%rd$)1shHEd@(oyEPIiqOfcyHa0{-RZK%ite7N&=8b<-U{13 z5F-3@7gixWEe3IsW|<`cx~45CjMw&3^+?*10a8W`XPy8U!!X5&o#`Y|G<@8L}ARo zL{0Qw6@KnJ5@^d(&JQu0E){zd8`}+N!mRcy{GlUbeADe!hgx7k+72&eOvUp&-e?)S z7%N3f#<|iiHq#Z;$37QFVELda=2T3wU*GqAYO#1wnLc+mzTtDiAKA^bth5K0~zi%S>7l*#Tl6N`)98IexS$}w}Liu@KF{uX|d~2 zGsG245mgm`Nu}4U{-k2z@4i@aj~m1%?{h&w4!~#^%Qnz4c5mHiHfyskOdaRV$4IIl z7_+!HA2ORXi~}F7UaVE6>oIZTuqrehLB8cfl+ zN?i~f`uC=OPigEl*;DNy1*4?DeubC@^8DfS9I}dZL#cZw0hBNo%y=E!p_qD=%;6ct zK8ZIiM3>dP|Hn7r&Uh3Qmb0$%uD-}aK`KHAt8w$p`ZX_&-$j8|u_?qZTkI&-w-3Z8 z4^yPQsf*4^#r_8{me@WtrbdmI=Uj@#{!59GOX0CiPO>VUEWgzr8Qs@Q=R7D(lDMdg zJg1IR9v}9^@i|YG0k6-+(rI74B6?}*LWA40y_VH+z4{~+EfvyMA-?8N_j`WDq|dzB zNwJu_s^M9v1=CU*$hJzQK)EM5v&9N3nZPf~niL;lb=Q8l?Xc!3@iPlw#;4RlydemC-7Y4p&uxnR+}XuI`QN2$E;`e3_?q=8vA+W z`qUZ(GzXajb-kkv&QR0If_MNQjenW6(xG9Ja}r;?+DWe2WQo74Mg(lNQ0ZLA!dJ3B z4u;srAWx~q#~TWuG^jIS8WA3j)&AJdxrGa140mYQysa(Fk_L@S>fwIO zEa8ifJztFRn^kGtOANWnV4f1T-A<^8WRzZ-RTTy|#GlXNrlp(#d%)hX##37Ubm8^W zzWU`Ji|Xe9Z9~A_}w!Pvf-6+jbs0?$vm0g{H zV1l#ze3#q`ro^Tc{F50idHVQc=gAzkoLn{iiD?UIBbhejqneu^gDmE;Ys*0qukcOr zNAv6vW~v;06NYFAz*qBbgDAF3>w;k)*A5AgDA8FcXFEr+%d);yMLD4npQaCo?X+B% z`D+oI+?gO<`=4(d$|cG`ZgS1uO4JdkGcJVd-SsO7hEu|{NAo>_YrPZK-Wjo!MxFq< zLo=r?OML{Y-`UyOfS&*i#kEa4o{oKoz`pfSbHGk@J~ll_Y9|~80FW45>`%Xt%t&KL z_|~&_7j3i}N-6~(xr*Idx~FRl*YhcV^wbVYB!Ff`IjoWW&v65$ZxJ`PURk^t zkxdvQrDhbbO^6Q@JZL!LsJ$GmpJ-s{QdgGQi&BZ|1pr%5vCd8M9$$^j0GTFk4hkQ{ z!|odGcR@Okk?ch1NroE|P_x@2oYAaOr*cdHQGO1~753`7!SYX4-#rKFA1E_YJRSUemF1jB{Q!2WOlo?Gbir>p#)#s~(8AHr($pbAnfHM}3KiQutJ ztX9N2N)=osvf&|eVVKC-3$nl;g_gKI7MQo|70VVCe3@rvsA92>Yk8*;IeFs;3w^EE{)R@kl}3R zS11014Kix3m|lt}h)WncVf#7exS_T`CR@DR$azE)a$Qkm=x;%<^-iO2lHjmjs`2pw zwy<-0RJ;D5bPNybAPo}W1oBvasNR09TWi|esug%^Dyb7_YS$t z+i@dWRF3M|jtV^_t$bo0c_*rpdBu`cMi-F%sATL!Bbga>HJ!STR5mdlE$BE`FYiRh<`gQaM=ef~+$rd^%kw9+22dE!=xr8}9hplkVW=s1OF;fQ6LlePZN$ zAE?8M)Fy%Fl5nx*odrkzSvQYxoymEu@j;B42Z1y)CUf+I?Yk4~T*`sL09iAqdUmQn zBefCDY0`l235Bfqk9d`C$zAEMYo~vc_s}!uGPmQF3X}{cN_ALNF)Z z&U#8q(2QqP(~4biuw+q`iiu9XIQYrow&J?w;Wn8Df{w+ZeqMNG_1c6)X; zQy$8?>JPJ%6ERE!|mZhX6JYJejOb^@KxUV)uqhIjLbohLtLm%zWv>a!t$ zN{D0=<1x7a_mJ>#?CD1$BRt_v6#vJ30GvDxJ0cK0UBsnHP-JYAn`~2(pwvaD8qiPK zT&d}?lWn`}%qa!9LIn0o2xX(?S@z|VH;NVjR*4doLd|g~*-DF2joFn5H~MKJjfDI? z=C}LNYMFj~t$xTy?e|)}AS`(iy~U{eAGhoG#ez``u2nmgd0m$Bu~qoVS)K=9Op_-S zlM3cdzesMR>4S*_X5)-0J(X?j)zK+MzAOVjzO-?@Wt4TvN z36439Ov*U2@=jdUop4>~{y$P?o@cu{c=p(^EBCxK!vmEX3@x+JkeliLZ>?y;oUUvi zR8KC^Qo_%9IkzfDxc5-Er?QLhj>=qr3m9xq)fRhD+v{?1?1?s&YTCmu{hkb>*EWWc)6Vo&;a2P86$?VS?S_E;Zd^8{2Jnq9Pi=5JGve~e8B9} z|CR>f;r6=kWFyMa=ro1F2Bmq!whq{mXD7EBo^(JsM>Ujic6NVveN8R3M@FXsbSQV5 z3J@IIDE0U@6!3B9ElHNntr0Ss9rQYvS*+RK({fl~Z8W|MHYg(hJl7DM1jJi_B~xxn zq$cgh!#%yLG@Dd<_%dg`9#6{uZWCEh%;+@5>i@9tmzlC|bs^NI0{G3Nh!ayEGobrLMKXaw#@8MmfFZ1N?L zq2O;sv71b*mh(V2cm<-EHLt+J(8R#?OSi2muJplI&rbmlcUvn1RYU@e;7-Iw-sicn zvst{T!Bp90S|8VgQ5r!E#-L7Ot{+_BICpQ6SK7@`%;EOhTIOKn6wYsb8uP73Gp0Y+ z5YJN~nEGQIf-P_jfBUe&Xymi@qJ@ID`Q(>1WuH-=PUiM!ZlA8X@f>LrkOi@0t@bnb zD%~yik^0|Sdn7sb?J{`Tzb)ECBG48;%;sUf02Bdps@e~4{h&8QBPMzy)-?f)a%^Y{_fu~`{UnLhHB{5rnC;fc z)En*Gx~#0@Ry0cSy&AMwA3XQwxt+&fhK65zKcu$xg)3J@T+>fjkRAZUON(dQ!~V!@ zWx}EZbP8>rM;w)wz!NRj*o*rj_9Wdd!qesg!(H0a1L-KnN7_{|moANge2X0PP6V3H zOtrT(4M&T~QwBr#=jn_(vHf9rU(AdyLNg;vxbv-=wubO7k|vecdB`-2rUxii-ceU! z?AIu**dGn&lzG%&*rw;tkP%~}+{3%z>W+5>3S{??+MZ94Sb6M;2}h(4Zp9WC(|wVZ zT?!TY-{V6?4Jz+Svk3~YIC1Y}1BzE#?1SXqR%+>EdNy*kI>uaQ@PLlqsIRO24`VDKOS_;(}xcEws=0~q5)+OhyLzc%q9 zOUK)>?`= zpcIm^^!7{ADW~!qp!u*Qubip);>^V!+N%;X$po+<=MTD#w=(h5zT8S;IM4hwce^|5 zR1rE6q7gIgVLp`03EG#WnLPl$z@*m@uQDDBqOltKUw&g|wvv)k-|=U7-Z>9`d!suLrrVM`)5V^@ z@TtLiXv%&%#AKJP;_)IUX!4IEMw$fZ+zu}+ab}JF)$-U^HrGi>3AZdbbMHF1mQWY@ zLATVb)iZoARZBFR@l4W)*#LN24mNSLV%oVg!aF=cnLR@EfDxRM_qpB60p;%Ap?(t1i-k$5;mOEnd*du{?(9rw0$irP=I-tJ} z>sHC>+@3kzI*#m^6K0uOvL`=0O|SGR=V>#PoNFG`6!c?V45WoxRqp;qm{V^Mp?SpU z(O7<0#CXKgI50b{Dt=JiNbZcp@BQNj-gyrr-aKjLIhT{?awaM`pO|&=Q^Wtfx1&~i)&2#= zW##@3B@>pXgQPjJyB(ys?c>?=kMf?jTB+=L*>fy^MYL4^=UK4z2SI%!RzOb4qS{EI z_9d{R=k}8(!1vA;TbS9-2eS}m{3~_?cqqWiljJ;N|VgK{aY0JNm?LjON;4EK#v zoR-tPk-dHtdbW00Eok#d86+#a_39)Cn=zr_JwTE)H$jY%2ukRubpELtM0oZGv5hGq zeXyKgD90~>%P%`RSc$?h!zbnD@~}j zp4NCaT3U5ztUV%`Np~yRG~OCG_V>sxuNDVX9Er@;yx|*Up{^hB(_HvS%G&X12@L)o zDO4s1{Mz15kN#(mEM+kXa4KVTE6txl2~5juuvbhPslDs>dOZIHB@}J^p@cT@?MSD^ zN?L11(oMMmpN>YZb1UZ^nN#5dMrDO%4jp|8DGx1s(Evdq5!Z<4Ui6r>M&1m4Lnw># z-x+57Zg0Ffv#2)pRr4Lucx*69)&F65N$83D0p3sW=*ffyZj_@-qoIl6fmnr}K(lvQ z1*13I>j(4HVPn-01Tw|wa6!MER%LgZh|$t`Ktvb-BguX+t`0x?(_ct$(N$LWN`uX! zAkfz$VM*$^eigGunx^02(1cg0KcK?|w4q@$J7X3kRdbf+&koG?lPCRWcg+gMp1Gr( zBNeN&q^AjSzqoj(B|HNv8*g~j)tf&_w1}*CE^>D)e7(%^5>>^Sw+z1+;F2%U)u|GAYL&R6uK{yvoiC#yB6iNu{AL&ConN>L ziBefS`u{vsjFT7D>BIq&*MA^p6f^=#c=bHTL%i>h;o=F$X<13ZeQ~=~KTH;>*hzQ! zr-5c$a%uo)eW6vftYA0+LVJ8HL}vAy%UycOO=UQCKm`3Vt7~?=sw?3KC3}z}{;d_c z)t?4^2M2^BQzlR0b^&XY=L|lQ1i}_rd;2mswG@K8eKu(JbYzXcmw+B#wukxYvl@kU zK){ppI!K$+6=fHYoHueDvxx2Lk{JNOCjYOUi7+*yGAiK5W$$jKfBET}uA_+(h~@D+-Ec?>E82ktB;k@+*(+nK%c8z-<^(6T7(F#N|Gq97xR~ER_`PIPy}5K| z!ds~Gmp*q{+T|e*J*mS0;#2K51{m0Z$*$=x#49u0+^)1*yvV@v*Z)5F)brPB@Un8F=?FrX?~;b<1&9$!%G?&ASpBtUms0zz|Ktr*Ti;g@ z;rIscx^_Y3)j9ZN4i5nyU3Df|t{mQ!z$iiDw`M`Ql_wKjqRBA6fhN`-TftzRMQXgy zc^Z^P{E{5zQsbqTe=A=yAbA4U0erWTuBjNIa%eIlKz#Zs`;HIBNcz;JveMM|uyB?H z^3=1mS43To^UQT*ZMVRs!~#OqufKmg{F9mQ>(HwZlMNTBiJZ-{yqk)LOfgC}zFn{d4OuNkS-rhYWnhPVMwx z1DE_(*Ow45hODBa`c5$zx-F5@CPnU%uU9WuvLTL@98rfE5*+#F=>TAW)TF!TxM5pJ~SO8yM1Lv6r5UGHl5)m}CWH<)UF{F?u4^CqbkBj1CvC4}d zf)OipLWG<2*T@CL2J<(v-nzr5XpG4`+9|2B(Z+Sp+JJoRel&)4p_IS24oKn@tppa6 z2IV*Ynha&Hqm^)nK6RTrAi);0Kj7&7?e0gfEL*rXXf|@>alL8zb%p((0yXG>Y z{yJ{tJ%M%!#E~$Y9Sxwjrvr+<3k@<=4P3lrhGRmZ?kTHs*Xhy3LJ&%-g4P~5d61)r zKF%Dlop#>#*wuP77^XO<8b-9J9}ds&vYnn5KdLAT>v`vX-HVyhsr~EXU_p1(HC9+P z&j|fSHrfS8b)XXz`g>ZJFUmg~OQaEt0-8z?`URxlZl`9?ObRmn6gB>XbvO=s8`idG zX*&w+w*N!v!(~$vCS%rX8(vfDkG&w?2^pTVZRz*R2~ap4Zd%s{rWl}oo4z4h!UPO) zvwuOX9=9+Dg7=UjAsf4E+=7|ORF^?!2pMc|%z^}VrW(R?%`fSIugDKT@g)Y5VDNo)uU+%jFL7=sPn`g5V^@`Vt}9SLx-DykUWEbe zB9P5ytgv}QT?_2nO?qCGja!6oeE&fSR3(@$C!GG;Z8MxQ8{r15Id_~^zw9(}B0EtR z+<(pHdG$OU&dGTdI~TLdv9fQaMuTtQf0^ z*)eAzU~ZBetaf&n*2X^V06Ou9?D!;*!_B&uO8`b=4vQZe0k{dGcjf2g21C~tjFh?= zf$uw|k9t1HRqN6u%Ak(k1(?rr#_{p!Dl6CDPYOlVzV!>uYgFQXQ#7FbGS7aD3@Xil z9Da`N_p!|b+AK0Ponl=#>}`@WYuy|)R#&fviksRpr4x)0pie3I<sDC94oZ*W9KI zoKU<7(fGxmCa)JiJqa}_=g90n0r_GuebW!`_v+=9QdWw$947^j-F}567iwJ&YpbsL zqim67CYP4f9540AT!!|UpkG+-OSsg+PdSIBvowQY+x>U3M)se`oseAeGXi@vW&u1& z4~1H&j0_`NMC=)S%BN$U){ue7^=nqX!mnZH(QR@CK-HcS3g7#3VFhHBc(U;q?K{C$ z+Ojna7#|8ysAwmnA`Kmkgj0iHrj#+DQsj_l`OYkLbnVkmv6<%mh6OiQ=%@PPO}{r~ zewEQU4CJ*H6l#)?3r4al4ShgGe+Q8p!6dlM>Wls*$eic+Fj0sL1Nqyl8?s*ec~f6Q z`btBGE9-4@iy75vM3(*6k?$^Ochv+HVEREt@Yhp`grpD#IJ>ykm0^woixaBFtQlN< z2D{o5ax)>Qs7~J9Xd_?~Ht^FYwqPJ2R7OQZ3KVy!KBOvTSlgLtzyJE34QR^DXPMyC z{A`+p5OKCGCt8f#@maA$kwyawwKxR}F=Y$+ww-^K721NVEYX~dDOR$}HUBavpy#xZ zPc@oTJ9gT~amX$=rulOX@)aX~vfX1o@d;foaF@#`o(XsO1ft`l(yhNY^(jVw0^p0> z{FY*SjYs1m*Dlmc2!+=F+lkF?$L_Hcvf*r-KNgT1J;)6fpw4bu2|2#>U+tl@&1?~7Y9zD#|Sb^b1P!KKOG!jUz44R zbvaM(9;>YKmS0k)Po@A?5@7hMm)}HM=*4QXKp1N`a7|_~sSN&a)obEdx4Vc~k3rq3 zR-?!U9dP*?*h$vIf<=&ITOkn80=1a{7mA3Gg*8kLD9Y3u{y%1A&;lNizW|afSdd>B zUE*PJ&1(a2+B1iwFEXOFj+Rq`VBU#v?3P)N*RDk*1|Qiix>0}g;5}FG5%>D#i#fZm ziSY@E`>Mh9S`2$*rO9_K4hJ&!pX*+XH>%TKe$Q{s(8vTf9@+M%o1}po9h=W1Bb4G* z3(pL-7nUuxPms}vbR`CC#wR{JWJ=sa=YGyxi7x_s%ufEx`0Vg;n?&y0DwiHq9y%8N z*mf0$+MrHVblK8vqs~d^D#6T7 z0w)adKi@h~Y`0ie;zH+$@^9Tf7m)Ru&4lu?$QC9@D}O2Y2{XlAlg!LH4a|$k?}4jL z8aRJ1emF4cnKq0wreguc8l}s)&`Rq*eer#bbcES->W;9GTj&ZM zzbH&5oYhq3y2cz9&(ayS>W}12h(}m@=R(sfGeJesq`|wlN|aT$1f5iDAh`~I@03HA ze*}$qU*@WQA|2Cf{R02Q~BM(NMI_avR1EH6+MbH79C~%EPW#}Is!bl#b zusUDp>+>dsvgeDkMHS1T(}YHsWIDj_O|a3RenTEa_La3634;0V367e97Bwnorn%EP z@gZe@E&i<1URfbeXb@@N_2PI|QHIDd$vrZkAOE0B2Dq&#@NZbL$%1NFxdm;C2M>+|wxEW_G^$7yZ2-ru*?5Xa zyU_r=bv#H%E_Jrwj2Ln*5La2I14IqEksJ9%WCw=};V?Bp)aOSUj7D*~@r5~xmr0$m zUk>BF&s+*_oKs|&2ogM&Y`XbYFM@mf-qTkYIOK$7Ihoku_2P!m{L&I0iS$yo4}<31 z0$AKrqkuLoXBEqMNH@Hp&6G^RMPfmH94=AQ=}LaWPXDTLR<~7IkQ>pD|6vjFf-F=` zaPKnE^(M|}dH=T3aKnDi?Q5QEGiu;Xh#mmYm@7|gapo<85%$=`@0|Cu(!bE(?C3ou z1)Sa1=?;LwmEmI~Vx`w}3 zG#>ACU=b2q88R9M6L+nuq*{`}{5m8od~8Mn!v#Cz+E)i&?Uu9+mHu_dAs#WKx{ z^;3eS8pVNgapFLsg!0seBPTg6pQQ37;GJmASGpuiX|rlCr=Mm>edmx@vC$^&-m0NZ zK=UP=0}$6-wMW&EHka%b#xYjipcOt=PB&ri;2$vRc#Ta5G`wAP9zsV(_aM^T$h0-GC!c!BP4S6- zHE1PN#&aqd`D3jPx+P8q1tMQI zXxe*q3M7-is#Mt%OJdl%!;Dc>}-|r2^1g70H(d1 z$l9VRI8*5l)K1<^Eja1VDibKBaf;q8kfqA&m2>&t2HV-7S6`dpM&G@)*^CljZvdG^ zC38-Y##kwJBAJGJu*+&SO;1t+eu&fsudoM(+iWaKX0rCXrYh3CMbuK!;Y^S%21v87dxzH&p;?hWgI}X{=w~5Q7 zc@V*?R?kUC1kmXAghtU->)H#)DH8TK_TG>7h)ne^6Fow1WYrl%cMqqeab1YVv~Qc zZzt^F9k%Jzs^p$no)U&)`%f*AG_U#&1^y#brY2}GcQyqGqXBCk7JnX5VZMx!N%tJz35)9A}3>?RLV&2 z7w|)TqFA;nA~Sus|3ODKYx$$WyuR{VVmQ7*7M_?B5n3oVO{Vi7GsXUtp$Qul4{OO5 z26j-9#d2Ly*f|rL;^P?*p-TJM%w>zmgf8zyh*^8-?@@Smi@!nss z#J^*DpFGhM`2!RZXy`nSi%D(9z7Jb1K7EH6PI#GAIyo3dX9tNsv`t9t%R1tC&Q?mD zW5jdg(b0e0!CVJ4*^zI61RkJ3SfIg^PhEt=KqxfN9G6WiE=Bng>#?fV-*oiF?DS6_ z9AiVzRn8?|$;uu!bEL>^^-Q;qZIg~XV&^WGh6{_g9q3@~zFOxfmc|;_2~f{rJu4~8 zlOybyj2gqan~$;p?_S}_*{}IEm*j&+V;MM5pfvtoJtxL|@yZc8*Z~_#055g;+wP#M zK{wycaGyC*hQzPO2a-nPww_99G?%;OZneSeef;{?b=qXTK*;97@aBOZ>|Vmd`Gya$ zMxXp+X?~3F(l!<&x$FOe%|>o&zF~c#=}VauLLfnBv`N0@7uSzZNJ5b!=*4u9Um02H zpRPgI-d_0bF&`I=2bdP3m(~jgybrAFTog3(ABk^nN`q`8x6(Fn@DbzWX;Yd}MnbB+ z^RUap&RTmK3N+XhU*>#w_^X*Zg?s%7kiPhw00%$;F9Ro9<@9Z#Qn6Hli1Oj=-(r>y zdU6Sdiq6}^?aqcP)fLp{{|HSFsdK3gA5>k-ZP2i#1B4wg_vMD+9qA5d+k|rIpEMF| z)cXVZ2o#lRjm5``kOQ>`a$7a{5@pc?Uw_tUQ=IZ-UwiFEcuJSf06FJJ2OgCG90Z;_ zMd~HzK*mBEoPY(QOgllKtIv4$w_|_YSxVTQy0QVI>YI_1Uwh(vM>2WpoX2GsH0Zj zS)p~Sdz5DGUA9a~{&Qa%7y!>924!iEecVQLHPoGAP}kv=ReA443&ZCz#ogmw0EfXN zqXIal1d5SjK-!$S@fiqB+%9Sa1{w+Qx(rhP@P;0S(?L0G7YF02+;w|O7Ixez_Ef&F z9F4q@o z6?`ILFj|xM>+(J|>^8hl~5x%SLIxmPT2OVok~qw7V~aG{#^ETmAl2D-Pd<+U;6s;jU09X~%!Y}R zG8)alhw8(Vo_yd=`4sUN8?*PdB}^l-zV6cb;W3HW_m=oXQ0zc76DSy(XzTj##&f{o z4gjoBiBGu3n+V+8+=k9^&N$j+6EP(-lge`4Uu$;CvXy}BexQ$30A)-b@cal=B?crn z7nIL81vYfiZ2Aya#pgHc62@N1LTE*GY4XxJH3BaQ6)d3QRzN8&xKQ~b(pSzK^V&}k zO47R6o74ncd}pFX`Z8%xNfjGSAk#`XmPv#vV=(fxDH2odLsGal5dfK=tLG!Pi0BnT zX#&uTh+jJ|O;+?vjK*hH>N_18oZ|FW3>-iXdUS@nkOGt&Hfe#gOt$s!8M*}(;B%$` zEKd}taBJo!hwTwowiX&{r&n%~Ez4T%2xGN^6JXmToK$DfmGxh6tPs$1hZdc~^)w3F zF^Zcjjob^j8Ccbaq)$JyDrd-_-*J;UeHiOP`VjMP#wY-%yMXPFj;7}C8xG_tVyHB>^W0!w;b%5EJ;89j(h;X%rroc{4zws@4-X3-zVAow30Y2%9qbZcCqo%r!IHljbh%X+9EF zA&^hgQ4?zL@7ku^?Kcs=(KWoJCLd0x~(L z(5eD$8;-W{vKNVRrN2-YKnv5jWqHJY6@l-ww?0?LFPn$YwTN!$jhR@4q$)b4q!o>O zoPM+P93LV)5u{pa&!ncZ>`vEc*Od-MGiRHP+jb)LR%~<)4aGjp+}}x}_a4qh8m_cy ztr&Tg8LF=~-Z>t)5&HCg4KiInO;t=%NSaTw+ucMxP0|x5v2Ad>MZCFkTKUxxemh{U zo0+*Y@FsE1W5x#^smNf~x9jxm^Jd~*PjMuvh?^|UuV%be*B{M7PSM8=v z^PwN?lIl1quzkWjoTryk)^&J@^oUecJp9GdLrl0d;<&-4GcJ{x`6~x}9_V^{MK|9b zZM_SEAHwrRnTHxat4v&oDJ|8@Y-V0=P#UQ@Y)j#xI_~Dt$3y5IQ8B19?F*0MOtz-~ zY(~jH_tkAqKlG&k{?P`1Jg=}d?iGURF>P_Y$K{Hmtd8EjL|)Y5K3pkp@e@(YRoW92j_H!~4AaO0rNq51-?JY3Fv#Kc7#b9fk@7k=y?Wv7u}f+kVCHjtz9Fw=cXIit zb27Jp*w&1gjBFV&$aaBGOyWaYBxkHeR`lsD360wPb+O?6G4iUJz@v-sYlPf4Gac1h zLn)|?96dix^5n-!>78e+I!YT6^DP+@ELa<`Cq+%CxQ)QDYXs&kkMOITek*}reiv&2sDAP$ zzb?Mf?Xt#QcT1aog4Gwwg0#fIA;JoMc5SZYuy2q2z;!@HNsXsrgw*?W`tT0h#?lR@ zSK;g!hHbod`i{A@x`eDWg2UxCPKRx*gP>*@yw3(Ch)(Q5c}$_nC-m9b{5)wqb&^J8s) zCi2-}kzu)`C=H*$1ICHftag$ig)t{cokez8|7J z{X-tiY{rD-4fwWe?-k^qvMI!!DJ%s;&e&K> z|C|qu3Z+*F2YxY`=r;Vtc|rqL`t@V=n}5AsArO=EYPQQ(pLg0?wZ-_jh*qBLmNU;N zk&I6-a*gUbqZ_0~&=hd)%39@Ng!`fg-KSjE9@d{bNIr9zsL3`pvlxp&*H)VBqio*K z+kdVmZF)|(g-o4lTy;s(tS2Pnf{Lf*_goRAz_}}E93?aqWc`AnjIq(}#JKTf^8B;- zLiRN*ORU469h)cQas7Lw_e4@As+8NFULPea2I3*UA(%EUx=$jW8~cp3v^%4@kD78? z+9Wp5Uxh`+!#6FBYz%dtHrXCdBz**R?9wX7+rEl1JzXDM8;Fe9tA)4hxH~p9+E!M= zws~D$tJUqGvj(qi`IjdZ*W1n2*vpK|(O-6X={A`VJnaMs^#4AZBAi`~}ut}6l)}krD84S)J|gV;(31D(*SXIpC!q+iSHZt*n{UGMV&4CN zeN;~$!V)Hr_-=NdC3ez1wr()6|I3(SXLV536ON;-UQm5cS4i?z>zU6=$F%>#?{WSc z-{6V;E+EcABmzU88?Py~}ad8)CPAYqK!3X|EMZU;ng53<5V&7(tby38IBm7ZyXSVyzG!4StbLX>V!s z(v9hihEWd3*z%F>m^|;Fhi9ZYZ)sR$gwEl?vfi_loPl?!(r@$wqnm2A<_?H8diE5z zKVc>36hl{C+Is2F&!wfn%G$WunR1MvWaq5XB8fC1MSnp9KMl5vAoY-7Q%|toxTN$1 zCA0N5=-k{vNHDKcMQLf3C|D8~0tm#M*E6P~);m+Z76?=6={_K+GPZiE31R0^8m|Q! z%@9+X?fwuD3qAX@qe>L(yD+yzIe0)*uD2qJ&uym`=-`h!_4(PZ8;gmw`q6F?UCF}3bi&+!@;C?jV;gfizS?6?(caHrb)}Vu;@$Mchbf;4djx4>OoV) z^52If`90_I@K58SHYgJk4Cy;UQ;Y_dH%WCaZ1^sykA04j&kk%~tEH+%vm+O-#UHwA zGn+sx!`Sg_q@8Z=lcNUSWwL~*SeyzK#%VR-P`{b-pN zTnT~DsA&jdn24)fy{_tEFs5;4_x$!NO?NAzLp6Va*0T=`cV)DH(tVyQEZZq)TmqSS zoG4md_jW`tz6THL@S0$hYTDOSGM5K#zvVrBVo^(PLX9ZlOQ)d)_jJ_yFWBB}qBaO% z*GNP>KXu5);3N{uqK{3kndSR6wSJ!C8Rp!PFwL5X=Yv+v{a-vy%L=pJH(Y9Ns!Yhg zaJl~7>&nW7!v=M(IeX*j5^IXEnTWZ%#_=37a92pY;06b`W(VsONagWV&{FiO%-ZwS z^Z9zZ?FviVv*o?N%kvMmQG?M6t=oD8ljYwXL$^2MM(Mrddtsmkk&Ke;e8ZAJ%1x%B zzLyT02r9S=_X*AX!mr;>oO1?I_e0k2h!jSU;M{j!`<`lLkFuP{ zciLsr>CDJ9ovRM5&^f__dR`uZNJ%e$a-QOf@+;NrKaEPiH@j4TuS%=nxg_HjD{91N zZ3n*Tlcj%qmmgiy*>ztvSZxhgA|@|cDylrlYMS7sc|NOK3RI#8Y*ET7Xwz_@rcH8i zP$lAOxqijo&KvEN2B#;Wis4f^sW_L(4FWfxB_>Z}6W^@8ZE_{|zt*RoQ^YL9LMm!4 z_j~s@WG?9QJPQ!`3!_QA7}Uhbla&iAGT|jpr4u<{ z%|~-?b?{AM25C(MQeMt0A-IPG2b-dypF0Q-)?`V0$j#cnmM(QyZxK?|%=JP_xkE8k zGH|n8jHpE%&1x$7GDqoSpyA3liS+#R4@*p&UN_Ph7q^@%ijAc=9>fU(zW!Je^p65w z44HnU5J-W@Yf^pD;qk-n`o|8<(<$1Mp29g=)OA=DhL85AdKLA~{mN?bTt2_q&DK9; zH`l%_vvPg)?0TT9*5tvrpR^zEaF1n(+!auz()UCBZK|L;If+kB?b{Bf$Cuw3AM{tO z=8l*Q>696ulsr~8x5gz?$Eh{f17Z*kiG>J0>IRA9+u+Rtwn-l)(=0JJOb(loT9pGm z|Ke>-B{LI5;!V?BBwnI!5VX{HGDl^x6PAlbOZ7WQ)bcOWz9drPa_Hs?;La0<5nIEX z@x<`G`X_(N|JlsLWWhtN<(B?^Jb|jpoe7(IG$h(ra$6_+Z)gL-O-J8ZMo1O$EXm&P zryftJ#`)Mbp^nMBq(!A=AMP5#TK$bhb&|~UD4&oJ(Sp_n!cp%F5}lQc+)8H7RM8+V zvnspuGdxs0+^k|5dd#dhG3_y?RyN(G&3sgaG@=F;aVb6yrIyBc+d0FKk|n7-2b(A| zny^YVX1j?qyy!SiZLm(zUs8&2zT%8}z^jObiac-9fd$#j!HJ9%NyXHg@UT1T*7}jt+WQAcw$OHZESc77AsocN_T^MWWnarBjyAoYp=(r7|IdjkDP?iVTB#b# zK=twMVZq9`o>FetM!tatVLl8Vijp_xtq^hHP4t_Y@22m)PrGdX%=b*?7ZWXO5sbMX zwKn;Lz?+oBAGPE%f7}s$IymlGiwinkwLo6eMD`|<^}Z>#4;e#eu}zTH_0v>c{oRmAw zi4zDD1Iu*WI2h7>a8guyv-d<6mZy9UK1B7G)NFy#%tZXm!MnUd9I^P{JB^5VnXsKD zBd>~2SjX33ocSBGeu%inW}U>2;B=t5Fwd&wR%sloU2}7+W23&`w2^`@3z|pGJVYH6 zmU9Y=8b%-cDGXmh4^y`mQ}+}yd|!vuM8?Ns*Jz38e3?m$f{Bt_EQ6G)oqFSXLYbdm zbckWgR>wx#R{an8RE|AuDV^v7QwK&|`?v6VR?p+i1-ssWx-^HJAzD10-YlC(c%G)CV)t|6gMJs4%%WjPX4VMmI55~BaCpeh4E_>DX-^l;36RN2^rsei)^8~2 zG9y4(#`-N9@y0kp8L5QTG@MU&%j|TDO4l&#prV#}_tf0}oWMgu<`1yK~=W&}ws5`Tx}dvkA!L1=sA3aA5FPeCL! zMM0#Gn05RLpQdyE5d1mS>?3vk!8(uamfv8WQa{XaC@8Xn5J;j? z5K`Wq&z{k)9H!zDhp)z6ZSB?|BMURXNN;M-_OrD}!@EPK+M3pa$9Mb+{j3PvTVqb% zZ-0b&ZoH(nYTVYMMDYwXxjz5WA% z=2{hh6gY>RjhqaHr+%mF+|o|FrgNM+h^-rnpdxU;TJJDG)|&0jXV zuI~3QaoCD!m*vq|ls4Thgk;Ewc%VwmM{WpP5JuMFi^?y2Y+&plLVXBMU)e1b2`d`& zD4%(oc%5F4g!;kB$>s7J|MII=MUk{j<+PjsMD?w*_@vd1$t}JAya5abjNYVcS{B~9 zu7E%>$A85uj7a5(NEU*k=-=`@MYbc2b%C9C#60bB5>us2yWMr;y_sLBIN!D2N@)+_ zXr!lr_rB~-Uq5@3NIXlJog(CW^+_!pbCPv()%0`~gVj0;n$>R>3l~r}y%5A$EKG!s zDFB`&$*H3QzTTl$zsF$su=d5tIk}>#f&Ck~f=Szbzp<$$YFabyMN;@oQK4^Q`c#Fm zcKc0*a#Qkt3aEG0_@{)cN1ZX-e0?@t9~PM7YcVn#vk)#z&DGZ?)~w@T0?fApn#Bm; zg)&KSkttEPBMU)?CL|>3A9M;h+@jdPx^+GESe-ews*&Tup&n*07`KZ9tbI595c8MI4^0387cY7`Y^FQ3n_ORVnZz>@|R;rNevrAK# zI*+ymQps}>)9!zvGl1aT2Q;Ed?!{_6v1U!kHEWThDXTE;Dc#OXwQBY`Cw$Sb9B%7T< zewWQ`y}*5C%$@k_CwoMv9lk%oe`t0=V~}R#C)WNg$mWLb315h+fE$66v{Pr{71)A} z&^L+e)9}`xwbbZ7r#@W@RNQU9eK9hgub+Ay1=*W zVVfr$0NNNuJp_uvFVw6A4X`$<>wwZ!xK+yW1&RY3GW%itIlJbPG=h`qWS89!Q(|w; zLzkb5cs;EwLUk}~*V?aKtO_bWJyHBg%ljF@Rd*d8Y-)&oh~h@TYlqEP?195r=7WI& z%Iy7B;d}=Wvu*DK>b|NBAQ(-x#ZM&|jLcwVWmO3QU?i9Y8`<-gU#n&J9axDTmRT-! zbutqYzT0*dfJKn}i!LZ9dB1vY*)^0Muy<}ZD!tP4P~8AdFj`C5d6Iq1LOG&SYUjQ7 zg_s6Jg%D+TN>qJ(r7D%5-j672b}~23X}I7#oQW#&9I~2s<(s@#*r$5cr<-5x{(7ki z*IW!ZfkR7ju3+^fP_fwM+hNfxNE-HaURlINDIY`=iDS^WXm2c@3VPZycA2ScZT7SJv+mBtYoe5--(D?B|%b!H70_;bA zv6{1Wyy`5s-+ps*H>KNi@l9$gB}D$XC8AdX{?Fdp5ddp?p=jqJS8Z>TPUJ`_18X(= z=d!6zLni8#kDRE}yw_u4rYGjTa@`&*u~jknkk?^jnIOcUR(^I%cG)_TS9x1U<=~rA zu~E$pfjot%bFZJ?TmwJapB7B!C+Bs%Ce*Dioplvt78#ZeE;pMofd?YRa) zDE$=3)opc4WyRwg8cCv5lbBIk)#1@VbNf9+mO6CW`Dpo-dIj{K2k7Tsyqp*}u0%&4 zcM+Fr_sCr}dMK;2PqP#(zUnyRDP=t*L+|m~lq%A~* zjn5dtA97?6iKiyCTHZ+ODb=WN?&aCQSl^p;|AtE%W0=Y>qm9mK(7R&S^Iu9;jD%(4 zZsaa^zV5BRPH+1K!!zAC5T0SmrXl<~V52nb_}8_~1CqXwLA*mxdi}7|@Fi%Rs&0;| z+Im?m(h@OfG6NQ$HegP_fOJ~_gki~Fu3RfE87%&AVEjsGr2 ziqt2}^bw;owY_h(cVdq(*epRX(ltf+#P(xye)$&DnI_Vz>w#d3<#628CuIWi{Js3> zL_Xy)($8tf)=|ZTXnL};HHqe3Ff@->hSs@di83uH|xeB`db{b|N_EGlGc}dQ{1* z>44dZWiMp|-s!o|Gj32=Bu$Z!WO#7zNt-=UtHTn9Ax~MzR83t)@|R*~AMKa%s6Ss8&=O+Qa#K zRATZLX@`uhMc#buN^JqxgXx_`-BIfeaX#%~-9Eh@%a0LLCtqckL~;HJ`Tyv8%cv@& zu3cC~32Bio0R<&Rx*L>6y1To(q+2DWL#4aByE``B-LPrS+CJ~|jd8y59)IyeV&C^# zb6xX_xmy06{5{&i=&T1t5TAxk4Q9s^II-v-s^+oXLH5t$C5f(#4*+B9 zC(G53xPTeCHl-q?=f;U226N1Z8b2SRcOBRQLFd6|x&bP5cPRP%WG7!iopsEDX+uYn zfMYZ?eJf2&X+5~5n~n{)j+@x7E6U`+>9;NVC5x4aqu6Z8BRz~L1|r=mc2R@~ zp_tb%71@?rGe2b3WvHP|KSFkcY)g)hgSC#~=iq)2{R(~pSr%wFF#xo&SoK48*|f5u z=evjdMft0KbMN>`(RNHyk|-|a1!_nns%V8amzGvA#Ov%@xB4?yly8p1uM|#XBuv{H zkHvSMo68OkPC8(U&5>i;HqsxoBVjqk(qv_(*Q4|YZK^cK^$d^jD^GR%%pSciuZoZ7 z=WykgxLvTNRQTT)Y1*-Po&C}Fz}ZbEZ~KIa;d$-doLhi$`0ZE!MtExHFXWimlc@uq z&omv2fk4FmNu?e&3hL?X&*lCfMro5**0wQaUV4xw4jwm8jJ4a*; ztz%h*i}*n~>#gs_MrlZ1nYLPWCeW6%P$fJ9MD!5t`u=zW81fX~-e5uJIN# z^I}W5^UK{Khg`&2*Z182CuX8v-G05&}vm*8=S~F?9@1?hx z<+{5ZZfCJ}`?M(|BcL*}onJ>|MgT$#y8FBy8mjpCa`&Y81U^0Ap9s5DcWleDTPAC= zB|yBcj4c*DpC6=X_PxX)*lx*czI$o(_g#w#o||-DPG9csdm-ID-Pg>d{F=tta4BCt z4`@bw?CM2N5lMQbJI;}zi286#_AMDlnF5G+6xF&kpaJVH7<5SIeJvqW6oV~5ag8xw z0@@4r|7kB0UpZ|abm?33MaEAx=M`K9(^S_10uu#obdLXnrkocg1LEZ9@pP z?$XxsL{vxj44t}EGIU&3l~NjzNb=L3zj*xqDg~vn${NBoIwV!`t^RTLu{B%FDg95O z2vAW~k$wSk-gyO_kFrWPuVRJWkuO&ufXEYx?Gy^+LjJV5Vy=9TP^4i~TlVQ8QR9Lg zpsgT6XIJ#~D?MlKj>Kq2wqwDg>DT|)3(!t=#T{;H#O#)!j*T7ndZ;9{z0Y&~lz@lM z=YH(UeRC?-w>Wd%#zJ%4xu^7oj!aw?zqY+w;&|kYbU6UZN<%}DHExE1PjIDL!o);J zGCRKd{WVtLcOPHAbQWw^?uGYxIjvidOWL+M>B6cn&N=DIe(E1j3GoGr@;G1%hvg$} z98k%Q*dYR4z(8F$QNy60DdL50poziU7Wr7rlW6D1fnS_{n3ma{$$tP9`(wHRH=-=j z+=Uc_Sh9rd8E)XB528G1e|B;IwXXPF%p9~A0s}GB_m%OJEu!3Sd&WUL)9i8Eam;6OvhKPSh0gFor7B%TocJz!! z`DjKsGvrj@SUb-yr}mEKr=UkV?HJ-!8$mOIq)Slm2a!zKPYrf*Qn( zm7I>zOvO4TOgucgP|R?TMQ(6d*7y`tk(1T($?8CE0%i%=L9{9rRk}f(-$eiTYT^dJ zR%gwW`ts4dK>t-oPUB-Z!4OOedIuJ!N3w}MWgAk2T2=wdU=Eub`YX}P%}1c^FtS;) z62*m(jxng!cLq@{pSs4c?gl8M;OT|Pf_;sIpEtoW`5_-pBJ4}7u9Tg^Pr&zQ4fznp zX@%94;B7Zn8^7PXv}yrQ%Wq0|0F{8IHPGhwoE?UBDysoLZ$$Fd+hplM9dJralRr*+j284Y8fS3~^Zt~*W4j~2ep z>ayr7xn)AH#8kI~Op4zbWfPe%26>{Y+h$(m6wGJziuiNtMMjt)r>v>Gjk*?nwVnLM zLpz7;<+ueChS536QeH$KM;4V;nt&xq^mo`Bfd`0+)EGyRD&$vL(1nd*#~)php-+a^ zF2bFC)5SMs$%9#wCS8?++mfqW!a>3-k;vAqP^+uIWpnW+BFYM?{znf7%_H-i=TuK7 z%~-6|Y5uFM zdFB_pS7ikl=QyPixK?dwIJps>wQ3z&^ciHT3F!_+VUhr`BWYGeAx26n+J;6Vt+2|g z0(7xDPn^G8I6Y3r`ea?ntBpgmc5&Oa&p(;PiUtaa4qgqAQ;=tl0tI$`t?ilM$whwi zH;(bm77L<9*yV}RRabF}pFXew>_@6jWvmWMa8T~sQ6wz&m-(D?BrOO?*r zkdVv%M%2h;XSeouY_9Im5jW9J>GXEWElkvNFZcLtxD8^F8LfI8_dT|(+n8nz~Y7OfMuOi?4h{B4c@~z^ZFF7YT zn$8HS(_%f+^LY5l1-|dIDT~@$G0Ig~OJd!0E}>Cz5jKL4Jkh`O~t&1;#5{b->no$rg>mzs-jN&gfjd|q2F;hOR{`l(&q z6L+fB1Y2bvj?3ZM{qrjyf}^92@*@jjd5L-`^yGXEbKBDkD^u`x&{1tcs%&F!@kXl0 zybti{FBWuOOc;);q2k5EGZg2~V5dh+0$n8T%!0RfA^37=5l-yZ{_J$mcBdf>lLYT? zLAWH)gsarJlzv7{MYi6ShNb6L{C#a&I*wt09z(a#mDG2ur%m*H#z7i1(uVHv%Fq!- zN#>7X&^OeByjaqn$LWuBm)Vx%62}!NJZB*uVd90puCtc1sF>7Pm_NMPZ3^(}(MjX& zwM6xN|e;+j!OEzQlIK^9^?iLT-H7P;VPGe@Vg<5?6&bYsbLe!kejVQHeU(GsqP(=pMOD zTI_K?dl7oud~1`D-1{0A`&4TKp zcj5N>wQ|qMUd;9EhpW{}+#a7TNmR|s_ZA%AfaEPVrsSjKe8f7T_|A$5ve!1cK)=5< z7SO;`q@_M8$7a^gV`6m-E_j5GRy^QIRcQL<@eNVjqVCCCH73KoN*nzO3wu_Q1w82& zpWP}2hRn@&^TdLTxAk|KLLihz7s2Sbk!e;HmM$a2l~tZ;qBj+Gi@f!=T1WG&FwoE<@D6JO84ozFZW(?D7yLlox{t2Ihffm(klqE65I5}R3G(^A6M7YRvj1lo4q9+%G{@81b4R=QF5{xxdrE! zl}QY3TSaPt>oYchu0$hcXcZdvUiRDv;H;6ZN4j+v!8>RFVrpzEQmQ^`*`Dt@zicdN z(xRgcKi6eHUOQm)ql?IS9eDHWE+5cbt%o1*9p&^Oenc zt#TWGdFa+?n)l@GnW)l=XGuDq^W8B+<6+c-S(~OZUH5R4lixhwA3#st-aKFceL_%C z>;orvHlIR#^Kd}oCpPLZo=@|pMMt6%`b*R#jdCyY$Ll#J7Nr7wKC-NiS0YACwdfGc zA|#mXhrCRDqKT+S_dy(6YTdtr9joxZ<7mRSi@3*pNF=*nlSBqBNC?+o@>XL2UF}an zxStie*-K;994#Y}8*lY9Xi* zN|(YfBtQP7h{nKlutG?iP_D0sd9YSw*QAe-FFw`O&vtt-gP5TTk;gA0TF$TkL+E7o z8k$>?Oj+~2M}VeTwhsyA6j&-;x}E*e>E3Ya_gQOxeE8%<)Du7X!5Fzj{rJkdDI|d7 zLto!^ibM$o5gl#&*zhCpf?4M1h|Md@^?p_P8W5y?hoKH75z+7O;e?yF+aaI#>U^at zqL2S(9ur$+xS{cB{MmAry*xc`#W zYmbv5Cc@7Oqz7b@!hq;NQGKKVMZH_zJi3@`a=&Ne16OlPWHT7$Fi8umdK7SAkN$nD zQmL<=%MzQw!pdO`BZ2Os5y`w9pe!!i*^i^f5)#2*={F#Ab6h7g`R1itpLGGIFVWMY z6ntjUw|Gr4m%&nnGGC@LfK-Wx>XY~Q^2bz_bVIc%x)*eKy3`c9f(Z~c;_p;%XRJ$7 zt`DU3g3UnoIV8d-A!+>3N=(uHgB@YjGXI*B zf^#Uk`9+25p~*p4(_o6r?UE(C?F>tvY>h zXPl8lsVhaDLp`x-9i5yq{$?N*7cbD~a}WLfnyd3^`*@8()9rbPXsJc7C!&Qg6%KZtN!eoS0&iKIzFUjE| z)G(*8s0rzssz26zo9wWAKU{cVSXxlRm`Pj>xg^Vq7>)IiKWy*Z5?e+}GZukxX`5eJ zY>8jYz=+fDr4eLj(By>IOB9>gLv>GnVuXAyAzX5g`)$IQuF~wyz2~oBsSP=b)&yVX ze%hIer|pRhc_$C!f=fpi5gT2Bf)9hZ588bZkk72uQ0J+o$HNkALd?;(%cl-~s$8q- z$NNw7nJbV5i-q#&)RhjyU#pj1Q&oWrk_iau=;+wJ=AoM^sXNUNr_@fwAHH6?-S#<3 z*sLU*B061?=E2>ULr*fa;N{R(QS!r`+HXTp zN2jJ~l#+>TOFOKU0|w5ft>qKD{a%``E}rg@(OGZ8yJA7j-~A`Xj=?`NHR}!`{2V2+m52 zMMCM|9D1L%0FTTgoPN5o#5-<&+nHs(5>j2Qx7YOTsZdJr8!n|GX%VNd!~1RJ-bO<9 zgs0L9=SYsptW0sk;;r#TYVT>hq%BQtSL}G0N*~!`(O<``j+~SY7d4mXBb)Q`yd@p? zJ=B&|d19r)fz|x58|$*(UEO<7t=&uzBG)=%cGDz9UL@Xq83L)`U|VvpgIZ!> zbJ68-#^gk1es)BR$T`H?MLeKe_>{9F!=FRI6t~N`_Y5w4GnJaSG@vt&(KCZ2-->C~ zp07%;6IB#XQb^fgWJX>H;6^OQLzIr1F`bi{jMJQ(yVT@jse@6n87w+Beduy)mD)UE zRMpZRe)-_S;}PdI`bl-s(pDTH8|lLOBLu;fcpV)!(rrDvliyab{`R$_U3DDh)QC^30vDVdsTa2~PtWk4F=w?X&Qn8rn~t4ZXf!CHTTFk^fD6Tb%Bo`9ZZI(EE~LffrhT-0rDUJZf< z^dq{Rf5PmKCzI%7o(|8RhEX5i0c_KeL zx6jW|HN%Svn)tyMHtSl%9~n<)JaaIhyjSFX<^;F?V&oMeQ7=28M|IEek$x;f%&!Zc zW#<(kd3)UOG~D-ERXCn|Nm(-De`KIR^1fNwGN@rCI$meZLIy&v12DOMlxT$02RP9A=GU2y~mMoeHdiLrMJW07q4xD_eM zy?vMUgfOIwslLRp7>(yuGC3IkB|WD4C`Rf}@>s)j?6(L`Kx5<~{KN$s)d-dz4VV12bB55#cL%I`|5@cmx4Xr8Ci=l zY^&9HseRnqYs~yG>P?gNO1J{|%~^4q#TZS#-|beu`N?hf62GVZ(X#zy<F6$E*H zTyDPsZvxsG%%@U12?SrpD}yB`LL;^9x}^Do2`FjWlg@fGD=EhCqJ+)i^#;C_n58Uc zcRQzHjhoIaaDNMbgnUKGIFhJEP85{^@&modj|i z-&iSyW+^g5T42^>uixY2zKM}`M8KDFcq7HW;(4;LZ>E#OBu?H_YDPSgi+aioD4kJZ zPBoQ4&}E$y?jj0EC!siWaZccylgHFvjEU5&h5msHU;B_EEZ#-`G zfz%T^LBV$v9uS+Dm<9B2#Kp3uKILsTo7l6tE>qQK%{F-x@JEpyWcaBjFf+oK>agIA zyVElw(?>5fn)ozR7Cs%%UJ4*o74Nh}Msb$d7dqe8VAW4KuC`1SdBKqpXgg^>SsgU} zTs@D7#L(g#J<%JrP6RLzL~V+Y*&yoeRv6fe^sSGSOOXFTkmT&avXEC+EAI+OfXLW0~ zBl!RriWM$!X%MfpM`CgNcxdL7eox^NZU4H+Bw7krtoR0xOKE=`c<;i(SUWAHdKM68 z6hlSFK5^^PmDU;;IaZ~i;jc=RaforRvV=BpF!%iheu<2yU2vhTe%8h3toMKliBV=f z{ld7iIPJamSSK^^l)AV``BEA=OtIa7P^)AO@hgR8Bo2<)um#ak|CFw|)ppk+YKqIw zFx^abEpkL935R`9R~#aLM}u@Z)J3mO|0%Hfz_{=fHTjna^!yG~VW zvgAxfS%P8!#Ywu<3tpXV=H#4j;K7ViJ-qAS?a1!s!PCx%b%oAI=y%C~o?-%Ty@`Xk z`K%tEx$3;N*Q1LTB~?aiQVK6ET{n4a?>8Pjk#>+q`#3x|a20j>Qz_WB#=Mg%S1?)w zu0|r%cuu7}mOg)rI#E)P8@DcD^87&}u+r{L*oBkcm8UKlEIle{O>|BCIzS?=VKLGV zQa7T&=d#B}z=63Nd})a6M=ubk-(=CbIV?%QWJeVC-D2?DsJj{l-3$srG`AEm`vd^d z@ki7v%PHad$`r@F?~sxgAXL=H!K~Cl#ahTo1)E71Y$Diry29#FVrt!N{eXNTN2Zbx z9Xe6ds-7ej0HkLu&^g+7XUj(XzeeTJGP~k?BNBz)q{}R zCSRJk!LLO-cW#b@*?__MJiC`6m;WQ~b)<>=zOi)zdJv-h&+^KeQwd59w|FE>q^#SI zUp46-suZFmLk82Up(*{5mys-li z9X~~9WlUdpX&{fr9qb}+E@z6&A|l(p&u?U@uF~)J^{^Nmlvfh%G8pOKi$p>vIKjbL z_wQuBN*V^`gx#r(XklSZz1Hr^1u4&rFt(9|Ac9~#dX16G0Q1`gX1mwHhrI6&@Uh9& z(cd#V-#I7PK%jIrH~mHf%@aQ79GtcIl77l(M0_Iwb<|4x;n@yY9(Y}mEMFDlTVC-ED!nVmw%r$$ibCUZb2>(E`qD1-U==h@^WO~%piw} z-px%fOvxhi9U<)F>Igdy+8&2{C~j@my2Yas&rb(J6(QzZ8iOH9!HV5F(S}CdD3pvY2?5&|W(0}g`H-jA z@2rY^Mwf_Zk`zRNQ|3+iYC9I;SX`Br*+aLrhBrv<-))K!UnF7|iNhgrp5R_8V1>J2 z)%L9)1u_}-GCqqQrAeWrwX<3MEFraA+18`YI6`X^A%(pN-3nr;@X`CxeY zrHOBJ?vblI)vV{US`RifRYzQ`>-ntJ-F~1gW7a`tUD=et>6%R3w)fxau2pKbj9p1Jg5x@_Vf`EJ+4lR9z?(LO$ z-3lm7v-Res)c*Q1Vupao-cWsnbhN=YS4CPWel<2S5el@LnSDzm;iDs}}$OFrSONp^3m6OLrNShzMLq&=g?FBsc zv-XK#$&ns9Kq-9S;GFTJRBl<85O$$xi5sAxJ@_K6xeu_S0_UAYEFP?v;EWu#^mPJ%^R88C|_nuYe6ANK+UVQr1r zv#PIrPA_GL-wTs>CKbx^b+h-P4OY@wb*tXz3GIqwHN0Ss6%Tk)k3C(K z95fTGV>k(Y+v&g3_F^2NA!5}l6rYKqsT#j?acg5B{q5Tkdi4PgOV6zz%&w$Zm{SeM zi#YgYuaze<3-Zt`jk(du6K3O`zw$;#H8-IRHf%H)T4Ob8FFYR;XXy{rOR1GN$m+| zzJm*ylDt-P_;^sbz+UnP3&MBG`&YeQHfC$rSxefo7JAVKEx#P#745QTCj{}aT*lzJl~q#BaVK`gCvB?O6i3KLW>(SGe?BTZoyNq6*`HKNHe5jn&|i*VIRLBIc&2G2yJQ zvpE;WpX00>nV+MCi@slk+^iGs8H$W-@`&-8=f)8cMgq<|7zOEjl%<>qDnM?Tm5q-! z($9TKz1{eEmjIvNy+}c|KF2Ri!@}6VmA7G^K7OH%`Zl=A=0&ICwp|KZ(6?mPd6$!C z>%Y6aa9BgqTb36%d5Ck4@_g&-trHaF4_h-1G%-YEE)+@DkFYYN+0+Q@M|T<_nLO(D zLe=wjY>pW^%@Xk3n)mY?R}REK%NM0eXmU&XOo4zZsy`K!^HRlv9f55ZyX)^Ka)>=bkq)m_ZURigoDw|#tm ze@+>JE>}tcp9b?AN3daL7A)Dy!sFam-h9nhKy;cru6IA=cQ8pW+pQO_90IIud&5re zr}J$YcjNeEE(JB!X_~tbM5om}>~cdq%0*`0q@0mi&u#WY?|sL+-6MvzOVS5?T+eGZ zyW6%#@Af_O>V6;N^gaU2@Ao(A)o)mMpY3`X9fadzwgk82{W& z;1nqwd-ItzkQsu~k$|=FVDoybw&dU*t>JoD>F$?o|eI!Dq+Nv-6Tiab%rF-4JmriQqF zmvz~|oCJ8iP#lNgY>#Pq1O6phQ^oUl*ELf7UUY&0^h&EVxMmCaj(SEeU(se}pk~Wn z8I9e#``}V%X*AjefT5lXFKl7#FhITk2{T^ac)aCpa^Kq6dfhT0F4JNtVN4ElcKA*RtD4uygORO2_1j__%X4)` z@Gwoqzpq3hz8XSt7KSj7@>()?NTzhq5+B}#VB?=%d{kxEHm>Qg$IuNGi?lC=y8c2A z;+3s30CoD5<*peEbM*PnaYa@Q9Vy7SmNc)(1LJV|czu`b*4=jQOljJ<2h0|knX#Vo zKH73-4n6@&POTj`Ab#fMcnL`yh!0URCRgbFg@X=xV(%+#UWpET#N}b3efb-SSG<@n z4Jo^qBWdE6ouTuQ+TOP~_#L!qNxmK1BVLf4kP(sR?6?elKa#Fe(>xDEH9j)P4;^tt zn^-zugw9@d72>WXB5%H-F0c;cvejhMEW^;QJA8)E9PoV3P`%;DE2!;SQM<{}_E#mh z4za-H(2eq##wOuoo4`3<_#I-wriVE7$I-OvnJ!1H3uUYsF-Gp5AVWXDLxI_xSe54fsr2C@A?R-casy zI%qr;l!vsCy<;V!*)*c2$tLDfjV!En&9MaDSvRc0-oM!$r}#t%7Qr4sOzu zd4K9@vTqp~bG0Ge_bPH6r_wR~w;q4DOWwQ{4(JK}P^jdGkmiXl(|nDjHpdwwagNc9 zfNcDG6{t(tM)FIcslaDaa|3A>Gqm(8`1(qCwJFO=p4R|3MW1-rj@=glIu+7trn3eO*N&AmYwAw?Nld>(CS|v)K7d-bS?}WhCF- zyUzaie+_WL4wRUb5tK8g7FO+4T*{6!ts%QJwGn+h4 z?B7_O9|hh-(AjF}OurysHzP*OYM?OwsBsG*lnc@1U2rdnc3SsG3{y67)RgK#{f^LvU9qO*jlHu~WfPbsNI zuQ(m=Ocgnw?+X1eO7fNL)=(WCmIboQ-3w45GZ{B`F4HF$$o-O2D)&n2z2M9^`&V`k z7wnUUNK?eoqm7QMh+!?Xu#l;lY4>Q`q9VLn1U%FDYN>~G9T@gQZW7GgNimxkAS?8G zf*P$GA3gIT(WSr-b*0B{pxa?f@Ms6I7P!vW89;yoGon;fnfNr(zBKDlpU8BM0#f?J z8WWofSe(GWZ4J;2t$fx!;t&~K@kK#9F^ui{iD&~nQVGO1 zPk}3@C9@P%dTX58N~po^yn79M zPWErxlAN4r0OhKyd{5P5J{NW~WLBkhrdy^b)(>-dfZtG@`S)oMvYYZAVwkaXd-8y= zq4oL%HOqe8lixFCjSvTu8FCBB=R57+tFIb9NvllZGo zvPzr#F$b8Xn7F@96#^8%&dU4idjPyYEH7(oQzv z2%Q^xskC#2bh6X^r&YNn1Np0xXz~9u$s(^Rq0Ys+&-qOg0eNJ0@>WRAEd}S~2U*R| z&wvqAsw&6$Y>3|)T@ZK+SRiKIr;kCviy0h1p#R+{iOd*xwgL_R{Cm{NL+bSZ&eFlQ zowWOzX8ZZz>pD;U<23gSP%@-@b%)y^rzwH^_lK{`1uY--&=x1Y(Jo z%Q|b)3QjRC8#@X^8NR_zPMLyy0~mXr2YiH#o3HVhMTclk6@h*WOOSgtir?48UoL;&>N~Xg{Dv2w{NPu77A_6sxb%nSL($O~H?g!h-OCF23gJ zQJ|LXF<=Bw`(yG(G$I7H2c>}79~jcVR0seQe7A4c!Hj&fz|i;#E%77893$Q&C2Xt| zSgFw23Tfb&z%Z8%5Ybmw+OYt@X8J|1DkoF2L_?Hvs2%uOqw8(32azCby_ez$bfk9a z2}i{b`Rmrg$nbt$1^ciak9i%$ulfFK!8oR|w6B(-tkRC;N*A617tWi0Ufk9jL@>W0 z*yMC29IRY{U`^Ec%_mPD&%X!@D7bPC9)e&jufLE~y(|dDOUon`ERsFG)|c**+#&FOCaC}@$E_C4VcyBZqpMaKm#PIwARw3 zyeXfhs1E%JW~KT-vlZAH=?xn_ezDHbM6MEeTR8STT05zO@QHeX@F z2o|I2j6)5gx9JPuT5VS&pAfGy(y!AeA(uiL$-Ou&1^+kkC9F2OhFrb?)<4x>_Bp0h zmn;?{pa=j;q3@l^Cyca3CbNN4{)B&PNIO^2OoiF$w@Nvt`Ii6@^B}j-hR!TyLQJ#Z zE`Y2JW{fEY@UoJ@gJgCDe@o_vBZVCKIW5>KH_VZrl%vjfHnGFJuj&*izVZiG!+~Ya zhqWRNEYafyOc=W1(WE+w`xVs0nUW!(hDWzX)`0mz!t&r$B2PT+I35LNq0+)liglKa z%+5~gl-|Casvemu*w4X~!omE@{(IJld;)_dSG_DZ9ag!GIO54ie?HBhD*E>~UHk(( z^5W|1grT9Ku)ulfyiJD4q|eI>q!=<%%Ty!PKNQq1CU!5k(ZtB@Fp5Ium4;Fd0v^Bp zEm8sf>1g3-=E)^r7UWe<0md~#)+l23I8uFzwGsI3`PD@2JW(D6`#<<thE3ga2z7UY>Qv)Y7uE zRKFYcrO^Mky}oH~}sOoBUCSH3{?eYk!|74~MU$a6lo7XL!LSSf2m1I}v@Ezdbu=KY@spdo1?? zaUy_hG?6U-rv>NM?;`QUGJr66VBsI@K6oxoO-&OM6Ih;rI`^PJ@QWC4U)@W&X-!{g z#Uwj4DW~2*3rVIS!E~ejZ^Xk{+E4Rj-n*}}F>~fLpFGz={)!WP=ckk(&};CmwMA7| zI2s^;#|`sVX*TYhy#i4%aR80w@O$?CuP^!n31kE(aSm@)#sTJ}*&6xX{@=Mz;CYZ` z=87n+1+$u05@WMZm61}Zoi-qFVs9SQva8jAaCc8gDbiTH1Nf1*#B zJo8Kc_Mgn-!60KD;=Fr54Z`JQ#rJdP&GlTK8P^~z54KhKEIzW)0{3jap@GACY7TvIvohK-K znO`I-h;oh!x&S9Qqgsy^Ko_*U>-1r3OhNyYPTQ+~CWMI5=+s-f82d8@+o zeNDsvt@>Z>q37V&agxI7SLExsRc{h+xTV_16Dk)#sVHb?C({xn0xXz0UTzj@cC1@1*zUoZ!lmw&;lzV@Cyx% zgCa@rRIxhVE|HR7%otZXMlq|b>M24Lc2;enCP#LVXrZXeceo5+G+1Yn-kz7tbV87% zE~lWi)>s&EdZ{$Q+_J%|c>&Bc8wc{a3P54*E;tvSDWu$(x0hGai~_7{4w#N8ogZ5* zhKN=|!Nr0@wph$u63;gu+RS}MA4$ZZsN|NaeZqA$tnG;kRKc>{qz>8?a3WF9h1piP zGKfQ7j!q~dfH@XGCeVqq|4^0^w*GWoCC_u&sD}C!oMgCQV50!EC~yuluK>LD6w?2# zSPfS@y3NUrIT3-lM@Gp-Ft-<{w*ul+5XEdb3a+=nnr=ukAb0o@eXY0%c!s znb|t7;+rD_{79U*UWl547Tt8cMz$YuSg+5q4%LJiOPT!IaXW)n#8pnIYrDf%>obj$ zh+D?QC22pcpzRWVSI$q&82v`s6vk>qfJ4CiQy_2MeBW&j$b7o@g$43TZi1!O3r#C1<8kwaDH>co2sLCvc2CWuQqZCG|F(1KH=ZCVH8Y->n(QAWQ+r}Ab*F59`w%4&qbR6k|(=5W&c6MSyyA!Urh z1a&+fDftpswXwGrrk0B}lh?^e*ww5b!*VVgUn?+K8jRQ*+4L@g#iV6YNI&S8UBe($S!ZTCA?&_h|UgyCr48oR3rSKeLP01#`UW3*JtN}KrPGU?>1sD zSR_9wh$_# zr6;nS3p;Qm?Y}gS&|Fg=6@8E*Dp0Fn1xRW3#uo;hOsKJ9xl}J*dmIL06`9ut|Jsvx zCPt_}mDrFZj_7l)7X-WV8X2Bz&0^uw59o|2)2fyOqJsVyhTSPs2|I4t_uZq*b z)l)}p#iU*@E~My=T_~-W$w<1#`qUb#Hpjns@`(xFX@!P2))&IP{7ZJhe<8iX;b z6)l>tFAG&T_>jLa>W-LE9x8u9xoF;54MOC{mmmrM z`TIFe&QiC7OWrfbXT(YaLm({TAp|V6RO)9X*;W()Ou$lkb=)Z)^rm=j--ucG@L}T` zh2MtwbOLal$Aa*q>x$7jg3@S9I4Sz$Lg|y-Dx&AxLNhqOz}rH}t(%QDn}_YjyNg-q z`rXe5jT#-!!v?$DWPS3@UqIDbNs)0{$=KriPB$uyXiHfr>en;!q~U%f3X@=4R!Qv0 z<;%NNS93U$^Tn;Rw>6|m-!Ap5sbLD# z_?-KQH6Q8A;$EM0gouL48(3mSI!-ZbuOO5z4C-3vu_O7mJ21+GMB3b)%88_>`+)C` ziF5c$GwgUVK8aq-$GzDOB$v3!TV7Yo+-VW_F(@g?U_rZ0%X=X24~NH8P21g)xfe*7 zYj@@56QlG8Z{YuP{~V7 zp_m}U!c79IeqWVUc5s!U9;K9l$(zg#i^N&rgh5Yo_WZb7ixbsYa|nzne=8{uxutj}iMih){Cw?O?Jq=BFwR-&Crb3q z_>QTVrZS6k6_JQ9R3|I1@wMiHzH+Ikyx*0b&_C|Le?zZfkK-2{J;H06zMNKF$k1& zk8J|8a7wsS&`nRFSXS)gcBKj<8x~Hq9H4t4;|Gn4cf5t=qItQM#6ig7H>i&OH9Z&_ zo~`?5uXdAZFo}Zu$Sd6(nin(ssG!?lW;6GTha0KjurZlKT5@U{{wn8CFjrOd#yOE% zE(B#S-aU_MsDR&dCb}v3nadY3SUA(L6pGgCMcM zJn0D@36C5lX?_~-75Bs4ITyQbof0^w1Nfi0${YyD7#(>6B@|a(OD9EE`H<_ z8OW-2kwbe%?CCy}7y&=;bgjp3aj)ziyvOU&FyVghneQq}ekA;>-~lh#RjT^YH zWHqPXt0$+_hQz63vs|1Gx@l9){b6CQzChBEmdjNvD-&5*qNO;3hKi?XKi`qCP=K5f z(8KEJhq%;-Q#PobRqE z|B8q+LD0P*NzcRo2TL$7X!uD(U#p{mqn^Smf?hC%-xCi^G%$?yi20Cu<^Y~t7rtML z?a1P2064WTZ~s0R-x$4YCvVFHfnx(32YTWD;?qQUpvn2?$j(L{6kgQZJnH9{ZV#pD ztq?=Ksx&VO5sv-%xFd|0X5aexK2XGjytDE{v=_#>I7Lv0FR;Q~={)7wNgI608J%E% zUcI%{ed+hYtC+$u--<{P~&Eg*Opr=DJnXXn0I8%i?CrQ&h2nn4?faNO{vDk@} zvuHnsm5WKvO6WteRZBaASFq;d9?*Rrp?3q_yXR|vk~t7Zy$<->oh%BxkU^MYWcRQ+ zmh74>Y0B5(*5b%lzpDlc0X8~$+1N`OukZgFlzD_t<8|hHxXv$Lsdv_jtX2Ke_s0#(-d?GA!PoD) z5wg%&uFDomZ59JW&H?C$_lJ;Ybf4bxTM6wT4Od+z4%&w%&{Y^H5qLe`a~=0KKdSR8 z>D}HAKfo!|r}<6PJmmWV&Q`(c_1aqBbV;ETs+uMpa2ZmjvG+3_|8HsDN#Jud-}bSA3f$3N`&m5r+Em?gjS!y+ZJ_9^b0&Nw=4dn~ZW zI&@0xQ-e7^!YR5VhR6ZJ-KZfZW;(qgPiuCs5VV0Htk>O?MI{F?;^tJ7<)uh-O_{`88kDrLN6qQLP_|cpn@J6IMSyOwrZo&C&v&N^iHg1N|1IPf%ii% z?#aacQrp0}|L~!_zA`Zy+d=E)oMYJwb%41R@D^I6Po?ylBII%3sxt-iQ}x9lMo)!e z2)+F~+o3c0eB-CSMAn#_w^cfn=F$1z3_a%8DRVwLN9N}!w?x-!n%Ret9YGHl@d{4q z@?pV9d(6zikg~zDP-{lOQApuEXitl&4v3lrx%k~{fXdQX?yT7NAu~Wj-#)2fz$RG9 zLhsE(J)v_}P61U1p3JIK*fl?)6Y-4&mSrc%KBoUJQ>s&iLjMZo{?!)6*yug0~716{q3DL4Y*Vb6UZ6C4Vw zJ>BPF_(V1s0gB^gE&6ysBQ^djvNQu+)ibniP9yHkKIoP=hT-2}9ArBMan8NDO6uCqi-7?Ke$8_!K0S6q1Y3iI{W zuW(o<-S3&T3Nt*?N>jnhqA==PLcEVrO?PMz=-K2UXFx7p9cN7P&}l z^}P6I^PfU5NP!ktw5Wx65tpdfDkJD8p+JrBzIh4ngIPS}o~#5=VfsobJ!hFrscrVgH#&1cklTE3?9IZDLO%#Xz!<@hm39aI}zykk^uifvqC$>tvc)`t90`;dG zRQlyu!wyiB2rR^ukP3eAaWMLNGK}9JJLTS z=iJW%z;#U`cs-kNzd5RUyFmEy!ISHAzChN=`yZ7UybBBYr|HfOw<^2aPTW9@goKX- z7jyio`3=90MhIDtWL(p@2)Uw%Qg8*wo6mw#%Rb|6<1;kozmM-4QjU5km%4#V@DbO! zLmhFuTmrf2;#@8K+{MOd+lqgalB?6>M1%ZOD*22ANbc{FpeXS_0OzfqeZM`8s zaEy!h%U;l!GYry1bpAy0O)LrUK|tx(*{2u`#Rg04^;;r8C5ybyB+OdIUH2mt2a}zSF_ZF^mw(QDlpgVJkXVtOvt$P3sFSL zBADitZVWC_2n5=3(Cn3CneEIU5=y`nm7hJ3QcToj>+J`FkpQ%(#vfCi*FIuQ3BC=czyL~*yZN3{2KT5kL>alse zex81PL{`N0gV609;te&)MIWSLSOf##eB;?9?x|B#Swm!dr;Ptn?8~ZAhIQU$rhEXO zp`gm%hsKt|ett;#6b&gO|L6P4kxn3ZwGJ99bGPxz+~CA)5&5T`iZu2uEiGRwcBZTv z1bcgT`nYpN8>fnVg41^v%PN}X8mCWW=xoxhBYfeY0B89^MT^8Qt+`UOJN9bEV7obK zhLnA@z%tWkCA;``-Ed|`HfdF|(-YO<^f~Hs0^V1bX%|9vzZqa z|KKnfyZVb!UtBy7vUYT1vje#wLE$hH#S~;H=N3LrWU zLr3+H!QY4d4vA54*A>a36lvog-miy?1xhQYxs)hedfvREuHF|l)-8y@S+~qvH(4_I zrK)VqHxqgt3j;oQO$A34l9IgBMCeQ~GK#`QASU5@XN^tE%Zvu|nS7IzV1pn1q#TpH zOOdfq6g;0l0Rv@)vk9w1rCFL>1zod;7mN^yPG=U(lGk29Q153+g?lJanLoYql{*F6J7D}U+C zU;rT>tWs@X6v8GKrMakP#|V4rt)I}(9%UmV5eFR zX;)@@7dx?5WUk+H%zhb`9T3Fj;f^O<3B(>mRtw|;0YfJ?Y||{~ZyFq?Fv`W)KwCpP z$E}~6f`e0N^vfb|Ul;r-0{yvs6wuA%Y{(+Y5gq1YLP}JF-OV-1*OvCff-`Z;@IIv{exkd&FkvQ_-vOD1Js$rh7;rvMS0|`$3)^ zeHBXo4dPwH>hHbo5Hbd(#0v8sUG*w}(aqJbcX&ExSqXZ|y#*-_W}XzAy8(EH(MCf; z`ClI&OM-;6$NCCL?A-rs1LOwg-|>IP$qLhv%lGazq71k>N7M zyJj8iVl{sXb&#Fw#d>FPaS~EU71gKiDkq2qgoa!n$S{34;I-doH9j7}r?U(&2 z=N=A6xhkUKB)h0UMt#TXqj^!|5(fS2pOJvNZwv6V(8C^Rms|U?&ieb+_p9&aRNSZO zDAWNzI_zBhhlBfhwk(K~Xfs;Adek3cr++-fxN|@Mc(}t`mpf`69gjb^Rfi#>JPIlh zaRfIerQPhL;i7Z zDE`zF4&B}RQCxaqB5&eN&16;;u3L(pI)vm%Hw2b$fO=XJ+_-mh)+R9L^;!ice!-Guy!?D`rzx`T(1I$aIYK?>_(AQ z@S2CMdPzkm8-x#GFE)+O-8~^rbgp7+X)ZkL1h>w!mK+Ok-}s%07?c<$QaVh;#0<{d z%(|TW9S=Qkh6xh;S|tX_xHT;_;hpzJdQct@&F~q&?7 zwGRfsGa*wMg!~1V^4^d!kT)G_*xU6uQM%>ob&7Btk4rj zb;txQu9cX?$>qfs-k=PDJGzJSHoR&3JH?lA@mXJI{SDn4${egHpIzK%)0+(Tzl)xr%JQ59T&LvVIR?#uUr((w5RwU9_nZb> z_8TOz5fNF9LeKtmq&+od&Sok%z)3-gojigb&`*u2+i`ji-m~mXBr7 zY#aAM3e+r4Nt+9)BzKDIyQ%eWL0)8Fy$l;%^-ZXd&<&eY1EfpOHXXmsnYQKc@abB< z0denN2*1C!W2NEy8XYAY@>1mHe zeKKp%<)A6#TN#D~NI++6@V53+*NmK|pwn9)qfKy_UCg|z8jg+^>L4-QD>L|m2!woJ3LM~IrIfsaX%a@Xn-FVs%V7Xw zuAi%e3<^--UC}@bra(}$J|ZFM5)ury+)NHtL?qhwJWQz75gby>i%do+Vg=uUgvB$o#BB4`)%+k4)QLOTCeSRz-0-P z_(n%YY^`W-iJg_bcVAtE7y6;0`icgcp?N6a=La^{`>0G(B4UjU>B>KWe;|*6&d@uJ zuMy;E6zZ%h5WIz1D3QH$B$!$;>W))s;`n&~NB5aJ*>Zg z!<>0~5al{AqDCqd;v3v5IFDktghIq92pTf>GB)SZ#T8=|ti8wT?mrZlXb(Pr$flg^ zI;Av02ss6n+Kr9j*R#!{?u>^&G-HzSBjy-&Ni!N=IWWg0V8d3H1LSQ%S*#&hXT|0(A5l!9A?-+&kNbb9QkxbfHty`Jj+vG_NazN z=)G?U8p>3K(Y|B-VETMlg*sV5)AsKJk;*(w=^F#$Z!hJ?pA^fGo4(zNSS`F{_VhMJ z#s4Zc$Vv|gCA}J%;`WiqB=Wc)eqiWc3koY1{z708g8)HZ?>f3^1iaqLR@efvco5zm zRB_h=q!q-w=zY)hA)NV%M1@KC?Y{Ng89Fch{bK~lQqF&PK;3P^LZDE0?EI}3^YC!b zJCgdkdM^M~D;lE^2<&V-Wm|QCK?)WM7kx`fm^vs`hDJmtlvxhhzYlB_1S0U{$t&DO zQZ%IG^ z$<{q8wRIH3_=J5jU3r7#wO9}dIfE7);i&!^3$01n?b?rIUxj zx;NCJ+w<8ZM1V!5fC3$Rfj6^yTSjhd2lwL+YBg)b+znd&G-c*(j(2;z9HjL83hFpo zVTEX2Rz335_Pw`NPa?GQw$Mk{iQDl~hJ9@f@fC`>g3DhIRGn()u%M#R5rTPB9cd~q zOHa7fh$1bwgl5F78-cc}EX4Zvssdal01yOsBm51`ILeFJinE3t163Iss$bx5)Mx$F+ErZPlf2G(Ub{^$r4?2yJ2B zjxyXopx!@T?F0x-vA$?~&2?B>)BLWiO0MFiQM;CXwA@#8ejp8pv7{pGFz6JL`D=8K=A7#I0;?*@{ghUvq@NU9r`6LhLfR9Ff>Ug*7{VYp4HNQWEmAp5Utm)<~ zoO)D821elT2vFoR5&%KJ&j+RRpvc6`+-8>?X%%}!=GZKKrZg{T4`PifmFf+NYMCHy zEQAONfFykw%x5ASPW1z$bRg)`UQ`lcRPP*x@A{*&0ILwB=1_p>AYx0txr;qvvZ3(r z`@loq=Uv`}!AA)?ud5u3k5)_dKq#mGoe~2Zr+gF8B_UTE0|@3%Gb6K(@-|uZT>z5- zxCA&)C8Iy|fp7}3SfWs?-8+PsavIeapxM;*O@1XcwltV*59crcHRuRDi)qO#Q%z7sdOZSvK-73(`1Jn%-SSACtGhdi*WKX|ked6fnUJ%bS2L&#FrmNd`99pYXpU;@ zmVk*)OrHBha^lj87r`DGV%febh4CpZ5O{UR*v3w0b71kbdJQ3ciI}7NW6U9X@tLDJ z0fbrcstg}3xAy(0vlSt!4Zna71iwn!J4ZlFFSGcNax@E$Z;3 zDnOu~juKI~>5UGM+K}Hp5bLEC2*|C|P$%|Qbl710`D(BwXkCB~^M#+E-?Afj{HsR@ z=$e(zptTuWMqr2?8V)IF?3TGo0^A7((;YgbI;a>$s$Xc~i%TOR&z&IG>(8RV1wr+P zX~2|#j8U`cpuo4DFD{B@MDrC(V#4j4q~nqWaPYU;V^mvXu6kVnOx^`1Ku#aI@^0_# zDP4SEUA#Ht^tvjIfNa}bf^5f#bxIy^Vfjuj+1Dr>ZHlV7VB5;QPBY}hbEw@#nq`Ys z2AnMoPE~=``XIv1Yei*7-#J|GL}kVkX>Ge0Sz5LWg=#1xm_<$i+5<^d?D>t4kZGNN z8@&G`aDo)K>TmSk|In(VWhcc&V+z+*3V?|4W}*a|Jg>SsQC{T&H=AUL{YCOcoqb89 zsD&v6x%$^)eFnjqqM_;gCh4P_z8Cc)Cjc|Sh8@42cC35eWlKHlwH*XwG%(vEm`<+$ zxiMLNA&1@QqroA`^v+gK6fC(AIJ)|22*0S2N>!Srff^w&ef&V20@xWt*jEs11X-i6 z0V^q33_{{;#||^APDXr55huYv>1>sGS)SpG^E50*ug^!+P;r>@U*j7NAPH%oVH|LC z)kUii;G<<>iY3DIgkb=EqARsE9)Y7>!0v&)Clq>!yzigqN(Jt#|f;Qj=udg z5%7hRkwzAnoA5fpQn&cnY!qClf1eA0%18c#4gD~^IGYfaH$|v)(Yr*KLFiwq%OyO+ z<8ok))jul(QFdGBI1~@c(+1>niC_{aAo+lk2`c6OvtZ=YmgYV*M~8Au>0V5WoY^@- zy$jnfWEDJgFhjKJ;{enAX>SkY-0-NHEUUWEtNeA{^6L(9Yb_+>VNW=GFL`0DqyoWI zQVd_v*y$Ho9O}Afv{VEiWxnk7tCq*84MUAxFvP0^OelkZb}{J1LGfEvqY4=u9tfD< zSRz_k9?V7LO7wX8H^0RL0@3b*%rO0LR>No#>Z{kJKJTif#H0D8)N@A1@>$C)ppD{X zA`Q`Ei;@w}_)8G~y*Rn-0Lsg1LWj#^FhmXNTtVs<{-NFmVxoDI1E`ch3Ib=i3$SY2 z3Y^{+udYF+Nd8%@6UW*@q_`|J@e**Tz}EsH-mj{7ob%o=orcD85W0a7(7~%ndjSDy ztK_IMfI#FwOA(BpBE?lSppXJibLS!%;b#!Imo!^H^^|DXfX_1+>`THad#Mh}7Qcd+ zu>r2_ua%_(jLc{fMserr;CQrk3R2i)Nzs@dXb^+m8AviEUaK=3Zvf!Mf7a*>2#_uYO+P=RCRLN2NuW$4RCyvWO+K9j&K(?3~C@1{Ga#W13 z__tqU)Bs>zfVnAxKe(Q;r+xmfbNS&N#ByluOo)jjt)iMH2LkHNc4q z0WMAC%c{OyiV*??Gyd0NeF=G72bLO?Kqx%8=384)IZdHr7=pD3Pd&B+Ghjiz0pb&& zNwEoEA{6M!PKVn7VKV(^jc!9|D9^|WoWspoR_V;%RhhWU z0zoYbtoTdW!u&C`N=Jw=&SWao=kDiKq2BiY)L6IYkF>=|zsQl!mznP+nWTK;JY~gk zm7*Z64P_=L9$=ES)(61&Ix->pe|N$GT$f;qREZq1Xc9zf5U{a`b!4*BY+V85Mu?gu z4Kuz&wE2)AIG7q$I$VH?@Sn9k53!`DLP}-eVa4dZ7xv+Qi7Sv^lYQ~v@Y^rOa!4rG zkO9DdfYFb4_~#P_cSDHg+%|)_v<5@pgPABhvOxa!ARM3w%0e=p!Ill|3+1Szxgtrd zm>cN-Y`%yg=>DRmsro|L2`n30xK%{T`@wSJgTo=LU4~#pBXGo0)rsQ;j{W3+e?Wk$ z0qiwW?n5D@(cZEVG0>A$KlaR;&?vBi+k`6-31!2>jaT9Nr}7Z|;SRFzr!s?6&&Gm4 z8W8Ifv-r|pPL|79sYw1d9@6XrI@L~t=c2o@k};olGc$QzRU z_Wj}<2bo?-(d4tApWgm&78@->036tLj7|@Nl;IE&Np7$aKfimLszl^dp;!6<;y|xG z-P_d53&4F-`TfsD91Va_8Tf?@$4EeX96%7bTzaqy^nuV$BmK_BOz)2o(I3?EDp-G? zA^@330VlRMFW8o{J2l+wk5x&pDZrK&G@8^v(+?G{HvEd{)Okl#t@E!zIfX#tuJPk9 zb^XYZ0G10pz4ZqX^x~;KjZsVn3+d6tee!O_zlnJnlzehGIsZGl`JSaBC0OCR2qZAPypH0igp+ z&U2D}nt#4Sj15GPmS1vkDukI_?~4^2@XP(br!h_)_kfM#+BB3|eh!c@ZGc4gmvq%% zC#E?7@-0}w9iwu90?lRHZNIpQazkK*!Eh2Vpx7S&1)52??|Z`-eUyLSs}yok`9}(z z`qeoMSn{Ri+#VxfE4u?f{A)zIsXi!k|JUUvU4rPAdZZ(00!@*36(*#E*hp2s=N}x+ zlr?5P(vNCz*cRiswy2RAU|I z=mu>W1RKev|Le(+p1?z%rcJP|Dae2=(@aLEn`p@j%SQ8mFyr^zF^E+=kO(2h6pj7! zR7hJOyvV!I{9o^@Hi#!vpF?JoAxj4cfzy(N<@(jK$z(KC@88?f+X6YvEu=J-`1Qm% z!@+JmAFJfS_zfU+l+X0M5-P#Kx>W)|>4EwmK&k$7HNF0v7=lp+`u3MWB9d^RHw3Vv zn8ng67l2>cZ%zP|4&eCy>(cbQKr6c>2bz_3z5K0eEX~X!W>&%U65li~-;U zdXuwplK)&kG-*P}Zytseq`FGtGKAGa1A(ItN*WvZ|C-W1h=!N+bAxJ7$cxea*XB{s z{>2UYd(XW5!0-LLVlp6y{NJ@s2}vFPUDwb4-=F`zJO5X2ozTR7PAUa%wRdAm*sbn# zE(pEr^xxBXJrWIkRk#TGaAr;1)VpxqFAut8y-qNW z%Iwc@BbcZhKnC2K!8!LFb+T3*FJRc zC3SEEiHxc$cPRb@xUT?Q8pMR7)B67;xm6ZZ1``EpX9-dE`j48 zLGT}V_UJUH^By*#*OmU?l<=;e9>cy^@+2n1z=~CG?7g4)AlA15mlG0N{Dex4yXCzrJsv-ERY&h7qi znH`?$g~&sjB|8}>C-&2)cMJcSie8XwSypL0fH@->CDxPfMFM4_T?ih48TEAFZ+cl; zcM(U1#TGs!YF{z1{i1-_|4Non$W1b_LTH_6b2u%`#&cJ*cDJMfuv`DEFXoVs9*oIu zUMvVc)cF2^UtXa$coYsRtmc`@mE~Vs&`Q02&uSl)kdFh z=8x4o@l+{%wX(8;@a&nq%3s#}LqT%I|88MgyP{pSG%b6}Ll-*+Ai$f2(pZM6EPa_JGNmZQ)&o_DcxhcWWK zXDueL%VBQmv~cvpx&uYJKshMQjAXd-YcnKP*ne>)tkNj1;Bf97;XS|9FWPqGen`Xq zsyK2l_sExVL%?~y#PXPu&ik*i9X0LL6dmC`MT8u;mzbC|g&iC=i@Zrv6tBt2SH>h6 zyRt;X;vBX;%+1YXM|L9#cz#Y!1~xYG#w8@sY^7gZk9ZSGCNgxL?Tn+ndX)oeEd?bd z-I_K{pE4fK^7{MDT$VgJ9&Q05cm|KB0U+1bq68wRbljx@@fU^vinyfVfZI4KaBaQs zv+MOnV!6YXOkGAwN?3h8x7*Fj?9pHwk2SoL!YlXGv^1buim;*e$J_n&TylMg?zhFK z>zxWKri48tZmQ?3JJ_bQ{Fj&YjGfj&JDnq`+GeSwDD@&`=}xTqy$1vM{DD)t-UK$y z*STOYf@|e(o~3JK1qFq;_?Ykc`K^cEkM4U54OoCh-EWQ(*5!A7y0z9L@W6x!=U>5H z7g7^S=y@i8w}Rpw9d-YX3m^G<#~tk;c&RWT0ESMp8gc3I;WA$Gr^sxz72d;nski>) z!#x!vqqwlJ|6QE7g#|4)H;}SP!^FXP1}N!;6QQZ8`12m8Luys#@%9YY=*KmZcl-B~ zHO&B(cd>Bjh2?!So)OnnPWG_)2wpCpWJS&GW+%_Chrk8Y*OC?*T-M@Mb=n?H4|!X^ zQB|{RnJ}4(h`#a34<^loi6_PpL4zhWkeRX=)){BqgxO4ef`k^<1oP?E2&1L0 z_kGoO^~&7u+6~n+;QRWkZXzeQL&GDpW{XEL*n7^RiR0^EEV>^q$CDej`N5h@m+9x~ zws>`w$dXCoh&$fjocp^Njg$g4Zs#}i_5<|8$3w{RM^$~nlyakAh2gOfG&PA|c-&nb zo9v9`wAoYy>9rAg->u+Ba1!qHa5-!y$qL;6;=*xLq0JVFqZ;NpY+Unm92^-Dc$^=4 z<-B|35i4PKJm38I;7&>8>day?Xgj=Kpj=pA_cnyF+{EYZ3%_T*+vyO8|5DKtx8BG) zkJAZN-Q>$f_mQD354*Eb@rEa!ZvM>%$Q3*c^Js*%y3v=XLNAu6dm=n#K zUu$Cfoi&=wR2W;!oO-%i_d-F{44TE8&^CPNiA2qsXL36+V$pAhG4DiZATmb(+lqbVpk5$6Rb)ye^eh9u9+g(;;XHKwaDHAVe7 z&Peifvn0OD`SJJzy%g8xwyRc>I7VdNqqmAR*4hy@_$1lkmLJmj%&8pk{8dx zXmQ=7>Yk(`fF~y*K38ou6ip)Rm+}1Vr7`6Q*H-FU#hK*u;oRGUh7m*}e%`C0=Lp?# z)t$%|Jsl_ld-%WXuC&=;On?9S{#o%!V5SoI+S-%u_?$}`RpuMcyp|Px6m2#G961Z+ zpJ7HmEf59a+Xp*vt|`5wMZxb6pKJ4nPdRM9Ww^5H(~CUbOQNM2sou;a;IUwUQBrfD z*2Dr7>=8C`0w<*-#g7QB3|h568BZgb*cl;{kR@~c zT(Jn;gKM@AKU=};@-w^>?+YrfS>$&8r1lw`!#b7LR0x$F5$-`0OM5+)@*2(*V`>=7 zG~F$i>#*@u%k6m#4s&Nxi;?@uUGhl&*TMbmREifrJZ~>#2oG!5B4yw(L^_^~reX>O z$10{n^SpfmCyujZ7)D=>;FUt?oZWbLGH`(C_xj~DHSRRYQ^Ozv=is4}?x)GpPkdh! zb`y+P-0s()c+s9bWB*0qGQT2$-OYz_Mst(S6oJ3oG8kkwp4&y(dFXb3J|$ggGNe`A zqIO?o=1TwtLym)|6-6xU=10KZI|}SF8+drG47?dqa=39Ks?Iw<#d!~3S1&kgawPIK za%i`+lrC75Hc6H0G^L`D-|5J+%TcqJcl(olxY(O9Ihe0wXyBNa6Ow4~OhMpo$24a9 z=!uL)Tf7>HN)Ao;(F*r6uw7}4&M3|#>+Jp1R;9XJU#Lb(waO|h-A*LIOLa_%IK7IL zH&Qr6FbKVJS&U3Zo&BGzI`qh}jc=a4ZxDbXaG_LC+i|Mc<4jpF%HC}6_ z0ij{|Ze6kZn^Lhz{2e1@Z0Eh@&aIZ~z}Z#dAY>#j8a^1W8Eui}ue@yxTu}6JJDG?< zHbts1TK4o!g3$?el1SUPhb>E78&r5cMLAtVW~uLCe5r!`SKb{39kvb=B4VzWOTZzJ zh#V{jgmLf0?Czx#`{;ZL*M_<%s+zWzj}&;gQb;r~93o!nAwnmB=YD-dEQoE{CDq{U zdf#x^ynFNMFazf1b^Ws~)Qt*^{G!iFp%DaCX*@2@ee`#ZZdfA_Y+cY8!nM&rd zKEmBc5q6aj#2XI?Fg9OhIb3DCh?4|M+6GE_4$o&YH9xCeF?@@enccNx#;e-TVLo;9 zrR|GKXt7Sy$6OFYb{PrhcH?N}yk}6uXj=-#@5(^?v=u82#dLZc!Lj&eO3NXlLFc6r zWnXWv&TvI`LBOHnCW{5;3HB`mZrQo7tMhWq>%!;Hj8UEU5Xvb|2U$kG6}>euRgbAY zsNc;6!5dynW?zuGBoT}{L1itz-i>>NT5@X z{diEfrKQ`cy@X3dZIK^$bz>Y)nc=}~jMOvDPitw$$DSg!~lxJ=AfO$l+%>9>bIt|G0E?4r<<^){m7FUc5B@^!@vkCfgs)R zmMPIBBH@=+%|b#2+`E2Uincu9r+O2B6MOl&7|*))>*rJ8qMVKjVwqfzmX!t)=)<%# zCQEf&mX*na*e;7IheMmZYJolY_#K#ux+&AKN0Y&%mC5|t-GDh2O{;kBhpRQkmWLaQ zqv4YgFx@BRBk&1n(m4B3;E!f!vUV~bOB+x6m5Md06kgFNQ8^zVcfFS4r$oFcg*_j0V3}AlbSY2I^<<(C3Tx$roGzPhhUmxqWEv9oO(%uI)}TE*r!|jKBPYa zFZi;tkVe{5!EC zx^2U1_6(1cK9jxa@-~~Uejhr|-|UzBa~xOFeJgb7U{?m>w+w?)QAt^h;>#;HfqSJJ zqgwcEz3cswxs@u#p^rSVmaDf;BIlFSX9c*t46lzmC6ocES$AzW9oDiHeJlhwvY1X< z6jfAwd)s54i9Gal+&1EWkQ0mS zEVpWg^YpWK+I6Cd*E%RzH(#47k{^&B3z#B1WS+zPSAOp~9*6Z__XEoaFNXUj<8nu< z>{PBa-X-C~!1*d9>kQQ*(tTyTQSQVWH%-P#i1kY!c@<;DGd1FKi4@yOlEP$aWsg>v zf2K?|KZaj@DuKueOLNkZC#%)$FY*guIUImfSd!)D+-2|}SmF!|Ugyv3To$$x!)zG_in7^>>_5R^KP3GX+ikuv|LEq}H5FseFDwu>J%R5SaIP=SIGn^LDbkf$3S6*$GQzXSHs;Jyn zFX0LL3AS4+m*)Fj{PS%2v+v~e0jS0}lOpTtFNf_drb^NrG`{&)O4vR@?s(j7|A2hMrT=j?@sVLO3Z+ zcbh_gi>Cg?*2A$|ILffLNU8o}AT*(T?3q0NIS4-EZh3D)pTJgAXY7u2<;#8byOr8g zzazea=G;OE0wcNexC|+W)!o}*v1+?<@ZZ0IFRNO0FaoBXKs0qRRe1n^ zea1l-8V=F+Mxb=7KK1K12}19E`3qm-!IdB)!aH)7D<_0$TN&E6mnB4eQ80M1yDVO# zE6;>TFpg6?$)Aa>s7EjNc`ldb+F*YUw?R3HJNBW|w=7$G2G3i8KNI^53R(+;Ye;Qt zX@M}v0mZf2YnbJ6dP3|;OlxQm+w{Ot zZDsM8Z%b$<>RS4t6!~XyB2U(f>y`0Lzo$>*v@GQ}=gy~eiQJ{Tt)d9|2C4H>(eH9O zdu=@OU@~h3ZbFW64)XFX&rBIcW$l_dQFd?Lw$iZfen6qG%;@)SUDfq)wi4|bV^qXA zv~i-K;Kii9?|t~DEw%9+ibtrZFK&y)*&}}cPktQ94PB}$T8TH^d6+w`#R&H@67BLP4Jm3z|uUl~?tdZ5$ zzNvX@lG7N+GrL@fbL)6rZ~OutS8MN{)q*N#(fjd!Io`mjk|M3#@KskF8>XM?v*EFj zl%QXux#`XnkPOo0K{K94*}#HRwLpp52~;)mx}%6PYB%pe07`9PRJSX)P#O(t)8xnjA2Xx`%l2ok23fI(c@>!aFEmFEj^Fnl=8c;uXm_6~Zijvt<;9^1|! z#~Gdu@gs1)TyVi^z%2fF{`)(~fmvLQ?Y1vNFevLyTP-*iUDu{}JK36PN6tbh)8E|# zN7MG|GY*&iH#~R{MTIr#C9jK3@OZWByQXy=c~uvt=F4mRgN+~KGLlHRti20( zrXB&0gi|05nfvvOGIcp!l%m$`*><8j>e;}p$(4H(*xDoXY4he4omI>Z&1EO{oQVO3f$%N zk9*}k^SlzPUxNrkJs)@SGScghyYUq1HGcGdJRm^<4?5xIxQEEUFj051GR+~o$&6%D zt+!SW&;HB?qw^ z8Dto;xyKX4NY{;QQHMs9Op18*vB(l(MFK^!nQO3FjDEisOn^^jG}1*qri8Wx<&`#F z)mBWoo5Z3bM@^lMX!*CfBuf$D7|yWQ-KUhhK2ogduae=&eZygyWqJ>Y2f{7k9&zG0 zmvyjD!wH(b+1LE6N(Tii$l8~ZEq(&|{Fp&H<{T=F-usxYj@{n%vfT)Gq*CeMj5-~r zG1M5vQ&Mnn4OY{eDRD7yMq9+Ry}5>1Z)=aGibrjnr}cxDCx9g1pnSzwSB8hhZL)4j z&bdfC(Ll^c^2*7Ugd!ULC`7ypSNl&}!wxGNqx`vtR}8$0R1uAambRr2x3^ouDa^*9 zyAI5*(c(rkwvTn~scSr+nKoUTN2h&7+39%b-4f1Tl~#5~4E#V8?#Aa40@F4HCqEy* z`n1Q@EvA5U9P8I;V*PohQ)|7Ea*48;NfoCu&QTBfTMC)on?qMGUe!%PoT8QTJ;{_Z z+L-1j+Pb!_imr^UZueTajq&!G1RpH}{BoTWZvM|)lHG=`d#9e$Y<)NTTg^fI6qON( z3O8xxsPLX|17E4(1I!2?d`ZRm2Y38!2Kt!c;`gbS3dw)9b&(@tV7wPKbh4Ab1>8nC{hJ=f^m`=g$=6KUvyq;-_s}eliAZ%B^D)QZ? z*t>hD(a=wJ#&(a!R~YaH zRbXJs?P_HUPy%{vxqseY5rz-dp4$Hu9Tn_o7~|c(-xeK&$E||5fyn%9le1Gh&}ulIcn=-#CNfhn7GlSpAj+Vlz&uMcp)-RXkw}jlJ}*yqK@Crq zo@zPrIv)S$@=wSFpWFf)jirp|qEYnS{rIy-sGGWz)^XgTd8S{U<`ns+IDi6wDy7ibK?{SpyWAljuE?pSI!GWRe$Z{#$ zG{Z=)ZcoBf6iwfgaXSkd@e6W4@``BbwQjny;tJw zCtqF@(K=waNa6Dm8JSjBpt(g!ZPXp%53oGn3ua5NA}L$bAJu=sAz>CP9C9eZRSwA# zucVou@2wL~AM5dE6Cde@?`^LJjJ0NoAT2mf3F8R2t@1D-yE-#A9{J4~3}kKKy?7)*Wd28er)2MrRDx zgm2MTxJ4iTXmFb)j@(=O3kuAMZa2$)&-5~XrPIUh$V%O5Q{4!tq`cxM!WF~p#$1Ey zK-X$qP*qfB|Pg|i+dmSQ%`EXAOv;DV1<@;HThl5H-Tuh3O z*1m8NA^rA2Ik~Z@gv|h=-O!BibuM0yKZQ~1GCJM}yrHAM^si)^V;W2~Ks4kP)8{3@ zLU=;n=5iv)xTHGD<#R{+4kI1-c{ZBx@v#BwtCTmwP=$w@RDy86oruF*9+)I6Sb{QDU6FTJO4U}(5q4z#{GI=Lb4 zEz22re$`^!HT)64u}j39X8-04jX4Efh~Exw+Z5lNfZNeFlJVw$WBL1&)|$2nIaTWz z9hJHX1lHmPhJcKROp?~$OraPOcr8in`3tjA;~9j}#r zzISA^iF?^oNPkN9)R2m~HKQ#vC!MCv$QmP*KvXsa}A9=Gv;J58_aJvBi_mpWaRU1qsH z2~_3n9S_imeQcL|HM(F<3okJ)-}xTmLjOxXRbn&3iW-d$Z$JxYar9>tmDJ<2p= zQ6AyVvzd;&_Nn&!@u}%byNVd54qoHp$j=es@cd-Y4O?&hlQYg-s`_Ozc1 z)+ev>FwFw{js8!iHG!FP3H{z3Q(x76^As$;jOR>mN2%(1vKdE2be`zHAHp59cIUCD)NWzv#7KT67Y& z+Xzyf)XgYS&X;S8pMN%>Y=K>8|JJTloIx|2)799T_iEYSdasks$P0b&sRe2dx8pYI zZCb3x&=vtwI&%=Z9S@i6%{JY9kWsX67S;x=*k`#@#mPaFF1}HRMe?gn#km7 zm+AS^KJ&j{f9(D^S#34<>(_V1g`0Oc-|`Qd<0Z;pnVrd2I|DF+OMj4lqG)MI8fq1q zD*4YeOhooLa{Bit7nPe zm!>)GI;7r;3*-BZ=lPE7-6rxSk!gopqgTD?x0fk4S2oUzZd;I6RM?eW(&uf-?V&T5 zGf#LzSG~P01;3Nc6dQ(%kLg@XDNLxJ%7p}&owE*wU7{-CuPo2FEzgd2-9WXf`-ve5 zqUWa!pK^?*JZNN2;ws0TF@uV0(Dic6Vd*G6w*oCe7xLvYCf+NY?HZ6(bwxTXsfbfo zYvJCH|1cZ>1^)A+&6}?^ofq5SyzIlpG^J(EEpyl|JaYdqO)*U+f2F-p?-bUQZV`^l zR*=-#L&WjnCoxBN_EN+-&#+AgXO|3e%Or=9AU72k!6pAXr7<3hpuhvjbGd5D(t@>x zyv6S=22K03ri$iP1dzc_e#9Mul%a?+nuDC6UuF`Ge((lSpj@gUWa= z%P*E&oYTjGfypCKP!%y$BhePsalOdQ8Ju^F2kJ4!sEJ>pRGRDesF>?!x(wo*mKl>t zgoDDZEo$o*a8;5KAJ9=|=#1pbVABR4we+x_!XL!6>7e)eP(r6k#Y#=XP8f3?yX~!p z2oZg5dX?C;LShlr!7=m7;1%W6Nb?3>da1xwblUS|y-T9fydUEAhr%i-JY2OXH}t9J zvCk@$rz5>KlT9fIsSa0o%7oWJPCfGbI)LOfYH$f;Y6Hegx&Cw#-8AqyVlFD(ZRb>s ztqf)H$MX_?M8b7HUb{B)60Nh%ILJ5dMSZ3=VvETUe|&w;lQD#CSj4os`mUZ|ZEfDT zu{)k^0Dk8tNITj{EglAWES1Z_#By!)$S3bc{#7F2RZQO}_FS-h5hT5b_E ze2*B*z~{Hk<=Me*3;}Al#AF%%>DlVXds#*Q@@#5Jh}^w5Dfe)|Q{@o%NrOw@??`?k;@vdu)9ow5NkwNS5f^=? zl9>>}?vGkrD#-tfy|;{tvU~VO>F(~1K^hc>ZjcZ}6hT005V;v*00)rn8X6R&8w8OM zq?;im1QiBFIt2mg_TAjif34?v&N|=Dx3k{)K+Mc_T{HXI`K$d4{hq4K+r~(mqesNL zWA=s!?MLcKlwyd(T^Z-@VO^3dWOr)0AP$M94lTKVmeI7 zZ4D+=(pFI^usps|d}N`--@*eH6?|th1;cXH#dQ-g&s&@|Kge5evCoPJM(*vD!&wx{ zd+6`f7*(LD=N3s&h&5`YnU^44ko~|CV75(0KK8abth-m&=+nq)4h2m)zjdiW)_-A` zVUiUHzQgCQvPzAQ)QnLbL^R%IBf@FiFSVbkb12-4y-OW$qmn)(&g-;}?Z)Vgi zbMRw3U`IWEXiIe#hM&_9tawamQfVVlOm(~l<+7wzS{xC*RcLAb zaCTe z)V;fKeN!-xeVioPjI@^_h|V&1v!X2bSGCb*Fw`Bs z*mn12bxFu>Hl0mrpONKeOgg^W3q6PczM5q0hIooz%ASL$X9Qk+F<-ow#QGA>JXp(8 z8uEB?t@zSSW%0GgG%yX=(`9o(!=`1L+){K3DeSUK`EKX*~;|@ z<>d@thO+72kBp-+Fs@mLRJ_8w3YVs9CvtSo&ekQ|UvJr}mIfWNC$V}npD0cXgBrVD z`h=sgKNp`^H)T5uGh@S@yWuehLVwtW)H3l(<{?IrZSg%mq=bW?)0AU-U-!5lUEDg%T?w$M~M4|XF-22thcYy9GGZFgxwgwOpcy-*<|(2 zJW30SBNf<3-Ee^|;Z5c3@tjeJP()a-ni$( zzTZ(5)u($4Eu#rxrz_v5(9;K^j~qMt2>DAG%BKW}P~s?Xv%ZEqRK1%gu~*{pssW6W z^tM{ghWe;F-RK8wGJNEB8@AIHK#`EK?I>E;xnh2+vpcf&4{-QZ`IT~@u;>)W)}u`{ zPyOjg&|1*Ds#jK0d=l#JdWD(?=1@-rQ@u?T_x<41H{(P2XdY6<5K1Z7VBOLP-8@E% zZoR2=9K)UFWk^RruIEAHFQ#^9E-_>*pspToIiWxICk28c!7|^iSscZXi)(HT?YU~7 z*kwvH=B32X+2%P{H0&r~+TX5dWxZBlmbIkRA}LcEkJGLRAyXhjOQ3{9|Iw{uSWXCL z#U{;tAH~H(KyZaNBv209qJr~CCk0YE3KV=zk&CRCqo{jHblHbr0nK=Yx|JKXEBg-< z?ft?$AEVhzbM_jA$b(cS{h8Zq65|s8l-D;(=Jf_p%u1`>or}{TZZ>LNjAXUs)as`0 zl14q##*XCbq+|_Kl+IZwX6+dl8e_g|${_Wr zlOil+JuzgNd!f&@A6g!h4ZB=pT_Xrd-(Xu%uvwwvI;czL67nR^?6+LL!?A{?Ei-||vSLmM1tyV0;al45GOti#Eej`e^ z6`8?^ao)(mJOQ#WccjVfjU>D)RjSNz$vJ&4Zr%B~Fux@-io-7G`p4Hn7!6xLCLb{j zn{5DvUzT6|_wQYbjY+Rpok){0 z)mvzXdG-{6S-$Z|;Y<$`+xBU`0hS)_Fku2_k2^sG$pqKzT2A9lRw06=sL_`h{PDlf z*AqvBzB4ZV7OSp-V>#J|ZJgXu*!(dYdSEd~FxRJ0Sc?IF!0)yR-fFn7HkE{;7*;>t zx)L^9V{uA%mt%5fxYVxxz3|;AECOZ+NM>^=Z<>{cV>Tkv^YR?+tx6p0X=CS;PuHKm z!^0Zm(r783QyWqkBGvukIkW;Xl~NMZ-Ti7YW!8x4JQ~hU#+94XDNCovb_lVr#WLMA0eah6Qo%iQre*N^Cpg(nCF52CL&@Q8TaGJi7k@-W6cMi~?$q3_-9$ZAcPD6vYA zhl{Pb`d`{|HSNfYq`;mH9@B@ z;V1%lNQb@az>qe#<{4p5>K)#<7B$+*bhIz1>b2_?vKzR`JjmO{S-759s0I~D;W*pN zgF)Je|93w9S=5WnmAG<_bWNx+51FA_O4N;dqi?F+?*tQzhk0(x&KxD~2@%8UWs7S? zHX(de_}dESQ@8fbX9$o)Vrcxr9xm3DL!s=tP2_Jb>$~XglVIggQH?s9632;|iAPB) zn+wgh%Sq+hH4%CuGr=6Of$2kTYI{O&@WK;D6GHYf)aFsFjKS=5C)qvm;q6Twc9xp7 zQ%wUKK3dkiA<iUIrN&}`zWz%PO6J02~t6e2P9nO2@NTd$lUXir5lfQ z)Z9y`jj~e15g>Nq`&`>fmWOK}7R=}JNPv(k*pj|cU4G{7yyk%a2T4~DBcd)-JqpAd zdSs-8#$F8b?gA3DbdDqxoqU7GHlaG*zPV}rPdL<^xHG<$-pg8$70#y=df(p@?h(4s zm$_^E17gUNCXJ6Y6@wH)SP&L_h#GC_lcD?3KjK3}HMrhCDADV7GYwvh{}aQ_G0MdK zgAkvkWGZZO$T4Y#7HL_Cb;o&XL{4bP6eaH|NswU={ZTYb*cwW7cNHgL$iztLn_NN> zwq9r_UP=mCc=05ODghZbJq|Xs-S~O^OnPLf#$Rd2OLKlEj)}8Way%mhy8a-9{r=+j z+Cz41X|nu1SA0qW>~X>EV8INnHIik1GD9yMVa1C*N1yvEzNLtI|cliFuft`O{E8nFzo#_sK7;pAw-xVKoBA%-c5|6a!N(0S&faOTqJ-UzQ znIfn6d&j-Vns{m8cW`%?!(3mwDOak*uX5ZKorkp!@OsmQSq)Pvwm%Z38Busdz45`g z<1Zg|yUvyuNs%a&5y6&WgC)k+O4$Xm$r4^0pX~@J_mjCqr6imvs5DEgdje3lI*kIJ zTt7ynVDfc|Z^o6&2joL{`lQ!0`M!ms&8hB@QK?t4=YL%6iXDbb;ZE5dmezz(vtwCt zku>7#@;VD|pkOCFUnaOyxw>)MAz#k<_6g-l5+o0{;+;7tU+JI;%>4a zAF@^bemcI?nm>(x_h-W%F-9q8R(lA^NsrQ*WEgpJ@dbC@zGxA~S6%1;&RT-@z_!X-(2*f;rD9E6 zxJw8B1qAM+>6bi+RQB7LlQdkiAeE;Y*30%Nf%a!OoMau;G%G1KPZ#re|FE=%aZ!e_ zZ&O|03*8>ZQV{#aiUYH=ax&XR{qYZI*tYpNypQa8$kq(9Zeg*siuvBMGGTw3 z!9}@eMrT19EK3d>rA8rSFwDlgr;mOTl|rP)p)<{ozlyMoB@H z*7n1-b=1LOaWCQcvAhpOwuyR|+>UhS`@r=_w|nOa=Xwqp!X#b2$t~h$;x#&T^3vT< zSxu9#p$EO)l4-n8IbZDH&L^+;?-os$+4fP@EL~`Z1zrahhP-XMP!kfAh)Q82q;N~~ z#`&@;l$s&g19gyxqmAG@k$kBVg;XqJ%?;rY@mBaej(dSuEU6cqUlixRkFxR`ok6CTAsE z`=gz=T#UTkw=T?4>+fZ@m)l}Gf)s>E_l^O%1%1+&VcrY&oHm`S3Ru0_{o*NcE)DXk z&ur?c$+Wz?k#>S4-5jAsm!cHe8=FX?cLtwlmJU}YmtLq-j#Kyvw#1cvkeI58=-s!h z3T_Tlio4hA13RI0C}Q@+Gc&tT#K@fvJm;^t3>!6CB|$3;)gq7_t-L3?Vhys z9jAmU8pu;JW+`W%vPSrhx<0&AMu-kk6i~La_nlfY*HLF$Fu?51)k4^fZ)F_9)4r>CGN<;O$6d&XPLw`4HSXo*Vaek_1lqpXJv!IyPEFDx z+vu}JwV~)ww{Sb-Wv%S}5ZOzEFM4M*L?Tg%<*C%Ygi)@u4FSE_D0R93aS|!!gXUnK zw2C$w1^OlAZyTbT{IW1ridY5<-D=ZoQzut57+ooTD7=JUEI9_#~;Z|g>er#6wzH&BPdHu4ecAM03ISw-H6aGP>qq;$uO zQYM$sE@1!7HijICq0Xn$f>u8(k^qV<#cC3QbF0~ldMCaL$;A2!`6Ra*3&dq1Wz_h1 zzHRCPm7&Mfe!v7{#gkwYrJ>zm#o^WL9RHqreZ6Dea`qF&Iw!nuMaxN&RPScH*O$F? zBKxcaO%7s00^!B2c@5NPAvN}tm71)}7DCMRDfWgAlE7FjXMf{1T$0Ah6aPo-PYu4I zKi0Xz$29gaTP1!nn3L>pdg{pAwRIo3%9dzgA;G-IzBH%$cat1GntY%}>Ahv^YC3LI zT$a|A*=H+Umj4k6UtzAMm@yNp#j%cl@aJG$r40GqsVpsL+ozq>0$5BQOkoi}Z3qQ2 zmQtXDd$7{*0Z^=xcYrW<;gN6+j+6r<#_s(`b*?+!JF`H~>v+1iSOF++`c~z4-n(<% zm3oGG$^mxo0{~uJ(s8;JxA3UqtIFTY)Vdy7hxACy^u;sA%!l(G0b$-HmYkgjV+umT z{&DS7VmK8y=BA_dydZ-2PZfNCFvnynSa^yPsb)A6A?x$#crOotK9j3nci!mn8CHJ;%Eg5CpBDAV_qESrog4Q-_dk+(t_R}NLf?4Uh5m0hu{{wKn+G$Imm95B>c5IpM z3z?$}7zZ|-j257YfAAlDc=BP+bKpiIefr2lz$iCceygfJ*2*t0;26=OPE(b&5xaN{ zu;;Q@Vpc3|cK93J8~benm&_I}4Uc}EE+>jNQlvS4cDgZVDh;+UdQw5C8dJ!Ye;^`5*x%w~epGda0rC8zEwy}@!?JsaOj{LvctKGnz_on6V^@Rzr!U8?I{N&GZ zTuQ|xQ@1E!)R;N~-OBRzZwApP#kU@oC%phCx(fBK#>xurh`BI;xjJ<*pz4YD25Qh` zY)}aw2jjf}Afal;Tu?w4g1-Vjjiv3D^a!WFBIaz6Wp2D_?EqNwu7KHq{pChMt&Buq z9B|BOpU z<~D5l=Rl_JOCu-7%vhqxPBwP1YCX6$QYJ$d4X7!Z$zZdmR_rjtzI0n*QCNPIHyoIZ zzIUCC1q*)aV;SwQ`!;=hkguSMX|<%vc1LP+CO!4xEi6b8-Q1JTx3;ZHfV{t@42xU)j@{5hVk`C>&Bko za=S41O?I7Y68!!ZTFWUKmO+_;n{Au6yG6pS$p(ve%dr(CZhg_S9_I_eezSFcf6n%2 zac56_V64IywJNrnM2wn&sUO?++89#s}_#Qe3jq;)RacIN3^;^i=!)s<8u8PNNA zFS(}rFOIh^XB#Ld{&dP_47<`g8*5@0*}gk+LLr0vEBJ2xGzZ$k{d$;^15LM^7%J}J zQknmd@2!bdJQcK4RYCG$-a;YT0@`w_nrtdSnqJQg*p*6{zdGA_Y=L@$yysFRpL$7!6XQ-x9T1f9$LYC|tg8|`7j;K%80Y~4ol$T#wxFSlz>%ZwjOyWEF*zu5epU1Td z98;4wF`lPN8m8DaO}Hs0KF8uGu730+Gn68h90dNZpGP5*!M4Fm8`Z4BFs0&S0+icC}TaY>0I=a=lK%XqCS-*(h=C zC5he$M2Mfy$P5>}zjas2VGh0d6Hksvil_4c+p`t8IZGeEe;P4*!fT&E1LxcKfZ?pp z)R=(2RgCdD>QyvfP3pL&vdyKj-XX?5@n12?++Z@Vm3f0Djvn50ZRHY9pTB97rL41H zog_&2XB*_3-wSYm3gsW}UG4;Kf9BPYK@PAqs<2I=g-1{E#FK6(Id!SC;&WF#ImB;O zkyuKRvV*o!;fnvv37ysB>O z#nWtNudR0<(aMQ%7UqXv@d?nl)w23eu@NjxPA?EZRz$k&xp> zE{=v|z?}E=T*GoD%bb{qP$8DKZJhpl-D_zkhKOmk3IN0mrd7n7*nlBx$3M@%*UaE8 zetZD~vRJbq*E5bm)RN)Fhgwp|5Vp?!0fhjG!<03w^?=&2TZ)bKE+g2v^jjam5}f4o z!P1m@HXvpxUYNHmvbSJh&;bX*=sWl{cY(2KRccQ6mad1P{k%tptYIG{dw}Zkw=j6% znp(@xy=;Ds2wb+f2?j&gZeP8c8(X8NLR8P`Tdd8L`cu}I^p5P2D)yiA)y5{W-W{YI zNR|7j7v5)^9p9KJqGdnz_s-i0Rpd&B7`kycg}Fp15AQU1?OEOUYcTA**_?Do*!&^6 z{y;<9ave>&BWl(1bw&e~Ts}I`)V0XgXf$zhm*}6VD)YT8`upqx6`*eRo$6 zZhGyF1(#93jelW4dcd&X8lcOcpz(ne^WxdXLsxt`%j!wXV3uQ$t>T{uD2fq-H@AgO z?>*aloL}$w<7Qs8kR!|qKPt|*ZwXw=tgDZd%OgMaHZvmjnDZQ0iii*9Y$Ywy{eTyo zX^m%~I~F%A960u^Y$$u|;cxubNX=8I0#R1#0)=N>EfWPpq?$JMpe-TiN!-0iw&)&z zk_#T@zuWNsc~*AZ!OvN)EkZ7LaABOw-d(}iT4ey}Ln=Vm6!H^A46`C4C zY)BPFbNgx6{m6KRi0!#@nuf+uHSL&+a@BtBh@MY9w@k?)@U&gKWK>v0R@mcS3Bn~5 z%|Zhc8NU-$p_h3RvIU)DX0?^jb3aCd+Ka~6iwVfx*T=bHdVME?n@Wbj3ta;IhIfZ89W5V zIGz1s$~@vKMr2)KGUNTG9#JW-d6;p)>0(sY7ej^7h8vV-4G?giAy8y3^#&@e>N7E| z0G`S!FiHRA2X%MBp+@q<;1Rw%kZP1z5A2J(LOc{+?EhkZ=ae0Xtl6 zb)s4+)mM?yj@UYM6g1hP6H@TOGTubR$su`eQyJ(wv3k-KyUR92pdh{q9x4yrCrauI z;r#<9Xhv0cyV<%PQ;AY)a$&K?hsQT9P*9ji{c+l0Uqp{Zm+yw0Y510vyT)CIu=fNc z({sLv+*i7LtxCw}Lw(JL^B!|J^Q#D2KauL*cC(Z)G&)Ey3Q{?gleeVv8WyRR! zAvQG*!=d;|{7)u=tAw1*!NwYP9KNFkn?&_sGUhWht6Sgqd_G(1%WiS{z^AuUFg;gV zDle@cL%Ye#RK~NoutKsnb|kOIJX?U#dUicR1$W+TmO}D8m2Vz$NU5fgt9u&gK1S6IYVAv%j z)GbR4l*;yjg56EMxr&>`sdm4q&cjscJn31fWAANCo>-1w`2kFLa zbA3H)kv!ft>XoPkL#r$d2ukvm6!Hq|k{u)9nz<5;J)XhSJk!hDcajyhe%>v3!M%?a zni%PR$c5)FU&~ZbFMasDopid&`G*`Q!O_OuSBnQ=X z)`*q~a0g|{e6w?}+eOeQ3$jdi8RGefkz+o`>Zq*(&HRT^Dp9`ZCQ^+}K4oFoIB_(N zCsfFjV1slm&`bu)lY{W%cu(&jmwx0lkiNchg4AEeFA!ek>SC)JtA-?y@<``p`YG>) zSuPz6ZpS=pk{}#P;8qaKQHxyCoDZj0$)%r(8l;+|oF%oFeiA!8?g?{{XJTdYTU^&F zQt9ysmyDiZB(HW?f&GBBaHd(HIqg3M5sU}>6`oxiwQ!z?T*SOeBykeuQLasrlC z3(31Z3&8G0WfkV5tVmhc`dra*ML`=<7q3dl{AN2`i$S7;>4##OI~ro2cMTLD0gFsW zHTu*=0&mn9a#Rp|Rc3J+kBxcpaih@*8c2EvmN*VHFM+6(aLd@A0HV851Eik$VAU>Ybp@J?n36jcTXR_)8NSNhOHs7-uNv z^WzDZ_UlhAhu0Uoj9==LcdPzxaF#?gq%1H8&>?4Qk+o9`HefHD*RmVV@T_v_^gug* z4+37~ztkmg}fJ+kz&{GQ4|AV|t+zBe9{gG3@V{9kbo! z(5A`#C!#o`<{(3^OZr)dtwmH$8s^EV+$q_i^%_?T%~j7j_2AO3Z5CM;M?l&^PNBR` ziZTgG&MUB5XI^rA;^UW}l*@T|Ja7FpU tDn>OuXd3apMl1dC^%rLuPqR{keUmQh zJ?IL{)$x2t>n#oJEi00M;T3%f^S7n6%acs(wZ~;G2B{P~ z3hwh0XZ^EwZGLB;lzj7;j%n2ltJyn@CePl`Vj^%M|cJ+sUWc^!Yq zdJ+p?@vBKC%VarDb$6@!mNuR^=Cnp0G7nLd-G1x!O+ABFF$;ouh%sH2-|Q!WaPRj- zQ}aHEXSq}O^pWik5PM?s+6rktOp=sFe_jC)3A(dZdY%ffY*uo>*K|)`QUpHy3J~V; zaq!}~DN%7%7bQR$z-%-3pXq9=+O01yPj=XWEgxj<2p)e&N*bASGl752y5H2POE-WZ z5V{L8XaC|VyX3mi+dfC%y8*(I`Q2}~J?>~go7E+)j^ys%Tc4`XthzXd;al|;_K8Z3 zm^=QJz?eFG>rZ!UtjDyKcS%*K)wa@lwx{7ROGS6(lC5GcaDEFp6VW+=v=g`CTP59< z4}_d&oBlYFr|{%;kDGGeGw&C7NFQDJrd2zvHGE4`sPdBWBKrm}@|(xh+MDFazqqw8 z*D;|p5>UB^_g*aa(SrxFKfO%0{eaV|>nljxU~Q9HMOWO}vAb}sAC=Mn3Cxn!l#}s} zm24$dQVTnxFUh}WI_759#RcyDdN_P*>0j@f1O1uZ`=6j@ncSd6t%Q|W_`7D<^S!t3 zRwx}~iZFg3F?MlZbp>hx@Gj$@IkxKS+kKmDW_un=0MQzQjvripOdSdDqr>2vv(28#<9nmG{t>Fg;J+7% zeUxNkifxqRF)UQO+Ykw`wbt&+1O4$_C3}^Fn7;@D0d9x*lOApBGK)%|Ik)zsS6LbpRZ{s9~4!xEA(3vY*d#E(%JP9q9$> z6@+>ByO#RnOg_$f#MOrt?)L8XX2*WXOkoKKA(wZas#>6pJpj;WDaXqVZpTkIfe*FQ ziwe>*OJ26~jk6y5g3f;``9dfk7yku5eeq96n2MSY zX~{b+QmRoL?NT-Yw_>hR$#1;%OIM$h1~%4#nqAkWX)!XQ({ZFv$ualu4H{W@8O>i;*hIMPu>yv{bgdbVBSl{r zKLU|b8GO$5CF|Wf+$AB*;O_8){%?m7z{-fej(_(a#+CGqb~Fu3YGUxY%~;YqDp#0# z_;}=^f_pSaCgBJ#eiJ9%lFK!ZJWIH^lM#m4Va}~GzB&#_iIMdwFWNL-3hj$ zo$$cZO(O5jWPw$2w+>>sh4GVI6ltaI4b~8AAj!?6U_uezQnkY2>`W6s?^#BP@cxs; zD_pdfIV(MW=Mr=P-!>8$Z7Pwsi~8p&&7J)XXE9v%(K&`N>j;%76lYys&n~xT6#K)8 z3v4N?p*GSWSu}xt_=xHnj{S$WE@`9Txz)2X1Z|{aSGK*uz;6rjesI$`4Yn< zd&OReVqS5bm!itlZ#cI}9DLZl)RfJiLm^~vjqH;krF~NE>0U2o>&QK+Hl{pVUxU&) z5cW6Qtd>TyxNAIErFjyq06znun^@xo6*B^O_?yS=m37v!R1VHif6M)0vr>dh=VnCQ zrVI*EV^sQt`)ogN!=s36OGB!|6Gk#hHCaV}QtV2cl7066>^4JA`%S!t-mEfp%4v;2 zf#0?Dx6H>cvQAM{{81S8o}KmLpKchGt_-aVg}llt;wAxV;p0q*p*Gq-Ta*sJy;?nj z#C4i>Ue=It-~F1W)p>AO$25)Zoa$%7ExI)SLNBvMoBGL*rvXb|y%5;Z7sl;e;vaFw zr^;y~TDvfK-wZ#B7_E(Yb$lAQ;#u9aMB-SF4{7SsFe4{kb5lQa{#2JO&JKKP-XcxVq{I8)VeD ztR*U+I)p0w%=kXTsl_aVUE`76=WsQlXZrtm+4f_<1cPw%yY~}y#PDXp@-Poh(bT@H zq8hXu*nW0u?q&ennZw|06my>qPiRvT|eZ|ExRYepp*|!&&dl;6~6dXC{g>q1X-+@w(V)){a+9* zCT|t!j;|Qrj3S)4>?yRn24vH4pSqj{OJ+UPOFSEtM)lMSG{!0;Kl6M8b~GFRY!qy4=wh={8tM$+wLe%TIsS^zXt{70W2wej2#3DZ z6ZotH5k2aoHG&4jc))dS{!(f7J}c66*X2e;)yqRP(yK+>*!&4;UfvBveb+(Um){1N zHzUK?sLN%AKmNsJi3)7se&h_F=53(ye~e?BWOT+l@gF3U7Ts4Vc|3a=D4Y{#wkI)5uTVQ~~_Q)@m&e;2j9-wEOnC(mRtsDSrc`M}IU|N#QiS z@P1>k=mui9$mPR?wri{6H08(Q-zOc$EWtD{Y$wz0NM)T3N^d=S3udtGt+I_wg7M6a z5g<-N9vu2F<0p)qy^xvnu|I)hC<1;}l(i@?)K@L7?#S*anZ_ihhYjNESs6=7oO5)@ z=JlG@kd&qd1;WiVD$R&V(ZUX^OZgwTfQsq;$>^~pQX z&I#GOo~J8HD96Uq0X1tj&k;$jr@O3q8cuk*LiTC3ddw&Y;6W3GU1r`}tDJmqa*kk? z;)x5D)FNB*dy35;wz>AmAu9E;5*Tl6>1&42U(q37bNC-n_bX{vG<5- z>zGvn{lD*bZAbuRr}o~jk#ahVm!Fp(#lFLW$4{o&X`B=&P5|D@LV6se6ZaWOU@`?}54jf*{=IQ)wTMKRLw|G5 z0yhexO2wKMa2~5?O#f-~2th==i@JU>A8^u~@@R2)ffWh8t^lrpf%LIL&gEfO!S0hK zPsmWf=$+}N{f8T|6XR+s#j{*>#ByUzRvrGn?&&b5B7Xw=R~M>+Iyy(w{cHW#D&U)$ z0L;*Xe@ZCKfcVee`h_Wjel7NRIARabxC--XO_y?wGCh*6p|bD=>$-p$-OL@c+oBN=~@%9-f$3T5r&4v2jsqmC%vH+VC*1h)H-{zG)(<>Ys!YmHEwe{p$V4sxgPI)#GaZo^|{A&i%$T)6Q> zlFYUQVOWJ3KNos2$oiY9pODXq!_pKj4`dOnBsPuZJfNhH9{)Q!*tM2uaIAfv(jCznf~z`tR~JCW(9p8 zS7RT}R;igMW84Fq9c~#QY;hu;sfNU<)>ki*;{;8S@L!QhNk6u`3Q_lTZLbjYXxjM@RauyrT6;oJMTs^+v?`4QZoQv3u= zDghAD?e@59%bd13&&wliTunv^tg?1|wblAz3C1kR26y$&(-~9-HW*+|m6s2hItaHe zz=E6kvUDY1`pfj8vCoOXOK1zRGspRt?>MU^ki^)rSHc_hx4rDV^1Ymqx?Lx#%W}4U zTHPE2*XswKGwgH(tXNi&zIRo^ub6|C29%_fKfkY=x$`<+IW|7{I>8WquC~4( z@Py{`^p%pYlH{1-O8xU$+N;;DkEuY+_AQd_^0n`_) zk}z#ne(8}dI&`^OU$f54$&YW9d$yQo{Iy!66ZSeW50`(&I7Tcke)uS0$hG>Dp~rbF zz=~`w_n+uRX?0(V4jHByxX-8kjNP(FI&OoEq;4brsM_IrSli_EQ|p1;GNVmWm1i|k zf8zhNlw0*XpD725>V_&mo6Vp03_^{Ub7EyBtPfuAmtPWZt`oY zypL<1oyn~Z*qIq3*w2Yh$4+HygY@*xWFF-3$Y1x?O{nAJ^1jp>I1cNzK-J{|Au5!Gg(w-7H=CFGFh%X9A;>KcW?Nc`8 zurAU_uM}Kq;%cL2tbZELwNwmRx+s3QpSfm-gwUi@FqkNlmWkwP%?pGq-ndK^aBTn* zPzzv~kiv)pBdnNl-Xp!qFpCoiB+-;F&2Gh&{7k4Kh8&7>j2Q>_B zh&c+)xB1X!pK&AG$>(xBzb`*YSlA;`_5~lQI0#vja#oH{iEj@56JMag9Cj0SqRNXl z5+H)|=x5-jaSh?GVXabR;=b6LA`i5n!$*cwRpNM66(8WX`|~mLvnEAUDVzm2WJ`&0 zR9mD-vo3d4t2?UBU8s_55zCu@73-VnZDD2%9;at>OFLS9+9OYXuHp=N>Ci|a%e`cl zdA{+B??+bxm3s7LL07_{0!%oSTLHO^S&#)qg_hL#+)Zefa^FT!iDa9u>~;vd6w`C) z;o#R&%8FL~`@xQHD!Fw8aM7Z7?hmmoRuQq4W~(J6S*Cpzw7?lL&Y*{JF~nS3;GrzMFn4)vDN(SgG3lEc!jWM^Uabf z6cQ!4GYN!{20;YIBM{v_-N~w)lZ*%ju9SQu#7Jsp=s~t4gLf{`d>$8teaiSsHaCvZ z3?6f{&X_DOQP%-CX^bQN@6NreJ#%QZgh=yy1FFNqJWP9-BF~^&4GHeJ231&XuzjxN zKs~I?$d`B<>%;(9I#Cz)BPqHhcncYBsz~;eD9>oQo=aPA>kRO*Y;tLO{$;B-=Lu zqKqI@r7ev26^-+R2i(>jVik0CRBBRY*zJ~L{9kv}@oTy97&V9ig?IVhRgLIap7GIv zD|boisSYS~7ExU;)cZbVGq8T(zCTF_(o~zX5+`y5(w=N#YKus@5Q&+gBAGY}X@9#4 z?(N|_O0TN>NAg}pnY>ETh2GUCvGAM0@4j+(!QnzUlw+v-9P0uy{XVV0$LKJ&vgCE4 z>>9J2hOXI1UD-VOy%`0X)=p@WkR~u4Z|*Yo+JY{8ezwjujtVx^&7W_Qu}^|qak}rV zEC)8pbZfC49Xk3@c+?2nhpekhgxN4Cs1tcba%0`y6u!=KYueoyW8(7Ba(d$rJFlyc z2rQL+7QpqHkhS|nGDIF_gRHGF4OiokFcOk`gXftuUXVw;eYiFK9^)_gX-4HSj(qd0 z-&`5J@DQ`-@xUPrv0_Nb2h=6-@Z%dP1Io&~=;1}(EK zqiIx;a@}N7Y0E6;H8XfGf^CQa-#9(gP%F*-L@#b%14cN|-oY1-ns*-(R2JWSapunC zNv@pe7Z*ry^=gzE51jDdv;O;InBaRYuPI5ku=>!e=ekUkcmCFk680w04L7=vPO4Ky z*$}Sd_&}t4kBZi5((pn>uFsY8drlIcRbjnfY?#T-OTjbA=&)VKF7fGcccDm{c4aqa z`CzZ3@Nbm_*D3k0KUYXg_T$oVS|36O!ZZ?>pNQSLp1G&-`?_gsZIjU;IeqzVuPK2ZaycLsnxvI>1>4m;alU5gX%kN|5NgqF zbC~FVevyCQo<{xua><)u^8a;BV{I_p|L@ZmqVjLX zivRiK|NUV9UDp4%#QwWx{Qt3z)ff)bT($qkS>7qdgI% zVe>b|wpsDCz2zpqvI5X5ds&4jW@+?0@tZ8)z?+82XHuY!53t&cjH!1PS>&>_5~nT+ zR2U)ADG5Dx_bzO_=&%1i>iz-b%DtnF30^bZ*aM)~EIS7K6JSo>R@^-L8{XV;E{!B( zrm5olQ*&HH(BE!-@8%&|y?D@blHPYiWPS&TyWMCJ<|1t`Lt-U5uI0_<2JX{PHT zD6W76jLKS}{(Zps)Pu~()*w>w`4I+LGNU3u7ia0S7*Xmto(lr9C)@qJJ%Onp3#i)^ z5swo=Sz+IRO1^)uPDVlR=m0olmlou2mdg?-qqTpnh2PDd67@g@M9>L>rnzGe1~`K< z0D&4@{9M{=aE&+RBHyO)qNM#qr>URNy45~|0Q23BBz)jhKf--u{Bmbkz z5TqRbe8P8A>XKl)8IyNU?i`%V2RYJXn!(o{1%>~g!wi8yN9i(?Rzr-(8=Ha^zV;YZFYX21^Yewz;UAV z{Jc|na3n`|97NZubXP`DH5Y=>)R|kqis$z6Z|@OWBjx4q!69KIALj!YFB`U{UA}uA z>LToPJFh&8a8TTq_PgShn#TB zbYo2C{GlKH2((p^+r2s{k)+D;Ln${h4!gjsqmK*k?e5ubQmSNd_eN%|zf^L|ecAY0 ze{CtV`Ptu}l_v+!R>~#Uz=kzsb6^sc-XLF-st5A_KD`3F(ERN}`Ca6o$OZdQ%z5Uk zm%^`YdoMikqqft8Osuk$922bk1Ie#>`bawJ2p0TygP0U3dR;EVV%}DorH)adhcZry+qW}Mjd4I&M4o`^M3yv zhd;*a>P1p~Q=jWM6Uf;$8N%2j_ zW;Vy`><3ctZj{Z~P0y0s)L5m=$qt3flW%NsWafiAtveq_ECs2)xEw%BV&vHBqY8q@mapr)xTc~4-ipZZ#ed4rUZsm4G@}Y$NzUDUS{yy2Y2lAqm6gWgCZr7zcSIS z9=2NdcbBG_o44H&q#&l(6KEzOOGNPEZBeB46LaLY@N(rXzs&?x))m;C*p#Yfs4Yw3 z<_)cNh#Gpm#i@qu3 zpOGrDF{1__?~k?<*}3m$xea>wJ8nOK=prtDQhH2W{I%o=5n}g^8l>W)7b$pSFpB@p zs#lzd3Rjl>bcDa_X;Nt>y3@hssw)=~1iv=r!#4pcR!^PR&>#Kl`OA!8oB5+um?m87 zPr`q|FW4<-%i%Ri{Z9I}`QFAu167^(QvV8lx@_?6acN#!#wz+({MhYvb`XBvx$1TK zF$Je2Q(vw^boRFYxge`@5{b$4*BHWGSjP5F}57Z|@5po=ba?L}O!?UH?YfeYElM$my6;IvB= z%%e+19Z=Odi=dyaFLAUj@HA<`&-=#%v3W2=Q_^M5Tr3IuDq=WU2i*>pj!`s@<&_BQ zMUt@?>GMd80t@)paI-0Oaqe$)oDCWb$s!eoV&WM-aCYDSMPupW53305` zDPf2xB%Jq3QwRQO-v4_>u4;-e07gN6qpRYXjeRMLK+&8YVabdQapqIEx2~3xLr>hU zq`g^xDCYFpD{o zAl-7i%GP!5(-3iEBPZh(?*U#HHGl~Ow&{H=?zP1}_~&^*^r;U}My&8aZnz}0PQbF} zvvKIZHac^-ZO^F{x@)W^uHo5>1EROt2F8e5P*NAfjM3l9+cBmM*U9 zG**ewpL%6B8WQ*Vo~lii$nUI+xvDTI07$z;iw6=!E&G$=W#*zTwWZ4Hg7_1=$!JJk zrAHG&jYXaR%+vd>Vnl^0g&QFjj<5Ec;+H1=!L?T=^Uy_S$BVL-Dmd`N(qT&WVM3u@ z=lxBtkRSN}x3oM&s(pvemWoqffys^S~HrJI(W&x3l1 z>!3sL)T0P-K)bxVS zWo-70-+;Vb%Ey-jSWo6>i-&aba8h-4(z#x^0!m`f70l8lQp>Q@n%g@8n~%t8;0ksD z-nRpCJXdeCZL%}Vc_4~GIIrV9c?xL);@ULWV6zpFrmty2g9)CWyy?Bc%MH4OFFMht z@5nQ@OIl??fa$C8qAu>pCwRkES=H!hUxTl~LIX9B#AJBIxJlLes5UDEXv z_;_xh0R5-aW%o~!m_ehNg5|@pT2UkJM9g-oOUH*-Os5RHX-A%4$8IbdjXZ03ReL77 zZr4{Pt^?OlC#RS3REuDZt|THYT>@12yUHQ#T5La{HwYA`szLR7azHFskeIdmcszQ= z!JjD`{Vz_6**=pN)1fCGhnRc}RB-u|%sx|+>e(*4H}%=v_BUdXn|!`D%zGdtdhZ!-Qf%v`t>PsYhlaf$Xe+mzMepO8GCGe!O2=w1 z2jK{nVERsq`**V}CbBbP1w3?tZny!}IEX{)P$8Ta$j8 z-&C@O%b_*&46ZxWYl3M+2_ZM4(jkGxrvGxa(oFi!uaeH<4lNP78PfWc<(8agKO7T@ z>BL6sU6vRAHDI~~`OZIeD`;=6$8bx|e%KaxZ?7|dUj0g;>9)PSiK%+3kPJQ`fkXqa zbJY3G-2~kBe{I|y$&LeVP)yRmvroReeuog`car_(t_``A(WbQ1VxW0eW>OamGTo=T zEn)s%zudT(BLhW@)w|C1)9a%jZO>~;`6k2-VIRyrc%L3z#UZEcr;O#;M<>Y1H}?D^BW@m5sU^M6&{G;k=;u8YU1 z8HW2!x8hFiUkT6PIb09kmVdPIDC7|6W<%!?Zv9WvsKYR(5t~au=;04@DMs(FKWg;6 z>r(H)33USh%Jx5;l`ZD1bdmvBVg@U+NPI>%X^+?YO}IepX>2DJ1_6?zM*-WK+kc!_bywJ z!y{_>Qh2R}y=H+jr%wE*4qX@rw&QaG3V0|p)e!=W)Wt6kCeHs_ZhsIZlgqA!&fqgTX>&*` zYP>7iFN-IxG=O~=GTYb!J$Sd^uI%YGK!QAZ!%uSGiV03be~DZVox_6%yao`nn6qJn2p+>b{!({&ghqp>#|bBnZD70T z#r`Tk&PJkkQe~uNJklkV>2y|KTtnJtIXGT&_CnnN2wx+y0aSg+9wL4bAcBoJ*eLyA zxh)q})nwTf90UYbMy@u!O3zEKFb=tCS+V&|=bGcaZ)mVMer)8@3agOOlwDkGKRxih zlo~d?M?8KI9wT+NX~b7GR?0qGK_eep5r)16G$@$D<)zz$M)*07^;1dVqePg^?UjlP zZ@<XyC9a1j5HjmBj2ZqQu={*WjHQ=R%JEf_j>owg=E`I@gdTN>-tjmi)=Bq@( zpn`w}`aj=y)|!IVZ;y(TAzjb)1RVtErP1())^8ahJ1lz^HWeD2X#MR%{|iZ;vO*t3 z{`p-I(Y>Q#lf#qtSFuax&cH}VW{jmw^8$$Q^x9t#y?W!$*Lh{pbW!h$SgBAfa&imI zbkVrL1kG^A2ff##`*-}e@@9~yi9t01x|88iv^1OmCaH(cM0umJD zNMR*KPuG!$?$)}>Y0zpuNarrh>DrfC`=MpYKQ$g4vM1nFtu7Als|Zn0pf9yfDXDgR*~IRE9qY0AmV$Cc_!9VwmQF#JlQUOM1?mCHiC zi}`u%w^?08@4}L4wqs@6WT5mlh3#Rr(bc=fWW1R&x^{)~q0wCNiKGmN132I^FFtA zIr9GOzxG_>E5L zCu^NrW8TMw`Ual+iOc!?SOtgY3HEOvKG&Z(1CkG@5dJl^`yNeIke^3j^#q0yD0jtr z=l$^bHnB~&(&>fn66|_ig6KyjBt)W^FJIbbx$SN|AWyb@HW=M1`Xq#&#LbCxv*QxJ z_Qnyu?&ew7P};AVWmnmCY?Jvp_i==wtk%xBfFbLmnvGi4R- zFv7dsJYR(be{;;F7zN+WK^0&k6q_m`gveM3aINNV2R1SVU%7utehh6ov`)02 z{aTYj$D8L$d`tVv&9QG#%Z1@Neq_+_wh*4C%of{JUs9YGG{aQ93v!n4cOR9xbh_Ot zjfFIRdU*KDv}xd=t#qG%7HHKsf(wi#&Hk97$c_RKyg zkl63^8^LZ0Z)q4;CW5M!gpx`Z&_|~O0GKuWR?^@(nS5kgq*-G0aJ;-$M=x!aWpqg+ zn}Cs=iC1CN=+-&Q77$?lr^(fHs>Rb5#Mo4e*e3#HCpcSco{AvoNp}JCc!|4y^~sNe z88pu6hNhx$$Y4JUrenk5s(AJ;BII++Rc+a*x;VlHmpG91${EATQt$}7sPPQn8p1HK_YfK$yQH-8H%q@d6 zgAZt;44k55l~BNSWkKRN`zJ48CCa5{821@T^DdP7 zBCAx6GE+$H+aFv6*xn|74TUnF>!3 z)!mcjT`NM?+<8^YqaHu2oMq(NHj+RsDy17;W4VJgIuzmbl7Sn0T)&Tk6?1`{x;liZ zS50}k{tf=D9j&oWL~=zlK97q&w83_1cAgKiKy9M4T`~^mZn|8?8&3!%qs&==kUAXb zK#Cu2{|IWN9DKRjo7@7~W!!razC6)&R9jbZ(|O_$cQy%Ah&D6`iD_{6Z|&Bg&LbN< zGji!F19Gj_+SdPbSZVF>14aN9jFPY+2d4@Nr389smyQh=&qW3w$y@;j>{xKF!?dU0x`bV$M+^+^gi?2!Wrl4`)nDNCK;^+!I6|xxs1xk%AU9PesmmW;43>#$r zw2MNG7na4ER)S+rm(TanKT&CJMPOe)H`uSt(uSrHOdy5_qsSkR{_Tjh4o3J8jy?Ug&+S9ysz&hIK5!tY7*J+Yx<7cZn z5J8Oef8wCW&7_r46x)E`P0g#rT#E>%`v!nvD$|bFyE^uzcS)>i$7i*G6|$C) zwpcD7wkrSMX^=!@mCA+!(c#3tm7y)eVrEF|-p@I_&NV1uRB?GIMJE!BZbmb|qW`6s z7>87*HE@_zb0SVfJsjGceiGGv#&5EBkO=-()f?2c`z^P*>q%d+rXXPHJm@^|wo~5s z#(8+rxhWssH5<%Fm`_UPpDcTwAS}oIzT5Xq{ru<#z3fw{TVbk{;Y;gGQxfBn3dYTli7>3O1sy%kz~k%Oe84L0hzS;Zj6-C(6z0R(P`WS=wV z1m4~~n*`Xrhb=SgA~ypx>Mi8iPP{xG-r`gCp1g5HZw{qv^sVfo{nMw0QI* z10A@_qH-7`19iKV6vJd^VnzU{xlm>hN$1)+lP?*05E{54+_86>n84k|vCbBFH93&) z4n*c!hlSk#(lvN)I?3PB&QVO?;WhAqGJdCc$CdBR&D0ORsUSlV=p!kBX}9di7__Id z-7dGHLz0dvZmT=R3m|M6_IXa$JD$S!yTx`qZ`Y|)_Fn&>lWuJPE9op1RLNgI;7X02 z;SP^j4Zx3c-<@MDcZNy}y|ZEE8G#cWd*|2m(1aUe3rvl}c~Rm4&1r6`DA71a?I;Y; z6Z8ZM$)@G852)u>f~X4cTE$AfsQm2FNO(0va;+2)R#k*OZMQRDz23enxT zExq8(YzSb81Nhm)2Hem0#bTq}BHSd0+<9W=H z`cZ; zSu3QQ5FiH8+#&4tKHB}+tY!i6`HJm*_uJjl%kPR2jJQR5{xr-T=*n!Euj*}2k|PMi z7zb&FV@K`y*j|p&MIgK|m@28hbtktJF$SFvFXA6J+WQRt>Z|i+D)dOISa> zzdhyGV+dbv3o`W^+xN1rfOVrB-{!bTIq zZw`mfFM3xF4+agq|CI@09;*5gdeGAVr+H@bK)_WLu?mV0Ur{2iQL6KF5&fh69cW8(v4+FaNZDqPVE=u2JuBd#n#|%A;si&44U7HWepJ z{c#p90Y0b3yM$;L9;wTd{UTS$HP?e5qXT`Kme_H}B~cVh+TqoPRCb!i-g;v!6V#4% zs$RvisUu(UFtaekodQo;hJu+G#;A(x2H}uO`y^y1@e`km-l1#icmV2=wsrJPJ+C*x zLoEF}RwGO&*m`un#sy=c;wNN&v0h|=VQQ$}1-|60vVG4LG4mM)bFmO|j~hNUpyoe| zL40f6waEoWCW~YAfo~yHR&s9DqQ85e^#jeHXS`vvmOt~7o|U1@ofAJ*n8`&isSR*u zJ%T)qON-UfjTQUnD9VdF7#_psGb`d}ourpZzvq{OlPmVGl&LduhXY%dA+EzBzWY>mp zYI)l{|JvP!k~+Fx^>QJl!lLx|O?Nj95bB{G`R`NW(z4;PZA!z(ejFNs6U5XKgX3Q& zh#qtIZ~ox)5zbEn|8a(FEjaC3jqoY)WD6uz0MRwMKnJ$FvSD6IwUKaGeQlF#!0N|O zl}8kjszxK*^u>f0&C0b* zr@I)J8sG5{4HN#;n0UT=Ym=y5;GRGyKK_s5k#q+c1pvcY$CJ>)6n``)N&C;_(C;5dgR4GrGR{>@aS`mH`k_eiU_$F!Y~ z^Cos}WJ8<>O(0&ASWCl|KwT3gp5SYBzZo&}8?O8h&lAZ1Hr1zDg&LpG#0_eyuDb2> zHUVZVd?dBah}u)4D7qaYuW&>3{&pn+*UerE+X!(CeMvO1yg@duc^2Z~IeNo#^e1MB zO&uP;rFR~bnMH*k8U?De$v=u&2OJCk-h`Vy>4(f80b6l_@qdEZQaXm9vq@lPW(YI= zpg^+=xL_LQm$dHwDS2~bRP!xj1?kgYU6-RxpMJ~5yVoh_`g-DS>hGm#&}riJZC$p8 z36q1j=3ZGfxV7tUsjBm)|1RG>X31z`xa9+7lMlvcE{Wg2+a|eF{Svflo4E69=S*hG zZ)N&3uuAEHlS;pS7VKq9OG0guTemfC^;0LX7F9MTyg^Iiz+`Rv8ha=PLW&#x&)!G50cw2ZF@z*XJn2P^vG~f z{2pz0z(c)buSJ`fljR)d3u7mn^AiaZrn6`sMFo59JneGx3PVPhaT&DQ)NG~XJZ<{d zFKYIRj;5Zv*rK;J{G>mP2AsO?ps)a@X9<@U=o9h0p9Mh|!1%M?aa&cH!RswV{IHkoU5T{xy|~qsYu5wv zUi}raAYkvn$y?RA;>eo;xk)8dC0Ds{ZPzu`2<2VF$K15&Tx~ao8vCax%-`-LNKAmY6f{y28&8)z1?9PU~MreW6?-^=tr_m={RAZRa>K$`dlf7W6fk$b|L z!s->}1DQ$)ZI814718pSMhLj9M9VQd89ty77||e=C!c+tId3-CSGDgya1;Je@3L9TP5p??K1#}mqnyppjD50j>v z6&^>zlH-CDhxqqCIXA)=L>%-H^gd3Up5YzhR@c#V&Sr(06r3Aj+XIFNWwq5d)x{a^ zNlXf|Gv$= z8F+bf0n$@_iY6?auz^-scD^=fcKhP@(3@=BqPY^%3^_gi#W>lSKVv|`v&Vz9ZEy%^ zM%y^sl=C)vGJq&#&FDudZ$>6Y3iqY@n|tSmh54ynOQ~+ku80fRQQW~Q{er=QXASuk zL4ON9)Bo++oj02Z#R&wIxo84bT5K03|Mcua689TuMi$`{?5Qa=u_QePEeRk}g6#6i zy|){I%(r=?HJAtpNRQ={;G5oHMd;J3gJ9MACOr`isT+q4+v{_K1mB%})kl0vH?^g=0Z(-$;jjeKWCPJWSLCO-HYP{YNk5 z6uI~;UQjOK6ApI7cb9xAmzqmNyb?sEc9l$jZ&ePiP<`6*cDh(et}+q}MCVfdCE(ea zIEKhUqSEi=oSNc&UzVL$AuJ-2Ix#qDQE@NBBmlnnQ6SvA2dY{q*yAgKN|?&CtkTF| z7(Y4=Qk6}WfniG9j!QRXQ79@m3-{v(zg{yFA9tpfJ(l6T9n$^G$T_Z1V3gQ#;W1}p z{?)Yw{Gfz?3g^n7bV^YRV?g%3;qO{^I3)k9fB_y(f?&R1iqI=~OU|K@vxylTd|xlo zp1o^FsrC#K3El|%6>}Jhxt6G&;0-Rs!L$zbQw*9Gpq{oI=yW~|iPaGZ^WPs}bWU_f z2Q@bh%On1TEb)8m8B_disB9IgdG3@&lmxqvVRiIsJ?r=0+hV9^mJ=38;_NM- zLCjsd^}aPjKP+->%lf-f{C;ap-lEyvoQAy-6{fEIK6e-k28)YMt6|TJY!L z6yCtcgZ>w;5W@O)<6LVtt{VK|C5NwI$USD+$er=`-WFnbmYj}AbuY?Z;*~y8b;SJj z{wxMhOwCZw1*!iQV%FtXoc!)@uE8T9W_$^rp7&l!o3Xalzri62MRbzy{XsNQ6q{2( zjJWmEW3igFMp78Cg@Rl&`19TfsKm^-2fBjsS~SDMf@(bPtHs=_%4;NYipLE;zaui} z7M3mCB7fRUBDN)%FV!0Q?!BYtDd^1Q#G7b ze)>@<%ek_no2Zp!hQ95OT8W-`+vJ743(@KnxH07L1LH*;@5Y5;@PL9^Q<{d?I4c6t zPIZ^@h?0GX+x$b?cyYG|yyztk)XhrZkjq`s@6d_Py5jpiEO1QEEJj=NN*}*DMI??S zU`Uawc~xS2^~NvO6i2qt6d=;6t_m~F)wQxxy1O0q4<*`}4UD`GWX7gnn7{O6SVYb3 zE$)8Cq4AHB)VqT`^Qhc%kr$s-HhmTLI42oczbcdEIjhyo7~v`gLIrdv^>jD0fr>cU z`yI>!V>~7r+h5bNJIt9cHEJ=$@Fr5cv$s_XH#A|FL&7J<-q$4u!R>jQEzt~cwkeWY zu{8e7=#Z^;&j}&rdPmfFl==So;kl#W(8`&3Qt(ZJ=%W@oamHTYWC;buFs0r1^qb4r z$XM5$)22^y04_y!EExUu40(< zA$x*+zOWWSoWq(s?V2LGIfT)2l%Q%jwj&n?%PA=wf9Nhmb(G`rcCK`ViJS z;fd^`oA(p7T?q_0e6!r4P>g%NFZzKKTL}xageCQefV=!o8EWTB3lK7bY{4ejBVS68Fh}-6G>_S?!8atfM_t zu?wOwU?~KVL>hu5dCC}^rv%E9%-By}28VtmkQvf2!KT%4d4@%Aj!jhbp{~?^qc<<* zl`qyIgDWK|heIdx>e=|!v4(!#cO8vTg};)`QW}F8K?<@KA{a@4<{ z?SVr7)@m7mKFYT$=Z42hB_A7Fr0KYTlV+<;+sqjM0ua|#Ro>?Yq~E5^dZ`jA^_2}u6Jgg;!z-=3W~ zp8-_ERcr%ea_~dor)-Jcpr%SoE>_l|uj^(9&7+QHdRA@Fie=+V4JI$lS!$;4W2Y9y zFyV=aCtKD~awoWLCNr~-fw)_`U+Ff$ugovMU3u{<&=z`n0+AT?jE~HhqbH14NOcmB zJ67&YrQwPsyT*;7ubvF)JAF>_o7RTkB%}vI#mC3HM1t3&8_yc`km?z+am8C95 zKu?I^^B`Gv>YkOeY0W>>Z7g~KgtA{eU>DG_N8#MU;EnR!^$XG@-f%+Oe7_ve+Err} z)dQBbB#8VT$HNS!z|OkM%ka~?32lldI5F^V;u zISHC;8OoMqw;|6AuXDzUOI_nF)^Z_5}ZwHcFybl$I?lt=u+ZP+tf^qUX zt~W=niSl&lQZjZ@$6M%+4ZegFITpAtaJ$%IlC2(x-8z{U)YTQMqjWhcU(24fYenFW z3li@cGEa&Y2A4>D?+#A@-%ab}6()vyQ3ralNgUG#GBAcu=QSZ2jzagY7lxUkkNs() zM)!QDp8)&%W9ut8e9#5&HXz;F@-d6M6SCYODmcl@j3?#o__#La7F&v+trSYrKJCqJ z2>Ijrpv^+bV3lzHcf3WKMV7euQsl;F9o?Onv#R8icTaG#KOY_|uMcd}90~su5wKr# z$*{eo4^ZRWwoKGbg%z-7uT=4x;~TE)uzvrqC>9T0$Nb$tbWsr-oAJNVqzFPSHpar> zze6s!r+Moi^?i{$wCSBh=nRz{+m2Bv$=mluQ(DTctM*I2mNM*;xpKsAkihQI2Cn&7 zpegs7_+Z95B=s&*u1QY!wfrUE`6t#1`uQU2TsV9&om83h5jhur(KEJ?K3A~0Nqxfv zeh22qy?s$2mS*-X88f&y1EXtipwRfKlpcOKS^c9(Y$HQ>2bt*GA^`#c;@Bh`%rDBT zTj?&4{GvjVT2LT%0Z#OO0G(d!L;+DRAMOL>t;8VL5GVey-)X1U12PzY-d8?{gk|`? zznhZ^bJc%(E~!ys?H=XY)k>R@XRnjhpz{jM<;nAo75w zH2-hy+MR@rGM()UMR|MHA`&iqY<2ukzj#;7{~COm2+wp5jRwbK~2CEe7}hiP&^f~ACq!HG`mTe}i#F3AQ;BJ@5HNd~Us5kIS7~gmh7hg_&92-o zM_T-Hc_=aK+wj`I4632>QQCja6WM>Tj)&p0JqK7O9d1n|pj58+HM`$QOanulFcc*MVAh}DDEZ_C8y>oW;e&1a~k;(j?{6FfW66;l_ zvSRWC&d;d?nEdlW5>KY~iP!$&SUZo*5o=J@%4T}yg4t1}$ZEj9(rpa$p?65bM?PS_mkmUen`VS4f zlPorDsB!;$tzqmSceT=D$ZkXPM`*Xd5_y}Z*EYHqmm*dYXjl3|d+!@Q9JpDR(xZ&* z2VOXq$+BSGU!xCu&c!JbzK|%_d$J$@xi0%p>~e@T(9}s4Y50_Uh&yWgL8(QIe0t*} zq^J55Bfyy;!j18nd3%y1>@2f`KWViq@%^=yWmULUkgIE&o%SbyIjiYbJu4eRC8J*} z@~*10)N@wgn82#Kcnwl!W|74_S~v9Fq9_mu-gZGX;gw!L%iVUhebPe~1|0%Sx>>CFPVw^XD9#cg%GAp~ zxo9OWo>Pd2_~KG`O}djgbLqdd>%L;4h}C9TbBEQLmw8h+=GfF-?U^N-2uT8aLHtEx55}e);uf7#=Dedr-+PH>T@Ois3bURgLQl z9|}Vhs%8JZXNX#s0;k2DM&VYpg`kIeL4=XmrlKejfQ~ z4}a4u&xPLE%F!jeD{ZE+F*FEYRcfPkZW~KqowHm_Kn8h2)Z!$MATVIaH zN%%3COr%WZ{fOJ(qL{}-&y;72eaG54+iBQ6x}Z1MU}Jcs@;I9uCv7#jCYNBx@5KV@ zxP1NQ^MB}TPDe{BZ|OpvX;}{ZMw&q)d+0B6Pp43XrEi#p3Qp~U{X7j5 z&Tf78c$T03bP%j6#(@Lb&gnJUc6|9$6(6VqXh zlcEcG3c9I1{&GZotAR(VEPQ*lL0Cw#U7=Pe*qcv8c*U90RO7+Buov`v?$!rs@-V*I zyLk;5vzr{W*Jb?bdbroA7T&!Kul<%lnRlDsJNyYS2wJ|@G>4Op?B#p{n?;Sie!F^m zI<8*+IfdZk8h0=94y@wlgnfy%Lp-0IM@7YZ$`I|jFdwvG&TBhh*C<8_r9(L^Nvygu z-sIJ~!sXun&UjamwPN~QsbP2aDq{M}$kiJXdB1o`CvV|FvNCk{%e=qlU}(>Jgb&Cd4nXW%HzL%*b+}CG-`?;NPO) zuXt}?e;Kz4SSHeWighhHBqYW_X^&Vf{?^-nyFBG@_Rra+jrkrm!EbS*lbq_9 z{){7}*$a~*>yIR`cHzBfi!`bN^XHJt=Fqp-t#(W=k0P(Fh(659_$=Efk%mkqFvyha z)7|>Z`do<_DH}|Oh4!Ja@UlUN2U_c^a~@?Pfl}VBl(#slSlb*kVCEK+AbKnCm2>N8 zs^2NHEhGTn>8Z13-sC@LY6e$aV(+EQhiMA0eU;CyLkz>+?` z_&D2#VXWG>Jv3+3DORhI@~I+k_xp1vkN7Hs0~{B^ApObZzy z)!o3&kjAF~BF4yAX2E6LroG8QGA3X{G5sR)JRg5lCi@K&71B71W?NgO;7ae=Iqhb_l_|I!`N^h()jj7A%Vit{xqjQKR zWcjNuG!4SOHkkJ+^$nw{Aq}lu)!Qrl789u@qm{EC7ahi`^+l-6&JL?U~euH^K!hY5BPbP_M2*$ENGcWnW zuQlLj0y^I`ytmymD1iwZX9+@D8$Ucv;pGp41WTrE>F|TQOIZQ&TLdp(x2NU=I-g7v zb4zURZFt7>l)6qfjOe#acg=gsWv^um-?L`@x)wKArLB(rxE}Jw@yR~>cZy4yBImwX zt;%taEa{+%S39OcMuVSX9rG|=?@xkU_FWbl9CdsfN$|QlLBZj>r#tUQ$lw$0?1^ZGP8H>-n_dw&Vn~Da>7lu<^tx`>nPUEAD5k|tavV{1~R=EU9N=81c zKt#tFF5Oc$gX*oIwI;>6kO?E#M9&kl`41BY_=>AouZU=}gZR&Zte0`c#n{whirKy* zgdPBHV67-N)WBM!kk^Rl6DMkV_GQcT8Ug9I$DaF}b@fecijt4tBY2W`+$@vO){=VT?6oNrlCRM_FZs@}0Bk zvJ2()Nm65_tmb1agm^0-``#rM35}{We^2Kb{+Z%|4F%tM8Jf3f%!V%Qm%Kk#pEEvM z7G$jIxaaWf(;CPEl!1%veglaScD3h!SABS}@XxfPt5V3NFJJ8mBvn2Dd-xCk9$m`G zH#HxeG^%5n_$n-4TK<9uFE|PiGaH)I{QM{fQ5k;iEx9JK9jb&+@x@ll>QAlbD!{6B z81LHArm{{5_h4GFX=;5wSpTv0QJchO{P>Ix#nN6Yy;-A*QKscJoAMR|y&a7);wUDR&{+eA}&AA&1Ze2WtaXZ!9h zytm2z$?yD9e)|A&)U1Q1C3D=z!BaSD^Ufa;mP6{@`%?}nydJ6UP1iSG%^I<1_9)1H-b^54VN_MwfBLRij>#r?C3sE-S?>AZ z$oP-8RP}&D?%D=qFCH{6N+10$M&f~}?ve^v7n(%Ja>czLOftySiuzcBQ`GkPd4##v zMrp>b!~YDfky{NOTaW0n_3$+nj<`InaxFf!<=tj^Z_Pz8+@e%QryDB!=^?c__s!Rs z{1;8|^9O(k=O#~x$?+~3Hb5gzGKGq ztGZQT59pCNJ8R;fX18>knLV-Ger^9ZeI5H7)Dzw4JerYdYVl7WHnmx;kY5zXY-|3t zoq`5XV3-8{sA5RL!3li;C`$}8U(rtj3r@s1J z|F!8r5VW=bVk>@Jc48*_4DU~c1J{<~osFq!`f3lLMkS)UZ=fXcaa{bCQC8S`?~jt) zvR&`H90hG~J}tO@ilcCPDN7hG9GdF0`-`d%jP}}paA}@3_MgPzb4$};Y|d#Knah(a zC+o>#1<{-Te8>GgZp;8OEp{-|mW=jO$ArJSPEa7Vw`wFN=|cuq+x=i5mJb(^+uvyQ zcA)B9QE=l5Bx85n6xD+~;X(XqpsRc+EHfIDBrbDQe+o##Fo0w0Ed`Lzk!hOWB(`yZ zFcqU%g^~@cGx_tsxQBeQX0|t{>(ams#BlG<)*;ix;|s_(uCIe_-n;sC;x@yayr?~_ zH=W#xF_4^o{$ae0!Z<43HsD;0zkPa+m`?C#wb!%GrzCc+v-R%g9y329t-ki=ip6CR zO+xoY^>P5wre2nQ2I=o?cP?~3QLZG@0-EDHOxm3 zpF{)nK)C3~8+^C7EcGt=SYzDFt^MH39P|N2d)uv?fyWD%$d;Y(Q;S1cNZW3-BZT{n zDmtS-v+E~dPeeMJ4Vs%dEau}Pqh5V<<@Ldg&+WG%8tSYDaC}F(ISZ{cK<8GTUm<%- zu(Mp(gIXMYeE7m}i*Ru1qlVkNY#uqWcM>{2@9Van8O;+kS=83piP;B!UsQn^;KIem zmF%ah<=ry=D9F~B`Okk?s)oUkzGzAJ!ounQ3F3U~I26P6$|4|8_>bx&iD3%RP;Bfr zp(@Vs(>shk3zl5oLb6aoycr4N6u}%i8Q5gEGMaRGM#vP?p+xFT@v7W}O9yz<8$Al5+zkQ;`92T^gcByRJiwfj z0{$WWg%J%5N-RCgYdUhfsEBA;)yM(@XDAQ&R;)82YWr9hJ`W@c+A6+t+>Ws4k5QR+ z>U~`NRSkfy+(JF9_63#+)8?50bScevzR{b<=hNoUvPMG2^mBpmL0o-FB;R$r9dqfhun+I-_N$f%_Ser|{=Go778<0|y}666Bq* zAMtf|D{@19CxmQtsTe z+-P&T4E1@?80dPqMb*OKHGn48Vr)Hz=~Ih0dmi=7zbO(R3>kg0ztGzEO_Y*8eP;oM zx7~|j{gsOM147-Dywl_=aq$O#r2aV&u~2FQObo}(CopbH3j#kn?k_fq0pekPr~QR` z=ulc|-qMWMzDGFGQF@8cf7h4~6#lYXbG0}fE1X%G@8|Ph_rR6p&G15ODz~2n-HrA1 zuvw9N`H-kS`eZwwFKj#B{>msd*lDq~c?5WCE_>bauPtEV zILIc^zRQlff zVMDc((TYCF)%s5*KRZzNj3>B>VIh2eKw@=$c-3$0dVe3hAX&ZvbSJaTp*oGXCn)}? z7?Mx8McaEq$BUsfdU=H@V@do4+)C5mCoKsl_uiEeulE4;-k9`E%7g+vY)ZqXra z7sYrX_BgV#fDaV`St97KnTFHX6E(8z2_Q5a)Ld$iv`O$K2t zdry7K7O9`@=Ii)Rvc(JUUTS-QDI0%VGo^v9IJIj{dM7gzWDmj%+Im0vL)S5Dn?Fmx z-ms8A5t!b(*piUVms$oW$ox$Tu~-d3lEj;c$_yXD_hru}<}K#zeKQ}!5pP2D8?~V{ zfUgx7!@I<*)#BvvLPF;6bsT$xL)at?v&G(X25TS9Dy~l_K4x%maPt}yJo-;{M9%_K zj0jcMlo))oyWNA`=;fw4Pb!h6t8|h>=E%ZE5pq7GcaX;q{j+ac_KBxPsSOZfj9HS( zzMVhlKtuI9shAAEy`Zvwg314mW2mpkm8wPP#vJfKAmW+RJsWDhvMbNyIh)kK?)RH6 zD|?{{UNxD|*v!9W>%O~LtTN#Ix*_Z5(}TS|f$?!EN7R1r9#ZhnN&#?nD1~X_+zqBu zE}lF@Qv~cHh4Pueu%P$d&-$Wzs^KG^ zJKScu*DLA9Jfl5YrmKv9AWSg5l9S3PM7-w*6DE;)9Mo{m$yi6{!+i{LW2msX@X5jW zkLFhyncVJ8`&VcGJv>;!tg?U0HpL<2dZIdpA^zqQgmbC$vTHOlfz zpsvE>YqjlZAcM|q5ek&X_=LHdjTOE?sLF5gQzXBDD+vSRtrB3tEZJR&ntpL{N=YMN zcl7_rd&{;q0(NZ|rxdGDihFUVxEG2QDems>P~4#um*P%wihFR1I~14TL4r#l*gJjK zT6^zj{e}IV1HvSenYr)lI?q@2R?5?BAC*gW6$a7_4cS%Toimf9f74`RDJD72@R{k^ zRpNKip@yTwdv0nguAcSd^QaNR_1S*&m26qYf5excR%-+rab<)IgUk}n92Ag-flJMP z@YP+sA5gKr7!>f(2iUV|@=dtt_|Mtq!963Pl1Orh{18Pa+{M4iDWhf~E&+T&Q&Eo{ z8kJhpKOJALbW%%qnFDp>)YNO+`{O3=K@Wj)cH`}1&vkcOTLDnmWPfp%(Q$z_p#hVm zS8(W1e-EqkUXSSWQ~iAGGlyfxvB&f_NG4dMfO##gb{p^_qI6|)+v?DKot#FypDVZC z)eFjb+6_i!Z4vdkA%VGShu*I`g8g2;b?nk_7y{zUW?VqW80IRV#U%2uzfJYiyi~7Y z#bNguN-;lPdYa^w;?O~@ohjKjhdt>XV7uVZa4r=FT#)A9VojJo!i*eGJ9X!%j-KNw z`^FZ)zDZ4@BbcA1gNBH*Xf&Ts(YQ=@tm>f=?^aMM)K_L#zpZii(=Hws3>eHQMD=C1 z1#smV9>dxQALkc)Nu^|c&o-rt!GQ}=r0se-PBdPYZzi7OL~@{Y2}@=1xb)4t_2t6M z@nGM2_FIM=z{dk(FOQ=DP7XC#$0Cka%I(8t`sa}wyN>JmDj*vuZA6mHv@M_EG)x|y zdj6L79}(knvMcSqfp-xQoXloC-sc=HB=Oj%v;aPn?$1y2FE8I4K#zhyrORz%dv7an z`TiNoYGU-^GdnBAkxpA~#GFW_7~S=|Z_o9s@=UPrN>+a9XxvX>>R^^3y3`x{fuzwN z(fdC8d@o5RMKKxpKyrAl&CZUcoofTOjz1&e`hwSo^UmfpLPH^T^Q^)2Su1E}l&{;( z{qogx1lXtIc`v&jW*(!dBl}o+apsm{3+-2U6yKjyRR6>HT-sHORGHyI0e%25ed^J+ zbB1cV8vc7QSnG0rs->l(X21r5=qBSqbs036Va9%Pbk*d34-B3YI~fQ#q>NwVAJCy{ z$H2p1BY$ygZc-j+H-N@*>G^r)4G`Z++Shq*hyRgUTk8%!)V?oJE3Jua{0-wgs%q|? zuoPCZBsLD|oOcXtJq?NguwAt`;@ruAEN1b}e|}&CN05o!`<8>t>MXOz@yB+emFUBi z?`wgrAg;zj%O}7cG+(T{@HPs%C9XvGO;4j|VAC$i=`&KS9RYv&x7G*x+Y_|*C#l!Z zV*vic^u9{!O!W@cC)UM+?4ct`G(EBDBe+BOWHL>ry}Vmmc8Sk7C2;La@S|n!q?;&R zkJqQg#>$SY!JVEb`?;;N4Y~?Uh1|{gXZgsqT||J#iy(bysw8C2+I1MP7g|-4rfjQ?s>*Rp?v>}kiTKP@Q5MRRjT znf~5?5OITfuxA#`IOg1ggM=I>eBN;2^(;|5iOH7P{D~0xF7K)5S`Qfo7YwNa`Lvz> zv1HLp^K}wI+TWBWZ}z)POBbOQ@m3mG_RgngiO%I?FbyoX8220B9M8aWu%pb}a=x$R z@Rf=t(%;vLLJrvkIyoGk;FVd78M{fvA>MyflvQU~N0_;fkuK=W^@Y}LcZVG%-|(G+ zbc5rOM-_r;c%L>QIf22nmB1&D#k1EZKE6g_`?)=q>OC= z&!*MVF=xaMZT*Bc)0cKHN~y9OfleCldkix3q~X_Lr_;AJ67m%(EUObd15_aQoPcxl z=9q<4{z6u-OAFK>Wg|T!NfCg_vzNrc)YKt$XR-N-*8of^+S9&V-F3IfSnIq}$Yj@E zTh@BI9`^VN?L$xbQ#G!eXhgn09QcMab?ChI;C-4ke*nHbF^_Do4}CKFaP09Oip9vx zk`VtWl27a-8cY&7gOPF5>2m#foY8lH>C@Pn@CjF;bk{DHkz7r_sZ$=8wiOVCFV!*E z^rYba`suA#*w&FbkD{1ClH~748-Til!=(52ZUR^ICNv2|plyaud%gt8%^b}2U0yZC zMiD|Axa7>|j^@Kz)>*t;QkhQA$g1-Jm$|S;`H&D;FJPFu;eYrPf?C{?9%_6_v zB~YGZ*<%kL=V9V|@OyOE1TrwCylA+Q<0<5PRyBqB!;D`5^ zbV*d*dV}UVc21X37$=-xlkI#HkOhwN=zCtX^S<`#S(-9u>ER)zkX4c%~)1{jA{N8JI3F&_Rrn)Yp4#M;>Yd@l83#o_4K8*@51XV55CNc!#xxNN5mR!jxBjG ze19%xj`{g+IzJg8eXMM5UiC=zs)Aka?P?mhiknUv%vEI0drX8bJOiYyQ{SO~Yww<3 zn%kAxNugA}^?A2>(DXkN;aHCF3gBBCbl(VjZ?KOe%ZB;YV)#>&eHVD6$v74cG9>*BUD*xNM%gOW;_E-Mc%%rd-$oTw>5!!9&3U_ zK|L~w)a`&Nw*Oj;@Qf4--nQCFR%Q;-Ys+N@GB(+6BoK zV6Sl6L;Y&9Mc%fD1JdUDBs8Z2HocaqY}9U8x3j3F}3?leG%Rd)qdA!t z?eGlZTqqg!(B^w?Bw9L;11L#?L5o-Cvwtly8>vY8nB4@GwaL0-n z%pu?PYJbbF3}8mJoe9>;VAc3t%r7>tWtr6W?+!P~;7Hgl=J_YM5)1J$TPN?B8sy(n z=9MLk0O!c(y>8&(r)yTe>1`EGf?kRR8FS?6>A5lcnhf%6XSzre`~Y-$n)#-464)oC z9*gU9Yi45{sT^&ccYDP3jikg6F{DP(+DNRMcOe||U=Fr)(M`+X*U~&VkKI~@imVG| z2EV;)`Fo9PlS{W-&-Jx^_rd%gv9I|qx3PDlp8N)5>rtBwJ=V(-t=H;4C**02?nMap zwN##MMX_Z7*Tf8faJhoo;(oSL@IUPZ(&BFegqA2>?+SK&ZO!3>8a%J!frJmG@3&c6 z3lPP(Y#c0{odKA~cnqeUS~D7RM3>b05!r>g#m3jIuV0%osUZoes>eE$dUVQ#Y__%f znj(!Zlj!LQ(`FyJO9PxX;W;=v)Rrgt2&Sv&UY&~W$H+GP60di8G0d*fnw-f3m7?=} z4|~IgItfpF6dR7~Cp9V~D?!j-JD@`88b!4_Kx#|}&t5tReUR*=_L2RxiNzz=^io+x zs{&|1`|~cuiomUGnde_Vsbm@Vg{MbqOjbES#8X-IgTX(zNv|loE`#Nj%rIT%M)T#d zbul`aMt~lxKu3c9N-}yrRN?PaWH59Kl$XnjO@f)VxyYH!awr}rb-HKAV~0{QsLemt zmKb~b`(zw;1L&%5fcY2x^wC1|f`sP)nIv0;xyIhMq}h01(B(nNY^5F90W?>+aG(X{ zmlG9{H=p1Te~pk$A71Lhius4?#|&P85f}_6aFql#efH?|c2cS9G}(40LkH=0H9z+c z{u>}=PsDJ!3{7paa%Gx{!Qc9X!^kX-9di%H)A9@f1DzQx$*hxhBUY0Dm<5;?J`^2HR3N6l7#wEAiZAI}r=zl(O`o=H7Z<|XV z$7dCxwU+Ikb9p=U+Gc)rP^8dt^&craay}fna}+tb>7TxEwf<6CHdaU{C~LtmM;0Wy_6t$fywnZRU=_n4Pm)=F>2kYWHFT|XyKmX+O7WK-aV&`YbU zz!EE>=lK(PtrTT*S$VIk1-91NoTh``?R^iqfcKczKL7^ExqsD+)aqY>V6M(9!ORnP zGVI!J{eca0Ai5Fx+*AeUBk`?6rSLkB$r#5;?d(hsj7}@Cs+(wQFGb&xcQ_Au1AK{y z<6)EFCauD0`6)$fd*Q5r-G%4U$@tSR z0h_78q|ml+R)yhJiJt|gh+8auy>d+B_}dn}2O-Ykcukz0q7TNmi-fbeE8H3UJ$iTD zBBn?J$Y26!ENeIK5SKI$;6rS28pM69Js)@tUSmV;6wFzo@mO50jbrWRD=-EnLW3Vi zvdlb5k3=!nr@8Jm(Ecale_x5o+CouikOsid6S{GbURc`bNtCiL_D&0q{3P*`6f7y_ zB=pt4=L5q26qO?NA#w0T`U7Sc*S!gMS=v}#5=lSh1J*4YCW>FvI{ZkRx+cLj zWty#sl4OzLD^jN`TFA2fX_Z zhZs#cnr7q-Hygp@)#UiDwa>vk9O?*AKujR==5H5+SdRTDB@YY-M9v=h>9dTG%olG# zxuRF6ryhEFX)e#xUa}HFpD4}S6YpSy7oyM>zCSThF+7Y!ize+`Q#(FRU80N5e_FQh zLI;VX02cZXiCGo7-`|A2SGIq`k<$PMc|!UOibwp>(5Jc|?9l}@s2&J|uX1}HXg83C z1cR4oRBCbr0dZoR7r08PGNSVH5ZXwphsX(mD^TQW$-P@orzkVc(a+T4Wq0%aicUWn z`T9*WPw3L$Lwo`)rujeMY^a1rE4QF+DO^o zg73SRM3b$042s;ihrgpdX?hw(Ob&{ND>QvOcvFI2=<15gXS6SBht-MRgxN$G03Ke> zi_m&}$_qKfKNoev<;zKVZe4w_KlYnvv$M$3TeROt(3@RhC)XxMlYR~4$KPa=)6o`X zKBnW6k=XF*I7e^F^jS-)P#*E$yj%RQ^*~5Qh+tkkj=hi(VobBiRwr8|wxij~Gr)HtI zofA742%6Vxq?%CA#E9tybc`)k2hOyPP6%dp*g|&?SKkus&o@T#KP{S$^`OtWE{wdt zJ|wZaX+~p63LoPn>6KRm6O=-A%3L=PS*g~J|F9OX$m&CPSpC@u*nA=NlYHaZ)O}Qw zl16XxZ(W5$sd|GaKTSKXS4Q#|t4gar=Y<4U{0R};>0HpJyU+iFq}J>}>biGO(uNu( z@Z?K}|IJ7o7t9oe0j>b(LKFPpCWVT{H372FIUqD@LbsF8dbamo2tWK38rqZyz8-V{ z2EQE8Z=K0zg}L;e666#+PJBHVy8-xj1}Js6213**azGe89U02SWd9X?i2f=ZCHBV| zB%C-L_fxtL?MkroBEVyGwP{WoQ=Si$@eKb&fUk@nPV+*`U8E|B_=qKg*YzcW6s271 zGb;fP^WUpuV4i9j%MYh_9XKfAi=Y8T%=AEbZT@y=xtUmc;-gNcq9&e(s&rp#m+m{_ z<0VHG3N&~Y;cq3s_ApJb&%thgrV)%g5Uid`*yUuRFZ;?DFd;Oa^>-J=GK(YNmQ~0$ z;JNc$Fbl0c!g;M!9JvH46(T^k_p0bnao@KxbCer;|5dMV8+-d5YEfBsSM+!+soAntkpYq9P_T|DKeI{Kmody20MT65Mx*2is?;n-)0B9A>Z+i4cX-B*QlgtbgpDi=S zn^-4ib0jbMGeOOc0KW=bfU@4Cec-stR>${t(S*h-|I60u#SJbDTFxc^TsigBqqyO| zh_Rm)$RbbmzR8(ZOS3Lvq1}QQWMBqzNU&aOQm~yZ+g6^aDu{l|LvT8n(ATjsW}HpQ zK9E03!x;KS%BO~W$V2pcCn+m7&Oc@xI!Waedr;lQK=OX03UsMBaZ?zL3punW+NK!L z=)UO-KX&PLX7w5uymq3xvwbtQ$=}c@;18fY+;83<`7}LJO0wFu?!H`V$#3NEFf97T zt~eZAVk2X1WcdQ(wORRDqhOQ_*!wx0zy}`q_V#1N1xJQC4L#Cmo4%U z5)_=mNs(@KG(OM-P-?dO_Th#JQDfkFiGwLZra*C`RZ09sE0HJH#u4r$i*Iky_N9ZR zQC6pCz5JnpR~O&Aaa|I`P!Mey%g*#Yz<&MOy|V08FL-ue4zGUjBCmtpnhLf$PXqf- z%gLH~R_*qdp+_U2XPmJ3!|;Vq*%!$6+M_wLX_3)b@+xqvBh9%sVp2bTj#0dTs3llC z9b4}=;2oNq0gDeEgioHRolM_XEAc(?X%0;0ewkQYSw19M+VEU3L+S{#wE!&}rxqkl z$*j_zg(@P^m|u)dE2U8ny|UAu-?*7qj6JlC)|}`SJQ|yoUjX_Xwrrhz1FtqTFSoxp zJ;85Pn*V`!`~Q}yu6!_l<@Tk#x3RcR<8%>`D#UD{?lwU2EpGl;;6M;)bOJRf?F`1>u5AhsiYet$vKMF> zqELfv1yvglt0rngzJgt%Sk@#rbok1LJ9~}S{}BOHC<(T(h)2wv$q5kU1*irI(59nT z-CDwmW@1s_F0hJ$jy3x2jKJg_(nJ$1C?8{~NKBv_XDi2B8$Pf0#ququCF9SdTkyfD zVbKWh#j{E&Siha#u$qj==B?Qq+N&79h4~eNqZWV>EjgXmtJo~_q~G}|7@i(U`q0Nq z*@beoGAKz5Gt{ii2+uhAm#abDQTQ5fv}{!m*Pr1u){x9jn)~Ih*z?GZKQb5Kq8vYs z2^>(`HKy9GWPiv5AJ)R!#UxpnCapeTvs_tL1z=} zb+w)#DF3? z`tqMY2lL+(uo1TDQ^75%-9o24WIj~9`1QAZD7H)d-N6+ys0o^6u|RSPd8+W0?Mk?Q)Ds-aPFqW*|B4aImglh z#4>ayYm!>qxCiQW6pf2EZB;Q>t!n9_IN zFus&D|H>2dhyQA*v@;4hT0eQ;ZAkCUFz1I4#;)hWd?D}rl8y?IO`0Vs(?TCZS-<z*tF--Q^iW(*O! z0FC^JLe7=ZBe4JT_`$={i0yo&`-1+YdfwUNL=M=72!SdNt+am7vWp0F~Z@@;$W z50u8=G4;8PUAi9~53{~KU<tch?k?@pnyEdt>CM47KJJ`!JIP=Sx zA5twO+b3nN<4D`@N^C(U#iYbk+}7dvR2ciiK)m2BBfWm?aZEKMmIpf`_1?RSj70hU z&2*CEp#V9xEyqpJ_9^8sC%FiT62QiU4Ry?y8jKI~HdFiG<%V4ZH-?Z31FLjrj1l#6 zC=nB6Tg+khd8nV|eA+H#mHDtRU%lS6$+#~>{}sZ0@LtH8@?!QRS8=t?-15S2a+<0= zB*EwDK^Y1dIN$G{r+*&I_l@O<9iy=-wESn%JLq+bW&n$s6>!9x}1 zKt%_zCW**jKm7&nbB>|4cDHj?{n4b1qUWUhU*K!GuTxP#r74d!V{3w)hIX`W3G(7^ z$|rw4DexO|i*V*3+j;_?WijfY8Ug>0g6|i5LhVOA&HU)5TlzQ0^+czuZ&j9lM+x;! zE*E^}@?4|u(iV)-{|L?7)X7qg$>m5|8@t`Er1$87L+0AlFB^_X;W$6@(#UW9(=H8~ z3u+IHpLHtd8s%+msmWIkta#(Le5WQnEqBSQe;Z#vta+}C4|6{(Jb8DX6AirCev$48 z@Due6%`Vnkq zdFoKr_2m8jA)Vd$)0r;tLj`ob1H|-Xg8PnT zIcaJg_NVfT0m%wLw3sPXrsnd0_$G03^?#)x9>+V+4xQuKQ9-5z=yt$HcKrGo#CU7# zO%J-0uU@Z`B-{`p|NHz7c3OvE>Fy5j{x|4MtJ)vlq+Ut7G`{-u4Nwc(3V5DC6F>8T zBSMoCh_|~M2mQygl}O2fm=2ZS2N!NHHKp3w};4*@qG@*^bcQB zTENa)-m|nZYh3_e98dax{Y3yGNl`z_DhLXwzqbH}CmA`oLS_yxC)NqMJoJo-Io(>k z4$~p4twKO-c5W<(Yk}GDsYLLjKiCh(0jRi~7Xa~B?G@PuE}ej$#5c3(M!9YF34E6Ex;O5QuRvBKEB80wbyO__z$@U3I{0Z zTmPxEqURPoZBOC{vLSn++i~0xA8m~Ojjo_NUhnHq+KzynJbW>l6sBl=|LuZqZvpJj zt?StFv#ie6eD+^5EKQMge0Z%bV<3W*NUs)J*T0vU8N<6we=7^@Y@+`2u(o&K@mXeh z{AjHnT-*Ecg^dRo9NYb4l>@uvU-veDnl6SF{3gjIs2RTpLF9WsYx_OTU2ZUDT=}kQl>#0`P)&|M@NgO!WW1nZ$ma z0iu$$Tz~jKj4Qa&5uf#Az{9!pnI>>n{nv4>RQzW=NdBe5a0v{G+XlKmJHLcmeHQ(V z9Xj|Aw9gHgKWyOF0c#(v*OwdXCxyYjLSAxbnhCgCbPfNELjQh`2Cs1!=`+6hKhqyQ z>Lnz=`sBblm%q~b%WR!Jx+_54$}}HMK@E79lM3&&5%4_TgCC5!J`q0vIV4~UkpV!? zCe}`svtxO;=@eCpL9Ytr&$pFC)0bTL&_oF@?;rrHSoGo51I-1{4)n8|;JX)f_?>r6 z&i>Vp`M*i`yFwR>NyW3ybbO#4(dUjKpw(e>_P&<(A6%g*aONYIrJph29I(B-UI3@{ zN4?JsNOso^<#*l70CFl^s>Sl|k5_A+VAKGHu79X~3m{bB4int2pHzV9I`Do;9nq!> zr59udLN0_NE^dJAAqLne0pwnWTiKluK;yJ7pK;Zx%VgE{0I#@%kRSZ_;r*^Y`nPO$ ztl@y`<%0(}!0}1%$@1UJs#j#hy{`c5QWiSR4)*icm}r1wh2zBXvMg!m5wX{w3>(|=2AHG%y(a>PIfO!yU` z;1KayV2>gC%j*NJ^*PI2)o$1q1i0WBdmXqUZlb4kmx{#zr}$rH&Q%nW$zeJOh_H^$ z2Ra_B%D(0YmuW)8X1CV`#{oCQ+xAb<1pFgH@LN?@F0bZtb@|kZ!E~-kK9*Zsc-zVJ?;MlMkz6US>$8&s6tO0L4HIUa3%@hxbNJA(0{)_s&kxoa>4RDBU zdDh%iVVBE_5jdn{dEy~szz)#L`vyw*&_urm^5k7G4Cp8`?CSil>)UIKd{RSGs8-g6 z^`2b@fRy-4VmCxY@5+L=T=%T^Ngw!0$}&Gnaxg6WLv)(X!XM9dGJ=aMcJDV?fNmnOX1y7+*9e`Oy0Z^*Q z_x})V$-V5q#dvFnCbdQg4M zpDnLkhzN)a#N~RL?K63&*lb6ZdWtGK0R$SUNs-S;^#hK%;UH(XmaW4hvzZ+R_&w@7 zrlAnn+VQ;pyX)EnIN(NA67Yc1Tee&;1Nu0cK}vYStYFdQw_BJ*xRE08mhJvZQdFTM z6}&by@Bg8G0Z6q*5{pLN1dl37R7$@F(RAejvN!dgV%3xPBU8qG)YZQYyQ=Vnm zSR?R8UT~sz{oc@4|CcQBr;;7dUeK8_iU*(L8rVy7ASSo8E7$jeK_%appMQbjin<4W z+Lelmi9Z4(_K%OAI1_qq*PCgOr5^S@EZcm(b(HuTLWq62#t#Qhn3%Mh(MPM8sE3=QMxWB{8Da{N!_2{)}*NxvL5HhgyP~e$Nr6Qc9qCu4PyH$u>Vc z)K}^?>KN90o*G!{y-%%wqy6G(VSlzHu2QE_E`)~$?T^XD&M>EOFUoMXV6)rNJj+An zYqMI=eAnS~<18sP)q{gRvCs1czom1g zM5#LH0tjl$R_V0@KGS!cLxf`z;bdCX$*k%s*HDI%tT+Avy2a6Wu63d;H72`^g#3)j6bL92Kk!(0t~B?~ zxTNI7CV>@mX4|^eLREG$i!RSYY6vBR{XjIKN`sa10Wie4)!w%1^KA4u_T=%pV8GK-iNMn~S?&m)Cu;%+ z$m5a0GS{yOH5z-H-NlF98;HI~^WcHMok%fx*f*QvE@cjR|}W4+~;BI*0`_RR~{GaZ5QkEMo6ilUEh z*OkJv&@$RylTv&#>R^f z4F|`WohemCkq_!`#6ew;RahiKl6(Jdvm!+nYfVc4A&+bvvat8BZ_kBfhsay50nU(I zSB+mXo05~FvaOjs_8OT@xhpbPyNy0K&}lI>xh~%ADMCK0@3EN(iB>NxbOytz2LUc# zQNZ(4TXDy6(@KNZ#@8ARJ81%{9Re2F+Hb?0cXeQdbarQzTfzj+qA%7dbCF}iC0Weo zqsjAuFE+9UWnD|5%Z*>=^C{kw4(S0&3Zoxcl2V63!L;3Y#KFBsGkTkj%1WQs}<2{#Tx^Jc@RQcZ}UilkfxmF0)>P zjBX`j5J|K-<2#J~Zx_4c)N>W;*rIO;Ib0gh{D;+eqyL)R$C6(7Y>*izDDFx+T`F#g zKCix;SkIt5ZHMJw?cbFFuH5UIohoO$Ti}kv4A8sTps2nhtdx<*meJ5^@@;*JjdhsW z|1^wWB_y@}+45Z-8$r~0@6&cmzxWpA_l89l2i4w-cT9gNo{pQpO3?+4y!oKGDrecp zm(VTJTQ+wTC2~f>G`wSsyxYNJ1e2F5E<`nl4-VBD=y86C4n*W`8clo~D zKv?Hh)}8g2ncufdmcd^iJK;K$7)Fj(jb*6rSk&6gHLlS;&ceXXn2^`FK!lcy$FQ^A zW&3);#_V1)ygUW4fXWoh=XXyi={E80c2drGQx?TXl960tnRV)b@=ejB-aPx^GwG;t zNvb3PcMGHP<7b4FqIRpLOC%-PDcUv*0>P9`~tw06L>Sidv&UriT z#QR=003yJuTHLS>5--);ft#N1X?Ixs)+i?&NI;UvckT?s(mV{W)__D#2H~Y`fB+D0K*CO%a{h4~i*zj1 zTi%ToIDzZh&*ud%q+p(A$8FaR;}1%=Yg-ZAUeDF&Bh{O~N+=%e=b=kPPyquY!zo8{ zzg5-8pg{7l%lW%Hobm;%J{clRD2c=?liYywqG(c4aq{{XC2bp#(&UNW*N0hWKWRz8 z=A_GCQfBKVN0=Bu*&iAH&TK_&W7@7)9B6iPBR_Y8d`@Tj^B{n3D@lU|r{SI|egr8P8lT+j_r)Wj+9i*uKrekJ zk&>|n*{spF?X=E+g>@`)KpVRlT&OYm18-ju9ph{~q1be*nm={At zX7XV#cXyg}9TP2cyyZo{m4FUeo5T%aNyODh#m8=G^&vQg1*j1l`IWmfNNTBEA@4j+ zR`ZXu6D5$(CvJwlZk8q{M#^2s!6k8fVf2tv>+l&F4dKq0yv*1|iDf&dhvV(l`em*FP^-xs+_?Gb5P! zb?QlRF@3ieFSm=w69R=0N!+ZjNQz08>gm!DTkMuDWCR;_7F$hAnu5RgtH4AWQle7x z`=q3Fc%1xZgJ;>Jw~0?&y`Sl{y3EHJCq-iFk3L&_D%+On6uBS?f9@_YjkviOONx`d z|FSGsr3X2@^7ZGrRlegcF90^pP z13dca!7=cM7sAF+nbzZ!%#YudR7xY#xqvHP5)LE3Z*`U0SP_gth4a}zLMN-MHU>U`*GJ^m`k z{%{AyYIsb~POXl@3{{zPVfdzWKGp-!Jg^G)a?HWnPYEXNapeFUy3K_tJL)oQ@0{<3vg z(~O!u%C2{50bNKFBJH=w8O+HM7Tz_ABDs{QmpKhsgSu^M46Riwb1wGH6;zR`SrvDf z_(MSXZrg!!QUHDw6!Nh&L_u12Hf22X;1(HovRbQE#j=|0i&ejL|6CrW#I;sFFfEuV zc4sQv@7(@-MNBvl8T&gpj`M|fumF~{Zk=11NOAcu=RP!-adl0aA^BUjz+ML+Y#F>= z+(i$e|A1ndgYYBI_wK?5fk>J>oD}EQA|AT@E@MR;f8wAdyQ+(v73Eaq;r1gTwHUKRJno_`9GL#%yfRIj23-rktH7-U2Z%w({0HnLM){E{HFTT zmcSn`#WT7j-Mn5Ym{%q*d5Z@2wv|MfS%|m$2c_;;y&TVw+(BvWPG@Kuhigic8|COs*jvQ3mK z7yh!IE%uhE(g$f=Ktzr9pBK`2{EFS$9!Acn6L@~>4$WzL{B$uNJv7Z4QWCyz2ec`& zK7A6G777i*iv2HlubtapwHmg%{Ipwbk-dSgy|vCpGHahM&a>L=CAy164gcw=A3ocv zT|ba0YUs~@w<8bNQ+a?NRcEY?pL?gRIjwuO_Qt6U{C);~nZ%HYFRLq4uLy5{R)n=+ zH^M}2WxV~?r6<{buy!J)l!#>%tBk=ShM}SQe%-^^-UVB)Bk`0S z4R+JaJx2R~(%7x*d}l7X(uH<){rt>Jq78knKaUGV?oH~0+`1d985wn(8lIkkd01K8 zTT{>qECGn9KCbQHop)t8i{Ejif0ONj{5EV}_fs0Z;uwTh^G}R=oxr;}29c=9Bm_RZ zVlP`b>z2=qT4_pPYV+{8*Qb4uw@a(;rjUeJ`AlE!Y{$tgux4L*R>cvXwE3MEX_{x> zW5v=r=B%EG*fQt~7}7#^#>$@GW5Sbh33>DY9!jIkq)YgPzR&ni9hUdFz*z7n&|;8U z`7U;CqrRJuelnNe8QIeaTaBt70@?{5H`F`zv|HijRYX${Y_o0dwuGE!-&X-8>IHCg zBXSjnRF#B304<(UwF(V|Vx?R%|M?fdBkYTVJ8MV%KN2UGlc}=X(jPK+B^#Hc*t%$E zEkn)tgrjo-Aqa4_kwfyTahbCrRgCG&hP6K}SL^~{?YMpNr%rp_(qL>NZo9nGe!Uc} zn)t!AzS_yka*BqpW4ld&u)g1XHEd7c-U&8u_461-n@PR06j{?6UDlk$+j;#VY}^2tM&Ez3;C(l#oqX_ zNLNZzZ*&A*j<*IZ`!w}el7#}%=&X_m{*&#e_>efEn_mWBvgq6hFXDORB6|W?Iv+b- zTyE#ixC)Nx2y#R`*M0ru?_=oD!igl|=uh z-j}!PT2AZqR^;nPDtV`nJ@G8Mm82jV_ z&j_~u&}uzJsXY(dt#5C7wG7|p5pnVSm75S8n-ICTyIa!_S){65>S#x+n{v(esj@Ih zW!C8YdY`UW@NPNPaM9dFnh zn#6y(*V;I$Q<6J-a6Qv*y)CorYGUW!=9U7wN#WCFRxOTRw!k1xDsT9u)n=FO+kE2F zVc>a!b@0m01Zp=Syg;&$>3o&9{V9R!vs!En3BO9lN9#9FUq0E)S26$r+#~>%q(|n% zd;>q|!g49{8)R+sm!@^Xt#xQWhdY#cE7Xhh^P*JV<@&)f?9T}&AA$c%udtYN?Q) z`CK_J>%ZYXHh)(znL2Y>zrOj5=e0eT`b#{-JF$Psy=Z&4@yz%LzfJ=3Y30m%^%gN6d3<;?|1mzj zmg<*HU8RCZh6fEGcM=l#wrOPp|g&EJjnfN5y;XR&OIOdZfuKF0?k15$4eQf8fV{ z*|&GX)PonO@NbINf4(ecjK@L6A30hL#(>O$9eR=}C8X%YMk=Vju(n9tI6fnSTl`+| z^KkI(_r=tcLI^NT_HLX`AQQTI_38ladn+~|7_Ir?iA|?nD?1r8%LD<1FP11PEwUC! z2AcN-huK`!eB}dUIH34e;yZOc-+&nTvwr!KA(p=R@S`6K2~}LiGvHoWJr?Si!dy~D z`7>=Bon6ccVe^WJ(>HW1TNEsuQ~Be^q(Qq`WTo>0IwAq5pyGJelHX}M?#PG;o>Pr~ zvqQG;Zs%L2x0YY$E0tM#&W^Rt9=B=Cl(#5vgL{4;yX}^%+mNstv`emJ)XJDp*a6ep zY!!KdyfJr4gd}LZzA+t}=6vf!jfC|F^r*tgwcr;JYD_eeg|WT@PcM~){Z7C8Z09!p zcVrXhbxCQkkl=f^SwIz&tu8qg{wGiim}Y6y%BOAMhqe~!Hal!u#!xV`TU!Sv5w*}c zC4P8cX|8;C+(6idfNEy{*w)C+j@O<;r^|i|3^|b78!?kG>+hPwW$Pd<>QIq&kp9XRzs)I4`_<3ktdJq z`%g_|nDImLA`;%|J2w~+uX4Ga&f>qX7OCK`$?UE&(A94UY3{C17zU_s+b=g9c*_eQ zJFcR*59KMn=T9GY(sYw!_9H^MyI6dybOcdFlt3f;O=b07?tZA#@MhtXfBhiO+BiM> zNt|r^k3_hN-y6H!kRb7j;5cvJ^2^bEIx@TbKwqq!FJ0%NyQ4p4FND0Uz+&<7mk4GO zP3{9m(^O&C&L3&M)Zm|Ay4-f!%vCzahj0wFT_0U#^vZspFpF{XYqrgD$b9QX?xq{& zV!pU>mm1zX-Vb*4^I&S zZop){gcFgl+wxMBuUgCV?Z?5^1O`)DcNA3j>t;nI;p?%>1iwoRjKDDoDR|c%w)W>5 zu@!8=p+8a)B9E(4iXt&H&polNHR2s85kM^rs`43_)>i#l}a^n9glrYe}>_P{_Bk3X`&@R$dI>P z-mfghQO(M)y7i~2h6i3$6dEOp_Kn#|!7%c(YccGGal;y!5dYu0jo+8ZaP`Bt<7|^9 zKkm82=JF{oa6oxwDJu=FG)K5cVP4)TzpB-#ue)=nfJuMH;Qu@6hfV(Xq#yaKU$1vF z?7bg#kH&n4nw?5mdd3F&IjND%YChF>ZWEVC$LWnfr*rpFXN6Nw1Zr?jxBQ=6{;#lX|*wm{KpLi0d96Ig>%*<_G?c_DX;$iiB zW6l>YKdP?10M>p-RA=wR_G&>uV&e6Fdovg)zoI0vu%52Uf(Y_152UzE{#rGfcZg8d zM*pBcW!uT{Isd6SH?VPfD~;il$6JFP9!F5dv!P{)42Ky_yyBs?#h*}b=jy%dgpB91 z^{w6)oYE;~C2sg(s2Dpsvru+vfwNMNggWpOgO4Koklr94RK8;J$~{VS6{PZgYHi|k zs6IM}bLVhx#@E+x9Z29bNw`>>CRTRU={Bl~Zv+l{b*Z>a&E~gY)7^UotY9y* z^RByj&`kC7g3XJ9p<(>11?d4SLUnCya(*2$-yLUewXeSGQz9fU@TN!TUi2i@PYAU0 zgpe#4k+>+PZ;^;`s8#4_X8=C0=&3(U7DDvSb(q@U@A}_i7T+lzJfB3L?`*(dSjgGl z`1ijY7Wg}aHlE7sNMj{+L$(wllKF={7L_mZ=?)^+h_t{EHN1aBsJ&euyQ)vo#ZTH} zl1f1bAJmNg=I;Wtt`WxP%<_3ZIBlvV!s6I**@;EODgWM^8QT6}PW@fjbkFXedu2D0 zE;m6*fM%YLf{0#p8}}Z);LIm>V*OC86aw|U4@sIDRSjkxiWM=uQEeS7 zAQtXxy%|77B#1_IyRWG3Dr<}IyUv>&;vC>J&n*Z!<3oS+lCNofo^*i+QRtdD#lB}V z$Pfc>T77ttO+XM#u+?Krt42yZ!WxSjL_m_zjnt(+x|bLxmOk76r1!8>6*6r)cyL;~ zkwKM|Il_2KV1%8h^)(`TkaA*!BI;rZVmKqG8?1R{1A|01BBQ-oz*$%W*!R2hN*&zD ztowQ~Xd$^qubhNmL5a3+Iu_phH7333(4?_#8#}gb zTaDGAVUxzT*~X2{9ozQzyyrTd<-5&)Tq>UI7 zk=QQ0rHVPAkz=v}ZjxT>F8@^zlvAJO!LD?nK!cH=TP4DXaeQY zPHU4?2~RI1mY5H?5A=#O{pQ8Z!dX< zZOtGRX+i`JMTo?F`>)5aQf2I>QKdI%A3d90U2?11S3`+F4M1nAqNoF`NUr##L%wR5 zDOk@)DiQHa2u*(=E6A+9`|WuwlJsG2x4%%t?|xD`?Zz*id$JT2mD()qo>V6xTXHCI zSQrxy(scn=X5@J;hy&6VdI6!X_ecTa?5ZR;DZtRb>I7z@rDfeECtYrlpPOm(@I79W zG7;Sb&l+}N_K_~+oj7}20`utWjUn~Ia~9ae#g(xv zk%m+uarSUET#*ziI!J?FYr$LQOJ>!T!i!)br%^Y2SRxlDKMNN=C5dwt@-+9MPG52cs{+pafMFe*xX_r zpo?ZDw-Q!&NP-%zihqNzf>i#PWZ}>t;QSgw2=7;vPAQX(F5DcZ5)Lwj0WSrvBMM|Y zA<#0@4XfYVuXs7G9E08Ne_u!UZKg9DmWe{Kwoj6ovbQBa7a1AR$GddHkN)%r{Z&9C6W5DLpjNCv+q&b;@ly)D;c^3 z8T&Jglv4ou-0<0bg=ACt^ZQ>b3{Oz$3>^HJswXnHWJKFrAtqcTVcNIhabSVTyR0f_ zC{=;O%rIS|j*xVvpxfj{W)%+|J4sds(WOB4tDbFTIJOv;Sdu6&xKgC#3V)cNIbnpR z7_->nY^u#fFnN=fjAeV$)TvMdrk#VNbob8)Ja-R52rY+L?9JRS9W|7x#}kYDQ)P2m z6C=jATIa%d#SloH{o$by*qf6V$Pviw&_BX(h3Wpn3I5bx6C8Lb_&(V$%+CS8GiQ$y zhf1=4ogqk}y+C{Lwru0T z9LhQ`dQ9KYL&HlF@ChhCm_H#UB2RDZl3oSjNrE7x`nP|}?dAEyFJ%0dOP^cdE<-`& zRu|~JIft*O(F>01G+Q6*I=}D*`I2*q+GmLMxq#HwG{RovxR$NcJD6X%@o#>o;3Gqq za9DJ|K80~+`=PcEt(sLPB^jE%zhB=lD-M!Q*`MZZay1*7({Um{GXj$cs2h}9c50f~ z3c-pJrwtjV5+aKkA(D)IoGgS&`%!#oT_0olFRKcF{FS*zNYm@(;6~72W9V>Yb;M_v zSH>vdV+2o2=i=0&qy0%mBkfMxes3-W2 zaufSX-a})v40jhRpf&DMl?C5<*ihZmz;E(II~bxof{OoT6pY|jZeIOecYgV2gXSfW zO?2qRE2Z3);HYt?Qc^*$RZf@Q*g;3ZwrVgZn}Dw$)HyWq@*2}!|DRA~qOA%Z$l0Bk z)^+Tuf&4yo5Vm;t?Q}ys5k>?G29o%6vva}#@pv|yXL8{I;l)oN;I3_khngbP7Ryx7tqJr11!B) zZrlbwvj|C|V+Ea(_Es*z3 zQ@T4GtZo9@=By6CT}}6G2>)ZRzCwN+u74U@u$o7eMbFGcYgaqk)Y?;WoV=-r&up|Au@J^Z3M$vJjbZ@puabj3ano$}eZQFt!AF7oYo^`Rn)UmEpRD4wzx z-_<)^k=4v7Fo%*Ye-rl__^n3HTGon!zoJDfoVK~TV<0n*$^;papuD)9Fmc&1Xrc8L zEE)umyL^3%prj*V`Rv60%;*LR(cfq@4Y?Cxbo%jwSxa$`D6wr&KJ*XS)PXy4O_f_8 ztj&^&fZx7O!2-0IS&DZ$9?Y{=r8c(AYpG#PJa3d~uxZPt7IdR5rWGHkLDC(6tGWqL z(HUHFwb`z6_?fLg3YR2-ykl5th z>sC8Yw^#`ym*kt`f{p-{c1d6Ev^upN@8WFu z{o(j2$=4x>|FzeDVu5c8H9~g#LFhQiOtmj5tnwP5BYpj~DDBp^_3U7C;;Q9bv!fNmY7G4w=SL82Ra)A{>*(?%OZ2vargz z5Miix4!RPVkTVGu+n2n-}9B3f-RFd47Jptsr(ow zzZ}b7q~5_Lk>&9*Mafp;Ya-qyV1YPk54epYzrIGAJnq~p8stoabYtNjV zZs^AuEciBjM7yHxTT4#j)GThYjy5&xdhjpBGS^=Pn!WcK-1l-Z-4r3@(JVz8Abciy z%0niTe2%UpbknJcPKsLTo00Gd#>%?s;ZGsj(F6XgpW}y-@3)H~ZWlmHMra6lgL`Tn zDp~1=-GL|&)f8=_Pn%c4Vm2f7pMH?T?hIIBj(Kv>GW;FyN`%2}^g6f`18SL*D5x2| z%IsBt_8O*V02N6;s+4|qI+*w~wie7Pl+8v`{*Al1C=A4(7)(@YAeu2V%g&w8NHEGB z_;MEHnv7g|C~|*yvXFFM)AI}I1(kTkwwa_4l5-+uGwODBv*C@}o}&Us6mGq4Ojr)_ z)~Q$_(-8^qnt$4LOjn-JLGB zcj5?VeQ|M-eGLba{H7L(|CvZm7To&u9%|39;NPN%*PZ(raM{gsN;C$AuC-0dC5d2n z<-vfs>=4YW2nVObV97Jf)VgzuGxdYwjX1UWAFzC2beyeUJRJJ`N&v}O%e)MZC8v`7 zg`f+H+qV?%HvSXYqnI*<;RZj(GU3|rCHdtR3I;OdtpR_g#3`;WAUqE5BnVIu^XU=% z@Gp3<;5MHvj^nV}+WL6=_AN9>6XLgAB*^=D@f%%Q-)w4H=y{56(Zw6zlN~d-D=8Qm zo|M-rNDmOSRtRQ$gqG!TTRNoAfzttI6LJ%4?}UW8!<`Q?hxA=6V|Wuvmw&S?Z*o8h z?Il&6kP~ene~edO>^DQL<`tYSJIvJECISF99$lGYZ8RQ$Q!W1!3|^if=4-+xz&D;w z#Q4zosG-b^sf$l?-KhD1uvbo*=k#uOF$M70ex{ZwqV$} zF3kUEo5Q2!x#`_O;HM{7nc&3g68P9_>GCRX1T=a716v5?*Rl;ewqNmD&;QcV+pGet zQe=juQ3#r)POu#s)`ueWA}{}v3MEfi%)tSs_t+wuD>LGo4~Ji&#?dwcTbje+&bg9P z59LVFUWIgbB5=zoa~bEY2nzvw%x=4F%>uH%uqLB!H*2uM58=TMCNGQDb(ue(-t|9d z)AoXdaLQjVx?U9BJ7I+qc7B0K2-)#UDexz2k`g#=KIF5A>H-fpoVxfHnfV`5HP7K5 z^5Uq3F`;r@+k0Ce4>ELsEP=~v!S(7zU};MH8-xW3j_WiM30( z_bho!QeAo5IOHU38Z&T8gTn-tT6dwE|$VTu0!WEWYw`t zCPlvU>$bmkdcIm_*Zy+L1*rmK!8(YY>OtuB5UbvKmd4-F&H%HkcE`IXzZ=XpHM^A2 z!m~=~g)-twU)=yC-K@=BZl8wU%Tx;6R}5M<{6e3r;wVF5J>pr!lkuq% zf+4u4bP}DDTFRYrXa(ikfh!W4I&HU+c)6{y=Pl&Mlw}q*9*`nBXlavbVzY zlltN5^)k4?IsVaomtN0I(Efb7oQ~!5$Aw!QkdDUrCScxTm|6!WyNugJ` zh4(FB`1x8yGIe;XIadr4E~gfs&3w+)w-h~=s-mATwi8#DojQ1$%FHSA5Vfm7K`XoRdTxcK4i0tBp#T z$DZVpf#2y?WgIbTpfE^QIGus*c;8frZ_6`{pRWRY?z1XxH$)~(-2Q|}7y}S(xBE|I z2h+aDRkvyAG9o8+6*Z^j&wn1$?yV$~Ivf$_#O1_s{;0RW#L=Pg76A?g2WO(u-20c? zi}%Nn_eS-6qW6fC_M>gV#!G>G<}!5ZC3!y9ZupNU^`I#eu=EM(=-UX`1YIccw?ZvV zU6Ta*gu)Q&8KtYAXcb7S6cQKZ?8=YX(Lk$Sye`T?lyfGxm)7%iwMS`DV45`(nbQp_ zjnNM+@np@FgANPj(w$WDEo;C)I9~G_Xqw44$PrrzjBu_LHmLt`{cx6K0(ffa zZ4)JiGe2REj%a8ZRA+8Q|AUh7EMNn1^TaVh9?IDjdp)E#r|R>1&g~Xx4^$oTDgDAi zgbT78NJsDJ?B?fm*LYYO5ZW&(^6&tS@-L^IHxns*4jmF!NO0oHE zK?>o_G$gu|!{r*bjUjOUp{EXOhoVi;wNh9q4~HRufsFDWz#nnxO;%Vq~Z3=WLyQH zB!nrbia_!Eg3hcS^kYNDiz_#FxnK9<6O3y;PsbnH6WA?&44550@T1zZKjBqMr@J)~ z4lEM3@=h)ieyM1m^TamBka8y&Trk-Y2@&PVO9jj2?Uai!(FyKl9DTK3+!MKERd!LK zI2*D;3f(q(hfEBY_i#7qOHS>D(vWi**n2~W4Oc{@;sD!@w_YXL>Cp#7`H-R_@YVlQ za)BBZL*mbS^5hR?eMs`}*2=aRNfH$3fA`aefS7(MqA!Yv)jR3qls5C7SvhoKMG<|N z3_fPjNFDnjfkP;QcN?XK7D<7mih~AweV{eKNo_P|iI8Vmp7kqooOK0Li~9U%>CBUrgo8$;nUxLpwyr<$W&Yd-gadVPC=ugX+}M#)V$1T3RWN)7kY z{{sJD4cyyDHf|2%vpCAFooMUI5zYr*6(=8$341Z5z~P+S+-fEw*k~f8XcL00X~_h= zMh^Lfkl!$iyUN@dc{zVpJq7c#no|Ej=Vw|3cN+36e!S zF+})SL6bS3Gq^(gBz~hCF)CBI z;|=z)$Zw31(%%5B!jjaB9eH40Td{>dL2(hWw92V^y&-?~Nu{#u13Hon-XeY89?0vDdTp^Hx-%h|8)2enkU^V+t{%1I|jO<4% zvt=%DL+qyoX-Btz_$}zwYM>3Zc@Bx^0<|VI#2D`b<;Pdu;&s(m-uF46zvELPEwc?)Ew8!tI|e zY@~=RqaQifPNzWvDm+^(c^4J?)o7!*q~RZsC2XRY-IyI<7Y zd#xEug|C+j&G^!o4He^xcae*|`kW=j@zFu*s)^Cq>z0_7Xw(^inv!svBZar^lz>ZQ z(TK|Zn9Yh?bn8N%uM_fyEQW=Nm|&1}-C z>*?arOE@yO-L@aC>3s-)XKo{dQShuTbRx8BTy2ah zyA1Yxkda&sw!?VeC)8z;&%d2>pat6sk$(L~*FuXIW+3;ez8OE@sLxu_h@y->>c(X+ z`39N|FX+>+E@v~6+>{$Te}5Mvr!b37;!+FVndeB35g>mhJtYZ0WZ6WbDlQ7EE8OVd-4k{%AE2^)f{| z#*@S(OL4~ZK7hh5Q?D69dT^Mk6FsrD3)y8IZ)2VofJshSXsKYMJ=6z`XlY8Y zu%L6qBa3#8-d!utp+zo9Vfn_hlg^M8d}7@@r*%tBj!TAZkG>~rYZZCr^Cgl7iqs#r zB4X{$P73}9^)9RF<9-Ide7fFBtY^VzJAIhDrMeyW_ugjv)w{N$UPz*xrUs)elYjfM z6Ql7P&wk5>78AKx0U8LYF!-gp4HM%9{*?YPP!z2QUYix%2%>#x!jg zWIuZc*6@U;mP3;`CK1a7e8mQFFia;Log|gm4T993*nt?z(Z?OvwNA|>8n0!SZy<&i z&ixfsW?173lEr6}Lu68(<9ySYjp0*LUOO|Y3jEd$JP6;;An0bFCL4mazUk)8FU}!8 zM4L0e$56ktis7@KEtM8S6P3(bjH{NI-a>tMZQut_Q9d57DgoIJxJ70it=nic2kzpe z6c}>?18qZ~fWzY^E^B(Pf-2>J>Yvv&>T3~DbnuN|DO_G1lSp;@bZ`mb9C!YgdS~vi zQLHhIkH%YvY#X5-0eu(5_{4U&pcW4W9xj^=o!e&!Ak$Zf&3DxwOY7(d^#WcUriL5& z)|OL8RM{%~1wa%oLNMA}1c$+$FHYp}(jXY%M>w;nADu0ElRjwrXFS01q%`Zmbe z1~BKxmapI3BG4g8Kn_DYhaAY1_E_}?sMUH2u-JRu`W(@kjc^L|M_Jz`7JhWkm=2&3 zy6=j0${a*O*&YKcWLY5bF>e9!8aIyl`x7-u=`~_AZDGYI-ovxDVD+K>6u1Bpil_1< z)mR`5Mh@;a77RZ6s!$JThVGK`Sds?eNQDF z<@nP!@th1$?}Cv5H*@pu;D-}c%@<1Y|FoRMr@3Mcq_)Plzo7mh6bZ*6&y5e|7ebnY z49WZ#c`CFTLQX%k`qeqd5^AeRvBu0Or@^3F+Kx7z*}>$0Z?^&zh+u$$tQGixehPL~ zVHIA`nRcR9wGp*8oTh+(mY(#s3~|T@kfvip$E9+jFadAn-)lIo3EiHfHqQVl1wHbvQ{N5)Lm^gg75Xq_#Y!36p8tZ&fz$4+~IY z;E&wMACvoAT;l*_$q2{bIQyx>dbYz-sYnA>C!VS*QR4uGXMkJfk?>{zVoujlt;~}f zzf^Js>IF~dqeZu;#9j0ze{>8x-BWSu8eacdB|&K~zT-SaN$})1cFs7^(UJ1QPGcYn zwr{zyRg-T9OgD6aqxWF$aw6Kpi7XnfcK${r7Ry zjD8j^G8R9RT1_?bk?m_xFV=lOkbd47c+6{g87IvtI>Y1(kARyfFkd; z#F`q~kd<>f?eHXR{VMJA>ai00->0VV#<^9goctHH=@TLzqG(9q&xI=XFH@^-fE#$$ zyZb*Ty&tgP!siMIp{dv!+iT~HCjve=S>+YOwi#H}V)>(}0gHcGP1pbO`;O%mWfRYw zM49ojA#wcZ6IE-sYKPjiM$U$GEUB|ydn>V5>aGHX#aAtLeXXMRSEZbh)m2%v*_y}) z;x%=0H7bCfsNp=yF#M{9=YVA)liulcp*kC4i+W_hmwWDeiQ#=7o^Cz`Hg5kgLs9$n zU0KeldTqxla`p;2UI^6u$jOVYqWxtJ?-Fo_Q-9HGResEgCw0jS>}I!Jkuf{fd4`r`0lqNx86dw%d1oGW*w5L2-5|4g>VC?vS6*mD<VeI8LsUz=iycv<@H z|65>VB?vI4yLaBUBCQd?3v;EpMW7+#2`dk@G|JhtL;)_oUwv3;H<9&`2Pj2hF)NKT z^482hx$Ks}8=e{Os1@rzLzebQok>fbp z^dGHH#V5ZtVl0vj)+(_w3tEJFemaen{tjNp0<YWuBmG@$K=OcfE`})ygv%nT*i%sUdk)nxZg^YeRE)2q|#-yZztt z;kW)P((g(capK*nZmdzT*u~=?ewXhZd$E-|?uOEsLqGm|&9d|34|{i&}dl|NABM%IHYC6l;g+QdANmj0-!AvXu_z?!QHYA78Kri>!X~ z2gG*1ED$1w(2#(7&;=fnbctl9_o896J6V|Y%fr}9wX_12>=C{;+ALE9&;0T3@fyK= z%lRRSgv&IuDbQFWB0Xf%|0kMW{`HE?X*8XIHbL+;zw@@gVdi6O194)u6eEdskjPCf`aVWYUrPXPqXbpB zH**y`VfP6j7C&+??EA-l)nnN6q-SxnjRb%zltV+H46ww*9RQgV=qV|v`Wu$=(9g^u zcG#*XOd#@4o?{K!!&~U4(JM?8zxAYVl@rQn`Bjl|f}brCZauj9=k4qL2`XzUm%|#6 z?mv!=oJiNF=UMy6v&v7AZN1Bg%=CQN_;7++H^Z|=v50azn_0$xhi4TJT)A$rdnGsC z&N6;o{BmNK@?XAC@MV`0=vLq}U0F4}JuAWHmw7@%-th|kF-5#$m`&DtGY~;N}2ay%!UNV&aEr^@)eV_k*&*kc&c4<&$E)WfQ?y9K1uoa9XyTt4VEplKYa^9sa-ys+l~N=E`G{KTo~=%k>nXI?aiD>oOZWXCGMnl?xRC!Ow_fr=f5~tz7Dg*78TIe8M6F3DD6a;fMhdi z2>j?-X&#^Sj`{tehQxXUFtsatC^tduw!d5*IyV%)02CcH0Vpb?{6K+5rAz8Zpz83) zXC*4R%COlN5n7xokEsWkg8Q@0nxEVE>EfTC4W-07;8!rw1Z6 zglFw;O-b*2f>;0bd_Q819e)yQx7_&G>!&8u9-z@A{TF|T_UzU;G7LQXZy=0uLBpSz z0xBCga;qJ+nk|#E%X&7cO#Xo-GNxBi_6)sT=0}Y-M!@F~vqLd`+94#{1%N|Ajc^LC z83yt?kJFHvH99Hzci(EZxObLAE+z?b9>@!hrL~k*zNvPB%oS>31rss{A5WX@<<_ne z1U}2Z(ij2N&_blUld=w!!9V?_ZJkaVCvOgdnh)Hvqupai>?;RUcX%xf4@1Jx+HBJ3 zXRc=MuaD&>$8IM1zckrsE`DtB?!3H#3rScVH-dS}B#&LpekVJls}lg4h%NzFJIhQ$ zZYD=)&gJ{_PmyPClK0n>8knjUR9pwU?6pSWD@lwSj312p#UZ3pSpG!L6TLhKxs(8L zU~V%320=Ggh}nsb;Vv}hiHClQq{3_jGui_0qwPeAiS3PYtv*gZY5!lAbR9*wMk!lo zT>e5d+`zoYTiWeuF(vG#es?E4GruZyHT>tHvzGPBvj>QhtJAid-9jZFo7F5Z1vL+y zb$l*uAJkEcVc;yMomm6=Y2WZK0pdLTN${gKi?G9Dbz%eO&yEKyTVg9bS}1jNl8m(L z8o1pjeC9^dz3%(DY8^3XC|cITMPMq}DF1;MiT}*qBb|}o(^O*e_d4sb zI0zX(63-T9{q2->0R@{rHKs8u2Ch)Z^L!0)KjWKJ+?!e8Q4$KCCth17EAg6R_8;U0 zP~+I8lY@UPBMFT{?ly8U^dR+_n{Auzr4?`q9tSP2k5_4AkRE1oJJ)LpzM)rxSYsO{ zeOe+IG?IH258oGeo5N#ve+FNz3dx8t>=|7Xpe7X62NoLtYq{@$b|qB$=T{D-8@LsC zgWi$A%A%G|+FP7JHjvP8OuFVa!)8<3aR4da>q2i+i9OYy4TJH1mo{mD` zFmqI<`*7nTSnZ$Rn7Iy>X%dB#F+$hXu+Ghee1bSn#z7YLS(wtvC^TRC+S(@48YTo? z?h)p>9w|cXXS9)2tG2NQAm0XGF;_C5cYbv(0h2KGSKOjkVvbxo6W=x`LZY_6ai|Z# zaIp$t&Es)hAbQ%J-B=l9AmKhwJ$%H9Tkm)%I?nDN2yYV@-}s{0Sjuj_rrfUY|6B9W z+p4U~=ATW)3XkZ2Pu$QxkuT;4W4y9Qs?_t^OMp?Fmm#Bt&GNi6^w+LVy#-I{uIt|S zw3S@SCK2n$pxNfL)uucF)Fjv#oU+RxfM;N>Y2R#D@WF?fo%DumJ#*L}6)ah!5+c`V zRH)9L5vS3rOkUDDi7X_GAp4CQSxD;s^q34;@+ASg=k49s%tFNkjAz>SzO%0Yua8SI zfq_~bix0Itq8spo!xj~3{5Hok%T$8`w!$`mJ z`{az8meY=0M#d^$V7x|gJf;Zq^L{%7n>&yN7kFjv-}|mMPZ2NxZ~^yqH8Ofcu(W81 zLYi#ae`O1xH@HLeeOSdjLA48v0cqRXfWqGnPI1FdULe&%O4Z~HETnRACBCyDd84?v zSr&swL*|OizRblD!_yqBZ;+?jm^E($o>x2jq-tg#WG8ASALAL{7)+EXJ(KQWk^baD z9`r#1E#?>8L^ecXHseX#K1F99FEOh}PHt$ZC&YYcR93)t0QrP1nZ81^>@#A9>A)nA zWw&T&;u!IN_jWPxqCxRMm`4G^!3)(iSG`$&Bgy8h-lHvBmZA+FNZ524`P6S^t}Q_W$TPU%N})$3EKwScm9FKblf@hc$89Rn#7HeHSA~ z^%{@_h~2SxWbZJr{H9FY3`VSyie?`f7K_Q_{8b}m>VqM1JOX9!Ekn1(eqP7+(Jp~Y z`#_^gRVRzfmLP!U&qdE0P2hWEsIwF%0C<3Av5V?7Ts!~eBv%Gddg|0j07{G6kOWI1 zSp}9Kciu9;cq3q3B=6hED1-dI0G}ipb9`Z*#O5}rltC33&X%uiliE!xBZV84pVjXB~b{Dlfm<-@Tj=W8H5k!>05Cf~t7 zOWi3r+Ok-f2_Z(R z0^0=M0lv>`B4sEfyz*>SYS-S}P~}<`m;qGn!emav4?DzpsTBrG`Q314s-NJ%KbUmQ zHih7!6p_XM>HuSzULxl6v`1EuU?G$xkn(V55*$$%+kh_!!P3vbDM(wwvn)zv%D|&i zCTkgrB_>6Lunj7-!XQ?}M?oa*)4|R*g?f%Yinc0sxP6BA7U%ec-4VI^{U(-eV(Pni zFC7i~H;-2sdrL_5LaO2#fx}@V?5~QvA!0xgJIEbtG4XZGLRk~ zE0ZPVqXypGQ`S)P0DoW&8Q$pWr$gYjSOz)K2^b#|)WrxTK`^@QawTJK=UMOvw};<$ zl8F@1<9P4V8Rv}@D)9v>v0~ae8Y5BJ{a@}D-PQSJEC@O65)l=9U{au+)Lqd?5V><5 z*H_NRYT#P)*!m|U;O4NFq7_XxF5C#t_Btn_g~*##WkpC+mQWu1EG9E6H2AKv3kP6x z{EgpyX38~OtEzt`$3PJE5x+W5BSKIOu?6%Q^1U+8pTgYkGzHGEQq3<`sp-1Y*E(df zd3G}zcB_(-IvUmd9wZ{A6a=op$kwaHO+Z{(o~=KK?&+{@XP!raw4aPFso5SBP8wcs zPpgz!-k-i;CX+^fzR@7XmZxGS-k-vng(oWc6XwFJZ}@or1<9TBiBlEeIs{NbZ%g-w z5Sxe2A|iPG%j7cr!Gy%~sts1%ml`oN0TX?)xJ~AeUvQ|b6b>;ni|xGM*S)rU)Kq%&{hq@@sa|CP;Ym5*?(;3s z>GP@eFLxsB2i%#Tk=;~?+02hyo()EVyLW&? zw8Av)-bI*D%g{O#Z}`TjLe_1m^9jjJtThBFwA#C2P?3{X!1plO16QJ4--DZ&d-CXy z^SLjX@aHjz-8B^f;bN+2wrtw_@F1;`QjvTqq=|WKeZC6U5MRhrdSVu%e1hAEBpmC7 zqPkRi4GPT0eA#xXNrSB6FtNNgzV2`rbb0|NFj-IOwf9+E%dUG1hWP!7Pe?)cC&G?! zU+>zElC~Dp5h*iB*5DtU!w9U!92KvfdgTN0=HWuoat0R|rP_O^5mvtBOio2-dCj;@ z-?~wmxgnSP#x9Z)VT;}tJeK29`f=xw!s)Qt$M5e$Rfx+bEKSMbA@*=w_ zz2gYrbr?e#LLaNV27iv%gUnfiO=ZX-=Z>&5qr&zXw8qldOM2=#b;F;C-E z=tUfdQN33?xc3J(nGLX)h$*#F^`a&)HC*R;UJQ#SBDS`#WqwMPBicXzLiOm~j`R6h z)53#35tKy&c9L)N+4)}$yX~l6B7UBRryFR{Dox=lhP6vH6lZnqy(FXJbBY4V#(2Nw znx63t9yvz~7=n-tloIyQXuAl&)gnz2aEn-F(F-B_9axjfEwlc<_1p+8ZR5%quBTHb z%hMI6HMmO#ft72t>-7F)PSh#q^8U5YRsNy5z1$N=U{LyLK&tuMse#);x-3a7B2ly* z-`dmj0u!A)XqOVc|;;K^IPtnS5N6n@cd#&$KUbW;fKLfXed7*y;+S+z^{%g zoY~*c_l3@#QMFZh7eqmL{x{t)B{%P0bpWbw)cIwSd~}k>{N!OC!h63B_xDBmo5nIR zYW9gyCZi5}&+D~Za`11la2#Zfa&=W&R;S5Yypx-~^k@B{n@n|FShkLWk45Oa4oDmg zYPr;@O|z^)L&>PTNlhyAz2Cy?wT^9`rTy9HclS zR4rUr-stqEQp!~}A~j_zRhm}drhm#SGU}SGw8mN|y-@$HP=#ieZz%hj;YYrdFMDUa zM~)e*12GA3_*sTGz@Lr-g*;yEsX>=<8seLyPmue?&ALl3Lx``psL@7 z)&-p2z1t`N%Q!Edi{R`UaK>`qx%r_pY+144Y5^lN&;z+`T1U$liv??AlkL7)zCV4leb+Z1h9Q8F0l6F$3z6)7 zyiom{bypkCgVk&tJ<0WO1mLj|nK3)@i%5N34Nunvej+vd@Oxf9;2l~x5$8N5izoUpw+hyzFhVoZbv|<A zd&kCVy2uNMVihjmVbY%ip3>G_H^r{Mzwlkeppsqs0>($x28%nGG|7=*gZ{Rzza!1y zgW4@p@&a(mq>oI?iDRv{U8C;N_BsAVByQ^mX^1W%e(YUPEtFS12&-)Y&I&4 zos9$!3Q*pf7Fz#p#K`~rXb8(}K6!Z%>TyulAsva3D}D9dGyU~q!q?q)vo2=NeoCBI zKOpV)slLa8#Tknhw}57~$}W?2OA66?skZwwvya(otvC8oy61Uze!3I-XC~;!WWxK{ zt;2@EZ$$X1-izxGV}PrgOhE8Sy@l_h&CsFkY(g6DCMifl;y+N$-%r>SH__-m0K_u2 zV;!nL(2;3Hj}cAay4AuVDMzL6%W=K=n3XUY5AX~Q2PjY+uXJVz$*@no0|_!W3un8Z zEZH9?jWPP}`XqxA8GlH`;8qw35|u8r?QS=3Ns$l#dfZ}xmW&qJw*>OY5v1XG>|Po4 z%%64mSzv&*5AF~!o7U<=vR#I6Rm1G(I6K(M2AcS9udrRf^zG-Kjjzbe#!)ND~w;(7n>;oKhMk`;H z6ZI5Fz|bSd62&2%1LEK)1UfzR5IAAuoY9y(>03`I@Rnp_PineA<#6}#L#sfpE>%)N zchsG04YLidL(e-dw+3<(Syp*kw=8*fSrML8uJwLop>zC*#}Qe@Jcj(?y27#jZN3DV zQenDDj<&7dHa(O`1xBkIpIh6Z{qBC4Ag#(^pqPGg!~cU)r-AbbpWaOj!Gi~!yT7qd za$xZAAixTt6C^Lp=~eshtDQ2}>)}$LUsL|SNJ{z@1F+YxgKtERL<}$R2l{Fywl!T_ z`@@!c0p?upM1dUOpD-haxZF9S9)YTf@(Zog;lOW%_Z+X5^dMPMxO(9aHvZQ>v`-~M zxxwf|w8~utfHY+0_i}1ldFwY9H@zy@9Tq?|3Yi3TJJU%`HtMsRD3n!wdP95;Jrc)r zW!M2UVd;wi9lKI*roa^n7zrFzqowC^sO4PD4y&42|3bjKBmkq0|8*$^af8%w3W)3F zXa`MZP-=U#7FCr%V)4S=+BFk zEw3*_sS;vX(&qtZ6Q!4(p31Q_czr-vbz|k$)BgW|{8_|LhZw)F@0JVBg2r%JoYqA$ zhl5FdP<%sp*al%fHNM;8^8w33IxO=T^La`JSE34C)1J!{=rX(qzi2fSwISmYSek^)`g! z4p3A|j7}**IHM14u9$|q*b0g_<;-0a5=pEP)aBz*s}m1F~VfGjs3 zHI-3C)uFZfC%d~X?&TsW6RVEplkFv+W%gqeCJW136_-KUs$<(zA}g61VSEd~26I0X zhD93+rZk#ZYhw-U3mFaop25MkS+X&aB&}S*4Sqcx&2704=Mst$3#9>K9xfj)Z_x2) zzUQ6#&=4Y-!WA2e6OygRqhkY;VGH4FJ8LKJ-BNI5NUy%C1rNRc55vI1{K%}gK+C2C zX9?o&crJdN$dt734Au^tUN^xrzV%HFu+*_Tu5iD~AMzc%P`C93!lki@=p?-R?lh(E z#eb}qH!LTIh!6O<+g8NipJ3j8ZU6CqlYj>d5pgZBGOnh(+%`mKrr-rk+J%OB7+l-AFDM+bJL+ED zOy9B9?y#jr6E;bwRUO;;y6rlF<4rfyJT;a#<-_kxJn&wCX5e-H#hxi=DU1J0C()+> zRKg*g=g!1+Xd)#*BpfS5TW2Qz+9j^&;h}bGN#U3wv~P|#=j;OE?^UY&mPG3)+Cy5^ zm3(rYl4V5c(Op?M$|+Rgy}Dq(Vt-iiZb-{*klYMlDCqz@x!9iwL<2T;E}vx?Bquxh zkGJbCHS$zi{z4w6kigq>z7N`#HMg#2_}%BNT;-QRVp=#P`sWk21CY$ z8UW($e2Pf?lE~-TFKf30PS7`YRu57>l{xwNQSf&qso^wLD^=R;PJ@mNh>btXxwW=4iKi=#-E&}Tn6-kK77MSZ)hDGk?-*y5dNf7=3MYi!HUOIOW_d7t&>cfuz2z}T{KIzb> z{4JZNj-U0pbDyjDi1=_k5Q^PPZ7GYwjz@ zFvma8_lv{8mUTunA+J2PRNL}_xs%D=;SBOZUffV#s^7J6yGy3QB-`)SKvUy6uH>mj ziB=>sVYSa7%M5|uiAHwpb}nV&ceXqZ#(E}@X8vNUkkwAa!0U1eX}T1tk1(+51R^Ss z4~z|U)rT`W8%OW$-9itXkkHBYul|$1E|V@weA_?NBaVwvFeFf3N;qNR7|XQAKs zlT|HiR5e|%qEIiibSHIe>I*H_)Gd5nL6r}E4ZCIn_Y%4M-nY*?O@~_L&(1bi8;5Rg ze7xN+CpE)+`n*R?+1X!*42Hg)EW&3KZqRyFZQuX?l+`;gHz`zjP#V#3UuTF+Sp^_} zHBhj#(c0WU{&BVY=1Z`BIXe%w(uLOPefzUeY7DIqGdd>D_TIMFlv z#&MVAba5?M2NG3r517YJ2Z(a~^h#}U#&?hS+X{4|KmIVJ=)~01{Jrmsbp8nm_|3yGX zx*rWOB6c=-hdPTS^gwX?~5I1!RL=fKaR43DhXni z5RGG-9`qxJQvpd`HaAh&?dN)MeR5C0;aKfQ%>5X60W8Q)69d=6PY(n@aMn-CFvAhg z^*^O5=VgHtv11_5LhPQNDC=4tTLb`r;m*&YJX8eXsI9Jr+OIj8~ zz&MwQ)oq2V&dT-SA|1ml$-R$RH zlj2j2F{JCkpB@J@VK$yc-I0awAKd)!>U$Wfikxn%Cm<5r4F~?eQJ1$vDPM0fzu0@& z{|Hr6ka$E`!gyESE?67k!r@NFTVS4uC1&1O;R_X9#{ zwcH>dv{a2+?vgC$8mh5Cu(uuj*3I+wbfRtGbBOBELKmkLU`vmC zDf7}Y;3qTtA=AK3x3u$fy#x`A_|A`~WYA<5$*friN{Gi$wT)ztIN1luGBe5YB6jK$ zgbpWaP9fq(+nL#mUKznt;|4X7C;6e3^}; z=;{?g5y5(*udHM$Y6dx1l&p!{mw<`Hw%K3 zB7VMQuG?8QEJIiUk*Y4v3(!**GA%SWEOGsp42gZms_g(c0_ro6TvAIOSZ z)}S8S2G7eeeh{GWOnrJ`(xLe6xgQtN578XVj!jobL7ANm zevzcUnHsUjUnEOiE#bb;( z8KT0s17R5N&H&;?*_<)+YjB?g?mRtFtu4V_GyG5cLWJ(A2IYLyO%sjf`pHdhBvefkAOoKN=Sy<>HV<`1AwtJ+(^5yb62|NLD(Y_ zKx_H~`6%i=RszH+Z|&3b&(!Y6RPT8R0abErkyi0mANbWVou$_Lbf~V+$e&a7mDz>Z z{hu%Xy1XSS_CW!b-{eKs&U}*&B@D0}KnN)T_MVdKhvy7AcB$W$%S7|$2F}O@Os9?# zudc)Tf%Cuw=xgIRj3pV=Dn3=g>r$ca^Pc~SStmc+M`)bw&5`=i#qlaq+-Hcd^rZ0K zMEuY70$0rv8*}G5x@NUmXp<{q*zZSLIQs#JTM5G|${|ZV33d|Hs`B;<+7(veK8T!-NUK-z4W6T$<&+5VGLk z6@JKu?o3t6g3iFo57Lr&5W>$QKetbB&R@>d!kG${G1qj!opaFRwVch%z+vLWTK>Im z;*Kzw*!}*QqFAFg3S1sQvdqF&3-W~8*@F?Mu9KsVKkJ_HfA zu}qC_vG+;yyaldU%MSS<#Tozxm`=VofMnx1Ui&nH28;+oDad=id7wf3QM}<$LK5Qm zw@t7i#r`L2rd0BK^Z(wA_qXJI(tOANzR&;uK-x&d0P_Fi?f>_Y;0}D`_P@W<|9pS= zF?fFeUwtaVczQmNh5Eoh1aYOIc4M+NF_yv)zmh0>pYLHK#{TCsSnezb zA_|~a&jc^hDX%RPd-moU z+e)QpY~~u&qC3Fdjsr+$*J4L3<25FmoR@ZU{E49eLiF_>$m56B1}VQ>juts>Fo55A zPgSg%UxLXv*z$b+t}Mg-2(59)sX{(vv)o;6vuRz=HGK2p_R^_56KGoCy}QWTGNS#x z?|(xL!s9PR`*B5VMvtR~&O@)|{&Q#C&+LuKCcv{H<#&DSPRwH!agQ@|+DFFik*}Sj zE@o|@tC;hdNj=)yo?8WAJJZc!Q}rpM=aSz|KnNjPSFPsE+g;Yg#(iI<)urvZdpMYR z&%T1f9*FCv-P(8^VcaDGa4%ZUYa6bB$`jocDkL0x&&#^KJ`Y9Eo=#Spm-<{B(qCx} zG*)+f<5;$_1g^6e!~%Wg%9i24Se1X zh@8zvBv%&QR{i?~C86MkhS+_`Kxpj`AVF5n(C-JRb5g?M7rH(kwX zvf<#lI>x)wWuBGt}+aN=lWrKSAKr`wpVtuB&Io22rV%b$L@2;@{xlHKF|F&B0yWm?9>j)k1 zlbfR2#)*(m2rG~~_3?KZSRO(tU-We+7M}Q-|9H{*Xe3uEbiwah-s|K>jz9;3m~@8u zOzQwS?bAEl|9(mg_n%UdQXFD~K(XPOjBcIJap#Qvty3n1ywCshMjr?f+j6Z4BW{9R7sL!Vww3l@ILrmo6g1=rkby2b(G>5_ok)rB+y-k`VKF z0ipk0s8n`bD|^(lsR4+DQ<`thU7E137lUqI47Zi0LBG&CnJ+7$^IC6WGM9N5VBuAD z+qy!dh6L-Kd-9XHw)gMhUC)>jlK88wHjM!y;iav1Z9?Pz&s(b0?lTbhPgbO;QS#Lu z6=!@U%(xX)VW#6de2j2HHm%IcQw44?%*;b)CgwC7>{y4!tAO4O4G|>S`aeq;T{I|q zmcQime3e~byDl_#0vxaJU9Nle>+f=-$ieLq^+SY!!m$jAY7kr^eA7w`dwhLS9QVgPc4E(e+@mydflPw z@KGT6aat49S&vjImZ))D@Vi}hGa+q1><8k~vF$XIDDS-n3A_4DFL!^{gb_s%p&LrH zdyF4BAtSdT|y{E5Xf z%aR3(q}+M*d`>Ha2q2}h0}uR{n|kXcVd3eW*?N^#!kD@p5Cv`%-lAxGDJS8ce# z6q7-(AX}IS#oP>V`?5{L*PKizdUAlWVb8xm&iDGf2R=?R!yZBK?H!S7!oBZ7!+ z7bw|(2*XW#2KWg+QUC5G1`nA+O_3H)p63YmL>Kmi9xYL|iDRu<>F&*`pO<1Im&1$QqjzUVCA)-AM%(PLob5&*QaNrqVjyRa-C$ zeAk8l=l(+dTLfa#6AVpLlh~{i{%ziQq9TG%ZwLkW$(T%rt_|`xzhc4-@LvL zyBb~^mZbw(K1=2jgFiM(;ch+lKqUzTl<|Ns*wY><@^aR1mQDn56t0dpWPYu5aXg?` z;d||t`sC^D-O29!D%S6ffRe*s!na_HYe2M~zn~(tnTb)J%-sd?PqXWkbJq>dVvoiE zCGJVGy9v;bqK$>Kr0~Y70Fny+Y%<(1qn|J=bP#V@Xu3PnJFO)q>5&NwqQbQhusQ4% zaLB&zT-`>CoHT*!eb9V0udsHvwXzFvmx(m>ort?H%nM#+%;pREsz$H4I-cQniZZ62 zoZ3+al`i|n%vD`t`QflgI9|MXNeZ%)+=Y@vY|*>aq7plN2(Mo8c^LFZHjn$Fk(K(B z`%6sNI5)hkQ{2fyL`awpq%~EvZN|g0qLL!FKPD4|W?E00kMZpNe6$dM+%N20B(piE zw?~5B%zEX2nHK234iWTk99R>;x=Y1QB`W;$WFiqkSu&G=rveM%jY@@g6e`|29LlFi zOh6hIYLkxrm-#X0HEzHw4G@Ps5m9Yg6Qno4c<)1n8B0AoqdzWX--%@Mt3=N~*ua%t zrmMEe8{xK3DdAT?(DqUqYm9)yq{QFW={n)*hM@^@O^4Y2wlHog1zYPW>UPBMMUR{H zjPvsKp9&{u@2eEvL}v`(=AVWZTqof2sVbHQgLyjlx;Y7M z9vunqq%2NM>$*$I=q*NbOG3o*Ph-D?8iDP|3WF}0($E`qC22o>Hn zTwp)Z^Oz{FjR<&)ss%j|LRa3(izqN5^Qiu{)`!g?iGM8lVQ!k&mv|29`pL_RC?`Kv zNoS|B(TnVLb4Gt6NPUl=mx3@~d`+Wx$dENm{rjQl$E8Yrn3Rd)uwZiuRM0-&Sp_$oPIlx6XERl#80Oc3#~ejF!dxai)Z(#b%2@-efqil z>>inDx>hu`rFmsH=Ov`xPkXIiXcie1hnu-1aFMAfatXrH)b8AOosxB3O*Q637|S2e z%7eH%Vqg>pR}~=WV~VsYq-S0itA>(eRGJUOQnyK3C^;O)<$xuDy(emG(#f=2dg|#G zNtAhWU!}Z5k zQA6L^_u~DlWZJskxA5!qToBJcJDj5;b$2pCqHCxvAyM2Q5 zxk}=@XQy6a!fOQ2`lXhB1%tAUWw=m-4eS%L7i?(G9pvtLdzMobIa@T#zk|;D^-3FF z-$<{_d+({%e*+mkZ3ASz{PejZ*Eg8CrCtEo<+(JKoBciu^Mw@6kyk)`8zvFkS8&C; z#N7uT_{D)Tb%l(jLEUikj z;;r9ZD!JW&&C^zDXr?WbTl#K$hhmK(=Tr7heRkd7BLycQ3RQzYr$^nMtJcVteuB*! z_5Z~z*(pD$yx(dZ+5g+63Yp-rrg5fFdFPeuLV=6$>%{BboNKIQeeF104jw#wkhA_G zt0QWpNDs0cVdIp+_1K{^M-iqE6VUy}bGU=aCcEHg&^7CMYzpdF?#dog1;6FP`*y~k zg0TvpKkxct`oFSRSKvUPZ1*{Rz_Bjfd}kKJtks~Knq^Y;Xu*H(YJV=!{>MEbHWDvT zz>hU#UBLY5-?@iemlhHPZD4JjvLo}Pr6>B$1eg$d7 z$xb_5{qXF?Q^?ewL&!>cYn#bxx7w99VewD%4PBcFxrSgnf3v>iTw+Jl)Buy`p%O4* zUlTz3BdMb1pdyT?SNw`j@N(mH0W32ImGrETA$mDzHattwF&&S+^J`$i+r2qTzY8(b zqydGuAtvr##Ar%JK7gljua&#kigZ^o{MNdtFyVpAl_NQ6{8KL z@9Kk4JKXJ7Z4QH)d(<*Gf~M{DE21tdyhwpxzfA#r3nc5 zI9tT$e8S(ELVqq+xrx7=dM=!>Zs4VM<$GGKAN^DDz+-9>QLkX19vy;1Nqu>|npipO z$j(H~;N?H0fvkrT0o;7ChOnVZuy=aY*5}0 z!teHnF4lu|#+kKpY^H{t#QtC`A^z&?GSf;%e3f77E*1~Qun3>Xrs}zg?SLg{80@oV zNk$4u!>(+BmDS`bVv2Ke4mUiBRrc(fdd6?34xME7!#9D4Zvuk^X1zsjcQB7UVQi*n zv!3YVq(xGYP}y0YaO2m;?N|*I>EHL?gz)sG*f-}}Hv*hR{eHC#SSKWurz42DvMq)( zhx1i(4=UTkexH-R?&J+Hl$57edH0mV^a;IV-Hzw5QjJu&MApYT_)ln5iY!Lz@ zaIQ;pak>K-Rh6cy;bP001|H`Vp#jN6+9S*Skqne*X_E9vtHE}yC6o5Yxj{G9ndr^1 zg|d1k{#2zOF8K%kF&Et<@OJDR?XNo?xt6p{nTW7yR7rGRuLNss_0@1x1QYSTZqO*hz z7?svV#KM24lVc%uG=p0UG@^5E-V^fqBD){rJ_sh2uCZE2OLPE|9ysoaN2Z<8XicM2dqQQbZe zMq$Nx#+;XemMqi5x+47S6$-t+&!yU}U_Vns$QJPmcK~Hn;_>Ioh=Okvdx&<$$B&MU zS{-kW-G6=6yC>_Wx&k-Y(7;+_4t}0Td1ga2r1lc)PEu_h2;@p}SWfhe$_!AtQR|(n zJPz@(*9eNl&#sTVLnI;=k@!D5%am;s=brx<_?F+m!r|Z*MIt2mwa&R=Y?I_930^ZP zP1w0ZSb^e^_vY_)WQe*InU1owyEa(Ww3Co``pANUS z?YB;wy-vS>7XA~kBhuk70`cHODeI67}hw!8f; z)ZWxCB-MhSTdWA>z@l}0Oc71YrQqLz%Ao)ATY}pWwK}^6Y=mGzAtfXErfU1|ykw|J zzmL^QvpH?X$90fR!2fvEPspLHdy{Gk>Vw_9mmvD_QRXVC4p3pL)x445{N=sxzhV=R zT!49DgUcT=kp}vN4Fell_dT^0h6R4&6d%m3hX%L_?z7(?*7sYV@68cRQ5G1VYR+Bb z&r)03cMd-JPKJ-ULUN|5)coQtbLs_%Pf%?y)~r|7QS)j7=)3s(&5^=+QCRE-`|u|9 znE=O6q4T23$j4@$9G8{ zW$=|jnmW&}v6(wLW0e$@cd_)+FEFW~wN<(AKiCo|<+`!t>lYW;B{J!y6Yp((a|}Tb z_rbzJ5`|zHmoJ4tsxFzd0(p z`i|k&&Yg8Fi5{unBZFO04y!=HhUX=@3=hlz zdM1DtPx4Fv!adkp6-S--;2zG6LCBo0RXh~KFmv8hYBGC)e$=t=FFo($2{^^rReJJm znA_H4bQK?c8B_|g`T1hPR38uPV@<`<4_wBvNUq}aTOqYVGSI&Qk9raltu32X77Ok6 zSbnkX`wJ}XUsSw~!ls=u`WR1NIVcIR2J-t}V>@tKmvo!0a37t7Y&XiN(06nlyzN(a zK8iVZ>*rgyK;?$@^m3mi9eaX7U8kABFeaLx7=)TU8ih2QJPfJ^oUkAFX zQavgZzff@8JAGWcL`$2}dyccqHg3(FZQh+PY=+3Y`vBvERjp$N?ons?^v_-{nXjaO zNsZ@9WxkxNUO7+N@hH(s|6@SwE^n|mKY>V%+|E(T5Jx{t@su{W({ubD+NU8=NBBCOjuD;mI@{{aZvuMphIkX2Chp*u$nDtb2j>)(d=-t>(XMh%YHpUqg_9hC z+1W;+t~c2X3*C#GKXs>(DpfjUY5mP5{8k5& zM3halINj&Guy2f{9N5xz9>1xw7+0Ov8`UOx{Zk2fx)VeD_wHP!#>D9)w(2rfPZODV z6bh5up!)`T11^;;$+ocTAS>n$ZpU}lIT=Rh3mX$Y>nBl(&JMs9Am$-*vihxj`PBbm z$}8}=H`BY&0Bl8Q+A_ow_e83Pun#&n)<31^>gr!mcl$wPkP_;D2Q$Iz7a%9d-0l4= z#OwJb4OB{hYizUIa)=uDYb0Ap?sk$cjA!FAv&cwAC`j zU&!yfeUGXXw3@uC&AyoSeV6y9u6(8Vr!I=0%F$-DP(v#9BZM(IQdBfZj=6>=7<1DY zN>jy8`z;_Nqttc6)#U-I1EIoq?zfRru=0rGAGX`-B(GZe+P#;b(=a8wNw4!EN7nG? z3S^a=g#AV1V-E@-CyapzOtCLc^Jf*wZ-2FdY+N_^wtx7HN44xr-<~jIj-^m>hJN!# zV%#M<4Yq~#0E-rD?kuUw8P$L^U$gut)VrTAqmNj4hpipoaB*ql7sFcs^sXN*D&V{m z0M>a!W!kdkjA6WvC&|}h=I`;yXcw@^iV)#N5)^YDbJbvlG{&xb%B)01gL$7m)Gf)W zL>zIm-ES~P94QfW9%XkNGU9*o&5?mO(cI)fod}qopbn8^?FSzK;L$tt3}kc%Ox)#TO(pSuU;s-$5bSsT1hFWN!WDN8IQhUE9O0P0<^I zjo;*!!Bq#|5Mb@#m9=>Gud^QqVODCkt0aZ!}xk~#k zk<)eQ90ae{k}9~HAW-1ZoFjn;dbbTUWMr0O!4y{z=I8aIX59PAx zaF1=r*<5v+2g70Fk0j19t`eL(T}4ys10Us~GC2%3zd>`mxI@QCT^i?-pW|sc|Lxl1 zI%FSa+oKaHdH#=+hhq+u;@!4YHFd6F0g+_Zy1i|_@?9b#?gBM}k4B!TO}FoiuD0`zP{(ho`O{&MH^HtvVXGc~(0yE8SRJG8m(GHxxX zD|JE}vi#g9mq$!O?hY9v!GIdP);weN#P?6cOPAMfL|$&8@ynCC>oE7UD!Vc9Qx29t zEP2jGUaW7Z5X-9vwYu%a2A6K>k2UL#gK-3_Vb^RhN!3bW$)sva?c0$Q-?$dAD zVnz=M=%vVXnb#aQ76oaXAtao?zw_GC0GTf{Ih!rpSOJ|#XT0yPSohUuNrgjrJr6&N zNxUXTeYV(XEQ}J6LiJn$5@SicOs)B5RoMu6Xu!E010fNNha*Jx{$%xsIZu=f(5SH0 zB{}NS{h6&lu3mlikcyB1@<~9?EkA)(jq&}Iv?1y`UMn(HS>y;GiMETcfbDnGkDla} z<|PeBPQ@Or^@J!L@9u6c)y(vlR%oKEP><1w1?+y-yl$KN)%=xJO}u1d`YoEIl)$fm z^>>NP47_t5si@7jhl`?redB0BD}Ga#hiu(MI@Dx4AngXFVA7x<-2v)paUI9CK6O>w zGrN8C2uYfGR9peEhHiGtzW#9w=Fuu6ExP=@urvmfbku#}?XTmolDcd?o%{_K3aZlB zG1Y{0qQLYC0AVa9o`g(x?${~8qj3CQ9t#rsByAv-BrYC0>Z2>6h#~I7Z;|mGBx|^; zZYlU*5bD;V43&Ub?E|h3UMBUT(X+QG0uoU}J5FIb#=7(MD!pgYi>H>AAf!n9AyPD! zPWu!oE`%tB+bU0D#8E*~0MBvOi9$9py@3(%O&t!B98Tjt+z=)6IoF)rRMpl@p3IQW zEezG@(Bbr^)QXzY9VMrusTHE2;maCkTActDeynv#0}WX(LtemFO+$B8Tza`z3a^cL zsbDIYCKqPricj!pEU-|hEXXDQSQdRkwTV+_(dq2t)uq>K)LnB?b<7nZsCgb9Vw{1W zX!vA1*rF^WMp2j{;#OQF#=(K^)#^BMl6$k<1U;EO_l6@Q{Q!EY&SU;3c{GPqj~9fT zhG(2x%2yoMUlA6WZ#X!(X+{w^aTo8gd zlw@gDN|1FZ^h(z#CQSVoal<@=tZ1l>#lp*9U>Nl%xfDhJ<41OdFWO4lxU#q%#@wTgY- z$!!R39XL1_@wX=Qbn@`@)vGP!miYeRB20`%0`yK#XOW|L3eq~)aKx3uoVS-WEvY{_ zvoT0}U~Gsd8g7RslA8WrkvbmxdcDnzoujF;Jz}TSWkb zr{$;Rh*_`t_NG*l{uxSF@^O*0=%?dwJLx_uLXTAG91fOJ^dfpxwTSgk$n4PbYU>H| zD!Z@{6D%ZenJH)0{lTHuf76j=&HEv?2=?mebF$}-5NIIj!1W6ep@igAyh!^<3E2lN zo$1(O?4;KY8FRQz(0r;d6e#4_iGni3uSG*{+bR7=7aZRVX+Ji%J2~u_hA3X2Q?h@8 z+{JiWfkXfAJKt-W#|lsP=CBv+Klk>C_s7VP$r83tThOny^5MHg9L}q@mpyb=%%kZ_ z;IIsRZ*(xPDmiHP$l>`jW(Nlz6Qe;tMol#dIO1)W>TXQA+?ad~SDalCfGW2Tl@XNDhtAP%z0G~_o^|1ZK*JiB|md(7=DI5;d zphOlUZ>Ge}O;uP-m!>O2t}3N9+@MaEi1|#1eiz+MxF*T0TP2<9)4Kh#xYbpGC(i!# zD~B;^F^<6MN;cD~;C&oge><>YvBr~kp%`?Q(K{rmDxrNVe!Hr0aLo~c;7sN;Rh^P! zry7@(<-$jDYv2(<#{|Z>7w5Bxm=RNvkcq~w$VHwr{<`+s~(s-ZUvSh@9t&Q{^C}RfBR%y2hzq50D-g;snka zw>ZEjC387eDa7$r*)s9LX-iO)$oOlY3u)U{)U8*#4RUA!q#=$zV=|-dK+pp^NZ4X7p*y=e3C81$yk45Zf zeb5TYHNE_T{O};#@FkAhBWG`~H=83FZBhU*jAj!ux>l;^-vulsO{``mU)51}w^2&1Gxc^Lv_q0X-XC59U493!T&N1W@0KznUT3`**Pxs@cDE;5* z`|iqbUQVr}zeW|lrWgw=_dHf?AYY_QwO-^_|nMQD97f@30J-Nrjmz!uhZHWM*ibRdKEF^;r;z z;*2=;SiJ6iehS|R-RN~|E)Bunk}W0~tBRzUz;LFzPXvkusa~STB(?&UfVJ(uSl+ zLR_xTWSrQg?2WqK73L-@Ifz%bp6Hw=ubFz+Q(&(qUAP5=#-f74PT zTJyf1WHhx19&Hj#<8&G=GMMz$(KTemL@A{;2p3`UL6*0u)^DN-b z9j$8IRyVyGRIe0k;#d9=p(aiIZ&2m}4hD|8Ee->f12}x?7EKb10nMlRrqG?qa+Mhx zpQBClX}8Fa{rAm*yRIs|x(aKe`y`l`lO%eRqWxc-$3?XR`5q8+1&6qez`l!Z@ait? zrY|dkR`=JGli8d#K_C&s!A0W6bn)v7Xz8C)h+4(0{d(xIpR4Ge{yj+lqI~aorRJVn z1C&;iUHg@|`7{7mr6lC#rMj5fI#GosG(8wnY%|3}?ZFAevHU**lfFVn0;$Eoq}Ti>a=uw9DJt`aW#V^!l*#1-kLyxMVY)4y!> z0Aw_4Cc6^kaSbrz{0gG2WPfsP{b!|6#{AbHP%Bg&Z~aP9u}sLqH>j}aFa^dva%o4! zJ_5Fq7sQT|I*(kX{~9&`gCMuklUWk(yuxDQxW84s$pl#Aba-!NvkYr)4pEO;oj#e9 zn@Gd-F$g=^0377>-vzyEuyzrCfMWzs1h{Vd^TjMBcURUL7s-P~iAa0GxZ;ESkVRN` zR8*2a4RBW)_g)K?V_(Xw>f@Ph;)xy%>RW*Ub12=wNRDi&6Xf})%s~43OLIw!n6v%& z#}A!W`87g6AxqpudS8#)H6MD_1C!(MWd)8h@Juxc*Hh$U zqyjUA&-gW_R^_g=j!bf+%^275(v0>$~zkg`aF?DL~5=EeU zhg9_Iy(#n(A(1h=3U)|1cOV42b4B^2YCI*YID4gA7qilpsx!f+YR+*SJ#K^YaZrb@LvQp)qTP!xM40gS z|1KN<1WjT~;*VZPMJbg%bdW6f_%ZyFCs3eDTjzSt{Yt=hcc3Rxw!~R-;tx^nSkZ^8 zJ4%e5O5C^y0py$)!(XURqXlZW$Ngbew(;P`a$8g~TSH@^ci+Y6O1GJ+kO2oPe3X7; zF?&A~6Dw62eCcj54Zeu&oQI!VC9Y(2CSwr~%ij51(n+)iV^Y@5BKCM#ubnzj2@?D8qpc z`zS$~VACO0a6SK!HoG25N!|-nQ0Rm531m)G2uyWEq=8e7_|{l4Z7@{NxvEezS*XK? zN!#+-QKz?dGJfxF1vAaj z>_`8kWD&*z_aZ_h&P*_2eU^V`n0E{kP_3`IBXZkJyM|3U-RF^RQyEDKH`Qh{q59pS z5+NtVI3#l2vw^qyqvCtgqNHgjcJPOxWReWdfthIfq(DAP^*}+%{OTkbNHJ8K(=%0H z8L-}LJTN&!)p6Q&FGXyZjwaW}Q3rYNiv~v~afQ-6b1VS|O`HlK=LNxyQT{bkV7zX+ z+^$6RIFx7?{{6LGcJv57R7(4EJ8daCr_J_MIGEEGm)J7Xa$nx06dHA1$3_)gOO=X?MLVh^DK z61c~5djA?mE-&%hPiL2}2#YwHSA2Lbh33L2w)Q*y?F5)&(b_f^sOHmQZ%rird~d55 z{j&2iT}}#DNz3SW$2gD~0M)^XyBX?h%TXnrt=s(?ot0uQWD(Ek!MO%Svfovf$6UqEC1(VI5i%1S4) zjYo(|Z$MlQ`SC8;WQZmhunWsbM0Nv9m^r79g~akSUOSI?dJ^^}S% zhVC_jhsFo)g6yQ`Ak1z_Dx+JcKojTLnie44&mLqgQ*s^!B|4}!oo*wl@8x$(=5b~{ z1Mm%Iu@+*T5*y;wa3J6-HwR^G*90;yO34UPRpp!7>yO6`r`y%o1y3k@m5%;vaEPb^ zriO^+km+h`86f!j6)m1@cEaJec9~?-buTT748Yu494ZEVssgDbrYzO_lze1CVDHeC;}?`fW0R(r0tu^2y3+p}mWH zgK=G2=(!b#p)e;q^OL<;X3yQ;vXIpiJAilr6b^f}>F)zyHQ$ zUT;$w9Zm^$Gwe>@=}^gSl{iAd+%%_x1O^p;f3SK(E6(h{ z#k~2y=L70DZAn69IZZc8gBJAZ?uyH-KcNHWrLOOJD-Xnb%5=zlkJad5Kq{~Av8m>H z+gEFPa}mbr+g{X}J`zF9`L10q;|v9IO~|bK3LG5O%+2tyNanX!H5LY2`wmaKML^R8 zJtQMeW-iN|jd&@7!A#&yrP;1FIo2&W>2wo;1octNkv4xm{+OYTj{{p{g~70;8HfWJ(nE!6jo!S~wtg+2VbnZM$`q`?!_Z#D{!c+La5H9Io(>&>Crh>!Lr zI#l0u%S0bqy1yqaMmk_&&B_M}-U+xGGJk)!1O#C1o~!vM^I5wEo`<8Nx#7|~-=8$p zL*D>N5JguK+f&huXT4{;0xNwWEDSyVB{{doZ^{8Q%G1F`RTvw5_#D?srm-eYK-cnf zrR`AMi=%F9`|iOHZN{jZ=v(KYI`TZbOfp@LPC4PqQ_9yX{4tuZ$8=+0NdOInonFoQ zYLr?8%X35FVnLzZxTF$eRH^~&C_jfE5lOH0CZF;t?i~-rhtJiYK72NfflbUWC+x6* zEY_^a?A&ljkh?r+JdfUa+TIKOH&4OkN76eSW-7IIuwU&)ekLsTTvN#NNEQ3(!wH+E zUypFDo+T;gUA&`o`-yKO24~Z07F1&#Ne9Kaa-*62?Of?U)6VAYTUe(q+~S8^>RNvn z8b4+@59)&Ay)^sw&#sF?=!fSWIBW~Th)ny7WV}dH9yxY0M7GPO3r=)W>ViL9445|X zxn4#2Y;Uh{+KTxjgfix0pXO4*f3c|Kzx+_Fa3I9-m^0ONJlQJ#f{Mdg1b$=n(XKpQ zyfP_%m96>a`}LHRtGzPI2JrR$a%g?Ddg?y@2I$6;LMeR3OX~N2bh`EIx)%sz+6js5 zz5vPo;}id?^bcgV!yZ-&-EY%9{y_= zcpk0EUDxrxv-T|$oPIZS$->U#Vp`T!y7ebVo!>a(xzZKJ2To9l6y(R5Un-_C6TFlg+=hc% zbIDv$H%1YqRH!pyf4q3Y&`N1u9xAAo`U}V&miGQ=RkU{z6+C&Mr8<{7avv;)hkTjz zyjS;0@Zb{dtgwO`r~6OT9Q3nc@j{1; zPdL7uA+Z{e&pxZPT7x=sH7-*au8Xvk=oDCDtPRV_f|ojpOen}-o-!~KsR}B!CbQ_M zr`Fb3j+jlYnbhL!xnIi+ngz$_p!8avxBEHiht%Ja0ozxa3d~in+UbSmTtU5h$!{ZI z-ta03_$ZncLbUbsQ}zn$x^+9SMax-f>$z=t;23YbcaFwNlWhhoXnx z2;ZKJ@N@-OM>-Ewzagb5_+jB)JHAO57E!B2oi?98LF#@2=6VQVGT&js_Q z>x~`yxEJ42+RmH3`o4F1ANlDjLcN<+IycWQ?~`j7Sr*hkClhXr@Hww<8*l8C99PDF zv;Qc(d^J~YoQvvQ7XqXPxUBjWZ=yrbm#5FwnhutK$b&5AQE60R+coWKixN1RSW<<* zD(3YA`mu(TIddiQ-{q1EB3Y8rL)Lxff!pUJcSgmJKf4x}6Qb*}g`479le#N9+h15U z{$6z5o^|i8KUGr$LCs@^`(YTI$9=pD3Bcf>0ycbl)1H^Y`92YC+e(xCN=hhmZsMz>aFc2n z)TX`J5VooGQ5Q5W&5f%{+7RpZ7GNTY=jS5cjVs~+lrWW*D@9k zBJ)_QYBYu7phN|;qC&YK=d;e)cB#`{HGL5;N*f{GpL{U`{qBY~9aj=?U#Wq_=g|T! zJ^$#CZZIwH$msChtLNup1F79@h$o~-#;Ph;D@e_aOr;z6`E^JIW#5{%cKODMcSxPM zE9#x4@$Bn$+%If%;tw}S3VvSL1wIsA{vB|g2P}&X`WwSJ;ZIq8(gWF~qbY=DZ_k3j zFkF%MUqKp?V^=%@_xG3YtlzDf8Lbg~dcOl}7)lSmdmzjsU*jXkMFU%SWS< zw&vUZm@8V<(|AooG^yU?y;}e+Rhc2SnXl5ib1dz6{?|8&MW-ZW3gpt79<4A~F~`XF z_vsW#e*n6iZZVwiwB(c2c@jNj{avueyU{<4sN^^*!rnmgw#9t5?*4_#iQ_Z-PV>S+ zkeTqvJgRvDC=?|?rWhIl`D81{Z=v;l3HTE{s?Nye5G|7yA^951NRqLf?3f?m$W5bZML2CU?b zs6)5RPgdB($X!EMG`NuYk zZ;V9D`_r->((Xgd_hfGyT(e!b36(Q+r-BO91v|VALmwdRFSPo?{Sn8tsIM+4XpZ=A zZqv~qyinOS^fBNTn${oHhKF0|JwWk3S|7-bs)K>S$HzgK|K6~1W?=k&yMZ@YJ4+VG zSxzBrm$R3aBMmwH!pCb}zA?gvB4AQdxWF^P^bk_AIaY|faoDBOtftph59#yet?JGy z*s1^^S}}4+LJLkQeUM+}wGGr2vh`Tiao$3$;}DN-nM*c_k=Re|CV(8$Hz#5abNw-o z3$a7bOjj8?E#ITuW9jEuz=ciyP?NPBz7WTr{C(xeTt*-Q8nl664lZr-#Z;n9KKZJ+ zn{XvH$$k=FjZrP`93>Sb*EZUoJcP301X8>>ZJe<4$I8 zOfAecgjEw*lQk1Ho5v|01VR1v|F7oGJRIt^jpM0AVal#-r;s9BBBUgTX>^QjtO;cq zGSZ=k!OQ?byc~HP=O!)?Y(OFO=JKvGq-|YjLdwh-R?0mtTXBi0#EU@HAL3JCY-=xf|ob*QDU7 z_6VKD!ob=RPNyFG{kDPlH4x(JvXhBI?1%Q+!_%9Yh8h);S|LS?_+cAEIa~vl5;f`WWK)~B zUi|do>P(^*Ubu_7C8E`H1W4ELWl({S4#$)HZ8Lgn7Y_F;`}HWO2N5G0+0aeM0i6Cd zr5gjP)$ZST{KFc9k1F^Pt+S?-MR-$s#!Dv|s3_7SMXm@%!>M~~pgIqB)P2h1L#Zm7 zrMlF%ZEnT2kP_COi4}P;zjkNU;0A%~~z67D|_f5pLs7ZqGLBNn{S|}w=|r}@5Uaef-Z5w*j^`5E%~)G!w1=fzdS5W z!x#Y(LSjeS7~;wAuu5OHebg(y^LR^^w_5@*%%jwO?v`$tH`vkY@CV{S7VM(h2UUa$ zYL1;pO)Rtz#1$1?4>S8}ebGVU#Vz&7xhi}r+o(6-xtiM8{7m;eS4vCw+;4T-sMzdr zgV}3iUH~ZIXGWP0k8V54gqix)UJX;SE(mR${}(w^U=73*xS4_7#U4I`Yt; zA$bcjONqJBGSiAZLXT~CoXF@Tq$>7&pljnXiD;OSac0Or)osWZ{^)F4p+tmNKzSE# z@2!Wj@RQ}mZRyb-^%#2i0QXRSXF;YqQ@2d2HLSrmZb<9E*$f!hU66{##&W2vN1^hu zoPA;F&4I#v2g~qfVo%99EzhFzcK`dMS3$aE?lK2;o(l=?-~6NN zwr@Ey*Op9L3&(^2wEeSG#sd{o^r8^DeSDW&gEz@bZ5ESrEfX_o9@8f91BM7F1H1%C z!V2L9Im5NA94c$6cN&Q)M&k@s*#^PiJo(_Q2DZ;J^rJxaRV}E{VWk?#|Hwi`x%F6> za$@N7?<*9O_p?my(CpAvxg$-5$1N7)IS^Eg4UXdBFIlxU?9_H4Xw*~Vz_|uz^RVLE zy~?CjKI`o*bh*=wr|SGIf%=%K2PdmOj;biP0;Cj z+%I>T^RU^4)wzMVc5VdXB4+`9Lw2bS)ft4FS`Dn3zIy1Y*lZwLgZLfI_H}|XedjE2 zflr{lGwv#Mrfw0GCaUaOr%zU-h)HZ$N1DFG@+o33T`F*VRi}MHq9R=?Ex9?T!RyhSweC*~bkM2m&BI=PLr9a-=Z`jD?awh6 z5YAPL4F-jkLxm|iRoOINP4sc+9&*x0}}M_wy> z%;&6Tzi7=}UqMu5glC66HJp(9BmwFS>8GV|EJ$%mAgng1>gG_v`8qVEGEt-`tJh}J zg3!)=wfRt%$@yN?*b9TzJ*pf*F8ht!lTK$x$iN!5wt%Z?!}L<{o%Xdaj_TGeAdfD$ zCz*9x`vF3KD5<32{%@n~GJi@aaqxFd}7s1td| z?>U-Poe!jP1(D2+>%=Q5I_`@z=(>+B`~A$lXBBZq`EWBwP<8S%l;$O&rez+9kV05b zcf~_9WItS*A1p}*fxP>soiTi;wH}{4xGi#A&S4i zYl-2PYj6QvmSHV|PN^Y^7B1VlHfN5TR9+!3nTpu|T#A%Il%Cdd=i0jSzMKfyUcKY? z^Y8f1*|M=tz#yhb=h1bmb=~kF`elcCm z$sE|bilQZD{70X|fS=#^T}(!1?4v9Ms7-`3uJe0Rv6uFo8>RKwK@e*97I{LyX9QPy zh&zl z$$mH|3*VIjo%yokfUy6$L)o$aE&vkBY4prcomOtQ9I2wVMP0LSEF%G}SPo@`*v=rh z2J)H~6D@tgg8hNAe5u^0%T$q&pnvY6{xe$-``$uas?Qv5((@(-281xEAjk1Ks@s4G z^7iE!kTim)C6H^KHy3S=V8M+HUviivrg%JfWlBYwVGgX+v68k(mHqy^$}q5^T(ovKwelTNYv zi~j7;?BLxFjZqUbJS9bg3jHCVVyeMR)c_$bP;5L3wS$WR^9CplmCT3D)4i8u$aw)4 z{SCb%6ry|b1*aN=>m?wST53~s$86+ZW?dwEYcjY{C}yiXk(3`J5(*Ks?X3r>h?uVv z6Yu!a;z!tC8UQP` z+whkujbq>}^y|~FH~+d}_{JdX78L{dG6}#Pgn%o}3eSG7tw}|bh9>KOJb{MSf4*e` zmF$$OsH73X>Y&A8AboN#5}_?Se}rqYxpExxK`19(#ZP~?K#>L;U~ z^q=g}nLoSCEg2w!5 zp~rYHt}6 zpKBSHsP*(rT$`TYwr^@--Sl#AA%PEcsZ+Mcj`Sm&ut}}Tjerm3b;3aSmF{Q2T1#&> zK0o9Rs0@=x+A1exKQ~w+wf5P!1OXx$o1G1QWJ2=NXJ29);0Dry830oD;eGCOR+s=5 zz&S#Ih2o?$zU{~s``C?W0YKec~MJlc=_bxkR>{9g?A fe|LJGJVI76aGBsvk3zoM170w7`0vG6EQ9_7gwhj( literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/fast-gen-overview.jpg b/blogs/deepspeed-fastgen/assets/images/fast-gen-overview.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2affbf8a4cc36f6c6411e40c4f4bffaa53a80c7e GIT binary patch literal 277661 zcmeEu30zEV-~VYL3S%irW$s%Ng&|U?##X7MO`_Bw6iGs~OmiYc+fYb_p-^cVOG)db zJxO~>YTEZ@X4+;w^S`FJ-MW|OexCPv-~aPI@6)H#%sJO-&RpB?`u@JZ@9*M&;x|Kc z4RrN&AtCT%AY=eR{03+TBqA&<_ys?r;CI@LY12eSr-_S+O`kDKeAX-paS4gpbLP*T zJ!jq=35mH2=gwOoDJ3N}3njg1q2!|Zl2Vd_4-pap|06nW=Co-uC1*>_mi)Va_}S3B z8Phyxjfn^?g@oq`iOdt?=Rs%)5}GFXwn9SC*B>Ea@Exa%&6p`J0d7d13keH}hzN^{ z2);GA+Y6LKqVuNB|9RcE=?hE`i!F7LRK0ZT@eGCSFW*U-RnV5LKXUf+O!0-%i)0on zE>~KyQh9^gMscn&7Zes1zpt#SuBom2RR6i9 zwXMCQv#YzOm(Ca-866vEve<%p2|*&0)%vAoQ|dJj)Js@YR76xvP%k0ji{KEMCpzut zb<^i>GZj1RvS6v|r5TdjZ#{nbZl=O|Gn&+qvlZeCmu+ZKqzkGxQM12RvCIEZ&3>ua z_j-MXW{C)a&J&pjVIiJ*q~8kYKgahrcxJTW0xjcJTE67T1MO#ppI-{r*P3Qgn;d~b zolv*B8+A_pX<+>E1J7=@Sm?n>iU;#t>v*B_T>QCO$_qSp)|iLTDaFY9S_LTTy&z~P zMMIw9PPSkDzPHmHpSI0e=@lQkjBj#%Xi^-Tc&yY6H*4PzH*_S?(>@>7Ad%Z7b>dXU zy2Z>8Q`_0uN8VhAo8%~5G2&QUjv6L_B28HSI>NB8!BDSx`CZbn>z*?;?3gbTh+>YOvV<14&EDb>jK%$a+YubMz2(43la$urDZ;>qH826Do29NxtpjDL-< zA8vC@bAgX6_8Q_2x*es8xrE*~aui;0;s`M~M%~lHDWI z;Uve^QlUj{e8}>QS?$46*6Vju?utSr(butE!$y3F@(QM>jfbz~Lnn@GA-~1ed=RrA ze>j^DiKFt+!>ng~h+2tApa#xTm6E&lK$p;juF;m?=276Ero#;)rFar ze#YzKM=roW3048qTFqkC`r-{#!3rg#Dx2QdkiN6I@VV0GNM}2xRoLlLgM3JW51qoG z++rhN(KQ`gGIiccPxH6h#E0A%B-$G6=lJC&?=8H43YhCz_mG*}+!UI)Ds-`fjVg2a zDmwMtU}jbnLoMGLeQaf*zx3Zua)bEix<>QZyXL^f9IGMA7VGgwG6LHOte-jjbBn}jb$ozhplnnvFRDh za`i5*sV8|DSM~CO-wN^~_SWw>ih=uV*Ls^WD@*=;?ffLGkH9V+O}b^g&L#9FoR)#Q zVAYd*xXUkLvE+Ej%lyFn%Iw2kH{yAp;wTJO|5;cbd$Pn>sL}Yb+Yt>{t+Pujz8o|R znHD!^W3$(|Sjsl4rXQ=d*Vqp&*)ZU7HrLtde+=w-ltH=YkLjl3KrFt72xJ; zVXT=R;v^aM(@pnVudBPFtF-U*Td78wIV$i;_3ZiC`X< zK6W~Ei0AYYthtZzI!Yhv!cVsqH?9ifL**;@&{eTk6pLF05A~C{r!zbKdXY+UqgMg0 zcTnUtr)>I^8;DF30vIBsQgs9==0gR2O%72U!+DKf?~uepTd)~4=P5UIPx%1w1Z>hv z`MNLY-p>D#2h1Qgll?|Ml$`K7`3kPeEQul<(>bp#%j~1CZ{~4LPF5W}c~@S$bA#!=dmLgXPjWr4F^Bc+*m&o~7+_|=myN!JH zqw1$-W!>`+dde&iz3%&4?w4(Nx_SSdlhut+(#IU@*cFi{S43vHv75gRS7C|1KCg`* zJ=0Tzn`%W*91GuZRYzo5$R{D30)9V_oNRYfK4bd7PD2HPGTrF;AC z2EVqwgpdzK#mj{ZhUIZL@!~Zto4w4mR0_&VV zI?-;gb&|n2ev(-=7=ky~D-XJIh%sZ6=#p(xjEjL8Qpeh~);$i}w%9<8d%fz-HPtIGH5?cpHz&Pu zOw;-hr zX}5o9Q{s3`n%&?l@@PavVU?TA;29Lo-F*(pUaKbWvWlC+iKDmDlw*pq-cqcA*4fTx zyvN$gHnv|3S6Q90>rjy6h1VFr_j%Ly(5qN&WRteFoR|KdxMYo!e5i=EWa&ryJC#>g zsb-!Gs<25<>EStqy>2xQqAbu}iz}HPqE--kxT-(Re36A7=iJfx#;GFZmRd^ha8kW) zcL&)uPxw%@f9M5nPWV`+l73Qo|B=_PE;$XfE!u0!CYcXestiW?1<9X1d?Kc>3Kz!u z^6H{Luk)~##d7Wl3)5HrS8QO_#OTc9Lvi=!lnrUhPI0iXzwy$D1FQ60$HmyG-u=k% z-}cW|AeDI+acAIv*2iP6N3re>8-3X%dn^RYo;CqOEeErj0a{oqPik_z6Hk)VUa0xxPlZ~7SZY1cI$abMW;_M^3pIJ zdB?81ufIH%v(=3;zQ5+ULjptMcE{>AWKon@13RR+t9}D!&WR@3M$dI2~KMCI>+FFr49nu7Pecw4k*#Q|8NeW;Bo~v8H2^#$+Cy2Xl?Zm4Q;8_ zQmC!r%a##dgc~z$`H&o?Ysicby_3Uok7pV#4+tOV%7M!_U};i_9;VubI)b+3L&9*U zyE7L92jPatW8vl?G_qdr$5Y#Ccm%uHa+_L~J zWE%1z(KCFA9*+IZmC;qBU| zuY1?|Dz8&kcBNV+fdPA!#o*CxTIj-Xxy4z&p7Nu&DR~*H>aNFn@QMeIK=EC_L^OkuavD)SOyPc2czVx4u{FaLhQZJ6q~- z|28SQ`%=8R}T(%I)#)D(01@Jj)QScISRD?B8j}tcdi31jjW7 zW43m!V_cY-gq!J;-~)4{BxeO1ADV_I(jCXLtCLr6caqJX$J)nY6vR|Ghuf92cIirb z%%92L=R5F#ZqO>TiaYz+?rue$)N1_elFRT#>7KViEUOM=F;^>v7DSRS%~=(mR2;VK zFcO1q#u=Y$${JEGSz-EE?)VzbkGNSL9+W;rtb|fBPc=yfgnc5U11Yoo3|(Ja1_!*|)jgNfEJ62<^O=L0m9PdrefVZ^SES zx2*6n-p1s3WDE?+;gHd9a2684+C;rpv3%St?z)%Ki;<`DNkQq6NyZP|)A2Ez6tq5y z#U@IRUFJO=AS|#7#2D=w$}nRJJcF2(=nXHfB)@s;y zEvO|WIWJnZ4C&TXX1fGsdwA`$4)!{NK9H|%<+X*fqhqgBK&beL2$R+*XvZYd0yr;O5!L7m zk^lGkp8u&q!T)#v|37{3pFa3sKvRj~E+I-utjC!zK`d=|^etn=TSfO{Uu&1Qb?J*5 zSLcDkXRAD#jCKcE+nZ=wded&9k^@8^zGT z^LZKG@?`g8=sTM|la3BJ=Q?@mHB;rgv2!=PjB!|ee{i2K?%;MRl0=sVhZmLIg2 z)ArnIEi#~f_F1zwWghnxxqO^0=f`l5*EFRm<(c$_OO0IqvT<4t@m3wSw{y=GH1hUc zPsb6}fF)YE*_^ZoStWc36Z*x&4d1t+*oxgYHUdWskcU~+8gyimWjJ>pv+)_;uDoei z`To9*k0eTj>kB+KFX=XJMTx!CzJnvxJvN_hTKPO-86oy+zG*>4kOFFrL&xfFs`8(UDmP z1Xu8DkkKI?hrE}mz}d;fwN>^8w!B`q_(Iv1j|aWDZRqjhM<}!J)bsaH_QbVR(se#G zcSG4#WNsFHN#eo%2fQoY?UaJWM#&^fODk}ZGMy$4@cO8nF_ns} znMk$vTqZr?FwHCOj( z+kCowMyzG4e5MqmkgI9e0w;{~IFfK~ewH+w3R7QhMVnXPR`DUh1=wR8;rR^`<|4Jo zOlDQFQ z`OeNNI)1uje|21L<8I?)W3=()R&h(ALsP6@Be+n5?yvda8+ z^EI@~a9;sd$R;Ujmd(mTad_2+Vv^Q9<$L1grDrH|+KSwYUeahvBzD^hGL|`yE|j&5 zyROpp#1}1g?3;{2?%1B0*rC}8fK4H9C6_pw-dz$OYwVVee%10)l79Z@>rZ&~%j!OW zai7;SK|V1H$2w}3MHDWPuksI(%DR7AYKzFS_$P?A2{*34m+3p!#D{{(fTa91AHzdy zd0EcN*DP2)V$9H%2yAZ)F6eP@C4 zN}W?z8+ChkSA>g8yUNC{*L4uOXqpzX;kftjpNf=$-zZ>W3d@F>*#s|TjFIe zADh=ZtW??qDMxi{K4w{x2g!^1P(*kgAL>_+g}coyUmd3$$oG61CH3kuc5X4R>qgAD zt0Eh|H#zVRDpmZ&nGlz&y?i%*J>A;4CZX7G@5b@;(3qomd%&1U4IUtiF@Yzi;6>uO zMI!;hnVOmB_H}q}D7)8+zBv6^ikT5Q>b$8AVLPGwn3ZOY2s_%JeIE4kjv~vS+|GAQxz&FyD)9f<3H`o?>%i%6JNEr zabpW|-zoY{H*m4i~H88-(+QSWGv2yD7a zC$+&RZ~iAHdPk*#Q*4ZF(kqWwd6bJisnvk&)A3ToDG{2m!wO)4j3W(o*^wIr~_9eu_zt~(0I*X}_e;}&ZGd(}&Xvy6G!#tc9U^2i$#y;d1)23>8*;1i1R@Z?N(yu$wZ zFAfK{=)yil|9zFtd!2$E9;)hjU75#+9@ixw&w$4+^PzhvUIsj{dF1RkQiUe97$^XX!Fx}1B+ zMiCc&tV^8RTb$K}d3b!NgjcTo|{WYr5w?BSTO=Zhr-sIYL8eQ7YHM@e$ z=;cYXNUQV0p1g_|jW%S4jNA3HxhbPBi``H;`~FZ zVNn}qKB)<739SG@Ibp8*vBd6ks@ZZ$iZ_(|xMzxcXCS^YZ9}lI+xwDMu`B0GVPlPy z%NZm9w`^mL49z`B5YIgO zQ|i97WWC;y<)VJtab;Yn);&w!js_nv9KSt2Oytf+hCdd&pcG3N&f!C;WE#sDnL}0k z#F;LbaHzU(6HYMc{?!gtC?9H$q%cmWz@t28Fc+IKk9IH%y^x6C*swGccvfuK!pL>oX>9Ajjr3m;O%7Es2-{P~bxUH+Z1 zQ11LL#K1QY?w>7<5G94!`Z|&(Lceo}&8C+iyZ8|J$9`FL#0PiXu#)0T*aR1&=`hdd zIEVo`(r0>1{O`0(mA$DIt|~U$KkksKiT7W%XN&56Ki(tW14P*xe~4)s=}3F)c=yCt9tYtGH`NOS&P0KlHkgj#xlu;RcT{>QOeHio^ak9Q z&xaBq!qdOycv!wir&wN>IZR8u7RXp{SRX6bz$Z>tGxKD&5lm(;r(l-(n&^c~;qvSVn zof<*m*X%Z?0Z=aTxj8K{(MaF0KNXWtsRq!86*4$sS+aiSrBK@Adaz@nf~ix6bGFRG zz8UA<@lryvZ^7NHzFww1qAO8B5-LP9ltP)27bOLq;U$#wm-<2MKKCLYx}{r&P(t{S zPBG4a@SeihritN0&N#M(8oq`O=6b%u2|k1a+O5cT@>BR2j`h<+usWxZdCK9b!(^7! z4lev8XtzINu$eO2$A?be=-U7QtsCLRH$Y3kt8?!cPynV8gmt%vXI{&2rL=ddeFr34 z;zr|bt0_6eTn5e;U!(f{HFBF_Tnid)2DjOr@@B}RT(60;TD2)9q@X+GmPxEm+^fa%` z-$TpH2t*fEA^!!%hnBkWA-`=pl3>;oCO{Cwc>vfCRVr_Z$v#alTuQ$CY{P6TtaDo5)Hm1wX^K8L!=+KnRiCu9taLD)0wp?5Jqx6LXoR#p0nNh;NlOC zd}vIznliF7?$brfzWtf8@Kv;LJ7tanl4UqD(G!Ux-T8mi7hgNcu?TiRvAltAQHWHL zZ5nNNaJHgZ(AV=FV1~F8%NX`$T~x-JI0hIQx5f%{uq?$T5`5@|VkmOmpwBUwsR!yG z^$s2r6;yGijT6E82BOKsbrRkIceM}&NS|1Mzk(@9N)+tZnchfW}i?v2i8|LuLioAoii;rpFwUS>SY{eZtcwai7D*! zp$`3aF0l+|S~i^-aX)jx}~*8<}he#k*9bVC+gspU5@IvIbf5q>*T|8NNZ z3l-2yWOJLbpB3~e*ZciB1~U6ccwe5ZB1LI=F6^aI0#7^PU5;($wK&S%_qc)Ps#Ihb ze%`b9VWkmTjbvDF5`8_0cHauC+YLrq0I%rVNc&fOWxMn2H`w+scrb?#0s~WCOF>%>5}y*@5gO2p(@T{3P$EEg-Hte7+_+w)_rk17@brUV z9-Gj~yi1W_)VF4>CMQvvL$Qp#nqX!ga-fX3JwjC)Oo5f(V5Ww15RRe8#eqBPDI<%y z+od?3Q!(IPXy4VWtDdfrwkPRZ%MPE|aK>mwcTkI;&!~6}31>^lq0WiVQei5!)?FQn zLdlONq!lz)Zk3Bd?a^yDg7ChU=WUJ*<$lh;8*8Vs&D~zLO!E2}i`$(~=}Nvsno6y+ z^>nN4*4cjAGscKr#JYWp^tixrqc`?+mKfJKnfZ=ZxPz{hjtO%W_PD`ZTHOsy zqZ!pW<-{3kUZWJe@F&6sA(}5&GP+G^EcE($s_$n!>sr1})8iNPPIYD9FSFF!aC2&@ zff+WQZDrCYn|`i7dIF=S;`{b&;WA-@1y@>x7HT7^TMn#%E0f?khoc3?R@94zFDIU*$Hf@$0KwAdf6g)h5nuXx{zL?0B*BJ zvV*~VVX92D$J9k1n@44=C0rkDhFJ>=xn^KaWSMYikwD8rz79JLv6R`yyh}^Lq#CYk zsG*|nH6%P^}N;mqkMXrDaS-@5O>EB^W z5QVYgGhD}jyHizqoa-%-T}oGhb?D~y zyL>1Eql+&D0P`(&FM%kK8w+tBeU0G=wNsNV0pTnOH;|D#J8&Dg;sR_ugiuHT8yDgX z2{kx?jZFc(qO*FzHxS4o$H&r3s##yKw8Pz$pGVBs>pHG~%#Etm}85E*X#!Gq{ykjh%_XLxdl0ONspm^q7f|$@x*z?~Y#lOyLBCY^8KM5sl zm6Vpibv~l$IQNFvnIV}YpIM|9XB7{YQN*ZWj!}9*j>xN0+;4JOP~v`Dd_G)Izt;1C}P zXYs&|ARrn?AFr+BLnke>EylaJ^UDU^#)02F_7w6Ool2E`Jr}z=ffcm@Ry>R4&32@Z z61v%z{n7(30d#F^C66n7$Cc1(FEDvUQ^CA{+=kRuGUDpEf%4rc>sIK82*lFLLMC zA=@8n@}Ub2EG@hO?jrndHQc_kt!&(h5>hUzJQfz1J}H^2P}g^oUN@W~lU3#wi{DbMlv5Isxy5}@?4u7ujQ$8!?(q#xGuI>H z7Ib29a+NlEzp?L4|8!dP3-nS_83;mUwvy*rF^zN6u8}>L<~!L-?Cvj=6T;m9Eu;NY zBiDp^BY_z_P@>k6y8*cW;ztM2cOaLV8vJS|>`B*aoqnUmUv%}F$7f!?w!2Dw&C~M#fSEvKiKif>&yy~9!vlt{gP(cpWC`x!YDw;F1nlO zH2+f2Ld~O``}TBlL4FS|8J7g2yd+upld3aMw668}2pDUCak=aQoCvoNBm>8;d!IaN z%w0f44D-Nz=R>vLSaXNpXg?+ifp2b_N-Mg-6EVmFwAP_rlo9;s*eHGn#~CiRtc3e# z0@C?m7Wo?7ekfChCI_Ch)E)LyvIBVu0)C@=&b;~!M<*GL$xDFg_|Jt2hG%MfY(jyx z-5?9B?T7T?5xmqyGYfc}U!xdOt21j6cIU-2SniEBWF??Ha4Wo+0@yW`&ird$LLtaK zIo+#nSuUsW%Axq#V3aGgW%81V$ltsm_E3;ua%)f3H_mhS6wb4ho(DKjrSDMp*+0a2 zdK&T}rPZyduQAuYFtv7bfOOq_zLHfQO}#uv>0}HoipC9XAy06gbBd#uN2CSTWP2^@ z4_7?_bO>9&UaE%vxvp+-A(|z-`YROfJw_P<9Pzq5+#sq39W?^j z*{#2l8>mlF4}gp)5xCtPn4GWbfhzP5z<51=4wkMjpcjULEq>cKq8WjsZt||f?Qy--G0{*!eZQNazNflSEChJmw|qwQE&|9@ z^bp%B2n#cOsS`mJd;|xuq#uZY%cgRiaR)4mDJev55%4s?Qubza1g{0!Y%^|b90@K9 zXhXB+wPFXTyfw=&`ZORbzr$ZnoNNx7v9g;q+9rnhTh*0%9WxHHjsDoehkQ^vtT}-2 zJnc|={QTE9&=#PMNsZb74OE|g))`PPl46<`0Yp{biqFCeWs%iL97u(yjGVN5E)Yc+ zs5x_H1E3iB6L}>QjD)3gV!sy05q>C+AI7jG4Y`IR8mdFAl`BCHq_AFhTLHMEfj#&Q z=3uqJ5*l)sF2=hyis6Z$9Q_c^SV?K`$k_IsfOmE!B>R63WtuVxhBUC8lY@C;AT3J3 z-U{%sk<28o`}Gp^Cl~61@pusFm*%DLq0g>;TjGidT4G;_6Nz<+LxXOi{=eott;k5C(yo6;T0Ea(#yKS7%c!u(v%&aTh z+>{x^4CW2F8rG8&BjQ6g*^Ml@OxgCHxn;qPehh!0JFY4pRl*CH-#%{ zju_Pfwgg~-k#)E`8z4(636RX`ZB&jca@I22b<$s83jBrpw}r-Q{y?0)M2a)1B;WyC zJ)%Y~tl>oU;9lU~6k=Z!{U~36K?kA(FCaR&4JD5eC!_?S0l<9BMk7Qix~o^zR}bQ8 z8nQmEInFROaG&$yfSJI;4FnjF6CxUQeWr>KJ5Y5UJ?8-HdZHW-~-Cc7W6OS2`m z8m$}ud~##Y=79D@T?3MfjTC+XXp1^#X;z&X(=)~xiI2$CTsMl{j1u?*13({d;bdr% zQhA;e(U-bU!rCGH#L?4!elwTxphqYowmFKrh!3?$({zZJC1^+?YJIKp%#K2p#Z0ru`vIT9V{InDc1s^b z^9a$1e{fa=?+y&)0XU?IGV=T+i`aIN>vxwl$|kYb7_iTcflx{C%odpA0b;!lakYBZGY2AxvARZ1(82Iwf&i%i|CdZvl+%b8AVSgH)6pMYw-ZK zuNVsm@*ll|CmSZbf?mK&PNa0W|HdmI0iZSM6#&2pG{sf$4_?7K?mSR+Gq3;vN(s>A zkJ~8>WfGGQ?IMMAQ5dF`0A=Dh=ZQOuqLvC6ajT$-H$pM}eyS^zOcww7>HP_oSP>^+ ziI-8>+Tt7!AdZ>pA(YY8xdG_NElf1BlMdoZmjV2^t<4pX0xIoP5XnCZa|nPu^xoghA+GQ>+EN3{&GWWYB}p^V(c+XJIlL? z5xIWh7@qqMfL3dhze*&|O-dxZB;TtD$np0pghtF)JQAze9qEnMQS}2%`TG?jcbEyk zM)&bQqI8=t zT>H`!ycNfyiV2JP&?p1qpk`%YiAhi{8vjJaD0b+??#1WT% z#gSA0sT}!ICRk+Gm3`&NsZ%&|R5r{Cg8!HzPna+V=|--9YXS)*-hVv3{tZh1D;DCL zrT8CttygzYuUFBQlff8H=0{#zuMkmEXF-a?ll=<*_W=qS# z4V&9%){UFD@gaG&b_?$7JefCeaHrat;CzGPCQODcwVhj5d$xaxO;E1xt0&~Ty4Y@* zmhM_M2%Kbm1xScXIf%5Q$4lKiq=rd-w(ek)4>L}gF-J*=4{4kJ>fUm~#y-*)AB08L zj0|PhJ%r2mX3>!1Y`E;U7q*!)4?W+Z;^5o7Pq}RS&nbOAOkgyY_V6K3w)l_=yK7CX z@>^xK<%3_2P`?lh*R3-pYZM-(c1khO1LszWso#`s&#sHC8(!{sgWb6~fiS>pz+@x$-qWb&*_Ds$F_eo;z`59(_L49F zVClf^jlg#Hv8e#6OR>yx-#hH@dC7r1j3GDL(2>R81%VM{7ZGc;zAwC_?s0zt~) zAGnGriQfoPr2kZq0y^`u4P0zSC-E+U;ZTv2wT7I8Z4QMQd%1vbI0QJNkuyM$a`GF6 zKIQeV2vQ6yfx_n34gvRJ*1QSM{w@jl3`4+Y;AlYvpUQb}=j!ULmK0)b;#|bbf{pI% z>DRg}Ui*|(jJBR``@A-GWk(%-liKhxLKdN(Dpu86_bzkWq3halPkr1^W*>VM)hW=D z+Y_Po?ca`M|WB%+EuGmbBBaE4TR-y=eZBJI%NJG&&IS>Z4t3$qdbL zzQ#sh?Z0Jl$|?I!D6>N|-JbHnxW%u(JMtg#5TXG(kw%gk$zFiS6a`FGFb%AXz>E3s z3O1MnW%O;!@+EG9WHTLTO|QC{=|tw`Z4+TYXe&xiq+kB8E)c`~9eM`hj!OVu8l&mE z09s@17|IdM|G!`V?7BtJoL)j*f~?lR*_1+EU_)yn*9^pwC*l0XDR90M9SBe?5XJOP zh+>9*6vZUs`w$%y0r9O6@iQ1Qeq0PcfZzG)K@ z(lVZPA$yGTuwP{_HICl|(64aUC_H><_r!<*suw{H;-6N8JfGPJ7&wb)p1YDD!SGdv zaho$Ya35*3{fF2%Mi3jXRGx^9U-*U>H6y0Vn=zMxyctkCJb}E~YYWE$E*7Qh@zMYm z-m5l@dWT_232gf)Fq;Qm*-H$;dIkV1apVR|8(4HjDGVDFrR*A6U48mVWZIltfKm+h zgKPw66RDyr9!&!*v%v~Mp3*M`o8Z45M)HtDfoB4<{a+VA-$Uj(GoNn2u^GO=XGw58 z9Z#GI7sW-vb+@q~8E{|LtLQ?Ec_%Oh?|dVR$vT_@w@&vJKf|{(bzUSl<2u z2#{i?@3;6;!Z(Wl!Z(V4qTyA%C(`?VX?j7@-_$;-=|Ia9x!zy#{5}ws7~n%|u^eYU zl+;%aR*NRPVQyHn{des%puGON_&FDl{Zqux#rT7uuBL6q9AR4!RPrTX*A^wQ=1hWk zXYOwcL1gC~@(CU=;Pm?w8FA*`$+*UzYHLpV9W1HH&`HayEQ3vd+}B1_6xQ3KXfjnXI0r+ zo0J~0()(zb_;BAR#Ff$c`e{j|0Ky9|kya^l`Gi=E27)PFFeK6Pxb~3o*z#;{AVIHn zx#8#+@xn@5-fC{^IOie7eN0n5XiTbUiE|?f*ltnAVqng~h}FQ+e^pQZ+6VBwP0AjW4;@)6+%DlU&9l#$u725;>>>-l$vufYvFwW2 zb@ccJ&PPLZ>?XC*x&ilA>?FUTn&vaBem{FNQ{2S30xA`*PU$tNn8Ha!o^xiKiQoWuX+}FMaQ&! zmU$fO336Lty(g7BV9Nhm=uH7^)P!~?l=LnCF9s~5Yyjh|Tm2EHpyx3W zuB!)O0YeH)`60lSFEnAchGl`&7|fkNgcw8zVEa*MBnV{s8UK+===U@Iz=Ml_`Z3iH zcM+K5)!25WHYyP6jCnA(XAKGJ2N#(aQU)8~{t+QI3Oq$b@y7DNwp%-*F2_ed7K(kQ&Ag^w5rS_b{>A33g&(@l4qDi?!Yd zy4|?{PE3}HTYD-n|8=#TBd^G+d4HvBkh`5q3UMwrq4&0-M1XT^n*6eqP9rqG)m{v zb?3XaSf-vjTY2lQ5}oV~H$S9ICkACL*KdvJR`2ncJL{_N`(bpN7Tu4xb23GIJ?|;p zeude9ybFQ5ms6xHpAu>al@!Kyyc6(*iYX%taDH*XmtjXCGl7*??4eBZ=49Mn_}wPB zJyTtfp7;n=-8;oe`UWR{dHwC46M~ec1}ej1gv1fn0YRBAa$F0EwFu~O#pK|khq%@o zRi=XYGh3ugUI)l2+_Sfa0qk`YU|AE$0;c?uqSv+`%gY|XMbFV~yk+N(Vi~5u zG70!!;{yUAZ|o!XOkSPLyEH{-0KR>~P|FX5wZ+UR^ zG;F&CfWzdkG&qpy{#rJU;YwlY290oCAHXs5>bl)AT)$`pTdCB}8t2~LOX;ga8WqYN z1nqxbUf~a${)bKxQ~xh}Qa{k$@o_)0;o3_;3q8PaJo%8meSr&+d!3FR8$v=CEXV=A ze~!vGKy11E6@IxMcWIjXH}?Ci%4CM1^CbO!5U7?^R}{s6ymZ!~_z58Ry`_OLYqy{3 z3ZXx|u*um^4_TpQjMT&3)npsNytsXmIwl(ml+)X6chGkMHP%evVl-fQi|_PN!6XR| zv_^8zg66%hqFI9I%;!V%d2{4a5Ca2vbO1Ruxibc;Zu?gX301a*%-V#R3*?_)De%2y z6`*bk38tgIX=GN{e2cc!e~Y#V>`oj2%YV}h*_zXOedw!a%!t6Z9y8#wY zAT@g6Ap>(h5a1BaSsX;ChJa(6X}A0)E%5AA1v$CI1zSjp2i+HiZ=knn$1rKQk36`({m-gTWX1V=_E$V(JWR`1dxtwkv@p(~Se3-b5&D zPcPj$5XVrn0uriVxclsH5Oc*uYRLaD#0*)?KBw<*T{*bwnT);(xqM8G;P>&V#=a}D z)t3JI9EOWK%2Ew^+(i2mih zv#Un(o4dzb?6!B&WXOZ7q7P1&$FvjMiSJ&9A5q4tvq*A;nx>^lEi+L5^2Y8Frj$lc zYL*?hFW(U^OlxrSFR&_jRQlHFYzHAsL+TQ)R>oGg*G+lv7RB*Vu5rhQss*DToUC{a zoHPOmj#|dJAI2RZyw8%|m?qX>;~o~i`(c(3%kur)oVo@1;=CSt6M~Y|_)w0`?DN=} zpSMe&*k6Nco(>*PB7jky`Pxn?vXhKPh-H+R9+v3t+&`6gLpIU~aMPT<2^yZYaO zl~W8Lo2hj4ALv)GG(+6z!!2xUc#}%=NlqJ(d?2q2VMFIGMaYZyM@aT6BFK%@R{5SA z=?`)vDTltnX)xw~g{QYl;Oyad7@&1@%SE#3z^b+27S+{LXogglaucA_C$O}MSdY;k z=w;;83o$GOL~!RRDtk8JXN&BypIZgYQ6n?go$U3^l*C!0N_lgv>e6T`dVlCe5l}J{qKI! zwv1qkHJ_mZ`HnZCgfim@6GMY_D-1AW0Yz)bPYPG7c<rnhzjz`RQ2|K`2O># z$guGBI^Gk>nX5#)pB)dVIE2<&zu~w)#9kQ|wPF7pyH;QB^)%Agoj2b7xbsHFMM3H3 zZ>9M^mc9n*0h1eZw3fg>Zp`t8cm7-c4VRGt0QRl#02^!z+%Ux2gqrtNhtf$lZp0&kS@psxn%ew+0=RKuK9eS^9TnRw~%@PL4*I6J$7n#V~ zd^p_`G-xjc!&`|yi&p?$XxdES2p@pEuOzu%#}8uNLkSD#0U zUB~?V8^=3{%Y*$?*fGdq0pYiK=;~(7*icZ?CxgMDjl=gn=a6Eh#yg3QV_F-VFgrzg z#>=zxd3U*+#&{l`th$b(8`cf6yshLHaWul(d{2CFmGw5o;YP%M@C*9tn9RwT@d)RR zSKP~hMfcVIBCsedA4D85t;>6W4(HoOCw@Fj!1@X{IsgS=5U+}DIu3cCx91bo}pPK*Oa5ZKp_Jh87`)Ithc zAU6SdssL6CQc!})twS74k{CU`M<^ z-`MWQtkk+3wJCY45aG(V1=M(dWx{Cb@*(5&ZHmd&Mv{KHKV4pj3xZ`S_JE~x6KOr2 z`s9F{=U&PS{rZ8fq*uXv*QF?QS}0@^KA1U)#Z33c&XR5+20U@qq1{bSy>9wI_5ZQ= zWog^ntyaD!^k#I&TTAjS+r+7w?x zCFu|m#CRG2@)8(J0N zc4{sxnfV_*=*|!E;Ge*bO%t$VMCB*2`-2>1tFYyWbTagfl&vJ3TXHm@6Pr;ifa#E#YxxtpIvA19PF|!#E z38YX>a^p)UmsWIEv?)^u&%nTXPw*#-Oz&gkkB>tK(LnYDbl3(g-<7dnzQ{C-W@WaY zFnOenJhp_AfbnIY1T%0r2nvHJ?oQX?!Tx48KBNgXDgrnX)*z`Ufp+G$0{8C+jzlZ@ zGI~%Ngm{~wjyc%iQcctSX_1l$>%4pLs zF9xj&Jep?qUp$)4X}K(TFFG4LPOI4zQx-Gw?%=qtpaL{Os{upap0NE_4UY+819Hn9 zIS$}8S37nYwvv~S+xS7?hD^Wj)PSy+;-`Idj_?^zFC zBtVmXX9Ir(P*R%}+|&+aM8tq7;3$ZNZ*^0mX~Je$dLCDPf!c!V6XNJIu?wNM{k*U! zZqLZt)#0!6e)VMkadS3qVx<8xgZ=`{E+{z{1A8GR19cZc!^%WGSs4A684sK*E>2s; z#c9o>9>7BXyYbjB&;N9D{cYi-sM@MM3IPwX4|}XfgzjtG$IUT#93PS}Y_73)!TuW^ z%q@pYTT7C3u5Y>(liBLRS$o@cG!Myd_O`NNFSxnA`qxnAP6{YPR>`(!R#+|Iyb zTckVnJd>p@8rQV>Uy3~X%B+STJCBiY1e)(0dZ4^EfPz6vp+QW0J9|SWOZ&ippxdG* zlC50H2BuE-QcQw(1*Ikw0Q_q*-c& z7$n^VgpJYc;9}O9z_Aa!UEHw;UdONhzSqX!Do=f@H&#I_X8VdnsPWuDx9G_Y?5YKk z$CH*t?weYW6I%#0pj?6A0)hh;zMljO-$(URK^S7&|8HNs8Fi5oa}>4jGq7dUi43C!R@unj03(R2@fJCz?y-Oqtii6PoN3xC;1rx zTJ;fMJWi*(#iJPV!&rYH^L^mL<2rq_Fd%LGbSpPy0W`lWB)BPy6~Hjy0g>m~0+8j< z1?H$+x0370t|6J7 z@ZK$7kzr{9me_Qj8+{b|;zDen7%K6S)b9q#ug4HTfqr@LHZr}2oBMy%mX_6wWoM9k zli{-X9waLp2l+Qz5C-Kw!$tHY;G+I#xL7nfxJY0MLNzFm`ak=9>fcL^4|-=Y0VO#+ z-k%WTk;iy$LZosc@@O<}8i>1h{(ow!f1IM6DY@7`(dh9a=k{DQQ#9@~I@OWmqEpfz z6X+BeIO&0e(Xe04-oHeY|Kj7P2PJ%tB*ueC;v`B->v?UAai zd5ihN41KhZz~|86JH8n4JLvD}ufz(7u{xR^8rA-_`~@ftOlc(N>It(~+Emg=kE^ZH z_*PYKcrBVYq?Dt-O`;loihBJrfuRk^PbBo%Fqu|G6+w4c2ExF=cT=vGY?A<)biw;T zxY6q_x*rdx%dt0l=TOJQR*e5(qzSMLQlKnLHxm$j4%cIWWIR!my%Enn1T}FIZ*UUe zQVT{6IKf_EaFQX8E4nd}1StO>K72xq_Jg08qSO%=@a&Dj6TjntyieEryDST~QRx0e zKla{tc!f0**yEb%&N4=&&K0pxcPiAe!vpP@Cgmdgu+e>a*LkX9rTyF9uAFI7LrO7+ za5KT+m)Zv4q#L)`(OC~r%mc(fzQW_L#XLaNlUvMVBIwC2<{>v;%%fXP+-?7WT;~1# z?fYoC;nlWv?@af+SpB}`{TU`j3f!z6d!t28@tD}J8xGjzzcw6-R*m{=(}LFL9`W#V zt4y?hHChDfu!P2&alB2uziAVY5A!NzyasU`0KBMVINjNO(X>7^tr|P=GvGW^qOdi3 zO?{xjg|;|;LOcMCO{+F!Wp!^urBG)(#*pW`$?u1bA_7R7y88?T-}LpAwEOndoKV;_WArG zxcLu}7$#sol?(GJg-RoG?1cnar=OFO&#~kg(fEb)GX2Z`e|&k>DSh2=(xW`z%|OfI z5a3{jUq^JUNz^|o4i-wU1C+%Qbq~iy)vKS)4G&cp1_WNq@HhW;hbsJ_sXKuD5waUy zBexvo%y_C1TMARD3l7=SJZOeCgfu^+c`!j;w~7;rTp5C(2NrRPG{Z{9NI2HOl+n2;}F zRXX25zb|@lgAqgt0cW){y}`qveg)uYd|CdODe~l-y|AAN0Ef1r{pS)AN6$pT&=G^$3?l?>D#}4qLxpD>r+5EjP(q#46lAAR-E?I!~!B>_~OynZ%2jfWlX4e`Zr&oDi z8Spd#rItI`T1o=B!>u+%!IC@RfJzs4$N`>h_!|f8LUfDP7kG{T7G5_|7vKRG9M4Gr zi-n$;mVKMwe>H^VMqYQf7l0lV5e?e$OWlnR9FMO~9jzSQ(wwILCzoVZ8U?sKfxZJM z;OC(@z|ZsHT%JEJZ+I`l8q8ckWdfKAj5yx;cNSs@oSYe3X>(QUq=F|k+b7W?m6XOxAcS{HwKoS&(!UxyN4 z8`H2?jdC&b;)5NPl2Wfe^fy=*9Ix4=dj_IicJ_unMxoIh+%;n>I^~p@<^F?9rx%7? zX9;J)?#lp}o9Pal2>~3Uwfz zK(H7?p(c<$soQ9`rdz&q=q-G*v`>n?-ju+(4$Q0CO@MNy#~M}VSVtZ6Gc!%m_n|r5 zkR;yI_Y=FRuSM1Ay6{SK?uPJl8o2!nn;S*vmsBxJAm|Zb)Tp9izD|92kg^ST03(die zVL5)d1aQd!D(ww=G1M{t^G+$YyhyIx7*7?{fc=Cp!^R!h$J-LkmPh5H$EJf|!ua0( zAK?t_;xN+7+lfZ<*VaBq+W=2JX{<;cae}co*H$jGdh7j6&y?f@uQiL7dK7q|$`2+1 zJRgiJt>>x0tz<8uq-_lk*LqyP{A!R!b+iByn@Ls`f#>uO*ja)b{@Nt^ph*DJ7&O$M z&8f$5DhBwN_anF$y-$cEPl)G{pny2UJ__a(-C%eo zL{}OsR!87;2p@Tyk(}x4CJuT8>R(TDJI>`Bf`b6uP#UWMM`=S@)3|ebm<(ZjuOfl= zhAgUpZql*0d1s=K^$biS-s_yZwvrEAkgIZ1{glU>vl~p@LDR_%pm(GW zptGp+IybJ3b8<0Xv0i%V^z5UUw)G%x?Z!6N&m|m(dfUgJKoLAaA(VlJ<>hqO=99V4 z^_=*Gd<senb)O1pcE)vlM6AI`ET#+tSYQ)2PX6VtuNfzy0+KLd>X2{RWy7RUK4t>t_x+2MsdbG3JNWLR6POB##auEU z*xWZxIt7AL+s298fW?g%=X8S?I!^cg=8&bDxk#q4EqQ`G@s#`xd4hC+K?bftMKZ^o z$aM{{-@69mTKYTL%;PI6tAi7>p= z@-%OdC1d-(q>glO=@=0n@cN;vm1A+;8 zcn0w)*_sII5v~T36Ijk#FVIHYxouR;aZiRlZjWtL2migAkf%V+10ms^6Xp4~qmM3OMU$bPn{p$qfO`JwXVmj=>ho0^ep|i|g@oR|59XwQLPRDstK`AEobfCG z3v{j(FtD)t{V$<->w|ip0HYr%g}O_*T#ftgNqT;SW!vF4iR*)0(Ybf^%^i;bv5Bia z$Bz%ct3K@vSCmdx&BuJv%=dOp81}W#Ucfceif_GuG`%K0_BGcDAK|)b)wkciaGCG<~UNmL5p_pw?86 zi=KQQ8bY}jBDn1HNe&z2TcQF3WNYLh79ja0c9*`8E>Z&+;vEVG%fWBgU4m7)f%tZ) zPiWk6=$ee<@Oc4V^|QZq9Fj92rtd@${@f3K15N;XXT}Y}_l%4ok)ZGpWkj=;TSQ3m z9*SYk+)(yn;yFzp-u!1Q#w3i&|HeazC@wgZR7&Q!5Mi&8?pH!BXaIy-%(+$ujCqjk zTgp3tzgG;h9)zc8?Z+S|d4j|E>x?|L{d3gvGo)h3I&=7i+pZ7sKI{Oi9e>nEoxJG} zj4J*;DaOkrVe#082Io5E!2I|Ce4C!SrmcKRvd_M=;ylRglaa*u=R*Y;441O?pQluE zc@RIPd9K}Z*E2osMWDykLokoxp{23$1<((RK4R$N2^NfN@pUtqlBK@t>HNB)Jf9Ft z(YM`Wu#(yDnZD`shgIi=kNShCLLfkaIN7FT#CumZVLhPwY$~O=e<-AeWk!yaC7ucAP~7L z^p~9j%$OE1sKa15V#k`U1Eeb&h+M6OALyCEfQpwfCfV^bF7CJM?I*83y9p%#UNP|C zD3RzfMBLgUS|0nls~*R7CwL;hQerY8yVx%b5DP&m3ILh(%`Sk)P`?BkU;ij~ydC0l zoQ1ixSoKToN&}Mcu5``uw0aKl^tWksQ_0_bYkW7E?GpvBFuw&rl%^h>PA($s;PQ6w zvKD{k?bdzQFsDb7oHh$q|sApnQ@1D%0zJ4MaSo8O*>Z3CV z6O;M_(g?%fo?hr943UP~jMGZH6AD+@GKEo<uBG{w1k2_4#9)kjyzLY4Vf@ zLNqvCOn9TRob4L%Aj268d&E?sH!+=0+trpk%hoiBTuQP-)`|PqXX>BGb$c~U0MC#& zU|^zxnn=;tqfL4qknoxeiB*mlU4l%E_>u$4Dy+$i+$T1kShUeh$M6VtZDEPHu#PvxNGIjvjGgIhOHeMST*T`E4-8&a#f@c9b0n#oyF zaz2}t+PDgL%8Dmn#kwYQb?s_PLe_~)ZTwv$0RK|L`mzo!soMgnz??I38rd1hI7qiC zWSgIHYQ~EkVmN8-bXa!RwpqPnXnkJGW{a1C=Ndbo?Cm#?QH?E}-(%ma*%WlmC7}cQ z=#(XB(TJUAQ@YH+{cKRl!n))8wIp&P^VqXUA@L54)Jwf235J{&jq`*meruW79u07nC5^a}bAS;twd&=XqM1MC zdjh|DZ8az`3xa^4!0h*gPx3^E7-ONfoT&lQhee{{8UN6R!!Ij`nHLRGR8e;iDLWTn0)-^rp4m|m{Ah_Z02BMoQXJW@?CU=d5fVCPb ze_0Z+#nFzbLRyefPHarBDis}6#@*dac%vH zr)z!QTYDZ5Cml2_pBYbXW00%Iq5v|S;o}+3-QJ|HA{gyJQv_lFM{AkRX-~f zz!`HcoI!oq{wp{WK?4b>9ObWw#odG({FwPwesjxK z0ntsH@AJ}Bx}7Ds9pE8X3JGsadCk>( z73Lq8dj*i^=l|ZLW`k6Wj~wU{FS(L+06~Ddx+}$!IbN-xv(LkLvo%1lX2X~7Yti3+ zEo=um-xKKC{991I`d`40KkJXL;HM^ri;4p$LTECs5?w<@M)+#!?7Pjd6gR3S<2u)d zu1yt0gMx3WL>S~JxIEY7@#->h{;bE|N<0hjrxksN2&o)POOt` z<>1`|GlOp^jvNzg)VYwm9gaVTdJQoa#H&5&#~#Z%bTu=LR_1yD>r*p=-qNhPuGK3F z9~s|K(9f2ou!9R)b<0o3+VOSgH^~U|hiTsT3tSD0h(l^*Kbtnv3;i>v*D<>s%?|QK zZZw-{KLazJ$+g-))WrW=4>nfLj9bUv#d)ac@$|$?ABM@M-qttD0xWS407iB;@m|gK ze3-h6@%q)fGYz1;sB!~(zcht>yx&VN($eXHLb`8MZseot{6lB|5DbDC3s>j8xKNs1 zd1#&Bk^b|qFflN3P_;wf<&o9SnSuwfsvHq!?~rEM0gyA^gX5Mx9w+Edeg3{3+;Uil z*>eJoZ(x^n%o8DBTF;5HUCfN~)0k0!P76b5IwI)p*=fAZwP( z0qlG74x3g29{U2Si6UgH%}hQNzbT+ubI;mWT}LE9DEfI>nK-hpg_*v|;Ltj=B$DX) zL5tQZSRIQrJ^a9E3ANc9Qfj9XO-{ex>oX)M-846yKz047P|+vIKPyD}}et zFCU*)0Yic-P62o;09^tWOB)cU0Gt-CI0c}%0C5VyGXmHG@W!~-5D;3pbMb%Uj|wVO zbslwe>DKj}c?Og_kM=>V76)V5m;v7X1A?WUsWT(t6M4s7cAL5&!yms#3xw1^kZ@Eo za#F3`du%6vWR%PKBMD}C4>sw_iYl_^C}WiiSAcnLohINYvSVtX6U7%5I?m)YyW^$@ z4}rREwvi}B{bw0>y-yz9&GRDQ!T_d0u7^MuJvxBA5q9{rN&r>UP2pje*y(KUJEm;q zEw314=GyI5tmmKA5BZp|{fA%8Qpe6W-So(OiBW^MopO6kf9YFLJd!kzv+GMVRJsLK z6OlET7wc?c%ztNPazOq^WNJZ8IqLCwKwZ#vcQ>}kp?cA&LaO}G!##xp{nu<|m!8GrtvvSRYIwK8U zo?jZv-=!itj6EyeugxT_$j)3Jeev?^8!tfeizr#uz?RXjh&fpy6oEdkW_21Yi{qBk z!1Ce3zSu|{&dsWf3Wjc?22rv3)WUlH9Gpj70~ARY$VZBqm{SJgi>h7IOnP^wDjz{$Kp#j%fB_> zH0yrIYYET1-;UK%;p#b3)349WqlDxityg7aM0N^?X|>g5fy%asjmCsyLF^dKT=cxK zvW9>%-P;yvM-LQuON%tuQW9`)Q8W7#YZ?7Vyib(nmUKRQJC_!hwZ~mq;4yvO1)tQB zAQP_#s~-8qs5_k++#2})dE$#Pu^go}q-%2B{m(~~KnvbPrH}hMz~2E9X((i26aLDB zHDzHYL+OD!VrD^>Z*Q(1UAdk?$pX}yq!okxv5P%eSKe;bkf#QT3=vC07;jt73Av;%5_sS$5NNW{m{O)Y`wv+X5+^5TNhNZi^n$_mJDZgZ38jJ+_gbzpWDRi>G zJl!g^U%AX9qDuX|BlCf+@1dR(-8&nrTxHPW#T6-#To5LxH0uB(8Amj7d&!MYnr<2% zcYd(4aNQB(KufiSrjh z70wbavZAZ}Twj^4QrTm>wybKay9D7$gT==+ZKCFGO6Y5GTMcyOLN)W%1YR2z^IP4B zUUuwIaRttR9b=)?N}5ZFrkTb$J+#r^`XJwD^E>&Yx0cS!de%Hl)H}?x+We183bXsM z(UeDx8^^ZlGu2G4-B^AFQ+>nFQ5RVx$`xyM_&D6RMe!sk&s=^;_U|_Co^cS5l~a8G zeAi`%ElUxr6ZmTmI-nQES5h+rHS*dX>{6FV*+Qlt8Iz=MC~YZ&1(|)G2U+?L8t1@& zOxN^hvdndU!(yNhFDe@jRE#SGcM= zbLPdvUi>#Vnf2|}8l(7U3*8o2Do<<*d_AZ$#x7!*2IieiIx_$-l`J5fXSv#pMw|o@ z@Ar7hj?ck5WHHq*Vqju_^v%4p$epIIScYyQ zgeGP0W0k{KcZIYC_}#hXMv-@1z4h?kLvxwO%6*u28ER@qhk|u0?&Y+y{oivuVBz7s zI9$TT*aQKgO<7Hs5`&|$cNwM%f%y)b%cZUM#+<$08QP? zy~}%xOm47ljAI3_-wlXDwUv9uDA!`oy;dKJ-uHa1oPp%COyG&5@1Lue9_W4K&7Wnn zyq5Z4`kI31*QibtVh@3Keki+sg`R}XnZW4k-Nh$Yp_Gd9*u$uvhwbS4+|6Qj6*s2) zK6gyivT7?9PBRb1cpdht$d9)uU1JAz7d{!GcnuSJP;)z7j~>k|sFb!X{Lr)Q6-QbF zEC=e;%73L6-S6*n<$&1>kOzqzUmX;S6?N9xJ3#Sr6y!I&Ui2ajZ`RzurYoN08##Sz zj;%gMRQukork;}f+Gk>T_gZ&$P|o4bItON@Qr}=DjICUK6f5I~R`|#V?`!Gxch8G9 z^EaV~J~6Vcv~sjsTf{$ntmfm5ToRBIrzP)T#D#4Pjm;Q(bfkJve#Xl_=$c+M&mXii z!+F^fDh#V;IggNceu}ydQCp60Vcbkg%u<~NH(ywQd?*>h**Wb_Y0=RL<-)`6rNc0p zXJAj3OtyWi=4$m~fqJ|XQmVCmtiN90_`LUkoNoE;{k^1&Gas1Vt8soc1M%@ouJ)NB z1!C4p`1;Of=bV3ktMj$ip{I+Vs$3n~A8>+e+ibRQ1&{y70>1C?s8`n*|6+FGwdRZH zbV*NCobai=+G~4*q_>z*;#9XDrzqQ9ERjUH6o&~Fc2gIf>bV_J7TlWv6>Gfpg$x)t zN2t;JO|cN$z9OPLNpfRoSc^#cD&5x<*0VJ%DmHaXfb*7ud#iWW95*kV;j$o4Z@&Jq zvkaCt{$*iNI$2MFkV){WP6l*LaZy?2jPM$Tq}~F~$B5$jQY_ozF1?4WPl!Ebw&!d& zS7sK>e3_7R?U>gh)Y|2~Yw=-1cN@qvIzAzg!?P9L=N6sn32$i*@_(k#N6pR?*DDK~ zo@_G|Tca3cW_UKn;6~vH&Z0{d?Vl$VUZ+`f#I*n$K8VHF`U4rAOYVn0hd=&bmz4OQ zc)e-4>>Z4qjds^6Nv5wKtw`ouFW!TGor3RZeoWsaQ87gJ0c`rvl$HQJLFXHy5%3cC z{R+NYeXq_&kG>_wKPF=5gqq1q2l?r4*@|f%Lv6WZ4;zK3x#_MfsJYsED%M{`Ge+Ep)GtW;tiOMZcMxiGYi>A2 z?x}c8?gf#sw*($2OOI562M>Vi){ukUv-%+R&8h28wf|KAn;L&p_Gile{_mWB4kdGfaWdAv z3;Jq}9(GZVgs7lK3zY|nX_k?U2+`7cSSi@dSTLCLGzk?8@^j1Gt~cs^JlUru8JJCy z^a9aYAF7V;Yt%7G4_?loO$XkaJljo%mP;x8keKb$=(Qp}bToK@yYa9@7Ekr-wbHpc zZ6DvAZH+V7Td2;!MVGX)(Lt?t1VJqaCs%TM?;xm(Ak>1|^(8#@<Mn7!Zu0+xtGtET8jBSh?~3npq5|a$I*igI02j9 z0_oZe3zc^QE^7(~dPX!0+l-ZiPbScl9s)3;XhJBCU`8}Ju(q%*4K6;apSO8_>~=gy zT4{>VEYt*)@^i=?OF?Rh^?DitwXPp`-`#ILhO(uhN+y0U1d7!B*Fz1Bgng(G7n z+KtD|D&yIOAqBCdj#rXUo)em8oC+#qJ{GCq-=lNaRiKoI-X&;|{-29;Q%B#Z} zes7K?i0!?xmQNpkc_F&PZOvm`K4Bpnt$Hmg^a3N{c0ev${Wb0E9dx^rCM??CtjFPz zgJ@@X0YigGvLr?s>-*`{UqL+uxc5NH5?S)A{gD{+m3F!77)!o8JmS0m!Ox@r+OZbG z5l7MO94wm;ED)zDvo{py62@i$EscgIB)nDgyZ^CnC7NR;AiPhoXA!`HZdBJs9vA{7 zhc^+EN8M1|Vls0S3Fsz++yAXsuQ*voKg1x$Z_a@P&ig;lu<6x0q2IS-1K};Ee$jq6 zec#U6a)F@!di<Iip5IqLu_@E=HL_2vrV=?VFa|g?R=b_(VRTzWu?N_14B~H z7kjZg5Q`AA5K-YI2O(dgA0d$_O)e}!nR^u4-!+jlzjDckjg6@TAzNe3M_+7;kmiw+ zb*h?CwVkQ=BQcfAb6y-;Wm1sEb|3ReS67k}aeSrV6LKaX)gqrd4}E;&g=?~DOJeet zYPJcB-9^j2CVjc+;ySKLa?>!04Hv-DO3c6U^+j&eXH=wU!`OGb4hp$tgM;z5zs?qjw>=WAA{ES>JBH37`_Da1uXP`7s-Qma5=iX)`G)JX7d0J3tACs zJZ1d}G1O}gRtuv+l0oyEQ*{rg`k*UavBSU@4}|Cs;YZErj;+8pp*YW(z~b;CDIP;Y3}Av!;jHoowwR>Z$@1%1+#H_2qX&}7O-0WI z6%~UAtlQ?D>kif3dIMMSc-hNjeNMydLSP=ZD~&BwH#hJR!={9Vvy!0_F@uI^8V=8SY4O`eRob6PgsKY-jC5CImpj^M z4_b6M8?O`)T>K#`FG%_XnV1dwPy6P~FO=0Zqo&cG*NxO`!3JGHXE$Ds&o8XqaWVXE z`U(6Ez0r%j6s2B8%)uo>M!W{q`+}ua)~77pXr}ccXS=o<6j6uX*=V9e*Jgw^vaT)E zwz1tI8?(}n=IvCz$rv>h^6+O+rA*Uj{gVqNmq+y2#XiA2=9JO32Yh-W;}|*rL?C84 zRDOyVNH>8lbWszqnbc+hT^tt%Fm2lksqCfOfjY+HANIubhx}ylj!g{S20MlY!4Eww z00{dT*(kOv{=5H4Oz^5BM8KN##!)o98~ty;fk(-lJ2+YEqSpE&q0S#0LW}k*8(;7| z-%KN>bwqG1bI)B7$p`wqK(OZ&xfje#j9@MtE~HWD>&nhuxYO^3eJuNY?YzQcCq-oj z(P?=C%{BZCN@*p1;D}M)jQK0%b002HDPQ~aq6b18A+%CxDCHeaQe4(;7m+FFKf*k3 zaJk?~so9XWGOy|0 zS7X>s&Q1g`XV|8XPrAdN38Uw}a$yFKCb6B^dN*?IZy5LnC(_*l`bF%r?A|bzkuCIm zKiq#3Z6ZISYl=W0R#*NJ$ixFVjtDTMC*Io7_-li?-9`@HH5niTS~>^VsTB8Oz$z1& z3f@fxFQam<$MIU?&`0FGN+@$ zD<_loE_&0_54=6}z?j3`)rJqEqigeAj=As|*VOK)2{=@@Q}RjQ>_6~vMnn&E!%Zj= z&t6Bo>;wrqULbD3(oU=8wQW3NtCT-D(A}zLMn8BJLmi%E*54U;sa6B^6#5wke-kA3 z&f$U`vTC^F6{ofK?jf%oDofc_<~4&2BVqgyl5E+}=cdWXAK+<>zkB9Blu^Gu^Z%0X*xgGs zX1UZygq^@@!@k4!j})7h#*S=NYui%D|8{>{eUor)KyH_)xSsv?8Sa+T>$r=sCh@m4 zItP2T8k>psmXI6>dcjhx6E4uncQ$)%$;7lCl$5!6Gg`C=#Yb9U2}D-ZmO4A+h6s>O z2gX%ty3F0ZF4Vhd36$vf>d^VM>Np!H<0RR#m^sZ&tn?AwMvQYKV-xa*G)LUu!r352 zKf|083AJpY868fA;pbs4e#5Li{A1d{<$v_Z18w+)ApCcq(gozkh_uyc%YKG|;F*LR zUJdx8Zq03@wzm1V5y`=KA=tG{LNWKEiRn=^kF{fLl%G?SLpyiz}86+0O zY8de>+q)qIb_Bv>5os7yuQ&)YJdpMytrnxu-@N}2-S!EwzeZ@g0jjTtPhk$I71g=d zr&UeJZFdPQ@1{a)_fr>CKQ%abs}l&BMybkc~i|d&${nJqLWcIwmE|9B%(j(l zL}lUUGXNhdk{2!24m(2y?JZk+m@Qj242&pB3ub!a5_UNN|#fFJ;KD1 zE%0slE|x#%Y8bMV;E7$J2>GJss%~o5orP7_s_cnK^IjpWo7Lt=uGXlH-6FL7^LZ5D zQuLrWLngKkf)0I9%^h=pG?TLu0hJkGpWd2e-(NI!ocdH7{F_)=)0#mvFgou7wkoed z?i$->eNk^3+M+DHeKh#2`Ek+N4Gu%*!FGk0E6Ml7jSi$zlyr);$FOSjL5q|JCJ&w7 z_8h)}K%5ACvzV9;O=|j}yje7@8!TwU<`ouK5t}Jqs%NJ8$I7&^Ha@XK8C_o^pRj7n zO*LR8tQOIU9W1Si*pc3~#Rg@4W@*pTJY{YBmf7a7JN?kVe00V<%}R36TFw(NUblmB zMteaLtYa%x^raO+87VB(ohqwDid119@fH(bb!6E#6jU;aeuQaCW(y6`K}^>ZKM+vX zn0t3)uSf=yg(>tK?E~Srbg^e2Q%pr~wF7dc7AH<^OEvr+Jpq2U z%@Y;NthYp0eUZYt+Ui)11 zlz^wodz%NdrG{xiQaM)hev8t#fnTEvK6;m-18GN|IT-Yi@fVkHlcGw;`Ga zwptV3NuHS$O6>0z#K@zkwfE+ND zBrLJ2nHe!gYJrYR55}i*=f)(x>xB7i?%X)qb)=#P%#9JJo59@pbgl@`GL19$Fw)yl zPYJ)jW%3;qn0=J+mp>uqv&W|&>4PP5W?=g9bdLtpPn9E>e!P~GqtKI>q&zmK{w2-w z!@K|d-6p-s%!vy0QK(OBFfUAXGW}lTK}UJ4Vl3b(ntKKtg{2)PR&Ri} zVP_>iA-{_0#>!E82G}JW>{;}4Atq*<41)?cAwV=NAa&}1V{Bso<1n<6XEU9D_79N>ZOd=iGN(4IsC{HkH5_Wzlqn`(=HvMmNc60xi~o!HY% z6Ul~^?zGYCQ3;(T4y?jg3eN?u*l%LKC`@k6=r+y3`cP?@Vla4w(YZQ$-eM~29r!Ha5H51 zMX@CrXPVHWHgXBNC8r|Og1xr9#NWml9#{A%%6H}-W$5q#rfK7o0f#2^RknF7iKBBQ zJz&suk!>x`kZHI*v*E<*_ZxgTK-%#)o(-`eb^PUYk$}@xsfJ{0PB>k0M!V91(-j~@ zCvcrEbIRFgeT#kITkvFjGx%uChygsikhL4|Nt_y9COC39X4{6AbkgsL4Bwje38AZt zz&-staEochvW0E9pzNkdM%-GvK5I}LV}zv&DMFZ9e2+Tr7W-e;N{CiVQ07ef@9r9B zNHOH1kmoe02|bQNTCJ;B_5c)elbFZ`C`9)4ZEyQ*09H8aX9<&W=N-@e*3rYP;lATw z8on`a{2do5tG@v6cyV8~8hFQ73y+>ylj~q&f-?k(-Csuq_VKn6x$%eSoZ_qSc>G~R z#+nY|5BkYxK>Wdwm;d6;2$lEJx4CWblwmj^d5D1?KxI8p{9rzqWV?XA?#JJagsJcUZJ~gFoh#g_*9`SE(z@w4N{5S-o20y< z#$_+cPTE{QM5=MHX?6tXd*v_27xt1lwgU}WMYpH zO{yWehtNzAbyj{u{Rpop!)4KRsJ$1MY_bAeFU4GW|7PooGX?i7ByS4hwgE*&Ar=Ry z?5A+vLbO5$0Cg(_!9L0ffYnR^NmL?jee+cTzhidFRY^L|W3`91Jnycry*NZ< z?T$`@G^_u{T?92>H#gJ48lP~BPl($b1Hcrp=^!GUk-He!hB~Rt2TKwI^0t>g6s|n@ zqPEcfxbK1;E&&OfcvM8qJP=^TEYJAL$?#uP{w3zvF{InMJ|sV?d1Z8#pMq_*+~pxU zF?dOTEg(I)Djj=npSEGmdrc*l0amVyO4~!NN-(aFQc|CyLunTY-q{pHGSCi7h+uKb`&O*&3+aQ-RlrvH-=xDED z>N8G2Vef+CR%Vd!Ah}q2!@aZ{E@r+B_J}Yf2r+{7ckmn?mJC~-6NPRB;n~bQS}}VC z^pRWz9sPvROHCSd>woD&?pmo^H5kFF%B^1?coXR%O$SvJc)1f`W&uOliAKSs4k>*? z%xdnznVCN-GbYGMkPO<;=9ZqW#2*_tiLsM~WD7C)n^&_j4QO7iFq@LBmCsUjPr^q_ zQnDm9*Y)_$oz}W{_KdUB!jO)?YF+OBN%^v=sZAs~H&L2X9${8~qnyFq- zLX3F{K@D6#J)+*ZB3+kY*Cu`8)#NuqaIJ^S<8BqkCm+CiwiZ^A1--S*lQJeexKf! z?YUYi`I~wze;2Uk86?8|N-PcNEq_DhZRJrVcbnER&I5D^I;KyPzMQdO`EMt^|lcFK%56Dm7otBdYR6@tuX;rfk_x)?v61TrWU;vl!N1HG$oJ0m_FwAWx zO{%ANV00UenpMXS&*untXKft2Qc1OW(0;Dh(P@cVXXJU&LlKL4QF40%2{%F2?(rXD z+2Du03kOEAG-|dCKfFLtn3AO|c4Z^@Va=&-M=LQ>#|aIU!zUI?p_KPr!~@FO52Nq{ zZvpxD(GHR$jhDT!rU98r6}3{6rX^oIQL4H=( z_E{vKA9WhP5COsG+{q*W$RWj(3Dpb-4qRv`zD!bOY}Dg(?$(5BKdcDZ;+4ocP|5V4 zd!XO5B&B#yQ=0VmU0-n;gGU`l#eE=*_s%zffJs`(2;2p_r;Lpz?S1s)Ltq7_ZI^<} zh8&v}{v=Au+K=YE0<%?G=3}oa0)Pm9`;YhRZJ-3xBt{LS%{>Z?_}}N9oPRP8)Zr%I z{Sm!fg#BwDaO%6Mw)yw=0VW3K!-K^EQUl6xBVp`ZBJqtF2L|>K5UKO8u>r9=#R9%B zBTbv-!3RH;W@3n4lML^$flE7n7P{NSSwecHN5{wrsNpd{ce~5!Z5Jp_WRV0`&22>?cI0 zKt)$lD>ezw4k^ynV=Yu7aF&I5phF5kq!S&5uG#|YS~O5O*Fu!xn^~qYriLsjC_k3H z1V~J60E&sFi)|`?J3W1wx_AbRV1B?km(p#tV=a_{hEvB#2qKtN8j_&FbM?aDd8uqe=ZGcXDryhMTAj}_TCLlH=m$gk zGDKfxNgdlM=o;sO`F@ZRCqlKqyyWLwK>}A&2Ft#$>`Cs~lLTy9cNb$i=@SAxSOqZf zWzI%$b|ID{OCox#=IIB>ba^y5*azy_lgd3 zxNt1{1lP(Q1{V>HVaA3{2Wh%*?J>5<9DuMFPQZ3&kXZ{A?KsQ4JhmSERTEF`pVT|~ z@@}R+SO3%Pg}glhVU9x+5GGnjn>~#iTlvmhSt0;dBHwm}pH9wD65pV>ryy8NSiJaB zG8FTJwF*Ptw&4ck%|0QN65s{`3&P%Z90Pb_#`;f)kg87zdp!<$vf120Vs#$CK%hq8 zwUA=?X3;nX`hIvm=3&N$`>f^}GvEZ00k{GIsE>CepiTiLF1Y%3fTkau1d#xMFF>&T z5gKuk(GM^x49Ue&J_LZH)^KqYAk65*+Rz_yIhsH2oz&G;e`1Ztflx1K)Kv)sfdB}B zu-@*I#I_dz^!{LuClFRX6(B6Yg6Y{)LLT-KLyQ2e(Mer?DnC5cPXDexpaxQ2KYDKD z@ctTNBl?<}3S2-^rs-W-*GwJ|enLmRs43jMgV;(nddGBHrm;X-^}$DCK9X?`P>jK% z_Z?mYpsDp*58kv5^#^7|AA5;P>jXJ^9oSP4{RzQ)p1?j!cAb=$g7}$j8{A0;qxsSW zwk@NXW#~-Jt)EA01-^*4quw)4P?O32dSOU5w_c=Ri`Qc3+Ou0udZ5>jr%?Z5`;?$8 z5XH8V8(X9Z;(+4S&_DsJ44z}bl45Idf&szS0?N!Rl^G`X7Wa^!F-1PqG9Tzwv)?gK zoo{}RCmOT%@zy6!LbY4ydp9!uAW+4H7nY}8prM*Hz0O%4Y(7+(ru4+><-XTfkgAn0 zG{u6X`1s{TRrwIG`D!8&+zW{OB?fc|QXFPhPc`k^&F(B|mxp`Se#p(^#nnZE&@ z1qpP^$B~|gGi-Fr?2c*9UEpkMc0*lY?&4Yd6;zb!zZR03_PGMX&pvqpsZW>)?+wkQ z(pMb;KCk~i|-TxYVo@TJBK zzZVaS_qxZl;D6FdBtX!3#sPqH9}Vbnt_5SqOw6!JO3YBO_Z{adECXZ;B$WCWI*;y1 zDjmK+7Bsect~)DYX2yo3iddi5X2-YHEW9_|TuVY(I9q2_%!;m+n zf1KqQz;&NX;6#!;_JJ2jbFuvf;6 zhgTfqY!lnQEKF8M9c-v( zsb6Zjb-?skp0;XP?{2VDy^UafHbK645n^l&SvhI*MIdD}sRgN%S+qKJ%v4+c*YyFV zC*_rouH-5oG2fb6dS;LRLuj)};@reU0XuC+sYNF*@~ba8&5d^#$@8s_v_=?zn6)Oc z>oe~p{B`Q3+)A6!y8-p}ixjjJeTDS5<*#_)!M9E8z4IcnX%ce_>mYEEcc!^F3+^A3 z^Ahu?1z%p*7gAYcI43l2|E?7qq9soqZL!~bo>BrFq$r=ublM?Lx`l%e%9$Y-16f#;kI{qk$0>H?OIi}d&Jo8kli+xWo0W#dn^ z`~RDEH#l-%*V=NAnd776xJ{|0%G%+(jwtWf*KXLdDAdYI><*twq_y_iZH5n@tPY!5 zr_s23TKJurs}*k^s;8@9jggJ%*3^wo8?ryz*lyM4 zJS={sU#wO!!ZM4u-x9*U%?Tr|MqNWC%H^QNB14A>b83)+R{MDdMU7kwy8E2uT91@^ zCx%?ReVC=P3AL7eqB6E$RhM%+N<(INtN6^@_jJmveK(-a*5I2BA#;@P2DTU{T2->i z%>MXu*8XZ!S$hdym9>yK`SzLAqDCt@3;!2~0t2Z^ne`hq?E@}f6PX)bJJ056a@t+b z#e*N=@KW_4MaMs45N~eWQA}z=Kex*{$0%YcF*anc!fqTsA`{oYDZ2x4Y|Mw=HT3>N zWWSEq%R8Qr{P%`?h^;9s%w1+Vmxi5@i2xzw|-*@ zEY85_4@i(f@uf(s0nz9O2j9F|I;~5qfp6y~BP{;GB0{s{Xyy~S84}V-FK)1RuE{J~ zg23&p^~<`E)Cri$1DENZ6t^=hjV`pv@RDoi>h^9a57j!^7*xORYK3=%tb#N>ly1ft zcm$WCc+csHbv7OgYE+G>*>~2qZ;@aJ?k?)S)#lT}c2DL9-0pFW+D$%pz;QXJQ!r)WxkAEe!{t9D`MJFGzaV=XRd}+<W^g7Cm`xo*^aj|FHMoaZRmj+AxZWD2fH_5CsGQ=^!E{qS8gAOO1$h0qHFy zDk4%NAfO;bnt+r@ml~QB5$U}q^qx=yDZUTxv(L=lXP^1bnK|=&znORc;ZmN3D_Kvy zuj{%?j1kt7{HTUu3%i1f86+PalUmt~v>>gveR7BF?|^4~15&o|%@|JCr`^Z*Ldj z74y`kCWl`T4;4L@9NDE6{UN-gVYZlZG0z=JlD?P6TCaXtCEs&MBD$$ zQ*|acA;VQ<7z|Q=mMlk7Utc~s7i@F7u_zxk{katWIQRVo)#(eu42}TDQzTrKlw|dZ z0Fn&XR1y~6>K=Ku{X896PWeb2r0Z!*;|pu&&R0-x)61~P(L$cZKs=BueYBJd6=*et(1N%{8K~oh3$0f1pGSPg!meWI zyCo)l-YzXWp24qZm#PH^GDVcEbD<}VE0gi&SM-LCKr4ss^2Bj!2GaSl+OO9PId9gT zX)D%ZS&AsQy#|ai|lY0y*ie2CS4nWcFryQqax+ zE{XWR{#=X{shv1l3!z`T;FDu~+VO+-hPvpGpKKGmA4!Kj%aeVYZs-DE#NB6QmIWC) z7>PR(WxmXc7+E4_aW7#Rw-9kN$Fo?Ys1&t4P&0*hlY4QsR+ypyKTABg@tPD*z6n!H zwP6%QC5u#d78LVcL2oTf?!^*;yP9QwIg~8npzG_X$ER`cvVI!nPynsjbWOBTTfgpm zYu?cIw!vOLUNtuTC{SwF4T+>Mzg>Tk3g$H($;6!VvdDbtvoMtXgZB{f76-Zm_~GI3 zXyDE=!wA&CJw+fyEQoj(_N4M~@m;RG&!y0j=*emz_9LL%Ac3 zCEyu!h7qW)ECT3MOlE%(U z7^OW);039|)f{Qc$-i&F0DtY@JOlFEiJIJZ9~uqo{U(?C8p%alr&;A8Yrn{W|I~K; z3zWoY4?3RZZ(3N#k-vM&@8*~f6T?dL_Ex~3^YqVo`sY0Tb3gqTo_&Arr+=RN|HIFH zYa1-#F0+4c2VV<&SsL%$<`vcJKj^T|B2<1*x%oH7iLOKCZqhZJa@bdV7l`WTkKqB8 zd$>`VV-S@KzDk#(-{r%^eUry90|#-Vb-bJbH;E2t7dM`v!nZBu3%+}t_nMOqtnpDw zi3|(fg$8Tt_j1|JSDIo85)iV0HAR-1D#wFp)o-^kEN#xR{`gpYQM36FcCE%i6;wWL zrnQ*MrT_g=T_&#g4=d`iFrD1xU^@DDfKv{bvZb#atTb<$)W2IB@or^GhS|dnp$7f^xGkJ zN}P!S&l2Pu-%{T!uOfRh*3^*Iw`gZGkBB?62(O=r?t^ZLsQjUoEK~&m;1h-oUb(3` zN19;FZw)0k%rD00+s;2ty8k}tF?bmunM5ShQp%tp8ZRwDqx3uu|3MdayI~DuMsg!| zR$*ClAuq}m?0h|no$}zdj~QGhKQmoly3RRLDe#%ZI#!a*7i;_qvaICB3ni@{HT4*0 zetIpyA3!pDoZVosp98%5#A30?czk1XNcf`PY^(;uv+NWuJY|F1K+oXirj6>|7_~Vp z=IwT>0z8I%p4M{`azv1m01H{VJ3sfPbDiO=`i$l`$H8xO4`SUs^`5B8A3PP zwK(*Jkm~NL9#=a7w1Ko2(KQ#5j2Pfz~SruHlM_iJoT3%iz|Af*cJ)f!-h48h#pg>m>I zr=o7IqjuI6$w4wJPNak^yRMr1rWdY7!atcH8@HK?d84lg7OFUFoA_*TE-W9RjX)0+ zNWZ;MT-q&B@y;vOH@r{dy=b1*e8;ChK+c3vo(*3@b_kzf$GlbyDb^r}xr$pDdff1L zXab`#N7_GYBQ~=ALX%BEa@q~V{r{_wt_bx{nD@spk2NhBF^WCtrV=5IUd+C5Z>-7d z!D;=Wtvk~z`^q_gw{NLwlIU><8x8Ike*WarKK9v?li=Lt3!_^rcvYGFTJ4SjCKhmY zzRP2y>?9BTi5O}kuEMg_pn!pCBk#o}UpDWIjr)fjdneg81K?(4sOPiMy2>-%-1icZ z(JMpd^ib!A6QT+4CLp$CkaLgvJ_{k^DKm8Kixpn_9CxzW>8pFF{n|1Mk06PA*z30X z${BJ|g(+K!TiyxW=8@$B(5$;LvT{`Yj0KR3-izn^zcJ$aO~}H{MKdP?4X?46Bs#gT zvMKcDUHQ-b`_3uGm)9Itp&e0_B3E36%Mc}4ui0srG5BSMWu>!!=D}64h*|=qH~Ta?dk+6X#n=mteKh23NkMk-~^b?4ft(k>jhpRQ?GJ z1BJ0m1{_Yu9a%<(jN<&JP8B^B(Z>2oMw9~tV-bH}L6z3cl+xD_irW*6 z+kJv-MSoM%)+1lIb5=g{QupgL!J>8~Ol$!8uV7OUzwikbe^MWRkNVhQypNL{zU^u+ zzne_1?--;9R2+Xj_BTeF@+dv!m^`E}#V0re`?nmGuf*8spfpZDBqzGht|e)xWS926 z=T&LYwa(X&-~$_dmSK37XdZ(H;WxQn%Ij%~?&nO$w7`xbzI#hd9Nw1b+IIUcr}MVF z5!Fiy=;fKc3-TLL@yq>bUsT@DC)u?!sazr}JP(0=z2P)%lrU1}EqP zQ%r(Qnk8KOj^H<7q22iFNt2N=*KOH z`dz~4)pcq^A~iGy*#?@bkYkWU^zWYn#d+=Uv`KzAM+%z}lIXStm{+FSA9UK1G-c$* z02KEIIcEuIF=pDJ(l3v!{_}}RS4PjsK}?l7i;1HzKy5?ibsu-=EAP35W&8EB)=Tz0 zK~p30uVYUh4w02VcZO0Gxnbr88zzMtRGewHrci1Pib`R1m?x6xL3T)Y&iSH{

iw zEy644bk=QQC2|$YTjO%FwyLh%>aFsCL;APnI7b$`LH$DNy`)M{>KmlpK}e`mj$hUn z_P68v9|&s6U8vBVBTh^RpL;zW^2+I*FSF@ITd~@f&h)I-$U2pQLX5 z8W)}qm&dE~*LWKbL>HzH%o@U>h^ZCbP=+Fv`NghY0U{c|fgm{*djH_m#T_^G)D#Z{Dr96XQf#6^u6HQ_tsf z>MPa2p5YvAc7AIfnXP5~x@UP!X@H%BT1I2F*4Mh}=#^pTb5+!)N6Ad?NuxH@G|KOf zfQ8>IG=J1T%S_X=*zh&BuH&mY{B z+Wb$|;6GFi_?F_p@v(gJU`6DljLvbfAel@5kqHqW*0 zV}R8in44dTtQ%#bC=jBzp~pPUbxQiJ2YaT5Lp)B<7E$Gu6%~s7i>Q9JjfCKF_ejC; z>^Jssvj=e#VIqY>9H-cw57!-Ey)^quiGe!A^AYzXp1|f`aY0kP>%0PKWx9gw^7sT(AF7zoT;A9m-j?|RF=JMrQ zYW=L&$!p1CDg{62oS`__&KYBsiujH=S8&(S!O?MIdG%%I3OzL2N3)sGRgkS&JmIS& zYJ3=STV0>mJ!}+75n1CJC1^6g$SpdlLX=4+H`zF%dYnMw$=_3%{Po&n9?x_iySgLY zOXId>rGIBB|Fp5L{JB#8!tL*RF)2j%!$uzL69!puv{;^fal*%wcT;RkdhB-s;Rzq0K{=-e{^nNJed z#WG1AS_%9X(kan(F;5%LsUJPXIk$gJVm)Tz2Oasb4dkehKmp{GJVQX?tZKvF0jU!) zcRU+fzTMWiuENlJmGuiuApR;@RVYEuh}sYt()h96FFPQnARv0a8>00GAmJa#q4E<{ z^rUGU#vVU4Y?FP*tnovkCZ|^_y|||GgT>UOS8W##33{d^MOE>-cWB1mzr=D>O@U65 zWnt&pp{?}Ur*n+(DAa-+D@C@KV)Rit*I-r;_1MGJl72rnm=zLAFSnceNyg%2oyMS1 z?Eu3Yh!V*sRW8F-iFdm3NV?pZ2&Q#8D7L!0+{n#&G1|lOfkuUoY3?{b^<<4HFR{Kf zPYsV@ACaI$&#uZ=WS$UFSX{}Y>8`ZC6mny`;gv#Wt=~GV(&xz@HD51RAE7mzpdne1 zHC5DCR}$CqIpStp?EP?GC*7$q`~`Aj3W_?JUPpOq{I0>|g{je-hqCP@tYnZ9b&nr4 z8B&ivTnKq_rlp6fU0Z_I7#XTaJ{sPt#TG}vG=$?V(BkNn#Hhf4exY09x z`7C+wE)Ge*G`Q~=Rn10VoqYXP(d(MgJ+uZ`@ zrOy=3Q*siY$c$Wy;>%eIyBg+=Q~rl~Jk-t(bn^H@NbYuM_nsBL3MG~VSb;JKkiub7IBJlj}$=2*L_)Q@8rRqovhmI z?`iR1jCVP-O27G9QKU=aunN=DWY4|~&Jg{AIn&rE3`FePgsQ#iH|k!(CWYnZ_=lJS zN^+UFA9P1P{-A3%dU;C_xkq&lB;bD*2RFS-Pco`?KIgk!pezQIzuqa=Ya7XqD8Uu^ zb*aqY@k93o<$C&Ze9R^@{VOetKUTJH%K~S~y_ZQ%SKt*VYc>+%KD+{nyb4?AAOvmZ zuoKd6lL;9O((=y8wF4ANWL<0u)_dh9sS7%&KUnOWn7^Lkl-L1%SpvR_$Tk#*B=HxA zt@Vb=D2@3FX$M1PqQy!prN@r!^?gc!ZETYai0InwsNo7qEO+(?L)pm@RQjUI21oRR zcZ&`QC72qGMRMPZMZB@Q*_}vJHPTs^!uSsQynEl7JLT4orKw{)6zxub-=6c`MOK7A z?x9_Qt8|r{;2E2vJW#ju&EG5tRA0D(QEv&E2E?+Aohe6YmxX;!6TA2Hl%uolVowEk z29ENToa490**kpx-l1*1{MR}JVgq0QhKTq__n80o-@6MNHJWvoJQ?F6 zP?243FW#yAI6a8bQm$&ZHTyQ_w;y!YFw_}X6?$)F@>R8L+j&<*hXf5Vl`w;{=f=-8 z+T1v>vJ@x|dz1nvs`m!T=%x=D%X6VCOJ7ks#%Tt1*Z9KTF&%!QJIY4-@Plq6fi&p> zUr}cdJC#}r*>caj=}+fw$-_fUlg%7jTa+7B$$j?y`^9Gidl=7IU!~LU>HLtlt;$b| z7~WWcD%G}!X~-)e&)DhryTf{CB$e79!J%ml7HLLGHg>1B%tOqHa%n@W`Zguv4|~sg zcvu~Te_wE88HyFPJ~9WQDjv4cr+=7ekGy_3DK1h`o$gwYgZgQ45LWO%@>Yi383EKw z@vX~+fVW>%(2LpQj)>{y9@M}c4sT_g*DgNq>kGAfb#;;9ZetVyXtTN&xU|Jmq zRE0;FIwzoBZhJM?M4HeO{0l;TJ(EjVOYA?I!Z6WlK6F%xJny@SIOMeM%-L=&~wkCaD#alJ>$#B|8*|D8#Fgw6pGTq3U z`U<5P1-#RZ=gSB*e;Zsjd#`$Kh?1j2oFck_amDhaeI zeygf-Hspntusnv9B!)kTdV(w{ir+mXYNIcyX6Evlg0Yj@rRKeG@$4Mh-^X1YK2DZb z9zfQq%S~|x>V4&v2YZ$ZH}q8wg+upTdPNG#NJuq5;S7wa6M7FQg{0NhUvsk2zaxd> zN#*oovdB7@y%wF9Q%m*_vqhp^ghm@GGb6j%f1?t*xgsYQ(dPu+LYL#Z!sy& zz4BSDCy(7Losuf`{!*T>S7$Vq|_wddhWxMAe*pJ34MM+m6te!Wf+`y#0$7w8o z9wI_#5jk_6QkApJ{Z{;ewds{f6IFC#W}4(3Rkg%yzOaz4@zN{Uu>|Nbo}O~H98b9w z7f$Py?3Op?jnA|OU-|GqLoTNJdcYE(ggbFDunkle?P0Umaib_ zxsPlOr9tc7BrNTKdWMXlG@kBMK* zGqe&~Xuz>AaSXkS2KkVCxstd0^-wtPiDt=`%G;BvQgiZSR+Sp5KEp1ML-tw?RqT*Z z7QVuF&cwO-a$1&M>agh)M{UVR(yX~>Dn|DNW~K<3GTjKa>5E~c zyz8w%R?8LgRFaOn96YAiYVi4~s5kvNEhkc=ow4&qpK`a)m%#~BM@ZIWTb&yIc{Y-V z$~hZ*DXH=yX=gXwnY;9>aU0#l)ag15>dFXnKH4oT%se-6pSy38Nj5*)>6b-ZOxW3R}I5ATwY6q9^+%G*%3~|-vBF>^?DP} z`&7&uqyr}O&={2N3r<~dGO?%pda8x>;Gsirt^!+{9?H{+Om#JzX+Ekap4_!ijk5~D zoMG8S2Dp31EEOsi9 zTDHaQ2H)(XXyK3)_ND}oh}3QEixvC;$%j24xglpnAzpD^?1Ld7D}6N-g)ZLAhQ386aL*;t4%KR1 zr3~)My`&ZZ35^KUgjAsb#F3eka#Fqb-gFQWPb8M^{yafDe_(6=OEj`TV{v!QJYFL- zg%E}k;!o+5j3E17sTW2bH_}eNhCU!Qr{@n1M88L$#5#Zk{`>T-3ms1k+MaoK`0`r* zMl??+?nNhH#@PvM*(SKxj{YUxkXO!E8)Ske%2kNk zb%qlWj%6ouMetWyn3!*`=df?D;Aex?oi3+Q)40(WFD(R3XA1PUwT#(|z$E2FO5Qw0 z#a?$>xSAgDe2qOA@AY&^1%fkned^)bq8A~tmK@g@b-XmP%IQu?F7cLgLh~hUD!(8Z zmyHGtWxdJ=w_?o;uiA7|$qO$hQ6j`lSE|F*Rh9*K)&f zF$c+w(4Y7W8LP z&?{oC65Roqg^GUu4=z2ai>%|N(Bp-&vTIvm`yTMTkgW*SXY*^_s3()i zwJ`|ho)|x_+!KXwe$aUtTlw{r<5#PC8Dok(m(1tRoMbxW_BA%9LPO_%N>60_yOKv9 zDZ>ft$xgWV8-hh0=6$-7oes(Kd)}^nF>e<9V!a(66ok7mw%yTusAF|P`8_hvcHgSl zY9_MkY5EHPl-w1*;m5odk1(dT`xvs-!?}mzlS`7Dn3U~90@5{ya;qTPFhyDmHaD#f zGH_h?sf=t&lUdRd#D;h{@pRb!-ja}WC!tGIl^}fd8HgXRmRi#|Lef?v#SpLblI4ZE zjBN}L9^*X9u4mx18AzLL#AELi#B(&?=)`^M%o_WY>+h~)Z+|LGoc-zbFAt{M&s>7J zfEZaz$9_8y_?yq{rx=MFbaBcU+Uu)y?=xMi2e(vH!vV&tyDK6hf*Sj7cV%oUC}}Be zDsb=rT?!&AyAh5K3V&*A#{4z^qNud%VglV0nJm__Y9dwd51rv32Vumc1RcyX2ot-H z8k?oYNn2$(tyQZ$bN;$C0)%-1Bmiy4ou(bcoV+ z*s~E3E%GcA(I&^OM{nA>7GDb?^5+YOuWhw>Elw{klCO`5@Zd2o6I-nvf0;_baxvFu-qsOT=;cJTRbZ%6bB() z*00ZIpJMIF7C&U$bk)M>N)X5g^>i8fwK9ri2X@VLf?mE83%fYMArV(s&sCm}UbOYC z*`vMP#xQ%46?i2wB5hIr=(EXB;TOjRL}?plQJt|ktp(o)JZ|1!=>BhTJglpzMcnu_ z0adNkkv-Axfj^jT~Kf;T-ZBbXQ6{e(L4JBwDvY8^r_!Wep1qW*EDoVyh8&3kD z)SDLFsq1uC;{(=dX~LBiUh~c(uXm|;RZ8dy`)PGZ^|V$vU1^tb<#?B3ia#Vl1JB;j^FbcM4I z)YDLeQdtFjeaNi9^)#O(NgIW(6?vCikwcJQwng({TP(E{WD3dweElhu->+V~Pyw*C zBFl(SYREM#&0p3h$F9Z_w3Pl^OT6z3eov%7B+Yy?m9}^L=hQOhiT*j$=>G0iVl`m# z0&rvef};8*1GJgf+a9$Vv&i||rw$Wa=Klv&siH-vKG|Wyz*K~B( z`h+%5BZJqJMl4_}yl4`r#gUL3D|S06pX)#F@n7zaVaMW;D@GDn^2sz%UntI&dVFWJ zd0=3)9cf=?L6rW70WjQ5lE}B*n~U-zW%gE;Ywgn;|JAL0I+cqq84I` z`SDN^FO1TOCI=!Hv%O1T4|oW||bjSLjh1Du6*i~K79Op<>xseg3cRR?zfc_d)j$Uz7d z!sY{|9LLa<6N-P(nKOY;P`5V~AQsb9!U})TJwmR*-4+260Ww5*k5WU#KvsW=Lc5os z7EBBY+t>^Is5t@wg@_2~DgXmjP53pL9>Zigdc_0517>>wSU7EI7`GPB+oLNUrvI+V z1@q~PBQqg+Z2_@{t-@+T+^u_*C_ zPIQ!74j>)hA~;>yq0q*)mC?;~IB>d0yo7^191%POGVRME$c>?roHOzn^%T2uiv~gV zRiuG85_cQy{n|t0sxa8<;UTTABj0E$u!eu%IXqamT!>IZ$1qsIQZ|2X%x$%vq7bXe z6GdC5vh?;mbL-ps}FI$1go8!YV(wD8goN`CnjhvLT#GLIdf4r^hWxDkHRdE>C(VfEoQdU7ngSZ zuek1JQ>_`*>s-2=^GeOp>8uh5`<&t-FYJN~-UsypQef`sJat3B4JN4;Bg+&V<4;S; zY5zfIa>&~aP91N~A6rJnUMW1;*luC~qm(;`fnCF<*PUIiN8%E;0DHW%Mrod94rfu2 zM56`FRRD689r18yBYzevZNjuq;*K5jrd<^cQ6lS`&HdeKk)IX$LF@?});zWg=U z1V%8~b0@a-Vr*Zis_M@GBerFPHyl!18U9ZL{k&Fn~YS^v? z!`|wRX+7ee`J&A)*Pb~hSjyhIj*4y|+C*HfQrql%1z%XHCs~T|1=nb3Gi?M4BWnwlo+vmBeM6@p?#`v$DPX z5F?yX>3})+;fj^^_FdIQ3!- zv#*rfwzi8q)B!zx*C(XMxoBZPHe4_IbS&eZaB0OcLUZ*8@kV4Q@|vn@s}|}8!Q)HZ z@GJ#6b{KKAhT>HXjdGwvxrh?hpbL^DcwbVv-sWi}p*tW2Ax8pv-wU6Sto?wO4_ zNXpr)iD`1O zak5Iy9pHF=udb~I<#oO)TFve&Q8@x{6sq1D^(qzv9W^Q@yoKtNce z++eJNzM=GXEG$smN5Jt8vclf?z}El*N2NKjm~xzc%p)iz+E-e+oRHo5?nQf!%ySrN zKkIrn3xBqsZ)>XwC;$ei-rI}Rc%qm|fdLy;LmXb1~L zSi7N*3dqpI*x8OYvSAp0+Ek4QG5yW32C^8cX@4Kf(u5g>bQS z1N9VBb$d6a=+(jI$ri_JED=}qwxR+ah`_q8f_!Yyu^iYzGXiq)v<1dy`*D5&aqpA2 z-*^kzm1{C3&pnKu`Y7hSdGDQ8HzW(1?@*CSN*+^sS}1$EeQmZN{kEyW9--#WdDBQ( z1eTzJ9q2C?$&svXJwCFhW1W#{fUPc@_VIz7wTf)q!2<6otf8XUdR9?eXKJn6MHZcP zx)ycF5bGguVd>jRrKlpmQzA!qdHtZh;CH*S!I#+rW?Cy6($8;cc16qVx(7LixajGE zgZQ~YSUfYe1613y%th5?(oV1Gvh}^Albt63UA7JM+e;F5<8`-?xWEDYcx?im%7HKD z2N$9jm3DnFYhuXx20)Y@q8mccXFdp*v1?_f2oDKy=+fEUoQ7-as4pF{45r!?8Afl^Re z_U8BxI=EESD9L6NxQH1Pea&m(x?(87|g^0U4+<92&4e#zpx>H1PSo}>p|>aCUI zAbiipv1!>$8g)7pWRk+z2Ubo5HeCIF>agtmb~NS)H5FmthtCVJ;?Cw!`F}Nvnl(0KY`J|2g0NQkVjy(eljVezoO*n*Tr&VJKniV`70ma1xWN50N zBI)TFJP#MwwO909Y02YUnA?X7c5Pn#SEgqCN1lan2jiu13gix(Yv1R|4tWmD2+gZ6mssJ`Mv;uzpEq0P15ab=KD1zhvc1q zrSBR{AO0yhxz9;zdjNTftu(ajr}*^$>@VD0-Oa434)nLfXK9Q*4YZSbpG(F_2y+zK zs|eH1M*+Tl&)fipd<0O)f+@&Vr)rp+@}K109HJN?-#qdA0QnT zV0iZ@0V|t<_IjvyCLZCAtX6 zHgC@2pa1HjTkLZ9u$1gZC+&7Nqbz;+`eC3HI?PGlrT_gecv&r;L*ifN{-FC5gq+E+ zn_uq3(A3e@*o`glQTe@%kOH$GbV1w=v>M)@pc(T!(5xE@KpBdJRg~$aDG2|dgUoUc zf^>;n$zwkiE1+62H``z*mdZ5%QaYRsPRUl&$r2v^3nE6f zluJ};NQDqVsOv8?>@!OEpO|Zv&eUY=gvZXiI3s8J?Pe;vSJ8{7!Hg~hz&Y{8!k}dtx;s30{HolF>>}V>m&WhKRg`2j@LP%9eAl`3E<^} zHDGCJ+ryIFw!rMzGru!CKsWYPKxwZB^=IN;rXkB3?d%@M0A#P;23!h%ot(~2x|iRA zWZx3lQLq~TMS2S0z#%n|xy7#g>kqoo3#*j3Ycc@iaq_{S)QfM~_d1diDY8H4?$0Bs zEQKYdpWp`pU{(4z$QH0nbmbD7)H~y=1{C~eH2Nn| z0QN*kbSJWD*b_~wX9Hs6&4h|(H&2TON_=Y&biu3}yCyY1CRSHT?0B}_Stt(bD zF;$~m97tyDf@vq1EV0)X=U!}mn<+h=_%&DU$GZe|rZ)>mFlCv^F%n&e}{-CPm_wpVOE@8Vh{9s(_F}Zb*54qu14Qd|?Yd2ReyL zf#RbKB;?(m^|-!*MZTI72~%^H1L5zN)Df_GP-GFDBYdO^xzr2|H{1`p{`nfQ>C=}6 zBK#fYcJZCuhAqO zuF5cHds}pG_~AyG4N1_d?T6kFzsth8H=Zy$o_cKYwlr7f-7%-wxyp;4rQcf>atR-v zTAT=df3KdSu+#oE-_29Aude0FL@t{uW!VW^d;1K3n|oP)$vgzDx-pCEX7=`>DLpkh z&Nra+z}}~hQzhVQU{!+rsw8C^a z>wAN3(lZ0peY!cD$Zwol-yZX?rO_Bi$ww7zY|`MmL|WWNsN&;`Z3RZp8~wclhc$?< z4IlWT!0B+?wODO5O0GaZEsP|zBi^L%o5Y>`n%S+*jXSsc1viGx42pg zv6pnXMgq2wCv4{LS_&mED!mxt-r`0jD7;O3XmKkFc&>VJojm|4Ej`76ImW za0~K&%AwvYCVe#XqxXTNU%+?rBF~QvJQPHP*CP%_O)t^ zjrAQ?mB}I$-Z(;be)DWK?Q*A#8_rj(?wUY{sg2?5gHDbzk4}k(&(9r}-`w|kn&FP# zwL}RmnZ`cgigWM)uN2Gk4&Wo?-1{aQ9LwQFk{ zUNn4s2dOEwyJ?hsz@@Z{_BSA}89#1F6$`7Mvg!w1ufcq0x6FU|K`?PSTrj{7c;BDJ#3k)F^s{dCDlYo7bPv>?u zYR;8xU;xaqKtAn-6(}_b9rEPc>zFru=VGPXRgrSgF~!m7cCo_ zIoRvY`6{temf)Twp`0COF_uE?9?~<};#hc!(iLHz=d0(E`(qf`@?o6{2ybMyDbli$ z;zh_C#9Kz%lo=7)2jPt(l}(kak{No8s0Y^c#Govj(x@I(x)IDk<6`z?LGWdc9Gu#D{zupnnX=57ffI#n$#!}=~ar)rkJA^Pw5|=LTxlOmvz1C9gL!2 zDI%os95^Sj2UHS1Y*G=8`EV!yCP%{IYZ+|SY@tYCx#q(f?D;oBNYF&jLA6VvUZ;%8 zbfvEJ(j6i~oWi%-7}DE;-}3LeLEtA7aDSnVzLc}JQx zVMzMB`q_?-`}ftg{~zCVg6){U0?YR0M2-WM6HD{0!k3x zPB8lwnoVW{6TacECj8*4DFmQu9w$7Ce9Fy{df0nofXt)^IKTtYI}OWfl!NP#ofiP` z@&Drm?51JhyDQlVkr{Ysq0yBAn0Vk%fZ1Zp!Y;6*w?z%tgrW7a@J#BYx7d0t=++RZ zz^`(zjM4_610$`7G1L+;IFDF~0OQYsS5RN3K8O9FQ&522NBkC2%2MNz6Dkz|paRBi z=5yxkC)g6en?%b=Zxb${`F;DaTTgt9Ot~7lC7V;oKy0`PCNf_BJ~N*uGIhXj;TK zjokoh0`rv206usoHT!EdL6QdV!? zFR1=94&^sDqJgaH@{6o$0i0_AO(Ht|i>$ivtE`&$tE?LIUy)UvAm^|XffT?2<@($O zLOK9~`5QLi(^0fx>_#>skm9_f*%BFmW_uH8wkeqJ*nSMn5=sU<;6UM;7m}O`)A6kb zn*_&#(nl)u=(OGqB1#>US~^j9!GqYbHo}FIA~qnDS>BCM5jI2K0efo^-oII z-sl`&fYbQ$F@g$tVmXD3LKv>kZ&y~>zBPOTLZv=fR>x$_cwV|SX&MJhKQIs6H8!?l z*HNyG_wRHy>Nl72^=GXt*KaM>j*PgYGJ*Eiz4)!+#zAN4xUj+~C#N!*&LwV{9=Ez7 zCj&N{dto;Dv8}NVpT)OQK8mTl>rsC#sv@SFVqb2Vu)t8E6CUPkOQ*vA@oQZ|Ay)`r zR)?39GSxl9u~Ex11EPYchUCAbCL$!{1vBiYJ=JPc9hO%&Eh3+VBcyl;m?58or?PMg z%H0Rj?|u~inEamp6D9&W_7-^{k>o*^w$aBNF;S{JJP=j)W+Dga4qcAM`0YrG8O9~xBow#|*z>WVzetZK_91eJeQCCn+^?So~z96EIy+d|x zD10;BKzZ@@?6u_p_DgyZjw0g3LX^gexyTB znqJM3(J-JCiP2aQ!{*08m9Sgr_GM1Lo}eu?8Y7Wxs6bYN3#f}3y-^x(W#$-|d)!oe z`{R|Y+TN?vL%rRF)jbW?p^5|AvFcsVFADNH^D@5QBd)0+P|xP?1Un<3MaDsc$6h&v zt~EV{5Y8+X_3k3h>!5CugvO1w-f%s(aS;=lkuyw$_{mA<8dS294jaQ;(n^ohX0OuR z#?}4K(tqFU;}|s2S_|EMqxQ`LmosyT1U~3m_#86cTdW_J`$c_@s?n!Gx(KZwzCp$f zW1h;$L*MvweQxLRU~skVi;&rVR!iU4J0pPLqE*RV>dGmX#+yj0`i^iA-euqgc$M@+ zow45*lW` z{1X9bPM{0*>jxeCH^*Rs6m$LyWrf8e{nr=$#20JQH|LOJ5)_=YDH|^Hhz262W0@o%S#RJdqkx0HDAG5|wm&lG{hV&?O^5=TM4wBsTipk$ zGbdwqKY{9F$AY2zl2PhJkGisEC$324Qd6|ZxrTz{SC}g&D4FYgm5yM3d6M=C1^l4P z6MAAl9^4e8m2aZuTE9O+chB_bQ<(DIU4JpE{`&X55m7=*OsEWhrnOoLcC$RBw*!8`@hYY}#U}`cq$%pt184#N~)k0radu5)Vl53JKeFLT6 z2&>YsV$Hz8j>ACdzyZCA)aD6WLyImC_YM>;{FEsA+W|US>ZKh1F-+ZASP+_#B7kq^ zEhkB+?Vnn`Hd!?QD%GqMx)=PzoA-Ye&HN`(NE*)-S@dW%j5ih=RD3IYP@GK^L}6N>$+ag?FGZr_ALHWA*lpzWVYf zIk0z9fGwHkPuxNUd#vhE-Y=O?D}f#L9;E({9km;JHUq#n&BKxuN3aw9EIQoWJwU&Z zIY8M4qXz|6RR5$Xb*0E-$Vv04nf-vwk`P(5=B5PWYS#qkrLe0U%|NW9q517DI1})D z*Wkye82(LdDdr^LVvr9Pi=yeukd`eR*NhGYAlPOHN>&3U3Ixdvj>)VaH{?rMgEfc`-6W?rZWbTKMfFVq|p=;z&3=J*KZViLK^GE-UAfvPyv=@ z;|0z?UeNgS3$pE@bGyGXu;rYlE3toN;H5z_08O##xJdX;Q7|sFPSdaS{*8v6b;q&; zJ!nJ$Q3FVr3+Vw0p4wAC71rNRJYK%?r@P|E75U|+DbpZDfFi~9TZyDSMbzlI2nxtj z01rSZmr9SK>e0;$3Sxz{DJZM~lQh7H0!5DEdXb|6hL**2VUFOS9W@LVFhG(7)y|qa zpUd(>J`U)1k4XS?%Y-N4eyN|*NH(yA zW2jyny#t%v_?4kgWG#743ckvHTR~^S6R3jM?|-phKGJVZ6aVyt{LlSqkp7zl8*zRE zDt!;sL<3^Nuw$`#mIZ)LsnN9iZGd4{`!mCy2c!*uV%Yz-R$5XIv@h@eY=>w?Dpg;aa%Hpy(oWus;Jft3Xk9t0vtia)?X% zQ>6thk=(hlPb|;$>=;kq6huRSEy7?LoGYJC_AL@hzBUvfHI~hgwWG6 zc-W4QZ(5vSzBpDOOAb;sDQt5RHDftOa*_rd-C_PBj zc_@_I=FF!#roO_=)7`8&<4BiY%*pmzK+d}gkD(l_x@OskBI4ThYHcT;r95(dHZJUI z{|T8tevC|S$ks*_8sE))5TepAP;S*BctlSDeTXCVeTuo)#g4L~#Mdl~vLAV>o);IbMsr`Y@6-TmmsfB#hu5^q~9U43ZI<1SjvS_*%CT5nN({z|;pcx!H2Y-@y> z-$&g+#~Y@$a$VO5^EDpY_p>x4viV1yY8*JfaCZfc%d2`@dSi{6Z-sC^3Fivgo6ZRt zD#B(5uL?D!^Mo8pKj?VXl0!y$Ec0^h2g@$6^neLn$h4XurOjGbHOn%+dVl}9ZOb%# z`*C#%)Oi?#wLlA!hkUN;;ODT3s92rp-DBlDtiwIKHVsmH$ST#!l8r!wtf zt7s|a69N1ktg^1k$1>h$wIpWT6tjQ2xX9muR%sHvyOTH0daO4Nm#tl=UVQrTNK1@% zNb=BwBYE#W8`gK?PJ=7?%2AYz8cNx-x3;5^GHmU<#@{Epc%8muIo&;8glzwUF|Odv zN0oL6;e{pWmL1osGJ7>oH~Yx`bp3tKDb73&D$SQ;qBYvnT9$WuHtHPva%QL_pwQ~I zkbK7-Kl8K``4ie}$0y+aUm1Avq4#jR%V5Nys@RaySA|_CK7afp@M3=&Un~z*bVX+< zpeB1Ke&wDMhKp=FYv&Yli8D%8Qz|9&M!;LKd2?ZLd#tu`VAcULw~53~nG zW1TwGfpj=~uo~iZN^tT2VU7Jn8(0qq5_^Kt7Gv zJe+d0mYd@$*V=IgI46G&q; z9~lImZ&u;(m8X=5a1q#>*!R zH_G{;c~hN7hZRjHp02&()GTTC>(TA3I7Bvj7yKxHj|yE8vXlUsv>O@^(5X_~^2zDO z)gTiyj!i0O^Uvfwh=qQEr^fCMjpHx$AEm|GIqiH_2n&#w z&3hmt|1eJbCfep>$4UtDEUI!h^}0B&1`D;=tzOV$BXPQDc=_?Amm0K1P+O9GcC1oJ zMYK2arz!gSQ`O! zPvsh}k*5U>%{^66Y&n5xnZ}fWinxcWc3TPOQzp7+#uO0+OYy~>vPghTM3Z=EiO%yq zj)OOiq=|eXp>gRA>CKiSlS6(6yACjo35|~_Oiu`$^VQf~u@2l^07AY5eXs+!n*Yjh zHn7Z*zQYi_ASBXugum4Cciu4V8(LQw${z6y5w z$3h3oXq|aHYFC|9k5Oo4w?d3Y5v0Ho&gPa&-)TPl2%8FLbz?tQ@Va54_nKwAwadru z&0`hUX+W}Nf_gNNe<%9{o&WWb)YX)}>7L{_Z}&WZUaY_UW^9qs8ow_ha|Dt5T?jhZau$>}ruumPyS0{E^2rH`C(o5?O?ZRDO0 zAH(T9j|0~zBryDaQ@jq5yVDIUd7tOH3EzC~7s{fG7SSO^B5y$F+2H;jHJz?H~{8V^!lYvqyEKPWv+%Kl#c9dLuz8Q+Wj3SLL@U zHM^+%#GGV_M4J=VJqKxRb8gR+FEiJNZ)3D8N9KHoCsgE#J=H>Oa*dVQgr?==W@ zjjhZ^gR^b?ioh~E?}ub_=lz(^HId`IdIo{8WZLuSidPlu*^dl@x>vRZvlEhK~cRTcd2iN1BMPl_ZOqSJ^T3sUe2EWwnU04Cf4c)52KiQO6N% zOPu6o;!b>YkfJCee5zlpYks-6>vknpf>#Nx5T;s%HojSNw7llgzKg9F$E}ir!!>Rl zW?V9_g%R`z9!7_&^9KxKA)gJP#Rn^6RDGyWyGbLv^`UJ|o_70iyQ0lz(=6w!lhj3Z zT0Ckn3d)H*-ra+c>p2g8O5=f^G|PZh*tU00L3{S2HAQFac7QX-MdzJ#8FI#PA$Ujn z)UwWf@n+CDv5Dx9qjGu zWL2`MP&Cf)U?5<2DJXD|N(=GYBxdlC-}cebo{Eu+-GY;FODYR3pS~OUX_tVTW39}m zsi*vRo*~uC-q$CE3^KvyB0wHp249L?~=B!BZ&^7!E(D)_J3kpKCgrX7RAaF;14J2V{2^gv<<;)JO}cpW_KA1>OKis z*?U+18ARLR*n_FNj_?;+8{m)5vEhM7{hVZYa^p+4OrzJIW`;FKbH;XjU^2{HiJjPl zJw*eSPs?mUk}g@r8=pDC?eV<0uwm58aI=k5c%lU!3> zs=a7bU|))&(i@U)TT-wGs|Dk7G(LdJr8ea1Z#ySO6-}ComvDX;N;BVZB&fXpEtZ`g@GuTVH8-lRn(9gF$SB0?45O$3>xOPe*i| ziJtUrr9QudP(|W7&}~h zvLrhy-<+=><8Ftt6h5Mmmnk6I^5ijM)fnqRiv@Ft-6Kq{a<=UnYpzUUqA29_Ib38) zfz#YeR;fn4(x7o4^bY!~Wt%)q0(;OfiSJO|2o_Mte}U?LN51|O z@|HJNroBSH(p3xlu3n#6Chf@{qH`eQ(EZV5?x_tO3Xl{3n{S+f+SR=N)@>K&I@w z+HT>#n9(I$pze{VcP~)GGy#vO^`Up~cVZ6qMgIc@%72SE{rNnYpGMR^!DgU<=}77< znBG|lT{t)a{eif-W0g_t3RAe6=u8mXcj8XM2N2NLvJr3i5C0ujJdSFR?u5;v>7kca zVC$arFau~3%^g+tmO)7O=fxrGm&HV{cyMm~-LygROUH?0}+YC1j?1v7iOn#qpM6(I!@?RsEzvjK`5D4cF|->1tXQDKv2xcRFX@Z@_lggkP?C7qE%;HWu_0f z7`bVsz(1}}<0g-;%h~Y|AZHJtN2tQVhBY^U`Tp7r1=whihdtt~_ze;Y4?Ci&1_SM2xF zu7Ft(2JR&y#P`9DjKx*S@V(gv^q<^r;c#|@qv7S=$HkwY4VsS7^SrhYCdrl*VN20N zI@gK2yp4$Hev!dFm20~?8p`__PSrb7r~6aN;||)UHFemgYE$@)iK3bmjYlsSh`aru zk#}8X_hIzgiPFU4PDwN2Q(?F`M;-2cF}3Aa$f8PYhU%Sf=7gF(qVQf3dWGzs?H6C? zs%yQl57ZNyzF`=oeMZnEgOfRA@p^Z;h+yZv;IIO;M~2XSVn&*+)st46j60dexvHTe zZ< zW-|YZjvrxPPI=aFn9`{bRKqVp7_XjX$+u&h*4F3q4QaA>KMa5840=mL={sN)#Y6ct zP?m0wm2E1V^<^LNb~QOk6%O6b#gkjHz2!Zx{0>#eWbjh-Y<FzFmWx}&os||F z@Hi3N;LEjF?Jx^*_AA%J(k6%zGrF%}?{PPD<{{Zgu+~(bx7({p%c5%T$;D&F>eb{Lb-qCK?a7@-6rf+?-E-WldXhaHDN?ZW6KSGbZ`?XB%IwQ} zmOt;#E1A7Y`wGv(Pk`jIf)#(5IYl&Nui7zkT7BygHE*ugGgVvV?cyG|9G-qRzBNO% zeDkDet_eU;(w%VKU6riFSLH-x3QIvyin60no9yOur$xqHG*%GAmHPgJm}>niFQQk{ zt`TLh4qZL7tJcqQ_C$_F7A4%UJ-(o)j2aEz ziFsmDw1Q>olI{(0nP|E@ZtIW{kD+-iP1asTA1C7kn`50*?cU&aXV+HuU0J;7Fz29t z;q$yQ{%~{lZ7Yv}nY7e6I4PQJXOU1qGZL_GW|XF=Q8e9dx?u-V!P#N5UwV3J1qGY`;?#4cj|yNUM-`0u0gVa77twJD`LJKO z3Wc9~-r@m#rUhx_076`wrPm#^9y&8!+lt7`gSx%RR$=CgSZVAP+OxgpvzrEG6!I_P z;uuhBqAMz`Q6TqxP4m+~Vl0@(=c!z``37=7K00(t{Tj+JpPUQN!x8i{6QMIJiIx3^ z*EB{8sR*Lg29BrL4R#I6)4Kt4mG7LIB-38;Iv(_fxLUZ#Nx{Hk(is2rAslD{x8F}EhxRM+ z&URwmO;TSB?rnAox!~iv{c&A|8toC#%{Lj4YRTQlty(Y3pRzoH>yl=jFU}p!P|$lC zD(B;9bPBH#*wwRAh1*Hy*s3mhzPPCTj7ERVt$WR*d$4_A|Ac+J7v+P>MczY*zp2F{ zE=nt0IZFk>IgSLj8JY#-K-u2GzP&>$4*pPMAELjTaS=TpH{% z4O2APH6uGHq0WGk&$}_}c)nd;I@8e7Bl~#|G7)hg5P|R$nRq+s%W^FNa)B1-tQN;T zJYHSed#(!4hpc^aq|$C64T{#N!=+LzFLx3zHs;4MaxT4_g!|zLmw|)IPLJF|Y*Z%R z=oSf~r)3S>{qsQMZ;FN{kv4;SbMG zu30pH@bNkL#Di1&56@3b^5HYI$)K-kn!(%uy)(Si>D*A&i5S9E0d`8-05Kg>iCCHed-@Dq*`$)RuJ2 zGsPxN7TOaL{C5wT!RqeX&mb1NNV4Qp0P5}~1rt#Ira>Ryc&3(?qxVUdU6A*oX|a*J zmTb~KSEhb8(9^FwP*C{DI8b^bJz)#u1Vh61Neh5RlcDsTJi|da2>qj+?IZK!m9*@a zb;3vL$dS;)-j4(quoa@iD6&XErMDWsYo^iFHK|v+Gi_)(rTxRPkPifrqQ{4)G(_$y z9A+dS1GnuT4CXz|m`*mpH#TqfQYH*M6x&zto(8X$zg&KnS#-_))#D};J3(iQ4m5rX zN)`y2G%Dj%h70R)JKXNn^{8gY%nDaL;}+=e)8z^)4}iv@F1iqcJl&G^`CY|5u4PaK zY$DSzmdce`*A3D7!?`@DWDM(pHhbGw7J z&+o=btX;Snba$_lS^3jqs6SE%hyx5!mEvV6&p}L2S9Ry74^{O#fr!)To2nX~*nqmS zq+WVK+`AG?R}o;Ya$6*Y->+5f_;`OP)BRvkk>Jj;5~;d``0dR?tY;D+Ujo;W{RJ3k z;Oh#On^|!}@(sd^LfaRkj~mna{rSzZ-|}c{dYC7Y4wXBAd_Y5a*TFuOq4D5l9P?Dm zs(w_2tN8QjV3vtmRsJv9oZf)G*y2idiD#pwv*z2`Mv*+0{T%(zhO{PZR#qH&B~4gn z@PNEUhr|oeF!Y0#ibn~3ZD*a$_b(;fY|$F=r)i2%r3h5qnqLB> z4P-n2LR;+E(4q&T8mwDjGe)3zxMYi1x`I0m8%)A17@W|ca3=tQKQ@N6fLckNXnr|_ zUEL$8>>Qikr@CJK18tP55`)B)tmd^dJAp9j$<@t3KXs6U;eY%SOaKc1)yTc>^OpnV zZ2+x=r4Yyg7F8F2<_P?&ixuyz>&V+U|KDAmJFe##1vrD0Gbm&<<`JEs4Uu>-dL*3)Ar0R77`)$Fu%ncgE<3oW0Xclxd|VeZ+}uGoR^rfBISJIMAZqXzBZpmg%`km8lmbpnPyX6* zVk7wVHBbt7K7@8C9Cr$54qbhUngy4~s;dP+F>{gL*H(BWsX?##GnW+7)<<5#>wk2c z{L}mAf6NoWOmPnNZu0N<^`{$)^XoH{VBB|2lIou|N%{W?O;SnQIa51$91L(L9O|w^ z%>jy=3mxFP9b7|6OzE0B0C>T+=!_O&8=HHJW9tS8SK)r~TwyD{y?!>BzAuT3YuaI* z`eEXsxx<`qCiywbTdWrzPs*8U+o#jnc;jLEt3k>{<`9+-n#Fe?=Zf+t2d>NqRUF>j zTwD0&aX;cB?$I84LVUoH{c00KfoF%o3sGjw9A@--hw9?~TM9>5 zY>`vQw=ws+P^K7$Q_vHZCxj=m(IcXG$)0MhPLu@(-7kf?vK-=x&p`73i+ATV9NRY|YOq4BviHw5W$3nsh^Fi_K#4N4>TRG|XWd*}@{TD~Ipmimd!&V2I zwk=>kh?up%-j#2;sG1W8jefSKhezFWso)_S5*aIz`}2MA6!iiP?#XMf-iw9Vx@&pv z3f#7YO{?H`(>U8oJ(h@w=P&d=#B#3ci0Lc;hC;Ecwg*&Cjw>94L&?>o`1`|!HM4ft zTeDugY8@nKC)yfE^BYUOeQR5}JPEg=o@yIuq4ScptBebDPfXm0ynFyn%wfJ@&fAPp zF&+3YRqdd{oUGn=s1>Ib`pCc$+lku;v&}c7Wy;%?-XA+MK7}A1lS8Lt-=WBdUE+IC zvxaybn9EcTPXtk1uiDYg)~o;h>A;X=?b++I2`XQ-dye0ZJzx0};5=5V?&(Glt9{qx zJEcqm7Si?Q+C3f0Uv593a7?c#D#h%Cse_19(dwDU=1d3fRq_v#GMaMco>beaTx(K_ z5m`};rNyt)yKoNV=Iny!E16(Q-C4WvmBHM~@0kF6?anF_eeF=i^Df+AE*C4IKG92| zMDv||19n{VA$i3n#@JA%O8c;zVW~zk+$>jD{5Ay+s zyVm;~*-mB_V`fX@2VkjQ4KJ;D{X&i#Os(0JURG0cm9NuFQE0eb({%va{S3;0=@WR&QA~=i1>*6-E zqE76#D)qai0_ZQyo7FXP49{ksXXaY-&WU4I5qx*HxFU^5I=IIGE8^;@8MP)5oe?(* z({!BIkGb-=?P!eg!}>m^;>;3cu+DA`vM{;xNrT}2(1YS)ek#28I-U-xj45e|vYSu1 zQg_4O9@;C(M;@~vj=ixg$WpQ9BU80>CU(CHzF6;*{sFbgNP{9wm4t3Xnd7!iXmCe* zOTWUE4MzwioMM`cF!IMY0`_8_Ws3d2ufQ!B+r=?*s!1JcFh2KKEB#O{m9a)sioRc` zs}TS(LEs)G7TzJ?&~~WNxVoZR)O;A+5O`mznr-2`rU>ZXSaoj$k`~$D_Mu@#C6196 z#f*ydEVsw-ESV7@RZ!l^I@pOX6J231!76F&jmr_afI6n7)z#@Rxf0ElE76b;mgKx- zc2aPWlZAn^kGj2_{*l};I34b?>hW9{MiA54t{=}?ih#0Davydn(*cdJRJQuSGK45W zOKCT3WEK=ut`wB`7b=}V2)vrx)})uP8Q>j1e?dON$l`_rk%WNCtL8M9n`p}TYS)bVMw%+;Y6K|M-U5DT z?~onMPZGCu^TwJD{27dJ6SB%y!3DM^=O|!TReJ^+2K2HF9_X1$!VFc5OdGUg$-8ZU zDAyjC6+w;rgk8pv=MZZeZ8J%2);8Ju4PW)vy%ty#@nM<}B>?UBfg!Ul93n)Kq+9dY zvc$Q)hOOoOu7E{spZ*39MZkpOUc;(jJp$9S72<>%^BbTW+MG?!zG2_Fr;-2yZgN}k z2j?SfVDkw+bvsEhXZ2gkmBYSxl2Y(4)I-KahzT6ga%NA(F2bFFLB6fMN`$dU%VbOK zVYVKk2JhT7CDCSc+uTyAw45r{E`do7ykGN0V8Q z6Nt4Z`icQ4@6)Syd*3ePdN8DPEp8v4XE$_ft1Ia*_Nbl1&W*$?3kx3%VYt--_C#x7 zdHq9!AjS*aS^REEV*^elHcz-n5o|3fQrj86GWgtsWAj&Az>D-+G@-unq8;ZLR(Okj zD&+h^SYgB|8uzzx@l43_gx$VCe%UR0x~984?rS9ZC^FDeP=jF=+sLUoM&zKkYq4&U z*NaWWdOXfowCZgixrgaPo9)4FA=FdZV6{5DZ(iVr;}jd!o25S2v^JtwbPv-I%VU8I zL}+(3={x~4gB@BGD-$&vpNbYqTGTDslq0ciW!wrN8LXlp8*QUbK1#J2jAO?O`!7RS z$w5~FZnVtfk2c%fO^m$9r;^BbcHe#YGbk@?Fv!k6^QIB?IPvCD{mZox>YX=y-@;0S z3~K}0b*ADtHm$M~dgXIlO&lPZ%Zbo+Db#G7yawJ5laJun70Glkj)iL z`2_5Jr<%5<38cBnqME!Pb6H5o)8TU0rSF<6Q>%W7PfJ{NshOWYL)Bva(5o?3MdS_w zy-U$GqzuMQS$hKL#ybg+4|Ze;mcUy4yuZ#FuQt7bWYk+D2od|?5)J4SeLuQUsufRl zR&W-A?3dK(s-A&spgg?OR}YIUPUtN)KJ9ovc#W6(Ibm}JnCyL}r^2f#*R8lVJ$x`+ zy&@$=I%p(@nozD_!4`w?RMdiN1nP4kAg+`50|-U#zYb;K*|7ZN#PUaecbo)m`M(Dx z|IU*TYQEt)`Mm^!+>Ovf+5e)@!>QvN?m0sl1yR;-)*-H)|E>r8mK0FqzKaEg1AG#Q z2}ZDQaDS18zueqnSu4sjKA4yk}7}i z0Tz^$7SPpwS-9y+;N6Ox@6CJ*vw(mDUm3m~8Q8GBqt^QaWIx$}J8pe?blg(oR40yrlO;N|se7nZ&m z8MQu(eRnqM{`(#{!hX{i*z_&@q)hFX@DpIu`*WBDY4n#?0y+TF@>V^;WsnJ}1cE*eTYzC2ffajA9D#7&$hhcO&$##=8+i8dZ|4Wv zeUNB;hlXfcS}6H6Db`|%3UZttfT{Hw?d-6`mT3sU6o;F?L3 z+ZfPp2%F)tKfpD<4${dD;SwfvL%8(qI(5{Ve0cTk+FiT#dk?yT=`EN=NmOq=y%lTP zXCzHgg06A{kb4;DEp_fd$i!8;N@^5H?KwZbdVI<57e80cuelf%nuWkO*Zk2QR*%V~ zRs|l;3cU+_Ym5(nk(%IgG1L#RZ%UIT#Ben5iamspB(0!p>Iaby$d75j^7jomS$g&@ z0^=hXD&Hb7%)iea64l?_xJthN_n_~0PXZi9{{&k&rHAVB1#U3WfkGW0UN@Ri}) z5!^;X;#Y?Ch(rT0Cj!08TW}5kq4x5h9Zw&b61ij=mysw1!$b9{1>_E|H8hp;{uNaY zsn5FSuJ^7m#?UU>l_}=08WKOsw1QjyrP|%<&E9f+#a|h6HW$1oAKo#rslm&rLzUpY zIM)zmkEw|v_{YkzrjO7*w|!+e$2kS(o)1Q{+`WQo1*7&Hc4tY4%f!l2^Jluq^PDr< zohlJ$^|vb2e8K0YMTl8;aq|rFzF2V1IiC@FhdRY=(_V+tH@0rwb!sg3q9bU@xugBK z)`v+b7dw;?!;P{E*lg4#@<4n&TkDlVwy>G{&9cu*1$&OeHkGV)ZpK*>v zzS_g#qM{O;nw-&1tMsXRtJ>&VtiR$;GWd%MJI@mf;)c~PH*QTR?A&j|_GWj>?mg7E zulA6z3Wa1pn<4cIXq!Mu9k2CqF2!;yGjdo4=eABc&+~oqS;+jEM(!(DBs9!yaM@_y zsNLC|lIGAy1J1?Eb~8Fd!l0p0PRe1^0Id?=wjHohKb8{I+sr_k+GCN&4;e1L4LR0k z6%6^LGg?1`pxC;>hFtPWbV8^w{xEYfHH3M#?h#`v*7Mv(s$aG>FyJCN_6TauI>BNid_9oa*Ltrb#k*NwCU zSgv2*_3T*P#KOfjYR4MLh4<OERn2XzO!c>}s&TrL%`Ud(n z9Uwx()qG!ijHT*M!_-S9%q(+k3OJZCdo9m&nP(-tAi(+j!J@g7sqX_dEovzzrvo*FJhUs{ z=e|FXZ5Z)O1Cu#H&;vny$E;SNNW` zlM;QLTBff-Y74RqS&7N*r9#MZF;UqJ8k}yiHhrBcO&s#p=!G2vA?D&t;%rN0;iB&@ zvAkYVS~99CYa2#Cj8-B@*#Nov^tK*nWA%e3n~Cci#NUt@=2r0w{u z{*%njho|&DGGy^ro|8>4L+gT40X1QlLS^(NR(?!%m;R`MdWxQeUtSl4c<~Sd$&exV z11IS{G6j~Uw6VSDDRr0W6(#%Jj=nF>QWoKO;XozG-VE*QpYB(3tcjeV+rSWZ0gF^A zkfP_E*u?WtB2(eRMdf20Ii?!bh#dyHTtKX3<^oeO$ zg~}CBXMT-5IfP@rANImh)JY}r8QA3XIIj-D&m#v0cJIZz&KeGvpEYCe(d(CqLkK*7 z@Ic@kroF5A1n@{0wm5kICX=lcOisAdKLEo@cYwbupcmF2`1{pft{$C1Aqfb!Y4r9^?-5?rUWMJ zz)n$EogL4Yu&BZ|fgPxR%T4qf6hSx|Mz}DrD$)-FT$Bxt9&iv@{k>KNFvB{{{|wot zA3HGsxH&Y|#0Je8aeV6%9v5R7>y+cI8$*wRSvE;8qq-^eofiK6Hiyi1oF>AX?mo!-J+QipCG0)aV6bb0@Jw~K{{6*zyl8-6ccLHa z5@qUFj(fc#MU>P9&sE+~2E7CsJIew{k7*fYJ1IxWs`+THsr`oUv&8a>wjpBE_#z!n zPDXsz8AifW!&+tt*8&+ zNjJ&<6zKw=3dOcIjaYrvc)-;hIvJeq6)6n$M#(Mgpn>7%JB)e9~%f2X0V&dpxhS>!O+}<>A~UUQ)TSK<2o?rcmJ!;sG}Nv;t9W}uX9J<^p1+EAarJ{+A+13eK;TOzKS zt_wn*oiP2xxN@!J85nsC|MURU0busTf0+W&GC(@0a?o>pT^{n6?t`0%UI3CQ?ZADr zvH%wBuU-}`8+}NU$ZLWdxdtO*KdnUpuqqm+_`_}v6O)!)JuS(J`W`QY9qB?(VHRBY z|1|0hv2Qjy4=yviQcD11K>Iebin833e{TZn3Zb_^7s1lEffE>G_W{4e7#Ip52j_7(Of=hNeDK6wU@23Bqlef_pB89H97q&)NcdU7KfPV4KtNX!u^m8+daJtMA;t}V?(aOlE;OqFD*hvhQamS5VLw27t)OHT7+{H6m(8~v zz~yV*S?=9)Q)iO0Sc_a=qzz8Ue#%Gt4eSyKPH)SHzE%@K&&O2@(>EjZ9rA*$RAOWl zZ-Fn zqbv0k!ttYTp0$kS>zI4b^N*#hzW&s1b7Uz?T}&b5kfnUBac4+%=TrZtLzW@?;T{hx z3qooJuQ}-)5ej!+dg;m>n3{5uKR$m;U3&6?t;16u+Sa)zLW-Y;TS-{RSzgFqNLK>& zo#;D`-8F|_4p>;~W*r6TNV0N%kT~D^K6EG*QL;XCop9wWF+c?ZdcA0%^0dkTr&SsiA`u~j%q>GM@0Nx zL#;^Y^&V{Bg=ulU<3^m6WTLb}R949TKxG~Umfdc8c$>nW^2lUT@TGm@pMr@1==8Jg zIEK3xiOzDVK{uapbYV3b%k+8cD|=)La%rWx+kjL@)R?Ky%6?@jJ@CiPlN6(%;}N6G z3xx5xur9{UWrzCvKVA$I>_`l^HhKTp+M42P`KC9&bZ`(TW*tOU^mz=DVP14DsulrT z*UZ%GhNu_5tSg6*-=(mzw zDtO4&DLb&9N$aGCF4=OrKHoPWF@>o^TT*Zt|yw&%3;1d-`}Cfx~V;kR^`5y?-jYvC*ADy zlZ9>b@K1cB!Lm`&+PwZLt}d?IjEnM$Zl*kR6M5S;<~r+4VVL6ITvSdBsNSjP3}=4V zB;M)2OI!i%V8S>rY{W1exQ;RuN2S#fnzS$6?|eO6tZmLP)IMt_&#CC=B7Ns{=vYCq z_~{s{s%Yn{lq&;IpYQM;uaW6QP#PGJjr}i2Qu^yMRZ3IOwS?&+a$;Ys^0g^Q6>*A8 zU9WMs8LqH86*zh)=a}kak>?)aO#>`nG^JZ?BVD;lxnk_Rv?Pz!vFKPfdzf54Vz}jz z{^je8cQW&yH8gpmX3{9)h=IKEpv?)S-;O`_-QW<{&MP3u`ma`DbNrzpI3CO_&%Hy5$JM4y$Co7J`st@DA&%wzAg zV}(nKBVQCf`w)I{{NU5);Z{3<^rB(}?7%eWAK!T0AkS1twa=6_vywGz-nnycn)`V9 zwCZ4_Xt+YHm*XIeaT>O}44&#{UQN=p;;l%PtF*J*Gx{{X;w?zaO(R{S#V-wXV+JdN z2B(khD}0@XN$;zJ4dX|B`l*YdTbc41JXwC!Tu4)}65NH2D3c9hu4M5#636dfR0XlCvsAFqq)5AZ6Ln2{LqRjND1j93f8RzX0Y!b-gg zZhBi!eBqYToCy3A-sg@c+sDH%q$@tX4)#w8kdSo16K;s>54pXTpb8GpvocZ|Pd~NW zLAkgC?4=KMO*SDxZ=U^%MHxlQ zc7GuY>BWA3VDCSXMb>Layf%b5LXcH6-MaK?^WF84<`#PqH@_S`D9BWB z*PVpb!q#h$#a3#kU^k&<+!mIgX=L3&x>=5IAqyb>d=u z7hRlao3K(6(q_yq+~_MG{<^MsRm(i}?vrb0xR>~LfQ>9~M^r>=PlIQA|NB=4y;Z64 zDkeS*oRmfiGqS3JkM`6{5b<&<>eUo#m)&iMd-_Gjo!02(YaW}vtYDilWFg6_y>=jk zvrAF9UAz4SU$mBE8n^0yNPX=o-($Wj9o*}&GlAQfD9gRapIWCE(v zUV^Pmoe;xt)+w)DFW`j{GfEI1y~OYf2B1@YV?~n6)=a;$U`OGaM=hBGC@tXUM7#Cx ze?b6#BMPBuy;vBz7G`Lkv(aje!}MDVzc6K`Z#~xj@4!ZHxj&zet=zw@op&(b=$)ss z8_;dcp4>j|L$<^hN-nT*1t0h);gmaY~DogV048f9REDil!fFy?@D zpWF~E^Y9{MP^bJPD0lOQXgG1g@tSY_!7blGO>b!Yw${ z0SW%YgJJLvRC(TLLh`bh-A0wS-sE%rBLy!8{L4@9r_V^xKw}n8>iu8>`yE6pk>K4A zfp>4TeccP6L**u-bwKUlW)w^V%&sUN{r=qdH4DP_Q-B4Wzzik@=-&~8JeZBfstk$; z)r+A=Rl>Qwz{(Y-xJbthfO5MYu6_@6rHm%yE_jo!0qi9M0K`j=COtv_Qs7PS{SRCl zX36z>I;zQzUJK6s(u*G4g=sKGe(k?8@)6dLNshF8x4)sMrlyO6TvIm*SGLAb?UuV2SYWnX(@J zO5RbTGbn#(h6!r4i@ebt6U=cp&7c2!<~W_u^^oN=&9cI_ zeBt-*sZQenLDMcE0M9)WfyYTm%1kSDzkHiuIu|^Qt_<-$S6zOGj6R!3R0&uX7oc;- ze_Uho)$L7WKBi(2j%0T$G~gE^GzU2QR`V8w@hY;OdA6xvu`FO7e~V!Lu%bDK?0PN6pl|nDN9+ebQ#<89Sdnuo0%iBl zke9uO`abolA1Dx}DSi6F)OM=Y!)bi`!4ESnm2!@nF8cibt^_)Ck7Z+qez$Zctlz(` zThMm+Q{b?)q<=|vCPw7cJ`YYIdSoW|toZ4{^3+U>x!{_%b}Ho@!d2ojcN9tWv%nls zVV9*|(jT~fxGJedkaf6HZnSjY%*M(WJLJU zuM97)ZGoja#GnJ0#jibz@yuwLKaq`Ylx{krl33zblAIiSIY?uSMBhf!gvH@5cXDN< zTdF+N)QhFe-GzzN!z>~Wrf4xyl4GL8#hQXugwonOU1vf#H2c0_#e1N|(Tagb0woUB zwSTU7Q4paN&e3;RmwO=lOj~a+Ci=MSq*`Qox2xffW0uCTPFXkCTr#a*_hQc^_aw%w zsh(Oj;rF>ggH){*K6)CoQ`xO@bxvs2M8iRS`{5ux4xutwXxY=bg|_`@BBml11>wnmQsu@*-?d(f1s;yXXs$fsgxWo zebB2(F3)%#k&bi)zI!0eR2r)P~?#J?)kd4irHXs zt<+%Qj?t4${7vu#_;%3MP5|H8KaEcdycv>+dv$_A&>z-qE+|Z}qLjX+GZ6-=p^)k% z@^z)$Pfnkoh*X>JI`?_*!&Y5NOo#iUcT9mCc9~L<>SE|a1dSOq!EtEz<@sdt&daxx z`JG$ya{9;Z<-444EYy8L`CbQ;w=7Y3QT}$K(id^k3d}j(Be2~tQ>-7HV`)`9znO7z zqN#PPiCHsa`42WY29<}QvP1RBj&{yDREcs-jlR~c28l;*Dm{YNx*FxT)~6l9>2|x- z>!*2J2p3xK4wpPUoGxt-Jq~+bcSkSP*CM5$)W1O8M(#swZqj1o@l)gifAY6Ml6T17Rg%0!4PgYzx$b`p5(&V>=!$o+^xRRtq;J`d{ z>v4OPiTPP2JIX;XeQvqY=#c1Yeu-@cv%JzN4^F>Wa{8>I_!;XvAJeN5Wz6@k$bZkn zu}c2iq>~DwvD|D1!|y|n(jHzF*%qUId%D4}Y1c^?KHaq)t>P;72fbN1;-~c=Kh%h@ zKDl_Tfy9|;KD76c^?RN#@kYvH`QkIJRUE-0nTP9OrL#9A@(tW{?;j%ZMdiCpUBY!A zjHr@fq+M6znR(y+vUNh|80zIywHcZfCvlfX$sVmRa=Me-BmO+c>kzMxMXIyVl~+JD zm{*2w4L?m3ymhiUajdI*%-mSlbW00oqD-kdVm{f_aD(%6F`D9!$qt`^~#8fK~}ASaKA_wzG*O+bD_kU2%srxUNq#0za|8e!QkQ4BHPq(!3g};|gs}Z#=ON z5_#B}vmVWLq}{^OZx^aF9Y$^H8T*NuQeHGvV@>W3k}j7gjydD^P20=L-zmyP^W^xs zzxEpm=^1}Llf*0oRp zef#oGPaebb*RDWSNYH0wk!p+7fbz39VgfMfXI9ttotnL-RKZ&Yxc>}Ew~ch( z4XN#GZQL>D`ZX}0^&w{(t_IgH$=tOt3hRwBt>|2S?}yp%y(LL{NBXv#alM?X_N?=x zDNFZ}X{=)zG+pMdLYfn#7vB(A%{*G-MmbYB6&P1bzIpr1?i0d+?1C*G@B^)9ARWxu zw7zL7vT)@kdC+{IKCebiuT)e^U=QbQO>kIk9jzgJ#!Vaw2ec9G=HYX1PGE(JyZelNnot*aTEbLPP)DQc7DHY>=S-jb;tlS(GeUuLRWut1LSAz< zt+ckuV}x-*EKsMt^Id;qL(E_`PJH00+0qhl1_4or#r)f>=_pyBbokdbW$wm1%_6yW z;$*QhgAfA0njyKSfp6$I*)>@#F~eHfRQ!f+@4kD6Q&!e2yo%2v>XUu8CuhmJd#Gj^ zYVdzj<-Q;9B^uLBGI&t9@PsLv-=e0NZ0)R4(eD!6)plgoXYBP}mU)A%cstltJVmT4 znk2m2;2`2m0Ktqr^K5msTs@|EaiwB zJ;DtFWJ*dPw9$+8WL9D_1xUO$v0V8}4W6$aZ1oU@8 zhl7LhCg|Z5EXlwx_R|-Qk?tCo(>VeSu{jP!x#M{|_K+oJJ_x?#G=@jqO6JepnzeVQ zx+cqA!X7mwlJtIrFA!m#%Dq}4Ocm_7I1FvrbD4I?mW-Ry05h%ji{vO@4V;rwqgfg{ zU30n6++XVsFI9!PhabgDuQlt(lJv3tO?bO6$a0a9^_wN69jP*i)<_o>0J&`$vSnV9{DOSTFhs(?a-b%f$4^K zhhpe*U`F3Z4VYmI4MGyz?rKO!-M<^{VeeWSP!_(;XI}mGdDeIFoUhbeMnE@lJ}t!V zpT2mrqtLI(O|0sxjkeSox*lo;oQ%p$_Cfl7M$}zF4y|L-E7sRdyJY6Z`5kODHG5FO z{!6&e3ly>9SzABa8H9EbIq->;xf*Ny=e))x@g}4SHj=k_$Y$?|%GJFqw6}+xd7PeX z&3MY9y5-DXTos-o$zxncv-b{ZcBw0%9nfGlzc>jmk63pEaPmvWn}x!i%d#OV>WR(U zGXy!V*1wNF+Ys>+Yr_4kT~m$moF9*G@W6~iAT#8YRZI8CvnVG}XMwn?gM<+fa&(-4x5ZgJ#bVs@HNH zqFSacOqO*OD>~l?>hP>>yR;*F69qAAf)r!sn&MHPnL9faT*Rom7oJ45yqcZ0z51=) zi(dGWb#eI4mUra2ED}bmb9Ld-p-j$JGNp-mbm3ijRHm{i-%j7dV3p&YXt# z+OOJzdwdWb3Apb2r0cpciM>|vW~BRT0vsD{3_Q@rbK8U0eTj(oL1z4YwYKE=oip=r ziU^keW{j{ZJp`;J8zqSu)Eefl5VGZ2iZ;#EQ3dm4!M43%?m|6M!dkydBvi+oJ<7Hk z*VNI%wou0_R8eadNp9X2q#!48I0+RIaR^Y};1oa{RB^0YzQyTzk_pYq& zl1N*WXiWH^v)OQ9>xJWL!EzGsxC{+JysmlwMKA^cT2k9yUiRU+VR%(EArYV}AzO~} z3Tq1RT=~E2KMQPvC4Zq5Z^I@h$(#OP#7VLn+~`f-UfZ63u6vj87|5c;2L85)o-CD| zlkLF)icS@NbsbQ2^1JKMLVyVZ%qegnVY2}>Vt)rUCJHe(y<3_AmJxOt)<8YKy21Im zO4;|zY0J9!-V~rZIe$lWzWB)G0(pt$2!2yuBB1VDq=}3Z-^6)FVF2eb+?*G@4YptZ z;**&9)4jp3uWV+PgLH|Oj?Gk<4L7-;sWoFj1tuQ?N4>5_Eg1X?A>R1`A^sZk{|Q30 zBte*(yPK&@G=onFnM`v+H|0}4F$VyeSHw=i4I9vdmC-A7pze`Tp&~ecsZX_tUgtH~ zs84m_$K*cn&n<;8a7ms4MOk|R=GIdIW4`P+n|$++jedMKjeanHGWwwayKl3<6{CW? z{sBi^xr1M81k@ifBqt|8u|GyN0|uErAAkvY6kt^H5B8C%tiKLO>uYO(@#cpuTXfay z`Q}rimkXh%eM1vs0ryBNb_$78?V1Z)m_d_o|ttN0`D(YE=@C?d6Kham@Qp` zVIc6GMFw9Tw7FU3s0zFP#SN$&z4{Pve%Lnrb5&M82QNbUSeaL0KLADCPCOEV9rHbn zrhSH@!(r!{fA(*&(*6^wIF^a#vdN;ibJKmO_|ISyZk*VlSe?Rz!WYjm4{ywZH1w20 z#Cin{`km!7nmGvWP|2w0e=6PkrMdSz3)7acDQl9Pl3Q52P?<){!Z0;g7@#y2@8+&> zPv!rvP4-_rSV(BF*C~Vk>JHw}Aq~n&R-;tWjf3F&VRLLV*D%Y)n3^8PZZMN_E(2zg z;rf}s!bbY-dw!|;`z*a6>!eQ1t?*$iMRF~mAX{TNH9#>vfy=qfCL^WMc0miMLSXkT z==GQyD1;o52MS}R%|qRH)ku16Gn$g8F}ue%Q{1L>&bk1_cGDv!C(oj%y^L}p8migC+6FRHGhH4C z>56AT-2GuDbLDO$g0L6nqPI`glJ_%?chjZxOiE6iBh06ZWe(2Ttw;LXQrvF~a(Dn@ zOqsmdHPZIvvW2p;>*`-y%@;-t_D#b@m6o=AXSq1+_D$2wjbn!Dk4|#AuS~Wc6h*(i zrcXY2v9%=XOzYy2XZB%8s9RT_teT-(u<4NE1J!+}vvnbXQ&QE(R$hIJJw<~?xkz_B zqS~FPuD}p+v*=Om6E1PO2!4;}7pr1PADl{`<%8+VFK=Bwov3d-ACYTqKuk9`T*fxZdj zLl{NMQZ&?dr8!UbuvvB|I$LgozP*AVyg{~;u*_Y~sq6RF2nXnfG8M-c6@*0A4!Dm} z45Z&L9cZX~pCJ1zIjLq*8_I@QNUwU14vKi_8SXl?3@E*sqcVAvpVh>v2K zpq=EcaWF-cJ##Yd54=^!ex4Mv{?z5resPCq^Bc zlO=`f^F6<4D$MZRJ9TgUvTtX@oudPz^GU#|FZ%iD-rMMd6kR5}5bdb!vx0tHu+855 z-6Ojt)|{30$nfnlQBPI`Y7BQH3ZAMh27&x!$7>XT9Zkk$ucvat^}Z}PR=x}aavJhD06lkm z-z_nX?Q`ibmCM%a*EE!`qZ@+bY4wgiM4i(hs@YMQPZE4_-V*8*7YGm(X&2eU9f|cjW)Y2@3xs@3z2jYVt@bM4+UZrVC8>iq z$J>e0H(cAe>5$>$H?jp2GdWJBTR2cdyP6HV6m|6k&*i!p*i^a{b!Et7%pt98Mo4-1 z9mG}&-!+nb4Xy%2_(5(T!i_t65N*ozCHOh3Etp#!gl$Z=Dafw7ZgNW22JrGrYrgp` z7o#Dbu(G&UD>(feK5=RDcB8~sV1MO*rnrH7=@~VwZ5LBe36@Bg*a%&Y7rjljnC_A{ zCs=CwILggN?rKp{iK`XfnW=DzZNh90+^)BUkcu&JOK^(i-8h~utQ<7_v|WC5*_#S; ze#X7-@rioQ-l!4yhd27VFn)?YP0yssGWo339{kYKz6eMVFGI?N!{bc5nqJD6R43rm zlb-tdSk4OqLnRfBC3JCW1nxK`<~@xQ-T@g zlv40DDk(zb3p(@J%VMOI)WfP`t*`3X?U}Eh9>axiaCf*pLrx9S`9>!dl}}3^$aY!1 ztsZjBt(j*_I0?httv&K(uc|W);%(GzQ~d0V-<-m$oFr(n?l$^;Tv9nmUv7UwIhkV_ z*dbMz&8|-#{5eN^MpgX&z%o@UFSLbOAODy#mQbijxS;orS|fdq*TO{FGuN zvfQYgd4J97x#oD!I}WVmCEYf+!uPcW?OF4(^VsDlPqFDV*voF;Rrl8^TdFK z6IALxMX*LTXixV|z0XsGPiNnxGDwmuOGMZ#o`fGxI>EWiTUob#{w=@t+3e1FyukTE zYIk}I+f&q6e4arKt~XrcK-~V;cAA#v)(NvR#&~BHtAG@0q4l<8@8l0ZmrWi$MV`r? zPx~41Gi5sww#}NtR+~?`GeV`F&#~E=@VqfW?u=^Zd#c%x=Ms7@g2q1u09@fP;vno2 zS+hGJv2|W}Sg|nS>&<~=_a(VeTm{R9D66Is?*L+wV_*a+yPQFZiOjFelgxGkH{M~l zv|M`wQ?i6CR2j2yG|5AQ=epQ!*8RDBr`c7gW>YsHq;u#US}-3n#Ox)wctTHt-RO89 zfP%Q`+n{S0#Q?yDSAfveSx~Bm@Z3b^&SVk|AF2pK2G>l0djed2uQA0BfPR%Au)Dv} z)WS9%1vq1vICTnJI>AYJE173VsJ!yo^^!>b#OI9}IxK;vLd^yiD{d4YzBvFIYJw*K zRp6x2luqO{fuspx5sp*oVq=tvgIWWZn!sGrBS|6&PYzb2`(Hp9ardEIAlx}i(=K=d zBia)dYnWDZNc2nuY8t^t-tl$>?~mf_Qgn_i@4(&dNi8Y4$*9yxz?U@9+vyddS#s3}_q-AqMjvIedsB6C3jF*CJDs>*T#S_CdxSh9DnXZET{IYm&3TeF zRz2P~KT^|i-|>1j#YJY6tG|)WC2KE%vR85*1G>1%C5MX_eh>lGIkd;E<(r6CrKsz1 zObC8B$f6<(P!V=^*YYcFV?yS38ysfr?Q&WQX{LzC)~EF_V$DfOpZY&nM8H9SV^1~B!c+BO40eE7pchmY)KDkEi_LIS@vfxfx#kJg+5n<9(PKN)FXZ0KuRmWSXPlBZ!od{tn z*j(|(G>{C42HWSko*Wf)U5IKmdviZ-*R4bLyi^~O!749BnIx1@LIUyb-2BR^M$Zn< zOERh2%td+sd=d^FzX!tV&G$3KnP{zAgF4uLt?jP!&PYMOD|}ln@3|3|L=A7so(6b% zzM3wvLcabW$7)DcuV$K~SF3{Zfp!_Ab0W{3^?|vIHRBy@I6_#8;+6)dgcA{5-q`BY zyI)c=ISYG3byyC$8TJdPnSD?Ri1kVuY;s0AD@J^_=j5ohQ--IB9KmO>FG51FdGKn? zv(+VBFkLWZVDQV@%5HB{L2&1Dx$Fwe($@=diLjds{rMvXv&HVK)(6O*SdRCu;Tj!5 z4)W8F5RIrnUh4MI1#bEE=3zp40-@~t(zlPM8Yt&*%nsDKMdGs|( z&G|~9${u0J*unei6CZ$K{dbmtvbxtgBd|Tr7Dq{j-^%?&9~C1@uJ=&8CW2tYxdb)^ zXWOFzQ8~^=_kxE+HP%ac$uxA8K&A5f2bpZ52;?5^?gyrg$vqSY82n?DuZweLwM@mf z(}kOc*}S0->-cp%)c|E39*t#NCGtTy;x;iy&%&CYTxDK}^S z?Xw!iin`={k^4TzwDMuvL7E2hkY_vq^(;EjBpt8^3upSm>HFELyUHGM5fGO;WwQ-y zbX?0WoV#JQ_`baTDYb=d2<*{xv^&{&xZz66&fVeAmk_P3@b zd?;*Wq04N2{RjD5U`*PE|3V_!e zVD}I2D&J!T#!CND|88_StNcG8e{PzxZ$O>5HlR)$EC{4sGMDvBepsJz|H=A{459Pj zn2ybiTKsBPt}xqC3lt zEgYQgkwF0JI)R=n%tpklq>|;Qk0b7^`~Yi#M`+l8OMmZ9ecrUJAfl=Z}TBUmC!tPBzVS%*>tK zM03pqHql&wJ|~nktNnx7z1h?`1_yoejrE6kw8FI6O#5*wtQzfx$q11Q|AxVPo>H&=+9ov9yE}UTD@m<0SW1X z2t8f2Qq7*D(J?3$G?NAd}IS-Kp)s3ZhLHG zT@0GQw)Qto#jhEd|K-n^j%jCO;HG`PV(Lp*^ktnfbfH zqdIoxJiH+pgMaToVwdC$TMfSBzr9D z+EXRdz%Di4$Z=Nr*}ScFo70H|V5Js}z@!Ve&W2?YVgAgmq&@Dbfognj9k;9KxLwrI=R}TZSqYrL4}RV5U0F!3`MK2uU!bPvykjV_*@|h7J4rU1%Os6z?tvAHb)*W3 z6kGSMj)aIpEnf6OjIEtayF!*4@n^g8CdlI3YJ-Dep{DGTBO!!-$=V^Y=MBTyZPo)j zh_13P6F5SX%}Rro!{ zRi0Tg=^3`;?aQs-pvYaLyh~oeP%*?+Mw08}?3el8^3{2Vwmk|{$YhcxR&Zx2F0|tm zuj=hG!rh)-#pRuZ>I{{3+PaJnGK*)Z86- zIheC-&*}ovWv(F@A-Jt9@R8k}#r`zuqYu=Esi|*qh4ToPY`ND~qfAzFK4FPY8UL)} zTa(i!qcG10(>v4eDTWGRUohR1mnyt*@5?=C;@%?lX3a<^I}&3c@b1IzS3&CMt=!kU z#6oC%6j}W@H#!t0=Pn&)M&_(JEt{PzSZr~>n8o)cKhtx$xmp*t8v)U9OY}CamgpH9+*wu1 z23#$eg}&q1S{LEDwb<0+=POHmyak66^G{?3R8G?SN@Ac0nWrdX4B#i@WZcRJajWR_1Q`V=)tyL zd1vQrOTmyJ^i0C|1tf=0cO551yt$rpXM@5Tl%uWO(=L?>RdC$qwGG`3d-W|(Keo5O zBjj@EG&Ar6$wrM!>Pf$quUAe&*Q(H+j(UO=wzhOa{Yg%EJ%oLk!F|bDkmh1sbuh<$ zRT|a%)(FX}BpHq;L7pfauzXphd>bd}ZilE7A!V*5W_0@O>x+zq$*;09cLmW*tP#c} zkWSvKW^&J-3=b56=M8)&sE?=qWRQNds%edPO%JUKV?oI~!PuC`DfSHqY;I=PN8Ak~ zG(1%MWOn#MDsw9bTI1ke9On}SJ=o=<)5Vp=q2Hh$i)8_jucc0!S{AoFe>St259u&G zU!O0fVXVEjee*A!^P5!5f27p=|GVuUDh1u>Jcb5znQV^SDiU1;wz}kDGi1PAt`JzU zQUrJB?UP^j#K<|F@MxUU^t!TkO9v$NwKWgy<^sMIhOL?pYk|{9$6dp#xng<(ib6bB zr2a?0fjiXTg46+kya9gn?+Y%$-{495g)mZCVNnBX%@QX)x(L^LocKf4C36(q!+)>p z`cpA$5b|x*Lf+1^LHkHR_f)VkUk-E^gIjuqtQOLY1Kn8VE3Xai;Gx0ux$`~Yo3Eq} zh=R*`VnWFFAUTJow8@03>leXjsz|)M^2aw|pp_DDF++>9CYZDTgBeFs4$-MrZWrNwXIi~=#=bxHo-fj)D>w?m-nK~FSES~EefgFO3 z-P(vtMM3EH!t5R=B{ek?N5E;J%33vf$!ue9fBlY6n0;7KX$nvyi(w4uhm~R!!Hq5< zFwih&NGRx%LfBX@{U!ji(NxRr2ptj(^a!5H*rfhwu`=r6BaHpj9R`1s22QSTtPbPc zivP6x|MZ4raH|8s>tCVxoZh7PJpQkt_^=Iw#7wX*U;+G~^b}3klY(HRph^77_(o{h z)r7Ng9AhzU!@K}fY$Rfhzs9c z`Pxg4d0g-fe&&;1BB}4xk9(nPht)5X3Q`##QKwmFI&qAv}ZZ_G=TgvU~q3tQymq zi(Z_eOZ-?Aqq+a_GWpXug^QJk14FAHF(S?kCq`KTC9YvKH=vN1!Lb9j&&*rGefZqT zjdIx%8xu;@fVc+gS)-ZZ+SV2Ss9g5*ALX(S{paPfzcqh=tpi-8pOcXe2l@3wJjGS; zC+s_PNX&SE-WNMcE?EvA`_A%wc!7oiZ_$lHmvyda500Vdc^QT?r5m%;G*b0{_NM*6 zJiHW%8=g*^+XvWr!S+!MgtxBS>DCUJO*F1O=zbl^_iS|I_D^`sqWugY1H<>BF8u1S z_LqNlr5anraPtp3zFeKrbF;Ydux+i`*X7iu`(!i*o9A{Pn6PuGgEZSRId$d?YT`=X z7BP*59_A7JX{I!@9uVH3+lF-AJyLDFGpi@{WtXD|PQh27D-EOfn9Xh%$!HQ3NxN|^ zz*JzHaINusX`MuJVwDn%quea7PgdAFyqcP-o!=oY$~w9C6y3RN8QZHhT6-z03w5$f z(GR|G_FmRby5ub6DLPlSn($Ods@qr8S0;@fGADgb6k?QlpYRNkJ_7GC7!|LI4f5m+ zh&x%*mGzun6z8WYqsl3TNL|V{(OeC?V30au-4461u6CYd%&S{Ojgd17>v~Vv3#?6f zNHj?f=ocZM!FTznQr;0Ie4$qz;c@ZSng*so>CLBe_GQIA4ML0Y-N_W=)n1Z)kie0O z02evpd)Q#}gvFAELv*Q`b)|N$Ba^lDZlU1eA+yvi+EbRdvy_|;@W0s4Y|N4SG}0Zo za{f`|nwX}@;T+DDbit7Wp@hY?dY{3zQ@=VWybjiV&}Y^z6=5_uC$v3<6T>0}tNJ~pzvwDGQEMo9gy zde&x%m${rpx=M7)U6Qol^gTj9$eO`Cpshw%+A^VNzS2s__CYq( zoi{tOn-A@ED$B1q$l_V^r#&ansQY5awdFwUv#K7TXB`_>XFud*nvpb6QHDc~JJ-pS z>tqgcSFWB>-#3!qn-pEFC;5>um#ZkFmC5U_fOKyZME79M>yUE^SLsQvCKpU*PMpXKmQ=7`6QKdF(Lu*OmvGA#EA(X{t?V zav_J-P|M!;1o1_Iq-`0cb4HJvXr=VyseubPgyUUl`cdK_ytbYD{%Ry8aN7LI2i#Q> zZZ?XnTNh^)Q}{J%xL*LeSg-e`5i*}Q45?3z7ASjeZ6FbySZKX;A+o9aZr(OXdL~LR zw-bC?QRRUa99`qIs#l#r=uzn6DRwj~+=>DUMUwaeEF|C9&!MC{_80qY$*%W!JD7HIiIYoRKDAGM=Wh4KKL+)*S&Mn`uXD_JRf zxM^U_JnNzJwYh+8%)&J!&&$`=Kt>o5z3vI2bHN%cBHz2YWKn3EQ8n4e3kM>*a*E|@ zIVGT0H=ZuRQ!0g8UFgd3Q+;~zlFqhnLIsv(bDHe(>!y!+@4QY~5Wi!)v&!Q52gicx zonOvsI5{wf({dL!NH0IV1y*TC9{#fc+W!~tuiOO=`{xcg&Zp%NT475?x1?k39`SL94$h zBdqB3UsF=_jBfncGy>ddZTDsZ_OUoluL*Z@qfPBMZ5IdO&QCToWR&c5^y6P1rnB*O z&p$b*ooi9bAf?s2fzRt~kezY225pX#-!ru`4XBGhWD+1y@b&Q`GqqM-^E<6wB1~ z_|5{E;%-^|vCIrx|HX9{O!$kb)v6z<)kC^}O0E7Ok@i;$zxb>@vSc1p=VO+GBv`An zMSmuw8YDV+QbIuIg5?16Yc8F$1|#;J9TY;^uXSHG9f-hA!}6cif7$YpZsJCX_cYiP zMi^CsfZ$KD)Hg6z&%mz~P6sY6_+=-vwoT#DZ(LZ^%z!qf);=2Dm;xVxzg>|ALn?u~ z`|FUROq>aeo>1_$a|2DlQ-pV?jCH@70d2y+ZF9vH)^6JTAWS>()xYkMW2bKYlRK&6 z7+@H{zOez)3Q#h<@hzHO#DoH07W66sP)r%yxBcF-@V1vXaRsyx!xisd zbT-Ni!TEksG5bPmrRfV6iu#@Gr|8j6YA-wz#;>MA*p7D;avFSDm#RK${j+;j5%wGZ zs&ySI9j^KEQ;ed+iWA6ouMV76G~y)P>^a?C-0HaMVXaU6>GAIO!XNf@!DelV zr)cghH8OJNi@oi|PZ-I1L+y&im&MLd~@Xea70DRToPB z2zKwJMde$Mlu2jT5nHjN(#Ku<3=ns7bgKdF6WEj-kbI^iV+*N4?iXiLzxd+lyt}KE zQQVZdk^b~Vbgnm6zqm3#I&e?Rk@{)%MM5$PS>R`KWl=EM6B%K9s%bEBnX?8&ns(ut zM;Jx*2W&gD?IF~?lDK>dk+^#t_^*Hs#OX*4(Djz|j^0YlgRJWCzIDrSQYvpi%Goji zHC@H(^PtI#X;icT5Z{mFUbL272ioz|Orbxpj z1+N&7+fl38*!hpgKNc4FkC)zOi1N;x!AZJ=Ehoj=YpE;v+%V4NtX}+Zl(6vND}+am z_6h*+*s%^UC!p*Zbqzm<9=`s(;ox_cPv`Lqxi)#Q;V28IV#q$v_{wNFQ!ED@tW~lI zRm?K>&`jGBT_%xwc`0HpY*ea?%%G)l&m2w@?JS5z$5@+wVVNoO`a_-bmeNa%)E5J78l|8zwcOUOhJbjq6;r5i|dSfpz9j z`9>-kjhri4DWJ2pu0}%o4%R*zKKd+C2m6+nDh+09uQ-t{;GUrD)~N(d;C33m<2jZ# z^~|de4#sTvq;s)8X*Cv*E3C3b*w&4{g$eg!Us3gQN6|bWDA`G<#&9`pshlmoq_%6N zNk3vT1~S`9UaJe?$(xGdd?bx!qQv;PwlF6p{ox5Gwcf>_@3e)_$FrbbG7CK9gv_C- z^<|uN?&&onhcR+$E%E{b7TSFto|oGGqI0H_jq#MZpR(g7D9fB&`PR1_2oktIVXcNe zC}u^5Ms{OEV@mh#QITb5owwa1#aB0QJ!ZMwVHfTxr?~;xDa58ZU0}JL5atMKm4tgK zxW;eaXifD0?8FEEvNI$$)x%~ux%gRyypF4Klk6}^;xv7TgH_E^TF!k1I7Cb z67t(PMD9FAQXEgBbKQ!G=)2aONx}UXslsEr3S{&tM{U&z_G^Z zwCpU|=(+n!j!TmH)d&p{+|L|HB<*K9TN_i=TY7owqlL3+Y;wEj*3$tO?_>`oUfl`} zOygbuaCPWq5d1ah2zI7E-jEEP}^IQ(ZZE z5A{8CWUfk@5Sqb#OQa_)!@@P)F|#+h^YxTF>+%iQ+hu5{^O5{~3ptDYe2Av}h8+8+ zWn;Mw(Ka;LZbOF6)_QS6l?^E5(HkP|U*5zznQ~)8!j0tlyrJmkOo#@G?*m)@kNU4B zp3+9`=kxtCo7#r9O@Bq7okphR`~H;db;H2VXD`-S`pf4Kp8dz{?@hAcPE#C*GZ*?a zMomIv!al~Fzjm}E7&bceW$%^n-q?A}>g#)1v6;>LfW@>XE};Bms)5%8Y(Co-B0rk- zLO3`PC(a5CAaT7j!_a7wc$H)78KVV@vVNK5v@M_8Z0B*eLtd~@Aqf)Rr+gnwG{P^__ zt=4g8=jz0;ZOQjaSIt<^{TEj9%7WvLE6==LM!+l-Ja&3`meE$#icW9K&BP5}-AQ-s zY@jQW6@1qM4+p(Q=Cz)aOY!-@Z%RKyugWPtQV~>DtuICW`26m-Vd{^Gq4%`mugoH0 zG`Z}In!FvhwmVpR7{gRtlbv;TUB&0a{khHt9P>O(RUG&X`&n{nW<+SNAyn;J3dKw7 z8O~LW=)&WE=jNLpnbze~kT<9lgbL{ql!YYxfY^8;&oR=_lFugeW4t^l85Ju|&dL{W zdYq~^4}uh2`ABn8`mm#+U-6g4{O>IC34o6Djq}0ZX=iHY+*g>LHj?IAyMZe&XJUCW zT8npzH;Q)4TX=y;XmFsLxNNb8S?nLG9i8_fKQ=dprm{CM%PyB`WxHfgy|d{n@wT(B z=@1^K`PU(nwdal*X9~q=Pq)9o1t!|4o=RTQ++0{@>Y-=<(&Cw^t2}Y8o2isr^u*4r zDD*I_PYw!ug}YW#7W{QcXCmx+fTGlV1Iw0gbwqUXza54 zf~4%Cdg#V3#G4)3`j>iVD)0P}XKyN}9~uyK!mOfdaopPRlsn!L4gE45HCz*~_XoIr zfP4;4JNxy5MB8g2oiv$LaP)7RD!xal!wqbql`IR0h!5r*2AsRIia(Z$8cMJ^fud|h zemu5Asq3hvvs_L1HqGKwtM;PkeiiC`Iif8gJvgGw0fF6T`s`LUqC%>2t{TvEMb|fL z72*IDKKa5uy2>p?{PbZgeLLtcnrkKKnum|Qm0aA2`Vl!FvL-a~!naxHYfb=HBiepW z3a_uvt(We%-Kb@`^JAX8YV|LZ2XPgO)Hn_zmtb{_5Ft@?|LB79eW z&HJ@gbcyFO>pb=oBrK;!tm0#sd+bHpg;sY^@H})=m2Y=lCD;{&Cx9OfazZVDAr0`5 z4URNGy#Pyk4-%~q3BR~FSbIw}ID`kMu3x!Vz3OU;72Y<{=mVnK1fgEPd1hEoW0=N?^jUKt{%#BuK%FBxl|d7V+;I5TB+IeZ0(JR?R&CpZb32FQ+NS5@>kqH^c(_S!%beQ95gZH45 z;x{Y-SR}te`ZI^$doO)B>5bDZaxBDgraPJ{e$unkxAL2X!P?Q;NcHm{g)Y3?lXTU; zrCUq5;N6O?sruf}!pFT;$T$N}%9{3gYBgFwEHfn~+(S&(#ccmPY(_-rI zvVMSga}ADmqV`I3V~P_q{MI4KpL$28vTOveRCHek@jLFr+~&Q(s_WEWIuQQa0R6bO z{5QXm;8e&Hb^Ld0U(}orvnz*54#*zkYwyILwJ=US9@?sutYwn?X=_v+fp6C;1y%R- zg*@vW+X*ZYGcd7@!B&A>po!51zOy`5CFQgU_Nd|ZIlXuD6sqA6+`srUdcfG9+3IFN zBJ>be%}Ol$(NCaCdESFp^pd8!b(iFYH9rTV$t^z3Ul;?7VjGlg5axJwo0qQFVB8KL zOzOvXr@ZbKhR@Rw`0R&ZIbxv-ze16J_2;o{{ggt5pPqF^8{8BxRCZieIJrwxrhU?( z56QtIYp7QTb6ifU@jZA|yeq138${#h^yT;+ZF9-}bGil8OKi>fBKz4wd)g(0EK){5 zB#*2WQ$L|o>0aIBPb-2A+*+F7j@pAbL-D}TPmoRD65+g4#tvm)k{&6Re+p$>USsRR zr@SehvoR!(f`V}^RDHS4~2@5I#VyS}~;k*lQm&{K)l8NoYPSl_+<6nj;- zRJW#5s9o*o+g=B%;?sk!QTF@+LtTSgMV;7Kwpb{OC857{-*duc@C@`Y1lZGY4;a6NzT#?(&yQ^U*^$hYTc$8q=<*^>W~C!VBsv8^XT z`h3imbmyVqzVSGbh@tL6kCo%;dJs|p;v~5L$~*k_euD+?i=7Mt38no8!AIdHVjfWU=e5jzF3y*r*`(gy5;ul1rXT0jJjCho0(l@< z^26(KBYh9ATa>;+C$rdelfgR=qq?^2mv* z9XLZhr8h0@?q1s8U=0SnggRJ0F6c&AfxF_`M2ALT>v^@R3z!qUo#eBqHwhK^wVsp&a{Ce1ha@Q@UMkmg)rQx{9QM0zB%z87u zJ9iJXx>~(xPW*h;7)!PP$XTU{6q5S-=6u>nVg|6ms_H_{X9Nr&obvet-61Jo3IUaixOKStiX7! zW}+0Ha|N4UI`FKisLZCBdlWaVvLBum5j`TiX9_CYcC%fvLaE!deVn%19VYd}!1m!O zk=$v?&Z^Pm0R7k93J`-uXrR=Wi3j=-=a-C;sy>NJk_Hrf{Gb*`jlm5@*ZEJJA99=) z{t(p7$nIlS5!eeqMT->|b}H!6*L{&wJ)CJ=nWIx}?|_K%j{(pMs`No*@e62#UFqdQ zS6vVp?nSi!A+m%0Zp(%1qHD^Sg23$Etqw-v17x4ZfGj`C8K~2hg1Ug3hPYqq~U|$Y1+dtxC;O z&o!{MF3Bli=JkgtykOa_XE)I`N>0avQJ9(W`qhH$!%RQOv3<;QhTwGRTDiR6W7D6v>6`4!n9A0w$CfU57aBW49 zQdK1~UmHDfeT<OIL6E&XP}8PBfqFdNlr`>u_>&X0a)BRwtXfNRIZ$ zL2l~ZsaaZq`pJF0`84odqVNS|86Lg}mPWb%H-Iz#(EC^FGI01CfG*F+<=tKt+32}) zar&UIP|C8z$ho<1EGM+dEGHY+D+?0JZc~DY8OL9)slN06SaR@VAs_DBd-+A(8o+hK zm8;C9ct33O^79zamz#@|JJmPc(;~32HAWKB2r^{8oD(-e4OyOFd*{#9Y*r6&a2MNX`t|b^lZia9_{IFBm5D-g8Zq$ z0+~0xHQwAw znyB6NE-HOu)=InF#Ae$#!Z6e5;{y~vnkprAHJB{r+x1Q?my~q}Qvh#d_sQ6M(RB1s zTz{FYk@~+i7tu6OX&w$ zcrWi^VVRUYvwbFQ&+FfGus(_E0dR7mMtM1ldHa_(au7VSbMIj z?d2M~4@!L4iUTyXQJ44m#Qr;8*ED&I_HB^ROx=RiRIiN?IHy%)Qdvsc7l!?OM%7y% zun7m2s6W|Jkd_jwKj|JGqUX9nIw5c%bZ95(=VJPDcPaX0@ma zxuoT5&Dv0={N#u+Z=dB?-~NFU+H5WMI3hez_yox&aCDB{MX*-o#=)Z4liTA0F%5#AKb2JtX@&O2{6CB1ZOovL)+e-?N*LUG`g<*LBTsX&q*;Y3%o#*Tf3xMu+j&!T-2*A7SgG`7IS(0}Y ztg~wakvrBvkl=yYY@kMXJN({EYlJ0W*oWl+OApxS2lkLJmj0^?(9aJ&an?dpAHqJK z7c~S5u>i{*ypGj&AO^LY0jT!Hn16k#$rC8js+G)y12^)`u|k78w6Z@6Zm1SSu5C!o9*+`leRYk_j?PT%zp2-N+m*?D&rVr1UXS5uXEp?|a(b+1( z!mrhy7yCx8YRvK3;-QqD)<7RS;2Q$fm5$^7zt7LTkf+tu*)p~=j*o6;gtT%UQA(r>Oe#FRo6(HJ@T{z2Mi1U$fX%8mc|yi5tIP7S zgm{8OjzZ+MuH0#D7&LOeSLbr}(UmebIh3+aA_E)0u+3;e3=t8$#A0YEL2r}9Zfcoi zDe86yaGVE4`I5`tige5O=?ITt`H-JrUlZPN4F^B`9)fCSpS)N}bV97)1sK_=nF(U1c8)UOYAJPDjh#B|(QpmBV+Mvk#OLw?qeq?3Cx^xt6`VEM89=z0WKS zb)NaTv$69bzA%WN1b@vrhX@3$LM{ZUGv5K^R&mg1+_g>?0g)`FEv6j4||E z8ktWA4Sd>-etwq1x3r>z?SDF)Avt3kCETo*ZSiL1HGi!u_MI;dz0a+34Yb`whNC`W zbI)Z|XKrr+2l%0o0oWaN|7;k05-z5FL8>&`)gv9JGyf&&g| z(aS_7RJG#;A7d6rsiDhJXfk8XQ2fd8P6@>`)S2)nG{g&hd01e^EH<)yfj@1?wqqPV z(#ah%2u@`LSmgmRLDmO>X)l8L2l$dwZ-fH{`5*%0r7VI6uJFS+B>u^*p36dPCAkWRjU_iFgH%%&aV$|+gYh=uoG$sKW(HQITII#In39(g z`G`|M{eQKT)<)|df8FQ%-O$K?>zbsFDCL=7OqXdqzN*KVm;YR%jD<@&^u1lkw&rUd zYMJ!`j(;S5#qkO2qJm$2(4`vT(ETUto`aaN)pT`kASCqGv{D zXH%~DsU|Da`kacx52BCNvq-z&k5AcblzgP~FDzrS#oS=~tYKkc#CbDg<@hj@=e)A+ zcXAltfi}sD5)2g{hun&~dwEBW3gZEk+!$riZm1J>}c)PycEEhbFJ)#C@t6XXb@7FdU;hX$9mrE74UOQR#(%)<64&+>S_+D~9UpT}1Kyx$Wp2VkF6(byoWn7}JB#dBLQgyj$2NS%u(JB>@$-n^P$9=pkEb^F0Dw zwQ#Xmc@Qb680l;26y4jm7^AHa$5$urfsz8C_EC3y=B>RAG(d8aD@OJnw5{~ zT(;AsXwMq;a-OcYK79Aag?jA*-U(@*&Fp+jBGvsuZKdo4xm!43d>mW!(*%t=du!2b z&fCZQM+7d70tcozAHbW?v)gfzssx|}*`u=bOlLxHd`ISsj3K-GCuhqlm3fR@V+5p> zDmPu{pgaU0w|D!lLGGDslEI5u$Z`kcirLBy_rGP*{_cNQ;`04% zP~N3`yCf7c`s9q_C)KjGKQ{6E9NZRnrr}9}J`y(6lf5DP_w5>A^OOt2i5X`7th z*#-&M=gm66Dt&$iEaabT9@7ZD2+vw3r&oxEEyQP?m*YKUU~`67?uf?+%#u8bp4#d| zy5{!2ABv z+JQ!?!|6#CQ1daygY7#!MLhSNv)H#KlIOy0j|WfThKMqt!~l?n8*mx}Y;_Oc zu~(5j3RGtJ%nF0&vL@<$75nyurSAMUI%dKQ-3BY=X_pbxPq^n{J-NRR%lPSE81Rla z*FZgR>i+O|>MlZY0~BOcv_PK`L-6VDj9J8|+_<~G)4>Z{X~H=l^39Nb;>QpYkbKxI z3uN{L>5^H`jQBqJp(e@qQE`{Fo@qL~7kB5h+4BC@vm@ek`)Prma1) zvL8eQWm+Frx*5@(YN+H;K+Li{XQs<$5>a4-~-8%`ELch7-$Oy%p?`${l=n#{4hTikUAoZT?*srDFXzH|iIkIFM!?WTAl zjO!qVh#u|=Dn~(79JX%}VK8kHy(>PQ=1I`451i5HqioYe4zD|JeYkwOu0%=my{f(K zxS4&}9lTM8^6Ne&n*WeB*eq^;&jg>Nt*ywGm&? zp9XOoa`nJ0P;c@9K|GL=e`Hiv2roamdktimkfFDtPLlXwJbRNIe7cT?wKpWwQ(bN= z`iNO8!Win`gQ{5r`LlE3sx$ZljwjeOS?btX{(L!jXc_;xNRDr5oVtVgvk4R#>=PV1;El++(Us+#PKmxN zYM1JECGw7Yz-%4tQhJ;R{D~+C)`R@w{q_Uc;IsGRE~lRU?EUt!zJqof;2T5$@0ZLy zhpDoeFbwIx8OCk=NfrpHwp;;1szZQJO-3QtVS^CFB&fUhvq8IqV*?sWuUj;kk5>VU zCci&E8hMs2ZG!htc?rH@6oD5p3)FAWD-80N9&DYbHV-#5TjY4xM&9khqZfLV-^ke3 zuF`@ULM}4lU4If9pre8F*m!>=9@r|eSvyuAPV89p`1RIT0#{EK8?nCIcW%=-|J6OL z9=@Q}IUrV&DKCC z6Ik!Sw1S?T+F8K4_X0Q!yfGZXwAi;_iLM3y8a2zpJ=09OOh8oHi{!eomRin~n2VI< zhPn&yoo3zq6f??Eg1CIIF?=s)*10;yWCNSCI~k66e1t9J!3z-^L{v(2J-L6%gDCa& zl_m8w+1iTQb+yFF^S7V6833v?OBzy=^ zl8PZoRt&kSO~1#(NGwbuYK>_kzdc+ z`;Ta_^iw$m5i0(gOM1!Tzuu<4(93)mRyFW@CReeQQS+GI{}C%u<5CDNnme~+V1KzsImj-#NSWb@B)6sNd5GP_|m z%y2!9QiArO`mX*IOYm(7S83X=rU5;<{3r>)X$lssML5<^8r_{@cB`bzt@-q!^QsU| z#q0~R_|ddiYRY9X42_mOj3cMhaBH9uIv*FSGzAX%6m1dB@|jW1UXRDIO7BS*v|L_2 z-ux04^_yFB1-QvCQTbK3mM80M;ew0ch~+w4sC%?{Gr$(!%@LKO7Cv;Fb(?}M>s?%7U~ER+QNun2t7uF4PxYp8XSaZ5PYfqs#I!R>4`)?r-Rc9m7RtbB27e?{om z-=BFy6vYK_tJwb9?=mbWmVSn>Xnw{iZyY#MpnrAMo_+pS8+BGLBpDFm=Iif3mdZ2$ zlfSUA8^NkS*OY%B;@z;pftx-)3K!5n{1Grv%D$o&~kO|&T5a_LV65C_k#T?;DEIu%GE?2Mr zGE6_DcQ*qQ5e|$u=R;J4N&%-ITG|0G4|LlVA0T_f1*sii;JAn{`py8@gt`}JOASh% zf)(CzO20qxRDu5(ci63mPdSyNdSAH-`R{F3g2YFV)c}*)2{hqpchIC(Hoz$Yr7;Zd zd>DXF0Q&DDs^EFyL3s?@2(oJ+x%NE1lfsH>^0oyDzKr`6_RCxhJh6MWo%>)l&3KH| zMK}TgSxk`8$zp(@)ic0Q!*8|F0OwC`5(d^l(oo=x8u{P=Vu8&Wy*?{o^BiHCbb@Tt zE}<*vNNoW?0>r^AyLZi0dcYBkp{Qk<`N}I5gD@=?$DB6eoYkJ3*C0ZVh*|<*IPZso69Gdn0qkq;a#3r;VWY7z;s0o?d4&1Xu^9hdq1PuT>LWz{UvG6ZFf2>~mmu6|W!~yic=bN}!Bf!(C*&I`tr9 z{{dca&=V2=;4buivwE}{cvTDuFlxumAVg^~5JK!30CQc@*cIw_Mx6IB%-uz7cvhq` zn$ANJxAld2YEP9t!XbxKr7}DAt)?u^!?(4Y{l_;iXL|fV*l$vvqDQ;jcCCRvyof_g z?i8EKr5S~12f347u|71*~c)@?B?*U>RWNQr=Clf)n2d; zxv<>x^MTb2f6cHBqJOZfvx@s^$h~fV&9F&41T^|P!-jI_q5bQI4Wje|Q!7W_t3tgV zSp~r#*G5d*2Do*!(2ZICZ|vGnc}bd4WS-mu7X(gx(EG8=3GBelwk7!9sfm+-Y^%O~ zn|~$a19d6>3E#Pb!@IYQj-mMxL-k1Dc=?100WiiMtn*Z9R8IsCEw*=Ng6)Ma-yKZS?pIihxMGddce$hztaS-n=fBYoJ$2K)!^!(-OV9 z8?6E{0MigiL-y~`y9khBsaNq+{YP4cPYVqguk33&+?7(v>YpWB)_Ka?_@Rl6|O1@tuAtDF}Y?Gwc+8uWBK}|wvY?y zQ=|9Ja6FA5XZ>g$OpRk0Dl8wvu~(4(kXDfG z$Oo`W-$Fz>z`}3rH-dpQ%nyJT96f-&iC7ha_@GyIu7TPU65K#7>mUokyG*&cG2;Mt zhpfMcRP%4|p`8GrI@llsnF&mV<_ud>fF3y*lM7&8%v2J2PYu<~h*Ac@Q_;NB%PQ1u zfVzq0PzRAD#07Bbx{v;|Iz6UO@Vp>$L~}Ki#3+WO9E~VZD^9}$X5gthp$=GD!<;ox zyoeIrqB>xC=m10fjS`x50_fn*Jb(^75cuQ914QXvP8omtsq5eB2b%@^vR4OCfI}-E zz#rED<_Fm>yU<(!f4^IpFWdb!$_FlP=(8PRKQK(_1V~Q}RncgKo8lAEOnuS>Z!!S= zz*|tw_=qasU=~rDsoV{$_Mh#>zvA`(rr-LnSik?jyeHFoAhiVmI!_a^`4pcpG*R0X zo+tK$s~5l^*Z`1WmVvIGvjqAheRyHJ21@7>!+wb)Hv=VZPZM!aicdJ4sI7kioF}?L z(RW$r+~t6$83~bVpnJ`fa~iP0U4J)fMW`~7(=f{$Gs#=r0GR#RCX=Lo3w{;tLP?bG z?J3A54W5d`N`qH3kSm*kA8I%u+yZsT(2vV4-eZ5CBNBKxVFW zA(p&Mmo*kpUz?B%;3k9htWg_sCTblCVJ!h91VET8x+ql??>9yS2z-78fak0b|5~yFatIW(}~>z`W*Rxvtz&Q`gd1FfB%G^{RJZ4uN?+DqMrI}4fJ({ z2n2|rb=dx`w^&U_O+xY4Kf4$4b@u{<&he*O=u-bI(Y2BN_!r#s-}P&Mw(tM-ufc}% z$=@Iow+_tr7ePM+mzW>HP>@^42IZb+0T?I;re1@70f3VxaCb2hOKrA<53#fHw zZzGdU>D* zOs`v$PV+jCg#{pd3jMxF0h;V=a)a7Y3*`j3A#?Dlb zffUxmIDiMmyC|-K-WpLOX8Bf#Zz;}$1MV)WY0y+NH#%&8R?0{7MU>@|pX1n5Q%~*p z6SszL(mZ_!vWsCDWB=M(FB4C!fp)H7bErN^9>DjK22f-5bPN_*F+fjXn1y5g31)KG z`JOkFIMVTQKKOYHyZMCFt7In5O6(=@7WY`Q87sDn{lW|m*`5_|;Ba{v(5AwpSfdX2;|GpEneBa?oOlqeq z1e(_-N-|?*HNKl}u6OImn|yz{!^+9@$-Y~eE`UI1K&l%%c=&9$uMbj0_95Hj#xXtk~uo>V|B-Vq;# z_a6q-A_gwM_raV^iSQ1SL&DPg`F*$FP`!T6KrX@dr#$`dFztuHHCb$z6b9Blch@rG zRY_K%PfT>run_0{>UU`zDF+RXJh?dNp#>p}43sw$k4G~A;nXhF8i*AT9G&l+j|Tc% zF)j-D(X{(yqpO@QsQ!Z8KCw@f%9i0HZLi zl7U|`@Jj}M$-ple_$33sWZ;(!{E~tH3K?K(CREf!FuTvVb!JQnsi%NBFVBTQ4KCJE zb@HOFK1pCc7g%R&m^*nrxAUBp1vIuZY75gM*cWp5#LHqGww5lTRtsC^k5EdEbPZ+6 zEI1PXkPw$*Z92ORvN)TVhM2h8xu|;PDed)3asBCJY?#3fiO&7HRBkFvsy?<9 zIFaYsibGJ{=FwPXfQCC1M`f0)S_7?;fLh|R%Mh|DP@kp#dSG!1aeoaYl0>;)4Gh1S zLJ)wQ+$c0>0_6`GvT=a}gEWff$@`@NQWqeyzj;TBVFjW&h0;K-`1SOE!m5Rjcvqzv z4xDUikC_z+-2;x-%gaUd)KnEJ)+!oZ%jq~a7kQ{3`fdS&>v@Yiy}jtVa8N2xP|`SF zw(1K@;(@lo$AXtjn8v^iXk*JA4hxs_Od@4Fo)2ma?4d>C!LNvi!KO<@qJ-4(KwkE? zN6*587jC5==h&jst#@QZ5HhU!19k7+K$Q*KM(&=A%(B{eKATsznipV)RWwSp1A5B$#R6Lx_W87Ey}ht zbXc+JXj)2=gy8UGNW1c`q3N!7$~Hr<^d~q+z}O>vMZ1uZ?s}d-Ecv(*71`$3YJ5&!iwt8W|{4G>OLlUmKlPz z^Tt)EZRovuUWVCMj;KL+aIwnp^9)QU(x%v0$p07JB+LE7cV`2Ok3_ z{G@ysSfxOLQSb<|58jmt=l!i>fw6^Bf`Gf|TcOe)1yU6^|3_00zaIWKFBryaq zn;7rvDdS-O5MescVAft^meErebN%D?Zt@3_xp$k;k84s;ck>Fpi2djNOyb@?yjZ~H zhqG7j>!rMDzHeOX)els*KEAU3DMnb&|%(BuE#<`3Z1}Q zJa6+^_(TB>5x(??*Wl$#OTKG9`nDkAyaNYowE8XHE<3p4Y4xQhUBbytFAmF@6|KZaOEiG{qPwN znoC-~TFoQ?`UMX7EIa5hH{z7X?x{~$U-OxcN35h`n-cgQu+F;}FFi|)7nN$xuj^oI_De?9$ATeYwB zMC~>u!HvUsCd28;^;9=qd;Qm7{d{w=$`8=I?WmjR#_%$rHFVAmhBhcUC<|!gSIkDz znmqTcRg>U-?Ww%sdBwj`5$DXO>9?le2?)CRv5q^4MsDpZ}WahZ8R}J6L7;=2Y z!sY#WzG{QEH@uNS(bMx9%s?~#B?kVlTJk3rrJFW3GTpqBa~LFPRx9C1y6ylk7m-YS z$yzC%_N2VCygaF;?mc4g2b_mkDso#XdyV8kI^UhFa?#jowfd-2UL?NlpR=LMmCa#9lK&4?Zx_+F? zjj?f}CVIOf!ggRWYsr-ty$@dUO<53^UB`8nUFR#ixmwoeO!%s=05QG_`jLeDUe5j$pc5 zm(zqQ^08>-j(b)r`fHtfTVf{LI(6RrYpeq{n1=0Womj63Q6!eQq`cI&e zzU4wvZy}5Yea~5nUvH-Y0?BixDC)yBD8;aRIZ12*KE_=}mdbLklg+t7h?C8bl@^o>77-*W$bO5GcDX~ZSV_v^K`*CUqjxa5gzrPy z*z;?4&$#V-jVMlFNKlxR}Olhwg>qLM6l4ci&i`=1QLSUFBA%q!+`| zw4Sxv6iG@6)GKFlk)gmj2DWr>!U)cEGQY~o#%JawBNvZy;yHuK@Y~E~5|+0ZZ|kjN3M5H%bbKRsu9iU@S&)2nCdn ztK{i2KNh_r&~cb0{dwicRK^qNFrS8t{DY^ttFNZe^9=sQx=-Z!oXP-t!aHn&SRq(vFf$K@$-06$gpFM@9N?po3^wzP^Hg<%o zbpi+yO!dBsrD3$V9V13wQfz!x&i44GD>kolD(WHqXg%C|qyl zu_a_V^$tc4i!FI81b9!gBz_y=!d_5wbe&YziB3EMU^+Gan(p zG~gUTZ!<7aU8*6r^cHqt)-|=~$}!wlA4`Xr!3M)>X0Ynrhlss8i-S<#XjX!s1^5bD z<^EQkdPIl=HSMEwrd)~B2Z;f?K$qO4c{?Wv0m4yUGwMJtu*M|Kt!SVsETaMoeR)u~ zvHuerZ|wBNraxlozrCLiqJi6r9O^w?*kKXUwPmy@nwR@byWh*ckg0lqgR?wU+Buhe zaoYO$m=+sSwQtYX5`UXeFsN(ZgDOR^tlGEDpg>`V*O@D$cf(Md!uJ}erjoYlaWqX6 zXPi}qC8|RqX6FRMiIWABxhkrUd|9sF^!K=qij@GS(EYrv)&HH>>u_+NlM>*eCTaCj z^EX{qEC-FY1j#7R$%C9DeCkDq$GPwLJ>t4?bq~z)^1T68H&IRU$##zXyeo6d(+S77 z9VTa?phCD_Njf*zT`p+-A&T{|UB>DB%>HW)LJeB-4e}FvZ0I3JKin{}g1e~U67V_c zSK8v_$C~PD;9*=#pq;K;oa&EN`#yf9D!VE99c>%>bc&i|Crl%wlaumcWf1{1IqN?4 zRzrOklWfTK9G#^iN5Sdlc0M_5X85aPehOE*WPp6pd+{rXK{=vYsRm8e(nWvqR*7?F^d|8_zG+q z^KVm@d8gIs+`wvnNsxDH+Z+dVCi$RSp5*?h`1I;Xze0)WDloln#90AR>rV=pnaukc z)JaJQyHw~0G9x8Muvk1R9%)CbV6K!2rn;%9tx&$D$rLR}C3;(CnjSGK zTW_K7{*WB865N4UF5#<9k#%69SZ;~a%nWcs=x9VoMDf=nJ8;txf<|3}7~*0JNp#P< zy|oFbU6gP)E{G5Lte|(^_Qt}nu$c0;!?|G;Itun4`cmGUDseXh^G@?^n4}!?`M3p% ziY+PvmCd_g4phFlRZoDsO^ROkH%yge7$HsyL*%PEKSYoV_stPsU@C@a3Sxq z@<@EB>tso92Z>1NR^+`I&Lx(uVZ6ShOp{uRo_~K&TO+RWy#0X~Y@0hdT=cj~LSED; z^~$RU|gy}{fV5mO>^a* zz~uw=h^6BkXqB0BCN8&e3C&gWKF2DcFH3_>$NJmpPCQr{t0@&(qUXzM&2@I&aoy{E zr(Mm#(G~r^paRr;tLHDqnLb?==r}T))Pwj$adImN9Xv7m+@ULlb@(_6)eT*qUokH| z?SWmn($aFk)6(L6)4f+{t_cXT#HL{(nb!RbaBRs|CE>ciqe*qGE#J)y0n@-4EnpA% z%wGIp6Mv%1?RNqU{o{SvSRbYf?>f2PeN;jk+pN&p8J~AC=_B4ad=#PzoFP+#xl3_7 z1PyxS;aQV=rsaZH6z()FoSNrU{6O`7#UAaW&d!7V$j;r?-=)PFH|DE#KY+@f_erh% zjKcImxjJRL>CrV1N~Jx?H^5lpw2*bXE71`(-9ogk=s%K$yY6R_I+^3=#~06V*kgIK z-I39~AADO;ArM34V`g>++_TdCom|Ec5+;CD47Lhqa%rfV5rpd)E`&{nN0 zq&2gVTywU8D^xFvwO{a7=`?j~BS5GvH*Z^~)%ea=N*&3AeK9 z=NzQ@7_t6ACYHblox(lS+28ibg#Rw*-r8VfnC8>6X3~q^8GBf8gnk#?+t%Dsa^GCD z{qY8_h~ZGy9zJ!c2I}tGaR~(Od?kW)8TGoSx~ovavXGC=aUt+u~K0dDSc*E>3asIc&C5mqY1s@fKzG6W#H#B`S@dO|$(*P4{Otr*)~n zILf|FtZ{mP!6v}!U^<_HRganNj#ZR2nKGRio0W1k%6Oq8pf%%MCr?8}WR65pg&Pk} z?j3rMhA?uS;5k%euCIvbt6Qks0gh&&>z7>t zCm&fsA@icx2ejK0a|*RvYtKk_;C1Kt0tt{21>_^hK1HurXBCf4R+aQ$XQZ1`47xi$ zaSL{O2ctxEvB_1~D{*tThBc7jbf@8SddGft)s*TFkNGC?ArxoVUUPf`pvkewV8C*J zZb`<{e-yVR$TL^71a{-4X1B&GM@~o#Z2>oE_b8^iA3}&4h6E5XAI$a5wAsb`m_qVv zV{9BX3n+9KB%L5Vn!{Ua{3Q&SJw*#N6j--!zRzj&DQpyLO&zl%v2~i2Q&~MK4rypq zc|Rk!H3Vv-E85UfjaIfI5|>}+CYUO0wi9REGM*YNWv7c;<&8MCcb%Yw4ggJB z-_7a&NDeX!&vYFumwmO@fj$m&qg6vW?&>2zrE3&Tm-!kZ;xj)R?vloz=Z2>|Amp&gz@B1p^K=swsX$8osQzNb z5mIgc$ns|P`TfHdeV*3BA!M_E{~3B zdtiu;E7fRhbxxHM?vp3(WZfm$kl#Rl2w>uStMArfW3r{-k@fMh1);a|J$HsE94(K1 z5F)=+r3e?Lb`AiZ!JN#Rbgkm3VTVLcgVFP02yrj_w2p;tfEV05)N$H8+`g5)R5ko& zlaO+Sw89)o07aa#S_2)BIz0_$vOOybW=MZ+e7EwSV<(> z=%_(dL!BJ1B5uADxFwKgKi17wQmhudZP4eazs;2n9bxW3 z9Y$hryjfO$r0MHMCn#v!>ar9%z*1TLW)gkYr-(w)q#HI!QMMPO`1xVD$Dsw&2$^V_ofbH z_kNV+@{;MP@7snF!OrD7W1+(`B_D6dbNL*(>JcQ>7u1;45i}Nq*$v(+5tePz|0p_b z)djOmWYvFdvpBV{mH)D8JM~T>WE!e~v{6>N)nAsPuG!a9p97o{&8|AOWUKo`uH#TN z`ii2MjhjTfUk*S3KL<8HW-})bCzkJ%WdJgF%ZiWKj-Z`{t8xmvs<)#~E{BPtqB|#}b}V4`N4k*xSd2Of_Pkipbt;?OlSbbk|d8VZmfktnZ}SiIk6tmIBN? z2N7w~>%||MU8UUOHL8^`YIDWaF2cYh1@VFiVfggSSc%$RRTBXmJcp01fuQItg>qot zlCIa8IWdp12Fy&d8{i!s0@2=JjrsH$^ISr4(JhYrAX3&5W#Y{x*m3YKxm~O(LCq<1 zdP1sM0?S+r6Q5NYWBp{^TmKMo{e5Mx>yqG0m`*OAYdal`{*=8dn}M*0nl~Qa55`Cx zoNAeNVj8V~{W7D{Mbzz8&84IT$ttHNj@VlNz*nJO^sj3<+KOjncSj){opUtB3MO=M zc|4?$m6#Q%8GA5RVMGR9by8TtIX)KR#mzPPZdk2t#1x_-y{;uv~7hE)i# z3M=D7E@LMZZr(63et+n^RT1u zA^@GsgfB_fBIuXj5LTZ%rRco%TKLJ9^gA{>-G}qrQA|S!Pa(O(_UFokuQ-x;4+Cup zi{;=k6w~<%$hORIy@-VRv4tV)L$B9BYTiWTCj!PAj=b~?*G}vgSHo-XTcQ>(L$t;c z!rF<^H0mcmbwM%2g_ei!%X_c;HRR>{_CV$;AIMFr0nwDxRI8fY^WLXLUo?8pj{>Q` zF9JrUrR?o_@dm{?zIw6}gPJ@{ab9>d#HE!f-7?r+0GS}@NNRJc=?Uy7IEO0imBrqAUNF@55MUI~hO>INzhZCXOGmYtsCt@Or789SihNeX zAG>TI8=D(DL%k@oUoG+IdRa959NCfbN4 z5ALV7fcG&}3~ZLi%NJ+pD$Q;W0%y;G=M48q#GP z9OYtX9h5qizLlv53g9y`&ktG|Z0KW{n_Ak+)bjz5F4FkjxrIJ%rxA6h<&FpFv;&Vq^-1+02_Jhc8qYm0(L2OG>CUf z44?|E1;iO%4SDC#Kee?8~*?J+Qi(tLt#d$v?<4dk)75Ws7lM-6Gf%nDKfkJOQ!w0&?aE|TORoH zc_kdA0;W+t5r6b{vMCNOhIqR)Vc_r)icPK~znIErJegM2csdB-HvoEhdNeg^&*db#d>eBTi}1%Gfa!~v#!JXMnfgw zO979~x_gGT5ch@L89HBr3Ah>J$X`?Q-s~BOF?|Za$ zzu>6Q4GYwvXH2|S4<~}MGL$=0PL!tGDYA(6by#Fvle^RURIyKX3iR8R7`lJx0^QhV zb{TP7)@0gdH?}G6p@Eu{5O7g$$n;EOTvF(%cVrp^2=)-^e(5ekuu!{;m4O}FQQy}> zQWc<(8R3p<#b*WN9#>?<@u7fThqVh1h4gUM%92k9kU(h_u$iteRwsa#F*KuyJUX91 z4{Ygh*XdrV6k&~>wmVh^mbB4WUZ5D<^W2pzEJ-_iSTp}*(cu{#5P1b$T{O_kJCTSS z(zF~3(ez;;Z za)qVLp@axEV4`fepq)DN0xu%NktNpDJ5RGJv2+nrKg&D!tdu1Y`O&oj7(Gv~z-kfD zA)KGgaio2k?wz0RmJ0?U%qk>*FqdMXq%C$vJ6pve}g39?$cWR$#gjxk`MT59lho85Z(BSzxM$SyN+b z;ThgMjzT|4mso!*7Ek&c-F5JJ3GP8^$N|*9__76x>H{s|XoJ|Xv~kG8p7G)U6~^U8yCk1g*zdeAcSG*uN}<0&?Zlk>94&=JbB zG>=1Ct|;(uy3ui9Q9)EexgzPD@WPL?}qfX5jpRsM`k?89R!Nbh)SVApv&F z!vopct#1i%38VN!&$5E5I4mgO8Z%VYl>FS%ztn z{el|IcX<{{dlf6^^H2#1Z>(owTP~ z%;g916ZRo1W>M&n6lz}I9IE+#?R{N6VhQwv$k_tD_`;KWev@CFSD*fwBK?P2u-oE1 z!i*Eb8C9@tUG24*2IW7GP8d1A9im-8HXDcd1h1Ad z6j`p0wdPZ$=$26$l#scAn)1~V$s-*J{%$8&5i^~E>Vo=_S9h3H16{%?BK&yzH;VK@ z4%Y{cqlbb_fCE7DK@FRL=ya>1ux9x}ZLhuzf71l0P7IqFWJ84z5)c_$0;3_0f>-mg z0Ru(w3cfXv3#OYYRd`xz3y!UxGGWvEf^LWrJPAKUJ{&~hxEWrUZnhc*Kr;@h@Z2ih zWXU{)9bh8l=*v{f^|JU}kCi%8y_0|QS#mMwkNZbhOo=Pe{$@Bp|C|eN(Y`%bO0mSg z)GChViqdi*o!Qi8(=3-V%LVH{EcBp48{w#iQLM zH(dtZnb)m7d$0Uv2vAxJ{}>AR&ar*Dup3d+qtsa}=8lM*v5dwtEQuzEY}e#mNqq??Ux~b1tYi(}M%vB5cqa=KO$_C@v(<2uxOWG1}V@HH4ivxjO&J zauITA=pYZpY+t_r4LDHo96YVC;7$j+2m0)cI)kVGB|5yj4(RB)XVNUgI6v$zEepp{ zk?z72G7grnT$re=;^+Uyp!)j08-Wukq$3WY8MaN96$>E;hPjG-phnVp-W^`xl_pmO zI18EP`6OUxU%k-@ksPI)*b5llJE$DKmZ8GHTkLXpPRukBDN@xDx3i%~#>L7XEk@h| zkjOV(l-9iy4l{cqy>58kFKYrTB=jL(&Gft_A_}MiWT&OlYAa?G+u5~Vq<`cqwk`EM z=oDYGSY2QO>4MqLfr?Zu6lAH{0ZiQGBrrv8-XUlX->W@0imAKn)~>UI!LvelB>N#k z{OuoxaQt7gi2GeGCJN-m_I1DB+`zIUx~UU7?hEgxs|+wfO}l}gJ;^_n(P8R*jxAj( z{?r<%ly`1w#EquWW(^e7(uv`-F&!rgL_ zV|f&Y96-b02WR7mu{sO?y1oN@WEplO{Qn&sjZ|@8Cd6OKSzIw!*m`?D2`6GnUMxOHCFTyW%WHDMo8Id8#v5PWn;K)t-8KKA?`S!7teyX4A%RHdVtE;5w2)rx+07s@8dh z3Klgiy%t105ab`IHa2DJ<~z=mkVI{wp6p84XZo%j`u|b)=5aB$?f>`?LWrAGQiCLf zP)Si^t0W}Uv}uqwqed$2b0xGVN=0Z=5?ThWT4rdaeIJ#YrbWAErm1GV`kmwEzVSTI z=W{>z^L$>f-}jHXYOayFuJb(3<2>H)_i-GT2{p>fx~Cei1naP+_1$BFSg4k?l`?{9 zX^FmUyKF2W3wJZ?Oq;_tK%4BnjJ@7U##&tcq*|bNu%p(<#1+%Fp;GpB*Ttqi?h$9p z+}oAkA{RLLdA7Tv!}h$-dgD5KAne8EJOX#Lw{Hp( zP&ddE8edWL*@3;AUXMI8518DNNGfV7lzcq2n#*5UNVx7|i?O79M@&%1{V zK3r0sOcD!d61frP)5bi0pQpkuFzrzIoGcYIX=WLdLD3OJxkh&gQudIEkspZ(sp4g@ zauZ+r$uS7pTpsk~<`-;Z;QYROQkpHj)iLQG1}_^1PxxGj-n=u@dt<7c`*WY2bj1=a zIfIUSwWX)BZO$c=jR;Gys>Z>xN29%tg}QN zJ6(R-ZHL-cd+G%rCrV-8XGZZFh-$P8dyy8JfvvJSs~kFNI&u$EYCYm( zPNB}d??3s0qcp7{evK?v;pZw5Hz@M_lX%z88HD>Yi6NE|reXJLiPZ(-i?+_SFGB`e z?{IT&=?_X9-vnIPnE@%_!M?Y%{D1Z*`aLFloPRf69e%eA8XQU;Y3GPd70nj9&jV~ z;euEZaAXAKC5kCu%mupVkvq?I)KzT@=F19y`2EeJl_Pkd)H&qT*R4>m#*w^tFRzc}5|IS7yjS31#*njoCms5>69VI89c{g5_%7;U4Z|-k?sG0$YUvR*u^32eYOi2E+;T}jwqHjT0f5CqsS{l z(TVQ`l9Djv6yI>->lozUGHO2MqH90vZyiNK*l)nsbmj_tiLl=!OMSExTqdT3JjDQy zf!Il@L#`zES>%9>1u*yZa-=}P19D#|+a2ygv7VY&0ny-25!l;aY2;mp58a(Xj10gX z@=ahEcumZDAYzDA&mbI-oOSRzU?q@XQwYhEttb4#I)QI4>=(^aACnh;VJT*69DLyp zbRBX5umo^TMDV5czmddL!B-Tb-f`ax4246be+ni+W3TpvYxs_poT74PqaEYe4@G(DsDqP7O9aCy&xVD=(sg-B;)6JqeShz`dY}}?S+0iU z+07u5Ru-XtzC}++_^F%tw@(lL{HY5lcIk5wr=x1_Jb>PZK)sm9UsR2B@P7E?LWkgj zT2AhZV`sweBq6TCNFVY#Gy%Lz4vMGc>&w^!>k~s!qszeXBwCU7vG>d%c0fGfUNR{q zb{OW)60X4{3Zz|gBzmASWIgKzxL4I$+TXJKZ>y!{K@X0zauUa7#m*r1sXdOXVW{&% zy@&!YFpUYTDv9*|!5aG+#BbWqJ_^FfpDzO|MhS=*14<3@RV=?2IW;hY$Q~K;n#%Kr zb?%bK#s!-Q5@i_XZsH8W33TP(0cM{R#@Q!a3f&PmAm&csRVhi(po7&LD;;_)tZP8v zvT%aMO05%^n11A!)D}WMaGK+W4PR%^s-gGq&EmJe%5eWWLD9fM*I@**i(RENpr6lN zHh^`dFleqUcpR2-7?5-YhD3W1)~^0NtX2B?sxjV_w=)Q^L@x-Js4iHdjU>e&RM*S+ z*B~F*uz_tTpvVk~3-Tu0g_fxJ>~Aj3AC|Q^gpI zjSE~nJ%iw`$R6YntFMd+>tivGdH6jSQQZ;4UoYRQBIZB3jXa`1nduw@O$u9-B-&=w zsg?VHT_h?Dj2W?D9BdGt_XyKz%IscFQa*tfqYM{SG40yHRpVpuTZ9+?uUAK!aAa^+5Qefae@aC;l2a%bFn6S3!nw__*1z8ag^ z09|Mfd)|VJ?3>=vXG^-3ShN>4r+>wzVVxRIE@!m_J69$3$@+xCNPgYch8ct%Qq*DY z8ApXktI@^_c;k*C{Ne_d$5hA11EQ&G6geiH5|gHvh$7eFxl$Dk@&ihrcUPksFDk9l zLiU)R1JBf3@$hEwpq`GZJ#cVjoxGghJkHu?$Ayurux|mpBKl)R&m5j<2O)X4R*Gp<4VKK z@lyMw8*=ITJ<7{QKylJu%dK0D+dCgDSgLr|a#DMeHC{}*Ca8z(MsyD7G?s%T;n=wdf-=-@GiE-Y&Ov0C*fD{6<@O5pCl|xi~XXqKJcMj zw=t$&)NB=|#P^7rW9HF&nA)PvBv(PYTN2LkiD`M;L?})!dXm$g6xnC|oYq?JDlmOWs@Cl<2RGLY+`* zPEEN_gW&nhAlfBDX^9YR^|y%KgfjME2C)+ofY5AV z3B3a<+mj4Y(lA^Ev33M{PhB>I6eu8x{h+=_zQpTmFrV`6p40C9H}CMzVN)AGN})b4 z1JN^i>Imyh%K%EU7J3bh<-#AVtQ_DWRP5)7Q(#l^4yEpU4u;(PZws1Gdu$Qthci6Scf}RvML8wZ)KD=fn%d--qgeGWISrQJ8 z`x+AhAN51<@9~Dh4;}GB?k@qiuL=%9OlqJVym1T#w?l*F84(-a5Ca`&5G(NCV|~4R zlEnge+>o!tMCp}5g{LQ>d*8^iGkP2U3;Nd|V!?OfZC1ID8L-JK9WaRk0wFK1;uMHp zB9&=Wo(Bk4?_&orp`!4n4jh+<5zHDHB(@d6hslA^ST6u(Hfc;EN2YlOAv&@=HVD;c zU?>dHUu*3IsftCYEntQF?fo_Q9l+h7GPen1p$=$p0p<|gh+^vSb-o5o+*1-hBlL$e zh_XE{GYAYAngj)(l~@o?dqCGG?{EnKcfOBF5db^vIDhq<`DAGZ{zn4#zo4!~FHkfX zOTnecK7%??0;%l!FBvX9@*w{F_P#03W;z6)I-zyy=a5*4{YJs&wVrZP zi2cai4XdVCaUaIQIxJuDT950nEgIY9et|qD#Npl0&|yvo4Fs;S-%tTI%F?hHjNOW{Z!oLfi5T=2%5P$!Oez3qH*EgK_-9P` z{hx?{1Eh6Gc2f3tlwk1lka%KBLF5>MIsrm|h`R&Zg1Vs0*^nPAxCuN&!*+_c7_0}z zT!h~ij{|5*1GKKf2j(^CNN^W$whOuGYW{KI66>ks%{c>}6CO*9AOV}G#=swe06RG) zUOEnNI#}E z=|BFfe%P!_K@nb7gNBO~K*YQl(`nYV9>-2`!*D0R;*Mwxt*Rle5nk_W^oA!5Rnx#a zfOuAr+_2;VaF&29n7{ zh2c55&-#z>yvTY(;R70qABILv4pX_S(5YXDRaYXWjfd+xGjY=?gWuQ9l%drBi7i|` zb`Jj@!#tU-%$s8XAm!uqqLyGGkUj!{^fG1mRzMyY75PQVse&aTWaj2_=q5gek_Cd_ z+jwDf9E1cNRgB|EeSt>66W3LVB&BW~K;+CFLhk{3Vme@!mm?e(=z(M*qm!ZRj6oJkFUIxhkvh7WVT>;Nep z$$$du<-7-bD|ezB#EuE#0bj^{MI(eMNiwjHWa6O@M9l(V;gcCc25AN{gE$aVBrX*1 z9MOCn3w61|QIEe&p!dz=O-5}0Kbkpbst{-OjDE*iXYj*)+F)D!P-X^EKpVK=xWZ3{ zyjDrRGryt3P*5ca?yR4qGPY`CBG~{XA$0x!>a~~Jt5_UFk{|4QTQx6bY4w$twfo>z z9I;ZoSgX&bD~yY~cD{XadHp#*ENf+<%v4(5E$aT4Cqq9UsaAIq`fWcqnh$r+7_9)*pa-gQGxf1m76JTND=tAitfc*Gcu&@NN;9^ zT7RK@ASPMbbA6w_p2cj`SzE-Vbf6X%?%S*1eP$brG-Rbrh15_uO5DFOQAD#k_*6PK{IKZL}{UlQKX}_y_@usb6 zsi#VsqRn4tnT)nZZ)%N8Id-P#Hc6H7l+Yb6XVEcad|CVOhj-SPwIf^*9%yTNym3sA zIXI8Mc~Il*kR;Ns+cRt=QQ8ze8}IRRXXx2 zQoTuyBYWgthH9DjX7zw2;%GjV(v81=JRW(24!5Q!heS3&8>u&;g)H!Qnwom6w!}6f zY#t!XZaJUnUx6NGTW(z~Aw{Dz`opg3eP%yAW_iZUBwgvlO&>Sjfnnty@nFN`@{ahf z6A50nq@rd;-qxF?Z56RHs6G{p4BH1Pc4U0k5U-7@HBhfd4}5Y~H@StJC7EuP%mm-} zYU-2x+~es!7jjbO!FCk6Wlc@T=v|LyZ~L_GK6gywmC?JU0V~2>brzx)Mw0cC=N^A} zHeU(3;H1-&PgE1&2rk+Ry}|DKuiiZk?1lWAUs(Qo)OEfFsxny>8 z4B%I_*L-jWp~j6WQ{?u6F-Kq$0@eL{-FIxcvpaHE-%ggAf=At``?JnrY9{8|w(*P$ zg#|HhRii4e)Vj;*6K((b;{Ztq1VpGW*?~MYA4G?G$1qluU!^)7+H!o0qx`mPx3$iE@#BRz`9BhcM>C5JTm@*E!jinhndVECWpVGJL41(5`C9v z&?}*Wvp>i)PQq36vpn;%P#wftUqC!!^$N%{xzLaDjAQ`V2F^4B4zBg9Kw}KvephI~ z(&j9q71;p#{%`}x1WLR8S3x{clMSHKb^;HC_^;84lt(^bue)a`0L*`zQdRW~qEnH{ zZy2^Aj9=!!M|ooZ0LrTQU^|V((onoP=Fs5Xj5xAR2{cyKXUuB^#gZWvPX|};I)t|%St%}M^q4zO_uUP1U zmkrpEyzr*N$^8c;QA=2e>N}GM7{{j|#`ujCDtH0`B_~uEK(d~0<}}gt_@TMTNtRGn zcxOvys;7YeUvJn41+K%KFgXFBbZ|SkM>}D9$X^mw(HaafFoFl+HV~wjWq=_J!Y~zw zk-zILS9-{u!(OQO!^L2#PEBI&50hU@@}tfez{S9%2Jc}%VN1wA1A{zD%LOu1IcX5@ zQi^1%5n({ODYa{R{DYEGwDTY2mfC9aCoFJwm?*7FDX{KthykGZ5xrlArQk>pyaIY( z;0%E0F(L508ck=5Pn&unQ>|wZU3S7GEcb`A{$ewG(Vqr@Xe_u5% zecn2QxCBjk@x*%E0e`U)kpAP~n~HqA>4ETh>J-3!cEp>tcpD32LoGbL5;Lx1c=sfJC7(>r}Gpn^a1sYZP4s1EZZz{F!Hn$T#lL8 z4RsJ;3Nd-)bHv9g*iM`WfNA?AtH}GI7O<>J%+xe|w=SrM$XSgRYJ{wNUM$jgzrnycE=Fl}P+5tC$4i60EHT-8dHIa23*F@vta-Q6)!;VVsQ7XgD~ zU1uZ+T^#Wm9+*lQFcqux8N}mc-$~4D{zmXmLweyN3m}>wVaJ1L!mC)wcsZgD(@@iw za8FQEy)A=@A=Q%y{%N0*AA6Sk`TG*RZUXpeo z@Qc*gGq6rj&%khw$esPG!-<^2__rJwPNP_eOlaH4J?*C3B~h zKM?F20Hi<17S8m)oH29_A$ziQkTw6bpiL^IZRccuvTGzOJY-lp1W&n+FIp`}Oe0KABa_Q_RPZVtaxg^bJRNp=#V$A+;aI(Kw>@Pcau_>u@n0t<{!dAZiO;b4Eh&Mj zq3K7shMODP7Db{GR?M29juChl>@lzFuCKRMf_kt|lpIo0X_U)r4*OmsmSDm_y&~$C z_Fin=3lE}M+~oA~)n~^}i{^-CJK-rvsYy&2Rc5cw^I^$j&W?`mTeI(NIcysJ-kcGF zkPclP6*aEoOgyddV{Dn_||A(z^$>Mpza!B!3yjti?6epl;R0 z;EPRbToymcyHICchQQl4V2HU?x-MqEAky$P7HqKg9nA5&)@n3pDXRMxknmNr<3>u_ zZ#{5#l-k;VY{$g>sUAP=<#Z)={)Sb$(y@ZIo~J;{U^3(nLUR2=g<7b`LL4)#18JKF zj*bAH`=-_ICP@o=KeI%(^2?iB^L8wfJ3eYHA(+RP1ZH=Prp%iMB8r3%#F%h<{!d;? zGyp}B{J6b@sa099EvbhX2z9@NnD$KQ*Bi4|7$Bd_qVneG z0NU;;@h-@fkR+e>%@qZ_a|3iUj&Rne|3R)-~(|lI~QOeA`dIX zK(TR9zF|Knsi_L;qD}DcPbvSJa(15$;?VV!VrXi9763Av&Zyx-1K>GYl6`n(aEBhx zVa2cAn-tfee+4+s#3e#p8aD)Xeeh2Y)mUwF0}&XvE3p1Dcql|TaKt@(lW61okd?@9 zAji1Y2ZdrB{s;kbPpYN9k5Wm4I)}K*?*&Q{AP21lwtn^$loOrN+XxbmR4)iK4#8Vt z;t`PdglSoTghl+-k5;>v0R#I35+y-DBO+(^b8xkj-w~1fzd}UFrC=6+fkZ<3=$}9$ z%tGi#NEH4HB&wic_@R^^AyFV3Wrlold72-55u(%yauglDT1Mu}|HSCO8wh$2zymsB zKw-5LW?tnliW?xUL9(BK)i6eY!_523pn_%=$Va;=%tsq0tF->2n+P#2VA_wlDf7gt z0H!@hbSA#|0QF(?>iHc&n8RZDASyv;`OyDCRr0d=2f+B~K9mQHbPjogSDPU}446o7 zAbdCCA!!8|-yZ;yTNz&otWd9BnDhWFu_L3eu{1jJ%S#r@tO6>mYyLVk*uw7rhI=5e z!M(2ouisPjtcX7)C(-n0!1Wzc)!Z9EGUW+u%Aj{Rh%OQck;3R= zb=@zr(Ky&{%E)7}a2@8n&o6@0Ez)mLH5zV%Mknb&-wk^SM+yv^^6X%pdjy6aOKas< z)0eybS8e})S}|$kzX3%00to74?yOXf8lh(=M3eiDzKo$jWnHWR`_81^!T(V^`U1@M zyLKe`FSH{9SPKltp3RA%n2vbM?t{gq@qx9c5djI8~-wP)z`66a&ek>r(4w}DFi~d<9viYVGEl!2rr%J;TT!NJlK`~lJS5m6)%XfS@z zgbGIE_+;gFBAYsh?+c;XmpXO| z9Sq*B5RT#0EjmA5iox>A(g-F#nGg}T#+VNoaWoxYKv&g}$)3nD4VrVRhwAV-b*-RY zO4J!X+Rq~`no0DEJBb0jiP%FQ^>A*^r8gw$F6DvT9xr7-H#T37W+9eW3%Nd{o|uPC zB+d~;`<$90F8+{M;1sjc!E<;P(VGa#S8S=+mJ~GHNG6^ttD)mWZlxYf^zw3Dxba2Y z)8r*l=hQk&ZiFaoF<9*?tJHD#?y0+vBKO`a`|vEiO`)@I*Y>?l?_ZTCDctLMqsuO! z9sayS)_hI(Sh=`4$Dk*rtVXXQznOH1X+G4jV_A9FX+pXu@7RXX`%(l;N>5vC5Wcb? z<=T^EiIzzg%J?l^FDo%6^jR}Or>U7&QASv85(nj-2yOi_VJHZQ<+oJ8&UNtNVd0Xq8t zeYnsB=-67MU>?~1dmXC`5y9cF|c>di5DzPTN7-M z;*T2LBtDTKKBQMdX3lz9$7f*r&D9>!sGM1*U`x9usHHC~DfPP4x+*%&J;r4* zaZ%`P9Xt--iJ>2TB#*8iFSMbTe^4FMZCUhsXsM#4C`tUHVEfbjEkTu)(xzvh588`x z&XQgcjuz!*=POk~jTqg^j>MemRiakBlXoU8dy8a{UIat^xxhB$~j*6cWml~Ue%VY$W8YrW3( z*D+PP51t)MroQuZ>QKB^8q{?I&BF0kdsKV$PT}a=630pXS%al*@U-fzF;ye09lR6x zrYg zE#wd@ko1$#HCQ~--~z)^ozn17bw|=!{BC0Tc)QqC0RzN6b9?m1Cw;B31_vstvNG=^ zM&K^_dG-|Q^f^dnzPh?M#yZkoWQqoLr&AX#9*8kJ1ad{pVu;7-;`|lv-4VAGLRx&# zi=RGT`!@HElpUiCLai8l>akv2Cd;Wv`uKdly$Cm^!EFpxYVIT1?$+#Xk_oDmcZsgc zQsOiSKs0j+CkxkEZOI|OZgEmutbTd9S=ZsljIKa?#al|#5P<$X+IYJ>tv@zIb-~Nm z;h7e1WO|*$7(;Og78hvWzvUr=39|gfbbO1`UUZL``fGbR3)vl00wE({!CxED z`wl-VUWvRSuS2W5q--|JMmrpH+iE8}v#)B|!0jz9U zhNDHSG^c(#yl>g<6W0e_tSZxFZts_C;lFmsZ+)Ctt?O4Bk-RH&l0Qen} zZZw8|=T#GXU3)D??MvvM`3a)_ZYx>@`PP`m;@vo`@OmrW2bsWi-O6#dkP zClOPXjF>iQyeZ;cg^VV;l7J<^-uhr)T!nxm`C`0NK0Ddb86`<<_tGhDOwy+I+vQ92 ziAP(qL$#KaW2<;UV~&X@$Jd7;5a@10{&Gr1zFi45VwDq_NEO5$&2(gDMV@utJz|c` z0luM!8Ub%>R;|&5kEs;r#EQ!nUQM%hFgEqm7rkvSGAZ7p$X`M1CSMz0|0=N5`J#Hm z4m0a;i-o=_Av9GgTM1p<@!(PH&h$bv9w<>DmTIi`d}4Q}t@D<_jt{8tU4Ez}>J2DI zv6v91xf^IED1~f~2ps=y= z6<})YMyb359@nRE>~L(4@524pgEAhUBkbJ5f9p*d8yKDDdHblbhY~vuyJ3ArPut8& zn_Lhr{eaYEFEZq{d5?F{m&f{#-69obqMQ4c4Q~cK978fdmAum7wduuHBdhkq+Z!t1 zNMMgPIcE84ca_&F__ub>an>!8(A#hFEMaz`p@kq*d#Q)JNq&=?xRiNESNr?(F&j)O zU+$Jq_O^jksaes@BaGe(`PwLCJ4K2Ufz(J42ZSx#3Fl8>?Ey>)bt^3Kk*@*@U1u?% zQv^dgA!uz&Mt4L_5TN40I?fV$R^ia9j(^WaWvl!pL1aFq_%E@WNoab%}ao`w*Aqm03|D)#BGV z)JWcda-ep|lmnV9FjTRS)NJ>%A<5RlK*oK?HR&O;=p2e)q-o@F{RHt(TH^4+8v0Xi zx58v)n_htoY+XS-!quZGj%m?5di`I~-Gdz`b&V>oglWCRa#GXl7I;uC>ATV$pAE)6 zswKFH1WSI1j@Fk;kkdgxMt9LbWTP9RIvSND+AXb?GEN7Jho;L5K3!j!shhA!fc+w5 ztO~29{Ma)H8ljE));(SU^kujiyda>#bb%dCM>HMbH7`Bg%zl@O73Na~)^TC1`yD^T zLJqltKsSo;*A|VmaM)|MxP|N7ryB-YqI;=hJ9w8ocvC?Ph0&Aq4uN-XqJzeL+T*6tfqR|Ze=?+CH>{fucJZ)p9C=|hdOb56G43IWap+*^@eK_ z{i(O5FCd)Oc-i`Cy%5H)dE_9fR3Va1Fy!Gv)CzpJQF)EK4Er>QfVai4H*R>u(;jkR zV~SmLe0$BBGx|#^u(#&lzPgvvs~7=q^&s}xX1hEe;kI=4HD#fj$htOpv~G6oF1J<2~v>LO;DqLz`?*^|Y5WKR?uA)vjutJM3(AmP|LP3qU$9i!qu9wf2*X$-ixgGYDR#f5G8` zlY!7)?(ai3C>CQnGHIgaVk3^~x{O1j=zC+l^&z01B#tT0-XgMv@LK8h?Mk-?N0(mF zjN>18*_^>Dhz8XwagA;oW@m3FiI<|AVGTjnBr*i^s*QcFC(yg|D#h5}+oIadU%T4y zRI8u%iT__vt@o2u+m-6H)2`ar>q_2?i<8m@aVUg7399NfQ2L) z1JvI|5~O=Tf`PAv0F1oc|E&<978HZ(2tdiwdGN<`T`@U0{s%zFa)6IhdO#oNhjPJV zeE>8}SjZnH^TQc(cYq>i#G7{p5IK9*)5xsntIUKq_slv|*CFeK$lp9#c{8x^In3|0 z>OP`2JM#xt^*$&K`-xUn{|mHg_%~X0Yv_N0R($|SYd+{I`;AxC$Z&a!WN!SM*(eAy+_sdfezyD9i4%rK=omS^O_8%MvfHmQ|&H?+!MBsM$nM7UHi5j85$G@Eo%H>l zcjEY8%R50X1i}pFk9?ehEfi^bF08|g;Btj^zORS@Z=r&N^#VItcmieACR`$gw7)wd z!qMEW{o|-m(%_$|;`IP-@k7+X(ybF#LwiwwDO#wQ3+T5XfYzRBRPciaU^T8>FCdc# z75l!XovZv(zN^VL0I@NKXCo}%{Yd_)?zfk4gE|4Ixf~2n%SWg|$ZtVCA)jKw(qNiw z75PR2@q;zePa6)i0&T;Z>=Q))<}s7618f(54%DRAz}L+qnm|G1K1{PPjZK)i@&k3M z|36cwKcw9+{}=co^y`i! z-}awfz0fW-B7UVt0a=PQQqff3L(^n_MLJU#LC=ij!mH(#mVKZxCo_%{(hU;LhB@Z?R9 zX$q-c)%!C+elwt{phRPyGhbFusMWQ-n+4%?b+kJ zq#>AcqqwD5c$e9QWNn*&M%IDZJa z;nnRS{MtOE&FRXc)G9TqN9fc8dRNG5okh6=xsfGt0({`|eyLIMoOGl82D>YF2;RiK>pkgwqj3oS=#+D%Oj5dgy}rutM=0o`bQi^(``Mf?s!mEsS%9K&=_2YLxb*4>Uypg&=_qfuavsvJ>ffb5J z0+yY+5rvMT^yl_@-1AwY6Wx0A$j5lkvXl#1ASx;$sAhO7X4uyyV<@gz;*#N}q3Ksy zr%FGzm#n_}LV~@5w)x8CxJd`-x^1PD#7-=5jNLLzJDO=iIH&#G zTjEigx}PNCWI+-S4^~(GI{j%3M>5(3QV~dFJ7LGktHlX%O+s4hL>gL|qyroCsiCQx??=48Rj)tytNON7$ozhPN&Jw= z17cKRCJ(+6-HYL`G=L2@JOWM$2$6%Tvp5v1P@X3A!IiEjxVaW~8!#RuXQuFD4d*}~ zPa+5QAsZtCp}x65EVo2+OrmW;l)4WHGG$P-!`HMd(E$obbKmceYvbv)tX%4p*pd|f zq+-v)IKCyZ17dHnN}58p5svSh<&f=BZzpJq!UO}2X z_KGoShV{u7oU)wnhE}`4<6*g#FTeDL2!QhD|H|idEN%R)f$uKqxLuUw$eQTm3uN{a zOD-amM#Mj+PKi;-7eiB8oKm9n=;4JV%h=(B%-sX_qB(mg)$uFne9~ry9N|gv)hhRm zVavxc4Hcu~G{PeOR`x+!y8JP3tRZV-yl>hp&#Q||VZe|)dntTI z+Dql}zFw38%vd&W;V|_zJ{jlj=(+6`4f}Li8MOK4xwQl5L>kFnWZL@18N>iA&KpHn zlj$dj;`W54WIJE0O&M$Mrx|DTO3(RXthG^uzJyc_zQP7>M59FUs_piYu{%E`VZX!} zb?n%%yWf@vmy=X7qG&mJT{4WRYcC$RPsQG;7O$@tofLh@)R16S@K+$G5|Ej(?>M9~ z^=Km2Kp{OWh`QHlp8LXS%=P#+xD`H1^zv6qHW{PV_wVsS_!q$U5q*y4G0Ix46phvOq7Be%BL6pIH|ml$4?@qj z#|mk=Mhn(9MoUnv5eIjS`lhy!HbU=^{KY-T*5n3e2eT{&Puw*0A8Tf8@zWdgUf9buXVW(I5 zp=P1FQ$Q+=rxl~P3?Oyu~c z#6zDol%JeUl5f(&h6oTM=(+E_>;kjA z52rhhtvUR$JOOtWvBlS3gjoZ1hsAwpFn8(k40xQoj#_o>4Xeh-kinZe&!`*i1#@J_ z&ON%T;XO4kPq!;l+4@d7mTO3iM>$3Ko9mTh{P1fHcy5ssN3tWQ8s5s>2uTqEMJ@bs zlZZf{{l2s_S1L@K6tP&|CKLM&Rxe5(zd-L|ntjrrQ+d+#W9p{Si<$^H0=gg`w#>(t zElsPj{3P>8SIM-e=*pF&1)eqD#(01=-bwN=*s|?<(&y3I^(9Z3t^F`qpn*)sN7ak= z0;Z#PYICzT(j&jONAhOYXF^HbbO1`|INW& z^J+3yw*OJFR9}ekqWSW&Bg~O0K;!uv-QQ5iFf9ITnkXIVxZzmc`WsV65d75Vqqq#! zj=YR?TW{~62UiXU&NY0N5n6!~?jyL0$?_+FXmt7ahq%OhIO_H=KKUP34Q~LXQnjq9U~x@xoFkP_X&&1ZB%%{I77boOVr9-+<8~QJj)gNcl*K~cL%yC^k(4_bCX7W|2JDRjxb9%Hc$v(|}S4`R)t;?p2eaN#T`DM9* zSr%i*8T; z#WH4(Pm#u#Ylm`h`?5GP*f*moH!I<-Op6{NLU03Fh ztF4)}vGf80$kkN+rY9|vO4yAaF+KZN4KzPQvov_CJd{eH1z>%G-a@AB*+uDNWp}Sw zk-Asudq?e)+W~gyIZ^C$xW{7BP8Z?^Y@4puhjS{dvgw$y67bPd>S4L0YmyfWWUUS0fTBax^M&ZC(gs&l5E;SON&fVyqdF0#g}#j1=?|Dw zO*huSs3=djlP|*VOfJmbDJyf**>3A|#VV@&+{GcBM8+InM#=832ciXv2 zQ{wD1HjF*JQ^6|}Pq#T_>bY;u?K~^h7T4p0A4^)yo5*jET(89d=8N4<+;l@g@5d*E|GIR+pXqVginOt z*%a=0deNK^=wpc6a^=i^r<~Gqy*0UB{DK7Wj}h$kxUo&ButFs_`ha;+B+ukkr}HId z&XEp31SWyl!0)ytg9Y3jr>zl`Y*5xhscg%DDKZ}Rl5O~ zyeia*nJsNrDc2`=m&#Tkj}BlNQN619t*>Tjx(D7-KHs%7^017Ia~IiVUGSQ4>+qGk zo6i~RBINu|i1wn(l-cssFun?H!jq<@&)4xueUomp0NID_R$lIbJxRwGc_$4|5$&Ft zRo2&tg-M*fjO9iXD}l@idKpkt9Jt2_;9*2vd(gk>*<86DCTfc-+?OC;85dUvYd;3e zgz0^a~`jn8fG9r65(SI4W!InuYhl0JD&a1BGkXt zw|zd$=j{lqeBbR#({MnRhd&|q>< zsee293rhg}; z{WHm&+Y|UB$&C3IB=cS&QTGeUEG#qpm1I`ugT?s+$$aovl9>vqmVZGqkN-(BgSpPk z`?u#>s~1$duK1pf+m!)TP??*kzqf^{0=$U(^Fd(JDPnvtVH=p=8}I$l8uFjDfhqjG z4NT4M|5h6qG7ZNMo&J`xY;9==;zf?+7u3iQnKdV)m?Sd`9uq$}S?JEI7U+S6T2Q;} zudms!D|@>Sa4ze@|H#12{W$|S^oKDS`K6w7wIu)6T3D3(kn>jvWDPc87XBcRsXr6Q z&`sn7NJIVy0$J)$0vVXFeDnY6hko%^lY|~+j^rfjTVIx|-xy>7fPQC?U#Idlyi|^0 zN+VuqLO)eAf@12Q8n?IylKadI;4WeJB)ShE9_4|BOpU`)O9AmRSaGsgIKS%22Y!fLVn}eoF(UX4A!mb|tm#TU};wkNEUM#WeC% z6DuCjBUj744uS?YTE|0dB<_s;8@}WZk0Yv1KL{URR|0AYAy7*=742g|NBwl->g76K zI)*|1fZS1z8P6h*e1bHDY?76al+bO^%fZ&DKm}Q<-VxGzVgAaE5lMn9P z@jD+3)r}iUI&pnnmg@;k$~#8J z${eMQaf!srCVUW2Zh7g+lF!a=XENffQmRkiwR<+GXQTruBkK>f?jpBKwziTd+V@NG zk1Gw47s8eKhpy5)F;<^c(?q0l!WME$^qSl>edp8j#1k%SU?hmD_NoMi=9y^RtT6a?+lE(cOIJV6WlLy}mT91GU2!Ud5YA9g5m@ zyk|FYgCJgw%GQO~W!!DN?zqFT(azr+-BGA3B2p@K$z-;X>x&GNq=wO~p-&uzgVJFi zlxVN+TufH4Cbl>U3}4wl5#4+1ZZ=P!v%e3OCF<`gGrnMWRZ%H+UeY3)NMlUg!9nZ* z^#*1^M~`Dku%fs3Y0;gEmXw!pOV|2cn%fBHmiIrs`rAgWS73Q&q5{A($K4}!1pTF? zvzM-}Bd@CH&pX;1vQoCm@@4pd6VVqxF}YW6&+=&cbnYpi#^@w`sN-TN5F_G>LgdCL zjZOQSINY_-jvi7q!la$0oBGUE&=Z?0l@Y|JN`w-_o(=bG=vM|+x=LAnUsoGvd@8FI zu6ikN>D&6+WR5bR08V4%&uBfq?64I;`L>5wt4a z63dS{hJp*hRM*>!Ok@;EPPu{{^?~jP-ok_Yk=oP-$j*?nEq*>&h)+PrxZSbMJ5F-nn1i_XAPRF)`;U``LT#z1J%7unL-eGn&`BGiZ!A_^tIPd<_!` zuS)1XVWFKy2@^(1G4>)#kkhB9u?+Ue1x}q!)Y#18TuIbSj{>!ZbI@-z^JYC&?WpFzP~#d8E_^28dqsbcdL(dTneNoGU5pV`8aFK;zM z7#H9nW*BriySk;QI64Ar*{zF0oUjTcrSvHDTdT1l(%BqwGp$4o2^ESIhmT~)`n<)yXS`2{Z9qhYrgso|3nHJHm{-Gt z$;{@2Z=+?2jkuq#)Nr<-$U3>Z#j0D5=za|2(016kwd1KMlZY=wQl&yaImBb?Mdcs9 zUlBciFg0P~mi?T^tu=jlvWFv)JYHJjIHx2_PIx}$bie$FrT{oC={>b?iq6sBiY&Z)~et zk#pom`2f3C!gDVEvo3S%+Y7{NK8DNkEAD%G2XW-|DNZg5Ul;Z^EN4=I3{+FSn3yeH zD_^`R%B|}zItKM2lEJ%@>#>6h)_32%_rtv5APjTeZxP`AkT?w700x0^6yUK6J{ZZ5 zwQAb&6&JlZu3o5g!M!L`sUl7&F!&$LUawV`^FlOw9d4>U^wtd{qTo}w)xa^aCaj6D z@(ugBz@3&-M&ft&1v0zARH7M{L| zJ#Oz!FSs#u|IRJc8T1&J`edHTG1xbGOlaR2Z$pN3OlN$*HG8h{6w=%+e_hC@j3#Ap4&F7kx=|hYEzz%)j7I&$Sv=J=ZFA^^b#Wt z)l4KoUD|IP)KOj<4f2;WKT#r*m3fyXi?#8tI-ub_@fR%Im6FG)d$;sPAJn<8wUPkv zk|iea5SScbU15{5XX8(Ht|95HCS)}209IUNQkPyy)p#Ap^9yTyf#tmP#F3A`scqeM+ z_+&l&D4cTX4@0B1^f=(?K{WHwbdgw$fDJ9ZR|gUjpcCk(cAUD*7%#KyJarMd(0cSb z#wPI0>QpQVUo0-QjZTbv%lgN3#QC?iZP{^!?PlR*`mlYy+@F z*d*-%ySX~&_WhatrEkyWf)+4?Tr53pAAH;{jtv8zx^}xpn=_@X4 z(J!3mK?%rm7m8&rv0&uO3RAHl+uo>1wXSrGKk#fMpn`K_Js+75W~A;fdr9L}8ksWu zmJS>fflknAfF8}3y7dUt?3{2=P;Dx63m>h9lT1bT>8u;||}REt7} z`A86oA8-H>aTQo!xxn!>ah#)nRcJ!dok$c{dIq8nqT7JzW_F;z0QiucUy#2q35_IE zP~SwYnVuS~lF;ELt@?5#P>vYD^8|c$EubJH>@EJ4J37^pFuCOib#&|x>gZRg{~Kmc zm3%Ldgnx`)1od*;O%WwWaHGG1dW#+UzhZiZe*x1A`%y|N^G{%Ue)-n0?0zKO7Bq>A zW=`72kj0=9S>VZB1v)VX@bLV01)vk7F^`PDM8L9I^na`nfbIP+)(WQ-1qB}jC1W#_ z3w5N>5{M72!Z?YdY{~>t!=r^F)Q3y(!KZ+UYX*qB6n6pKh3^dbS2+3=>~RnaO;lF` zu=hDa>8GCjG?E|V762al0HB2jC+kvi8GGs_sQU;wR9PLJfHj)uD!^A3S)(&qFz~Oe z(ISiVtx+BSj85cF+h)J?J2f|``vK_JT_8Aib-v(6A#^|d<@ZPdc;ZYgtaVa|I?Xif z_vHMk)^bjiJOQYvv{T|h^3xqaAS&Z3GoKU~L4!18loKZ-#2(Oug36s)Y|NS|RErki z;`S$u0V}6-34a{_(~wQ2Da?t2oJwSGwTUdi$3l?|M}SI&O<5Nq=GjakVil;zHN%_$ zqsFbrICgDuY*VtGy`(vrEgSqsK2FR<(LxvHcN;q#fBfwq5#s%oh+NoQ4MoxZ0j7?0 zE-lC{W4LuvbkK`P>nL*R&FzREsA6;V@5@rPdO;X_iu32P)ZJ9D$ad*1mZdhMmZFMi zA8TX_ZWh=zYy3TMn*5+idr^d(+z#v@ize;Ix_LG83-|6nq@Vo@{dgl{-2>+B1YmDJ zu$uj|-}?EB!CUr4GCg&{g8m^$V+#R2nqk5cX6PxcQA;2LptbYF=_QO*q{C~7<5|%PWCwjh({|UMTu3(g}P$TU77`p_j)IkX@ zLX}G%iICWZ@I_5-0L21$D`0%~tn-tID8pE!3ef$n3U(jBm`|eCElk{!>enB%(V^WP z``?($hFuOQ6%oF?{YI8RIjat$>Qg)kAUXtsW+LZf$be`N3g@P0NNS5|7gFZI&_> ze_rJoT|$v50DiJtxFY_dK$@!IZ8nwGTJ!efgw+PP)Mr~i&c0Xkw{TJ(_B1+cXEyHD z-i>bF!QV!c=*ga0M!Hpi=cVZR%;0@#!$lq)8r@G{g=Lz0h|+0!5HrzBVfKr3y~$ai zExPMnW&^P=N59Z*yZRcpHf~evwO3*yM&^(&9p`v2QG`EU+!fv*{2{x8$-`BI{ty}P zK{z&Wy!M*>L5|Ol#k+G&ijW`XeABPqH;=kHZbDC(gq{7!avJ2hVg>fN2>hi;!~1DM zHws#7PP_{6fre9S3XX`~OG-RrHybgMHSZDK1HDVU zG;puv^v&1L@De!OG*+4VrF#{;(M!u@ziei6s=nAj-d=pNmstGkNRqc=GusB!Rj)z^ z1jvxzO?k9I!xp$S)APv_gRh2f=I{sWF)D67`BU;4Zzn5w9XDAHjYU)TK>i6`PUk38 z+I8j#HN4h+Lxyu$LkVfH>z%VCJuc0IpfvhcE}yZBe$KYVw|quUFjexdr0UAtGl!{B zBs}j`QA{HDXlEf0qK;ruVBLJ}0-&K8;mzU#C zX@DfouEBUCO57MpEF{rAN*xfMFwS-;DdzW4Ms`?F0x0dqw7Qrs*R4_u+_6bK7^-CX zi-Jv*VuCaOR>n4CKzvd6{yplzHQKwEfwHV4`d1>bKQapZ4np1k8-$vTz1>oR_k$JO zbWi6(JL-t10?(V&9_s(-v6X`+SPNZ)?nTfvo?gXodtdrIL9PsGoR|z19L-xhXlLwg zcb@G#Tg)*r8gXi*pD@=t^-6Kqa4}#trptl$D6aKnx^C^bBw0RgJT5Wiu-M3@Fus{D z7;`TVfV5fDkGj3|wkW3E3lYm09JRR7djuX->3evJ{d9A4^&2w{hpT+mYDof;A=*|l z?*!}**El2i?3!AvBJHKdW~%(VeK$o|9aw<@>TXE#32iSv_t(_9?(LFv{e$6aH(&nF zcKX=F$iWT36KlESjP0$X-^~oqI_wufrw(ne&`HIk@c>=c4mP}?!RLaoKvdiSQO2ALDRm@jUny}n%HY5e@2Av9= zJDC+;*DuZeUb9#nl*Po6qWAk36_0G*5bU=RR;I!3ISQb@ZVh#Uv~_c4DkZl(9m`SS z)_5EF=$bu}e8vIW;wK0zMSR(HM>0SW(qUr`H8S*LsTiTiGmYCvB~NMFM>VPA&X`6P zp+n6DsOj#WoPG*xCz@9yHs3nc)pG}-!kF?p#*9;!WEf{os-5Y!Pu5WwPi;ZzFjYWw zy&JIfA82%!8F+xFH?Y%m(E%v3p~5{PMubZ4P@LrUzR!Kcy84E!^~HAM4k zM9Z13nf@qo^J|&$v2C``dA#a&fk$(bYk@9fvEi6%w8%{qug=$-j|b2hrFW$s40YsW zm%#iDUWo1!T8A4ifC5U9=+&_ogLm#x!5&1`6N(5y>cZ3~q;`~Ht+VI;#}k{*AC=t~ z#V5H(?rDGa8rT9O&ak3j3q6aRNmsA;=tja(8Ct#e?Kja>+4!2X9DmG{>1m$kFZb4c{nrrLEnOFHL^I@@psrXnGQB)T zgyHJ=ZWj>j5a`}pqt?<;3qpH)M5~qIJ@VCG^*uyVFdyBmogZ*<4_7cY2O&kC{V0%|Ex+d3Dl`gQbJtuywak zZ$z9ug*Pfycshu0yj))Q23@T(e-|t{eP)`nXEG1UH*Z~oPt%Kj2g1k9tF^*`JvJ9S zi1{s%W49>|Mzv_|EQ(KW22b{JQy%s7p0pEE#nszW>W&TEP!bQ<^I)^wmSDwxXbYL` z75)w30;(CS&wP&AWjmFd>VO!Wl4_(l&tE>?a}Bid9)VZm$aW6*aA~;@>q{jcpeJV} z2ql$0(KDSb8SYmJ)>!eX<;mQM@u}z6b=?gs#~-hq4m!c1lU{%2(z^34rJDm)2H);m zGlgiWhEG(NL9;v#{aZOE^=5%DdMCoIzRNs8+OJ#tNwK=P{T*|Y@H#!-yTJ+>axFw}k1##h1jAZLG2$+sIR zJSfB_oSwT+A1-OLo)^V?Q@9Fg#!6;XtWL%nMX#!Fpg(Nz9HVjYJDZ4_<@3_8W_zOJ z+hF#b=t_zZHs?@xMQ~P1;tiM1v13m2r#MGA8T?waF!>dRqahFJranDvoId$MC|px% zRe^h%~`T}G+!f&l%V!&$* z!RLsB`2L{Thb(re9vq#^6dJT5#I?XS1Y5m%g$;W_@-4j#_O;>oBG&|X#*tW6kk+g& zxvEs$6kJIBY&3CB>C=lgF7^p??bBuU%H`GKSB7)KWA^|?8wV0%jJ>NmL>=*tvhr?{ z@#AZ;h*|g7=N>6FNA4A9>$GpNin>A+iuOrPTBCS5+VWgkLo{rmCP0xZ#vyhXLx|zD zp}Jkm=dYGx9<{h-&n8jhdXfb`JhLx}bamm*K7O973?O?=X!2QnE7wVCGG3!gVU4+1 zGH-D=?KlIgm~qI|Q7aeYUY90#$tYa!wC&bq{Xub6$s>aMyx5{r;*YIkk2!w}`IP|R zw>gq`SwUy>Kp{pb`$VKc#QcMXh`kkzk;zPWwKrY5igKfS3r2}}MHJ07baZUT$F+M3 z(0%C4H81$L3bS1RKo#gF`-n^$`r`C8d0`)e`FlP5DUpd#1C|h74 zyqS?!N1xj#CaojY|Y9W z;qEF_9|L8k&WKKSXn&o%{6O$HsI^`N*6_8CVX=?J_jMjB^;#0U{Yvbm zKTuQy$7h7a-ij+s;z2Ca75pm7xlVga+6-zmkd0}CF9T_U$in@Jdj8SY*q(cFJ|#H@ z+beT~ME0uB&^gRi>4sh$smzz^>xMC6S;5-Z{p*{1Z$6f7J&63$?x8h0M{ybG!j+hD z6_eAr>`fkeKCZc(bD=kIAWzC`f0Md^Tm4ybY4(+-mQb8*pvJ|}XCsfnYQ=Ve-Q%#D zVMV?ciMPp0f8w1X@USJSL09BJ;DE~DtpQ+s`Vsjk0kt#aB7LC_(Z+G*s@n|xOk||a z+!s!GjTvQ13tRc1^oCiX(3*o#z0nNWCREoeG;+o#T$S>r_>?LmEp!VqMXt!po1z7<1HN>HOlqV7?#w56 zI;5-*Snt<}71ZjJlnKAF^k=eBqbcjuuJkOwqbS?4o7}cRo}3XEXBtv#T3D%-w+FAn zybtwuGLQfmT?dTQvfPKkcMv*k@C8`T4Kq*(#G@J7@ur0}YWwby&iN|lSzLh6RT>Tn zF$A!)=kmxC!I9nOe)i9JDW|#oHyt0nzf2cSpU*ysG=d8DLv|opX^jX9R#8MBZF;nQ zpv^6)|0`t(osAKrl->oD8l9)=RP1#}Z13;;pd=@bJcU?Eh}4;O@WXvjEk^jbyf<;P zE_OA($t(U|whd*>28ZULDXN9=in?pgYwb-_)9x(aQ8~9tyk}vY6v1k|gCU-v>v$cT zRDNyawfD8vs#uX_r;{A}^B@YKxmOVOI`vG7vLn+eVV`)&*zNf>Ret$dePwrds2N4@ z?!N7CJtN$_r?$~HlRtK-3z%U;H|(o1pp{1PqGe0(?MOXLiK9<&P2kG zB()YEF8RxU`*eR8977Ju8TcZ!z~zuKr&8brFuTXri*Wqs&xCespb9eB*Ew;Dr3xbV z7gRxXeyD=1FZu6PK~P<4P@x=fO%~vGz_CzG<%2a$e!=6X(3^bV6)gDu%_RtnGMo=u zTpZffe|56|FdTorRzkIa#=|XuZ;$k`1VrL~V#~FoJ^^zm48Yt!xf`PoU~!N?xH}Ut z;DlrZu<%B}W~=oYq;2K~`q){>>Cd`@UXA23_@F9sW|AckQb$-q{-ELo_Vs;kRuX`9 zHym0=88o@= z2b}yR5MIHedk}W$DQd)6zx#>!fJKYERF|U%jSdINf;{Ytc|Rbp(nO%%0hp=HeAmy= z`gG;b==**!4o-M@HRkxtd)&M*37kg=rYus$3j`te*OoE8Sfcktim4yamSC&?B~Rhp zjK2e5Ko;EtBnDGZ`ipj9d{5F}TnzupaQyW4)VoW@iTLg(y5J3zPDWs6=OP1yC*r59 zL@?(Mq&NQO zUNn7p;S^9Ai8=Qdr}M8LBMD1LA(E}$vi87lii|0G^N(3VCkB$Rh7o}bi0J?@`-^@b zl>=Ba1?UZU9W}or zdTvn~Wv_i^_s0-Wo%5OiQg5CUOdjL0G<3uT7w{1@WOjF+D5}eU@$~kv{+H+VfA~m3 zz`f-Jj!O|R%`;P;ETZ6)yDHTTkhy%|TMsOK{~U|9-Bf_ku_&d2oxcrEv<1@LlPuER z>Y_pxVcDPl9CloNK<+EUiL%u;w8_IAfzrH>%r@WP^z6cW zXPWDbpN*BixM~k)xq*a>_pP_O-QK%iuMaA}?l~ch$EC~L(G4yUUDxi`^N7t9pPDC< zzTpM8(M>KoCO)87qYIE0dcGD_^vdXY$zkC>?T5-&B=ns3^3x~^>~6y&x(%DtvmV8z z-yc19t(cf0w^BqxZUz?ZKCa4p(ab|GLcdPm8#%A;S7jx z_e|6f%d;Ml8t^GQjIK_ZJwcv1m%%M=J%#Tdl@Tz=?@L$Aa?CNSdo-gx4_1YnOQ;f( zyYN*~V@h&tX4G`%D2|lIy)Pzt+clwW%Ob>`@y-nl*;jTHMKjyk$#GaZ7<|$*2%M6E zHMZA}ke)OeMQFAzzsKx9kT_KO$wx?j8~d=4;po%N&CsVr;pWh#8{s--0pDtK=h}{ug`iM z{E>*)-)h=Fr{Z$2Cso?lcq!41R#=|!Ag-;=`l6d&ieN?K2EC6;X(`uBMC!;h-@L{;k*iDf*1Eko zTjUW@;(PGph4dmXJL+rqA5oU>WTvr~j2 z7LZ0xxw#iySN9uKz2|k#vS+Q}+QXFx2ZvSJFhgZJt@tR^Wu}K^PF9^Na-(mg!xdt^ zGt;nWKK9Eb%`HJrdyuKm7^}T!)1_w!bvgLtskx~S6v0xdp>D5CwH(*E zr{ZRfLgi`(pCgJpu}HEu0ZHYZoH|YZ&_v6)vwGK&ob>z+_t?>!-0pU{Vsn(-r2wdK z&)NSxkE+<--8e@lbiKFL$Oegez{#C^xKa#^+UvlzY|ze5@@?vU?RSI5zTer#nNd@W zM7@*L{XNf&HA1Tk_cd1S%AXv)jH)BNT1l1Z^8Gl4_1hl&>Fd~|I-3rWI)?fy+oqi@ z%XWHwSO%C5@<1(^8b-lOGe&r98XB!uqT(B4IYvq_nV2HfdrRVIcGvFu^!*>lXU~y3HJCaBeB#i%vR7{DLRW5-`1d?$Cj-c1-h}c1Ufn}+ z=o+1jD}BJs7QK$t){JzV=`x%a0UgQ;nGm-c`oiZ-b&6IWv?V1d;Qa{vv@Q*WmtkY+?QM_xJ+$^o` ztC+BXm|KK!xU zHd-RQx{7pFp_K4J@WIOw!LSb7IlcGCUY*J(d)qgnu0T|2R3Jvez6BTH#dWQP;5#nm zZtztthmyLbYpaCZEwC@lz+qVHb_33aj-Ydu*%^6gFGb3=NJ%YsQnEpFLswEp=qt>n zh$<2`C|y=S-b?9i+kSC&i6yl&g2d#~Qbnmu;_;k&e`$I*{-@<>9t1 z2a*%<+sf|-vM~-b0o^Vj_~2FFWo}|V+ys3S-S$qE%QH(hkjAEgsq$LJ@_?EL@-t$@ z$n9@mYAVTBy_vGKAl`76R2c&4G1K@GJ&qEJ9BKR$De7sNVBEC}R%h-gnVIpkQ=KUU zc(n;-e|{t}EA;+VXK$%zF8$ivA2@%;l#1+`;Az)g3)Z^IV{YpeEtMzY+r%#s z;@4C+DNCS=1!2t>wgPSF_OL_146x@2;S)9v67#V%R|j&!7`%ZRGqOR?Z$=hsOTWRC zqaA~8L5N-}D{^jt)m|A(8(YFEFXXHHSCzT;92-gpEWq_-{Aephu7sG{+DzSVbZ}t9 z&8F+y5XFJCPLQwXRV|B=Z@4nSKs(#N9^{5R`A@HR#M~92A8~LHNRs2`KF6VYbKHJk_l#%(BtF%l zpt1Rv!e={rEWBEuTy}{1sb@kQkUQ^oYHx10!HTABl^3xHsfxL0l_!L+RiJL}WJDqa zpl<-wum)|R$aEnFA85c_?wpvIyrvSp-pxZx-hNia*81$UBd77a%Ls`Wc;%I>6RG{A z5U1GXbVukBbgC={vH6p{2}$*yb44(*Wo=yXy@4At9-ycFX7#;1m8RE5NnUb^?Z?^$ z<@O?`@$XSqpo$(FjFCn++jS`G$GY8cRLnPuE7CoY2?<|xm5Zcx?~2-f5`F9F=+(_eWf8ZJKNfELHtao;z#17WpxlTrA_#lQxP$nR z*GR+N+R?}L{W1zYK{9X@c%-EcEnM(wWRC*ZuxktccFNV6wQRfDsd@S9yv|{VECRfy1cr^5-?2Q!6_-C|ev- zKgb_4p}%eh2J&zj2^o^EEg-KZlhl@WbG=b{XSw0DpK1YWx#@A~2SpJlHtt>fK5MN| z;Xag~$vzj{idYMg5KvYOH~_p3FmJ&-ypBc#3n{ByFTK;DcELg>V?g|xiixGm52`BK zXNH;C3muK5@9$fwr9*=kli>5%*&L>C&L;ZFA$gbCk$ATl0d$sp*edWwGc3aC(YvMlc=!!I|Qdh1zTCuaJ$%*g= zAg4FfP-M<-nE(gL%=cuM53SX-()af_(w-XDAy)%y5P=rR+k`f$@QBDd`BYe>r7uuB zvu4>o;<$Gx?&|COP3gBc`a|4yY>CS3WnPkGZipL$w8;|>`HZ`maZH!*=_BEkQlPph zRH}}s^KvJ(j`5kKh0f_^2fPg5%P+ocS6EoV9KKwyB@$k()s0;}A{voPrj|gYFC|T1 zwWLWmzOn7aHyeWu>(`WraO*50o;DfYAkdinD)_!vl`6-`({tV^l0&eH?e6n#76#c~ zJ?okmCb0_drf`(9_iMuI)(&wC#f>rooPL-MRM*1mN<;fjq(8_s2w7>vLE)Tr2G-S@ zp+U_}{MKe~5PzuVh;J7C)UGdSA5Uo+(_Lr3F&~cQAn5Chx=?<`TO!kn5?a1?3(x89 zXdAb{k48IVwQJySS+dklXy>5eXcknfoX1p*@t|3c;1%ZF|btiMh@G`e004vo&s{R2fo zrIRWnsPnCQ>s^qJwILK%48HCUCf@gc1GG1a~4;*$q`bjt87iHz<>x6mUi@BDI!pcvgl}yBdWDbCl|HIc51YmqELXF|!3bQ8r zd(;%v7ARe{4_C~J=VQPzwM`Uj{AZsZdtw|xWM3IM^| zojjaRMlk|GUJ%7xo9L=S`G)Q~GmQ?T#mG>nW}wHc8CI#g0d$A{9+m=f6NtYsuB{L9k0$)wZ%|UAPW4slm_yZPMS_XlDexs)4fl!841u<w z===o|gV?nUHxK;hS-qdDkV!w4=C;|7UPsRh)lQ@2yAkuEL{MD;Fyt|$n!r+eVDl(* zYdT0YRR~dpSh>%koBIE<6*$NEB(>^L~j1jxASjM#@! z?E#W4%L2Ly(*h}JD&#%LIU5h;yB<`B%rFoNn|dV1=)ZctyR(Nu8N{OmAbpKy zO03n54qAHQI(|{pSs1@qK8(fN08Qn=xu3o>;h_E0+>$Syhx8}>T0i_7*jdT4 zN0r{I{;RLfIB;L8R?(cwRZF69)8jT?*wBUL^JbFA3UKJG1tTkjw56Sr;V zS*(CdRs8C+;rGt9?u7!>mEn?fxPE%hD3>5mC~`2ZxL2|w=^ajo{2OfL=Rum z4tp9k=v9+H77HO$*OK&b}mXr*%BIn%tV;l4=&5 zSg}`yh8C_#^sh+N49_-zwv2y0DVTgR)L!S@&66YjO@5VaFh6_X0qk<;IDFp?n)ppc zZT@m{+LpH%gEMy!=^pafBAmoX1o^8>p3$pQmFHWR8^{!Pa<;iumCWw0)iiFtywbX? zU0}`%XO-n0ELeX1-SdOqw@NU$LQ^GX>d82}sfQPyJ<7_9&v%l>OuoGzr@8HnNA|Q< z`79xg7~(+5K7F~GG3}CU0yXOu*R8we9CSS|Y74RPrs0rdmuZ{=6pfypQs?oy7)7}I z2x@&ZpHe-0Tum(TTknC{2PPTN`_Nk->Bqw}Q;2xim$}nvPa&Fm0uIlS4E1IHiW}EMpy1}e-!`DPlp@PsW zx2tb9dGZit)lhR;=Wf&ic6>~?IpRfNVS?quhcY}ty7;M4&+`brrlM}+&)fDMHZ#w# zF1dNH=RsYr=E&?-O7WSjA`9+Rj>*)A%k1OG3ZsNXGxVz%$&#mAZ}gQGWO|A+uK98I z&kVoSMMzzs~sGiQjYs>Yz{c*yP7lCTj1;m z?+vX->PcxILM(ix*02eG-QU;(M^{`g?-LevJ38Jpkv)uA@x+SvGvz^X`sLYau$^=%as02N`8nkJbU|fiRw{*P&tm_v@807;cX`pZ@;+K#8r~mvz^gv9X6dLmc zUr6o*_?Ab}w0(B-C#6B5gXfG^t0Tk+5u%<4H5$7EDJkA13q15MNf(;|4gc1UMhp5I zI|LWF&x3sagPUqS-G&Umjljx-m0lA~IG8v!)vdMU%1D|Y)UnGvl|_K|6eMBh0Fv`V z=y)9QWf?1TwN&VMA(BA+8cCS>{jQ>;=O%KVeW-J03<=+no4R87n-0hBbAZn2_@WMn z2uozZIz! z1CblaqV73j*NzV^LbV;K#b^zG!}jQbCdvgtjuT%!d04^(8ihry-}N2ejn+nAi zojM2R4(24w_xyvL@8{07P%kHnRe^9PeY$!~5)2T!7(F`yeC}O)wiW`Aj1ey2kxa^ggf7eB zPPwAXEM_8mZ(jV)7J@i8A5*==InONxbI)@3e#f%r9m5(Cnz~<~DTv-A0Ru%h%dEx@ z+{vl_i7c*O?(*@sgdKtuu( z(2EgC3ISDsnsxI)zPDgSBnxZ7rLE^bUmHKYQ#fW1YpKq`>|m)Tdn)`?mO;USJyrhe z5a!~R4#{=`tO@Wi4(3VjDflQ@IigI26KJrVJ~PR(-yBg?5W8$4mf&VrWrLnH1yJLt z0Lra!%=f@ZkBoK(bWWb|Iu8gXg9uFC844Ec$^fWUff&NIT|2VjmGlE%2+&9jsQ8Zg#hOKSxoPbkVWC&hAgjt zAF??8tB_^E*_Zh>LXlDFU)UFWQA|%&_+I*u!2AwFcs%z!b43)CgF3n})gDB(?7y+` zf12E%-`N0x_ST2Vx^Mw6C~6qAZRi9cxQ;;_n1BFi`liYk&mAKd0)z?!h0 zPtz{HjyKT7_$SV~QYu)OGF2eJu?3^*72YRR% z+M_TB%5uh0Pw#+O;GOy^VBOiMXFzX7cpsi++UdYBM*x`|u_UPCIsh6*6}-WIs4(-v zoVoLzEfRdl3wnnnDHdK8%CgbskG{(NJh=az`Uno|kv4*~cLKnJhP6nnK4ex!-S2;vDXloIcTmY3Ns%eBHgCXu z;png9D@EmYWKRT`Zxxk}xZ3TVc?J==;Ss4f!pfpC^aP~l7FCQr{$#DfUO0@2E~lp; z^P$gX8bdJne8Z`VxlN~ATE2KH76C5XBu4>z-6B{hk0MbX*r6ztoyj3Eis26nuj>y?Grewc$M!@8b5!*ItW55br+wolOzx2d_o*7zJ2~rSyu( z)NE)Z)@nfJmY-aZvV}#QJ|QoV!xNtw+3Woi0^STaxL(;N_rwOj(alODTvgce;(a?~ zxjI`Fs|!y~4w8`)m@1c+lBu>qL;WH8*Y5DXPZOpoPv}y^lcpTbi6;K%-c|DV`FQJV zc%8s-SELG<*`7}G=Ck!KYtrvdY6?-X6phr)I>8)}1ZIr%I*dG9R{744Z%=Xft_l?* z(4%reB6L@4>%R~XP7YVM%Jujd9hz)=LwklbFxt2Nj9l*@y2d^~L)V82b^;8~MlZyJ z7-~ULNxx@&=Q&;3zc@iZm@HRGAXU!cruQ@hb{GUuEKdhzxLow@0QFM3hO)*68oLHr zQY12-%cG%3qVHs|PG~R;ziMHy_Vmv`0`re3a`5Wa&OOyg`@GG10xc#fajsI;5&Dza zPxcI0eNfp>*cwu;ga;zF>l1J`Rx)Ak{7z5f2qvS(f_uc{Y0 zGfJTG84gNdycwE{7seMZNVz&r&`af5R3=ku7a82Y^SD13{?X$Wc@Dnt?>uhY?>z4A z4+_BJ26bE$3of@Q`Zt$b^9(pFf4JOxW0foAbe|a@qm_=o1ffKKaQ#+bd*{Jr5NBX) zz9oQ>_0`OijaxQP%#!o1&X`!8nvr?qk&q+3zFoSaNAJbb5dnmF3da)OHpvlQS2D-u6L>j?{Y6<}Fd1gEiZQg;o!hQ(9l&Iq| zEWUV`%DOu>+5>CWd}revk(@~Zdo}EU9t{8?(~qA4d|NTFav=4;Srk7lr$P#W5jaO6 zBbaOL?cOo2g7sK3!2q*z#JInh(56=lYf_!*gRBey?*GmJjQz*}$oyj&0JJYyG8{ir z0LH^ppa&K=&t46N97HM4lbz4akAuIY71c0tf&>Vq&)P0>83R)P!1h@jsgmCf5U|~V zELS=!lIhb29ORwEAh%XnBdC+g3Hzpxp_Z_b*i?b+9fFK%m`dxM=dEVgPagnzV0(p&yuc-oGE-42b8G~#Na z?l9@o3lErUA9M8|n_!GBVX2=j3oT_tR)!WCk%g;&pC?db9As^Dxv41M=3OMU?YeHM-g{er#cX; zS^xOz3)~@1Nx0k4;?sdd94Pb-jQ^2A0DxGV6Eri9exTj}9%Dv{d>@3g3$Wxbf4*(; znfAE)G?Vw~b~}?11N+K6dAH{ug5nNROC_T8hR{mI?u}}GI}(CpMvkRSdTt2iSiSwK zTb+sX^d5-czRu0hAi57bo{1UrUU9kR04G{!NKL8hz>(t9D_332gf}z9Jq;;7`}Q#C zVhWZ*r<2;-0d%?rlPX(VP-j{lv2|c*g{n>N;|N8F&f53BT zvw50K{0mE}U!0CsdCTdow17CUXAcWZK0N`Byn$hCMi=1Y5iMhE5850Y7$SV^ZBZ)IsW1h$oaA8hj(_eXzJb@UNuD(zZUy2Grrt!sudO(=tqv%&0}oF^YOF(3;@%O zDfprQ%mgOqLXaMl_ffs!UUIi!Ce@wf_Vi*;UGiDx(QR_E_y)($e7~0rYyri%dm9w6 z5c(df6d-k`PcR;ssltU62$+X9`(nucv9X@7%l26R8KbDYn6t6-Vdgt-fH|ji_|<@u ziV?CC0E23J1v0Sg69p;~_HFzW2HepL9ir(RRNcFv=ivd<-X%{G9KlWq(9DNv8A#xi{u{TG_!+t02(oC!-`Il+&!iF?ptK};&H@~d95-)TW|#GS zY)#X=0wcr@fVBbL6jer`ES}~LTC2*Z13?qdVU!eW*{J=DefFckdCT&!fQNd>n09wI z&?la%)UBc)F1V7hobrISX8s1L9x>|aGcF@D1#$#y7;}4pSDK6>W8R~TNw|6bv*_t5 z0OkiYEv%k*2NM7LJ@U(t?BD#3aRus*0td}PnaVkz9xEkTx}=y`V(P+M&f{*DqM{?GMi+~WaMfsSd45Q-G6SdE>XXI zXDdcn0@a9`k=LJ9p2jHD%Q9&%}+0@SN69xw8_($vD3ypnQ z{lJKDsC}j%k1&L9dO8DVEf$FdnhFQSt3mg%bRIx0-f48g0UlHWDC{UEjN#yFv@Ju+ zCq;dO+siE??#hXA$TM7PsEMS+wJmq%HY?2oU__LKRyr`iMFBtchh) z<$`Hq|J)$#-~49e6?AbJIWqwYS?u~w5x+_Mz2W=LrY4TD)3Da3hfjkZFJeL)rW8H3 z2WrK^M0^~eNc#3w&`8Zh7AdLn7W;+|{QDPcXs7e=3*sFEAShItQkMnGz67GXerMAL zY7GL@T9vFpLE<1DsFwu{X98e&uvuWhoFlk}8rUyHYLYoJ!~F?9H|RVsXu{xfPooL# zY)U6yUN|%ahBr9>onzy-E%E0kjxH-BqSHVaU9SL#`o45opSWuEV1U6|bZYmM^s&hEjle#mS{S{+-SNIn zKE!HGE%c}c#pCOsMGQGSYc#MoM?)l?+K2b+dS*Zd*xd16XqOq_z`R%6i0-tWXiWDJ zB!o*YxB%w2tNyP!_&2BiAKq%9x8#u$&g0m6RFTM}Eho=O<=mLwrGttRs5s!{=0GW% z(rBpX2qG}(fVuO-Ab=T8F*dTIkKg~4Q49!3zYkFf;OowQ1@ab4{qq06{Em+%T)DbZ z6I+y37hL>WcQ4PrT}AN&evHH+<6g`*ssv_}o0xv>q+nmB?UPuGRylh@xhae z<{(E9>}s!fE@AxgPQMAC*&y4U!xn2g^$VC9HsdLL1zAsq z_T8hM7_>ou-o|@!%`u)exRbcYDg6XgVbi$;nMNC?Uh^*g+-I6u2mG$ZI`0k~*_*WA zSuBC&xn_klu)N%M&0&^XyKr||QKo(t%ja!<)6Q~o#~Q9JIK;!M3m)+A-dI7|{UK{d z^tiD1%R?Dbb2E%>GJEQSo@QtiK1q-g9{)_>8LI9<1sk6TmfYrGVK6bN>AIKkD8#!J ztq4~@hz{MLPGNJ|77pX)adRK z?AXxWDCK}d*R!xl?~CLD)*koAG(~eqM;5i3Y3{;}j+EH$84O)x4bx=jiGEaRs+r9? zJmv85tBkTi4ev+Yc%D{JPvj9A;u^MMRGw?6c#0PHfr0{t94UxDi;M`uPn(@`m*;=I z+M9!8^4+W?;l)bAfofT`9s7hM$zRUX@KiyDUXvSoUFvx+GIZ@p)-*(Rph|SG#wNw0 z`LDqSpQ{O zt0~$$dWhzo{#N|e`t3SSp@8pZTM576$m%Ug*!^Osyz)^jx8t?M0D@ z8u>0^SooVD;5KkY7UoL7%t{ZWHQC$NXfH3XE%%F+6_%V-`4aFlm0w1K`HI-#+(kjG z_K){RTgJ~Rks!aw!V|kaE#_W6aP2B$97;)ZcIuF>*`dCkp_x-R&ZY8JwPF);t z@>FhXh2S@ge0(5Ayu&AAyfNz?uL7w6zLZD92dFE#sXO@K_~JfhfdAK{f4sHc9vOq8 zuHNFkCYp+od^z+rLPcIOcG?p|xJ}brb;)8+Q@-|yf`@t{R8QFr_IhG-hlcl6adHOr zmB`B1czNZ<+L^EN`*$tu|Hnoyf86+gb1`3s)=k{=xFT3Ip({O6phXQcI~u-yS6cRR z4Uz30?52^vH@UDuxb;cX+1IP=#-_?SX>WugA$n%*WP!BHJNSi22oKFu^|FN9EvDKl&Z9V4j>>RB3+1z(u?$- zpa=+31Oya>h=53q^iHIUh;$;oN-qf|1X8>k=FIPowVv~_FC^x3$|ERohA#P=Kl-{nS%wpt0`DoG6iRdw;T=Gpe4F#jMGo5PH)d9-su#lVAu7GBw31!Uu^}Hs1~9Gg&MrJ}cuwYN)xHs6Q*E>_ zbT}*kPIfUTTYj?LJ;@AZmgUl$a{d(6**o!wapkfQkf`D6_S?L=&A+xWf2seJBr&F_ zIVYNrf<%jqyXNkFHT-Jmz-sI`!jO4oObz5^2T;))D4J=7UaUiu6svPadBp6=sPrkW zb!??9Qy`qtwS=TnY8X-1W+1tKA;=zf9*)#`i_Dw2`w|x&@)KU{5x;qwIN?T{lrJsd z^LHU^ee=uc;1LyJf+OAzUhTdPo5926hj(AqvM4%H>YGK5FgiUdsJ3E# zj^Wr_(Yf($C!joIyfjK^_ZDgob>vDD4)b6 ztRCTc7<|O)^$L5WmMu!X)BRb&w1l>VpHqHU-q_9RxzoWpXis2hG+gRZl7+%0DUsz9 z?T%xkf-ReiR$Jlk0=M6+)(t!e5Z_e^7iCjR(S=pxa}zk0RQ?eBcv1R}_5{z;BTXUI zugR9EeR>)%$D1ZwozjkTl~cl(^_*Z(3-@Zqem(t2lX`MqWDztR^9gQ%eTrztFvh+y zqVbu#psCq$p|n_g`YDtXL67Z=S?(cSi=Y=cwtX+EW(5f&9)cYk5wNoB4pwe=30&1_ zq%u?ia_YMFd?O5wphH=$;tvpe-5f_(CDfUwj~E|+vmKz1$ogUZ8+XU-`P-2eX#b`W zM*7MTsh$PLSvt7KRewB-9ie{SBXBU7PBy?+ZL6v(8V_Gs?Y}bV%CBBn34Joo4+MDZrXfWy?bJK$(8ZV zsrIm=b8ME?^rv4Gg2>Q6<)Vxc`C0ooRYpWMWm)|cY2_EB+ZHB14WnHQ}!5)ln~1TSUA}cahshsrt{^4LbIy8x=PP zy1LZ8dnf30teX?ZYvbQt&RFg>Wu%OU5gb}Py+qf8V?RS~&Io%piK;$Hw$)MBqzwD? zjaLW>H&}S+1im4wyDlT*O(%s|x@H_J&%cP0P_va`tY|KSF=g3|)(mjA5zDg8F3b=fI;-Q*(ecNwLQir& zdqHkpeVp1>lVv5Q+vFxv2Kd`jGf`JRl#B<5eKy41z(neyLy15Kvqmm}tolegY>t(E zu+P!!T=t{~m>>kCVtRTF&@@~evs#VBUqx0`bs#5#(G=VuSqs`D3Qjl3Cg?89bl9M! z!5LrQSDzuxY;e*rhL{E$zF`4ds>4!=wp6~*&9{YK&T!|Lme7g+@@;Uh^Py{_AhGBv z0UG8h5rX$D%rm~ymSJ>aOK<-2ZGLz0Bt?IhRk{v-4*!zi&I#1--iRz9s6=(`0TB_D z1Jr6z63>^Ce5RQ~&;PX@UQt^U`YDwwyXq6>6P?^9e)m?tw~ATIJauT@*SKX>qoiyd zcEkd!mcwRdNZ|^8Leg)ChC+L}dhNLHAXoEAOu_2s3dC{X| z{fv@K)=Y~UR=#!3GXDQxhD%>wDtPQ)ZqGb37v!D)(0%@_Lb~(V z^|0RF-1xi3H<8PHusk!j&F%PnK7*oTl-D{UXm@llD8sT*Vp#W(Br=4o#aWpG1g_Z9 zN{hFcD`VFg>&kiGfaD?qon|OF_|SGcYP-Cpf!o$xOg2Ibk?7>u?wVozsjAJkX)3V`_j7 zH}^91rNXCPxw34iXb@s4dma$8J@=Au7++EJb~C(rj7*GPkJ5+EvF!Qx>H_MineHU3 z^b;XQaVnuPNrlK4Frj9VhVh)Msag6O-7ZTABAaU|R_sfVzLLekH8(9qeNjQp5*m@H zGNc``e(^GuR>^ID7{f|2J)eguFVyu| zpJi1HHd&mv$}oNrvcTmCLB?CQltD=X#KamMbB7Vj$ksV^0akgjYDPI4KG=c!WH0_9 zq<;^zx?JGnWI?=2cE7*}7u6FhCJW1xQoB~}tZ9p!w(Lcls)N4E$df2o&Dwisnm>a*HQ1Ihk4mS31Ff6@JZ4NSk$j#J@Tg8W7JMd$FXK>D{=hYG(1Nh7d+l zN&2 z!28eUj)d;%FK_nC($UR7@=9;3#4ec~+$YG4vXN_DTzqx6n@HQ{KqhglkxYf(U7x3e zRhrnT9^g?yqpGN0vd92U0EL1;?a=-9>a3P&R;^hzmP(RZEF<;^#o{iWn&fXOZ?$9Q zu@xxv(E23pb!JMmeMzaco=!4wc`RnpGyw=3e+6PN8v(%%{=WjE{XfRW-#V`Ui-9}0 z^)DYf_0)z(HxXJ>S46TtU)$5V^TtcVi_KM(sHSJn*6<1n-Bqi|+J1@VGv#XMd? zp1ym^1)x~~lmUC~^#TSEloa`~@bEKU=e0YetzT6qPIl#9 zV?o_i*#GI1EY$}yZ8-zK2iq}3gb&a5?HI?}LM)51ad9;&$P%gbc(2o-!Se9QySR6) z5vq5S$|B`r)`$EeDY<0+_-cHUb0mR9`w}j9z3ar%RWjGqIKnR8ZcF|7=_d+}s%SU& zH8*56(Ba@$J>~v!CW7Xz=VhXUF%ET=0!LmZd6idMH&h!pwj z6AiMOlPtw;4|O#v_8)e3?MoFXG{RBz6Gt z^qlKOi}5P$pfa@Ss*Upo?Q!=+rvB8^Vz5Q>6B}%uiT6wRuMLQgGofNns+^3SQ z2nz_di}A~R<=U&^uhg1pQY0c<6Ufu}&Wl2a?7-R$BfC@TDo&OJT3h=Cn_bdZE#4Yo z4LZ`Da+Q6rT@g-nz(vauZ<{qFziBg4d9LPl=e!EFo66LDgE(F;AZrl(J}!m%5=wg& z@o;@De!GP9o7;AX4=R~P0iOJwpCRh)4*Wx*RcF9JbPtIwXY+gCu+BS8nA+_}&4*A^ zP#4|RK&PJ~PZ?>!<&rpU=N8qD z-c8rU##l1#u@_mLkXgdVyQAab{mypL6fk!!P1D)(YO{~0+OU_$P;)?1PR0H5y!&eo z5YP0-bbgv2Nlq-|g?X9o?Yco?Wj+Yu*23KwUoVf5&|T(6BCVg=#U6^5&fy9MMN=qsCbp2m#F3H zU5ixCDw9)}>oBG|m~P~!CDQug(pUZ0dkEm`7iu1YEw0U|A+DWzqn{X zRsM5`m6BMEGd?f8q@HzAF9ecxZ15UhHnTPgB~V)871N0#TV??HV3srrO19Mw#`u<+ zJ7?J6?m(Gq4DMr$g1nUB%g2iQ8t-efB-8f#aQHtTqux z!k>mlmx%Fv;tfutlct(fJf_4`ccCC3*-iC)k;6wzt_hgZ&4!_dn^!{}i0%^!L}Q#t zH8{S2ciK{rm>$`KU4`6sz}{SzUqiVsTdcoF^ThEzWDascDYU_G$V~NShYHS+W3iP( zTB+rOddP4x`5sVz(Xv~At#nCApAvT`L_4Iw-$aYM95;^QW!arVUYUa*BwrxfW{{cj zB5J$J>q14stX&l~u;?h_Wc5m&QemH#x^|mD{+zf6gJO?V<5}#jNAHTYp%WMj+|tD? zmY@UbQFh(b(6{Rsqa{-xcbI2c*n=a?Xjt6#Cbm8hRx zJA?wEXhk1Av!1>F5_;h7duH+dIhe(ARx0MDlL3fEUc4b^YitmsD?a1Fs zYsJV7Zd{+Z!rE;4AEkr+H|>+(3ymp-c!>x3mjq_@Z@PQM_oK*WdcMprO4oJ{S=8%^ zR+u`r30UYo&^i)+36iB$VKc^Mh`yVplf$RGeLf}A?LdmO z&%?24(pE<1(YFIiBAJwzgPcf%>NAS=uw2xesXVs>wMJa|rw77f+0I6fKZGrIR!$Ch zFG{Y|$p!#oL`QosIlA?#x#yruA2~w!y`{+${|8wslO;Ue0Mr@X{j?c}u68=6t2w9@ zzsB4>nK8e&UO;%apQiR?Qd@ZL$5#!tsS4eR;FM?poWTj@BYVsPOsn%cq{G>R3rasR z(HOyg7*X_=5Uqe$0K38We|CdcLfVwJ`#4I*my+k=?2y&)Q>1+c|Af#3XpTWfu-g!+ zWjjzR-N%4HXsDo(0&?Vuk|#V(bOd+wCD0B$b-lu%=82UR(~-)>tNIR(h;r2M zRCWVUW9@sQeW^!@!kP?(uvgoT*qmC!<0ogr*1LX=Khrs9&=}Km(5~mgQ}s_Vc%I;3 z8NNyocV>R_BQ(YYjq;4myKnz11zK}cDGvT2Y+prwA#8v9hEAowehbI%Q;0VyPh~HN zx~aWfc=s8YKW(rMt3p4hkgn?&o~PHI;aON0)m^Eqw0*-`G!kdpc z!i7q=9P|TeYliSVae>@EN1;wlvYI>Z>?-cugnq{hJUqZ!S2tnFT5z6Dg0bTQkotB^ zi>!u7In5(jsg3I}-`hTyv7ZcWuLSw)%GKss$%G1hhHM$_DX*xA5NfdKc$IyI8mmP9 z3~8huCVGwroKSrH`l+qq5%2>tb$%EWA@reQz&_d3;&#Uzp8v@;St07g!SqCLt-FVh z`gG{jLs?}Rr=j#K6J5&}lDxL)41^!;CuqwUBMDwRXYaVp5VVx+Y?=D@%S$VZ3ZGC4 zQP`c4^6Kac)9ARKwe-92ptz(0FU$}JK|Qwfo#oj|2St4q8SENr@1*R>{LV1r$C>5W zb3u35DD4z}4<{GQKFlqWjC^uAF7n;SiTGvCi(;(N<00wmq|i>>$M=}i*FIXL!ahZXQW@V(fJ|oR z8?{k2liGr7`t@weerchO-q175bMNM z(AIR@=E=S74&)|i&q1TW&yYLu+tWI`JwL2EpA~yg^r?HTn|S_KqfGO;(MWZ?n6VPk z5YJqH->PkXxWFdI2$`r5dQF_3@wwPMi~)JOpR0Pm{3};Kt*({hZ=DJ?9-4jh;DF=b z5VCsCcsk20w!9pLjz^(f$n^1-_)j(FbbK;;@hrMK1U=C^VeCgi$?oVQ`e;6^MQ|6~ zNJ(@nC|IR$YM=P&BXo#e5wXf%LeaSZxePgcFn@Io0arblIGZ}XUXh84Fey&9>cJGG zztHDt6`FAdNg;Rtpp6=cAFm%a_^}3ltbu>m8ejswSgwLO=+B8lq?<(g$}XL-hG9XT z_eO$d&(3^?T%eJA_g3*b+bzDghxf{zF@|3U^Osrj(B4owvg&BCzHXO{cxTK2^R4Y8 zLqQ7r{T%yLjX&+)2^qn}+OoOUYZ1X9On6O6e@ApoLAn!9|NFMLi~DCCf{F$2w%mFK z0RyBp?!1rQ$DJ+NaNyN#$FsW`kNWg!&WxbJz>PU@hMyiy@bIr%&&0-!gl=7Px9aT` z5_?KFn?oAbdI4K`dtfzvPXCL89luiFzy6vnZ9U4)KfG|?DRlpgi$?d9q*g|>i-;;bRHmWWj(jhhZtktI2hWhCtRwht zZ!2=-I(*%pe=d~HwNMqmaDN%O-z-|yBXPPb%&6y@kW!$+?v#&1AyPU^=A?6;CY=Zv z`8r-GM9ln_xfuUTOIO2Z4?XVK9aLc%G@}oFd>RrFNC5}r?ujGX;J64^UR|P0M!!vm zBYRI&@VJDK$pUH8pW+bd<8^FVA41`K@tsvlXdlq)^2K1GA$Ar!vfw|!jw@^Jx0rgR zkUrzwaFqLFLVb6y)l<#mg4GF!t7*gF7R25M7Ugde*!g(U{0~v9gmN6v~Uv1oG@&$lF%`?y$FV zEd;5I*aHTWj1KOcq;j{Oi~G;rk~eyg>2A zMghh@eeoZt8#vuo!9i5g1>;kl*yAIT^&4CqM02>Wa|Vr`JU6z?!o_`84$6eARKVDj zqPx+xgN$WT!)}Qso$Z$&>U7P8%PD7{56vy;7Jhlz!2jH7KRV8ek2sfyju;N2x)T0a zfMa2aY^FKWTyoX2E9ful6|!SAA1RK?}dsbJX%~s(# zV_Vwy_9G&~sJjG7PW^}`#7Uyp@F0B?dskM>ef^3;tMDyqfk(XJ3K}AX@AXOU3p#a{ ziGU6+GfwuD!G&ax7}`ur$?&S=htH3j-@5Qdfu-)6Z^NXpqVg6@EVAdosCYU!BRNVJ z;Axkx%B7rYC}L1GJ7&=9^Weqh4Q1w4-d=S*g7;?Oo7dA6sE zRBO}+e}>H@sj$s3ri08+PsX`>f7 zC1ktRibGi?fXSPC8=d8eM*cmPGo;cYQIcL(*2|Edo`dE*F($jhp0fzl;2@B2(0UE_ zZ8vt1hEz?0dEROR#SLIA6El83BL8TY;34JA&K&D`W`0rLer|2JC+$G_9?VhLP%tVl zn#?6D-X+UuDPiC^=i!{3=oVaM#EDq{=`+NT^YWmsmp%qQ`JHu}e)IDTvR#(xsfh&z zzO2e7Pwyp}ZC5=l4S{e+$U%=HD^)R;s=x;f3kHY!*^U-ys^u43yiZAe1t0Hx$GVyyuRd8hmUTa-myaMhQgw287(NsFg-6Eqx-TX z+56XE5Jc?V42st6`E#2=ai?p*Mo@GNIPZ01#e#zT6Nmq2!QjX5z6_N7A24R>K|MNe z7=5Tb__|C1OO9BJCR_R(TDv=noB`)0ZFsU5{kD4B$w8lBaJrZtICZ}*i#c*fo)l%> z=HbI8)7holw@Pj-{@lxbbj`9sNA6v{4?Lzbf7JHyG-j`-VLCC|pd@@ryVEAAv^YkO zY$=wm&kPLkY*hEZ}L~aT=5MYL4B!-MThgQ2ZQKB;Atdd83K9NfRba%BL)pkY3~)&dZPa~cE+jEQOXia1?(u3R zik~#L5x)5sS9BZ5BEgyKtpv4-tk&jz1QuXkmsPze&S;kY++_5}vzSr>03j z5^IN%98*&9UL13`V)_lXjP&Q^p7V+qzfE{6liqChoN@Ym<$z!t*<)A+S`)+5HZZxK zpQKX|P2V~acK!t63|B~JHazAZ4ragY)|pKQ(-Y#e>0mx>G?v<`=j|@t@m8f$X;zH~#@8G!xopTw9 zKJZ6klbCNsQIgLY&uv6x3lu2(suS7>$$d+ctCk=ne`G#12SV~JK3Awfeh%K|;!|dj zOkAyUhXshLCYQIaSYAL|pJYy~<@D2>y`#jQzGjQYE_@>v`Cd@AiF|9RAS~j7h!Uzq zg@nP`j{FRx8JKA-1A7`O{8;x4Bp{jxS;W-40eEBfC3_^z zyDfXDA4|AP2v*CH%H*)bdFa8Fb&JCNbY(PdL; zA-QKxRdIeSjyOg?tbFO{5KB!!N$Q!2e>UX^j z{+jxNPHcc-^C36PuuFAFpyF`r0=0Rhnb{`S|81;Q7Xh=_qfwyX^mdIF(ikq!MDXhG zWA)0w>S!uiTkjGLZxY#dp=Ec}O;-p+`4}c1+M7-YLJ=NQ62#`5*#fw6V(`*W{JVg6@GR9kQkf0okQ z!`rWg??6|*TR%XUrX=!$>5gV<9)^A+%F<_3ire$dRKp+Ivee0WiY zV@}95G*n~H&pY0xD*B#@OHNoXUWxzTcCF#fPz z=sy8i^0EC1Pl9;f0Gvymk__53D|>E>u4a`1t%k^YBXeDM)(%+53}C4SD&lXmilIlZ zv)EtY#L)9ks(<5_TbuV5|EnzrZ?pXt8&OTtms6-ax;LJ4Nm zzlaUbx*Hj~9dqd$Q98V-5a5I~vOL}yy`kHvHW_G@Z0a^{JDtzgU5k1%DSQdC#rw3` zM!(KUp^rL*pa_B~<%K6~MCxe|jA2lMU{vsQs7K`u%Z`=cd&)D5`vxk!00KzUHBo&K zt{ZnNphLLih#%b}`S}woKt0``vLRXcVo5U9!(Q4mAMi@|Jj=zc=b~fY`w#%FOKXL$ zkL@IN5;Fg0GE6sK93K*iTdk6;4{$1dHM<-kC9%Qagr@qc)= zp)k<4U`!W+Vp3+oKRm3W%Cxyiugml_h?Nv34v3w?BIcQN-qxNqb#?_4KIQJtiMXo(?|GJpfw4(bjmW-ZY?hXzD~hjy~lI7IR6 zV5qG1T-~hLVdDVoBXvmVkl9AYC8klg&1PgH5hM zttd&DVHSKrHqJw@q&LN~H(!wLz0H37fz?*V^{M@S>($d8H0z%s-$demGwlNovMqlD zU+4a-l*nes4PTWFob`147QD%3Dk394_}KhLDq=dkuJSP{=x)7xJ&1TO;oYRsW6>lz zPj(z8)G;n~d$w7QnoVYe?VV!11lcLh9MTZS3TS6Y^KEwO1ZLWNM&GQ~9|V0T{yCb4 z5UWAG@urXxPNWy|6ZHV+X^6SCy&1!c$9l!e(d2^1)?tSBvT|>P#~vvl4(YU@zD==b zTI$~n)>f`kwIwOuWqOIkh@VR=1&x(U0|{-z?n&9NZ_ADzwHv@B4StJ_$W1y^>i_?e z8=hK!3t(3?|49l6yHrhbLbOIu+faGO%CoBNxZ{ts_f7SP@-!$E293IIQIfm%Dc~IK z9UzFxv;xd*%$bfFMuFN)pvaLw!L{c#y$stgJYSi|#F?M37um;~VGL#1|IH9frLw zFYTV$Tw7s!hn-F#GfPp;P?S2*$Dj(sH%Y5gHE;kcZHLyMjAv`jXWVP4V^9_>rJ5f; zK4P6_SfJ>gdqeBV(4B+v_sVIhEK;LL%0A2kC{c9~iBDUB6M5>%DAQ^oQUDkuoi&d^ zzM{pI5}0t}X`FhvLQ04;X8pXE95k@gQn}4Ge@?skV!J?=7`qDL zRzZGlUO;k;n;}T@1@1I4Il7IO`_VQnv=$`-Gd5KL+GBj2NJLIKib9apyq>n7Auqsj z1(1gKSfGj-`2q<>gz_9rq4*35GTKO6;%r9_{380H{cOF93avoys^9S+GU=pZAS5+u zS~f_1hjN>A8gIf`t zWU;mv9c05H#)GapYB4vg*~beLuXR@~Mw;yN4?B2>FT{D`%oPvTk#s@|l^zVJIF>tX zOFA-#T=rYXFumH+6_GS(*^?Rn^rhZ43(3gY25U=G^8>1)i>2Exv>x35{=M9;?aD`Y zEn|qc7J|XN4n^uw$jwyB3g8dAkH!XrsS0v! zHd!xJ9bp0t+*^lo?KL{?J=YlkZ8ewlz&V~cJcjyUu!fP$vw$IXI#9)lTzKZ;tjI$t zBMT7M#Htdjc6ze~v2xb&56q9#PX$Roc%FULq3~=~#7FG%1RR2_1d91sm#5%oh;ba6 z`e^Vn-AKNEn`7*phgNR8Xd{2H(v|Y3f|*uxoaQF+I-pL0^Pb+*2Q+dG+%#I2oyc1Y zYm&A;NW$64!F59h2=os;(}u#o8)VjC|D*Af*@*XlNs7!2#}1to>;>ufRYQ(n@<{9Z zUU>7FONLvrYQ>T7^cmDU?Yi{3zpCZ^-N&bNg1v9uc*b;euRVjo-&CYGDo={ho)^2Q ztmd#i2zFQA+IY^;YYuklgOjXkogKMk;fs$M1Bv$-RS+d^!8)FLF0u$qk zKOBs&D@vmP-SQqNFKWmj8c^${;FO;#(6ttj{G=&~9zwe5OYT3vX|8;_ebnmm&@q6^2Y7jJr|l zoA8wO_;oZJ{|v-ad<2m%5?5zRE_>%|Cw z62=tM!}yQvfF}>p_kV;0vbUYWW3`{lDM zzB@ag=XIq^)$-xl>;O*JF&e+F-85Wb_s(#Zl>A8!_YcGr#wg5CB8Mm_T=EAFzI2X$ z7~|`yJObNC4Eyk`-PPq%QHck?y$FA{w|`LBVrnHFU}J%>ukY%Y;4eII0v!H3L4?Ye zbdi;D?R6Z`R%_?A^^nx({RRfr#rICGahytzx~&n>a9uKHh<}wt-35|P$9VA))M3=R zV;=eqD}=ujIUG>|T0yn45$^&FQN9ZL#H0Dp(}0USQ>Q~TBlDpDz~V$z{H|ZA9h`pf zHczk*NCRlS-QahV=ZAHIvZz0Kp21B#SYcqijO$8#)vYAc6`H2sU~OkAewthTi1(?U zN^fraQ%YNPbbc(iAIt4OcDel?wEy`2m+guGoVU^eGY-&xAHc0v{EJx_E!a@;;L^{Z zA@jhA6fmvn$U(J`l_z0iBv}?@pZaHrnu`;O0{4dDR72NfoBiOc2Y}Ny>_>ctd`Cte zaD_j>Fgk-PI6CqnG*!*CYLIXi&(dXr$ZE@dO*a-a^P3er}BRYVQeh0F^k0I{p**GwRR}E)Nt(O68PBj5N zbqND9FJD{^ye;m#I5yMQ1^@{74{Wu8_x|8g_`aqmG?Wt?TEm7E`Zr%RUtbJ(7<1~s ze?$oW%XD8oBI@IrwTwRi?tfgO@fYL~&IcX_XsI|>Lp=r-t;lD{rKrdiEl)t>wkZKB zcYqiRPJcC1B=s-T4KaXq0PR%x-qr9&A4Q~vV+MZ*R}hdE8hEDyl5d4_^*gt~AAQt+ z{at+v`Tt+`H*}xiUkWxeE1Fb#;6_B%siR2J4S1v@%?e0|j#AZghodPqfc;`!;FwM# z&=a;6f%QF0ybaEg!YE?>*oBoJj^$tISpLBG|A<0Ni)b1k0QHbWN5AhZwQND8MIztQ)CdyTvMKo}>-?#L?`k?>jyZHfPzg`7D0PPpK z!5@J3pSYwSfc97H`~%Q_A7uYu23-FcS^FB$5;Up17KW{bYA`sUoabdC+Q-0{4mV`k ztvW0Co%b8R;jI3hG*;( z@Q&iJI4)yJueZ{NxDv!jGG92nZ!U`XgWMk+iLHa;v@*9>*T8e@{jKv4%eL+!z8sJ% z?MB=%cNK}`$`+*|su`t|z%coCPr&%In-#%`-$_eAK79>_L)W(>$7(QSEq9^WLEK<< z6f6%`HS)1+{vG9WzYLD+wHw$&;d@ev41Zo{R5N=6bevRSgD=I#<&mB%)wsJ zah>!t|46LcmWA7EzR5!$u7=cLKS_}J26rP{0AW7H14b3RqJ8`|CC3g;{P~sZEUj#@;0d#0TD5%t$#N;BYk$ZHhEua@b*7f!R zHO4^4;(cXndk!1c$XD6|5unWRJC8d4(K-q2B;^8hs^4NItw54?b?(*hncIsFv>u}` zj~+i1?Q`~u)B*V8L0r1$X^{MDCO?S=qavQ~(IkM*ToEscKvpx32YJop7VL5g+z0g+ z7Y{FU{1qtwt*>9>bT2KqBv>ft@=6UY33B+A$XO$&JXANS-R|1xC{?6lMnK%=67LkF+%l>Y9|boDX4M@rVnEDH_vY6cn*@c z)e2sqeqdK&HFOzdtr`N58$&y^iN@5ssI}8VU=)WOwDn&tJ+-X<=~d~BGX8#-PpnhZ S_}8#sIw<6qPmktv_x}O2@m1yoe)*Y_}jC@6@4C=x0nF@yq&G!`h`BQcbOAkxiHDi;9>m97CKqz0tB6=|e9 zB!=$J@5~UQSMUG-f8Y08>-*Mw*Sgnpopa8!pWXYn_jBB&B!o}nk>i0tpwo{Y-j@M^ z&X9sYC$5}02K)_q$`~c^2bP(P@LdqHo?;yMA8g$_;&(uxPu}=j8aTlJJ4=blJplgX zg_H6Rm;j?&JeJ0mRZAQd=OOc?tdX9MFq$0nTWOhrO7k)v&v*RZlgyB~J z7tp*-tUf9`u9fQjbVB|!QgFt8L^O`d_onpmd~gX*b9BANa&n|X~$9QJYyyuTV7nM1Xz{=v6_+)=ysFEo9VBSn9M$9+-JcjD*LVK{K-;X? z`@C1Js~0TB=U~pQX?0PXML(@Pu~I5GEx^0OwA!9M_fo0YyXNHb*f>gt&bC?6QCK91 zSUVL0Dqt6%xT4Tc*YY7Q(dbzK3P!Xe{AD$^l9zhk4Q)n}qJ3)p#ahejMor_g-36$oRJSL~D}H z^79CSuq;TUEv2E*{)_I{AWPb_U6ZM=*-kC+yVEP)@_Q14bsd}8=38aM+{JsjMghEP zXC|v}Bh^(5W}KSB7pf6$X%;y8ZR03PLS<=JMk}1MZnxdq>@;9CYIjpxmy;lZxJUL| z6Uy5*OZ{o`VVN8J^|Le99q!yq=F7}&(6Ka@8nx`9@*-qXPK%%16?t~C%)+gww)Kld zh>teV0p-;DyMBl#gEKR$IjTIXUsr4FZIH)GTza}V+&YuO)H?a(L?0%X<0kQuUB!dV;5f)PxNm} z*OsqI(Y=-(%*XJdr;%~f8{MtB8Xu&xSsfcX`;UIp>ofp>C9Ccih;z9Yp^_KoP5d+}3(6iWh=wCBhX81@DR z!<=iy!jneTvmV~YMY(9QR3Sl5m!FDP!wnwuD>LD%j?z?VZAT8{CVsPi*lmgnAR?^FA>HjG={t2}I33t)6<*)|f?BegD^h67kOd zsGHo0q$KLkUQa@Zs>grSxq72{pjH?jWh2BQTQ^p*)f9*dsG|$fsEJt&_3&UyH!qZe z6YeVDgO;K#7`EV=qstBTe(e6V$+2(PD2mBhJcbLhSg3Ui#)o|Clg5%{lM;2pkYj71 z9|~HHoOJBSmULi|_Iws5$oo(+U!o<$NiM7$TJedVB$gE65u5~_Fnj^;jbaIErYFdX zdpT)*Q;PP46Kilkyqn2Ii|fnNj^h6M@XM>KBz}tlqGNrKjq`0gAJqcd-Z-&Qw5P-% zV|HRvioVlMO9;pnbzBh_3Z~15EZm35N6raGlewJB0 zx8YPx<#p|71ncM(k#;}njnZ@VHhyAc-GZYZ;_G>RCleIoAl%^!24vz*NUp`3&Fy!H zSG0Zd-$>3x($m#~jh_X#oULbKyfSwcamuU>s+M{C#926hUDb*dmj=U#!SW4dyWSJE zttcvw)>y5_I#lOk{JN%S&)wnx(zjsLw5^R@quYAS4auBg0c<~OA?`M(R)q3dkUB^y zEUo6fMCXPMHH!iyZo%MW5(9f&x=Wc?t;e!b4mspmIx^TIFKWolsB9%Y$a%d$R1xwW z8MLT!QuxB)V|v$YqJig6WvDYYZrM}SovUVj{wz6h_6}wL%=h$GD}E!c!PmMAzQis0 zL*)*iPYf%GEsU}ZSFg#*vwmu91g!2#$mJl0DB|(^_G>Go5)W8;x+l)zfyN{fWi4%* z7?jWWJuWt5Y+&>2Vq%YjLK5uFm8`T>kM6J*5jFj4-4NcC&bE`60Jz#njH|5`iF+-<#Oj%w zRN-j&x!%!ntEw?b`-_Y3ai=>^Evi{0xwY}5&N}LaEw)jMBvdoG_*#yBR?s8fVjlf$ z8pC!AZC-v|6FD($d~$tx$2>kl%qR&z>-!P$lZ?#W0vDVXpLnUzA8cXB5S_Bce8YB-yfs}I|tzSwC zq(%%_NkENZ3UN~l(XvWs*b_BMQao;IFHk#W@h@oc2iFHM7AitNbHkcV5_S0Gs#UDi ziTeFz2yJnL?NILtt4}D)5E#*vbT>xxKV9yfw~-49lKgB3?f<~NPm1GAR1>i__Hq$_ zPD(L@Qph)+&$cHN5^UXE{?frDiT8n|+>#%+{kg=QMB02Qkf5Vdm}yOO+Qx$_iejcV z-zrCt5{tYo^UhQ5lrj;cHu+Cz_rCGw8k&|HtIFdMtR`N-66jQY)q1#ka7ZGN=ua=X>6I;4f9W+z`NvjdEAo3Yuv%mpx}Oz4XjIy}CWU=BN> zM&5anuc5o04A3#jL|H?d-~bUvXQi43F}^)XiJD=^aqFFXL2RPE;rU7C z!8cYV#@*P;T%#!Zgq*S?AK5ofD1WN=OV4E3jSjD-L>E*D~u14k>wG-x22%OvssnmJKt!#LRy~ zc_QI)+%s;&)+ASvpcuo3Sx%9L*4hzN9|NDXP{%t{>x)=@e_dO?lQ`?u@L@fZd7){- zPOfs1Pj0jsZ~}<@?OYELNnfIt&5eP!H-J?oiTlM{KRgsp4T=00gQ41N8;f}kvk#9# z0@PfvB6yM|f=2Wd7vbECoea{lv2)Ufh7#iixxUNLs=7SJw((#ulz=aTS+Da0Tjiqo zn0M(_rv`38?ncojuqHZnTOOfw?|dsUs7C9P$-6y)-|oTfI%D06TXV!1{UeEH>nL(cR?9U#Vj{DsqA|~E~f4< z=E?)1=fHs8@aRV&wQ92ymhNBB5UsA_r30--qf^q9q$1KumGQ-oY!>4^Ygl_iT6YjY zCsDcG4d&7p?i8jF8rGc{p1rEiE`SX|TlCOP>iM@zYt1ed*W^ru>UKYWEVN$0H;7I+ z15yE$H;qi|FQ+#_867>O-%aKX^6wE`CL0#RJ&(N)ECo1RwK5iyFKHztjmD*DJcCW$7RJ|ZIdJtOkCr9*l&a*2@)F!Is7&6u< zQ~;!M8_0244yK8uaA#>%Qv2RQv+@}?f3YwxK?9O&JFR2 z%|XVO2);c!cIP~t~%|)ATban2-r8FsLUuCmzB zY0s5~(9s%wp|k903SE@|z)O3Askv(Rk&n^hE3-Ii|0@pDE$#1Zu8{4MS+()9Vj*&v zb@h?7nxve%jI@wmm1Ezj$ttPH>O27S2=+?RZ~sQZZ(1kL5EcxP0X+O1f$zRuiv~mc z|7!SuOT3d#D-v z3CM2b=HH#gqtm|rLE*)?d}u7Unkt}??8y>1^RCHF)fXHFtW||BaEZSB?6<7J5c?)H zb+Ms+JRw9@LV{P0t=N@5EI33SO%dO{{eYb};DW{uo(iRAQRHt>HO<4&nd_)hYI z`i8C$k>;dk?_6$kmPTDbgw5XJ`-~0_O%uNJd}~Z;A0HDT{H5E})aJV`Fc-Ucb0*3w z*+@>>=1Bxg`qeDGBh)F$l-zg zy1o*1y%;?@lp6P^XUoYuym^d)EKIJ*qD`Kixdn89u%|#gL(4a#Z=zJ3_X3tSQYqt0 z+t3-S%NJsT-@b~Q2UO^(+F<`yVC+-9%c|4VBzVaQ3y#6YS(!klBdin#?vIPPzJ8q~ zL6L~FhQ8Y#q*Sfl{RTQ7eKYK1{4={Se-vl7-1mB=gqCQDbO=aY{5`kv*9qlK8*{hK z#hPkBYJFaItVkfenOhMDkV8vqvrN{6eiF}25<-Ozm$Jl>^@rc@&~Q-C&r|K*NlNE? zCTEcD<9>Y`RFIoM`suOIY{?4NSHG^0)e7SV7v$xYf$+V(H5#sf1wwT!+i>8iz;fMN z=X_eNpq07}^@v#;YEgZmi0aEUC(#{=F_GDlT%!sOS$WqM9FWoT&F)T5IkH4nLlIVW z%BKm>S(z0e7G0u+H? zcnrM|6ljM8j^m)80Kvv6(F^6A1)veJKaz%Sd{ln=-`e<4mW#a~qu|hwjEADlheB1~F;AhawH18S9o+vh2#>~4?d6($z8*?eB zN`|NZ+l%L4wDld_gr8q_8^j%g_II|6e+yUjs0_Q;Ic%6WKT^PX{s}1>1nyp?fX$`_6 zRm|N!cKZ!^khijaHK;_(XXI`#y>vWjAA+nvKwwOTfKa{c*yA}6l{t&9FKz_8kvp{K z1%tpK7Pg$$$O_@Wb$u;i-7baV0lgN=IJ$-@GT&2IOf-I?e%2bQ&m}XLtSQcC1beT+M=a{NMU%fz11HCJ3~!V{~%X(G?`2 zS>sBX-QtVoWRj|n1sVpZ>)R#33a#)bqOSMpn>V=W|vpm40wj|lqOs9pc}Y3l>GI_jHY@{8fJDM zK5YF2C1IS{i|P9sx_U2`T;#EOp*)Rg^BVKOLvi~$a*2ENv(fEh%=_oQ)Q1fy-bTfFsa_u%@!6!4ioL0mC0TP^!Z4EL2v8DdYk&TpKJqBh7a?bHA z0Hm&#XQEfS4H8B7!i{~*1XR$pI~_}n27$gT|}{&SpF+T z%zdZQqUSs>gPcw+)MJBEZd;v9@0vcel;h*aK@kFdXF)w3xuSusfZMvM4N_Ya@Z0=RNkC$sw(g)E@Gzr*R|{8aPY9t9Y!xb~8X z?4|iJP$RX`Ng|gkpuZyA6wUweG>JRt{EKTk-+oWe8EJa3`ibDc2PhT|Ktvb~t-4h6 z`BE8eIn5K^*2(&T8d^t*L*b!tN6aUez(~a3@ zPv?Kx9iAA-JnH}58~Op81(a^t_bw66H4yhVxt5BligOQt5ir^vkZ8cql)xi-#Ri#>W<9>TgB`<{lQFY+kOfAd8DH-Cq}`NGNP za^;9@k2<`y^PWbwcn?SdL1Xu2rHRvt(LU~|KckicI4}&`-k3)r1eR}Jzk0CEJ*T6$ zbvp%MV^zq;GOC;~#)(MpfaB(gwZuOv!__i~Z}X${?pGjvll`_vQ}F znb;D?YrFY)sU%)1WbQaxIXuYetZ5G1aNPV*0a}(;10M!rVORm}QjcDYo1$y@>W=XwS#vNsS_Pr@x+ij2bU{8m~YTg)B*N+J010Wzr4KYgi%B z6hrIR15e_Dw+FttdCf!>P~rDH*5^(Ura>ZngE)^9czn0#aov#s+cp8J&!itZ;0d@l zO!Nr*;=PELVvjTaUy?nfYcbSc1T3NI>)lQwGb{SL*Xc2Q`t8-tK-K$5zRuunZLP2E?mFzisayiAaLlhEzX!A?n|FQ~fG@8g zol?>h%8~Z=g$g-(0fmZ-ti7^z;_|&aD*dC=E#AYg|3P*y2D4yT2tT1BS(5RFt-u5` zHF2i91XegpFb9<4SH(WkuNJ+#w8w1BX}(M5v)a?e_PGQE%4bk)Nxpusa&=mgx2VZD z{!L03tpko9lO0iF9{#;gGisx;@S;uAPLEHaE&6ZDijsIQRt=)<;oMD$IU#gd9sm{F ziC}@8(r-G1LeX|b0Plc$Z6pA*AgRzYP#12Hw`G)R%z!*eC}ZaL25gvQp3kd-qAZ5? zG88HwJZ}HIVnjiqy`k z;W-Sm>$i=c|5&k+WKZcAru#^($qhDqmUUVBm^k^RKPI-g3$UwE%56Uxug?p?YYvF& zROcSEy~QZ`{b|4|4Smw$BQuY2gV-5s`WKslV>LN80i;<)Qf`V!y7yz2`a-7G7=uAM z3hAU-NnK1eAG3|zN_a)+GhN$&D0W~j&P85PWQ+#ZeFDpt zYIAsAs;|}?TxPyH+u~XEA^%Rrbdl=Li9wa7_B3-YvViWY5nBX%|I$!6$}oQ7DSg{A zi;G{D;YRctQ`kbQV(wZlqg}$B^pnWWgne#*`!$nch(h~CN?!amk}HfOYf0Bfp*^=N z{WrLl#ooHoc3)AXH8T&Bxx8FUE)Tl{4;lS<%j(JozHO25V@u6>6QAUrBum~5$iNP4 z`$pM7EUGO@Lf4L7j$_LOr!|LhG(9voIx{Nz#ncDV|NEact-zEtNlri`m~}-4iLx7|ou* z0S#I7Z6{C zmhb2kkrC#?u>8Cj>z2NvuF~SyF|actBTpLW=O%T zoc-3(W&fV?VAgy(JKsHx!%v~|Ao5YgY?JiBOp2^#6o`wU(tbuLBgn0k&q7QiWmdKz zcaW8!i)l&o9lo8ts!YcSvTK2B{P}Q*tf2~T-4d6``TlrRGTmx;x0Ur=u85rXf!6;v z&3yHGFHA%gqU5^VtZ>pyq6umI0!hlMIlg$(zf#bT6vy%~vHx^?K%Rm2T1;2ca2`0o z@0@~pa?8txhqG54n_b3w$Osra-5bhP0^Z66tgsa}xy|$h*mA!sZ$`Uj?%!v`SM~$J zO1P|EWo5`bQjkRWOR4tdtIS%gaOc|;@?Y~+#$`pAwpvBRs_V2Zao~fstCcW^$8Eaw zMD!de)A!#W4Q?R?c{e_{y*EI-t=N>#XYp$r6=g8N-7~vCC9vptWn9{4#yDMJ5>I&k z!&CM?Mtbb8E>kY7bEuS(t>R7Qu&=m{ddnlDZ24e*i0tk7&oS(m%_6x1LP#WMHI@-q zg8?_c<)W8sLHA7fz{>y46>nNJ|xvVlB8y%+oNVA_*p0E z-0Hbdx8~T!&>C^~%5o*2b7mLxCQVkFBQ5p+UGjyJ0`Hdu<+Q|nVEjIocIG|*X?c6+ zqyfgyOAGb{iDh)H$C`JEsMHp_i9Z`!jxiE5xOa!gMR zY>?I@(%oE2hheEDT;(Buf=NF9S@*nGsx@rtkWPdS(D_(aVmaj#J~bM7Yv8xk0v`qx z)^S9kHFc-+A#?0qgmYz@{K9SzLZ_Rj@RGI%87w%IFa5pB+t=y+_X?65Pb6H5J981Mv<_qd{yzYbaAXW*Q1hy|UGz6f&Ku@-$zzn4;W zb>~7NIxM_yTSz~Cc3F@`3xMopXp5w(np#cG?#!B+mHdmAqyhKv*$5>Q!ROCANZ*kCENMnbEPa% z#v90daHq%+yGhqnH>vUXRWC{5e(mVw%FZ1E&cC`epaT#84)g%U^Io(k!#1pN59+w> zDavd~SkeEU1w*M^x(uxz(DO7zYC|t^5$`Lrx2t?|1*e}^5ym8yEpfv4T`JF~zg4j4 za;%>(CR_%?YLzoNL=YX#Ar2Oc$YF1_mEwurH@ib6Rk`_K$;7v&xndksrlweFDYWN7 z0rg9t4(u~$Z3N~V-L5!7ly~6|{O;Rjj+`ghOV@$D+;_g24`zvSSePEUo3YMuVi~P& z!4O|JYih^LU6hP{Zf-47@LElCpQ{YZ>f|5%9v!SRwY z4O2=7z;fMrO+X)A$7S5LX<}qp?8fzhWzGqHO06NR**HZDb|J@jGlc*#h^GfB{YrEDFNF3BqKZ7O30JinC;ND(2hGTb8BiH)^R9cfFAg{xUCe;cL%aluu zOzjeJ43f6LXf+L1cPjvxFYn2FZ^=;QbJvW8lYwfrm9|;LUbuX&M>`tl{nU~O-GtW2 zLjuWvYr)Rjoi(q%bV_Y17Cb)TgWrj-cFEI*#^nS*vZzNz2O7a~ORr~&jm@436M_Iu zFPd*P-lU*C_sCL%-hKv-^4rfceiYdNz`JMEKS^!s`F%Ume6T{L&ezAU84<@rT%>Nj z-4)8IIi>IN9;KN7GT!*N?6fD-5Xu-IE zbNjGp6J2hA_)u}BuI({0LCh%5@Tx1iSOoBxqa`^SgFKXzu-!2B1!gNms8SStY!6tvR<6V>+qz#dA~HTLH3H&)B2HXpNBNWmGQoC8T6y* zKYi7Td*Yj0p0}T?(jB0&=W!Fyfdr2-k9x+h+|Ci@dPu*en~DuGC#ji zlwT2rBDs|e>(wY-3(s-vu8q{!S_t9Tho5!EC`kCJ$&J5~rp=dNLS42j-6+0^P{RS} z_+iHW145xlQUZb2>trV=Zx)R+?D^xqn$K0Ly~q*@!!M&UE}&eWLaF-+EwS3qzE>`c zrvb+JZ`vd<52A`RNC^2j8FcI&S?xsT$K_b=_& zBy-a}r1aP!gbIr|fU`T}Kv{hYIK^Q!Ev!U3f{9KS8AIZz3)qT$SZC8NNBu3RG7&OeaJ<#SRSWR}^HpI}QjkN) zTj09Sveps3`w5{~KgAzh_sWw$=}lSB;`Y}1lZC~&-DA#`YubK{_L1wBzWn!LF}iPt zwGS~f0N(%^9$+GfQpbBLZlqRM?2g|KLYJW5a*OMXeg$UuU*d14_upR?V2Z#trp4s% zzXK+8jUxeA;k?IBr7Rj%=h&~E*>f5N@*xBDDmzv!)^B zv6}|^jqrMAc5^H3{)l_Pk8d08%x5{>!da8XGLQTd&a-}I#}#L1i^veP)uf8$`0R{5 z8iqUs0gmEg{4TdZI{uKjo@V(MfB%}<=C7{K%}jI;KEBzT~jfa8v_!_L-ce_&qV=BqR+8~?XwSVbX1c%Uwm=iJ(c1Nh+F?NQ zgO)9(5p^WDI&7ir$MnM<3%Y&r2cZA3JN2V^@!orJ5k-hXHKES7kj5t5%YM8iq>Bk1)K3b9k zgTmL01)f4n;ycbunK`ySzf8izxeyD=2P>jY_LmrLbJ!*sezH|PeUdkN6D&M~Cl>x- zlTqOUdjR)p4az&0p#&-DkM?YTOg+{~V)&e0ZyN|$LkFsgoRmW7pD0I~Juw1Mc%S{I z)lW%MKW|K*^iW7R+Rsp5=^(M*iY$GUBYx(=wCgOM!C&Q~!U-=%;iwD-sdeQO?N6YW zgb!ls(YEe>La=_yYP*OWMF*hVP^qm8A%wq3qvR|;r6=xQxcrjjAVoX4!UU_3C(J zjeUl`bfNwNl>e;d-84RUmlx|9l#sk6-s*hppI_e~91`*^i!}{9T6!RCu5>CvBtW*0 zwDHS5Q~p!_y#Wp7-z`vCoH$J%c^D=ZVwHnDCJKy3+jqA_Ccd5;S+q>c2b1>N8DRqJ z@0v+gahLMWrwDAB`UWN)oAve90ZMLj{`jW$;o|z^_3`p35-8c-aoUL;BJ6?5GL}hU zDY-a%_TnirlDm({AYE0xJt`~nK$Ap!UbXMtAKBl_&?A>ne=;BJ!U&6HlmX2UYio#C z&GILWwK;x-E?RBd%c#^7%5S`GfaZhENCh0XH?&>8v$SOlrNBJB6GB+iCj_1<3!=^S zckQHw8ImYQdExnZz|HA4BL(>#)GDi4<~isIxZ_0F1)6q!!t4B2xn}xX`%h$?w@>qJ zS4SG9Cxk7BakJ2wDcYRr|Fh=wu0J#*Jt1JxhZK}6(PfNeYk2WJ#w=)nXSe~G&`hea zG;4NE`eqGMFg8Obyur3fA`dcPrjR_r7Q^rKXMU^;E-HR3v@AmkW(iW9&mT^YupSTH zjmlu~am!<#Az+7nxw=DybCX!!$E|y^Yd!{zr@cVa_V2X478iB)rg^3{AOy8s*6oGIimqc#e6jc)yrwq0vpI8Mg2&*=&f^ z$EoK(S@i&j2^fM=Nt`c)qMJ;dap&{S^XS5WqWX9r%ZDE1hkCwll85ESto4@TV((oh z-%nLrHfeU^!>IDXZx0Fp3MX0|kW(bOAY2%QPjQVRof>S(8H8K7rtwH z3}rBw(^er3=#nJ_K7U)$>U)D`{?8qly$sYy;>MB>0O07hx?jky*siFYZpcr97)6YI6BbayI_yrYsC-8vMMu6t+dj2XK+K<+QLYf85>-iMe5A@DAZH z!wt0$auY5vEPF|n_(xmf3z?oXNN(NYV1xla$%nZ>NVXMpz=lV8#9^H7>hQ}f9E|V& z{an!=kUEPTCB(0I06$?jhIA63R{72@(qy#L&=!{d+&M=ZQEby54p{|C@#j3i892g0 z?^E#lte?dE%NWxhw%n`Ok)-9%O5~b1GS-4nb;$`z^3m)YfuF(F^`m(Li*HepHZSe9 z#JRTkrR0nM*APto_fHso0iR3o#eOFRjy&Jn4$z2svJ+X~b(a<5V`43gwwdKB+L1kd zFlW>>NrG~EF<;JnQXi%e|ED5oc$%T2|HQ-hKHHDpAf~Q}!4<~wmNia>4>j_qnp8ik z@6K|1_pYXf!4L9k{!x~wfKi+jY`D|1HveA}D?qJh_$qh@zP4?Mkaf3chGFMNkt|QR zk7P7_b@13$)C!#;|F~lDyy&q~4eKiPC6;w$hZoGb+Vb7xcL!@1g?_(12l3lMu{{oN zuRl!RlAYnE{wu0lPu|6;0&-^qi+qZ@oq7BW7(pi{-ULv3m!M23AimC$Xm)d*1=UjH1QUNYWg zw$>yc>^HrpS!z)Pd@Y2E1DH$t8xW2ed|8yU6b^GH>;9%K z5>$4sJby30{o8ETfzP7WAgCKMDuGl;08UdNirB2E=?m)1;8a8IU@MNkeEchkO4z%t z&uvo1ZyS`X}&DyX3WXo+O5N!*9rVW>DdVL6OvN35g+bH0A_)t zTsF#O>%DEwe|;7+-$MMi<`lZ}1LEie11A5kV7x1-xneG>?qv~1?RA#03a6S?5hu@m z5?^{Z`plYL%zJ(wgDS()m7Ajol@oTeX1*s{0EeIX!7%^R5|@I0pXH}3A`YOvJpFGq z^$(zSokT;78vjFkr@r!~cd8M@Hs4Y%uOSWK!Ybmr`SLxasXIDZIM)Wlza*$4V(v$1 zKqC^u=CJ66M?B&Q11h^3sBt^T9W#UzNwd86aZ=Q)d8`9>^*C=D{51j|sVjJ0y}RBX{BN z`yYNiKFJ**vVvW$v8|c-a6PnF%~)7(fO6J(O@$U*lvWMEpqAInF2n>-v1U`sEb^PD@9cYXl7Q+XFkv|O?{&@&+ zehu z6cs(Tygx>^Pg>MLs)!zW-Y+0UJ7TWJdGw4Ffnz z_e}_G^?T!aK;U4ir|q|2907R_X|Q%aC*AWhT*+#Jz{x2Vx9g@?#1U}wa0*W~;sMEO zL|j`~-myy&F8`pnHu!aNiOvheh8FY^JEy>m6p~Nj!pT zVy_Fyg{AffjeiWKA+n?4eU$yp>J56SGa&mLmsj*D?(7dc9}F@&9fSA1Td=W$OTOfb zE6qE$^!Zgs;yeoIryLADM`NbFOHj-V2&>9fj>9EA7)XE*BLQQYW&scPCX*VJW%y^k zX7y509FMJ>it5A9=(aoZ;61=>a4&JA0Z|Mr`ujHI5807Vah-0Xj7UkRtLR)lR^qmw zeXaKb#f!VXzc0Q@hbMkW8NohX7AvoA=R1II|B|YhXmGL^;PISV2|YeZ(AeHd{C#Qz zYvBGb!`LppIjIa&2kg?7e4Rdt>qPv^fC!o8P+4qh`98P8_%7EYsX-FIb+PS+;(?EVFI8Y|D$v8zmaNjF@Q3HB7*NJ{o^DDKwk;|;6)^zU+CNhQ^c3$aVGbz)QPDJ!2LO;9T}vF~)a^z2Jz&lsPVEOQ>Tnt0dxYm- z?w6lGLS%N{6N{CU;vxvmvN&y&@`B#Mj?XMZBNbGhH^NOtoJgn)6Q z({c6u!yneyB~MHEj;u2Q$0gW)isFYt`~`x+c=AjH_oSAi>!mem=Vj*q%EQAu0};E@ zSkF(>H~it}xVx3g}G37&1t z@0nep!}S~g&-Xq%8Sr`g+w7UKRb1dOOUYp%+nq+wGajVJd*j~+r@t_9sKucuMQtS? zJSlK{jPqD5uO&`B=YJxx*~uYcZ2uIU9bueUR=^(m|FAS*R9-sx(Lav1p|h#@ZMYh> z9hTm3iong=_J9qp|H!|Lgnw9+rRMLR?3Fxm)zP>yrgxSk6=Gb3&O0&r-A*@pR0>2p*mpRZ{Lf|P6CIb*7_C8?0%B% zAD;xOg@h1%Xfl2Xz%&cs>=HGJu`fF?%L@OIR&L+?!!f-ai_%dvmN?|VwDqzVx*j|XYd;25O;mF9 zVI64Ckyuh?gq7s50Q8SBdme>3d-h*Wr6#>?0UhXMy)iZ1ErovF89j8(c*Hyo+Vv)b zoPQZ+3T_l3i^C?0oD);{_C#hGVuf?7NHGOWH;WgV&pZ(F$JZ=`0B42O71Cc5l&>y# z5-O%+TtRww8bs<;b-)$jX*~FDu>EG%mKC{f(N%kV5p&?H6<2;Lz%j#|K=*#z4Z#XzmzlLzVH;@rKX?I0QATQkZhJFpOfwf z+AJG@Q^*G<2VFvxiUJ(lkR=84#^0Yc{r1tZ9z=KcUgCLlb>$q0&ol=P!um9awbQ64iwe3~21F_mjgE?33%FfMAcduZlYO3XG7CF#n#TObi=spyG1o?0^ zAN-o*54eeO;)9k|KbZ5*@yU9_EQpWYwAj{oN;7cW?~1*P-Ye&XS?9rs~puo;XT<(_wCg>NymW5 zJvI8Fh2i5ekts><2DAlkxg1<&fTfjorATlz6D3)ZGWS$+;PPgkc+Ka}aOBx{hIm(( z;of$`W^t@#z?@3Y@SAF0M&M)M|17Feu=kn02$Ek1jG49v626P$-Tgdq(`*d*?srUx zP@hDG$$W$*G)j|S_9SaM1JF%2T^6&^dp|z)Ci^^<_LjvIo_gC)ns}HR|E!Gb^vgsW zENXRI_!ILYTz1e)&DQ6m1svGb)UL&tep@Zi3v7@r>+B(+zZI9H@}BWVjo&v``X22C zdE?bZA>!}0KUvEWxaudBJ>#`We+v%FIc1eHp8LcKhi$px$@Ah~f!%6Hx!Sbid;#vc zo4e;q7(-VTMR-5T>ByGsSl-FvLd19!+iz}c1n)-OX-T&6F5Otyj2G-@%BD!9cFAU1n>C|R zxylV^OT^afddku};(9V*&icM~S=cdoDiy!cjz!17^khC-z8V#!)X*x}H8q^xw=#bO z{4iGDex18Yx+BsTm98ND@JGdwuAB;qxMWT6WV?oeH`BI`A_o#dX6PGMvt*O#=5w-_ zU6P?%%40BA5V`p6wV2A6d!tP~bFV%pC60r70%smFWc+)3olEFwwdX6Yg`b*~jhn%v zf2@krDBbX;;ZzDVWHL80=!_O{4!|@7{H;#(_Hi#!$+nTE!K%#b{o~L7d2g6{YH?;Q z98;g^X3I|%imc@*%ukl)gVoZF7E?xqu`(DvzN5^CYSea?Gy}3GO)qawSn4Jc*%*&& zEnZ^KxkuzDTmFchGnPjNZ~FJHW&rdPq&Su3rCUDGZ?DH`JB1XKplDBgM61KXUMtQ< zLG3{Itzefsn<1h|SQ_4iA0Xx)FBC%s7dpHB>{SP8$<#07c|bArqlo%}Gv+5-0b$va zk`_@2@j3Z^5jb6Dx@9xlw?ta@`o z0)0&L06?yXUD^8@1;5dw^#Y!vI1FIa$?TqB-^fjC|M=nvA{%mjclD)5$}FKs?RS{7 zjPG>X%W!PHU2VDa;{FUOM^Vut(O9eVy13ntXD0U9s2Mt|I9&9oIPmGCCz#1%ZY#G8 z%&Ap$X?sKzPoaXVtE9sHakF@033o@Vp(V5ss!Eyq_A`zb`>pu4`y2|C2j1z~nONRf zV-3%NSP9&e9C)yYoc^um!T?v8iCCBZ^W?)`^A#Al67BW0p@QSL&lhPzFOm7uSp3bjZ$zQQR@`X2Z@N2+~lbarT|XJC7>on1H^A- zpJpXBQK@EspJp}n+&wAu<3*DoI(>`KbZ&-zNa-i9{ z{2B%vZ?_KQJZ{@tZeZK$fIkh>Tu$Up-a;wZd@^0k5dwcaYMXuBNZou5#u7RCWA1?0XbC&etx} z4`4s}qzi~*2FmKa_bo-$)6AfkWcv{G0BgVYHn^ej=cDz{uT#oL;m9fflZm9=DGT-e zq+eAEx`@|ywl}lgBn$E&>fWrU+byDi#QTbsKu7}|kv@5}NcnX@<0#&kuFR6O@@E+mii9+h64Edr zF*MR00s|tV(hULz4MPb?OLxc6-7(zH5Gso2dw%ym&;9p2o?(CXUVHDg*IMt_ zN~bYdgmP;a0S{cl_KyAB{WgM%&EAq3cMW7c5q)N)BU#+{)!;_xR;%#Gky-qRjs;B3 zyDMNsePoG@W|lu!1OfW4TV$Cxrqu=eFrswSg9?9{!Z*cL$RhKjLYZF!(mJ_n)~NRG zLe;=Fs_#XiwZm^$9CM4EtLz5CIk8Yg$sU{w@rmx;Qm*6fs0ZYrMpPiE8_&dH(*9ed za?9Bme<$DH_0k9p$nr(T0Wm0?s8 z(;x@Xozw^5M#u1e{BN2p-us(Fy;3DqK?7zg83c8WFdm zdYh(CFpDv%j=CH>$>OyxvNza+_ss|sDZHY!qnPs3%^+c`@o9nd2Ng#4h2I`i*vlb* zFbgf^x->w5#G7bp0<}1RuA@oj$BJAE!keHH1KkHB@K79KcSkNUqdM%`Z+?oqE!K~( z3K|mrTLL0Wks`@U3GfjlloNSYQ@AkF`F4cfZW@r1%~)tF-Z6i78HQL_c50YjK4&71 zc-wN7rPgY%E~2!w5T$w^6B)^&r=P>v4Nbi(jpwH{kkM!~th)!}#-&bUMvV~I6L$%go%+2HnAE18&3hU=-(PXCki>h#J+!21g`j57En}wo%W& z1MaF1ngnml;o4|JaTbEYw~Uh!CgQ4-%1(|q^N#E#lX#X6Q?%1Gu}S)PepGK!Q`vy4 z%B`dVJhsa0Znrdrp_m$|Ttn~zEELkeWnQ{z&?8sG2Js|5a8~bh-|1&UEmuA}&d){d zv)UOOJWbz=iSV-_QPP&Ej}Mgp+|5az)Hv^_?4&oo-D!v7ynt~t_h>{IgIzg!GD39z zmg$EkoQ@IX|LklXg-ctnj?PLPx3+cJ+x?=1@ZSWxzotkUjWht8DUZ6Buj)8}E3c)K zNxdyMP2Bi27d5jNQ;6U6d5eMab-4{KWC3#?L?9M-=__{gmB_1K;qFs0iL1|q5IC*7& z1uyRa*l=p3C{|N5_8Qa1o=IfvSBfK}vFUVc0%a6lpxLK7k9Zx=wQF36%#It-<|>I3dC#;zJA^)gUY9$;9LLj1!Pb_KyJ6uiY(8Pg$Q`y z_2PCR@Ls2yp9QCUp?DApnl?ci@o~Mv?+Gufm9pt;g(!odG8limG-7w$nlXd&T7Z%Q z-m-$Q4SqZoOATW_rB5oHI5>PJhzW>|CIodv4&vn10MpX~aaz5^e0coFSOE@+@`J4{ z>hB*hCq{--@$V+Vh3qIb4LN|dlb^0P0(1kjm1mWuGnTz^6SybVclU9Yb_bXRZ2>Nh zhC*MAO#&suj~e>ZF9hIOeJO+L%DY-!8IHikXDQaGDgA+K_|AlKJZ;=oimDm6Z$!p_ zy&D$Zu^t;Kp$dw6uH>jIKEwo+FTUSE9bvQQ0irSfzIe{^j?qS-pF^y$%khwGErrmTW!2zc#7ju zAyDjS6NdZO1$M+>#R+Zq2G+n^tqQPLE>+vloEW~+?PE9@W#B4>H4HvkwBLxy@xMyl zx6|soj(U4eup*$p>1mRkmE-7d3K*c^r63;}l{-$>7h(alCcrD=3Anin;oKQ;1g@mb zduJ|(&|GljaaAawQFSck_H(n(!FK1d%2OOR23#fqwLL{U6lR=R%s1BZPY&jtVnCX5 z$hlH>6>ioNZR&yWaVPXwMX23M*=8qYSH?90s&)g_hi-qU-K3MbTTLLb;2wrzo)&cUMo*L^X&66rwIPtZGE)Alyz4Nl)YA2_Z!;kxb|#+Vyf z_~o#60{_*4cb`w{kN?d3%I${%qMjGv()n~hWMb$Z72t0v$<)?q(h$cublM~KR!Ba` zA!HUM&9;g+9uxqW6!wIiD38jF^};zhaooDbew%=u*bZUl!bMbjWtiJ+Txo&o&`V4& zqmikJvL5?ns~IbQP8R9;iKjcFvWuHIubBjJ@E)2IM}BlHC?yyCO(Ff~2pu2qZ|neL z%!4=`#?63}@Ii-8T7_I>{2Tdi31eTQs~%hkmoFn_D`wDr(rhkIlFHE;vHV)JF62IW zHGxTPc(Rq`HD*UiHFCA37U(N1f$#U$)RnziVH_GE;bhNvFUlg)bPab zt2JEmaP^&3yt-4-aq^hek*8=0Fq#tP!(iLJI+sxMGq+qIKzON1C-v@k*Cu6DKTayj z4{ugzIdXm6jUjSv55eL-Ay`v$(C9GbCDgbGeWhBbG%ljb{qE-a1e)BDw3pkW_KjJ- zIe_AT1s-zVGIN8TZ8aQ&f|zdZzG9sO;?2O0+z?r;;+O>+@_-dV%wMy2Y%vL2g@bf09jF z3GslWfa0o7+<%^qKTX~-1wo8#GK{%hL+bhnBh-E%{?49ADFYj5cc+BP#zFBgvQ_hK zg65SB_R5w4=}kw=o1b;m+j*XVjdZ_}w(}TCbFYzbxbuMe1Eeq`!zbwM&+fGDDh@-K zw%1=y*?hHk%w_dk#;UR;Kc~pj0+Tm$u5n(;4G5^S*5*4tQ3cPONWeEnUNMHzEk+*y zaa?Zl@)BL0-Rq<`BmME8sfz`RVO9~Rh-~98JLNaYs1MIbmaF$N#DT(v$2Ug zA?jo>Em%M|Bz}KM!YeZdC-${z{gfZ)F#->~O2tqJ-Axg-c*DC=r ziuf-#d`-p)tx*y1YTmwB8&H(h=g}`D-H4!oi+QuDzBgS?s9k=o8c5S8`x-WnxeFhR zvpWiP3B+|VQC)fhTm+Y0zMGARQA=9p4~0@qJeN?ty;j=ke_^KC$Vp_Jr(*7@H=K!R`s) zT0{WpkiU^cqCg2C`2K`Z5BvuQ&X&MJ6x){p?8g6X4B0s9`e1JUw@b`xRQI}YUfe#& z1_{KaizoV^G~F>-y)MkY>7n%E8tMeC>{Z@vv&v^@!|7Pe0MZ}+QxrHqE>C^|Qv(WcF@S4~EPn?@hZVSdQ9T6J!crnS;1?2Zi#1#|X4k zPdw`7K{d$rkaEsgIlu3WV#aTxB;U+Rr2^KS4#ZzFFr%x^&GP}Ai6R<_`E23)u}c`| z41=#e-$-XkK*(aW1lz4m!SQxNyYL?TtgZ1<=95))!hpQjSm^F#s=smxS>Q24&Zi6J zj(;Z!D1X+^haP0o{Cz2ArS%#jpY!lV%8iuijPdY-=WeDH$ngI&RX62Bk%og>J&#d; z)il3%W`#5M&8Rs_Gn5-Uz&Hznv4F%4mEe^@xqVOWuh$v`lI?O)Je^zfp?gx(>d?80 zdE_v{`jM&l>4(I?Gk}%F0|>5W{xC;TwMZ%bt@qNdXpz2{Sdi?M*u!ZbH*VcAAay&K zggNM|y>!0(!Q^YKtx0w0{O16@MESk~7b#gscw!}K@8xzWUlBQv)ub8s$AuEJ0>ACN zn}zSUYhjM#iQg@OvVgjSu{t$rCzP^c3{5?z|-cQRu7ka zMFw=rk=W3K9SEZb)EP>>?~2*BdfvXwI;arXBjW=|qqAy!KvwYQMl_whURxcDJbu1U ziW;U&yXE5pQYx(`dyXa;jzb^x)mroynkybY`75-^>EJi*Zi~s7UGT_l-iqBdZ4JhA67v9ULR*E|GqA&*~1vN zj5q1XN%&w~_Aa-X(e(TCA>nbR%e&~LfEV(Zl9uWEo9XP~yvg-|T8VaZ7Z*kEign=R zl?m1?T%jw}^-6g3Se3OOUJIqH5h=8kRF=As#;3;YwxicR{APUtQY_XLK1OM`NFKke zIQk)Gd3;eBb;2s+=qTN9duhvZj>iLMcO8O(5T%=DvJ2{rR>mMo%5XVr=R2)S$Q86LqM(0h%^T z2TnxpJ0P&a`lIIUrd>5GlU#CLYN8I3f2p?wwk5h?7e@;&1&@t7(4}b0x~5nY$<1ko zChC$d;sq?-IET=;UOodH!73p~1r(CfC4&SK>eql*>m-NMMi-t0`!zlXT@v{`RxDCX zRLHzkj5w7qx~Iq!Jz$YFu~(9CYTD0&RUw~jeb4uJac#7M!s(=<^j3Vcv0#4wkLO|` z$u|M#i=<(lh6Gi4zT>GE&w(ab8{IDau+;pHNbQR8Ker~3X4pQ9!OmlW`I)(5YeDE5 z?CXr$7f$qYgiaxf+LZ;8hPE%6?`zowN*%|=7vO-UHW(chZ%t1-&SFm1sqM~ra5&|5 z;icy>pg#tkjrP7i`l3rq4p&rt_r4+IA__e9!srKPjJ(vGRQV19`RdXTU-?N*Vygjq zYw|3bC`3H-?d+jPoyy5Krs}iopJrKVwF!Y~+3q?YsN0eh%T=^~A&vuqc#W{-Kw01> z@U^(&Q(d;rs}x>2Bcw@mepcKHb(I%`O7aW1)lDcl4EfYlX9o8^FxEVtX~q4L{X*Ea zs=>xC92=k}1&nX}c33qoMZL<>B2hnZ*Y zfD9eAuEX8nN{YExCZ$5QQ1QypJ94u!#I8^IY=Q53Z1jx=Z8)LW%#8zFgXA3cIw4TB z|4TjD&E9~t6eEzTYYrSx3#W6|RL|s>o7eRgqLjrB1{CF=5s(q{Tb>CQONINsytj}svZn1Wpj&A2~+co3yr zJZ=4qiavKG;k#R6y4&o8e&HCYq{eSmf7cbRv`wU(R*N74${V#Os6Ow&x{BUs~Cao8YJ`)M-1E<{}|=5$MqH zx2sFwz3xd4zI%Bv<3d9)KjNXG=a(kxzDd?zTiyN3!+kaNFFX!d7JrskRx^!LCHLT& z>^C~54`y6Zs&U=hL?m_H{P>Pv)};SSlalhTk6fs3Tz2519`N?n$#qWd)&Ko-x^L2J z^Gj;bqFlPy%biIfA^Tm2VmQ(HNd~DIzy)ZQbk6T^7@G#Kp!%u4G$`55Ve zW(rxEV!B6?ucLa1^asD-%n*pYo5x;?#S+J$UkaNw2cL zzI!rs?YvtS12lrxLFGG~g1CUPkS6`T#-!*AcDyA}5}y}aZ5Mn-F^w(S zg}$ouF!zscPkk|{t~^*o6SZoD2FmS3UM>fUg2RL(edaRYRh4em-XwbKr14ykSl#~l zcO6PK+>r~u#(ymcKW&U|U2<>_ z%G49&ad-ME+wER*5aypC8&y?yviKP8paW*vnsaU+?^ZrLC{|=XOocU_VpIwLI{pP@ z()EZvub}w1s!M<0*x}K!>ZaMCS22UlU^K7f!FP1voi9;MAoV!?a&8M*af;iQ7a9nb zK>2hleOm?)e82eNbTI?jLPN+S#dCRt9FgO&#;=p&0tJSkp{1_=>q3rhn)1l!@SmB{ z$2k=3L0NAaIKy|joi10jMm}0xvh;~Ozo;SwO{W^7^JXTJk#iL?r0U_2&>POWZ1?t* ziCw-5JWd{VcZoe{M~(b2)15H6v`@i~fkpH+DaO0xabACCP&1p}=gJG_yUqU`)SCa) z?KIqWM5W%_z0}tWbX;)hKNt_r0>=G}#crW55pUs~_Y@rHyMfsY4rW@Q1F1{(x*6Z= z=JdL<8@Japq|&G)TD>B*5B z>ylg5t$SNX*CIy<7t9AcOX>VVrGMrFp-0{MtbxI1TXgdnV*?NSTq`0zrKRprwf-zM z{k3NO7;^e(C32IjM&}6y@@vMu%bBgW&snXl3~*BJ+Z1p`9xM>x^Q=@8NU2Yl*A=2P zENzqIQVnIgg(|wT<(iORvU>~5nNJgabb77yQROhOK;?GdAdkawIVl_>s&lx3?9*1E zlVDG_hAHw7I#${3+{*P*3f}qSZ+7F|U(B|2)C&FgvSw`r&HyiQ5bzy*KpK8_E#a%3 z253kNC=lq1v)2-z@TU9xwH^-CLJsNAAZi(ooP^`+_J?xwP?=S~OU_PCR1XX`8UsiR z+RBU+15b&}-)7^Kks*%p@jtq19M)@xY>C^bFxk-j`{-1%siMBVt(6;Tq&Zx_Ilo{& z&!69oIJINP;ctij>zAh#v3_;3R`b)&G`9qT^h`22*w#kFVmH_SffxoCk?8$2lGCf9 zk?ZH*hgMup&MyOG^i;fsif zXMv%2m~9?|0Z|!~mFJtRoW0JHLjequzzb<*;>CS+mEjjVJj+QMG~sd|QyJN1dfi0e za*RNdrQiNPPB`b%mc|5*-V)AuFzwRBokv5Mo0q*26-mMF@yi&DxV%qyLeR*iYvEk5 z7N(~XV^l>Af?zKgS+SLs+Z7!>=X%&v^dtv_buB(dfe|5F9;-{K5ylfsa$Gke z)4tZ~or$IjjBj&ei;=ttyR=iNL@-dbvHPvaCVs|LZa^5uGWOzI0BbwaBg~|0a{<0A zCC~Nq6twMY$$5VP`fda&Vezt(N?f?H7xr3ep2d9oocR5Wu(bbXv<;D)xbq}LYZ#HZ zfkaPvX~AQ@#cqhqNb!{*!sv+>gs2}aX{^$GjOEPtKKO&#^=PpP&N5@=01NHRf`U)E zrfg+TsdPwRUU5%|tA4by_x=Xh*jm)DdM3?oOu!l&G;SVvk>ptlR7N}yBb9X)*687j zW!Wa6{o?sN10^A+xK^s>b6%&nqKVSC08gDdq(~n)OWk>LR7}h2^w&bcnwfhd47TI> zMWSr9cDivnfjs$UY-vl`x8(X&ex4nt*TO#v;~G+siQQEg#!@BzP;W znDX{Zz9?2c?KtNx@!&@OIs2Q|Tq$9Tq;fCA2)3YUE+FPrrc^zO1dqUkQ0|^d5aKKs zk@vRFD3DOp47*q`Z%)54(kopK3BPf#)AftzXV_b|e*0TgIvG}1x!gIg7>Ln$ow(`-gBj?m{8#GE;F^{VrIMS$#?S8i zCK?vdZ`8J0#tn;^XVdgcR|zjYx$}uwdRhkQx7%e#ZSe?Ge|dVJv6bps;=6LqZ<;#< zrZV)Y-(Bg_j4HKAP$Bsh7uIW!&P}BJ`h@Q72wf1Ou;?oOL~>PuZbztjgnW;3b9ZMr z|AjF8wqxX+{uY)I-b)_O8UOhh91*N7+U_)$roqlNAybyYR5ywsP+#0XI}KKak&9Nl zXfqG0wQ%W{9r>Gd@(rD9ut_^MI;xs3M+Sx&540zXZmjXeW1Gv*Anr%W#yym~UX9V1 ztF0TkniT2TBvmetL+CW#^3nstS9JBYh3iX?+q?|k&AYEvJo{`d2)k?M9{{CIj82V(qBkk2^V+E@zsYWxh zWca5WLdG;jzSBCR<^evUEbY0gjqw$sHwW=ZUq5`luGFSr+?Y)kxcYp-Vk*2g6CL!z=H=mdNiB+O8Za*u5*aw79XFS^z-heZ z+dU{I5t&Al*BBeA^!cOgC$!+5=f;dAA^M#g0ks?|9NCr-%7g$orZy`*`p(CfOq_z~ zl7-1m&9Ua<_UDpsKuq$@_h{Fy$T1GKClQ(M`Eyt&QO6b)pgF<%c$1WzB31%P=~@}V zPhEXflFCH1HJllBNrLAx{kTFxa{NP_*HYyz>ndwsN;Lj>FdeI0e-y4F1r|0R+JP_1 zRD>Yv>`5$Ggl0$0D_$d7@;<+Q6q}=fHRJs)DBM26vml`A1+4})L~}hv_LKO$47STk-w!K)})R-93CMC*6FgfmnC}z^7st7_0XQcdP zWz9P~m}iWMZw)8hsAN;_44P4CcjnfaCnlv!0d3XFtvnpZq{yHb=iYd|6Y3cJV(+e- z${Ja)A>QVohUCOYj{Qcdd4ATub?1 zSw)4pu%~gku2VNvUNTNX)jbB72hDKiDR>w93zDu4FK z+bOi2sn%>a$)S7bskJ@;(wNGUds0fGo3q`$`41^O*BPOA%JXK4#>7d3$^}YOZX_-N zHyC~TczVOD`q%Q4gOYEbXEnb}5$;I099T&g{T?_<-JX>`U5^c8@N2Moi7hh<@8ez6 zVeFh^_~JLSzTw$x?~*o`KE3`cJK~2{=vY^BIScVMzR7l2uyXaD+Hh73r`3fmm@4)? z((zh*EM0|hDn`?q&#@%a=2M>w45;B@^6IcWiFW>i&imSiGlupVbUC)@nmgF$T5i18 z$H0lZlUM6?BQ*PmlRAzFU0HxoBynsaR*aa9JW*=Z*J7SGvG#FqY-@9!)(&NGp(qMxu!EUz z)g!{>3Ohv{k1AfK6=nN@l!a_;oJa?R0sUGat!(pF%l)k@)wgIvEOH){({4;vx**4| z*~xx!b>g!lKOy<2m82JMcyu8d9`dN593q6oRQ?4%;Y zBHIFkzw*kR?_qXN^Ai`b!3x`y&zD>Ro;K_oPw;M*#C+Hej(W5&__0~Uv|Y^7GFHwW z25%C;F(3zZo3_s}BA-#my7c4zea8@(+hl+d4Vi8+gxHf%Lj=Bv85!KS8OOk`jBZ8` zu05}@6fYAjD%ixyW4uOzzNtqm^=VQAR%ch7t84Y_!E?3%;do!6o4g>3j%9={ z+?bCJkP-L^D6=Leeb1fL-ISsm)FmsNcoN0D;ggwG;st})?V6Sv=hHaxHq6~@{vNEy z7T@|4FmUp2{b3qHFEq-ae!5OP9t=ytv^vuhUl#)BcHVY+2%3}SK zFd*&`RgG*7nt;R&ZJt;Q?0Bu_sG#DQ%5to*JC@0?ixT#1e%HMe4&umXWb7MId2uGd_{+s5}SLypF{@A})}+fv%k-dsmVRdyXuGq9Dy#!Hf zVBH_GeC4xYfrq9Ca#G%k;BgeUidr=|a~lKmLWW;O{{t{{z_j`ET=GXZv=6Fe5;_~) z=n_oXbMslg=AZU}``OPjYwOiQ z4gV~LX!K|0kU1zD&T301BlCz+(L5V^#VOIRetG*C`R5|4x-$`uh0&8x+(ko`uh3Ew;hw;5V0G2#AaAil$bS7> zQ|=d>uOA{>{_v%aWnk}W1lT!S1(|>SFkmVHW}0XQfJkJt9fZAm;&5%i@Vx!FI}`AF zUN8)WltW)(-=K_q7RTdwFs)r8V7co?44s|%twP3_kUBur1gHxX$pBp`o5|(GiL#jF zBRtFBPdEFoG50ro!*Rc%7DUh$eB}#l4iI@8(Q)M+m@|+9-+Vm6Q)RJ#{|?L!dvTnu z_EN{dKwLLtSpt=x9boG_w1Avp893Yuo0YA}W@G7+lly~b@Rj9SdU*z;<1<#o9(6Nb zR%dXu5r&{__~_kcFE~OE{e5C6_}FPXQ6+@V5{Zz4L`R~50U_A(UWwuAr{mx|Y^hRr z#=fTlJ+w9VD6trSSsI-GB1+k5M8mzhi$F%5cRZ_RwbPopm(RAUb3)*|MFr-Fx}>y?Dcmi=Sob>rr#&*m#&LHzwLMg&?LYV`c7}SLx(yD>X$I zTId{{MMj43aP7=wqB(A*;4JY`WylFi6fH7`dQv)dI}W#Z>%A_kFYe;?{BPCa7o$_Q zUnG5&Z?iV$9G1e1ky11oQIt2zq8xIkt3`{TU06e|5QGXj5`7}^Ipl=EHOrTj$d!W% z`yf{tCzt&mQm)e25+W}jprO;PUqMv>D+6%!QGjmo*N(WHNcq<7xNFz*cPZWo1H@P& z$Gek?z)UQEwq8L8(_%d;fpJi8LFMUfp%N`mWtd=dN$$&lTF5r6hSLI6+8IeTdEYm` zF>iP1jX2e?rr6KtkgJ3&+ox*yOfED7;*t(?zq@)G+iT3Ab49N;9uUTrf z^t22B;N7S?6~byd7Fz9?IZN@-E3h1 z37(#o8#0@u+SVEIPa!MLm@&Q*VIqX7R-&8&iyJ0fXYnpik@is{;1al_9nG$0uybh}X zH@=v`0Adm%Vsvv+R#|4U6FQ)sDI&QpJe;xBm;T75XQaAg(?E{x9xeH;SG#$eh)Iw_ z0JC|B_ztD1c<4MsB8|WD+MNrdg?+_MzpAVLjk&_8rS0+_{)E^Iuvq8h$2ToPl3%Yi z%ec~oei>)GGVp1e5F$MQzO6LdQu9sIx%qMF$`q1c&#!y=qjGVB)=o+S>;qi@QEj`Y z0mophC-7p#gbL{CpTS_B-5;#Zlg=;upc!hv2*G@AnBV$2M=7VycKuG(Na07X?aE)Z zHYYMnf1p0<{hebz)?&gi%MTN)($2|oj#N*I*mq}k6}&idrX)(b_Zex-hS zDZ-dBFD(e^wQB9?w<=!yFk{O422^+oo9iIQ`}0sZYQX)7ijO^FBu0pK9lhcSCLL%= zmX|?x$eq(3te2xgF+~BZdtq1b)C%QHNQpaS<=InEWWVMNezsDqPr0&WBk~v@FMX(KpV=0?lKZh!Qt@E0+dUC3P>l)iYhfd`2K@|AXljURQ(g&UI*hJ<7f z3rdwFq%|vZT?+i7m=&E2SKK!UV8K%?VDcUat_xLsPm~-ZZAwRE55s;0R+Q@5FmC7B z3n$AS&z!maRiX_H6ukeRrd$1?l#(jCgIq@#3Qe<&lUrl6_MLpr$<~db*Rq-uSpgU8 zH>O^*K=NTkU6QmlYBD&9S6adO; z=9+Z~j4k%u*6u8amNhC6$w^dTvYVggU3%id2ayFB8-KFOkZ8Xj@BE+n%q*U-8q^Y$ z=jHWHKLx#@Y;ucB)uL@Ol1(}HC^DY!1zvrK?28Zc7&1x>=yrG!fUn;@Yx1VH>;KX))9Er-ibD@^}J{^Dc6r zlD=Mht3`IE*UbqH)sMGtP_f5K$QagTg}A?IU?+MN!s>r%enOR`65?#SKfPdK_^7wW z@=>N8+Zv@+$@m7iO}O&)(;foza*t8yBaB zC@wJvToPXkhvH_&<5(#wURJ+0oh;S=t7!Bz{0=`rbi37XsLnI@F&?~KBDGIL^be6L zo(&|CF_JPvDs5FaKWBI(n3mqoc88Ktp|w9%(Qdd~v;JI^MTBwkX3TlCu7cK$iMKiX zI%)sQfR6KS&u~j_xz9uGnGkkcMVA_ffdBb~is=~g_^ZhF7&4TPZiQN9Ii6+uY-wo- z*TJCh^8Ej}r+^-92e_XJvM#)9oi0sP`w68XCESZ;eO>Ed+W&P_ANO^Uh=2Q^+o7c!3+(%sha%P(lRJKgO*%RrwhsJ}jr|My@K@~8nYU5D zwg&uD_@dK!h&wuwr3V(BEhr*$}KW zMf)nh<8N>NAHQ_<%&(RI$FHOVUf?lr=Ok+J2Q2WQ<>C3?Nr43SE*52QAt_=9J>!n4 zh_d}#IFJ2QK<^6mc!~frq>Tn#%mhi2J+A$qN}9hPbk6HjXZtWT^vVJNhx~)fSdhf# zHvURu($X?7ne6KhlKE&ge?Awv3xVR}BT%Jmza(}1w@WsGrw;I- z`1P2damqhM_D))qI}Nt+$3H%Mh8Y0s>K+aN;5owkmpHK2KOU^r&#?k@HYD(3=)e5t z5kVH95&!6u9suw~8NS^bbp-+tzm4GJzgFDDAv;5PfMERLi~Tmczfp=4DbAfap=h;` zwZ^TZ$ULVK3qqPS+aZef<4?Ef%!6af*#Y(Tzy5qiTH?N^6=5OMPDJbQpVMNion|%eE3j7Ql5C^d7%V)^{BwNDbaN*G|RnP_)+#(OT(|_4o z$GB)9HgxRj9ItCe8Zg9tXF@`nfBCk-#(hlHJILl~CsBpSk%A=Ya&*7^%V>X~v44jZ zB1cR)UCHDb^e>|Z{GW@jh+n85`#xu1)g08ob#(@eDDf9}#WlBdWp{6UykvcfWg<%& zkFJ+rx)A^fjGa)akI4xHy$sL)#pIAElA~Mf1S|2t?C{pq!vsYB#q@xe}q|BKV{e_F-AdLNoT<461XKTPPU6%=Pwkr?RzWDTb~eAmw$xFY|_luvg!FPuRF z6tj`r@RZ4X*StHk;E31~hj1_*UX%Y)R`oxAs%*E8BfNx)k)gA7XC7mB;@|O!bAPeZ z;-N^`YjUS#;al6v#^6|NtO#)Y0Cab6fdi(sZ#}qK63{nSp_HOp%`RuSO0cA%wWBM_%;Imi=I+r?Xet1f?n8+YvW%_{QYopv(yV@_PO z9km9yE~ogCfSbY_5nHEx?N7ntI!&~j`6*}YJnAr%dtGP7nFud$2l~no(_^y&1*Rjs zI>c5AEloe2=XQp!EB8HRqimbI;cwSmA;B8}YzTmb@9FL@^aISsx%Z9=r5mEB8rm8mC>{=0AR~W>t*8-swg6S2|bd2wu%3jeKCXN=yaqw;0hNSsgc+yq?lZv>92g89L|75qp=F=4 zYyQ*6^x}Pwyeu(aABh0H1Py^Pz7&@bFke8#8n<$7Zgx1%i9u8J>SZRy@B7AL1Uz5T znGg;Rn0Lf`iMAg;ta&StN<0zIx=IiZ>tK*dBl46bqQ7yDbAL&8{u1G4a`*51_&<&) zy@V*rASC9wBNNtZkR8fp$0otpa^fJ z<|0XVS~^EGLyP>Q2id8aQlv2d@0+0xTDsg`U!CTCtp_b!vNRI&Be~=irHkoVrX~{0 z3nEX8pn$5xx4noeoq@5sE&r8$=GQ)$uLY(iqX)vFV-Hvq)$_0CQz{&S1B;yn{@08k zyKEj{;L}f`gGMiz3toRY!0p+mMmk6u*WgMY^4%;eqvc4+@$;bfq}{cH zl>kjqwFq64%%pk1w_2C32qRNyW32s+!cIM?zb@__#05U@xRtWVPKo`}Jd@4>Bg+Z78D$07fNch7=%T#9Jj?w(vJfF-S8WzPiu;0vSKiB4D;q<1aU*Eba^}!dZ-na7RWPR60&lme7QSIdONE z^om^163fi%WR=IXdm}(+z>JTQtVWLp*cYSqEHQ;c*0^H}QQBiN22^~$a0Wkg-R==v zL#`9>oHYJh4}rhViYgv}tmQI0#HbNy(p+oV|JVY&wQ)55wmF$m)( zLpok={LkE9XT|dS9IqtNMgd*w#Nrz`orlKe7flMjjczpuAEjGVOX81w(l;Y6e;k4Q!3?2J{}_ zNK-v@H&&ZihTDlP-03b~ZrCfo`=)7hC!y%<>Jb?<F6C zoSFs~s3{VtAdtB`l@W93z2?1rk0?ecJW=&BGY z&k{OVW4i)v3BYm~4J>7z^SNq-fjeCb7mF?v-gr17DP)DJRG`hEWY!L%8=x_%QF2MO z2dY|%Fn1fLT$cJx)Q8c0>S4J3nuwnNye?DUx13RsDC z#_`m1hR9m#NH$kvGU2#~b^t{ZFo}?Ey!V1Nq~>@z%Xol-h4z$&89zCqe|u)DL@A$% zCF4wA3t$93$zFi$t@k?SNS@8G7|F9qas)4}V`3R8^MVb~KD?)DF+})-S@7tx>HGO6 zG|&XlP7zpy-FSgJSG)^FjJyO$|G%sQAn^7U4A;cD^UY++Q~7N2!$ijkvGW`oQ+}}* zBIWKT=TTjmL&u$_v}S?nSI!G}2R@-%qUBzDfoog_49@CETl+`e^?yAzH(oP{t=%|t zxz;9cA6h!NA2OA~75j$5nk8R9Qm$M;$Q8{Bwf^A2^lKXBPgd#Qp2PE~A&=VuhSB|P zuR;`I!y0^QM%glk4bNu zbhkwo=ppRIhSRsU4Z67NQg$bU>cK%^?SCL(-#pu^-rwT#+(G&2DwElZoy$9nX%*ig zh~K7pCHvK}8(GT~5iI>QwJpe?%q+OI5D-`AJ$bI23qW1_RPURN>lv+G|6~0Dz`iBJ zNf@F)SDWK*q);pD)z9DydSvUFqlg{dx5`A?AZ5YNH2^*X&fUQ%S6!AQ>52vI(q0($5l)K_@9|v@rRfWV^g6?dzq5&+9iCd8X4Uh%L05$K1y!Yplp>buWc<)$k2Ffir zKp;i%OV;=}@|C{6W*D8;JT!lf`G8qi9tX4|nEu;$ElxwZw^*&!UFT!Q$|EJzNA&Fx zbEY7J6))>~jDW(A`MN}~!p!+(%YFZ$fmnTkdt$+DL3Xtupkl5N$p!tbzsRdn=6$~{ zC<)_=#U!~E|Y|=4b6M<35nI|`G{$&p8!!}zU>aUE^ z!rP0cQna(lfoGe@Jg4n+@j=EbJ+U3i-Tt=~u(1S1UGu8K+g@e$vr=-9f8s|>YnS*P zIODpQ6&~H+dM#d1bg3OOJcY^3*UU|ZbV(#BCs%;8_BZ*na8la3BewVbmIU6`fH|WV=Exe(lYi zcJJ17QR-89{s7kcB8tVil=thv%lbB~yGgOnM3I~0c#Pe6HkovPITr2Nw}Ar8%n>T( ziKM8z1D4m)ZoRT>%GFb%-B=?XHshwD%#M|iD0a8HjnTe6%YblvYMv~pk$m}y3$5~z z@c~hyTfel;JWenga}U2Hwp}F`f8UZ5W(ZGSp#(ZW!H8Y5$OW<@0YAQfeCymx?7@Y3 zq|o|eSnWK90l>BWdGRl@U=1Ax@;6?S)YsPmoxu&1-@##hNnDhwEPE8o~QLb;vAX?-C-@_s(x^b}sgP9$PE|L9PY&+x@`RgbL$7%Vz<56{c8*=P+ zORR7@2UCTc@`>k6KVGk60*J0&l|H-Y3Y2pN6@c`;Ko=nXRMq(FCWcX|elj`n13Q6qI}Kh32+uE_BSfNmBhX z@?xmaJFS(@#I4N-qOKpj`I|N!$C$$)A(&0MJhjqBPZe**J2)4rfw&{QDnIOEj(P<5 zB`T%u&;g<6dj?5)07(09*2!jtI$_eWFZgmllfx;G3DY>u zLU699+S?)$=aV7RP7>UA@jCO^bB?X}0WRPLBwl1tf~Pa0lKJaiuWE=l9r$*9<#lCI zrN61564J^yb8M>4!rv4_4Wf%7be{Hm(8r<*x- zc;6N!MgAG zi2VNGv`arDc$;Wr#I{(Sr06BDcS{2xg@!oz_dhmNgxK= zU_do4_RLKqqlT8!!^LxS!&oQ`TAT{%t5_e!8O+2!%O9MVtorL&!cNx8mpK1hRu-9H zKj`%Mqc#p%cd=Pz?=gXZ@X6yh&^!snEY_}N2z#>1)H&8RB(+aT-)|uIa(a#Trp()5 zoq};_J%4!^^Vo7Zl$7y2hcE(u-cigI4`ll?BM~uT@^u4ebxQ<2d?|;Ok;1Ez45avy zx|}KIQlYeBS{~e6T$lr7;JOpKf%-Fx=ox0Pg_`v4x9lgfu;4`3;{)g?>>*K1e(%Tn z;Vx`_m$T35aSz71$b3Vc?)^W^VQVijsEcJy^5*5pOsFgUdx2qgaI{`mux9N^`i`Z; zE@q!wxCLYXC)_J~EN?W~9E{KlH)CQu94Y7}wOo^Se*#-}-)GsYt3y+x{!&eN3L`m) z(fyVf;8Jx(rVgtbvTgWh;J>AXPqz%?dO^8)CUTJn@vS<5+5?O8{M7;NEY7AqHcXVJ z`$_eDUa>Zrz=5@=FK3yu!>+(TG;d%EQgyh1AYUd5%B=n1k}AMF(~2mg87u zUajMPXQU&74&v#0&?RyqLP{_TC=UV#tR>wyu8}c?NtE88*4N2w+Fko`>Z3ymI7BvR z=sRsB_0phr7VUy*gEiFy1=|#l>h`3^^C8Acu7TD3m-BDfYHrceeX_eUPr1-cJ`>RR zc;4cs9~*RXG$5_a0ZVm0k};_FL3-z8Ondc|O_7CAqO||Ayn!iWer=Y<@a@5kahdN4 z3sr5Z!+4;jO%Jxjv~w7=rXs@;KmxAl`=rVp7OmAH$CE#9ZmCL~M#~00_xB2DvTEFKN%%aW1ZFhyTS@$2q zL^#-I+BDnc7(xE+A3gZ*u)x`tHm6NR!KTT}1wG{0^L^~Nk6eQDKCS)w8F=wMc5<211Rr#!C>Ljx)2d_M^f!At|hg;Zz6up%SHzNM$2K z(F69)VcfEsIp^ye?@cPtW^Lfu`+wMb>!>LE?_F31hZsPqp`|4TkdhMVR*;ZVq(h`b zx?8$L2?Yd1lt$?qS`bte1nCkHNs*2@_YjJTK0e>?S?9dxUB9)S<@1l;_x;(Qo!7qh zp7vQIx_9DF&#>Pz^0{#kE zMN$YT?e_voix&v^aVobx$^()UwLi~B+WT=1a{gv7$pQ|`sBzJ>5dh?zsFlFo-I=w& zX0xWyA#5@&-#sS|i-}mPP%Amo%Fnb*4KJhG0yx6d7%b){6kXevG2Jx`otdA#i@8gP(9D%Ysm8B4a) zY8A=oJm+H)a2>f)Z10}8t5@QuBo?84!gRwyd*j9<2Dt=SWN3Z%wR3JxX8Kk)Z?JEM zxVe`$tty@Ex2URVKVR zGb`l!vB=L#T}U`~h%#`=S(;#V*~=82I&PQ%sAR*Do^>AunIix;@AoSF9C2IWzz>i* zYr#t&``>MS5)1sENk7N}a{TaiP-_c8X{<7vTzu# zVaxrjJy$`GJU~I3TAy}mezlDB>t?!y6Q3L0o0qqv2?iqb`DZ@US!!RbD5uQ-c8aFL zs1^h*$D$Xf&D_KaOWyQiLQ6tDb6C9)eKq7_CFtRIbJzol#)WSKb9IB?X>I#a7kj1_ zq4B&Hg1E@62)0&OT^{z)6P$Qsi87plufvdi$x7oG-l6TgTWVlC_PFUqZ0$t#Hm9ks z4_KbJ8z8P&S^V&XF)wYwNb{3x)iv6}n#Dc06CSWp#`W_z0$@_^-%5EP&3A27R23oY3u*yzH^vLEU8lUFFO~i5BE2tf zo5aKg;c45_#pkxVQ++LWvp-*x8f~qYmH*+R=g0a6>UWhxZl&0#t}G~{&VLKc z`}o)DgD`EBnao`--l4EO!Ml~cU+@)zbp3Dd9d6+iN~j1xFHBlvnC{xhf4 zB!w=yOa#}Bt0C_q19wH_v@a%pI`M-DImNOUrBgoLHye6+=iZx>9aTwWkJJa zW_;63eMIz4)YfaUAFpq8P;q_$ZmXN?`YiJ(@lFCKSS2ul++HX%LZGDzF8n#G_`O0O z!;HCKp=W+SH#UK~B*ILs;-L*HRqo+7{Ysq3{MpvD~ zArzs1zBo|SzNTd?j6Yqu7Mr7@7epPjeE)Nn)Xa2S-4I+jLg5om0_K<6vn(ri?uV^L)2r;Hb`R17P3_^d%2Hk zGfLMk_dQkUVDAededZz#r%#nq6@7!tbycrTSzLS)dhB%Z-*il0Q4CLnrJ==iB%XgI zsda18vdN`z?JC=&b4HY3U_+S-ut#|cagRZht>u)4@9r7uDBXNp9vq%B)A2g9_&P%z z_w+4xmG~e>cYU4yyMdN1y6t6j4c7&(uQib*4ZoZw+oG(@f_KJyr-6(|*xWS&MK1B_ z-4>vTRdN)+7m(0cW%e=)Oml2s4dk<~>rRo=4wg>@N*UsmLtk<;-!F$25fqXYOJ!R0 znD}Lhc3BMMb==n}i(m!&#>3@%vpP=e%Ndc)peN)}JGv$nu(hIbUdgRTq-bh^2iaWTw|#*5Ud2 ziS9~;{u9Q8Wo)ALEfcSn<6piMm<-4^^Qj?I( z9s-}He3OAK*UQV4B|7-?zJ=?vIeB&Sg@7(Z{-g^HG?>PWjaPP!_^S?X^=B`^19i|8 zYALeQ&}@Ekt}6;1mLwZ@A$zPg|0=Ii1J;VwKz`FTl}{$C6&1G~RvWx;(G;|r#1zmM zndoeO(o-FAkYZrran4LD(^|9{B)&t7Wq-FuD?WQro2s29yd(-(X)~E{%J>Fl80?v= zgUfSuuYxPFd=$Wuex{>L)LnEx&Dsc=4Oi-Tj(PDnnJ&z-L%_=X38VPU-pl zbo8-R=LpE{#^Z!gCRFRuO0F&23Y7Hr(J=oaf-%;1`bjl=SN2;&%eANW{g{46vc{G5 zp4zI7mmSD7~w zaMV1bN%f(@;ip^DKrh~IYI6y+)+NCd!@;krQFmRTM=1y32`AUO#iwVTA9mK9%h9AO zl$8xCUk`6oe@r7U5qF(D*=(Cv#3aNyz(R`PvB_*%`>XcIKKvTtn*y08m#Jea%<-1m zgp%f?T9?idXWR#UEG##f^R~S0AJm%snUe>QQ`1-&ymuYyR3-h9=uxPGz5Uoj<;t4p zhRMB);Pmz77kAm4Gp!j6%?1bfg4@HEW(pY?1Ff5h=oWcN()0P;Qb|28Bd(DA6sv;h z@Y^okdb%5{K2=k@_8=UorEnRn{q6@maCJ0@W-npab>w3X>KJ>smE`E9+h#c}5m=f3 znO#?N7&2eYPzp)P@?P%=(XsOH1?ktClqjQI58H%zWoCOh7(Y^Hp5>~~v+OIrLeCnU z(}a4_nE9dwYSQ$0)-iRXQ&#q-^)6kP>O5^rdB2Bqe&Zf)GpyLQXSc%h&&Wj@NKI-cZ_-v!1ploih@jC{(= zdM`QI-}&0Nwd5O^f8o=*GdVHq9Rd zjdbRjM-yayHu)GW(xW*UoX(gfr`WfL7GG1c&dk@fFdpEl;xc>}SOq-&A1RZ`(l z(+XE?pG|xhhN5Gv<%dtHl}NoTtof9^6wo9Gg$MS^)wCy(`9oLuWqav_2F%~O9#NtG z_HYUr@(mZ>>$G=rJp(XYRM22_JQ6Q6Z(n|#fvqZ1Sa_zH2(n>Z;A9%SVFAKNOS+ek z>%(i^f2$*OTM`Jx7u1LZ2m9_YO^`Z1ALTWyFTII_wWhda({^{&|Kaocfrs-%lV(mm z6f$iC@fu{7p7-HH?V+YY-Odk)9B}d>uRR1j19NZ@bHuvtfhsF6)HNB5J~TTzi44;$ zojYsSpP%Ym@RpbFS+mhgDTSa(ilut?$sY3MxUJyF`1T)8RB@QK0FqJ8&$R60zQOAx z_i|O5nKNgx-bi~@4O;-lQ`K$4F0SxLw*dWcoR5Y3X|LoD`>*MaEGkSkf5J3O0HhdzTUU+ zBE#5RV%S|}A0uPI5U;nnbjHs^DJZ`_zmwRTs2_BeW43uWKlwUX52PoFpHuAAGA`kDuJetGQZ(V*)#uK|BMR|FE zg*=n9wNI4-8vh(8C7V5hXn>NH(Sa1TBg-6eKXe-EvOrYBl5TY`k-Lq;URXs{RJB3g zK!@kis_&FGqTh#5EWQfIk;v+FMh8{~S6@_Pn4HW9S_ku#{yRk%94+HSWP%jZROC9M zmaabNv+MWlcpZ~%A8RA&6KFI=GjQf7zI7A(nunMb4WX7FLKDOy;Nr%u#1u1o z6CD$&YI`B0@4CkQgsvWfXyH;Y-6DnI3n?Ch`{w5cyL-uE@#0=!y5MswH+!mQR7s!` zL=;I?F-G!JVAa<>jg(f8A3(iSi+MP{@T*8Sx|^u#;b0ADcNrKpHd*n#ZT=D==2K}w ze59D&kYettKT6C-nQF-o^acp-(5u(aWQ_5-m}@g6n7$GIT%3@@8o*2pefGr3_vSem z?x&gid70^;XePc-mWBN0`G#VSX_`i}Na>okKw`Id9;EG4@Mpc@OA1_P zh)uM*6Qj0R2x&eL3^9d6wmgit^Hf8U!ic(qm&!*=w-#qg~ zW1m&0YeP#iH!D%3=sB&1u^jUFEp`7vA10Wt~0@jQEws%03-lpNI?cmpgD2 zM~I}sj?b|s8|~h{4x_c)OoacbGO=ij&&G9tS_&vW_r6C)2Hg0_AaicHzv)PSMiKd~w*4tzs5v2N{H5G2QZ#iA_9NG_j|`x<~k)qhzlB8M~<&xxn;ygLjg%G9h{K$FGRiRkU?IWhdX zQ$`BBF~%$F)M1+uc1u}R)ScR7TBla~UtSD}4A}8X@dJfe&EThgWC$`$wWJi4`|6o_ zF{S5V1q1}~UJh`x;Q-$(_?FGWv!1XHYW?6Mi(1DkW`aae=T8gOl+3W#*rj3>yAlaY z1C60_US8dpH8?8p8Aix_E1ZAKxy5zSDWO?zDgtq*WG5F+_VRK+JeSKEo(8o1%s0g5 zuVV~Oym;H8>r?O!V;YG4R83F%sZ|(EDWacxkf$Y@i&k10Wzpv! zw8E-6)o)@i>7Ko@|sK{9O7QcL)f~hPyq! z>WU`3Hzx~jc;F2yWx)8;f^RyQyKhiUlR$etdw24^w)_2|buwun@=x0uo6i&^tLXso zN1PMtMBG%Rp8pSCu8j*_Gc@~pQfC-tJmAzR;r@+&Yp)nG_?EH{ri!(1@fBsCN8bD4 zrCHghp+2dx?b1r)aH-c7T{V+^O^=d;4Fh0U(X`@%4r z{4CLTiIoNx>Z&6(gewOC$ybLFthjQ6wx%y$jBduAk1vv|J7k` z#hHj-G8yuEEcmuCEHz7%r&$iT^}F?ywEKk>`Khu;Cc&8yqA#$_h6o( z`5w;%-*DjPF_2=u;1G0gGk7@jCQzo*YUQ>lKWXhyO`??ys;hi7Z;RYw8>0RQ;9rkNs> zQwltHQUBxtR+@X_nG3zTI1tQi3#m7wWQ)~g%+8ir$&@t5+vP+0#Oc*@co@?`CD%QJ zEFI`sZHXeIuVUV#4B5m?dYA==0Pu_p-ggD@ASpBNJ|dH&#EF2A#C3dDZ$~_?YYg9& z&#b8YCfU>8cZAR~wg4f?{nW1X-0KjyEP3KJMcKd`7)1o<9BHFccuBjV7eAbOPiELw z4Ku}x)+`tVd?jwlI0KiwrzyII{ZK=@a zuKrq30dd7J9W<%fE0??Rk&65JL@9L6DFRYeP>_m>eqVJqM%*+406TY6lAg=G5*Zlv z6P5)}ZL(37OMD<{} zToHKWmKpywQ+ZPlK(asB#A0L3miGwfWQk(_h3vw^AtdEw*;gMKFnWq)z!6&wJ3L#f za&$?=wcDOEb%ZM}_$cp+il?b>6c9z(R2A4DN?KPZs>*NZEIyZnalGz29niZn#0y)< z4(K9a{z|>CiSdNZTH%!0>Pcb-r`s=Gu#Wu*OK{wlyM5t#)qNjd9^6mQ#vf~2yK zBwq>CHtkp(2dYEP=`%_(7uD7Yk+HKAC=abTBuULr>1>ZnEpy=67j}M z;{FDY+A{zeFW_+MAGjb1F42Lw4fLK4c+If%L?E50(f<1fl1zy*S_z$!eaG_nJf2BQ z8}CDJym1}%Ew33x;0e?(x1}v#w!g^hlOGmlO_nP=lTv*4K9jUvAVXEedvRN;FDLkJ z0GZBrKQa*H<3;kMpStsxY%l$RGD^BdLu7G~EAX>XIB6K&klwD&v#^Ew`k7fH)sa`v zFM@eif}|6EX*DRJp9|Q~u;LG>Q4Zt{YP@iJHd%Y8^nBymN~h)HT6VULm@)@dIL`$> zM1+3y%S%q(0>Pf!9qKC@T5>a4!L?`fk8y3~6bNgKmYn%m`AG~rEO)TNj&^i`si9z8 zN9%!gb1oA4Wv{oSnuGb{jvaSMV){9~2cpkenBG}OT&Rtm(@t|0@RMTC=02mDulJ;! zc^;EJu8Oq&^?nAH5D5P(R>H1cs{YJ0Q0bxGL&S-*TitHec!9j3jTi1&+M6&iSH3Gz zOSnXUZs7;YO$V8$lUKHXFjj@zzuEYz%h@@9uZQQl60Lb``RAKnMizk{X5);phP@7D z9W;f2FXkRt8O;;Npt+<}8l3hMgaKHbL^?>ihyl?y%avQxapUTy_h#GH=P&uKYWknq zLjlcLXaZyoH`=kXpCzclo7GmibwgR%XH#WLY~TDi)L$iwYrmfsPm(b>{ZeOVu9gNr z;jzjCeZa8Fj|}mj#DjBOA^*M=C*yDrSv&V zl)h0Rbh6tgOO%9&Ba%K1WX_T;7MzpuwtT$sc44ce>WFuPB|~t)sa0Wa*y~ldLcmjK z(D>srX7x)Rp~2B#I-hsvZk0Sq@~3(E9=BV+&6RvN5xZoni~v)4A?LAmb^ zdHtcjTQ;uvErKq~v?Id!^DJ-W?EXj;608!K;O!=@_Q4KRducUBdGM7GAPBO$cl+E_ z)k|Nr^t*WK)HOgM6u&D3&6eex6aue{m|s8zM}q<_M!`O)Hg8B6)JRCu)5m{o`PZw=zk$% zkq_y)V|JI&fE8%f)$Db`@Wihj-|bh7g4i2@hQ)78!0(R4dmQ+!?(mFD?yl4i?Sv-X zw8ANRu0G##Paew>ZLTLSwScIuj#4Imz%tmy>{|QR;Ay?02rgRN(gD zF?oCPLb8wY6K8XkM@o^gZNV5_ep*UZ?{a3)MMy%@Puk#*OZ@lJN(!}S5ZA8KuW|aE z1!uw19f)uhVQe5OHJ`q60VEclkfL+2t1CW;aR_1%vbL zJlg3zPrp|rqA$I?iwC0`m*-ISxiQHyl7)S`PL0Hfo7f{Y^APGJek)XOul*cQ-Z!X*K|{9S zqIE$#^5-CgG%Pe+zUz`sS*8jmBru!yRWbypdtMO}47Zm^54j?_IeS2`b z@bt3j@uI>Y7B3Q6^K}`DtpQWfY7)_JLvOZz0gUs^`7<-e(Jv7H|mx{jb!qpyo z`z2FV>YawIdEU@XZWkc#3FASXyY-w*qYkRYT>zWbzYOE2rpO;)BxJP42uu91ES~dJ z2rq1C`FLLUTCJeGS#z*aOmc){V!52EUN9&BPAcj^GT@J+`;CFr4ya33{$#L!q2S?g z(;3<9Kq~LmT71RldL5j`e7cDUzZN;#Xt|EE0@PLv`z0GyZ#0S}wx#-^d%K}v#dA;C zg}KbI1ww!T@h!}M>x+@vq74~0q1U|@xtsKx)(J1=?Lpkh75t6 zamdOI4bwY~{rK*P6jlFnm-Da@E%Sa(_KxgeAeEi*cA14PSg~2OQP{{?%NvG3&Z7a9 z!M9Mf&l3QU!+SnXR~E%MzP|eK8v_dpM>pm>PN(it{o$2}$&UL2!;X|~LllvZ()GR9 z)n5u$APAfLG78yQ&-eRy^z~>)+>lrHX+grAzbf@D?u9cf!We-&<-fzL3Kmmm0;9Gn z9Z@FNHN%3ubR*-j!lIvld-}_PcgKM$xSCjUeq;E=p(9BhfRHcYl{stFBpV;?egjA~ zH2W2iH?XR+?3mWl`6xI_gOX|=Emd#>?|AQnJ!{T--OAkeC>>p*J;0%_k7%pA<6>{y zm$x3>9nl{bdmmnROHC(5Pv-N>a!lhH4ly4w$a+A)$8%BX9^oJ2n+c)L*Z?Esz~zM6 z>_fSKcOkdRo0dR>vLjPZ(WIB>SLwEu7ISMX_In78q|SGtRL3O{*M5f< z_c|*%`f0#s`LiYldLY(=!^lJ1)U`z{d#Fw7rpt5tE#!cd^R@Z$o?l@9?9}7d@9=G$pb42Lo*8H1CCVR=7(KEwqFSGmbdx z>u;X8_8C7B$O-ql zC=mWk@&H2UecbE2ss&U8=l8*qaAP^8F|Ui3Uq4e&61;OSX*0Ib zkn@t7>ImndRB4|Ef7k>3>7cL1Pbi}pHo_RgvC0TxqX6n+Qkhh57gN4VL+r7|hi5jtU3y?Yxq%07FnXPQPfJ^i@@u2DTZOK`*CG$ESJ_!ugJ zL-puec-0^c_EZ!L#&fR^0*9qTceCanpZG6wg4OdDsw302Z@vEaJZugRx`x;mCS6=O ziA5sA0$U)7q7}cT1wTv#{p*})&SiiewtzxK#1R1yBY^w#$1U~VW%lpFJwGM{f36po z%H&kxrjysbG~xVLcdIcx@c|GpD)VFAw;y}|Tn)S$w*k6)K;_pSxApdRYg})BL$9re zO&VQu=WbPb3wMkN=v02HL_%N4N_q_YkX8Jt2y>%s0XRtQ!<$YSScn_uiR;@ZzwmB- z>c%7H(4~NT^5y5{eCItq1grM*E z&R-|R$t}n}6hS=3i=^7G))=02BgQ@-y#nB0uPUR^_?LO_O(Gy^FhB)*kOz33`5hq! z$ZIM;E}nyp)Q~Pzt}7qI??)%#M=uO+?m!n4_AU>a-rGdmTY*GHBiv!xK zgqmN^=|W|1>nw99nJ8SmEd0SE*0r2Ml;jT4_KjAr4y!}=FKIc*u29bma z-}%bW@Pg)os2QzU<2x~syn1t#Q_ELHq(q!4{-cM+Zzd_O1PRzed07jHFk=X8BXXsc zeV*E3N}K*|6(|&`p^U<^Ag9?*H>80fA3U3XJpL7k^~xwJL9m&^J6@#)Ci%!+ zJ!=`=XZ-!WP6UNR;;HZlM#hfkL#q|6#vvu@_`K4%Qy!P~mZBbC?|+ldA*8MQ(@HZU5`UH@oZ zyFIR3McD+kT=Dy-N)@=q5XjaiFB>?%Bv>A{HBr0nd@M47q5dQay)Y@qM<$rfon2yw zIL2bhA%KOoXz}G%1Ian{G~G@U8F_0$D9Mo|O=QNqdLE#4;%B6R`pfF>G`4`!E|_e| z)}wNsLbAWk*9M8-x^zI-*)<&7^UrEWsSxs03tqO7M(7r(^w`_|6rr;bX&}ENac(4J zlq>`2dZwL$@bXj8V?{Jzq)=j)#{1dD^JD<-{R0-pZox6`Y$`$v!QP4`rx(QS8LsUp ze0}O43gTqoo+&!VytWow<_R*lj+Z|c*+B1=nY`{H&v2SG1k&|-mwfG-JS;&Y$@oM4 z#!uO&i0huBtL)=Cb>bg@msjJOsw0u7NebYnVARo#gt8KtK+lS2A2j&5s)W&r`9l>P z5wsF((Eu8(9k-qyb``H>S-A5jU~Sih(b99&b-YvdnMy*7th=cRG>N}#Q9@|W{eTiQ zFTVCnO#s4W1~e z>dg&)103@|t_u;0-$XJ^z5`L?g&pY1J|7Ur!>n&MmdGNH@s8s0&g`%YjSlx8WQ7MU zNc7qqCQJsTq_ioMj&9vHP>H#R;Q$bnQ1lQbq4;^CAMd+`mnzhYn21`ycNFJD%gVx5 z^z`t~QY|9(?idfMQ|_-c?Y;bySlOp*71fvDTkeBo?PCdLpK>RH{rOrc;4SwWe?AeT zFh2Y)2HNg?a3fuz(0R9vjZ(zcJVISH)%$3Q(DyeDuLDhvt&eX~7}o3qd+rz&e$=aY z1Tg%f(F1T9ZIQJ<5OXd(^8+gGac7IA;H+a*L?mlaP(O9BjDef;oUE|{2*{9c(C=d-jT5SUXaWjdJREWyf&ePgCTz$O1 z#E%E%2=Q8$Eecc^@;ON%pCO{MKq0>hQ1sZpq~eazBRMF0Nqmc8l47e_jkxGEVk^Zf zKI2iYB}$UMHJRgwQiLw$&PrsNoSUEjX|uOPMGqjrl0Q05A%QzgI`X2* zK1nLS5|s9hF%7hsvObGiBV+M0Qz^J?6JD9+9Thp7o-|I z$r2TJm(fRIXIh9X-YSXjpwoXJ4#_w!a+;{hv1&{s5X=F5fH-pGZY22#BF=&S9n&SY zSylKX+j)7)iRBaGnHq!dv0C=--mkpJ^>(wXyGV8cpb#fB=Rrd}N&?+6TMre}HUdklR-%%)5zPw1H3yNnOzV11vb5JSXO9ZAkBbO zz0+F;3_w8w_lnuztCls=Pt+aMwAgHkQ2i7SN0$aGjR|ThC!?LNLEWjsmjb`^MlzxK z+@6vfBZ2wdtkq({ag%TV!Z~C*rcfZQc#4d8>f$XI*4iU)ax-h;Co&d zM*0r@s6! z_;9ySgCjt}>Wj>N<`7^hSuvP#QX1^bJ5J*55!2|~8iq|7Xt}c-Z6?A{APtThyYYJL z2Jok;R2X2*KYhlg>drkcb!zf+@thCsns(Z6$5(Gnu;T)0pVNdG4W;}rc zG|(BQ5zN`@EYYv5>lq9d(iu$>qgF9M8K41ev4kQ;AWFM_A0&gb{}8MM)K0bLZI&d=`McOTbXh zRzeJ=F2pzKuP|J%jTORpyAetkX8=m^O9Nd;ZeR#Ys(v5NwF|Pka(elUbW;8HK1C2G zy&}BMAi9UuBj*9O1%CfT{^`{nQ7U2U7+@}DMGT*kVUx?c|J_rWCtflfAt!$o9YJ;c z{jI4;$kDQr*u3E`po0s=uV(s)b7F6o7}&f`Nz3^AUK<7x2Y zzg-*JJ<_k*PT;ZTy(cMvvtl>WLx}&70RJH+;rmn+&GM#WQG;ZMj~9ty2!bl3#P&VB z2uvQ7yue8Zg9GVWgm||geQVMKBNVq0R2<1mb&J7RDbXeUMK=v6r14KmXnoKBx zRZS!8L@Ft$o(eR|=V+9t0a=iF$7x|7d!Gx;r+xS3kJ66Ogg{g!maw;pn!v)qlYjC! zmu~9sWWk>zP<=54qXB;T>&l%<@!HF>w=c=wc2zR+L(}^}gevN2zB?sLjK$n}Vn0Z; zz&MU4UZ0X6Bwjmv@SzkgPsyL}8xi_aKIHuNCTa+ziutQxfaE8C$;aqYyr#$0tvHVv zl{O;@LI;QuKV|=xch<3B>|i?wU}Z@)Sx6zwK11c&$eFF{cf2Ckyjb66d%dU^MNj=k z6C0Qb2Y#-2S0;WUh_&%Z5tr?2-`6?Rvi!w6-2saH7n)<6)WO)9n!s>=@yKi%C`uUPS8 zs#643YNL~AsA7(nfUaoBqX(aJNspSdK!$;XKF)5xm)#%NI`oIy9B_10xX^`%9%Pqu zj8*lvTYZ!Ca$H?9&Jhs<4;DlIH}8$7_NPM2pC4;?1I*F2GcpeW))Q#il^1HAffJe& zKzmsnIp00$k$H0Mnm29f zGWSp13rt%g6Z`|7NC57b<-wrbJ_2B|!~;~o*(kjI*>*B%L0+#EnO5E^qa=rI7~8Qg zl~``C0fiNT4~o@b;vaZ92OXl_(Q91{njxY?Jd54@MVA1I(EutL$dut_!KR9r&#&jW zz+o#46QwsGFtkySv<*{^e-|2|YJ1I_{J*?+4JmL5EGZS9*_;+s^6t)51a=nX9zk%n-u*RN}@N$RJ+ZcVb&HcF+1 z75IY(5E5l1#C^Ocuz`wcyal`yUM2Wnzt@}|PM2rRwtRBlX99iedKR_WQ)<|V#>Y|@ zZKaW$K%s;M1lstc2J-*Tz2pZ&oPCF*pPV@ZMFYE^8m5Jf1cBAhU_(?T!3?+l+lPlh z{HOa7`o~VLKJK%&GCUvID8+lH#dTTn9_dKSoPr^^=~j&z{m7R zf#0QI#@mRTiqhLO_PZ(InUt!as%QS*s>*;QzV3C($S@Sjq<7Sq@?ZG&QiE`a{O+%J zwKRIl!o?qZ?U>&acDhrpd#Va|yr6=vyO!lYm;Z2xf~qJ5oIhtcdYbn?cY%w!z^*mI zaSkc4^pW%NNI2vlypYy|EK%lWeV~Rqa-jQfBn708Dq#)6)VqUg#(&2V`88@-12dBk zNrF5?RYw^9%V#jdI6{t`wk%l0=tMhUDE%7GOZQHK1d@M4n(;S9-F4`Vz|^uot~hHM zPVerF=Pqb`_%?onxzY5&_cn3)*Fs}j{K$`*B z*FckW7MYN*sRL+ph;4it=;Xw6hKiHF5?I8a`Rp>@fqyIM-S556%xM00WM}FlEBgPf z|7?%;bp$=ep4;BPjED^bXklTLq9Aiyf7;)ZSO(fDX!4}6fRmi52V@A#JNDq#J|(Wn z{+FpMK5m&?=AFJkmkD4u761i|DARts+;x%Sm+6O&cD%8x&!~a+9Kf@j0WbjEkbSn= z|9tlUNonttZ#9We6a0(YJevIQ``iED`1gSves>N6qi}>*4}g2Ul_bA$1CJPeu`-Fk z?RkT2)*Pg3hZP>#tyMwW(a*|j`4M+cN}y{jJ|Y}`GnL=;c|V#=V%W0nl9hi*LQ%e1 z&2NaL|bXRV6uVP4Ds#lH`XKmfA_w*h+tBg>C( zqq+8f(1RZmD&#n|RnbH2tp+9SNC7kZ1){_I9j^tV2K)|;Jv3){ltQm9_FE5oeYtuY z{%vhxEv|SK+di%B(12!>@~;~13>NX)TC-YP^t*?(zSAF84`S5&wx;nQ`9$Vshh|&H zvO4Ot4)6T4Aw5CS@rp`1Am6=~murE5MK%QV9fdz!$kX`8m+jA@@gE_B7W}5;*c#2? zdt&rbDBpMX9j?OJ z@YStlu_619o?yEs<})v1QllkhHAO{KFxxWQF3rSbgZ3XiOg~4=hbkEjQFt7!A_%+! z?AAN9tNCBZC0)}OG^&WPb<@^o$W>d}=iT=WtA|?)hZbbn9--MM3@{<|!rAlnp!;5S zKuhaCd`dktVuE;WP$A|%08Sm+&G;|QV`xJK?Oj(BsRwz8umX!}%>Ofe(>v}m`DmwH z@yW>$$g@D?67=0!ryr^r@msbtKhP^ZuDj9 zWqqHNC4?UT;`nn$OpEtaI|#c1n=`Pe?818w*)5-F<(jMNlKz!)^EKW|8)$hp>w#|# z`o20Ch+-qSGwpqdEN>RPj*b+6EY@KH`yQeKp2}gy`1jAGhR`(N!V^#t)p36|Vj0pj z$f?hWdj41pY{GYCiwaNbhW_gA2XVU^YTOG-<*G2hZ*oP(Te=Yqu$|XTX+0ldqt>9{ zqK6FGtM}v`)w!WLIC+_}c2rMn)07(65vb$h6On0T>CSs~)J zntW*GDG*m9j@hPM&Xp6Rz#7)PAc#X1YL(Ar$IEX{+v<_k988N-yET-iNHWg!n3Kzj z*5WDO-^up)r>K}yL~eH*lXlq`dwb}Bn9mOqEd0X#`H_!5&BJs?-}j+`iclR?3?mC0 z%ULP#7oJOmQeYKYPXnzdsm%{-)}6r4q&$7(e`Va{zIO4wcqdmFg86oP#Ct1B>x}tr zj-l6xM%4#v*Lc^uvv_4ED!-W=9&VqX%dM3XjeZPM6in(N7zjDN`hbiPa&`-UbmT7H zI|yG>M9=f3xgerD8zs3bf2qCH6ee%R>W1hR8GaPdp(-~0 zX*_gq2BIf~ua4W|)E)UTP~aThrvR{%Gf-O05gLZhmHH;`Gg}Q8EJXF0k7lLM@;71rP0-gIaCbtCu zn>YbC>n#d42HWM>T59LV-Bis3cAQl-1Ehyoyw#M4vSw*8n4VTJ)f~p(9c7e{P-;3d zWCbml4M+&$y-K%b#vYp=+Utvq@!t0}j9t5h38sdEW%>KVOm!HFUI5E#%){h;?n4N+ z`POw>>&Z|F6m-(!Jb&k7_Sg3zqst&}IvGQ!bv|a3@sgZ#4kEJ8fAiIAR>5=46U}N@ z93;|kesnyj9O*il8tEeG?EZW)Ab|vrE2?l4NocbgWggyY<3;Da$S92jiE&Y^P#yQz zZ620mPK+kK6sxd>a4>ikQOD*U`2mwTNR6-1j59%Mdg1iX?`zQEor^xM%fVNbP<)vy z8u`ME%Xkq5fek!GGoG7xc?E^L=LAMAfl}=FF4n(gMspfSmPU~vJ2GFxDHJzw^=Xg1 z8QH-wY6EI)dCSI2hs&rwD9I*S@dV|kR~9U>5ml0|JPOIRsD6rkWi2qxaar&h-D$=e zL=^|euC@RZra(P{w+kf#Qy0D!3b%J7lO}0l4W!szrP>+DK-BNG=hUBOesHZ$bhy_7 zKYW*aw6~*q_byMZYR;J4awwp?XVZ!q-(CZQ6}+7bE4OEWjoc|1EmFk(`DFXs)jsUm znRt8ty)M4@EWz>`Z!k}0Yn`{1L{ZUOVvzM~h2aO~|ExLC;!%qi1}UPy~jTVKj>J%#({$Df)p zUqvxW^5uUqdECG-TiHT^9LLzcy>^16-CoXC73Ieee{4qemV+bOrsRbWfo<=KifV;{{>4qdKVD0P z&+!X}baWOcf=RPgXdmiRoyH^PA{DW)jJ_4psQ!~l(dp$yUof~X!y$N?tcS=Qpj%(6 z;b3|bafA#(2*@}D`Y1Pk@ z>T}D=OS)2lJd9L^4C!azPAHJvczf;Hk2(=?tTmB*E49&6mAZ;RURs8B(-YXs`?8bl za@jCt`(3j=s?G1NObs{if#|y)aH_O0J zjlBD*@KNL-z4oIxw)LwGmAuU|VA@#Xda@He@gU+Frf$OBq~vH2g|G1m8a~1ETWT`B z^vHMOb2Ut`kDZRXC|%#q99wJDuR`DVlSv&HT*IvXoC@daaU|U0x*?}<{OPOc=T@hn zy-zux*l2Y1lbgrMm~X8cGUV8_%tidseMp~OtDp!d&caQi=ky^7=3IYa$XlL;7n`W7 z8E5c?LEgA<Isa zk5`Dq6rmaAiwMy+Sp+*gWm~)2FmE>Z78sWBy}CK()HNx|(;R&=cCkV_iQrwSp_9z7 zZm82600@PB4gv*9XF(jrC7a&ch}{oJ2a!h`WJGw}Fb-Kgw;^aL^kY!+v*Big5)~}b zml>b7=Qos}lwO}R9aSx94Z-H8g32bu&oCy4y||fldsHvnDB7+!h^|ubU=zakICTqc zZ-#l#MM+9fpU#O9z0lh;`LHf?b>UfPho-Lmix6TK6M2bKgg+@f6Ok=B#A8)=`eD<8 zW1G_L+YhK=QFl6cSKU?h=vI_B2$FC`>*tKnBj3f(1h-Nsz%$pAwXBQ|pz2uohCSp-HBg%iCsO^PF?f z94QD~#aR;`l`P5See+qy?Ai=p^%b3@P>oT4^Nmf#eo_s7CdaboTx*7FS7Hng^7bu* z-8`iZ*&p}EJ%q6F(X)|95j^cG%C}evi=(73iuaY~NYG%gt~YzI9`_iAe--DaT;diL z5r&*(m}3a~>Deuzv{8*u7B;SZFYOfVsPZ|DN1Y&Zt=nv3kH>iuDH8l=+Y4L{L$B z_-dx~6X&dV1O^@q_h&Vx62nYf#KL)BU9Yg_ZlFB9>^U%-&XYG>wGjOIGvtW080b4( zz@Tvy08qMJPS9SC%0eNV!(Wb1Ec+?-Z0HtWWR7C|jMTfR_k=%||AiS%GjoKYJa!pw zP7F7A`$x>41Vh%ANhi%Lbc{>~mK#6PLCmnR82qq34lR!%sC$H13jbQUxZE*IoJN`x zAVL1&hc)H<-*8ISnVP+(1uTE=f%fG5c|`9N-DSHr&bFYc|Hau=22{B%ZAm4hQ%a<} zq*FvfN=mv*VuOH`bcck1Ah0P#De3MGX{Ec_h;%o6FDM@5c+S1|`w#n_wbsn6Su@W( zgP_hn_~BN@HGjng*2auhslJnFpunTlCzGdt$JOYdxcbv4AnNKr+)ZJKX0|10 z>v^6BiN}5^-M??xWzpgC=S@hIPUexA4&SA@qt&W$70MC+Hp1ZIjllxVPlUZLNG}xS zaSXcQjs{JW3*`|5LiXrDV7+sNXa5hZ<*G=1Ay)dfaGdfynRe;2$dj(;srYauQZ z=cEf|O)ql8`J?R9^(`YaxY2Id*TYLluz`qguGCe2ZFHv+yQwD7U@cnr4j(2}urAd?Kt}KUm3vtTy2zvpqC1bMDY+%90dnHOnN5EY zb^H)?EgcMmH8vRNyQ{D!Eq56AtE;7s=iR6@a-DtMUtIGSzopv~Zwc@xWz;ywbMGoZ z$$SDDC7JpjuzJyys&|bQ+5`Fs{ZE$QXj@G;S*21dJhOCT^G`eny#!|uE z2n|&?@h|AdDei^~#IpbVRTC{Gr3x<*y{>p-V-xUPy$Z}N-3w7svx61wpACA#}(S|uSuTxTs6@p#BtOGj*q>`a1T9pBU7?z?O6*Q=1} zQ(4$Vh@vg^H)H6$1b+nQQRl(CO77~u?6hc6W5rmp`66UMYd;C<&o5WyCJOZk<6Gfr1|8Z&dN;|X%4MC53 zBD7fJw6ta3y>dsD%nUy4(GL-D-f->EF?rURk7_~8$}Q~7{8LYTr4)6RGYWXSVQ|X< zIadG2bvg}&z#U`Y-0G1GU?vM8YxIDTeOcNPg;y)@zpeIFeyjRK*ctiqQyRdh^1*MN zti0?(zk!UE07Cpi{f_5u#LZcQiB%VqUH()m`&s+_<)bPVs4b3U?5&AFtBMaZ4`4?gxQz0B-?ork5?mRVf>@&yW(-tdpPLhw=J#|K1ydlX5dIlPbH;O{>t!NUIL zWwjKx@}B8@qS$$b!1C(+Dd6SuH#M!Yd806M3ZiP_HiTz!N7zG?kWZKhC9w|O_o*+& zr4fE*H8o^B0rlF z07$5B&JHZ$9-X7G|5W(Hx>}HOySM*hjj(AvtGktKOG&ZzBZprPbMHqNw)*GL=`SAf z+zbr$E6UXX*hlnre8M($rYAMbFV|F?%#!S6Fqo10-&x;Ym)zL6^CyRcHK@$*f_Jd@ zn|pEIngHWM#k*nUJSftE%aZ1&(?-r@mKWGUBerHoJC#czdWwEWE_eFZ+*)2kMMhFB?WR&oY53OBMa=6E4sl| z(VZ7eJ>EE&Gf7$PdV7nkx`V>;Hs#Wn27&g60jXN8jkC8Ii^Fyu``d7oF6)g5@~Plvj>``Fa9`L#Z&(OcHXV+VpZakX~DwRfKLt0L`$1 zsCK4Ae84yBkzZyJ1LJ6SV>A3FM;Md#6A9DsGLLB(iSPL@~&bDBvh>P-x11 zN@7zS#B9Irxi?Z`@6xr_Ttd{X!HPDsqN5^V`|)3O<~&`xqSH~l@Coh=7N?=klIA|i z!2ns(Co7xPprqGMZJhe)oH5>gUc%1g7mEO(+pM|u;-Fl6zVUC1f|4I-q*p6d#*1^m z6+7f03=}e;88lb4)55k6O|>^dn?0PZH2pk0Rf2)mcYiHrAS8DK6l8EuzHhgToAaB5 zic-0U2qu_S<#C&_LZV~*dKf`kQt%ry=q=2Y*s)?5(uvmqQ-L0Q7_HC0I1eDhEW(Rd zu%3H+8sGdCZ$%T;7QX#m$w}KMpW~5S*rJ-f<{%&5zI@^z>Rt53j?Bm>Y(!@*6N9m+ z4@T2yGo`5<%XUvez0*!^6l1i~{M83e2MCr5LF?0;yZ8j2rhGG^HBX$9ytR+K7E+4e zKaPMSOOH3NXB&r*AL@;g>1w4q8wB}}`Y%-&-1Yukk9X)R#)x7g zS|rME;jw4qd2?9Wn_wyWWj}l@Zd)D|aU(1jXT+irFinExcthQs(OOPRu^~jZh@DQl zwiPA*E^YjUc{`=id*IBXq`faa_gZg%VmCX0HdChF$cS`Tly=pC zN1Z32SmeD(>plK02f=llG$WI*N|F>dX|?#eTCzjDm@Qc+Bo>`DM*D^b(8u}lF{Z8e zbErzO^OkiXVo;if1Ir)l@s*!%_(_g_@L5PRs4jAKqf>f~RZ6*Ddq^$KNFknO{)sK1 zNvz-%4hc7vB|ol%-*O%dX(Lj!#D~Yze}HFPdTAgf^%bSrTvo|uuW3>v3g%;jpnExZ z7*j3di?uhUG_CEFi(Ca-;tjKRE215UWb+NshL(5d85+56VrJQGxhaN-RuXO#rQ@r| zYlBBR*8hK;yn_AbrlzB1xOOm68cGcbLfE`_BeU0b6LK%cR zSp&m`+7WfJ!u4BU4_9D%Z$U@RWc0a9+N>I;2;}8pIyk8awvyhUp*v)G_)E0uTZu%+xH+G79H!5(?x z1tHQXpmPUY+e2P7Qw)WZ>`0U)Tp`TQZ`z~o!K3>MPS!8{KCJ=6tzsfz+ScQznns|~ zykR|p&=2Ye%!co)d2c!$=yuobzHJmaPNKNO0?wl`ww`%#yFBU}O=xyxjDbOB5xcd2 z>&_C};#a_OB|AhO(}(H*=S4)-CbV)nNSBhuD8Q_KEBb(2?P4QH4MIp1S8RJ=XA`IX zI7|zXd-OQU76l7t?s(et&KZu zVHh^6c>fZ70*sWTBX!XfQ&JF#wSz&Zst;A&b&T~$$j=|gA8-uTEd_iGYTpXAY1jX7 z@-ih{T~^K;OwkJb_`i#Ml>z60tJxXObJbfNMp*KZL3mvx3Rbu3!R+i#MmZ zVI$$FV(EMIF*Sgc9ybM7%9Vpwhv>D#ed^d}^>YsV?2Bp2yv05tHNd6i5F8_(Ugmh~ zUmkWp!vu^{%y|CNCVHysD(I!WH(Sq<_r^)Bo#NCJ?o~xSPP%j|Gac`@qQ*IjFxmtn zf*ABshh3Yg?_xMxcnzJb_DIQ7rfeFas1X=bsUr-bw{6@8hwvH_AhZ)@e_?vYIula% zUQ+BiOL|NiV?;>CNM0w8!%D0~2=OgaEqzLPy)GiJ|7?D+t0FCR%#FMbRv$uD+m_$l zI*Tph?_#(ek>LOR0D`!9@qiE>4IX0X#9GQg3P;*q6oRLuFF3$F!Vtu9eq}OQjm5_v z)&}N&cCnka7FuCX?6R^sh3RmlPDjM|$Ars;Ym!>UrO|d3S1f2f}GdGL$d+nG6RgOqwJ5y6=-*Sd^&0==x~B?S+F{fvtYlH_IRBo z)nbfmZpaw##x3M=v+&4Gkt{^Llpx^fm-?tl8V}Rd(r_LGRD8srW||roj(WSvI7Z?* zNyknOtDU>!EGYtyA*-ZLV4SLk!14%yen?rrEecyN^Z62+meeU8Yqq*vao3{vhpHg^4KN#(>mnXS#k-P>c=^V{y-;<$dV9!2y4 zIsf#J`6#5sH(zD1hz41c6+O(VWD|)re`Kh(zM;Dn*|`wiBdbX!DjoJ*EG`c+1`Nzw zQcGY@@ARUi+WZ;8!}c>O#n0$MC=) zk(SrKHv{6jT8Wh>bZa_WR__CX%SSPt!QCvK<{a^T3WNS~3gks(YBm;ku!rn8zJ6#8 za4ysbzlXLAv@+9!O_2OExYrWMu~|Z0E`013JU^FTS+C0sGQIaV=*yKhgxm^ZnYWH~ zIf015AA=~!_N&bs5KglTNU*sqL2=`xv_kh9EI4CD;mlQ8gTN)-F{5Ecyu}pB(8)K2 z-{_CMG>{kYEhqjHBzYoyF9`9QJ=_S;RCZqYv4HX-j%)ZV9q-|Trm6F_Xxu*w7}*pl(f+-gJS_i{@jrH z;ztqKsulX`XxODjcSX3Hv&1Op_0yJE{_sYaVlW}b&ec8;+zURNZSK(ZKt)8XbW%*R zKbe6=ZrgoU68BvKIE7i9_zDiFJZV!uHAh?k`})-7P;H&T11jeU@LX?9TakO;|Kx$P z7~FDjKVVGkoWml{EdrqCy6XF7NR)ILrY#(g;3lwU2O4i78%>e;et&sh_PdVmcaWR_ zKvO0ccK?H9LnasY#z4f-A&kyMVw^Xci*7&B2=$pBX=L5^3fziST^6G@qLF%&!O^Fl zFAz(1=JXMY2AYS(Obb(10{3@B@5b8?%T25{n@THwi3Q*BZ6pF4CU8W-Ro%Bw&ag8_ zD!1zt^0y9Bk6<&R{-LLB4}*JaGstC{g1V`FXE?`y51K+WS_8 zW8UKu1N?KZr}TN-#NbS`f(t^2a7=f6o;+Bi{_a0+Jz!IYC@o66{+hMNAZ9fd_kkBB z)482IP}EtNj*hps+I=T+FObE8pH=>!(@7=3;Lym#y^}UdA5V*#<3pH3gU<4Ml zfgeJ_EP?xoixuhF8qT`7a$l^yKFB< zz){^_=yxt3{^6!?&nO$kRU9trSIG31@=nWXwc@)m28uDQLAywFq!NvuOj*>R^#0dU zic^_I0m~2d`rb%zvF0~_z=V^w`{rJy<@3?I4-;0dA8=a4&)WNUM+?dsx0|pB7tE5oMK^KjeCv1PsCLSJ9w- z<3Wj!%@z~(W(gNNt$-s8h;AAOh$VD_HZ}|39^PEusb^~z_%>Lw;Qk_#u~$R5K*~rl zB8&RF@V6;`-a}xYn*ZZIc_83o0dQ-~swLdi;ZwKiUMpt4Oe9jCh8f=Tl9E0b4)%~j zKINbjmz4T)138WqB$fcNT4kxr+xmVM@25;@gSGV0+q?;dWz{UaMLwFz)#A{j%W2ip z+!2_(cUN$a_|6(&9FZuuK4$4G?M)Ce;@^@BvR#^YCAv&f?$tkz*UV~Ixjk7!B{>Sl zJ*MgZ9eV2^SuA|q61s*dn6I6CcF-GjcuK~Nsr?9UtRbBvy!n_qlvzi2q&Av@8c2as zy&Po9$en~cIox@54_a8m{U18K zBa70_!8>{j1wPQ&mbWK%Cm?We$sBkw@rV-mGjVY+c0A@}HSgpC%8kvX0rs;&WN+%G za&}^mI^@g3fRo46(9vYSf}BT;h)(SU7qxBd?(^@qA9NRGFMaA4+5HjiV&KW1b~8>Xrwl5bB z{Fg9;zgn0tvdd!0(Bj zSs{P&dS_n~0WH6$)-e;rLzFwl=i}vdIw>HG7{T@5t!|BC*73T`j zw&Sy_1O+HL>2v@%E(`LgqX3IBUdkjOn=H(S{gXBa`7^ydeq6Op4Nd;>fIky72qge} zZ<#X{GB){z-`2RD+Dri zpq(%E@M!8}F+9%w%~Q3ZKvHljFgZQB1E@EwM{$u_FsegStbSTNP(P~Fz}`XXPei6M zC#Z{F?>=E0Rjnbw7H1a@@KMBn1Xs@JKe2PQ@dBzyoVJDbGv%Y^R{_DS0N*1YjPYsgY#BP2ZwN8?k1ilveYxNX9OqSovg6hATbv+kP{55W^#8W~+rLYxdg zON_FH6}l`EnS{P@&6jOWt7V5FZL3ZEz^H#&{Tjb1OZ#!256<H#*WOr+4Y2j?Cdq}Aw4rV;Bpsi0O7B8)Ua`7!GUiT zp2#TH@|z~UV1ONe-VaXXec{8gwCsE3HoZcl1+R#cPM=V%K8(l)7OOWVgAfv+$N3|Y zpUL)#BOdxxIxS>L8v)0LMsN)GkX6`KxA)+YV+#ct`)VYq{=Iqk_?qmB_v$dn0A||9 z^I~y<*A_>|gKRSIu}nHI{y3J)nEZECRf&Pz_-UiHTc$Sq32aQl7kgAoqYv?=U@Bvc z$YD`F{eh~b7&wP*Sa21_-LDi;9BzjdjwU;+GqwESju9&%?PT(Gim4g6a0td;bq^n4d+uNeF zsgxO=qgbN>h|KYv2ORC6SU||3;j9MlpeY5v@r0Tuk{JND9Gud$ws z>6Qd6Y@(00E5nU&Igg~e9Y2I(XWNg(wh4uHR8BdO4Arp3467n=!hQgh-prlC&>K+!E;ZP`|xq5E%K3y|M9u2m#o!|w~ z>^C42z_}b_<9xug*ljsk?VE9WYpDPZ)y<^$9zIM`E165jfqrtOrl5I-dlS@$-l1YU z!N*_R>EMM4n~=|TjC0QEnl8{NxN%d@2l3vZ(U3SbIDIw(*jqZOZ2;tUC54Q?AL|6< z(54ge`gvnMFvd8yY#KT@I7e1rN~HsGsI&_3Y}$c*thh0Bg2}98|0p58m6_cs@g&@l z*eMGr;7Hf^>~Q6R-sb{v1oP=JWqLjn-xFaVo^ z0G+vkXapbDUU4xqwNRuhnUkDHuB!T~5|eoW`6|L6Kvuxu0gFyxNVLi0f_eSCS+@h& z3%2z0Jknn6=U`RK#_z4@tvPV3nU7miA|j(F-4fki#s{JN4z`^LCn8v*P8S9J^meJ_Vhyhy-o}E7%|I7e#5F&7x+sDuzrR+w>Ig^aP z%Em^TCe2~V0e9A$G!G2VJT>TtaVXG*roNDh(}k>5l-gz8pUO$38}wiy6U&(o{0MI`IOOTlm@WfjFmfYG)I{rVm56s#h??z z0J;!40ankMBq1i3>ye6`w*#cv*Zvja&`mEKpqO2C7l^B_cu8QRj^Lrr7~Z*>t=r_W zO>JILlZ0ScmK@CbG)CH7SU5Mj}&RP zXS#R>-H$TpctY^Vkb)?O;75tf?G4x{>7(zTAr6t7Qo{2}f8WJ@)(CNKSJ;gkuLM6? z0#969P)9-Jjtc9_p97)BEPMj&FJf?w^US!v>X%?le|);P;+aG8>1j9YSpn`vz2xE^ zy==H^HwSt22fBcOLT4lryGJrh8rgy6zJ50&^ zlTif)A#R`TK$B|Ko0IoPmZzLI`C3Smf0AMi7fr*GCw{7RU`^2`PQyHEelzUvo9Li- z$Vvi$eUR_(0PqS8!46BSU?~;V0E}+Oy{(lzd?xBqn2-Ns5oF!?60&Kl~?I1>Zqy`6Gcg? zU&jw~13l|NTaUdlkWC0Ne}8nIs2{)ciIdOq%>W+^f3%E;X$&wSMimu3KS}~-0|0|7o}9BuU11j(kM_n0w4uzLJLo)J zTZTMK@{j8N4Ty7WR1zKV~?LW zv_x|qIV;u{4!LCA_xqO~4N!n>yLT(n2T4j=w6-*R+WWzjshyN8_h@x80+k0=F30)(sSQEl^@9(DPbbXJ3nfpV!WD*(4&=I26JzZZZ=&!%V}v% zRU;vq=l3RtTOsMOu{kF6HM*zO>TT8Uty!Xx{hRViz82**FI4inXi*|4>m~c!R|ed} z_s}sp-vS+CUEuNw!bz!T`JOc29y<*e&hYcyrOhp1t(oNe*KWF~Of|ZIoM`IDw67S- zK4~9^BEPtWbS(8#MkTD`}>!`@gdxCK2QpxF*~5g;0nV*59vd>tQ3lk>zP#Y9oZ^v*{AX#?~%XW|VE zA1Ls<$DPo(RRX(=6sU6Sc}iq-=(~b@TT}#&)r_qspV8-EZ1Q%if6%w5vuW-M{DFuD zc!uXq<0XwUHS#wT@>hSO z-JExs(%poFbbrL-`)8xNLmeT78SvNq`Oq+H&8AzF-a^Zsvyv>r77yB9LJ!KS`DsZ| zX;^$ZfpKC85b7jH1{5u7O+x+MV(w4jBBjeFSzJ``D~QBNnwTr*o6 zAP|vH(>KV2c^kuo=FGabtelfl8qK(Kma(*lqXlA5uFK6tL}zWsIubo@8X%RCm)1NE$S z0Yvk&xdZ!v2~9-TJvQxzG{`_^|F)0O@3s}#KCDTw$TT~4^Wuzt<5Gd__y3_&12LJ& znssu*9`OSut>ys_^XZ zze-ShCq__xOTYP5#`xYj6+6%?s&Gj(eT@xe^!Xe?8#o88PZVxI&{mDZ4IcD+vuIK& z)FSGj)QiB6?Ejs7oREcX5<$o$_I}>+cb%KRhwM!Z;(l-zM8FgiVD?x-;Li;1e;^?V z1W&nvYQL~W5B$FMedrQ2*yeGRvGT)i_uI=>Sx%IU43v#OusNGjN}` zUey9F3FBur^v}dV4YS)dr%PvtH4cuxi;HAd9t{vh&%fxmE+E+2l{Xw+|FcY1&*LD` zLLkK{agC`Y?*5;YX+L&<1A6Y@kq*!<9`GmK5#0&A;0m*P7H2M%BtW~)!}$Dj9Q^<1 z7j)t@q>_M{p-LK4?%dL~hCf(uFNpL1eMf@7V_y(G&CYXWv3PLEc{poKedG@Q|NKG| z*8z*iQ#lU}U~)2d=k;H(Dc;omLJufK_~it=--x4Rn13*#8MK%XU_Qge9tuS==rp;- zh|&3U}AjDl1lTFTJG1O zc0Cn9j&1r!9YGNNK^`@36{rsl402s8t<44+3GQ7ecoWj=1@rlaBvqibhy03{0*H*u zgMLyrU#?|2NP(8}3lHt*0JxjI&X^_^ShjX7N;XZd>-Ww1@*GCc;MsW4gq_Wnhdw`{I{L{#RnMp zOHf}NTW|bKP__rXXwR$0F!bme4m!ergQ$pHGa3cJvZ3K_HM2~s7G2ThqYc91I0l24~se1*sc1CVPq+?D};>Dyc zW?WsKJG0CnP-;4TxK<^+|Lg%D;T*U60rN8)GMU=OG1k-Vy~wEPw%S^vWN2;Xws*6R zG+Z(S8)6L{y_~LlPoBry5$mRUzK13|e`6uDJw97I3#PA$I|D6$b;?15uOKeGON?}R zj#N5&qR9NmF7Mw7K%65xATQn(d4urx2M-mWOqC}e*~&pFN)B0%@bkps zdh`~eJ4MSTEz2Z3atKHWI3<=W_An@%ywkMVs!J@RD&=20i zUc3`vF?VZV&i2mKSwP+ncw44lJRO)n|MU%U?g246L&KJ0J0U@m`(q5WxeLTZU6QrN zBmVr$Qj=jH?MC>GLk`;0By$9)Xp`a>8g%cxriF`d|7o$nZj}Te9ls4g_*SKS>y;*| z8d@maPo?$8WM)wa;yKE4YXFbTnU+FooeR=xEuKsT!#3TnH5~C*?s-7%S9O3V#?YQN zDr?QhPs_Z~q?(;uE7m~&+bV*3-HvaTA3r`_a$5vGL&D(i=f!7CM*`oe*N6}uJ-IJ) zq6BZ8jjZhOZE#2~! zOjf5YKtQG5JYHaIO7v&hI~(Xj+(~J*x6VTnef-q+S?IH!8fP1CG++OG$OkC`2$2sRMg&5bd3-hL(1HQ8D2 z+(>gj=tie12lv4kbv}gnaCyZ;8*|FNV5GwP(IHf6ReqmQM-RHFSJdC{s(EoiVcL-0 z#_8uas4h=)x9@rZuVKCYVZ%uTF3zSnm_ZjazBAOb0nm`47@2#GU3m{__fCUe=L54X zLQXk$ZfJ*Nkod<#8K>U(OEth&<@Ag~Qb~3rw>Fpo4M(}6;TG7dK;u!SHZZ zSW6#GD3^swO=WTvmfAGppyw&2Y(~&=6Cz)K8PBce#qiQv49v864q-mzYl`6Uy)c8F zYwjK+Ls;xkjbuVU{uK6e)I-oP({})q-s_SRyuQEE&u?|OI(HL8mlRc92bhcoVtCQt zEGNlEuH5l3+kEKFQLcMp&16fzN4p;~>wqBU%PJYsjR(;zYkaevo9)YYSei3dAnQ1I z6-C|xxV%tZT{r$ZzxP&cO>orKzGv?P@t7!|@SYo-8XD?V-SNWQdQU7N3oFmbbz@;Q zd{(k2VE%q(+FaGv#ywL`kU7J|FoIti^1qn5`;?+)u2+%H}VNZR=r+U#qbU zXJ>rI+pOAX-QMtrNkoa)%AWDcePvPbZB!p;P)r2Qw^U&am+dt5s$OBt7Lwp5)#3we zf#7JNl|u9x-D_SFnc{tKX!}fmpYr4rlGwA z@K*NTbcPfWNMuHfIk@Ode%`Ta@pyy?8GqN^uH=kn1hq^*XcXPQgDwYUwl4-EIiELh z2-~rrc=Z&;LwdCUc!sgK7@X%n$P?NjQ^hDI*h4U2LyAw3A1XsH9iAZJwwMSeU2^16 z^>oTvrF>CNYV+-5V@Su;>WjoloapIegh1UjcH{zDJ*OIT^~2y0(wUcS4uJ z8Qo%V8brGXwYXh4;1`cNm-^WXNO#2$OKc54?|lXGxET(rD_%+MiMiS3F>D`eWp^q?$4H~(B5Frg=)z$L;Xb1nS zA^V{q#%xfiwU}n;Up6*Gl&YxVu5Rp~3>^?%OT~yq%6d&(g0L)J%l5=$g`qsMPeoR* zT@3jsjgn6{nw+5o$7WEmnaz*ry<>IsMf0s)@uLMNlE<*mZPcqIJ zIX1UFyM36$Q0%WTyJR)9(A);omWq3~do*x-(5 zjzJ8)vQE10*8b`Z5X;^x1~=+v{MLgV#qTeJenmB+MsYKxH#ob{04^XF4_OUrVFTcj zdy=WSrPcbtMqYC*f$6NVLe@(inJPuEWulsMy=lTp=v}D>QKr!;#gzKQvq-)``xzOPdZwZ(dsfF%`r3<@VF{lpud+Fezznmq4_ zLGYkKhbmYelCt9Ih(M`d#9(S)5g&6$fgd>be#W6Sl*yM<6OBm8lu{OX#T>Gr+gqk6 z>N^BQ_>jrP@?G%gFw}^ih7t#7!mAC%!jJ^1t9tIhA~RZgBgS|=8M@-h>*8RWWdAG(DeL557V zSI5WIOxG<_^WcO_%QolR4;#$79BN*DwS2AH5xbPBuDKv6sNWtww5VV2!7&?Aj5Mf! z{&?vK&THVWiOn(UlMX4l!SEC-+@D#2x&ToRt)90=!4Gymh22`AiiUZGsMw#u0tru{ za*91e#ERFG;xT;kI~xPIbIq4yEL?FCB%2;M1y0n!wQxg^VrFCh}04w3DC zJLHJN4wF)Ci_#pN{8%#~tFwAnt1Eg=FV=-og{*qI?j%wfPyz-Tv)7*-z}84drE-`L z!n~qWpNgpw00^7l*l-)C=F)?v@9qc+Je)#wkO{Tz0v6Yitv;>`9K>IdqhE7t6P$QL zohY2vKkEb)&opRr1%a#STq7nn-jcnrExjprtaAy^rX<-x=M(*QflUQZpp07NRaFZqekdBdrKHAXESmM{<33nG>uM* z^anwk@=)V}7<>1kX{;^jn!L}HMipi~B^G9{stWi_V|S3*OI+9>FUF1nloMjLo!DcX zvXQ`z-qK7)mG@2cx*m^2Ht`j4ELYHwKkIuFH_CQk)0I+tJd*QlHjp8G!kNB+iBep% z`Ny=e%Q#_nifg#^MgUufoX1-hk5PIo($xnS<4%CY(_fEH1X%~j6JFi!B1Y`3!_^*d z`|3if;qk@tr{AbtFjvqCmzI}cAsjjS9pm}sG{w?zA6&Va_+I4ehCb23VBn;VpQ6x(_PRc1;qxs=V6?R8*po@gJ%mxX8}Lpbw4e0 z*3pi-&A>fM6Kx9;lo1x18k=Tot!@!GLMA@!-IhC_rO%4$4Cqf-(&OYX@gEQ0VY*oL-5*c56z zxGbU6mRZTrbrUuqGiEafOXy&9&aIsrS&idII!4-3=gB1id37a8Q^$4E z;7T+x#SuBH59RPt{iY?69gSeh% zoB7X4WmwdDgG`ZRnFR5Kl)duXyP@ukbq9Et$YEZ?*XQ*|A(`6DH=u*P6Q6z_HBh!;jKdqqqvi^h$(<6+M{q6mOb2UH>KG~&k z;DmYFg0M7DYm3YLC8#{*uN1~Pr{cv#t?OAXDxDcvDn1)wje?a$>ovUSnv3)Ad}m$n zMnv*XlMXhh?((==WG1~q!$N8rOESA&;+nkR9EQQt0)XJ!8XTQct>O$ni}hVv$HI&<*pSM z__M!4q`qJUn=6w>eq|GZRDe}6Rpq=?JL%v@5G}czy-7WIJxiD^|oc8q6lsW3h)8R%F>K~R3tcNBmN z_&N(d9SgOHK`|mo3f;*_T7bv^RDEi(NeP<+{a(}?n)?gUxZ8TJ_yvnI!ed9M&HIZ# zv3~Dtxa<6XP1CqCK;}2^HXOGeMYa+y$;0=Gb=slIhSQ=xUSkD4w$W&n2T3Xt>|IA# zOJyE-q=oyLWoKcnd|EV+P^x|lwULCezKX|rBotG&|( zlEUuj(wAp452$rOCXBZ%sHeSaD({V^d`e%OzSfsl2ZkLfuaGK;Peg(evwtGV{_<_- zY~NQ#N&QNE&%jC%uB|V`FTLIRz&jVN!CHyA=`GP=zFJN>z@Y(C{1T%Yp?!GZLlmZ+ z!?vyLYYo|S1M}~bWlOZEOI`7fG}HK%Lbgi{+mZPkMFO8S?@X3=>J?DVoYj4yli-;Y z%h3XFNR0XbuK);_Jft~#n!}4wU7g_fV%A-LVCB zZCG2&YEK(*a!M-r9jDU|<{{iRy|I-7--ik}FVFgTJKqW2^?uIVe91N}uuTWjOO~3T z;TKlAy@RqX|Ko8VkQ_z0Eo#@P%Yr3Qn#_65_59;oN^@&|p;60wZmMj{Wj4~ITPxPv zwk`6??|4tkp*iAfDuyf~bVntq^Ii-#@XK1W*pAv7Kh8BDZ)jf$_~;vuW8{ys=h)6a zDfWG5x#;gczL0rqr~}na4S$)Q6!7t_nV$4Z_bK#SU=u63+ogN?bM9@am@+VCq98a- zr+x<`Yf>*_LSf05B?3c zLxdHx9?XctTO=+LZOr;g^ zMr%#CO(c0UaGiz;6uh|#YH=wuYwP5O_|!-m+TY>6#B-|@cczUo#F}I+UX^F2!m4%5HgwXpIF#!Poh6%qNyYh_k5H^_ zEE-qMzDec|20A3hi2@VB`kdwVk?$D*>ii_1jl7& z^Ru|4&nxqmt%J~+QUoblY2`J&cJ4S`_@|0NldPEbXfT~a z96OVk6jo4OBbW7tp^)BrWVV}$2w$O%0OGfEl<+IuD`glN|1-Bh4}MH1bClOH6$a*K zgZa`hb#Bj$R1t?@cDKE&>UW=%S8*1kCstygtqw2_YCyi2AuSUom1rJ3p8jZV9ys}W zFV|SNjsNQn!^3Zc;@0(lk6E{xrI^?}>hT-ioLlp-DjcLWt@?G@7d5YVpT$S|pTV=s z5?HLN+&1RDQT?3LFLawXtZM zj#r>iR+zcmomHve?gocsXF_$hFF_y$8^t7Y##&YIWgTSHsyt2?GQ~!!R-0)m`}cT# zSkG;j$@?Ltx8d$u_)ZfO2FH|J^5hhDZfgHQX*o_{GR`PsHT+(Iy@`&IX_Vr#cYX}H z_O81xxz$_gEM&~35qR<^#^#)AYM+Ute6b)=zR6?$_o#)eLyAQ~pSPECS&HK4Ce)?X zGo6+n*7K;CK3}@W+7QOsGEw}lu9lYawCiOY%ie5t^LM-5Y)jnE6{kAeXL}mvZgOE-vF}Iovr~1&Zg+L9cWn}Qnk*35 z9e(3R`+Zl8Cyz}NZX4xCn@mF*j|*bqc~!-k)2m<4C_h-3tcn}%65ZA>z2RVRz_fw9 z6d_|Pzo_BQJ3u}e_k6cI`UsM^o^56^QAwF%kSk&ig+Z6YH?LG#ITK=UX=+ujvQ=d_ zXlx&}NvM~BI$q}GSij|-$dR)!f7Z16=IM@XR*wBbIHglp-rj;TTMX1@UQ>OSGh;K+ zpe63P5j}dMGozy6Uchl#nkk-r!QIeNXDnRkuG*h)?YTdO2{CebBN&8DAcop0OrC2a zF6RN5CbUyT6Hb1=J1-sUPUhLvdmh|Ubnu%I`hoAQDl8{k2=_mMxJMpnInr`;aM!jlVRIMhBRERPAi{E| zkTOrPXO!@}wqS$jrO(Lv%gp4h3#PO(vu9HpI_oD;ZL}R}^BolGAjCCd`7A4ua=I&X z)wNhV#|09W;NyJT7)RjYu+jxwHO5bMzN8TOf$p)uCkAe`{M0{mH21+jxuod_q+NA< zdROn8#rptIz~8BUOp1^>)B`@CtM96$(m3kkccHUbk(CQuw*zVxF%@ghMSUL5Y0I|% zsOwpQxKYh!Uy!90wa*|&>SL)oP=K;199(#O$gNivy;?p6dzw8%hFYBz_CU@LrqI2~ zN?8DR_VrT9D-R=c@IAuZb=p6<4L<5d?r5>AJIWUC3)I(vRvPeEs4Z$jMO!rNg}h(K z&i2G7#G6<~{4{5R39&=lg38Tn?+9n#nz_RI5jkdd zHS<|S9mPfW+-S-zc!#;!Kd3^3mvEeyW8+n~*K2Z)NZ9i(b3D8ofCoUW;##`ih594< zV=>Y{xmVjRF7_aBIG{al?5z5JzG=Q0!gFfs+-U#>*{<_U+}3au>z(`UvOoEg^(3?f z9b%SGBYy!uUfema((>zC(9+(`Za7ytfwvU&&$qHZfE1I6qFIuYYH#bBU;L_vyZg!i z|5u}TB##3T<8H7OqEBLp`Y9?GOQXa454}QqJ^=X%vuL)9V>FKD-bY47v_gTPUv+&_ zspn7`z>ZS@rb?b=LV3lhWYuaG*~6$YKHAsEj&N4|L8$53fR5aHeU}J#Dzozvv+^@; z!Y9U>A?zUgal-=C>Wu)F;@6K>|N7Bh7VVIlou`}gODJPCfhLbWxLM_)Q%h`+%U`vR z+%RS#k;x=aMkeb|6}j10;=4$_hCU6t>hx41xe|amR3SS0YN%M z8kCTh?ovWZ0TGl2=~y&K8FWjEbT@qOTBJwaXYc(x-`)SpeX)7hoTKKLV?6U2K$dXo zl19DHFnfSP_>D^^@E<5GcrngHsb&7c^NTI+iyZRFcXI%&H$JSvK7sz;IWwEXg3aZLMV(H&`8G ze2Pti>Q_0%9@|PrSizkQio6&%w6KMqwDwCbmXG9DJd1(XFpUI-#}P8; zCHrawm{|er1rhT84>FQRbKT^8Mze?EwRjm88ivuv0FF{!o(zMdfMnj_1j*i(nRzUD z&^;a4mu+Yz4-HW8Z%S$StD#au72draRVYM5zhtEYS1JqJr={DsC#Pn5 za-sA_TJ#YpSac$G%WGSvD`g9238_1q6oB?Gup=40beTNIa#Uz)EEVv@xcqzCS z)~vJ?5n`kCVm^J>hvADV^O&%rOmEj|+m`hxC%xS8JBREQ-A#BoRRr3~xgaF`5J!EE z@qJd??cJu80R$Y!L>^U_?oUesT=i42Q7(NSMt53#l}he>8|dJxD;;sAUG5z^pIEs0 z9Yx(bwQjBc`*cBQPPw5kUnk|$sN3Dh64aH5ERj7~kFeew(uK%egWV;t*8g^-e~et>LygdebOoXk9SZ?A`T0n6uqsL%SXFr&g{P z^rq(Iysa&2^#E`)=K0Dn0CK$QZf^4yU&NWIb=2kH!m~!ZBzT1~?hv$}+k;nBH|MSzSBlAG@_vry{6WQ>1Yg*k zCF1@p&8?PT-FD?=#fwf!ZAT}ED5Sq!aWXuz)q*-*{8n-mqXgaRQ+iyq166w@|6y!g zc8BzxKnk7j6i)nv;H{WPo88C9#PN+Vdg91U<7(44eUoggiNT|8_B#bZ54w=_)Pt(H z?sr`uM&ug(?fP4G-jEpwa1sM zyLf*HnPm!acJD?~r0i9zgEP3Yt7t$~j5&FcF;ycSEq4-QGxVJ!`mFAdKdcul%w%2v zR2N?hf8O0#8jm6&Prdn_S92-Ek{}t?S5up1hAoxdiOfL{20naF17r*X`E?O=1Zz5M zr$I1Dc3BnR?qT@JLu|KJNHI9{29(wQ&l!A&omWm)n>=B^7flar~7bF0}1+%_0}J+*JGDwzYYix zRZeOy(oa)I@4WOy9YipR?rxAAr*?;yh9%-^?KdqN`Fy{I6cn;mQHthm?qY%EMkf2mM9FRV6H^{=naK$`6 z%X1Y;@S-w1acORxd)|u|D_Hg`j(aF)7gGgpxi;v<^ypO3k^mL@C)PddT9MO_tqWw% zQ?gFvSBw-Jte$Ay`r>j4ZOh=IiP2b^8=6EYM@nlM_`H^TQ4WAPT9XlQv7)Qkcswq} zN&uZb0XNDZv_U*Z+>l@7k2)cWn7IW&ISG7*k=XCR%nqPpKfBFC=WHJ57`>v75|=E{ zRD($HxMKoFbJ9l_SJP=^_#rKVj?a}OfT9d! z=u)Ml?B<7)Fcz6}!T4j0;szh$NO>Xx`=(PX;}}AgVfd8(-g>OSHRiiwgg7=02x|P5 z)y(tnSVla^8C;sKi0ZCx4KgcL7cITT%=2c>XlvZ8n~ctlNzFg*2yE{dybPVHk#yZt zGp2dxA>*?dl;p=QBG|1^9%|Sc--+6M#h@vxqvY1YZqalgf(zE_op*`M`#j@;x9(~$ zsyQlU<`~S)5dasix7HI*ta{c#vftuGE1Y%m>f6b&$=qfZ)=5aAgiL>-EWDC(S+zN` z5&UmvieddD`CvHe0pB|f|ecv0Qr7=X&6NM$!zvq%F^6;(b_{b$sKtk z!zT(Js*S;?sI*@C5u6O~0-n0}T!-980`Im!cXGN)ZD9WOrw;)@CSB<^*5eM#-A4`i zHo5~&vx#@`*I77b`P(f4APDPf*FJw!kHMF$2c1w1r`L$sW zD4X2Z*yYXbmY487bb9{Hex@-qKk07O6@D{3YOL5x=_Iv-QV#Tz?fC5aR|uUBP`c-wL( zU9Ud5PSYWK3DJ?^X;ZhHt^t2m%M^|7TQr@cuF(nYy{j_{H@;BjeZv#Z+9c! zm-6aG0+6#lJZhz-MUq|U3ppjSsiYXUJNV{qPg;7IgFKqA z4@m$*i5ZA(Xc$7Rh@P3p0$k&s@-Z9OpQ5LTG9+~$NyeLoyM|*<--p8w)&8|N%lOm} zSB&$~ptZmihz5Ma*Mcn<@ZR23^xoEQzP+2mYtmo9%h9jI%~>p({@$`dWf4)RmQWP? zB3{=hr}-j6g{9r}bgO7eTXG>XZ;N|W0b_pW!dx%=;1EsY##iT+>LJ`sNbpA_anO{? zYM0FIhGM|o{?VKCF zH0EM~5rYmzg3S=f&t)eo{H1e%6yVsTWK?73XDlNGall)~EcH784FUWj+)r_!zsgLR z+-hrrd^R{JrMp&hi`IA%;a0F?jX(~+2-~}wHeJOqvni(=xf})&0S2m-DP~`^h$axc zJ;B9@S8Lxf@9o)74a*&)fa=d$Nl>5wteykAIpctHK>ROAKUzhsbxAze>1$5d+8%GT zW9Y^Qwk`v(1%)2B3T%gk0HPZucKu>Xn!64(cdQ4OSspQbw3H2}0{2hd%50R2)x4hy zu^yq`{~&UivkbMzyVcJw{oAW}xeLy@y`IG==!&{*b;NWwdaX{JHr$zucys*JQ1nye zs)@`K<^&c4>i|lF=gli*Y=z4&*Gq;qcNcwa?rkhqERFAdS_`Et~it0SkN4!-u{$R9 z+2*-IX>O9>;ZE|jLm@}kQic-q%z-4wTD@w$+NPKoBK%b*y&9Iw0B3U6rYnRYJP2D_ zLAm;-zi&2NVckuA6(iQ|bu6sDGUr^6*i8#EFfx5V`?&Q^a{6eTl$@dtFXIlWlw2KQ z3iiOB(tu6;6&pz=!fqNG{w2BM=M;se9hQfCYb%68{*n-61M1qBlut+l<3nr<4cCXB z#Wqsi>or=*V-7JfH$s`~ZYD6ukDTPz(F$|yCiC-~+!vUQ^P10lLp9ylHQ;TS@j!>Y zoV%wMEAm-0{Zo3$hn}z3rrOaL?1nd@n{!CBiyF%35z+^sFai}=+QbG_`r?z#<6c5r z%-%?eK70j<{-`|hvby`AXP3ZUYCNQH97e){)wfw;E;c4lbVz47HSlaWrDDP8mS z=RU%6zs%rovr)d~#yEp~G7`F;ux;*el@zU34O}5oa~aU17R~y~{Wtb$Aqfhdf9)T< zUMwr7s~4}=tY6kaXn(M`nvL>y3=Eb!y|$Zc&5cg&1H1xGSTt&i@0FJLr=<{m?5y{e zS)_N8ZFr2w(G_#eHXfILjL$i|pxQKuWvqy3{g6^l?cv9Y){jdC^(lJJ`lIeF95Kx~ zQ&RCrdeiUYGYxgK)`n{CmwERZJ)11QUBa95r#i*K`d>3N}6 zX~j4V*VqZWiPKlL=kyD3Sl%z<4L-V>TO){befmM93tIFIliToY5}Eg+YKy`*$_kV9 z`wDR)-&(somgrOWF`=amyb)M#ywF31*k2z0);fixamHa6UDc6wv#?Rd`}>2D#40|i zA#ajU*0Otlc4G?(P~3)Gi&+BB!Uh*pp&C*WJQF2Aik1woBQ@@I-ng~j?zf18_R5=O zh*6zTp!yDprW&H-T*=~;<|0kQ9^rby4BRMLs{t`sE?|v z4flkl+VG+c$f+iL-2O~xU_p}g`W4F$mX*{Jtq6jLtG5z9hS?IonVw(usU6O)h<1d! z_F9%Mrme7Tj~334QdpFK4zRQ>5TBlHTl|Bsot3EgF2z4Wi85JW(<1M^v;5vvkUsRF zOH;x9`uQ+6SI}|2rTr75L|kJ$I=c~%-S-FW&Qu;s?}YTiC){6ASUaJeZ2tQfb)v)< z&I=ejsY4}{QRqF0yvy#~kKtrXGhTd)!K0~k;;F$Jpf&eaXvTO@6LtfmK~9uUy^W*L zrnh~%EATOz*G~b?3>x67S8l5w$p9|(x`{$x2m@bj;D`D!?$z2b?H_aZZqe=3*b<=| z>q~5#g~Xtz{xEJ4zt%llHk*y|Yl25M~JNo@um)QhDSj2uR4)HBYwA z*zCn{NHaM{!~wp}_+W?BLn+&Ru*9f8-lRl-%X`8TX}D>nxCW~BrAgVqSe&BTG*gTR z5a`C+MflY7cg!g|b4}*h-x}tcnNOpr8yQFox2MV6gRE2ra{$n-UZ)Ncp_cLHIU>hh2bu;Noldf^7UawDd8u zg2sk4cwT9*HXNiw52&fW-CF%>n>6Z9zItF?q*FI zu21{%F(y|9J;sz?09tMpA^DyoKiBH0WMBr}bRON&8AbzPw7{m|bKxk`Nml5mh7TB0 z-wwpxg@LI^4* zt}4TKweT{A7}D6HNHSyFmxw*ajcJh5i!Bn4^6n@7xrJ)E#nz{@=tH`7;OLmIeipn0 zK(^%-_@>{9Y3)gplEGth8S8uC!E_D{Tq{Jn5t07aKDJSMwBYGr4K^pHW(r=R(M!qA z>#-&1b155x{4=j$TLyUUTx~+1^3FMZ_}Y=-`Jpn_iXEnC@8W#;j z0)F*1SOIQx^qoLd*=09=gwq}v5_eTkOCI~<04$YJ%ZJ+sJ3Bj<59bdjiRBucnBo{p zQiM3YLqf+f!lkWeVkUs_?vLo~^WG+XF?wQ8(59XvzA%nUph$T8<5I8OyN-J0p4=^k zVi8_Eu)WnBLc!P1GIKC7K-0F?hUNj}q^#O3U~i+tlZbYcmx^0=jwi zg5WZ(a(vsX8pC3+M)q$~nhs=FGMpyYv|;JlIecU;O0KI=OW_kW^&_@=|po-2Ri85pkm2YYm^K1u6rFPJ%?GlKn*aqs@u*Ok2ehNg+@18NCks2CZ>N%`{RL@54 zc*-ojuA}HNn5TgA-aE=RYV%2ovFHv@-(?k^MZ^RFp)6-Gmk1sYA?cQG#pIow&|Vqy zm?UOwN7q)aD`k-{i7XuJCj92Hmzj!LD{nOY=?ZhWtpZ8BEv3y)oAZ6MOA-_ea6pX_ z8=HS81TPgDJ=Lse(K~HyJ8FAF8{0($R$&tb-%;9gRuSahD?UOFm{SN3oRBFcR@lH03X{dsEzAB~^RG`(SDiV&ND5b;;C@#nG&RI1ZrV^m`VR8_LtR`&P3?6I2? z^k8Vs&bMb?dNC%`_=qv(Mfe*D^FCV9e#<;TkFJOTpkwTathl}W{afXY8vV72EQwwKHFV3vQoQkD!hArEs__X$T!7_#P#gqC>&dnwh)WKQ)@wB;T)lR9GxG~{LIQ{>_OV!{<(x|cwH2Ti{+XRO z1itahoeg=l!4kGw#RWELF`xc4kMEvUfW~dEqR^)VV-?jlW%w-Y=RhyA<(2H4gc>?7 z@sp7|wi7Kb(}OFN`{M8cyF3;9K^LrT16NuvNro^j&PshvSemJ$*TOVd-2Qws`8^)v z-b**SJU+1>^~^5f)lTcQ1H$b};{z0LwRo{s=EwCv$%1$f z%k?7jM34K$D0y9-_zJvfYV>dBX}URtvdQo*RoR$|PD5onP?Q@K(BY^ZK7t_8xis0~ zxJ{YHXgw|9n)4Whbuelc;!PSvVqR;PQhvNk&na;=4gvpBEgC@rh{)g*D_k+o9kWj>&|uIECiYY|moA0aZ730?qaM(D zDHkUM=&9PzddArpWY{@-(h1)&PId`^`~pYZ_U|H7z^EwyuvTZFDRcUZ75H^R#3`Pb zH(R$Z^=f%xCcz~Zz3HrA{hfjI(DXuLXAkY+`;srT|bqZ5G=kCuxxxf6LJ+S?WXHsh@(Glgj{^E6 z0CwwdR{xJ%KW!c7zgjM;)uq!;&zYcj6eR&ZLejhk(53;y@6>uz;|!SW{QtIhoPzv6 z?LkXY|Iax8Ua0v0!}p+d(FVwa_y5>yr&%m}-u=^%G{y_%Vf?2$x4(SRE)!O$|0Q~! ziR661zfGJDfYCd2MzikrfPWM7MJTdh1uiJS`6e>fIJdSdqg4t(TF8IQn+RN>fB7jF zV%PoU#>B%hO=ZXlI{M^~m${MXC_FLQCAmJE0sVwu{#G#5b$`7l(YxwkN5ygbV44ik z3-se8B|)a}^j#*(sY0djEEXQ3{`k8L&vV8h(A0bolC{ zHTHko4y29aJ%n;f6@Ms^(59w+!_6ZbYKGv1Zs3u0npzhC+QC`FhF+?aV{J9 zm+4RG3+C-Rqa9V(Q>e4NK-LOy1GjN;ErmF9eu(_T?_AdOJ}jYKaBcgtr>II!1Xfi5 z4sMf+kDOWli3pJ1*4ViAb8Ccs1DM)V>QMiI9LO%xJH7daf~K*Z1K_F!{)ygLdDauL z4n09kPO58Vx%$s#bDb7s|C~f&~g6S4P9CC-0{o8)KdS^sn9N{Un|JZoJH<7K} zOJ$$`Cb<~Zv%Jc`>32aTDGgG4lJNc=bERAIbJI}PUM&Mf!VDkB)gZx1SNVd;A^iaZ z;z2x<*v*Fj^uWJbhZ3zA=)eGXCDGlqi1WeAM)c~chsY%WW{~w$P=)*w9ulGh)dl4sL_%ix-lMV|Byw{DwQ1aK^^k@aZ?#` z^Ras;x_?|JV%C>109b}>f(_@JF;4Pqx|=ryR-^{HjLUXNbgabOAAswhCnk^;Lu17e zjs9@-?`+m)$(1&-X`kH#uJhK!)_q|iTKdrRrrmJ1iPFc6nU2L|XZdUWWn04$M1VCp zZ|l+**fjy8PZ9AzCv73?x6F~GsNKewxX z{uEta#B7^`v&+Q-P&NO}Yla}IpfXwLCGZ1qE!3XQ!_cftGQ6kL%vDE0SmtKw8{1#w~Sx_{#9>pLV5xxj2(G?ySPU zFLrCWh86l~Ee^Mejvk_ZjT!sfJ1Tb%zd?2Wqeo9rH2VgiR!!*Z9m>BcehmGCqCI?F z*|`-dPbyGH0P1rWvCkhqFpc=*4PH9MGwtdk@nr+I?XfVZu^gosI2e7f$zi`sGOCSM zva7HaNKq`J#II7(*#oULAd4s$D?2>j#&hrnd?PKa~CFvIwC@M@JYNHV8#_uvUgVVP6HH zM8egG{-_11c7Oy7xpbWgfF21nym+kjU`zf0<_Xk^76NZ-VSTq>W6aET-KPqlnPjil z`7QIJFQuiqb!bS9!B3b2kBCg30*0(@NZIWJ6qC<=X*fn{n&IsAf)w zgq+sp{c#_MmEC{>ulrg`eHgngIz)%Ft}!%Y#1wTDt=2EepAh}&l03S)3@l<3Sj0r1 zg8SjLma)cYM-HL`5iR|vmrwvlbf53K4$$;Fjs$>mzYh3~84uV|XRe1N)G4Ai&%Q`j!g$>)TDr=j^TsevT@PdE83)Fm2Q#cX15^j42nd9Ysk z_!lb83+IEJc@5MXaE`srd$@DO1-{1wekp~~x+s#0&XXrlO#nkx_>amI z1@?)jNVh;-MFzG6bpX$%THW2Zb;Los$8elECzW%qBqyIyJU#3Kx(tBy>L+92bnYr) zJEcNW#{6%`V2&XevdV^*2K~`|DV}o{GsI0++zv;!0x@HOqj#RXn7S{S;DG&;^Zv&# z%1j3s+UdAcD!3Q4^5YL8)(B;#Gh-rK5#|^Er7AIBih;IL;nLqqg33!O@3r^S{DePf zw-X0>+EX2S-eanLr>eX2l1cEXq(yj2QQvDG&p7S)E|){i5~5UZX^O#-9&_GTH||?f=Vg zr=-eY9Kx*jn3Yia!4I#MzR6I_jAZ!ETYQ3eV9U08cO;7?diqeVNEaURp591@&7X`+ z))Xp%t0zWQw>Bg|uPvmiy(r=0?lynR2@`#PcLf6}eSih+=_kdk-wlo*l^szP6}mZOZKR_GPW@H9=%=&BqPbzoP(y4S&A~8{q@Fe%{b}*R~Q7YF?ax z)(sJN8D#R0Hfmu@5)Cxr`>M*?U6~y$2o<66xlDo2v@NI^f^qqeGV|k1!rbtu*h$G* z#bC)#KT33qXbyGY#TGfi)6azYbu>~>-u%9T(m~Pr)wSO>a8aHrm*ROnQzQ6by{C$L zs;Z+gkwZ1_a1FGOE~U>*BHwkeCJx{xG(YIAI9V|3Fjo!}uPi;0zLB<0^Vws&Vyx#W z;`xXZJ@Ctk7538AL%Zesldx=oH7#=OGHfygQRW+Z-K(!YX#BPeQjLV=4N|;HC?8o? zcc+%!WTS)%2dw%G$r~7{8V7b=v21wg;U{1jdmo#C*jp@t0GJ>z+*v&#k%8f14xJKm zoO;Fwdx~6WsZMkIz_$}x-Q!utF1}#iP{8p*?vU(BH@J{tS%hURiQUjoKJ>>e7$oV} z>`t8AlX6OEluyG@B9vyGIub8AP?pi<20f+^?seDg*0d;B1FbCWr5Z{PcxE!-Nm}jP zxm7@*b@uq+IDb^)>#+-8Hd&WF7rWSh_A_z$kIOz$wC(@9=0FJf~PF0n@9O>8!w6$upq;ihZ!;GtjkxOq@UI$B_mZ3Hw z2$E+?fgJ4U2P$sfBFiLWINUaYw~BEtA&=7zpB*4vWrh0+`Eu8;H|s!a8IGkFwkUHp z5m<-x7p){VRkdPVx^nRGwr^=0$Q@eD(|_r=d{or7%9{FP=5 zsHHix-J5mHG%s)reatMZ3h~4MnA&zniSdv<&{Ew72=NsOexMaH=qKBo5SV&vmSa32 zLi$aSu3DweES{=P_N3HNCs1sV0pt<#br!%5AI9|`F7L0K>A`0P&ORl;L}nMB}r`v;kdA=SkrSAf|c$=W%DPnQLIoyS*x zU%dlcsneiX;lTVC)>StrJdP?kRf2KiLx)>=KLp>$G`@Dhh<&RQ#Z&A+nAl@ZpF=FAb3SPF9jE z)gFWi1)**!`U_C|47e41TgZ7pK~%3^tIfVP=FC#ojPQr3$Jh_1>&by5CVwc+>rUC6>8;=*slzC0}kYOP9k37VBhwE`cBpOK#B2JM+WSeE%5 z&D}YNJwE9|LWxdDdEhBv7eQ9^)tPaV9|t=TH#Hi~tG>eQ>unk&q*{wCTD$Bfmw%Br zu)vWhR~{*6c*D&j917vS88ORu|7lf)dgse`?JQEQvl_J$0!dk@szG~7?)h^yA0+tX(hY6JLuXX4Ms z1#hu6WSKpwq6>9lf_QD%%Sr+!XoQwF#~Z_EJ9nP7aN60#^Dgh# z3btp1yqyLNQ*DQ;jJGPXbXW2%Mqg6tG^@@;n`EIk-#4^r4PuB7%{qh! z8qD;cWbWRD!^Yb9Ccr>KO~YC@g|zPzA2D3Wm&;GjK{)9%#K1eo77^jfwK)F;SX8lg zC179EZQI$0W2ub)7aPcUDRYgn&OqnmAQKy}axS6@UA9u&&|b=I&l=8z8Y^kd&VF+| z^_!J6kt(<6QMCF~a71M+-lMZLFQ!!Il&A*Ed#1)do_Q0IUfmg2o18JOmvN3kdfJPIDJT~kIwK=c_lO$PPbVpxKCNZojcqdj&Vt! zO42mXa7Cns!I+!Sfrvo)1%ibPyXg3$siqe{yUCrMLHppn60;D_JmHIal-hS3p~Eq@ zCEQffQjeGM3h#dkQ~KdYK}8+cD$hPOZD!mX#~6GcE#VC(_S%O@7QKgCrW%j12Rf+J z>%QMzmed(`);)3~3R;%W98Bj=oRh-(4g4;KGaO=VE7q9iah2~RU9fVGPjkbMEWF}? zLp`}}@J=9TI9T75N(>O6?^H(DkHNP)Jdf-et`xS*v3?lq(qWsL`l`Dt$Dl(tKvHf! z2=qtKH2_-p%?nmX}g4dNJ#+G5?4^-*aX6X2iEJ&qw+sPO?mz^4Gg zX`Lx#E9T?M!RcY^2F_rhjmw!dcLD8up`r{TrhT|NU~AKrmPGMU^~IL~Co;TlF_5l; zg_`klT5+5XwodB>cKcb?bfTn%TF2X4Uka%=gL4Y@=1O~0iX^6LS+b&|kE{+ESvdc< zPV!wtIr|JGioq`F>j))*`swh=$j}gZT}?ILG3>+EYiPZwY`K-Y;dZnr@2lKF@F~VV z&bsmO+9vgQ5*Xf_coA4L2Wb+|b398&1{+u}#!im33`$`5FqIY5%oa12uzl-*pQ2>wd}KOKAuh9CzD)cE%ZMoTe!khE^k=0O+hyI1@NRhMpfm2q`;-*SAQQ_r=A}(I{7SM!mP8r3-GKJHlcf|5}qbD@xtt)pAq0 zc->ncA7)fiMkZ1l@#)>Y;x|~#(Ot#P#RuMFuzpd)Z&Ss_{*?S3)-_`47J<;!htWpn z+*0-Ak+aWF=+qn_({)_5F|3<4;an)dAxzYm*RuIwZuoJ_`&}72p z)z4`NbGD4WzA4pIkVnq`8mP)`d8Bw%nteZxWAWOl+7MMy%M{!Iw;8zMsq3^C3m&szNw zq<5|=N4cLy7Y^JaUb+fOOdp5lU;rlOA6E_^@1@{jsx?)YJ zb`h)J*N^pK?@NyiX1DU=fRRPvX1_?yW&4Ycvw3?*e`zzy7w{1S{3u$5+hP%_tn>@U z`(HGEHY5NEY@WQD1y<|<|DDrB_{Q|oJN7T1%u5}%xAmT>-|o~^{fuppaZ5DJq?CUp z%Iy6Uc5r4hhv7ib`buRCP}SBknc`COqn^H_OdiPO*4$UX?qquxc7T-c)HwmjGC2_P zEdI3OxdZ_}ex}^StMkd!sd(a}6Qao@u0ueaDx zoD0ob-&_yL%?J`6Eo-JiuGi_RF-rNf5)zM8Ke+UM%%v-BB*hW8ec?O?P!Ue80 z+#9COWwW87$HW`xcJlsUw7vu!Zn4Z2q#3QhTx;F5MO@sON5 zWII-?4m(^Axr=lY+sp1rJy7{Or`kPw4yx;exVy$^_uOpAXwa!i(r0DEOctv`d{yx; zupD}R3*0rXfMih3Tt%T`^NtRDkrOg^xh0}IW3u{dccdP(?10~E$wb+5Gc?#)n{7Tk ze6+mq9Y!9)^UKqPQf$V(`lDqPkPj1|hA8s8!C0xL_ljPpaunrCBq-$~XmXOhkU$Zo zvToaN4I%R)BQ9{3d*VycdT1;1s`UY$>>Zt423niwz=|};pxdlq+QOITrLH)Zbe!q) zQPee~48@(M4^{R35D`r~w3VKXfdJs76y=VX;*iw+Xgv!A3(n^!&7 zr;TLteyRMeuX6x;C=U*h1)%N?iM->LTP(jw10|Mq!Ud{zu83p+Z{TPQai3U`$aK=v zvsVHR(Y!YY1Muo%Yfn>Cay+HLwE}VX$CtlMnJc{2MCeGtP}AritpD-SF(9*T-&49C zfl)Gp68U|paZpZg3ZT8*lAJ9>yj zmdg3xo-HWu8o(0?VNV|bCe;3tEhmANr_SmI=?EZ7fwm`a0v~vnpmiE&G*kAK`dUSC z*x7i|>iCLfy(pMsp29g`UKz;iKD;}epGBZXc&P2(_H9^~%@!b%jH7tki@C$OQ|}QU z#Uf@?UKw7E?~a!FV~Hq zlnSv`hWwCz#2Y8WZMWKnUOMFZK?vg7GV-Q%Xixt+i|5szH9TqE9ymW@LEEKq$5}~K z6q1aOX4F9TmrDeW)@&=-q`bFJI_$Xah)m7?@z#q9I8i0Lm*p$@xkZ--C|9Lb-^!RT z!-U5Xr3+66Py=m}x$dnB4Z1Qx)f&k2ccmDgLxH?d{Z4(sqP{mQ2*LNQuD9tJ3K8qG zX|ECj9Q%?qi0i@rqU-&GeGBW$%ksP7WMp_bnmp?xsT_a%CswH`($1yDVo)4ddb@I)^Bzo$l8w&2GGLRg9 zG+O!`6ySR;6Hq(mk0_PJ{mp&lKBmRxlY-hX$}XSFpofc!utq(4pC{Lt9{ERXdOi4T zZ15<<1+_1WAYhq058!;(YhE96nf?7`ZWOI0XQKhg4@wuQ5dr{DdRxca6sV7~Vj5s| zvAW~?p?ZqAyB-WI$-iA?lvX9j2j}Rx0~DxHzI>`g7<)y=e1BJ~nK@{N2a<=CgH`2P zJ3c6lI@cAHN8hlBa=i}PBJj;~Lu7UD@kPB3{2Ef;?ud;x`DN$~>mi5aG+{bw^5KZX zY?y;LK#6U6>JKG-ZCQ?=|D6`@)OB3Nwa0v_N0e;Y8*rhP!{-eA&!~-?G1RQUk^y+8 zyYkFoVS0{gZyg%XY`y_z~2$njJ4xEz1sD;$Nq!kB!XIB`uo znK&r8W|`zlJ%T0$Jqr?uEquTSR6}s|#ehRIMVV1{U~fEJP|?ucu{oMo0ej|ms?VTB zTBfCa-w|9ERg+?TX7_jzpiAmCUiL;!W)uvxo41X?Ma)#j05}cDMDJUbtUrnqJ@9`t z!gddOB`M+R0z-QCc3 z$VErJR69f!hTtIJOtM7nBsS7Ner<8qBRE?l6hnAu8Hzm*`)<%hEpk;W7Wz>S_(STRfZgvR zQQ~(-M-C403-r_1hLZJS48^F3_&YM?z(=V-pqj%dd137w8 zr*iaT01w1tr_iGIWww46qg_jQ>87FV`DqJ2I}S3^7ch zce&e^N;CE}tA9WO^wA3L9skSUmZJbJt2ExfFOW-bKk$ix274}y-8%R@Y6HhrfxY|w z5sn2IOj{31TY7lXKlwSw(3yHQ_210qv=y_(0d#jY-n{U&TbGjQop^Vp>bmg9WkUt` z`Qi)pAydci6>t8O@Bvi9kAAZBqq?LoFaY8hfN}KTu@2mu6&fU8OK7(+vdxDJ&Y4+P zFEf$`ex~$3_3nHUyt)+}Dp}hxR`utL7eM9qHw;zshc5%PLXqgrp5?4eeh7NDp3k+N zuHf#Z;g%%1dffea{Zm2!@QI%D{?j5{9WY2@NEe2jKLUBd8c7XrzXh5ApvEWNR|JFb zdMBnOK-R#bN&h#C{@OpD1TV+Uyx%uLBMIuZg3J5gNJAn!?ZONvmzILNFAtOORPyhu z$^hlU%kjYLKRr7IdTr+dA?&&As4V35g$rd?(qbaYBvv4xSUvvL^8Rkz6Uih1OzsEJ z>gi~>W&Cf1{W?upDI+NAV4$E5)MU@``;s$@%^jP6;Qw;p5^)9qmULaS@~shSGJNvw z|Bc`HqSj%%0?emexm=nD77+nxq5qo&6StR&r%qc|;Yf|iD7%Td2R)6M14?wo#TSbH zC;k{x>o=d|;FRFjnVaTzJAzjGedeq|p3qf#?J@Yl|4rPwoCN(e?VF76%r8@pQbW0QdQsJCvvIcTzF;31S(W&R!nd<9-l<>-ZCF6m5a& zS8d{mY)*r@tLQQN6X_jmZ{@ENr^l?>AaXferO!cS=SMS1J{aKe|Eu4#tp5&yFz;NW z+c*!0hCC}nIZ2Do&Ns30r4BDp9vPLT8Wr(u)kIm`S`^6C@J`1Y>FfG=@z-Ls=R^N70U19C_aq!Hg}lrF9o@?v5)6iRwYdw3~Dq(3_qa*kYz{ATC}iFiM` z)dqi@N4|MVN}iYCq}zXaw9R2)J6dd>&A1>4d z914@{YdpMxv3Ml}X52TDclqWR*tn+V+~eH^Y#yM?rBTr==(5JX+)R;E)jNrWE3_W! z&Sudz9;WfuxSd$OylqDqo6YE}q$%KUeSOPVa^mRC85YKGmiF6j4Yv;=?bdD+ghIQ3 zmq9r&8v-&Ba1LcXMGtgIM2H1&-IPCJ3Zj!qTxc3vj4^{^KMIJq)1)VU-OJJSB;A8l z_A_J9k*)+hedwFB9}r%@Psrwpbe0kbGz8tuLM-Fbu!Pe5O++QtW!_b zk$g5vQF&yDVq^?xk|{I*?dDIfT|^8OW)od{0r)WY!Ts?-abv)9hZD*wfx?&q-0IdH z=IkH~)@!ehDW3wohNf%5)y9z;J#Tm2p`!`FmjU~!6@lbGGB?;7v!ZHoq4(g#6TE_% z4g8ZxV0s{Xq~U40jUlLifd?vB3G{G@f(;(XRfS=67MeB?S!!w6e>vjbNepz2G_N*&d>Uf;7gsZ z;Wv3@-t~Qszq?uI01tMsI;s5bg9{b)AliQ#j(eaKC=k`=w~~77e%os0wIcMdE4|l_ zfr<*VQ2z67MU8}2f^(?=NGldl4abN2hW_Ob)3XdmyGhfAmA5_0Xd5?oO%6RK;aKl) zzp_bYuD-)Y1R$q?B?u#fPkU>?isJ!mo^T`~Aa)-ulfklL|AJDALL5qfcs{!|Nttr+((-3lt2`PHu8i%gn%TOq?e$jya1cx?DvYP^imeaoCrFFJdyv?FlssYA zk@xn6qpQ^bWZB)e+r3_<{os*=X+PA#K)5BjvR}mgi#x?}^xO87uSHINMX4;uaT{3h zd2>$Rz0GEW0rE1+xtQnxM|-eJ#RFjX_`mp?36Nl8F}cs;5H4P1`q0~vz12IjU_Fz* zn)AS%<7VqK2TCoGaen<*##6pdx#GM||9ljk^xr=!i+2vjRw)#F%D(s0ygaQ!H!T@g9>Z6WEo>E8Gr_#k_BH~v0#x+B5&R*Pbt@s&aewQ1 z+Nq_WhL%E}l!T1(I>qUuf!5S}amVr-rmvbq;#E~e!7;=IgBSBf{#c?g9N$yyw7HPq zg-LZKBq~DDvjRxv5%mh}k+Gttk?~DIW(g$BOZlBo{FmM+XeOYdq!T;R_cyoxyvwYO zSwS();~o`=6uYjDbq~)1LiXj~L^Ziohyto&*aLhl1@*{So2hF7Gzpf=m`a}FIOE?- zTFr00g0EY@FJRr4htI<5^&_qTE55llj&%K<-nT*0Bh}a+!Q%aPhRY+}%86JHlA4Xs zv22tTzqSc)G~;VYEc4mYybBA61u8x0XP8hxmZO*@L{JJg^%+2@hcC z?)m1{>p`YKhxB*fkb`FuXT0#RO+loi@=0D(^7@9eidJ^iTDU*pQk~RFXi`&sECf(c zhdN1ACzVe)`oo8a6vR?!#c9CA=K5z}8zDtXzK?mCC_8s!OUW}mCPZX6JG!vd6F_q=QW^KPnv4npAK<4=z{q&w2(3Ahnl^Xag{!#L)bXDyJ#LSKe!YGVK5kd!ZoP&3H zdpl0rc!`f6muR8JNuOcS?w#`*jU>vxV|gK`4=(5ZeFhYhV=}_X-09)S`u)g&L?KVi zt@>Ku2&lJ0cNG}iR&M#67%yK7OM&dW+9C|QbdpM$3M`=7VuN$VT{qsEq#G`E|I6IF zan}2I3M8`5p&l!|W$^0YIB+K2caZ;Ffh$&azzI$qsocf7v_rh0d}K8~-LSWXGcA8n ze1T_A;OG)f5xu~|wYU{S(1ZJJ#&TTeUOFxjwGuem@!`0tLv1bqXrrGeBpHs*(t|bw z9XQh}*t)#pKRbf?`(ZC~wP9izR^<+a0Q)h(BZZ3IjEP=r5bu4?@(OU__dst5B3cT| z!nQ~^zGALcZiB`oDtuwy>E&pHGk<@XyMy7Ex{ShT5>n}s=&tKHNvW)QU?RPRxL7)Fzh}>Fe7BjeF3N46MmQVU5AGMb2zvT3JLgx** z^-00J+7Y$Eid&`|QSODag0!5KWq}Nw_&ZyjcT4%(U{q+a$snHLLp4v>wFwNpdMAJN zM_N3+ytUm|A~6YOW@`@6r&cLtp+c$wnA*R5$cx8>B#+;f8XzvB$=MO+y+y>@=5Z)a zXSSEiYRilwGu>BzlPK-3t@`+#mL$sQpI7GdAW+XF8U$&l8{mu_hwrQsYm&zOon?!9 zs6l>kDUl^-;=feBx$MmHc=@6;!#Mk`&k~MQ&dgND(qHq&l;fEc?Ik7ucv8eGP&u6s z+TpAOj1nYxU^`xLpKdR(W$_lnC=h{nl9EP`M-ir_JCfgo+{H{;CpjGSpZPs6Z z7y+n-Vh$3m!DkBspDjO2660S#tNY#)`Hw>if{pK-#Z9+Np|Q}Xm!MoS-Wdd6=jd?Q z-|Vj6qRs;kT)oWc0GHa?YkMF>$8)qHSYbO^3q!8^6u4bivLc-*(whxcSrRBXr=I`F zVc&%5^01D|klXR#$&8&ynsw$Tz1C_8L`!2%P4r+lPlywYX+aFLt|zn>I0AhEEP(Jt zI5t{=U9?dJ>oy( zx*U6V_z6bEEj^4^#TJ&l&KWUV2;~oi7)g%bfb!N!ecf7kpJoTGz#il%_|@l#D$SO>1^FMyvgMsn=KIvItX7P56t9@%8q`CHHqUdW@rI$ zdX-x`bcb;;%^J5qy7@yJpmH26Ia^_)g?y|X)7_b$odRFo*dzMwc~~H*5#9<8#R_v# zA8=nHB3eg>-4^ed3}>A`vgD>0r(-rALV<&0xBHu+(Y5&v{o5yPrk=!Mno}Lv(1Ior zLa^^TO+rGwm_Kn(_$5vf29Y{SDKIiVxYO2C{1}r(;hO}mCn$Lpj#fYSy4jBts`jb2i0mvt9qp6Q z^*&h00tl*ZYgKp$rCrcpo%?~z2XsQ-j_%=Kyi}c#4RXnTNiU~zuhY55G^^iwEWdL8 zUJpMoO$XOjf)R3f9Dy@APkUM&>Jp8dgms|yruhHa&*8^K>`a4^gE9UAcR4?>0`)L9 z{gPU@-^RB^vrr%L)&m5>KaJWR5e+Vpv^Usrw)!BNWlHv8bWhsvZo{>g55N}(YEuJ9 zd-VH14)i`vY#$r|o1Z$gYCyWYl*e!vy}~QnE}rL3vkKN zdOKdwB607Igt6brha*hJF6`4E*G){E++2np6IHsChC8WD>and?y9V5S&ll+NYI`D$ z4ms*Z-Fev&f}DAQ@?0QEC~+J}^=_2$xyk=}Yc8GM=ZB1)o7Aw}ou!35!V?n!1sm%S z*P*G91J&MlljV!%OTUOaFeVIjJKi2GQS^r0E-48KsJ8ey3v(`LDtU5tNzC0HBAU)? zGT&wK!9B5F#y-A0xPIR8=gqiJ3JIIs=V9uJ+ntP@n0{O(ukQ(j*J5+3s@}Kn+{@m8 zbM-N3>KLU8qL6+%3J)U4K2qUeu!&9y-$(8kZLu$FD0%X@lxxM-@Y{~O zzw8JTrQFlSI+`3DIWKaLBg_I*r;EozUU{RIi)!~702XOz2)V(=rkBWXdoxw=hfMF0 z6yUety%gW&8j&m^Wg#-)eL^PqYCi_}#d>)Fmh-{gQx`_HXJGk9Wc+zDsIxD8evYzf zexv`y8f9d3+S5R{Raxkbu3K|2|ufO{`i<=50#MDUJ!&Ra4nZ&#HW{X<_RxJ~8 z3G-BH4vqEC&L!oGU=JjW%s$soHI;g?&`eY^bTNOnefEi!1Bhyl`n;PJHN7FqI0=YP`d$H@BK1oI-afZS{9^`EVk?>rBgJBP!dp zdso|(vJ`!E(*m@lN@6{ymHq1#m9m|D<2+OG-1ikFhAceS%u^tj%%|1+@@tLcqIDbx z@vhS$ds;KfQb_qS3SAd7=x^<49xq$G$K?v+NFC}c|J}qDoHJ!{qH~PJ_AlY4iH2e_aov~H>u&@8| zrO2gZVFxP3Te37^+Y}gD5 zoC#b?{iFhoRc#XT+qQ;i%n+P<+D21h8AHphAL;hNg6E0=BX?yqd;DW7<@0B&R>N%@ zySPpMt7Y|U0{iK!Xs^Ju8xeu%1<{jfGCsGSYe%Nhu--q@?^M!bm&;|{FrF?tgKNIJ z{T6R6%4l^0U4BMDFZW$JU0{#Ib$mgxO1r2W;|n4}ruc2TmdZ$2zNs=;IV(cw=vgOlG}7&v9{@e~@W#ixm``bYPDSA|*L zs7YMV!!p?^!yO0e7gwc8X5W?#WA-ayhQ{{4FUZM@{!(C=j7JnNk?+v+7-MY;EB19x zZ+BJM`#&wFMowV2ehj8r%ZIx!RiVSmlrjUq(1(>rm|dNI0Hu$xc39BXs?C;z(?= zBg*dmkuaGom6CdWm9uH>gH%IDr7$xY_GdfnYcI<6;Sb5hAt-jFOk9QLp0jsT&xD*e zT46E|IDY(!e&SglML=c)V;on=`!psM^691atK~UOyy6CeLVm;P+tyZgoMT*1?cnc9 z5N|HsmF?;H1kM}!A%}ltu1g{!=iRqkSj;fHjb552Br&6)ssYsFU=&fYDUR@F1H!e? zUXNiOe=5W74+eL$cE!3K;fRRkYH0A}Wy zw}UzSCcAH|o;<%V6w~_7=h!!J95^g;(j_`5zMDXAb#ZlvuIFOo$4%I8ajRT(IL&}v z(B86Br%^D1v z8~jYw*#nu0g!rzPa_V|>XN$!8tKz+8PTW_2f0{V1zZL#&m{vk}l{|}TskyEe&^cD8 zzs@$`v7;MXzfprET2cJjttG0BLJGs%HFhg8WS|na!OXA-uDx(z2yo=UDzySE>urf- z;Hneh7oEcB?yXS*R4y*Oy1{GA#FnFaHxxC=-Cn~3aRf6CM=(f=mG|?j;^yH;uTNw) zHQ#7{xNBb|lC>;tGp+du>?2;*qVK>Zeq`u!EHtk2HANHFfkyV(b6Xn0y?+Xa-wlYnzQpXfb4Fa;*)DJ1*>I-89}d66l>z+w*az@@fk1-UKCcg9DbrIy1!< z56{TSr9Z%wb3t)0YJC2Dz>wAXfdtCexkkQZ1#~A?Azppi$p@)EGzERDONK9p-3XqX zyBAvD^T!3iRE9saEANq3W&AAHYvVZv?T-AZG(iA#zcrSB22Zv3W*+I)dv=KUoE*Oc zU66y!0n}4iu}zLsi+Woy_S`+nazC;VyswMR0}o~~+^0P>EW@u`K0CGbWr5|;?^1G> zCG1Yw_lSZga-L0tU#kM|s@o3=ow-hTPG6UfO*R$MFE8~p<0VXBa45uufy?b6quQN- zsfIedIvR_H$L)6&rwAKl1!7Dux%)GKO|GqDEvr_!RU=n8tkohYr%ttfbi%Dr%H>t7 zrmDEAcgR)lU3Jxh6r0U`Izf^OdD5d3gGz|SW1UY#)Wg-;mpdL+dwDKOi}&)|9H)ye z+1Kmv-zFi7xTz- z7MD?OG7v6~i+nRB54ElDH?eFa7^M7BQ>{O__iAm7*~i4gu^9cIL-25($#dq&>u=wD zKl;u3^Fsqw>ij~GM!YY;=d<_*gwG0|6wOI|@Y?W4O6}(Im70McnqBW?P_NgN)3tCH z+TL0ZdS){RHd2j!%U4s1W#w3scsP$2=)%H$0myrz(+^85S!n62t)bnMgm)RSw3~9% zk8vMUCUAaswc^a2j_n))QWk(ym|Sz-zcUPAI>n@WM89sfI2$NU+Q*Ni`0Bd!M^iy| z@6%g1U-Db_8^wdT)GRcgKy>%?D!JjQppe*{xC+PXzRt~O1jsL=$aVRb5x~Nwely@l zk~EZsIpY0L_y4KvCNw-C&m#WJtGOxA!pHe>OCeBwE8);te=I^I{S>$Itx(0rXD{g1 zR(EzXL8Uz%Y60z<>B6jaC6V^*Fz!Jy8EC%+e(6D3VFbQ{SC%i$y;M#opB;FxlHasI z+chBdbXDnWCZVEpVbXFD`;gP7XnTThLF8jYQyEoPM)fg430lku4dor3bDcxhVW()M z3i^9l)@13idFlFX0V8YFoP2!Mmz&e*M%{iI4)ZlQQj*j`TixA#6#bopX{|*o4tV!= znxt!9ZSXB%1ik2?%W>J~9Tqq#6?U?Q8YlW^p3Zum$YP!mS?42t?)T|ov5TELR|U$C zik6jAg`{fLt!C+_zTl^7X`BR4+m;7-7GpX!GWw+GF3C!GMU@Y^lngAped!D{=gXuR z{^qBP_johHP`SD`+&#!Xc1t=mJGZ@&4q1thpzUtsX}!5gsZ^h8Ve-LcPO1uy35J8D z1=vAs*M{-F{ZWt0I-U`_nEm$awV1ApEH+J1xa#pdtj+URU!PKgo=#~_vT~7Ess23f zWXnL$SQG1Q{5`ByUf}DV={&qtxJP5A2DQ>~S`RQabzzU{W-u!dCLeq;G*AbJ@m@HGV^ z-qhmxMRdv7h~l>iZ|+#AR2Zr!&VvnLVLYe9^N331#xfcjw|m84$YKr|8P8};_0Daq zVX1E;w{-fpi>k)ru=(iJaoAfKBrB}d6SwIb)J%z?36V_%Vgc>dGe`t!bt;DyjcE^o zyCfpTbGvPnELv+<6-ZP2Wyl@?GUe-=K^e*qFr#WPX=2I4@60VvEO<`!mS1g<>mDA7 zKKE5mTCULX7Mc+Jkl*twbEAbTYS|*O9y?1PR277jD#$osNsXO|`RGxc4G~GT#j?(z zPdUbTzWX_oxdBECwN-G}X~cBRTgB(45NqdFGA7^HjUQ#Xf8617?GEbd&hpvdhm(Hp zJtc&f=ii>PfW>W|@^#~!wKmvU3?8u*m)?c5_v=^cOQK}X6IDZ~)dX|#(P zDoLm2#&M735n;#Kd*x)*J~tuG$IdLO9wz*K-X`rB+mL>oYZ53AVwd{G3{cn!M=^OW zT^*iG85%Wh(h9Qdp8KR0V6%~T_fqO57i!)?lUEnNBIq(6PQMSio<93lr<0PeDjzCK zEq$eUaim#H^cZ$O?~9Qx-$!ehO?S3)I7tgB!WNa;V%4(oc;8G?&hh2AS_kA5eX&-L zc^$Y;Q7G^L-!F|KGRX0QQo!fs?^eP7m84XOaV>A_>lc|hy>IrdKafq;l%hnITrE!L zXCC-wA=K-Gn~Z+_rw^lKxJwA;vN9VQ=El!r(;9`#xf1Gw;-UjJc5eyReK)baA}Qit zl1<7lu5O)-EOCzlN_Kx^8j}Nq*Yx8hv>#G=u_^lnuBva9J~PQ)bhk=h;+-eJMD47u zeUTU|5{@(ZXmGCgDhsymO(zPK^B+Vpa_Qf>_+9+Q^1b~T#jT6puDpD-exm(kqn4e~ z{520!6@3zdFHkUGuX4}nq)sfOPtaiu;#@}+B5HY*?O-T4<1!^uJ@2;!1ft#Nh9(qx zs&j^Nn+SW@w(y8l2wU{xuM3yICTw+02;J>TFy`kIuw$)qA@@VsGUE|N??5iyy&LZ; zcv#DvwblCGquWg~Bi@@{rNs?mRanOjbh-o?otigX;XLeqgAk&MwoCc_P#iaw_vIb$ zOgZ}3TAESbd;-45ci?FuoI6(C*3lh`Nm$AIz{_SRTk#TW+R$ZQN5)2X{|o&WFTN#g zluu|mu9#qU>gm%O&gZr|$wYb;Em!w;Y2G%WUNf?xHy_@6Goo0_8z?15PxD@wlr1z*z zJ~cY6VBb}|k1EmmUjIDfTDBK=uf}Y)!)TNvb*B7fTr=*ca~zb%Mx85ZJ9*{etZDHV zo5C2{d+FEulj6Ld3M~f{Kf_H9=*41`4t#!ClkKK-_-aK&B)WDPrC$k2ohzp3gQkkX zx^pFPJg_+*fjKut-Sv#=^>_R$jt$1)$?2P7?+ayam-faLpXkQ| z)ngZYlST6#k>rEEPU(cvu8S-3+6URHF0D*o&^%MhV)Ye&{rabtp zd#*{=EoLgCB!-ULkf(M+`!e(~KwPDqG-27RedE-%Y+jX5Or5*caf}D~iBnRJk|+KA z6lo+AWggyhmj3l~@lz^&7)u)I&}^#aN_x@&`$pQGS=?R$^yz1=35_@H@_1AlKVOhL z#mNimpX{-pDm*`E@|)jBz#+81w{N{gPPE{j9(o`>;iUVbC?-ymQRUIhm^)0#04i8y^Fuwx!PTN=8WC_CXKmg z+#1>5(PwoHkjy)*`W|~L51mS4NT#tNGr#YHYaRNTjss>x(tWdZVvJ!6*+9fA z1!GP1VKn}#naFTA?Ldsk^DQTOa}DX&ok`2_iQvl z+qhM73L%Sm1{AIhOm^;mzq*!x(<400yqtWp>{!D@V53Fd1>*_`)7I;ip>Pcu=(zal ziSVM^VJd*NbI0VmLX+z=QS`-$oLw29@L7O>pjgMY>^_~w{{jW-4-Na6T-3epo+x4z zc+3dh=WcC!6@7nqx!<#Td^Z|E-7-?h&%RdK8f#}^@EyhLRW@_jSBxLZsR=K@(umch zYL4XjAjGbv*PypCy4W2P9m2_kK6_mbhUSxh*^;D*u}HPR{$vzO3CWHC%?eoZE+xfD zsR4{x9a_yM6HMlF1bh95oMyg`-Fn8RER}9BriHt&h+=v|pW-e9?=X)7p!(ei}yEaG@qKA*W;wE#xzc?tyO z1tqxT0am#w3{HAF(_Pv0OC~!4Hbz*y|NL`szf!Krcm|Tr$MvfKUj-~GB%hXck8WYp z@IpmZ^%%Acixq=CE+79IYX;!7>eYwqXnOh)vAj+4Np#0*n(8I+Oy_ql_1%S3ewx8+ zL6}AtdDvOeseRIgT5o^)d0}@6vgBBq^RVF7Bb%s144BnaRSob?odCD!L~j_!FcEkS zVZ)wQd+;<4TJW@x*z+mlrRg`QFbjZGoiGIv7uU%zR12veb0vUHz`ZZX7aBJoTm-Q8 z071|i-$b5iuyOpQy%LZ(JLAU8f-5u*Jz0T?Wkj^p{@vjO5PD_2=~&|+d)u21Tnp6l zacJfZR0zlV#G?9rDXo4@M`2;aeRdRVH$S+YO&yT+2xJXtL>Z{ha9dpjv>K zC|2X*^^p@|0l0bRgw5~%#=&y6NPO)9p@Wl1x=~lg^Ri7w3}z!`6z%e1m5td2BrI$0 zCfIOjcW8H&k7cd`l(w}WYz6LX_d0;cBKf|yTf$|O>yvf*7PTSl;ae-k)GRr@y|>Y1 zt<_(Pm*&8ZIJ$Zqusq%8M97G~(uJ4HG0B*D1GvW|0sbvXV6>X>#pLCo zGVkY1LRYhBnd`ktgvj`A3_J4I)C)HID_%_Vz?RNK{WCJjn%~*$ObtG@Aiq$2{A>Bu zg2^xwhr7h=R~lGYhM|m#?`-*5MC%_XV9!ZyNheYlu*niKf3W*B0wFo6*%E|#rN;>k z`-R^R<9pa@fXUngdFN7@YXwBZmo9G;Agwlp0~UKZ4K@mWYwz0Xlo&j;wuBXM*7+N# zHT>aq{6*$t?iOT-*9@Qi^pdZ;?@PMqY9uNK+RL#(_WRKHl_9SRl-dib8!-n2N}Ro? z<&`|`VMNy9gK@dEIsG{k=)I2EiR4BxZT72Tr9!S~RAD;k@^lR(YWLaG1D>hr1n-$I zY5|H3Y`*PUCw?~_Qfcd2-KD`T;&C`wi6NMT<|X8g=I+qfVY0T0yup}0;m4d$tpJ@- zxcv(3%sCX>AEHRx3}dlldh32K(3+^?3UVRAvxkL)k|-{5uSnDr6Y|barhp+S&10Sz z)|$Kldy!xa>S)u@>Asb;p|G+~iq%w^!AA(`Q(66k+x?u-JrPUh?p+YSoKPs-X$f?ab3_ zFZ&tfj4{PtS&wh5t#ub-;jzi1wic&I){&$@`jd(ypt%#psFV0K!lb37b0W^aHHe@P zZH(E?D$mMoTP%t&SMqdKIZGv}!H-8|bM4CfOwPvwE$s@i;rP5bJ)VHB;+xlJx>Y!aNza{g+o%*C)4HCxVnbTiawA#9TNs(f%-Is~&c1t5j_~ zX8q%1s&Bex^_v#ct18Z9xr|yb#I>J6uJh#$zs0nH@Aay{coQEhhcBE@ySI;UjoqJN zNKHjmFBxWN1CKH=HFmhUk^*~Nx@cx;t=L+jqsE6KbGob0z^gb3Sogszy=r5u;ji5u zN2U=zrH1^XAC6J{m6))zhuP3>Q7OTuQgLe&aEIsiAW-rYQekfYE}*oXe#t;GQ7fnS z3Y53_RtbAzi5-O$&9Uy!eS|R}6TAge9{hQjfar!wfy&Ylps?-gW4TG7RGe4~+v#x^ zt0=y$s+Y=hA9_~o87qx2K`H&8rcP&gxcteRo#(4`9n-IGD`rCVlD{s}W_?rB@F-|z zaZTmln5*8TJ3-uC+mY^JY4mbtFiZ~ZS%j9&4hxG7-2CV0ShW%K1hApgG<>nY-(UD! zTcaGT8x~R}?tYyb>7`{Br}eMhVPqMWZ}LaEx_bQ!){qK~26tiUlRh1+E5~625{lY$ zrE~gnPxS=qiDy$*e^@<=0E zymqOp%ApusV7RZDm7C9=;JB^0QLysdxjIe7N$$4J$H(dkz=7l!%cBj(+z9BmEv=vZ zygkmXD4#ef+Gnk{20$nr{qszpx=);}I^C{Y>9!MmJN^9V>pY;?)7z9TY7w;IuuZF?_D!JX ziY>TSJMp_c^h(y)6j=83(&?x;r*PS$upSLWN{O3(%`_c*hMD`V)uT~u62=>-n=Sf0 zvif_6wHc}OGL=u?+*yZ)wY2iT_oj<>MBTm$)N9hx+BtT7o_I4uMqW!B#fsE~*-7d( zn^@Eshwmqy|NZFO8>uKlTbOOtBSJCLlsvkWJiX4SGUpPhye+$&Wj5g3Oc!2zttUXu zXoxOrYDs#<2cdD}eVAmxWj7oFg$mtPz^bsw_4hV!rDvgE1nS(Ygd3vAQz19MXzn}$ z-b8T`(SxiLy3*K9l?^2cX;?8bwEpw=jcWR$&SibME%QHqssN9B;b0_57wDpE-D6@Peo|36?C5X2Lq=f~o*jiuBoSdbs>}pG;{7XAz~B5{_qxH-woSzP zbn|vW@t4{f$V4;|(LI$Jpa0p<#!xdac&GUm9eaUOT?Tw6tb>x;yMkB$!UXR?-hW>r zK=jd`DT>GQa; zBxt#;kCX{RJe~*18N8$K1O!jlX@ANl*)H6RYN(z*b0n&`BEA1o= z`pA(y=kY$+jC+1M=}b+brpVxs)|o0sMbvZCxds}E=~tv5899HCl_kd#JeRZi| z41ecL4YKMwn;*zTXDv5ZMVrb`txqI)r>x!H;L2&}MI$k8B%LEYGsk?48?S+S#xQ(I zu?FI6Dt|FIQhg`@ldumz8%3NSp#$`vovqgnzf@4|B-Pc@ucK5$E-AvDi;Ppk?NE`5 z759p;)hnpcFXk>j;(`gYOYc%W$;xk&{r1(Ua(2FC0y+=PSvXGWeCeYNFJQQV#p6V< zd<_I!6u|%emDdR(KKU?OIi57r5paw#Q+vs&Jn#^8JS~Rvdq4f!^f%kLG{mC>xq%WW z5pBhaC-T~@`fv4XjU2OOalLnq?FPnw_p}GCvaHFb+(P=%n{46ov>)0tDtG&XHyp2% z^P1jI{p4kOD+F1|Q#7d`d(M*99qZwaZQ>FeUT}EoM(6YYqCsXZ%;0JGe z9RNU=PbWlAYcU$CwcN7ZB-sj{2{pi0I>2a#)sQmW35s4Gw$8W z-TSUrONi%`?%Gks5%E1F_E(dygED&GV{g$=X!(=H_`!*F1hzbR;T$W0qeYJ94Lbi{ z2$6>b`<7K3Eo7bvP7WDCz9cU-prK`vEYC*78AQ>1D+dNb{YOiN2`Mzmh4j@WLdx+(I16v2ot-EH+al$J93mz60+{?!V{ZkbJ=h>JBF0Hi$v0lJm{A{wkbQzmWcUYZSg(Z$E2_nI7t9NlQhVR>-s zI*~v{MGPk+Guy&TE|33W!w;xW%G9w@(P`{PRO(?mOQ7s&YE!3A@e>kpt%qV0l9Pls zge1dg+Dek2e`(%za3%uUN}bRlQr#ozVXug1f4b=HLcP~zJ}2kd8AtD^RvCk=T$wy* z%^7G}aBw`v!75|6UwKx&`uhbh_d{UuaK4aO?ek=v&S@T0W20nV zq00reT_fxh7-knoz;2G!E)y%mG+AJr_z@@&)kuNNLNZ1N1fVO9%pZr|8>>0R;nm|1 z1?20Cr>FH3J_4un0u!sqy9T!bw)u&0tIy=sF&>?D3hmdkU4uGk>{;SEsAUU?4JRcorV{xz=b(se(N|ezxBi1d{N&s_UrRibi(cRU3(f& zY2k#irgRVU@kXEWK(ltGAZw~*bXh|hYc8@Y^PW#MN=VI=ebbNakf&<M%zW;h0hcUIMXPbWN?SmrrMJ+DBqbTaTI!Cp(5N&A= z@(&3jT^HHi#?bj_S%>w|jz(@IzKs+qxdtm$kCV^mwC%C=gQD4imH`HoG4M%7d`5$d zjn5eJjouvPGE94o=bqnxx@a6_i61dWIGDU_&$#r?9s_23=frSWX#fAqP>bc)$blY4-PM^@{(lw& zj*1f?So#F0#Rh^R7-`UF%IE*|u|F!B^}}xGKm`O5&IvC0zx^--?k-@JF8)#)6Je7e zu%i)5D8F502;5r4$_03NXXG^;_8r>qn#1h-9Q(idhIu?fB3KF3Aryrcy5 zf_?7s!SCI_AC&*gKL{?ER))h_;7SM|0-(VQ>Z+f=!X1B(%*5hp)H$uaU?oKgwAptO z5~Mr}_gnA?A>g-v^py(yQS*xEh|+%?U_t)_QN4LQi>r?fRfh|wfs9AR!L+aYdq)9* zB{Cov1#_JNdECR3SApla@5$RR)Ls?H>oVmT!^08!_18J~k4-Nr;x+F~E%&G?X z$ZCyPpHv3_F?nE-(*GDJw@TT>?*QSHZbkLp*Dt!5XslV~lm`o~@T0V}Q9@tD)bCJx zPMW@_K}n<4?#QKv{N`s5vS9dhu%?6pd#5uZ;jR-7q!(<`7QL^DhY#y&e8hTLeB(B8 zkC*q5dewD=KK`LZHsHE8fR{Y@HB(E~dA!7+LEv2e(QS@%DNgIwQ*H6MhiAX|K90Xn zDnTV`QiY{-A)D{$+QCI=!&ushszFxUSRilF?2i_N&Lqp1dP9$zH8kX#kXK}ffD+Q~Z7THbyKR~4lOi7o03K((apK~#N-m0>i|0y&c&8AN z)M-WqS5L6Mu=X<<7)tlyG3gv5>&d;fK_N((Na~&6WfLL*{ z^AL~ptj!B%wQy^L2TxX`YUs!SxIaZX;4+fg<@gG9sIgkkZ%0QUC6RGScoVljrU@OKwlQ zXFW}OGJ}Q=(H1Z0q@9nn=aZ@I&FSvN{EI<@F`xI1XobD5b=IeQ5cDUsunYvPjM1SK zT%O?BfoT>_N$j*NrCpiN*jnL1;dLB=xgTpEOaAqT{rsi z7u_cEl2bK6`?ud0-D#NuQBCAQ$CR*)L=$n}jkg)u6;rLc`?bF?kVlsq1Zc=QHqJq* zsGI~Q8wx}vS`aw&`&bz%8!aahma8wnd`UnOk=KD5`grywYg7mynXMk?G1{EVDe4MAc@EnjBr|0#|OPFS7 zpZJ>~MtQ{1keFXZkkUD14yR_Z!fK_Vdi5)w>7s1&%bD(;)SsZpCps08Q~^Hdn%yM% zg(W0!l>hc+xpf{6XJH0H)KZ|^JJU?Bg?ArS)irPurU5+ZtTZX@&BYoLbV8z{V?QFp zPXqj`A>h_|fXQE7zRR<1y^Mp-);ZJw33AUYoXyM_ONE&GCV%2-(OjK)KZ!;B-;T$} zmGJbc64}%2Iq~Ie&6YdI&4!+>rJ|uii6H7~!7`cEh==oG1lqIlJQT2lwF6@B?Hg*s zurOeob{U_aCtTBN6_A{ZE(*cZ4#F(5hrFqLXET;UAl$CgB#8OTaJ1K3{B;%zYukVo z7eA=9tnf7ZThD-I;B}n}b-LH?6gN#G&2)+AXl}57`gM%|QvwS^ ze^6Dwo1zeE^+&-Mn#UA)%AV$x&y@vq9j^9&Oqb0Ek&G26tkZn{5U-R=7-BG zUr(xDq;+-dY~J!s78xbIlohw}f-3ubVxd{1%gqqf{mxl(G#=cOE zHevddW=Ad9bu60aQVA0H9t$8cGRqRi$w8)T`^Q=OA14$Nbg`NCT{+{Lr{*k0jJKLwzEm+%PB1_4DyXx%KNF1OYStb3}}+hLtbw^4)XQdmTJ>` z2j$7V*5Z-8{qWjl3y;q9cOPE$HZ^uSP1tjF2{pT?`F^nVW)nPCn4s_-lak@fA+ZTuJdk6_4d^6IW|4>{?d<)#~x;*-s8O^@-$81fAIFI zt5O$*qM)eH*VJp0imH~>ce#p0v&wZPQM`0Hx!5XQV@J%Apb*a#2uao756tIBLjc*4 zh`#Fx3B|%RN-OS~c``9%L021IQs9mCi?im>(N$_hSW1+(hFE*u1*){J40*xx-|fH9 zI+o>{Y@~><9`~iR*Z=KD2}M}m8oa<#C56J_vY!*)Z1{~ssA0kI{$CS9=WSVjKP4Y_;%4JY$o58nQ56BfEc z>s-U!iHND0-%kTUba(Ere_C_)dXtq$blmqyml=C!rGE|8th*nQ$50)n)AlFo9Zim1 zv%{+->~e6JS?%LNJYPYNI0_tT5mRDKK9xy@ZX2T;Ifu^U_$YbapB+7@Qyj^_K@fH4 z2du)J@1PQgP6H%H5$?UPdp@kaVz@!^Z6z#O$4&L|rT0kIUk=3(e?*);7TEm37v}5} zvjMGC1pA|97<{xCsffSg6QEx#rM&ao%yxvbXUFFj5$*jZPWE7*4S8wCv z8OGsMh=}>cKm5$hs{6hZ0OA#rY5w~94iHsn2BHe3zQ}#Qp0NE6%#Zg#xbd+6@Q6M# zkb8ZPW$2LJS7J~#(1f9_|N77NQA}(p|BYIwbPEMQtV%be)x(4 zo2j3#C5O6NaAmhWm#Lppw*P&B|KlStBDn8RBel3S3A{r`cF7R&(g$7g{Uz~JLjrc9 ztr>Kl_hXN{WLwvs+}oG6z)|jXG-XaS>6|1mtc6rsCX2`jK@jnl_Oh%1iA%S5>X)m< zxw`4*72~lKh@YHZ6%jMb)mHdnQ|5DOMMz(cU0L3SV0qg@U?K5U_`JY~N3s3zc@)*5 zVXoup9{NP23-`T^(9X`5X7+#5Jc;-*Ew}bmNZ`@>*N?H`>Qf~Hg_=!1<(L=gqUYgT ze*xd>eRFMF@&{dE7OvKHehvVW_C^5q9=9Whlk9)LElP3n^!g)spQelA6WcfjVX6X$ z6c;}<&M&C0DGP!2P$f^m-c8Hy-Ely%*vcIO_A{+hpDTK*6>h=0EA(#vXgI>Iy$)26 zj${&9)pd#mb2Z0YUR%v0KWuRQ8HZhPFyjCr(OVysi>N5Km5@#m-&4H`5cNv_2@`)B zZebgavtz!f5f7qFPh$IgGfU#ZKY9_0vGKO*A#99N4D0ZjQ}Rqku4ZF9F5XUEh1tyz z*odI>+fyxXw?{1c^*4|VJP+?!AFBtoi(aZQJ2J#i_$;RIZmB8a*|!2`Rq7Y7kx}20 z)~r)z!wI$rc;o-@>xrG07L20twE|R+8F1da8O{AuuU4|d|JRh)jFm2ATo=@&JQ!2EAzA;A29F3ik%3J}Wk^()^JV9AM) z=18&}>uwJO@P3C6D)+DsRnhmBYf@#3sb6>fab~7$473lf50AI^k{7d-HEbwwj{eK1 zdJtSt8|^6_7;*wCBdm*2&pe4AU)Mf%ByS&PApOYR8;Z(#8AZG0Epo`)x627V_)QTd z$*EV<#LZG5QAU+!k+lihF5J3j%p<15>S%&m9Zz1nh#)*{fX3H~$g)7}kiwF63L8k1 zEb(rdc^=*S70||NqGjvM_icZG7WHQ6r{v*AP3VhLlhN<;i$!**dYiE0pdnY93*1a7 z$+h+LR6a|wXPwFfOx6tjatjMTtWyh}eit5w<%h z+KbHZS=o+W4;Jv^aS+Ycp3xr4rB2YhsBBBv&une(>g#@K z3|;12+^H4b8Y?7;$x$&lL~HMNR^jWH@;MEz$f_{i?sN(y55k?SKEcbW_VQ?^axc(P z;V=a!`Tc~c=5_>>GMq<@-y!M)iZ zfy{#$d)?r_{df#r7^s0(ZB@Ue>|+i+isznREuzE2c<%6NiTdLuEc9T>>7qN#&$JcZ zGq)3-@0I*V^hu9H4t~{Kzy@*A;r&9yaw&CudM?mq_4{rP)UV#!jZg2E3sRDR)RVHg zyE0O9#mLv`t~2PI$nt)FBcZyoc-YOvW+UmidhMugGjzq%eJ5QvMzZEoe;PJ#Ep$3` zLU(gPSgXA5iEw+Zpn=+(gM0t20Jtu19SX-~lm{QC>g|Vs_H0~^hh2T>{JgcWzFq_- zZW)@r#ShE{OvY#bCf_PE7T$$NPpnLlFWN*KsC^NaH0v_qEzyTzW*A2*#MM`d7V{&k zS|96eTlSrFx9TYu(_sutR<^Zx*M0rAYnueUX%h~m?qph*u>Y&@S0guUZ?!%3<{Q=& zU$};T&T0FG5h_l$-OUF+wdE>34_j?o+MTGveZ*ZK!6obVlr`jQNJ20Zw-U>%1-}=i zAG?GkQ4T#@{Uyg^CtdU_W+b2#2yq!+u)Rqq3=<`VDOBwjOAhv9ER{N4A2G(NAk1Lm<&g5HDr9=VCP(aj%7w4)DM)MJUiAE? zVVrqBa=w81UDhEYVm}+Q=O;7dQ{N`U{5Wm)x|M+S5k=foLuyl<(W)fDnK?=^R<KxtoTzMy-4Qo1G6IKK8{8*}Wn`+n#+ov?@0(k1Zj!(Co8y z%?1s(4Jr%W*gt3Ajhvf(#`WB~K2oL~ zTSh!blEM`cPW5H$^Lff=5|bzJ3K~j+l2&ZVqs)d$%bC_UsEELpV} zov1M+P$~npQAECa za~N&c&4o#;&!|th;yLf2Knx`V@nBF*{Y9oxM?`%rCANE|{{uutv@in`@8}j?r%4Ir;ZX0T&GN0pfIn}CFa^5*|TK$~Q zTip*f1GleST=1pIeB{ihgi(=%SibvA>OhJ4U5-WjGzX8O@<(I8Zw7Ab&o^QwwvOKT zz=g(Tvg5lQb9k1yW#zjC`n>lHJP~)V>2)^-B{<1$?y8K8qK7 zGeUDbQB6CtZ;q8|qd48;&7*Iq2^4)_ZZw6j)86ruX{gYsv|Oa9BU+e21oF=pWInC; zX}$8nHxge{nIgRDuI7WbWYVA3B?ezX^ZN*ewDmv$Gyix#Dm)W_`2$IB2o*Mb>Y;vA z*P05My53_npuL9-ImJgc5UJ^&+_{E zXChoiM^UhY7)4mLHy4Ljgwf*|h7*>~VlaVA6{b8`?$Sw4ZdKZR!=Z~3;mZr+LiH(E zVedQIC-0Q;m)$P9gUQ&sq2A4)*O5naj}&w?Me*LN7xlv&-%iP$JWH}u|ERtSXAa&M zyl6s4cj4khJ$c&z>2*+uS6UP#oTR}{TRogd|&qxn?@CCo9-RpCe}JUNPO`) zO&(Ca^0k}UgMKG1(+f0;_Qu{NOAIDf3pC6?k(w0x3%HQv5Vt3d};C5E% z;Xtw!tpQ`g$gK9q%mKc8jvlC4(ql~hy@1;vM3yYq&QD%B$Q~*5{DMNyXZZ%EcV(Pv zgNic1=bb}kTevu-+vjSA;FvCl<2 zQ=r>8R{F%&*L_Nk-Oq!OiAy|YC@!TkY|^&>R@i;_Re{y_9d3lx4gi0gpt|=Egj5vT zZzRV%gRz(jZPNq~+*h+nN|SE{wp>m5Nd1L`RtHM2@wi|*R0c2Y?^(T|5M%E9ImK7` ztxBgW^UIvKY+N=+(hkwRz*6TxdUCOTdv&v3cc0Ia6x(N^c=q6!o!pdM1S#3Y6M)gQ z%&Q&|=TNJ)=ZiM>wCWw-J`X{s-H$9^WTlHVRptF8%#?j-N9uklIE`)cf_eE>?ZivK z&d8h}_AErYdGG+@Q6e9Y;krW{uTD56VAX#0;z36uKNNi`U=TMxIm^g*#pVMB!?jPA zi59m2UkwGvRF_qC49d3Ih3(@CybI&vzNMrpT4-`Qb+$k-N$a-k@%P}Nw(TLDUw=J5 zhT{Y2eSWPohE17~Sk~1ByZwxzdQ>|go+(MY*%^Q|0r;iL?Gnud!4w9GA$_ZG9&I)0 zz2&-fPSYqUpxE}@4MQ8U2~IZEb4>SXfj6aR^NAgOc&1P}I>jw7V@&VpMtSJ@1R#m{ zbYKg2EqTSQbiFK6e9K|H6wU$H0t-TT)en_Ax^f2kwegjThe2wIZqlrzdPsXbj}lZ@@oshW zJsu$;ocWVU;r0EcP2F}9@AlZnPo*wW5tkk773RIA?_4o{RNR`Cd+)1*f*RZ26|*CI z_JseBYM+PPI&VK7fACyN=mN&?9o0@x5@?axodp{0&jm^c9U(nr;3Uh7y4)Kv3zRnV z)QLEviPUXao&^q0yS|y05LU@@L@<-ywhnLVJ%}v*-tD=Q5y<3uCX%VVFJ(cI^CAC) zG+5Ilmum@rs1-pcUe(tenK^l*jWYa(+vsE}Ery=>v})I5W`CgKP1>FbwLi}=2SgCw zExAw8ZVS}8l5&?I4lJOWr&I%~UUaHQ51Set-9a`u+DJ+yLAosIB8ry7qO`N==^UWI z?ky8w&*h@A#pubi(Aa48xxBa*h2|rb96-5f7%aR1b~&CRS%+b%N3CtQ|YZz^ugI+}uDNwZ86liaV2!5FTOuVrpDg-cgmgl;*CW-) zRn-%1Z3D$RlOdsLH{gy56v=O?U$A;u+X14^Z*`_b4u3*7HGQuBGL7>zLL_-6TOXji zPVsY(C0#=_1@>p;X^jx;Icu^`wQ4U(X|@@~Y+uSQy~El>0ThIhGc56S_byU);p7cT zu)8?m=wLP9%VaS0{(#IOI)^?1sQ4ofeu@V+&JoFw9k@J-ZH3C$_3nMwZ^v^i-bfqi zz~OMIHk&v7YKjkJ9wU5XJ8YA0GkHz*;iGr$P6n`U)c9>x~VXr zBrXnv**(gW4_?tzPeiVau+c>Bp%BTU%VF*6mSR>Le2|8)e{0-Zc&iM`hGSyK z9{X3_IPhAOFa_yp$i16RLKJn_pdP;@Uv>h|&^7T)tiF6H%^M>7`mowg-iYjEG*H~g zVAfqsMw~0LAuxU*QeWK7eH}qkAV#gG**PN3o8weB&Ohsa`_}o+32I?MV-a1otbCfk z(vZzk5CAkc(FT>-Tiv}ZuB`OhO-;+=>YDt8XO|+ao318Y9ud^^2?)$RBJtqMvd+CG z$KX(VrfQjC!f2pNS~qlk)gi!Ke_duSEC$b1k~mAhVHBO2at2>8ejgD;vpKS{;3f)7;x3i z7C0G#<$~b5*`v6DT<=#rwY^G#FXXC>#~EbZ4+|K_p4whmA2=x! zKOb3j*3a%LMtoF+3SYi{I#P5eXZzJ*N=3ySa*Xs%{KIP=HHueOxu%i5mkW#PFXh6j zhHg=+*+C@%bo_6%J;QN^rpl#fvV*cL2{!o#IsktWO>{7w&2;O z0M}8bVV+}t5%rb%s==8rXt*A?MBPvENIjGHhCb>cw6(&(By}%c-6!V4`>c#@!VB4X z0>2m!A*s-=3Em`nJh<8@!LV)5r5&+wGCL$zb>__z z;AQ9QDGB>~byY%k0))?>5W*AZd>9(!;61hQl@QJD=mZHl_?r0Ad!IGjrRyv7UI)e_ z;G#B;Q|8g3rbmlDN2h^m^lwM3r+uz0N^lX?3m6Zz$b1?TorcVn)!4=Z>i>m)Y0X&n z)7M<>Xw7sX=$tX#rb|yid49C4w3zn~V zV;`@^?ewDhk?VP9Q_iBi&rIOX4}dq#9s1;v1;u`#B`FTb01oN)iod&cQ4_OwvvPQ2 zx>(J42M_gL991vH44Kn~TDfxW-pQnAYP`B{3dk3OfWzFaEqTXpjh&?b3LI0sUz@_l z#VvGHVyv!y{yg?By|zJFCcN>I{mGXwp-5OqZ*BPu&2bGLp7$&h9hL`Hf8{nwM9*qg zOZ6eN(gMke*(eK_S_C9WcSNn;G<=BZ3bq6z` zcI=~f%N-pj@?UYIb3kg^XQ9;d?zCH1gugErphg^hTeAvyE|h5Aat{j7IbRvh#mHx3 z5apqCskXtimUE)X%o6F3yYS)UXHx!9q3&7`)wIgD3vX$xZReMJbgT~0Np_%*ZC&vbqT9^mOaP7On2fHi*8?Y*hjc+LP|9pea(xG97h%YE`jnKcw+D?LE?{ZgrZ z+7m|wbn5N1ZOh47(VhN4AyBfVpI-&VN4_M=ZJ}GC74%HxoG1CV<>GYB-z*pV3vbVI z&Je{mvwu6bZ813G>~9u>cQpv0WW=s(a9vo%^^|muoDH?1N_0%J2L2_lw}>>Z$f3Y< zwg!1e1Bf}Fn`GUOddNRZQ-bu6svwxCG{d!B`18llfVbn6*0gxJ(_1z)C==lz(G#WBTM|8T&%H ztXxqhlRo&1CbW(hef&9*&=*caQJb3A1|#|>HToOR+drL-&$o*e@641M%b#OsF8V4m zJ9=C=SU97gaMAbE3U(EhXJraaRH9?gh|N!-Mka;%`doVT~G4E{K_5d%-gy=NuYmp|h{f~Zh6FqTl-NLDu zd2^>8QAkk+#NmGWH z{C(4-Ai+xvikQqeZu`-vg&*8n^d;0|tUF@o^l$Ao3U_~~|EkQK!uOh~M|{&Qkz!jW zaIX%yNCjk+BXEE8mzK3TtG^f@sMPq%-lqH;@ug7_{VD@87JZVoFZ7*1Sd1y$R2$Va z|F{s%zb7bk(p`L_X}G|W*g+49;sxKVe9a#HOB zOKaU!woV%bhX9+Fe0_VaLNyLV2`@;G3`x-eZ9Qd;ei!3DmNnPs+!5F|8ZE}{_4v@8 zPSZnXy0+O}zNmv;@fP|0s&+AFL2f#r7qke5Op2s`PtwOiHzcZO04~4H@Ufv;-%L*Sh-Rv^@$TaBCWuW=>PXVrl^}n<21*5$l zummYyW{T!Tq|s16f-GJk3nOe?s<%LY-aV!Oi%#fy9VP zvolqTVNU{U<*VeF{mS;X)rR2 z{09^{#jyo1R?-9cApraSySoBWn3Hi4SC11T4Kh;zv7`{>Aivj(CNA)zMU&4p{k4i?GBP_RP?esngdY5&rbM8CK%vn1@C z)Ii0OjO&gE?)nd}7Ja|#BmgG>^b$J_TaA}ye%2gKRNGC!4FpgA9c!e8!%xV4v5A$o zre?7wMCi?+LQqf|ezId!z$u(n2DyK1j*Y+(Y%#8l_|4S+C*99a*}@;R&gf=tN}_|1 zrT-Kg?iF6W+g$TJe-xQZKqxVhjL z)-efvvykA_k>nxfFuKGjOIaF`*MK)%bq#g1*d{Qe#e_FT8Vw~q_glc{1 zz9|HCgd?>!S&k3p?nBI)|G>qqPP`>FZ0~(F3?PiAO<_1f0-O`;rQon(R^;jPf2a^q z#r)*Aol{bXWNjiDZi?ELv*Om!EjPckOna;>4YU>G*4Le!PKmQJw!l|j_hce#Yt}kP zuEM=F{DvC35YV26M5szBc;i}LjGLs@q6>^6 zg8FRUCUfW@FRWRxk=(FvZ2U6DLQWlyJtbUTbcD}Ag>fx-3*-A!ykI%@9Kp;y0zIS_ z(=*h$vR;|M62Dqs>GW!)qO^%Yu%5iv=&LhnLcYXgk=5Ny5JniwMqocTGP0jSgOnC6 z7-kcOmxNsqOneCLO*1RZO%ma?nlLszH*2*bSKgp^yn&qa*bI8Mhg^@U#Af!{wW@cq zYtf=b3+|?wa$c2lpIZve9Y}_V4|+btvyOig$k6AMi9GWa@Zu_KIPsU4>+}pk1)1TuX7le&bjog5%jt z#&$Zs1dgL*4z=T@$Ye`rFV9|I*qRIS=IBt*9L$NhO=3l^Kggbh#``+FJ00Fc?x}oi zcz!Y8EsVt7%h6+a#l6{#U>d)Z|LSq(`5N6(y!IT$K4V{-0`0~%>}Ro`7wQ%x++(UJAue_KRQ*kMWkKwd`*+6r=NzLPh=cu=1QB!;1B(g}ecYsUe>jTb!-8$tf*& z-)WfVu7zVTce*ro;{mYs29R6f;?RShCah~bYA{00&JDY9>r8p0+WG9go^oCu!_5)S zyt|Q2BHM5^VOpI%yAd=7bFe}(1lG%kxwFDf$V<}*E@Ar-)p=+cNEl&cUISedcEv)c?hJ2Id#aHtqF5?j9;v7$+D9mY?brDVlBcZmc|)x)H3zd9`zT)yt={s&6UqTV?`XBGGQd9LqNRbmIohG@XU3 z+3mrU!|n5n6VnG<>dD{xYg+gopXn8>C9@YdpHQUw$a(4( z@?L_O?S8^hBeC)e4Fhgd7xPe+_qbw1jj97zM-MqL&d8smGd2|S9=i>%htDtLSMK3lY6S& zx|ZPtxoz3++4owbm=q*>#4fs#ObzJuz74lK(|M+!HT4&mQA#Te1VobkOTB;K*2SjM z;cMl?#4p@X5pnRciuRV6ueEaOM*SmAqI?ofrem0gCV;s0LA|M%K zDN8Am^qq$I36)A|eE>;_*OO85@;gj8*T$8t!*Q$Y^AmvoPViU@PI<#V?=&jRCjr15 z+@gihswbM&3nG? zvSz!O4fArf^DHsZMV%M1i@q7ty)-}R)HWEJi9BXD#{jM0Tw5H`E%giN9{j+9Qh8jv zuwGtwa=nm~ZpNf77Wt(in8Fix5Bu_+08ERdkUP6#ZeqotEkfoIU6CES^Q@=T{Jay} zGc&0Av1FX)cU)~(?NOJ;D$oReVd{<)38TE<sHO%{K?U{f=n^)XTZOBQmD#)Dfc~y}VN>M&&(}%~AV9=!~BpTdgMt zWrnlYZ|fSZq^-KVF5*$@3?CmTN8Oia`f&1%B2m(!un};Jq{r_51_G5n%SRKNLN8g? zL+i*vAkg&KS1GVQSU1`UoCm>wR@O!6q#7bJfor5k7@hH>V+`zkz=xT5pq1HXDsSjlm^lV(C(cBAnfOGGL#RVBVJzlYXRd+3L}W6MG1l=+DfAWydNdgM&65*!>cSQlJ5H1cF|>RR zbTe@a%P(z70;`*dO31RtZM2sMJ(c@~lmrzHJ~+}VLt;E5P88Ax@&wFt zB(}Y{>KO+448u(%Bm_#1J*gV~Elx_JzS!e9)(5;~3OXz077RXE+5{RF1+n3clG}A^ zX6)3=*y=kFB_)yvO&xCTY0(_$G7ZbJujzi12ll}EYG@~;l58{xx;acC<9pwgmGHLpT zE~{w+&L2drz1eoDk*tm!>9gUNky9_6A@(6az511w1k2hvG!SZ~NN~)lQT%wP{?a*S z{})+42jxAUOB{koTv(&EK42z^f!R_%&G`@@$Z@!X*ZoX&8}PA$-?N3xC5~?CQPu|r zji=BW!KJuP<|P>;!EBRPj#4{H zVq|uSl10F)Z9kwsgeJx|ksEqnL9Hynr9Cl3#Q8Pdd5}4)SdF2F=|CJZc6OHXkr`PK zj22gRTfjuXj$B%;g3!E}<<$p^i#0!URM7!~R8DTP6NE1x(zdWN6-?Y4&p6OKg_R33 z^zKHrxhz~Wz)dVJHSB?pYcSRc(ce#^%AIlWo=--fP&;LvlVeJlmnQTF8OaQ(#Qz-%^CsP`IUz;un15WZ0@E{0k| z@P&X-1YM+naCwZ;s{ZD>H_vwMD9Sqiv2QW@5FbMVMctJX%zb5PDG-l*Pcgh8 z7*%ktuwor&DeXRgSr_5m&&KRjACoS<;DGSfr5JGQ13uLQ>f$*VF?^UL$#iW zv`wwtD8!ja>)~1v)AIm@A_KQDkYf#_L+u&zWO}BJ4>l*=A2QU9HLDI(z;l$My@L{O zU^@wPi;N(jyvFyz-f)%^3<)obS~#}|ziTZ0sB_@P8VtB0j)%_yJr6KVCx1w;wX9TL zecrs=yEB`tw@5~$Tt5n6GZxshD-z0OoX`z}uHT7nPxZ)bwk0x_1cTKE|E^{cfJyI+aOWW{s}GOmYG5UtAA{Vfk=HyXb>N>zHHn*2V;#Z7}I+sfQHwR_ZrSG zAeTnR85jhV0i=0DSWS@NW3nbn<*DbnnP;a5IxD)`HC~mkenHMNc&+LC0z2ks^Pz1$ z!TFX|XlN>tgqkm1>ex|&$6APDsPm!z_;__OyFp;BbNcIYb{M2fcP)zgpN@v2+ zbGnJNWW}`>cN><+Cg3`K6HVPqQ{xt7#*Zv-y~F#lNgX40dk<|(A3?h?2wTtU>`2LS zaZN(yoOinH+Q7~0$4NW&8lTKsq{lelZiongO>mQ3vX55B<74DLfRQa7YwP64p=Pgv z^<=%}jYVzNYCl`muDd^IDJ+CJ$CwwXoms7C@QkJo1|YLwxUw$rOvyKq40R3aSJn>T zh7Ht>9bQ$lxsB^Q4sL4$=)UZQb(rAtNP8zhhFa=_I~q*G@HQOuEgL>^rb2c-7FosC zU6DdPKDUHTy=q3ED!~@BI?y&vv%2iX-Q-YgqV5GGd_YTaA+@7;;UdnK+s_V4UK9WV z1K`g0>4R?xk3jG<1!4-NOXU8){0G#Xe~}dc{rE2hRlxoJE3nzWZ}U_B_P=$eLI}Ai nGu{|Ll6?RAzh}7hMI1Q8Pt{+lJADvn1X4buuK3~)W8eP(P&0l@ literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/fastgen-arch-light.png b/blogs/deepspeed-fastgen/assets/images/fastgen-arch-light.png new file mode 100644 index 0000000000000000000000000000000000000000..9e754abde85d1357d63576b54db1e8b75904872e GIT binary patch literal 176270 zcmeEu1yq#l*DpBgC`zc5v^q$abe9N&G)PDa3W7+7G>V8K<&e@SogyJUNJ{I_T@E;O zcieXdP|rDdzW=w@UF&}Lu6x!wOOCwn^XzB${_Xv|uap#|adF6Su&}Ujp)xmBu&@a5 zu&~ZhodE%VQ>b;P1NZ~mRz+F@E4!5f1^mxR6LEQQEUetXb36A=0skMSB&Q|?{0aPX z62e7j;3?z^!TfxK(7;Q;^(y9PP9lS6f^OGNG4P|G*pVxkwK;k2X#UX)~`mFC8hmX?;Y7cq~$`#4$inYQ#R z6RcOSUIp~p##Fp1e)ga`7^}Lvx^?KwQu_1#O050;{dX6sw1MsL>{Jv~v4q6L#2h?4 zJc{6Ocw6Kt*J( z*#D#d>aRBCd}f0yg+VAk_i7+qwpJZ?WNZC!;rj|&upu&4y{Z0%rTVV+-K}ZfcQ{*T z%_Z)`HLga*^1?md8+&Jg7*DGumMW=EOCgZ#sj96NEH7u~X10~S!~Ig`6Pe1dg@Ygr zhK)^ji&LLm$^D&OXlXOIo3OpVS9CVL<$2X`Q?2;LoK4P8T{IaMemQlj2^)nHM9I#7 zL1ICDBbIzCVzy)>jT>*Lz!_3rnzq@9jb`0&LnWGtc{gDr1dWtP__yc8yaMwUJoTXX8GAcD8 z8n#0j!*$yaHF8Xz+pT-ATz~Ci4g2*PnaQxr85F7s`?YC5@Y3zWm+bq<8S3H1yScOa z?()N5TU``e3BC=IAPiQ&YPsY4HewS256=6E>mp}dz!wH%+_WWQ!z6dW#7Uo!Hl7eE zIVZTs_Nmt-@Xyd3Nn1kI6Faw;SeViU#Kau+Wv-ct)231=joNsww32qTRs>#+PhkUp z(7tjzXoWTTY!$#L>Hr4IH^zIgis0A~4+XY3)Z!$s3M25kU^s|a!?wV`51YvJky&dA zB`h{TyqvVNwF39dj|WZIS=-Q(0Mlm=weU8O-BDR_d`Q46LJ$>SfM|eZ_7_39J5M2pX@9q<*b^@ z&mBjecXR6fCvvKvl98$pue|wux6HHUZC7Zqm4js!A9mN4cTTcx zyHNR#`qI&9mEg#Sr}f?3deY9`P=AJ;*w)}i`nw%{ub7n24?7KK?(tFKQpUha2vQR7 zb-fdMYf`Y}%EQ=V?AWK|5@kYSYtrYJ-uZ2uSLn1VOBT8KrEA z@6&YG*j!Xabpp@rrM)OMR1iWPOX4*zR?5wVmn_sJ9|<>tUtbVv zBQP@Ad_#*m*$nmO3dRL7T6}bT^La~m+=6`Sg?o1n3m2)9DvtS6^(ePf>l3KY0m0gn z3&Y7+jFvjf%P;m7!AOJQWG@wfO5C7GW@D&EZ9J==8PECDqF&DJIS*>xwg*tfj)549 zhxU$ZZns9f>P?1Cw_UfU$NlB$lau-8YWdbhViz_(!uJ!3EQ%{a)abf;_sZ#MPqi!c z=20k-R(aKqF!4aa!(5VbEV_6>9j7J|tG26lR5%o$cmt#^J^t=ap6=d8*T(Pb)V>yV z-k)c?q*zdKUZA+Fp<{mOs$6mgpH$_Fr~@~Aw|0x15wVe4Y|9PN1$&jSr&Gtc&9~aN zZ$BLBA7~$FT62|r?~*sKdSV{|a5YXrRFvv5eNr9!TwTO^MdWNaA~kKR+zOU|#`g-d z;=V-)?3wAao?Hs{0$oP`{m{e)F3E*`!~r*xx=$4*vXQH0vJBly)xO=Qq6VrzIGdNHX>No^4z8xf1Bm7MLVTFo8!>8 z7m?p~#+N{mnL{p9If-1PMB9Dbit}fMVwy;#$<(e!JG{jw5TKa$u;v%rQaV3{NR7Q7 z5Z3!?*UgT@U*5+V2oKSfaIS5IevK43fwq!Ntyn+X9E~1=Qo}P6#(z(9D756IxnK8q)z?Y3 zzx{9%xc0riC6Bgtr!DyT z2@Avb3-uS%HiZfK)+srX%ZIm2x`j17eQ9dA*SXh4>&XOTo9->Q1OQ69ES6$e_rmm{ zzx<^2&04NCo{56A?fWkl^ml?fE+r=;nyPm8q@?x?Ik*$ba&VDR)??Ae4Xz*M35TG) zHmdWF+JwrlEtHL0r-)PCWuN7AjiE5yxseu=qlijY3RZ`faGka{RTI6jJ%^@SIeh|Z zHx%gG@jh=p57Z_=^*!NKyGZ+56-sdsk~7(nxLc?B zDk|)_J+4+C?vc5vsx~fHzT+#9*&_29=USS)Ts2Cc#?&m|5Y^}>-|i}H8jC-9+1hzC zzAQr?74u-o+I=0`;&M|BpJ%I&Q&DvYC)#jPNy|V}=;nsaliZ2d{R*3J6=z{l67q{8 z(2}dCSM!XA3)jDG#oJmrl+mB0&pc%Rv z1|sGBzRX)YcFFE_U9QbyC$Y6y@ei%OVHg zbk2x&E+V?hh%iWH7@~t})Wov}nDG=eR{&2JJ4wh2zGIB>isjiBVOYp%sc2s-OD4U- zZOUX;Xvn-)F^?lR7;e$jLtdNTVy8u~+UJMiD6-6<`i1L>#=U!rnLdW>R0cC}E)p3P za72u6`=zM5&Bv}8n$6K2{;DIfJgUV#^tksUGs6+R5$P=zYt8L~C*s~g1NAfIb4wRCOO-P5s z-nCJ6B`}0>R9abta=l~n1DbZb@*ugQ4)e7Fli1{Zq7A~^%+E?5VC5kd+WYSkRFiwj zmn2;+C+;Ae4^CtGKuwS0W5mPj!QB$$Gq+J0PZe`(uBMg> z0uG=)a%JTlKNk@c?t*=P?R)L4I-eBv8VLG6}hnqNPB zaF5`}{Y)L+cg)Zj0|!P0MRPK`n>cXJHmRmhZ*(6|+zRnB{g@;qyH~YjJ{0VW;bTbQ z3F_NTXVSD!vuk_bShZH24f9<|RvI_D=}dlpims)?WumzF0`g_!OxDKPJT~u~(|ZNI zj(vIpyU9pTHXWU{S793QuGfvCvXGuhy2`RR$yb1U)FAS@@oS=|rY{(KW3NC^cFhX4 zs8{lGFTH!&FHF`7KW?-X(ViFZc0`zJG)B42Mx?Bt6miHaQ&VFFMPwHbX&uwc!kjor`p~)#uYuZG&$MTC(*x{!*&-bMx^uV0y4-F{^<@ZH4YDiZZoRJabm?toYq9HDVh1U}CtQE&DwlhY-;6kqrtZS_#`!DI zNh1eFdBye>;*>^VOYT7UW^*a4SvaiWiMf<@UJm8*N^E75rIA`!RjfpEw1(U_<-c*k zv0g~S1)yOtQjaPMEsdUu+ykv(xY-C|gt=40i(InBHs#bObt+DtD}4t9)k`-;7IW_j zo8i2F%M#c|0J9S@PC=~Hge|TbwQ0AETZ$}ahw_#h%)yfm2&lIR3-@ij+ta@MVyUC0 z!~K8_G94SFZ-!2LZ9((qNrt4GZ2KD}0e60aO8|8Mw6?RS)=ayFRBd8JgOHer(0O+ zCk<`FGTt<;m28QTv$O&johbb-*(wgAdkK>Uy>){xh{z(>vhiDflZZD#8b&kciccUp zwJ7tBLR2X0v6IENszyX&7j&M$_~rvWtkGmn;Vo_jHVwl*uRxa2>uAuflNTY^%7i zC)F?M-Xpur=yB=+I5km$L-0X+`^pqYn?*)=@XLldjUmIe{I>zc4{uV2w#$`141UgC zciTE6M;DXH7^OT;u3(yXZdVAg&G2c&B|^z3{GuvqNSef7Jya3!A%q)tGx>3YM?qn9 z0k9q(-2vq8fb}U^>1GFJy=S>Mx@LVoLuUK-p^vTuB>ANMTaZ_2uCP`Y~sO7#TbwFQiB?C)#C!vzf zj#}=pcN!Ql76eH7(A%1{RD=`)?lx|WnfNuDLL$Cen}Kz}(QRu6uiXF&TQR9VHtteZ zi@xm&CMAKf4r90C_PI2&_4oOglM)}2Z6n`=pC@ZQ8jTsM=QB0 zzp(r~o0n5H*LKqK^*iH!MJX+63iz@uMdL!Z*-k`dbX&@h19C;XV&QA__!ITFO7!1t zFcvBO2=ztqv#|Oko-SlXvVAs0O6D#u7aNa$=f{iqwA@Szx#`P7_aRN6&CY^UD?X=>c% zj>J1KR)2BfW-7(EZrVfAU?XYOsI&{UQB~h2(tO9_*=bc4S1Ql>$mXb;^_ldEbR{nT zu%kzF-(A~K&V_mgoe3!TFS+m*HeT(ylmwA`e{HKU(sS}&NE_|+C_ny`bx!fJbxI@P zJWY%o(q)GgmKs>1_8kfmD4wTrZC=ZoHs#!K*siUw6&md5&UPwcSSSh*n!gc&U&tN< zMh@7K0-H^MNHQS0WL_nVF=}gaN`e!`c831#JvyAL%N@Pz(5D`zKzS0#-TnrwstEWP z=B>v?MW*v!K=N#}_|mE3b7$;)dwxHV|DSzb%*q+~)`_APzP>okg{Yc{DuzaLZ9232 zj~tPxRUzJNc@6HPakGNztHFo&xDP_~x!T=v9_HtvbN<*$lJez!Wu)#S0Yce~-{WR)E`1-P~$dI#x_PcLGz3co(O@J&oAwS43V`}5RiyI){U=G1MnZqB=I8C;at>L&tw9D8ToYO4> ze!eiSjqT-??BMsw_uT+bEJ;6^T&)7c8xhXbMUJcGxTo!FL71}r*?kKqph5~csE}^w zjJM|kv0rnobbrh^Z)MNN5LMEWW#-7%`=~86gex7lbv0kb;uGm~X!eAAI@$?$gIlSx`wc^>?9Vx|WHn)Nf|MoUc!Kf6YyhlXx*2=B&Z-Bt3Tw)BQ zAda8oS)I+-2KPlG3v^b{;$k@9hhmE2rtwrbz>hodZ8lb9!!DCp%h=2DMc9L2n=`N3 z9d)44VAn02kC%(MKB`NW^L-8eNzI7%(#0G79M^^(OLoq~GbdX+SG1jSw3-negR|3f zLTD)D$Xrs zHmQ3tO&id+uH89!(OVY{&-B_0^=^o^Dm&R%b@a4JFwrUwrsm4cFOB36KyKge8vz31 z1iWlx&U-KOTg~v-_Bptr)#QyI?{z6J`J`RXs7x&%?k;x{DDtrBRj$)J2QGUX%Nb&; zs)>$-ZRZcuqnJ&^OzF0p-8c3)%~*kLcj1L79{{d!LD{=bg#x6P4G)s=+D_sq74wZf z`GB<7^a;{U5(c1^neY{S=krE+JRmk=JVt+F-CDRHUvDIkIo@nmcx73w7EyjGX|4HX z)ZKnViy^&^^V>h77s69Q zf$*b`E97hiSXN4ksF@aCPe5+Rw)wiGH5ZksTP}itF)nOdLq%n34Zr+!f3>%9l>fuA zqTrPYhnAO}JY|V~mLdtAw=~9?b!pO3bSJ!x&nYRiPzBZE-cY@5f23*nUoCx8VLjR)=>s{3MB);p1_(qWs)VpQ_?p4 zVgT|!IYx;9o)Uloi$S0QB=bSQQ;0V@ctH>-qWFoA9q?4ZcVICnthM&{E&qpr_N?Fc z3;fm3!E>{oUq9zzq}?$5Fxc!d+#V19?f%opw6$vg*-@d-R9q4Qbqk7o0g7msO&JU{mjVx`jpztX)tzs{Y1mz_N3t^d%6tQvCm zC#B?K1;D8jnZ|c3UZ|fle$Scr4Or|6=(MFHC^ZZ%h6iQC-CxZ8xL=g28&_yn-q8(g5;x|gL#B_z^b5*GT z6YwYA;!hl$XUOv{?+Cz;@&T;EG6qp=PM@7h@}T&JDx_s)?#t&e$Q}C|x&*gyvxr^U z+g#a)G5lFL-CTyvYtDp1p8kN8-+UFt{~bK!c~YL3)dLhD1qp$qr3a&)VGiZxG~u<0 z;rCNsg9+TCd)iPoUQZUwM#X{VE20ahArC)VZIW9tUrm5O@5IQFJXZ$zL*2|t-3+Y1 zc=~)9Hm7AJz$)irY;z}7_c3F<&8NIH;=$fc!=6SL&eB5Dr|y@~_yPRhFo@kWsJN&7 ziq~KSH^K=7uoBUPYtR(6bcxsY3w{~)*+^!cz0(x0W$2+Qcj(%s=U*aR{E*NBVm-_6y#|v2NcZzGMyU~CmKG~P9+Kx*|iMuBl15a<>k zVDgsD*sQ?s)0)l2osICU(2qVaK8GN;^}G%I$jGT3G!y~S0e)0@+xu*#j$#D8R%0`9 zBJTuNIPjyIaUo4P2e=6MG4V#;DU?5dCEAQa-?IFNOC#)egx>*g>$MzG&^_u$3q|DW z_=XQfuyw<-t1qJ|alWXZO{?p}83svXLmv&L;#wSLN7Rm89;PRH=(4eXWBB}(=h$Fb|Zs^)wd;vkZid=-~ zYJycY!B=ETUMM8LR;UVs#I#92OFTDifHO^V659Q>!R{NB6qdNWDWkbL3=^oT%$z#r zS69GqI!>K7zVpwus&M#UCbF`#XB@M+`WJDC{nMvfo&Tivg9P5yv~%a1)A5o&;7J78 z;T~yTT;Ove8v>}Vfq}X`ySF{Tv&2U~?w9<4oQEPdT_Kr4UO?9flsEP1C4wPZigvc+ zH~O!DH!YzlhV(bsT_%_rO;dgy7$EY@+8cn+S3!9ut?ytCfPMZ(X3!fMfwPit zp7wH@C_@DS-*O<4`F)uUmzyY78-g1bp>+4qw65*tr6$s$j@||g=j)Rf-)ZhcUD`5!@H3xIp6 zzzzlv=Cn7$PA4I(RQ}=}+ZOd`EhI4kkId@!7Ky>DgoiNHc@%z+VJ@i(m-?n5-i% z=MfbKGQ2ZBjQ8ZTO{ML=tN2nLHY>Oc8?89$))-&3Me5&s0uX5fQO(LHQ+}F`g5)uv zzx`y&8B2F>Z@2D`i#I>*4+Bvx)*?OQUMLcWs&py@r;4DtaK|6x-4OK;n~A@O zf04-^(IXycu)#|DmVb32e?^#TG7at?&w1>lH1|KV>ELZNv<-?n@*C_`dl15BzHM|c z982_b<_)PPY*EPoO1S%)O=S_)LMDJ+P7a@cLXNhaJL4F43WzlC6rlGorZZX#x5Gd^ z%XgZvZ|-GM&pYyuw4hD%9|Qwb0Bj)D0O=J{R0e{Twiv@fL+8<==;XurVmH4!^QVBW ze#em@-g$=fH06#jqF5NOpN(7IUL1Gd($&W^XexOfXEYo#ZGoFo{wW+}7*~1%uYPhm zsS)Zk)e`Grqu`=o){Ro!eH`1m$JBf64<{^o%ul0^WS9>Usbd-TnXA(9;^0;Y2W?ci zTd{R#vVlV+0zI;@KC!EhO)NY(Z}^gR@Ik&zaIi;@m6B}2_=|oG9oD{@Y{^j*L_^t` z+#jVj3}K5z3XEMoY?D$Exz|4x-7;l9x1N1}9qotG>ww5bx5a`99$6gPG=PZ|KZ}^} zY%d>d!ZxjV&1DtU^L|F!A~Qa;~Xj?MOxr`G;aeD0;&c?neSGoIAd*W+%??OxZGKXkfBwqjWb3xql9+UuYKtY34 z{AatOnM6r1%J6qqTJ1lcDQO73v5)o^+Q0s&a9q|QFyTvO<`VKQ!k*6vFj8)eLcA;< z3J(X(8NcPw*j1M0ng&w`jia|oE)ULzw~aDaLqy}Ei1U6xRtyDVX}>1Sxga8Ddq<6{ zyCjCEZ@-DI_hjq#!M}t0W|AKfa`keif=OHG48~-UO$CY5R&_AQD$?=-t=o(Z7Ib_) z?o*A;j@yRR>xj#N+U`3Wo3?wic9DeNULqt0ZoCkg_%Los?^n*I;J+iZ<9No%>Kre$ zpt^fmpEUYzo!&&y?X#Oj)AMMiIF$HEWZ5274^r^wX#W z%2qY`XR;4D+!+1{iV6QyuN9Ml2nU!!Q8?-ixeAWqoyeJBb_FP{C`|964W~vR19R=g zwf7-cKQ6|vghC3PaLc`MGeiSP$WtDJjSQpJF0~n^c_BR)hHLf7HBSiWYJsgB-%%?uOB377g^GfYXgtFO20F zzSB_r(!j)UDwnZ=#x&UYHOJFLD^8O-r#zL9qK2Zf4255H({yKq)#JKGSLhz6Lm zOCFPr9)?}OY-85}=M^DF44cSo5lCWQsZfzscwcmH;NrPjJCQVIX=~NHc=4<0?GiV0 zuRqe_rwGlM(DBV@&AD77izA+?#p_3>rdgmF;}VAQS=vfXYHH`j=(86?rWXN3SMqOh z9@sF_aT|BdHv-IxXk+fgN8YKUT7S<@!>Knpn(J__g&8t@;pvUeo{vMlN@zh50V0G1!S45jy zA&Moiu=+YEo%SoIyX1@?=a%S~g}pl8)~h@I*L!NA_cVN0tcRyb8#JBy5%yX=-6Z$3 z06uMZ%(@>fPF@kOxA?A=H`24GXn}H3DX!3;i11=)wRc~pQ&&+L#fOTm&i5vrxmRzm zXw>5-m3`{*ts=tsYD&3#l+S5UWJigK2=!Sebimdi_*l06kTUZUB7z|s(Y#6~PRp6C z)@JB6>~THWBhwP^T{QpXU^l*oCf>_C-gR1R2|8awnS9O_Xf5wh5bsZn{uo7i78wm$ zT>d_E0A@vmF#Miq(QqpgDf}#-OZ}e8;olTKa6n75YfT+jL2Gw{} zVv8ZJcW;S6*gOlkMxDDh;~U+-o0VatMqB+lGq@(x@y)|J9_Xjg=AziRYdo5pdq$z1 zD*7BgcSUtR0ZD@TIc()V<}Rg(tPG=Gpg#)Kqza?QWa~gG#FY~pIIo5Gg;Rb_){OD3 zwXmqPxspr)iQ)T5{*JBDEp&c@GFi${yjPeV@E{yiGX95k=3h5qL$XxU8x%})0J>C! z-#?P&&Q)%qFx$1#Wd&PNF79pzil66tOw73SJ_keB9N#gOfXyk6mW;`9pkE`C?6L1Y zJvrHyOZZ9Z@Tl~OedkKaaHo8xt@%Bt`8F191vH14)xJyXYt%1QAhB>X>d@(YvzFSg zI{Tp70i452l)fT!`k2092QnTvg(Z^h`3JYfgTcJGk z_Y$59Wh%X)cGIP7S_hg(hl+e`WdnG;we?k`Cx?AZ*3yp*RxqKUuBBGEw7xdp86@-5 zwjB4=FM$==0VYM(X`pZr-sG`7d_H4&4ay9ub&?RcKKp9bF@Xs!4zMu5)8NgITcdzB zePo$lO2w#=zP@_EN0GFhDIbifg&{q+H9orhvxX|)zcvcgak__I1MUyW4I-R8Yx?C1 ze4k?9eusM618lX>hx9BRj33GlI_(pwVoc!zWfh|*G0dc3?J#cX4lnq)T-}7N)giUY zWXFNo;-B!wK{aD~WeLGbFzn>p% zb|yeAmDmPAVF#r;z_*0O^9|^Xp60$2`Q{TjViMG-z!hBR{HFohdNPsd=h+`5;=)XpIACjP36Q z>q8;Q!?azGLp%V7- zTO#MR@m_-Ub>=iB?}gpbV11Ao)9N~T;JA7Z%FdrP^RKEG9Y6fJEbDD5d+Lp|HKBC# z#&1I@cvSbg98=|BJ6of&P1w=fA3qW*>s`8oM`D&YG2IlIBXC|5_sK*4cvt_A=Fj5qX1?yx1FuyVVAp?Z3$uX?+m9D3_?z?R@WwH*zxvzR6Zqk zt6{`>MaK+gZeK#lrgY}MsAA5UxXW77tY|X)+d*Io6Mre4zvE#1Arz0{)O4A>dET1v zYyoDMpemI5=TjFGqycyVVB|L@+ArTKke&=}I#GA;p!LB|Ed-J5BtN#jgS^qYHQLyO z4UIdO;gAbEj4*AiOyTZ9(QR!-j?U@{S4ExEoab!rIk7qF9>ii{G??*MXaJ~tP;S=z z@vsL&e!|orD3qE9!}-mZZNwHkDsrM~&2V1juz(2z+yrfQo?YifubI1Lf+9n9kMA8x z2#tHDInM6`c`t0lM$BA{;meXbBMh=R8OV^3IPntcnQLtrS_I@I+IaMkuTFXt(+r2g zaOJ1b#y>OaokOLI*JJJf%q`0q?ZssQus{WvRyPF6eipFwM>6X|{FA3Kb}p_bi^_Jo zJeE9a)rs^R-9ZdISzNy$ZGFl6CdQfQY2sZHRFge2yZ^E8B zcU*DJ`8*n4L_ttUvG)m-h<)r6YQcL`!=xd40qFA1>;Os?xn)!+258FC<*}TT-?cn+ zaEk`u8aeBg=*)=w0Fb4Ae^A|Sp5MvXnO?!z4Zu0w*vlNeI0_He}G_*@X0>_yFW=i9xncg z1E&8WcmKa(^F1nM0dD}Djrn_xbBN3ym&zTXy8lNJXms;M`+o+?>>5oiS^*Yy-$}eB z00vygp94QNvVMIyyz%x4(Xm^y#XU`qvf`;)z2#$KF0zgtUgO`1tyiz!+Vz56JxYoG z4lg%=8;C&`6(l#-=vcrf1=UaT4a{$M;iZ3GEh7Wm{)MC6{}HW?c9>aC4}3p&qD=>n zbl~=~X(6Xp_NjHwuJz)7e^Ipq*syDJD;NN;n9o?{V-xi=eqJpryX@TNPws|oHMFzH@$LPotpq9n-D*hP`H_xj!?g}@^c)+yq!?ad_ zBs6!2=vwr$muK_p^ogTI1Ko!MXrR z7`J$Uh(e%1Od@?Wb+10KcZ+`mD6d%(Dqw>d5f|2l|npmNVVcFTZR*mo$;A;=y76TDPv4&dKjU)l$iD4l(k z%`@*B5a|?58oYgs|M86E>};x$fTE&xoA$i{Ps8VqftScLA8F{~Gfplh9~T@g*^=Kz znL3!?FnNbwJFiD7JR!M)qS9BgP`CwF9lpLi`oO9`1KX)2h8x$uI!#W#mkNJ*8SY{1 zM#zfX9VZM2IXqt)#BP_9z=TF7o>^@)_dnAGI?<@&1_j$F{x>=J)sPg0sNK3X!syA<0cfjsa;HC*OpPtZ{n_@u?a#=Q|sqa89(`BiNFq zAdq_${#q1^@A{-yu`r3PQ)FpCo=`hY2cbiTe$A)qqeS2@1as8t+^=*jx8OJEq)}Ii zZz(r_f|ng+pEA}ZStcr;CF5}vS;$9lci?@$Nfpw=R+QbH39KkOx(#LJtC5Be}Ks9+unG$mNS2!yQH z9!H+-_ndN|uE{fYzZHjQHGhJWO(4Z+k<5NJ6m`W=%3TRBZ%(8uIcuGg&LLbcu4s6v zXwjlzFq7$LlPl)OUz47wX=DEQkTRe2@9d=?M>&dm?#0#zR&|-@(g@z9CkeQWA3RGP z7|DZA9xkQk$69hV<05=atgC z!WRZgPqQ6RxN;#qc+1{N?JRA^kW+M-pu5HU+19elXzTll%OB)l=*`0c?9kYnQ02bt zgYnC?`**rQX5`%{7kc7=h{<_zb;`!|z+gw-*vhdu-W_(dlLiISZT`9@Z%0`j}EJ?b*;<8#XwBo={ts2SsgC9m&z&5MfKOL?NRJ{z3i zpKbrUSAd?6lPx71xF+(!eeX=C$2b-``1G;b0~_Ne#%>)-uCDwB^VYJ1-i)pjZD%kl z17v#W`GK@_uglZPhhqVl7WUG_m$$r^W|~8+HG}}*k{q&KbUP!h};#ujf~-=X|o6k?;Og%@v_R3;N7>vb)LHF|J}Yj*(Yxj*5E zV>~QsCXQjvpR!CR$3(y{T&YhAaWUZcWSzY4C>WTm+;FvM!~UMILFu+d7-(-B?vd@X zk+_P5@wR`+N52ogd2B)Vb}?sIIbtmVv)cSKRkBPtOGW;LxvNEs-Sc^?o@QI+#C>k| z{3CnTg?+op@-ehI3mT0IV(CORL`TeRwX8*^9gz)FmCT8f8N1 z?EqAQGs3N1HkI;Qk|*Nws4&U?@#7CcZ^}wJ^piCSwKHu8obQtHvf1=Y-0{8dI23yy zF0~pGATr|FM@vCSrIe>l^JP`*nnNyRA- z@)rpmAYBQA8bESGoq`xdZ1|kwo$GW#4=wz$C>WwD-o14_5;^4v+cw`cAjmLtA*hb# z&avI9sY|Bm7s&=I-Av0)?8=70O*#`?_xVES)aSCQ%gFy#HoLWiw%6+Y`rGHN>Lc&M zqS&bYTsjcCYZ1KlzGJ+d^GP+EHrVBCF0Wgb^xR)~v{|7q9ekkpu*}!%KF#;8QF)l? zU$t@1&{u2!sY783Km8pi7-I@B51=u^5XA9BO(`pNt!P*B4P~Mu|#<;jJI9<%9U?sG9h!v zhwE?i)}_y`zOqjiV$34;<4D@BF}i=jyYK$Iy;DBWx}UT*eBN-m-gng^^nC>;-#B(y z#q}A-6}IogMjtEsxF))t4Da@;WR_hHEa#KROnXq?uYDUx%#>lPSSNI|7;DOdH6=Eg zny~Sydjk}m|Ua=$(w z{=5P}Qqy3tH3OnmVInj)uThx!}A!~qf3*z1Ld35l=BnH2OgW>yEl(4s$gyjY@GG@o`TqedJ3 zAHi*r#!UC1UmBXp-3%eFa}^x4bUN)+NC_Zz zV-_xGS4B6%#-l`*nLR(ay=aWiRb$Z;Od=Xbtzh&d%ULl9Kuk41M&_D|JyvzA>x6_ z#IkOTJ1qR}s_%9Unh6Ba zE2E?Y=kd!a4o5LX=F#&tZbN>xkC@pCrij*Gp*eAd4^_=ULrT@Q?xdJ3csQDgB)pCi z_``yMliOu3X?Jpp6w$K5 z`wV=T1pD*ClL`>^S=fsCZ;kx_Pv+kg?vR`0HV4zMwNLLYkTu=Ta<2W0{?;#3ZpZ{B z@_s+;0{^QFEx#PLvcWGtAFF7NBiF=XLqUgW!M}>6{L0~&ECuQ5oZPaGbd3SJxjxhx zpOHlTdQXEVLc>o3PePp~4+m}kO%doe(aIV-y$h_Uh-k30HXtplTT;o;vyQ`frf0z- z6rg9Li>_diiwa9eGwSH^-G5!xX`eI1?mFPI(Zjs|Lgn+|pHuE=-1uKkT?PHi;QhaF zaTIbiCl5?D@0Bb*%_II3lYZO&5x-T}agG9Gf>y9~QAwzYO~4yF7I5p`|A0f;$V#o( z$J;)T9tjGdC-p(-!Tm6h3CnS2s#FT1c7}-rq38ek7qrEX+cUwTA|2>i#dsdKQAaC! z;6L%wahmJXhq)(YcjDNcfzK2r@~)w0h|}nf-}dOf!0vGz!!J0xEUO+cNi^yMuh)|pW0_ZsbPvz^k6VL5dSlt923+LuMXo>U+aY# zeL|*(QVhO#JXCOS?oht|23-GIeF=mDAdvsFy|&`SiLn=l0E18Oci(r=s~1x&F#n8Q z9dni!l1hK;zHODEhc#D!y~S(00e?7dQ_1u*1AVPwDQeTe74vt zsobbWPkCQFiq1eVb-4)zw9E~i`)B+$M8{m~8TQ!s@*RlFx8fg3`pxT23csC0PyZez zHvoFENCI4R$;LCzi~_-lYS4}scwKz89@HZef7SO6D_H29sHhQ%Bm`2^9rvkQ(vLO@QrTmP2M>!KHT(y29{Au9x zM}H&ya3J$A^gNzC!$ZGUBDVB*J^}fG74EOB_V*dOLkA6T_;eEdI;jVrCpY%gJ58ib zWgfv_l7Vh?-}xxr`$taF*IKK&>F-$;2(IzQDpVz@1;)x_y^nG@ObjeK{qIa;t=rOI z7>2*v|1uH>h?(ja$NvNJb^KP@+;M1{zDv$hj}Ko;yzzIb{by_BaE$8jmSDjdM{Fvw z|9h-A%=GYw6Mxuw^vs&bAQhMoSpDz5|A`tW0v=gIjYnDmsO-PIJ_Q(;>ftoqVWwAT z%hA<;RV;RC+3}=Ljt<@v;a|xqkTZ;u{j$ZMpu%As6(ljj6}=R+%l%mi4=w^6eEO$J zVzR*B1%Cv}93}w$#iWlk@n|OF+;t6h1fK!JZhC=#l=;saLSUs9efqYxXTZgwgMPX4pt`t!+miIoLf4l^nT;OuPR1!>R;v(x{{&XAay z!=wmMO!+X-1U0@4!ude6=C;P-5+e5ZZeGqJj=MZXf66;Ow8vmHP;_5bq$lTMpw#Hi z#f!QR7G?=r>w)hE9ZcE%&QU}IUUUJvGD6!6*Q2Y;cfVU75H+ysANRf`u)1ht1%~t- zL7qFY$WEy6-A3#^PzW6(o4>3i)k>H_Z2xm1p!;mXwP)5i9R|{?pdrAQ7`@2~=Gp=- zsZnx^to&w)=)piJo!~)Ei8ct5%c{-0B%TK2*OZQVxf_!0t?#&<}7<*B!*|+7y-p^mWlTIkyFTNXg%_t$m~x5TP#q zOP_eFtKp!0#74TpFTdWlN30Bk$R9DtMO5)8A!QUwEp?OEfp+!Acuqr{N9(>!9g`?* z@eGHTbBTxM*~uyvZ>qzTwu0k>*}Mpmne->>MUpgSpY#;U5>v&QUg-r z1F)34z(?vz(DMhXFgr~83iy7WKI~?Q-XARo1s$Ak+u0eXDXbAO3I(O%R4|o2g6f;tF zkP$wfG|XfLM`PH8KzfdWnfI*3cIKM^gP|6WUpbL8=fQ^N)v>TiSU6g6x_ErlqXYx1 zwJR(5d(scCx+ibAc;sl~jl6f6zqD7>BYG$_%+;!+_>aDlB=8S*>453KZ;PkymFI7M zKzjDWMXZ)5hS~GD~JOS&IyB9vDAbVcC1NdipEZoy{`}acux7`pXCIr0u zt4m1FDY_jtcYmx-qNm@7nJB)4U#=3~nh9RpB$*_TgYGZ`*Qbi_i_hwt$~nJ|sr9Wo zRg`We!U}d3KOXSWGXs2A*1jI;{equ^lmWaNnMFsc60%ikORfkOvHW_UPcIQU#`Z^6W!mM&Ozl6 zXI4cLlT;1!P(;-#XvpUwo=^l$Gm#Yi~WB#kn{En_eZ{j}K(U6U=CNzKFQg zQr1UDjTtnrFx*llR&>gS*5A%dV2_tCxi2=^py$-GVO{n%y?$KTudZ?Pj8V%7qn9PP z_2p=S!@<1X;l*GK28ss$fD~TGk6Zq>>=#~osa&%+?u6+A#P0#Mk%A$yJ`OW@_*19Y zwU42V6-%TjV@l;gruO;g60IWc59tRp8TkxL$dqv0?WIm9j3s|~gjz~$sZ&}$XLT?7 z!V!}AGaX0Y7Db7nE7IRRNkjoG|3ywppq8AT5h|$(`+mdgW+$RvdiCUO+h$~s0R$47 z`%J==+rWNFCP$p0(d$9!&Qb{QB|OxF)5Caw*e2%3VSVxoJ<}b&iqa!Al(;>%1>9tX zU-wXeJHs2JQRbn1lT3Qn>!pG|OGWw&8v4<5_pPNYB>1(tzeM_n8+HbOOni#j4_Mfy z_gkinKF_fm-FPFPx3_?N@~#a~d@Z^a3Ro3HExABTiePz;**{P}Tpje%Ewo}%`MF`2 z9~Hx3jW6@;Bn}Wd8veC5`vntaTNYCUB^ETTY;*!2jNSpRE)hJMv~#_@&#Xr0LWdkP z6DtH6?h~F)h~U-Mu=O*yPq4d^!-CendUU7#u)dmF)&s8McTcux{uL!Ct+*pVnb(;> zy4-tb7l@R}9`aUe-PF8<*Qb%5cbzHIgJobgJ2suRLc?WoPh@S~=qF5yvK6}Sgmvxe z>hZ;q1hfG|$UyAr0HZq!f8M6k1}l6!$G&eD@O_`NDV)kF2z&)C>IYcYKJ3-GofqD> z*mHhML%2Mim64(4VPAd6d)aApW2cO+o${|G(=Ph*=pD*tP2Qt8`wOx~yFMp=&n1C# zB8I@shQCP#5HR&{X;vh#65PvmB!K%gtU5XLR6trXcyv$jC=C2|z3uOSRnhdd-4rew zUA$B>A)xd4txew}x8J~4m}ckp$qb}t=R8-3w}a&U&`zwW>MLOqyIz2x+JVXLZ~tYj z_zR|-+xm#aa-ZJ44!x@n4771(oeAjz?d>cFihdcBR?TDHiTpaZK4n(9!IAM$5AP*# zE2v_wQo4#8eHZ(vh43%Pv!k1nh_K0!+CUjF9;fFW&$socoPa>MghELdmsHdk_{z4L z_p&wr2;X0see{`ve*lQO`d8xcxV^aEx^1HGMSl>wY0YA=0Ni+#*e*%{0v#q=QTY68 znf(*LFPdih-VRxPUDC}4CUnqaKE1P8PwjzcIQk9gx!4{ui82R1bso?7cb)!i0{0)N z@7$K`rkz|hN%0JJ6ZX!NO1qxTK+>JXNs1=M!Y*JezMWY)u#=EStq-`Z#V_dFgxvyQ z@;abjx!j$9!`!*B$dp>Lf8LJ-pOxaE&{I`I(j_hv_Lz2mX=i zpR3x{w;jxjD!re?044tfEPl6n6A`xofU*7+U2n3&Kwst<9!K=u=>o+9`dp+3By9d` zA{X~&J(i?+_hyGT$BNPEtn->z+bzQ{P>+xH3>L`?7js0a2KS7-BxLtEEjD-QA!4-5?+!Eg)TYZ7@;K`MvKQcig`nve#PQm2*DNXNHgO7N2ZYtK<>F zViY|Ez%}zJW+kn}qfWSD|BB5)Ao2#4oeDB$j=WtXkq@;{I2}l+UZm~nRYs8?VfCn# z{tEMeufGzQ$>r@X_BDjRx?iS@@(g%7tw!*DjP)ItHLN;;BIg~|bwg$4%1nnq3#=(e zF%^^&>dTfO0OoTE4rc3#?GJ+&8aB#I`8E=q!gB{l>d1u7r7r|6L~UkA`j<|oMaE%f zJ^m1}JYDqA#0R9kU@o}}4CpK-;G2DktCv?9Oq5yfLhn;&Z$R%)sFx}qmX+X3q#~M} zMAybyoWv_QdoC?5YWw4~V!RKk+lwy4y@3JGu8d-P{PY5tQi5HJZE$<6HJa2O;kH{h zSGNgpv=pFa*mNrSpRBr%C($rnKEBeLypF-yI_tVR7ROIa4$vTMbg(ZWqTC1nH6t}B z*P$4WB9Ee`%dq=oq(W@daNQJGL$jqX+xa1m>JlyAU-XA4-sF-hth67rC&pKCB9(5(1m zNG+6@Rt|nO`M}1-UX3Z|a-vK5db8!8YAHU@INKC~q(>y+zu;m%wADgWS?9LpFgyKRF4MIfYc`gtTR(~C6lm%W8vhvCqlG5`4y_2_)8fyhR&n*6_5-QH;;l#E zwp*rxVmtTc6Zw=bbm~6c`i^9I2a#QGE>8r^mZyQ~F)4q*O`|#b;qn z(vcJt`hAnaXFIfMRZ$3e;%*&jSaWAz2K=wSEN4zaU)GTMRm3Gk^YE?EE1`k~^2u~< z8JZpGuNS@}hWfktHiZ63mrYe++BoXyBZ+~#**&;$$fEHoX4fH`5Rn>Eumsm=<&pW? z3d@e!A$k0WIU7ifz=nR%O(qFIxfh&Uw!SCwAT~$AsZ{dkK<7!F^1}<#?|-ggB*K>P2F& zU;%$n%wRVISB5;_6v<>(E{2TYVBS;UN5eme%NdquG=r!WZ``}1daa}edSCm&D@RY6 zvo;auB(1qNQy|i=FWwtl;bD2y71`kU;*CXjE11qUB?lJAGYovQawtM=^j#o?VDpt1 zb2^(18LcF*f6cZkPm;I6$p}txCH3a5c45{MtltnUUr1FCRUZlr%Lw+7L4CS!e<5XW zaAoDHtbP zUI+74c1b`fI#ZZQBQ~~d(1-VfROB@2_)B|}Q8oud1(ve$4LsL$DQ~r?J;&YUIc(Yw84JG-9U;Dr?+gyLxMSzn1<}OTe_}$J5;&3U~e^3sTEG=n6)DEv6^-;v2 zJU|bJ58~-zeG(^`nM9|NlW1I(o69bpnMg+|mm{wBv2E7%VVeN26a9N$pmDwQ&n(si zKdkMH`45G>OVkIUH4J=sABJ@;i`6reE^NF@)Q+4(n4QUbGJb1(S7uBJ6~_-kUyb^F zetT;Y{XMFo!j|6VAaVlc@#4@hT4waYI3*s=SU`1|NTY{eX1T-s;Pvk#zXe<`Ody@f z>0tYVN?Duz|)Lk8K-#4^8jtQl1w5ckWm9d3A7z|_4Rre#TNY4u83HxBLV=FEj|qvJi%Rv{hQ->g@3tB1_nD1uUcU%W z0_SlEu9_BMn$~mFXF&R+J1PNDIK@-aJLBV%^p^o3Dc*Jf#2* zjX1BC`mR_)J~rKU(FCsHTJl~rf7WJxPg-+XW7=AIPv5m3LLnO(S!%H~=!Wts1MSQ! zJYD?sSZpC_`f+nbw$0bQ^Gs7wQHMi&dF;?(CWHO-_9X_?lmU)fn{_eC&)2%Q+pjf9>>3CYXf9{N; zXTO&kVMxtS5aAYP%G6_5WU-r`AiD9tbV3)@%~T|wvTMa|RVZt{TRVMCR>O^;(DwEo z1SLPS{OKdI6>uZp4pDj@BNzE;FurNTa^Cg=Vz5N1m0eja*yMzQXvnz1&)c8>`J;pC zVCL#98-QA9396JAz4Oc=r(lye=!rh&Xa@o!e}Kdj zE|fX6$>Vu6)%b?=6Iu76>#j>x1xV$aLd`nES9MrvVx0*Wf{IN@NkA zTL_jCG`e3DYxCU47#)iqBe=(N$fW$F5PBKka5U$0nO|mE4-Gy^&gylra>)hDUVUl@ zlMny=tu6s;#mCZT-7|%U*_;0QTFzgBNB!sVk~_0EWWDQ40BMS=&fLvK%Dq0 z)HN~|HOQ;n&KK`nB@qqUZoGV#4c@`k)svZgPk_L+vS*1Qf6-P5^;W$7lF;@y18*XH zBdp77=eiuqstme(!V;}`7O3rE3>rzWTE#LXACrc*J{Y1)>D}@aj#ps&3u}+!>5zDA zQcl=a-n5VmSyn#{w*`XFU4cyZ*2SV4A`emo>%l`hVGCL_L8+`CF5?Bg^KOLUfz9cY z?KEgDVf!Is&?@jJv0UJwUHzqhaxp)}x4|uO(hW$Kq|L|DD@GSBi%RhojNWX`fcyOT zQ4ik{rtu2L2MC9g7RQ@A=s>?Hf6ew}_JUA%*FyPB+uwAuk|o|9LXatntV~uV#17AUW`{-y`z;>(;JQpc zjDiDh1Vd`iqOJ(V_~i>@?3Bj&1X35XPsW?g^`a$6htD2$m$6%)+{33R++j=kX2@B% zKlNhsO+~;2h2G~ou@0SitU&aF4^#Z*Y&}H!t5>7OiyC3fkMb}5R%}DEu?8Q&f_Y;> z28w3AAF?};S7pREvHJA)P_7|_aISV%<$f@vRX#b_`*vdA$N4~l$ptkg0j##2oXJ9odxE&05{M{*L15wV?~&Bt0Scg$W$6M*fx8OGFC-X-FkJXE^WqEQ2~ zc_QS7LG(c3&6<}Gayy%0Lt^9wXOne*|AnGKP78?Ilk0C8=7uS3I_l+7g0yGF8 zAWLNxFC4O^W+V(iZM#`aA6tZPjsDOAiM)Z3VQ#{qyQt^Axy{D3$Ms^ zD_-&s*c4z}Rt-Tx4_E;O+TdWF)C|G6I5gxx4ZUAko}JYIW7_ANdQ;f-#r9u`oi^JK&|&khJTSS3TXQCi^t zSYTz9H0<%se`fjW6P5i2w1&VXwEH7Ks#YL=(>c5j0bTF&2hj_4`uBX{@B>#($dW}h zpFU~jk}LP@IR>d4`}>@P9_dzN1-a@w(3%j}$L&8E=v0vK zpo5Z%ZE$%&yiFuXirAGvN{etPxM2Cy&&>dLuF=PbMYs}5*LFY7y4sDV%eTAc2~jQ~ zhx3BgqU+dhlek;%}96apDIM zE{B}Y>}$My^P7Y2UGS~Cws7|(vG`ErsLy-|e4pnZMPJ1YD;bA=gB1H%UVdNNO# zpW%>&G+bV?EU|w}BrqA}<<|RJ`u^Z!uh7S;^GgmnmPzHA>>r9R-{E=vb@Q5A$b$vE3;34sQ%OoiwS6rD(qoR;K+3~ z)QxAy5_R-=PnUe5x&XJvd>dD1iR|8ppJ-hkV2^WOIMs>nRNHXVT#;D542#QLvkeBk zx(6g_Y9e;z7xPHO(fO>NsJ4U+w&;+F2@u{E-$}oP&hfx(KY0)>5^s>Rag}9X>VJlQ6{Dq9Est}B2ZX3s{=^@aW*9K=1Rb3a}=6ICD zqpAH_AU&5-SP(s2pZGw^XUOKo{W*?i zL_{}w?N~Q5os?lX{X#bEo7Ymwypd%UHZb|=5408U_ls#3D^&M3inE5~A1;t2^&ezg z8Wpm4YgjGR*tv|Kx?{J+)M@*^fSy4-W}0u_Z`Kvk5eq`L3j433Q^T>v1twHb9$6o0 zqG;Mha0xs}>mUFkLjt$2viMU$6u4eB$9WU*Br{pOT=y5Wj`ZziYC&(-vfeuSz*{dL z%;JZJYO=v%$D=7jo@rE#V<4pAW);ctP*TWcUdB_V)p2rYC0Dm+v*K34N7}B9`SEb- zO-kvKo`Z0BM2^S(_Ejmr#_AtBE=K3#(vd8?9MYdeA4f!P&hH-Mwk7=?V_eL$wljEW zb3u<`*`7~(mmSg`Q5ER2Wjf|?b0_w-8hd9}eIAy-{)YAs)8v~j%sx!3fVj;sur=Pk zv!M9{^YLp79^c7flvDz#$@tl>xZZz)nHa$(p6In^k#Hxw)f=|=ArvRiC~?h6t&_kT zPNF3!JWqpNO1;=2$jM9C{viCIDZ{3q>}lI3zBq@qWR<9*9J{Qnn56>OL&$N#)5p&} zGP+NlO6L+OLsLoH85+8P23O~j!c$M+SIXL0yBoaktMgbkmL+Lw<6Ry!Z&P$ae~KptYk%RTH}w>_^{ngVc_fYuT|{J@fU2ZHjD!_JJ0QJEZuL4P5$? zw~E3O2m)Wu*Yu*b^A5eXqPmG~h-4v{4G>l^Wffs9|x)n9p+AUq}j!egpv9OXx`P5~snQJd_-q;4v5^OJ2U z$rUEKLzMuR6j-2=|15^YX_Z?jQ=wS%MGcSv!PCJEINwJyY~COEzkLkh;Htf<+c7b! zH&rHgOzP#u)$6$RFPD8x344|@j~>-fSr@LY?vPhakI}WM#$seuZfwV(h+$^Fr6IFh z_B?vBv6XGAzV-?Q zpor{jrNbMOU)#p_(1sT(kpFhsg!y&O^tc35&4Fuf{#NeYtDZI@M@E*G2XePe{L8vj zu72L0zgVnkTx#D{)`Ll_b0Wuh3LQAL?$mMY&yE6Mck5tJ`Dwo9)WdT`VNa<*iy#rv zxJV`1wYedmrwfS(qqewF-#iX$1~L9UMIaE~sKaRZVe|cMs(9iZ?Ns-Uf6!-=Q^UnQ zzYi(1R-a{|05cHFox|#A40&}}{@30r7$ZBE&|XLBAI?kQclk-{>+DMwlc0}NV3 zW$Ms;*;(;Wu>__1gm>Mk;_$N9ak*XEpir;nIy-{rO`hjntxF%XK1@LG6CdHbBMVAZ zvr2*lOM}kd^*fp2S3xI2ga>oVrtbW1Bl{}9?4@dj-ggqSfe<2wPvOzJx533_>6h8Y zz}G(`c$*ED!KBM%RBAB*apumE+2CuRTxWI^{c{f-2AF(7J%s+ar>B1rM3NEQkCjnA z%6xCMGRj-%$E2ZWb%`Nm0!!B%brZPBwgi=F`glrk>DE|~aKK^r30mVGAqcf{dMAh z?>jp+LZ34wxS)*l#PpjhEZV}+__DFev3yei0^Xue=X)Ba#_r^d~7(k<0=KD#? z`eQ1Z%P^9Ee%4j!7SJ=vmEeL9?tEmj?;wQE&ny^leE=)s`nT81_euQ~vagRG6An+I z;0M%5pb&N&T+Ekl)rmqOY9F9A+~4d1FsoagAYen4`F_va!v>}WdLLp4WI~imC?F4S zA==xx1nwCng5Z^KU%kHbb*5XC2Qut%cKv!dpO3LBNK>c$32X3@azm_8d^D+yfTH-zsK%arqc9dk3T@ zfMD|cAjgIcHg@LY+LN;+(4_&kT`?!D3j5IdAx^0O1sH<)ZiciGz- zG-a(`6(rFdRdsRVWKH%`WpNmCinzh4>b9xxYi^O^&<1yKoej?unof;`=a5ih?p7R7 z%G=gcdG%;e1$)}rKfEIOmTBe$r|&i+^<_sOw`k(g{Y)WRFfiyI)mIO_-(Il9W5q-I z2|!R&mt%$n9!}$&u8@bu`@}$|M?Y5iIL72(`!UIX!&L zHn=XWcgWuOY!&CopFD~@3|fs+eFjaIo76YA${lf#dt)AFUOK1Kjbw#Xm~orPywL%j za$~f(q}pS&{9vNN9udhGDR(58T?k1XN$$=byT8o@EtkLPEam3=mozzyR!Fn;X^Hv+ z``D|n_y;(Df(@G=qLQqSE(3z|^1&MD&@(Su-1#oor)_YfKWqAf&zY7ZH8s*16Yqb3 zC|qV_mKS`SO4wC6$7JwgTQqIV$>MrOJz8%ARbr-0Y9*f2H9a3lNAYW!=U`oD#pJ79 z_N~N8P$nQP?dhY^V$~>GT(Iuaz+G~q9fVvSUhzK|H;~iKU!@q3dZgtgzbu#-BIWB^ zKj>k|q_Wsgj~JdjD4Sq=+#e49NP(!s8zdk2glQP+I6e^Y%M zKdC3%IlB4WN$_irlE(~0Qc~lnf^rnXT9F6*o0urN4$$q^J;`s5_VSS@%J8BudMc+R z&um-OITdjJZIqWfI!5kU6|y>h!+4^9?75GJ46FCytLhNbPe)66Cld`qhy~Mkg5GJ0 zPAN^Vley;z50YJcL2z#=XG(;Xr@4=+&(wV&QYR~CYW0)UflOWjuSv9zKUurRYU~uM z)SUsgLdbJq0^dchukmN!B*J_^GTZ0k!Cy=$)FJ!D1W$Gao6o2L4`A5c*6%dfc18?VOW z_UvX%_S!%-Z^qA!4mCb+2+ozHKi=NAa+eR;w7%>~W>T`25B^NIMwe{nt10=?Go;_llm00^T*VZfAhM)XnfEwx z<;t*kI)5T7a`(f>*WV~|yA=%iG%3$pYj&ua^RT@*?CZHb>}0BedRe1(Y5d%|=ZnES z)YTG(gqs_c|Dcp8>3c2nd5A|O)q;1Shn8V~P^Mo`5&2EU5B{1~QdowvcV=4@QZcE@ zUq}AME0b?Jpg=Y`>`ht5J~o)(uS{d|Hk&GBsmg#^r&Hf0PY2D?ngllEUfr1WH%U!m zx-6oPiX%GZ*`wCok2#oflU{FAJFj@_q8au#uhfrS(vgeL zZ4Z=P7-3;o+)@OBc8!N9O&rt+E!6N$!BQy$AziexiXl6Z!3h-_v#C5ibG{-D40 zN=QmrZFqwJpqBZ2Wd*XqeA`uDL#&J*4(DJz>zzt(G}HuJxO`W>l?p@7y7ZIWr3ziW zAm#*XE7xVzkxolIlT2k?yFveVEp?)cKp`NM86K7aM-AeGiuTVf8L4j{9ycG(Zm=N> zkb4|8H|cZV-45#r*8O%Fh_?pUA-ONSw@9-;I)>I}ap$??_*2=CLilL;xz zxOcN{L`q!$hR^Ved`sGpf>j?uCC-NENY@2`%^y^a^REF`QTT&!j z!-9XWOOsvOz`M6djyy2?AbwHD%D{tBdvNbne8eRj%?_GJsiOOa?7f-Ab@R60yRfdg z9c*O-x}cvNt8`!%mVb)57qvKb3pEi4$d20>r(+S6V7yeEQZoIM3Vvczih z6TK`0r3@<7U9e>y8)GldlMM<^lSz|qEip`kQY#CI9kDTbZyxDY;~J|}k|JeI6m2mF zAa5F0nH29#picW^yvkil@2b1QD+*1HkB*TcF6#F5%d-VNvh>=jnlXG}Q;yoSmzy1_It&l4FbD5=#zaSo{hENl~0<((<=>7);C+`a$mo zi%}NQ-&-gQ{mB`~@zQVW2E1*^=DKo*j1E~pwdISLc8M`Pp+;~oYku94v7;65KubI; z^VZF3qoTZCLa8BYkZ_ef<_f_>A=cpse2}H|Tce964YJqc^yBTj{ND|OX-u7Su-P@L zdVEmu2K|$CuX(KbtLl#{%nvMnq6hry9XPa95M$ISMcup%Yo<+1@v``9gLF|j9Nx6& z9*sH{br?P#^KzL4hmoXSZy75k%pXst0pyzQ8@sJoDoZB|k|o260F7IV^KO|tQJJjn z(|0E-E-xMOpa;uOD@c@Foz2V;GRi1lZ)BvFF2^;yG)40Ad{Xmwqt|8;8Nn#3AT{tR zV(Riz@wh5eUqX=n|i$T z{xB%FOFKrriWNdCP(=*~aQEKgYh^!rpEnQ^VH8>p`CMNQ6FyT-d)5~pG14&SO zZD(Q|&FaJUiQfidQREv+dmCG^@}@1Wew%Pan`Ns%!ybP3xjdptWc>w22M9xi?TAIA zdQ&At##hwa%bQ3gjbyD@Yw_Pr5w)!dznmD4;Rb4O6f+sz9V$u`TdF>*Y`Zu}waDWnN`zpl5r zh(U!QF_gPGfy(aZfL@QR*h8kNNKx0Xw2Hx^+byd>Ndx{_Vj)WcNORHLqqKg>tpAD# zh|7*%d_FOeUT`_-8mo`!yq2CWHkwabIalUpwmb!sBGSq#S9Q?Z4ypHej7Uz!<8=)u z=8+0}adhm+5R=h;?B?g>MniU{i0dZqI7Nw)O-0Li;Cf9jtqq!^IPOmyd%8<}>;M(A z;t0q7w^>HL@>3rEWM$q3`do-mv<``r#mX6LtYi* z_vm(CY1wX9J0$LNXJ`>gy8V{v>@S&z>>-ouLhHholsq<=O6kJm*$fn545?k@reUduiWXJ(6#9CUa z6J_{CZDJ2O*9_rBjdUiXx7ZCnQmwaHh!*;fVH*LHC#&(5B)qPq%9YFOS2p!mTR#tQ zgfezY?AXF~f}!+cV<-QeiP*}WE1&xE%A?)l%EQQY`8p--R7J$|xmI zf3r$}01P-rfl`3rG(DP8sU-I}L4xOln1Y)@>upvdajzqN>5iC>$;f%iv8u-#OC3rC zk+UtD_to4(KMiE7Po(NydDW{WQ^i>t7(8Ct%9spLUS;j=_feJL zK(vA_pCWD(GH%CY4});{?BC6nb;r0Jtuj~*uCLyj3{mw^a0N1QJ_%#Jg683M-{|EB zUf4%^(ksu?oIAd%@)VDv0GDPi0ChveRHnfo1v-6DsE5>Fp8xJ0$r^PaNeB`$-IC+!kY&Weaa8EV5G27LTBaj*cTFlhCD`y5<&KIvp-4#GMLR+ zsmviWy)X5~WATZ7r9`Q59t(1E&***;^F&l*(4*8A+WUKD4Ys>kXp1EA8z!0Z0zIHz zrXGvRtO<-?B6$^pb5@lzKViJSXK=on2!S6)^p3Pt3=T9(t(#Cs28fHKX|PwpT|lFE zS)ShYGvx2g3{w!}4vyi2I5?vfytC55Xag=X4vyKI7gc?Vr13IjkFpvc&I*^Zy~sM? zAkYN&aAXWQgQdTu#n_G+1D0M1VQ5OQeGv7dh8VQH)TsdT-zh@eP5Le=E&8IFxYk!ary&_Y4(%;AHSq4YVoFj!O;D2{^I$ybeGjeZ&6Xn!9VFqMT`{F*SR~&_HcDXmp|MEn>cp zM_o*VEMkg7wsBB3<2nPN2}-J*DL}qd|57RvH6umBG@Uhp^lX(~w)vgF_*)L@i$NJA zmY6*x1g#!wdP=ush7I_GHsu@fbeM)3oS zzozkSk(h#epFK0`k}Mq6|6oe!bIMt+=sk!({uXien_Uyq;r=1HPi@^|dWN<(WuzxHsC415 zZI*99ew@`O^V)r(tiKM^@oaS#{|^@<7jU8Bi)+s0mR6_ ze33cM&KOR1Ep%!6ufDZKOwEO+P}cT5>W15`v|X!=%Bj-VahiHZ%5IM9on?tiIDXG| z`w13wTHP5Xn&0JUYuj+MI;`M5sl1j_wb>i_bSy=#TfG&g^_|yOnD&L<1WrG9s2gW5 zL|7HAwS*%JZ`71=v3-BFsPl6r9-Fce$o+g$|Mfa@egQ8J<8&nSo7OB^3QR=pjyv!m z%R>04q(_hj~%l7|a6h$9qV1}kQdDkoi_Te%|HoOwHOk@35G8Dn*7T%#TMUENS`Bjy`&X3K`JHvR zpt69hpfpiVJ=4l)&2n7>C{6j~BfaUpA zSN6qRt!v8UTZa2C=k==SVLcpB2+EzApLYNG!yZfIf3A>qL~JSe|AUh_&Qyf^9zt9S zXOaQ$KWuEjf^K?&NORw68R`l~o8pXDBjEEJLj!ETIhRWFw-5p4K|T%w!EpBF3X!v4 ze))g+%fJB+xS{&6V%tz)rttbd{B6lLxXCKJ6P}uX3S@WXe|i6Z0sH;J$N5gF`R)P# zj_s@yge4(P*{5I0EdRy>JVp1wD0M%$Eq=rK1hEJhw)V7II}Gvr=Qf-L&rhqt$HFe? zXGoV{fvF%VP2&d>w4WjxzEcd&RP+)+Fp`L*3Dfug9w-M3X4hdx1SKNzrvTH%bo%3Nz$J|v!JE8QtS(ozqWL6{R`9Y zyNCSO#=@q6J>?%zyTANxB0=vihte=hWuV^t|emZU;_O#8ej3d(=bFY`^!9(L)laO&Y#=9?ij=Ql4#*3c!L}}-2$LeI?eH4NTugH zeEK#ZS{VK}qy7K4|EFkw(xAD}u>Ol52u$pA;|BCn4}xN|zE^JeiV&o{rBwKSq;TrX zo9x#S1pD}JS;IwWSF@j(t&Dv2X@eEOCq&BSBYx{_6@1~~>Mp4vE)iZ796Y-YoN`)x zF3M4tSQ~#ew1BmGCn*@ek<|Nay`J8ve@iVk=6{-ju2jzQHB9)w4{FifS{bBlWt^|h z60oQrEhGx83r?Rc>_7956<@qhzciSQ{KfnO+kSppM_;hIML$;I{X)ExB`P_|;wpLa zUoFm_Iw4QH2Po^1vx>Purh1CxLc2<{kX>Y6_+MiMBs|X&Y3S(A6ZgRHCzg*i^-urD z_l0TGPLHFn)Jc9Q6Eh3;Sj9e!Uu!|VUA0SE##a+|M5?HZ4-kbRI?vhVlL_`eO~>2+tXnkeS9mIKU?`Xdw&OhMxD zewJ+X|Cmhf=QSCip66*5g8!@LeB<(bD-!UP^5=6g+){4zzh~!vkOu$fNpgBe`WC4> zttbF9cAj#M|L2)rrFjfP4aR=68i(V>3DZs~14YZ~)mmRRtsyE&SKt31NRs zGaWT1r;37!TS1t6lA3qNgR$WPcoT$yE8$s{g|o}$$e%vLD%|76iZ6)7T4+seb#wKJ zDT2Ev4MJ+`7p(=%B?`Mo7=yE|N7;DB!2H}Pwc`r0%079@fefq?@7=dtl?D82CE56{ z0lA;bzG`n2EKxs=MD3KJpJHLB?B9McyjHdb8%o7s` zAjhF!SchLNe*gX@gVqw|f{p+5_>)bUk^6U{eL37{duT>oH8kV z?_hBd6m98p*N_tnq`C}5oyDiM7=A}k`_rDl1&~x^$CyrlGz|NAgYfo;>#mqsUp}7f z`Ly^~B=w*6hB5@0NE6UiAAj7P%m4~<(?Az4!iWDadP;9~jBxUr;U?f2PnK67LND;( z-%le=K|qHGI#R_yLF)dv?cmd8!}o0*NgO#?ZFW>2?{&G19d33SJTThsS{Jy%8#w!R zA#ZG|$vG9Ke&02v@O0bQ(qb>%=c(U)btI=*foa1>7F(3x{e5r9p(Us zY4zLO*Dlg>7mpiR=S(#;aH>&oFwnWs+5K?HP`lxhJW8||Eosb;eM30*2CpKtSw_RO zkHR8tnmKkpRVM~i8f2MQ8umx!k{I=F>{$Qb4F?TC-rRFF5Lqx9i`1?51;_-Kklt>7 z4LlAX)y9qjCP|ch6isH21faVeg#-$SqJ;71vmNczPlpW^<>Tfa7-)sFBHy`E7|N(u zxBB(P_4~Rc0I>Pz8BIS`F&jMPE5m|i(H>9Z{DSzkLKEj(Y|+UQv12F+Pxx!7NkhHBZ60KpF6H- znq$rEQO!)cFb7dScQ0t);geavhMmXp_3ij;!53I<~3tQW%xK@K9w1FP?eN6ngRqIo1p4L(zp1u`S`@G3ue0ffcUm|7V8cpBMRaDun0ZV#7L-Z9!(9 zIg&zR&3$1~fb8qdcz}|6MM(dUBKbk;5a-N3^nzltgob%8r|&uw`W`dA_#okSq0$`% zm9AOhzf*@Tm^DNff>I7Zie&LxWH<7cK~f3$Au;*W3-@16CjNcws@VLFh}+Fl>JY*{ zEd{+yQJgTTB?x>nE+}JI&yV_m40sG~u(m6bsC2Ik8^6M}sZU0Yi=YKid?k*{zYAgzBMA0HVugmUzC)3IK z3@xL=TEw^4(-&SW{|#FRYZqG}9Ihqz-&DG!H%GJhlDFR2P-Sl;TlK~kes*f^e&s?(Q4Z4BhryDgs7}tCD9fZHM|U>>iOIrf~j$x;Fpb7fZW)A#EDtsnq6ddHp=Ys zHT)Z}s&&f$idg)YPerj6)6*|=H;Nh|1UjWxcxTiCbuk!4BhaqiN(0v?O1B*xBDYQw zO`gw9gN#`s_69}BkV)R%TLyFYq{HB?mHdaE14l3rt8LV40LOm?(hnc!*PzB9@3hzD zsS%!|9UI0h>gX@C3&wdMZ^q_5x>GoD!=c?}Os!Jn251Xs%JHQIgZoK2&=obSyp2nK zTQV(S@Vs_i{Z|_KGrM7EPGP(vA3T89qk#c=z9$HO4I}*i0mG#hIu$T$l{^aQ(?-}F z05RB%zd7q_r2`uZTn7q8ogqM8OH|^Jk84klAYB5P<0H9ys7L@~k%8)ygre9O0LlWcHSo=ppA!3<=_E`?tdi%?uvv6p|(85VgN|uAK$j z|GiMfp9fmvkpy1{p=exA3Ail_=c#jjSkBb!O~Vyh2mrSg5H$U5${vOHxh3*h+QQJt zHtC4xWbIS+UR^dD@hDuvX&O`6tC?e9WJT(5NL=*Z#2_@nk*O|HJNB7iukmt}R zp#zuWwUcu52d;m3YjeE{dDS%L@?wzyT=~`OkE}oRgcPCT4DfY!vCB$a!O+i}!~jf; zP!FGXCS7S0x6khqTRYEs$fNm*SNgsbY4VgCIoWi%9*KVOj}Jzns|JrW!qfBs2tuBr zCck<7Djk>9(2YprBH!jQ3rQd3L5SmBBf&2AqNKZyl_e(Efg2AGAOp|MhhfGa%S}y% zvQQx~QD1rf`-o8)Lf`E^4`9+GSM}!#P}ERSL^P|6F97WI)x9u@o8m0^ERGNu1g8oc zkn`EU+9kg@%i4b^k_En*HsqE1pGWllUK?B+br!H@#j(mGcCd8qY6D0vbqn@=wtr?r z{Wv%)aLsm#mP@fY&M)(A@H;Ecdsa1r24-7GgWpFKfE9M0Uijl9@HMDfepKfP$x=nr z1q`4E%j?Ii4SWS1jK?59&-I~Bu(^Wk4YDfrOjrKC#DoR&z-(mY%If(Y4pycE`v{w0 zd6F#sL51}sjl;^B6>}^XlqsJ7_uua~Rn4h}!}sO5^< zTi}>wjb~3&J$X|@wi-7eXRnmfmxtw%U@PW8BUGwR{XwHTZ~b|k*GW!77dRlUQeZk53F6(>(vt8zer_8YrD1H4@pnqF^v~e{d3pIiyBv$7M z^6r*Q+HyH9!q_4Ev3ujCLHM&VbO2)r&P1Dk8^gQ%0L&`0#Kkw8a3^h*!G&~7^2 zOohWTZCmhd+!sAzLVO8|)DY=1|9-n^m*BzGDr`l1B0SHH*vFNm;BW)62#`7bDJm$( zuc9J|Gt-2Ts5|Oqj8X1uO@+XD;Qm%qHKNJ64q9UtmQ($l_N*%1x|lH)f)UN1w;ioL zZ3T6RtE1J+@MddYTfJhoi-PN}KCdF+bh+|rlY$PY9~OFg^peE^@5y!97LJ3&6!2sY zK^NG-)VSo4^Xw0Ux$d+gG_sq&Y;=&jVsaufPS{Z1zhEf#tkGYNBo7&e#^Dd2MlMqt zqCp6e!T9rDj1Ph-7IRvYE|M}LA5Z89r7ZYFLw49?iMax_)loC1zgsDz7kAnW`v8RyszL#%44qGj(TpY53`D;AB4r8 z75axe3zw@yF4*Mo$4w*5pb?aYpBC-&GmBfLMv(rOmV#0qbhCNula&WBv(2h;yP=mg z>xo~7o(PV_-gT7EIrP0MpC64&C@S3A1u+ZMO`c=4E@PoRB!TEo+LM-lO7w3Y9cR!N z;hrcwh@{bc#*A8Z_xIm_kU7zRGX3T9N=mV6aHKK*w&c#^_j1t^Y>U1wHAk9Z@ETPT zeo+@64@xQhuzW-@#lgvO060?72((%;p(97Hqu&N#t-Vt_hW2arD2KTqh|8lq`b&W< znGEQ)C=$4T>6(EeJu4pBCUGc3!YmKLI_HHk(&hxD~lKXuS5D) z!pfnZ=ImDCfN?c~VM>8sQ)RxeIG`1BEMITwDzrJ6PuD%9hcO6w=2WNl8IIuoAW}zQOP)<%- zK8XzN^9!a39VTnG8@v8ASv+Ys@~)%TZYzvjLTZ; zx=hT^i+{0}oZQSdvU(TvTp#kv{(&Hsm8kDegWV0T#N@D$+_hr`RirD&`~@qr-F-8$ zO*u@LEA!AlMR;b=H-75zt^Jx$1hr>X`&+&}C6L+DWDEu&oJB;a&G5J8{->2u3si&nGPvg>Q;odC@yKB^i*u5nX z4kXB+duzAOb_F#ql5^4QNOY$jFh*?TLRy3v3zpU2yyP-PLO8Z^T=ESHG26_| z3-DujqUF(+i-@I3{bGzeUZU8~k6Qk6<&)R%-H}Xis{C~~jjOPXRFayPYA{Bfy&Ve3 zhprh&RH8@ms%{zPm9VzCd5Ufd1}GEV9IJFn^M8!S! z2V!Zykr)6TpjUs|IJY#W=*|aUR_;&g4d6}*|D<}dez11ZGKhUjO#x`JKAL1rzB;tv z+$rCPG||0L+UmQ~k0I(f)%D`DG#0XKtK1Q{t2^pF%SO2;n=ZRSG$>3Qde_lRvzo_nUGJ6(0)#eZm2Kl1&p9`d|Ab_eoX#)d>Re ziVBJz?)%pEg_)I;JCxU$6(+C?m=Iq+FD|4-xW;y)QT`F?b)3a#)_S*sn-)I$-o?>W zhw@yaiFuc0aKq)q*{P4dTsc~N1lWdXmPCi3hS=6+%nA+7Nmrfxc@%lJZnHxB7D(6l z7UonW+#2(&k+}T1bO-OgF!p^gR#gVxdJWVg1vXS#bwvwOB5x_J9<;%(w)HA7Iq5h;N`JGKgTO8zH;lRZ@i zo#!s7;e&K3BlL#lcCGmGJFc_M#AYm^3S|&tKgAxaMkkJcy;vwOk7vm}VVv@*F-wQD zmaEtlwWvc>Kr5^)dP@T&oQzMixy5-*YS&-|wSQfsUn1_7%K)WlA|`+H8A9?6wDfIq zDg<|I7iB^i-4#}2Tsy2q;Zb?3pe%WCv@x-F;gs-G4`2-2L0~R%p%yaX zAm+-+ZtU53Wy5XmU^1C%#=`m`tSrKj9wgsbU`m0gQTZRwwVzhUms02K+rw>e)hEXU z0<>5DANKw{9_qIJ1IA4wW62mhG4|})5*quG?2;^H$<{7Q4M~`hk$p*IDP*tgZG;#* z5m}0|(_&v*80$Qr8I-H5TwVA5ef?g)=Xu@#m-BNT=dr$z_i;>3^^rhIlVt$e2NSw> zujQ~Y&u-p%DMMdg+$wyvp9y-0gWvRU96_zx86=Oo;AY)+`vivNOD4PNj{#16%sNGZ zLuf)uBk8M)2-C%`%Ex;qFJ^0B8vN!KWQGZUm?ma1Kz%Oq$=eT4jZ4e~nOnno^&p86 zW-B#)<(9~!#olSi3evY3Vi}KFfX*5lp}YU`g6MDE_y)QUU;g?5U5(09;)n*8yuZsT zU>_WSns4qQmAV}D=vcFwkVBPHlk}Hxm%)Rm6g%(UVOw=8&&PZj`uoNN9p9hln-U=y zw$6mwm0fZb=Zb=lPxznJ4O}(7nWQl)7xvARL%u~uOnjp6F4;oRtFW z(w9p0T05@PX+Tg}(e0Ca1ldr`JU(*i(~A(`=T;vN{jFWPh1BzXN)8VE;N{>qw1*z0 zQOdbwV1=bQjm!QY`t$VWBaMcR2Q+B~tx_+`?Bk3286qqc9PjBy+~T{@)cb{>{p{Nr zuQ&TM=u1p8;Co-#LdwIP7#k&>YF8fi>shEx7A3s*va3IvQ}mhTIMc_^o_*BkT=2)E zd)*&cQ}wQ0K&!4?LK{c+Bo|v1IU{?HyiK>p+(w>0aQEr=SNJVqXLG8sy%4|DdUPdj z1(FeXJ+|)3su|BQ$N{7x7e^HFO&+t~W!yC7M~Qd#+`97dozzGVZ>w5}H&LC)Z<@|E zHGlRZ*s+b1$=O387rVJ{vhzkiYVzE7fueBWTov==?Spi}?O|uLnkL+2?^UjzH=;v4 zunMXC8o?f#$?J$3cJC{1tK^c(NvVIFPfMSlaaa4Jj73^wDtl& z+VJZ7-&hD+DSGENe2k^$N}N~2MJz64zCOuCfrsEIfuD+!-`tEEKlNVJsJThNHrY+M zLTIZ{OFuWR44&T}p>Z@_{aDK1#G*TC=KO%m4n z6MZ6}(>3nXjPpVtOZH!3IzE}^(S0~pGw$)b7{iZ{6Cyh=W$VZvk5&gXAk8O8t;s@p zNZM)t5Y_y1?`PfO$Y`e=%OF)jhg$+ZfF5JQ&vo%=$l;i3^WVCEA_zGJ_xAAVD*Xqd zkX{R5i|JkJZCO{FgPYGnagvL7fCZJ~ym;cS*k1^_F!8FtcknYO{Um>;btAOkAG`GV zWZx=7F;cDQ!PkP2i;EBTjMtZ4ATPb%?Yw_X+u=q!<6EQHt_i2ZetoEzr_oZhe&yk~56>|~iiOGEo#9qaR=!I+ zcdj(7S z_}riHX6Zq)S?|N8*u}N;cvp4Z^18B1LU%k=2o94;US#3*V|EEQj}FPf`-JFH$n1m9dQc`T*4l608hYOI}1o-p8AzCEL$*75bHF7i?1 zEJ~Q&Iaf=ypVQqa+#lx#8%`3QhCKO1&@+)Po*G!iu3LxVbc@`FJTdV0#F<{D-s72) zmX(BUu3`W>y3F)FHT@kAl;$}*P!|)Nj<)2;Nd0)4T0eZEI(EeQ{XjIHV&&lJC!DVh z@q4i;X=+CcRdJt{t74y^T$CL}lbGx8GKl(#u*-G^oNJLj;ENCT4gD;t^u`zn+ql-p zMP>OcV?05DxY@2$?n!*__%!jDlYnal!^Nj@4g6=% zjPl5<%oU`FQAo-R&lZkMiO%^XSRzsijN9iPvg=r!bsK%y$u$W$4UAhh#feu)*Nc9& zZozC_k}pp)=~$>vJ!9HqoAu%e`H6aQzYjqJi#1cvyxZ)3lPABN%5jMW8sk@g0lL3$~^68c$MT88yOy}c69%$0Za={1-4l1ysov;lq= z>akr*)zyTul+w4c&np5o-0BSEr)iHEy|#x5TV7kGTbv!0CMNXpt9L%FvW`Z&AE z@%e9FcRcbsY}L+AOFJbtUH03J;kN+%{3KAAj&=E&)GR2u7UeTrbpgLAW?4&Ai@mqN z^Wy*P?OPYF7-q^7`gCHvK6;-YA7~ZGW<8l?kbkoto=B~e`(kVs^-_CMm^!H@ckzn; z&=FzIr|u7odt8@2_wbq5GKdIi_&>P4+;#TTMeJ(0n#tHl|M1TpFPzPYzy2VcRghm2 zdq+R`$|B#Sv9?fk_^|*Yo^Nb+Q(FMX|UTt!5zRWDEz&n$YW2o za4Gu(qhoTW_bR5CN(hmzj{fdQF|Pn~c{%sM3CkddFRORmEkmLT4BO8ihZ?09jWE4G zTD}l8!Y<=Hps(;Gmwuv+AP1Q}O`yI1oEO!9+3aZ(eS_$Qq%VzlT*Tspo#rte#k0Qm zYp0ky7o~J1LW1_c9Si+{el~Y*na`=Oki9^zqc7{lN4{|6r;~>Z!-33WcY4ZftqpBs z`3LmbnI$3rgyENL2=P9u`O+Q4bi_|+5?~8{=+$lUf!B(LRYl8(SJgQ)+!ET&I)->2 z0^#snb?@0Lt-ZT8oGhcd&aAdasmRJcL|T&lyjp9D>aCF${VKDO@*V+(@u-^?G4-9d z`!^=l%c!PL^GUBw4=zN%2OtVNAlwwgmLOXZrlp@19 zW9H>7ZVdQ5LwDq=Hj~XS9BI=zjM?X?9(1TjujODy~>ehbhnrpS>|E z9J0T!-Jg-Eyjdz-wr_UC#HSW6Ie=ccqSsC3dSfItjpdX;*?YXO9+@EgFxfk`%O`CF za@-MtelAls>}m+^B>$M2k_XVfM~dSeb1}0VFhAIhGU{?Y)42}}w%4C@=Gdv_OMO(i zMA<2{+nQb>!M&oUiUE@8BDc#f2svf&kr}@u$4WuGy7c7Tv(}Ydq^;S9a_cyEq55ek z)D?{mwrQQeXu`^PI@(7NZo?QKUR7Uq${}9S_1Lvz%?BPl0;V)Z0$yk@?Hz{}5asYm;C=Sz;9BJk3kLA@Kh4-`1 zs?yu$xI3O}jQ|AnIAp0Weu8->)S`|zEfD6boGA*EERfZkLdUZYyc|tV{x?-*S8# z!|8a?1gZvV^^=|8H|m~@$(FlwS0d+2g0HkSwLIUoEQeRD8MNn2b2De7qf{5bB4bC& zjHJb@XC&xWuYja>SiEdFu99y-GwgOhNl-5Sf#vD@?N(3q(o_o6DHno$54Z2(gekqO zK44`{|CERA(oHIMTV~i0Bgf9uBW`3B&o~tgJzSU}DIHkdQCV-059UGK>~yTec&L51 z_(Y%-yhL!Lc%*Oy>Kpup6VLTS?jGEXSN_Pf){P%Z0!)LF&83R+E;pH;a> zZ8d4H@0axEgG#zl?=!jnN0p9|E5U46&M5I2WLl^tMGtZT0oz@W&}) zMjq0<&if0Ps`YmVx!ZOJJ2vZ!xnQhRFlHVMcfz_@S57gUHmI2!e`U)uF8>8P?>&E_ zA{XUqQx?x%Cja! z`!K7VYc?e>TVE@O%@pOtzb2cc+#l6etMo?U3|lO1<4Iuy$gatF=W%g;dds8`c4l}j z)>Q6E25gXSI@tSPh$S5UTrBto1*K4^`GK>sm?vw!B{2a7P4`wn2#%=P=yvk`5$eXy zMM+)cy!zgGl=Rw~_Lm>DvFuVg%Xh!QbkF$y94QvreGXAjxzyP(RwK6ivXW!9wjvKD zWuIYb-9fbdQG0|%>zTe(7Job5?CZ+kkXs$A1+d=;S z_T!_1#M1<2>4T0RpcuO8BFCJ#^hg&SPn`pa!uGQ98(j)6()H9RIXk*?)c9-;&)Fj8 zN@Esn;iuVAk@8FuX@bsl3Rg6Sv|wzlocien@4z}0$a622ZJMWH=yHH%NRXRh}7vq~)@NSF?TJ^-w zm3$22!DLD|@c@Broxf*TZ&r0xWtW=9XBnK-v7P8eLD}3QyLV!F^Dg`8C&XW2j%azM zNyBHm{6Cy%r^2w}?^n$xGfISKg4fO&8pmA`gBN5Vv#rM?(!o$cuk)I0<9&3~sat}I zb7NK}_KQ|pZGBk(&DITW?#@M^Bg3pHDIl6oaj9Q$~)OZGef0e7G z_pKAF?UalMLzlG>944f=4}bJsJ>@_`+jY+v=$V?G8Xmez2soEg%zt`1Cf2J)i7Qx7 zmghz8VdjUZhy7655gd}&TEEc!Skw#|2B-HNZo+GN!(e|MhmSB^1zLwNr~P&-9wj_w z@9+F_G_4)Ptdcc)Nf7vo$p7doq5#<+yQU^?Y_64^#(|I0xM!VAC99}pW!D$m0b*-J z&&BRk7)=z?f`rl>GK5rn?A$4E|MQg#QVZ^)@Y%yd7Wabq}QG@NDY?;pR=F%O$ic( z-SJMf4bR2}O~dgHQ!&q6I{fcPLvc+fKGTp@NV2#{09q_ZJ4kwSj*`;gh}+Sz_{G!h zAR5N<{b|(SgyFBq6H{ALkZ7FBdSsEwZUdzd9JW| z#XkldvFanU)|@q+856?MR}gTQyW9i5S_h(`4X5eIt~C;Gjd<|2dF2tHk>6DSW6<{L zFha$Pi}`>UkTu>m8AFTR-HS9MbFVs=mjztQ^1nNEqJF0OZp9_< zoqAMkAe`Vte#M&Veeg&Cm`w0Fh1jEZ&}H9^xC&?h^iVGYlgtCiTFIKl5}kG3hn>us zRtOu~(Sg_#Lxy0-Y^{%v)9I?5KM52*is_;eFgf%o=4??6l%@M9?oFsn&0)k{TEszr z7~~l-<+(a12!FNnvH$c7tbzFy5(XK1S1E@<_7$Lu?9r>Rj;h*rkXd`Hz2&J4uvn8g zG$UX$FRUUdkQmM^C@`WP8p!#CyC~oL_07w6kg-a|+PT#>BuSzB*AP;}46s*2{g)Tb zC;_-clnLPrIg{9s+xA5kF8r}drfr-jffZ;Iwi&uKGG59vA_16qT^`pu*!lB0)4h7p~M_v z!)1c23yp@3>J72-VsMu0%4e;h13`j*&h@Os#TJIpAsOH`+?3thC~#aEUimJ=6|d>m@Kye@G5T2|G;pR_KgRA2ImgbH$^cr`8G|A7V&KZp$QiUyVOGzK(VRwQntA0EM3# zi(rpqI-UjQ(89HoKJ1)vW&67bifQTjX|ow`21@0HCtp;!o>*US!w-KOG7iz9h8bHe!lYcrsa=_N!?0 zEu}F4ywZiszXSE)8VaFrI9yS$=qQl^&eI**6||B6PmsKt5EQLZO=0XfW%Mm&w6oNp z>J>m@B5L8e(J(>LVrj!nu-B;}%yYm=9Iv&HX5z`HH(EuT3Be1nK#E}Ufq-A1kU26XojO8fAc zC#s~ZvD^<@NDIJ5y<@fM`O=#Gfo8@zeQDSG1FJ#}@r_6lo~$vniNy0`G31U<3T4#W zq8_MI4#X{8_uMyZ%?gwqUtTaINve^%O2G~H{>DpK6(}D+?boU7!+C{&dOmRV1>kV$ znDhqkMp>I`bMLDEQBJzhZ80CPirPED69ZM!HuASd&6S;<;7D`T2J&g|9!0zR?yLXp%M`pKa=M+pqT?w9W}GWP2UYc8=qNXcG62(6z2Q7o03ZV z+t5wX^UWQfnMXt2j^s4T|`M{4+x}F89MLli|%sKG62jBsB{ET;$aG1Nm zGr*a0msdJk z8$nn06k8v}SRb_I&Ye%V!F}Dz(;^_&Sc1Kf4v^{M?XD&I6xcVDBY|P}0N1@Zid11q5X%9p<);-ynzo|G zPTbIleP&u{6Vm~TKO2t9U7$(`H9v@WeRS9sD-DN*N>N4MN2GssIV`@co{hwU{I4?l zvu1!XFg42^zp;G9hS1X{+6C<1*SKn?#c65^arEXiY(ro6jgg2r>7J>b>UoRAVn3In zgMxD`P4@VAqX9=1+iC!DP}|9)16XaH2fu5}Wv`b`XSNI)>qtOy|{Bq)L50mXaGISo0gJhlmXjXA6oq?IhOcLcN1(r9tZJt43jK)p#SWQzprQ$?Gh{kw9$uUY1*Qv6cC^fwugm`HCn2;{>VB{!> zh*cG}7zUQgTU_{il*A32+G!%1xLBqlRz}S-lj}F~xJ#|>(}IuzWKs(^_%wDHsUn^F z`sIZ>wUaRnQczj!cURZ814vGSV8eVwNQhFl7%-yn!MfqjTfcz|x!hx!?uoI@I>~hgMS;5UU*;BZL3oCR}i_T-HGFFyqK`T_A1I#tphl zEf=yx6g%x>iiX1E88oQ~$s6GD|M(rQrWRb`%4Kcz@w4Iat~zHOIDMJ|D(QY$a$Ttf4Iv8Sg^D8{;GFF7SBgFA(Z(!$3H1e# z3|-4@sOwF4@uT9#jFLNv8e9+}Eyk4EasI7TnYteZojvR^R%<^eXen0hY2Nr`kd7$> zcm|L6?m%VX3kU}k^PdTagUZ;JwX%jT4a!Q~b^xg0N zN<70FKCXh^JGL`}C-tdsOu8@Jn9$@Z4}w%wtj+h#aNuq&zzVwpXpC+_J@|&!HqIgU zd$^sF3VgmIz_hKT3h(Zx4h`Y$5_+HLvHxq|#Css9@lsrNR=x?nmlvhR660n9FFAcR z@FL~F^S>#qD^K!eyP@rjEGZ1Ba-U@Fl^TO^to`#&JwQe-_<5>iXk1W&Y=G@CVk0I% zFl+Yam92k$QQ^~Z^0{)a{viMDJkwRtGgbk8hU9}OvomWmv}^NM{TnV%-aAEzgaJ;* zEG!{30L)s+_$*ceN7-;Kd+GzPP9-K-uHC17? z_xJjf#JZSG`uz;A{e zvDxT3ESAZ3fVbKo0hcs{?%~FCzuzJ7M(dEIi@obPU2g22UdH9W0UHx42+NfNS9zCj zfC?bO_2u@lhdMw9-sU(Z)c~e$a90lgir@z6zF+J+=(Fjx)&XV=Y^IV3d~WY;L+qZB zWe2Ox6`l)nM&U1fyWml}*rWLni?jE-c0O<^a8^u6!=VE-@zj*Dxk;x7h$_#a-yk{O zd@ChkPr6OC&l7@*6L_naCj$SZ+%;ESh`p@sOv;Tb90?hAxr)seggG9X^T6Vqr!FlNGvo*y}1y7~n|e%=K&cmuOsH05N37&MBQveo#xqoYCYO7{cSy#^ar zTt}6+Gb7JPHs|mG)bi@WfwH9P0ijnJa&&crRK~ONZp@^e#+()?@gl-Uj3p1Fz6ei8 zOzPqW=jq4Lhrlucpmk62qh(9SU8B7P>jtcNkf;9OKWMzKf1zb-I>;TiN;gV&BNwtz zp>KZHAvj!!4D^*j)Jz-F+~^-P5KmGB;m`si1&~pz0+(0#Ke@b3r5jy(U9uZZ#k@hD z&Hcis+2VEhmJx=aFrhHLMY3QD_b)C8G>E^>!hJ{$Q8J0n`Fn|wM=gi&QF?(qIFsE@qvw>qNiN@+Ic3_ziTfpTsecZ^ojx; zrZX|ponb{y%!DQpXJEYv$DROlWN*4{bRfNCKrQ)`)Mfn_l5t!TtKRFtq3AYAB(tc_ z2hW$B6%i|QT<9mtU$bs3o8Q)`;kIumqn!oV7PlGaeKD;DHn*1)Z`u0^!ClyXc5FL% zdKvmr!+zMW3m=G2U9e3CVAZg;rZO`G8a^-g9X*{hg?C(YHd6c_ z9m-9~#!NE6|E;TMhwbq$&kvV5C<(PmkqsOackx4p6S*$aff~+u=UVK zre*$Qu>u#=RgT4g-E%5cS^NdMs~k;*l|Twfy(O6ChZ?k*&aTmOl4e$j6Jc}TliP$C z%)xw1WnXRf)wUk>ftT!l-XY~?GJ`1?cM}<3NO!f#Rna=lUw{4#Dujl_-hOHMBGI>- zZfIe0Ha63=-ZDJ)`{biUGJJ8R9*nELtk;D`B}3@mBx4ta|KgEv z3KF5&&FV+Q*rP76dUow!T+pd(X=rsUVgS@v&W8#mp7B=LR&R^r`|r>239X#aZ5v;4 z^%z;8@|Ln0_QiY?aLLnQxwW2Rc4m!;4DnH(*7{L267j}ii86d3b8T*A_aEsqCu(JF zxoK`32VcCQEEY%Q!l7l3+9B`&vz5xSi9K#4f>%ybwz9ag^rLD*hTqsY{)tEZnMY)h zx{k(g>DTzNSh7Fzhxy_h){JT+liyyirM&NdZu)+^!kr;e)Q{C1f(0!vO5jb{kFn zxp}R<-g)`%I)xPMYivF6wBo@Qeua|H&j1)K)iyEdFA`jPf4GEq9k1MsG5^hVQ}0y8 za%@Qze=-TrwzzNzY=~LloJ%d<&%)LbvW)(-_f(H2 z;CeIUK-D3nMEhOQ#HYdA_~tDwR~n=6c92>bFOYL(;xv8;xtL}f&$yMqrEx$9LOvHx zOqo`O57QN0`>!tXOnZood4}|d@+^SG>u4ilq@oUr%wNyS?dI~g$J;W#2mFHGy(MrHfVo4(8bzTghONKfs3zFA<46IB*u^1 zK}EhSK&+$a-X!e;)z4-e7+CM!W4HT48bipP)!0QtoCjLyIFU=jgDEEc&R@J;^&?sM z=d5tLet$a%Kj7>Klg0nv{Jn$iAkP1ri$(xMGg0&Mg?g}F+j`x|U9-n#lQl;u;{H6% zx~vS%0yr1m?<1byLYd!w&eY=_$NuT`p51oA0QUnZd&s%PK;P-Gng1&3U3LX1P|LW! zwIHt%<+7|B13pnCV0_DN z$VA*u_x0}N8ml_@tbC7*DxrrqOu9~4{kKmJu;uRTVhQZ&8JIWIr~uCvRIXO+wRvRv zUwowk&{93I1NX8VXK1^MWuL^$NLXELjoWR`*|C3NaDITPKwAXP1!O)hM9Sm8eLRjV za84!ry)U_F%K|*xbJ$Ek%IXsLQXttAMBYr`OT#e|Zj7Ao33i>VyT*?h|DAvp^dNjG zS9dpA96YS7?bUq)a?EW zw6xR$TQ#!Ih!DbwbwUMxX<}fl(oMXDg-OBLLcia#aMSoDl-6GZ(Nj1qLac zQm;YHw|X(z zxg3WCr~?{@B)0>f7DhfG(XsQXxL!lZdnKy7NUR9Z$yzVzTz@`;J}MMD@JzK3bQV8V zKS|Y5`6uw!LUjk|(6w2B8MRsjItiwoMDY$j$^p{EzWz6wO<)ND-h6r7NBZv{}!$Y8*Q?c`5xzy zihTtJB!X}k*zE#XQp$Yo)bXp(|<>&MRCf^dlzC-i2| z5mEjX1LF+*@kGgsFp>f0QQQPbiOb(&c%Rpsw~JY*Vd7!aU z_33jmC(a&oUZ7`EiQv47#V)?~hwWXgP2n3SWZ8wvN?!bMY&{s^V}XSb1=xB({?@L; zh{05wZ6K2AdOxJfv=c+W#rujigle=VOa4vnoT&2xxz8A76ai912$10dW%!21B~vY} zwhRa=zMnzid$0re!e!TMs8_~=+^LHA$oTzjsTJhZO`N*HGoqud59nDzm< zECq*AZCXzQO!|BySa!Wv)^Oo;^aO|@%tW^jCtye2oSaI`hF>m#bnwe4{L5g~!&B5{ zRnEI_yI#hLv#c!@A+;)0E4AysY9^({`Sja3h@78WP${_3=;_NIwjz;2KqFl3ziO|t5il}9+$$ktXo(ii zhPA>qPFowL&qZ=_3+E_X3=mv;WTEfXR<-h|8<%_;;?@LJ|KJONw(l6{{HpI^Uh7Ba z7+#yZ!pDFW26KonE+UR~8}cZYd^ae2o#;O1^p0=LiU9k;EAg z>OPuO!QJj)~16#!FRqfc9jtAT9g!r2O^k8TWiULsY*+Oul) zBn)1yG<%Wo*evnZ5R@Q~VtXafg3P`|kCi_$P`k2w^<&`*Xa7n5CH#fyx~7Tlf$`#U zUYU~wMNHkv?G!453)|3%HGd?@Kub~b7YiRa*SO0IGQENRXf`U)_Kp79qexI7YR~Ynl&t4(XGNyt#qMQ zyYQt{ni#x4z25iHq0oo$ZuX5qKB9w-11|<~+ZEw2^~8Sx?Q^x^ynOTVEDguuVqTfE zzmS(wEOb^rpdDmemPA!@R=aU-y6Tjvs~jArq>ax$=I{WvcQQ|0X-)F9K=Lje7D>X5 zBg2dwTf4yg^$I`G)dLfFG0~G)=;mY)A@#%}B8ys=dtwIe;CLw)ItEDMluk*|78NYZ zu8k2MKK!RQQ@RP&Qk$Ft{$F_}B5Wv4LSd}+1<*j9XeqOjql)EO;zNG|r@jd9o(;H@ znCpOjHb$a0OHE$&2zJdG+o=T204EWN?%^|7u^-eQxf0-b0zhDNkRpa7(4WR%b zC=2Te<%{dDB%u$Pjs2=p0O%{j$1E$)6LT!uSf>wuW9T!@(zpcyp=R@wJT9_cxcZhB zcPSff2+Z#jH+if2uA~4X4Vq}E2N{9CBUb{J_uUw|SU$SOf_hBisgyuj(*^>&e&=inR& zAOP+n3`e-yK;}s|2b+&e3NOClgaW6J|J8zQR}^g7tpOILlZWf;SlTR#oZ>e%9RMpc zlhd+WkN#^V*}x_D=v_F$P04r^gO{&uj=zKz%uu>`8#}^*o~#waUchTVu5>1J4QkgU zd~Ugso=}9t>L*Pd??AhwGpJ;6PamZ@U;}^*WgmfYQs)QnX4{;XJyUaU#C;=N6CR2S5AdYDC`uO4tfgIKxH{365i zM?-+NCLvgaCJ#9qz4B6k52EIi{aiWp3YYlDNR&LVa-g#TiiMR$ci?-#Kh)d(vJGl; zh$WSpbCxjEkkCm+e``WkZC5aItg~uBA#z(< z8&Z7cwxtC8t^*bVka5=U)dE!jf4^7tDoX3{7+Th!g(o1j4RkOKBz=q*KnkaJqXVMfMyQYNd`-5Ko6#Y zgEhft9;%U<7h^#8n27Mb9m)%5iQ@zcvL*?I%oAxC?5FQ??}sXYzJFg8ZU}u9|7Zue zJhEQ~K6}4h`b$g78Kg?v84`Wd1`?`R36j@Yvmxr*oqp@_&;yt_z|O`gX;Szas*j;h z6Na%uSK-~?N1xAB;(*<*_y>UFcc%QsA9DfGI#Cjp1UCzB2Qj!eJifXY?&4r-C3)W+ zcplQcs*0)4;uQFzFDLNdPTE;B6|q zB)A)XSbDe0O-3nf)t+4{--IkxFho#=kiUf>u`8!hsK_eln;ZiKc8^CX5>2;*mAq1x zi@o(T3W1M?gmwL4*W!@(`vywx8j(?k3p+P%1yJvO0MU@%9QS>6=>vw~;7NzT_z3kz zF0$6@&hs7H*9}`5OavCf){>U+@c{r~yW>;gsr?iSU*XFoxvye}Ajvh{3Lky-!FN+c5X2{VqMYwnqNOucciwA`3jaSC*u94TsBUX+dzS zeE6@mts)XSJ6^H6$_YjZ)SWmq%iD`%1Fzca?8#BPG=BY^&>kRU*t9_O@HvB`mZBEQ zFp>^V0X^~+dN;HG`tg9jtPLVfU)`I##-tFrDvBqE^Vu^2#q4yIWNQ5d;^s zv|1bw(Ho#K*XntxXrc{SnL+U#2>jwBYeUK{;VpR!2IMdjz&?d-x+&Yc&z19I=nq*W zWSPm3hJrO~)pB@+@?imZ!7ayqMs3tJ*sq4O0JV+*SmXRJ=o4EV&p_}VxWsHk1%A~T z-1Ceiz!Zf)RDt!1a$M4`q?h!6{&8{zsPqHI1j)9k)wXk?&b7jDE~pJ22U>w&DdOrG}e~o`TC?l>4Kt2J6<#uS>HjnQm5)BosWtay6An$=_gS3A~ zh1n9ihAmQ%DngkBG+Gc#On4U4e=6Z0ciN7p!m9z_vMmVc-CW5y3zKn=Z_K$6#;-!c!@mU(^5D&vy_0B- zTrr0IRt$#%5gUUn!0cNWz(|X_2O;Dz8{dD?-HH6q{_pTU0>+Co1$G4m3+c7AP*?ue zK*^0D>)bKmNHs4q0?IK!nE}?3^lyCojQjV1;MSW4wxYlHSAWa15vU2MzHW1s95^Q# z1l($a%Je%nia{Vv0CnX&44Au80)OvA{+5Mv2gV6yyEgUX5=0i7buZ-CZbOb+iCu=Y zM;-VY+^+9!_2<9k=%{(gIg&pO0KR69^?T>i)0Oi#7KXvObi*8JfUjA9iFFp(u+nb; zNs76l0`6`O;HovJv4Fm;GY)ILf8$8#hpTPX3lxF(`u~dH0X_o&R@48h`wqn7U4ZGF z-)QJBA8ctk6{*7d_erE*Ki9&+F|^d*x5I2a*0;r*>e_(%6cbHmgY!t(&GolMEgQO) zrj)#JiVjd9=ywGID6_O#*(1rH+INvq>ey3=nNu`9^$$^dVd;*_I+{q?&g-tjl35K+GdBZzJ*DssIX3Kh1xuoc#a2@cW|xVyP17 zUp-iV6fq0@SJy7vqXUWd{X-3!yG2Oq{#hT!zX$|G5vmo;nQQH!lmGk#oB1GtRu>{? z1;;aB{!sPCFWr#62|TvYmQ7&W32knq>qG_q{qbllaR(}(U0Yn7TA<&Snb=<|<{<0J z%G^ll+z4d9z~{cld0tq0;$~jSYG5562F83a>mF=)eKni!Hh1X0ova-9sb=Al@ebu1 z*HQ)1ggKvZ3rzxtt{7kdeqU?0&1JCVe2pX{%@`~n?4S_|5kVEL{x)uO{s7ctH*r@MO`)@EyfcEMe9uuVu9BiHa_X!|#g&)vR}Y=!*V4Cf`yyH}KK z03zu6Lg4VXx3>m#fi zk>gj6qk#UB4A+;!6(P=7xCPP}=sjwpj331k7&Wd&NMLy!bbVHEJF`uUQ(K#i{hD=~ zfek>BE)d79;+_-Piy^7TN6B5+*D!A9ASbK84xz)6CwVzv&5r>4NIS0+HSS-^L`kCP zCFkrXr3@Ls6R@qnwD|8|I*;7W=*8^l8Y6M-X>xu$XlilDxCdH$Hgeem+d9X|$BuJ- z{|MYae+cKy+zjmD?4BsRb4ARp#3DVr^91WZKED-h&AadiD@y*dL`r~HxenxU!2PM3 zoPMrjx0M^|@@NlzRLtOk0iMIxeAI zbtPU-dj6d>CfwrVwUKrtHjRS@GGiRkmZ|&tV(}s8;x$2Z*GXA-5#mj{F+cptoHf9= z{GQDL`hMV9LFLThEt+v2*CUr<5A&{XLW*Dv?9Sv{DO$%tJ-XA1i+GBUV(6YJ^Da?S zp7&Y9Nv`c)j%YUibd9jegkZYIN4^MC9bJtV9E5A}oXXdFb2768n6N0W*I?)#mDq>M zra8_u&k}UTE#Bx|B|*3A`O^tJeM^SKRA7Gt0Y%@8^Op}=J{ZHtA8wiuA{E7pzQcZX zzI}iV6#6>T>WaB~3Y0*8v26#|l(HwA8Qo1-8S>&UzY!Mt=yg^GxQu5^V(PO?#65w& zs&&S|$j!3mH3l{E!qhbTy%-qdsK+_t(i%e8v!abX7rv+cUiE}fvmU;zr+GnIEJb?A zR)2S%`AH*l9E@O860XtEGvTe-4b$3_iuSVCVxF`~8G9c`U|&4z)aIlxmzIX_rybF_y{#~&4M;|IAX7v zD=eAX;+iA&a)gfNt-w8O^7_}qe zULZLg*ms2=E^)zj zMF6$pfIX_x^=WTCo=goJzOU+)b1+Cn3K)9>{w})@-FGa3W6IU$UEJ(ybo9DXf2Wa5M{ZhPw+TNMB`CD(sOjte zD7C#rzx^KA6m=jcHOj?pIY`c-C8y$s-Zx;I?Q+H3+nvOKtrtRU4g96R12k8-B1!4| zeh6_Y+FyJr7{y8dum>ID&VwS*0uk12rg-Y>#$ip%5f;WOE=U&t1wphAMP!9lYe}W98JfeO zpxKt3>4ddQ7;mE_DlD*%@Ns~-TPw~eWD2R`C2%~HK7M=xAQH+Sy4ozcS}^F7U9uiV zD2QQ%V0n~VjoXMtYZxMK%rZO57bc!6t;&$oZEUagC~w}5&f)p`85+QHVRVM35ZHIF zOyTK{k+V0%nqF8aLHn~Lm$JZgH%U|8->V>kj!}YZ zXmm-24pM?S5*>!AvXj$p2^u`?J}G)^)p9CVuoFx)b)F0`Vz922w1i$duIruVsFL_` z9dO(D?S`YkE=|YAIAk&F%jK|)rZmO16-m0w#oU+G*F%^WA%p>CCv0Zt*4QE#3OoLC zD1cdL7tBB1e03#``IAJA&WL}2fT^raV#bY@Y@!ou%mRKIBF=)J&%N02ESH!eE|v04 zTic-bzHWnS-_jI0sQZf$clcQ`i>5xkX_r|<_Pq;$+YpwE+kY`4)C2_3f%$tHZ&s6$ zYU2?nc?f2mDlzzB@2z{aO$?$(3$Jn{>8XY=j@Jtl!vT*F{1uqSrUS;5>Mu9Af6TLR zyugR|P0P_DZ3v$b)dN4M>}~^{6ls^ohfor4GGuk%JCPGhh}#i_Xi8|SJ^!Bf1CEDr zXMJ`iwu;I}aTqwW_#9)Rd8IvFHowB0jE@yP(ed0MtPusF=&EKHW00w(~U5^Ht}BkQ9f~V7=IG5BRbCXxESZ)1YgOA!i!~gITsl} zJCRmCJD06P3UGQN=_Tb2h0|`t%j_Aki8yMMFdIPsjE(qvCG{VPJ4OMOy849^nqjz! z*C`NuA_G>P+>X5Bscv)UJPZ{ zQqma?4C&y{a}T?elA-1Y7cn?+cDFL`W%p}<;k49@r>^#`|LnncW$xb%{5B63z_;=O zbr0HBi70l?>SoIff9B_SXwK<=V5ht8lN0YutcY&d0oYJae@SVuYetn_UH}{=4U)Y! zKB=6WtU0Ecm(YMXH<3u@%`xqo&}LbrCrNj)+piVgo%Q_}!{qt~D(7i!o*q9?h!2Px_p> z5oL9(4-HWriE)e3FlStkAO>S-C(I%`PVd2)v@4zMT{AZZy@TSzgd*onIBMS2Uv1XL zC0@61T}!z48c4P4Sd7^$GgqIi|Dq4z@jE|#>DMEkE#R`=FZ_|F5+23Z+>V@Y<=VI} z-{E&PBQaxk&~AldR{1#bbL$!y-424kB@ZsTrI~kyJ@a(x*Bf4L-9yq24_{Duel3X{ zDmSa!_kwI`>~X|=u~#Dx-&DTPW=Km@V!pDEg1hE6fYAf!mLs8Gna1)LdoPfv6*}u@ zNfLcJe0l1|2a4>Z5mw>sbp-@YR%|>}cw1MLI{YU3vi*%|Q#ND6e1NI67Fx)buS=?* zm9(g#p&H>@?dYDecl}hR_4zwau*~u|whxy_CT=&0Yn<4E8h9`O+dv2a8^PVcR4nHd zFW7w0dnM(QzgRCU@$i;(O3326yI zy1PTVLAs>jyY~hapE&2d@B8cf5Bpwg&8(SOGuK?h;ECb7eIM$Mco=fQcO~ns7e@&* z&HGtUm6J5;SAHH)gbN^*rgPSOdSvAg^YKA5F}aZ=GVSjDJsakGvlhVu%NKGAXb@Wm zWN9INh`>20f&_ho1#cwT#N7c7SRf|}aW9l>`J9Rm=7e0ZzSy5Cf!t~|zFpS;P(E+C z6u-WgX~2c@U51!#a3&o>-=kBj52hKxypgQhX0R}l1}Cc^SH0;Dx!|P4i33!F&ZSbM zTQC$&67=?&+j&_V7=F)y@MroQ!p{{96$iN6M)Bku>6z5p-^S95)A>bha{KlPPe@t& zN=~Qpq64%M3cm#Fg~8=sU(c1!!~rIr58W@cxkv(Jg71aS zLA&&Jf=GPWtVMEUj~@s=(gRy3n7ZrfE`Le<|6KVHdS;DE@)RF=GVs!&{kLPGq-sTY zJdbhWL&>kyXC8HrG*sc^@)?+3Q3mvp+3QAH zE*Ki%bNy*DvCcMcE&9>QZ@;PNJp1HIEzjy|ho5v+P(izBA_v8-VnmlqywhntsWmK-g?S?k_RBq7ynW{feb0TuURZU zyFaqpa}ION`UdgfYgZZN)3Sij5>|o*po%kkK{?pn0Xjr8P;hxiUAzedoj%zq(5#ra< ztFc+x)KyVoR@Fu1BaxI}}5;*AwVNlb`8ceO6Gm(X9FmsV> zq1p;d9AHunlG$@Ph0B6a+j5M_E5tRgXHvUDQH+aT7 z=r2W-6&&d_w#(~`S|5;xl%nuTZOL&W<9YJb&nj|+*PkA!?Q$le6|{`_)$z^{nSJbA zg7K}iIx2r9Eq>pmXW}lYg%U(vehm)$XB{6$$~45l=2;%o6fg<7eHZ@ieQwovu-)r& zOu#vSzh!M)X48Ynx_O@RZK=<352|tUfwOu^dMZJw!QTkp|5AJUX|Bd+_p#PoPfuu% z{M;7ZkLO{2INjQtK4eFQ@3FpUstcvnDZpMqSo8A_HCI3r2ILoHafCBD7XAjNQxSf{ zy&r1sI|)DYw|)hbarVCr4*3`T?6Pq5JDB%>5ShvMOve3W-K#s)X+oQj1La|JNkh~O z!tEvZ$UCkU2O&d}#J|l*`4^#?g2oLO z_<>fi-3h8Z32a2eW(vfI87)gQu4bYRtFJa4)LbYxR2|TT`7Ze~-?Ktjd z0J7Vu!4(3*4K0t<;Ln2OFO^&sx8&vcj+2QUA+I1pw5}LW1p9Ior==&A z`9n)TDy~&(MHkiDVDOnBWsv0&j$Tb5`A`gE^@`OoD)4mhfy(pfhce~4mSxEQDoZt& zL7cZfWtjvdEbPb^2pK*MPmrO{_dl$L7AWxjAn|`p2DJ9=!gzsIC*oY(P`evZ{b4K6 zI^1w;L}=BG>spX%6920p?+M&lJ5}`gP&NQX_N@oA^1Y@J+X3%;k^6HXe`AHLaB_x5 zO0=Yubpff{tKKy)nJRw&2)xUprI<1~^5WY+5LlAAKdGiZ zb_cHaHe>s63QlcMjeSJc&jT!5+Ri!vmHr2_(T63lpHJ1Czt(0& z$IKGTNB+}DAut5$uhYW*#f)sHE@AN9Hls_g@L8oa+sc5+XnF$VR;hqGqbQP(eBXxk zEjwk03io`C8dT?Vq7iy$LbcbZ$~G&rA;%v&6@%v*=3%;v6wn+v+GDfm{dxONrlK)R zI#)Ff+*T`|yf-hP%zw7kqwQFf3BbUI69l$YMv7w!oK$|)*pX!hin50x|w3zDonHAsO1GH0l}&^l8miS{|CH+o@4G^ zObT44`lABDZz|=<#74@VsiL9}1AWpvFwv{!f>3Baln8JSi?RDvwxfp-Y7fCnGJ55n z6tBGtq7!+fv+xw#F9~zVXpKm?iV`SxHzL#iOBD~@BY4+MKN3YLqji)1*$RVZo!{m3=Njb;@O$>4OAx!3zER;>l>nse=n%Tz`- zf+QRsQOLQMNLCSXq-j4IlobT6;5!Bj&$OJ(RA}X<4VX00Gq0fiPk{QdQ3M|i5d7{=f)@7BIq#I^FIEeZ);TcB$ zfa_=Oo$F)3tVlef89LEa-^cVXFI(UxD7Hi0eq4c-0W4H`Crg7J7}84t%32;37~d0j0doHH%AxR^r||%vOFHYd$nCr zzPtY#9hvf_H$+KdY&m$HPr^>mB&}KA4mrn_t(C3j_dS)-AWo{5P9%fXLY?fSmYeKv zi>8kDELel8VYg8KrcuFp@pLxoSbB5sf&SzZ!x5MXsc3YT)25m%c6_Hz3tc0gHJv+7 zeX|chgy#3A``S6vUX*xtuF5^Yu4NpGT8t(g;@?1PTY$P_IM8rlz1aaqC` z>%YKhwr+8aBWG5e&II2p7I}aPN*?HXfqrk~mF)*c#r6kP=#Kcp5%qvwuJ2xWoQ$(x zTjD*rCZu*0V2Oe?h%u@_1T>9L67Qcr zB^T;)UO~N!F8?qC^%=~Plu~!x6pLs2*?Vt9B@0F_JYpiI0w|i{?crXOUPrR|P5#$V_N~7Ts z!sQM8hxOk*HP^fm+(%Gy?XpKc$)Qa53%N?BEzjqJv_++;+Z2O_`04Krbo0f%!>S9< z(YIP4`4Khnpyf_FrYfo@8+%~MtkjI*#QSy!YqMx4vs$C&i%%R_wwygiwBi$&5*gWF z0qm3KJ(|CbyZQ?(34D72U%zhugo`c;y{s$G?u@lk0K?Io$7N4vfJcj4G0t%Pd1UR2 z`0-YBoRmB3+)|V-9=ua`zquM(=;OlVuv%wl+)~Eni?d!5$xBtrmk#niQn!~>6*y{7 zeQb_X78}%8Adv58y@2W=br+jE4P;o3#be4Ld@o&6Qwm$bT$xD4=8^c0?i*yS1nzfF z`@dlb^EKiDGui$!weTiq4&l;oU#J9|;9oq5~3xbo$alr7;q?mGL>+ZC}utI;S`kzZct06;+AT-3)*0 zabmwTvzN{q!oNa;x96|Knnq90FS`M37f zO3G&&FPJGcu~zk3O1>m3><&uU7#63BCKvH!Bmj?^hA)FS@Ky0>u_zI8c54dWiR6L) z)Xew1pqS|D&0dWfg)>@nd13_r_xMBvaHMx(_@84~1q+*^1v z!@K`Y9SG@yyGhF;#^%S33d+-u8W<;%;#)uSatl0iFnllMl8m7=HfCDqq85cWXW!$$ zTIt@Pb8ea9OVb}G=iRkMAE-Oupg!3uWH|7BodJ^=^R;6`$7-5$NcrR}T7x$*BGtqQ zBoB+LNF;eu^tyk@+azyB&q#UQ6p@ll=f6HXeC!blo_n{YHr#)QM8cwe5>e457I4&~ zVr*wd7QHLj-&2+q3OqHk**4lpI&1^{*)ObdF7>8RUkVhEj%S7fhM|lFP9AmPyBL~O zoU(zq-?k!?Ur44x6mcOAyK)T*mH>D77OTlC+=^!$2#o1jS`^YGAI+^lQ^pC9GM1P9 z9~>fHFf)E<{u%T4C;DY*23E2&#$WKuH*HH=2v)z*>3w-a^<~%4*aGy|PuRb#9 zE|Za>__fg7woq#~T<+)aeo1kD>5@c_kCfFbi$A@!mbqFa&XQA(ED@owZ_B^OA^*Vg4NZ<$1DE%v=ZIxqX&!tb;+pASpq?jcW1of~|a#%)9X#mcHB zBHy8<;(Q9~#P#!2RLpo(qC)0teY8hBnHjh{jmf{PS~n z2R_fybx`bBv40GElJVi0ZqsL8_lPNL=E`bD0_D(^S0Y^LFjEmg3nM?&KS%lYk}lD7 z%)3568tR?PylXDf>|G%>wQ>+kvA`px=4o7V-><5Cy`QPMPQNP{fAKwKXmoTaM5kQOa3NFJgAeG8EV zOzffniByu?Ki}dinynbL0^6j08z{)pt%Pq-^1p2>OhtUU+_OsR{@gnu)gI*%9W|WNjhVi#mbqVR6OoiLR_Rb0QD7>vi(1Oxyk?}SVQ_4U(ul9 z{1|v$sYAabJJuS!+UE@AxX1iNYQggS7fvAyauNn~s;}E3&~*;sPT!RPqeQLnk+Y4| z%kuGhab`h-J}aqoIL8F$$YvU{C`qMy@GB+oB7oj4E0q7dh^O)hMTFuV2 z2zoG64@tKj%rm%erRKk=9LvcB4uajch|zT0NIi?RXI3EL87J}Xq8#)uuI*NA3VBpVhGk$lUOr1Y_oa&ys0YFZl zqXmnEk}dbTfMy`FDQXfrks$$@d-fy6ZKzv0k*Ypoo)5w9&o=pDi;4Qm2CSDrzcn#> zrb&r(4*ESTh0fXelzXI21^=8Rh!S0O^GX^y_qqKX%(ovsMZDPqaBx@>9A3U|<;I>f zBH@2DHX2d7t!(H_cv+k_e0Q~zh_+B?kSl0^_j1mV#7BR=VI-~VDVTlYvcenQtJ4Zw znS4TmxLNC1CSH;t=JO){pq$DJBZVAU-GG*a%Ai+9hzi~RZBe(DGXrI39U=9h`(L>y zl7L2&IiAh2^j1#nx!2%DjYO1=B3_089#+yoIfev>jQ5EI^;c(x)lYarwg643R9qH6 zj+biOJ1T6$+<3!RLxs?#`j~u4ToaBFA4a4Sl#Jm0=nc8CwYcGd@qwioz3mt&(f=_U z1r{Fl1q&{aw(@$RQM>q*4vXwZ^$wb=h3~-O&KYs7oQql1~e75r_paQ#0clk+;N9$1Dicp1$nNta|3b)(Pmh1J~~GFJ+%Yn2g=Ki4%_j_vpWlv9gOh{YWZ zLwK=TJ{-I4(HgQ3scCKaqsls(GIWZW?+mW6_`UfGS~VZ)HhKcLa4^@ix5f3CTjxEx z9zgJ7&lInX_)~|ry{tE^-CB6!gwV>KIbSB4Zg9&V&1boTVQ5XT($V!gR!CG8tgmuowBiB?ki>vtTE4H(jFU8K zTompI`sTa_vdW0sI=&4lvRmi~K=d5mT_mJMGS(zy%D*f9rReqD5$XP=?62SJLYD(m zwkP3U_Drp%c()O#=%p*#Y0H2xW8S$<^AOBGMS#>4cK1zqkibkqK(g9i@-5<7ruc5NK1%h=yqrg|YHfYl zmzXG_&)c*>6ecko`?vVl`97+ooa&y%Bndd@Zq1{t8_k?5ukAsK>>kj+yOqEjLdMj2 zO?kKOC04v*beq@@OVQQ9w@}WF3=1V+3NlS{qSCsQmVsbPEJwxW*jwqygi9iWHXZ6N zwycLflLDKPq&Z9$v?dv&)j$qoXJa!AIcI#-2oPK|EM8~8L0@b^nrLAREemfMd{dL1 zO<5p_0b2PU?{Gm9Ph$iiAsD6WSNkmVq?t_`f9;UgJVdQ6=N4}#H%}i$Tz#YE$u+^? zE>RZ6_;Ii?q?rMMw%`}$c43>*k%Lqa1;5$xsaIj*$YT`>26ec0WnAlJTpSfrQAxD}If-{v77h_z$aza-{+paNCK{pvU8Y-~-_Kc;PN zNq0z+D-I0pf9+Ot4DK*5%cYG@B=br&QmAmvF(#vQ9Gq3YyXZTj7`-zUHq1P2o-D z?POF=hv)D`MLor-Bq8}d-HdT)JS`Y0@2*T_-Ou!5C92T7h~Rm{ZFY(y?}Iz4yLW&w z>F{7CHsHzs5{t@0&2oAJd+&l}J6M#-G&_m^nE;C-3)&)uwC2b(y9TXh$T7^VS>!{m z2|`y6nGGbA*=I!%B91?sN|ag{uL&d>BHp3`c)GkaII*oQc9#T zM(^cPv*o8^6#385DfgMvk%pX?$%YfKn&3=0z5Hf!0Hl$!YBex)>E9+jmvup_7>JZ4 zAS?NxO#v8+noN1w5oiJzk9rv)&{X}2H)jw9`{$2p=B#}9O#@>gE@V-=tvYa>6f&^}qe{(y z7kJ=}L3>@uQO3AVIOdd?(%;XJmk-3`_Er+x|0q;-Jl)_~$ft#py-j25gs#aI7O ztJ_0s%3WOjP1b^+RLA*5v46B{lorf{5misjPhM#dm>|mI|xPEj*d_n@D5m}Px=#`EQ zPcSO#yRXmfBI4geb0v{b%nLHoxax$OddV{=^$>bv!2NwMZl1w>1~3l7cv4WJLUkgN zkDgr_iJk8`?|sYnd1cj;#eVYGhvVBNbmLKI~#390!JyBQndB z>p1%=<&+e!UE#4gdM#XSooNAN7^24Yl);^r z^5i)oPO&~w$kDih4}ya~AJ4iY$7>;4M%cDI!UV!DI9DKm_2U*L8n{$MbZ7873Y?+L z2Uvadz&w1wqjQEfhH9I~_=8>quDu zs%ti14~q?;{JR}ia2EFH{axwc+IiclbxTm~o$yCw;s;Y|@u$&H6#VJ-jmmh~_vEv=UVG=7bmq)mHmog}-IAJvZ zOR&I4xkoTpGI<7?aWoN~QKa~qn`uIk;MW06LI0$X>-hYXxRU0RoX=-}Yy7663r*!O z-fGJ>kauD<^w0Kz{f*UVv~6^RlCwOwat;XHn-msK|k&!QX-vRLg5ORnAymUvJAnl+BcqQ^BL29lb{YJ(` z=~6>aR}{`^f@;4P89O&MM;m_bhqAeUEwaT{#&${O#?yLLJQmA^@)bex`zrj`{;oBg{y^`>uHzV+3?$XfeXKm6Jcb+s&Ih2Orp@M`UTy$aI$OA^7Joz$C4mraA4)S!R|&%c0};c>jjwGC zKsn#7Jr+*KFzDME)E=6@2KA8R2h}S}EF4s5HtHl&6z#qyz;lr8W$e+duZ)5~!J?!9 z7W+a>_Z}v{{>ESi%o)#|leyJfTgJ4BcEj2ogAllPkf9(@WArQsy7#6Nt5gDd0e600!n-wcPq#XPjp+V9cw<2Y@A&tdHtN>)o{D2KxKv z5opPDA;ErmwADlqq@6k(q{oo6kMZP(wJZVyF-3Rgd{)z!mTSZohiCa{@)8X>uxl(-(;@IZ3y!tt!)JPC|uPFK|yj>~ENQ zIFfsREL5F-2Q22lV9?)Nwka2S`*5ydB(?J? zDCVOyq#Vy~b>vrmzWNZ>Uq3Wj6@9k|=&l1v4cddx=GfXTaMU^fsSl)S(@+w86@aho zy!^!g@FI#sld#|mgZ>yW_1RJ(V>lzbIU_eEoX*A08_~l`hL2i1IhU^pkfRUa` znPlyKNUDjRQT&w;at~yr!S?`s_=nKHI?TS5Ybrcl=<5LyQEo~}dPK^6ZF$5#O`HJ> zG2S~jjW?FhId=8aD7SI{;5gGX7?q5+xqTz#hy4KU|CeIss|Mo%pyc#7n|$Cwz~3BYz`)w#-K(2c+A$^1aZ3as=s;waDQEfXK;19 zx!R#o15hFs98K2Pvb#_axSG1khV}25BC7k28+|wW_bPqTMfou|C7TPY`vcx8pJWm< zIeIBcSv&al&QFPH5Fq-Mj0~aP0z`C!KY^2y?wcm=QzZ!ml(Vx!e=F4mh$Dw7yeXO@ z-LSXl?`4er<`6gDrW5Dx486^`qxx2HEjd=72JS-C_iZ8O=nQFv`1L9HZ<*kmRBVJcM2h+^zAOBI23u zJa*gd65^~_Gy-qB@7b3-w;+Ba2JANga_c%r64J~7$1`C!jowBiVsAT@Ri@!-#+6XC z(XqHw-nFM!r*`Wf)(@bei~1hD&Up>eM7E1S8-@Dy;L>o}YZCtAHYfN^byO+h0V2O~ z!Iy9scuL0t5s(|b#~9pnQ1a4Si22gG_y_Z~R)`B6o46(y-+Qm5V2vRgU3h(^`}=gQ1j3lg7sCU9PurDsEJh+O=>QjU zIP(LL=}|z#!6#s>#6Hq*`+}}bCS@Gv(>#WCE<#)S1s8y(47IgJMpx!-O{>vk z#cB;@v|iVb1Ueux0^8+AC!kf45R6hE8b#lpuVl%DnG=6l0Ho;R-$OBg{0t1Q&6#C0ZJu{=y0EXkz3S$?U~Yp7P5e_$FqXCTdp2b%Q>W2hH7SJY};;MBPS%I{xtmx*9II%nLsrzI6!7Z{XnLvS>PP=vP&0c<7l-NX2XWa-;- zSN=j?0}?v@Bdr9}4o{KF<@p8ReIuL*2rdt}n8zDAAy!J|x0OofA>~qwM{j-8zN88f z=FTxhnAbhYSDenj+F)tPd=A&}^3!2lm>J#N=v)^eZfN0(APVx(YdS7?QSB4ALa$cP zByMB`dB-8Z{mx%qwL1UxXVp_C1vv6ztANlr3H@kliMpEWZ`)a>mgW97>U3 zJDCq4(SgQk^1vIQ@Lx_1L~@L``+XcbKLaLGzzmcl;5-Zq4&Osie>-Cwe2PoUSa&hW z5~-^o0^zFr0sPebCpL=n=CiZo`Ie(`D1oIU921CX)TGa~1^rZO^|#LaGyUL655Nse z{imP!2W*mm0EUZ5?W>qaIafjVe(kR$62Qm&tNr3v$J;-^f6CFEAM9ZM3TG?raJ_uSoH(m-858j4 z+DqT0BZ3gdw7R>?Ljz8@C-p{u$K-W?$8Q+l#`g5(!R1L!xwH4|Dm4;>+x{Qm)fBHa zXTpIF!p6NDy%rYS*DWZ&*_7hmwYfb5W@uqSUn$ft$}EUL{?q<$@NJd4)5E-!N8pn~ zyvijZL0C@#nkWdZm@f&z5e+!Bx70LXaDi6OJ0=f3#MNLE z0DQW_ZUJjlzfPAT3FTuFngljj<8IK_{p~oa@PTn)5IsS(uzbiImIY}h++07HUTOvH#vuR?U13B8cYso26COyO!j_ja|1yH6+=RVBv|u&8 zrY7_qcXxl#URhTFymRamgMvO|2tQ2*h({s!jk31smn2`OCWNC9#6T|F)z*E=NeQA; ztB6qoz|P$hW2g>%<-Km`&Nf#aqGgB~t& zY#|t1offsu4-a)66ve)-wqnBkfc}kX@0N#ucUBBMx;zUgd6b$qYHP2qggrD!WnlXB zw7u0rUu?#I)r|#_lk!mA7n(=ooBp~Pa`r$dwKF47yM*LWSeed)FFowftS>U#z^%C6 zTL|eeX1A}!(1_hDB>Zf7>$Zn?{L-7mD72%w$H=~6P}IBKO-$>|NtK^IJsA6FU5lMu zi5;wKiK&A4hpOrn^w|J^9El+aXnMRQH;P5J|1UZ1z*>j&?`-W=nAAGtojR=EN{O^z zsB&D04=^Z1k2A?jvLu{hG(+Q89ne)BKL0TP`hf6b*;q|0Z?Z|dPXp*r5qSh&EML-K z)j(giT-s}mAzHFGPqNl#-7))ltH3+)HmjZ%D<5a4#C-kPx;7I0Df%7SYRuph$I6~L z;;zW%C2H zy{3MsxG^_wWQNA{*31ODvnA7)N_!WEKb(l|v?f^rg!lmvO*GzXgaq1dU)+gJX_*n==_`h=;uKdC$u(Z5@SY zG%im&I>dbA>b$o_lqS%xO`H!L1f78#?3GH_%9)i6U_sNsynJzW zkmDi~KkYntQ#P-7mtm-)4w#ajUPhjiZ4ST4*~dKN{Wz_BDK&+&^MvHN0V2cmI-jFR ztLjV^kh9a==7{;`K(G$$G3n(L@p=Z#N-zJ6{80VA`{iT29K@X#hDIG(WYyahoZ_l# zdOV6dStv}}ryAKL0&j|nPo|MIPE7GD%^8q=Rw$m{c4SagCsez2v#p1ok`-{?|GZd+ ziG2&s*RTEVB+$f;R!b+e5MppdlcBWivdj;7!#A93z8gJ9J0Y__*-S=2woJWcUXXyM z5{#E)l>X6y!2kA--a#%hqZ4~Um70ewkKFfKkZsfO=eYJ&zL##<1_j`R8@;(x1xHsF zAKlZ04@l)9q@R1;a|r$i7Bq*0w`+Pp1O?&t_ae+WeJ~5-ob3(07bV*x;d(;mt3A;Q zHB~7sOAvt8IkinfW`5ja#ZgJ5b8iRKOtzT^8n3CudlTCh}|kz9L$1g$;-+r?!ZleN15n`HaBX4Okyy?f~51TXMHN2%8cFSYhz(wQXhTBD=y?> zkWhhIIUx{F-0&j6Wi9OB^ZHc0ChpKQhySEd z`>@a5L86`dnB|8RyZ%mQ@xAiH6*D?PZrk{mswSR<6QrAar%==rx0k+5l(>|?L>zz$ zhKZegz^|N)VVBmyb_tZv6KI0vQ_*Ly@OA1pj-_h>8+yXo8z{d>{_AQ2xZ&Cq&v7bM z9WL~17y34*(_c9!xR_gL?k1=4Jgkwulo>%hjPjili{F$_*_wWFkz4&8NKw`f z`UsG(YKJuiC@NOIB~fFqu*N=<%2= z`C!Yw*h_&z>cLw3jxr;p;o0p=7Kg&zes7uPk`p1i(`mOCji0zsO0*R}kOM#NCpfW5KTt~VsLu^G>{l&Fu%Q7&Lc{)P8F{TS9Izb78%gu0yo?rLCPMm3|@o{CtaXB?4-sCnj1_;P*r# z+o2q8b-(D+k;z+LSMKMhu}CqHw=t8CL8CXAl7@MDv~ga!*Q+?KxEGHvAQa#`)a8)3 z))(Dw9Thk5*vg(Ql)=aY!Ab^)M>aO496^8C0jlTzY877U8?nX4v*g0!Gpk&3vgcx? zzFhVNiCKhe_54*{4@yKjYLo5BUFw{Nf!2c z*5^AFI*;LUgg$A5;HvNKE&Zr$&(F z{;e+S`!`GgR+P;y`ZHRVcQa4EU{|@?5mA-PH_Zu)cbDCtPBuo`ldsNEY%bBY|7xP)U7eY;|n{2y-&O0?a4;{vVeSe8rl2Dj{Dc-9RrprOl zf0<8qhd$%cEUfYBtu=$dZBf5xrUl7a1Ubr|)-&YmnCtG~jfGi`EJq(i$F+&b(mFXi zft>5yonX*Ci3+~WC4Wm)y^Xg|5=?~l9rwrd>}5w)#aH8G&W!A$&z2heJ*!2Z0WwJZ zEB~zZ+0#bYt=#~y2XQyF`eP4}iDHJo5u?o&%Auu1oq={(vyk!=lmhXE(ZjnN=qIvX z-k)q%Lsym0%eQMgPrfYBzuX)QpS1Wmw&P%kS=WUKAMJVAFSse)gjhQ9feM3vIIL_k zY5u7SPISy?0)TaSb;E)eUElT;4Zm7_$!4d?<-K-oc2@lCOROof+7tHb2RUdeB|qW_ zTq@p~QU`Hrh<}Kh-m-|dawa;**Cm3Y%E_Jawp=frd*p?X@$XS;b|Ya#vm1O<6y9;SHLC9l7@$S;X;?ZLfM`@=f!lVDQ8``6(3Prh^DcD!Eq)Al=&GVX(2uLi&mw>VND*6 zDnIT+csi*9vnWEM>h{$Dc>;&JLjU;3!5mFNg^8K+%w6a9p$dkCGAp5vVU(l7jq#m~ zp3so8G{93}r$aFX$!>HXRlSh$fAcEYYsf08`wC#t55!bIn_E>QX(Lg0D*H;f0=ei`{Y_^r4;O+d1I~%x&>kr2Z?MN4Ypp*11L)ZMg@y zsTNK$2-6X3mbaNEZx7spm1EC&cdP!cQg|ZX(4$GRenI`Pmpk=E74+sV{;>&N?elDf z0fls^d~a}{NEC45<#qTRK}&l>t(iY}UY+qwRZKWs8mU|K#WmgUv?U zZ*2P;yrQt%9J@VeMu=@@NRPU)OY?W8;`d97hHNT(>fsDYFkl1WHd&ahmDgFR3y2=# zgmP_epV+`xcG>t(-dl^)Uwa}PF~GW1IxuMPHfcvJx+yS;h=tQ~b*Lh0h1xlM~@UZ&6pb{-UK< z%!X+e9B*xO_IH=9mzW(xBJ~N`JeC4 zQKX!W-JZ+NCib(v|F-q)Rr`@iR|~sMCJUa)^yd3;mG_0Brs%yMi|%P=Ql2|XBb?f0 zx-$eVoWP5!L=A86pQYZC;WXmC98S7FDTDjKFN){fxZ;^t#8dZ|g~zp-yGxlL$HsE` z@|e}cN;|`b<^#fn-#)~?TMwh+3e%RagQjAq&O02ohKWxf+OL9CY-SMrshJ%P_WWIw zNew!O-pB_}otO_#CO?Tf-yN(~34d=Kb(B@k$9g`185*KYq^5eVqJ!VryKpFtXeM^o z9~&5k_Ghw64#5)R=H!B#1|Gw~eu42rNxYF;8M``FK6K5;Zb))GrvFT+BUyj;BE7Y==@) z!Yh9wzs(tv1rj`as;(c$(s{TPCdobQS>EK)YwHfNO5IsHDe)TeQ*M|hvQBoZnr|+gib@dmM zHuB9EFw|2NYq->FuLGcV$@Uj~*+J+&)K0RyXEa>0SeugJ=|9&Ig6g9vi;7YAiuKLY+#ipVtGbX5XZc` zyW%++uPT6NM4@Qoc9dn~0K$S^am;`h%1nd35CoEbiC6Cs{SkSIJUmAf3srA8!!7 zOl&S1{pgE0vX{r>&Y@83V4jPlU-=*$vbCNuS$qcvYi-VUoe%JAN(jD5s_&PnDK*V_ zqjHD9s;XTj^IX+Zv+=@mLO zIW`&J5)Hq-$>>SdS^ep~K02@Mn@oBJ=9hPHb1t|yd3}#Qh!%M1g^A zMx*bDzG!)?UDIAvR6h`>JJULUo0&4`3*4ydl9uhC8cn=bGN4j z4@tZ-alD(>m}__xu9$#D5zyb~T?YeKygh1!H?tT3kmDco*+Ay4F8{*GuW={Q1J-UhPD) z+I_yJ=Yt1>^a>e?46^7wWe7`T@YS(?`Ui0O(@E6l@Vkypw>y>+tBx6sY`K_%*Xr;J z?!B{kNLbAW_E5YXu2Fh#o)l6Fw^sA)$;Q3x4Sg6n?#Uh?OhxwUPM_f8(4a3y=PDoF z3%jbNUN_2n>4j5bJcyU=?N1{%7QR1XdPo>YX6obwaZ3%I`8VS80RS=nj|q4n#k+u* zpi0$VH(!zl<<)k7qDIdt&bSM3!rkPn7&n;XxTLVOzD3Fz`f(hy1&>n{;QKy1;tvBJ7vwdx# znQy5Hm14cMnLCP23DN9XzkBI(FPzBV6~>6E0O|C4rN z^850s`foJK4x(pIAKQL->ttNnWA|wi(DK=x0^?+_!qwEVHz-&4cPVN9%jF3RGuq#+ za~|c2j4x3^_V=4ro?mfiD~4z=%#ucPr>yPPlXo%Xwi;Y8%-Ouli~)ynTUuW(J11XU zM&lX7K^g}b{(4TK+WFcz4Pj6MLnBSAsm zQ$ZhiJx*?-ThqkBWc(RxtuR@~Lp=Z?b{Dnyqp4EqTC#kUOCC?kLno@jcIvszx35#; zW~OS@i)?iQUuaS$xoHS!_Qe6YT!G8tyOeI>fj0^yV4w4!cO~H$&wofbau>WQ+ayQ# zN|=gFgbXx$R^Oz+-qS!c!hm^o>qFv(4g38miY>ers}F}+re@h45+D5WptuK%IS9t$ z=FG4%_Lu&F1xJ`NRuUT6C5Z!FIo^vj4Q zMh+7BjD_&XUbVxiKGi&=gPP}m;a{<%^I=6>u|cho6hV)eW2Kj#wJbqeSzvgyiNkNl zaYT^42g&8xl@=>Yg`hD=BhKi~-zY1Awb>XEgq;GPn|%UK_x(MftaoKZnV?5gc@V;D zm!)#4rkewm;+oMJLtO6hkZ<SScENhvV=b6>Leuh_dqhE^NW3%86BLf~+U*~u{T+;QvPHH-W zhRgA!_Idz&TezXFlGEH#FAg-Bp;6UKDO=-2x5EcLL2t@dIyx#r-vCyACx#Vn29~}1 zA%=j>UT+4pd5v&`;8p$UqzX~e6+k=s9H86}^SmX;l^5{m7##0pc1ok#>}+%(rSf-2 z9C(?bfg(rOXj;1U>_`Md9X*v9;?VGt4`bh9~~CDBJLA z7gnqt3Q~FH4+`*TV_RQaDUqL5rGx_9cO>|5a`u1TU^q3ZQo3PjliDbT>-~F3`e@Xn z9(bX$9yhQ2+;%3-tI!T`xeLo}gj2NNSF?9;nfst_p&OwvunsBLThRV2Fx zbzoT2iK40~9o(n@|9)3>I-i!F`M;>&r?wHHHWDt~R88lNo1!}1KkA9tIUD)xK_XgA zhUY3B4n^IuS212`CsILRrgnS4lj~gjvO-kfPS^J#Zg^(M=`N5 z9CG2Nj*Xe7JS-7nhwa17Qx+Fd$8v#p)f1#@ss}3bhb=jwq%dCg#rAAT9am3pXo-F( zacpRwu^aVB2=b5UZMP?o)5zxkTY1^n`0cwwCRYMqa%@ zdej7digx01wm{=HC=d0I;H8b+%u*>h_7bhj4vr@w4d!@{jAtu^Ykx)>bSm*N4em2~ zL590a{cBx!^Wt}0B`*T+$&vy#?XJ}~uE(uE5_+yivpB=0t-9$TLB*TgsE0#PE2u&S z!h>!~O5{HQBfpBzjhsTw@u{we>XVI`_$kc5YnLDSb;}xx3p+9c{cz~rz`1QMKiB!m zSkJgPDm{6m&2%jrkhQba(Rulg&K1rtz zh6fG*XAB#MJ;i;65Z;#t+i~kf(F`YhA*@sF%7kGqarEC^6w}rj?>Bv^i0j)jq01C4 zJqgVa6ZV@YuqU)yX(Y;1XN8|exhCDKbsgfsU*Mf@+;zatA=8y|rsOtPgxf(MY$fqjgQ zA{owtk_#`y_|FwO#e|pIJhr51v*0V1I=zdam|QF&<17M{+!UqJG@Aev$N$CG0DeJy z@yr!cfG7`NKm@={e0L|HZiiP70T2Nlz&20>8*Tqze)!(=>W}$A?V7?{ZYH1~Xl|gK z8N|AzH=%{KD7pc5_y&&XIbjn%2vH__*fjp>8H(e-0FVPq*${=*Tb7pA{!@`-HoTYw!~S z0XyraF}Eo^aqOs~Zm6QSlMGOw-a*NwR#tlRvo`j}FNp(?K}V5WjGK>Q|F{1GV=GSu zj~F(^Zt8q<(}Xa9s{~x-AD6Qcz#!bsyn>$qZ0#_0spaHrH*Yob5+2px&k|g?(0^b9 z)PTY83wv@z(k4=*oA$LlTFnFOb|Bx8*A)KA_pi_&gnp&~8)A2WEh6sl zWAON&V$#9j99Z?2&*qWee3>Q?g~-{XyHdR0CRM&&{dfKU;n%GNT`@wzZ0R# zRTW|ug5%Wwvi0jVtO#f`{mZY+Iv!n6*Ip)~nO_n+bGt*V%aXnm7|ir*F$$d{?kws7 z(EKWvm$Q#gV8MtczsqMP{~vqb9Zz-p|L-_*Bo301J&KSOWv{H1j5NrWLMnS?onw=X zY*}SQ$yPSUN=QQ4TgM(pHpe-?w{z;gySuB;eSbfnKYxGu!{c1<>w4|+yj~YK7-*5< zgrR^MkJI1RM^_mGiq1d+FO0z@!d8d@Ak zkdZ!6v8%KQj2you_sdhi-BTjKnqm39Dw1pL)}3#&k|F`AE{n3d8qvbQQz{L7>= z5wMlT8RDU?#f@ov!w%1YiQ_e1ds8(lrTNA$#M4=>F**1p!=6#HGCrhPYzPK2Zm(V^3)3LHG-7FZ8E@-c7iFCCOC1Y{WG`5J7~>Q{>@Z3VbWPqV^4ED zB;j5s$S=?94TWe*D0^C53R)S~_O#K>cki6Y)6)FH#yqBYrkxmg=z>cFBK1iVKlj3#!r<^`%)Wu&SMX`fhDYJVxvn|ko@G}A;g za4|Tw+)gl=0>4@4kDKif_Fp`x>}CHLics(v7*xrNW4^J>^-()FrKx%=UdOTT1RJwfq;8s z*dxi65<^&r%Fc+1sg;OVU}~GVQ0a?1FyB6_Gi(e^toobr?v`V%+XOhV$cFzY7DoU; z-YHFqix5C}Vn$~ej_>YNz^v?Zoi0DOuoHBA@Va5cWSPG-6*_Ym;iWDKcZjX$w#9An z-ULq`-*05N8-!L%iEQq%vd&3ixbHo)*BqWI2gAgl-JWTgUEb@q*=}E~A~8C>F6LZX z(iT>ttIrX*!?c)#Be9i0Cdo|7bpu;%{p{}1hx(s6Xce?wd>R- zD=zqPBqjR{emf102+5CGLX!&UBZm5>#~$W6*vUFYC#l!h-BziO_iBOo-Q~1W1zx&~ z_J3mexv}yH&a8T1&O?T1zfp_&UFrc{iFV*=mNJJdN4TpCj;CXpz!O>JKG!_}QyqRb zzu{}^9JrG7t#$rgJTUIochI+&T)HeQQ6I@<<;t@Y;<2Qb6Z}}&Gt~9s$A~Q%Oh`@m z;hxM{f}Eg~HfdLWT+r0F76UC4q>sl%qWkc|zL3OKP_&grE_$A5)F*YX$erCegsASL zi1O;gS}zB|`%4_p>FN21vnN()Oy{2;+HD12nN<`Z^ZztCS@qJ_W~GC9fmF7bEOmr> zxeh24yw<8E*6~P@D>wWUGP=-RO^4vnc=)y|jW`T0n5f;8v zeis*>4oL$~Q<@*N}j|c%;A9&-2j29F?dp}^VpFcCGbmrYbQ0pOQ-CIzB zP3$iEvZ!#iLT*DZJXSRS5$HLp(porsHkOAFRM#`oV>2r;%rRBqH&Nf!%T50 zyMR&R;$mS|2o-PW`V+{lT`lv1-OR{ZL41*$`0B4@Lk*$?JENJt&O@9fHG4B8_5>NER4;ZummTZ?!F4a09@5%`tCT1J|#rPljbkrb1Bzwgfrnb_Rpya-GGPO^~8Xqch9AxPnb# zKYnZ&P+h`DK50xT9gc|5S!iZVt;ARjtN($VHx?=A^A9klJ^uN2Y*g-RWpR%n_taCS zr>Z{(98xrOla3-M86M}aD^G4A_A?CG9LGt)s+mELY^nY;u(Mc%sZ;sgLcYrbb{}%v zQS)x$nw(k>4?aW6=;*F!h0n7?>hI@HPiCnIE$nQ_r$u=zf;Sw6DRYE-CPwa11d$04 z`S0R~Xx#Grl?F@#6JczZyw?sJ>kEfUbJl0HlSVX;#jGu+XdbRTvm-wW1C*e)kkM{H-z)+rtVv&JYzi5O{`y|M7 z>-%Vi3vT*$PS-T;?)P;B8%>9I+*kxWWCZ5Xe7dJ&M20R9_HpV4a@IkUfU{pIHv|N~ zM5KAwvFh9{P1i-~InnI(x13&gi3oLjt&ffaqw6Nk68Kz*(WZ=!Rf`vC6Yft8o_J`? z0=Of>=z{9ZcDa?sm_RFUPh8z#&M??y{#Tx8HOi`hh`M_MvXtk+K5|W-rxTBJyoX559(-;;uil5PvgCJJRxFw(@z6 z6Z!fn*F)Z@B;9kLj2R12x%7Hj=0ke8h@4fX-n}?0TEg~sHxm)3EbqM!KOHk!C_sDa z`cTRu`=h!g{)@|ayjlEW4Dh_%{+<0FvqRD@&}A)p)RrU(A{VVLIhJMGK9C8HBwhyL zY+(2KA8g=*y5M&bsp3UZgXgU5HYQYgAS@?NcT;zRSE-Lg2|Vl9I&Ut5bu?jV4JkH1 zCmUByKNs})Ttu0#6$$R<={f4I7u1>fe zTX3l#Sg_)4zF~OMyrfKKP?4jdVsa2~7t%c;`&agU)*T5T&JUb7#PBMOta*`L`x}FMf~%?LtZDc8KIwY&eb3TG z=Kgjqvz++r+&2a?INL+7R6cGsm06zSKn&RLS-uWbBCdROdS(w7x;^Zy0q>8;`B|$_ z(YLo@9VNs}d8Xijgpp^cDWydx&RmPcw$s zBTiw1OKzF&j@WZ&HJiVFO8=bqrCHal+hHhyhx^B-A{8zYbD;B`hZM_joyyp#IE@Ph zga=#*fJ>x5ZJyxDA`iMiEG9bVYWWJNYAkYKJKa3zAxb#EXZNaPaDQCppMH#*10D6pn!qUIeg|?=c`XlYLE|`i11Vg zid@+F4-d{m8vlOQcb{!6lCUkFpJ_7@7m`yl&+P){9nfuFE@rZqd?#|5Ce>cNoi@NY zA^gh^-aOCw@xy9``G;gpE-L9+;iEF?ns%N5*5Q@^?KVe)h}Xn-Z{|6F2wIlj*Y6fm zqc_8*ccRW8j-Z;uS}Ys7^EE$uIc#@hh@Or(Cc$^pI@#)p)`6-SifJ^~*hT3LXRugH zn);=p4LA7G;pc}GsgAk4M7?$}T0FR_(`Dx%wQ+KFlBOfqB?fJ&vaQ%aMxmiX5n44H z=5UzPV>Bh}d})3*Y4^uyA4N29iA`=!NkTo&L!qBY5Co6e0e*(Jacr1Y3mW& zc}kOK-9q%VoG%iJ8_$wX>(mE}^N|&O+%|eFOY_j2NM^M->edE4xXkxzJiVYE|J@sy zQp}B5Z_94`55?QJSK9{L%mb!1KS##+C^ju_PgQ28`g={$_zyqe=VUmFNikuDwMs5Q zmmUvA!hN*+i%v~tWlT;vFX+8q^jpu1WNkkIv+VQriRy$u5*xPDFnh8*xOzF*ajdL* zArVwgcfrQ}oq5p6`Zl`ziPKTA3qPxS3^)K@iZWzw49+%$PMH{}qK>juu}%fH+rxvG z)~c38nwBUgyKCmxz5THOagVZ`OlpM4(%eW!dWARc)>uc(s%JW{L>FB4X6r>s|V3iTaW{F_$8k-KC<6YkA8$~u=^ce9?v)Q}h z31pp>4;09TUj3TYXFe4MPhVEj&6nau=OF?-Ix4p9<#!z1V!#iC!sD+jt{Bah$wIu2 z1sx`epA{twXbN&WDRT0&ZR}d!eb#oi3&{@ZkB2CwgHpB+s)@KGl!g@e=%UI%anipo$?$Rf(KX>x*Geo$!xNHCRqxq-ZnqadJajtMD7t9= zklY>0uD1{ON@Z@HD!Y)wPiJgCmWKqtv`0{)XlrtO#i?Gk)iFJsCy{ON-E2S?WQILj zSXn0ZyEGK`(&dedrxCR8!)i4k^u{|XrGYr{-aOyDE`j1yh-O2mXPF9}RQY)0`2x}6 zfi|{YrpQ|n^V_2K9QX1hriVnPGt?ZI6Q_WaTQl3U2)}h>UK8M49YXS22iTm`j7O=& zHq)60(4~;r6bG+NHg4jM{bk$g`VnWor||Zb&C3^!v1OV|s9y6}qUeXbjt9OT7cu4T zHKFQ4KQb-Y+=@Q#(d_Z6c8yxqbLsiv*~EamBD}64MO;N%-U=wgssd$gp{mPDxAik>Wl zihB%X^?rf+EM#%Jf=%#q;-J<6vT1?b z+=8cs!_Dc;!ADZcicB-Nh{@Dnx{qUiby2&gCf}0n zsnfH0y36@FH=^fsXoMzu2HFs)a(9&Er570(v{b)Ov|D(f8B7VVPogqqy>{xF|?jqSnmm z-f2FOpmO=}xVvZji>S_2xphS%)Vv+_x-4)WEH#nx4%PSGW!G~Exw4I%NE8w(T=Gu6 z?==G0uhyEz)lI)S;R$~C0cqr>1sbEneO{&C_YkMs{$V$`HwWSR`NfN>uquYz;R&Pf zQq}KX9~~^+U|uiJI=`@8&kCVq|4R@;>;#F!IrI}ZGczxERF+wua*bo}?Q znGSC@krkYa8iOXj&L<(C!LDagTWo~a3O%fx?qO^y z-?3ApGLT{`ljyu1A>$N@iKABNj|wazg@f?_G*T^?l?d7tWv+ z_SUxlg#w>HtYS7cHdru>7;6@d=14jLIz5N?3=R4-3Dhka|Cpsddd@4bddjfj>Oufg-@{;jk1CkD1OjS+$sSjogj=nMvH_W8&7AH5H4(uQk z98|-|j@C*W1PO0FpU$$<>)Sp)S)hE@*9xN&a$pSG^gh^?p}5BzzWQ6loY+`pQ_Eyguxn z>-6vw`G#z_)%%lo?%$3xG%mX#mX@f=EU$+zpRkC}^{{c`!Yu_$z^lcrhnu zp;x0#uG{4SkvW7Md0Cm!8fWSr3*K^*7d`a}epc`ESDD%TtIRZT!;aT0MdwA2MWTdd zW7h6Zxv6y-bgMt6wuA6rxKJ)tmpUna^Qfho1^3G+l@wIK$xVv@Zs$A~$PQcC^&oZ+ zu7x2xkd&U0NpMwLT)lcta7MjQxzC7aI#&@Szr%2}_uOFEnsWuYRQVmSNb0um=9Mjy z3dEYrScpgdc;k69k?P%r9gf1y8-<ST zo>1KxGdCq}J9XDRa&Y#vkkv5cVFx;1M$}DyEroTX-~(O1Q53Wm`P^{BlWK7HwfJ!V z`t~4;(0vvT@(R}@vTt}HWFVXIO35H!7M15R4Y2hYvHG$1)Mf;!2c=vL5E`*` zx-XXRJfpL4sR%Z26{;DF&`$YWH!o?h&s7z)S>bHbg%JzSFvjwe5#Yq4rka#QmI+qc zU5?k;W9>y*&U$yxL^_)-4LLe9yAF0-d^3R{SdsUL_<5{87IX@>FmpqI$D?*+$Lgi| z@O&J|XPFayvv-{zLIk`%*Z8}RuTL<-cLQ4_i5=~&U<8_cax;?Al_jvbh`GN_hOM+51QBtKpha_NcY$%Z9x1Voc zT=WFLH*We|O7V%%h$t2O)1c6|Z>y+g28@#d){8OVltZ8zl7}-S;O9)hJi4=0Oy;0p z6RyKs5aQvR-!BKo=@YF+44CC#{ayhGw-4dJ4g9vz#s44>h3=!f>?00D;3T`1|L>A5 zOYA9OB7_UiPp0rff^0~sb%AsHQ3`fog)OTIf4AjuZ&*fn>iSpOC~D~?rycI*-2IL=XO#|Bc4l5rKZf&g%z5laksj_v2-TPx?m zk649xrpe)UOTyEpxIc{PtjC4R&}`phiP6CPwI362G{z4}0349%+fUMEfc#@0^Tw`3 z_q(4FAqlR4iR<@$R$zRQ^uVfHs`^=y&PW(N@N2WE#AAkF*bya$3zzf{sh!~|?SuU2 zCbstUT0yXE2BE_^b}GdnJ5v%pP90=#6jQDQ=5Dcx^$QR`R`~4-h4u(m*b90gTHF${ zFL{9(y^R+;1$h0$-(Gwj;(aUPP2?{bOKJzj{p=)3tg1V)*D>Iy;Q#;AzghPGPp1Db zO+PsM|1T2%rAWw;Uyl>E?d1qP{yf+J=Z5*n1l}a;Ql1RoAVZ%6fYy*S!^(Iep9D?&xv_x&`GGMvG03zJml8)xAhagbp$te7_V?LYYyI-@T zIOGGE@KLhsfbW>2kJmlKt_1$)zXTlBgb`4w@~2D#{U^_1kKW#0hM}7Zw+L}=3!o7N z@cYdPR$_kO@1fsqR_t@W92VWMGMx&0)w>9D70h8lIpG}bmOMTRAz`nf``rpIlbYT0 zhhSa`;gp7q(!I-UXMe7LznJ1T_sC1=I1iKD|{Up2) z4?>(2NwF+kQ3kQ?Ry>m(G9OZ8Xl_(M8_OM!uin&6tnVmmbt;)Sf?`zzbV zxF8RR4s(CY0dlV0LoZU@5W)G7Y^D#?xQJp376Z&N2FTc0>LVJP@tQXB&L!Y=&A+vJ z&h*XGl~lH@YhLjL#em&#!BjIk9~YFe`)ytQUGy?S?7r9KLa^|#6(ojJzwm#$<3k=0 zfMtE?(}W$yFwAyyW2(Q@vMv-cAm@U;>$9D#5iS3wZa=R3$6N7FwnrFiFkq=>oM4%+ z%G;^XV1F%Mi0)l)Q?yJ#IGG`^7}J>dO0=1W`!IZgpwVP6G!D;9n0gWG}M89W>C zvZgg1Y&@gRH=kJkWb*PA=-V8SzneVfd}i|~A&!+*GJgcom~e%8n%~R9+hxvyxI-7Z zB)WP_c_Al15p0G~Fx^^*qg+^fLYDL>Ojv|-1{&ImaU~ugyW7{anXp!n(=T$xQq!yT zkdk~uOR>IDO0+}gduRU@D+vVlBH^eh@35)G<2bPE5W18gR*sVL6ZjVEI~bi^%?U66 zL|iQ?(8c0@P_qJDkdSkq!kx^&P92uPP;oY|@_QUxO>X9%C)zJwT#g1yN7GpHvo%@W zHKAkXfoz4L&S${PnU^Vt;x>nfan#wE+UuDHoSB*UerAcY8^hWXPoTe$ciDGx45=sf za=bwXF*7IwiP3B(t+I&COymi2E6rBbG8WvTm?o*bj^17KilcunTCUc08l|d`!T4)0 zBt|G>vv7r|>=;dp-t=NCsF0_J&fxRlqr24Qbz(TH8V_O;Y_X$*55HWt!s3 ztu=N?|2Rk^i&vT&b)4KtAG(!86!(sf`2-|dshYg%aop?BTy_C`+v@-H1->nehkPLi zLgG4KPxag^EfUVrwrI=?iLah--zZq{U%bj7=7Z~s!S4Pa4j@9xA)Qa63+oBB)mUE# z%;RyLE_=AmRK@7MOgm1z>+8KLF;-4FhCNEwUcVbN=SK&v+xJ9F7jWyfx$;G zMK(SXUBOWJ<)k>#EoCIZ2=DzzB$daW)2m01OhEBH%=}z?gabPwdu_jwN5cL-9f)J> zA`n)j&C$LmQd(`>sor*9bG>r~hQUWhau z;luN+#Q%yeyN6+FY|#xpik!!^$2Z{lFJC z*|gSs6gWS>M-njJ8V`SY3=vFz8EV;%2q)6^AxQ5E1X^nBe&XTI+U$)A_lgY3M1387 zf9;2sUh=xdbo4xj3X;&m?=1Px#qRLyK z`-|PoFJVKex*qUk1U%_ue4Ae-lLBb-10(fQ7#Gqskc_N0SC9n8r0^$fLa(0Xpj#-g z4_))$O8QYYm(@9s(_01%3UHnKWr432M6!A%fxA5>F%JB!aLOg&pukfL@K0FbuqVLq zU}tnR$wN*T8g!3LAc=uBDJjnxd$3lT63XV6J-)M8FY`%8?gj>#1b!vH>)H zYhKf+SvSA%OsEY3J~9bx1%>cb%nXF9`}=9 z-Sxu4Si3*XDh?$S@cW6kA_x5e%`AQN`;YH}LhN62g+?g(j}|eFk<=b5FfiZzutkhl z$|WG^E2lKMyt5A;@uQqBccWYX1gFTTiXje*0eg^}2!OU2rvs!?#x!89P#HBH61(`E z(nI?s7W@r~ZR(9-ZkCzx*0O?-zTn+BFjwpYgne-SMG7CnR4V?}KupgWe&k19d8scmc20s#W_sxm6x_sW{Ct##azgZ)4V@4`*2LeC zb;OSgN#OMy-ZR_=DlT8Wa@=~E49C%B;1EDL^G|sVap_#kK^x_<@<7Vw#n;NLdHB>- z$Dg|f&5EC(Az|k&bFGX&3J_D~cf>r)5lS5y)ssOL!MlY)!b9DQwHI92yGk!CtmdNi zu)U-1wWmP%?}vrWJB+FNoKX1F8ct6bie5q2OadnrUnA<(Q0-hA8(5{IEL!oS3@raA z-zP-Ei)Jpy-0w0dz^otLvRXxczq}*#OZA4M8NH~ ziOod6PQ?u6JuS>!3GJ#$&8#ivBmq~?UJb;9$M&Au%Wre?(pNnfx3K_?qN77*VFVB&{ZN{xL z!Ml?7(XT5XRTMKLYN8Ogt&U8wnG6fFPrVMOpFG)1mz{rbL~KW+7f;GP3a`~KaK;L!7F2RLRYRo>H<~@>bR|q;=i{rGcCmix?|S)P zg>7GdOnhxmq2J_w{?@A{yn-4f1J#iqof=>y^3iRC`ug%-w`<&HYWc?A9LYcMlrH&b znKhN(ugxAKRS)u1WU^hB+R!~v|C{h*u zI_56r&_YnPg)flLpFFYhxgvNySiAWdNxlH2@fA@=s)L1St8r7@ps*`4_7mL37?{JL z_@{BQFflVsd%jdO8U2H2|W)3OcFzAmwP8#}MB!?BX3~lB|%BSt- zuXjYsZHL0E9mZ{#k6B39*vu@Gz8g$;Whbe2h5k-s?!AN&6BL=rYe7jcT_z~~?Strp zgx8Vw2**#-OCAr-2#xCyn}3RBQ;YD&!8ZF@8d&8au} z@1&%e_TKwMNbF6Os=q?FywQ@`rrcZ8BY?}4Pqyy(d7S*yXm2)!p!G7OI$1>T20mpO znbTY>Bq)Dl5toDRKL{uz!6`fKP0zEkw0|nQ#9HLyy*B!tNIB85=}VY6 zo|Z?#ulf8ZW_!hYknr%jS{|YCfMj{W2dat% z2lpJd+n=nH{)V**?C){0d?4^>=ssc;pWfQ}Y!5aZ-1Vuum!{pcEhRU-IPAs0vl)J9 zEsWBPx>f*CZgPMEFad1y|84o^Xz&O+itaKM<;0l+-cCGoum->GiGSCzp;!_BCNkXO zSD>h_-H)a59i!!%m&j~AW?T;Y0VZ8-EzMOFkVQGVUzg?n(vkr@R`B?@$VS8V1=HBa zP2pF`GV`e9E77B7TdeAp44Yj$SN+OZy=1;xo`1OC@Qtbw?(TNBuO*r!;2B3RIaL?U zWkIb)O=5M<{t%RAPOXs)q8do#CcL(g+4cx)`F-YnEzJGKOKJr<;EboKLV@&vLzL#T z-fwxYosZ2`s4HK>gDncpYQ51At4TT!8-W2wmfD8H z349g5N0om=WB6Hu0Pxl~i~s<`$7eY~)E?>f@Pp`~tE0(SBX}+Pm6su+y08^A-6=3| zNKE9LHe(I&bqon4O1YGB+(RZ$&#iR#9MMlK}=#iAo}=tVRSBEkh5KU=<` zd8B4Kt_(B5*vDz*N_v%{Ehd3oxy>V7s-QT^u0|?AsyM2pJu~jLaR@5^0%U7`Sm9ol zHSve))%P5Vm6n3&q{|Xxr`h`aFy%w$mGqXxcyG>F&i5nSMm>mD2&j+)2BgH|`(#$U zPD2852qmxNre_t{auj2mDh!jux=OH>$!x2V0FQnj$lXH{Fi>HpI$nwzXqp8rQ@o*o zWL4jFF}v^B<+Lz*C*Qnr+CX!Aq^NXrM9jPnX>5C|2LHLj(eof+nKEY(X!9BIfO{nL2~7;v1#dawWoSn?C!E(?U|Zx zv4z-KLhR`Jv5G$IR-#~)zRN=?JEf^FTj5P_QCz(SWGmPNjr7@GsKAu{vc#s@QZ03o z!`LGVG5Wxy1l#uHxHlhWd$y+(H)*S@slM$$7XYq!LAQp5~3YRZjI|-RA@xm6-F!#~7D4sF#}J--_CtUpv-tL@<(W`I(LOgO^KYZ*Rhidp4LcedGGp;WF+r7nyxk+G8@Lxjz=JOKW|KxRx4Vm~DxWRsGMPS9v5h?&`4;YdZ@DPT-WvY3zj(Gv{3kZc zH8j3X&7D7A#X(PU{=Jf;%Jf74rxc^*QoVzO^RbKPh*oGZMJ8-fGr{blT1mQlapzA29)FMuO+tg;+4!SXQV%B&SuBOxt(5o#piC4Vn+IU(PrIZ<%2Lq*fsLTvHj@_Lr9w+kny zeMty068`J}NR0YtHrWTE{EuW2!S{(wLt|@o!_B+EvHctqq^o~#yaL%Zb5`}^qvV?< zNO9@y7<>7bQ!)u8;Tp>>Wh%NS5o9Bhu`^+oTc|$jnv&Yc8=TUWFdMIdSEU^L+fqTo z1ODr6&9{(gNE&owg?F{EI~rjZF#)3{6hZqs(%-K&{OpU|T}Dh;l~kQEjXwP_J&U`6 z>aneKzO(@DBj_a|0otv+?dl;@M~AJhCGi?DUdpn`cK6k8_eWF`1IkhM^2GpKgZ>+~ zYEiZd8-+<)x*NE6CL641@3a(4QO)*db4pcaJuD89!R>9{EX<;wvEfGed^Qh+$k6nr z371Fyf}?BK(VZ3A@%QgjoSG9ir?NJ2RJ(GRs|ojA(i*3W#eW^JxDXxnbdb#D2wtTD5k7F!9ot?Dji7N z6;E8APGaym?=SF2adso7Zy+0`gJ08>#)D^{R(P|r({|@^e~8=Ym}D$ma+Dba*m!RE zjm3@e%CSnF%v9})6AKQ_2??F#oiy4Fwua7%E$ODIQbQhAF1llPBuuzZai~^W7BYQ! z!iG5VGtWjy_k+{b2%`ImPxFa%4IO8kE+OlsYU@HZ57Z@OBsiK|a5^z2yuO)LxC}%^U6@TJP|LG_B6RLfZ9!8t@2(MrYNjk2atL)k5itg~v zba|>(Z9hf1j>57X_nDE=y{T&Em?h}$kln3Yj*5=TAB7C!hPA#}_Dw^@j3uRxMkR+a z$0nb7fD>b}Hjwd25~01X>DBiE**{hOC&!F9(Fej%9K8b1d$v_O+g{~S8PYqMTikF; zM00uSweDD-f9+??WJ{M>nr|WY?%}?)hIe!JaHj(e0)ERe$xuyaK8QI$nPeC3%j;rOjtYp2)d#=j zB605ro3sJc?Y^7-r>1*s{j-Jyf#+Nvso$-xhd~)G&ZdA#4E>lpdwVx+celrKBVvM) zD4#axmgbtGhf7j|o(`16nqa&C9MU37agsD3psYA?lU5ZP3@Etda+>9wO?wMqc<%pc zcmPPUK>BJP&DnpmFG|a)!^J}n8y#Ls$#IFVDQS`kuHnsM1$#L9|0%*k?FA)!Kd%{WZVJ6Y+^BjffEjr%h%a)c_NX( z57BiJr63ds9{j#MK0%a#COfEk?*)48pwWE?CNZWw=9amOPV?cT;n~-8FiXbA3<_!! z>EdHVbN%GgO@ij%9`Q7Hah1LQxKz4qaL|4s^S-m$Os6}w&2E%FUf@gCII~ZF$HblM z2EssTgY-XcZU5#DzO^?&^xXWR?KG_(;{!Dp`jE*?8tc)8mCpX`@t}{nQbt=_ecRSP zUhifZ9zfb^&rzPUpP!&IC>Wqg5;17DK*EtAU`ToSD+z0CU(qTy z=%j2Jjb0yZUC_-72f`Lf?j9pmIX^5$(`b&~i|cDXZyZPi)ONC;{6}4wu=}G@U{WvM zkfV~%x_2cEHT|(}{G=XtHD==$mc!I7N=gE&E!BH{=^;bdx#2I=7`UFPh50qNG*yR| z=G{9tUL>aH7&d!JAW|lb-mxlOTkWaUyn(bZblA=0R4jBn{qdFa(#7g(x}%R)HGQu- zsLx1yx0=B(4t=vHy+U-rcFVYb%l7YSo85RMOJeW2ya?p3OG#^&-kv~)*!`D_VP<;m zh0Vt8D}L5ZM4Dl5FF^WV&C-lp^@fqzF$1;HiPA8E_tJ+`oEEF&pa!>y0xG6Ps-Kqv z5rc|v*Y?I`I`EwdjVE6-DRJw4p1c3vmV8IQh})XBkGuE8q8fX~!wB^lLRQY@=kX3b z22wn>+u|Hf^RHg+9^Gz4i)`JMk-{>0an%YHox-FO)dT_&%~l2zB@KOABJ)$@^xE5~ zmJ3oGis?2=WnshaaZ*;x$|_{UnXBy*+GGX6QeQ30PP*Uz1)!yazqKvrXx*2b+XtUF9bRf-D9uXmaN@C^+I3ldWQd90*2#LX?4xJJ;rPqw<3}^ zlbPHNnzNAb*-t~OvS#IS4el1ORfLm@;qwd^6U#Y`-cp#&!rf+P^G01Q$%jl zG^0ZiJqkciDM0aRtRVm(4gg7Vdp91rbxIUP4Z6Z^3I;Is4=ZW!%DKG;J-VB zZGMbL^>7?`2^HgF0YEt>yq~-w^>|8Oa3_4Gi7|=>EbZ?j|CC8+^!)10I&wdZ$oYT8 zQP;yuuhtd`*aLhR+O{qUoAk4EhHPf)lG&Oc0C1vbfEB$`?f;b!r}$wxJ(+wl&TvXk zh0x<3h4d)HR*=VF?YSL30)Klr@Q?9oTg$8w-hW336pUSNr=7rMq4K<1D>@V$C<@r=icuYSU9D7@ycLgg$ zN0jCU^db^n8RfRME(wp(XiJQV*8svPPxDrs{W=1Vfd9^#>Gf@b&MtA(Et`i>lC#?c zsx-luwV=8$KnHSVdckK`^6e2^BX5Q&j?;jRsAIrk`2Bf5{IjO{rxi+h<9?aawYF)L z+}71voOVJXqSXB{4Y)=TXg#BqpUb3X%mSySwwZh%82JE{|DYArys|HIFXZ1G>?8z6 zr|e_o|Ex+=d|q4}UU5#m1gE43sParR6&_=Z22Tk>?wq(g2lC#zhFfb2w^qo1SPOti zA9y=BNDHn}kRSky`Jb7#x|q#W5L{|Esi@WjzXs&FlUrs0<*u+z{!vZ+lk@-tc8f6f zwU_|$P%V?mgzR235xC4G1{;Hf!zy8~P{!Du-Bm{CeDoBO03Z~A)%z`>0QBOavW)8c z!ebaD!NOo?zjK|Jgvo`M*hDj*Y2)(urCzWyjCEiWtcC4XrJEg(uIh{;X1=zH_lV&4 zM0><^=+En6`E&p12}U#4+y-isyDYw%zeYNr46QeVbC#i~ zfXx#9WwV?TwBKLpKsP)&9-s(M(7+=Xg9JZh(%bIZ6I^$nZUY(Er!hsG_fSonoUSqW zlHcKWPD1{G_g^S^F6=MOfj-R(WPx9YY3yHx@bx6Ec9jZ?-?X6!okYT)Frf0+tqBVw zviSCBy?yyuE7Q44;z@VGA3NyA{T5ra-dd-e;nH+m6_|yksoo0`)^Vqjp zyb%h=n!mp?!m^O{JtY8OuBBjK)`t+6^cxZcAl5acOtF{q$zca(Z>}RMWvnA_G-aa! zT~NLZ6%X6D+gb-$$Nu4pzO4P%o-f=$5~QfN739*zFU+UlBS(_L>5xmhMeie#he(upBu(}w$C0V&p zh=lL?5X22V{cPs?{!Aiy&yG~9{TmZI{@HhSxl=oNj)n#LN1cGl)4D` zX4E1&chQJ}x+cN+>Wr6xgh>t?qD9{!YSHY;(Vs_*2CBNG|B%%U53P-%C$M?)UimgP zwy}f{jS9Sx2i-Q=g9-V*MW4y61oMQnWjq9H?IJ2pJC=$^Vsz3t1Uw6`vEi=Y{NXG1w$|oB1_graB85wX{(+dqr%&w>V!taF z?^A-($!vTYNV~RxJC8?j7#Cc32XA755em4g4|umJM)WlXV-kcM85wZNzezX>^tCGO z{)1f&k`!KT*fm~y0_fs7F*hPlTF{ zVJpo1N1v2UI9;JbRvzu>v$(Q{50eA)5H?;2Gs%865FsPz!ygGy|7cyf5`G`ll z2c}Q%paT^y0D~KtJxD(sUc`M*Gu92Uj3rgVTp~NyMzj{V4zFSoGNu{1aLqQs>4Pqh zlm2M!#7=#?5F;?)rtvH6^6z!=d6EktZ5N3umxS?k_#CMD2-TAA3mgw8jrCF^0Qaur zC1M%ujM@$~s^YPhF*!VhwZ|SAyF?m)u2xhCOI+tT>+u3ee&#^YAZYIlzXw_Tqk<>pYeGD)1M1Q%DmBzRJAZsVw(NccQrl zyR(B{u0HQ_%8vZh#^ch8RQ?bOW2?=H(}2w_v?*LEJwyMb{~c&QpZGF@v`x z>TMb(=plnFl=`VADl8|#IMbwR{=9|F!{56o#@o^Fye#RT#POpP5XTlPevtd?Ua8UT zp!&LkPNk>ls(}z=|LS5ltL#qb3lqpf$n^{r3xhCP6tj1f;UtjXk83$&UH=A!waUaA zV(+gt2}k1Hf#B)6jfaoqNGM#c!z?Y%R#h}K)DTaUDIaM-VPBnS%9r%u{uXu?%m$WG zhP!&ppxgspgZCz)?Jq^T+)&PbI>#vUbmQMU9=4y7RyuoNoV<}-scPZxe9Wg zq}GJHEx$nDrf(JrHL_~R_H|L^At1=GS64ow|K(2Ji8Q!tk5;q>YK5MjF$Ccc!8Ge8 zF7Izd5hRt0mFmXSkmzEHu|-Nj>ay`;sp0c=CA)t42GYDv)jgjh$MxFkg@Z zz}Niqd4q10+U*K^PJGBJIWJ;dKU`L{Z{=B0XN)}k&x1`C{V7a9m4j~*|C>*O_Ku~c zs#o#mS8kV_7md{wmCWsK7|wXPQl7RFv*5oZXl1~NgxA_PNDk#-f`$QhY#m#@g??KT zYb;PzAw!G0<8_Cn^BBCxSG)DyTZ%DQ?2P8hdo-=9{QFOU_RFjTrr7jJv|!vYWaB z8P60thx5zhU-2}TTaDS3Ho7lI#LVR1rm&J56*$KvbJ6_1=2|NvtOuumCjN_~i>fAc z;=?W$9V}6{f6|QbW>dF(MVu|G=5{7j-<~7y%F&m|$}@SFcSF23elF@WP!$v=rraaz)8cI)lG#4(;4B+3zYP^xLm1zY&GpU0vd!&tn^N5o z_(7zrC4)qkmkSMCQb#gppC|kWeFpeT_4mJNIrHPxizLLsJ0&`=2vrWo%Iz;4`K*2E zDfXaMs-$ZojLH{uqyO(?hR2HWHz= zs`ELNc`x6~D_Vv!uZ4Hydvxna_dvUtSUz*+HPnqC6lIpbc|J_x$Ihe^Y@MBNra!$0 zKL1XIHD6SA*X zvo-UBbPo6lmLeo{6?RlOl@Rh?xpdL$ zti9ra>JMSd4xh$lOh;vicsGs+Cd{(dNHU(=4d3p57-JOsF67J~E7j``nGBBr$=G)) zR?=!hFm$I^^+@M%_;Lw2{)zlUAo;Zm`v9Vs2#Wc7iOEA_^Jp}q#uMVaEnrmW7MkLM zAm>k!KvXDtn&->MIUCz2tYUkAKG159+Jj1xOlAd{;4tJRhytST#f=t~cs(lMe6aeO zGAs>$y^t?et|;+Gp#+XTWODSU_BwrG1U6uGcJKFvFDTkWdr+3YefdDQO1wVam)-NF zs{ZlIa$u=f+++&3{}`qIREUhICWJ_74c1P0)_ATMOUb)J=8`M{xjis|A1wMm`n1PO zRz>r`75kEk^!OJkTsvCY!`+5o#M%?wVFM9ydGQ-a8QAF^9!t;_BAe2%gxu3Lu*~yv zC%b@uIPR+R%MtcopDr-VxSyf1^;4Ln-LX(K(3P9)(#({4O1cF2p^zra6*PAGGQ5M> z<>@awxRn3W^6!#Lup7m0L=$aTv9eUR(!lHdgYN+X;@`$8mX7;6`6GH1LeIn%>5MV=1|Ud;e=kp96$@8{zpO z{KtQzUF7E)SazHD-m4{eMDef$f+JzrhSVS85GBPlh$63WW`?OvjJ@bHHpxCcNU$k_ z9V%dsZqKRul3|XuR*`tks_}%)6Hg$)CoGL5f)$uTV?8!uN^@zd@u<0`>FNyAbF;&R zhwgP$Sqo7*b~`r#m;lvkHI8xZM_!2MstfzMqI2S5vL5BYS& zcw<*ewA)hE_oZD9!{9B+PRgnA*!LdICdRy)XXkLV ziEcQh7Dv^lqPdMwM-F!W*KV8SI#WF8Ne6FX?*zpL`R$r4NoYUzj;28>=@@Y4NfIBr zdiZBfjBk}u)l$Y$$-0Hrpa-i_DZ-A+Y1h`ADf6{g7jYr~@_D+R3ccPBcN9{}w&>xD@-%W{)V-Ld*Pg4rPi8fnBJ(7V zy~oAr1!92kYR!ltoWEp)jO#By*yEIm)2v(l;7<2??qGiH+EV4jJCei;vbV}VU7_i> znLTUM(k~+$BdPs)l>D}|&GpQ7ly^0uaET|59xE^8!#2Xh@f7jM4bbe-oRI6dyQ(zi zTBlNn%_InX@7XIp$uKaxQeZ@bK9wiQxZtGH;@IR14rJ(8(d56;~JDvSEnEBVF8<-rj!Mn;aHjxMo6kl<0d=!GF zV}R@x$-UzWR@dLVS>3O}wfNKk%17T7h$!;ges?BgN}?OMP;gBe0&JKr{c$gVwQIlDp$0-Ges-NhDJ+n;*Yx1~n4Fc0Dyuv+rrhy*p@V|)YTYTY-t za{h(|E+O;Vad{|sVs8|GTAdwE#|+QKPmea+OURK*IX0Ftr4&lTaRCep1(g;M#yXyR z6LRK=*9W{CPmsW6mwppu9sEX?wY}SY5%~%7qwU4>4I;ZCVdp^C#X4p~wkue-PZmogNU7;L4CVBM*1)ov}06V2XyqYrJeJ zR5yncnL*1?y$g0daGkFBuMR|phw;jkP&*i$xUE=$N=vX=#>JJVJ9ci+6N3y)EM6IJ zgA%FF6DqjC?G?Dd8UMmT7=?e~5GOZG*kq>w)h2P51-GUC&AEZ{J)GMF9a*4mgagOp zx=nB8D{V0Bg8T);W@Q*E(gSu~@US{)L2VJ&BZkX<;K@0ZnZdtUs6g#^8E#P|gH~8U zd1S3e_wTQ1o!JKdv1ui`|LDZm`DLK%)hM7_%R2nS;jK41LpQZulz6S{_V3SOz3cM- z|1DwPgZez&mG}=LyT$BE5OdwQ$6uiE~cODHWhnkuyma{erH$14>(lt@oh zA6LPy0r-EF+29YaCWA7=_@-I;(~^Q1?pF*Mp1@i^_j*WAk?#jb@H9$a)kb$t#WM?$ zcCc?eD-f*XKdbfX_*(znamE;qehMPH{UvsFlEmjXB0o{svfmE0K%_@ zO%18Sjjusqa}4kd>zk#{l-7zE{_Yj>lQ8`hjqSXb_rZ9ZGC(}0M3(tU8-%eIQ>{mT zcRcW)FJ!Brp(?mGUj8XFhbDg;8=DX`>+Cc0*Cg)IDV}c6q-_0l3Z11Dt38PE_^(hW z4jlOxpZ-`mbPnh~$GYpKYFN#&DEH80*5l>vJfjT%{Wk@IwGS*3R1T|#si0pz1?VKH z+oHSVGTEPO@@Uy|s4(n#i}%>KvN?Ds`lcmYbi8pSh$GN^>TM9cxXh73uV*a-AW$Wn zE`zZI0o{wIXmaH?Ws%|+SzJem;NJD)C;JYKFB7g*rf~%H?`}Tzbubp(FXw4$_=DBB z4_;pZ^24?vKOj8%8~MrLI+V(;h(JFh;+RyS$5@L`ugG~uPdboJa-6rVO6qL=hNdO0 znwVSY%-H`X(|M*&r;S5=t^w7#AHZ=e%wzARHK1~5pD*uv>0a$J+H0R*;Mn|QdPhT<_MuKR`=UO&V$@{ zw5l*RBwBuw%Ux_2e(2bAkGHb6X1aYX&f6dB5as!tl}=U|`CMD(T-A-tSWy|Xy-IXw z3p1OZkwh!e$j>cK4s-UrHnNEE1f?Kf9G-)N;NM)TaOT5TooWUz%B1)0l=2yVKjg;E z-`rSKyr--*wyVg^%yQsW1c!+*>Db~h1FE1-QguNX?OmP37j75fos()`SLte1!Q@cQ z5Z@ZyEQ!aDSISTD@6P3s-}C*HWnn0C8tL8)D+SOOw@dc(+ab|M?UG;krE)sd_g2Mr z7P%n@Rw5iqhK83`MR)hvtYou#>2(}uwKwb2L)D6&CM)+VPI4F8g}RPKaY>M(4K4N=xqcQkbnE&stQ19B(a6Lcf)GX7-am)HrxwQ8mRrBu$ z#5y!LM*Ku}KeiSz66>uW?e~q{V;PH`;EAnFX|WxdOyX43%*{-S=7Dz4bT2TB}6c1h) z=@caC_I@zPZDDTS!jDhhx#a4DLY_M<%1FLk3D4n|y$ugAbHv#p6kIg+a7B3HnW;}A zzPt1f?^5v4QT1r`lAgy0%lOeU(jP=iucT`jI0n<929Rl8OOlg1Izv1*aauPHSY!*4 zvgVYtRV7Ja<^?-?X^uWKmYLCU`KoZ4y_!wc;22U)Om@_`Jd@{&Jo%>32edEq|7t%$ zsu0h9t|N=q@%Z~Py?)p4f}s3rDLw#<$@dx?syW+yuDIqNR< zkEG;q|0G13(bw-ifduf>opWhy{!Fz~i08H;GE~diqzche(n9~XJa zFv%|*fk=>92jJ3wM^)KhR74&j2D3OD*gxE0GIx$7^6~z`<|W3c;ty*?FHkS@8gOjQ zH1?>NtA~+X71z%RQ21mgLfiQXELcc}#SY8k0^l>}+?L@59}DlWZLT`H+na+s0Xq=e z)nP(CfObIbU}xX%MtUt}>>%bXYrc2?%lDyuB>?3uU&R*4Kg;9_g7XdTMpUNQUu;*J zgaUHvyu{hVj~n+ZRDWIOPddsMTi+^76LXTY5VchHVIEE^lz!j$y9jRQ%J0*S9&-rH zJV`@IJRxS#lx2Fu(tLW?%3bEZTP3dI*3xD9oxMXJ^Z6-l#_QumNL^lQ7<|v?=drKT z!c=TS3Rg{J%qa!PIS^Vo!P6+)r6(C(|(pwozi1FWHspJ zQu2g5!$x(+=S$@*Z&B6`u6~bH1i{LL?eZaKr`=Llr7(4@QP&30SMZ-ym%?9Vl*Rj& z6B605B6s9bsFp@!@sX5P6d*3WS7cMoKkqevGXLN%%-c~yIGDE;|NZWLw7-v>dcl_W zjL0bpsiH5wu=YoRHaAc3IMWLJXfJc^3`NQ-`L|-v71M}Kr*=djHE@N4fKAD^w z9&>|;A}&))CmT(YP-YsWMAdYJ{LL}cu}!+z-b1V8;rMX@gwH87)W3nITq!;+9P9bL z%N2<^6MWw}vGvFBVO`ECiF??Gg8B1ai*I_m=9{9@NO4cB>Mu|1(4g_uIcI&kierha z4iu=owD2hsy$d3$?tYp(pun&0U^n^Z+z7rdlfH~Eq!%}Vz~-fY1=nYIG>peby! z=fk_`gQ-J>tLjX)_ZXd;=sj(I$inIgdA__@y>rWjI%vVNRlOkul)R#H_})+aU%j8D z3^m&}!jQ7C=h^em0XgB6NED*U>I$=})>C7zr{ei8KldW#ykN52o4MUf6)gj{88b36Jc9 zNW?o1w%9t6J2ix7SE&a2Kf}{S6sTYZ|2`hC-)*NclSQmCwpW_pJlOc>EN4Nv1mLLs z`Zh~YrB^AgDana`$tc6%n)s?#6?%z4J_0;gN2c6NTc)bp~J zEWSU%ZwGJ5-5qELES6`iCf$?bDK9@io-5ExPAiNVzDV=t~6T5J> zw3Uzshw56+0jRE|BM`WMJs1ve$u8#Ln6qH=xBpV?x+l@NDn+Ez?nYC$MqqQ#j}cE> zYt(SgguK|ykh+4JSccJd@kpHh`gp6WViFW>cK2iG&HF|!AS+o=O51!#aZ zzTn<@RNDP81RPTe$_KQ6_8gRd6`8|Rt-2i{u3fGgsDW_zd;lg2*z3G!x;~4~3E)*8 z?`f!XnxfsM(?0|X7apjkF*cSleJRf=T_GLXYT+zT=WKI#N}NE=r-z>|=3MW-Vc1de zBDh+uY$n_Sj|MSB|zCCN7=j;nfiIiB#w+0MSB&ZW`^I;Q#~Dmbbk&2;0TB z{xhX@_mEEGwpI7&8m3WI^gfrP$sIcPljimhHofef$_c2#91Wli1esD$dvt~-czWyr z?5_B3H?6I*G3gtdG0Po(ujaEwNFR&yR@6t<3eD{O_A_+ML^8YUhiTDQV^h`VC&sAm z1ZTALe6pS?{6QcGC&GOf@!M+ngDQvGmuEzWRll~(O{VNheM_N?tl^GIRWVa&3`9W^3I0jXL zvREUXI*U}qa+7REtOzZ*Ev#AS_d6 z#gAqrM|Wn+i%1^~yPgzL?3{Tp+7hc7{)tT~sJV3fH+=wQa1Wb`zFFpVT>j)LaB2KS zIQOxPMtguEHFRj+k$ua3Zd5s1xMN{7Bot2^HZ@u`#hqTe)=am+t-lj(~6S z_*5pw!slQq>f-tyZEzF&)^reXuD9P(z}z`--AQFG`ncF<8f%@wikU&Tm-%jP7puBH zP5HlQwu=g27T}6h_P5biJ^CEmn$2^aX$lp|-A9%kE71igTIgw5JxDto_l-j@)$SUt zW4EHMiBj}R#-+Px;^8w6eV9ugA-&7(?UcNPNb~XejFU9M zitJv~Iq_7z76v|z7nNdXIE!c5zvZp2_DmAAv0ReRnz(uKyhyIlWGnL%1ng>J_5#ws za~o;I3@YRWltK9kx&@gz!OQ-0Oku|HqYshB@6k7I+?8QdXme!GZ8v^r=QKGG6MUAX z(Ktur2~n0;o6G8;@T)OfkIJYs(~{PjVkB<^FAjP+_UuVZv-uGWoKD)RiH9d{mXjoM z31!;)vCnIV7CHbb?)@f=Dv_48T(g8maZGRB&B?FJzC#X(XINdMWsEfn*sfGSXFdOG zRsk08u)?b0pVAhyX+8_xBBZt!k)=(qQni$86=vem%EnPC;zsPuJu`Arw-uk+*pl@#5Uk}K454gCAIh%RJ<*V*J8Dvl~zU^1*0x6(2^T{u- zSiO4g=v<3Jj8)iwIP>;J#T0|5c{;bL!5*fT&Lc@Sx;xS}_YHPlX+6u&-7O;coyBS2 zb=*9`uHToaPlNujW6tEM5@%@mN5`0c*2AZfGLq?zM9m!PSWbbWIy-wIQ66IRe7?XK z-}mKHH%sM|f(*)pvA;U^P@Sqy5L2>5=49}qrD6IVa@Tu%lpvvQ+7>XikShZ4$Zu7=2-etD8DPSndF~6P~y+2pUzOk zdUr0^H*uy-TH;ZkUi*EQl+UaE<|~1Oh_}`$kGiWqa2#>De)wJf9ZPGPp|nh%a=D|K zOnuuPHWlI6{C`C#$RGW7FzJf4DfWvyCOg$$(bX}=z62BPby9`1X@W922vz5!-lU7@ zlkZleAn_OSvq3J#(!!HMs?IA9dR48@QOM=0&ztqZS~-W1>`=O^u4DIDKiqQ0)iF)x zmj0^N5>*u4a=ie1d zCzh_Pj>QiTHyCv#Az6!MD-zn61Zy3)jh`o{rT8R7)TC`*Y5g3e&$dmr$WQCFQc{%rS>n~oXu88D{T))KM8Zp4$oXM_T-@v5E z_ow9#4(`Y%RL9cr~zl^O_Hg>2!<3aR*qcoX3|Fs&Q zx8*~QDTh;IqqLugM=yK&V_1mPk8myv$|VYQNosaYuY93h9dEjMf6j_TqDW>$x4jeB z(jc0^y0k$NYaNs|QQ(Sh>r5cjQ5teRI6+?}s6IngAQP{-Y80e3o8H}Qn^VC=D(42c z-AgeMdK%r_bV{@u>fV>^=%(0nFY#{AS-_ZdIKu`M z_pNEysWpIJ;T}9$3G|9^G*TYZ8yzOi|0)@DI9{H+Y;-`7bO)&(s83wTz{1?Ej0gX6 zgUn#u0QqCNt4Sc!p6MZ7LXT88eK5S>iutJ&G^uc$=Hr}a%B&#G)Mo;cA#rE~sQ0>8 zee|*8@U|83%i0-}2tR>6!;4zsmIZ|kSQCAhbbG}MVyKJBNUjP4M>h2>q`R?z^{#?d z7NB}SFzSVOi87ofjB|$Sc4UB((K9D+f+8FhDZsLypZAOujP6 z)OD0DIYLmz(W+WwCQO=3*}s!vj+I4*SqtFWs^M}Z`cCibL4vJDNo7c%WyTzr=FGFg zesVrdE1=QOv6#N|h7hUN)Ijjfd%mM{Uq zG%SvqW{&MoVeL7rl|3BhfTDkE4w4Oy8E-#x`jK&6Nt~z1tpnNOcZ?X7twyE3tLo}! zpte?0$o-g;m2c6i&2aZawo}Hak7!Shd76GLV`B$&Bs#(dZJHwDV89mMLW0~PG-ghH zDsNYIH_de{;u`f0#n?^Ahc+(z*q02Mj?(WjpqNg8%c~t0Agg(n63e~S4BvM1$?0Gc zFttNq)vNdo>t|J#hNZ`RsnlRRLoNS-uap4qNxRM=3*ds15AW+Xr(3xd#y%#?c6+hd z`P@ll`ASi~(?kp4A~TR~lK3Il+M-mz2Uk&ZG@;QDKE-NVjmceL-9GA6CuH{HBLJJ~ z_FmZ?2pYL}QR$UnxjF7onAIeFppj{Q_%^rXC+BOUFQ4Qvj-*tJ;;>&ared8XuVS@v zE8sAyt{o?mkldB6x6yKNOM*1+856a5(TWhbFl*~p3Jo?uhiomuRdqW?KJ03^1a)K{enG&eRF3Nke6_K7d~T?xpv?I zqRkt>SPCE`5f2ar?}lB>jweZ#PYmz4kFH7?FU-z4W-Ha3_Ne*&1H_O6HV<7kXosyH{H7yq>Xz}c z#Z-6LFvwttPrX@DqT@6yEdk+M4P$2O(Z0NMe=iHxgRuKVc>XkPuN=6wEJ2xW{T`#i zy$hsoBBHMidc}H|n7mt7UR87(?Xhc`{$z5|7Cj353)bHjX;~oxMsjJ@BMb72zYIH~ z_vNJaiVP_W%oK_skw~|kwvwffJ$6o4Pfdcn>v-u{?PQm%TaW#*ASR0nCp#~13-ll$ zg`oS(%xE^nlF5}lGR54J4>Ki(s|n*{PNHW@HB_WaL^|Cp%T3G3UZk+h(CA*`BGyxmp9zKz(;WynRk$3brl^FgY>jrN$z{Xkl}i8)OM_zk__?Fe_z?%Z~q0E>u3g z+j~rRhbuaF>YKH+sZ~aBMo9*S-G?r{E@-@u08wX33#LYGJG$kxYLFPpDQ>l9yBOF&Xr_il6ZY z`GPolh)Pe)F~$$6Te+|ts4z_*8y$Z^a?u9Of1e}Agr;Rzt;A~AkqZJ`=OgIbD+$4{ z8c9IV65#dwl;ts^$v0eYhDAqIuQumtI1eZ!==z~WoPL~^^Rk?*nZ48aDSh{ViP=;O znQ9FomaDOHA>amcw1!pXhf~r|{_+y?#}Zh0w{cZS!PLU#Z>S$2j{Wr~%TnI(lopq? z8a_R1lIe9f_7%_~FYb>Lsx77FRmy(Vd0=*PBH&yOe+unXop$4}xYw$#_|Id7r6>|w ztxad7Hh9Y%I))roRr0uyCD|>3r)oZc=!sPu86&|0U!}+TsJN*&IBW@8k}2I zfXeQ(|J%J*>?@PtlGB`;^K)dZA$+)$E<}1Qrp$bXfpGR#z$)Hs@vVcpjbcx;IA31i_I^r6=%esF9fx3&b7 z2>MuqD1?(%uaDfjTjc)D0>;d=7d#GVv}|L!bgR#;!r%EyQO#@?RjbfQ)1ff)M3`Tt!dOF1sL~ zX$Bc6;nl-t!3ed#Dm^?t{NF0cOc@#_K0-u&^SD|HEN6GwAkXS)V#i+?*r)L43j9_| zkJy1=UFy6zeHE{u_>uaLhU*tD`d=YoH}~oz_|A`05>3ROBEESE4S;dG8{3m~MuH(FY8vbsZM-KM* zz6kUoNM$O8IOfxD<5l!5w+WF_5&o_u1n}Uy<2dGywJMliiw(t|ZsH8)Qmf$pW;b(s zJkgml0vPq9C1-z)>u7(O_l<8P`=}OFyv9&ODeO}xi$!|Q8-g(rxI!twC>wt$F|OhN zm#dRNFH;J&M3JA@U@PvR3?YU}*^`yvLmtXH|BZn3oI=hyBT9$D9- z%F=Bt1u6CF-(8<+Ui2~LnVu07p4+3OX_l_aw*IlS%1%;5<5M&GR;n|yhoz$shvMY+ zuO1~2MVF=}$xeY9UpvjTA|!@IOaP89PYz<4gM!Ilr7J~YG z?}C7jig&AE3y*?oljBcW?mQ{S-S=euVa`Re9ON#|+36Fj^6@Jx)Y`>Pw)>;nOBi7O z_$TuU9j2`I7G>Qz=SpwZOYi#rKwN5|!p};?$^P|&?Ohg{9Y}un;fbsYM6aG#BmhLm zf_BB3c%Ofg_f6KcpVQ)!i|;%L8@sa-B(JlMx&e zk6|swTzf5U`Y3i^%N65=m*^WK6mE=n=8PcZ`)H3m0sxY%xIXdyi!&fqN#nR!SZg(R z%6B=d^k&b^DL_w~PdRr(GSK&OmLxJ3B0U z{yeA-Um0$4K)xJq={zqfD&$(GIj6#se@x(URy=AsbkGxd$o;Vbo1AcNRMM9xtaKmk zyGt`|{7_4QyiOztRm+Ier)h#nKx-)1c`_8e)St)T;_DN<^ObgT_IxmEm}u{McxAW8 zkxA7`{}ICTj)AW_T!%)p8oy+9+RY89ea+AMVugN1Qsv~D#M(pJX9#VRw?p;6*yiWX z%tt6?rNc6X-M!2gGC3JBnaa+}<--ckXzPb=$>nvhsajOhYU&1_)~;Cn>I3!NAI{aS z`11rwd_BH7UFJl?(|iWwu_@y$DVi|3G@k7kh;N&L!qa&d6Qz6ewTme|Y-zr3ouANM zFBR#|U8w*+x$MTE=?^+Am@1Im|D+)BZ|-yFBJG&@I4eZXt}V?ua;u1f;OW(FVN!HD zs#qmA%Ym0C_bZq;)Z(V=VPhw#3?1J7F~o_jeD6Oz$iRsOp_u;N&fOiup*(DCadXNB z%GBkKtbYEN8PG2khZMOZMC}P*H?rzATCBkFyg0Ac9rhD-XXpHs2N^_4|q( zvXw4jW0X2e(?fI|{^IL0)qnrRRoRs#`;w&T+kSuCmOU$1-|^d zqp}tIBmb08j=9e5#xuiRuf*J=2T!w#&p^%mVUn5qrp=oFT@hs3`EO^oY)ZA1H`Dv)9Ingh#L-LhucR5vie4K@bGkp0(dw$2XB8sdu_J%9 zHQOPFbqYI8zdEi6au$2R@AMtC_)`lvFRndmscnmtSDwy{5fA^4lRpQfl)1v<|jXOQmctqwnlsEWoBAtq{~&T*byE zk0>rL4Y_)ED)BUaX}Tx4l*>x@ST}B{ie`ew)9M2E>8a3>Y^%h8WBfca{eg?#Qck%R zrmC2I(7jQr^Q0P-co4(BQ}Nxu8E>>n$Qk5toEE*ZQg>r+W^ye3O0Etg9Nn}$+$XT_}~vYJkcOQycv#T++NK|7l|pPEvLStM{{q=tT;3jTQ{ z$rkExg9%jC4Pp-39Bd`tj4(?mex~Z*)4$SrmGGXmQ(d*am)xLv86cn*9Slaj2lO0w zb{3R7raU7JF#B=!rPcU-#IEMum)I6k!caLmt>Giq;KxNNa6eA=`;V^=g~)LJ3{hOJ zH85B-p$YFjHo3T#Y_oNJeK1Xn40P$BJuV3P zH^51ufFE+T{vL9zPmQ@S3Dc$m3a=v4dQ(usS;1fEMUar1fB^lgy9Mt6gA=?jIaFaPE@=lKFB%!=056sF3MEg%$ z2s14Q?O-8`+92UIK?cn{fSwR=zqdM}PM#K9A<}PVo}lU(BvS@868!zzA6FB;{r_m$ zVAvoOzlR)v21Zkq{ z(e&myt6hV1c-_=$bHTR8jrQ^7>V#fF=;V- zdgfW#HkmWk=8oQK;b4niq^J5weum!Zr{O%D_{E3;h7CJSh`k%wG$1HN+u$RE7&uz{ht*w?Zn3s2O3q+e{_7S$@iONj~gEq(+S1`gGaK0x66G~h$P3p%R^ zKP~k`u;34aV#!dHCy7a}P~pqDCuI>7j{ZI3dv_|ENQ;f-Fwp2rvO-C)-l`Ko9Mjk7 z&f!H>=n z^E^Y*ZT#sK=gQ{+!LPb^0)|RrwARL0Lg?H;OeqXss{bCA_in^26AXmLxnvhOMwSXl zsddY}7m_aRxWA&4dErpmnFEYt37`Gs;CDcq2;PB-_nMmre<{d@5LQ-uT0cV124sNy z!uqt|kj8Ekj5F?UPd=BU!^}RaPI9~0$~5(xb@$D=|7d^hqNEoU22kjF#kR9CB#F*T z7E6g*txQFhJsIbjn}i|-LOrDc3oAp9ge-or^xHk~4a)magEH4t)Sha$rmmV;^~MYv zs?ROC0UXUb=9j$XjIxFLA z+fy|y-?N$)#;<4`xf=L3fm>y9x*f9KfgBn@KkDG}L(3Q?I?rJTx=fmdHJT6hHL<%eK3w8uOJxB3NAkG{6LqZ{cdf z_qQdzVYLEbdxG?SJ>iNRKPYa(sVRW>0P@%jp02+p>36S@ z*16I!4h4R#@9_epEnc&;9?#Y_OSvx|vBa0SZYs#a0v~}A264bQKg9$KQ9})(n?dcA zCV@i7+gTH4s*Bi)B`YEYB=03(hCzb-aF%gF)rm83LR{eyFny14jE=D32DN|i%~0en zU~0@FAhI*$4rN%O;#Qi&zo2ai_!fz;BDr3X)OJ&&cPB5*4id05Xl1?fSCFNEFZ9bV z7YY|jz)~IY>KU+B1gT%>0Pj>2W_6eZzapfH6axLSbhj|NQwD9Z*s0Ijl<+AA{(6e* z94(NMw!4dCqKC~9sGS2b@i;=E zr*^aBWm5iW+xde8dTooeUC8%HM$_+@BC?F-{+p|hG2!;ooKXuu97CViaIh{NE5#k@ z^uIb%(rUs*Ub}}SA@d$`6X^<^vmVQ{Z@kU1*9m2E&ffl53Kc^X;?_pl_Y2qQXY>2Y_GuK_*E@HWYAm24bUBa*!$+y zw9)N`oj&ndWMO;IuWHRBeM(YhQu~<(3Ef=uX$)(ZdXHfaniz#+YLb6K!{C2cf*X`l zTVCjGgNyxDXz!o8-=qt7&-ykc5z{VS%BdXmGA+1^=JwT#Unx6E2p`Sy|Key$TvU{% zjlT`s9W*^!CS#EtXEI~%8o*{tjuG#vQgOVge`fS!+vkLx1xH{`Q@l z7I5EFYM8$%)RQW4e6TpOThOrRV|)_=^BpU&T)Mh+mOz39oLsgLtT{<|Vb{2s{#PsU zPY1N;`|>?STF2sYKhGB}f@CQQ&9tv6=9U0Fkq+ZeRJ9aHtvB{+X)6;3_pWXo;t4vt zd9Gj<&AHL4q!ua6$3GA7H66_?Uc61Dnl_Jkj5Y@N9UZT~7GwYPdcrD-q!oF};-#if zVV_MclV)Fqx6A?|8tP$eH9TDt8mIDycfuZ=<2tOn@@2$pC$|nJ4KTd&m;C3iK(#Nm zD1o&k3h`h4D`1qvJS^j&xK^?FZ$;Q`PuI`zk!}o&?{#$9KR+-`;}gf-ppLn&(%kl( zU2W?&7+9je*5Obbf{Rj8AT8?+HE+`5Sy))h;zfPDgY3Z`mI|*)k5iA#4-&8=-?YL1 zPNnrtgw#PSjR*aUKeOJ*43kTN;W97j`kh@u(bXA`;yJ21_8q!EDy2B6o z;vXI|G+GC=_u&Q*lqV5@06A9%Ix1=qg97dEwy#*pn|d0JYlxp3Byc<9~iGK@%8@&#v1bqOgC3D}dw9yrU2u7Y^oyy!Rix zi*i+L1UK-O3nSvVVUqiJ2OUF5n!K|P- zilx{q(>cJPxrlzZUp=dhDjqx-Xfx7XH}P)!$(gGni+XNUgmF~99v?WRv<=G3)Y?Bl zDDy^$#qT;xiyo}P92sSV(+|@&DfT4}>y^Pjebf8S*_i}=8rvO0L{6ZcOcW%4VGmc= zJ2hUwEG^+XaZL1l)_92r_Ld)q0%iK;>Vl*LIfR#Yo%Y#v&bK8H=BR+sd}00@iWW=( z1AGEUV63z6xg*8v9kY!E-ehSgL+(3xM%PVfBKeZ<%JTW6B02~9}WoTxaW zv(Dx9v|oxnq)i`9+Z_SS@xHKgmR<`o`M*K<@lb>t0Mz9Sfw#iiSq3y}KM9+X>`S6b zS#OV$fjBIwq^VOhy}^VAjRlz3gZZoD1gn@9o27QbI()3+pBBtP^w#9G+PSH{J;U6S zW+%*yK}*5zwHU^Q;#y~8(kSb*XXH!|5ES!Xq2IkuN|!VvtWQB?bnes}8fi;~kWXtk zBCwY#n-$_ep{ssXV7Uj%t6YeD4FbGtu`T+QkK>;_Bk!ysRCwum(~u-I?x#X|K%(7Q z9IXM9K8?%1wR-5ER?jpRQWfO3WCoDTeI|if03K3Awf5d*x4k<&qHa{I`BlHkNtAR? zRLV}m!vSKR6}Zgfe$n5c)A(5;$do`?B9Z8dBR_WVZs!g%9f&Wc3iB!+G9 zBPlYr?<`#7&5UeK^oS<6IhpGGDQ&w~!`y zzP;lj8&Bm)GPqQR`?g$V`WI4gr$d8Y&)F3nh|5W3ap{d&pZ`G0z15$oS6*LzR{|Qd zV^yun+J3#%`TIZB`JQIIpK%lsAgv9#&|^ZIL3pKu5l-w5ZTl5R4}0Zn6ne0=*rie3 ziIA{4+|s1hKIo;zVw8`*8?V3z>N?*~sP#fAXQMQbm8#YEv_4OkzLQ8w*Mx!Q;}SRN z#H_)6k=}zpN#*WPtM$KwAJ%_#4K&et(|sQ}9(d;E8`FH$*J~q*Bj}_cf`rNYy7IWa z7-|1u0Q&6_to>$k)|F$r@GMP+*}d0{Urr1$sqj8>d*dmn@P%ekL)uHCvip_b{+NPg zvso3Dh--?>fh`^#wNFDr%Vo_@zpW|&=`{QvpRFFEjA`{ptDT-fqxn2^v)SBZwXX!e zz6(qbCLIeYYJ}_Mk@RnCHeRQg9+G}o>--F`Qy{rfzci1q=8b%@c&xQA$VI*$oIK_3QSa^eRg5u7q?XN) z$@XfkE395j`TxRn75D(r)rVSidA7Riv?{1oo%*uQnuB%+AU3ECpxyi@VWnvqUzB1p z0Y!$ysPGlH_cyUhGc&@2=bpQ>!K>W&%PL<{j<{xl)7M3K z8Jr5mU&>d~?(1UePZu~Jo<_p5$GvFM&{Kw++atZ_QAwL&%vla<^DNfm_6!F5Sa+|T zZuFe;{J6CIV6r~5AqLH%K7JXM$|C;dNJb^=H)$+1Tx1g=| zDU$PK>j{Dyq*B*g*1IFz40H!hpCrsKdw7b{Nai@BlZw8XV%X=aX z#c?`h_d>6zzenVYQQ}8Q`seOHNF5?t7~BxOZvQ&Je( zlfzCoJni;ncgJJYloWG7p&oi)_l13^diezM_0@;9y2Q*2o;kWcJKnpw#L*<#IXBZy2~wK#hQAefR!UQXifbkG8V~Mu2!7rEc1E@gu88B8vz5k_Z9lY? zGH8Tl7o{hMal=)J>)>9fEMA52Q``tn6|9@OKN%-Tej+X81(k%@*RLHh=Zr=)uD3lN z=EC0U<8>N3F8$=9jmw2A7g{g9|C)56W$#YV2+H3e`aV4JUYHRroFcbLs3vS!_mu2> z_#YrJ`F==|lA;ylHyvk3eVLcpBgfDaNI)LltId_cb3m_<%<6bfn1Z$|~MgRY&e#EmY#@%o^Y)0*OMRYr|ZjE9V8Qk}hGU&5G?Y86f;hvgg}^cr5Le2^*yst4=5Et~>nmnSNmAf8dM_}+-bS(U=Vvk;{{;EreJkSs(eC1IvN~_(v{;8=z zd0cd`i<{}I^4S_9PzUBc)I)qdaNhsJVyL!5Gitu>oNtWd0X}z$qFUMTD8%KIi=9S+ z_d?g@^2sl9$yWGy-ew8d)hbYjz@1x6}= zURkC9x-30zHLa5v==7YrZsh&KqzL8uK!!-FW}Z&dwpqe4x@X=Jb988xIr(l< zyTgbmW^r0!@hG8UBhm;$}0tUapK9M-f6y;J24>1=lx7NeYcC{MO=l0H4uUdL(+($6OPd-#KGS@pas{QByg~ z@s;giE|n)$i3}w(BCHp-T2$on2Fa+W{3o3&UtZCXuQ+BGC9gAmq8sF(5u)bqa?P*c zK`?c*v#-tj!VeutiWGZLHKeJ-Gwan|j2utx>7=c`a7p~-A3Ha$EU3Tg%nY1iH=CAf zPDpO(un(|%`N4^ygS-i@*^+;~Ar2NwqD%S6TkTDue{>Wpcj5QRL;ht^KSKp2P*F>; zT`a|Dv^K)%(O@gesRK(9DyiCxfVXesY_j=OZmtyXkk`}bg{?1nN}(thwnS+Jbn)jo zYM)T7y7E>1YdpT~es(4;iKgLAm(j|A?`hF-D&DuJTv}5|@#%;#ceJtWBroBaZv@HD zZG`>OKta4_#PE5jLdojuryrQU-n87Q6fexRIhyre>V}Oc_*rn0ZTQqho0{!Y#kt?gD}(p4b*Dyzzhw`L_phD&`o~jt zZ-2sFmfBt=bEtgphvQk^GpdtVqPl(j)*=Yk-%_W}J+mefcIo;Xf#1f<>wJFPy`s;9 zmoCct>GFYCvQ*)gkZ*mj3R8|n0Si~Mfb7(mZ^8aN_@1e}kDAcFbk)2#<51tAE@X54 z^`VSD9UUI!in55`LJ;o6>z$3VuSwg=M8Rll4?pL%4gSOT)f~4N6Gl~7d)3A*w|4D1 z)ey?h*IZx>EP%+Crqlpu*&7KQQ`+I8f7$}R#@!)a_PB>clUHsnTJ&E;*B@T5^U}5L;4WVe^!A80 z3-8*Gyosw$7)vOjiZ7xJn@dWVHw~IrkO0{w^ViZdpu@WeB*2gf;x%tSRI*I8{bb&8Cu6ALWEnJ zsOe}^DhOWP8oc6igXyqK0IB{js{AzGY&FS5Qi3TSC_p%+TKqbVR5@{Y)gNPMH)4FU=Rpk-Z^9TC*gd0x@R@>^M#!Cg8ax(m-o*)XiKx|w`k7jkQ#pIXWE2$qCOb%?YDX-&}7R0|z|_p^2*KQZH?f z9QGUEif@|jf>d8y$zf?}bfp3PbliSLm&34+~!h@V!L5$GKBL>7Bh|tWD%ylag`IwTsYDU!_xLKDn6j zaEg+z_qOB+&KI*3C(;o1kGSvE?$opxd!_KU&j$2mHyrp!vYMyraO`Zgs0~_2+Ww4ak3-I<%iyX|^gg++RG8 z>UIIrM@>2UU4oEb)L z%)Xb4+p6_^U*TyewVOpQb`Irs99+v`&ZlyNT5)rQM+JR&ZD`JPR$#G*68y=A(7lGK zx4yPS>3-%OBbe0yAr?jK-AM{54N^1txmCEELW%f>ble2b%X+h5r$R}Qx7TSKXpX;I zyL)TkVYxo+IzY4-dBl~@7&f1{JiZ29xC!@mRLf~b#GF%mW<(4eq|^p|$jkfj_uz<@ zJe8tE<8I%!XFE0Z6^D*+GD6hDPvJwVi>gJyhPZD*u|w~Ehuf}}+5veK)3O(7{J3nm zoNOrh>?QE>3#CEV;;iFI-7oaqTl*YwHG2kTUQjE9B}}hPm@8|Cxjp6 zJr5rCbk(aD=-ihU?ukY#*1DI#P8|*)oR?iOOtb;Yc^0CqS}$xz1X&8IKDp=VNJ~w5 zZWrlA&Ncu{S|!=az|i8^h{6j@NM)7yH~+z^efZYDsVMQJ!>NgHhCx(4&Lv^5a!Yr<>c=U z#MK4POySaY4j$k?oUjI`1}EqwMt(SgKn19ovW{WylCE@KSEG8->trO_KHdXj7r)Ep zp{LBQ0kL{Tw_VwLT${dzx8;hl_t3fc%xu%*APsxjW95NocH33fsKhBLOq{J2*^lgw zMU@q^Fp_GVGCb7Lx`cI(_xsx0(WO#*N@NY23H zyrhiX@H~ItVAJ1&j#oJtPk-W@N$Z(gub@iHYvT5;F}fak{F3rs$b#0u{{+ z!QnLP*`vGK{5`y)A{oHqxq_Qo4`&Khq~ESKRb5@=-4opyxwi0AK>;hS71S^!6w4`z z6qpE>c~49Kvu_1(%8FKM)^>SlqLE(}WP3hl+$B5lV@(n-Ncn~P6K5k-}}Z|w=H z`cRcQ)l?LFem@-Ke)DsweS~Ba#p>j^UQ>j|n9x*{aTuh5$4_M9Ym7#_@1X3;xT~HY zsD;$-w6P0*@`3rZmSV#&R>uc3x&9mY8m%!0x$FU#LYqyli&nTHU_PmO#YRU+->T^K#Or0uRa-*nwL50xQlUmuYtjP!2}z+T ztF&s}{kEirXg-skzD&5PxnmbRk9>ANv2iw~+as|2XzzpZcNAMM$eXUI$sPBx(O#(( z3(M|0Z`ku}fA()Gd%sv~A;pc+lRJx)DQ$Z3sU`&L+ct*@m8v5BiBV8l5f$1Kfic7T z?mdCge&1Fh8i08d@S=%Virzjp0|h1%vbe&oi6)Pwt)D&)2vQs?k8X&F@i^vY%Sjc^ zA6b$r1n^oyRUbXsWAl}0`dVAB8nVV_`01U!Zk>0XZ|hz;GyI03D}Nq1v2E>OK-3*6 z-rALuRwlqy7Jjy77xc8z!Q(P~VLtlZ3Sf_DoI-%;t>3KE@P}VEO7-NbW~yfY_PE4z zZ~~xGxVYiNM_Uq*g9Iz3M5#nkpkR}nc#@9JW*J3x`3d@wyN0R76NfE(Lhfu@Gz-II zO%C78y4ijFbMFO-0`K_eH4z{Ne1!Exdo8>rSNqY92dVEt59A)?R+yy4s>JPb6LxtW zaj*I9%Jc_rza_}qpJjDT6D>+f2}2m4(i+zAI4R5~NY*-`+5TDaPMbK{aBsJl&GGBp zKdkRZAHP0oCWYv*(m4l@HLOh9C?P#2 z%M3hkyu$u*VW?*2MrlOqR!t;|jC!1B_GJ>~prVP2Vx+-C(zPLOG?!s=FLD1i2hH7m z*mmMz8hkJcNri1)5C1mynCA0vy>Rar<$o4|AW$l z5Mu_V>tn@BT!hl^CcD+y4yxj zh!=|6-rICEK~?T_ogh$0ML%3h#>?F)cL^wM~O2L?B93!1@hG43cm;kTO>8k0OC^-=w)&kZZ=rV)Jk_gJ#;*K;v z^fZCNYKD`E9p>u89=Wb0O^BGbnC6=-b^Zbp)HEB;L^a_BG|l#?Yfe9E|KcQJq27;? zd3wZ4&F)I>c?JK@`gDCgH3P46j)W`zz2DCC^zP_rfvQk1491kmYZBhn#}2*Tk(!#B zTf8HCTWqCuk)Qty%}gmoc0I%St()X7#ZK!LSbN@XB88ffUEX~;j))sC9Z%sUV+h^-DGcw4 zYurR?Ogl_=B%&&P7jw+b%AK0=mvmrY2#yvrj&i1klM4#l`+K9Ynl#j_LVZCzG!iM& zMeV(o0+)mxxy6*93hwcViC~m_=G)%!WO>m#WT^BIDQQ?%DPEw^^pdiSd1NZNTm1o^3s$Sih*8+xgfIDc^XQP80U$`v0)E6q5>H1j%z2#F`+PjlZ z>P9NCLs}=EuBi*^HyD$x z0uGA5huyNy(A(&h{OcgOI}rEy`?`Vx^^8|OSBGoJ4Fjq-CUqc5lF%Sv#M>zB5eH3> zuVqETnB)mQSRAt?gWgNM?WQd|P|NU+c3H1ue8(|q4QvEK%GFGbF6h$Tz&gPbrMq5N zibm%lar$;C93>j!y0wG|c4IX(ufa3MQYqQ1TX%Zw@vuFpPIO)sb*V^|K6dSrtMeBp-WKx5my^BP zl#;sw0z=Y4fl*GRb0rc&eh`Q1y0IZln*@yS#_IP~dj5+6tV&P3N{7P$WNTi-WR`+uL+V0!vkK}Bh0!B+jVFNyC#h~zTBa}-27rek%G<4o6 z8FFPv+~)6;@tHM3a=A=T=Zr{e6|G) zKR1W`82kb>l?BWMtgx1UBMECYB7SF4!Y8p>{5l?z%De&S<><{oo0Cz3LCY!q$g<}C zl`+J~H%f85Pm4lxq3uT)Z%p@LQ}pa0^z$;=2Gf_P8@)l&R_l0&(}Vji7b)yi-f11P zv--HbeO!ih-t{v53q~fLYg^aUS!QfhT(93yl=p@W;mS&}Ll{`x9Dd`v=EKX;?xRPB z&hlRTSZSAW&QAH%twq803ZF&4DoSi3+O>}U{%D32$zsmRxKQC2)@IipeSM#|%%ukv?MqaZ0tfTQwCbwt&oRj@8G6sgl=NjvE;TgZu zh5Bm_)^NY<+=E^0t{g$4E!Y@pbKjE|U@nk3lA}w5D~66g@axV!9DfoQ z5fb(cpZ5k(MIuLx_XdCH$_wUG6r?|gh9FFn0SCi^O@8FPi!?{QF%N(sg*=X}>`ax- z`GLd!Q0O3#r6C8Yjv#*o*o(}Qh~F^>b0FN{f0ZQ1|Nl-&m>62in`oEQ|6EffXVCzu z0w0cDSSS4h10?qC6jyVFaLm2V^dI;YuGU~LavVBiAO$;rZHag~Q~V~t#7W0ja)D5} z_udD#ALO`lhT*?4$5{$L`@?{>CIZauZ#aI|9QSV=dKM~CGQ42*VOVq^D6#pEnv7o( zmqmWC=)r$8QZ6{&9RD^vBl0)n^de86WwQAX{x(ZkfR?$)^my&xiOJGO@?)+qu#u~_KV>7!MhR93SfC*O2|uME%LEI0y=Q@hru>wIF4H9sZtnu~ z{5|uh%yUubVHXTQq3;*y=KFI$rJD=E$HKF;V-=`&foYdD{*+@b3O{UXmND{hv)r6yPx5u|IBG6WzKStD!6j(lEhpw zR12t!vDr8Xl>a9OmUgB-eYwP!_?+GV)lwi$0f8_fr#TpOL`nbilH}SP3KR#}2>}R1 zTFp^)><7zxKv{R8y0f9;1(5UtRGb?GGCc%@1UVyv+0>X=x%j6|v9mvdI3zB5YWcax zGs(OP%4t!DFJ{KhR#=L{X2yy-hL*Sq&=A1jzH5poGeL=AlN&&&wg$6lv5zu4jRJc9 zz!s8xYPM(H!c+1H$PyLe0ZwlXJ=Z9AIb`o+B<+T%hwsNAD%_ zqSu}Ugte^Oc)F6K6&Fg6wf-g#qKe$->{8l{QT_)Ah*QA$ z0Y)s?YC|e9-rz_6HSZM}yHsop0Mz@bdCgR4zDI(Jkt98B(8)COOF{}h>@v*v{s3xx zg7#PCYRvtym(QW;ayOa-;$>Z$!|-L^I}iWM9BnRiE%S;0+qgOtL;h!7#~n2uc+yZs z=%tWWVO2q(OlvdI^9^>UOOP117R_=K{^pWkWU2<=(gFrT%A|qU&9-k#P#AKOJc5SX zX%R1Gyk^ zf$c>IiMF@P5t#e+`b_S~^i7EQPULTG0o&>?>u47er$5*16*V9}UQxr?ErDR+(3fqP zh$iLPXaD%7ps9k6h`xw)v1H{jUtEU$ts%OR2TiyYXAmfbSmo`NTzAe2v--TGV`{2GbL#TVZIFCUxDN*o}w{Wby91%T|#2}13 zykC;Tc-lGLG!U4@t!X%jeXmR#b3?0cGh|7c6(qN^R29`#5cEUT%kwaw>-S?H+m&d7 z;ZC#T2Q2d>Sn}z0w&K+_(b>d~soK5*vj{pOc@QS-MmA{y;~ge5oJacgBq4f0 zh~`^}kZ#w0CK9m9o^p;zoIQ49va)Ox$*z|ZCX9sC z!_+`sy1hXm^eBO`Xe-ZlzC=cmSLY`(5@+t7M7GfEuy*fiW4SW3HzDX>fO#rHxS5~g z{E#fN1BTE35aM?T?oVcsnLV9|WT$T06m8#h9}@S` z^iU8X$+S_fm)uLOn7XO5kn{iCkVF%z#U?9-ylmP=N%W6MTurkpik=~X==?(N)}qBk z+Cn4;B_#`LQ4<^Q+<-sF;|XnQnogclRewWDh$91Ir5-jMC99)TrIJ~E2ZH==PIhi) zPzS&zU2%EfE^!7*jYx9ip}9<}q+ngg4{8%SV8Cpx`!~93_(DOT#LhY9hlWJFcdkd4 z`kte9lqWWnd9jWq&a^fU@{8~;*+-`q2Pygm5u~s~Q};YnZlt%v+t07Ba<>%;$5Q`+ zj4BPuq}fN%-B@A`6D7C145mBzit0j!q~ys2!@-if(L)&N9#R%>l{_pehvcYob>6vt zCH)G+QR1x4Vn^E(D$!C&bhzg};)hO$b%SP~yPN>scC~>0V*XjP<%VO`v(3_7aqVRE z?<8XJi|$Nr*Qu%QE=4g@Xv{pHeE%ctwD1JjYxYZnse95KGuN#eX{g=bxJpIc1BV19 zEs#IKteS`)QcW+}iw~1}v8=e$!$1^>jAUh)>Gcrso0+7qUPm&y-)O70X zMwHES$zM13&8reKc9rEP(1T_SJMa%0oA|;uBu)bzD*jt7JWl%hq7rkzoLSI0ju)O1 zc6UGa{DdF;1=c*2Im5+>lJPJiw~KjPWu&8SaUxDRW=k9wpyI$>T9RH%imMB(|ZNtA^reL7whv=CI2_ zgnwjAm12~DN=*G^4`Y-xI08p&vXI1fM1liUGV!B!Nts(PTt+2)iu!!49B;KdUs7iw zUHB=|A&$h2-S)6iG`H%I*KHW2b&u6$OeRCBOBor_1>QdQwFRb~@JSkGDZ<+VGr0qS z^%WJJV(TV#ko&7TBo~1M!=!F`Kf=+WBORBguO)a;|Ev{d;kybt z-{ED2VNosU(MLm==#7E8f)L0he->9Jcj&u$Z4Zvb*53>>T#|&&M97P@s41o;uBuT{ z*l>u}0a7_{0)t7hwtHf$HM3FUE!-tt`?I*AliZV0V-xMP5<=V%pL;=@-Uq}}cGhAR z9e6H{7@vWSySSX&aFV3^HWTQ}Pbx7<8Dk@6C63b#BR3R@h(3+oQ}951k4yC-`q&rL z=OTIw80KJ7v29Zzfk}1PG(0FjB%}=qTZizvu$IGJX?d~BSrMK-8)CFAn*|%E8iiFR zGd|k~v*2h?*7QVPn|IO?W{-Wb4q>2dumwzG(b5#@PCW|mbUdg|qsSwfP8!nf5%bNz z`APFg5VRs|L-N>wg$JXX3YV$2o-XecZq_yPK-j5--%ulN^^X<_qU`mIC}?Aj15V!n zKt7wX*J(UzIL%(~f%5{Eq%f8G??k=Q!04vg^9kw{ZbUG*j5GSDOFo(wkxNTLBBfz1 zSldrX+@=X@)`=ac8KfBWF@93sY9+m z$s?i=z7sL!h)qJf-cWJzm?}V5rUPp_d>Bb9=~Eu58x*$V$e*TKA{fuOC5@7k`9h0; z%29L2!CRt>8sr|-xT?o~IxH_6qGR0M_$ioa&|ZgeMm@*kp7O~UMG_xRPa{83;A9wiJ1@J?|z8O%vJ=0b{(kU<$Q9-&(p;% zmq5vyKw>_kFq_~9D}c^N0olCEY^ZMhV?8LI0F(gKW})A9t01Vp2dEp&t!q+R%?BEA z0*VdLSpbt*!Ht>7kpa2;%_i~ze}Zw-t*b$Wb4!Dc7K6+^>{oy~=CJcWrWbHl046QU`1yPM{UF>qzZum;BjS@ME;QZ~&)vTS?%NtU- zhh*5WJBKRdlV~!U73ugdn-F7nCOj8aR2n6vw}6MdPifJkA?1*Pk7cM;Nu=Fk)^7tF2b}U)Jz4eTUYf87@GP z0ED5C%vg*OG(`mX2L&my!~lu|K5glm`y>Gi0|OZ{P=7e(ZpYUM1wLW25ZN0@TrFTM znl&{B_km;4jDAVrfS@Up`+gFS}OcwMM9%--uuA;ypO5B~Bda_1mYeq1@w zSUdEZGEvvhEg~!#rl8&i7##>S(${Q>^93G^l|ZD9vTl+Gg-X)wIDikW;vT!)dc=`}Y4OD{a7Qb& zF)2^pa|MxFhCt-EveXqlvBYv_e~&A(VWg7+WxCTNio@k8Saup(k^Pa~pR|h(ibP=P zD-F8}ajUez%pdKpL8M=gva+{_NE)lV=ghR<{%p4WY&ZAI*aNKd{xD5xkI4P!$sy+` z!4-9aX4C@q85;m@PTnx!z*mthg&67?DamlKgSdO-JlCC8PiMyvoB)m1Ve~NR&1vc1 z<8zaC=Lu6*`uf!ZtOhM~iDj>m$+^kK2(9(}0Q-)pEMWDmghJ#L>fX$-Tbm`gZj=AI zRL964!gO^U;-p3#%OZ9sT8=(c-djpC#V#h-e2#$LO{DuVl#^Yv_#uyUU zxr1X8bp8Q5{@(brnq}rf?pv;$oxqJjNEF zi=ZmO2Fq3`bzgWYr^5s%53T3hY%4sQJn}18>DWnf6qgFRo&2;G*bg&C)8{w=^=_ga zzw<4NNq7o2b+{evM6|ezrm_>q@mx;(?d9DF?LfGboZh4OH2?-(?qxJ(swQZa7Js_58Nw@P7FcV%c@195sf=&X- zyD2s~5(Z?E@-w2+R&5jM9sf2c)-DFHNq9MHx*69QG)W0$rs5t0)Ro~Atw^i%Eh*oI zem5gT`F@nv^M@^IW`0pakU>Kl#ELR7=CduJfmH&e;Ca$f>!DCj?MtmfY;pp^2_v#M zRlgm&F5}63NRO_gGcRm2Tc_e))0YLQYVZ8o!oy~-rKAZy+8XUm0M)>iQ=3J4-|>0G zm!etXAw5{ntVVdFP_G{iJsG2!2;1GLi`A$`;_%QzQRIm~$S z%cf;`-vfR|<0&x27r@D#JXoDydng6?s0ap`*+*kZ=m2&O96kb2&NQ@r1M+={VA6yX zo8_3+0DUb|ib$Mrc`wf;Nu?jhnP+{2F>fTFWx3u@D(PyF9FdnCIY{s4HJglqaCC_} zRpNU8Sqea>pcvD6v$)pfAieTJs# z55b1mDEROOcVlU3mLPI4fB*iT1aQ7|EQ`VdQ`U1&9W)*XI0*=J?1;|c#6#xZ F{|6GTdo}<7 literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/fastgen-hero-dark.png b/blogs/deepspeed-fastgen/assets/images/fastgen-hero-dark.png new file mode 100755 index 0000000000000000000000000000000000000000..6ac1a775805b1f8e8b1be81786cfb6716bd71de3 GIT binary patch literal 146609 zcmeFZd011|-Zo56pUUY`vD2d*)W7!ZIA$oObK(U5JCu$N`OcZ_*S&f(Z1jJ{`dX!zSqfhaqTOcTzjv* z_FBK+{k!jbtsCdu9C!a?|36ezRCdFhzD200?DSJn`Q+=r?*M0Zv8%6w!$)|8<3CmC z94#^U@?Y4~@Y5w&T#_yUeU_zWV!T=7Bq;UwriWb#OZW^V9Hj3 zEUI>OGsSM3`@&|kb{k(gzeKa+8cwN#=KbgM?;f}i%G}=Q+%DsTGvEJa!2KhkP*o-G zr|oj^{TJwKTdnWUbj!a4U#@(#Y`I6}xuv0{|9LRSN`x{Gj$!= zM^(P}naQDoql$_;`~cK1^l#s<3)?E+p15#&d1T7@OO;;?aV`VksG^d7G_$bA?4w`a zEKd%-J@N4Cgv(K7pQ>Eg%g_x4M-`Q`BuCD9ipqtQ^~kWdC%kXlG`9@=*V_g*f}@H` zY4>3)Xv4|QWQ(S^Ck&kqjDtI?8bBNJf|vzwPpIrYlDP=(tkQd^ z<%sgA@=M6_vr*87-V*}&+Y>4~v%1#6omGPG_+=@NDi_QWW-fy^%;rkV-=6sXlWug9 z^649%9f!T`g_9H;Tr+3`=&=8E%YVA%KcnUUoJn-!ryem)u_fK!a(5rQOfgOlxIS3- zw|$4A1n&dHQpR$DWBF7aSya8N;tBri zUvMjgl3ge}k;~-KH!a@k9#a=l_zuAUA)BkRbkl#}ecao5K9Y!UrDdI}3Ql)7EhJJ^ z^p2H$a%1_)=we8a%dwq!dPYOrWy|oAZ z-T!g&n}ls?V(KDdxxo9)HLrZX>+P19%Nny6gj!_$ovftn31h_~a=xOiAurL+wHBIo zOYu?Wq#7Yrk5ETRSc$lpCrmWhT+5I&XN^CtE=kpRVvLSVebo2D(86Sm3nt5a^Oq08fH*yi^nvgr#lpMy)9}R7duZmsD$N!x> z>MjUbkcz6(_MAo^jHR}4hcV=rAC@1+{P6@;QkHynDBZjZVusdy^U#-PgV_O*;GDM! zZrbCYVeG`kJCXD!g{-11_TOg63+BV39(A_p%ypma#s=uBX+y85P!gaRYq`EbhE{0p zyn`GBH+`2)2RrVBdg)@$3+?Qlh%H5?pH6&|(0o8>F;vdEZ_$?X3iIHWY0M-$x`I;$ zO{o{09r(>>cYw=mi`2Jx*|$T=9t+O5p7oWNrB{l>1v+z3iGuvfEZVs9M$p(*(e2Rp zInB`94TWMHbQ!zM-VJ6*G(T$Xf~#Pq<0 zI1qEfOx{K4B%F9T1$|QK{EsW&Gvyl)=gF8leW9tLgMT5o4Vr_!$(2{qYaPr8Czj>k z#97ZI#syZ8bm^4v$w2~YJsCDeWlszW-9-K+<+ZmI@HYc2>s;5m_apQkdv_>!ss!80 zhEUdX`LzaBTgzXVa)o-%T1xN3y%jzcr8?&-tWwd%4S0Vv;d-@!%yPvy+Aa6Rh&+ze z`C$8(_@}GQc{=H7`muw*%~8fmx4Zfgrowz5x3{q19Xyc#Q?R*)SSnDC#nmg1AjLH$ zEl=g>>&=!9ys(=xi#2G0V*hxBSdd!V@mX4+0Y_}cZVkRs1zt-`P%P0{J2BuW0hdM; zsh0(_Gnc?aRJhiD@DP*&>XfE}+DDVlo`igpNs!8nvzY^n)f}h@oy~3BCHtCC@&j63 zQn=_RqiUB`uE**kF4oq~{@O^9#XL6;;1JZlJDXNqtLb7Dg)!9N1;Upurkp$wwg4q3 zLTqo;9}wTv5XOA?a#n2h%xe3#l85E~4}X01Sa&X|B(rjbWQ66*M~z$1ywLfC^aPG` z4S6vn&{`pD?d{Owd&v`YPo&jf5Vp^f7qcDIN16EBjSUpMYngO;{SG}byIt#4J|R4s zRwud=IvwyY*P0I>QIpE-+~4tYLB+0$X``u$$0J0-vt!b#WyTc)wJqaEGvyMW>}agh zEhH=FyQSC#8|N*2G*3s*xdbw5d{&3B8kKIouN=woaN`HU>3uR&Ns?gED%a(zPENSq z6X?hg-%sGnBfk+{%+V=8u6P`c8($aadXz|vDZB4qZgKwcp^Dca?7VdUomAtUcOgcP6#BmVx`5`;@EBHc(S{G&K zpuR}2BCHn2#H$|;=QT^G!g6Azn#eni{n_i~x2_eU2?N8Y*;m- zhTpZj+zwSD-FGO{(34TguAJAm@PQSet`B;!iod#$)*g_ktq44^2T3T=uy77k%ZIb9 zywaB)V3pZV60^5(PAI|E$c3utpG{iG67|xQ{-M#fqcF@53i=(27AqZh%!Aj!eup?>mShgr}f`G90w)iMkaxiUjMzCWrYWMfo#kA9Ru^ zJ89OHaV)eUj5Q*+&V|Q5FO_cMOKmK46oFQafdg37tQmTR5UDNOc=OwvQwRmDl144M zEbyoB9oGYHzFKTM8;Lin7?j(Anas5<2QwKK-ybX|ammzJyOY3*G&`sF+L;23jJQ^yU4^_kzs+>h66T3sFry zvKyv&tX7>=Thh=Y!sgtcer_(Uik^P%`x`T{Poh`UfJ=6cB{NB4q-;=KR)*4chX0RwF6f7h?U4EEybPHgY#_?U!a|rhfK$kPMjO=DoUpKT~$bEBAuczS2gNS zxmHP&7jf*>XJW&KK)r~l>;pS*+$6W+31AhhFL*T})gCLv8+)U_7&Bi@ReH7tV7HQ?$YDSII z*w;z3w)3LZcMZS_*2i?yOzR1v!=YGnVN$Es_*1@l|HgX!1G4rc(_H_0^%G`--eO4Z zsj0H=7*N^$S#0hN5oCA+z!8tk1-&lqKnA*uUD&X*r|_kdsaM2)Bs8zRIzX+;s+<=A z-l*8JZpo+9UBR0pC!Ayw`2Jg5B94gsm&Qs_0kv^p$XK{8(Hu9z*yt|TYWIbc-Zo!DlpNm7=qs@x=u$yjdW?(dBPazCm#-iq$LS9)o7_LGw9dX7b+D~)@Q(g zyk>5mn5qLC>!=r*N`4nOOKM}BrEPbGCee?kAj1*X?q2_ulJYTDq=s%N+c5`CmSQx1fd{kQd}$E2hATSn)jJ- zy<4yA%$|rtSXfi-m1lls#~^wx3|d+8^+IHMU~KQlA!?OYs|~g3hs$VLag8`8@;>@z z_;>E9Or5>=rHKZ$t3_Sp3pnk#XHqAVa#upY7E{L zJlqK5O{|#y=Djv*!JAz@Ls$ZEC0Bn?Ngcea(!KAN2@JCf9`C9^&%erMy6D_k?xePB zThtz3bG#;Ed21$ZY^T^)>cKjSHjhN?E!%!dH=atDx!Gt<=*znO%Fw-xB{4QDZK~{; ztZ&Gg=p$pAAE)r$Q{2luho39ZecR^<5>Hl^S+FO)tT2nX?2C&Fw21gTIJ!4sG}~{C zWJ1F&Qg3xoCisLkD&rWW^w?%aGAiE%q3TjeJGhDuK_R6x__3Cw=pk%g;lDnpCmM0w zV)1z}e}s0nPsC2_BkZvwMr#(6!IufsCnEIhzua5qHjmnXDT@`Ym;(js7~7mfQ6w5Q z{g7_e(imk4Nw0A2DD_9f#tqUuv+c4@d26_^V|UdEwBpyKBr^TjG;F`=F+S3*TQcE~ zCGg?0v5TESCno7KL~qj)GQF~PR^}Y|rF6Yle^XdSyoVmu6IQWbXPp`;hH3}cr=33@ z-R?pbDYGA%PB8(*#J$W9zN356T zM(?dy$)#r^W@9*YZe_iPk@HS$eFcw^elRqCDo=9z3C-CaW)0IoI>+r=Je=lbZeH>F zvW-OkaTi2nQ&Bp-E;$I-hI4%E?VVG_t%o6c57 zU=4HHIc4B%_A&~+Iy$7WC~Kb)JqL+~DxN+xr7DN#bbc^MYorI-CnS2xpoK!5MHcT{ zv&~&V@T#;55Ty{b^FHmSE$5U_cQyb=ccb?#U{?g9I0)K+P`*(*Gt!U|`wz^Ud^gAZ zpnQT8&Vwb*z5*lw-tNkWn%&Wp_48~4?F=$k6O!d0XVPRF)uj%XruJ~$^s5Z@( z-v{k9%of*cqt_bTTdS=^t@bA-y>@FO1&RBFt7ENl?Od`)wlOssHQ+X|Vmv0= zD$9tkG|2|z<{d?L<0r1)!j)r*bOKscJR=CO%BCDfQB z7*E1xHb1UL$ehjs*=6e{Nb|i3!iO|tzP>K#gnZ2Ud%mu9kg~B*)QKi0?6LDSz20Ec zlAEI7y;LuAtIQqHGiXAzPRH5zb;|8j*C5kw`GIBJtb!-RxDb1XqWT>zAS)(CD|q)3 zB8CLtBbvh&m^rjwQzYGa>@?14IFEnH`Mj=Bzi))2ARxlZlN~i&2J*U#B`<7Cb`la` z&;n*S2#uo+N3m_eJ&KBPFiwV<#@T#QB3Wz7gV%PWsKf5 z`>iXjOgw&LY`!gW)-nx~EZ&NZ%uM>3OrxObZ418FF7hp1k`Pwqd4A5Fj)PsNQ(w_$ zgtv+KbXfw%5N17EoS>b&xHb&Q0zjjh^3wld&|hZ4ab(SY8RNSLxYXhoG3n+^0{4)E zc`N{C^-;;NP=4h~?(E~W3`Ue;*9f2c8M@{)ys{|L=CuKktAcF+ z9++V!x_q@NBttNFty5pgD72$86g~z&M+bn%#}eP5Qq2fGLBWNtEm00 zU2lnGnEIZti1#h`%_jSp%SD7vd1112N`Cp|b%cGu?i>V4xLU8a&i5E%`76|uYAA`N zn}HT<`-jMO1CwX;Md6O=R+)$sE}q$q)-~1P<>MXz@SJ8Q-uwWo?NYuoX#6|c{pqhWE1u>Sum<~GxtdL&IvaYmI-C+n zikBVsUaz{b806Nz-bos$>b4fMnrlnaF`|s6b1hvqXLQcw`?Xqj-AAU{aP(~(jRtJg6s>CxrvE&d}?6|vsDnujv;Tk zEWVLEDAtjA6220@Jd#E{Au%6M#WJeaYJvuqs&U#F8VY9Ztp=h+2;019KpPVXvvv~1 zo>RcaLe|nJlVXnwTf0N&z+_*vCHC2;)pZSd)#jm##jRoVSa&(9FOQ?xND-e1h<7#4 zA=ANSD6dN1FiKWHiVseIW#m(?R8jsceNIAcK}8A+W1Z~Y9h>jRUKpCuTDkU3B?=5U z6sS1*6=BlEnnnwcfD)6OWnV=6v4n2vV&_oUP62`_wO%_rplVIcG|8dLkKwxyqRMKi z%Y}ySD8y1VQzCsXSxSI5hBryG5LUA`nQ(!1 z=gutLL#8jDQ#U(QceJT=2+ppiM6Ya~OXcZ=_IQI>!({w7OoHNlEe1%sG^h%28S&)0 z|5wi!ZP+FMysP3E+r{%h6lGXNS07SS+`OWeNiHugcX5+4%GNpX#2%V);u6A7Tl1j)oXEVk(7 zb1QRDbX0}~+AGd(e^0>@9{9Hg+a6^sop?NgKfyl_N~y6B)`HX@E#SM~@#Iwd%b=gs zBF9Jr{p~#Yljqg;L%Z_15%49BWHrx9lywiKTgF!pKUEg2g>9RnB))~>$N?}}N?uNE z*8MfChl}OE^{Fe0!&NhDYa@-RYaf}bZkGG8vc6D-1}}hX%7RupQbO$EQ_NVXIG?~( z-?u98S-M}(^r6rWN>X7g=J~ZD%}vvU-YtZ+gl@y5$5eXVjrko1j~lQRktY#d|P zV$L8)Z~55iY|NfNvRJ*AA)W}&j;zWuq1f`Dy|j}n19$F+f!m4kDO)Vow{V9Q4ATo9 zAz2o}c4YZ6h-1%-r0fLY?Q5$wzZvxzJKB3@$M(I@c3=?>gJLUPc2>|@XfoBN--C$F z<eN4U% zD}W0cSdeL#8#Ls64)z9FtensK1oIO9%Gk|Ds8 znJGA@?=)J6-jrCSod?B^kdU>kEu1*I_%*beR*f3iVK6P$jPeL{x59dIqTEh*G4WZA zMBT!kU1g;j;2Jl6D*!Z`Jqs3e>&!mRhw_Y9L0EI)F6WBXY20D z_Q%d&%{4-PP!`?BeMuf1tRrhucU6>5Oy~+jARgJ|$7|K%e*2J>;iReCV`o4;yB3C1 ztzBoPWW|x0O%Q-e9+KxRG1^O%`n^6mH0mTQ|00SpMT#S91)8(W34R(_HV6-2W0IlO z=(&-e*X%k%Q+Sawb6$arYgU{TPP($EDObNo@SgN8$_@oxcb>rghY)%s)NvFk zo5B7-o(~Dzqkvna$@lxV>35a)0<8W!R>Dl7@ORWCfyu?bfN_NkN3*RabTQiXM^Fx2 zN!c^As_XuTLFs>|(ZifTTx0frhk`+UH5@KVMMHJd7_GimFue6e$3sy`#G=jQwTu%G z>+Dy0O=v-@YlL8g-Okk(GieE%)%<@dBX+uN4wHG4nn4re|4?Fd`4^Wyj2Kx+re!r@Db-ZJHjDL>&{3v2R;tAyBN6 zMxp~xII)&4|6LCB7Bxv;(o_=b)ch)s?`8a(ubH{K!sY}NCbgV-eGtJ6yJ>#0^DoF` zpi=%ybo+%T7$?wn0FcE=Oc>h(H|JJi*SQj)$}Olb0nkA0R9we6I~hqwrz_%5H>g?H zt{w`FM+hc>j$+OJ%3Kaxr?ZTUU$RcdCQrhanhW1^t zVvmpp0F9|^GbT)f{o9>DFupd`y3BrRfoO69i%1XPzHWK1V9b!#=fe>Qg_P z`rN>x0MZXYcbUIqzB@vVXE6m@=Z0Wq1KcPn_1_8c$|hgSHWof7nv{S-Fe)Gm7qwcG z+TU)sv1cz*+-BsKFK3#=#~!c>UDT|rOt$JHHU0*=w(mewYX*_}=3n zcC%Ww+tM;ABmxK|S3vqbc!xvv#D09LR&`Xnk}J{!1!H-p2mL2zESQ0d7cl|R3U?>PjBRnBNu^XupnC-sH(h4z#TLzk`~0 zadS!@_&`|Vr{QoNZ)~#C0Bh0i@*hug76Z^`1PcCoTVOvVgPHtVP^JW0|XuQk`oy2a-L^fajm%6CLgN}&3ho8)c%>Gd< z@Aj79+%Ru6E_OXP7X3kF@b9ja+>0@*%~1H%gA#>MoV%_E5O67_)P1ezLu{;vxrq`M zLqi)N+%MiX*t`qA<@b%?5efGY!n=D^*P5NEE4+khhm3p7<~LqT_o;Gsj!PfyDiix{ z+)p6f={%iL)_VfEXen$RX`#b3hw})tw;QZ59nijF`Qy_9IDs!1;6!M+U+3K+^VB6K zO-zO%>Uzpvp-tI15tjn^@wc;Vo@@uuMDC#vv)j)&Bn3Ix-`@PQY9+{T>9!C@h|Kv0 zky=P5N>hwSjG{8B&D?4Z$MfY97)==6SOo}@TSUY z0a=q|9-G+dUVdJ|J!g3hXB#ZGCPMeUW%yp&p-mf!1<=eUBcw!A^hMEzul{D4gjm=0 z_I)b})DveuEG#~Q9!K@{4KD0CkQBLp-v-+E?JG78Ukz1gEUM1gK}J0k?(F-Qu4=n5 zx5@{46Qq0MTx>JZBNp4>VO2-H>_e4v20cr52#%eH{(jPDi5R?@nX)ZWs$7_ zti$$WbCAzJzytVC=*!c79@m*OCy?-WajFLtW!?VRLGz)KXE~0@Bx3E9NyMb zNjTLdZEx-hEYZZTbf^49nSkbAaW8xkL$1@qL{sWA){>s&H7UrxV7{_$Hphkm9y<$(V^)BT6o7H{~M@|&%BZ) zP&?1@Qj8Lz?=1TiUk&kBE{Z}AQExV8%e`i>eC zepES;m(z6BsCbIFOZ+u_t6xHSH(7?9M~QbfMj3&y{BH-NpqU*alP8wm&Vq!rYB&jB zk!oWg>+!4tS*W~(DSvUKBmuq3sNl?jO6#%`ag0EF>E`yBnhg#bu{Tt(8eiJ&w}3^P zHWx`p4~1UKW>k1OimgMj*-KR{J1>G{yuJAuS4KUb(P3u+kSEP<^u%9OBfUYupH#2e zU|mzVfeNSIwSq{oXYG7A z^p`NPuGoMse5P($6=iF$oKez9(~ z>~rzH3N4dfTJS2&bpJOje%$3>g!nm;^FLXDWrzbOJ;BZ z>CIN0Up~5ZqOv-A07zQ>u61^AXPI)udn{Ua!N#kqPpgU5r}tNiB@Oy`m-)H1nsNVG zZwXSd;MX;P9oSnLwP2BBm4@S(JS8$%?O0b*M{c-M)j{8^qsD-n4_kB|3bod)0qDyr z7Nq(Q^}X*4vuLxkp=)Wk@e}8At>CVTkC$Wvi;oq&uOX##e%F+0m@?#QP@&9CV>{td zFNHI|gZ&8iECi8=NK0?)hS=woynzBmdqyAds&<#<8MVN=<*^_RU(xxTLfJ_9%=Y!0 z+1oVqyp}q#(?RxFQ^1=ANq7_Z|%E@n~(%oq1mbbJ>KXrww=F?L@3)5Nnf>#lB96)F5 zQDzeu=q!pK$+|}l3+g;GaF*W^i9aT}bC6hGIV#`0GQe20PAwBqk0bJvuKPC4yu$vB z!e>}ddKD$P5=5r}O^SB`fHi{_l_G4Sq5=RWfPLVzFm8ZGJ|hmtf1FExOc}#&d@fHA zZ(TqFAqk+>UQHo0oL-m-EJi~Mex;RKR#Qyqp6me<^pdTM+kpG2wRP*239b7e&VW0V zM;&Js@-xb3^#vcJhczy8De&dHol#6Y`!mNu3qHssJ3@)l4deI~3Bgrk)~)aLzRknP z>(DCDy&OVEcKFvq>!b5hD)$94Q|IMG_|q4-{Yr=TJL`8VGAG4-LSA@f{_5|t{cfK@;R{KcqYY5)SWE(%yg1iSqa#PgBgZp<$shzUmQpwLG*)@m*6 zbfDDIgvGfuf!3HYm1mU1vKh<3Rf1Z;K_RZ^sG)T+3oG`E$C{kCr?BuVZE-e{PfFFL4D7=fHYAaKgk4IH%2AZw*< zFJ>flQQBP(?6ZR}t4=x`Aw;WbZx3+@L@uFYwMO>sX3+}NpG96FIuLaZ39HJz53A)4 zeh9#_DpA=W92T>P)Uzsg13g zcqfpE^*jC4JV4XuK+{WmJAhP0I-WpStLe&QrT&H@E*L=xlIS{bG&p4rS= zzD0VGQSKiP*|b6jWJ37H=Qkl_3G@kYPBz8UlocbI{?1ROEW?o1?nv4Fx7=gy5u|c4 z!8?mUo)7-xp?HVRpVMvwm%!X^G(+riBd{#x;&UsiYuD{WP<+jegehHY>Wq^96tktO z)%Gl}+2>FDE#3$pk%2NFm`0SH1DujAt=>+2pMN`c0Uy%YGp7PWy9e732A+ppt` zY^8~%zlYa=x{QdP-{ry64FIZ`C%mMTS=)u+9dfYo5wsGA-33-hG0{Y*9H?f;&k7Q3 zTk7bg4rM@zOs7eG^Dizz0)|&wQp2)sty4Ge z;YUTcqjL(WnZezXL5t=ayq6bKHGwby%kM3gnJG5Pq&9t_3*nde@0xt77QPIPr!kHB zV=n#dtP${09lN`{P(-i~RJWH#(ys3SRDe>!N#S zY4HZ)L|GM4d?L*=!**$Wt?zJXsk8VujqsWcUSmQ;qr_epU`Mno#$V&*W8B=uUHMNu zDRqU5UDM^Y4XJJ`HL)7*K<0`{Nn7E2mGuz2l^_Js~#I45}CO1f5}#mR_E2Q05r@ zEO9Sdo^)8~Xv(uJu)=On-j({h_KUpnmvPulyHV53{?u(rfwYk2FR=}`;w4o}Tk~gR zvtMCcHKrc3weYVmUJ60sYm!=Pt>pVpZ5D0FX8vb*4}Lhj-5gO2A7I#}7_ZpQX%iee zq}Q)hL**IP5LL~F^CiqC7+W7vU%FBPEX@h*?@)kaNUSF>s5fLXR*Nn#+pRNgI z4Qv$KE9u<`96X31ipq*6<4cUJSW68<$^^4I?^g<{JJ=|#lJjX=PvwX1vAMR)s6fA` z4VQp*j(iPNVCpjH*Sgl%SMC5AP%BC>mOkMwFpvL38V`>Y$d(RmTlXNco7m*UjHdG3 zi-47FIb4{4LR;$+nu5NYM(R#O&6Cv8>%X|lfyTs zbkdbFe2rCkEiPl`A&m+zPd5kQA>;5$KpI|igK*_%W_Wuw3b%J#N5!=8w?{s0_&IBP zkeoAUy(a=<(kHR(^V(y;)X~gcHYq3905cx@FWm_|;ZHszaR2VW%6+SxT94*UA4u9A z2YH_hi{1$KucgP1 zY|1Z?>1TkY#yvC+Uu>wVJ*X^1ST>IEnj(wYzHpJE|MlGzjydn5d(gg)j~Lac@_LYE zPozFOF5`R=Q;3EuIB+2O>q#UPA%XU<_uKRH;ztL!JKBlv*+M@mn%-wj-WrHS4fqNZ zU1Sn+xEZ8j)Vx_!38Ei3vu6~uL7eD;vnWnd>zulbGAF693{HVk0oHWtrUkd#we0)4 zqV$8%dU%j#HEWl046_UP8WYrD|JK=k`6kWpNz-of2#b-2)410kJ_RKN9 zo4{%cXi+nPV!r}>;YJJ!U}Wm0;X0X8ZOHd*rM1@bUqLA5lFgbZH__ESQ|A?~a&1ga z7B8i_oN*Y^AhB-2S4Z@}(|Y37D*Q`gfNZ0@hOz9ek`3l42KPzj0Lfow%TYn^>gH0F9Jp~PW1etAz_63c2n^Av%gD?GGXl%}bxCzd-4peI;N zR+xfyay9|hO*6Sp)xN~d)7!R#d7(se+jOyYiOcQpis7v_VZ8a+l1~ZYxnv)5bg()K z+38ME=%sF#9Ge6-h!|iN@*z=+40WrBwk<&7sx)WEUl6o@1w#izwLdUSYUG~>8e)?u zfA)ds8W<9Td@qFBso)wrWko^Cj&*y>)B@|c6=naeEvcbPfBb=BuVvh`4g@D6=kC0g zB#TaC&8G@LQhTVR&AQWV_bIrd3HCbYB=^2l5f&&(md=zB@tH}vwz7Ooy377SV3OO1 zf_L89nBX2nxoAnj#C$I;1e1{-=Dds%h$8Z(5??#zy92sAiJqkNg6z1nA*NKU|hN~7?(%cZ|EqmL{eQSDuQ5dL)`mc3jzJIwN@xXkLwdZdHfFp|GB$606w<&xN8>P>xi0U#3j{w-cU%Pd9;z%FPVY?k?n)f0`%bLmb1yDJUP!v7^x3`YUBbnSJFX(UTVQRGoJa}MA z!o);~g5oP^ikZ%nUD=ntj)hopI!=k30f`R-eoB1bJus16TSp=2H^k+0hz3Cf|%0jN}wy|_w`8{?63Xd z`r-PG_+N{Yq-Wlo&3zHGOCv~aQD1T!IGtj;ZZiQjl{Xz3RwT@CQnDvtj;zjC-68iE zHq=&TuSQ&yN+Z&W)7n7gtyDJr2fY9x72TQPYCC18+KU<3lHSK{H6-8iWaZv-qeW>` z_=)y=vUqhrN{k&Ut@vY8IZIU>+j3mwV5}qJs0TeGX3&@wJLuJE8CUd{`X2mXvcrIo zUikosG;0(3#JItHvPXIc9p2Gk<4L{bD7IJnBnxmUSso51qcN9GHdl6wi8BV7I{(_JmE=S+u}l!S}KQLJ*VOeyF!0M^DihjnS;UHGW9 zUd%4cl(>FcE4;p980J1kA&L%K%r+W3HiHy$b6sfs`QvZ%VU2_KPF?w1sC+ z05fDk6oAbvauy0>2eb7CkpK~%%Z$#Abs6hc776-*WeS8+(KB^R?Q1vDV3m#uhk%C| zVeF6)85iMcUR$zhTDOu9=$W|2k7uoB?xo~NNhZ9Q1|-7gE>o6E9yLQtHUe$tY!UQL zt!1mU1NJ%0SR=MS$*OX|y>-E|i=qr2{JWK@#w8*P#C;l&#T*UL&iOQx3TnLy^!SI7 zARhsuDd4}2XA%yl#T726M~{ziwJ?|8&SrvwRiw-w%7KxzS_4zo1E)ZM){sspcoLRw zANH7Pp{!xsu7qCYrr{11K#(Zjp-P~^f!sdH#Gmhtp0acmto!|%5YBoiBOrii8wgb7 zmDdn*Q{PD|uyl!Otb9MV64Ui~Oe@Vw_=a3p;R==-+zmHL^8&WX+^>=ESIOqV1Q`Pm ziwo#6Q&wr@47(7_thy$Ud=f#qE>?w1v+rql@#aDCcDd`>=7N8z(tH_1R=_l>E9>N{ z_YaA0#Xa+9c{jA(m5vqSGfaM@9h|nBp*YQZtW^sDN14`Js`P?<0z8vhCbyW9s&r_m zYWH{A@S=uEY*`NtEyCYq#vac)l^+98C#@;!5CqX~Yk#9g{u#4pek)w)g_fQodgnL6~-D>&LAEzY*@XGFzNh@HU#Y>b(h~ z6;_=j#cR>8irRNkFhb#Hu74LQO^vZCpOWFO4nL=um!1DIsdn-HjPT^%V-++s2k%9% z+f_F}!?+VxyIE{@`K{4xkU)pXY^d}o1jjd&_Pgs@*_A%|)$ytfFvq~JQ zh&5daGb7^BvhT7X1x}=5 zfvF^SSCbd;f6K?1vf{>u@%RHhk4TBP2{N1Jsf^i24TO@F)mT50L=pH9Sg$XO=>82F zXB`1hQiaH#-P6YR8Bcm0k8#CrGLMTa1LDDAEN760Y3U~_K@4z@zT7dcgtT8HBi8nB%yS$BKDCfV3%q-9h6FJ5FodUC_Eq z1CMu=Hbwi9I#+b?^1xgS2iD17E81Z2L&1s=9Fr7?3Ii`*+E~jm+F<21mR{<$cohp@ z(B76gr#RQ0154>*V##3O#A*B!WlBF?q6%$C&nM)(DzBnJdC&l_-2nmzGNZhJh_89V zlzEsz8&Zk|aTkI*L^TPUcenrEQmZC=CRVzG-tD^}ez`gsmAf&A%e0BTq;H*FxT_L4 zTOMFD!K<+*)}R(N1rhKq4{j=9HM3?I>uz6aYmf~z|1RUrm5dg3liAt4(v*#!?`}0l z5pJ3YsDdP`@4R*ghJ^>pN(@s-09iW7aR9{*A*&yu5(s&6BdU%eH|o z4BR(yz^1?95HF35o_n-0Pu>p-LZY^mVKz!vQ}M^`YG-sK?RM_5 zqU$rnOCf!Aw_2Cn@^^}m^!~znrToRA`CDuJ(~-de2Ec$=V8cFIhqY^HcV9~nt=E~S zj1lgKkuzx#9vR5mgPHA$EVnus?EVBjb(z_(8v^9hT z0>PAp*^PVcrdzR+WWBuX26uzfqXPD>f8)U_I%ss(N4hHt82qaMM}H=7d7{{g50+|$ z#kp8VE%4&udctY|@#wFpy9cyV!73M%*%%LVROg1-rdu~QP59b zg{{L1lQyAlZf`g1c(+V<--*HezI)s~U3O2*);qXyX+s9kGS1?&9&2-G=~nqTUj(cJ zgPZ^ix`aU(vk!M4qOjnzY^p`FQYca z`uk7j{iYdaXo^tvUj+6elc%%-$pwgl-i0KG9}U=bK8!k1ofF5I%N+zEzd_t~z3XZw z2vq7-TM>{Tb5u!Mh@x1iDpZ!64WYA)5w0x2c`?S6R=-#LFEDeJeAA!7<34=V|4MrB zE>+uW+SWXRd>z@hSuS8I7FCGPhsM?o$^U~m02IxgZ92`JCP++&WcVpm>fF@|UV8YN zJQjYv#ZqV!{4)w;OKSal8)#*b6}QYk9z!rWV)PB38N8m+94s*R2dRUc>KtLOoy~-Lr5F8sP^Dz1)Jj7=l!x0>;-y6EPrC20p<|Kmvft z1Ax4?mi(Cssn15kwv!*(ID0)VYVs{r_}qMnbCXkTLgj z;Q_lePyi=pY$m^YNHt?89f z8KU~%n_7niBFD`I=TTBRwESRBB7Ke@KWSDA-?E;TvCV7+6VAPf1&pOC1Z`vy3rz+q z?hYVo-Jy16Q>E#qC*qF?FA%0(-a?B40R+)JJp`8Zyg65PmfhjVZS|koL&kUiu@}Q< zC`By4YUV4Q`~FXai7*CR(7o6O3_z6f&hE}Ad0%KR^SY&+ zvJYNz>(Nvgd;QtD9gyi0;S?O$Fr!Rznw{K}ha~R2jpV;~ikvZ!v+_BYO@IwBX+Sr>YO#YTn-x}dJMu6)Xe85J*jO!#M$9*S}TiGPe#rFX80?5ARY?jS6^ zvc@ZY(HEMmHdaIS$TUdTSdDwtvg-tkm#<%^SNeH57p+v-9}Ry}<&x zKWI6YvlJ5!Q?k-*BF!p!;)t9tztxZFU3~QAzUau26BUn+K}ZJt^Q{ZYy518a@Ssbn zuI25i0|BT3P8bY5ui4|te^SR*FL;zEc&fCY-w`U==64d%g*Xd~!W#uTpGE$<=hJ;N z3|W)RCdbKY;!_4aiXP3k`sF7!R{mRZu^|2Ubn;{#>DNzsKHT_e!kBC8@)|Dxm(XU> zHW!v=eb0aJMnU*QliHQEY_9z2ndGJ1XkDX=?RfTocB)eO&jw!q+1Knp+r9l~7rg&} zHwOf}{HYjyb3pWwa=*SWz?VG#17|)`u8IEN{YS99DcD9ZOTcfx|K$H;KSXf;KcD}* zdtmR2C(PH1CWk}Htxf;@U4wc2>BTYnnE|l>1-V$bn*Ht$jTIR)^YD zzSaNyOy!}sND?FvSHi!%%&p6ZmB1T@qX`9qvs~4{Bj@pY0~#h zhO`=7Zk`<&rb~QBF7N*LwG4T7P!%~heiY5nl&yXKeBbJ0tTD(OYA@FsDzy6QAL(T z&;bQMdC#W>H4M+L(4eZ3comYA6FU1#~){lzWwka%Ewr9_X?^@ z;@LZ9$dCSxeD?}+=knwZAO4Ikvhd5@&8Pnl{+oP_6W97+rN@g9C+-zmS6a(_Pe zrNUal>2Khd2#E=cI`>!NO<9UZ#Tm!N`(BLxKYYD+T+{dZK8(k2wXFiJwxY~h2jT#g z5jJf(P6R|0l#R$95h5TX#8w3tplk_{A_8I{VI|BYS_A|PQ6LErJvvE|vNtU4qq;lR_^Ip5gajHlmLkc_^Rj zI*8@ai?e!Ftm$)SlQ6Q3;f4V(ckV)eJz?tO;^1YdTdLnyg@nC$d)vwA60HqkRuF1w zv3EgqmSHMidoHpd^cs(SH*Ru_mPW#k-K{r?9v{14jkyZOz6~bfkp+q^IAZXBy9dKl7N~{v=ReWs zW9ZS1q}b2q@p_xJ(L0ZS4!5;HqAZa%B7cc+@IY%~pJQJ2K;sKp4!JTZWW8o}N$E2k z&TzBZpWN^QGA)S=dr_6QWO{j%k?ZPDcii>}dU`Tdz42@kbQhdI_Ijy#Ah!SY5@fi$ z<|+Q!3^F9U@>SzA6C+o;xZ+Tcebh1fXG?y5TU-QfTS7#>Qu!j6z^6Wgd%`KGoDv)i4F-kvdp8p!(?s#xHAZi+!&O=l9q% z6&fZL?<7Hg;dc)Ys8S=MlgKo8Yoi$}vyue6#O^@)VgckrxntW_AIn1(6qoIi?ba;& z;Y2d+(_&Bd^@iaIYHAhHp?5W_O)fcCG@OW&dTQa4pSI5<-ia-km*C57@6aoO8(sAOzaBB#qI2%P4n}b$8@VpYDZ9 zZeZhrYX!;Wj$xt;k7z6;^bw3g@bdRluYAHxqVy*)mz_8VjrM4YRpJ)N{{3V9UF;Be z%~R7IYtU5I&(`$N=O&$nSp7M8T8r{Qa6Z2p;Q;}8tM+{XVU9&vjjZ@&Ay-@z!+wQZ zh8eN#a5JDTKX`UPD=B288H=ody*rk|TgV7TJ7M)A`ni3x^X=it=xp3TZ1Y0L*2EW* zbN#=3M%5Dm&;6M2s#?WXh|=K8n0Jdwp()^=#(1{6(epyJB(@rs{lx6jwu?$u{os{J6-ojGfOJo4Miro?T$O`eDd#D1AG2h{XuQv(u03 zgSn}G@Rp0j=Yg9erELo9{)Mk>sUDcmIomll)B;)1Bs#$Q$Ic=vZo3Q?KjSIBXkl$@ zhc3Ez`*B-Xdwe{*l1N9CZ=d3y+sj4FeUgZR+g)$9W2v{RAikR1mnNEKcF2k(G}Zn1 zYgV3X_qlyotEm=0ZupXB#6C?mn>g9$FREZd6!sy1LA6ek%*|R$URtX+{u%jv=Lzur zE!AP1J52h-U9M%4rmqgNB3RP*>v+*)q5dIrv(k3xjn5J6o#6%8lD$iQ9D8Fc@+Rlb`lXP>a(g9>7k=DE z-w?cI2*#0^9a{g{fOJ`RZ5v@BLFxy)O}4dIBDZqWt=HBJG=_or6buKaO3xVY+@yAu;W4_8&Jg!%t+C$p(^J2nm+$3#EF(5I#)@HV=h0%U13L|_7;Jw0YsLVgQ<5VBLf@pstbngRdt0Ft$lf_>F@(I{`EPd4s8?70i)WxApnWk;I{ z9Vd;>NA)C_E!;Jpi&ut2HN*uIj%e502gHox1l?{(9Wo~(<vd_B)R%FW3P@nKKIo16wbmQCC;V*W!`50XxLO_^0FF&ut`hg$f z%nHbF2W0zy@%1dN&CatHPncQZmK%n1!2C`j*Gd=}_dN1!+w|Uoe7?F^6GDFio&A>$he|fKu<8<6`2JorwQFZ#{GB>Em@RO`xs|m^1IH`q5 zgQU>r+5rF$!THul>aHo$lk!k$j%HCQ#-TMrZ-H@fsx`S`AU2SH?9+F>&f^rXb5eSZ z{$zx}b*;fufp+Joz!wWpjH=SMySA@On<+uUpnG%2$hZM-OEXMGm(Gs-Z&h zmZ)Mtn>ri2BNerXAZpmYrHr{uC^i%%hHkQ)YesQnHRcPf47+9bKf}P@e-AWYmMWr$ zLp;>+Gd+IXt%<;x^WDINW#~WqBoCQIz!u_&Y;SsR?Xac`;t?#~FF(%i)5)LwGZ^BM ze}0=!8FWyz;_AhP%{`L9lV_)J|90P##>4=I;@j-U$r z8)2aU^UA(VkQBxuQ^>R(BqvT(1}P!m`r``VpJS1LGJ=E+4KqtzZVmbTU#=X>^8xsC zcwgFQJ__DjR0eqdYnNoEgH#;(MFOCZ!v@3+ODb|2j`}i9QrO$s2O>f{eJDr%YXI** z)0Rj+eA=2lKc~;)&j{C9MZzR+u4;uklSU@;l@X)E+GyFkSLsxeptFf)B^l+rkYuhb z$(YuC>-v+>#VpA3o4RExgvh3vc`pBr=H0d;nXpOmFiDAL0uH_?z_LVJv`%BL>cM%6 zR>mMpExI336WM{*@Vu@5$&cKf>3Y&<TU*mj!z?L!D7` zv}p6?Y{yLsUUvf_uAYb^yhWzNY{a)8Z*MMGjFsHR;FS>^zb5e>YBH|0ElP+m#K|rT zDA}E!wsv(d-W1o7=C0-nrjH@HRM|-j=hA0ULfce4n(9hy^)z0qkWzOC_$&3J$Gp+( zce7d^LG#2+w3@&6o!RUREM8B1o9+)z(>$#YVG8#+GpRBMGi{&4y%eiZsi=uPj}cOi zEhs3&**1J8uS)mm}w;(O>F%?Blgl8I2MoAE62#rofrD* zA?RX>zPiGd{LjqN%PE!JAZ>MK3hStl*6uCJq{*3d2;#TIj-y6&u03>!Pp2=(R}El@ zGO$O(mQ4!3WqxMmFZ^Hx=`!PRf9ta-Nh)$NXMOg@OwulQcg_}{8h1%yHVH1#yKv*9 zBmvFCRyRb}xFg`90`3MM7$ISYwf~iD6b9i^lY*F;Z`9pkWbcHKn-}}V#m}DTr3saE zCs_r;2;wfBE!HN593^Z&_)W_6ESif-07~cL4BbzMILo-h(U!}1*Q~Wxjp%6J zn4bA!;r*R6i{iJG@(TKn!!Vw7;>5bor)aAiz(_BLZsWS&APdD-6oZ>+-o&2Z;K+Mb z?g&`PH&3>}`&B$iJFi{d6a#+!X618HB^0A2O`OZ7XNa|*`k zj6}z+Q^6ylAI1B`y(Wg!BCE&8M7};N!6Larx{+*o7M%0dFjwdwisw=q8WR??#@M9F z7@i(IRZY)`NwpGBW5v^2Ie~0fB=71mN_0&(dmyhD8y_#3t}|y~5~$Q(VUm$ddLQ^6 zKiFf5#yH?}2xbdEUh>_OEqz9L--ml{gMO0KR4$q;M8M*DCpH7KmUU2jsZdfZy#a&|62slj=h$ghdo zg9;bd;1o;$cujxK{0LW9nnptHt%fB*NGJPb09t&}KiW|7%=p+YWVHEW2kN!(d|&T?>4I9rF6U7uAUqj7e3#?(nVg9`hrzLsZo%QnrQT6RqpUpo$Nj225Id?Y~BWY8p&vM?H(i>jkCYWn+VlWesV2H&F*?V_FkR* z=a}AqqFbrL7EhPKN2dEVfe)GKSmf4Oz=u-YiJe9qz5ep84CHJbn!OKf%E z)#NCN=b5F=>S$6ykHU#ZUs<%Xd2F?KpD_8c^?Bi_j#(q{qjeK-rP}~)<}-*8j0b$C zok|DCyg*87{lMoWAVRyqBFagxv*&;>= zUG_JP{qpi0u%YyxIEw^Mi*gy&c*k*a2I@qO9|=uccmv`Wqivf8IpcHj?vS{+@iuFj;|=qO&i`?VKOFE+84S5e7X)b1H(S& znp7zFeC)%y?O-jdM?y0Lk67L!={!q)p$SSi*^z4^c?w=L>P(n_YjqJE))PleZ0lm< z5@Z=`jn9U*l>M+^IE(+N#5cN^mnVSEL+gQ136e!}c{?M-nSGKQcM{M!RCP4#eUzgb zZr%)Z$S<9{kSh?7am1vyJFlUGyU=pidx09a6(OP7ws4RuV|8}SfBJ5NQN0--3q1Uqr~l~2)q zc6P~RezdR~BAH~U=KX-!TCCmc2K96pkG^Qm!bp2FEIaowg%mVZO%okmJaO%W3=mJ#W>qLYn^euT39t;xk%8lxK9>dr92i| zm^-16Jx7>sSkfa8m{$lLmECjIFj5*)I`!R*$lhb|O2!&vm;V;(6+!IiAXNGt&i27o z5I!CP!vwnGKf*kHBKH)NV(l8BLR6pFaw~Cn@!Ek9XGGH}*u`evxPPH(IwkhlrB)Qoz4*s+QAU9q3?n^UEPFchu0zauICl(aeh*G; zJuFYs42z=~AzJa5c>RUpDI`0k*INH;ZGsdEj{g3(S9Rce&9hVXvg{!4e);JZ9?p#b z+hw0qYk4VWqlZ8YEn%5K=~}AfTxn*;U|H1bHy>C9)+T^{|+|$D*(?Up`n{BcChBwIZ9b8 zu!U`{1SZNV=t&P2FXb{Ns15Sm64g`Gz`zjEH2U5C5ISPNFS{rZ5n%?sXzot9Tdh#^ zRVdq^4+w9x6XZ6aVbtu>WkN)xWIEf4?h^n4qz^cDh$SY)J0QdvmD^JkGwRModGLb0 znycMq%^A9C73_;k!lxj}wgS1he=Qd|A$E+cc^5b%Nkuh_Rs#;$Q_jIYE^n8Ihy%si z(ZUNtUl7>u$n#{!*weOy6tobs3(qCpulB8pwDA9}SvcN;$^b`4e3|&kh0dwq-Ax^iLV44egX&DZgNmb3!ADyJJYqC zU6C7yf=|Up0HE8GDNSkz35ET52cXVuH(p#yKSt}qS%vhk%-~`p+8Nx}F|?N0=ph-q z1jB|FSS}h0gU2 zcS⁣)9}RW3iI8J!C>1h%TwY!3GAoIEfFqYlLR7+}R4sSvfuF40Cr<15F05+AY88 zsvNhkuKE^KM@VJ(-!I*f8}?Xnbn~2b50yYK7wSE#=>0AnPs)i0Eb;Ae*;OCGCCPX4 zEi`Ct7gGYdMY~%RjBkH+#uhhsYyvFw>bqTX0|u!5PwH`JgZBaS3_a-b@&M2?z=k5Z z6r&_uAk$P2cS**?=A9{I;RJ8Xzn%K?7191ev1?TG#w&YoXAy7p?cbkcNlVnW)nV} zH|bfOlMQ89B1_m-;c@DYtz)qSyE+B`GoN9yyTvc~^Df~@;9nJ{lCiP87Q`E%K|>t5 z(B~Q8*qFNJ;atO>W&6e2r0I? zQuBrLa=dGBpKf!lz6FeAC@lQLEFjLbgF=Uu`=Gk%AD}RT5Kf2zB2GP8e4lWib!~DS zvXRkjtoDv-<3OtXdxfkJ)E}+`T=I_ud1X+jxldb>v$NHX!Bu%*K#{RBvf}!u{hn!n zO6;n7fhHkN=m-EInr4)k@)LUw@P-59DViV?w!)ON-|RwE0IDMB3D%Rqa76o?&VJ%` z@375FHPT|n<+j^No>>?Lzb`=Je@d^G+`Cpb50C;%g(RkI@D%hgJ;lF|r&!3D@Sb2} zV-FSvWCQko0)jHDj<%pPBR3b0qqIC%C=-KyZk0&ccFqblkNz9^PA*(-lKKKxit01E zp0W|GXYZ^YP~#4_hVM&iEz(9S1J%ppdnIk*yYT!Tf9l*)qU&(Z_S}WneA{dVf9KD7 zUtz<4a{}~7&`R096z+Aoi1$aqJ}7OeA|=p0@r^clbcsFJnR0Y=R@Bd|CkjIrjd%TAn;Y=``WB z#Tr0eOHySMPN>q>D9HgG_$IR967unIjwNxowSqD7EB~j2;JMDb)hv(PK7z$q9&|tO zQf`Dt`e;!B2ApO7!?_C>JlD#T8+Xw>W`-O)a(i%70T#*&vIngH^)YsbiN{~nF|-KF zmn6Nw*LoRk-3R1F$c^K-?tyTEI_)ahIAnGz_7d4}K54f^F_kw_0ah%*%Bp%7Ud((0 zLgcW%`IU~Z#5{#WM_Vk5fX&D}m@iEZhv3oJVP-)>+fwoJquk)P ze5zOtgc*$Vs@&*=6UjKD$`9;}t^Ybv^3_pMp#F&n@ZHD>02D%^GJfkj!(fR&6f4o0 zc^5030Yx1&7b5w`#nW0I>v>>7dB*O#@zn?#z+jIKb4 z<)!11(VG6|;ivYP{3OSr8KKy@bh*Qbt}G?LFCLL&OP4`@y;hGV)=-6t(9n2NMgj`J zdt;WqoH~*6{udW}9SM}dp&Esz{EXCAmJ%*S`pY%kS3V2@piXjhy}E%M%x{bpPw&z4 z#3CG&kf-$vNN3gBUk0?v<})nM)vO_v(%R*{6KR;dB=WvEMFMOiR;Sn%M3Zq(?3m&m zKRCkm`06`f8(DTJ9xVe?gRndD?#p-=B2*$BpRj9;t(Lw7XJ-Zp__zHAW!#l0fBzT| zFoZ4!^+9{B<=hOd5P1tJ7-xT*nq9oSJy$GqB(R5zOcoH^K$|uu6lS*Y@G|GNR>uF} z=NC!vXGh|n*=ULGX^C=wh&}+`XN@=xPapes<{S3>-wclXtt1jhk=mdIq&cSIximtN zE(Y-q3~jXJ`8)#YZ7&WC!H7~vw1icBdHOo3&~%{?NNRQ{SZhfw zG1P;l0*&WtJ1v^^w7ZfdPlL52 z;Upr1UpDjRcALfmbh#o)_Kkp?9-F*%CQCNn8l(lJxc&b}ihCT4ezUJSw%9XNBPEf_ z8ZDFB{Nr1-ow)_-OkuE=;AZb~B-dWVL!IQq9$pq| zyi@7o<`H=;>Yh7X*Wh7MAv7`Y5QA1MsGIdKawc0nLF#+;tFosqsHuq+2W_(#OxHm1 z31B!EK{zE}7%NS>WY`F5qIzY*m{;8Y z198&C!!=`wgP>^TX0=%EG6)KNoQ2~5QKY(J<5v@@EkE$w1>sC)tWVoQ3HlhQ5=u=n zE%o$4wisGs_;^e*OQ1^Ev<-v0`r3zcUVzQH2H`+)(cW41`|;K46<aPb6;wa^*incdWE677n#W2|fKx<%Mlh5rh3CfF=fAhBR`M=CT?n!W7--SRD(z zQuMWq?~`fy#@;nTG2Q9?qT<2vAbAOr-X|tq3uu#8@CA&ed3J|dir zl5_=0@;ChN4VWO$ti*TBYPOdlD@Y}$U>V>v5?BxUoZHJpu8ap-x$6#2I>;;i`gus~YANo2W=o=L&;XHgm$kaLIcs?o;0;J=$oDc4pzE z5ZQn%#)Ei=cN;5Areh&mg^S2J-?tT5|tAQ0@YFHfUW? zRz@{uK~-y+Z|Y;bPx|lUuvNJW6Kjq5_ein=K^>^;lR3|$p`hOZ9YP}!vVTBF)87gd z;TJ&>39&5iNyl@|G^Jf8>UO98JJYGZfYe4~;bGRro4soyXF_VGCS;HLtabxtgFV)9 z=<>LRp5(6I2_)kzA=|d zgbirxSsD-82LWR0ObuUur-Y?tJ6;n+<2eQoo9qCMl46AJXoCXi#z5%Z!{b$A1juek z1JCL*xlJODvwORydLLU}-j8 zsE4^T{P6PJ#6;=CGhy2YhwS45l%RQNb%iW7?2vdkAbKXDr?No$q{TH`$`<}tmpJuPGZe1&jtesBgI(Qq6XX%JAf0ct`) z;Rg)e#0sO42&u3{!UPEwI{`ytnP0E25=>BqSP%iMNcV^71~8_PtUQ!B*%k#K96^M} zO#=5L&Ve4Ew@zG?ru3jOtlz_<~RAaS_N6ICN_!G zmu<+LVT24!@5ISFBK5J=v4tYH#T#S#Lu^tk$3Omk_SxXpCT5`qh_Cqj? zagmZiZhJpB<2e9I%VaCWD{8thDT~~vC+E}*mh631EKTXen@paV12lwatbQu`myeK= zyXXn#Hb}9!eKJo48shIjHc+@=n46?U0{xgSY(euPOp5>$Vu`$QHJNDdD*w|; z%ybhFZnnI3H!e>#AhP9e-dEA;n!z6)?0WWX5J~y-NW|1*Yls68QqX+Ynx9{Nua>y8 zsb$P@>Mv`oWHu)4{oq+nhny40yBLehi5IF8LV>}jgPb%@e!u3JMsXr2XoCxtvq00> zN*+HPoaxGlK#|SzN&$!N5r@*cJ9;6B!N2-xO5#e>#G>M}1z3&nZZd5Lcw$HQkvkK2 z^!>v>`g;%q4fwZb)!#>4U_0O^APM4hXL~vVbQwtzpoEP?A`*t=x(LH)#ZM1w>%~7A zi_=OFI}Dn0U^83 zXv6ITK0;-P97|+N6fn==2?J@~4xn@&EHg2TKFovs@oB{6 ziMzoh){u=Q(|n1-NaMXBzm4{2h)o=fd^zujIyZUAYppThCXA^y-r-a4^Cl|jeSxcR zL7ton3P{AEiSdaFP@~l};%R{iHLF-K<>H_4R$gPwW=VULJ9XS@8PIEAh|zD+GnnWJ zrX53%BB1>t;%O~7Mhn=+&o>PVX(r2^uE?BO=!&34WvdqrI|G6jyY>kzj7T;X?az>cN#s zxpiRDL?*9P#Dzg8e$LiJ%W7_tz-*fIlz{I^&l_gTWf?=ZvVw!eZ<`T#J9C^fTR61V z0`x0R=w5fACT#oZQ^()!LG^*ge-G>z?X_c@$9o#NTDdkcmQ&Zsw@ga#tM8Ui6lL+i z$YFalgnMgggS=hr8A|8BmN&d~Z>Ccdw7*#G79-VsxNQ(TNdQK;HW(E~bOg~a?dz1~ zY)g_8)5?>8dvWMSL${VT%3J3zMlV(~_+NMbKtq9QCHn$XaVn0oD*3zEVHbvvh}{4b=Z6J#}eM=d(8Mwlaib_{6B zbdK;wRc?s&>e%K+Ktp)qR42&LEi)}7m$LZA?R#ys#A(;337z_%a9S~K$xTh*L6Qnf z&Rg$oyR#ohS}{#$i@@u(n-6>k5M{`*b=bKSqHa*1{R*+2hIOM~tAn#VFk zi3L-llG3KeQ_%#934hhY@y8Prja~Q6P=~>pbN*FvNO1Q#6B^J`Z@|QvA)z}{pFXQ4 z{PCc1XV-^0sNO%ABOUkwb$qwaOVgX)!6%BsX;+&+jBPt^pPk=w4`TIlO+=s(jI|t; zKtJG|b;RpD!(*$1d%Ew+*w~f8!qC|Z%n&74ZR=O!rTo2#724^`w2{$&uD4hgjgm;q zxI=;0|_%4oUVuZzH6aMMhMu!xAjOvXHud@6N2bzwc-RlQqq9Lb{QNUe9~_< z4?I>J?YT2X))c+dFXT~vq#B<4FmIBxiLuijPi9RIRlX%tlf{u}8;@zaM#{|7T^+vDKsy(`DszpvJ|AOu#?m*Z7Al6$xz>1xG7 z%mrJsx~%w^lAhPu;VzbzSLrKjf2X1fxzikzYLsVFleuo*P1FyrP5AoSj7r?RCLNHH z^{%Nf<&hm~$5ZaN-xV@;BBE|LbPM;K?~m>C=9Hs&2HZWNXI!&aR`jAbnj`-H{jqZkq~(I%^;exjA2~L)c~@voX+(Y-y(4DegojL1l!-(ajm(4CA)C=O(*VV(+G!ZDbJDzOa^a z8DUStjS?%4Xzk)D`qE z$#U`=U4zspmh=y`zT$Dp^|sriMcvo%{KKc|7tXER&}&n%T11}Io9mDizjp0Gu1QFK z!VgY%WxJs-V~by*hQkp>(FLOxAr=#Hb zOqwI+-dKZcLbCM!LkYRJX#M(9*KMNAg-TqM-9FLm&5Rf*8n3(JdXGnH!vtBn{V?30 z?zt=%)@vurp%Vd1VgHV_&H6Bh=jYJq-9D~?*Az-HkQ*J9!pj{xOLxpRTTDn?H5)Z$ zFfq07IwtPZQY?;li0DnqU6iM@5~@ejDJTBQ&R$Nq z<_7zpCY{%;jS4(27)uR{9(I|X@%w`r{iU`_s~qUooZR6 zJgr}ipRKOC6PF@flhhkeJ2o888ZRhamfKpOZ@A)cf?Zd!tVTPAH(|x!{Nd5Q&9*fTye5p2uMTOlCvuW-*|`gDXKSPTiP22ks7nxcTaUkV zBByIS4nk70#==nX_-=@%X5^&L5Jl|4>&{|6LoG$>!rmOr&eU6byY*e#%N0%6=D15jg`4Yv^qlGq5k?wFOmPkDly6=NBt}lfR zyWE379oAe9beSCsoelRp%xw9j|DRg-8HQ1(ayZ6Nl{^TiRo37vNxYozwmH6wE(`?lI8R`jYx_q%vxbuLmOW%~JJFuSzu)V=Rzjf>K)s4hs# z%(7yRi`QDNc(dQ{mtMW*6?s{jP78MDW;Ko(h<520@z|e$aCXErxy8M`VB#-PiI~hE zk+j7KZe~IBEPOl#;AhCbrzw%mw3g}RF-x)x$}VKH#G~&3KOpi2Y*RC#Ax~9-X1D~P zLKvJpU(lo2elR5e=tB9=)R7ii;FO~SXW+T6sIbmTfNr;-Wsf@|AK8zk!v_>gzndsB2K5qvh>aYhn5kwSLQb z;N_u-a9k>S3XQONXU!xKkZzj}W~@J_O{y@LmdhT!UO4D-t2w5?O;cgH_1)?YG+O%0 z@S~RMVWrfPFAwv8cBqk4_QIPRLaE5#wzRoi=JUm~oaj*I=RIdG%&|R(8q`U?CC!s+s^)Uu@?eGPBnl4QH=he`4avpR2PY zpPKUg9Zr*7++mO6E2Ab+T{&hX{&cM`WsEjqER;ye%Ie%Gdbvs>xR z@*+%{9%YZt6O4TuvrZxCmN$1cy-W8_lwu3u0ktE=K4oq4KyAnsV7%UZ;5p)Y3>+^Q z`<3T!f12R7->Se0A)TF_0UO4B-tg7z82(-z`z6>fqMMoe-1W(~Zk4XJOUC;eUS2L~ z{#n8BS!JZC!&l{SQ=<2Q5DS;to~qN{{VrxnfZOQyNK6a@*SoDeFmWgNNOZ5~k*Ceu zzn|B>Y6X2TG1h!8BXz%{7;poN6(;HB##}JskNh==g?3emufzWJiO}@RoayE8BVf-#RuPlH6?K zzGSlTj0KDsf9HPB{OeEZ8#eak`=^-QH;&WdN2Evd}apt!5TZ3>J?S* z{>J5z>UBl#EtQ>WcwN2;8bG(4?!c7NU#I>`hdS_Ao>=f=G=vRn>gFs2T6sR0P(eh= zc53;S)PkV&fn`522H+RSes1T>3dyXye&$zpd|oW6#IiS|?%Th2ZpQ0cHB9~!K;K`! zcg?9Scv_}KNQH){4Y7IuAtm=NofW99Ou<7^qcsGFB%I?0inykE9KhP+Oz0=9z0-NU zZ3_cZ)$7)UTzby5F2@Cg@tQ?tQCPpNb-*%X&_fMOm#nMNJ_%%=Y|K9Fy4dx-LegvrlCnbM`f&GBA zaWbJE{=ceBem83rOx|ad(l>v~UXzNQ98H+CK z7xTfr^;c35qoKdxRbn#t&PcOlU`taf;c@dIBZ4(ao1&EeY(?=*uUqfcb|Az(A!6YA zD~$J9UnSSlEnb?`&MQ-~oXyQyuT-I-54UXS9ujO8);wR3`XiQ~a<%p)es}|bQ~gxd zQj?cuxn%N2N!|Q;LBZ^g{>Rxs-yhH*Wc#dN>O7gdchof5`8l&s$T-WYzN%KAu>KJgDs)_G* zY<$}vfr+$W7%J^S%V9FCVG>nGrAT5?_Abu@SZB}~>(ExOaYevUH} zzha;#wy8|s3zt<$Ufd`l3`4hS!Aj;NMh{|FILwp%`=FXw#Bi?f*eR74YqvpJJ@Zjd zueW`N3HBT_o$p)3=1=`4$(1&qSi0Ff>T-)MQ&}^0WTy!fp#^>e1aU89{#9UfF z5&wso-%jALCO==XO?l==t&IG2aDT|DgqC_P+;W`O_6`-D>#w78 zn__K-amqqT{>E{RPc2+P!E!k^9xKm%o($%N!E|TmmnJB;13L9d-CVXu`44Cm5s8K%VGnMSctU9#$8ycIkWJ*)5st7dyv;ibu(zxTOPux;brZCw>b7I)>xXypy$Sws0!5qrW@;kK zlZ_yB$hwE>BkE7LnUI>dt1v`P#|4_bmaelA?>*y+i%Y)`7P)+0o5Mr(p?-d~{sM8I zR`ggckoQN(iH*91tm`9%=hy3}D=L|g&!#=~;*euH4`_GCEcuz(Ry^Y``7$Iq4;Ntv zBNk88V+Kp|{nPB2gXaHMQuewPfHe2rkzKo$hXXOh)B?!XaQq_wZ((9Vdti*0v1sqy zTW)6Im_xVWq~H~iIJr||ewZMOv8rR3thK>-_xoFPbILOp;yxr-)jVQM=?FXGu_XS6 zJo9uO-A2j7b3SEmXYKnAAaPJP?G8~}o_fNs_aL%D4a7eQzb0XS=8)+baR+`c(Mjsg z;4<_kI5yeql8%4(@V=o(u^JwqR#p%9@ABZPta>U&ZxcX!-StrS6Way4pSC{ z^mA&=-=RsEa&X34pCDgJgK|&zx%dHDR>F!j1789neTbECYH8pHvYS>~ss&73@Hr@1 zV30iM-LRywn-TAB^4^7sxzCZPV9x!ABdAN+j(v%O`2_~yX!^RHX)S*IC%(FB-_#s2 zgFbFH@km+!zXX}t{xURk4F)2Pucc=UY)7>lFSvW8sqyj-IekiWGj{bC4e3rU#r!}H z(e1ApDFV{^oaerL!M7Z`<**;oM+~>k7Cllo)xQV9_E9n|IcoOVE%ZN{YDSG1@S~0g zYX@h?Oe5a8J>wiR>gW@sq5`|T|5I-3T}w+r(PCDt*nVb9J?ov|7>hLDSe-ipV2SY4 z%!?q-zTIn9FzGi9-s{L0Kc{(CZ~SJd2xq1I;C@B0=Z@6h#%c+lV%?PzT!3apLT_%h zQ#!O-wVZY?g3{fP; zG3yVO@P}D7fJ1 z*E&n!nW85@y)ugFJgysP%?M6vt&EJ7BP0JQvwAJ&hy>`)&Lwy2)G%pj^1s6xx)$~qz1|6zg(WZuGc)K6IbVMfgY*6ezcnt z1;Ip$hi2&AyG;yEe*c9pe9-AF(Z2S4a^RX&kP%a~y9%c>3)z74H&V96y%p<$ zgK^GLXyJ1l`5;G55|JXr=Jm{f$^F`oN_{V%nAz3iCb^@GtjI0JE zd6N=e6+87)D}SjpqTl6)>6i}#z8ONV^6%UVQ;8%i@l@N$F-u`4BZ+D3Zj6S{gH0Tz z^cK#}go5bD96My)?@EdphnXWfdm0`p24Xcfwa~@`>Bi;1CPAG?4qVEQSL!%WbJ?n^ zeGBwmtVlAaEGdaA=y4}S?FHGz9~JSeVOFouMN~dIy<)Doh_)#otZIzD);9thA!qYx zRI`Yb1)<+NvJJ*ysede&dA&@TDnXaAYfH*}Jby_dCky5sCmMQ%+9n}#ng8?W0%|wC z3VYXjtUCsl5~Voo&xm+_uilG=(b9IGSz63fH&wX@Wx1O1x98)VVs5BwD%`!t+CSB{ zx}obTevhxd8=)YRdyb!LWQuTTp>+h;`Y2{KE5{rwTVur+PC1gN*wY0wF%S`3y<>QkY~WDO`x)entMVAm!;m<7Ga`WvE8*XHD{C* z6`myCQUjGW$YK5RmDjB7Cp@SeCF=GoKei(M9XrN@jLba0j&DEuu@Jbq4uVPf4;*j}i3)_<@ zCz=yI9~J2N&RDSbC9P5L+I~pUHMsp%nx4E5prG(1<7TrYyIxi*S^jGRSbX~Ijgdc? zKHEQ*h}XMJ&xZj=31l>IbM`#hAjtS7|*wO8w&~@3WraD1Z!5v2w!89u!Q?mm4Eh4oZC;6(Q6Q4 zzyGwL6mue?kf*=lV{LrRA(rJyQKI3Qi05vt9MRmHnBK9!Vo+f3pAd(&RF_}$fDB7J z>BJ6?Z6s<6vaz)xsk2aflNSGlVLR&YEXZNG#Gl{Vf>%-rYTvS#41KYZkpC9r!g6c> zCN9EBhIh;*xCI!;xc%qG4Pk;Z-5ddZB>Oh$Psya`>{!hVd+CMeL=NQ{!eUE;Wsdqm z=NUuZZAx)Mkwp}1@@~t6^DZp&a}sbSvg=4?S?yHN9R09EEBWhO;#MJYBP|&IStTG;WUHIvE_k_`MNKrvfZvA2EgCn`)T96E?xzC=1SD`4 zbPVi>PG)Uh9Z*u%hYI@NLoT<+wt`eR*sA6i7xz6m{vAH)>q4>jK9-92lffU*7?A!z z*x*?2=*vepuF4)8Lv>$TVhvi`#m(xm=XLv0ft;hv@!Rd5u1Xg8MPd7TO=$Ito$oK% zh3GeRd!*sB|6?9dVm-psPeTlsNcGn}kTT*Xw~f5&xQ?fPvh^H4kRiPyKD9&2J+9DU z)eL9m`E%s$TFUOwnFA=DB-B}u5C(8^{!V>f*nSWpb#OkFl4+_kiRda^rEqiy-#eHm zbQxCPxFXh8v!T3g(#NA@+H<5<0JvpOP~@ch$d8@>xutCR#`SP2MWmf24RC`jUXlflk8UG>H3b)E5L2S3ueU_{!6NBR~8Np%jY0n>3CYRP) zAwv$bEpauY(9uixJ7??m8AB6xC24d-LHQ*b+M)ZP9=m`O|ALE8s`!6&eR(|8>-)Y= zsdGxS%U)@bU6#pS5tS_=$=Z=EG$G3v+o=;uj9m<-$etzpGE7o-rifuMDzlto$V|pE znBOzbDdY3)*I)kV^~|%}>vdoE{cd@HpWJ59WNi0usUqyr!?l7448f?_o^=Bx8!8z1 z>LZ#2>D&nPr#c%-@prpaLd*JbL1n^%0#?^55s%O-7nZFWu`p)uf+(hg8-Ddl5O#d` z{L6*XnNF7dPM6goyHk7#U7a>Xj?+0Pi0uT=syg&QmE-fyk<#`Xo4B4_PzLo6bcG0lBnRFLIU1oC&dX~(Ip6S#CQ@m) z8<^sis}K29UzzjRU_=&NuBPuDH&yhDP=07^yj z0zVOY@rT8HqQ|ZwKuh{v2uHwAbc$B{H80mjo}Jv^i=1)8#pNzsvrlR%^qW>AZ^uve z0RWW37M~{vTmGB>cg>S>tj$K1M$Yll#)TN4Rl3!IF|`ye8g*r+{P4(OZWBQhU+|E@ zgvv5klthGnc&Bi+A0fRac-{*VXK?cGNtNkmwgz|`R}C}X#_%0vD&D=1i0)Z|fP4D- z;KskP71xW%*RwTuGFnsdSc%A1EZSJKc;l~lll@~b&dC_f(fdIq=mwI4qb)~zeT1b(nIUmJDklJzF-k-n|Q4}U<5{YPlIIUzclXM2&ju1|6I zK`kt!2A*9qeEq@U$^9K%7{6k-ktWxfLlI4e?6j%$Vp0t%spZF93$B~3+|19l!C$c_ zB9FV?c&}wDS~2RnIdA>J}RvIr`3U6V9AIS3rb{6pE?*Yp~f z8a0Nj_a8aahXS?BjL zso25z@x8?YhR^R=S>73BGt$fMB$Ydtewbznvwhv_mC(eBj%D1JaqU&Sx1`~-d`A^^ zXlQ-X&aAl z$Nu(m^kf>n5*lC84)bYtYd}V{f02L7E;%~!PYK{D8W(33Fo*IaY}QZ(WhYxSuh!wb zpL6{V8qlEP(db|dL-F^lQ)_AT?8g1$*(v@VLXZ`6t7=qwO@rUv8o~wR2m#={u2AWj z&#$&=81mZJRzn4;Y2_ZDj=;@SMSdLNRlAL=Rx^xwj7s6tb*d_Bv$GU^Z{x;c&5p=O zR>Ohj@>=^kJ=)7JLp<*_k3on}9c?)8cJm0b z>)DR|Fy=*SU^)0r5!L-aHyj>s@A`q4u|KW`GZ4Z!WA4r0nVqK2sjh#wV>Q``2nqSw z@DZZX%=m!-jz&IHN8b5jLl_r7SGV)1p$q%~wy5BBU#1<-&08|i?Q0={Qkwg(ak@r)%u4Omg6x+qk zj8oQjJXCnOcf`L-9Lc%O*gTL5Fr7~q+2C#`m_5@2)va8H#`6wtx{<2K;mhQBKKvQRgsT6uddzRNF73^^>>c^E+u!LI;dl{=6NTQvLZecIqhZ z<%udC`kSk5=?A&#@uu-Q2W|s@brz;_rN$A_7$vL*=|$43eE5NbQHPH&ClUVzaKS## z?5umYpwHhuGJra^%c2&PGb=S(H=NB%)p7o%Wi;mf3h^VPY1pa0Bepo5_aqQ^{V((W z%f#3RBTe(QlWw>KAU1p{I+QGG^!r}x0K#&E&7QE6j3LEkR8@~`mfBCMm-vMc(a-3` z#3jWoS&wSQMjAS@)73rY;(s`j*av2QV*Yyp;93a{QtUvM(yU<>@4u$$48YT~`uP!< zYT@e3VCItbt+n3-qUurts!YWh6)!+^DU-6*;gW)f(FcPc%oCM&yksj=`SHFGzcWfZ zWo&LjB@sT;m5mpE=-8|0$t+(-KNQ0VdZ5lhOET}=*7D`zcO8dbf*+by^!KL*_9c7B zcRo|(7h~XWAiWR45oMcIaR3kb32nxkZsulXm&xARpVEZ2nItr zOFaQo0aNPD)U9vd`S#}~!~NmK4+H-rxI8sqa+t$2AiV0=?S3Ol(t6s!GNmgXVXl&N z@;GUB!`Tm}(EX50Mz{6YV1RdFD&^cn`1`)(md~u9o3xRn?405n?cmu3_pb5YXb=nF z8%+QNdZO#bj=iY)y!$wXi^YxPL2&;!8}&ZJgG5yCo_ez)LUr%kr@igLD&YPNI}D}e zr>xcB4_RijTJfY@w2TdSDQ1lDiv;PN|PdKUv^m z{rl`GIABcgjFGk7adFiPcC2f>84pElnS6NI6`Zm)>D{)f7P5MgeM&j8#t9@xhJUQz zp?6j2beS#w9WhYBBg4r1;Tcl&D&I}vz-086_O~Ei3!S{Tf|D4s-?=Is=DMUPm!jXe zVY}wDkeARtOkIWs{}l9Jbf_?uc#i(ZEWGi~w?^o-HniI(A%>*)6GH%s-kGyG>GRBV zh%eUWM8kMzV>-J;uy=O0DF@}U9RCci>VaCA$xvOWN>KLSp{f*i@>Q9#soYvw^GwP2 zXB~5nLuyFkp1(T#GwqArmMS*&=dBf!hYMIdV0%6UQ7HucFsEio*pyLjjSo4;(aCV{ zwAB3%$S3h|vFu$i<{4F+<42$&#px$}sWGq@z#^V1E6LRMEa?w;ARO}NteWasb$$GS zdWIPBox5)H@)fvI+Gpi}zl0B{K2&2|oJ~_7TWqy)f3HywbcKRNVbIH-4$?fUo0Gr7#UN0Syn7JLc-0j^p>lx!yazdcOeoNMp}X#Ft7ewN zs-@=qQ9!67?%V4HY6b%ndEKyvLlIZs+SBu!afXSI(l{OQZL&(Y4~|pjOPeUOF3V91 z+H#L^pCkBq|J>%z$LlmR?2Qu{GYUN+$34g)&ivZ0Kq6G+>r8p9T@q}LTy~pSRy{1Y z=@dJ}lYPRv-Z*clg!HT{)nGh$$2jscUz}V6wnoO5AeE`#L)gL$YEQ870-xq-b-H?+ zy&11|+|@RZzFq|{+mTqao5!`Lv%xuP$u@frjAKz9CyyV1H9XG{wDXglW!r5($7n2+i59P5)>v?54G(>&5@?yuza`$72Pbh2{S4P?7b z+PG`u9zYjkt7IBu>Z{0_oEv^nMrJ!;8vH_2T^Ve|LQiz%oGU!;M_$Th^dYsOh)%vj zMGi}QH)r;Pn5d-x^r&ymXz1Y~=35q1D-PA*4=;dWZ$2%XiofAgCsL4(ANIbl5wf>} z9&HH@(#MO0=80%0pkbEYd?fRNxAWhifX7vsDSjVhgk1&4_lQxP@U#7ccIAArI?xO7 z;3Y##-l+69*iN5>v$s}O5lSwhmJ5+>~)gQY*3)5`6EHDqB^hplXnbu$Xo>VWTDClW56vk3Xdfz?R|)t?@(^T zmMw^es5KEn1^l^`7I${dPTGz_Eyd&}0dek{s?)7mI8&cGR3PNTr4iDj;2xuJcGjms zhc`O?&E)5sRDz3X(3JKY?7Mh6*$m)wCs)T$$OJGS6YWd=JEtdZA4v&2+4j>Ayt-C- zwq5#%A-KL8q4b^NFAw`eUIFJe`W~Rxo}o`%uSTI3XFB#sJmzqI*NR2D%_^aPP7Xb9 z8JvZzj=zsVo@evN@>1LH)$o0YLFvz~{3ANb^_^VlS9YQj4ZA@b37B9me_klwlIbN3$j{BJ-HsrsI2LdNn@)zF$+YeU{)p?RonIr^^R99BkQZz&x&kI!xk^whXcf zOV33Vi$1%nTN*3Byb`V@6+5YdlB_vY)sDouwiyFF1C-v{;+M&Hr~zLTQ`(=!B@iJ- zZSFI_t*oXb)tzq2#hxcL_4rOn;ce;Z3#mcxVgRF;PQlvS$%^b>x6H4ueZSE@1&Z-( z%Y5n5U>#%hmQg?IkNgs1%C}mFrh*yjo*uv13YMwW_6P4d%!3plAGwad8mzh)UMvon zM}FNeE`=+R%=bot3Ri!xLKld<#G%?%B7o%PH&yu-^#$hrk2zrXm54w*XXmC8b;M7U zxT8eIDfVvj7;hN8F<>l6Vby0gqj$uz;hl!sqKstmd7nwPMU#R4!F}VWgLaU{%il~^ zxHfFkpr3tIZky!EMb81ePZ7A~D?UvM{M>YX-a3QE!ig*)cuwK@c&(x9gt~< z)B}Bw1-G3?y)C*&+k}-n#!3Y!d~M8P=>i6|*SE)QmK9>UJT6vM<~nnym5*F1S+mLU zPmcRuI&;%c!DB3Y3}qUh;<&k^qIy{9>du_()={l3+s%3oE+`)kH8CHnLt`6>a?*f6tsajea5 zfSVpjoqSY6MUp?hmWbul{pK>?(7n9Y+P3P>SKp!r&U+Vjd(1O#o62K8^g%^EA8u(u(GT=#uz+M=^lE zp^E0s9@2Wb;jQ;-ASrF%4%K|9wOF3-6id^D64mI9?^G&&)(odynyBLfNQ!$u z&~v0$D?XV39hHEIRx(ZQi!AS4<_AE{9+oORu*fmZ6sP6ZJkr1>+o;!NC`NPG?0;Gw z48M*2?3Bx>xqjmNq~KMSGD6W+T5kCAYD?Ad&U%-5WOD!2H-OelsB&6yIeS27D*Nh# z&~MLfA85K!`WYQ8fJwanNIHaYU?Nir@u?nkd+0XXyb#Hbq%FKXFSNmM#{Xqb6&k8n z=>V*Z`2GGBc2$3hZ$nv)0cmSFC2D49exl%|iHXVNh=kC6H0{d~NcwG zARpbc^uls=e9>N_P5U`PgZ-+%KxH7C?7PgnhW~TNlE9 znO>3Oh)?b5HKl`y6nE`3ZTV1DP}0P9OBMo~tEHluu;(kI^pouBYr1>aSL|P2AqjrT7H&qp<@xR?04Nt$yIf4PlvT39 zkME-b?Y~OaRvbCRKXqF*_!(#F^(fy_D>rzrH4c#4?Eo!s^4?;#-?F)HcNgj1gIh2k zYh1>?8$r<3f9e?nJ5umP?KvjhzJd`&luM~}20xuiE(MGSTVB?#xBB3~rpu#^=j!Io zFOgY9>@2o*a(-$7o4N=f+9H4#vMV5b!nNH_Bu6Qp^*FfhqT;vf;3O-V9_C5YseZw* zV|g~&-DfB5vUkd)DD$907I#`B=ys%Id;4Q-b$P5Qa@(EFM&G*6i$A1w+6dy%-G9UJ z@6Y?x_%{ccYwODfUp2Os;0(ErthO%f?Szz7$p8ZCn{#FJQHXhD^8^Ap`6M(!CplLH zlXAZ$ymQXtkUxH!eAv4n)wEa96m&4iM#$>I$E%TUI0aL}YLu|3svFEFW5EvN-qV=Z zNkT6g6MHD1S$x6pWtLEBByPcH0y}dSv@%^?i@>z{4A&8@SRqieh3vo#s6G3)f9bi1 zov~e>CcJ|Z?-yXrt6;9YwZ$@J9P|QV!ME}*lN*ci_36jheNpU431p6-s@y3~>ESlT zX&8yNJUj;fIO~Pc3dq#(Zf>$Pw*H?ZJJ0n4KDVz7VB`Kh(qEfR-|L|k0LlsG$I#BR z=W#JJi0JYD+jD_CBo|LmY+Xeg0bP2A*mr6_2IKEeF|rP76id|MFQg*%8Zs}&>xk=j zXls-$BaeT_h85Z`Q;4PUMpFS)jNIzX34IXFbhfGQ=0GeGoQGaEwG1w;3`xLXPOi8o zmfoFOc-&6+4-}a(`Q>lVK2_fT)X1yp-neP@eyH3)nHQ+);}-{$2T*`ZqwluZ^kx3W zqM~&Z2JdbO!d9^k=$05XEUG%?LZW2{wlDsZD~v}mfd{H?{|YN#?;Md{e))fF)1(V7 zXtVW1xp1}d6q|Wy({}qIf+<1`x+Ow?DWN1L9g%+&K%WS>1(6#1W`-L2VLr~Z7a5i1 zKp>)nZ`*AilsbS8{@ZV7LS^twHM0HF4*u~9IX9frSyVFnI-uteGjk?er!|e-WT20O zD)+93u7@CiR(yQyvWN0EwW`T(DOHlahZ2|iP%Rc;i|W@yAsA_gKrORPDMj#i~6%PsN~(eLMG0wq4hT4GX$y&wGUj^k;~pZcKWGqky7cRQAaDoNDd$ zE!B38A_BmGOCbf}ptZ>p8!LeTaV`R01uAvP8sT*h(Qa=su zP-m?^&{J#++TKK}2__*@hEKZLi{|S+yuY?%Zt+2NnUnZ z8Lsfg7S(Iz_5Rgmqq05s3wpSe%D;z{P**#kfFz#mwq=*j`Dsi6J=%^`qWOb9j(m$N z?3Tf5^AZg0!ist^y16v;S2q-i^)NWdmVqK4QyHG zOChgHr(mICz4t;ITO>i=zoObE;xM>1pnvCCg!ZA#X^mlR>kfrEpJF4sQ!l-G>*|U0 z!hY0Qi~Ta5s?ysTw(j%4KcsbW!*5sS!Jsgn9XJMKcN~P@4X1K(2Z3N?Y`WoXxt{R$ zj-=9dfL%;T7kp#w3N*)_3_?6-f>v+Vguom5r}F10_q~m0SS<$W#Iy7V2G=J zl+2sKb86pT5Jd+`*yn|j2dDtl(pIn4zL4nSk~!!}xFXupSE)u;ovm8F`4NAGa&KjT z&99S=klT1Sh7k#JL7$Xu<)Jtgkpg_312z`n_2*A13gZ!m5##74uue1H%b^W+L2^!E zUIC@S)8Ye;J=X!snZwuk@y=X;#l;5l4X4?Vct_W^l1o-L&@|sTw4VB4S$RlPYhAcB zE+lQf$+QFI(y`J(Q!Bt2FETQ>649=55Y&&ttZF_|$I>D8gDRp0oythsZsW&41IXA-;l*(&;{^9tWyFY-R>eh4{oQVBz@4?YuCAHj z8VBf1M&*GXvvi2|2Nkh`w6`eSZrZHT|J_nQVP3*X^LL~^-j`1*YCe&yDpf9B;{WYn zw)q3lgS71vJ)l}&)Bga;YsL3VKEKw=t671a$$O&lUD~ssH_N4mCNCLYeRk6}{rgN? zAx_I!j`=O>!vufani;KBKBudc^Ka+%hiK5NiP)ch&JEVSB} z&26FqnZdbW8wC_)6zt=E6#jWrgsJ)`O*wOft?rr+>mU4m_xr@wwO*pjki037uu^)g zt9D)}w82<`AWiC*`*E9rCFYWvTya`pf`Yy~xRXTZq0?Q!r1q9JYDM*p-oJ>P({3sM zI%KKAd?lr&Ieh5k`&!T7NjjhK+;UzK_U3;0r!G|RM`CGj!HiR4nu|!bXiel=w3u7G z`$*PFkVygAc^Y$F1kTSt#b0G~q0qP&d493|R5OQZVqit84TSWB2UlBf=y!Fw7^gAg z64A!IrZD^*-wdciLa}a(C_wf(>*=q*k&K1{kYzFQk5b+Ye_q39G&lOzY9iprv`hP7dfhQ+}NJSfzX4` zD;1G?5^T$uRnZp;FOjopUQYLZN_ZGZk-XCf4!^y7zqHG!U-NxQXaI1+@#wqKBtAga zwYT1loj*D$=jnax#aeGMC_Ovvn}Xt|Hvu!ggN$S&l88^{kJQK{JbXkRWURcVr#V7S zFg)WnAZD0Djuh-sh1$u9S%5~pA2q=Coh^f!Zrko9;5rWH-eZV{TP~OR4$mv~R2-5M zF#!E!6Y17=@eL$*!MPew7tJY6?|28bA0F^V>!~hV}!*&FGoArTY2;8C^vK4oYsZerG_(2=zSZnL!I5I^#;aN z^z@1EY`2jVGjisTPNT7i;(nfilA{q%)bMZtJ5@log7LmC(O^zyIDWgX=QcWN)MTgL z`wgUwgE0-&8Z)fs2^Wpa+5s!si4^#YLO)jm2rNb1=NBV^OJ_cN5H+FQp*|z%(=6_;E%Owc7y$OB z03}F>nQgeN3~*WKdA96aglo5S-7d4yUj0@sSZkbVO;xm?ff+M{9tSpdz^ZW2QTFaO zchKT(crvBxpb4wP@y=lI!1Wy^jwtHAvEu33-K5LQ=oQInd+b80GknShP*r}KT{9pO z2(cVp__)i!8^Uwg~lkTYYq*xEvZCT%nq zxot6;3|Wx$RR{LR6{amtoE$TP#3N)k-ks#nyVGqMRA$N4;^9JejLl>V!CHdr$E$Fo zI-_sMV*RF>MV1ct5-iI7>L5K3G)!o1XY6pFr$>5sBSrG2w5E^K4*vOy-FjUY^YH4l zWsy5?62@ZtMsLO|1I?VexAN!sRv8Nw3VWi`DbVZuU)dj2`u#~a{d8TcDHB8kEj z2$NSu3d3y)hqiRNe5;n+j1%Iswxrfro6wNORTIU`{Q=5E&*&QKuFo3B^A#UvJ1hGYPWC3W%C@cbdnFMyNlANY0XW5yd++ViaxI_ z#rpX7@vmu?lJEfCx+ETk`glOC#=V1Qfv*9KkPen~G)Kw}QX@-{%RK>oYo6K3g|^Fe zjaFK%9H0*%QfUPNE=DUs|aOaNHFNhEmJX{Me? zlTni~JUH!Ycu2p9A{Cf|7ghU}AHT>@$2xn8l#&k!&HEFAu8m>V#CXKkbvauj{K>fu z&gHU92lTqzHuybIiU)U+E6y*7O++Ydd>f@7Z!=hXKcNHX-X@Tf6>buh)!6oFKOGsf zx1;X(;tM@bry(n2_2iDyQh)6zFt3H8L5n~49L}0h?r_oEi6>|b!|3C9f{3Xsj2=pL zFtKHa#)!kY8xu5heRZW7VQwZ_8Cj|1TLoH#f?lL@7DDi^eyVo-rc1Cd7b+be_tjET zx8j+?c3v%mQZ#{j;}E!C&?wJVXus_^-X~8;MU+Lq7BYi>xm9Yl!dQAdVoHlnnnOGK|eV|?56B8p*1KL&^9K{YH4m)$w)wsA4UEs9MVKK0T`ill;$Q|UM#Yn|< zDgalQfccxUIT|!*Fh=p3W{A{xB|vgx7r)w?GHjUc)i9Q&w=BCeyjUy$p zV5^*kCoZmiUBPv;5HRTmP)Q(VXAKnNcQEWp6TKUcPXNakwzs?YC>2x&V7RK0EQiq3 zwP0z>USuPiX0Pz(@kY$%GvH0dfJsIlfbRKui!eRFY_ttmkT?%I@v4>-whGPEIAtF! zhNcRsLu=gOKDsc{y~V5@(I?s+BL#9dWH4`C;q(QGAuR9hcYRZUg4#lI#BSW^D!_Rr zamh9j=q>wS0!HTf<|h9;CNUB?Gzs#RLH7KUo!1LZKq*NIp}TQ0nqYP7-8%$fz<;;! zWJLqDbjx^I%FYM#6*lbYNjKYzBG8_vp^O5dO7jXp8r2u)Uw4vRPtNjK<5DJ-Vtfce z`e1TTrxyPd>x~s`3jlx~Uj0E>J=B;Csfh+m>FTovbKPJh*D-|t#w{JJEG2E`D*KFRBF{9X9@25I>MfSzn1Q+u0h{MmIKY#&J?Ph@BQSjqbWp%4v<^_AXssQojmY> zd_UVkTcr%Em+0O*0+E}lW&s2CPJe@?(ZgL=s)+&9en!%dJ%XobNY<@~ejt0yIRZuP zp5N(&ik2u1;rFE$88>}faPM^d{RJl{5Pa0!>)C93dF zS2Hun7-Va9CCEkk${(zR>z^b(ZDGUD)_eGW=6f84%lGpX3W&Hforee5nrcL{JU;NfGAS&kv)cC5C$On#x(} zG%^LiK5A?98pq!s8*aFsK;uaj=$CtA)9A&D`m+X=#nnqGb_GE?*(}rWIEl!}{N>Eh&lSAjn|5+nAkGhue@UW1 zub}EQ)?S&7hUZs;DUD%i!{eu?uz)=47Z#bEkYlPISHt(1$>)iE3{0yg9L!WZE2?wI zXFZC#%mYL>GP09;dLF&GboH}yuw%08Yh$(fYjbs7;%u>YbN@Rs@=_E)OW7mG9xj1Z zrAkJE3h9hVe29Rl=v(ZF7?|SKv)0577U%!Jiv#dlW93zHjRxzC`}=%kg|1cQ+!Un< zYcIN1QD!6O=GS?wH=pTYn+KG1H2zLAzLxc#(HTjT2q84>^-tHhccx^6B6b%7>e@f| zG%ZJ-9P2D!P=&t)U4`%|fmJNtncr08-zjQOx3H!6gn=VDqq-D!p~0QMQu9>SD2>eW zv_GZa?)W>=m{S2>N#BYfzn6UiEf4~G#Jxns>C5@A%m2^Sa;s!dftYW+U>dQTl(>U@ zA81#pM)(FAD;HOHJbM7dt!?mI#GaW-Ai)}I9bvn>K?(4ORhQTV{$A@I^pdE~Q%;TwK;F)Sn01$L&$dbD3er|n`8)uzIM0Pko8Evi zXK1H?x$0wfpY0K-vE_VAB&`#P1!n9eP_3MG)dD&wUkp%DIU0H8Yp%5ESx_^ULEJpN zZcWbo!T#GMxz+H_SgWRu{7gUNHCW8I8{-KGckI<|^)qZ3(; zjHKls;0P0bNRDCBgP7nmM5=yeupH1mhUZ8y-I%?}4et4A;Sm5-0S=294A#ud*8~<1O z^m5p9OnjR&1wjIjx8pYzt#epyU=Q9@*pj>PeG6UXlxGBBbWu;8|!jV_FwlfRUgd+;$paS<4Hkp$zvkK##N|X*P?jfrGK1$t^EHrTiNk= z5qcTOKzOG;cazc}P!Fnd4G03BNUmFKplqFBV2s+@q`_6gvDjmUuChb+8BGDpRU}i{ zU%W_uIv_-KanofyL!%^6-K7bdEH<14-U9&c>s>}*-Om7ViM$t#!}nava%&GV5G9>k zClrP4`OoxUa*H5mWIL?p&v@>3vZw8k1nqPu5r^cfHppLM^I~hiIFyB|K5U@0Qx)2q z&{lW(7e~HA25mAy`#j)x?4u;|n%*(Uv$TaxW*cIsGjXHStyN%vk03WQ;jidxSp|h6 zQv$SW{EvQ?*nV0;^IrBuj4Q`{G0EQk@0V^+;R=Y{@>;@U!{a;0)ihXk)%y`vvqBFl*t`oFw zDVUbsxPypty%(|vv|+9h(GddjiW_c^SkMTiE2r1Ufid>ih94hG!^e+*kVU|r{Yt9O zH;eA&rgt?5%YqQBdD_pG=>e+@OU+)7P(XXQ-1|%!e_xP%j)2v^qF1PZ?P&}9u!AF# zy~O74hn6czmXdnV_2cK20#SC~7`tf~HYr}@IITmyE3y_$5j*1s#+;}@n_hUl*<%#k zJ8~YZ#sNb7Tn+l^3Vc61rPY+n+8xM-$D|Q^EtFuHqCuzad3tB9Qq4bgu@@;ODfCiTF@Euh z2JOK*RfOw@mYOw#G!oz*Pn%}WR{}U0l~J>UOJa4ORLABCRZ1>?YJBRCdIMmAhG2p`BDfI0s5PM^^@Y;edeR<$whbq zziB8rz8az@D)RII$B$jJ;<1|SAvib_hEQcVg?7X#9=RTbLYS#hAcbr<%E>nPGzK9Nh z*)cj9H7@Spxl+kJ@arNza^9$?b)WL+Cu4OP~I)h_{+Q3Zg6HMGp?B(sTWW zg4AKUVfgjz~l#<^6n%2(e3Jb;iLf=q(D9tOf+lMI=U(}&uEy7r#X z6~tD1lslLrX>7m!S7A`ryP!KnD&&KvLfTvt7!dG~Cc#z3L3mgk0$Gf!XLim%@sqIz zh0jlIElwbQfA!&AremWHmfg|_CQ50mL$&feUhi=Nz=0(om7>@)-82G)9)THFelI|r z5$*l;mxtyD-V(HuLl`XG*c6|U zdaiofKXxwlD|3+!fcq@|v+4JBU*1jt!G@YZjOeoGhmNPDqb(yA%hbVy(d^aF@V#R0 zjq|k{&+8eh9phVbU*}I^I$3wPb8&v-XfTRc+H-I6nm>wqoy_dahE$7y%>462klVM~ zx(~`G3H`~OF};H}hN`*T&L|OOXn{c*zdw9Jc)Na&@}p>K^(1TXCw2=m^}deqMr7PAdI`jKVu{P)$5YHhv#j2+bz^^7{2>E_Gp zZ1yvZ%ALv4g1(2;sIlc}^qzQN)dlrW_QWd7_q1Bc(e^m}=`w@FTuEGZKhn6ESq*f; z0%UY*C~9s9=;`>-n`ALNw+g1MOhyy${_>5EeeE7gvTs-L309dlrS0B)ErR1Bq^O5x zPwTwFPeMx6Ftb-NP0^0;>TS9(FIV1G@UA+~9@b%Xg_<-~f5dvgR%Gn6{cNRh$L+QD zC(k&s;O{RHkm#e9Qu6_gOf+ykc6F3~zwt>ArYJn39SPy}XxihHNhfbEX&|lA{&8|< zkA%r8(X>`~QAySGF6`q};VBo81ISqkD&`F+Bm9~Jf=eIgzVJV9V9lF=IPhv0hs7*5 z-6p;*Aq8=BC~P9|SZQ2J$}QMsGfeA^{hF25e=A`hq3_yG^1aFxx@*e{m`I-a=$i(w zJGDhRA`}LDf>DoS836i&lC# zNng$f%+t9s3u}_ULoj5Iyfiiw{DlEX05KiNBz-(Fk8_XvgGrb>iQ?R4n>Z99xw@|0 zYhtHpu0WvSR$&q;Cynfmy`Ba?EJsb?d3{?WR((}bWhr*%pJQ#^-v0!`e!r%T%a_4m zmF~ZJ%fmBCu|t9n%ILwX`)a&rNOvXPgG3n6kaz zxtvaALp+kSFXtCAxT4fo75l=eB-6J(geemS>tD`)_7?fy{%5eY%Ay0FqvVEaJ$_eq?#)}KHuVF#sBQl@Voq@KwsWAR zun``+JV(Mhg1sGM@c;A?j*ELilQrU)${+IJt?ed*UID(a<=U)eX z@SZTb;QUx?6*2p_*mw>zb%Ofs1v*IDE&ND@aHp_k+FyhLT*LCx-TT!YUWKQ-t)@Pf zX&p=_`*eucr9Uh5E^efOkKzPJyZDH_)k3UTcm?+svfKZ-nH|kLuVqR5G_M|} zUDW9)dW*Y@_G_Hjhp7qnuD!yU!u_yGRqww(G3p%|cW1hq>veo<7(Ob-7xu?Go9~QY zAJ3UQkwinqgQ}jOCOv<==iLnQ2JDjXjm^4!`UmfRCd z5B%;^V$>9Lxo_`Z;;WLh>fL};84TPXG940SifcL}+IVT0trHW<-;Pu3 z^?Q-5#Fiw^WbTLi%im9-77YX|Nh$%lRzhe4DoC2^)CRSI@UYpjmAImnxr;@e4lZf% z_@*lCa4kh_Zt>zYn4NEISCSTVaN%7=B{q0b^i62#t19(JrrO9CgWvNh{rxkZe=gPv zjgJ?;=&%}?_NW8mIa&GdjRw_%>Ci_IcE>#Dd(_VpR`dK;vWaL|k#M0oXPEe(n(GzU zjzzn(TeC(Cvi}4l0Y+$KEm|B`dnCuJ=e7UUQ`r&+xQjWB1a#r+J1#Hhb%m-2n~b5O z&j`d-FvE1P46Dmr0{t5G{WefM*z0?ibO%}RyCa`mkKpy0)Y8|1=@x5skf1;B(32#( zKD$hr$Ubppk9o?rbfY?;4#Jqh)JBhVSefldP+sZnSFFpgp7~FxtT~M`uh}OP`_@ z0qI^CG)_%Y`X#{^xzgc90zoRBc2O0sAt9^YrZ^8j!t{WO!^84U?uiM?)HT@mTs^c_ z^{p9~Y#W@GHLs~8hZnlH!oJ^Q2HDAgJ_P1`{Z*Z(Hw?0l^P^qtzQ`kDo5CJ*%4%hF zQncex{fpxw2Q$pmuy>qZUFVHU^P8Hx1V{Dp$afhq_szbrbOYO(my%1pkT#z=fVW!o z`3sCdr~=Iz;!tB8du!bwpIO7w8Vf!*jV`b;wp%mqu3$jz#_jzF;F?%jqn*-Pzey?&>r9>1|&8y6_gSI;B$B`8iQ;cse({nNUOS6L}c7yuN`e;G_ zfSOA{$u_qOB@MswDV*;OyK4cyS#opVRNEjs;&=iE9s+pN^z`Zh+CN6eK)a(th&Dr!drEKyZ^)dfJFAH;#YiIJ9TTUIEOD>ds!aqlnk3cKlKHL)A}Ah zhTFlcB(r&WOO;N)Li+8xW~Wttm(l^67qnAA^8yZ`eQ9xY^YaN%D?9>CN$j!A>h>nw zwJnoNZoB;z?DQP^c2iNKWzLMKSMTo7V{$V4|1VmbUj=%4e609im@x)jUw>vU5nQp! z4V}eXV-^4J-hei^?BxA)JbByA2@HgMW(MCWaL|Y@&}neu)Z$8KX*0Gi|L{Lqy%lj@ znSa189kmr^Cr6e=+K|D^CV=@C{V%4aVnx^wfUe3?uL$RULc&p1-DiC^rRcjy@Lok`tiTJ3TWlh3OuoQscHjl7h>aazV!=S# zsBDLpk7Oy+R++b0yccb#=Y^}=G^4~(D;v$i5Q8QxXr!E9pc}h@{dqL`g5a3#b!r@Z zJjk#vgKIlSdNzDf@nD*t>nZA;=V|L)Y*C;Bgq~Y2{|K~vCCd@Mp}TU zAYxzv+CS(ojG8pe@GzRhqG8uHXvNX(ZR4+D4Vv4@tna`WcP~9*e?=i1LjQm4y?0!b z=lVY$%Te3oz_zx?u3EGxL$+bIRY8b|GGqp<2*`|#fXvWZQOgoUq6DGJk`-h_2n31} zW+Gb%5S0;=Fk~e_Nb+t#O_xt1c2d}=GJWuZXy07bfUH3IX9Ls@-S2{P{ z=m%=B>dHQKlv~dEB6$vClhhu*rkxg}~hE_s%u0QqOJCTv2wx9p3*tI`NQEZf?)NIhd%QUKnPbi1FH>#2^Wly z|72MPgaK&Zv`!GyPc|yje8mN`H4=8y7Ef(j})Gsv5VlIdTeUt zS~M6iyVb0JMgMZs&{{T3hV@|`Q7Qk;V0u7#XlXb#nS@n7{E4VmJq*|;^y3Z0(=`7K ztKfO^Bj?kg=2W^G4KrEi{0HZ$iwk^yq~pxRdsky~@z`J38Z8J5=jV3TpwTT~3QWnQ zo)D)R-NPuEMsG%;rhfxhRgAgT`J1Y(n|U|a`o)Em_wKYufwUS+AzuaCK9fdz7-r1; zcW2L*-9@IH>e}TWg=vz7tQw#?l4+a&{OqGg>pLAQ*$O4+Kd+of=+{x?9_X0gr)&yp zMUVqvKA8SCcOh|)H9)rLm7bL6HgSqmG280 zJqpgrt-rRt6T5c5yN$L=UIKHp zyO7F_eQbo9Wh&9o(XomEYCHt`(V?er`+ozb4&*#N1QHEi6OlcdPLLj~0n2EDa%Qkb zqWk75X0f5;Q#Iqt)}Q@5cGmC0cP}x1xQi+p$tVbx1rbYwYCYJYbqQ=Gw|&l+R{g?O z1;W4m+Z<^Hm**kj?rEK>={T?qHrWYWw;m*NZQGxrXFVV1O>jG$z_vnMy6^lzBO*ai z_5OJ0RS-qdYpDgMivlM1tMw_>$~Pooq`}wR9?H2Ijba_AtwJ zS#K8+G>3tz*{JQ^JED?nT5fwZ2Z3f{b92R)$b(fO+wBjQjqWTW?tQL)tj;~Eu?-sC z2tQdPQ&JXI7`($t&FlVUM65j=^dQ%Cys6rO3`PgU% z_TJ5Wbb7~LTrr=1GFOC+ifYIrAs3YM%ha%!+ls`mP2hU;+HEN`IgZ<} z%=MP6iJFx5^sNVhF|m@~XViKZoXMA*(zo8atz9Wtnd=z2sQ`y+tuF$ZdT45C4aMyp0lC^af~Q0<%x$PKCj z(_y_wfvpOdMhv7{jdeCAtye4WuXsnIz_1@`^Y%Th%q>3AA$8j&I#`Vrnid&J#x@ zY{GYBXimTIA%vU}+WVrko=SBa!w-r6n-X@rrw;xD~eZLD$We$rDE zNIA{FLpRvODk8^+_>~JtZwM1hF}r0Xs`mT8$HTtc zELPCVIhq3pH6_SPsgI$e>RXjxfLT8unwkB`=@LaLiQd?`Q~CfIp>sJ!of27~e+5n? z#;2EczL@-;Yqaxhl{)ulN9|(c^iE&>`QAh>am1~B#4VL`kNHiy|BDbTT{uR*)0m}7 zSss_G-V-S8Zom#;7X#aOn=}u`0QnFAfgwGK25m`=W%Szjtdo)V_P&VxiIK9hS_-p0 zEML04bgXdG`fTr8@d1rd%im+w6V*i?jqY^LuGMNQI*i(-=jlFOPE`MTKsTf9YCC(F zJXR#Tn(`&(PFO6_YX_*{K$X0Fadt-dn3s6%Y+B=(L~>Ek)QgY{X4`rs04^ao{@j6=SMm8PuKIf^~lP$#t%!KyKsQAFN}jgT4$&^Rd_4XrCR)TBoIC z+pb<;aYZD=>eSzLwecJBx11Om(|tlnQ0Lk?rrgX6+j)2gPxpZMtBxXLOzBz~3})2- zn`!R>PZboDcK03V!v->e)vB1lk_911(4#Vayt8rpCk5x%bNy0a0BB5w5WeGSzaUUd zLiht}YyIwzX?6;f+~0z?CRloG2CaT>C7i*4V^<*w)0qXgup)Zdd)HAvK?q z;)gK!haM#yiyg}bo$=&~p6Y@;cBiopUb*SoA=ce9U0qr4JpUlSd&MN;V}7eXN!wcJ zHL~h7;8)aHU)@HapNz}SI_fu_d3ana3OET|&y z;g)pjwfsE5MQ3j<)R};T*^XzlfKq4$fs}2W->)I2tj0Qz?f8W5zWC>`2K6TBQeAP- zt~^x>3tFg*b!}>2@5i^cy;|VDbx-$p$J43`woUbDt04unqSrt58AsiW2D3YM!++Famb>PHd*IE7>D>*6(!^dz z9d~A0gK-344q|x=mXSyNg}u79rR_23c61!dJsB`b3Z~FX?LjXzumC9c>s=Srbqly> zAbfQA#7HWBD*jw%d>M99WXx|<`??l!Kb#&p_Oxa8p!5MHP)L9FNj>_`mcp;CBr8sT z&wX@*`txCUkwep&77dRgv+mXuw=8mJ8l0sa`mJ5{Ashi zL(_HM{zA8d)48j+w_bPj7hfoxKs_^sShd<-MqkO}bFE>9+Hz@^&deOIx6aij*eR%( z>aFjIdhCIJYV1vVQ^h`k?R z$mqRLUaw@K|29kB;umlkx-#+ke~pjWmkDbYD1GScA(iBD*kF94uOiT=Sy1jd> zQ>5<`*&1hHAS5rdxyJKyyhb$euTJ5E<5K_mqv=QXfQU2!5y=K3(qCYxJ4ohN@&>}% z=J0sNp{iMMTQ{Ws6UjLuo87n0?-py98;tNAd^H;0bggOs_->fT`W)#V-M@6tev*3P(y4QnquQy4qM}v2Q_rsYs$x)DAUBT53 z93kdIRSFL_^{lTWc~nCLclEs<{5^ayqI3t;q9{RY{_JA|+fu-vzW^t@A2hzGQ^at+ zvHLEW<;-I+QJ|aSL~g~MTN@$rJq&1!vO{m@qOz2PbkJDB(R0ChpDdV=BD0wnry*Gg zB$60jxmMC5bL87~68YYm|IM2oI<)xky!jMc2c?8=u!HZ0tzP?hjfXv|SyzB!@=CiS z$FL#7<#Oqo4Qvq9Q1@`W0}W2J=~GMo4U{wwZ05Opn*|ON`5`@^ko6esN7;{VGT50Q zFnE3j%9RnU2oj5KWB)uq6^<)UJ&1&kDA8^eCFGTiyLat0+C8vO7 z`{_Q8>sHW4O?{5@>zp~v=GeMKpB^Mo%SL)KB~X7w&_^-Q8(KbK8E?gVp+K%qeZt)Q z^Ejw3mJa59?HwNohP>j$0%@3ASXiXV-`)!fejb5Yd+E4K&+sEELRR!{hd%&;Ga0e2|pOb!O3-zEil05P@w z#=y5$buo(Bk@>j6u>zhPA0-3w)43M$sJ3>S+9yknT zQxCuuQPcb5Wp1sjcYL4K`|CgNf*V8D56_DfX8KbQ!xu(n<|@HZwsmj_dvrBf%Pg=V z%y;oeg|mIt=P}G!$OVA7JvI;z2r z3{H1!eLUt}@E_zNNB-@%K}P@QV5%|nXu%)prF1Zr8X9i31;~2!MsaOgeK-eec!5{9R+iA3%?4f%Ipzpe@ zJXHV@v66t-`0i|wL7)BB4-IA0Uovrqvj6MRIN>C5IaPBHh@sD_adgBcb8{fSu#DcH zA2Wp_w}WFEh2$7Bs0}Iym~Y_`sv=0~0xAD6*rEnhT>=X5s?RHRMu8IoQ#cgC_^Y;z ztw|r#aiKmq?n7&8@84*%*0tu(wA1q)I-uoG*B0;&g$N1-4U!3f_vnZ9?)C=I2L;pj zcI*V047YuKja$9ColTip07(6SsA9Dm!!Mc-e3>*%9zG7X^fzBB#5e~L|Hdq)kmo2D zpR`ostWy4g;{*hk36>yv7 zKmT3kL4rPP9p=GKU9HfH2mN;o*3Luexxtg_YbYA76+FH3zdvm`00(0i_H6Es`HfF` zabDqUay1_56+HhlV2ajH=p|4{y3X$ai_9DBEEj?fFk9ZP92J(?t09_TB>G7G-}M8S zHTJhU*hddJiNRjGCv&!5zX0TOGcV^HxlD{$Tdf=N$H& zeZo2UccL7LE;9v_P7We60%%;XWnMF``yKX13|&)<2gc-G`TvUBr6*})`||ut)J=Lv zoMl0PU1g0g9Ui-ZF7siM;B@Z&*VFm95Otu@Mt|eqfh5azWK{_aq-b{D+73E9ZIrlK z?x0CNXsve46YwXvqw`wM_3PTnz|Uv@SN?9_=Y2c+5v8T;2dmvFqJbj{9#>0_-yZu- zuM|vC7(XlG3cN`FY7hJ!8|oFhG^=m0QWavmP5^H#1LQyc3e}@EOnU}9Pl&cVIm)Q~ z`R#v&C)b3EmJ^nYi|&>`kC`GQQ{^NMicRL&gMRhF1(9~2xJCROBkrCUJRo-VJ}AHX zVw>0ij&NT8(T-HVy041Rgs@}JfFV((so6spdp8^ez z`62$RGl-3|$DyZa32M-Yz)>JR-NNRcm{Vq`S3dRmf@wKD#j9$OtGX6nkow)|HJ{lKA* z%7CzzHQ48Yj0*&qHHD+2+wY7Q!~R_F4SRR7`A^K@4o%+!4H;$zcPaILhtd}V=|W?^ zup^(SaQaVN-s-75u=RL@W0mVJtSgv~ilhB9e}7yB_86+9e(wYNFNkWf7Z_F0+^;A< z_X?0E%lb|4Weik*rXK;@zwzy;XgqwTdqR?ub&s`;hHd|HE)1F~w7 zyEiWLr=%HE{`a*=om+z3Tjw|kDB>rp82QD}(ovJOuT3h#3aO|RwHy79S-q^IQO9@zQMJVO(HCVE`71C@Ij<)%t| z9pS1PX(e(Ko{+rPpA%%q?zA7GNGcaRq{bm@Z6Hk;cCmae(t=~xm!@NVGM`1xP zTs-&c8)k2TeQiUF-u=pIfA?qLlnTB5zL;%I$ek^V5E{H$rt)9186>yGYmT#bXBX$` zf_Yg3J|*n;`nZk1sYg(+tk1vZ6tTK$CPnNyETbGT%PbOh6r~Ge!UF0k8-LS6T^N7C z8ecbbSVm6JIj*)6j_DA-vYWbj`NEg!{9I^7TH)N^imv>=w8ETOnu|6Q)`eNte-k`lcZ1su zO*fpbZ|nOxEaub>*n!6tV)9S&4YM}`sI;Svef8?jehcS+l9m^{)91$lkwO=*Ir(Cj zmgCcfQF4Ozu;cZ2HvS$7gVk5_nAdIb>4MpEG0aoN(Qe@=!yrP?CU_>hT#GT+mw2&d z!-P_~BU=|CINaw1`?m6?8HRxcqz1(v^*c7cphKi(QT|~1Tq9T!jjjAybbeDm6*Ec| zHd6%!VY};NH~uPZr#itTRq&FXBV1%{Yq%f;#Fphjbec0*YNIb6WbA)SXQB~A+{;@X zE53c@e9;4TA!(=M_HRSU^u>8u z3`cCNL?mboF!cz;$*?LBW&3TjDykZ=yY+*PnY)T4?t!6R?U%f7BW_s7or-l2Umnj9 zwzsx5Tp%r>!k1Urm_Q}1%Wlw*>p{O5sRMD8?0yT8$S(?}PW*E=4`QzRm_E(w->-}} z_uTE95oV+dmkoNhHL#Wc8C^QsZ6E-QAe=-Az>LbAF>?`7hrP#se*GQZ&!wJj~nE)S$Hj*|gLvk*^T9jS|0nGvjlW%?+| zzIm}>>6kSKFtW-JzFC|X`Y(=OnEh5bqihhdLKQx6T1_)h#O}EA*~T{r2Pm?T?ik!) zQ?|4~TAo|%#vH}|^UA*e!y7QK<5x!vn2_+}*vgWm%K+u*JU}9$q~E>BgQFYfY3u@k zXW>03ApA~+bpfj$vy0tAx0}MhWLL+XR+XE_H>KU$Fe5qki8^6M>B4a`Rmi=Hjdwbv z&qrZ}JhAaam|6RV8C!YkI#`p-b>;6!r&c?lrD;u=!^|lEmlu$PT7qB$aA;Zu#2dQ4 zVJkwxDmEP>L<#sLj&i#T>;R<9HYWGjHTftK0yrW$^Pi5jykh5IUbF#X8~L&7k8hZZ z8eUy@5=_#05I}bJiX|~_8mx|!)A=~R*7}PZe#=j*zZRuO&|9~)Qe}aR-EI@|UouEb z!@`HDhmP!p5KIPwt$OJzKqbW|6leW z#&mNw5qE8vK9659Fw_~r*O1`TZNeWcpB)9-NI(@fm}IvjWr) z-#cGf)8O1H3o^3ahzjQ1go)dwul%%ea{n$1ED#-l`Ak{c;D%+DSHNU@15wF3Y8&y$ z#&`D;60w|fP62g6tFO9+FBm{K+N+*=@pQvV`sxnM8!U;PKHu1HCS2?lOPl!yZ14en zHG{-(O4<7v#D1uK8_=sev{!$13KR4LLzE#LXGK%BV$CoMK)m?V^p!&TO3uYgu-FY_ z(CdY*0OYx=;AI9Cg0B`#Uty}AkTe=MEp1%0ZeJ1s;BFSd{cOSel~}qU+CcbpW*O_U z2ex(NryfaSiIEuT{KkH&u)J67VbNFVf*=UYVoA(O0TJU71$*4OVMyD)B0^eSDB?qg z%+r_Aoi$|PYY2g8em0S}H@+zGh>Hv0=&T&RyfuRFl;E=|YU5134cK@L3|AC_ zp@O>OwScFEvZU1|rrTR?Pn(8OUTqlE0^f|YbLcfIs*B8}^wm|MHjD_cs*PJ_&l_fh zp$0-Vuz*sM&-gXgA`pDiDh{(SPf{I+iEY>v+M=`kl1yK+1{O@W&}(s=RS9e=;R0A4 zY3Vv?N=qOvFY^4_hK1~TrzmMP7Sj*Nv|><1y2NKkWh1c{fNwW2Nf6 z_{<_+N09-~I_r!2>5ZTIH(|@O^Ym0~fv`okPS8|zaK~kh4gdO!e?|yIh0_KQ67mbz zAUebxW<9>)!~H*)5a2GjaozR5@2*|9{ztR7*UtaZ;eT{k?}qpvKm3m${>Kmh?wMWi{wEjz|Cfu%@YOmwk65|m)~m&L!hy3M-Zyb`*UvW?F>hV>eG@l!{jBT0s)&HS zDSRKSg58b$a5)KlLyMrHc=i~^%<6rT-0!=7{(SG+pC?|kGzg%bspXl?GCDq#0l1c832A{vo8spF{{rAg3 zZ!+6UlL&G#yJlp~ds&SvQsT`x5?oEHJG7P^!Kab;$0|7;F1Ond*}AcVNrgPk$~XA- zt&S&rmQ6Jdt}XG+?vldUU6X}xZs5yFbuMtc82;XApBd9SQSkivgUf~C))BAgn?3e( zl5zEI#l(P)GCv$@t`_=nU7VMv?(Wc7-9PN7i;Ifs&0c<+_`-7pjv2@(ZUenH>t+yTbOZ&dcg#%pOcdzTr;^=$GmZ;@T z4Jao^6s@%0dYR|fwj(!Ge;k)wvG{7y)_*^;zPq)M*rFy2o(W8(4~lje!Ya*irPWhW zisdNbn+!fpqidWN1+Xe5O>VfsqHT)keor}#2fkl#(NbZE9Xxz$;MSy?BU(Y?BVsvJ5mnK}C!1phmyS5w!z3g4h;FPJ^GID5SP^x|~v zfPdWCJSAu`&<3rb4Z_lXuFxQW?x#W1`lGl25o?tYjHC8J4+9k>dcb+-UwY%`M&7Y# zy&QOTTQ$ziN{i50-ZUZBuNjhlj+eBvX1w^t*|ETpv52kE3jt3MRz1kI)1Bc>S{$dU zYy#OoA8kA^kUYABPb(Z2>edXk;)dAZyAYF1o8z42b!e}*(70VY98UO@`pE0Fy`DfB z%j*_`?|H!>(4iinNgz4TJ%#6{vNG9w%Z=_S;j<|kodVJ?5n?C|ul$cB-QQ779LrJt z1I~IO`tB7tmR6-LHVOGX5{(!z?JXBf5HL&Ilds$em3M|~P&E7AJ+XUK;Cjl4oRXdI zCyC0_w}7PGVIQfyrYD}4ge-^{%AxJm++3AC%+UCoGeWU(?^2bsl5j2hkJ<-DnG!)JGLEqt&W?KN>a?d_^w&aMgPvoB z%*aJkrfIJ))`kz!fcGD1aQHn9AUFZ6eb&4pf6$_3FClB;Lb-APWi~#yE{__3(TLZb z4yb1GMyGsFBlst{mwI#ZmFVe>aKDk^gp%9Ifo>$PVf)Ziivv0oe@i+#Mr!&(zswBI zO^VwkJi+~WIU>dwl{b(00+8_X%{4CABu_KtH=A*X@x>}u9e*C6WAX+1J=VX?dRR!_Ee#Cuo?#2O7=I>~(M;ruFf zRMsNgDxE3tNOs=tZRwLM=vsW*HJ&|FWWndkDXl0d>BR?hXEmjiM2`_`_PZQZoFYue zXAL*N7y4u@!Z~sCQoTXa`>)qLFDLUR#L2Y_m|6b`p2~sRT#7Q#E|)75lxuocv|O~b z2%OUc$cvvhnn!%SsPp}G={f$4SmfE@)PU*}40%J)NUnlo{wh4ez4}451?aGjK zjo6%%)>QauaJp|wAlR6@x_U^mzFg(ju>hL0LV3fj=|Jwu{yb}BJj>v^wmpHiFr_g# zPh{zKO+5|(6lbG%puN?en994e2IazwJsog*WNt0N72ZCb<~tM`7tCbmgzn8XMXx3S z85}4fPWly=qvN#$qdMhWf)e&_uNjvxxGu$+LWxECkNI|u57Kqurz(?n)PV%kGNyf3?qG~xK>gy$WE{%!xB^5zH1b5BO=Shcr= zRqjReK&eM!vGhS?vh0}Xyy`1CoJjc>Dj`IF2h}DG1g24V;D&$CV#{7?qK{_pvL#R3 z(YyxE9n1-9rUmOJN?1fy%F8wSKQ;=lU<&s2QZ<^mwc3U|tD_IfaC@h(t@fLSrhg5C z70I`?DYZ!~_>iVp&&3U~Ny8_iE+<#gH}|WDD3Z|8vzz?eMl;*WNkPei>zp7;@Rsa_ zY!=aPsWdAeryVpTPgh?FPIrbkz0t#UmG-J&tJqxs*u9Md0nyaGjoJu8RW9KGe089! z*A%HeZ9lZFma_zvSVW$P@C>qzxaEWvIeML$T(Y+fzT{LN^oX5wOcWu8V?_`q7`~Kp z1Fp6;@;-u4UE`u~OxZp?bVN_7R>>+I@AD9UVrO^)*o@W8wN zTJ`r*htT^5b2w+y-kigcrSDEAgRYn{YCGsu62D^cz23!K9o``Fa(hjTPL}BYM%Fn- zYFD(C6Jlt!CWPw$yb;blYs+dM_z5vw-M^A8^g55YkBI`+6mM}mwWj!)XFy|yrh@sP zVwM(~X=sS#2B0k%4)nx1(~NX3oHgiTgH)>RL!Ztd6QdI5AE_`598IIp$pO9=Xw$be zBm@jqR*8j7&5FZdKLWT&)_6S}(bik67NaCz6TSVVS3i-06=d78YQ1(id>?;Xsb4UX z>t?sJH!JOd{II37nJt#Z>sgMAE4MUltsb_Kct~RW#S-PI-BN zPD2++arU97RHxJTVI(SgcHZZ`E>7l5c%{OpOBvR&LRxdWk~!b?64>9G21f87n+;gS z_(;XXoTr2ed#Ol(wWohsD5WanFz`M+ljdO5n(l zo`obbDg`LCO{sEVua!$9Ee{AWy6F~OE}3-DL*A6(KuQV>PV&u}RM5c7osOuCjp3V` z)bzr077pbJ3+IVXvPO$m0?PB6(z1suR#qKb2@2u3Qu-e0pmH(WY3tTUnD6?|6Uc^-+4}s%x@u+PrhiE3v9PU4Xbe9rpz|d@iP-IHv%y7|sea zBHQI17#}ttENlv69c`mWXm-N3ywvN7Ib7ngij4kwaBAtUzd8tV%fkKPK-41 zy~oTNlLsI0Fbg&a-u1hsX1f1M__z3JrwQR_+!}t0RX3I6>ARQ$FNAC(DGOG2hir_s z51Wb6m$V2pj|Jn(dS@;Ayf3yRXwwW%X`GH+BP*=?X?8*}HldClBQxrtQ6U$|*DuoV z=HJq0j1*heB%u+^?1b6=yB2+g9_Dqq+>j(7Mz6zIGQ3N4lwsT^*qd7=cqSy*gHE3 zrs)+RJ8jq{cp$6qseDcK2bgtG7@m3?#Y@i%iTCuYkLf0VJw)=A=88Nk7LKr*I^B=G`yTI;MORmuDrF9x#CzjwOr5TmUgE+50({!4$DDT*! zr&Phk+Jagug-5iWmI&UGK0hQaKHpWr3q2@@G+a^a)z88i78$ojDUt`mGp0ID(p|<@ z93E1f1HAO|s8SbBHg(q+>x9Lc6WlMZYD^@d^HY*gtXi4zn`AxhbOKp_kc&%j2BN59 z>lkR;_b2D2&c)BgX#O4PO3D>6?w3%TUxdCT!f8EE|0*SU9eGnNq+i6M?rANYd5L_7 zmL9u`o~k3rLx>^GHND^z+t6#+sxdXijQ;G*02_!3)+s2$n6_g1#N@-K#EG}x>g52< zBV*pDfmE{n&XIL)+h2&-xOw|ZJBNwOjS3OJdbW*bw<*@$W>sSwfvs&^8}2yQZYqWg zDSFw3551i;nC*2kRu1W9VvJ4N4n|ier%Z=t4Q3bE$I6du`BL`8+pb7BN)Er0u;L}9 zSvniH7&Bs)O$pD-q%wHb+=9{5##sNRkx*nEx6Qm_aBfOrqoXF?|+o()$A98L#_6?nUgzqQB*^`O-!z55fQP6T@Gd~ zXO9T_dNe+qU{{r=g8PV^OJnlrc0F2txXC8i3ok&C>WlXya_tM-xVQFBzUo2?%}Hz& zhoy`Zd5@a846mL2axO-T9P24_zlwnJT!uCf3`9zh?Fw-q&2MJOOdLMqnC!e0({#0e z+?9LAxaF)ynz|k{c@)pSmza@;9&#Vd${|_g=_9=dN3}P>hO8i-Kwd&#xR`LxJkUR5 z)ZsQhurg-q;SQ&0rx$~BRZPF=naq&s@yo!{?P(4=ZbC^0rb*B-lQ@Vh5eI7fbM&!E z5N9Mr^jn4#1_N5kCEyfCrD{zyo|)a&C76s|$^%>sJ^tJREW0$9v7$R$Wi;_jjyv`& z+^?k=q>h@dX~{?7OFPB?@=>auUaZ+WJb8gR^AOtun>z|E;(jONVr!g9scC4c{bIuO zX%_rkicd(HFIJsiwh%PyXgBk`O@ca~Kqe}ACCkcaPF~g2Jus1^xDw6ljI@CtDX#sHR^u^n*RJ9ssV65>2gf75LFP?lktReD79@xwEnLnt;{}`K|V9y13;QD;2X1 znx}jz8WKb$tr{-EB#u?vR4EmNq$Wu%0+Hxe(&OJZ2e*=LnSOAOfoxD_2=}Y~p5uun zHO;#M@8GX1^n3MV^3l2a-Fo%5@2D`or)Ii5XCPaTQn+$5=wL5Ad+ZU|)umA-_qyCl zk|Gn2FbOP8JvK!WExbw5@T)a!vSi#UDcvD6Gp2`g#Z{~@J9VbBhLijQL5Y|_e+@Hi zn0*B#-QR}P=8M0n&3NOW8rvrHY*f)x<;&i|LjawF%=Nf_5HKC;(46x1=`|wgxU+3) z`AJC*-y^+E=1svW^w~+N-eY!CzQmFR|9LE)eHR{vZD9M&Zn;cgm7t00G)u{sB|1USA8pgn)iI zIqTw7YOfO!FYvFrZP;{n^z^{MmiV9rDInB+F7BidgZSfw!KA%-yOrQR-}o>et@}lx z(xR{zK7dKgx0NOh4>do0oH;pRNDgiv=_h=PXRdew5{Rs^$Rt12L2YFcaEN<_d%S<7UN3m@dlWu3#msTQy@BLfc== zo~D$m)5?0y;zxXyr!;4J!O@-^brg)$`VRZ=2+6E5h1L_jmDD_Y?4% zE;sU%BMkK`ag_erxUI1I@#OM>XCYcHVgsHXxC2O!@YHp;@t?5?oLbT*!)dZdy6}0{ ziqF4tEZS=5V1sbC)QlExoo zFTtO^ygh_xKUmH?A}nvS-%(!;@S-Hx8EZr`_Wud4zm%yxEj&tOET^@ML5+= zW?309{Vb?UMY(aLCr{H+1RamDJ)sV_fu-(kHz>m&;#O()`?0jaf&j)TB~EK9l1qVe zRkso-&&l%V+T{1rQ0)ScTK4Di^QP8f`E^~2>ollTUR$49NfRzx^x!%IgocyXe%!<- z4^L~}Q<+J-&ARJe=k6ZL-ti)X+u2-&0`yw^=cxW-b?nI?T=J-`hzLygmOWmYt9vd5 zT^%rl-2-R5W%6#yH0N)24B&+-{(kQGg8gLK9F zArS^|IX|liH&f1y5uQ_8POYRkXU^#1je%OxRP>IInBG5HS8mbI{H6)%Z4sF8iSz#U znjA3awJ9>pkmst&4}@7u-@aFuQ!|;JZm;{Uux8v>F0&le+C|*wE+?x5g}=1~DuAi$ z*PY9b*w!`+s*MN$#Eqh-m<}BaL(Ai{oIZ|J%3M)w1ipGfOh=&%>K|vVoqypkT-8S z@N0rE-7cz%^?@49p776Lz{3X=Xa5DZR7Ugw0LIrxLsLVCJ&RM60?+cZJR|d)fk|0S zZF0<@&KJ(caDotR&*OXuSuKsWO+$JeoN_v&}RI&r|F@YS_dtOk-D+i4Qz4 z-8A&F2OV*;;~-ldrTQCy2f~EJL5mr+d3$-iymx#E^l!xNY+Uku*g!33#^uT| zOU9U)mk@d>-qSoB&>KO^8S2) z!)ld*6um+JVZ)W-(G<&n#hZ5wkTVV3I@8X!8LDH@NapLopCP97+d?TJ!6mU z?qt9MFJ7U?t`$7E<_45moYt_2!wE^Rtyh9{Yf-tp50HS*|+Vo23)}8uXcxZ zJJNl7%?Lx5T-#QeCOFsKBwm8)|KSia(s}UySLfa$@82kKu&rE4UR<@r(6q()Ju6UZ z-ig6-OLK4pFL-w9*Z4Ghk<7qfA3wM~hxIL;ofqvOS)cZR_gzw#fN zSI&2LabC$5T-}y=n~W$`3XS1ApeMIyAXE9Mz&Lhwgj$su&YdX?jGP%W zL$Yk%GsI%8+FWXQ_x!44u#6LXh%yUF!~<`;rifX`!RGQq^7eNQ44YTr0%$Jm!QSwM zY#d7>d(ulIEub>DDY-;MBUzmmke6;Bdd+t&Oah*CEs}=qaeO{KE8<*vEdbbIquIbx zqErKWX^R0aP;cI;rtGTM$?n_9&Xv5&mxaE6oyT_-*36)v&2c|c=)RGe-fV@rn zh(}`!w`SHT$-~d2V`6i(z(QDR$0id?GLi*>_LZDHx+9&{_V2kZXAY_XwAU$Ni%WQUrM((#NrjFBd~ zGUDP6@1=7XWAD5=+D|b2_whI-dupTqqP`Jnk#kBfYcO*;+w(pR;GL0)j7d~QE=5hw zO>#K~(YmvTk$wQlsdJ(n6}0d$nWVzH5QXpndZ|W5lJukE%z}%(-R70J4*Bt^q3@v7@6SG-dr367ddi->t%NWT6Y6hqvUh zf?Pcrhz!Zh2j}TA=6|x~1j~Q&rn-TL`+U+U_$x$6B|zhE>wKx{wbEV!1w zKv!ScuLB~M+KwPD#8a!X1V?X62B($~Q0`uRdd&#S>npRZ@CMghJCW?k%!Kj_CWImX zLVZ6|MCiI^He(oH5F75IOGzHTRxAuBpAxs_YCfW;PWEA^tOmqeXv$@W~Z^{7qxo3Wl$U`MIuY4yn4 zr`xBX#j7h}eE_Rn;%DxM_+-_T#^`em<0{K)FQIZ+-Mz+oWNP`VH1vJ_q6;9lBDjq0{0&| z1$bUSWb9y8Fcv{h_aS(N*SYt3=aarExWB<9uOL!kkZrbCGY>LZ=#pyZk;dk7MR2o0%>25Vi!e#aJZ%>j^!SSO2Y#{*jf%5d*)vkbI|qeu_mgWD z?tw@vD>H3$fnx>yE}Foh){#9jni&i*zH~Yta$Ht4kygV4)cmnCuhl3S41qYSbj^1e z5XZM=OqRxl_6G-?nz>6~h>?ax^2BY8vd1z<7B5I^$sx`AT*_wjVjZaO72KoFdBKP| zX_@!EY0bjI*2^XUrjp$#%53;^@ngJ0_Iaf3!OzWg%1BOA7K7enVb}^Jole!BR+sJ5 z5p=T6x2)*1LnCtYt5pK{%J6f9qpXy0Qf`GjTlIM*JoJf1)BTBI!zMk|5tJ2@0D~=! zmYjaxQ}g-CImOGs%O0eXG=Zdu&pW;kH8daX2h!)WJfB^;GA;~Mbd8rnU3{M9IYwxA6ek=vj=m^?1A}H{VggvCAuz$syi{ETh&_)1%MV} zYJtx)iweREave3gtyP{NW+R zeDJ}+3c$>~kl!74vNtOlAOeC~`%j2gqRNeW_5`NpT$nW9@g&H22nhY%H)q+_;}p{s zlKGv+%h~=s;ko;bUwH~ee*a^IX3M^zh2T%PIgm}}OTDB=Q~e*Q08eKtVbKArl`#q; zS8MO+-(%iY*pxVD)kasY-3ld_$Sx{HQEX4fC^$ln?3$ig;5^vRa0E`r{iC{pde|`T zZv^g@UN?g9{0G}0E6sRuJ3Hp(j)xZS4su2H7j9HLE@2)y7#LnD$_b`OxV#8LwKj>e?bV3vCAc>U%7Q8r1|I0W-i~7upm5F*A_I?rGqvW<5k^6T62|A z6R!|(BuHU;q%f@MU5H@MxQBGXJ)a~64oEOG79FnwmtgdEU^<+(ql7PMMF7zBH3w2N z_4SLpx~F9ZWvGO0X~T=u9XsRFSKr8(_hv>b0f)xUmZfRc=_H5DIBc+$hWV~z2zA}k*w+_*5VXCBrlmf8!qWa-%rnrP*sOT8sNW`v^@+Q ztCdOKc`BI~Eldb83pJGqY`YI6TeXc|1o5Z&s0?%seKn;tp>we2x%FN(AKX0&J*kpq>q@n=U7Vg|N;izo|yb`2gvtR#oADXF^HRd<_S*Gouq&aGsB=8BY3wIe;G zhC_ccAc+&Vs~oqXYP+TT{p-=eWO_F9ywuV+)z36RaAw#yy-3f@7o5D+6aZCfxBZdu^hsBR=d3Q_M`K&8#4+%*+*SIT=@iTgQNrl3 z$M0$|{dzypWN)w9B8y!f0g$Gv_c=OMTPR+CkxW=ak4zSGLvm zzchVk9Vv-$|LW!ZkX-jJ;(gt3%8k|~?BF)A?t)B$G?yU|_YQZOPfe~y;kZe1%|m4N&u_>J^94X+*cZMIAE_Rx|(;% zxgc3YNJ9F{*7jYE)-$38cchVR2JKcqv%W*_Xsqs=FIwr?>*K@by}X+>jaUB3bf`6AoIEpBAe%Urvje}BL zJjVXhGuJa|FIrNZr$L=hEiTH9=o;_%a0-m}lfqC33=D=xU5v#hiq7Sjy6Y>9Ba{7$ zMd^DXtJue$Yr)qut;s5DgY!kg4#(fF$(uJH`2I9T)dKoXcvr%_wlzpqqIHzUhB7_-Lm#Xl`g?== z9f|@$n%mMP>gK0i*`5#1O__HoW+5p|jB;)N3)8iGxL&`%yFce)B@phjAE|JLN={?E zmSHWYG&&ADQs)So%((*;zJhrjbBf?FT6#2Sl3uf@Asq2rE$#c&3ag~q3wIpV#p(wk z8!a)+6&6Q|2Q6jv5I1bmp3flp49=^8a26q?pNR7OOREhtGmJ3^^}ee_=(z&8vbO)# zdx@xNJwph?#D5mkIyyGN3LX10?H7%x&Y80R$KIERHGM5%+v{yt%MvdJ1*!xmzTf~=9Sgf$TY1PCDlS%8r6oq)Ez z_tM_)dA>it`j1LT&hN~bdFP!uGZQIel=b+Duqofd5)-mCjj&S0Ppg}USn{ZMtSyK# zQQD)gYc|*!14!uPQ|tt!6=~G^T~DiOnSGG;-eKe*b(Q+jii;lhoK6T8r$UZlh} z$D#&r-plTjX4R)wI95y7_G4DArmHXAr{QsJz` z<{<8#Vsrowq?GT_Po?Vy+m1AHv>~H)crjS4(zMV#%sq%uqyx&{BLXBHRTPjkD6`$SNT&vulNXhC5m~Qg zdeX|(5Xjj1tmVS08%TJ}W(ugz3O)eOZXvY2wxbkM29V{-Q)X^O&V`L7uSHS5Jv!rc zB@D@#GoH>g`bq}%<6STKl`*3!GbW6Yv)%Y4#fCGK52hqgm@}xBaZb@L*Bf}^dbx2w zpK8~{OtXVPWYqW1Y=f~?FKHtA|KF--W{1NS6yRMcWJYtK_xdASje?EBDJ;)rC zVI-skYjQ;4vshu+7X6T{&$-77!hPXa7V%L}Vq5n>zyFzDHK9{}vr$PEdl3u-tO!PZ ztO$D=XDzr9PtpnEIuPgupuki|6oSIGk&m@`ae^X!Yne4PO<+k8eIvf7w@!&nAt47f zxu6!;tx2^xMfw>of7q~a9T?RTvs^4aNSDVqn9nmN{xnx$oW14Y*0CQF7uI4%=`e=b zu8@DOJsb&`%XmP!GUz@CiZ(A~Cw>ew5}v|m_Bfs`Y%D343*Y?|xWAhp`^6e>V1aaeGt%@v({ZrI>OtwpVfR$s3e5r4;z zcGZ>HeLi!d)v(6rDz}_`21duHg1V%zQe!4G{y$71qHou`Ys&{}+}DIx@X6@gx3gUp zKt+v3M_7ZtzC)EL;KQqg_fi$l2 zxkq79M87XgLO(ZYrj)zgDZ9B_6wJZ0VK5Tb_a<$I5TtNoeU|CsNC9CP`M2Y2P`CJG zmxv6|YPOAfh0lkr7yslnSbd_5_wF@H&9TzP^pK;+N2uTO3LvCtqO!uzYe045J_|}EO~TXKT~Kan%zF{rVLvB(q$d^6 z@zifQF@N=GnDd*fQ0#^8eNX7{v^Cc|JDzN#=$e9G2NZXxZ+v_pusy+w9yrcUkU5;{ z3c^5OBZqMDABCs@6R9SSJkUEvQ}F}^2sw6OX9zAOH@G5v?iuY>$RbHlJUT0$o`8zQ}aU%H+F>R;g3y&K1 z#3dyLX4L!>^Z=nF9Bv~TqXXcH0R#Uxf^?jwwm!!qPZW&)uXQ-$4}(9b2`NV+b-=m- zYtYcxShg~}GU9j&ocFmNtWJyZpD2j+F`QvDNY>ptG{;`7?V*i<`fM@6b$5_n{BvVF z`%@`Qsh0Q;&2eBh8m|Rr6H-^Z$9>>!yd>x)dvRd;KaR-~cYZn9*`Omh+2?;L#`0lT zG|%Y0Z@Bs-;@jJ?@LC8mPatt=A%VoDMS#T1#KowCn&e;qL@&}&lH=t|63UL`KecN? z*=^^jbCn-Hwl*L%xXNoeF2?B zylqen^$jT+XKiA`8nBIJhvJp$UEu;W`t5=K_}lBt{Q2i83^~CEN4FhQ`q;@nvfAU_ zw9AJ!KEcPn;LN`oXlfauiz`tJXuviwiRnL}1?GRQ$1^-9_mei7i|2<206_vHD?KVb z!^wEfhrk;5#)gFVuo}|9iN4mr^sF8HkKg|NeSiP!ZwdTANI-O{#ECbLTeb&NY3uDc zuAg1#$`IT2z4gs;(6caFnhDA;tBnV{MC1Yc-qqK+FT|z;{7Ck;*o8CGz~}F)+#{9d z2gZCt(_Z+$)&#tZS^3YcpM-p7`-zCL=B`~$Kc2Z4l@EFv3({rZN&1b-!ISdR#Rs%VO)5W5&O~Qgpi>ymDrG>ycB>1pofXiTU7>|K7~sC-b+O z0c-l(X8uOA-$46sX#U&J{OvUVrZazYv%iVvuCsr$=D*{azk{>CBhBqd|Nrl>xqRWn z#H{pH7pY%#iyS6J3N?2fdQGV@^bpo8{5|KfQ)c7${KZ%oI)YUKCPs0-#(tL}cQ!Tb zt`;T~lk$%{pyMwyZzjRt4=NKt5#;w}w6 zsluHSFW;x%*nHvQeQ!kf{zoAZp?T{6ollwky8rXEweKR#;WyV-z8dUViiPO7n)|C` z*w2S&ATP?MeMff4wfsHJ+G zKA~AV>>|vghFVw4sGBP1PcDvBBunT7b;KK06=EDDlNJWt5h649b%xL`(DA=H7<4fDkdwX}0rW1EcS^W@&sd-W-BoT*pq2&~t%&L6+ z_Ol)3H(xY;!zv#%6_RcX`@hXX6sJh3c0y;tYPLOzM5`5D4D%z7#tZMMik+7U@;L7c z1JgCqwC!}F?ja4#QQbM2ZJuqbR-l~aqmXa@M5vmnmhOHBY8a_4lFtwoDzr30E@5FV zdxTQrnc@1b+D{v8w0~FmnRt+a)11o6_${Bem9@K2x-Rrty(6quZi6?St8u4C*G8*A zhZ@j$s(V+6`CqPB$S&>~-VeqNNZat$x^cIE?vX7F{JGpc;}?r!xCUnB?&BHd7rC9e zAX)g|W+*aey<5r7dUV90w7V^3Gtciq@}?X`eLyjP%uwgy?oPKV8ZPt#=T#K9ln#~Rg?1}MvJiT+Yshg;%|H&8DK`e#_AQo`*G4D6BkXQvQ^E(*MzJ_RBc?^4Rq2kJGpkZjKxK73PEm~S0e&{rK*eEHIww$YaHKR13ud4=!{BtqyB}oWU@vy&UBJ_ zdOsaXjNkYkzUzDF+WdF_ofBd8`<*HwOr4s4)h;>nJ*0wSN|z`8PeFa-)0s<%Ig|iC z8TtI5|Kk8ylfb55d2`wMOg8ec7n_Tt6ilj)k-BBxN#)I>ndymHNw zOP#JE|CDo&jBvkl-)-^wNOfH6`HmW`Fo-dKu-yH!(VA&_ZxL-ZsLWAM%F2iM7l#$$ zOg$?!&k4z&z+bY9 zm4|M$buXEXyNRqkR-imvl5*(8C=ZhRn0X$V1Tfpp;V2kK=ajNhmThs_h;fQy+sO7V zb+2u#p`>9bioU(9`k>8YeobmQ5BE&p+n|??=Z;B;a2Jpg)|Vrhzl;v0)wS1jRYxOfAtvt6~lj@g;YILHgg0?;0&vA)zhqQfTunhV@Y%kQ(-q|Er(ErO4 z3d~ZeA9~>I5a3-x7I$lq0F?_9;(qWQZYMq%Ra*ojeI1+QsyELGYglcsM<5veAsllG zej?6JyG>oSRfTIfohCZ6IV;w7%vKv$G=5J*C=GABx50VYvh9k=-szG!-J933wu;e8uAw|@;8#T z7#3~gU?9Z2SD-gx{4Qi&sjx@cMA4jbAuz^&!FSPE+AP$Y(~t(wKDHo(arp}j#1#Rv zW;H^(8@!pp>Okepj2h+dC&PlaM&P2I;b@kt|)3fe~vFj&Y{%LOA^L~$SC{6r;#H!sE;cx-Gh!G z`R%WjaU9+%@kXYKw3V~5`k^A_GZh?!?^0E6YwDuTv+#_KDVtF#)xm(O#(C|i`qxs< z4(mhSd~2tWY2_q#B}rwAq>mZO}|C2Sr@ngVZ-At(l5uBeamV z8=IQ_*Hs5qbQ3_sB6t!XA4F`trfS?Y4DS8wVvsiuLkx7V5EMs-wtB*gao0id!cUEz z-oyXRxqjme&0h;R19?<}mtC{zTa5Q?e-dBpZ`ik6l~g_KJqAPQRPJHja~YFfe6r41 z>wCp+l7mNBksAVH=4LMP$6}y;u(p)`MUShRFAa?w3rAtRTJp56$hf*OH<>8g4Ypvy zN@Nb3#yS`Bl^Pfv(H8mi7+4X#L5ZyuW!$=soj*OkI6tPm7#Q1Zi74Q@PxWW)Kzi?3 z7e>6+Z4p@>dGY!Vx;QRbPuTd7&@y#g-6E{Dp}KzL6EFhSc5p_3@$lV^HAUj; zBMTxwDFRnRt>F!lh`uJEoK}MEMP!Q;&pWn^Pp5hPIi!1#k3?x+!;$++N3S4Tlyr0E z85w@7x$_PL=Ec}hhOZJ2vJnv&x&8}=QofnR^qJQxrY?`u1G}Xzo0jZEYJyWoeFro%cVhy2S`EC7I+5K*WPNNj%WC^La!M3OyLut9I?f<{Iee>s}gFInLRa8%~3)^UMy*OcVS%Uf{k&r=P}UHI`s_G+#DZJ;Zepj z3IL|jmX?Uts#kTrw1xp?>X3fhx?DFnj*z$gp#$XkvA$PA#?9n)ZwtU%LAQ-FPjM|5bzL*4bRJexo3~~gwvP4%dL$0NoKX%YGzDgX z-kgepIYY5NTjIv>RKSinqHeD{+qWPS=ygQWYu0UPlF%Bvxy03>d?fj28U2e|`V;%;k#YZB7Xhis>L^yJe~2LudiJbgvAs?losOaiBjBb5<9mX>%|pSm z=lxAU$1*_FGAwxlGGgEUMI#H!E%P)d1iUVapuw;dxCPy!*`pc<2GD`=YmQ!Xv{Ud{ zd*ZYHioPaeR{cv8jy|K;~+y66u)B5XW!Ci(#be zw;A>MkmF&-4CS;S>rgf<7893Q;F!iPSy4dh6nXT8TgIxIG?v9qoA4C6ft45T2tq5R zx-ZfzyuGhvLD*%@#u-7z!v(#*!46SXYB}Z+Z&OBp$7R?wHyR+&nWrB zm*+3^P2W|S3o1tYwLj8cmuhGRv*&~q#i$t5SWOeP!?fIG_kWlvVFLex25>(CVVHje z8pb4I$=AUo6TsJ8$LAf)`Cm`>-F{uPgn$J!wY37>mNJdR%w>5t2b#=-ns7zd`>0~)<-QOoYA0MUuXeFTLsnTZea?kCgMLd(V6&jhQVR@h$!hq7JmL(-4ow?&F7<>e24@@FjtC>0 z8Z&I5VjmKlMg;+`4VWv%vYvzkN;YV8Jnf$>viFQ;$QKPt=U zPvEzS8g${#oCw!bv*zYCFoh;85rQ`9h-;X#gMqQp5=n&J=`sG@MV?s#sr4Uoq~znq z)#m_Bs(}oc0bN!!r)DWy$2uhE(=8lERdk;dMJK+Ts`c5K7l`Du|8cI#hb5sq4Pu=@ zEVEi)+H*df&@Dt|+JfGa;+|vOQHLcwHQkyp2@{7csVN0+MYPA|RxBvPg0~Rp-8X)> zmk-8XGD72@`4(_@Q>9=M)ItRWA2G?XGR@u59Fy&R2g=mN!m>xs4Ec8z##smo zl&uunzyN@Aj%7y4GV+k;Z8kVdUOsKNJ3+yvM2aqsYQUu7vauhnm0`=h!OW9(MZ$D( z1Ju2C&K5MkOAmR*y!-S|H7d#+{*f3IvmT3D-rY%VU~8$Z5AkGcJAjgCrKg}9Zf8y3 z*koZ;;3vgMTW#Mp#heB&htZP7TuU7jFhTWMFF(wYBAySl0Thv!7q5VRp2G59GX2r7 zl;?s5{7R#BXaWBHuWQzzj!^;rZwbDPCp`)+n%A7osuT;?Pv!CuFz3?7F*7K$=vQTn|iYoFt zcT*jbQ>2DGn_z;rK>+>v3onjADG;t!L65ehltHqI1Z9j1)mVca00U$soFg6yOwVKE zuMjKrWX++3iNF%X{S@h09uHX@ucB#uSa&@9&P*s2p(HncSYu0r%0_4zUJjgcTG9;2 zGTbU`1>+XX{cdE?eJENY$TWDk57r}mF3d8+TT6Au9I%k_Jlbx!L#_q2N`G_C!!l#6 z+b`k}Xbc;!ixXK{K-bD^pAG9tAE|R~1QRZ$>R4-w3V$w``7eCj01UtXyE!9`b*$Iq$`T+u+G@7eIsz zx#nz81%`}0XZjAnzh%%;2)l_|@v%#~UBK@Xcy}S?7O|S+hleQP9x-!^7lBo^){D?D zgZWWXLlmtvXJX~DyPtXNFU-Pt97fd(*VnAF zEsQ{s{lP!6EJ2#`eM~#C71P3-Zij(#`UZ_VrYOtg<=ugM2@6M!bxYtU1%p&PVze(4 z*KUUhrL8-1`UTMUt!Jry4g5+2ke9L0*M9VQ7Mwo z{Eey{^SJ)`W@~D)k1j#~h_*4^uEZfK)m= zc=Y}mcRxa_D02pb5T^*nxk|4jrw+DDfGaFhLma82}XrQ4W?4H z6_!8sf<$D(1@e2bjCsO^=c3a|Fp%@neWN{Np;q4hifZ9}HCTvn`%tGZiLZa8v1@w% zv}14)8)(G_T_%So;oQ`brClJ9n6+4sMTh*(ywdM<`TY3vy8J!f0Bp+3^@M$YRx=T^DePHG!S;My zE|2m2b0+RPuO3GKIL{Un%WvRV%@4_IUuy%gr}Zr3iBp}nYO_Bx%OXQ+R*Z4qf@%dQ z4ghd#d*=-+BjE*v2@c~M8@0%YZH9*7SnxUiMaCnu;X1iPD6?4Q@xuV@6c7xyAB$7W zkSs6;;SnIf!AN#`tTNQ4=di!W8BEH=h5W202yv!WFf^>=stGXF|ZxEBODq|q&V8^=*OlGcKU?rkQ4TubzB zRkljw>k3~^%v%3yPp`kJd)UuzO~*{L?Ex{h${%+0j$N_`sKim_ky=$r{${KB9D7{) zq6jl?EhW~|Z|S{ViCIy!1{hM{>kSrv0b|HP-cInb{^ej~RJC$zNmsqIY3y{`xvatz zi|=+!;YyB{h9tX!?+>s`m21=vj0MzoOEdt0x6vr9ZMAGiydGCOYP_{dI%xwfAWQCk zX%^1pN@D#lIe(b}XN_w8ZbH53Bc^mSeq~&0fj*!POlE~^QR67|>U7_fAl<-t3cl5V zc@ZC?e}7 zTR^?G$Kt=a&>vvtceS}}n%yD~+UnoROb!0e#N&H(nIOLtOmeh&pFYdv?irAa5(_Hc z1afh_Jx6_dXpyjSD_shWX(kfL!`xu^Aa#52bt)Yntcf$Wzqz9s8mk5TOTb7=NV!w5Oi}DSWux zMj;BrZ+(}V3*js*v`@1x#qY6}nNB#@Gt;aay>^siz(~}ym2EvjPk3Ee?U$E+tV1JR zhZ`IgKIhLgo1eC?aD4Da+{WVQal@qJ3is^CFMTIE{UCvhhXc>XGP_HANI-_VL82^Ml*Fcf>0HI;<1?VfR%$ zo>gl-a~Ri0QDD?7RGhf#sM(N%i=!L;IQSxCpA66gcd@Z~xo=udn*ErEnTOo@bm@!l z`!Uf`!qt`!0u$6sO0y?24Ms1A$qwpu$YOOUqV7?DstE)3up{2ZP<(}3Uf^XZ*;JeT z?&-|+f26G~9sorvSC1Z%aDNBWpyC515 zCr(#LnVUxEh?SpnhV&k5u{0RI9A+W5$;92Gmm;f;ONclQ*GPFJlX?k|ZpeA96= z#Zt#N-k9IEd87_Qx;SzJn-jli+|FIaX1loI3XoDpxqm5?Eltth7RsD{9I!|E za=AnIWg-8?x@}DNjZ|CzR&mkgsDGDCPbjyJYP4lCL&JqnXxhv2-bcNQhomFXl62&Z zl;o>-sWu|=g_rSrtRGk8%qC8p$_t<~gWY-B&YMNj_Cf?`qaA8bi1F_Oj#Y4k~#E_Ma4PV5IlfTa5t=p4zbK377H5{-XyI#2S7 z?eu<5-IQ5A^jg)jP1iO)g zsr7c79$?>%#B4+cdkm5#1{2X!PptH0OG*CnF7+IFdgUT1cK!}oQs%S zWA}_iwUXvmoOQVJ?imBr>sLpa^U&rm|782t&97lPqMscTQCj>;MDn|$2i5MPUjnzMmZe#TuYiD_rys=x<$G^VgEMzYQp{{!k zpysAcD-PO~yc9QRJU-Z!pc_5{z3>LQUq0QNZg}7S8@oeuCVAm|Mv8%nyo8>!%$JmC zZg$JU$hS3~1Bfm2Cr#$5WNDx|KHBAgno03=mmo1&ywhRqE7W;4(u!&T|* z<=CO(OW*A+mH!ebMN--?Wj;q$VwNu%gf%W-j#!cD)SF7+TUGa|tlWJ`%1H?NR5tZo zTT0X~SSUVA{lAPOxV51l23rs^4^FL6ljeU+D7YVZ-b;49>#@v8r#M+K9e8Bveo=IZk!MF8#yZ;w9w?C|ys9>- z3bQ>hnz9psNZ&KAqQxm~EFJk_V*fKj{VVsl!L z|Anu)&cEn^*>ngyMJA(T>51nLl?A;Oi1Lh4u>)O1W1=fKpZE@j7rz<`oC=SnSPUcB|Gb~3i zg)b(*DIhg}Oz3}cQ1%Y!2uh`T`lIE`*$%Ihq*l!R zZ2G;L_*5c|;Qt^>G`i^Ye(HCNfM)@ohM`^Ew1+PNf(NIhQY?BV^u!qbHTv4Av#1pN zOSr3+nvLq$+rG$Ll+5gJSsc0t+2bU?;;2K$LNY#hBG&rnd3)zr%xiv;IVbc1le5|F zp6}xLkAmXIA_+iEqRiBp{V*%Dm_P3gs?Iy|p8hW~oNd%#xu&SPjWRFbh+m-4b z!>st_W$Aob@8uPTtmXHd_xj}dSmOZea7A;qzof7t&k#h5NtNJn4g~Db`ou7XS-#Y3`&pV$6QyiogvoevL-C$--tw?gP!79!jCzD&gEu z_7eqPzty(DtZif`Rn-T}wCbEQHS7A4a`tvi=$!w?{^uzl1btt9Z(+V4Yj3gh1tC=7 z0(Ux7(hwPsqD}^!*yfb+eT~*05AbcPl4I*prm(70VqfSD&3;7OmW3P;un^eS6xCEpl#bBRs3Z&Fo~1ObY|TF2%dtG~`EKb+jUs}*ui*z`Vb?K; z9Z%$bW9a$T8}&zi@zPDH??z8JktD`=7oT%0>k*1D`chK)P@mL6&ztu<9xvI|4oYu9 z!`=sou@SV!72sK0;$o!XheJ~q9@a2hW$9--PW2v=$IO<=3>u6)f9CS2+=fiL(Ca*P zy;9!qq}V>n`AaTnOqFP}U(jCsQ@6;H4jI=W2)Rx=K$~Ytw;2KscWBNLj@`JzAI#aCO*VUJzVC_Z^MXqT(nm-tS!o@wB`m0+ z!a8QdoKAJkB)*g`{?(l6R`g*gT@;K9D14NDFesKSx`tQ!H7aGJ54E!A`U5G0`r~2F zuRDgD;}^`PEiy6c?tGq?Knk&! z;^{zt48E+`mt)nsIt1ii()Fdt6_isHC&4`rLrJ>2y)*~{`CXQir1-B4qbdI`_Y^#- zo56L{Ne^z7;P+ZEFRqncfkz9FQYg={vWC4)Yb$Hw$VVAVN2^Pwpd++j>ZN$Rh=^4D za|7C~krv=S!}G?D4!*pR>bEeTgM76zXwF?mO1Q0dqy}F2BdwAA#?bolX_Eh1QwDL7 zdp2$b2jIRy=eB3owK&9-+vZ@1{n1UZN^}u8ktot8b4de-N9*w}D_S5G(vr3v+4`=f zf{E%I9ToQc)wvV+U#Rlj<~5@;?=P^?o!g4E_xUII z87>^tvziojZwZ6nY0XWC;xv~&!z#Ny;C$4hHH zf~-r5aTs;hXw#Zq)4!O5C#?!dm+Z`u>7*h3$kLa}1k*Ws%Y7>NipC^ABiLBUE$K(D z8D?XnjqG>z0*Sl-Ch^r(_hrUn^HA>cLuz&={=%^y8lHd1==p)vp+g{HQay)WL^Jn> zWNyEqS=`wD==8xQ+jf=I^QcbXQ6q#$Bu>1v!9nA3=z-`YJtO`T++kV9Y2=;Y6C))T z)?*Pe0k&eE^SHe*I_s)iSV8sjP+#LE^i@mwHdjY7=ECVc=h@T)@OK~~B3D+(84eHK z4s|t4)d%m3QY&g#Q9mM2EGtZpw+bfTQ*kYTyR_AFe zu|q70`tk_+I7^l5!xQ>TledPqq3ZAx;N`2RwDho zZH1<#a%j`s10+toHUH-yCS*UTjl8rkvz#t41@f< z5bpWNB>Q z$A3l5uyXprJwN`1%j5Tsx(r{v>Ja|Q&DL$v0-5|Y?I!<$W?Y=UVc&30PR%;}ik&sH zSG@3LK{mQaQaTiU5e{gh{N_t>jc}Wy&~`HG(TOnEkv>r>q5h$==pt&CK3@mFwecuQ ze)BpN2@;>O!-v7z4aC1 zM$>1H*ws+n^iY|)^g0r-rV}LltRMa*57Uz{$;&!Ol+g1&%8J4^SUa`3G?U;S9kK7S zpX!fb$JH0w)Q9WY8y7ZrYm9HjGct3hUs0vdi+V(jR?qeLN1U^6J!aYm;&4qE*h?K% zPI4Vr<1_ehlaH=1Di;OC%fu~((e5=vQIhaAQfpS%IE{XcZKn0mSbZzgz#N$5b^7^G zsXqmDlenrR=RAI>i_)Zy>x$I7Qw6pMJ4<7r&g6*hMV78apON(4&U0ID%lnwABhgvm zj#r$7{jEmwJ)6?fCqi=}}caq>t za*jp8t5kn%sAGsi7IBpLZfG>*x?jmTd@c%`!|!t2)oHqwr+gy&>T<_9r=uj%s|X`B z*{>@z>(D#N@SG4Q-+!ITYuf7#lFCL7o_nf_f+0zSuZSA%O1;`h^rFiRG^df-Q_}j9 z^{UotjEcM4m%US<%xX3x$>w~$>S!sB;Y7qz^7!scYrm$Axw5dXk5elbhjnYENNpOp zULm~=f7en7oQH3%Cg{Jd+$J~fA7v3b7Vr}!w|9q8YsimAt>w9Ey_;=jNyibV_$hK* zQQey)b~Cfhgg}b!Z+RLy7nM3f+FL0y>h!|@UFam_mHe}WWilx$u3s6;te$Orp$1z&J1}ppid^fnKa?6rxE{D?A!+LW(S5p zFB1QjFuGeh`>1DvYDjd z?PEDtiXnZ`W;p&8@zk6Nx*@Ie0+!R)7GW4m`mXBqpnm(!3J$qJiR|iz;{8Fd1!-yd zvBmek(vHEseV?(Y56u&awzEqNxEPARv02vKB!b{uq#$M^3VwP$JgVfUraKClzVWTc zsYCnCg`=^2Z2A)Fm}y6XIj55PGq&C}y8e;X`lk*V+qQ$KCsB{e_q}w}(zD*^x)4P# zN#QZIA0Ib7@vcziJtgUHtdGsSPKads+>-7gg^%~5TSEY|ZbE5%e88L(sg_|NMzHt1 zPsv++nkQfGPw*MA$CreO)&vXx`Z{#>0p#1)NY~ghHecu8z2d)LiFi}FQ}0${o=wzn zyC!VgIgKTG#*_X(#;2IL>3G_q+)I$I66y~6b1Xkrz5L^k8v-~Cp4}P_-;G( z73rr2p%Mb0@w~42r1wL_5mo|r$)NFog=*sRbvFAT~X9sjWl;8G& zR0;CIzu#=sWHXY}BMc6b^*d?r;>+ce^DM$pZZbN_29W8s^YnQel%jg6%%EAvXo?h+ zEg9gZi1aGU_bu9v6PeRtsOY?_7BeFoQp)c)Q9;ZzNXq+YL*C>)2(Wy7(4;%;R$<^A zS{4_q+XRO(%$Y`d9VQ-hF#+L%@RO#wMOu9f8LotwI@CWKeK2!zfm`7;ja=6}-|lQ? z|K}J$4@YG!>s78o?Vydi4RMl2$K6GxWlfeeFJ%aSJ9-NBPSY_gWZ3^rcKt?0ObBtv$7%b_I(KbAE|I{@`*xz!kHZPv&@3m zGLA{Y(Nw4K7?s3=dqcGyC&I6^k>y5;UItzq`AXY=&Dm2S))=x4(@KX>Rat}rF;ols z5e0qg^V2S7JYhLQ!e_q=I5qQUd+D<;O%AqL_KbZ$K8|J!0?DTI-_(70z}~6OZ~U@C zi^_PY$@$B&gT{i?gh}39bc$y;enmHKXFw7Nm#*-=O(V9QuaiLMV2^-aZGy;(H7~(i+g32EN8|S7zR#d z4MN)%JU3sakU~jEYessEjB=}PITSn_Q>7ljMzow@H{~ncSje0A zqT+Sv8Z}hXd9e&%x)94uKkbiXy75okG~Ty$!}%XZv8waR?^G~?_>kdywfCzw!_+)I z(vofMx!UsO&ro=6YS=u^GxXnfj>WgS$uh_|^1Fvf#u~4~8Kf4(N8Wtq<($7zLdN|& z{*q#6V!b6hC-ZJ>o4<=iX)1_0)r-g2DtBZ8jvTB&?d-D)^b#d&Tr7tUl z_Jm4=@rLj#_fv`2kjg^Sn5UR{Uf1RgudjE?v<`Cl2igoWtdD~1n`8;ZX`5w;UpcIM&j&W}^Y)KS_L5#$8oYg~)&@zL|S zXoae-Ow0VIV@2JgGsN@5|TCF&dRKIb_5u*&0SBSDvvxzOKR{XTr3g{NDXnk-T_>MYG=~By}q<%^RWNP& z255$1;&H>>s}>pCxfA1uNCdAM z+v<1eua_UAWGgq6=bRf%QA$+q$wiYMY0gRXUfB=sz*Js^8D>U4$?OQPBI7J{4>1O5 z>Iv-YMtDs4!3%%x1mTM|mdxF7XB>cUDn07R;WBo7eHz(%I@xN)ZyQWomCA6p~&C2LQhxSnOtf=mB3s7!v5 zwfgS909Mbd7=+HfxzlbvCdOig9Cdbo2HZLBQGKgu^i8D)6{10!XSS~a^Imm@>(-QR zNWq0_YxYqSzA{Ww`Tmbk*v};(X?jiPDdYhd8gF|`xBy%RwkPzx;-BjVjsm+6n;7(| zOXd~!LJ#7zTu;vjtg~Y(z1A>?f(U{C&#%Oo<(XK{Ea=3@c|SDMaZn=Hc zs1xYt1LPI{%T3eO%5#GX49t1TmA%sk1Q$7b`vOu)WVX8B=Y7`YIV(}14kj?fwHv$C z1Mc3}Q-)4NX2`YPUd?~rwEdU|51>Shrmqg8x1tf9pCak+6KF_o!IKs83Ga=PRTTtAQVT6bAgrd^0 z{%dE}VNC@HC_5!ohqn?QoO2!m@~J2w@>To>>=*N$3Q1~KeSl~2jj0|zYiSWqJyu5G z>I{`gShmhbz+}DmyB`n^Ci-4scD`v5owUlw^%1|)yPg-t=5<8$&SHuv6_0r@z$TD~-! zj7>j2_fLM-hj_?NU0=i+mJrmc!+I4Igu^aL8JUtwEeK!vON*Vpr0CM}^ibq*KP^>9 z39=BsWLWlcC@(OzX;BNX{HSZS?mwd!H6`gZnHjhH*O4L!f*>6R2|=VpYNP~a zrD-UUVnR(uFhEE`1X3g+@H+t=XTPiazSrenqt`Y0ob&YOzVG9mM&#bcx`hNsS(LX` z5^Fo>JL6v~t44j0w;MduT>7p<`wku73e#d7BdPiOqE_^c`w=fK{e>S?!17583vfVV zF`XC9DP89Xqbz9RLs5X&9C-O4j%&OtJa$D%`!|bG~i%fIdj@!{UQ90u-N%q4DruSLavnp;-0e0BrINGDst{1W?zxo zLdK1M98p`ltF^;mar?OPO_+3{uV?^U5U{|gtK%AKlIS>H4W}mkxw%K+rHsZYR1DsF#6cYtyUPmH=!508mc%k)0S6FZ>5r)4n%zyi!|9EL=w9mex_(3ZsUYwh;pk%fD8q(Zb z2J65zZJliQw{Ltoj(K#9SN87qf)%QHCIQvg`NrdY{@+an&tBm)oGGsYAEr?6XYCqvg@& zF-1AdEeL4cmMr${Bp&N7@XuR}QtNEM?3{c!5_Dw+si_121>dvo8#Z?K&Hn^!F+OFj zaJeyDs4dI$U%I`^#L>OB<80q9_4CHH)9+^*m&s5FK2&h|6?u_vH8~kn z-k9a{PtI<(S~Y0scywV?>+xjWU;0fvBT2e@hdaxIyLaagD!ah`ZZE5b2+lva*Lk}7 zLSQLew#5_`mDfH34U5Z2DuKiWpBZsnNObLN^(m?esKF!G;*@fwB1x@M32RfSO|`}s zY{cQ)j30vCiL%N0Ql9WLp?w33De$yepM7&6n+d%0?4;vzXul zg?!O(<U?1WZ|$@Ny_42TS{z;gGf;U}{%JrkyOGWA)Pnp)*liR_ zto*BYaoT)wcpltPi-H1Qv(4LtltE<8s~9h~d~f|}hxQ*8A}RC<0X})D0Y_A{@Y%F3*zs_%-YVala1jE@S_f z5BaY2kVO+8l6R@Psd5+61vFR2&|8hieVe?}ZKPjwVyp_G#&&`Oj{ZQLL@X`DXI#YCa&NB2X${}|9q4zWU&W5aXo z4B_0iDz!rKNoNOn7Q&dUWYx8V7>`r6M<^po%(;QU$Qie-_yw1z-;h z4%kFMu>scNv}P?D)r2~_)d~mRU_3G~EE(Ao6%>)`NYY2r5Dpc?F|>C69ca%A-N zDF-G|ytGZ+c}%07XXAwSjJ3)!u{V}g*KJ=?;n-aoPD@_gg#n1Zy;Oi5IFsmshRl_P z&WMKPSyE%&@?~QUX)Ywb3;Q=E^!YO(>_+kG0R0}98!a9Tq>dn5aYN?94JJs-|sIl0f#?XtGJQ{%`h9m+KvXD#h*0?wxBV-1A8(aIr5A0c@z6~jnk@FN6%xiWYe!z`mRTGTXUq3Na?GS1LWS-0Y3^H-2+ zw9!3pHI0$d66%DZ6!{^=jDv8w4pEXK53X6dc(`-7E^6|j#K@$`f2fCPNGkW3=FpO}TmbA~)b#p5vZ#=BK`HpYQQ>`@J6&0j zf7$(T@Av8;HC*v2-*S+l(sy;Bv6tNFo7;MlR3VS(-ah?+DMhQ04@HAT!-Fa{bV1ox zpC33?PVCO&%VA&2P9wd-;&da8A?9VFA@bb=cZ0@nN!b<#zod|ogZEZehoV+jMh*3( zi%72y_c`_?V_yDgTx#o8wl^waXjA{AGnBlStN`j5vkf2ChSTc*u%H5bxR| zQ>s`yBBDXFsn)l_+xn|VSCbwdH6lH_={F*}RZk5!6(EjyZ>p*V1fzJ&P8K$I4dR+v znI6hjG0{m57)pTUhPgMJaTS#O`6p(2qJr5-V&ER|YvWU_66{C-WMv`0mGx^A!2L?b zSUoy>9Om{B1E$7mq_Iv_aHpn%1{^oguDdmLhdu7nOHwH%>>$8jNb$hn)YdVl>ua?V zA%lOzYwqnvIbeN>=4eKRMx`MvR7tBX%c6Yj4L2{B|IXldxNgsk7@S$#y50GVl;oo4 zJdL6&*7h#jJ27J3W&?wvO=$*pEvP8geKS($;~rdJ#D@?sVmb|3w4w$l<3~ma$vQ2M z$(s?HzN+XS{GdlPK9|v-8-vlMJV6TA&Jd;`>a_%C9>d{jgWOz2>I>#%4zu=p=GHhU zgGS)6wqRi{BBvgr-La1tm+HkApzH>fuX%>wV}hz67i}#>O9#n&=fM@F5?32Y=Fd83??3)j9}l9 zmyQY=oj}y~{K6@{w}Rn%-nGzKOP8P6X599wY`Mp!3G>1E(*SGg zgLtWQiqfQWD%Kd;EC;~e>cK?#r?y-gOFlD1)SFe-GA*3LyEy|~^{lMQNewGA>^wP0CemE1H5q`}A)UE=^w7~Ak~5U;ort6T#fU7cZfnwZVsOer?S)gH zL-MzThEH3sg}ZZg8)b+Z@sL7~Gk8IxLsS?1Z>Aon&w`_3MVYr?&@j&B*e*}H+=8D# zu$neqK~P#>houO5nQWNhze=*L;4yF|Xo;)yHj~qRE}Nh>T>XREZRkSbt;*CcxrS0b zdZQqqxEKm4@6zuPhO}7@*XnerwY&HA+J8Aw?4&1Ouw^*X^U?GjU~uE6iXuTQLObMX z5Gyqm7@y5EWQ3ix>+Zw!S#2``hgsAe7^Zm`@Zgr?M8Zd4j^2 zh1L)N`T#sp&mq#wC22tGFj>nf7V=zGj66E9=Xk%K>U!ztE?2cx{NqRkO`rUj`?ARQ z9#N-lJ`P_(6&}1S|2Z75QBBsfEooPSbk!gyLx7@xh@ifS44p+S#r!-I4?WDdApEhz zP&7Wk^0s6Dz18)_O;jym)o}WTz~tAYh-1<{`LfBa0d~Y_b)YPSxQa$R&??iLk|#!} z7L^P}^uV!gyOG>cZ5$JT|2i!&CAoRHCOS#Khg=-{unKfPXsSWOo-XS@TC6{fT|=7f zn%{{BD8im&{U#^{`w$Mh^S#gqG6s0dvyZ6NFt-^XGA-m6o)TpY z1BndDi$B%R(jrgj!h@e!c|_ao9v9NQv&-U--}xV{VE=B9BbJTtUhX<)AL_;lEnayP zqd?@RBgQw%b7w-y?sm)gY^z=d>s|rvI%+jk;4NrvEP*H@AA_0oN?gR-*SKK}wVDd? zc5|6NFoBg(t8L^~Zz00QXW_2St!h0q{7&2~=B<{F5pxFb!s#n>JT(R65`yycIO{JBj6Gj#7 zCrXz5uCxH7n+*(*a(i|$)B0KO#p>4&;U8$dOT*Mlt2>Q9Rd+egfjqON%TCm-5n6*T zeZ5rAFWVa0LlNK;Aaa6cYzLSCRIGWdN#aLa>skf2Rz2Xb((>B`TsLhpU5R#V(53x^ zcd<7r-y(aTAQezp)-VPOChYHHvOUYO&3_xv-D{A9J#WQf3b{1J`?KXLg6x_D{l@lX zKKNDVjH*)J!dm333YG;(wWC(F>y*gw-H*~g=daH&eXtx0e--MA^VuztHFsvb45em( z2!BlplIoPm3vAm9i?MGEbL)O_PKt^L8J06?vW6>{Z9H+lWcB&<#r!7(GTDM&Tg`RI zchvOV25q#!w>3WqxPa2IlwTkb!KOr=yX#NwaBoFDAqiHAn~AwBYiLOKty%n9sH3HK zYI~f*$kjocms2Rgwe(z(i(KxS6$YaC>Lqvxjn1eu->Jazs`Ufxdob z_RR8wZT|KsLZD(JXc!))4#-Va5sB<_<Mb|;5f4!F91|83PEQD+>i#m zKoIcEUX3=fmYe?iwT{A$wzBtWcb|j?P_Z!VJCt$!h>M-V!P=ilV8qsU5VTDz$F8U* zCWl-YG)2kytKp0QZ7+eCLriMn3sO^CD-s=N;dAwr;#m&AwGTErEG!{=$=6BwNlu=2 z76xU6B&#5VOi=ryfOL-n;&pa9yyZq3kETM+Z$3?riIWz&LOP- zkI9u0=xYN$sm*n1t=pV}8@HOKnDRwm?YwM-M-H zgp@_}M@8o^NN*r&JjDioifw(n901^U?oL7|uMt^XtTg@edmMeq{4C9sG)i0;1A7VM zTRCxH+U|YN=T40c5}$vTDB!d4)cv!05lpg#`GqM zeSGL^KAL6GXX*Pz+K4o;3AoRROB=jR4*BgXvKKvPbk@n7g(8|K%>7Lw9O_CF#8s8K8F;~XR6T-}3`2>M9~geu3V~($AGI7eoh=uLe&EE=mJTv}UJF1BdqJzeC;Vqg zE9e&iBamMa2VnQs9Nuw8bGCfb``lGWSWvN&FCa22Iw zVQ^RZaGwek?3Q)&^T9STgil#LrbRO;dk899>0k-K%+)C!H$kDc;x}dN`YD4_N2|{y za@|!x{^Vp{NN8Wrp!OX~ku7G@9Z+=Bb&XaqC<$vh7NIK0Ottoe$X5kWvk+r5`Q8nQ zW!fDo!~p8<^KFm5IOH2yTF@7N)5_SW>m<(YYki_MJ(VuG0P(gVMj~AM-4CO2cddBV zBk;b1G1?g{f{Glwm%L3(QLY&}n^Y~}8-I;?^)GF>~x)Oq)+^vlQ!LX$^E^H%X` zl4@m)jB`DvM>DkS4pW!IudI%Sig~-9ec1>nx>UQ$+ICs@<7Ohxe!u3iryzx-oNSK; zAzGRutn6B+e>vo%8b+SX>+OoSWqu}5vA?ii5n z{#S24dGQv%RctM`@s}ivLi(h{wY}Kf{yo{L6~u`BoFXeb0)2xo$3-*CIe!82FpQ(qR)VJuj4 zUrrexk@tP%?z$%>?9n--7nmKy1^zVRWw@xYa=I96-yzq0>W>c1Z`~Xpz4}v0S4tTh z6+RWb7$V~RDE#}^f*y=hr{tJc*+@oK^unqt+>b(wHBW>2 z1jFzq*j3-DG9hnxyvypadFPPyPNXCZ1%O~>U3pG=1N%-}_;&UJvG&`3-loxkpm)2W z1_XHHCh4iU7D-R-lMA!U{72lQ!6bR1b&~!k^9y5SVIuz453-)=y^v=>xZn%vb;9~6 zu)K8pe%z%YT%kuDlIOC3hWx8Il1XO@Uz^DEXI~X!+Iq8e$v*;q;5M)}1mn=rb^}}y zsVtB{z+;JP(9xG{*0c=H`yN>68gZ+8mkdq0M+%{o)3Yvpw%ee^e!0|>KnWn5<{!c9 z^#P%BLYfI>s(aruQ`(i69E-nkivJ zdS(cEr+)Cer{%=8mN0YxGoZjWVtk;T5O`3$E9tN%#2W=P9?Bs+C5wo4s30ve zS{^gv(7|;RhxPCS0v>2MQfe{fX0l9_iCS9!Xo7Hbif>M;pz*g@U^3m|)xFb{X0x1B z06zLhrELH4SoJS%G$t@|T_t1uz2!o%$I2q;L7>TjtR!FgrvSSU@G4;ijKdf;7Y>>@ zer)`6DwAN^!pj{OuzG3p5xR@kT+w&t70E+R{c6>MaJ*f=YysAO>a(S>8L891LZm$x zQ40eHGoIC19WFac=*k@w8HE1j%O}PprT1>}D(lf0c{}5(ExK=Hl?Aw!8chCfi_zd; z>CkDgoM@)I?c1mBNw;eT)(Wm~@SeR?`o+O!d%Ei*=V5Vwn6$Lq#c`G>&M;?N z?-G>;zlXX0>buRm7{=XLVAE$;=hwIt zaUJI8HDaSpYa!BAfJ;EE2W+-a#aCA8d57v zbw}N<0RgXjs7qcvv@X5TdNlzEU}RKIZ6@~aX5|h&!c{e!w9O(siOeb(}4*Aq+ZDAo@xa+|K@X@Ia)*4eAo1^I$V@_$64)C?Hi5<9DBbtaLg?i}oJ5Lb%NOYy*G72Y+a8TLY|E zPl)AALeiSL;8_xN0MD_)ryq0mqFh9lDWeOpKarlfdqA*?YqR*?q0YQajFHa~ zD?of73C#cH#-e_d=D`B658CT_!@ki9C_?m^cZ&U_{$5xHgggQ+;PvdI+Z4|NSO;!J zP9F?KN$mQ>*>X5=C3ak!K^#$ti`oMN`l*Z)Flsfc9y#(@r`5Rz;@tgJ0oUmD+N z`L->i5%LS-dn}|N+@UOH(4>yiLYSJ#A0)YRs%iSlB9+>Bo{c7?q_{0BQb;OjJv16r z(>ZfDrS;HyyFg@Ke6!}mjwz5jM2Hg(Zx@Oa5%UyG7>rFvQTnWPi;S*JvCuB4E& z0dij1o*b-VRjm_-rQTUAu^I1Yat6i!TCf3<1a@)3opLhi`ZfC>eUbVvS)uWKJA_)% zBaQ)F^vXa)^J}f@UD0LTS%rj7Y1IgR6hKjif6qgsPl-SvcpPPK%i=419%J?FE+tHD zUVLx@E2UY7$I~$ZbwV&jYs^=|&YjSrw9Z7D85h}T>VeJhMJVEXuHTwbzoY?N-abL% zWi_NAc<04yyJ6FyzhVE} z+vmg&%5&`&$?nlz8XtiV+dY22a8=dV_mq3;-kmx%{ufG^d_(@_eiL53UI=eTSgP+m z1*>Q?h|GYQMz#dZcz|ThU@5m})PA86vlGw{Y6y9oCruSq8?9~#S1{CE}Q`g%1oOH z)NZ9$@HYA>dnwPrKWpu~Mc0YkGvH3#<_YoCx5&RjlF7#im$F(t_oli7(zCNqqC4nn zg7&K7#ggzVaI|Fk)GKgfq?~kN6;=3BL|B)BdX|YeTjw2wBA~M?W5D($CQl7!3cxAG z@kUl5W=g4h18TS`lZm{glio!MI?IvqYa{W(Cgu!nZ{H|Ab1`9-zr&Mk$+>4_C2LSM zMJMj%FbM7_bJ492<7AYkU6`9F;hq6le<{pzA6DjkDtr1>$1a|mh)med^m@L$PDQ?E zInTK7)jT*1DU&jt-tQWHQn37NMzUmf%jumK$6ICbu6Bgh8=Mw}9r&7~KBW>o@KUrd z-O1?X2g!-!vIDXPCqiP!0DcPhGH!k;DJ^IDzO-K*AdQ65TLAd=xL0=p*@pBe@SUis zbux<{@NT(HF4@UFDQYlcx%CEj(Qgs<3V3J7$U;UGHnBuEtr8araIO43;CAk_e|)!q zUMY)elD*#w#6jXdmX`h5ZFD{36tGw1i9IbvY99`L;;hl=IZm3jYq&nh;HJCm=g%cs;=SM)6I2$rez_F>BjFG+Y&YLCm*(0k<5 zm#5s@HT#=)cUgWKAiKRNH&dPK`9H2jyD-(|y7XcQ?7a7g9xrbWmhsQl>a2mxfURJ$ zVO5MOa@HVrH%V_Vn^hOFbRP=PAk$ASXt$zh3O7Hj%@M4xy1`XW^)eO>E}kt*Dr#=) zljMw35wvpu!2WG>V=Y>OE;!kxjw0u8Z#%qOD0{m8bJQZfQd_AbZ->>c;G#V7DJ#0h zdF$E3qg<#reFML2`0vz4Pxzrskc{U;AFj9fT2(ETH%Pi>5G@aL(oNRg$(G zC{~T3#c7&2_T-~-*mqb3-N3Tfd+@3YQutiPtZt#t^TSh!*;XqXVZ@8o255jH`0xAD znK7DDup=?gQvQ605ziGq71{Kir1gwjQAH??Cb~01$*MCFT*4nn6m}e)SAjoLT*53Y zpQy=u>GibLs^@Lg#((y0f;qxE;%jV@y9Pb(e}ubW-N;-*xlo~VA9>m~h1dDM_}8sb zd;atI^-_;4hK&8@-Vny&W4zcvF@Whgep=f`%D7P&Z%NUUC;)Py6zwea%NyHR$cD=p zmco1^C!ZNqV=YDIk@BJ{d zvi!G=o_?w45)&(tJ`rHCUQ=F)ya)4aP_69#=*=KRad#@cN(=BM(FK{;%9-25X}r++ zzV!$VB2|k)C^XPLvd-x6)D=<1a0J+cij)~CGwI4|g+2xi=H@WUbDbNuCw{IXA@bBb z@$0J*KOe|np)oxQtOC6$M{Tg>%$)SfM^rkUHUu&U?6V17Q902ic2bcAzL0W?1eNjM z-5SKnArOpxjx5?mJX#IELFgS$)-ji54#0%^PY0Xd(T!guM~J>Xa&UbCtJ5w%K$~f=at=~?_+{VN-Xylr|DSHLPPlEhs_8#PDybBj4@`G zFB4{gqN%a&8_VHYt_Gb%)Bjk~h>OOaZX}(LK>lQ@E(CO+UkJg2- zQiA%*$hyW|6a0BdwT^j?k1dc%thnut2cy`-K#|MbVVw_M%85uMZx1B0xI=xxDaPNo zf-THIv$ggWuOK`IasQs=>X2OMAO6CpD1x;A#}zrF#;UKf$IpiHR$VeDN!2~$VzIMN ztiK{6ul-D#9BA~>EKmVoP-)G1am>vtkroD4Ryn1Z_nW6^aDGChYKE3S+VYXKaenOev%#*3JBKbHZ7~mT(?|Hu5UTet5Mvr;5|*GN|}OI z{iC0~36nqRls3Oue=>|H;qFrRYnVjsG;#9u0Ad`1G-WlU&nT(~Stj0g`2v7^UPz!c&Nic6X&j7?g%_*ncDja*-uW8qK2pG_svCpJ9lD%je2i^!jP+D+&1`w&fsmb4XFS0QLXP%TU!AJq21 z;dzLsi@s`D(l|Q+()w~ zOlZHxbho$Bb!`0^1HGYm>SpSpk`!J@`0@2mX{OPTOIzl{$gFwT+~l0yWB7<)L*ew$ zrs&m!q6`BnMM1$>wF9)qa}Pkd5<}5f)oogl%U_7{WNRa8eeaXsu1j=we<-i?%vs1< z@Pz`vupIikh$abJLo8v?x^2T*oZES$GGzT~+lHDQRN6&WYjHW@|5|K+dTdKU zPQ!tIp7pZ22LAc9=$?|o^*nt-&g4|m<9)sJNzHT&&!Qz8{(jv$S1oL61D zniRaGc*Ij_D2ak}%oFR|imFJv^`cCddttF*VSddJ1HppNi75naiob2&3Qx3+?(d9V z_oS#X3Wja#)oYRu#X3Dk+TYlbuYTGY=$4o)xhRYaGd*o_@OSLy=1qQj$pNI~2u$Zv zYTZre^Q7R}q;h`1Cldr9lRrtGV&iXF6T4?2Fz2gY?J?K9Z0J0MPt@}ROrQ53Q(b@h z(2`hmFj8NY32SDqxTpM8>Fj!;D7nnwj=xp9Xntdv<14wJAcya1DJSns>IajRqR3ZD zE}adDO&0K?!4{rkad+rU)~#6nS<#Mbjd7_Cg5Jeaa$!x3b@hTD*Sq}orw#}Fx^?vN zS|1d&1yzE(^}brTFW92QMez2Zby=Hxa#cbV4lmL?#oIo8d#A(lv>n^ApWb19ZvB`% z{xv{(M?Yd2dN^NmW$0qR(_59pYrcVnAI{t24-IOkTU*J-w-g#iAt5lm{ltKpm8ZEn zr&lX=pfPERE0U;hF2w@qeOKP77C>CK{8e2qS?fqI#N;?}VcF8$KkR&y`)7y#@6&WV zr6fvSX0H0F(_>Z7pAV0v7}EV=4-Q!FZZr4(_!ECvi3}LYvUrGfxfP|s9DToKeMH%> zbZ(`Wv-0{xfJF>x$oqP1aQ1fGh%qB_vc!k`4v>M$*+fb@Bgp;^}UZ z{z5E4Lz?>{wjb@gM=hiMj{_y8#Ta8CKbpL9UZ8wm{OB)#F-o&E{qdX<%h`S11Xkos zQf~bA$B(@=D9a(KYiC@W&&)8;fk{H5ww!^V4eX1oXyF>Ff_Q+Zy__ z->j+Vu#=L=6lMYxt*lor#WEbhY3`hUz;GmKMkOd)9-KeB-Z%E|`aM-=iQw2j-qe!L z^ZO9HP8wh)L8B&V0(y@4!zhNw_0=7`iA*9ZO6tH?DEbV@tJ^C(Wk!Q}t_ia}s~Hbh zhf?o^e>38@kMV%G=$I(J;}~)L=$m`givM|-2{Fb1a9vaMtcG+C(nrc#){Ndi{&Nzb zRZ7vJ#$&y7$5l<+e$IdXG6++`*_0omWb2aV#IFy;zwd+B$rg2{uM#rgJV9cri$M71 z8X+##WWwkSyw3R0oVQ%$z+NFt z#%+m+`b7;0@Z_lv+yICQ&+^pc0WH`%QJg+L=|j#nzUZD}*HE^&Z45PD zXpWkHT&|=jNu%;u_W*Sk;34`js0jb%M|@`EP0RgXkQz)mI-%|uxA8a@K(1`Gl@{7A z80nKXo<2Qq(iY~WSmdDTs5+&}2=S^&HKtV;Ce)GGf2gW;4|Ndwf7s7i38V*N(YmH9JCGT6k?cU${IZi!Z_VY2` z6!rtB@byJi%1@;7khvYU5vYPDp#6HR9hWIICPt_*@^fL4!;6gSYqem3T{(@_9Vf12 zv=CNdeJ(&Y3(#y^SE_X))d;GyQ19KD+NMT{578E{qFzUW!&>q}s!q zgb;0NG53Y)gRkVpbBtM&{Eg8*_coefzn?7VFNw(~_f|YyU{i@^PeN$cmPPCv&;kAWl40J^XH=KL~n-cvfiJxD7z z&>u6ELR0|*SvDuh1wXI9>vZs|1Y-g^P9IvbJl#qT6^c%rKn{^$p7 z!+}t<{8Y!tB!`tn=EmD#0iDi6{~!gmDy$q7yqy>+HvZ5wLXORd@mBqQ!f(BS?2oxe zzG~~Z%{wKiaUm?|1IVk!-~936-w<2EZr3E4VA7mmcZTbo(qrkj<0}UXx{dVT{${b4_@ab+G!@h{ z*);S|29ZlE#d;f|X}J+ALoDsrSGPfdvNoVV+w7W7hu+vSsr}|b`J$44B1GtRy@A@Z zH&DRlYwrU!D6@k1YIB-4f>wDdnxgb?Z@Vvue(!f{b9t&YdD}@Iv8X(jIS(Z|GJS+f z-|mayxhOD|+Z64!_BsUd1qBN6tGM_i^mn798d3rH7jK!<2Lvevc9{|2L(zZw(Axsz ze0=1Jb${$a<^ZwfpiPSLx1X;#X94#CV=>ZM&~C+CS_m9dS+njvQtLi2hst>B-BJ77 znY4`T)B{q*cZ-@KQxe2Wseyb(C}5Tm16MawPrh40o3US4d*JqLope~)u$YJL*dx&j zA9!PcSOzmgK)4ePBzBv7^ekFfIla>y-%L^ioG7X0oQ=N``W+t;F~6_;#e0#8KIg0E zzp^KMW}qLFu?=*_*mp{i59APf?lDR{O|M-gghRwRV^c+EyIAnQJLB|f6b4T=K{H+& zB}FE_Gd4Xs0Aa0buucro(272^E^+z0{FTnXN6#Fd>=eHfZWu05%r}D#SnqBdJ@hdj zyb+b%zYw<6ddgB`N50{nd`lYDQe+W+dfj+xo`1k}LQ$f3;+q%C+UO`y1PlKLieOC3 zntOOY1$-WKtWR9QVXW`HK&$JF>mH$x+od?q;G0jNf}}xahUY+EJ*4dN2V%X>)i#9D z$|Piq23`pUpYD{>G7_gxwiwM4QYU*N?#+Wm9<$h&zr1n1w5gj)Yq(d z$mcL(ujokZ3Un#PAJ z6g5^p*Q6--IN%CCL?BX~7Ry}p-KUzS!U( zBD>-YAi4P|)CfvcP!1i)O7{g_9dpNn*?sGw@`QvPtY6CjzCK_0q(Fdxn*=2_T|4;& zk?fBp-g8jmIrb*nEWy2U1|V)2$=~~<&$`z)iF@%tp=&S`}og@nn^tVnSf&Bv9y}|g!AiVwzNL_q#T=Ai7iiyqbJbPk1lY%`1UcdxEj_!M6s4dKdZ~!@&F#REU(kiA@Nc$;7-n8!-G^ z4l?hco#ZwK)fB+)=T~bBUPn5dmuDuYkgjQLO}AYi7D~lC*3j8&itW;AM4C1PH=%4N z1O1|&E?t)XY}r+e6c+;UkD1=3vghcJHsL`Yua>9EMqT_a^pPq|v;uDqge?Qb}W z=8W$oOiI;|w5se3#gb&|VUAAw{7m-hg?fyQ*RnSpI}RS(NS;?oud=BY&iFhupqa zzSU=+*LRGxYQQ@uUVVO`-~B2+|0W}+Gf(q0V6X#G?^}S;F^tQIAG}FUatDwC z?qwe(d;*ktkaqqv;O7q|uo`{a(1Bw3=&wCiMy1N&SqV6ZfsLWkL}>`{cdXr!b@EcN z<4(0D;&h?MVi_zovXPv^b+<#UI0vTt*X3m^MkWbF;TdnZt+X7cY1Vzf4>0_t%1Rzd zn!Z#9Ms7eJ%hC&L2f%R=7PqYreyu)m=|R5^-gvPrhm*NLN`L`%r%WHM>=uKjT>4W zNgra!65IR%!CmkYUJMfpK}kQV>E_@}eXnh6rABWuH}5;|;99(l zw?PW0WA&3xJ7gN5K&_fKb6W~<1bReQdBm8vhrO0t^F5NSo$sbX<*Ky$)l}2u5x+Q3 zn?-F%xA~eUYHWd|7;}`*@JulPF?!~rd{``HtGMF~PzqLG-MoFV|0h+EDrakSY0fbD ziIm(7cB_EfriDOZ5a81&&;NgZ8W9P8eBuI7%jgWU@p3aE>#OAVOSPAZZY*X_4M((W z^3&n$g&>93?AY5J&Ry-52P87aOg-8WPj=TlxOQ>ymO~{83I3*QsK}6HcX?**yBm4n z1sr++>>A8f>udE2@to0qPZ1$|{&%)~QCf1$HFe@fAMOY_H_v!VnY~8akyhqN(e=zq z0#Vb#z{_XM*(d$@bY41mwc$TAM&iKucCwL$*aF1OH`zXMcZuf7L@eLocJR4j=)F$S>!38z;{%NHC3p5 z8Jm>}1{m&%a02oXQ5B z#ltwsWfE!K2wl>e8{t<$sGOU<9~=^;D2IznO*#qil?ua)Ollxl!llh1rvN*WkpZu} z0gx_kec*kM8F->yxZ6tT-`3tois7Nd&Mz$Y)$00Zel-Ca%)g$V|Ka-LzsFF8fjRSo zPK#!0HZq#O^jl@J`Y+1kpp+j+tkc3ke$x2-O@3>kml=bXIa{xV!{&b_2k4L6vYJ7ZG1aJ182W3`0egEC?6M$BFon5W z=mWNjhk@4es_grF23`8;d5h*ssZBdLY{140B9afmunGBaqvCT`u8LnXKF40rD~9mi zQp+~NHEqM%?_hg2-YEog@uIYTYLVnh24Jp&3-mVz!XE-DVuKO&0@dq;t+wR|x-D|87wqrfh*>RnJC#(!Lh>v8}m>dFH%0T{L zHG}o&5oj7>-UB=JChwMCC{JuQ59#HrlRZ2%k@C}EAl%4(<5Z9f2p)+QLg$Eu(M8ReUNRjcPAvC#Hx$&J*^F*NHK=g3;^UZ5r&CX=-0UXoNxlY z7r-T2N^B%^07c_Ml zNYq5`!R-c*9?V8+odzUfs^Lb)kh#$>5q1CNX@1mReG{VeZLPjaa9flw-WER(^|!UE z+nw)<>mI=D{7??B%UfvjL84al56U0$Unf6T?paOxTWJibaUBK^(RT3*$({duXhW9^+2-9iar+GK)Npx50I&;ZI0k=ovo)0|73%qwn;f;6)D{`Mq6C} zUl+zXW~FCFXY5#2H<*v=et(OzLYVsH<>lR$58n&EczNyG5Zf>yY{j2^?FqhY7IyH2 zI6PLG+1k(9m{V&)ZHT1dz+>CWk0%2>HEsxDZ+96ux$&tS^NXdf<1y;^jOv8I(WG*z zfY=?%VRI_xoIL77OU}$Tp|x;g@Zi^GkM)xU;H4&@(xaE$pr*L9{g!#h5rcuqpZ6Rp zZ!^Mzb+hya+1l#o0Lg3%U>?<;~^XIoHO^`Zmlbp1jofgkF3 zsVd!mw9d5HPTgDr0$NCCEhfF6%ps3OVLpCPAw-JjzomMywCOQbtFu7K?O_G{XNSD6 z6prT@i802S2U?vbzS-;zKqK51|EVCnl-mzNiB;0=wQtf#)|`n8JT;>Fl^X+ZUoGzN za$WWK-b!zx{7?LdKic@HlSb6vYxM3LVZu~Le9&g(({D@x$q(FL`)#1WSB)PEwK9Iu z*UbG1aD#)=RvO7#llmnl=*bR){T!7x18gnHjo9qjtJPaHy)4b(g@~rD<&N%eeArIi z;E|$alZKt0(64%5K&MGQtu;5GOo(93wW7eI0DRN+OAADDX$xn{LfxQlfc*%0%`mW;KT-s+;vvb+%9zw zUV?&jhk3iZmxDRfX*@eZjnHR&H4V(&nh6yAyVrKSC?*0lzu@Xrr<56~xv*PZ zCY9i^5Kta;f=n>kIH3gL*j99MIl5;q0T#55gciepyOZ+_AWsqJH^uw9)9O`U5+%0l2ut}#7ioUnF-)yIMvdx1p| zq`}y`(l+|(TKo1!(o_`{%H*F*TtcumeGHDHHIDF7xTSDH{Ue|Vj1wMxb~97%a93+G z@)-V}zlYcryq>x97tJ*QLq@RWT8UN_&G{;u*u1CM=xZG{iM>(X1rnqPq^ydx3On6B zurdWTT7^#0*AGv&EZ@G_4JcIsY=q(qW8eP@`TrkZ^)f0K=>jp&Yt7k#=&a6~x%0Y$ zm}lN`Vjud-3=V*|9i&x1R6yBsic+Pg=uHR{F1utjOUdgu4vUX5h}r>j+`573{FTb! zU(1*k5`#OB4i~a=FH>kNj6Zl)Z0+%#{|@%u;jgzs>gDRZQAfcj+rSI20oJc0(krr) zam);E_Owe(bX;hjQuh6C&#Ln-jTWVIv&J(5G zi~jHvUf}JE==MmhHe4abeL+ycESWyq+cyJrEe+F#C9o!GN=OAM0a64s3psI1B%cq9 zsUe5W#qmZ98k{Iz-Zs4)=k(@n|KkMuVvJ$LjRp|&sW=6oDWdpYT9(Ob>`$Tx!^9QJ z9(db`#oh(X7+~U#Y-cGG1L~6KkwX4WYzw`v6wwhRfe`B9!7%6Vsd~*nYJjgd7^!D z43N)O*t(Ren=IIRdU!j~=;%5Y3^;A<+yS83c+^VQZJhAj*i$CjTH(aIP)@a6c1?*p z=#`4YWI^M8nHXu35LV00Q0ch*@-6FMRaOa>R=b$1InJPApqXZ^{KYoxYh#zRLP*5L ziGMqQH~N6Mo2DYuX8v{Tpj)!_@KLL1Hs5N*8#5UrFFrfbINd_+vVg$A;|eQZJ{Qj8 zce<6d1@sbp%%XHs{fiaCIitzn?e+KP#k6T>-O_A>C}fKgc+LrSOl4j_JN?63cN3o+ zZMiwW-ZB-zy#;A@RYgx90bxw*^f>TInQ8!o7xvOk*yBJX@l|=`B}qaJC?G)DX|op2 zdVF+%NJc@E8bzlczT^_o%rBr*P#>EI5z*c z)E(VXPYfn+4NTyvmLz>~1}?dbzZKoTb+P}&gIhaWH*4rD%5OGl9vg;1WcrQ=q7n}+ z2Iq%91ZOY`&cM7vSSVJS&QEbm?o!JeUyaoB#YXX>`u7IjaGh1!-`N!!(_`e3eRxl2BN%THEII6hwY(rd(<@_0F z8G~AF9)PxJhqy5t|XUH=<@oBeRsBby1|f;O5(t+-QqZ<|3b@KV^X*Q9|*F_~ZEN$p$s zTJMwQT2SmJgSu?SZ!Q0Ce$l@$?`R3A?X&CN?rWltjavT+_|@ae<@z7*A7#H~EN@`@ zJ?P)}XU%qe^Ea736LkNc`K->y)r&A}Yyx*Ov=eI7k&Ekz=!_{KG+tzLoEPvwo44(Q^@7uki9~b`E zvbpx(yY=ct=g)n!l241#T z)Crw;+Tr8ZovsR8(sJ+B>|eUYF?HKK*8{J(66Z4bQ)0+|`Cq`J;CZd{uD*P&di{>` znd4XfyX0^Ft?}6O{`51YuXd*Nh;Id+IsSOgKDkBPuDbVlpDF=*+4%UD?|j8IN1n&8 z=07GIm!G|6@r=()=QIDZiQ&WU;YSbsYziK(E#>6T`}ELv&x}W6XAG}=5A$DJ=#_KI zd7rTDzCVctz+=mR7p;WyCju{h0xp>SB=ycbOX89Iu~|P0c385vmje$eTYUaS>qFr3 zzY?bW#aQ~UmzGt`JGA<@j9%ZzU0)WTUY(Q|r~LQcFQ4#VZ2jMv-giH%eqPwKXZzms zSH+&U6i=jO_N zei&Zq`*QNeS;ee<^+)!-kSnaSygPN~^SwP+_Z)Nj;xKE8xi-~a#g{8PaCDYpxJ(@2RB*1!UmZ-|2f5hq!K zu5ky-O#U&^xL!+piO=KywM(l%K$jhXoDc7^fLoOCgP(yiNVhmoQnCHHKt9_4Z~nz; z|Ag+jVHrk&dmhWp;FFf9>;K#M^}kWcQTb@^=de{zq*(v*?+q*9`l-l2GiCv|{V3;L zQki=oS`)I?8)y#v3~tnODAAVUP5QFryYdU*dOF*;qQ$_Qf%BJQNp<)GQvc7&J0kzj zUQVwQkTnG^aq=ykQvzIoS%ziU41a+B{dB+n|6l%kz1@%6FYfu*9{y*>ukYIZft$3m zGMs^FWiH9djrafCKjs1V56VY*7k`*WvV(FfuJW%-maEUuJ$pDB=`3(~Cj;mo#0j71 zA$vv8zG>fQ_v`;oM9u2o9*(611IAJ&{J1wnG(ir%eA)0Hn9}Rh{=S&C@z2DqCBpZz zNKRvpeS_4F+wjB=UY!l(A-c>!b%=HbUO%ImPGm+zGaWt^4tEl^{D+>+iHlRDE4M(V!xtF< zy@+($COl*??xqJvGD>j5Pc}s_pwLZ+k3fKe6MoPl&|ah^`#>Jzlt`eu(a3_{k2V@v zs7YZoXJH%L8_iinHX25Y!Msu|Cz656{n4B?nzI7uabPBe(FO*oEnu|f%V^FT%~_*4 z3)GSY-@!B5&jJ^+r$IeXTKBUigF03=d5!IXdccDHFXtUL{>3rC%Kn$|4qiF8olx29 z2WEj{=V>9gO-~U}-An%mW)%`oLH)4#2Xa3coKl6z?r8({?Efdnf@C&5+BZG=LE=4Q zZ=ix?sV3KSsL`@=56umyn9AI0&=&VGoJhGcE%$)bV-uM6CALqHQuE6WzyWHP<`+V;E zd2XNeI<;+!-WD}AwQZ+ee}Ji}ZS_@C`|8`xUju*H?2>o__-_LScIvnqdB9)}`0{1g zG0$UaYIRv#KZbk-d{@8f>WfiRv;JN6Z^MC5KaQH(dim)ej`>^>bO%(Zguk zFFg+fjzyn{KK8@YA6s7@vdBuUrAx&+==w8+o?A};inf+Y;|Y>s zn?PMb@zLnL)?Ncq^AKQV|GWN{z@Oi>C6uu4-uiU8%G9yf)qXKu;qCh2(-oTj!P(%K zgJtxx&)18xR#0qV0?hRo z-{z=xcB{k~_^i=O-}y_`hf80msr}MC(XvsE@-7JY{CoC-IaQ?c$ZL-p{qiu^pJ_D$_7@jQXj=A9Z68EwONQTsdLbc=+SvzPWKqEVL`2y|J$?3vB;2 ziLx8knXnZ;9&=_S!DXh)Us$jb-yOUDg<7SsnCFhh*1U^!2CN+z)pgtS4e!B)QmCzb z%yXeu_@E^c9@K(MYH>Wmh>v!lvP744PCiS6$uH!ZHfDq`C?QMvez_pCK|{(pH~&H{c`^_ACMKUDKpT6rehEn$zvmS*rn+4B?k* zmgSfviwP&WXyhH|RHQ`gYy=F?q_;O6316P>{!K{BDBQsu(@WMrEpzngpeZ$vCegr& zhnYq2k>$kV&9^Dpx`Xh zYxshAx3|JU$ygCje_)g^+n`o3WZ#}2K~TmbvNq&DFnBpEU{V<56mA71ziTV;YZ+jF0oms8N5~&RtD%MXd6A z-PRtQb-<0RaeB@Qk;D-N_S+VbYPi>B{FYuHDM8OYw~ijRyfT^Czd~B2wTs5P9Pcxv zU;fg&a>>CEW$^?Y@f+GjW5g12^lN2=hn6Qw4o9Se%a~o)9t$tLFUqLvLfg-7TPID# z)jX?+d7BW$T6S3wIj$8Y9T4XZiqE>O9pyY<4a{N)zfh?F&Z4(1iuHb~SkM2KJYk%- z9V=vvTz9GH@`Wn;N4K+Tm%#c78I+%UcohebJj$CQ%|{~nYZBiX{J;dB9?pxge*nKO zj&h(bdWB03K5OkM&fUaqXp5WQw@{U>p%v_pQTo#2mI`6TwObjBm)!yzK=d=mU%|!e zR~kzE#*E^X`880I-v*jZ0j4OP%x=fu=UF=fzUnKC;hB4{TNxl4rD(~-opooff1 zES^TCt!k`dzt1gNo`9}bTy<^kAGTVxg(;NZ6=RP^@?`$Q3V4$_vM8i6@P?^}eZ^p$ zC$rF8?n=yqq`~#0w%IbZ)Jz0-=dRD(IfIA~cgrxF*%dthTO%3&yA{C`?BACyySmJIHpO zJ`EI5*KXtk06PH?#l&1^UWzK z$U-dD!`NwGGr7S8>Od95KV>?lGE*>|pLxPR8Uwbf{kB{1A!vwt;qkFZqa2Qwno=~) zSsueLqiVl?U0mT7Noek8LfXTDDc*%C8!?Xu&s0sClsA^RlX&F^YEc;x-4p{aF{B7F z)oti4wlC=$lr}J>4v7`1f~2bv(={(h1Z}lbFSUmk$6L>HdfTpM7|U~AYT<8?Qx}z%kWSq)2XgNCq_Fh>g>el9Mi(w0Bb%k{1{+Hp{2eX7j z@U8H<<}O>CnFNktg;TzcV>^yOFuGlJwbW&&rb@28j*At?-NFezjwk8sN|oP@c3*zXgfKHiOYx98;z*)hA!5oK9u9l&Xe_U0 z?}rXB>zw6Ts8;ZVlf`F8xjm@bwsthOBH)o`v;3g1GDpR*RA zq#S{QQga(>+k-;opGOdMPjM2wX7iGs+MlFhn`zx>qOQX6+eV`t?3sle5X+yVkVyMq zGH^ZFOldZR)r2xE$P}unjl3}cf|aBfx4Z@Fnql^wzUMuA`41eh6KWgQjG1Mccb#lF z9kq`}b$6-d@CDfv&v<3Ez(?mKM^@00Ex&fu*69YDkgU9XK*`;YS*;Zcn zQZujuf@`VhtLm?5`Sw3GaocvTEu2L$voIfx{Lq-8p9ea*0_J&X#nU!rCNyM&sZ0QZ2DNj?V~x}I%eil*H>M$IT<QRzypa!i~~YNXLvW zJ6nx_f1BjTsHuVhJ%46Oh#c1zOid$oj7AXY&R02-eS!malE9=xr}wlM(pgbGF;>g+ z(^W(h<1xFR8z*;?l6fg1jn11YET0#!*&fE#y;QJco=-AfeR}>8%1}YYBak)}U|)E_ zmuk7v7uO8XH;pJCU5oVmd$KKtu5N+JJ#%3&l%0L|dZ{s3%dTRuUI==HiEgM{n8l@I0tTlVlA}-4V2d_`u2e09rVZ=1Q^y$`OgA2l zb>%I0QoJav(9NWzqGAsF$tW57P_>%jSwblu#;nRo)$t}X7%5B?#0JgmDdq1Uy^zaK zvj0I`tY{=6Do+U}YyL z=igrI0G|oOXES8k3Tbv7(FpEjgTd^8-6hM1&j`0i}W3X#A|{H&CqOQIL$%{R zh)Eq*aCLZ}=mJK)JJWs)bS=;>%~kelW?WErm%BPR8_$jnzMt)H1j6MxteeAd;E;qC z!XJ1cs4225F)3Zv-;cAnzfkE}wQl0=+yr>Dp*RoPil%xp0no2#XwT#(ka1q3*obo- z8G{gP{QaKvg-<*D`Y}D>iDuEwMt8TU=dp}qALRHBSwG`~|MKK_dmJt(mxsGcF;5Ez z2eRRN{N{^~Q(25sw6OGAFlCGIyO=f4W*s=jU`8<~)i`9oN$DmVo+*$_dT0A<;h8tN z0!4Kvw7lzIuuF$ga#v5JEtsAi$Xwz|LpP(CwSkz4*nn`_+Hp72_#l#PQC{&QAEjqO zDpKVF0WIr*Af?52HqRmDPwJLMr9^JPwbNHBXAB+fHgQ&GmcvjJ6v0&N+W-;gOmJHN zAF+sMn-nh@&|aBtnc@npxz=x1Je+L~Xqzd`#9L%Hw@)6R^CswJ&0;Chrm?RkjUhNL z#=!e(NITF12Y!*+O6$RLln1We9frD@vNGVD59>bi$!Nh4Ip8=*(XA8*igd%DJu8>0 zf+IV3z;2YxJA4MDHcRjUC5pMn5CUtd&%vff>tihX-Uz4%4=M9HP zBXf(!km7hMvY3X1Jg1@L{~CJ*Sxs-7f~_7=hnpJ51qF^pl1s94U?Mx+H-3YWQazPl z0_11Qj03{iHr(tow!Di))t#JG{tPt^JUzt zJ$?WrG!81m=u=_3=xJ8KBTH?Piv&MXzzJe2%gTl@PfI|ykmu| zd+RdpN?eJA8a7nQ%=M(v71g;o!Z~A#b8dmxd_qdu*g5f9;c+qjXy_inK*J0U_m~c7 zV>Gll_)5c>7SbZv)XZ)_$XH;^x{xo3k1UA|4>RgUpDE5|0J;Zc+gJc_@kArv%9ZZ7 zg?s@z+64d+eb?Q_dm9gmki3#5!e@3+hC?RLlh=iv!k_Q!1=q6 zzS_ndM6U1&)8RB__>*tbSZv?)lxWW5&;;E+EX%myrTdP|A-N2KW?iX@XJ(Kj1~=m+ z(!=`LBs7w|r09`RBmsK>Q>Dg3J|SWmr}ZSKaQ;M+cjPqY?FTd}%B1!C)ReX39np+r zFI1)n{khCoE=YU3@4I4ub!DbSBAFJ8XGQa-JFQyJBFIH~&0Q1(HuKsy z{6nK&u5+~h44hK;X_k+pryI`^B^3vi43GtLWbmj=3P0dn5h%Lp5l=9s)lnxekl@D= z3Pab7+SB3PADqtY;gvN)$MibKTLDJa+o4jP!=;1DzNW>=p|k8Mj}HL_PqLo|Us;pV zyN#N=T`9Ud6iiXKW-q@WZGK-_3a(q5k7Q9ifVIEA81xvm<`u|M2oC~=F@$D5WrY0> zK?{Gnh23u#E1$oHmO~MqF{>YS+_FqrK|Q+!g?guQKf{~XE7lcUX~eBMz=mLcw*^V_ z<0j?zKN6~5@mJJTabYw{s)wpf&3{xHRMtr=XLI}ME?=fl(H4cEwEmyTqQKd*ot~j`!NcMfT^6m+~JAZN8cRhJgg) zsfKKC3DWMPFL4{xACD_%bMBkNqgvL=`3G2Du0&d7V+oX~{gG;`r7JN-`0uZ*UZOSC zv<=>-ZsxJ(qvHG?03PpWp}xg)G#z ze!hA%Ho1}Ww?46XcQM!uDB!%(=7wYa@voW6&G6HdrbB^UgqAgs@z^O&YYIg3o<96T z14KGCX5td{I41e9TkZV-Y?_la)^vh*5zdtDw0`dN%3OZ9$9WTs7-d~f7c`hP1yv?M z_ycC4TTpO4EExvB0{j1v01=Qt;kzN98-W&4Y3WftQZLFxS*?#hX(_&t!FWMln(){r z+d)9o_JZ(qZmWUyly_dyBa=#}3($|-t(Th*#cR^)@;^Y-;S3;>XDbP7XyyklxHoW^ zRQq#egz`#DJHr2oF}+f&q@h$6Sc#kAU*fGSJVdpL_QeF`f;422uE}_onhK0A0brC2 zuphhv4${Dy?t}Hz43E3jc2SicIeQW7D%mfMmxj-Y9a5L> zfW^-#_nnSxs4s!?TWg4r)Sz}!EHrie`JN|S{=u2N@`0w)Jf+p>`CJg8*I3dA5_enD zuSSqXCLw0aC2^X$Sk1=3TyC9;a>I>Xn}23PsTqYa|IKal>$I=z8ZB$|G4fPEc49Q! zQc=ta(2RH;#J>f&y8jl|AKR6!0eIu(FMz2#49l{7{5f-+*ldnIN|4215wQtyPrH2U$>t38EQ$lPkWSmJ$LY}niDyf5`xzSe#HsC=bxP-a=i9e1m3 z3J3;4`(Km-xJB=PvrC3guxCY(g#ht6I|!I^cUv>J$pg~SrTd5{r`2hzo<^itTg=-v z<0eg8n^2d)<+Y3<5vosAvJZy7^Qg?=(kUMgt1iX z;g5Nmv5tVL*J{*tDu#&e0PA4z|yUEO71N^;% zhpoCF2oHsBer*85-CV%z!3bj?BlCe}WyoLJ{}sscg@U$$O|ZC z3k;j@!|`;hxyfx^DpYbI^?5!VQ2E$Pt3F0@kE!Gd2Dy^0G`6dR*3bZ9HwScbi#6_B z28&;Wi~JgtQ=4cp-gn9Heq5B%`@{pLU}}*K0C&Z{0e<9KJ)#nzIpI(L6YF21jvPl= zKEHddswO)J|HMZL!UbLR283V`6`AYzzPxn?P(a#IPLHeYfUxl1Q18_|B+w7fQ~3P?NK zlgrpzb1iFLCn5t=M-=)#>a-XGbPr%VwY_Hnlrpzhj)OT=w$d|u$;hc$MOkJPEEhN% zm>eq(I92k)!@!78;}Eh|byjyHVkoFkpf zFQ3uEkd~o{)jI2@-ENT%7(9pUMy0q%1`YT|ilK-PrI!xSFfL2F)}fr1ym_0rTHrL6 zy@Qb8K3smPu)db7c;bmj2jS(XN%6Yg;wh0M-^U;1%{&&_c0cLCjG}5Wp5WgCf)+eH z!=`G1me)CNapw_k(kn=6d+g2ML0%>ih@;5@uX*nSAorp1typiI^bFv{`%x8d+MF zvs%X9Y~K}czdHma;D1*pX^YoD3xeLXdIO0o#wA6ilw5a|@KR6l`=Td5iXF*9N>2@* zy`X?Z4$x*=(@nusCOD!aC5Sb0y$E1E9;sN5K%rykV;Np(^B=W%4X8zy2I-i%txYmz zGJ4*XLodIZqa0GDLcKLNbEL3(BA0#O=jwQS6Iy)KkbKJNYZ_KL$g@CW`~k|JhW$DH z^8z4m(Po($sS9QRPArpRj7y3sQ-e+Nl9;?>5_7Ml53d z>P!%~gE)H7P7ECP{ant-{X7KV6Y|#qps5FwD)imNoAOO4LESsp1M7ofm;IG0r7r8U zFV8e`Obgn$oLi*KhRBlv-GwW?HpABGr)!Wl^2y{Lu#RxD{4Upqx}-`-;v>y8WCOYW z`zer(1AqL*W3}cRbUH>OB0Lcf^rBpetDY(gz?K8fAk>8hzAzo7O9;!d@l&S2H$ zQF;EF&T{w~>O0oJ#YO*tqi&FHcPcnFaC+N1B+i^t!x+&)dKGZ2W(hCKS+k!Lu7^6wT~lq|{j!2eL|;t3-c03Ir|>Ov966JT-B=3bsZe0)mDgY@13g z<^*brY?`QP%|?}PbAbrDL6vU23qos1o&#}xX=Tq(hiR)nvf|Aq&U;-sG1 z8B`3~RzQ%#?If!srl`kZVFMtXArKT^{6*x9l2j=LO}nW`q!;%=9$3y+pBp~!Rh5;U zW6@Ju<6{H@({BmH_v-BKd%p~9GAU=lQE28vn1Fy6l=7QIZ1JJE+B^fSPL$X$MFV_c zfuN*g2>!`n7r6KIlLA;pD7BBRtokqW3K8FyMWltZ!FQ zL7P*q26B7{B;^R22iK`~cHtUM70lDhG~SU`ta+r5<5UFBW7t;f0!v}UK$uIx}u zdLa^9hQfkW;i2{9C+Qd~dl{_kj+ls+2PCUmwE1!&zSK67vRRPtATcSQ5qDnQlX|v_ z8Q}Pu>4x;rhca099Izm*KSt~@R*hN(lU61QTEd{3_py8Isz)# zTIvG(GOBP~=BMdA!Q76#Peta z6kn-LEK#Nn6E;}&qY$kiZw5Bf1J0Ztho4r-OK?#oFI+`tIBfdA)W}rcd2XBjWA4fD zTZ=bsJBw%dvxC?6#6{Zth}J}}=H=lzu#V<%^3PR7*u;hTaj5lJs_{|-)4NJS^ zXDsPPF@mAB3jHz0%5d^@Hmknxx_7rxPH9pyXZL{B==x>1k6k$kc{aeS0Fg3p!2%F0 zs4;LFH!BwHKz?V;3UloZdr8HKGfb^t1Q-p4MH;3(%W`LjZ5v6@q)EC+UB&vt*>bd- zn5}8`z{kZib&wj%Da!uHP+qyvo`WWjTX3FFb=ToC{8xL~c|Ahf5Z4_b?=Vq_1=&xfOszb!8c$l+jjuU z5E?;S)L5)ON)@c^VjZ$OMd3jKW%E-pTO&O8j;Wud26o7YQ1ONQAAP!gNxa1ge1y~b*v9gO3S`}T{H59sFHA_Z?)09B^NRuL4DpkQb zRNDI4S=+q0d+`F4?%IS}OLMyb9r&XH$l8p9voT1rBulH5!8O=+AJ@$Rw77S0xsaJ7 ze7Gb#v;sZP!8~l_2IM)_o+L?rHElnG#+*l&MC!evo8KQo3-s0&w$qfzTe2V+q93n| z5Z%wwlaA|h#)YDBg23CUyPL3Yx#@HDYhGeZY2PVWR!_^cXea3%Hj^kM0Ya5RRXwzi6UG=pv(bB8ftZk(=OcG2n)jzq> zcnl6G`bA0kg0xtJG)kW~a)1X<*ReQt&dkmNSa+(1%-{@=1n|Rt2a^2VDz-`{&tXV8 zUUSY#Ml@UpXu3}nj8@0yW?xOrwd@k*akFN-*p!-LZh^ZSg;(~3N}IaRcSvmv>>F7G zy&lKSl0%3o%c(hz0+ZL=PJtUKK8ekLG~@JTIj~BV7qf34S@d*c4pV)Io~g~jFx+VN zP)D=7%p2ic!PeA8wB-*j2VAk9X;5sZg^7KnQ_Iel)DfNIsRcwhRz=ch_44CCl{H59 z%ZCE``1=5CMYp)4jHDk2;D7)$dfAn@tO7=@C*d_WEh|Zo&^#~SI;tULmMY8;1x3olcx|Go~c(O0659kxnw4w+Z``>fZPBO^?*6rGNStd_Zqp)#5m}C_e@9mdv=&+th%5C zARo(sJpxes(zbk^D3#-X-bQ04rZP23Q8&iGKpWhvi*-n$(c z>&=PR0F9p6W(sD}?9FIv-l8;(_^ds9^1?n=M|cv*E{uixu0znOvODv@~%c zEW9eq1Ww*IWLvulpg+AMH*7<&(y6Fz$99 zyS#Ta$+ef(TSGIqcte`T=W!SA>qXVD2^&1!2IlHq8PMF)TAY|*F&PJJD^P*(cZY6; zd&)Q|Vzlt;yvO24M$IB0Yu%)Yyr;*>GhNAiWfF%=_4Dptb~2Ls5?TyF9!!~<8Y23o z#27BtRRG0gMY_ePS8h(b%L9YBIs0+rrc(#)?P%bpTJA-6iPSy4=`s+5U)yju&B}pYXk_aO!DH1IsKH(B59_D1yGHsY z7Xi}KI>oB?^zFY4=sj)2)fB*hjOCYn`n2q5p5A~mAR3<3LZn5A{be)Ymu=bk+(iBu zm%f?~9{ZMc;X?!bLCbDWH*y_32YWu@Js5aw+l# zRc2ek1skA~V5?wLRpM$6q|NkCgLqJa(i-EFKGkDldfSdfMY>hw0W1+wy9ppupsin4 zDAea{YDuxj-GUy~jU6u#0#HTyGJ~>|1qNz=zG8N4n9B_Cr(YXWSCkjSh~+<(AHjL2H;&72d{e zqB_GL-#~r=_~AU2A4aMCa7)~zQC`FoPjc;N$f)wbystar#fM@Nd$zaluLM?UJ;iQZ zJvm&qgo{3mmIw6cNOJ&5k=8f^4XUC>>Ul+gD-&*}t6ahR2a_ZvBZA{yhH`V7v=HcG z%LuOmiY-XkG{yz+)qd@jnSYRoMrx`2uP)z31(ejMduga zD6YFhbX#h8k^xG3z6Tl9wPWnExVUKK>iQW_oFp}`Re)x9{dnBpD3J9%KvF|%E4di! z(6?)6e&iH;6@mydds-NuOrZ!1>IfiHnp6kwrR;~FF3s}cfuMwBGB0}vTk+PRorXMz z*mkDPaDvH?Q8glxfZmhkDN3t$W!G5QlR)J!paF1`;H|381VbyEbJ~sb-tK3eWUYqb z>-ax=(3JFEd&yl4a75KY0yZ^I4+?Nz$Zwm-T^G-glJLdS1jr8vbL8o{Fr1gT@F@~C zn@Ku`2=MGG=6I6oj_4|tw(!c%%`@%sy@DKe89)zJsxNbytXHp zTm#fdz+=xa7B2xXsWtcBn&1E_8>tBKt|DfG6ncJ&rklQXFImu#|4A~ns|1?&|J947lVxC z0VRfvp{@nwm5?M%yA9Z0?i?MN*_CL{9}BvTwPq!`vSKWku4C8(KdVkl>{yzbE-MBK z@f_*5F&?hlgjQg(i8pQ&fQM8mG<{R*ktIx{$af-!pvs^x{nd^*e~(j~0+d|8xbXhx zhO!~4!(VRKh>XSz&)x(y&GoitX9bs=FGgOWVSrvOKch!no4uhCP;xs}HYi&GLhNY{ z&4AOE;037kHkMfvbvZr=Aj%8haADI|>Y21WOPOpy#*-DY&PAH-Nf4u$ze5DJGd68C zqPnqt%86FnzvKdvZz(V9+C3zgbXFBfF=qC)G-r?A=rk$}my*YA;PWpB)9M}^nVGs4 zff5Xr%}^jGX_RlQD?qGZe-4gN>y0BWG*C$w<`ck$ejGLe8I{c{D7@A zVn^3RtU0I~4T1JfZUlKd?Zn94U4u~qwqPPhUzm+)VzMiO zXhJ&*7gsNFt{2kZVzTx;6hU-_E5Vn%wC({uaS(IWgDE{()dm!YSV-Tjg)k{wg|&QC zP6g=*KG2rs4yeTYrbk2AOl}f+X3f}63^WI%U850b_NG7gtcu6kNjo>&S4ef;IADvF z;_(Xz15ZueGjjr;Y(!=6SvJ|XXBR62XmAJyP9~1HfXyn)+%|{Fx}T%;DnLwO;9GRP zBM-w9FH|T{K>wNU8=4W&hNHssfAHhK_*BQ;C;#`6(Yqh1JMfQBieI!o@B4|fHO!S2 zzinEjukZC{`4s>SXFtCe(3LCeXKD7leyedofLs5BX-^HqRB!-HFu`?Em_={**(Dnj zK^Zv7@H_kY`~G&`FKcSa^1!N$Y>@bFAZ*o6SGqD5%hNO)4a>DEQ=nkO($;%eJB>MM zB4bhtf=_7!5>lunw2S9nF||+hXlCuGE3k-5QO!(4)3PUANJSaY%;MI|_%VO=Vr2Vx zSnabKltc%8SlNvmD1TmD@^5ay{p%-}q7S!cG&--RK4r2A^UZZc7*8Dl#oDk>S`S(hAfLCvXW5&OQG2|A(`57L;vf_)t^ zL?M=TwVK(j<;2Q%uHm$S_8Nn|c03Mg5`wH3wi2tTv?3U1yxD0}XP7rCqC(Q(0#xGY zsETXxQ+_;|dtg(xX>A|}Hx%P6@vDWGz&Wed06ZzuFpUdD(sl=qo#!}gn`=JbM3KFU znPLwlR|~(l!?QL6%wgpRpxu8@{R8>*5S9egm^H0tqMB9hJvn$m&bt2hWTH8?hnr|* zJ!P3nZ3AMp6re`}fG*HhVm(y|xv!#!?8jpXIKq=-V!{H4Z%(6A;W$8P%qm-BY05HT z!oXywcL^Mzr4RJ) zTMRf5PNYmw`A~gj^=*jUYtXg}avbm3u}WorumdoR&KdD3t|_sOw0J~aSO*jtK~Q2C zr_KI_Q)SnyGRa5rqg2^gtn=}TBtjIYvX&}*fLAg!;#cQ@#)J)@bqFotymoeI-KSMz z``QL*P?>9%_mp6WI1kPT>Tt|@KC*o#$eP83u)>3BtG@XF*lc4K$S0B`D=wp?JBPyB zh*h95@6*a^2GvS)mE#(jjw7B6IShLJVK9k>y*4jA*)kG?w7$=68gHed1AR%SY~ydr`2gSf*qd%{6u*5Gy+JQCJ7 zjzLD-yvBe+aH$K}fF6**kJIcpqmtc<^Lnk2a0aO=9A^4ILZ4kFOT zS}!R;i)Q#>J8B-6@EGZDPm`L`f?qN`FOY0)8<@{J!vF%@eyVP!%jq0~hpFu1e%CJ; zK{J09r7Wf@?Z5jdFc-eg7W#0SpWqmaUz3om&@D8qC_C&Jd9sd$h&&APevI-US)~}% zf8C70=Q2{FJr}vWYd{b5B;S7HC!YmsbwUN*HLYj%UCh_+G_hAwS*<@sUM6ZY2G_{U zo5{R$0BHq5iE{9{SEtHTRPkk@VK-HTaqbAStqsW?3A5Xo*~f4vE@p-h?R6bP=*au{ z+IL}K-S=$61hSIr4KJq(yCP{|=bPN?nz&5QMchzqu$Ro)cE(8nWNFsR0WplGFD>O#ZU$CWJ*R#+>1JkG1jivd)hDuhl;6z8S*90aDyAi+F+ zUsFXVCLsbPA1_6$K-4jKy;D8Aq=qwHKEof=ogc7?N)GTG3eXM~^(pE(EU?d}7vf@c zg%FDgpkwTEdkb6AeC%c0_Iz%{tgV@Ga1lm2)qPe!m0yezu@JW?xGQet(Hyi8QZ~c3 z2ze}{uPn?zc@T4f1Kwv5vkjka)|Beb-S-C>;znsV11SAc1^1MA!f{4^qN?2l{$wSd z-ba>UuQ)Xco8sG4&Ep7(F;uj*Ih5yokyS@+6KKVVJ=#C&`{UAYKa zzJnLEK3N4NyBvLf7O5N@IVjiv(G_!sb?KJO5dAiOtX}K}snrP*ujwmwIn-2lr>$sQ z-j(EFnxMyrhXQAiDq?g;6kOiyc$fbZ24>SIC)LT1DcBom%cU`xvCFt#uRR<)4z2>S zb0}tGHpe3`v6RtvEgxWC%liDrAK)L6)u+dcSpgi1TSmEEF=HiGcFGn$Dc(XrX{)6zNQME=b?Eu&_E@Z8;7$yWSmX{N?l%aapsx8 z3O1_p<+l91#C9>>TzSh}nUBkx3XUMSlSO103P^m0(*Y>X=mTM2`e1PYCrDzZ3u@qX4Er)EwsyiDM6h&xdixF@_`cQ!qpVLGnnXpcofRMlfDLB-2twvgIK3 z+o0cmXnrZGv2Cb@%C$kH)j={%)U4f`Q9DC7VV#-(sCXIJ4prJ;b55IXn&p?D2Xk2* zQj8V>|70ZL@##pEIQMFjac$o&A$iH(l0FiH0=m@{W^G+v3i@QZXu|3y)hffv0|Ig!U?km zM22v)%6nif=H9Ql{BV372Ug~f*e}IPBsSW#Bp6X^&VkBZqs$~7PXAOZOadUTDabCk zWe0hcjOk~%*TCQ<4#uprrYL!02HZ=1y1!;vlp+-?4A`fLVQ=Z`D7$G*`v*v9%ZMz} z`T>`{E!h^RUW1@qJf~wdcfO@}7eWun6`?h0WJJ7JA_VN{F2jC<%8nkMWZb>%di^%n zLn`SEd(4BdUQuR7A|RdBF~3a50QBW$09a!_E=+Ft#qYX>J&qC75!Kucv~Cu(JhC3;j@|&0biP7GF_t3+))WYq;_6>xgz0m62FHHVCKv^}i zwtcgnzTxkiN5UR=%=-+G#+##ljk&cKnHh)(tlZ(E`S%TH)1Y^L(egY%g~*P-8HtzZ zpBOV}km*ec*Bk#u*tod;?;C${aIzfY1(o!XcJJ(NuXBtaQ2uZa++E$NaJkz)3KWU{ zx{mHG{XP20pV2wRe#fGZ5=Y`=Zv)f%B0MF)^#8d@?Kkt!chx@!?P=Tb)fnhbwf%Xa ztVND7HUwU0@z-@WYtzAl_4M@{NfMjri>J(rPtAX9J_*c<;9>vt5y$`MKD9reJDvUU zUW}{3*sFu*e%kY}rwD%EkLmc{=fCA=j+t4#8b?? z>zn@`>-#{%@%C@66UC&!o9$5_r5o2x1XmLy-apPWf2#RoD8q^^*S%UTcli4USHCvw z(e{-bt{PD9Mo}8#GO0a^-`~9V+2xB#hauxnWiCh6Z2J3duYd#BxCavx=^qvczu-aF zE(^w=ztVa5yMs!a6@54(t$$AMF>@nu9;Fwbg@#1&-b`Qr@Z-_4f>yIT)k=S6@I}@c z=k^L4q2=0yKE&~TVmC$lQ3tx@^ekG5Uu0{IF|Li^NIG>qq%~oB>}$*$mxMg_y)r>0;I0iWcX@f;LX~$Edia; zIKKV8V_V5krT;3W!gMXeGsQNC*cY1j@U^DYW&FDBmM7k;u`icfH~zg@gTuS_NA5l> z&OJM#qg6Xa|Ej$7Z3A?1P9JmU#29wqMCHS=cPkJn=fU9RUgY+>$+!>f53gC7@xN(wA5bAzSLFBD-yf^y}Ca)I}bjxlD2Lt^9C+k zg%kG<7V$4R6Q)j#y>@aSBqNW#qqBbV(KOXCY*0Iz`^VjC$KtTMy7V^9SMGM{GO3~%!X?LtAQ<$*iLp)J`EGZL`86i@XmPb)SM$AN z!1Ajld#5*xA}Ef2yL|sSs2YMrRTA`A_<@9lzz-L(uiwy5G?2a!et$2fvg~fPvj4TB zV8ki1dSWKpEZl6D*kDH>3v>o;Ne@NrDv*l0sP@)re0cpgQ+}$hm zia&Pj>W4UgVy{VmoYRp*{n13^POUTZ^+B;Sq!EhN`r3+%@WSuT7w^Sf!9J-CU2Zc$ zjiL6`@6N^Or?_C(nEw2x3gwQ!w_7m2G3^^$alh5B!(#uDBFB8l%{nlHk>GSAV13W& zZ+t5u`5vrGQn$2@a)}) zO@|bRO;wuJ+xSfn0edgTX}u)*?p3d&kqtG}=ba9|Ml9LK9AQ(~hHI4_?+7@J*mh$D z`~K-W)h+#~m(3Y=zWR5n|4|M*w)qjKZ`-z@DxXGK2_Ywn1s6jncb)GxN#I1<<v84O4hi$V0Er+b%w}yQc{7Zw?Ze zBQg7Ka4|}<@Y2nJ)q%i}BY(@)GJR9Rt_f2csJ_u$3Dg;jTqI{s=J%Jy3dEUdA*`TvHl|MGuxPZ|1-JT zpI<)jV4nT>!H0NhIKyMPsaNbM4!T*l@0&k0e|9Fkx?7Fiy)&!u#@*`a(vlHaXvy1s zg&!6l!d5lzUG+pZ|Kq@a*(d)#>mk>q?)IzlE?D$Qvy)n@C>gKl`jx0V)x(Gb^_n6S zh3HX`d)@U%9AA;G5%W(R{ax4LH`myaFW#KIWB@*OPqt*w0o0CCb?2@BW?e0#a|nZPK8@D5S2zgp z;JxYRvbkYXTK$jZ|IB^$;f@n`YSeF*ExvnI*7albF5msuFPg)P-S5SuNexgLsr0VH z)Xd3$0`*ee8tI;BqY+_jn>y}y6@_RKxmPd|pX z?OPmtRihL;of2N7X9B;9J8+R)zu7+hP;qDJO%{09y?^}L#_&5|r5(N!B!5-;?pp0$ zp-kY=1UwHWTe*WVKQVTqb}@Sl7iWabx-@Bew|b|@WT#6|8~w%De_8LJDcb^%j5p{k ztvolWA0v8)I_C`X&ac^<>;5C3ZyUgLu=Zm2q4Q4?_*LT0Q*=HZLQ6DNqY0z|c`aPb;Jc>V3Mgx$Ne ztmE*kt1P`qgHHNHT$^3OK<^(ab$?A348LezwIg@ttR6v$@gU~E{+3X|#j5L1Pal4~ zaIRpY$uf2yR;CYWiU~kg91NIh`aZ_SI|B?TFyNq#F?F%oFV`A9+$W1`59{##Jz9$tN#x)9i^18IqgKq_`~4Xe(81U_(2->gb$KsWU6 zTi?wsC4#L5aSY@KIWzfk7QG|vO^8LdBJnHx#|N(SR+tQTbs@x$6`}(*h z^ud!f6n0K&%S{S>XQ+10KYF&){hRGxh2#@2gz#OyS)qW8?^fR^+ET;Wg7WV6D4u%3 z4CFf-S{K9^4Qu;exEB16>1=cbet92~?6{yfdSYy!Z&uj*y~5uf92Yv@sorn#^^}`r zukh-^dH>!eT3U;r+-9!z;M2dY`WKt*gL{kj;_lk8x)0VH_9i%;+vU3nyvtA0a9J53 z-7vf!RX?@vlpwFG9qc|&d{i#MsNP8U`8xW4*t+h3CeN);>#crH>H-JDjWSha$%qvx zRzQ?J6J@F($c8`?tyL;05h)-uq5?7_AOr{`t`K%qMnFa+j0gcDgb>JjzbANGtzQ0A zc*Z%uv(I_n!v)7*t~|WJcuL#~lyfoF)+#aVU|0^1^b?`LQ`U;coHCuRu#46@DOz~Y zEw;4xGLHPg;P1bdIddzUl1>S}uzsd??b5;X3wN=1w9@SSA{o2b_b4B7Xe?R{BBMU; zk5@siQ+p~nF$$Lk=pW6xbp~SpS>^djC29Dh6P|Oa;2|ZalP46Yy-t-#>PR2OJciT%9R+XZ9+mt;EDd+kU8g z7kfyD9TBY%y9Z8qGxp*qLe3o}_J^PNRz|c=-;i+~$qB=z*H~=80G}J;@+r+;WbnoM zWZu8?uuD(w@U(5~78v*D@Fw-ciBr7o-D9QxC7fMmUBo}C^aQ8wsPsWjz)!qiqI+*8 zYrSk*Khh|$Zcrm+@~fp}@;!=ovE8oCIxJT^`NcHJT+m@Mys|E_e!8Xn)xX|v*a8;h zx>8j(N~+jj1tl`ik2C;tLFu@qYcnp=^ZJ53nN2$Hq$>@7vr_vTuu_F9b>DX@h2GR$ zT4@;Zvsw4wG?bA!w)xkEjAC1AetAU3mw%PHzO}sZXiNaRI_uUyTssz@I>3+j!B>;2 zBS$MEYO80e$n(qwCQ2)yHPFpj1KpBLEv=_7!)gxlB%`m$lO>Ma_{jfI*=^nQNE27% z*>*<%&j^Sg@MNR1$KdqCf;p=AK$K%Np0N&53Y zt;EIubAabGB9cp?YdXX;qRP@CWwt1{H+W`JKqjLZtq$=Rf`k>}rY?`raczT%nPW`x zbhOBQaF?#N+3~Wu4rWpH0I%LACo#4=4p)-EQX(e*F zEwL@vd|D{rwUNl?RtIii{cN+JR)Xml@w__jRxF38asTAv2lk=Yjd*KCwdjW)MSdcf zEHqMrB~}GBCA(S{vG9JUxAN}vR$=_5>TjF}b|P;FzszQfv9=*j(WS+5{sB-H+-PV;jhA6QfZyUVgBXfIn3e8Wgho3;3 z%={zoAMD%!$@Z;sv{sc>$3xALwZ0q)6)u`DFgQBV?-X;%$XMprd#Z;w8?aux{!Q1b zwm|^%SzbdEEBA`524+HOgZ>84yG{kgh%s1L&@i)P_`r_`f4564Xf#k6{Ks+2{C$Oz zd!1kHg-SP{A_@l1!n*_63@giIq?L@}OkCTcPe*YfW$Bn9|C)sr1;eLU&1(31{p|VP zZ2}}7&mG|Dh;Q;LC9MFH!;V1a%SmH|Zb+=w4M_}bshe7YXQ<4J|8wIrbD8Q+8h3jy z&7}9yZrBr5%Q(>dsNan`CfBf&$&ocmKU}@@3#(P>QKr3MDLM9(2nZtRF z*DFRvE%&CB?uQXEu<($SQgs&jkdD?0UjryYw%<)u*wY5@_8X~4x~!PF*ip7xW_ilD z)b@z z^O`1QVz^JsB4JOT^kT8pu^;}|dj6G>y3UW^tF%zq9u<_L(KH|JD_L6>CA0jm?=BoZ(^DC7O8&`w3qZn$p3t*^ zh_Nibk|k7p#SpP5P;Wd>4x;xh)^Vvd*7vH|vBTQW@4GbNs8xjBEnsjG|m$fIJ^7|n2f zESSsyP(ygO_zrs*V2qqJQ$(H;Yn8TC25!`+^u-|Z0f6>c#of20EyL3Wkvio4$vsDoIef>#UVj}Z-7Dx4_KzQFJ(D}SQKH6hT4(s9A{XG@5HePR0D{UweOra=m0-0ZHIv+yL=ORK}==2e$<1>Uujvu{LzG zXYNBcN_-5tMtm5^puhjN9cuPos=@8I2!9(V8S11*P2@iNbG3ITn#*%WiBGgLFw9RFOd%!%uC|i8t zrjEy)D&&}M$XI?M`8)ROf@4?}J}(D$f?`|N^rAH-9@6fLU>MEy#4}8EQL~x>nuA(8 zD^L|AIcTr0dp}YG6;w!*q+!AULD`-ryc?Gl{=#g|PzcZ@b0e^A^4hNM!sm8~6(bl*-VgKND~+(x83Gx0)~i>|bkp5{)i>%3NJ6P@Aoff; z-!B=%XaR0%hn^J!SzK0t7LF}x{j%|TcXd4KRAC7T*bxj&*S)oNfF8gPIjQ6X7h08F zyjq_dXKB`M@RbE|Y75XT&0nP<9GSQzQS+#{E>$b{CWzKRaqt^_XfFX+2?bFfGtN9v z1{?wVt(ERK(MD18=*R-*2U72XL#?oVx~Ql3p#1B4UXoF;(dqA4RjgE9vtSPZ=&HRK z9oRa%7}JvBRk=V6Y*j1O0{hlDX-dLt^`Duia0BDovg?-AZ%i^xN98J-<0qNJqGeb31n6+!V<%R;{;rcA>j*LKW)=N<3WSX{z9 z95pGh=m>dZ$WX{oYOTdGYPwiCKzEO~H$!*cimC4ac@aq$2oX!VUU%&q=Whjg3ne2r z&NSLrYGtOIh)!*=eFgx{ZZT*5kq71Vlol(&+d00Htg>;SeJK&SeN(9`Qs`(~|Jn-R zLg?kcpUmLZpnmh;_8|3JBYUu}NROp_Pxa@ijG`hx?CW1k z0EAG^eLtNM3;Qrch_fOf*MCu_-^`mSb;?76YeRIl#LHgBS<6H?D|k0xgRhn;j2N_&674nVJs zNmKmRS|m{OXbr%IL%CxQ zebJh_hSj`k@`1M)NE$#$?IHeApvC;#-4&xu6^{zT5*U|bL!EC9Myu$qK6BzbEH)%} zRq{{E-BP+YDsu3b7{zp=0Tw{}qP7XzJV~)zrGg*!c_2Rix6ua|z~Bp2F5= zpsxWdcd@_|>0PrRTjqS#<6Dt^eh_GIg34+$p^pJ{t!J*c8nbwB#l4WXFzB$ZaW&wG z^}U?jZ#Yya$>6;Sg9;~B9^sERAnDuhL7prv z9wTe$Vsw4<l4ql2hYfQRs@EYWnye+ zD~u1jYMN*Pn%I3yll^|cCGsddIrl)Fh0)f&2gJ)Qt9Taz|Q{5BwW4}027~MK zAoD^fkxLuCE3-WH8~;Y8He=vE83J@v-bl&WJKF}%s7IV>B<>l@$E3v5!END0bPCi4a6PaLQ)CJwIsjZci>nmJ+@vRNM8UWyr&0l3|0W7>xw6_7Kucxaw`a_vH z{9bxK>9dQg)Gzh>3l0Ida1x&Yjy}MkU*^Lcq)t}Q7MVvUzt#3aqLZe{QTr_}yQD$| za@z}|Woq40xAbJ2Jx5(0+9j3Y?OZMEgq*+>u4~!&5ULX&#RWh^+#B9e?=NG;O=P@GKHLB_1n;*etA;C3sdCoSXM+<;JML`rh4ceNa z*e|G++&KY>f&P#5M;>8^EToS(EpHSk5fzC)ZL$AW2Q&f|0@)Sre|7>GBx|?&s3}Kk z>B(v*U<58r3&t+P!%%X7HMMw7&$RGvYXF?QHA9Iu%oo4pi6E~_Mtikg+i~&6Ki<~- z7MzoFJ|NgYSZ=cekx%Cp>g8m&+kwoIno9A8A*y zogSFiV0)y6B78k9^Xqqzf`62z)OxWJ}eZ+3xjU@~r>4Kt zy+Pi_QA*e@5!H5h>n?VP=y8w6^C#aP_SnMqXgumd;in3x6fOE-T^k#E*pRA)_)%pbxHmDSc6E>lP~k=JJ-L4SbYe(-t;@|H~=VSPq>+Eg`*qd z@mOU9pzHc2KawVmJAl!~V6CtoDC_={cRc?v+88TyF#TJFnOOl?z4g{{nuio1Zj*7# zq~1(gI)fCouuKVE)HAmONHslIFqP4Jhpp$?;kob=@B*##8xH;2%^%692Fcvs^(`z} z0CeOe)minF(Ms8P_n-&BAf$~b@LQM6lRxrjz^Z5)@qGK8MML0^5wh9%lL_ZvuR)?Mo%@}4(J_67b&?i{i90>7CY~iqdN>yCVT_ou zRB@RXZ11crgv0zbSrMp&o}@IPZ#K$f22%G>4~C<~R8ii@!if_%Pc8{+Qmyrg6C-yP z8y2*;e9nPQS|?U2n3C8;GMk5A_$#C7W&4JxR0ek%54B)e_MFT#nzD+e%q`q z0z5K&;*`89nR@akDno3{5DWb9NaDzryhbVmz02VA`owpid1%82J$iMAKw}adUAOta zEqI9=6O;q5O)O@?FVP6%SAo%D3z$Tl^i$T}4U+i*jL2FNR32!cj*zUy95rB}xGg4g zib>+X8aYrV>@K(-oVQguWime$P4yLLdqK8ie&iY2UaIOYo5P<$Jz=tbeWG3iXHG#2 zHW>HcPXRTqq+$ujc%suqW22f88mTW{yp(ut@7d;NV}Kc8AWIC+4qJwuRoCX8D3%Fi$jguha->) zsmg(7|LsH^`D2*s`T&W}Pj%s9ARazi+=uEhREgKO7%8)crkc>X{x2CHGt(i`;saOg zJ}_^B;krT&ApJAj8i6(k&s;3#GsNyqAmH;vaBnqX7<$T!2Fghagg;Lfd^QXPLa^81 z7IH^OohVZ6zb%iuR1{!?KOU2H4Ag`NmAQ!e!-sqdfMw9S<9_JrQiM~x9EAVjFr^I5 zF`z7ub~!SNsS9*q-+_D~Kz}u2492^uE04{E>JUC49}kJn{jmJ*Kekw3C$9L*C8hPo zqqV@?8QDK|Rs~9-tE?!A{~-|dDN*X0bb&&^GCa^NXBKEnarg|@eCA$i{fGvgD}fR4 zuXZ})XjICW?EFJVII$KU0?U}l59|&RCzdV#(i)(tWXX7CAZLo4lusC3W@Yw%3Ui(LqdYsrfVW-}>Om zcRD3KzKe+^JvdfzuC%zcK4NlW&ke3L-b*>kj~T;^OU#wg#Gck2ngMV~ahxRE7%V9( zTO3r1oJ>qDra}4lruglmK+4z(kQ*Rdh*hq&+5;sH=P zKxC@6tKS=C`hxL7(x2{GkEQXzY(H2oA3vChIq6?oy+RrD$!r`g+mdXKCmV&oZ9xv8 zq6-JE2YUkVV3GNfVUrD{DuwP)HZJdkzs0do>QLiCn=*d-K6J>JraRqh4{|>NYlD#| z8aT_hP?S$)MdPH5!nVojfj~33kY-vAH-sb;p7982Qq?#Rk$j2!#%RG+Ic87z+0G@? zlnWAEyn67u;`rlOrRbTy`~-H!6&w}Z^IKo)d0BK52i@cfSC_}A*xz;D15zWD($n1( zkaXHhtH(%KpOq}>iCPgQ@VB{$D?oqqLl?J~Am*C=9JN5W1zxy3CJCL)tL8;Xx=vu~ zwWAj&FMD(tf^dOm1qN+Q74GaF%~8X`qw&&sq+Q8fbW@t0vdq#~tZ#FOCpK|mAYc{wd2%TwR3Y64mC^`| z-a8>;hN)Io^th;q!XZ1gVN}CnZn?R)Vaz5AL9|=>kTU0(LWdpJ9k?W3#8jeUfWrve zp7GQ7v28+^v>-L>QMr&Pzv*1I`xr*;^lw^AOd&)^P9xeG8sWBf`>$B@994m6_;zyz zSwn$^aEN7H3rdIPr4YA+T_flNJ{2}M;XvwQR%E)Fdw+E#I~0S`hT41P;H86+Be~<@ z0&kQ%>Cpem4k$b-Dq?G8l*mB__AyatgW^VHRVK!wcHl>6O%t_nTd6ae?u9eY45NjQyof)oI8%VK6Vi|O5X%};=TVLXeKVYh7!S5|xD`*#c zDQ|I#;H=8!VBI(CNvombCRjZ{Y3%C`wMzr@oc?L{-D1uQ>v*dOT~|$sMPVDsS#QxN z@WR#yxNoIbA!8|gi_@U#iQ|Dw34H3%3~vzNqW6XC{K>1qoG)Z}wgSG*2u&tPdf0r2 z)?_L}BC*4uDwJ5Vl)PvVT+HVbgbRJD(xG~gb=LY$G3mw?6FSI^0x5#VbzAm=g^TnM zF_%#iXm=N<%sbq3X64MYxb%`SM#=;}@N@QSpwGW@!LOi7n(X=rmi7Ig%Z5B>c-6ZY zr36&C!Dq`;>%1oeeqMr47w6F4O^Vcx1BIggR$PLWve*1H*Wz_x>?_b z7aw}|mk+l8VAc&lo)}PZ22^g=6Ss9r-O_tvWwSBZ{D)y#lMU9n_X$y5Jm*?6DGkc+ z30=>wgeMe!?g4<{CpiwZ!MAo()2TJ*HxK1_j!>sRlJ~5KosrA-5 zPtJs9f~))Jl`=rC7}Nww#!|!7Cw$#=B69J3Ub5g(wvnYCU%c#?O<8d)%x#A0IJ@&EY(sY zh5~4A%cg0aBDn6*W9fP5J<=~c&IVa1h48aWx%(!`x<*(07)V+%K~~*Ft;-i`MQqPl z-ed7?K$Or13=p-0lFUj6z0m9*B)UT&Rf{T%KWmJ*Q12GwiL|sIF<(l}uOn0+3)z)hv}WQi)sL3cl3UV9R4@L0F2F^BYJX^?rX3x`Z}ct zK;lI*5ihN$1^r0@wKnTK;M?+Q7l_iBMqQqhGy&ah1tHn`4A%b{h4E+WIsNsbhkdj` z!R`F)dXX`KV=^OzDRJQfuj)m%wNGU(sA#|-+2?QFU{^2&j za!Y1;$i?ZAEmV-x2w?}sxF~e*SSs)~kz`LUxVC$8L0(~4%koZ+lt(bERmgu#SV9=D zd5TCNG2821@tE_P5O<203VSA@F@l7HCFD8@4N)s$DZ8_VO`sK8%4Bi>MlIn>Vk#$8 zxFY*LW$b@|5Qw?xbm5cP7~8zYrJAt+z#t5BUN~@AiZxy5I|m>KHUMSi=lVs?8Xr&x zw%{@JDzYPA4SbOmWW<^AF?ExY_oI=|RDEE%ow2?FcQ1C0I~)WfAF+ zIIt92?25=His}scnz%X?=N5xECkscvvBlTN@U@WhNi4ji+Yx%$qp`nSuJ_be#Wr1_ zj9DtHRcX53pvjB{>5hzwGTw=w`2wx#>|c0P+9071?Cfix90&zvsi*gslB7w{9`S>2 z6zbPYYd~E}(lhr&DoB>te2h}eoWSt)n8Ao{m8bE+%1X%ijn~@UeSx2^#Lr`eOK~U!Bu5KT_t1eG-1Ga(N94x6S zD|+EBt;ecRgHNW!^KZ3brD>tGk^4xWfk7-a&tL-q7d`tbfI;L1R>eJ3X#y~ScHbM*EFck8~xAsJa_}_G& zWDevxYJvNPfP*k_G3F%(t^qf6)qsca%Gg?Wlwqd=^{EtPcN`1hgzJO3;@GsC%>F9i zPAg}FVARuih4qaext|Ss)67y1C%6Vw5{o*bdQs#j7xa%3I0Lnn?w{Pk0IB=AH-;ht zQ5=A5NxOg{P6OwyMzLyItfM9dM>$b~=@Q0lX9WS>W(!TuUPg$Q;nDGt>307PI9X#k zspyfvCt-Z|i>x{wfoBQ1OTCdIJ$Zb!le_CpGTjqYJF>P2k?9E$0PmhdqzUj;RPJn*$kmOqjs zj#t6au4{==Sc)mFT_D2sW@C*Xe*Ei^eD_Ru?O-zfwW&0qz29g{b0-O>qEi{~HYv3v z0CjeY2P0#$$i#N~U`@p0+*Gu6%=-=+9R1oBPmCD_Tl3g(?t53Q#b)Iwo@+yax9Qg^ z{B)D4;_6s1k$^-^XN_Il4FX!^t*|W1&yz?SiBj=Kg8TV#9>>M!+JYBzc)02Qdfs&c z$K51^H9d-Ofw&gkAwp*Yj>s#E$gbxV((C5CEpX`hB*sh#36gcY)q#=w8l_iG+1$8S zlj)hr6Vw@sFEd=h+{{^5{1A#xu9OaPoJstocbGwnUbHQEd4B7zREU8jVM4_?Xk2>l zP*#mAA_{0qTD^F(#1*LUMMF;Abdt%Z`Fa&}mjFkHiv^C*?~!Q<|0&vl8(Xkp@`}V>t8mu80VL zWUa`FL4nIay{M6=LktpzS&MGc0V%@qi-WF#o3*^d{&mS!$A1}dxHbYeAb`eV16YOe z3dKi_>5o#JI*6Yeejgl`;tpaOMre^QJgRm~*bn4Mphe&e3Wk8*f`*P_Mhl3kh;Xnq zES%Nbf+dIWy91-!DhPwDB@K<3tF^BGGd1SO*WQzUxUAwC{g~5c-E8iZvGsO9+_x@x zMmv82e| z94!d5Cb6cX3E3+t)xqsGqT=d3>%P{ySJU^#DaVf~#1*fWbBf(>GJ0Vz`0KiTOCw*$ zaswFAvtvNJp!MBj)O-_(shESBvDd#&F+K3P#iTwohz0bHGO0ift^h+*<`?V3Uop@) zmkzCypGCNNfE3O41dafY0Y4*MsV_oH8EY~0uL9l^=R%osuhaEFED~t`IPNtmL|6Gm z_-SzBbfg%)aw3})!Nvy24K^=&p{5YqQeV9In*C=gPW<5YHZwReOrOwcpuFX$CknZ* zUH*|{{U0&RSm=#d%JEsd^`)8N*+4s;4%+%}^6FS@IwsdvAm=P=w=}o*sAClU=Te{J zl@rIyrht0G7~B7;rjiaU2X|i9=uo`HzYCGz9{&gKAr$*Tc^AIx4}zREUFv~z3FA-@Iv3=X2Ib7VRvr-<1`M7_20R$=PCJ^Pu;?P z032Ltl4f&qNArLyOil~6{%ia_=`|gH`O3q9B4vYd1uEOHoWs(!wNK3o^`-S&K&vg_Cr*Sod?GJvUOUIYyf78e7wYPNRrjpLa-K2}bCt$AM@qP64XjdS; zu4!pe=@wg*qp*JXMD(jN|>R zOB<15PIj-PHDS`e@#yihQ``e(;q<9BChaL};0g+0^en_N4nOr~%t873XlJ&!R8C^6 zx&5Q8f#2lEnhQpzOYjM)#6Q0dIx^#8k(-WW+ni?#V&;!RW0{_ZuFPtuWm-z-kTm`B zDtA<~GL`NP(swHmySg`Rebk6#aMg8FR(4-ony|dO76-k;efpH;s|B@Y*#(tmo&ZYS zevUx-ZkIdRr@$_5^TvEVxzTy`{}|ue#m;7Bi{GN-UcSF;;unc>V?}#ExP8_uYl06N zjA1>8N~F}fYC9hV`*e#FE}S0K(=)3HzEl%)7(TJN?zgXN|j8Jx4pIQUAE1>r_IDZDfkHIcLq#Jn&aISs9oj#!+YnUzcYCmSgx<;)rEkMBG+E#0sf2G>gNSHAixu*Xgj zv4r-_8^0J&&*LH!VpTTp{}F99fnEujdaA%e}OumnWJQuO?A27lXEWI_lJwI zy1D>QrcCXc&~)%hjtU{$7RS(bU-inC$~x?o$Rp9N@A5qPtS`Nv_ud=j7Tm_axEgHt zA+v6*k^m284VJ(do5q{ZvQ)@(ICN(Q)xz~(od>5Wrb2mp(jl+2Pf5W4?elv? z+a_&O)7Jy)f~OQQJs?<7cE&M5kJez}V)-uz$ZWq$zxvUB6{^$yY9&6L)f*mQ9PeSB z(|EL%@#03M>*cCkdL6Ik*O@nZWfFdU+*>fwmM2ocVDW~ zw=LdL)5DWR&5!7V>}#VuL=-eTgqW8$YfUnB19YtdRXGmJRi7=S5p-VCW%V(bK;8Q2ULU~qo!9qlA?59Fz~sQLx3AdJ7Pu* z4qfEV*vKV)f$5falVeN)h&F~mfTxTrk;6n~5#@_0Yik^IRO{8DZv$Z)Mf1|i2lev zkbhXw(+2?J003e)(vx#H1po4F_6=}aI=IP%o>F&Z*`ji7 zilGyILZ7=!(&cQbHCBwls5DPJW@PJsqc(M^yw(`KAGn4$W__#o&X z1>N~%vV`ReTgvGT94;>^TEv5PE?ikVK?QGnO*&LSCbeTE9y-vnp+FO?`?AToH@D=S z^BZu@KnGOpcSz*Dugz-ryIky#fWTv+)QNnn3n4{VFgCF%@eG7sdxAcB(d*#0v*KiL&lsG6kcc`uqarvFF5@>i0DZHekBdn02{T!OcricIxrNU2HoT z7qQ> z%=*`t%tv_Ng({9+h$zD43{ z`h0s0a3Vr9BzhP0x-Um$bCx}NBW>S*kWUvaln`oxThtXX?Ji-RsZ%pggyw&up+tSJ z{{@C~2rcB|Zdo8mfxpQUx-%Ce;8CI}4mSUnc+h#zv7!wJjWl0EZ?*nO-!!M^O3}ho zP>5^Uw=Vh!+hnsd`5?;;_!!b5Cgs;B*rqws{_SJtLxHB!DBnUiE@E_l@S+xTue$1& z;N&S5l!uyj0}%Sznw3yNm%pThZ!FOaTyGt#_>%FTtaP$xivWp18twhi`p5BCB`;rr z0aeWm{mHk*LOU<<%S>-DZ|+w#<-Rp{hrwJ$7w65^$w1bURnd{AV94Y1^)2LCE(oF2 zoe=~`1>W%RD;Gxya+Vl9`&Oi#b36G!_PUiIhrjzCH@q0+hYYi~UHMhX#>+TriNDTPMRX{8j`EqUi06AmNRYx$o2;^<$^1g$mX{5VIf z0cfFjq;z;rH4g`3Ai+^sMtYHLqqH_UT=1jshn;5KNh(W>G}kI$XERmhRnDY`F&_2H z|97N!Ala9>Gszjd^LE&ihkl=jk5*ztVVFS zZ8s=w7yEV?&UUkN#B@aA?~*VEs{0JfW8fBW5HCw(k1khiCAUHyJY@p-@cgDBKRX68 z1R19`J-Ls~Z&}iml2cus*;k>j4QNJMksoJoeE^hH5$mn_{vi8}QPI)MPXW9S96qp2 zMRU`ZfXxb3zb1zDE+K&Rc9Sz++v$hNU$2FfWA7ZA%qrzK*a01Lr$jfx`{^QrApIh# ziFLZJ;pc-Kn@cAWOhoMit_-64V>N2F3IhgbevSA80O5T$l*0Vv#Z+pCv4!WRcc5Yc z)wLtwtd8!**Pm;Gpqeh>Yk@JJ60R{@B`G9%zyVyH<>TxB7e}*g# zBS*InoMB%M8IV^ihv#=xP8+D~jVtEYb5^(#*;RV<^JQq=_x*bI12{k|8QVx+5kkCL znp16Ji~4X{t;x1ctO3-l`*5!2*NmHDj`0ww`_E`(E?UR_*ar-LQFo0oC^?`&j&_I# zgI)`0q$}byPUO&`QSMTMleKUXRJ7SutVed;^jR7hXfYI%8}X&3cg4Fv;cp+~rY~q5Dm*X!t+I(@8i&^MczQ~BM{W5ma z_mXiiQ`JTzF{}?4E$wV9D=L;&;Xw5f6AW%duyZv7jRT>R)5UJm8YpiO!ppZ(bEJcR z@o-7w^|B0jDf*LWfPNV@y868dXepx&@F#@mUnZC9pOxk9-(&BUrC_ZYSNv$%s(91v zapHV8^=8OmWp~78XTIwHF&h1L*a6Dqc+W12u601rlzmK_FTZ1WRp za-hiH(X>6TL>`pg`JY-Lhsm+xVm#ISF>ClLC~YZbXnEW{M4A=g*?>T!y-10DX#6 zvF??VDg4Y+r6YbjL{Xv`)}+f)0yckITD7SL1qN3-WVvaqd3~dvnK8)x^IR4zhC3ANA@A!6{GH@Zx4E|3dRn}dd}RPXNjkrjCTSbJGHG+;58J)# z3hXC(QudcmUtF3Jk8U5j+XN`AIjDFrRt`YK()$g0$lw{di?OL-Vo*9pxupdpzLldu zY~_n^x~rC-0)Rr^3!i6xWs5R^gOj>1-d5UhEdoVNP)29p6D!;;(v!pg*@nBr3zOu4 zV>Q*5fMUGjYe70JX@USIH7KA%@~4E85AR+9Pcc=VQtaDc)@@uu)xMY*cE&j;Y+kU( z1Qv#rhef3EBOA|kqb%WGietOiNv9@1sP{`8tws0pL2Y0Nq&OkZ!X1vM4}$bNd3aRo ztM13A=@i$gCppzvw z)GDC*t=ThnXT22dY%UBSQ7?U~64DOGZUTl!sb1LQ_|db7cF4BNcqKU6!DspCPqPqh zFzc#waQ6)f6=ix3xuGdkHhy&+ab@Jd4RYMo=b^G2-0**{UbXG8d)LYzQrGM4=&(0l zw*Jqv|A`p>{m<3E|Gw(s|Eack)fTGkY<&1atXlej3sd1{*uCSnTSjDwJ6T*KhD!Qe zL19R{upKqc*2>aSFPfxJ8^sir{$;#7<;9!ao0hqn^3u^x{o#knxhlEKZTp@*EuY&S z)SJ^stm8~{8oLteB8@Z$R<2$lhr2IQaCzqIJ)RiUo9N;M7w|g85w4u0VQw4l@4R*G z<*m)w&|eDZ@ExYz8oX@ZOE-&!dFS3B%^B9ruvZ_$ZjmE$mTxUT?anD?Qq2n^aKEi} z#>QM#0OUBAzA2*!;ojNHrYVu^EukkW>7cr<^j@NO8yP3V&*#6piddo%1i zdrSIO1a{?A%~D?nZ>jCPThKY`A%&ywq&vL!kn#Om_8^(_R%mim({RJ;7VzFxbs#Y% z)#T^YsXuNki)Y|O3gNvko8NILd(BJvisEAvTf~oPLSm7Z)dIQE&BJ zBX5w^R^WSq!CV`3@&2EkSB4(Pnb$t(^wYY#lbwR3X@{A1`@UuwmJan*y=unyg=Y}mY>FmKGF!Ar(Qe3T4RtN|RKc2XfD0`w z;6tEQNpALeILa+kA`P=YZvbn7`nhk`068;mLAWqcu;n` zpP8tXZ@e33QO`k)r(EnMuk&gjZ^`Zz)>3>Omi7$Pw@5VmvghE2HmZibnJ z*GJRmHG7`mC7lnOG3BsxYKjH3SZwa0ukqAFEMLGj?Q9vSs+)Ti z)RZ_rz%uh4h`771-}jKFF<;I>4v{We0~j&Oq`ls29&kRY;H2@ZX;)El^+>{8_pEjY z{h}jya(-*7(yeIY`Fq0LX~UBMn~R)uP;0#nEqG9f z^=qt!(l68X9c{T<|C!4kou`#TmJzDFR!&5Pj%U3g zq&RvVc9NWzx$SUrt(!mioe#G=uX<<5Ii&n*vZnlL-Q4ZY`o8eDFND+a7PR{MmUx88 z86YUdjQVw&sBpL75pW@lCa->~YVf_N3|wK#9gZ}ga9K8YJ!l+=dig@jEn#ITfd>CM zb@|oR+lKG**j4VsZg>25C`!rsR<){|M0zqy?lHMgw9e66k2<^M))#1_>xR<6&FF&jxns6y<9K$TPEEwd^%~y~Ja_-N{=ca<16p1X*rh%*NK4y$$ zEE}FR1wfiP)D&Imf8KkX!F(t4bhoDh&=oA8E2Q4h{X=G1ta`~&g9}CN4tBk4&s(=o zTxztN2rdKO^Howjd4g$Sb3LWx=+lJ*Ksz}D2^k!z@8h*stJ_`}I%I*vEwsCYy=^9~ zpm;~*5(d}hm(puTa~XFmWUV;0rdv;tA1zBY($`!oV|X}WCtE@TUA;9(_d2QRy}C7M zYNpacw`z8G!4IwlJ*{Ztl;)F<|IwWvi3^8>umc7cdLJ6MfDV9`1WS$ytWr#xausz* z_doAFb7P!yIaNao^&p}~_9&`rfNe}|8VPf)x^iyjR?zr0UiMV94IfBy$S}CNr{}hE z^gtCISL6=e^am%9+2gnkDMu!L%hk2M985OjXW{HFbSfx8N7#oY!jBA|&lu%&NTjoS{_DdMljHs0(1r zK?gLA?@Gv>cS-1$6nV%1kb?-m2*U2vX;(!{74_AvPtT5G)paWv)@!>@MLbIupHWXs zJo%4)I2b>#sYNn0*zrDZG?zv3PTR$oG(=-8Im9T`tfAlAmoIW#hToL{C4ojvyua+~ z1V{ejR$E!G7EjKs4e!%DTFoi1J|Z*O{3Nzm2qR8avL?c2mY#eAPfq*HYj`IQ{H0Ks zJZ%$hrk?TjzJAmEKZGc+OZv|)0}Eh{wa_*LC#gToLsecpA2pLwJ^6gTRDJNTT66X~ zx7iQXbEWQ~=f=-A`BhzxD!Xm5?0M;(^NoVAK6rHLnx!Z4ENeVn9b-;7`_lFoPAb!s z@{%`XwJIyO`0lYW&9h03m(KOJrr$`WsO(vBLdr+zMfDNebF^8e{5&ZYHYhP(+d2EL zgE_m*HL_olJ7Vlo1;kW7Xx!b^$+=b|tj$3Kv;nxu$+&EN1=JX?AUwuO%IStcPj0g? z)qcKiWJgJ8g6`=t**klWe8p;?w!!3FdX~XXX3wB{|6b;X(4*bB(Pw?hncE&Fx6>0#L>(Sm?|da`&cSc5S&)u^-ANS0Eeq`*#4jd{)#9-XV%;2b zj<4+mEE^w?-)t|r##1HkPkHy1wB6qewtyvV+iq57swyq| z1Fwdgd}2UrI_N#9dOxuhV8`;qdri9yYnR?Z8?G!%Hm-S6?ph^FJ7riv2$vj9dDr+w z@b7O18h<|j=7|;Ko{n<-hS9C}B!HI{*~fNoe`@k50_#^b5b`u7RcTwQNxtz4K5W%Ri^E2ibpy}BS=3RHorB`sZprQd7$EO~)5|H!w_1LD{`@nZ zulR1ya73MQ9))FZOHU5BJ?l;0-tw<4ewUh!Y^4pC884ejoP-wb6C0JvCBxB*Zi30k zzseS>sWJWa8)k51*gXCd>dG>klkl!C?$ z-9+|anq1J^tq4B942UNExteoeLgW`+f5<>xt+JrG+cWBp|K=0E`K5Ks-6zhX#oK>tY4Zu9F_tHtAg6Z^NI4}uFwe-0k;PR7q8 zBZAL3TyjlzSG4f|>gbjgKuy zixMqaGM=}7czccb@dK%aDR=uh(1-FVe8>_~3L+MiUpaSChu3-;_~0UJ_2dB1bYklX z-}k_EhoINqQhO;QvpoEni?H}%JNl@Og4)_gcT_Q6>V&U%xihcO5B=GB_2Y95En}^K z7k}O!)^p|ql`LXQ_)+-l7C1#h_9{R!aw>9AmscA0n2wylRRV(^%RKboJ*vvmjO&WI zjsePX$v1liK9UDc!Ml_G@lwvcS66Tn)B)`B_r{)BaWQKA^~>FI)9VT6Pc-Owq3Cj$ z#fQma*<0rKpJ3gc`!moDO+ZUlZ^nLpiTB`BUfX0wylS-TrTq@ueAt04(;Hg?zUD|$ z{R%#2`sM7qXj{{V*SnPm6Z#GwPuf~;VvFfP&Lj<=-RHx(M&X=-F-FQIOJ}}%zFJ~s zEqGeqTzY^l;#A4LI`yjb$tK|`ArUN!hU`->5a=-yaWfBci_-})b12NY=-uy zd^q_fT)Y3wSj__K=Hx6OyuJb$SgbB_=)ZX=Eqf;$9gnHKa^)h%=zY(WK{eF(j;2GB zTc?hK8%eXBAi=D*U8Q1~S+e--rT=r-cF}%bt!$0Cs*aaxSWOQ|_dnk=W&9M{ZgV~7 z)x{YNW&V9EF`_^2P+I9ub{{4wDr`x$WA(!8h2$eVx8aL3=FCG&>&mzsnCRl{XZx!e z$f(_^o1}UiNU-}Nn7=tY_3{VCU%c7*tS*!Z$JURagVY-QX|s|*Vn{t_7@Br=juE}Q zmK!)BrZLK%)`bGMY5XLIh_vV6F%jK-y~&hR_;##r_-d?EwK-?&Af{XUP925Nu>14& zlkdZ*xL`i;ksi>=d9jGxLw4S=p1=L;3{Lt_7sTKUFMtg;ikk79$Jq48ZM`DJdOh$z z&E-?%8QFrJU#2eQw7=I!2&UO=7Eb7!}%!iK}=(+Q*mNz5rDa6q`qrX2jFmkHqdsd=mR{<-50^m z!&RSeLA2?1ME!PlRLf#*-KpWy9yjrXQ~K`YFA|k0b+GmrY&v=yoE3#172~!NJQ{#I zFSs_sH#!FPAiN%>PD{fvAut3&8=6ih_)R&&XJGXtmp3aj*lP>UJ6_zNSW>Q)Flu1) z1@aBd^|-n9QkhRwSfzNH01rJk^R9Y(hzF;cnY59?JI2GU_PC)z=d zyd&WA`&mytw}(kO%bOW#+d`A~bqQ`wCVAfan*L{r3$sG^@(Azou)f$|V!Ock5HbCp z9O23!!?D(vv!p;P6i4b53b@sC3&tY>%C#W8xQdl<^tjw?fUhzR#@Aq`%GJO8q}d!4HlXnf8l@o&^bR zUod^Pm)nCvJOh5+;OiY9uYfqVYw)tk$GL&~zT)%Zk!=?kI$-hfsT=syM$|We?Sebi zdzbF;+@K`TTUL!N+U&cw>Xfa`Sytff(WAj%81d_Q&Meat@EWy4=RZd?y7ygBTTf8$Oyp}7)QLcAQOJbzuN};!;r7oE@;y)_g5<*HQ##KT9nrC045Zvs ztoahFy{fu$Wjy2Eb=<3qQUU9aTjvbC@Lw+(Kf4q)qs%P#EPGnM(9y)&a-Q!-;){pp zKd(w4{y$yc9oJ-*z3;5MIfvDM<_=$?pWE?CkgVuOEEh``&xXbDr~@o254`h_t-o z+)%vgNs_FAF)#&Y@`Df7$N?q&EiHNxCL3`L9`NKzKRLCDXR&C}bNRQsH3Ol@>)2Zc zubvy@of&|#n$h&taR>6VEaM~99i6;m&;lqf% zx_bu z(h2;qAWG^uOntjF0Rb^fFd>pMAnsG7&`H^9|0p1lBB10Lvn6RQbbL4Y#a#`aU_n1 z&Ll40?GKop8==B$>2pThPo()$@8ts>e_F$CnBg7C4mAX5-`FsFx&H3a)fa!zD8Le9 zw5kU#aO)spId}w{KfVcTDRI_~wx!9Ne67`0c5YLcb^toh+2e`-E+G;l4=ND)BVJGM z`VWaZkTc#Wc+>hu@g`$^qbyDhNSkBpIrE50B{bSIJSb# z5Ef~BHBShI#NScfg~VGdFd{UxmIJPu^4i8e%tqrEr5hsFdIm*@h3`9iFta~oDXVS9f845X5 zQrQ>H-E~5d@{FRhtLFP+`xc-Qjvs$Ih_A%=DM!ig8967f<=kS=`r z?2@xzNn?dekhRQj|JL@j%*|zX)-SgT_DIm9m?w|V6ZGFU#l4(9tcZWkwiYy}#-^Wn z(-NK&G=1OzqyO0u_j6K>=0_iyo$=CDLlT_-ko-kDmXT|`Os{4$mns*g z!!f+rbeTJP8Gk_W z@7c7agF}i=w1Ut?Z1zWQB}GSo|m4{yHjYjDdqMq4)!FnZ)Uh@ddTJH z_S{sf8RrLTcOHGE6Kz@2R<(OhCZGHh@}z*bT+ZfCbPSEBXQmHv4TZU@2I5rVy_T(~ z8IR3r_TK#^wE|ee-V$mNx9%X4as7fl%Geg;2nJH2Tv{*JK7D&8?i~<>FouNNN-{Mes4W)!P@GQB4|w6 zY?2<5o7GdN8MkDW*8a=ahQ(QOP3`C0?3b;>rJrXi+XjTSP~AePZ)e>&a6Q0#6zEZ$ zuL+~{c*IO5OuFh=hyul@Y6 z^4Q)EnLI-okqH^d{c(W832bFW*_MsFYkUy1*i`)xX7?tO)DxwJdBsdpV)3!)pKNk*GA-Ek7VHPf zf`SDXvoXDq&>$o^kB7#K5yBYP(*?eaxejebwHE?CcHR_sgs(K68r!I772MiERTNf) zx?(bCJn?K4oyFF=(xRsr*&=$0aUCy{RLfTa7&XmjOIAp{-pnirYP0T(h9a)akOtTTIzd$JxcM=*Xsv zt4Xo`>~uizoPI^VJ!^WlBRd|&cRwWkgX{0uV70P)#F-Ask$vya85d&Af#<_%ZI7EM zn?m}fi%^3xVhjH;k zP_Hd)!Miyy(B+g8cKt}uwB3+^$@`58=z>A#F)tTV#5s;ksl)uvoKetNJ1q~?MA5(B7z5+__$bRivGLF)R^QIZ}j?0Gj~b^WbA*gHssc6S5VSH@7g zAh~iOJovk*-eDV2>?La|qBHrgbI%L2=51PJ5s8a#+dZZPs=2F}Z9JZI>hYE>E$b>}wK{p?D;2}ox;gVnP5!m)AVJ#7kvs24q~o*^jjUsUk<$7hL66brs-(c%9x zpYS|JAGEuHw?w1i3&3c0utQ(*w2Ik0!XYFs#1mk<^=!F~jnz=lX=r%R^UK_t(l=ja z++g!7NzBdJL1btLA$8JwTMN%gOSGGsv+AC;N3;2hHW0soTd=dU;EYC-I8-V8csBc9 z^navQ0Qa@0p7WA<#7l2cCmQ#@GV5<6-2FlEx)eEUz9|mLZ*og1JNMjsw70v9>nuLx zjaKHy4NGxPYWR^ct1-zQy)r!F|&42y;fwVti%S`vH1U{D3cf z!UTyokq+tKA-bZ!ZyA}x0Wj-Ygb?Jfnu+86j3a09RPJ*)j;mhu$Y@BLi=>wq#b|i8 z;3E%Nk~L@f$ejP`nLT<@Cj~tSot&G*{#uVCTKi{K0o?_<=pSr#&4VG6TW^S1;@;h7 zUWq*OXLflfLU|D(rg(h|9PS~!4;d3D8npGU(9*d?%N9>B{R!{ zFnc)`J-hgBN^5o&d2hv6rp|e6xSH_Q*SkzSRR_~mU+aBHR^iVzhsK(-v>|+i(rV2l z49GTsijthfQ_}@^oS^AQ&2UgpjZ_o#1tN{8Y*kKj(UyX*l71+vsE}XU~HPhLm{{=*`9aPWd+61 zR#<0+{jcUs9!L~SdED4$4Je^@4|4Egsapu8VN_)=AX~~gU(F>?+K_V3T!&o5k)LZn zRr2)S(JQ>6!weB`q^4;+qM!nUOR4p3fyVd}Rley-*1Sf@wX8wQL9G}3&3HmxT3(4; zfv+p2191x<=sCUJRPe6BdKGonJe0V$+x2Tm$nW1~$I2k}4Z>X`Bm3@~c@^-nz?x22 zyCrvgOV6(oV|Nd^y*XO@WaKrC%C=|sP4|eOIb%46`yu*>%4!jE%?ETe!N-o(P$>>3pu!WgD>6o>Me^8HzWW$|6%R4%PAX^~)y7W5 zrpt|X+bSx7xylui%W-eC?FR%V=J>P323xjXs?iS6O}H9IIq;$A@{w!KY{gC}OQ@JvqPCJx#?LXU#joln9yE#cSVVFZvd^e*gKht4%w9B;#f7 z2x092;PdEgx5REZ3%(Drd1Q3-un6>gX1OM8cN<@Co~tb?jU~^(2GjUA0+xxn7Uf?g znO(RGX5?Ixh}#lqV3?4`ZiR}uQ({88uPPDiW}L6(H0o`M@tAD~&{OFCY^5@HwCvEn z-_UX&>6}TJWKW5tA01reYTyHj3ikhS1Hv#btAw4_ekk2IfUEd3xp~v#O|@|geu_&h zUJ=Rr>k-dj0F}2^o(%%XaK17@9Uby~j zzxh{KGZZwh=MrmY<27zH$>W`gJy6JBpx;A`&&)ETlGnwKZ-ueLCrt~NvP02r^Ww%y z-r~k-s-y>kM;vi$s?{+P{S}9KL-J1vDgQj1cxADqmL96xh4r0%_@H{U20uEl!Jbd% z)cOiSu=WP8|FTI%F|Nwhl;@QQ$e2xsq9Ag*)!oD)I`yCPpWQLyR=&y9N76S8^ zbG-ET(Q3!=o&yF@$Y*$CBROApvQOEm>ZAk z3F~5oPL=X~#iR8MZgMrpFfvupi&SR=U%!W0JDS6m!n+QC=rCn#peX3kzru6a(%4^- zlyYV&I|^vx>W?2rdNNYo?k{?(DGGSzAu5Dn>|+T`1lktan8s5gFLFA*Mq)57XfD zydA6Jb8@{o2xUXtvoC^=w1Q5{M@HY?y7mx~a0#k4;I#yfwPCV@o*|U@1+83$gHSFw zG|xAmLyKe&p$%D!lS6Kw$BegH;QYA0KO1Q@{5)-Kz4V{TrCw%TlB6k-R1T#>MN_)? zQ(=h%-RdMYE0T!CIAXWM^r?0GpnbEU<;utue6s?5eytbow5x%x?%hjlADk&SE=2nqyEMg6!#fV<6tjOq@;4I6wKQjE z@ojd33TH7S%k3RkG<%5IQaNaqzREl}YCt=WaWT|#C$0vP?Q4)@z*vm^QbfN^75XeE z1NbWHlr@>%I6JC5J}N!$#~y;p;Vr-?w}qijMfT9BXDAkJQy9eL4-rzuJ2WkU6%YWE zW5(3NQ6_myTTq4AKInus9oF#q2jL_~S9I~kFtS4QPOf3Q_6+N(c2v8 z+(SlgwZRWQ+chBwKhoI^(!4iJ=ny75-&W^?{j z^Vsk+*BoI{J{!Aj75M37v^hHWMCA2vyJ6Tu?GAifJ*c$6iMZ0F>4?ryp)^)TTq6#i zz{DVFb#7t^trtQKcV5Kk$HPW7zQ*H3gOs1nJB;rKBaBSBbFv12Uwvp2G6C6hbBX%x*w1cRf&;e%1 zRGRx@>l<~YWa)2H7p^fANBr77FQk^amx4M)pnUNJ@j)Hyqw*!qgt@W4xoLvIi4ens zQCazA&OA!0bjy{g?%b(oSLFoRVXUDrHRh?U2H2+WzwUHSV36W&5bKefMcY1@r+?M(FqX$@pt<*FX8*S zjDTI`1cm-QDygiUL~x}YSI-yNn{v|yY;IY|5Xns~FK}^C#Pifi15kHw}B$b;QrzP`9=RGqC5ls83346S1|ADSt%A$IvPZs7^PHjy- z0k{9SM(T2=@W7pYIP>;CC=~T$_*j-A@i$P>eF3!w?UfGb$&>Pf17)2#gy&HSDmPxZowM`p*s6AE&_S8! zS!%tx1|LL|5c8;De-wn#)VvAoK@moIRZ~@J(602Ec#mDg7yXQE%FkHVfBWCYz%l*DX>k!;t@VWUP6;An1Q_-uSi^~L!!G1TiF2m5+2s55g_ z@gGT&l)21z$4?sIIfs3$%7eZ=Gwk84>c&MgbeIulAgzr_=BjV@DlemBJaChQG(z3E{zMO&reTs}9!l`X}q(e&TrccqG#M>V++m?RAQ?bl)c1c6A z+Z&g^EhSGB{X3-iM6GKiGqZ8uTZNtnS7MPP1Z12||DN|4zuVcQ33Q|0<_?S_dSR>E zTNJyYsq^S1eA4kB6vcag=H^azNG{?oTJrTJBAoo0Ky|}4!pf)I|6mdYszM#5%ZV5i zW7HV0!dR*c>_Ek$Ep45q)~$1I7AkD{B0#PJR@(Cceoqv z9d0w4zERgM2vtoF>Yh;cC{fICd5g-#Zd<@A*dKin$lxudU}Q2k7mlh-{T?#%UU*0~ z+4Bf@v!9Td1}Kr0;)eC*%IAQeT*8?!=99h0qwmB=+pRLN8#cI_8be*cygv3@+J2Cu zNjh|%U!m&j-|i|l^i;mo@0*~V^}J&O4;%BadS)S@=jFKH^c`!-Vrdq>lfBDlQ$bJ$ zl!w`Qa~1oE{4`^;ae=)BSyYo7mpP~$H*9-QWu9SBNAWK;raDL{O4Y2=bD(c6c-G!b z622R|L96cBQ>9giapY1;9Ezgu`FZ1u29?`;=z$HdN4-Anp$AVE=tB**s3pwQ$Gp3p zxEWYq!&k#SyZP^-2i8sZHKk%Y4$|@~Y{JnQ8iyl$3NdEJ)SmRApd4NPPR4(|Jjw%B1n;?EfMaF=5CoSE?R5JZ>aZ@ zqn8yP%vt(_cHU$=ZS1L0bfMdS8v;zY1a;(vDTL>lB7@$ds0`#W!gr`2J)MUHBK^1b zn)9mwNR3!i!#XZ=W23fvI(*-9rNBGcmir(6sgM!n%QbwtUv62ut+~^H*Ac? zBho*chh!Yu-mgFHx?C0983La^@}nv7J~BrEu3;o;wku%5e(js7`HeRB1X&e}$jF;V zoZu!KN!XoEai}QblyaMnPW`#}z+BT#bBrEs{$meUpRa)l_+`D>C#9BzbOCF~y8Eoz zS5NA!$5Cco^izS%gRNfyfzSz6H;Mla)(<*IJGnu8~VHrh)F(Zi?q* z{_uEa&MAD|bL`l993<(DO`18e>3oE{MR?wUs`hGd_IdWn0ZJ-bArlcjSV&|=QPBoB z_WHIXGP33viu5Umhi+jV-^v2x|HguKxJX3Z6Y3TvCMVC5tT%FI|5`(CIRX~90D9JlWS3oF$XwazTW2F{K;E4N>ySB~4N zmzS|_FLN^G2ZS{=h5()>8+S2t*}F=z<|W$~lgo=eNB{QbVHTypq)-c=HBZo@e=;^T z3p^d>qV2riwo6rT+RKNn?EIs4H#(c1N_fYM8wS_b74E*2J}OFeW6yK(>U;gI`Ak{v zNO8LZij|mp>{ZSb^`5s2)Uaup6=E`07MvEwqP2_XZ1jQkU*B1nWr?WCn#0x7uRqTy z0-@!pFz^3bLiOPL*~F0_mDAg}QSsY;MGkKH3Bp65rxUw@7_+;0h%O6;UTlvS383&^BWNj3Ugkp^Q>g3*5nF}P2Br(k5G=9-wHu}HLp)6mEkexe_{4$qVLX9BN zJGoU_8XNjwo{$n5Q~&{K+(_NOKFj(bx5CV1ecxByksV=I8-T|{}>Qn>f#lE-tAvfP%>|Bh&0?O$Xd z#u6knG#uuSUrMRt-;`|k-JbdzZPL#y>MSGs)K^oKgf(4twhmiyyBEPy;4irlnmX2(TFSrgX-!S)CoWR};2pA1~Ca zrhH^vIX|nru1MgJtd~5fyEhx8hEvSFjk8@ht-@QMtY6O1gY6u*9p-^(+BJ(3Q{#N; z$ijfN&9OAO@n?+e@`RzL!$whSobw0P?^{t6_xG-ERqfSwcq6zqfv9U7TkRgV@Wi02 zh4bjlY#c7^2Atf&ybSY#(V%~5CXP8kVfxfj&7NbTp9j)N*`l&4zJh%_E7IF%AhVM= z`S?2ZxIN19>wsQNnZ1kL=@e*Kz`DA07&Uhbk?Yshu(qxnB4<*iTlF2-ui99?MKg3l z#6qZg&}_%kT~qTi7^;n)w^r_7svmt}E=#%Vat~@Tb@cG@N=-Ro`f#$QHnQf`+|$fC zxJOBxQ77@Eqw8~wIJDsTpJ1#O3LwHc>GvKJ7bwjNQH4SOpS3_+FleD-DD}uB@UQL? zrFHMwaCy!pPsCDf5>eCs=+W%ZdnMLG#PN5z3UWD1R8uOl<898IgB-}SB3<6@UQ;M- zsQ)ZQZbo?Ho9z=clDqUR%s7HjzJp!Kfx3~WtKBp^-K{$ebAr>hoWrxi%B1e^9#h>; zob$!wOWb!A&m_hDM&BZ8ZPf2u;)8(sN%qlu-;9yaO6C)9uwSPls1*<`{Hm(^G&zJPUJ=v_%|Ser9|mC0E9r#9^}* zGf5{?gr!JsBns)nr@%dUtHddc6513NZ{CUIFOi1xCZ0RVVST~|wcS=yAGBZ;lFeq% ziakP2b39Zbv~}O~@Go>%@rESGd)X7XvFS!dCnG4K;e%T^y5Fv3@IKR!!1l{$lkjKA zc{8I_3r_j3`r>UP7&jg+uZtNPzmU;6Nkzf^sQfZ*mlV8a z$cfyloZbsu+Gkos`%0V-+;{x)iR_UF3I~~e^JV?QS-ijwABOUlK9M6$dQGl78oDzh z?s9u}hux~gMx}@z#$~e~dpFdL!t|sAeQKnsiPp--bBV_W6B%j>h!_>xQV}2)Eu_LS zVrhP6ZwKy@ydM01^7gXl9Fjz0>}ZY_{(awKlWwmV$*Iia+C5^@ucrm`$|)fRwyq!B zHHcNX$tAW-&wb`3^{gGefJr%1d+Av@mO|x+B{8_I#HM|&7hp1aT+I4%o;f)*)NW94 zTDC}J^4hxaDPn0&jixapYj!t#)a4elp?W@D^w03sZMb&w7)&6{Z#_gQ}5ZEGFH4fR)&)qpsa1xm0*w&tX8gCYEk7K!Mr zO7-Bub@0>a)$QcQXytDtnX$MF-DIP*_K`GWFlM>#isFdIqiN3TLs{i>RR`e!08oet z=9_coD3R$(Kdz*m4Xsc8L|~{+3!PE}%7vcUAK&Dis9F^KWJC7dPaIA&E{heAgtGEm zE@wBdk;)q!l_bsiZ?7tIQ^sEAshO~xi+2ClB85Y#p8^)u`ulmlZ(Xl6PW1JNnX{;q z$|u}E19MHmr$U^EzFpe$bxVD3Z|w%1cg#`biQH}GE|yIXcYU{F=^G~A2MZIh1YoB2 zUEXG%*2d%56b)>Qj&Ur(|9Lx3Lht^)^=2_6)qjEey14mXERs)flWNTgHRp^vyWjnD z#W*(jUnXgH^Kb?ZqXkTCKeG1EBbCc~iddOLa&Gw&&LHD4#?g=3tXts=b(ltph&Sik zyt`qIEZ&RchZJf2*N4--e7HnV+kNI|FOnzDw3|cpLion$z=9R!=(%;j#_btcPi?4M zT9AWck@mibN9I9tIa7#xLvy6irXR)EfBQ!)d^=2CVhsBFv*yl#WFoYm*q`>mVq7t- z-M~`wXL#tIb!k-0qB}zCY2JKPaWkKqr)`bIw^{9nBJ^uHTEAOj9Bu-jb`&CNg1_5q z2K_vMZowmY1`@iEd_8^Jc=gC8QHmx!c1>di7aJ66_Q3Uno2-j2 z?fF@AnU9clRlM^S)z4+I7vFq+FwD7v;aLRN1=mb{A1GXDZ&5>e-PcOPh*^ zg$Z6C{d4{cS+;zWS-o8>v(It|sWK!lKLH4^#^K?VomZP;O=U1gG%+QR69W%XiFmSEHJ^D zWJ>kuxV2aVQ(slf<6lvC-E zGwXM#mm?cBo7F;HVJ<&T%@W)NK!Q+TGXc+(TldUrP-`oK zVlW%@v#G)`>k`WgrfC>RTr3>ug?gmr`@q9%R(C2Md@b!+bF{De-Pk}g(EmlIA2V6C zj`|(4(AP8{$mqM0ONl5T>niOFp(r^mKBt}zMfZ_8`-uL z>TDhLXf*;TosRUFK5tEB^udq9K?#TrD(tfQkJ5{Ek=_mjTNEfdY=~<2V=tJQc~W-i zd8VzN+OO)PtTj9e?DD@>fZkg2u>1WN-=kd0y}I{ zvkk2)U-V!F(80|7)a5dkHPm{)qJ8#9WyxYW589TV7KvQEye|H<);M`G_kh)V%$*XB zeDa7w;Z9Mj&T65xA>j#tx8c*VgrgzwoE%xfY*bsW*1otsBIaw}b}a;8WyQ`UvdzQI z#T^@%-Q9=bA*<;t;eNk`BA3E0un#=f0m?4Ha|k$p+WX??trL$mLv2^#q%3#|6Xyea z;koN5z-u99b))Y=6NWSICUZ@|lIqRLxz1H<;sB*RU+f{p?=xu_25+YxLn%FzS8~+- zl8dUZmx~ej^4>M!;0DK`M@(}E6Em`6rs$>gJ?1Wok^5AyuGm!lpI<1afLFqOIGrM6 zkQ$@AHz%v42O35%t-_T|)>*@ZE;QJRvWf}%Qg6>SfPP?GQ?ca9l?3DX`9vm2KHY}_8HmspT3T*&qBN145 zg75jR>_nv2a8l=HTN5$8=^u0-H`q%zIc`uwetlKnqHEw2n6=SQiPqhl+nmbIYm;Nh zw0!?&mHMcc2Mf`CI5QwqH*KDYVZHY6j3|Q8aH9f$v`YF6bQ-b`m7V$3eVg*7t@WfY z)|3EZJVbU@asD*bsUr)>ORPM7N*uWe zgM~0GcxQYGT>sqinqeYasqE{D+n1*ni<`H_BUy^s{zu$@H76U;lgJeiG}kCer91HF z57SiZm|HV3(-~qoH4D2Kt^2V$z4N&Yqkd01`jHJT1}u(6v)6*K40 zXE#{6{|I@5@?(p?{Fn29zT$UZl4OE4QQfa?$VnM%0dx5&Y8xTa141$eK`9!RCb7;B zGOUZrSW&Pi*08Y0R_1({qCc8e5oW^o%N}tjF<|TuzBtJI%;u)_B+LQwjB453Oj(y0 zm7*{UuY;x!!$(3$Xd1hickE@sAb>8b9OqTyTmRPbCpWep`a)6vlw*SZSOe{EpY32l}D*o6*08K&-J&xIZ zfN|U9Thl{pIjR1bHKnxKDdwHa@P17`0om83p-E5zdqU$D^iu+c-CxWw|%UPR1Y0$(iZ_mA2wt zmv5EVVZQoby-DOA5T74-2g~}m9IeYY8aiKp50(1)%)n?MR!e!=*9HO>5mqy7gleaS25(u$(q5R^+WY+Lbd5Q2vj!i zj+$UdSY7|r^1bW1b>Hfz&ckK3Iy@A9DIOMapvgp9^6%j{qj^|WbMm>bU>y2S#O#v8 z;9{)JAPRn^zuxo=CCypK;Rt5SH(I&x2R|!Z))5vZH?0D*6+_MjjW3OtEpV-TS*t}wP#x|#P>nzA2O%J7^mH0(Hx{@9;48eJrLa84u zOi#iUk7Ui$lQ=h=CSU5Wt3H%G#&2uK$J%g=(zjhqPt1X_&N!ykDt=iBC9x_wWqEmX z0)tdPUG1#i)@G`=ec3L(ZLaH0#2Ci+W+5`sxfyPPHpLB5y??uT`bvkb6Kn>MB_b_H2BO!FZj_ptC9cXan)Z+As81;*7Gop zB~(GB|HjHm$#PxT6zSK&vrdA`91QtZuI?n{oC+Wqb%Zv=44RzQD_# z8|%VJ>-~FTT_KU-K>MU(zH~zW^#fI^jHw0r`0`~>0g&#;c|+S0)wmy9^H&4=U5a;ZJUWyB&Z^he}v#?**HNnb>`F$tQuDR=Va*}hBB|FN|kTGHDFcUA{Lp5JA*spZzlSq zD49J>yLrVOO5!lr-U~BSKO^B(OhI?Gs|7RlKl$PdAXugaxDRJ+;SQHR`ka;8`|no$ zVi*3Rh-GFH;d=8$Q42~{<-b~!Nhrgg5Zj#1Vax(*3M}n=3keI@UQ>Yj=XfQ4xUpb{1d&{+)-I-XR*^>RBNy(!0tzVkVaRw#T~#Uen8!Q{_mKV}xF!&gYs4J#JAUkF1X zmFOm4&yH~s6x+K*L@yb@?|}aC#DDGWVWZM7v^wRtU5H*(MZ-q#YkfqzUn1xiy&U= z*Y;PWY?{4Mq{}xctD?|r!;;hwpHk=)Rq2cS?{fVk`(L$w6gAB3zM2#K-Cg~VIhqkv z#*5yY3p+mbH)D{_%Gk^D_(!4;XR6-y+VaQB|l$L+}51Qzr}nyd;GrR2zd` zRe~<|*7$~$Ra*7ha9`W`v4f9ef1~S)JZs9o=sFdEYwaUAzMMK_RB%`f$>L^1Sifve z#x7K|mS;jHQ#$>9P_&IG&$)`29&x+vypryf3VbEdC$gq&U}OnK8_p(2xbwM<&!Med zwByPA+D>aMPnl(JjPvWk-V094K6d*XI>s+FK|C9oygX^9WVTHFKWm={rcTxmPP}N! zdwgPs@BgUrUD=t1?g@)UR6}7k)6OfyRx3QsOQo$cDV1`isJyajGt z^t7*Fx|8=!%WvW3nX?tvqI2SdWkYVAHQiN1qm#p=5(MFMlhXFUYC$H)A;&Qg&?aEJ zP;=S_3kiENRtR?QNn>ef=gBe?66dc4W`qrlVv#Dx|5$1LZ`gh{X^-dRJtv#iD9G_b zJTgP#BuqT3V4;2w8?0X|KX%<4T2x_{8**4ZC-^$wdxAOi5s#c;fXfuCfzR>;=4!(i||E7Zm+;cUbmS#H}HI^0U& zpmoc}@PCfRKRs`*3&^koNSzEaireU>e|aPsh@}1K3$L2cMGL($6_xY@9_j`LH7W!7@D9sX+U51uvO^!PjYRzwB9sKQPM63H5t z0<58+wW@|UvA2AleY|?lL?^d=5a->m&cle<8a;3EtzFUvcz@n_ydq81G5u*-72I=n zGH&Q?p0BIq?NCu+?h>y^*!6!bo+MDLflQi13Q5c$3=8kpgY0ngj`ol(RH`M##l?NM?h$W)_Mqow<{n=;?lU zZm*J>D@nrVr!LywNLlsN6N;T?@jS9lEZQI}s(#xl-SB0RW&KY(`NYfu|EDA4+hb7$ zTaHgX9dJJa%8wM#&U@WQlC;a*B;Ew|1tGI@qzw{VulcrjVqU!Q)5p?Xm*5^z z>lvSKS+BIay?cE7H*tKLczIM=X}3#(d?`^z#cxIi*3Da^W&mA493S zbJ}@OUeq{^M-2+8(I5f)Kx(eJq>blx0ntNZ<+_Qrn1cuEDOF9iGr-)8%%mSEHo8~1 zwKshgzKLW^Uva7pv+U!=k%zZBmH*|M6`UHO*uTxX!p4m%nCUMV{LOQoXGpU)ZQL8B zPfN`nY1*u!N5A$`gPM^mecp`YZwd+Vd~PwNPso4j`8q|aDLy3jJbPxZxhp>9|FXd# zp;SGnPNU6*L1(rulV){CpWIxxrmpY(S|}6hN!s1nZfe-sj+zsi-=ZeBQWr%Qv4RCe z(i9hezns30$yOYwLfp|6Jr=V0keD{RXgio3H7=C^}6~miBP%u2Y{^<8^{p`+y zh5N`PhRcq|QQPxLNJR5Uk&Y9n&Q1u9z6|aewM%-#J6Q;jTP{+q|J?3cIA+3|VmT+W zK@^CC=8Btr&LQU80cr?-YO)j?dANP8>U`@BOIDBc{!0jRLx5{lLEg)_z0k*$wYgom zdDr<>Jh`dUXG7`WwM$9Y->Qoqj<%_}IMj)2dl8N}z~P&|Q756GN43+lT=1B5erl3Y zIUQR4N$YY&xQ>lR-`TX8cv(!?&oO*2Q~K>;nm}b3#$0#uC<@715c$nm{le~-70NUd zW+8e8n zhcK!7`)grSX*&E=lj@~GMSHD4Z(NX>LBn}!Fd_!DGc;<)bYKE|;nUT{3shPd<=xM? zE<*xcg$ufo0Cj9!AQs~lU&rEjmPE8OH4Pgu$(Al4ZMCL$v9vASv&_kD+2N_pk0NlvFI{dJb`#mQ0vAa8~)Yi{%n8en5b8(sOYKYzad!)l+P-qB>U ztXTVfW|~C6`GE7@bq$@k(GNGuvORIR1z>HR=isvzVHM1IP zvDt^$OF60GWkKy>6EJQ<7m_t_=Ub{;FkEz~Bh*%gjAy>?NID5KGHFBhoM=+0x*O^}Bq zF^-&7xRm_4^7;>}?9kdC-Z`_o=Ig5PzO{L=gWD7`HwE>JAS%d~pjJt#HmV-xZTZI} zosssAr}yuHt~V30Y(;wQ!Q3gHn@lwXrDlf^xREMA9-kN_B$4pwdCzxjlm(d!T}?l~ z+JM!0&J`N=V{Mcy_i=IaP^^}-X%bKx6qs9N!SD}0D6F;q)>=we%H9~e#qI6lE0}A& z`kBR|)c14U^24t%1}<$n-5(Y)+i_il9}-n?ajSKcUXY6fi~*GFAK9nM9JXE6-8sjX>r-}Y6#%2&4NUz(@O zP+3{UqG|2q(*9z0o~89pnZru^*Ts0+sqt5EF{NdVc(2e=1bvmQRt6s65`&YIQ@s29 zD*ql9H#r=@hgQRj9xo8j)h+g@qYVg?D3Y+xDWqx80MoN&Lz{GBU3v(Ux-wAC%q$Fz zRzOX_c*n+O$O`g5&))uf(fptDUK2#!)Drh5KyFlFyLI-#Kb%+FyVqW(Z8K9y-b42< z>!(Eavy>;r;&UnSnB?Q$EUKVB>_bv5R*BeNfH$3!?_+@n5IUW_aMJR)*3#9y36S_60 zEohgbcYfRZxnRk#x7E|Bk6{N%nkJ3WqUJQM=UIZ2MoxsbnBrzap77pok5}Xw%nBFvf ztAA=OH7ZYo@`1W6-w#nROZu%<1=lhe&P}==^sKGqjh>D2CA3eB^s+#`OUs8%wdC&S zQ2&yL8vu1r$4VjS_2v>jSZML}MKy_HgXQt<5Ofvc7Tvbn#4!@+!VQ*sU;mvHc&P$? zq6W|o&{^{;`yfy&Rus$hEnkEvBcrtEzc#S3{6@IcU!783so8&hBy4SBgQ)W=gJsX!hwMQMaXVThVXX-)lF|0wtA7tv2=`-l zU-n|whUXDwmnGKzo+Ub?>EM%?oX@il5?(JRjN$Z3oJKG4-@{FQhrb*jclP;#IpVGK z9(R{A*7G2~7es|<&*(zC9|gf9YcTsuS@2;u6p9n=Za&p|*ji(1bX4@e6aKNbHLazf z>T|mP`L3T9BQH$FyIZfe^tg66wcB;@wch6P&a(F`;+;W{iil0_SFtJvARwsAU%(lY zvT}=qnAMA!KAKApiWbyB#c|nzQBpN)i0gD+dYrq|U;k)bvg#9QW8H~ooTI6U`CPqmcUmiHc592_NQwJ=;%hw{Pvz19VIerR zEQAk@_Dx|ijtiCOho3@c1o*>1_~nLFwSd@gOO^y-&%XTH$@lRKtPc(gh)h7KKVU;; z*I0!wb@a=SJY4IHDG-S%XhU)v=mKWW#$v<5H0pP(5iE{XhF^f89*j<$L$5k-xBQC zPR^Ege*D$6Pvwij50jTfgDz}kOIJ@B%+A^A+RXq^IgU@8S?#6IjZT zULF}@|K6x~;1{Nbn5=!ZT#N)r>Cu_~-&P51d2FSRACLm3-#bIKrN<327rHrbSPd5L zy)ZH4DZIs|XrcEypRaNC7s1`n4w?V?d-HSZym2AsjIo>Gv077df6h|ZX>^xmtpzXtyPsA3}p2{oglJC8kOX|DN`>6g|Rr9h#g}|wFs0hzH!6>nhy41g?1-8J{cx3j-jBoJC^<LRGRt3n4&hBx@3ZA@|W{({XWcL?Gq_??A!6Fl%P7J~%{DAUIEy&9^ zgSMO($gTBeS=5%6{Xh2JJD|y{{TmNj<=e+W)e5q!h^T-vL}b^BiGoUTu!lk*U>E{I zgs@tlx(HB^DFRABKnchQD+XwlJz@~TCMqNB2!R9$A-{76#nHa~`u_WSn?Edt-1j-> zI@h_b&oxhSKg%YyS*3S@B|RK8wX~$;8pHIG06l$1IR_puAsV!~@z4=@VTYYX`uQvA z4S(WpdvEouBVCCLUyvtN#eUa z)L)h-U7{UkUdntPPcYzY*(kz%0UEAzYcxs&4`g#w$FWm%A1-C8C|TdLhq&#aC%Bs6 zDxNDQ)QZ0Jj+-3Ryg?%|r0dewhNO-P5W=%Jk}ZW`T*z3bnV&0|=t?viG_zVY&(oV?BLF!{8h%D7y3#-C?6e(9W9{N6j`;=jOa}&1%t@BN{XXf=sL8HGk zvNXrWIXWG%De>!p`d+v5(JWB5FMN1JqOTbfjN)+ubD6*f@)~1cx)T|F%S{YyT(W5V zWhYHKl(^r*k#esb_e(`<-pB!p5fl}MRPdf&0AWQ9cK@6z_EguW$C;TzD;Ax~us$BhgRizg_G?Ena#3rU#IJ zOB+u0&As`5aGvqIB;QzWQ`QVbMZPjj67kq7H9~!YSMM{EU9jvikkt z{#SI&v)RokTGbEqS7o%9&jjt^LF>a~AP7_mu|(11rO1va6KID~)y$8&?Sl(x$B$)+ zww7%RAqo_IRK|WR6#_PIm4(#zuLc93`vkmJ393_T@zW{obUxu-!b!MCaendqiCF7! zVMP4q>Y_jqRQXFzR6uij7y>QtOTLQdZK7P?2p4&Bl38tP4Wj)|a|-NF$P^nr&jHTX ziQ%QA;IKRX^{{Wu0>#bp6obrXp#N)ON96aAtAPOfq#WV7@1kbqBN3|_)1*JDp*-2s zi-n^d!(}&JTfxD4b;u>?&Cc<9k+q%&W!yVJ)UVbU?z|t=emxUIWP=2?>@1e%qs=xM zmw?la9(REq)Ipyq!q&7pwtU1!AH3x%55UzC0FsMkw)~V?G9eAH()MZ5P_0i+cG94V zm+qgrIH-q5bZYaVjlO5VT^w|e^}(QSeK{kvZG4ZVE-iO*mPD^dsbzukFUol6+~m!w zq$dUzZ3_k&A~qm_YA6Kip(AD-8S$Y$@12WPH zA4>yvRGi%Ov{>3kiUG#my`GQ>`Za*Io!mXdGtAyI>+fW~(PG*_QwZ#{2km;0$URy} zy@;JNC-)Yc8B~JmEfc9Cqv!7G+(bcWW_D!7r~_?VFaGr2;k@C-AY{7&AQr7FnH&uQ zeXjnjZn0{)wLZI7nFOyo^~cMjH*VGHfs%ukqvJ4UL1XL=-J~5lg8!= zNrzFY>ECdzWm&9kvaVwC&(N8(qlXCUCqUQVn;%bp0AA8Xs>Sm+cUItYi3F*>Q0CR; zn%={yzR#`>F4C(%o}Hsd1D#D2fGTxScqM&c#_wPv!7yx=9j4$A@3~ zLb^Mr3>4_5=iUXbxQWHo)BM66Eiw}wr2gOexP(2`G>}rxqbuD%;syc8$RTHT6C=6Q z*;7eR`(q;Gdj^BQhN6fdF{OHTVKm|KjKN+_5xqS!-zNC1YN7<(t7BIrVJ2;n}U{HEF%Gcn8fy@HlJ zZeaWj`L-nx!$WDG8%yLSFQcYeJ{2M$F$vB7-gZ56$+7%({_JN)T*)E8e{E*?zDuKbh`W9?BD348K z-W7&6tRW4FUqyH@TMZ~@-ZS%zts|DlxD~H*JOv6zm4OJA zIgp!=M65F&&OHL%C6gRRk4ArqA#q1dhv=zCtfj4vCzCU>%#&aJbaU0q0vma3Dq_i8 zT8!`i?%*XODUw#m&z4RHPf8aV9h(ws9D6BN-Fn^Uqd##a&*WFtBQwKGjv(%N^exVJ z2iXA>`u?ANVd_sPx#X2no%VX#d?eduDv1gcythdNcL zI7`lTGk{m%ZjXpX#zS+b$^uiJBGi=+qWDyzX5~WjP5Mk+$^F(PPXENwde)ijunhLC znhwx_#FMa%QsN^QA zw@KFA!zm;|8yS0}vl8?Komj~P6IuVE+h0|TDzYybQc~G^Y18b*5*{@>-YqU$y}L)p z%-&}#Lz6d~U3qOhRl)w?Nz{s!Cm0MQ!#=QvYapAimM`nwGytvr(ZKXJ8GCR6MD(o# z(33r2J68KNs5ou3#3(?<A^3RZ{&~$1o`X9ycg`XX09U_Ku@##O|lZkA~!yO*U){y%+-OQ&snVD-O&%7 zUi!VvxM|Id`-|ifvhG4*bj%7E1f5jUBT#h`p&nr2n^sbqmR}#Z{KiddWyrK&KlO z13^m%AelhpQW8btR_qNcc5%Ob>i5PXAIo~)r^B~B&4yBeRgOccV)~H5E5TQ+iufpP zLjBLVl|_T&OE?*nX0})_01VO5&j^7oaZkzZ55HJ^`Vi$A_gQL0K%Ri>YwqQpTGL zPE00`*a@}T%%8oO^ASV6f62v^*P+`rZme;WIJe@;SK0&b|JU`v~mUq`}~*6?epDy)C{%Y1V!lwpdpD9n8S}D2jPe8 zdC=MZeX8o`tn(@zSU2Dv8}oEwhm4(OP86L0)&-GU8N%m4uE5HL2Y=7$qTnhqBB*b= zyEeb|PB3r0`wD#-hax1s6+t&038Bk-?xb6AV)+;>?_7?qDM_vJ5w)scm4WAqf zH6FX1ue|3nz99wlog20U&#OF{!NmC%`$4^$0h9UEe%&Gx-VTl_0A0E~_;b#obl@u{ zLN_{C_^Fh98-kyKO0p&k50#Fue71ZUG%PC2nTg*mjLay$us~`g zOpLvq16lFLn}$>VNouK>oCWnl=3(xtUbP-Avgc_FvZ)$-q00l8FXCHdB^x(rKmR%c z{0Edb=7AC`a{WU~*IFk7FX)Or3$vH|{^_Vy>7InH15z%oZWHuc5@|7F8J8B(e{G%ZI$j6|i(RL2#u3eqI(`wXaj@Bd#C3-uEA4PrTLCDbcZ+8Oe!sTr$ zgsWy3*lEBF@Nro9G4m^*&&-*r6}#YP?VV*-PM+`#&>JWpxXF}hT(A9=2f>IZ)IEEJ z5K~!fa7)&B-T!};PDbQW%+t*TG$G_R; zZ40lqT`w8?>B4ffH|ynf$9*u+o#cJ_@ZVvPrn^w7RJ()qDYBr*;CM10>HR-o)?)X)4B8OWS}$D)C0n~-YVv- zRtglMxt%A~`Ij)s-%HR(F&F$!gIb3|yIB4)CP~jm%HH-?ow;n3&6Wm(!kGt5ElV%lUbve}hm943hrxLiM zt)}CDHm-J7_&d1uk@X5!`VuiT)7SfJWL)Sv=mvEXoBz)*JO2@md|ZK*7i$b9dVzg! z@W;GISre(?c9e&X)+UulDzq=weH{lUnzx~a!DXE00j1{WUM24VNbo0Bf_WJ7BoPuRFS!m5Tq^N)^40ISR5L5ucd~)IXUp0CkFPY!j_C(}%q;vwfy5jR?~fr)WqvoQMU<0U#ab+Rxlz;~a%$HhOta{tZR{8Oa1(taQla|%4eB6lQg{W`gA zqZ7C^BLeCs*4;jO325wKcPJNN{@O45KG^ZFe?E0&xmuP9XaaGe{^6y`EGRAk>~(>m zt8aF(;p;X(`6(#4*gLjFYlG#t*n^hQpnvn(z?iC-uRDC&jv&i^a_tN45U6YxszQ+6 zsZ#xQgljnxIQH(QXwm}f-18URcAuECks=9ysoMS&;jhAuF!h%jV;{ zQJ`tX+&)m-@8K{~YD|6xb=nw^6LMZ{$@-6Chw2K7zJW<*UA#$#5^{lUK(6$Nfhs#A z*Ccp-CFx?VyJAtBc2Y3DUJi4lfyuQ#UsjuZ=D7Gswr9^XH-}uYbfOq#Qemj| z?X~eLsjmb3BbEkHLEfO5$cfE+D8|-Z`ETdP6soP)5C5Zezp8GI?j38-11T!QgP}7@ z=0ggN+N53mdsIG&Kfa&F--(+I&~<^CdxT?A#VgkDG&@uaWCMJKioAl#oMIn69&4bf z_f*&MLKiEdKd^o6+hif1zr-It_4llj?P3z67v$uEX!_*f>zR$`&vz*gnfVR9d4AL2 zNdMNCLSMCO>%Il1T9i|=d&Z4~%rbD3Pzk34QQZU6ey?6YBjOK1woiOct8)PPM%p9? z?su%K@x0EO;R~r-BSo%$9ohfz2PjEN@1^c(C(naV0_SSZ0oT6nVGx)52Sa|9-PbX^ zAn!&a$N`+a!Wx7`9!O11xMuU<`B#aW4F?`%naP^c{K-L8{;j}|{%iPR1Gv-lo4*Es z{+{^4AQfZ^+<>Q`F`>I$fSFBEs1&qXxuy#PNyW+kJ3@L2g5I*HMf4(xbkN0Z=-Tek zzeHL-^O~%hMH4?;$}avFY?Zp|ylzVIj98qe^y2I_snEZKQ2rj@b^hj!o>T-^rbZuo ziGK|I@H8}9(B0wM#?ZpAQxOwqK`V@mESAV_;k29dCt(3|7kedM-~T$wTUpSX4sTsc zeNJcy5tzlL@>Tv{XOIZhsN&gMF~r8^KMz0>-&a5z z5TO@zQJ65i{QKhl3teH5m2}$}w(;u@`^6JbWWxgTR8Vjxe zI+3Sh2z=9-!cR=h-JJ6jL0$Uv0PJHI)_;UHDuxJ;1|~2o^Av=0o=j9joe;xh*X`K< zl`98*_zr>#ALgak(mu6^gTJ{r@^yO$e*m18hl8gUqfRfn zP`7tw$lrd*3#y9DEw{Ex+iR3|Ev?{}a|=<#I-NP_?qiR)2-+@YThE$M;ajv_$Af*w z2tu%`yM`NVc2#PCZ@d;-sf``8U~{U)cLUXf5|Igt$)e;FZcc{9R3v!|H|HfRolPFK>$wvMGgY&_y067Yhv)1 z?Dw%W{I4yC@F~~y0Gi!XjxaPF)Y&G0m4=+#x9Sd3_;ZzO-u3+*$WDj^p!L?ur|#SU zf>y-|3wGK67a+_6xP8m}kN@SRI*IvUs>tc%UxL+(`T$$+djByAm$h6=Ke+fX`!9IC zsTSa0!8fyDp}0lU8&O7JJx1M}Hq=RTMYd0;IkPC~YW;1gQK zU#L7A{Nvsa6PhVnvga$V{86Rpiy5r^+V9JCuDr<>m>2J-YnUQoPym2E_*@hN*1>zC zL}k`Q<_HOX{r}EC^S;Znd_{iVJwC-Kt}@=;_~X(O+6)vJ43SfBzSR#oB^}P~SK{1| zwPjfwa*oD$dqUrDl67149G$~ZGUj15Z~DAs-+e>vjBp@GgYR#? zqhE?Tx$v%}v%0+@A*^P=HXl|ac$=39tZEBY%11dE*{5*X#Iq%ZF3XE3VSj_XouXP; zeQ$&L{sCo)o7VfdvHoZ48QAirRO5yax;1IAWkD8syTW8;umg}5nxvVp$Sh%;df$T!mh1uRy|4;=D6jfU_Y-Z znQ0JwQ;#`9W)3Pwt}DD%&Dc z`5kW$T={Kep)d}Ic{QRljCy^Fnt&d%3??)|-_{hsS&pEfDk$iYr%PAH&g$oQR z2}ERNNCQhlyTvT`+WI&j<$dl;i9?Frq@c%(AT=`7&s|v zaYoiwlQ!MjJ)QhwzdRzRkf2OWh&X$B4JL#;qp zmmG#pan2BqLrKx#sM9Vs`1qPw$Hi2X4Rq-`lJEnDmO4`-TCwhhHN!6U&PN6k$UYI? z6B=ZVIGZ4mrXy@@pd+bK718sIR^!sMmyrK0OeI4w*FQd)t7mS)p-A=+YHEImP;e?l z(h1C~{_4mlDRlXaaD&1nLXqPml&`C?C#ka9`wW_KH(N?k!V_J8BshCZyPYvxi*wob zx@FBdSK}wi#?~lFKAkp2WCs@$_$Ff6=X|>4CJh3wraEtnPIxf z&7|tjCl)e)Ig6#`(`Ujb^|eKff{RtV_ZVpzz(Vu3p1w@;o_#)ai+ZvjaO>>w#J5g~ zBk_WQ3N- z`4DAw*+Ps>(NspigmDAgyniizU=Wq-!F(&(V;iB>Jxn|d?}599Lu=jItlDvQn1tC^ z8_``9r*nK#D_V{hJeahnAmX`5E5nJWD|yGY-zVKc17Rt?w1>- z>gcBd&Ma=J5H^|w%Th;`V5Y+-TduV%xj7b{j`0qL!Je8QsP8z=ZDb1D-Dd|ID`E@orT2~5gI*M z|3O_*6poM+>_sWQ+c%|N`_pX6-Y&*Buq)*dU#c7Amf(VC!3m zMv^?Ap~5`fIF^n(7qKyN990-hE6u;#N{zucx!%} zi>bJ?92bT65#_kq7jy@)Ep9DiV{%wRu+me9vMn2vWkuv8lr+!a;QQNUKaQ4`@ z4-k;VOk_TyGKQm$|1uM(=i}vglKp4(2-n{;q9P>io@w0l)tWV3{S}dy-G`dE4tY`e z1xE%_>h3iLj_Yv=llneNh9P0mL1$nvoa91o-Qs};ZD+8ig6@jix`jB1h*t25eeeoN zEEip5Usz!)1WEXoU?60f6aJ!b-o)Bn7!&7E4uyJAP0DUCc)# z351ELIo#A~+S}^(4Qhz~qo(A&jcZ(IOK=mTTr$swois|;Q;Sz@k+Mf{hpd@hp6;A3 zuu)9Yf)-&|C=bpk5{D1tm8ooW{${!YSjwG|O2lX~HxgI7YmtfGM-)8VZ$i49{e2tw9OuOc%~D7|jo%=RO}T5DT<9|^t-fg2($LTO8IY>~ zH$oIHq6;5n+cG-AO|PxF^W%#@#$1w&)!>mr(5NolWx$S?py^t zLq9-?Gogd+4jNqEKgmB;)aPPk`(7xv-h-vMU;Wb$Fg>;q6}@zhhAi2Q*MnK7F(`mP zG{g0eA$oj}mj)nucJ*MeqCs)UB5Wi?)a_vWRtAc$j)pTc3JXhQ*%3Vyuga2|C^QF1 z;F%mPxuGS%`SiMy>L{(NQT{nDsYEGv&K5a_4Rp%Tmbk!;R$ zmhpN%Qq5}Yu&fp(9)K%Oa=??x#T7UsL)3<`vM;;QR7r^Y28VB7+OHyjVKZ%pyr8+rl~qks%i81%`= zC`?DnTyoUmNZm?+NT|zt;^OxB0VJhUTCb)*rr4>n15wk0YS7{oXSj*AyJU2fMDV~7 zI?)#$RUApPE65B1!5~OO6foAxH;&Z)2}$kxs9?HIzO6AE@xzb11D11!?qJSAT&A?q zkefG@b3B~7EEy!5O1!AK|1mYGHPxDt&j`}l|2QhPb-9q4LCt5WTTe<#9|MGtYPV=q zf_UAsQE9SY-LAOXuBPrpQiyXpu-A@s)CJ~qNQbtzhqbDkTZk+mnfC^os4h$H!8>4Qt(~sRNR=<>EZcx7TfefXx`N%*Q;I+5T4C) z8#V!1Ob$9yP+G_dQk|)g-7;AjtH)4LGh^HIM4^9=k{e)VQp;@Ww+FRhld*Y|QTkzH z$Ho1pg3u&9ucIJylFXY)lneR>_G5)hOL3fDL8S?vM^jV7v>aZXgoFxu!F5pAAP>O| zrj|Gt^;xlVPQnaVx47O-+&@KjdpoX`wUJNr69Y(Bt32hfnq35r?mcDNTnMIRpcZTt z(WD6HI)`EXtk0_(nUD}u59K)HBS*9r-C(eawz~AOx^MXBXsY!tD1q_irdj)#Md$H@h7vFV;^JRsnhazW<{ zYDXL{z#BBQw1sy}CPkCetk9!m8nwcd9g)0?o1&Z+gl}q_Qk=p&%y!uo3&SqT$gyhl z0vKUr)(aOpAbqN=Z+oCFAU)9#Uh(Z6JYApaht^u$v>6B*Rdm@~o_e`Vp*%OJZ)#In5_tdU#jIw#|J`|1~HQ?k#(U5Uv#5!gn8wE!hQ$)9l=s#v2X$*V4AzUdp7CDk! z7h7c5lQ%T-V(eDbNy~cZG-W_gqrZN5#KCix-r?{;1q1$$!g$ojMPa&4O`G@XQ8+G@ zdl7jjJ^47LKcT}u^0uY!npUehen3qQvFMEKjh9OFK~kD4=O7stiJO#j^uBsuN(;ox z7vRG_daReJY1*a%mE&na9o~c1(^*u{5+G46jgiQWvCLokJ!6T}qVA+PJx9?WfOQY} zjf+C|Wg?A>Ok!|8<3R|vlw$Y62DixrfRM7%xi9=lbI=w$GOKBXnyg?TVwG0XZAv&2e+NxO!FQ>`k{Z_ItlE)lW*+Eck5S9 znZsajK<-gbvXc}%#5(K)3I_Y$Ksuz&wkn>bHCjL~>Kdz^S&C$bdC%Ia&e%%rStzB63f|ls@pfKBmNwYgm7zq#uK%NSrOQkLl{RHRQra z2_^E6$#HH&LA^a82=!KbIroNyb2uqabQ7Fw*!B1ozNF8DjA|r8Q=Q}8^VJ1RFn7k! zDB?`6TI6=Hdi?tSNL*;Xc(z#JEY2UuW5$s;*E>n z&w<0B<-_}*i>H;F1{~+Y>n%FPN1x>UmOc|KBb|v**H+ju8jo9mpT{U$*M_lg7x>D~ zjeNNg<@Q3bx>umLs5q4yq@jkuvF(j(q=NFI^p(pXKm!<*d3v0y)*Y-YPOpX?{~+o# zn&M`1aBNFsEjHEA7he>JCW!&r*WjgYNbK-_f3wgK4 z;I86j9zzM@0Oc1%2*U!iU4DHUCs(k?>^Q(t0Sau?j>KUK0Fc;ck;+d18) zG>B^$s|}A}=UOJ?1-RR)pyFXkt~HpXT1A-9oe^B;P)`ArWtZ77n%dK&E*kf0BgMsn%__wJ~k* zawN5W{JlDL@HiCQ5E53*OHA!h?RHicmvvZya?t)SJ6anv3jT?2YJgZnE$J@=A}VU!EFy8|V`=&EYcSa0ona4CE7r0J z*{7AdaOu3GGTB}4dENbI0>N7VFwP=*;v<&3ngG7LW_LNm|KM~K&QG7K+n6xwG z9T z$jx1N4$zkR673eoG#gB0)UuL*x820`n5m-+5#em>RoL1E@LI^gfv(FGDvFcns#gmz zOBi4gOSG#Tx*Jnz&^WE0%Z!V1M9!WB-jayc@~Nq=B_xsLmITNOxlg}4vz1m*TZha- z0Eu6)#~h10V+T3rlno}6C)iVBr;G5DwCL%b>z~Shglulg8*vqs+*8Iyyp1b6Ib|_V zJ~mnsb=V`Lwyu+s;29%fI=PqNH4D&+rweAhJp9&Wfzw(gGpa0vQUV9U$ka5Ph-~)d z?BK;g_78EklbhVjC44tY5*>dB`FsoG3h4%TW&(kYqmk4IjCp{2woHo2qm|tUdAhH- ztUBL*?Z~KZ@u0Jp0k0P2lCXyjEGG~2ID^6J1ArUc-VIq$fMDJB_cCq!x(+X@ZIZ9_V7!X?CxVWsl!Lb5{j^zd&Ipz$BPH;LEQKV@^ z2|5Dya1lA!?{3TSB%?x(k+3zkky(B*ISk-}T}WH7jv;6vnSoa+y(*6j<8sQTWaNBi zq6Hc&GKyJ8nGJpp=jLXV>;XdJ{W8(NTT}j|Vra3hKxa|B^UAToG4Yk9QbaYjLI^5{z_WN#DOqfZ@< zU35>+{f%Bxv~=e!AP(=O*Ot}kf&h}qD$`m9^&a+1+23L!s`pc;gIsdsQ$PG zk~YI!)GY68f&4fxgb0b(%cz(fVo?^`%2`8uxn!z3DnQ)PW$+rIn1srUU6rk{Wk=+n zYH6nqr%4#7tL+2Er$D%Vrq0km$+}m@CF*YsW!!;mMpoaUf-42c#=op66Vqa*HbC;F z9O+X#tByB;Ptj?WPwq`Ba{3s|K}D^1HLhUkxE+FWHmQ+Z7Mj6EA~_3Q5-p8+TH zgRdHR3%d$i`d`IGm8wx_oCXAMi(CFHP#K11(LM&?4Ee4KEA8mgQ|~C9DNSEmJ8`j zb>R76rbH{FnzyvjxM%M9UfO*b9rHvi+Q@l3kr=Bi$2nKiDGRi;){MXy()mQZl=PPk zcIy^!NEoK>q4$d;^Ga91`D2tW?#|^Lr$@{~uSxG+p`$^y!H*bPM+%mG8Kk+)_j`vU z7DZL;2U6Tq-ufzz=r-3^7pUxbM%7!Sq$UT;wo2rfo%)nI3^s#nl9vN^9sAB@!cYz| zLPm~$b_I=^S`WBVFz3$$KWtE@5MBAb55JDBAFI_d_C5zH4MSKo>PBRu);pI_*w zD|di?`T&7ZJr@N`hfNanphr!ifSn$z`lLL(Rt0#L@+rGQCgSIdn#jEs93IpzyxtxZ%P6R z%68+=?;|Vl4*xiu?pg#5rXEu~2ke<*(yi{&I6Sn<0Zcl@IDm^chxRnt*tv4{)>y7n z7`E7?t!A%v8k>-jY1dd+s>J&i_D(9l(a5s`S1o!}b&*<< zj`5ZN03LrA9V+wQn-A}QV+`-!GATRM!kIFWW4`&Y(pGcAKzPLSiD@-&8F_2?m{XHx zS~#Vh0=p3w>SO3rwl_6^ z(Sk6X1G{><$edmBIuTf!f^GJmwHCeziQDLD5W>jI^7DkJhRD_Glo!9!u^-WDfwr3M z?!+vWNWSY9o~-BSb5%F@_T`cS-(G6PzAi-|4-_JSqYA?gDtc7XY#hPf0(++-cQ-Ds z|73kiWDIcQvD+wW+Ct9+ctj#OA;RW~3!jJ#I<-a+0|0R}I9HM8$2$HBaBNu>Uu~-JC1)wDe&PP+(@soIM!t5<}PeWA2L2g`d_*ij)6NOidI-t|d?#CVhsDQ4` zM!9lE$XBb6?CbK8g1!C&zDJpG+pH)kMlVepjNz4ZY1Z+Lz>%@5J49J-8wadjYT*ai z$adhxiJtb5utm-%Qc(kS(-8*_%^vWWl|msO4%TrU(nA6ww@vmm^m}P&+jm{&pGb|D zl0FzlWsPV}xiqghG+enWuzrH;a@sQ~UTRVf2Fs=Dh}czd0~o1|Y%GX6dCyuPa4oe0 z2r8fenXUyhfMj(HF>*#`q<*Bl!7L(&gvm(d=-z^%P1blTTNQm0{=&0#6u%)fk^qmdeyR$IS z&oH~Twp$r94JS8}c`Hy~Ev-ZZo_7w|8d&}HFgO%Soe0v5Wps^@W+>JG{u=<{^hg(dv5b*qNtx{+M8?E4h{G%U1Xi|y7IJiyz*&jY+W)w#q0Yze zsJI0`iX0w3LD>fiK%qtwE3LeSw6R(Azi3ZHiKz zNu;vZ!YZrk3>CM!i#g>13N*YJJuuX1nnV-GUU8yEN?QFwoIp+p?*M1Qxd}M9%-|`1 zfKUw?POp{QATQ_hC_2jjWKoD`am%HmKH~ss+oJv4_?YE8(WaFMEi>p3HT)FTUU!Kt z?JiC?u*J97o=!LXm}PmzHE*e^ojNoqfYHw^BI1H|41u1gKhkf9XAMK8y2=V%Rg6KB2Sgv8r)E-t(yC#U>O#enLFb7#<`D4@O*`O)p}q zBWmhU@E(v~Ku*)@>jxmBKS~YZu`eW&O_GDGb+zkRw&+nFZG9({lv8Ipu4Fm>Beq~J z!hPWp2)#@-L~)$XGh1-?8&9!_cuoPpHJTJ91q+gqxX7UE$mX^9ZVYl<$K#P_2@v+^75$s5Csb)Yi!0MX84OHW2sb7i z3^XR;wM~sC)hVMRN5CZ%RE8;>;)Ja58}9X>%-`jp!7H~Hfqt*Q5bgZ zfy>NyL+6+;CT?Z*TNR65q#Rllx&nEy1%pWOF0LEd0(h*>QkI8NdZE;mZktE7S1D2|gf(0Yd+2O)}wht-2f?zh=1>{G3+ zD@gQ;orx$LTkspwwiv!g)hyC^yxl1~q(;jr0k`YLq9b|<9%)wiaHh-FsuxH(_-|sSXNTf2t4ayVF79aSHx>#>1t1j zp(5@Qkwv^75YYBat>ce9( zm_}&|QL*3En%S=B?1?#luz{hw-{G$$pa^$c^w}NYaT*wfBKLq&IeExooT} zzU*abY|c7dQ*}bh1XrJg*hzZPKv5|`ZE*fYjLrM?@xQ^ zV3S>j{SdzRC0!bowQsIP=3GYb)rqE<;3JpQ4;pZDK=PRZyk~9?o~NuH$%_L~XKMa; zL_*1MSI1h|yf^imT}D(LA)rLc;tCa-mGO65S7pl(>)6KEN>mnOEQZq}O?jp+dMS7h zHz*}vr6l$&zi#f-v*8~R) z%yjj1rQe$=aY;>eZHK`|QXb5naZ}$T3mp75df)O6X_t{8a3|o*hR$z}SkEq%?HaOG zf9`X@am${Dt}xMzX7Yx;a`mNq02(*oj*R&3g(!@gT|gDITmtlsG+(A5DEU&{QyJm( zYm{I|Qe?z$S?**FN`(xtRpjpV%0o=SPxN5v`{xf(ZmA22ToiENFFsxTn&GJ@L}rdE^lx_LyO#Of3L)GiLbBD)Xm zc*%9(6z@MUU7N&fhIf0&`|c*<-)o4EYDyqI5_H#^gBWw-LfKxG@j^lodX~pgZ}hjz zmdcaT1R0ER=)3aT@R-)Ax{(EBNo8#-6Tz)WZ)?!MdG?^TSUaD+dtUXoR%nnpr=EbUVpg|KM{^1tURx?>%^;Rkw$wp(u7Iv-Ob1xP(b@hx4$70d zax4v>|A3g;*+XT0Nx;Ir{AD@P2KXH2YGT}4@{?tTdVWuSm1Z%o7x|k#$eDzNGDQgtU`+?hKsl64|Dk?^ z*_|jW4^BTQPX+0ML{Z>E55$ioi&p>1^oQKS;1gKlqNjO>$Yw#t+H3P8i`U-}WOqm2 z8ly)Me6+h!iCs`wN*tP(h$^hYboNY`!yAq+WKM!lg>_MiWVIF7AIhVArx;W3cBp>x zG;}1V<;Dez`ClJG8{X|rL5BQp2sS(z8I%eHA#e*wbvZXH*QH24$$sZ2+g}zktodsXEGc zextK2=0V{r3^wmU4e$W@s0_$KQ3Vq3u3BpiIPTVN`Rc`imt?0^SFi3i$eSS@A1?!0 zr>V7q9Ant;`>kwn|0ja}-#-pb8+LIQRm2}ZA2sI>suTpj`S+iHDewgaToyw3c}sFo zVaUw87=PxL;by_llHSOEw-zPn=SumN-|d=|N>n$I*VcjxJ)#tbZ)K%G8Gpf{-vKNANTW*5yOlbHHb{{?5}7%foq$z~X=X|H2Nk|_v;wn;cMW>;WAMCs zLju#e4M@yHrzU^Fs5vs4^q}5vLDl}voCOavkhPy+S`(;=2MVc4$+2Z+)Vb1ysev>9 z1B2-!n=yrBV6r&1VUl37&f-4(f>EE;He+NlKY=p+o#=|B3!sj{IYyN(n5=qZ0!uKe zrJhFxw{+VYSg434t1m|zOcsw@@DNPavgZEEsAFENzJE#dFNyx;qJIzCzq06Gas97G z`&a4zi;G}Q|Kg(mKSIY)A^t$fvM0ZmKcC89}C1w$3K4c}oA^?0~8Rld|{0BoKWYWp*yw z6Q;uW$I+qJzc39#^DZnul($P@i>Save*X;Hzy5|Q@_Q<~^4$c2wj|08xi}fpCH=!6 zUyNIVlo;%r*viyPv~r=%`+YLyBj0^upj^rBhTF23#UA{f7x%{}V{y9xgqMyDm6U|%uTq|G2zhp!fVHZ~K-{H&+5qiOo*7yryx^o%v6YRij)`G|Mr$oLtEo%|K zZmxhr4`T@Uy@c=eCb?Wv^LFmZlWdXKjivKid+hiaK5Xqf*icXfGVi0E8}1sIyN!3M zZ}9uPxyJcYY{!d+V1tf|6!!B|(3n<~b)aQKX&Zi+RQ-tsTeYt{>?U)9cXWa+lB--r z-&e6`0etT@^>efK)h$|jT-)h0R4WnUp5m6s9DlU|$%_(Sl6=&P|KJ@e;^3}0iRnm! z^3VOF%<<_vAwRFV_~U1$w4(&|UR!i|{XQ#n8tnybVat9K=fJh6225FfZ$Gi+f3RJ` zGg9kqe_X=mx2_pVUVbLZoEMgdtw(<5DR{Jna8fxt>HeRQIh<*^^C?a9>l13eSm%$A zX$E_q`Lw*flx@BI$DamPy$36%mXCfLy7S9TL!X5<&4mzVwauD=$kI)m-lP(aXvTzn zVv7HAe%j#;N=a&Y{WOsyVg%b*|M@|av$|^qj=^uTmdt_w(<=^5u6rIG!XtS!Jg@Gc z8|)QQ+5MSsLp9vx^yQOaJA;wrNZE%o$QsS6Loyn4WtmTW3#~6U{Kjl+u4ovSMKt_0 zQn_rqth!q}Z-#CkqS(_+R4Fggxy(&;U-r+DA4>4OTU+`hGIgZge4MvXXfiimZq6@C zwKfjvu)PT92Y${>hV(xmI$R41tF2Y0zws>`?IquM_n61=vbVJ7x)MoCB^iCbEc;8N zkwks}xjDUYSG@r@SI52~vkWrxFt4j6Dmc)c%P6bxSteeS4hgb}2~2sl-F+;-XaS`Y zUrcLmJzi@YL7p}3yBc3O9(heV594OjQykZE#Fkl`(d^4Ip5rhyg0!r$d)`qHGH*WR zi)!?x0cPYZrUG>oYngtVg4Zl#NFMw(M;6q+<}D6v*k?+>#jyWB#@;AU?O-@V=2 z?tR*7t)|&Qx6BSRb?MC1lpQixR#uiLJghKDQJ5N@5cu48ceV0BnTJe;-I{4BkO#mM zu&or&1>RY~=V_sP|#rn6q}0r2>bvoO#0AX=SsUY-Fxp&c$VegquH;2n&V36HNPZXWP;& zk03|v0r&|Osud;;7fD$$8SX|D~(G%?%WomF`T%ig>M)`n<#11Yt30x>;KyXPPbdVQ9*bXG9NoE^dbxv zbmW8^Zq}q8F>3JJpilSMK)2=~VlH65s76!Jw~?oE^9?d{YXgDb#EBcIh=`RO&eOvR z%2EgjxR^`Ez$IVCw#DKb2G-{n$iV3u*WU@b`|GhPXVA?mw23Sk2FTG~t{CZ$zI}Z- zMQksR@D~`5BW#%pNTyRXFjH4;9AQz6-L1$|3d?`ES8IxJIKabSo5g-w4~hwHxxYq8DX zFPG_m`g@PtH{JFJ|BD^G_Z{R(@j~<>`>68uC!~x~pVpr@@+o)|!G6tg(p==zZZB~r zopW4%cl3CxpZn2yl361_Uzg7~brm5tj^Z!08x@(DXJdY6 zSb-Vhz2lC*Gl8(pYKGDgeLJX$#nspCiJ{tUTt@su_~2U9XyZVkP`S+asYo?%{fbO$ zQ%`BTGhhp$TcmgKd0e$Gma)eKs26>`W1~=!C}Ep&wt0NM___RD&DX*eT{ZYi6SAiN z6dD!sZPppGUngV}HvATMI(rOa_Xo=KpO5L*8=qOZ416p8R~L6{*ta+4nbvP=tWVK} zt*n0(6)HM}ybS&eSy6LRcLj^vB5?(DK{$x1JUJ(`IkMuegTLHaz%8rCH(lflN@3=2 zU%%d?iS3U1FG_RxJ2pCc_1$w>@Z%q+7{5ONfA5^P_SafEucbj)PGhCqC#M#hQ=0ng zw?+pY#OL(WStexQiauv6)SlaO3y3w^;>(4^q{dO=s-#oC049PM3AWG1?-pk^pM&RQ z2j}Jd5;(3fxJgyIiFzHp!H}^1Lc8;;0$U;GV}k^3|I656z&L6jJ+26Mm_|VrV-m5V zqb;3Xmte<6pLlNO&lujofgzSmH@dQ?5oKLzY-EE|)z&+#lbG#(9lDfiwn*U>&=b06 zkI*^8$k5_MlC+z32T?_T^#3KOzR;4WBy$+vMuEgAl_-bP*A5e)^ym+a%yKN^s$9S13SL5>1qq^iuT@}=_DJXh*g*4&?EBGiEkkwqu6R)0TAiys8JRiHqK6q4-I)AziMIUAoa7Y{7b^dXwL(Vj zud3;KS(sYVIdol<#^FO=$XnX zrEgvng1c-lCz{f?HHvl_vCsyp%VfBmj~S6~orLA&Tn~n{8pi4zRSna@hyQ)i z(pPh}MBh1P?5$=o+`xN4@BGS>&^mXshUaM1k0XR>z&z;adWq>+o5UoB1&tN z-sGhErHm*=4wXUSN2y4Byjc^9M}jtj=kFbFwMFZ}U%X3Ta-(8}GEJJ%#IjTA2kD0f z5+vawo%<)(31{aMRhWfwNf3LKK{X{1s$1voljWxc;U0;fD8_}d=$b`UzJL0E^$iZ} zHO)p1F+x+Wk*{3lOE^~R=vMoqY*e?KPn3yMQBE@jBge1#(T-+Xr)JV9=Pb&FIi=~7 zFlEzw)2DTcR7j54x+=CdH(-9|s|qiNFeyW1XEeUAVzbLtrmVh-Jre4~;{aWcrTg)` zp>U^@tD53k;F&=z(U6G@+#{OvFnjH{op;Z)#TKwqBvBrlqbr$5C&a-91FkJlC$cBE zQrWq=xqbWEzLP&0PhnF$rNljSpe80y%Tr^r@a(;ltx1`bv_L#n8yReDjbo6Erg#a9 zB|=J;G?%g>CV&M=Jc^4poiThCk1(bT+wdz#QI)d&^+W7{KcwrQZz>4O(lzJdWAXZ1 zewP)#2d&E$w5|p8JsRx}Q@&H71Hbl#VeoRn6ba75 zUkD_jwB57e;g~X|O1c8$FGY26laGT|& zRZOZpqdl52z((>EeM$B1SLLzb`ve^B91J8GpF*p#IH}pT;Ud4k#RC;gF}`9IF^F>) zcJwinbF+$>4Ch#tZYR@MV>K?e$QMbVmSLQ0Q4G?0z}je) zBubl)6;*6rHHBSRPbK}yKS!MJ={EvGC>zM&*WXbSAq$@!>B7)uvhu(f+9D;BicrN=_mt`8>-C z-N46IBY9P3TZmZc7h&cuV4U^DfNml@_ze6=lzHJphP<%#2*bo`H;=PNd9C9}s;&r@ zKtjkT^4hI$q_Ra2(+33z+8+_i+VSwZ*|3exg0qQ8r`HM2sfXNsQ)1LHwUBv>MjiaE zGjQ?RB+Czyy18kSFsD>2YN*HD(8JsAwhgMe=FYPU1+*?=%Nt>d|32~;FvUeu?jw4w%8>q^;0bKjNQ z?H%>oDYx0PBl8l_c4`iH=U~gQ%$_+un}EGkrzB%Of~)`gv~BpafBZMkaIFb4uZmt=-h(*zD5MsD*ebcg7~pE9xS>N& zG>&ZqyvA9hC_{Rlx zWasA*wJ`%)UaH-cx%AljZPh!Zwmq!-kPZu$o7kxVXTj{Un{{IcE{amnw|gT-i)wI3 zI*lofLxHbGZ*Qxvyo(Hdj$c=q>AKFTJA9-o>gf&nl*>;s&YyrtI(9+2Apfv?R=BF_ zGlNOuIW}8r1dkvMoWIm=b{yAlp160=@fBb->%FX8SU$lGg(RQa!JjDg7SJf{Rj|EZ!${)oeOg z(}aJ;^@gJUQKGusMmunm_N@l|w;aGbLi&E_!8n(^TzxoG?}_a2~G z<$T&f+45Vv9J!?qmM`RAvk2UE=?x1Mhfge(RA|-A_T1^QYQT;x7jf>K(p^QCmPS zcxFv>wZ?!+)lH`^Uvm?T5%*KK=*{*~0ks5w3JjFX9rkg`J_EO%GUth1w`=BHS7>uv~_j*50fVs1U8;wP~htA7gDd?w|83dY)tqIu!V zSo`AH=kng{3brFea4wqj5uHD!OPVO7;SB?Co(Lv-{1IFpUfkAhp)eE3x&md~n~E7~$$3yl-n&K;d&ubVV-qwRaQO8i08U6rd=8vy6Vb(yQq zO7`tOKpN_7)~VA%Tdc?ij_dxnsGF=2vb%1n+!h(HNthfNpTasDXMP1h-D7@#CV61{ zcvR9BFoNAfxP)u9f#9QV{}%Adi+8-Mag$sfYR@8>TUHA6nijvTct68VzUM? z#HzN-(C@%Er;8EXidS@KP0>Uz>~F}g3POQD|pw$WWSs}3wSuXR+!kM?(u zZE8OsQCFTaFyu8=wdO9U0A0 zze)c8Cn<1$Br%c^+%WK7e&L^q9KD6e*^Y=jf2#%eG~6ZKn27O-!}On%+#PxTuGnq5 z7@%y@_>7OrtCMs95#=WJdKKPuH}tun@wsL(t9RjYl)65dl2#k&I#AuTkHS=K<0;L3 zIk`VNc6NkE&*X?wlr8T3NRBpU0^;GQfBMRs?lp`>xeEQ$MmjyN<4Ii9xnbGjjY%IkTk%F`czAZf)RG8DJna33A{Q^`5# zp|Y^d4Lk>{arxUxu(A@FjmoW%BPoIZU2P+SBc6n`3Kp3 z+RLeAm}hI$P*fXeyN(F*F6am^e?dnFSU#PT*r*4k#~k9}-Q%#2QnKt?qA(W@@?C-N zL%_Tv7vYGukW~jAV~#5IX;BPDN+F2r1XCHL^c7xz3UJ^pdL z#1eQ@&5KS}9l(s!>kMm7r*+e$SB4_;uS6WA{x0IT!IBlZgp4qh)M@_x_rGw`*590z zV)w*7Zx`&n|1ht9##yJJSgoQ4gz?!?Oil;Te%hnt=*U2RBj@v(B_#iD8dG}%vr@ej z8pf^TH7+kLHz|8tj}%`7%bK}tT?>I#JcHLslDOoO2Tsv-6JNao7ar+2*zSXNOGyBz zzI)kuw0p>gJp#rK*r`WdtL92etEYb*eEKEM6Y3{gp63;BD`P{RjS4>+dO)8MmM3bZ zt3vNw_@6(C_C(SG;_%fXzVZG|8ty4$?W0|uh_|awM6hG>>;#cc@rWa-7$K}ur1#e` zznU1(JQvLi?&KnUkvs>q&WU;P?BtcKaDA{SCI~R?9p?{#;yw3C$YS2*95tzGRBt0S z6z_i3Bfnmc9djv}dxOnvE@1B-in{d7US1i_Z+yil<4Gz{bIo|Xrpvg^B3E;465Z$O zOeRuOTL|xXl{J<d#FPzv1`kJ^FHch5FuzYKna^v-T#+FW}zP3Bt4ha0>!9nt*?$c}5>??L!SdQz)$ z2L(NHcA=Li;sk`usGpiZQ+)W-|8oi&&>-Bh4R%Bna~5cgWUxjZk+NlrOEO6BXef-R zq?eKArL6ZJjIt_a5?{SUYcJ1l9lubqAN+k!gAH+ibi;cWUA`-0hmO~d4!8-%l*J~$ z>U4Aaeuj1SsuM0_Bj)`$pr2mKUPZZK(uWQdu+_P$%H=h)>Lk#@FM%f`K$tgmL(Z>7 zuPe7m+#-7g!Y^%k&W$1<7+$deaRe)WTMZJfk85;Z-CH3|ZQ)?h>C@2-OxXc{Jv-du zi^97;EU8z(OazKKX=~o&8$6&zLP`m%2)>QV6&kS#iWaOAMzdf7tf%YR3ZujmzCyV! zd^9DXXFW(>=k!NT_%ro2BV1|jLb<#?ZGPh`5L_DodocvHzFFiVYy`~ zF`d3hzL&{flnU~EiIJq7R9w7MPhQmHzN8OZ8lveD^>^NTy8wfHmHbXxCBJFJ1~_(Z z*1#2*FoNX!j=Y^b-uYO0w%-WPQeV3Aswc7FA0D%hhgsq(CJP?Hh$7Poi^31}OM$Cd z5*W_{y^`OI92!xoV6c0?i@6_X7Msf_Fv#8#zV0u)ayWctPj~1KJ@3;L?5<+k4{Am^oV$OtrG9ufi z=C6!TVC+Cp@Z=Rg#Y;iB0pj}1<3K4EWBex3T#4`>7s&lJD^Hb2!IhX8n&(5~iD))e zD|bvW{_KL3g4T-3V#xbj#F-sFCQIESXX+BL(_m~hFp)!-V(|%c{Z zW+I~_P+?-zouVHu4*n4heElJI48Ie$zV<@cuQg`=NyM#(_mHWfXMpm-H>+z7&s1b8 z!6YdI38x!8THUJguH5$(hu?#q_K-A1y2$LH4c1;lEyx(k$2CV!kQRGvi*MFVuwU`h zm0aMcg(gES_{akkD4jxkc|liJ#a!#g1#`$on%2!<$mH1t%*U37^O1B`t{^ly)w_|C z2DF@Xm1N3)f6vWZAtatL%p)#dudi1R;!<-)S)i!2w3w|4l9ZA}5#>Y1xo%JW@a2M6 zGv&Q_mz;LjhW{OutFGi}-z+El-OACCoy4F&qd~8X(Ozf^?6i&zmqmLVxs_1{7Hl3m zwRpcj4eYtKH3^n`>aVavBxt*uO0OY98W{>j(plI2J0~{>DD9%No3S#&KixX@ zf^;05qUAX%)uc(-bc70QaDmcm$zXgGGR(vY+cZsXHf=v4?{ZBid_J}fOK&t)pH<+z z(}}gON(rfQSTFj+mO{#08aTcUVGq9-? zZ)YxVeP`l9&f{0h*Um_bnIp>2b3L*!qDmAtG2!@RVMAa6KYRq9M^#u+!;_#5dFptx zMB9s}TA&i5z()}+&vrmJBB9y_$}i_tW7)z~)20KktYYJ2CdreUDr)-O(ZRJnQlOmp zu=KOU$j7bYHx)p8%z^~VngdDFY7FOW1Ouyt(=PPO(=*KX(;8h(s}xsz#cAnkY}cQbR^$ltU$(e-4N29T(c2B12LV%IDB7E1oCjNv&8V zP}0cxAxw&qhqGKc|K(9_B#Yt%JPRG;YyRYp~~ZXKH+pBQ@U$aW=h-Q z1y#n2R!wEhj4!g>?5$pM;v^_YLIybnFGKgGM(>r{3(Rg8N?UuKeG-v0hW6M+1|Arv z1p{Xpa5GU}k2>~c1KxL)+Pf2cu-Mzz?k(0)%c=)R-pVFdhhZc_46#?C(koPEzX!un z*oT5|GMemLW%u!ezvL`ptD9yno$Aq%uG7P~CRBBElo&k1K-SUPs__DB3-jWz z*Wp#KzPGg!ncQWAouDkk1z4k7l`>P2x%9DpV&t>&8CEL$G@X3SCuZ@E5BQk^I!mN$ zMo|abpFNDGW8k%ks)Of$EPyY`>@6hKSXCc>QcuMPeaw|e<%Ji-o^h{%g)x|$SXw)3~2!4ffK=pppcWbqdW8B5k0@i7RqRqlI*PS;Jgve0P4 zR=J1{;Zu%g%10ijg2A~!I(qzs+Kak0<`1+RbG~a_H0EC{-R*Cf>+=oWuv!}eMlc0Z z0z~%$n#__t{QG6xd?N6zBGcDAE>RY4mrr$7nh(DX3q_O>!d|}hThQJ@A?k%_zf{$V}GO8k6(E&g3$U7H(0zO zAg}A;8`YbQPr2P=9zB9KR}T)Wbn-4}%R01xuhXT)1kz=@$%v5W0ilJEy`l!<18SLO z&tKOne7lbDw4XExfpNN4PWMp~Bv5(x0_5wHFH*^)fJF*cHm3g(rM|0ApeC{CjwuWY zpL&W^uu#JHLyPyto;$y~0X6&`3AS3sg!_-_Baq~P9Mc`eihGGgK0fB)!&CVhQ(V9O z{LY~3g79Zv1UJfs0rrPft;XiTwFQtO_r$zg=EI^kGYGoD-%@QG7lep%--o4_JmCoi zcM)iRNq}T|0ToG-+IXu{DC+QJ@rK_B267zaQ^RSb*a5zF*3lN3kPVnuuG`h~dK*D{ zU!U>|T$rrlKTG(IpwVk;Mx)bw{Ckr4LM@g{ADXFD6@}o_DI}+PAb-e=-au542wQKR zecO4ExMpgq3$rkx+lFU!qBY`e)Qm$}Ira(&k1CWjlMTgu$4r84JAZeA}qb# z6I8(RG?f89T-wqKN0u76sZl7qd3%s`WzAlV$O7Gt*|(+@uNw4MW15AF)FX?KVj=oy zk495I#G^d{s8y=S6)z+R69<4SvSOOeKYkzN)nnpRI`?fyl-{_kvW4#SQmGf1-yJ*C zl<7-uHpUb`yA_b$kY|M(v6d3@BhPDg#%SfF@w5Rs-`+#} zt%Op>S;caG16!x4rJiZqFL440><%NzZT&Y9@{TZ%mg6J$EijqBb1k~j9FHOaee>?H zQr!V*`Lp>gTz%sqY}9EGGXB(92%pNgZ^6E$QmBW!^q9;^XEdIu(&NQn74Jj1 zOo53cv0^_s4!N~K`)+pU6wepk7~rP}1<8=`3=_Tc+~I@K1M9Wu>n4+Xk6Uc_`Y`m* z$rqw-YSubxvR72Mq>5ft$Y;fww0)`r2u=;&H>L@KAwUNp0!A4a&x8zp(kX$LfKS$B zBc`1mGAR-XcBque&!i60=@f-wY}*zT00#0N&uL%kM}rVMaB!;g%p_*HF8%NiQBciQ z-oalyHkZS5r%#!8uM{tpaX4cP!gvbjspP@6D`78SM&{0TRyh~HtIiH1X*bYRByALD zwpOoakvx6*1fl+fsO~`%OmR@Y^BUO>HlQkJ3zv_2UTf{f7p8niPn6wVKIJ*U+cXzo zz7wr^sD_I|gag~&ki%4Eh+j9a!;=uGsKow6;VAYb*m~s8^C5Q6TCd`HzTtRobEuqJ zgT)Af=iBqta5C8G$K})|Ju>5 zgtO=N+O-q&+7lQjI(;UD$}ZHWjhkBaLl=ef$hwQ#YFr;}QOm{QF0gPokx+q8&mfdz zte1-D6q=f%V+uxau}+ENDT+nI$5?s1>JeHA&~7PS10s{#%aw|)*pA_p4%CyEt%3^#MM zoDcKF#q|pAWSGeb2*rHOg=2Nvv~DlbiR#~}r9}~-gBmrr zEWtFa-EaSVVPg9Z2uI!+a*O^5+Z&#|>lVh~4129hi-h2QNCI#T6(DrU`f(y`LM7d&%dDxJb0s0)2TS814Gj%z)%w!B@-tW(XB@ zSg6{-1@gZ`sKzgcD-)tq-Gav_TPT^+w;Zm?eM*On|4vD3+0!} z2ctTl|Jotdpmw*Vg&jW(#=4t4@ZKHxB{*{$W%((R#4WP}Gv-oA9^`gFZZqMas&Pfc zpyo^jWYM1OjY>ZB`&xJBnF>@ln36;S#19ZD=Yt>isT-#YbKD!IOLU7`Itj~&xvD|3 zL%;(HAFQs3e|n3TQjg5PZ$#ZOB_O%PLWZq9-VqOP%g-cz9Yd)^a z1|%^C=PF61{M6}}szpM%g@bCwSt>I1;cPRgtl_v^M%n^jTs;H6#0YXmOy+z+t-^IF^U zHTWB4%t!C@Bf%Rbl24TQ+jDJAR{xUkWLy`3s~$sD=sfB|n;FV2Xyj=6_=(3yco3=Q zO5&t^Hp7%{ng00;jophgesli1D)bw|Y(~R9DqzX+MJ&xQ!ivtPU zAo)tvI9?@L9^7V!5(72H6{q^oS8^v2bC%x(SO4X25=1@TQ}6q3&UKOl2#{wtYR`4Q zE6)ir3C{zeyr&uMtJey4Zh&l%WEi2pE@_$<8nvn}6H|VtCNN;p@wElr3uDv$^kuDy zc%2WLTJ~u@dIE36B<4q%06FSH8(^g=-cJ{l40eJu;18lbN_vvtgBlwakBg>HlXcoA zgrfcE{7c=Acrij;i9x4C4zirk!4jDR5*+bVwa$GE<^Km+6hDKo8gI3#S|`6h zJf1@J>)woVsv?u?2M64_nmSo9-JZYvaQ}&i2UChHb;N%u3Rqq57*w1340@5lL+>fy zT;-(RCzO1z_LV1=O+%R@6F{P=WQZ01?9m`Ahe2=?5aiO22QC_mv@lujoFl%WI*brAc^K{cd<*G=wJnmg|)nhXKC`#gFY5 z?8_r+4570}aQNc8ZZ!Cp&o@AhT{JUOcIa8Xq>q&RiL`&bw&MCC7*AhzzPpi`oHaB~ zftdDXq++sl`#Mu*{_ZA+#WdxDiv>(sE7leUbC7Q)q7^*rQp2;xNV2K+MW{Y7j7H9W<~jJdl+3-cHIE{e`B; zXu2??4xblY%{u_b(&U7#tQ3ym*hpU50F*ml?~61c`D)Cs?LZ0=V+MxItNvzErfF#c3*d_E<&{ml1{@R)y3_3npup(^_01&z?V|`+H0^E znRdYRXdvp?fktMFT^D{QW_xtJAIFo(+TRJxxDT;!6%N?}m@@Ln@V>7bRH8OO`EgHU zx&^}jcJ(CaI=dR|-jwb1ngK~zd>pxZ3g4)N1UCg*bkoEyHl!@J!=o=>k1ra5nEgEu zH}!uVuIK}o4#*XwSHbzRz|hj`1T#1+XeHM#x8Yr>KZfar#8wO^p$Qvn(F+$2X;6+n z^tf9I^KNts$fL)dpU!l#tVe8cK zaSuvKxfGj;gaTISlUUBKVTHWq9+f-5z6zJD!wj3=CtqVdFBsVg+91*1QC?_WY1N(O zOh3AcEY%;THv ziVFv;^1WqA@AZT(bAL?xZoY8gK}%ZSVIlm!?fnNn_mwt*prNz_t|T{dI3_4yop+^E zfNo+-^!JNe3oiTh&7Q_RDqBZxOPS0De**yOxu3q){L|`I=zqG#?RZHdaIWpEfr~2Q zl^WN$2yh#qFmo*Ou&}kdTzm5PAkJa>07%+lWkYGl0bpADPc7U4`FZ}A)H#&Yv67$Y zr2X}IwtRL?k>V?RBLgr@ZS$Z&RGf=I3&y+&h+BwqlgM1cvtaUDt_ls)?_E?a{C+ zf3;(Hat9c^QWDOOfI}*k>Kt7s4Ht{akD|vNUn(w0oOgblInOZA%N{!p;y`;TtuP+V zM_r8OrklBoqwxf)kxx-&1x%(EnZ;=e^9iS<=k=YLHfpwqohK~a2}rhT703?A z&W9rw6wF-%BPG@ad`;9=9IRsiK3K$MZ|%JsEAP<>K&eZ4JzrH{#7aS7_|eibnI zTNWBW@d7s(RC7AUH{k59ZXca}U22SyJQ&2gv`C>$rt_)14D;q=XKL?Z+nZekvbHIj zD%=R}IJDu+XaU1l;3cuJ31+7-yCAA2Q4bd_elVyu4=nj4j|lFJDa~4diYimc%{3>E z#G$}kacdvo#eobz)Ek5nVC9$-dya<9s=MV?B1jbIBbMj7+<}dt;L&D3Zn2>nsqmHC z@vEH3tPBU)=3^DlO&!FpFD@zh4HzEf`F>;WV+qC;BEuta*1Ar~yQX&p_w&N+7PFJ1 z2ABB|590*Xfc%6buTHvSB17f`yN(tCzL8TM@Gq%A1n=wt@!_u{gNP<|06(HqHZsy(s zMExJF`ibPEB`(m?w#ja&e&8_>qI-J;;Qe=etvM1-cfZpdT#&rdDtt)9MR*hH4CvvH z$>1FFrP8=a$ACa~al0Tx6h0UnZ9ayQ=*v|^{No$*IfU!mH*WxaU2`;PG6VPpWPa+7>HH2(1Obh@*Bm z%678Q5HHVc!9E%Eux#l*@MZtAa-{!D1(rZ{c!DM^lcNNW(j;ANQ3Vt3cepu=)2Wcd zHenhhL2)ZQJD9yXvsILn*BwZbVebt@DeCoZRb&}88wsB7%%N4oxAKp-fpf#U2yWj# zXcOS752q6fHFaCj9n0{DM%rdf=hOA%6wl=MglI1Qe#E`!2^T!p_2O*SjTl@g+v=$Cks1p_b)ri=EqijUo%1{VsGZ{pDtO|Pv3!%yT z)qUnKr&s<2Wn{cFI`72Rpi%SeP0E@W0!-uMT2C*Qn^D5 z;uG|3XQ(XFL#RNbbX3HJ?12a6DMA5cKu0^*W*Q64Ml7^(8afv$u#^`5BWxXp*ceiV z?{v@BVW(Shj>h$jUzXS*w0Md8{u3uD;D_~AS2SPxT(}*Nx^D8Bp7;dZsaTXYUbL&l zfBe1iV`#Fe;-|0ihX8hp@mktMp3akIm@KUKWuM=-of0_|4H4tJ>yzlBYK}FXa04v7 zR(#~P-3{7=no`A(7r(Xp65Sx3-DKLSUM3KRYdXC(2f4*tTh_UO?M8A2aXGps;@T^` zh6!icsef3Buu}U5=%{Hib>>3=xZZ-&HXc%ZGM*wnRnA>d^eWVh7)j_Si1kU#d^P$; zPEr{7BQ-|(E+b#dZ8vz}%bR@G7H#!LA`)`Mb|B33(TDNwL|V*7a6!Vhp&T&hAgmO? zQInCj6SH6(vGli4Ap@{i+)J@e-Zz!?8r^CD9iBmxLt5^5phDAUO#6CSu5?b2+!gglOw-e zcD+X%WnsX6yNvgrF5c@$4@cSYJRuPS;s5wfADs`Z7lwz<&J|Yzq>qsS?Ojii_->Xd z4`Dg)Q0FcNM==%ZyXe3&^H)myiv8Lhn*%~TrLEhx1*FC4J0=pjyv(A}+G1SiS5Y^4MO1G(Bt(SfU5_UlH-2I?@Z;6N*Ac{ zP7s6wcJAkj2CK=N?WF*;Np(9l3X5Ke^1|deRTxxTquIh?Ud}#Xu16-mRqvL63&)&X zaifGasT}zSVFF_T|7tU?+n!kU0zgs0+~cxah}ySMp#|Co$@X_$N^gEM`3|u3NX%4F6&-2mVCR zVrqxS4zu0{M<+shnz^y1w^n@i?t#srajAE9xKIAfsRx}O6H+W(NTFau?L8(Ll+HQYE&-RHH-aS(u~_l-rKrff$NMgg0?Z<{>)p^T&{8=r5RKxl)q`r7rX?U<~<`(=+` z37V9ZF|)VS?y=39zlH4G3HHA%TWvK>y9CucfJ)q<5n_JTak$IBBhVm|5-m9Cu_iJ@ z5gTzdg7b&Dazg=-KH&oKRuY-+TA2D{`%dUY^1-c(UJ6yeTA+dhV}qU;A+K4 z#welA;}7Jo>9@yEz`M}dZ*T@GjS8>UqK%_>Fz54t#&F z=wiHfdK)ZjJ*|32?QsRGHv=Wx>2>(JT^Q;Q_5F%s5uDIEqd+z1m!Sq)$L|N1Ki^&e zjlsV?PjYmG5)yyia?2oxmB7%ZLt^tDRMKK2`KAxype=G58Tfe=-|Ib}-oVkv^mxA9^nVc=rux>&iR*r8 zI4)`rIE;~6X8i6U^c|UPLK#M~jy0)U+)>?c%-##Xk97`-YAdmhS?#{9aCGiWTsub* z3J!VMD54w8Ws{khM!n5q`5N4GR?T%@&OS$e0aPQ~~Dv7_nhGk~E9 zfWYc>JyGBGZM5*EQ|yXF(=jAFvAfwE?hcmM0)^lg4!KbRlaQ0o$xsp1T1NM zxEdm#obkD2>AIfjrH`zR1@7ANXb<@ZmMyQL0%N}&e>u-CjQi*<%^q;V zV6YgjgWYPG!88TyHsS9kgWEMLZfM4F1&$k8oJ&xz&bp$yiGWkeiZx$1ou%&uLq$ga z(b5Vp;W*w|;oIs=^OV&6IZ`Y=QZ8?zlp=dBqS{qp6uhK(lder=aMK3trfFPv#NWIr z*hGO5-@&AHMYQ)=J(oU{_~AH~-!MvS4I38FJd^F3I6Pt#@e3$_s#RVeg2NljGi&ar zde3AL!8{p9_zz3HL4`rv0_(0`SXpfi{*KAjjA6^^2^y=>pB)GC7(xkUTn+)x3Y4%$ z?G?DbBR*%qrOG3Rqlu%0Y+BM;FlDAT1p@-#)c%-I%(1NdRBoPc>ES1w^s3bX_7_1D z0VK0lt$-a1GZU;$ed!PeFxxsZlhv5K_QeF{i(9)JTP7d3j+qWHk*D^E!3kh+!|KZ) zo;$?kg$bhm{J8EXPJR7{{B|!zO4p`ZW`i}=)z8x4TmEXU^rFzAcT?NZk*jHRDy%)a z+XST55;})7-%8vjXUU!x4e1o{Qzzy>E@f#maA&~t`#x)=Sn3I73TFHNt&3$)7YrV_ zKU`aZjT+Z8?f7@W@EJ4?zwdr0a5YZ!DLJHYY`8@kygGTfTRb5~Ta8$(UVaqNbtc`*()q~hqp|U9` z&K$dM51Ts%#s*3v5WW?UioOj{xNr6>Mg9?3PO7Rw@~HYo9x$+!77+pOW@&}3-H&R5 zEx9v0mAvwNYAI&%LM-I)>cWcg+f^Mj&bPVHIbXL)2!#4TP=*Jk2{&|3;4Od)dFDw0 zt4xJ!nOte4g1|E6VilFE=X1Mj&B>g}<9~*aAKLMvndKkv>Rkyo57;Y+pE^S5+7%`BwEU_!6b=|c|Z(^SoElGr`o(8hFGxF zcFOwc*552)f%A1NWGk7bH!bxGq?Ivk5`htym4p3%m;tXqN6-6EHfx;m^!WZyIZ9c*5#vz<-jX z4*@GB6Y!C{I%6BoNva57A-?Q3{jIVpFHx&K;T{wwQj|Z*jMGfI{Dh zKbk-Vs>xDYBwzu4|J4Mt-)O(-gG8XmgfcdZBP8=3b%;+%t5~vRlz?!x0EsX@E!JAk zcl3ch5~;uuj0&G3fCFuM852Xa7xqbw$zVMm-RMM35_(ZdaU@v|UlH)v9zb0uo;^Sv z>A&kL*rEd4pgSTj)q%h0X`LOozCYQ~!27AkQ7WyAbW591qCpL0s^br>P6VBWa&z)q z=N{+|wuSTA$Y<^XSW67?%}z4nL!ZjbZA>mDYiG`P!rK}`t{-rB(TBr=5`v9uR z;sbHmUAD`_+8#l(y6 zh9dS7CLF;s3ZrE7siPxCi#u{wQv$aCa1&Qm-kPyn?)%Y&KO6zyovZ+RzQ53ryn;b7 z@0!VG%~t;b43A)_@o_j+N*^#6*4Oh`S6ig1R9Y!y><)2oGSxl? zPgyG{b@D29a4R3?osSnK6kFKt&wPH#VEF^vZmPvZ>HkCGF@p}$z!Fl7X(F0xv{Ev1 zs^20qJnB_}!)15fok_E^*dcH^;&{U1IAf<6FWJ<+)!cCw8y2I5-v=QPa5wi8*c*xMdG4ZpMy+2<`5>-=>zNQ2n$DR;Kf?NH*LFmDoB12*SV;|P~A zRu6Dq$s&JR=Oxk9a}LikQwkdU*_cax9YPV_M;8zA2pC|$u8L0Gxj9{+FlEBi%c?&^AkY&pKgGN!8L<-gh!*DUYvb;0S04; zSFVUV+NFxb(Xqy{#wyW_;?2TzwnGcQCS;;0)tnelB;Zq`u3#>x*g`)bm8o!G#n2M+_ zb0@f$7;m0(v}id>e=wj{g6aHbj(<@SG6U$ZfNRjt@i55Uwo$Mh=q;nH0gkZ1Gtc$V zsyf`F+FZ{3iEk6wo>;*U%Q_lHq4nc$WYxC;Z2u8Teh!@m9-C>H zi{ol+6|ysCOq-8s@3(d*J4LUS&m>T7M}`~UDq2q5#0R;Hp|P2C@5$B_Y%K>o1hT4u zm?k$wKkRrcBxdAp5@-ykAk4=9uf4AiOZsg8wzaM8x5};crunimmu2SCOi__-YPQV0 zolWz-GBYDJB}7G}`) z?2y!T5XveJ>KOCuC8H*$<)PBsYP7bCrA(`dyYJ0vk~L@-HF!}n~tqJ53b*kXe>veHUxM(>{Ck?gofh!#=_;8X@ymT!+84wV+^`Ma` zw2fQtFFqy_9gd*K4!;zCktqXvV zd|4-eM;b*Mtj`v!w~R-UukH41sMFohdHztZj6==Kg60-t#BJiUVaxMLh9yLt0^)!& z)7VqB#RK|Ynrwu0La{q;*xIXOh&>O0X!=TPbbCmos7SuKbyR2*(-fOGlYCrbwAfjY zK%J1$ay3MFM8D5fMGm<&X6c4&e@~KT`idG`?v}4Ds}~PS=?zE+e8ZTLYH(erQ_$;Z zVlVr5ldH=y{D zAMKXB_mjm(pZW15>5E8g;2~~7uYOpYc44i4&KOM=?nUlhJj#=-M53z-#aGg%h(_#J z;@cF+@5<729^HuZQ~$O6EdobLbFS@46+a#>-xuQ01p>x(wZjQ17*Zq4QIk@R6v$F0 z(6`=l&hwOMz}W#8lh9a@h+^Z$4oKu~;9$fdR8;Q7i7UHT@K^a&5L;9dvONYlHu`p;jgAyNF`8Gangd9|%9_=&#KA{I|q`BA)Fq1P9 zzfq0?RX$*Z8WU&*wep#g^c>Z;c=zK%)^vQA3uE)Hs+mxIzXT?5Ld8xS5h4Bv>Wbop zHA{)T4TDgYgVS9bgwcyDRFN4Ef&QEp&D9jn(RUqU5~aW$_L;KY^boo^9*H5usDy(^ zV9r?Dgf3cFvRg!2&ge}poe4Mi_0H0>C%L!VWfg$Vg%!*0&8`nY=i|oPRKrT+2iV8z zynR(_OM8Gpq%l5vDB|o~q_;R7=ooj^ZyK2IJ}1DFBqIw@jk;(CoW6(=bsW~`xMg%I z;hY*;E0PJhLzJa=nGk_DM}f&l8zD^^N0}3{JyzYVgvSe95~lDzXc;MbZS!z|<8UK9 zzpX>m5PclfffX+D8w%>?X#BQ>F6+#?!Bx)3R?_} zNLJ(!IR|IRKvVfnuR7#^o60~)*LQ=mzny6aPU#*UL@Rg{*9rC z+ry84=45?&sgiZ{+RtVmKDfh!LU%*rB{Ei>`SAFp*6(z=_oC|_=%6D9xWwqhA!}^p zY@P4~+tudbR?lsPx2EoNiqszV}@S)EYu?M`R{0=AmE}MKfY~ znyjgyu2KeKc0O8V@G`=mw|U^CrIG`?ex@ZA(ej=r*YJjBbkdecbTLZ{p(pEe?9z>T zOBF53i!?jVq{y$|-N6%U?U(1DF&&`X+1Zrid}*LSK~+|VdJ7`sF{y*@%RHt! zrQU;$KsLWXF4#}}&ti;`BJNNgelaZ_vdK5y*F)eB^@H5l(HXG=DhZH}M&1E=1k^iV#KV)Mc*xR|q(+J`QF|}N@@-+ z1G2X>aP%mXT`9i)K#7s3u&O})$9*KxLS)3x;6xrt`5aN$CMJuYS5&dIc?+7zEUnxn zoo6z1cSD~mjU^j2p`M{$FIa8xj9$atd*`4t0tBR#K;!%_ zjfG;UyWAd{aui7yu&im(Z)s+nS)J&oU1wNwWSb~O=dUpaq2u`!grM3i@pu#p7FtlW z=5ltG?EZ!zfut;$E>Jl|JoD|UL)a%?vP?}En6qe?8%eSPEp?M=wMRbvZnQq#jgVj? zL$6Q$vas5tBb+}VFKtpbFA0!SUnEZr70|(>k%CPOPNPC~&xQFFcy6~JZ9}w_E^vA-#g^zDbqjn{Hl*TH5oR~eVLYRxev4h=OGY8G&Q&mRO-g(r z=(1OrIj_E)-Uz)U_{cTxm9CDI{tpLgSnFb*27M24Y|O1tRwmW~j)~5$8Z_N}=_;ZQ;_Q*e|XR^~hmBq`ypMM$T|6>`HW+zNqwG=OahOlSmyjT=ZPjn#duhLzI4+U*}BJDJbSl znaa^zg7sW`YQ;O$Ke8^DroSAQF;JfF>`?Cpe}C!b@!biZbLuPURwBEpCp9;A?R$oQ zSZFUa?kWXfK)aAHD9xNW;AILJp&p$0QS}X5H>RrYf&Tm1qf-)o7mdT79Wcip6VXOn+q>BN!4tIncWwI2Ej5hN4y&D=}tASM;hUi%VvIjFvpo-67N$KGS&P!$jj$(mXX4xIZbrAj~ zMR0fr*Mv2t0QKHVq=}Y9y_zz+7@QANovyT&QQ%#s0+4E2X4hFM$FPi)>H7ZX=9a;X zWxeAZHGMyRARQR+e{|H>iq!$S`o+k<*p23&Gfu1Dv00Gr&O5@K_jXp5Jh>qYUbS#E z!D&;Ph$(2K!C*OgDVBl@lEq{Cq-m1~NCs>KZi5BIi;*7_98@$Xi(pZioE#+yw_@z^ z%lIWL&ymsCp~0>K@6l|C$oH}k@m(CT^XeS=Sf!{5n(P}&=2u0Yji3V;m>|azq^@2C zFdphUdUJwjfY~$e#k9`cX@Ht@jfu#dmhsp2y3#YC8z&DPqE-OvJq|(vwmrK5su3{w zf-d##di!#X2_lVw*P7j2;I$;>sD=JVm@ycEyjjRM>*gU-7;a29o@&2P9opwMHUjh4 z36KrD>{yKNqSp1LjgqCABdA!(3mMqOSv5_~pFk?N_Bm`C?`%6u7YXuEVk_Y&%uiXH z4@EV*aRTZIr($^u2BnFhO*#|43sTe+Q8sLXU^cAQ?;WkrFU2Lc6d76CyjUE_MhFBZP68HdGajrF>lZ9CmK;wPEs(GxWH*wSju)u9Wltd4Ejz3YUs|Td@wNydt81gRD!8v;RU}?33Km&G z@$Rjk^uUbZTfa24o8sk|v_0m^iv8z#u1|H;MV5Y_Tce*kSneDQ-96LFlymG&A#QPE zv-7N!;2)6}pSFg}TuERqEe=CJs~pLYIs1A|DDxJmy^!oB<-8I^VM(tg)~62o^oqPq z9w$7UGszNNRFXdxwYu;%IXR*^7mPEZtP0)@gzAH+EgfW0o%?eMm*q|0Hq%08n@)g0 zBmV;Ev)k|TdjJ=hS>cN47+m*rau0TBqrf5U2j-}40%vjUsBIvP;_g+y`k(;Vrub>aZR23u_03Y z)OYIV2y18iP^!xtU+G`?zBa%EJDitLad;+`dWM%h->UeR!;_misg4$F;-9G9N9q*s zK3QY?j)A#V^+eC3-#$xM<**}-Luf8&v{x~@qDsMQ%|TIb&l6c%%#(0moi{rw47o!+ ztxIn17()9T^Vq=;LxkC8DD z|5vKea1GWfTqbg8S7ot-PxmX{aot(X2flh+qds{Wz4hG@Mz+DPYdY&v>=VDZEGcmk zIfefiukcYRd#hN*$#)Dn_)W7B_wnvYNQs_tZ{$U7LV-nI;ahsS9HZ zSEyXpc6!K6%j{ZAms z(XYYvc|YrIsD*9SYBpl!kmg(w#M3nK5wMj0CPS|#EF_&YnlNtU%J}$d2=F-N-ZKM~ zB{!Y$qbT)C;@ws1Z+f0vj5g&BzszL^e|=@!vm47NWC+!M!bDhKbo|c&t$|YKoxPH5 z!t=<}H>ExF%04@cI_y@M^_=R4q8==Z8VQlrv#J+0Y7@u{#Bb+xu>S3oSY8FT(RL9m z5#5eXwSv>($}B{OyRLwN6?w$~5DM0d%CAW|4blj+^6wwPQg?+9QeNkh&NsjNAsEZ|a&>lM+~-``rtp5fe`wn6!y1w%i~jl1l< zY~HSQ-#IRXzlV9|Y4B`yRDYWhX#O7|)Gq$i%X4{_nDS^OU7h~8bucx7)EQ) zinXXtUTK*Am;ljy)pas)*teMp5)j6ZfrgUl|zOyZ)W*?D#5LyAP)F>THrQ zwe9S;ZOnv-yw^4J3K#96%YoFb#`gSY_qRr6ZGru#c-JOQjNWz|F|}RAAOi;ZZLI39 z_ZSB}lW-(WKuYcw%>`GJS$-K8&gW0Olyw?-UtHGk3o1f_sO>&-SouGH4K8^k#bq|Lw%+k9|o z&$5}XIggeVP0G-dTUl~4EfUFc6IueWeI!~Y>Nm{ER}edbWh_*O?1$bq;Ka&<#n{oZ z@=MOq8XfZ=klBDJ8UcjDiY`$F#CRMkwMT?P9$?v&3tSl#Wpt-TQ|dov5)Vc7nMhsq zX@%&`Z#sgs$I#Ec$Po!Nvc2QW(Kd~6+PON^C?QJFg{gP!ZypUL+Ay1FdDWr|!M@?2 ze&MVu17q=YK?-{!RXQ>TaUCPUTa}O5eN7HHznCCcCx-Seb?Oxd= zcPhL8R9NOj(CBS}5x41gQQMg?4~E7}rL+Kcr4#p|9djnYUZ(f=Op*2pr3zn0e3($Y zAM87eFPmT3D2SXHv|@|4NKjCpae>Mlkry4_ypI8LBY|co25H622!Q*Zo)^tr6{e2< zHYDn#Zvk6pk09DyGi%G0i^iCOQLSMS81b{3q_n0c{~h1KmWSB`2r^~Y9l zMkJD!zD~y^x+S^Xmp=H#S`IQxL5E(3e7@`*-MIpx?t}MTtHXBP5ntZ7VgX;ysCXtQ z_OqE#*$Z<|V25OIHq6H_W8^d?une)TeZjk(agA}zFXNJRx!eSP_?$Rx1eHr7NVEkG zEGtB&TBtv-+K}Fn3Z~O^0?uR+wN>s~tS}wU4>+@EvqO)#i(xYuyIJ|mkn4}pY_H1k zU2nj%6ixUR?y^?mL3dxYnrSNLRg%ZFLm`VpGGy7nA|=(QSv56ievD1{XCNlAt-eVH zHV>JdAUftrtY1cqyCmijfwbi^PB!bj_F0nHA?`!;wKc8Ff$!OaM!H;$`>mQ>+`_dV zzDtc4rGA-!2h4fcgAsbjH0c|SirPXuuac>yO%Gw1s`Vdw3Pmk)HD3)37zn+4Q&=fb zy-da;P{+ioBmq?##BU2vP_#lJw)q4IEH{wT;kD=Ojg{}Uv@^E(C1~ULzJ$-wg}Wfl z+Xf`>2OXM*7L$~!)ajeeko(npk1lp=$J|PnQ(+T?fHLgRV&xPsTkS)m3LE1a| zqUAuc3MYbf$jwg@6viYEY4*nk-`Jh!=a{6G`U2ZpC{8}llmtHyHMC+n0HLBnnR+CY zfDUHVxu?gVYE*(xszHF;SbW-Q?%jXt#eEHlE<9rH(y8KRWB!%Ue~?tulZuw4*b=nB%`uuC-obug-s~m_JPIol zK^RI=u%DsfD<>axh&^oHhLWD5+?E$l9hs zAHPmrNNI6Abrn&4Tr;``Nkib`Gzv^v;7o zRQz!p3ns|(fG&DEguynhIT3GgOdckb8~@qzSOPS_a;6;>BOT1L{ZNTb;X7=HFNSu11EZ`_( zFz&NMK6I>KhS^glayppnyoMebp~H!1yB!3bDT~36AFsZ!heci~o*96sAD;q4aaZ;#Z8NJ%^S>bb4jK+tJzrLXMMjcdHKJ%_1Q0Iy>VZp!Hqqfv za%bBM62KRDYv@80RXUm{juyApJxGGeUeKUBEErGQ;*B_K&1>HR7vl)n*(~eaV%$WK zY0{u|{IMUKz%1fAnpI$&a8Drt+BIiJJJ6+o4u7c~?iA62k zXBD30h?z8)4TfQjuh%?&f4Y}if!A8G?3fWU3G%!?G$du}`h#H2i3|fyktXI1a|$5r zyo*MZoMQsG5J53d!rK~&%9js9KLfB~nctn(yEQ2(9H>LN*koKJYt!2yA3#S(Ec29@ zY7>ZsEZznOm28s9;9GE^@JE(7Dgatz0Og|QKtguNM&!|jV#Hhc%_nY-J7)tZwUdj; zDljt#e9mEl-^Az*op=-ktE70C-VZrzLFMhVNx=0EG3jXhxQPb{Us<7aJOA?Q6}WF; zFt>LFxnydLU}J6PQc9AUEX8|7^>^MYp$}mBKb#2DV~6G+v>>*_9$z>=;Xv=2ck<1w?ip!I z!DYhChldr=39e#;Xz#7xKKiB;@x&xLY7DqYgIU2PpCv#d^Hg&jWtON?|ND4(5@dRW z|8NN&$89myY+@k8aHu=(Nzn1ssWx*Ui}!VHD4uKqGL@33o0>^Ekc2#Klu6Q4d6JX~ zFkQH4sWi8U^cAv-CpDC;eMO>T@EYGYS$Z zo&zs6oDwPbEk|m~>Dyosr{9Y8Qp>-wtiZ-W5`;AMOfTGa-K zBtLuN+Z~Zq&$CxZIr;KJrzD^BBSMgRaU)0in^W|MJz;MQM3y1Yh@CBZzkxZBpQ92d&?z~*Z6Z=m4 zR=U)3Zc=T5!9(kZ3L9!ACYDli#MW=gNpIzB%n@^h!@K#N=SponQ;vj&#ozr(!=&a+ z`0s!olXT2Ae16)6(*QZ_+M66sIha2!3C^n)hjqH{`t{DN>47(V(9+TgTLREoqQ%ho zdptuI65>C)->Va<%?7q-G$p#hrK+q6C&rvSAR-}eQFmIy?Qsc5qR<*kag&>(0BE|{ z7Eo=Wz>@|U7di*~l!^mh0zt^7>QD-7sWSUV?LV96SW|P=H!}dZbA85O` zaCo?634t6gm$OTm;Bx@GAqaLJTV0JFJ2&pk&gnjPBW^cx>p0{-375j&p1~{lbpDqS zUsz=k(5x^>EnX_$JiFqd%?V6?+zL$W(sY;9lhhBq)M;f@26nU(aH3(?CL96KK^7rG zSx^*vcW0WN1;AU^a=u(#&`1?ZD$k36gvQgcSD?ME;Zv6mMoD5c+^|CeK}1{FpPa*^ z(D*^0b+?^Vsz)(TeHhlccLU^8C;Iy{Jb+UXu-L79&X?JuA2}~$BfJ>+n@RXcHsF&Wuk1q?a97SF-E=b&THVLu68Di7V&L1a2&ohylCftA4f=&<_E^a#c29PoVlYX zgcIvZ2e7&R9HG`jTHc(m@|ceBU*3cZfj7hj4N_`wNX6~CzM(>LZ9-m?6{80fz4n(s zIgHb@#z{!yQuyR4|47_Ged}h$7t71%v6RjN&f$E~j|gpJMXgsw0eZ$#tZ=}ByoMGo ze>d)O?u{@|SG(U!M$KIg11eMP;}wj-g*qS1n3MyCRogT%)Cb?z6A$EQ^s4;=3z=uH zlfLh{rPxZ(cZlWP_4T~Ttmy{FD#nHj5QZ}|olP!pAjbr`CFvB2{U!3@%;$@r`ZT51 zxmURYI4K%Q4ZqK?z6k}6JX@(Hq45%}^{tqXGT-);IJa*CAmV*1n88K;3s&f;hwUK)G$7mf5Nkzr0dc|?98S;{3skP(wP-Kg# zOr!aPXT_1eZZ>cY~q~IzduMipW<6ev}U*XS06r!^tdmb(6 zMk#M+Zhe@@OVv!3ZWx9NIt$>mf9m%LzIsE)>lGmrbq`uM_K@2BQDy(x1U|z;!{t`I z*g7r>v|vI!&29ZUU=>yC7bb3FA`3MOg97_J2wFhp(`apHz>ESd9qLPvxHHLP7JpKs z=_y{k%1e?*Q-1UBXvb1_O`S3XynIV5R-%>PP?$+Vg|sta7X)a_&eHo{utL2iLLXki zu$ww{*>#*&E(A*41_n=vD+$RNiq}(__qmxgxuylmMXJhdgS;51hdPIcSoFcEKyN}sL%h#0 zr{8}@SF%f(0+a*3ubL4ExhNx1W!3r@M-H$y1c6mpw_6+(LoPbzJ00VfXe!gHND(7= zSMi%LGw%LK-@WWSc~(+nEw8X#m<+o!aQV)1fL#*3f=<}InYBx%-Yq)N33=Ec=P}_I z42MP4`5p2L0xIm_tnu%^Is!euDN5Z^1N~JFu)KO&+GZAFY0)^hsUb7J6HSrqFj1P7)0YV46&DSh@6;1#1UN3#s{& z9oz$U5CZ+#r-A=Lg|SC(bWMvLhg~^!{ol4q@M&L>{)jpB-DX(s!VLu*Sa=ONrv3Na zw+!|RKSHDt`~;$kOdHj?^#l{nXTccx%SD%(CQPK3vv4j>Ql5OS@_A?WRNAjkuEfVV_&h z(}k}u`J9{O{fYiyRqxp|E}T-^Ai{X5vZ~2E;phTi^(|y^`&<676c-*D5Wg$Ll>?Gu zVQC|Pw+50rzz>oJY40l)bmkmKnyXSuMnQna9KfmUYMq{L2|F*6$kOvgDFVSNuF4(I zeKs|TLkpd0$21p40)t&Qpwc-B*?&@FCnx$w-r@BJ#T$?y2;3+AonaX!8-T7S91`Nw zz5zZctRfdaYE9>g-IAokLB;YR~H`5_*tdrMA@n<7p5X9K(&5kcv#oB)z6PpsCAp;GWvOurddnG0Q_ zeL>*>1;3}cyOfvm+43p|6gymqJo6@)pXdayqX5M76Y-Pn!djrq0(B@`(HA?VeWxZ% zWDN7337y4b@8hlX8yV~oCcOUE{D-+B1?W8H_cc~;ddSre!9PZ|)pFQRrCvPEjZ@;m zK=tM2rC$osw2%;NsRJ!u2OkXake5}jA8KqO_?t*R62P>4%kt<<$^@5Iqn?Cs=@muv za_j+p=T*ZYiew3_;%&j#k-yzR`7QC6g>AXm>h+*l967LRJ!ogA|BgsVbsHg>l}%h; zdBCSH&)MI<32iH>=MW?`W9FB2X={3xYQ_czlR<}iDhjR`HV)kFo{Eh+uWCLr!jWg@ zJ2Dn%XDZ{foOcc=&CF7aTHxz=!*_G@K@6n2-P0AjRBN|m_3iE4za(OyE-y{p*F+e1 zOXWRuk&&x0gMHLd(A)x;5!(ridqd_`6Gt+q-|q_k6o_}Rms{w?zfZd^CD z*2tu%+Jmj=i>ot-9%_cAJT{&@l2XHZeaZdOPkVYJ41uA+)>HS6Ba}O<(*PBZQQnkm z!N1STVIxRMy-GW)QG2b38rRgZ!HYA*nk3R%DNLG41XEJs2#n+r+MYTt5BSAK3EupuCJ(7c#bn8KO4xu>B$Onw0p2xtha%7idkE7G=Oq(x1tw8lSI zkGsF!A9xACVtg?4ouWGT^Bm6%FyS4*N(ss9 zWyoVMH;?m6->4rfX?pe2R`xWAoS^GkcIP=lR1QGOH+Em7?qiRzFeE2g5>cj#u6fjw zo~2C9cd!^H9|Am$0Ist?qQmaTA&e|l#GaG$k#D*AlN!w~knPE1s3O6FI3E$zH3P=? zkfyu(Kg~+?^-B-Wz-}JDxny5Zoj`aFyT5Zk!yZh;iGNy;JboY_@JSDpylRayO;i2| zjOYV7W9%^@;Vf;fvZ;``3yMTv@(JO%E6scuE^@~@C{oQOKs5r47gJ?ye(SBem1P`0 z6ndy=qC2>6I=VmqK-0FbER~Ms&E*%n>dZlBpm_M8r5``a1dI@i&Lbm)tZ@hXVUhU6 zm4;>e&Xyo|S#<7CWUpNeT%eV{?o#8Zj+-Bn%?`i7HUbni2{7!-sBb2Zh46*EIE=ny z*}2pn>FQTiDCY+k_7)69cE8X|H<{`;vFKkKu_A%kU5JCFj-}?f>A`rt;l^~nyTXfN z1EJ*9YoxVGu)vpqWMgCt(SmP59V!q+qq<^`U5IJjjGjpGH0==lgQU{y8W^4y7nZ_kZ9&Fh{YKm(C$@1JR16?FwX`z zK{f2=j#~XDQMz3oA(Gn#6mz0AkvLdVV+KHlTB4-H_%US-@v>o2d{#_^3fr|{mX%lp z6deYu5xAG&C(Fkx1=6gnGl4#sL-Xf!U@W;U*C$^^j`@5SxHUR%IG|?=G1l z^$`y!<>K9J*=fN9q;gtLU;yNbLsD2}{y#A8Z!Xs{NSX!p;f>S`LH9w|(Eyqb-XTUM z>uVC$Kh%95vY28U#5q+`e{;ARIox`v23yqBR<#(i5g1ial|#r_!(52|RVkb#6<&TP~Die|hFQK;D@OG#)LAR$}5wk*s3K0DRhBuE~BuXIZ1Jxr0d z8A56~+CG1lIJJd`m%c!WQ(3Sqk~Qb8hIVdoTWNVT2>nHASbTbV{7&5XU5J}x(KX+p zmAj6jO{lqLv_wG*{F!&uwbd6F8VQt+Ti<74s$^d6eEqltT89lsbeXc~yP05!ow6|}8-jMlhvu(++wJ{wAb@Qg>Qy~fmRR)uQ&f+t zY77G09XghPK~`iy{W>`LW_AuEOot7%AWZDM&TXP7t)MBg-`soyHvfixFs*C82`Jx2 zi5UP1n60>&q%ZDttr{V;sEiD*Xx=!J!G66H2Nu%NFHETkoc%@~CPPc|m&jU80!NvG zJfZlSbpp#FYRQw_3YNF&jnfiZCE*ItrMVH0GSBxY<1~p+q0D7uCR_HPP2QZI)8;FJ< z`1N|#7EJ9Uz5eqj-t3FnfBg-fwywT2Bi?MpVj%Ik;W82BRo4v2JSYaJ4^m0VJk>p>5y)%y60d`(Ib~RIf$r)rh%SxoVz67r%5@jw>S?u4e!B z_iUmsF1pm{cfeB;THE!DGFfRlTWuR?`vCpW_}gXH*sJ~x0Ne*IDx`m)(-^FUrhes| z{)rm@@?Q==|Fril$}oau+qcMaQ9cebm?I-6ys^PVY!= zFg9D}X#Ka(89b@}E6p;vY!R;S%fMZ_&JMtbdDUrR@19e(Z12{4JW-qx9dtW~Fua zCs5&U(flo%H!$bl!Fz?o_9t-s|G8+w_8tCU`}ZGzeXT&4Wrh9t>!rU~bOibBTl(P# z+YjXut3sE5_;t=!_iLL!U;b(PwFJ*X-<8jOBCNM6S@~S`A3m3}eGwh;iOsQp|L0Zx z)WEumAC~@Id&D(sII%ov`8SJ2z(l|0ujMYx1o)C ck$zGCp}{)MOUFGvOP~41^Oy(z>y!WbKSIqY5dZ)H literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/fastgen-overview-dark.png b/blogs/deepspeed-fastgen/assets/images/fastgen-overview-dark.png new file mode 100755 index 0000000000000000000000000000000000000000..dde598a985d81eb273b4162e0ca51370a524a0fa GIT binary patch literal 232115 zcmeEvc~sM9+O};+siVkLM-dS+9jn$sKt({-w6%g3ky->~jg~^#BZfs1lGIwIDni|e zge3|BvV^ECVNFyDs1PEA1X+@Z$dZH*APIr2-}6JAuXE0O&ewU*@clJCrzPMy_$>Ew z-Pe8HPcMFZ#NF`EoBq6P*)qdJ2fzLfd?{SEY&qxS72qcYjepmGFCU=4bN^~tW!GjQ z_~uW@FFn6pw(LRXD$U8|;QN)Q4*H{)E!*=5`uBs+e+U>{woG4i=<6?!oe7!Lu58Nl z!ASS@#&UB2M>O{@=tp{`t;{F3vjQ8lQr$uq?=W&eyl``XW+?1ZSXUG zjx@ljVzTA5^aHS>|Nise6Zr25{PzU@djkJGf&Wh@pxKMX;T}h~<++Ya7`wuz;FGpt zU57@ByRvQLp69}*Y{RAs8hl2IJ8?18QnbaPYsY+@bS5GCs-^`bM$1yd?DXj`_JEQl4c*JWQUSUl!s~4`6e8aLvad*m9Xl6{PACAWm^wjYp{j~eh8h>S0hKr zRr(NQm$tmQxXS@Kbfdnb&}ienGz}v7QJvSAT-K z15H~kFy3@dRNp6M$i#ZJb&leo#r~4n{73@dCT!+^_~O#RB7S!0J9ZcPN3tc9ufOOU z6u)soxWtK?j52ffD7z+FzB+X6Zt+;fi(C^@)qY2}7hy>{>P=SgdJ1FUShG2^;qKOg zmNTq`6MP#pCbsw^S-vm=e7vPR><&IYYC<;TldZw8Am4oVHxNyd+j5h^HFSk_8~k+b zvqSJ2)#~ottvwHZdSV;x5GTrY7!D}k62)!DR-(Z##`A6RTCv44@Y{BRk7VUxWmQ|DgZsR%jy>Ll8|qCF7{N8yKRfjNo`d&ZFM4E&S#WtBN>nJ_2q*vWbmzvw<{b0mPd&gy5Z1w z#fiqh z^J^cd2F!BGY^G9!u6?%NSH_wiBabrS>>NU^{l6O8FI_8G$IxvNU7Lh2zx*Jd(5CU@ zij%^s^Wl6vdy0Y=Uo$9UFffS<>pp6!E4Jf;=m+GZtlrKC*+`qda5^LL*{qo?gFDaCh1ZmE1{d|9jVbEJ-o+)_Qi=N?Mt>8K)ZZ#XwESf zIFV;I8Y3SLIVSsb*om$@q44T+3CNpkX@6p50_-ulFZTmd)feb(5nU^~vpdvJ^L}k^$8@Z$F;)Ls=me|Tz!)bPeK#EJ&%<76kFTRY zs8hXC!YD@=vfaQRg3BD7_9B0oh?x;p4;xCXR?+KzdFzB;)PBA%rQsPB! zd8t3Ex;lNWG<#Mh_Ah^;h?>ne!SQWIA%j(T3v3ipWH+g_IfYeCBy`;%Di5b5X;LFa zL5tRF8A~M89wYD9#H@r?HE?3cq$-miEgDL@fIFcSsA{rTl`Dg48XUE0nTM`@Ch=wZ z7dUu-jl2RA-x-U!ma(I{zitQGYZdq2fHW6dp1D{5q5u6ZQMxOJM=}%tm4c|MjJz1R zTbg}ZmLF;K_Dd#tEe^6*M+t`2PNu>k@!2RBl1hdl!f`9;d0={19_yX1L(G2y3FBa4uy9|{v5k! zD&0Yv&MjUr{t%8oN9eErnUsiWBb==Y5mfh9|Drs+kits01SclTHwgqsHSS_l3+|8w z!pN1JJ!({;V&UkbIR#|Di%Mx5?Zdjp3ymg7A*o|cA~ON@zT-lM=db||EObk$L+D=Q zm9frG@(z$lRo^()zE}aSD$kU18O8g&e9RP9khs}Cu`3t+X6o;kR4fTS!^v-4*!7xJaN;#v-iG}j?Jx^{Qp_+oZC0l1t!Ruk z9a~OlJ{dbdoDV{@MReC@X8#t)>w9gHS!0veGIpn5Mcmq>gcT^}#THs?67hIdTsOu} zoST*V$!xx`vI*%1LS^!(iw&^e1`BC+_qYo?IE9t4VbLz~euXy9n{Y^EelTXNd^utq zGW&$c&=8fpWP#sgvr?ySno2~S;gln^rfO{)>hQ}^yrfSR60BDt24N~RIoS|yv^SH1y!4>2v)TV#akvX9 zF_K&2UlQI-N{r@G{oBJ`(!j4~{oE`PbP$HOT4TB`LIyOt=_d%?DAAtb-X!v2%sn4d zUeqe#r^sduuU>ubOC%XbYX5Mft2&Qdgxha|JjP1y=yVfdJHzOOU(=IlN zQ~R0HtC6Gq!F-$Gej6RS4IBQ3ROPQ#v!)j?tQ$NVIx?5()ZwN5LU1w<&C5pSFIWg8C%!HBbxG9}d*;Li)z97uB)Q!GzV*2iSjSnX zoaXy6{j-fkO9Wo2RY+GdrK_BDRXRON*H}D8jmEQ%Klr}Y%)+;_sd+(B?n`N?NvTLP z$IOqV!SM42<$hFpsjhjn&D}gjenACCosoCDQxOcjR^dUkQjcwiFFYd_5x-o;yfwpI zY2+e*VH?e@sHbNq+bA(+BHJ@>+Mug1)vI%I5~HAa*r+0ZMAI`A2s&DulgJ?SPZlW! z@RE?KdXFgvAw;VSd7+WrS#bHg*5);RjJz&UDM(aKs&%7?4AXC{m!CMcSHJ4pZi9GF z{m8Y^;j}Zr#Y2yxH(c!c+>>dt%?!CQOwpu2J+H~hXzbQMOnGK2Hl{@L5XD7Eds~~$ zc5Jg0){Q)6=R{=qzL;XC+58<&z`fX=r(QyS?VZH8DJ;nzz7O0UTRho>bUe2RfR1#6 zRP|#4FM;i8jZa{bmitJAC1f47Hjd|5unC-mHVCuHvG>w?3s=`v33(0IrlgHzE1qJ^ zKDWq0xfhtV3?E&y*xx*gWY8OT=T8yRuW@>vlc6e=`4~S-m9l5(M6%gttCoXF} z!g$|SWwi7q4fQ%0-v!$R@d5?5(}FF=f>3J4T%E?^V4*FWnCTqYi#g}*=DMKGeQ!(b z_Jz+rowN+BJ(lS*cf2FY%4#TCfFmFB+MgCs8`ICI#z@GRC=bQcEtyujIT}Nukn#p7O_mY&A9541Z!ffHfJUszV9kk(=zpYV{|xx)X+S$3Z19 zl{Jjk6UBW$M;ASe;>OPp*ok=+*ntn;fS4(IqU`K-@B*phN9AEZ07D9Ir?7e%($(a0 z5J}&-*!)?8E-nXo4z?IVx4r>+>x--5Op@XnF!VE(9TClu&#rCNi}nOa1DaLc<%97i zl=`~r_>ne=fT2?*mk_^NrJ1g5bRE?2O_{0HXcNPYo&Zkb669wCufzWO&9%EnCQEcL zrYln1OG4LJ)-IMDseE-F8dc}^k18nM#aB&5uKE?A5j9u{NO4mjsnvi850_?h!5(w~ zM7;&v*Jc4KzP+z1oHu(7IOCaoXnEZQY8^AInqPxah@uC;@=6z6zVCkBO2fbU2zKTq z#v%PtZ>bHU=JaTpK|V%QdWv6|*Aek3!X#Oht5uR5m3M|=)r)3S&c@H*0*T3;o9k0x z(;4DD<~)=e-xHs<0h~C0Wy`pB&Liu*xNr|vy|qI>sIs#a=hrUF28eRRg*^(uN*19j z0n{awQ5UwhRhsRHvHO1Cjehb<*P!w>+OZx{wUaqm@;?g3CfDXg&XM-? zjfY#>#fgT*X{?39KFZ3z17X~X@oC#Qu1o9(dw}!4KVay*m4}r80tE1Fcw}Yg%SczW ziK}?qq9qqCTJk4JA#Q>7FMxIIMRqlLHGswDD_-n33hwLbh;U+<#kSO1RMtINUi?X& zuFW))GG4E}h{#y~tY=G3?xw zo|&`_8|@RPa4krU5Ek?C|7*3z4CL4u*5o$G$N_+A3qdHMLirdTU;C$Je;wAySo+cR z7QQ=^Oeprud?&p(u!agGC7?#60$YiW$|rOewEAsEckYH0LyS!_oTloGd9)v%Ntn|5 z?$q{wfb;b=1+Y)p0{<2}lPxRDLo#G@&;Vu#c#If-6@KOSB@)5FpK(}COAkyrs2SpAEI_QeHI9K4tTtHXkG zHcwOtvR>Uz$NXSr^tQ*Rsc$Ipqp*A3SBj$Iq~+@4$uf%I3LSS4{1v z=CkS}=%F_|BEGP=;uNaOxj?wZm1jsRKZgGZS;n2Qo$K=ANmX8sD|Ua?omyD)ZT|sr zF7Ueoe>WSyKmOTr7gP+o+srKGw6?hu6_g%wAE%;#M>A> z4#jBqoZTa@iPi{uYAFbU!pg>}g&}XR`gCp~UibK>1!{=tQFvLwN}*!LyH`r3Rrpthd)v?=@V#=nG4^?PRjHb>cBCn4Sps32f%_$%cHvXwXj^6{-2v&e(4F{&2EU${i87=kGS zF1T#v__t=a4@<~+70OvY-4dOX@@~mnYWkif(^cDD_m-<4)nxt*dkA@55W4MRmn$>- zbF<8shTVq?7s@9K5|4mb*1fYZ```G?do%60b8m5p>#$d^@j&;(n$-a((c8Xh-;&faonm*O>fZ>uzYo%l&ldg4dUP>jx1o2}+C@rv z+Ngu%nljQL>K(L)4mOPGzuoa8TV!Fv;lF$DWkt;WaXbCJU7xU@hu9oZ4a6}`e7>EW z%H=D0w7a8%ha-CeOyX3JEfI+(6lC)&oth}%ULrPfBIME9#Z{tt3UerrQu>e!0cYk2X1JSSb)`yK;p`YYnmCOf)7=W zb9(~3tE^#XPPF!{^U9rlD92z2K=G7sY+0|@!Qyxgwz+)m75Vi5LqIrdI`wuM6Ck8uOEC+6cl+7 z`Q4@9`cdxD^yQOdIVkN3OtXt*AeyQjxdMQDv=#1vEGij<*~Vxp)%px`+UOa+bn4BZ z-Tyt<5(o8axREXLZbH!CTf90SSdkk4_<8=z`M(QXQ1jq+Ct8BgJ-_MErEQ}fiKCk^!}i`@ zx7ht_f&Wc?9SYQkp+MdC``gg`wpm>8?UpL#6_3blAF-e>&I(6brsl^2>7T|3-IEazSn>{EN>O0*Gj|)vd%!D z=C4ny3~roz=8JO*kZl4iWvX}({B*pyk8d%TYp>^1Y2RM~eKr_oV%XhPmd%>P0z4wq zg*Ssl>iD>>6pi1f`RHLLg0W4r3j(R8(Os@Bto@)gvh)%{l*>8+x*Gdv1fY zS!oqi71nWmc=%P7~z7-n$pYi_)t3u1;r=$TO0)WutA~80$4mq7%Ze7T9mNuCg zw?WjD+SD%QCBJNymlui`U>VdjPg<g zd_M>PdEbL(jh~_Z$4Qwzb=6XPPOMIY?Sm3qC0#XO1A+3w(ccv+PMJH{M(;xLn<5o0Q~l1h zpt{&@VjngY3Wb1KOVcv%xWUf>JGjmkNgf`*GnRfM!{djIuI+{8Y9!HGr9{)H>;k)=Qb;!R9!5_jc>X$>+4YX-OuA6Q8%hH%DJ`JqX3Fu>zZ;z$ZQ> zh3-ZDame#8DP$2YLDAgO(Wt0!?XpKN=cyeL5Z`mTvSAD|Cch5rLKnx$Jp+w=#aUyk z;0AdmsEaNzw$a8MxD*6qv{BHU`nk&iKUF)yd8()bl-50$lw)kgq_I^{0v-yA8pMR8h7KuuJ3xh! zMTqkTVmCw^YG2QeT5PC|BD?tOLu!&Fq;@O%JT|SkWa1y&^Uu7l0}MH%alK$$Uxnm# zt1>tMKFN}#%Jb;5<_^~hC*ZRDRJ@FS;cW*9C`Cpbziy03Td5O{8 zDW99&T(+zYLZVYm%T04unRma7q9A~>Lz+$vzji3Zy*>KW+c(t}p!!ddjghaL)60X< zyhJHx^1)Ta+GuX5_JW~TFixQ)ygq#ka?*>gn=iY5y<{d1K_wkvR59!de?Ky)OIVm^ z5kYS1F?;KM+gx-62i^RLGNk@A=b^8phM8mEhn>9IPM^85s=cIS;?M2o*vVC(WA@C@ z3(g~<4g@|-5Gt^f@sH3wEqbCfRf+`x(`<|C)K1l3pbVCCx$A2v)JJ4=H~A!WJC;~} zQ5=!g-TCi+hGaoC6V5j>zFD>`bOZ!C0Hn>6T5tM!qM?H74IgXFG`rjpaU~GN(ArQi zm5clZdc4c220!7F?7Dve3X_9Y~Z{y z#~XbV)jvC$+$77BZYG+m8Sx+_e%759+MN~w<<_7CP;GDB^EEnsjPl31w%?a7i=VAP zQlK0dzmst6!(|&y$$AYDVUs-7)D9xKCovd5x*FPqhw;wi`%o2t$(W*J=lp89YKjl> zHp#;;nU~1&W49Pfep+Sa@I8}c>*ZXZf}U5B*z3GvDxqR6l)Yd7vos{B-vA0Q2js)Y z<)4y(f7%-_fqw!Tt{Fg^{JDn@=;?N`1Jxh-Q7{qSGD>)N=$l%QCZk#}-OTLFb&w?r zcU0g+;NwoiE z$+cMif=$yl5IQ$mkF0s!!^w;a z(VHa?Ej*J?A8?FrOSgB{Op6S+G~ewDvWhkznFBoD3CzT=@1ls)0@o1Og^SOB-Yzhl z588S1*XQUDU!&G&l!KXv^6ouVpViG-*432tn+W6Z$lDc`E7Zk!Uvd}|+o28~wDnoH=#{5q5Y z0?vZWI9E^x3VEwzw%GYBz)TXs?KnuAWFfLw=Pdft<;qtu<}gGBXQNNO^H@Y<;u58# zA4B3FebDYa{>a)|k1+7;Pw4nL(kF4+zFc(Mo68^2Bm?cin-{}rFkQ5h?#WR=Obtvr zZSk&zU4ZTdePj1CckpR!&UXnOV^9~ zllW^IFKKXlz>Ike{nI#3nU6)uk-P;P#-~$_f<&;}%}W}e8hD>#Cud|MaM~-S`Urv` znWv(7t|!BNuE0yChRPi0w-^idoO*L|+$(oqJns|t)m(ybVxa&k5qdf*=OL25JeJrE zVD+zL1i+>hzAnk=(y|mCZ0K|cytw4^C*yP#2a)pXtU*@wYnPX)>>*I)ECgXD$f6z( zYrQ)tM2~!e+qVJt3t;MdnFo*sfQ{L|w>^UYzoQV}0%|Cl2PBw(;|>U!%i5u{`V=Kf zk@qm1J5Pnz#6GF_vBd{8{KD0Y_tAW`Qq>7X&N_p#$B!R3s?}=3Zf(QSZb0sj=l4TU zHV3$4q&2E9thz0X7`$Kr*rg#*1s!F&hSbI^XxM<;Z58ft1B?uVMi*r8E)=RwHR0jq!<%A;T<=C;3QLxNEFau*n_OW#uxE`M?x)XOaBIW zK18OV>&xQQN>tr&9Lt3sN!KHrt%-sZC^LVcjxI1@!|GSc9=@7+_*UYA` zcFakg;`=X-l(_09GCU&KX~B-zVmh57(vL7G+l&nN3NFo^)9G~MMQNLDovaFZ0A*8E zT}mr$7G$P=V@xvRejd0>pj*mf#9IAYBl`Xr`avv^%+Tn{W+^TXTqT<0)2HJ%0Qr)= z&1Ic=Qq5RkH_)`H=g%95oH>J;2!8#}N(+p-fsCo4e&R4x!T&P0ks0{E>kq!|6#~#6 zc)4r8@ZO)6-H6aD^BJr!x9`+u&Uo9o_6B{GvujJ;skzk(PimPxZd4aFLDS!M#LWH{ zb>hS;nee1Lr^Gc;qf6%DSiTUHNv5NP%R%K|X592WU$fkjkv>Q1?1(_92m0{Bvj*nJ zd&_Vyn=(9e>|(V#CF&{0EOhi%Xhurclk=CYV_S!4v4g(guwH58FB&_S9E5%TY@V@i zCd>dAKkTkLA~>c0IXi{;0)Gsljb5XrS(G0-r{%nO@q(?@YTNIo!F<5R?7%MjBZA!; zcd6IDw4@1iSI1151?{{%5~Jk&Dq}=*ySbKi#H)L^Rs9>RJ81Tfm4=6OM+-Pm?F58> z*MZ`Tbi*72fb~L!5@HH2Eom9Dbpop-;^Q^Q@#Ye#T%oixevvf3c7R|@Da$y+TFH6y zcngR$nVUiQr+(vQQwEYptY85WbcvT!Z#kiK1z8X-m=mtD(sjX#w!aJ;&n4(w%gr$n z3t5(dPujaEDVCrGH;8Fn@0T#pdJoCWb0UOWg!7%-3JzSZoRqQ{r^eq|?vaMD=A|jj ze>{@(Xg31aZ3>6$tyM?-la;_vZO(OsIe3=CMmjAyk6_v`+mzQQZ1?R;h{cp0rOEaT z#DKf(r7>j9s{kDG$peKYXmusy8x(?!+UcH;*6e%bAlxx9WojI<53dk$Ox;!LSoFo1 zmJ3$~E1PBq1fE8DtxImB6NlS1Lubj^vJ5)FbNfzWd_4Er*B!4lRd6yDW9v5fuFKR( zL=(@hLwF#wWc%hsC;%n0*uYq!s=r@`a6lciDhzH4BzAW>iYxjcnNYW{Hl#PGIx2*f z%)l;ui@8+WsGjOaEEb*^bd;6E(*L2n4dP3e3-44deYKo7c3a=1wQnasRL#yHQ8>(R zf}iaw=&G4D>WSr?yN}OC_1wMarBEupzVUJrK}HI;_prMmy%sA-=hltqJ0PMJv_dF! z{G_@=h#R-5W{op6qkT}plol`pG%WT*r&=0&_9H<AMF+$V4pY2`Eg!C~e${DIHQ2|uI{YG$g zKTG!&W+G-4&843v`lAh5RKuN_CHGz5xJC%*&0$uNu@t1j|c9K)_RJD9MHXd{=no<`otv8D4 zf4DXEO5;GLRVFW0Mrkij<6HiJ*w^9t5(|8&=Ay+B6S(4HyPINdKp7RUu_v57VjKVZ zTYygQMx8n}eNI@Ee$gxSbrqE%t8An}JU$&LdyTN*`Z}|*2(fixursJXh(V{Hom7lC zbISzRN$vhn(*cOe4j=T|Jr>uj;2PDw9^{jhqS?`oafTWvRkwl!IC{yVQp14u`ghf- z$-9AEz(pvI{pY!ejnM}9AGpy*&L{iF+r3g^m_MV-k1~FqD^!L1at2-ZHIw$eM0kRN zYVfMN`9hZbTHDVcKS-{5(6JA|2hjpH`E^+JKt6ihvhMCpK?Yaj!bh`msYUHDZ`i`q z5<;Y~Ck#!C<87kX22<+y&AquESQZ#pFM+Z5OF=?i;$q`O#o_rE>#5EYKdUpvIS+-8 z&4@3;HE+&8I2t=JC%OmMAGA%Gz6Su!i7x?v@4@ykUH3lGe;(u(*f=M=OxH@_hrR9( zg4}HzsZ}YwY%@(<{Nk8Wl#z>Oy1|B(s}l=%v~}9{1u2caswV=wA)#6kLi5@fan&3{ z3myX*@Jmo(EFE@BuSMt^s}Os1mpxqf>Eu_Gw!|0PI#*j=x{|~$7yWu2R(S!h^eTUL z{Q{{Yf-NT2j(7R(vvZy4XVGjsp2TEpZx@ z#q0fo74D|?aTVr9`O(Qy_{=~c_RY*q=BH2(dF>WeHi*gugZEFiu$eIhebwc${SS)* zz?er1GT%YQ5#DkC?OoExDY6omh;Iv!fachluCwHP8FIucFuCJrbZw0S!fl0R{5f4I z=kD3X1hU=8sX~86IGSMsQaw~MRnScwzz9hym?i*5qxYeQ zH!}oLe|q#I?~<-mKqnrRNl*yK`$>|4?e;)3LC*?~ZMSa#^ERh|>#la?z#VJnC0?m1$57K^41;nvJj(7nq8 zJ~#4Yf5PJv$-ppI8wbzkp+G0BFkk-~7tYFVynEfc&INNL?K z8-hQM4#nA&izwu}V$#07$3U`}5f-j6(C=)X+?_a7;;JZ5<|opCeFCYOx$h(3giL#Z zCYedQTr)e^on=ip@yw4tdy{y#B%*$mb<-(L3z%d6?I2@4WkHeYJo~g5l{)fdMtCRi z$&v$Z{3WfDFlCOF!%>+oYcRv+*ru={@U8Z>`N$gFjJdVrn<8w}^lb!w*qWgjwLT+! zOg;0=V_VZ~t6DB;R`Ly|WY7KRLLcfG$){?-CawcCJ%jEiZIpgV?e9NJ1#r$NRiabF zGz|}?e%-%jVL_g}Hwr&Ow!-fh^fGgtM5f*NUU?a25bIS2$P<6VHEUc{+`ScU<^b+%rD@~HDTu_ zy=ie~3t}D}hm}zx#_zVrf|#+ysrO7c4VNNd44iuAnoL!5Euv8g0xd6uL7_U4Ksvhg&l8EdoR{lac80!mg z>JfO8hnngaitF30>)-UW^z5glFM0FUxhfi7B|tgjXZYgs=()-=f9g(Zd5FKu>=i$s z?F_vF2RgBxEDpf~JEd`ou6HqVjO0|GV;NeVt7#PaD2k~=2u}@DP`S-3|Cr&N5q?ke z*luWq0ZN7I6HH=7z1h?Npa3_Z#q_UrO^$=BlB&g2#*DPbySft+)!IO6EwX*d*luX( zs6h28xkh$1uzl&Bu%Xo(p%{(g@;P(T=7)7<9U=Y{Qawu1&QR8x8o^ITwB-m=f1?&N zC|QoLs7DAh^WNO9>|Hi>tGQz;QD4GDJ3ZKQ^QN+Gp1F0}dxaP979bKqTW_3xjUqra zvJaR;X#-JMd)^lh1A06L9n+x#^Hb=~h-heYW>XI;p=a6;7-zN4x`fa;F#ofXu>1qQ+ zx4nk;oz!G|#hAe@>3`vJi&4}J=Q$hB8vqjkFR(t%%&rarqDR-e=wSgQE^?s&qkZmE z&ZW&N%YY!c0N|czoio@y4KH6V_$V z)};YhTO}f@EAptANdOlH0ZDN3e6U-&1auiS)HvP%sc*1X(`@bk@J*IrAy1IHB7BN-un^5QdeA?Q`|d zZHsVD1IL5iO!XWe<0Mc}ffS+`v$njq?Fi|ZXQfyQ)11LPi=6?shZ$%E6-vdpi`gXBWo;q3OBgqBQgMRh9HP*pe#|PV@!MK0{~~ zSB$4*5r^Jxc#ZCj#61I!mR%nyX*GT_;4foFxNFIoKGzL0%T^c|5I5VU%vzTac^lr< zpM8L6R(BN%kO5$~9N|dHKRzq0N9dGYZMQ@lL zT-OC<*?PK(k0Ohuy-(U3|2hNQp@1B6dJrhWpzssC90A;tTj75lG$WM>vfS?IXOxs& zcUZ7Y{*HDH=4wdLxT9@P-@`<)HwVnXkC9V`&ogeBV|tgIcKsrMx%Zq)5y_{GR(NGD ztokGVjLtX{j$>b1Y4dv3$yf^BGmUU6z&7&L)2N7uh>qbjC1XePWbM0(tnnLq`rH(N zpD7BdwI8J8NVRPAmK~Rkvlu5bT~y?}{BYvV8EK;fBV<{aqGrNV*Ue=7P5lx4M5#h)m8V}pwyXGph=w123?+>bA6!A6MvUM zu#-MEfCbgdJ3+DH2b^!|`{+;Ym2xdT;0d@sRao^pslE>ae?QeU0tv{|((=VVdZPU9 zMb4u~k4hCCJoI>Z9PBOnZ7!F0M7&q`37Hg4QuUZ=S2!!*sSrOOtZJ^;MSGbEX$7{% zPrz9Acf9!VJbk`-Qt$gBddbACU(l=cr9eE1<0h)WzCd^g_0lfTsT6v`R|X+&;v9b; zl$IqfKeTQ;M^Xn2P+m6pt|*dj0yCA-?UR5fxB%0<()Y5OB^NgYmqp=Wso))jD}ts8 zFo?M$zq=nv)|Nx>Z46x1@%uF@Fstq==TP$>ATSe0?Zu+lq3k3-lpf;T z%+pFDfMi#41=c(}I^xBt9KjysCu3|=-uhlJpe9Y(9dQ=#Z_m`Kuo(b= zlx_fCIwMDEg)Yspfr)*=$>eMKuT8YjTMIzy@4cnr1xGV5RwKYEC#~uN$72@nu zTyM@JGMz?p-uo@6bNT*^A*3XgBN=^Lr6ebU(F=$xP$e^^2uj0ojx<@}!F3c?JMI(y!*2&Lz)&kA9hSkjk}P_rep$XoZH2>9sQyl-ZeR3~$Tq1w#FBA`Tf7 zU~>|a`e_A)d9Q4Xes0zKdMZt&Rx7=%8zt{Lk(F@`=}B!S$Vu8m?j@8(bd*)3t@QF> zT;m8E+NYvVAoufft#{BQIzWTKBrqC;7N-rN$G2pH16*PfW@EOE>-2zqrJn0*>I4ca zH}H0WfWX(KtqcE?@#6trbs1^b=!2-~TcPEJ+t>J>JGZ}YPN{@DfkX&+0*HETYyk4q zk*-Fm?kfxBY1wfwa*}!i;tsy?deE*G=yhNmw^lQE6~8LbrW}Gdy+u{wW?AN#!|$_& zwV>K0k?jb#yoyj8Ncqy0n^ukUHW7l*IA` znR1g>(Kx`Rc=>^mJ^DR{zOFXtFf^yKq@wH}uR4tsGHHd9<~wYlYe(MY}!Lv4!U$&^W7ZG`%ytmAwjp<2|SfJYmq%wMz?Ugwh*nw{v zYX%LCbILqNoXR5ya4P)sewhg&kw?4`NuxKfCpS&ly;rU-nb!PA?k(&9zQorrkqgp% zTUfPVzNDPL3;%pw^tR4ltbeYlQQ4X;5cMIRo1DuwZQdLJBr5Lozc){2fYEx0!AedX z39Cj0of)1764xJ~mqS|mZUv(K%NLqwz=W;K+z@y{>7Gi8@{lP_v^9+bpvhd;df6Be zKlz|tm%7M?EVax#V5BKiJ8;&tOMngD4k{!tc;PgSmg!@j$YWZcieFG>fqGQ-R+tnOuyOVw`7e})CNjkFfwyE+A8$`H9mRZVVFnFJYBtkXM>%) z=@o0%Pph=Gfe0#gUl5uJyGtAd735;xoBS<@dvIx}ivVO-q&kle)wV#dxW;it=e;f0 z7^HPHRL3(BbtR;Ik*62yc+e*=6mCt11_?lqs<}K6%;|v#zV0W2%7+U{siSoK0=;sG zP#tK?Pvrj~W^p#-Wv=4DL=jvWxvcRKTz<3gE_95rsU_)YbP0Ye<~xnvX0lQz7O>WL!D;w)>eYe$$?r4vz0pi z)$ztM1nY4Lg|dR6%zW3;MtE7*7rws@2!H?yyM`zKYpe`feLeKrRARY~A8=HLoqYRz zoJS16(u=9rPV>>?pq}-AT=x5gnCY6x@CJ|AmH!0own)@?Tn1{m=Hh;tgToaA)r{k<@OYSb?nE7`p`%-U2rp~whhVu2=RREZz;Mx-3V zq{J$;VyfN=+6+*pFXd)D7pBE(E-spUg)=&{^}&O(o5b!A^)PCZll+{td<6L$L}~$b zZ>;`II4E3x0SZ>T)oa$6Ukgccf6o~-X!i`nx0@+{Qcne*9)wideNMi%#&j^}9>l|| zjo{Zp0tVm1hIJ^+u_b~Hj#Cba44bm@xi!`oulRs5^`uwrRM)%GPD?LPod7yA5UNh* zxKXe2VD})6FR?oucIwdwtV7m~|1>ln9%u$fuGs6tv;(!a)Yy5kSvgcKQ_f63HaN0x zSWYYCK+k5rAEs10bHeMAL>&>_@_GD-w(W`{0Tqn}U zh4C1D&AYS@Ykyg~>D}+#`{^>mUCmH~r4Zq-TB#r?Pw}>k^=PWRW1tvs=FSN|MvEOU zS04g@a)Q?`9>Ck*#v8i#1!si>+|x;=RQ0#2t;@F9k7}vXKms01Hv>bP-Kq@4{P4>0 z<;2GWL15qp!}gn^ZZp>ga^4)eUUv5ES?glr(AD?lLr6oqo$Sdf@Gd5&>}FtX%u0kh zqPJ54WJ3;Hnt#p!GYJx3{qWp5A*-SBu1-wjX#G{Ap_TZlksgo!tqKI=!IYA14MQu? z1gqv7!Qbf>%%bYT9`KVT)Tf9x4$@G1mYlj>a!o}=cq<2m#LRkCJ>y2mbE{IlYBW*% z_6J@b)?vh6f_A9kT#vQE#%zZYNirsJl<%`XXrn1JAkidNbEw=p=Tscv0P+-o1OT*0 zPgKvNqnx|othQwAd!5|lF?H~tp(q%HA{XqOaX@n)xfGhEmY#Jthl8gxjE;B-j?(ow zVs`{tWmJtgM4&T$^#ewC(@bm>?ED&8zl=jwQrRu(*(wK?mlv6$tBGCEIxXSgj@e1| zoW!PwaqR~9yQD{qY{Et3P5WGMwW>bdQOvE~-tciBp|Gl#5f{umDLlC>Nsz7AsAj}0 z0Wvf(B{edpI;s_F8E82OVb1Kl_AOF|(*IM2F2ObuNGbvD?}pxJE3|&|9-1{LnMgW? zY=ZVqnt6!w&2><-37(U(t&ZP65w5L<(`5ZlV5+;?Lp$D|IjISiOqJ_*Ifo1O`NI23 zde~F7s@)fQTCOyx>XXz3h(etrN7SZS8@q%|?LZtjCd5I#V^18IB#0Fby9H~5N{UN( zmf&&kinvaCdi!OM?&$Vm6Q_*`|*{){qJ${G(Z zhrcUEhDhe8k8xNswQvXYhZf)u!T&E{$zG_a@$vEb$VqQ`_bQaMj24$tsyi`mlW|y! zVdTACOW)hIL<&FsAW2X?5(3AMELT^+IzT-L42y2f(?2-J_yj5_nY4mt;bmFbokBu6 z$vN_S+g-b+sQ|w0{3m>|F3iK~77V}(%p^mKPCL~eJ?4Po|8$w{o%9*-xil5)K2oA7 zoP5e5Y}66u_X~^h8LQarE626R2~by~*9)!gLYmgA6d6MMAh)HP{4S-`q6GLzmEW&! z75ECq5hJH4(Rkes7v3+N%)|uD%cI`#_E~wt7GF3wu&3q2Njd;V$-arBFTqsYDyu?C ze{xTY-;sUx?;u5*R(1f=f|h|{0!#PJyLatw9?N*oohyKdpbbpAckveA6WG-QVQlb- ziSAi>WkXc2{@dMaK_**yj{IEAue?flC@+h2d%-)}QNj0;*q$KL1YQ|Bw zD*49`4{KkD*M#UPICEGzXfO!Fs?P})TK(jCGrx)6_Ag_>2!g`6-RJrW^`Z&}hr=;$ z0nBy!9j)mm-7`>;0UCqCB0qC~Mi-OW`d~sD`k5_pXoF=)s^o32TIdL#o=mS%8C7b9 ztoxSU@Uaed3oS_dl5jP@7@SMsb#u%yu++0}mkQnw2nBlmC3_DewhE`7gL~>#M?HEo zX$~&(d-q^^pa`w`T6+c!x{lzO;N-;%Yy8w@FGO-Ht8wjw0<2P{$|xUf;gP=cAAI3+^GN@|_Tg%nUp2fa8Cz*}8y%DVKn<7N6+13gBYt|7dP1g8~Z@W@?8Rw)b}gqhDO zhpFe~d!f7JzCEDHpH0oLE`=oFyFmzuYJ~`RR8ezJ3vK9gjgw!_+mk`wT{NJqqO=to zv%%YYx^+-F+urti@pxi2e&(8?$8Vt#LK~d~N4ZIS0OE82kDz~;lL6PwkC&7Pk6RV)=DLlXyP(=_=uvee$ba}*d0g!vpAUqCEJpUwi;i~mX4WGX?*seI+oc$T zGXhKLyIT?od@KJOBg0+aW4l}v1*M@gl87xpf5md*>oxT2FnZjR`kC|hlfVmmPMUM$ z$F8r)sJL+9f_X>t4}Dhu1VM3z7(CYh^iAr)y+;Cx9Y)W3&Pmr2o6t zq&7^{?)VD*!=~W!0?D z_x5s0?4_@5xEbUYTEQWl(lX`9myntG2!{hJ@so^0UTi2P&KVfOJiIv2N1=h?5+dlR z97*9c{CJ3Ijs8oCN}|S!poR zlK?SKKk@8Z5zW^4SpkC#hJ{{m611N7S{!B~15A;IQgl_~%SpOVYHdaEkN*#QUjo(S znKnvmtJAT{Z=IH+EbUmeixL$DSwh-cMP;j^q6CPRO4y^a6Ow3cr51sztr)hr0Sbhu zERiKds|A%MN+5xR5ETI-ga{!CWWCQ9aO(Usw07>$bM8IuIc;k4`MuBkywAS9sLiJ6 z4+jdy6giyM1qunv*(eLy81#0KjB$~FgtpQnQU}%;nDfY zg?COa_0j3p$u^@ZyC(}8{aHxBNvE*a@0AN&4}4IRU|5E&)pn&PJ?651u&RLPHaN9X z;#tl*huB*K^=il(BNoN-{(i*VFk~U74(G~;X#bD2w9(7ctEb9APgS?DDXYIw1nQzD z0BM;pZD{$3ozb0MBs`THiNM(eOh&iWl={Xb#?*l=n*3m?B;dFaDZ?19ciKkXT5!oMwkzZ{OVK^wwx-)IML#}0ja89b zu@Oknq=kKCHCXA(7>4e?mYZprZ|R+U;*5ZJ>x9jC3#wWX@T1imXUdz^w~o;f&y6r{|2;y@0LbFE#gSqtirQQLnQQW8e4I~ z`WKr$%dsK&@b+{-ZdEAWDM*456b>I11jRDD)He+qO}9hpD!2e5MPe_?8N_k71yB8` zdOQ~whfh?8D;VgN=+@I0X|kS}6fPHEM3K63*QH*XP7lj}`*tGG4r!>FvgiXCSit_? zD%8D|qbjVL)nbHzou9m5zCv#KIYz3^;nZU51}T1)Pqq|j#eB<}A*{yPZUOL_Fyp?b zADQt_CqmLsY$GoF1zqK2X9G2kC(( z8TF$uZh;piBJuPW5TKy+t)?XKy6V8O4*TYp(2>eZn%rIfZ7`uy2xVm&3w2r_)JERS_GBH4 zy(PB5c`R64xM0Bo!M^rA$M6mh(4 zcVSymfgJUkq?)0lIg9}+0QXJItODF$SExZyF|4mX)ROFWWYlNejb(}|k16XYwGTbI z3nGy*1`tt;@F*&L4UgV6r>2`xbZ6ZRFszH9u!i=}xJ=Aw2#!9XHmxKuG@r_6b*VWl zzFi(hORgx9TCdPvQ)Kd z6;{b5>ZIpt6ZpVXpJ6WFBn3-c)Y^)_3A|`9s zj@=nOk@9e|bOI^rnfA=c8QbirPRZP^_)o&X2EU3KhpK9Jt8VLuUdK*?e>>d2YXq5|Y^QGYm2{7D)m7`%inS@18^R*u5}u znO@7HU9?F-W%MY;;CY#%m8DP+`tW=Pn=UIZuuG>|A%U~``H>|vDTbW3$43vhU4QUk z@k9-JY3+M*zvU-dN!oRgrMy#DX9R0o<=&gvbL}obYuJzPM?t}>7m*cMu{aB{`W_T zS%z_5Es)vMs}86p++8LbD!J{o4c{;ZoJidxui173u;D*-V2ku|<39DT0LaoIaWEvzT2;9fSDtB7|5+(I;U{!Dz*kRXf4- zfp-3~s|(apOz{%0Or%)P!vj_MEdBSq0GPhPv5taLC^#4Is@6z@s}s$oXHYik^k>=J zV}*$lO4sLLnO@CYyMJq&7&F!fk`%&Kf21UJk}N{XRpmP&P55xIV2yT_0OpU9zf5d5 zGxcMH!U_9yA<{I`SCHWkStVmFBh>JBJ!^|~nMSD}r{50&$nI~5gxe6+fXiB} z(n0T#_6%7c!w?%PhPhBIBMnE&oEzK<&x+ND2cB%2Uj*5FPCvpm`Jz)qhrwfD3T zMCp$S;9Wu)#$5Ht2**O33jJ1DlB_M;lda;`PZ)i-j$+dtVg3%pw;)dbFFJJkSm^Ew zT{pw&@s$*1#}9Bf3}Qpaw(oJxOBSAml5V_;Bt4$Br12P1f(&(2kHr>j7@7-n02d*n zbU$iouz{(LlkV^@=F{}mNq;ezAOn?u2gDN%3LCGpbM%7ROIh~rrzdQ@X2I4u+WXZ- zcjnJRvRof=-E5%P=nn+&K;u}ue!gA6Wlne7-r=6ohq_{@J|o2r(b5fSt}v7EZi>y< z<#b~^b2H-zSiwh?B@2kff8tK;lQxmxRyeO zWm3Siu9lma5qfCLZVJqu9esQpIw3%KK1=sJZ-#RG4bzjSF*&nL9A0ZF{fTA2J)_?2 z9=V1F1rvPz$T8d#sQ0H&6j=x>25z&}>7gSJwg7aCT|sg-09pRAr#{l@1Sz7n?HG;l z{t?Lez_=a_cirIr2TLQz%4VLQ()5k+2eq%I-wV{1IkzqjE&td$*d<+@1C#O+g^qTENr6EYq9(f!L;+>0B z8kWqQIkPJx;K1;P@l}GQE!=x&Hw>>5{4^WoEH~qF?;ZL>+`Z59*H=H424fVf1pFO2 zQBNn*Z}z1y1S%bGH1U>rW?zfh_w?I!5T@$_Ybnf%09OcLv_p9r6qtFro?FEZBXjWxz2q?8%9I%0VTbS^H6fXA)p8kOl!$uq-V6xu$dPWN zDD%|&kUpAbFJ!M~XlX#ygZ$a5uFw7TrC^$<^Vexq3H8GqnNuM5+%vz)L0EVQqtk?r z-1`_}d-1|gW?7E@G|KFF6U34h{uMzcdG)itlj)0^jGxM7$v$Y?PBg%j=VbOU-;{W> zPezOqQ?XXo7&LHj1<3!~KGqkHkTvOr=a+iGOwydwdSw-Kp5)L#2S7v6v6<7a_r=?B zNuq^T>|oiNo0AUdAX{~I7T9`h^jPPFb-HVNvR=~$CE9uzckrYitY^$73eg`N*JdPd z2Dj>MIQ-&5(Lhn^U+KD9RwB~{o(QYiR9xGuZ}NEO-7}A^=^Ij)q`mH8fB4qgSLP>Q zJo?()E%UbozCL&5`|}nzi&D1)7Z$zx+02*T+`AztIr#dg86!ZDpD^EUWhrS)L1L#S??Dmlf^Ra^<>Oq!^iDM zmA4e?0K>=;#f%}MOaZCFIEgD|bWI!`C-`D)PnnKmdfqaU{B~;hgXH@MNhbo*&;xs! zFgeF^Tfh6Z`o)eH9P$%h0H_y$wPsK_bU>{lk2M>*N7$B95&3PY3ye>9?dp?Z#%?y( z$WyKL7oOhSy|gn~JmMe>Tb*b?XO`zQr@ZN zburssKKOGyLmc0NGvq2}qabE4Q{D`QQ=}Na<&4)6)R&i*PzY-5?H1Fob9Q1b2T&d> z*fq67-3Gn-km)spIuq^)ty8c0Lm}>ozY|BfF4pR-Yq?H5&K%6KSWdbo$J8*gRNt74 z4$;*Q&-%&pvZ>~VLLJCfCi7M7p<0Os?tU~7VWN<#Tar=sKKbnkgZyT2vhZ8L`MHdm z$MXjkwR=x5s`n-cP4x-_4YaMI2bR6VFP4#LVLsM+1k>d`avIbOkWy35j!zL z3o-vgetYme;-f^}ymXC7QEsa51m24iso^SVVt{VzTlJ>rr4*q3eSm0-S(sxp2TzV3 zSjIwb6Z1XrUuL?In(h#aGz$fj4rJ;*WnS8A3KmA!z@unhD-!Bgv~N<)kSJWqePMi6 zwOGeDjlZtrBmJm8d{p56e~Z`Jp6L+}BQ?YO<8%utPtV7x2@lDUU4&+7SB+K6CP7)pB~2b1Amn*E#axg zI~o3%2LeqogQXPUs`#uSg!Z5z4>}Hec`TZ8!6$>tVwKQBAeFN@b3%QeL zyTvl&$@jgdd+Ar)B!1X~o*}0~dKXleHJ88o-1-5!s6(LK{`tsqi9wy-Yu(#*j5}qywbauUmquCdnHj zp}WuV;(uBB?bRGL#Ak5G_f$HqdL1okIGL@ahq*`^KxZg3!H5V@K}lp~FNzm|YT+j} zEfO2HGt}b*-C@kwVSwR=MjA1tW=^Bc*o??NiX7;;KG)#(S3x_nEg1P8&5Un_oQS&HJ zT+VFqwaV-FB~Uc<^u$|>tQ0pCD^21yCOVu7KrVh;UKzclu=@?m77K38n%Iz>u-W%M zXj;|e7ns4TtT4%GS;lRAfvcd?EsB=Gr!?}NZ=r$w$e$<< zqs2C@QR(yu&0BxY{Huvc_^w0;WfO|E{lt6yXEx){IojNRw`pY)_V=C0b5FxPlc#(g zP21o~zPWda()}hT*H=!P#y2rBS%!rT+|KCwaMOwC&;QGE8eZQpZAS$!-&rwTWC$<6 zel;&6uzxXJ4B2wR?Z11l@b**7Sr%?6hW6>P0O0M26qp_r125m* z0hXZQX$ux)+uXwct_FMd`spz=;q8A4uxFOVbD9b8GVKDaeNjMS5eVN;e*Euxxd8k0 zQ@d-;;~l`hhd=!9dS?LppE~cE&-kx8uid`(Pqc#%@CS41^h7Cv)+bDjhl~H;G&Q8H zR5_bDeqc}%pv4ZaM-0g~vYK78`p)(rLxRuY&HF$6y{_A&qmV=bXVCIMZN)2qrR$$m z4>XQ5!)xEs6q>QnweCo)?F()PUhcd)=Qt7^5;k7>@kK%`&l!}@0y4F=V4}`-z+u7d zGI?;uKc-i^&<^)KsbhB6Tey70CB4!=Sejs!*}M@#kvAiH+c&>RF#5UR(`4v+DGmn_ zp5$!C@|R$xefKES>@Yqg}C2khx$vuH(N1DAR`KpWzo0C1#ckOn&0`L9*BQIZ!Y%I8>YP@s49qtG819L)vL1V^V2m87Y`)Z217j0J< z_(o2oQ-gf}(7)l(-;AWLeR+blPCwgKTZf#+96vXBm^ON3Z0j=&{<4y7!;Z}Af2HJK zOWRF6opybKv3e&M>t1$Mf2W8bKUb`{*T+WJ?nDIBr`%mja}MQ~Zm~T1{2 z^54#+Ssb5W;ky$o{9si(D^IcROgrlf-SZhP1WA}wp&cv?156bNQ&oDT_ zS?^C`@D)TXbv9)L6#z{r`0O!DfFmW;u{%7Mm|yz2Wy;!f!p5>6|FgBsC)Cxa6Y9$3 z#-;ik&7iPmZ5T;iN?ueO3}mt3N`LstW@qlMi)rgEwq%C+&iGGGs?o?*A^nq^>P+*t zXDjovP#l*BN8zM|2=>Aqh2gnzh<3T^ICtgEUcZ8t1;M|am;M)a*uBKaRc@1l{Xr~3 z?1$}eXFNzF>00L4PY7RK$hY-ylw>;+hdzn1_~OWFVcFb&CFbAGg`A#HSX(Eh#kCoT z4m8g~-5(kq!}(F9DTGv17qYov#|C$NW>Zx2CELE-UG?t-39_Zzd}seBxD3B>DT|{) z$kBJ(X*u-Hc*yWK>$D8hWYLmPPc1bAgHpT86=#|Sl zvxj{Qx9IylRtIqtBTI(Mh9I}}n7WIcOO=i^F2isdV-tc-(eOdl&bAF4+tP*H>QhY) z#vLwLIk~8(4_X-$CrmnP7sFp#6R9HvjDqEs4aS0tU5cbj+27M6M`xlY)iGyw!$7GG$Nz$D*HN8 z73P*s5s3z#%s`kuylS8^%<9L0Vp6-QeshxfQV?S49{7yt{7O(mu5DM6mpjDB$1@1c z3PZ{GsN^t}}KY6)h3|tPta*z>q_ZW4G`=>=8CCwun7lmK4@j(v}e9DYRh^&}F zqekUY7a;(#IfJ}!7${)viYBU;QIrb!lcKcZacg&g{QB5cx?&ec-=yHPj3SQM7QJB; zKI2kkwMy4I(|GcK_4a1i<3KEXXM6MNn3{)cowc1r92n8o;mx{tIV3l~k5b#p(z3#w z=qe!C8Bc&jqKB!K;^#rFtYO~?82dFw*^9nOUu zS!m1p9v?L5Y+H}dd|2m>dQ3HU;xrCZgRZUQ&Q1ubYqpI_$~TW}9Q`zxdIiXMmQmbl zCXJZubG4Xs!#H~%u!c{8*?wM7N?)gyL(Z*$rvBu6MJ)KM0 zz`Gl@Etk4pX+@z8H#XYmGzlpB@MgZPZLDmVy5BB~L&LvC72}nMjYa}#(n#1{KQQ}< zY?ysxq{J&;q?Xe(6=R=zfkw#kz^C;O9&dinn@j!KY<8TC#L1=v`=)#8?i=3XZ_$X# zB16yxF(r%fOE`^J;j}?+{maLo*#t7apRnkzzXqJq$;>m0@|MLZ%Ox62catPpcZ)7z z=#Iw0wC#xU{$ne1*5uo|I;H>({ODM*?`4Dr{CqPwx?$$6HEm=iCs%h$Sp2ZHthT&0 z{3t%ioK=9&kEuPCxqP^B*?E;#a||CO6!>un%?WpHZFe*WRd=hXdmpQ)x*nq```*Ze zDce53pg9*tIe#_eFX|s=U(?^R9w!X1r-2OJlG*$g*L+b9x(m4OQukk%0#7!RVtuom zF!~q?hnPlqU~x7iCttByvX)L*z6Sez87NW-GwRPqWXl%(wl_5nF?m zE?bqeX(0Zm+_f?CO0zmeD0<9QXP+$#J;b^i!3B&Tt0LXh$>J}4W`CLDfxWcFwf|`!}4kiE8dh~*Qu}n2;&oJCcH}tBDaM58&A@76vEQ+3ay7^@; zb+_Zf3%2V9exjbwK@YtBQ?5G2p4FSnRL#luw$`-T(+%TMTKjC;XWY($@q(BHAFkR% zveWJE%Ic>GyQYRL@A?tn`xt<;<}{K)1!Bx{&GDtICfibq)}>SoTC(RN4yP{QkW&zL zjXpSefXgN<8#J5>{z3*(c1acHfqD(2zC`e&Xl=zc`bKL_lcEo!ijbhlT|z!i`IXq0 z(v}kGPigY10bgwnZKe;z{8=YoOJm&nkML zmtVV@H3-()K!~HaJ3hrRXvG;YQYhiRSScP1ElT)u%cWoX9xk&JaPkeuit@U@jt{E4 zE3;k7=?E`|6*)57z#ghO%_!s|9#8mpD=vGdJE=O1uEXdENVopjF#Nau%eK`2b*4SN z2kgh3lrTCg*w>oVsejryNXez@9#K0CKrTPqmKuO~Qt&|ru#`AuxkepUHObC9(~4?U zZEAR*>i}+3I#HcertI zPcD_!OYOMnY)gc5Pli=8@j<4rO0E(3k9sD6-)G_}iXIyB0ZpCY1G4ZnK!c%t^y=pj zSleL6dATolMb4aj+fOacc2-F!{hV?IWh2&5FrmALW=qf5sgSVI1`W-672TEY&TI=+ zJXIu;#A=*_4IB<(Hs@I1-elyo#Q3^LREP13(MG?c_)s_aO8i!?Ip^kSqbBw6cb+vi z5#O*VE3z_FRU^jIVH-X3ZmuR5S>qGQc4VVRcvx^-48;Cd&%d`xAhXe0x#;Zcm}0IM zSUiROsppyUB##N*lc*)NS_SM;k!08Ojwv;(C$8)591wij|hJ|D14Z3h@88y z-G=iwzefp(wL9@o`;`Xvp;8@3zu~28mbqq6`YI1eN8E2L&Ykt_36q1CRoX0jR5!fk zy8LCGv;AfN+z+nZlKKX`_;`W@YS^&r`nI>-3hE*|71$6WFw#iP|lmo`gscEuZyZpkgtLRocq-j$gi&$o|7$I z@N17GR}430WenT@^>Xl3zzTcq_f>?!`RrUh1@`Ag$R`M8_i<;`Jx^SAoRH*IN&TNj2__asQc58c^8UXn_Uzq~fH0dKC zTKum)GFSR&az9fer3aA08L|KCiYbvI9hrQ|lw{y^6b_k&AZ5x&9FI=Y+u8VlGqk>$ z;LyIB5pVSiW@88614brQ0MSnQ$dmWj6O?k~!Bm(fjR9s&fl{V?_-J+XE!5Z3TT6AlSHdLF%{7&hv9tCKyU4CQ!ON@pTr7lN>U-|VM7Gikn+1J7D`r* zPaY6zD%^&Djcn)%Y-q|yg1?>wWq!|8xIJEbs8m!02Q=j)XOB-}#q;YafUVzvkUk== zr+g%N!zAgqe=rs4r)ohSbm>xG9{gd7b&@yeCr|mcsa~@DA)F86hWP#Ko>Sp1__aw^ zuD?5l@CtNO&umtPf#BI!JeABSyD*8`z$um3Vlv3uH9#wC-uLa7|ww?zM z+7_*Z=Z@?UCcwAdKfFb+(1gc0to=*56kpVE<3^F|7x3Kc^a?qA`^)1c z7itX$;D=?t5h?_V{|Oax0Jg(E>9^pyw|yh};M-pw_uE7=^uP~0%LGFR6z{ZF&;Yhs z46CE?+=?vj6BQVI81JG~4bGw7fbKu)wbo0K&3S9>B z!~M`|BNU$t$;gAk7gzkM&%3BLVB$3ECRpg|$8J=Yd%gyL+44%u?NVq+uh;cQ!M zF2ME=Xs-X*s>1z=_&*W^PDc?T39<)xM(x8nOh3|1tiT{uMOLco%As(lD21KO6gLQE z{bm#bfnD##=HtfD2K6it{6WXL`>tD&)emK)26SZ7Qx&cDVYRR}Aym@C%&vXtBVgCc z{fV1O6Y)**9(qB*>8LE#YL)NHGrOq{N@v^IlOK-~;7_Pd8`+)3n|uShj&pC)?xdk9 zOAvSd;Bis3`hFaSsF*{GQ;w#3UF#X0!;bA}z(8IuxV-dw^pexfGr8tx(`GO0>|pO# zl4=dGuiLOM=kwA-m}A6s_1G-UYQt;^%xXS$1l>dJIP7e@4l+QW_q zwPq16ts+GbyXZaptGE;LN_ld!mFgO~breKBRQH1{g`jt>T}`_Az6D+MQs z@5X0T-7(*{&?73l*P$sUQ?|MB_3ysW>`8jKAJ(6R+PdR#?uo>02 zmD`}`OX}kCQHCRZ3>-=B^>fA{sN3iOrs!X`BIh-mq=)Ei^Q~NJpd+z$<=5yK<-o$B zgogfwT|e-i`lBV*m;~Kj*V+|Brj~?>H1+sNPw`k?h-KI&hakv+4yB=O?Q8DXmO8;7 znq`m+m0uN$&i3^TEYw-q{srNctzKQnt!Gk}2xVI8>j;pZ*VSHhu|&flkDmXj;hJ#t zahMg8+{KRX{l?JC96bTKSIbzg`NK-w*sy+_G@dUGo%$EmidZ}lIXJ)X?iR!9rDS)h4WTn(vMSXv8ZOOUkAKK9%n2D?`C zQyp-(}0d)`2l~BYxRI|w12fL86WZ~*bi5Ix9p^Z=3AOBFf6g_<2s+4>qLm!+J z0`*zZKjfGG99_2&pB@*Zw6Z5}vGiH{F5U}g=m?3(M+s^lW#QiD@6fcBdCK}z2Dxh9 z*3!k#m%~(dCqSi4gI}TA}P)wE{o4|FKyHzQ4P0cD>V&>)QWy1!PR9hWeIk}q*$><>({ZR zsLliGx`G0;2yaY0qrL<;-h}!N^#lrZDmtQjPd8ijyB|0k8$@lki8;2%wYjVNzNv>~ z_$NmE4nJ3E9JeY@;jFn~?_72&o0`8X-<%j!H;8tzm0}i#-Jt?Q6?e&Pp(rWhY0a{a z*!6>E2rIR?0PT|H_^W$+Ca|&ZGqc+3WW^kYC#pR^cX*T6PD7`v#<})1R7W_b<(Jk+ z&)kGRnId``iVGL#lv26}HDy%x-YCml;GehoUQop*?#=X+Bw#r@UmpwRL= zPGkQwfM(5m&y}hc_x(5hJ2nuQziM{mh6^)Z@o%TSf&kCRn zM&-VNtAkz|>AC)8Cz{_k3&o#yY~I%xsed4`KEZpil>M{#FDnjk)p;>Fyy{cANU>w7 zYP2!QfVPrSk)1bckt#h@LnGk-27>>x^aR&hoR%6+MOfD2pdwLM;8WC=BV_Aylnh$C zTFdh((|btv&>0aVlEy$;r{Np!se5^p;(SAbnoSCnh?%t|x+t7TuMk(%URCtaDV5?D%_QvI7u#wFM{{%1S?(=k% z-Bu-IPA<0x20HlaP6l6A>PvSiQRLCHu$)L%m4c@<2rntAlvcGuiP8IM_|^mumApqv zVG|^}fm1J;?B0l%?e8{=sCBQthmQ1Pk4kdq^GDVH5oK1tl!k9%y9iMM;PJfeHf6YTK=vZOp3mhhiG{G(=X`r0iEi*JY3<*mrSmEg5h<<^_vX5 z1fSG0|51Y-Qo|hXSbf2^X%d)T^+=NgTf2~E&0imANAK%hzzhh0iAUT)yP$sRuANYh zo*Bek36%8HNMkr>H8a&AG2pFJK>Nm%I}euRQ2Dre5}37n0Fd)oyrhru7eUgJ^e|Cr z9oT90L&MgIH$^|ClqyWSBv&Lf;%GznKY($^pN_7eHao{?_souTAl+nyH;>_KyBF{G z|1IjVH^Sfs2D1pr*hoG5KHX8&b{NYkqA?%M4YeiimB@vLv4XH4pi6HVLL2!_W_YpT z-ShzVqvI$)_CTW-OxT;#7JY=;yeg|NtH05^-n!yU(`ZhXt-i9p=I%x6GO=}`q zr7*K5zayft#kTY^l*g}WiVQrJOZ^5c`P`7mg;Aq5aC8lk*=Sv`QI!9EBrS<42-i`= z9~z0LJ+xe%LFfBv?-!nI8|vm`IpLyviSZI#xl5_yAg{j6<*cHEj*9Rw7-W&?N=1Iv zt)n5Tr$V|cSl6^*Ih5Qtsl?rN?1cfDbecTDD;?DVst9SKxB#T3ycGC9wt3_E<@bJd zw%&7-x595U`A}4@pqLQpwV?@+xGBmr?3}T~35MpJ0%G^V5-!s69Op^pxX#4SB;LLh zdZcutfNfwf^;;xDEQ*o77Jg(`DkwZU&|5g7yN@-Q1$k*N%CLrvdNpAHZ2oRLg#Md) z9q^8tR_MC0!#n-R`}9{k(ViO`boo`Id7mX$AofCPcHHCYQitwKWlJWSQ%nw4!jYHL z&G;M-sh?&L7e;cE3{>R>-Y=o3)Y(kFt8AO5lUY`KArdmJX@iK<`9r{I)&n=nd$oQ@ zPspWy_mjl(ig%Hn3qzpXbhUBFk%@`xc1*THrzqX( z#COqD(!DMTJ1!|JO+#2$Bli{v`t;$Flad`KCTE0*$J^iAaCIq8er@avqzC6#erZ#0 z0^V^(+x~CO?fM*lM0@*g2BVzaH13%l@8L(@_=$wLCXm^rMu;$Ygcbi*K7je zv3_+AX{7%M)RO!ydvWGz@OyPb~9Bsg*>rdgVp zB!b@2lm&QWOv}BcQG^K6F45h95S2paTU6DS6KbvbF=in!u+cQ^Y%<&Hr5VFMa$5rc z9nWfBo7I=wzieCm-_wGl^1Va!zuvR-+*mhrqxeYvKJIaSSX6~~h~=~Mi)#%nPu!`~ za+}LJWKT^7dn@y5h5$|8E!kuyi)RY@JSlF19OMY!neff~*LS$F^_3G%m$yCe+0jZW z%z?=4A|^yS8ukd$#e`s-&! zGM`pDLfGO^(gqR8*v>jo^-1l$ZdaFz`|>ARO<(iC=QPPDoA35zB~-Kw{$5xv04y#{ zYS>5gLlaG9CJ&TwSd`X8tgZ&ZkN&HwAJ3L)^rJj$7*q(<-Ss&W!WOdLQ$>V$1%T6+ zbc5SH8#$}Rqn)Vuy6PN~P3+AgN&KyOow2#K*K2S6+L+bn#zm&xHrnbNK_w?eI{ARP z{sl5q4A!~>qBTuDo}Uv&)49?$ZS>bK=~fN3k>{|I&Mos8*1mNH^pr{A7=8UP6Ayua8)L9-!++fPxW#TKu?^jd>;}c67c3Q|(L7 z-9KEP>A69p^(q#c`)cx%KGu|LM}Nk(MV51PV!Br$_9c@C*AW%X5hi`2Y@L;xisB*= zwMO@LI749TyyKGlAyKEiLw=)^_Bonf^b`@2mR|t(r#DeO^TwaCm<`jHKGd3+pmXLP z4W=3TW1Ld%kgKA2YdkFJb4}*CUx&eztPeBoF3WtN6YG1O3tA?{g^3A3Q!45E(bK&;xu8|L z%vTfRl25*~^c+dAY6A(q?_)4$^q*v*>T}fxOC^Q_+>c;ebI(2BTmAw=2c8o@ITP{_ z6s(Z>T>r7jzIITs>|&a?<4UO&xm=MkaOy3F&KKv)9z8kIYCNo49YkmKS%8K7m`nQ` z?gH53|2%5pzXZuOfTWLSx-{ovtTlxZ+5b)%ZbV!;_7TP*qSwAn3Rp1#tPB(FaA|1O zY?{^dL2?=(!4%Wn9JfeatOA(Pp9>for?6L+=8JAJ`2fh}eGtbUeicv>pl!oUBOvhr z$P7u}aZuywGzQjjtu(`6CS(}sW=r=U5TKJtBA7-nI~IZ@xN;cp@n(9`CrQT<&IVJ1 z)0+_NxCoi&mKBx#Tca?ZSW6Ym4|r<;;$k&>&P=#hNuV?|h};E1{W){UNi)fr$VzHG7^@n45k z2uOpJ#P!p7P&?^6I1HMD<9=yCEh{o5zo~mx=<20~qI#i_-I*6T+vMP@|G^-~=g!4A zfC2jGCI+PMejlaE(tE2V_Oeh=5hY(lz)GC{0Y>_WS(<_MTTH z$Y@tY#~Me29$mCi?mlTXz=XOoK)T{l`kT`9ST71QGW|^IlPzXT%dG{8=z_>OCI@f) z{y~;6x_M{Vyfu$AUz!^0g}Ny0MhDwg5mUnmWyHjI5%dSlfLW2hcWB`Y9(w{Ddzb`& zRJozL09iV=%mGc`P~qirG1AoJ;3w0!=VcHcT4xaz9k9FFBJc;glul zx_-VR{n4uk`-;i#kAMH{&*$|Ja+?4M+sS$0Y40@IKC=f+mVx|FOw>IvE`BK{rSqkT z05R^oaG^9e0LPqVa{b`%pZNWpKWiKW4uO-&L8ca;Wa6KVp)uKc1Y#%sil}wOJQ3ds zaDG_O=#p2*^bwHI=OX6;jbHfP!&+aI#$N{-<cTQEskg6tpfsFsu9uOZ9WVlva`J z!;g5`WX1QR2L}7BdaGA=)Hm0DoAywVFMPPe1TV1ev|_d?%oxwlZ?Q$-$SV@5oEq zLSt&lcELE^k5${#&d_DuVQHEtY-U$MrWa$b?bVjw?zz7@QVDXgo^5a@K9MIRC`&>Pz5oM6b`h?$YW z{$h1z1+-?yhZRT02bOBejSOzLPTSG4m+Z+`B>ufMW5;ufW#Ns)q057EVk zWvX`YB3nunJYFeQ+^TL>QbuNkUOzW&x_Lz|s0pbV`$sZE#E#4`7g$h+9<5b*3scgI zuzTwTgP6;T<#cgj&)O488D+b~x^1u7SzV%@#=}JDK9=y*c~;fRU9w;gOy_D{YIGb! z_;7BpyvjRFjUA{xuNdJOO0En+h_HWFYh)<;V&t6ND?#PabhR0vuV4I9p%+y4b&&1s z+9rnS6M9CKe^79k?r~;cK&VRP6jOxd+ZY0*YL(R$^JxxJaAI_(I zqRGJcTRVIm%PGK$+tjQS_IIq&D5+k*Uc<+DGf9^P7A|>fbWMS5^8VF}l+s2{<(KDv&K_KU2_2uI&U^Xwcc)D@?FZR3D?n>ErsECCalGZW-dR04VZ}tc zs1M(|oZwO2iz#U^Eg-*ET81--!=#T1>x8W?Dtn0?Vy7{Hg|%yIl#;RYyo^><#1;Wc zrkC=S=Tg`$J%wwueTs&e?Zl~Zp7-jVViOa?=%&v;**4W>TQ6ueA45!b9LZE05T+Wx zUbPoep>or(GGdXa+j}6EneVUE31Zk66#n#r@&L*(SBcZTkyfnp#T3}Yd;!s>{i5ch zVX6ZWl7azVMTx&A155ebAn`2cefYH4uALpfOnY%%=26JwM$z?t=Z(U15 zLGcaTLfg3V(U&;)r~Vf43-1>VmltnXpvQ=g&`cBWVLJVF!77I*v5_SIyZbl>x_$ zg~&UmB69CX0!4tRs=|g+Ex@RjD+ofWl1dA(My@cU3Dk7;;znMH0_TeptH0*qO41Y3 z^C|fXD&3mymVG+blo4(?z^__!kacp}?60>G`^#S&t3p_)Uf~o+K`chRgJ(y2S1fh+y-k)4X;q>9O zaWYR$V{&V~j1|kkjvr#h<`|fYZ+XeQyz~n?4d$Bv(5Vz3w7JwO-#WhLk?H-HvcH}- zx4909OBK}e5u#1n0z}&qq^5-$cQ{{iDz1<;N;2*aTYXb7Sevh>BW~u&fw$uFNxeE> zwh#02o6AdoU&FryfsvIfXRz4-kHFBts^gm|B!yq1G(O?wE$%Y~-sKFsS;f0Me1w=rv? zq%Z+BV(^16G_k&iH4tl}{B+u)S{ed0w-|mV5X2N| z0Q-Do*76Wr-B$}_U12Kqp%|f4n~Q2FW;XF`q~tpR3FmD%#RBJHO-xaJ35kcsva!}I zI5KvZTB4Ze$}1V(6>|sl+IT*c@nG9$OUe1xZ)vS#QjnjjA-}mdeY+;RzO8-;${p~n z5hjD*OdBmK#fdEaEHidF$zXJHg+XotcT9O*wMxJuM~ak42Dqy#b-oWXx>}&qxA7P} z_Fiy3iBY1VE1@*`eR^DLeI)3=0-h@(IqIqgo69&!n@QTEVazB)ECxG{l?Y;7N^Pp= zOXcK3m&85+p0Z0c^ZL>0GxCtF5E!yBF;N%AjcRNgtZLhC0Dpa771c77hJvosqb>dh6TkJlmdl5W}y>5@}Z4X zPLVtkHS$Xor>jOmg|jY2?t#;xwNq?f_OqEbTS^C+YVxJp0+Otp$Z{3gA>z*m1Uy3| zA@Cw(K=a-)3zwEeeh{h}M-5mD;Skdmge#$ZiXW+UN4dnM(&dXV&jXZlj>{eP{-+Kx z3$gc0RF*OQIiWNnyXu?y_gxG)KHVhy^fYn}qpRuxkb43VbDrZ_kl^cA)x zgQU!o*m-UfHCIEBEhFE0<0ZrdSmD z_CQ*&N__`JG$2|5bX6GHCn_Uq9;zv(=S6Zez7&l2L0mW=U15DUjISS2U!=oFn))fQxK&ViqDiRf$WJp`4DheVp$qD>Mz>JvC`FDc8G> zVt_9cJ>}o8M=saFy2}p5KJTkS%T;DKLX8DWC!JD&Yk}RxnE-r1$*O;>!o@ysw1XRQ zsM4+B#!AMEmsi686vLHpPeGtep#~Cs0w2H^NGUG|I2N1bK=1m9lV27j4Uv+6>Yfr( zCo}KqakBBpGvVtwB1#eN0QgHCp&kxJvb8~@FDh}uo}PY$q+tZ$wsJpNk|%nR=ZBAC z!@VQ6anxi-=qny|En0Tx9s>lbX%lFLY@GWbg6cr|c~nZB`|ZB$o2Q(txs$-j6b;n; zOz&AfA&;@-;H6jk;K8vc#Rxf;gOG<#I`qq9+Am{IR;xLJoC64bFfFfVRba6u{j=&$ z>Ji%gvvQ}{=TJWn9!TzpIJvwSXN=+yJO`o~qH!;(*`)5oO*_y)k~4`?LMKLC(j zTOS8OwscxyTXQtMF%orZR%PM=1b9gcKyeBS*br z+*`^anEh%N4SZ2OK_@n#A2~=cn{uhdMRJqy=lAuObCnl}MBK#V*s~6a8Es4IQ0Y zmK$J~RbnM&96a`QBbGIN20#5+)B=Pk9IB+jNnEhPgWd|hn$RV~Ob)*ZOqz9w{g;|?zQf%1^2A|S#$R)-W!czQa> zC(+MkS(?(7wqikEsSV_<)GM*Vp8+oVQ6=3;&Tuo1uBWGZL`grZuE30EQAv@%Ss?B9 z0%+oDuYf>mY^e^Wzx6D+Pf+Ui!Hc_#=ZXh_D=rnuW+7!117{2b$Gy83I_Ggq6!8)K z7$B`bHYQApsmFsKt3;@)+cUg|n-m@L@;zYbaaLBx-^F5>IGC9t+)93eiuY zHnTxgQusVjHD_)k7oi>j9_Q{+{oy@_|CKkBtD_@6@TEESeu3D&?5Mr!oACLSqn59g zx`sbl5pQVs>FXlf8*9y$S*&GU{5Isni{~H5CH61*aMiX?G<1JFxc1v6M{0|VKRsm? zlXvbLRYQSW;n~XuPEN;+!h`5&K@FL~m*Cy#9w18c=X#JP2hkJzjykp%jcI^C4o9lg z+e%k2JK9SqB39;|%?CxHA+9?Kqc5)X?Bznl!4_jYF`m%j6akDQ~W#>fYYJ-~` zEs?O8Hfmpxw+xvUwu)o=RZf16AHRww`6Y2=&_Frr4#;k|nCn=o%57=sHHo+3X9b_M zuVKwq(elh!#m$e?HFrDbvgiaGrR9zus>M153GwmqhmCskzpl3V{Alm}qUwdc>``p= z8LB9xYsaH+9CsII+YH=oGAdMBdOW;jVs~KSCRXkTpC9k7D5?PuXgTh{i3=@mSaN)} zWI|`*mrD0)eF?NH5s2EMpUNgvkkl-~oj?bJo0eGkK4SgRkim5F#@(X%e)g2RMH*sN zV19p~^ku;n_15jI?fsIe0NK;V`(_5p=sUwl)rsha(RjWjnce1YNvv)+*FG}y>fi!J zZx$$WXT+|2DJs%Qaa8HH0Do@hp0&F^u|qcE-5ZkhS*Y z&qW!Iowqp0M{*r5EGatRG@J3j;Yyh8l8=kdy7sz*2Y46Voind-99Aldb_&m2_@$DI z2Vq@gVfNPAGW20oo>5Z&1C(yR79^dNfybSGi0kw`APr3o<%Bw8h3)dZodR;@~{I#i|A###N3?EeFKDLHN_j(#UE$Ep&~fSo|&_VpeY|1!8UM9I;# z*p&26wAssvBV{w4e%)k5XPH`}VZ#^08yG4= zQVS7+=UyIo)Ok z3z|m)nm4CB{l=cwdpUk2)>l=qOEV{RnRQo6%(RARRdbXTb|`6Ob9A5@YU^-hM2mu` z0`}uU%K`Y+IM<@_W>bPqoJK#FchqK$N96aYWhZ|rO7iNp+z*LrM`8b(QB=Y|s?{Ip z;Xlx=0-lg{wn%z3DBbWvQNAO6fhkH#PR`VDd|J6j@OLu;m5O&1)qky^$gzGS=UZB5 zYf=~~Fn_MV#8jEICnSZ^;RQRa%186xgsF7mtI*d+6T>90*w226pOf!WuG-(djH~gP zwqd;W~dTF$s z$)F+8C@Q3$T!9`-TGfn>QHwq`d;^gahqz%lR>*wC(&?97GwC>H37^XGIIy-`Cgua) z$3}kl*YF-Y@@+kew2pk;1KD+2rxKWbV8!I}bwM1_k0ocG=d8WFzx*z*T> zbGjdQ5&+l3Pt@de7uaF5>2Aw`#^e{)?+NJ;GheCbtiW>Vo~rqV4 ziGNp*i5<0cR>#wQ7x<1!V~f7C%yXLdt;Dxam0mVvU|SpzzY=(FY*w)(%W1fpqzCHP z;*8FRJj`jB9;!n#8)&dvP!C# zAyMd8!!LG4=$@EQWDhYdrR3d|sP1|_;=TPLQ-b#Tjw6-AG@9gb!(K}36PUZIcsZm( z%2mfh@`J3FW)##KI(b;EfnY7P$kaY4bHhKt_#ff^KR9j(p6=f6R~~grSbAr|wbx?- zO_feog?k9FyoI3YS4Gei_&Tp!wu4AIXJ43J3!t7)p;c9_WQkvxKwb?Ki#m0;U{}PI zddcgc@)bJ$k#8T zSP|$!^~ie5`c4JwGln*D&MO#S46w#>VgI6q{U1S{+Gc^R-^I&d56!Me@XI3Sr(2E? z?%s+Y!5*I()r-Zi-A>}-5M#Rvj#U~gHeIJVfWyyLejYqHqqO_>CnuiZK`^bi9zTAUOtgnxHXn!}KaKAB&~R=i@MIP2z7z5N@~C{loTt=cg$1y`42Qry?b&pdSdz9vRWeaGDNABpff*Or?h4aAF*CZCR@$1> zL;5zCi{<1Cj}5|!kGCeBVCXf8lr`mbZlXyoGkLHkE-~hie)yf?s|G1vQg3E33VrWD zfnAW5*O}9@*i+uRyq7J?qs@q`^19H1V8m)8EuLnoda~*jpBb8vvTK{D7t0V%w2(C5 z>xCCvOf{-0cCNMHD1rApcON>eow&6q@hR{h)dNwp(Mt=;F4x zm5>ImBu1Cyj_2fbyWF`EhP1%!;=zo*DBmjMsB5w;&32Fj&J-x)H%A>}G6-nuO8FdC zgREL(iJ?8SP#0|;%U>nrvWN5<$Ru6fo0d*q`(NOYI(-Q_Q}xMJ8x6Y3bgez8q}`Lm zX9aj}yAI4!6`LMpMf5`X0~jRh+M#z0&%?7kP914foxk3qsVPtGIqhSqPET%n+z=yv zl^ad9iIIslbA+Sd$cXVnjYcO>M-(#>2{dxR$m9oVK!&%^I`(nPB z1eQ3okJzbd>A}d{=6R-@0;eb_+ZJRs*}RKbxqdhX2I8AULz!m9u9B<`GP|T{c)jtO z{*}^j-A&6^g>fXgs7sAhGs7Ic*9_pPh*3$9|hH zI@U0a)gwO6jz1;u*dRa*8GW+5$=N3? zPA4V&<3?dUh>epfb{%$4p53BJ&NFBHI!?~%Haxz!IfZ$C^S1sPjU1EX zA-S+F^OV&bZ6@7y#E@|KY(8z}HX52=3RNGY0H2*knN%wYYT*vYAOCn%U;WYB$W3vMH_6~x>cO5V_n zNRL=mw+GNtFvacSdje(K3}`w2IiJ;NDDRqzuioADq|;Be=!UP#RM&&))Ah3B`e?u5 z4U~>2SJnsS<#eiL8+J^haQdDzf}9H4abS?ZohgrgVA|g*|HKFXY$yBwglI=NT{^ zq`l+d^BH+3udH&MZ<1LYO|8&KK-4_H zWlM#t!OgugY;5XI${J2gu)cVS z(9UL^G*ov}-1aeQckWsKOcqTtXo+ zN>A4Zh2OG-BeFdpQ@Y96;vf!r4nky<;FzI~;(SD&SZ%Pk|rm0`=h$-E$87 zJz;ACG4A@<@x@Mx^*;@#JG1)08|F2dw6Cm2t!$+&Op9)yB!(? zu<$&Of~|Yc8zkRHf6s(RScm5Nv6j%z?|=v1Ex-ZGdfFhVwX?0NNzmus=%jd5O4mf) z06)OZa_~eyaY)pSsM?Ufg0AjxI+|x$Ts&n(^BMcHN$TAf4#EPt36Xfb8|PTFk<&x4 z0qGl2XH-AXw8+z_1gnM{bQmu~^)V<8IHbZTx62(dB9JZvp#&tZss(XfSBQ(-^`yhH zQh=b%fK(v%fU+L2HR6JKkwlWrM1aX9j(`5JK6~2Xzn#`OuivFH`}gK_=GTj#_4f9a zc3&|}S~)9WX>-~V&CLFbwzLX}wrEuT9N523tIqx5(~oy;M@b<^e*ShJ7pF+*eo`NiekohDerMdsn-ByVne9A^S4U zA7BB#b*wy7QgJKM3<@pQg|R;}qFW_BNI?=nPHrzH$3gu%+yE3rQMsAy z*y5#bD{oHL+I6XvL4YW&07O+uu4zvKan+j=iRZgBAys^$P{n9D$RyTpke7D^-T&^i zyV+N~B%{NiL`8z}c1Oie?k30FEnp~o*M8#URKR8}Dox(zt9(Rt#F%&WriF2E`*vkV zRTi2L{jeNsORXq7cDm?@)a)MK0g4BHP1^C`Akb^Vl z*3)I+!Z`ipbd^>`7E~>9Va`U)|2FJTX}=962HIOj6+dD%a!!DP)#7FbfXpJqFe$xKX0h5l4 z-VQeFSTB9|uh$sskF&_X=D>|kDvvs@fB$I6I4;FYc}UPeH8II?0d)y2>yOFiCt^A; zRU}+4UgX%&&yE@n>!cFON@O3^(mbr<4|ry8^ElW|#&?yK{Wy-x>0YnhB`D@ScW!L< zrMld^u^gc$B&fzgg~c_(UrnB0^a%rmy#j>oyiF&04do5xU8I)bb5es_VQ#<=3M_`Q z>9tPNM&*eyfr_IsW)H8LR_ebF&o5{JhsQ-f75{7s!e1=}71B%_?B1}+f}hk=N8N$B zxgjH*znf|9^jp53C2owG*tJze$nx;ISTKca*?I$bU2o?4s5|x*t+x`xLUcFTtP0~! zrCM8xBGl2Yc-cq`%)0&nJaC%@BWPu;sHw$Elag3x5*orRXwlTLA>H*aAE(4w)@$-6 zQ<2u5tD+7vH-_jaCpE*m>^zDVTVLloKq>&>3~c&gMFCcsG8Oub{S%_M5u~y>Lh&w5 z5!1Kwqa?nmrd=%1-~4P3uO$UT)*32EAt-oZ5KKP*+)W<>knM=?{*yZv-GMz1r;qT% zIlFd<1i9>;%hu{ie$i$fN|bGBFDpwPPlNQ(RyamP)^941aa+qmS_=9ENfPS}*}dHK zxSXVbu)77IXrM^?&rR?>p|W#@)1p*(^+7ITo3M5+eXSJjj7ndcxXEl=dkx5kVK*xq zv_QMf%<0bfvP)?S}_Qfw?KKQru8<+!nSw%Jvw%LkrNG-{`xtj zAi*t8@!iQxC3wIc(jL=aF2Ic^<=2!tD$ja>pLzWGH1{ebC2lgh=_#wS#Swb*u zDj(g-+L?nWGBhx_xx~hPvE2tK)TWI34}kBFBeQodwd4e)`?wo*S7Vdna*)o>hh{oE z{cKjUB#%v`AnhF$hPq;u_zZL6`5YypI3{v<12!)%muS*(tWj9A@e7y4#kQ^ov#_%i zYIS+Razf7MYe%BXOJ21l)JspJ!T@W5FZeK|i=0>oULYFr_xNV z5>M#xtSXoD?!{>+6OMhEr?3)Ly)`x5Mi#61<;>^}UmCCU5x|{B;UF?(tXVOVb z@5bySw$85AGuCXc21YM8Ce~alz$)yBgQJ^@)5ExIjb$rm#ZPZg%IUjDUnP0vWyYE9 zR}@s=ZPOV(w{7vVaW{UFHFAUp1nH}#wT8qNuvu}cHUqHK^Z*ff18~}OyJpDUMfebS zff*iK4eSf4Vbx8>=V9v*bnzvb)+Hy$)X>GPk9(BtC5xiQMIL2C&VcO-69cj3<8)s}S+ryL5hddl41I8NgHNval5SCyzxSJ$K@q;O_L20BKAg8Yx}nEos=f z1=7$XuWr+h%&nRgwX~!Yd);yk$B^DCD;6r|e?f;u$&Y1|`Pr}|s@nB)p0p0occ3ge zz@6_8rueSko+7+vBKdMUT)sg#9PqJ%EG@qH1+cnUcU0=+%JAB*>NoB`jrZgc(4fn5 z!nbl;Vi;MY{T`4hJl(@v-M)BvMZoek&4XP})N4(cz?J;K`sg@42F}deAVne7g zS7{c_ny@c6Q04yKr?`gen-9FXf+5X|46K9FJC{^$tq&oYk#VKT<2_3&(C$fhy0;MA zArzUUk+xf4>B$c)LZ)!-`%12lQ^}TtD~Y%}-5i29Xjh80MZJ1Dx)lUIAakM!d}PeM zFpxf2hh`Z(7{W6|<#8YgH|g*qX+KT2^7oPO!I+A^YLy=n-^Lm_>jygOV|kr=#5cx}Flg9Q!= z{3g!3M?seQdZR2xvl9qckpwh*Rh4i`yXR26%tW8qcYVAXAij2&+#=&Nh{N)8>E5_I zotlKh?Qia}n|^5@+3onefAPcaN6$d6U_TNpkk- zQR2i0dq#9|y`WRVZT2R#MWD>2y%Ho4e1osY*Aq7ewhl|k`Zo8A>a~f760--tY~pYu z+6j>%Y(`s|>R*Qu*&wyE2hd|@KpA6!tWt4Qa!vf4Y+H^Q=!SIQABA#l+j2gJycGHz z2Ilr@H-4E9^?Un_@)xCDXXE#ZuCY&q6>xzSEwR63;A7-uJHx4>9XWP=)GeZzpt^TE z>@O#08z2YNLrN=zpaBlMOIa5xj(9`G5kziWR+1kVPB?R6uEF3wqmzrYQHhPRHW~by zXceDra;Ew~yGpF^KHDh`%2GijAzDI9ejR6gMC-k6K}M_-CbejdAv!G)IPFeQy=A0% z7@d!5xi!(e=WS2d|Gs)DXLj2oq1>z9!M`>0_xCJxy+Z0INE3!OM2TKNId^h1e*=(^ z)LB3F>jG&KPa!U2ATGNYb@82fE0l1mx-$LE2tcjb88E^nDqEi&mlW=Z;+2cFjkvF)Pt zIBFlEh6`FVHM@GQ5P;-lW+u43z{=bzAfSNjC{SZ_#hDBGje-i|6voD_e*U&dkecH| z?K|=Om62-(s4&&-QwzF5KkJm+Oo*WPMHMLfV|7o!Cf+>PZUy-UsT=bFjh?=q9xDne zFMoAPxMSIRz4?a${-A_|>(JTqcTpnSop`0iIjbSzfB9Q9CJUZ#8`D$ZU;TD(5kj#LfbrLwO)O;07)EkuxnDy{4 z2#<@(CWo%V4yx|%7qte=J#E-qrYMwe=>!GlN_lFkfb|NLBI@;l%DTmvi&G^-4?rnO zU=@^3KVcIh!{&{;fJ7tK0(7K5uQSZdWZm68und{mcOtp@5ZU+yOf=7APk}g$uFbQi zb^6C+(cfVvzV7Z4c&9;@?T+#Xbwu^fC)yq}jemi6I2P_vodw7R-3rO(S>?hFZH`BKL;vg;033XO@S zMunTO!l=)ku5sVqm4Y{+q`M&J5;LgoKy_^qk zq|ART%n=?50@k(}5WK2TdOu`{ODlcEvrw;x6ZdZnu?_C%aRc>yYe2T=^WlT88KBS= zV}*(WA3>~(azhobmCN#kuR)AN&gcX=Y&)(kjRY7_ex;;A@MCe3nMbO}htLg|#Vi=O z)Y1Y_{IW?MdaTdzx#=og8&t9o>`0xAiDic;l*fLrUj=ThXokGN_M3_>4#IOjxB-r{ z!VZaV_5SX->V_vLNJ+9PO&4w`$b!3}`OwxdY2|Ws zv(bfH@xkW}5`wRmzmVovn{>zVp|V~fT23c;=HC&d5scHoJ+eNNPJl--F{p%rbZ%ge?` zu*l3Mi`~2HbchV@wxc6OoO0u(tckW(#%9 zcHR)((}Osli?lK`Ag?{yKP(>7YA_(B5g8wV+Yk4i=D;(yc1^eQ&u4BIy?My)`l?YV z^~#^9i2V+gu+}TgvhOzV%KYJ1mk45T1_%pcb`q^gLDNy>Y=k}tYRZv(kW-IvWol`T z9*LY5$ajTm^CXgM&}r@7XWH(IdnchJmo}C2JSL18$#hM1w(#6k^EHn!%|&G3YgZS* z)zze;l$H@qN=n^6-nABW3sDYv>iWX3*3gnMhz&SOLfuZ@a0}NOqcOaF|4C_s0ol#! zWAf;zN#nGbcpt7u+;0>Tq`sLIDeP+1IjAaU0c}}Z7+xxj)qv(0qsBGI*8*Q2 zFY#k$L~VllLt>dxQFkUse84rbb+O~vM}FVU%px?LtvojiCtJv)a6K&%rdsuf*hW@< zj`|15#}+yF{m2n?N*}7X#Q;NwQOXY6<_@Tjo`G&s#u0Xh!zpgT-ys z6v>#85B>Aoz!@F;9doF=4?h>^iI^5*)rB(wzTjRTxBbbzZ|nuutd;JC^L6-kio8m) z>%!8-t|Y#k08~8h+BKT_Rf;@^mfjj47VVus1Oi~PE^*(Pko@X)2hZB^b`X&~%)ZlZ zZDEVd%3dp#SGR;|N{w&Rckp(k=8W#Zd=oQqWP}^6PK%d($8;p_{#*qi-c4hJ3vkG> zQ&10Yh2-2ii38Kh=<{3i0s_iR*HSKy*h4xc(|vSuu`|GIZ~#~4)X&5k4bOwL)_z#m zXd$mM1GE%x0ThEfa5njKps)>Y`!%ZB=ahHtJ8~$OD%Rj8*O~H$B$ME%j_-Q51w?Xk z0{Yi8Ne5(3*!t=DDM_*8xmR6Zt}Va|Iw?Z+VJ;&T|G^#?a6QgWZT2)2=l^dmVgB&>`H9ytEX&4KIK<-fH{kZU3Z$AOqYRML0PSy~mx-Q(`` z^i+65>6BW!rfr{;e_UK6NR|+xs*F8z*T|F2mO-lO*fY$6JgjyDIWMPBLCB(Bd{Uy` zo~$KW3}uyHM1Us!U|;9#HL=)Z`Awkt8?m~^1LYIgUaRUV@;R6aU$m{*^THX+7TC>x_Fl% zHk_rwn79$RhML)aBH_67igIc@3jM$(WDtuRG$2;=6|Aspt0(~p+|3new&vMzK?@xD zy6;%7JW11_{hBuLI=!?!NcmGFT$A+4wRk~@&<7d|5K9ZZ(7o7w0EY44!Y)*_4iJp; zfC#Y8%vGsw$jtTIS71jBr%W{|3{K}TxG=Y>CJ=R|)N)Jx*7k}-wG6tq`rZ`0dXz(7 zU}-~?JOo_owYk5)8y7slv3DzA!10CWpiA*i7$N8&lLequPU=7%Y|+dl5CvQG?Ocp8 z1dld=x(|}+U5|o1J6T;Qnt$^XbhzeVEmc_t8uE&;m)qijz4P?IQAef9(Ab6ms&g!4 z1XSR00%r5NuMsx-y8OpGa<2-+E*PsV(}%h&pwawUAaGQ}-rLs-<=1#LF z$n1#<%FB}nw-PDg@dv=+J6D_T-Dp&=t5&aPhrVUO(CO!l9@;D|yA+ZSzY$oviF@Z0 z)y1ydD~*9h2yUYG?Cr{83;2+_(Yh#G=HApeYs~&an^;tUpxVeUv}S?RQH z&j9UrV=Z!UmjP{`1iK<}X9g22jwE0N1~zm4OjOh%s1{&O2mnVElsyPAi*{*==1z5Y zQaN*|Dhr0^+8PH`c%|QDBW(tp%O&SfU!0s;bR+tl{#l*Fi4oZj-iT{g)mlC)1tcjG z-8ZNU41TUFhIRZXamek#TiRNvsYIP;7V>63wZ{=HbtutCzRRWHn6FDg&t;$n_8J`u z*tVq-2xt&C&v6?BZ6+Q2W*N#|ok2?$UsNj`sNvR{ls1(C0;mcu7j0fW{A9E5r&K3+ z1gf~glQ(m4udu_)ZS27ruYS-yTAV0lZqrB5x2V7ag z4Zm^#7a!wp|6lk*FwX-D%l_zE@C8dS`5{dMZj#gKSD)tlJeJlHLdW4_h~6BDkaF0%^Cpc@_!X~}86pMMWM(-=P@ z^=CF$ELVyH*R1A>5K)WhJzOfwChPa`ZnQ9%{yZvh3Jw|wyw54zohv!r4}ZF$l9Or! z>V3OJ^IECFZ9xVg{7urNwDcJ74@ri{$9SlrK)IK(zD9!>2JXACtj)}nm)~3Tn;_ai zPZ(-Ot7=RTH{q@!DctxOgP&uand@CYD6@R-r)g8|#G9d6!{Hi~&I(YvbinHNRwE6x zw7NJtsqnU*Jdp0UOkHMSwheT>bl=)+g{ar%717nYtSbc>-(xlBSj=A*Jw>K;!4cHX z3$3TtybiWCo{6|^6cDmPz9Gt&xl99}Uf8N+-;`7C#(Q3qe>#w^&kIX6u$F>zhqS*% z*(xZD4qbF{v#!_Tm9&T})`}A;((=ei4-=WM$q-!4xz@LvEZg!5V+4-U@TuH}*|m5% zCDqwV6!EQi0wfyGEH)y!IC^*)6k4i_Bh1jJm|7Z$+`=W=-pimZ_K8Y*oKUnf##OD}%%XgFASgsbWjaflw3Poff+WvcMoO)|weMJhGXYskfu+ z6)I!}V-rp)KO+4iUV7!KiiO%bEQ3OlZRxGw#@i^|rnKwkV%xhD)M_~cdxIM^1^&_6 z3S9hbpRdLys)83|pU~Ph$?IMRA28OHi3pnN`ScH+zA5QcPKb$RT5FKEM1?ceQR{h1 z(|Yn%HT!uDElV{JYq7C6O+HuJzvw%^3y^9=1F@mTjX9txNS-k$mAd?j6=XMCXya+-LqHcf}|+#e5oIXTP89^|zr* z@606qKkRxtx$a+N*ZhC9>tEKce;asYr<3PD`1kN8p2Fp@m5_&CjE+8lD(_tUe|*;e zAI$_V|2Hv{MI9~j?-q0|KF;SEhRVoOB*VGHI5@fPs$xE0;|v;4690n1L6clxxHX&o zZDi}eHX@XFK=>pdjvm!4=!1?UYwXBe8c`+mN=z%rjY%)jrTL9 zGhyvXPoWPu93~5BMD+>5_t1n=|GsHS#n}I3PT^9cXx zP?fSYFjA%EK(iiCIN(MxbL70cd9R?dVIxaEI+WxEll_8BS9u1d-EZH(a0wCc=iF3i zhzol1MPX6R9}JcKr$d052cQwaC1LKGcrwJ-UeC9a`M?WL_S@!$$dVv3ha%u!bBWd_ z2#Cnt&$zL3Q*k2GV_+WO$>1&G<9|5t_4gBd|87#MyB3&}ni`(s&*WEE=f@WHzZ*3Q z-g$T+L6Zd!0-{6)n^pbasM43zRxoSUJ8xu@P{Jui;)~|qji~`I>_H9BshFIDUKrD} zrhWODf7BgPIN*!>AIyLKr*nuy^PnljeMV)e6vj~W+}*RVw_}yT3-<(SrQ}8kNGi?w zs7yvKX_%D{gA{H3qN>ip2oubtzx$U&=|_Z2IUCGlTOjlhne`9veRm33`ae+F+9o~T zllzb~MZcFOqHhjv-}_YW?8dwArcN)6r2Yf- z{rz;JWiJ<0*2`TJu1dXg=GP+f+i}Q%xsPTbC`uD}ekp!}Hn~-ILjgs%X3+U0@z0px z>jPhX-0#+I_HHomg?|N#UMjs+*kpfZoLBG>@$uPr6QzE zxSv|IPV9;^M5WtIF8$t)Un%~U+j?ed%^P!v;+z4OE89OTVINP}hsen4}@%j{Vba>?7t)c?`| za@Ds8Ya4P!^2VJ|hS{ET8k$%RUaF)t`RcJIOa5Xg$=gnP4FQAQQLuHZ)Dy{-6*AV0 z&c!c%Xl1JT4j}(EL0)(($V4DW)~4_)$GyJ%FT1>_W*gJwWmZ*tbnXXktW}gfG~xQ! z(HV5^S#A$EAruV;r0bPVytU+ijjwOf)duj@-7&B@4+}nP1>op8PH72-NP4(YMtJ3rs zycIcy+gMQ;(;D0X>5XEHtc8fwIV81ZU-V2J*lP8{jcV4r;rrj8@qa&^{txDkeZSx$ zuS2FW;KnfPb$R<~h{lr4HP&53bog&YjJhM;gd7IJ0rnB55nG)of zetc?~{lNF{IP~{J`F{_yzn|j2Y=PA0!<&S|?l+rK&mR80G%0I>u`-skN@sp}pF`X7 z0LX9c{*}>w9G%NB8wo*N9r)6!Xu|Evt1yRA|6ZES!v_YPT8fY&rvQ<+nAV;wP#WM;OHFCKx(C z;)@%{MAL)DO{s!0FulJXg+6c*4ygvv?buhkE#qo#D}PDY9zGua76>7jrvf&!SOIT;_lpj zJCl(g5N(0VVXMPqM~4oO-z-r{AJGWBtxh3%sYdyLFD6LD*OX)>G-qd({}=I+dAP6W zzUTVpaOgV%&X`hBSLzrnW=5P5UzavBg4_pD{BS)%Ih+2nA$6o&N9X5G9Z{cFrplm# z;lRh->Td`5gegxOP#gQ|tayf0g&B&FVvO{`SAr?{T?lw0xp-8DEpU4#s;;P^DQ zW5$c5?IhRrmJ^sm*@6&7q~xSmmI?!J{~No_bWBF-h#XQp$}t}+hkp&LgWj#JD^GFL#55xO9BrdQX z+-yapLaV;F0RerDU1_h-5=9Y~_3r z_wpBOHPLTcU50Y_T%YtZ18Z;O^vp>VGHCoj8Yvjbcerq|BCDK(hya*Ls7yC>+~`Yx-`cx< zhv7p;Gpc}o`aITG22{IAzPEi{6>5Rc`emu*E5K340JD85ko1L%B9$c^(jAnejwk~t z)-a@4>r*kXV;p{DEp?Zhh~Dx;M*Y_f!qeXiDcDNBVIQD@>PpcOKTIKJd+t z6fZByN(JOM$#vtNL}&SW&M&3ZHQks4vOZm-a1g~ekhNCz#{kni-!gs0%YzXy%uXQX z`N`rW|JZyCm67iuSwqiq?5s8Zz9;VcuE?C-SLy;Q_F+siBcO+q@Brr*bxGm;<{ZeU)XGMPeiK-8|PqVcEts46d82Co;9JQhD9O zbqpPNygJu#WPV%k!@8Gf*v{tTPP(x9`lPbGKgEisJAd)D1%hMtQ#*v2V@Gwh`a1TX z?$rKicHb4Sw{`_Cxmvkv2FC$CUG-K^Rr4N7g$xgmbHtr?vbtu9Z2U!!FAVD^_a*|t zwam|;lQ1}Yj;SoTOPoJgcJC|bxKpwzhar?SQ#|J%$>>Warqg>5!fmCIjw3H{!7KE_ zPIp!UAiiIcHRsi-oih6jKyoijflb=`sH{y!(^U|Zn3(YqV0*GnuZT0 z%vOk|v4@RCcuUCc7;8snE-pktUZJsNsoW`RcTI%eUVY4V|NU=m%zWlYO_jm11qiQ~ z;!HR?WFxdJLjG;kDoEVF9wJAK3s&|H4NtrlXh`$Fs@5ur-;kri-BMi4%SA^cFE}2C z>-0wYc;=e7Z+-1wS2(5~uSdwf@0(FvKAoS9{j#s)Y*V&tetkW9LV6KME^G1x3<9fd zMwKNi6>g2W2l&$c4I6}6CHLQI()yEg-x4;POeZhL+^t<%ZD1!IM&PBdO!N1lr!V0h zJ)Z-&nmTJswvKdSuM?S?S7pA)!6-lDT86qk{!G|OF_fj$cNvy6AEmvR%LEJ8TSgOo=|DB`(X4a*-2WC+MaS z6&P`I-wcqm{M?n8DzM4Cx0}p(Sv~b;+|D?J`GeM46=s`c$|HY|_=zsQg1?#WId>JV z1Dk_vva*M}L`3?AtTit9xHtb7aH<^_sMUP2v2?OFuFIED-U8nTdQ}W%x2UsUq4CI= zku^wIQ)p0ATJ zUxXdo+1CwV0`AG>=c(5Q5PQ)5dH6lzOTiD%&6N=F`#{D; zDq~+}$UA6B*l|}vrfgij({s)NYl0S<=kEfJ;mR#JyiQK8lfy}JGD!}vmy_$|aI&0C zW-PsqzT|Ji-YX(m1sRXs#va=f8X`+I>KB%lBXi}${>h{Hk}eh;GnX73{fu415Q+&Z zF-Xsj)N`kXK;Kig5GSTF9wwnHp~(DAag*|12_-WR&tEBb5t^t|%4t>7P{ zn=G7+sgv@oVi#+EFBMCPY&na33JI`1o!W}6u-GkU*= zo0<3J&W1e%ctV+l)Myw$3H}#3#l7)BHhj`&Htgx_vHL$cJ?W|qQ@H1keOFOi&~wR2 zqsWS0$_;#Li5B$c8#mYBr*GJiKs*LVz1rA}KzMXOF)!^1d);ZuQ|^|~%^`kevL9de z&VwDW8#pQOJxxN7+4=U+P>F~1A##dgF-x`JY5=3 zm-3E9E!=Je4o6s{Q(Fu;b!9y8OwA zoHq}T70Bo2A>+ZYg#!%MF4cArkw=&=gG>E2V&zhv?b$C8_k z1kV+waSs43N70JlBg@lCxG9Q~A5t}+#eN!g2cG6?2XPvt2t%l|zn+xjFphmn{Y`XH z>Z-LTKm5ZH7jq>gc60u!vF7Lx6zeX0N-Wd*u)Hbw?vFprE)^4OPnt}A(schPC8ZZQ zQ_T+Cg5oMEt%-=uUz<#scH8;ON$De5$(jw613!e6Y*kX4^P+sh_cDc!IN*@;qX+)& zzdzK@88D^bVx)b|D7^)7XN@8-PTJvs>!X{y$+xamoqpeIf9&Lc`dcxOnCtTQGG*`-5JjM9cY4W3 zlgs9id5E4r=1!}jqzE3|B+#1q;2>BdPd$8htuYJ6JdI7x-zn^*4 zEkILTFgDxL-p{<HMJoA3zevoMQiRTr@vDQQHS3D=-!QaLK;{89?J$4elwTpxA-Jbh$ zhwZ6zU6S51@7Q~> zCuBDWKVk*oaXa6u0wtphqV?YKeqwh3OkusOK3=i*y*z%g_ARDdey?#4-=6vPWNs1= z?RW2`;*>-50;a5g{5}Ou$QcHhVhccE`Kk9Hs1nu2Ox@sw{xbhpZd>fvLy{O$@P z7n_>Tt=h*V2P!Y7%uSWN)@6NeYb-n#0^gvski;AV&zpjpfR`rD%fk!TQ zeWl{<_@2$s;%hxvuz359LLnfpBQOcZwwtBGb zj2Xn-9htvDOY_`6P%J%sG)SmeH5#grm*T@tN6ZTUBijFncA5WwtoA=v`yW954<T8EVl1ffT`!y{zCr=c}C`D&cJ9DXzOps87&jTZ=}r-{QO293?wE#Q8o_%7{kCuItk9`_O0Sef~sipcJj)%HZl zuTJSPM0@G2=)rp9sPi`m`X#+VX#QpTT$3>}_&|g{no2?vs_9`5_|xlH9ln{?I>!1r zc|&P3t_VBXgB{8l;WJhvsPHB#%5Ag!MP7f*7~t>_{k4}hr1O?oHv40+$D@Ced5#e@ z8vdC_^j!6MErn`FhE|h=TJ*j#w~=YMZ*V;;U~Uf089J^%<6F6l?^}7eQY!P~ACEXY za)yw$89tO}BGRL`28$7dLVWm5_Ge%(_pJXMzfye5#dSFyI!2T2wP~Ai6F|9p0)xT# zW558n#K=N@B!Tn2JhHdhJgk?RZ>FC#M-1{GVce8PG&YxyKS+_CkKm~xo|AVCuwNr% z&!M2y?!!NK>^Pa%;fti-84!)(#)l%ukmob=rFrzerZ5fpL;jR+a9f1lB;!>8I*hfw z!hg_P|Enn7ep``#MDSs8j^U6cFw*Bf7fW}#|H*Noip@4 z-%Me13%@|Qxt1C;+3YOX>6L2Bp=y4RpR{yGuT($`AjvEgW#uN@%+>b9sU&&lD$9dPXVus%Cy&`FOY6I zcxqs(^1oIi2?iFrJL7J)f_brbuo8u z3JDo2ODY18P>|~iz~J|W%zsqN#lJUrTjthbD;c4T^Gf>L#D z{-33?J)=d?B{ge{h>1c)E|sxXC(FS48h%E%Ohp$_x9>tTzlTOI@IhuGq7f(oPZ7)d zyrCKYY(v3IaYf$sAT(BMd_WupEs%8Caw)92T93Msvb|6D&kszh3ixBZ5dSm8IEkzq*$nE-(<9LVA{|iJvNI5Id%Ar$DTR5*19!VzCXm6K z@979c^HWLyC%;amS2Qr9-?~`xuL@!Wez~&vADKnhX1rzCA38ws@&**>1z2BWGf|XR zpjni#g)r0pnmnJhFnxmE+&-t4qsZc)p@<2Q|FM6B5?O^Kb5C@;q>TThyzg=W#dCpy zK(y8}$2T%XX7vsfaEmJsz`{y}W-4xO@bY|O#^#WGD)j%fvZi$m)>a8fbj6W81HK2@ zDka7ia8x^xjTo??8^Rrd+w~jfa_H~Yf?ns)Njh$OH-%4l19W{P?AMm!96Gq)E2w`& z0RI{Yb>VN`<4JPI$ij*u)Er0~P?W~Nw*otFbU2x|wegpv@9XBDt9l370H0l%$wt21 zJo*V^E{Bx}qWCZJ>wiS>99iJdU%4JaGb29cZC;OPW=f4>?-O&0U()S@%*+eP0eR^Y zmML%O5$3m4hYB)cu9b`oO|}5(up=xpP!wvAuNVPsY+2;x<@{x0Mw^BCRnWT zCg!*SMZanBcfg!|KzTFoCFz_66ab+{tg@sX&pCE0x&@$a*o`&8{rx$nsjzO6uDc9|-|^Bh zdOn9d-`$7;Wh;@^Xm%on8s!naS@CY+L~@Gz{uIp5O{$-y7PbEe8ZD4T!`Dlw(7+?X zb1ampFg{Oi);2IYI73V;{$XveJ}ORLWtO{qix|L{AifmjF(SIJFc(N>P0DMX1xDU@w8_S1f7s~QX!zFLs}LW3UA%*CqP({) zP2(_~u{CN~&~0^y+LuSB`FNFy_wFASxE(*c%*b`+#|$NX2I4xa9>Lp@biUxth&~U~ zzqSS_X^`k1xUQ7k7G7lzjg1-Tk@RKyjFYJ&C`@S(a ztxW_q$WNq+SluF}PX(Jc9{MBW8^rJpK8RAo#WPz+6xh4hD_&Y1rYa&tk&Vb@8BRzH zWVFQ{lmRA1H!q}=H(+R77A0&_>~EPK_6dar#CV`0IX>@}+!f|GX>)4)-0l;`rLEWJ zPWCHm?%rH$eE9(!z7$rPZ1j{83Qh%%_+0jpuBb%=Ea-of4Yi#E-w=lo(Sc6T6L9c< zC;pzlJUaotK@a({-xlhjCul~LHw1%kfPUxZ#_kQV!&l=%jAB+{IR!ct&!swwRj=q1 z@BIi0gjWnFAZn=+Y(APY?_oTE2r|<8qym%#4RCHd9!Xc^%a5y0Vwef!UXeHpL|{I+ z?Jq!m=%H=0=Qo(%TW2=AJ$iY-k~n6>4PNA$^P1C~F&q#<{0=r-(@2#3lF>ubZLOrt z4Rd+p(P7IObt6uS204byadjwjcXjp}8zh+AqPC1_)Iwg%e16@?UWwNI^Oy0|oxCdk zT#%WN8z`Kr;4Fscw$s5T+Vy<;*Qf@F81bi`h&fWvdk@xlb7iDpd~?OH^)Y36jyOxw z;b{U);pi{ehTY;czr4>IsNseIJ_3pddz6|p0&X0Ha|^&nOa});V{EWd`hb$5fXcL4 zT$5Qu^(KVn?{nu5g9vXwEHy_;VI5Xj7;}y3nfGz=!T4c{Mr;*NoLQ2tqmFJ&jqLSd zH8aTz=AjFET<9yUGcN=eIMkF$T5PB9(^>iHg! zYJ1-GV)jK`Y-|Yzp$>0sX_8;3h+JPDT418**pdNqj z&#}A5F}T6qUS-v5q?;`4uA2{`R#G;`x1z#6v>45l%e~%Ju8;1uOW_zm1pJQh4MuqM z*4%Q<$XxBq>*EiH*ms8%i;NxdMAMDU*>{Uv02Qrar5)lm8?pI+HwpgueQtAI<_>E>xN zBQnAM)y$aH{(L`VHq5B?Q1W6%)MS$H#nP%I-RGPjB-5urex=1>i-4#bxe%1a`aA>o zzqfovzAirIjie83{KEy*_}BCA4UHQ0BbMdJX*3;jKaD3B@f)zf66&f%B~{%rCMWK1 zAU0T3b5Rs`dqo8`?Mmw`Y{O;igl`b5i9S0pX+@qd!u+JTWd?#dlhQ9|EZt;>8lh8A zrGUKs%1QHCc)Dd%r>{6bodt|is%4^wlmEUnafr%s0n|=$Zr?mn=6}?a%9kHE(bM^MEEbo7j3soS&DMa zZD$r1Y>`7eSUEj;-Q&@{EV+I`4o!fvqzEKY4Bbt}vU`s1hioouZDWi=Ie za{;#&Djxpy4V_OeEk@j=9jZVsaPh=}KYXp_UHu~&5SU1itPEVcq~P}u+* z?uKu~isT0bIo)B&!Jp=W5_>maf!AdqndS>m`{TtP@@JY_+OvV@eG(&oMt2Ht{AN%0 zXV1I#bqIs~FjFz+EC}~KsK7QLALvsYMRPWLzv7k$ut2&O!y|IrjpxTsuM(nqy@RL> zo}}kyx95(CMiArm72}9`dGLOdg%S3>RmD`%WJ?0D&ZMSn$2qI|$d3fy{|##K_g_~w zlPo`rEU|RScfB>VV`0o!3>NiKv)PKIki-2zhKHjB!{%X`xe*C#YkSDLxe>GB8*{uq z4|hXA4gJGra)qeU-u0T1G7l=KQ6gssRa|3UEQX&FZRjqp1p!NyuY2_c%{6+XKwsAe zh$13bV8+@Kh0~k)qF_r^0r>dy%^{<`VWqm5lSZvD)Ce#)OMBKUd9`)MUSS^^qc8Kv zC>V<{bn)E4)WC@Qc;47r&1oV#;p)#3m|V8;b4~?-zuq}ga?A0a`T0fgpv;*qmfi*m z&un@Yu%M4$!L(*9yYF#lV-R@bo}MK1#igW#LsIuStR}_YIfegNl+&A(_-?|}t!_#h zO~jZRm@Gve5%(5e)TmF{28|W#BgLt2u1l`+*a`xC;@RTjg~QZ7#QfOF)#reF?guhc zf^1K7+Y4A^!BWQ>#frlwz3VLvyXD8_fb37A-fK26;;}X6_^4LV`wq_G6ddr(RYkn` zcUk=Vq}D2Z^s+zswJudlur$KT>1{%yK8>YJ2R(d!g81mOBH2pzWfkY3c(#EeliaPP zcfZ=vIP4CN{=4QA$+89pJ+p3$+=T?{)`ypAoCSBXeJQI~n39J&LwT3WLg~-4)#hzd zbqj>XOQfhC?}ev~fYprfv8ICvF6#p!kpz7Ct?P}*)gsQqg5rFP7YKkOJ~GCC>Er(+ zk#?Ml5~ND;#CncGHR6Y{hR(PHmA@(pn0PX>v?t=Mm{EcI<)n%_YG56l;?zb!UAEq= zEVMN&@!aIL6t-5`o0RaC6tH7Sm?&6u^eGX~Ymw{o9Gtuh91$35RKMZp`M|(1rVqvy zH4S7d1`U)MDDg8r-B(L{H=gpUe8z8xL$(}~Sp&Q%U>SGh@S5#i#8@6beslF)1xnL_ zy#Poxr6Xd?H`{{`EGGroxPmYEO2JAqGoT=ZH-X&oILa3QwnFIbwP;jzmxOb!wUz0{pM$lug8FL*O55)&6J|W z+3!q}esDYbxUlfnvG31+G&yT>@Q^Iw?D<-KmLVm_=9_~uU$>!KS1=WgyE~n){SKVp z%}s;eh3OF}-x`K=(AN|=|9cdR|NgzTGZAsMGb^>?y{U<-S_N~Mb}L?lfKBvaP?Q_T z>QvlEj4~D_{fBR#$PVs{J5@$5b6v3>RbdXN_gc%wn3QPEXUp7a*d7m6({t;af7DGoaF&=jO`Y;HRCiQ#)DQ{j64e#EU?i-G&vt-Gi9M7e1O=$nGLRqwC_V zx1neU_O6VDp@EIw?cekY?JT3he-CvC%NPxI$qK$Bd~fMvb$nsAlfHelKvB2hqd}UP zM*}OYxK5@)D!*@)UA4JngV-u94x2OMRp+X;zLprSeH*}9h%_XWL@24xSQWumwB261 zAC@c_pph1Q{6hUD#y+LLnbU4^*_tOe4|fyXGIi!v(u`ZOqY_N7NJ{?qLVfiW8mkta zx*sgXyByvi;a$~-8V;#P(QcPc+EAP%;>jbXyp@yNoNkR5nxdC6kh24~y7p6%%LeB1ZDcWy zDeD(d1?G0svj$IUUcgbaRMul+NR?5m+txg6w9$Bou|XCIvjwO+5xU-uxrujN|E=}B zi?RC0|Dd&>m+LGvE1Omccb3!5T#d`a4{5z5>{D8~Sv?@TeRfHa!ClQHU3*;bh&r`L zmJt#^fj(6dtfV=SrH%Od@#@kJ&s){z{BoDU39PA)l^;w$A*PVY_Uo|P9X7LiELbIY zx>cDyI>6Zo5(eQS$nNnX!*h)&sOmHG#3TUUFZT$1P=6m1l(cK^ERvb-3I=1D{EsL!l^RaVf7 z(|?bcMvkhxClM&ac0`-Cdq%CU!qwen=B1Mi4fQJt=EH60`}fW(rO;e*dOB7D7N10A zH=iX?`6_=8njb8b9T>G0S34iQNh+ts^$s|j-AP?SqxM^hgVR9&cq{onoD2901#_^>~(d zYhu4(6RQB0O5hRI!&nA3Co%zW_2#!5j8XXnf1zyl?L;>y5;a7>@~z zS#?3A@z8AKEs;@n4^eAe*DeZq#O3bfE^&YKzdiyZI5FTbV(}dIZjC2>9cyAt-#FLz zk&S~!{9&8_r$a%bsjdAdgj#MT%>bXWpaa)(+}hl6sUo-djN7PDU4^UD>lV{#=&2j> z**An|h8;>w6#H8R_>x1=o8_xqSiO`wgNiBIT-QxA*%RuY(0vKrdSU6VC`U~z%C-P= zkw{;{MHg#!9~Nx12sY$#IV+3mcm-G0mFRThp5m%g)*(;flE(QmXGd=Q8QGkmcJ~!g z`2L!vrs3#W?339nLdGD1o~!@i{Ni4-(RccknOtc5Qaw>bh_KXzozgG4GYgS4QlGcE z)Aio>TZ(M1{+v8qQIWuSuhX(i-B)jAWo|aHw&DleP)2umSE~Q$H(GGS@SA!aLKTrx ztN;0go&Q{~DN4XZs%LxSD+yn}iLLu;%2pg>|A5qy{t`nS?~a=uCH^Hk>iO^!@(zcg z%W?RKgS`-Ro(p~`G_~I0IO*eY_4jbZRU|YaLh!B2uqV@_L(*-;dSBB6M8SZayPEnW z?q(M*3|Z4W^|I=&H|#+gWZ1mqt~pfg`v#10DH{35>L9KSv+Y z96P{47OUx9AwFv=kxm*k%FE|BfDo(O=9%6PL5woo>`^bbAa&*baH{UK|5y+=Igq)c ztv<;2dzk$3jw$Z%F!a^Yoh~L`uvn~fzX-eCA|GQ%7e+7PoQ|-(#YIB(#pIte< zXOz9eqBQ^Y@`S#nInv(4jPTBOf3im49P_-E%THN%F@vqvNMU&EU`&q$Jz(+L{BY9P z^@KIYE1Dj1Fzo&!#!0983#cd8m$OvNaw^x$TgN9zC*REf&y?!F2h*~m_q*y9FU$TX z*f_4GTJtk$L-?fC?v_F(OlhY(gLVTp>0Ib}5v751eWtS0T?4vS*}AG~lOsZ>DeF`v z(aKm!#~{R{uP-vSyL(%(82wyjQ6j zIe*&aW=5P`&NO{{NnNlZG>(u)eB}-E)jSNlu0+@>*{mx|X4p+NtQAh$wrub9b*DaN z-XG<|;I`cof)3rviCyaj0Z)n{he=xSCEAKM&G(c`R~a#i+B&{JELOWrH=u3h-`C|| zf1jnM%5+=r(s>Otsn}LzttP7i9rjBI?pmH;spKKGmNYiWM$Z`K}1x_MN66?bW*LptVz zirXO0_!04UZx<%*CZ# z?^d3_#7YVL5_*28fX~*}on^y(TZ>bA>0G;qss_$_9Q3@UPXs+1eswTUXIzO=n4SGA z;s7H34qiI1AZ63J_0VVn?ueEy98-YpP4axM$E#Ml;hlW7UwWsXEF=L_$@sX)t06M9 zCC>L3dsy(F?E3;Q+Ke0ER23`8ABKkiF+A9atuAqTDDy~7^}uZ93FmoLObqk}zoEu7 zU^QG$7)c)FM>zLQea#Y`_oH|ANzeI-p(n5J7WW6bPUf-nyp^%lG}5aH-1SlQ8VJ1% z`+_^Q?v7tioy0Z!9!m;LcRp>)G&;wxK@XSB*S+xS*OnJtEfiYQxkum%eO%rkqv=8c z)v0!>R1n10f`{iro=r#Em|3WOzp~(gWEU;7r82a}VWSNNb3gckTIw}^4>@P)!+W%7 zysR2lAM5)mD|r*5cUidh4QsrLIgO*Oj2e`?w3JN1_2T$g!1U%tyf;cW5*TgnA>o$<#k<(#7DBQTU6Cc#nZY5J+a$4$fz76OiLigH1>+Ao$nI1zYru- z!cVtLKzXCU-ahECY<&P*elY!K7%|Sfk@V;BB+rW)Bf=!Lfm)Tyk@JlNZ{aA>O4{&| zO#8^*zXE=~B1Yg}GJkLxR!S@P|M22mL%4)cL0QuHkW+vDx}OJpxu*H(u_U8oD<3tn z+tebBeBN^}OAz01g9gURg=0m%hJU7%PLVDMXM+v7X9SIEI(5u!AUh-x4~$3^tx5gV zg*RiBDHsBC!LJp+xfy}3Du+D#txQ)cqY^N-_Qq2{7*XsyPI-1^r5BG^->-Mr)i)db z$7BjlC#fYlY%j$n;2>@Qx7w&u*wijP-PAV9q-&6BlGn`|XIPs8a_{?vRFh0Az0b{v zOhvblP?TAHbK>-c$UH(?{Vkju)ey43rbp-$(c|Ud@-$#11X>&cf z<%xjF@0$14zXh_^GID`t?c8AIXGUQ^ zHeDF6i{AJUd!O#AXEypNI4u@6B}|Mg)%a3PKE|pje4&OlgIQsT->u%x)){-h4c_9A zNV>Vwz`I?qt8K5HIr-wO_wT6mlP+(SswCxJJ$j`R7F!d&twMFU7qWQpl7vz5e3^xs zo>OCyk|Y&NjLx`gtOH%MVu{@RCJf1WN1>M)n1x5GxN>3G^bo;$uq;Kh`;{e}D@jU1nz8A1|!@*(%cN zZr2k9zMvee%r%3@(t9AjUFM_Pp0)t%_>GZ8|v zyHKfHj`boa)uN=gnEvA?A1YYD^v~OnTh7-`F_n{g(^QR2M>)!dy>`B4+}_N*#M1)6--xT2j78_RGx~1;XvA*ZjcYx{U#zJ=NC>vmn+?b z+?7b?LpS4_gCd@lCM%&4F86d`GzN9q@B~9Ed#IzE75I0cILlUs zFKa;C{gL|Y(921k?|AR3$FJS5-B|re>qp^e=xmPB}sYt zKJMGl0=pcDY#CD>sZj%*mT9G~~V*RcP|f??Gej4FJXz+Ewi78cAjx#kj8 z(@ipm3pLRqZ*(*zaH}lBgl~h?cjf0~A93HQR*OMRQ5&*VaR*jDx}gJ@>niAt3qTeC zE~{vR|Lpx`9sR(@2LrbkER zQTyY}_Y9q#>JObm8Jx&6J3+4 z6y%_XQnjMZ`u?=CPlE*XQ|-|PuKXy+Mj}@TdGek%t-*V$yWF&mocE;H?hUPRI+$T8 z?RI?SqYBy3bo9=XODX(b*cE^M7?1#p6$~NBe;|o!J$^p1Yy}isS40 za0&USf+4;zt(8AZZivMi_t6S^66B9KFmA>FeUHCFr>K!@R7|rWMkd&`Z=i)cmiE%~ zt6%s!V4hqFdLfOUcI^A8-7SGDRZ8-nUJYiAI`mFXb4s=B{PdJ|B5Fn<%|SKR_0 znJ}!RT}#fmy9ePtuOE<~phx`dEOr_0`M@e~vHrMSyDj!o(69-JPI(!)w_PFBzk?;SHuvbPcST45;AOYM(}VX}LTasykb$TUWhLZ72JY zRGPl33be+q0H!v4!@%*3iZpxWQNWs`UcS*24#!H*mPcPa2?I`y0iH9ZC$C;=DLZ!^ z*5NYBIp}_o*bp327Q16gnEzz|LI<)mnXVxi7rrc#%u($g9RB;Z{9fD>+E5wT_wg|H zZb7OR`O1>7+jlx1;p}5(Rz|21CKRyL^!F{dMFU%J=2^*!Y{jwznp{jbjyMKJfu7l{r+W1vfgdE0Pd=nZ{VzOP=6@qv|o zn5L#V;n+&JYxZSVl_1*P)QajuveMYjXWo~jl1~C}TV5B_Z&2hhbtY~59*6`1t7>4{ zcTFmrULB|Jnf)^1-OQp$FwULX5LD&UIG`lrAcFXT+njxM|8*JQ-zgbs6|;(1c0YoB zWq-GH5%O|Ues2yoZb>^$8)hZ@V{qxNW;IP{HTRLA4 zxCBV_`CU%liTj%#;1?$Zt!&0v5qI(K?Z&A~g z)(7cq%L)B8uiPW%ahLD#o2(Zp?$TtKSweEED%rXB!*##-CHpd+7BU)U^V!s-io}^d;Z^F{OsLhtf=UZ;Wp#yyvAvq9)Y9K5*Oc5qA?Xft5VBZ5=((Fc6Q zf*3#1(O1Gbyi@&6@X`(z2`byo=sF2M=~(;mmon}7AhPoiDqgKGB(|Zrv+LKU63F8B zhA9}o{6|{3mE5qpWNzJx0SZO=k51gbb_t!M;o^>Y?rQ`uOyesQ#6>@uItXfQ(mO?T2a{u}Y)|2iL>1ND`cJxb^YWmVzi4Lk; zT5;Nyl48`Mkjc(H84Is`Nk0X4+;mA?#X;|2EF9Od11jd0F7;Ke+g|7CoIId@(&~78 zmUDQ1awdCvUp|~xZIpi;ezR)ej=fFaQ~oFCi*_X`!JY`i9LwUjB<4uk~EJSrv8D9)<<_+DAsxtMCoiFSwlM z+U{5KR!+QX>bEG}2fzg?j#^AAq;jNC9oin&t)^FEm)cW;E(Y>NKj3Hf5B^qZ8s6I) zXWpKOw3V4X)6>5cW#5RJy`muQjf8H`X=43-qkb11R$sMzw_qu2rvqQEgwSV86mkp< z6JBIooO|H7&VhG#bjhX4v_WM>ZrE#o!VzOv>tLLBO zD~i;d8v>kO5dIn(%r)BJSdN;Ai5lhmtwkR1$D zC%D1b3ve{`KDnh0KT&fjrBR2oofv4nAhYNC`bSrPcoHg=WLcM-pooLu9((w-CijkD zW2N8(MlP0kPv;zsm!$!4DnrGP#l2-T)B zwafQP`xgN#=u`zcqh5Cof|#IVzjM#jBHLrDU6&Febh(Jd;em1u*Y1wc8ItQ(56pV0 z66d|eTW@Aza0(T~>+W{$7*gx%rohdZp7;>Bp)-{V4&J0|Rx|1HNP{2Ko!1US+s#80 zyZRPSse=t|%dx(+{}{Z#mAnwgt2WKx`$b0f?3g;^HfR`BcqCN+Vu&mM0j88V8TN$) z5EyG-e}DW`&2<&1O1#zXTs8Pj1nS&x)d$p->tcO1<4`i`BZJ-2d+{yBQvugCM%&Ie z#P*JyCuJXbPdo_Mf?+?pu|#JZ3p!z5MOR8MN5^{fxv@GS>3hf$zbu_6WFj{VefqB5 z_I&1`T#@)FxrEhq*JCI&e)$-vyR+-#hqBf3>r6u@jFG|%EQJ?qf5-l_2cbjN>Cofx z7HVtF_-ESKI4KKIl)LDv549qpOXZWuO`%Pas$*hFO)qqJXWMyg;ZtGON9d+dQ9w3+ z!w5>nPqgz8G8*4zEc^hQLMuz^4lQbQ?JZ;xIq#2k`{Pcad3sQsEj@VsqGnx?bfHtE zs9<1evm&3WR+v+^o;Im6N=mUpfB)h?=C1yQxJMQCv#aCOEYegr{v5ddTkYG8=5W=W zM8<|`-_%8DI}Hp|sN7evuGXS^?#ryx>CzUiJV;r-AX@Q3s6u6~bvot2lvTqOG#F2P1tL1q3K!8@}4 z55QDHgW|E+$mRyn+%A@OI|yC-bJ9!x2-xpN0Wmro%buRyh`r)(QV{Q((RWt*UsbWBnrbVGn{L!=h-*J^dB$F=-}tfON&0=Y z6R&BPcgHx>UVgysSgBD9h_>x7ZhA&IUNX`fszc!JnhlwI;Pc63$*fO7s&8ovo4ly% zzW?M-#l8?@CVqm3&_~d5`hUm7ua2Ie+|Eu6y3$VY&LBR@aBS3ds?P+(x^L|VXkq+v zaHc0N8d>qS6!)`W?N?~~4(oVVHMLwqlXZdOar@)?xZuiP9Hd*%Ul3!W)}LWAg(D7c z+PI`hNo^=rRiX<*3RfI1J&}%mv31$U58_Jc{+am z_lrB*IZ~(vZ4XhpFB)(8HP>M#QRf;9J&vTSJ_Y^2UpD!Fgso_yEDDq~WrrRLKFTY}i=dr_ zR-s*XjNwI*3|vyy^HnwC)A=5aZ;zV&9dgb_bOK5PEmZgAg}v~V@()Z^{V|{>yR%Rp zvQvE!lT(~sCS(!T0rU7t?4d{Pe{@&vRc*z)4;7rjHpS`Ta1%`{+H(VWMWnpG?xaeg zm!x7kP>p2Z>?f$;#UlowGw?_?kgjvJ2ti-HOU!cC8iJ%>5Cc!WaOK0`ml&=WAUSp&mLGY#>?HWWKK6kVXz6z*!s;2S zIpq{mlpLFS$zK+tDXS9-RMOIl0LL5Lde>{|*fO&Bs?n3|8?_((OX}9Y{i1Xo!#lEU zux~A)?PqXK*zG9G-7!|qciOFCuV=g`Uuv&3wL|W=KIz?Z?VDORfUZeo57g&G?f)gK z9!5%aK7G~iS?@C0xIDs%_E&{opz_S{T9fCqPCcVA-KwMt@&!S9Ga4G5Jiv)kO43Iv z)cZ%#;KCl#))#B%t8VF^p|YFeCOPzW1U9Y_xVPo?raNA*#c2opk?J{UsFYwkgbwV( zxL@mgIn|1J=zQ~T_UQn$CMj6SL$Epn3+HzMe_;waehK@rAE)R)_bNP0yPm7I(I6)nFbZiK^t3J9RayID5cws3$MY}op+`BLg)NZ#ff_%~I}UFD|5^Djo-h690b!33Q0DQ&O_@E-P)Toh3d`@%N^+MZ-1 zo%eW4^gEAF#!rL-p98Y9Cu8Vi(Jw`N?*OE~KwR3@CoLJGQfOG1UCdw-WE~mvNuV3+ z)NWo3L?|#UH@#xDqVhYk_JZD)O2e83f?4qCkQJgd5L4z6^>pCQAJ`vVXg7RIDI0=0n zc1hw}c?Jp|y6bMgHam^1_FF}uI)L_@UowV2q!EA64*dA)vuXSGG_;Kh572PV^qeT} z3xvIRU>uMzICk(1(5s&uim=u8*q72pUdoakGwcElv6n-6Qh+r_8VrOQ478~%6#kA~YEo_tC#7%lXMwntH5u@7bhYv-Yyb}zsBs)6{*$z7eL>aEsrf{P;(L|NK~5ENO!2r=Xe$X>2oj2^XX-b>#kbTZf$^=up1r8=|9+vr z8IgXXS*PecbVlkz-`PWUE-aidlVE;o%Po5r)3-msJ0TdXAe_j9IE@twMG*B%mW8^@ z!)3UTx#O1tb(b~!q!B6&k4={}T^@nM!S}L84UQwxJ<^3pt@)u4gJmDUq@48Hq%6&& zb3G?e$gv|}=4^D*93ojR4!L>pVIKeWd6WsV%q#mc>oRr{4Y}X@NGoy}pX$2z72!5A zqf?e&`@+k+ZM@>?gy&G)~C^>&j5uGD$3W?{hcK?`8 zdV7S7{*B1}RsLJ`sjzafivZk8K$1B~*7ewHT`wK1y&Z+0MrUiM0-#0aSFeviptm*ghQouQk}hoNs-_FK$XjNF z-w}6tLLb2Fx8W-AZia33Ix&A)pZ2zC)EE2&?C-Y ztz_t(oqD?nuO;ZfQ`5XiJDX%75+0aDK^j2w8t5U2Nz7uIU7f6!S7tOm58N_=I zDUBJ7C`r0!szzJP&|j1o`aLS6lLAr8=frI(Xa&;> zG()Zu80gz>G^a>;uK(Q3HC;a=6##(U%bzws)L`mJs9WQ!kjUz& zt2=~GzeK@qr2*FhQZbh+lj1c|j5U)7tNje!C#&`H+Y3WTO5}$MKd&&;hMPE*8&Ox? zxhaA%8N9;6yRi0K-eThux*IbLq3^sT5mU}UU?FrXGTN-~?YWjX*zD2~tElPZv8FWtz#)z%tEa!p)pbTM?eeZ*%9E6n zU>kJdYpEMJCHjZRytxOyOQB!_g!xOiq_~$NiTxLUKKAYhI6$BhjwoSNiuC<_EYD~s z2D~E*co;jrOf^PsX!ZX*uLcjt;ps%keQy+?9+-YwJef6~A%?(uTKrUv3{-mj39aiP zB-}Y=t8U;pEvtr7ITL28cmP2sU6OEKj%4noE6*UK92(VU*(nyX8wTno6+jTL9IR-% z@IyCYqm;9JVRQ$(uQls~fa9(gMg3K<=gnsOZU6p+acbDy|{IwD{HS;bi1sWd`2RS ziWrkVWb|207vjhgE)05m79wFgr19}0m~G_jQq%I-P7KhKWAxWhh{@}(8x-^Di^2zB z7G1HRgu_1z1{U9a$)0`H4LSne{8!!+Fl|hq!3pSEBDAQcPeI{rSd@!5nDesftIfj? z9>jJCWa&kb8#%c90_TD>s3emc`x{X1x8fgXX75az9ceq?VpN2e2J`hVrGGr}W>)h; zVJAcq8khjm(&+jCk=28H>eW^V<{5H`x=rIm$_iZlw$@so4hVl|IWCcpBz8 z;g`u>D%-$_>H)<@=r3^ghT~GMwKL)5D82G);Fk;$*isL%?8~>4dB`iytA8L#Pl^2vf;2XEXh(yJr3vK z`da?B+hc~8X_Tv6NM!o#jIgcyd;vP2f`}3@Ytb-O6bU7^gPNqX5 zPr7TUfEN8c$s}MAr+`7Ap4qP@dQQlbHy;e#J}mBHG|EN(A?BHW>4zH@)H9sJQ{4L3 zOTdUh+nGq4!2YX8oo?p3W#;|MOG|$Mbta`njva8?aGg)NUsB{J^vUSCvm`c%zQlaL z$^cbZDG8@d&4ff+&z1YaPupVO%7vL~HZpf>IMv@#5Td*H;C!@F{0I%iAn>>digJDb zD_FUlZa5MQz5|Za6lp`V+HK@9>QTgar`_F--R7RTdhOnHc|fU{T!*_~i>yGt+bmy^W_Ma5~UJM`#-bQWb5wg^&RklnN9sFfg?F zv;GE8vI*;@p7asycP?aB^^Z`s@k8$S32JMF1BLqH=av|6ZCaG;v?A4RTP-{P*eyX9 z`{D8FSo|{u7DrpqqGr}lvzrcU&Xe+Pd+9Ltg&Cf&RQpGs=Hmy`AW5Q`A8sjI?H)VP z@HV>hxITrIaqsg?=0_8BlH3vl+P*Hpxq|k#KlW8vpswm+@44U8t^L=6p?4A*dbbb~ z^sa$&y8p*!BNQ0X*@oR#P|Ik``b4-#dWZ+@ZWH!UaAeBrMVm&%s2w#ua_pM8WALTg zUETFOV?{~Q4u}*Vx6>JS)9}(8j+)M&dvzK225`Ed0qxxP_7FOcDrWkA6Cr}Nz@wig_Z!0}{S^%fn=3H+3ieQp-rXBo|CMCm-^V+| z&t%{yfDWH0qaCqt&q43J)^>m=kdPF$pT$LAfm#e65;*}~mV_4VLbG(W`4+qLk2-mA zW4FP1KQ-HVcm2NiLruxKQJRcjf$^W4&3}bDCV?(kf!quE?19SkiA)!m@Pu%*()S)-;aRbkr_ zZ&`HkCXR;Q39hSuds2 z`)(HE7hw%sMsQ)5w6N*I-!JT|E1v-BW8-8jWxD}NT?i@Yd`E;D>X~7$Sv4{#hi(b{ zfJ>XM+I+!XxW8Kq+GcDi_Ff!z6`r1H0c7?E(iUEAcyOR?X*n{f5gjm!IOnijZWQyE z=Y)7N{J47gjacG~;l1xZ&-G3|8%42)jAr|4)=O5IEx7*Bu7$&CjwgIf>4bTSh8oLUJpZodP2Uw*pynYBWl@ zfkS#&e=EGT9HvI@2MsDhmS0BT)-k7>_XpbTxGIH7z(tndZZ4izEL)5hd+kokd*OJfmuhS`&<2Y>;Aw{Lg z5mMPoS;mq!$zI96CHpe8p|Un)4Pg+G2^+u&bgoaKKJW9 z&+EDW`JLZCr}HY&e7@KB`druhdcWUSJ_(lZa(`@=XTk- z^tU!$TT^}ek&T#(hS~zk(soNQ139!%ZeM6CjljJFY_m?%pczObG4)>KlPWoZp`4+d zGcXQqa4xGYQrGi*`A&&m@IgX4A`y_Zg#A;ZcE4-0W`?Ktw9l!audtXqP>);&?i@o2 zmS*?TX()%f;0|G-V+-Fx&-S=<%^+ReA&G654_&-kHD`wfqI|Wm8Yq-(~SN+ zF(nqbEd3d`F~Cg*Y}}dYrPndBh-ok=$jcBP5kp5R%+ESP#;4gkF0!2-#tziH58GFi z4xRAJ>er5rquSF{FLhz3-isa0M?iDa?l5miBiiWlLB*Tm4#{o}9j_Z+TEd?A^>GLk zP0_wf`-CWQD89kC$Yr@M*L@tViP;nXm{hpek#1gcJGR?0lW+JSYN=4HU8(6&YP{si z6RP9tg9OlsA^%=v_HN?-UTl`S((k1u(08ZIMvN!_MbE9L#X^L4&=vjRDJ_N&DXrad z8ckX(6OF5b%=UcYTW%7%aKo*tCc|I71ncTTBe+Z~_*`b)e@J~Rz-AfHh<@|F$*H)f z)^OAL9rrxGt0~w;+-}u~Mj+b;s${Xx7_wwKP2Y@q{NtqiL69!x^MaVeuu;dGgQ$5f zRIQ;H4+U!fbA`?4uPQo+GK-pHr9NR0eGZ>c25#**nnrPFJ;xJ8#2o}qIDI^aq~Ws@g{ltA=JXTCdVD--Z5h59|OaTHX1!YyEXda5iOHUBd9iugBp>f z1-$hvG(N-(^@T6jj__{69Pg)I40^N`fsFYDPL|PY+(M-_e(W}_lMZb*l3c9^Z{4_P z@>5m`vV;I5^kPn$6wn1{MK75pQH{y`0ylZhRfI$>uD>=k+G-8vaiyu@j} z8`*%cJg-_5c31f$0n@s-CK#WOj-(I#bSzq<0sB!^EYJUZNAn0fXQ>gX<3mEV*z?_U z21NUjTd2Hjlt6&~fKSQYS-q7dB1S!sv`Eh0hYMUY!OU0;6%hZZw<2_|s;$wE_w%RS z?PfckPeeP(-liL18;YEi|6?W(%l|nlPWzIijYVJSs?s*hXN!cxkq0$LjJMo9@maul*_(LUG z=Mm|e)A^VIwsJHOvJt|zTw*?bHVfZ)v8lF2EaxH!W$#rj954M{Z(_C*nH(3Zr+zl< z*G`|*6VUrg8-<&EfK{2(*Z@~?LjT+Lcj&>)$ZSw>xu;S}^v{;)a}FE%=iF^?>X+u~ zPd_TVycDTr;vKp$%K$uIyAfVE4O5e_aM>KoFSA%lPI`9mYc_*778#Arc&& zY*SSZpzbAi)s7(Qd1MqK`OwH_8vXgzl}${1vJo35_(7QBeuA#{8>zP-S*F-l0s zQ@3IyDBub4=O1Pv#GiwTW>h}7{CxLW9s2dXCFgU)I#a#xEk8xP^lzORdQBdc-RMIa zkv%UeRn7l{bbqAg0eRbJ#UMYpC(fn^cmg zS=N9mbx~&MOLw(>e_n&Gh;!>MsfN}rC7(ymPB7etVNY*hi{aT&^A1jYRwMfwLlC{H zjHbEG&+Z)>PFu^z?ZZb|B+0F2-~A}QvecQCSD1{jDU$hOqq;`Tid_P+?6VijO7-a& zx$@q{oZ?fC;|PI4c$da>P4iUIqKfC2H|Gu8_9@>hWj3By8#Z(|r7k5Z`+jxxtgc84 zQE3=2NocHjAlZ3OdVF$8_gpNxoOhnfH$38%to=9M1z(R9+#ylA!eU^&+<1uR+lt#E zD#65`@Akn-qlE}NqO3%wH?beGs~fyF6*|^=JqC)(pMT2UNZ5t^DI$|bKSJh8(E1=H zzWS&*4>?BcRn{3z`$ne`Pe1ZHJk+3v(IAlAZLKH$GU3*n7>%qLS`kC^s?#F};bQq> z*sh(fQ!EtN$w|zod0zq)WYJ&3&iN#d=~?b)sYibN6HRUBJ2%e#h{3(jSVovBtd#D$+mc)>iZrEX8>=K&Vk} zMq8)Xy-KiLGiXHwNc>cm`x3ENVP%$nIuH~&7U%@X)r(kQCBMD|Hq zo}Y&z$z9$H=YLAC%thj4i)CoXu!N^bRiI{fe2#B5I=nH&3-{Vumi1}1qS~97Kioa+ zzB|a4o{Xg4#pNUO0>kz!bKJ=OHiP~$1EhaizZ&5o+b_O(m?6*bSDo6?kyJ4HH9}KC zuDzw4KT0L0VvswQcX|^R_bKAS^rn`^x3XLjU-9%AGk!H+ML1UbGRqDg{rQ6`yQ1(r zRdz|L4svOUY$xoHJmRl#04?FKa5Df-ZRPTk?)&|&(g)Dup8M}h!_kbZBMt&$wqI#p zV$Xz&g0B0>7)!0bLP!R*MLeQoiS?}}fQVg;CotQ`lmYM?A@f~_-+8odKVfa#|P;(@=%}&V%CciHv;+aE3 zBzqu4#LS%NEwC~+ITdHj*yt;zIoa#?b+Kg8DR=pKaR`@|)bM)!{OiUIuVcm{hFRl% z8C;x6Lcn{!q*i5{K-smbG@q4_Lv2N>sx3->G{aUm!g9XPG*k7I^dLuv@jaPv_-;Y3 z>uhxdui~XC?}E{W2oc4IDet*(S&OODC8OOuW|CHe+@A}7Wj?}3vNjm?z$r4@lq~6y z#C`HJo)!c@x1M2ZPMKhvtg2F<607Kn^+$RBc#6wvsSn3X^`!K$XSmn?-|jR93mx)& z+bd()7$I~)ADt&Ea&5R^tjSUFh9-?Iqu^h#dp6;L?f$YNUfcbQF7I4ouGZHbUvc9GsU*yL{07QEaR*^D@OIV zY~hp~x+~$hzn>|?ETIEfqB)Kkthdv@Qf0&Ty?=ys#&-VR0r@CzT*_LsPkV1|_gBX& z*LF6g-zj7!mGPCqG`l9B&UzsBQaZk;p$5@|n0%vL+M)A-&v2|hj@DKO)%42^yA-w@ zp~frevHZAX8S$XLE9ctqnvejgUfN%^W1+R0!IhLPt=z8U#SR)4;o9b-S}@uiA+7ku zf0Ry0AMN3}Eh&zK>Q?5XKSCi=xkpsq`?o=MtQroOeoB#M87q`*(+fp1IZiB(0iz;BQR z0{{F$NEE9e{OzXio#NYYZuPcA+lEQ#4r1ZGGlc_<9lIt$&*2o@axx93#cCF3A z{s1RT^EQM$0W@5o;2eMs8)fr8p?EV3B@o%60`lc{CAd1c6Xp!srzUKbLz`l)&sG_l zcRHLj!+i6~x^YGyib~-Bq$%o2ZuFBX5n-M*^1$A0mrqQ{|4I)44mLG3C~Cuo`B)#y zFN60N#~*t61;+d7q~PbDtgfLa>J@QEu!eUUlZ3qRvr9d_vkr{~f*szG5EXa{*bW>FrTO~eoR%a<29wO`3RzPzd1v@Cd1xB2ojf9f!-`}5*Bi7GBbY)FC$(PD3b zeIFxM*@C??+99docf2XrZ-Ql-;e#SGgQ`Cq2OtqpRD|q3@J|1Iy(=bX2AK*{0hB}TQOb76)QVG?R;Nv4GfACM*a^YmQ|@s ze$9K9*?KI-?+AZ4Q`W2N_2aQH;jV5*V|m#am43aIVJR&m&H2MuEmkK*#LrIUSM(@D zids0@$KzD315ZBiGiI3E`o6TcK12uyi3gLSFTFNpa8@c4)J}DuiS*G{HwFIzKlE1W zJ0-8WvblnB!Hl)051M0<{ceT|)w3s%;|iHcxYhSv<=o1JrEvJ4s*P0Xk*be{^)^3s zh}7CVvT|vnZ;tcD?GUZm!Ae!vxi9_A3yHW)=qNcD4f529t)VcwhfbSWDOCFQdAlwj z?KLj2z}lC2%<`i*G?LoJIa=UtQD%~QAo7&sLDYh*Z%Uj4VHIK4TCdCuAJ-@9~9>GiISNR0I0}_T20Yll@}Sr*zrbUB=Of9hffPJ^gSa9;|l$$T*&zE=*u%1 zTS~kG9+<@V@c$}HWUnL*i!int2^WmEN9d5f&2|fsoS#~5gIkd;1-GmU+_Hoq)`t!S z*4u{{Y$64g-j+$rGa7!8;SN|fpf4P_>r)?rQapvW>}iOg_SyWpe;?#)Kvp81>%Zai^SDym{f^~+0AciN@s~Z+li`shIv9lnRw5i zG}-q~*ySDFEgw(!*7M_}cv_W$aQZHf2~=`FgELMB+chGELMRCwn^^3wKHZvs!`e`a zu_scB*2^p?pew%Xv$d!>Q1Y#;-=gkrhig**Bf0g7Ys~DaR=8Uyyj3rCEE(6t^4hER z8;l*ME^1q5NBM<4}-30uENtG+%#=tS;ot+ zj(mI}%i9}ctNqov`}y*goGNZRUKQ@0I6$&cMS zqB6sQ%5}H6zk+ul6eZ8$9j!)k_Kx)^xwS7^PQ|oi+`bK4y;q8RI$CV$8TMseoPkW@ zI?k(fePVbmO=3@adpXg zW*Bm_`gZ0+W-xsV))sP&U!BcQSFtk*-)j7_{T|Y;!4FauC-BdZg&+%iDQCCf!!&K`XkD!^rW$}xnY0ZCNb_ceKJ#FCB@C1mFpa6gCM9tnjazp6-O=L zoHQclwMFPS`#{wV-WrRI^r_{9BgU16M;LpaFKECRyEB*)S%RDQ-E4>&%8%fivIL%6 zmLj^E2f?$jACJd4_9;~OprEpjaf4>=xKp5MdhRMy(Uc zn-w307gqG7?OhCaQXHf8dUbAi4vFQ>i0nrU)KJd_;Ejxc<9Y+fRfQZ^7aZ3Ryjk!k zNdMUK;4bI`Gzc#x6OY-&YwsKuwVToF6bSH7*pmTZ@CAg(qr6&-r}NIUZTbxe^K5!I zGhf?y$Q&@@li-{-9c6Ac-~?mn=(j#LP2Xdy^)WMgj33(?bj>2_Z5Mju^T$>T%*%2x zn*+Nz0*6qYX{BGAlU0U;yS_`ghoKD`2YwmCM@5#2tpszFbth(f{yZS$-e)e|mf}_2 zAARr#E#O?~=&;_#)9@0!sH?fvghT=$8sf5E^RG?(*>~4MeDML|Si&E)HO=hD0h3u9 zcSGiVQD^f=ybk?f_nASi*k9m6Z->Myj-0{|$S+OEi~8$&Ef4k1Csc-_yc^?Av9Wh( z*CPgSkzif**683)EFYYc1MoM@ik}Y3g`kkHGHwpW=h+6dcOTJ_>AeDecYz& z=l|5aC2zr4B^zCOu8dHMj>A1L%#P>XaM{sKla4N?bCvj52slFa(QgY4tcZwG`xIfL zn2#PE`m#wuTszcnql8X5f3-3sR!K;#i`FRx%*Gn_{V$xc<$^e0R&*3cKrxHfyBO}H z$Uj(oR2(j-LO{G+pwWT9$y=?A7mPajM>1Zpbzcdy@xOcpJ?zi#jpQ2j!#t5M<*F?h zzFsoMOKxp8P7fmntP%?9tuY#@F=DG3c*?a(S{EVkxY3fhGA?qU`|V(U%_RgHhZ`R? zzd0vp6g0}le1e??wpPV9#YAIy1^d9w2!09v>0Ai*B{agd;47R>Vp!9v;usvA^y40W zF{on$2?>Wth_>vq7CiLTIC^6-#d$TB$#q(7#f@*gd@2E6Mns=-0T59{Vf0HbC4l5JcMx_^64k(P#Oz&;!ueiKn!`eLR;`!^H=`nB#HSfj? zF1d;vu-eYts0v3?_y-M+WUOi<1d|<Tk7( zhw^7geO1wsFB>22buYObws^`t%taFQJ`yCYLKbDz5`P0$8=orYX8$oR#22Hj9Okyb z)7bqb+GbjQka?jW$`*O)VOJs1pfDja-qnTjI7uKAl5332AU7ZKzQ_!6@hSLqnb-NV zXQ%eq__89S(mzGGl6%YCYV{(oj==3>3HZoqh@dcN%1$F6e$0s=_)7lXnlCtN6=$I$8S&h&f`MRFJ2FQZ^^?Ng`9AWlOL~%tJasM735s8aESpsX z)Y|!m=DSz(rVrb#8lcI9enym}w%9gjBKLRD^_LIDR#v+ps^9!CS?t0}fQ77B68bQW zUEzqlG#*5ZLs1#qD$#Yibm-v{%(Kbb$#;sq4))Ewn>QBDv78F+_gecJ;HM`-ltax3 zA<&YLji+Qe$V!jCg7WULa^SfnW_a4L$lz9qt5c`xXBKJx_pW8QLr+y0?DsJcvaJ;l z^Wj=NK+&6&S$BrxMTY|pVQmX6YbZ06X9E(JO0-x4JmL<4c_a-?nd4tD1QIsg7vWSt8cJZio^ily%;yXcntMD3ClNh3^+4uUq%tU}8Uy&FgMcIUF zvz)IQT01No{fK97QG$|1oO5>zGk;Pm=Z6S;!g`QNqmB1*>6^K~EhB7|uurL+>8&4o z5S}dPoVZ&!b=y{W6`32Mq`tYx&?l-vlM3LwNNpqo!=>tBuZ0{O*h>A7aZGu()C?4%oWd{Gp2uqwSru5Eu?sE_M441Y#j zw{DEwYLT#8#h4?TC0jY9iFg@5I)ZR3rlsPxQ-3)XUF;hV9H2x3McvhI=}`(%UvDSP z2^n2j#`=4l8VWQR;LURT;R`~)SZb*%NP9uAhB06k<8tV3?@KGLV3y{q$J_W`XSU0) ztcn!ZFek{+&s{=0$2wPqwma@fl-go4TyJ}gpD#mQcA?r)X2c5In#2)#-idR(%S&xj z+FmcDieKd>WT3qw7fozi4*);{ z1P-f6TuY>H!7nlrAjw;;>E433`N{fZ=LGoYJX4jllhZ96=LS6F>#D(8z|ArzpB)iW zx3trJU<}qxG5<=L+J435N4*|OEHQ41FiFrOIi>HzHLhx4zD^LiizZdp&&M=`EHAVH3kD%= z?fx>={y2cd;3~?NKJ$^aNl;WhLWrqMdGx;{y7JnX9}=sT*XCFsJd_7G^Yv+Esaqv{ z#{v~h9Z&T3p81}olLlyK)bz~It=UpH$gohL^0}}&`89_F4r0flUjObyLGGUB)%A$h z@@fu1pm}j?MBr3BU6_lE&x_UTuzcuiz%XqKtG;J%?B=y2+N^!@-RvI4zTUv1Mh_2u zbHJx!<*E(cf>cyc_Ii)!_B4W)t0=FaqSKoQ@*ju9o-y9|x68YHzo>RI3-WH%K<6ku z6kssXE`tzJotdxD7FFp6UX{d;fL~7?b69S+K~%-{XKO|$yHy9&9j6P|J6nO* zVNRa(2$yukq2KyQ1g^F6rK<u%6)8GrNbQ4r}-IInd zlwcq*P0TAX5HfLdIj9Vn-P$!If$B;dAhksw*(=MsXelf*Y-Vb0xHti7`H?%JNyEmb zRw4rw8eVpx_lU9SiyKQ%@Wb9il(@q@#LoT4OPWJ~SCu<0H_R{V+}eCq1fPK?#?7;L zqH+8S1!gr9$53@dl$7vYL(QVRz&~gu$3KX-v9yepb?w1QNWIaE->N=C4w%VH6&!TN z-B~zdCe^a)%NEhqYpOmi#i_y*$TxDFa z#S85H10YI%kiI_jR1TC(^5T3P7U|h4abap{4R^-1XSlIa;@ZtY?yW{jR@F$7zdIW< ztMa=|*Xu)mr9RXAyfcfuqE=0>+eDm3>%W17CHm2T4IkWn_EupY{O5WTXskMv0;0-fcYui!Y&Zj<*w} zZW-Tnw~m$y<5b~Z+mAHc`g}}uUO0muGL&bC^JG&O&!gEE;GG8!d6n0E$k>1tdrZAl zzPBSD4|d+tiS)6}HYH=cDDZqL=2g9Zb}HAVo!^7dQc|NZxp)F)sf??Ratd~)fW>i} zPN}pzD_w)A+6+X;3j?m9WA;PzlV}>lBlTv-x8IQ*_ts9|X+RU9@UEl?qNJ405vc=} z#EZ&fc5M!!D{pMbQefj$TXAIKl(a^bN#VV6EYHBeLFFp#SCiV+L4t;9i=<77q=l6} zSC8boAE~9E+wfmg!gxVi{) zR7K=%xU^4o*I&=Hh93akorQVIeFjZ)YFOiYg<>NXjGx|$VX7ar@=5jmBxiBHvnpJw z(znIl{tkC1W@ZPf?oWdDS?5k>p<1F;itl9Zbr1$=+K=2r(+Cb@w$^rs?N|YM^ksFs z&?GTXp4C&dYJ8tHeJ{YMua$1L7O48(&7Ei{S9kPh zW1+rNO<6S+S1jlU02;MySTg*K@dYy9r$h7dF73sAq*hkXBFg1mMCto`Ff1*$&Y-C3|uM(vk93QpUfLFFQel!6Fq=(gniRS0P_Lg-tDkp9E1d1?I)fM+=NkFMT7yNxo}Mh~q!&N7t%F$wbn?Am_*c5w zWV$Oh9)xIzWiRz#CI;$&hNdNE+_FDKn(DB24Ul*Nm5c(q<$!0I3MZK}KucuFUz4M) z3>GV)x^9Pt&sT&j(Ea8;IBy~YGb01B11=K;v)8OwX^}~peE!YF!6IdOcm6p3&I+(H z@OCQiT#-iPu$cfjmMkfuzN_}T$;t)`GA`;2E371brs`f!)o}JpGl)D@Y zn#5_j_6z1yz1@YCO_v7GHM>-&XA(PFSmV*m8!Crj6|*00}HiB9yb;XXgwK+!w5FZ}~bYm~ei^aU-W zJCp~>+{%>3=(9G1^AgYr|MndQK5%2ydTqiBBL9Vn&sK?Uo1Xwc5aq-aiHY$`%IO;b1HZR5bx`?()?Cv0uiiH>U!a3SANeEya9#--$Y{my6!qq#aQjaK=P05mQd_{{ z4f<5u9tPWNV&qxJz9~^qM;GV)QVHDHqhZh_o?@;(oXVm9JS@_=RpQ!|()byLDSy{% z+4su%fBUYVFui$@YwK*bE|OWM_bfV2UL&uDm04_+FqtwPKchPJ%C$2aSO({ws`@Jc z_yFTb=zX5@z%RPzv=UU-6UuD1BT}_rRzq0F;@cL%mAw+60Ii)WqTqLn7f6)SVC<2T zde>{v;kYsOxM!zZ{Bb+?jIroBT-xLh3aP(57kH0)h+SF2en)F=iu;<-oWWyYI5`7})}<q=o&LI+%nD4BcTaZo}}RffV#YK+(*C z6Tp5_wK}!GI+3IoR#J0m${%$1m*U~T4{PcG;BEX;Hw!bZ@% zz6JyB%NN}=cvkNaMLwr+J~sWA^M=N9##F&$d&qi7Qq7U_Jh0haBp%)Kp~$y%f!cCu zZjoEmDijIrkYrismEGGRif18jFq(dkEOX$u`{~iUe%aC~I7BF*!YNZ~nBWX1hslAUTK5 zkVX4D`&(UCZ^*U=3}WIoky}$+U|?~j%lwHmFI6E=rp4JtR+r( zR2Vh3{WOG6V>1O!eoMebC_GS;+i*?>Xk0T~%7F*`7Hu-BRkLgL?L{5|W^ z2-JR`Qcn&c73FvRoas$45hK``q%KqU4F`x~9ShvjD68+?8uU`tAX$qdvm3Z({S{7k z4oE+Vk%EAS-?dsoh#b+=!7K6_x=5<8}D?=`vTchzA#mbS#(j1X#vI`VW4trVRCGg3%i^ZpHX zf+VflNGNT}AV{6vqmdlDnzDLns5&=X&ta}Ch?^_1`vw49KUSs$Rja2fCG zvB9li{?DmFtB)WkSq|OilU_JTxE5om>*XGVy61ZTbI0(0v*9Q0jOW-9Uh!ubfA#z$ z`oDW_mT@`XS@>KkrjR~VbHC3tXET$!sG)qs#M?7RY&H^V>F^2RA(b}?cZr6w28<`A zw}TINqK{0dFdZ7GqXYBg{4NS1XZZyl*moQcbil*0RM6H)+gZgU#J@@`DFMVK6w-%$ z-8c>YNguY3tfmM7upk(tvFDs`3?P0!P=iYP8yX42>0lS1K;1)t z*fL&I2H!L4^^$90YV%apsuHQAk#OXKvH$6Oh(p$bLJeC@tM3j)jktW95|*SA1fZna z>PQvRf;Mx8Zfzv)xE8JZ!*)&*i#0EKX+xCn0PO`w3)3e9R~Lm zIh#p`>CxKwt*l<(;SDIKd1LW@Nbt zMpf!`(>y9RIIy!-CUVjvUa>a5(WEl8d8(v<{bQ0)rosB1Oyb+!uLif-xWg|Br62TU z&D`Sf978>m&La@lPEJ?S6@{udfz+lO&#J~2n7zaqux3z0Dp|{gN_;LfWNsS;r`6lX z;3}jzFQ+jZ<^JzA12FE4h@~oCExgtL=aXG>{@}Gq%@KZq({7$$uh8HDRjn7run9^~ zTR;xA(_GvR-``tZ-P^>c()Z}FMcOqmcg*Z_i_dK`WHg7qG57E>@3@zk!XV5K2IfTu zik5u4lM`e^bxVEm#oNNf-x{$EZ}{bAyS&E(WWHrfsZa+nS=al?zz)&9d1C>FN$5s{(2NtqQ870-kIm^0i zkEC&O*`%MpVOgpiLV6q)&^b3z=hdZB9F)Yg%d!{o951iZ#C&t7zHy%T>ktK&t|G5H zVc{5~mPQc6cKu%h6j;kkro>YGOtytLD=4Z%{T~DL&tsgM8Tvqmbb0i@C(6%QAZ|a? z7~8CE*ClSZq-NfF*Rbopx?asb)WIeZSa(u&EZ%wHmGvJ%HmD`V=qH3OTBHZTrT5^; zrrME8b{O`hzIOk-4B{CG7OMD{c!cm^FQ|`^TnWWnR|zxA$9CjwCdYv1!jN-KL;lk- z;S(6#b;C%nf}lHBx!0sEs8Y|Idl;+PDwpP~Q#%6v>JXY|?5S%o5VWEd3l-%T9ODYZ$=~Ah_iF+nB(dlYus2LiN zhxSX4s0l~J&AU4S*{8E+*-;u=`$m&QoF{fi;NIa?M6qGBZM<;n*OM+;1paolVQDNR z>1@z`$9t;1ab8 zcLBWP+EKjc%F5%YlMM>0NU3aLf<~+T_2Zzxowvv;v2trlFA@(IjukTn$(V-uFR{4d z67RH|xCXwAw!O~Wx#q}ITr!?>;D7u)Mwa>~CU^byJYYQbdaW=Bg=w&O zDK_yrl(-J}whVguSlPRo-^!#cT*c2_7cfAE9f(gu@=Dd=(p&GqR#I#>Va1ow-Wnt3 ze}&=K*)#tvC@+VaU2QJ?)whaS9AxL5S!~@@tK;DiIzIM;jd`MPdw9aFGzwa7d84H$ z=e2js4#OILx_rr^3*WXTnDy{aOOD{Bcs1e!8V1|N}r~DJJ|LcB0*ihj^@~> z$5=bU|1nC+pQnEsU%ImLgy)Jpvwbx6Hk|mj$de$gHk)ll&I#)0ucW1?cV}H; zQUd8~_$!H1WtS(XO)HoB5h0qFAVX&>$8H9dS4CItN6ZcVPDYJ6E5o4U-gB@VhMht& zK=LS7-{YkDc1@T{Fr1?xb`|PeIyZ)7312|=Hmtw)vb(l>w3wIF#**V%d+VcTY1mVB zzT+EX#f3#kRYL(0c@sI(5--xeRpQSNUprgUO*-;Fe>eXMy}weyU!nJQwrz(3`oE=u zQHMO7{$D>1h%XN~Ixw=(0v~@=hE3blt}s{*VFBk>P;aZBa_dvRX>2g5KmC|&wyTrL z-s1PC$A?bBYkCmbM<8!8G+_$*=kW1SF1gfb0ILw8?4KNm)B}jxzQj8U>w1l;Z8JhH zu?TPpCAPLscMrY|LXUZCxJv6IrhfxQ zLlbi}o~{Y^@1ObUUwHk%dr%}wvhoujI>G0VF8*vH?)2L-@L+Gt4~tbN#ebO@Ej{o# zdC~8Vp;8NX2WE$fwv_V=817qG%rEQ$ai2r5HdvO@q78;b(Q4-B;%@Lpi>as+m;HQG z2%r_Y2}1;De#r~qIEYNyv{*A{9x1&J_Xte~6;Dn2O5sI;1hcG^%WKpI{!8aXmjm`Q z*iZk>tdFkT>lACYPs$gy#>s^K>!i!4vpSY9rK>t#7Kdn}1coH+mJ&+?v+j<7+Dj4* zQ^nHSm&j%2ri^kcNEziVkvVrnn^jUL8MtLH>PR}_?u6>JS6VaJItqD zG4wv{M;^6AeIOAgERglP9c0uoTJye2VkKEbm|mA6PU!K7nG1rni2eJ?qj1C|L2NY< zdnL{eC(tSE#sAn!`FWkM;-%sWSz^3laK*#-1jc`wAz z2Zhett2cm8O`+`Z4Or_i6}QVgCV#OT1dXv=r-BCZ#h=)&-yeT_aQ}ny<{#ta|1E{KxR*GdFmf*)6Gf$i zjv2Pf+*Y680T$SqR**>IGcZl=^yp%ngp zcJ*~FKPy4h2#GX^3Mh$#$) zPTXvm()Q^X_K#U)GsQxuJ6z|UL?rc9)?A{r;n$gf#DD*~=*-{&w@K0}YzeWz1iQ^W zdJq26dx96V9cNAan9m(CPjI$l@nTCM&iXRDB8)nd=dOyDjDFOu?=!K#2cC>X5$gm6dgT%5bk+^>m< z)|8VK+qSbtSQS=NAsXKYwG8&@#oAQdzxa0OAdWKNZVP^iV~0_-m!j(%bWJYR+d8Y4 zR-cySQIs#Ek`K|gJ5R^NZI^ZXk%&0PElljdT!i_ukM;nkFs1QkbZ{fAq>5O#gd3g& z334IT(Uj;OB+~B2dNpQ_dUQ#-u?4Au2Y6~em~)l+elrtLJiKuBXMlCogcLy$3h7em zUr(2iDBe1UCSiqzGAgt>#@nm6IyKBp8#Mc?Teci4-I`yNs7V~0fSTMf;y3PRS5?>O!OF#R$XGsqb6A|Y zZtgHVIpraf+$Qpwd-LQ+{)N@)4yCsgriaNzisuE`1yNRjoi#f9>k;ZgeD*-1c_^9c zz0avJ?LWfiU#{ihA%9P=&6cwVS637By^G%TJsDKi^R{%*^BqeHm(td$HJm4(6$|B@ zK)Hj>ZRX07KO&r$I;|!g>KO-HyX?Us#Vk5D9@ZsvCi45E&s!*TGkuN7ljo7LOaQLo zZ=N&{>r12kCrc``i@Ed=Um&@qcPV5Mc^mCh{;)#K_aV(i=#VeZ+@tz9_tr~aXJ>ci zG#Vsm-#)P6wGy<3%K;Y;{YQ(}NV;K>`rljbzxOoqcwlnq|8vXzw~q23YnQ)u_|Sl_ zsG7ta_S};%xAGD@Vj8WUrReIVq<&XNY8t#tFJfjEsy9gK-{o#M?@-E4$g?AbkpfvY z&eHG^(~E-(z@z+gO!F6@e{qA86|#8vGZ?{Qfys}smymL_A#9c}?~mhoR_#qIiM)1v zT06xmYpz)M#pv3B9aPWcoTvNGAt+^mcNG_j6ai(Nh{GMKX~6N)moE@%UE_DFkJX>>>f0ev^JDx?1`=+mv?`i zV*No+3XY7ITMEIGlTWPmQ2OO)C!7v%JFK;HNUG!;?^SMG_vndk3hX*t(@%|@r)R(# zv%8frQQLRK$(Fm*RF*%Qf22DvwcV4P)~K4(*w#U-12YAXke3g)oKe<6{0ExxXrbbPbH+`y*~5GT)| zFA=GNi59REJ|h9c|5kiNc9{b{AX2||kbG~H$iK!%KJVdZVa5xpzvc-GcILxEQx~U=3Ap!;D9cETr9{uX7jnJ^ zLZ{W#bqm6Zl?0M>BIRxM2T2~6D6yp^SO={O8iIad2Mfb8+@#xz*JaW>sjR9Sk;|3N zs@ZG9Z-^p?!H@)5y8=81&k{L=A2-o`Xud|e9nE<`Fy9F?qu0|G@?hFoHBB`*sZNv$ zKhdF-UZK;mB#Ao4rQP$sd*Iz%aAu2cPr-|eRXSJ07X$`#+hqj4w9n3?g)&&w^1oEUbKC!0kgW{1rc zkj#}YHzq}t+%r`<(J_O{*=&iexqSp264V9$I?TQns#-386J&!JjIYZ+dH(H^@%sz7PGF;1VAFR#fNOzCl55wgBw;sSE!Er1 z-8{A3#w;;srI$ksmAX~L>TN7NDj9D6Qp{gNR!=07M5;FKua(u~L#x4dsp;LEBeE1X zhnC`rMc-Bqg2-dOU2jv6Gt}l|pSM^`^-YPS_VB$19#G=DgfyafB9})hAzn^6kH~Ju zwo8{J2X?WuJW?~JQ=gEi4&hM}FE_y&!-;q5_gfyBSsj-yo`lVvERL2kj_gp3d(0Ah z_lsLqTHNQZrMl(qzIN^v5h2gc3QuhhfPL<-92?~G*GFw~>j?X#{FC3p*D>|Kp}PN! z-uz=%{*NH?KNVbm!w|uu<+9ScZVUF3fC3a#&xI0@cfjXVm|<*loh`tG+&cAEt7O}# z)M1M=ZE*=JF8UKHoeQq`%IixpU9};4J2@R)$0&#~2rM=3>;5vV_wg@-YBL>J1lI&h z+mq&*{(Y@eG)9&z`J%PM@!y=&u!@z8%g3+pu~ycMOrbQ0SJ8tauxEVyUhK&4l# z47RNF=DRP>-9a&E$Cpc4WZGzUCZ7kD47-M;brLVVC?)QX!JRFQ39X*1&4=Zch0R}J zVUUjBS8x@G!83X1tJhu+gm^d@s`^PuN9oimIDtv%;ZPcxd*0unvgU$Y0z&0j7Cr`3 zPg=88>0SvU$bJ|!SY?19X|OM7P@>wQwET;h)Bx(8?~}W;RNn|fhg-Q0dF?7{6bn<0j{)`Tg|T<6U8n#o{KkXb9lq2r zNqt!V3Q6tHWdM17{rRV?VR85JF-If!!W2^6f`7oEnJJcHV|ZStPII0PednDN#5qv6 zjOoC6vmC@h-}XwoY1-|{Q*=BQwQ~wt27!e==Q8vfEg+q>q}R-wYK}WtVCh~j@f={T z+33J(M`+wHHrA(|y^-JLAcMjMnW5-ca;OVhwHb~D>uiUPnSXh_bX3K}-!ime?tA{1 zTe9GIAH#n;b9Myr8ebg0%PzZuFi-iCQ~CX@66IFX0nAAZt}f5+DNYhD%S3!Dfy{+? z)S27}QZ9YnmVYXizoQLSw)jnZ8pxu_dU`;!0W*j(O3cr&hE(+kYRRn4LX0PSvWG9D z&tS;BxojwqNFvahog>Fp0_8!Mt;`wlomxvMow@1MIXw5RK(L}LkTl2XX2=*V$pcZ@ zc8x6-_$ufj@UD;;#%#Hu%NGuavL24@KGeU5eWGmh)3;$;HXO}JQ4Q-``ZrtI<|f@< zzdhiQcNw;Z>1iT%86QhegUi|rL=g#&v1jM$5o2g**>7cO3YK4Bg;k$>S2DZF=eC(i zX8M4R*tRu^95YcV-)?YTqrL&hKY2FGr%g8TOeOL9kV-C*oH<}YC6h76Gd^wEbKPxV zi-;_{ld5hp)J=-#|M<>IcQn@pFgN~;zV05-YJ&&2E^q3@j)04kW%JSte#p1eR<)%)6 z*pgxC)PlwY>nAPt@}(<_8@Z>B>8~ugRHAQ|)t3Ki&v06#%h%P* znDLg<_*O7-^dLtfRChB2WYO>RaDdEmh1l&^1dW{>9k(3p%iAY&>=cll<_JuQY=)t& z?ULuUGjE@aqsrIxUA?UEp?nP zIa=%j3E|ds?g($?js)%v>d+z7P1swkJMr--DnkbBi<{C3;{=c>ig-9L?XR8CM&Yf zpnW+10gsNM$|aA%vAq7wMc)uRl`S3jFyM1LzTgnmGjYlK7equcH~TLq!*f^S^JCkr z?A_mEM-0WojTzrhR&6)t3otBvt={oEkv~j)WHao&w=~pYWpk9TFzH`uvY6iQa#e{p zPV1Cib65O=ZNJ%+o@$Fffw~v0A~NZ~gP0OWt1HG++qWRX7O~mQD=tne0+9)@mlY8& zDaz0P1Sa;9;3Y&`8xjxq_(k%V)@qzNxgFFsusE4@=e9Vg+7kceG02AJwcs7yGJi`4 z7)6*T%%q51DGfiem&% z#9fNxA#2;WJ!b5o!4@9#NYaAu&+ra;xRGgQXu(Gg~{cZ^PVcY$$jTPWt@ozQ_w z?!k1!g^EV;ZNgoRg!WN<8r|!mb6{ShJJoLpw-)dmY4!aCSt@xDZ^-Vdb?+KGNp%EL zGafbjvDT-^`+HOKvA{sFJ$p9p8Y_N&=E0e_@#=>4zt@2(=D~Qg#gNCw`UfWsc4;-p z+ZdXS-uZPhQRaK@UOwx}T)!>B2X1eb?N*k(i)X1?EbO}*Og@@+1WiGwQqayG9^LQ! zE|KV0J(5RBNvy&_LH3V(QSZr>j!Vq_Nefe}l(@l)RW>bYLmez={Owl{SI4$8m7u3t z%^{lEXRp7@J@Ryi!Sc-b^`P8q3!7xpkBJB;=U(>64LG1-VD)MA+Me7?-MpWk=K6cP z=c;w*tUTO%tv4V2G^y@7AOzIxaG zP$UuUKV8DCzK7eIGWthJI*q@&FK?n?y915m97=0_a~09YZi}8$z=FYU)h{u z_gr&Sjcl@XTXc?X#p`ewg9A%Cb-!TZS}W1!rIKQa%o zV$Pon1lR9#r@QK9Ja+K>`Qaxnm*l1#(766bUB*9rc!QTaBK80L@P~5eM%w@Dhc`%@ zxXwNomfQTaG5gr{Uqzcf8c+VD`U=d)K(6)AfC{S*^)glj*G0bg(>3)09nSrl#hR8Yi8m z)MlDA&!_N^0wsze0#i*o9hAyEqEeYUrHDsR5t(wz6p@rv6p=9%G6Yi;L!pG;pQ>FjtgNAG?(3MTY1`b%sqHxwQI;ZJW=Vya8}M`0(a9d--eo= z+S2#ghriS4bk*B`e0a$J{yX&#`QLx%52sD$mi})}&|fz3x5qaxrvLQ>rTi9j+V8`c zgCAS-^T-QUu<=%->_xtiD%DNl5=K{kkY%;G5Ex5F+S=_W+TU!VJ4{H`4wbahJDuLa zRH0L$#fu3tLKG(9P-jmo)D?wWjUI&Bx)%(#zjW)s=`7`zijnZ{XmaYXw4)-k--~o$ zRDAu)>>WR|;i_qxj92V|QKrvVhZ19$C7SYauLyUBw~Wx+#C+n-h6L$sXgEqwv@}{i z-;%2M|7S-;qxMQ~k!Y^ZScFqgjv~lCT&rA0lOayk0YBiAgNg%(9VW{fwI*;d=gkrB z`)v{+*yjEqCibROS4q`12yE$yYtC>Q>%fR}$EDx2=RajDAz7MmLXrX)9;oIPUZ-FS=TOi9pYC3*6EjZ&d4o$0VS!gy?t2)5L`lf#F zYTfH11omJ5^wCFyzdXzT$%XhIgc{F*$7z?}hZo5BxtY?^yM;vHy%kITFC`)j-tLvbISIF=RmPh()5+`YkNbL@ibJ~!oT2ZQ+38dFpT)z@uEjF$%ZSQbnheT=zmzRg#9Q)7t z!e*z90qyn4wbnxw`ycH15L(S1`wwjVALo_-LCE<_+y+0kW=&JyzC!Gc3Slegc}Azp zXk&rvnCVCBNj#3OINz)pf&5;Qt$jy4+R&)k$JhBK3ai)%QQhtQ$z2WRcRIyxhm>CC z(VcU5+Zq?s)q*z*w}y+AJlpgqAS%&xaY~Y{TRSe{RwwJ{Ajt_jv-(X+lDUPZ6b3;s zb;=6Ja$dy{uEky3;?F9Z|KpAOJV`wE+<2{_q|389oK$aKQ!m{$SQS30eSsm*+MM$x zG3XG(HEX{3>h0@-)?YI|B-#J|waUQu^Y2%08w6WL8BMS##E_vP2ayF5F?o^SzK`T6 zwpbZ=0-IP00|J=*ocy?=h%g@xJFFeJflBbjPG6btcjh!iUrdqFvRM;S&eYtXpA>~Z z^y^?kXr8)5GdXI(7=eeMQ<4(+5Z-Kiv=ZNJSUz>FuSKDb9pM;rbWUjm(cNLXvBA3{ zS$0P&i<;mVZ-7@l-jhNen%x$KMPFW=CU!jOz$J9U@iv4eQ81~r6k{{OwgB(K{_I~V z7fytJ>-ipVtOAdHqrZ5OQcEeVfth^T z(4epTo#4(e%}l$YdeAM#LCHr}+dvR`S$U`5^;)@iYi_ror(bvyD|+>PqN68QH%{f+ z{d~~09|WJfYH7LXjI0TFM@>%EBes=Dry|i^&(yBtVQ370wgaZqMX5FIQn}KH()5K5 z5$T8LE$-=v)Q1-%gJ;WRS0jVpc6-hC6?@EvwKy#w=5unS;y}?j`hf<=7pKl##OGsn zR`R8GcOB#TO`a_2_`M){ekUJNr@`)4db^^8ap016wzVz(+;(4<6}Z;C~$+6`2^tWD(jcr)l^+H5q5}W4(0CeE*=F6_Ep(@w%;ZQ#yLMDcS&j zMw;7L6*aZz0rXcMXs6x}z>^O*Iki{3B6pv!NBi|?`dX;E8mV#u{GBkqcDK@NcOt!C z7M_ohj-D_KSVJF`Sylb4>09wPN9fe&o0`CO+Af=7dZXug#SGSA|I&h#N;k!8*MoSM z2ylD1jJ6sM3gKyw47ly0bj6N^nfaPj2aab@brBQ^fy?2mW1p z{1tuwblS_!<3CfxU+bF3dv}_X{$1C+N2PzsKl*HOFJ`{J4~t|jibfImfqfkz`306! z+839cQU+>8uN0-|c=aQ2%Fgk{wp1ozGOWt5jolk9m;Gr>KXx~lTeQhQSZ;W7LuSqH z5gPzX|H;?sSPVrUI(^}a!o|`kQ<6sJ_SnxXa+30g=XhADa0H1@WU5;Q%10GL9y@Gc zj6S$d{%)&!C+Ee+GUKm1gZvZZ`C}Dv{xhjot|EiUIcHs5r;C-+J6vK#VhpVqS4nS_ z8&fqE18RsrpToBMnfhF2(-9{gp2bCAFVbMfT_7%nFOM8#J36O>Vyw31hgIItr8WaG z2zM=pH#Y=$C5S`(y{$SznQ47+a3YTema_~VX0tR3H9(*j`NBwceP5+RC1*@vfy#+% z6_~MZg#QFP ztBT>eyWk)3Y0^srZKH0^01Op&dU>2vGGT}AD)0s}{t8MF+MS?cA?yKi9JMD9R&s_^ zwpA)`(Z8xD{-#Eg2hO()q&?CkQuoVAl)YEtIwBjx)JY&T3p}1F=B|MQ9Tzr+OA0jW z4F}EOUwr=dHLnPP``c>a;GXk8Cp5R7!9)TO@~3AIzAKvSD`Tfcc%TPs*w>ukR~y}M zF9n%Z53&f_N~sdkQ&v=t?7kjNc!38L1)djkf7&9gbNq3PrY}}hjucIe6@#6m1;cca zxSA-G3AGi728tcdogHkF`_g< zJN0LUv-n~_c7X9_BY&~_M^z< z`*eQ7n6*tI245u}_vc5;zfEBN=*LZ9j=pvMl8|LeRCkt_&?8$e>Z>1HeO)xO*Qtw}_K zn7LEwtvky{@I8#XbWV~>?5`VA5xygcSUhiYWxRsl;@rfDHF?`Tv6J8w9<^u|nSy_n zEzgW$PSXq&>gi2??C#!sL0W}9M6PyI#A14sc0#s-OBI=rL{17Tu7!sFV+~cLY+r zfGTWn)lF3g;xj$6aU0W~m(lWx!X`~yweuoL(dbv;`ne`DI7{?U1JBK8PWFOye9vug z#D+ki&Ri7UQkTuvPiWfAW_4%SmVAY7M*I(XqC7;x37=`VpQ%IPCEo9kX^DdFR7f#3 zs?5rAbM-UCsNRjgFIU2T{tZJec&KY>o$RXZi3&3`MYt#A^1Xlt3Q^SdZx#g-15!Pps9tj}!pz%VU0w_lGhbGNb944-$XEhldg*WC`78hg>#VmP@N9L|}Lh7&iU)KT== zANhHg7y}+#Q5i5@90bpi>3)R?V+h#OVS^rD5NPK4onc7<`04FF+KwAC$I$?$#+sjs zUs~2EHg!bS+ok(wbirpn8%eR|&}F!udOW72etu+j{)~j%;^b1b&}|(5D>gWy90b+h zK3&cF=?f#pKjJb)e>9%r`p^y_uos*J@bl>-$ob~vOgVkJHzE6b3|U%1ua;EMhi7Gx zd`!s7V6w6a;gcE~uMCf1LS}zUu6z>QU#eK)*Sx6+c4*+D>K^=rT%HjC`}(XtLsNgX_YCQ&sy{iN=-kJ(}VT_|3nM9`!uHVsU{ zBma&SP`s>x>;r`9UV!HY(ZFpPWG>S;rQfPj#7Azfur%6SiQnT1YnQ-^ z?9S#{dc`_u6tO+0k|-&@vaI6L&L9%NgCC)zdI42n(Z3*avsQ}TI5^$Yam}-_7dJBa zbA9=fb+_|0D;eXT1u+-b^38MnNDVY*59$cAyDarR!5z7(FfsVDEQha*G5Nup>rV8$$GE+Cvhk!WXn zQqp80STJhmod1n73=)lMM~m{hpnFzYr{gF>{f`n8dtH8u_wS*rs_w(O>zRvJ;FTP+ za};L(O0foqQn}_m@q~N}9@FDhtz%x-wh|QuA$QGd({O3aXPp@7FMYiI_J3zl6jI>Z zC?(8@bHJnGDCqX?7iXQH~? zZEV>OGa3oK#SV|5->d%HQ1Ky5<-Y@B;DvYOBBU`R#RD@4dKArp)ZvEvb0-lxm@_5C zH>3$2^b!UR=oaq$um~l&$votAMq46L+RjSE)ChDKd4ggIym~|D?AL>(F1{N_&+JOI z_B@aK5s zKFoS)fhzxzZ|fh_{F=~$BGDoFFRw!C4ml{6s&>1^_qV-tW4OhhJF+}FdAX5iuA6@Z zPkTGrH~rBS+Cz8ZaJ(ht7ua>TI-NBVA67;?;M$Z5ltrW zIN=>^*RRq4EDZ~K)tRbjhFpM)pT zE8>mXPlt`A9WITN=68Ar`O78bF`5$tWI;u$RG|wZJ!i|M;wDipO8q0h#?+`fi-t?2 zK|sLl9M*Uj$Y-26Fnu*)>7@YEl#2#u6*kgT@yKivGJ|rxBBL{A#O@|a)b>0K!WW*u zY9?3qTm|UluEdB9RHIjySCdk%#YU6PO}^t^hmTCWF5_MaNa5btO`m=Nwp@3Q*zWS? zA;|~TZ!x1NfCcbo_a88Te!`Z~`*EZp>tge7=<@}yLa2Kj6r*;xP)+Eo?y*_0fRTxO zVr2n@D#VY;Fhs`SDGoliHOu0ue|fcb%89Z|DRSTK*GM?PEKrl7Eko@t${CIybfED%G945^tWsNH~;u+JB>KJTq9uf(_l(PK{Zi z68ufm^Xun>Wdye`ZOqVZ1FfV1?bFfi*rVnaSfc8PT9qK8^w?sG7h}%mE>1;Bx&Xwp zTcG`c(`Kox>6#sK0SX@&FXf1?il7}I)!F^(*nTymZ}&_}`X)!$Uc0*&^UQ<+iRyYi zD>TB_7g8JESIFt6t5YE#4lO?9Dm^&hlNvJML#UM=x1>)O9C5-TYQM+=VK66Q^9kbu za+CC@I#;s0Y+ayvu;K@QKPwvRU6BA%F!PW^%nW_yW9&jB!Ue_6F!C*srV#mRBq@^b z3#h{wa|FrO^!p%C%I2v*dv%gNI~%yJuzPP9busdB_!LfEhOSOI8MfSSFgYv5v*wEw z*BY*$?*AP9`j+dIy~v|KP2~q@q_=J!N$Ikz^sklPxaiX1g`g{=h*0O5c$4M5jJLEQ z9+p+@S80^VOe3>0*Tw~(b7;)6#7d;_olH+ z(Zi!^J;DQo{E$m7##ITxb*xd(i0M> z$vO#*oG(q6-s9!fw;6N``?EW;A64hk_Sm?d7sfn$f;m`e+Il1ZQ^2$d8h||dVAW)t z!)BAI75oz^+*6iY9`0OMvM}8RUoL@h>@6)_;|2GKHTykdpaqGCF2>~TXjtrsg}zB- zNDTUFBuz}E_^x;h>*rvTpR6d6rO?u|1c>O4YPt)0^&q6p`ZD$@eeCu{C`JzBWA1 zGDgs4T=so;tv^zBp9TkNU?I@ofsi{^W~iN-b2jc}-0lQAn0<9;14uu@`ES)`DD}6J zzFDbOo=AruN~gAqw;NxBeVw`Xn!F;jFO7`sAq1N>^kg2?(}l;udt!L9;iac%epodn zC_-I6U14@ZC5@*YadMBT&oLs-RM;4~OdnLMy6Zn3ap?3lCT+;6x83Ns z-2092Z9Iui`4*n`3?%4FTT;2wqK#Z%+>7A5FC zHr=6y2s3%OR~-Z&aa8gHra&ck5i@@NS?#sJ)cT3$s}?`*`A|MIVrSbDdvoT$O?b78 z50Melh1$)r1v%vbLq^I+WAJ}aTyG8{!e0WK3XCt)6?J;+Z2e4sTs#0qUHtLQY8pLw zxWJTeMg*BJjT{LU$^EpRaug=TJWsDQy!7u z1HkAvx$qWzBD3x23*ca}Q+mnr)KW`v`A$T)X3!Wwns)TS`MAVqZPVj*IJ)5%Q<4pP z1Y22OpfI2vRuhQF> zsx5}$39OpX+rEpa1nll0l1iP@OzH~mtOp!mkc#8 zA6t^#5{9O=gi*4JTdW4Gx%(lK%luaPutK>cb{oUd7DcXuL|X}WWc&+~@jAEzr$Fra z3;BjrGw4*SZnWB5DN1Un_V-`cv2Ms@E5A4|5~M31%B`5rq^o9JmFO7s3|tV*6=>ek zK6#%v#$Hq*DT_ztjiZQgmyo8J9yBrbGZ+JUl`@6QtRjPLwL>1|Cg`y^-$;1ER1fIxYQ4wKiUGci3Vrv7^7&c zSbc^m%g?8oM=0G75zLlq-8_`H(TU07n9=AY{6ZbNZL#@&G}+R~U=3!cP5rX?3>zuE zU?8qjRL!JPnwTBirP|bja65b?gcM0&CKJz?zz*Gv!PCMdcK{PuXEroP#m0v@ic815 zK(#70Urk0RLLCKl5}?jSjv(rZn+t>oJJ*#oRv41V!Ak+G?l_6TGwLr>i`}-Lvks4( zKT7otiZpjUQ>=YrjBnr`jQ6&>e9Fb-=q^B&?|W=th~Wyqg|hkx+jw zoD4MiO2Wq8F%#m@fsz)W;q5}FKZLrw?NNG#MRw}8sZ|>k$V$DljwB>u$}BjVKEL2l zQbQ+x`0YxNc>@arAh%hD#MlsjAJBo#q z8O=|+_{iRM$#M8~4t(wF+8ex)-owVQD}5ace>_$@Bl zhGhQDDbMeSEW4Zj{!!bF9yVbB>eae=r2YYitW;FriXnxhHV^(N-o21nY4>h$W%_6L z?rvao>>cYEZvf8BgU5sXx{mWX4P$NlY~pEh-is%|jm814GzWCl0=NUqgX6yspOI*Q zMdDZ0*YX?S8d&qSuU0g*r8xCI`QPMs#!m(^Q>JQVu*lk(istbHBZQgRloNmo;-w`S zBfeg<=64jK^$=AgvGDx%l-}(57e)B5KE?kBr|W;p&HrJ7{NEf&wb=7|#`6Hg9rZXT z%=t})u6X95{m}jb^Oe0;w?paMfPInQABH}0u839{^YU20Felt}-w6&ix*f~)aphOm zCS+E`+yPi2cv80=Xc7rh;CAp9lC(=^GT;@-z zRX>1T=egM~|1!Qe<~ZN4iA0K5TVH|NG)h0}$ZcERX)wYm14fKbB^}HWsndv@&>JnS zVQ2w~9Xfc$n9tEgGS$Tb?Hz&kvAdQ?CCrXQ!9EJfU3B{*o#-#t$V}H^2fvhE2a~;i{K&7e$}bIXEZNv;61&Uc>}^ zm0}m_1f*+1I>FA*i}cW)0<^$tXXbs|4Pl|wLxR+#w4;rP6R00KaKGHh#uSfV9Cu|r zrRL@5=k{e=jJ1zuD}zzR^1*%I8=JN5k_P%;l;4TDTmSkDu=MOA&XY}I&{MQ7-I$dhj6*CGKwWb`U3&NCq&95={ zs4&d(fgqmZJO-%DCMkz)F^rs93=-B!onUGQSCI{0!BrF(^Q1K6fGyh**0t)f(VH~m zCbbcD0)IubazS266;3I5DIH99o>rYWlakq&(K(ogk!AJPD_=2IqyW~}3d&Zqv`tjs z{ix=&oQnfU;h4Oq%eOV^r4PFK%Mr)=Z6hg+#2?PpI1j>8`=L4O`;BMXJ1RF<;6sLd zQe7{gM!Hxs6B6bkMzilI$}+AXqmf}nok0dub+_?d48?75sid3OiG!BVFgy0!eB%6IabI zjoxlcTsR7lkA|#`oq{4u@A{EwseyCsbJz#&vknR#c;B)wm0y*Kw+L^6SG7cqE$l@B zQ3Mqtv@q(+-6VbEE~bPP%lO63=#0MkJApI~YeWv*!d5LpnK&#V z=}=r_x@_Tl=yu~TU~tx&5ECy+Q%yJYWmWZo!316ZN{IP(RU8+AeU4M6$T8@|=50fi znP%P43PYgHf%(~YR+pB#i!+81i-K~I#f*wcRpTSI{czn=#p1-Mh4orkxC1Ln$_z_mtncP(9JkzJpe7Gx{y7DGU;dQ9l*+pxH0& zey?0^Zn36dL`OO{EzA`7kJxGgO5>QVhJAsdsS7kPF8Y0@Ts`YW=>597&A}dBbbH_Wg6(1 zSyEtOl~8*`)})IHb1`+C-_%{>^h@K_#)%iG69HnJqrU9nMEALvDgdEvL4g28VXmq8K*o49umV8|Q+p^NGX7p&ZKy1bNl!ikDAO`Q z>M!o14W5>w_T7FrdtuAS@Zx-eM}y(5;RN@+fMJa@CpM&pjkKo%4KlR*Gk^02+-44L zN=j@P!8pFh{#b=fQOutjyMeM=8`H6)L0=twU`o(W$eC+nEzQ!a^aQ>%m9{mlALptj z?Qy8&$uVWdjnOW@XUlh@ip6(#)ZPz9Zj~ltu-xp*HiSp9Wspq|=%B;4TF2g<+X35;HE55DGhG{T)N8Eaf{mnriU4la!vUX1vV6LwCzO6{e z%!#?rs3fZSPp3QSvhBpzarV*XO@2{dyo1B8)s>;zG96nYB!z_+`{3wuiz~NNXCyBvxX&rHbg_5=uyfjf(^?4>!O=zeLKYTkutLLgZ`eU2cK93Zzk`JG~I zTqjN`0$x+BENg`b3nzAZhvj0(o-=bu_FDPUo8vQcpz2qwPgJ+r#pUHwNkteW=_mg~ z4J3<%CXa47^qAc05ANuPT$=&}+eKB&@$vyJ4rG^DWAg4DESvAg%4*Ayo9pU^U=puEoBkq(Sd4BR~^3$TL@rt82 z`qw(~yW_x`*&2myi?FrXSvSS43Wd3)H#&rP(BgZG^#^~iD-|-SEhbD}yIp^oUmTM@ z){ul4iv~th$R1K(d#X96E0lOD?0hGP4bq-qM*-95MsDuBWBhWJ_8qLKEnQY3|5oCl z2icNiV|%Y$nQGeYC7tc>RnWS8C=RvJs!JNIsP@ZkfYSeURaG?D1-O%7`2jqw=P^7@ zdN0QVs_6bcjy?CNC4Fw8cIX~){`uMn{vGI6>D%ZGnhh;la}~Db#QqaDZPi5CP;mij z{>Rw6Fw|DEVNeP7WOOI$su}+nI&tzhARx;#s|rfM?jQ#f6g{e6L1&jDW|~F)PVWGb zFTFSpGP)ey<^abA2CwVLu)3R@JNc7E?8x}7dkpXH!y^~f52iOsf@V34wIo^QKY)6* z={;kG{d{%QtKX^5ir}sCdpRP*)@cX~Gji6Fynk&N^fQ?Bsl*ruyOFnPw8jI~Y!_tW zrLGgCP8MCAo|d#uo3m%wAfi?mVDm>2seq!&7lGm5$mjlNV8;V9`m{;s120MDM=ew+ zeIvmraq(UH+ZSOf4ia%H<7M}!anP&wHsxvE-O+kBX_Sb32xArl!LcLqpoF{33=XWD zNn;d$L=hlN(p_h#xE*nt!T6ZKI@u;YN4QgC4UDixV543Qd0bzut9Ng@q9h=dYo4&B z91A25YM1Xa108Q7+iP{79C)kfQC3Df^)%~1kVmf`vHP>|{0kQ69Nc2w@{%p9klr?E zuPzYWLC~U!<`jRNuZ`Dy8XSz-D!n;<-_n)JV(Ypq&6b{uXdLEoy|rY{)7TqrJXtQq z0MoIyBlq<(*`lB-^9qI~o%Pqo5GAuzoDd2K*@^bQvSC|q?ccXw7yyH2;;XxH_=2F!%fIJ?T&9M z)*9U4LP8c4jr^GHUcny9UUVDwED--4C~c^MM}`$Jw_A<7tgX(I-5wgchvs$$g}(qB zwD4Z%HFn(xJ>8t{r5o+-<&-oo_h03U4~%RA zq34(J13`f%QY0Xbhw5LV)+iZ6~K%ha+;+f&Ox!oX-#ab*34M=0UL22o1 zU~C|mN!bs2qi4n1;)(e`inrX&U0k^Z3a7`b+JwtnF6R5tO5avgoRPaM6k)HjaW)?*gs*DywUB2 zxtIoX-ob{{c~4--|7}$}!XhMnARlc5Tn@UcZ+iR++~w7Z;;dx}$+i>GagGL7celdg?rG!~=Jz>y_t=cS=8F9Q3m(iL;) zFm?V`h4buG$3Rbw*$FEpV#SFeTN9*uzjJ|wG8+TqfS?BE5tZxBVd408suZZM+7^hU zUA59eoAU(3*>9UymHn6878b(63tbjo9&s{@YfpW7?D)|B@7Sk16T>~@B%$~o|Kt0u z=srR2)DnCbzu%oP$@bD`awHQp79tp9)bJk0VEy>dKbm-r^moKfrDR&wXXSnLTa%~| zbwbFQAmiB^%ardtGqUzD;+M%=f+^XHml}^};F*i~4!|F$EU(Z)AO)b!DHu!zh~H1s zXai(ybad%fG^!y=5xPRT~!{d#1FtN?11BC;f~XG^y?H_;&Z+d7%2gLfr&InXyAu_ zO;#M?T0tLODrZAYF7I9YQzZx1efTtRc;%>fT7cOD8yjSknWm)`$6S}#JJ_I;@Wq7O z+|BemNz>RltorHsTzVKi7NUc;=1UzE@~I9cEO1e_PpwX)WO~LjQ0Ch9p>4cqj+sYl zt4`<$iJ*n^_qo3dWG@|`9XhC%&e*vB(Y_<`yc_7sy^X{b$6I!~4_n=>w zGsm9AY;hSaj`rw(zoyn3s)MfWS9<0VHCpyQo#^cj?Jd}+u7kn+Iw`_D${gC>TMU&+ zjlr4Wt^C`CWL!8?v6F4#i|P-gGLi*Pr|;L;jRGShbqnjWpuj$|DN~bPF(BVmHS2R~ zJr}3<1hmbty+FI(zP0h_;bM&F8p>4}q{~s06N;FoqX${EcL#z@a8?uy2&hMEkVNf^ zYg8=OoFpf(Nk+U5V-v>*hLsxvnzuvV$XbSQ7wyMz{xVv4VQ28{98*kUbJ@8l-RX`a z9qWzai{#ZqR^8txOZ#V86?kD#k$NWt$T)r@NFfhAzh~&K>dq->gN&pfnBc7Zm_|kX z-H`Or;YHpSL1q!jZits3c})XvQhCT_+D3AQSx38QOAl-1T}>uvrI$-ZQcZmD2G*oyY$e|~%Le;XU}|9v|T|8w-R?%@hcv`Ooi2UZ3z zM^_6*1tjV~?Zm~$Dj#{#bQ#@4K}WJxBI^$e78eMWo;kc(&OzZDKkz&Y(A~eY|B&=s zvmVD?xjNkfit93f7ovbj-Lq>?guWE{}g2gkah#I|=!_fW%J^bQE5@K*I5g22e zEOxJ$6>U-ry3=%>8t99Pd;1oIx-v%W<|F;eS%qWh~bmkpVbY?mhX>#F*A2c^pSSKI-8+z zT{#K=OtEBlSMd{mdGo4;yWyA@&K46ox9N32m;NKY9fG|PP0`Ha_UP9-EW&+QT&L#! z;>txkSEd0TvS+ozTlv64^L+C1!gB!nFFB*9AH1qw+|g)vi#k_3M1`AtLyxKx$R1X$ z6D+K+lGZkLKE#sqTpcZr(=Sfzoe*g!!8CwG?MpXTF(~Ic?)j?rRj{1>I-YD ze?_%YUJfLL!YSB10Y1tcVFp<5hE#pWfF7)0e96)e-;hLE*#N@lUdb5q=*FU!`fc^VNO2l-QoKpa$z zIxWGeTNU-BLfqnZg4t|Ux>K+rhD}kigmE)1HVcBpA6Ba`s|0qrSiVI=`bq$P*+8vd zR*`T%pV`MlhQFl`NAVK0e}rqjB00e@kG@pU0)4>#prCQ$9?`rZCNfQP)(Lwd=|~W8 z!CLLN(o=kp$tJREQ^d%)G%4JYT%Xns&8`V5mSOeW?#9^4lC~A=Z}g)}lw#9c>Ybfn zj7ujVCgOUXo8x;u;^Z~i{_FZG0=}g`4Y{ob#a79v1vU}ceLpz)3(DrAjj^+LIi*br z`rZ#Yug!H5(%X6}OkcKuVL9lPdq5fM!x);8Qk3(l2t5NnE#%NU-Z_f_y9ywLb@%mO z+K;dK`EA{71ENqHgyGiNbOa|pv&E1nR-gzwDj{}B=IE*jazekPL~z2RWfD3+YSqfO z;F4lFeTB4Ba5mAxPa%MzW|}-iHkh1RF~xz#yB__&)&=3V(PqJbls`rGs=VaPf%RNg z#pZ{8h4`(>+-J~Rl6gt>O{#kfd>o8`+tl~HHNN=mL*=kEmi(r|y=}(&`@l>4lq1HV z>18e83}!(^Q`Q2|f@36Ti|F>>a941qlU+}FofN)_EcAeY=}kppO;wBac3(|pg5e9b zfps>u#&fd`PV4!c>2?sSM3y&zX#cy zs15keiBK#@q~{YiVIFPqW4>m0WGh`itg>q5{VF-?{UEl(U3{FRZEt^RRD-DU>L(`# z^yM*To%DC=(A`bXfPN6AB&9qCEF;!0Z`lxK$6d zEQ|3xNGh<-m7H$pvojdayR=D8+xbJtbQ*C3 zCaXa+1$U%UI#qD}25$pSVZ!m&>B#)#{^BD}n$CMUy-d9uzHVvRGmL)&&`ee1d%3vw z-&f`%j-;_)JzbcTXMdkM-%s1dzSfZ14I1LRHj+I7dFKhvC{{+>pP+BP=`E--=z&*5 zbDQk`bkxYdPzFVdwjlLJRmNnE;t|pSVhS8h>})JLR4R-{fCUM`WXSc3f29`V$_t-l zE04EJ*q5ccaqW-xm!-2E?UGg8!+{sh!ZDqy?`DKroj04-5TVBZ#(V`nVF({SQQv=( zEY-Z(ImwrVEZ0=8>7$U2|^Ulej{ zv!AuEUnmf#S+mD!3;ndc?CV#QFEo9fqPg2wz-50jy4`u0inkpqLO0h3nw)Tr9$TQU zw0+HbIsC5fC}w4T{QJ0o!u$fP+(5gH(tF%n5$~aR6S{H$+_xq{N zodt~dpA~hYHw*6uN|I-Ovecx#Jhr7EAQbdwv7ib}=b$RC<(%#VI?;{D@Q~PrZ?4F{ z5Nd8lcVZqgpWlpBx0TG|p4VZS{O}mBT21H_eeBECOF~)2@_*`g^o{_lwC{~satcUD zA0mFt*KN_1>$f?uJptUW?S3>O%CyIiQ!?|AQc%@>JqoB!)d&0+?$=14H?7BRCFj%} z2q=IY*b;6BaUb@8!gAK0f$kwsvs?Va>=JZhhyEB;Aa03jSPva#R4n>nc%e(=JE?|6G<4gEY_5- zycrg^hvzXnNf*w5{W=w{I`CzYezM_TELs1b5BDn}d-bZ!w0?vbL%x^+wk%t!KBcv@ z&yXUW^v_i7(e>dePSZj5Y#u+@!1ZPSBVOqsSf~ZeY~ssfmPFO{XMFKzLCzwF7*HL) zif|y;sqxg-oXvi0KZ9yng&H7OGj(Gs<9_q_IVhNhH&u?*+A768tcji8VE7Uf*oPp1 z9+851XMuoH+OG`8*Q>Q>m6|6j(W+rw>UDv(N#&_lNk9{o2}Xti{f<2#{mb>U0RbOA zWuv#)ub;F5(C*}Q)FHBO~()cIj7eox!-cPxDp>U;G@*H{Yc9ZIR`C!#~$fh6q9C; z(j8)(z+5K-fjbyw2^euwzNJ_AZH%-Gv>bczaZjlh_K+y1`ZpM)v@_VEQF;M*MiSv>uW3QTU@FNc?pG^VUS740zDO-G@+_KB_e7bFgY!7(*9bTYKBb#6sAGX#-;?ja@*?HsTP)QA z>v|dXLEG3{5N*~3uU8{RV41>**iY17w)Lbk0&DXxzFe?k8=~4E??Mrq6Lo6U`?Fx9 zuAzOhTAD(y|4wl_UMKd}1BrZn`eFRRtq6~gsh!YNzt2CU2^)umI@6fBjrCYJznM2;)*3WeqO@O2~TW14{xSl6wojV6T-6Xo}-S8<$c!@|}{*c9z zlBb7knh%f-@d2cDVBbSFxn?Tna=seAntQfOyb`NZdh3B>@P}1w`T!$;okama#4JH) zF-O_Y6^kK42Vh|lKKFnc2C!UoCDV?2d&3pq=5+vt8x;kwElu;s3MJ30#4CC|1N{g&E zt4{_vO-YQ>VF`I3H74~yk2WS@>T_z|?PwJ8XPI#h4A~uc+8N+8PU`~SRGSqM9tDvK zWA&x*PHZaC3`3?Yv&Kw~7WLKWMf@8II;RfXGbezpVrA#stKw_@Zo`dil<#X|oTKKB zlUbFjyqyk;&sV*QAbbICfcUK8{chk%cK1%N>jR})G-v|N6oq#m9y;B5!Z?*|u{5$d z6!KNTx;`MLr(LWWjSV{l0vT_1b0Nn`+1^Zqi+6!Y_q$^T)^2t-OB{*ot= zp3`;#5+s^~CC#3@5kaBlw~W63?B1Q)bad4wY*Af&xHF~5Fc*!@9k`ofG*wkY{X~(b zJjj`QPuRG&S6N?XjCsyL4Y(9s4lhb4ehW0#e}GEN7`$;#urTrDa&jNgg^!xN38H`D za0;4Rm$mM9>aI%15hIs)&^#C>IK~WfhcNKKubv8;GzRs3c0a>DSKFl@m#^18Q>=^T z3_sXnG8k>csAMRdDhRJLqB((Y=)4C!$yi5QpVh-*q>(Z-@xpuJKK+ybzJsFI@kp<# zSCpzD?G{ALID4!2Q}-xZl@5OM9nDX8$n1Bai9gJwSj1?1jVGShKJzbLyXNOl*pc4)@%OG8n&*G+EXF%<Xr5emna#?ga4fY5p`3lZl5md8i*ApUN&LWbZ;q^X*&)Rpb8_!7HA&{i?slDC`OrOOm=0rswkZ~f zGd6w&=HE3$Lxxsb2QDm3f@ucp3BJb7Z?fWP6-1OG3DmI>u20)1Wr+5w^zo)y2E=MP zUq~YcYyQ@eQWtF%{%3eZ`_M|B3Mv0eE)m&^{LS8uY;`RS1usFeuxG+R+rTxILlyKA zjP%A>@K}%QEpe-Ith~7g>669bXWNB9_IbRry@FIUB@D5y7eby{wE9GJB7 zm4`EPU-4Yo8)_Pc7WkK&2x88_pK8o zGeK|TkMe@9`Uht$$t}ygS9dQpW@R#|RA??0O!XT}65h+H*;ezbqx-K4ti`vOvB&V% z{Cr?e0_X1Q)dD#BDa^mKV~;us&eao3&+Z4U@}P47q~O;2>g%ZK2NL5eFlz(BT^36E zbUOEC_grpdivxFU{FVZUJtVPHYm?AA9bYw;Clj23h3(d_J+X4t&c6QVoXphU)_EeM zM&e;k0zmO0dnhMR{mvpNN?l|(IdvtBe|M^2(7JeR`R9>{$3)*!J%_UY1Tfbt!}R^C ztiOXV`B`eZT6teTKL5{(9AQ2j4Z4P+PdmA%=qI_PaCM;DehL`R4Y9)2`Ky;0FCLGW~$+x(2^q zGS^%$Yl3`}CTEFEpzo?cyjTykeQT?Gq3wqpfa$NRa~NCF2!DC3vM9VsXak7rOQMPW zp~bR7;0*vxeg-!G1WRAD_i(5wf9Y@AqJH|X>xKHCb1F#@Bon$mp1$?yi<07v37 zSg4X3>M(pPAomjG^`NPjmDT;w*0{7iHD_@9uOx}$0bK6AM;x;sI7SKL`uGKddiO1g zn;iy5sjXS_)2ezv&vW%@cbA;I5g7daa*t3zGKnD@-yg;%eV2%IeNz!aojSCiHW3K9 z8Pl11Re*o28gE+P{{(f@uO;z9>+CZ@GX0d151nUOX4VE~VFBmLXRCTKpZ}2Y(x2fh z^#G{hVm&d;_DzKaB{U+HbgPKP`g6!@ZB+16yEOs+5G&6(Fp_abe8l81F>Hg>RBnta zCe^9K+hcx6i<4FG(a!qSf3{|ISFd+16%~`rA;ZTIrperERQEE&OQFcbbA!W*R!{xf z)BVF})jflzq@wiE^Pfq-cdg-qX`)Sh(iwpd!Y(}Rz1?YoTME6`%L0jg&ChG}`B=C3 z(JBb~`WTyo@dilXGWe;UDLkSf6ZrqU)_zjZQ0`kSIuKsNk+L+jW4x)8wlf&SZleC=5F>~WN?uZrN?09Y4;l2<;i>Tyt-tT}vmwFEW& zJ4=4v?uAPoNg3exGi(Y;+W42pNC#}%08|O7J7b&$+s}HNkk3!Z8`=-r+p2ox>~M;# zf>87JVMm3iaxIYI3r-7?!6?lleUZ9m&5_lOkXt3$$5JhlxeflQ+}j`s&H`qt4Gf>T zbim;t)j6~wB5e;lJF>*=IPj+CqjUz+C9nM+F7CYS3k1DPz4B6jzP~uMsz^&%30Mn5 zQKh#ZiaVZDvT*f?WeG6pvM3?NMXb%C_%n&St*^u;u#Zbqy7wkdQO!E;w)D7lWlx<# z?L!2DvTf#08Q0S$Eqa7)yj{QN5vv;C-zh+$pU>-Do~?3B45@{pqSP62)C&;baVHEXs$ zSk-3GB0FA_UKQTuU71M%M;T1|ioN0+6rAx#Mn}~83WMi^#sv0VC=>b#Ig!Ot6b1N< zNQ_#izu}k0D6!}iOU8wF)J9!vbiY7zUVnntfLA;4igam^|GpCO{>u6^)Dz(tP_iYy z^Z@{&-(WXAzS=M|lp1Hh;eH|X6Y2G50DSpSesNq01@;W71)IaKGlfy5iX1)sVmu(i zr!w`$?V2xE^}af?KlUzX07iZbKsC1D>_e#T2*J}=A?<}3eb=E5jrBUaU;Wo{*OeRu zVDS-E)2PA8o(UFibxrh}6~S9m>|LdknP3bPV6l#^YLN7?e^a5>9~Hp7S<~Gk`6>DP zi&}E~1}VQ#QIY4mZ}CT_#~Z+?%y=-$v}2`2vr;lpvakcA381*RxYvUlzcdc(F=__| zHxLK_BuHfE3kjbLD57K%cMlI%&6WsrT1qJ^lGEo(w$HajfCEr0BVX zG)qARJe{?jNfN8140g06M-hHp2@8@JEaKS_Fb|2UeuXBT7cd04_u}`xDd9kiuG4*j zUt`8@EIl!x`<_%%hm46zk$5#GRzsX~oBGvXyu?HMg=CbCU7(G_@in;`{s`dAXyO4b z!6-ge8qO`8x|*pRa~@G3AFdnRRv`b-?N+ilBJ{NlixwsaP?9#b#JFL%@q?_r^Pc1K z;C3CvAeu9ED%B0!OxkOZRkeet%clN7>6oG}wqBP$;XzQ7f>Y=&1_X}UUDytWk@I0X zrGpi0^a&FW5#TNngI$Wjjw@Vt@vr`U= zNRL2fFMh0^H8>yCmvjA=WTH6oKsnjyBeHxg%Q8Nn_H&GX$#%1)^v{w@hA=HwD^LFnb%yKG;r z4GJ)CH6MAU_+#VmkG#?=Bqm!P@G6lFnMr5CrC>d9#NsoFN-R6(&?l^ra_vM{?(;k8!K}y_7XW~x;2qtYSA@Qr(dEoMD_1hsXg|T7rH@cj9 zx{seVoEuF%xl^MhH{^q>6Zf+k)6*8#q#B+}b9^N|v9ikBTR`X~JpozL`M(`h*zq_h z?ksRS0o+Oo1DNt1l;-$ZF%%D{jgC3rT_FFaLI5$>F^7QKap?_LzRUUi_=X_gXO|tJMW4EwJ*4f{i?K_Bn9fsJ zvC$=ecR!pq0GkeRPume-sm}Nb7%CSPNrmn61%c$tn!|&VbB5UBvap6Dilgzuar!F1 zU-=%;QExlxHF-EUhr-B%G>PtOftGqglYh7kH>(3*>TyOTmhY5K_)dO_(n?&-jMPMG z6Tuz@Z8ZLX9zOk7nI70O4k&rAPD75O$fD!RjsflD7AKL-iuxZ70r^N!;XmxH$m{SM zS$KZ&`+&)hXV zIukAWOh_>hIJ^)3`Pp4MPX6wz6Dbw*8XQr6BJJ(#tRiO(DhL%mi%-mpxYq@uV2cJp zidQ;;;FdLr7ac&mnr=+x8ZpTzo--{{!+pc}DB!YXIab4b58PjIbrPrKVm1??Fv81e za;ScT$Wf0=9@JhaNL3@u(%G6ruYt2Oyuq@`MI`b*5E~s zCR(8Z(p!?IUi8N`(f*#LCQ#aA5tOi~tB%rY05yrC%|Qj=GteLF_#SW{#^qa71n`Wo zJamq=$?yYD^ibBb{r`Ki$JcZE-H$IxJHXZ;iPnXnI6__|$Jwt7?OKLbV-CZZscI|G zDuOSpPm#*uo~Oy5*0M3uRkx(8YtuDHf*lq))?k$!q>7Q?cKK7l)ee9H7%8jS8g`x$ zdt>c;OIUCp|2qoWEq{HYJo41!>ZO*bi86O=lgl$wcXlAR@*|dG*Zz1|j)ymzV|q|I`~`_=XMxxV;KFCICAA&t z*Z?>ORRQXb9Hx_|MK*A}ItPw_D~iadcAKg`O0qfi4X)y6Yʁu63CNo3 zlUv4{0q5kpV?_BhEuNpQR%qN}Y}gsD)YvTRHqKS7cOS^9BCF-Iek zEvp}+-TXb|Uu>lKm}T3aUHt9gultVz%L?^?`3A41CU z&@cb9nxv}M{~ubPo_U(HO!aqP)qhmA{?_rX?LQP^2>JGOn^Np6xI#r@agYA}A7q-g zYv?Ke{{h;b+`l#De`^tR{YlCF%$bN>k%K8m4_{6|aqjBnzzoBz=9+y8RR^oSuNY3{$vZ}zK6{CsyTjXsa; z^WPNG{u_dz;_@{46UW0cA4&G+1ZlHI#8Mv5ZuV$QGS}lQq;G4Sm_5=o(KFsY4r7X) zIv)~=%OR2aMq%FS`b60Cy5$=!4}H&&Q66WXlVJ~u2c@@=Fg@&9R9#m{l6DzB?fDpg zlJ@my#Jyq^MOV#8cTG~wu02U$9q7uN1R3?QRoDR1D%yUT zXw{|w?tGYG6HtPnHkBAkwbVw$EmDGBg3F=1CeqGfu7<=YUrQ6*kwB2)m6IGKBBg5F zrUFVP^nEY=8^jAm?%AW5x&k}h6Q(45U-)z%MC?)Eu&75Uny!!$C~o6uhkga`RTQ@; zj^(i08{crn)}VZo7H+hh+V3Od2KRM`M~MoY$6}6+gUum5>vm(-BRkYsK~xni6_yPj zKM2D=!&0FBQY7?L)YTpF8eeELCMJaP@_m{P%Hcm1nvKjsBBi2Df36`z{3p?1x z3caMad7XlVdCGKT0moLA1E{C1B0vLh)0Z9<)@ji>HR62^Kk*uCgR<%5PjKHl(un97 zunw^?y1uVdZA|3*3pR32++;XLuEFp79qf>K8B1bVCozS9o*8QMaVOo=8o1gFD$|k` zo!A)xI>2F17*=Sj9amPrd+_RbPmY5+qY_;rs6%HYq-c$Nj4&FV7e9O`*ab-G#({%?b30+x2V!L!7Qn(Ycs1P_QoDQ1OSJ_$oCD2SZly zUJ3?!PHw`qDhq;B3>Tlto>MjD2a%MpCU_|4`{~iEwr-y%2xX00TZJW`^-nhW}PwL z0(wj((=e2OLY_>fVkl2XQruaE_O_O5c`s6f2YwkVWSboMnKe@T=|%$CBZNBzkjp0| zh2WmOU)8+<_rU-47Tf>JgTtpcytT%6$w#K^3O|t^+~B`>?xL*?@A@LA?}4~_!=002 ziF`C{=!_SwAWrtWdb59P!VgP!d5Uh=-cNtN<^&MMfOSC3omH$UD5NrN%V<^$EFn=?69@G zK+Zm5mFBsG#($-7#bEK3n7GyVH6I>5wj0Rd`K`VC&s4qk1LUJ|?S>jj)x-W9RgWCD zelceVG1$;9fxPRft-@=wy~xlPp9A|3Fw$$FW-v0bUf6U%Gv~l))!-KF?lxGT5b$u1My2y=MTZy{aay0F4Ek-yh@d4%ofwOHC{k}`$~-QZ07Hhts7 z*I3LcptE3F?i9cX-N6BThw#i|i&qqFp0mk5TYtLR( z7=0pKXo@adI4$gnFGI#J_rO>ka-=~hN6IA>%gpKea-p+{y#jjw71zRtqDGDBo+r*NLO--5a+-?4RoVP)yfUVsVN?HOh62VPXGXVXMjlZV^>?4~>k zV1ez@_R)%*{*Nm%?cm(A!gV2gS?H*fOH5~yV4^k7XVDvJ)aA+ z)Zy}|t1SxFFJ16%aJq0`>AseN_`nZ1*=C`-B8wyE&4av8%1rtxc~S-5TY(|%N<4$> z^Hc=8Kp|!gbp?@uAJ*hQ8-Q5oY|sjRj0-hQ!R2w?cfM)AD@Ppt=ZzEU5?j{fKr+6P zTSrLE^e=!hwGhC&b2!?{(y7aFBhLVrV8L+9U=qh>oj%{3@mLYaCFg?|2{UBuHn5Gp z_D1+`YLL&~s`(Qd7dVSlE`;*&a>jj|xp88)%u9N>$PGzx@lidStv>h|)h{qzktsT< z{FH9Wi#@;6PME*T4ps99B)Q}cJY=t>`VK2W6-1sk0}r9Z1Ozv)P_%?O71Hj* zGk6`qsHWe=!z1XreB?K@{sp*v_1h;yFwmY>mM&f4H|e#%g7q^N4(pvx-50s9 zq;P$p8BX@;$Z(MqC-dC*S{19RN*h*Lo_{w{@FmiTKNK3-h<`#4lw(g50LUlo&2%rq zNEOur^z$FCLS(a}>*z!S0;Z@w*^?WZCnO%r1d- zFhElD0hgTtRN)m6z^8>9O{PGQpB9dWr~B98OMyf3U^7*yc}ajSv`GC-6vALXOyT+H zIX^f$iZuS^EV3(@OeOj)nTpWS!Mn3J^}*Ar=UEyLK3G3#p{{-t-|*tc93y~GvnP9+ z8+tPIW=~}456;)KkqmZ256s{Zm_hL0Uqu4(;Qy!rj)Ah6qB|Tebu=$g^PVPrkSd-* zGnj84&bvME4ljE$jp)~88mS;GnR+y~G1$m)H{;c!VgsfUel6{E0X-VL9w|Y=zbJupC{4sLT5nl(be@Xd zX3*P1?D75EhW zAOKYt?Fn>?7eTH~L*3TfoAkBr4ZVWxRF&vySimxX{xLV-x=z>@;Q0G+EE6!!&6 zAP#SgN>R;sI(#nM7sldTU=De>gN|md!xxcicS)--f1@cIJjEkmBCkMv@d+STHZWlI zPu-NhY{-)r)sQ6MGVcA4z-2EXNI!+1TMig)!KU#-et^lCxcq?@;5fNnV|*i3-vAAz zgp}Y)J<^~e@xfr3K7#K{0wTq&eN$%P*#2X+mQ!7l^KJ}ii_v7?a-E6zv=cR5+G)`M zFX`PeN6toSXU^R*ps~Ebzcc_#0&iU1#oI0$J@m-&2(41=Ai9x|RJBh+KB2Nhb_!V= z1#`#mDOJ=p*m1oW3rVEE!DX)0yY3lnhugG+V|Y-LDCJ%3`s+!gJ!TbBe8P<=vQUh2 zlMJ^CDb0Y{$On&&?TA#=Rl+gY!uYr@pnos^$US*y~> zzE6*UtydF;N2VKiSA><5hIX2SFe#w zS!E8eq2BbVHwaw^XS3_sqA@L;6rw=$B@xyEY-iEn(6A7-lwM*M@((mq4sga zePV=>XX>A{A>7PG{GM=yFosA0M(w0ZsN9Yl&MuxCZ;Gg){St#yS!()g4`QMNi;TOZ z@qGc);KbPh+oG^qdJ4lt;~vU}MB|>bBCiKVY&q2D;ww)r$gI8LvLCZfC|%ph>)e_Y zA=dfE8Hk%s;@Q%uG}Jrb&ktcmv~gYH2wyyxSvO>)E~}yQ)bQa#G6ICU<1a-#=9Ju@ z!G6{YO4V2C&XbGVciS)jD29Kn#VC?b1FvVeVC2{gaJRezHXRJzVbmVpCT&(T76&%C zmW$7jlfZgCn^3$&a;~>v7*En2yf#-Wy(QEdW67ta{y$=ksvS?RgRO-adu;;OHsI)F8tP%7B ztZQSMuEx6@2A8!=wT>8>03~KDVrPP6tym?lHH*2I;42o-m`k6Mui#j_h!e+2nl87Q z2J;iz4l-b|6VgR%F%=wuHxp1!%7U;YEYy|s3wh(K4JHLGA7;>cO>_wfs@UzMpYi;# z#VI8CP!Vb0FkE3NT0pBBbK+gZlt{Bmi#W5+q^D(8zIbO@`&?;T^!fCTGkTT}3u!+` z-V~HPld9ihLmnRAnH~LJZ?1)Cs`H_oQe&WO`a194UoK3QV23@)?Wh63YKJL z2X!6zIn3E$yEC4X8{-yZD`y|arjI#uj_v{k_BPN(?I-!V}8 zq;C*Mz|Et4-D?cQp%Ht&sK`~cv;DO;=(|0a(&RM6vVY}9+?Z9dg|Hx9Y)qk{Rr_OB zDIub;?1&MO0P)3cRW_`{N)BoXhoC42w6vCp*qhl#X{?%Ca@!@X`*WBApdJ`ceI0uu zUp(C4RCfn_odxv@CA(9#CEc*o0CV#^gq8fsK9DY~weFD3s150mouBGtNwAbGu8FtY za7qHjw<0-o+O1e`3sPkAy44&=f2Dop+b)DI+mfAeDSbh3#vd}&7+*^hzOUY4{Xtw) z7a2GuiP$iarHbpOiL9_Q5H9d47&eQM!6(G@kWF8O&dlHm8C8u^f4*j<_2nP_9Gh>E z{b|>;9y*uLY({5lZRTyKuohpLa-b%$)DjQLOnDfEb#B_EOjLc>WlR|}DP~es&PH|#I zLG=0eJHSVPplsZQfPy!Abim0f`4RXS;u^3Sd#ML~6^{ z%mW*SyL4=yonr2|54ym9>J;zP3D)i~8+9aWrMJ#AD>M!6?r?l+;DmAlRK8xk<MuzXmY9|W$W1)oQpd1G0g zwxuM{?9LKDbk@}NjA!S*P>gx#R&DPTCO&IwK_(=Ra^x1q+P;DS-s}>OetIgSVZ^vs3 z&=Gmdz3LtXY^TR#CHn5x}<$?irzQ2{g3slA#TKIzLVFNQS4qjNV3&B1t}eNd;aIIm-}Ac>Z5dq0;tpB2g5i+L)ED+zHBjV{kB9)eKhTHf99sSv*P#;(Q~ zGz9b=>+t36kgcaRB!y6w#`5{K0;^({4@(u?5)OB7jKxBiv#gF&zQm3D~n@TC#n?&8~dFM6d$*7x)y(%{| zSr{3-PRvC5eU(r<;0^tA{cZ>ud2aL-mn$&eDHswDb5BbdnHaIDFv5E&w>fc^Lc2R9 zapO?aAzLY|r`GphBRQ!X_&LIF^%VT@im_J{g0MTAR=>U!Tb{mhkeriTIES2jw4hL^ zW5QH*K|d+oh-S3vEa{bMG}*NycI<tVsk&zTAE zmp14Ns>>-f~(1VIpeF~J`hj5Vl(fA_5Tf(8;{9fym{s(MWIw%Sxe)wzsPmw#hR znEja$t}#1(G5Cl5?!-t=@}5HI=3=MMxD~}jd4&O^^j;> ztY?IyQqS9uZBAaNC7h6q8?%bsG96fsq#B0?<-Ad0fw{gD;RScxHF{2d_(*NeRBga` zpM8*iCk2~4RPti?6r3&PlZ=)^#CC`uZ(AhvgZ(SYdX*ki!=a5A3B)N)Cw7@u%C^&Yres`~MpNuc?4CxnS2>k(d0NXGuTa}>O)@lE ze!rx|i|n$?&qy9a7}8g1J5HrTXWbvO_sy`IA$wAy&>Saqpod?WIYj6(?im)BYENvH z75h5o$q`S|bK~@VTkeR?V)B%kWquDy^}tez}2iYT{Vqa@~PL z;K+tMc^(>A!-$h`kw*dF5IIqc`<`J(3rnQjCB`c}H=X+W`=BWiyp;U099I%`TTwLI zY3-4|DH;LFdcV3oGZ(Mjwl2%b+;u(-UJ8%R;R|x@V)hSV$)x5ze7VB;Jyn6KEfg}= zprCg;(^xppW^(&#_q`mp>a#sv5x*}_q%J;o<-x-fx7cf^*d55jDIFF+R_MJ}ODVeA z{-(AqSvx0tv2({Y)tB8~qh}!!>dt(>gO5j|^@|{L`u!gsK@X|Zgalnm&#@-?T`;E) zaBy$W&Ul=6ZkSEfCpGLh9hy?5%rl2Nkm2LyKgF*L-9Mta$JG`;oHby_Xbh}Bc-xj` zpWxvW>Nl3K%}g^d_2V|>x(wAFw5dn}TX+-R2Vul7)4s1}=4j9~5%X&pjSkq}@78he z?smz3#_;hO4Q$d_$19fMNu777;}xEE2W$Md4k&{y-sQDEudkq*d(qPrin)1N_bwa7hXc3{`z&<*OZsV&b9o9Is4^NCAQi{5ZQ z-3ML0W8P!bYv;Nc-a3BRK6vMY9z^By1^rc1i43H7udqn{bQV1qzD*us7&+@TU(4=Z zZ!btZS(*#KpkEhb)3gdd3tm?mS2rQeX|Bt$T#<^gxzIGnC;*N&aO%IcpxG9KQTT9)iC7@rpwy@+zG<;uT4XQcO}5EKrn6peO?~hGyHkeC?gppJopXYT@O=|J7i#1P~6rN_#uT3cXNzc zfGsAfOQ%F;nxs95MDXFA->l7QzEB`nOV+OKcF;;GcBQ}2Vdrwi!`4TkniDyt;}_R}do)`@CJ;~CQkUBn5MS@X5*TPz$LOSZdVJrM;Z3iRk7 zIghOSxjrr!8B+HawHdhLS=lW2n;1DR1ZrzmwEU4bif3;mu@X!!$e6 zx)mB5Jso-bXsFX@QX!7gV=vjvCslg1>1i~2`bGf5;cuSx)R9i~&VWv&CZoEAouPz+ zF+}Zis~ozY$=-426@PN^G`%BqIYhgztA(Y0_K^iK-rB(CQL=Ax$Q-g;*7Xz`nlhUB z*~Wk1max0O8v+aijt)gll$7QFsRG>xLk+Rui%eDN6>n$v`e94>_FwJa=wnW)5*gheSRIJT*)v$i3RFK-RNVez zTW=$>x3$}lDW!UGt(TP;Ay92}s1v$=-R4|Hf?#)h@jj-XO((K5dj;*RmhvV06FQyZ zf{KKkX^;Q=FLcH)ePRr^KMrtrZ?7<$(^Q9hOKpGdr*rP>B!{tviJmm)l%uIr+K%ao ziu^QIpX~7@^sIu*Z40fQw;?xwD`!$G{Vui@4oKmJ6<~GJG{n+4Q}(~xrVepjAF@8L zo`GjH9;Qxb1n&6<7OV7qV801^&f~c~!Twv;jZ2Z%$+(>QXWGR#?Rm{4&+wp}aYns# z;d?5(!J-fwrjNkjRb&r=;R{aufP$d?|f^NapyZP5Ha0N+UZ z#)Z1h^H529@wfU*$7~MfPh#k<3!2+IApjZ*sW7>R!Ja5SQw)scnjPK^R8Odd=0cxho-sk*3HwxvX;4f|G;l z!b1V&b8=>O{mvZv@>8;FnTp-AYeAI}^7YUaD(#vHpgHq17E=&};rH2Oyg9PY2|YhP z%Z@TC=ra_Y#!!N9@a5N+2O%2xD*tkl;;_L143*FQ^Sx((4?}*hx6|0AM~x~oLSy2F z8tyY@z`dBi!~13&o9d~cPt=8*Sq^fyH~BDynbeDpdjtJLI+T^Ws&d9QyU&$WDq*Ht zg=;QQm~|+mo9VWvRp?oM3wat#{I|C~r(lWcb3JsAz2K=I&lI{pCkq>UtbA|NrD;Ahow4Asrx*x0K zvTsm^cwmpdxS_=KQI@|&Y}(aX%hAWyH^R{isd*@>_RhL}Lv{Vubii3%-{`OMLz5%5 z(6yaVOWGeVT~c_%cZk{b3LZbioY~BokWa}|;@Y$G=C>*Y$YrO7Jqn4Y75+DQPDmo@ zCf$+#Ymqc)s)i~7%x<~`vyAtl^N{pHwI-zFlDWPt8al=9uJpv!ON6Q}vDNuhs z3`eJ4sf226Oo~I^DAu|So|^PEA&z6s<5|X$hx*owPg<*+H^msiH2ER|LArg-q-DqK6 znWZRQ`m4V)AaU-67mDVDrkvrq0N%fO(vCgowzO*0CS)bm$-B@Ftrn zg3+`-E!IMYYd4(LGr2USn`)4Ma4}tj!-7>kN`lEct$N|<96_|wI|U!V@`3Qhoe)NF zT;)0t%FB`|lxdLZZcXu}xsJ9=`H{#tL~g?yZPT>x%uytz3O}1)SgIvI1Gk98?zryX zV(oO z7gsZHus<18zeXH0JKX!{ducYL50&5iT=orMWXoRPV+RdJGkpoofN?6yTS}k4uCyoV z6ly9;Z)Y-}_|Uhqby}c4TFY;VPN1GzOAkVRLHRomdKR(hHIm%Og0pU*#)s4SSmHIC zU$Q!@yyP8Q^m3_TPO*Pe=k>^^$NmmH#qY$-ITOs7PT_+((e1MF1DWf}H#p5}tg~d? z*qeQS*`;;zalE|{gWouL0LdhXyW;u>N?^o(%|QO}gyO z`9gV2#_Mges*O?TDpRSH_|XOvtEh1Ihs){$5k1dz5T{GdalY_jVxLpAXh@AqqVCw( zm*tXCUbNcIw?%J5kz&%+19ki@P7Aws&+@iz6k?ShkKs$XdW z#75M=ghStcxnT1nuOwlp@(@p$A4Ph-arHrYqs3Mc!w=VYoRyJc0q-F9C$6e7I2C#% z+=y{Hy<#HL?G(qnqR?j74m|}ndU(|xen zTxlbr!!-g?suN_C;J|f{@t2`QblaN1AT1Iel1cr5Gd>CBdfzri%w3x!YQK&-ZLj&j zK-ao)Jn|tcG)Z#vX|B2Pv^%vNgSfPM+hxGb?kaE~sZi`~PEFPEcKPRPi9r;52Uc{J zoFX2Ce#GZYOIdUolP@N}ufpr)D%VnLziebRvPJpLvK@plQ@ZZ98JXtZ5axRj5Cs8I ztgqMR+d{9-7G+ph6ojy3lV)Xr4aLp|By%P6Cqt5@la-V2Bx}zhOPyL^9eaj31y;7Y zHYYIUHvWzOJqEyxvZl8iEu6w=I)jB*rE2j1}mhSMPWxwl{lc2ZCl z2Ri!I=vf5TrFs@Ir94YBtkHhB`&|A6dHwH0y^)`BEqhyvbK|0F+~}=bUg%hlK=tj- z1zb`VXY}Rp+l}EjnbM=NkFf!4a4oa95&S{%!H@*)GeMma;T3fXEhem3T>-gr#*axC zwGsr~d9K`S40uRm?7W^*trlw0#Im#W)wQeN^>I+@Vs)$76|bAu0_U)Sm+Zfehcpg_ zVCSHZgIWrb5yr1^)|V|x3UT+HXx~i5(Y{|$T4mw!BWf}hao=`K8+VyAG7)asQ}J6k zaqm%Rigpo>}8>EmW!GBKDl%T2&E_R4FtgECJUkqo@N#Ds?V& z?`<-7dk(u+p?A7ksV$*Oh6s8g)hH5Q_ z*dQ8G)>&}l}MVtgj`IdP0W6RIGplYWwAy;+Q3 zUyUJ-tgrBO?R#gg0KfcO?)Sw+IGC+3jCSeE=g3$m&@0flnaifk3`0~c^TjBucEf93 zmq&B|VYf8q(vupg=<@=DpB7FLWlSL5kvMIRwCWdY1kD$#&5vDz=3N=~<&x0G_t$X< zW^A0~vZTP^&zgm7ww#FWZapnJGJ0iTg#`Q2urcK#q=$Mw-(_&nFoDktNCGk|tuD@V@OD!3xN zSDqMuUTNw8`%Jcq?A;VIk)gn)6xZd=N@aYRFc$}2Nq?|v6tlAyS3DS^bcDEyi}iF+!$Z`EP$jBozvV{@nhyuR~4q5qRwF$<*B-RcP4q<;p;eZ z?+5Is#YGEd^^OOEXOicg`ht3>Cjb)2{=%#KIgjlna*aqMIZUW90zBucnH=SHU$(eO9KHdog1tCdWnJiE6lg;m+{;U>4RQh?RsQFWX+||jA!ZON1enl6{bJ~B%kAq#HE3gY;6;p*=FjU^>N{{ zsDqh>ZcHH_`=(uvZm)9NL&iiIPJb2P5OehszrW6klN{Q zjHoN|_o+Wspib?&xzcLlNyha?gLyiu zENHnCm1=?d@D#MKUg*tH2kX9q?Z*Ja#2pfb@(qFK`~gdDP1ae`m$Jq}V=Wp9B}&E0K@SozOf}Ld zDe{+IZcJx+A4>1%|9XJy=CipBo!jVF8fnEaG{Z&N!)`tHN_9qVjHq&_md9=T&2?kw zuu6H*l#lXhS*vX3lWN^@9`8SwUA9iCUMTB_TM*xYy}f>NNOGj>YY`14-aYSq3fHG} zfI*~tg4V3L(76321%#H5La~l~D~*lYW(!CKO9TaUwLxj!yEwnRpj1n7#wIz+)uZ4Y zFmaOQKKi!p9soC@&n9i{z8LmQQD2w?W=yE`FaPxV0&s$gLjen$?&LNU1D(G77ppjh zx`5?gYk^f6+MF<=zw(Ny=8erpU?Y9Hc)*n~H#Ni>CH3ysj&in^2IC+%v*X)wP5mIq zS9;S+X}FU_qa2E{alg~1qB(u1Bl+K`4RNW`hAU$bD{!rMZz1d2t0hrmx`N@YE3=4g z7Y7fGP{_Xa>hG8_pQx|5E|@1N|H@86JMXG0`j;L)zSa1gI+fw)==ax=;||PVxL;L1 zp%4`Iq~{MQrkczJsrnb%je!hg9K$;WaXms7n+v`HqCwsl=n}bc>3F1#4&^4bp3#k@ zrTyDg!H!d@@bTQ>sS8$j=~O5p&g1yKZVy$BWl<}gyOMB`QplxY#^~}Dm{Dl4F0)Mg zT=wY<#y^+ye|c(epz*GKcMG8FwUXPzMs+r!j$AY6$rFP#sNFu0-`oDv>@@OrO@KVL za|Dqol5{y-bx0H$$&yWw-Ci>?eX@S5?Ro$gy#}K?$H_Y@0)XhZPNQz`+1#Mlap8TY z#Q+)xlQF6=$$i@#HISA;%ksyw6N;@@ zzqImAWbEV{r{3q3esFP(N+5zCBX2Y3zLNdPh@V>h;XCikx7+qQozRL`j3g;HD$?lZ zCfx@aev?R8k-XFCVNrmHSymRwYH-lW;rpX8Lhj8z&<5~>wU4xKeE@RJnWX1Jb zr`hY>f~`f)6=d^{YCpEWvZLH`!Y4t5BHE@6&pNHvp+Cgp==ZY5u@9q%=|VtDdDq2%gfiJIq_)62C0|yG5y4SIP8{mODA*~hSID4Q{rwGY#x=CT_43o6mDS|KK&%TTC4nd za=(L^$Q;^QRbIKh8xf1|RtS$ZU;3jO|Gube<*^Gt8{0d?4)H+`?f#s-TFKjW5@ zA|3@_^}FS`up?0I&FDN$^C#>rdwn#GDra1wxvtP$xau7Na#xqx{pH1b7e3*WbIC~T_EBZ??5p59_`%wS_~^XT>Tk3_ej=RYH#xhNYx;{Qv*K2h}2OfV#8QR`1j z|J!o!#dqpdH{|LVW4T@W#y_2A4S{Kf@cxkDx-Ts2?wnSbv-sN;VWcQ0IRJ@^)S&Mx}j!Q&|~ByN*?skCU$dEu4u%q z*YW2OGA?2nw?;G`yKt~|9j;l*n_D7TREJjiSDIqSZ|8KLcXhbUsqru@;#D#+Cui|4 z{E55VAYV20IaYttID%L$E+Zj9$}>1 zLQvfN;eP8buBXIM5Ivat%%l^pn=_$V#n~Crz@(pNT1q^M58VhK8VwH(CtS^@J$?sI z4Uv@B)y9=DX66a#W>Va@JTwZd!ckG5m8fO*-RJw7A=wbcz0shOYab zR}KjJQa*jXaWSQU`U|_!E#&Rg0PfRs;!4Ir_0jt+gVv^o>ou3mB?9g?#CbdjQ zl9gRHXT875_dmht1b?w$M9Y|`HXs*j?aiPI;L=p}{voVjk|&O8K=ejT@#)Ckr=inV zRLYu5?9@&qDYzwyj*jC}G$!36lsOR8#i^fbdS)Dnmwc{@i|{exZjHI`8v_=%$Cx;c(;|%_ulrId;P03I~=}irCPFtRBJ_N1TR*J09$t0s`uf> zVDuidge%hLbJWX7tE2BCV|rwxpdrx3G7zQR{4A9# zgz|DqZ9HYFgw|NZ+!r2w#9v7iVsZI2LkKbviCMYvK>dbEE1yx*F1;pW^o+)#OK-x( zZ>;koH|R4g!*AKtMrK{0**TEgF7HP?u0&pyd%6~dIPab~V$wP{yirXp!ARh_D<<{g z;~M32>IlZ4mtU66Hs0Yx;5X-)n@&Y+LRMZ~_+{r(=J{-S6YhaELQ1$qZ&FhmOSwk% zV;@`8U|z@%>rG1k@TWYZ*Uux^{D4WFf@4d@T9*GBxoTm17o}uDyoTzR8*68zepR_x zd)!{05UA$RqBWV9l0)>gtb7;v5p`!!6!96)1hQ!peL4zL36g)v;U6~U)3U3S+svMh z?pqlJRMd$1q=ra_rs?(Dg|_9yvZ*%%l}{t`ON(iTAieNeR)oTPFEj_Ca{T9J`KVi9 z%3?p0#`M{gK7JMCoM~yc`775_25A}a(p|ew&b4*WniW#X)Qea3x$@3URphj2m+gn! zt%?dQE0>vDHg`F@6}3Q$<}j2Q*I@Wst*Mq)O*HB^Q<`#6{#DnvbK6XvS8Tnz`=9E# zTzU40+CDkF1%1uDlPZ(?3e)Z-wqFQ0)#$XlPWe8{uI08SkM2rEDqd5=Da+kRO)^}C z2Hd=C{ya6#QOxE?zHeI9?beAKaY5s!vwLMfGoXI{UcDljdF7sbxR;5@ZlGhu9whpn zJU75VihucW6#gjTsCe{~`QZAB>_lM1gUtnKJuM-(zpzqd8(XaCsuc<;YY=svxMcNo z0Y=@SD5&-_fZ^9qNCmxfTEg2Nadw_q4eRiz$K6_tLac(7JxeHg0lChDN3XvlSw-@^xvRgUhY>D?m$rxsGLoEBAzJ8ZG?0Y1wn~3H{hs@r6N|%6POs$wOOeLOG7Azd7e~n}~e4l-~NC-+|Mo zF@3x`ZT%l_MCBy7dtupYtl4;O%1PAmZ>RJi!#|Dr#DsC7KijKxmgfx|sY(Bppr4mi zi{{6_^FpG>sQ(g$fV9lHXEE64vBTZ{B|lz57es0?E>v^g%8ITFmz=G(B(M>t1xz=q zEtlW!CCL85YM((COEDz#-Ms5(*Mg6miJuD+lAVeV3wxU=J4CJAJF*!8SBkp%ymf@2 z&!BoaT#co@L6gykO|WuvWGABE!Z`+Z<7s8Lg4GbSzrqi#eEMW`R0RGvU3h_c{wEjm zXOCD)vR%6MomsrF}2_lKSUN4pk*imIJXjXNhmcUs26^ zaKD=Vk3XkDY93(N(NEcXOCB_0U-|)~$Q&Rbxsv+Z-`=EG3FptGw#u(+9+cQpaH&jB zI&V7a>-+k7NRG<-4TR`umd;T3MSATX+2z`vC?I_5|BJEr4r+4mx`vO43Mx%Ff=I7M zX^OPa1f)wbQRz(-kSZjU00BW1rAslifOMk>(oJXzLg)xYkPcD;p@ShnfP5F8b3gMw z_sskG{*j5}XeRlUz1LoQ?RBZtD`*qNHO;AmGP0;L#z?lr+PhJ!V8>?uVdEDG%zE^( zw#hhSrZR~O;h%-jKT0xl4S9*5^I;L5rf{6#zK-*!(R}7kBiW_xiu#yFXVu>KkZj|= zm(CifPwq2{!BXl#kpQ1e1G~J!7hjLj`SuKkV)ld^fj>UCmP1(1%fljjyiIONj`Xo$ z(;RhOh!NaQ7Cy9Q2&DwyFXvUJq-pw+%9r{6{khL+Rm)2T%`U}S^v4#t{NKCd;7m%g z7hoFZ;=R5%N#`*|`2p%;zD5s>&)RHcWgCePvO+s&7?IvtY)(Ri>+*z{F`b27#O_83 z_JLH1rNC9k0ZdH70Ds{jn#eC+!}g#+mW#K7FTqInJ8Nemei@C=Of?+T8_X?@Imstn zbG3JYu$^c<$5lMy{HT=(3wM->z%Axt1wLhZL{Mqm119HX(Hg zRk+1w>NwQw*0>b82(d4&dpu47J|x6L`9|nce&QkBt1o^DH>fJFKCnyd1c8x~2`g}M z;Q&Ovwde!IyLwU-yGL9rTcGeKlv_5!8G|D5Fcp4{%XHCRhLYT~T`OI*ntadX_npkS zvxgBAs~JLL8lPrQS?O9|syUb>)I(0qUDimudq$|+VB5FAp|ZDtp>P81d&{+#+vod+ zHHhv}@l#HMRBXy+&%@#{cHu2w2&1RPx@)PW^+4zpH*Vk zQ&nu)H8hjQRY)pER=kvZV!wy<@f{{=uGT{Nv{+Y0Kx+Pr)s%|9H>2&_eOvS4lS&g= zxQ8Vr!{7wsh;XmIqL(?I(J_w6-Lv$0g28Q_HId%9OwMNnBel&AXC-g?pf^O=FQZnCGK`SD+D2*Yzx4yJw4Ee zGzpzFJgCcZ%5oQ9ZsA<#PsRnxub|R*IhSkrWqVtK0s>t*k~`;q2%Pu1sPSf$h=1sx z-+`obIF@99CCP<^A>F63Ku}QHN0R!m1|J`O2RAcq@?>6wdkEjrXHAeHg!ibo`GLu- zsaR+f>*O851QkD$jOD6U+p`iExKpmyz_H=FLf$W%!FVEqaT*EI7B7DJC31f(IWdaK6;PjC^VmXoSk${qg>8mX~4F zUq61lc~H$_^D|YrE5+@*7D`l$ClSAnX!hQ>BsKj`D{5JRw@4#c@ab6qawlCS)`wz^ zT~IeRmLT-c8N-cEXR&7;}$JZE?i4@mjeyeIlyQ(yYnO}NoHm3E5ed~fOb5}zNC z9oKm`+{BH$%j)IwQhBq>a97ab07qE9S%GJ|Wd+&3L`8ayng{J-)U;RsFrFzsx*@&$ zin4wh6b>`=YnW9VQ_%jHwBE1F<&`-GXb_MAUZiP^uxqN^E28k7{bB;KT-i~pcKEq0 zl)if5SM@SXmO|W0HkBbNXq4)YUr}vI@%JR$5Sy*Y zj?pJ~i?F^iXRF8#f)&(G-{`cuZ9?vG-nCO-rsPyP6UN_MSoc5BuLQkx%FKxPf-Y(W_q!^(@P@+XEtPgseIL3sHWc0HUK` zGuROIYv^hSVk#@#tP65NttRAFd?!Mpef@xE?ZuYZ!|5|PW(7F>JxPur@LOIeiw!$M zO*62@KvlJcdVRb8rYF^0%GlUOs%osnEXKrNyj3iLlBD8UJf?%^4HrXof0tXNOCb(} zs=Tt2R$b^e|<4a3{HotisB zjM;Va*eCoUaKT`RtLh|07z!^A@{J-tDodD(gha3^HL~;+ifX;U_VQt+c?+K>3=n@u zoQ3o@EG3t~Td~+L*OlG|J6A3&(MrXGu(Au=>b#yW z!|mWh7aB2Inr`%dLIaklLaTzn;XD`ZCu~)3CFUCeUIm?pz`b1p?yblDf3enoFxS!h zl;!`xGW`te^J_pa@wJePuknY)tq00BAq??#*T_=@xe5f3jg00Hi*=eqS2bDa^BCHV z_O2b=YH~r0sbNNoCeY~|YB{5W&JcK9&<)bJz@79qDb}yKot(!&+tU}VDDF99twH+u zi*(|r9u6d;JAKrfT#5<;VjF*e`)SgnI(Sh|&ccPyaEi4}#gp4vOXonESLt&Z6zi4b ziJ^AI3k7>5pY!#+4`J(`$%W)kPC+XaK&Vpt6Es)QN7jnq%9d^9KPJsf&At2s31)=0#E zCF7&y{C_57zrQS&`LlnXnU3HP^EKG>C8wVkm$3EqL%n>cazf@jCAi(iISHeBV=k&H5;YNj1693o`4COOW>n`6asQ zMtyhJm_Xo9M&GM0Yuy6wAe99cy)F{=Mve*8lQKRm-5aVI^_auv$)PhKM@b%S$ck?4 zo^ymM=t&ZJlw{(u2$J`1noP(2@m+>R?-ChjzE?OO`BE#chf3w~18^P36+w4h)K$7C z06*m>RSR2GQC|}zf(5%MD7SSK6+_5f>uYcI-xX;;O{-N##@wsrws-gP>;;Fs9~TscW{XSlmlh;yB*X2#=6+Arc(wcel8 z={#5z(h8Us!>H>c?=Q6hH$WB=p~$ee-g5BJcOWeVIIWl~g`TKDThf`|S9`mP@P2dV z0H&|_2mqS7W0Zxl_&76S{K;YCYT9M;IH8evnb$w$Js+-cb4;0tEk)O|C0o`o+ao-2EE{SQm&r|p`#CvY1ruq3fh4El z%Gupa>PixoH@h>a=uk-ejYsiVYr5vb)Aa#6Cw4(ToEvZYi1N_TL!`y6Z8sn(tSK*% z>LK`t10hH8VryV{;|IY8QmTi9{d0z11-gZZm0(|_E z9J<}S8Ff~%$I>rM%A0blZ(~UvO7s}7c;<6){6G=K()ry6>e!q(+>;J7C4oWUhQ13v zkkprP2>v!vH<9snDYNF{mnxH@n{{<|&3oifXp)-xVy2!z3ahX~tEjp3 z$uxl)S=wU9c7nCdz>S}Z!z^@x@y2XDd9d;D)$D(^KY3G0k^B!gm?Y2kUo^`R?ec$8 zExb3+2ZJl?gtbwrXWRs{#YcC4?uY>)H>9*P`wFtRO}CaTvit>n!RE4)mnh%91o>QXeTmE+$kh<#Z+r0?kIG zvbIdLz_;%Z!;`dohJMa995&@YD9-r}@+0lQy5l45*^cE|kOFdKaSL<_| zR+ker73%r(844crEE2wY3w0qDx&4~kLI5&vW@n>eQcZi^wSA_>a&wISE~8=D$#N&D zLAh54AttVbdWovp-E=C?YoUnFpdAc>#qAUD-KUF`3;cjMj8R$30^!2@FbP8lrZizf(xnLcv48C4U5g|vsaC{Vja6-tWT7#Wk^{x7h_c@#>1z=_00@a<*R;#49u+_ zfN;xw`X0j$?^J?mW7%;K_F5^XPpaf3j$;WlJ+?iOAMg)jDpy6YeB~xb(hQ@B|6>DN~fk7Fyn%?+Y zLi5m_EdNFob=WIo6kNx;IQn9?WYMf{Qwpkmu0*a&#l3gMtVeYGAS)x6I$G+340Bgk z<_tx52IVspmFyy35g!;4vHZ!o(WCh?BM1fz6LsI}kz+2Ew!&wtbYrN!9)r9}TjjUK(z57GS( zdpZ*KppW+W{EEqZqO?6Nc!^uhu9|k8{BA!WSk-;}0slgWAAB3=U=`)7C0l;zaiJD^ zX5lwfXbIfZ6G6bQ7cVtoGsRhvC=dn1@#_k*Cy%}O{x;*@_~qVaPAY#^ppk3Ys;+ZB z!)vz6Pa&?fz;*!j0omaszJP(Dk9Uo(^@nXW?Haj?BP+EYv#A!2V^hxCe>JzEEwAY5B&nITAm1@keG=~k0B!tL) zy@_RUs2zaPvi3q;mx&BMqeal-qMBP}@?#PiUdD68GFWYu>+;39%~nip;leL%XFXdY ziw_dWB)cX7YhcXNibrKKBXZD)+EnSfR{d;!^;>LzIU5DoN7?gH{gZG!RBk8J;Pc}E zbsSWrGSrTJfG$iL_xe!-G)QgEFutQ%l=zHEnMotU#;SE!{avIM&vtyGC1rEAZBOPa z;u)5`i#^fZOXt7wKHz7_yQ%;TSpvM-4N;N&PwF|xqohBT-ec3(0FXo1j6ZB&Scye~jH zD*XE#1N6jqucX#iMr9e&9LJau7IoAk6jWM%ooh zUM=6+8GMWYEBd`HTBz>puI}6CQplx#{g*@|Nre~yd{dERLP7bX@d=8G!d3ur8UG42 zk<{|Lxp>?z1qYJ{{+UjZ>VzyGExOO%aaJDZ!8VvH%<0zr2~o$bt@qmZg1}w-F>%1$90~=GY=v7KLK?ai=Ew{WS|n&*u5(`1zb|}E!zYfLNo z^|S79C9wUUWX$$UMOB>-JtFG+^7n0J5bIl9h3z7M3jAZjXXf$s)g+^V=Y7m3<3i*G zj+|@YK1r)p7Vx!l`KuI$0=BGt!;3b+<~%As74hNT!KeA%BQ?=2WH2e!ns>#KsbW6B zb6b$=v&%94uBFK(@1uFh{*v{CB+9W?tS5zfRVvJIo;_RGQLDSYeQBysF1PRwMv6W!1>PRu0n-6gR=xcOJf}WPz@`mN(@Gv+UZyN60eWWXRuKL ziqGHosn~)`lAWrD=(^ f zxFkpAe=X_S@V0-iV$K7C+OD+?rNpHzWW8ZR0XQKm;T&P_YG9M@>GCo15m3+9dr2HN zX$`L|D`&oK`_SwP7ercuS1E z8c-Iiol(mUgpzEBD>G$A{)WKAsMb+iPG15uJudpie*SdOS@`f-pU>zjc4ZG6>_g&* zz*{mX7U@gEzP`5t^hRd_l5TXT{U{x@@U#3cPM_K5jXwECA!yP}=MolnxU6=>G> z3aGLr*UD#tsw{_1xLOL>YpD&{x%6g~;G>+^H%%%8(cPtwDm=~ZaTCF$9Ozt$E;o5U za)AkPwpYu!K4YGUpWEyP#o>Q^TT;YIqmx5k-49(D@%4<_qBvp{3bT@NHmiYmT<>bx zA+jC;pD0;e)8hij!p$}sU~xRA&9{=NJV`r*eWS`YK1DlII&R?V${9>ifta6;#g*X?z-bv**5>4FNTxmI z#R$*I85;Uo~->5PfAS&qd(ScHNRJI6c|@JPvyCY6Ea z1EE}!L%Jkah4RkHTL(xI!Mfg{e)fcqa(2p)v8oJN0aH;OCu3I)FdxFAum#Uycna`b zNBzXX+iH}1D=bYHMgek%^x{t!?cE@$to+Hqg=f-Ejncjb@-`)W)t==nZoE3RW+XDE zn{Xj^X0wNo1flOfzZBy)Y~8OKd*g47=vC!B!P2KC)^$bF#x|u#SZ=yLvJYq-NsyoQ z8OOaEbvm!3#n%XJ9hPm$#QBi6G0yo5?m90Xk&GdTv%ISzHmC^e9;3c)k}nASWL!Y| zg^L((|FwHV=sw61j;9=ixMCeDoeLH1LefSg94bxn7v^=_j{c5P*rVt1yvL;uWp}Gy zl)O&(+T(&1GAXZTuzV6E<8GnIlS%IvH z(brpI1_nV0ZVFwf320t#BJ3_Zoa#f9SxxrPc&GQ{Px1=a?J@CC)02+#Zi`tuB* zf}*~00jX8<#TBMJCUl?J^S<)5&r=A{gyZ0e`Dq=Ci_h;PcS;F(NU1TKj+(sh4cn$4 zQbAac_8ym~?!=wUw7oU!Ih8p?zyuSF$F*^%jylAlv0RkDnhegh-2`f3nGrH1G!;+t zIC6kUB!mUef7s)X*!qoNPW|bM+E*FJr?gPbumbisUo4tzZuOBr`X(W{np|p(H$-%Sd3Lak~A*9;j^Za69}rCTisARk=P8a z9Wu-kuVL2_{&bU4iO)ZOChtIqJPFd2kT&oR&7y>0P7!*SZ0f|DsZAl{qO~Ch1VM5l zphjN17W0}oq14h~CrRERm>og>K^~n95l3&^6*SJB^vPO$S#N>J2Evi+MaU%t1t`=v zwc~)Xc3(jIP1I+02kv>=-%E>T5{|&q&mbqSP@Aa6JMjVpbL5^iRkF$z-FNNlAqb~XFM^`I z?v%aY50q)F(T~S`p(vBS+smw%)%mM6qNXmOP8BhXurcJm?=>bT65F@2=ssf;w#rlv z9qOYf`>8D$`4B5o+yd*fi6dsz@?Xqh(7YXSX}cIz2RQ20Kkox%b<~g*4j9pGc}DT* zSeZe*jK-BSfVeVh+v?P)Xu=?O;8!v!hgzNtf=qV#(v{Jtg(QuLjdG^cvfdIl*CbY1 z6)^YAk3+5mvTW|yHoK_M4u?_fkcfkZXhewS!q9mEsABi-RK3i=o-3|`sT&jI zH1K<>JWF*;h2B2SX#5&k7kij#eNHUMDnyL6V?#xn+-YqGS=D_JTk;(K0SAk!a=ic6 z=_kS8<|6d<&OT&tllxh|k0pPnizVwgRYAZKuU0*!s1 zogRWD%y)F7t@olgGC#Gnn1%2an=#Z}hjWPfYc7}fp!}eV11--8g^g{u&-d*$KXe@n zDK?Yb+m$i1J(W|nl)q5CT}ZG%EYksC;zBlHh{Km?gfEge;EL^uiC{#~T0?r6j!J6Z zE7`to+xRJ^Bi@|p54^d4Fc+)Zjy{PKqc0e~R>W|@rtBzvD$^^MBqFg^QKoHpNK(U& zdbA8TZ%P*Wry+?};XC`xXY|{J*o1zK;YwYmW<}JGw|8q5?~Mcd0))+l*GXfb(JKdal^3MbkUgHyiVb48k z$iK#Yq~^JW93l?AWlc{{=@8|`{201YwD1XD#KAj1k-B$_q_PbIVfm*y2gBlv6crec zt=O6_Oxfb6R;s3Mec81AtbE+)%GONyUy$n9 zjG`qQ!xA=t{Ji2e%MOJn&(zx$%Z|7~;gPkL^4(_;^Lw9=<~8reEg`Fh#Ck~#wa{F) z)sL;)Z(dk%NFkp=k=@>^FF3Pcd4yF*_;L5H*U|OB%vE@H(*!b!zm5ovKS%^cHXLr4 zJ~h^gU=>4#WsTY;;w*pHx;sB<-=P%4dPOpZ;QFj@+h|-RKFEi}i}8v*k~OjEjGxB3@DNLvVS7&)&MrnNj=W6uWbD;=H-(4zG6)%2o;*qNzG?| zdo&jv-*KRbrwO+pwzm58W;PnwzT)nv)&(Jqg2il<0X0Ix-(P0+$HBcA>X?F=g_&9X z%#>x@(iBgvWfbML@$*R{*;9HnG>N+@E~3UOwR`E%BJDu&8o7;d3NmTLR+9A=ppzk} zO{JRo|MUZI{2@tSPh^SkDr`XKviFy|=46SFGWyhk)w9f5tQ3ah$Dvl`(&EHK`xYYL zTLkooD8}OgzJWlvXaS!_s*=n11BR^o|8_6sF0?wL7Z%>#{4whX(&RWI#0P}~$lC-* zh?Q%qk+`(~I~hwUK2&bT(R9+Fr00h@?4m9FgY|?ud1!aH7Rv*uXvDP@KC2sz3sw0u zRY+~_2Ng1#pc8N&QB%EFplc@RSKOoDy-RSDTad<-_tXME%rr_gK0_>K`#Z3|t#Hdf|0 z+5pMgc{v+l6uqBax{!eCV#Tj9g4L#Qwq3EHpY$2=sn!+f8-V- zk3^z3ogsF58z(mG2$rfQN8GCdCS#T)im;X%619JY6o~PIxH7z8dve6~2F+cC*G+H0 zgDfQAH{ZmU*hgF|$*?TXGiBNA_v>*R7Wb_uT!9_n^Em6!W35V?9|%-PuO9l2mOB`M zJ%Zvy6(#ll6EBT}peNoqfqId8+jFlm;)N4wVCB))4v%F@1|;6VTdmYh{QYvjF$TTl z48F~qgmWYq(iN@EtihiSxpY~ zj9Q<-9%XtDPli`RxCgp*W5jP9HU!&@y_=*{a$|JZ6=6(d;R2U`I|0X+VbSLH2FA~O zigDRTE##6MC`M<2$`36fUbazSdGCOGJ@kdRsev#W>Yd9jYLE^5GTQF<_$&jm%x~gq zhHfK3@GhTPVx+28=Z$Xv-T?7z-=+Tv`u|GN{%_4v{n1&i>Ejq+cwl(Rxk?YH4;zj9 z$n&cpT=q9fHybd3ju~^uB(x**Xc+o0O`&g8$nS< zD@TozWoF)0zJ$H^1aTka{U=vy2D}rbOejV+vKfQ*L4DT4{F;eoU30uX`Y$3pZPg9D z`HWA$=-XIO1=0D{@k399`OuDr+B1Xik-(Va6^{Vwy?YLen6x;~2-r$zsULOLMp2MV ze)?91pHdezyO0cmOl!v-YJj4%+SD`8jmu4B`ktnH?U2lPN9Rv#6$x$5?g0qT(=vz3 z9dFM(<*_X&rH!2DnhH=Q%gIAplqN^0iGJ*gQ2NS_MHE&`V0!}NyMlYM+ze<@EYXSz z^X~v<0s$R7_<_nFQNaRGBK;@-OA7qoST*0HT%b{oT*84ikSg{K%yOqqa^%Qq@+jQm z&FWt2pMFW~-kLm$riOHI&U~>Gnt(h3t3GbaFv)fUVM&7imgE!nhByS3!1BD-&F-fW zFX=5AhBC3_zk5+Djrwf#KNa_FNPfe0o;IKFL2ZWs?hI@MXzfc0-ouq%09&~^X4bC8 z${`pW=;Z1yesi`uz91_Tr(MFAVp${eU1YeXZ@;TyTn;5|T&Ck)7uxN1)GE2IJ#<+U zN`EUJlk-}Xk`2V8agH_WIK-^C&^KsuYafpMS)YI9!WOKIf;OGZ&Syxnm8zPhW>L>o z=XDOTL%I%V4#nO;hrR}tuXxCVs&`HGoRj-jySmnod@ss9^w44#N7g?N=Xy zz;^OAembKm_}R^`)|)4&w10FWx24q5_LlbdKU>?D9DU(*W@9$_f26reW{1Msmu5Z( z?i41Xlz_p`Vvb{I-J*pBf79Pv0H*;HP;WTa2J#Wd!&BgRoD=DZ`$WQ*re#hWYDxOu z0f(S?(;dg=d&9=8FM5!rMye9N?o1UDSbz(yghAj*!R}EvCIx@D{Jr$+CA3Ywep2Jc zzApVVKet)?z?FEUR&bsiqvaE{wq%zk6<|vDJgI(ByPqwaf)(F9U%Y2K`0ekQo=Eyv zi~%O#Q=ut8VJ`tK-mKSmnG8+apL_@-mRm^or7Kwy6v1TtzG69FLt*(Dc%`r&BA7J0 zzi)ym(->U&&I+JU(QMeod6uSO&LdGYFZLfb-TyTc@TqY@fvTmjQ5dc%p6>`<-un92Hv!5cro6t5=Ql#=E4hO| zQkqjufN*v@s>ejygl$SS_qm1aFw@|z`IQw&-5RI+c_mN{hgR zXMSz|;bq_X!TJmd%A?bH-6{C~7UCK(foDF|IcuSMh2#A0;b#Q(LX z42?J7g#y}>Wx8WcTAOvnK`OMyBC(TtuSR=~%Br*OG)ow+nlMD0vN+(KiyH1|NhOT# zNC~iFc18Uhf zOcqic>Ui9CxM#POxb+k=Wc=Zg%+=Z(T(O?70E@r4eV-1Qbdg8U4(GAw)TfWP+W`vG z^3)%qkEOtwtH+peA@yK#y=6AA3F`$u4c5kQ=PA%D%zhtk`px(tdPWPgGKCn*T~Pe9 zY87Ym&<)<4u1IzOed8&vcXmbYxNx>uH2DKs2)M z1Poo0<)YuP;Af-rcrOaqcmz#Cdkt;s0qyEp4T>IM(-Taep^?5`lqmHh2uhe2(>N9Wyy9j*^@(;@gxHZG`u zJ{v`LvlL`y|MX9|GAQFF)T{e>j~Tlk&0%x4_ZDCjcGO=E9)0|uJfQW5o#4L0&q`ld zl98L&eW6MQphj=lPgh_7e0~YARHJaa0`poWY2yLbQjCEc+B7@Q@rw3;Wc z45mr)J&1Y}|JDYoJXDC8D+8j`XC;lAtD2^b+1+OPD|eeeyi;Kj)NGRKHL8~;Be&fE z%Ib04!G14+9b)#@;e1MD33WFUIi>0l*LeA;*Y+MXm;gw2UXs@zwmSk1$}lN=Vt&Bn zT1q#($SR3^t+&c7YiS!qX9urTajtZ9oUOLYZ%DfD5p)X}GBbV7_3YE8B5q(@Sld2L zV4`h7zdET~g;)z=G3G^$6g?jm^S2tf(#zuP__Um(jEDB7eRF3W^ZbBXvz!_gW!?RE3(2bcR{*GU`finPC>- zurmH*=N55y*Cq_+bN^d#3AaIFzSQiDNAsbEB6S`hoIVKt|4Rf!VpDumrQ`Wu0c`h{ z6>1(hUpSQw(JiPCxfQGO>A9sGg0Dd(B$}`UoisR*sCsZJY0UXGKqmE#L!{fjMHGj8 zBwNgP`T;ZXaFXvZJd(f%+?uwnExxrZz?5l#uxSpS;CYSH2odAh>9dx)OcDHX(|pMv zf_j$&l+F){9D?k0hCxK1Qd~psN5j~PVwn*EieTzZlmzZCZ$2Bj4FC6nW=2_0|Kb>b z2PjPtN>{(8jY{bu)yS_2T7Vnhks@2|+?(k!l}jz$U#Q-KrOrXa##}2`F#fZbiR+AV zUFEZffV7{9v~4o~^8yI0W1}|Y|KaE}iR%Om2yDW3mipZ#vX$!XNi&aHO@-I1ls`|i zY{iA}(fSFY1q&ha+lb*Tp-RJ#Ddb&lxv#*lTT8>8O&PcHz9bP}q&4;dQa8C~9J?YP zHP#w@5lm+5P<{(2poV!hU8oQF6;Bqwx^nv2Rb#$TcTxxj@O>d3mco>7d(v)4*$Zj3 z9(Q|tH;MR0c;qFh>8IB%qvcH{eI`-;{^zJ(EEDx#T1xIuz*u|k>HTa160)Y>*B<%i zbAFsW++@%FsFi+rJ1Xws`~Gf`nRC~2Mw!24;|mvlT(1@$R>ZDJG`mEcHSYKbub)1% z9R2KU2wLc`%|1d>`Ag}BQho7ajV#pES^TOi2>jhz+;?~xZ=iJ&t~MfzLUqom0@ru< zQHe?Ns1h^n)zA98qzSg+I;SsnDw9BA1mM`--m7oe1-=v*KC~N@hzs>ifD+Y*hG}+bvuQx>;rDk4))QhaD&`;LlHy!= zLUFkp4WSvxBN;7Va@W2a-0YrCLS|oypF;|~-LJv!uMzT@6YbT$RmFPxQ4H7!R^%u1 zPlz2bzSPbgw#PVwb8eyEQ4Q zm_@OU+=WJndZEBlqxU%4I%oWPW^{Ar2p|llP|Pp#;;fpTUY~c$)aI+<%FEg87f>I0 zN>jdx=ICptMxT(wznccM2UO6g?2w}M36FB~aVd0t{rR~T8inv4q`uY9+u z5^*LC1?XMRQHKXnvPV)k?IM|`2k@d=O<-mdO9MVV;u@nZTOsmZ6w)RIk>Jg!sj6aP zXr8(nW7x0;$i3bc<(mtq$zHJ)zBVQZQZrPHKhGpICqEbmIBN7Lhg;Q&F;e> zpCV!zo~R0-=ug^Zv1%pSV3?qvPdn_gMH5LTiN;nrs&}SuMgsxb;+f( z3LjKBv&$lU`4a;@NJiwXD77oYYXsJaIsd0a1wGx>+H-2e+3>i$x>No=y;qoTN{Kmc zO|asE5F{w3jMo?iX$BPfIdd@2=Ktvb{{B-Lpl|VHm+n9g!g8x#s0CFSiuLv2ldho& zX%D@JLIIUt>2?3AcCI|>Y1wPIA&Gg zKUHWeznk^LwISuS1y&d_Zq8fP30fgp`cw8)?@blmd_czsI98)`88Y?_OD5H;a^n2#o&t=6xj6KR+Ce%D(W%{ojdI~=3F>|-qUei#-^_b zi1_=J^mmi|-97=SUbpa2afUVG{zN%T9`mE&>JOvPkgPeto)s_&d%0{rM~HgJ6UU~Zr};AYwM3Ed#&TF?4=IG|#y=^$ zHh&B>7!u+Le?%ch{ei=+%rE|;#R;Ob3QZOIT)x1$ z@QTWMGcYJB4(e}$PVUw%SOCl2Z4;u1i`;07jJq7{kSV>%J}6jswY0*ZgeF67xePXn zWr}#P)z#$k^sM2FX=V_e0u2q$CZ

LhX#%!VwVpzTJ5+=1Ku`2l;RRV;zqEJzVfW z!FAYoj$;_Bq5zh~n>wuFNRh{zWn;A2J1{OuNAHqvb1|mOIq&A+4b!SYCLzJru8Jq= zsSMBLc%O3QoTdwa3xh!|SS|l-=X}N0%v1DX44L~HtD(W9ewFIyX$E{X%QBXp$8)mn zh5*MW=hc13kYHn}wT%0#)>2hws@-=s1L*8ffZ4jXz@iBhUMO@{X@jTRL!em4bnP}E zpy2k&aJRb0B|=dKYyXu$ZS7Qu{i-<~~t^F4O!n5O+q*M8mGu}J9>mM=1zk4)N}!gYq)<__id%V)0!TVx4b z?4`8<%mBX$n^f9Rf#lF)xi-TB6M2g^Xava8TqFAY4B1>1MO(H)AN_j34n#O(w(C>A z>(BQ63mFz!d8W^7!arNwUKflPeWCMdN(_V>0p@EXt~6x*j3#qB1%oyF80U}eBF;Np z*HnsF^h62$+Lp6jy6?l^ux#=dFzdmi&~(z@xGycz@nddDAVLEFWa`YZq$HR=!&2<- z?xc@C!@4ecztMor=grge4rjr;AGEnVejO5ZC!S_|jpIC&b53arz(s-p$8zr$>&dDt zR|}ue9(_7a`Ub8@&j2{r&rB<0T_La|+r&&gz&oKW`3!X>fWdD)#+z*GIakgJ3e$li zLIbiU+o5)%Rk>5V*IG3&83pE74#l(+7Q#93zK|O+2fImxIk!~6?E)y2e#B?b>rPNS z(8srAkIB(hX19NDG*XrG9kOopais*l6P>%7_vChzC%kTr$Es(obXGnK))(i2O*;rM zBjtd4=AODPq_qXWb8v=HyBl_xK9x+sdb@5%Vi#U=?icb_6rSBuRb})Z*o@KMMi!-D z?48LEovGjpk<{LKi3o6t>P>7k7{6c8t6K7X0IRlp%Q zX;#PwNbOde_HDO!ubS}%>GNghvh#!$4?y7RGZn8qAHyGBL_VM2Y$BZ5oS7p{UwQ3q zp0Rkc19yv9<+Z^R-HzPJv>zAxswb1M{=)<4dUp?$L`PKZ2Imj(%&aS)ZhXIgsY>!= zdiLU-c5h;}2|%hwpCpn%008^AX)!ns@X@_TeDvi%=&d)EvYvc?FRXMlSgnj zrKKpsE;m`DJnz7@up07qqA2)^(2kb42)K|fAP?Fd2BH%k>)~u&6lzBGm>6w{B5+^3 zmM%xQD@$t!JU4BZ1!U=#6*An{n5P+VgDRI>8Zzt27dTPWu+J?=vyN=^f@7NI;l@UA zo;>J}>Y1AlFMeT;<3yqKd0(yYjNbOo;0WUZ3o)$?*R9_!0cHZe><7?($J%ePF*nYy zT`Uc=XD)TSzrZ+_CA56w(hJzLznG?F;sC#{XRvrrFwax{(ojXOh4b)RKyxB~ES_1t zRKL&Sg#e$%8xJ2eu@K5-Q(5xL^1TfAG4`Ih-u#AiBlU^E&U?c;43Va1zr1mFW)Hg% zcVar}CH{VoFD`~R+i;#a>`d$KJy)y0T(4sJ@lnKr<7_q&_jsDx6(f!$BKbo&NDjlZS50V>NbecTLhdTk#jZs@vy?2khvajfVfH}8+s(0C&@OVQ;@$%UGFz>9N^ z3Ywr%od%-jZ}2GtlO+_ilncG7?y<5DHSQ@MQ*{9*QLG?AL0lLcx6qB2UF-q)GquNq z{u_h+IYRCkfi-2kU*D<~Y_5SmfkDKZZw+nKZPPd35kMdILtV|#{3+v}M3L3SrQ`UR z8*{LVmjdh#%5?d8y@@$KqhpTEzPPus^Ey%%!X@e1N+zpGU`h3XXX_t8gV`&!M7Alw z@*cIjJCB){v23*uHP7DJ7kmof&NQgUW`YIB;B(k*B@QgQs08w6eazpTk81&OTpBU>UWiQ9*ra%BMX@8iI@lnJ zeBUstw+-oon)dAtI>65QvNW1K?qPKawv6hDrzMTnwEL>DGq|&-ii&h+hS-q?F#Lq^ z)_15&KzL5>EnEFJJYWA4-DCMmqA9p6Lwa2u)WZ3tQQ0ybl|W@{5J=Y2y&O&-VGKMA zC$O6B(#`=#&R^hgwi|#=(rs)kE1TW

rZn(3&xu^$)dMH=W;mlPDnsg=aM9jp8Bz{3R&^A3>5v@w z*qN3k(d6|XV}}ImbC(64=$KuilZ3-}Jzc+)*;_?1E3$Nap9h!($CZq$i^eU~F3#OX z^}r_4*$`sUf~AgPm?@xJ;5ptpT^ICVwN*(Lc;ZCE2A|Ja9ikCQ9}uG?HtKEgZSk=R z32^8ud^%B%{>(hXgA%GN2U0;3b5|ZP_3~>g7WXK_UHFOCr5JnBHF13So4GN^FLl%A zDgrHT0_IKLZ`en+VEj1M4L%v&`EIng55J~3{j4HwH4R*8I}!CTrp)2`EV%U5UmU!n z3sWs+Q~RO}cH zB11!?fDtsAOFr3sV6;$jODgIEN&$udhwCZm^=T(MmbrXBg=Jab(e=OFR2A{UoZmL{ zzVWx)8}YEABxp0Z9jl2~9XyE(VamReDPG#q3+H?hRCn1zcwo zXnsb?aJikxj#g1epF<^6+3K9)Hq-P{R|-IMb(#EKj1!Fx4eLr@f(3cHVBY+LC*zLA z=gA7L9V2qv&@(zq{fKv;fxX^(hn2N#eKSL1)e|;t3u+@0a<_aFs2f>l1@FzDDt+3k zKjCP*R1c*XjEnmdt?fe|moR0&>QZfb2X_*AqPVuDG{o3mK7a1yocV@D)5x#gF%@YQ zTs`14E8?NIE3vc<^j`drdN0ra1Wtr3BK&(TyHI$>QC5i@$F?7zTdc8X1h3%I)=&5} z%o-P6pgR{~!eIDKV|fMCq#L0JIO4D~eUS}Uq{r-Rsvcu%5NSaAONaB*v+ww*+I=OS zS)gYgREd$dujBY~qG6|J$?DoC6!nSpKKp4w)|&^HT9$+ZK69=yGCj^zw53nm30JX8 z7k4Lr=fI~7=_p(ax1mRJ#d~woL&^cOmkIo-c@jYA!1eMBzw=((9)L!uw)$Z%&iVnC zPOxW`y>Rz=>*;#&d|d?|u#Gp*bH^JA5lcKzr|Sgw{XJFv6i?O}#~quIR&=gdwDZyU z2q>6r;DbUjtyrTcv)h1RgI+kGdvq48NK8|%dyU11m#F@K?Ok~|)N9{Amco%Wg$%MK zhm0lL*a-=vOhsalP>E-rWUFUtC zdU~(xoWI^He;B{(cU?dC-1q(a-ke?Om3+9ZUzUJ9)-3A5AJ5d(tkeB{nomGPEn z;*6t!QN?{3X53~9DI!vrCn^Y{eTvxSGV*p+e!H>8J$)7FK1#7r|uI-l+B_)nV(KKC|xo5X6L1W0`LfpU`Xs-lCQDXSQsT^ z^X&@H3^>prgivt`s70WM-r)Tv%_D^pYkN40zrV41e!9ad9V^p~TvJ022`g9JI4@Uz zpNiU$;(f|?F5QDXvUvsZIj;tSQFdj|%~)kXxh9fOLFs>NMXr$5 zK!w1sm$$6#N0muX4T=mWA?h!n-(J1ANQb2(Q*YEd^OW16quWJZb^*6RK2okmSz3` znPYqs*hds_6ir##T8n;cb`}qMOBRXd>jyy&~tIT_?X>n3nb9$guyk4s?X2 zsYN3hzeJXs&fO|d&7Gaq;|YRp9=>H!is(`Hh6o+P(v<`P`cs!Sd6FC9YxU2+H&@RRyqXewa9QC^N+V3=_&39_ZHAQ?F zbH)=KL6r`3ox?8-Zs{%g3TT3$JD?SoH;GU(?P`jCcE}E&OM3<5%p@SYC&LiB%uBGF z^L|Kh6L`*g7_ELPmGM0!z3m8p#LX#hR->-!h)nJ*-O;O~OV5QSa;MxIoPH$4snmJ3 z-ytqMEMh5oT~y}R%wCx5J8B$l)I2quJ$cg3gf_G);|*k?HepqH)TrO0-f@V;{R+za z`!Vro`w<5aNPR69wt%6ShIldVQ7WG`0w_-Uu;>af^mF%Lro~`L7N(n3Ht=_`l(5P( zPQ)+=`SIDeMkKMuSHX9peMNV$ON?8mR9OvEAA(nA~@M z35Pv!flgWrtu=DtS^{Kzx*gu`dfzd?Ju_>7J)fYeQ^n8gl#3F#RUHm%SLUHKqcar2 z+1}FIN}r3;cjA1!!r}0-A)gX(uo9x(?u(^Us9KU_!E+7^|J=OY0P{JoxO5rn0uIWO zpyqH6;S-T-_}(ytQ`9}zbKa24zbDJl)M?_0QN)M6M}SVMbAu@Foq74FECu#e_lVB&XGhq>Z;+^NrD7dpD=u zF-=}$UTTSH%Ths{f=}^V_j~W;I|}IYUH_uu(g~9P2yqHW+pJ=={@k%%kG%7WokdT?g^$*Eb}q!0U~nI#VN`?dvi zuCT2VW!W%v>SYh_Oc9s0jmE=?iq5RaAKqGvZ5$Th)z3ja}@7Q)U zFh`jc>_zAqyczEYzpxyB}e zC3%`vh3cuwo-%J9S1z+DBGE@G3ls%o-do6*B)#rn&Y9UNq%Uc1MJ;6bSKOP^kRTq0 zUF3)p^=c2uAk8Pej5InQYu`{=kK&-ujhN;$`^aEttz%YXgVA@oYup?a@1jfS+UJ%7 zAsIkXr|~;T^eaAY->ncuIyWLwN_R0%xFz(Rc1y8HUknWuD2Qy zm;&;<)cY^v#v~j|4+`6_fIJ;NA%NwPE0LaXYnb3^G+MaR&2sLNM9r`X^~nr0+UZaY z0L4+yh7e}dc1QfNu7?Y9o8DZRE#U3Mo*fa^jtQ>G@eV5ARY+`Leu!T&s9sTqZF_-!NJhHQBcLP@p#{Yk?BHr zY&cQL*Xb(CsUlL=g(r&DnqE*}sKRP7?wW>sDfR!vg>vk}YBxLXX19HqHuKY1QPC#%nddR$1EE8y4iMU(>FreV8Iw6$ zrgj~puDMiPU*Ozs+zuS$61t23h4c1wSvWhe8fbswlLSJNzD9;3NU2d$_CwN7`< z+XjuDQpe!xqH9&Xs%Gd76%h~ktzH#T>bCo?@!6kQMReV9&_R?cE_3!`Lt@P)tul=g zts}h#w%vdIasJ4OAx3#fx^RoznAN`JhPckzkIdGWXA7G>@Yevq-~q*CV@QSe!Zy@G zim;peYb#DgAOG+eM(ZH2YA1*7Sss?ef`#ZK3i#^92JZR5(_|PSwUgP}VWU)CND5!t zg$rNiuM4>}k9n1(1`R$)%Z(lvX4mJ2z;t%~UmeCzfZePtbuG$wH=81+0Lkx?eLS%v zRMjM#`TE@_!>;kLOp+leNp*GX9(O8^Q~~o%X7tEi51^w4sHpfI4s6!_BV@)pJz5EJ zZm7Lnnqb2ka6_zw4IxSK9q`?$BE>^>@;^T|!Du=7BPomGtgkZjWxK z&jaQ53snXI0dQ%i_uX|L=r@NA^%)of6>|j7OlJEe9q?~VX#%*tfP23yZmPhC^s1YS# z7$nsG)CLZ4e^T_!5jw$`iKKY6CnO+0$$U));2NCY3E~+R@dA4VTxy?N9PDyW<(02r zUP64~uHm3@X-!Y^$4;wspbdVSK5#Z+y}h@s447TqIa@8Od6+B9o?|3-BSd9wk;{fv zyir#s2-`pTL16U4Z?Yk*YR9!srn%^vFMrfV zEC29Yno6r4=m>dpy@k^KTU2J<2!ah1_C69}sgu=#W!pb|B*{t$uWfWI+DMM!6_Q*e z$cj8W!z~0ux_{hQRm%cg(>f)5tF=lZeOLceI?mM{dIqW?hd`A+kj(rjPG%k#s*gmw zO(<8XG&xW2Wa@G@c{g}qnT3A9*SH68SmrPv#pNn5r6&oay-J1yC{_ZAMj6fxp@3FB)=N7S&*!K+D5TftsCp%U*K0ElR zrvNoXEi_0Rqy=6N<-4Z1ps(PK7n;E_Y%-=R#$|g(5oR(f=Rx8cN8&d*;P%)ufoJg5 zLGLjYprG9_{C?M|M&{1!OruXNz8>cDB(mYTEe<)5*=1H5=fO&yGzcvt#sPhGh8 z?jt~q5G39~lHH_2Rb&<|WCU$+*Zg#XjeIm@N* zmfR`r`~&I%Dz&m+_Wkbk4q4vTLOe`G$6ZQUQ&}-hp1vUIsvrCryqiDaIb)`Ku{#ks zs4NyaX!%Z%FRV`oagl3X5LKq&ggxUkxmvqtMzVHu0V1crhSt3R=Ls216RDAx|Ci;P|%^QA7~}~>b4buWhF;H z>}2!^G%?bnV8_@q7dI!t-QuE(${6+}^^UHAohtpa+2LmHp-T9Bd!A@Q(M@GZlEFag z2&&RN{0DwLhF|T?(5HmCYM(!f=d}q4yb*5SttR5F4UldISnRq0%cnS$sOOS);?(^saxeiBvhI zW@S&7ft3g>)vNl2xU}EJO==~@qA*58ZlIARXL`;Ws9*QiZ`M|lydV(=>>l?ekS7(Q zMp5g*{)Q_xW$@)1)F|tIjepg_`fR|DVrMOhwr^ImWnCVXH`QPAl`ctv$u+qj!%-_C zqtr{jPk94@{ImO#DkbFE+;Tdiow@#gQ5GgNl!~J!qoOut+F51}P#&n%X;@2o7&nVz zT37Pqg`8^(+k>yP0_~+=qlpi>?DXrki#*!tkm6(-V637+wUFT!NSn|oZ?$KSs83t}_&(f4)&3RJd3KK8CH6hZ~ z09eMF(CI%BD+&9d-cEv6pB}2L=?b5ABBt`?Du=;a6zjzda#)QQkh*}KLjZ*l@l$&s zN%|(bLyYo({U{-LQkGi5j)V!9ms|n<2e}$5ppEJ3O@)Sca(> zD3iqnn1P_U-zy2M0g{0Xu^SSh;AfXrEX665DDQ=f#bp9+#>%F%()XshG8*Xf712&r zxz|5kasH6tv$kWbmDnk!y=cduGk+3TE8c%ETz>fOi4FXeWh{Xmb9JEe6@YGr8r)8g zzBvwmjuNi0;0uE$x^S{|oo&QK-#Yy+&h;!qKC3CmvrI<#A>6tCYRNXA@t0&}DY8Wr zzBzYx=L|XR0wMcY9$?!gw#+}uKBBEpOAn5JoGw6^i+gD1yCq=#ZlfP9*=)E!>aO3s z1nU0mz22xuw{z1frU^o#T?k@&3_{q}(8VAH_&Sd2KO53GpDoW*$A=VL(nJN-9$Ql9 zkLX|XC)E`B@UT=um&iF(|0F&uL74Deh>}C*iIRv#*X+qyOX71Ln4N3*WbAJHS%_4dE4(#ZssaF4h&;u0paxJG)4AIei&eEIqF^x~SUjfXjS$!IsFaK9Xbn-Olol%9->dKBfM z%D#SCn0NO-@5YJVq_iw4I}7OdY_<7T%?3)ADB9 z7nA^t+Ob_>RXT;V9Bd!j^>TyZoP`P5@#~itO{V|zuBc)6OzfCD04zbP0{dD#KvGd* zOYh#nuSb`Ei#=ZsHgP!}B5x*n^oMAs)lUP}BOtEq{CmX5Lw|D(7Bx&Y2q4HtzM^~U zdjrs{i?IHepHytPk*YPSJ4(Hpi;UT_or8={?z47Z zh=VT<`n7=hAB&1aF@$&?yk_=M1TuE(##~jeDMItyuFP6nF9|WczC28Q+$$}iYlwU6 z>)BeBsMIlyKy7*!d^(3q#mh8p@7?c(7HY)8PEpli4xg77K4gH?KLH9VK#T5~TB~@x z28TU?$v)nzYC-OSWACfdlVrJ6J8qvT2gRBb&;Wrcq5&2B;(yH zbCm^%(%r2}wd0AZ`G>?jep(uF^%g8i0!CXj1qS))u|D^1EpEecpdxWjYs?djR24Qh{TOshO#<`biJmu;-_QEP#5p12(Njl@rnG zz82xs-IW4RG_0O45Vz&>M{9)tiPnI30j#f&=awqJu%-IScI+#V=<}nx^)CgIT)vIg zz5y)uTNS5RxaI5mpCI&W0QM`m^7Er!T<9A^eIe=bl?33IGbEqCU|ddA;a>XCw|A-k zeRuS^RiZ8z*Qmq)iCVvgc)tR1_xnopW_IBlvf{UK{WpN(=kLgOX;;1czvM;#Z*c4L zqrOkAe@iugpIYCi)?Z`V_o($fYW-!o{Mt_Y`-1p|1jO9;1@Zqo)|vzGeXCZga6hHw zj`9_|1X5P-Dk`hv0Bzdi_V6&$aQ_%r(bcsiXFsLKr&Mlk26BY(sMG<;g8h_abH~EN zi~E_Q$(i+Y;4y!IxbMHe_rUia_}&BmuX_Nu5xAG}iH%>si6Gh1^WzWT$50P`rdY=| GE4C_3evREv+~KeVtI5R0KqZ)VYd)3NpwL1p^{enTL?XYE=+5RT*SV zQ2`kvL775`tqdg$QIH`Fi6#bQj0{Q0{OHB`?obQ~|_3QfAzfRY(BBDIceeZkk z>$>)}Z{v^J9$ftQRezs1Z{FfV-|ssPKPAqaH^1|53*a-G5~(!!>66go2lvb?XtG@f>*mo-(9yw$TRq3s??)#$o z6CJ%vKSetX?q2Dzad%?s=bxs2zv8QJf4=$oS8aGUEyA(j?c4JbQmDvGMi+}^`J98+ z8+?Xb{78BtR)T^KPxZh5{qG9=?+X0y3jFU1{C|H1o_zbadFPio5%>Soqdq0q5?9ST zb*9s6*`J?%^0rWLZ^673E<)$e|NQJ$`H1YnC-V+CS}yY0peSalngpbt>s8|^i?cUREHG-BjbHS8L=-@!j(+r6kwfK&)kMgqSVB3;Iw_cRQ z5H3fm^b85Pk*ZjFm3$>yS!75UOhL(-SN@{4a0TXIn+z3kCv>Q`Y_LB%ERvjI>^C9z ztJa?=-`3}J&N0<1bI6z`5e>!;)nX>X=bv58Os#bsbRHTuGt#s0k)DJv$Qx>T_%>tM z4|5|tE6eEVZ=t#5B(tvQn6{s%rZ*hAb=eT*`st|yFT}?_n|G}!M(@)%RF}p!8Pj|X zN;PiYk{LNwWHz1RWBAqKTlb%z9d$XyZsPkNZTp+B^2tM4gQe;|Rk@G}ub81Z*g^=A z*FM|vA`>f?L`H}?+HPI@C!V@ELe0mYyF}2&@RipiDmmN9iN3@^nF@0O%2`s7VpF0 zEQS_b8ESbsWMhkxRBVoH)Y9nJ(4HNB8vMqX*l^6W=(=FnXSRnhH!xY45=<@D?O~0nz+W@V z!=-|P58s@+G0Vt^G<1X-)z%UVuLMgu9o_aQ0bUXo_++^<*0&Z3jvZVZ2{U=H#XJ(0 z&%=$lX5P>Lw0Hk7k0rp7gLNonsfD@lFV!jxqQ6y2M@~#C(oOFhehp#!swCqRtMJx) zG^5>StIzmcE@G$N!-^g>E?w{fWhuIm65f5zxBPNn3}4dBd+4WaBfcVaO~s}X$_@I` zj3bq|C^y0n4I9B02U4?)287cYsa9D=7c)^l3y4E4r{H0uAR5e>sjn!)A?s-c5@Z2R z>MGW8aZT_o2&p<9pKlgjp*?LFg@fG)%?-|TZM#^UsMNNT$p zMysYxUricL^Yhjny;ba#U*zL{Qd$?9i;He+(PO5H_-00BS<_~rlB;2@JIatx651Xa z!fy;KVtprKQRrLi583xdsQmM>Urs_K<8A(o z`{pv`j@0r|AM1hXIAlwg{JC=S17$Zzs;rs$6}@!x=-oAZO46`eB=T_i8TyQs>&&tK zg&D_2f-kLvmx*48@v~>Yh@5@Qe%ODRvCDEaeJgd;H6?8`*Q`Tn;gg#(bVLRZlmJAJ3Jtw(TcA(v`e zY=9S?MaPNyE$3ZAhnUQ+`aJZb!l@=T)~>cR^B@oRbGhHA z`kMGbb{6_GQv;_*j~nfW+_Od!>Mac$8Zk5aJi6_OHlc0NL5TlmW(uUrBd>j|=lQ2V zb$^pyT3YH%jR?>~*IFd&yLOGTv4ZP4)+r*Ta9D-!BcN)r0^A_M5l>6fb!vdG7;*AE z8o;hG(~5q5vrxB; z{pxnA#}x`0Wk?G@`>N-WuYK6AOzGJ>hgX{hUR&r?6t`hOz~f^jPum*(3$kzwM(Vu` zON!B<{0tWcD@V)g&PJ7gn|ujobh1rXe!ITU*`>nYvUykc2A^9NK9_ZwsY<9gB~W-_ z>nf)*hGFYoLHT<)D}iuW^%GKfh8#Et7vt;=Li%H;TPJA(oMG*BreH=Ys-SCfq+Jy; zs_JnSr>k6LGknL3LD%XSz2l`Uo#H4I92G7jyvwE3yERRkv|=CsZsR?o)v6S?kV9)= zsp#bmc4%>4#W~>XyFr#e#smy8NgV2L$AW1oM#hUj9U2Kg`gmoM_$R9h_ z_AQHb-=WO_|JL;S4YX7>aJo(Y4X^tZ-v_dq-S~qd&aAQisGJ_l%OV_i%8DOpdjE@AvHKXo=8{M=(I(vF{ zSRp;B7=$2oZSF32s7*;keO6M*y~^g@L2y{q&Fqa0fA`^<(TLiZ3h?B6sn#Vl)Eou8xPM=$%qvc`4?K@xpyw_7a5J}98WShBn#LwJHsaapE z6q38ib>C7)OXdgnyt*Rp3J{I4n9j}k6SYn?(oXUXCSFuF(D*}lV(hJNTmx8?9McZF zSgty4NEBA$mmh1mnXteu>eZ#mSst(TQ4WPeN9G092lc)6>f zW744Vf03@S+f@Vyy*269KhD4*fwo><_|-{>qvd)r3crTDM&{hx1hl_9!ee8q;Rt9`OC1vknZyV?Z}}@>chhu653keiN8hOgJjfED4xzS zo7P%|n!YthdUm%mljr@AeXTY$+y7D3(F4bUx@YCq9cz`{Kvbv(*iwI{T5_IY34qPT$N4R{GBMTePHnq{FP?tQg49~pz?3Rb zxSj5+AD2zSe1;UjgmemdOp3pSa|cfLsd=qa862tIY-jA5^bl!IG>MfZf#=CcgqRV zi#Y(uO}@Xi{_)+Ah^Dq3u&k<_n6cAW0Ys|OCPS4ooy6SHyAh6Eq4HG_v;X0W{`t@y zc{PLE@m(9{&|^RotmFY_fQ>V)Bb|l37xs4H+5Y<^KW)D{74LYk?X6B?5}Delg{^a< zOFNxwW213rcAT(!HJ;9qdyi2CwZj>Cy6>J>tRZh4t1nfBj0^@gn=S?L4L_z(Z8I z7U9mPkPbOJ@eLfwiRjO~YYon!F_w+wU>9-q{Ihe<_iqR61yBbBk0wLH1d3zH@YClF zxp~~)Om00hx)zGX11J_x;tB-!F4wXmvWwj2H5<%7az78f3D#2v&~7P`9N%PLsXl7* z+XBcQW2`bS5>6CPj^*o4KHIxQXz+chf}umj4ViP|6=DwEwAsI*15)M_hqO8Rs+?ci zj@RIjYU4HnrM}E?*=WuKIiN9NbX(jZ#$bB!rC#^)ws5s)M@^dj2hT;ac6IB2x0j_? zmRdM~8Fj0ew-{5_;$0GvQXC4QpV+!Rz3SZ`*gBu#r^mWBN7Noa_EE6fgCiBemY;sI z)qgJHyNX?^<4$4*fz$=toimvRj|AC(=<>nR&Vi14ar?eXiEjMS|waj zxBfaQV6&^*W2%29WV>wtWL8V(1SPx#BA?>oZRA94?y3s>WEI0ZwbgChz;y_=Xj3}I zB=p@MIQ93$Zs!JOH-f^Cr(?TF`mLD++dp3d$=L>p#}+XzQQ8ftV_(T8s6nw}DrX#b z*d?zcpLom1-h1Zvb;LRVFd9KoPglnnwRoFnRvle%Bxz2pkgJ`}sc@@ai<>-hYkj^= zl95eBS8BMtW?(cp?$Gbv>BSfS>hd4rKTh(meOG1JWlH~HX8>y}4*$ym7~bO|QI_@v zQtSITN>Sx_c^|h)hg1ZQJzb}6S7%6&G|#@)L7DNHy!b&l11G#&Y10Lhh8KV#P8>q- zKr|LFzQ2*O#>lU&rVx4N6{zmx&PGmZD-dj5F z>3xt!rMYgK%*P*faG@GvVm6;zTXnQc3t9d4OO)dhEy_?9w&9RY<_lTk_AFOP7^TAw zW%;GW+b8U?1*aFgZ~83vw<6x4<__ijhwWH$)F#}o$dX@ADzk|g-0YJbqvv^nQswRU z%fAohe}2UOH1>~Nn5#FZnMjuW{*Nq)B^@2y!#cY~gR$JuV{8vAZgrg+hB{I0S0lg) z64HZetE>)hPd5#^(HBIw{1PR9%id74dp3-63pt@ZploUr36rCjeDtph1|E4pS%0v_ zCu4FbD`YtktBr3~8jUBnqW^u|{v z>Ue|e<8vo62ujK1Wz(AM$S;D9b}^x0X^4!57QQw79cZssh^nnH= zf*n1CMCwYYJmise{NaX%22Z72GRZe2REaGkf9?wGyF^hlta%1cy$6Y6dmeg5sz9Q6 zJw)*;NPIqWNnS?$$Yk6XD^SeiCfL84GA_&spx2Kh#r@N{R2K^o&=-+C2Kg8&=1+@9 zE>WhOBRDwmPUS4mt2tR;_?*&O9X=XEnkOo+pp9z}8sje;CE0EIY@=UShNR*e;klp6 zdcqFvAHzsO_%X*_KFeeSGn;)B4j?w7ek#?fFT z8=^gcZvB2~`_X0#S*woZqen;uv|Ob%GCfAZ^t+PJW>^1i^~G4JhA)tl~s zAg{7e&L_7I5o&kUM)Hv2m|c_+PBA)awz|{ntHqe<7H{%jJzeMLPS>$!SqmxH#18UE1*Tjj=FN-agm&M&Dx-9JzHEW7J4VP z25t<^UUcW0CJ0KRNyCWou)(`C_1oZaPZwlFH;^teSP!&rpMA5kY%ICe{jVObdG8L_ z9XMQzs{ZA0#by^-WVuJMaw3Dzz54CaHWL#^ZQ+}duw6*OuJ5iDJ!Vgr$1+>JO>ZNG zqT$QXp-c;9eMz@R7(x2m9pGYHPqbO^n7jG@4#Yz{W?EcgJ{#Pc{C^hU|6bO5zpgE? zu9w~ZSl3qAc)a_j^a6CD+a~O|A9{7@U`xDBx=H6j%j$A})>kN%gvnC?ov4xX4eflo z?&ICoh9PSTx}c%9%ClRiBz6PO8#k4&&u_zS^fWm3j8mT`^7s3eH8&kh(3L} zcC!Y<){Or(&=QBLqvcVPp74P~%{bKcdjTOFh=Te$Xq?PddE$IyWJz}mjUv}01gtLw-oiPV}n@LFhLbq zVNzZ&@)9#anxY9@?-j)EVp)p&JJJKHj+msflC95RC#rnL=qhC?$R*<1AiK_!NbI~Q z@){bqQtmIwY;#^gAk6K@&9iw%8j4<*2k&2oH{{Itj{w|%D}Ht)Dr1LR_7C?XO)h=p zjXeKA808_^{!OEM??E)-eK4=gZDbO*5l5a4xrnb|aI z2fQ2ILcpGxKX-nMGH?x~D~y%-S}A8w$=6u_kHFr)gxckISv&;E;&uN<(Ex265niIC z(B=Heqz zaQw<#0_^zYfgrkqTSij&3);JZ9SyPxR-GD**gY$sz4$V85a%OLa9ijRLXcjVJM4~a zRy^ij{xt`p^^UjK%p#iq`fk0RQ(qE6eL4RRM?`@F2!1v+``TT!1|zUmGFUm&Zq3sa zL>&~Mnw1qceK_#|&xj;B(jU-mtcj2!ji^SS*fmQ6dmQ==^ojz)|u9NrG;%g+Y&3|x&|7M8&Nh+TG z{H_eNA?a$4iXvrTE?v28(j`TmFi0$RDcfizu3`#n4Ug25-SO8=YbUP(25bK#^(pMHPti>#&Ea#(TYyAP&u9Nd=G(P`&yv^ zGdO7&qKR+4H>~Cf%>Ajend};BNo$&sU@*q*?vIn9dYJ!;O8jed{WJG?*Y{u`Np0!7 zq?Tf7Yd*dMNosV%#3VJ{MYTXvjSzUanR;|)Mh%l`t7B00`c{;zkE4Z1hEjUa@#t9W z_+zKoS8C`u^NGrP$}cZ-dU~8QBH7W1zL*>8ds4EG62kHWyfHaVQuPB8IyJV{NBJAP z8g62Um}3}+uyC(`m+1Z{tnXiU^j)Ge`!g*di4MqTf`;LZy-O@#=pB%{phrl)u9lTLQI+oeFQY- z<@wLJjVNWXK!WaSgq`@gCC!NNbXKFPHu`DrtPEX@Aia17%SRslAmippaQuP>W@+O8 zn=gUDJbhQk_b_j`@~EST#Ze4!gi2Wx{`ypRiXCeU#|x2Adn$vy1b&MO;dLr1tjc_t zSue)QTgQV`q98B8FEq3)UMK5^>_sa!@z5`~U<@xKVHC;&beeMZ+jK9JpudQu-)#?4 zSRRjAI1mrdd*t#5Kq+?H0b1&OfHZZa8-iArofv&3@dsNgqX zy%Q{%rB7K;76a`%sJuyxT|WUpK>7{fnFV&DPrD-tk8R-r?laof6zrY!ou&~EBjKinDwEjzJ?7l>K?AP5uB*=L^P~tYyy2~xwrZ!`s#F=^xx9128 z7ZF2Pg)O7SLAX(}B1`AE%k6I3M0oIIy+&ih_0^+JZv>Ypk+e~Fq`+egf53^I{2%}h z6pR6zwBGS^>$dcYzbJ^jdnY7$pN!BdIJjY6@7!T-04@pAMtzw3CS6$o;FZaIJKPpg zj~8)sm4jn#cPwP0Yb51NUg_zlm!(OUpI-e&m#c=J-7`iLvu`WIDDII1#!%&n4U8mR zEGf3!22wTxQRopr2HYJf{o9I9SZZ!8wbRIVc8l0p(70jM?}K+*UH>Alc&APwTt(k} zN$0(<<^}w79rvs@Nc@Ix&uFulZb3S6o+<7Thg8z>fIhDQ9}_J9WN*`TY10AlLNFuF18qr~6 zY6*}85>pIXaGNB5ort4N`6Z$S1%)Le?mw?u0IEHDO=!zDT60Jrp2#C``udtZL=*&r zwW~#T1ZR&;Hfod60o;$i^1D#kMIFO+NVm9An-D9oM`&oKa+S>%Vl-VL8iiE{{S{3= zPIV_Jgp;*m8d*#W+wONftAmUk80A6@nP|mZ0F~Z-oAxA$wiH25Ef}A>2*B{a!K(SI zq-jp)1pRVV7iIdW<(5yLdH;dX5fMChdXLL7(ZF?6c8}~u5z<*y7^gYb3gkkxDqIla z0Hw=BH-$Ucmf1&L?}QJ!?Q(m{Lf%JL^i}e1n2(T^qezURwueQ*G$X3F5I#hR zGPLN=bl|d!oU1}Gf_4YVOi})98Ul!rV2%?IRIeh(3J< z&BT{zDT*^!LtE*M_=mOCa}klt<&`JZ)lnV0FH7go$)6khiLpH9v+7wC=v^<+ z`>Sx7sYOq`px&s8Xer@gQyzfV&zg!XeHkK3TYCXCEx`A}>&K%i4AoisJ*A4Ie2r?JejJkDAJl>$KbF zR%OApcRJ8wbx)maPFJ7S-mLMDT{!yESMQ0Fwg!VelWW=tiA#KmA{^~*oNHFdUoOOQ zE^d@PZWd>S#E*Ds%bZ(CD0(>QczH0dfsLuGsNw71jxk6!ON>qy{93_|84Ydw_5JH< zPJiIdTDnD}$5a0JVT3!Gq zfwMO`9z-zrm`>|HGNN8pegc0TCt~Tep;YpyvC_s7rc_TaYg&ZjRjX6XRLnBBmMEduR)=;oS14h6u4V$;h1j3<`<2V#x<<)_2a3ieH;E(mLQL&*j z2DcPpE;_a0p>^g4`Wj8UCa*Lev{3yNmuWi^;aP{ys5P^tDwUOrHY_*$n%}8&pUK|t+K6`SLknd{c0ueOr>YhQgL8|kw_n_!u)O}*LCsFt z{qm-kUqFV!%f>Z83!Vv*hf#j}Nas`k1iOJgoFBKa`=DH|j1vqjpKNPh8)3H2vr&Dm z#}A%$1eC8IbVCWKe06N5aG!iExxVy=SHC$w9=X{AC8P^oWG9V=cIw6-vi%PE(i6;}#V^G9@(M@6Ja_xH=0;>G6Hk@;w z4ocF5y!~BB<-nJK7@j`>DffnizY1{L%`N8Yi0*MFv}LgPFRI_~`l=xOK1|NMGIH#T z^CqAZKD+ba#F6v>+a1 zt^DO0LMcFw13%GR$x;49A>)OGLfvh@drSAQKrs+-yN5vE?STWFcSbu|uy^5_ zAHK_gla8a`F?Zps_ueoYywl@HrnPgLMUHr;Y^=;qu+~RnOz6nF{S5&3iyN^&>SF%^ z-8J7k%^!$!_@*|R?03Tw>LP=N>#&A`I$h?C+1vIN9&2<{nkZ@*LX>A6DYMqh3M(nN zLw}hS^|BT-hv0QZ#P^Dk(tQ)vmx|!!_UEZSSu2AGvc9~r;~mgK=Y^7FZb8H)1yeF5 zi-}*Zt$ce$M=|Wrm&krx68kM8y^JfTpMI=ppM29`>!$jyuHSsEmAu2YM%r&HANF-M z{Mk}fCtu4dSHAoy&I!UiBSN7F7F#B8v#&TdzjJ=H7QUuA4L+lWCiW&-wMJcWA3| zcO6IG%9+vf6JdL@%8OkjS=d*XG(@%JAj6LY#TWfjYFj65U1r_AFC*GX$4Ux2JdzQY zlCf&?icT1JzRsXVBPB+o&33F#W^htn+)w)@@>d_k4Wuhfc^}KWtByJ*(g&e`bRFp$ zC?2OfYBld(sQEQ`vL+w7wFhvaH8z3U#rkof*3RJ}HfRPb4{C-xfQKSe=)|%5NT+t? zR=!bSGWq&Wzuk2U313Z)gsxS-eladvPbt#0yEdE9VcMvHpZZ?Uz4F^+eQJexqiXo& zJDlbDvq7Te%(J~TyUmDp7DF;HVKv&(umyM$TZ?Fx@Y z>2EGkhBB8Q4Snd3;+R2*aqjDF0kSLyJaLJ|+wT-xjPl&1 zTbNWtVxl#k&_atU3HBHRzI{If^RUC#yjZNL$;X2|>g{i8Am*)V@_5Pc_!}ao4pNMc zSsFb5&ZN_pus8vEqK`I+qb zl>Ex?%ZpJK|KNr?S$w4?sMSr@bLp3To;h6oQ-?qe13)VuEKk7C_M;3EX^?`m-d>`F zyef!2>;hgE@Z{J?wI}kaJU7JymV7pK=?5JDJFes!)=+I{u;?dB| z>Ar8kNQxeVQ$J}+m^KYo2)nz{G_#L%re|ELCMeN*+=+t~9_z8Z{mTr^aibW@RQwsf zeI2Fw)a_zEyF2egiR>K?toA9OulOQZv4UEkhYh_IiU=l93 zc1BzyFuGEMPa1v<5OkK|RNF3l^r%Um#RwIKlda8Vc}?_;%A_@Z&-yzEkK_S3JGEZVH=X+q7Dc z4%}f5Yk>dg>Xff1-}AV{xQuLklaZI@9y?ht)ICBjEUDh|<<6JqVcMXjsQJ<(B5ur! zh7Q^;#bqPs7$DWz^&cKAr+V6jukG?5sT*Gs$(^CDwZethO`yi=Zs_cjSMVzETU;VO z@!tZ#nNj~Z({A7^pMi%K6D~ja^bslPKREU4mK|IKnRF0?{X_AhLkWmA& z2{LMM?flsMlCeu$5kcGF0}&^xPSg?7U9C3zk8U83 zO*x^qkno~|I+5I8LbZw3fhSibXs|VPol1&`VNyt1E>%YWhnzw@To0dq2~P5#z%$~G zqavNZfuQQ!VJu8a{c85x%WQG-*r>5@pEz#`v@j<-NXi{1h*o!4ihQi<}7jcyLBD$Q{5<1hGTrybEdu&`|p*v2j0ygc-QKOieqXGAg+4 zCW;etbL^5%IN;8Rb~UtOLZWk^W;W4wO$6&+aA$D>mz|T~CG;9txU&_3D$q!CSqJIO z6+AZsVI4O1VERCWB53dwifNcFT4;$eG6*y)?WSuHt)yoaTNIBG1*|*$D(9&OcKjQT zVj_~KlnKamX&utdO@n|4;ygV0Av;#7zl2=J@RH-OS_wM|$Lf4a5_-h)xRgw-$=_~Q z;McOk@LO0NeUo)@C`*jq4VviJ*-p*K#?QnYjn#QCZNxCLGDoUbKqnrvWKPZpxx@r9 z5*dVvUH{vRM-LI3#-v zSa%f^?ZN@hTB02S#Qq=l*I)ImDnNXrQ)()7pDVo}hhG8v&0NZj#+eb3moR^4$hV-@7 zftA^jr3KoIP{b-Z>Rz&iNshOgnCeE5-3#hqnqRJe$wwkzg|=c<`MztGtnv7`Jd;k7 zE-mfu%jlry4(SyYK{4Z`)z>pvOea~z)R-5L^b&f8YKl{S&qB@9Xexi~B`zHc=}n%U z$ssm`R~TNaw@$9B4)+^fNAh?tANKx`U9^d`KH~R(T&V+gX*Ntdi6=oP5|W~69k{Z; z7IO#$#TI)%AS{z&f=XNp0}Qs(jwZ(Z%BM?EHDW(oaDRLhvf`T5e8$e=>(asUxPxS3 z+q&*B>K5)z-Fwq-?bd14dipJK*UC&viI4?XXzwqox!+g++$A`;?&<{HP}QqYdDZWK z9{u(=k8SRjKX{0@DfK*`w#Vcb5wr%r4j#M^heG*{>fFX?P`wL=!nK*Y%-HQWs-(YG z`m!x2pL_%JBmWt1UeO%4q?ZV|L&{aGc3nVxK>zfME|R zgi&nf6m1)JQJ;P4Frqo9xV0-nWB>j~()_?xlwA#lvVe5Xv%@c78YpUbaR1|%6V4TZlu9@v=wCa)xb&R^foV)>Xo{9{7Y>YqBsZ_nmZ(Ns* zH*6@0J#@`|#&tlszk(oHqY!=9S@o9GaNDV7rz1xybS{nEEwCB%!c~4RoK22v7F_R< z^~Y&st|{%9hX@FLO72W+a1W-Cd)a(>{oVQmFJ8lS;Y~X_6j+#88wbaDh)evhoj}0} z;z+{K9v8KKfMCG@Y>%)m-38aPE(2c;t?d|%CrASyugFb856$^E9dtSBdB1^ApPLcd z7OxQXk;TRvjk>op$+&U9V6gLdqH!aY?;*?wK8^S^+d7`L*H}Y^zv4`3$gxqbifSBg z(EE?9N_l89Cjs3c2_>?BE^tElyODA>mUZ+%hBjQu3fR+A^&O{nqlTFZ3 zS+P^`(%gsrMZX3T^7M zXu>YnIg2PlQ-{0z*^C%u)-PDXPw7TSpZg;@1*y1*LHC0-L-3c|?glmVFqP|LNPCjq z`d7Vxkt~}O+D^y*w>i#HLpC()PwXs>!DYLvG%8%D7Vo9w@k9G55DJHv*v3WmNmg?+ zRwvhAplUP}oGA1}W-?0cQ15v=%-A{B`Y1=7=p&Yc`MqTbh!<=lAFXLnF5pAqeR z`;dBG4j!!{u;kn;e&5}$w=o@f0a|lCRO&TIw?2sDP0aSi29`4a53-`RaOIbl zkne72q8~TnUwB(Jl+`Cwif53PLzT0^ZOxG1 z|3Mo%OGmUXWN3G%Nv@{Yz=@8l=vT&K^@7UbGezo!+ST4i2F2 zM6j%2+hIH%qXHZ^fn3U9yxhR8XSQQqv#Qo27ojzC2yaBP11lQqA*d8QM7RWdU>A9- z@V#fKyP|~fK~FjI0oID}kvfw04oBF)D zQ4?D833W*#Wh#n+7AwVc!f|4rJe*HfAr%#i7fmz2_()Zh8TjVv*7I>mYUs2U3QAqDyROMoyh~#^$@^)v+6(Oru~5% zz?LnGluYs8QF&hv`%*%)fF8TVeDID}4D47k z^UMkDKwLZ25J5~gmHYA34AOx8>UCPxJ9ed@$R3+kkJ)V=1y_##0D7co{xTVQDF}K< z4ZNFKO?%i!BGqoFohrM+t?tvP|4ghuic7OW8UEdW38ekLay85q&4@eTFfJ+veJ|u6 zBj5%his335#7%G8#53ePWrK9lh;$;@03@{t^2th3uODxnw&Z)T!)U2PmwkMaFk(-- z3d5Z=0TYtxN8vZj@%Sri17oePDP}p15CcgbuU91^ier4<+S;kYxJ)}@lrmnzFG+m> z(_PP-GJ8~4R(nk_wc}hTPOi~)vUny=Sl)-U)Pmt+4xqnxp@DpM8Z0rJiHLI_@e9!G z7U+^Dk}PvKCap{qoJI`3Jm@VxKo3_XsRsJ1LcPHAnEs-^d=5%Nm^6SdQ=4cp^*zT> zllqY`QXrguF=?fjlhXH^`z*5~!R=!EtQy7-ph0OqbWGVHs^1&phzdQhjg)(E83vK9 z`wVw(;}UHrz*FF+F-JvcXAfCakKANqvQixl5+y%To}Un%s=9^@&iKKacmlknC0GN~ z%^lhE0&c)rqdog%A@{0o$s6~RSOfm?uEnPPOXu1%9BZBF6$~^Mp~%8+G)lh=#t=V7 zO;oH7TWeyuG&7yE%nPL?c#Yj9qUHY)F{=$F024KyP^>>u2X$Bk7i`>x%Sx_F zriU7l;i#xqKM$h~^eaHN5(c3hKD-NG<(-*#w`{C#1y#!Q9Q#ePz1i*JQJ8!#fGL$waf+GO?mEA= z>$4Ym(+TcI&&%x&%bJZ$8lO9rTtl-nxy5d`Yt3x0funPD8^5XS(9b)c8Nxb}4A z(Q;^F>=e{Y#~@%~2Q&qsaian8YtRs~rlCG*n!O6G?d z0i1SSYZdBVXW+AMEb2ES!-!S~5HUSsLv%zaAFqtC0Yz}1djj_+cu#j zavAz#THjK!2bo#4e?M^oOXP@*l}%PLaU~TgWsRMWreZgdqC}vx4Z9dd6`y>(WzNPv z`|_fWT+T9?M$CrL60dYF_jgfr%7S3>11=x=CH}*A5ZG3B0oO|Td38O;gk@eDO2?RN zzuZw+l5o3yHN)IClpA*yq0}!o1}`p^6YU$hmKr4q34v~_#Cj)J6>4L{P8qP?8*x@7 zt6>lwJOc(cCv#0{o99$xl}CB&J-6~SF!mbZ4PjXakqf`-%0WjxGCKoX3&-eAs!OW4B)be2)0ez=enS&X8*x znlDjo^E#-r;gAq#MAHs(4@Rgb^mwl!PAr<5Utc2mF&rZH{PHPUAY5uSj0U+zkUDqA z#&kO88ZwV0_OsLYNM*GzM=YAw_gkm6Q+JmWZAXKFa$p7ZO$8|!5aek0CCcLTbq3Ba z;#oe6*ib5ceW@hLnb>h3U0# zaA$}l{L5W$eyId)VbqBw?RTjT^A*4uLmaoWdsyXy6hSl-%cV3Mq~B_GvYDe4?8zHQ zngK-p5zG!F3mVpTv9&Uz37TFgY2>=l?6I5Im#L-d) zqGiLzfb?cWG)4e}VAKMxNeJdAIW@S#-F@t7@4KP81Q<$(S+CHXSg`uof}P7_6BdRc zC!lgT&!tGL!-r(q>8zTK+#f?d$9xE1d8Nn;f2?-sA_J``PT2YOf~_2J1T52|_8Y5% zBf!}MKVm=mV?tNm2GfDnY8r;|#)aCeAE=r95eRS#%Q1tknlROZOgY=K1o`q+eD#VF zx>CySM($CHg_li$?C0|vGd%!CjMI2PGR@bOjlkVGnQ$owGA#As2X?}Z08Wh#9vmuP z(%6$j*MK}YJrmM>~ko4H=yk#NX4KQah95{P5Inl=8HX~XrlVJLUyn`@a}p-vPZT;X?$0w8 zPOJaZEbUWJ7-pO7@LwUm>bx*?irP7Thek8mJQc|0I0QX{m`nmebCkv*V&$CCN!k!j zJbY{Yb%zf$aVq>gY;Aeb^)?F6Ypi#`a}GSeO`Cug;8=?8T6WAKorqnVV~~!|-6>~} z2srDAthT{aKLML!z>>;vBl*(b7w$zFBIMy$ClT-c4S#CO_2SkMEkfRavk|T`3~t%h zctj$uXg@RLU~n&KWhNZ>(iq_gyeTKlZByUF(|Z-?)IXBVX(ZkiXCD?QJ9W)nirp6R zJSZ=vCbtalwuyS?P+hYf6A_LKcb2LXqNRhgayzXXs|WRO$ONuzti7UVJ%MgN*CbYF zY`QE&TtGQv;0N#%$c~GgeF_F5IhIz6VY@~cOyQ#zaVUv@cEJj z?*-wfvDst5cH;ziG1b1hLdQN;5J(+gju_=3a8YpwwW9A^wWq%NI3r~-AzCs_SfJ{A z*eudWeMu_SS#(VeQ|Q!aZ)YRl!)iBPMyfL4MmG*2@;_=A%n%?+7uNp!!Hc9uUXKu#e{q3xIef6q$# znX!-cj#;tmJz4SFB7|e0ZJ`^#b>HM_*`$VXG9X3`VYdAYdAJ>v7CYtHaJQIg+T@d| z4YUN;i=yE(`(ey<07j5t#32q*K_1+@G2-rg`r%^PB?|cSW(-V|1)+SIz35I};dTz< zth|9z&a%avROHIDk1!G#Gu74q6=YEDIhnHG*eHdO3!@l^rp-%-%rX(F2H^JDyg)+> z*ty__!Pya})>Sxg@!u@z@5b&`YyzoHe zkC9nOEXlSU)0`~*n~yE533N^Yi_Rej9IyJshvW~I(RL+Q+bFj^#~3M%JsVVZP$&9^ zFOl)4O7K>-%)IESgCy7)Kij@|nW336;0U0<1!Y-rV30Ok-e0y;>tvPZDZ@^LLxC=V z+$mE7F{yiFCpQjm#Mp5bap2J=VAPW|g=62J3P+2pyX%pQKajf%aA%AP5!k5|ReM&c)y@l4l&bMA( zkF=cpFVLBI*8%s7_HYBJ3pfs~li_u_2VTJ)eLEf^0@8cGvuxnsR3^BNYt zSK0T;K*X5IQe>V?`Jr~#_h(v7Uph5O?1sDF_v1jsh)JAM97yOICs!k%d*B+)?3}J% zd{>387@*RM`|_x`hkf_~Jbm@EjWA1YpJ%i@vz`-QTIiwWrJ=Ibr_14%*^xc0+jimH zXMVd%r&l5dkNrB~o@6-FWkJr%BM0D2N6%*eSbMYsx&-{~prTzYhC|Ajp9b=eb*XPh zuC#N7!@Urg?FJGYtDGhgS!F(XL>hV==x;Cx>`c|KuSifi;WmnGq{*b2?+)<^EuncV?Jt|DjaC1K_@eW+oR4v**KjRTa%s$wiP z$vXDC)o|OPUGd|JgZ)$NH*Qlwcdu(Oz%c{-GD?hDMhH1qr`F&OXobNqM_EPdeF;>% zazeB}Sa7R;60WWg8jLS0z!M<1K4>3KWF~qoMZ)v22E!z?I zl!ya@{|q zoV{a{?Wm(x$G#Zris1-+#aOAo9eLbj0N<{U)uZQk4k!$LnlA1L>P+M7nm59J^`Iuu{j-;s*kvya;bD+4vs!-h(d@ zj&ugMnGMy!EoU9r8b!5(BYpt60|l|USArR2)JajS#&CQ`2D!nwO;5lx|YuvL}?{JZWVVS0p1g>TY_gek3$P!KWA zXQ;J(VLf@S$hSOQG5oeIkkuX3yLsUJZus{z1~f8-k&QiM_ScBQg-D$pYNzZ6;BdK8 zc5%m1H=-uL-Q7-co8oVk50V#9czIG>&uIL z^pTgt(9(vG>b|r+t92xhr^*exJMgTSgr~ZS!gYFw2c*Licy#mUH-=3DK4^7IoiVIV zYTR}rlT%DmKAijR-POX5z<1yM4S4ybRwK%2Ps}w1HyklJn(L=9V5P>g8`s&UapEo8 zZou%=GV{gHE>Se?;-WfT?CMNm6w-S6fF{_5RZsA1TRY%BKUm6T{||d#9@gZw^-Jq% zJ$@>jmeZml(AFwi2h=bJh^<-&1Z-6hfkaCIky(Z?C2=^4R)$tl1R0zHDw7P#ln`x@ zs6bGV84^N-00BY>A&~K|H{dDXz4e@)-j6=dy-)q4j}_hT-fOREui>{$p>G5fb$10) z4BB(HcF3C`s@93y08up^b$u3KaNJFZ4n#856Hv;aiRlFLIi_cU^3FlV__*S`LWLYH zQmMq+=N=%e%Po8N)!QJ5i5qm1YJ5&VKeO5M>3-V5z0!a#X|Fa=#v<7>l%JFzR#Vxx zt)s!O9n8KSVR*&dyciqyNRJ{!D%2|%*X*b9juA?TmT*(R+J!R$c+0-3k8v}&(Zzje zhbmK$@$D72)jnfi%dYAF5U%>@15)(vnW`R}uxEply?PhE8@TwSAZ$2=<&pL!4-)q> zXi|$B27rij%bLLZjM`T=DuxKvS_Cb8O^2$tk@?)?UNF-l3-9gsW7*#7MFWfr@2PIN z(4zAYmT45=`AK()1=q*XU76b<-O!Rx*fuxmhLl~n5l4N-elS?#np9w!)IF*k7n&Tr zM_}E)8G=-(gOkXk>}OIs_Al4&rAL6F4kH1Q)`)7Tf&9UI?2>}U^QXlF-}e?!MsM3k za}2c@Jy9Q5QDJh_hL^>h&ze#m_&{0p5WFozZup)6kVaeSxA(05a?%b%Y$y{0^;7E1E|g7HWRzLNS4=!E4G^9}CTXdwM1RtH z2#$6#JYR+lHY8404EZL40Ly5)0dqmLw zO)DjQ(}@>cWUb77=!&Gh)P19aNYrYLALKDxkdoO@f#gMqW-kc`u*WdeN#J%f5+EQ4 z^Q~Glx$u&^ZwAx9ssg8iz<}+ueEFOaC+S!W+p_-2rXiQ@7=pMckv?-xl;W|wPQ3+t zw#S6^u+|qo0=8+gEx~G3^yAZmKJ_iM8;4j{MrxEp zm9Ehb;mQR_T+=s88yzHv+OsG~hsZ`CgK(PY0#k$#n-B3<=w^Q@6ZQ98Fd8ASztEwl zFmG7l5p+g#`6!(WSPy=6f(|fPFY>yoVVK*t$eSIsk>-GW*p4EztHsZfl8<<|ruiMk z&*!o^+_PP`;xc1%+ZDV_1*EwfS>I&Wl}Tp?JL( z-~tg%{kv)MD(Gctsy;`fnIJE*u;temdZ0t2j+D4LkT!294RT(GqkN(dnZ6~u9hDAl zx_?0+bP@PF$bPN_#G*%M3Ui+?w}A#0N%}&&zWaXuYPg=D>ZJ(sT?;N~6tQ3EO3*YG ze(i?v95M*2(cgNB)#O+1+Ppq$k=+*taW=Vgs+7MB{8IfGbE&XPT3dlU1>t$Br!4XB z9Xo_{6eu~Pk3NAJH|aLlTE6BzUI=M*qxb?_WfSO` zV%JZw?zt$_ill^bbe2e<+>x-G3{j60+3EGk2*9QvCGF}5*qev#g&wx+TuV1aBZezv6jY z)`zy;iGEshFD`A8xjwB`F?5cnmw~jdh22w0pQF-a*9N$KxNxK#r5}0JgTKK62#GYR zsk@2DCt&iHbCfTPxSE8?hQnXANyU7=llxwz`yLWTc2nsG(sk{Q@g;zXv1CgHeyzbi ztTOWfW#is2RLVR4r$;|~+<2L`fSuj(vWqDQJ5Vy_kZFwNu9M#F+o-*BN1J z1!YqIlCkt`A0+awxbIT-NPH1~f(wgsFVQZ)>~IDsKEAeGaUFv2Wjo*ZgX&T1I?E4S zns@Lj8^4441ZgxlvcW{e^Cvy%etC`#oz59)76&0eD%V~C@rZ*I(W%fNy52%}jy0Zv zgcJ-SqI{`)L0^7-pzN>Qua<(pw&4%Tyr6Ctg<)g6q&Jqxgrsw7llVzKP<3$*LRJ|B|KQD&TH1;cooTP0C!*n`E{2L*ZXd7k%!1 zsA@}K94OfkKZW0Lj^T)Detuz}Pcxjtt7-xAp^b1qNXHn`&pXrm&d+YK zLB`#EJ6CweDBiOJP_=bDBu{YxsULXq3b)!tl#>+qFhs?z(2q0k;iXL)WGx1_6A%!D zhK31GWJZUy&FWm1;MoK`fWtFwQ9e;R&!wUTGwCWnKD=#$COP1cCBjpu94avX2pr`4 z>-O8A)u;orOkFOp#8hbOVP^xl*XuuM=`;{d*sf5%PRvazEv$i}b%QX2nzi+v%Rx-& zL40}Cch4@8g#i2s*2}i+6&~6Rr)-mT?-L;Mg@0&Jtts)EWEhyv#{M8kRCkrr#)RzGHx4yUAAzf($EnzaZvD` zy)NSC%cn$85OCOoyf73s>P4`$Yf`Rz2k-wDwH^wziwdk=#&`!0GP2z@$;=l zLapCxFf$#XiCs$k-4w?bfp+`a35ro**x4Uk;g;MI@x^z;dN0d*yEcWjk%pmi1{xfQ zS`=>^Z!pcB zpOx6F!${8s$K<(3W>&SE>0rn;$xcJfvP|uIbmb*1q~f8d2MMbtRGE})h=t0}^DqU# z$}(;2PR9&<)*ngm!_L9BNJU@>Q7&k(b+2!lfYM-XhwZM!004{7%bAHWbOAa^Wt8}b z(jeB)fRNi;0ONL~Pt_ZSUbppb<;I$5`Dl1kHWrDMgRDZq)nDOG{wR=)$D1E+>QSO@ zii$8l6kJdpC}B@MKu&|Qxv>rsXd}SI$^ zL+RK}aT2+CV&JQ`wOHW~(42vcN5Xrbf4slBj|H$0_vumyNp_WoinBLCN2D$?d<4NT z<_qRpqU+12)cR-Q7Btmdf}SzYI@E^mt;fQvR1}*Y2Zyf6^J|eVg3F9C1e`3NBtIyh z%v@Oz{;c7kqSgK7K*->VfGi>r49lKC*)Wv1nLdg{<_VZ-3L0UqE>bn4jyG3dM=lSV zDd6(_HIr9hk!unS5K(Pw{G6eeau^)F-+Iql{o2+!8dX9EMh@Byew7rUy z_DkF0=8thx1N5LSKS`CkFL&?+Ou^hns$I_s9^8fkidqxRZc!Sr_^f2lG9 zXC!Ihy+WY^`c@pQJ__m`vtJ$(R7FDDK&(J}_|-5{6yQ6Jklq_v8gQ^=|+NoRv6Q}8;x%CETjl#bhI3+*cf;Fz(JJIA& z;WrtV)M+kn`ziY@zEYG%hqj!}%Q0Bu%QS$nC>#dRJKDG{4NxF8!*BSOQTD#?n=*ca zZ?=Jy0Ef7_du|Sqz%pMRHLSImkuD*@e^d#A(GT11b~P>ZFi^B^lus?D2hzOs4QwQP z;X!I-DLmdm3h&zybK1$(jH1kE8B=)Gct65lNt%?s4}H;75I-lhfuFb6G(Fh#8a%OU zz^};MrB&K+8o{EpNE+EDF5h-HLPo_ZNhsa6UDzt z@kl9q%WXk|Ja~Oq%;13qC(@hhzX_ml1>w&F<-ItYDx(R!=p|C<;TKRG8VcMCNH?q zMp%&QZKLe1e!9@3=D{VSL;?>_%!2TZ=fFEok?J8ryoR(p*rcOB_Fe7E{m%|Co82J4 zoky1KrNRKABB{k-orIGe!wr~hlY(!Q^sIVh@Jt~;VCmV6$R|RMZN}zB8MH5r9c-eZO@J=!4m2MMRf-9$@~gD8kElDS3kofSrrFe$wi;VOkKh(MY4jL zNF!JY_KVWqD=rU}yz|v6g0EcWZm5*J8f0UHq+onG6q}VCQ$%Hj^{`8%0-S=MrKDC! zWdr#c?K_~Ikh3C7&KEvO=m@|GY9hGMDu5%1M~g7BA|FWhM&Iq%A#!xi0+@rp57pjH z&?LsZG*4=xh23nT;lHcImjuNR#i^o*CV8e5jM%&IX@&EvlZwK6RFYZj zzt$Q7nNA5Fu-~6=YrHdfkI3;q6Wdu_hRj!MX&~IeUe@S$RxB6yp^3gr{3ma5T5Bpp z%qQd|S08Yn*9#2Js^qChCVw{cQ{6p0(3?j+DsOD4oLEY-mRH;QFUJOt-{YhSDE_5l zx`BSY0j5&SAL+8E?hT{Kx&p$VZq?t6G$GRg$%MkWXuDx4cJ!3aA_Op=q2k?*BdNA! zlLW75G$}2k9R{BL+d-&V$)#)OmHLM_USnBNWL<*vriC9O{CvK0 z!iN?X-oMqLT>R}zEoHY>x+7+ME2g+9O(E8{?)f0$=wJ+KGKNdDkAp5FHpx=)G(rB< zF5K1hkj+mTfnsypDg~tf_ey&V^>izF z3I6&U*$p+3~}U#;iOX2fis1|mazj4 zRKjGkP9;A4nNPkxi&$qukFg069m4gu`GDGTAPnxqzdo}$H%s_lg3#lpKsh4NPZy=~ za-dJa=XoV@{FVWAg;An6Aukl15*7C}vi8|K9YL^h6I%}_shASf4@nmV&Yys?eG%)@!AAUny3)J*r^gkN!;RQ z))~pqr3<-?5>`I8tNu|aJX1K|y8lb;>W*$Oc=oMO6z^Zw zdqv7Nb)$a#&cYw(_;kXA#q{+pix@b`9ZoAb0}VzTL#feuZnGD+I!f7bmUAIJ6*t&k zQ;oSXmvgg>&ik5@3pL4cH1H|Vz!_6mOFWiTLSa{QhWw%_UG@` z7|$Wbmm8Ik!k6S#95tVKp05{gEimZnUFcy(>Q(^jMf#cxLg}MMn!QQo6NN^R4F#O} z>5|7B3lgzzG(MZXnF2`09g**N*@G`Nl?E=1fs$hYWehivLv+u*Pb&+qaf#vPtDbRM zFDPl8R*|wE4UK(tGtyymX?nt9Oz;y=>fVX>$Rua!^Q=xA`Ojp*QtU*MlU56jo5Jb4 z7LeYQVQ)J2Ra+nL%;q={8wN?vh(RWi&M3wQ$W4U2+>ZFZ->Nz?vs*qp{7%EXyzx&E zxH2Xg{S9@)%TMYy)w%p<5;OaDz*%;+mE4Ft@a}alaqufenSxFi@;5AxR|!eWFR@ll2KG$>Y!kJ=YN3Z#Wc7|1cGQ!_bRX8}Lynu7t6F@ps~hw1 zJ?|yNVB{P+GGQD`42CYFtf{)s#D3Vl6pMCzFKD zoU{N^$0z~`7e=yd10Vewd-4DLZA_jJ{ zO$-6|5;2V97+mO)UnV8cC9Y!y9NhBfMqzLlu-A>zP5bu*e$&R2CgA zJjI~vPu$NmyltyO-(ae~KO5^*&m3aU;f;CnisvLi2FzqWUUyGd{hmRKd>s5BcQ~A{BcP&^G zCwh$Z@}{^r15Ij=9lga$kzbt)lp&|@K@*`2Tv9{=4@pUW6VL}~Bf5p!uH zosl=GML&=)kK=68!J#@OYf2z8bY(`N(;^#$9GZf&p8kXsBp&T6q4j}2-bb<^qE82#*$&BGwr)U!(%Q+c<@W}|DwcpV+ulC3-C z5N=__>t(C{F5-e&kcWZnXUNJjboIxt&xiXwtqaz!YK4BmEjX&mnjd}x44Go|tEKn7 z=hi0@2|3wk6R^RY(dSOq?vrP;Xm|*B=Gtwv19=1;!V!ICDB%wI7=;0RPaMuV7 zcI?($hWiKl!9Eph@?HBuFxn~xOsg7^EFkNwaMZ<>E>ZnGtEpONwPOV`$|FXlKpvnY zg(O*kYS=I&L)Ph)EqpI6%X-+`2M|}IoyXdJ^7V)VNks3->Hv9lu8Ff-$Cc&+B^Tc4 zH2k16y`R}klCvA730-YccA9ly5BXZLmc1K)A4t=K;t4~`Pz58YsJdEpO(3fTmwT5_ z7cw+Ni#E2FHYR+4F1TXY?Cl$xY1oJVZk3^Oih)z<5MJsm){1z#^(;muf?+O4&ew@S z?|ok?AFGK8Q1%5VCsp|t1w8vKBSnLeUZvQD78)*X%%~KH_JuV0d~ja_c^p&b?r~ht z5W2)@RSE_H+oWYeQsM3WF+RJeot|bx*&)OyGa45mSElXQuEp|<_ zycYDnp|XvrXdw~!Dx$nm;13w8QufcQ%7-6zoJ9skyZG3Q6lq5ZLZXJ;4baPNqQm*` zJq@MoKREZ!iC6Dd1FBzmlZlxYMpBrth%KRLE!I-qx)vi~T_=hQ@mfZabxrgRmh=jT zDxi8#7;TLuebp2Lc9sLfke486W{m(NlP3!w5M=bfBC#!_cxd)gT~%=z$W(9x&LK|4 zYC`f#4*tmQY@yJ?e>+@j(McW*nF-F%?; z#H*1(_N8Vib6E>Q=w)%Fo3F~_j6RBs>K@5y9Ngo?3+ib>E2FA0oS-zLErmp`G%+cWWUd;QL(5wVU`KE^b)fNSXa^drPqYEnyUx z7FhQn#NeLON{vWI%ns|d256f_0@YI?C(exceQ8gAo&i?g?sK*#kHW7e)nB0FGTNI- zthDiaEwoT`X?4k>09v2=&5Es=&w1y>8;501O^6(oN<6Y6NvWb$Uyt^vV ztB0y~be^J0r6^vjiVy(WaceVsEiDsfNCZvx@4yT?w7he|+Hr;%X=7v8fQ%m>mWf!yqs8pYW;k*-z_s+dcyrc1PHul1a_jYQP5~q62 zAXu~h-W&vLmT4R=m~m6*@YYQ<9(^~X|L4%>I;L^r$-TJ`rf6tfIPizw{~V?b=V`2S zoC$9#8fpUODo*p~{r|_g^;z=&$K3xM2Jhe3IG8hI{-ha3G|*+<-^@jrbIv7G1AbjI znlmE~tG7naS!x|tf7E9?S5*iIX-si*7A%F;?@z&l5EngV2U_aF7yrXKCRpFMa?T#l zu=-&UdA=)cvT#n^7FhpJ#+l_i&&o*slX1e^{vT$XxhdiT+Wfz-&WK8nIq{C)FwWeh zM?Ed`Y7V&Fy}f?PzdZF-kV8p8p+E58z9Y<*;I^(ceN|vs*KG-T%;;hSA}#P2fzG zsxyKgW}j&o9rD`bf0lm8k_}BUt}5|V%i@HHX;pG`*0~Oc#=HNf63Sn>!e|omy0c)r zybA=p9_6*%YOEB#g_FbtB+Ne7|M*?W=Fd;C`A+<4OOlPqPk0ik6J4$=u*sI(5XK{= zW(U)_2j5zN_?DQ&Oc}unHV&@dmMOV{a790>QWS3$2w&n^_uiY_f+V`*O_}B)uXkl@Rg92ccEn^v4Mc&2dv=BMoT>BU@V6rEZobs> ziZEun{hq3U!|zVtSMBtDU5!I&8YuFbV%#z$iP^t$tdJPzD-E5$D{{4JRIQq-ejy<1 zrzjTZh||CaMv?evbEup$j;XG&FP}Gks&7)fQmC_R`)@8;e)N}08YsFeR|M4mr?T&63F+r9FJB6T<37;NVFs_O!~QM;el|_O=Q9LMBgC*q zFJ)^{M})V{p3MIq(St`R{2Ugk8tOarH$5J-Vl)i>J_1g2;O#dAwBSneFI?@{M6D-J{VusmeSeyOPiF|2OK>nB^TUTv9*BGzoPR;92Cg#Fm(pm}mMh_}rV|F~BIe`^P=$B1s?m1k{2_Go z>yyaWb{5(3f8WVcrcY)vbFxhndX~pu4xYa6Ki?S$oAbDG>ka~9!s@=5`R+Fp)|`B$c#wX-K!zKinA8Bz z`!2}zo^tLGdR%IB55BlBCE(D)-~18dS*}d}OM&{)Kx-GOlJWWm} zF>50&P@?)|XCp69r3H%LTU(3q^2Ep1A9CbT6biB~w|wWt&;5>7miz@2%A?QHf<|-& z1=78~Lq+N1)tqmVsA!u#U~by0h(k{VZRPaa6?IGJ<0NfjDSg=xI%GUI8>PHq+oi{Q zRSGZ+R(Vlf?;0QX?8`tEVVSk^V9(~7S-WV{!;2MU;ZnOpx*%tE{0+2{@hUUcEi3^s zbcfTK=95yKtrlZMQCw0r{}|lswN@P~aW1~J-Zyl#KflRG%coHA;$@8!PyXUUz&ZX+ z8{L-R&^ShDK6HHipWmA~b%j@SU!f(x;GZ9^nU=_d506Z~LTTzp9EjXG@lk!brOkfi zsJE-pw&gX6SQA$0U~Y_w1-O4+IZWxbbiA9ZuDI>DoO3%@lBB68M$E!1k7?R!JOPKH zQxv;(2m8U;kG|RS6`Th`P6{|EtBhF(8ua>NPw((wE-)9D)0YjSC(p9o(j-Yxcv5OJ zSY)Kd-f7T6^?zsCFPGVi3a&~{o^{7vzR*U8JKPwm3)6H!+plr?`+X-HAN zBRuIm;F3X7jG~MP`S9Vd#koeMHY~U5>Iyq*-+fQ7<-SCWh!gZBSKw`#7(8~**+EUbXmx6K>dV5v3ru&@ZFNq&!;9&_3>+& zfu4QWP(>)@|7wTUtF`*I)M*?1sB0tfw2AdLU0F%}SX?N)Q1Eg1pF(yl6@6OJwXaX{ zqG>9@T7Jf?I&_~w96zb~!T5{v@jb@OK#xop3lsrPks5S%EvoBz@&MY)lNXC}O&)M$ z?qS0zy;3>P_mmv>lrods7EM!k#&bGRF`y#$_^4Qwr+4J<(Q2%&?7kbIrytrDbv7g7 z8RA8Sk%c3hSYI}DeeX-Dh{a!xOgoaLS1A>wt6xq0(K8NFwyqfQ%Ti*tk7b@HD^=V_ z+vk*i_xYl<^oY=4y0xc*_#O@_`MTO<4y9ADpqi zFL?3O3U9Yc#gti3`B)5qL2FTm|sGsx5qe3<9%9sF;LRxxX>%0>AoW^MW9 zGJUr+(N$EV(}-sUqa``wHNUoR+H297v4c4;E9pCEX4c=MUuI#1`Df=_^U{Pz(t!68 zt1F&_Gj`$=wBzuJk#0v4hN6D{DZti81cJteaueC#JhkHZ#(7SM7(oQ}clW3)P!%=F z1BkA+Kq-JrHh5*KQj_a=YPmL>k?YoXWa*>Q`Kdwu{S}S|57XMM)8&~Qnw6w{vfKXR z=K~|gtd-U1?d3kH4HeZDyQrD>86X;w{s(X6YKh}8Yix3hOm7Jtu%(S!xGSN#wzVnn zF<|kCZXH%#7C>QuEHW=>z5XggmY98(gy(d`s3zq6n2boFo$xW}z!W@J?=WRyEdG?x z|HqLe~zcp1&T+qlv={=uX~ zy@#QmntBA6Ll?9yP*F*!q6iSYA=n_d);Ek<+t+4^8u8SUsb}{X)JvI)9nNAbft;C$-*z$Nmg7xs6~?v)6FmfDdz|@GDfo(vaRK#Py!B- z6|=<{#YM^Pi?^A}XlD~}HH`cLZp(zwM?U$0C{Muf zG6T5iL-qS)@N5k#tw-hSR|?Bo7PEtdi{Xm;yGBm<982Yz*9YZsqXnW`pND;E&74=& zp`uH2Jh%-K(27Bx6Eb=Mn-Uch(>ATYRs8^G-!P+CS>Fe{ zeaxG(FBX3xvhMm;nWmn#S#GXr5n+Bpx=T{L_2#@U59#`Oa#ac0whVn?Xl6bTFN(Pb z2c)d;$^l)YoKk|jacIYkv5xX(YsW`-ES;a=dQra{Ji#uHUd>Icm4gpXG2?v75DgNP z`{08(KDX9S)5}v742lU=G-ZLh2HTaxt@PCT62T#faav?CPsE1$ohL|U5qgev$l>&IqX)PCoLAh*E}3y!2FFRo2) zJIbtmzpvKU*voSsd@==9bkYJL7*zmp`&__$!YQAYa)XVBL8BH%^{E zFcK?|%W#G6AdUgnUeyt8>ydlm@4+E|K=A6j6qMQg81AS!*fTXoQ%%NY9hLJZr=w>x z(F@QPV~>V2Mj~GaA*&aUr-169-V33F)CBC(TCj3ggJ+4@BwGG#xYtJte~3j9;|yrc z2{;iWU%}P8j-7w7aN2PIA4Q@UO&46Dj>UqsVzH4v1I@+xubRynmi09yeHA7=Gpq2_ zD7TZ1Cd9vr$Q`PSZk*boquk|uD|}C_xlt7C_GPt2bA;3qrNgX^H)R$ugR_{mhTJ{5 zpzm(Q1?PxD%vPwidtcJDnFlB8k*6-^gvBfQ_oPxE=ECA&|HIVOHRjJE5C?1e*VXP< zT1TYMlz}a&I!Ae(k0LGh1zr@r`KgA+i~d(yhtr=rOvKbioD5?w&PhVOXK11I%U{sG zt^W5vS{S`xmYozI&+Pu3hdlJBe_j7Zs?m2IzCXF`ZQswc`Nx2vjBa=wdp#|99=~_b z<6X>oZ~nc;mRl7wE421ewxD_MgL8iBQtJGF{)H0@Mwl@h72lKI_Vcgtym?m*$1W+H zEqU*?TbSL)GXI|X+UlPRXzY10_URr>TLwA!Ejzd}O#!dFzfPi!j}z4-Sa(_U?r^H^ zK_jpKvOli>%lgagrlx6drTQ%-Or1W#+kPHr|K zV5%3WO?%q{qL~HOzHI?*gP+GQDU#pvbK~~&Geq=%8xie^e{FP_@OsP7UmP{WoQU`g zuBj~WSFh>OUn+#^YkbSkx7_NQoW6!g-dh+cOtkeXu@FOj8y|GZXBOyv%K{TU(+j+f zm8oxbzxn(!A1Jk!$^s3frarS+StQfc7 zp1H?;zb^2HPua=;mYFeBpZPW-*v*{5H15}VZB^@QH81)u`*l_`Zh!q|3+BJ7a}J-< zSF4nwHNVbkFYd(5ERg@UGo(*1@HU-zC+p1uZ`}a~QEEj_Z`%Q5W`UBo?qGU>j<+qq zd$YjXc3=^$7N?S5|M~To%WbB`DeY})wQpQ4PH+49*TFM-IqZkG+~ZZQS}(ut=M#IS z8($|LUeLhG_;B4W%X$0WeMlBmInJyS?r98+WlrJs7twM zv)5wf!BfSOtiRY^WIjqf;WrSyXF=a5!^3NSv03YOOSkm?xiK}_{-oGHx#v$F`%^0a zR2P3LSdBkT(x0~SPw(P?#2*`LFnQ1R#|dTP;av?&uYP!ceOkhr$IEY1t<@=@f12Ju zfAc2={-nVFM-;e@cA!X3MlENNn&)Y;`KdeLcN1LAjED&`xn;SH=+OaQMnw17u@-*| zRc+CH$VfYN;OI3iV`C-+nNQTzG7K@7Elq)c%|LIVYTiGb(5AXdBPgj%;^=!aExCW2 zS^a7^qEb08V7*GztR1y0t0^}$Jt--@z=XG15Hu3Hpg$H>#I-=7xV1y!%-UFUBXqjx zXScCb*NeBOdY4+Pm#6zOL{Xa&qPP0+hB3VKn0l*Gw-`x=;McpCHcW&vd4683 z)&LaXJp2NG=D61qfu?xbN0qBJ2Tl&L#rBa0b3`c;*ZI;Bb{QJ1ksv(zb_g4hheSqm-?&`16su*MnUoKcnhx zQkoB4ds%A3?CcBlG=AQs+UE4s1%HZ2qs?krvOYXERvfSS+>}(2c`Wo-=W;yY+^!>FyH|ro>?9N@5lXKn_ zGRj(UBAm09Ul`MtXJrUka{g1)*wkwNfaH5ZujG4V`9Qp$r6>p&qEr@T=;;@CMMx~Y z!u#fFYLUk69Q3f0)Ovla60<=cEA}y3svIxDnaZaulC<(IBnRj|TB`G(?;6Yz`bnCX5}ix80jg;@mL#5pB4U5s4+7rm1o`(RLMe`@hq@NB|{W{5<= zPv{*V6_I2`S^^A@GJayciOnap$%+ljH>jp`7CY%nxszKQpysDSsJW5#UN!o0vspbo z?NbwV)|z_u{tb|+ZX2Ee6Bv4ku5B%GNjIg8B^h0iZAEvOOM;Z%ZZ<4s5!*D+-R;}3 ztEi&3E-EUx`gBT@@ zG#KzVVd+()x0So3wukF16?`s^{Q2x^pWwE^Luvg_5~BTGgNOaN<+sK6+#Hjx=xt_( zd-NI9C%4r?+Q?+~|Dc$750*iFL|g61AO-Z*kWLHP<8-l=A?b*a^{0@bXSl6ZFn2k!9Hrh!3txe0eoJLj>DgKm}op$BKX|P z`Z{)BtGkCC)6MNxL6-a)Ln}%3^MxKuA5?jd4zBoty|-dFuX&|sU3FVtZ6uT&Kt6Ax z+Pvekb3m^;k2j*rE8on^Rm4DU54t1MA01?S->PtR$Tr!PVTX{2h(D_uGYa)!5GWi*=6p|gooJ;6{fSmZ zV0c|beymm}(9J$e7{7^SQjOkGegUd3E=T$u(LzXk%iFCwqc6F6|FyQs#-K(^eJ~S0 zmA}e)X5MJ*EYK}K?MkYav3TiIYwE@ zJP9pGC^H-R*pGkhwU>VBX6w&&EBg5v#uRB#5+lY3GP~%P;gThSPX_!8 zyH2hp(4!^GmAn|0fAZAgZ2ss_K=|=l)iBNyE56_r+3RM|SxDjj;N_`mo34h5Tm`g9 z?`l(5!vIO@fh3F6B(Zyr8moIP`Ce2zEjt*>430kkk%u!j-l9Rol_-PT(BYkRF z(NgvGU%Tq{`q~r0v%X~ZrFXVqlwS{^cB}4NG&QkBS05kk!`a@Eq}WN^+eG`}b&Nmctq_$jhItTD^38oirRks8q~Y2q(NcW;sz zh)T4;LgMn7iXU+0#(9^e$6Ogr5A<51TqS@<{JOTbBE*x)CUu*7d0KjTvZgGdLa~M- zGVv6IBj!93n3~Fi8$?v`;Q)C6*=P-v(;fDX4q8WTeQ?+9K&0bO?*4?Y>a0 zd^$Nt4%V3={>G|@;>DsseA`IQ{8v7jK4EJAX^@gJ7NBp?vXfw%#~WM*^=$x$3}s}z zTFCvPRH@10AZ#CqwoFYY>)xePEKqP$unmp-=<1}r>Wp(6mCxJMU2-57~Mc+=q4`vpA0wU{*6j!0Lq zZeRn76$I7q)QS76JRa5`OcUFZ#@-9^n=Vxi8dwUYsz+11p07(wPVTMEX)&+3>}-ib z)g(J-1o^KG4#9d}5fw&F6Nm={T8)~BDoj?B2%F(7ebR85JYLxJVU+j4H)PU&*ozx} zvUVQrWBn70bqI<}J@ZgVttphj-A32F=anw<5x3J*63eiamb@AgXM2#_n~jVCn;0|w zJ+3foYma+2`~0$}dh?RY7M4b+nus=QwYak?io9NPt+TKZ2a7i2fI1Nfb@;-SWIB2L ztIH43*vPL2Y~bPD`jOf>|3?gtzK3IcJWW()r?CYo0uH$*&V`*Nl zX`+eXv6`Og!eYEKR(`_kHTrsOv*4L$I-&rKF?%BkYpClUbWm@N4>g=@aNmtNDD7)d z*s3zeL&*W48ocE)noH!Nfx&O3^&NFACuwej3VMl7JSgwRda3xM-9Dw@xW#Hc2h|#R zlu~mW&Afp!kZcz(tMcA0vh%s|udC>10ZyBkhx7@E-=q{E2IX$c$k5WM$*B{O?|o<2 zklL6sT|Ta%kuxbEXYc0f)KCY>WZXIhqXDxvEnby1j@wOhE~g|aWntO4H3|XT_`wTw znU-s@^acMJPClQGOXVAS0s|6t1)I5Pp5pXi@Us~PmQR*-9(hPE{`g<_9L+*hKX_Uw z`5{{*Mtn7KXPWg21B}7m{4F`p24fiw|J>+)+=0kl-ty1NblhB|O%2K9w2=CSNw#R@ zTWRP}h*@GK4H6Goy`#z^{j2(sZqlvw=e3?9L_9`myj{TzpMS6AQ~Rwceuw7&>q7Qf zsQam_)ST!lijBtKUYaG|t48wH(D*nY`tjhYhRG*EM&-ASmAk%=;@luRU1`Yo$&!g0 zNUPt`c=S0u17*G0T9ouKwqaF_JFug*Da+4MfbP{C{~=PZ$2f->x-R$RzwVWp1uSac zI2l#d)E0>-^uC5Ps}-aO>VFlypDog}eZvzP>zdhLD8BWU-z?Rk=8eaZ?MAyBegg80 zgw$^!9D(M-k%x3yz4YPuBmPQ^lEN`@2)_(AXWcS^=>QYOIy3m--1!FA#%2GywO|$w zPPcJqZ}8aYr{4k_u)t&Ck*6}zGyOjr_jlsHi6nT-PnC9b?U!;IR!Q3$gs>YyNWvFD zp)PO}PeOxAn6=1?kPKoaR4`M?LcIz2#q(>bSW|v&eQ67UZnNhI~>M;0+blBUhDXDK{eWiW=y7Nf_zH z$-_vK2d^4!UoMM<2?o~Roic7TCjDlf+4CBC{T$j2wk5~7sg5G- zu3^GDGATZ7#F`ucwAg~YP)WzdU$!(>_E0AGA{}>{arm?*S$#M_{;pV6O1HB0p7_lI zb@gfRaVU_yZMA}Jfxqgr<_()^-2bbJ9OCXRf45Y}n)2#aOk(`y227(p{yK;bJ`f2V zxDsB33h73%T2J6U`d+NmqV;Q!*mUTX7Ge{wcE^jx?Zzyw{OY)g84ArBE}*RrCR|Ab z9m$Ii7Dp9M|4WUlRQ4BfxXFo|9k!~xS4Bof)LV(lsj%{v#L67lhzqlW#I$ZaorfU~ z+4gKzcj+0kLL2msQgoYqK8$OyYWz37ShLt-<4LtU?H%!TW$-Vot)WpQF4?Vfx!Nc) zs_FFZ;L$cRW(}ECoi=idoNoj?k%dH;@VPdxBdh0WGNwCYk5NQ9tkcyM-%wY*JQ{k% z-FxJ>U&tf%g;c?XEX_vg*@~w7;xsfo2O4dYlB$iuqnftv4j!wwZAogt@cqa2K;3>k z0=g`tSo$OXJuQM`^e0FU95f%*4_eO zw8r5uz|YGBZRVdZooHB2uv?=LjW}Wsy{mEfG2&IWLwl=H)c6xR&Wdvx^jMMqEJE~a z7Of-yo94_}3ghsp@l<&)2EQwn3W~=SREAgQkxcS0dvP4WMUp^VOPuNyU~jy&@#|87vrEb;hS?f$L=Km5PbzrT4Na6&P4 zcK`lw>2sgu zhaH_ZUB}aZ%4g5&cm+g)&Whp%q%WFVx39G6-G_DDEovK9Mga>XTdHd3ndR7YOB>k) z=U=&l5%V4F&okWr)qtGY9M})RaKWK8vRY@a0i8|6r)IG))fFJajx@G7*<_Dj0aoG4 zn|Rny!K=CRn}-<97H(u_7TB)oDNN@VfDeH*#EWe>P8u4j@m<@#+z&IHG)}}xdVs%t z_F1}qCVSOjM>Z~@_IHu%*S({$ao+F1diwV}nEvX7F3`Gpl{nwFvkbSETsC|_2RZf0 z-+^)TzfXriY>IjcSk+&=)8Du(8jq@_pi~z&uL|`(UlCUSZAuq*;wmWBwBG?a^&6Sk zH^(fo+UVf5S9IIxau$j?JS3jvk)y?2IXtbJVI4z^?^*iKX6eN+?HLNEaGpN4NFBfZ^um?z~ z{}dVMH_ECh^9)>OQ>Hj$cPBw7v0=HS&iL)OzMsHnH7or^*4aba!W0I7UHK473J9g`A za}Lu935WNCrI?-UZpOl9o`(?+USZZW7X|M&+;{dui>Nib4S~WPRW&O zqdnK!C$s(+d+#2X^!fge+qTxWR_@eXv-I8CYT1%)rluyWwQ6Z-rAQu-mP_-1W_U;t zX=}}G9o#B2k8}bR1U#P*XlrT*m4XmO!OD_Ag-i|#BES2EwNLxJ-=E%xPkbMb@Ao(V ziAN83y{`MZpZ9gXuPfmb85S6Fo@cS3qWe0~F8_8v)qhdWmiUCd%l{Y`XQ}-?B@lQ@ ziJv=vXqEv`!q&#WmQS3>Nx`x~}zK9)uHL{agwJ+FxHs?R+(f%)L zZlWsCf6#%79L@hAK;V!kOz@;Ooa$K{bBb54HB%`o7k^lcGAf1XCwbLcPd|S{K9#+S zgjaq=-E_(cV>~L<H4mSm83mi4Ot>aMaZU zNae8K2zO3iTizM@cE90TX(QWpB`GlWWp((wmf!gUFzz`mZ+n4%JUP_6VKwjiYTo5` zf|(&S)G|{v%(` z$?1h2b%xxm=GxVk+SSwe9GGKillJCHkYs)Cet_-Fi@*qAM!kdE#}5Y-o`WbPHEJ9D zhKCMS2Tqk4x-QMrLCSlKC?xFeqUEw*5t<|y%jo4=jR zhGMpP4lroHD3sfu+>s;rkh(A_Y#=zSA!!9*pi@{jd(~(+#jjqfqawAJg%i(38Ql%t zqf`y7^M=_E<_ZE^Uhb^h6Qy51m$JlmM}K%4Z1obUw#{K+2c^$1wyW@_$bVQ5uE-xM z99ZF7qoo0a%6|`#kiow9%SKFog?AW!5j55{)*$xV#O_Lyp@6i#K}*khV|ZM@ych^dl#3>f-dohCL$Fo*#9_6Wa@PG~Fas zrWI2i{gjew2~rnC)lwSJW-^tVk_g~TOWPegCF)Wi!-9P|QarMRDxWG968)5v7(enh zLJCknmJ8Y|M=T{tuI6nk%!<)y_D7_Umv=6;otu#+ZUFYv=Y?`dzY0l>e{6SQx8YP; zpFzh(+Z~YTWyi)joSM$remEO!R(F8@`gT~iPAF^l^^@3rH&+U_7QoPb1m2}gjrRz>pBel_yik_a%(HWZ?~LL8VU zBzHv`P6`uE>C|wrC2n;CFr@o*Y^V}}U+!qR$%Ya~{{BjdTzXbGv3xH})R-*~_Op;U zHztt*pCwST+chU(>0j!9AFuqaWXRB7}h-iz+dJuePsjjb_rEG+{;k2#%9yCWc~^+ zuQpsA)ZBD3YjF^9lhU%Hm<4H~sp5ppb>q&t9#?+5Pqn2wS>CFWJ@{_B(S2m2 zL$002#y_hess%wb%u0$6kr>EP_TBIJpBu|rRhGW>%_zwa+rHSfYvhA>cfIAVKej-V ze>QY$vijAk1O2mRzx>Ke=d{OWfAYz@M=>|ces~41+4{l)wZ+g;!Pq;|RTL`d#rc}d zSzI_lkmiL9wryUx2LF_!Y5s_MY$5kyVkDFlwkKih>@!Y9Fk`OE)yM zAsEw@ewHboc9Ym6^GkPb6w0_LK$Mc!eb1h*{lueJJ3riXCqUdR_f^+0aK=L}B&WXn zjFktte~d!)2ZLUvsXxw#2ED!u8rZrX5rch&YPcMN& z9W%eM>nYHb(_o)1)lyb=#}5p&s_IO)5WxvV!X8R%4eZEZY7V)e z+pOL?fT8BpD0vmqIx^af%Z2mhSxjj?8ATQB4o5mW?^OIDg0Fy{^BmV3&yv9x;*8I& z{#?ZAsfZLUmw z-*YLD%hBCKO-&3x5b!80=?{j(OHn^Agu;?Dzpy*NoGhN9_b_D#n;Zfmf*dk>H5_F{ z#9cm0N$G9(Feh%erwiMCaXDq(xI9yQo@UDMoeL=dZ<5D3!0Y`ef_h`~D?3YV$=7r% zaHnHi3Ea)O(Ofn+qi!XvT&SoWP3+>A3kUdY+{5Br9^KCwzu4Rt_hwd&pL3EUH@psK z06g7S;DjB$x{|K-!kr&r+NHTX)&Zx$*qyrJHx@%%G;ae4z~!~z+_`1%o6RO^4!Uq} zy9=o$RSQCDYgvG5M;Tj6v+SS>q84+Bb#nz^Z8c1I>2p2#j~$Azeae8FEdRgT+kbG< zpuoYCVgdwcsUSc*O8@;;C_p>&3%iRUnOodRV++qJW!Ro0q>QsSnX!Z=S6NeOR%)k$ zbEoAbwLc@3csT9^tin_f7sP9TNHwGCkbH0mqZa#CN)KhDl;gc20+wCcGT&PSIhlM4 z#L3oyQP-o*U;OuT&U!*fJ>vKJf~bOqN3iv)$&}mQECRux`4>{0OgLFYUVMb8c z?Odr2fp~jg;Nmf835gvt^(@0)#~;TsARHVsU3D>&O!5WdX7kQmcz3h0R7{&{a3OWz zQWbHIyb3>Oa`EULSu?N3k8;g}IPl8OzfZ^Z$H(Q_om%2yipCf=Xsdm5nDAT8NWHtgF9)=zW#tDfiy`6GU3gJU>Fu(ID7u%A)}VFgi^ zc5gGvSX3{&J0E3?Z|4N2#&hF!6)t9CcW)3-8$Y{{ic<9wvB%WrT=;Z-f&2qp5WEqb zO%cvme9zT%(SI>g#0GO7ie|{R1S{FxG(7f8JkQU< zOV!d_US^s8fe8n^{SNzn15u5KFIse z#K;ymkmd}FQ^Ieu{5q^@&KbdDm(Yq05jjv^H~#ul=l;)fqj@>pbJE=Q5TNkV4D-n9 z{n()5Hbby-fGFivhLr{ay%(A;3GVyYye1DY+W13U&~$n(e8|J-ET(Zuxo~|w=iGBW z%{CO$Q>imTdP87BTOF!klBa)B>^gF7?$rPlk&!xKbTOM0Y4u?Z3cyJw2Ub>p+l`?e zYY=A#7e4Sn8Ta>%3^I_$b_m^S*98~5|8U-p`=_5v!<_rG0RBVaS?z{Ly5+xHNKhOO z%$a*6pz$**5{!FMAEhhj3O+0bK)w;VRUaaDJ@Vk&YUS*ZVeJiZ5};3)Vu+LRE&!qT zCh4=fc}Zw$4xcG!NWU|b3)}mm?U^3RRp3B2$eIUczjiC+k97C5$3Z{5C2C8cfjvWM zzzn7K%5QV853UhU3CZ&{V-!j@+oamo4?cyeQ;07w8|- zT0(Q->Uz`FS9T5qeWLKa6$+BTbHE_F|H!Z~+n)zNf5KQRj(ltkH%oL>G^! zT8EP|p7aojysp;}QkYVxy2h|<<5v1nF5PoNld|eoSb$NH{wAcU2etuCEs|};T_poF z_ot=HF3t@-24VmRSVB^&x|}z$Mh&1PI4QiL!3QG zn)7;EdWo?}HOvsu;S~jP^Ink?it3ecPJ!;z@=(UL9p{B^L`8XK3)lQ=d$|IK?X!1wAYucd+h)S1f1;0=`YRz)mH5Dl7Yj-$l{yg=|tTI1*&({>- z!jPH3)a`8TcEf=h$Z5tJm7}^Zx`5JNk7KPp(d`>uz%bd5_Ila%9Wn2B&(o(>iICA|U zkct-ozmRn+O%vZ%=dk9tmIlYZSg1b1z?jatkP=Ik;S8KffHFqZ-sunHw@8LHQkS zJ!A^SxL8f0tTAbf-|_bs2S0QFujP!DbvfvSwl9H*D~?0!!O;M$yTT5RonNRPq>~;V z5<7!%0GArMgM?4*i_4SK2$8w)cBD35To~Nau?Jk`=9c5M0P0rg%LAn~T_Va7X;`T_>c!~U1wI@SaETg4|a*%bf z=};X*98cyI-v2VTQqQeAlM4vMa7 zN?9H!`M~?XhaH{>@I7+;|1Qu1LIvAF=}#JJ*Y{BA&mW#Xugr=}SV%&f$m+`>(46W( z*oVYuAP&nRx;Xv#>^mZQOmOTp7Zc81SGNiq$vp#Hd)J4yKGkE^Q=J72gtW)*LmfMXU;LbmEjdvr&|_8~c}k|1-4 zX*4iIu?9x8PC*s$IGItMI$CcPI2M5a$HLn8K2x0`*=rS`#rU)wg!6+(g}jzC2@+UEU;P{)tNT>)1=&g& z1CNqAmnyfSIM5_1IvL*tFV$3d=&jV!?Mnm|o+%64>c zg`_Vyw%gS6JOd-mt9e)`UXrmdcc_xdYq(R{lCC?5lG>X}0IBl^BB*AG$YWq~7+C#_ z2*%TK;Q`~jxaD~v&RcOPPz*d*vUWY_nQ-P=8-?{lSFQ1HXH->W0~-q0v&+(Au|Tz) zRQ)^^1`+FoNq$GpChX1C|N3FZx`l;lG(RG{VSrtK`x0F{M(pe6TnK>G`aAO*l&wAC zxwezk1%WJUDrCiiVcqMf0v)`5*x@ZT2m`QyWqlYDIy(N!Q=?ckB7~}QUIqIwN(6SDUNC@AQE?xc zT@Rwme)his;rHx~ZxknM&YAGO26Tml%m6T84lvVj5DY#57Dg!$eSO@h)_whFy|L6f zwK9*coc*^vDnF(V0z^u?%P2=wd&L=x;DZ2MB8=ss9A;>f6?&q$`;S7yXC7;*SN$l~ zZrC$G+}lc!F+i@l!3i}g1u|jeO~7L|ef2?{;p~DzqDHk26*eg6Rd~CaNsyz^b3woo z@_dk@rv8&aw6(8ACZTXw4I#>?uF)i$!a|%5Jg~DQjdHcWKjgiS2`9aYW*Noqz$7B{MX$nGns+A-Jnt;LZk$Mj6Qbtjx{M&B^+PzYn0nF-W$y?agJT#V4b`mIFA%TV zZ0UeJ3lHTuqgE3jGNB3|u0k14mns+2kr|(XU{WmMeB4SK3gGmr0^Vz*#H1Zna7%+9 zUS672E}a80+5i|$hU`w+dY86`8-QgW>q|6^ss@6WZ?=rg5ItxmniE|3DVGy#JdHQ? z#hsuzz)Ev`8XQpJCqQIl$~_Q7SG}Q268}jyQb~RX+z5UmIIGq^PDpk8R$D30*6u36 zHJ*h?XVLY)LZ~hVs15~^qu!+Z>66S|x)vAaU{TgNLZPbrOUu*@>8t3qnjt+0NDmOA zRri;Ggv4U&N&F!$-j?ich%l^q13uUAZ2`o{3xsJ(Esi-%JnK{c;Jd*ADau}M3FnFIliUJ`1_FLK#VF(Bwx)=7X zWCYR)RIBh3SV_HpLTIAL?k0Q2k zdN{TaL1PWd=9W6FNxF7|q4m;gL&YJs3yP=R0r9%lJ(R-8o0kZgF#sQdq%CzY{q5Pi z*}%~*MaRzP+<$$6>Gtys;L8*Th@64d)0sP6Y!!T`Vo;D7M{~G<6CvnXRN&t*m8!Tx zd6kjtLUL`OC6(JUD)nZ184b44)Nf(z$Y~FbZ8HfiO#P1zCvhO*E?ks$LlZJ0?{#4c zVogU?dU5OpvIc@k{XHZj8x`v;-)>-}-F8Of)xKyH=YfFXr3u!%i~p_!h zm18M|)62sHk_fXl6li!UIdPQz>w5J_5WO9$WXM^&Jg1Sm!>F)4k-kM$%nKy1Y`PNWe9HdHdEzyf&NK?4VP*-W9l#`Iz)N?T)EKI6$0m`_5+1H@zH)#++H&Ry zrpaIDE7rR16eJ+|7L2$(#B|p~OfKW`3wZSi24SP_ew=YEx)j#&%Jeo}tk@ov!zjV{ zG6}({yWL65{wCWcvASB4b;3B~xmq>`mF%5fAxtn(Ddqa35Rw*vP#BA_zg1sUPIJh( zP`6_fy8>PrmZ$v$vI%EF!E{@(y4)GS+s`B&Tp6GkrmyZJDe(-_;OyOfhHi+c<{edw zvB6xr-*6=hem<=935v{9;{TNrwB8haBFpnfNlvlxft63($~!4eA=F;=dh#|IP3jrB zwcb&5xKa2~nwZb%sc^0kj!1evLxxYmyPzXy%P^*o;Tp&|#9}#^)=WN%+N54@DAWJq zaB{RQDLD3Rj!ODd=LD!3>+02hTZjD8IUZ7Nwk%gDHuZzFQY{-vaUS0SYx8%LWk*JG zQ|oeoL+n`+K@AgE8~t(oIrw?a9H^MQT%ye%st9Q2cztJAf+O2$zIO1L&a z%mX%{tdW=-U3E>3u5#+j{G_)|yo{RD1jNUyP>I8^f}I7m2>VaVcmBt``^ZsJR$jF| z{k|a^RiI$`phw+cANC#fK>+}buw4Q)dU1W^7X4Kfla2(5OIt&TTP4tJ*E1K8SY^VG z3}?yPX2@w^fb%6IGf!(e=?Zjgec^pooJiO%1J=WfBVT|(#Q1<^H7KS!=<=)!9aox+ zuDbOyrhjJ5F=w1`erK|9N&SwNazpBvqDg?CQ+y|(I)tuXf+5MZeBvkp=1 zr1WzN;Oy|e!;C;s6Etm&J|bA^9iVI%q>7rGvd$CuAfHQX@ZW;?c(i#Fy5F0E24q*q zgq#PZ7qZDbh-~5pcW&!P4PD_V7{I${FA)V2H+tLZjY^nPB$6U3qFG`{eF5W^402o@ zE)bOJ?b|LvJBQ3lkTrxT=g=^o-0`=U`zS|0PQRh8KW8P{Xm;uRT0#nMW#VzU(T!KK z7}8UqqQ%Cre*kQ!$s$@#S^d3iP$}9kjSRt!7xS%=NJK><8c@s*hA7rASkRDQd@nD$ z;TjDlUpfZZOaP@s)x(jI8QX@hsJBulan0@MzT+S__Eb+-hL@geL`?Q*(!^?YV&)JUW+kCOFDC{|7LEA@s!=*LI7wmdGTb*7yw+ z)gy_Qkw6`ry1y&LF5Hkqw}exjPp64}S5||174HZgW7q-K4v)TWX3h_GK|Y(KZY zjp&d+;lU=c_ey^X21#jeE|7n@>jbY3=+Q)eBu?0DAxjk@c0Oaa!}*b z#I2QH$Oa(p8qIMzP&^3=+iuk|c(gmIq?Fv~ItIMu-CSUFW-D9I$9kBqkFSiBEJw$N z^r=>e>o4R$9+1_9nYm^CtgAFfgz0FJqN*zqjk~I)J+F)^p*2)^+*=6ZdU#VID1Ogs zffUC%1XEsl$9ZinxMR6tAoys<+^hOaB*yAOl2RJ?g{V&+0NY3aZQ=HiXc-s0aE!yx zl5~oRS5O0tw{K29F7;Lhu&s5EfYUXRj|UZCdAtAP(H?QVd2B|LH4?8=~|Aai7H z*3W?Ch~yLYEa8!~8%yaoKs+dGKi{P@n$jHaetbEKZJs3Ws&nOsa`k+o15$Idah@_N z>A#HFpF!RBvyRPkw3A|OMn#P$-Qyk|0*QHB-R9w}6W?1fvqt)h>Dq%gT?xs$p#==_ zAY=7;oS{fHkAB0K1#V6C0p=$@eBja^@1cyNcs+NO&0NuWM`HYn4@8%N?!$ms?0i1$e&UZl*~m7eNIs&hrXY$d%9 zl|5n!j`ippL5Z_N{Kr7ma(W`iv%JTWzM{~010=RJcP5}R;Zmm^%u9J8AbN4w2~;O~ zf)s@MJ)62j9K9D|}^OAZ~p!jSC z2PEZ(sycbqVFg&GkUU&pC|92>1R}~!3gwIr=9EukL162ej2V^(!V2v@D0VsgLRhvE z_pRYx`*#wx!G;yt7vcMdRi1PuzCdIM?$f%8H(SUft5WNwG&eD*ivj^o7|TvC1?85{ z_js7jtFFsCl@Y*y?WCoD5Oxk+JOB|#V&6}NEx@U*&GD_5QrTrfg-+^A4J);50~f)r zO`j`>+~Fx9qV3s2KGE4&GgJOZG0979P*gQ?IK9jXi=?0f4P|M!WuQDoiEpOz>ZON# zg89Kzf%24e$cI0x0|cyV7~1E1JvcP)w?vGiqF3vCb3*)epiDbSo+g@ol)N_>H2#XR zaEy&{g|vn?5=b0T21$GC0uU&$EPE1C97R~t3##rk($Z;Lme`FHD(B6kVMnrzQ;Fp< zAdOj$7fw4w8wSzZDQ3ZQ8zSUtHgaMKALLL z<3Q2t6*;(FDdSw8do^-E`&;>H1}Ty7dcS9gDCd`9{mbGvLD_qdi2X{{%c8(HKrMWG zWEl4hAxlT5=b@Dv$Fn}@hVGO!yMRAhcsy%OYgf6(oB0uFXRyCP z4-h%9A#&@fQDggQk@mc}m@Yc4Zw0aPvwqHlLqPdDiI-(&F3i8Pod(O%T;GjnTR;U~5(}27oobF_wU?pfG9a zu7&b|^sr%1_Aw4FHCu&1i8#*(B;g9DNe#3>U|NVkQaP_j^F5-0;TkOBxX6zS$|6ma zeEDYWJeLEZHe84fU}$y{TE-9iDTyH@57rxNK{ZkpNGxw~-ueqEAvfBCz^<6CFrH8? z7!@$j6G0uf1=LKtGM&qp!RSKlzy=A*Q!y7@Xqt+h{jom?QsD>2gKZI{+EJ^-q|#sn zj%t?pKFy(xYN^L*kE^0UIV**+`Y=X2w!l*pPG9ZGO$1K#?L2S`B~R#p&hw@Fkp(P* zpDqbpt+*;LGyb3grYEsf9X6}uxXpt5J6e>BDYW$deBQ+9xjCrwpw?u%u>OXKZYZDM zX$X#W;@lr!nb>=YvD!y=lZ{S_#9acV1ztzw+QhF#K6kkFY;djA=W|JjU5)LMgY$nw z0CZl-=h>;xm7*Ki8$aRN-HoAZ2da;%Ll@JP#aY5FX*YU*I^lcXZ`?ywCHYycO9O$IKOZWr*u|F%BnVUv3lz|Hyx^zC+O3cy)i0jS!fX;fIgH83HM(0j9 zkw(R0KI;*ows02g_U@I}5Hj6KpY(6AXUVxzkRUrq6=XOf@lCkY%QOcOwRfbN6I?Wc z4RK(szO;VWOWYU2c@PSW(T#D^O%zP=T~I71(8N==%_<55N2y z&NaKs!L_<$q>*bY57gPN?tc-uS6>o}#J-b$V=hB^XtyQ=*XM|`P@#}%15R6}(gU{( zB!Z9V&#qjVSZrT04Jw8s2#6pgf;A#74OWi#8Pq{K<1fHh(U&WWVY9lSE5d4X$$PfA z8FP|@vNRO%K2y^vW$?3vgK;O|&Bog*D!54m&OIZC(Rc-9(*sO5-4`@;jnWmU*kk^q zJ`B=6iwmj5)KiOwR~HNxL;7}rV^0d4T^VAtx+1&CYw(uJkz8>U1ID1GYnqYnl5muB z5GhJ`IZmP}(ykQ8g!JECMK_^7I9!f}%u>p5Fw}ATRA6Qddx=*m)y=);YEY>f7|PZaB91x&g@nCU7`a!UAtsmeXa&<#TG#FN zjI4U24%|shyWt#SVcuEhyiLY29(L)BA1I7`u%`GILwV!wO{jv~ZxlpU>W+zd#!X@v zSjs?ZAmnxAl>%>egq~Dp=V806!WUQdWT$`R*fdw{iOwD|(=`<=2F>A@mP)^xck5!0 zlB>M2zGsX)RiDfBuf>c3hHz@*s9OAzyvEq%V)h(4r#!@?pqPEyWBWxuFVS+EI%&@K zK$#_Mv03HIf$-Zk3?(L?20P>3v>YWm{EGaHcZx{9*)5FF!fL3<0s*IY%d^p!Rv!~q) zqANG>l=$!%Nw8+?tOTU(HrRdQrYciM1T~#$SIn^>88tGO(ZFonZ2^}ZA_zbiOrMsU zS9b2rL&AP2$chN3`-C7cz_n=dr5b-yVd$g>FN$S!MTl~|xX&2?q}+nC<>eN;?h#)) zC{%GcjT>CN%R{CUN730VIse>1xbseJ6DgGqJi3(m4L8OaY^J=*`z=ifxFyh|%0(6U ztCAR4mQye(xlZF&D9<0|&`W_Aq*U^iL(OLi0q$JK-gXeKwp}gt&PyAK-+)@TuLD^; zYC+_8b_tR>sDTKiX1}h$rC5}zRjD$>2DEd;;%3lz{DLt@?0_W@j(WK9l4!~C;4;f#-pkw zVsx-`7!r2oxdUfyr#5=rlCVh4+bIfF5W}(^30rHT#St9xO=$ZHr)|DG0s;8dx5f78 z0W=0^lg}oFDwbFqusR(GukMU%ue{UegVuy25h49mM$iIgu$3IgM{qf}w64LLU|Ich z2R8h#t)EPFy7+(UI0D8J?zBMG|Ey|Y1wQj?HSl;n)@o%EpyxmHq6_f+5x&iey1>&f zPTQ_i!~YA!4>t`=er*d^95APcY&+`|40|=(?$Y5S;w5r|&j{9wE>AUg-Z5#0w^mfl0*A z{yH~c&i?^^my-}!!w`%o==>bB@ywJ z$$2GsjS`K8F3DwyWQqyWp5~=cFR5dKWS4)`O!a^21Z$nemkrJOO3y3_iq${5WoN~l z54qLdGCc7-t29d%HF)V`aCSfNkdclJWYc`MgPy3jCl-|o|FgxlE4*@}XX=ii#R0)M zj*b{VISA(2J@_D}Svx{RYy*|De=GQ3?4xyxn74I|+~9bTU(MqVWZ-rBfk`*X+FM|0 z?WC?tHeXtLqn>dLaf@(sIsAS2tI)vqUAEh^Ar)s4`J(OzANYM?jYjtcw(kY;v=(hsYD6Qc^yk0I}2^u z8wF}_<^0vMqf+-D13&}VN0Q)QXwH24SF6z41B8}WEGu^2YHO=uuq3+!$r0(TO<1q# zGsDRg^I%>HNw6H$)2ym)&+5K}RjBbniYzK7=k#rzA_G~({iDI_KkF|T{m{B++rgf7 z4GYAAE2nlqEv}(g?p!1(ZW<~)hPY3d$hK_F{1A*|to@G0=^8T$pmr!rqxW+&LjIlA z&4}bLq1CZHB`(Zw9(C~rFZ=k|xF*$-jA>B`~Poy}*gqk&`Z(tNtc zh=LS{ILUj_-f&U*YVzOr(y?~>{`)4nMc>W<5zIZUIVMPRVIF*9QKx`Do$d#-nPd15 zmfyJqf&TSEH8h!7c1T+HLq0UVcF*7U7yrlNrb?)emW0V{4u<;Fg=qO6uasP zG>#P^@H&ieJ-dBC_xjkZe1Q5dk?92~l6>`Nqu#%-xX?>W{kPu)gjC-z(M8?spE$~* zNjHTPY3i&bC+n7d1flzeAd;X*1^)d0=69==u5Pl}V*_{~#yX>VG~5F+sR&iX$JJqaxU~fL_FPP|VZ=Vwl98gn_+pzj!rM>xA2}iQUeA$#Wz-bt^ zkKZfH$MeaVZGjxq;GO0@9W>oqUYn>HG|XQf6FuqWHhiz+BScIzsVMKBq9vid(|0)~ z>zJ20!i{+pV1}3KFSuN43g8$kH)aLYPfZoQ3dlIH0hmionoRDTo#Le$?b5MtAbi2~ zmwkLEbZWOC`=&cmO@hR!`@)d1Xb_+E5WUu$b+|1dOfVQ9;adv&_9JV2Jp5y#zXI=f z?^O4ERGfe4UPpxhs0A$z9*U<2ZH=MoS{8LeKPH5S;$xrW4$6ltAbb+R~@+$sAqO4-26`OsgnH6;?QVMLuy;T*J5ncsotfUMOo^>_>OL?cF~uYO1OjM ze&wb++`BYD;%tDw=trku9cQfYUc-ZxmGF`E-wTSI24DNPg3$x~`6b3GbSU zqWn(4nouv8>E7>(yUq^}=9Nt4<}_(5&W9lY_vg9GJe{VOAl}j}^&DC_835!$^}`vt zu*V7yZDoRL32;S(fh6N4SKskIvIo2e^x^kLiWWyT$8SkLmg10KpmrIR{x-jef2W^a znwlZZ^Yvzx>uZ*iCeLG&=9SWPIRDOqHE!4L4+%Ql`8Q4#xF<)S`o*ngao4p^_eJXS ztPpj`*a)8wcml*DBpK}5i};t^vB#YXFzQVfoqdE+DIon$jaiHw_!w#uLlVh$HAh|G zv?EeK#un3be#2{+N#SwbzvnLmBAQc_ zKZcL4Z_ii=j;Y1EJN3Ox@OoLG6m@9vIKF!9k!);S2z+&2u>c6ibJ_>i5wDO_-pbs<0EzaNfosk!K`DdoSmw;R^Hp%~*X| zq4~uO-6hzMA0_-&QdXCt8mnvWAG93Co+UdMc7MP`50~R& zdw-5?Ymc5bc)6MK-D<)`M{+u+H27$}TTFC1ZPKaGFra7|iD^iQ?7}8@iloBoP1bW) zc-igY?lN~E<9>YvpvGKRrmJjd2|B`5to%wk<~02kB1KsWAO9+^#Hha{=Suv(%ozj+ZN&-4u~OcjHvQq{GR|D_C<-vhyAea`ofb{%{i$Hv@;wa~T zESpX9`8CD!%k=MWT2JNZyVdhIi8l>sIThEyTS&jJBAw4vZL{27l2@W#kaNtmz7SVn z0+Z)l$ynw-Yb8}8WMbz{ZEyEOwc z`uM&Ll8ntoXIEw1Hx&~b#$&ot5zEQPAqXt~dYtkLg?KAPvj_ciYI02fYl!828V@c0 zQ^m{R&>Xlleru)EgLRNU^U36rzRr~Fc$n{<^G^Gf%Pr>AJSS7AJxBW;Z0bke_EbR@YpN^|^2#qzQ}DpjVjDc+ zOYIHcgvT{ww-}D3-P_@r-Zl@QQ`T^ga>6SRk}!){z{uo*&0}Sx5pDsW@1`iF^tAfz z(SoW(k*ChJZYu5<$L+%tm)a#LqRhU5>ht4x%@WbOi^AU=O>anvE;_O5boZ^Ef<1gr z9X7@lf2?~?fu%}!JxfB2d?RL0wiN#`=1?7owl#~9h9Bp_mr@$t5;rqTDJPDKiyjm4 zu@`wrCGZ6n@tc%fs)0Pj0-fh@nNyd>}cB??CKvY8>mg;@-PY==ShOpTW0*WS>sI`uMF6 zq+NW8c|h0fCdm|~qKizq+Rp&cl{6J=Yx?&}^>xPb|D`UzpovXS&H5^m zJW9jEin^(*ZZPjnCH&}ClOt2P4Hk_Y0G~RI&|cN9>$(O{qmLh7v!2*y_3`(GlQzNu zQeOv1Z4WR%LaJTj6>B^}!CsRN5YC#sG2B&T8k{-<$XNVEs>5mZ3nJOh=B2g6ztvwz z8krQo*KkW^XfYz4r{HMvuRtzl+=O-x)$b_XBkG4l^!?byG~FJ5(WS}W_hO?(k_T@NwvmP2ibP1Lj`%%IYi*K@Hxu({iIcFw$hV?W3$IZ5~Cx32usRS z&cDujp7S%WJl%^QcK7YY#a@IH!|J`ArEBkYG-=qR@yjYNNwJr~r0YXR$o;)h>^czr8M{@%`R&5u4yD{{tPn4#|Z)8j{d`C;$(|33N639sdQ30SY8!)>D}vIPkSj*y?Tao$JgJRx_G7N%{9J? zmjc=sK4*99ef5z_c-NaVuYGmrV(VJ>l|7wx4absHm@f9)SI#CsjVCWwc3IDa*Xl0& zq7JWbMEGO3Qi_(?*#+F`dTDrYEo9&W@2VtRF~hpMTc6Ilp5Lu7S@gusU-2}Sj5Hy$f;9U8oK9VR`c8OQcjD{vf5N(>k60)CCu<$DuJxm% zK1NYm?&SZ8W`I)=ZrvaUMm=@|ADK4KmA=@@=zhIRTwPuu=V#m2|1D|p)X zK0fdZ9};z6Oa2LQ4<8YSe(LvTeRSKpuW7AM<46zQe`01|6AGVB+y|s60T77w{{+LE zAIV|dKf&%fJi%FdVPZ}Xmu|MmlaVphQ3%fkHAZ!V&zI1K@fvX5#S zC_0eo0X|#*61Ezb$%<~hMy>gH`Ht#Zh2n+mkBWcv`t_`z;;R5VyN}S#md=+i$3%h8 z(BH)usY$}W9{(3}?CkzR&R@ssFQM?4#QaO0{H1IEvK0O@qyDln|Nm$qUw7+%Ytgxs zw})QvKc$~4y7e>F?UUtef|YAKUbZ`M&(Fa3nl+gKKHK|7rX}$-GSXNfc$72sDR7+`D^p7R;J3(ye)Drh5Nvy10E{Z>{leUSyN8-#F?Z-bZEq89H5E6>1Ldg~nj3>kRak_OnL5o7SiO~>iarqMy^u6;8p|<%Yq0);u$1&tu<4xnJZ(`=zo*Sw$2X=%n!zezy ziGgSW3lJJduep|BxEG8fl_6;=9?dF=Bu}@iC|r#W5T-wVUc3oo!7wXzm}QvAa^+6Z#Z)8SGlVQ)s=i;DWWNABl~3$fl|+;yhPKRr6(8hBzKZ^VgV;O|e9?menM+vFPiClV^IP{qT%1H-d7F+7P^8kq2h#w0B1 zQfm#EBw$g1m46S$^3oiBz+`1tD>p>tVL$WWntJAmI6p6Do9i60UwfF{#P`Vh-+ILZ zAutBnGS?LVi!^pstiZM?RM`HB#NCm>)|B<*_f}?*NwoyS+yS2PIAA0qlyB}l9D`=> z7jr}MXf%gxFxe}7lj!g@S5k})6rIRvqkXUN!1^JJ28>tTn0JC+`xuITr&fHm4uo> zvL*~;g!Y`9Jy|)X|mi>gPaMS8P36eET zM|BOsq<1mASAL_ z``dsxla~HTcwUSUamd_Tdvwv{)LL)6xmSz=fFfrq0Td(6_As^7Pjt9DGD9Gy0=7uN z0W^vZgQJ{vDc6%1E-j9PWFC&<`p24~4|uC&lI&PINa1Aj8wpWUP>+DuwFTnyLt5Nk0t&&A({t5 zA`?DjT!~KS@2w^DYrn1La}6pXL#=CuO-NHbL2Id#m}xTr2fIRWFHhBgjU-Q{jf$-0 zt&geP$;!*@Q>8huB1r01Mu;SGO_+HKoP)z`fMX#R-Hz4PDQOPJk$loUZXz9?H^2}m zox~h$7&B1>-~IS8)_;c@_t!R)bnf6`zC*v=P&Gh8Z)yRMri4ip+q8>~(|Z~sq5bZk zFp>8GF`VVV3L{;IWcVbcERs#Nf7;&rI%u;R0RV4+trL__YXJsxMF4sAgJ80=5zg>; z=wd~tm~W`&z!ovp504X>iRsEcseSs`Kf&+%aI0YlCV6a6>HyfXX~n!v41P1vGlk05 zuVI`|`91ND=1#5@wD2@$IvMYr7smv`GKp8xS@LF*^iONlHs9W(lLL*!8w9kFB}r^P zknjzhtaZH%bzQSI-B3P&;1I^Z0b4qZpZd)>ItSaLQBi6G9iM#8f$)8hOsAd~J8r2% z)D=XQrC#b4i@|fV#=-IMXYdQbQ5G%+GlO#t;E41QB^oe1(1$~~IS6AP+q6%~^!f@#Xopx8XFPOd= zcvCUi>c9>o{gD3QpV-#805m@2JXp*iPdONNm=A-Aek?>7sly+9x2^TO*mpRGLIThG z?NALva#O<}`hF!kFYzUqRvN-?2KXM3#iW+QDg0)jgTE4EuxYi%Jdt8x8vi6K?B-vC zY=qnKynQ;Fh;s~B1Kd%6i0-@rqOObGC;{+P*Aa3YkA-N=)kKyYvIW);vj>!r&<-;< z!CIzARO1350-xfRF?PBnY8=cRc6$Ju}budi(=-vMGD7dzEWl>$-OqCuc%I z0@hrdl(&nV2@zDR#c|0*?mWfQDl%}kFVAj75F#ZJUmg_OW(Cb<7DuYCg%CIXkyP(S z-%2wwaA2YC>>6;DE$yiNQq)I&3nF*Kwffd#mOuD1;t$};*p<$*nZP-w-A3J=4&M?dxQHAi^SecyOL59WYv=eKBqOoHo%(2v#1YUv~Y>2T$zjvPUzzg-Rzq_EVz)B z`-*n1g8R17c6v+7kKkjcwt8|Y9%MV`beF=#`QrCA2^r3H(^PThY1xl#Zi$81Dn z*K`{QTya5KYjn8-Eaal^5dI$6#c*r|UtHr1^5Zo4dn1I3-Vmn>6$x1i@Ui?8y1W7pC~|a z2II;h#I5HWu7A(2SaDi^weo_z(%+|7rTWjSoqZotvOY&am#tI2$9vFS2`kT6)>Lj! z*%X(fr;>}&tecf2;+-FYWdk5^t(>aWwo~n`#V9ra zCmZjRi{Va@m3MBfEz>}PGypg^ytuq5VZnQG(rD%Ah2n%eb6wNBr^~}Ht)U}o@67#7 z?4;>grjNE$XitC&ImscsgY2-j!O0L#m$g|YyWUAr*l}^!Ug_VrN4e3WoN;a@*CM69 zmtXw-(9ySnGc2=BstPD;&Y|+i`yM!w#;OaMnDO6niMLLMFa2p8=KUrO zGr1fl*E#gIr$cX2c2=x5U0l-w$F+L-4}bc{3#vqho3yK8gO;=*lejO_+i`IVlc>VT znYTjFP*`W7`~eT4Xag4o5 zN|mc$EFC!*z&Pkr;IrW2KZ--yS;EX;VV}e!jUFQ2l z=5{)AcO|4n0cF`Jo$R$&SJXppo=r!*PDQF%`!^ld*bz7x* zYJ>}4Yv9@ijC&$A+_?-3hH^%rA3o^q;Wi@{swdd|DVa`u8;>(>Z9GizRooj{D(ay+ zP~qMGnBq6hbfSE6FU=|nG%}Ky%xnG6Z0PSd@t@;;dLIUw92$7dz9?*%w@X#kmj=U+ zVyHU`E2&2}pq{AG+Qaf0HYKGfBxE#(;^@u8T@SCf7#a7jX?4V`o#%=>X!4)MU!xOu z4~nA0Atb4<6>x}7Bdq0O+6_Z@~!@MWqrFU*tH z{M>?g<;CONc^3q|n7fi~0y9_%OBi(QUBBXzkBBJX6|Af~`gHnx^*>FjTxZSqnKWi< z-jZ!Val=?d)0CQ)4~CU&)Qfmok>QCVE@Vbdi5GL)GSQg!cAO7=B!Tpg255GLDr|A8Xkk^e;hHDUN^j#$G`N+^6b|fK z(?rL2M1FNkk!57MTeKf%?tYLzC`SomzFjh9DtDYC&Oh-ej_L?~cZOx&z36%%Ac5k7 z?8n}B&e`fpD%&`@&C#Zh{`qz-6iWEi@>hApS8;z=nBP( zPX>!{N#R8-Y(U|f_q8gcy$xk43o&gEDp}poQ+s3c+`Ft5o;$pnb?|`~*A*oX1XPAa z6vJFUPR`h3cP}opqD9dr5Gi)pkX+3%m<;+zAqaX=J*RM8uLBH-Ge3?*q=;Cm+H<`5nGeyy7 zArqE-YPSEZ9~|WbnQu`qZdP~1WaRLK9hQ6vktfw^vbUkkT&g0A>s)Bn_H6&yX}*&)Dt_r$Df(@FSS3&3EUr@MX5lGt#ub%Uc z8E{;!BGu=JmL?iAT44dI|NjX=gbA{1V;w{TO;XLn95E_G! z7a0s{KmPSjM4@4umAu58NSw}$h0n$tEs3zbJ}F|FC3tMBF3!0xs^X|sX%9x#Aw7Sr zqdzpV%tKg~y*ziPJMSMA`tP^*+mnCAmg!D1V1ZExmL%sI5>^h))!fIy;coh#t>R_~ zGpn`{N6w@e7;~?OHoElhM42iwE-bWlK&DaVHU1+>l2d%6=>;2~02MWM-%_+%M;qY` zdk~p1i0-AHD`p3A$oc0A16nLJeP2v1JS4QMnl!p-YV?UArlo@791E_<1%BVc)S9h` z-kb1vnjR9oM*mP>#P#($jK_r+6%N3AsrIewxx8b9wOg;9N)WR>BrhMmn9q?rsJ0ps znUwdqv*87ZhXn4S(}inkx}9-8k-3X!*)DoIeW)ub|DXW=aogWqb@6^lXmdiwqy_Hy zVn64pEmo3}om=ZadqbPEnd>I6sFrrjC{39Epgt`bj!ImOHl*28{6CVI<9>+1ltD+r z`W*uOYt4_fk;?UFC|1NF@|#)1H^lte<3x4z#>MLn?ZbtgKQoSxc#oz`?t-BS*Kjk) zms^-554JxMC;YzV!CO-OdaA*3-i7ek`^b>yP*vqQe1D_{Rii^HKllQB;c*Pju5vZS zHXrJRDmT>bQ|=AP*y^04o}iTGpV1MMd9bsxqucxNR3*7yDwc0fjN@L&ch?)ZH4^K%1uTRg_>1h3Vr@mY#Iq z)ema4U#2^ghN&Cf=^7hl;B|^|7WcJ~g_Zo$5D)R;rRw9WD%JSpL-!sTQ0XL>`leV1M519UqAde^ZT19ZZA&&m+MavyGo$u61c zZX^y)V%tU5`Y`S1Tr|h(8{!|+ zKW4%b;_1)L=j5rkROS9|l)oQ8f1=acn-0CGMag_FtXJ8GJ=|NLyO(3q?$&9e7XfO~ zgbx^FxgNCXMyaX}{-pjpmYlS(-ip ziOBP%E(F}mlM9|W=LtM2VluxrKx+mHv!=W+wxOUw^t^JRU zvF&h+iqeTpPv(bxH2V?vboYj5?lo5Fi=`nk2B!f+kJ(6cSnA(1=FKH#6!R|3hnQd$ z$1~Ja)Ev{{M8CHnSBMvJE$*9Xzo|QiNS9^T_;gyLE8>?T5v(5Q0Ud^yuvRp8ugE~R zS4!q>dH1s^?#L+jd4v)TGx*}`LV7lA(t9~Nv|a46hYolCu0{XcI{zl0;VjrSQ^lFJp3Gw}wUQGu0pJU~WL zBD6bu9wyoklQ?da$@?)2ihFW$Dpz?ZmLa2qQ+bFHM49s=xZ3!2(&szB$nWWi^k#8y zp{`Zc?4j$8H&xWfswX+;ao2M@$-OHgm8-$0)%Y*t7gWRL;h$0uj3~I4@%7Y1F!duU z!-1vDOG||%CgvMfZTsL3R;yWgx6LevBJuLI)Cg-scem|IK?m| zrDXMn*P8J=t{FPz^PHB6`$nJ__lanFG`OS>K)AmwI@t9`CKjd8uX{Kj^cmW{>&C-Y zPdbA7o^kc@n@(3dl}dlqLJ&+Go8@>sxnR znl2AzDL)&5NTJR{;-As9qU&sF5&o)_OqotuxRGz>b*3+0cC;k05k8=Wt3K(%YzHxv zv7KlSxl`P-_E^1<5~WK~kQi9I+F_?ZYqeT#gZ6j9i=TKP>?G;8P}ypn&5-MD;qd1o zrgT*8B92OL*3>&h)kDBksYoj6`>wzH$={FNSlz136B4zMuf#e>Q8W@Aq*Xi%5N?OM zw1aM+ZpQvwD}}B3_V71O*^9V1YH;HmU;PX10-chbiejywBs=$Vzhqx8-_NPoJO**b zJ@qzLE_xdVib?D<_j|UG=e`#B3*u%3gUZ_S(WU2t=A(S>#eZD?wbz@+A#Wq7PeR9vH*?Qb94F>8s_WezQVg55-;~QgPhw&+Y_{+#R`jnRtD($wn^wE#NcMuP55Jc> zFqFFDPc$6Sn0JPZ7BfeJ5&}SXX??J1~ZmTfs z&MUEVpZoYIXX`m>m5}}}VslePnH6^evA7RPoJD^ zf$l`*)z&qYxGKBIJsih9ZjGT~LE7=4U})<_9;^E0OzrW4JvHO__&P(0lJ+cjix~A9 ziaoxpCpfqD!G(96OT=XNJB2fy+hh14HD(t0M%e2yu1)tadpK@1-{l~J`Z#@D9Iywb>+N<8`q z8*zszQ8`ErE>P$`8 ziwbD21br$jHn>bK^s!;2C%`9zii!3)zQmsX+NTqUz4yr01X;bV^z9ucw(a|0qZiz?_<}_|@ z7qy&BXXRpZ-rx94k5Xpuq(;58-b2inIjNi!OS%9z z>>;7&-CX^*>M2NA1F}-fdC+HRk*kaIT7tC9PR;%}ii{?wc^~1n7v8thc)KBetg% zdjiz)c?Ir?y!9+DI>dcH?Vvu?)!X>v6|x-&)TO1kG>5W*1^-gOki1k%Omw} z6yF^#BnKX3FWX717Ap48j zB8qwCD>FVyBe{s#eJ90sXgLm-EYh1=7f7SY19`|S6sx{;Ec{wl=d;&)IXON?j5bb& z?g-mK6piDZ_QSS{S=Wq2`7^Cl8D2jiOjvk(ldY{TOgfLU@dx~l)lgO`U+2;iTasjl zl&Bx@*~mlL*4jlkh_%2O@!lyHOl1N#i|U=5ym3Q+)tXjkMa#)gYp0T z=eG6-+&zZkS?U(>EX&&q6_}n3kq9NwEe{UFP{BG@=8yAZbntdsmhUvJ=xT<}qYAr3 z#)F>7&b#qjjK^EYnd1-p${lL1XDI_NFk>k}cf{;geTP9V3Pao2kn53)01U3Dc+6Jd^lQ5nCZli|H+z^|LVfK5&C9kJkEza*uklWehYp( zV=?HJ{AjX^p>x@TEhD?ssXL>jJ3{_~K|1BO|37UYc(Y$Tqbs{Or7Fq{;f2Dzm{gf@ z3EAp3Cxa`5NFd_&1bFp#uh^5~4|`mLlG=8;AGC$sZM7iC3P*8u_i&Jd{v*6SXrFZZ z%Y0d_t9a0BRj{zyMd2Zd1KUGkXPM!a5v7Gp@!s`wBWo@$WwZX>S|P1+rL%z7^qK9N zbgWzrKm_Iuk`=Vbw&;8H{EB|FKBW1Lel7#5-|NHJ6>SA9iFQxlswfekcjdvNsf;mF zqB}V}qEF7H4a_TX`JQdt}VZn6~n)Y~y$ChBP6i**BEkS}gynWDg%KVW8NB*h zah&G6Kh+2QVU!s5fXmqJ$23oW{A7UIFXA#ZW{L#t(`&wSPu*(iOnAbJ$yTW84@6v@ z_;(8&{ClP!v-B}vLZl_AZD~qBrKKh)KFvcl`-O$S8Ie}--`s(+2W`2F^W+d-Z@%pT zLErKdK^b=3u?u=tOP{4HQghyrcBs6h4=cGOga#k>XWiX z=DfR-YmQ-?jb0(jUO)eZo`Xx#y3#di=L2X*3`;?!bq@VE-iIBWQ?@T>b3P`hoU(`-Jvg`N8HqWQZ%EbQGqg-s3K5pvapG{@< zA=AB=bqV+{!ect-X}dR&0(%T=-qc zBQ+T=Q3KMlNHjNxzpBPsQjUwha)J6MEYK?*H}iYAwxwzQ9L{QxYgE!<>J?tVl+%s) z2(7h3IMT?blj~2YV~zsI>F=~RNDBsutJunE z3;RLx>@_`xa%pOvs)U%2>8j0`*BS1)xW~ogc1(;c9ZcD9$Kc0eKg@N{R1{1*Wp&^7 zi+?gkaxP(?wi)|&6iKZ4FZG_3v)gZs2WCS7?PAT*Yk! zuc5OBDWrN6QvH1gmsi0Er6u3D7lxHUI#BF&8EoTR2xWC7!->kKjDLVK&8l!*GX*_8NJ=Lvd||dDI$WlAac9?TkT8V zpB?Yb$DLB`xK$9Ct|)oKPi#*xzh2*VBu*Z*P?e{DwOa2;hb5kif{xDpvza1;Hs7{> zAVriGRk5gJY_CU{bFkcH?<&Vx$@WkA63kB=oF#a4L1F6S-t*DzZSj3t0sj3Q&Cqt2 z!JWJz+myH-B+v090xhBtGj@P z1jQ50EazGkQyVJu@kWvuC41d=U1K)xmES8v2kU(HTMK$paor`hN>gwgR~hZY7aByj*FT6unipzyIwoGlAT5=+*vdB8_HW=YY`! z##WQq(IVt=7Sp>3dA1OY*s$imvBl<$mUsZ}PKCs2M)u?Er;Vq@tY3DVCJua@I*jMm zLCxv6mwjFErw`+!y-;cCdBL>t2C-zlJl1++s)C_Rx=#%Vh)c!($JzRk&v7T_CE{4S zLA59|J8|n*Gs3%zjTg@jc%sA98O5<~i(xzWJf~oqN8gS?yhtw&I`{Yy$cahc*CKE? zzY07v(~U`eVc%5rP@C@r2BQzDEnT7QhOc>Qrwdip50v}CdBNd^$`XfAQHTtbz2cV! zUOStoMab$sUH&XKNYu`oQ;+`P$2|03u;a{QJX$lF+vkc;84pfNhw}>D{q{Z=y}>Fi zZHwsM!JC;lNem|i-QD>oj8!{bx5teeCkXuvVg4mFK zB3r#0218-mt@1w0&H&sdE6XZ0wcwvoI*||RvSH;4YH3|X10#LB8@1mH>ulM8QbU%W z)G{BtQvi@b$=m=}C|b7HpYgob-zZ7@Yi&r2Jk8fM?Uz?l0*FR?NToMw;9jZQtfCs< zD|6tPIphR;WyHNa_&Vf7?4-H+^KnupyEmd7Ko0p=C$=eEjws@SdF%!6Ct;V@ZN@xL z#klx`XQWu0Y&8GWI<@|T?CyOq~H zxH!fi4Ny+BuUrGLLbfCJ`I*Lh5As{8f( z8*ToBfIXK3BjeEaj*mZn1SSp`RFtX)706&PrXG_Y1?xG8$oMG63(Ht?l%i=*-SO`V zUUlj4>*s`e|K}kC|9>mpU8J6u9RNN+kD;7ys85D#Rr;j{OK7Y{l@k!@7{6v~Hbh40 zMh(7@xzua^ymLm-a%o%trDNDL8VNl$!1Jeb;!I$0q8h(u@EpR*V`;}S7nO?f(_!Se zcX@&9I&BV3gAI(@eYH&3Ss1Nf203(Nr;uLZHXh~>{;hFb*ZVTmz6@1U zHXYL{(lV;~bs;#E1tUuHRFNKS<(0xc>K^Qm=^j+LVug>!>#(W1V~sM3uG zRF1I53tGs}#bQ6V#cJOeaxt`giiz9vJlrY*Ys6UpZ-%ju<2+VtRj*O#_>nuF7cI*E z(+={(fl2+HNMZGN1w)bq>XC|md973Ja`^g~zueY(P7i8%3o(6P!BqNgzm%grsI}v@ zq-ns&<-BU0-7kkua(VK#n2vspx!TgU;c|BM^|SP*ed|?a>^4;6aN#V^?7yI($QAP{ z1HkOZv?;f#*IW~$X1~0z)?rO_@1hUhqzrr}Cy@2poG6y=x({=O)ND3gV91a8zdw~Xc zy(W3!G?>;(#0HOQmc`oVyZP;kC1|Xpntgee%iH4;jBfaSz%)y|$#CxF)bhTzc|H`d zbeC$B;2BR+^r8?0>3(Rol*^$p&Edl-)1=7PxNuW+ssUno-!FUTeYJKR6glo_j!qV{w{myUS6Qe4whk{@d|B~CFtanOcZ(i$PJ9dc> zgrei<3_dyA-?jI4^<@wIo4KwW)j)#21;_yAxXm7C1-_j#_cpkuH8WM4bHlmQ&a>R9 zh+T80zCxQZ!}gfa_`8+~E-3FKAcHQnMOc0q&2Ve%8aMI6{qRLp7czCP6nKXQ zp(NGyD53)7ZfjU4>vMuNXxk(*yvs{7cyn^jC(1O|>O z7(BWZ>?7)-;k!0Z={!`%ALVE1w>*cjrt~mmZHLAS zBc?#8rS(hS+Lu;YFpx9Wb~zL#?~~ABt@ROOs2k0=>QVVe33tyUgMW>!i;zloC!j}q z!8*E>ZqyB05PW$n>&?VT(23K>OL|5RKsUaK>is!ZmXEucSR0ai7a24`GNK$(S~ic% zDp&+6?sf?etkjI4o5Huwp+LT&E7y7^TitV=<()i(^0TI8Qo-CYNpGkoujw zVSih_e;#_I=q&V{!8$LvU6{c1IW%Eoo)WxgQLu#f=?*O|2Dal& zYMtMoO0UvfxjNz!L_ykKzT{+iXY`s*`M@3}vO}vydfq~=RXo-Z zo)JZ`fXHz~c!f}GA?{WQwIQL8=L7jLr(W0z#2Xe8qUgShs`l9So-&UZ_ZS0v`l`1x z-+(l;Jm2ZC?pFgV_XULPf9d_TWbS~DCPo=;S#w>bDY~vsmg5o1GPK#ESKh?Yo>vG> z@oPEw_NrJ@=q;k3XuQO8rSA0<0wF*TztMxtl!o7)-on()*rD(q1kY?aSn*@#Nk)ka zqnuzV`EW4B$idAaK|# zt@lsI{NGo1MU(Uy=9`^)Pr$6~2Ax}N3sa9t&2aVPErnIyhIE{H;T!6&#~5D;d;CQG znNr`%S%MLu=rTcv6id7O@P16!4q^heX$~1y$OY~pozXBOUtw4{x3_AljkM7g>~T*A z?Q&7Q#Cs?CPx@MpPJJH_bYFMU+XH}xqEQ`DT=`2C>AMG47Irf0`w$88JGJTy2Z|jq z8J1i4*RNGYN@CjkZlz;5u0G@;AjQlas2bBHAbAz8MQd(yoR7EDgDw_s!#z4R%tygy z=`X9#c)YgbT9&e{q+^>654%d!=$uO)_yffN0K&lVYvnjI6Cl9u9~*?S(-I$8r4eBy z7M4!gzefGm=JUVC`aehFS>{-?*=SpQ{T3u%_R${d>M- z#kf@wqghZ--rBuXS>*B-rr!ZZY~{M01N~Xm1jhLl?+euLh1jN}`9OC1FSv9pKHgiV z)^2P}Qff02RssEspd*foqjq3`T99%Mq24co11R1VOCxBMEs!$u*<{P`=_lV0GcHuF zR_gFw;4cu*P`^Z!C&Y1g^S5ac3Yqejssn^j*a!fvr|0-Ufx5D;A8G+P@{I1Ye5IBA z@K(=uVkwmpsHuFcStNJ^Js8Zsv9pcN_)%BS_bk3%o}ix84n)#dZOH0CE0_1`pBgkM%9fb+!Hm#&L;;&bn9elK&-5v@~DE!imx0VAK+NtxDRVVSTWLv#_SA zryr3O!rc9RjW$nm-UBb_Mj^Ed6u;nr4pjE^w&(F6m+Sn&`HeLj!mi)A$eYSkx#oBy z-6=UC9%ENU;?6a~nG)Vrsusr?`G?iZG)u0m{m4r%ju0($l`JxPnD=YWvgOh>d}FMe zYrA^Q$?{e=vW+wxqNfpI^oJk4_WFf{J%99Z9SDtsyop5n;wUa1NSGg}>Fw3m53=!X)tf9J z@y@1x|M6=w`OIl=Iw#bmy+cJU&1y@HW;xSh-fh4QJC_lU&=9Ch%VC{9FjQ^-hQH^F zBDbX`i%K8VhpSi`YnNYZkPhSShq1>FNEqx(t#FyL2N0!$Ii*I2X{kp%uYAzI2chgg%s7+|TZ3b{E z)X>Cd{H}XVCZYOqmK?;@Mqr*H$)9Y>_K&!kQ?mJG-`WT@jyjr+|^v z^J@q)eng!2Odmh;M%ET>S92Zw2GkJHY#K<6Z5NPU7%8`nWH+7$BxFQCC=p__t31Q9I*FrtM;ds{z4Y4PTHS(4Buq*b)Nc&+S2s0i}Bdw zM0}o3e)qR1(QOgCKOKHCvL(-BIq+`CmLJ&h(^qwWFn_b7=xB}Zd+g!m+4ilfK}h`P zp9zSvtN_<=OC`lfP)$fPVEVYD%5j&u=1()$GQ7@ed3(V!SJu$@#z?hfa6?StlyJGorz<(h_X z#6WN6sp3L`3mySVL_aT7%!*FPRSy?#^1NvrS9>SmHGd*atkB!A_YG<7dUKz0?i`EX z@afD&E(}SAJu1$lCR6FPm&RP>X^v8NPEo&p9sWKKc}Gy_jKYbLx8LNp2D%xaS^C

v@MW`8{hwvQL8zGDwhtq|tBAY4cI#UE z*x%BAqt~wSKydcw-qdOU88O0W(6bo{u@6%;wf8i10F~^b19>)Ps~3>zf34Xa zOLBgVs^`5`l8|rSy<%SQ2i$mk7-;ZJ+Ff$^d?ah*if3(AZ16@vf4HFq+oD~XS0QMy z(kT!wv0)c&Uch^8>_L-OLh@Gzj;Hu6YfJSY^Xn#Qs+j^B^=_Z1xz25N=B2S=bx^`Z zCe38Xj7aZ?)e~`JO7S&1>!UiR#{cTytuJ!Zfbg4TbT;DOX5o-5QGFi2 z`}>UK=%8BsTOHo$KGVis{kvC>y<0p=-cA+M7rtrGq8FvQB~tp+$}_SkRZMW|1UU5v z_>JNFY0^KvqEH_sqdg*iLQz}?PhO@zUNP29R>`$bOzWsP>I;~~s{5y)a2=pW0kX#N z1ZHrJJ@Y#yJ16m6uA7A^2)T2$RRZe7p@GUCwn(;XZO{=Lt+PSj3!{T@&3Wms*9nw? zmumc-qRs@XpTPWsEx}?XeM`PT%#Q=!k`imB>$Io|>tPSt}h5KkG#ANYhz=p>wR~ z+h4)kfXkFL1w<_hcW#m~o{x7kXc!UFdGRG6dJ#n7&7IBQ)Hd)76{*N3ya0G8T$;SW z%mt&@T4v_QHqH-yysyW>Yc>9n(uK_J4iUU}8t69Q6apO?n6>2~4AXq|5J{34O~NdF z!L3tZgfkE^c9tJ{y1wqurEi|B){&O8N-F+vRMN~;!uRD54Q^u(J1fDIKw1L_Edg0{ zdcn~V6CRmMmTJUR3&v<{K8>=!t$PAo4tB>7`u*A2X^?t^W=&*9@7Ro~#dm60Usr0x z|3)iXxGhnlep~F*%jC&-x*o$HqFw0f-a}q9;b)v?<7*)CT^df;mEH&ZE`y&>bIWK5 z1V2)H7#i}eYEwL7nd#=-08SNyQ>^#kx0UD%Q0yxo5DlrIf!fPCKqzPdjKYIhH!|3j zwILKG`w{3>Q5^DCFM*uipAi_da&c3)_;?r>4_*J0Pp%aV8KMxiqlTxnrU3An8Av3B zz*}2m+G6glH0=zonyQa5AE|pQNZCM|ZU<2qdipqgDuD4B^b{vlVmzue9sXE;BLFla z9qulzv&WV@{J zBc$I4)ql}0%cvtcA5?p_*US!kw8;9#FN{JnZhs5x{&rte;C+bJfGC`J=O!oSyv#m6 zuzcBWVEMMjfZxG8X>N^o;hVtU|CxvM&k3Y13t)@fN|bE>&KG+ztY>gy$-ay?(=P!z zsV^hqXL~|S8f>z+9RIUJ0vjsEw*?EWl?BYH@9M02v$c218LJq5!ls(JNB12WM?I zq(iac`>TLA6rb<>fOO@#RF?Y9daHcZFj~$u=QCL|B8DIORO5?v>PK}Js3RMN$g~Hi zN)7^>LZ8ALR_;&!7QYpQrR#*5RAj#NWODij4j!^%gfueh|lx0TtCy;RU3 zu5{|cJWE17ZhDO|%o_Lg{{|8h?f`jvzY54%kfuPlwf7)kB?JKr>KS+k1gtm+Sdn=M zxV+z>k2^ps;1Ae|2eQwE(+&m1o0youQg-%pGq~Qo=Q?8y5Ud&Y#7l;aC>gM4ZBN57 zAEvKl5Fl*VYB2VQSFj5}8SMmO{s=JHlVP_SNlmmFBwOmFTw94!N>e1;Fv5B4(+0(! zJp+2A!8{FtlCy6zq*XXVDlGzQ@l;6M2Q{P+k_Ruhk(jEDT;1vt zW>>j9R|Wh!V`5n41Hnrs5XlMkxMi{k`EgdJ)f?KV@a}&?_h(e=3-`-{8s}lZt_8w7 z4aE8CF=*8{$UB_E&&>Xtpk*on0%x$%43;_IdD`KsZPgW_8jf=;iSs5Sc}L5Krg?#! z3Wd$7oRxnl1@>HoNMbS$Cve4ykplSb*!a?&cZ6AVgYlWn+Vel^I47 z_HziN0t=3#H;hz)J;8{Wo_s7dX4Lzm?i%u|FPx`ciY}-6P3`y!R;>Tn!twY>m(JX| zkdoDN2$M8;ZJzomUl{yr3b2w2bxxA@THDa>Tf9zY&&LNBKyn`XL+0gT^TS2R%SOCg zGD!N;QQ}Ld-vvb3c+6bLe7gaPL|KsgiabBRzP@LDX6F)k*u5fiB*RcmqE2ngFBK0HbMdSvJG+LVt8dS&){tA)WVa zkiUM`2dGKlX?dV1Ea;>`6&WIq7&FBvmZ9a<7XuVQe+C5PAGMfC_k=`T|mar-=W%7!tCNR-o;_kwXoUn*R2Ozzp`l z3_`#R2LWDeKNT?gWB1UHqYEmTpQ!T4l6|y7ZB&Xh?Y7j+w*!KO_XG$;zk9-7?k$b1D#QQ(UZTuIGee- z_`<46WeqY(P)Du8if5>@$ddgu9~t~GG@gpl>c$ESN9=an|3$lD8e|smllS_f$)Z@u zd=+>j;CDzgNG##ECBS#s3&JA#HDKmUu#G^c|2f;==;JLHPhxHcpt3j~-gse8ppza9 z)>AHLRuMM#4lem~`6>E%aa0udsaEvjMZmSb3i+w;%zvT?KQRZ=c$-bk;*>TGa;VYov}(3FNQ! ze#SDf)VV?CqC<%Og|XWf#`3mN@J%n86X9pIJEJb_=-dQJ0@ts(a$N4m_t4$6ahJ9` z_>^EFSa?gE_9#>{en1CRrz%TZ8A{L(+|XFPzV(OMT`vHbZACLTLyv)yCI!60D@pJM zuo4DXsR*FHsC`zMTdNPAo`-zOk_ARNGqC^U2{sWfDVhfDh>_{-N3Ej4n-|kJp*8+Hpo;5}VAe;_~30Y(vHL!(bM=e#O8X&Ql+XdW^=}Ga3NG ziBlbgYR9v59@ODvX<;(_lcHxpsI)4Ox62vo3u8;>Wp4ifaOQ>0y4G#a|2eZ?Uj&>H zl;*pZ>S6f^`VQXvTK|IP*qF8*r>(c}--9MX4;bN#x*wSEy)Gryx$-xT$SZYwO9t~- z0f{7*g-CGkF9=Q;t)~pR)l3iTTP=+qAwmF}?Hk`1t%+KVDZntG=tP8uOkCOc+DSMH|3-`dYx8@{sHGUJgJbv}q?c3jE-l??) ze)8fK=^9=|0p5b6;;CkXqy z!ERdk0Q$50 zV5S>#-|Sy8LRC$M{uuvvD4c8t(onzwftImzqG27nrKUiZGhqws5%AR@MI4d%Gz1u~ zVK<8|Lslp(FgQI=cQ_9TE4lYTRy6>~FA-wwyU-)~=$kiO@apcLpU*U3g=BV2bw?Sf2$OnAt&oIhBrY-YPRDfCNMmK4xVq6+>x{psuhJIz0>uck4gSx_!vMkP31IQG3EeIk&R#2tcky0G&?5t4Xovn&+{csSnC62%z;qWFql+W4DRedl;M^}9T;`r~+C4V{n2>=MFru^4{=+IRJfU6}>fP=V! z(vke-->xxrLztQ8T(tpI53g(00C^qiT?{)e21NGVF!5#qVSmYD5htb5JmlH07Oi+@&daT+AWRsYA znQ9|YBWxBoZD{ErhkUK>SnmL(V(++i!v-kD#V3B^@e_MNd-7GEvEfVSrkG&aF>Phw znGUV6cEh3F9Go})AwZiRvXf2O*ImJ&Dnx07A;wEU!LqvN)q7xA34x}P5XC(pZdqbM zH>h=5x&>4}m~<7eIXnPtdlAh-I9ekm%AkHJ+8nQ#t{}=pNKEt7(aNh4Cb&@dp9*-} zZ1`xkv~9%peo!r!^7gx9o{WW!gmsb5O(jM8!s{Xerymc9-Tt# zrvUwak0c8Ow9im5?Qx8=>xZa{Z%sQ|4w+!#t#1xfyfLiT9xTUR&MFE!SzzX-^bxR; zG5q2MH_T`RZWKIt0HNf1u(NAr`&t8!ybKg;T+7p`a0ZVHHiwMqQ^(X!M|wW#=ZyU< zs6Yj9ckfGW+cMSqcd&rTU;$5n1#SUt{C$x2ra{^}0YY8@^!Gm(7V__f z6 z17L;-!%ix8z}f50N@=$)W^H5$>m7~&G{{7z7t`FL#R#3s~toMe?do=A6Wm*M9f?abw}*GLJ%cEga6fJOv`T2@hz)L!I{? zt)SU0s$bLYJeOAaP%%gk?_IlfKq~NX-leI}+M5oMx8D{!Ffa4y!-$b0?ZCSE{OEnC zZ5>&3!a^qU%H^bPknI1A*!|~ii39&$*JppkQU}3U(-ZZ!_P98zZuQOTyLX@q09j1% z1ej8PbQu16xo;ohynYARf%Em{hR;^?h=55GJSap#QnU1uF4*M&bdr0)gF93Wy>k!T z46>~_#|28(`n+oLNeRn5X9Y845m*0emH0_5aw}kIJ#mGchsg9LA*R7k;PY6(o#N+X z%fG%`f|FEQ{%E}+GwUMvVfEr0%R4_W72tcoC(LXg8y;OGuVr}q*Z$Eu{Vt^7|0C`@ zqnb?Dc4x*>bPy2{P^#isX(Cdj1Qn4{R8U$dp$bSBC831Ch!h0@8z5apia;PC5K16O zSE|$mLJ<%mkbr;$A&{Ks#o6b~{`Tze+v}`z#y`WFC5z;}-{-mObzL{tK=K^jK%6`I zR6a9&9hcoeV5BUbnmn5>2G=O!TrMByKAPIoqLl5 zCaNtGF{O$9c<4@phu+28U{QktJ(?s|Fu;5*1Ekj{Ki{dZMj#AYj~LU&D{Paq5w^g%mP#tTl!V6hqgp3h&dAxQ?3%OH69Pjzh@3i**>yv6_LaAPr2Xyav6Ffn#0+^L z#lETDe08IZ{tzTLYhg0gUtHS*OWBXN<69fOt8dBw?`Mv&m~j5Fm{&eWmDdiP7LTjun;r3mJRL}FIM6RLpYz}3xJy` z3$Vj6rRP|aNH$7?$_7~VuaH-Exeeb8Ub-Z#g?RNrQ&X*C!ztn40K_#MR5>q^>27_D z&G@oHrKr85@q*58jUH4?3>DdoFDwICYk^-C>ru>YaJh@R6K{TYT09hEQ@NmD$ntka zZ3&*2tG~FnX}!68=i-6M3l|f+fYY6?HeG_E7x=lNdS|uYs(h}&H><+?tSqUaG<#S< zhT>%UT@R!dU1=#>c@x$Ds2V_YiQ8Uv2W%HqzY3 z?|mznkDOHqZ}Ss*#GVGbt1`yh!mADeFmhI#q(GZ3&nGn-`P$!J7`%g6P z(!pX6nt>NDGFiL4^V?rs{r1QYe`#))KOttU0Wx~zFwXrs4Q!t)znH8g4VX|wB@O+i zk;=kA~|VOG`Jh;=cIGdM)wT%mrbV=cHf_0O}pvh{%D= zg?i~qfMh-f>oBb(Cyw=IOxzY~DFxtl?MCH|fdB|uUkeM?ie#>IYk!UMk~Zy9p4lF7{x>{&4s_Nn8@`?O<62w ze9dd3r6j{xV(u3Sw6z9-R2=v^Ui-vNC~@7w{{&Jf0kUU+sxl~=nkd_TUS%Jx!b|m7 z--w~R&CvBMEZ1)K0`W3|#_ChkzUaM*A1gW6epORjaVJBUx5A?XfE5!rVD_D$uU{Ew zgPIu0@!g`ZddU7@^my<@%02(b*RqQJIzq;h`Bcx}Gr>i6fMB(6cq9_W<7rvm(Ll*6 zgI>5Ik7;|%djV~MWkRr6jZqz0nh+L zG-WhGhTsx8>g7`iXa)FZjKavJrtwRb5Sa)NQospUymn87k;tr&$Yw2zK9Oi(wpV(V zK$yw)XTG3aDGR-GL;uONo1b~PbY053h;oI7oq5+=@5bFeFsXkrwF~rPI!iBh{FJuo z9@!*d$Cosp_xoj*OX&4?H8yE$P@?(lO&*-udE0Jct<3Dyldf!IkgaqkRP#aMt@OuW zG4S6zA60*}&-@n6R!y1TeA^2__ZHB)6G_CV3O}8wnWQ5%%$PKu3dY{mpqb*S121m_ zlBOynQ{D7P$ht*SEWt&$8YE~>i}z}7hl;$&aEak|_pCAB=+=pcV}2R4{pLVxBvs7a z%&C9jY9Om?PvFE`cq`+V207uULpc+IkeCe+aVhX7Z-6SFFMPQqzT6&Xb89Je-gG%^ zkA2(&f7~t4`oIgXjnqLfTQ!}6{2BD{cl<#PON?U=s4}BHt#;-)wT8t3&v)r!`ZbW3 zo&$#*ie@8QKu)@6!fO*a9HvvMXF%`yrB`nnB-U#9(}q6M7~`@s+C_j7qDjR7=4$a; zBEV7|tzcXr90rTu7xm7i^ydq?SH{c_1389X#oH3W@OIZN=%6*SJHZ7R;xVjQNAyYu zfO0_fRm-EkH>Nq9lj8@uJS6!yd#V>DImOj!jTD+QfKWF=3JnVx{!U;cq)Q9j5EW;? z!ne`~zZ8%9kte!p{iC9{l}tF-6rEmarHCmV-2X>}zdsHmF6b9bW{=E~ck1LnR^&Hr zk1jv_K;vNMl_}tqKUro%_y63;F)g0SRJGZ9V3xHWZn^iF{3JyY)ArbBN=GplH-j61 zS)qYc$iA1W@`>)Xsa^p5FNpjms~e8)lV!=b!~Z)JsoFegKo^V8AD0eTEJs6zj%JsN zu|i(h#*$Y20Aqr4@19cn#1Jdg31>!6OqHAK_10LREIcF2qg`dm7n2i0!2M4`pq`Jd z)N*VKjnPm0s?Oj8s{#=qCwd3Q#GG)60zYj<5HDcxs|Zq(@T>0Oa z2mZcS70|ynM6(JPZ9xv>*)k&x^yYwK87i7^J+n8aFn z#e@d1eJ7W=n<7`-^z6Gm(8~a88wNyGbZ8TT085NEc8iSeo)57{nF!>|@stEcX=a-c zM4H%UMDyXkE&g>ot`52_9A?_89t-kyleNy7Uw^zot;l}^W(ZE$wZ}5rzD4Y_+wm70 zHQ}dJ?a8ltj!J0q?!@GMfkG)bUi8yo9yWp~FI+J*@P#bnAfKNirRaQZ^(m zi;iF;wt30iy|5co*FNlj?}si}du6}O(;zRbVtu&7;~#Vn_#OWzjq5Jpa53ku_}bjt zZ~E>Tj|}~MYr<3Y%)5|7iA&q8$pcvnBpfGayTIwlLQM2kFEUGZbI^)_kP49%U(yTUDgj-d8bGJbTSQ238GWnk<$tVmOsBn*tS2 zjaNv@jdDwTU=VOwQ^Ovq@;s714P_*JlEfesD=qMqXE+yTU{`w><{PXJ_#IG!cv`L; zG-LBPlm45_2SQbAvf6AGWS^Ssq5tOg0Z@-3h|hm%9pM1JON6E`pg-Q$dgU#I{&P

t_1*<)WK+ep*Y^I|@#Bo^+K;o9fxhX+yaHJ1 z{HWUXE0x6jl>K@ajeh3VFsJM(f9q>dslhEWU@B=4E@w8VlNSv!+0 z!e5^|J@Fez1xA;;qj>CvmKF|xNcnkny5Z+<9|JKsZQWnk92J|WVRaoa+5yT0vvpXz&v`I%eeP#kF-ZOP zMm+>l306yXFRWe$)jz|=M_vYOI6|FV+;YQ&pF+=Y#+IoqslDu64|-FaI#)RR!&9(@%F47Fu8W!;ruz zKQs!j#uate0?zadD;_oE)EJU{r=nqRbAcf{Moq(fz^4SF;eSzOn6%lN3ov4D-D$sV znP!d!yaNavb1J~Sy)skHT?@cl5sQkTks(@$&QVmaqhegn7kxb?G_GYo9;7(STX8UL+qy2R#jq17Ju5qMdgicMjC*{cSD0HByaVboO&r#B}HrQbQza8&sB29<%BFh7~y z_qDDih{QE~Z5o$PRJVX<-4^o46_&W7&OtzqJY8RhM{3EIDeLmW;`_p<(E~MUfGL8Gsz?QJUoBykiNy8toPL=uC%4f4C+>OV*F-&QLc;V{8%p^A z{+UJDI}Ce41ZD2*xB9jU zek3^SHpfQxtxOiJeV&}Df0t)}OgY5Z&Q{FEv&{Q5sEvv9dS0x8)JY8^1Z6Z^r5xUFwb6_U&UJDBm0E+3IU5qCmzh%2w*# z42vmEpnZH;sU{qh+19rQ3^Q1l@!8YV^!KM$C+t&~`PGB9guk`|v6%s%DnaL;AOW{h z65}=gnLcGTS(DlY@s5D{?ImIb;jz{;@2nzrR?P0Fa-IH#M(oQC(>g%t<@s2dcf}>Y zouJ#2E5vy2G&KDBSz7j<9dF@=tT@246@>w)RI7cXD)@T*-|ZH4f(GdUf|oMmizbyH zx4V3D&O})$xdoEWOotbbhAY%$fQ$hZVlm_J@tN!X_^~p$ zqI=%Dai^of#f_fQ-S2i^?g5x>xK=i#0gtnpK$0#nvHmzOK)kCg?ibAjz#EhkH~aRQ zN&6GknM#qVN+}6_T^LlDRuB_DRkR-&V_Z)h661X*>>DeLc3HP5O&!>ba8o%i4I~&v_Ue~*C*9-x?&h8LmN`idGS0B`^{x%T?;cp ziY6MRiD4PKj-^xT?ee0mBzU0evaHiF^l@PyTAHYH(^L?~V4PvmP+-uQIC-T-&&j}>ZSV9KhqEg@5`xLjp;a%~ zQs{~6i%4yh{~isEEHmxUW~0LJr+Q=K&hOc$pf*SvUUZ*Z-${e2nv3e$b4 zB`-1wqvycxjS0*VZl0dzv+KTjgHUCf`8_Ixp3}`@qeCg{)cx*!rG?l z*}iWd2@c~A59F?m_T(-mbW8?3=BmAn74;F91)GX+ zHFnY~ux(KJ#T3Xfi>{P|YY*uxV#BN8lrOr7ku{*$M7UA>+^wn%=VpvB(5Ebx zf)VSz&Hm8T-KYPL&(=5jIEewZ zzdfS>E!DJsgMkdCu!fgK>F?fw;le9$Ht>eB!ViE{xHQ3G(Ka|fRW0giJK7?+ElrTO zqIf9FB>{QS2B)`#>VE>{uZ`;yX(wdVySYj_)Bx|Wy z*yWI-7vab)nP4P7j73%(rCfiJ8wh=nr9x!99z_tvRRxJ-q zv;uI|e4D>ls=#;zunmv;M;t~06Ippdh6%(+vZznG4!K2xp1C2NYpUrhz_3nb*P7yD zLFi0@vdrRsbzS<%{`U4mv7HMm6$OEi@RLRCX2-irS4+fsGm;`HCc14|(=fm#SNZiU zF4{p6Vnw>R1Dz=A9d7PJ&#ApU70iuU#0KP01l&yM8u%;J{-f&rJ8Lq)IBQeVkmpCG z4(I~0=#sT5_`PoJe%7UAtnz{f(-U3ss{k-fcHEK+L0cc(SUS& zp&~wN_Y>CR*L~*HrRp&2nP&JU67I*nwJ9YJLi3l_GZkC`bPk)&EhBek-{DaDcf1ryheD#9x%7 z|5IK6XA48FaVEW879y#%+>mr1(Ij*Oa%>)ASCp4UO_m5B`Z5sMa!DvTf4hi@UBno3 zEa+sDVWaD;!>Q4YkvTGk)g;aHLO0?3b;<6ykaq8ZbYa`9RVUTv)* zWSRmv5%w?OUO?H42OT6zqtx4L4z;vQaJ#7iNsY7}_|LLT2nP$&hVK(IE6!BIRFaUz z9=9NHyC5HN1)y%*0IuK?5PpUj$1>EfoTRGUt*mn6SH-eYsV^8n+;~{Hv^xm&j0=Py z-8BD%s;V5C2$Cc7(eP0}q9Xh#Se0_hgE&y{S5K~&mM%#~weO97_nTEaQ4Z2Z@7$9P zzEHx}2P{r&iORRcyVlHcWS8uz-kP zWJgoC)&o47(6~jqc8$!{A?bcmB25`ls&Bzdvk$ z4p$I_8?G8^E7k$1n}WPe3$_Yi>j!NhgP?=~hU#|>kR`eR%{v(P!j9=hT;E6{g(>Z^ zyiKm!5~d;i0V2%ps5<`+NNs{y_j9l>B$Z27AF+L8uHTYA3yJ4Lc)Tefr22Nj2r0jm zsX*=aS<~SG*w%IJi*$50Eci4%El0(O-MdGy#J8CkpU_}mpw}!V{S|-GHj1e zC_bT~OVK~mf3aS=2BPXO8wVW_`XFZ_9&FOXFuC1PXK9Z7PPGfp^9E2* z0JqK*fEyNCVJ6p6_<MRH^JP}7BiLyaPk7dNrPHe% zcfc{KuVPm*ag~?pvsX2qqg-V7MY$%&5Rtr9eQ}Vu#rl0pe={S&BKW0!vDfsW zUx7y5dH{LMh*CxRjk&f*b*S6i0w%kW^zz^NUdChH3p0XN69MWe9W0>;!JLy%zmxBE zwG^q|{Y2?J4Jz~DGY+FqS@)HmHoVL{SX7!rD*<-A#5~F53S)wA$l6BI@9Ao+h7kso zG!c+$9N{S-@2aJe^c(F<8$i<7`3uX_{~?$m25FD200AQjXek*GC@k>EzGF+;v6+TB zgR6=>%6e56tV9JWe!c1a-WKh}Yp0f`*weIw)AZ`GQ@uo>hs&m5EF|CKCtX&o6I(K8 zNIC9jC^;2GVgYm76bL>BRVoOtX5Ksvv1BePq8)(@Y&V9jUzreTDf=9t(H*!OphL}{ zzn8llI3OG_kr8$s0VoOy3Db+p0Bu_`ONOg~MNi21OY2_fbStW|g`qHSg6WYYi(iw~ z8zn&3!H})L(%|HD7;N;JP@fe97BtAVZa*M@9Kzg!T?Q4;X{?@=L(>*s@BQuV%w&H~ zdJG^f6h-h?;P+3kkCbG~=22TSQSlB+n)n?z>xAnr7DTn;8^WuJHJa37QY+ zsanRas^|@8Em5Y{xDTtrzA#^Bm!ZQYYZ`L=^?RaQcd|yl<{OPyMNL<*msTjZTvJYY zLh`I(n|9SbahHisR6S}sVw6(Qw@yM>1I4#5K=0(#|32StpR;Xr&P_qB;y?GM{y!J? z${h>lWFXX|#2+t|LGTA{DWaBs@K5e$;K{jmk3Es(Sqs=By8_!;`KM+Rqh7_`yYZmdXC-12mQftM;nX%mut zC!U+T?%C^Y|D&mFjM&knsLi+W1}EBq;OIfp1sdlHJ3Rpi%Uvyff_$Lb_*01*A(6~3 zBa#lxYCQX;xv%MxBmVOodG?W8dWPGoX=f%!h4Bz>m~i%|`)Vpa9WiktGN!6BPWXGD zvm<-&uUto;-RPJ|{`9@F;kI4Q%dVSg**BU{)nO zbVIXUdN1$mF}!1J`!a7TJ7XaXQm=)Upy*pV`n1!9mDw??HzV3RW#)QH)~?QI1I3z4 z%)=iIN;oGx78MY6(XI}6Xn~1t{x|9?mfpH8tiO z=5rE}!y&;lOwH{l>(0!zUJCArOm zJ08>h4>_;jBi^8ECaSjT{4t^V8}neLL9hw56X=%!!McUkAnWUWv_!kq_PuU}&t-+x zoRl1!^I)dQd?QIs5;fEV*h`&#zrdVcj$UX60=_qfmV1M)vIAcN{jiu|dnGC< zGQ8>{-R}&**MXtMW!M$iG2zn8mH@*qyq*~V^C4bnMqkCX3-Z$9&es3b0J?}WS8N{MjhuMi*y!tU z=0)Qmrw(u^TmX#r2fwZ0Xy4V=68YO3zkS=Wwi+K3F@~J0V1ebVB`l-&mj<}oAn@C< zQcLUTMuU!6d{MIL7{QNq3HQ&-4J!K+H796!7q0e)l=(< zUOT|l5xTd0N|=uYvK~>vkuE>c$Ln~iVS239bHzQS7a*O>H#*;|ZI-`x?17~}E*5Bw znfEM>n7`|DQZzTFizjNyv>W%Gy8Vjk@_Nmjo=v9IQPr-Fv-KhGrA6=Nvr7ec4YDx z(lf3l@kbFt72)87-~rO&H4sUuMRZ zaCXVrWHC9|4qw+(!Yf~zCFPpN@Rr3K;18++&;W^zBIqObM-;<^pCAraQ<@pqm)N=% zr8Z#ahLp3G8{kWUuFvkYlJl~rT}Uk>dXTRM;WkpMv>~LBon>)Y_)CMIY&!N5ZQY^)_yq~{vjI`lOpbX z7+M(gB`<{v8U&h`zE{NS%>bx}RaekQ!5J;uR<|ixmcf2$wx92@O%d|&*@9oGh!Gp8 z;pDol>cZ67)R8Yk;^n7DW=D zXulPcO{;X0QgCI7dQ=Xb5WfDQjZhTtWoN zrNaRdoPHGcWEHekM+rco^wfvQE+9x(qx(b6~1#rm-6?B``@da4%qI~ z-@Vux^MnpmT@wZl>StLvpOKxlcyYwVLyR}=nm6_4qyAcylbl>f?}CT_kmuXGL-`%P zQ;P^!FA9C8WJKGdCJ2j)hXrFz2pykpC#-sh^hqea#~tt)%q|gQX-bUCV!5TPYDn9TgmuEH0`64#e`I6ZeB_XCNnnP( zcFrD-k)%c-(c=-M`?+CXw-1Nu(l5XKS2)vOc31xb?DKyKVwk@AGa4-6+q=-EN!9Xq z(s%Ig`tZx8{Xv*w>FW=+@%ST>N-pSzK)@OH|AJNi=bL<{xBAFkE%33DD= zMhXOwQ_=Ar0NDV^^SiQRb7HHF(AhCU&gyJE4eIMfgVLdENBVd{vJ9vZ`an1;ynvo# z>coc_G3dz_>y>AMQpY(Usp>MmQ-zz@=Hahso^po)+speS*cDsklF{OAxwLp?We8d* z_zO!~V)=eujb!GcXOjGnlD~sqn%E6^!WhRgI+eD&iG(x=dQnEU1DnCVfrup!eXkb9 z^do?(|Auv)l!j|`xSOY}xHjy?_m$ucUC$&S7i^<^v;- zRxB9GN~fDQod}HgcmkGKF~gxbYRP~$MbTk?uz#A}-Q_eB2Qc=-+ldK*B?WoHQExtK zgF#U98LZ_lhm2?2-TV{d=Zr|lgR+^w^gmu-B^%7po8fB93Yg$KYi|v~cBVPh#q7Hg zcg9f99yCOm$q;>CS6xwDCysqx*gz3${`CF)WlnVK=cR7c(CT7q{JT*=Eb8#YQwCJp zbA_BK7K2&~D{0cgrkj%?k$oNjqOvC3br5(~;BQkx@^RAoa_MTSt>q3MntsS= zm-{_?8Ea6R8_;NDmBOY(`p&x~EY)tP6EM4N2=q?_)`hU3eZrkbw^TpH{|11kpMH{C zeqD8Wz5F6=-F$r|E?xnIGmnyV3SAz{#o2!e2QoyeyJVaWqR-uE)iO}zq*YqZKiGUP z*o3`zC_^OcGE~+Ie77`75^N;FZ)K6B2?xDRkITjGe$r4|P*zz0rH7uW09+JEwg|dD zx}8=Eq`@Fk5B#*swY8E;%pP4^NB~x|{k=&k5WF5~cJMaEWva8hk`pCppeSvLYVMMs z38N{?b%3*op9(r~+7yd_Mg& z9uo8I9k9|dR%6O98Ks`lgi*tXa9@6{I>$MS_neUgY)jo2Wjxp;;TH-G5`q>? z{{BIN)V;++G)dXbx752zag-GT?NRgU;BFyG>M-!g`q3Kl%;vm=rzXBltw}EhL zWCc=!#mD8n5=Nd%MJo0cdT;RBCHH?nHwH$v zvV#^XL+TRcWFT_xDSO0H#lt831cC;(4O|i1jxwZ!EVsQlPM$5GctXPr*Tx38E?uX& zPX7WcHFYr>*5649n$};!A8s)3=#qB`hXlSUgAc*h#;^s0irFb$0BB?3`oNA5*>Ewy zot+wH?iz=-8ad$AJaufZtEfibT~SO+tg;y&cP+n!C!cqiZ&fn|Q_{2FQs*DPn|Tz0 zGP+PALLjQF@!}ctl%pQr%Wt>atL{d+6^pS_;K!ArSTB!`d#{iay$}xe&8%Q!wDN;S zQYj2eAa5XR0WC~&>0Y`byEPdI`-=i$uTQJ6OgE3I3Z6wKw^R z9u{zq0OQ;TrcT*!lZ6)ThVE9o$fmu*e+^L+?uJzMK5zYA{WQ~9O=|LDD@VNt*Z%nR z>m0XC=|ZH7>=CgvSb$swaRw=u&5!C=lbh=~)uf`IP}RH}tjB%JK@4g%Z>*QiW>yW*dM-EKX>T{j|M2D^5D3WhEj6Vk<8U&$D%endS?}@ z>6=S3M6v9*pxO>hNtJCMhyzE%1#JB(q-FaI${-7p$6AdRWPlumfOGc64P2Z?y_?4t zbV9-0;R@|2$JhdN07#)$LHvqP-pOC44Lr6?gP(Q-Y_fsAn_V8s-c_RQwjuc00coIZ zpohO>G(P|oWTdGp{!T!FP0pC)j#IRtW-iSP!ruxzAGUn%_P%7z1LxUP@NVx5pP2Ol zKK8TKWg|J8jN5~<3Zh$CEYK(EcA7~AVsZF8F`Z(^$>?{#fZ3jRKX zL!KUx2R|~+>X)WY_&Zhk<1+Ej+*aHj0bb)bV&y%L8^zDu&u9>$wurHEeO|i(@?BNW zh;MVpObYkNiJ1h-Tfr~>3B~;ZdfrrnYf0w0@k`A$Bq6Cq&1iscV$2x9>95-{VBU>O6DnYuAZz`gCV`mO_#z8wu9bK+Fvf z^6(90zZ#-@?=;t@cR0_D(OY8uvcroh-nQM`EedX&Gw)s8*eIfWh@19H^W;Zm4F^z% zeIx5^{(dR#R-%m(RS_gai0iBlAgrdJZSH&~!RBiT5x_cScH{YGEG!lhKos1|>n|T0 zW({~e50ACant)Zd1AOSzm~}*|+oN^@PJ$s8S<@o?GtCXU7n3u52#Vc&QcwF8e3?lL zea5;<-Qy>>JD}}uSII=^xs1A zwLI4G$BFfFp66KVj8s{s+nm_R_}vtTvEQFEAOYe2fi@keY*1sePW*%q`9PhvkbLEF z>6Ey>&zg)wO?>QsZj~dnE2EU8 z6O!nadBDf*l?PMNu9D)Oc8|>@lbM2^TPcdJrZR-vUxpG_1Cv5hi?mV{W-|mA(sduy6aOZib>i zj%(J#EOniWi-LBz-LOc8*4oa0^>>O|Du2xy2BZ|Q>+;nE>7(r$8OtoLG#=r5TW&!n zr}lgKRM~X;J^>m4&j5`2`Am>~3Q`U^^W%L&VuSR4CKuz6T`P@YlkT*Nt{XA-x)6{T zfof_6Cp92M;uR-dPmd|7Fi zT`Gz?2ZSLJ-!z)O8@sD*XA};s#WU|=wn|eAC$#V%+&~=KBKG>Tek0PY)-PX;ZLS%l ziF#yikk$|Sulx>YfYMD1HYpWoIV=QV)3HlhuE_wVPW*@f>)n~Q)$e`j#1P~X5BoI1 zsjw3j(NKFFv-}j4FKcVv9;Iq^#)qkmuUoR_(()XF&tk~um~HBfeBN75%1I@um}f&0 zacQ;&?px3$1*6>+X{T-_uHA=KTTw{*vY|jYPxMfucFeZuJgFnY-j`T;>+y|EXbXhO z?C9=e0HXM2)A)CSEra9Majqp?II3ph_zAZ@E+D&E z=|#Tt$q!R^a;`uM-!P|>&Nr=xlC=+cQS~9wARv>{w_18*Jz`uLG+FZ|%(1PUmrT!I z!2a;uWFh&W#UeOwEt|Nat3mK;PGs6oFf4yt zf__+iPQc{}@}dFGBPv`=rjQo#k>5wl17YxB$rn&*T$JE-pAweNLlFf3$?()?i|7|o zU3=28^NZVto|t`FBzq@s0ATYWcIF)^%;xVfID~KrDOGu3;jFwXD;KP!h-{sGsAEAc zyccFGBNUF!ZHavnnQ4E=B7LJ36%X%00e{);7~PrztvJ9x&()^NuxD5`PD8NF4sA*h zeyiLpNme+gVcgaKenGh*x-e!X$9kz1|t}g z_86rsm+I;p>8h=1*DW*k;m)5dQ1>vhqNs5Dc^*W!G@|3DCHi99JEAX;tvl|`pl1)n zM6PoVG(;pKa?AML#0QTUNTj%q_s;H$lK&>#_{;Z#6}!d5jr%~)3LLJ&t*kIvpx&4UiamP$I5F^pe1e_*jbX;c1eKYloSJ^T&!1gq*e&$ng4>9vI_e)t6}# zr9K#L7qVe*w^LqvHxKBWgP~FQ6o96YyFK`$^A7B08!4YJxHoP(UwwZ3YuRRJkW0wc z5B^31b^mwF^8KN`T2L=W26V#!f8UKISHgXY1^kq5fIWc2iNGMbjxxeeu{YM_Jq3B+ z#a|X1@_Ge7OgPH{np^n*YXFwnaMkpUD?-@3ove%8A5`Gmp@DSd`&*0OS<{YwdahNrht8Y}4^K`hTll%|UyNxv3EQSy+g zu0!U!2~`0A7uj_%UvNwozQ8>yU;U_bI%(F`k!6Nh*5kRWr2+}psgZY&ua9l_UoeVX zm-BATUC*Aap@l>14YGiJaiZVtC`YJ?8OaU;ff`SkdNL(p!8X`f!*a3+yQJglfbZLd zDK%!EIF&zfRZm;t=n2asJlFNxJAEl`)g$jZrq~&lH7tlbG}ROP%nu8xy4`U>y|FE= zvT90yzUVVU9qBs-G>uf>owA@v75e%Mq`;P^7kdx%VU;0K@T%3d@k+bI0g5otal^lt z6E4z?0MK#DW?G;}t7$KwEA*_mRTKojuF|7dU&Eg=NRO>r1|lhthnTnCf$A@NX(!dG z_O)OPW;q`_zS0G@*xEat00p@+plG^%(6c%Jw<8%#nkK-m9RKWohb=(KK^JG%5?ie2 zd~@|oT7UXSJ_`}v^dt)i&`CDFCm8kaWt5`~KPtOYvyj$N+;+;stQ$55W(o}eI9NLf z%}<+6vH*N2>o%hpphqRQqD}+oZMU*i#$rQA+3M*bSmA_}%gU`1Fk#Wdlv}n`xh6E2 z_Y_y2o0*fHW<`KSqs`r4RuEtV$NqF>ZEBD@ry#ZW$nWTV5N?LP`k=rrZNPuJL!nr_ zN3tIX6B#7eh&sL6hqvnhDum%<9JjquhsIk`BE=z9$2TCNdEaejYubOLqz|miExlQx zcih7mjd{_lbqZs3(PnsJb;|UP8Zzs`+KJ#7)GuY)ihwaK5?XCbb(T0%`6gT0RqcXUzgv zFE^UOEQbWsoH~}}(4JEDmUdvrOyAn)3W1q~Uc64F%%$VKHFh#md&FmXZHg!>XM=~Z zT$t=s-h$TAVLsq>R0NPA`l(xWTCc{%?n(V^UvU0EU!Y?M-(>(Suxq@bTH+ z*QmW;Q%Jck*_$Rcta&YzP90%PY_gu;pS@j5RwkC0giVy1q@u`!#JM3N zW}T>&?pkQ+jde)~R@L)Nk;=6c4sR|Paj&i(kXP_r4kIvD_o7zJ+5UI@%X(KUObz_b zkhC=$B$ww+#}DMI^5dH3$YVI1N8mJiV%p!1ZNghOdv|?wA1zBMg3TzWVAoKUg~gNA z1ZK5CrN*V>CPG9Ug%QUJ;7|iNzTG_glTj<9H4b&-<3qrvwt`m=m-XB~{)drvnB+}w z`zIsmnZWFQ{xDMBu>*>*3ooXP*H3?Y2d@l~xk3N@4on<>ZT#aMoXTAg$ao^2YuYRL z?axExlxOOJM`os<)UmMAjE{F6nGSUaOr^vm$8}V7bg6~Cf~?XJts43CJwqeGrLIAK zJ?|Aq_fv#M$*etuz|Hut4~-)E>sN4;+>E+eVk*Cvh+W|$X-(8+fgQ!^+C9^J&Qypsq#u%KkvGI%0nzXk$+Lku=3p(vUpcv){uZnGmq%A#B zk56Q@bACgT`+km~)%3YNf0!;J*JSBu#Xp;F<9z*LI+M6Zd(!{HbpP$~{nMKPR{Y^j z)|gBvpU~*3)ZV^;x{d)|_^b|sIegr{l#9$!UF-F%9`>w$VG>Gf?NZRtX8&&PRTfyL zur|n6O)5gmva|PSErIHzUk&>A8TIc@0#GYfV4lxQ}Ah@4z2K)6+JJfFaKlkc<&}aX1CpRvTPH%2N z$Tu(Z7Z)hE`L_oK|LVlPU7&AH%*tf%{rP`#Vt+bh(21}$dL4N)y+&H|Mvk^-sY$3{ zTkv*~7p}`;Q5f7?`H_#+gc$>Clh92O0d+J6F=)|J2V-|1VeME=vb!9dGid5n)*Ixx zayrQ$zoL(>EF4&maa#)W&7a+-P}JoAB~F*x&eJ%sKKcWR8^~VoU@l0pmSq@TWbH1& zGznFKLw9s_sR0vgorEeD8(wXIBilHySHD|}s}hdjUyg9zAti!}_JtOrT|Srk;o9Na zKfLuRV{Go>dw+h=KYQzf+z0hv|M1rT@ol=5V_{2jP4|nDayK3a8XiCciWBBNLRY!6vT(8`Ln5W)g#k<`V23QeF9|w zQ&>8a2S!(YfkF|UU6`2i}}Z{F6+50j@_N$M2| zU@RE3*e-1MA}=hurE*KMhJvE}hj4``(@=x$tk8r4b#;cd zkR56I1I{E+9q1th{py1~R+|wK`PmiiL7pg#M(jvTC(amfC=s>ooQg&ggfBky^2*KE zf_8sgkUxdefAR$XCY||D9L=As3O?_L9|SMt9oEE~gc<=GoLx*Bm0rt0gi1YAxbIu8 zR{%TZUY)xYoqx2J%u-V*u1070hnBxzOVOHKW-bi&RhYCCC{b1o`=b|iO0pD;StKzA zF^{=S-+?7@**xgFYG%sQ>ny?2|!n?~TF5i#N-C<6VpNp_gWm{Hx9E{-A99~tbI zRCQ&}btxWwlcWx7=6GE@ z`K(1?0QHAH+3m|x41ONqb6!+QMq!?}uCA$)0@H{3B<3KA7r{{Vn#?Pq`@iMGe{zBU zl#BhlQ1a~n{n^9*ORD#$bQ{ur`r*>Au&=#IQr47dUMa6-r0G#h+B9E6QUvyB?fTe| zN>c7tT8VII1lD|~wVG9E25o(m#?q`N(D$0tq>iD}EXVit-}7kkK&_jStQ-@9Z-|V# z(+a$(BZ1U{j*^)|iD~|hr86ZZy41Re7K_87{jlvOX}eLfz?GUh@WeirFm>$ zlS04SO>YMK;G~AUOf6QxNZhBCUghk{5x#VDgTvZWU@aFu{xY&^l{W%&rw_wdr4K z0n3F66)&`nQQ9`0j6>aK$Js46z2NrM?3$Z4YQ6F!GqLz->gSv|L0r|*qE+f=wVBcT zeHu3RXQld@TH2Jt(Y1Lh$QwJsnVpUN+b@ZU08#STSzppt0~Kw`ET0Fd^2Lxy(5XQY zm>S5xX`tK?pCfl#1^bk7ZEgIF@)QC=8XM#s^DMZOJ}n09RpMk<#!XtWi&-v?g`?6|cFWow#AP;$rpKI{9GK(ZJH4%=Bt88k zJQOpq2UGlZg zjD-@nm>tWpzO)_0CtFLqGGED-xR2ovN}^Bh#!kiP+8_He<}AGeg(}?g%|;d}_3Udxb}fF8y4ia+E_$T03L(Q4et;*c@(|m4iG9Z~Z01s;=wIvEz z=zQd7s;L01Z|T)Q)Ms3|86C*IeecFbgXT}Q&7T`I|1S&Z>xOBf zM31F#3ctu`HdA`7x4_21dT^HSa6U{$^U$zIhB+cMd&hFP7uBHv*R+dcs^}t;2mDtG7M@fNq`PnNoQc_T4KC~4X z;`WuD{x_!0ccEY(dGq;7C(1^xg{R1(YP7=6W?TMY@tt$Tt5&|(soqEfKW|#Z+sK%>P;Xol*J z|omt4O>p9YejpIwF5<;^u?qi@C_rqu4?orcJkZ~k^)e*>D-FB+x5FS{*AWK&hRrTxydwUtpa z$XI4|ZqY40$k%AdNmq8c<8aSZb+6ZQRjqsnZba@lCWqml7t0rGiS(fkMNAzUQwF`k zw-SGEfA@!H+XFLYY?y91?aE@4Q2x?@y7V?0O;?T7E49P=YA73F8p{59+Gy8L(OsmV%<|fT!Xg@{_jUPJk$w;Qc9GGhNwWxa-38X- zBP{vm+s`4NeWWZ%L3U-;g%!f-0bMxXH@;IP{7?8K1zBJAYtzX5R8~;GX1BYAL9iEX zTIaf1e+R|#a^2)*QwkR4pKOjbSA9>HdCG^JUa})euiea1O>v^cih|4xgS+_^2;~Jf zYZ%g?vd=g6?@o-iz_IN}U#+<`V(_v7mRcZah#3$G4ZW@ME!% zuHRQ%0c)q&(Y)aGUMWJaSUbLjD|xK;F6MaNk=poq0~JwpX_;ptZ6^7@hU18bN)Ly{GnEp?zPvx z*F9XCZM-!(l>{Q-G2_KzsnVZYE{hx-Cz3QsS;B(zSso(N(Fiv&RCL+PcmgGuJN z*MZIExZ)B9cPC~u4~7<_;;gJsGMn*gv*eSw&yeJlEMtiM#)nr5CGlDKM+ZlV11gH3 zShWZ0x_Z_C*<{I+mMyy(}QN5u;5yH;_%EyEhx=M zG>a?KKWy&OVKMv`v8!%RkuQCI4=${*=8PYmuh*AC={=b+i6_(0o`FM%VSeCj8)BFV zuS$o}_*!(18jYFoF(8&6YGZJm3H9e8=22zJ~MW%e0X$eh~*JBeWsI&~1kK8YVvvi@e2ZiaQc3M3yV%`XL} z)^?_CYVSCgzU2X^l?%U_d$EouJnRZU4$o{tz6_Eyc_96W;63j#&plvpx0mVU;ULQF zVh)R#ENbse5p^h>6Sf~fOVz40+U0wPZw!pi-DN*eT;^`B+A-cLabH@=q49r>lei^Z zw%qa{P4QvWeH@rRI9kj!U6L$0w>{3Qaf1Kdu!Wz@1<7w*O8Q3X9z*rno%+x_W#DqM zbPcGXXdzl{mNs zQ-#`5Oq)|G0?w7!br7Y@9dsF}RIh|{FMwK1Phzz4)0IOOD_SB_UrK7|-P64doU!>% zCmVy838?F5Tj=+_9Jd8<1moKX1yta)L+bLU0P-``T?ddl1pDos7?OBQp*FbAHMo}I z=)^>FBo}8F6`ZEda|qEZKcU9XBVD>U$8J(uKrYB&Bx1y>LJ@&N9C&^TA?6w`a3?>@h_kx>fhXUlc% zdTu<>6)~f9Or;LycA|=l)SNbY>OR8%+WXx2BR{j3nkG~wN#5g|`sN1y^6-Ca#?-ia zVuMnkvR5KW%N+cixu zv&dSY_8FOI8CVS>>96)_COogPKo==(1~GKj?Lh=RJK2|I~u;k3dQ_Sd-Y5JY7wO~hi<63duR-a)Vh-17?5~ndL zXURRV$4tmmTmqGdt4($7P+!W1iPCj)uNb^tyo;y4bF9d^BEbpXIr|_S z>GSHHrVK;I%18Qg58bk%YOcYL+*D#1^v|=`kj=B-B88=4=q(TRj&{So}P`xT?cc2^2m|4R928l*)d>3wcxGi9|C7Sj`+#dc^M5g$0FkOc| zQgV8#V+try`|LD1`EHso$gQg7o+P2RK;2cg%Nov(I8x>XpTyg21%ahkd4C|$6F0h88*^{nI&SYW{ z*>F^4i~w-#$bDDVdvP@g^9Kx(X_T4Kx%`8(Uyi2=9*#n!Gbm3&K-KaZkRBQQsiPu+ z?qErgvlXRs5aW_&niPc-$ESLXT_yBj1l%6o*)Y?QxlQK@Qz_DYE?WzIG14zH^}Wxe z9`l?P2+*&R%%6?B-`+Bs(iXqtB+{yKqVaXD)c`Eydzf)Y0Gcr)yd?mkao!!KPKq6C zReHTn?nPS@>6E2zw`%Z+E>p@De0X^6SJ{;){zqPZ z{LXPjBCyhH*2*yb0W_lkfu)`*OQm81NOQfEo&-yX;VX09NtmuY3$GQ+ck!B+D%Q`4 zveT-|vM)AE>H_+ePN?t`c^Q*uI25! z^KVR`@F%|q8dTb(q%9y%ya|2?5!YP3VR`SYl2*`i%I&IytAqiSbKPu^KxW*sSb6YB zFn-e9!hiqF01L}8<>Dz#tZkfwEklWk43La3g<=feOXiL*@hS{5(Q~2OtX1bWD$pi1 z7pCDVUG^P>z!&0RlkDweJCp?WYXYNPgn38od zL5ZEKmE!7lEd+@Ie8*^(P5y*gWp&N@ zeES}RRl^DP1ik0QJH%moI^MLc|4FzE5h%Z(%CK!#Dihf57p(K8Bia0gES5t=_6x$~ zn5s^ZH^{kAdj_8KX})!$u*uIt9OZdK2{_tCN3s0A9$m`0gs_21tvU*2cFK%4iR?OD zifrx?A}1zclyjVDbixadLJxSL)PKm#O|zO@V?>lJE)tDTSGfL6NVlOCSNOfVL4&+J z0>H~Evzux{c8ucy)Lszd=I>mg4=O9n+wbctkc8Aw0%OWu)b#1)26?ooh9i71yEugw zM3Xr1h$hBz9GX7PDV)z-IPO6sf7K0d3^qaR^db$Ww9i6P8W8y(wH`g!^g1?>3xAZCfranK^)8~j@rhqWkTY^9*S=|S zsB|;CNvgek3Y0YPC%s))E}}=PTwR%Dch|U}re{kp@wDCRNuOI@8j(F6m2)rl9A5s5 z>EA8xIw#H!e~OMWv^A?UAtwIT)F|l+|70EarmWDeG9L5IH41gT-xhP-g0!J%;n$-u z%{>5y^njgA7^tET4O0V!Lk!cgPfW(1l&41K7ur5Mg&4#)s+p9Ci)$m|o1e`OFb0=W z*tpClNqpFyphm{_wrfQ3dB@S1t^byjGwLAB$j>AT^&`~hrQ1_u zC0qFM;E%r=^H)`7YGnIdnq#u3>G&6#C(HiK!OjTu`-^hzU(>CimFR;p3pe;>bq8sW z=Mun~le)m!(wcyNQI{X*IID>@!0H*WU^qtQ3F3JF?!&ar@B7X~qJ#+;;@lM8j@+PH zE-rC*H^!m*tqJUu%$~h4goisf9jn7xQNMO;Q+uJg#q~+s5QEp>2i`de&bas+pc2dD zpI0tkC$TrOzq5Tfqmprq)KoEZOG9dCn@1vWhv~bFj+jg(A2cafZ$~e@$nbi$VOb9r zcsf3PAnuM}&S_gLXVHQEdOC>xA_nMZQWymF*tBU}c0mwXRLzifLxjdqNIiW{XI{^7 z>?3)u;rnYHdSQ{`QqnDm5o{~!&fLM5lk<~9>J6myr7i4=zzfXmPcOe)Ss#IS@vE2` z+l36U-aYfL9HVW?|38J_{|OeOuZL>I%)a|ErR)f{^hIHukIBa;sA~8?M96&`9dK^u z3KqL-gPQ6^3zIPo9bM-*fxdNUYg(oB1GrT>E6ukgG0wl?B(iw6rEzgM*+8h@)BA@n zk>;Z~65-*j7l1sW;*mng99HZl9T`u1L}S{Vg>N<5#$Y!By}*_d zHE)V6JN@?&>rIm{yy%YSQy7Tlq%u(UwV*?s`!(B7wC%}L$`FJ;QreMbg#PxeP<3ze z;_F|Kts2!%=$_D!m1`um?ht;G@@7!Pq+uVK3|#Y4foawys$Nbo7#5mPHEr;w3E-KN zL0K_OHGLC}U8bchj4mV;sGmm?4SpW34O+bsB<(}&9A5|6$8_Iya$mc3t@y=}_M<&d zG{(U0k2)i;86O^%#*U@&Ygn1ZveTI87PP$k&dEI*a!M_u#ERdd!uDld34aVdhP~R* z0r{ihC<;;N80Tzgi$OOvE~Vq`jGIt{iB~*5^#ZE`F0@u;zkRJm{(eB4Y_;W1P`PO? z>%sJh&q#zKorm@Fbh9vc>5+uYlP(_k z3#&XD^Yq%G^<73erkSjcYo@lW;~{wnEcv9!tuskEU+Fl6v>!Qay?pMzD{jJIboQvq+DSX%23 z@PaE5d64Blcdzf;H~DH-szK$LHlzA6&hP+4ejvq9?8HMGZh-GD4DqzBnQcnJo#PNE z&kSvxd{v2Xyf~wjjx{>Ga?b_g8-SqJo20LCm7h%M#-3NW_xJu!Ni#T4hBjFjMd&Kz zAxs?xP2nxwh~HDOS%-mvsyu=13ppKIC)mx?;YzDEm4!S-MV5c8p(idxlUB!Sm0I~p zWrC%5RtjJjKNz^o{~UpLz6y-^=VcwWRA^hRX!`g76THVL!?z)Ok9+m zds=r&3#k9}Z(>(Y0&NQq1+tVK8wVk_}kCM2*O=gAE39f z2>0guoI-XTdw|qG`?bST1;2;ZdhG#H_`K#4i~1>Mb}TgO^i-ao4M|N^J0m4?HR^?; z;y(l|U^rXn-q%iiNik>4j7;z0P5>#&ol&2!KcZsbOwmEy2v>B~WC;)PdehROSZr1jwa$NuOCaw%QYSDlx0^MSb0GBV>gx zpgSSga*Tq@)l7&chDAFj!AZVJ%=Iz5Z=EE{&3J298ry5N(8jv5a1CrzX=8)oVHwJ} zEY;3q+xBjs)&$qzxb~QEhIy;6Adg=SoX&p4Eb>p|Ems@?q&OK89qTYq33G=BsAZ=K zy(401mCXL#bBn*?dEJLUos_?vQL0oe;EhDlOR?QJ%1t4$>#c`YMel~?yI9ld``t

7n_^IRcRoy(nnI521?iXl6m9=|al9=~==Je~*2iYAU^{%@yklE9wT zc@|*W6iJeOIPnHj;(d>OX6{M$l`O4ce#ppV^z33B8V393h8ohUV@vTAJ-;%fsWWS+ zY^3*0UaQl>6o!Tt10lF6(BM$Xr5lXj49bB>u?b)fzv~Yuvd|>2yO797UW(mpF0i?2~)Ajd4FO&4+F^wClJ~`H|enzON z>{ZX#5XV$Zm`AbOdoR-_&0C8xjbz$EzCeF2CEkq$S-6L=E2$oI?JJ0yH3j z?I!xeJ1cI)NDI8BxAO$R7d>5)IJ2?mVV8O1M@enBYK&OfYPa{E#{?H_^<&y5DG25Z zlpZvQI@SSxATR9dtR8_8;=)iFJL#OXycr0_03hkSw{@^1qw*a!{d`sHJIeY9RR>xLQdcd*+!6}< z$BqC~vo@@aKr}sLZ*we>EaGY6Pk1}Cp%#Mfo2U!XjvfX*^1jt(n9*#!9jOsDIA&jH zkNG{uZugJ7CL`23`_t4#EL{5Nx$=iQlQy->$P5%U^7Vg)KRJ+x+p8_|L=k+^J@KpkNweyYaw z1&i7Q+aY!Q5|YlhVEyfcboQpURszMBza-H?#F$(nVrto#dNzi@M%)j0u%OS|6Y<_X za4dj&G6sx|G3Fsr+aTayx7roA0 z$7c8jCT^xGr4U5M@`0^OqiR4< zVh@P2vMQEV+RBVq>-ZF5ujy9!u4o!Og(KXsF;*n4)whJ{XurN{=?G!6leK$^6ynS&qfgpc!OoHX;<*o& z=nWQ4?@?DVw)LaOR0-U8J^&B?!Re^HT;r{&yqR0m$*~zRmim>0;fK)0 ziC3bepR0?4a;Ao<7iWcsmsT+Z3OUZKn>`y#k+61>T6&07%>)C~GoNkAobaQ;bv4zj zTmI5|HiACsSXu)|H^me){O`t8*IWkD7uX0PzJtK6tCEEN<)MXA84HPO8ZHzv{rg<OFkW2;=B`xAPpok~2q z;6! z&X;jXs4dWtL5)>I)n*-u(yzrn9Zw*t0f@5R^i~8|tUNsZj@V6JIWu%()6f<)EuqXE zGX~0^E@3-zc9D*b`8p$+&jd2Lw;n&q$qk03zc77P{lN4gJZr^v{_Pr;?hv6)>siIU zAowPjbjcRi{~?MO-bcLVvs!MLW%6;g3|hmiUfJXG*c0|a6rn+u^j0Jw@9t*i%o2Q> z>Lm~HpTv*DFK{;bmbBsPe#ljN{NiTy^q@kQW<7`W3So)!@PIeBIl=?D1WK7x>I~y$ zw{yfUO{te8rktqzJE5{6`0A`0D;l&HM3PqTc_elT zBYwk~sh<7}8`E`2B98d9Hl1c_irTTWQwTb9-__-~uHU!qi<|g!{$T^~F_1L>c4I(H$nA%$o|Co{0ztfig6ImCK9mP1c>BR-}N$Cmzq@kBO( zlfSgmws?$}h674gk1*$9c(9S2TxG|DRHjZR#{DEp8_V~AstOQ{yI`9kID@89;Az)i*o;|xO2)XKbj z%D1;t$pFcq(C~;11UCiy%2_$E9EbVf&og1U;J9z>M>RZL_~V%hL={wQNQH|(Rn&? zy)aQ=1gXXrZYA&A7U02uWXq_3TK606uh(PqN4TFLp0R4~#KaMBELgWowe06$uHrX0 z*h6OCy-6AW@>#%iczPFlOWvDwewXU@D0Sav)hqgKBH>Dwph z{*Be+x~6F|L8dIPsPKGQYL?R8EKJHYs*fC@$Bb__%*gc>Mpv5#f?sxGB?UNcp6RQo z8-^LfwDL>-zhsdo)WUE}K;kU2dgq5jT+kyiu1 z3oYw(ZT->}GxY+^07;*IvIr?@rbrJw30?nSV;mu_eHEm3?JW(CN#RE}XA0&YE7>;0 zbz~8&{K2uMnwMg)teQtP>8x8Q;@+X%-$@^YQU>78 zUq=lNHRfSGkN{CG{kM!2YotpjesV*ALQU9BZ|I%jIxgG)&Xv%(TO= z0M`SEKj2GT0dI1FR+-Rm0vN~~#6>oidWLp8>}jQ)sBB;%h&Lbbp|T92JVf%jyZv6x zmw~x0;tq?I3?+5J$B&UUwycgzlM4mwO0g(XB8u1=RlY$~<+5uI!+7M?IrFi$2Xvs! z0jMpG+yV{G@aLaguCGn=2yT5}34l_8HjUM&PIGqJyBAFqA{jHA)UZ>-Y`sX{pSFxE z>0HK@Z?^&^`NMoYAU1hjC?%l<)x|%K9|+qHtm})Xj5`e*bqD@BO*~HJjFq7VnIV$D zP=gXrc!0jc)DSKm+W}?bjt!N-k(qVf9Lb@a8*;07%NHluYGxZ2L)}$XW485i9h0my zw0J2`+t#Vz@=#FR?38YD>inW3Xtq>%^S7VBg-cStQa|9Vu&x#>c>d%cNWIx3>_bqPLAg6;HVpB{P6GU<&KkoMh(18Sr_rW5uuYvWnP;$7%3TpgH1 z_fc^v-cFh=j&}r@?dRS7i#xhT7v$Y_tzbtj%$Ja8^O8mMBjOC zeNfDwoT3}S)h{U$)<y0KZLv4U(j_W;CkQ@BQ$i*%xRT70Hp)A z$9T@Ea5o`s()uXG5P%Ldky~r=wd(H&G+_r9chvSj{p2w0QaroTdxo}Ui8qalYfl{l z6d4(Gq`d9vef(M}Ss1#_^IGB9+!oA9dey;ER!x@Dx^(VDsezHh)tc9wHXEQ=<}~%z zQ;5O}G(~X>-NAAAVDKKW+|f-kggc=9VVCc`%>fdXCLl*EnFQ~KjIfCERP9~7sMn&m zeed3*Zca6o*{NNJ8=|w)I7^k;w}UDUqIOdXY#|1v!@{5*bjO3-_{rlCilQ@So z#a&jnOQ=)S1j?Vi+I{NuLMj0uzm>^5+IK6YRIN{~rG|!FPib{3B6=2Z&3(tTkOMu4 zfAhRdFk(2XfwvB?GW&A%ExZRE0lln$&H*gDyv!%f$YIhfjv3%JHW95o&aHXcs-;GLJz9=Z*E zjh}LhZn4rGio9BXueM7eGrs(4mqORy`}_A&=_-=ooJ~R4eS7rv17|(vcLKFsAW(Mg zn%P)WD|zDA+WZ$dj0;2T!P~{nZ%2oBP~x8g(mhWphsGHiiHMJCG;8Uq=7RRSIj9|P z^cV0joo!?ry!J2?q%AMcIEWp^jW(#jK_p$4U{9=6=d`KTPPs=P|2$9cXbo;KCyu1{1PmaFO6JCP1h4_r5Q9^ijL}MHtSmvoxZk+l*34Af z5#|``!6%)BhqE?Idse%v$wZJaJIBX@EgIe+^DW{vqcw9PTgy>{_wX)KUs`$1+~y2{ zEp4T^DsT1@uO-%3Z`28pHvl(Mz)jZqMhy0OzsPtO1t=5+a9TiHbOLTek$#Ql10_e_ zv6EL$17jZ&Dh~(BTld^&BZNU#1{t}`G6SFq5 zEa1O3ta_Xq|L<*9vY9&ez99T5`w`Fu)yEKG8e&S6{GuFgx#>DG$2J)q&t#w^kXY+Y zfMHiP_3m@)JqXpE1P_-ah0RMT{ws2JIDPjyT!!GrBc)6*CClL(sBU!fYqAHHvIwyq zF;#`VH|UrKv!BuF*i94V_i86cZhI{Q+41P>vCziWMCSa^$5+ys$+#_uE7WDK` zr;*XCDPaPKTdj-nNsyL{#VSAu5yDQ=CK;VVO;EQN0hOE_{0G@)+3)khr>z^puW?@C zT||HQ=T_R0-)b1+5}E=+150O)S9pBpfW2~n2c%Y|PjtQIBOFqJN@3D^kH4xap&5{! zn?XOpZuNF7m{Fj{Qs<#z-xt@_eAuVK1Mxc@yR(v{3V{?;(s!3b3@RD0Lj0n*GCeA! z@}HJV=_mE}V$eMH6=V%aHrs}?s6h&PC2V?=^1YHd6`jnV^^bTd9SauSW=y4^#ha=M zL+crdwVX|&wYgFNSY4ZLBQsqZxZcA=JTpiRL_**FPEfKZ)med*1(h7Jsji@q5^m#FID|e5*tcx|fqhOVMxvZF;yyRUdJZQxZm%oAy zNAKA*San$s4Q^-e-XjsOWWy64>pw zj6`?;r?W@m+BS7lN5FdsP0UR0?ZCtLvd6QBe)dY4Nz$f!(nZ31A&E46QG%l`Y&bO< z>eRj<3i1^LkW+PfYk zRRd1w^W2uAQ_6sDF(clqt$^~qoA}-UUoktmebs*mp~?I8CCPwg+IBFHjq!glg?r4n zDxC22GLwFs!(0U!5kNUaZ>&HaF84bKK!IP4i_XYuSXlU9EtgY7onH9%RjYx@h6)~= z?R3Om-c6VB=_cM!%mTVaB4}YEuW~B=Gsz>8bpyM-aL!BhUvMsr+WoUM^>!YR>M}s@ zp$o~RQ?q8_X6?)&)<(9IGJA7QUcs*?BO%eRwK6At%%|ZV325k>f?ss)KESjqs1hZ^bn~Z4+$C z6ztEtM#qq1?ERlNoa5jo4L!7~clL}t2b~<%6}1)`ajwU}94A6*EdsZpM%Rpm%M#r4 zZZpbLdQ9WO-+#KP8+X59$r&-?BY9V?+yRJRvgyz0YEUN*8@KMZgJ#3MIx9n>6uQ?J zKGK;GyAzw}nLPz)t(<QaS= z;F*kIViqfiWfeqj)vv6lV1~I`0LXib7AvsIx|eAat(UE~%_$c_BR}?E03h?D5X=v< z%29tW2(TNjMn!U6PE?!Kp!$`w(?)KiV<>xZkKYcoy;BVCaR?$&dI_E*9*&s#oA{B3 zf~@c`4!O$$YWS>(?!^2545XhLHa{5)KO%q%=PO_UT!ArMQ!b{OFsh3H3a6|H8Yz*D z)dTe18g4#0CqNkyvT_mhe_kev>vhi#Dzj)lpaL|D`39tNpnC@umo9o!20g3aWeoX( z-3UuS)4oX_gg-R}yU~V#vyWG#k^mt0VZY0HrW5i)j}{fF@9w17uZeF?KZ>FA?=1G$ zR<(A}4rZS&S#azJqnVVDE@1ZHA%M7D1ydxw%C+MCfyjH;%$C$1Kb-!PJ}|L_Na2of zt?NYJu_0HrHrsdxV$8N9e4ty-)sS&n7uChGCgt=9>D_%G6~Y5hskph81jO0n8W3{A z=ZIT_mMwtzCn2io8T@iX{Am2Id^U~mJm=Q|!Li9ddQs6f22TNoZRVVm(;)14ZK@6L zR~hqbYxPJX65={kz)sY>mdt;KChiPaT(7K&bfPgkQ=r!V_iNssUUpCx^y%91x7VBo zh-<&;C3s*CeKH^P%v9BV&cUrG`c;(3`(8A{WfUOb&FQ#3b)KGz{rj98fwQ0UyA{OS z+15V1Bjt8bf>WD#eg6Ve_J>Jv*&MXV>g8a1m zNZ_FLF>IZuT_KbIK74=VTfiFuN8}}AGqeC%GtH$5XNQTQmBais7j9Pesw1%-T~|yk z(S=}y_JwHwD9AdSRlf<5ybUoF0H)Q7>R-2&+&c7pRhFL5wrX27-RD=YYviix>SYt6|Rj4|n+u zx4B-zgaTqL%+OX84(cf!g@l{&j+mVvwEUzn8vmuI0i>LVaY!UQkWA!AUj#|dcSTfY zNBDF0Tm)PU*luWmpJZ8 zPAdp21)U{kwd~HEv}*m#soq#usA;Og*9+@>5Zdo6G9J|U>@3*l0D#}VT}9B!nKS0D zK(5;IJh{2ljai-AY&=5HByuF4wAp+F-2Zf)>#4vT=bg|(<pK&YZsp`HSKVt5yNQz~eTc)ni=-;hnl`*7sYCT^X_H;x&5AWUxhs|7mqiM1@M|kI zlC7V6#j*yiTpkj?PDa;Fn$0nVfb&2@U{P?hD%HRDj0>$Yry$71&+{rhs*N#N&h(}W zmjj+Ao{I?<-w6*JO6_`UY+C9Hf^C4y@R4i?c#!)j!$}9w1>wEY4Uk`aZ8+4 zR@)X+TQukh5Vz4&J_N(5#}%4QJ+?%)Hlb!$h68BgXaHmTg7A26-~+VdUkYU$WO}k7 zD0*?Uy0z;$9|3H92#hmm?%g`{t>Q{aj{W4ftyrZ;G(J$$^$@NF`MH- zpwO)^Q2kR5{s6?d1g%|eZij4P^kH{!E|uK$EuGAYLC(>wH}GRv`Nb{qRs=~46dd2O z1pF?-)OsEl9D%9Hzenz(1=6kvHWjqh>~nz%Ii0n24{nn2x>a9P)cOLST|oW7-1y1y_R)YU(CT z`?Fqv4UaSGxRQz_yW{uG@cl<=~z+;+?7PjF5KrP4>%xJ^_*HUjq?^C+s)_Qk&JHA9lk;CcITJ zwd)Bne+aak2W#TlWjSfW1vK*(&VsMd+0(LVsrpt{7qsTf8M2-Hcq7jmD7b;(i#Bj3 zfeA$Vb9fiyGTr&8g|C}4v(jP>(R%pwDG{gPK-^p%#cyaMAZSiZejd455EL+1A4day zI*m=x!sRj`eR{a?gUpuf0}RcP+@cvRUoQpk-)$CP>`wgvQ?g0FabljDwdT!EIYJpl6ZLrkOHvdR#1 zKLE8=1!`~~D2yXGcwI}sp^z3IIaGImH}>nd34mNX=0uCMU9M51?SQ)}Mn$%QLSFJD zAWU4$LcxyIaYhaXuT^QUGGz=!=w_|udk{_zyNF!keGY#rX{xX@s|UPgUj@)H-Nmjs6zvl-E{H6jqk_qCpyAa5 zh&v*|(3f^i1src;Ywn(ZyCKgb4v~PlB>Dr=ERzA+&k@mu{$ZRqVDVBwAl&rz>Fms3 zt^)eAzx|*bUCvU*N;z%&V>yvrdk|C*@Ia=n)B(=ccT|4krLA|Aa6oA5qH29yr9owO zL29Si->Z`v@M*S8bnnW5@3<)d`tM`c0;hJ>nym|u%49wvbhzwaWcL?2!UM~)(nesj zcn|oDum4lP)%*nQ_%A?g_vrv2>yE^=1aiW(@RXt~`GP6IW%sPvwlL7}2!8xF%Qn_) zZ$E~OLm~Cp3c$&Q0AXTw@BTccx0lC2god#PEdVvq?1yO@68R zE%BV8ZJaZdpMlYPcd=_0qU|H&N+|CvRPlqkQPt4Ezn+m#YXXUAFm~!iQ}L}UVJjeY zxfu7vjdCui%A>lTLe7$Jr~}SDZ7wJCV)H&b&uxS>?AoqtACk2%KMOUZ)5^^NXA9Dh z#8M9=kac)Oxfk8k0q>Xi9bGt$dz2TJJ3E7S2^^V7=&Z)P*(b$-P&*anbeM!1a#@&O z+jKt-8QV+7Q81?H6&SE)Nh^m&M4f=GXK%jK6mSF80*Hw@0Xw*^(=E8rH?ajc^5?5T z0L=&igWFaogN>bXZi^u0ttrrvfU5t#3}?(i_=?|85G4-4o(!NHxaeClSgl!E<8}?T z`f>Pr!kupJl>8Mm7-L4H`)rPOI;>J&gjDQrb5o4$(NZ4rL=@UHYfK`_T;u zJ>dU)^lg5V760$|!{n=8kI5#FxGZ^w)C0~oJEp>StGL+AO#t(2a%a6s+~+KOENFt% zA7&!K;97Tgv(Ni@zLSOxYQXMO98=kK385!UPe99m<#$VVgBhT5)KY)ouiB`l@}+q& zgY_xi4tZ6R@A%Hh+?!qZTOO%j;e*lfj4hx%f_1~PQ%S7sDKyu<;3Zo_l!HlK0llE$ z2YXn?md?n$7+sMi&-X{9#TI!;fYes<$t7mCeGc^Zue^Ra6VLl0A}_|X6)&ne356%n zcvb_tmBW^tQ=sXiPG(iqbepv}ROP(y1qMdSfFSuBM2xK-X`tDRZ^=?!uex6o#}z-hW><7ec%Q*iKJ> zM?4dIG2Q|%1k*`;O9+FB%l@QQJ3v>3SxZjdE3O4e>$9>7lBJKK)`h4< zaq;Bmxm#csX+Ie7jNo@h;U3ed`uWmYdbUH4aQEcnk@FSj#2@i?nBW+v<-&ynYv&Fx zV@+r2W8mqEO=={gcv?fP`z28nBROgTx({{ck5|BZQ9t+9R*5QJ3fkRVAX*MV5-okKkeWif_e5PSs%UEo--jc^Z=YlEMgGER7Zgj5}q zO$@DLSU!+hnu7e-V|B;cQYsZ>9CmhO_KaAA$bE=%7@BVm)CAXc|5=F|po{ZE;&;gm z4EzZQs9e?>*4yDwIrjfrNDW423t7AU8+o#R4;6ITSKa4j+d{E+ua{Y}twFRosBKzM z7T~Aef?SuoPPTr_(m4KC73ktd|9$bmw!|M~1p{t>=j6wTe<(Q0`@|ZvlhU-7bzlK= z!NLS_2=pvUe}Mf&9~>QsPnzaBo@18WZ$rj_x47_Lvl4nZmU6iVWV=@=9myn$=(q23 z;;Yqr0xR$i808+VPKr%_XC7Cl2o)RxDN8Kj5>H>12qoSCtJ4xwv#7`0R7&<=SUz5) zD~|5WO(dt}%4wU+w(c6fvbt%+Ex+wvXJAjAr8MPTE2BlvbD8oh?*VH!i`^XzL8_PW zV3BPj!xP=xyD@nsTtx6m@#%}##x8sMzoPA3@9mQ) zrF|y1BxDJEcveS_gQ=f#uN<_EY?rR{uJj?S9J!98*)cB#VLZ9*a()}2h*DEY|%&_pJiAoH7`+xOT8tpRzott5KF+}{sit}>7>mm@cM$*t`n@Ec=5bnco&Yy zb#Ha*E{Nz@0ZbUu#kNW?vT(W3Bht6LOl_2fbB6$>he9Z=R^WRJ~74o;rL^&Cp=y{(uOh zouR7G#Z{e6qkD@whm61nuOPQZ^wnvb!V2?t6+QcMdfi{y)b*+gTLq|B{_mOm|I_=NG0+a_ z{cE8?5ffWDyU9B^cIUgP4&Ch&89#&nd;cQs#)O==b?40;@$fS(ZUg@ERLlD|m*xxQ z?67@tKcx7q`wLF=0d`?|8X5{K_sTNmrp}FjK{v5`!X~`II;PCzKyq$I;jIPQe9N0} zzbHT(WZn5gy8NAyC+YIXS-pb66<4>i)d&BJiyti7vVywt4J>5&Uv~^!)k4>{>wj{) zJ9w-oI+9-y?2?d2YHzkie2&mIRVmEbRq*wTGxR*2Uxk514}@Kwd-5Q?ChFTWeM;MY zR;j;Nys3%qNz)6j<9;}MqG*eO;(k9V?Mn%`-$=00aX-9yGPd17b<@^{roBKGeR;;~ zDO8X98G-x#<<-5W2vcpzFGF88*8FcS+81_Nuq`5e7n1#DA?5x5;;P^dsq61(7Ht9M zmHWTj+mq3~D({NEx(-P30X;+M=H0H;*DI*&p5Bp&0CRmT>~7NzYBW48_W|I^oZ68WwT1&nuHVb6A|MscVAxk_t^IR zV=ksDl^GLqERCWqgVWqp2h7$8umJYg)9YBMbnZsg^dIj;!-v*v{KS9Q_*v(Z(PXW3 zGn!uG{J=nd&Ayb}=WWsl)TF6j&K+?9Mmk1QeefeVtz1-L=S{=1^qLs{pNRL8{j9%d z1T~HJM@A=8xXX8UVSgEi);L-B+Q?Axa_=JTa)G4ncfP_&WdW+zdTjgUUe-$Ct8wQW z*Zm_-uO0IC|2X7HS?6b^iwkNbr^GMU7T@ghRuoFiOt({}ILr5TUhc-q-q#BaM6;;I+3nsl1T(>yvF~gC{Wp!jYy!)Wf^#_MZrr)9+s;&F zvVAK<<6q- zi@w64#Ha(_PQpYp(@ljBc4d79yV2PDWNi&Yd4jMn?1@0w6R8eO?M=T21M{%}|0VXE zxBKH{;Aq%i>s}WGP8r-V?DD^5JY;Gh-4A^XJSFhAz%s6aTYxNs-7R(o*S1#xP9Ctp zePC1eA8_i%^0F7yT1OvI+iVJdF}Jj1G9497U?5=Mk+L@X zr7b7*5|D`9^QbquSJpvh7GqPZm@W@KuQ2%9U%uM|oX%Uz|H`dvQCrbh-K6br2%Q0k z{|s~wcSKCBhOKU_%{s01P_28b{pV%gf~alkz_~-l20JfThiiauDrh4N6<4t!NNDV3 zre>WtJO9t6VPt9C5pNUxpWqs}wQKzINflES=rDd^f>0NcwfwK>WS?~`^|d_@_|H8L z=ZuTm7BtmfcQ190Y1-r;D1FLh29TFc-dw!`VqLdg`j64cz`GpnylGq)Ki8)4BC*~& zwqzFOtu8dJUP_>zy?@>bx<)2+jsN;!2e@%ZVEMbj@lRhY{8AmN@k(_PGPJ&R|6yo$ z*;K6VQ|sggv#9G0OUE~D%?KY}3Tz<9DSH2czYW&=OXm+;z#+PjE}svQfmKWX^Mo1F z71%cY=me&dsKx)lyMq@&v@!e;_dh-d;{2=b77}%R;q6_i8o&x!zi>Cs6_Sj1UZ*-} zZPWi}|07@jhY!kRPjRG0GfBf;N~Tq=yxTrsFPwVQW5+B?AUlajQ~S>ybOVn;GhCxF z&Oj0TuG$z^9mAE_LOD`zE}q@DjgM6RA#)DtnaRTUQH=V;J5THNZboQ4(s6D?KwMCZ2T!$0fnChAtZuua2!ER~z zh=2-ED1{Ex)79|t9r;S+uvHh%5RX`}(c&(EXszXXjby%X`MUw&yL=^va6beXxpJv( zoc{7Zak`tr?g?&)3Fh9>+1j1lVd^-PR$rSyntDs5lm9v&;j%OT6#40DG;oJgz#XP% zgDq*HoFN_Y=MDixFY^7}+htlD#yCA{t;cMg9hliIpib8&wLfkFb+TE_ zfJ?5!djRBYM(sjer-^wT0lw(AUn@B31 z)v;RaDP=HxsnNbbp>Zp`06Fs0eZYX^1N{hWFX|8BA#;DxgVs}$ zx-~h6I`KA4jo>2WFYm9_i<-h5Ss0*A-6L%_SIrq14544oBn;K2I3(~h9qQy$&pD92 z{2xJcZaTf*1{j2I+s$5cVSC2$oi9AjB$27X@(m+ z{oq8U+bQP4D#p#YTzyc8@rvG~K0%fOc8`v#JS%?`+i6WEu8j|0+YA``Howle$c{|! zI^*xDE?B7n>_QZ}Ue<_Ypk||yY&wT@i)KJ4)MaLrfkXTty?o3nR{bKEd7w8=Jdg_N ziDA2ddby*)scsvj)l~J2*6-ud5CNHcv1?82orK?U$7upo0{6idij>K2j+31Gl79rI>c1kc-?+sjQg3? z4YwK-uo!2DS@Ynb^jB^!pT@~Zy7#8t@}piMcQkPD7yj93RbhkByUkyJ$Pr1Yp_{;# ztr{rJSdDr}=4(afgb8mj;mGR{*zZ~!0jp1M8E=Uh< zNe2d0i)$YqnNlJ@eI7Qx<6*a-6#hse02V$2q)gfn#5nS>^x{sM z7;rU17`i7`pjrn-AcqC}Z<0PW)dN;}Xb8}4Er699N}JSgY-udpq2Ij4OTt(SYs#bCq=9;Oq9`RSA6v-3*Gsyv70kkRK~hO-^_Db3 z-jpcIi&YE6I^zg_6A2W%zSKO9fSh6)7IlqwY~}ElFkfveS;ti%7tzHt0xA$y1NYF5 z6A&jQneUo!REOJXqIOHsF!6k98YpMs=VWO?wS~ zZU~=&d|M&2!t4x6*F^%i(5Zt0bn%O}$5*arEj~al^~G-Z8Gw_U`@AQ4^AEV4MG}r_%j6jLcLa8nwV1sUEX+y zPyFAXzTWnVU29e!<93UiGcqCJXnsiHOnB1PMi3I}Ou~z7YOnyD`^wgb#IhO{Ak`~W zuFVv+laXqmr7uYX^}Jv=6S$<`NL7RTA>HWn-=kURU+w%?YrH{r-}lhY&&2h$m%1q? z$gD1Pwq7q12|2=YX{4I-tNK`D_WlNgbU(y^=DET5=!!nf!>a;28n4T& z_W)|xwMCzo**f%AU+6}2COS@UXSBBT#?~Kh-XN0knxF7%51{|Gv~&(5^rVD*{1e6- zTE_X&Yuh~Ls@w&L(%So(%LdD<=hMz$GFLD*3!h3Ey(RFEP5{PfX$4!VFaehBGLIDy z17+^>@yGS-5d-89@Ima|2?79h4gI5s*b!LMVbZ%r0oSda-=mM9pUpQr)Zp#q!*8pKF_r0Dvqy1Sm?PanA?T%XmnkiCi0jf$A&A#{ zI>)`2%d`Z@r+z789RO3{QJrc==KY(v6sEbtSu%#|jJGiQ$N1VWPEc!_M+)@?<60t;P8Y1Xb^|m?O z?t3&nqzh|`$AX7I>ydERiTdRXF7Zz|ns|TUkBB$?nU>g6qaLJfDU3kH)6@yW?BJeQ zP{{TyTpg%61Dqks;S|rXS;y5OO3PtK4HnnOIqzN(tR@K0*^S6cJ>S~)U*jyj0S&Jc z*+#uAolxY`+cLBrnJs3J7?KXJdlblhYLL%Rq$QLM=nQPLHj~PhG)P4EtYqoYwJ?XE z29?~|z>M#xL9GHw2zVL0R|a6MT#z}(^G`D{OdG0=ar;6$fiO8mUviu~%8h1I_`}P* zcCNPW{CsoKmUkK-H6D(NQ^S~aC;t+oLwjC&=a53-6AR6j4$VpCe641rH=VDn$qQm? z^?n-F2WF37#WL#UKUhxVJgEj93-QMId>2|nt8}0kitjj{stHR@T99L?&ohdB9Fic> zC}(<6ET2S#BeVwh0)xWdX>jqj+s4)?W@#?;H>C3gf)NeTLer1%0;#sAQROqnEunQX zFGzvgPWfah5E-c7=={x~)TUjK90xpW(VHGA1#y zQ^V&_O4WDIke|NVCWZZgBJjJZ&t}Hm1k9&Nc0piu(X?RSl001z327<(!8W9e_rZ0G z7#&1Q^01`lj%|)O25^Mr^^H`i8uo&!7jiKAmCDaef)heae`^` zzy=`Vklo9pXZTm^L5AyQ2E^G{SW`MFXaV`m!DD|iytASWI8@1zc8_-UB7Xp@K5Xr74z2bAA_GCL;hUL3M=!n5O9M(nMT&JmDjl9%a~N&i#hnu5e)(l?N|G`CgzJ+wrPy#K=b$ZOIX{PtAyWw`6805ah@=mY0KEJztWv|=a^ zLHJ-klG)za4`|X#<8EYo55NOsn2w{Tn4WoIkax2FBZ*zgh8%H>*!8Q z4XkF_V_#tcKGaG&v*y<&eUYc{pC}(EUnY^*-_^&-auDKj<4HOevx8jaJ|FW&UAQNe z&m9qX1HU#cU_^`y>fY3luEoAwYi5V7&O|dnG#jiG1iGAc3Ykvf6OO`jKyg8*x5x|+`sO7bPxkQr7onasXUI!vjg7}#040q_=0_uLIR57EO zs7xL{stKs(5d@H3*Z@X1tq@R9x+x(L(up>ZkUY4$m z@{M ziF~sxO&#JKQZ}Q{jhbmbzm*qIQJ;F>V?%rhdw0VzhVXeJ+>BN{Bc;T`e6>6kWU!`* zsmMQxKz{EQDnHqq!?71uiufu&(kpq{MtxvsdV}9+hbJ79_jjS)ri#=m8Aq z#xo98gdeNmURxK|+qdWgEkZg!D=k069jBYG5x?}F>|%xo%)MZ+Kx!B{RpidDX#P)D zO;rlfFA;JQeWKaGX-~!F*piejXBK9ap8WQAUmKKu{-u%YyJ>&l+L@h;v z5lv9~)BGYs0`hLz<%RdBBDWJO4qW({aT!Aev-FLtny%=wF*nk~Ae%$R*1|3We@cdi zRj3#K2z&t^@X~!8o7WFg6A&WmY1#R;C{cQ|1M0Tz@B%ittE zkLYTJjnGMM62tvIzxG8Xz3t8s=wCqn#U|`j>H%Sq40y2_H%=W7euAq+|0g z@tA?vJc-x7Qbm1TRxuRzX*wB=OyG`aW9YAHRS`S3I|;zld;n8{PtsPY2Lb^1r)W)l zC=ghvZ{w)}fTz%}+et0hMacdlbLSbp-)F4pAV~@HWu`cqv-BL~Rj7Q)61TN(i-%tx zN0zVcR73i<*vc?R_dr5)eqY1RSx^Cs_I!wB$M^Fma?JhT%oGo;i}&y6T2%Vt0LJkzt6{e~&9yEYYWH$lh*Nd!;T?MJF!((k|Ur+8_dVnJlA zH^hTx*N+u!kc8j6wC0Z)KY~Be__8Yr)PHyjxWw-;qUeCTx+N@aOHO7&8 z_0wjHKXXDkp}k0XI`ImkCjm577y>oIzCwTNj(&;(0KB{kAE$wN@1>^Z>%C5#;jDyq zY8bd|6uEXFkiy@qZuY4i+&EAKx*W{l>0th1?aK!_J@mS(xx3FBiyYy;C!N{aSQiQ1 zVw5sE5{CEEwKlaEjPY=*d4b0n{FE7eCKk2$uptgu(u=4GrET$R(6qFj!P(|Pyf>Cb zH`w)`kLoRwiuz5xOY9-%J|EqhKo!5~!Mw9)wbJe{BykTiav9>yl~@2qxi1)6ME-i{ z1UF%@2S?1@*g6_`2k!`-wHqO5kau14Q^Ioql`DTMRzN+|lU55HD-8sB>V1JT2f4Pz zA;@d{qyv5Wbea3Eb6va%7vaHn8p{K0TTjxrlWc(tbS*T5iD_j^&;!wsWlTF;p@txw z#3kwmV)>c2jk|#uD$NVf&#dJ4<2$DRT3Q4el?9}D7SDl3W&U$tAf%E({49II+8q?= zx4q!>yvbHy=j%#Vl0O1sca$f(a&2GGx_Be+%I20E$GA%;A2d|gw4(g7nB-dkoJ1FW z8ZD><*}=hc&h^^fSZ>TOoK3>k3}^NTS?obTE(AZFX||B89sJQYCh#1jig~Mv<)QDj zAuZ__skYZ`G!VJ8K&%B;dUG!qo;#`7woXwMdqt3)6xhe_Sut#?l{~i8D2_Dm7E(o{ z(3797Hv^WPA|j^*r|$O(q4WLtOzI3OpK%U8qH)Ohf{T!g-lrUaM?`&_>$P)=Cqr9T z`d+PRX;nT3o&dOhDAQ#_B$%(##|nM`zXv!dzL1c?3@#Y!CF$iQ00 zdveM>UNHKtKIH65qXXm=DKHbbW;Hj5|aFZSUEXvPr)OvAX)(Kw_h;kN<)9X+|KFn4QfyN#t8jzD_3~ zrpON1xDn2(%dbD08zl7vj(68D@|p_~B*yjajB`ifiB}6r(awNmbSc$G)3r+R=ceOk zTO`Lm-{J&6`D`ii2I-QE~e!XIarawQYu<`R``$CWoVd zrCIGy!?`_s#z*aZK-(7ckSyCtS=X{*AFL)xdmgCtY5^GLR{2SE$}>LlM5(Mm=UH(QV^=B0MPdM$eq(u=2PwY1kuXl;gFnWC8BQM@M?3ksb(%@%Z~3(B(E4MK)xv8KNt&aBMM0+fKV-0JProZTih-L`w!s9f=%oyy!PP>QB30&o2(SbD&^w(w_~$ zXM!(#$PQ@d-P?>i2K+s;mmIziSk3D`rnM&3sNcwR)F1h@vs+gF&!@&9?ZYKPz=aYM zuE2%BN7gq0_lqGyEXKBP2!x~(BgkEBWiof9p)en`izCj|a`hpvD&OH!G|hW~ViSCpf!J@reE%}kZ?nUww2>OWZZOkJbV)V#`u_| zn$?uzpO?}w))0rj@i;i0xGUb9uz~7HbZ_I%hF^wz%_J|s^L`j zPT&N5_o~e&I=W^?QNTi$XqE;N;T*V5nEA~JKSZ2;GYHM2(}lDXlO5RiK$>fI;}V-G zGse-xa{4u3U!(g`z@v_L@(K1;4J&U8HoX}SkvB<5G!8R4JN&?RVjdOFsYB1bH9h#WZ%sZmgd9 zH;2KazeGtMK3Yttqc$G&@(FsxqkR|lzmFyJ)0v|RbS)q@4xJ?%Viy=5{38T9H7-3v zqNn3eSsz=R(dFFBuV9E+#-rd;)G6fGa1poNw<-k5sQCFcQ!MdpgRC4entB%Ox_he-nbiB z{4MA(M_Z~KjujGX2#D^{X>o`n{GL;L@1We${)4SUXK$wg*w$fRn~|W8!T`n=bO`_J zdP(31Eom3ER{z!I0^7u$B(^!=s3*1eA5(u;H<=-P5tyS4z%R}GU{@(KRx$OcsgXV<`s6@%|M zWU@O@li|mI{0!29Jt_+Mo4HdANU98lA>c1n$~-OubD-Il%{(rF z)bF&7O8Fb@xKKD%Yfit*&#oKHx5oTN^O>P9OHWJ~Tfyf?1Z*?a4GypJ$XupLFFPEW zBAx=jFv43pE`}{ZNrtWvxeZT1`}8R*hdAg%x#n)hdniC2W*f(H`Y@eL%#Pz{hN9Q~ zHYty3Cwn45(i^9=wyFTjv9^&oqrZHc184Dn*MjVRmoMM$U7Xu{3a0tX<~^gD z=W}1$72I$xr5Sp#Q5y2vS_(}*Vzui_){uJ@=^=WkDKah#Dhc}V%eKITz0 zCg$uW?apBeKD6aa=yku2iFjcXa5s0pzJcM5lAVh*xbi%D>J;=$f!qOm9_ zqJ>CxRd($1G58Rod>~Yr4H4^onCRsR@)*o*789cs-mZ1?Ol3FvRtnz+U)HHYA1?Do zu;UF(wY5sJKu~bVFVes#JNX6mdjmD2@QI`TA2>?C`mt8l4X&9Zte0~Xcdr1w(+|3Q z;MG(GQ9p*wJMQHsUR%cZ?n|}a>qAbADOM)x1tuI7iHapu&7(iMFG;jLI0I4TM%M>o?-KjR!gE+i%5ezcuy=gSzQ0 z@M+j#jrgPG?T?+GM?G!hy$d&xSNYx3&E8cFx%?w++$lgNisbYI+Gus-UMf3# zX#@^3W?5-X?#oxwKT&hZ??r^Lw3g|^kXgPcREDXQrs5IF#BWN-_!NK}Bfw_rpAs$x z-Am0rDCR`VT=Lmmp%x5W&zj%Wz91%}SMXEy){a6oeWYGVcl3mSn3?PXdRRN_b0Z&< z8o~Ct+9Z3Jq7Poxw$&T5;ADlQd%(1qTu8bVTHC+e0|BqKYZ6$&1;Xr=6rcE**Ys4E zg0EL&GjM5XDw9!9&CfxG42HC-Eb#BhyjBLv-pDDcU2RCK__lxu9I;+w+a$1wqr~3t zGm;S8McD%e!?Hi#ka_E)hnu;J<@bp8pNu@VD4ri|-89Ct*_x2@eV9M}3`@qeZ)6X# zt0PcRqDKWKyE)~a#@aVTPJny0Z4vp-r==@on=%WuTXPNbH+&(sc`px}N#gc#Fo=4W zdez|tqVQalMpK9fw{zaXzc#|((E6nTIT#zTOp8#HpT-kRYan@7hq=P!o6@~!!R^GiK5R8h$RS@*<7RpQ- zQohZOmYcUONuJ@15S<8a(!PX`H?-Ml#+1Mze;lERniAFXAjWpy^R|ImSZWJc8%}~z z%dj7n%WqV#^)V66Aue%~>vnuJ!MHG~2e}yjcz-_XFMdZry-SnK*R95l37JuY`#PjP zjFIWfUGB9>cS)HZWmeD`KARwIoNq#(bmpP z9rI}_yH(g=0zMi#O608a61dN}Ge2+5z0WuJ_JHXv@;lp4S-Z8qLFwzmHK5z)pV7te z6`yJ`?|cn@m6L+MPHwT3WggX>+Z=0@*VCplvF5bba~nv13@K?da}wid!5XElujd;ro_J z`w*7ZTA`S@`L%zB(ChW~?fs9?#m+7YNnTwP+)8UuqBt!G`WU><6}VIrzT8|PDs!yh zTkGz)fv#E+QruGkL6C^^CRyI58+LT*o0d-RL}tM#-KI$=(2qHBwEd2#jAOy(g*R>+ z_IiI?NX*F#%e^!Ky=G)vUG#`(cU88;M>OwFa+8zp7EkC4dY#TPQ5fG8o_zi|<8F^Q z*<#&2U-yo7&3(VB+_$gsU#`|W>?OAu&)vQyB6vH;JnM3s@m5{*oE)z3usM(>QduvA z<@TF7Bhom$2(EhwJbyYjJ&tZ3WFBJ*J{|Ne({HDVW;ZK4{%^5Bn0&oP*89NHiTC{q zIP2N$%dPdsoJh5uc{u}q$JnZQf>0srBiUXJ_IbIDI``SFUh8Zr%X5K4E_gmB%dZg zo1W)p?wbbCoqk0df>t*(jPSqG7~yHu7^8qhw2V0nnPk0mR_c8RtFl&e8#~%teRpp? z_^O>E*js_S>D0v4zSP8P;NE99&xa&uoLpSJA`fRS5o_{m{HZ66HR$QN(F1ZNmmPe_%DK73^eC}o< z*RS4NfZuG*%z9S6Puxr!f;3H(7W0g%4}XqI1I5^uVBzjJwH9~@Gn{+;&S0kp>FT3! zBHGB~P5r%_VaanXJd5NQ{bbQusPD@m5>8g3z3bMk>vlG{`#0%SQ*BKmA72_b!@=_c zhpr5rl$TvlkDhV4jam$!@ho%T{5`@w1fQP5T)YKKkd^*28~tU=-06o$0yiwTjd%X) zc<8Z0UxU{_0Xle!`p-+ILa}~-O$tLlTY>G)a*88`1_6wlbY4E#d+tn%%1ntks7yT| zLRbNDZR5OAp;zVQ^S)1%PW}Dhlhsq%^L^C8dsC#a2`R=GSrR>X69t)Hc?&eMe~woD}<>ph2|8}F!EWC&ez1U3{^`C z-nTGqjA}!dmm-8Vb@tHa>QE7^3#w1AO4ww;U{Z&s{ofi%FaQt@d*&xVZs6(a-=O>j zJ_t`#;XIqsNiq+WpD~emgGwylCl-zWb5X*VHj(B5lfvj`sWg9HG}=G?I{Q#25(iZ{Z@pu-(fyZ+=%OVhDbo@k+lH%~UA|)^R_Zzx)qf~uTmw1LNGMPKOD$J&=e2P-IjlO8*=fY!;)mwdYQO+JAbMU9F%;Un0 zl)2*M4*kbbt~RhoWkSC6a{1GWNB#8Bw2!5~{2{4JH%-b)#izr^s~t%dDU%&F;}n=>5e!A|RL@m4A5SC!K-9=dwBEtyUOt~wNB2m-I!{OEX$?_8ox z5gl6*;e2>HJbJvC3dO0xI;>)|8jL+6{Qed^S$dFBEF_o5l0%4pF)=cHX4G2V{s6oK zKVq!<jH=nlz?Y^(-HP=_O9O2d_f!pIxmg3>8kHqbxG;B?l4<8 zQgb-a9F`KQe6x~ToZyKN{bH!=9R>MinLn?u(T7ubzo`p6a&gL>t;f8cio!2L;ws5ANY>)9QxH{-HNp}yRAVKDu!9*C& zDrhta)}+!nLl!3JFpe%S=?d!S=ML*OdG=#n5Bm(aG4i&P}GmHg90FD~GLboW?OzhP3#2m2}U_2IJ!q%K|296kF0; zFVX2|3J7o*EWp>>PpKp3PQ)BJI4|O-KYhs|=nlmRJoU@Jhr!69DqC zLk$7z-xC?Gkkmk_5=lxRA`tbsNx$a#F5Bj^CC=CBZ#MCy%ZV6x`C|nSxb|qBm$RDq zWKWgcgI$w#tj3I{Ih!L>1`YU&LAd8ppQ_u2sNkRLp<(670IA+3iO#bZL~IdZ^ZJlF zO=_b4$PHl|odINx(9|8!n!3ni1TE(*I$;84!*O~?JXfsK?7djz zlG;@%;s$nxjaaYWbW@WccjijxPg@T1&r~rHZ7^g?y4Oz zJGK+st!UGT?>0*Q5YpuVJ6{|Ev&6TQ^?-b_;WJqm3>#h3xxm*S14VXf zLBWndU2zNeuNQ(NmPaf5yOn5Sy#m4YvOPIDKzjfiv^E|0!W(Ano^-QB|yTO>bty8x?K6O0#C za{FbHK!H_T|8gsbvX$KEGHIZ0Icgd&g8IS1`)K|U=2+Ix3pF0Mro;6!O_8|Izs?2s z^f3>_Tt$-_u;8mN=Xg?gQ;u-=%{}kAn(wQ&P#)5Ccnt(~S`Kaw{qkY?+j?nmG@5i5z7T$iF*Jxuf ze7$D6^7`X@Fh48Dcj%=>SL>P=NUm!WF2_^8UaS$2l)ln;<>#z)$9k;a<^HL9<)Ld+ za;NmTU^@!RuuMFpsWa6hhr6Lo!om6n1Ypv`u&jnl68EeX5=?KF&@aj3RI6?}q-XzX zLFrQL^#MA6{u0cFB@7?Svin(UVcW{d10?wSDk#gW@bpkK*gYd-=tS1f!CP$;6$)X? z4mDbd>B?)O&;%)0&@It|d1}xxAEiEDi%+kNB=RCt z6^E%)@$0B!csBnP9gi{avtRDXhzkco1#?`8j4(Ub~g#kBiLCVo0=I2?2K<# zM7uI#k>h6+RvOZEqug=hFdGyZ{U8V2FLJ8K?C{(3Fl(M$6Q|nc2g;jZ154e<7R{@S zmb10&PWSRZ+6OlIYf+M-hEqv2?{f<2>qtltg-Vs$+;&{Xacz12_W%o<<6M%UkNfJg&K_A4qXOGvN z*v}Q`{%v!ZgDIiX;(rurv%$v|96wbmw00*Nx*J=^&xID|P7hX`(3reR=>Y#-Og;X9 z?@T#nbNHsq?-s5T!NvS)V!n23Kw<0xU5s)y(_AKH#o|t`iuoIEKsz}DlUbq?aqd`L~pStu~ll=Kl*46_s1vab)=U6R|yE-tVmVZtsh(p0L z-Kp~;Vg;}btILJ4A0qw9*L0yN)((GsXCjPFOn%9fV|5FGI^mj|}GKfGO1w*cI!#@h>Zf;ixid zd9;u6=CNd3upHu?>#n}ygjBiuNVT=gltO~(FU2YIOD^>6mbNk`MAFvuaX=_6MQtJG zs?Bf46y~nqz5xa-x?H(Ly%jV+j#zPqjBcK>m%p7;C5vOiPJZCp*W-^qu8*jwD+&5K zJ*h5AiP4AQ$_%U#E}~P}KNKgGC^bE%rI}uye#3foGQAsi#<+#UA0W=#gLJR|--D$0 zUk2%!r@Y>??z3-AKxCVvz)H25QQ?cG4!BCi^Jt1en$`WAEAQ3-986QrZ?%Fticso| zBtg+6wHmjk&!RGQ^h1>02x;!+A7+NEQbn#n6?Tu{)rY+;U{uvI zcQLB-^*TbTY+79KbXH{HC4{}M?T?%St2T{n#9Ds@irdrzAAJznS&&>emmV{ohMp9KZD429|?s&S-mNnBSWIdR|~sI_8z7 z1^l(r7xDEWM~W2oT+47gO~DVmClhZ~~^^?C?e$NjG9rUsidNjZjT z=-X>ZQlwg!dV-i(%%NH(y^*5!xr9u4u;WxZC`L(2p8Cr4%A7-0hLJ??W0;Trai>8L z(ggzBWxe0&snndO_0Aa4eQrn1LITbE6%ZK;HO5AbOTUE^(u%0Z5GpwlO{&%xX3|3| zK>CqtW?FUblv>2ov~;iPVoU^TG`{_T*cv9LppR)`{Cs*MLd_nub_^O~-S39!R>Z2A zCWx*qB)ovm9j=ItX_A+_9rJje1&%BlpRgw?I_;88IIZAS8g9_M^35dEH-eWQuP%7r z!G+ot^}G#ucv{f(g5fokqMHufc)etl`z#oxNKN>mL^oc#+38#AZ=k*}VJ5euTd4Ki zR!+R%8SVhQ+qC(pbG8EArQ%0GK<7V&%AFSaa(X*PV2=Iol?pp|r^&cq@8j0j2~X1s z1j7!t61-ZgY*L3TBF;X_<2-Ke%xkMOXM3yF3SkgMeK?lRdJ-Y*&prQU z!K2+>%*;W`HlPky3sbS4&;I-K;^#b|T7oG_vIsS2Gcoo$-0HxDOu>%TpA+Xy1L}HJ z3}1?V*JbNWs+(mgTf!o*jujI%oiHTz;75KHu9ZvGbJQGO*5 zGYd2A8$_X-ZoG0K*3ltI8p|!=oBil({msAI;l4v)NV4X7jVrC}pPVN?XVZ#GQb6cB z{QuHmJM;OU&Ry3>&lSUrxJure0Fmp+A?teXj1~@)m6}o6QjI^A zMPzZOiU{U-cq{l6^v@7N@DMtF(jM_p*?PWi91pmN$psve@tu_!jlCGYYTorH3j$!9lhckqW_D42ulP9Iu}a{iq%IPtaK zsjQJ&5c=>(KEsePDr!zay>OvSD(si(Q!)meelbY z{VMx3Ig~LW-CpPak9vc}E(;7F-qztkK>WL%z5N^JwgVO)<^R0ou>)LO3XpQJOH)Md zedBzqHi2)5;~@`>gk9+{?WJS7y{HVFQc7thP@l=aC_AW@8R?rUg${A7*0X0s0GEk> zU9K#}_>kZZ#U^$thFiUJj*-ZVPL-QBOU(L7Zr;G1+B!QbWs6KzF@M&sfe@;_SZ)^P zD0v1qbUF*alIKbTOP4~}d)@T z)#z5mUzbc{yw_lfR(D;WIsijCd2p@3Qr;2)Q@#2_d!pUd+%oH>pYo&;G+b5C2RY{! zlRfiWw0{p-+xw~nN;vPtnyHy>F@imw;W2qF-|g!ALZW8YyS_EC{=JhO7{~PGo0pRl zpWj-j>x|M*@<;?{GTz)iaWm@aFktY;ottFc%4R*q;B?}fIvpqP-_)}R5g%}iwnMkj zHcNsj_D-{p+f5R#FM7i}U>b0_tk)MkN;Wpqyw`56i9k1fW!SIkV;Q#E1||ETjyK2; zx77hmlld>d^k3_w7t}4#4KBy)HHUZhMRm4-C^DNeGKWd_<%Aj2eFRI7X|?unQE2ux zvW4b#*f|?N=O$w=!2*BcTD{EliyvRPqVNOBh;n7zS=g{@O_YKa3-HK$f_Lz1gvH)< zCn)((O{54TQ~j^B=WGl{cs#_-ExuI=h;C|uFIn-WMQQRTg;>*~akuQFRECqVNs2#o z&aT4hqC&KDOHV=|6byx3k=aGD$L+20E^^6?D!hivD}DXZ;>G|3W^avu_sz_D72y4f z0w>X4lxV-+lo8J(nD^o*f~alZ3dz(OVCX;H|g`V?%-1 zzlTeZomT*_cW2H^GUB+n?JXZzL~6Me)KqtO>iWMN(OJ#^7gGK^-Uec~f9IC*dkjsZ z{JLI<<=b|peryV#rz0Gaht7?Ltz*O~_ooj*VvdKygomIu!nXpuRA;)Po3cGO}01S??~4P2P1#P;pVP`=B7+4fuV` ziQrHpi3hX`N1Jx*$mEhReucS*+#dVFEs~zFn$AkahDEQ|xKACTYyl2!IU+*AaN4|o zJ7PvnW{NHLQ01uR`-R=U1j-IUq!@KT&5q%(k~c6ISUU7}PP z;JQa(OH%4JWa>x8)pySO=x38HR#NY9Q}?%O#U#$(s#`oH2FK!;YTnDXwA8TW5$Y~T zyC^OmmFj`stbxe0`ehQhc&SToUOZ*2p#AImxWBHgZ;YfZeZlQlKoRAVtER1C<6Pkw zKOFOh!V%W0g1^BDCxdJ>CHMm|G_|r_R(&z`Tl@_vunNQ|gj#B;oc!v+5=P&XcKxU`l0c^V9rk=?Bu;Ksji1%` z(na_CEGDZ2j!(Q1&a8L=#W90B!Rn#Pd4<!DK4fSz*KCh*2{@?k8{+IbQ@_1GV)%(+JJVP=odW9EiP&laK8zuB7_*t3}6<(pB zii=ppp-&aYrh63u2={WAOt0hNLrXAW)dw?{@mzV_66}hV?8wf*?orKVPkPy? zZsTrs!G!CoGtb!OWdXF+_eReb{M$Rk7Jc(4(_dTV>c|!!ru@>FL&l`YpOjK(&`qItD|Yh~XkYv?Z>3$ z0rCYEvtn)=Q$3QQ*8WL!uUmYsamA~DOCs_N2}lg~ce`Ogl+CY4a!1$-bvuI3X|SuTa{Gy3z9~ePOwC(9iSB z=mXtCgw_^SGk>s5wfxjD>-H@(AFu|@RE2k6dJas%Pq5tE{l%bvOJB{J_iN7)EtQ`3 zq-o^;)#|FSD4ic^t$ACh73AP8d<9eBi68kxvZM9N&8t~|n;d*Lr$cyo0XD9Z zH>)8m*H<80J9H&Tx@oRF(?<$%Fx6J&Zjak5?3vLGn|Jn6@sp>;%o5RwDyqH!|2==4 zd09IU_ckN#vr9=Vw%5sfa??OP>o3#Bb>(=INKZ2#h=q1Exc3II?r!HIG6oXu7D5(m zY=DuRp`6#I9vQ1u-rEN5N{zC37}DARmX45%(oi{~)`ES@VBWx7}JwnOH) zqjP`VOk9a?Ec=S^bSxJwA>xMkt!nf~2y z{Kx4J_OM#trbl{kTZLUc=OE|@lm{g!DBkZ3y9*#ckKMmFvr2(%_w3>Sf|SiGGBWm4 zJLiL2k2VP~5s)t?ei;!$IiM`7wbb@Ad)>q@!>Cq?@1!d3&$vZcnZ5h#yhOBzI)HrD z;{T_;HxGw;fB(l%i=!b;$4rZ|q|DZ0OZKHmnlYFmNr{jo$rfQOl|fY0%-AA~Wt41@ zB3q5n3{lB08b(vuiNRRDPj%j(>+^np&hMPxbw1y}zSrq5U9Zb{&HcKc_wsl=?)#Ns zYVg>>*8j{00(gcm$!Wrf_@ix)O#vq>C4^=OO*#V;*Bxd?$hn)9Hr)hKM`x63gRp)^ z@SP{?dDoHwFr|Kh^gg}P#rDJ$T*{6@@G7^~e@PbHV+GJ0%;F0pE(!tx);pj**_(LL z?>xZz^l)}?2VH6-sFXf%+vhHF_*1+%4GX4jlk`Uk!fyUsXN&fi`kh{I_UOF|1^aY| z1B;#?D#`lRN!rg$>BG@1BZ}(c#TnNa58a&ao7w0NqF;X1XO-b>d?$Bmeq)u@s`@i4 zeadU!pM8qU`hO?sSL$ELsL;Q7dpWT$6xJT9Q%e@Ql7KF|#C!$CZ4d+>O=|D+0nz1> znORjT!Ru>c^5qAdMM6OOI_8JRmPi_rLQk8D$4(zGGEE}P#Br~&R6!#+iPq6kr62TI zCv<~hf+n3RdIND16WX}6w^p?6+jeors^bM^Xqub)GaI8ey+KrzDco8&rD8{!5fi?P z5$o3CiJ_fVq<^R(rF9fPAj|-ay3vgSN{PVy9v%9@R;1U8M|{VlGXgL3$jh^H*&-*?nDId@WKKP_gIGG10U zr)`>gfJ{LTzRX5i8H4cdsk>r?q|fCL!A93P!cRi}^wLkv=i*;DL6m-{E5)hB%BfKeX?!krBAPQvt^<=4<9i;ecH*F)#L-S>JgaWx>kXY&DPxl=E>C42-t2d$Pj zPzt{PNgbs9y^ZyAGT3QdwEp9&(fEt$gkRe*u=khZ2;VYpO>xM~*Xg87*63Q|mo&v$ z9>hrl2`!4Tk<~CL!qz_|_;!=iMZ+pgHnU6HXje<613!M-@4~Gl`z(cWEmku%*>j<0FGav z#VR$eRYxKslI`Lp!3e7WWa_&&9mIHN-_r(hjTnI$y!%96W z_)0369N|i{iVDh%e+0k_GC#E$8~#G(wdRyF;QSISr5AcPNxx-Wd?}Zyg)C1d9FaW> zB&*EoW$Vc>ZYZl<9H;1R+#7s`ldMSp`i8X!hCJ|5UcttY%h1hO=pDevfmH@zm&kKveIuKl$;pyUvrvl99aJo9)P96d!jy(YKUME{Y4m zRV5PU&&qIopwoecwi@#6q!`l#zk1CK$*QA!XnC5T_m1@TL#qcP8J;6^jZc0+3xeOF zh07=Yq8sf2x+0tc{=)NfUg2whq!BXqC|JYMB z>xGAnWM8Ye^!W0}vl6)Df&0f7&vxp(S~H1%{@jOI_UQnoQL=09Q^%z{9`gCggkREZ zo?UwU1)RYA1drxrYkiD|?K_@3RZWhjD<40;q!zhzNCMU-&=6+UM ztIl|^PsBCOVS`0S0%bam?Be=}Mvv2w2f&%g0~3`(&Cum(%vWsz$mVuyhLXCdU(dL5{jbio<5`jXUfxFQohD&Xe}= zLYpOKAdFrj(&i~|VhHp)rF!aHDpRMrz2?cyRA(Ui2At@b)ZZQS=AUz-p&5jkoU`v# zw(5TkyDkKMIQ1!HkM{`7@Ma5wXX-be5~<$$m}&w#?PJ>e=m)3WSry`&ctZvdr}PFC zyui-{hlQ7@M?`Fv{y%O%6NZPAHrz zzQ|!@c{g40xYQNP4;;Sv1qTP2K}nH|!v`wc2f?#CQO-9-D=5j{nmSntZJI;D7*f5v zHA0Hk3to)Cbe$6v4a*AVySR^7j?_meI3*$6=w0ICSzjF){S%_Y!jh=v#+0J<0K7Q1 zAPnU>yz6CO*42$Ktxni!WPf|=P|iT76@#>=zV4O$Oq=DXXPb2D$DS*#8VTW36KULG z7X6dXWQy6*jy0c6rdJ%gD>i89;IrBldf+eW1*tXhc-y$fB$XEZ$V?NoEA`&Y>43R6 zWN~J#I1-}tTka2Ppfik8kFqmG2RH8NHxYHhXVQBWfi;B<1-o^o&hK?E~3m8*xXv-(i=tTRPUD zH{DMl_+`DV3x1&)ufo9lM_?4qmTH9 zRz*#jcG4am8-_J_NV!uz3W$Woh)iJPAxO&xjZ}P2Mqn$xeM~k?C$`e3b@e8vM_LFO z6ZDa$JzkXKXvaKd!0zWgj2_%rB)_!2WgX_ECGX5Okv^y?I!Q(ZM}|L4Y@A_o2e#U4$eA`uQ-;Hwtgm`{LnL(Z;3RVDD!F{M<&`~(L_rC7WQeY z(JTBSGh>sL%7dXwkJ1Q7G$tz{B;nKpQ#0){a?RL;5B0CAh)wmHvx~_w&BzsJY!1+M zMc*^kyTXrLyfY?KdL-9vChl9uJ{p$d%h?#5*4^I+M}pb)|573UkubUbl_5;}aJHnS zThTta5_;YEi;$$NMWbstNk{y-HY`hu5fU%<=m#`Zr;;o{jTz4dEK`mZ7}1BRkilYC zxd;`E6IFe=!U(BB+?lBM`1bf!?Y@IHY8-88cIZ-f5eRAe>xOPdnl2;;73760x9k%a zp|w6XGoTL(VfSn#-snU%$?_$_)EO6u$yrg-{33v-`m3AtCKwy&+lm0Btd2i_-{Pgi*d8 z&L6lw?LEc44S~H*m{%e^mabL&CW8=~wyeD-nmWd4QRELc zLSXjpNfq*mRtZ1OcbLKaj|TFI-Je&2QZW2}|C5s+$q+PrTh5!f$=Z5x_A%{Gc>A+` zh7m0&fCm;5UeUP(DXq?9^jpeOLb8mKpk17y;|29|>F}4rpT%HE&_64kwJ3#-#u5q# zWS9Jfw3yTW=kyPK-O0MF-|6`Tz}c%`@?oPUQFrk@S~UI6MDex;9`##CHht|&Tjr|= z24zp-t_d4nE1kyB3%`$ZYgl@yR>1@PRkuj-l`)l$AFux~=t(;h+hiLpjm)5dZxA6= z{sB@?#dX-&wfE}>C2%SWN## z_n%yyi{76nf6XKE3EHamOgQ4|^3j~zag)D5OY|^w+RM~#J{f~7T?qCHW=OWEQLT{hsFnWSRzN2&=4md)RbxALC2pvvmG)udIFiH-8Ip~< zI%AS|Q&F@gbSkl93g(y zw+av%cYKsDJEpJX+t+k=TqYNVs)z`|=QuuFNjy&tej*!BWd6c2cWDnVr*T6e(D7wg zbGIjXJEj*hH)YP8MrU00nB%5EpfMI(e?R4i^OF$GxCENMXL^86lx&5x#bxQc1o1BZ zTvvFp+%qgXim1*#wEQh_BcrZD&0pW^HW&PKu?3WBZJYV;;wf7YvI7aJmpnxD#`m9 zo6HY7ET4(`QUOUHmPy5@P!(5Enja2?>v#HyN5%D-qHTeS8(uU!9CXmDmHhRLpe|D- zUu^KoQjJgVWs4OQ)=l%3%$KLLKwefp_~%%MlJ_WxJ-2bfQOBfDx8RJgAkbd!jMA6&1? zv%fg!=bG*kPWst1xrzQThTok<1!@ek`RMrxbLEcgV#F2G-UNbf+6VKWb`tcQ%I_VC{(qo6fV`hu%JAr(BGtxrI z>5C~OEIX*bOVM!D9?~8wDAhd2d@eXXInVtJsZAr=ApAY`H6nGcyLh&e&h87*@jva= z$8}(7w(Xq7RYwjkT zTHv_}%`Hcxw~KTGSm!V?eO@0F9Y-LM4?XZ2+@8Esev+#|N*jin_9h3SI6;6uxGvQP ztH6|9@mj|Vb@4VcorNCPMPY9LLG2kp8Ps z&J`r>EgCxvgGO4M?LS14W~A~xG*%K&(fStCB}IN~t5Ga51d{ZdI0w(cQsokdRAKK%*tHA3oo~oAFm5oG*7^ z-W$|Su}Vkay#^|kFgG2znps4VW6q}P&+z&gdyiKh5#{&Bz!-_AF&d6cm6Cz${P$6T zP!rwz$9VSl?BtrwiAmQTeokm=|`Mov604Y1j-j|F^y^9?MNy zzCRB5R?bf*^wNLGM8F)*e;4{*UQhg1c46`=*+m*J`$)CJ1wec`C!FUSkrQb;2w~|b z416J5$sI6e4+lceJuEFZ3geq5L@pGZjADpae(zSC@~ zo21zJARra~2Hge*0mthFHby}BQ3S8%-(&exgFcGluUEJ+<(}yoFUdcQj#p%j4nX+h zB~+j2#Gt#Ci+fr7U`UDM1xjM=gHkLh@fjM2S1tf#%G&|Is8}jZxiFo`7$4tK9Vu2SMX@R@wa|H$t{3?Y$Mowns!lE zxOeNw5|9aqP(PuGw^d%cNJz)=7720m@8ffsn!+s$+Vo_vkL33sjtXV|fDgqh_R$Mj z@CqNKwwnlPHKN4|iJyBI?Y%8Z3Qy>7PTphx{sX#n`9yq-neM&Dq*q5gzm`J=5y*Q= z7#HA?MRv%F-cSlK=BKG^SeH~>X#~&2bJy@=xSuH&&HSG0rfW8PE~TLOqI>q+*C|sU zxaU}Nl!5EY;=CAr89urtFM(zSP#fLk9u3ZO&ps(sV6~@%S2H(=RRM9@*CQiijP~}_ zO?Xh&z?3D)lGbMtM&B92v`_uCW#&)j?=MZ+Dsl+_&3`xbbhHXnW@IU?x~mgD^mvH03coc&9lz@P7A$zU zfxGd%8cc0!=@A0+kk)LmzTO9f9EY3IBK3-5ioi0ERf?W9-2sGy`hSty7#_lu0pjnw}M7x3?CeKtX}>0 z3W&+L{jV}p>n|dv*r%7yI9WSTV)|^0+7OaLP@Wy&?quaZ`HH0p_fi)W=f+lU|KfAP zuk6d(51PnGw*FLfsQy}nY?LhYD0+A}Ag6n;bffdh1_^+nsuEW_M2TC3E$KdNG7v>% zjRD3-I)(7Of2lj$R~NnKVh+3&D2Van&sg!|B2<)IInWufLWb8gn^q+&aHb>FG z*in{x$cbz;bIfTU;lkkbmD9c&?p_0AcVCP_Qxc(0uyAgYK|avve-gxk+>fF*!ZV3` zJ)y;^kDqyc)QC5#<(AYV?2Jo|c$u51asi&Q$B$qb*B+TiJjOu~9=_@p7Nkd7%gD3)2-@-V>KyKLPZv4S&wNneIwZ ztd+|PG0AyT64np5U?-3#uXDQ~(3|et5sD(Je6!yB3utR6O^MDU^kE7sh}w>znS9pa zHqZ6Hyb=thW&;}ZRicTcfkJ+JB%~IKgiO3xCZ0z6ybP!kDi+|vQfe*eh z+V=i|&Mi1Je697Gq(#_s(tBfjc5_Zn9tbGlbyBwjs5{}@Yg#>^u3Z_YWtK9+ zm`WkRc4+)W=Zlih0gJX?$0}q-wW6@OW&V5&$A+*wV+VF({=r~XM!%0jlAE5 zTJE?4l}=R>;YE5Q!L4u~{DE#P;d0vdqBkr>=cDL@N+SaYSUF+0;=uGo*O~!*G@Z5J zj-Bq>mt5eENFD$=%}EgOI+a-yB+hs`p02)&he5YzuiOWXQGKX;1Hdr#H>0hNye+ljx@v2u`3< zWVSGsC7(jTReBXg3OuCnnhwkT^MSh(`Bj~Ix+&V2G8bl-osm2JRw1+)<$Vi^Lu1?= zzP0Xz`=YCqWrqtWPNePh!|X+NmG7}(zRmWg0|Hhv<=cU5`=H0wdMX^f)xJcpd;+69 zh}p~LwT4kL927Xc_Ve9ZK)SwSZ1eX5YQ0>$SySBb8YlHpgf1T~S&Wk2WJldOi4t!T zvQStFul)Kcagmzt6o-0S&Bfj`oz5JU6^tuMWfJPM2w0Xkf3ElVs=_sggoUXgX@)rw zPn=N~6%oS-h~ZA@4~!e*r&tVvI_Hp|bk4cwuNy#Z-{R=F8j`5?+s%s^0Cwhxv)rid zNcirY_LOi8f|lo3x|9MLyjd)VpX&Vn>S4I|Q6g^vQk(e*m3O%2FkKkQ!$(J9Ev&Gm zrtZc|st=o9s@Oa$K;r!e63}8^xsp>)ar^NxKua;snfGjj>jAxsX+mJ}L-C`)%!)$e z#gMRjT1fcCQZv=aslaHm;Es6XBdTRImGT?EDMOrnrqZGqeNu@S<+V%bH{>g&usc)D zo=S;srwlc5h`dkJiRAa<7Eg2d4NHga@@yf3p64y!!H61hN78-Y*Htey?U}OmFAl4Z z+UNH%X2hyC8{RiY;UxyD9GjD+jW*3V^g;*si$ub6qN7ThOPs3%4(pc{1&n~&jVO}I zoh>8>joJW>h}#GPetk|d?ciZ03MsUNU6#=}{Uq~BKp&hBb-n)3tr<5Pm=C^s;BoK8 zQg@mAUKSkRy}Kmt%iAlx>d^dK*fKGV566_ACW97Vg=WEbp7Kne|Ij1 z@lnZV>?8!?(Ds%wD%PuI_*xXHW{QRv87-;1)W6`PX$>d#PcdSxSI#5*^}o(z&0n0y ztx+cUYP!u-G+ErBq|9_Gh7}KZ3hh@R_sPOJY7;Ls4IvEM2@h51D{nj~6Ew|9iy=pY z6BbmgRfNuN4^t+jWZW=_G{?(DP~!JSD>E1>*O7>i)Jzu2k(ASlKsW-TGUY8 zD=i$P4_g(dtwX=)am)zxoun|qw^~U1SP=^;n$)wMwg;_8xXKAWp?W)NU_O8uWRQ3( z-NF6c?Mc=?fEu`)^@3G2Q5XyO5&5`_&kqaR#`kZt_=^6`JUG)s*?pwwS;g)$TYdlY zA4%Ro{g(w3LBWj~%PB!)-4Nc5shYKNGbcqTNiQlrZUcE*!ON2mbl03~e%FT>c7kIu z)fc_yVXik8@9o^^YTwz*)=1|ryd*2PONp~7N!kodmAqZ*YU$c%@hBOS4#9~rxKm<< z-?6~WlZiO{q{uKC<2aMwC|CO>zM^pED4u8aXEQknwBR+rWV-> zW{cI%F)=^FGuw0aC)OzO=aA0*y79M~gJQTn+P9WNGrMfEIrQ_=@a!A`|6>@Tf%jHN#lh1?A7dYhV`= za0aqb%pdnG6iBvyw0;VdTVPX^qWn%j8qG8N#j+tLw^PxUgLNHjP@UuS{bcyR^Dn zVEw%+yE;Op0_mh|8y^{AGxcrJw5BBE{84|J;6_39ivBR!F@Mca_;_;*>DpAYe}5ox zLnTrFYn*b!_tN7U3S#wjKAG#-WZaA#Qg`BHOMA)|h7<8t*3rW+FHlP1h`8Tb)h=x9Y1W z;r&_V4y4k%u`EdH0r=8!4VjQc!t3CR`g6C%B?T(5lCHV);M;3IX-tWT^VKb#!Ml{3 zOw!0Y*Bw57pY3p;!|UbvYTLA&LXBY=hEs-|s?UO7k&n!GJI18MvAYEKf|mES*tw>b zHdu07nl5h3{QMj8J1*yq+UVL z_9etwxD~dHkE=kjQyVo5rJ^8s>3+@?>F3$s54v27=YTY+L{-RAV#eTg8X{oM?&ozH_=G$-e-y^x3 zAW(tDWd{cd-tG6IPWPq|@EWw@_$SR}W2Go>Rsr!>S6)oSK;kyT-y$g8KriBo-#D+Y z|IQ&`y(G%w?Cew$D6K(rKG_5qPlikB5$!L_bnpE0N59EpWOg-hOB5 z5PGdgpLv%HFBIcA65JoC$hj$)lwfVAHY1A>3T%8c*4G;Fb+7&0n0n1|htXRJzn;Cf zEz1PA9ltqa?`wa~qL&`wg{$jLuubK1K3PC!?#vPvk`#J;Lx$r3`p^=@A>(e8F!1>a3l;f)Vji z6NV9S;ZxP4-aQ;mAR1ohQ$Y6aT#-Nvwh81(Yk1_d)X5nmJG$>=h3M!(*AQ@niQ6G6n({>qROYJs*T;K$nF##E<}&(hDGCrioA z(o4UJ=6@abcg$0qcB85zn}l6MzovmjSFXo7$rE#pnaWF&Eo^ibaTJL8A`~}K_>Doz zIx;p^OQ*GD$bH>yFuTxZ*DB{vL&{mVrID$$<-$G4~Cqmpj*vqNHJ5pm}9B6 z@QKp+j`%V)^mR`gJk`;TLlUo2OI&Omw6$LxU+N5u)#cuz7#`}K^a)~*Ger+u{9V2CDz zHW=Tb2P69JxVN*;1jdHrU(z=+Rklli;k^vDy=ezBA%|B65u#QfMaPXFcqG0GRyA9x z-P24%uw{#@vz?Mae@q*nMmJ5^`kx4lb-^SThnS!@DkVlv)vetUB~N>Sen1o{zof4e z8F?DM9es=N%O{&jSM9e=HRB7m>?5@CgzvR-GzCP3J92~P21YGhA3G)lL0mK81dfmE z@IsY0)y7+j>>dduh|FS{)5~Y{Ik!o7-;37nM1CSF4lL0x>aV)mo^IoFpKO=SeGzmB zGvN>6g|3z`lw_=%c%e(>GQxzTPSf;#b?CSFZ%Ulg#qB3I@?Tvp+Fl(Yz7~;Kfb4NB zz^B|Hgr8X?tfwos0nmS)(^{NC4~q7-G}|x4-4rPLo>#OD2PrA^q|(l-*YnI0+PvXF zB|KlU6E+k+DKr*8R5Q=h{gkB9WPfT4T)nifjF0hY>=j)o?k zjm3_WN<;ili=1-C2jrea51c^rA73i7!?4De%AB*?ZzyjWB7=o!HzpPMb2Z}ZBNgs5 z!Z^JZm*+c_Z%O#}{VI9UKR|(S&e7}vPQ93(b~rIAeRDxEi*?o_c}f)V3~!s%80YZF zW1#}tww@O$xZ%y#9Umr7eGsu6-kzK2yG(#PgwP8+M`hYJ@NViEOTk773vnl#lgmZkVI^b5S`IHqLctsXCk z3bcO?5}=;f2{&)Nc+EAI@FZiw;PKr_k*vge6yFTp2 zrmx{STDwGv|AH`-?(cJ=-)hfFHo)8%7_2P1gpRWuioU*9Xft~|T;n8$EV$P0aN6@y zsS!F#@nw{NYYiEMR;i9rUG)SdyfDX1kM9)nPEmn@d)bAYP3#l#d4%JYx)x=#K5ar{ z*D%I-N(Q0c$Ie&5!zA&T;lw~aYxfYPoy*q)a!Uiw4bt;^K#4`k4+A2ZI-q_||DA{L zH=-zMXBKj)@f0lTK<;m&!Z?K=!FiofgApnh8}(>7G1)50r?owwIplc@ufttRfBkyx zBe(vFVx-1WjGRr4<;*je6KV?@qaia`MmY$@Ue@IyvpfL;fAk{XOv`S;TGwcHeg8?z zg8gFc9xcFzIsz`)v46^|b>|LFfQ8KhPTwWoVO#2{z$4zN9vjrRkQd1z^#z`Go=qoU zV8Ah-rE92h;pt&ZuxsTq{Z>%i*s9X{xm%xtZ3nyXHDk_VL^v0`;*-tC6sMN;>>1AL zfzp^Ak2=eFLHQG7n1hoNbU1$D47PaNS;01lm2NsI{D8=X`xV)OkvCimj)e_9nEA zhV^@UpD8{V*Nvjd1OM|SqHD@v$X!IrKuMd9bEe6;ej6~k!b=hUs9{Ho_PcSPrIu_Q zeEXpSZG5&8kEN3h?a#z|t>PuF7O}8>%8`UvxS^fEq!Ab}R8p{U?}Z9Ia7Xs+s<2}m zd6NQ-LiVf~0rGP~ljDWPoG^(fXS{3|a5mSsA{5$&gr`PCt`mfRaoQPu5Fj=Q->1pb zHH$CKRzH2Jj_OgP6_%PMJowPouIVI++%e&d$tS9CdJ6Rxpe!O0?I`74QLzPQ)2g&F zx-3tW@x-&*JFjdMRHyvvFSaMXZ`(J#W%v}nzTBEQuWs<(6;uKBuP4mqGduK&;d(WK z{H68w1%uC(uf;?r&ou%Sb3ZFXQ!- z3)gbOT>F|}PP<3NMmLR0jLI?fFTxrTN;7v-Q1f61O-(K=yudvOmycd3=n)3uUVCS0QpQ>};3 zcOoq&(GR@cCxee$^IAM~)wc}qNfypOttH&jobWWbDsH^&70+jSdY$V{Jp&H^GlYb) zkV^%C{S#+^C|*YAG(Su^y3MZSDQTxmU~AWCWafmWYdMNiXO%bdUl%U6KL3vkl$kGq ztUMTFJMrx&&cub{d}hrOzL>)yNmDM-V#C)Cx9;4jUGl7$!juLj>2?)L8^^Kt3`0!m7G=Fcn8j3~ATz+l|uCsl%q21);M zR64xWF%joZabd7|^sOBs$U)pV{@Ru4{**e4=95=m^B=aCEU| zic1$MA?2n|voNRmnYVdm)5VVFg!d(4)%OegbQSHRWV2p=+xj`{evGg5x#pMOwV&|y zDp5HFtg0pzHO2>`Do?Os28>XxM1zR@`Q&eZ_}KpKc^=Oj`mg8tj10kF-mxBFA^AIY}}slp@10a@DCS&L*P?iW*t-Q)GwrV8^DGc3^d$ zzO})_VSp%;_kRp?k*^Sr&d{tn9TxEkV^3IbuixBFDF-l8*XGfO#b8lLwj6+26b&>y zT3PnpA_dJKWEp`(-!nTm@x`IbGOa+3#6H%uqH{fC;29pN6=R%0tC5pr7+8h`x6t3d z&|$7cGi->+jJKCTgYg9Gj6>&|oA+hCTIFZ}LGYPKOJkCq)Idt}rB-NO73|lxu>|I~DHG31RJmi7@eJ z9g9Bcj!7{!FSQ^N@6=W;Kj4b9R6RqI6SUR+jVE~!fk3nNqZ?Te?r<#W*ZR3M#fFmZ z&Fls+&~K;z@RkEw=*{V%(V{Cw!JP#dYH_%G?6|-Fsyo-Gv&5_79b)pTew_m+x~`X0 zdzg|L^guTw;7?B7&&T)Ja_ITLA4hE0e)dNE?sC(WjguwS_lCc%3YNxYzdj=I^)m$e z#iHYa-;HF95g=gn%=|i1#p?tVJ=3Q^neyEuh0K!UA5BRO+)JM&NcH72L%-p5Ioou1 z+OK|J3(KESy)VxRw9(+4EJ>F#%sIUY=4{{}A-vp|3id9yf87RwJycZ+bN#5_Nut!t z!R1Rs)PG${9B_F5E`2oLY8xZ1YlXO@jJSYsKD`WQAOR|lSsGOcJbhOv#~ z@hA55G<3fJTWg)Z?I^$!V~1)c9?R#Hc+v#)kD`I6+jbPx>ef^qgz)THK2&c+(i2m9 z?j|a_-)4`CE7gbHMNeqK9g|cW%cxX)-()d8{b-!jt?5D%aH16u|@r^a~cc9osyueYO;826z~dP4bTa)FmDZe*P>((ne9 zZHJosT>#~(H1gJmDMWh)Nc-sxB}a6Uu7!LK>Fg^Lp6&gT0-?S>cavq4VT=l#@r>?j z=k}BgaAvzQUpdPqYhfuf_VaUBXn!Z+Za7T3nkt-(S$jX35bbBdloOZ!%%fkBzG2Hh z-D0SFmc=^e2k;=W`u+4XS>% z;8fjLOdQF8)Ka&gXoS%eMcJE^tJ+<(r$NB)xi4 z=38RWh(AYnK-WG`u`}lVM!Wdw#^;PDhbOMJ_kPx42AuPf;t#mNkWU^*_uUPYJvlLMfsZqa3`@QwXE&P$DYvM1NP}VVBgTec+U;dUeW=gPVJ_Q6Ixra zvwVEaKycbR{zL9|$XLvE$+2hi^Oow_dD?Y4wc9whvsAM32L2d8Nm(-*vAIkxN_UHT zfsG!+`|(rj0sO&*K>fQnNR?L>9v{;-V&x~5P26H05&Wtq&^F*Q@)=U)YhX= zhGo(Wkq_J~N>mf|10It}p${_w=zrXr4}#Z~N$Iwa7PGyKkD|k=xe~oveFcRf6LW02 zr#VlW=Is|Jb;L;WAt^o$d~~uw)n@+Q>Y00FQq-7~9JVXsn|I;A=afIH{Z5tS|3^`n zx#nf&Cg~H|k}tWpi{-S{gUZPfC3f)uH;0v?59H1lYC0)1e8!zIeu}Z7lZ_@Y=_AhH zL=2-YP&O~C4JzW)wF6Bwz?NkH8%tr3taVHA{j~!y5j-u>4_ZE%1mWEt;8rR^ZVE3Y z((2YMOYW~-%l6RD?6n9?+j017xbkLb=G0qluo*PyW_oQbRTN2k?e{i?Z+E}>^m+a} z30O3b2?@jGi^1FgEkfX~B1NfbSJ+kZ{OQ9l%F+~Hw5KH;J``T?CKJ)5#S7iZ?i7zY zOGA{ICVkKXdtX|=KYub+>mvx|9*6M7Y9;dbkKJWC1>XNQ)EvHAh$@7kC_RoTIrQ?H zi>{=CkV3v#LD>k2T=2%97tB3Lc>2jUjW%(IJI5-g9DmN0G+%$Jq>#H!nD~(_ImEU85c9h8&R=@92tLJUf0i6rzsmJSOzSdtA#y7%- zQ8>H#YXraYHxwAH8bUyL_nC^kFb8<;*638_M%|g4hUzksU_Y`TqLR$I3Ew%V?dLl= zLnht#u4<|@P><-*vw_VVk20XOmZ~=6XOsKyO+M!xMXP!5?J;N>db-C(rpZVU_<3Gw z)t#cxR2aK&I3lmqn112Dx%D0ybKF1*C(t8!ZvDVi?Df$DU?nVkW4H4g2EWa6{2qLR zf6&bSHc=-%WUemCerzFtx!CM{WIEWnrm_WW80|m1%q5A z2xI;#gppR|9SUaepk=Ltsm}r#5`k>b7KN5hrD++8;(mdj%cqU+H~=L#*VIgr`wXnV z1wg1r(MV$Y^M;4ZYRe5Zz1qod)(p~&{62-GqFo4kGZej^k@cmagHu;ETd-gU-u*~# z6`d>lMkFVz;ja@e*S#vXLTT@9UNed}79aW-vJ{X#<5zWrYM^rX zlNoE3NV;8mQ)6>J!7{uUtOi_iZW|nv@VvJDToJ70HG1pCQ$UC&IP;XCdUH~woE4?~E`7RA?-?n~e7#V)Fbb}pc(C?)$6%21n2wj5JgxBfShBVz9nPNS1{FOc zd_I$7c4|#Ae>LwK=^FOZumbJ9Z7ktKnzoUBRR3?J^z%hf5x^`X3?D}$I(M^}LMu%9@WBW=KKoK_RwOnlUq`9Y)-xMIbu#LaLPzAi0C5)yp zXB!H67?0aLf*8I6_&%F7ZvA3Sz(CK0j)+~vf&xV7U{!WMUutrWCMV_I%Fp?as_YNS zwT>L0|ES~Sq}16&p0WIWn*%`i&y|U?r&;>>38MGYduo;^KdDgf`t3;4Yj{Dqpmws$y_`Apm!*_e>*g3 zzyJA}3X_&5Szf~R#@jHF3vkSA0KPc=#LG3qkCN+y`UA!bmkp^6DE!0cM*#G&^RSjo zYvtitezGv~l;@KAmY#?}F_E=BR)KE$di){z{AIAFm^N1GUfP`42vTNXAv?g8FT$7L zvj!=urz3Z^SG5bN|JplH2Ftf5#n_njqDZ3ciTuvqc9j+IgS`XF3s_~sTsb@X_cV(t{1cLMuciy9z0;e)hm!UP#O-1#W8d!@I@dSIwJc1T-Fv38SjJX|vp8P0 zd1U*(;mVP~ap&2J&mhVa+JUXra}CnLuK{kZ-FM2vuCQQSx^%g$Zc5kNT7hgg(3{vR znrv28eP%IlQlp{3V%%|zv&5-*6I{(qsZYUo#`ZmXa7+RFUe&Vi(agVU@jrO#KdbjY zDAMIz)Vp`nT(m0zz_=%ZFiDlezIQR6uDtN?#rz+Y|4%QR{;%@?=byw~utsJ)IF@aGB2t_RAh@e}3ADwEb`0#)cvxjM z@^^3o?0b*Bw1+EX4<+;NvA=KN&Fl{=ggo~B(VQ{MzZL5#Dw3j37giLWaxQXz_n2e1 zMTaRLSV2xGDwJ=e)_+Gu{(r(>a$@!RiQ9nuSq{4m`;VId+z>q1f869hQPO|hBA>k z#ne<}YvsWurMYvHQQYT|KZ5(n$?2fKn;}|6ocMS&ZBhAQ)jL@|tnQtK_*eR!6Z`kk zbWn2*na`oG^$>kH+e90MpON?Uqbjq%8@|!B)3Rlx$$Q*-tl;zKKI(Fc;{D-NsyhlO zsT0SY_tK;ks_!SO&1G=Ef-$hKmE$t3<=6$+$v%n#BIF+7VDXRGf7-S0k@SIoX9oTa zV*W`P{Li)#q_VkG3XGzKY&OcRW-fZ$&9F~PTD;kZ{mmkr5! zX@=C(I?yAkhhK^zr%N~$@eyVw`S*UNj(khKw7+YGCYlYh%xUr!_cUvZ>6>h7bIN>k z>d~cVTnKArT`_Q^EnriFnS5%G8j#~vr*0b!YnDqjM&BbMNj{T>Ts_o}r18YH6?*>f zrZ2F6X?wXpka&o{%NEBO$Y6DYzv+c2`a4a2;T~s&!9<RINMZ$+%6dt zfMP!$uVHCX;*r^MC~%t~as4mu0sJ;V0Iq!#e)1o+!XK1Ne^4Z^{5C~yl)k{6RbQ&*}ds`2X*N|I$$S-$6Qmq>dl) z|A!m?^WT3g@Q(%lvB3XR7RcN+f4gsX`4^{O)}c9P?n(Rt{$UMG59jNjx&D6ugI1)u literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/observation-prompt-v-flops.png b/blogs/deepspeed-fastgen/assets/images/observation-prompt-v-flops.png new file mode 100644 index 0000000000000000000000000000000000000000..6d45880588d912798740a81f189b16627c5bbcfc GIT binary patch literal 48387 zcmeFZcT|&G_bwVxEQq*OKvc?R3j#rkN|CCFigcuf7OIGp0MY`4CT?L{0Hq2ViVy+> z2)zeTHXuYmx)7T5geE0)tZ*=u3q``v8Sh-hn%=L>c1Zlb9J{D@60)?0Dk1S+kF!c80?fa^oJqy zCf)%Ct9Dbnqio>wVQ$FRA8Uu*+glGMeHY@@y7}pSi@l`HuomdoXc#)^aaf z$A9eXvn{3R4;~SLORp?heA-JtZl?M?{Z?D$?_yN=nV z^$;TcP9|vHD;sow^=oJP=KA(lDT-2zqDYIHm+M+?d_Lzn9xlT}|E(56f{Z!!#g0Aj zfT`=r59F|M27`Q3_HA5-2m}F#yBn#v_Ac@L`<}tU9G#%8F29-7Fatxw9@mH(mwy zn_^4yh5c=%__I(-=iS9^ZdCl*nx${e5;dZx+LMXCD^v6;OCdJT+1xV6AD^E`rRxMP zM_1B&&9=iUrV^csA2y%CdJi~&pZ;V`{1J<==)fz{h1ELe`is2QhtT9%dfWohtW3Ml z0+TBzCue{}=GrF*7^UjY-Clh}0SMi`0q@`}w{M~w#ZGxT0`t>z;a^0SkVo+|CV-mah@RqV55v#wXf~L zKmIxM*><#);-asV9M6yR?zeNlvG&V$;Y7ao;&3i8VBTr_IU=gqf75+;{N}H;z|}O~ zc{GtkB6)I$-#b+q5r7W#^7Q0ti`-uzO!w-s{Lt?qOea?VJ*#s+it%Z%OZ5^s1p_!v z)T}f{iT7q;wM*@;H)&^?Y+Cs6Md04{$Slyscx%q6boqsH6`%_DYd5baXi~R^@TY3t zH;teF>2>hi^~VMNR+&k#PwGI|tYAB*SEkz$NFQRx?5%9Vkc z&C5L6TBZ6;W6{MP)7ngoPXmAS)Z|st;EKDeU49D{(~#EQ+@54 zc_sgCnRLg^sys&?wV{+QZ&nxf&z&am>_3`CeAC~e6cvv-zx#*P$)`Z&b-|9s=iMu5 z*@V?_SAW~Of>7&KO03tPU;#ZI?TUhGe!$nLGZ_8(hfsL)qoQZ5`K(4R8*@eNJU@p& z$a=)IKB8}oN0JxUdEJNV?!VQ>mpmE2r}a+asXcERbGy<#k)(jvMhe@Uxjt*B7M984 z;w3WKY=@0isj4@(-Y%4nM|4gjECWhvwucsZ=(RIvwQr<)lEA||y$lTvsg-GIDnHyd zf<5La<%%4yxO@C|0)uLQvT%y}tX(x-nzJn)>u-A8P!!qm(mu#4mcv2~n1@ZA9hEZj z%e(d`E%c(xxAmHJQXrm{M>yVT)cMKhzmMZ)=P6~po)O&g-Z6O>FJ6rF(>lmzC62KJ0|TC2S~9{H!g<`Bot&Jgej`U{ z(U-O0NEsO!r`{ZsqaV^LDopgQ%3`nCwJ7&NX$lOcoAv#v*J0osEt!eH3xGF4Uw*2? zrTz9koeE5`+@&GH<|b+%cYN%mV}@6fI_VC0?Bz0wsG8q=NgwNmUzg;*dxc9!j+Cwb zvHz1+H~8x0G$B71xJ?;Rhle6Y_I|BPxeiw=lf2jF`f;;NCpfs=%t5%JVm5XI1Aqs= zGDOJEhPSQ^dbDTZaCUXFNj_u8OUlYLqQpq6oyxff?l0xcX)Er~p(^w&aHkg3706_9 zT?|LZKfV&v3kV?bQzzpsdllCQi$>n%<%tdXue9Qh%q8zk2(rJpvlYI%UAu=-Yp@lq zUB!uK;qmy9-4dt1xZd+DN4#34`dIwJ_ty`L4dpQo`!S4J6n%a-JUslu^+&Ihe5Vpr zYFrv$pLcY2j%lCn0%pk##kIE%iIwxElvSWXvtzHsjc#xmd3Y4@`2SR%2gVd7>C}4# z`dGhfjYpfT>N!jQf+RnhFll#YPn`(WJ4-Gw%Yg};ql)}&8sOEHjDgg-nLGrC* z`)kqFJ`9%9p~@?YEGsUa@krhtOtDzsT@141-Tb1;uLg`T35tL-7e_WiK{T}Ron$8g zv)d+3eLf-Hy|*2-r^6FzUnlEYH7_j?tp@p`fl4pJua%p7MI!rS8wq&*QKURA)Oi#CxdtJ6M#u%31XOJD7na zt=y)yxhfY;D~oX@W(glEMvUvL5+<0hGs@EUV8Nizm0{+&cR=YE4#Gwd+ zSn(s4n9gUWkz@yl*I!Dc^FRDayMy1HC((hJQWi!RF&OF|faxs==_fUkek z!oOTDvw$;0`Mlsulv6TN}9SXLFUf?(P>L(jcu*{_av z8=SrJT!XJxRT~d7*jN$cmT9fY6LHI=PllHfy4SD>G@STQ4sVc=;@N8ej~=<`Lk~>9 z)=-dPN3a^8sZVvse&yZ?^Yzl9c8+_4wTgVN>LQ45WHYm&(S}T9^SS#E%YOv$Pnl%ewi9?!cwDI zv_aSGH9%j+8-TTEq!>tB4)nw}zf+%Prm0PL_e2|gl8kBLcu3I`a}|zn&rXwh!NXVevKOhR4Ypg()y^4!O&%A17_)o!r7OaB?#2XW(;bY4U`g z_v?@6;}lItITADDV*PpW7tSd6G=*e`ZyNHR$`rd0(>JS^Df4Bi61(j8 zVYwWuS}f8`_54ER+VLTo*Gmqslb zVGxh%miTJTPQFN?ePrlqJmmP#(b&wj=IQ>6HLLSEVEJ|wwTYRNo`Z55ZDMNbI0g8s zkCfMK1B>L^kp;0LHXjq5d!v7<@xWu#(wyNxa|E%7Yq9!;=rF|9)oZcM^P&!OCrD8V zA@08VaVU}g?ajQ1PwOko7$k1F@z&zam6Q=&%1B!NNbdb|PXD_q7KYEOf!IiV@oC*% zgzgdYkCd;cIqw|pYO=MOjNa?2|IA+ha|^MV1 zwm(=i^U=ZO?X`uRT09S#-v!RJBN<@s?ZtQ z_Rp4aaGqlJd)yWQyQD3pqi+7ch5Y~=u9<07$s-R}?za2z)%c zD9Mxo<6Ge`QnGLJ_zr15KV-2<%Cpg(e9|)R&N7?tn925x8$bBWU)6!Ki}9JSsv@|` zAo8rTbR}ZRW(~s@0d{XPay8ow(C|J=6G3KzHZ&jG#vVhJFDMXSZJF3!U!lo<4YaT* z+fj;V1cu1?YfPK?rajlHE;m2Ob(EdeM}0CiB7jjG-^CY}suv_KXctlU-~hRjaGrE$ z%5{bOS;LvSlcLyx((75y!-K9l_0}=o20iz7vtKcFzd0s6fhJEsfwTRbJ4qPSt~gEz z)qD~k&lj)%Y-G~w%@jSu?p%EyT8daUXCq;he+74)33yrDazhQb1+MD?MkU`M4$f4Iv>_4j}R@g4)YzO4& zoll{7aUwgPm~)|4wrcb*j0{yPjhypmzhXkN6|E}w5RYvZPFOfmwhT$7-IcDejn*1mOXL@vBfLk=7I&uFi-n;ZuWE-&$QY_p0D>m*?J=EJ9?} zB1VKGR@Xn@KUi+c$M2=Nc@JG&J?N;0DOxT@uP|pfl5w^2l^{cJ10xP==(Usr{0MM+ zw;9U(dE%>i5Dxb2_mb)zEX3K9i*`dGF)q%I$6^pGe&j;Cfl_x9{VeAiHXniO8HE8) z5#Fw6w`;x;?2`j6kj;Y(nEaFf&U(jv&Lhc_2AStP5^XaVFX_%LPErGghqS}YU78rs zJ{=<`>KXRkTX10k%B)Jh%e1IMWww-~KD$-EvZU%#?rs`fc+_M5tPvA%(3u%>e-fL7 zYl+{}zDy5Q6}3$A=X|bvU2|_+o_T`IF?U8HZ~|$wL?Y*|+`a*ip_o zQP))r64-6SH}d3Q6|mNcd$z7YMD$VX7}+u718CLF{9RmbZz*sH!)YaE7#&9Ed7~zV z;Wr2-#fE(N1zl8*@e$5N#pC&lY!v}IjGkwVesIbUOwqTGAOor@)U6T227$>_^io{s z?6u91&hz@snif>eRGFUxly(tpvq*mxMI=EiY38v8ZX5M%XOoLHt?bF}Fw#P4v1XU_ zQc&&)M#qhX>7_tL0PdAbnZAqtVg&Kf>bf7g2ooaC!f_y8K7`P_fyyFPr&2bw5tJyz zVHIg=CkF+Ej3z~r4s)K!6f+t$0sgyr-g`~DewboDH2b7uC4xBah;^y>MLf7RGX}Es|sWUSm8ooYJ6>3 zaRIRw@ogF&D_e;npFdguhT^cgumKOXXhuv&1#&ds}9k#G{w31$`PynfNfv)_kDufTfPYm z74jH&+MMj`K#L?B38bUdh}qZ-1`$L|Dcg|$wCF*e_nhk@l4lUK@Y42k#}pc41VT+8`k}hO9TqBxJElsb7gWl&OIZ`KX7k~?(J$7CvJ=zGEW$` zuxOYi2}*Okr=0Uc`_b1{S`u2OG)vGY7wjeZHv{q@NsI3+mrt(L5I&LG1#0_9?P}o| zvN5KAN9(L88Ggjx`$^NIN_0HpvONd+;<|&y2-?Qh@+Q&1^(L{Vi|Z))hXmUyJNew9 zfIUjz*WuYHvNfTqN_D>9=PNQj#FU!eDfiV%>varAYno7jGc3?+1#QOx7>c zPg?#PC>oY|KEtgxF1(zgH~c~#k|ZvQqWsbOZjBs=I#*Ay8SMdSrOX^WX`)69|kMFce<;%m~hFEg;474xLc}?a+o(V`<(!Vzp%QR{t|e) z6G9e~(S74F?c;i1Ed8VV5JzUbxf@GaVV64JN)yYj5Y}TFpT6%~r~mXr18oP1LThR% zGZUe>UIZ%=C;7-v2c>DptZuhL)1K_^kJTH#7W0)&eYaZ9uaGu#e&b>H(+!0NC1gN2 zZQIYMr-AVT;+)dQpupd$ro`>Z<&4rY0nB>>+EGO#zO>kzAR_BlG)c=xoLsUoHoQnD zi1Fz1PsH}k$BfMC35~|Jk4N>5!{>7tgHzqgtEyQCs%zH$Mz}3DLkwA}zuJ6@Rkttp zUEv5nfBl|Cq-w?RH<9-uLPL~x!s5S;4_#+^5WVFiX6yUC??BY{fRO7X^gk;$wG6#5fEhT z$g`<}qq+wYg$FyB(lj5Hcx#09u1OrUaWTz2k?($YEDLy9l-JdK^*aF# zJcbJ-t>^1QCX;`HL$=MZT=8R(-FJVZH^}F%lz7>7S63is%e)*5tS7vk(--KEq*kUvBT`Q(P())md1Ko@tS;re;clvFu-2>A2tS-Rrg0e z$_$jCqE%6j!+LXwl2n{N#hacfAItkwsvvtzEf^G{jvoVi?jFi!_67cHV&w)qv(H+n zp$mTSZ$nt{gO{+`W1xxR_$h?R@7LyZiFN%EfYCj#x7KQ_?f;OJ<1a4SE*jko>x|EcZ+{)MMUPE4e)EIqj*@+9GHmqP-TuPwMZR05RRW0jv zng3Rq%RNw>ra|!*0ROml(e-}zi>;nN#6t~V7~k^h2)-Ry1`>Uce}OkUN{B4`*!P#= zFj&kVR@_hTzTyDT|5nrhg|Zz0+8{{Kcg4XOX zm`Vt2Rw4)VH;J{sD*34ZRUZ(qj^d9e;}!!JF%aSpb$k?;zg*gCLDc+(FZy0TnuJ*p z4&Pot*P1aWZD|7Gj%@Xot+wy&w9`QU@!2V*OsL1Y+xUBx&ry_?DXsa1`%d$5+PRd0 zR641(haz|VP|f8T^m?e_)8T8kYCl>N6k1PA)ym; zbG1P8^I`2z2caoYT06(tiMX1&%d&U}S#=lW0E^8<34^pGbqX>X5-=KiW|5%GsMBA3 zk-!)Da?5I6K;RoI;det^e9gst6||r4xG?!I1j-Y*fI-8!5CE!mt17ut~9eQ3#^wu5v}96G}b_yo4e8~!v<&`_c9Dz zqp9)3{cv)jA@`cF(WD{BLtoH>{Vn3nwN=FM&~P(%G)m@`&D z0*iX&y!5DqonWHc9e>CBhc$MyX*LbfFwyJ8xH!>7*<9@cnwxOu@*hAsbx#B3tcO&} ziAeI+we8D8npNiI<;C$!$;ialNPE$`pVb5|Pc%opG#l>AF-ZfBUym=4+_*MFvXI;k zq>Wx9Ql)1Vz!Oj2Xp=~PGm@5_0-<3a7;VS?b7VqS87ZufRg^Ue zt+E_tiqWsJ62^)?UWpI`0NJ-n6!!oK1{AblY;(ePxC~1ca09lk#Pj8AW!54;V7xk(&0Zo1{A}hKo+8<7RiKO}7~#D5+dM3bWcMKvkjF)yk3r0*Wjw`v-`do(f}K)UH|2 zv5j{jf$<;2!mf11iJAfZPY@Q0Z6nA-2o?e@htGSslvX52h)BxO>@tGu>#BDqC5fpW zTAW(O0E9^ayt0}70M`795AmIFr<+(w3;xU6wf5RwglpDEvf7vDq^(2(T5!d*tR&6& zNO`Ub2s}LlNl|w5L2M~a^SUa3rH2mZ!yn8e{u+yKnvZFke*-0*i*X~lfnF=u@T^Ms#>^0myyry=h7{?p(q^@CIn$N$);C6woLPH5r;MF zVm^vd$pb3>$>s|%QaI02mKY6jWh|adjUktmor{0<#_8X}92?^jyhvIJ$+{;rP7Foq zPaDEgJ3UC^wSBmZM36-d3?mIY;K_q7_D}3%`*3b`q5$hAHW?0>0L14W4`UAB>euV- zGx8&w+SHZJm1zR7&Q)K~&POg@6g&BFWP7{DMQF;WOx9zH>*w&=Jca#LLyR`>h8DBL z8M@+%!nV-SuHtTNi&I0WH>o<^kStTYKD%^1RVhb5<6o%)r7Q7d-MU$$;+EX|U(!Ly zCxi!pmf2Zr!W>J@PrsKHQ6;54e>Cbi$SG&b~~TB)dBKBNgAAw4pxd*lHY~;Ub^Ejka(ZR+*%94}!Qt9~ zd3hk(8)q)cxF}@z9ja$}eJoE-d9r&b(eCQ_Q@@l{h%swP6~$!-F?( zmDj9$M*rs@1fSzq&A2vG7_{eW9IM)X;VYxY@R5RXN7U|VvxYa0AQ!r8ua7^#DhTHdyt5PwOHCQc_4yYM?4%Ywv*%j9*q?mk$yiEKKv(fv z96Zr`F(TI{;Z~Z zI|`i15_Qx>_1iRUsCiOsZF1kcu@RQxr?^Hl0J3ti-p>2fGO1;^ zoMGxMP~zeg4~oLhUZmRmV+~;}_bC4@wI#=7kSX-s-Bn)&Q2h1>0FMHv#=K?Dyfw7E zn#3~WB~M;n{Ci}dcmgp+cn3RNLS#9PP8dXyn6te69tfYSH#VQWe$eYN~-7Q(eQY-gc0yG;GZZ@EYS{PpJc<&ey^#t;shnYj@{e zH7u)sgkFSrLgT@E9|bkv=~fPd^b3?Hn7U2$-1i5SF3!$eow6`kljRbBUA3tS5*Tf~ilK7km|>vdbA{!~uti}<)-b-vdf|5Ez$28> z^@Y&^!BP0TICok7gpe%rwmFjJV6&Y`!;#gj*I6J}$m$f!8$Qc{d6ox)C}@dLot3H1 zKT=w!NiO)$%4OM%JfLs*@^277;K`038<2P2zCi#O6(kI0PK(E1HC-}R-~cgnJ>7jE zKUcpMq{n3ob4E@WbtaF!1)3N>web$rFQnNj3w%@(_=}+e$y$Ly>oM)RFa<2==}uEB%^YarxM3Zr5tNrbPYbnI{k`8**N36Ul0)Rz;S0)P_20g}c!HJAy=UuqJIWgLLWM0R z^}W6mQbv$(_vo|65-h+QKlyTYVwe<7Sd;@ecq_0(X3U~VDM&VeO8L>#mm4BAEfkwS zRCdXF39Qi6AN6!#4KMe}tA=F_aUXS2T6C9Fm#@eUA<0*KwrKdQ)N^MSz?kge81@O3eZw)kRb0jg*r6Ku(-7 zn^)yh_awuCM(?wX=hsSBB3bs2>V9Q_-6&mtZu4R^TWj8riba`LKZ>u^-|XCNyrXOX z!5B{+w{@3030hotKQ|ziegsdy;cI)`d+r#wrlpQpS8e>w4xaECDVTFxrp82rB zr$SH`=r$Fi>{j9`qfG?5m;}0fz3>1{n*z~0yQu8p)r2;JmFYkiwkK&JC;e&FOL?x) z&w*v1l_Rr58j$8dg*xCw$ogI!F*r*nQ!r5$!V~UnIwoAq0Rdk?73(-$t=dYFEB3{d zmCn;-N2za=lQp!Y_<98QabkULU9az#@Lj@@WWMC4tdv@roMpCpcqCcb7^ET*vY_l* z{cB6%qouLW{OjzzX5}yflD5>fHE*>}K;4v^ z07Xq`l0(0G#)>0KgVum9rH04mg5Ch$^X8}rmDJ_vMVq)__*@Y{n4lGOribDAA*iUj zmKXR?IO)zz469z@W0O)%LQq`~;`2N5xtQjJmPyeBBUj-xyK~zf?3%z+(bOhmHP3Zp z;TGq}>_rVzE6pFSkj2(!)Tda>kJP^j`DIgKQ(h*HD3+N}zvG`B4pr@BOGhQkIr8KJ zX&7NC)K@8pn>ye}K=`8_LXTUqpP@F0f8v2H$XR`KORm;lF#LWm7MWLf2n zqGAp|Q$~8UN;N%Xh1GQwq0E4r;8rvt6%-lzv6M@24LBbV0_u8zw0;lHQ@*I`l(h90sH8y)AmBT zU*{&n^gpWpjxmB@7DL*o-dO^FPH-`MCatmlcohW&v|@C83dvmYX1!c?uz^8 zY8_OhxlOol+I0N{fX3?b7ze575uz3;R`Q^>O`*Wzy7oWh`D^MY`_G=P{ru_2(3=3s z^UjueXc_F{iK=B%Km};MSKJPL~aNzXv2fi@;4g)KI12-e&S%ziYKMXXWP9d-_sj z&`z)iD?n?4KV26!Q_T9{y`V;oZ4mXdwIfK+f2ZtynSDYl&MJ{ZIfjoJD)mQ!M)sN0 z$$zDG^t=fue!{BTx$cr7^;0_!P!F=-wSui>7+x`)nB0Njvt}0B+y!!##v+9(mWS|J$;fMhAd}T*hS5Wp>g+3Jf|3~1vnhegx z4b;`K!T=dbE9;xhMk_I>k%ha#uz2toL%y}X; zSYL$B-EeG(x{Dt?x=BeuQE*9S0lO6nAfz5AfFkq$J%9uUr-wLX$;WmrC_3!gZL*Ro z3Jqt?yVYi4y|@I9GI9Kc6Utdz68+A!c=?tH`D^c><>ob&@QO7Md)fSXG= z_Lt^9lp{tYy^hOu>T)8ZA8&3+{N_sgPlsd~|wVyw8xljlJj7_IO zuhgo$`Qa;kEoxvGS9^t$hwh#0Y*VCm?t+T!x=v1?u>Kilzae1xO34~89FzazOAb*w zcd{Q4+Jndcfu0MWI1%TkCkqg@S$eGe>QoCBP%!k3T*Fhy4l@&K$FWKz?Qkh&z2+Fm~{1a-p-pbzd)U`w|=5s$Z4PrA%+0EDU9i!!@df>YC9htd6);9XbbFvMrf>M|(y4>Z1C}ZijI$JU<84U8!$KJnNc5r}a(1pRUB$|#X$wl;ixGbZo(%l8?d;<(5H8PRzoW$i(=E`1 zl67Ll!fODVQNfWFdm`Vh-vT%9G^T`v&3zvN8XZQ zWePR+I`RIP=Ft^3H3Y0oCBrc1w5QxjtpDxX z&>H-a##D1<$i_n(`w$&(xWNi;-Gq#c?`_LO55Qs2s_K9#x!RV9HLa;ZDpLI&zP=lH z)0SvL)W0nFPy?!4MG4>ED2^H4xV66l*fvD^;3_a5z7l(^v=j%viZrbZECc1IhAn8> zK$!tJKLQ}Z$wecEivfl6K-;WzeVSmf8&?32Q7~3tRmTX}S?>35d3Lh{#eN=D%#RB0 zeET6GX?t4}TDiOaorLNv|4&hXW14du`Za*aJ=<3y*pa8e&A8n^@cAMMhGzEn0LzQ6 zz2YKjS!RZ9Hg_;rQOGp8KD(NfRpKtk0xC(n6CC@82jgOgOpU>AC8IK0w#4gEF<{sT z%z8WY4@je2=EcI7Fi5|}V$)V2fW&`vvdop4x?=j~dpEY=cW+xT9qeRO2vVXg;dgPB zAffTL@9B+wumUKFrFmK+GQ}c@e15f1KCO%AdU6uziAqWtAfJ|Yg}a(&XR!*I-aU|_ z3gn{y1n`aKbWg(9Y|{0#xFJQ;!cr}AZ%GzOME?Rpf4(p0JY8=0UsK;fZ5&{i6g~&8 zvNAtAglF@&ME_Y5I|RE|pjn!dX8^ga6107Zd%*j+JlC#J6c)0OE*hkvx zU-n^roJU6X=L4U$PqM2MDlAgWnnwxGJL@>~Z%Kl( znKJ<6C3Mb^A@ak)v-jhK1T%;;3mZf7eyGie&c4f%1`fTi{jk>r_|(+xmH|p&$nMs4 zpz(KqV2{IIGwSTxrCFHOsD!4k=um+%T5%3!I$O54mSO*Nw23+cExA&!J zYTsU&UX|%?}Lne0c|DUIWI+?*CGdqc+2OOeOqQU zdGg!8KC#)qJ-mbhP?@@8c9c?{-4#=y_G0+@Z|u`aDF8eYwoO|u??>%Z>NZC23&S{5 zja-M*jwUi0F0i)d`7wkN_?_S%FVNRv5+R;z4_BkZ;=dF-V{hK8-;y2eo2M$yVz(Y(9|hW%rNkR43)r|*nj?(>eN}P^)b|CFbkOH$@(&%02{T$y5t}9{o z?%ch-zyFoWmg;NJdTw|FLWkA68Co+9><(0BTD-hH#m652V!%to-wS#ODUo$uvGeZo zDRM~wS`SobyncIk09NFfdAwOdh#MmFNEGg$N&aKCS4jc3ulT7gr!@4Lr z=s)ISl8yP)GNBnxI5A7=a$<|zFg$l61_Lh6gAYkVAL6@0 z?p;hW0Srp{N#F&0n4&QnM^)VAHoZ=Q&seRFai4}6`~#xp>c*>Cezka#>qkb5_zP#1 zpYJSOBeFn0)Iv5~0B$5b@1YS|X(xg?BR6aOW8&4T**+zfC0Dig4-DpQ#c(SDG#mE< zY@V0StG6tAIV@D1om{Du%8=&A*9)yLLqC-=2g&iPg<0=M{9dI=lxL4`^n~}hZ3`y< z13?Ev>8@Q*@7vy?NYQAN#$tkL+VSwy03K7nvomS2f9=_T1)P^qG&oZGNO34kgAPgj z2Pum{qCNn6WDhhb_{{!?rbh-pBI2CEE!cn`Y_iDE#_rI`b@!Tm{h50O%%>&A4iOnx z;r5)Lx{H_0gqS&!nZ~iiuOj$CO7?B25a&`Iy0X2^{E+mHZ{%Mw ztHorPw3^9HxXXFAKG%Q98Jg&*0x0s67W>yt_pc7!U^Q}n++{1C6U~oY`}Jbk7YVMc z7;=BycUmyk=VxxiZZrZgZg(TK0VLxYuXB5wqsN$;27$)?7NxS)Q4ScFN`~Dm6uZpO zYk>4e%U%&A0|tO5%Gl-HfgMS3LULe%pdAnsi}z&{S1?j#U1FaL{9!8AgQO|e5WJDyA~o|gStvAXzxx1Ry#cF z1bK1-v44-3i*@)eL*3P%pUuoxJLM_-YP{I%cT3Y76H`7j==h%7KjQj7zyl@zg>dks z=OX-S_Q``Om3MRt@^wl&ZvA#6^=Bc^mg@hhTZb3<5mag~{?R#ymDL6abjHA&4+nZ# zclyT=8;s>NrIh}!$VK`Z4S9`)Y8!N2E5j};OFHMx2y$U^-@)fYzFXy?tw#JpkYg;r zL)-ar&_y=TpGSNgu>2hae%?FI2V;5aDNW>6EH(y!7*Jc%V$2bMa7*?6J>kju`}|m( z=776M1DA_<0$p+e5Bd`56E%q?!n=xl49*(Wss2_6QK0HK)J0@UXW!l=#sV5h05Ci` zKq>$?lYYBq&htWihlA6;VZYLvPLb05KXp!AJq<8!BT$zBnA=@GIZM9eV1 zNnGw#DL&t!Hh1y;E3x;upgn4e*O0WJKT)e4vif!=rL?okhG8T-i}R3$z1EERgR#!~ zdrg&Ut#P?#l;n20pHxpMAoT+k)H~Q<6J$D8ZK?q!A$~%2q8xhT3W8?yzlOUZNq`xaQ+SW$gA)1M(R;wH*9lk ztdl|CgEs!@x8J;{1e{_W-$Qq%q0MhL0ECme93r%|a;jco41NYU zc99BwIfn1ozECA47oT;-Hgf=?n9!Nu0*yGTevD!C&tA}2s&C&v!s%X9a1^g2@V_yXP!?l%XrTaaT+}f~&?2<4>T$8i_Kl zNyFe?H}_gQ%u1YGNZ%O~Lk%T>nx?yHO4Crz?AyPM@aK$4|2aKCXX1m#ZvYmWm98y< zXc5po(hiz7;NEmEXsmu;oR#;U`vxwAuda6KK(hn-YIn%BUnE|q{R*smTgrfGG=n5KVARxoK@^ z1P~YBNJbpJR|#C$P^6e4sQ3XkBc|j8T99U-CB~?95R_9s%z*p;HU;loYBnDQ?2LmJ zJJrC?UoAH`*R;$93og&SyJ{;LxU->5`lZY8eWE#9qYt89u2Q>> zmfW){_S+J_NZ)EPIv+I^;Ff(&R=My)Ive||P<5r>qZ2?Ixr9EK5qjKz}$QovMe|SfjsZnd9{C>2J7$b%n9)+XbL9#FQ0yn=p2TTf$@MLID>$;m>zynh9+XU z`LbrS)wQKCCvY9`ST*`O9ZlP225j3cz%$H+>|P^YtS4GWDGO+G(Xx+1RC*YX-+f3gwn<9S%`th~ zjoknARh7*-jPx;lCzqlW*QQHM3sM1rAk|4cDdV9uvGQ^7LHUsRXUrcwg52EPxIZ0s zHkP&kM->{$(YXtGnm99;jvSMs)l+*vPr=!kD?vNbliq_M6cjv&G7w@@A3w!zHIE4z z={xxlFe{E$mj!HlH`rXoby1(SF|eqD1ZZ`%s5L76RU6l(vihhd>rA;Bp0BZofz9b zO$W`unW^`#ap{4$SDXI7Qbxq%$+3Au3u!l%QR9u_s)|Q-)qq0O`agWt6MVUdzQjIhefhpQb)U)47>%F4jjZB%A@%WBVnaoj10SE3n+#>WfN;eAp`e z)}NrS59XNMmwqpbiDjK@+X(T1Q723H38&+h>@CI(_;NEld+x zH+)?nRdcxW-N4VlAAr#dN{ed-p1~znpBMq>`KF$k+4~C@FCM+(WmfbkT#25glLaWT zVDL)KaBpu0H0@&x%!0^-COIU6At6iVTYXzYO<$=AF+WT)BSJHXYAr{00VkaC zN=n~HoT%SIs||G&sy~WF@77!NG(7vA9*tUYOJeJa0L8Mw#Wo6|m3DBdchk@&aR>?!$XOSb$j;I1Hc~oI`rPEM?Q?V%8Ya`Pl9LwO$OnL zc)8QZ!iA8h=vQtRK$ZLJgSL@an2Huubnl#pWEZS%^ki`Wv>Cya8dkLi*fiuP$l|p* z@{83)BjwK01c=Z8wLMTy1H}%N;}P&6XBiJe!$oQ>o6`&wOnw6IsZQl=7s)HX4F<{3WTwt0Z~SeZH`Z7IOIql3^_0=cY=o6lVt+WoFKPi@PVo9h`orKtzFrJyHc_^Cx6 z2?aXXH_L|DKa}=%R)AtfTXSxGK39_h#!Nk_dqS*Uepy*nB>--XLvi7d^v%8LpkMaT zj4?pP0OLe@UkE1oX!Zq3?5(-(K_m@rs;4RBm`cD?s)jeLt z*SdiICi;F<-tMR*0Ja-01@<)DKtWK<|3lfEfK$D;ZR1OAX|@|^Qg&G)sWclR&6Wz8 zlc^*TqRcW>J4#8GsWO(CjG5<>xs-V(q0CB#5We$X=y~4v_>SZM{X34H-NRbH-#uRU zb)DCFoj1kduG5lO2H`kwqdZwCtU^-sJ8uF&Nx#EXKonA-BxW@^T$2a~kLdLrOaGiJ zaP#Y#CY#YFMt75v4(vD|fBdv3f8F&)qG&Vu-SM!zsp+u&&s$?98r6QH)|Es=ZA`1r>v$A*FW7_TWI`gXp12P&qedT|?>`NKd=Nu> z&d*v3KAW7duOdHt#iMn5O~yLBdg4oVVzyX(a=T2r4^OonHt7oXBK0dOX(r)NL|eW1 zo(WtauuUw5F}E%)tjzLZvuG`{qab^Ez26kmFR#wC@5JiCfo!c+q^&Z~&{m|P!LfNN z&>En1K)+UIu_gQw(@#QyWnt7lbCp&Zp`KT-6Rd{2kGRBOe>xv*w#XSQd?P%N@slAz zQ^k=ZK>%MnV_#tEZn+mb(@uvFuo+b2v+gse?OG|k2UL4^>e%Ge{KuB4xuCMO1gv*3 zE~IRCD_Yt#k7?s) z)Ja%%33DvxjNfhDLbG`xb9JHO!T^-2T26KXF#dYhIfk_-qavTzcW;&a2HCmDbm>91 zHJS(+XyxEwFBkj@p*dIiUOUsT&vZ83C`HeXIa~~ji!;8;z9`@lLOzzM1Qq0s+Kq)w z`dC6Ew*zjej5*$(BO=1Er5=%A=Dk7%&q$mQL^LKbQGr$L*Js<& zfy_~LTHcfrwm1w}*^zJQI1oemP?i^Weu`{R<4zK zi=sW~rQI;x+vCkmv$JkpjTSsj{W~MMU=;GJEEG#t2mV;=%qH&ZuvEi9>rJ#8|Vdn{6D)%&nU<8&H*pYaV`A1c)W`X zSa31Hf*ni!WQ0ux)_<)$LO!&X=5W?NQRm|(Q*(33!BP0^4H{H_4KCPmzG185W;OF4 zCc*K{z&3!w6=yOOt8)I1>|FVFE*@v!p%SFJlPBC_UMq{(cP5)|Ie!walw=&Gbi|rn zbF`fZt)d5I3th`VD1p4>bAUYQvv{6s$QE-=td~h5j%AqEwKenm+a(|@fij4OHK4?u zTO8H=A<5ubNH`-{%^YpM;J`MImjy6Ex=zj2rfMkwU8INLvpFeRDy(7kw6-K;Vy_kS zMRFmjz;WcXRd%>om}KBC8cdwv>Vmtm(Qxzi&H#yDX*8?;8}9s)siuWSP_jPStNuBR#PIqGbJT{$2iBI zK(R5T$c?85`errrSq}veUw}RDG|p)Tgp*NLBq8Y6>?~MQYZVEWNTGIBM{rNO{k+|id>3sIgw9Dz#t~LOtp4ue?rr<(Q zL)$)ObCKz%a2>-pDgV%Rq+1P@9DenLnmO%$^6YxKDE(4-xwFbloU^@qj;v>xAj(=v zZWK)9{Z+PVwx!2Shc3%+Si-yMhsW%4YI{UkB7Oq4(9S|T6BH89+$YiL7HRR8a!mdN zE8v*LcEbSj2fTa*eE}9fl^wLStJQ-0sdtLk@G~$^Qx8j28lcKOEemrS|f8xZp{jr%Y zi;GwPCwl96!lj@jwbRx}og^|KkNyRnNY$M~{JlI^73_C!-17l@Vs?RXAC1IhEf zyE`*j6uwPJ?;&==?O`1;YVp@3}&%vwRzn<0$SFCUOTT$VY+}A3GXTou_WL}=JX zPA6_fJe*0BQff8w`g4GVjECBd^N7O3j5kOhGWGu$I-2oK6ppgOp-o;T9kQqfRZ&r4 zXw;gU1A&3fC5Jw^go(D}75>F-7nN9K4B;+4P&&}rSCtOtmgyR4C2;)AgowTHdi`ZD zV{wbG!%;~pcOfcy6Co9S$Zs^`6!4ho0HWt8=6^k>rZy=G=cWQa;6mgVVSh>CbdSb# z8c}BC_}DRjZ4Tw^mqV@C-FM)#csi=zbMT+ zkRdB|WZ&+)BP@Jt&_f-U8o&NWE30g)PR^e-#sPN}RPEs>PxWL^MGlP-H%!gE$u!tk z`7Y7G?1VZl{Ck{uU@uRLR-R@Sknq&*Y5i8M2jMv|9)YHqXvZ^KithYRil1WiH`{2m zF;R_9H;cl4COCmjI~S+Wm(-o9?JI2{r0YQ?053wjVwoMQaxN+-#;Ljdwv3{7#g9tM zpM4i&LQI(iw7uR(GQvaoG}u}MRflPGAE{6V5I=ltdPsVj>J?er-IMywPo#T{!r|v9 zFLl`VY8>cyZUx6p`mLWQgz`i_57YReXE*B+u3Z3cjf5v?*g74b7st04SNt)P!4N~o*p>PL z60?P0eRgY|%uZ$E;JhBKS2aDU&B*xhY+r4({Q@t^0NP&9+jIw9RthCd0B$M#U79Sy zyC)ykJbuFi&Vd-@vP6a_z5pZa#@oVjoeN~DnLn>FE-<>cu2VWBGf zqcZ13ylIajZ9Uyjg|fUa*gUB_FFBb;7c(o{2g(zkg>2i{p;>FTTpHg)wCoA*3s47t ziE#HAe@Qt|2L4=ueKzAmo0T`)->u6M3D@{-n@ZIkp$!WgUpfN@J4g+)&v-0>XMFRs zj;4mg9=V8(cb@w=O5)kquV1$`QXV@awW#K@^BhbXpz5!-n)EcQ8^ph4}XuUT$=9IpVn+;<5m(Az7>?g7Hc-9z{dq_tI#fdqTK)z0XYe*ACP&lzZij7$>LBnK@EV9fc!iTUVcn}e}ArDD6p`DO(t$o zgWi!VF!VtB)=_|d1)6yjs&2)|&Zg0JNm@Cx&ZDY-^Z87-zGmc$2j`BM)>mKV@zc_} zLcI4GHVV?pFI>+ICf02iY9j(Prb7=;(v;LKgB7DxxS_iGb`cN5V~Kne5gJGLj1k5C`jO6(S1!Li(6W44UHaQwdnL%^Ld4!6A@YiD? zh^z!75CLotE#{7?L8a)2>1Y z+`HFu@UrOd@44fZp<=9F^}~{z~kPrVKImrVvB!0#ft|5VBGxAS%F$$v9`6AD#>p z9&2|qMY5_JZY`8Q| zAM#2R1_h}I;UYnvREJHGjk<$ZKK@Ji(ImPPj{ zj~gE+1=&tbbkKv&gC#rI{_lv14hr%kw@S|B{65#N-48b<&~4XK0G7esou#1lfB~@; zoKrWHb%UEB<#?{2n^1))CwY;Db-dh=#2_JNy;N`rLN;WlSH%lr`wb-K+*o@pkTr8Q z8#I?iO-x-6rgmjus)IuAkHC$E)W(V*Ub9Y}>tlTG2vE7EvG13-_cFs58~-%1HA2Kg zDd(FaPLtUj5v2f%miF_pW?d#Nqrm#`ssK3jFto%oO7%@8__x2C&lx}2M%ozg9y}b5 zPu(BD9J`6k$>}T7GFoS4h*`^<=cF)JoWbysa;}0IrYT-zqJG%cDDH;kX>jLETGwe+ zdA#4&YaY3Ih5H}-g&Y{Aw*b-2f4G4?JN=?Fbi*}O2)*gsi`gyy8_s*}$1nTShu)QJ zpwez07NEu5m-fY*dF~)|*H$NNV#uF_rx>gT>@@!5T~%0NkJjc zA^511>a~!(W$itJ0BJRhJo#|qp9M=xBTuUg{r0{*v=gAt-1DPjpx7&U;ZO+w$t{_`atxlC$_9$zMEmW;4ybYlIZ}a zCrC_6&`8i8@DG#Ux+}Tl1$Y*zK9?KGc-Rj{rkIBUVt7GBS6U|f%{$Z=3w+t3(Gv2F zWR26te;1*8jRd|Rh|Tlc6T>hT_Ai|pZtqtniSP2@)7-@VYg?|Fzx3EWc)6lU^59+K z)L+GuPYffp2?(pQgkdEDF4rzl-j(tF4ndfZF`dW%dExILRnE|1 zV}#f>yW{1bT%6unMXtSIT5DFHzL$^|0bhap2tFo`7(_%6n3~-@l%a#Y^RXQVl$I!k z@pveKzO^R9{1qy&olgBGsqf5vq~nwkgn*Csl++AVBR#%_HeSf=2J8Su%{ZwbG(5X1 z87y@oG-gs0n3vxdDn+~6T^oiP&nljIvp`6gGMS!1kc1LmBpXkVjDQYfl_r+WQDk$N zKyC~?e#-nL>mfD$M%y3f+t|LRYM%zl04zWy#!?8l@gtgTEqDL<6jjl#481^Go6iYb z$ytmd(OHfxB}|{ZDptSAaB?gviXIfX>*-5~`+oYSXtRYkH^7(_Hc$cxlai}3%eg_? zTP1MSdM=c86OZ<1$}FrzSss3=CDY8zEFw6#BVcqwza>j8$$(X_IX^!?P^ITv1(SoYuv$DGm1J%+E-m#0M*3p|n?@H^I*wP&f;Cy@ku z*LKjqFy%$^`ya|sP6|RnM^$ZFEWHj;&e0n}LNiG_fsF4` z>}zj>q^a(0W>V5)*6nnYCt@#&^TQ+xXZg?+qgp;gmTF|`OljrXUTPr-ltSM}GCBB* zs;v|KK7=4BSV#KF(qpam7fk0&R4(zX;Ub2?Nv{l-TNG>0_5ESXE;%TW96J>7T1rnw zS~zskG6e|Q&axdTr~X}eRP(R%X@=(E#uHU*q|TZ1MIKMoD2tutC{jKbDl~#cbt)eR z`;aSlJh$)eOhwbU0R09PTs$_E2`q_#wG~yruF))axY#jy>aQU z#y@YnM_IC#3AQY4I{Z@G2rKxYXG&!GzsPNw`>jr z*5U+xB2Tc)dX^GIaZm&fkp>wa6Cf+@*6SuLF`A0y?k<6jSEpRX4>K&I^1SleljqUW%9UG7DTgnS$z zvkk#WVz5s7I4z4M*SnDPZHIckExAs{>usc)lrW7TDBx-=U#A&${`E<)N$5d!uXQ+0 zdU0$kRUDVnuPKJ&$ki#avR@;KH`ZL+@T3-vuz=$D_+ zdF0m6B_p?mH*_`|l~Z`!-t?&yt}NoRQiI%4moP3kG3m{6uA7it;HbBJ>GeKe-5tF9 z4gLQ)odgwR#_`JlRfr?mBUp5KPw+LDZ=WyTtw|q+A}$dP(zHlNex2~c!p4E$=8+m*^v_!>*Np|A+Y-P z(@xF}hxb{(icj^)uO4^*sO|pI!_t^bre*I(T>s*L>t>)GGk4t#i%x@Cp0AOHD-$B? z4)wQB8j^LOHgLGJR@LzO`c|`aN0Y}~xTz*T&~N^!Qn-y{$0Mn_<7@qeM-z%&&BOER zgZtP-hgNmpx`u+BGD!0W0Ofs*h?0cer-~t#l z3G1DN77YIPRXO`FzWr-#(&vWKPM7SIBL+y7Baexbm(9 zDn7rI@X(=wu;b-4aF7(*FWnf@UKafbkqO!>Kt%oXpe8~(oA?NaPerD)x=02%f|G9 zNu0u)edp(X1RY)hd~kg=3Aua)l#7Haw|`Eb$v3T~K~%*J7{6Dvx3@Q|uvyE5T&UE> zav_0?4AelWtI00qI((~i5q!xAY0!)1)z>rJwwyTJJmqu7NATQ(<45U3_19uHn&=ZM zNw;llh0Ocgt$~^<<3+g-xf!%+iNZ*T18klyJJ`W6*o@n6q&Sv40=C;iD~HV+Lc#*u z{(n(DGN+xO*SM6^Wy&&2#f%4@j!Ee@XL$M%&gZ;I4FYAYjl_Y+f#iC(8>vn!5@X?mc3c+;n5iGc!jFKIj*fy5JxeaO9|Vhc zo{q1nry|(2$}vKJ_yMCEorGl0ZyNil0zoOhPV=5#w{8yA{{%NBU!I?|I$O)xtRC~q@L8;Ib4)2*cQ^r3?&|5`B| zDJC@&lT#*8B1TX}FJ7hCeOY3!5MLl|3pE4J*>1|O#v9HwE7wKTC?ynwdh_{w8_Hw& zAj~})&uBuJAiBA(eK5p!TmmlYt7oR`l{gh`4xu=AmN!q~V`XBOS->h<^NOGlOv%mW zl7aABpxZHLMx9Wy!l5HlPq*q!g~ueM^GE?g8W#V~1dWvCsQVBs?YA-s2X?!vJ95Zx z+kVi$i@VxgwlEsO@;3d-XTkBF!Hq;8fn)hGP9ISz2+*R6A=Qhqsu`C}K6cY@28_tQ zA=%&Zk}sg4u$`NDe@4vo146Ai#)BvVOSmbm0)O*cW-<*z&5b(|>CQYN50yuc@1h#i zs(mbx9<&4{P;rSD8RU5eeU6elpNZY*5s=07P3I#rbahcErgJ7n!R8Pz24gl#a$rV) zwsxW z`+VNx7cBng1DiIkyElK6n#or|vh;*!m)>V100aevNvL3M|A-(9r5K8B-MP>#)>fZw zDW=3|DY6m%V)4hBE2{r>;6q_;X#LgL#*&$LC}U|nzOCeCSR-=Htf))Cq;7%-r8FLm zJEu|@i_?f|acbF8QsPSDQ`}y4B{Rgug>Q~)Z_kA)j-_!eVy{mc_^wl679~qZ-hW}S zz}3bX)9H#gIDGP~HWVE{ij|Z1y5QQ;W5>l@BL8#~82a1D-o4`YH_Cu^ z&d&JSb=X~-zc(C?I%ll}5Jbt#PxJ_)om4LdXDjDEfE$irsJ+jYbzz>`Hbk`F7(6v| z_P=syb=RC)z3tDo^xLBB0R32Y(=-QP{ zd>_gyhhxLNujtj+#qT>i`P}>2*XIVQirIS_8gE244L|WP?Do8($IR#C5%T51J~kJ7 z`^AL|e@@K*_Sn_!AA!-Fo-vCdi^lghO{1zg``vzi9aqJTKjOiVDQ>hMu#^1c_ls(o zlCCQ=YQ)SkI}YOKF^rPiL@Od#vm3c+X@Le|Dvm6L4HZpAnT)9pfEk=8saupbU)ZiS z{`WQO$mxyF9UAG+U#Bw$LOd9|)BSCkoUY2oVR#tgHCy#o%U&=vWak-Ws0jxXBGzvL@8(+{{x zpWUTn&R|~KOG0ryMe3APh~9^)dkrjRtTktc^+ZxnF}^Lo3_IJjuqdu8^P_pm6*rXw z(SiNR`<{e4roT>~+1VcA_ROk|XGL7L_h0RQ+h>;+8IExfNk2XFfORl&yLX1@booc# zTw^#;_TqN^$-jy&#R;Y=&3s@Z`M}iVux+q{9a)(4B)@H zJhDSoo?P^xH|F=Y@4J7mfw7IkuHw6!Z`{xypIDRbq2*S;ZD)9ZZx&(;4td?kKwzohH@_|*?$v~p2jL)rAgG^u_D1C64vhWyrU za+lPa^4F_XLmm1vUq3h`Y7x$8*k}G*d@Rw%oxdz?GW)`)UDDOfrL3LJ^oR_F}%$D0`54pdq{t5GRvt~p(6jBV?;mLYZZ3m~m^w>X)7lgbG zH4!g&b<-90keL0^6|_~tsovusnqKq$!6S^1y?3B2QBZGgVre)hHq+m(@3f5y%~jL3+8SRdKF^j}o0xzQv{rT~)caQmc zI>YyI`=qOg)jpqVE6S7=6caLUecxKlD43)O?48FiPdniKL3uht?egDq?^)|#e~x?7 zZ2Zp3Qnd^fv9(`%`xy6DgcVcmXqN^fplLQdM>zZ$(yR`L8EuTVm;N5MXKp0+V_=K# zA28z_3Y>lDnxkkeSGS&h)z0{xvmH+PTP!b7QqWBJY{5s>6m+Ld~~%l z^yKg@-^LA$m&mOAOF3-iOS6U&mO7i6#wz9VSjt=B`=8<}QakRdsnpywo4G+yd=VN^ z{&22WUKlOOT~Z}RIu6a_so-K^;r}D%r$BY21Kt+Y0dTP^`E3KZWfd3 zXL#up-eohtsL9q_vw|wkYR)stsIjEDFJNNli)z!^*@>w)o1KR|^77et+ZRJx!-IYH z1@g<$33n-<4?8$bZoUp%l0G}c@FVAb-YjV2^EisJlQ+DmP;6)HXl8*OM331rg*5llwW!n7f(g| z6<1-onU9JGi(`!PvMj7E<{#c4GBb1blBDsH32}c$NS+;E-zT+Iy)uSJF2<9!`p<~3 z`(b3p67dqZn_jC7J6BVcguLtc_gvPWqK2>Q60Uq$Zv*vvr-u*atgO=R+__U3r*+Di zLr4}-!Hyn7Z*`OA!8bA*e})yBB0PddpL3R}`9FIWjth&g{`~r3xeE`K%DZ{<5rT8P zd$4LhSdAO3HYu$-F}pN9Rx)|6s-6D|l~Y8{Z&$p5xVU&$L5Q@!;lE0)t*r#wVq3d5 zEYo^a5Bz*dY3UlkLu?!zuWrw0@wA-&(0TdG30SfVVzuE4v9Hr-OjYsUGfyonlxRW{ zmOJf$u&@fEU{|VXS2<`PJLtGio2r{*&$NEe<@+Zet$pF{zM6zxsN0eUYp-}@gTQ9T z!8hZ!rF{(0g2F4u|3&@HjT<*ieto}4lRbX?_=6aY42w8wQEu)rl)CJGyFLB;ccTwa zcORp>)9EkSwktMjx~SdFycMf2EBWMK%51%#-`vfQ%l!8{q_~}& zoXV@CE|@|~{Ai4*N#_$|Rn=cr3A%DL`iCnwElP`dtwyVwSh(gIzvt45(ZyH%ZhcR^@(8q#!b=^5(1BVYGO3ne^y40T5nT3di2h2uE=;HUlZC7 zp5AoU*?l3d{ch>2#aCEkSFTz${QKv}Z@7|Mcc3}b>dOP31Qfglp z`1tXopcIX^&BpO;DHgojc$VJL|EwTIX0k4Rv=unVY}l_CgrQSAckUcXENA^5JT3a` z^WAu5etHn<7*A>3`f;yJZ13I*!7Q`UH`gpJE!iy8v{%`fPDYxoI5KNLWoYM)D&IzZ z=ho}juWMhq5@9njG7cWJe5B%zwp&o=+^3blG|{k01-AsaQAy!RcSWd+UiGW+?Ml&W z4Sp_=XD+f98!%<=0^axHo0P9`FyJ{}F0Q~rYr))S?Is(=T=^cJ(A&_j)7!t$oq6Kv z!M0kK;#FFl3T%J=W?#jL^Hfy}+m|g2Y3=c`Z~mquz2~dK&gE{-D8Ban^Ba|H+e{*_ zkHu>P3c&h(SDyg>yOI#65_Q5s`TY6w`NqZ7=<_IV` z^k~ds?4zK@%6(h5d^wqJD10vc#{5@USrK3J1Q_Q}z#gpyip*r@aH(?)uo(#q=!v8z z-2Q8fi;)kWrq?H5-z6^#YU@+)6);#`r!Ag0Zk5pRFu>!fvaT2~4 zs{WXAAuhC2EzM;2W3L?-6tAYYw_o0&k)kxb;)T<)$q|O^f6zl5h$ORUw+1P@`Y{tV-9U+VX1TEqPF7>9-v2B8IAEj_(AkHST z$zZP_6h2NP%kg__F5AkLFVP9zuJYhClP)?{x4SZ2JAVF)XFfEkk8=OzlPwmvS9p_V zJ`jzf*B+EZ%fJA?3=Cx3xY0L8&1IA5P;&Owu~M>~$+OUNXzVRQrP9aNjJ!)_ zDaLI>IrCZF8%A6C%hGLkaW0LS^MVzaaNG>_H7IQqHu;1!5~mG{wa5FCTP`Q65Mzm6 z>0-NfH)mSEs6jco8t1L=w*iDjDMTu6+qqL}%}&)P{{H^xc|cjrJ3C+DgLf+g8)L@Uu4|Pn3p=>XHyDai-n0M(Ol6*T93TKUa?DH|=)`54-h z6ZaGT*^axuvtfbE?S-+QS0=S@BTnAHoJUT;B(J}YFv$-9AYrd%*^kg*)n)PJ(5WIw3*~O{9 z`fdM=YcH?xm$E?ORqHX$zvfme`AO8xZ+jE`XZ_}gLuf&@rv@hI(Vr(FA`W58+m9S zX*%oq%^3)zt$17My&?WSQ96~D7Hkg$ki!v|Gao}&Yg-$lkYy!m%s6TN`7AdFNJ!LlaU$MiITe9DONi_SOv{w{iN<1?d7C=J5+EG?raflSqzmNOm;TJy$aq zcW%XM4vilWt3C8y!}L~!$Xw$d;XV6s7%z+kn)9Dge1bsJ=j*$=hr%8fw0=W&Q~ecl z5fQ}=>!qUBoKwF(M|!xY+U&-4MnCP(uh4v>+lx5h?fmwxH!_U)l)}ul-(D_AHRaW0 zHuq_|p)UKM-tWbNkb9T!>-k}86>CIzp?CM6&c8#WR-YWxbfYMYju_27{~6+#w`xgO zdP3)aqtYtK-ku zBNVw;Tok8Ukkv?uG^h|i&kr1m5x-NOZbTGS*}7_*q*OZpb!l_ctDF9GehuQ1#V`u` z)rA_SJ(YhNw;f^Py?WdIIb;o)ADj(u#DAb=7o~XA%7F&lcrB5fG`EaYInl1CeTeN? zmkcW?-*=hYTdGjW;(LTcOTmp|X_fM4q7)E#A60!Ht0#SnP7SxY{N&FUK-=egSDLc7 zxHN8Ciqx`1y_KJ5gI5X(br-uP---oq{oA!uhH%#z0#p>27C(m7ftOS znS35}`k9h^WLl`9?tk_AnEkyQYhGi~$7;6;503u$GwG3=bB?CQ1Gh*`n^Nfc{@=rx zvK}vrY?0FVRM&0-II*jmebye@(=qy~PIWgW>8mZ>{EyDR_Gc>DFv+UqRQ<=?++ilv z<`r7W?Cu>@xpbqj-xk=R^J?GBX2RVRe!RhZ<3yPBTSw-Yqd(62-3Y2p)4FqTH_UC*|)%(5&(v`tWBlQ$V1sUogsF zllsPNM&AgFRn6_{&)xWOi2ecZxG4qi|FNfy5Fqo)IZ6ldZrd-X?q}FtyQOw+PkT_9 z*o)noXcw!C)e0XH?>C$q^9`ra)|-QkCpL=M90BUuBsm*o@N(f5 zdtqR|Bl3{{STJYLJg^drADt7crHZTj_0OFOVf<+7)>4ha%BdDQH!XQzOwa7-+7~0g zqJ3$id2f7O&;rwI+XlDV>6kMzloa_bc*iQdL-Fi9`M~F)bBmX6)ja;j*r79#j0pjd zSNP}c%b3^=Yo%FAi>$a$-Oj42+1effFg?_(|2kWSxi|Y2nLi!U^^KC{m{)Ux!m~Yyy(a`Xx5s*uk z+`&^2!Z^?jycZNT`L@m&w&G0>=1cZ{f3TW6_)2j}#o6Sfu%0>0ND?qxFi!VJ`kgUp zQ~T#;Sq2vJMP*W7oEPf?dd1J@uim?bkbvjr`r>b2D=aqsE34YwHWK zv~9KjnVEI?0{4~c{o_hex%1^eH01ty6!{H*$6Ia2(ce_gU;F1bFlv(e1z8oCobe-x zm9Jd0=MV)x*hu%MHj|P6dvXBCxl4TQ*gih?`8*A9u=tz}k!ME3T>Y2(DO7uKRQ(7K z@#eZScVNpfWBDH^=YaQLg`pwIb5k8LZyvvDGC8?wy*6kj8q6)fFld#f7*0H<_8d)9 z*dJjTcI@FF>m9abVXRS8l7_EMRfMtqyjziT`#T`wJMbMDs{`I=0ezG~Cm}i|MKCxt zbUQDv#6}Ub%0gGJD(JfGv59_~sWku9RwkU2+GevS)yuy%Q^VfBKZDZmBd82Evp574 zXVIcXq*#X}rct!)Ui8!V%$b_-cc*WdCwEq}ztO%ilPk6FIbz@O^8XFPY5xAR?~suc zy)q%oe|=??Y3}HZbD9F;=uY1MU$D(VcHzE8e~g>nN?1lushJ0{Ia1#;?@-TBrnk4Y zzQCPKYg;|k5VHY^f=DWjR)#V+!_F~%`f`rbuGA)ubr#1)XLbYcwt4((-2Sfs&Eka% zpHOX}B5-iXr+WK=I(=DS`TJRyq`a`6S+DM^nLHmpaCTOZl6p+s+_x#Iso`1$kLf`O zWVh|#pIG@(_d^nmuCgyWu)A5_m+?k>rxW%amzHl7zRYpzjh`c*eh|}xGl*KQ*U%Q8T%Xz;FDviKsaOr&M&< zZh9(S8Me?AIf(O7ik4PZdq&O$3yhpo*!}14L$mQA-PH7Sk@OAA7sjfdL2&p;E3!)E z@_i2v4{eS$Ys}cVd3a748lI^>R&@iPrO8T2+;_8{C(r)}vLNt;kdKX-;yDBR;`@IfhdKCR!0M@D+ofB#Zar%I!hYqW zbrwK|j!LeX!5bu#^7if9M{VG#u+>$Iam&(=9y?YNIg_%YbU0xa{&jTRP~fh@7(p8)AOl+6%}&y)vH#?T}|)p)$gl~J4VOt);Ar?kIhV{(owt! zji?Mhy@pmk7p*r$siiAo?&v{THWP}56!PrfuK?~M@HcOh_QDlFK4^*8Y7_y@4LKhEeFRD&{_Pzd z3Z28+yZbHsa`=H>RA$@R`HEUAVlPz_`J$r@dq!G|zqGX_nsmvcS6v|g(SrcNcK4?b zR6{K$2;jqNtEeiyDj*IK=vL3C7sujNq9T-H{D{Vj^ZNagxVCiL$XLmjP2x0dZS5}9 z^i-heFc|0ruc)Y6R#p}+5_Mj%Y<&>GjlxTDtF1%_CXBduW&e&R3Uwfq5qYV+q-neo zvh%V;Bu7_Q7uH1vwV1ZEd=c|Ykdz)X)^7oe^JEM|%BbdszPIEW< zj?ynK6nopbRq5^ceoI{3o_(iknD_N8p_WwC#`lmS9kf*Tb0Vg*f+*Qwy|9h{);4Ah z2!gUzAoRUtjg5`(C3Q7trZ-#l>;N>$)gsh9+A5i3*c5S!(?(uMNC$yJGtkO z$hGV@`Yf(v#`c8T*OIM4B>u1-sWcc<1z4O8tTK=x_0UiPA5p{41)dxSldiq3tF?gK-{6&@rRYn~yn!;eOxZ_wOqKp|h=8#YGz& z90ZUa54XR-Pvn6_o#YhC8YEki9sqarw zyYr3ndACJYMF1CE->MLJ(wS6bZKG0;pL)zuhJHOd{B^jq>==>o$g=*P#i|qsaRpJ- znGzt4%C6L*u0Pqa+$q%B^wsOCk$gyTOBUPbK&ykpeWd`BnTx4hKJ@1oP5lo`>28Yv}e1> z^+Xi|jNF0x3@A1&|5kFT@n`+*^FbjY^=W2(dXEZ_L1_RVSwr(ezL;tiAa(tHG9!g-#)zI{fv-~}cfFsQF3Z39mC!943# zKa;zBv+*cySgpqta+LCLbJz7_7y>^4PyB4;H)@mO1kkO)FmqVZ(e7ggG)?X$9lrG| zmelVip*J@6%;U_)gYoflz({LEv(2a)k^_=-vh*<`U$5;o?UvV?M0w$OajTC_MD4v{tk4-t)l+BNksD!d3H_gz{kX$Sb?Eb3}IeMOp*;r+bm}Ly) z_5z9@8fm!R6CUa(tD;hGsxgXk?G4je{`;?a-mi;j^Uw zlqK{V(=4yGOW0|$$%ajfRe2=ncvGmoJ%v_I(jylq=PxF8r)?==+q#>Zo6|rY=f*vI zwnLrgXxe)jHuJ#sX8l*zM6huSF_jT$$|KeeQL7=|gnb`NN=~ETu%fdpkO;)^ZQp*H zzIx4?P_9#tCFrC)Jpwr)W_3)P5svoL#fJ@vjnwrKG(IY8D5E`{+n9)*rNe>j;MxMx z&m{f2mpI?Lo6K8?%Y^L03*gK_wLM6=C4h%clf@#&lBk~7ka)dC9XUEWH40ZECmq_p z?ODerUWjW`Nou0FsECo|3;tw=QSGjc3#1ZursUbPQBS6rn2m4VlskU>Ul@wu*x1-D z4PPgh+@IVfJabAc#VPS9+k$n&fis|PUvTYPH({gz;`sjRW9trC7`Yl??}m3(g!Vw} z>VVmcCfsFboE28{z{SNyB}so94f(Vif=cY86v;p!zVsmE3ve)n=Nzy+7MYQ5G?>Nu zdIFV=&QFp6;Svp~A1s|OISyIKU@_C3HhEco^vC(cvYbPpNmK?frUX5%FfnLe=$CJd zq+Dvlw@rsEVpS)YS8Kn!zd{94ac2FE=X+B`u{n%;zkTBg*UuM?+4%H!rilQR8dO`S zv{H*R`Wm)NzH#%jtvDB);EZ3s`^U1t06vQAM{UmJ@bGY9P*9&W&*ybUJRn(Y zvn$bSP^|}2rvi`DVFoyU!a&J@q)RRo8;9np^KN;cjJ6)Dg5dUF)2m)L&Ie!b4z(b= zx-v#1yD(|EyK4J);hb0Gkl?iab%fqNFc1TejK=3pc2cQlvdMQ`p1m`{XWK5A6_3gJ z30_`ackbRb?XBG{IyqE<{pSrM6;6&2l2dxJ`PfjK%i*u-&0&xvh~^e6;SC z?CiK&!}E`<)kTa#GJ@~kU9e1cw7)5)N0``1B!;LRp7#de-jO5)WHlm@6%yjr(g1?oN-tG`-x9pBp~I4OHRwR}%e`N@wzU{;sVSat zNzABWx{{z@x2J9X8GL`=mG^7gRA<`1)w)YlM&=1hpzif#J&t2l($*N|!N_5AlN`0f z(seI!GIx4u*p|U+4ZT|>UHIX{S*2*zvIN}zsCZ&gYAa8?Rix?A|46Uru!A zhH<~V;9tir$A0|^gO@XQCH8xkOEKTTFTQ@4#aMb5wHls{ivKv|)5P@jSaY^2`{#6!|i+NY#FXk@z;3wMf%L=UZ#mko;$lA%UzoqnKlh}Z$=MUfF?2czq9=X@(Z!>aU z6xVgJ8#7lgu^Wrw;FLfO zob2KLrVCS3Q^Lky{++^6+l_=0vBOzY}%fwa1qU(37ZRG#Jnc52bZw2;io z2!WBPu;8AFf(@USBGFv~UV}61QY1C2b8gSySnr;f@-1<`PT>XN+8B+)GyDU;IG5so z_Uw_BK0Il1`y07eP0Mw(v-7GiN-@g^+LBUsZx?!e0n^)-LXL%e=r=ll*Mny_{;SPp z|BZGD1bEU~`H#vl5#bke9caZ(JN6E|S8X0WcizKHm3=6yzt=jz;xFYVRIMWtv@}$g zxElP-`W&q@en_T_q|=ZsD*feOfb{WjvYFmXVyFfHCkDwe_^MW%tcoLCPdE;3X5e=A6vsa(nTE+ zvu)uvM%&nST<`h}bvw!Zo@-%+EzRrnk!5r>fIOf+N5dMsWS0gwTkIE-&NbF%yINDY z6KyB0+ALrf?DhdFm{`}FD4 zVE0OyEwg?qsSH@_;=D;zh4=z-J)aYciCc8!#01(H_*9mO8` zZ6h@-FtrhIkZYO_T(6TQQGZK;0sgZ>k8SHf%7=7rE8-G{LuBWJpdnVaE7Ys>bvbZT z?$MP46JZ&Wdv^Zx+tpXQ^W*G1-Xjs^NO=CeIR9j2ey98Qd24K$YPb*z!Ae;6)2 z;z1BT#?+e+H`bYRwcBudC?KA4ZtL0TMQ@vyeh=rI7?L1Ft{avV=DmBjx_+mK$OTAW zgR*qJx$$BI^)o*ru5#q!xjUu;4bZE5JA1vA7?vgMDxBW(^nt7k?RJ*tExnDYuYXu% z+fGsud8C1PPSzP?8@S$OWoy%Qr=rPgCMe5a8Jq1xmg@l*H<3=kg5wSMozZd zebv!wk;sYDWLvUqw!g9yzX)xLC0buSz?vaicFQ=$ey2t6m|1xwD@#f-oPH0WTu`9i z(ZON8OT)BPDeMjj=hpAnqtfb;uZJLonr1per48wcWVX1K5J2p&E+sM6P6X4RhapsM z_ef1YW@9pA#0}H?4z&G346-f3_rsM6I``>EF*slO7|RsyO1|NHNR=A@YCEAWV%}eg zbTLx&<3q+z!>@W`Rs2rSv>*|Ghv9?}1jf`dEXrDcx8^zyY3)NwfRm5DMI@a^rFu+8gqYU}k6z(wb7aqNVlJnJi5rA)06lj_@K*}j67 zy153HeXx%`QLA7%#L%qiH)^7~o6^m*#N6{l?dhHHx8okyBk+Fo!vk8X6&aR46QFQR zOKnvE;mi3a+5Z?Kcc3xU4%)Jtf`M_dw z2etDKs}1|E2B0h35g{RFAcfh^*?Yj=nFAukiCWvs6A!5!rF)(sm^M|biPt{u$h>MM zVtEG7@~v{A4YKUxkoZy26d^}3I8M%Lb$T_?Q5cvA?GjH(Ny&1eFW3{4Em^!{bOJ#L zNA6o z#2wIUN=sa@bZx-r&zb~Y(6Ai~-u*fdt9cxxrk^|h{xBh?N;`Ar(Mgw0J7_>me4q@X zakT*jK4)RUJgLlo{dE;)ORq5{?lFhZOGIV7+qTi@;7ZB;T zgtN-ZPk|0}#&)J{BnSrgqS*M?ulL)Gn?$Lm#(z2h%0d;Y8kg*7*o!?VIFG+w3lu`D zVLiN45*R_6QR|)GzkgT0IL1QkDtHn`W^0&$>xsN#Ik&QAfwF8Jai;_x_AHB~V(Nf? zGypDGOiCnaG}_?LpTup)DIcPqX~j0y`nlL+yGqurRG6H=U%xD7L2X|V^J+}XrB~hx z?u>Z?mO>p6-Eja6_T8hTr>dBhuck}X6dQn5%+jDeE2~Zed|0oRu*ul%q zeF7sSi~YHe;4lzOuNlrDI1D)cGq70rZk-vfDFxhK2D_a?k^9eYQOVBuTu$YGur6FV z!uTVwAF3OSKTp9qpZV7rgl#kb`me_kNDO5M9V=F?3PfP7b z+k6ljV4bL6zWHqEB7xM`Yjo{h&oNx!_uFIUmP7MT^iQP&J_FfeqYxl&iwHHeHjE5FTPb&R5(+%Z`pEZH-;*D32rrE{Q2)*lb_{qt7cf7^S7Hy z#bNylz#Jc7J-D0lXe}vMyILE2F#@lBQ8Nmx1j&c)`ou`F-`|gs$ZsHPBKq+rQ9e66 zJ3xO@h|?5QRGw420$s$)4U(G_KW7qE2s`M2(u2q`WQW4sCHiYPIj_I^Q5krOmDski zdVSmnyWxQTC*U}QTo)W1T-4BT!PeH6Y-{Kj1|ujfZfX*3N#v38kpa}iO#@s~t>q#b zGTLIN0k}g+Qql=0U?`q&4KC>5j~~Xr&hLbRIB^YO(ma41eflIrd`%LjfS@GT`LS%B z;gGqVC;Qk=|L5t(mkkU#`4P`ikGz1H?kL*svJV-T)5Fka?Tt9Yd3_4W0$dSIed0+zlM|yoxbveD{?Eo};DZ{&kYQ6pMDPrDO1b@~d{F#U)0)db z1c^&eMG47Ogrm$3N(i=}AF_-Gy=%}{#A+xEYbXMV0>Y=Xf|PPcU!Pk0PxY)@=Rb-~ zyu!Nw_T55OmaxGHM=kiYS2y&l`Hk-0#b{sct&J159(k491_u)mLlvew5n&cV_O56A zZ~(5T-un?-;5F)_D3p*SaHijW`5h2J6@Rp8i6O*GYr;C@J0MX6x z5X^b>3a*o#wQ*W>DkmqWDcqHd7cUMD4wi#jnFw$QQSxc}NPkl#0X6W|T|*PZVjR=F zQWL4POKYSW3xmKr-MFcqPT=q`bcn!_Fb)aux4E}&)qDSR_dPM`6MQ`_Hb<&sG-8Oh z0uX%mhbIUuNKns?YhKTvN3d+N5C?wZN=2A~V!0t1q~LIbu%YP$B0v&@tWT5$6SZ(l zi`7#f`^ckq@7_K5{re|h^X72$5M)EX0|%-X9x(5#4@RKf4ZUfjlo2q(tq?AIJbeoC z5YAD;rOwF_$12x_xq`sKj4Mbm=zX|Qe&wV-6f2Qchj5Py!DHd@Jfdj9*o7C)u00y+w2!l8b8M1b zDcO|W^e%~mq%A7FZ}aX*eRsd)~j^Ki_}$boO+n`Tgd3zTeO7`99USx#Qpf*3>0fbLEoocU6lR zZ*tl1J+;=ZRXM2RrIZ5}NTnY}+-^_?x=rP9K`P`wWmkOA9$IG)CQXjisnC!Hyl$7M z2J57{zL1E!^j_f%*#rmQ*H!(=vw=%6NVOe)+W_O#Nvf}ylm)NB-}?k)KN5&+Ey*{h_rF;%m90Zm7 z)8vqqWYFi4PJr4-mW8o*V^2vq?u<)8#sIezFJG?j+Kj=ur-b4dv~^6w{?_z9{u$VW zOdDiYKl{bSBhQfp6Z{{_WODeBQk)iiy7-%*iQNzbQU2CYOqMFx_rHi>K~Ob?asm#q z2ge11i-JK^Is6M3mSQ0wAi6s)Huf3d!K+5Sb?X>NM+4bDt|PDN;RtrqXNz$h4&rJX zpv6fF`8T2T$2p6RVqH+d7&(cxbiNO|f(U*PXDuFR2g>l?a*iQ37p@B0fv#of8S(9I z_Uw?@-NVB;eC8CsJL)&hZ%4O?i1G~??z(%te2QT+WLbuddk2p_^&&btdH}o60iS;v zM<*x~3@O~dd}(vNQ0M}SfD89=^s4%npxYE0O-WhN;j;-BnYU=BNQgc4Jz5=6TKRda zhS}*ixXe5qy+B(@NeO0B8W2TKJ{9yG#owVmp|pkz#gegszNLp;%0oNk6|pcf_Fg^s z$2f&~=Wxlu`j0k0@f&O-`$L^l2cxAB<8t(9EWVAOT5q9ja@eQ#KvNh##so3+hUYQW zT9toA$s$3-A|ou-i~9^>e9+YB5m3d_+TkoPW=14VU)=nvGhv0njm8tywE|{vVTnqK z1_$!r!-Q2?+O`=gQ_lUFrV*1iLocjp=DpPBp)v^3Bx+}8j$>YRZ?S?x&LAvU@`Nbn zjBnCJi{p*^I}&cAVDKZ~4X5>M=0l(Lf>K!_bQeBmP^+9uM1>LI+gW>idq=4Qp!Mh| zbT=WcQW=Z!c>lypIlyTKvoQ8)&#j#RCtKoXYyMMurx4lfXs!sXQhBJ|%c%qT2}=MH z4j*kyUGS7j4RqEezGAH~O-;=uy=U=6fRo!0pWwA0T4!MJ%xxr463;)|jA2nx;BZwnOx(poxg{z)6hLjl&;+)aR5z_!}sE{GirPG)ooBNJca`>d9%1> zq%5Em>FAxE%L<)%YhzN`6W3x~;4hr-gPY}zazQ_yfv7I{*vjjsWS?%HQZw8Pg*%CfegcB1PK;tzvHENSZ8ay6PE;)%wT1cMg z`mI}Dc~MrN1LR}Hber#@C#4wk*jP}X2uEPZP>!A(%yurmK_bbAlIwC^KLr9V?QC!n z{McGJTF(&#KaYY>=RGyrAD5WufV#GAcFZMYp3qRFYnm4)nZgYMB6LZ(%L9!N$FDdA&o|yGc8n z=)21H%60EU^qIKP#tM0ovDe$@4{$bO)b99Thb@-&)C@^sUg6;yWDaH!*nr zNDt;y-=jnm+?WRQzBgvIJ@!HipLsE~Dk+RNqTV|^tdB;4QtVRXui1U}yW5HY|bkAXE{wyJefSP~x>&UFWAT0z!mK~95a4U>R?_olLH)m>;k zY95>LfY^%0!N(TLRJK^{M7qI}%x7e0?pY{&0e$?-<-gV21`OWI<#0IkY%!l#9b2F) zKoO^>M9hYxptt%ZZaF~`L5qnW>^VAb2BcLhbdB}?tryKAFIbouLGXRtuuV`4{Tlw= zhWyz=(YE~sqccLKl^bg}a{&iNjHD?7aF$SUSy>sm)}di<8hP##PEwLvd{mt3(xuk{ z=#D{VoyJI!acKD~UPBm5JI+u8OQVahWm1Um#{0XW71ii@b1z;@M*&&ET3-0PXS~*l zQRLgXNt#zf z9NGc5_N`_X8zo{)y~AM%ddk1rN1iq`>zAc-sS{)32?N~gq2o$?|!mAD;q|E>gfvn_xqZ-WgI>z9*&Ytmq-H3z{Clb zuiEUt18YA>{@a|(T&~}_Bn=IX{*raeWn}2^KAx%-kp8?epWrrbu+Z-Q*$TUbu0KP6 z_cNWK3A|ISTpeu_6BB}n959z7Pn@;At%D^VWcHy{^Q#d~M~Wn12GnWmliK{dp#5J4tpLWYa&x&+ z^~0iLwHGvxjLy(}bf6@0?gPRAqlF>}&!k+va|G#BDD^^@9!J$%4~^}LKC2bM3L;|0 zzAx-RSx`8nk)nYnEvBI=>q>Tn`TJ@xoWDbswFa(#_39PHh=|MMm!)EWrJN4lOCX*s zZ{EEEot?YphK@t&mAZEDd$pKQKbukP!4VTOJZ_$oG+2`v-@^(IKX>r!V0bJa=v(WnGYY zC<)jG`Ijg>1x)Fb7YrGg>NXHP%g=y+;9{UILyIqd{{e@kD7-`m$SP(h(!e7zoQnZq zW5DX9CU87LlJ+wem*1=`;&)OZ?zu=P_>EpQ!G#y4?0cxUSU@`Z`uf1%nUIg@sPTR{ z6oz50EEWsPzr1lgXRx%iv>EG}iZv&0gSwzU*@<*bvS+D^(vJebD^*hz759U25GaL@ zZH?8HjJAtRK>?KXtEOxlywRI;TzK1AfTMOqcfpN_0LRBKCW(iXSrV2|%4jg#&G3YL zJL?JwLK_$xHhb+O?;=_{KHS~yX8r-Cp58Zg+w$qR>*1U4v#_+h1jV|dq5_G8Oz7m^ zIO7b)U&5PnPK1hZ*R|{*?NnUlbe%$KP=B%PWMyUH9;BY07|euX)aB3IAR|-nnRhvO zyXQyY?)oPR D08#rA literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/observation-prompt-v-latency.png b/blogs/deepspeed-fastgen/assets/images/observation-prompt-v-latency.png new file mode 100644 index 0000000000000000000000000000000000000000..7c14e2bf6e53c7cacc37e92f1be796a089923751 GIT binary patch literal 91276 zcmeFZbx@UE`!>7*0cns%5Cs7dknS=Fky3Ia4GJRN-O?r9A>G~GA*FOTZ0YWL*XDk1 zpLyTkH}lQBe|>*^d*&Jkbzj%D)>+4Koae&ty^J^(IvF|y0>P4aDc4%u`k@`HvB4kzJdDbKQf(Wx3dsT#h? z{7E10*RyqTUFmuk9NbL1czDq@nHa0zSQhSfxk&GA5JyY(@8>vzW(3*y-w$~pA|u{^ zJ~`tAT7Liehqn!x5J{2$d^R|dq5pneu86+*!GFFG>L85}_3xcV4AG1K{X)8iyc^BG z-#Cu@{~z&xu~9}pUcB6BGMTOYpsp_c`t|E<$$xi!*=W2_orfHoh!KjrA?GoIB#GU# zdAxLZ&LYh|lF`ko7e}j6q&!#|I&^4Ue`NHKTX4QpaPo5$X@fqd|9 z)3dVT6^-|D&pUZ`Eu&fUT3{+FC=}~-pih+>r0GxJUq9#4aHtH5PcmK3(4VgU`{VA- z@#Z-52+h_T!T%4MUJt_>!Qy~b8b5hZ{4wr*YkE4tqVb1UMIEUEm+u*gFl_25+E`wz z``H9;#NED7sn4h(^qB zBOoFYd;J;_ybKL9b4Xwy=5hN!fyg_+q4zepZfQwp%cicC!GHs^?J*4va^0+qxw(0E z?>}y1a73WhWBxgg7QCv6nAr3BFtH96Oe$(>dM2h0j{f)2)OYbZluh3pw)_)zxOjSr zhJk@$O;6R8(6!vMd%Aph{_keJO#{gxIPc|SNDARj`iCiY(}|>H%YUB3D-`j(8|}xU z=Mc#{?Bm_aR@&eH-(-U>-$59LHQImDNKhCPNL%FHc}0ys4T~CVe(7SiHLqIWQBunG zMltao_OaI-4N02wkke(q%g)VRsuF`%R!dXNQ-Y*tX0jO!r7v3qA>|Yn z>ivwMmx^WcqLGX$c04tCCn@O{82FTnE2`3X+Tvn8Yq-i(v$d@aXKGAEKd~c5eak~Ti*4kLwke^1>Ol2IT_!CBB1y(E%#o9y|LJh8-)OErmeTbq)OOxE zI$FcQPbQI%;&ijW%d{Sf==|GSe>~-QEm)J7hQ^$E zgIIPT1%H&!BZvFL90+gUy<3Wqm-%5ZoLQ~*;yDAuWVJbUnO=Wtaq)NXYTh`M1GUzh zc3UM~uhkm<2?4*;^SEUD?dc-YivrI_@j4UG70{o~z*ZNig5soINaYA^qr$54PZ+;+J)^1IAUeor0$` z_)fc;w-?d06~ypchlaPNs>jzX&bpc{9&Xxyks~ole)V~n#YSktD;s0ltF~V34MG&h z^U2A{PkMSNny|Gs@4^@qqp^IImTgJqhNI$vn$;*(-@yFD1hdnfSw{8RBfE%yO0gVs zOlD7$Yiw&a-8_uB=nag?<|{c0KeJTUR?Zv;KSSad+s7lHxyZ;~KMNX2!bQG6FmLr( zGKz6=+O+9&iwCmNbY)O#D%F+^|A!AB%1vhD8l5jKCO=1Sm*p9j*N_qD^0i8 z%;9vhdc&F0Ap3=dh5Z@LBlbkWqT=Vj{BBGx1s-QsS6X^dJOF~tYI7`z*K!S)fx;TAwwo)M@UHDV2q+R>i>K^*eUCxjIykV<(b4%bUCE@M?97FP zg4t|RMY+%Uk8j&3M;**`ht&LOLRPDt#d>K_W`>Sh^|Vl&tfZs&+3gjQk^YR7U9{wD zHREM?yb%Kb$drzUuwZ&B6@1)2tu8HPd*pod!R2hx=eiQSX$yFGjDS6xY11_q{OXV6 zx_P=Dlr;o!b_fkkAK03E}K)ImG_Y4u zL1MJbu^SPG^odsNK3^Se6Y|a1{14IuS`^gi#pxf;|D=U@vMI1hENxoKksI}R4 zP)OEI7IHq@Yr{c+?>s1Pi=7WmpE59ProbqS?J8i00udk^@6FoIBWP88tT|f>IOSbi zjn-UCw4Fn19O&u!0czhjAD=SoO(j~{q`rYf{>fr(Bu`YFwdD}uOzX`tamA{;ZA}nx zx*?k*zowzMiR|6A8SA7+W@>##HWKDrn8R^u{9~&@cwzPkW3+lQ$wE-QmcqFUF-H<% z?8EaHFZ5QrL#2`gV!^WIL8TTExmWP9VyXUcVELezd0u~F@)6k8<%`WC;rz{`o!NMh z;BA{OrG{7HO&b@%&S-V`6Doij{{8z)5b-r9Q-2sSmmeb8S zERc*7-ECw)07gA!%;TnrUz_sh_+LuHUJm7!Jo9qF{0BQxyy)B_-mcH@8!V_K*6;3RLZP z8%{lnC>{R%(9){z3K6~`!7NHmOFQ4xysaH%)hLRN)^v%zy*X`)h>o@#6g=JQ=bk6u zJ5{`^Y@eI-cM>*IvlWJx0=UyVP%wFfy*{*`Bxr_KBMan=@c5hbQPV%u55ckVH zbjeSjJXv%z-rk?k4cX?dN37m!#ekRS7(VOg5P&~0P@v*V+g0<}Cc=BoNSEJ~A*7_F z^#?y0)Eu`fAWi^JTw)bN26&b+lUy#=o#$+8)C7)4QU>F?H${xo{pHez*l>pz5jO+leE9&GV0QOTCL?MWxy`Kz4-yQ7_%{U|3`vzDa_P@Vq)8aMuR922`nG8Gfu`j-I~0^!)t5y}do>;5ty6u_wepoims!r;q0{P6hu0>YcO8 z4YqQX$t(%C2|0Mr2g{v7X_=V-ah&wk9RpXu#lveERo3jMu$?KHuMLSZh!Z&Irp||! zAi7~w@a=(&w`Dog>V@H%C$qM)vc+9LH4XkN-WA0K$Qp$IFzmM!pJ`UF_Rj7*TtWRu z%jLFMC!Hl|=gQdMetGa}XeuYd<(%VYDAB0=^poC*v%;Anot5b>si!X_~C-u2Sr z^GO&ls8N1z)})5ixO@ZddTa z<~1^p51Zv$--1n%Mguml)HcXc!hc)>I!XtxKHf422?_CpI@{Zye?WygPC2Y+M8od~ z>NX|(o`)up?%A-n-jXF9y%BWEv6BkEj#C09|?2bkhUv(wtI9O0+5s|pHl%8O3 zdGcvh`lrR_p>$OZzUMF{h!neLnM$c&@SEZa7f)}s&!vBTg2uD^t$qV^CAZ{PRRwji z&Gytn^vcck|kj2O?VFRCG1Bro;1mggVenyfCsxkn9NH1+{PCks5=s$W4 zl5^RhP~Bkkd44`)lHC&8BfA9ze}8`x&aWdvZ3Sd;TU%S#1I!gwFCRA(YdW>;Ew<(4 z8KB_t)a@5VLS#a(k%77+Lds30bYnOB}DpbkMxbdbRo>+j6e^d<5~9~>Mkl)yt4 z?69Dw^D*K6{(jSXEkFlMruF+M5SzM0*PFxSr!+L%rxSur$`DGIGi|d67i}LD6-8gY zYI&vXN^pmlI|nt_KA+%I8wk0e5@Fi-pibBs$g4S*nbG?Ewz`zdy*cFhTsKxWFc&`= z`X(E_niccQG@n2Fps{#3?FSz6u{WAh!M=$3)GUrCTMH*G-mAQcG8ZS?y;;dFWSHV7 z`E!*9!?e$zKX;pY1QpK(Bm(3`1T&CbcWr~2+<=?9!KoCWDUG)`mnf846xabBt&z$a z65wO@WesfOpfvZah?SL9qHbPZ9w~sOX<|V-#~Y)lM#En#ZZdJd?Ai~xNXljlwzjsq zSUzlu?9ln4UT0HoI3E1av_5oW<0~9_G~FB*Y3;Uxgm(xC8i@W7@?+~LMwM2|+q15P zO}b>KgB1J*klrYdD^3!M{!!9aRw^tv&~30Tqlml^nzi)AJ+oV8OR9W-_@Sm|HAzg4 z*k<4V#>#$x#`Ur86(qHlDPfd3;6$D6e#rZxyk${@F}z|Z<0pOnGUI73HgI1t7_A-R7K@#Xeji|6Q6gQG3j21g>Mk0QYMiE(6QEuN@{Q-(zV z7rT#deg+O|?M=A@u&=5~jaMXt@)+-+PRaMUxOafEQSezU02-*>8Aw`xbFnF|9xw>w z->`g$?$g#%151*eZ*&HKIKr_t1bECSh+=-5t%U*p-SpgCbT>fJ*1lURY*TO~+0NOE z*Q3MS*W&U2@sa-QZ0*@%X20ZOFCHz`^yFk_>6A$oijwx4ZpJxCvY_aCr`HPK1smN4 zS8o55hBC#Jh{c-yw#2*ORug5g6B1x2s@LKS;TQJiM1v3e_Uxtp?wpXLH?9dNa)3$e zRDA5|?gk>rs@wKXY&sb>h1)i1DbU68l#7IN36IFALb~2KI5@DR{Os+`E4yt$rNrC` z0CXDw(v|XI8ESrh$K=HrVNfW`V7VIv>f!rfyBcomf}c5kQsFk!S|ZJi2zt^QK?2(S~BrfN;QMK360lS`lDAl?xO&nJ48rU9Gj zT>ZkP;2Z8!o9qbZUEX-T{y6LI^YX@<3q!FWvJQr#`UhL?R;kw8Q=HY;03>XkHsYkG zq&z0k=;-L6Fy}b}wI5Y^!m4ie3SeQrEiNFU5X~g@Y$GQtEIhoe*&Qk5 zYox@TcNo~^(yo=BdvKSyHbp^bldTGLI)AMtI2cfp|3)dW$kKAXbg(J-*jcqG%g(+f z<5cY%RLZh9q^IU&Eh`t-`MjR51+Ye4YVqJtGu7rVjd7BJ0K{1hR_g%_hpMUn=@VQQ z0XP?vh*_f#aMyn8adl!$Q4Qx~MX``a>RH!rZ+Urn+s5iY?FKC6loSv3)8HR}l9N994+Kph51 z;mYL}uV}5xc93$I184K9C!7vd`SHDKz#h!`5?d@!leo7)0o1uV+dtZ!Z<2WkkWD?P zZ5Qf9-W5z{|QiwboYZwSq+%slx5>rY_OTz%` zWv(m)UESG!MJ(#ocaIRtPlc<>bW-7XV)Z; ziu5?Lwof7{z!|+8e{7jTlZVoeE$Ez&q|h&m(|D_Yw*Wzz!)m+NbKOe*b7KY%2k3P#WK$UfJTY zL&#Zu!<(wnJmteNRa1(+>eklRfbC`RF<@{3*%DOQ44K`sgaq=^(o$%h4O6Cp1+dev zGJRRlXYp)X)Y=>?kV+O3RJcqM|wA-9sh1;vlPdPjBl1@Gcd{K|f*B6-)t_AWo+Z2RQoQoMcY( zzXd+$GkUq>nmh>S_W;dy0Ovw)hVy40v&Vq8uB@!!Qc{Ktucw2R16WY{<#*}=jNs9u z0ieCVgF+{S#^?B(r^?`Y1SkjWXTtYlV`JO9y4ZEv?*p00{bkC4{q1Ig7KcmWq156-dCJT{1)KnmIh-Y{5441qL%1igv9Sf`MyHN%W@Kir6|Zq) zWTaisMqbGNZ_O)sFH5_Wz~0ZnlqSjR9O^!EMs{PNK{NZYGX1j^PsX%mj|wZ%>47<8 zxn}oZsOSDe?EEN5zr6FOihvHH&!q;#jl;JbdLW8?0BTDG92Nzh!a%u5_GfMmh-5$EBE<2yCeFyogNu2`FY}Fp05UMDRpSnGoNtv4LjFuwF$3IXvRA#> z=7Ud6+yyM~Mp_yxfS6ftf2reAtqdrD^`H}is#Y_|68Od?*zuH$Zv-ERBP-Vb!0EfnGbOD47=v2!Mr*nWDdxKPGuecK(6Dti{8Vv~0R)^!qn}AH=>!B4`ue1%s!RvZVac*tGKtuBG9dgkU9UMo zLU;mlgANEZkjj^A>Of-&&3LMO(F5bf0#E?jSuo?cP2ruA3820LMmD8yk_)nPJW$tn zL1_Ac)C#(VfZ|QUTmZRqdC_nrCl*ZJldn<+j|xRa#rDq5p&Cnl6l`*mtqSYnOTiIX z_37YT(%{X@lOqy=_jSbTR(Nm?9WK@bn9iu#m;hW)PDu%6q%bo!B0^*^Nst$oVEJXd za8`S~66WA@n1Hk9y8HWU#X&O?j@hW9GF;GOH5_9E1n?OHL)-HH_t@ATgRI0(pl?XO zd-v0{wsQ1qhfc|YO%x#}CSDDNRd#hGyL zvpAjYE#NX_Q*iNso|W!F3J%AQqd7ro5Ggjl6pv1JMh{e*vVB^COxzDHXSJ!|z`zwi zJm8nWO36G3p;xEkA|iY3__FYeytDwIwSAx(>=$p&=nif~`YglEfLGutUJL z%UxBGrMx&dG&B_a2M9nw7P4P(N3{c2u>JI%dLVZjlPR&9-4RjGLHgXuu7K9`I9E%D z`Qa%{O{Ba_bRt;of$J6Bv?VutKYIeADmx@h7wL$fJ!os7U|-}}c76yzV_tyY19;O1 zBpT1%S9@PVW*v|^7>E2Ra> z3j{QIjFTh(oK)(2abzbh>xY062Fkloa!8FO%mPH)&Z$$={NUC3q5D+N@+lBN{^{8c zk;EW``-c$AU5w|vS9}xMzoMC?yr)=2w)17)C)kf3YHRw`xwFnlvp&oqNt>X~P@IiZ znq_^Iov27K>gfR<4N9P!(mYY7<{6fHPw1~}}JTpsjbzdw27fnPjt zH7pHA2S$0<0LQ%73_E1C`txhy+ms`Scv`4F&_tCr>{qqN|Fd^;$+0(()!lyG(~R4e zDl?2on{eWbX!FZa8mqzeW_dy8R>j(YqjgjHlMY8G4QF9V^=LQkC)D)KDUuI(3}4DF zv9phE<|6#YZKd7=?Dk$;YD5G9@a-5L^Q9+)hM@fh`Y)it3*=}6&?3jaauH^3Jfi}A zhy}1D0H!^F&%w(gkklZ5ey5?@&X{)zZ6AY<2hi=K_m{iM%obn6i{nCzCp`IqItK&o zhadIzsa60w-v8wD zQhXo%pP%mr%^)*2!dW)u!^h=H?U~$NU-{2AcI0lKHS;0rb_SxEn9kNl*6q}877sWA zZ>YXH=tcN5oP|?Ve}E49Dro?bfIbFXb#xx!-OMjL>(#vRXj_~xjfXF8b#=8bhSd`^ z-!)v$SNSfn+O)dAKLM)PA2U?!d zyt0?{i@$nwSXoIke$E2X#-_}&E+4KD=PDs(Dt?sW>C1tMUW_eoWW<=KRKTaVO+!m- zaeIAg2`)Wy2X9f$*Xa5e3g*s$7>~r~3 zUSi))IYEfZJ+!Y6=D=dWyq<4~iSW2}F+*7&m7nde zA6xmE>6_pue}~aP4_#5{oo1p$2P(#6EjbFk)tZ}(KtmP3-mkh}e1xn5d|T8GkZvT{ z&!${S6@#B)B9_bKWZH&28=9)@AWZhqCi&4w!R!5Z>g{c};%$QXzV7_;JoJ9^(232? zcA2M6h=kH7L>!wPpE@h<%KhD9Gf~8c+uR<=spP9i8(Zn>SyDRjRh=8?INL$p=0E;U z&Aq05|3P|vK1=@m_n_8m>2Hb30|hz?WmAiu(OkRj3xPHYs{t}+})FH zav~GP1mhyc3e4%!b)k$;8Xxj%{P*47+QE#{&-`}LjOY=xoMm{bk{i98^sz#*iE*qp z#4AUe`S+=#n1UYeEbbL4GXulk|yCwA}ucb-os`(}EGD<6!HT@_AaXox{|VlS-)FmWnRy?|2~Z4q}I5L_LpRE^{=3R(tLF#t=(Vkg7B! z{Hg6d99zDCl~N}zCk}5KQT3dX>f$&JzO1$3G*$cuWpT?ULx=&hq)T<7`2U?rBv)1x zUXx>ULUg}EXGz&O;@H`(2s-Dqha`lC^=~ur$)5}}t_^ZcN-(oLo{Z}hgzr;+-_l2*&B>X^P^ChhsgdXCeZg8Ub3=P`W>de+)iQmEH7@Q* z*MTzrXn2Cg`!^aaq(<_4vyLia?YfE(@mZSZID@c=Gt82CeU3A{AedjqLJjfRpuBq6 zc~$xkSeDFUK3mB`-*m}OfB!y5VEffDf2aG`-1m`(|9wO5oJr{0$<4BCvMNMGTFA)& z{XLY*#qDZ=xw$i>xtCa>HpHj4Jf7uDUQ+h1grYY@S}$vBzGVk?{j_^-(|TL;-;&7j*Wh ze5A*%J1gE>J1YdLvHUuM8JyD1ZhAL0332!`!C%E)pk{ zg-5u7$um5z?83)dYKj24eF2-m?CY7-)bYBs)JJ(W`ht=>&wINoy}=C(Ytbs5$QOx9 z>GT~)2JfT)fmJ?F-a}ta)*WV_**#Ghg=iT+3b@RbfnIRwTJ?Or;X4(yQHnzR^`a83 zigWOXo`g_%A* zk(!tH%>Y+Byt&-AE~eX$#qq}V+s28=4CU+wrEOtHK?M2f7Ya%e8>m~LyM!+(gPC4U+aKM0Ko zq!vPQbfP7)73toQB@==h^EG_x0ItIUxoSi1&Xy~;v7?*{E+yYU9j7hMBZo*PppIv3 zo2kzsIKCr@&y%#X8Pj*FGN-3qD4(k4D`Tgqbbt!wk{8XRy(~2)6-)^4{xhE3o!Ma4 zp0V)ui#Lsc3%Zi^vc>E4bf2e9*vH}-VfqVHLASa^~_oWj`IYxPbvSmop!b&tEqOVG*`emTnBOMf}ecxB%KAqkaH z0}pgoq}ww19rJ==-dJtm5TjA|jlp&~Wk1zY^^;sdud?3rnqo4uWpsjY`D@ zL;e|7v{u$PsNnCVy$CgTJ*;I{`u!ZAsEeMRR;zw^350?`xYoy$R z^l_a7x_YZC)p>re@7w;?NAVuiCA?$BF~8=rIULZ7SJL%JP|}#IuhS`Blia`eTHX6@ zTgg!~4VDsXgG*fpQuF}Q<(}nI)30iGmc~BGw`0`CM$~irP$Rs*G`Kve@d5GWboNse zh00Hx>0igr5%SZ|h<+!QaBMqsg{qt`1xj}&{USH~l)6?EhW{FjOCN#JUET3Bqq{9M zIFlMqcNjzYdBcl8A7)%(%4f`TS=DR_9u(yVZT;GG%Q#gLyLi~BeRC_D^I&t2WnwOL z1DWM<45Oz|v5a`f#K zhb`{#vYx=cba98iPKadJl9|s2(dyH`m?|A1@8&bc=i7u=#|hFErW1@jmmFPSuY$3q zJz5ll<_CQhEzhT!@#&A0AisvgB~I%J4v1Br-80f7q{!|3#cG3O4#T1q;6+k3;yd{! zm7kQ2JNv`;Utko9(o>#hs55JlDUcHjFF|BVfcESSYcqEL!fV(N$4; z&!5lIPt@O$Vf;)*=ey}g7Aaw-JlCgPLXqiZC=ce09tX!LZ29#Ui#^pM`FQj0wdB9( z<}K2aYDIXdvi?$Hc_ti;ERSH$jIS(g#d*^V5enF-+OrP!QFcunOR}(w$nH$u|DUMZ4kRt+7|29j#v5O(ZTyI1>WFA!jsIB`<#BuKnKwXRjcVO#Eq3Vak%vW7XK%5?eHByn z!0oB#Qy5YC4?Um6=glTczj8J#r|P0{E$5EE=~uQ9CBmaGsuF`*9aO$`yH z-1_q89S^e6AXUMEw#zJC3>;_dAAMy3)-|r5 zAD+Alu1<=^n_PRo#oF^v#_TMHzP8Sz@_|F)Z+-Mx39VpW=RcACwqapjQS5x$G6w|U z9a52KXU}FYyqP5Vy~EcMX%Oes;KDk{bQV>x+5tr(8elXQ{is2X(1eWNot^ZCd$;SU z^_GOfKpBrtSP30h*Q0oEZOd36P@^$jd}yY$0W4n4AM0Es(c8h+xNE7IP=y&oC_I{?u(o{-P4!jJE_J5 zc@dhR{iIqlbYASQyWYR&pZCnwRfv%C=9|hQQu7pJrH#eOp{jNeR$K81R^hw)$ojLx zdqN*MqsiD`zEk1Q-!k&b=#zomKiy4nTy^d7pOP4v$=%ubl%2&?H`P)H>!bFfva~`>JOqF6j&+LwlVc_Q-J|8cBh)>nId9t`vgbmY zfhuaqZZw<^1|t*A)Tgiwoz8?>2%Ba3xewYCic}JskgEn;V(X^9m+}*_6a&oR(3BCr z%b-)M_VN>~X~9>f6CQTCXx{xU9=i%FXa8LXc}%2GeHqt6VtrI@k0A>NgXbk<4_h!qTf|V-An6=6`(Gy!Iz#s_v zPGps5Qd{QhQ1Tson`A0FkH4$YZEvb0-;JnHkpB4QcnfRc__lV~xNGhSss13QT6B3w z96`zaLT=|SEOC|nyem~ALQy$X@w?P`Vl33TLUR${P|56F$0dh^6|Qp&Qng%ZNk^ux zY$|NczcZP#r<2 zKrm7zx3S?1*EselPEmdd`k){4Cnt8fY7mi&s-cDhBsTR;KRy_(fXZyKc$Su|kvAum zc5)=TC>j3Wq9NX#kG7y%?UgHGxF_?`)7A1r;`LY{$(FjD`O!f8?*jUObT?x>%EaZ#^*?J%2IpfHoRs-5xI0y8 z)64c#b_FjznA(v_>)%me)HD7w+R7hWs7Q$yITaD3F}NUQ;x1T7e{v?pxZ|tqoJV#1 z*T3<;!pd@08uDyLoBe7=80Qh(Hg8W=AG8+3(ieDWkYJ?!B_egK0w>f-j$fsJ%Ou4d zaYY;0W zJc@CU9jC^htf*z!Q^b!|oGvaOwoZF-hB#`$7g$iXumiQNES>Isph1^>4mi|L)bl%)4bca&~=HZip|u zR`6*mN==Etq;0J%6#7UaS>0G7KRYpHmf~5uwo;HNVchtm zPuSQsB-ZboIpphX1)p5f}wt(o12S2F|a0&u*xBBDo9+9w zQNUkA&nQusX#{r=MJ~@J)z1_S6til=?w2&rH`E5*{zwze;VWLqBHBS(Hf^KYesA<@kbT zyrzCrhQz&-1cd~_Ve_ND_6OUo_btg!oRoxIvBE8s*{y$*Rhm3&QTamm^Pw0=s0jY~ zrnQmcQxbcQ;}=gs&SG`%C%JO9)3aVMnnTuizz!a<0gEr@6NJu$LaD=x9a34;@St0l z0|7KtK7T!LgJM|K({@0tUbwJ@|1*~REMe15ImtxnQt@u=V6HlwST#~Wm${vBJgwxo z>F3r~#e7Xw$T>>VS30dHp??<45%dYsqL|W^ucVaDIZ~bDH_CN`dYfnS9|o77@Pq+f zPE^a>8#qBT8IflPh2P>H`S12Us#4cGy@3AzW+B2WG8I+zH@3+vdn0NHM}hG z8zL!;Ueyj8w9UOhbFAV>a4s$84C;Ka-fl*IVOTIvsWrUzn)WuV_vcLFnGH(u_(&40 zMbXh%DpydkebVOFP*q@l7`+H_Vc~G_B4$_QB-k!}@*Lw5G1^g5qA%v6I_Bb9P;A#= zCOlq&yBCcCk{3A5mXVT@GH?kFLBP*GOqS^%f_dr#&~W-yZZM3iNDt250>#-0W`}GM|a}x(QA~Ja#xVxfoz43P1iz%Fh7&I%ArQ^}h?NuI0DjBEgWZ?w3 z4)PLhL9O^kTQ2E-fowfay^T;_ zw@Cg@_LMOe1_AF5C_~-mRTiu7;2kG$v=HIXY#ldfp;Pnm#evrK0T?a}i;tHyF?j(_ z46=jKFJ3Ui4LOHocy+n^r~JGYpa-=``S$o_36)mJK za*|MgIG9InT2414< zis|PckO!@zO|CFJ=uxi=XjPB2K6yx@Fj2pS&hP|xJtNYZUq6B=-QWgYEq+`##$6Mq za>_i@x;b_ZP5h&y(SCn1UzQ!f|6J$ieBL$cKykd|LSr>WRIw)U0opyO zN@FJN?YP-|Rr%&lEU$=Y<#B_y_Vizh1BJy8Y`h|I@f6!)M^#Cy9ciSw3fn)_XAJCn zw*|4M_1@xO7=;MFzzSf;H2g%RGkD`$UoDGN&r}e7Eq?JkN z28XyPoCLw49dOp@HT0j=GpybwNsmsfYO&sLs^dmB4m&&RuYni^4lIXCGlyw3mJRTp zd|Trg;$JqJLeT$+ny>vEUbDSBzCu&FCEz`Yn*)qkG<;eXt3U6rZl-E5O{y1BKzkl{ zDfiBQq1jh9$e)V_?loJg6Mr2r>A5r=Y=dDG3+8`!7cCm`RvuuPiJdj$Fb-iCM0GU! zIIHbmcpo#bvDQCf>nO`DVFa@+WxY6uZX37tuI6x#uqI@MtZGQS~~x`9k?ggEdyOM$M1c5-#uWIxBes>-VkPX4y};J|X%f{bgfX1BNjh5#~aLd#>pdXj&&yc4Vwv}uE}U+z~g1StK}p@PIfHyrBKGJby-)P zy@i%1h40`-3`QR2ptm=n8ZF~;YkSx3={zy@G;DA_4yQTy3`^kqDW7&DdOljeUy~J2 z;Yej%g1QKH{7pG-=(5StpWiLoo*>1+58Z=7R*rH}`1DXWtNi}rpn^;7ea!oH1$3Wh zhtC)KYwjV^##7mvE4FN=M{T$7wSPA|XRl_fgEuhMD}6a0N4+P0OcgW@&K<(X8Nq2! z`xQJdvO{%3U=>Sz3NG;GY%U!H+dnL#lfGMtF>mZowXjm{XVgnOvLMv*{&lp{}Xr)#&2sIHG*4f;$hfQ0P zeleDd2E6smfa8LA4-Z5zJxjR3c6T2Vb_QpHIo`PDU^Yg-MG=9qkQ#fMzW6dy{BRm+ zm&J~mTOEuuvrjfpPCR_fISjnCe=~AXSyWHSn!~e+`C*W|U)?pYNU}9*a5^Jq+kNYG zs-d@kpN)i0peScI4U3`C0BuTO=aoX*HI_S)rh=;jDBD7Z z$l7$U;CJx-a{mWBSbcrBqE*!2EPiJ_JM=GUuk|At3!(YUDFv1Z!H`orJX@3X&F$`b z5G!)koMoP2+p1&hU2TD9VDzuF(k%qq7-zD}``7izYFw?9eBlLti-BYnByOM4;%h3; zFydfE$aV2&lF99}sl=dL=gUAG+WGch4QMOvs-Q-}xq+M!WzjZC%wFa>wb+$ZvuP>) zSMKezD0c2aBut%wy5wRRm~Md7ed|fE_9)suEz50BPaT2*h zkVm*Rr6S`s+q<#CTVG_owXw zwp3r;6EHm1`VRQmw(7=0mdm)!Z7^tRVGYcf4#of6rD+NsAZ$#oxk`(q zLGMd~pXfbAFyIm?bUSwf{~GM%pMh@^2Sz(J(6#YV2BRlf#4rML1y|GM*F|Q_u0We7Nu_C0tg19|^G3s?`J zHZi-Qsik<7SL{c3mLx2#Q4-i3STcL>g0!4wec?FMRn;=A#_JC!7Ukx<*L|AFh*$f> z7U)5AcR5?FJPde0p6%_DD5t|snOe7Vzu;hfa8-Pr)oS&a^HHt!z{ zV&;4jp@6iFSIUjLkJZZk;yOVtn)Ltk9yuXmvm{+}yhMR=I#_Rm_v78af-wW;>bU^s zoS7JtoTT)0Kg8w=PBj8*YK>*`)?{4dgakJs=xvk5YrpER*x`i4Z~OMVRxR@FnTp1* zEAZ-C7$LUm4%upnSzZMq|5hgPA}^}_lYqmciYm{I?AFc0$)oB0ipFfcI0f#apfcrZhDgc&)E zS~W&R(&o>#d6aP*HaBI&e6I-~S>8meZCm<2sjL%W-5NHwg`a@8Ok_K>0ktddGGhV) z(1E`Od_(<)CvJz${rJ6&agy}-bypC0J}`>{01yjG7OaNn-60n(>o1aAB_YAAJDhgc zG?<1_N)dJhxW*_5=hHj>UUB#Cw6J`Iy%ptD=@0k{i=P0Ru~F;wy6 zKX`WW5Cx&<8TQf!1O|GZ28&V9?l*tDtw749Fa6)&pRsI_y3qCu#n)rzzZS#ha<+fP zZu8cch9y14ljpz7?I?2!CY8U0`PZHqsmTcFl7+85I>3?gMjk^X=-#@za3gKmi&FEX?(XA;|B4h`(38brP# zx2%3~VnbzZQ|-6QbS*MXG^YhaI@;?>9cV`kj_7G+LpGap>?l9J4JTS>yawV};`RbD z(Sr`+)eGdYpdjZ2i8xA-8Xv5(CYGA8QCXh4E8(4K;`!>i{a9;AW62XmqnEY~`^ zy1@1lG^9A4-*>{oxNvm-`+nwK$#ZbhZPzlb9)pJZ^+UMKD;m+yotK)^AbftRi|G?FjA&% zym~Rb&%9UfzZg+T1ZR`6u`w6|)j|4_{{OUMm7eUlG}cZ%7`V5_fqScl&+{EAJCU%8 zSML3~;5^zPrrf!Bu|%!xoI09#S?yDk@9b!b%hUxM;V)esORQaf9uK`#S9kp`%f%D} z5C6Ks3k!@x+#}8lJ{0EQ%0sgawklVvH%siX|AES=sJIVq2&>lQpz~wEgW;F`FXI)a z*|NK1wML2gWI+E)mO|wKJ`@S}oxnZGM6aF*I0*n!Tr@_XIyySKKN~%0Y-TCK^uL@* zegALHq}Bf2eHpdKi|J$TS~hA^5h)ZEoWC;%U%bG$W%YuSNtannC*R@GT|lR zz)aZgC**w%j~8AfCS+4OJJp=pr{mnb23rX_0vR~ko~_rAja0WIgN|3Dg5(-U(o+IS^@a}%Ii$-1y=k6%2I z8Lx||BCJxRQHW)oUu@8n7^xsN#hYVy4f6{=&lh8$O4h^LFZ?tw)l-^t!s~LC+yB%8 zOoX|=fOCb}nE4hRbLVWgXX{XVt9ooKiSgyMv%9J3`dzcf^w$ou?niz2kT>n2gq-JY z6n};YuWV7`4!VKn605fKo`T}KPH#gt#$^s6T?YG0JBHO~0;uGk8sF`1lG{~;p_d*C z#aa9Z67DCObKSc5ve_d%QV6x*XtA>>bQH4|8L8-+w^#qU<`BTj#)M30c&H0FY zX_)O={nu8C4l_KFMQ}%Q5^8F2bN>3kk>_HmyX=kH3sTk?1R;BRMn*hDOat9UJnUwy zX}%jvab$lk-l#hLB)VUNrt0$KRtGTwzbP1F9^JS)14(qe?nQJ zV+VO^F!%G#HhyZ4g%JOtVYyahK94Z1{8dXm~-_OTgbDg z=Y4Mo1?xfw3N-MD2nj}MaBm7;fvTbg%jTu%9QK?d5uo(gzzT)j*=Y=Z2a{!9r! z)J&0g(yDyiDuQGtAhKS^GkETrqGpUmUq zc~$Jk4Y`xxy8oG!2yMj6Jt+ItLJh90OJF3S%*kb z>@YWSSn};>S_-;=52Ahp{wyc>b3Ca<5HQya&+G43H)~Ee=T#En%#nD!6kn6jx6Fiw zB!hF9?{n%*NdD(cS1DYqQmcva+PA6GVK|u6_9_;_25|{o_6Yc`Plntuz*&tBzGsiI zr@j6E!;WK?7x@AVqWs{~CKGWJ#3o^T4SOICwO|yZOiZ53v9<8=0ucoJXbynLIULtr zR4%%SyenGVNOpS?xB^{7BN7iububt5wpcdP-)~p{cg8PMt74hC{igItuPG(y-M2Qq zt1|Ial$;S_EG@a7s~%8xu^v{pry6o(uiCxBZjZufFjBxKwpdJ8kEZEug@3V!v+w*&uy1soi z+c+i|v3=U3<)(pEsBem)rFdf8?SEi(uuzJ+y!Pg%yunm}(8E9O4Dfvbb^7AjvpVp+ z1V2?ZFO7wJjW6Wakm^tYFliA|XK-+;$#BkuWg{*)WcJHSZ(3IY=pq=>F;sg|RaI3p zc9N0-inkFc4gclBcJbNxFIcAR&4e^hjv8HG8IZc14_c#$8FQI<@&jTG&MCgM1aegq z-G!jN@F_D+;LtWp!EllS z^X!qf_N;ay2@OpT@LM_{7cULQ)?i)lzmo~V_2x*a+wLZKsLgu*k#4hQfkgJC^t`a= z;tYkHO1}c^nf;lW-fCa9sekNc%ZF*=p!RoW^RBvjz^;d&gSM)6@nCEqy-IJlHPt0f zF#YHEJzx>izz|>$ZunpR z5D{S~y&Q>AT zLQq$L7t;lvGh9zOVTJug#JTJ|eg6D;@bCsQ@DRr5Kd`zE%P9Djns>3$H3Iwt?N{F- zXR`ak(^(z(hC|BXk(S^T09>UR(y70* z(C>Ti-OTfAQ$m;iq|&~>MKX&Y-&SJDlvvG0<8l3$W0Xv_V1?nek~uSFx;cNEBbM&> zlq`5kewncQ9jvN**un{V;RpE45&9LtW~`*g@8Ng~yPo_=lcRyLX# zS&^Sd-P|$(i7_4)52^XGyr%8%UmwMC+3t2Za|^Hf?!x1Sd9p1kJr+ zvf!Sg{p<7&_?2w%D+deoq#%BH0%qut(_)oU^Df{x0Y4SnVJX@m_;mm8?-0fNdN%3! zBIwnXOJ3)**M5F#a~D4G-YGD?qNdl`quil+&Om>*a~&nWJR#+<9GEA@h$R0)=hNWz z0ytukpv%IR1!18ADI@NO$v>p@S#CRxqQXI`5vq)R%VR#7*M-mgxYP^MNNVJ65?o0XdyUG6h)Tozq~hPAMAhhx-*$7LjGFoR$$!~? z^5}P3c9u$cASBs}IwLjt3{k7NJlkC!s2<*rSf z@H?2~6-(YYOPK1MR%!2Cp+H5&*Tk^N=$0ApOy|9b&m0&W%lHr@Qb04sYI$?*5-3h@7Qb4n&8c-Z%;pg>vZCE~!iCVjNp zp@%vPt9rYgs<^4Xhjnm+%s4gcGz3QAWQ#&0?52=Zjwqak8Y8#5Bh$-eaW z*z+P5-OX$Jg<>^QQ&UqfHF5xqX$jer5C|JvBsM$-fEtv`t@SQ%J@`9IW%a=r(=2!X zpnfLAxRE&$clZ;Pzr&K6+H!G2XTW@IB1>}KXJo79;!^r$sUtW_^!vgCgEHD5VSXpsB5Vo0Y++`1KQpCH=Z81fOs0eJJ zN;Y$Nioh-nVXVGyKK%DA!cZ>VpS&Zfc@+ABJxC`OrS)A0^Gq4_>U;SvNCbc zpqmn69T);O^$LteW)Jws4}eP|dp;-#9ww{2kdSm{s0EbZZ&fA882R6SOORB3HEMp~<7;b?%L{^hi@cyMI! zV`iLd70;{>UpTD88&Pf2eTtKoRTX0)toFSju>>_!d>j>9;ZyWp!u))Zi*j5^o3GKo z@{DYP)k(+7$NMM;_P4Zp3}v9fqf$~*)(b7jT zEh(`MyTfI<)89xg(gn(NNrPG!SUUcM>>eDu*iKDDVQg(U7xM%=e)YUbWp`80aeZi3273=4L)G@vvi z%x4sT{%T{nQeG-bz*%AADQ|dU*ihkpJLXwDk`Oz-$f3nb!~52JmlzXD&i(JN$x`}v z9UceM(OPpFbD$~0k^2f5Jjl2Lo&_OcgW?0|&jf<0Fau8G4grw~s2E0=5fIKWLD2;H zFEMF+x4CUQ>3eBJ~lT(L0(=UCgoPr`=`G;(hE+1 zhW+?Rp_=e$uEmi)*!7ajC%)%nR|RIU;=Y-2unhf}`V28zuhxYu@x! z@J5s&Km@efr5l`_VZbpFa5w`hFsXo@X`XLHvyS#m-P@br{>=|^5{Pm~DGG0cCVl{; z&OE6~Rq)sFVbdh02*{e5lIPCR` zuVu*|EsOk#1{Q*D3?r&^!WtWR$yc6k{9U1a4llZt<)9>pQ8Lu6t7j7Kw;yn@ewm%A zT!F|~kj;&>^4*HUF#JWu%!fw${F;u$I^>LG1o=2dgYiqd!VFHE;-rJuQDx%^VlNU! zX1X`Mo2dxeRj(0zvxfzH5uwufLy{V3+to7x?-0Z#uxC2hDzm!tft#^DR*Ve+4w&!$ z>OABN{?2G_!r$t89|aHTrm%{4cgW@0CeQ_D1O;MaMDNwGt|^DA)*a3-J%2uT@do2y zv;D6?WrNKNGhv+vYA?%Un+`1+z+@BClrZw{d%s0f|C02&KP!X-UQ3t6L|5}KUDtZW zLcCeED~lj+hjP>d2aUfv?k+?Qj>Xh2VUfU*b4r$Uv2Ck&vozXJMxesaR3~8>)GLX^xlOwCTI#w z*YkA8?NAL3R4+<@?db2*8+W35@$TWL==VA(tb;SXQ2%NV?(sI z4M!9ehI_yxQadzlA?`5G{mjBlU2VY#%6DLnJ zWfo3M+lEYNc>y#W3E=K1*SZOT!508~9>8r!Xg1p*a`YJIV)i%> zlCx1G;i`m=Co&EyiT;S>-0$ewR4QD-dbQA5fZqT>Y@kPpztI3hj#Y+)=z#%nCF-{` zET(JD&bQoi3uhYOc8s7#n}F*Q&22;om#@$ETT@AQki3`|5*MDVi(aT~N6WJ}eRo>H zHX^`XKSUok7hqiTyGM>iT}?pk$3{tik}7SYV$bV?Zb6i_DRF|bi1$xh8x*8IL=PyX z9ZKIK`c+V>=P5t7@q|>sGtVn|0DuNHBEo3bf|>xvg|~F~r#WyUfc_qrK@@r(4N$XZ z0=|==fKg|vy5JWSvEK8&Ae%%p}My!_{=%|NHFsj?8nV?;|71ghWJ#&d81Q-8d8q z`ES}Hw~lq~le3TRc^(DsT^B5Hm2u+?^k-`d=n#7#VdH3C(udyM9_`lPvB4iigbhC@ zOUXnqB7QA|ksZl|Pl-7SR+tke^iiRd!&C^JmuM_Y?^B8E5D!_hvN6zF7r{2N4+&*1 zkidZ{ky{2*LLh?!SF+(EXQ&gzeiu9dCkw#{7WTOPyT0nj%-Q6~l4N?m&#rpwxZlBe zXaQZT>wUeAs}jV6pz=+fAkms&s@?8?H}*%<>b}|D&(V(y2eX3Z>%c}fy_uBZ)q1Zh zrLp8IDk?{!L5@oL%^|a~FZPpry_OFNLZ^{e1nFvAbA0IC056IvHoN{KyLI0x0D&M` z(84VPyPXh{jB*V|$jI-;$%PTn3e%&{WdVJ)(n;2PgdFwTBUb8tc*|mDmfV5fx9-{K zTg+8jC_a1n5|G3l4&Cd+aj&L|1-X~Gs-~>XM|QY|&EkT(G=u!#_ExaBx&HCUG-@3i zyhxSv_yzp0eZSPxKY&9J`o#cI{2FlN30JK}7dXy0?Q3;GT7-bQVFDqn8#pE;;CKgk z5)#zAfU_F~f4~#Cj<=ewgUm`o7a**kbR!i`=3ax)3bj7jXQ>(M`9+;`XhEXHlv#W4 zR3x;toYW&q`JO@CXk1Z5s9@!uYR!1j`^}4MWEFOVPp@A@9TBQN3w1>st;ho}%y39K`rrcZv~LnY+omLa-E3e-RZ_avZu!lBdTcfWOH!wwQzP3ZW7 z9)lU1`|u5|T?R~Vr8pcpK+J-!^%SR-tBk9gD@Io)q~7E%>FFZpQsLG}!FPU%@hM@C5k z%1Awn?xeGL$C+Kvu6MiZrV~nUnx<%}>AD-)u}e7*Iu<`h4!TW$H>K`~HqZ-=Bks;* zX5!LWd66^59cgl}r$JZ4LlhgpL5g zEI|l1IO-0C)!hLFxnGcn0lcdqBneu8CO-V^BeK4O{6uQw`b6ZwOe%?bf0=2s*QSUT><(v3`3MkU`_HZRe?E3&QOX&gOT`fh1f|#6%SJ1%-tnCT8&P zeSFaX-s&MyIB0*-`JA0}hK~omhlhVel}Vf4Cts(Tt%kDt<6QC`RA0=K-sP;N*Luwg z+P|!qn@SBiw}_@j1o3@WT8or#TR2{NyQ$R2efok>K}C0j@@9ZkPvm>ezQ~E zezJMDN2CY1b{+6>7G0+Bsoz=b2vM-@dq(E-cDDDA2PHmG#4ej_LM0nhu-mxU8s7ac zXaR!Gq1S%xuaw5R$tBZX{;cV6fMpcV$Jh}M^~6RHuO;}E6Z|CtX#53c$0;v!XP3>?fUzp!4sP&gQqzVqY^ z1C5AMQ4MZ7wvNv0UGtYk??($dQi=S|^NEQ;5eGi3#B1Bz?0?TFLwgyZ24`KGPh)p3 z;S-yqK93PS7F$LeDB2L$-!`;H6lU41Y|VG8cTaWm}LV_r%`Lz6q}NxMHV zb+d4It$F1QP;EOjcEt^O9zhhtul7=5==shbK1NC81;h`$d;(FzS1qaF*Ig$Up8RoU zJFYZgdL<``m)7&@Rgr1*0|&40ud&2;^_pDOs@xduxE=&G2b9XV2MB;i^a+!S{u|mc zA~Ga+CLn+2u@AHULUCg?6hx3#RF4!D75y>^Lp=dshG)*MUhxt7C2V-K=4e#kFN+Tg zvsPP!a#5&H57?B$4BYAj%I)6S(v5j9XLR%RbhkuMKQ^j+z0OpdcNA%cDpTW?~PKkud*{FVO7MR_X&hWe|XzCrd7k~@5gP7yCgRc z&@s4h!Kfhn;mcg+&`UG0MD-8~-Fwoza~^2WZNb(_`CxPM)N6nVraErOE27V5UlMQs zS$;p}?Z9#=m@8(UY=X?W1EW(JpUc4P^_hKU`oRf=D}(n#xlVZL7!s9XDy&DS33`29u&`V9TmfH~c zS@+A_gG=Pm>GYrOCV!%GxOPBOFC8FvxFYB>*NTdzT4c7}0L$Tt+x4$Ke27!zU{7M6N=^7=oZy-8AdPYl?pkHP%%l2`-shiE&jn zHt(%C5qu?h>ec+Um-c?BvXb)Utrwz@^Mz0gX&wX)9(~9N19~?e!eXHT2LJEctI+fc zX=4N=e=G#B2?+~mSwlod_7vn~2vw65V7Yj{bOT=@*()I=U-<1qSQ^C42TxEezo;iX zZ7k!Xy?KOuHpE`K{K_3QoqB=s;7w5=S9fQ!x-UKL7qz3nl04fcx{uuy-jg`b^zwf> zwtVB>u)1iFQkphE_4vsf+Ai98tI_qFRSK{fP$DmJ;2V+qQ6mk~pQju$bO@fZY&9b@ zZ+t&8NaFETGO{-0Jb@4KU+7Cp0^-8wF~WU0uOu&IL3DKCKZ&Zr_`FiTwEGkFrPIrt zCMQnPM5esh{FBf5_ls}{i5m0Cd9DThVXkZy4v~NL{$dnyVpowJ$xbDn!EA^m?Tml2 z?u!d;+M&ugxMbG|j4s1*#sf^{v0@W(*gK$bxd2HFKyzy(fhZALM$Fn(;*a`hlL`Wb zk?FGvEDPlO{AO)_A*&o@KWFR`Sn{wAXS1-xuhHi5JtT(a_1F5fqwXT4Pp0=(1I5Xy z%lE~?CWa@lc|KphT!&LtT=;?Q)N>rILIIy))x9qYr-fy7natk2By9;2Dxyq;E0*3r zgT+J_koBuvbcSWSX?`m*W+Wfwi1G?9a)BsIH1VFW5-Oyn4`VAP<(WbWJ-Kx zW%xGs^y!&=b?+J9Iy(>xsB!gwpl22vPN=fGmBqzhY$TvzI_;~|-?V7iOM53&R8G`S z7!5S^*~nl)4E!we1}uDt?Ax*q=Hn|vGqzxc*s{KT*RE+l#xhbcpKmW&4{q0z)L!Lz zLDEn*9Jvk3Qs#)Qg-rH{=gP{?8j{BJt|z=qnPP(4+IATcGFqo;`0~}Z7f}>KR%eA` zB123?Gc|VAM&WT{{ykDE4mhvmu^iS_Se}Tc+%$poTL`t#qyGSWC`CV4k#7yyzK~HC z11ThYx%IX{X%}yg$|a(wF+zXaDF{u1N<^y=Y-{5 zEq2M4DK0AKEgVy$yUK)uq;5>aa#Ce4yPm$z@sgyMpEO)+{u;It*k~;O#f~7U+2c61M9#wl5kE4V$%n3qL}1_(*Ag`ja`t89-)-tE zSAE4{PZ4|XWf8SW;j5nm{Qg9;?Q}e>9{qjN&8qInI;$ca6eg%6zOPa4i@0BguFqt2 zqmxDVe=10IBAgVyq8sLC;~27u4+7udC|$O?jO^@YfR^7&{42>MG;0vZFpR74X<1n_ z)Uexbe;Xl+Ut(^==aj0u78bzM7RWFZ#L(hBYk0gYPOnp7-lehBGuj@oSc&CLo86k0 za`JQJwtw>6wXm@2BNx2Y$m+%+CG*ts*Q_Y#^KWP08lSQRbZHNv%)d9-zG?riGIOcX z!sOwR$=*6em%V~bv@*fq-I+yNrGd#78MHLHcKdisW`H%}!h@zweYL5p2~rymoANBp zHSpQixINmn(1{yAtLj`bO7I}IAS5vI4K7Dz0>2(Q18w!e^ER3}Y%{T0(fQjvk9JW7}0hx1?~*!-kJi(ORIXzx-D8**lCgG$!*>sz9>ja#M=03JJ+Op5k|7 zS5YHTlaZ>NCkjj3gj;)-k8?xc1j~FCRF=zN@_83Ue}*?Ke|ikAo?hF|i^@pnDqy+M zAuYrZ(J=IlQ?nb@L*%(-Y9ePO&uwjaAo-2B@sNI{aMx2&A&2*0r=6~WlpN@tqU$+i z3h(vrO_7rum^ZNGPS^J8*ALzP#!$2x{g!7kguzPFaIqV=5=W-uy8iEpjV$%T z$nPWyIJoGXoH{>8_(QwJ+M1f3w;iZPSQwKsCSUY8hL2s$?JfJw)%mRBa<;@HJp5&R zg8~5pkGHoS&oo2AvBYQxeBQlzFT*&V=rNhho+2hXRtR%=b*fsxqUM+pS`*wrO6HKX zN79psupCHdaYS;0p-?M?nA-jO@&u?E6Wva3)41*tz;sG3FDL8ghH4U|A^f2FD}Vl% zZ?{l(?HhLAUhhc+<9gve-JyG1(?1Kex55IdRi)_cq>}MX3P_~tyk~86ln>+vxn#o4 zRGFhi>?x#3wo;OA4pguTPswPviUw4qeuyEjn=hWvQ}r8AdUo#*kMXkXXSaQv&eX9X zU6#Awvbt#6gWlr3h-H+`49cvGt7{{px^0p{*D{-QgV&$F^iTR(99%xksBF-PJ;pbd z00O;?G#s>WKcxVHeQHTb1T=2r(39*qI|La&q~k?3G^q3Q^N(DFt2Ef_wLXA6Qc_Y9 z0fz(A7223(?N6kR?rih`oA-Fb!X3t5eQbd| z1ZsJ>F^Bi$?_H3Ed`zCLRPz)MuE{%0lVlRQnTBotngPv>g5Rd!P}WiPOc=I=F%-SkQ*hNl2f%$8h8(n&H|-c;Iy5^Qzy>z zNh{`B6TOwqmI4-^+DD$j(3LQ+P zCJLmFYDAV{8)+^0aUe-Ptu8ZVgb<97*drKWGHm1pjg zHm)>=jr-z~O-9d2Ef@MYNv=}7mBg75zgS&frYn*WLw;F|_Km+BQzJ=W^N%e}kPWom zrPA6!8Vn=7tw1V*gu=(r>k*2*i#NS#L7EEUCr;}HiEiOjQ$z{!sJS3y(!ABU__>mj z@9>)0nzbli=E;c@bfmVctjs=&R8NTS@xk+TRrO=zN7U|+G9PR4HEi{vJg8l=>PiiZt7YK zy~5Zt@q%$VH&a3@*`CnjxBg%PD)pXB+2ro+eiS#?{ z(q2AWTp1dR8wHvAf1aS5Ui2TSNYmbm2mm^sR8`udI~tSET4wEs(Zb4zW1G?f!&yAz zj98WF!_)XTwlojxjy!ULkS8BlRH)H)q$6?}k9q-P~ESa?iihM_x zG0J2{%6AnJpRm*w1=Q<1J%% zbXcKKviJ^Nea0)U9TP*E4B7=CO$*_5?1#AeJEa;S{}vM)>+kPx1j8!2z-Mr&k-K|L zqTkuX=;Sq#zF6b$AYSZ)xof8C3ijehYmOJn9Iwp-muB0K6Vp)>w#T59`LF-e3gvomiqSnpjeVe^i{PrMcr9X*VPTf^>f|Eu56Zkd`m*chVFX3)0rJf{M_@W`xtvTw`r~8 z0>@BN$-b@gH@UdNq32Eik00O&hKxBhPuRcF*U~V{AtS>&ouZPh;&c)VH_$d&^w}OzL)zFTmeoRie7}VT)QJYmGIt-=h;h{yb{Wj;*7wciD+Vf#9WKr z_-n9s_i@tYzsF0JS(bWrq(sVB@d)h)Do!4<@!@d9M0;W2n}<-D=<@SmYv>Q3`E%!- zUgNrEDij$V)w(*MqakyAy z{@_r!RYgb;`sI$A3M@VzZnwOIC9!dQ4StRm?`moues9DaD>LIxw-uI&yl|2J3L*!!^=(7-@dr5&`T~?Jvz?_;qbK0 zJ{8w_uKGlU;o44Um$<_p-w&%;9Q3S-H;s6m!nr{hXBa7Mg#A#m8rBSozjL_lr_@M+Qci zx^&-+rWB3;+zns)qt4GU>RT;NUc&Zm%j}_mm2#|+%IV$fiPfU#J4T|bcT^7Xo#$W0s&rEL z$470xZxo!df1M-pw%&+}G%P|cf<3s!(ywBP(QZ&_wqWen_~b(C(JuitjCHoG6HXaX z(VDmLMOL9uGyCJ&wr+jR)!vmibJpc!W#q`E@&j4Gv)vy|10r0tJY@w-O*(Li%I3Rh zVoph1i~WBN^P1mM5n2jb z|P|oEcMl6iBQmx-@HUOABIT3!*p0gz z{iBDQ-}I*9vR!HnnFe15Wp6pV;_YA)-h#t|i}_M+YjCz5c5&ia>?F1ZU74=fS`9(q zCz+?-(z5H{PpureRaBq+7{|dr4=*H(8+%)Gn|}nC**59ML#{`U)1#HseCbQ^Ya6{a z70_Wz=lBop${Gi}Mc-=HT*;?Zusn(%3np|p>u$yPu->B%Jd9Q47#P0zs>l)(d)dOE z^&kUwJv;1C{{OM-iCePcVkf4!{1zp@$BHWan(g6VKjbBtN^iGt?sdU2pSN^UyyQECsi3w0(|ek#s9B*ZJ-Yv;s~kiq_uuRu#Bz!du7 zXS3cAq1wrv!Cl55d3MBJhq-)Zbc~+%gaSKm&zU6S-B4$COlc`gcj19#t!{hK62vrU z$IxoK*kQh^DI0o;P8CBAl|Jz(xBkCEh9Uwko#%&pg2pt_7n3eO$)f04thU1W-ZFI) zO=>+-35!=IoN7EX>Yf2JM6Ag*ZthpFQ>@X}a)c(&y?)+zR%UjyvsmGq>#|sV?|)KY8RuM?lNiR9?lZfWEiFetj&C&%>DU z)?=fKdva}bT@;Tw2wON{zxuG@PdBs(^SQyS|4tv2J?xSh>ngyw$OrkkpOA%B{YT8+ z@#O9vkr*_y&{gDpH~H>#!kkZzwJ9Odk73mQ8UD9y)jJM{*@LDN&wNHXA72e@zQM)p z6C59Yiv_lVA#dB^N90qrA$`pj1UYw%gldL1mpoHsU7}tz?L^^7H{R%>==f@MyeHd6 zSHGAv`hF&FWP2=rq5K?P6BXOKW)?4uS>F&m89DS51}!cV6O%8Zp+s`Y@87>KVf!m{ zUHHKxZg?~<9mB+eUKd7P=PX9EUd@@e*sS}D|8Q%JQh!Xq* z7?wPn^i>O3s~S#0?NrPvAx6Aeok2bGo*#9Vn(RN18mFn668>Q?MxP9`wj8HaW+fK1 z!+gp{lP`<^mR%7SN)`M=Vlp245F2wsy-2@1?&+;-4cs!&?fyG-SVpp-{7~B=~q&6o^0fte@Uz{9vJjz zS|qGIq8X?@xCL7VZ&_br3uc>`OWcJ26D5xtt8Zt5{JsrMlxlVq)Un~pX|bY_TY=x+ zc6*)}LzYoAHXSSP0AkBo#6S!m9gy{a5!)N-4GJyp(UC6Vh^P$aW_fve_IUvWrsJS( zM$Xd&Zq|t>gFzMqqw3hi=;KD->l<^7ZV!k{Cg^dbq&^Q!z^tq^Ia-(xIu%#P?MhPB zN3X{+P^fjqLd6PUNz1VC|>bhPf_}l%7dVHK*!;sRq<6*NO z6A!Wn+dQe{w4WL)U7}C<72d>R9bWK@ZvumoElkku0^U70P+|WaT!HI@XhUzo=6vhU zow*2AbM3|UPcEQ)LmJ6IA5{W>QD_W@2A1F%kY(sW*YdNU;vzbr%rgU43j7q&CB?sd zF?mGX6=)B;Vx<3aL&C^?CL-xKS~;1hnyQA+9eTPyK2TQJzafa;!8VSlhUFh6y~%z* zu>bzvQgYSQk&y6Q95vyD@C7^)jo5#CZrKm>3t8xk3kyF$D~3SGpzXSmp&?3z+>n=- zcLx#(yYNu}7Y))99N0W#|D;j=cDM>k3a-oContpZ8Rb5KJN!L(WM?j)-pT0qT=W}= zNv9mlY}e8=<0klgaJL>{ zi(^rlkfVV9Z@-W(Hc(4{lyvn6p$b~rq&f7)?VhT3juvqT!Ek~3pAITp_e@G8j_Z!6(QKC#_DHE8T=6{C1<++C!*os#4zCxucFQz!to=R{vLzT8g z++G2?Q{UwnT?(F;6Eg@+><=Fg^!<<+C1!quAgy>(umZLK7X$fo)x2-t{P9K3J|KP9 zpe<0}#6$sSN6n9R4jg!0tx&{igd%A0@Y+Q!5Q8E;^8~@&&VBD5Qcs7zQDsm#8(UXF zo{;_oLun&v)u+v?^o0jnH-UFq_nJ@j4Kx!KRHHGPij#UZw>g2{SJp3i2SF8*g_|iS z{-`}AP|UYT(;{U z_M;zy30E8D7P9%m;lLhG$I2RnWF4WEq^)|sq}R3a@J$GwEpo-9{uN*S1#x*jzBINkqFZ0xB({1Y=-8go{1 z>DB?4D>~u7M&r#~z9N|tAJ{IIdJ=m8JJ1Y!R(eLp1w6cLV1x02u^2I@fwyn@X8dke zVfP%2bxMAIC^*Qd{lT~Z8LW8!kHZR%yW5X+z)iU!9M9q&r+5w7{~rus zD1!x!-$$GBn#!mtFZAwgM}`hkKURE{(<^a4+B4SWhK7dV$*0^5`}Z71amD@_s?+r> zxtOsR52M)ldVana5+RTZBWdKg?A;*kdOCCc5++`El7f}sr(KQDzmsZphCUH-X9*6@ ziY_hSEsX`ct|asay&It?RI4eek%paGR~DWm&%Nm-sQ;iB)hu*lWE2+u?VOmKlM}qR z_vU18r4jON_)5`8>bkExs!Q9LI_lv$Hg-LRR0hq~b=SPZj)n+|kmm6nvDH7p742<7 z^G)FbA42Co{|=g-`?G#gS}U9>ROEX6Q*%WnHzjVbh7?n4bpNvA1~{g_E5R|t4E_rw zZa`Xafl{r}S~hpM!z8!bPPzN98)HLZH{wc0j(hlz)oK)<&%#_y9%e}cy(!sUQr?cp zC+hsQq_%@!lx4wNXYD#mvRVZ4$b!HdQ8Q38L- zG6;yky%3JHQP&)^hpCPj+7aAI2YYQC02^cCao=ERcS-agDapxK3y0el=0IO~uT2xu&qW);v;8{?%2l@SX47!HvrN{$V7o zghuJfy*l&xXW`MYgOQ2$KgFcKKMqNuEYJ<()&-Ntf`DHJY7H@gfMLtjY^etJ)Z7V} zV2ckS>LI6~KzgooKq4F3nWYaRIwjiP=J!zp=cio8G>N#{M08U@%K6u^c*&=%!^l$-+3T4;d#ZQ05>u})t z@%U|S_6aHzHhpr z{adPR{7LaRh<5A2N{{sN`IeKzc=P64a1Yx+TNo~f?U^kXT;%Px#S1FE*p&g;qwQ<-KQ`n$ts(FnWZ=Fy~_6X!Ozk@kKmji~IWad#>&o zq$JauEY=yrv*veEInnii48$cnH#aK5v`lm62stRE9;F0s%g={Wl9IP>-=2jwtn;7^ z12M|xkE^ywV417M>z)347n8fH;u^|CA15L400F0|nhW}jcMq{%uj1USlUK5e<6sHN zT34Q^k6-_Ns>f{xFuB&J@eX<0FJ#--zmLXW3Z+SWf*h&l`NaUr{hJL@f;jZPeAL4$7p4NX{eCkv^mFw~ct{29* zxzL6XG15Y^R|_gR?`cF5c6O{aTz-iW5)vLQvb!Hs`-~xN#Q6x8Si-8Fg3p5>h5BE-v8$y?!$S38 zSHOJ<1c68a0(Fap1({lq=qoa>-EP1xhu+X51$x*aLId7WcNlo7Z^t66O6Qzd5mn(0U4>`m=4+C< zo=UYXAF9iy=svo1F~99-efm5af!@|=z|IIb7cepef|vqc_i(Nz{sR}6Gwn4}$_+;1<-McD?@ zswghMNSjMG^5%D$mJjJfc%xa|DR~2ngK|MulX9M4WQ*pTp_Ho;mD<)B%{5x(xOLJGVy%f!1s-*f&vlTMA2$AJ+AwUM)8`d zto?F5JveVaciLYSgF+VS?|)kj`nx>8r1GoYqB=K~a?s^46>} z!*P3yEBaVo%HdNuMQaG#&z8EI{3&Tp-yG)^?at1*8$Jn)t^~|hB#`WCr-e>T+0cIp zluJ>P$cAf{__3oR?q<jYunGpP^PMl5c}NIFCH>rXN_yG)6aeJqj-=L+-225=geC3b%U z&J9(|bJzy#n2Zvy#MpF-Z!HZ3(@2J7T24J~a9=thV zqVdE}RWP9#?%a9hTIRHE1f4t=F%DRI-np1@%U_<4vc7NM2&wTPKbLK(IgG73hZH6z zLmDj%Q$j*N`hdP~-O&7jhPL-nOWkSW))!UQNU|S!y!%h;pe)Y?Niu7L%D20(ZxcL(V48su?C z5){QA2YT{oO60*cn!K^X5;~j}EP}6F#g(V)sPU(43f2{njY;U+-O9%1=9l2#b9Ay> zVE}SSFPmnl0GeY@wvws`E6ac&K6cpeiuQt!-r;Vz{||MLMF6WIlFS0c;jsB4FZdA_Sly_nHplkK2_SBTtAA(^N%6vx0>vK@vMm< zppEBMch4u5$Li36FjNx!!X&FHUg(5(tNK??kw|!0Przo*2ktywci!LBfrMg}=^BXF z=M#nJ7nItwe~m)py$Y^Vh{p!fat8-6uv5do^#0;`bnwmU1l1-uh`BaPNAg)*(t7rc zpije@iR$skn}UKdi1o1J#!I-{pI*Vi2%mlSPj*(Dr{m`?oN{gF%}T6}Sf&7BhC3U;_4?y>_dh1(Qg0t{Au z;>Dt*@{6Vl3;r+~5m*%_Y-$xJ5k0*UECmf)ZDLc`WcZnN(SShg(BHWO*6AL`=|zcY zk~}X#=awe?4ESttKvB`tC)+wh$Gu{|#r)4$g^Nhy3fm38UOk_T!V1&M`?Z z?e6EJ7Au!ADRvb=M!23g1iQr}7Z(>}HWx#ooe}wwp&jwvg5%CI(=i+O3O+U+0+$o( z2#Q$jiHm@;AJ&Xy;WpAn5!Us7;H4E1h|TVFLS<*4+oJsZ$f-ECpPR30IaYpSOBQR{ zZ2z03l?vWTv^&8uK~7Pk1?q-QP{Stec-6t%72g%p!^FcOxB1b0^I|Z?WpB}4!H)cK zIrEzwB)&orIj{q8s#(kJRHi_yQMfnh>j*x~uguJ5L9t{2G6@vs_K1<0*-331Hpbhu zjAgq+ymWUUhzRYGTbcZQ_#qZ}7wO&IcBP+hehU2c~Q2G2xzGiX3a^%Wt)tjnu;hbXs3zsqnA3S{ct6&AxRAIds+rTw6 zaUGjl&n*{Z{d#a$T3R|SqatnM2=Wrw@`m6e2*X(ds8->K(Tl+tc6Rnm$4&i_Av=lx z`3!jKn{<(eg~puEIRF?wUTQ6rWuU&r5)*BED^XfaSlO_NYe9Jc?EtU&wm|h0lPIzr zJG_}T4RMQQDb#G;%toKgcQJ42nS7bIf3C)gdp)?F<&`B@@_&AU(eiw|zL9~=yk<|ryC z?Fyqs(oIp&ZCtIE&G0#rF~hmG3rGYw*M%D0XlJ~ zuLo=vV(EI|9xSty1^}B@9Z;pN9bq*4-xfvkt-c%yu!*XH*3@c6S~Dk=;6Cg@!$@ZXLy~} z6zs+tiV{8MNmo=fFL_M;<{u{7{@rme{6-aP2LJuTuW<3kYq>?m5CVPMcck?DZF$ez z4p>l;r9Q($`;0$z*$R_i-7EucI5aVT(pg9e7Q&U2ybUVl`_jP-8KptM6n0x(`!`;wQ^1EHexm}tvi_0t?sY;{KdHV=3px=?_Oe!ueSSpBlnQT^W<97o4>cGWH zV-i<~>}sDanfh_A8oH-fLLUm7+!s8|>RCmcG*v_as76G?FB^h-uNZ-Q&(>mJ09cOw z2v6b#OzdR8%^m$yP|$`SS}%y$RW)}ZR(snP_7#9{?u=v9aK{5iQEquT0O1uQ2;nJbR0QhQm1?r?B$tOI0?V)|YSoA^j^i%u3Ruu>pBTMq<i+SuKb>IFfUII6d*FPYp z6o43fp7V0SMxY)it!4!|t@(3LMHEyU-{1s)(6j}*xRO=Lu3K-bZgP#-$S)Gaz{QFd zw(PI2$ZMJ8hb*llFfJhQn5JW8&RhV5GoYZdi9MR8dol?G45<6O#neCjbWW1DJF#Y( z_%iZl@nh@Ws^GO<9(Az4*j&rORwc`uo36E7*;k+EV)cdl6U-EaSU-cNRp-t^H%{s6 zWT$ud-;dezc-GBfvYg(93!q;-&~d=3!{79KgCD7OFx>QiwGNG(SSstgvAABTzY9 zd*xrez^DWQ_I?6N3SMVg>171%W%Kl}vnM2BGfZ6j34MM$*)6ek_;+e} zWI6l7UHSpvwmE7w&6d`r^aQ(=CRFNXf`xfU_;bu>MR8rIgO@h(H)+G!%~8Ji9Ah&` z<+2N_QbI}!oD1S@r{!`NIuo#={G~<=rS$al{?Yo~3@ZQu-fS1a!zUv<{4!SrmR!j{ zerSO`7?cX@^;7{ahOr3Zqvn3R?R7|id;;8(@dbhI;bV#xNtG9J`P7b<@W6AQ)PO{p zIwD^e|3qJ2dEnv$88gOnUeiJ0Bq9&LVMew{91kjgK03VraAwF`3or$tn3fw7eT7%u z0xSl_bd$wrZkgbA!DhSg9GTw8O@-5> z|AC0ex%Kt+t3*WBy~+lb;27gJ_0tDt3xayMLHU~3}Ld|8pCJM^! zC#&C(_bB%glP2WAYmL|bMc^H#RWQk)ccm}11IT_-yMH!SH$BJ?%6XLJ;d{9lMQ`3vlMUNbig0Ld?SddGDt6g0`e>hp#6 z*l%zAD5r&F#8?Dz{)I~`g693}eU^V8Zyb3(c1#hcM*}k5GI#=iT``LMC8eG7)lx@o zF8QWu=Why*w!r<(=Fg`GIzdDMeP|D6J|+$Doj5fo@{*?EQZN!%9nz~42SzkGMq_r} z`XuHiML<#VJ&hT;NaF^>O%7bw3THx80n^r587@M;p72|Yj_*(K0Cun_!AlFRLUbK3 zv9F!MCs&ir;%6z@48sa@^@_h%f7tx&U)Xl{S4ivw>e1?Myxz@*x=3$PDNs%bI3;%E z*wZ#73+m{ofyzB@@RFU7HI1$B=y@%Lo6$nHf^Us92`X0lV^6}P|6CEH?jrK)`Sy4j z24`6qF8C1JW9n*3i~+Eka*qBzAnb|=LUsBVA%v%2&Ro7^qk8M-LT`NJ$ zjNh|3{o%a0@W$2-JD84Hw_!4bcG+wfO?oic75Lw9p?7Pc`z%xVrS`Hls(;ykCPu~} zlb81GrggD}iW|!Yy3aHQfu+j#S$stNjw{!FWJd`e7OH z_Zdc8L@T^26G4BR9k7p`zqvE5!0y%&UnW0Cjg(L&p@Am7VW0lOQ^|7m55v-U;^Fh@EtOxE3{`T(Kr?0u-1GnM(hzzU&9~B z_D+38$&*&!N_hPW?|A}xZ|PbC`}_e8C{^kM|0DjB54~x!%S&Z0w8BG$ToqcuEfY_WxB30@yJ|$x5uh=@wGu3B^)vQC<)irYC;^s}z@H?iRVHRwl_50Kn-(iq zIQO23**g}ce`P6qwuwxlCQ7ds9n&VHiyZjNHV0A9y=-vpyo-?FfO;y(9!ur5q3StXQRPA-fB=pl=LU6|9WJD zkdF}i|Rd+@9yFo9(s#HVt%2B#=yB7p9?ZsALK1Ca_Sq^y!Rck=g!yc~yxsx5ui;7=RPbVZN+viUSIM)m<{^a>KcV zDAFrVk5^yn>E$Ujz%snOEMw8o8q#y2Z|Op1&}`l26a{y=l6w0^ZiMe`F01iKM1&0m zBrFncYOk#=)N5k8%h(!vP4PnQr3U9vmsf~m8cFM?yVFH+u+If&lLU`NO1YCOmvi5D z$tpDU=Pc2OEya*e?wZzxx5DYg3wI%p$=5JKvqTdvC2TGF@#GOr2eT7A= z>6bo>>U;}bE5;kxuU}?2-6!|>R9*cp6|MAB?_1iGXcu>WBD*D7s9FtUh_jE^e{+wR zI)I@b!cVzdc~DJxQVACn6yPIB{h?9jWVx7DK#>~@5!hA@G*VIAfwQyvU}#(f{xo4A z*WgrruKu^AuEKc&<@CKgytZGd$2!GYV^YIJJ0{BNtO5`Ejn+oHIb4n=PH2}$oZ>qQ z+^;7myO}tLX79!g87ghdHSwg`B^qo`Y0j4=XO)Os6c%O6yE{?844TCHMCeV|F;mU@ zF~8IXsr)cRr6Y1|=AVHl<~wcTtJkk@fSemV`!LoLP)##dme3los=20U0l6`VcTaDI z=2<5b_Vq4|9;9<6nvvLM|I~Wd`CV@ax|TTJcu^*1>n%qmHr>qgY9>mGjlUH`p6xse z-sxr+5|M4Dp<|)YwBY;FrYSk{jjQUT7U}hnm)o;#^RD!^2bLrw%PWW>f(rnAa4or8i5Rf*h)!hl z8lAu=mVc37I*~oW`~70ypW920SV{z?-CXQ!Gc|3%b9>Un%uIHS@v;M`n3k<73u7lt ze1fhNznEMz^W*KOVzy39wrX`zRQGu;ADbyY<44OrvW$10vKU%2wW)y*Y?X?TR6s)x zQw@w(#lWfd7G-E`Y#V|y0joke&!`7K6$xdCcV!j9U#MgGwxb!V(~E(IG%Z~xh*Gc< zFI1XoPaG{vIYNx>Hq*p&VQuucUbBHAs0n3vy&KHWYF=tdhaX`ma98v?!&YOG_>9yq;;GsbL2~Zp~w-aS!Wgb1MnK)xZ#j6{t z6h<$Uj8T=AH8qRw3WL&was2hY+I`uw9SEuX7l{Rp$ z$qGJ*&!5?=-D9L^qiFv8;5sn|NenUbN`HWiWiqw%(exgnn1y4{9FnF}E(b^N;fou& zDizKw?_BXj?MghqBU8lXi-qA04)91fOch_?%pl)JF+#ix}+FxCBc+LTXE)A%syr; zMJMNj{{C(_1J5hLt=FEB{bj||Hth#55P|su#-iEt)QC`t3v`50G8J! z#()bI?=J+L^cD?B+*eI5KXm;?rkaSZ!lrAXIZQY=&hmn2i}yzc_D1h-J;tzt8>l%V z>A{j()2c8#dRsCfW1jGln#MPTMQm|lR>qRt9Kc4^<=whjLwK`=P$XZ7qT)ae_UroH zE3j0Q4Wet}#e-j&m`ua{c^3R=swXS}y9Lx!4-hHQ+xR%uT0pRRl8mTF19cT0%$a(2 zUB$Y(IZ)7jPG|FlVImHo89i6|?ZQwvzu|>#O?JaJ7BMXgIkxf(TwSGBMokTjFC;F> zj^l}sC>z+KKP3)suDNj-v>ZS;aN{Q371>gGSMXvJojydW)%s7jaKL>f%y0wf+EREE4L%(&_UAVwQ(lDvYLfEy#8)2w1~~HNdUkLRERdvk%Q7*(+C_6f6PbRHK%r`r(aS4yM!}Y8F@jac6NL09``62hc2S73pqNLnw{8l zKQ%~LEt1(edWOO7eyFHZ~uXkJDT6GsL0gtJ){6D)27P59kv zq6}zTF$p$?YCdp>;U_wy2V8?HPPSP>nAIC$a}MV+r$8)*?i7M@>KnNHM*%7GtYT3@ ztR}^VSl?7Qcf>7q=Z3zfRai2;+CxX zC-;Qn;~)Vl)slqqkIN3tWi)XkAGTca`84l4)#-l> zHRJ87WtzI!2ibp@9{%+_V77sR2rjsqC1ubXyiCUVR7Ir`IAsAkdN9pyKaq7%w1rPF zzv#h|i0wcBW^oDruFYpxWnoC0+ExtBGBc$|q1nHSi#rV}-)RY%iqcppj;@L&;9XbN zC43iw)vvCm)E_D+K5#zD3<{2TN8RqT-Pixa568(r!$oXhULI*y?1S|F#^(FMDflxG zIp4&`Qv=u2@!egRPWXmo`m9nTIh#LRb#U1|RT8n`C~&o8A%8@#6jTav#}gLsNGQ#; z4xA4oaz*fl7@byJt}xun9K2T?)A)Vf)iJG{A}0$gQd`?b!`{=tu6ETTZ>MPE=iiU; z2%5rGp=7y~cMFF63?pDGq@4W#>qK<$#Z;WL;XMqx9G_kzpm#_BAetX(`XjjRxNBIoYHlus2dn-${qAa z2@UbN>1ZDQSq*A~Swe=RN}zkOLdT*maaF?EE<(H&@-3wrVGfLSKSM%u82J&6}FMUrB zAIBtn6;tqbJc{X^{d3CFw_ToTBK#ncT5@H}+HI~3nXLoD+2EJZp3Q@lCZ!U67Sz2! zb(}Qkce(ACOhD_ud%|PS7aT69j@vz%Ke#U^Y>tsMJC(&(zf}VQZuwYX!r=4hlgp!B ze~_V3w$~u*_6u~T_U(s!_A9?xaUexdW3^ycgB9moO3#P3IVzwmfl2ZMhAYUYz1M)_ zX{b>g4ByW%EYByG!`YC3jgQawH3!3WXtBda*8f31D{(J#k6g*6wO3FPCU7{V`SfwJ zv1&waWXqO`%NGj24DH=I-DT&?0d++8<4m~X60LA>WSW>Ko1@GemSmolp&|7RqePXX)iKy;cG-l9ZK;5cE@KgX@4dK(jzdG?z zOK>CjcMKJuN6H0}>&bYXrCNN_Ih-Br&TJEYGnlJdIa-TX0zexCaLST;S7vRIVXmj) za5kE%8+vbsvhV=AbYs(ka+wuZS5jJHx*+-k?JmqMbC;N~*0*=&w^3XQeVMQ7^ie>` z4ZXVa-ToOTb7ZuXCV6-{H@{t7;L$xeLA=SY!`=c#jyHKzv&FmFYbAyUHiT8mk%k)i z*bf7b`G~c|1FAHlK9;GH?u>@_KXYNCN-f33Daq3-IB?qg(D8z!$1hb?v2jqrS-HXv zuZN>0Z3unh22L7+ohdy$w=!9Nr8LKtjaI8bi6pu&T0Z-Og_3E!lgfcw&EMTkH1G!= z7y=%3c0cT5_YuAQ?%^$9Er6KqWzC0cr2xYs>^jru$PGQ0D=zKYVbX=xF;C?#n|3>@ zUcWNy`}r%$+y^gpn*7Gb>KfNu0<``7ZgHA2YL7%b>rESJ8Wpg`G%W~gq4w$x8sQI^ zEL?VYtkJl+!W}bKPdz!dxc2xO!6{_w&cug!uQ0AjStTaFA4ReDWT z&>mzDwOxzGczQMTL$y9#6VVe-mg@!5XT>RXG4SB>psk#7Ib%nF+->9#l)?xd6w)dv z-b=tjqgDw|6o0j>@g~+Z4q5BIb`Ny_e061{FAu3*9O<&eZLW=2Fvm*e_s>wLw-_t3 z5ADu&Q>QoXThi;QwTd{I8;%jQENfb`g&-wTdE2%*dAKPe zI_IMM`qEVXVA0o1Jdv(TnaYguPCE5Fb%awklqZxyFq_24ydzMHK<2Drmj--LHdb)h zX)(=aq1(rHvDXp;x^!~}!g#D{f%OB_ui57CYNRqlV5B=&zk#Jflw~c6sVfQ02dj@* ztLcVVse&U?YdK7!jf;4LW_!`Vpw|X*-*$dH_x zDA%0<-~m{VI?h=FZ`5KwAv7C8)lPuUVGSIzps=w1kWj=1&y?{Q5jD3v88|Bt9@(%> zNF%#Zffj?3AO@BwzSnd{iVrA&;R1f^%E(no+jPbzjv? z)?-2{8~0lRB&4Npahl!Hh4<*qU`lwY*djc1S&G0fI-aZf*7tquAd!@;j*r^*Q}xR} zfdWBa#G}Jre<{lANgY!EAa1z0{-XNfT;X#kTK!HM!Dt(Ek543Ej>pf5iB{qZS6k5n zN}vfu{yu|uascNyGE$+qxVf=Wo|^`Y4#W`<_^j@Rs@mGOkW25$(LN-kpL+MvBcG!Q zEI(E6V!7oP>8XsDriGod*|r6Z8_w(YP~E6&=Qo*u9r6Wz1caIIAL8emCBHH)Eq7JD z7Fb^;z^%L76VJs$``wP#u%*ee{r;sUQn9KSwl@NXkD4#OyjPO!laQytzBKCCzD_bq zykX*o@sg;M{@xAWj{DLbra6Qbh2j>DK)rWZmVv92<7b3mv@iT^Ho0!uPF_{jcjROS zvk4eUC}>2jeW`<^+?S#6gYV-8r)k_ov;t7=X*VjU;`r@MqJuK+WvlOT6FDbM zvyf)EZ%Vul^loUq(F1!aEO4k;BO>uM=-U@B2AT~#c)VQ~-{=0JE_HA>xxb6I<8#NC zHbU%hGrY{qIRDN>+WwN1?LCi?M{V&^#We=)_n1jk&)b=d3A?HBVy&RCm|4)0urpeG zB)_M^`994Jkk<T9hm?kJ zPul8|d$p0);-~h`x}UPN{wD)U=;PvXHFGTkfbp?E zxnuhdc;tVbi33if=wN7Kk!o_W?p@T5TR68Yb0r5mo@7u4O=h+$46}lQU#g~RV{sl6 z!-W?L9I*B1=X@d{qkY@+@^gjc(MmY+l6zw2Uht2WKaJnsxr}ZkW>fb|1(Uq?Jf3S< zuPflxXj`Y7%v7;dN`=V)hvbnRmtYBDL5mFZA206+EEpn3qm>6tZVWt%W)Nl|2FSoa z?H6w8x3;pfC>der5-+&UEr#oCpTNV&Y)59)*&g>BZnq)V~~7|`#o}NI3q-G zKo&&kbrM91YT5l*Et$tm4CyF=k zgAqQV7ym`3DtkiyBFFMr)olFTJT_l~4M(a^HTQjWPxHlylWHDYJ|tD2?D|6-`S2$n zhF6Q{kUq#ascP3BY3bSLa_S-1??4fxk<&CYiSPpv?ibK%5ywp=OGoGTMHB+$BVs?$ zFDcqnNAW#dP^(VmF z7uMI;U(AR6nRW0*^@q}Hm%&;6c#^2%D_|c_auB;h9(35!8$sTzl80Hn`rG~RB6%6x zzRLtIVpHTt)^w5cJsxlER90{d{!O+K>GUj|)n0o;jUD{J2m#d6P|00EuSe@Q!pI}5 z31n}RvzD=R%p@2;#CT+x48|aYiyPxvJTP~l>9{_4V{b{uud z6ZDd6r!907yhR?$Y+Uzd{En%9kjpR_8lq)e>C{W&;#VA=duTjLv$t;$*5bDmNlu)_ z+aLA0?MYTqDOsdsk@e|S(a@g9p?fwTCZEk;4Sg8Fg^@8LUbzCGNwu;4WDZ{zL*^M^ zKM5@+^h_m2NXHpxD1f`OEL9pj%3yMF(roA2!ZX%NE<4th9AD*f9WL{-0Rj!%`G|+1 zEA%giI0;ntX{+|TF0LpEkD#Ss^p@x-eGY3Y$tq&}*h2ftRC&4Wjj*uPZS_gj#0#@^ z)D!lFYkX<||7mVzT21CYo|y_pYOCj6gkkWv=zCA#1|7D^qAWBA$=VXi2C~Z{LN$)a zyWUNzaTl%QM`(u0=xwjO^&i0(o15C>*RP$+X$>9bw% z{tcN@PCDku<5!wL;b(gv1x($pbXv-)Llj`+`9o&LQ%3NT(DceBZTcD>eT!7csj1E0Qxp8KSQ+)srLjDW^iS5{JXx&7Tp|01 zsV=A#r*imw>>a_`VQ-0f`0u3k=c#+-!2#?eIOi^$#d`a2#$DEX{xSM|i}sT1{k1X9 zD5vW2dAhldr(;q&9~;(M6m@PjFTr*yv)uZvkUS#Qj*<8(#v}=8tgxI5X|#^Kx1}S% z{cA%bFxkMRr53FPG4>Fh?zdJuuhibTrGS?N3$=r$v>;?7g-)lJU3>Y8?0CAo&+lT# z*^ah5EGc!Xc@`=SdHoMLnkDXxr!(sU3QIx|`cO`XbA+H%+~yE|J{GqMIzvE9oB=wr zXWX$KKCc8nl|cQrt?=D+*Xa6IBSvuzboq1>hd#*q-SQk7RCjXmipT6%y%dxidkNq`B01T z#nT}=aGu0&0k!~7#SjnR=;PK36o=7B+OM~&rl1iMhdmi>HF%ZTUz9eqZ_}_`^+n(b zKlhEXt-pXlUm%!u(; z9(l*X5>u(3gzvw(w&Xr5r`++N@?6qNDF&Iel+0okS&>L#*;qJ^*dlD$uNF7XM&iy> z59F(u(mtb%pB(J-UiwiRG)-gCM^Ef%S-Lm;(9R^Ic7~O^?C7ytOG2|u{an^B`NHx( zF;4>`xMUUB?!*6l+v+p9weCtr;pB&Ew^q>9RJ$SBH2c~BHvEhny`8JW<=OZ8bO*oR z*# z*YR$(kahLWAoX~?oE=!vwQoE{j0cw$k8?JYp-W6WemrxNU$;2>)jD%xSq_GWaO?c^ zl9S)15--)_WA7M}M`}aX=XqW`xOSY?J>tZqLx5Y&Myb91f}2m6nGGmt6>+VHXySDPR3ft9!4zTpoE92OD@AAlsS! z<(@ndvzV~^Y37WHnp;D6si63^^7CYLd)^Wk8a6fBdiUW>c!~0>rH8tM4 zB%K_kA;?ujzCQzu6Y_M+@~`B1uIbVVvwlo_!h160C9g}&zd6kxK`Iizw?Le~tw%Rj6{hUV((L^$1? zoI;P-N4nqieCbsZ-Ue1)tW;%~1-k-o!Fk4Exiz2pOg=pb?mqBDJl~o<_3ygv^G@7Q z03H0^D*pv4p%|hVb+LwY(qS2H*IjdqM?b@ZdZX6|-tJ59PJC>pr|q3L{5K_uznf6cw`|beM^B%hquN_+F8{f3qv$_uQqCAx} z=u(VG`G&KGN4NQ%60faAYEU&`{njOZtQrDRPQW{f&lo>e{Two?A}wQr=#?PVC{^6j zTb{iHFN5Pj(aK}P?CLM^cgA)9HZfbZzYgWB5{^I&vJC30w%#GB3SJ6pn>U`1Zg7T; zbMsJ+onDTM{~3skDYZ_}>ga>9pu<O#nY98y|W>uS;a&ct)$)WLpywi+z+7sgjoVuDy83-@eWs zLVeA`xI!-KlZJQSiphW9H}UNIy0PZOg@4d2mvK;~0oT!!5}BHM+3}*Jxtx;t>j6~1 z=1b}(zs49?d$Ka+k?EG;89bL!)QOJy@B3Q*{l1v@s`2ldJ4V92pOxS4sNiy42UoC9 z2BYqjr`6I6?zE%pE`epGUwWg7tS~HjzRHQ*DA96-s`67^sSl)ve?XjUD zJrMVy#|t=R?-WFmwSn1#+z+54Oz+8U?gkh}l;fIQcd5AHY5ej@qKsnHJ)DO@3JhLiVEPRc@G zb!$W=hAoE&xA{AD<@a4rbMAbUtkwD9i?oWp$%~;>r0x5hRG$xQ` zR5-Gd&kTIL=|Tdv?BK@8F?2%mYTog~inyg8uf}WUu4_j;eWfY0Du5@iyKA9auxW`A zROmcTTul2C5aFS}i^=Ccpghviyw_flJlcljChK2Vbh(y~kTsIj9&!oUFWBE_*-Fx^ zmEb?uUTc=$d}W3kJTcsWMPuJ-l4ot*xA|iy$wKoq=TB^q@Bi#v)R!K$>>We#3tR!Z zD2q2|+F?$D&5TAs5ZKpI+FbeC0(6?or`<8b`o zM%A%6#|9akg6PWS>a5)+onK9o{iVt1UL3y+6SyWhn)c0GwML~ETLZ8UEga`w18L;n zViqleF9jO^w}ufe0AQM`Bds3Gq}f$Ua7b0jR7RwG>4EP{`D`zP-O?|e%K8XrA3XEf zk=zyjShlcWHr|bt^lGu)+FV-%KD*tC3q5?>stZIYR!F;pru1NK4>BQy7BtjE!PNn^ zc2X*|0t61{6;6G(Tzn70$e5dDhPSZcCi`{mYB7zzg>h_Gy{|X~z~?ibrGO3kAdd)^ zk_n+25}*`QtT?y^${pjyAUOO6%WCm@<*dgrQ3%0zH|ba@FD?qY$EbF?`{8ej?fZ$v zrdq${)Nj#Q*q`Ri6Ni4N@s6m&K^7BF?Z9BMhe#fRiv&=#AZjl@duE~i2HEw(JjM|FKi(T(zLiV>(9U{tB zxXqTVm?>l#?eU1Efg8Vc-B>l01sR%UG2o|wphI|({d_em93|3`oLQiRU$vn)+t$aJ zbMxN79)>G@%S8Rz#Ur!xU*O2?c8Q3jdO65>|K)0Ivb{r5`OrRlk?y{H1sn0_UbN7b z(Z3~$7#|@A+I1Jl3%^#2Z^s>2z?i#~i0PfXrdZ*r+yT*gAyCV863uFZ*8u2n>cI!Io497T2-P4)F=;HjXX7s$X|b}(dyoX@?Ws$c!k zs%?AIFU8l(%FK;rGU~r=c`zA1Jj8_!=QB@`0TOLc3Qp%5Q{)Hf5Msf?v?YjFXxV>N=zi##x2RyF*Xdl;NXnl;6+_{3hwrr0_wcz<=rYBF@ z)?4|W=#i*^s4u!z1C#^~zEQJwi z0tDci3>O9>(gY2SRwQ2f;lsaBaz77DP2A_3l{lnHX=o|0PY(z!Vmkxr%`^1(vk4eI z8z%`#O#E9i5rFJ-!Y1qp-&WHCHbRdp6SH1E%R z4N0DtJb+b((>pq2e8!r`}c7V|NeR|H~j6kR?JH48PxEc z-BfT}{(WQpCLQ|f1inm_~b*R4u_pr*; z>ng$rR#x}j!KeTY5#oV;zQFA{sB{OC)~mO>kM>1SYS8@O&2eMYRBRjyRzz>F*uSor9Vi}sM6W$4k2dD4U5zwc2KjoFF}sJv)v zn!2tV@$_oAaKYfZiEqPrZ>zziy;EVy=_@B{J?;NFq@68M`U?vS(5*bL1*TO;TS-ECkka6l<`40$Ij)CxD3EV7A}Kn+VLVE!1EF<(UOI#* zx=ocI=C1+W>gpW1E$#F&=_J~1a{<*w}JTaAW8O;hXc!Yh$|0hq|yGd z*AnC6CNg^nAYMD#ezXA$L+;$U!yT^+QGi1S#jrh&6|TOXQv-UPwg4(f2|-Kuyz=f4w^rxj70 zfrZfq3;=X-jt_SY26GM7Y8R>w{y4|&OP(@QrbMx7R!07=C{64imV+o_#A(<#X^ zgQUd4-!(=8Tl0iqkMYX&*i^cwtFyCwqws7`;wcZ8zIE5_7-OR48nLC>k^RTtjAGLj zr{S`X9M8EKUrzcm@=^OeS#ea8j6Jy!J?ZZY?_I%n{@;1dgNvAUfV>wqY{muR9-%d- zieofTC;NL*I9*K2nTebPEAR2l?#?t-E;Z&pd%^V5(xb0~26m-M_|;$u1WtvL z+)FM!R*xVu!wkKAnntI)7+bSv8Jh!57#U->q|mW4qyuG2tjm4iTza{vNl09R@W5$^1~c0feCp@tmzHq@w~lM2gvI7? zYznuIr@YY{pQ_w$2nNhMAV4Tpo^_@_b2VmRG+(y3V!02msK774P4e*l#ZIPiK33 zC5Y5)&4=hm#&#ADhlgfmIlI<=@uEx&>~0?y$EEf^Yv^oKL}&fF&a)9=uxa>MmZFVB zD!Zw!>|AX-3lC<=O#7igBVJNP$#Urbt*pgp&V3I@Ur=x`m`t#N_xV9<9kWvA*KKms zhQGRC@Y_7o1>@nQAs?2dV92$7PVzABpTgM7$sC6Hsp}BCcNjOu96{my%lzk-BYMw6gYD&^-HR{wmeF4yx0qL?B_F)*x-m61g|`T&(DWYu zpxmRHybNL8i$D@O;d}h_I>3eDvfL&?#%`^Z5p@HE!Bw`=t(7{~7RF;;Bq0NGXRw?% zIUCFx-qq4gj}Q$Rq)mZqs89w|-U@=7b5Bk|;6$0cfeZZe|D?L*m}auFvO=k6>q`Xg zvma9E2V|Lnm$V_vhXh#dNJseE1Ld z$nP?cVJY3^&O5RXlfmq}8Y}s}x+&qtp*7qv!I$)3pZ~nay460cAwgH__CLmU1G#2 zKPCIH`1#>zSZNR#(CTpfic)dJbhGAiU%4MKbHWb&b4%y3u7<6eyubM2ZB6=xXO~Mx zT9MQspuAKfnWYex3jvN*GgjI;17e`;1hAP-W%1n$(Bm?R!lg?~QYyYP^~?roFjZiKLdTX%GTzNz+T zlqqo-PR2fV!-OsW&_0gf$0n-2*h+zOt+Ya)T%^)D5ee~abUDuVYY(6wq~^NagZwd7 zZ?u2*|0{qMcXK!)Wc%&9MXZ8(NYzC+zM(dLEJ^lZSJ#x9=SV4F-lV*?=u;(HE+w4g|v*mravPO|+K^apyqK&}|yHy~>7zIwYytu1q*LY^?R zLE(K8VZIW{(6VT2)9U-N38TMV5qutMxZ9)gZ`4x(a8&Dm<1p>eY#{qR7#qF4fPJ0( zeIt~>{)|AQWwoT|a6cu^VfHpeJFI_*u88+5Ce~vjv!lPixw}A-?XZhU>mj2vcfx|q zuX3dm&AA*%_Sv%(rgT$b4|51DNMjL4%IWD(@X~rY1JD8HKlt+B%a?uCM6r zBZ$AA0L{YpvvuxJev#R^@YW)p z4wdPmjAbsR_Sn{z5VgOffHQbVw}ksPr8r6w^!u4x57QR~qn^utZGo5rXODmR8lmkBtF zb>>N9TUQUG#qDmE`(;R+Jx$3!O)p}Ba-Y1ucobsQ&iv9kr&-Q07LcF8b zUF+5x4qCtE>h#G1X$+ce+rm`4l_h&xk*-ZkrSB_>XAK;&lpih;Okh?Ecd2m$p+gcN zwxCtcSizT)W!0+m;7o@;;B*P%)Oj3=XXs*l4He>m7jxQ&w>wY+eFG4P4RSa;y8e3B z*hGdBTmNijH9)z(Jc53m!GCmaGvg@OvFIf>LICR9Ca;?3QMVZCKKBfn72i~!V<@5a z_xBM;H-yk{GL-KJ{>Q+HyMf>~1RPRb@s^DeKUWX0-?zvcw7?_J0`B4ZuyNz4z^a0>D)$+sXt71L0{a|&h_l$ zdo|r7(rN)l7vJ(}eVO+o+h%d#tdCq9LKMe0p| ze918$XF6J-JpLc8GAkvx3K~(Kek@L!6Xr%YBa5Njt+X~YPn}1@NqOG_3LS23dS|;n z^iS6*C*Exc;VTuOP>NMId!sEl3YG>X#2YYBGExlV#W7U>y+j#gIu(P-^B}^~4y>jH zPNagp3{`0YItobHrF}TRJLs@|`-DMxyDFvJI^33zK5nBzGJX2_$aYs7L@wkj_`3T~ zq$n~>)X*Nh>1z5@=;GEn_i}^a+5i44fU;qTce}^0o1gV9NWamw&PO?<|M8uBG6J5* zRUQex_8YZ}3_sEg(Z8}L1fubo)hJ|Rp=A5yl?A3H%to?3gVIt^#4#N{vG%0iStTEA zGD!2zI-Bdlh~~OpyO-OQ36WF9=i;J35djbWOD^ioM7BI6$;JCLDVG`Hq1JQvarUWjI4)0?L+j4>NW7cafpG148K+A@1G2h40}+9n`+)vvV^6i= zERH0Jno>VG_@Ua4A_+fyx!oM-Pft(Hogdh2pNMOMyB#ia#|NmdJnPS3JOD>Y-R;@* z45)W%0Bzr!E#-!0_KL2+#v2qHsAv#xiz z?!Y{^6>vD^r?jBbBdE(0kmG9fygYosUx$cHX$Fwt91nX;4C%9*5h&%)^nt^YllKt< zXluG7j;^iPZt)=EV#>+JGtKipbcM``7=Q`;%CIt;kEt^0?tWzZR7mV=tAj!CEjn>aY|hOcUMgE40pz|Gvr{=0%KzhA@ct|hc8 z4)o8VJui@VIN<%l2Z_Q{C?iv(>PSNs(i1v>T38>_s$$#u@~kZ!xi{grpJP_uk75pA z-51tj5)Jvhdg@a{5wl-KusVo&kH-^Tv=T%>d&~Yb*Mi%XPR1aAR=9sl%2UsX?E;eSLxm`55JkiAk z%ogA31ZoQIXvI!3ATye>P1fx)H`R7*g}n)n$k?|Sfsv+Y`0q#hfFDUhqDHddB9yxC zmN&hNpr*tDA`5=JER&&&F~$d1TLSaX)xOwCiDDH*-s* zX=ylUsP)g+^+a}E?&?gL*akdcy+GQfZBLr@@7B#jXQbQO+T3mqJD@|#AC#4Gk>D^k ztANA;s2jq9v4WpJH;|M8iu+G5U>|^TuX=hs3Z*JiwMx+pS?|7^C|z#+ba2^BE}dHD zT~qTILM~WEtZ@jb7|4Rrk?N(r+nH94A4^du`0Y9PGxr|E+dnS2g&f3_G}9l>F*kH% z9^~QBHvmt!z22=n28ow`vD-24IGEv|wx|sSs#ah&84wFU%qD-~=`%ftvIYCbmY-;p z5zs8uI$-sukr!R0#C#n)gwCQ>R^WNX6A`w^$3>SMVDzpb6bT@q)#GgkYTFKsjn*A) znl|3OS2kX?r+c$ThU)&xq?9gGy zA|&hpN)5KD;eZ_fQA%Z=ce5vJ+D6Oj!kJf#^dPz>(v=~mv~`+Ooz_~vq<(wZZj}Z_ z2ptglV6mINMPt_dLRds+_ju`-u?qgIKYg5xu*I07AjOW5O6lL9-2Wa(%!A3)!rx~k zj{V<%N-7H4 z*yCo_OZ4`a=Lbvury<`&V7!Q2+0Pv*T7SkS+C!l}Pvl7-u{K`a?t5ZB5O8_H!2ZhH zo-C!O^thk`wS8JPHcT(j4=u6V9AaFq(<=1jfAYi|unYTu^2qpMBv4%xU1WHgzR>3! z<5rFV5&fC+UHg<~qehk}9UPGAOY0>u0=8}ln~FKzRk>|+$)<;pdL^um>1@EyKLaH~ zG^zB29xb1j+|VwJG($)8{#qV2d@GnSFh0*tghoS8{|yVIAA7S zh|o??u^v)Zk`yo0B3bc|AD==yO3;4^2?;y1wQ*pi0jO5$>gtw#{NoKMtw+ywfr<*m znv#*`J~JVvZwAJzSc3g!!=}vOY7P;#cV`dF;`G3Hkx9!-h_`5-Msl)L>DTkcu35wv z?*gR0)E%kDPBuH~?Oos+X_kivwdjuhTbZe`$FObG=K%~$8r;X|i76=pFP2aL4iJ(0y!epF1IFpHNfKeG=VD7tKu;@B-ZBpZ3Rw?6mB zu}l7U=|~@*@ZI^h79$VEcc&!;$!v=z^$&aMx)9q>xMn`nIt_W`SyfmeFO?k31bbVN zNjtk(5xzSIv$K9t^Ebbvw2mY6>3a3CK^=+mk}sHPp^Nz2m9eH~;zSstAZD7TLEgXL zk}M8I5oNqQB4f!iI5Lu?v-qypfqlQ*{_}AjtePHaLQuXwvp>5aLUd$vGiR1tz5ggv zDFg|ng#CMxUj};kvRW+;n@*^uT&v>1f<(2iF%OG@+iD(@I6J1@(P_=6AF6)?w{^`% z1Ql(MOIq8t-kec)5;TP8VBS$}HajpdSwyQc;Ha&ZA+;*CWYXl_;$eOfQL&v&BTpIn zS(-V=8w6b|707{8U7lKMSIoV6R@DUji|di62ov=k$1U4cGptZ^fpD96po>{jlovyv z?Wxh1m&-?HBqc0t15FAeQPsiI!I1|nxT!7+CPyy6uY#)s-@+TtvtQj64)n3;t{zFpvh~rNe?*30V`H#xYRC>}Ix1M}kU}cP zet)UkBA&^=I=DtZ{?%JUGTb%3{ogKbDTAkK6;{?`g2CXe$jXs7nk!0Z>S5!pLPCuB56i~UY z6#MVQmA6bW5a`Fl`T%}F=jtT%?Xyy|E(=Zv@>5Iw$|MOBLeX^tE)GP`47ZUl)Jcpf;Y*2gu%F>L$=RJGeGq7J3lQsYkAqGwd* z9Afampeexfuq#D@@}MGP{BVC*)()n!PvFH{h4q`3ULjzuEFSF0tl z^0gtmOoB93jqA8dRhcSe*9|#c&4I2RJMZ72g?S6LBgx9ocnmBF{O#YF?!;BL3^D$a z6g`W=Av9GoEH>++INR?zx5Ongqs`RBOKEn%sXG>lkx7&+&K+r(!x_gg?pWk=)e07- z`s!RQ?xff4N+!g!{YgcSQUBTYLkWRqfdQ}&MGCC5*uNLhN2J9aSE#Q`1dB(fzdip* z-?wExVs`s9DmAw6DARSK$U+1kzn)l4Ed4V9ON$9*oLo<`r4`E^g4}zauf<;(geZ&y z()emMLxpn)>5TE z4I=-AZ{V=3#Ff~zX~nWliaIzHdwMP|I!u*w|C!@L>}emcF6{okGa-;pjV50ElCM_d zB{2OQh~m0-zD2c3eG=Lql2${w*cCHxw9~qj=}Pqg|M-*CB^KiH?n7 z2CJ@18tGlGK-LaSOfWK}a=PU#KR~6z08^cZve`Ay@#~~nhCM3%?(X)-nrp5ZTcyi( zScFn^S9$H^6mgspd%V~Km5?y@GyHWfeslEu>GCl2N!7b+%OxaT>GdX!$zQmC9QtaY z!y?!wZy#|s{<>KL!2`O__->lprbSRgK^B!Op|+^Q^k7B&6CZW~&rhAH(VD$ShOR2b z%+cOAocFtb{$R*cpGPNiQM7DrF61<#>#*xc?Yw%|fem+FMIt3FCEfoPBZh$EnLh0N zo<_$Mbw|pNz|RiYGQ?Qp{<-RoRJ2c&{yW9rfggr)Y(oYC5x>iG@Lv31gMN3Alz1x0 zh@J&Ak(P*Gk(4Bb)SENTn9W#47zR5=r^fEiw;OT&$uc3%Srx%ay@TI*dQ`~OQPw`W zD4#4`&_0)kD`reHX+fyK#dFk`gq)5{IBSO!f)g4oxY=EbL$tEQb6*PG+L zpKmXj)f2n6@Ld@gtS>FYgL$GpX-V|-dg=-l4S3cv&>)C!=_G54W5)y!WC4Ny~wsR&NCOm7Ynm?cZh@B8G1JcwFW^8mEdKXIs zm!I~;?ikHZ-Bxwa)|;b65GlR*wD|$&EVQGu__-<_>RGHX(iXS(LP0vaI8XWnPV0!f z;Kh9Vo8I|KAC&M8&-pJ+*eBuUeG#n=cEPeVy=#po#VHN+53$ahw#}0RIYV^dRpdzn&|Ip25Q3P0JvssUViOd#EzTjztnRdwlBLJ9LtmxAx7H zIxBKshMBRL-*l%-k5f8Bn_6{~9ZqNZ8IsV8-Dg!sDQR`iR9v|c)`z3$kR?j&D$b;@vd6SpY(1$p3qbHyB#(#@&L|n19Kd!cmZz4@6@AJ*PnGU z-9%=hu<(WgQLXUW6(Avys{R1Wf@+w8ZFQHl}EgxMUn&3{vC;I zm$iLz6=jEka2XlJ_+>v;{d844af15Of~V(}l;)UOmS#){^cjO}75Iw|^i(L&;j!1R zS)HKfk@ZDEA0oq@5o~5O@BK&wIXk_$er*PNMUOmt2RXPECw}4sbC7$tUH6d(wb zHe|C8F81b`!_Qa=rJwqZ3h`+R6W+w}7SQxBJw~ddl2M8J*5Wul&Tc+HhoT`eYn;In zI9HrsMMLHH*;>MP+8e2S3SaFZ!VLTdtYswX+s$NdlzFY!ZDXQ$(`bthd4%Q7&z}&R zt~-S+f>VLW=k_yV^`G+&@x=s5gQ>ghal) zv!UblOA05||7Nz44*%-b`T!QGAo`JfnuFT+!eN zIY7oRiiymOK#ErDuvTPuXu(q;^5TsjB7_>hj9FV+auRKciARS$p_ulRrfL{zhupuk z#b9|@G(lSHO4t=OT(Er5Aj{P52`ugurx3^2te0uNYP<2FS)almxMg3jNJ9;u$?G}_ zwzhADSs)pHx7V)EtJ{z-s9Sl~O0=9&R_2|_HLOx$hsi%Fa+)1vwNIC6CScyZu6s_P zC$ZsG+TsHufc}E-PunT`mukz8yPkHQS?WJa+`yR)`FvPE$jHhU@-7!P_|-#(jGbc$ zy;~&6=bs-PdE5)H_MAa<%n17;-yfEd%%UrNV4tp}FuEZ$1sV{rjY}+P_Lpw*-9}i; zbDtSX`1@1Gab^r@65g|1(p^KMEAy#Y_KWVHGmGJ-{JeTa%N^|Q4(QEI%T-bWf;B}AoY4?n2c z4D7s2v!n!(MLd0B%#BFj_mk^GZM72@Id_+ZlwOamU2B2Pn*p!%>EtF42q2; zklMm3R`B`MWqUS)yfy>@gr(FQmA6G@fsOiJlL{53I$AthY;ewDR@Q&!CD7QKt(KFd zQG;}k?$E|qZ|53BU=gO?a2vkRAoYJHN6P^6yIk1XN%`>bhu$GnZ$0p>&oYQd58~kQ z?j7+V?L_qb*VoHLk=`6LO--=yZkLlJXNIe=5g&NMJu>f`C+*?(%ZYoHYaJF05g`cQ z%~fz&gXHu;l-q?}?5j%7hxga6y@f!>4u`y|cqnPkyBMk**C0B-tMHpJPdcK8;0h;O zZ>j|^b-M{>EZBJDo)hBVM~{t>uV|jqoX3%N^2Niv*dF}I8YHEsa{NJ_dI_nL-zq-; z^rG{@YQO#jovO)zIj9lFbFWHnb|m;b?L#%w z&Y}%XY3uw1fdl6|2li-`!T=<7EKGZf=2aF!ow1Df5QzS+#Os$Ia4b-tbpH5vpdLLK zW{7fRg4CRXf_45ZqF1IAS$_b6-Zlv)4_B_T(cx}uXv|kdowL}X>a-wy_JoYSsXt1Ne{BsV%teM zO3R2oWpy|_C1jc`TYL4Y| z*~gBlq5Ksfw&a3dwnq_ywShP5+3$Ji$sr2zcukmEzWJU^A+k(KyyPp}*4sc-t79Sm zyZC0D9qYG{-}h{nu6lCy-80B9@VY4y+2Iy)^pgM{sKev zhxuy{Sszhs{I7$*_mqSiY5-xH6+Xx|-S=e~EFnNbtkdAacf;MS)I)R~9CmwkXVy*m zz}?ZpOzFN8507F)+Or|P-*!Q_+?$j18yCGO7Tb+$%eMMBPQfguAS}Z9_g!t%q3;Tb zDeQW(pOaH{zV;c$Omupwu7NT(ub_i~ZqYiHPtcuYoxO;^Y3eeEK9UqC z53C|-xoeLm*{=b_hcNMpycEj=(^m?i=Kg6doH||JJ0rX#Ehmht-Makq*W-_1 zK$qY&O9x7^da`<2Ja5%kx)5wtM*tZ7lZSshvQsw zZRNsa^0~zq=&PR2Evdo z*DF_0L2CNg$WubzKY!LX#2F#jc*Al9^+Ft?%6Y$8f6HGSOnU{H`g;5J+mc*S z2N(lAWevxREwM`4k7=^_^F?S<^nnH*SZwGp{oDMXoxQaCpqQ3F2`s5C)RO9K$9i?> z>5YYu_a;}shIa(FBz3g;c~bxR?Z2QkUXnFnE>vz#TfIuF5=!ThFcjb_fc62ZzW_dw z`p}n5hAkYI>f$A3a55yA(<7Ydg`1rU{jOfC|B0kIlKM@^wHtbZ>k;ImYJEIR+dlS6 z8Lr!E(oGvd+d3%$B!Y;n33oPtt4vk!*(d0PB(S1k&>*68rJwc5J?1b&QVN0>14GaP zB3tuV1HS)bR4Ia8qbuMnthBWOzA|Dl=V~W>GbyA!mHS|wNPSvABG{)+L!|li= zZC>a7+n&IikcV%5CqN&1v!1os#efhEbG;G9=Q#^CqL=W7Zl`yjKw9zWg2fJ8hMaMKcryT<>c-ZTFx*0sZ|X(38#&oBG2BMJq)VlpM*tM4&Ru0^$K(iJ+_u&@7izF15%n{Nh?h?<0kx zMV;*9i(Zp7I$OBjB@@iArW7*zacOMj`OL621i5MPLEsg7D^Z}9Nw#U*M&Tly7SFv1 z4@BCF&`6v1;|Es|X`9Wv#2;!GWIk>{gQcj64arvDgkz38^wl_}@Z=D<+}}!ofLL!q z;{UdZ`r!+IKn-YimVEK$V@T}W7r1b8uX256R-LVJQk9aAypXgxQ4RS65}6r6kQ@h5 zi*_7s@0w0TR>Fc&=;`C1|7}k)2h<#o9=O!uc^3P7^RB;2AQg@9d*~6h7ZB+u8|-zh z_6&?gxMvV~of#@@YTj7py_9Vh|J~h6w|m*3SZd`w1A^_4U7>i>=$c`gxBs2{{ut=S zf3v1!{%bmyl2i>X#(f_QeM;cCk8%gx+A4RPLHE6 zN!MTp`58aY+f%+QoQVCu<&1^y5?S_2K>o_TwZPmBXHmSB#ys*m`ET~N0PBr+RN5XD zG486D5E@11AD)Y~6TKGAY_1{CQp65hYpPdb@bx(ot7xlv&mG8qKImb9Jz9SYqa>_a z@5T?n0n;X-P#6QH`^g{VO4r|7S<(s3fVy`Qh;M?1JtyCl9Y+KhWP zexVJmy}2^qUFQIRGm7ulpJmN7cT&isyCr;p9)qUi$?mUI0KbvVjt1g+yBYfFKYqwo z?EVkqlU|7rg7OWP=bx;G(>>_p0!mU%;$|p zr7-M3;}eR9wWwYtu(7puzFdtX7O;ET)YMdd4S7w%4vhO|Odno-Dbjhdhte1f3o?Z* zcC3TUQ`Yl?Q~JLVD96rX)Jz=m@ecjZWvi;C^N`unlxkCBm_Y`{&)GGY*^a|XaH=E! zY)3`~L*p+*z${o+#toUOFn4XA_5>Y#NPM3W6k_k`qzK^TU!d^J4hq*0(2RsaQNXa4 z^ByAWP@}W3R=1H^_QI~;L!y9GzF3#b8)hm1yiJMWF;zVUC~S z;yPaj=gAUxFXtLZXsX%ey?15X+jT1b?^2IY!K<9=09{Hj5CL0HtYQp?W8wg+AC}!} z;RS$i<&rysB{>G@!RVBfJ^-=$36Wy`s8twubXoiKV)_wd#g&U9u{ZOCSO%7z0wJ@x z-cv53cp(pArdY#ASQbLfwAF0byK{c|ZyaA0I7e0nNnM;TS;D-v@;`oExu`Lv@t+OY zN_>>(V&w*V&V^p2RG{&tV;ROR8*{AnBli%vMIJYY;y`6e9LSM@&aAOx;}dk1$r53_ zJ9C(V@@qHV3F!$|8M=+E0u^~Zc7*APa;^aDZ-t8>l3U13-xhZv@bH~9cx<)-OIE)j zyQje$uv=Y!U<#74aQpv{3ra@3T2sUXq{n;O1wwx8fH-%y)={4`;p3kl@PLSV)S_0u z1LOWEr@4^FdjhvVoNBy358Bbg|1Ccf31$7?8nVSb%;Vx@5DplyY89F;UCT zhYwYki?w}#l>`@(ofGl0x)4jd7~58w>51pz6I>s?%VY0ChVv@q;tA*P$B!-E{pY@q zC4s&9nfw`_oV@Sm^7s;??)SstOAgBGZulcL2yclQw+Jq=CriHDxPey&+I z*d(}u{D7wy*N?i_uU;`oBY;8~&BofEN^PqyQYmZ^2=MB^)%opE*AF{SyTlFccWM{Q zOIj6fABMO^6`PnQl-dhG$f=i|e{~>T{TK0cI%h+2DZrHy4sE<zLHl+Y8%h!MOh8rt+g=J)Hhq?)S>Kl(}kWtHp*2U92q4 zX52PX({nS^){l{!u;WAi&(zm5Gzts^1qXvKIJuws@dlN`X#wWyd|wXq_rZ9t+v%t> zbSxfpcc6@JU?B3cjm-n*g^5Xy=kzyi_k;Dkv2R9S$Qm^5{V~YZmTSPUA}%p8uwGpf z=Tzu&hBBl!=EB!U47WP)`16*q;e-%{aTVrq`_>sIY&l;ad1|r=`}hQrtjLNM!beG= zS^_S{e=a)%R2Z7393W@!2M*Qd-rv<8m_nAw(u77#*hO)pkhuh(3btK89pv`El05w| z-7a%#UiZI$@>9pR(W_z^P>T9X0a9G2bP}q~2vi54cq2GO#Ov|wjEsXB+krpD`r;nv zwev6s09(!pxZjdx>GARL?}|GRKaBEw+#XJ+C2gjs_6cBBS$w0{c^*^XmvGi8gHo6mfhZeY7`%_{lMTXHKaZZgc1$5HDKBNDtNS(pyey-|g+AQ|e5mv_x z*Kf5315SG;=x66+G$ZP^HrBO)E}ga#*s5@N4Hg!Y6M-WH-PBthZvA~X+Ww|dX0-OUe0WxXBr+V`%ad4#ujAU&sYb>W-iiHbP=0r_ zDOuOX3Xx`^wwjs1?>I0B`6G))fV;YCLC`t)hPaWS#6JD8wk#MC95AQ5e7c5|>62csj2 z)Mo8Y2yOU|(7^eO`cCiat_}%H-|LdXceKAzIo^!E=0O&Z(}(QjyW;Ygq4|$cxy2VJ zczWDF_bcG+MWE)TJzuF0^ZA(dUo-_H4xxpH#?YTbRr{~}(Zr{Zu<6F=U z3lA?G)yK^-ohv~JYETnjyp}pxyC7jCNkUk@ctgLVT;4%O8q{Re4%$!FEifMOb$%@X zCWKTGGn9USJ4Ke_55pFF@u4{&tZC|&v3{m9y210Fir=ul7_^#%hF0bME8o)2WHnSh zgbRUiPXPJSB(SlQHTF!Clan)mgb?IpBYbBIO+I(4@$p2EsHmu$%H4@#A84bn>J})P zL*_>s!1zY$c3N;Up;lZ{lqIkH#he~UePP%^`@NwC_RLg3_X}p5P9>y>NjNDdy0XaF zkrl$D0I%88kahb9IsRuO*=TZ6(>cz_Il*1r?{AexfRpZr5JZoxX=bE?e!Oy1qIAIu z1YC{}NaH6sKTkq;g_6K!0fS8ts9HG~WXfwV26)axHpks5d@qaH8r@{0wgTk+ z?}pil*eUbBSB)F`kPKBk-}TFG{X5e8{f6L(jSM9uZPkL-y=al*>}8r zvz?{u_eTmGzm^%Ne^rq{ZY8Mss&~J>g3YsDZXYGmSzTFS2Y}$lacK(xO!O*5R03oB zszo|5(D{GxzoV%Xd_d;V#>NIh#BXDuDTEiPws`8okW}sC-N4qGl8Wv{H4$WPzzC`} z%`==R>BEGas}wA{Q<&2#zirsG*oe@yYqJ_BpWd)S8Vm}?p{UV5%>(_sKog~W%WS1mfgk0!T7ZDj#ucz;YamAa z#R-Huzj5fdUnCXHJFlns`uSas&x64xW|bmHTu&R6WFdgkN90ufOy(~gQmJ|j z>*$B*6}z^3(tU@jo9a%)?BMM>^Xj4qadHJ|{j+_ks*4UGgeN_rN6l^_%)wpaO9=JI zr^2|7zajY(V>y^HXQRq#=fgI_YB0nA2w3Z`wsM&@sv@wd#jtR2n!(YH2Ykr^n3M?L z9rx#E08U=f@bX z0Ym~>fD0F3>5D-yhTDz~6mS4`5jOXzM!2ceBAv9%GOLih6pgjte zSxphfh~!$GD*6U3B)N52%%dAogF@GGRVLHeSusd7h|mxK+$dLh&v5E|x^sXH_HlER z`2I|#iBkO|*oC*%@dr{JHW|*29+mbC^Q=jq3E6cPepJ!yN$-m9XR#nx!t1gVZS8VE zCoe$TqgugBQ4a8%N`zAgU)&rIKnEmh9rxQ;ZLrD7WdMscRQrI4`{TRNX+bg!sFo+4 zTxu^+PX$I$9S|}+77|LP5O%Qwh~8na3C9;^!zto;{ zMozttw2o=6B$U5eRtl2|ZysWb{GL>!yE}=dZ-9l6&Sua(kIr9P2jjR#4}@h-^3 zeJd=0&Be+rGmXP@0wXkAu=G1b#_eV~Ag0DW8{Z)?WI4>?mKfmhJ_g{XH>qf2LY08P zXUhwe1#;~7@`O<4)TK8alNxv78;71zbcAXaomy_97D(4eY%Y9$tTr*oWqJD$=#Xx<^jqMaYUX>)wTa?hXGY9wTIz#$IzE|wT_)wWA z@ZlB%m6D9HDojH8i#vHqx;)sp{j%q_*I9fPTS4wROd;=O(Tub^W0KUHPbe!&KfR#? zcIF_em$PQQE_^u2=MOPx{LL>B*sGdev8)Js^xp>&6TXK;9_gOZNY8rug8)2 ze`n5u@3wv-1pT8kR*v?#E*!RZ%Wc>_)Ez9m26-qd2B zC>-t1(UIgg`^gP|Mt-A9D|x$Wice%>a*i9UpDLEV?GZ$K5kJ~gZeC-pZbOM9`lL9C z<@DI=FIXc(3zXa$-=DO73c4#tv-BW_v|qd<8@F!;<=|KVU*VbeWTOa=8)Z7@ z+`TqmS-o^o_DizeOu5GRQh-l{Cgv>=`uJw3I1eRsy&U~mH7*7IMZ{$HhLYxmVr@N* zU&jKw`t~}^-~0CVHxwI@-KVbg^3eAb@75MMbw>{p!XdlUun~CMjiIviii6LsG##{z z0fK7jx=WqUt_TZcVZIc8lz+192*iXYrcdZ^hB(Z^_dC;*2#cLgg&<~hTyf9#8&h&_ z-vZ2*N%80aEYPxqx(F|5J6^U+eKBptla#^Dks}or9zyhlj8Dw5(L+YO7g?9c$0L!v z96(|e8m6V&3(_Gfp{zBrZf~xaB;u`g&Bwj~b5c?$<9SzQ4?k>lO9-`p3&8#*Onoch z_yS5~>-SAyCJ%hxHbT3D&yHN`V(yB*f?KyV7mex#T6pdsT*<)CtA0Zxwl3E+n&#to z!)}k7#6t$xXmBu?wj^G&;^E(??#o|;zbLv-T{(;|^sPQtg6nx3DiJ&TW7aQ%3fF%G z0mh@^yNgp9v?l;2Ab)~_Stl}BC&SWVXmUSeG6^9{Rstwwq?!WMHKj-*Q4fK9HnvJ% z;^B%?J=KRxFf>-{OHoa^bYi5BVlFvBi>WZSO*hv&1L%D3MEyD(;hR%A`Yshec>;I= zZ8L#ylg`h$4nRu>{d|@J74HiG>Fp*oes?eq|H162??zg8Wc>( z^*F@t?9{$>EF1RyV)nT4U~;amf5A8tBU)JJIXF{J=JO|{^m>#pU~YgW{7XZAe1D|; zJU$X`^5P9=Chs_%eYV6qB%|~kEsDQwlR$1EV9H4N3&(>6xd_L%n7fq>mek6}=Of6! zm%u~uTbcUIVcUZGyEx$--CvqM9_;IH$BN#tJ2N4MY5K8uevl6L4mVNfL8aCr%MnGY zwnH1(4Pemz`sPwn zTW7V4^|@{`2M?TRE>f5n@kh=o;U@AVU8E8nG)EowTRXM+)LHR_t62ij%7khBWcMd? zg-HOE7ml2y^@JfsU z)Wv!jt<3peT>g^oR^DqXcc7g6arDba8bP|RnFD~7WX*%h)rHoK7GsD2*-5l)X@5Qk zBD~_{3TVxEB-eZm-NuCIa64_hyA#$7u5VmpX~r23JJ8ji9IT@+v{Y?R+>M4~9eD!2%$a=&@fLsSpPT>w1pTZCyHN_QqGFtoCpoy3`RJ42_k>?_E;j z?g)GHb4C_X$wfyBB+u`wv^B$P!|FUsB%b^s?3j6$?fyXsqx5gwiXAxxtP)LQH73Ek z`m2deFQ#+(NC&t&#SWrZX7SNK+&(LIs4#0TarjNT>waxn+BdZ41gM?0J_+Nd8O^`Y4!*>q;BO071KZBd>tY_s_xKPxpUAX6`6{k*k=aCU`U$<<)AbjB zeKe~XIBkyt$G6*lGX)$ey1L&h?X{J%-TPX%Z#?cb|G`md>J8ukYa>}iDM|pEUzBvK zj4nj1T%DvT7+#{Q2@||tsHd&;+@g|=*i^4G!|~3wRmrSf2a~F7Ul>mVbc^fQd@R>I zLp!$Hygmmnso5HT61i$CrvmZr={i&+9DLxw=^J6P^a}`LLG(n)Q!-u@R+G1nfRR2- zi=S%OF`0)mt%U1kq2}PruUB}KpMB2bPz}z3d&~*>WX3MvtF=@OPaQ>unRuCjn{$&z zHUB?m)K&X3Rc!SEmA0jqZifaM6!X1#6acA2T}}z@CtINY+S~j4@YkP0A29=9xq?QR z>8*!Ghmmsi*nWrl(l3Edv(}L(k8EwwV$WoXPox`XxKDiV6tm^uFB&w=-fo@nJzO*Tw>ezb3cNtvH4@nzaE1lY9;#ZMxv zrgNW>)CW!aPX6`)u=$n2CTvBle=IKa94&7ArPo7crHmelhJl(jR`0(Zi&3oN!rg+m z$d`VnxPY3+>9`&h8!K}aW_GsoT(WHdmMJ%YZyRPQm=M>|jxH_r_QT$+j)MTq;L zC6OiaGT?fORNhNM9~0nC)Gpud*cj6cEJM`gPSy&f`+EiCNugVS1B{NVcD_acuv>7qoj zl`{3!PbQylXBLe!8R)c&=7F2fy}uX49iBG*zH#rt!T5>?!P-Y#u867En9mZF@qUH( zFNAyh8)|N20d5Q5MBFO>%ujJO_iTw@#i^N3e^{s2Je?z(r)`LqLdu4_U1ypYyoR<; zxuYEG<|a53j>cq}&5F;8N~UF6;R&(!=QdFD!U~oz_AuDW2|Y^^c2g zbXW?PrMJ7!WBmNJ$*gTE5BDsi9xfB3efS3=6N-(B)zfI_ZNjKn$ExnTSNRSFQd|Ax zAZS7R()#0HQ2+W72&5xPoKSkK@@LN#bjo!_VieIMiW1o?gIbfxe5yqi@_r}k24Os3 zemEazMr)Tz1^fMclFrVf*QJwvt#60}dH1tan}l9*HTc}NDYnP!tLa;g;)+iP7deu& z8^s0qOIf_eam4=j>wY+Y<@`JB(Y_cClhJQCJZ8)U2DZ@b1A&UdG87Jb!!F!|qA5&;Jp0vy@U<}}0A zIu4@)+2OoJ2ukw^9f{N@T+m2s>PuF!HUhOp6~J$4Y=b$gCao;9mZkusKWm2>|E!tppOP3ghMukKSa z1XJSeLo9TW!OKybSoq>UM*&zPEhEF1CE%zy(eBp+3OO(kjul?T<|TC60~!l@vNrub zp1_CQHG-mds_MvNfsIY>kVO>o`MZmTs{kk8*I!1xGnF7ZIUxJ8Y)ZPd-EW9_N``BI z?gwUHW4Sr&T$B)mF)WlzoTp>%7ya1^jEW$dVR}dBjA(5_16ydQ`n_MkH=`h8jzIzD z#DYIxPE4kvJgg;NPk19$Ot2H%<3kmt!NUS*^>5z^Q zT$iU5W%0BTnF*%5y9jBW8IlXqva=*73acNPTyIaeHOQTn+D8 z3``8&squRc3|q_{y+i*@GUfy=!v@*XuNaYscIHu?->Z0N#9MLn%9pv-2YzD4{XO_^ zMS+)QMVdQ80Rwm$2iiRcJ3WYvSY|B%!rWn>!ciI8PGonq(7k+UqvQG9M6hHsVeeX5 zO&%B~72v{4TiJ2aXIta)8>4vwhN?>s@F*59w&4?-m)(_7r767l?U8cp#MUTg`kHSy z&883Ef4xma6PxmXmS$`B=yoY;V>SamjR%p6fYd1cO8M^^y`d+mhfl zSH+Udy?@CwE{1`r*NfKr=wpvT(LvX8m{7d7X(z<2YlT{sdE#6D)Fm#1tulHGPV%B% zkmPsc=ry!JQ&9!ur%eQc7%2pQ>zhvq0XzE8Q)ej+Jb-imff={RuL0*p)f;vj^Z}ak zWZL5)xSqZo1i_;!yCvatC3#jxRZSyfjIU7|w-j2oV&}_=XK^JgrB4&;z1);Ht<%v} zmgL>Q0@VlJjJYQydB$8Ka;j9Az>D{+DNZ{69F^_pbgF5iv3ak71LS+K%*t`iiuJgE zUOrH`0$5WOo{mKN5(8_t!HgQzTG3BCHm8C>TK@y%nYLZ)5GRUH{#D#<|B0EA%zRVW zuJ}!*DzjK(yMIIkk=XGXc3If5&cyGCl6)=(Ek2)g=j+^^7CaoEe0w%B^DEKXm(1#aCUSd8f^Zfzd zSB0B4lAycakM{rIKtkTXoG3I4UnWdp=f&*M5k6eqJ9U|?-(@e&%8o2%LWKChOwn}= z=17rA&zVU8feky!r;=h9mCUE7RF&-)b;zk}NM#KONxc;E*)m{{(QsotERV<7Voh$jlIIWgg))YHJDxP#lW;Pd{(+1BhR28hKJ}$qZretUa1_dxWWKnkrrxV9? zIpEaIz<==}0xa+TI^Pw$r;SkvUD3*te>|1n-^?pIp>{*qdC=RR*HD9YxlR~rE=QGG zg9Bx$ahA1?Ck%wHG;-ujW@DH-sxzdp^M0jp1&<7Xw(Z1pEJsPz<+UZJPmCkG`kO@) zm<6kXZ(Zq>CL3lA(llQy2#7yRcQlg{c+-w+g!*JcuccsZ`|mbAF$OqR?x(5gZFufA zK}!S5%PfqDjYNb&%ZV5hg1 zA9a5>O~xXC;>92owy>wXY(5)j|0iEH!Zwf=(Xm*6lv$`F<<&=SqL26_GSE}ZYa%;O z$MVGv4)?+G*sJ|>9dB7caD?s+P_Ij#g&7yY&zz*(lo7ajP+P3+#mSWI8}UfJo7|Qd zoLS?NzV>h>a51IK=>IM@N2n)OlPEk;UmfYPpV*p=*DD)tajXN}ZykX3W-}1>Fk|zu%+%vkHNx`EOY-8^< zNplT{b1CD}Q@dk@xAb^r?Z-A!(pug!bnkFSI_ZCJWcK)eLlW|1M-~(}GL5CBYTCJ} zr#yDbUIgQ(zw8_TTL^gFPY-=;_ga={0;z&LP1v@8KIJ5af*m`;@)!5aprCG5K2O46 zf}I~_t-K0MG_7LismLd03ZIR;v>BY18y3#;OL?|^FMnts?hg{U)ti#Tiz`8}^fA}E zl~Sr}>>^Zc%(GJ6t_HO2L(81JvAn|SgpB^f0J3gBp?Vyn-8BD|MkG!7rQS5 zS#EpBE3y{|VVWF+sUw7z_tO^>6fJ~;b$!1w7Fh$zkQWZeZ2~c=#MJ8 z941g*eDTVP4_bh8dHrueZL+}EHYReI3vzkP?A=$t=|OTimB}b~eAYJnjmCK0W*#LP z-gv!-Vbc}W5->6tbqFv*zcODm{ql`GuRncO)c@1ocLqh7b=@{%PN+BrP|#6eLFq1_!j99D{`O(j!cR2&jp<{8U((Jm zy`q|P&o{YPGSf^TsrthFvDyODEBoD24JEn^G9EJC5^EW=nXHm_L+#f~`2Z@Z)32G6 zqXSbneAy@DGt+;6XcuWZOusmk{^u7!XPtWeKXaFC-T6HOc%v%|?{89lC}sWUnFOyPH{^fi zed$t1|DE3v=*HtD$>WfT-A*xzd*rXS1KgcA*7U9F5 zg7^pvo`sp2UYv)I9gA@Ij`p*%n%Z)TB8$mzmQ|QcI?;@&3ZaB^oKCh8;PaW&6R4oY z{o?bXXq|M57BromQNL_r{@NiXzPMEvAl1(T1L247ZS&Gsizw)uc2*}=wpst$BW1H@ zX)QXgq49t!y?%W|hyh;VHOkZP5%TP~XE_DG&h&@;ud>X}X@=XGgVS>rfaZ?4!@I!5 zWi?Il$}~*ltAD+@`10ma1&XvSA!ju)IBD%`=qzTZ2ca`bL{+CciXM1zcXL$Ij7wQo zvkwY;6?Y7x5Cktrn)!y1|KRaVvwAeTaf2Qndo`-yElfVxA+?77(3{Vfs|!bzIG6MAwCgIU8w@cj37zQk z(qH$W3}Ty$j{lQanKf{MUJp{Zns!GoM&_P8+xyt+zV>coh` zzF*7lw0@6EdhMvfTxnU@1+0U@9_Ld~)V{m*7vWC7aqc(D81FlJkA@>Q%ombI1^`wH{qLNSHTQThgI! zwsraPHDvZxbf*D92pyXz=HhWQq}9O zIqfbiEHr>^?8ir9*UqLooKREScj(YuFq5?%z8Mn5?M#z+jnvM*ro!aTglgpukJ$@X zWkTyvty3obfQhDg*~^_K&r>R*Kx%jb?sAKFB4PM3>FhB@c8N>f>>x$R4}!Y7l3Y3- za*)k9)E}6LAE#g)43XA-`}co_pQ=U1h!JR<)YHSO{~EjSvp6IHw~NS^f*}|Kla2ZF z=YwJM)}rPyOy+G6ycl*err#A11K01p8c<;k73v#^Ja2LaX`OA-HKVgno2%8S%Bq{w zN1Ay(pt>I1>j8=IA|k*poH(nh-&*U$hY8L$`9Ilu1pd*`k<%vg%5=ORhtL{_Q94@@9qg4Wx}Q$))K7U z=gu|GtG298=nU}L9euaTG0Hyd*5lVvUfFuv)J9HqS7*ka&tC?1yyIyDc|NWZMYbHc zU9WA%U>70<5>jcXN{c^zTmWX_Kj+zVy>-`~i#q`(sNu(l+achCs1MR4Y{cVS22EIF zb?m-c*Y3}_BPPQqc)QKZhXw)mY5N(ANI=#4_VBRyUON?I!qs&N6~>AR!H5 zmr^nwtpFKW0HhW0>U&{Z5etqqEPb&Mh#o(2!mG6H2ZSx7Vs&tb-dHmZU|(Fy zWk^T%Qek*9uB!}vzoe!IXAT`Yq=#dXEGI_>8{o{`h6T@}0Q(h=a4-iy(Y{l%+&tS7$5!kYEr;}ik^ay7zc#uLrTbQC*mg$2lc0P&B8H>0m z=x^fcQaq5(`&ax4ca=^%Q}DiZ;jUbx+7%rU;SIZ$Ao%%wG_s8^F@2h*!1JPtnZfSy$ zKEvRt^(nGdLzU(Ay=xP_^t-+%*9OVEjs_S0_;DV#E39FjT=Kc9s;cQ2jHtzN^X|!$ zCua;NvM$X@&%U~-UrjOc%&r%2WJ{|QMt=CRtfT2Hmv{pm>V!QT5~n$8 zvdvJ@?jt`gn$r3_PR0W3Ea0MhdTr+GTkWnU#;Vn-&M>O2>(l_54w>atxBR__m%k+N+*b{mYXcMl6rt-)P2wq?>f zHwARMuAK2=C(W{`_+^5S7)(nuM%y=20Oai7dAA<%>KVP@ZNeB+e(EMtVH}-;sa#4O zuW@hV2g<@VQZZHi*Rvdt9BbMnEM$7n_X2sk>mpC+Wi?6pkmolR+IXpn=fn2`79u>_B;I%yB@ZaCP7&CuBE zGhsA1>$rq~v4 zO|-(FfQqjLrMEIA!(V)X9^yoyU+ABPMnJy8UdA|Ak92|x=B}!*lgpQ?M>bCz3bxzN! zg9Z*e{Qv<6%d^`i$d8tKvbc9@2u1RAy_8eWx&9*0sT8h!NjO6H=>dHrL;8MtnS&~Ib=djB!k(=Jrr4as7A^^rR;!`TIFc(P_^Zl z${Z)0zBG9y4qTsQVy(+&vP&IOC%hA6eOu%qR$tXQ!w{?KTg8x_--8FEu7;0}2||WV z%h0{awfchiy+@Do;pvtKU%+s0tQ`R^xgLtk^5z|dIR(96_!evs+ zQ(e1~LvuFI*`t@HAYv!s{(F(y&3{I@jsoXGHJtw3{gl}gy5bx>KN=_HYOgPcB{$g?-dIEO6IWtyCaJ9BkxL1 zMOK(3`i@E!G&Vg73rtGI>c~-3eq0 zp01m!A=~9vS%z)oQS-8W_VR2ZHO-z=h3k*Y5&Yy*w^KV;k(#R*T`rv=B9SiHVaspR z3wNlNJ+b)nufJMD7@Q4**6U@z4F~QT(T*t5u?NkJlJP3UPkAu7t+&Ic zPPx4V$d=HfClu)jPU7ZOW$b>OljCeJ2k0?lLh&jvH8(f+r6QvZ0s>daJf5bG^9Onn z9NpU4R~F+Mul%}3I`IAYdR03+ahw_Uw+sd|%#&s`rH73MQ%x(QnxJz!bgj%F$NL%Z zk$MjD+<}UrAFX>vK5BM4QCWU4CIneyTf6PwMD&=_-^zIoV!2 z&{pbM-qrEgG)h13&U(*o{{aM$@eTkB z_mz{17D9U|SxI4D*VgO$Mp&Shf>4Kmd&HRs;U{I2k00735tvFdAVJeOJ7GXdx+ zudnUoBo(e1ZG_Xzn-`YR4q(Gu)8OkLMee-T5NG*L)uAyJWf z68K4EUugetXo|=}QGZqVz|Swx?q{#%TUDi-_fsVF-_7K6{;`xzy$z|~MJkFDcJT&_ zW^&3~R2a)qL8_j!LBz?g*2Q;v1Q!Z-5r@_f67nZrXg_3-gRnt2Va zyXtjDOKvZ?=B->-da(E6ElO}DI9U$RzkI+5DpSSK!CMt7slw#kiFET`6y`9WzDJ#9mt~1>=Rn~%ajx(BU-zd12qbD+hnQ=_L3ga<2SEZ~n>1+7;bv3VT zl<0AilVgwf6DfUGwFP~`fd%W9-~rL}(D1&1iQrPx5D#_DsYSxta{6r^+dl~d24(FT zgBgpq826>|(^+P9PwbPqqzTrFF7OY>KlfmTlvn9~F_ket6dy^oKZT>T%lupHhq@;w zJ@r}f9-|LxQ(KzNGJYY&GxK9+dWM8-WcOigL0KljCOu2}6H+xwNhig29(>aT+Y)C;hyl)B5Ja0#?4weT|&cC%W5s^2EP}C+dnVTJ^da17D7+ zOfHCUtRGfWse3I4ayd>5{*!SCX}&}lo)ru@b$|tV_@KUR~xA25DP1X4Z_pZvQvy*H4 zcci5vjp1angk5Dqx!T>Utw`$I+InFM+hk{Fhr8{bci;WcAUS!-A9e{Ii6}WvE*4$Z z!9+Xoy$r$Q*a5O_Uoitb=AUZU)~Ng6*X2j|eP%MDEQr8!tLEFkqc>U2M9aDi%Euu2 z)Pg1*pZ_SZcp;+Zu>I%ax{${-(%5Jb2de%tfVk>?;05(T4N~P(sSdzA8j+_z;E1Kvg&E@nAshJfmzPJK5mm<( z6_KG652n!ciQeu`TwN7aJu-On=FJQY^vrgn2GZMU;o}`@)hSC!R@YIBcjeHPx_{@6 z)#>V)1tA0x zxW3^?I)@EHDgx(?jXA9V7lM0c3Edf5ulPv(;xcF>cy=E>h(kxEW7rzLj5N7K>S$?j z5@J1STn*T0qDVaqAW*RHz=2qNq@Z#@E1UWP(V<*thfaJXuh&TuwE~K|e;o1ZLn2o^ zIJS6Z@l;|vySv1|Ao;1^{rjYtw?af)tJCFDRwj~W3?zMYR)}3t4Q+O1X2-`irzSs5 zCPtAP*dqWy^%!y*(rr(b*?V@V7b`_dcpdd0cRJ9rowr_G&J(ORi zOl*aPA3pahdw;wP^8FS|>KLyQe#ar*#C5dz$8@=)7^fKL55^DMgTPb#J@1Ucc{dD&w+A#L*Aj&)tXpP zBLI@CB`wyFNn=zO;ED0)h7A|%GUSO3@r}(sy?0W#iKPZ$}Ov z<|s}RtaIyML{1LVCQ?@>gZd@Tio|!h&|h>8d93256+Z)p3AL!j$LL?0g{70%8777t zczh_4F8{)O&z?Q)trvc{wUvc&6PwCg9V@qI}Xy_6OR-1DnT_2<9$Ca3;k z(E)w^_1r?r=TXrjX?t3l9Ln(sER5$jMH+K>EjO-T|178T+QQX)G&EK~@Vg%KnCB3v z(oohrfBrlSt9N7cPAm5OJd~Fy5C1)cNBX$V?~XgRU-I z{f?k1?i?Vn%a6p2V@=A1QTy440KW6dt)3n$qV^r5pN%=yje@>ym(3AT3O_yD%i;9% zO)e=X5_g-`iXL*w$;t0?a=wJB@bkefanIhp;d1^k;WmI%Psq!c4GyW-eSC}&+lf!n zpF4KM=Cr&~2r%tb$z7@Imc`Fz9;A35*dR(D-vO+v6mPInL^}@+LM^ntkZWe?ZTl2* zm>chv9bc(~g@4v0EqJ={o=Xo#9t9VJ53pePnq&=Q;Wg{U#qXDusU5h*^8s_}+j&e` zskXSczJf~UkdBhF!uYrw?(O#N+XpZ#fxPx_%sU5Xi=pRrl0CV^4_(@Zk{3a-D0`i5 z8bS>P`|cawKiI8r9zcsCB5vKLVM#c16Htazi85#LntF7ghzC2q9B&D#1r8vF~6ZCTWvi^np@eNTcZ9++Bavm4*E$Cd1 zhWK=llWmSE1xFfEl%8vf<5|>dNQI1aT2c#9)uoxG+O{aO;0x6U1JQn9&%gnX;?Sxn zcl(Tqp))#ez56K`pcKW(Z+wHgg@p(StvIRW^v@nRFdOem?BYzLlCP4L@YO8Ax7PC1 zsnyfXo^Tsi2Zx>b>GKWmu8?A*+ra08&e( z&sZwkgSzSjS^gwDX0wEmAZ&MIT2VdYoqEslL5qkxK@m3(mf1aQZCH_FrW7C`d!9i{ zNiY|4&)waZW{+5kiikAB>rWrTFOU;ag`UPjvPe;hVqZM_xZ0e`lR87Z%5E7YK+7{9_z3_ zEWuPau&p$-f1p%_G3bKA;dq-0wGsA@$PyvvLQWK9q%$%bthSxcTY?dLOe>qgEfY?e z^)|_H1YZV|VglK6%%Vm%440}X7Qx_W+mj+9ePY-8i32rSlIUeMqaDc1=4WMM>sSx4YI@p9YTtZ&2Ltl_GnV##&yVy+(eYL2gO`0(Bz$QWin*VyMS-h z=tv(6RLmqkHY4y<(KQszhjrBl<6!q3K-j;5+>vwGb=L->B9Kz$Zt*pS!}F*}@?~P| zn3N>7)3WxMvhrdabQ?EryyNFbU}|1TNwiJ-TD3rT^@}5YO+quZ@bD&^mPl;AU=OoX zu;xGU>)5w`@`5D^Y3OkejiG=ry7wBo`}h3()EQwmehuYl>u00Wa=YBb5Gl5}W$kyx zW;iPLT+`&7v-k&i0<=CR2{q&opqXZ;dw`1liLu}_n6oEdCl-mkKqRCrhE1W{`PDec zIfq~8EXjA_kcsA-`fA_1ce8L<5SL=)Dvb2uT1M6!{)k?d|JE}HfQ~Q$?|A6Q5n8Op zyUX+B#(u_8+z{;u>?PF{x(CU~V7dk0SG=Rp#~A?-#DELU%jeJk&i57ShehjWV93&` z`+j};s6f_Dwos@%i$ruOsiXpPSVu1w@n7KCkbO;4HZ@0Yn-u_YgH5s%axV@h!gw*f z&P)O(?ga#VM3w0h8p1>rfn`r7$8m+3jNTZ#wv$)6_;n1c(Ws-qALt+=IGGoRJ_0FB z6(pgjdO}BMF?00%+HMrG1`%uf3d^r*Qs}I~XaX58Uc5-bx)}*c53)cW$mjZVJG%r+ zYe2LlP}X06?wXMEY@|@*eO1S*kuG0x1Obg5I`OPJQ`8Dk&5A%q*<-12|3rn8^`{aH=6YgpVD* zrm5(dFW}$vVWlo5qjWBnPfxu#v%NY|6noVGurK0teEVeuEx%B%J>@LTrluyjo=i^S zu7l60iZ^xqlIwTxUL#iwDrd?vLx@CS)-_lS19v5yApFpqVjk`e7b!RF z%&L@aPoz(5#g3tnn#6azQYDGo!Q1rr_>Iv-(rC;10ZKo=Ag3*!?8nCv;WeYXBRI`B3JMCRLDYY?k{EOb_k|QAGHpW@ zI1Sjbc>A1$P)+^|RC5 zMr^1_=r}ZVTiatX1oS1v#l@iGbF+6TVq{r+dq(U84}+?@yTV-+n;_o zCvCWH`I9NZ)bs5$TuiO%CTEJbuhYt_oBGX+4`%28DJsuSI zhH|me*y=A)Ku5FpEYSr48=(a8wSXkxq|}M9-pm!P?AUw^ZRlGafI;{=EbCCQRNAWy z?Uub~D>8PZ#eOsM>>fxhP(2Wi9Gkkw$xiS#b&H~A>BWHZk&*3Efy$)C%01%?(h*Yd zCMXm|@0@Z?-6Et$9$&t>0wQ=RPd>MedVxw8j<5a_P!PuB#?L>W?97g6I$v~UN3d5p zFlOe;*)Lzcibca5LTYW@m(-m{pJq2a#lsZ(sfe7XZoFBNRl3wiG3x#MeW-J_!6*`f zS7@t_Z~+C%wB*{oGHGAxpipMm$wA;7+BxJ1=hhcl>p4mr;OW$%LN)q@;2$YC9@{<1?;STa+Fb#o}+ z=*^mLrpJlU$CyloYh_4CNQ`;)&IclAZ&Y>|BC4RXPf?tuRp0R_9v9Zv*Y`ESoHz1J z;!sIWdmWuleWmge5D8}69bD1StxE=prxh(k{Di`KDVmwM2N4wkmkI$Gm__-5lx|=F z--*FzBBqMy6?M_s9u)MH--FSz3D}5>%H-rnDgLm}ta^~-<|Zw94I0c_=vI-H`tzq5 zzE}Z~6BIMP&6Y?tEL@@EF=EIr=K8fai8!eRM{KI4U0m6;fGsQ#%U{@sC7P4*G2jT@>ufxQ*_ErX@ZT)1>eo(X(&+@_}E3oO<^c z6`|WCW?VeLc8lG-DCjmrWrKE#-6oDXZvBwKf!AZp?;#! z|E$AmbfG2q_FlZW35U=y`)99HapPU)nz~9;q@+S?Q?Wj{qg)&n0;8*h1O(LQIpJtkw4)}1c0e`Xn zufOK@T=9p5Q@n>99=iP$({<(+6es>!x`9&n@dc@fgfQ4}onb~ZSMc!_?_YV=PIqZm+uQz4y2G&3w2S>r(jm78+EA(i;515YOZ{+rkh}`9lkBv zMkU(TA3*Cv(d#~NftqYFqqppEtBqi>aVN@g3V6)(m`17Tz!m6kc~n#|dl~t8d9SUh z$T5yR+{WRF-}(e9RH?#)(mz(L$Z7m=6RU(5M3d-cHY{yx>ofDD%Tn{M)D+D4?ur3E3N1F)foG5$HukIRNUVxHyJ$nST$9~C%XmR8eK;*NTCbpGl9p8Xoz&+#_ z6jThH)4{b#7Lurj&1cNF^W0M36A$Lt+p6P{X=wof92Ir-9>`eYYk%s(3f%3Jr6|th zc^Ay9-ZqeJ3HItjTw^aIR}A7n?>*$JsvW`T}4E!lnOBvKiNK6IVG=EJ_ zLZhDvNJv;6I9#={I~MYY|ARD_n}??W%$-NFj#2D$9FTG&yfk}<$#qZEt}!*5Mxe)? zJ4=FGyQFbENh3LAA2AeY5GU>54K-XYy&=<=hUj2&7; z_{)zaOwoh_hvyCml0h-iXakWncvM9xVgZ5*DdRMQ*6i8#dRVEQ!!Ga>l6$=AFKlc&byzifO8Lc#Rk|IaV{e|wjP*fTzU V^XIl$`;LOw;r+_{qW0+h{a=r9yFvf} literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/repl_scale_llama70b_tp4_p2600g60.png b/blogs/deepspeed-fastgen/assets/images/repl_scale_llama70b_tp4_p2600g60.png new file mode 100644 index 0000000000000000000000000000000000000000..834c06dfb07a507278422db135a02c5434e27448 GIT binary patch literal 28961 zcmcG0byQYu*XK=#NOzY=Nh8uoNlBN0fFL2=-3<~_ii&_BB@NP@3I?DeC|!bdNF&ZZ zzTf)3nKl2;xE_~}c*lwB+WS}g#A|Ebx{ObYk3yj?tEwpKqEP6@C=}WwTx|G>hbUtf z{2}h8WbCEqYU|~D&%*|F`<|DZldG5011n}98xPM1t}eoSqI|-<%=TVhZk`hS{LcUV z0zOv{JN}P6&7a{z@Z41Hc%o25_mJOcC9=g2P$;)#RmGe7e%YJTzQ*Kl|KjW%F)pt% zE1^-aC$tZk*H|kgwfE)NR_7@f2y)Y!el@m^AGOrr6+G1|s9}4|d6+g4*q?xo?ve3) zmg#Ky_mbRDkIqchN;8g+kL!x?lzULt(-)ah#OR9f>lJlo4F)qSE9;PJmL&=N;!4D$ zgg;Vy{(t<{5RFxp6deOY_`FThcS|mfz1*aJsKlL^#HNw-idtUi--`jl$UQU;V{xak z*9MQ+@zU7mb3-pwYbpKzb^xi zwvOi_i7ahwqJ?aREWeiM#Kpy_jPIvg_$s7OUNc!`C>wJSbo|Ea=kMR$(}SMLV>Hm> zy{?;G=e{@)up%fX7W?E0mCV_O@Kk50>=)C+hry0zLZTQTv^Paf{yRX@SazWza}E5zj~3h_p{CB-nw_4SOOd^hn{M&G z=j?nLIg~OZqO+rIE%&LROdg#|GjU^MTH3-g?$W+(pavr~p2 z!gY)xpvixCQv1BL27|MNmzTHNXVXN)enhV7e6BmvZ)Xnet3!P%Yr)r`!|!%(zP@kY zGKCTl5fPmq&4wQDzGbSheoL4w?uiX2@IdBo<7|k!j?daxE?(nm)HTzFuRKh$AyT%V zvg5U6|B#D1e(U>IVLDo7#IKST#V8%H%3xYTAPXXKrudx z4gwyYGL8D9&3&1`LxM7+s)4+lk#3tmL_cJU3+t_j6fsEpSne+L-%+cTS5OcPIrn?> z;0yL_$k{cG+qY{TeBt=LI*Rdx?{3f1VERGwY@Hp^`T4nag^8#}mH--D)F%IZ4n`@z zJm0Bi&y|t#JHi<}MueU#pZX(lhy?6D(VKOKFu+3nD%k)_bE3(e4ILeQZgVrqpv>^) z?dR8V@$gi4{nN~^YG%t#HaPFDjqxB~Xc4rF_0>VtWws*;PTvw%@(sToW?E5)jMDv; z>$g%E`{8g#J3~Wn+_TIVhDJiMjd=HjX(6RP* zQ*Ov99`dmgzFUc@Y_}G6=6jJ9WMyr=3ya6DYPeXp{38YOr%~tPKSR z1t~`3Q|M+N!D1pu_2ozh3Q;_Hxj)d?7YBO=&(YD5go(*@$6r-db>U!r{ON1q)|0cd zhvgwlu+-JzhTW>tC>u*Cy8Rr#-sBqWCGX|b)YN>HRMv%p6d6)@+k>1&@AYpzA3xq! zR90Ra3b{D3LzCN>X*V_UIX`wsllzb(ZEAFQdf=#2dtX(j(Pdhr{#XA)D&!4KEiDP} zpYRt?RE;_G|EPEDk0-k-&GvwUxuqpcb#-<9Ds$;+g=a~0VwyUKqqmsQP_SXM=Y1SSXq|q%!IP1Z z;WRAIstj8HR-vMz()Gk5q?9L}9S^qF0Q~-ZV3A!`#)Inms^6^Xz{A6XR8T=dLBGR| zr+k$x3Ne-+pNW=ezeI)#qA{lz+s7FGV*Cby+w#2IjO?_4Uc2C_jWUOYP(1 z)3VnQeAY8CfK7MZX(ZEms)>YygX4IkZrIC)l9G~I#GV*xk+3Llo6~n8ses?O`zyn@ zprYw*JYmO^4m?zZ^Ew#%vj06*%Gt6v=5p3`XW9e`9_&k(E{QmPBM@_$QqdXTU3@$G z+JI+bahCBAKx0Xe_h4 zcLm+%tc6GGi>9I1CSnig<q9P)!VM6qNb*%$zvHm zM>aHP!EUGWWPhc1Y>c8emME6EXRcZh6%`f53@dZn69WYsnS_iiopmPLa#fg(%~|U@ zbW{6!lc0k!qwf?B4RgdVQ!_Yq&|%m0LuVan@peRYeR?5DLPHY|mHt&$0ZWwJ+2JM$ zDQU#bNbHvmW3Q1qubje|6idWpHPhzbVl8{JM0S30w!J!5_QWN-!+ElS%gS$W32*Id zX~7xWuZ4w(7m~hKBY8LD5)$O)w*?KKR+M->q{~GkReMt@H)3)c^ur_wVYPV zZuB>ZMWFSyv*?0^hINAL|R&!7n30#9$r(>i4e!_XHmL^YA-|noYExoT!Weu5g#8v zQ#}NCu;yMbHUW(g=BH1eICb(d`OMqmpn&&$`b20wkQn~-DNV@fud5<1Q`9&_jEb8- z>JL6W&m1R58CIDm8C98g&(1QrySp1#TcSO??ws)V;*1E&K=(>+XmWM6z{Ygz_?-4; zll$V%dR6Db;pQZKg~zTjQAs9(N}G z&6_vvf4t!$WRfX&w62Bg-UH|b(9rmmX~1~>u!s>91SNU_FrjfrkT`Ow;1=FYV~20V zgTmGes}DK)X~&-n3uz9f>tkiH6BEWz;s`V~HIJcXj&=kGI!!g{9BCOU{|H#LV3N=Hhr! zWPONmJH+<>t4);K_KY-O6SB_pz4w8|_m(6EHe`6>(qeJ7a-_II03Y6SbmW1q9Zf9n zwfVyUzElyHMVElY<$J9Xide5$b4=0#ws#b}>cT)0ojfiJ?4GNrIR4Ia-R&mx_JBg_ zXLQhE-8?Q& za+gIl?ReoKzn2ZHw|wIoD=3k-w%fM9Q7|#_!?NduMf1$|Ln#e>x`3b{30%bYJDz}j zcmewPo$M8w`vb!1$6*S3v~Zl2MU5tx$RKHG^mB2M4fV9@Abyup?ZdgwS4JOun78~#UA9$)>z-V=iGAftTv+<;{O@sx&!hCM?= zY6J`t`p|e(kv~zy(gD9m7>;21Sz^t%1}v!QaK!D&>PJz_7?vPj8m%#4bIeWiuVLZlE7 z7+CDb)Hg7I&(F_4ny+-}?)O)!=>VvJ7LW_lo$}s+Yo3tnYN8O8larHg(HYXR85V{@ zlcSdjGB!E{q&UC4d_^sT^CjKIZCBG-nZsbcE9Px}EGXQoSFhf)wq`?61}rWj=mTZs zRgImUop4_S{TQL;w;!3=zOsuGkj|y~`0&;oFFGaU^_V*)+(UFBE z9TXglRBd6qPn=&rziIztp(-zr-qYKw^ID$^U@LXvJt)Bey^{5g<1xhOk=U0%XFko! z%0nuI!u9QM2Ybuv0J@NS^weP;N=l8#vRWdoDE`dM3{8LwG}McWGgkoX9R2K*0Ey_@ z-mx^lU4-KfK(0#HOnYFB@3#5IR5KaUKJVXWht^-Q{&+Z%$EdQazaNW_j}Pt)JKXUvG#?ma`7gHzXe zW3MyqQ(;oC-Rk2!z1@2TWE5XMj^N1rLA}ZEt!X-h4Of0kNKRIEa^gKb+^M6=4>UGKL(dLAtp>75=jZ2JT3e^*w`t|dw2aU8_bZ91N(kpN zU|M&P?O)m^CmMChxbT#gHKl%Fe+=AN6S zpWh9`a^vu-j#GR<%e}Lqf0Oo>-&@1Sv{=JhtT5W1oC#*KdGb5R_EqC%nAAoy+=>sm zva)9ppOooYMOa~#iMY!O24p8!1$$ihK+Ag=%YStGd2^j7N)6xh3$Za;r?O z@N;V|25KmStDy1E#&==phn8?FpfWdiOF$oaR_bH(b$z@_(0YIX6ANp8zBe|&(}|Uh zO^<$ie*R^nv#To+iYrEWV{>yC)=hgYtXbqxk$xgVn^bweDZBN{{dngd0j&sbM@I)B zCX<+)OQ`49o#oBU%o65r%3ZMS`Zx_PEG&4P&DuIUSNPUzl^IsR62$-+00UNWACc_& zqxoZ$xceC(TquTHY1=)~mx^zEdjDSC^$d1=TvF2Q-5iXiqcX#?eqhNuFXgfB`Wu?- zp|HuOJ{PL5f3z4P1R&854XHZdz(K!MSIO04Z~beXomTrEvcigUL*(u4IosOXz26od zKI{AwVE4N{l@Tf%_R4|AQz0cW%!5NA)~6>Men$t_^h=0QtLE>Q?$^N$DIK7pp(zby ze6$yggW@-D>jH*Y10aE1kouHYHKo;eyNLQT_HWrh9>ekyo=^)@Vb+ra5nZy6&;|a0^AJU4}ahDWLeVO0R*}8 zt2lQ7pr$>nk%XDVjkNRGIbdViYq2KFLz$lJk6$XpvT}0bqM@Ocm^Sjk{qcTJLj8=~ z$;!%VM=EA$R`VM)=|n&i1yH(;YwzPtO-)(a+Q#JNF_(jqF}(|@k`eCat^Ndx8cgg>whn}A&GV4?#Y!4o^8c9}li%1NeJwSmCfU;T!m! zpnV(gLk!RbD*(_XCiUD-PEL7a4oD@YH|^XEcK(Qqs5>C5ba%7#`_G0j`TO~Gef)SC zYLDOP?^1JJGEg#hc6Kz{#e}%qp-R`lf-!3kppgCZTQA{K4pNw*x8^|`dukrTaaEvO zs5V9B6c^#pfWaRBw2FqVG12JC1g|U{f9MUgW%^>N%A|r)%ys7L`pu?*gV!@BX;lIm zYHGS0RIGKNgvG&n`E~H7ufLzG19uL1vE{vc;nS_Y>zC_b*A<@8!&?<$S7?JAwJ}lC zJ2x_PhVT;>78cMOI=km{;QWAkQvgt{of$FRWB>m30@cpL#l71@w;Y7mK{)CzmHrd##e*H(jpY*amOwN{BGWI*;3Ec(@D_)rmV7C>5C`35pv$N zb%gflZ!5N?fV^Yj;{WwFRLsEex;(O_L=V47kd$0e0+J}Bf^16 zAtfV=Tz>zgKrx;);4fyOk@92i)tPol%s~5)xKG z@dAb|TD-VmbqaZBdikihP6qy74F@@_*blYmSnI!U8>!IPj2Bq#5&q^U1N;8|z5+mf z0gww3s%Y?9zZ>M2YVUPJOiawx3k<*k+yRFh@o?ptfvp^@jVZXfRdgK=+W$S>!!$WoIF^>$D56_maeWT(0J)Jy6tY{C2G}xDD_DWNruM&Vm2{p!yQ9zi!#T=LNoJ zk64uwUNueMcKB9)2gBFbH*(zpZO`EylxKvz!a~-rv%L+777KRWz%x-tT>=UKVGF?X zahQiqp=NMD=DUjz&>F$++s3MCbBl|S7&wHlc)opjCgIJEY*>(}=Rh=yDlO$g_@=!G z?T2Tg>bs|EncUIvZ_I#ByIyVB%rUvSOt1AzBcLVZ*?xQ`AS~Pqy?qzH5q%`r^wn>>SlD?WzaUzLeu*|3 zpfovv^Shv=f$-4<e8d1s%P8=bOKB*G%M)*Lj+d)TCy~#J28*dv~>d}%^RHV zM6TDO?VYzyO?eik*HK;jJo#SaRsV?E7PNO#HswTOXp=o3K4|y_20jFkvwrs*Z&cgb zCLC8(dU`q#eaeuKkOfd5u=$gM_m<*!j*h5+8By8V+V+f%8Gs_C!E4cB=BuInS}7FunDN3^Yu(knZez- zi4-5Mf=}Pf)WGSf)hNn2=$f)z+*Mgj`vn*e*TTXAVNOs7 z6k&h&fxdH)JhXMuP^^5DnhYyP2zVx;E?-S&l!7a&$#@y4>Bj422gc}+f zNK$<`_Y1pP*&>Z0(03G3}rV&(ra(M@+qHlGyKnVbHbYWqkcG=hh zD0jr5;EFayDF(<$>YBD(sUOY&9+{x|LIMxiAV|N_AXTn@t9Xt}NcfE*{YRhz?g<|* zQ}FLp(6NXQ4-fMJsI&6&eweeH^_>mA3XOtWpF4BL@$c43GO_$ro4=mdug^+A+;D+w z>wq*>6MW{=ZY}irTOyyq&e9;+XqAQR{OT$vuzm2L8oI~La9JMnTkrv>XZ!H+qtam7 zm3}yR-Sec024~ZlCkO97sBOjEf3vx*wk~|NF*WMl6=CIgPSeea8VsoXy&xm%l1$ z)?7&iaWw@#b%UqLs!JVM4JKe5IlKcg1Mg)oUC6@izjyB*P#A;Xk7?n>8ArG*J3z>r zT3hWLZr*lLn_! z-{O<=66SgePUFpMD$ZCFGw0>-H_N~zE^cmo*i7TgI5hp3>!E3}SeqLgy6navSX9^5 z37m-w+k2oYTC><^kvj8DprcT0+ zOT^T;9WEb@QWf%1{2uwwFScoY#m>yV_HpXVf#greHwIt8yM;EsXob&)q<6Yb5F58qm*@{R2`+PNAU7qS;B1ZO zgu5G>EFt54j|09##7Eb8CPHiZiGJ?D(Q-|{x!l8M?psAfO_71Ez`ytCZ%*M6&~fn7 zI#{9qUdQnZe1Dc@R+y|1g`9oQ3rWx?kLJuw_r)Qi z|GY!a-Ca&UV=55qow@+~RpSir|GfTDpQ3opXZt1ROf^Q*H`M|zW+#`CyPJ>B%Au@T znRdq%=?AC; zNE=&LRyL7Z5FJD)qgQBfBiPC0&4Z>6P8L+}lP~AqrZJp`}HXUprCLy6P zhSl#piOv+ve$}Y|AqJh;kU|tA<3Mud+@KfHH_$9WOVmoF5yqwf&-CDiYMM4E?HPP` zX@gIHcY~Z#14tOL%{%`b(s5{JXMHDz`-j_=epPOHXwSsZ*0yVPv`cLEU2o)l`lp)J zh)51p#_rf348Y^1bS*?gpKA5xuNxyHAsPJ(d*b&mwj;A^X1MnkhK0F!;Z&chR?gU3 zS)uduSMk1xjEgH_ZV6$qpS}3FQ%0~6TL4>wMpF8X6(`7|(xN;}D z8UzBt5(0LF#7z8J4zmrnXY5>DrY2N!t=26WB)o6|?!(KXDtX~rfu0|el!S-^NZg_D zH5)4{w?Gu?x-fyn6|P4tKwo=oC~39=RZa zj_z(W=r9A|m}I<8gTg%iYJL66t_I!2fe3!|uLXft|#A3WaH|&ccr00HNz* zQ&MsO(=x7M-CT){=Az=U$+40)h8u%WEf-04cFZ*PZcylrY(eIG!f)|khmw|s+vbNB z75iS55|#Dz!Td(~ay!O8b-#Hfm`eBH=mDC8K?sZE7T}A6EondSBgpy8j7z!W}e+jj9c9qcH@wy{JXL$fPoy%x@F)| z^R+%2=Xz{;8RR>FKLc=hh_(#dnvI!Ru2B{)9vr#?xI#ifTd>!Az{&y+-V0)5>Z3;p z_-FlDp&!~8M*`S`?&5r3i(bl)7<%~X*V1&-+pu1AvkiHOaFi9_%|6dUMdPC-C(pTA z%H&O`bE=Yd{JjVa`upby`v_bz27ZfV<2Rdn>G?c_F!B3 z+td_2pfLdh=4)D&!=s}lVN$=d;1s}_UQgySZGh!2MNCWkZ21(nYONaX+_3EO47Z$S z(Z=TTyqu{ihtbLH$F(oOb+iKW63B`IsCJ`Zl^|Xu{8H|3(nQOH?1SI=aZUF=ur~N_+6Q!Y6f+A88kD1!y`}A0zV}uCC*as~Fy+6C zuAC@vQzZ`H$4Ax8<8#rvzk_$g%=mAW&0E&@Qn&Q(Cy*Z4zG?q{a{b*}pav=HiTp?+ zVD%p0a0q91nryfX?q@W+YKd-P0^j%7fbSlo2aQ8=sK#?e^L&3;7IFa8C`;HButCDi zdoWSpwa5cPCm=M%{tsv48iEyY2~q$`)*P()5x))`^}sB9;52CnK@RS$=T~1kj+d?e zIB2O+C^!P8RtfMwF%1o#q+~0BtgJfTw^lI9VnIlmgD^@DgfSr01lJmhC=PPzz(5D{ zWbpS%ECEdf)EmSRf&jqh%#H1ivVN; zWZwbj0}GYvn#r4YM`C4sfb#GS126qUYqIZooGH*Js&Dp;TRdmx)nGi^-~VFuDVzO` zL*~VP$rxUixF`QWvNKu|NAPCw3TxmCA)C(@uC|;yuF8|oO=Zj=Y>`^A{$hzmfOE z3~diH0C*@S4o(3`QK&FTO)bDq0s*(O+w=99L&?}0`7+Qu?9c~#Ji{sf=X&r6H0}J$ zNG)l2(Q#kZa$Y{~HXQ|uVqrljL?&r~z}@|qqpO4L;}2h-lc;@rK67ZhV#-{$UFFnD z{Kjv00ZAl5{f&VxbPkMlqSodTh+GlSVHqG8f#~bw%|L8xAYA4&-9iDnc#FRQY?tac zZ=%}UrJ)!{LAI#}whgdtAZ46u-zwho;Ks*szCG9`;x$GEom>G6`Md1jc#mBAI}duH z<|76@SjnJwr_9Y+o`bXdD(Z zEbtXx0V}HjxS zH`mTT*h$Kfkiz*j1om2Vu~x2CR~VW-D3Ope@$L%?nZsS7<4*N8Dh8%UavOv_VywwtCq|OCI}E+V4NX>5u(lf>kl3o$`~41D5($v zzd(#uAGGJ!(Q-rUKVpZ6IQjAwUI)nYqA4*_vuD(9cAQBA<}%Rw?=t z<7=i+%pxru2p{Ymh`Kms1VY&t7Z;xgKG(7dHwZk)Jdmkdgjf`FTHDb81!>5m%8cIk zh7TIypabOXYH|+?FC>rmB5zHo_E;X809P3aI>F@zM}ZQsS!{iM{ZvI8+hbGZQPmmrDHi6EQ9ddG3buO4xD(il3DhYW8v7)Bqb%_>_^e2tGZ}KtMUbqTTz0YdU?xnqM8MuM%jG?7bOS^ zrl#|dk#)`Fc$P!=eZF*2i71G$NWXgr0BGhNbvSTUbh5?1A%V`vr%}ndPJX)@prv5Y zu^Ty;E+*KXNZs0$3^~7n$flr|@Z3RAN{oHIoHp^!2F=di^7HtOuxw-ZN(!!%?D==&J2sk2!HYb!!FWwWa93S`2{!=#E+Bgn2VFuYXJw1PMY}P`!ZJ?$h zSrFsS5NXgMqG1uOZ8pMA1?YXaJzMqfzW>_WyE3;;^tBw1UXsIt#@_ML4M~ijKL?VN z_t;>~bY7fZf~v0oF&ACwA`o>UN*bVR%OhwJa%5qRjg82K(LtkPBur!?;)2FK-0s>BgCiV zn)#(8I1uOq9b8tPWG2~USUHG52^Rc77#hah#zr}^$n?9SqfoMc*hByBzU?Z%gDO7N zb;+}jz(PTonOe{)>T@x0E_761U)#x-vF5tA4ueJ)xdLTnf`nCu?>_wUa5=6>hY}Jn zIz{kzI>k`&FA1rlhn%MlQY_0n-RhR`S|vv18CIA8j~a0)WoW?omV} zC&!m)&qupQvZ=+wH$Xb>feZ4gZ$I0wGt+R7Y!{=7O5`4lqVF7H(_Y84{eXIH zDC(oAIHbqh?EtlB9Q1~K@??6n$Gm7Ee+(G$-&CVuzdJc4GSKK8!UC6Yn+ww|h8G3R z4U$=Yptd}4@NadwYx$1&B#^@>V|F*-G5~CSE7oc^7VVbb&zIKHwZa&s6X~dQGLCuoA$1R8GQ7woqNqs)@DPOH`m6(`yYn?s82%P zuP5^51Pbij{5%?vZzM(-&^UYoe#Ja^!%*x0)wX3)dDLtO&s8G;_U|8~y-g8r&W})q z%JwdmZS`~EB?yQ@HYW@O*axp48?Qvm z4gF*~rQKaPK|BvyDey>~Y?)wMI95n{X}N18(Tci3Fzw%tW|QzzG3R`vbP4z;BI5Xa z`{YMx)-aYYzD+5BEP$JjkLE!Z0u#YSS2IvmR8-zsCE3}18md0b;bB`J4QTxPboY+Uka%Cr9|qvCe4afA)f zREUgPzk9?25hDyJn5cKsMrrIXO&eX1TmZ<*|H|9-5&JOMPd7$JNmMQ-e!dmcIpu0e z3w$Xbjre-JW=+IbwQ~9ZR9V{D#eyt@#JSZBK{&z>JSO%#A&q;O-e|xrS0?JBq3EBv z&^UyK0)ZSXG;T@F+wdxN6%-{eibE?T-R#aCw1x(HLiV9@Csk0kIJZ|v&?cJoE@nPx zJ~?MXnB0-j#94(bXqJ$<=;J;-lLxgq#a$bVvOY|j-Kc|B>43g*{pTAd#SbR$-DwQ9kP&4XXVd%I6I$$`s zO$A^AAPW`TXyKqOST^0m!$jcA08=+@a2lNiMLAR@?bXp7y#Om~aCf&{CU;ksVAs3U zu7^|}&Fq7N6eL_%CyV)ieER0LzuZxVVVPhNW^l(aRt*E#Lsj4IhYk-d~s4(n2 z-69G#SqwgIgnHZzl=t*y;%|L5(X z{f}bRH8nROr31D+LC#eH0b+2>HFY|GlYqQvBO$}eYMt;TvsyRbl$(nK zNB-W-I@pl*%Lff1E-^9c#f!T!xmCsyqF}~FE<>c!?|6sh0#fu4F1G})1c_P--A#Fk zxqcVGk39UO21+)#X&>eUV*;unC7TTP)A}D~GE&lN2)ezoA4Ru43A&f{Ki7pbYje7W zjhc%8)T$d@m|h|yPRK^cxAKDwzjUSf$AA|O-srZUrXJQv z)V*J%`5$T+=$u(D#x^nYVf-I6tnxLKl?2!YV=&IuW`(+lLf0VfcXotg;V{L#VCG_W z0N6cvsz?GyBS(q`(%>-=&xIUmJmkyrAeFUqaPYAq3-p!r*1#WU1qIsZ z_gAC^Z4+q*68-<4^!C)!vXB&B(V80AhrVB8SV0T%7i<)@xCcA@PYeuv^z`*%f)&pX z(_AT_oLWPitOQ~$5P5QP!+jLTDJv+3lR$a9uPdC&-`Y0cOc5FbjCI^!^G{>H*As>J zfbmO{yY(wks!#}KJ$?N=jKn2VEQMw%6#p`~TsOfdGW zW}GgM*33}{_*Er&r1Ag;ld&~!quJS=e7l~8X>39(DZ#3i1g}?7!3{Z{$JzK^hXZ=> zP0)#SUa63P1cwQ^k)G|&zl7k1Cen-RA{1$_zUc-tH@?d0Y^@6h#3QN1IL^vKb+WBoa zYq(M`QS!K+kIht!59=m>WdN8<@z{wAAqs4tu z`PtD022!^XU-IAKz3^BLYw=!=U;p&UW0;}}igqCRS-rQrdpA&az4jkkWn)&**Pz`R z0$$IT+WvMe&q9`&Ret6%yXC%Z*Qj8iQ7K3MMxk*8$hZ9GXWN~SWkpQLgF{##gy3;E zK|c=gjK9aq+7@zIBZrp=r_%hg#3gjIv@=Ag~MCGx~2uP1=A=r z5O!mZlH33JQSjHM76}*|_LJ8wkKO!*BvB%c*g?C|?$)fVHeBB#Z6I$210Hh_YzPOI z1d@R}$gf$?pBF&YK3;xefoRo`8sh|HqLbsXPC2&%>J&ey{|RVx*HM8-L!MP}oPV{; zaV~+vya_X4r$^gWOAp}wMbij}ZB8yX8?fC%pAES*F+nZK%{oMwfLKDmV~F?6Ksxge z{2}5WT=w8gSglPG9UX~Hl?D;X3<&)1Gi8I01>UbPvkoP>Ord_%V+0%q=@~W&R6ig~ z=gE&?{(|?ou~7`98Fxwl$F>qzpoT#r5E;HPtTcQ2eJwuDqj195azQaq9?SAAq2-F^ z+z(OfTS|)J*S_F=1Gom0F~=B%YMF>HgAh%qsjOhv_SYZ*4rs7D{Ztl6K8`8$To6RI zQP8A#v9uI3OmWeizO(YT0%Mm8k{YSKvNIbh_vct z+%B-IODE#vYG1UTou!bV`yFar%=|FvNSS-YvbPxA(^E^yM6#@zV5VemWfcM4{v`x| z)UuP~IE~2UHS@nZG{>r?0na$e2`_59{yeVBh88SEc0rQ3gfucWvaX((nO zv$r}DK;^weh{^Vgcrybh$w(0ESM4FO+1k@Zk+s)i%Q}RWBa4H5IW4fdtt~A2(lV4o z6}$vnNnLJXxoF}h@Z^2Artpq_Pfa|G$y5EOrn(@F42jY8lx5m z27ALnLAwoh(vPL*UcgztH#n1?|9QU9f64z;K9V{naxg80GRu=25?-msrP_Zm#D2?QNK z0PBe+68mZ4`uV+k_?W7Szhwd1x(>e;QcPoqAnx6v@HozbD3p08wt0J6<6X0*$sa1D zBd-QA&JT=i7O_q`@`|dQ zZdy$ND{Gn!OyffCVFY?13}4d0uxu9u|7%;-Scl%eF=6IbGmn>+G1n}2a$cU{sr#LT zO-f?ZO5b{`?Z5G9UN%^Ido&f4euo5c@u`US-#0X%toa5Q$wT_~1nojyL}3&MC2W}u z`kUcFF&U z9n16$6f=mF6j5XxDD}thq>aCQ>*|Y_lk0kpr3i%|ABZE&G!|+qDPoNb%VNl0aE8uS z(CzdR$<J{Ef!gOC9|h6v{wKn*ik ze}$`Fqu_H9Fv-xvd}svJeOTmZuCA`+-1Yv5^%{*mP@ngL4usB}wW6n2#WX(^pz_Cc^|d5<;7R!Fw1|j0MuP2!B0>vf=*c z)Ej~ANcr%8_f%N!&uLfQO%D%)N_t}Y4@(Jgdsq7dszJ(*fjwz;`ui8x;T=;`$V}2e z>h;FO={i3!=X@zP)ZE++?E!?tw+D?+LO56pbjrdXQY|mNw$qv$ zqOknR$pOf!6OffbxiJY6OWa4KYDDX!k?{q}fn*Uuk)@lGh0#fb??RfO7sLZ(b{_y6 z2q!j8e5^w)Uy8$*98K8qj9bWY@zB0iVkJ<5o(>5ZMACik+R+Vo7@-uAXGa^~BLN#* zogeSQSHL8lp!b@7(C^V(T!(lp0-+Z{L1&wd;1FSO8&xI(I%0$QekCm`fCoEna}W8$ zK$p7Lb$|Ii24DsrZr1$l;tBe4SRsfX4#O7qqxo30BKCQ$2|Iubk(?`t!m(gUTR|>1 zU>FjRu0`wZ!DrXOh)n$HdUq10Ia1&yNZ1u-ump~!K=(*eE|^RPj{^nt2=>)Y2<-ov zv^g=j>A<61{iE!@>H7@TI~8Sz23h@E^YKRG!`TEJ`N3$1d@77id*`tWnF!Oaw!T+1_mH1VmWy*pHI+?Dn zZp-@r@Io{cGA{b(^niqt5*tAHv^&&@NOsj@$aVkUtguj$F+XDK)EkG_!klKjiV7V7 zh2nhO@P$%;!n`sxJkEU>=>t&14}u@eYx4CD8-Xy!$*EGRw;Tu9@7~h8)Rs*!l*swb z>5(TQoZaLAj0Lm`xw;?rPRghL!~2W|8#<;VAp#J!GJG4Q@`xQ$R&TURd22juHZ^i( z>lt|!r0q&|%E$yff0`#UOrq1?KK+3(Kb=QEOdtJtt>5nC`g2WRe(#;3Ev&%<;m;LY zckkXsL&5mr=z4bm%+Z3m_-t(-G_$XRN2`7hVHToWR<1O2<<>u{=TJrJ9DMeJBvTDE zl6z#>lm|E6i^k62yjaIEX*8 z-ne0GL`gvrjR<8hk*1uO$oW^?9o_FNR|4#hek?b)N1YIGd%7SNs0lR})+LA~scTcO z!Q!%6&4!#g%&Nw?KwNqf83UqcU_e3{FhQsBLV^N@$iM{whDA2FwN?4NSV{5D{<*k- z3w^amq14Oisq>G01{{Wpid=9FuJ9i=L-xK6!p|SAg@sBPP>7oWbGWMw=_{I`F%rQ@ zL{FBWH8Et*k<2k12V(dEio<%i(Q!Y5Ggn;=J>-Js_$S>7_D7WEgFz+v__&!{un!^n z(+jO^Zf-6NriPKJN&!)WpTA)@YVNNos-_9p43Q&qKdGm2xHm%RkN4D17$Z>H%|g|F zf+H2r-4{U`z62B*dA^8WK)?c2A(+h9U3;T=A2U9F&)L~!7JFkOtW+<&I}$DW!VX@3 zgm^|Ue(VL$AsM95Va&{6mqm%EJ`*Ea9Vcq?xlApQ0dkR!GUsDUe}8`)|Btw=LdhfA zd4|_4`xC@$7$i_-hG#tn#oX@3$@d@a0zm{Dcu74(d zcAFovfHvkul7F_RTZvv8>^Q&e*;o6 z^pb}^b6?VmhK-e>Sy-H07lRCFn^Y*c8EVt1Y##DsZ z2kPPR%*wMG8Y>P|D@0?1`W3BQaMReB8q`|k(K3)s?}m>BrJ)xf*+|iCI)H(Tu}mS_ z+?e_3x6@FUE}fLMOV2Gq+6+X_eK@g>i&IC4i69b!_uALdYx7VvK}d@P^1{N$M+iY5 zCCIt$fK>`l1R(<@%?SCZD?nS~(Ewc7eESBfUsz^lwmocJcUW;CAW}I0coVK>bDV*U zrGP~ZV=4GZzZnodnyF>xQEq2KNSeVtV0!GF7H4Sy$Qfo8`Nz?_%5;N zl)P!i#pBD){OW>F{*Zfop@8KEQUmJ-M9RlJZ4*C|-qb{tFFW-uyp=56fsDS-SF(2D5yMWC}su^n!+3`bU!>m|_F&dEuN?e#gY{1RMJw`kIy5FTyQC{Drp@AUQ?}7^_I* zX<8+iPDtI&c>S1DiwqqEC}sg|a}oVQbZjE*jx$0|{dW_i@L@>z0Ry?H*#hAe5Fl+= zN?T$q%c9P6xqGtKc6@cjXZ{J=W3q~djp@|}8C?B#e01-2a``%*guVCmEp>97zO*RZJ z|5NB`L`A^E%33fYDH?!h2|#e31exW5WwHX2zj)8I^yVMh%fvKz>{kXSTYWJqD}Qa* z;9q)g5(HbN&V#obJq)66i!jCn93xfd%9TMe>*Rk{MAtXU>NiW9&Tx(D=+&!XGa`)x z2PYON%e#_=#c=RXPYlC?h9TYcsSa#KTv#RW$Oh+XDVWKGFh$?#!MZz)_aTp80WHi! zR9IO2sctDMFwp-w#Bs5KN&^yB5EA&wa&ffX3QsHfUUwA^z7#G{vl@7hQxO#vQs-Ay z!_w2UEx|*NE3pKb4Idw0$hcCs6nRJpJg9}i#l=O<;8q6gcR8;&=fvm&f@tLx%faw` zlcc1$xI_h@k4#m<#pw1&E|#m>b7MUIlT&6P#3VZ1j|+@QNSCR6x6#>@QP0KRPa+~9 zjprkicYesT7Yn|!=V#^YIqt)o|Tg2 zpMb>yWgvAAt^Ruc1#W;m>;k53ksxeTbTkWGP!J<~0DL0Pq=2VSF^?_N$9a@V`i3Pl zgx6TlW*Ai9W8%Dr21zp73ikn~p%MU@&mZqDB3Bh2ox-9@3InCcw9wo{jWuLDIUn=h z$%jk*otuyFcuO+ROx$wa&~n#m?!*%jz!d$%fdT@cN46T z9BzF9A1_#ZMo?A_U@i%nb@uXl)BOj6Z#}?tw8KLSIjQtYm+lXJC8w6A1@&fiG*X=( z?$`jlB^bO2KC?Ioj?kLfzkzWe`-$q8X=5ZdL+j&RN3$@M z{Y<{VJ-W0MM4&W25V=@~xM8&;{urbhU_$3UtCI5KR?guCB3`SRi;zFkZr&4)ij|x@ z+y1!o=V)6t)HsV&2T~Iu2xkFIhm4)8y+~GZPfhK5|HQ+iP!}!w(p4A=D`*oJhG~8X z1C>GE6Z(<%nwMgNy5_6$3dpB+{V+i)3+oUim?aHh#egIpMjWwzH+2*b&0b{BNL{`8 zLyx>>xdugNzc@5I87PNH0omDcLmr6+B4o5~@s*+KK1F}don>`@1NBkO;1UX#fS?a~OxX~O+(d$z&S^rl z>{3&MfRjnbX_GrN{ESdOnqb)?IlnXWg3&p6&$R6>eGB13-(ukR!%6c>MetL|9EA<*9&yHZ85uC{r1i3HDwKGLgaZ8m`~HK9xPvX7Aba z?L(NixA59BL#L=acSde$+!Eikr|&68cHf`*q_nFVrSN^8`e*ODh-djR0m$D#lY;X$p3syDFiQcYa#w9?CW+bZd4zv>e1;1RsaO7M6%I z^!@KUk8JB>_Kt$~Lz8ag`H(uR1Rli^>Nt|NU4F~65Fqa8xN5+uSvdUVLn8!G=A_(R z6MfY5OsySJI(|b8LS+GC0l))c92RLG;=(+}Fg99MNlV0Ugkxy*bn%9Sqyk-Go$;p% zp72xqXyZd*@&AAl(D=NfKB_@aW`L0q>E&u0D1k=iud|E|y}8vgR7CTCAHxyp&h9;X86thV)obvN)kMCSxesbjE&2pM{nc|bMxJwVj9rzLgB z-Q@ag_QPX0$7G~g>eC;~7tQqXao!7cL2+%X%<|)7IrrQgM9 z${CsF6-^2eZAw!P6Pv`u#3b)6Yc4m|&57D>_xRB%sYl68$<$QWnH8;g)Pmt9lBDOi z7H(N1#X9s&eW*W0r+b;!x2J0a1RzowFxNV}q_9gEoGY%r7_K0}5)(J0{X?g_fWXi0 z$zbBV_kgqh#CA*O_n26Ji#<+~Yu`9w!((F+Sibd-@z0(~xgEJP6MXH+KbQI2uwEwr zJt+>)({_vw*2hn{WGy+okmR8kn4K2LgPj@H&#cK^b!E3FjDL|*-W_>=cA%=Yb@`*% zpPN?2sZO_AEc^ItcBGzvEcsc>t#`7R{ra2pWwV)$xsTu7a7so&RJ2^i2Dw&fqynd6 zjt<8h`N4)-mPh<_acODT`eUt9_-F=f=XYxaUiywZg@c2!=KB`YYk7I0y)VOlOwSAo zP7oEv4LeZ5Y0mUmHt)Lgv7o@zXXiUpR|NXeR1I=)G^bEDEP@wIg#Ax$k(YO@p6x%q z?U>NdVp?_#K{#PSJ{GQp?3PzqzkmOdI^6X%1uH#YsoCj(>lo;T7)*p)&B=36c%SMS zdfBYA=*HfJh47`BJ5);Oz0f?&Gv1ZTe#d8hh#gZwey8Xgj0~`2E~(pp*0J-zz(B9I z;!^gQI90YbRwMrFg(rFWSVlgjj()+I2@`9Gu$7uNtPEw1Gi9#}e(+YuNe|aN!DqHD zXUz>Y-iZFb zkwYT%_kzxjWfSdYbLQ~Oq}y-SQX4j1y&1g1a!1%Y#N@TLb@Nw}erzx?ParSIG?(pR zWvAG3qZw{)jQ#sZ>C-#QEVY@~;-kv;+t}Qjg(m|G`ujWVS`lnzvYJ&s@uh2Ls>&HfZn(2` zbbuNy?|S<4TgYi^UzDfQKjPvzjuBgj~T~K;csH$`5Df!N# zkg3r9zlWl>77@Y6ES=Tu*vWf%Y(!i>^8U5P<}sd|#j{5HPl`uH9(7)Ah##?)9Y1|z zQb*P@6+UDn^YZT4$l`M#D320$if6dN?Yaq@C_wB%~_b9{CPCkmE3T=3ahtQ*ArpbE7%|klTooyUlJdE z#?p6EHYok80l$eudj10F9E7Ss2AkNE3$)RAEm`I`ygnSDu|$jxGX1}F-BpDdAD(<@ z#IIsb((hXLi)%lFNg+Q#GA_dO70^i>3D|nM)@;aZ(9LYH0nnzUP|leeEfRuT%EhRI(Kx5|F^eS|8&LV{e9U1GY<f#9sG&E11|E${#%f!nn6CGb3XQ3JOYepg(GQ8ET*LcX?qW4tP}@AmgmI zsIQ;TdAg)T^Qv{harovT_wj|MDWkC@iyeD zEYz0Cad2=bAChhAf-7Jg4a)n!sHDIp1NEYM$fjwDoE#_YsG*=b z1BpbC*PyYM!zpEr>lJfb05r!tq8Zr`0l|y1|*lj|9}i6t%si=Md-t9-JDU&E80mMgk2%p)-`7uu-sJ zVqjom5?`qQ*oxedq$Hv8U#QJ=`vP3ntAd^4i1WNDIUoZNT%>r&I12`Qs)zP9wZjHe z3**vfOf}k36pe$XK>X|FNHM*+=xqDH8b@_{Wz}uWEyc4&BXblxp}6{OKL8W<`r5Y* z;i^`SlgmB?7kCz7DE0wObBe@gcz{@JHHpcYS*iND@BTu8b$VJGx zk6oJ$YyK^K)-n%mIXNR-)?>nIdlsl%P5V63YXMT8j^Fb)7k^2`Bf*KTns{)?(MUof z@aR8LRE;WA0fQ;Weq|pMBrU3;m3^&){w5VLHsGEWQ#6ZF2=GhGZL8dQ@G|Zm?QgnU zgqbxS8)B2j^Ij}!buW;>G@6%D1%K>;r8#cQ?SbMb{fM5RXn$uhoGQdT@X=YBE$q4< zUtl%bIJdM=!78!kGocBhb9MnF&EEPF5L`gzL`Fr4s1&6k+sMiA^7UQeBh^%kC?ldE zje!YYGARwA?)pj>NjGyKm@IRxxPf)Kfg5_|pNJV{!htZrJJ2B~A)-cae>>@10A-lm z#Q}qR0|Ntrp1G$Rw1;K5`?QpiOY87*rI)U^sA4CEDqy_}wA0)&8(lxclamSJ1FUfC z$(VWX(5DsACI{q~+OU%f08EvfRhZJ&EDj9q+_G|fqW&dRD1sjX_*eiba(D>oAsfI{ z)&__jQd?kRf_AU~jkwRNiiwP;0I)*) z`iHQF)7u%=e0`8;z_n^aj(s{>&$7|A&EaYc7r~$3v4KYI>XeU}P=F zTLP=zDLOg-=5__f1t=5I!A^doDadB~{s527%6{_V;O~Mer!K{sSn1B(g zt*_5DE2rq)2?_n{fNp>)a{+^*8Xg%tvYpcy03tmHuEW`L=h_jPB)TBzdi`}bphcXN0|#koIM~y9xO?lm$tT99*9KZkk73#9YHrPi zkKqH>6twQogFRXR?jw?PEWdQT=&u=R=g;qQbv>uFrWhzVM-LAoG)dCT?oy*G-=anF zctUV1_e(^U)7!>#L{<;jW@>oknu$`3JgqoXuFHXh*Te54Qzp8*-FU~_u*C>&^02>m zh#Q@sVVrDWbxn<4_KP1jYAPxXQ`TuaF_@|VTV_yA9(rMIqImt8blY;|3b5ePd>P1~ zc+su5@z&?kqET9}fVfq>*lH29$14x{#i;^7i_GSi>TaQE4hyhjHjZG5M^6@(6B>A9 zu;d6CofgyS>1lGb(-gwa&aNem8_y8<&?1^z&P{LMUOHZa3tMbF2%* zE}5ktCx#{(3=j-80YUM=Eq)XnA;@k->mf`K*rwHRHxX$Wx484aF|It0gew1=dSpF(uUx}aZyJx|qYRw;Rh>C*MsSoSnF+2jXnB)j+XSuPIw@J#%V(IUU z`F`Idf>#gG-OZMHHs*h!c?yFje4L`0?$9KvFzY`W9UTp&j2FdE7yw82ku9&`QFyGU zr^hw@D_E@Hpei*g?|R5e##1b->k;6lL2klAU!knVR;*i>zm?46=ilA`T&**>VDR4X zuqt=2%SY^8G*^ORdz2Exh>0HiJmZ;^QB2VFDitaQ6(4F61;k+XpLC8a<)7e-f8$@7 z?l4ZyHA?aV-`puWBL-3er4D6*T$D1Z`&xX__Ll2B?ANgFjMQyn%AOqb;5T75qcW+R ziA>{6qa;kiKtSadJJA0|1;-R%jp=stVyl8Nl%_32;|Zed<`00ck(~=`F&F$T5wO?q!s6Q-V7mQ_I}9{jMoA-e zI8+1LHI%_n*!k(m2(ScbMvk!s3qz0NJLC#kydFh9onL@&>9R-!c0xqj@bBD6SAwCi z3(6yG`pSOaq3*VIch3S@2e$^$Q6HK;%fJE6YxO2&6r~rmN-X#dMb4XG#mw4g;fnn- z6i*GM=^;|#H-pX1z&&`dR%&j&DcF<@x>;MiKH)xnp>t$-(M?#EeqI+ywp7YSc>_)v z2MvlW-1rgj`WjM{1>2Wk_ofg9G#f)m(h{2-B?N6T*pkj%TUi$T>tRj_L^LB2r+o-h zC?ZlrqoXyLS{?^2y7#y?7IBk-`-MI4~T|#_d;XSY#Hh@?wBwL32&Y&XAu9znPHG zE8X8lk_QxodJ0csRg0#>O~8PXdo~a8a+Kj0@Zo7F2-^~QP|2^1(|=d((TRXI{aTInws=8Z*9e=*B`F8%b%?t`j5Mms@ zUCbgC2tA+^{7n$k2f`E5&VY^c<9U)(QSNO(kwbSR3{-B|SBYYWra}e7N#E)bMYKS8 zjT5v>nxg=yhqS&n0r!x1)WzHuh7K0SINo`od-KW_uCF*SI$j6ZEZxNDquxUdCy?U6 zeLNL8$aP_Rc@Fjz32+F#+`(6e*-SW&TPc6VJQ61>&Tqi{*aEr=gBKkt(Af{Z^~tm7 zqS)Pd*ks?n9L8dN6PL&A;id!8%Vp?3RJ`_<#4v<>fv0Y0Fef* zL{D7=9q$3y8nYc3SQ(hI@g^b8#sY(08jNb5PMatW9vj$k^6c51!||w+abE~Bs!)<< z%J016b&;{St3EwFW6Cobp(1wFXs0ZFm=!*aOrEmm6>?ymN22J)3|W=4;K#1QPWG_k`0G~po8wQ6F@FcYtf z(BFj?MgmXeiGkaqnm(*VEydkpVSh>!20Q{jB2pZam}oYp`BD|p+Z}^u5ZLP55n}5B zfGutT;-v|6(hO>p^oolQQAd3I+K!<-7Yd?i!^{fY25R#@BuQ@~IU1Okhc8}i^-Rys zmWIjbU^6Hiusk(r8zxDOPfpG}p~T_$`@L?~3x`cJzDG?J8TfCxsCQHupc*g~h3G&G z1UK3bTH;KZ^(m>TcIPIM%=H+u$N0*`+zJeEs_L*S+%3W;%o6J1*S`Z5GT1=Ial^>B ziR5XhKRO6_>#FKCKgwy^TC-S`{#`9(vOC3riV#ZIo?3=c&fu6BO439a?)hNcs73Y2D=M0=ac0>>IDo^gEozEOswXuEkoWZuBf9#^~$GBCMGNV zuF>#H7WG^h=r4$;A9Tj8&tU8rvXwNE#jQgSbkp})ue3`@x#_R$EB+bSTmT28$ z&Gt+$*jt0^xRog_EiD1^5IHD{I;s>Xiz7q$QH4bm6ev*>j^0EV8-u>1 zqvO>%1t4+rVZ=bOM!;X<@e$^2H~M5nKy zE4s9?D5;>MjuTW*Q*$BwgsLh7bdQ^@PX|@o4-eTMGP?Hix#}D&imMLx36j)-E8=!c zq9h-u{Rjs*0C8B+E%hV*-L(IuGaxMyw-nV`+1P6EA~rLKC}Q!XQguYDj7Cea0#y)w z+Uq^mP;@M#T(z5+^QktfL}pGo_?s^sprSQ57pn=X1?cqqeVz3<;umTtRu5Smvc*iE zj#O$so|ssR!8(hs=jI%oU%-#SBY_qxvV}+wQm^gVfwgo4CF39;3JMfKw1LIcYXss` zwGXc`YBM3CeIm*D9`TZmg*!n?#J0trK8aY-?-`hNfXB7oB=gyD` zX^_40ihLX5(1_#JLgvg?39;g>>sD`0_A(y#rYw56=5iZd>{4-&D3THtK$6hz2WbF% zQ!EZ~pz;D0U={X`8*lr^5X%%jx59;t369g>CusV#3o;c790~Yr(NXK$+o!c2fO(OW7_kTT$dl5-bYtG<>jGSiT;g9LF}3h+Rj3kR)g&+tBf1T=esP zDkRJ<;c(Di60?W+PuQ8s3QCZkv&l?b>^ig}kkyikZ(x|{fHHhxm_lUWCGm06$oyt5NC^!iK~QY)*Q(xN4%uq>QMR z1MM*uJBcIo8UP$%-X&KABw8Nb0d6n5w9~b+HZ(V5p?^WoLi^Cb#>T>$gMq>H|9Szv zxs^Uc2TfTIya~3&TP1506g(~D2Q~FoiXjS0(EWD;e6k;+*T(HX$TpnRZf_N3TY}m+R-gzp_s~RTQt<;J%;oP@OmFW5_p~Wid%dtS5V+6Rg9>=axfa ztQDVv#q5GA;}(Z0Dy$cxSqk(E99UTN-=qXfW08B_JPipbu6`r|F|9`(E6X$J8FO%s=;iFCBd-wdrQ*<`Q%IxU8Z}JBh zFD)4)Bqpw_U6cCs&P`5up)&^l{dG!jp6Mv7-?}i>qeqXrmu;?*FxK99N?R=!#Twtc ztXTOWugXuFI+R|mHB~Cn$Je*Mt<4YP`rWO1T!!!8ze}c``wxzfBN#A;xUyMu z$x~!J*!cUmhh9fyp2al3pUxX0p{NH<9UcB7MaH{BIsvzci51q$7haiZYCeQ3ovzrd z{;kPKa9Vm$Xf-b!b@kxY`QB|jo}-8REB$bJghx#ux?;ISBcFeHCKfrhzAn}g`Qi?{ z)e{Qw=-kN5rN4=L%e|OnWMpVpuH50YqgBXJFKSj8>Utyf^y$;>gHiL|a$DoG^K(mk zdkSXeA5w{LwnxotLXM8U&B4{`jT9RGX>1&+ptQHyo)J`cm?tc^UG>VzVc>}Q25X=+ zCn6%EbFenNx6<$T{rgQPXXmohjf&$LKZfR(mORI?Cen{z;l^f{m*2&3*nYS;UMzDt zcMNO`zdTL&>AiiJBcG+Lx7?qhkYc^qNvl$v*}Pc%CTFm>)KbTW;{l6)XJHGIX-mTd zdg>Av7nfqLCMvuQv&dwn1IzaIc4v(vJp;dTcO| zc_)lsZKj}$CsDU8bZx9GsA8kMEtrjmCzf8Vd}?#Ei*vK`wXiS_tijyY^zXHas(Wl~ zY%&uGPOLbqeQ7d#8{<(0`dwC&&WGFGukog5XJ4zT-ltK>eqZZ`+8)Igkf+l`w0iVy z0eOrkd&>esLNXchbOolygV{wA@i-P|N0xf+;e2op1|x-3oOWxukqpnCdFjVGczxWg z3R{eGC4Kek)j+1wZy5>hhRztypv0L>5beYg-_`%d{gU&ZV`?dQsTst8|h;S~!dTH?+R)_}6kpM*YF+ zIc%dBONjyLuy@j$o8R`=xVlEg-jkF`5vSX6(eI3wa*KpjGu<^EE0sbu`kNq_Bz@!K zXsKnQ;K<>mYYxhO2iqjUYWQsmaaoQS@rH(mh@P3**^rEk49An@6fr{#4?f$CF{RJm z-Wh9aCNhqL*=lk@&%|iIox-l7;>}X4C{~=$w;Sv0lO-f2&91Dfl9QJY?LW7&vXXUl zbj<5i8i{o|G1T?TDTZC079D**DOg%8^0|~+g*3w~mjiac|XL z(OZ__IsUb?wHYSMit3$Ojf)zU?WvVUm80APcB?;+&SCeG_J0$`wLO# z)V$7Hb(l{%)D6GW*Gt=7o*!DDpV$s$spt#0wYF++&oq3!!_v6I1N)*9svqo&sf`VB z4~*;K?AFhOsS2DTA|kjB$E>fSqff7{w*BPu@CgVgactjR>|)WXziQN%`l+IV6BY?6 zcm-;LxYKLH1;`G2ZqSV)At3<|yA{q2ZRDo^c*bb4=`1{or<9aFZf+=~u4k6{`T1|( zzMYEqCQT3yp_WaP{v4*^GCWfi0iXVW!{(<^fBFYU$DVY#41Rk<tdA+F z%q%QQ`R}pe$voxa8Vj!0Zot`dD`_SlE5b;)0lG2bwJ0&yfQVTUD0U3 zz`(Fkwn!>1BO|7!0OjIc&@(323#Z8EhVF0=845WVQ?;M2?5&Ry!vS(~aw>(wFkEJx zGR!36DCR$V1-5955Sg>U`}a>`V`FXBM?xM8eGA|5Cwmn;T%aG-YT`tqR{0@Nr|Db# z;OgcP4=g6ZO3mqw2OQx=T&R5+3IR~+6`j;WX_e~TP_Lva6(mz-2?db$-lxPaq%Ofn`sVwTzPS_YO^-<*Cur{~pV(FpAmo^+KGdKSHQ##R2# zkIvN>C&nv%X&!J3(08<;sSqu)+pUp6HKkx=yora0r(9yz0aZ{twCN~f7%DKV5&@KM zw(|>ug^KskO23jYp}Cojmq+m6Hkz>+>*Qvumes>9Pm@Z-ASESzZay)VDVUmS+rPSh zpVRKEA2EZnbItC~{{B-&#;tG1@O0iF=d35oSqOGDs$}Au>jjnM+0&B z8}<3X(GKsyzkM6Iwl814eCFw-$*JUro%R7Xs7PWsE4i9yQYO@fLi5R3C@+!KrSKO) zZS6;iLIDdw50&e znf2PCG%eLa5fz7XLf2|>yrYXe{9dw_pFbbM8#loHy%QGJZVjeNIw2)lA1?5Ib#deY zE#x*Sskn&=EmGG0{{4G)emF^_rrO1|dkt#`GhSPp$e^BH0&6~Aku4jDEw;dK^ z>=nzbo{~9jq1unz66102eK(v8cy$rZ=Yg?Zdj;=%NJsU5kxOjZxIzmbRhe0vd4Xy;K3IHJx z$o1O$#?xodFccINP~b2yaM-QARa7L&socStn3y;_+wHn`huQO~NN6}3dSXhfqc1$z zi4U{a;e3CpcMh{z?)j`A=RA;rOLO_rC^$A&+{Wg4ig=7RtWjUKT9lw45tGZ=2Yh_| za4rX{z*WbKGl#n`jQ*5b&YIZmO7dko3)wFBpi@v#Jd=oRcfHsptN;D`c(?17!|8g- z9p?8~*YB~?_A@Ib4z7Oj^Gk$Em0SprV0-H~9&GjdjR9n?O*+v2qu^FzVO1i|ugqJl zjkw5vx>JF5xHg=KG~fBv)r_8=o?vNVe-b8sSZ6UR-C@sir(-Mh)RGCuwzUs%jG$n1iS*P)9-cn@LOUnctY-}m$bu4Y1j(Z04o15m;3WoN#?sf$Y zetW=4@N=OfN?(|z+P>=Jp*)61UGPM;^T(v#<-XezTw20yot^U2)6+!bY{y+8At6Z{ z<*W4l@XFE9vWG~c?d=2~N{3$BD%Z;kYHI4%N3KRj>1cg$khO(Z zM~X7gXi5F=zK~{OViKg6nV4lY8In|kUWWWU;C7VdK3yxok-UO= z_5(8J_dZtE)-ov)ajZ#dr51e0CnxCurowxKq3p(qK^Gw;BQqDCEHVy5>$C3S))%f& zBc6umMDVjSnqB6I?cI2_Uw0&#%Ml;y3ZUQ@NpPB*W(v&qdPEqgcy-^ur%dua7^cjG z7NvrI^{V2plwZHXvoE0r%4usS7>?Flo}16FubXa{xz2|u1Q!$(aM*7|OH&`7$Yv?a zjh9;f4mNM_CB%`ugwi5vh~Pz|k;1gJY+=LkIA~LnfazJ1?!01wr$P?bl|E{JEz&$OTXPoaj6Pc78Vw{*+zf4 z^^qbvuf(&-%kwxKEG)*;2@*E5Xu;|kN>O4hz`vq#S#U1Wpcc^Za__dYBs(=4JEzoe z3me9T6q}BwpKjJ@P=%b5iV$l(ePK2Cwx_rE7)s7Naq$JH2z0q0E#_N8UUG8ot`&4e zCe=XCr%X~TG76z0(+p&AH4c{cg%ibSF~u7!ty*E%v9~l{!3O1eZL((a&5Nq4%8tK| zRk5psIWi??YG`O^nW0*MqV+Cg3XKPEBoV-td7i}1$%(KNZEbBFu0g0a??gp^0ibF~ zUv)>vJp%6HF#VZ06&Ow6ftaxHOv$86qDGC2UKZca^Cpw^XPD5kdJih$Ks5kqQ7kcg zkRhL?4W<6cvu8cf2s7;C2?qcrW8vUD0we)-YI`{)t_RShAN^dmT15|#kls~5Zv)wC zZQsdX1HEHT^7iy(gq{E^q_>j5NZs-Rodk}smX=oXAe_x4z!7>`uS!_#H&lbAlSUkE z*M^vonh1}p3pSeQa^tE}xUqU)*Bh-TPoDy4*9p8BG0us|c9oR;K5XBcAn3O=R>mTi zF3WVf0ptynot@oeIKN>{G4YK*0&S89p>1F=s4+lyK_w34Ev5s&u05Qu%XN9S+aC8SPC?xvNw)2Mq20Q2uYQKZ z?gFiHp${x0Gq4+(*zMV-#9(UK)&8+EYj&rj+~JFF!sS?PypobPxgGa3>)g?QB_#z0 z1^ot2leNJ8@qiYH+tJBMWYWE;BETHRC(Cnl0(2{1iKO z0>vT7P!|k|jqQYU*cQ!h{nGWKa+u_vvGb-wQyR+A>FJe`q42Y1HuH(+BB!ut9aghS zUVhkm2oN-RFM*Z9e7rmaY5W!T<`XWb>yOrPN5Z4Ne!b?m!PkB9c8)Ge>dhN3!;k9F z9rl*~{@yE>h~u$Y>H7j`cNYlbcHHGzO7r}c^?+KoNhc5Sl;6PT9rstHsvK?ooPgi6 zestbnk%Mwf%VYi1SsjSKrKP3b`1_}2$!`59Y?nt*KbG4pff(_~U4g}0&h?ZOZ7 zhxST7n~Sao1m_jM7&~LRMS$4RHjd9L1{p93i z!qa%l9ZWOyyau9XT2HS>y4yNDUopD;3Z_ZjITpc#7c+*F+h za)M~x(lly#_E!fntyG%7Ucpy zsPe#TarV!pNuc9D0lfhhJHpI>abn1JcR@^BM+f`*^`g?{Kk83vsUIJX+mitXe939Q zF$Xl6no$FI1Qb#VcJ|>+!quTXo#ge#I1G9^QZDGPd@ik@K3hm6gc>IYTz= zHx}p?1)-r|zkcPHm{L*{38k%rEexD8+_s=Sf+eN9>810@bLW%2c-Ynm3NspPjZpqI5(Yg;nd~D=H*Mrz0~2bF({X!qqvU&V+IEYKMKqFee#)VzRU_iNqvZ4T>s;5X z%Rico;YHf8?QIEY-@(W%joS`P={1)7K9(x?{Qy4`>Wv$ckF< zX65b;7 z0$|n*eHWkrB|!Un$!epnxIQq#{W*bVmi#=4K*QzuDYO}z zrS9o(i$kTBkEy7rfKXK`>e6Dvr@(^(WI@wz7ElAXX=tSe)H5p9`gmtvPshe?bD{&D zPv_M)!8sfSiY}{r6dgH>*4}CONI&pEvBY+ zSzu)_r)$0~j84y!EgQ=M_WWp}YJ=R6;oVs01Eo+p6}Q|^4 z67OEVv~%~R#l!J(+u5Q)^~bENp@3QopjF9dD%J&4$%yJ?D}(d|J+2FAjyRA)fL)ru zy{<=d*tX7uZ&n?WLSy{}ZJg_5De<}4SUCJO*Y(n6AV;GMcE7J&96}T>D%vz8jOmE4 zU|gH>CUdEUGT!Ka@85dlXqekoWYg_!PR8NAI{VJ!e1hNwgb0J6kp_g@6&~n=!_D3l zJb<^AXFF{spy-&SB?4zqq_vm%4jlCM-Mh+D_%wj0MI)F$gV>8wr7jW0YN}Z6#I90e zrVA1dk=ji_Q)h=Ym*|BI7qd+PyA8w|Q_IVSOqy~jy?=|06*5Fjhk8`Zs}S(B)R%S} zVe9nt^iIIO-s|b@00A34a{{K21iBg5@mvu6M`q&zS`b={1|IdM%hB+jGH(EAgGZJe z(dEZ1_8XR6R4oTK?9&%7LV#7^cAL3wLcRVu?zYp$P=lOiv$0aG@;7fBnqHjm#$CP# z?agL&z`rGkBIw7DdT2qsy~$z@N7LS7$G$tm;hgqn@n)9xwzj%}FAzZ!=n0d}k|dz! zuO%gHi(PSxAmdEKG1co0uLKkhtHpryEI=8^6#_#8jKK&x!ZTLZ;%1q!_E>J@J^mIN z${?z4LP7#iOJ>kEAA+t5aCdy0C|PSFJ|=R5cAmH1xVqFI+A{~ag$7kcnTz&JI@Vnf(|tO z+q=Nv@1p{p=f(r=&)0MSO+hvF{&WR%r=6wi=AAn&uo;y~Ey$shJdb*llpB}WBMDnt zx!B}gTPR&e&)3Vb7lu8agKj1OZ)5(_g75O2R{eTCY$te54 zuqRnH#Kpfkg1&*g8^Y2+>SL-98>75@ejR8jItE5MwD&aU(}Re4Qd zFpsiva;hNYhMJP{?XnF=fw6k9;oYyg1qaQiW%LF(*JmHsKwV3PwWI=3kMQQrpQInA zP=T&L^>g}sZsG)TFpeaYwQ8N1U}SvWv+)p&l=g9+lj zlHpx{W2#+uYX+OXJ&*y0&*pq6^yW=Ev4)dWQgU+SJk84D+q9XEIWsY-WW(mJz!bx7 zD!I-#KXAy++Denh;$M4%_McYer*k_mFAotjpw_s-Z~vWWu%v{Qc>)D=yuyC8f->QL zP*>pkJL348gzZr3?r=NWe8$ces0)@h87XW@P?7dY_}(S8{2fKa2WJ8HWv=}#fjfVf zZFE@!Vp2=ztYtao==uNW^D!wR_TEj_XLSGPA5cRPPteo9<>8?UAT45g)!M ze}&{YK-UogrzwDptt}`p zunsz#Sbw@4I8#~x1XQb?hQ*)$yGC7kdbCV9Ub}m}5(U^@O3v>EOr}XE!>A=Isw|a@Y@MLsEJYKU7%T@T*bgZ%Qc_mfrWv?#W-&` zluHhv?>%fD`74?2?V=!ti{(J2M#<7p%o!y9umNZ%zo5Xf4|@x=3?MP&PoJU_hgmr| z_=&EVoN}z%a6nD*vo*FdWV%qF*f=am2;@+>A+6se+mjzskz@tW5`@`DgpT&LP9=034^O; zV6=hcG$DA4--kIYG_*0CNeAwpcA&$2eq-_+U~nWNp}S|yI6`}KfDS8`<6ok4u+rbQ za?b%I+aEuE0L=$?woQxTv7B&VWTY=tuiv462Yz)H=M>awPR%s<0T}Yg&1Lldpt0}B zy*iME__jA(2LrR-#w2wT$~mMKdFVZf6B2OuOl zNo+-x!!}#JiUYPjqb)H9j&V&QVN73{)prig&>x{0s%I~xg_+=Ep$9i2Vl=2b9mOTU z8bH+qc}!E7)8*{tVrNVuRM4k9HF;#kKeJzxQ$GUFkoOexHJp>b!G1)U$V#E(66J!; zfq1pRLQ#N2fi*G%jw9%E$kJh7yY{E6D*$0^u+c{MZtx2&J-d|v)~uQ=>z6OzqpuCf z3w{Z})X>;EWH`0@vXiEIRS{9SQ9$O_TeeB#d?28~XTF4xrv$OZ0*h!41Dgp(|d-6=Nfcc;~{AiNzEC4ojOMKZ5_EtK+~ z6w$jZ>Je<*m&ymy!ZFnCHM`YwXu^m#{pr&u*lSWb>fBJzguuX4Yn35@GMIeGvN;Rf zm3BKXF)=ZMS=YDU6c|$c$HNWqmC!*&y!+gs7S;;Ua7CQd$vvNOsT!Bz;J6p)$G^q| zS$b>=YWJvF)~O_xF)%g}uluEy*xJNKudXM(`cMjD0z##Rh7=K70@@hJ#>@aDA98Y5 z^mWAzPP$%-#ql^BjF-PqDK?n~odjt^rDbIy8K3%q=PklcO%^gNj=Y#9yVQg|TWIjt z9oWHVfGq`Z!T|Jqfc_z}Udb=}=GLuTXHHK~1D$LRyXSb>J9hVQEGh6cLqbZEwnXTL zy02f5B%Yrog@&s8LIcOXapM^+?N``iJ6*09L4-8&xkmle5QQ*q4FlQ?hAuzoI4&+O zP;O!XPDMO02GUVzGK>#W8+>j!k8`=w_vGGXAjJ$zqa_Tmxdj9Tfv{`$9)K`{C{UnT zb*b5p3xdS61Q>YIY4=!RB5Gc$ujD*{+Pf=S0qNeM?L z4vCAJs^&0sB^eGBvjA4H|VNTV-UGUMU zclM8KE>B;<2I7%fodyp~I+X)gN~tdc)1kT;i zn14u90}#5m(!n~EzJq%cY@1T6c^pKB1bhkW_5*R+fnlKJ+OEozIxdkT76EL44h0S8 z27tYYMbPo>9P}L)I7VOq{sDWZ7IZ;&$34oQ9SU|3%77wdHM*0Ol!T-?Sj;E5AYvfS z&yNOtiq*7BBNU;qaOAf-IX4N)d0t(fu>gdajj^B50(|G`=^4C~>=PVJfN(Ug>-Sz- zOrZer*8&GKn$!M<8?f;mP}lp(5&r8@HlfoSw>j-FA!$Az~@dr6!2-KWw{;P)2QnIz_{9UO_5H`=aCdY zRgx@7=xE6QG!*ac+3bT8m9e+Cmq7NCLM2UiTDT*{&09Q8_m(oxJ&2Y_fw(6N+NjZ3 zsio1@RGs`T_yxist-J+u1G;MnwD8vEX2v#vtPh|iXiy~u|HHkIs?aGu2hJu53RCi+ zo8!zEiuN54oyEC<1A#RH{jUFLdzSDg?5TcGefywf8Xd%TMX(5=8bN_gE`)`p=7sv( zW~2m;e;+UuLuwJNm0qwzz^|523N8XsRDO~nDVPHsF0|m3l$4!=16FRxqL{yM1%nfL z%k*Grqm~~}Tf4d>K!;W87zM`m!eQrqK_M{4RNxyCN#~2NE}1wFfpun9QwS^7@LZk@ z)`;dn0E6Hsln&h@(WfV4VeXlkv_;_afa#*UnGk#@&3ZV3O(N}ZQXySWEQvQQl`^T= ze9+~tMjWRC$71N}TDTc8F@YDb<>Cw#I+g5p%C~;KJ0avr!D;Ao5V836>lf0X0RM{y zfEs0SvS%n03<qNkv}6Q#!g1o3XRo6|l!9`rPN&*4hE@gOk4qjG6*Uhy8?uH<;5F zurGUgfNjDNY}_l?`t5b~>ecXXK{d2cJb=?S$cEPUEy6yAh3`zr{Jagw$#QFouQw4S z2DpMPs2@mL1dM^L?d@rJVKPx>EcZwJJ9pm6$`XJxj&NYTO%*M%aK=v_n0PzDe!oXX z8kPSp&K*oqD$q+_1oigl_}CI$ibr9f-OrslytF`>a3=1IX?tMZAMbI%LM?pHkMKd3 ze+Pa7V{60n>+{Md4V;Tq!aVojc%>S?IfG-t&(A+pXc#2OGu!hT6c@y|1=OrOx4pGR z*;DqrzMcd^Rd7A?;LNZ=u~`HnTv&co{NlJxn~~)0yLWX>O<+U>@b{-Yfe2ZuN=dfT z2q?NrU`N36{sOlgx-YZSkwx!#AV41QN8%tYln0Up;z1%3A>20(tqjPyL&c`|9IGY@ z-@k@ZiHeGvT-%YFJ4kWag9i+X$eJi*Vjyd3A$gonG0FGX?Z5|Xy~vW3Rzri6NH&-T ziau-C&C!K|Pf%ZCfGg~dS$Cm!=j)1qMds<}hYJ#9ez^qjP=TjP6nod^70BN^|w?b8}y?yVVacEOR!)H=%$L>TSykzkhVYsE5%L`8EDn*VI z^N2tLr=%4`nod|P^N$zx>SW}eWo=>8LrhzHo^lz5yfl}x*Py>S6tvIL`@TWT>>=vA zQ(0exx~>?-D3@Z2x1vg&($(HR*OMeNH#dhE_fWk$5YGTY6QXs30i@hWlIgYAwQ2ac z6qoZ8#G{^cIrWDKTL2>VH{LU6)T~7@8pv#cHf(8SH3J9aKD(7CVyW6xBdQtrQyjJ{ zf6k6~hpRra!7tOHAQA#59_d{5=|&=m5${3ggsO}<&xj7#%LI_wVIe~Q2(l#>mX=ze z62Oz-PIjYlRpQ0{25#2}sQS2aaiE&8fY=IlxELVx2#~v7&-aO?q@)@g<5v4Kn&7_f zfYt@V1CaUgOx-2;Kk&J|;CV_W2@}d^tA44|1W6qO1yM3U(z*qi$AADl06s`Gf>`~- zC*%i0WXwP|U~`s!*nR{L=mAGeXJNOsR!+dp`mtiuk*+uXh1s?4&1VV;sqe4J8(Ld= z$&*bw2D+M-sN|@0^z4MVgGqB6AD;_N(10f2yJZu~sW?NJ+l!Kt_f^Lk8agBr*PA>_Q z_Ek{j0S3)1E-L(5eeBt6JBx9jJ%M{umE&^OS6DPfmV6*S?J9DF4i71qnEvhwwNXBP z>;ZK`63)@105;|?DmbUlqVPT(`gXf#V6|DYG=$VtWYU7Rfe-^|SH8|)fh016%>t(q zrz(JZ)r3J;NkhG>AA3^1_}_mM zkM35^Nl2I)NLS_$rpC^t&Vx0>zj;$~>Hvz=q@8*O><93@j<%b~=!${uU5ycM0yrk3 zE=lf58DDHVo!9i!-uz>jfGiQz)lYu}d;&_U>3Gyf)zv`}Vg%Q!pb$hTB%4z?PijE! zc?J9WXO-0|T&WdUG2Jg;x>EuM7RiY9tLf;)ML-L-anH=DxrprUmb`}Hp>zTmHiA*h zU91xc{p33A>doxP0{E<+{e($c+aQ>%ben(Zjpe0Ol|zBkm81?@(=a3;sn&t)WU5z1Vwx1%I00rtYqGd`Gk=}@6HM~OO&o@BkI)|YkB{g!;6TX!V zmNW$2!eFb1_Yg6ti$WJ!8pw+3=tNR|NQhe67c^7^g@Nv_RWOT7PF*r{k?1&$cAq=_ zx)~lW&8oCtt2kOF^UYgYr0;Fl3`kW&7My7DxVBJ$b&<2M+;#&+oCOjo(a|RM%cqB% z5|D~Tm=UZ;&;V>nF~UPKIh|SO0==6Avc#!GrNJPvZGlrWNO zhH!Ng6n4`AZiBg*nR@WbkRTM&qmH*f-%LooOpWElq>ykA3TprObS)8hEqoAyg~3%! zfV_9LQ`ztVpaMAZ0v|rS1mc2N1i%M+cs|qds^0(lz&MuGE_40^n#%6hmL?cah+_`W zxaTL|6_9xicXveruz(W8@?lFeT{ca77gj_->rzUNx?~@OMRLYmlU51=_4ygr9Eg`j zvYB^!V&8@cT57`q^uk!AcVAu{)*uxfD#;Gu&0BZxBF5>Oq_(v~6Sjb$p$3*t(*mn) zm>}CUG`b^j&5>kWNAWx~LB`1?3442c;80h25vY9tDkXkkPA%+&8OUh^As((ggM*BI znHYJ=`TQA1XEdt5KA5tiew3c?%CK@z4%S70yWoypfK$Z?K4OA-*2FR7*0%v<=Y2dh ztH4y;-VU9;`c-Vaac{|BsIqKN9{i_&d>g#TXeOP;a8r?Y@1|itvx;QMq}+fk6`Ce+ z5~5MatT$htPhKiTr$B0hmA>pIzk%;gfRXm2(sKR``Ke9wCVoHREO;#TJwUZxj^~M= ziH7qU7*K*a0`ZaCh~{uWf490?va0-mYm=CF;$eoPEuDrT(!1+F;}R_++CHcg=o@VI z8)KuLe@O4$`#8G$?8y^?Y&U)^%X@<7LPfeFsTvx_b@=t_)BG>oi=pGbSj###qV+c|V4l{)mnsDfzCDT(t*bSN# zi^^m3h4u*K8UdOIK^p~rSPzN4R`AIHM|=0ribk>eFZZUX{DP!ht@iyuxGT`Bki_o7 zf?n?l)^sYQX?^_sW&vQl*ViAdc*#E|w5$?Il0B4+5x@)%}A?Tnu0jE(&UwDvc~r{ZU}^f8DK457vc z+P)a%IQd#v=H?7`b^6nEHhZuFMbI+&!S6|gWr58EVLhGA83F`~=(Z6-3P##cneuIjSm}0= zzWgY!($@beEdO>$^gqbZ>c3|*RH^iFB1>f>p_BjRe}hTE_f$%Z_~iFjz$<< zOaCUJrgK@%JckR4S_D5+O8kO{MDF>z+(`*HGuYjze@*Q!x$hfgWR~jX{UptOqjmk+ zM#+leDv>7wf8V-2g?Xy!vwh-d_AB)CBfk>(gW3LJAPY*zs@tmrhOV(3_FmL-YpM^) zefb&@P-a2tdM+2TW@0)f*Ypj!^M=R&+!^ngjaUzHOy^T0HBbIvp`E|T{Yc(@QO-A* zje&+1jcBsIV*kCx%ZZ5t`cF5X{`*ZI=AdvaXbH;^5nC-3BjX}Uqt?0w zo$2Bex;cGCPSckyi=jg7gaj*(Ow-Xq`K%~5L`trim`Eb1XZ^Q=QhJXbwR>cC#!QdA zFlGKL7WI&}ec~znYpbAJ5$r>Z8~5Kr69Fa#kpq=EL|g;g0s=&2=H}+eXb`}|*6Liq zmu57N&|Hq{xYj-Jeo8$)4F-Ec60Dv7wQ6wqv&>#WbOyxD1Q0#I8v6o*TM+P@2cQ8k z)2G2Eh*eFlIG zYi6b}Pl1U^vVv7K3L~c`n{Ms?14w8LmAUZ`y)84SD}UfoS8HB#{sw*-(b|ELr2yrV zueb$EG$i5VWM)A~h#tlE`!$~7m2rQa{FD6xd#ca5{#Jk8Z8;UPRl(Z%jL54%ZT-NN z$%VS@b>l%Rz{O<<EX z=%(hpFxCx>D)rMcEe|5q&!6QoIf&l~R&;)rB3)7K$zKkZZhfOm1r_!i#GIMlH{V17 zJC7F<*&zA7!X=_*oWBQtI>((t8V9QW6|CQLix}T|1@(nbXnUxizf8D5{uPhTD}A&9btxzsX|vzLC8$3w43)3>1Idv99s) z2LQSJY@ot>n+R!f>}406sE2XqCb*K2kj^i9uvU9%RIy0XQzaSUDz_;maFS` zP^t0idmu?d0u3?hz(m!@Wx(XiMR~^N|7$LY!U3@L2V!PIxIGkR7IWXX;DeO$KIakG zhNTUfZiF<2ao6whSKoTTFZ&ie8Q*W;7Q}V7DMZ6-0h6|-3cz{5F+MaawLsj_2mkb%FQ%vhor+R>@RNdPXcy{#wfC8o zmF6(>x_;uDZt&Sfo=_7|=$lewWO$63qpM}~N#kp(6Ijz8buD?X40;49p`Z)}yU)#y4dSa} zN40UPWPOxNWgnaNp!v5X0@(L?uEjZ_ogR)>JW1eh75DC)HmA-Xg+#p%oTfk5)OjsxtuS#gt~6le&kOvqQ`#@cuYx+Pb5(@ zABTXxAoV0KudJZxa1&21eW>VV!Xqi18P`w0bd>$%=FX=G^){sm{r>fI96D~b)w{Ad z%7t!v9muYXALB~pwh_O{bl=GcF0nzrn&UZA7(*&J$DG-Kie|l_^c0gbykWRlTS0Wh zR%mZ1EbwnI`SZfu)$wvej}f&w2A^EQq{5;Q8}X4R350Gox8Q-4CPuN5b9SliAS+o` z{-E=R}?CLl&TW>jFbJ704)V*QjpK^^n87FajicV)7I_o_?k+rwB|0#YV602Y1hkBGSJGC z0g6>_H{L@AdrzLjAr>AUF38pLIH)w&z`9;!?jV33XmK@Vw+01=a0tWwqtTTy8j>hX zgbGgsV@7h|faG=|8Y~4-7Z@%5{o|=fb0@uqF*Y{2fNpT~H9sOlk(uYl7+K206#@zT z&EowSg^6!1(milCWXL^eD3dC%z{u=E@?cI*Twt&zfYY9-Q4<^ZSO}kpsPWasHXh_c zrS7IcdP3_;%{c}Ym1GsiH4;g!OAOz9&E`*SQ4j6NRVS}#9mc1Rb?7d#5}tU-rT0EB zEiGjPrwnmcA!P}&Lqki;mk1Vp#7ux3TqBhHuhQwy{Q$Kk_?+y08YwEWR4_9O9ld>< zoM+;sdV71sj@(BDr>4*S`uJ1;(@2+Cm837u1695|9Ui)jBts?UWH8!04c&q=3DP5) z5TYJ)y*!5`W^@pR1X1ZF=x`*U%0jAr2}C&t4Ry^4ir5cmARxhXl8MGGkiDLup=pl2 zOqK7jQqfKzaR zcRm;dVAnOmkB`6qJcv`3pll*jbrf+<`L(kk)AxYwsMi(iJi1E4Tsh9n8)3&}LZ}c-F~3aa>f^y|z%~Fkd=C!(Q|Yja!=(o*!SZOyrfJv5 zi!+B78yOD>}~hx>0rhsB?Nc*HFfck?_IwS7#_XO-4D1 z35Dy^)4DQt({9;jMJ8p;)-=uK#aHU7CkU=f;-Ja{Tw^mEs|8{JvxG!s=*Dd@ zh>#K7E<;8V#F>GZIzA-Ls17XQfip9cXl`rMm8MIhuMAY}^4z+$z$f4IO}U(@)Ypfy z0S&jw-qe12M^}}4+-fMW;|eWg!Vp6igavGDY^pvUBti)w#Wk;s9floLRaHO0pagD} zgr43rWVT@N-Me>g186COtA1)kA6Y`HxCY}PZWQ9tSGJd_k}~@T`$2X#qe(cbf&yhb zm50al=_dK+WjBl@ClQx4TgA}0{dxTI>3wun$(QjKXziGVtAF(}`zr=0FT9h2L-rh( zg8j{>WrX8t8aS`~$M$uazmub^jg|xePW7L9Vs~CX73RI;)2<%iT`ry=Ll71{8ai5m-rLiIjKf242y8S$2Un;7 z@aJgmkD+jol&{Mo;Y)zJAox+Mf7(S?gfWpB;IBEU8BvOY9C?LjoKrbHgokJvsCcP(n_jq^f{80S8{_7NR-;HtH%f+cP zhowP0ZJl^|x}y{04{Ld2gER{Kii&^Ro(z_u^|5naAMdD2ZNECS-JSKPGLG(nBbUx&gvP051|K9 zXkHf1RTv^%(lGRh zgmgnE`};$trZgdqB6;QuT_1h0h(;P~??O6oT=)y9V$P?_J!b8v&fcXlu=EJnF(SNgOdd-0iP z;U9jaJ7nzy1(MVa99kyO&a^{h0ER#I;4?0=vKvsJv{m>UFujSFp(i~A z?c`kKlz=9KLw|Wgq1Y(Lx+ilH%%i)17Yy0Gr0F6b=Q_B-0AYJ0t5Q%z1WcAqQxN-o z`$pem#ewc$fzR*LYW5q$^sKKrQ1ZvA!`V6Rji2u?GK6`lN=q@a*Zb7H_q=`{9MA=M z6s0F^j^?*ie4*hKhq z!yj1^bim1dMZ%_1;tIDFIwB1{x<6AA#oXXZ_vlo*S4;XZBZc740%~YR1`U$-(YQPz zhLDW^u%;bE`qdqmpH4XKig{PjALQGx@KGgDH4jCUG zAH*}kJ2Tm~unZg$$09coqf)GpuQ2X@>rNx}K5g-#-ZvsuhXFT^+l3EB2qHTYqmju% zFz-Od3JeT<4#T1DR3Iljj$*9?Ftjnw=#WpV@Yt7poAM7G>TA{yLcDV%7TbmPX{|BR zN^8Tty-2dDyXvF83Yj|ceC^hILeo%}##P|+u3>PiDkh;TxLp6aU}5p2;==Fu0_p&` zLU%arGOsEC2KBl)yDgv96KlVUVfubUNb0_Lw&bcysRI_jc4vNM*KYl%2yq3GP%xX<{pQkZ>hUAST{qXsh2dcjfW53#BIan>D~~!eQ|N@|`B|*85iuc~dV7&s*i{hPY~B_LK^S zFaa6lLx2RyS{xphG^g|~zl=HL(T2AX1nm3T%b=H~nRb2V+do+1#w;;NQ8%Gqn9Ua< z+V{+a$FW0`acX9U9}-h(N(IlLdchnPDTFl#0KEAKPtORN@>Xmc_qDu()y49wEgM7RKF) z2!1sf9vcaS$)!=#mR*9#cVAZe$3cO)hL4|zNSNk^DqZQ$o@bbP4@MEivyr0kgvz%y z2BS_h#lrinHIOZcf?5y3&sM;&BD=EiV5WclGI*t<6E8zS!h$h2UNp7R#!zGvbK^%) z5~0YL7a!gd&Z~WQ(Z8CvlX~2X;EPPa!r}v?;tn{u;3Ke^j>6cAy1F_tm5^j<1gM6Q z-&{Nuu+=O={L9jpH2}gTplW`}mTtNPLR$27ygR+KtLb4^yp` zLUZe(2_RDie_>N1QZnQLgaAKv6honnfJT-FV+H?;5%v8l^`@<9@vB#P`O(_LwMbP3 zn+pXK9xjJX_pyzXX3X9G9&{~>r@DyA^}Fq=ZHCg+6MNAi_l_%>ASL|^|3V!7U6j-6 z3qYsHP%2a>6KCYoEmYMA@pufTmYsTUpKp0wRhVJ@@$Gec$@h?-9p;2DZjGfI$S~Td zUsKmwIIgNMvWZBj(=*}0Fa~A&8Vn6VOD76X0cHW_;>f>8kMo$5^RTRhjaE^!+IdaU zunu3I5?!IBZho?8k*6qr0n4JLz*FsyB-Xvmak_99se9IheM-vNChfq9T3T8FRGt?* zz+gj5OP~WQl+4JQORp{-k0`dH!Sj>42o~j%se|>AE(E_0Y|s=v5eXt2O9&lj+UzyF4&lzZJ znEPKe$oxB@YekyB!2HZ4RqV|d_j~{_Fxzb-ZI4F)z^sVk^m-B|qy}P1+m2*7R zhMleryt+J=K)ll$mvflWz7LH8Z3?ob`jGlS@&$nCwqb-~^xx&uQ=!*f-T-YnXS4)l za;qjSNP@8sP(MOrH6L)NiMW!1UvW&Ks*GFIur*}wPyQ|!nJ646Y zDn_LQv;b zOlZ&cZvRU02*7)8sEst!-z>6Gmg5b$_1*?_iuGC{M#Nft@NFlS_6${cNcRTW)XmKe zZ5PJ1endnVI>rPCUo}ws=2%tzbNqg6-r9WG4Z)}2kWll-;N5qdbg{12blb8JR1BZ7 z>rcb7rKe7_zVQDtHFKq0WmoC5KMDmK<5GKyKBa!^Z(Dl9VH;s8xgLIG$TRnz9J~M4+4Id{|&0AM8|Qm;k<(gJJ6>R7f3A1&EeG z%dKr|Yt1ZyV36DaTc;$PFNa|lM2si-%(!t@tvc> zo%v@)KK6K#DGFs5qimMt$p2B_q>bL z$q0!?7vjpZeS%d%B_}4`-h_5mFDWp)n*E1lY_X|W#W@^r(c=UW4SvHz1Tscwvl(C0 z*c^0o1mXr0{+M?aA&e7VJC6i#9Oo42bQAN^+W0~!U44hyu_)q0PC{u*p8ufR10pN^Dgmrm zN6uT<=hZemx!RDV0BzZrHN%kM3f`sXvXYLO9waIud>V}^| z_pQ9*ZZrr+lL-n7-zJ6~P{z2AH&9{beP9_6#li@|ha+;Zv6k%$zZm5D6Q5#lUv7Q7 z?2NFVd!O0B87*bAVfzswTRAyn_V_jBxJ%@9d53j{nmCV-N8$kinob9(IL;&0z(T3N zdh=$VhzKE%hprE^1?&_3Ds|VzKAG=wpGE$;klw(lNlNaM5a=uN#P}Du{Al|_cQofr zics^hxIb5qUis;PoBv(@SxQLPD8_*Cn#K?xJ9i@RkD~)O0)g;vM`10%OQTN;R0=O1 z%hDI}nb^27cf-JNV93@EhIna}vI7mmcyAsv7NuWNO@5BpD?4#$ota!A$e5iim#&JL^slS>k8}5O4nJWWdDxpoqhTHWA4=@Bt z#&}x*E>Y-ACI5a-!_#WPGFOVL*2J7do-By|%YL#GO{PrSkCl86e$(1q^=BQ)R8f<9 zC5m>U?*q70>606%pHFRy^pKAeFAjJ(q||s0D?fbGg|3q)b?AHGFNTmI4dOyV$%y{O z9ZxZ0=j(6jc3>Z`L%p8bEf%%LM!VUjF@gOj2Oq~fUilNxE_7(^n+dAdm98dZWfva0 zDA!HrYI^qX)Kj5PlfNVvq^!wp9GI=rfJOnj%Ga2Pxzrr3GRH*<)D$;*AVQN0P0uG# zJwBrk7$8c8vs}`q^pStZq98_X%1!>eqZYGfUX@0rA18LLo=p8w-TgDoW`W6MD)nEF zyUYomt|FWL9O>CLlDf_7+zMICKTq><2UBheLcK`HPiJRmiA_d`CYGM9n*aS^IXA-& zZ7KZgFUcWoeGrZAiB&itl!jp!_4Cl6^b1C@=27%N_k|^FnYl_C#Bq84#Iy9e;~gVB z+r7@k`YFI8+JsNEfC5^PhndiAB~lrjl>MdoiQ+uXu%^_otHA_fd*_j!|=a)BeDW)>A(pbHGlpe7=ZZI(V{!`E0*FzN+2L`^0ITeTg zN?@YWmT_x(!Y=^`?fBw~`zu0BG(sHK&ErRv9noa26DaThJ0Q-d~DSFV<& z!CN1lFVDY7j`6~GM%pEE&;2ScJZi)UMAsGrZOQ>Dj}HIdTvF=l(k5OF+sW)oub#BC zj15@}y5g&%rh^aTrgxL~NL|zTopRHZ7`tS2U`NC^x9;)@Km8c7;f9xcf?{LE@`z4) z(JAj9U1%E9KKAS3*<4baw9j#Va(0RF+z{eP42T4dT)pbF?JY(&nL8#XOUReV+NGav zI(M^nERElxhbH z_1AfYXn~GrQsq^Hd5#U&*k(#9t2x{XPt4?~GZ~YD)K~K|0&NEi=9f{juh>fa(#(6H26Lq`XVLUk59w9?x!l!6CDuTE3(>@pjp;QdrMY)KPR{ zDQ7|6P*a6IOPCb0;2}h^cgWy!n7Vn7(0FJ$v(e|zdt_y#m=xw?B12Nco(l;itJLSk}Aq))eB~l zXTAT`=C9C&>hDqL@f({E5UR^sJtAIZxBjyJ_=17C);KZN2{sGPQ%4ztl3lZ zY=izp3}&&O;)!p%x+c!fAIgxR`?b^lnqHIArr791K}>IQ zNU~9pgY>$`?qBQHIk(CiBcGO-PQhSSnNPR(x`ORRm#;!a$v2wSi5c4S)9N9|{VZ3V zPpbym_dqv6I`*=W>d}!3VUm8s(BJ01_dKti>dEytI`!I3g;XOc1yN znEpilzBy(Oq7b`607+|pEvuqc2?1vSTfh6cxu4!r6$s|b5;7UPf2KGld%cHLI) zv3D1e%%q$i?&6|*&!Sm=`7u|jF$34KAgft03j9>)HuTH*Qy(04U?JQ&JHYH9Lq`l) z)6N4HHSwTp#ae5HFnUMEqzxbbxJY`oc}rg<*?dnIb>69y%^Q-E85u5>mu`1npkcLm zr6J7uRHG{H&}o)rChKgd!ph*KWP@&A@eW5t+2GH$wJuZ^sC>071ER5EO6U&>`2yLc zXB^MWEhVIol|AI-Rq)8IfK8}Xj=KW3sR401agKqxdAnEM?Xt`=~_foh~q*VvU=F_=FRsuw8 zkU2PXY24@EMi0LrkS6xiVOVC|c5iTlG-y=#HW0=4|5Im3c92a;Fj)P(jpX)S^QY8q z$!^@Zl=fw8>`3y{$&^$TrW>*2B0cV#$8w!168n7JTPxUEFD$8fuhwGM zSaSZ&;|s4chH#1SQ8!-V-W`H+8^!mRg?HOgVf(SBt3gds)KDUClO_b9c7*4bEcT##y+gORq{C;IVR6z_t8MbT7qHjfu&gjtBbNFan~%J2!RTCTAU#{_Xgg z(?`Um%j(vEpob1rHhaK5`Epv zzsg23hD;}GcO+PlzrJcm+5L3C?cJ&(so*cg-k*c{q983y{r`3W#I~PKn%X$2Y$O@w z7l9GLR#pPE>|NqQT#C8x&kL2`yzw@W-zw6YA@|%lw-qD${Lap?kG12U&Wihq9UFeV z5diH?`M=h~N9gtoNL&%A46ogu`e&1w{BDw8`#X9DBV!jHMG0-FVsNi^l1!Y7@?00X z!Jq&_L>M|25dug2t=|eB?2b$wUif?>LPJBJ-8>NdoNh~Bpn%nXO%{7Ax8zxbzAI|w z{d4P9aGH7vkJJ0#j9LfY*|~U1nT|#XOHuBoj`DV*olh;X+^Dq~Do8C@M3PBTRHnwK+?t@$a->?wnHWtZjq1Mqmz)pm6?F0(Y>B11d*(t)Hc znZObZrMHAcg=#O>)|={2DEfW}%=^{7qui<1n`YD@A=f%ywk*E%;=Vjn2=M|U8#m@V z?E<6pMU(w-VEb_zNX?w$WcSppZ6=Bv2gc+VUzCkm$0Dv+mLJutJr!A69bv+}7NZ`f zO{hMY)CmIC6kIhL7vYHq&<$ z$2JEvzwNAUNV;V3i6mH*7fwe!F9b4FV~Y!Pv4?7Q`6Lw3mY#d|toY_9KP^dY7T*%H z!65lXq0abE^Ti79ecd1EMoM_{xBaz$M7gWj6=l490V~-N+r(5qA6t@EVnk+^*{-?n zYn7hk(FQr~?>V;K9dzUIrnw36NN1bePwR|cebp6)YdhbL>6p;x2WKdvz)OxlyrZ}` zb31QV=GSTH#GYdwGcN0jh z$EgO~j(}x+k~%mzrIQqwtg)MxzANwmSF_@rk@#AXSQA}c<93R{6FsG4b5^5;8^<}{ z3mVbx5^OuyjPa!R_r{mcd$NtQ^~`5yjmtRd!nTgh*;{^{^O5v+-hbaLyzuUQsy62V z_f<}(D<{gxGI}|(h7zjOY5%I4T8EtvPYIh9SEP4#7N2;t!*aLmW|}S29Ne5IOeXVe z)E6&|ygq*A@_FHtS|@K!X6G(aFlb&#j#-+Z(%KSqAzw^6gRy~d>-0ZxJv^0bR>BE! z%Gl?ZGpg>^yF4(%ufehL8*H-O9D|H#cadF-W+03Z__eFa65y+_`N_{nz2h z=!~hi5^}w*LpS7}$uAn1b2XDg6P8aNBsV{>vf5`Hw|B2)6NAsfVd&G+GzpZDz8~#0&ZaO?=;<}CWK7m1u7j|sK-|(- z&HY}yeBKxy8BN7~N`A7*xF$xZq~y2E`VsF@>z3jeAue`#yp$ySt79L|JlQbgf4xdJ zfLf=xiZX;zX>aXO`60OxyKQZcLmI-!e5DFB^RWKZvEoCo|fr)3pp$ zdtUA>Z=V0o+c8!lB6yFH`38>C`?H5^DZ3+tbY|KW!dW_!f=Y+`4Cr&?(u#myv3|yw<%-;#Z{te`xv0bCBGfu@VpD0m z`qu9obtyf3RZV{ETYYO#g}Z@uI+OG0m+(}5+_54X^A%HZVMavXywg{?VqDtYX=Z9U z(>6qPcX+r&pMR43B4bLIoeuBN?Yz9+N~&ogJ>Q@MMruY;nc0J5P5z4&*#|e2c2kpH z3RSGspvE#0;}YP*i-vfvq_L>6?THnhqSmUfpC;G$@K`#su_xh(&(hDMj90G`?=2}uk+k-4TFxm~pkwKmD7E3p zR?RiJl`HY_Nj)#imiBhTur&O0%hR3%LQhbL;(D3VK6zbCY;M5_nb`OH zw`kc{ei3c@jE3j#dw&1saY)$KaF&m9AA2v0smAg9RWtOdp=T5Pzq+Q7N=O@p+??pOd4+PHDqHey#fV z38EWzoBlrTo#y#uz*uvtFT-}8;v_>ze*Ls)1&wD+j3AfFX}nbAy?ez~i?`XYosqdV z-&aCDk(uf8qcG-FdJq*oRMB7!F|oVQW_#zs4xvN}p$?IeAFR7Ksm9nArAkT2 z{rPR2&%A@SrsmSd48_JIrt_vJz1PV+3$NDmds#S#YtONLB4r=X^SZO%%okiPgxh#{ zef`)0<0z@%4aHa66cjL`^ZCHFLrjgn`uZl(C&i)ybo#l$3ZV-I{#=d%Jk)J}s2}#d zWpi`UvKw0<*INnHJ_i-B?7afkDmJM3b`@AJD&AjV}y&Y!j>y2HQnNq0KN*okO z#skT(Px|uCqN-LE)tF#oCmv7fQGvHq6WyurnA}g7@ROVNba9jW(LGbGx@oTaNmx-; z`jr;+3fpjjX1<*-srq$Gt|www#h|IF8)c4w-dH1RiEjqowyjAerffli9pIuD+h7#-lA->+*IeTG@7)wVY?+2IG8GGwsg{tc)$Kt0&K- zW5sead3y3rZoK&@*2B^ZT#G89uv0IMAeGb0*%^TF^V+yXV%0jfAmubevPo zuezwM+~|v3+%z>`w{t5N@CAuI7Wk3!?R1r8&GDo94w)3R_7>BPa8{lPq0J|t@2Fy1 z!HLZP1{Axm=a4#Z8OC6$B|@xX^O#&-xoj_F?ssa~?>qt~OW(8nlAeikQ;Vu!!hOnP z7bVB1sXgZfc8ckCXI~ikdZDGPy@lR?ompGESomf2%gvjwk6r~C;JB3cKyKUAHBz3t z?^u{;?{I(6(nMA@KZXyV90Ha}Tbz4TzR-_{c?vs0G@c~gU6Eql#UJhcEou3+<)Kdm zrv;LeJwQMLMnO2oeSy26ANo(G65G=oDgsA)Li!D5+Y(l|x#|;Oo|J!neSo6S$VkJq zVnXTkX@R!3sozhzUnKB{gjU`ts|+X`-+M#D^R&lpiaI?W9o5QBPp%a4kjSU^ zXPJUCXo0x>cZt{^TyX1<;t{02%di|0=4_lF!i-FCRbVB!;JL>^$PjXSAfZYBEw?xR zewgFZGqJR}F}d9Z1(iddRtadmVbj08c9`Eep+z*)ogbzza_#m+)d;y*H)^*|wYVq? z1>F~~hN_#%B-6F|ubW=TbXU^jn2V2!QiL+?=59;9FyRJxv2Flc9D;V7vIA^D;Bi}U z08A_c{xX`PO$Ok}|10#;n!B5@spx{k{I6@J4C*A~j+}>Lxy>wAna%P_6<$hQp27iK z>9*mlGn?+YiEvHR1Ban~@Y4;7(Mtm&N3DN7Zo%X&rR{cV%2XC3xju%9z-9lDla!LO zI<$kX|EZg*YEV~tlAX?CNqD%JY)=S1c9@S`%vJx*uUNrU#`I17dhz$6BY7^iUtO2Z zKfdm`C|n!0V=>A7I>awdZA-@fQ7pPYy2;z~2Wf2OY3~wOD$FadU<@O2f9yI_HiBR} zhMVp@6iwfRjUTr7d0UvP)Q>!K`6Z5;o}|ykrAc~JdS$Hc-kD6!;Nic@`fHcX2JB|A z@nidKzfWIaxBK*cpSk$r7{M76(ty^aDrxWxt(Jst7}-3G^i!CT&VWulfr%_^vCBB& z1Sp}u%GO%o@Yd^_LuQ`zO=o85(H9nueL8DcL@o3EL=`tKhZ1Gkt}gdgp@YK#@=UM( zXTyl|*y6{B9hQYG(iJ~P?2~R%JQ=8{XU3B!n6Chr6AD1`OIN$xlTD)?68f1dAA#^cV7Y7iw@dKTFx}VqGNmV0}8NVR4d$jH5@j zgwQgP>r+`Cveh}T&`bwK+F^>Ims&HmSmK4`bUN7Zq(M2hkmcYPyGTV6S{i_xyo)UKt?40Ddw zaS)BiY%rYC)Q|Bx1vLb!GT%$`abobSQI|TteqZqN%_^IeKHOoH&94J*g(tdG>q;OOr>tIBw1yozNP=EqHOXk@YZ>M@V^4}-R zW`Ff60R5kG}?=+ILDU7ut zB}4QH>$HzV=Ga{9+RqORq(%W*16k-xey3~Dc<$lqEjRC z1_~6`c>h$)xbaf%7JDO7n|E*nm0i0*?i;7f=3VFY2+?Fi$|2Dwqqv7jF`40!t{XcK z$~7^6$|3i>_*G9zBFH4McYTJ9`f<{j=X|;Y|7FgE_@tA?i+2Qr8AJAw$Kaq}yRuds zvww0&b|Rmcg;M?T4xmZYaY^po=Og0{HO;R>c8zsQo=GLBK=x)DCVF!d?tvH zmP&8Xcj#8VXT9kCRm&0&={$JYzA8=CCKKtY?(k6ilI<6_rO>zU%+|&#f7+Y5a?}r; zX}5IQbm&jV^ddC?4RjiKd5qnz{Io5~nAu~>3`>gX7P{g&QVl7Fl?RVAoN8~0kHk0R}sV5BeA#8v(w|D_(z;wQZVQ(Dif z!#hYSvr&4`U`PORpzKpVHSFh`u9l%kKjK5_|C^1&PtBo~;zLi1zcmZ9qSwyz=3SL@ z@unrks`KN`Vk?)a1--u*Z5JcUDk*Vd!^}7k*BH~(lf6k+r-Y6lU|6c%(-!~Me2-Mj zF7(-kZ35f1H54?G#h&O+AAMC$_pWT|R)*-->Wg31tEg>dhbfBn`R>e(iER)j^(_4^ zQn8U?l@FT$1U*Vg(N@TyX^akB2+~^KB~Pb#ech<_W?uGqkrVoXEGMi@Ljv}vIZMbs zH;QlLD3z(bA1Av*#NC4CNDqf^)he0sRj+wr`N)UuA&d=P1SI00t(=zF=kXF9wV1bf z`;^1Z+BbecH7pFdaB>*+E% z_9qVS7Wd3LV4@?;@&1->Z+Drq{CJ+W+=Twoa_LFy;SS@h zz$%5W!UMt@9L6n7wq~A7diA6^5@%LcsmHCmak9Z%1crYS22+gAqPG!)Gc5r&un_FnxHNXa+L;=$8A6bp(8FlII5&jhVF0wfil5;iYu=N}Ih&EmqY~>MQfV~kDp_hUe|zOdU`A+hwMH&l9RZpl ziu&W9dZTE3sUOWSdpMlz7ZV{*hs|^X9lkh6*TR(Vbhz8MfHA+4MakXqe_D(8-D-;r z5-qb4^8VUIeQ!o@Rmn&4zT_+8Zkd^O}e<xKOD=`fC8cFVrE%aR>V-~nmWrM zVeI`_X5;1g>9m)ZX);}VS6{#G_$Eqyv3|HD;LVYqYj0TLY)-xmlkj_ih2=xDu_aOq z%cUyT!`)39v+?7(eQqzMD)g=u)0a`l(oE%M&TLKDB+@=8GQC~TMi0(qE)|`Sjrna) zWZMa^5n&fXtp=HvHPn~I&cg#EN`iZq%B30dFE*17E1PhRA8c_%`SmI4{$r^Q>Eb7) zcW#*w+TI&PQH0qNF$5v(m@q?td@t_1yblo`1>SF)@06!{8*)&CvDRect7KdBc$Z0v zbe3Ja+2sROU*nIfJB~*T(m=fVWU|_$cScPqOAM)nD zNb>V<1W5djxkX*h=u|+#0>(z22a{DL>DXdBtWB>koG+@TesZ#6^6}-JNeWf;XOf;D zN!tIzIK?!mzJ6p+k)*-4e{8D~7^~LsdHruVDPD5MDo zs3piMYux?d*%;3aIdpJF-M_aQ%pT<9EZs@qA9^-!)*b z&Pn`{75~0vQTRA_-XYt=lnvCzqobVe-P5MUnVXV-WWQT2kXZF<5iI)K3+1|6^e(maO_utnh_XcmJd9Kp^H{ z*Ksd4Ghq$T9+O#GS}L!qGD)ebtc+=Gylp~v)i7?`cyY|g=To$EiR!Kx|Bh@O`1XdW zypZcy?fvcLK6bv+frouUD33Y!a5G(A61@D-eiPM&oHs$tTEa1*{6!8kC;o95{cln% zE{5GwUhjc0w6TM;fm1}JQ-h~6s6p*U{+d>XtWn&VEaSByvKG+|zjg*1@cC>Ax^Src zJ3JogeUw*OY8?xuu?o^^okP8R2|;H-*iaI|2;exLUtJo$y#BS1`14Kpk*7h~h_A|q zE8EVSUz;~_1StQlH)2wpp{uJi_@?qPEuvf<<(Ci2(2%2*@gm-)f<~R-t83JsC zmZ3M*;(%#QqA(n)x^`Mwja!Mk!AUQSRT%K3@%{kvYuU~ZD5+E5` zEe+)W>w|x{*qlbPtFg!CLGxsr9Dat8D*KRCelyR*zkkb!+*en4?0^2>@R36V5Ndw)fq_!rXx*dC%qA~5=$IDB} zxOMBWYJhQock<&$i}&juS>;JaRA?c~Au|9Z@Ph}HP{YD^fWoj~a0&}=hBM@! zac629b9A@SoT*lHsHEWkP_vnM)06LH++)cx0mp-O%a$_l!;Veu?d@k}i&K^6OmR2E z*yKJn_Q6wFqb2$-gB$DL1Nm?1yE$}C!@~(skWVZvf&W54CZ{<6cDm6D))__r`Uz;f z|NBS(mF?S7-wK#7w9H5~kb~~HA5!;gzO^tE61>%D0lf_H1u@vqe zHs$(dzWTpq1f#A5Tk$QJhnUk6Sq#wPmb}u5S}n+7E@?yfL@Io!1SVx><`j&bxJ)CA z9z#}YWI6+%Wz0FDn`s%IBfvrSw0|S=ESB|B<{cNkiNFPHcj<7Ps%=BAq%h(Wq4kc1 zfJPK_1JD5cU#_zV87pE;3v6T}UJyLIPDelA?@KkiH|DeVpOIaE0l{}WoL=v0ugU@s zmG1p<#I(Y#(f&8yu2(tH*2gP#Cy0a8rlClNzl^VY2O3JPuL9oqvjPY^XAk;+3-roM&8lf{FY_@9g;&&1Ib_j>{Is;n+?80ik z?CVu)x1(GPA^6k7j~CB+B58m~Nx-RaPWsl4U^?6~bKnjiONEkEV-tsTf3H%>!{lCH zJZhUqM{DaQ*%+vq+1YbszK>Dr7|Y8C5+MT6G!z?Es3xcM{djsuMjNjoz{ArRHT|Ja z;@}B5Y8Og<336Z$&U{fZ^uC#}$;Fw+ez|f$H6*j>D1Mbb5_pb1do1ahWj|6~u}G(X zlTK#_Ul%xm@VlZqY36mwgC?bFfcG&2h^ZoYrq75!6-gn_33EK&AP2ZmV`0)(xPr7xP znEfWEGhm9uVBc&Kv34;_8GRMvxC+mH;~*zOkG^oco4DfP0mB{ifZ?Mva6oU{xl;v% zRs!FKo3==2qfe_Bl0DjKN`u zgW32XgyLdCS6!LX+lbGMEj}wSD2b4iM$1D$chZ1ROs=*T{r<`Ub)IK!Ro)+`k&r44 zQkBi@?aK*&iC6lgPXY9T@5E_&ZQ=vrmH}=KProCnuze8H^Fo+4Yrap== zYNZ@bXlaQX4c0UQqyY(f*zMakuo46~)v)eWIMarrA_TLX)*zZpDmB<(aow^WAk0QBIg5!J~6dHbP@I8fqm4M0s8{NmDBy<;i zb^<%xH4joHf&Q=gIVbpg{o~cag7ems#6iYd1J6P4foHhmnw^2yCjho^zdeXJH8)t6 zH|q;0f^A>!ZfHn}AWYm}Z;96h4g>+Ilz(5BAOOq!{J`xG%(YZZj^kylMxgl;_h8YN zWsd|Xi0y8(OH625y;x8(&S%%fzN?Fu*EFQSHnWh(>=zgRUb=>(b)Ue?;2?HFmpeEe zaq}iY=|b*K)(dOl%zNGMUct|bQ#Po(XVRZX!p2TOIYCE9=LfEP`ma>;r*Y{4$&AQT zSj>RT9e7_fnc@HI^C=EQwfe(|6D}P63m!J?*85jHYwIOGw(gUgmlLA3FC#wa)@~`5 zW5>z{r#-vprT+)tkHqTJK0e=_R#^Ve3BZ-d2WCFrStFKwHDp0wgR}HON&ov*B<8FT z)eF!%h$Wntdd!PZ8E7m)K0SBeEmqNWKHdPnXJnp=^`t{uwt;Uv-uCMJbrxo5&+vh z34}AZgano(UF3IN`rOFc0|RL{yrkV&Nc6y()5DWU7g2d;a}O-E=B1%kFR^Le>51#5 z1Su0oi4f88@gxF#tMeG={tO~BGjQ!?0Ewi7RoJj6*UTg}Jw4WUJ%`^uAp(}H6zJxx z@0xSuBCM=X(q$}w_(RbRTzAH1cJV~)+h<2<&qYJXh_&wM)SVkQwBWy+0~1FZ%ZVP0 zH~MXDZT9d1r%8KX+OtT?@O&(LwZ_fWRSVw}Tf3rh`t+eGA~v!cYvbR~f?G%rBw*sI z?S|*Nx!S*%sibYsl`nUVL3<=3vn3H@_y|=7<}2Db-N&XfZTsGRKtg~nV@Zb=3eELXowcwA9px%kTei4Axw!B7QXt6S~CP1CD}-?wNzxGff(|`xTJ*CCc&4 z((TDH(g8dulL+2~9rIDC=i-1Ko)m3gBt=BD$yx58_7X^qQc+bUIMCYgay9q#=qVxU z1-X0OXKNnZx~V;bHRZuFdJP2#f6pu;XLG>^N#@UR{IhUa|7NaY-IzmXV6B%RkE$wK zdk&|bP7^C_iMT#2E=_}SDESlDbqAw@P%l!Tx)TM!v zd8ZY@RCT?r+qNa65z{p^d;kG@6waaUJN*T*o5nL324>)|e5l%(ckIYWe)_Zr!4=M~ zLpMDVqC3;}pnlVZYow&dO7f>?slG+sf_L}8YCjsu9E5-65id%x(((MUDSak=*?aW1 z27jsw$ybdij+3|D`&Wt#0g=JiBK{J4pnE6A(fB9Omq0#2{A02G|G)o4JPb7!Xs-{% zoE|+3*S1IEE^^wC9o9RdRPYc*AgFozOY=nj1hkDf#f9ElbW=EtfQ@_t@c`*K@`gP4 z6{#oEyO?LW7I-BkEBpJ#@Ju4o(rhq^0j{&9lq!24xY?N~b3xxe*y&F^qOhi1KpBTh zKURA-m`{e)kM%^@ONQO+grBy&h5GFCY@faeDzwZIm{H zVnP&79E2FByjp{-wS{=#O~EKIjtI1(E91d@gfNDWs<-k7fUK%8qJ%6_0t!I0Q2bfEzk-jff;aOW82~;QF1oti(_kAHf8WIy zOgR7wIDusZybOY%u{{nPgJd1dBFdXMmN#ZDm>=23S4IvWMC?iE0+BGP)2AWFDfr zdTDux$uNy@bC)`qAQcK3FXwQ$iR7IFR{?;5Squ&~ljC`x*PGWaL71C1({xI1ZH53Y z5P{JMWXN#_rEvyJ3fJDf6u4H2;9>%ElKs(ItHlYSPH#|^6N5FtUh&u?aw&-L*SmfF z%#RTHQ;S?joUI@bB{SK}b_IO}BS8eG1`UdfyF0gt$UP!Q1{CbR@$OLEG%&E&;&Uxqtt37LVmX=pc=Lk`feK1D*rd-v z-~}wbZ1#@3mU@C@G@eRs-xiFDatF~Q4Y`iuc}wPks53)H(H#s7JCB56J?p~M7GSsM z_+MXQR{}3l(EMl=OC%8(N*Gg!O^+MAkMX#?L0k_clVfkpTf@)LS^zPM@ZQ z%6JMzBmf~>2(78jOBrNkpkbVXyNAuyi6vQW@FUI9)MSXxVN0rm%+V z;<3T<(b~27IMNTPGA`EqXgaWtl2BmpUU@1is!T+DgkLJPeJgKP>pwifPCOa|2L1x7 zvGqx$VuXZ*WWrQi^o!`jeEvXykwxYS+hO>m1+)Wo?vEwho1$3oV1cGZ59#4b{ti=f^$@;x`<{I9q2xK8e`ZtDR4r6OIE%9ZP2EVD9cfhmcE#&@mkyo zwfuG<GJy3L zo~GZjg@%)htNxq`t_RX)4dRjx=qZqDa?XNsgQY5-upK7i?63^tyo83U)>|A@B3WVK z66!=mh%J&7@pRaW)ByqjQ2Ig4b^WKO&chWcj}HXF5fL|?=S+xY4tJBP#}a`;K~D$c zLXWE}@?bZNp#VXkW1XW3b_(dP5G)Fce1|?^1o=1>90vm!wTUathA^=<;qw~d`GBmn zEkOH^<4#J&lBqBI1g}4d+n8XQ+03F{n#NwrOm5ZEIixoedlD$|Og0VN!s=H6TX51m zkD?L5xudaSmyM1+Ouz;T`t=bGl>HLDUIUux#6-^QHF!a>!FJP3W3mPne_8+Z`}ZD$ zQk{-^r>riWJ2#O)?`VaC4v4PdaLj`Trx88a<3!by%7+L{@;dy=cUW+*{eVYP)x&?% z_b-U+GGywy+4>P%xliVUxIu_w2pbB%7om~<6zlgdBJyR|U@5*ytU8WfIoKw7fFq9* zl6~aTyb$pR7~CBaZu^M9!yNxybgG}1mjyhY1@qCvW2Pu!(6R3!S|ns~;2N?oScFW2 zFZB(@?n$8~;6+$|kf;tdUCA@8I9C92M}M&r*Pr3QA6*e_4Yz&z?c) zrX@Fwyb&_}i%!SVrku16jcoJ+KrH-M(^6A4c<7rgOQ@`|6RjTF$kAb>S^{SPcIDA?aV3>ZB}3S$0|FXM zgxh87LX*6?IzW@B$nSN$Z&F`NMCuM0ep5=n9Jm+1X_3@Zx= zYWFD4Q?*-eb<)&G-wtYqyq+F2LYD3X22zC4K8lO`JlnS56EA+P zx>_0?_)b_~eO9J7nb%yt0fc!}Z0x}uepAiN2)ssg+b%q>9>m2xN1Hk3B7S71t+llg z1PLuoO|r7V2wV~MHbqg6;%SP9k%F>ckgaI ze$0XLh|f(yQB{@x086+46XI7VP-w!s9z5(E+GF2{V{U9>@_DL-H7X`1YiVHN!-o^t zYo9wL`}Wb=WFw>E)l7@U`bVR8n7$P~-e>!TTS#9=hkJ9Vg4+K357hR%inC)Cd@3Qf zgW+c1d^C2~4)2U`!=v2J*4EI?36IWuC>Zgzdk-G)fiWKS_N~_8xYlz;?^teaTD5Hb zc$}_l?(Cz{)}GGJy?9mRp#vN&k^=<6qHKQH@G2em10q>S^ieM^_ze zY))R5d16S?2c5yZ@Q(fAvOysb1ZU1Z$v~y1Iw)H?_bm1+uMwY%<~IZzh1FhylQA+@q~$$Rn^$|5ms8< zn333*4#UU8GvQS^_JD4_oV<(6>(A5lGfx^U51r78(rXo! zk_yM#J^`yGea+l{w6`O5hJ3d1Tq+j;BE7pX&*Ti@Qqf!%`W)R(v&xf}0d z*Z&2odo5$*tU<2TuLnd$b&Avz{PR%PyNueIg>QRXUK@*XSD3!z3dd>h-cB}37JTz=+m3wVr#zYhkj$c(B1(<0y1jc7{;z)Y aH+h+bp+9xyS0yt1r>vx*m@97?_`d+L$dRZ3 literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/th_lat_curve_llama70b_tp4_p2600g128.png b/blogs/deepspeed-fastgen/assets/images/th_lat_curve_llama70b_tp4_p2600g128.png new file mode 100644 index 0000000000000000000000000000000000000000..8b69a8a1718b0edb333713d6bf2b4e8d95d1d4a5 GIT binary patch literal 36645 zcmb?@cRZJW|1XsYAt56q$sXC8$|yqi43WL}9+7M+Bb&&c*?W~_la;;6&fc8YRrk4n z=lpU0JKx9Sdw;w0aec1qJzlTpdS8LhWhJj;l47Eupj?0YL`(q%<&qi-3hFzItME6r zTof_zA3l3=HG4%XLwl!}wgxCNFYT?(t?bQB^zJws*xH#`S#q#&v2ZZmF}An2w&Q1I zwfOG`SgdS~Si2aje!@krSwB&?LqWlPiTsP2C6Z}^f)e)Osn|m$=Y+KhC+7&|lg2Hp zYRzgo`=6)=!Pw&alaHgBNZ1K@DqC|4HyNkDZB;+?;w5f>UVfMClQ)yN=&Q%{1a!|O z4y59DPUfgg9xE72k8-Y`RTXGiblYvu&)@sZvy`wT%%2|l2!rYk-lJb35p`sD;13w` zeNVh;kUzZC$Jf2~_h&Iz)c$^o`#x$k{DA6B=+pn-A5j+NPYk11%pf?ifq#D6^BeXg zSqwX3P>Gr3s+ZiKYYne(;t)f-kE%i+E9Az_pNNwv+gffh#uUbw9uN>9n%Uo1s3Rg! zS~SmXH5nqCslPSbvTrGstyXPMe=y}G(%0AL?BXKd;OU|5JcIV(!-t?<=eZNh^`SIL z+>+|6zOUg!$r@#*Gcz*}Bks$n73sci;S+G$Fn6Tj;o&hI%xOsr7nX8&cMsHA4PCm9 zfsrB}Oyqtr;Y5u67C(77NhM#yVeIu;6r<|lv@h=OUtg$SzkYpqa?)SzVCH$c)%u8X zXIRJUfo>zprAwFoY;VtY3vAAp^az(^Rd}2@j#gNOfBblhii%3`Y`6b&db;{RM|bz4 z&c*SqmoHyBySo0&QDz(})bY|R#G3w@E?v9b&d6ywPHbgmB`7GErI?9>D}0zTxXB<{ z4p+-~7OODtRkLru`&&NAZG&&L+#=ZD9~}m`vXVF0wdKs5Lc}xS2DJnp1w|Nq?R8Sl z=}?}@JJi%^zbx|9OBQ7srb2k;ynbRu^YQWBAR?0JNfH{Vb+P-BJQnx)^F1FQpZ4`B z>RWC+DXBYu{^*{c9+g^6P38&SzyG$xsAqO#!!k2EuvzfK1x02sqO7_}b&P@JUVkB3xJ*@2(YSN#v^@BRRH&q-< zB~Ps>OFpf4**=8$L35s^vor5z{Rs;S+}Uk@`+&mwwX1nWSA1_XqL@FyQi(a+i zON0(@T4=99O*&Nr@ zA{CJPA>lCWEi+T~Jlj(}-d$GATWaWAEo{(E75&mlC(mHm6_<{}<@`rixz5!geR#Di zOl129b&ku&Lxc8 zbH~TE6kRf96P3~=LYcZxT5jpkiRNoo6o{=CJ5KKI+7nVx6yzvnr`p=u<_WW~vdS44 z8b<%-lOBT$4X%5ZwS51c1N)$a-M1;F+HpN+d}2abLnAI*kdIFt4IMrAWTQrIVPRon zx7Oo$SN_wdPYir~BTD;lgN&m2FKULA%FTvZ(6V5|H7Z5N#%7xi7vztYnkc3Y+wZL? z6OxkVJbLtKVCJhoa~K)+RW7e{cbM}H{KN((g0+KGq&g6!$R_eD^ItFi#Hd=36&Xp8 zKF~GxBB5t-%&H+DWv@r%B4&_Mz%lRAz+kSbRc=?sicaI&i41-E`T2r>?SH2pZ}2l$FH^zQt-dkR`7|8IhFKjm26k zDkgRX7TZ9fPBN~@DPDS{dfnQZIZ`1mUB0}vwe_;m8KF!G&B!jHRY(}OAY2IyBsZiWIkT}iYxNjVys*#SLM2#oLn4_mG(TG3rS&fPGvc z==9_X1_}#HIa$d5_I4`l+s~Pq5iv2wro*3~M#S(~QG}9mHBU?=a+wXG6gh2b!&>$& zEM%RWoU~sZ3QZJnic8#aoKl=SZiF#;2m3K5D~nE0NN9N=`{u>zPI&g@;nrNNut#-v z*Y|k&M~^N^hTVQxI6PH~3r{buq!g(h3*DwS<>4zMqnl)8W(>(9Uc$FnwddC>roOke z1;UEV?a!sBr_a-ANJ6J@M$ysHQ9ObjzO}!uI(OKfChMB%Y3uNVt;8a9D0 zJQ__L(fLYl;gcs%rX%EpKRxulO4RDWPfjfv_8D4_;ezp!>miB9pILp!LA*RGPK!it%RVR*CLu{g?8qjNIJOFTcKx&2Z7I zub%rTbB_IORCqojWL$l*yVyjNL{YG&E@5)6=>o$FPfd`S@lQ z7acp6O3J>6k|{#ffK6b)P~(2Y4kJoGWIVvpBly+hU_#O#hg>QxmYYiPO?n=b8K~cH z(XfbO-ab|>)RwWeWv7-1DN;-O6sZTx)qJ!x@Y>D$y?N>!Lj^B7;C8=3=|4Q23zu=A z^gMj&c{qbE5lSjPR&H?w6H<9F>Bc;2y)`SYQDJFuOie;e{9|P>HleLoc zQFIBf(=GI|3M-y`^nl>txenHb&vr}Q&}@F_1+8XhCf&Vv?+ujYV!I_-!|sH@h6WM% zFfBX#2d7CFe<-Y5P3V-?ONmbGu=xA+Mb35=uVNE7{_aa_YioNs)9i;$#vKAUL1R5l zv&wqrlJ^LkvwXfrSrBwqiEs)Fm3mmEP{1C2eS5vDU9;TWgkRg5fsU26XIDIbc$Lrn zkS&5%E(1od)NHum*LrpAuU~2jiHThq%AP0dJf6qPwD)D=#P(JOvjI4v<5GNn?2mJJ ze7wEg!8*6H(tdF^br7ezqC@!ogVGL0-jt^TF))4Z%^Q}<8@defIY?J*r0ZD+x%A%Dy(bR$E_(3eR&<% zh?R4cUUtSja0lq)?CgwUJ>7_EYiC!rl;qagy}Y<6QDWHjtH!A?+kA7fZXj2+hY8R3 zwC6iCG5~XQDq~(p3%6hsk_b9mzo%3j-wXppFf%{@Ets|b2`m9vvBGJR;fBKnFEaQW z&i2WmRDXir^DIe_gqD^TR*g9PXr$U9cbSXyI*fO_ z9E;*b*q&N7j-jwGW7_KvXR%;9+o2ogoueN}KzGzT*>o5%k5^Ys4Zfxw%z zkqT1lV-<3!D{ZvNoVh%v19ZXJRj^~SdV8Nc9BmtvRKs?9?&;~7;TW|~@M+5hbM%Y8 z|D}QPDmxj2@Aa#LxgXhyu`{;0Ek@}QWiu(fPHhy`EiDT$adA}>`5o8=HbZ-&A|lc@ zYIkG@0f@zySFs^oyt})51&UmY=h{1}$B{KPHL#xI)Q@0fa#PdNm{&40GZ{te0LEq% z!%C)aoE<89`IwN5OmmH!P|(Wv>UAYOy;P}Z&y@UeDOKb|&Pp#SU7VlE?4c~o&ws|D z5H#LDgo;QH*xX@z{;A@~(aus-cz8H`XIl0eQqPV0`}-9Wh1`aeEEA;MHma7M8Sup4 zz0G5(6_^=PpjDL$C{_tdO=o*Nr%5Vc$Hflv%NPviYYxxay?g)u6ReSR*WKSQ8R7+9iZ3BO zoO#rIz2;Su4;s;#TM2z0bR;gzao+wjqht>iRMXX={EXs(0VSw4%*rK(GX1+&@;YWs z)h#a=1MKgd`-{+8%q#=}LVU6eGURZbZKZG!qdU+twb%_*6#5V4*$A zPZqa8qYM+;7P6{Kwr^=WqozliP_4lxL19{%I24geCy$H2I5F+&D{QO*Hj16wDWH}M26cl&ZKN;+BeV93ll9yX#$*Ztu z?$Ofr14hTM8-n@Xfr7r>jK%uA!g4}Sh?|DSPsi&l3QC9uQ*l>kXQ}IZUz=~#931_5 zwNMa{_6vKz#C6Y9_-Otn+{to(X1EDM;k~fie2;W>?*L!pH2?Ju&{qPW*I_+UxG&jp zn^y9S6BiekJk5$QSa3JU$t@UguU|I+TxC91-o+{Y(q3Sbz7eV(^wYP?o*mK5CQwzE zr@rQQCJB+HD}9kH#J+kNiq(C3`arn&_?nuA1}W^{5ORU&9^sRaCSUZO5u@a-U;5!+ ztY;pDQHtQqHzfPjEul?q&U$hO-Gwd+JTG|rf&TZH$&1ED(eu4RFI1Y74d*{i=v^6! zA^kkx4{WFEJ#GW5lTVYdS!gHhKl<6`g(VEy1>tI$54X?fNQS|_+;|V}hbNN=qvbH_e%1DY zMsTuUF zZ<}5hHM^u1)a>GFSK$|yC+h@umU?KZseNl}1>fS3o0e_BoLs}gDv1f^k)=))_88sm z2M7fdA3vC@(@;Nh_U)S(aQ7HKyZdl6rN;>$Sza*otuhpJc0QZ*ICuf`+yec-bLbbW z6wQI~kPyzA&7~!D-3MH=OG~X#pE}!B3bfiZt?F7}q5SCX=8n2`I;TwlaK->u-0sR? zaK)5ojaoRGTJfuE>gwu5{k|z@@9BPVr$Jxr-||5kGt_*U1Hkk$P^FQM7Q<~pdI2mk zXhOKR-n-%a*bG1_1qH=GzUIf2lsmz$Wu}AdHOVnCPb4JXwiw}*_T)C{-Vqi~0%Q`~ zjUD zD}?(=x7-_oPe$l4DJdy)m+wqE&s~Sse?>Q1;`6;Ja-RAIm=83zx{u#=?Zzu@{K`3i z$?WnLosOew)wxC?7zP$xFt^$X3w;o9W6ftKzDG`*iB6Ly3=M;3%#1uI<}6y3?||h- z>bHi)4tYW0CH8GBInT z9=u?R|0ohbm>e)dY3_&nAkq;isxwswhViKpvFRi z=c=V2eH-N-S=6pjStdfg;TxbWe~)>f&lKkG=V$Y0`eCr&x9@ib57BuzISGDvd0|OC zdBWz_@c8j#KBtX@LD%m~Oa)i3lNNbhc<#Ci6%-U?Y&V~ueVngJD>@iH)F$mY0(gkv z)m#Uw)2}A$c0BHf*5y`H-GE)i02Te&-~X=5)?4GW$=)&Lm^-XNWVttz@Ie2n(}csv zkrpl#cskwvwUM=nnu;eE`&N?9&XrwBb3jisVL91vjE|J=w^!~nfd~RD_bxSchs^Qy zYenrp09v=|hYNiNkY*K^$Hu@A1i~uoy-xTQ{=W`$WdvM$G&?1z(^?Y#_r?2Rb<3R8d{); zTs{iU*#roprKNSa(<4FxG%{NoC@}0$4o0(~Jd${B3n-*5=AYbbM99#f5rhL!D>fft zFIi%EX5#KdEIIDs;UVqb4Ve0i&71@R$e>eMGF0r3&~+P8F)-xx_7rb9yC_8VEVcyS z>XNt|8lu{_xVRW?N@lU%8OxEtYtwB0)aU({=^9)kY6Rtbeq=aXAwEp#B5;UTz`S-h zry6=)K@Fk-G_U7*Iu|Y?k=Y+xAjWxKyT$MgnDw`{`jQfk$8WJ;!T=k7EFqid9!;FJ zb#yF0oSFccOcutgD_IhLDUk?=(4W+Yi&V<7 zs!bCGaQXPqwYAB*_!l*fu{2DavLb3&KUh@EF40nI`x$pK#`VDbhWt{Rn z$yYz(nR53H4h}YX5ty7zIZ@-ZWftHy$6d{o9Q0}GRSw;r5938n~ z*(0rk22(59dG?x*udfSqeP9tWEn8$47SZ3GBzG3OvR~A21Eu>kctJq0knsI|8rhiJ zoD=2vX^C|j)be5w)=GBimY0`16Zj(HxXj|tXtT0{A~HVl1|5&&sTNW|#V&JhxW%OA z4GS!`OM6^OzMr#pZ%C6#v%GU{9{>{wvmLA@uqbM@NYsf*niwAQXgcW*fDi~MyV3Vgs$@dO0;2T%tPQNpl%qwXr`9SWcB1MQZ= zW&Q@D))i>iwF63N0pwl@NF-3Z2hirTi<+7(2aS`FMhuBFjj*Mf}J<6^J)P zV-p01leI2T>Ikc?IRph0;D)0?xI(GZmIIcdR`1S-G(PwUks=YuEDT<`s+R1lH9kSA{(yZ;zXplq{Id8#NyoD zVP)VGEOEiDZ#O{_*#X}1JYR!L+w;gkF>$dg{_-f=ojTXO=8=(+NWz9J6H>>`$u4-( zY?XXc7*>6_8N*hf>t@F~yDyw3S{~yHeFJ4k5;#Zp)GcO>Z@@oIxf2ie$E^CxEi@5j z1C;r8M1`HYIF|&adKoHXwsx*cJ~6jBG3dh#01{GxWWXNvhmsg9Y6)9Wy!xBaojcwJ z9npx+8^>+Y9&m%&xGxnAD)zV5))M2syN&J^@ADf35)u*)VQu{G&%}kjGYi%GGZfkc z0jDsS+nLo>*xRSJ{dT~xe}2)IobLCF2RuuT0<|mBcr=^WgcN|OOz=P`UlT$LLH6A`IN-^T05xSN{iC)X zpalEhy{|#t1<8O5fifUpxE#*-cL~RfpWTjno0?%#RPOuQrZfk53pOcdmOZJOf`MC~ z-hw=v%-1fpj`^gxxEw_DBP1fqfZCu8w>gMT>8|UjUZS93xCD}v2JA%{~m=0%Kf=y??1v1>d`>3kAI_KRxJm4=pTy~4k6ccANOz74OYdep=I|8Yd5rWR2 z||%q9zo~UMs5jOS~l#TETF~lgH;m;1BWB0 z096Jq&JRQgZ`}BV=)u4X2hPtZ8_rai!U{rf(VW6jpo6It1? zICg^|_(uU&O%5!v6{zUtt{Wgn=YVj)fQyTZ{NjDt1mpJU7tak1Gh6{PD?fP>VC&?h z%B)fP99q+ttOFo1K(39r1ZudhUAwjd6RqGhZsWhbyQ>6xEdxKlMku+!kSS(hs=E_` zW48OtDS9Yo%D(8hDxjBcq%qaUF>CtNqET!@9Ic1!&z$0Q;4JhP8kF`nb$|L?K_R>Svy;jkcN1 zPflV=t92qmLW3OD@t(zk7d1-B!X8S0fL7*f zXuDn9)kKAbS2Wpp&G&)Ri8pwj{Q`6rO2$*}WM_UC+81~-h*Sx< z%+1Z|_vdpXpPO-xRQQ~cfloxFk}6h$ulvI3{M03WGyXpE!&V|nN~uZ{KZ{t%9S@Y>t`m{2Hh)4?h4ugNVcSUa z^x8W*fL)E%kof&v-2*f0T~w4fwAQ+KMJlBsVFx=q#=i76aBc53GO(u;09d#~Pv15v zQ$?Rn0AdCdJ85B7vg|yMmHBxw&>^ku>`I&zVQ6Nacq@y2pJt$=!(e4Cv_0`3(q(%W z9sRbVq9R22*As&^Mg7@{H*c;W7;}B;1qM|+hU=%J3FsY}r3&_>?6?8TI&t;~pMGdDMt4X9;>STl(5` zMFAvC)VwP;@<8^wN``~cqsZv_SHCC7_#J!>op@^+j<%Q+xJIH-A4~q+-n_0xK#v_;!Jx~Fg0xhP>_NR*1-J?rd5m4(wDG)Gv6hyAc1k*i_h3nrHPsPtK@?*4KNTf0-Y$)ubZij%pET5P+cun2F zQ+|~g;d-4P(E>^`m~MOxY9gLD@WL=1&m%u5!#8f;{0e#nh?zx+n#{MmBIy;4hw@_U z;;G(brQE&S>6_=c-szhgfZJk13b#x0_-5lH&0QaY|vWQ8v z039zCKqjZvWCCDsL|}B>7&qC>7#jq$qHozAxEPf7YdAO=lHnBpT!tTY# z!}Ep}2NHk*u)_KI`CsWmEjMo6GFs{JsQ`y!y1Zus^kD>R=e9;OsXqcKHCv;M0UTSz zI`XG*4FG3ha(@Sts$E6wiu*;K%gDi$w6*dLzH#L@P+pldN&~?508{KMcvZctd-j`? z6H1vNpNL6F6oYq(i0^xXfB*m>#BXGN{){r}vrl$2aAnUmQUMJ0F|23UF7GM{i}4$& z2`&kEN7%LX82Kr*?<&%Z7i!lXZq#i4=;)Y(o|N~ZCfX>;cz7rhotEv@zqYU|&< zRIUdbT`W_{-JOR~VR}(L5ZT$E6yo7D=cZL8hdPw6DG3(hvsNJ?Aq)H>QW1SL zM2|$!tl-{&B1p-xpu6GUya^+Gy|8vWXhf+hp9K4Aq{8n#*Mvli13s?v6Na8^ZsdWm z!J`b5lLXLlMX-~Sjtt;GxhfAsNx8kMZJXpp(kjNZ{jU|_+~ySp8uFvPy?Jab9*_rU zOHM^rQ^MopTameaIA7DqrYt2m7z?&4;Poiy&nE^1x>J@Qetiuyu44)jUXWq0}}rBpsEgH#GxSS z8)z^cpwazkYx@MxReQES2A}$6-whjfYtn7g1|kDsUlQ}#-eax(<3rDu0}~H@mI|!x zMKJ4Wd3j?HUJSef#S=7o5K*1i%Z8sVVPj(>zFtP7BcuUpk5_U*{P%^#Ln2rweFZNB z5KEyMi%GUpSu<$c74e-BNt~U7gM`x<51JS?QZ~$|=!AsGQJ`%G_xynV>jIZX=v>>6 zA1T^(uC?dKs|dkWT0(dvY~y(lNixdb*e!O>0L;7x%R4597%|5Ck65&8e*mg!gx%@9 z_NDdN65yyzaL1m-akk~t&OYHZ(D5qvoMCcTh6{Uug#acn1lb{k zj3+E6CI&xw6UKhwXepUzLhk8Ps=If6U_;EVu9~UvFfs;1 zS2UJzwHW`;vDxjM2|2u%@m%^+^pF&Q`D_{-B)oO&R`75?a0CRdBDNK$2|fz6?G~_N zVg6=7;R1|X1@Q=8+j$IBR8;ec>L}VIXCEMzZ@}`i2BjPk@B!ht03woE@&tt70!|f@ z%K!{X83R7p!MGhR2m|*(9{`u57?vl$i{&#lR|QQ;a4HJ0@XX6B!fFxUB?2^8l3I z9B#|;EEI?JQLd7?_wV1Yz+W+JAHnV-Oo9Jl4uil!4opjSXJ_Xu4=1Z?w!w^4SK7OH z*=&Ow8bV}Hx)hZAo~;0u_!|YOY{*7%f!czW1^ArHxcAN;EWP-> zBZyuQq#zxFLBRd6=+Xc*fuMot)Y>F)b+RB4@-sCbxJ=n^M*ZX@K^FzocCbzW|F_tm zfCR!s0RR!kT_M@)yb8Vps0tlSNYsZ9dS#Z7!$5pneU%{~po5>E-YhA;INS6x>`9D_ zRQfC|FXiTSzLR_rvyqvePA{6TT{mhPZ@FscbrbRdrbpWg479YZ8B6(jfOiTxk2&d| zW)I~@1%;`s?3(Q88+98mAHqMIR@~~ zHeuD9f(np1V`n^vS=>bktf~yyheQ_4L6lyn@1dN;S^|zoH!(4(g>?z;0-l$b7i@XN z!vzeBngZQfOk5mHg)UGSA*`Ss_%@l+^*7c2`dBOC)4oqivedYxGH{TI0G6pxY8yoG zb`*Gn*D|++ZZ_1|>{q0RcAG!ai|OtqJ-yPo4{dd`)&ca*p!7rudmKaV>Zz$I1Atp2 zP*9**5`m(6|5;2U>@zu0Q4~O(NUwt;nbC{3VZgvKGvpAY*dqah@C=p}$`fg6YEI5b zP)RihI0`JJrKR;(2C`N9bqJh}vtT$vNI2etE%68Vu|lD?Q2p@=W3Jh3Ue(_sZz&0R z6Ho&lFuEa??P<{UFHZx`SZwyg0!;lJY_{C+D#!-$x$b6(>)NCt$q}M%vdt-n;TL{G zh0rD;Omj7$K>)|KemOW?}~j7k;m9*#_DuRLySCi8`{xTkw9N);3ek-R5Dj z9sXT9_nKU4VY#HF1QG~7^fF&TFLQ=_SspG-rmsZq0)fU5YWchTb5+A(1?|A>*iLyI z*z^b;{5f7NL>dv4Y!ZHZI$2rSwW$WUxg&p&hdOrML0p;zjRDbtI^%i5;gRAeK_zOq z&M}2ocr3+;s;aK2jNM@Mu*&Tf`vW7 zolcGrN)fnfl|KyP15owBjd)^jZy$JE1&tl>T_iw;wEpEoz&K{y?4`B~QXq*fgK-6$ z(f}|9G4@qx)3~CbM2|o`1_m8E$W1&v1eH0@_}xPOMG{WHtM5ar25N)o6F~b-c9v7? zMIK?CY;C>HOdcw4{nDgcO}A-YeYf*<)aYHO^ho{^G0je4IispZk0e5bp z&_N+apajpa3L}>YOpU`0IBXE&Gb^d7#UKP42%#ycs4LLAGx$LdkcY;n3|PB;cvux7 z;#~=R{r0z?gFmAx=TmlDG1u|x;kcL89(F%xuEI;oa0KC^2 z;DglO%Fpy*M1_ERT!H8Xn`ULb!>Y<4Xo2ue$_vZO*>RjEdP6)=$CN?zP=VXVd=5Q< zo&030l~Q@BJ25IbIT?9M$km!c9W}gliD2(cz1%!@L?2X6BoxS`9j;LPV?zFk*Ju(D zWwUu2F}YwUNyIR}npRgsyc-urB&4 ztaSXVU)9upI4>2_JTTFwxGjRSxBA#%{P6@Jcb;IZpk}HJ48}JBP0#gH9LsbKeaaVe z2-p!uPN#EzjOBSoxH9C|m$p7ZDz5?*TUxdoZ+v7jAA|yV30vDz0_+qtwbbn_!5#9{ z0s3@HGgrLQ>p30$yD4~&#={Z?%5u?jRs0sakl&BK*N6{9UFxALzxi0k*<^lvbkaj5 zihnWe^gH(@;_qT(`AUSY!Dom5eb#)fo$zo+ZXL3l3`>K)QHe~px8M@yOtk1ezGd^N z!%j3eqeA}Ww?qBjHm8L@Qi-n9yt^izf#HYuC_o{l|8-_&ohH?q_$5RdX_g41=+YTy zVc@w|#P9`F)j6*Sa>>?kO0WJ}6$m(pW%eU3b+A4lf=8MC4?u4lv&zDI#Oqxvs8&P7 zZSiwkC$L@SQXc!e1Z<-o4H3uvJjZc)aWOGASdAscP&RGB00bLVFH_|jmDtG19B!p6 zVQt+fWk$COiC^(o)Y6&wNL&xz$|l|so;+1^uT`0>aaz8j4uuFI6 zXp`}Rtq`jU3=E{IZw!Xoyh$wN%E`iFd)*H&6#?;6_?c#^)=m{xZg0+>dMmdpD9-$( z2?+J;PTcX$J9J+6G|zJsw9JZI-voSh>-HWuryH75l@)UZV?5N z8ch(vF)l%o#h|+V@L34aD>U@SC5HQ!p>((~@dh(9MJ^m-l1u37V--v2s(pQUyIx$8PJ<_s&jMwlaZOEIo8qshXXoUU32+$&@%e8Tf_HF( zhqDs?LaW}L1hnG#3LVg+f_-kWIILl4mK#q^p*1Xp?wy{ljViq~g8MEmF=D<|N;dq8 zq(i`Is6doH)Nca?^cxy51;GAb2a}~lAI5SV+Pk3v1P2x&K8#{MSvxX&_~zP8#1UXh z%7kjfZCQo)D3bL%)<;mK6f;xAvDAuCC_F27S+(%P#4qJ3EY4ki_;7R+O)>}bs`|r+ z8g)NsA;<=L5OX_hs~(Uhusp_L6EDu(icz42n%+2k%XFyJ zb4J15ep#pT-fdqdTh^C@=LsHa0Lb)2QnDC2|>qb;U$kaLIWgk_MtNo@3NI zcp9s5Vun6d?@AebUBdVA@npc`fbM_z_|a%{kwB_&xWu4?2#7BoH7>V;`pLlHpt296 z{4zaS!+8#~hMd&T^41%vbg=8M&e=b8Ij3G-TYF|{$pR#Kp*k^4>=>9la&RTc7LLL)xH27+=NF5ChX z_HL7fE$Q`s7RJBdCcjLAU1Uo$Ki{~}QA{IS_3B5zL_rbJnKtX#PM%8P?RSXoE|y;5 zR#Co`;*Y-jo9e6Tg2Jpy$9qPOzl3eU`){*iD2pUGT<1ahTI1MQ{rU~RAJ2*`>J6UYH6k{{rrY9G?+C0;dVcLs}yQ?foFz+i_2O%iHLF_w(#1_UV+kE&TDCLg%J=rL|i{+QFmbd zSHwXtC=QI4c>?CjZYEeQL^4vCpVJ?z5wHFFq<`j0uC4v2ljAD>&9VXPBWm@ZndS3Y zXR#kXJOa&M`nE3L0Kp7s#A;QxG?0?3GJ#eIPQCY|%S=vAP5`Ev+h4qRLB4YNc}&nY zfBVh<4qh`2t(hu@cJ-Bxk6(xKN1OMx>!#ZYjjA2}QtqggK(*bbr1db>3>!+W7*p}D z40=xAf_x1CQiNOrLa$!X2ZE0-YHWP;!ymz1!U~FI(+u21)eimT&DrWVb5#Ai6U(-x zKbE>a`V|C*z)f1&d5BRUC#?<-50L~F47Qs`Q-jKUO!~MFY$(s$RDE!(PBZmS{|?B zKv?V;;2aEZsEUw2K$0b##=YK<@O(}6`0XE%U4K`ux4I8ZPm>$nnGg%+<|?nfbQUT(nTLg3nl_1o*Q7uC)YL?OK<&I$Dk)}s0El2)x6Sj-f>Qf~%5q$L zehkVhvSII~rT-R7LRm-d)6R%B=i4+jrRysnzFMpn)`le~?THyl@O(lJ|};1?FO-efhXRW$iy} zlkodu*XV%mU`}sc4d6YgwKdxx9k2hOPOrqMQvPk{sm;by6bcF{%8aq3_G(WrM1MZ) zyM};#N2@iUlq!&YTvGZayS$V(hG6ue$!XS6Mq7^|TLq<8-}4|5z@1 zs`o?2-!z}9$hS_@LG$kwW_oeRf9?~Tk1ExeKF=(13KskF?LS>b>G;;rYa=`{>~ftN z%LpxTI#~KI)6<*UqeeID>+!Yt5`fk@WVu+Qy>ofU7_MWQ>^SnT8{Gb&pds}(1RLX8 zUwQ2foNwrA@&gin=G}kOlc>nLW*{t^d2k>rbB^9njGD`u@yptRfuUETa|mrr!;D56BgUf01Emw32=A9phiU(C=tc3BIxVjz6=oP`2i=c$4#8bltS6 z^Gd#8k?+?qEt0>7b5q9!+|{Hz{zLCzA#6K75kQkIz~kc-39z+&JKgPQ5trXa01kP8 zEi<3GG#zGd0r^0l$?#&p*z6U4T8ve!jls2v7zLWY;G&(`7d4v2uk%@!{AEOH3jSts z&HQrxUL$kkBs0AOyAPZhZZ7x`F=I}3t?nZe$Z<^*WN0D8 z{P#i24-uc7tSarVaa+9&2sqXEIzG-t*vX2y&MZYGAcl^*>qfqy?S77-%=*_->R5+<8MX{ zWbEkS(Ej>pk78!&_C!sYYjYFvE?xsMwkx|mRo05d`a~)_&EHw62?}7tEgmiF_mq${ zTX_F|-|`dP`!{9pTyALPSWl*xIqvDwC{$?!zYnmbzk8|vVzPEE-l}u>)iv}F?idB9 z*StS{7u~jxVs(lA3uM+zYidG& z=KMv0-KCH16&pEs;%P~Rz>l)(YsF~rQ3+dCyX zWa2ko(KNReHZ{<{*)JHNR+ZqUY|4RSMo0!2js<|irkSh_01qv-zv8a+%*OvoBLgpc)L(9U_JtRjBbTRXw6tj_X2HA=&xGW;qeic&?zE9TmK} z%>H|(Q}6lJKMgl?%8H$W+Y{Mi>9~lo0Vn#4oFDo^Po%@6Y*co-N( z$c}$1jB)YIPe$nF3Ei(>Xr2Yv0QpZ8+cqt_6hZn0#(f3^$@};31Mw3D|G*i%MdXkI z;9d5?>({Q$!0{hyIHv&-{n2WAWLyJ*R1E8sP1V@Hu69@-qmX|@-`Yq`%wqW_|M6?G z=)+eD{p|uwZv&bfP4rw|?w~L-F@=B?J-fEn35&)BjK1F^B}+1tHwg*FAzqb_+<0mN zuRX^0cFM!czk8dIYWX2HWufM0RpXbQFBk>WuCX}#f>>NvRaV}laVOv1-QLy%Pa{Ft zBMzLcarVkf&R}x#IYg{FfMokZqty6ycsTCEhYul47XZpY++bUKyB>-TJW1{A3IJEB-^f$e5dde>E{~3Ff(3h`mngT_5FKEBA^UY5SRX(E~`ery} zESoGW0~-Y1Jy6UGI&C*3y^#ZgAeoBuFMC9o0@?u|RKpo<_;5igqzHt&Gmzt?xk|jZEOhQLD48Te+s5t-c?)0 zjAt)lB@r#aPb>JhAuE<@{ImasT1CNqb@gsB@$}vd8K1lwTQR8{u{Ef24zV5Gp;fc) zAigSPu(PuhbD0u=&Mlr~!}w#7S`B#CCs4k4@)c_wEr5UOr0&Q;L0CmoYw;vmf495J>S`7`(RR`P$>+rveADi{IlCljms-*+1k@`gu!~6`)>& z71HwLXDG;yZ5@oB<-=gnd;yoG+x zn3zZ%9~>O~B#h#hO;?yE_+Ge-8SHlk+_P)~CiP2!fzzv4X_Akl7{c6QJzah));!M< zWm~|cTF)EKT16ClGe2giBRAbn`q8qZ@DmYd5+#$(AD1#eP_1U(rhB>qDk9&n_VVi%6v ze3!br+4Qau99|Sy6G%11ijS7_h^s%GpT}-W{IvLF-H4{!&4D@|T@BjKJoEW})5MM7F(Q||d zfVS&04V*a&)$6h#T6iS+KYpY1fF&}fPV1zf(-*?ko?0)Y{FM7Tl3X_B%P%^Zv2P)V zrn`fJRjRuL8pb>{g*iC$2ukkU=BC`)@t%!i^pQGb<1pAcUQP;vurj1In5~brx3+rB zOs^|Z=$=lZ)~aq5S&!(VP&$5$jKo66j#^ZN3S8tnu64Ktr_o7HPw(*lnqX|;pAzQvKOn96Mqf%6URRP&Dd46efHT@+&~SRm#%6j0Cr#pYxLhok z;;#=8SXau?D{@A;d1b>QBtF$LMxmI&8BpZKA#e;25_UFFm@2C(BUh$l0XXSBk$pQ^ z_eqg9Ae={ph_W#1U?YDpIPmq^NmJXkx40ks82wet=m+943c})F)d&{5L?R;oM#F_C zXe`hHfM$#O&_i8a`X8gBEJ5YQ4GrtTsuC5PB7O9=@$9q7Ldt;xi_ELRV|IJ2f3ndF zCPzJJj|42t(u+q*GBau6Z3!u6s_JyKpa|a))WXUMX;ND1{&RM~W*Fiq!cWR=;UjY1 zyHa6=>ojH2dNue%O}Si+UT+uWiZD($#lEGy->V&zRWMe--IwrTq>h|JrF;E|!}q$xsLMC^IvL2G&eNzd0;*WW8+EGEeYy!O1UStp==Rf#=vJF3J%?-DYo8o@^}E1Gt00;Y{Qbp!t`IGSG-i-*x?s z$17m=TG_so+je0tZSdxT(y79)9qd;GPg*w#ak@nO0-C00O&ZFT9~CQy0pJwHN*bSd zxv6b&wlm$NVU0FgngVEC4StD#K(O-C`uWhOQ5*RTDK3XqNM#_=F~A%lRa?abu!$FB zpT8^LSXogv9{2%RN?Z@ob1U0^f_wES^mTD%9d=YZ#t_6^9ztNoc)aof><|6Z!>tT` z(A1hbI)ahZ5NDDGaDr>U_l_EdbI`it#|oKKWAH09!n9)L-`yx2ylL=#5p@V)>GEh9 zBiP5YkiJNhOBOb46|ny;3ZPoO)cAdV^52V`p7FSEp=s5dZ$<~;w}B=s^ztUOCUB~- z8zEfx7j*Zf`ibRF!AXu%R9z zX-U5!;2`&?9^N&9Dw27VR=4ji8*z$>;ydP74rpM`MBhheI{?W8*Wx6Bx>a`n5iHkQJ zF=&ENrlrObygnlpk__dZrZCRpc(>58h(s+}1`2T>zx9r|w{9b-9;jPcziBm(_q#}} z!%g`eq6y?JG~pYxFJ62GoYD%%K!LLO4$i?#CT_yZs$M~296K8ALCw5ZS$s2osp#k#84D?^8_I{ghkb;rLbv#zK^~-IXSvX zLD5^1=`ltvMR$Uk5O=wXWSa1PJkLk{^lh(?zkeHhUYt20$tVaySGn#n0CFug9c;gA z;fq|tVG+AC9{okN=;{#CJHM?a; zicn7~H|C!*_e5zIqN!3#`sC)qT}gH3z9mwtI+n6txU*(^hP>TkW22Z3+yJSsU-bX< z%YndH?eOx@`dYzlg&_mBS7uFwKU8>dzTZb3ie`rIZEa~$k3^kcShzt+*;DJXlcidi z#EfjXZ`WCLhJ140UN5_s`XQ(Z)!@ZXg$Km8RIat+gT}Hbg>_5+Xw>QzVrsC6Y=hvkak-c`PIu63t4PGGxj;XC5;&m_nJokyPqq zDw#^ixptr5_x$erKIi`9taZL?wN~}%%iDhK*WSG<|cszN^B4mkPTqyMOF zA29}8i|Ac(A@RDSV*^Z+ccbs#JI5rTa{KdVYq^~H$qu%g#g0!7W$HD(Z(%EW(MXr5 zYAv%SM|4&6Q8^sHFs$C0%~l4K9s!HtM8= zN?+Qtd2()33L*udTG} zFHuj_!S+`Unh$7ll7k*DE&|)Oy|^?4i7!NVc{nBf{X()?pDkp)x_>{*HabVyE9)j& zeNx)N{m>0#*#d@~Z|_o8y;wY~v3k-&bGY~FT0N}%$4{S@so;Y-;ZKr;M9Bav=C$bE zza|igv;Hm7Ek{%IvQc$Y?b>zgUVvfIME6OT{jPh4XO)jBW}g+|l-sx@$Zdb*!psUS z{Av*)LNi}OurW14R(m@rh>jmFX&m0QD~g$KGYYIaWSg({SdSk29JwkZgu|J+OLe%^ya1jM6r?|(_hN;6 zfW!1^Ed;IS|1Q!u+1Dq~85G&t*k%8j;S(>-(4=*rY)@GoNPEka(dee~>$OZ{QmwV` zBZjBH&)Y-@U&NP(?ilzA_Jao(Dty<0n*%A_BY;1k&_1TFo(!q?-|fZOdi{FTlV25H{0x{_E|(Cj7*^WHg#PLJeN<0}34&+n(toZ$q5ka&u> z@Im@0|D<(-sD}Hc@l4eoYEdYe&&*qJ1@x{dkoMSst(>c+s*-A@-f5l7EG`L7SFXt8 zEwKbSAg~(_PZ=2HQu_Vrn40neQ65RD+1PMic=NW~bTjM zJvdUMe3Es<|FAtnZ|Z42e2t48@(|Jm1O$X?-VI>=I6pcE9j*NkOTh&jyPU+QGur#G zebVW_<`>(cYQ(y$DPiE!F^Xwnd2B**|0YcdSrtFsjW|&5z#TK(6z$;Ty}vRpj^FB8 zFN3L;&y=Kk>iE0)X*ja}nftBf`fTxQR7i-8t)^wIZI_mX zr3hwmFst}4#x>W1xD%XA#l0$0Q}s-OiL)f>wA&A{x5F$sU6TKFP}?;gcKe`9yJTea z{2}|}r#`AqB~|j41NAA)#CQ4T$^MAS%8$EdJ4P?T)_e~a;E_$6B#dFM=fKV5$Ip!+ zqzm{O1{Cb;k8>_mzWaY(+WwfH7Im*c<04JMWEK!2$^85n7mv25-isGk(OFU|3_ePi zrkzy2SaR`QGM${fsh7;r-b1RY^zhMa#oe^&(8uaj_N_JLn)dcPu})qx|E4Kgj3M z$KSQ4Jbw(X&tR$KY_4=_9b@NXFYaJJ)pHrq%j^m%Do@A=2E zY$RcmPFcin6sSnmM_e_82M32G%9sCQe|o}fk$sAqQ9#9@p(U`$Zuo&>;bMiy3-$2( zccxE@l;(4@b^A+MYx`*aydI5wWjCREgA;-iF9gUfhx5jjC0d-v`}z#uXVhL#a6C&k z>`l4O3}aLhy77i$1`0xVRLH-~d$>Ue-4nltW1Lc}An)qmp(*wLMnU{(>Ynj6nTNAX zA4;g4`+LX2oqm>ktgB<%$n6{RDVAsbnUdxWha>WCY%Vi;_rxtweuJp@6*^5Ra?7c* zMug6HnJ$Ri&W7`nJm+p@8nK&UA*-x5C7js*bns_r5Z{r;zn5&=gQFOs0%#0z{>H=1 z86S26osLHjYWhZkBj{RVV=j&!n+)cay0TsE@7;}X%8Q#tffHO@AI2CH5-8T;S>9yp zE6wAT_}>1_tE+}p{#Ie{r;hhuAVTGq2)C-+R6tG>3qs0$2!lo+H_WS1&#;qlN(v)=qoW?IL!5GEDou5Qx> zleSdc3*}KS*rH(GIyKXgU$A;rjrUI_l>k(&!z3PD(!?&a`&sDZE0``AzO@&{+ zy35LwwQQ1nr#Q~6$aaBShG})k6)mHNoYR4w=B&;#rN4VCbQAsm-nTj8Qggf;3PH)M zvam4g@ou)oXJ(W7To)MR?0M6Ds7hXiIsAiKsOTCwk){M$JG}n z)$_TDK~c@)%>bLnXdm?w#?Qovue2hGCMeuDmH<&SE;$w5r1^h*v?pKXX}<({)2At5gY z#>GeQEUP3tk3K5CwcIH(>C=@#xzY0cIi`v5*2A&Vu7RSJt${&rr|q0WWORGuFdJ?$ ziiG|XTq1CcoP{2_YZ;?u1STxg~ z3!hKZN!v42UAb0WeKvZ#+SWa_#d86|mVXCnCkIFA%qeK5e$T$!^o(is$b@F$l~~s` zX~KLK$5SCokoDm?xMEzP{5I_b2Wk{&51h3Zi_}9kOjrc7uG=l*rrfCipBgdc6Mg(D za{RcTtQ$^Fveu__pJdFfW?|y^!gtZFizesPlXwrOHTgU@w45iMWas4s9ebL_6kL}- z1f_H23_1O|9lq|wE!RUtz04ve#^kfI#r<1=n111;I_V-9Uaup7z)Sc|KK} zEUp!I+fEvC%4spB$a~!QUVU@7dEp{7sa*lX!}InV;v_(3T&Q!;k}BELQO6XYo@+kIrFqr^WB(x@0Y$9^~W5r<2B~Gp6+j#j}d*$Y4DGuj7{o?PGbjKjjyYy zJq~MFInNb)Gza9#{r&WJBJ_Pk12JetdPn;9ysbZKG{y2|n=ak;ta$$HSmDFy9%@sW zJfs+kk>!e>w`;Or8Z{(Y2rG`O`ES{ex(Z^drmfq~y<|u;(l@fx)$#Fxh+^>3~QZQdZT6sb$>3;{}OOEjmvqrWp_4lvU#s_H> zIWWb>Qhv~{RdY8{tgfRKj@-0!TN#(L=YHe9Y=?`q!LAHT+;-I`oVSavAwPy?of+1* zbN$j=&}h^w$zidwa;VZ-oHuNu8?JHf7@lUCscG7y(`+-M-SgaGLmdm*r$TQJ`Q8lT z7T9(AW=Ke)jrNhnc~-Xb-@|YBr<}g_H14`m?%+@6F1BmGJ3FQg*J!T!JLv2j64t&( z*RK~Xsz=r2oJDUoQS<2;dX_Bhq*vpKF-G-3idWVp`hx=PJ{TN@!^u;ZV-J7$>~+j9mw#)osO3$w<# zxD2-l2|0W8benLMZkr z;-%%gnXc5@k5XZeP1jy{Ff-Hf$p5`t#OD~-%QuXV{WdbHj15Z*8;z-1IhFRFMdA8n z|3QJK-Q>s83%qG`K`CxOhiZ`j9;@ksWYh9hE-sWyW;0(XZT``_m{wnCxHgkn^lH5I zo5q}D+Iz-WMq;P|w zEgcodE(ZJ6ryME8(o$=MRUVROktezz9>N>dE8Y27zKafjp82lq$5|vg+I@qx1=(9D z{;ouQ8*iz!YpthT@KvgurSt=Qey6>s7(0B{eKVGC6H}oh3ZXiGlXZN)%mxOO^FOzr zc=Nz_c0Dtv+4HHhy~jwy_nvf=#pnO(_ZEA_W}Pt92rJ~9&M%S@;t>B-eotqOOTR@j z`U%D+yfL!WM#Y~a%F1oMXyoN!W3ct2hrFdx`RUV{9*bA?{h3|tq>KEwA>1IF8dkrn zsAVSURqZpbo`M{!>t0j=^8xg2hkqTKe9ro@xB&~AUEH2oNs7Lu# zSI>Jjhfl?=98y+>MCDUdmOs;w`FV>t#c0XTHrw=t#a6S&U;o=UeibdTHz&JEeRcJC zw1VF4RWgcG%-8+AYW5pW>#sRy&a&IT=zl#m>Z|>LOJT$2&3ACifs5j=(9X}>WU-F) zmR-!V=ejyYjXSrA!Av@R-He(8KOgt3W$xUR$HAP#uh@VC_q5q@QF`v<|9!H|6`mDw zw&CW9W)Z3moXI@IRcJ0Rl=gF@%HX<#McYf+5;sDksQ2%|b0py=@A)aS=6wu=Jsame z&xJd<1Y9gGp>3KB%Lw?u4D0%o^mNXTj#7qB?b6BRdcCgjB6{x-3LmmfCq1QZ+1g`&-lCXx#a$Tv@Gl^wpHJl zPO1Ev4#$9&qKk6=4yD#4o5)B4F9F{r&9B7SGJZ)|LXJz8XI==g-Qa|DH69o$rCM4a) zbnZ^WfAPsO=JfhR9xcZFf+s(w-X=bUgQ+-6;!2e?&&Wg;14}Fa&YeMdd3!T-9tMSp ztqQvtQ67*h!0YtFAm-hl>yJO{MBkMV&+RGu<0fI35m4$tn4X*W-9cCG>WSdveZxod z)W-fd_W$tzkAnH%Gq48l@4fqUE5>$7>(!@g%NrY;wF_P!{hoIBF3WZ5v2NYY48Yoi zbDIwPhTHw;j-D$1ZJF*74$^>2JR7}nO{ZMXOOQ8)@Q-k-wx-UGT+y4-|kmbj6|){*Vn)Kn`&c9V&(nmmxggWgp5pAu|OF$ zDNdv4rgkF1c`USe$W%!}f+8ILgYh{vt=c@#ykOir<$$05eks9Rx#HG88Lrta9*Sd@ z(5=Z$C{D>xQ-3R$Fni@g-mH~KFfRL#j<@wYWWHF90EjMv(8lds6io(%DYyf9wc)hF zxuMs$8O4e9tJ^=$--h{+k@5Sog7DX7Em20R4I(2c^i4%}cC{O<`#UGzXD{b;C>^`y zt{`uCf@bXF)XEh)XYW>-^;-9p#x%usuZM8&y(*_T6lRqcezr<2Dg3+=${M+s`j^07 z6*?6oIj9Ie0E|T2VE1WgI7V0}q=7Tmoo%Pq!MbVF1;f>(tGIQFWa+ zpBW$IG)qpgrS6K{<;um8v zWA3OcAj3YeRJ8bdRRV;Z!+5W7B}OD8Cw~Unr9yyJ52Y6Yx1rVfl^|H5w8t$6uG2O* zUfZswd%4t7Z|1*>(ATS;(50?B9UEZk8Ustc>xs&l5RxI;yd?8c&`2iciBagiZwA(qM8ww7)s^6?~7`|+F^P@wIebPpYv;Oz? zP%gzfd1nta(&d~Tx_j&^v+mZybj9z69}J#av~3S?K89NRT%eXAe^@Zpk00{gxev~N zf4`Jjl8DbIc90+;k^~vn20l>iRxSAW(TuX!;kTsZTa#Sn$G1%xy<&F{Y4nNlMENqV z5BG|RYHYst<|5@n^XLysTjkEl$EOFEsG7Stl)bELqxAz$F0RqMjp{(OYWC8k=<|t* zB~1}O&l&!0wA0gh2wesOT>G)a}c}D ztMg1z_bwWG-h8#76pnU_XU@TepAVK!_Y0TtTggl@_o${aGTQ5l?owhHK7#=aQ#-%a zxHO(l=XG3qI%T)-$`aLy)DHiYHzygGbyls8v-&kfSw$}&ou4WsWapVvqp(h|Ht@TW zQvT(^Z_J*QhOn;X<(!D+AA+);hvv68D6ErsuHpO0{p8oApkpJ(9jeKdb+uw_3Wo)S z!dMeE*vxJ-9u*P)Rd#8Omez+@ujvQbXA*)G28H<#8TmO%T61ruW!{pp+&RwP*O#u7 zZTtLHbX;7xN~htDpSKtsK99M>Q%iQ|gW}fe-A80+Lc+If@zZg*Xz^*IY0EpQK_OTb zP}#kJT*HXTZS;|}vfC&dZeHqRVhRJlyw3Xi@S&lBv-2NuVTP=UC>=o`pTN}`QsUM^ zYBtB^3nXkx3dHWr%Rb=0#knf>+y|4vtC!RuZcy2c$I^5lx$t4y%a=<=)jkW`)X~`* z@{Lw}-N%)o-{YT(r$v=%)t2v2A$&wl$uX)}F>W_!c_~-Ug#W;sr7Lnhcif-nJ=Z9T z%r+7UNK3-NkF>vgYRH@ZQbl~9zl%y!nuC$JAB}@p#en2iqB|=MrYEgN9_8#TI2R@NmoIg?2N85XX^RyZA9!-9Cp$y-u>dm zW<~d&vlMk%$-@$#1xHqE{Ryh=GTI+rs?Hsm4`Y;S+CA!UU+t;%RP`Tn?7yzq?Z!P_ zfZ6lj<==_rU*eA47ZnIl$3UI@E8BKht6iJ8j0Q#`-JIS)<$tT!i}-CE*e`>Rd& zJ83)A7gBHq^BD?TfP~x zV9vi+t7~uXbHLm$In;gS^}Dp@2mx{R{$Hi;4%s=UR`=xWy7^rqqfx^Wr{M^l`l`)^W~{(tK$bfb?zWnQRl8}rrV|} zI+oa4AKl_)-bYW_5mJG>NdHx5>I4-Au73L1`ebrZC!D!2DqVz#ew?fX#U`3->swXsL&nl1N!3F%_tXQh{tY1&rXx5WssL}Q9iq=?$ z?(p+PvyL4Oe9t$D7fz3L70qch{D&2o{U53s($HaRBwnGOF$GWqdIBxz& zJ9o84`Uj7j;VK*WNBLw6N~gX~Wk>UwmFae+Oc^g1rpxR5I?q)544L+;ektd9vP(4Y zsq{@4OWA{0oe2xoq8O&Zl4PW!@+ak`qobkxbx^8ueTTdaYjmp07lvMs;BN4t=JW%B?ee z`rB#76m{iu;SS`1T|G=i^@j8lE9Nw|4K3Pyg%0?XmR2v?6r&Y=KBPQ?B0Jb2+zO`3q-kvKXy=ZRA8GzQCvV&7m!e)|URIXpp6%&v zF4g+MEKT#U*xd&=4|*3K8GHBci(Ov1dr@V%MriPE>-JQo_j7(CGJ^`%YIg$*6z!D4 z1bU`<_Wd5|>5{XvOAMuM&6SuqPQ7JmhGmwL9Hhfu{H?K}@zs*(0iyiaIuygQQ-%Xl zYo-hdW6Aq!axP1kh4pj(vF;RBteO6CZHIbckj>fmfwwQk4-SUL*7R|qNbPw<=4IF@}?MO zD2p^GFs0X9JS%~1dFz@yo_fk18FL8}Fhu`_GHLOW4g8cV#3EV_51z_H96rI8InzRDmH=Q@90-3D_f$)6g8c{lh0Vv zx?OH)tF>tGi$>19Tn~oq6&0*J9D=%%vyXTwMXouL33Xbc3JkQ{OG4Hkkk-q*_@{D{ z4*N+rR!(bPvm*O+y0f*ltZXzUCiHr_2M>RL@3Co-MqE*Gl2O@i^8T&2syX%#8>@^p z8KFy`4OMwplti+ z^)y97)@|CI@#(S(haHTdv{1PXS2g3EsP*0>YU2H#u2Oq#Lrtw~ zv)8O-wW;&oOhMSG*TwU=8fv63?Nf!FgK|)uI_038VqAzzhD6pb9VgFa7f;G1Uh7fj zLrgaYGEGcbzlE*v@CvXVkWh>-@AG%I`ngm{pA6U3rcle67gtYkZtL^F$7sUpl3Zh` zI*Y5QnUC+Qir%G%ZoG_|%qBHcY_I*}YTgVbS_hRTPL04DQC~yY=s2efm&aXm*c&#R ze6VunNOM16qmTPe%6U(gC*Q8f`_oY0_TpC=SeRoLC_ zXr?75)fe3f){7I3&%F@%w+_a58au9cd_J7-eNCGWjC@}Qwg{cRiHYIzn>R~>F@xoO z*`p)cj$h{Z$1!sT28Vv#!i$7cuDxnhZmD*lc8Qa|Ws$S}Y^N)axo?O4Hc^4c?kv#8bQQ5GFkX0Nq zXH9i{jxgn+qKUu~MQ6S3n4@m46-#c0&tuu*vboF_LwJkwfSy#-R0#+_ z+EhBoTyuxrP^(Q%Y!`R_1_7~WHEqowhB%eWZQU-Na(1X*_+;!|Xx4O~=IU_CU73L& zcU0258~*>##UIt@eRO(r z6~7K0lbIAfP-5F7LsOhTzeX?*K50z}kS5gb=`(E(ao_ajmZR}6lXL!A1D6^q<0ri$ zJgM^~Y@cfR1o6Ax(i1u-zBF^gy({*i*FNiCQYtLFf{izbTQfe&93a+N3H(S(fvvY3 z-+uk-2))Jc$OYwT|Jk1APZ4#+!)KD#HaursG@*`@&!55#x=P?y3~QPs*Snc~fti?a zZIMqg^A(wK;b(aYM-QZqYkPd_s@bE`yk=y5*7CQHPm?vF8SK6&b#{o`sjYzGZ*SlS z@kydTmcos)S5;@3xcNuYq+3!Q8SYKz;ox%;bdUD)D z==z|f5@o=LJzmy!%dsKw%}lTlef}<`U*2?5v8L~GzR`-eT+?4Wm6fnO%h0kehu^p2 zaot9@w*7tFS5%%-}#^>KMM^qbPc zdt1)ZNdc=mVXn;U#q6b*7jh5}+nwn-&n{aaTwNZ*)FgK2MC<#`oCBM!zHeTrpLp`a z_}_1cMaa_W_K}n3&cD2!=c&G1!}_^pnz<`wvC`Jh!y%0=uu0o^f2msY5zFuKW2*Tw zvAD)lQ>NMJC1!iYU(DB*du!^G_HLumOO)*S>tE)kXN#SlLi#8`tfhbdjo%kmIUOTy z-S?^~m|zU&aQEX?VY!*MW)IJ|UlCdQ#OBWv(2<>aOUv@4Io0F#O^^R6>^~5v=1c-X zWm&-vp*7!(KFyaAli|QrYx2ca%C>6M%$VhbPGqg(Lu= zWX8XW57=_NWd5>u!*l{8x7v56t%v@ID2CD{cd0{0$;y&ary?LBApvzpZv_{=Dhd#x z8hPX^!lH)I0jT!5N!SPAQ%u=$|I;0!#z=Lginr>$OG)(qOKVZQec8qf-tUpLWl!R1 zPi8#M_e~adb9M9r;FS=9ItNHgS0bWc_{xV2i>!keKnO>AQ&-|uxL=SP8s;7XB^ra~ zeM<`qNuWXCmHsg|CqDm4OvTUbDT2Ea6BB=dKKNMjoPU>mw`$(jty>X#<9b^@zY_dY zl05?z6!_n^-@oWE;KSfEK?YL%*C$SiL-rUn3RN$u^)Mnk8-|ABaHEeLx7*6&2;lqQ zDi;1Vy#Gr~GW31#Xe4Ig7Zx7=8m=A$H=caRXU@~GC+C%`4h$j4^wHAN;u*Vc97{OF zf35QWVJ&U`bnVfpSzSQG5TuyoyYdHWA{W4pp{KnE3LcyeJE!G{~x*+_ju$UAjF@RxLn@E5UnWK@gbnCL%VKe}%BpRC~9s^qB( zOdS7<)c`f=UyCK-ME~^!0}m}&<9Z0_tyCbl$kYQM0KEFM)`QSJM!_MZ0Sp!iI7B3j zlVP=#=qcY#g}@_0Lr&mRgY@$8ALnhL0EPTS5P2a03xNJp51mj`TG5E9%i;`)_5&sd z5wC0z248GDa8yrEkE^KoYJa{Z6F9&^Kpr9Kc6y}QY2h(Ad>27!!-Rld+3oQGFZcA4 zj)xmdPNSQo2LB?VsPyzifa-`~4obd_J=Y_u=GrCj^hhEPA=)CG<}REk?Z{#R1KpzN z4={=ZEC;1D>FmO0_!c3oXcFna7T*q$hvK1)B2{s9*fmCEo#Sq@4?>`IrQs!L&@?V z=%(dct$NLYgUb-n=mJsN)YvGJxkEnmNc9re9LCY zw!ui#BN$S#3ZNl%@!G0Av%H@v>}}h!#bY1=xgoZT^nDNSnaKsi3q`ziBreDjh&)h` zpFniAag+z0Vladu=TXayWO`->CE{sUL38$_*{=w|!kQ@ASSS@VR>78x0dc}_{UWUmLz;uw}+m~BfuwC&YY^F$wlK}pDB&@|t;dGlsf z^e1PMk-O&tGYm)j0I4hSxF`zx*C7oJR#>-^KtLG)GnEASUY+aGpLdKpOsVwTLq>@Z z67hbXA8k)2SgXo~FVXnVted{H#kFC+Eg-jqfsrvE5PE!Lo?}pG!pHLk()SA^2<{XV z6#Sg>U>UP861h%98w;CTa^K0Dngf3nJf{WmLLkRwNauniCzZ2{%P@>G1gLt6`)q62 z=RKQoI|Z{8SVJlK2>BqaGX&8BawOIz9bvV7v<)!>)dZV~5#aPg_m^a|k^l>&i`)na zS*k9V1&%apzB)HI_nygM3I>n75XhrQWR%O?y3)>b3tSE?1XfUMf&W}vgRg<1Q%`u{ zh5QYo)PlTyEhxUH4fb{kY%EbE4mlVOpo)tdm4=70qw5H8ewKYpuIXstW1S#@q1t(j z&J(@6k)|Z(=1jMJybXKCUH&G{O3^;9>|T((&AV5q#(Q$BM?G>%ix9^-g~)__*aAqF zv6oEhrCzVFD&bgsT2ZWV`;cRdxL$!dY>A>kp1NTal0&u3U*FRtAno)&0FH(tTEVdL z@-q^jG>9iXImgx%hIcRMv#9ckSCbuh#=?lT+Go=)=mvyO5fsGi8#gAAF1=S(fNOH; zvK&_Skx!qVZrXV!9-zlAxFK|)HS0pa(BCfx8g>z)G|4ZzG=!%dn;fjrN2r*gNOC7) zO%lQVG|%AU=FUX;UnkO+l0ZGp0x>fO>h&&o3McT1ofM+&BQY>=+cw zmP1pnVBD(fRc{FLmPH^S0WsdK-)q->l0;&_Apt;NmXwo+r{2K9$O`;^2ru& z2=9osF5F@*;jR4NNN*o`iG2f{`IAuWzgHL8hj?~m5)p>{_mti@APW2A>At8=!F3z9 z{@zN@f~sgOL9FDjUcCy$JKb9$qI`z%7LpO%&OW|lA{DlQ?LO=99=!(JyAfbK6*x5< z`t#Jb%E&}uw4|^Boj;I+t&$WQsMjE|HdM>rkoehnVvtHeG!YAtQK&bp+k^;rU?U%+xLd3U{g@r}rK_{3Djvqfx@=$dRLTT(et|HKUhn!rQ@@Zf**m-F|k zpPLq+;c_bl%f?k;wh0VfR<1GWp0MKQ4q+9!Ss=07;m<{(2{P>`L@O7RzsBg1Ojc~X z^)Qw~+guA%sr{A~+|m)N4EUUvy}UY5eSl`^;F1F_BS5EFycK=F^O8?QxWXU)VRZC3 zc8UiI(+t?Uo}yj}JQXdHOnxY9*r}m*MrF4fcc0AkhZ7tOpPi?}lzj<<7MzTPr3hfU zwVtAc1PiJS+^c6F%zakP_#T$@-OG-Z?DN0;5wUg{@yC$#7Fq?2%=U3JZG$A~YXhhe4T{ndw_( zjIPgtE58R(nB$Mk3%ef2leEc&Ou~(nd4LKFHtX{t~!qs4WZ$ZtI6>G}i^$=RRrbvQSz&DLU z2{c7w7qLRX0oaOF@)sV?)YOz@rY}MivFsfNBpGx?ekjdgc%_O(U=N|217%u^&64ukjxKlT&gv8A zt3Y5G6$w>vdGoJNUE03)5cUTySjv&(JWT9#koj)69S=ePl^O~?xyw`hSSvq)(kcWX z1G|_WDXxIQuyCU|Gz1F}oG;2OV#vj%zvFKO#g(N!*sbdk$B1DfwJN-V1h#&r!d<+= zeNt*}1zRqOHv(wi3ZD6Z8#f+3eoV|M3ovMUB2}TKwF(1Go(9TXD^V}(+9WcH}&cGitY^fof$PyQ0@+Ei`sXB8h$o| zz+wlGmQ(WK$`Jv|K{r{A-3_~-b{W=SPzrSY5bYzizYTFk8~+k^QQ0C?6`{0$kdsqi zf7H^-%BXM+3GvyzzlUroRAjr-?!)r5F&_J})sM~hmdPSrKtgT@XT3#Eg5VGu5)#tX z)RZrK!5(pWRW0u$s#pFPVQn#(wLDHafa4`y8pe{6wSHL%kZ$gaS0!Q|Zvp?G8y zYEHr(Ry%#VB+V&%5Q_|>25Qnm*&P7~2n%TBBqm203fq<@vM3zl9xNwxM51{?NpoQY zq8IAIbUyglwLS6XsXyJOxUGlzs)SSd(vTfdMTW*TL zZ$keBjT|pSQQ8nRC0IKx^-#a%4uAjNhA0Q|lugZbSlTdy2ruAcFaD02gqvfCFq$aq z9{2lvgU)3JfN;z@}#u8hG<2ZE$F)g-aIHKYAQO-v=u=YQ`vAKJD#70EF5=D^6{+ZxN4n z*Bq}--K*rO+=!3QU*oWNCe#y)IG!O4JSX9h0192iw!`5gVs%rbhph4LARQqP1jGK1 zXOKYk3%M^&<7F#w_D@ewBdfHyT;dy+W-5Yd!SWHg#SWU6hjCWGEK^NTiteJ+Fyv`# zYWj%NBB}E+>&dzrj>92>iO6CAr211N6X+?%0WWb5EN&91abW~2oEfU20c1hkSW!24_nF4#<^tJGJUsSu3TR#+d2l@^r}f-! z_zQ*)epWU98d=;gjU6B`qAb)3HcF{E2a+tGWlT}Wud`t|4L&Ya6l zmc^>XSc6H{ieDzyRkO3R{a1WG&76!oxSTSFB-?b8DtTfoMq&WTXCVbY{@WHmjN^QMi$G;Ul0-?qANhsDyCn7k6yGZb%f?7(_&746`?+Vk zia{_QEsycJ<5I|f+|9|1X=KP#acPuIO{CyZ{pH(U2%bCQZAX3Y$1{8{ytI;>_VS6& zpSt{I!nsoq-Cf-5c+|`+UWBUbAyC&{KC8!blastx5a|^kDmr^X{HvkMqm0dQUr~e0 zc*pv8+yp8K-vhR?oM*sQkSR>216c@Eh4iz|pg>7_B^(E!(5~mdwj!ULgq$GTG!7Gp zByW(2_;;!}2f4cDcdZ;_ICmO%gPKZ%;M7v-?Q)O_XXnwBG+6^oi`?DcuSZ||V`ej& zxvjFqF;wq*;3En`(T^D5?1i=A1)2q@8mSO7cP+7+zNWCdp3>0QLULmOXNLvE+$9TV z3otKdK-1S#qecK3{8Y>YWhBm8a#O&(;HI>9&6h$)H?$*tHFXbPEs zVZ}XH<3}xXfOLbs%RzjBj zaZJn`{PUJEzhVisf7#Oo>A9#kjnuyq^&8R@7Vw_5bbtN)<;yf?LQhJrC8i2!|J_*= z*iuRC@-y@LC{;gr+tN*|5B3*3e8K@gr5t(pq&CxQb1;_O!pCQtC5;zyP0hx=8!e0* z{$N>(q;n)u0N^f_q?r!5-wX&KQ;1}>B3Z^yt0}P+^=d;ygK)Q9`8x#4-vTHow`zVf z9$KMn?8ArW3{&ayAY;tDmf;fG@%X$2R!n?K!;|YSSXmVywCQd=V$eensfAMLh?J)J z3ir@DfhHz{>)Tc6x-h;Zd5)ADBwQVwA%CqqKok*P6{$l=EK7zHkN9In--9?0reZHK zF28h}BpTvDfcbvj-3Mo{gTMdT#_K(+Ng`9R{+;y91KN^gEE`;yBsB==&m@ZzPlfdD z&<7%(^vWNvxi7exULJ;iXI7I0Ipv*ufhsG@q$&!J|I7go)R+d9mv5-9G0AEwH=?_VRTF!S?tRy-M4C9MH6uU%{zI7k#Wy ze6VtajdlW|;Y5jeWmBR>S)4=3mWkaiQ&cxY7%B2Nt%{R8nSEG*l6%J0-JHU{Hyf^S zBid0#{jsFdkt0V$aDz-j0lG?E%;#o1PG^#Obxhp2RDCBFr$uAJ zuc6UFeYn@&}3 z@G~M=%-iT|V@t~ja$6DyLazK@qZ0f0~;4tBkHaNtU$N!-hGPI zeiPTISmJ5LLt;96Nq>Im8i!2_oUUcSQaO6#mQFRLeq7m2_H% zd{+iTWl0H1T3MtUD{b`IBzD<%zlS|7KQrwijjTso7o9ja9Sc0jkWjJSxQfZEwQM=1 z)xezj;@6IxBI%!1R8$J{D|2jAA1rB@sQ7o}QBj@$KmLt7ZKO~RC7_&FiVFWbe&m!& JjJK#jR%iD4ka%z4gp&OZC>lgOvqsx+5wUOIE;42`;) zitd>+m#EI1A#=Py0iJ=~yfg{^ch*f;Rq0Gg56cqx&3PL|EyXiu%44aHUyy^}Uwo}* z=yv7|YdGoO*;w4B)tNJ=x#}v4dR}HLO_YsnkhIfNnv3dc7YYfN&*om#L=Uj-j`)Q|CZUS z7q6Aso7y`{j9;kl{ryGKspt2>?>)cXh-#)=xBv4{PKv&n@t>z}3#;@SJXQIR<5fg< zd1l0a96zt9S}I=skK@Cw7*z28!^M?ZqRAiq$F)_$ByLi@`;Ti!P`f<5as5B8eU)j} zmiPXDUR=2S|G)A7w95ah7AG)o|LK083Tv+4O#VNXm%#8}3$B(C@?qM+Y&+$Xn3K^; zEjB>Nsxv`i`?mvsO^>c_d^_&#FEVI$-HIO=WJ)Awj zOEXQXy8Mw!*c}Fw*NP@CKj>YEJ0l&UBPMh@atE(0HTumpGvs8=Hs3+pud1?KJz8sk zS-z*Xk=yzZ+y09yV`6{V(7NXfQ^QJy#+<UU|-s@r02nNgikjqlGX57efs&CvKfeUwq8w(w}Ds6h5{o>Q#%&a|TcB;}oRb2cfs%iou zULBEXaU*&%;L`$<GmeX~oL8?dsSv;&C9L@( zvm$HSZey?Iw8@M@nk7owx;tfPA2Iz%z~@=VZ$}M{cr)MC>S*(HTqvFPPnsS~keA$H zf9#*f#il~uTl4!U@#d=?EN+@}$vzQ#mh49c#*c+FPLCJn_UB?@$3rP=IknQ@1Lo{a zt_OHJpC5fbwHf2tzSq1bdlYp`A+c@W7<)LJ#CA?H&rjBkZ_0PQsV0&wxP0?PIBQLM z8^4Lin3dDscKbH%=>}~g@imi7_(|{Sv5v9J>2k}7!?-_#nVQg*5xT;^PfDQB)W*kz zwe1t8Lug~fTMoJp0B)aF2BxVx;GH|e6#KMe_HQPeFV7AdA|ie&dDQcyf#qkfppcMlag^iE-ni2*Fi5k25xwDP^?LN3NUw~@-msDCWW_-7G;ZQvvdPz< z{D1d2v!-V>d6(eSAs*U-pBGya^1_^cXFs{payV2vEFz}-T-9OOe%6ZB7Fcv(F!$S( zLf$i?LXbDE!AswdH1C3II37Ys_g$;s%$FzHr8!W3uY-c`-7N8JLa4^~9&t}Kp9>Kn z@aSvXF?14BvB%cQ$WPQx0Ou3s{Kof8j<5N?iVyUPn#U3{C|O=BG!j=~>os8#qlb5{ zerQ$!Q^JQAPZG0KuRZ8YVE@wPZY9V;ronZS>gAVUo0gJBe06FM&DA%(@~6i}n?=%Q zXbFaEdqk>Ja4eQOmvD86AD|cuh0-|_p-C8o@uP7#wn}L!4ydjE4^u~Fv zhcuT3zcg%h@-PR|uateiOa?$lm}nqv?`~%t&vT6nYyRka`^$T^%Bd=#s@;Rrt|rxN zR)$Cz7#I7xs@<5M=7D*m&Hwn0cdzeMrDDw0KAD!I$#p4pp9^nlh9-wn7T*I^&qN2- z6eUlH*i(OFiU!8rKpc;`-Es=ECh>)ziAO8Q z<&x-6!*9J*qZprgYrd<6nEiy&ujbTpfxAqV-QV3x>VnBp@s{DxuC7G`wHbgXfUf7{ zm^c--j2pF#?LI;P6{VL}O)Mw&5Ha_1nk$Fknkm31IsZeSBQ#}(3yR5Mp(ibM-EBZi z9Otj*Eg$?}VZ`x%F-myyVw4qH;;~dbP|mm00>%my8?j3jjHf&30f+*X-)tn;O(DJS z5m>%*8l1!7r+#;g^otjax z*j*W`7jIhQT^hg4+ie`?(HSM1bdWoFZLMi<%1=E)Sj491w=8SYiyOAPMT20t-^lqq zi=z_=hJb4?Z|7&63=qD75ux9s5we7*`t$?p>Hqxl@3JCe?ze_X^KZxeA%lsf5k<6Y z*9_3fCxCd{cLkc;Y1*x9^1|7a5a>wp%Z)Ug>ttSkroF5ulOOno zUiq6+mS{F!#Z2j5j?w@g>R8}H1^i~H>>1HFyh;vF_FAy}2>(8fmYrgl=!p?~;}omM z=!;1+rZ>AsZL|SpV7w{wIqReB$x*x>u-CvG2&ufaxw}{Wc&viO9Rbl@l#Y>@jj&NN z4LmzPm(5Nr@J9)f$L1A%QY3CKy)#+;-f*Gn``smgD5r z7P!$yDSl#^Cl>(OQJ@QviQoJ5{fmL5u8{cuZhSn?g10Yi-yBwx{v zM;Gsu=?swIIo6onUl|tipcID zw}%4b3Wl%9;yeAHOO#Q0;^v$Ds5RDIHKjE~2ZzQYoRRjW0GUa1^4=XSS<`RX!105Y z=h~9FHO@$*uurZ)rvpN`6w_i7QdU0r@xU_%GX7-!Q)@hLA$ z%h=`+;UXz<_%PsykXWmmDAqd@ZDNMA*gCbBqCPuKnEfp!yPS8CW~%r=M(M7&OiA z)??LRv+4O_I3|!x*0ogg75LpCl}8V;QZ=g&!*?bgSA7LC=28&LbCIQu@>zHpT#ZsA zTyMt>S99lYo<^GjC=bXX^fOZE1rLofMH1j5v@?qBx-zol1xbOV=_|8X;U6-gFc1v} z=#nmEj))|Dtg*%u-bm2|c_Glo!->W}#?1$FgZkFC%Hdt9y0pSxk`>`yS2TJH2E@CF z6q&|QyMx6QE1&EPx~ojUq9ItkA%D<_PxS5A@Y8W>+WH|DX2RK>ee2I#6IkJc&zuB` zTdC!DJ4Ez7UwB+GTf|#m;gq245)))z5-s)CTC*w~+B)AiU`6@bWh#bJ68&^{YKS>b z6`?lS9=!V$Cspe@sqSV>Xn4z&4UhCJgTV0Af;193Z>p1VULSO3jXi5$fv>dk^ zI#Rp#t$rf{TJ2~r@@0+>=B*LnG#C^viZ3!PjchGaZ9J@gSNm=-!()iD9||8~zPGfV zV0FXCRlsX<*(@1LWotP(vpWv!m7RuoTk>s_K+SNr*(CY~%RmYAoBj%B<7TUH zGDN;M%ikv0)iKPW-uW@HG++mTsGr|Sk|w-hnJthEW&o_dl6mdNFY+Qn5A0Ay$# zpPy6*D-rwc*%0r;#3?oBik_+F>MTztn2&~#7yy{ZHLTE18M$zGROShh# zqx``yotxP7w_Yt1O`M?!@(EMjPxfM2F zs7=K^;OBTLx!NEXz#DPOu0Rvh^^$T5A>;OPVao_=f30mfT-D%^Vy6c5wT!<@r^6*e zf0b%y!=w*@WO$gL4Vc5=0tL4dC0{5*a}Q@|14qG!Tnyn|*2bhLw#@r0AM7ocwlr_9 z8Xf{9SH1R1T`NoyV)MyULK`Q+(;0&AKl@S;kh=^d;kY?Lj zC{oy#9xu~2)UChf%mM3uMV8x1#4FWbHcEs*r%v##mJIOc+B}14rTGg$$V0tEt+7gT zv8QG*c6`%N)cn>-=$0r%v@+St_}NrO-#AY9(InhpfxNu7U@Iy4oXCMKA{6P5KNP|k zc$U5go{qHi$$xAHEEkV-k;XeA@Kpr&4+F^SY>Dh!;=C`vPYQbqr(>* z{#7Ge0o_gpfBBG{e1d!9Gw~BX2O-GOYsX3MGmtsehe=2wZ%M%aP484T%f{DutaKR|%N6+^X|w0G+YjK1BtIxy{IICU`bEY1{H zUQRz>%0io!DE9pT@o*8@Drr=$)dPXE?EI=(GA#mBni1~#+{o;){=E1bda6S%#rlQz z&e;j-(*Ejh3Iho4F2LXTt*a3ppUKl(HD9R><)_3)ZmquqYP6&Z-UH3~EGg^c5IwuV zL1oBkKRuk-8diip2OshBJgPrgj8##gI7h-JVGnSWW=aS}3ff^fSrYfvZ2Oi!twJ&h zZasg3C_@qf;sD2e8iRxRb?)-er=n+ynTv82Qq-u9Zq09a4h3`voxO78O5HJ>_)I61 zZwsYx7x%W8+fqac!g?3z=}#;-oIDfV*LfyCe{e_(#{8q-8t8d`gA(74)DM~5I)QZP z4Lw32L$*#)kO<*C*Aha1wL<_Zi(FQvZFgo<7k zSbhOH666$MuDK!s^FAJp7oyAR`FZP^&EivWXMW1vR3Xd90qMXhB*?{m(x3!jW9-l; znlX&FXnMxzWSKa_rJQwW%z#;YQ88GDL78^8Q^)Hn`WysMTqW#WWIt}T8sXJ-bj2vQ z7I3--)4^v-$Bh8Oi*`g$N_;hs@&wTLM7)s4mw*DW2qlCA)97%#L#&jum7WQlda1;o z`AbWE0KR0dMCrgN1Rxsy^N2+2PXBCx)q`Hb6M+QY#UB|o?446GO>>z7h!`d>cn(LD*cOeI7^^20c-S(0cQSuq{m2+>3Q;M*p{dsT=uyQ+=-b)85Ma_ z2YWE}K)D(RnOZ&ase#Z?{cmJMYwtij@*UEBAftvm6|lq~CIW=M)1hZm$$j7-f$Tgf zzHko#1w*ooYESWaqGaWv=C29Z#foR)k=Hjn6`Fql~H_{iReCZ^eS*raS}fIq<5L7PY&Q* zy|$?QQ#xz%Xs@q+8xhV0O6KsMCA2n~3+rYK%R*>x~QVa*5f8@mq& zW$!~*jm7J1^X#Wl7Fk2Bnf8K;GFEsBq2lv3$g?&)LD}50a9sO3&R^Y^hb)sjgj*@C z|M8F{hN()(3&y5Nr-KL@40{R{Z?fVw;QBV02p{73cwB2A4#AnoH%HhU&zLLVJvt6jCSI)8%K~K2cEOkE@dEc(}YY!=$iI723~@EHH09?mvkOrT?-t( z!$&MwAxFD;Lgwk0K^gx_^1UnW_9H? zZj5oCNCBuH@Flyv5=Y}DvL-SS64GT-5I-=sr*N%zN$iu7O2kjEia zT|iyeQ6iZgfY(6~PPl1e2$0HVfaO>3o^8^)(Bchz9&8{g;o1I6~XB=;35pB?Z?m;wcc2 zg+;Ac|0YG#Ny1~P<@Pg5i5JKqs}E?NE*BuW{#bU756Bf4yss@~x?tG`G;+G+K<=|E z-~S!}+38$|S224+n(f>k_`|aR(=Ff6QnyOPGc)GDTu{NeUBMS%YHOp&)t^Js$Lkpq zVsGg&c|uo5AfNSF2O#N3AxPI^dyUDr$O7Ssc{aV~4zcew$Yg#~@)GZVp_(T5iDj!o zu#ldl;AK0}uou3DbimEpYl3oiR*ecKkdC3~ay;rM+%g~OULKejGv29C(Q_eJTLeIN z1Nfl6C(sh4ph62k-V9JPZAS%GLxx|jlO2T|wt~<-Q!rhcf zFVR2IlfeN=)FBls!&Wo#uenC--`JW`c+Ach2tB{{;=E^1kWKzTUwxV)T3d`u2}y)(dHdm zt_(ld#`${!c1%Fs!JkxSECmI|1Y(q`3~6X2?v68KeP}U#laFZoRZDU;)~~$cY%ZVCrLN!DcaFj~*nbrT?3=`VWd?D@C&)$3n>t>Bk_d7Z zDY01%v#jf0s$abSd@k?eZ+IABm{Hjyg#h7ljO^|xM(9RMWs@dE1|C1iR&;wWA?)~ZocEcg`fis@di-^DTm-VaV=Pe)`0aRSbfbH-Mk+_ zTsAMw_x~`0MBX+xKf;U*BLuekN-k@E4mdr+#@Y?!i3WHl(fUo3In|O&fCTNfFC9_x z>8{CXF&-h^=Ug9vnOQbJ_3@j(!){nAbM&|3_?vicJGChXTx%pzf+m_vP9=?7ouC!% z?v*J{o+;yR&gu*%r@Dzf&v_Ej01}`e2=c z6yNrh}HCvj6U$yUV5|2fdkEL3brLw=f;sM)gxx$26%4eKEV3v z{G@$UY|txV1v~5EBEx7{6XhJ5NyK_y>U@Uj(2uZSMJkk(l_1h?M6I0lPMdY#jh2nK z0($m*=oURYy()Or1!Hnfi==vxi($qZ4d9KOM+DzJ)D!aEUD-inSb&o4za@XMC>-N1 z!eHk0qC{LYT$m7_lPkb!3%l@NEZ)XRF+Hh42i}|sdu(#5t*xo$GvQ+qf`JPn(!KfZ+1|=iamz#6z)TK{iY>Ne~5_hy8zpj$3= z<_1jeBKO0%;paMDlV$6le^X(@bG~Z3SH8vLVxlz=%15yQcPe(~+`BBjQWnz(6;%L% zA{goX=Zvw>xKy4u!jQ#&K(*!|%K0;JJ2sH2 z)V=)4!9lPD2+j`S#%y9hB$jEKwR8Di9har!y8!>=&5i{kGm(YJYUGi~fLi1IWzwOr zqk8)`SrDn%KySL^bX%N#(}K`XGwth#rGhLaf>et_^mpaBSdfK(qz9) zvg@cTY6ufsEz#?TZaQCpOYtFqJ}$j}dq^Ts=Yg@y@5EYtSJarrfYxi%8InHR*fm`I z_}uB1z^SgcQ+2kM$IHOGw}r3#65lp75SZiTPc~!0#GVV#4F5RF zI6cTnbZVURqN`%2<}4IwN73L&y<=Qvs$t41sy6`A_3>qd!3|M_jFdj>IJLL6cckm0 z4iF`NHK@U;=KNfdD^lEj_s;7n?hn*wodho=6t9sNw_d6EtuRIP>E4&a-l>9-=G&&q zkg(v!y6JSiy>nlGv-z&{gE!Ln8WFN@SYd>By9$ z2P{4}?P3rhv57=~0g_yr=Ra!P#NwfFFR-P56FW@lpGl1}=pL;8MH5{sz!(96;Ptr- zE3LIkXgs+H!}zYhB&1=50_oA4O>YT!x!;@7Ir+?dY~&4bT$Xggo5gZd zJSBBsb-P|_nkObFw;bcZ?Fio-=Cd?r5TJW^gxN3D&c7$im;?`p`OPsW;`GJTingwU z^-|hnQ*)6xyHfPo{!e|Q%Tb$BY5x*iIyX&e580AM))TV%GzFgQMP$tg3lz5F2DyA0 zELhHwWGrf`n}s_50O%iApp1*TXU}CEn~E~i;%Aw71sA^?GkI&0vCAC2P%#j-yu7Xh zr*MI2wJLRqS&e*&U{Jvtp%sUVJ0|-W?fy zunu()P~Z^arBD=%iyx1!4A7?&RltqceP0*V*_23SyVjy&Luk z!axZ(yrHB<_00;>f_?{xZLXi5tb5BtxQ5-s15XwLNg1~N^Pgs(Nvi#(C-a33W{1~0 zGl5mN=#iazva{zHar%%wFF6II#J)K;u z*!&mNLda3ekaR*ripj8DIzjw8^`wyh(AMoov|Wo;=rVHeNFHdAdJ6#(xx;7 zCz5)oo5d-}3I})-;ohJ!nqv8I)kA@-$ljb(F3gP|4pujvEG~eaKy8@M1#(nZoX_H3 zgMDwY8fOW}X2#zpe(^0&grZ)w@qvNb4A9w4QlVg)5K zC0s!d6C@88vw8Mbx!vJ>I}ay|JG1Qrl}yK%FN;sQ4`3cV&(lL;C*x@PjLqb$tT66F z<)NfTLEDiANIf#D=JUNIS{TFV19uF}#ia-3QqJzRe6;FTcTNsQ*^gNP*5bm1d zKo+oH10bYJ@tvo6yJ^W3fEa)RyWwE_OiGi7co86!Al<7qCK*|2{-Bs)jw?g5_$pJ3 z0)%bZy<_X3zYnp?Q(KHYXjoBUqS=-J^zv$sa8|bmK^-(9x`=ApZLU7{#kCZ zt`7W2*%J-%G#zOgzfB8nE6*mqACp53Ngf!(U_$a<@p3#SYt3jDWV1ShPck)O1hCz% zOfO4{7#(fG)L!iPi(|?GEN;JrqFX4q##*agIU%gPbcQ%t!8YM9W0`rIP~R#eR8IT^ zM!=v93~#VjI_8~zR@)F$zI$6>)3)v@Pl^Cif+m>-9rh;NdRJXC+(F;QL;mSQ!+7Cyb}}v5c6Vtl z{bnBliCn}sYRqLG*0_cFnTx%8UK9wlKjT7jQ`J2{=Cbe~;z%HIBxz~{ufNplU=I20 zNq+P3^3ccE<@sgfY7IKffBD~aL&`tf~sk6K^CeF-*ZanLjt8cV50uYO<5J)C4Awjz{ zQ8c7zl$JBW_5gb_ERQ(kM{9VK)MU}-mlj)qEiz_6K-qMvuuC}MR{D(5(k}?%RdwFl z)4iTBdXKuK=h#Q5pAJ#LRuL)Mp$Eqiy8X?-6yHq*sX?Rht5{>-K+inPjM{mk;hU8^D($3mQavGS@!oi{t>-al_pvB`Ky| zu-}fdiTwnG2B{N z;-VkZV}`PvSw@^C>fdq^({m6Y-^%5qo@j<)#;{sCx_^}#)d$=52?Nk-+MtJ?Bhvb5 zdq8?9aHuryl-uS`kH=T`P+35;5Dta;xPD+Iw#e{NNLiMn?SWXh2O&crDZjLcM;4-g zFPC&pZSh_JL-T~skr*Wyhgg>Jwf5NbW(EcGm&`%ykfVp;$i}a}`m10FJ6LjN9bCma z+v1kW05jjNN(+R$=N_wt$ZCSOY38IXsNMv%Rk?pV5N$E(B$bS1TjiC?gbGC^xe$Kl z3DKJCg4rJpa~(5;EKzY=D(#xR4kh)Ry~bkq?T(FFv14brNGwCmwx%D{+#eob0x{Hg zIb6zLz@^euZM8@WuwgNh9{1adtnf2=sm`5U*)s9TEuZ;19c$)d<(oGhMy?2mtorbn zHI}7$%|ZZs9U zh$$a@{V?;wQJJH6by0q0plC>R0E>usr)(^$2B2uu+K-Q*JL9GsEJpMUQhKl=Ab4$T zWL>e2zh2H)motGY{kktgra{@f_{pQs*_NIRhpttle`z7Zz8_1# zrZG267c5+KA#5dh{0x|Ae#@36){0Vg=B3Gy`55r@x`nzP-cJUIKC?T`{3Yq3`a5qa z8%sL*-g=BvSEUBvV=A1J*V>I2=z04)Qw?FR*0fNFU2PFv!rX$d3;ZK;Ye}m&_#)wt z#}WUx@(#A~iU356U57F&NRvpV9DP{(lep}3NLX5CjX$?qbFiO2c-DOZxdwo}%PovC z2&(`Xmy?Lk3ka|_!8}0BVfCq$zjl@B(q9#b%Bu^T-LL?VJn6 zWh0_DLq4$23%#zjF^{spFuxjc2M^wzc&}G&hcWZ!GwTFXXaaO;dK*-DEZdgu6;AMj z+UQQdkCYf)&v9z2AO^9{Y!mkOU-c4Ph&NA%SJ7!+%ZXfhAIFFefC_TV#HFqmm3Ffb?8`Q)7 z6>yUj-EeC1&lg7?RBHWmcaqt}5~%p=rkRu#_ZL;=G5ee(Ct4*5hd}``hYd(L+4=EU zERiR9z7DJTTf30HP?HCMa=3QisqJ)>`P<|tNnAHTxBRe{_3wRG&gcY(eipGoZFaC@ z_aXnaJSr-$Ub}&UypC}B;~6gyV{Nz4f?9VH)0xIga$Ypqzu5?;a~2#C;%ybl|7!~4 zL)-6?hBeH-zSs4}jc-(z_`$+xbJBHlr2p3N!f%d2sbOrj!$=9;zSjTzK^2wTQ=ozy zB@K$Q>sBhdT#0^|D!YF>1#S9&pTzhOGryP}8v|0*JpWisk}c5B9~b?2`{92c{eQin zHY6#`|Ks@oXEuWW!zLg|{QumP%$;HDhs|AA-g28Y`*{}`l;3_FeUl!9;*lTjsEXCx z82|4rmb1dxb@<)bW zK*nqnluqMN!TP>7@n-0Iv~gaH8&+vKVr^WjZ45_B#?$2dK{X2IFH++MIc?bswMO~F z#HZwQsGV+AQ(=_s8H5|lOXGyA$TLKbD5J>1(U1<3DwBbv*LJMOB^|5}2C2bXtGnLa z(Y&2k+7WUO5qr0>=D$sJ7;8NV+rbajI}c>=#M<&ogl5^>DePeb1OLW7a0o!~)PK3S zw#TOENU5ySu-?^*fA$+^s^c|gSfk5{X_&wSOA?=P5|$UOb-Sr9euk1LP4wCp-X_pD z?thq}5lCI)S!A}Q)4UcpL*@Sn8Q3RgOFN#MU*FyDI`1ytcLe1oEeoH3@To!gCUjd06oh{bZ%I!sc(g)GN$)& z3Lb-PK-H~)08sQE4J!U01V)r;QR|@3Sf3mpY=XlifA$VmYq4@Qqc5pwzK01_QAGuM z<$V%|nQ5kq@qrD(@QLZ~VnQ-mS{Bs1;N;5!+fN}5oS_>ioa`ms5tf-@inq~Hn8E!3Z5{ZQtg8ZDn9Z_ z2Hjd{k8>X!l@gWPDb)F~Ah#!(v6MTau#0=!u`*UU2R2HzzmOVo>m~z*X%A>L%s3-!%H8!bM)n*JwU+zBS8@VAc&Hn{l5uXiPmJEPS-M*%KfopwS8kpV6uO{% z_-HD3!T+4RrASCX)G-8 zXuRG=amn$Dt9N-u6#3@)yWg3UXt9i$o?Fke6V*?%ra%xUVtN1WK^fdFB{`Jv@ zTIL3w$t~qWX3QkQtE#mD}S4445r>OMNqa~e2c?~W_-&b%d zaNixw$JY{O12@k zq4Atr_A5i;i-v+^&!Rrd7*KJ2h=^<#gypx|_ziwRJOL%(HYE`zmXOaaIxXM`OMAy!V?f^Uzz|--n&< z!Zw2$8ad42^KZy5+d9%OG53{!`iZI}H+rbp%Yw#Q@OF#v6FKNg`}Ky@US(#oqS?G{ z7CIG$pLzu*X`i!HoU)QR*`GT=r}p}A+FM(M`vq&%5tOhhHPSImw<4BDAIzyFP)InLPeq^spv!|$NWyeAni zv0PNSg?sVIssA#q{PFAKC7r-`V89gx=e|mD^zSx~as~V!h)*<32UBhJJkEJ%-LG_3 znWLIa?)eq(Bq~nBlX?0Ozw~EW0K>RA3a^%tM@QH(XcC|>-HiCmp(cZU^^{|MC z>|)F31k?lUl0B zyR|LRmA!J(+VVugK8Vt9oF28DGTVG{d3?e=Zg3;aaOZbjtFZ;E!8c29%`c$Vz?luJ z1Ev!hQs(vc17qQOhM{hieDZPM=lN+4?^;d8yVYdlZ||mz2RSKSE984vOT#qvu$Hn+ zORZij;;!y^S@eYj($8LPeA`J4wQiovHOnc@N=pn;3si2Hlhw8lDV|!2Ac%UcX*+#r z#oi`rxiM=hzRI*U%j(~~Pp0M8qP8K&N4)4%(gHV#CKnOB(FiI@+ynx7cfO&e1OgAb zw(t3p8b%sEC8WkX>#;@i@@!&+lI& zt@2>{nsb8r8v7Bl*}F%At(j&)8-K5F<8V-#ySVXNr1y99@t@udY{ZPAeu)WF?S!j_ ziPy~A`WG`rhLtbDWD}VkD|h&L|6~_-D&_%7W(>gc-Ng0RJ(nKIg?ZGw{|%gmu*%%J z3%fv^p;yd_8NA^XF5Cwu?D1^s>)lSh9-x`<45=YM75XPm9Akx9`V0gu!5S)>p=t5a zW<_!n?`66D=}}k<(RDqTq*DOL)qA5s#!|%Wc$T3<0~|*EBn7u$3%a871{`XQq36*r z8hFNQ-Yi2JFR1&I;_&j<$9tr_&$V|~?aR?hr4C8vaNs%X1C~_`T}gah(NybTGVzm~ zZO(MuDWb3X+1Kh|sLQ+&1TlJH{(l>Q&m`HQE5XWb9Ypy(a=edB@xb0%MoTsrn3_32cg z;M>8TM>(GN>Oh)v!{g0!dN2o!a*xIE@jb#4rWL8w% zSxcoTYN6Zn(k(Mq8~v*6(LdfZEF4w7+647g8G-siLboX1F>+j6C*dDv=4((GngJD@ zbM+2AHS1=PTB{ejZ~G~~a;NwZGj>2OX>a2q4yld$RYg_O_6Ky|8M@9LZR1R>jh8|@ zzwE{>92jzvW%?F&cAH*H^pEfP{R8qJ)@Dcb{>#WBOy$FNT~0w)v0xIQmj}m}XICnF zV@S>W&9?%6x6Er8LlmP}3?A>71`-{PLCzX#5=h+H{1EGR7lfIm^QbovFzs^1$e--* zgLBLbqv-!nW7?U3uN@za zt;Rp1rIDn-Q4YqpKS$4rt72zowa!Xhwv4e3xg7BP0!Hw@eCKsoJh&Hnmfb;Y+7+$A zH$FoJKOKCo^Fx_dHo^SJ*ty=yv3f|qeqrDhFXwm4L!(2ALzhD@8BE2XEQl-|DCw5i z5-@@pcT{73%a0`myK8Zaq#Le*-rVRvP5-$$Sh$)coU7`+RaepvZ~!$<$FVI{jIG3b zxC%tOwseeEBEKRxspj6Mc}_{@X@1_)l)wF*3eU<2o$|!`ABb@EUO&%?8siKDFOlOW z$p2_0vvm6NfLQs=(`K38?I@4c>Jij}{OQ5pTsn|(;_mlHBqNff?JLh$TYuvWZX_<0 zI#*hPLkjlzkN8|G8Lgw^m z>eG>61YKBv;fGrF(_=|<4X*$LU1pYFwVGG!w*#xi0-(q*xj1rU|P1S)jimu;foE=a1 zV^2RZQw_s!Gg95|A(6EQ?~S9*o9|X6|IHE;XUgew)=3eBnJRb-93D-OpB5@pc&)v` zTKbkWrbrN5^7kT zwH=+`%Xvjc;9F!BB~hSjmToaQzIMD#(|&-lBIn-e!R7f8F|23;lRr5wag9A`d?W9` z=c{ZVe@X^})xP3sYbEDu@JZ0-JAwP5Hkznv0+XD>CBX-g0l_E1a=~IV8aY!TzCu9^ za8PvU>1t^jMJ^9wZ95PXX`A1qO5(cqK4*cl%HFKz1xv^?YM+M!fldNw_P)zZdw036 zFjB37V5a`oxw)6)FKpt|^DyQQy5u<#FH3Q3{-dnyqaiI-P8^ zquW$jTvU!pgVdzL3LDIW5A zCm;FI`x$lM=(?`&6?La`VY3>HOH8=t+dH>}kJaF?)J6aM%4z~_SHJ3j-QS7#ca)~( zPq*bANOptWq{r|($H|Z<>Du;AYmA2uH>@&IR!^W^Q z<{zV%rQ_WNW|$&XN%?RE&Uh`~8Wv04) zejS{sKhSM2^%+7SAKo~hblLK=B1J<)=LG?~X%!BEmuB)&zdiZn zOJ^1>cw@eq?3dc)?v!X1WxDz&g;v>AbsBz40TZ37i|q{QlArg#Z9iCj6vj0ne&wy< z$M>7(%zBP|rXQry1~bm`TFNKoa`wrs@Pu*6NN<>+-fNn1MzhPywFKU!Q4wQFpX`VB zrI4bv$@6cO9Ez5TOHMvr264XU_NbzqFUZ(zrS4rS4Bl;fBJQ_Uq@IbbY*GvSR>`+| zCHg_N@>*NX_nt7DXQP};ph?&BE>xG}Yg2$OXblVHPz0rwvw3RB% zmQz(KsZN&reG||P84oZxqC#lp7@y~BcOPBSf28KrCe|;4hO07d6H^>YekHRtlR~Gm z;(5hr^4!`b8cW3zZxmyD;=^wb+YkoQ;HlI_@I^fMg5p{`nu6)2<(D-1#DL}R@P(FSP!PTh zr;d0<;t}}QNUH7si>vQ|YHHisRjCRB5h0;NKoAWjfYi{WA5knv1f_SYttJbX4D3)Rq}vE z!TV!qLx7dd-e-e_{M+pnSyY|T!Hjy7@__od(qd9^CKB29*ixQ4 zm0o(bAP4kGd-aNmb6#FLgBPz2c$O{X-Musua{P}95kxMjANqKQWtX#e=^*8Fu0+0h zKED~k*aA)bvM>6i>h@yYSiM}Rx9pEUlZh!4^;PTja+8JxBcVJ0Za0K&8|^N=6mk`2 zx083v6Ld{sZ=FW)y)qfrDHpGl?6_V@kr7~PUW`x zsZ0~jSE`OWTQE|8C?*0GJst8=6IbP`3?PI0DlD#!!dV%uxzs351~tUQ+*2XB9}0e> zyL3>cj2r4;T(CNyCNgd)mo|GbbwmtkBlhTp_cEEQ7yR~uv zCHSDZ9CS6={;?%Lp+>B*;7AkYX=HwW1BD^GY^v%vbgV@8Z)WP6YZfM``vU7U&#$a* z07~JXh^nX0c`fP;*XU#Ct-lGZA?8)4BO?dzfrOPAMkU_+7tVqoB)|{hj-mvntKN^k zh=%hmhwU!sEa75Xrr@ zDv3SX%OGE4*okEyjT{mBWomMLQzr}(CuuiU=-yUL~gG;x@C@d zR{k0Nle{dySo5{z+v_XK2QPYFR(|jLvnOH)aqoGtL$PjOphn2$^F{Wi%Xj&zls?8z zn(o^u7x2Bmx3_z5$|T(%-yhD(DwOve$tJn{^gPm7$};@^I#QkSjXUIHkM`3-j- zsIw3RFv7-Vjiv(yr+YX(qsk=fZ;$X?vr3hdj~9BSAr_pCvsMh9s^l926U;Q7WS zBOxBB4R-+NL;ITjf!f*U$J30>g^BY0aer38_B=e`AW8@2Nf;>EoyVCY5nS!rYKrX| z3zY+rVqp$es=%!?V8KOxm(HBeK>`wu>et*ppQYmP#<^coJlfu1ol&`SwX;887CDfd zCG%@D_VKKvO7{*^5FU0Xr`JvU#Q0Y8M7wkUEz#&79T8J|XS^n`1=_~cgN#T0cK^UQ zvw6K2){xHG$LNg<$P7&5`A_64M-lw8mEyFTBJugAq|a;Pc*`k6)Tft=<|mO1z^uy0 zskLsm52FCF--Hd5B|mPa`+2$8@Vk#g{7{hq>7Cj$Pws*gDG|7}-=!*3v|$VzFA9J4 zEh*$lRFEGA2JLk1i3i}aRtR7va+$YJ89NL*MqAqE?+?ih!a*n%E=OB-cDnq3Xw{tq zG2v#Kg7opDFf%JY^uil_{TeSdY_4FIjNcFpGVq^0Q$O2xHGy9MylKqZS7Z3zt0Lzf z>n7vfSeuLsHbVFF{TdHM*nD#Hr_}Z%NtGwCx~pmnefWkw1bXcQ?DXUre-wEo6|a`* zyk(sxbl1UuJEwQ54XlJy84At;SQsCd5^pOGP-Q+0T3X>J;vi^GBMDTa>)&2#@$7AP z>}AJ7SRUOB976Oavrq-#K3W(zj6$%>+V0M_NMIPW76Lk2G`XR&vR#wAJy%=%g2r6& zdf(3~UfbQJ%rPjg%Hwev=px6rK7&K0T@&^fF*DW-VgnJOPi?R zXr9paU5D;J3@_ma<-(mz{O**>F-I$g3DOroh-YZ_srp~#g`ckEn))ZS@h&`d*TA8M zkf<`GDrrhg3G*P=^OC=t-F<)t@WB zrjL5WM-FtNP_y(+do;1MA((@52Xn@w({b#_Z9@2<-3g#*ws?r~@UA^9MS(Ce4A58) zTgYvLMGyq>FLZKp2nw>D^aM=hfb*%vPN@v2tS3ke+AB{u>mjIMplAfhAaZjP#vnwX z!I!k0-8^QnndLvYez(NTlhPIM|Mjqdk1z!5@V#;Dz_;Q<0&h~9rPvs%AGC;@%MgfR z)|8#Suf(}c3x_>Pi%$sgnym|oxRqM$M&i|8PgEA!$Aj|gM6Z@?E_n2MB4^qIZ_V6wMNeVS4r_*Vl84s>2ZItq7Xu zs#8_(3k{>$^>)0qHWzy03;t4;vNGN{44|}BV3Xs6O2sj8kqzTY@KXx|b0~Oo-F9#I zt^;cvNnrEyoV#bj&f2i{8mlU(Re?NM0RVLaLHm)2m{1yKlITEN*qvXX9reWNdE}Kr zCC4i7j0l%+Zs_KW1t53lCw%JTdCKp$f_qfN>78%$Sv@&&Y(t{G`19#*Ex|fE1dcMy zh}m>sky3x@SExB5CY05kn&d9EuYpovuh|mmh!^nWM>}7qOjC(TPEjzjbFM8)u{Di? z;GLvKpGNWy!|9agFwVUfixQ390?!PUX~gWr0wW&pir|?4rRIS+hX(WdiGODUBTv}F zu5jRc{Ic+gDoGY4WyE_sJt9XqBVAu?{39ENGIt3Gyh@oS*VC_XHqmfSlu5Zn#7B3E zVJ{u2APOOf>JAdAPuN>rhJ8CUy`lCVseZdtC9*t`Im}DO16+PPMOU|Aawn7Ddj=Sz_ukTu+5mI==ozg0Dd~>FGqn9C| zH1TkM)*OudwF2pnQQz9l#wQx`HpDgFL+2+l{ZW^!gAthaAt(?RNvjy%@z`E@sh!9o z5?nwo(FBwQaD|LPz)JR=AC~kM+g$~f?%(gpBl{pIptqKuf8HC@-W9)~wK2XvCbCC- z=}D*6Cw_ht08F;a%X<(Gnhg*;NKvK;xicHwKMwMOc+anY(_(+|r2?HcbE9cb&d?Rq zGjh`p*t0@+LRzNQzxW#n_V#M!Abizn?COF*n5`|!sI|)BCGR1=vp@X-IFuoz{YY`Y z%#uKKKFNA_z2uZFJ@UJH;3q8W+u*!=vc>~$3#UlMoHHXmlUI$JPsjdSLw1wmy!jn= zXv5@fM&8cf`v~v#dx>{XLe=JlAgLD;5NfBe-><-YJ@L448lLGck1&%_VjSY5^^@Gf zU^M@1j|@L7u2Dw;+&`y+h>PC_EuUzry(n@$;c+$2v6$W)c;Vi<9iOlwZ zxd;$#pXZl>4k{3^HFZ1VsBm=%g;y@FXVe{*B_fEvm5#sK>pQnp4`W-vYI_XK8HKv-kznO+v$2s6gR}G5M0OQ2R^RRKgiK4=}`UJ$=Hi zp3`?BcChMKBanLYU9$;r!ubk9+B=5s@pnh6z4CqgUbf{0tU> z1AYW5Bs2*Ae(X^CQ^-4PB{%ziNP7R&agT_tG3iL?8Jz=HLAJO|{JFz8IhS|skLwGK zD)IiQqxJD9S2D621|X5h$dX=brUjw5x!^>3Y_S#UZg2PF#O3&~d zxt|R0g5;EfKyH%&Y{gVSyJEk+nja*hRRtBDO$h+MQ3l=x%;MdX0Ev8wSK#77L{i zDyS82qrcU?O4|8!dpy&)VRZ$^sC~#V6evED-<9^AQaM%$PAUUH@b7H7(SN`x$c;32a)@y{S46L)xoy|5 zO_kK6dfwSwMcHG#7x3uxqSQZB11soBaoXwp7>xU(sphu(;%4(8Q zy!X~wT-0v)B|a zl{JsZoquI0KxveJN37lGLTpIbKvxBu5hvO3krzwxmM-8vwcwv~!ud~yi|Gam+Tx7t zv*?SS;-pX2h@WxBMi9oYbvwgFg=hpC3N&`;J8bzw0x{?+CHq`a^@n886gxB=J3Z;# z|3P}Sj(fE7cEVULXgoHa)t`4u30Ki!r$L#_&^A^j!9w-DV|N_CR?{L^y-y4W`^zu~ zvbyJtgeUuPanVZ$d|C#IL=T-9vG)SDs*YCFg%n%Hy51s}DH|p(8{mNz#BNZCpT>$v}cCa=?#M*z7#Xj zNp8*5!=I)Ji0{uI7zG|`I1@XcR;7bnG4iPhO>PGr$+MX?z9YbZnaR?fq}51A z99>ls0apXw5y>8R%Z^bi*iPmq@T1RVT5u_WHcvSI>!y1lbqS zrw!_f9tNM?Bk?ODfE!>bD#h`|y41YPZT3(|DC+obX^}qoud_Y|=S*SHPO!jDZ7xjz zz*r-ghSr8*MWLj@VrQpnr)mdLVng~MQU0tTqud%-tNdc!V~>x6{s~Z#UVcgFko|QJ zsF&H$a^hlx{pgA9Er&ucFjHzBO)k5I1FYVhN_hvxl9kX~nG^Vrpvos|hlS4Tay|Pp z>4(+F0SVxE^_b-%$kDa&Kp6+@>ynX3%@~(E@M${vpN>aFA*x>_@QK1>!Zp}fT}kJ(T2eOqa4$tfKB(V zN6=o!7NU)ne-M3dq9rXiU6na%S|FwT%HaVQIO}@^jQyfL1EbE46a!2@|B5DR-%95c#lHfZx3fZS<$_%Wv?C*W`&EGwU8; zj?bHVa`gh>2tj^H7E$Tu<@b?Kk7AV?U6v(o`du);2!tU)J`TF(Lb~D-a&226a0q-D zx$xAHZa4Pm9VSwrC6iCr4(x5wuu)Di7RU>VOug?a^<9F6&B^ouqMi+ ziqH1dsY1EQW0Q^y5#JNJ_glTke)8h+ncmM-PJ5iH2$je7tLPJYkrm#KSzV#~&7&OK z?>mgz13#7`JSWYN&be^%>}aFrky_AyyRBv5t%pR=e@14hB{yMvqK|uDc?^_!my<;e zeYV>R1%d2eC*NvgmhGR6=@AuuMaB{!ucSbfph%rhHETkS1WD`&ZN>|Tm^aw5&d0=U z3YCtWdww9`gf{ND}xDPp-X)>#mspbepsa zwhtMzAm5=h@(4IVvjlk4q#$EF_}Vernlkkz!eg_(>OH&~=`~+Qm-C;9|IprWw<~(+xwF zv|gr+yG*t|EO*g&#)!KixoB5Y9#V0}r8t<&4T(=Brj`c`1={0 z=j=7D7+2g0k&ob=>yZK=i8~jF07Eop ztkZ_-PKgb~HhAU^zBGcLW0fnD!E(G!a$NJ7uSYC9D3t}8P--iVc*2wSq_IOKy&-{O zKm2=lLK4ox1uFs^j?qWD(mR)_NkUn-v96K%$Ova@ls#sVj5I>dgo0KK0R9IzAa%w2 z3c66;rwduQbGyXZN&t9_&(sMZ`CW?0BLn5WzMk7G|d#vv*`dgLUnk^nyFCU zpT7%rU%lQ&a9J-S_siQ_#`e5)fKH_Zjf`n1-bC}d*~B0iYM?4PCJHxNSCGP2Iy)R^ zcqi|$_c5?diW-jV30jqTjVc}z4Y0rYU}JeoTb#e}xMIU|%pL3yv^w+Z8Z$*n00Ym* zZ~WvvXA8=b4F^~yelpc(`$8{q%(DwH5+<%iC7>cYi=59M&VEFl0!;rX=VakQn` zA&g2z_3&};;V?KU65i3e+T!oLhrW`qrS>*+vHJ)E1tzX$!pqF*>U_h!>+)&@(mjDIK#qi5WtJ#j~#H0Sf~kHv%#j0z_FA0lwt5+T4)e#HYt)M2Ps zP1|{SA?MKt^|O}pYW&}r_ut3Bbm~7Tu2003JqWq+xem1}XEg?AG-2FpjjK1^&beW# zIxo>x%jN`%!do*Ek5bg$;FEDIy*N|HS70Tu%PwX5d!EDuL6~IgIM!TVm@(HCC8E5r z+4wnnQRByN-0^|=HdwzP587kPP(Kpx3={taS~t`Pn9gz1*257(EHsh1CWMQ|I;eht zt@?CCusXLHM&}%wIeUOokm?luP$+RE)?2&<=CF=iZc~{minF{0@xaL-;^Xw%DdW)Y zeEVj@{il~!(yJ&=YPzQNFgiNAW`xEEWAl{GP;fL!ywuATkp4OmP02!*wo$oI&%@C* zlX9}`$4gaF*spF@3jDgK6V_g{hyxrAaS9dic+|QR{vFf`pPA)S7Ed+Ci{Z9&D#QH4 zh9oQC$)}Wb2)pI*DeeOvvIG#81qWbzy=Lxr6Y6>CV#+&?4jRsyLit=cEHD(o!UqO@ zT=Ec&&D4?X#$p=MDuGm7Nl%hb?t1abpS_*jU4xPrve#Qr3FTm|jMR-4uZxnnyEM_R z5}wJJCFA#Zo`wv0Hwcx(4U{geI-KDtHY2cK!R%Kt-7RiPr4}~0T3@7c|2z5P*}*z! zEeGTe0F!8B=;%33G`_QWkk~-V6w2I5v>$!ZOGlYic1=^0kln*j-=7eyOpiJ{Q0MB^ z|C5LLo+^yOEX%Y@eKt~>7$P}uY4L&cpo0RVF8dxA<3kt$uA4C#vOXHSyl_uf2+MC6 zg^z)x{~q@m$gG|@C$m!}2|HX&=tBcHU*;_JwxiOQAb~%`h*^zLQ~{fc0UDJw5%#0| z>+X64YG_7eWev0a$u)HrM*QEeSAqe4T%X!|wV+V|d_yBlM$!iJcirT{2_0iBeqMx9#pQ;l!L#}l>{)%eGZE(+|A zJ(|(8C{Jzru2H3k-9$%!^mmh29#q4M7kBbL&Mn*=TlI7o+~VwAQ*jw9A1ri8FwsVK z??F!@ot$Ol7GQxEzzvkvdloy-OTE527VVWKE?$beIeOSJC7lt&zAOKtE+y%?vc)Gj zn#-PxGZ;0Ppi%`9OOtI`a4^cC>FDnsIM|7*&-2)h2xFg&(DoP014!E z76!v?f(2Sq3yv$M-aK!$=eLLOIYZXCN^Iz|ntGWk!e-dk>Hw!HWAR!Myb@kBvNMf2MvI}76Q+oG-Q@sNbAEv^8*XfdIo zHH#3&cgctAP(NEFuy(yRE4q?!hkLR1V$_J650Wl?>%RQRSguc$ITEKoz6S6O*<90L zsKZOo>r#K{NU#^%xQitJJ?9mD{FGnxt zDQE?AMu`&`U2?eTBfT)QLHNcgTSg$O3F)#1bgVGFQ|?Gztr#m*XOU4Xif*uW`TROB zF8iZ3DGy6o@rN0U3q^ZwP$LBRIcKnv@;4iTL<8#5MHTaUsogExBzBSgJ>yf9i>)N* zJFENA0kF5cj$v|%!EvK)wg_^|7SewYNAVCz>RIA}37EMF_8ekA3bwG;n2>gD-Ug4g z7(88L$!?&>1S&xZUA;SjeFG;nuZsipM6(pk z42TEJ^DR>4I)XYh*T$)SjONoiR~)V$V`sCBLFBx?r_g>eNK;~&JdTzUH>95rwhr9n=W_HbZ!%=cVT>9Kmt-zzf=zM{Sc4aPQwTM z)JIiGVR_^>4UJ3((Ad5zSeAQc$?PTA=}gD|h4$6zPvpIl=#MpGjMRu$b`mDQXFh;a zPnS{z06)1k4BlRdw0@nBF7`s~q$Jn)n?o&`O6fGy2*EA++k9Me%?;f6X8|uf)vViS zLjZZr=)esDZ9fefv;-+H10Y`0hx2cD>`s;p`eBWhW9zj5UBqp37tW`>xP#)6bPXfu zpl?3d>!O_TrN?ueNDa4^cV4cw^IBRia##bXx0pbt>TB?@@@#SH71eS%S6#zSPfabn zjUO&bsq!7R#fY;hvT@ewu}k(V=mObxaUq2DyiFPa1S9(Pk9>{N7DUqzfr>Fst12UQPRo&I@;sQ7td#@^_TevEzcqrA}AASeBiy`>9|sa6yCipN7k6n>LG{l<>9Y1~ng7xMdcAhiCn^LXPm2f-rhY4rVrV9mhd=Hu6Hlz*Iv5n~(`_iE*bFf-TFX+B3Ig@lr22p>W5u zF{v#nh!#J)B@-Q6Q_>MwCXg?$hy!?33zK%`rhQ7otbGRG=tn4}t=mGrP1G>BE`?K! z4x|QGgm(d6|A8Aw2dc+M<4?cGLpY8KyCbh$Q8;Wy{+K?&u^)`ODYX$43<6SN%JWPc z{-3sCAzEyzeC)?qZi)(PwJ2s;N=9QUfg~XbsMk;`p#$K|_?e~F>V3x>+SVH7SsZF6 zlAXpP)KcY)fHdhso3Qf7_#RwQW8$t)Us&7@3R+6~`hl&@(E$IHGl$v#NsYeiw6Pgu`dUHB?Z;$>TnDK{VLfX>`tgk=6T#V^@s22nT^_iN zu>@_C_5(Hq;V@PV70{Ozx?d1sa1*UmbX`e3F=HMq0gB? zol%9bw%8o&ZN1!TZ%h+9$2rE7@~|B8tU_{uMh2VhM|Dt68#K0uUn$BhZqyxy&F}+K6az68-y~WK{VOGt7KJTIUyBEFA!u^B)GIa zdeG1nEI52vFxehid*Z2B$`HumK|=wP0}M34?-jmj{`E${uFP1yQBLuBlc$0+RK=v) zn|hoiCLT1GX|0W@y~t(gaSR9%?ByHRb-@dxo>g)iY3{tX#PpHpZOmiXQ~JBcc{*qV z{rUkk#lwyF%~_m)UEz!)7Tg}S=aTWR5sH>#R$Pbrs-=@Oyh z@_&U(#_q;m(<*&XBg!eRedyB&#MgabUxM7SzuzI!v@9=;1H`O-t>Gfq(KKI$xVyj- zy%AvfP~x3>?J4n0+t7gB8`sMj^$g7o?88KbyT4GJwORlZn1VBMtcO(|f#R4t{#;-e z^9>YsIB=e2EJR`u65hhuVM(9u5tx6ZEFQez#WTiyxN3J=R;%t*nth zs)wpt`;0c{)t<6@E$@7!15i~PK-w@M3hKK4Dlu&HZZu1cvtg1)C1dMSc^47eOMcKA zY@Ib%VL&O46qlUq?~-GIedH=X>B2tEHS()DTOEL+?z9v=jwTnN>Jswn)GZgYeKe6Z zvOphenCiEnt##^gwlRG)4_AFyvIU0fsRG%}6y_@jmjF}|2lH{2UK2Vuf~$pwV1c|{ z1|JyrGc87Kn&{wm5^$s8DTA{24^!ju(RrNDOe&>8XhW$;pX**XiqtL1swb1&tvUU2 z<}|BSztxqUui9%e;|ZnhpIH_dtvV`M3)()iF!v(vewCG{OirXuC|8cEV@(Pnlh|Ci zziyaV9emTWvT89cqw0Z+6p1rd5W#~AQKh|RL1whI3l{(;r1IuyPVm>6-YMnYx^Yn4 z*5--MixBn@a!Xb=!si+nBU}zk26fqXLWc2P)4)s+w4el~T;rhj6D{N5*J2qcAc?~k z7<_3F>=RYLtnpaa3p1vu!~HaIb-2gZo${-`{&Hmr4lv=^S7|*QZfJcP6=Pw>3(^j zV;8avHw17eG19wTC~ z>6U%%B8a@r6={)htOfFm3;~{FEYyhLDGQotNOt-%+8!a4!gH=2j8(*F7N3_qu1F@L z&ndp^;~x{+ywn~!gjX@hkaPb2VBVFWpkxww%Hi>QCI1=0?h0w@NycCA!iLZPIZ7y* z{!sps#^(>hUjl}J1XO8tD&1RqEUaNy;i7!I&d14(`*%ODBn*52t|}5=SIa}e`m)+J zV7ii*bhc`d>J#O+w@r3A!z;+lZsSnHe4Es+K)*N)odpV$Crj)1f@Z?C=gJV`_- zs$sV;z{CqL1kXfTV$Kx}D<}8Ns=>b6Akf~)_(b)Okqb#=(WfIIm?Kqy70I|S4itIy zMtJeHws09SI^oGMo6TYbU@>-o9qI?7%aO_*3GYSn^ zPyuj}K>FxBoaxxH+w1E*f5|J$ExW*B)o%6}r42MIPD;G9BB1`>wYw<5h8@ooV6zbd zEecz5#M?|GJ1eyREi@l)OU1}-`4P(d2K@RkHOp!E?ny}|`EcBy2oz}+A<2=C!vx7% z5Z#BhuryMARe%=r0x>Rdgz2Ij4@?FN z;fjMUpDXz5@sWeV{fU^xoSGYkp;^MgYqv;zXMhSFlB07oerzYhstQ zU|%_;Gq_(X9ybN(%7%|bB|@x7&`dOla8B#A zCKdi_IWgSpiIXGqM6*x@-|669R5s=w3GEaet|&l6?sQI{AGgY zZ1eC}sgUEOEyT6FDPzfliYACUil1P3b#LyMVT4?v!r}BB%krqHquqw;M8<5%Q)4hL zp~o7XvX<^qq4%4bi#;>!f3O;*k(v>B<+tRKk$am@=Rw}mOvqn_e~oVYuM_q0vZ}$? zOGKBxBg+lO8l^whAgtDILxu}zrUk55?`|FCTFfcGTeiSZJvN?hmh>!1{nRcb>4MvQ zLLSX~xYNXbg;@rpge+dEs9{#g{BGuCZ<$fR6AcqR7hmwN@wI%LN1E=fr=2-{2O7QW zm+FMJeNtM|SR{mhNV_l=78Xd=alSX|W6Wh;Hp$K$W!q6ga))d2lfw!UNG9z1N@EjB z%nOxKIyyyQOK&+#Ug2JY*<>vGH2Me(c?k`Iu4UtxEpDmafn_I}-FDhV1UC|12GD|~ zsaC-1XoPiXNZ5nnHBvxdn-OAbjbJM{PN#Bf1Y+d5Mwq4n;nmgB#zf%Ff@$0vOQhW0m6u~+$U^=E>N7{W;ROtx0u5;PdG;Pk z$|xIsIu}-R{9au9U}f+va7=S z+j|aE);!(|O`yYW4Q-^haQn+;$PDeu)Hh3XOLf|GlI68rFbfIFGoF`43^g3t^)*bX z&fUEiZB2t6yN`?=-{fRBCVo*MmY7kmyZR4Yu?#<4cXP{d#L58 z!kY?JBtiL9-#^e~%9Zs?$ zm}$CEV^pVx58mUw@a7IvySp=V+N z>}~0*H{DONKBGD}%=41lZWlT1DQm2TM5l^26uPHGqK`k}t|7*n{8KqHecFL|p$&{! z27OXW%;CPb8T$Fe8;GN_0iv|mmc5<{0md%!ojsTI)p?*!yN0cL{`%9kam0RG!na+( zF4l6CrpDS45$WX&0-+lVoS&selSd$C{m+i&x*eso_im+mF#~rBKwK!Yi4J#hH|$ZW zjFQ|ADL-Eemo841IyR*Zq?<3z{^)0)e82-?`})i$@`oZWjl7AjpK6}=!8)iqB* z{Xn+g2_N;GnFlAx-x@JeB%|75*>8*!d~&|R5in}aX#Z06H*>$6@touQK;$MnU$vJ|^-sV? z{Fsh|3AhhHw~dbnh}WSnz5%uY=om>nrsFvU}#g7nN0F20!#YEl7OQ7HEKZ zU~C@J1mI(~j0p7>URDl9uo=JqiH%FGIeN~< z4z>P`0aQ@3wcW9xZILY%bKIxm_b&RN&!Gf1X7kZmg}a<8UNogNZ(j_beSZ*HpM@lAaKiKtrELy|_f3XqXU|04tdPcR%_^PJPg-hVa3 zl7deZ2Ms(r@E=}D147@wPyYGeM+hDw@a+D5;}ps)cK)S+Rg~c}I`cgGnl3w&O79za zUcVW(|Ldlb{oQ%Z;eEe<)x&O^D=Fg}FX}je7itD}JHXgY_u77TT_C0JWy15=Ol?xM zT6CsvMkS)ciLx#PF5(+_MgRBflTqEp6dvo0jzo)9S4y)%<+i&UJhdhXx}a_l@r_6o zV3Y-+C%h;Nu4L_990u4A(O}vT4y48|&vk!vEABE3P>yJ_XemLlRcgT`wVDTJ;9We4ZED3w z-!%Z5Ojt&HbLod$vnP>}VCM&yg5$gah9=|#^*6|;<2{#wYDy%7i8{ovj5=C4DVP4&6Nc7 z`on)YgUGyq@++Ko=CX#rDz`_v)ctG3a8tgayh-oYjq%9hRH;x9rTUs3A31iT#H?L+ zHw_8=aImN-#I}UTPyQ+ws#BFt7Q3vx2G|@DC@+CK zt-Q3}|IR7sQ|mVJ6PTvpB*skV@VI^)pX2k)mznUsd_4OtkdvvP8>0k;%E*P%6TE?h zumLi|?a@W{U=TFma9a~WvMY|b#L*H@E58LYJjnn_roOuLOmKQl%qj!1XXmP4eq|ED zG#I)y=M5y^R45iCLDuahsYfL8>@O!ELrn4>qt-*pXkd#7a03$pGc&Ws&jm)s7eyfU$tE>r%&aI7OPji-4B*e?UIiuOf zHRb}K%ps`v-~gjAdq7RNyu*BKvjvQ8K(3V_WDCm`H;Ei!U$vggt?8`<3hLf_}u<<+CFA^s`uEb4TKwt+^B>y z=Ro>hd7|cK@}xQ~XupeN1_7CaE^`H7jQ7?}kzZ_J(t8BchAPu%U(n#AL$}v;Ki3{{ zka^ED{~E%SAqgkn%Jui`TK|7x%Dhk&0H+37kc7%?&#@Hv=u&_X9k3o`tjDTo`i^pF z1-^}x9x{I%ExT3K)%htF?Qi0bs0^Ce;a7$ucMi4_J6SgL1-4sVw9@-o|C|u7zTdSj zt9RUn3qWRJNf2+SVxdOEi~HDrIE|=OI9Z3lx2NhV0d>Gdje@xwT-{uXX$M7%6w%F} zpyV>IE(7?49FGTJCom(ubC7vQ)u1`b+y(>z0yme0OYdWl?Nkbxj?tx=|ufwoHdf}uZx&g>bfAZZx@P= z4cKJUPXaHPpU~Jh{}GEmg_t$~Tgxr44h~SW`ZwEDhO>WiGu&vl2QOm#(Me|hDBjOepUqq6zoP zyUNxvVTF@ld`DiP-r-<^n$p@@TnAK)8Knqawi&q@j{d;D++5 zA16QlaIttHWbs@3I?4Nc0%#nAQBG>-JR|9Hn=IibsjiW8Nl*CR`T!`*Zq`KLs=%!6 z$=^!mFSJ+(Dh>xv?G6gACwd>)`kE^x{(LvT=)R4XhC#!j?sTP6pYrD^&B;0&sUH2o zn$wn$CjPy{6I%za5+L*YbD`PM^Ik|D)$P6YiOF zKSc}P4Z7T3(le&t*DDW6GcpFG-}WD=8TIo!f`62c51qdhMkse3HYHFG6#k>gJ9Dn= zMVjI<>JDK~Xsbo1M)T<4qq{d%8z0+_hsp?jiQfQ4XPm{R*Q!nl2K!C2*Ul>z2xoat zR}uu%HT~G%RYguIulTqrq+O95er6mNew1y$d2)@yY^|OMgpjN|9TC0jr+DQi4`zn@ z`G068iBXf<=C5)iEAH`Y991Yo1?bT1bF#P??1hIq)TXk_K=86}q>fFG9VNz|*HDR- zVaDgeod!<;!5sau(8}N7%dQx$9>VpHO<+;tCCOHg=jrwGDZe9X)DEf0RUz0<*q?Zy z&8&hO5Bym%#PcC|;+y&tE-nwMQDEtlA6ds!oC9#$34(aGsY<>*e`dGG3OITh>4RyW@EpagD zD;8t#8s|Ow;F`_Dj~(OWaLlr(&SN>zyg;+L2g-yp%AfNZ?hS}ajTDaQ4_eya(NFm% z9=p`cKXcKQhW_YjYq)B%DHl0A@fIFDAbP?6q-rf(M$@1JAZad#HQ81*0|7*)(-EP^ z@avEnd^!TMNWVYv*qZKUGDI>eO)vv9@i8+=pjEn1$}^#{A^ix1RIN@7fX56O(W zoTKof;meVCSwVsmO^8SE2-jElPdGIy`dt5R4+T{&gip_e>=Zf|)Q+1gY5K`XuJ1<< zRjmQuuaa!^v(uT?bp)CSXBG12d%)*T!z%)2?#d-{eaFWTgkuo+6?}&{IdyGy?32lz zmffbF{0}491N|R>!EoF1$TODr+ADl3?-ZoX+AcDiVyoANf|i>Yi?8`zp)r8Cp7X(0 zU#Fj8s=u1wtkA1jCv4KLe=mA$yZ~;@jKm76C%j;L@z+Zv=@Acl;nD2!?(D~__4;#7 z9m&K;iQ*fi;RKecoiYxpEVxDCXFDRgrP!T1BFwafdhe;xYFHlD;Ok{bpBsI>lh4|U z`8qRi>hpZb}@SER2c9 z;I7Sg9r?>I{;^4@@i{40@fB{?ZwNtAn@jD*a)h-{zb#W%+Y=h0eQ@L6A1Gcv7zCXY zetw?6Q+^P6#}Cj;6L(gjU_7`Wr`mW*@M5QQQ6kq&`H+0$D05Pc1cL`9SKZZx6%R5C zMfUSTWbv0fSiw~JrE^2l3`^PXy$Mq%!p_eCDTshSvpCv&PvZ(`_@Lrl^) zNCG(Lkg_@#JUv10+#;LEY1 zc=i*OMe)A)9it)EZ&_eWMPMuh)OJKDbS@I!xYTfr%}iJt(Q%Z*UQ1lDyr5(GC^}tS zdGG|A2K@Rgm*R0o5?=uC*P_Slj0~fe7aeNszg$${ZB64&fExkDAm!ahLr}S-9=-oC zPQ-g_4z@PH-Kzsq7PpYXI8oY@SR^m%WYSn~#BOi_K^$gOCXKc*rADwG#2>&=eA7Xd z7fExH3+h$mTnScaq;TDvL;SoLONqW%XKMM2%R6G{7X`Mer=%ySM-UUO68+b%wVqA9 z?*RShQRkc5c@N8oJZUtOrvlXTmB&X9av5Lqy}o{77j~6vjw|$@Z#oA}M@g*eU)ka- z-RX>tj8i(UNLe6PyD9HnXS@i7fz79SKv~$Ho6wqmA|OX0s(rarSE^)(tdcorjMgQt>W zO|kQ`zJ7nXUrZYzgN(njI<65H`+6I}_*MCJta|DrzZ4Nhdz#4{A^7F*du5HFI#BXT zBx}=zkizFA9109Jyyxh&WIuAY`~;;$Z|muK@2?FPQ=G^f?IkUjWyW(YyNb2nTiT3w za+Lc^-&i^?h`%RT7m6q~=L_mgcDQE}Bq0|y`TJ3E`UPg^;@CI-|73}I&YFeBr7yj3 zi%Txm-FoV#3U}(i0?SZ;>L#uIOAmxNf?T%yqw3uj&};R4c;O-0&c}2(C-k{ggOWvp z)L?Cv$B&0RxgB4I+jNzqR%$_8Y_f=YLbe+Q9Q6td;ZIB<*>3g!&~)DMRR3?>&xj+( zID}(!&atvb_MYctt7KNtRgtI?v#ljAvNUE|6r+ z*>fTaWH*FxRSXJ=H)yHMjCax+<*kID&ra`m9ZcL9)(f*w(JbRy*8|ELlTV2g zr6c>pUHC8Hn*iF5r};q#(6Lm-&NTgUkz08Q*EN1NfnZmw1Mg0V2G6^1DmTneJ_Nce zf=FR|Dec>wAUn}TNE*%I7DmJh18pxIRF-Z zQnNbFB`E_F#}&xreEVv@0uU*O@`k&2Xvl<8m>?;DAS3n|xavHQax_3r?wFw3KtpJM zkg$g_+5mc|6#yO#sQlnKe}2^~4hCbG`^xX!u$gU?r{iGVdSiT(V z6a$EZkP;#3{;Y6PilnzGfk=T0`g+}@#B)zV8<-omp2ucQ1{XZ zqAlq{m`|?L!I#!TzGrPO(p!B*jQCJEii~E}*cYPJ29VS`viZhUt2X|IK!7D?%eHyy_ zIi)hni>U@TUJi$pAq=6ySug!Xca>l=KLmf=M(1ON-Epb!ZIowkN@BIiA4O|hq2vMx z0%*Sm61%;pmscG7leTezX+Q!vtfe>%Tu)tznv=jP;Com;$4jfpIyHX<(Kv?=8yCRC ziClsLi-^d6^kGi`;X|+M{X(mbi$qzSC>-AYDp}Qk)reek{R9AHIkUwTIyQ+P-qcf$ z)OdTp^zN8QcPgrEe@8ZPoP;K7X!YweB&NP>rSK$|T} zVm$SF{eU2sM~z}EZNAE>%hLYl7;hMM!lS;II5pvu(J9YbsU|~t>L&B{D(v?^&fSk@ zI{@DG>jLyf;=L6K8`((c8q3;KHzPMIW$KC$N!c}h;zWt8wqod>&6u@kip(f_?D>`0 z&L+sIIKJfT9k_v@iwXOxhe0wzgMb~9BTX&IL?Pnn3Ur!mq6e_%R0z`vS zznm0(={pnM0DTq#LKy<9p87Nx0$)}0w;b*~IgP#C{QK>HP4NHU=*;%{^8d(wF*P>y z__gh?7Ek}n`ub*`-&!waq~v$E7b|}E#ms0RB}|~xl^*t)YMD=R_oD%v{|FAxWdzGICBL>wW8EEDLI&>#T<0VA{=}t)ktb(_- ztae6qXC*e5you>Cv=vZN8hVy>-gI1cI3vr6h?>n)u&#NqmzKdD@y96Y9te^f%|%X3 zsMbHJH@+1`5xkt!BthCKC{dNDj8{8jkXz{ySn1I+p1jN3<8X||!?t2&e88(OJR#gY zb=;a_NgNSlRkUWdkG#``9XSQRvOnPs;~3Ra4hTVFmN_d>xv z6ZUhK_@gjS`bM*r`81x@!Er4-5_h>t3gn64$5R5HX?}Oju}`E1SAt#snW@^y{oE0< z216hu1vD2|>Bsz7zs!R(2k;!L zTs_C_P)}eRsW@=f`V*d$`KI<3X*Rk5o*Xy<QI2w&48d)Ofcra7SHG91tur7r`-upg3@qj+b`zJq`K-;=JwX%H7Ys~5uBhSoQ!nlH>G~-cH(cf?Uk_m1*W-I=-15d%kUSjl(oW{G10CvCA>f=h2ED_p50A! znBv}A#}lBXHiZat>nmBehz;JydYB~GNg^rl*|z(MkN7Enl4RaddA2VAyl+a|uX(>` zr*?_(729z@1P?1Dbns|C%xx9x>uv`S=E2`@xiGpet*K|P`)@rbQI_MQNdAeb?x_PN z8udWe-f|UJ3m-bSr@QI45_;^hOW6>*fD}_`$r!!LF}tzzU`@o(6T58cbc`E@4nz$; zN6~pwb?{e_K%&`&?0Ny@a_Pz${+`}nFj7%&c*q_Rato)D`Q3##yJ=?|-L4yo$qVfB zu!~oA$#68uO?gevk5nRkuvbUX;6Pv{&=Wfc#D+%>Ms4F{=YxqUV>dNf7eS;9&ibsG zVlIJBDZK}pFVjrW7G0_smG#zh!&wX^T_991%Wf}G1O7wUc|^pFD@#GZ@g4)J+IoX0 z=>|wL_$ntr@=S_PEdB#LU0JGBG_{FviKB{n=5#Dd6>M{aWm!oQKrU8ZNY)pgcK+oA zoHGJ*C}n2(6UsLU+C53Ey(cS=XJTTt>3Ev>$+vtI0?QPJEw_a`oyl_!PvX zPv{r&vVOC_*CXOr%aY;)l)A}c0u5|=hBIqrr1R_4F-sh{xW7E(^&B@=E^C863ty?a ze)){4B7uMtvXa&m8Pg+jaV+1s@BqLk!3RY_wgjy&y}C6lE_DA*#>qR50l|ds4`W}%+mx=jcwCpB zhu5M!C4~c)HfL46&KzAH>z524Mj?`&T$s0q6r0ZKD4x@RC8T}|h8=1Yvn?d&%))?0 z%o4k*pL#vP6<~>;aRE^*HRzr0*h$Y@T>V@EwF6wn!zyx%0Dz@wIcxEc2b3*H_N}6C zCr#I%wb)*MtphO>4@%PZBO8qjGvL7S?U;lDCx%neD3L-|sM{kWT zj54i%8}pP;izq4|mb9qwcDzrzPo9!LN;9@D5&IQ+V<_1bEn_F%DNk;g(v*+wdrj_U z5Nx>QZT?pM$-HXkU!irw6E}*td+c+^y0xVlE+r0QElC1ZH0d60!fRR$rldZ6pHmK# zOLQ;Cq?>zuI9mZ9%20srUVB@^lkPif^lmCxb~gAF{0t{M5>ib~jJfPPQ9#REWXAO6 zF0(|?+GdBFU8Zzdu~Opcro<^M-9J$8>Dfwjdz-Sutc+8ye|z+J1077n z?U;|eDCRaylY)K0L2R?tOVv{RTcv=0{8fDGT<U>vJ^q)AhI^|km7o}ia91Mm{EJpx+#P96CO`phmTS& zm>H25V}I#FHs`32RZdE_T04(MS4aEYq_f1xf|x*XDz8Wo5@ex_buiEqh~H&cB&TqM z(!k8^ITDV)UV=yoIqD)#-F#o!pR>MK>v0H1+6+#0x!sV=PE=dGBQ0|>3zl1uG0_Ny zUEorUvKlh^2`cf7iU#qZ;nW?i5<%{G^1Zb8J*k|MBpv)o#UNP`zZ4!#I|*zhm>%&O zSjY(Ks#U3!AZ-PVs-0_Ke*hdr>V-)VhE{9qs$b*kRN?ysEsKFIE(?9N;lB zPPoJcHW-%=J@*gI7FsBFM+z&o5Z*wRGS~aWPx;*Ntg|;QPXskup6XzO`KUoLiFrKE z;+ehVcH|xUt;m<`!Ia8YGE3aJTFAWIMtqG zvQw{|`&SEz4~gTA$ElM&>UJTXp1XDoXKc2>R-KvCuh(DC$6EBi38IBQ)3I}0YV zOu-gtmiM+>Je`Cml+P7^d9*;>bU1ysxv=|sb78Y_lfo)k!Y=)Oay@0|r)L)>P0G}h zJ{)RC8=JaP!k?ON=kj@7Kbtpwy52b{dE~m#@zuoVfGA=+qWTYrH}vjY)9NUYPDT*` zu?=Bm5>BDWJxi8pRmU4K_}n_Y;zFfZMe}E5SUDv?FzrGtf2!-r_6ROc4Pwuv_07`F z0Wg83<*&+bOE^G8Ne#)jZd_0r#_y_GX-Tgsxq+cK9SqQ_@v_mGTom@uY6pY(+zwT! ze7RwnhD@iV8^Fd+>6aH{m4_f7N^RfiegYLQ8Wq;-K9Td0wK1VP{ z^c4`>>+xBODNgpQ;LQ18LX2;Vi@bAk_s{b4I$o9VRU1NQWcX_Sy)^ba#0s|{TUOc( z=Ka?)51qEKGG@i^+o8Ekh9**%cKdnBDat0g?wV9&Qsy3eyUTfJ{4o-|LUU(W*=Lx zO*fDnI%!IJmK^GYWb^N77s}Jl6=w)uZ+E64!Xq8mweLJ2otL3ZKA$rak7#J5lm4CGEjjyjtkDVI`LGY=>(@ z*Emc`B%?#M!yP5kc4D5k51Ny!HGYh3(e}5`7@p_Myb*^0$4hLKr?E~*lVw~;oGSo~ zM_+FT_h0<0>f`gpApOn6wVpt5M%Mf2opuqF_O-~Kry#fxcz_;^Np6rC9^JJ5Ylk&ROD=gzW}y7?(h z86BH)u4?02)5WjB$;$5^GDITjqOy$%tiuU-<}YwDUbNXI;UF!j75+bs(43JSeWrYF zXh(FWGQUQi)XPW|Pzo`i9}>KKsCTcM9_)nN(>^h4#8~65Yw1r32NZmLRZk4Tv6F`?!kzQ@#Rl%5Lz_Q z7tcF;peLa*XAZwVloBTRI-o&%4;%XcjSkD74C6-f5(zB=)Z$|*CUv>l77bp zcHnX8vOi0YN;;kaeFyN5OyD!?spYmxt4Wpjirt9H2L|O6l@KilMAQD*=KO8QlU&`J zU?6Z%!Y%1?XMyki&k=j1{O@AS-CD2u45Oew`5^ng?_q^-|WiwP|glx{<)KvVb%_fIinmwE5m^vNpP-xibm8G5xzqj zw=b3=dLSYgK@&=SO6l-Elz9-{D*41*G84!#)-62s;<2dne2w_j<-FwYvXOncX^LQ% zb=4a5sF|E_$Y_^m-MV2&3+MAk*8xt02_v<*+^;I7PA4>mvRLL?oiTGXir6ughS>Tt>-iU za&g3{o7*Sl%AFcbYlBc|AqgWwo{*H<*Q%bDLfli_1cN}f-rIxpI_%)A8 zyHSy`iqY_~b0W@ak@p?;_KH!V;DZfr!=(MX&q||n61pAqsfRkBDnzD}66;iN9&q{g^w3qW~7?>YJpplNXH|pG|+P#jUwv;RnMv_=@5Gb)iRpHeMb@{gZol zccY;0)f4OR7mZy9B_;Z9jN>$+!H@qoC@sQi`_aDq4;!dY+4uCU1s=|f@dCpi#L`od zZJ@z<5%F)`u<%T>Kld`GHECbK=5B+Q>iWO6|JPIpdPiqWzt-||z&Z#@4 zmtr?znCmmWzDV>bGa z$DHNV4sfbmTr!C~4-^(%S0m~Ob6*UH2J(jT`Xd&K+rf7Tkp6g$yG)?e;^c7HJkPw( zC8F%95jn56kT!fz(-IEe+o4tt&M@i+@6wiMCDKK%wqn&{>fk$1V~^v@GQhHTSs~&p z$e+ta{&;T&?Yk;wN8_SCrhyzIsfl~YfWbMC6G5vPP9d<6;DD~XbXEl{QACjHoI&uE)?>9}_mMwdA zoOdRtSmVgsNIldJo3~n0TG8X~=nR+%Kmd3=Fgix!u_mF(vp%uN9Rk+Mq zJAxRWf3Se%Prei}-1YxHC*CIRXxDgQ+W0V2Wz_XZVEd1g0L!ifoM~e_hY*(uleV!j-5OgWLv)ns?5NjvMJjfws#B1rOH!5 zA35x(TVbd@$)1b2vo3RA+!)H1;`s-ak{V-J5|ClpHyt)UDKqu0y)~*a+9UYmk$p#7 zxE?z~RLHibQ8T1GTLQvpNIlcQ+MrBK@nM?coq}1pN)1-FX2ozRaH;q>yfkM!XR7CM1VyQS+lVrDGIPx0uQw-k)$RB)o_2VA;j>RgVj;YiCzpl>IBa zZmJqzqS;Kk0-zbmP}Bi;STn!h>p}TdT$qlGi?d0FbO+NcDNnJj^Or4{0g)1*>w#Dw zYhc>bU;CeUE#D|>b7S?xT47@1rC|uJ9kEu|XvAt3IOIvK^6~z(N3cROIx(S%N)gdepv#9nn%^hDacbyo>cjukhJRElO;>qC52ko zT73fN21GtBl#rj&)JyI7aIg0Vhw`{#9Y` z***6n!v%JpE5pxCSI0RkIDJMhn`Y?=f_hVO2O2Y`BYnRS5wJyDsT-aY8fkyCX;1I; zWB2Ayk(#ctG7?XqPNo~tb^L1aw=^>9UKqH`y}HN>tbIVb#U`27dTk5)BxOH?Zc6)c zB38?5=M!LR`+g>U6G{a2`i5 zFz%GDQQo$M2P^lV8p?t!x`~~1E~WMLej-DMLyO!+RSxI0mc3ldlFSX>zwZZ&tz350 z(=O?h(7h?~bUtpcCl|-9i?6}ggkP7$ICt6uRSQ`PDj>_#7VE>&LKcIEPtDu^70IXh z_7^x*i9A{gApSQoN+ZwDzWpnvT>5n0KGw%K$5mrL`!zm zmTJG2)+)uys1pBYjUN}c`luwS*PF;xRDM2cIg0NUICC@s1X8WMKMbl_y6teOVS?`S ztQ?=IJd@5H?jg_5jX1C>K4!Df3I{^Mi?1grxfYNjb15MyQ0sZ7p1)Lo))OYUqItRl zyK{+o`hvw_l?RfYji}*YW7~G7$#k|D`af__!0Z4oEha>y-mGGrO_+D zwDLKFk%?uNIOPWU63Pkmtlq_6ky-E=akhzThb;ui;jcQYH<$f3m35n1j`LmRXb@d= zQ3`os;Kpc^{&tGFg<&h!b1hCt6id0vJJdk3ksmnnm)hrL-HXrP)W&L?xg<-Lr<~ch zDxZf@x(ke?n1Tp;=UBg}xu{Ua(;MA3`>x|mD;D2VbWRJe=5Bu%Mcz8lhh5N&h$;Wu z_2`Nyh|F|lV!f9C{_|$$rb|~%jZT8sr}f&d8{(DqtGHD37mWLu1j%5o`!k~Unx)P zpdjT(qd+Voe(J@9PJ!LaE92JZQ{LfgaTQ= z-3EPmHlJd3<&{|Xe-coZ*J_9O^e*64*bPszcd6xf$dOIFhIT_vS6m9#U#X3~KQ*{U zYHcFCmmv=5zg05u7Y_GQ?VqhBNgyZ3ehG_2I~0xQ<8M7%ZZIT*7HivS_-_& z-x{M8SBhVy@*~-|(hk)lek6gGQxy?W%gKip|3w0`P4p7JHU!*fcw?9!q4C{F+aCbz z3yj6|m#~^IIqdXf5{Fas|K|N)lgJ$4BNBltOHF%kx&YX|l!;#-CWj71Eji9Ulw`K> zhuJ?ov>*^SVnE@>YybyiI0(Tlqc80l=rhwAlSyJ$ku==2Dr*j`2l2A@8xf z0Y7wp_S~=@QFLNwmeAGHa%dg4;2h#{r+1@-9@xK+*X?R^Q`t;MC-`uZ41q_RMkz4K zTfhLg$Zq*$^%(IE zTAdlR&*gu=pE-EaNW~Gzf~+Y)V@!?5g?nB$Mf)=#1-{&MS+iBQby7~6)#XB}O$n_S zP06Fr?V}SHSi2puLSMeozTSTuh&q)x5ugE*ul@i%Q+2FZat8-?^ILlYC4ZsLF|Ohq zE1TuEhye7ppvO5@CV9eCc8QP%-mU{^;HCfZr41s zR+RYcSDGryTIY@SpU!9o73>py^V_UHOpFEQrZpQ^uZT1|XWsCR^?*`gX z12ukfirjNumC>~0pBaIG76ldo>W;2ii>1(qN>}AK4Ve)P7MU}76M@P*UczgzyCx>p zT$VO1Yv@ioFV{Z(BH!i0HL~!OfWpreyv(~{QZo7#F_{pL=;fBc2E@(vlWUi7~F z`^x$ zve;JkA|%{G3Oi~saJ>fVW2zt}5DTicI#vqFHMNtGoBM*t7Rn&?1Y4(W;?Ribjzfto zDGxnWCH`k%c{~q2jKj(vmJdQg>D}nbAi@O;=_K%#;{IoKDBhfD;8(Q$3{Gh^*+#uz zRL1xXvb~^{e67DL(*tZFcP8sSsA06|7#vmktsL%<->-kZV&4GC-rHX|UnapbIIu0G zMmH6!QjK$0aFTL<IvHd8H&i z&?0KEextS6`#3pYH<=-mx%fs{dj1?mj^M<+$A&dWketCL#`y0RJRRcQ^AYtcap9$J zYRt|@SxQdl1scxCeH7Jvd>0C1aAVmg8`^47Fvj7V4f7mIu}9;>at*dXVfvahfaFJ= z{f~p|IcE`kRQcrL2dOZND8WU%n(oh|Z;tB=d~}b#h>v(}Odk47N-xUl0c6BOOy8Ma zN?3x@_@AkqWe^=!q zH+kJH`j?j{>!~|Cn1n}ppUovIrhH@VVT4vP5}UX@;csTZ%T0T>%v zKfW3Tp=|pX^>6MU3!pc&QB{CC!NA`}M=)`_YN6z8l+zM~PR#n1GWBYj+anzCXAUl% z{C}Ag$N!8_SF`R{!L;Ub9-s2;+fP73Z`Zipc;gl0ZtZh5bjh$nwIuhl)xUggr{avK zdWB0wy)D+^h(2J@6H%$-YpYSehtZS>73|3KNl_ndF3 zFGN`JRq|s`PS}@pQ`Rsdj0oLkEk zO*y?!8bCxtY7 z%!{>_toH5{*-F0uZyI-&s}aLQ4`m_8`c-WM6+y?pEP4hCJdW9(H?%sQxiUgn&w*GP zC-sspd4~tMfc^M(!y6j(8LO;GO3?`i?b2OsO7c1cvt+(2Px`g5jV>gKbCMid632jR z=7$Zr2Q3H`4uj3lfHg?9lv47dHPq0?#bj_zaOT7xU^d={t8nSgSDw=I)J%qYlQ-%z zD70;({TW73b>uj31wMofg zd^!O%Er!BfQ8$RuyNH#4;{W!=YRti31^u+& z|ME57-A!6cp6Dj>57P6C5w&?zN4=W=Qr(X=58|cW(_jN(YS;ffYVDni&M>@~XQh)Y z3e~Lnn!z-dA_dGnkvTeA6(%(Gy1vdMukd~RF+S+?Z`)>5{#cG5kb0k}-Q>)X zp+8QQ)tL_>_al726~k_|LftY2(5>QbrX1`@`Uh4pGvvZ{HCAHJIyyzt;?U#ZK0<)LZs=J{WY_ zN^qqCMjTl=MEbdC09LH@%TZ8RoJTdsAZ3>|F9n9r(!MwLFxn`(JI*}1v0VT`=j5T~ z)cywAZEk=y5Kf@C5Ly;cJ3)ELBGp2~SgyAoSX>}S*m6foMe^CDN~~lvLF_*`wrml8}Kz1H{BYS4)Qzv5GM)D zWhZ_uhQ;QoMv1vHO~o7j!T@Mn7&k9ms()ow)u9>s-06K{x6Y@< zpYbK?Z1QrYZjA=Z&YI$dKvk!ZskBD%>udHkVsX(0%Dd1ch2++WQw~H`$QZmJNSV7= zRV*l-R?cOnr17oWg5DOu7EVd@)cI0%=Rtour1WlZKncEvxAWL#{ylHYJ3`@}WCm)p zf$lr^1Wxe}q*X3C%Alro26VY!fM3^A_hh1MAV^72SWljFiVwnY(iS$5? zu*C7B@tioI-U60j_qH;ib-m9DzAmuM1Yp|^>CG;Fio)W!-)(~_;xH^%su)%Mj2J{5 zJonG}Sb8LuuK1WU=m0!rP2&dbUDzlkS-sNV(6uHlVecA+@QX#dQjviTvUqrYcgo+q z;AIzfC9|c-Po-f8?b1J#H6K!=R9YbkV`}h@=O*>MsU-zu%WB{-o-)Z}SNjnU4VMka zhQwXAub>!bOR8Y%W$)He{NS}kDoq{I7_!}Dc?|nP*)8BGfbjujG!MPekNdgTH5;^l zR`*L++^83TNYiu6N8QolBy6Ni4;dA5A- z2`2F`A3ozRyv9`aRZ8K2$o2k|_BMis4M=9WATmKpU0vs%KIsELn{Zdlo4H5__5f`g zDqc9c%427wCNms>NVHkZJ*n=&EzX?30z z0A#`%Ri&r_qW!cgJGt|;xw7qG_d<4%jtFb(e+ym2PAD@>3}37xTr!;Y@G|&tUPkFt zhoAQl%!PG1{3KauGOouZRlb!Dz%PJVBEJiGMbrT=??zegWN z^Jn>&;)KdR?J9LEpQo8e>&byL``<9)?u+Mdo&u;c@qg?yK&^t4giOSr!C!YjJJ4Lw z{y*yB1QPC0C(9u9PrP14!|HFsqpB0bu2QYHeydU)cDCR64B|{uexD-ILMQw>!`z?U z|9(nMes%vD$)bLtUF;t>@geNUfF!OcLa4~4QlcMhOY=Oj0f5JyWK;B4HxB}b(#O1j zDl%+O(O!q9!O~xB))sLAfoq!IT{zyZ;fjmlpyy_T`(%C$?0 z0Xy$#00$*Ku4%U4RfZzV*vA6DSCv3g41Bg6Q_P2pRK`W0Liw*mFak~PY+WF2; zWM*9_lrW!#-GVKUM&6fJ}__nXmS- zgqB9Ui~J`D(~8>|n~{hkk%{YTXuXDxxVj>9Jv`}u;7h09RILUmOngnFSzX^!+y?*3 z#~)oORgV+{7Lx?J^V;^Se$J?sHu?Trpo_{MdGSb%ob6R&Uw)CNd3c~?trxpYs?Vf~ z7`v8h-12Pl$GaMCn(uie;^=qD#hn1Js-LP8mP92kQW-B9{L)aWKIb>-#Be*eBrHi4 z^)psx$C$?gn9@1VO}PolV-PC2$+9c3aE<5+>q@nV^6K4 z{Px|J9+w5-#&+eBS5Q0aPrY_AGovMsXWi3l$#URYUs9@I2`Ll*Dkw!I8lF2K5@CFH zO;gOro(I#e^H8UNSC%y3d$`A~v)8_>O0hsh_)qW#JnBAhB`erB+YaA5LSsXuw}4D^ za^31VrVCopilChFu@v>Ru(6D!x!|YHa35!b4Vrs#|J{rpaZZ z0A=LKzsXN@*ZN*dUHmg%l~GYx)gERgN5E0qrk0SD-?dS;K@aZPM7{aF!6um;Ye{cP zz{C{$(RNBmNE-|{e|qV6UiwERpxljrfAHRszp)B;v$^qqZ>>LWtH>39ob?+P?$^Du zG@QY2M=zFJFiYB8%%FBh{7r-Q^POK3bCU}$s+aRkwp)hfzRb0+47O+-ES-1jp?k6M z9TmaGF2Ki096kLO1cX|C{jA^DLX4L&_fyW}1FpoT+q;ZhNkQ~lcfzvum#7U0ozU-C zh$u7w!Fd3CPRzUkt%@!=LGHj=ljXq}4f=G)Nf{#`rZ1h16Q0#FNH}s8kIR(F4^D%? z+}iUw|K7o;5u#fJ0-Xnp&K+3^ij7Ru^JjIU$Q!s_gbHnzc3#@1Mg@E>F8@YfEa8D= z%yyO6x=jZP6L=+cG3 zdL-f!>1d%CYS1$DP3S^yVE$8914=vsu@Dnr_>p^Ht1r-XvEDCX#OMf)f8+nD>YtI5 zO*lVZChPc0xP`k5n(aB_h2)stKTh*7ZlvscgP6v9pQLQ>+AMY;K*TWa9+H(j3!bE^ zaAuka)`fUG*>#cFZh}YUy`wt(?ZF;b2+Jc|%nHLdK9^>`!P49U|4%!;Sk*aOE3vu( z<`tzd?IQmTyGCRU&cpflWkloQMR42Mlga6>!K;tJccMRd_P``y20^5ZqddtiXLp@^ zf)Ino+J|SWAI%q5Lag3M){u7E1LDPO)K0?u4W?UGpG?=`LU9PDme((H*s0>e55M2K zf0EJyPl}as;gA(Ml*p%cV8JNA#XG$^b}iwsZ#2ih;4nQA7)x*e3OF{h%axHveD<2Y z3`xf1GaxS$E-HQ#-F3?}y;u=!h?Qd&&wykGCkPg-No1#bH25*}U<1J`siPp!r2CcU zFFg`&|8+Nq^PZ!v!l8v~4{e7mskeKU+&Vx0yNN*{}jyPY+2zE-I12aQ> zZ_V8u29L;pf3WW$RRKJ=sElWSviK>o20^o>hw(jo$Fv^Ou}0y6Y;Jdcq6iScO5Yb)p0KJpR28bPnR52$tmR)40ne54p+QXAieLA3|#HRERCkv zA6gAhl+*~L(u(0Bsq^FC``3mYo2NS?9sc3%YPhGRy8@kxe9~SFx|9yo|HzBiI#ilV z)K)(vITQ3;891D89X@aECak{r<7zqBiSN36`LoWJc#)qPf?S5@4>$fgUSiX>{|gLW zBl|U`oSdwH)A}!k@7WPw!-7T60^SdEP&V|g{%xKeeIvSp4j8Z#xO8FP4Z%_ZkBZ}y z&^oojK@Ym`tlI%rvIM(ptg%oVHX;@0LR*EiXa4tulZT%2b@!ihE>Kj-0z%fgI#n?_ z)=2rEP15HQD%35v-xL&B&hq}_yEmZ4S=!?EtIw7IJnnei+@d9V-BK9WST+=l2VOJx z%~UA4B*KwnK6AFug83a|J#xul-Hm(_esjlrF*(IT!zRw+L1B}QNBjyzk5{(yFTw>< zip;|UgU$l!=S4|BFamWpj;@n)LKkTS945|FICof?AdAExRJE3-Hjhn$*6Ruco?ATR zO8T+;`B+RuE_IhasxOTXFGgTKF<6wd?hNER{90!r=NO_KyJbj(;yH2$8sUKhjpdo{ zEi)%_)0owXLjq&c0$$aiHGQQ{HJt0W(O{lXg!hp%p8K1H41TwnkuAwvAEikoXBUEqu|2Rq0nedv**x zW1Mt;=KMFrLw~l%7g@%4l7Z`J2-0kn#6NZ%A~4)_KS3VppC0AX>%VkRORr(<5yC6= zpB$RlM+bCAZ{4YG}g&A@v5{o&K}`h>96;XuAhzzRKd zUuR&D;-&yK#Xb91xVt0>UQ9ptOzk%&B_?+hA(ttumBbr=0s2TTSTcMr;)t)0xi!!0 z{C~FW^{R|qtA++dv?kvk5KZ!v5=Uv@!#xk*wHhyd^ZI6^lK~WbFj34y;&Xo7WV7f% zhF8_MM3I@ge+4pGP1-y$^W-SghNq9@J^8buK&PWiY@M&oCSMxFElc(DX^@aIA_^`5 z#4EC2Fv$@ChYa4yPC`b1`geI20L;nqe=f^>s{0;H4sj1b%_m%YR9F)Xpz0AZjQxrO z3z>G5RErq~M}Q_&#gczuL{|c{l(sDPJ()`#)x8pdkcMCuKV^jvNcQ6O+7thvOfY58 z(%S%r{ZYx)j5unfbiK8;^nkai(`PWqNwtwP>3$7eQUtj^G1`(1s-N9!K{wk;|0|p} zDM%!FOLuPvFly%`A|d&_(9Y(VFCD5p$G(7D8Zof=2_VRL_7M~rcpRst@HSaMLcj6% z6@%k^=p=cCp5XjR8Uk`G`t~)4Ak83hQcEso$xP*k;D3?ij>yVv=nMR1gUoowb!uVE zaD7tF<7_{?tIpzw%&XKDJdIGEHFe|x^Y^^(MDm>pU4+-(46%Hcieb>O0joTUhLtadfl6+9!g;)Z!q>Z`9-_O)TF^&H6$w{e7 z!@BW6vWFv^Tu8yrt7!p^Q^wB-Wv{m_L08Y7ZsK~#-o31QXIeuRqi;o1|4=ZQqkdDm z(X?=FKq0Q~4o}vqP*@GwnIcVoVL`!X-A*a@$#=9*i(RSg)bXslrzj&`Ur$fX zCAuOBBg>h`)OC5CbL(})eBD3~vh}^Br1v8HZQpX~mymkqJgU8lI|@p1`5?UFqM>pf zmv!?3<EVyMth2x!qD98{2@N#uu{<>^_E(f64)imKVQiwmj#qrXxm# zWnJ~ofPrE_PaO)$hc*n)i2BQk8G)obUo<9o)_rh(DNh?%<+0+Z2K);qz0^B5Xfe-K!i{`8N!mM5E+fX{@)!v~>J^mb3^4zAfPh2%Q9gyA^Q78GK)H1CbRk}-`z zI9z4jjk4sbYfKeLnJWD^((JGI!4X3lHJW7!%$A62&w;5|RNg;jvvS-X<*iY~3zJO5 zH79FbBc`MXhaZ^X38_y`iIDUg2ccUZO9^f_$k{2`Ze5ZWgW5AdIrx{%s5jiI`AV&mtF|9gwnM9(eTm+X12=)9Gax_0LUv${1-o0n>~X$r|zj z7(Sh#NNOG~m!^_6LtIvEQ|l4bP#~=g=R&20%Y9D@aBvK%VVp}p0l}(JScFxbZ3e|X zc{!ZM&F|8Nci3GLxEOox)ji!onapvwk(-Z4(^&Zg(66v)B|TFUsZT?x2%(Ownz7wr5IJ2 zhxKr|hd618!Q|Q7m%(&%2}ku@&bgotLJX^7Z;PQ9(K3|7rJmfdyCJTaTcAw$eU$jGPHqQ?7{}^83t-jD2 zYTzUx*?GCz(s>&*&)Gvd7gN_8b|O|o$rx2ni>NoFnAol%bI$pbbLS)VzzgtS1*~b> zZ~p@Kkoh^0Y;x40P#>n)weF+J8QPO@x$wtxE6;zHb28n>y3Xd zpvA0?JE1op-Fo)9HXk@KM3(PLINY3Cb5?YicCM;o;N@Zy_}f&E@4m#e$Q?X&yiE|p zC^5dQh*9#PaMU4$V_Y*tkej3gFh(d-0_hQhtB~>1PKkAXQT?|*{FlQCf&RFrVb(Y$ zeueQatWiBBA;|*n^V#2W(D)~h=hK0@e9X`z_+_{D&vbMe0Ulz&_}PMM0Qfe0%)cE4 zcTI$(@Ov9aEMXiYDy8PbuvqB68T+R@%+gymF(jCK*N|ZqL&IQvX9?zE>=@RXc#=fM z@Bo$FjVaGiN2Vmn(@v)n3>@PKqTl8?k7;7%IZ&P+WQ|I{djnTKRZYvz91@Bqwh7`% zkorqP;zZP@UhYsRzQ!|^t24ANaeg4pXX`CGM!J>9b3@-t*@lk=tdl zdR=+_GBMy=F8AV*H_IKh?PKDpZ|#vQ6Q8*}dnWN&5|Y95RQ;WczrL|f%E>Dpvw~&m zYIVH9trBu*esD>Dcwm*(gX;HLAWa6)TwLwmH@o+Z2?w|=q)zTL*j#}05hD6SfL24zQ@uh=qxjU}KyquBpw&$*PYcH$M?{)~}Yn+s~X87k)d7M|1UjCrR3$wi z7eTtf%DVA=r>>OMM1oLOsXPVB2*G;<6gE*=wZw8|qGv&*z-%dn9w@(RE;Ru5wEwq< zUs_03I3Rf^EVrs>Q0^m;*HSOkpJqAbpIAgS(oi`Yc*y-)2wt$+0dfxtC?0-f=Uzvo zeOUcxq(9s;ssqv=Q>aHn(2EOX1(yR zMccHkgI8q?I>q>F+adhn<%Mm*TMTSsla;H4hPTIuDS`)6R|5??9%?w4v}7myDBw-Ixrz__G{lwdu5iTI{62yz?Vm!!%oyn z)C(Dm?1YI?Clc0|5yxCph81fWMh%`a{+WNokw~fnQh!|kj%PUJ5q?XT)y_ZYs6JG) zX7DBjyZNmg9T+mwJ^#?tC55Lq+sr>g{^2_(YY485@hs=NLFL%Mk*QXRSy*uTeRaIW zd;~J5Wa(*9_Cn)60}=}P>%ZkEbxxAp(U`KBm3WM?fi+5MM>FgicI~?ip0Fxb(y~4A ztT$%t1JlGs^x5=0Ey+7Y=YKq}af{c3%a&)%c=aW#(~BTA}ATdzgp6tho8n z{7h#>F!dsRo9ghVE+rh@#2dichH#9@Qmn6&7NNwO`M9+}0Ov{>tM_k~Jn1C4gp?06PCVk%4NOT2n= zRMwb{&ki)X^Hv6xBGBonmNIxoZ}4eqnzqN^fFQfu6XIW!*+g!hnHF~R-OCnLgvuPV zXdw}_GMSbwLtBV%d~sMn(z{*ly-KKSMm3M8QR}9k*u`L)b0~3ESox zhUAkiLlYgBe~l$HDGrRwK|=E^F;9p1A&8?N zo;EaWk~8=NI$SBo9FW-Co~j@G-M-FJ$5N=zw?pR|!|dCg=LM}rhrVaHYl+?*$A$Y# zm$eRBAE4UxYz>LnlWC`D+kO0sF(# z`t_D-{8Z#_Pjk_NBfaHJ(}{n-;VwgFU0Br$R@nKVw8U%fjLTxHuCsi#D{nSv$ctYv9b z^J9iEgBlCHQkE@S)wq()UrM{gR^|-<@yj7H@OyMlv0P0SAGpeO%os4wBeb4MU-hxN zTQQzWSSIMvC!_XLEa5Ve_^(_>%mDkLL0ASFmUCee_b4k@RB*T8s2HH?HpA*E6C1Iw z+WU#h;x4nT?Lz89;tcH;T-hiC5!~yg7KAffTZuP`-5OTiDPwH_v#pnJ-o>u3x@V(u zfIbO?Jq3!b1Lg1GrwV^F+N24aKF_Xw&47 z!oAeuwpUmNV4LlBX)D*&_|vrLmD3+a#$4pm7~T!uG4Ku89>Lq`w&iCQo$6)%b!@~~ zt7T3y!1Qt3Pf*-*cU(dHKkChCE@1QhXRM%M*J)Yw-D%IQ^Dn9gQX~r@8#5br;t(1n zM62S0dfloep;74|eOLX(avOW{kE{HoJE6@X-+V~a(f7MGKfth4ZpPB+Fw4D#M2>}G z=zQZiHSDw?!(FK|gN{$17L{ogbkya!k3B@b-E(JtuSIZ~;h2O5C0z@4d)EJa1M@7c z!3e%0wk*}-6LGYtT;YHLLu72h9x!Zttq)YT`;xbZh^m6Pv|?DeX+Ulrb+`UpJI9W>gDY zb_2cpa;GsDXpim@T?^BUwe|DK^su67s;Gn|t1Q15)ZCJ&F-{(n>QP}UVz>NbS55Eo zbWi63Z_h-ZY@c~DOGLSFggA%d@HqYRh~*&fauYjPu=RkZV3V)W`kN=0m@W4@`x>6sV7%`CFoD2*Xthrb8Vh z*Dcbh-K6?ADznNk$BM3#{l0fx-t8#CgBFiFuZZ?7UN{J@FYODl^mwg7t3Pl>9=l-TMw&<5(7EZIPT21ev2OlM0B- zB-T+LlQim<&}yHyxlxlAlNLel54S^Fs%Y4xK8NOj*uY?`@#* zY{T_dGcYB~_MQ!-1Su*z;&iDE#v8+X#(NzDsVYbAsv5`d_x1R?X=G)&jq5|2e;9~P z2CY>_>l3cq`t1GKcCw{;Zen{yEWM4`2WH< zEAF4YK92}{*(qjL+;uD$V_!7mLod4(9sa^}p%I(aUvM3jn^?>D@ zBp-D{4<&Dy$ZU8S7Wo>1-NC%YAJ$g?M+B%wv_{!g+Hfc>E3Tp0ifPBU1Pt6RwT5f# z1so1Vy4y{q1e|B(+_C_flqrH!_e1l-cOCHcVXXg+vb_L@iW3xtIsq7Ww;soco0BzO zfAJFiET8qX!Riuy4-KvnQVs+_z`D;X{njpljeuRF7SxGF>%+dtuLJhUNAy|Ri~4|R+v3vgR4YsRxR3Nzm3PLNRw%cio*xpT zTMoxU#fp3ZRIF#XV+CP+*FiorqUajMQbptfA1YF@{y9Y_#gK_o3Zz9(W}AY&p)?1P z#1*ePm3!DyN-gWZJcwgitr|^aRd+69CAA9FDq|d0ok~dWzQ6i*k9?r(GR2(%El@G> zJ0#P~Np5`SXNmNwpr73MBL(_iK!Wn=6vTVNmAzp=tIyhw89*~65dfZy_y{`lVk`zd zIszDBP*?b(_Z46un2jESm3$zv-4DHh1Fesj2R$7Pp{M3lL^&%N2o_ft>cVSL6Bc=a zCr&$NA97lppvj8PQ2=a>%~7?BUCrmzyvxoAQ)=f#WjUt#In+c?+Zd;^K^;WvSy2&V zHVakM^)%N(V-{`ixLeBio0-c5v*qQI<(MAx)N^0wu_mF_i$kI#T>G1ADqwo;R-bU|Zu64^*}q-r9F9 zw&AwpE}K&Q?b7({g}iN`XKRYkO}ESI{oD*RB)y-;KDNVHk5BTk#5s2&2Bo1hZ@r@95({i4 zVQJOXw(%1F`YX2Rl3Lwi-B&GSV@-ZlkqTo>jPzrXa%|E#)_h^F-pSH*y&j+8F8qti zpn~{*u$4M)Y=2*mJl_sqK-pxav0?Mk{xYAx0G>-~*sA=t$1L4XzK_=od!5DEAGX@N zjxf*tZoUt<(VU0PuDJQXYf_v$TA{o5Mu3h=H2jnwtYg|Txi`XvVSw@06sewZ{K=W3 zL);q&l)~*NUz+_Qcwa{er#9Uh<*$^}n#*H=MLg=P;)-&F6#=F0$M-~4oq}VD>bMPz zx$k9io=sgIXWy3rWbxR+rl3@`Q|rGQJuK6*6Sv3CYA(RWT*Y z8wW_pdXWCD{FL$i)-JMf@j(N8)vu@Cz{^rJ5xmUOW3JCLtYW3TC2{CfxvP(fD82Y# zHx>x{CM1#yEf)j@=84ev8 zu$=$Hlti?`wBB2l7`^}9_t!PmX@h%u0V)ep30EznI@GThzfN3LIOCsI-w~Y*@B>x9 z+#dF__CvaM-Py_>fj&jI$M{ajt78Px=2LR5!wuS`$0&46%Qn8Y>s$d1ut=5H8$?%uB$^_<^tXtdk!2- z{QPtgE-=h93`od=>xE3~8wbSddrQtl-sCYEkk({)L{&LkDbpqm?4WaSgO7JDpcxqA zn&-I`?Powm9JBZD0Op^cg=93gEp}|rNcQuJHf@VP^!ncC0ipr_E=&e z8)6JQ>>WG}JJvk+X$U^GyuG-a(D5eR?!2nHWzR+<^57l<&MG&4{>dx{!W!e{)GySo zFc;JOyRk#3dP!%K)jvITh;8htHSXJWYw39KC4)%I#g%42v*A~xXW3S|sSMZs7rPg3 z1z%q*)rs>nROShbY_jm6S~+ILxCp+rvDHoVo)e_8@17Fb4YgS2?e%0`LP@3z4y^(0K|Dz5I)5enw>k*$^`Nl#A!(DPA9)@qi>~r zAFxRUI5u_7L8eQRnuqzCt}Bg_9M(&;BD#sB9Q8QYioSQ36x4X9eVPU_+`%Oc#~>Qc ziCd3pi=C)cE>H~9*r^wFGP)DiKxHQOtmfsID7paBTK)+k(>}58qW4T@(z{W7^Z^i= z^~*en=y`3!7&QazP! zf}}!!`ZEswWbloGy@}Y1IogGUlUMpwDQ`8YjdG8&*vY}9D_w5JRsvHmK5QDu90!hp zed*mbaQy-#=%xsP6FU-$4KMbrHs5!x&N&T{W-zEMWI5yV-fu%X!5O8mxuOYH%F!F1 zp%DMr6=|gK;55+pU}Hxk?vY`?=S<$HZUQFf$OqFXCc*d#rB{ZAwi+bxVo8z?h?&d% zSZu}Ye*C~20WFP67-eq)TQUo`*0)C<*ogIa-r#RmO!)B=iVE_UZ|#~EB@ir|@i&Pq zVu<|j_C58;U_m%F%S>d%FY4m{!iv_TAtr;k>F*tP+Z`+ss0xY)}E)=*FKtG&)=z(mQY>95RZuEaWi zhp#o@BsUb;3!yH4DbcA-##=zP3}UT?hKL}cjv7a#sOVs3VhpggwST1((zW*tavyJf z2}ohHfIZ6y*FNrc+1AM1#0@nc86-+l`t5_t68Du}QEy+i?fdawmajcvfW)J04Ld1& z9fQ^@x3m3-XZ_F_*@=LeLTd7xLWNhXrxuni6Y&PL*ttxB$P%7#2BU*iz zxp5LysHAh^dn+2YqC~X6&NpYTeQb?t=zyfOss(@%&e)B(vVo2`;5B3kk5MERpAjvS z&CV1FAMLaC+fp%8mWlEJm|MpPt8}+F8hc?09K`syjqjHv^~JPuHUuUCp;eXs7bv=1LDrqY9-gOJf!;k8Hye@Snl5 z+muI6c&V%?Fn{HmC1jmi4Dg|u2NKOcd#l!+a2q8AhPcnOokjFlUG(H49FDr9Vi5xP=a74+GJrPTPL>njo}T_q@?ywWQ>)HN1Qt>E~8f> zpgO)}Yck2^9YH*OgryTEjbp1}D?4{`FVbS| zlhg6SkIMHT?6LbHt?k$7(!c0zC39?+Uy=eUsIn#7rn|`fNGITv@ABe&w(YG_T|fX` z^JT#2E5Zk8v5h8wNa^OC24CE2s0REf0ZcPJ7?^C|^Ayv!nx4Po$~WYid@IY@9vH!3 zLUGKl*w{>mRt`y={1n>(xFN{O)YhzkvlcNA2yt4te1dn6a5~g)`AJUuhztXd+-;gQ z&lxGUNJhWU&X!LG9P@5l&@@rMwu0*zq;T3^@0X*B53uc2na?C^RnFSd3`b38pb3Hf5$QtdkUUn zR+CzWR`RI!v_?o*1E=!k=)9-hm%()5{%=Q9Br6WksU&-=T?wOM=$Esyjd+(E&jYQRF;+wZnWd)+N0IO z?E7oeOBaPOEgJy7La}ojIi#FWSJzc5rY%DcQ$Ij8;UH?rf^(Uc?V`v0_^t1E5f=EU zXFj#FarKXjJ)o*YoTNgG#0-9NHDqwTyTzWu1u927q+jXbv|4?1aU9hJ~2Suc#=SH+0Od9lD+J^|@y z_?A!i;-B55;7-jkoFtTUChqyOTvO`$($|a=ntB@Rbni};IJ(gvPt&2>9c|f2a~sVv z!Pf<|f4$W+Ny01ncLdXJX^213G=sg!FXZ$~tDCVKEP-v{HL`aKupnZJCu*xwPehC9 zueFsA6s2C+k}$Z-Rn?rt`m4nwrsP&J{pZw<9@IfXBD!WrpPpk+d7z3G$aDTJM+vPyvg^Yv_{FMHu+*2Hb&og?7EpBh*xFF4@cvuM zq%;%vMN{ozNu!i6ne}ue_D26a7Fb4AcFYs^OtNtg*^P553m;hT=d5Ux(OYF5C7n>( znhQI|sCpEbe?$Mnr|aBOl2>|ty6;9!LyZX=c?~H2)Xu+xL2_7e3%kBXgVaaogiLS^m+hS7_a`v+m<+;L=wgZkjrm zG0EV;dD=s6#;anTKbvDJ_Q8QK2kP@KI#X!n-wi6-^BIO7`sgO6PS@R}k6C-;JfS$T z_MU;HFqgL@C!LMD?P0teNL3i7;R{#o)$5|io$Xw&U`**U*X_a+rCmR`xlx;!^iQ9N zI}y^+LWxxm4KFCciG;n zw_Z6e5*S_nhHOO|(42H#uG`r*lK9p}i;8u8gX>by7@!N+JQNvrGHzibK{soJBex$- z37N?rk)IQ1ST#*8-za}s+~&+mjCEd_`}Oj78%2=vkJ7Gml;v2Um3l;kKNF@vqN0E- z$woR~fdb|fFUYXUlRr_>yCMfx8yQOXrOV=)wOp&E>%J>9TB0Sg%d9o!-fL!-rZJ3J zY*;vz_Ejny-NP>2qX~a;pui%_a@fr^MY#MVnQFE{=Ubj_C$InIFu0oM)QknlI4X9* zgMr4$@Az6R6?Y)c7LC0*MEr;HmO;T0`rPVzJsABlejXD1I zIRmzC+a~+IDJ&XU8M<-fa~%bi-=PL%s860%e-8my`=tV!>cq3RwPcep|pScE_ zmbQwr)qP1xDA63DW?|I2&?AG$U6nJhL9C2k}g342h)5^S&yZpF+GLC&KpWg zb2LMiwvb&SNzW1H`o#pKrFl8Vj$g6d!q$+IOJpgwehh65XH8w&(vo`9F`!s(+LCMH z@Vkd`qMj%$bU0vx=3Je#Hk*~m6+U9e{_AIs^G(mHJS7Rq6XyNE?sv0dc7G?to`8o^ z8rtOzatX30*CKcYcU98{)yb?UhwtbaV18Q2iy5zK(kP~e6eg4bz5K5BGU=UqC_1E>6?ad)?1@=QL`T|%gzC*`uaKsS*fsNQRm`oJ zCg*p_pnVmeJmd{9HZ21FUNC}mO&!9OHq<1zmT(ET3V1})&yU^7v?JisEw{YiiAk5y zs{YEtX|bh{WUVuR8*L+^RyBWIuDAKvcV|#WI zDdcqY;8g5Q@_d2|?;8^~8pwA~6W@i=v(>QXefW6`7%$Qk;|1d1L>FpUZSKW{75ZBR zc<{4>C6Np6AmZ6*#^3A0L&_Bhj2AzG=#(iA*B_>X6Z)E&rk@qfn_S3W13r<+Ot1M8 zv-_*5K{vD8zj_IcEvIx_;v<8BXoT0pgMl_}-GhOX*Ec<+*J;5~db%67Jw&^yTuuL7V+{sUo-*(&*PHkrPFyC`3$rErH+j3KcFB{8ehP{d5)T>Y> z7829Tm*-fB+HE~awW4y1ee>+#{S+_5r*;y*(-|ms(hM%RxSqKPTtCQ{6UgX|BNGAY62=J`B7ak0p<4nYp9ckTZJqPCS zt{$eQSJA5rmY#v&fS=Y~+{r63z&m!8srwm5im9*NH(lM5k1#-B{Wp+F7r2dzvtnbp z9ydro$LFwSV$}A@CGLw2pTVZuf+b6C#nv)CA{tm+U>ef2Q#BjLgpE1U5Vp8UI&f3S zKg$zEBei=%Dqpcf&xv}OJ@lqsROoJ(;4VZAx~% zb&+eZE!H0o53m>O&eup2D9{^Io4j|SCUUp%_TRH=Qd)yqTt>NZdt?Ha%Fd4aplgKy z{kfJXNY+eWI0)7iRC!;ng*;x&d4W7&EEodkif9CW3Yus^lcjLJ&ADrQ7*vp@#GAOr z8OGmZrug-dN!!1W3u2`ai}`VH>4o*3Kd>LoiHQoC2R(nh{xSrU%J$?N;2cJ@fC z=7z#@u-UduN0GFf^7-kcd$xa$>YU@Q531eRc>l+U>o2E4)*Sala_OWAl9)PLWG ztW}A2R%>Z*RUTxZP1q!N!0zpB&U{%joCAT$LSE*!O2j9n{r;AgLlSHip$dARL&5Az zRRJ2nMKDe8$HTPbtLF$Q@p%g0Pzn$PCrH4-OZh@NXqh6 zzWV(w{~Jv4YF)G-`&PAC=K0jZy=n8()2+|;`E+-D?RRtg$_?{FY}L7JQ?_K*MgZW1 z2iIpUiOKJHF5PlS?VXjZDy2lIK5)mM^$)qVCsH6f>w*Gl|NHA(aH+8Mn2M*=w*u}j zo7^R$HHdMbKnduPMWn9r(%suM{l1m?c^*=4G$f~FPkr>oB5k6Yz3g?dEoq2^=2o~E z>3CG~*!?XAq-O(X+myhV#W$>>G;JoiU_&Xit7wDKcj;w}`j^FI%HwlpI4-?KT$Yys zZ=riIL-y3e8kCv6Gv#%J85^F*>0Sp;G^)S1FFL0x@y!r|EnMj9a@9?S_^gAr>9ho_ z5BqGFi}`F?up!b$QJ(&q2SPfY`E*H&lj`d*?CICd=agDYu-^x0~eJ>clk5 z^jBv?p2)p30=D8u94TxMrYDnAQ>D4N3_e*(%_W$z?$Sfs?h(Tas&)bl-A+GgwtnU+ zBv|;~0WKSVbHiwrL!l~R_-k!=8V3Lseyja@PbKtoZkW6wLk!@qYoPAG`X~(Gn|@!} zZb&=j%q-W|DQ*%4Qlj_--5qe{`>(n(EEB5ICkTL)n%4F{M})~84I~Kspmr~Oiuai9 z4xREL9VwJ#{Tm=hc8-+3jF-Eo_h75BPYMLA-#miPNAUNXObt&;FW~nh)-8{3tIG8m zq%ZSN!eo1B(am7->PP`Whwi>;!lnW=>Hbt-^}bV3PQrjx#FJ3*>| ziRnoErvx5(#+{K3xB)S-b$ui)Pce$j?)K_*aJ!b;de*fM$tr#ToT6%d30cDS$u?tI znXR$?tO^n9!t0%(C+y0J5sEWXVa>o6XHpnR&L{k^w6Rf%Gn##`{%UNrA^yvNs z{90aj!_n5Jg{BP)DUoCmZujfYFDBCGU6N8=F#ieKs?#m zTJ86xQ2_;kn)uvb!=-VPW24LUz_dbwf;Er3`LX}AZg+tTV_mOrJZ`&@f!WH#_iVyk z(QVjA0{9L6napa?4xzu^C9YqDu3p<+VHO;2^gW#$QPO_f{nBoWZrN2tia2Ha!_)HE zG7~B&rnO-M&+)1{{qODd^p!j)8B7{jEIcCBu7#V+V~>Zarg5MtODGK(E`wu(a>(U~ zF##u3i#foPy7_oiGQso-IhTIiiG;qk_7}if^UZ@&4B`=pFbL0nB(A$*)-?*S8E1Jg z-}VJm;SwM;t^*8%t$+=@5G%KyzZ09Bg&Sic>stBZcwEUAZdV-i#5;Bnji z?Y_pkBrq@R@%=GoadowT0Rb2fup9pgc$*ghtE#aN+mQ4CW<<9E`?FX;eCeDymg~9; zmv`bp^vzl(lwt214lb@)8bCREi`bn0`7`dMZ=hqmM_!{%umCJ*xBQ7!LzRaxtGZ}! zBRs%R2D#r^7hEG-g zo;|bek%!3O2GBT9(i>W;JLxxraor_|zhci16QjCe8(~xZz32QN>}`!Mf*`iZ+fNq& zSk3i>X`r%O%5(44J!H#qrBDqyl;px@87eTJ1lhC0>H6U@zhF{>4br7&514;+NoDj^ ztS~()Kjh1oFVBZ8Csk_V(Y>Z|c5+ zHF9%k9Gp|^X>V9SpusDe9h4DGNO;xZGGfdBq^tm9g+#?L&Con7Z0?)zYLcc zTJDWHMXK;JPsN}SN@?aP#g4XqRzOeqOCIw1R$f4I%6jATmcMUinMi?#9)eTX2;Fsn(|NFug(`Bx9M zmQMCnq6|PQhe{faI?lHmJQRN(y`*}ZfMMKkfZF6pW7!Rn$u-(k@my1LLE zim&70;eovPbBc$e9Kt!|{Cg@z+->QaR@I)E)%`9`nI>Fd({1Ye_q%1&#AfW%=ax(E z7g}%M{$}kcIVEc+`{0pv-hWUhCPhxTP`{6efEUG;Tz5+<%o?pyYpT5wN40Syg# zUUhwqyHRWu(Y%SzRia94hBehijHpunV|*?SPjaCwQK}5H#OP4 z9d1kIPD%6k@U6=3E(^ZTqpJeECyc<@!fRr>)%fYfwG^}Y?Ywab13sncPq%e})tmr> zu?LuUQx*nO=V+@jvy`Wb8?ZMZ<({I_9n;$Y(wG(t7DAK7JO8L}b%oboiv{V1K@y-o zN-9OLDiHtF#b*`j+_u!O&4TF$?E@0xOn@5MA_G9TN}b6b$_R8(X>tzIQY-qe4v9ii z6GLIN&%<8w&=1a?3)}_(UzG&l7JMKhH=CZZmXF))y^i1|PN3HUaFWJs)r1e!DV5yd z?Q^^8oP=5jFVHZ92mKH|oU6h|fYr?fAJZ1Sx$IL8y!U#;Zia8ztqjdqA=$Ptc zwRb&(fo&iAP*mB;lhPDQ(4}_UpQ-@L>@u*lJ9ATA`*)IJoEz`@i)EnDk(k}I{L;cL zUpGfs23O;Cxk+gyVZPCa_=*mhhf#amOlA{UmHUJi&IV-&ba({!HXOKgy`A^f^p9x* z19ZQZq!!3$)owPr&Ph)9*;V631ll=SBzIuF*W&S39S)A|xaspr73h4O^GSgJC))m& z*GsuhwPz%F((}J)?opftT6Z#B zcX&49qILnKa}GcooB*uqs;hzt_oW)yYCp-}Os3g5p#f=#r^1`589cj@)Mh_V1X;px z#A#A3HL;nkES_ld{$B#`A5H$3EXhDn)2Kqf9Jmj9j25}T`O-xK&&A`j}E?9rFCQ!xp zd|I)BDq3$EH<&_u8PA{3L=@Qg1-OECKDiDo*onJ;%J;W2i(ec}+!I}11m+}Z4m#o= zRvYcVh$8j&-*5o-w@i=d58a@cDR58KmHE0MjaE&~#Mre85IXkYr3zK}Da8O>8%pJx z!sWf5hLpWca6kElb8fCqrYdg0N4WaZ;2FASu`$W>)p&ow0LxW5jlD{f8K>dy7jcUi z5wt*TIu;pqW5RI_>iTgEKqW*eFKG<;lU26!{v#6_R&7}vdji=cj^O|ua>-|n)> zy|s%4N!x$(G0=P_>!*>WnOk*^#|E$(G0-qm4Kjyv><}lr#sGtS>kFLGV+De^r~4#9 zI<=rWc<+B++v`x^dGS7M>T`k#+40bFsX}m&vuam;uxmmx$=@?o9?r=rV**STxc`$X zTnX*PFkghyv_D!J@XurcT=$e!oou8)Q`Fum?z1S!gSRuE|=@>`vPT+Gl8CF}LM zEFnL(j=i*pSIlvit?j1T{rCrx*Por+I+ORA2zgMr(YxC4vK9uGI zcHMPhlJ;%a#q8p3zWjK>vgX}U8RW0r;{V^i;_LMZ&{Db}3R&&k36`(fV`BtF2C!wT ze+|SpmkfV8WgazR64xPM4@y)_>AvHQCams2YNcFwd6vQGA9z0h3sFta_Z&Y@g78%= zztBfFfX_-PSl^B%hSG5)MMu{BArm)nKDYl6(}sI@yN^uLQ$AjYm`-EV3g_x|68h^) z9DH{c)9v550tzF;Z=2E9orw>x*HXb6HTg>(&}{y39$wo(0r*w2|Ma=;e50Lu5H@xH z3D>4Td^XTZk}`u|Lb&xS>9_n6fpwu6ShLK*yz>E!l>ZvO^Kr$tit`> zj@HkJz*_|H#Q!gF54^3WKUF3E5N=dJa!$Hs@9a0KohZ4S&XiinXdU-FLBv;h8y7FP z-F*w0ucQk^?|7&&up+G25?!x(19vz_si3dE&_6VA9?k<Tr!xi2L zej#7Agj!=j^~7C+jWI@9oXIXp0Aw%%5Vpm4vcs+LPZ+`)fb(dQ?jewn0F8T1B{#9} z*%3O&5+yZ^1oBee*S1DHi)=vL^Z1W==kk$ln^Rv|m_7=MX>e1p`8&SZ1wCh=M9vewcg~Cs&g4 z)Bbr3m9Sv{7ECQFF^bpw#d-N8>Tzc?N;A4&%BYxv#Ns2n2{u;T z4-x`HC@tlJLKLQGJDioum8WTP?KRfy;QHp zFM|4~$T~j-A+OzwUgA)ztB*+g<**2#J-*NLi@rzkAY_h+wu0yq=PyU6QHwLSDjQjx zZKvhPmIdxx|NHEdv3ssepTgSvJLEuo8paF|k$ygae1NbKb!SmBzV?c1iAUbVJcUuU z2v_`PJ(bKKX)peT0ss9Hm*~5mIi9-T{j?aUu8EoFSw*&;o!Y%(B@VAIY_&4x96nM~ zM;^;u_hdZ*RJBw2F6Pgbfqtp`QV3FVp25V|PL+7|ln^E-ap!GkEQYR}!P%k)(A)mK z%^DgdxyXM?Pgp<*b6a~nIZ}QyxwIgy|DT86_l{R+*(N`wR!QEMC||D=k!wLoY*mY) zj#(WLy%hAqv$0W(K&^Gk6xoTXqC`y;#$p5Tr=}AYFAiM>EY>x`*y_uVXFC7AlMzp2 zC3~CAC1&7}8dfAnCqDg=$iW(B=mW_Z`_C61K3#*}pg!$%h4b#7R{au@#zCf=Z)uh^ zz7))x$qLDr4<*l%GWaa6^I1lk1sef+CkY&J|GC~tHxm#NONi@t7?@dgv#2!z@De<`w|5HUX^b z_tDz$@7dX!+%r+9`LYmPascAV0>HF|OfCQnaZ*n{e<+!H$;~GEqM2sX|J_q*r_ZEu ziPIfTIqMu_4h?-tR72@#X`wHKP-wr5=(}-KlE7sNWh{MUyumWNS{ zZ^Jw-s`r=K_a$SnbwL0-?kIU^sevgR>~W|5`Op5Z$=bv_{dM+3+sM(|VKGAN#(6ED zA_am@Az0%GWGer?VPhp776WG>BCvXpf_j52>AxoC@`U!Vz=0=%)^l~b4QSVJ0lw^R zhPChFC(8`-K7XeC{V&cc4;0get==xkn#H_Wna`?j8>4EzOgj~&rJP$naj$__3!s=| zCaeJyj~oUBZTN@`5P}{%Qnj`Hq3~?`?+!)vVDa__6&%G|P;w`Qlvt|XwB-M(G7v=Z z{O?6K7^@RiRx9nQHGW?jF_r}EIW;|j8@@I?AP(TuGvO90JlO{%&Lr+{j!%I}7yT47 zxDOJS-k;8Gg#Yi(jw&$*K+G`v3p6;4j@tJmBqR`!UwQQ11hN2*$HC723xnJ^?nHOs z=s{}x{n6^+U|ipb+wJKTYedNZu2HdjmjM%$rylo)mK9%8_iF|vP``2E3jzn(Nd1rh z?cRDnJkgO#GH$AZ_0uJngo$4qq)9#hlqMw#;0E3cf!DkpH`RO1+ZFk?3cdpW2@--R7UCzv zO5op6@I#FtXbGz?mJ$p~o2mc5Kvo~S6h@U2W@kVRr%;4w4=#$K^LLy6{95<@x2m8Z zmHyd`Eu9HK-ERhH?ZCh)tKI$W`BYB?iLmP`2%kzsb)`RcDuOku*sFDgMq;P3d*aWZ zucVy+yxT$+fcFD;cX!j6*vGDL2%SLWoaiCXvW|^x4%djr11nCT-4Nl*ds6=f(Nro^1G$3kgpXLlY^|; zWe@V*V6j!dn@94*%K%cEav}n93O!*1eOauA4A%m?J72WRc*x@@H*H34={Wn8HBl)A z{O_eV=1rR#P1|7fRw>ETIuIv zTXzQ#YM#c?-16tNwn~9%nZ)CQWf1ouwpF4Nj)9+mXV8&EH{N&@{a`o*0vVJ7Ix+qU za-d~;1$t2t5f<#AIVi^n#$-$5#c#TpS*yTPjF_b75T5G=0IYFM$vQSBaP)HXslB6b znQfx}v$uY{9J7gcUl+pOfIqzRmM6AeQfnP~G^J)5)%qMU;PGeqTHv{ht*tHMT_dA3?h!39u92bP#Kc5?XlN+YR$PStacTv*f6zX1 za`J3wp>?#!Rmx|mH!+!g$l!C_aL0>wN$wgRw=7?%tN6);Sz#@g_8wcTRI`-t9~vD) zGC$Lk+LoePE>BtBqs^nnkdmNXqIyM@wm<12JtYP5yB4TkpVmxIi1M(3A?>B8x4X~D zL^=tCgWG99hVCpW&{<*ePUzgZ7IgOzc!BvpLjVP@Xg@7?vIFAIXzTl1B7pi+pdxHL znvuVD4)jWkT~_)arggBlMn)IS9u_w&nw^zWAYEbeYD4J*y%j~JhCMCK}l z*#KYcjZMhlEFRzs{yxeXuxEn#`#ods_>bki>4sE7ATck~wAf zyBg0v^W0+v#A{edwD!+ipo_LBUY0&#x zCG>W*4Oavfu(GDo0QGF}p#R};^w`V|V+q)5CwOCYM{kiNpd6dd)FYrpd+Qc_%KW!G zQ33IFeAQueh9?Y|5t7N5(&mX+bM;?5{&Fg8mR&Bpbk+J{t~uYd$H^#z{Bmx5jmRNL zvq(ql=)NFx`|W%c$4#jkN0}y6KFJX0Yj+1nSLqT3Pk6`UQ^yW#D8b*ciLJp>1`%by z3<+Pm607;imhc5lSsx&w0D8fG;D;g{klt~8m+>b(gDVvvZ;Bl_>q{8?Z2m9$@c;4k z)?rb$U)1Q&0Hu|Z4jDqaK_mneW?<;A6E^ku4&wzY083pIN5AA!xqSNOAM2)rq+7? zUJMtf3)~y12Uu!#zw`;DI~ZX<#%u{J?FBhyn)a*M)qwo!8!~g+n7Bk{%Xi@TO0xW# z8pg4@XC4`MT|b+cXv46`B4by3T;~??)NK*R0tW*2VvW2Z4fV|4QkID%-PTTshfLB^iT8YzLvnLaa&;e)_ zi_x6OkRBI29ypnHuAR)UsHi~cNyi^QuNS(4%Uk99xM8}tBaj}Q?()<>NnA!oIl|Mk?t~A8+<5_J-g|Z8J28pM zpChFC#y4V3vrNFa%``xLQBCnve8;sf$-8U{ioLuGvhy@_oC$>MoTHT_L_?GDhczV? z_7|=dzO#;Ro%FW-eaJ9zx4rjHOnck1My$VLvf`SGaZoCUU8QXw=5h1MBt&r;=+Ykd z2P0`!Dd~Jb3TiZWK5=STFW|v-&k80e)ei>}0z9&1lkzWkHvX5-C*cq9M?Zp|ccWxd zJo-?hP5vA26d!+c52xfcPJ!$SH1J0R($OLs(-|qmZ5hD)0Px56Mr=V<)sOPqxU@U z8pP9oF_Mm>|Z( z57*+YWmg@IT4SoG|NQ>B4_Yb-eBD%}_tM5#G0W2}RnTpjH0>mnK^o~8{C~VX1{(De zmKaIF-6DfcCRma`0mHjm^G_T5m^XYk6j>GIGzr_v$h5p(wtL?yJxNsDxWn-x5jOhn zP1(`0(=S+HV4#Poaiwh(s4wLQ{+twp<<@ZbakYjSayS!fwFwqh>TuHECQ4(A5?3fn ziA^2wrr5WAqG(X#lnP?j*!Gswdy)G|pul@O0xOV=ox<_IXYi%^_Ib+T&oiFrxt7&k7m_guvR4?=taf(AKgyZ{SZ3?d-+k zT@c@^MXGKuRW56U^e;6w3hYt#c>B9gOd8|%^_uo9Jp|l}QY1~GMd46`?kb^I&m73+ z55+0&#`X1jXI`XZ+2m8He3*Q_a2x-HlF7AF7>^7eT^|d9oU(> z2|_3v9U0u)Z%TkbG#({OdR))a5~Di}A>EO{``_Lcyo|?4UXZ!bWMJ?{R=@v>2LEVx zZ~!|yMBl83_YrSU%{}x&jAHV`iOGtdyssOvCI*I*Wfm(Vcw($Z4Z zOqKz+jz|h46)K<{%ekH!^K*A9jz!UrL;M^pXrNdvr03D>2}^;D4J0By(r5A&C4TCH z+T@{IE5zTYepS853=O1v_E3pT)a70w+Rf0SKxQ`lk-j7vy-rB$S!VV%lVx3n|DneW zm_)AuQSE*Yl1J;|!_a7{1O1Gk-$wwIy;7Un-c_Nbq!e#@T``6sgYXVqa0cWsiQLJ} zS|D5wp2>aq0!;oe?tH}-JTf*2UidpzO3#?7@+2Xxh z<0yPEno=#})s6{QW&Ow?25z9TfQ>-*Gk-QHVo6<{FR2ITbDkC#6=4&v_4k`iu;3Yh zC*;20s8?1_=GA%hsB(OKyb5?kf4}R;4-O0lg4giWTv!4dYtUfN-w1wzIi7=h7Hwhh z!$h54b2}SQ67nQZ%TpJg{rkvi@cuEGmGPr^lVCW~@j#aJ8iBGUf04CYen(OGLqLF# zBEjv|Tum91%Gn)RDLF9Sj zvA{~>k^mX2x_!X^Jf0bgkOgSJ98F7IzkQK{FO&qQ{BLadra!rpF++$(MoCT$RVKSe zxu<=4WN~XlsazVsI>Ui-;5bjOMdg(!Ya1>AWhU!E8oC$4czXXV`r9uHl1e?>$p-nJ zY!hJJuFLkQQgZb^U@Pf%nZEe|#HmB+wI*@ttpYK{3oN5z)qnaV6YvuKun|$F(aH<@0cBMg+ zQrIotcnL$RT=QQ&nvN;~r9M!)@!Wta05WlMKE`HZ>v0Ya4)Ir8c*FehsW=Z0CR|68 z^h^t}l#OG+gntD&+TQfMJ-hj~i(wFpFbxeQHjq4g_%QA6qkPr73*?O}Dy|stpT!sq z26d|i%e;og&tjbqz6;r%C%MX?9G*igeK^xgdWInNa&JN+f~(&$KaS@ZyY1!GDbADE z6w3j9SyPUMk18{k%>2$T+|kifJS@9Q_S}#xoY7^P&XgneQy5H#%&NlN{bBkv)C7Zk7WXyq5==E~vV%pN^9s$euey^Ce@!!dTAEPRcjnt^5n}q9v5D zin(H}qYAahVI)^6)Q`;i`do2o{i?_y=uoKj3v|aBu#-74$N)jgI6yu?nx15Uaa2j+ zXNHMmyF#JPGNB% z^`9K}ws#k-WzYA-0C`Qh6nqII?fb65DqQFqnuzZ$t)zv|n4c@!`ccre)6Sh>PY@=+ zCIE;`nuM1_e4SX}@t(EibMr8=4QT-qvc{XljM9=TH)C_cYf*3BaIg5feXDbaZ9e5r zh+x2DS4{_e4TX1#=N~9UfRE0N97`|<0I-8nIwyPQ61IwP_xr^SHOTAm!`GuY?e%Yq za7Q=dlX`N}vI0Hrr#6;LoYZ@+fPF3p%BdK(&`L&Udr2YM_Zpf-Q@X12c1X2b7k z=jI3JIPhlVM_VW=x5dO4uaKds8)jVG z>mw=mb{T{!>2~;J&#yf~W&iUhK2bJ*E9k#JMmn09XxhCWn&PM)?m*r5qi`xtG8F3w zB74JA8mkl@hDgyS|E*x>7@%g9(p<^IMxB;%Op|d)RIgt1gKLv&nA!in59sq$-k@icl6Z%K=sXQ7oGNXcAK@(cm042mwGYR#wz8{ zL&UE(3qMFh5OS#QHX9u!d1z`n|M^_PjxAECYS8xj?0B-JvG=4rxDY!K|M^m>KYvrRx9X3$^4C!x`+93YMt%N}e$n$NP!|Ow zQM*r=cp7xa_uzu*TO)BCb+Y4IA3s6}&`I~eO6`vQvzkUHBM%&=*-SL3ORmy$EsPV6wzk1%*egG3IG6?#j zG%BinMI+Xin!0^WC&*bq-~AM^0WZ;zy!@Qq$3GAeTDP)(Sd&qRO7zuNh{C< z#tU!rfEzSXe;>2nVr~LwVK;4?IDcFKj z_f)4Sm0~eZv4vk|le!^Bc(xDoao%!{85RyB)A{#s5$Y)nz6@3{{Fv6`AMLxmY$rF1 z*X`p(`;(EsO$~Qw5%Ytjd99w$MyZfK?E`0a8`I=dG6tNU@!6>h0Q32KS3+S)Oew-E z>+T=%Nq>C`>(DLOyIr?pfEXx4*#t#4;*Io11o54KtcAZ@4DRrRRcE%*^?Hv_mhc8) zp-=44S5hQkGW>g2ph65ui5^KI*O%6$sc06quh*8)=Db~J#3aHU2d3=&yM2OIt^+!N{}#aJU5opRQT9&P-;V{^up?Ml_ZLLqsu9u0PBr z^*ZBwFNf+(9oDW_H>t1FN{Teu?a`u{Pdb?S@s%wrEHE#JvGe@=hQfX zXCboaXS@O(8N<)$sCMd``PD3{&a<1r@#5Co?F4f-Iz#e1#`s~+Y;Et(37wy-U+-&3 zJfI0e3eU^GwLz?u>lf!e{s0VtTOHA8F{nz9S;;zuUbwFs!RFB|#9p_ihD}P=$ApBd z|D7^^I08oTI1}Jx8A+l0cwO)%oqUK(7>%uV${atMg>d)VRvsGW3L*{CDXnD*LRwY9}>-)gLllnjELPT1}Uk@V7I5Sg~UKW0v`rmTsT;Ke54sfp=($H(9S8K2FvG6kmEDi ztvMGHvF?D9*cXveDhl96Ud#4ZT`9px~0AWp>7ex-3=hfQil0w3)vu; znM)yyNyXoc&M<}Y_0`0JS%K}x+e~1YV%lpvbGSZ*g|Q3S52t0Xa3=h>^cUc=8rWS4 zLL|kl+HM0B7tVgX2NFn61&WiRd6Dfg%M+fg_dqUrChzW zO5=x#RR^+&U-A~eSv3JA!PoXEPlWs=jRpY|znY03-ah$Q`RBLYRjY&PpHp0od+}Aa zQ9~a-POG+z19?6!;T1S3ll(P39UTNn{UxwmePFPT_!!IFEeYABAV)RVJJZXPOasq_ z`khZ)`@lP?sG>p>wC7F7mWfv@dUyxV$|K)d7zY4CUw}o5K}Asox%PK&S$i9eR6p{O zGd#Wr4*%KdzHQ$k3dC@3qXsyW3+ThsJbv{E9A1sb%@F@D5Ewb3^6=R=0kV&U!4V;Yal&QIrpF^F1H@XQz#D zDdDCUZsNiI05DPSYK%PIv@dVPSGWp_r=Chti@Z|9y({;#5d9yIeM zcTLXuuB2!QavVhOFr5T6Fn-KaLw@SAzO1=!kx2ofRS>YmkI3X&e~b*qY%yHqx!5;Z zxVk;|*}Je2=1>y1x(HMLbHkt9_~-o}^s(x^8CpbA(6nO)#HWYA7zef5h=SefWWWul zXIQ)wy<0QVk;2=f?+1Kz_$B*e+h;3#5AO5uV2X4wLjBO7)4aWt=`NIP=5%JYqg8+K zs2O7d?k(=P-O-g&yO%?_Fz2yPeZN~0k_=KjJ zSs6PR*%(kVjMsu9q>yXGQra{@rNSF;#x7TL56MT$C6zTY<0c|cd5K})>LjHV*R=KU{yWON zP4r$3netUTJ(t24TqPiCMtp2MvayZ*^a;I64MfLiyd|YPRl@!#kh+96gSnG+iy&fALJe*zDp4+t`?^@S zr(?iZDZnnCW6(&mv>6D8{8rG(@eJ2%Yrh#UOFQi#!!$$z5Ztx&XaZRkP?w&ve(Cye zVcrDY$=f`?GO9Lm4mUv?&o1$Zr&SUedo@(*G=aIoNA%#NfDR8PP$e2pwE82T{hVUw zISHwf<_-b95i#Y2dIbO17{L@D(Q&4KyqgNtz%C+J~2^y@6@fRx>^UTpZS!;m2g$w4+3G{0`4&bV$DY?Chi>+-Sb$a z4J%ryKiPPnFe>Ieo@gMqx$@%K128An%mRLiO=vx{O1*z0;z^HdJmnvSBBE*|^3Xz{ zssQSRaaIh90QlrV+lb@X%*aDj+hXu{38QueuXc9Yp;()vqzEyes$*xLoACUX0#2~D z8@pFDocu?r^08R|2@iT)&Nyu^*er^y@cDMQA73-^z1tChKnPQwW z65)}ICMhE!?&0q4b*fo1peKH%$PfprB6Lc#vE$Txc(wK#KTO;{r@L{8+L)H>9qQ%K zA@I@I7zxXos&lj8aC=l`*%1keh|LMZ;(}PHt*A%|1V5u#M7{afx+^v^5`5z%D(k?ZZ(HIZtZXwLW?fCu0W2) zyIlBpA7)GogqoqWf8g2ABbc9u)03OZXr;NS`UGxnWo`v1hTt2D6=j$3atk3o(;zn; zavBf5eVCTk)Bou)BRFucqPCdU|8Igwqz*gvt{iG$Bc_0J*A~FBqMelzOyc9_G>E>) z-GZ-nMyQ${q53FzSyF*+PKPlQQ3Qg?tf7{ww01B{|8)?2C>|faF z$RffC<$HIQoxkMN>K%bWV%Fd%T5Z~DB{-8GqCQ+97QYVyu8DodKiiDQ%Zf5Zppk+e z5K)b^k;tU)@%sQ8lH=D38(6EkD^BPj1F3Y5Xt!TL?9~tkh#Ydse zFhgR@?z6I!kzbN=zg%gV@Rd-xvYIO#UP)0pcCc!R9);pZLw8ow4=#CyDpKg!*nZX& zgws5(5&k8QDG&?$8AIGpq=p>O5ftVYTXEq}#LHsmp3*}Fl=6G>UE?KdWbuD5!8~{# z#5?^jSHs(=ambLR+UX@M!)K~ z()q7kn+&jCe`lv}ID=mSn@Sog_x2Ng;68=t5BPWcfT3`Q+LQ`%73*d#0nlK;Vm)yL zyblKiVgP8<60Wv@Q%n~t&oY|Yp#-c~*RvxIF_^!uZeEw4&jy29kJq_dRh<4{xGGx~ z-0)=5YxIn8X<(Ou?tgMnIqus0D}G<~aq|n=KP|uK!zB)ABT$Le-3jdMQrbIdKXrb? zVphV!aCuDXg^fAH9Dj_xr4e&U2f;_i^~BfW(~lU;fQ&uHM%5EUq0jzo@Y) zgqFvz?6)647?I$AK+Ygh{~?x|GdgcylcBb5PuTRKY(A2P${tZ!$a*?mtCns zkIt(qwJJhB{!xZu))Mxe(q1eRvc7n1i(6a3*DBJY;yWS#_gG{XWhKhusP{I4k} znYTZyb{uusnrZ27v<3IEQ{PW;R_N)KTX-sC|fkEJ^Ru%@t#+gp8fKNg9 z2b4if^o-rl3T1)7MtlTNZBl)Q-#k_q%aHQ-CT+_E(H9oNrnnV;qEjgi_++?P3J=VV z2q55(1uttKxZbl3jEL(N=hNhU4I+xbgRks`k5czpu~e6wOgVP=cQ`ylLI-T3uB5dL zu~Z#sM3T0c8t`lVtx^`e@b>YCn_WOG6zy(kSsVK8Mf=zXZvmNz=fcMa%lwUzaSzM@ z&+j|v*a1p%ccl?@y~myPUO>SJtDWKm<&3xCEym&}PAl=skXEJycIAcHFF#x}edh22 zPv;rc7_fb9*tc8u#QsbHyBOzX;BR@!GajdTE0hls@sE5M;Q(j>m`A`{$K=b1Cw2NE z9(N+kZ}ko+hsz9twf$Pas&hG3o)`~j&j~Q@UEzx>c;2+P)ciOC0Dm@qMv#J#G`u2N z_C}iZQ&1>k%L>Fwv4L@^t4GU%@8<8{FCa{8z1M+O;LZX-ArhO_6t9KNwugP1C`AV402r8`t8vJ!}L6R;oOYMUo5^z|YQ(fR6D=nyIVBla)_db@GEpJ9!1 zv)Xjlx@FRJm=o)ca4>A7nX2Zce20@1_5-N}Np zACONO4^qYT#P9Te*bfnGc^51=yB4lN$w8J#^y$hdNfsAw{ADi+a!?JBg7Ug zAkZ0p$Em$&IFyg?%$t@!o*OuNq_=m(468t5G2l1O^_+0VY(7bzux@$58wsjl7?2nM0|FO$wFI1Pko2(2 zn#BLawzsi<62>gOQ8zF6s3DR$0I1~iQF%9n2AbtT0Q74giY6QZ(ntS5ZEf>;5*)@`Yz?$FIF;Gf~bnA_Fk;>MnE&vMCWF~TU zD!0jJ6#FT_$+>B=rPvlhNpj=K+NkpGBCPh?Z4>{J)~1gQDBLPt~_S}bYSw& zLaDco!G#AxU%P_=l+iQ=SU9dgGl|bGI|+yYMv9DqRiE!|N@fJ?bW5xjoP~n4#v5A^3rHa9;&&kSTg?xYu%GC*Po6W^X`zHWHP>~d$M(fQif_vo-?AOX|z}E$L`{y zZQm;~K5B;KP9RFLZ(rPf;DYKJ2H9~TR-h0A%rr>#ZZj=g%l3$1&5~!Vuda~**c5A1 z2snlNUyl=>W^s?PRu3!pIRf~9{=uBN`#XTa3hxkY3sA3j90P^j#uFkUOe6(H<~OrR zV6|}2lf_Vz`^OaWH-eVsF|yZSZDFDERg9(w-ij63U1{UEoKc7^q~h-k)yZuZe5q;_ zjAn1OEoRRsEqQ5~`6U69eQL81#pRKRJI+y=s=!kymRBZ2#y?iwZ0ZaZ{dlMA9&yuT0i{&- zgN|iup*ZxMPDW+H$YD*zlA-P|E$f^doEjD#`V_=pjY4(ib@a87+8g+7E-^i0JHIZ~ zXtQBPE0*>XWymm(u5d7dQQgE#eEQ?j!-MHe??5G+tdNZ7=P;) zCO|7nR8mwD-r=R@X2(z4rwdt9$>df^=UCuWi$0x?n{`W<(4i423Sz_G;1$3)mVG4d z#mp-lyZUKm*NSCYtw!fOTPX}Ry_T9h`;?k8{#nZgl+>C-JZy*gUM9o5U*)j)+J}0( zpj&IxV|rh01WsIg%qD?`+a{0j(AZ*9gBXJcG6Zkme|p;gH&|~_WJZ~f*4$+)45^0} z3{ZxqJv!1a6dcM_4(s&wFYr(O_^*w(H{`GznIQim&62W1SqH? zx9Z~9QF9K30q}Gvop2?+rYUa}BcDntssE*3hv_DH``z9OePX=ld5U%*qXI-)XwY?p zqoZ=z*V&t=D)IC_PGB3$%+tG1id#u`X(f0M9B@Oj*XR>_+xP_mGua2DbI0VHkjy4K zYzqrWpS7M|YP%s#=nT)ba9v?=^4|K|>(oE9`_5U}P}*4a&AsG^OJO-&G|3%>C8Cly z56q4@3LPAL2t~f-Mv7!z9YID+z9G-mm-28zlTRg*rYD=nl|WT z?b_H#ue4At?2&(50~90`hS5@!R}I$y2SVa|ti0#$7fQHb1eHu{5vf0k5Zb z{MXM^0=L=lGXp&6q{XWnv*aqw?5O7DNs?6WserU#Y%seKDF(Zr%Orn!aQ+NB|Fsn^ z1J9>*sgngl6%WDwZ|@#qyKb)@ua%8rdwu@`5~vX@tJk2?wy?C6cg4VMyb1y1dkRVf zPM97hDZD8V2OsyjK1pa;8h0!k#fJMI?(Z8&QLDILyX^v*a@b zU&6()MhLq1>aLc=A{AEoDb}PxBgD42;zCfSr`!4c!=wHr)w>TR4rdOhGUz}1sA2u6 zP||)x)??kbt!Of&{N%@DYL^Jj9{THw6|2;|RBx@cD=tJQR2JWP&c+Sp_ICDTx`^wAg z<#6b>k{yED?N8=yP?z|@k68ZFOK7B_c^m918SY`80}tIf=I>~=<`}&FWs0{8z~d0$ z(4u~yQ`okbYQ3^w8k+)whF{m7U4E?3ai4uIcYeDXTF4(JC+PJ7y?lzK=9i1 zA#Efv2}N?wwZcP*)FZ9E+%Y%8U7Zr`%ig)?RUs0z^%Spwo4%;~G?p?Q&;d>4FgfDm zQwxdvMVU7hpILZP2~F5Qu5DB&d!BA%g{8^Hq-59BWTd14>CU~*GXC-kY-{cXI_Q{v zrB`VFM*DnM0s+z>r7=3l0-s9G2gt*6{+Cf)7O=o|44f$ov9Wu%di(Uz=f`{4MjO5$ zsOQ?xWBkRDiWNVsfZqyKrS`R>%)k8LhTIUO+s10O0! zA&n+MS5^&K4WFqIsq0ZAG=*`q;d)BWfY$a1&cs~`v~Nj8^6Pb;P}?U$m8Szk5a1ft zm&MnY%F=k43lw@)E+)1>ad`HsoKj%fnA;h9H8T?kS_DDcVq%MaM)y;Q-|Obq1K@Sg zpOmXUM++F6WZ0k6ehfOfbLz_UKJQB4W`gui+@VAaghVxk=`Mxi1gP+R0T#|7kj;VJ zX>9ri1b-4MM=?`4u+#Z%?3x4CMoIAAT40)m0NFjP)l5%MPw;8bU>g(O^F6kdCb4xJ zH&AJT$yxaNzVkolU{rEv2;7Zrm%Xe^C}T|O&`xL}+cQ*gGlx=-)J{ZbBBZ(+_DW2- zShwT*hieS68_j0*n@yvtl7GI)V}fxo$NK_>7kP=tI5>Ra<$ zaJsF}=1rE4(A(t3C>Dq^t5 z(>zk!&-PAtF$8?=4_cQp6Bv9UpLc7UVUNGvTrz0TcX0X4{kNvv>-5?y^v^gN?B02I z;XJ8v7QJuE28_&o-)ZZx>gaWuUghFKlah2+wf$Su&fPOdIV)$pW7*U_*PpQhzBu1!P_w)$4-~3InX^GV8KAcIyF~v@nO6?MUh% z#dFjIlc1(VD`so@EqX*3@wMU*SoI}j`YIT{m2+_I?=}Cs;%oAjfs(#)s7}s7pN1hh zJjk_okAEyF6#%s67X$otqcxTyLyD>CmMsg1g3M$5I!+k`9FD(4-z;{t$tUyu=7}}LjL2hz>+50OI!RthtMkYRc zUnp#C9Y>X7ByUTdmTSLH*gj~L1L)n#l)n|mTuW;a=VEoz51aCdl*H2{w`bzK(pBPd zMAgs~R7%#+ ziF;PA!Nt0;p5!jY%-xUo|K9nCmGEAEoY?j_sQ5mXXS-W|pOgAF*!Lrg=Nf<(@?Qb- zv3By%SIR3YgeGrix5bBhk!iZ&qms7ysSC+rxL~|fzA5IK^H$X=)3zbJw=EG~;?*^w zgQXYiF6PUZ_r6g>6}qeTgle8JpdDQ?FSJb8z6+9Q@+A==TJwMBR5D=RYMRe$7W$S9 z)!!ME6Or`hH_d1P5E@=QTLUwtFh)~QUt|nX#bM2{rDO{$=eVv@6`Tt~HG@pxZ5PJu z;t(m1yM*5ZSVQl=w%zzE`u<96Uz|h(qqK&k=oZ##7Ev+gN6c=nVi(;NV8VZ1D>=q* zBdK+XP1EhP;0E3WvATuw*xnL&PucVad%6FRlUx7(_;$+hVp!In+B=UKvR6Arvb22J zdS5Kh$r;m#AJ4x$8{X&i&u*Fe`9liK32~{^1s)$lPuc3fH~^DeUDf!oAACA5qZ$E~ zZ6R*XvapCV9iOZ1jzIY95Akk(Ht#4X-kJRSF*)O_b6>D73(?4}2X-$CXp;(Fb3r;j<_U%Iln|z(3_bq~zl3J#`L?o|UE!wbypF8t0 z4in2f>$!Gw>Ti+A$*5ieupr0A-rG3CDY-e~l){y4SmvxWKw<*c0M<(S?9~_rN~ra@ z7BE6k%e3<&Pv-z0E}s;_g6@5>wmphqz+qQIvZ=)5uFpZG5lteE_|a6%Dyg@l=xP4| zwaD44MyLBmJ1sh_LvuH6Yg4ok@VM{Zg(rN=%T8Sv_;(t;vVW?S%XBF}pC9mUO0<=W zzDoEiZTcfRV3$NbqX36Ua1;H$rE?4mv>%#oolPB&jHhtJCdmBUYd^}SyxGPMDW6y z)}s0U@j;lc%pzq;qRreQC9(D6piN{OF_|?dEd|*%0%H*&AwaFNib>Lv^S3}E_ zW>KMLN8OeyiIKUjKWLl$ zmx*osdZ^yiL7y(X)ZP_DA(UNLWl3@HM12n}fHXRGOpjXi1)*ZYW)lXq#~@>CpSoNd znj#(X6oU7APzLrooizM*X%YTw6UC;#hnmwXd~|$^=21$ouu@TaER2%y4_m$?*wx)X z4xL@{MvR=i2n8DFSU`%z_;LlTITerUL1P~wSO1B%M7J|Uj1 zc$!sCvLdPZM6mLCdVON$*c#%>lU9{!#T3D;>AyhrZnK zr-IzsI^9Jp^TtUZ%C0nSCh^uo{YKxTem0wdViy`1HLpznF_D;mnZlvJG4S;4oq@4# z7nG748F>v$5XiZvG%RE#nT$IFAp?>P*^5>PQ2{nn0y0K#O@%J+k?42uz^H!U|;fxI7TnvH`HhjNji@%Fz zLI6h~Um3&nb|M#Z`1q16E_kqCGdpIr&euvhOSv{_cp+6oEw&bWXpx%3RU{}2WK2Dx zKINV>PX&%*K=F|zIUgQ@7XoGy8N1S8KI~+JYrQf?*sDm`^1))?-(>^-sWJH-dQJm> zBNlr@U!N1GeE?COIpPIZ+iph#)sg_1Gheu7>f8mGV;_Kw|40n)>7M9EZ5;~#8<4lL z%_|Xq4O+lpNlc(YAnn^YRH5NeoFHL{D*b#G%7|V}O<*tJ(q%8x(IG(XR@R8RLi}9* z=HIg+bpfsjsD$c8zNSC}^?R~jW7UC#&%47Ht0}u!0uvNf;?Azofb};L-b~x}h(7fe z7-l2AuhoUF{Mt zF=5Jx;uo5@a=~#^O^W{ukS8uxOo22-hYv-qr~$Cm*?Bxz@P>`N?C0iBBu>) zZsnorbLs~I=JV;0F|5lz%ok5X*_p!$z2W3~y+1s{f(tAQS-}~A<-OojQchN)6>aaj z`v}FVJXhX+@$4JJZd%&!skjEWyAvp(;&U{QpoPMclzK{`9TuIz;$Yik0OnWcSKkY< zxMLqNUJi9IzL{`^pa&JK-lr5+>_(ss7P)*#mHAT;9s?t9q#OQZoxp)$ovjL8z`5tZ>ECCo0pl8CwJ@adCx_HMh-bRQc3F8L zuWec)jn5#?-lEp5N4j!4`=wzSO%1Z7vGW5tCF_ z?@lMb($GFh;MNymJpeLQ69is|3#wXCLMW&%+d=Ch%_ zB~R4(0&$MjL1B1&0~NMuNNf)@BOqp`tp zOdKd7*yPFc6mw&RSEurXnb{0@!OE>G9!EQ4q$$~^Ue&_hPQcV1(f-^G-%Ih$-w{AM zzPusN;z|=#>vVmvQ=IN?J!-Cqs;7jIEb4>VEHChhVxNnA373=HUPuNG4v{gm)HB_9 zA^(Yx)?t^Gy|A^XCvO42f{lcd1c|E>qqXuSMG{PsE+`>NQ8wRm*&_a$vN9ThjWZ&o z2#9GQ6vv5w@nPF{51JSoL6e%!uxXVZ^azmvvd0nUp(dUbL3E6khVAva&o$zagfX@w zOHf>E>v*DBH&Bj`*4e*4W`f8=a=I2lPy$g6lJ3{`QeVuQ+^2K6c8?Zcu}0|rOOgpra2N}fnrDbi~bZFj125C@Y!Q*y}`T>`1o#-)>w|xXlYc*)1DsS6@9YK4j^+PYEt0vbg^`(<1g!- zs-y)~3;#U91e-yNH{({5q0ZHBWqrD5#is3qYVbc8D{DUBkdyQLDSun>< zJ;7&hQ}-k4-Kcntsr+v`AtIt;6l0K`QI90wid{qnU|h!q<#pe@e3}hYRDsGE#V7PYe;O~aT-6e4kwdjTEWJl@L!nj=H zbf>BP@xPr`023B}3T_|~QvpOl@brc8UBBL-jVSYHYWR7;KS?)jbkE3PBK_!pLTf67 zvc1C51$=GD2v?%q2CChc=SHF|9EkhNB7%hG2=PK~M;yl|tN>O_EVQ?^ z=ysSN6&rDf)QnD2T&Qy4-+Z+OUBHbSTO?OfWeJDVaFfKGz6aaQj6+iDf|%h!{B@|K zSr|=z2VWx|;6K1AA?|?k_4iQh^(f|o(`#D9X}9WwJaVpe%k%wEgXXv-9~*HzoxUW8 zDE_rZ+>zes#v7%F@3sGU1T>)_(U7(VXgtmlp-@kCshhokT=#hOTg8^JuoOvsV{RQi zWvfo8RagBb57(_S!}D1`A=e*l!-AF_#M05^NkFo}np5Loi@bLI+x`O!iyyMF%;7($ zjM4;IiKTx0llgZm+mF7?DvwW+#;ekBKg|<+l|HUjYCOls*j7Rz;}bcbQdI|iPM&6% z;ekJ5aL^)CMbkeeYk^QkDeY7ZH1wJ;?|3QLW|v~{S>Ti(7QDL{D(~@b=o=j^9hn-C zgEQd3=&HT}OOmsX`&#p5wF!Y<+M|w^7 zB;mQbt;|}`K|PYC7qUGK=Wp)?A_#duQq)^_8=IG0#Mn4F)#tNc_yrhnKNooI0e-St zN-8z=L#a`OKdrI#y={yp1_}&XJf7L;5W>wGHhO!7(wo*+1&bUjC zh$I?ERDPc(trWg)m@jbxkhetn$)}fD$4c&%8%#4DK!ceYuYXRzvksOT9pouL{X)B6 zc)!Q_H=gTmB8EI8@e zKp%;eB@@h-kaURsNZ0dwx;ikcE~S&uw*3?gc%& z)#Pl-G2Bdd1q%bL+1mPsAsbCha2y=Dv{h7ckl|Xl@q&x>x^lvey7Th4pNK!KsqIp{ zc(~HT1Kc8INKhTruu2<+@z7csCo=&0<&E?Y3b$m57FRw>A^=aT^An+%u?Y_mBLm}$ z7eaitp3mAU@6dhT-`6zBsJcrxb^0m=wVs~NMj)Yap0eQ5f%xSWtwY)i`E|h|Q9NSS zrzqs1pKL*Dy-s230=3~mO*KVfO%0q&H_?8oL5_#`H7wWzjm9hwU=r_u2Eo9i5_rtP z`7h42H6aKNTN?>ql>20^N1-*;Cx=t!Y5IMSYY`Uwp4*>{P^_cMuQCbx z?^%3#BV+j>fr1Nlw0jn-byhDIEh;_HBsS)+^^em{X9E<=e zB=8#f<5|pVVUo(yPN zl3w1Hzn|A2M*{!=8nz^4JMO$lzp~>-(ufTA@Mq$lLU+Cp+s0=MmuYFuCWK~?3S6_k zq*uk9Vkx;}GXpgJ_B2*z-BP6Y)C0axt38ts*!bplSQ@iS*6*obWap8_E$)Bd1Pm8t z&f0$f=36rc)FqLUwbDOQFN^V=5!EYyKQq7a6qTPOU=jO4)9w+RoK9H2Xv&C_p7Q;u zpimFX)sgF+agaL84yF<_vz*S(VC(C#H(4xmXI{Fcf$|S2PM@YK`T3KolpdI(vU(4e zi3gKB(cspr0ZWI>Byh@;U5YsJj*ZghAd=Eg`z?*wxrh1$gSq5LwEl_HubbNhdXWy( z>-Pj3Z=?YUAVl0QdT-1M)8H%cd+gik_*EHF{LzvG&}yiDk!GdXEUD?lWW{cF>vX>i zP3sa{+H4Bg=jhEKnWoNP%Js_zZ-3IuB2B-12wtZ0;%@Z!efM*{I7W~LDhlm?ibRI!xQg=?2R z_L!h*!C=}ttdA=Vk0uuu6!5QTxV2tItZE~6tU<6-h=S}YRGI@lRCfUjXmsFbGZc4l zaA5uYCOheQ%53oYM!&P+P1>h}91RJ)pVF*)Yi*HL+wO(#2JqY5If^X^K-Ptl@x^v# zabc|*>26)2rH``s&5ouBN*I}9V8VdeA-HYX$_rqGotmU z=_RHfP-eml{W-}Fh>Y0g^S^lL(oSDZU1cN*7mfKDE)S^4u}@BnR!D+kHyV?PH&7z8 ziRKf@gVrsrt*xkQvX}qC8a07#F%;fYS$n{`#ZU%xs@U%_`>ly>n34&|jwSkU`iJQ#tw1Nd6v&elthS{q>LF7XjuPP99pGBtqgOIucnY+oKR1)MvnzvExihiMsbZ-%z`rmWbjZ$vI1C|wvLvveipEtKO*Ek zYf1rQH~D;K@rXb|dEVN_3QJ7(iiNL5 z>Yjo^gAb4Z=Q@JGv-3ea#NtcTbg%bAWde5v#EK@uN?dzwb*}jqAJU;UhmhEj^cR&! z3Zw;?Q-gWprZ3ttS{qhXnQRsv5dtB?IL_`YYzsNBVb_<#5C=W*H=zLzfd7Z5vka?p z>$7P^U7K=j>mFjy4 znUz^FH-E@HYajz23Anj~HW=IFMCXYhZ;goLea0$9U=t!bN;RS7!)>|av;C~E;mx#GlTbk_#?MlsKuwMJkfN#q2ua@NoB7-vKw^|gv1Z6WEA78q`y zhaMXKv`rrOra|?As9CU>%JEyFgtNycO%}Ax>jA4r0Wx(Ah^ECny}dl<#TxcG6#R4K zA1{hC;X*4)FhBhpqU#7o&O&>oHRo~eEz;YW?pbj{yDZ8;-OI?xs8?^F3CnUqKsjpI zz!*RMZ-3$?DFF$+$R+c`@^&?MA1u&GJtjrw@ec>V&PonA7NvU7Y;o!wl1+|T?~ z7)7_|390KQfODh*sh-geAnHy7*%#^1g7f!=)cSuPRu2SSHjxCjJiJjs_s{q`Oi#U$?N#5 z2KC0dv2`5A)cCjX09V{5CMN#gQsBPt6}>#3YLt^=7#_$hvbxw6P*r_T-IZbzb=1gw zsb{GVdndtLu?q{I{l{_%%R2l5lxgfYS1?v7khp^_hy@h^uvmhd9&h(9${c)O!Nl3+ zZMW7Bk5}#3!VhZz<|#Mv-z3GkjAKehkpQZIrWZy;pvai3r6-wZT%4*EllU^c2SExf zJt>=I*=S&`+;F7n?KS-R)q1AVWHgN{_YLnmHzv52JjZ#+0+}4SAmAgx*r@>UYjd*R ztJp=8A9ufsEdCRg)ADed+^!@?nY&ggPx&vxqlSu}Gy!~iiv~}-`xmiokw-k%7_l8z zIB)eh-Y6yXwJN4vidjuLapzVxSFdpc%}img6!{h_Ibo6HNo(QSQQouu$Wb>q`qZCB z6jgF~HzdD8-0}<891GM>#qJc$x=hgJgc|oa|CD?wsH?kP982MQ1+&3Hs_YC^#%v8! zUylOH9s`NO#3~Fk1Q4iifuE~d1y}$J`X#BbMbFiMC6Lp~ss%ucTY-N=(IMD9tW`GW zxNgDzRS^J&Nx>wAAJJ`WY@|(QLnr|GU)O5dJtd%mL`Fi%^tY_?g($cWIpMeC)r@Msszll z@`Pv1u$x*pX1us|nuni<$SX3>($}KABhnKqM0(dB`KX+Ix43;?qU!q+Mu&TU*toq5 z{HN2aoxzD(J@J6Ag~xC|sOpyo^Az7sB)1i0Kc0V4e7cq)=t9BrPbpE8IYVWg0#Fq5 z#AXvTWh$sR$`CnJ1+l%wI~oXq$HOqTiCbhV(~2Z_|N+giiGo6(hCE4N@T#MfH+i zWYDP(fZg-*pB;6ZnA4#6_Pr9>ztwUL)MIT+K!j^BXw;?v5LJAM++pg=6@CU@RAu63of$nSrUk^G3z z9~TSs9LfD5QehZOxm?=8KRX$E)VbrNS=23v*B;LA)>F_i4 zT~-P2m;ik$Op-$h6hHGIt{V86fxWeWxf7U~pONN;fI%*QX}0!?-*z?! z$XvgLKC=G1cIv_dG_-9}cB(|CFhJLN!X(#C2plW}IE4CSQKsp|CWn?BF7K9MFbtBz zV4_CQ=U!bji&w-JXBWNB0lGvduN~Sysqajq^@E%-5P9AgLwr1%!!WGg$ja(O{4lJ; z{8LO1I+Vrq3`kV2Q|c8lLeBG6UjSLkOOt1J0A@f1iGA%BQ@HDX0K__DBS-0Zf6||* zE}s+{q^qg9p)UXL{q9t&*oDVg$=f9*tpMv(2F)}+u*U}m70vDS$3~vo$%kdui{qN-`Hekt1 z{YNWxAAs6!pP0CYC3%H0o20D2`KD!_G%sElsPWI*Ss0CCEBtDu zak1<)YoZqoR565~;LWO6lfrUmv&P~aC)3%KU*LXwtip5c4w9dvRKQ2MQ<9aQ8@BZwEzxfod zX+Vrn7gujMg|h-48-rDX&Wh3FX)?6nLxt~w4X8_FB&60sg~$K{kb{4^S-V3$ZvfPZ zjw}c}b}-;~e!)lDTs*yuRBoEV_^`E+17h8Ab}PIh3ECB!T_`Z4Fp*kg2y0y zwWI)B=qJ&_(5Hp($6F^Ll#1`5Pb6%csG=831Q~tg)E*SQcK@v@+(fCOEeSVS@p)LqsQQSTd zsH*NpM7f6k_vg0an>?P4_)IzD-@z%ukCLo5p~*3N3?okHa7f>VG&qMRNw$I zA^TEI$@DeiH#u@CbM8tT+R8OjNHek`1n8QGw|d_kRQGSG5&DrMtn6E zF4;nuXtYuoY_6ndRnCRHPuh-ycoFzzf{Y<3qaph;1D5MwpVjO^x!ca2ap{EI4`YB^ zZ`NZSoEsjhy8|%zA3&ip0t__1Z45VhBtWd&DhB~T;bPX*2VjlwD2nsB`JFslZ5~~( z1s2FZtK@(v3QzrnlRFHzQ#9^L6E)I%vfoYUU&(WDOyEa7r)Xv>+`}Z@u3sr3OiTob z`HsxgY1(CeXriayHZNYaIi4A|4>L8=>t!d=Icc_A;Ab~%nj+h_>>C~#sWp7?A<>=q zZljGs1{tsw@Er+HjaX`SU%37NK#Dl9!hte|tw;i7U~1SrUYRGTQji^lb5kwRk-X zesTFZ5rI_1&koD@7>w0#6#>{IVi0Osbm?KzVKUXwxJ*UrfnG+>1(?i1%4_K4N)=y#DAk=?f4?y|Ljf*ZE&mU$im z33vTLZU3RS%}n|CE~?Q$0u-kMm-RUE#w=CYh)th|Gkoj;^j%_77LqFdf*A zBNc%o-Tl_Jaqp>)x-r&}H163RkJLloNGV0CJj?Wfef2nAof-TCq4aB5DI)zlJ; znbH!QO6Z&24$#sLJAY4?>4yutZ0T@(V5L%`Tw}zS*1=)_-y?>X2&NH66EG! zs#RB4g`RUIzB)Z!w0kKGL789~=(P72+vftWRv>)f+tz?1lV!V>hNbz zNE1$|jrWkZE;gHVbzWyQLXCXS(r#|abP^asa4-YkG(*s<^H(`*gl*AEn=UrOS$Fv~ ztZ_}-5s239SK544cTv!TUf9m64YMeSmH`( zVuK)Dz@Kaa2qY;Yp4Eog7(xGjL!D?GH_!M?^UClz;_1F^0J)%I?-CF#wUYs%f% zQ5)#J({7bS1^%d#XF}>{r^~$uUMUo(gx6YWhU)y4gZokU#qmW_fw5{k3yt^#qpo|z zc_nJk^K;`3pR9OnV2S`S1CY9lJzkrABd*5rSnUW*;kQfsh33Cg7*`pPlFUw5I|Gh?1wdyH78cl>O^9bpO8c7>bJ)t-z=yqy;1&)6oI|e>< z6svbsnsjc2?{#qo^i!dO0?$#}M0%AqkJ~I@BMv?cYw4R^lrQR_zUIt*-vBllV*B+v z`u$b=-(JwdQ(bgl@x4CF{%gzTD}J}n%Z4#=|6<6N5_S(hpz$~xeX#+;r3nGj>6s5O z`&G+HxM7tx`?>wjKREDbzW_;}5l{f0vS4k-B!vN(@1<_e0eAi%@CPcJt!J(>dxV zf8w7myo39UXn#paX~sljQ75!xYJAK%=0lRXStf%*ixoGa@eXfuwsoPoCQ?l=m!i8f zYIQU{uuR|3yb?z_(2FG6N7rCDu#?;X{uCFBB(;L1)5P##nO;j zPr2WQQ2yPG-u>;n&oVG;+($I&(|=Ku)=*{92|7eQy}i0-G|d#JLTd+>KdR?{x}LfM zr@(xBY89)Gm~!~O#sF62{avGyO&a2_IkkVznA+zfRbH~RYSnWw?8M?D){{|7drVaTbbU>QMhZhKnihXQTiPzv&VHDWXW zN6(w_@;vKB_lO%dvIf#H3rh$JD=3qH_NyE0FL~n3E`1?2x6mHq486J8$JhoZ|C7%5 zylIccu^a{sKdnPRT?@HT7dCr|g6^-Wp~2;KZmEcRO zBOJq)?uA^RdR;hkcf=4K&R=v`NV+jl2onjMDy)X!2jJ3UEyn%~XOc~Wj~AyBxV1FC z_n_hX<*f$3GK)@-Yq9~@)PSzFXwr<@2H*$$TF|@8gX#PJG2eX>kKdoQr+4YDfx)J` z4@Z;479)lrH#K2P%m{&O%T>zl|oVWU8rw%s-%3vYIrS9s`Q@YSx_v!eeb+HZr_w>8w&@=_Y!*055r6o z;sLBY+vS!yYjd!N_kg@kTVzS2=PkFp-^xr9^ys3aba%*6qJ1BxaEWMnGeczC{dN^> zi2k7cR~K8dXJ~qzy>!EY(h6*ZN5eOvYq*!OBUq=UQPGE3vd$*P%C&PE8q~KJFi>#e z35YGb(umwUuHK&oI`|IDlFWPvxeWX`l_S|f`tAww+Z`c%ERt1>kUT#rYa;0H@)g;H z8C1kp0D{ozrsx%p@^2}Kj63OsN5-lp>?F9k=k&u2?P~_nDsw8K2E@IxT6izqL@&k( zG*@5+@Zun>M*Dt-D^FpVzvCsYGxqRs*a|qpa4qEOJ_tqv)0e<4~}loM8>v7?ggm4g*M{H-vG@L2~G@RFb(Pg zP{e(q2iJc#1{7ME@Q_HZ4E#4$8cxxqa~63?4))1U`QC1Oq*F*!D=2VqN<}b9{Rurq zxoi6M+2q=_Xb=P+a=hi2xy&j2>R@C%!7#&kTss*ssp`IjL~0}Ab-v7P^3<^h_wYnh z)pZGlS=G&{kG4AIyNy&6v$?yt+eUsv@9r|7YY@Yxcx5P$b3gpT0M_BBT*&n}EiyW| zVIx|JK*PKA?HhY-pD^!h+AZQ$=F|2{i2J1+!5J;oT1|3A@M}qXdV1#f4yj3~e?E$Q zF7(m|!~dET(NE=e1fir^`5Fm2?lj&eg^+*fmUAFgyJweIG!dPzbA+nUJi1@oGwO^t z8DMX5IqTUVNF#RLD6e+?){V-_B6~Z9v^4d5-xs~?ywHkH@$(~Fi}HP@6n=Pxoj*%+ zUXJ7_i&(YUI0c9Nxn`qvh_a1_%>i}`0hGc4iw4AsK3uQ0*#b$ z^(|5F55I?55LjCln_Wp1#fgBarN_lD?Xna5^f>`x_e0x@E|53&O=}mW83l~wN=Ss= zN)3y+f9CY(nGo&Y9vNwB$Xh;vuJIbNO-en2(2X?^y+Y{+lH}Hb>GXTW1*iz(IdF<- zd33hB*4IsKKKXz=AV+Wq+}IZXD_GvRMFzXI%<+1plQCUrw8_J=kvIhYJ;bFV;xHd5 zKS+3DsZ}De3?Y0S8d*~0eJPR{$YrzD7InieFyWXwE}YC!FiDUTL34%RFflNMY=+E$xWALh?8hmsM}QPs z=>f&6wJ=GwX=6F$-h6&7R{dlx2t#l+nzC<_G}{d<8I^JYi^xpvQ(PK48BLpt#)fz@yQG4{GBf~ zWoW8RiRsX0Xsa!QzX_BLyztlNv(6LFZJN}XdM}8;S0y^Ko;6$B-%9*cXK+X*l%Q;O z&T)HNq7T2L{ASDE#UGmCSxOS2f^_SLPvVig$1)HXPEh+*7neM5&V~Lsp0D0^&U9Ni zL;U0_-}~y&`;tCyqw_~re+mFBNt&4CW_DwfaCMArd{+*6p{_n9OP7#MZL*bstMmOC z9y+m{@r`rJ9t1}#{YO=w&&;=j^2qDtyZk)5E(-0Wwd;)+tDA{0^7ZS>i_59(IW!$SyYdcK#WAbe=LXGo zp^h*Z3A?k{1RhtrZCR;&3*0ls-n?0>{Z#;Gf$>xb6M5as;|CF4?N9ga0yr(9FY3KYq{s3xCkjo>Sx=4Sl>Lx_*qD4h7;2>_-~)p5tEEtbq$^d`fJ;OUUDK zM=BC`md_+uIG6_VeiSiA7*0UJxNLE z|Gmh0k3VgP?UG^hSTqpvOh*0_eYrC--tY`&6co*kT||j|I)!%_V@(d?@LJ^W#ng49()k$>|^_HFBE56QW1mIXIq=+ zN64Rz__b5`-X-ODUpb$7P88Q_J8&7@=ppz`@$unCM3zWZyp`QqYMwQS=eyGxdu29q zsSc2=OQ-W9WY|~VLsp(+Sqp)^Ihz9`A4j3vSf3wV3_xF}?;AYNENtZRxgEhQv2M0Z zyW;K6bXoeZq3ajUMlO|Mn1?l%6ZzcnLZs3ABxWQ2&D_i@Rwz&8BjXEL4fA=NW52BkHvK+V#JYG{ z2Z=}yU+^*WphrzhSRmkH4Y0Fxp{63r3gp06GywsD=}pw<**ZS-;Jd4y;?5;p{Vdc? zAIwj`8$$6W0?NamvUUf|mP>rn3ya*6wCL=U-pIO9>?C3ePZp@5p$|aDLeL=}s;?-2 zoPGU_l*jxr7Of)K1X{q0F%dOQh4p2=MVD&|p$%fBWT!}GO0=msbFfgq0et*5hJ=l0 z*C+A*w#kL+6Gv_e%f45JPO3CnDzeM zp1f}{5M{0O;PZ_L&-E-Od272EhDUI@wR_Do8U$c_J6VU{0*`ydM}wyRSiUVN9uC3Y zb%ZG1R{RLBh!19dm%h+ISCeCj36;ZH3m6Ha%Sg~w3xU7&C0dgSyzzb8W=dH1&Y>&k ztrN*Fak#3_`T_iYia}9}Ce_|J*(C?})a26`Iyw&OKf)a(CM{6NT^}1?3 zrKq)^+JCW6QYBO97g`#_@3~~ znm>}<#7D-3wjXUqBFfR6d}NNFO$}Iavcspuwr zIiUOK=Xkfu-x;h(MQ)kClj7LC%1vF5npL}f*>MomA7f{ozUq+c+xTm%-B~8dPWt_c zP>&W=Ki~L0l#tF;Lyd3&-~2>cok$pmP(kLqXIyHD;}^>;Z!Z4|-S8`KTl$c1LxD?d z#`NK@>eJsed&VC?NLJ8ygysy2(#*BsJcn&Xs(CS-No%r;8h$-E@_HY>J6w(4G;0w_ znVIEkP@J^t_a;m4pt?5>;g~)D!JJC*ep@} zk=v%!yIGTjVasOAXOHf2E6)#uiJ!l!8~Y^s6ac zDrpycS&p!o66wo7ll{9s;PN5-*8jnELH8Iis%;W<9sA=bFH`ma$h#Nth*vqry^fD= zfu4^AEKK$PUzj;z!w+=LxZvH@Lw^v9b_{C&F|1wMSyzgIai|rd-pYofnTY;rOE}-y zl=@4)8&&?b$@sn+x%oU;*2rP}TxS=ua!8fNaK5j~_l&m3o)m^DCI;9q&CB<5PWknXnELe<^2|Z|iWySAI*Q@sneJ`dPxzJG93!&3%)=r}2Wd6<~21 zb#7P@rBrrfs(PIvpvG4$O!1rKdyo2+H7!yg9LdS73CV1-%gFrEI?K5cCb>xeNz4}s zua+o6z_-=Jck#xXsZGt;DEKPt^TE@)7C2jvKGLV8tV%UTujzhTLAZA4Xv`5RFGP*c zlz27A__5Evr>wm1GcTaR#AxgxMPl{wY`^fHUWsu`Ulj?;LynHLTwX;*BNKdII>&ty z&*vF4xPqP4K_=o+NeLO=xuU;cmEZDz^RRcHGBiwIIlQ06?+{lBT+Ja{T)59$Uk3}i z6wf_Tv~X9%I)C$=EgI7QxiE#Bp7LE*Bko3@tGwR4LP5d%>L0(_^8Q?}2p+}@uKWel z#xhAqaKYveN(q9Ur(tOwcBB4BdU#agFx{R9kcc_~{=fb~d2S|RNq83Mve(tnY+Frv z`5eRrFG#$4P~HX04L%iDMnFQ7s9yS?({C(@j^gK04GJ^hk5@YPzTBU!MuOLLGnTQY zHKltMxT`>uTR*9rh}6WgZkbvqbCx$2gc%b7#7o}Gn?Z27DZ%~)L z_Bb*CDYedP_67Q_DHV26!=$mhq*m~m;QnoI{V(y7bycy4tMK^ADM@<9=bki|2M+!V zXb}Wx(wK<<#C>_J1|pa09s`z9jp!4CHB4c7=jL_E+v}4JJHED^+2bA#HWOB+W8uY5 zC@$k|tAk9B%M-;anxa>P_lCHXo@=N1=M}EpFQ$uo%mPpz~3hCYf2e^3($)foDy(wGw@h!7}n%~;LbU|k;(r}zuMgR8tSGrnWb ziI9^x(~%<10X>y?D~Ti0EjMZ!_Yb;<^u{yETNj*g&qnWQ<|&S9$oLPSHT&ukGtwD5 zp-F@mygWSnB4HMikA6SdS%mmA7EW~%lSMuZd0M-SsP=ZQ#dtx|Jl&@P-?Rt{E?za8 z^Y8|O8N3*OGrL?+N_@TK`5rRwN^#HT;5RPwAukuD4^fiAee|p)p0Z!#20%~55r|jj z7|_!GNwck0tf6tOH14YpR;E0`bekg7r*{pZe5$Je4ERG{-x4^DmZ3^4S#D#jUwSLS||#9q}; zBvQV55*o)L^QwDnPNWYIvku)$O6oqT>qj&>3U6KxxP*$xt@5)bCRtjY-?4W^uIEXW z>WdLK^FNnt-}J#yER%eVA&dPywJ<%khAWj(qnv*H{%3dhSkSp;B`Wfd{92ipQu@$6 zt&|nKcSMVIql7H(2WaO+xXC{QyL0Kg4EqfTK14JV;w<%i4|69 zjbsQ;y2))sei_e~f8q7CFPK25m(6$qc2Yplw# z)A4co7$!r9W(HpF94<9`$45DabO~V3vc5c{X$1%|Ej1!a@-)#|?igl@2JtmT#8gCK zei5fWe2r2h=#}9i+xyjLIvhWI5$mzDA~quxQP(KR$>cqiI?bf0`m1rijB1r9Z^QTM zN<^l{CLQv)Pd(u_%;1jCytiyJBXAtZ75*~+(5%0Vz?0>MB#}n=XFihu=-rHTV5c}2 z@4Cy9jR|sRfuuxrDrrr($%$MJ-4BBo!^!E}grzTjtf2N>@mI80Ts7%$H4ojOYzv7< zcH-mTKdc(OdT;9yqu4DN%1!m|)I*aQqYM4eewI5SOhN@dNEy+Cx_yjbXGhX>(#k7P z@YyrtI9wJ#FZ3B{>Y7)ge;8I{c%N}w;#}c-T8O1Vr<3R1Mz>6q+}!wqB23g~R~)b( z9nZ(*>j$sq#Fsa@S-`%UnaerQh_+TGFFhI*vbV;Kp-d$bx8^hJ?`JmcQmK{>f==D{ zQQ^syiZP1{3+K$fHk~;g%Il7N&dRbE*LEW=&>E7;LPRH;4YIV3QF^jgz0W_eQ60>U zdR_D~jpxR19I)RgTC(eJgl@#fvP2h(^40V+xy`~vZzLNs9DYvy6iDej7@!WJ9J_h= zEf%;S-2USJ@pd7jPB?apByfQ?yq(gSdT>MEnqeM`PtO{`KDwZx!6Uxy#{(U*|CjHg z){yPuX3E_w-F&|OEV~igxL*+xx`Xxvc1D3|OEy1Qix37Q$1U1PqIZj-uluhgmDh1z z%e(Xa`0X+xRO@RU;&{D4kG$qLE*>wnf_0&8O&T~nZ745K%C~jbkBh4wJ2Yf%JO4V^ zeAv3ullK9gPeGwra;1%rge!-vq=ZjFsg%E+V~Q=PLe=buzjwu0ikkF7(pqi`sRWnb zecV<-7=0|27a2$aGFl6+W#Jbq{2y${*A{dXq}dZ;L{n zJGOP(1-4T|FrwlxF-7;CJIz$ah^?a3lcSFVktEN?X@t|6j$4r=4#%(|j9in*Q0Nnv zQBKRwr^NVO4LEMpkY;#vQPoe39xQ`JJ@V9;F_ijoK|#M}8+qYxx#hk2!#zbv%CNjJ&YWeU z`fLLU#h+*H9IMp$m#Q#8Mmy6 z=2*K5_b*r>E{<3o{@HiVSp*ieCt&axMjt28WpBd{-LhGC3d9)=G>VX@d9GSp<&n8S zU;m0$5AE=8L&fG}^7iNZbsM|zJig;}ILGGmt7ws|pO-9pc%arjrhWRWP*}a{0G&5g zFKlzxnSq%x_gNGA%e)D-&s;da6s6Vi&9?e~eYUV#3C?w)q&s(f+<#%1!~fDSY*avx zevKw$RTWatf#rr)RM6Zn zlavpx5vywLYxg0Uh(}F(oZidX^E8w4Y0Wjp3ZTaSN8VLNGJquy=`!^E{aFA^^;;Rq zu)t*tkb(Kzf*V&Mi`-JlRUNtf$~R1sR3s-fnbasN9$NQ2ZC5~>UAF}H9HgXrJAYTr z7V{P(Iiz7xyTu*>#^8{pXz5GwUYua)Dtp8W*7XZ&3A8-o8x?X)0dQZ zbCY>d<(N}b-|+tBOoGDiWrOYgDcZCX6Y{yhge$?m6GFmle`}n?HQAhErK4!VMUv;C z|Ki0CLC)Q4!HSM;n=_iN4q|{gNyj^#``w$DIu$Z&8w*1L3Ry$wBh9E^-YY*pV^nU_I~f>sr=bU zM=J5g+x<>#&lhJJqQ(00SL3}hJOzdCmf$d)XZHx$Rzx)T31 zyC2m_QO3p*Tv@6>-y8y;hgGP`BPQgk$}9B^5dkCX^>|0`*R8M~H+2kXxcH-R=~poz z?%Vf*vFbM1nj2uj#UP#)*O5c|#sIg(&v z8_J`3Nwq9wXe%-_T|MKE#n4jx6C{7^o%aD7d%Mq!7ybrds=~{XXXsX*k>sq@XIODn z|5nlkzub6$oh0igOM|a)rC;luZ;BzZFhZKNgbe|t*#(1a6C*fwoU>5{9J)>kebQpd zACWMFlo&5&Kq^@rAjz98HDR3Ze6YSGL0#+)UejF1t__1fo38 zwG+DNeGmKCGv6%f@kTymVRWt6GBmurST-|E#@k~8E=u^}v$H6mxfWYn%Y#8yOI}jK z=yo&o@}V4p@c8KLD4Z{Mp4@)l>lI?(cC%GCqYLBf_8s|M3A2fCOMBj)|C!F1{xqQP zhiX-B#(7q%TeCmNt~FPEVXB!gxEm}diDq$smF55FS!&e2?00iRR-JmXAqkJJX@2xIWqJE>Z9)`f%a8p(g_D~Prxr}z~R zhB>22chf}1NcQY)H|^c!-j-?`7rk9BW6}T}YBJ}a0y>(g3@K)iv`3OiCz|V)CALwl z7Muv4Y)J!bn!yn}5_6x*iZWcVV!>}TzBX}N88{U}n_#f?TOZx-4uPN1P0&EM!Df|* zTrOfM2SF&cJXyu=iyM~(l9Q}UO4(oBMgG!he{0dXRnVG>SmG-R96vgZ&V>)Lxm9ss z+WdnIUA1V^;~Ly9*$0j=IVR#4i;7PqSI3_8b{ReNjH1lgIqmueosOE}8 z7U=zJ8abTPH;-!TvZ?1nT9%nK>QkLikoDr+|xCm-Ll-r3<{5YouIz?Q5CB zb9vc&i76SzRKb^@Cj&z*Ed;Q?AW0qs%6f=;opk))f3K#3k(D4zH2-lGXBWtH$m7)y}{<;)wTV~nq|7*BaC%+WYTMvV-|Jp9W)S$Q3 z1p1OSvuzQyCy=fx7plwmlWG>lY5NyW z9`rgyhFst4k4Njifwn zVUCcK&Wr%ZUarG%3UyG2@=y$>-SpGUC z!zw{GGfmefNK^zwLisd_+CctX9#kq^;8ZCX6C?#8OjGit%Set{Ewt#9S}f{fF)9yc zp75lL3??(t{{otvmq&TW!&gxjM2~i(vM4sap_V+z-sKV~wDaWi&(UZRNaSL^MtDqd z1u|Aqpx0WeryDfl&9=w*|!EEtu=0$g|w4Ql^rV(Ps+?gwF_?r{4 zG5FXRb)5a@K>us3-9X57Rq#RQ`)4OJl4T$E3gu=$_Puy)4rD%+cGbu>EGu?7Z8dR% zH20l}+LX7K2}W0j7RPrSq|WG{HnYjlJl1_V0JD{ODp?!+h9xUu0uJsK+-}ah-G*m+;Z1ka2h@}hWR3L2@56)wntBR6M%rPf)y2BB40g#0X^6V4BSSi9 z?_vbaqQ{D~?wk!{B=y?;T4NlJ1@LGb&0mOVOA%{!DvQ3L9Wy9;mI$` zD96xSVr8o~EFv@)%9+lKaV7c8Dn>s3wA=gkz~mNXyHn!}#WX5is9}$f6TGHsacJ23 z1Kz(9-iXQ|FZ(y4q~`Xoy-|5^E#7{AhsE$b6lV>Ek;FQ2Kbr`DV1F4nefbK~JivVZxXd#3xV zuC8pR6W<6NZQERA_xC>?jrGak9tC=jI)$`hF4vGWX7s}J=#Cj0FkFn7mTd2riokm1 z+L(&*8?I0|^Km%ucvoZUwYKxNB@tI}SXWvRUoM^$yQA@MQxFEUn~Zt z5W&OURsDy~|DL-W0%jmVB-%t^*TtbwZ!QThdh81ztLHWS6K9gDr|^XfxRSp9bm*B~ z`}NIGyA#M}DTOoCxq$i(7b-8Qwa2p@yqHgQs>yC2%-;$?>m3kxs($f$hHa;U3D`b)e|{g$^FvtJz~lI;Ney*Rsm z-AJ1S8S=x$>E-F~;S*q}AO?8-YF`H(4^iWeC*QoQ)zXgA|5I|-vOPl!?82aNK&`Y9 zOK6^*3~<+(q|w14%56-5EuNIaCByl_p2&_`tBw?IiJ5Y^SV`G*h63-ULkl)iAe16( zU|Fnsc?p6*`@sab3gp-!y%`1EZ}^&bKl(tLSQrSp(a(+j^&M;;^?RTHC;M|lV0$%u zX0YkFCPXZ5FzBX-3RHhIy(3bI$Z#jEqrQ=8foQbzg!?QFR``|9g?lUt`NoQo`0F}_ z!p;Tc5KOnjQo|e+|5?g-plH3cTc|t80V1s-Gf8eudCReUu>nbZlj7Xm_&)J8{M7f= zz=;g9O*_eM?~9Gws2-JD;{A!K-+gfL^K0Xisg7R%$oLBM=|M@})#SRj-&6;LJK5ds z!Ft>J4y~cEnw<;r>N}Q@x&O6XiuTZeqAi4<8@%cB@kBPGpp40iMbI&FZMzHVVQzlHx1i6&j)b( z>JLIhxBbr14{igc3YUk?N9!O5u@mO8d){{4h2}^=JXhveJ+~W{UKZDt`tNs`N7>;K zSxqtfQbQ|w9wmy07d{rwv^kCRoj8pZ4$e&Xh(TT}Gb*jjT_E4*H*dxu3gURNL4NX* zZukW_GvAVh(4y+#o6)LglYwdKcm5wxk1AoC+}lf=Xm^3!1>J{)8C{W)i;34W6rnv+ zTvpZvsYjrw>;k6FaG=~F1}~4$m8|KHORUNl&A9S+0FEcWp}gq z0Ga9hUE&0r%DFYdbYPbs>&0((=xw;5ofkmq>bUH=y#}Jbtcc(X{6K`ZDDT(8kP@ON3-RuLc1|K?@XA|bj#YzE@ z=ZD0^V6g5d&`kY77?~5$YwG77G;U|rd>{wUGLLW#*4eugK!j+`YA7)fz%eVQg>(aK zV^k0i`oEIeajYEaFiM11)dBbB5+*`wy>5Du7Ikw=BSMn_1%VRlPk7OlPqKjd59@59gZtY9mk0+z#^_?H3!I>T(3K;^uLrYiRhC+>m?#7os#yvFsG!JNkjH?SLl+ z*P`kStW#@he_&N9V@=qSaHB3*i_e6+@v@35I8?;^wC4#r(+75(!z{l+PwXkN;d@+I z^u*jJqW=kn-CysADJQZnZRJJSjQDUpo3?7LT3U?YITI#V(Y4E_14c}96whnnZ{Blo z#|;~CFcXT|2852D-j0wflt&Juh9`v*=>evHUASXG0GU+R(K*OJ^0+_Lj)jp zqyLWhKS94885yFkqFxIs%lr-kz1C+Qd%7b0gjH99m?K1)5*-DUalc%_%2&ki(Dqld z5feLBI}bk=Ns^(4TE609N(t@3XHu#xp^up_oZkKt_mhj5-8H1Dbqdj{ZG*=TNH&id z85nfVfQVNvj$)48LXwO;F)`6@{1oU^9f81Vfq$Q+B{P?Go(C`3r+KEZvm=c{Kkke6f{tW*#wihf6MUZeC|4lj?dq@3SK_w4(UNn71; z5?nM9Tsl8oZ||54%RKd{DN%uC|4Rd{L5=%$)7ngAy}1Vx(PpWNp{FF|Eu-wNW{HTG7+L%%7e_CM;}1dEN?F+w5y#HVvEvYtz3Gs>H~sEYy+7aI-yRQ~*STN!bzRTv zIqsD6=SU}x&6*zKTghWn_lcWPt47g($ZdUye*1Q$v@OsdX!UIJErqRR)$|9=GaRpR6{s3NeELG8HeJh|r#H zs(IRzZYf*uNiV%OhorVAgHxbH@jmL9LY+9vycRw&9QfJsr`nll4q!_%0@K9J`B!-|bniRB-=>;tAjue%R8;uInWNSy zJlJtPn919{yn0DN&l_aLT>R9q%uAIeE)*FRXX_0eb=Hllp&$({A{6&r0^zl@#Ra)#mI}{mw->l1JoQ3&Fkd)JSB+DK?ddL^@?66H!4U^)m;)NeHjLp4%R>}$5~DB0B=5WfbZ`vh?p!U>7x7=m zha*p9K#yZ9tNbR_zv7vM9Cg$N0dIaCQFNF&oSW-IELHBt&8GB+b+Yhe4L9K%Y~A@5 za-$vIctdSJv~{Z(#v~KW_Hlp~!FJ#to8?gx5QeUQ?{aw#%4Pl!>WlnWHz{uni!dM! z-u>^hGI(9F(LgR}>);l!YvsY7i;g{DT5SnPi@l-_(1_gX9G%-u;)0}OK<_3bVe#`0 zgEp$HNTF%tHGOi91J=HEB-7V<(l7wKAkeCRFETaxtK2=rHwD^IwyvCV*imc59_)^r zEZFTHyBJY%4W&F$#p}7;TP;0r`XQQ?bGtwLWsFSMh56Ow`2+;M(oSyilZB&4bir6vd6B(Ls-cX71B5S~|uuoybAHNFJ+u)06B-3J(-KZLZI8R8h+|ru< zD^dY6jg@)@geJcjIPZ!X7yP9 zr&t$fm0a&?5%z{vw%*|2rNNWN*Ti zEeR68d> zr25JRP4yn_-(OV4DM8jLms8k_$LWiZBLVF$dOWbQw*KUxTg3D4e;FJp!v1(CxPndjmqB(#Hxurn-)E21uo>kfH7c%F z@-*BFx}*#zqDQ~>?{-q&Z&)AxYNA_pH%&WSP(VNhtk>y9&0~Q$kO)(Kz-xa?0Wfoq z0evPEII?_1E&@J78$do?LHy!1ej%YMYQDdJHwyB;eyT$|$%XJbNziro^dI?tkgh#y zJsk;s4He8K6rl`cQ5wjH#io)Xg6KnuUK*kr^(46dXJ?$USIynqIW=EJLIebqw`NM`jCa~L z(%~-)dV71zJ=Y8{MkXV?$3S8~cQ5diHsGjXb{^D6N#N3BF%QJ`)r$Zl|B2m&2|1hc z7}E9uJb0A4W6h&~FXyrTe1{HQ0q#|kzKv;kjq&^qK&0IuefF0BqqAg#ygH?*@xn(p z2LhX6NO-kMXwXec&j=M_=l>2wn7yIeHJle0!AbMVsurB3#ZY4L5E;G>p=>KRu2$m5 z$-jQ-d@K2gv#^+leg+G}GX9MFdIa#gP}{VlCTcgJo3H-JuUI#;mw zOD_l77Rb@EJ9h0m@KWO1)8wlmKgXlAMjwK*#Ca|9RuW>ct?3!|%?LnRF1yBu0}ePz znEf&s_dD9P*iJCE0Kl|h+%C&mf3+LzHRHilKDHWmhmHEbBM_2-KYknXV*Gs`iV>Ih z5I>iu2|tW#UR5O+N(#5FyUUjc!4KbQyEqdJ_?$-t@DH94Kx!(H#cU{Q{t%uETEr{E zuk!I>M7D6;wFG1+p|lUcLAZs5$${yL3Ynk9Ke1hFLXT>;Pu|*fqsDR5Qo^SUxW0FT zzdMV>15<$sVhKDn8iz#+3-l?y$qoEgY1j^Q>r+gOVmK)ZOpFR+o=kMPc`d^)!PsgE z;%G5?VdvFuFi_du*Kh%Go$4Rrr=GNtf8Ob2`84wR^_`l3BT>{gm-s{n)wKh{`dGqx zou_OB50|g1jkOfc=W-h!H(EzFfPGd5z5^Aj-0+nfm1D+Lff(`Lky3OJhDDiy=L4Jg%*;pZlpSOr1R(*C`OjN%6&m4rivG6+Xo!azqcmb5;?N6CK z*u$b2PWY$_SD3v9U78uIJn?s-&knC?$M{Rd;Gex+0lom6QwFHkwFt# znXdOd23+Z*{P^utWPbe8@-mOK^e8jc8#yp){Q3Ik)x%CidpsF!c-^$bsosCQ4UTCV z{@!lTx|1C}_)ti4;G8mO?Rt(N$&@GHdh=0W`{ZRC-HtxWz!S_tix|3|OWjX;6XJ8A z%4Rh*^SpuyRVJWR{0)w6O()Xk^AW9HdC^2 zJZ^AV1TQe(s=gu^wC{1Pv^e)W8*F=mSE_~E1wDLH(UCQ{fpU!>L;!+BGIv1iSU+V zpXAt9kq`-DaklWZf~B|dw*Z-;g#B<)+2}yt;X5xF`3W%6(N<8RF`pop%`z|2JMBda z0qlr5Tc&rK!5038*dG984%w;xX{qoBAxFI6(48SnI2UEdRG#eHGe~N=X9qU;dAq7` z$uY)~!B@KhYLb|oY!~N`sTSH~bs%05oVjxVMo^@Q=+Ayer#Y@?TOyztz$)nP_oIp{ z$XzVhXZyyxE1+Ie{JyD z14As^hMYVS(m3bAFw>4K?yx!-&t4(+*728WQ;39IH&2k=6ja-bEcLAF1GU$yk9;13q|F(N<^QQ}a3|jxI z$8J=^4{*U~V9AH@y9!{ct6dPW0!4d1p>Lj#5LCH+$OSuf^2*l$Y@jgT`qAsjytJ6R zNVS`guCAy~buImq0O<8P3H}}C3F;Dw>Q-F>T?8}#SAvf{ee12@zdypFIcr#%4{Kfp zn-KfG2M@7rZQ-Ma8`Ja1`tBq;L*Kh@&I6p+J`)eVw4?^;Tn{QaVz3QBVC{g!x^bY( zK>P34huL$#=>W>QmWbb2j_g4IpiiJ5=aO#W4h84;6`){PVaXa@*r1?7h#lqn<} z3c2#$(GUofY3_FM=G$!o`=Ao4vG;cwSma-2WC&OOdkZ#jpn3Uv2!*axJnu4`JGD4( zuEGtd@j)j?bCx##Qx|m16IZAJlxJdji0iu-wT+nAYYkQBe;qRbq#r2gg{HZTQ)K(_=vI*OOOCoTV6#|Y zRm?~n+133NLdyaio&80C9x2s_ya7a6b+?`1*IGP?g}Ji_7hc5+_{xfv%C|ONhnJ zeH|R8sJu%xY7FU|qF0TAd?-Ny0Dn}O8ON?4Z{``P&obLw(@_wm5+1w{NAZI0_}m=^ ze8&zR!`2M4^V>~cP+4&l5>hZiu)zCUT@R4{x2S*aD@4==YFI$^K6+bub|~28eo&!L zb}lpjIPsD{xlI$fQzr5HyJa~Nv@rHldMtd$Ao0prDsHEu5)b@u5+oHe&*Dxd#8tv4 z_(%)FIk}`sGeL7KgTdhd;fC(Ym3dG<+JU|srXBhdwf3)Bs=C3A8s?H4^rVj6A_t=@ zMF{KH(BPy&Wz&1Emf#Y%FbeEGCifx*CbV;#4Z(h!*z5s}KC1$A9Sj+6PY#{tVnj{t zRJ@pNx{&)GFCMF|CjjBUB`QTs^X}cr*K)5uh{pYU55tJ5b(Z7S2=?XJ8K2<~ER%7T zD^YE*tMAa;(P;&ejqS=I_DEr)3ZoAZK2Y7f5C~T*)0aG$dIXCEsff6R#p4{IO!|

Jy@sfp2A^<);4}iJ~Yal)tk@@`mWS1b}D!kaN-a|mjdBTP~ zGFI%rPmELZ8kV-ko6;^e;7K35W(8-^n^6VkKA8^;amUA7TEj))$BPNUvhJqm5J6{w z{GFYR$>4XaoYJG@it2iC1;~d&g~>;^qp|^tL4b-4Elga1 zTH0-?PU)^YYjfw$!#5`rLxMuZZ?=hwjGqkV-ueu}fOo(!0o$0U9ud;qn*)~XJ1FS4 zxzdKsR^US(0Tu)p4@<&;-)FOnV8@NS34$>%C`8U0W-lnpnhw*(5myR?|5S0cVrsT6 zvXG1}T(fI98J$>eHyR1Uskon8?nB~&Vi<7C8gUeUH#W>9QsEgai5OuuUe-eyzeBn= z!{{c$sQi8b4vsckD&QS!vsqe-#YARFTga?NH2(eNVN~2&QLzN|Y$RlLbd1~1oW4|4 zSANNn87yJf#RP3s2Df@-N4W5~2Rt6b3I4j7D#?tF20+1r@PYrp8Tl zJ$%5W4$$ztIJV2B*Y^@LGMw>Sz)sTSRwKNR8N(CB7%Xf9lJX3b z!x)bpf(8Q!Jmm|AsUIiU+S>l#ZKbR7sdkMwy9@Z!IFwGpJZN-{@srjGqP?+yUWFr> zIddY{lmT}@zR=8NlE7)YzBUhL!<+aDXMdAC|2_6?bn!F|geoL_a@cyDGTxH*xj(tU z1ikQ0&`4dma-}AnvqM=uW`N&F`&JVL2m&`ySgJ6MPd;NY745iHdQa@%M9zvwO_+kS z?bGma7{%KcTk#|#b#P;weI=#w)eVN1DzUDHLIDWm3zn?)M=r@(3TqLM)>igr-pg2T zn)#Ga^VRR?e)=3RzQ}kv|KqI0uI9v>xo!5NV#f>oK#eld0GrXFlu=NQ!UmC9&vBk_ zVHKR*OjReS>7=CK&|}SJy_8WYp3ax^gwIZ^IpQ&@QCx425>)JO9^R}NQz_0 z&=ZxS*VK`+m@HK;Gj~Ds!tN^fC3W7ScdYbD4-FrE9>=W!-XX})QD3@oYw0PGL!E zw!u&$dOIBaU+AYXspD6kk$MQ(qt3p*iezyjA?th-!_n^M?$VZ*$^o-%PGurYG)s%A ztr2ed8N-24SH1K*3rz@IjFI0--j_y{@=pRP@F{k zm|lYZZ5QG}#c@k7XHtfG}0o~Oe5;B2iU5Bvu9&#J1!C=blKO} z*8yx2fYQoh{WjbD_WfHFNy-Z!;sG28dmsPbbA?hd;kJ9>TcM)%b0Zw+$eh;u<3rtU zG_hcS4Ycy(EC~{qS&9b+IK;%tk3X*Q1Hm;-zu(zogZGy8#xn_#hIR3&=X-!DBp^#sl7loUhi`%1&dWjxZ{anFgLVPs7HdQ z4K?8%D9T2Yvq*JSW^7#TAkR9dLEJO{MnnajcS!(yrMK|j zW>MCxUDKu+?yKB|C(A3OjV7v9O8heAAoWGNH1`0uAaXv<$&x-O(e3&6?VVwo)<_2) z;C(=Thb`x2;tOu_+iU{T7N(`e%?JM=_b?gFLVP<=mi;;Sr_sE&I$y4Dx1ex3 zEo%@CBO*ONwv3-L%|;CT%8F%Wm!o5pLfxX4z81?2KWa&~Ah_K69Rw(&KF5}GMIV9~ z%&{{4D2|W51V^&tY&J0*uUP~i&ESedGrsCpDF8I4A_f#)3n%h`KahOK6~j3ox0|)w z4hS>_bZ|n}QB!wvjL#S-zHMHj|1~g4%wq;d`4vR`w#LSpbqzh!8e`6gieV-k($VzP zX%agcp~+urxsA5Um)9AzO~Hl8)*{hO;qU5VGB^EL;kjbNg(Lwp_LsdkskaD&-MX{z zQm`VgzPB71^tPOijeMeB_I0MIA9DN+1R>2(O3;Z2POnenh-~^w#BV6}H(XUa3k@+g zc*}ANF3ot|UfJfoP`FG7HR-&jj!x+txmN1vVd_%5fov78CNVLwnJ;|*=F5fbA`!yq z%Aj}Q{rsQz+_M!JqQnzAQcnz@vr;*_=vs?t5mwB)eWL}wi0ho)x$wtH>-?H*Ns@2f z(2kRzSO&fT%k;7}@i9}xP0q!L@!Q*R(m{5a*q3J(b8VDnhALikODRx7pCdCn)}?IY zL}G|gr((*2%bF>C zvFL8cx<7xw4NbSUwI{ZY?AwUqXOLED<4m5lg8P;*PanQC=IiydItz3AA~7s^Me%oS zvBkE_4?f{$L5dB9_IMbgZO+voHT5am%ylLak`y{9}9;L?9VO>ny~Mha|ai3EUt}%1@Z5I=5l4ibIUdJUt<;(8q^%q_){5O*@7Mo z3aM7ccUS4Vh~?e%${uP(fF(hgEJcJyBB0e2n?2L+uGkIC=p}^vLCPT*h$&EtZn6e6rY%a?@w!S zckuh~fB%6q|my#Bk}hXp^@qqOZ;6Q6ajK!P=L|7_jI zV?k0y``I^_%i)HaQwI2G!Hts9tw-QPHbB(=@8srw)t$AS z+U~DYM#r>_P`=aA5|EVwxDv&_EIFeO^}wDHXSpa9?OLW_nEk^-H~l?eBdHy--rtoA ze^dkzvTfMtLx|X*pDMIPbv2gEj>@L1<@d$6s|~C~&FZUF5;DyfHBeF!^YzT){kLy< z^2OQ-bElp!;KLoKs~wZaVYS?V#pQ&kE6o4-kBvtyCu3q0rsm1y`uVJ9&oYnOE5HG6 zzJ)c;k;b&d?LPY~dl0>!xi)A6^^nZ?*cjazSUy`X6y`9f2Mw5#=B#Ia6|k6 zxlEv#koV0m`B?(Nhb_P4jXhRSGp&0{SdM&k{Il3BL~BKx14^a0Wg8R9QnU1K5va(JEU;U^_h4FfIZ~ork;0FPb_b?+Rm>7U%H?P4 zJvU$Nv?bkHSPTZVC_|(}{+4BGRa9(o!S~eDbCIBR#%^E|&C>7l9W}>} zT`_~di*hPyMXht^nQ5q`;WJ4#uHA($LVNWUtoHHbuS}@Wa`%O7tf}45HV?c#%QAR( zj2x08Tc&6~oI747Vz-}W#S^-=yV0!SJNkIS=&eTmyQn1a zA9u#Wy76nic`@&dBq%tKDX`K~f(`TjZz-UT&t8+^%=rNJ@y9fr8#e9@Stf6&a(`qO z#b+Ed&i3-d9={&uXX*r7G$@IFt&>jQ(`6mdx&)8I4WDMO%%5gAzD^~6e$0iif4OGB z|0n`JDe>{+M}*&z3*d`k1BUdKnZV!v82?+Jt)?Sjwaqa$*=Sl}LmchkegLzD^o7dG z$^t6?Rdjzw%oRZc@Y64k#ufkKixODS-e?=aPmJzA*bYaxrN~2n-V%A50&mH%klEwJ z=TeMh!;g{1y~X4LHdh~ugJht+>vDs!H-mhjlU0MA8-&bX0?N+?w%+mxH?Jv9V^OLh z;sb1+CR@%_PYf8*m$=WjHu`HnCB&7Vz)$NES#x9k&MoI#+lo88GQyZQWim-Wmm*uc z@KY{AvAOL0e06s){;G;_X&F4wxa{B{!T&BIT0y^Khe!RT^&vp)1ds@?Bbd!ja=w(!{<}eKG$=DICdlkm8!S!U56Lj!^ z9g}_n&gr|&OwllV$>;8{WRX(kG`BV&-_KS-3r=)RkG?yG-(tQf{31?MSESou8@t;X zS9&+so++(Q2zYUaR6C9?gfp180E^EUuqiq^+zlpVZ@GW(MmTWfX8xx1TcH`aw>nt+=? zT|xn~x3R1b7I?&cjbxD1UY?{>qOwY1Y=yXAfk_&mV-ghGAn>G0^SvbuRkr;meiY`B5-OO_4s${|g5q`M)@T*H!f1Qm$PWz;v5WwXiN zefGsB$N06kiuL!7R8&LxEjYRm`4YhGY}@}h*jH)wQu{14qCMV)jyH)+%2g_BQWZ9% zz+jDBotQxv18WEh}PDx78fTav8b9cKhDFTDjc+ zdIxkTtrYiSpC(prU$!;x^$AZpR^WejJw>8Lm_aklI6|!brs%;YxHsv`%UiJaB-x;y zX8IqNe%kLzZr;sut&~NY9x0`r$>5ZjYquuw78LYu0&YwN);mDTZItNctFZ2(m=xOv z>ay}_TntYTW1O5uIPQ>W4N%9LXDQ|FbfmbJNtzB0MH_#z*&5ir7^WWAM3aO+_&fR! zV)37+v|SX+EW3zRmPW1EsI?yvEYzo;p{(Da$LeiZbgQuWrJ(X&bV6Fv)}Np>;;*<% zX$NdWQk~iH_vL0DzUeHlghBhIOc>La7_nKp@%f3}) znDE!?yBy@FGlN=3AIgX|6Vk!wh;APKUkY{9G+U7t1JlcnKj6W1Bggwl940rW^M zzrD~DB3DMQ0`ER&!}aGz$-RV`VI4!iWATT!Dy=tS33pc79S_CGcQ587=7*tQM)Q|H( zCk-!>44WXWObkdg&=RC?-(btix`b66Rw4#G1#ISParo1+5?#4V@Cp)-099z=>$R+^ z5O#UO!>w5kz?b30fVr{Ek|b`K;`gv2i76?JmDoq(EPGV@8J3HOqzSUmoLdJ&w zy^y0WS{66bNYt{Ksokcj27P%G?coU}><5-hYMZLCE+^R0b%xt0+MRW^G3Sa@hfAuJ zK#|1ZSwGdYU_Qj!+Z*WVrvp_7Q%Nk3w04+wWx-Zz>qoM}0>_WC4064UnH zQuv8+T37$3zl_w|t=4rQPVq~}Q zWm)l_b6WMcGY#BX9@?^+z`L9h1&UI?n72mEkEP#_o_HS;TC}a_ZB=Yh*IHhhd|~sn zj7)SD;QM4w{`}w#lnA{n|C&=!Pu3@c?@L|+fsV}loK^tcED(CgB+dNdOQQ&1Av!MN z5pCM10YmF0CGgRZ#u)YiI_>X5;N=YV=HOz*7VQ59>-x%AjcOKtw zA3g}$|fRT6?Ua2p#;jv(-u`3r&O6^cOxGapDNuM>^=CNW8n z0kR*>G^+d~3biSPbK)5MDg*B8j$(VggqO>%f#vWawYLb}tmE=#BdXr)sGFlJV76hz zXEml|n&WHinG6z?A{AG!3nT2EVeb37ip>TI{RqX+Z^~01ly2De|7qD8 zCh4DzvOfQPY>uxl{q$HVUGWh7>=|YAmY;>I+0AQpkDqKDFDu*cla5I4RImHaRA00$ zf0H})K1-pWlrOnvQ!N0h6_jHrO%jNMbxd<%o!3pG@4iPn>Qr))2XwI!#)>zzrZCpV zM-bVYG63RXip1AXq2=tCa$!`}IYq7j=^i2K76}<99yfykGu_D=XQ^9ZqvK)_s@3py zk?w_gfW(gjBPs?|PtPlyhiQ&glLC%a$v zV(Ud(|6(w^2F{HH_cwWi=+EyWm)-@rJ=h_pF3p?E-okPCAZKkUAEwvfNp7~@=&ds$ zv1M@F((m5JNW6d?ojz*ng`Bq)?@ay>eRd=lu+nkOqV5>^Xs}9}%#-c(cw!vG(2GW zD|2GA*Cf{PZKkPDL}J23nW&tQ`8S4)j62S>4de&488}5dCq+DY8VORf#ZB;&mZg{W zdmVRx9cxFI*7e(3T2&c@faKsD7cN2lKi_FGcz;L8P6cbWPYhFf&{Ow})+f*JF$kj= zt-Ju2FpMop8z_}Rc~x&nh>7i+kvVuQJsTGmOIo{vB>rWBS~Umsa9NZ~5C(_^v0N_z zntoPOJO}s=@U6AMye@W?L@U@dzg6p5y~i&rVANb81D1EdCDg0w->t4rV+ns$J`Fsz zh9AZ}g&4^6#F)PC>ds|l0@yWx)5RNo$E*P-lu-UY z2YIjriGH`0H*&r$myz*Q zc{PSn#pQow<)6e)vT5iKv90S@U}RcDg?zmk)E=Nb989YlMkP!ttl!|>kn6zrk&bSC z(h_m?g=7Qy<1#H+jw18*LZsIkfyd^0Y-p}8X6yQ}b#3HRr5?`>zr*%Iy*!A=itZ#= zqVBo1Cqy3qbf*m&8xqECiHWx8uAlfxilVnLGBR?#_8(6b!i`d?m5(R|u~45g(IU_x zs{{KTICN+q^|0jY{M@{6b)9~S3p_Im>!cTT@zXTd%&m9#!{ZLhp|GQs{Ng4P;k!5c zU=q7Gd{?SfK+*=!bXMmjEMa`-^nJpX7~Za{b$;2m3Ax`hw{9ftmdZwqimmodIMf8g z^Gl5r@;kiufO;2kjILkH4Hh-35^Xqsc{b7~u{AU!TjY953Ku`O@1pU$Jp}U-5G*@- z3OX$?Vkg%*HaZgCyV;k|h95z`K7sEvHm4O5u#U)r-O+n7jq5RS&==-L_6-m5Hw%H`?6D3K#^2E20r9sbQ zlbe5u8vR62%9HS=3dg~bm>u2=q5t>}aqaTxn%@@pe_huE1O&jS=;1!X^5pENaLR|r zr3Q-tF0Wls$x%?QHoDaR-a_p|MkUi)o>{l)E-SP75q~R7*3K0jW4{y^|VVuhM1z6xPG{{hUUhXMT15utoPXbBeP^ZIS1bxg|FT2!#=`@CPD_IS+In z=frXp*j_h1mD^vzW(|-5uKzkQ&Lrn1V>#Pn))`VPA9cbPS^#%K^M01-m|nZJT6b{R zmkSklPV%Tmr$Xev1O|%Rp7^>im)o=6oxq>t)494GX&~$xWHbBC$ZQnwZJLOaTs*(U z#7aRC$d;hJpd7;#rWpyt+dZi`S-ljYYgr)9$#AN^Bfu@=*P*RS28EYmnawxy$L0!E z@p_+frfPPXY~G-s)dpcY;D!+oM(e5X61-`$W8%o%7Y<9slUEZ7bUTR&nN?1_E8O$G zy(j{h26(0)M8I-UmvyoOY7|`dm-iQ1o9J%&Av>osa z592ttuIsUo3K|JK5nB!E#K_7ei<4hn53}b4>pVYw4FU?*DBz?w$#2Q=IP!T(&La!h zDiN*5v8hAYKQiE;vXPK^qD!Y-b}p6_Jmu8b$&NFJGErv0i~%$;LYV~wJ>By_yc=*E z$~N@S=W8o^B#Z!=7mou+J?R+HwnZ_{CEYjGk6nl~i3e|yz|-{==*Sy(v)eU@i$MFP2IDM@5R-h;_({y1&1f#1+HV>ySo;lc_GmEX zX+bUgx64}!ZL?0b-rlb#1JlRPQNgUEQdx=(Q$MC=Z_l__f!cxYAm7Z_KVf3|BIg$* z%`N66Vi<#|EGVpltcq-{H1!Z|^E( z^Y+F%ePA=Bm#yTKC+4Gc+i6oFaSf;Xbk9X3F_qXRwG1LpQemGqaWyXFhF|@H17A)R zUy@%0Ic|Xte!SAm`znvgK-hQo?2^a#28F&JC;&*sK^rCva zE~^uX1SuM@l)+NPfE+rKvQ_|?8h^mA?QA_nk~;E0`BL2d*TWajsA;Ji46%Qr-ENYT z2R!UR%c9k7!cvXnaR|(8I*)=xtozo*$u;GFptegrB9LPu2oe6w*RxGH-9D|JeEGfu z%v7Egii~#qnG0aKLJl7l@vj<@KQux|R-^nCW2?#6pN#5mnqOFKrLDa>O5REKt$hqt zXh9k2-w!5#FW+bqkB(-Wd7ytbyCW;FsRgO?|LKnQ#0$LY85f{<@JTQwr>vC8?HSdO zl^L$(^z!hAI%CwDj8VFVxYC%bH2i{xXN^EnLdXpyWH5rdaeep0Tgo zNaGI)((464V;^~|ZJk-Oke1zPUX)g=Q24qhh=uy|K$ul*xpdn@cH@+Nn%8XYn*BBIvOz3lEMF!=Nou5xH+ZRvg|`mQ=qS;iFjS~#kmNL%^DBU}$O<_&(!qI8u!DJ?kNPcN-EI zUYEHzGO6xsv+$!(cI^j2!dFi+>!sb`YdLbTROJBiV~Kpcq8@+8Nt}KEt?xJ37DX)1 zm30NDuE3|`!r6Y&?WX^BnVtP5%)qwVK>td)j?`mzF5NjY@)nSZdeHoGGX*5OB2OGsi!GF7efCnH;tf< zK$O@-2!V?EuA*W9RRAxSS`?8Mo8^w^eT}royQCnOu-;xFHX%WYfA62%Pt}8{gwGT^ zU8F7dDNWnT{EV^{Gga0{Ad3m(O41#B_+(Pbj&aEoc#>k50bdLMGDkz#c}Q@Ya7;IrNU|YAQ-)R)2t$$V-LOY@rVE&E**q<0-Ep6(yY`k@?^EDn7>jnbpqN2yl5mUz~pE8G@$ua(^1 zJu0^N_(O8LY*Cc*F$4rD^=hCLCMnFzCZ`h+EIgy-O5hKSHPnbc4}j!Z`e%^%7aWa^ z!dBW2%m$9^G5DV1aUiXBo`9>_3KQqiMZ>XirhCxnfL=CT%c_2~2Hi_WSFvFyI9=X@ z`}}Ihb06Y$zw`oz(Ba19r$N!gh%~|?4t<{dPR$|MojcM0LvVAsbi3jkM=L3Ve2$px2iKnHpA)hcbHV~}Z1>`0W`vFcC z%GUmn7zWW!;D3Ms0;)=}t(w7BI%1}~s-IcIm9j9*Pe6oBk@prc0Pq9H+C+8nIHL1W z8|N*Jf8V)8O{`IR@$kwl$2p4Q;#njxu*0 zoed?9JvmOxRM?-$jzHofm(TrM-*~Z|SqFA1j&ePjHPmY3hzmP#YePVr9Fe|WJ>+N8 z6!s{x^L3p0!|er2=^x@=HFo^e1NYQ?gYp`u9)3TCByx)Mds4H?eorM}YCxU=lKD+5 ztA$Ju3r2aLWvblTmFxy-Na*``yfyk1V1giQQZuGjHN~g3?S}%NlDJ+p-*m{_=Z0UCEq3-zktuAzx~+zbLx@0i&WapnDE%Z@0JGC>Jq4 zaDb>81EhTuz@6mdLT)a1iAgO5NaE}Owg^GV76DAaX@eK|nO&VsO)PKYQIBl_d11j~ z)}3Sh|C%%pqnsQXam|yKj$7f^R9&So2)zUGBfnFaTZdmNx4~3ZRfXF) zMc0_IKq#v-wOG~{%Qk|^xWtaHxCCUnT;>0EbF4(^ibEYtc*c;K#V7YfH!0GYc>u{Y zIywkg&+K~uH5Gi~6AQ!#5yPQ#dxzc$mzVFm61twNVsS1-tTYBzW-d+`LCTfD#jX^f z05W0$;Kq0Dnl*-b3^@jISL3}P1g*#5%KiO3#%>*;NxcMkT^q(GMSAx6kO=V0QD6~M z4c-A}9>_3aS||MSkCA`QoLXc@0=tM2C0!a*-Uh$XN^A(-Gn!NEuc1@0AN*U(FOJhl zj?Q&>f}8C=_lLwXyAZAGj4gY>=EC9X&q`DK*z7JdI+-Tq#3}DJGFK)v_of}MruC1N z#}^hM z7TYPRO~|+9*?xZW%MEg|-klqB?l@TtM5D;F0?(+KwOgt|j>HB@7OEPx9bUvE zbojo!?BmCt$3epvAVFNlQpqzPNW~O&d~Rn9M0Tq3y}@K#8&yAkW0sl5GkP~eH@yAd zEaHY`w4!~VGp6@`1oqdPeFvE$@iMA9Y(N#anssub!(odp85eT(DwgBb)Dm&~#I@GZ z5gLOr$;6%ZbCxmcmuxgwe}z#}OA7UoivjzJTJ)EDN*8qQXjFMC=c0z3=c>)3IwM;v zKQ&M$^0NZRxN*cH!vD-`35}jWUbjx8_;4@Fr3*M{@DfP9#H46Se>ow?_@!WIWS_U1 z&zx%2shhVQzi!3Nn!QF(N*unq!p5A*X1!GZw=uE2cb2EJ~y# zMcdR3Q`%SuNr1Cn#BU|;;6q$B+1%k6a+K{ZFJ!mUcIkY)AvTQP6dzkJ(Zn^1hD{yX-aZ~V#Vp(|82 zk-gWq2RJ~LgN$=|LsrShNe8h+gUkaB?H-}p=~6i${P^(v5%=U`HI@WPsrd#;QJ<2l z>s}Rm1BrO4cPpOV=ON3RXW`mMg_OImCg%F2XPufn`fgXg+W|+417oOS3XB;rQj-n5 z*oUp>R$y$MpnY>*L9O#0?Ek_5?P98vdT%aF?<(Ezez`RvQ(!YEHDbj5SM9Wdc#vkj zoEz7&`4EjYfns!Eo}#_+ZdEI!gi1TtcmT!h$aIEtizOPWO%=O-s%mb{8@AGdaC!+e z#g9u&f8mLNIG#MQXv^5E@GH)4Q98gt+VDz15AZf4xI zf_OwFOi%CS4TmM*I+m!<{9jk=qg(N*^yx3?PhXK#zWt`$FseRHV87+5A2j6PXZFkrvfKr=73PQJ#3~{ ziShQ-5=F4^{Wk!-n0;Va`5GGmk7HqxlHn2({l)|=wvh#^{%3MxCE+AK;Z$5<8MnuB z!XC|GyelTl!^{T9#TI}jrgeS1(s>>d^CGnb0fMa>k7oQZIUJw#o@`Fl0o``RvpGzi z#w0(2CzJP2fu>4CS%7@7NpPnI!0Ey7!j+NC7-yD;3+w;`xl!^We`&DtT4JKIS;wzj zGSZEqt{Wn1g$@RU=AxDOLqeET#l;8~7lbpdC?J3f(q(#0ILQ3Uf4vRzT5*AcuGAk- zn~%k%t)8Q1c3oGKO+!Eo@<;sIzr{=PVL!yEO^;lyBh@%B^{N*$z6U!oFF6F-Nzj>i zmzPGhikplnTjdGZdv%a7+%VOhW4GSD4%{u26ug8W-`8j8RD{$f^MTQI4JS|fAUfQlBx6C7g$`>PYxYgXa?t0>0w)pJg9mgAi;G7dOdwxE z9I*1?@*ZdoK#_C!$_!87T#v}c$?e)6dA1%TWcuD}%g06C)>ZWR`Ex-d{de!6 zb`*-|+cz684kvwpZ-BGs-l#>O^WH8~7VG+~sHmt0@I^2vF#@@%Oc@JGS)pRr8xxp@ zVoeV74E|ygGMo~G;^s1g;2k2Xvd(UZg;sX)L(_qc2PA?I`N_l*-KGp8d2C~1xX{BX zKPuMwWPBhH8zM-{`UHR2P5_tMWLOc{lKu(46sChmcLNuo7-0DLerA}ulcDnU9O&KR zx1KM{%WO67mB3}%*BM{Edi8%=r{D+lSRLuXxn$Wkg1VWM4^qyq#{_*8zY<3}sGd#G zGiJ$*r;$L=nc#hDe}{xqKuv;yYrKltY9{YjSI&}yO|cqTh(Xmc%B<0b>q>sQ3s$s6S&keEoCJ)@|qSFupxV2*3HkK_BzqYJmBcd)Rbw7c@r}x=^QQJ8G%@2 zpBG=JOR=xn?az>(lzEAX-9`RCw$3uF%B>66bf=_*lu`oHDcy~Nba!`3FS=2r8x%z8 z?v#>6Dcudy2-0!pa_{|}^W*&Y(o0zDo%0>zdG32}|LFi*=(yn?92wqYNmzsm#!x2v z10xDfla$V{bg5}-S{nJnAcOWZ#t)Pk2($8u59i%sm=67&pq4mm5t2?f%aj->l&TEk zUZdN4R@2?zzbkx5Ntyct-XfrVUn*}_04i_a7eLQoHx(iuQXc&`Yix*7vG8lni0J-^ zf*3^zuTGW_K(j6YRDr>okPTvK+g4miUEEcgZRWx9+!7VrHH22%tLAGAN6SlqFzQsL zJZZrK+bktglWDPdW5D`iEeZgtF zZMI0D`qHH3cN_S9@v`wxjE%_xWBNr8c-L6TT)w4m|5N6aPjb+LB9pDQQabusuEyBX zvM7%B@Mm#CS$zBN@6TOel6OGycoqQQS8Lx0RBQM0#Q%Q2+_k;C zTfjbb2p;qre!s;~(56%vXGnplW`QQNL?~5qau$P-8w|7q8i?U(C=x1PnY^(n(3H~G zm@Q|O6D88{3LJinP2iKkp<TFC^0cV8|%44Q5f1qPP9B9ZU;f^y>-U>7i8y$Fs1T<}GXKd9`VfA;LzQej*O zEnUkgD7N8%D`FS;rhqK}*6hHFT90EmIy(B!Yz+f|tJxqc^yO7n`v3sq6ypvqMkcuE z`o)^kOAiAn^vw1TcDbHa*kK6hmpz(txRinu_YScvIl{j62j>4wPX*a0iNrt@^!))X zU9>iS^Q5%=KB6MjGz;!&l`I;5-{qQ?3rD3xCVA<24DeTbAYPxrqu`>1W4lozCnX6~ zoaeBFTDx89J|^G>{G_xE5JL+^r|Wtx`*{?(C|*r#p~;N_T;71J57Q(48eb{%-&#?m zC$@)q$4u)ttUf%he0t6iE({#*)V5$!B4c#?_%Xpcx8mpJBT%Gj=yY1|6iai8i&`x= zP4XJkeN_^5%fl<^`hjf@r8NQ}(yT!u0(Hx|OLpuZ2f0tKX8EM@1k13bLd%1TQaWev zQfgL~{X;n79c7#EMSXbZ5L*it3`P*AnePHZ#fO?_N{Wg#Z+=?k_|O-DJGKzk>lO1e z7x36>NZHCS4`LPvtbjSxx!jn$RVS)BmF}zzW-WBxO5%ZRr#=ZSeBH_=Y57V}H9@DD z1FacWcrz|*IPRYg^BkZ(uxJTA{``vl)CqtINTf>_-{T#i7RAESTAcvTnE*}|O7ODn z;HrRD$G>X{`U|}vA!`ACkUy&IrrDhSD>&_lY3!fZuJfp<%hVCPDR>0iDi*&R;Ranr zMOQa?lmyP;&6pQKy)nVVnZwg_eswmagayxFt9}`gW$k{K%07{0_8sy+-Yo)9NU3$v zY{FLO$$GyT@cK?Yohg*X)f-p3omQhN4LTQkq>{xyUY@XL+{>*x(AU>z+8G4TYZ{i& z-`}rSRV$#JDB>}_nDvkXsCFrmH*Qs;w+E8*@}xaqMmf2JSVmAVrZXR}+PbB~f3Wv> zlfP*huKmVT%gJ?bCfX;C6olK5@h9%$0$6>~4;p z*B8XZ61A8!%EvH(DX7spsLR5hmmU>Gop+Cf2#Sh}|B_)qZ4rl8_J5i5P@a8e?Pf%v zYmRP!OSa4*!bkk(4*%$F`uMa5QEc=n3~0rW)>z%+7d8vK9yL)(WS^FilC1#AmA2odK*CKz<#kBO>$!yYRR zD~;<{Fma*AZk3`@Wgfcht;Et@7@Q1pVY^ub0RXHy}r@fQ_hoE#XJf`ynYOWriM`&2@zi+JL zCZAc5qWPL$)f0SXgaZty&)U>JXk$L5dyJ&)9UumT8*mHoUXB|ZjW>)=$99T}$?fe0 zpU_Kw>}4YP!>2w8C+-Jwdl}9kQq`BL5r$q1NioI2tL4gpP30d@YP?84A>##E}OWPjUY)ot6l|j8IBCj0@Fq zb}bgN|5V$iz5C=u?DeD0yr0^q9-49G);6@ZWF^(FvI)MxpMd^0@2T+IU4nNA5SB1n zcq-2S{P|FOALz@F?@72(8)+HW1-iS}mThfyw3rV*B2W5@`vH@}N=WU)iK0C=RU_-M zZ{Mf`XlS3GDuLYEJSNpotI3n&3tYfWo^pozvlNbk=B;DL=^VWv{Q{f!djK~z;mn9R zIbYDaJh_@<6MS;-gp`2lxj$nPsMwgh17>k|tmC}nx@rF2ND0HV_(ZRQNeStSm{G?q zW+R!0B&2MJZJ9CDES{9a?uY&!aQqy37`1yTVwo~96WLktzArEw{4-c1W*DAVt1i9? zPlWAhRD9>?`%0T8TQo(7WN(-J$Q+hKT`GaY@jiGlz0ttAy>l6tnB(dOf4pjCTg5we z&`f#cpv7a)GJ$XJiJj}0!B^@%>xpuNB(j+Yj3c+_Qt4*zY(IX~$Z#m>rby681Ffiy zso_9KPwtl=r#<<@tbVJW;_8_|u)P6Wo){ zQ3EUHnZdNEcCy)f!DULvYW0J_rW9yAb|Z{ zPl6jM;0Y*p;~utM7%FA*VqYu=ih}v!nlQ7}Y3Fi+#ZZclK_EB-NJab_h0no>VL})7 zNikKoEOcASyZj5&wr40Ba#n;}$H0N>Kq3vyV=!qIYAt_uvyj5>cC@RIM@u!+Hn%UO z3HMAd_{$?|<|lG;i0@J0vt?V*Or6SSUrUX=>~HP9+e4hRYO6L9PC5MmK$3b!12BT- z%Y>*XiCUBi8Le{tsDGy)q+FItNtpr2rZ|$rZ-d`_-8owIJc0%b`*u^{!kP6Ww7o59 zF_w+Um~5TNSS+gCw2a=%9r89h~%01cBE81~pCRp793 z-5I2l^;QQ-kZe?h@bts?3aSnly_S1H9mN(m%co#1%Wb3zLf4VSD*kyG0y~NkabFrM z`x4Id$B{<>T}lf-peVycPs_d--vQO(q4eq)=48%AH_}?1?z6bN$UzywWNU=`OHm4> z6Ac)L9M_zPQspQet1?E$$JC?aV>yr%u+7nTC)i5qyn2drCHG}EKGc@4zz$kk+1&l| zNu2A`0+xd3K)_tUNQ*uB4*jsEMy4mE@Q@eZd)AP@=Gx^*_YNA{s-f8hS6hMSVKcys zk1viLbP8U~%@Q-oBcM`ryZ_R%=4@ zGw14btKJKQf}~c4Itn0zIhO%H=39)K5b756fImdgdK|*^3hp=vH|h^YYhK?m1tJ91 z0_*i!^crehQ=2tC82>%Csb?!MsKI$v0-C`yWVqdh?6u4Nsp>iW$-3P+YOZ1pO^g0< zoE&&lGH}5WK2w9uMyGcAY|m4uHGm>Z>dLFKWf_EOo#-I5F7MU>zhmL2PZh^;n^BD= z_SR^wlX3_*^>4TY0?g=Pk=Bt5H!x#AchPh+3b@+zq+QEICT~KIB!=XY7k>5O3>TG( zT?HP-L&+xPv0KF+4xxcx<_7)YH%^{e8}KX)XA4@^rY{;3R;o}Ma2*LU^c!dwWT!O> zi>i_EiI5{p=7K}Nltq4BifYVwDJ89eIL`Dr{#SJLj>ZSs3n_Ui+z&GD;vUK$1k|ir z%VQRVNN1$uJE&L&R!@)`VmjA*&pWFGE_|nVtvnlYhpIQa?hq4ru5P*J zT78J!T8$|j`#I12n>rrM>j&Q4<-}>4Zq&o)jkzl)^cVOGT00wWt&ihU{X~5V4qpUb z#Z1P{yg3Grbu8Z+z$)urS8-&x)w|5_DFstV)fF(Mj+^YZ-Y6i5)Q;4Z@W|=P7~Y`>oT_YiPgbsba{YKVa@RNks6#Smhn^cF z2RLT=&u7fa5_ryxc>{@~T^9e3r$hZ` zIvy`?kGMOw7f`thGEkZJ=e{5DH;{~<%I5phfTg;y+#D_Z`6MqUrvNetz{w@py; za%6wHA~$<Z(Mg;~ zX~2Bq<*K)`+~p9`pF7cUf8NlcGaN1WB+h2yx3|7RBQ$u0Gi&d7*;uPGrw%k7X%O(kBV7bnaX&KPM7V zis?nY&zt}Ipv5;k(^OhwCnf=X$L;KqV|%^jWE%hYW8^SuhdvhXn}ddRFc_<3AoZ`I z){>+dt-#v~MrJUeAQU6VnC{*534(gSL>_YMFn^0#f4p?Ah`(-W&sYL|8x}pupZFb%XMqW>Cf?ceX6eiBB+i#&YZu11mnWX{ z_&^DZ#N_x&DX&tRLp-T59((x7K1GY;qM)6~c6$NWC^QBKn_+4}>-g#3&Um3is@22C zT5B7qi6_VlRnAc)vT~tD%r7`R@t4h%x724&~vv}S|*JBs2ovmDA-w9ZJf}0%s#idGm zd=YXgn$kkHC$Lnwp0 zclL$pAO2Drpir}N{Ri%*?1Vv-byym=rUtlG zR62+twc$FIt85~;0W>g&k+N0nz41r17fe$pn39E5sv;-m1jGK^*Dg9ZY}+gpC5*8pu-c$+&S} zj}^#Q1_m16dK>R*7!M?dlM{l-`$fOQr_PWt`&iTfAl2UY2KTEIV7%%n(yhd#M)oAT zx8}YB7t-U~)6sR%Kr5!yKNtJU3x+8Q2{;)i+zcj=Y+TKP35A~io3oXQ2>L0NftVBl z2Pj=kbU~rqZXM7*iBLK~!I|5roi%7l8(E2Nv2ioqXTOR1vxmxEu$+N6ERh6-L{PX6 zJPDyHbocQA;8e5jQ1tpj{O7|q;p+!x@d79e+3xQt_3LRuhFtx@8NZwkHUn=#IxTyn z>{-_DWrv*(Wqd9!E&_1WKZ7YC$TKAhsn&CIeH+rV84PC1^?ANG#il*HVJf z*o60NZtI4LmpLRI-xG0dIF{Wekvujwat-|mOoS%`BqpTirRF0L2*QdN^rwFveXR(3 z(GIr-b?H%(JlNdMZRuxJw(Qbh&4>4qz(8aHb4*RK~$ zSt-7n_++aaL>U|)9zI97Aer-u6#MMY5zp`ZAtM`02;o|gFx4bN;ejK*c@&-Zl0qbx z60@YZ>g(n!S;{b&Kf`m?#)A?rp#wZ2mA(GOlv)a)t=XmWd0Aol3XV3J83Ls^bFKYj!9ZPj)K zRAsHATf>ITig?=6nA+;OXt^poX1|!iw-1!_^x;I8fAU%Hr^8pCFL9OAT-~v zOmD`d5k#9n(V^A{Q=Ce)I(8h}xUwCbG?%QiQ50jeie%)sr&`23C%+{&IMlOl=R4E*W z5N?p(iF1J|7##=T&H82v{N~1o&EhgpX8jYBI|$LQNKsSEoUq7wp*UnnuoX>j2b)22 zt@S>8n3AUGP38O9j$y?P+KdqAt7Zm0KF5W~zx&=Qb93_edT}4alHGrGv)=tQS15>D)vjI_P@(`=${>>5IA*hUH0l>EHE!8spX|CV9z5#)mi&W2Fd62q%`%fgMxb zYuQUu*+qB!_)e)u4}z$v#V&~tIB^gb702QK+n6B+;baoVB?Ll?j3Ji1*3QKjQqS3a zrm!Q^d_2os0KTM6R+`1z~1`V~`eZ!a%#*s*~c@I@$FA>F?Ui6^& z_)e`>kdS{j`U#E2=6Ttqpva+#d_Ud!yf4E-*3FRjD=W0^; zmOP0K6axmeqkxc4FxF1Y~2#;x-D@#%rhTqMY;1C~2wr%nUNK`J94 z*gPOhOHK|0yV@IXL&F>%bl7qclbA&ejO@$uiW&s~@p^xui2yWWYxY)F+J#=S{r z#{cP7)Z#OHB0ke8->rFx`k5tokCRl)ofa>5_c8MDpMhgh5n`|bDz)KK*MR2Y0(#zY z$3qwNId17=H?mQMDWH!5BrpKO*KQp1z%)<(ikrHg)ltpaoq@^c7|ZTvKCR5ogue3w zd1o4zREU_AtMAP>`V4~%yn^*ELc9!kL}0?g`-r0nd!A%w2>uw|ch?pCUvOKOaCZpo zY`d}Y2J}jKwcYTEWPczi zLli;*xpE*i(kQBCe-}2b<4j3)1Y{Cvc<<>`St*@Jb^{Y+x&T3b9X#7QL3|+^hC73a zC{KXD2dF_eJ=4JiX%&o}dI8Wh_CKVcMO^Ud_Gp7lXI2&k47y+6LLCW>q^_{gZmjNk z47{B^5sNb9?gnwaAF)ujML|0Qi71^$s((W)gF-ow#Lzn*Kr>UtGSG#pY7PJ0-+eQ* z{Sq9FF#QI=^2v~FD4vcFirH(qrLTMul5RsHJCwkV^XFp=ED(Q?@w8!M)49}YZKdj= z9uDP3I>gODd65y=;q^6OCOa`)G8gmuNu1k{BA*5+yY7`YpjPWV+~uA!lHG>)X8l#loTnzqylMolrPhHufS0&lB$ z@PB~KKiJ z(A5T*=Hd4Z9r_gV*eMycQ8y}_Q#Y?HA3N#LkadCF^gh=Vv88W9bUaRRkx>9TH0QuH zEEA&zKcj?&n8g{tC)inqj9f1y;O+~Dqdqol6Nuy^VDpIG6e%%lpOqzaCy_!9(fm7_ zy6g!5I@Y{FA|>j)#Q*j*wT#KqV!u37L8v6f?64HRKt1Sk2_~BB(>1BSBRPOrLwKL17Oa{zJ+ z;Jcet^)=}A85i`qNZkYT13Z5%PS~ITpTWP_+Ptcc}(ee3TVE{1(qvMo$?ro5>fBW zq2J3UVRf``{;rDW;3kvBX4xu|R&?DeTkrwC5E}f(!V=ysmdqsn7j419Z?Lte)lQO0 zqACovAuRopqSuxdb#fuq`t`l|txdDZV^6md^t^9FH|B8VrA z6uWLSoF;O^KkJ}KUPWw=ElOB?$Qn^i2Tn=~p(;BVNk;ID1MFRQo*tnWh)f|d)wcVs z$YYbi>Mz1(M&(A={1^~-!jC0KVrs>gf^Hg{1EtXaF9c z=3C9}0sO^v<5B>U7F)oWC+FGK)mRNX3)67RGu`NDjG)zg)E_pVK1Dk}D=12zsAE1Y z8-PN;fe=zS5N^@X(e-$a70xD9br5aMs~Z?NL^$2ATjb#Mopev`7x~qy_DnHhHYHm`nucT#T!-UO7JqLh$w;?I5d2)#F&S zANp^8a=g*3XZb5Y8_^02Q^KgxHyC7N(*5E6dJ9^WZoktw%-D@PgDi#mcGQl&<2mrS zF*t40io`VoHHm}>OiHf~pW%W(l4vgDhkF!NhKuR#tyE<(M0^I&DWKmkgD^BdhySmh z_>WFSC&jF914ZnW#YMXq;{-JKzSuv-0UdHjS<3Rsm{5V=YZi7$mg$CFUV7i$j3@cj z(6TVGZnjfiZV;?=p&EE~sR7+u=d;vC#SDb6U)`$A2Vc9DraQ&+L}cEC(=ap7Wnj({ zDn(>d;yzIXVV1dFfqxT6Yv!5oMXzDpLfEUe^G{{I)f7F)@tXy>&zYBVnf&Y%LPvl) zvH)a*X}k^@%Hk&YwRpqm>fZc(7Ota}CO!7~?SgT@crLj6D2hyu`beV5exFhqqm z!~O3u?lY~RZFULos+wwHJ-W(w1Hog)R|DTbd;I|S+E4TUt}M=P$^2h-t!Sk%u}e#{ zg-I7%_ICXShoX*uV(J762)bDAJ+o8E*o24*R8m zG}i=p89E?|4{5MrKZ|(a$IY|xUdKg_!_N$piU>ILi>gCDMaQZxo7Z0MB1B1uQD zjtiUl6$^P+B?d`9W57jtj{<1EK|ALb-vGmtZF7_G)TiL*o3G)~Srh=3PMs~ki>$J- zM*UJ%%pqmFY=?E~>N$;+dj5<`kImx5JJUL72EQ!M*SZF(}mL;^d)rREEZ z$QUhT#do3P2Byl&$RJt;K(~)jVK$i<`YySpkj)M~5m4seBIy||v4DG7f1?dJ9;CSD zt?LGuBaI*6>-PO6lip76YBUzH58XbrTS7bIBlPd@PeierUa@N|RP72-@NM&+`5e90 zdwPVML2KiE8|UM<+%e{|+?-yQvz{>9lH#MdKg2gI7$UsAx!<((=%$-3?%p-($S8C$ zGa#a$-KnD@row*)RjY8G!K&~9QZw#xKD%=@^@;!Y`n>bO`+i(f8s(X9)W42O=m^T<3ge?fY|msF$(66Z&yBp6Dm{ z-elNCO>+@&H1B|{lE?9v?=pXH8nvYRPqg1-Zvcw?7JS%au{Ti!o}}3Lxft%0PJQ`e z-@;)h-sb?l$$;u8b|+Y0(gS zvE#NcH!tsfSCNEnh)7yqId}@^Lr}Ur_Of&mlPhn%{H)2aHanh$?xcZt->{C{zltIHY5!Rf zzt)ZsXY6Ue{LQJ});__Y7$fBPzT;zu3AQu$7!LzCRGAOFcymJ9=rJ%`6RL*#1Mlth zTTVm=$~orU2=d+?0}kO_N7-SAWL3liiuAj?Hx4M>7vV$d)V9$<-@fZ<{+`+#E<0_r z`@2UrKoa+azyYQN$i6Bbkx`B?qBeOzF@EHpw=- z8&~UXe=0p~2C^$-Wuu9JBh}@vQPm-9Jp>gPEjbXD%cA`9{JNG?FIqSh(y)=OR*}51({r(X}$@?tNGwcC!IeNT861TgE*Z=5hpI z11pT?-5PqPv<0kLrSggR-@xKwYQ_teAqR%V^fILTU#fx8f5&G;<`G>k>wT+0nAFcR zt#9@<>ZoxYr-dahgq!q{Co_eRCriL_2-QmMPnDWQlz;xy*1R*ym5c-fd`SZf6~mhf zdG5so9*k2`42VGMa2SDTfCInwI{35y*9EDYP|JEMmfJlM`X)qi#Kqud1YV+O8t$WE zN3njfT)tD1_31Z9rivczUaCEfW|OB=PkpFq;tK{;Q(xLkr{0OfM;$4fA%AQ(KI0G%F!{_1tOk%eb ztzhMRNwxJPa>HK;VQNwk_sCXc8Bwi4(Udw+{|2KP;cR9(uKBa(E&BsKv9}wAT!I4| z>8~NkQBmjGMf2h%Jt~K_pmM;l~_^%Fx8|Th}*L&i1 zdyW=1I|kfH{)!RQOSctlelN(}M1pcfmr-eSqrrE#!D%@@43k7xyt29)CS;k0yFqYlxWco?RyQi(bdR8SQ)l|=%ye#gVAtu029 zt9IA2ee-`XXi0`UDh3fAedR&~2&N3$ndQqbME(#-@y~3x`@1){x|k}#Eg7O^dVJI( zq3z~ei@c}yX#&l&pPv1!T^BZ1(*omc^%HejfWNiJj4r{i#!Dkx$&YvvZS%9p1Vl>f z6T#ipP;?_vvNV9-tPhJK;Bplf6`9@ks7d(ABFD7~W|a=RXn!P@?f>J8IY5|YgzS-{ z`I@=Oi9e~U#{-$kIalEeym|CUwP#jl&P5$_w3;RsQLHLOzrHxWV^=n7&1GWOw=}?0 zH*oaH_bTHxQ%CWhCc}yI{ZmzC-Y=42k=MTGuEH=iUMXWR8!`8ZDoV+}7RGGsh(cbX z=n-lCzO5lGaN3X9Rx z@~&t>5VElidlYZ{Qt%`L0M(rDf@>HxlgrEK{7#ndKvlA8^Fx1?@wLCFdHP;!#i5<9 zUK`?tqLS0d?x29vT~64gC` z^%f5m(+si=O*aGCR3dARId=ifbg<@| z;-6XGzqEzMmw<}iZVYe~qr)Z&diwu-;X?6Da6K$+hhwSPM!8ZgHHez}*L>j(=o=yK zzo5#d2M;w~Pl7yBEN(igFKmC*%4;?Axl25j7)sRn{{GW%Eu_0|!9mg2Cma;H1qcPy zxj|@bDV#{s=E=f>zYKpZ&H6GtY`@L9a8OZI`SDcw2X|Az@mcFlR@Q0GcI)V!;e~&4 zgFBYVe!Vaby^^uWb_kWHJW)G(lw@nkRC*>yZ>B5K{IAzgV$y92a<#JK!0CtKOxLQP z%c>6<*_O{l1uH`pXp6m%T5$toJLVOl7Dq?@ZE_TNu(!B-rkdp|V>Gf7<`clg9?Q)7%Yl&pfbZ457*;Q+Q=S6f`!?T8% zcJb6`7-+~rEfFH38v@b|eVT}8(!p4VRj;W{DieM#r0DdEl=2TSLrU5BfiaF;W1-fD z;2xxij6t{%D1grD|G_*I2(kCbSbDu?C&WjT`jqx6_wmWI4Y}OI?xSD3g7oU1x zb#~vT=meEAdLMnTYi+V5y+nIvIIrD4pSeqxv+yULZxVX2UeTFNP;O7-vtFGkW3)2N zb83H2d+K!a#*|R@b zh3|s)&@R?b%>O+SLiq)c*P(G@-h_DXRu=3E)P*lH-ia+3*Ql6f(ZR2WPqVZv`zZ{+ z`i>jJ>rSn(D;KB0Z~pR~iweEbjaumrOM(GJFsT5siE+={7dk&J9Q+Jkf}j3Pg33U3 z)J#|MEom{1yJ5cd`{a|ikOqlwI-}&=r$KV;q-@*Q#*hRqx<4xlflnKwsh4BKf*F@X zRvK?jpO4tzn?EL5(r|Iky5E>UUm}djfqxi$_}JUTUTw;Wa+n(%iAF8(cmPE2hqcln zU6tkWeMiLTLCBY?#o_(24A4Ubl5D>eLcqgg=sf z|2!h#d!`_Fe&|~V8U#BPy3(m;VL zOjL?j{FICnn4uis)#~dPyWU8uo)l4}2KbR!vb8?`NxWno79$e-s+z;lSM^Y|_0zEY zUaQc_LJltO^?aboy^P>;C9Q!mks!TmgGIywk=hyQmyGp=*b;|m#_jY!UB~?Gk?-}> zJX-lCm7IyX5~6v%)p0P8tx$URLkPYbjp~iBEeLpYZXi~qns%yw*KPhWF^x#Y&=IbR z(jKlXP~JxLhwQ%brcQHtyL0qht(KUaq(Errbrp4p7a_)LthB_$pl{!#gec{_y&D|} z;av52RZx32LBqY$R9ur}p~eaq3^uIVgSN>3xdcdQ;W206Wg|R|gZ~ka{CXC;lF>zB z=_^xPn@n_viZkU8e#hf)`AJbwX3jN8-YPantcan>vv2Pd&KrszlS$O6L7K(J4fK>$ zY>9o5dRXSpJE!rY)!+-n5DwDrJP#tdTI=G!N}v&E6P3`5KFitgRd@D0;C-nh4iN1(MX1C&zp36QzgYk zQITVLQkwLN_aQQonoo+)5rLsJkd(qon-V%Y=?$O$?j-Oe&Mdp35AQ~r=z|A;shgBV z{4$he$UG)>5*mMOSsW)ojdn)9sA`zu_T2GF%}Nj*M)Bi*>`i@%`!=T%K4hpo`E*ey zD`8m_ameZ(zv$--!%7F&yy4>pO%d0z=3M3WxlF-E1gfya;K6m#4DZTysGSe%^9_$@ z(si3l6J1ahJE0(X8e{E9FQH@a{P#UqIY@n}A z)mi#Lr(yc1@53{1pEE@7fE`3{#%uW7&pYr7X*Y1?21^SPZPURB1LdI=@8Sf~15Q)X zT^O4+Smf}H&CNeLWYKq-ikBB&CculM;MgcrD6kCN=_Dr-p`ddm*|nDEv`GmIu5vc6`mO{WqcN5n`q7}O3SI**TA zPlKLUxF@67sQNMf_!*p}ic;SE{S6TXAC8>73#|l|C@CmbMIf4i=Mg8$PD}hHdap>9 zmDxhVe0A6(%xj2bG_sCCA`vE7yV#}vmn5nP9o(hq;~RO;kC7J{)7^C(grzcW7K{BT zu`YGwscyuhrgsrV22KZ%hkylERA$;A;$cMicb7Y(ydu8TDT7b9sZ?+Gg9kSIeuofr z)kvb@_Ivz3+KTus)^}XQX$%umR%-gw&zJbdCXM1nQ@EC@h z)fl+b5JODlU_$hAJ=9Wdcu-pP`;A^1&_;mG>`sd#4Q^Y7g@q2!#4r$;2tS6&RXVVT zgw~Q+P8}aEwm9S^c+z#T%V}$9gn~tb&MTVdkA7cb@&1!8FqBDZ+WN|kxYc?QB>cd8 zV3BA1?r{~x%N^U~-Ivr4hz-j*TF|qbE$!IfYGMF?{5&3BBW8j-dG3rH0)4nIi*w}Fd(!#aL?*HB97g5a=h9MtQ(mdko#3l{% z%(2y)mzw57qXnp}*W~aNv`Nsu5`9jXOId2pJr5Y6{KvAEIAS5P!tH&*=_ztEu-29%* z$jDe3+%++wA%&D8^3OrkhKM49AqroghCgs^l!^cQt}}icie{hLAWrU__&NvEKCo~9 zhPetT~UX1P=)lN{1Vl9vhG7+1@g(HYXBj`vc250;(E7zs@O8GjJl-Nebt9?^augOcI z@e3rASr;WK09=4It^2Y1bLchlia+o!wPj6w9U-Ow$ZV^WwVJ$MR1UeD_!Q}#B8On9q_5W1k&1vFRYMjAG<)yKY~%jN zB>lQBsFKzzwJR@)$ir`5lqM1!1i>K&vS}}sd`+9<@dV>5tGbODNrgV{p%J{liw@2I zLi8^vT`E=lopCQO$$UMVU=atp@P`-`bwo9F(YBLkxl`C1DW*l<-Z)hFxD79~{R&GQ+{}O9I^Xk9 z>W^Rn-{0c3%Y!ELtqoK6X~VO3`;LmH^d=F}>Y^RnJ-kNKZZuP;{P}D9U2hNuV|+;{ zHG(-N+JkAuetX{U60eP%qH@Li9s5$)-DkbB*3ogc_z7xd-dr>*0=#ys!@Tj7GdQSy z!9zpAb#}pgOK|sQ%^wm(J&6!6j=UyOmSXsY-Ns=r=5R)nS=LH zu@_Q$%jdcwM^3m;D5Qox@h(yvKl6a{@j`>&2pXOt2o=0#Hs| z4ypz~jP8G@!ssHyQAY9LMr{UOm^ICJ=0E&1(prL(B`)GggST&cTs6TlR6PadgR`)# ztv*0E9-tm^gp*4%2xctz3W1^?LSuEo)7O0CB=kV;Q!%E=}x%(DWQ4p`iDZ zBREao$u$AiU))B?O=NuKErZe>Q6Tb#!E#!QGLci4AI%j9R%AY7_2#T?;5|+bCOXqK zyO$O1LMA`z;@ivV4Se%5ic9MjDRB8TzoGudrYw5d&L`JfJs+J2S~x7jC?dMA;q>LP zE`CQ&LqYGh_pijH){-UEmTwRQ7)ZOixk)z1JB!leS7=vt_mQM$LH(J? z{v)V5hEyGPTd&nMH#~tCxcxu~Xe6^1G?;vJx0KTbBHoy17%PVWM39IE2LV{!DGt zoRCoh@G;aybDG;=2C2xWAjq^;V=aO`Z+_u;DsW2xGgU8o1uy6`FcL!COSEAHoM^aN zfB?2e45`IB;fhSZ8Zus)C{cGH=lSk0V8e6!{dq+i!JpjvwzD6KTjqiDC>TCaCiS9~ z#`XA`ehbU$zenMYG&@lgkzv8lpl=k^OX|hofhF4|Ygv2U{FPL@dMWBMEnP?J5aDU5 zMB@dImOH6?;Xjf7Cy40t}(DHLj{ofwPGj$eX$&Xn%Zj#RX>}zQ(+3CBn7|nhR z+d(~PyEdxC_NgnH^YAS53%6I7%(O-4dKSWEH0WTL zdsd{|P5r&&{6+8XuEBsbJTLE-iFB_g;bg{D>04VoB&g$O#v*!-~zaN znf&<|;~%=i%Sin3xWYKft!waMptq_lX5_o2-KrqhR6cuR+==x(l=~~z>zCS%Gy508 z1W$fQM@UGFKZ{i%rQmEIxLm%Xx}T{h*Alix_MT>aV68T4LU|E}SIognYJWPEX6LN` z?s(=W$&!1X?t;&jfJARBm`?p{tsgV(n=+Peo}ap#%yg^R%eOK5vxO5L=xeMr6GfMa z7*>v8dU_kHp-YQ8v@}Q0A(tk5`O@8I^(5pgOMutDG)?X*rs7e}&1JWUi0l1B`ey#+ z$P8v%NlIunEslB7oe(a)8X74E3AyJanorD7$7FIAmOlGiS{20{n`z|VUl&LmN;7hG zsJ+rU_6`>Uz1@aG1yzL*7A=Wc5k%Mx2BV3Jd`q0`1m{Ai%(eoKGTH2_e!VAkB8Git z6NzbIx`q*fZ3=z2VV+B;e5=-o7MuwEUUPm;dwhc73J6RG6d?}S?`Q6yG5fOQ>{ zBGiI9Co=JKYNS;~H~05tNq+{WmfOA*m-ogVPJ-)AOn@w-Z>aF3v}NVPN^o9;=uU1D zFPph>uz0a?V=R1v+FE@*8rJX2mvwsvc2@Wle#cFY-_n;fxUibDt`EyttF5mt{A^|s zm-M;N@i{JkME-1R%30ktag%J%^;;(GJYDt`;cUD(uGeZ~%`jPUaFYg7~bOZHz{OL{G|KaK^qoVx2w+{@> z(A^*%!T{2ZlnR0(-67o#(%m7Vq!NNkcXvx8-6v^HFcs1U0?sN9p z`?@~YvXc(_7_RXTZhg6{zJh>4jbdJ3Ef>bLNo2O)r{GjK8?^hIsSY2~x=~-Bm-^f3 zEBJxrxD~j*l$g@o+}s4rF;gB{y;Xkxzp$f{FW5O z!Z-B!>Sq(KQ zh7$dI1e)24g4pY@waqasy#CG0)?R;Rv#x~Ri^A^Es5{;zpXCV~Q7SvL$quR2)p%w- z;uE)PWR0=)W%tRoH@2L~PR2CRwK_iAZ9m4Q3H#+!{SR@cCGD2gYVrdtvgzWI@q&*p zjf+G;Z)A^Q>v6NkGKF}na^h=N9wWLh*kW}uj85M#AWik}kz3-4Fn))GH!|)WD%YW7 zxPHS7+L}n2Zn2j8YHHZHbZ_TTbK>FaH>2lJ-~#iyHT>gT!YAE>y`MkV#Ch=N&>x~( zgyc_H+|=gqmPz*SooB-Ayq{i7(0USv7OELg};=#cNF#*Oz? zLml^@k4EKG^)TO>>spIew5kU!9hm1hVTb8|?oS;yOj!~kIWjl}hDvc`+k%PqWwdq3Ie2ri@EiSlvY zt7H`NOHtr1#jT4|g@Jy?s8h&3a5u^r7|^^;O5*W4G20d4HSR!WBKWVF(U8bY8WKs4 zz!>AoW7Dub@!pe5q_o~TQ0^n^D09&vQo-qh={3oRwxt;aS$V``?O{2VG%k_Nx_!4X z{O}N@>R$5&$G$5=n+1V;UK{y&Sx(Bi_uTwJ|Mm%#TV?xdlIjj7Q?qjvn2*< zvijo7SAP!ZH_iC_Qit-G(QGBDWZ1-#=HTCDaH%i97$hYmG!6AFqUkFLf)dEf$;}!W z(Gj7%ME2|(9k0;DnABi)mqYRK3C15&$b;;a-PK*|Sm8wBI*YTd_z0te(n|XSSA?C5 zW6i4FJM2Lm?EGT7mnxkljw3WcdS~?WsnMbcX`L~ObP3;MYcn$hfaV_m0(l&kQo`AT zQWAofBiQaq8yW!gvo&ti^?MZqwoKB?ZXu&NDNc`>(G9(K@=^pHKJKjZs_N=S1F|z{ zdatdS$jZy>B(4DMVeu}yX+%c+m&1nty4$Y8#$5JP$(T?vax&l8XSVU*9o3O22PKs> zmhT0PP8qmp^mU=H_Kq^_1(4j#Tgtz8H?4;uJmyA~_d_IuQAQ}e(rbt|_PPAzV&6UJ z-s6d}s85r6!JlrqK;n^d-kNb6>K=*jHP~yh5b#Xw*?gSic)z5l8dv`9&5q&ercQYI zulZiXrA-^ty;DuK){P-ID9xX8@@3oUWtzTcpqe~H70pp4prSTNg4sG_REde=vVE_3_)f4U1$wfqPcM8BP(pNJrm6rFu$)JS?(R{su5Z>!vk|gg1N%)m!Bf9p^63? z^_$UoAtk^D)hUQN)^}!?t<)_^p_L4}gizFub)M)d{=&tnEXUZ{Fly`$nCi(Pp}ozt z&1ocVe2?}j;ZaiE_0n4g2;^baSIfiY!OM%o-o+q?+PCMFgZvD`mgjtR*4tm2zA<3& ziIN`~CzwyN8x2udpc#jC)a*X$wEp}B(islvrGhA>Z*~Q777@QoVhO=hpz4A6!uPOY z@k_mQC8fqyu%q=reP7ULDx|=~ywA2FB@B<16a)b+kjnq|7><}Mx+~NBUlOG_65(j7 zr^E&l-L5U4I7-Abn-W;cAFi@prYy@;m)i339qdgf!BH)Ra%8+u7@x1XM7C0;M};c= zdrkaud|~+wo(GEp=6ym8a{BG%@;yYURZ57{2bDB-9@T;xGwUr2V_xdyfUIFnLI}`8m4VUieynqFPTjbmtd4h zc>!FK%k_;{lQ&4AuD7D^J-5hAkFrC81N})y%HwYJ+mGtWxi@!nu#_$K!-H3yt~`Dt z$ws2w2pTKLOr5`6^1HFKk*okTi*)rtgmpO7CpaeOZX?~{cE5Hx4iLEU0I8eipykHi z293d+`&I`~m#A%~iptk?QAaB)ME;m^{YGWl)J&gJLfI?`5PWeZo=>KGW@!1d+G$-L zkb`Nyu1FX9kG0M=dX(;SXr+bmq<^U;s0 zjj^`2qyXtJh`}SsYr1paXb8+5NVwc;aI3_a$2J?K=8fmXV8MK8@B8j5F_rVkS5MMR zK9FnU08gUErMZlXVGv1Uc#czh<>rKqzP8&<14|jqpWj`Rl*Wv@jdOfuW$duFISdQ} zEIBU5tOH=%(xNN%dS7@3Pd*=zR985O6k)PyO`fpTdf( zeR0m0k`|(xuh!!P_btlCn(Y7hGC`|iR0eyRV!mZ!RzSUeKD%(i-gJbRheOVJ2bava zs34DSE*n?kcRK8{!U8G%l+kcpV`*{2A+!0>N>5L4esmKN6B~oh>_wW(-b^`Q*2GXp zN8DqTGEJ2l#sijN%Bf>k@wR;6hi#qrT>4Rf-iJ>Myksq>nTJ}_u`Ndum2ZvQ-1sRd zDeVk9%57w2yQHV{s765J*91V4z!bW22kSByV5)m$wZsz+m@+ONf5`Y^$Fm>p#^nxW zHe81DHm^tgw?Rn(<_7ZALUWyuXoaQ`<(iJO%|3rEKBGn`T@7K`vQxsM+Y8c zy&|SN?O-&+f4H4VpoYTDZFkV&^!d1?x#?OLhQi=%K)I;~@#i~$&O!43BdV(YvsesD zKo~gUgIWiI3ACmT$(Xm&mO66eqHTQ=;$yG@cxSm9qki#8>D@N@yG;}4Qv}RzRsb%E8>Z7 z9J0+U6zd}sxcCo#}6oswF(P8osKrzm-Y2NofCG2Wn<2l2j1Mv`8B9wa~toiQ^LGNy}e-%2To znM5A{p`{v@zX0 zVbok68fcwto=~GPs_~CFH)7OcMc^v`lts)t4?pQZH-@aW-*-%RT*<_&sMy4lQQ>`y zmSZo@_<6D@y^$?vLfWs8AP_67^a%$_Yu~^>-VB6J{vjGYM3^8}iPn+YSY5jQe{fgt zmKq+_F;s>rUH(B|LpU#a2Nv%cVV zoP>}E7}Mo1#LU^w?-dYf02NI2QWCyPv7)h&S<%3hSs8PO6737N1iUpx$NdvD+nbAD zy+oqdGLOlL4Nt z0;MZyHmt0qWEhMQo)7%bs5@pF(uj-}@?`gOB+#E%EVg{a!YXZ2Sd0M9J1joUkGUY)l zJiQ3(z3J2Zs3`*2DdO{yD9N*IA$T~;`9g2+MNyw~eESb&J7y}|gIJzw?r74(81sVf zxVffJ*UHEz{V|aM<@!Be*{LwldrV`}dn3liwkL=(+={X!9NI0q8X;k8Y%NWm*@CO~$ghn=Ue`PQGV5m~#EW zwHg3vspDmy(7R&&+n%XVsrXZSvU=}50+ewfqW1?#EEA_rZhJ;XeLhBK9|B?s2hn?D zo}y-2Z?@XXo0@wSk^jWgAS;@&mA4x~(xver>6v<)Qsc5|H1^FL`2(bDtV1i8oW3jj zdz5k~+fI_x{MA6&YRs~UYtT-7Lu!MF(TzL&W_jmL8pF;ZD%3}WzZxWNWi{-aQe4O@ z`}d;!&%Q<47UIU->)-{QxhE>>E~8xr{Rgv8XffH*!&XwS-s({^5BuL9hJGw7G^3RwM++gFiGi$KGS^Khx{YI-a1J86Donr;0f}nP(PNKljLklz=V=IvxxcxAM#>naTu@@9lr1Wrn5VM1ybslwChwU z838;R#fEh)PvSH6QkBs7otDUKnI2dLg7h+9KcvAk? zf&z;oVo6S)T(zE(b$PD<fKqw zAsp(BLwPPvTzBiNNu5yheLqm9(`Cf%#`l+6$x~zJirSZFZy5i}Aq-dX{ZI zHWZT!=mpIPgs=>U_pwI;jLK5Jp##I8M`2b~gjhXhNG~>p$}zb5OgJ!b1anFGtWDiI z%myfem}LqZ%?T`%;CEl=h{>z&HT(GHKjcqC-k~tmTQqS-fPpp@n<1=8may%~yd z6a$w*YWNrBT9@}b)&RK@hU;OXaF6X;y2?<Zp9T)94O;fb1gwR(>EHt`4|kAE8;CBN zEDYqg0kx~|jZ&=!&csv$nX?87%s%5yj0vK~zg-l*m5jZ~!Yi|uR4pfGokw`Lg_JBB zn1+=v?w=jZ|Mw8WW9>pb3JpI+ksi3+zP>xT7;OmMvY_7v(H+|8=klWm6J~ zLwy+}OyzYA-D|LBw~(Atmr%R(#`&j2%IPW}MQb*!U_GSCc{^xn#|yi1HL4TtIT{!` z6RWh%A+OH!x&aU9FktrUn+t?O4dGqmC=@HJ_BJ)87_22wsES6?owF}2xvIgODKj^F2rP+n8Te=(n)9Q0f8(KNWSQ9m=>RR z91e8~WC0D=YGKDLEGa7?j*xAvGcy^7SFhf=(IoCh{OeTa3}iFGio>l$^=iH=S4J6m zKEmU}106^*cMRBxHQiR_$LL{CKGi2ip-hK#(vHF$P~LBT+dh?_w%Q&a)2%Ox)!^mK zXDd1$85x00o#khNi`!$vccJfFTtkHCvKL{dBSetr>6=B7UsOC!^`6_9(9h*`Ce~C` z@N#V2+#UMIhtv9z#Y^6Mey9N?C}_Oc3Hm)~0?u!J9~zQ#aw_+=M@4;8(eMuhwf%I^ zjjEU==cRrd({Rk-TE)=Uo~qAIQaw}_lTxd6#nGX7UW;vnfk5lKQGh{0 zB!o{N{_V+==d410O<2Q&DTq+o19og{GVn+^&Rq4LmoWTWLJ?fyGVOCR7J^Etid6lXMzy1SIdAC~K^^(AtV40XBnI=hC-YB{iK0BI>%%>KlKGJosC=gltogwVE%a5u*$U zHd+CT$1g+I0>a-fI8%fO6v@_fepr4uqru;$-oTS(_Uz@k;W?ld9pzL@<)-&0(A^&Q zDIOosL~$y(_Pz{_3JN@O_`JwV`A5KCCvrDO+*PPULs!=(*tH66sn+3B@$K}a-FiX)|Ajx5n{!3RA7CWMA3EwS08VD zD!(*i)~C+`?dAipDUEu*DVe#nGaIrB*^oHi9uLn2k&yN;0g=h;Z4sG7Bag;7ASd*% zi0qFcBpE|n8lBfyAgcY(b3BUsM z9PZBkAn{d^j{eXVpw)`NSX5|uoS^9-at-3i_7)qbHI(6L>|m%PenW93_yh6vP#Qmg znpZ{+6*r?A_Vt+_yPcj_zFe~@Z4T2k^qtgrcDtP$U*>n`%@Odf?w`avf~KRCRdHa| zxu^ItJy0sNJ|aG#kX2{`17*EJ<`v?Go1a3kc;>7E`WSx7O0(}x`*Ro(E^3Fdh%%^PiqKON|iay2t!1VHW}PVdx6KjkM+FyJ(gpdh3e8u2;5 zc(xUiJJKbXX)?CPPi2r%;VZ|!YZ>i_?ci8ZYQ(?&B#GJh>S%S*W?DWvjGYg&Z~^1h zR%)VFqE+Z7@wYJpBdWPVF5h^&|i zW!wY5&M@M@7Q;Wo{*< zxpS>>@}8@LvB}C@wT<1{asvJ1y}Rm2LT9JohGY+Ywi+lQ3RW%|Pp~rFPXdz}b5Eviq=JnpzlE=-H zmEx_9x?WAkX=@VgVUxky1T=Vm9c)UHeyY{wScg~cG7&Y#B<4U??nGctGROqK2>xfZzRU_cbB3bbq+s@(ELIoPa#Dd-sA{Bs;^QRg`h zY{J#X{3|=S3lM1VwQlD`JI>qHuEvQTv|YCTx#O~vylTg_J>6=(Y(AXbYB-0n zd7+`)GSI=fh(J_&W=nZd=eC+C^@*Y04lEFzCuK=C;fGB$IDYil4rCW@_=88&oWqsV}5|UwMju=LEMUi@Cwlw1X5w*36e$7 zn$J1XnFmDMVbOc@D;=1NeIQ0*BU)&j_G#YT30N_ywpIE~B11)yyz?)as94Pq<<&l< zUjRYwSwKggeS>EK&xbBf_zcw|z6D z%ki5cW)WmzafY)|e5U3J|HZaxV;43L_|^68eMPS#9_N-u{@pz+Gs7^moNLLwiXoUJ z5OBrC!0N0wie&E2Di2X^77d#zQVhfDek3~0}Q>|+gh~h<--0|<4$84 zu=z`APW{mKtL#lA$qrq%I3nL-Lm#U{6S0c@ z$>6zgIZ9AKRgPi+OB42HvDry2xqEKQ5|cOI@tYrvMrW3%=g*Jaw;6wDlH6fOqWrB> zfAbYl&8Y6vq*DwAWg+52*m$g-Q}sO!f_@z>2_!}#bk2~12|ywu+z?4#8)DGM!D4n* zk+n27_dCvu{Bjys1``o_4DXg zRZ}x{fQ?S22=w?snRv~*J3fDA)D0}l&PD|GV&IV9t4Sv;vR4euiDzey_(*powq7+X z>cogHcIiqJ#h_eAZuZNENFlB5v_vH{JjQN%1M-?O@MK8JH?{Tw-f3+Z5BWzPTNGb! zNc-J`t_P-v>E3#O z&%k?34%8_U2tSIHKHI43>-VXN8gx~bie#(2Txu*3tMP!d^+3e&z$;^Or!3w@2)Ht94PBIBMszwQNjh)NDx8y7)xu;Qt7{ed22Qx5~V2f12&%!CS2tdcD zHESkbqr>)mQz1f-EN2K;SxG5|7^Nnx6}vlZp|+-GN_SkC_0d^+yreQdgJHVLSL^V= zo5F9@&rdPLM(P$mUDjeOC@2>frsIx?`hScU!JYJOKdQwO2aio3XU-1ar0*@0ECj`UnaZA`!KpTMZ0+;%pYs=c zg(dfO-#e#T{;S=jj>9H(WpPdphN$}^lkt4{)0vQ4AFkZ5Bi)WHN|rLupA!-gq+MdT zHPuiY(?UiRa+a;C4{Y17wschp%ncL=G#~sX0+7KvkP-^#ZLR^Gbtg!aiFJD9K-x^= z_Yh8);kTKZX*+qxrk0SL{Ii(&j#z^Nq)QV6!Vb*)%<@Lu+znyOKbvU^xO5WJ)1!g3 za$d8{J9K+_v?In>u2@)-SvezOl|i0P491XGK72bmk)Eg$ONa?(UazaAUIkgPNhyU9 zqSLhLvHu9FTZ-vmWi>l*OHuj|$OfFOd12E;aIz>M^1;S3NS&OVmgZWdXRt`U)QZRJ zRXv6c<`W zCFHkCg1wkQMeITuIFG(_jZS>Z_TBt@8`y&1e#{3`1mw7K4j-@n0O3+_xFM8E2rr36 zrAB>JcDllK=jDRudh8WElq&Pk2Mq)_M=xy|fC#R}<6;M2tP!zOsKnil$~xeJbOQp# zphNEwWZg*z@08*e1kiMXz!5m9kOIz(-QRjrxT@-S(|Uhi3Qt|IYaKY82|~eg+{yc( zgJ}FF+l?o_8pspoJr@+JpD&!jm=JXrexb-95N@d=b4zdp$qm z^0swcgjP;v%){@KzE8ZriqJ1SZNr$phGz8Axwx|aeM`)E>Bm2aCVGEnNr;ih8uMt# z6Pf=Z*6)tU6I4z6ITUI>A9JG<+k0|q`h&>-i!gJ4Utif}N9wD|#8aM$-8!4Ao$@&fBGf?MDhvZvl9N1L*5Rvc@-u|d7O@U>bcDCTPwO#b-Y#|fm}7a&?$ zos)^yAg=fD11KdkUl>m1_tgmpn)&9t2D*tbOt6y**n>(P?ldaP1wYkalWU zStd@p9vje@2FI{5oyf}zJ}5Vmlrm}0#r1p0L_*XKPL7~NE!B`Uqw*C-70&JA&qHOz z=m&y!@r0PpA5Ek%k*)X!F@ugDH&Y{JwI#?rCdrSp{DZCJw2`PaN?&fo=c~=xmz7Ze zig&A#5UYYX-9Ap0PH4i_VRjZaJ@uTk@Aa*SA|Xq8mStpDka%!cP;>ePD2XR>wT?jG zLntwp-#X=T*(Vk`gzOdm8{k4Bh5H+7mVX`|p^)m08kf`g4=np$>Nowlzqx|n{z2bz z-UIDo87gsFD+%(uHJ-wQoSZ=QXbR8G5KhNiL>zgjQMJJ^@&vQWQv!1=>FCz5H>R3YUj`OxW*k#~Mf?q{MSx1YY%sE85P4eGQYz%BH}MkXG)Z zdnE^zm-8gC?ik;bST&E`78(9_-%1y0@XR3Yu2dkmTz`x^%$r(JeyCd;pRlQZYIgip zl>eimj{F{6tU25!s4#Qa30Wt) zLev}@F!?XIKm;QK(8}DA&su=Cu-ByjMC_2(pn}S2G2-E~0JCS#fYF?eK6NoZTa{xh zPR#ibS7i9->Ip%~xK5>sD^C{aXH~~8bfXDYP!F(BeAa@!^cDigUpQiL0})QeaTt;V z>}V6In~m|+#~Odap!uUs$mRZCEc(GZ+J`$z3Loes(yE`e_2MEDMzw)*abvpw!8LNJ zrrPe#CZTm^nvP1fluYqq^AH9l8!6PLNM6ehv4h=%uE|-nqu}yEQAvz{Vzgnn0zKz~ ztwTxTV$1I(H#ZyH>O0_;z;ifv&eQ~QHE8K)$^v#NtCAK zPaEcsC5O4WHK9!UD>A<~HO8U7%KqOB2~PO;?b*vq!*@#yu5*cRLPg@1W!itfc@g$$ zvN`O{F3H?FiiC=~TBfgkNn`Pa;lnucf@9xo*s(%&saCsTnN(u$Muop%QMZTnALICe zX*|yuwcH+WT|qTMjc^s`rN=Tz`oO)-@4ygZ;aD0xA-s!2cvYoq=o z%`EZBQa%|)4pzW9!}qnq=rCVA8HMe4?hun?C++=F_>{=zfZ=_`ogN5`*>i74G^Du% zxrjXgDEwN^55QzQK*Y8UrCseD){U(O|3a&OYcZ7~B{$6oE-zn8*VY&8y^lvmK6?#7V~PPC8xApN@O4-Z<(SOwRRPs>>q^ z&8=)90+>YSY2qEjh}_4dJyN%HjWb4t#ly-!cj#Dxv5~^qHQq*A`t%+BF8D7Jqx*i# z$maA!LDF;HjuNGZhJ|2QMtqll>`6s2s8d-DpVgTtyN8^Llf> z(nYVc4q|60YWvzC@?T+bHANWp18f(Ir?{htr%Sg& zrc+xFUo}i>nP2n0PV-Xcc3tCNO!39BAC?$N7fo@QS_n-}7Htyl4MOBQe$wr3Tm1d> zebrEP@aF1;@rSAtHY9FR{~*fRTYOODWLt&~el-&lTxE`l_4?sNTYF#NLP-2IFAeRDx(bF~`uvtOlFl88lJBZrk|qHl#`4)=KOy>FB@1WQ4IH*OZ$=VV zMb4SQL;@`5jVIzrw7}e~(Z2bXG&EF9CDSv0eK`;kNHtD;%yj)hB(Osw3d)AQ!Bdx! z40*i6?rD2!F?TLz@bu~FvFtPODZn{7!w3E}W*2jJ2f@=DiLnd-)C}ZI{FXM6?l{D< z3vg)M7qG$dnN-(YEi}93g=FOAmzr_87k$2 zaAZD6)7eVlvxtku9{%WSInPdygJz%$i&NLrvn}PA@p%y=tCRC(r5XQYI8iTfaySW; zI)So>V^b@rxS)0J`@Vl~dy$WmU)}1l*^#ch_@(*CmreYzlwPv8cIb`kQn;kR zA?cy8TL#cI%{cZz(Bga>y}(TFOoCCi%jD;VWO~bh9iOXZN-Vi(N5a_%hpyP4<+w(a z4`xHhn{&8Hhf^i`HlcTq!=UAtF9=ecxMXpq%2(i7n}n+5l%I*qZ6p5P9J{iutfkcS z$sg>VBuvyno}8!S>6c>~4;HX5V@{ADoadQ`46i(4Y>O%WJk}PFnoqr@PDF(xn;NZx zG8l@tDCH%bkdD=KM&S78uH5A^o_vaQq~TkH()XiE3_D=qb( zvOy^M9J-aLS{2AzO=McF9|PqMF@lVUrFC#{1@+TI;7S>hV8gulqr}r}PG7IwPZ)ae z0A6Z`7Ty=ZIRQr5bjEU$nD}0VLA$AR{dbMKs1FSdS8Q<3(}0c?X~g2s&AJNKm>K}C zj0nSvoAJ48R#_E-nLpHScMsO*s;|*GmeJo8Ar-hIDxVL#3yBaLV|3!b7a?9{BgCUg zGMB^qwf{y?A;-|?;b9WV4~eL$d*Lk&xVYx{sA~7Pt0=wvrbxl*J}Y&JWJpz(3>_sA zNN_42s?Kb=fCYkEWTDG@c>q}xdaGMbdb!cEJ;d+w^f&15mxj;Z#FoYn1lQXw8j3P% zF%lsf@NAfg@$TEyECSO~BPSW=@n6^dfswt=b$Eh|bPFzUl%U*r!|^TdX@^d*boPst z1LZ$Qn*oL4L!#Tw^u*Yco&~>|VyMXDGBO-L=Q}ssj2v8-b)Q;v=dZ36tvA=l;rrb0 zc<6vFc+F>=oD1GX8Y8tVI_(O&Nt~&p{*GNAQSv#8ne;20YhF`_X~0)TI(Ir?%y_MZ zlVUvQN@qodXMt*)*#C||##Oussm>Iw40K2?TOqg-=wc@kgy=0-503gd+6K5GAk5|V zytM7m4^8a0Ot;3Fsmz6?FsQCqy-+I>ORI%yrtiuz@{&;+c5L5=r}^!l4sAb>SbfIF zl4o!J>?HBjw&;Wg%7Ol)#|}ht*T|r9T-ndp3~8{Shsj85jAV(!_`0>f0SaC7y#0@p z^OsrMgja%?WBSM+E!vkK0uNf7iwZb^er)>?H3znx9()zS)fO-FmQ-z_W}P>iBgqOw z%vL=0FDPb>xLTF!U?sgVuV~sncZ~V@DOlW1!k%?`Uzz_jIipf0>ZpaUCNCYk+p6g6 zwE9BpPeay1kYKB5Odaa}1+|&EMV@OUj(X*xQyG#;*7kww(N$5L;){(B@)^eSfaup% zn(xD=Sg>{V)iPLc5*q6L?3r_Oz4;Y|w(v1vXY8GIYrE{7wW`z&VO|Gmzij%>9N`j? zNhnGNSR|{(R-21&lNI?Jd3nUj=nh*iAD-u@IUQ zB}Go#F!1R2OIbl;hMp>P4FX#jWax;Ml<^4#5`2nNT4;@AQ#q=B+_*yaPIj|&diVEeghDD#V*1f4Mgel&v zoyzeH~ex4p7xs3QmUn{-2>47eFOFKRGdxvYhdBzAod` zPfC!BYj}=mZ0Xa5eQ7GKgbu;@?@zti>c{q{3>R<$+272R;V6G1Oy^cWStxW4{M((y z3xsXzYhu1SKzf==);g^B`4B>3qpu_?f~`n}_*scJo*|Hxse@LTDl+Olkof;>Y!hHN zPN=;wlEFa#-g6?pG_iW>?)PkURsFH#Pvy`sjlOnO*J&EonOsM1tq4U{3aX@p{h{ii zft&YXMX#@NBQ+nqdwsl(jEQIg2p2^hi?KuL`X)lODo;~?6`u(|AV6P;VPXF!BmAubev*H5lXJ9$&tAMRaR0x7@K2IW-Do` zPR@bghZDfk#$n9EgbC4pkwA=W=!rN}_01Pa;@2Zf{@dDCL>2L$1yH{e3{{<97hlpJ z&|Xs!k$UUkfXY6nQIYL?c>wAU{C&0`X<5m5=+Q^^C4 zVRzV~Tcr{)KD-~x5bqZsx{Zv@Uy=jZAz7{rV7B&^8?D1spFk|rnU%oI#K)HT9iG3d z6XS0=jd851i`S;-jZ#{|N0;*kmBo$PIu)}3M_JYKtL-%08iRf>=&`yIR8B)FjrgEi z-q{R58w8$3>uL~#_Td!r`w)_m7HXHy+bvb6O_!_QnZ_Z{*+`UBTR*pcB&a|s51xh# zv?QZ%#8R!h-JkE1iU_*=Jl3*H5NkP98gCORIF$|XayMTBTv!CeBY&VMYq#vpy*ZRYmZKU)S|03 z&J}W86ak)Gb!i<;TpDN$HC9Xxj>ynMH6CX0HDrd$w1cTPr*;LhKbXLRB(}@753RRM zp!k+FA|9Bdfcs9KgSYywIOo{Q>T{EH5`noQtouXDeK)KFvi zCEu(CZsAVWNu*#)chobq7fp;UO=KIty^5o5o2%v;%i!XTxv6MaS?GFCj83{)D)zNz zW!H;CXM#e^-*MnB1g|)Eo(;a}@tUdU*k5glHiokEnN$QOvODA{$QBjT&EWA7sjmjV zAQz@``#v5RLl`(=#pOwmxn>7Odg2g>xI-&;(=lp-Lqo~1%osgHfd~hO3_&Rc0o*rW zZv_4(>_BDZz5lkIsY6WKkxp~Ftph-D`H05%Yz(p{p@GX78XBrq=SUBaL!CV*B5t`o z8PKY+P1RSZsHlKh3{z$JUd+OMp~V6K@$+rjhh(DcaMgV00VcW&EXv)d}BLj;cTC zs-tMViH%BnZg z1)aRA#1iP|JdXqFZ0Xo%QC@bK34;hC?TdYvKw8W@Yn$fK%m8B4Z=EYVz+@yH96+BR zO$m-#pGu~3_-RjWNjcabP=#WyE`@pjTx5q4o~|kcdW23OXw&%q_F4$;eD^regGc$g zQ_kVId}`|tfKR5xyebyMZaz8pv(?P4d}sM&`vrUtu5V&e`7BHZQ+V9p{I@L8A85(J zI=ew_(D)wKp^fX(kh#W4?*k_w*16n`&dJQPm(S1r3`MV90h^s5ssRAJ22k4K`C zALf}tOV-R>wgDZJkR?Ovjx#+Q8(s)o&(*=H_ zg=f;v*X>a*U!lTVw+wuI~bUv+c0DcIW1?RSjj177Iv*B|u5 z08pP^L}YIM@!#n@`dHBRD>@)p_RH|ffHKB9O?CPJMSw^f+}r{nHOHH2}V9IS+Vjb3c=llfRahI;azoZ6;T&lmpyWvGVsCyZJqy8PBFnQ4by6 zq+h?RBeCGyFXAKqwCF^UtZag>9Eb%M2+8w-QN`x2$yo2~AI0nHg|^R`WVI~V8rtb} zwuhpKr{M*PbY*jK(!?4kZX(G|Qv$q8FVtg@te@XJ#+{f5`n!`=odC$a#?(W?!{>hm z04WBUxOY`G>PI1`2h2(-MsmI&Pq;dMy*Ivgpe{}wF&vxH{4#qx-bq4oO5tV|Oc!hf z2iLD}j)jDUg#q3cZ*6@Yw~OHKQ4z8|{fNJ94tmOy;HW{Ogtiurgdto~V6;aE*jm#+ za)O)cJjy}2Cg(so?1>;4uN^Rk5!{?7j>bWbsy1js(44V>r!Fq|?(hN-aI}uaISe`z zw61tzVlR8HW2XoD z09D{x7^Qt#Y7cy@FeC$y&MkH<3k=(XZ_3-tF!Yx|z7OOLDecxvjhjx1>0cZZ+&hT~ zUy_p<%Rk|PWbVtD<>*cYF^jTdR_-ccRfAEuE1z5bJUic_rX0#(U9av8%IN!?&&aIY zhjZ(2-9b23s6WADT zlqQsQ_Dw5h6!L8##FCe-`$=T~k;05)PZyt?XM_-aKRiBUR4pQxIuZq2U0uD#c?Fa* zM`H@KtAHxe(GiMu3|LAlc^?cq4FCMrgul%i$XW&l_8j=YY>Upx$=Ow)UbuG;g3&0J zd`^`)s5aJjOI!aM+uau{UJuxI%h0I6U(>r-W)=c7gbAl*m{BHf56A>G|7Dbmv2-SwVx?|bj}`J#s(tZU{vWBu3qtwhCli-tc!aRhB$;oP^QyerTX_2wBW67UjxZ zeyeZ%=1?^B9exNwm?BsZN;W^UPUrbJ0ta$6rs(BLf?b{+=3;}i_y>4h=BG|Xt``uZ zUvr3`fG^& z=WYLvZi{?#hz~L5OPG$CqB1A%ypBWZZ$|$O0e)FZ^n6-Wb7~OR_{2BzD_cp-V+F+0 znlYypx_VKaoPe#mL!BkIn?X-WX@VBvulUd9K=y>*b<{+u$*SJv@1mKm|BOl**zFiF zUoT%zWb7|wD)UTYdk5q;Qn2L#P~ly@1FthjzbIMT40@4WbxYD2*WstC*y2e6fJLL- z26Vp3UbVlXi_gt_0XQ@D(htF3LhtzI1F%naeG_ia@p0C;2UCeE*ctB!b#Ypw8walM zweGAXYVPFzR&8Jzq@APw39cqaxu58(gt@X3P{^X*d?vI^K0jGFx6l~2>>j$0!P53{ zkqwW@p5dTfTwHvrqM`^&WPqtk#K2=Q1mTtf%bS0AI3@>Ult43aEzWB;Sq?KTVsms^ zEf#Z<4lu+ra2-*oySv&h!97gWHhRA~`eo*}einzP$sf#Ms|sHs;MC!gP!4Q9{R&em zwWgm{Bo^;D)U;8wP^+ds5M%Y2BOO|N{3t;BGzBGJ8TI9WSqEk%_pB`$M)2lK70eNF zlNgd92@7(!HpyI$>i=$GL<05Zm(wU@WITrWAMIR6HN65FOm%u` zMAyRI37(fJ%1qU2ONH7bN%?0L`>jcTZlw6i1t5Gcg?$#4F?PIrTihS302GaW$@jc| zsL8+!-@3auUj1cc+!aHFI-aqA1+GvKq`Ez4slCpLkGDSs$&aVC5Bq99ja{-PUr+bB zG}UJ=XMQz?vW>5J^`86fdDSj)r0sWPS!XFa>HPW@v;HU}>em;g^76UyI`6(`bFZw@ zHOnKc!7<1+Y$`4}IUM+EOt$`ftw~*A`;u;o9)#FQ5|P-}%o5vbvJ(#@NI1&O-)lJ2 zdF_-ZKlTzrcrBx-ga;!C>UR)5h@kWtjvihNd%A4~eAj#?ZwWc=q0H`J%W^u&xx2CQ z4{FKI4y{%S=Z*a!Z6>H@Y$C)vLGj_R)D(V%M!%~wT7b%ykiP4p!R>5pkvXz4j$3NJ0l2lI$ zjWK=9k!KG{1&X>hii7VzHE$m&4UW0!kxdlbmZ>Tzgp_dB0U#&BOKF+m zMv6FeMv?P&&<45pq!;MKxpzspySY`9U^{1lt1I1+sMw3QH2=KY3~;x$P(4fGziP`B z4^gpVQF34Ulk;0eDY-JrXg!GH4-(ZPGzTF1%Kl|23w0EYdCmJ{cYCR;TaeDfiiq&v zoeAL5|7&K#P*mr|A^!fmrwYH}X9`8#a{7|+*Z-|708i4XYItR7ZOl`8>Rq;LNzucn zESbPu$W9TSpNJMDdGat3?I|^--dGKD6^^-U2b{zH7oG_jb$>O}U{|DzJ?)iri!-R<|^8P=}`~~!&V7WzpN{RK*Jk5;)sj}RP-?k)l z=6>pGUFINfqZKU+Rg48~2xw>vj)PcQxB_wv`p+D9^#HC1kz=U-rAd;X%`@X_!?cm@ zdoWx$>&PnP$_hZ`ZN%Mutcq zOL^@C)#`Byx!%NN@PEkU7Jr{-81I9Yur3H8*$t9ytW~+=EhTrZASns?M!O;BG&93; z&(k9K&KD9bomiW?tvuajq?$it!Ag%7|QL$Q-6!wApu>?NL@+0 zWWxu6DS+zF#DqT1lAH{VNyP2XpFfKY5B-r*Vc^D~Uk-bfowU>pR^=IxhbRMu*&ISU zrP*coZx1ePcQ4brK6KwVd7q?UJ@BBtn715hRW}|vKA5#w7Pr%vW1`&tIigcd7yUYI z_*E6Pvgdxv8D1*(fa|4@z!+&-24(-aqy-wWJ*0>?MfxYN^ws+||7Cx3{<8N}DDtlT z_9Eo>=?B+X8DxxQSqDQn08!91kH6R3V7xR_0#*Z6nE zU(Z&H8}CdM&p3r8?Ak<%^a%vfKN@F21@hLPH-nxmk2*EUs-B(5R^ekGXC}KC-Zm#Ty<7*%F^thxmC%;IZA|{1yJQjTJg}gs2n{ z$~>G2Iy*gIw~_c`L!VbkvRNrb_jvtb+8vx3P`K_83L_cPkWI9Z`cvKx>{iSme z-ph2wEln1q__4+@Phudp^3VYYI}{zM#JhHh18Ks*NpS%I_8HJJadTO>mTFV1$;$z` zBEewm;^5-y1}8_SUYfV?+pAOetnWU%Hd-)p3W_Pl$zRKoC&O|GzOdbDJAe=`-5d|q zo^=%?O&YPP^@Bi0Np^PIG^NXZo0(d2pPL`(h&jC4{;p58a(}mfUz3Q>_6mx37s-S_ zL^j9SH#F2cm?6Gf-{TTP!cni1rF|cq<>Too>b^5+LigtFAok~*nCKdMux85J}p&|f2ect7zy&&HKA4lMwwvh{A3|i zpa)mZ@J8oPZ>+qAM|%Ci5!ZXqVa4NCVLC!PQib~0#H7L0<$mq_${pSi))_vv9Jnw3 zGI{}}lcWX$0mjvb@r7?aJ&u)rb(gpi(h3T7^%l6Dcef^M)HszW?WvTF%(;0~gmm*( zG4t?JODg60KUN(0sV^uZJ@t;g?NcvC??#V543JT0dh@{?XyX!4^9vf-L)L~w%9=$a zE~fY|qbE|q%=t{r1ti(sC5q-_@34ljr0SirE}i));wMsFpAL&UT^!u5o>KH)N0RaM zAh1HKzmk^4=P&UP0gUN${qWLYpPTnCPZ2i*ayKAo2tfvI zfQS0Q9F=ap^XKRs~71x@@8#)3d?29FXiv&pq=VjYu*1bulYz45Y)jj$ApcE^MT3VM=JZocar* zjJAiy+1VLp(h*Er8ojh(F+{sAO+WdOSGTl_u0)>_)qMTO2ZL+GKPe z-|zl1cZ+=Xif-rkkBI2T!7HDnp6enIYy{ga72RrcHHgflOu&tR*wzO4GKhN0>0W*N z`tBYK&!JZF4?1PI(@do1n(biY)h}OA^}EurovJ_=p7-W&0dL7n8i%(R1sa|?@XLAZ zl=mWTSSc#s$LG=;K@6Z6Ue;_XLz_hAvfp~&LPB$tpHmkl^b?S%C56$Wy_8h@dp(cW z7J;Z=BUP6^(K|1`xZaUg*CZ+@0)GWlR+IlJ85Ao)=V63)&Xeoc-V*2|dX0VOdmx8bu-`NC=jAReUYhisz%A`(N3PDn z%=}$VS77ch;5?(o0^{{&WH9k}kTsPVq1qUkU%z6{t{FCnvPXM9AwEKHwgKG;PFv4E zR@ox9d`z8@YaRH60=NUt+79@sI-lMAz;UC{vT6W5hsw3V3W-JKy*It&2YZFP9`!r3 zq_Q%yY3p7E2BV)3ecz*Ze^_}Mj?T0P7nLAirO9}|+BS^N!l4?%y7$JZ&d|HJzB?b= z=LnoAI&ED$UkzxbYo94}D@ZvnFQ-*LO|<&Pg>D^5sHNe5km^&T@Mou-t4Rpo#LeQN zhH*R1u>4GLIGTXM=WJd^ic>w$Cx;Z25b8J-|^1_#!6~j!t(UNqH^D$ zJ>PaJ%n?q+$M<(RTSl3jFCU9ZQ6Qx}(&n!NFfGa})KC1L+|VnW_%?n?dU4%;80IVW zXY%T5zxRW6aICj#ivQJNFWpS?;4_vC^W|p^qKplcJ#_ORAvmM+`l`5ckgv6KYEF-# z0Ygi~lCHNG85J_gzU~{P5uO<+?BhvWLAN5>=b{egqAxGHpLiJke#XwkB)6nFvV1bR z?;@AzS|}&x`St61U3|na~2-wGc024Rl8Jx+ zV{N}bw;in_w8~LszfPBO)ur7n1s$z=t9W``Rm4sdDL3GAYXTRw%K zb8M32Z*#w7yYGPOZh7pVG+xC%M*P|L>^LUsU4=Ik-G9`%Hu zQAaQA_`5Etg|F)N&wl({Zhq$}OS{m{tzQIlI?d{p}1?u~z#BR0MuBR{ewT;_b$5(U$mobfA z9Hcu=q}~bbs2VUWF52f7vi3*4?I<^U68uJe?VVu9iJ-~&gGqzbWG7sSLu01+c*=pJ zcl_CIT3sJL-0V6J4LyOxkt^{@nXvNDas|m7(2`*s>-o0_%`zF|oZ8r~y`Jf9n^s@ix_p9I?^v}{HGZ!>)N3L%J)<$ zNL~2o@hlC!to&y=YHHMOg%#N_y0lHz^bnX?XG%&bGqWdEu>avKh|ItM?Gh+)JVP{{ zJ(1rpxPGoRZQ0SJ8Y|tQv^=&55S~E%W|8>5(7L1%+H!h_mLrTPp)P1c`A<5#@OnCj z(t};xmhuj}yNg~ay>fc~_sJt(huvf|>l@)ZaTj#P3%%qQ-suJ5!2#>{z*yG_tay=t zKLB7z-mcgu+!rZ_5N^Nli=eK^lOG8z`s!>Go5x#z7xQoM=0w8t;y$*cLFx)wCqzf2 z7CqaGiB{s4?a+*9Ct{tWQ}K5l;9)jFSU-x?6`;=cs8xCTod2CYu=}} zenFtsfCs21@?QdKm3b+oi!XO$X>8Gh!d(OKXk#99?e*M8 zFDU!qm3jJ>IUgP@b~2Oh^d2;2BZZTO5krp}64uc%>U}oTG4Ff3!MY$|aPaon5PDI1 zuHm7grqqvCaJoIgBqY>x=WA|(SFx>A9y_`B;WlXklV5T`tw*S`v0lh&Du?*A{CW1K zJW8eEoOAFV(m!GHoL>oYd_CyjBfm`~JXE$^p?w6#qzMJtZj?~)@Dy^H_ z1zgTK7<@UJ!6||r#CX0&4Apzy60l|bSoa96e8!;kcaXMx9 z9d+EwT*40SB*?nDai;A#awb*}<8$PwORGCPw;M>#z2i zqSIJK54?-fhh_7ab1jysqE9-Td*iletTU#*RB|ocVtpx;+Ni(0D?b$PIseG7jMyv9 z3I-O0`NnVREa*9%h&v$Ek!1lajGV_0gTD+(d;Ii}e+ZS4S7K+(H6<`{7k2p3-n(C* zc(HL#;;Ugi(n^KZ5(9Qjc`wXrV?W#cOTC1#I~|I$%k$t4Ni7`nJyTQiP(0ku zLUj{zzK2jI#bA@G6Ii+jBV8x2v$@Fc(awR%&~!#HxWxdu?h;tk8z1Nki3jA$vdU7^a{uNK!wp)m^3G?* zSMUt3QSK*LGGnz`m5Y?Ns8sLj?Z4TpI?D0*(4io;p+v8@nEe#NnmHL@)8JHwH!v>#WNWKJAO$+1k#7+rzoM zf_KLT-D{!)+iNZrreJ;~z&UNel`_9^;4%6jpzjs^BcgCB?1wRscot+3Z2Dv7pI>g_ zS%c#WqjwHo@LUkiZIMMg#}#9dccE*49!Sk2FcqZ+aksi|)6&`4p+ixnTN8e5UdO4f zMMy3BAX$P!q{lTNPdTX%`qJU)AMTBG)yBni>Lj?L%J{;8)0~w1cKz?*7tT7_+eV&0ryQ3x*>5^4YPDceF%T*o>!1~Cf z3Kn9^Ukj;(VYBh=HNC4YVg(+fQg()OyLQwGR5Ho8YBmj<_OI?&EQ;S*^Ue88Xgmx^4q|BO(5HXVfe8wZePURZcpYA-D)CoSZ1?k%Wx9C5Y`QmgBVdfHzYOxmfHPwq%0!HFHN;t~M00KH%q z(@OMqVsU&w(1ls4qn`84oVC1QzrPvYZ^0}O&LSGxzr+MB5y4e{K7)uhiHMTK>*nfN zoUE{+b*b~F^E-@KJHJl-nOc|}>9ENo&SET0P42o1-{wjmc-iZ2Y_fx#UB@z^@Gq+I zn`c%;64oDy&J><_ar}Q-I2hg}@rdRrFncLE9lB~!M5(zK5 z6hyKyji$j-%Gi3)DJPodG7oxm=HC5c{`M9UK}gXG<~#6#)K5+}4D!8T#luaWrAb#9 zBK_8Z(j7=aL7@fsVXuP0o}()RFSM)o1vPX9L|kPkoj^?o@9u#9%qiL(?n4hFpiz%J z`@S)xOA(jbSfYwlAB&VmhkAB{QPTm>t9N#YOqI=cWGmZcbC{PxYP^4~NIuiwvRn9g zj=SIMkD6U(5qaTX+>h$F6&+1$-330j1p~)Ps|^z!yD0M-7Q%}7j=61r*xLMcOUvWw zi!k~4`PkcM>dqIl=&okT+E2z^mVm!@F*nvPbd_0O%|P2^=DV|T)l&Ow=CtOL1`KR; zHoTSN%b{kNA4yByj_{e=sRAWshnhZ>NzbFj$qHs@krwi=Pom}YUs7hDmBS06%39>+ z3t=P>U7G)(HU1IBPMDHCjExuf@XN{zr8U$5Z&JcZ4!#>UWy)^=;?PTlK_uw9&wy$+ zQzM!~-#Q3rTPfa4^6S!_oBj>@7?YJ47YA}=eildXEa4~;X5;(K>u8Y~k7Ecr7|OE) zgq7I94f^{NW##CSn|WJ5XG~%;qu*&x2~> z?~n*d-dK+QBEh(vRC|L*kR4%iW+q=;ME%qzI2lSWHpaRB zv^OfyR!|0AICuk_m7vLNHUX6PsMWna?0xI$GK;DZP@)N2+1|1*H&tEIe(<+JaJtIe zcIK0nUVe2fT#JQ^;ig!P|AyC$!l4(59MjY>RGHW4f0Mv)U$o`xv1|XO#KHQoSQ7mV z4;Xn*^v$DSkuA8b$GS?AL1x-EJNBof#yTg{T);tj^B^u`_peo85jG!T#`OU?V0*Q1 z`Db*iT3bTX4((_YdNrD%3}Rtjfe?_&y2-OUxVybBiHjY(7g0W>&v6_!R=kEwI%jss zbYKU7q(R90yq)&@}c85{O;iUag%e!}ZJf&yZ z{Qx^_DoVnAm74|1X#iRAvr@#{7e!`LcyFMsZH;GVB+h}MH-maC0X0X{CS9Q;rZuT~ z&lfvVip9_nBDLo+nud)V;ywEs3lv)vGft8sx-5m=Ts!IUg4n2pg-mv*$%~D%BL4m* z-kBnL|I$6W04Iufmazz(kKED^B zm?PS7g?6sBZRP95-5EnhF=nnP21^UE9Qq<`VDn1Tp^Y;B5SNJF`0WAxwx_y(-$bWQ z?b3v-D;^l@7n46-oCjLm!BQm8(vaUBKjI|4gc#|Mm+0>b&D7X+f}2{j-_0_e$JOSS zLmTML+-0Y53iYj zIyHT?p&91P|DTOxj>Kkn!xHW|YLJLt9fFFD1;9UaU4ZKTUVN_(760+cFmm}qfpXeGZR;&=>+P1)hs|V;;iN|y6q*Cnvx0qXxJU@isVjL z;HgmoLn13%Lcj;@#!r0c;YyC%(_Io+9BAXj4a@ubn8e<;iZm7{FT1e=nJqppu%j_a z^h{wLpEAd|4kbCE#Cc-JN#XznL>}MKFM%Mp_s9Do(0UctA&CuXe1MHU!}Q8_XnWhP z>^Xf9uh%&Wl@L-SS!dK+e^sTo&mtiZD{`Ip5!SM%rU?{z0wxukl(ABC)n+kHXM`o_ z@BZsx0CqO1fO79gt4XE63Pz$2Hw?24^hrpuhegN8*fxpM2aD+CT*QCWg&_bvfcwHc(4Ff}41iOeyF-jWTy*E!@+A+5S;La$6Vj5myJ;qPanh597GOBwXbW359vLL z3En6|t4XAPuxsna$x7DTW@djkt*s9fW{SCR)mC>LHemVtBEwdZ%EDE~>di2`^h{SY zP7|wS)#*U9l_KLJu$|qdf9uv!s1tXsx*M5y?cz`_LwZ=~c=~}$>N-(qqaCXX^b2EX zJ1&E~>hXa3x|fEX9LEya?VIX5gBWG>2x`M{AVX&bF?i&|IcjBf0%R_gwI0*gI3=l)#iC$@{1#nSQ8KWJLf(V>OzW^Xe zz-g;ok_S6uuBx<&Q4clzq`GYweXVyaikO^ip<9Xrd+&m(Y?FchUXFjdu&PlfXVG8R zK5H>aoFNOplE4NhBN|qMOCDUv#@CBe%rZD4{Sim@;L=k_QSw^wo!dHx29?A+{^?#p#N6!YwS7JR3;_qnB z#~aLjIz+r*@x{S9M9=e;F-$R1R@GIynSw>wL3hmb_1dcF2YQZkb?LtRqjYNcnv$hd zdh*7uBPQ!3rRTY{U$%pNh;bma+J-R%83_9wSFrAFEK19$suT^Gkp}~J-2zHFAVMOH zu*CIAUNVv~!bb?VyiUXkSno)j#s6VzyXz0!&~) zxmD?l2nGLm`7Hm|H^mFW!073ciG%kLip;2EQuct=z(JAhMZnQxl>@Q_+tQVRaML#1 z@05#ZE(FKoKcnPZpToQHfaiRik zQz(Z?6TAgB4VqdM0F@`ILX3t-@!MRFrlNvFjm-$D9a15W8M1vtI`_uGG>v2mfH zy~ljya5{~b{b;_}#*wkTj*~1%l^jc77iNJ(KxpPZ-t52-V{sGVqprSIM6+)jih+v~ ziiaMW_Wi2NguAEzbT^=vT!cXAOsU;OyKP-XXjaXlz@=ARLO*;EoOBV6mIge1E=RGr zV_kS(2zOc>Bk7kfZ%3izWgYVezYNBtdhpq6Q5850SKFK3DtQg{^SJ0ne%=l^QXZW1 z4~T$Yq!*JW`!*1eUY}gi5Um=Q41_CfLv%=PD#*kxmdR}nrgzT=Bu3R zKO}AU{|m@XYB0?<7gVbw2C&yxyU1I=`Xe6Ei=jUy2iQXtNWbHcE9V2k+@ZDzAL7OG zhz0WF(@Yq0vPsKU6pIT*!hkrqQZXsQJZCjgcCa_1@gtfD21{R|5=S8c!DDwB{r}3fG#2i^DeO7 z;rNA)QAKx}4NU}@1C?Kj@Pv3aeVkT@W`X$0F8$=spN?v7G8TS(j-73IsJzg=i}sB7 z1x5LqXhk?Zy~3%yN=%fLKuY`HdkY-4W2n{L(yR8_-jz1M9d?CH&-t@^2DB()i4Km1 zk)Y?r6Z?@Zv39*Jz*3TCiQ=e4p)W$Nh+o(kX`3wPCd|Ov%phDl^q87zD^%YjcWv8* zdfQ=P?Fk-M9|~eKh{gIgBOcC3^>93Vep`C3aZoH-t!ruf#){dXhH&!Mp>mc=dQ&g^ z_l{XefP%E)*Qu82=$9P>tc=r9w59`313Nn6f>Z7XHW<_0hM517Em?)4G<$t97%g=; z0(Zr(@#musu1NNN#(OX33}%LXMYw(vNt~ltv889>+s;8bjq$v1ctHfN%sO#X<*hiO z2@Ki=bL0bk7CkWkc&odI5Uu3(d;^yMy-DG|5x7uvv@mOIa6{q#nl*+-@x>CwPX~=a ziuh>;cD^!lyrTGGHp&ZG>S_UjY8sO>#f**OUrd;wjYKg46qglBHR z9V)T|o7s6cV%rKWoU~VDUbQexo)W?ff6kkSZ^3R+p?(FIuBYxFWHDzf1%-66WY zX7)X=97^4E3=euVW8Dj1z%|=*#m)rn_=U}|feoic`0dJncG!9%oz}Vc=wxZ7X9=nR z@@e2EV^0JP64Nc9daWrlzZKwQ0MV-RG0USP)^hy+9?qqWpbzCLkbb*o<^DcY?1r*_ zZka`oo?7uzC+wM?{2=OIwxV|@ARycx2bg;?R$@*rxA&X2=>?9kvY+&3eM(}g%5Hc< zVMVwF2M1k589H&A*gkk}f7?zadiqn^nmoDZeCobG3%`r`4}V<*+P{(09>oaY3+Qyn z;+g1YO{R)u+x;7XEy)&t+J74Ki}i~PMzKTEUR%_P6>udYy-tIdgcZ4?et0%0D{@K> zZ-dg?rKV&;AI73S2o?7-(<0chJ&?n6`g5>x@*NyRj&Cw!K^f3z;Rzv}W9=#H;N2YI zN&Hc3kE~5SVS6;KWc&CIX>9bjJ4k2y7|)f+^mI2-M1674gC1e0|DUB-JxRv@|SN=Z%ylB z$)@k~EspXkcX8k7w@%A(O#d-ZfQe0`KaWqC6-X*~dnoxYED*7s#PFq3`oFV?BSqHq8ZgVO z#IeU$j>eQrcq5@0GF3atdE)&vr8rAquzPtKQr~s|@fml$lrK3!ij~ky+ZLYol`RS! zhYp(U%&KQo#Mm!kBD0(}OSf6}!KJ(t-+jo9Io_c-z=`IH4Z`MaZ^;x}u1K!+T?W2(zP%%bcs(;E|luB+NfO zU`_(hw^QjYF;Sj=h%7btf8w0Eb|B@rkd0_3Qk&HNWTWU}p;8VQ?u8<<%Z{MH6X2c3WzfUxtnk7nE>&*`buw<3dFj` z7I{2*FHsXsbl`a)V10*#a5pBj6G7CSo&5n34cr#ngADzFeYNp&B-VFD`9Bc>uoPwZ z2joZ4w9N_o@2aBe-(&CGk*T!|%w*%HQWA{oM2DA&$qNxPHCdI<)E*FJw_q2GS)m6w zO>u8#$;zlaVQJK9+0alc?zIk7;KU^CI*RA2yI5{)s^1G3gBthiFAdN@jVC_+^qEWF zz$W)AScN5wOq9AUN%!P!8Z0+X9Ml-SO^E+(U~<5pgsoP>_p*OE?uXo89H~@_d zr^Yvo2lYe9mg%7VFePO!07M+3|Fc8@BA{M&%^yvC&Ux&Q#mLzGNZk96;q$Wap`x~! zR@9j+WNb3z(^ZurGD75ssz{0k$WeUS$Hce=Drgl$$j<$VKA|bKH643z)anP&AGx=h zE3#5$h^S+Ze9v**&txmO-;l8|@xxWh(1|C%5anF1SyqVV#uH@h8t7s@-6zY46z)2Nt}`E+8-ghOsf7z~oQ~`86K2nlztd zDOV>SS#|nIB%mB~J(}fwRrZyqgFTQKS_3J%u4cJx;6JECbnc|#D0>;2cIA8(HogZb zU;^`aF%GPkB7Hi19Ot0;M*bwT;#67Lt)ul94~BgUMcpEVh4N;{GT;|CYH8C01e*O~ zOMCW;$B#nf(%e#?nq-wm9@A;^KZw2)uT+D$oE%6YM^H6E=``vqTa2GB$}>CFC8e5W zcBt%c>$Db2z6)IorJ8jY2x=4NCp!L=w`8IgpTxrCUlY+^3b61w(xDH$OBq|!vB-b< zIu(Htz}F*GMI7?nc!;s)U|SLnrIrW4iB4YUkX7`sucr+cRmflvsjn@d?O?I(>9H2< zI9?N7RLVRYr?ji*IV-EY!x~`sc572C3P>EZq1J)&$^P?sSETyF9R z#rl2}~1xbUJedXYNKN2TR8^ZZIGWD^v0p9V3I&ZnpwdDbL zQiQyv<^C#=Z@AMO8z&rx?FLvOTP)JfWKiGOmW7{aiG+sdZPQL z_Kt*b<}Uc6A~r6#8F@*?#h* zheH#r!5fnQIqt^jv1VZ%hwj@qphF=7<6bgoH4OwozWf8bj*Xh95g_7nnrT%!r|}3C)gr}$cC^2caoIy#6t+vwzlB+sYN^PmpHC6| zXszyNU`p1}+2e6rsZmYZC>3FC)|YLV4Ad_C1q&UVqu*VbBz)gWgpHY|an}yIHdhzm z6s)2qq04$<)F_TNJJS@QRm8*HgmJhP~WRLIM~Ac4;bT#rd)KsD;mBO#UDz1Dj zvw{IHbm-Um7aBhFNZXc#7=qku5m{qT#6OQ(sJPZnFX2+uA(uwl<6|u&60qO##xY7; z5I8rq@I5UtjWIhHjO`aQC&;Op+{NMlR#t<$_EX}WfBo|z$b;VGkuG4ariLGU83+U|~< zIVQNL>y>cR*U~WEs@mArziHcg60#D1RZS{r~>_&=KdA{-+AS# z{VdKMCw*OPTDH$ZCOj{U@i^p*nj@H~k@K3OWqKbyJASOvwk~z=c?0sOiqD^8gCw9G z2lS-~BHr`Qr&6Vy{G!WJ$>$FlLf>5Ry0@322{>y(w!NV#JH>+E>$hcU_n%}9EVl-v zHe1X*6%bI!mSiM}MqNh%X%?lM+C8bL9~>8djNwZsRUbdh8Em`CFT7>lo4I+}xD(>C z#Lp|7=VPik5?{e*DcEyt`R2;M}m3RMT)9sY#G>rupH4p{L%2d!~#fg z3hCUdSVuj4f#f2evIuwv5n2T$d`yxbdlKuFtxN2fMax_iv8|g7egraT3OnUNMX~*G z2q>kT-IFrXHw2b+>2OHwGc*5{LqHvvc7+eW0os*Cu+vx1reRI&DBCZIEOrqXS$k(x ze01atqs0P!2&{ptKf2{upR^EFEz+iQ9aj`%Eq6 zqb#+d8Y@5^X91coEh2=jST0f6t81AN0+8>W-o3p#9Hv_2-XQn}$|?Un!pn2O+=nje z-Of|imo|%G_SKjH&E|K@a4|b`^9kA7y(mt6lKM-M74<6WI>P zB^lJJlOQXcT;zkIb_WIrwilw&bK^XvU9TV-wM-l7z$#ToN){U{rrMSdDo33?TO=$< z@nPx@XJp3;r-0KA(8u)J;4ku8T)(QNLysWl`sVtH;4wC;pm52RTwn1ZP-J0di^?_A zaP28&*H2W684i`t8{&356$e@y`SQmBY?vr}B)K(8WsJvFp6rfpn3TFmM}{UQVc=AO`|%e6F-!!DxT8xm~;prLxSsr;|R zDeR#WF>L8MqFStqf_q%$N#xZ_px2a5M7_WvLv~}fzlLQqaN?~l%BfNFQVNcA5|XAN z%XihE=Fam9n?;nsCTUASo>F*Bl-!v68^<}QZT|kH`TH}0PPDVzYeQz4niBn439}KP zsPzp(dScwLwv=>KpB;h>*opqXn}ZPL_CR4vhaUEmF*3%|awJ~2MgpUjb|M^ZLA!D< zPF=u3N^JDE=I1NqLfT@(#8wxOWSr9^P)b1qxjcxHuwB6$e*Zsg;6=DmvwYU}Gk~Ea zknLnUd-8j8hb)79?ZHKiFg``$AqW0%=cT7ICUoH0g1Lm}-_o{PDhm=Bh&d~PL3zj2 zi@OBjOdyt!;0ick^jQ{Ch(+|Z+Ko3n{`nKs6@20eTJ_oT{A_j4#f1Ao%DF$}R z@L%#~H>|G|aG$)Mr2i9=cFSbi9Q6L~XWnS*BDmBg{LGp^{vbTTTQ?6xOR)05RJ=fc zTa)Mr;mxMzeIf!xGiC$Y=_F`ihF(Ah3dAm4UVW_1Bgmc4)mH;=S=WfHTr|aB z2O;gtNZanVvZdAV(rN#$o{Vst$>@VELkoejnJA*d)Q5nxtgaa7kq^6hX-ErJbhghM&m zV~$ScL>Q$;xSYm&T>;w@jjxMgeWxS^R2J>8^sp-R=)bU$nj?WSjoOR(`&;9`Z|8m7 zuf9m#vb9`v^Gt8nguj2^emo=Bf3CC^u5F?7 z;mr!xAYpjtvsPY{L@sw2%oRB`)EM$Grp9crpx_-@;e%UB?32A@#}FVf6ysS9QUi$G z@W5Kwu9EcqRk#a`>-9%r7BQ4oFvn@5A0>InTz|n&^n25s--!`s5mm&UOyBFo5FEC5 zhj?3_;6w{bPeJJSeP%i$J~4cAMqQ#jXDlcuSaj>=hP9Qa^59jCw{3Eh6Qr#MLewp~ z2woytDH4G>Z9BXu5jBLsCpt!J+>DeBRqlB1I*Z9v`6kTWqLDVDvzm^Z3;k@z=~HJF z)!l`ewsM>%T1x#hUWVNeP-__%s+*RQ!fm5LNzZVF04~7nn@_dGeU1l&mp|#-myhnQ z=7iC9;QJn+ir*rr!~9v7P)fN(1`-S%VCWNO{&yd|C78o0G(Gh9+4Du(N0$FQ32U{3 z?jL_}V&zs%L(iv+ElY+5=aHI3S6gKCWOQ^td(wk{1m5P3R&k&(3{=Nn3$+HnMof;z zf5a$No62NXGW?e;CTANg_NF*0;fW8k-uQ9J!&Cm5M8}=&r;1k^7G+~#wSyOB8mMha5Prpl(o0rjyT0-+k z)jfJD4LS_V(|9%`d%|Vpv54+{`OCN_k)pKa%KGLgsa0V9IL2?p?L9SLfa}Owaq_h9{!Q2a0 zJqR&3#QNxOi!0^t$4HA#Lm%ZDnc~DJ9#8F|IoqEX{UvLzuer26O2=cVj33Icq+=ZF_2Gx}RsY$%TX!@`bd9w9$*467xq5Jh8?UY;D@Cz}NZeO?8Dg-+9 zsDR`J+2k0Y_rZ$0L;?AAJdCmN5>_!qjsG?3^^yc>|v@6Acsv7{bk{zuUwmsb+w zX$mG3mggU}91!pY36vOa68Tq|1*|w(FV?@>CSbNv`K{4rjhS?)duQ=Ok#_vT$qYMVV^+m7V0bs*2u=CfBW~~0R_=3+tr+g7% zK}Qeb&`z~caPkLm4Fjicl5o?n|E-wY_&&MvRS|$RW-VET%${gbnD@V5yl+$FUEK3S z7SD3YmE{olll)x3#bo?&SEbbA$ir(V+z&*n^`j5el-l72i4whOL}7t3Lbf^Z!r(D< z%OOb2<8e)jsMvEYY`_(Ii46`f8`pn`Q53SkJr?Dx1$(6x|JCl17H#0w^KA3zsdME~ zqqo-;SH^PDwKDWA^dmbO+Sq?m*FUqCS;tcNYmnDX325O9l%$Uql;;9*kUBfbD^~Dw z?Si2zs{cXM52Odq2CHrPPKD6}GOs;IzcE#gkV|!xMAF$$tBqC+oV~Gdja!iqC3IR# zlIo#(LJ=ezj`V4;`MT`eDes2MK*c=#sv~XsHHTB3+QE}{xMiKmiMM`6S>AAN0Pke{ z18u{&?iP`OHYj>vF(m-sSGmDS+CMNE!b5*)Xc(%6wo}qRk(FV%f}{!ARmir3HkbZ% zQ6f;Qm0!DdP`CB12r%rv({pj*g5XnSVEcrz@I+3BGw7RH48oUeYYRYaX8!klHxgzd z9np;*$Q7i^Rpyb%A+46Vz4uH_TAl%Z#7b|&9mh)(t|IWO$nX$l&M&)eI0-Y@S>(a( zCn=?SNXN9({>J5FI!~Y#>xd`(;^`HbEw{#~u5P=#2 zIcG5B=y}7b)D3q~sZz|iOyzU(iM;#cL*q^BgLvaTrqTz*dMlAWIL8jU#9 z@6*8VXH*bVwNTx zuXQjgxha%#=JKolOhMOEvY-&%E!4jY6so7_LB$V}Q?Q`ChP>zjqM|CC8^T%-^8Eog zrvEQE)GD(Q8#K7G0d{_gnxjB-Z)37q?Bj842zYzU#fKt-5On%oSHBf-U6rKuTp@Ib zA>FS8psW9E(;IS=*U&^GbrdmTJLhxuVkY%!7yh5uZS@Q!+-~O#Tm47jPw73OU~=&0 z7<2*q_p7FCj)IfB^sDqnvJNi{*Qf0I_&a_WtcEt3`@>r0xqgmAA8qhBQkt0Z<$IQ= z<;ji-a!hU-I=mmB?;s1M?fILwXNb>A%sS{8mVNs<1A3mt{6C_BxA#g&XnWA$*#rq{ zwSom~b_Fu!b%$>IZt^{noG!L@D#z}wAF4CB%Y3RcQxlc7G0`8KH}JWp@Q_Tm3~Bq; zY~klpt%s?opfpWOFboPK3s8}^l!X55|O`Zy@xek zZzVhm78FeVT+C~)SB49?hOY4r(V4NiF8AI#<6SG@(yC{(bf%}`N*qDO(=RrcWl)RG=vVyC?u~ib()gLa|^CZ&xEk)A3?~)i2ol`XB`w}*tUIYiKV3*6p@mY zZjcZG2`TCBZpj7d?hZ*&KuV;$yQI5Y0qOcKe4h7xznL9p9RG0lzOOjrIDbb9--Ieo zzEAK6f;nt6<62p8R5p)*49UjEA1%-ykemH8kFYAMs!H|$8T7Ppk-ty8}j$){qg4d*Je#;Ua1gWXKUeNjLZ-HriGC(!|`LmS4EM zc#YA#OmqFbSP8Ax`*u^v5Z&vpyefv6aD6CLF(Dao{wJc8sDi2ie>c`ag^LzV+f!JqJS z3SuEXkxCJ}rl;=1m{g&w>+Xl)9uiY?rEPPJNLQ@Dw*hs6GcE?kSepCnf+xTA1sQ5- zsKr~s-%Sr~Nw(#5K9h~z5^XJ;HOI*mUuZR~>Mhis0LLX0Lyq1`g8=mU620F^f&IP4 z1+B0`CMp5K6kc3no?%sD!0~q8cPP6Cgm*iq$AA}1;KpbAT&@N?v_Xhwc8_SsorOn) zZ1upbju#uxEtK6boMdHx%}Uhu*ut`&`&c z@HB@lMirvTRc_d@t+uh)Q7+#8sK%-)!KlgAUsL3i;iv2sgn;^qTXY+f2Ah7N<~f79 z1)B*6PW;i4LHP%o=#Fius-oCmtrSt=$8=&E0904igT3d{@!DR!;=LeK!(K3L&aEal z9ZX|C{6bI7(ASm6N5Rq+jgyP_Wgjf2vEa|OBN!p^J%}BwZnI8(c-l!6E-vKntKRz_ zb!=JQFqpA7ZWHLQ|Mq+7>lkqOG3?%*`81v+*#>#S^nsW24;GL5$~m%4tNe$g-$jA; zazV4EG4omCts8GlMTI*syHT=&!3>ZTT~}aRXIeG^veDXl6GS}32a#!h(P>1;_1gF0!q?&O$MoE8*4m@=5x}P` zi>LryP}{NZ`IkI|dTbzgyhV8=B$cwA0Oz_cVDM#kZ8@PyrlI*i3i4~DvjS02k1^*{ ztVN>N-6fBpUo#r>mm*{_caG%YZFk3Rd)59FGP{kK_Z^W#0qH5;!ii{{6ha5<86!rT zg%)ZX`G6W}gu{i>j-|0(7HS*PA?Q9YBF0auNMwF%U;3Co^al)F0>dUququ+3oSUc_ z8C_X4F}z$RR1T1TiaZ5C%!l%?q5FC{s9gYhUO55@(gn2#psY%VuOhjuJkdLT(Hdo@ zGk=3zezV+K^I|aZasa40F5Z$5>>@AJTxZu^n@WmeElW~}_cVqemBdpN4=XG>0S>(Ew|vWCaH)ZH4taQ}<243Dq%ZLEd1Dk# zaSPE+&qQ52GK!FU_Sz9wxH0DdoP+YIO$?Q6TXPvje;>15K-FM$R zD}g(pJT(Ay?wPJvr^3Ae{$+o)nLNMyh_-r!nD1tH#3U!1C4;f8|LiW_j|TeR;HMmG zYepBt^`E!>OY=a835|+y`US661|LQfufDH~E|zS!rriKAu^}>RsA*b&Q`ogX0j+j-zdyNg&xG!bcCnLQ z7>eecNLav?xt$SkO60Et$92s(^&vg~TL+!MA9PvGf9R}K7L@35m>Ch&O5n15NH0%+ zKVh@kC2IPC!fAPSb-5mZ=6>t2B0n)SVVMG-bmBO%sLub^FzkUtep=phm4q{+%bt;I zVcZU#27?sh|?LQ4o{trbHs8A=z755oYlFAUF!pLDs?Q;m~+lXf- z3NpQ(m<|%4j3(OcP{t9_3IbS^DHCs>c(*T6xVIy2NBq z&<1-^>mrWSw&!c0qiqw%!u~U|WiTBN#-=6@TTz~%U|Y>(xBh@zm8+k76{buLW(n9l zIM^KXwdg?G`*}&|nC)26XUP#jp|E`!J}mHC(9_6hf!B$K;p5BO54OwA6joCuV#fcm zL@g;{wkzHFo{+-SuKG>wF}ww^iJIm0FZACB8|~YhuA#}HM3iV++}Y8vg46x}p%u^4 zU1IR(w1{7$+mP_q-n_cMJBgc{RAd*065c^Ipnn|&tD@V1K7NBu6bbEYIXoCNSJ$jBelX{qnuhWW@VDdDP>Xdl?D1}l>GfN{mi6m(3eJlm8!yxs16v9WpdN8pZy zduFHA0x;Vpdut-cmHQp*@9Xa`*qTKwC+Q=oM0Ahu?xEYPwh8CnK3ZUF!w8RI4dO)#Y5!N1s6SnwX{oXZm!|)i#i-5DXgBEv1hm> z>H1Nhm(6k0K(Q%MU`l9rh(g%8)PRG3mDdDVOwRxq?SJSWNS$o-^CS9(f)JSJ1;2F0 zORdW#-f1#5(haS4OaBBn3?3bX_2~$U@Upk+O zD~t^M^rK9%_+&lTNO3V>r{^*2yv#EwGBjHz|4ut%4KN=b*2V6It^MefPBQ;QKF-pA6Vxn+%&$S5+x<{XH^e0X{;?XP<2%PmA ze@L@>jWFQTdX0Dk>^}!>XjIe@9?0VV9m}sp{4baoYdX|$nXfxFu_iWqTav|c;}-A2 zCoQrj#^}w+4$vWJU9hRFnTXIeKRYju+0;u+cyakRJVkBcQmF{PL?zrVTxNHV|s9t19iKP4w$#3jdM z{Vq!0jkmDdN^%P4)!1&sk~1^X!E)AqZG1!PtH}Zf?trUSdo)UAEq2YAEV7ifusT( zM3Tm@?SJzPmf#XMHE@d=mrtG9EcvAL`X?fZK#7ZQ%|^ac+=1{(l}Yk@GJZB@hc6pt zvjWlCqG4!P7qwXGMtrFf(emJzBz!fo7?;v++#n&p+i#jlVfpQXUDMA5luM)vr8-tj z3Sq~4NKB-+UJ;ym^d(*f*h4Q*&{p@eu>!zfqUE5F0p*sj^sgz z@&%INgxv;z9%&vEk1(F2`h6CUo?m5hhrkmq{LwQvLgeo6#1^G{o}t)(AE}Zhs~8)C z5S}paA{{@nRJC(1%^5Q31srJ&q}>n}69UKhwYD)MQPXsJb3v_GJ!8(2Sc+?ZUoJ6P zY3O<-_gvuxFtDAHZs@zKYcL3&9cv12B>hkwGcAFE^OY_qEcZP%BjJ1SlP$5B>9fl`5?GHAkf~koF;u(rI!4Jk1{S4^Es5^(&%%N>s4FS1NV07m2PA{jW$bm zON0%#G5f`O*?Jh{hgS)BA%}x=Rpf|fi`GH30%RI>PcCnrR(>uYRKFX~^-Ir6u#-kN zlTvcrm?+nFebYuG84#9Bzx0T`yi9W#hVVn907Woj+SorJV5maZ<(I1IHNdO0vv~i2 zgdskL;hYL%E5HZath-6+-vR(@Aea6JVU5n{^Kvmffqdc^noneI5JPO21tiaTk2acN zslAQ$!F5Ueo9V93g>ssspB5ih_^b$5z zx~YnT-r1%3qK!tvZ#WaSvx6uu4wgokc$Yu;|75$2GOisBGVbW_^9J-3IAWTnp||Fx ztPs)-rQKCk;sQ)ziU$riO(kj_q+RM1nhDJt0PLfdcI-YuLl3Qn?1vE6{M2!7%vTUK zW=dV|C2U^A>G;$W_w*)zhs~nk!0DS1OBo)?vpV}(?!w3e^@eTmE8p`-a~}q3E19nEoeSrEJ38yi zq{=(6i37n|{0JU2y#>G(_MYri29JHx!N;jPGP}kxUK((nCqA0J<8BlYdSS00qyA@s zJZ|uUZm}Lwo;}=1+{VvvF;TF>mwB9{U(#LhBlb{V*C4mA8)qUbqz5quB&-SkxOeLG z^@SYione_;UyaFe@(=iF@)*Ude@Y4~H0s!@{c>k+_SJe}#-qf6@V!^ZXgd&N68E+e zlGoT_zp`Wb9wfs=4_0Qe`FrLskH1|IA8ktp#%mRgiJMVRDf0oImIdy7u&NAZk#3xM zryFQ`#S964%f~y2i)lLJt%Uo8wr*9ik06*UcERB-gXmj@x(fWlbQ$=(UHz4qXMTjz zZ%Lm|AewlfFlfJROScVRAwg@c0m^390f*5fS?G~7Uj7%hC>>`O#JR`-Ock)ObKj%L zlfGV!Wx@L+b@L`qX!#>_Cik_HR$?FTfY|{#8 z1oQw$Z97h3j}zm^n!jmdjl;v;#X3-)arrlZYy=;KgxzZ04&Pyq^=dF>nnm(>4m=J& zraSo78n&4ZrKNuUx6B_`UL=i`NK8GN%rg8&;XHTLQzXRI;2FgZ>uCLSDQjOQ%KR%> zZbIvLex$qoSwpwxC>cv$NQ^>2bCNo~%8i_;m&Edz_HA4J(oKD|Q{f(KiRZ9c)r)7b zo=v3msQJ9*@2K|ac=i1-g~&@}DJ7K{alN}9H<(G!hNTm&diryQ2|};kOX=k1!4_CV zrW$QZjq1251GOEJ@)UfZ6>VwKE@ov`NynKG`k*t8!3bX%f<%G+psp*fvU{P5eg}8W z=z^l-AU;Api~oU#^qBtoQBBMtKf3aJnN^zA<716;7KA50<*Sh@CTYF`;#1+f&mpI{ zqWgF-wVwX-(nHtBN05k%M;a*IAm8c_&$>1lQ!!HsnW4@O=Q`QdsKK~g$M*X`6ogxb zTDgHI72KxNP-3qLHm!0?D!UQ#QRArsr_bHl^!ic5Z|?OUiB>>9`n&n2j3isYtqd1~ z4~K8sQG{40XT7slKPQ0Er#eC+=!fv@m6a?4z`3?*df#-B-jnzphdkJNk13 z+ePZ3-P|Frg?mILcBbXQiw!nS=QTF28JVj>`?(SlFsVrEOM`keM;Umq1Y}Ke|6L6Q z*0oW-se@zuQLvjNIL|FvmJ73V-exzN@`rcS9*LND1MM7vpJs0|>^nXE7VsNd#QjU~ z^v#W!uEX<;R#5vWsGeKUj#}5cU`Fg7&YV`UEs8^D*tuyd%z6!oD6LLp^-t4VyQdE; zw#F+IlLBA&3S*=2%fo$$j%}G$k`=gnPGR{|6&wj)KN%@lrKSd4Zf^6@R?R&sb>=Sy z{}*|z3W# z+MJL#HwJt6tDX5wybcJyrcp{sUaIkG(ckkTt=2Xn)L+ zzcab>g;}TXk#Gv_#anjjj$7poX`a@UEVb%wG|%~CyV1Isro9~wV(qLf7eO_?G9(6^ zQg2=8K}&x%`LpQIwRS0;f@(*WQG-ST%K8lx#Lze^y$5KyRUv^CA+HeJuI-Eu`RLV? z1|jsbrRgtWy|HoKb3gv2YMtHPX>OA7z_1s5ag*sXoh_Z%%*;sLrjMV!Z_ZKGTCP9J zfBabAr(z8<-NU&b5+hkcxE?V&cqGGXuwTE^mcF7$Ebq8$<2rGFEWqHz@EGNAYGTxM5Er9LeiXZ@(~}BndC)2ibd4%4=OB37FJ~ z4Jd;*9+OevqpIKi*54!a*U8!V{M4|j~2ewF# z@5BA@c$@*Bc}EcDqp`1jH;EaqlNHY+aPrYojhxyJ_}nc3k|=>=CP7y`exJV3`HT_W z!Zjc#9dE?HGX(g`VP(l;@|2(Yv(69qcXWDfx02qyTJ&7#Q+XWG>1k<}d&;TFsKu-h zk%I13&6ee2l@}8hg=>2kiHxvr6lc7_WT-q~Fb^tWO9Y>l6gn$IWoA^_z zw6wH2z)fLiEDxo#(R6IHs8Jkrq0dK8#MX2t zAB_oIALf27O1^h%CEMJAd}N9W&Ie~&P$lit0Nw9wOGf68Q1 zt?*uH?*;yl@@whO`Yk?p(2nydF8Y(a4(d!uALxd$PK1`%uW*4x=ua97BR=ikzbNJL zrj-35R`n@@qV-;+WEL)uxS-^R&GSH;r){F~vG_Rg0Jx7J?Gaewl+iv}FVyW@;dx*$GHN^(WY><*|Ts-WRN80 z<9TbaqzAaUY|KuqKi(LtLw;)7+sr`Ka)>sgxFzXQ-81C$iw!<3ckeW zvW)&|Y1YrQJy73X{{fhrg5Iqzs3*_#j49X_G4{1wn7(NcySmSF>t{An5F6Eq_up;Kq6hXbr1;PKbBKDDe23;Q7f7K zPYzt%mY@mWQ4^7tpB8J>H&TPpR$bcKEaKM`ZgTk|rV*KLqdc@Sfn|S7E0!7)Izu_` z9D;E(Q-dNU&oLH?zq#@mF52Z=0pE2Zl7QbkjS~#y!Fm$12{!?!fnZ#ubKdz>#(vDq z9?FFB3U2Y(6+K@-2hxnC&f7Uyw#!6*_S9PF)we8S?axNi0`I^R;{TU*%4U8FoF%sk zf7eFk`E!gtM!YqjphmyJjPlO~Ypd;>(9`GBjF3hAzBW>IvA!&cbA2~-CHq}q7;k}Lc~ zjt>X3#UQ{e7LiL~gV*yq4kYmf`3R_eZ|LmfKDyddy05@<0`FJn2nvCz+H+O9+jI~r zaI6DCB#$i^3F6cx)UKe0Wc}Z9$c36lxM2!6djpF=qPF^Jla7jHJ8(Ed*|sCi#~`Z9 zO*FIl3$lnXTO)o7smWy9upieGo#XZ_bH>P@h3mjnA(okd4};7qjDKn1`mo}d_Jm#* zA_iOPk@)>8vqFH|`e^W_zyZ!uADX=O5-P{1Bk)gcyJY;44uyNd*5W2ONcCuC8U_&(caDp<`zp9d~jTkf$q7O|P zG;7`vb;^VY>wo77!YvHYuUsdTdFw@J9Z9U{SgJitY41!vIA?{viciep*6+{KaWU>P zqHdblv`T-Qr={;KO)cwK$Bd&Q+AGecb3u=af#!DJap$0?%vLy&-3PC3z>?DQgrR5t z3m(oyN6>3iR6o&`V@@OoBZd{uW9?ZLMRf3XCzUkBUKtHU7Y6UPeImEjzzKBXq+2Q7 zA!287Wai+^gkU+J=lzNH7ve{8-8W^D{qUobfc`*-gfSb7uAL?so5WB8k5gO1W5qRd z!->tkoPc`#m!eX_HKN4t=JPo#)7z`<>>j|Ju;|_=15~MOz6a{QWEEb%kAKarO>JDC zi}G{;BOTp-Hpt_I6jU<6q>BLk>$k`1+v8`FZDu%z4SnX8vV2!we|}qZ@1vOkgW~&? zTDX~$=0(^27iuYPI0%ISL3>n`>T0=|LW2oEq_>_Av};?@ww}jyH|alpIi^VSRS0JZ z0~OOPP55j1okX#l_N31%Zc3n5Jx^U3WT3J&H^dr1k8V4)Xek()?Y$J33)yz<{|1Zl zoSl4_qPYxP;Tz+#?+8M)^%rpM-IvoPV_>KusrGQvik6pdW6|+Tddb9M8I}8<7F1U0 zpaaE4Sz*A%uP^o6q|3h=gYsK?`#{$NMQL_@+xm!`U4tH7JLZTNED0I4_$3$LV-||# zI58Xw@hv0SV&-LS6Wc|?uH{&3d8x%6zhTV3g=q)SRYpa>g9Q4nR?hB0>MXLDsbt(LfM|PoIPW z(x|e>g4W|lIUuJ3UgGTN5<%mEqx{@Me?p799=GE%30WS~)thy`+4AS&C33kLOAKCX zP2w9LfCJkqoI58BA}eG-h(wA|EEmAxnUT`U6@wpwigT5HQT_qyJIv>qe?_zxe5yH% zyhaUTD0*gLjU{Z%Ctpmxhc5r@2ouL~kb*8fph|NB-Q|q8J#MrW`IXmULEks4WUyAg zCU9B#(R$cyZq5$x$#1B(VeLEP8Ru%mumprD<9$}NKhnNBhpKBkKq!4K4LHjtZT|B% zC-jr?`ENH(WX0Y)rei=ds;Jsh1#n^;&@oTlXqc_ST~G&?tfnu5AkJZEbuZC37O=Zu z_)USY=MZdOqwmKNy{uE^KOCcLiNzRlZX|#Y>YA|$!!-_sbqgsfe?~X3xhyGns4T%u z*qoIDui$)z#4*KyzW;6bU-yjWYlVy^>n9&fJITbTnWX)_y(qsBta9+9a$C}7XPGl? ztPA><^AwU5x{p$m$?__iiIz#fkg%1mpb1{c&ZR#K0Fx((KbJm!BWVHqGqNfRA;Y1Y zL~3;Ut16TALhTqYhR@kA_2KaVqpOq6!phhv|KNe}WWaX`67oDx@h-0V->peKlho!4 zO(UQ!g^T0UHqAtK^D)b#+`I{i>vRnlY8l{P?zPXdu2O#Ti@OqAO3Ls?%A}2(019`? zaS*fJ%1!WY=17ba5>KiXA1Jk060V=eG9?4EVRj}lzCo2Ism2s5MkY>kzxLZNH=c|S z5$PT6JP@Tcf8hw-+R5pmhD;M`TWw+L@?d5AVwHbb-T!_VGePBRJ#l-5;bIx+`pKoI zc`wq6pYNn!f7`ttYCCJ#71K!Q^`o;B?)X^r;>fi(>ZGK@HE)||K08b)zZ$r~Q?fwE zZM6G*SfL)XZM_38>4^a$C$xh2G7pX;WFv?Y8XJ9ld*n<#dSIVZ=XI$?{<0&MLkK_Q z(9E+-(xrTU8VdjTOdZ+eG)@M^0@Fi!!`my;U#w(o#*f-8v#n#z9iD&`Ol@kjUdeUOb9Ye_3=P~YTy%YKuw z-NC%7&yFC?*Im-o{RQ45Vr^q=xSbofCh~QcmQmnNl&IXhoLT!f9S6M6IF%H?&~tsd zXFr}<`9vq2LW&JybYI`VT@BqQ5lG*K5NThFkI> zOJQm7*kkaMGKsLgp0{f~KRm%pP!F0PYpjWV{{2M|xo%4q1=(R^IU5;!u*R$n7ta*pKazMsS#!W}FpNZK52y*7$%c+5wc}QllDV7he-lgELDNS)}j?B{9&P zVW)CB%>hm>Hse^hPv-?3$$Wm!U^PT$5jEdBrkb+Nf($3Mia-f6kDK(8?+?Zs=D~W1 z{kxg)NSeT1a+xtI&$gLTPfcztuFVV4Jh7 zmPTJ0Z~vxCL`ArdNH6;{kG~Sb|GeKW%YXg+Y7^^X#17|%?sJWg7DmwR;W8&XZ zc$T=J9ZZkhzvPE>_nAfDGV*P_+IG6am7Y199aTnV-`9;Pdaq`*((-Gva5h{ZF-7#XaQSy(F&RO7GADC0G^h`chxttieuX?{dycKHcyQC5|4f{b9 zo(q{4O_7m_hhpWN3$OXHC zC;r%&%ZPl&)-9-~STt5PTG#v|h4b*g(K@09}TM|V3JObSWrseci_Rhq6 zj>W55#m6^HIr>kTyNCDwyUo44MV4Z07Ht7EqJsvkCTxqSEFoCiN#ExX0HkU@K{MP$ z<)SEgRookftJYbbOhdEbcz6e3@3H=lVU#qSsi|yR7=_zOW%X zggz#iSCLxSs*e^Wx^o@WoD(_3nx?^QVHnzbGf_=rg&kAb4JF3&MLB<2VZ$j%Ny>8x z^|U5OPH#_PeO(iK`ZXu3JWLq^F8t4-?`*gRQ&n8gxARj&ULl$;Mno!@8$xgC|K^Ft z48t(Eas2oRL)sQfNl9B zP&wWM$SNIxFeK&pG4Tfvz&EzO=8;c0VS&yFx+oLQt-3!f6I0dpu*4t-C#91X#OJ}5 zE8E$=OP}!49xJwE3X;VYf6VG69c6s)&WOY%B*u!~QB~)@j8^W5oi5&HN~}4hno#DALC*-{*4=+aQtcT1@e8k@p z!B>&t5>lGQX?Za{nTg)jNtU}`vJml+yN0DWBEezL5JozaBZ(6Gdcn&LKv9wT8oc2`p~ z@z3c%|MT%1W1tNj}uB}wufc_Wm~&jm_Asy=sIsxC|*;& zUqWvZn7i9U>51{^>l}nlW&5c$V0J(P^rE(2YH%;M__p+^v-v8epAyO*(z3IUZ8)f? zsK}?wfYAcnTtX39uC_|T&Wc zb6oay`XGo27@<%&rwKpiK9WM4r1m36{b@PR!b0u_W1Ua)MsC%` zyvKpbc0JN~yRdrq>_lYJ?VACU>4um%vPkdGLa}4!ywEkQDuYiFdGG1JO!Db23gU@n zXQ?r(;`1tcs6Wc zaK$mmL-9l-HkLj+Q$F2~fXOdpqxH2RCVDO#iwQzuVN=Mrxu`Ax>KOf8gJPn7MXr<} z?U2=S-qPRrj(OnZca;bR1?J7gPl=0K@J;B?u+@x}=w)_ni^SV1k3nzS;AErEG(&tq z%dY9ydyJ$`L~KDtdVJ#-U19jpC)1kcBt}DzP4Rz;&S{7xO0@vRj&nt%d#5W_7{o~h zCZsH!*Rhi1W*`p7>G&ix=GJ>Z-fuV=ZYUb5opkmw4d(@+?w`HWp+6VPHoR*mt$PsL zhYsx4t)ZHu8aj)ji{B_y-;>p+_@zqW&X0bqi4jdns!a6O|EaNovFKP!e%|!fLZjHfrF?ta`UJ=)|qhM5;6SnOK z;lZlK0n$xt0@T?H?bbZ1PbVPG&txh|NJO!aAD z#eCzfxC|E~M*=A}#19`ai~WJUvC3NkmqAOdJ3ifo!O%C>wvRRnEZy@Hbeq9 z+On`r;ul)>oB9!t7X@CUVv7CVGc_+!(|LG0ug};u>NuVZBtZs7)!Y?drBpTXl`S3T zRH-Um_jImjO8o95x_hx&%W}Lk*#|xTGLeH=FNOO}!%OJlR);?92`_$a)Aimr)7wX7 zL!!RyRUL4dItKcObT-kqNzCjtp}uApzChQxf{5pna`Yov+Y#+CWjA}%6`Po7(q4o6 zuQvX;LuE08?v%)Cei#G`WZBINnn#H~6RXGAdb{TCuEKrJu5}45LD{YOHseUuIrjx&zVM_?s80yN_W7w(w4a{2RX32%v4QNDB>v0G*p` z0A!}o`q>ocTGQW`aaP(^fv`|?KOrYri*Nj@7um-Y@s__7qDkwNz5ezxh*~m*Dl1bL zu$OW4)W+``b`(UdMHpUTBN=|2J)-YpqJ3kVu=cur$@#$IMF+ys7wYLSB7!ayd}tQ$ zMVlXl4j)VTmHUYxnPr*qAIFDp4hreGQq(f_uYa`fBigY^6kKbh*W-P+zHio0?w5jj z{nVIk|AUGVU+!wdruIE5ZW5nK?KN3NP6AqmRr2k{+Hr;d}aKz({4mmqd#wRw*Kc{-RUJFh{Zm zy(=K@T??l25e4BwIgohbIc@-vAkTp5NR|$9v+==9h5Kfrx}x%B+{s!!^vVDCR)Y^| zB0Xic?oj9N+s-iw@pj0!>VpHhkquF1dFtli;2!?Ic`f!HZU_tU=??`pCCit2zDR{j z7=Ja722shSIzKCZ=Lg|1S&({WqfGj# ze7w#11KL3r=E0SYV{S3`E!)FXmdDg*HXHPq%%^-K2-M8OgemD%&jYu()ga2PWo}_> z=?>^wtsjKVOg08SNv2c|2OWIFZWOvC%#ZLOgA{p1{OD2W*ewkbc*}?!#{Ht~V*?w; zD+zHXS(8?1r1RA+TO->d0?CJz&?ZU-=t3hOYJ(dOm@r+Hk%e+4QemWz=D;v11e&RCXYD=)^5Vd9*~0iH(IlT3{BAhbKmQcz6&Cd-49$#TzAAqG zO#VPWd}J0ay5N?DdNdAYY=N~ZzT{XJe&_Pc9THPszmU=}<@MUNFebT4oD(1nSqJE9 z_VrDZ>%Y6TAodx$1oaameq7GE5Jx{EJuh4kLg__*@0e6i16Mxl&VsbM2o!-~Y5uN? z;AD(1h9-#9b>>g2$KReLY--PLCF+?RT%092$H+s!lqJw|@lzmzFI50C%UBm<7br75 z$HMXR@=Y5p^FsZYqfh&KNmqJ^sSP%(sb3?ybztXaZ=(Kc=fp;kskd2cexdL3a5)_` zqzTs9N#kKFjDJ&>m&KzpHBD8aZz2xE?+_5Fy>4n}8}+Q(bStAzgkxfvX^36LA!I?C z5j)_y%4aifPvb%%N7B??H+!xvtj^AUVqpn!P8{iPg^51g*Q^DswUs+c)fG|Koc+YJ z-wz|{#*H4ii;}xid+d92EPG#2Jo1h?eYR4~4CksXzF@dDTr2#vU5wEGL-lYOoZ1Y! zaj`KmehQiVsk(ijyZ`Ex2@%9dnpLK#Q*$2wy<&^&%5?$`83kJ4x$m%N8WguS5kGTe zYkJVt#2Y_H5NM8bS&DNZoat2nu@NPAH{wIxFckC+v1IoXbW>^jn0gM~TZ}^~D)(S6 zI|s(Up%6{GrM%$Mcj^gq$oogSA9tT~-v6C=M$Z#p}x-!c`JO&)CX*Xd^rU*Ygf9wi%uYnH2-iFIMyStpwHp+gb%j(j5>Mzmg8 zhP`HF_<|GCls++ zqIs-x^?{j)$OZ}~{V2hcURjr)D%8tybeCWoINV=tr|(KdlBQZMWfA^J3}GUCul|3H zQ!qzgBue`H;0m)_l}&&G6^fiY5C1YcPVNqeS`fK+1d-zcWcq$OoTZ`SCm-gy+xuC# zQew91*hYp- z;B~hx`9=(oZF5Cq3|oaoRTvhyMaF~3Y<)h-VI!ygO=`Dm)Zr-!9VA0?jvI8kao}i~ zfc<2Jgv^V?XY9VBUVXvW{r+m6L*(^6odi*j9(~>TaizR;gU@H_mfTBIhw+bQ2XM)* zyjaQ3yi=)_&QX@30u5QVlstAmZvpKcDvBuFn2Xq5oa@q!1nLv<^}P8;tYe z3uEFiK$GY#xUWYRUi^unQKtNA{C`EwA8`Co+Z)uGKWULOuiL1Ya%P9cjXn)0!&A&i zJ+YhDs%3VHCV<+y|B`ag^+24rN&(#frN8J$0V<|Kp~%Y-kn!A%x*v#B6~7aw(L1mn z?i(WeB`T;pu&SCf$J_zAfVW&vWQENG4DCQ(*;qb1Hittm;BK?;0AtkwyV^r+ogiw> z5$qs%z|nJG4ZH;2J#uDiz=GK5@6K3?kY`O*wNHiV5UI+$%uMf`Dhc`TJs5E)ujkYD zc&l$SK&=wQQ66C7BdZk3v9AUsq@nvJftZthZ+~=D@$c>gN~!r(UP6RN=9puj{FQ6U z3)am>X9n#;**I;za~}r3`fbEFs!mK;#P?DXLVG- z{^PgR;rF|WqU&#Y)$IFwOlOrRs9A!VNI-wzr9g?T@Dy<~F%On$qIL*X9(VVIYpf8dt z$H$qkHMe2dzr3ukGpAry?CU<{j`0rkkJq6ZUOGI&=+58r^$!M+7z{i`r3c*g!1g2=l>4{doq}jo z{L2m_2b~5LdzW;T2$`5BQcVM~Y0wSUuxMjLZSFML{@sEs^ph$)1hrmhwm%{y1mnuJ zB+;!~L3`=|@*U_Z_IKYRaiKbY$ORr0t9wgP610I14WQd>J? zw{0%^M;*|$7^r|by$H@>UNAWQ2gdo+w%q;LaKU*(K1INd#rD4%Nb?1a{^T%aSiVe9 zoPj^3@2$h5UpSye&>~|~9x$sW^us*RljIrp%$ohL!qYN-D(=w%!*ja-svtFszGl8n z*z-4e*-Uh$?YcLKw%FkIo$Tz!I@o}=s9JtX)&8{MWIf(M0Un1G>9q@7+F!_rY@}-b zrJ5n@JDI6F0vTO~iOQEuQcn#{GTAGKLxFKR@n-{tIA${xv)Ol-G?m}aMc1=L^P~A2 z7cewMW@{$wJ-c39dza}k%g==r@p!>Vw0xYm5~B0gu2;^n%<>MKh28I5Rua`WwwXkJ z1{BDr;{#GDG{fBx;p5p}h+64K>Wr^VP`Vevhjb?&=P<0})PGHt#x%lox~J{mCzLwb zjkQaySWFK`F`b3~r>du%6Oy& zC7+wa5K{^L=efh6eLqP7rTrTu_qC^Cw$*B?IO54!ztnieg*v!>k!w$sbB4 zBM9`WC^$gpcGb4VCAsM=qT9d#oyK(v-o=K1K%U8qYp1i!}fA2>54bMXIo2V zaqDUm=El{=&A#K64nvr-05Pcj;6k79L%%AP+phDYb6ZUfV*KwA92!2H2Z`v3UTgOQ z0-5Q|7lkX1>_Zv@c5X(M?SWF9m_rJDEZ6#8DOJmbc`41O_}9;G;YzNs%XfXFLOV>U zwYDZu{d2`?Hb;ruiD*_ud50)T`8a{bqf!K*#t}rWr#-P{>1XoA@PRJI?_JPhX#Ct6 z`>KcLgPg-+TPrHhm;98eE z$$fu&`cB16_VY>z?z0H1w`nALtYE6=T-MAL)yOM!_+n zf&RY&9}6MHi{;ZFbd&IjFJ}D>yRoQk!ttN-ATU_v>(?vC;lGbp?-T5UH$6!3-}GT9 z*^)6xEKq&Z4I2#>{WKs!Hn(KSu54~%zWOC9&eI0-HThGBjz=3OvjerbSTauERi)@) zrwtl*$El56k2n*`PLw<30?*LSv2|w3Jda~#&{9ynW422)#x4W;gmpsD(HQq zhvS`@WT*Q@nO9ZHOfZ)I^E%taL=^Rw?TAFI=*`(kJ&M1q!3w@h{{N4ww+^Z*Y@@!Z z14tu>?hp?l-5r8-OGr0JcY_>C>69*ML0Y;S>F$mL(jZ;$=6T*{-fzBt9A_MLM)$t= zeXVP)-(s5!iHWvVxOSto-1U+$p31WkMY}msrz0w^jA^NPB)pxwUWIMjb-V=)#f?%l z7@TH}k3bOaW0+)&7N6xGN zZUy$V*2>Dbd)~jc(p|ddQ7BjBdp#IH-Y$FA|Pafo3d6f9Frzti?t;y^;fGegUNF}pdBmAa1iur`* zVl(jTcQGrX|89t1!c2VV@76oL|v|(*DbH{d;FffoJ7hJR^c`RKhR3kN>u5uaWNAbBn=8j$jqiGfdIr(_; zvjG+LRi-2VxaahO?oZm+BInFUMW20rpIf;njlC${Yz_( zQEZ5OC{;P?o>jwzuI!q&F>MIcl`!8wr~;=j@pnhv*mj16?6F(>u;T_(Iw_F+J6-!# zG-h-bsZf^eV!{k;7AZFnhh@T3zZ_~hnwB>RJ`AU~%oH<5n6;@z zA3*K$C>}@}Vz=DR8kj!B`zj){YVu6T;W#FsTiAgR&DpQfPCDfM`e9-J#vp;mdi^ft zR@`*NdMF(?Po&t3GlnqHmZ)S>O}+4s6l z875u4^%f~a*nb%JTweVX@zW2z-|kw&i3zy+I=D-`c?uCdiLxeb|88h^yGCjrILE%W zwQvQh2y4#x)=oTv3As%5Ic`K@ZT%Nd{0wyX1)Xj6r#ltc=e>N<{=28aGroTX^4=I7 z$9)sr+Z9Z9?94yDX+@g(Ei-L_zH7Yuo&a%Qw~PTXiez4-`zKMoK;Kt1TC?UtHe@Wl zxAu_n{Ptn>Ay%m~kQCJQwb^0{pc3o40|D8f?CAg4nzeui`+OU%^&Y%H@t|!>ZyXbG z`?0ocFlBXH`sHepRB(%ucBYs(6`tVyg-|F)RWwuxdA2n9+812V^b#Sl_s7U116H}g z&|f9~zf5RoY0m)hdm~<+g?-9(A3QxM&5G3 z6FI!Sz1{tg%4EoM(sQdfnhyyVU5?MSJQp{hkBnGvE2yg{UO}9y*Nr15N6{(9utvpG z;!|)2p6un=*4fQpDwQhC5RgKE@FXJ?Dn!MEJI~bL_{eEAcU@;m_DtvY)tIA;dhHLe za=ht6Lu2sGr>X`%_mG`0H1tenOlZ9VPhAA}Xde`)$)(J(fBCo1dE>PYDaigjX# z*n{^Pf4o5pK<;&QpXC$NA?x++1S}C?#;jvW6`!{g2V zu7P8=LSV3Axq_!2_|sXM+I-)I2gtq+DJ9~1h&La@6OHYbclpBI%lD{XWNkSiZRHD- zbA|N%QTG`v+c{$>)O0&r%k(gG6RKX#^@KO2p3q9dfPlfU;nHgmw<;k%=$}TB00ITd z5BoOKi!8uQfZ5(Rw6B)vP=Wjlv`wu2Qwt&MeWP-e1a0PTES0SL_r}bU77b%`VKv)2 zmro z9JV58R=3!xYlC&IY>AH22(i_j(m_9}B!aATD6sLYOQ2>jXd#i&@ z{`tDIvJZupAb%N3ezZ?A(DUed5Z|@UEmcKDj-M?a0^f zz16$1RU18^9XFJ)uvfS@FfP8^EH@sczPQ4hpl3N1=-yNXX0i66qz+O3G#?h6oRuI< z6h8;2D~eR-XM*0P<0_a)#{BG#&cNlthw?SS)T~p07keNPdvH~%@mBztk1M z>sKH&>7%>0bky~ZtB$`B#T!__Hha5xMJXNpW68*h3ug-5%8hyaiFx5lsJUYI!=T9B zP@CX_d&YS2AG>bV#81dlPz-Le=SO@La_1B~7Vs;Me~#8{z78=I9jfPe!WRnbdwIWb zD!9mTlIv|>kp&1j-tB8T%4|>^G&PpBP{{A^*Jb%l(5%OZ+WPruYMwo+;ta%gyMoO= zPwlSfT>yga`p?qxE)+^%qrPW=;Iz@O6<)uki2^vhPzx1OP5J=qvY(e0*}Hb-+rU|P zNT}IAQfI*Qvkp9Iyj$;kp_eHybJnNde|icEwPU-)nkZF1%5AIz6t6I1PK)&fWf69$ z1R6TNl#`J?o0hYiD~L-%{EAH%L?>?&!`XD0Zg`V5ZOAC_OC8v z(xTfgJjsbb$Xb8wvWt$8gyx>@N4Fe}JX#yD>{E1r7i_DIhCaCDZd~zw%EoYQBC?`R zJsp_k5bgOJ3&fyxcedIt{Th)227kwO)ifgO6^ED<_^LoWyOvR^X2WdeM>3yve&j^5 zl~sBSsE#?7u_SnB9B5^@u>G^9Yqt}vjVMctOm^eG&_91yx|RC&SC^@Li-jHqYCxgA zdtdnZ-`!OI1Xf@Eq1zyd$ut?+{u#y}C7hZf4=b$oYQe^yt)Rwy^-qSji*)UzjV<6; z)kU;E?|`8IEn@Z7FOwTPRyFopCOK?lT+;5O|1S>u_wC93p^a*0ja|RL1T#%N*CfyG zwb<8PS8(!Hk51CYnc>{IWK1ACfj>>poqc<;$iypI00_z^?w&3K{0e3r??OTm0 zyPU3Ar3gS8=tb>rOVs8L_|3R7c1Mi=)+@))A*yGKXae}0&K+EIXHxf$wN}#k23&gR zm|?(z_jMxpfM@SjNZ^A{Dc9P)gjTV=}?ka~Vr^yLsBjs#Dz~ zwAcQ4MS69-65)QdXbeO{Uz#66F-b$Sgnj;eV3iQj3`j+e@w(Ur%0e+^iVpQv|KhpA z4J(4Gui*BV=wNlTp|}=u@iFhuNtg)C zdm?O20RZB{xG-DFvF~uMKnX{Phfup;fHrTZ{@S3|FF1|_v$kzdhR?PlYH006f95t( z58P#R{Ru+%p+v74SP2+}isV_6@8{2cZ#$Y(?&}^o)lmI{m;(_#JebIyI+Aq1m_T}@Bm zm?x)A%6gh`4;I{Js7@Y=v_bg(yCDH>Q&fBuMmcr}MvLblr!=I2%Q}y<{;+Z-NC~QN zhM5jz$D|YJc_LXeL99)v4OgE8g|;FQX}YRR18YVRSlWmhc1hNu=wIS!16d+Ju~fX8 zb*ss0WmZ8N_zeA6S|;WQ@P}3en!FZnNqIV4XXnrPOa?<_vtQ<027GnWGxAJwq4ENUmEDoCr+^KT2io}(3|Xpo;QgyERYIr& zNlCykqnEK)^TDkG0rLu46L zD4ZMIUfTkh#o{_R@^3BZ1xe8l-xuVgLw4E&zGv*a8o}mfKV`3>3!qGX&lW!1d^QI~ z8>`(ESXEBF5wjT-=e4Ch7;MtZMi58C;%&ECK_-sY?uE(`=kAq0m!-H3m9UJ<kJQKy+DB7Ru835ze7WLBeoP)v4Y#IX<9GAOwZ zq9;j+H~n!tSzSQ|NiKxwmuUTe9z7h}wLv8W4);#!Fy}=N_h;KM;-z0_K_2-KAWfzc zYaiJ6kXym6^p%ux3xjughD36FrL69s*vCXh;O%=QpVb-pZijFII1fcByo+!#C@>D^ zj|DKY0z@1iFFWwH;GXXYJ4CWEREMdH_@A2&;6B^V8)udK(+Ms~uf778r_)trkEH*S ziOsy$;ceG95Cr2o^GgLqx~ApMG|Q3i0;$Y|wP`nL!}rO}fFVLq{xf|)WkAb>s@u`x zOz^(&Zi?w={jgKFgIVGuEei{JWK5Fn>zf&!N{aZTq>w05UK0Rwb{;1u0VEhE8eJ?g zh&g)r(6Gq(#sU9@XR$9Ro!24y;lbB-3D9l3!wP@)Kv(~2N|k6>F(}7ThPVI=MJJ>4 z&s647_-(}3VuLfP%fM8JF_HdT%h^?lg{3wZjFmo5E&p{@G=XX`Y7#1<=lp%6nxemv z<>+8vK*m_#?Vs4z+IAx+5_={t=XS<9z#2LIHG+lmX;L~M1^h*D=uG-g)gLDwE2N%( zwLipJitK=Rda@pac1;*Xg{pnpp>BE5)lc^rlU8!cQ%F}g zV58zw@Aa}h>L1l!`7&2jntVZ2vBQ+!&g#0e)o^`z^m^t`u-$Bh{ZX(nwzf_u#aA%0 zH|-8QE!gZUqNb8UwQ1o4hP{_U21G&Tkj<+$u)VgRb z_kWMLpGe`4^AvGOR!eWyW_3|If$(r}@%G2cxZFAb;53(KEOBM9M6pfI5PrmsNwOrP z!XR_u0a8L-2&iwY=w?{IYP)-Fa;3Rxb{Nn z3#X`hzXYKmN`}G(A;Ul3E!?W*vl8&X#j!>@*u5TN3YqZJEU(?{_-$kN!C#Zmk$BE} z8r5n7!HTwmV78O9I_Af;6r{;z+QXamZQ0pxsG6q|rP1qE_9b;udi8Q-dF&QgGpuuU6Q5pM=)Di1#4)( zd+GV-$$FGEDEm8Pe)R&$51p)GYeAf!8%wFa6~-E>CVmb9j;b>}&5p{cOf}IG#U$JR z2`z7Pa1lbFsquuCOIWnxN4h3?l1Evl=Wr!1-U;^9mTJQ!9IoG8I z_rh(&*B9L)!WhC|;{a$I!*T*O0n{hY9=aBu^(HjDS9d`O!u2&ml9(sn}Ri4}~-@~Kv4)f{&>O&xI;8y|gE{RT%prN3uf zrd(`1UFKHJ*-J*|tE{oHv4yryLw`Ob2^l?FL8PXHBO4?YiaR<_-PjaW=a|}fA$Gz+ zfZ7egvc-D;8}sKSf@i+yMhC47TOtMXuX- zJ{^xqR}quNJdO5rc=e>Wvmm?mzOG(_=YIs!9xGS==XDi2 z;HsU8z-wDM8t6*&FVq+_KH*5E+FAPa!v-Ze!amj>>j5{fd85u~R`kR~-H%{l)|E@k zBOR7!g6FKh(7`_+QKcd<_!ZuYcdhYfJDscvO4WkFZIjO=%p}|eJvckuNxibpd98Z{ zzmCnr)iTWoN-Nq({)1EqcnMno;FHv|za~MxI4>4mb@raG@b<*%C4j@954egr0(Pq{ z;5RteskCfxJ69|K+QIHHCPsbyFC9(>JiwLoUp-Q^} zC4_tjL=N{)gyP|^7M{9glx>%ceUvob3XaBh_uX{hF_=#B$NrGmqJxxZa^FDYF1y%z z%ya*vi#)mZ&|$=${ruvL5`Nv)jZLRCrC7-Y&6tB1+)}3+aO^5eH>~NKovN7YrW=yTppEkGKxRIh1)GOBv#LT0*vtU#1X?4_B0lPmiV?!R%LNvEa# zfNK}MFGnhub$khWc?q-=7=Ccpd@rJiVceErD&1%*BjNDt$v%9<^rIR2i}AaeE62 zS2^3N(evSSIuOkKqiKyUyyI$4`}K_2#;COjhML4VFyg$7?- zc2{a4<+@h*J$Fp#Vm9@XJJ%e4G1Z5FA-QiS;Z+y(TP2s-g6b$am6rR(B|FYfKdhq5 z16LIe{gv3^aeH~eOtbEglqScssSp#I@B()-(Uq@1TJ&%DVrV-563Aa}3C=_;6z~b3 z3!#iPfr47p{DSato893{wT3ub=C3r&Vx$aB#Qhf`H>Nv?J~CJPh#1M)GnK}5C|kdU zqc0iqAGQ5Iyy@Iom7+O@{eugrX=999_UK;H-c+AGK}Rza4uSeQoDwS&U%1D?{Ew=p zPJ2&%K9`~Iq}7u9uaFETx@G_2X7K;`=|hiz=^OQUoa7UQr$=B;Uc)YU+I&>dm~&wV z@J&L`4;L1`f2T|Mu9sa#Jl#%#n-5VOsZYcU0k2TPidqzlf=#XisK$x_yskv1eUa=* zZYM1l(KQ++s8@#zri>T|KWlPe4!;Vo^Z#!U^DkTE!+R|;G{omDo4ohz+icy~oWpFG z)bj*A4>qWIl2V;~NYZYA|M$hXZV{jMFUjjDaQx@d%}c9~UMw`@O zqo?niR-G)U;ORacJoHt3X#~pf+D#oWy(uSoA(@B2blOF>zvy)^(P!(v#|?K+`ro+} zoW)``G)zc+C54yUC+*MCshpN+o-;shywOI21>>NVpfMb;MVsd>gO~{MG`}#O4 zBZEvG_g9SA2=qGoI{w-Y@R;)kz>@qb{Fv8nW8||wP?QzkC`1oV0+8!aLax)umdf|k zRdeJgJAr*CZS?Gq6!rUYAJ_}%vIzCt=Uqlm_|N{zZQglB`Z3H2(vb}^0J}Q-R_*2&+GF9=*rT@WWEEF!BxClsBpww$OQ|;B;$c@ok`A4?ir#!K2(T#f8mFV5 zTA3w?LQ7k(A4)G(M>SL*>=L-{+-{{(>ouw8ex*gvc?>%jVy(9xa~i;AI9gtn+j!6* z^F*`Gc{N661h78ItQ;!ncn0_r3k;?*MpxLe*b1YK{Yx?GyHz;#JKZVt;`8XPHE~ET zT_TV9&lp7OK*?j6KG8A^4oh(n~>DS&ny=%e`hHD3BW0J^%>b++^qA~uG z<*zCtd|QW$dRCWmauAklkMUcoYW8^iXav`H`Vy*)B>`_aR^QX3YXnP2I^r|-tBLfC z-^_L#f}2@>2?~5ch3B{7QSaC!=ibF85m>ThLLKm)9qahMv|rpFo`moPECghfAd*pf zKvt92;(4P29+ElT;Z9&*$3Pd~cpuI(> zQp)8Cjg4Jak6={TJv?egX9Q4wua;PfP>OqZ>h!#Xb8>mS<&{ku(Y~>A z%8}67CF!=$aeXc}{t|EYj^NxAc8QnHXa=I^lYQ%%o3i}f(RUQyiblVC_m^z__M4Qs z$cP9g-xwfbwCdyTt=sJWYyWY;m;NFhlSGE19WDEn=Ig50q00b-wD9ncX(`qoo#)I) zt)%-0Ag|`R6+W=`-6Rvl@RKe^f(g7O7T)L8_IXWN-@=1o9p}PB z_dp$A{|mz?xY+l}%MJ*Y9F~38Ir4-USsyv|>;n*yS=y$qh=~RN3L$o%*!N%u*e&EEG2Ub;qH{p&Jb$G?y|7s-$y`iP; zCi1siX`0z{FJ3#tm#P)WQ8{hBwx#25TCQ5HDnA93$F0J2dk(is+7HoOnU%Vnky_l{ z0^r%XxG)Ouf$5LQ$d@Me8YG{)|EPt@ViH#%{x26dg4{0$$i?_b4m?b&^Xq_JgIRrH z2t&)?zisePomoZ|1-+mQXE(~c^WVH_OfL}SnaGz-;J{b&TAKZ(8wMn{HpfJtxhT~v z4sYrk0tlnkH3lLnL&E#XTrgPLADl1hy@cLV2Bx8$d}Bi1kq;h|Pg)$9#Eb<=bDA;6 z23+7m6@m%!^WZ1P#X@sI#Hb_6alQNl(zIsC{nShlWPb69YWI{w@vL{Wo-biR`=6M@vmnR6VYLlkLqT>5nOxLB3#$6!;Y+NJB<+g=0QX1N(ZJ_OwxpL&QK zJ9oJ?WE;VR6CyJXnG^Vp&mUTuF0tE~^d|V$o(C~69CiQp&wqbg@RpTk$HPtUJ*(wU zSEX3DdtSbE=&jYC`WxoF;=OP8zIpq-R(O*=3d-~M0-8Gn_$#j1c7bzg273yiQ5bQ+ zW*pe&;BfGnTL9ja_MTr|>a2dGV#I;#D6xhKTlqt8~5ka3kwcO}*BN;=RPv#S9AmoPk!@!HgpITfYe9 zsR+us3=5B;Yp(ZbUb*xIMN8oodwXOpub-&f26b%<^_!6EJ4_o1O{WTF7{JoB3=Gp{ z_q`L!CLVb|u$+!uJ{hUqg-8<(D077=lx?{gj_mqgwdo()jlqJ)`qnG4b_HBoLl)Bg zR#DC~mc!sF+CL=M;2HL5Qnq^-RM2(_=7sp$@NOS<#+SX*PN-kf`9yZIHgnbT$j5Z{ z#*&fkhIb2^H=ZXhpFuQHG{Cv5mIdpd<9YIDYFia&5vo()zV(QSn!DSIED%{s4hwVZ z!c!b&zFKWy-RB%0W@PEvZHtN8`=&%a5*wQx`H{c*RM-7%ypN^Lzn;bPS}3gN;m2|8 z@^d3**yg}#xk!bLm*Yj6{X)KKp0qSq@wCJ8Ib)crsw#bjPSOu%_a8GwB$vE`U-L2LX=072oO$j|`q?Rm;C{;8#kA{&mE& z^pjKS+k*NU%qelky->CHeTD&pOxEo!<=bwOfmBh`#ndL;=z?eJAG*!&thE@SX^aFd z0BYq7s0~SVYFgHJ>;ze~u!!gV26W}MT3xn)_r_-J{%uRpQt19v(bmg?e+zu7n9A`c z$J-f=-+HyLxH^(Y_j{8YFoVKt@4Xcx7hvSyl4(= zU!4s*-4;*w7ak9-(gWbaR1^=od5x`FD+9Rs%@g}w!*fn$90Epq?^iXV{ zqHRmXg{?zvXg0?=91#Ur`QqLOa zBR`<~b8Tg1L`gc?kBeEN?K+5Rp9*sjJ1++XAYb;qA^NKT>bnNnoHF~n0XG#dz15qg z-GrQoh;RN`?{4tdOrtp6;0zB8bWW+=X%ADc#_u5NzK!7%aHFnVgh(6F?A}n20Q+gq7g&HWRMq)UUpRtzyF2Qc=DL z$|scO^3!ZHugcx7%$78?FRJErwUV&1##j}Xedf9m}Sm6b!pP&n1wW_vo z%&aw6mfhT~G%3V;>`9D=EQr3bGbH%QT~RWvEtDPc)0ck!iw_WiPdQUp;s;Rz-r3*6 z39Hf27llv4{^8~ag12GiLjb_XBY5?3*r4dp@6#GTQ5DsJlJ-fbT+?Ki-!#28aKrmN z!-!`ek~C3<+OqRq7jVmK@csYZxo|>sxzyf7-93r{rzHaQJ(_`HDKP5pAxic2LeDJ* zwE}NP-%6ki<)|=*U<-_w%7!&koeC7TCKd=>@8~95c)|@f7w_`LZv6uMsZUJGm%Lx7@5Q7{kB62V||oNCxva zfL~l^3%Iiy0O1)S&i*tf0gSe`%eS0n0~_c|AYwBrKmOG-&LkIKUTdw#lZ zho4h-(U+DsPShf4ODgc!?S=j(3P0N$trBn3vz>o2_+vyz9^5`f*3V=afGd)`$6+o{Y6EcO1d+mROv_tYIVe!8LeX5hvb^NG;a- zwf>2-L^>@~i?B1@PH4TPh~z8z-05&Qb9~}Zq#dF#2?6w7cD<_Mq~us;w|JOmX7|Nv z(#}K1ky&g~syX9p4o*&mfrV;;ct$sDe|hb}o*svRD6edy52fMjPZmY%|N4T4Z{~j~ zo4P2e!YhqcYI+tOeU!?wb$(vU-poss8ci%Pb|lKO@e)Cnas?dIqaOw%9G|n4ff>e= zBANH8?lx}<(RYbYoxreP*7sL4wPl;P1F;wr6S< zwKCXX-t!^28tSg~6Wt8)>y@1G-V{NJPZDqPt)@`A0~^8!M9V*DiU6<_BUx8M4E9kd z2Geq_HH-h@EtY>o+DV!9hOlSz<&3mmO3FOW!UFZzuLHd!pcqp-e^q#`ic`%S$ZEK( zZLz7(agXn)P=ov2^QnvX0x0MM&ta(LPW=|PNIKz>FMa^5kk3Uf6k*Lnb$MZI?c~qe zBUU2bkGu6Vf~^dO=llDbsi0A9Cttg*lX~aq2kEEtyQo?RoG@W82AQ!j_ua(EcFQ&M zuIPPjqV?u4*D=@hv1`B3*Ek(-RzG|7ux=kee(z%@Ug~^ z3I`?q&H2%b8m>kGoGa6GjC#x|*|6Fj8Ah^d{eEKJDV^%9{1%LK2qeG8$V{FJu-X-u z)^K81UjtPeW{(Q}!m+ry8DO$N`x^V+#o8aZe0yf~R?}e1pPz}jVFtrW%!|}1({xpz z%)>#C-!30k62hNEO0_FD85Ehdm*PHP>zscwxrfgL@!EX~+}=j=V*O;=4i?jS2 zy;{M*yix38_xw*dB0x7uH~NUYbrBX0I-uZIWF9Df>xyrtFVejsay~#i8sCI|Ffhsw zBb%!FKJLU8i?xgo%ImM3;>ZpOc-W%sW2j`lBb)tQiN|*R;?7UB)&;)bCi85{8x-V7 zfE7woJ#OT*t8BocFdM%W3_FTvB{lu%Le8_znK+TcZpvoV0as(7LqQ~f$l=sZrdGLU zKSu5Fu9p_jOxRdsUR)>upoj}UNTa4{6ncj`&Fpysfm;L;P+o=FUK4pAuxTOYrV<5f z6%S-Z$qQ?J79ujQtFLLk@?0`&5Qs{d0Nc~aKNa`fTHH13ellpvGBpie^ZUu_M z!b8SD=nfBj)ogQBrXSQBd)FEc$bL?<-iR1A=2`8Y?&S{p9&tT65 z#A44&QeW;%K4yFOC)MXmYB#-!nfjV>l+Rcj?!jb4xG6GMbbkwN_|n$>*6&}Un}2p# z1=LX-V}=Ny>IQ|*GEERVO24-U@9PFQ0^fZbzyRkTD0{HVim|0a@hOzIvv7hHeC%^m z;b9`{tU6aWSsqtmM_((5Ja*l<9NE0~1YpcA%8j0=O6(vMYZu(#3z6vUH=S0&q6EBs z_x4|}Jyxofz2`8P{^GfHP8OZ5j;>{tY$_bMf%+&p7Bxu5wQry6T7$ zh(!#anTeEytu4HW!!wOPDoaz854I=8acR;Lg-HG3^-y}X(1(H*6lD?FloTU^7ZKf6vJn>YPWK3&q!7ExxvEnMhtoZJoyOo)iKHGHb`nz11b zn0sQojbgrX%051H_AY_FCMOB>Y0migmwPQX>q`Qmklx=m7JESw<$o>XWa_f*x`sjx zSHeQA`||B$X!Y6jON4&vAgncfW|Ydwpv-6*8X5(Ii&#l4h8L9qebuAuoS~f(9A9lq zLP_cH-X|$_?g6koD~9oC_Eg0|M{f@`wDdj0C@r)%&VySsIW4hXa-dyncq^c{%2|4z zzYb^t1E86XCot^a0JRZ{Qu_K|#-M<4M2b!ldn{XY4a4h&iuk`vIdC5bF6EkB_Fh$G zw~OcYTJ^LTaeFHnV9`Cd==G_@Hen=rBA(nOgWVMS9VP}NZ#Uy(6iv5vj?rW+_KC!Kw%?M zvheIXC4gW1k{xa>(K5R+$Pzrz8k{asciM{LO*{J&d~^hi?vd6mCu{BP`(mS5m*yMK zPbbd`t#4zpN57gkTFy`tl=Uf_tw2K*H4o2o&9`&`s#sJy9MdY_=r9542CxSYrj`sf zTJH$7T50<2wcto9;%jb~V!9g@<>UG;e+!;m+-h*J#~N$@^L(|r`~_8G(!}wH?@KjQ zUC305+;7jaPTQJ7-mmhG!u(37?`OoGl#W?~3!XzNBR`@JEiz9DPEd*BVxq<4k|c(X zuagDyph{Vp(FRM`zqd+YL5(_y8J7&DsnrSK`g#;z+4l>FfY|PuaM!P8L+L!Dh#|Je zXUy_TyYc4J6aT;8ZG}zQLfP%OgdtEJou5xZ{ zv8rKmn+SwznN8-M3vQ``UG1I41k}(=Nj{PE(GxIlLJ~$h=WTtvS0imc6-WJh=Xm9;)1`-Spe24HEvwLC#jt(sW zS`SBt=P(y3B&X#yJRpy8HO%>_y9-=3wQRavd?D5|gM`flsg~26BTMyt z@&=mqgF`jf#zO?_W1@jl=V&@W_X|)N7Shr*TY(+#W4RX9*-1w6h2)(88ilhA;00zZ zxachpxqP~1WsNlQzFM$M7x+G}-v%f~p)3 zwAh_@+l+5E0nOGM_%u!VVWhEN1Hnf^)yTT_scEF8P@5<%{O=ob)F8{1a4A`oF8|lz zM+DJ#^$(AC>M2-A-xak_3nIus#t;5nmPhGo6}ioD`dTS4GfQ5ywue_|ObAz%zJL%l zas!wG?2w#iUWIr9v;lkL-YLK&ZHMH&6LN2wIvUa{{@D2-tlvgtQO~7~{=>sj$sapSMSvY2+6uzsMiV2)A91e?kj~(4TQ-lws_i+q zTU2iZVS4yjqYEqzrw4$XYaqY;fCBt5jg(@kYS|eSe1r(jNOi1Bb#D^qQZUO^i)-Ot zIVQpas*eG?p-cmvX5VzXv$vXBwR(Fy3ZT$SUmv&qz1~SX`{Y8zJU6JNiyji_v&)|g z-C3)|U4xGeWBr{XZUPLQn)c}CT&R~D$vS4)sEMRaAfG~+3 z0)6o!m+C`gMB-vnEDhyCKZp=PxXcQo(;qsKZL`|zRbsn@dA-7~=VgPbqY<5mAGTsd zBZ&soAp=fo@MTU}i za#MQ%sLmhN)*F5PiXpc4buZ|%NY0*J_VzpU`bTY!1oSc|ap7QGTxifau3l`&`m$ML zUHsh0A}HS;L2XvMzN0mQ?p>nIL$VvnzgGut&|lC{JAuAE>fvyB+D{0oUv2G{Tp`Q<)TLY(CatCSI^~me1M_BmX*4+$oY<-&ACKUD*1$X}(g- z|I?X}sVM#vADJati$otu>(5)$Zsbscq`4&nQiGHuglVK|PEqcuT{r2d8tk8W#F1md zYk}PRjaq+2xZv`d$wQ#PTgq|leyIbId0D701E4yK2cW&8%ep9&{^Ts&FU}CCFO-JJ zc#aG)f$0E0E>KK6Tv?cB(uoF*D!1=u$`m+1KUX*pX^zCi1>;utOG^m=n?gH^ec@lz z8jNh$-3Dfdoyl=V=$nOFsf4Z|vlEUckh9DgrQNgO={uz?|V`xJ%i1c`!S z?BpG06~=J2cYF6O&iu7iEwGl=-h)YZlg*#?t#=XlAxR%s$Ky`pvHJPcKL*|Vk{!B8 zbW-&Ca9Scf-(sH*&~|W+!Esu!ZDFV!bfP6|8G^EWEzTylJ(Z~7~J4RT>XEY`u z0Ir6;Wx}*-7!D_;{|nBy-Nw7>btFevq1MN2c7?zDm#YmQpNkSLfX4UWm=h4dM3k>m zX{~+`to===f<)PR3`weLb-D5|>2ANqO z_i;m2z&GE0D}k9u*K94LFoH&q^L6J9(z%qapDNHU!2b^o{&RNZd{x;gV{q6Fd z>%JBke*%5TqDZ(IC-DFK#7mXXakwK0wj<8AB9z7XK7MJ-Rl( z3wtH118?F-CrU~UBDPYAg|26;pvoG(Z%v?n zK`$?Qi>h=8UsUC=|IQzcp8;&f4Oc2ijkV$#%xwd-j5({D-0SaqjbvE_487ksv8{tXVxFKpR+J=z zk@R3cL~fTiF(cgxH}ZR!h+184LQb1BV7E>mr;erOYVmTs^eN$Z~Qzgk(E*sLuX(bZA@71rO8fCimY6;v_6?@w!qU z#G6n!B&d_?t&DqvU2)A=rNWUG)F&|Y%WnFTBi55&MU3w*z1ro^u0B@3Saun(zT#+~ z{?LH?K8tE`gnqN1>5KFvoCk zMZ2g;4=P^y!j$AKSzPCq&N7_yaB!PtRAe#<1r z(xJRz6AiB^NaDR#t+wW8-8bp~@!iQ~(ufic$W|YrfotE$XQfJ_ExP+NZ3XCDi#ZVe zwmMPMfsr1_rL?F`3N+i@5iGf@CeH|9EDay@)e&W}E?LBDQS+I

g4%76=gRsxk9V z)6=S3$+!D{Stq*)Z*~}eM5n5fnocvPbsWf>t14zj%JdXrOjn%6p)@Bpm{9ZXZS}7x zRP#t`5*;V%C1)rpoOo`$HhZ6YX0=wd+qd{fOWTeSS$lbZ#6qI-ehjLpBJ*h;(9zRN^ zu9m;|cean1n!s*@0_Prqp4TY(blBSZ+BUnh7u9K!IB!#O@>usxh`QrOmhS0xT#@l$ z&s6N`rnQW>lhEYb6=-clA@#4V@GdK!vU&sR3W`|=PPx&O_9gE_y<6+sCi@D$x%B?t zEtd++7*QOd>wR))O63vNyw36Q+}Z^UyH4u#>6F;%y4$Dko?B2=sSe3g>6pjf1 zA?u!H|2F;d8ZGkP5AE*R=)`USR+M||P-HxCchV&$2W%i)nTHWaHQ4#1d+@F-5W;5Nzmm-;!<=Ta%KwtU#F6rO0R!0DE*d)5Q7@pKEk< z8+gro0hoH!8~N`k`{(C?beZy=UNOzAi;R-lzQ_`gMmYiLY8+4-YXRU{nOd# zfnuTafMwWK#RjYmlir$s!sr@$ELh)Hm_ydS?fL`$OJ7W$<3dG?Sv_Y#C3DsP@6e10 zb?`Ou1eHD_aaPI(wZ%wvm*f?*S&UNoShUzQE*l2?GY=Ck(dYSOPHG7p4tGlF{@Vq%QXE7)g(s_QJdir)gdWUJVp#KMu2Hep~ zSm?4p#UgH7>YV8Y(n(O~IHCLYq9OY=BB`C$J#y;6K2kY7!CyeT9%VN@^>I*d1V9F~ zXwUIl@IKoQQM&8OsIm{B3F0A~gm!q`1yY>%xE70=L@3hIp_^{ISPZ+SiZm6vS%*Wl zJDy^kC}~1$2pTmTb;AAcgZ^&fn((o~>+A3pjiCCv%=((kzM8M-j&aZfCfb_zSYLEg znU4z)r{U`FGY7+cq%*X5*B^50VESn@=EnTQu=j7R7umy-5Iy1QbHH^QYQ?{H& zldg6?|zojy$ z+2FY+=0y(@ydt2J8$2$p8x=kBTQBqr*54x4|42(G7g3RIm6|gl*?2jH9lei7HGv?^ z8FcyZ8}%4d1sg`(?IJCX9z-GvAMJc~WDGOP$iLpwCy-R6-8=l_-(? ze0fKubTbZ36$Pb0NWzD=-&{?!1&>)(wA{$Ux~aL8W@-9TNR;bCL# zh*YA+~E|*K1M=BQnFL)zI8s${G~EUN)CuMNeLSE2$<1ujyfbp=&xg z*}>_4dG2ZTe4UCBVvHx_8z5uPblDsr0zK`b4Eka_5!ZzGhqUJe2!niOBfHP8qO||7 zm;9{-4S*{=mj6D6aA zXh53Vfe^yiP%M2!6-HD-_bb?eAB;Zyx~!T@xsq?moV+14`d^;_c?%MG+m%|Rl##3> z;ane6<;%s%qY)ttZG&^Jei!iYbB937`n};~>F1u8YP#sZ%aaU#AQTQJz&GO>^=W{`Wr`QFq1x!BIJ(~?x0FF$a#OYFNdd{U4>m?}I5G)Jy^>c9SlM=Fj}S0+y1 zHR=-VXL0oW^7Jy*`r%VnCrOX=+Rlm|QG;brQ7R2*jXFduVfCaL?ngff4{UbJN5V8U z$h&jDorcHVH1K-|jT@Oz&ZK7HP#(f~GOF_~0AD))W{eZd3k&yDAhAnsN=2V{e;oAW z_62P~$2cJTnTy~eA|e5Jy2CFE(L;f@;Cl|;2@}$djowIefPlP1dpKQ+<#_-zWqZ2q z?Reg`N_cqI9L{AJJVnvRbH||>NJ>a_$WY0Yeo0{t?+thtqN2&YaM=zPNZHaMw@UZ9 zrlRLiwt9z{)-G`MmdO#?*d^~n>*$RGNYwx^aO!1)$ve0dp@>i)vG$75r#$r+gWJF< z^p>_l>Z8yn5YTXs{1$9x^TC*oMVr_qb7{#QdOC9|m2mnSHs5g1{jLW@IK&sKy4tb) zfE2PkQBV?;>y@f})6mG+ zbI&3?JyA=>V95pl3!1;Dl?FO#abmL@6b0v{QU^4d(7fE*44Ia`wV!_L4Qdrh(tkxv z&Re@8$Qp9#xu8ceJq@z)2rUMuN(;uTr~dX}a5IGWBGv@^{ti8Esg7-yk9iPVc2sMv(c zGZFH+4t?ci$c&qQLX&2{u(y2yEhNgG+|7S=QQ0880 zXx+gE9&5P+y6SDfANiAsQQzd5i9NP^H9Y-Mrd)qKg1^^=>&Bpr{P}2|0G;n8780w1 zH&C+7G@pEIIkHvmKuK$R+#hjn1IwU-P(vpi?iVxJK!yiLx~FKKmUl0F4pYwd|EUVN zEN5%Gwl!*Sd#txOB;_KVX3>h12kRX4k#U?NNfbg$9M!cZq!Y(#+Iwhf{PytoF8Hc$k@|*KLZUu z{x@MBC5@}!ld_Ns)$%27_R90QKh3i61$3j5m;fmYtF~taor%c1y3N<3m7PN;Hbh~- z%${dsV?+ML3uYR*0qn)~B+|-G{<76NSFf%K#Q{W|uX=&4>BQ{p&s5^R8J8@i2fu%i zn3J}Su4Y8RxE5MMPv{QW)3vP6{wM%Rq&YIDW9 zWYRek#Rw*=@^NIP41=u_zLRJtSN};P>h?!B+kRp~h+7&V%-Xd06%r2RHbV&UrJ?%TxSC#f_AcAB^%+XqgV`S|ttPC6cQIlcF4JQGa z?qd;6(d$}ZZHfNo#a3AT7rPn!sQC0*B@l{oZpeTkqer?=@_d;ajUm#HLZeWR0x-s0 z>Ue?Rt8Ak;1%0YZYFF(rU~5pUJbLA`XscH2wOi34x5wJBR}5w#3&;0Z5#u~8(fn2j zqvyc3n3K0yM-O9qI9xC=^nHV@FL$jimsZeDOF2&s)ug^qf9J8^P zRdq2x*O9K}vvalEuhmAb>aq7_?du(9ctuCaQ$S2~sXcEQU?C-9fhXK{9(@uxZG$K{ z9sMX4!y9B;+SvHE;m*DJRX5P4OuUd=dAypq{;2n+ z1N}4nd}1v@5dH3co>Ai8F1>-r?+;e@&t|zD^Y28U9O73WGkiw?K>Ca_kX##U+OKRK zZIXl~-Q~EAOMm=G4)};7f$?IMOjWafw*OJ{P>zN*tUMZR%zYe6)TVG}+J9%}CsKjph zacAaLBEaOvbX{F7ojc2Kc+6}qNo2np8BYpOtV}Nq*K4vD!H7y3<$&r5l~T5MAU!vt zxiqzYHYK1b`dy3g;Sq>n@=lCR_JC1%2{tw&m-sC({E+l?q$6ld6gLJc6mS|nhM&*A zmB?`~@Geh@gNzKsh4dcKxaX99kcE$(Muvw1zczVzd2W+mR`XRW6O2c#~(J| zizIsMid6oWr2ZyvBR}bjJ-J7)?+cG0C|i%`gt!wfUoa)MO@mG`WJw!^23iOj#OBD! z-ZQ<=Ar2EqqR6k)OJXY9jYOpM8gsP-Qy$oYg8G`od5>o=~V&NCOA;|o@(ZH znTVr???@|P%WEfe?fR$7z%VpxmU0}1IPT<^%BP(gRR8jWa?hq3K%a{#C9R2sZ?O)Z zG18rpH2f~#Tv>uEk{>I3clqQ$&SajfE9y&l$=NQPrFg_-No>^djO+=G`N^JgOcb`5 zTBPSIh}nyHw$7(Ochp$4zr1MlA7K4!@v9^sH$OHAY8)r}(lE{nv%MSP#qm$wkSb4- z1bg5Gbs80O2UR+{f#k10MW>)Q_B|l0cB>(muRC`$|Gio8i;@ab_5&LJqLQOz3#)od5AJDK@Fd{Mq zENY9sORbeETmyD#66`9Q2b3j&ov$gpgPEv=TP+E-n5>ftK&FAd`k&EJRE=&?71*a zpkbw9^Np^0|2k7MpfUCw;+|1*u8P&77?jxu#EbsX9QsaJy@j=%+_ydo{OLKeZHGTe zmhfz*bwFPal8Ej68c_GxrGO_;7{O1Q*MPei*KrMZYmm8=t^lBKN@l&(_$)2U0765@ z{RPaTF}?|!P>G0qlrX=Ke}<1;#&n;Vnj}mD`+l^Nl51g2xdIMZb-&Iw*1YtZoK2-5 zKp!23IccnkOXWfQ{`uXK`$o%6;W&e+%?&d+Yr$5dAby1fL1-a9Ec&r7Mk9MRmmZvl;VZ`R@(brf($#tB?*vLLrQIM8?atEvWN^QX=qK#w&=-nqYD$` zOeT>e)J%ItA5>pyE0Z$a8qgECFNS&_rpibx3RmgER6xya zJsT4`LS~@MBg0E=9=#X+x(ZZbLQ^E}RCoWX{XRz%WA@7#7!esPAK6$}u!y0ZkERoX zQ1cJ?y5Uqv9(%=6Y-y%#;-*a-6aE-ThAgBOqfTn1@lBRf>a_z0YmtwiuX{yCi8{G{ zV{!>t~3a^KP6AqzITt%u$+SKp{m`irIP#2rGptxjbm*u43`I+sa9+!!2w?6w7X2ur>+W+g+Fb z*Mt*MQ&dEhk#zv!TDcN?OX1UmhlqMD_B>JtDHgimSDS=TRwho2I^I@r$2!Y()#9QC zV@oT~OKH=5b$b@4{=pGloicR>E=E>axyS!m#EllX?`y2Lm#^1I91_{sgHAU8c5nQe zd4+go{}UU?UBP2Nhyl5Ow*j^UalaA{j{iqhDUEosFdad7bi=iDC8dS*clYakfmOVR z62ELgib-3=%smTDQ7kSyczvopK_i`I_gzw(E8sT3l>s^ngajmuyYE`zivJemjm&N~ z#XI$9P{}#DAzFC{k&s?G?lrEV2T@^GPsI(|c}r1HxF7griZ&6xT%C(UsB9ilhWwKU z8D6sz{zpm;Hb+Rox+F>))oX4UM9nNw`NdcVti4q}oR`uS(Sb~jC`~BC zaEW7lI4^0j&#;Hy?n16D@;{GJ{!silQ`+EtPwhY@oqYMj2cx>H zq(~HZarmLW);$KP)lCL&oQORa4WG2*41|5zq z=zwJZ^2X4p_}$j)cUmDo&p@lcFm;nO+QuQ{8TU&S@Prf{N_w;@jIQvJmN!I_NBwur zRe#g4{$cGI5~oWV!|t~3j9_=}&dbxKD1{Z=lM1_b$6i^Q0S8Ns6Ts(ktz%_2WO{{@ zZP?|LHf{{?VWkF=3^`;^2Xzz5vsu8WG-D6A8?Qh6fHvE96sUYk$f|6fAIP0IM5Rpl zbf|-{$7C&ucc5j3chXv*lmkWu7TkM%P@6lcaHyy^5-=ErR&tH0#o^9*S-`BjghAj;pF~<4VKfu@{wkXYw|1T; zU1zpD^bQlN4oMpKxA1dWe5zn!QSxopR1VkFa-x3T$Zmi=Wiy_gywY+wl~tlw8)X>y z&hN+SwfR-}Dv-gt9q$lp!|JbaV?VB|U~fK(R#+bfl9_PnlW13@&9*N5bpN6J|Lg4t zX<9>f>!l+pxDLvNB2V2xoKN6(q-b)2}pRFnlADRsB25u0Ncep%SYiDIhx$Rsr=YZ z%T6?*B+cXnwNX%oAH{RWM*v|g-f-_s;4ss?zy2`#sy6xba`_Hx>~8I6e-PePa~0(= zf^pIGYuk`{LDn>^l|xu_6L5wNXjE%4Kdn)h9vB2%{@!WfB{*pFIah9HqVM?6RxeBs zdyceQwehN31E!PVc-EV;MF_iGP^KnA*-6!l+e|bi(-iSVSW0|hgdVK4wSe_x_%pMh z)|Zj*Nt#g}4iB3ViqhJk9~>9oi=VwRxe)`knP;p8`X**(fZz@@-4rD`Sw^bSVXoD; z$JQ~0EuWhBADOH7Jg1`_){|cBI$eHl%y)j5en`@p(x3mXTyH~7>5tmZRr$vA3IP<7 zD9fplZ8hWwz#F=j*|YosghW%obq3q?9>|vpYjL+nY{JG?kt4XTwglO(ki(!myf%|G z0GPDhkHn!XvzH1HQOZm9%GGnP_eR>tJBE-(SP*=pE8$>@fc*sh~Vpxk5vd=@ncaY&og1TAX3gdu+< zBJdqyXi-URZdyZHP@KANdoL8QVe>5*0!LPqsI25A-hh$a{cTCHi79LyL5`t}UNv}s z&aJHC4s6UYc>bqb*g^ii;*F`((NW-+wN_Wx&)r0S(6qIh>h|}dcHH*6x%{4b3iKK? z>{NAL;|f2x{Q+6yV1=j5_vP;@f57?7&g=i_O&Oorx38*a^{5T#)epg!j~ynGvR9@h zbCyqmt(@UODp{}>y-W1FXu7G|_$s+PPjCFf|1o_|i8=nkyWw_B>6$aqSOpLCG}AzT zD{(zPkpKS2+VZ#A7c9TWpR+(fmu@_EX2o%Lve3%lo9?*CK8Kd^LXjNSOvpbVp(l90 zK(`kBYZv{{!T;|ayNE&q;R4x`7fYF^Bi~^ z(IctW*hQ!&Rs-u$h$Rgf&&9jQ3A#IM%&kej1+Gva)XX+qfKte}nFl*!AaI2EM@e?A>Wy>-mRHc zo2BW9$mHCQ_pum?>1fY}7#XWXYtM+?mOEYp1Fer5Hjd z_nB$gsp++{NrrI?N-bshQ7}%{An%o~rO`s9Zen~e7=gEhyO))kcbch%b^zxK_GuxY zFJBrb_)QgM|LU$(vAX8L&mEWJ$Kv2tJ^R<%pYd)_f4y>euuw%qvjEZy)$uQ}{>IH_ zxX5acpjw{MQ}|lznOu(yA#QVJS2fhjorSZh#Cv~^h|hU%8a_Iip0U3({*-#iKK{F= z<(nGi;w`!L9YP%1^!D!~tSrn)-7W`u6K)jZ-ldWKDq~6O(vP70JD|q%EW>?q<8~}* z&{Txd10bPe8&`ctVu4;6r8zYI_2%1c=UEK}vhHVr^%(!Y<#Fq=1fNl6lWA1;>y&%5 zS0Q>{avI;Uhb(JO>_N?rRZW)yrH;SGA$v;}3%Dqjhtkf0-&2K4gGVO&Tx1bP{3q4Zo8k)}vwT^d1H&H-gPl z!yh)4n?{?jvqj-V*w3(^w*}QQhjD_d+?&8A6fdNkC^k+DbJ8#oNlwRK!Me3LL4usC z?F5nuu*E>UC4)eRX> z{zlW1Vt;5-RFK`VCLq-A$`kesT5xR>OGitZE3GnRr(sZ@vwRsrA$eWnlw@%bSf7*( zoNhiDZi#L~r8fz8Cj?ko+#NU!!lwRvNH#Jkfgue7 ztE##=9b@>H-sQl=^r+$kc*4U!oS4y_kX=b}u+y)%R_8tj(~f@rmpB!P9PYDx;KLgZ zAh3x;wae2nobPMAVP4Zg==@h9S%OpF_vn)~I&sX#-;s%oLo9oGShH51?JfUJ=zNKX zT|sM`HknO<0$#y8k|oEfU^msi@PH}L!}2x=EsCxYYy%4iJQcmmmi)I*6Y3X64y21Xoi(6aUC+b8v%!0Z2?y zYB&~W69Uk1!vPxBvFeFP5+HQ;Q1;5EqNq)3)XC(^m=RT1arNIg3M;&*L?uUy+ zzzI}Y*B5h}7jUKbyz2I^W_oygn7xXB<8!l`(~aOR-R>US4bYMkG4F5QN`5!W(|VG1 z>qcsELm0ztxXb1}qaygsWWb#tJ#XL1vWubQD}PQHUjR{;s1Y;0uW@PiZy)5s1A&+H zbs7!WyMzrtqDe+7UpaX!b97%V>P$Nvfkgyr_q%frm(HybNPoe zRoVmHI~=vjr_sz>NPlwP-Qn4?v8KwzL-#tnyI14ax$z9m-`p`6`&gL8nckk$AK0ix z?rX;DzW-NiP_-+^*YTlD{if?eZYp zYI8XAF?}x&!ut8y1W`r?Y0W9ki}{D|5O8tYE@9e(L7=9SLtT?Jn%6(uv31 zKKImiTBC{7a-exVVB(e&Z&qiHR~IP#VPK+`WuOML$_5+X+J?PNNctD6Q#5% zalW_$s8C7`W#ztMo&+Q^lleVkUBpx|Ek!OB#|F5t3e61Pm4PwsFg%YT4T? zDmOV9uj!7TE5H39y|H5Xt~~Q%YTRdh`MY&bvUSp*ZC`oBDPf10^tvPcOX9gwA`r`{=s>Z>3I+v9 z6h9_pEl1WY+-&Ox({fMGtY-K{-Ee7W7{tiW>>)%@{{q0W9cqEsJl_wnHPc8qE9rraTHZvhJ7VSi_1xF!Xv z;n)%o2BSvCz6m^_*$S_J<7h>2=H-uWf7j;-J;vw+Zl`3fSw!gi64kT_y52wwMB@`b zHoGz{jh4uM_pR!6aDL)}Jv)L$iCha>kmC{$xjo%TmemQ;5ZN!j45Bwx&H{_b&>&B zs1|9&#C+!HG~&!^q%JV5x&`|iJL?GkC7Gaj%r(Q53ReoGgR((2cLdAuT#SuLC>J_@ zTl*^9kJT=iA!62}=Wo_8nQ7OslE?7fg>-uXtDs#1-1RU!h#=qRVxxHdi~9iOHJ0&sN(?fN$P9Ru?2n8(G}*8~b<8OadK zl(oJNx|J175|L${w5XOQVovR295hiqt?qaQ<8Y1Ca3%)~iV+ZC^8ZErB{DAZj|ba% zMp|k~D_FohWVh0@Gi*NfItD&qZacdY1A6IER~bfYbYyye2gU*bfh~0v%y%(=5#BWL z$YaS{*r0^e#$ysysN~j%1LVi->Mp=1zFNY!BN-9>WA*AjTnDqG2XChTXsCl^6M!NY z|41jCA0G{9^4iuPe}CAoZ}3_$fYNA*?f578FtC+U{Gl%-{u8G(JZ&BPTmNXE#z)8F z%XNu$Ozo)`*=zx=}KS5Or%lFKF*7=-#wrLb8_N&Q6Olv?!MZJ7i>3ma~KSZZrN z_E5EZJqo{UaVmZCb2)yy*69#*cNT$UXk3|hf0)O8EnR`5-nIX*JaK=Q@Nu8mcjipz zA*RB~;3@_fZ)uEe-BHjaR6cx#i5CYa9y_*FP;n65-p^YA);vKi>miX>b3@M}@Nt6D z%0JHp#BVcoI|W3gye8X7SbGC1&YxcmW<%O%iW7`u^tSYsjd) zc_fQ(hBsN0m+lN#c{_m#NVP>_hL#Vu5g|u)olLkm==qAz*4y?^&p6k* zKS)^ktYxAh8r3TXi4|8_5y5lVE<5@b`f#`HwsARZ^}ulrb!d9T_4!%%DxXgF5eI}! zCZ9o$LcG4W3hj|FL|UpJk=>uu=76>NaN68&{s^Av?4}9_WAnN!hX-Z1VB6?YrJ@$uJStecUhX0nfDb-6GhG*d`28x;-s!v!ba$oV0=>wR73Ag1_cI&5=@<0ncme7HC;4ELnIVtbyyBk-Xy*34_F=#OT7H| zQhpRLwwD%+H62g=9yi|=IU^c(gYE@4cH^?rm0%^hJe~*N!TpccgyvixF2~iA_xqT8 z*Dq9}23JBNZh;K8+h@axrkkNekH6r_Qwp&v!ih+aeRry@$FYBQIkf%06u!*k<@5!7 zTaSBYHkC=s>2&jo17qzeZtJh%^`psl!!P=FKAvZdtO+1VHcK)MUGmFF3Uk&;F06_n zyS^HIJjlZ;^p2+!0sVuaNf$$^nHpX}90S118YLtP15#csw5q$H9pml}!B z85TM~q3l{waT1;Ez>00rv?d#$ihcNErYs~Qk5@u@XQU0-bTUI9Tuba6a`46#Q=n?K z#Y^f`n97g-y0!A2bwAtYExs#u9EE8?4tjq`j&%>I76Z%vSM};SsgP9?i9RO0%0N6x zW@4gq@0NhHFS`UgSQ8#B2=1pOff?}-g@H?+O)q!xF>G-k>$@BgqWBE|v@>STr*j{` ziJUbt3#gEV0H1M_?d=5tm;Qyw zzV5G79X#3n+Kv{s98`U|LBDw{BPc-o(%RV{5-UA7*ObFw!gY+{{1VhUkCe{*55HF+9h}6y*K`p$LSaO*@N-UyaU>e_t?DO zr3V4@3)0`Q$0EnnpxCJF9w{R`Be%!gB)6lp=Huy;ZJR-J&WUs5Y?F3M(HkpE%S_dY z=PR5=XpFI4spAGK@q3_0{cR#+n7FYPBGdi5p(qqddWKLFbjpEHCpDVqb|5`}z|AUz zmztxi!%8EVZd}FzpGcv`TMThC9*9;4O&2Do*hzqP-D|KDYG*142a|dP{3vSE*S=gc zK&!hcfbF3e&DTw+o-t3mkICUd?& z{55&BbQX40h%_~hg!!}Vy}EZ4?VTA*2>J}H^D4PdrSvI^-R+ez zvqs1JD%U!R+?|TWagU!ICbOvSmbXxHr-Zv3G~j;&QrPqgJOKzGV9^^le*D=CJ@ z>wG6!p>j$6ueRuvC@`al7zA1|o@S(s?|;4zBpvUz;_@F0{n&g* z5rnv@G2Tb>A5Ptj3B-b8{w-Ta;r@K=^J;XLkXyB*@_Ok!pZbjezQR+U1tydz|GIQ;bIw%2b{fUh-R~RY^ProJ;`VC|LDzfT zgK*5(U2h4^bJqrmPxQJ3I3b!*j;>!BSj7aB6PVz2wV~piKpZ+~l+!ZkK0~meK8R#U zLn)*^)FN2V8SZJyQ8B`{>^R@p)sSG?LsEU#`yK<>8JVTf?c8Svbm`0P1SQBcvdDb) z-2FkC#N1a;U}WA{Zmh%4+x4Ru=Gd&GJ{)~cOK5~_!Q-K;)lljYe0SUuI~q9%<@U^! zoCNa0?ciG+c}yVLlSbKIYOlcDiA#AlRiW2lpB?x0tJuuTYB9=m7VZ+Eq;bg8J@ZLI z5Ui8k$8ghrFd*xnwZhUk2>tPu_OBD~f%tmUSRVs1JkI9}PHM~+Fq{yeqc{U_Hk+e- z6RF3%FU`|hg3EQ){Y=^G!%c=L>db_IEts?ZpUy!4a1(;X*%gLK=nULCNXnie?eJW$ zJbW@J6e_CE9;4}JT8hRlc$WL-G$Y&9;h_Tm3y+4yPKg;EFS@pXvtf)L3^BJy0O;{+ zU@L}D5Td#;In?!t#B8%5C@*?Ezp&y?c}3oBtqmQDI(yqPbWlCN52 z6-I#ymU2gEyz?C`rFWj^JlZBlhSmN6R_;@dIUD#wW3;hXm?V#IVgmovB_o#wp8v;y zc_E*9$LUn$yk+=IhFQ)?C$DN$kd<26$`Zc5D+P~$S^fd`-`qp>5@b49z%6I5qta+| zZ~rj7u8W3nRiMPM50)taI0tc`7W2mhwD7rAG~AKZf6!3Q+q#2@eK zoX2E*BG2usB(-(7h9zpJ8Le5G`yBESvP&mxouWt64N8Cm3@5nIhc(`u)^mo5{d;j& ze7yf_A`!HIgnVZPERT1WIwsvnCVQGotxlJp*wHMUBxBHh;=zPWBX+r*kORy?yKd+H z3WSr!Akg{lk%F`dU`WGO>^9k-6R6ic1-v{3Fx3TQim zrunMg3z|3O>Jv>c4UEPy&^AkkdDKVH3CrZBD?RB2Kb|NL}-ceBsH_Pnn4$&bmKTIon(%X;b+ZbkHMCv{3mRrb4>R zpf5QVf09?|Y#7-tHHkQi6p3L&%a^u(>Am`^&8k< zYAN`;U`HL~Q<6oyu1-Vp`=!|UJ{BcIOMP7Z*kp%nVRn|wXuC}{Hl_}G@I%h(+q`dY zBjUY*2KH>G2m`m{yXxk8&+y5@8s&zORsd)|>lO{UL|6_Cb6{I)qoaX7Z-wt z!V&o1^5G6>D3}j@{T67r=ij|X+PKlc=%PPE==_Hk86Yj52p0tNFGBD7fLB^_t-fG1 zJ}mD&)b02)fCeU?tnd8Q>vH3NBR-z$+lnhN*5HRqF*UnFk<>C0QHUHz(#*qOX4yG`v+3T3{z4TV&xm9vJ`turBJ5YK31s{x{{K)!$1GzNu`z(j_?&F|>5L-0YfLVCL{D>aWx4sFCk zJ)Xd)zUrPi>BP!Fe#$~O>c(x)j^%$97(9xPR*42~xVfm?VbFmql?XAwUKqSziOkuG zE%fBi{-VStQ3JpEV3<9`t||?(v!`vMSX2Bw%6cyVXd~zWQ0i0WX@1Zn*|Qg7i-#el zaQ4$P9@2`1d^Anw&8JV_Q!z`_*GG%*ShYGVrVULCfguy(Bh>?=WOZelnQ+XmZuz?@ z!!Q1=v3JxfXP&ww;`9W|47Q;}Af~Dsh(ru(b4S%zZtQ@Ohmy)j%qcsWYZ1pld7GxU zSB!TXk4`(j3_l@V`k&f?2GvE3 z@sh6NnAx`L2_JAl`0&4qC(qn#I>YPsFWSLY6gMB=6~QKy20l!<%oG;@dHe0P>edG5 zCs*Z)Ab3M;Td$3n73??JG1?n1S=BcrC8&)Gyo+>v6@+I+ts}LoEKx^{cj=w@98le^ z@qZQXz1F>}bM!Or14M2zMII1&OM(Fy5^@MrvY8OYR!KK+P*lacjWure1FUp6cnz^_ zOv?V~`5yq`PM$6qgJ@~9{ImsIIXB4;b|C=7pL?7v`!g$ zN%ciXHReMzGJdlwgO4vK+0kR6&3|3tPbk{fV>|t7f3G9`-^0H9_SyE<2mN0B8KRSU zhTBJP=YV_CHoy&(AAJoask;JuSy$NK+&?D-N2z!ggmO+hyhZx~BeGl-WR)~Ul9)?g zwf_fl)qorcB7As1G9GdeOg%pFmPkf=Zi1dl06J>j$xz;HKcro)(oMp8rc~}igbJ}I zjEYItsqs(NGq8GJH2j208GH@74vFp&KVHK)#)4~94|uTf@6#=z^gu(8xF@Kp(%=uJ zUo$4?(@%xhc~E@E{g15ugk#PEy(RYZzA=ekuQtu+LHG8_PXnEst#vPx3+|}r;lh3r z*%a_cRwe`fPoVfSZ>RL|TCu14Ont{@RHwGn2(o9=vhy(OBPC^83+6tz9#m`TFL&rx z!8;aKHV?!_BUQk19A-(*XN3(|X(5g)C0*ZAajw4bx3e&2Ql%yDatsdRx7@qz#_grC z3X2ktuTUj15`{G%4UnpVMl3d z8vEb5DHT8qWG|d#)b3jtaCgN7Q2%5i-&~ovho_2BUzx<3Gvu*VI%~XT78f_n``x*9 z-AD-}lVt(uSKQ|~&7eXl)0r|A-TF^yF2aAVdp*Q8|8m`J^>+Yvq!jqUV*L-yG>xU@ z%8sW;Kf*I}fQ@{n62m9%^GqMH{-(LK-pIzsM+nGN^>dGs+PzQBmVx+5hNu^(W9y*+ zV5Oo=#+yUCSj1S%ud|-uRL(9E@iow{pTQR_x^xPvq{&xjodKZ346vS=e>F$kMn@I6{LueN-__>yQcJ`Lae7?_!}c}e!KMLqsnoFRSeta?u*$dw=eZC}!> zjKg{WyIY^$G5xye>G&LgZp3C6F@qr*QtwloSFUs$)pQlcTCTHZ4SIatnT#Bb({qxd zNt>f^Az-vxoPKd?d6J?Ug#GoEzVq`uV7)QV&%vj@I;ZE}OhyJt#Pc8u$OR6b>KGu0 zR>AY=2GZZ+&*ZXCJS_dbSvKx)CpTKXZ|Cy}zSojS-mJ`S{+xxdUjU1*_1M?evai(Ho)f;}VN|q3*dW>GKY7&-Sl&6R6t=XO78-k$#$6$Q?e{@Behd6_Iu?IXuZwr!5 zoK#vcG+|H>rS?HJx@C4s2(I;#QFzDt+O9X%q+^a@dB_2CTMfe%XKc;gYa28fjnwa< z?aQQ&-m7Bk_4C0{Dr7*VWHNxg=!{Gqgx&Te(S<8GA?cL<&S*I5SYYr+TKGmx1d^0C zdG%Rv`vr@M#YXl~M9`z{3Sv--&V7ZRI}_5*MB@krU!qy{@^FKJwn!7ID~NhwXTC#x zvboL!R?N4aNZS>K-!+&dP$NhW_xfvrby<(8{id3kDHoTFyjd1~>9fYZ;B&eEck8+7 z1_bDBM1EHz9sz5u{lMSlkmJKALVHC^##b8FK!>ht*u-;sBQYAs7zpW}0`UuTCMoXWWV?aG zX2}I7LM5IvucGba*QuRL)d>uGT(#*r3ToLsts@d!{x9Mn5Or<{5-Onp(i`qtSNcT4 zS<+k6F-v{sd*|LcmPEHda#*dUMFiol9I?E%lBxm6kzc1WE87xqVWq3C_xywoPIk{N$HG*@fcmx z!X}yHU5WxnZj0v7GNfrl2>YZob(bSxU@yDEvT5!Gx!GWjFL56O152_*}tk* zR$~8$p4wR7_|xlec*LEE_u40|-A zJESFw|3lJMg+4Bkh zi2r%7|6~rIgSqzFd#!uj;XlTehKbK7)wtR%{&(7-rQ+Mq-EI}n9K9a(=8blP7x@jc z$+HzYg5+4j2eG`gZx5JgqYQNqXb-b8`8n<{^7J}<`lL$L%EKsLai)de?->;h5cia0 zS=BZcVb%VXJGwG+eEPcXkI&;aqG}?qn8Fkwp{yp59QF%S)~QC-UJ+`xVo`X@{TqYH zb(V$AX`7f=6abRGWA9WLk*Jd}VgULWtnoa&YQtxTX$xGuWP zy?U>V3X44UK60wGh{Caz&-+I{rHek&C(X;%Mwmy&!|J5W;AfR-9(+imYS2CE_NH3X zx3gBGIOnBVFJ=@#nTgZ0C6u{%a}PutQLEWJJ~jF|BC&AFIhsf7o zF9sY=SJr1##wuG`x=vGTRfi;+S@qXeEd$PTS8LH^2RME?+O%pn&!SHnKKt zL~b-wmq1NqH+F7E{EU$3rT^BZymIMsQv7SoH5xIU%^d!%x{^vQ{wPLp$g|lJ`~MTL zoj{?NFqGMZ7<`R_QkrzkOn&3<6Hc?uX`hkq6fVr29H~8SnY=A%2x8seW>#;)M$KS{ zv)Dojk<8J!*U1>GkJf+&O555z*l0k;t3)t&huW>sU@=1}SnJ!VKEA=bb#Y?@PtFxp` zu7ni*Jf*WxhQKoQ)zR2S?`Wv?5A?E(3cNltRXIbpTjLRgm*E{?21V(zN8MxFKlqnF z!enHNa~(b$XRuu5o4*|AB((vLIQ z_dUYBhr{RC`F|u36doI+i7%*;a*2^V!TT8RsH+*^~1%E@odaOn!n&^!Npx`3qmyL zc2-+@f)4+`s5Pdpo{kR-N`omMDNWb>-%wm~oztp(GM>{@mAQUI9S5*{#l}M&HYV%* zW}kn~!*;gY5x$GuqQ+4-+VPh!4P+(&RRIe~q36PJsna=N0zMX-Ua8)W(pLM$2}AmRnOVHtMKLWat#0F7l430sOyna-T>$?mXJZbDtm3=Zqta5m z@l=9Wm!bYeq7q(oVEmch3HL}?u9YYC*L&H43kHz* z;6Wu&cXBbpWtw8k=L_%g!|%&WS%ySoAVS#~YPySr zeA^aGm&WeSM7`1#m`)*>XE_+p)h7IBylZ*$w204rR)ecOFL!4YU*k)X&){V3xLASl zq6#^m(3k^caEX7M==rsBQ{4l)L*OlbF&Zfcv0@RE5E5B69TSV5$-wqZ+zaNg(@IhN z%*9&vhvq|zOGhFFihRJx$?Nhll~p2K*UzQP|7e1U-F1<_Q+)4m{%FtKj0M+XB_|P)gVy#w>u%d)|CwF6@$;Ue*S1hk{qOWmNCXJ*;AWMN=?6bH8oZ zf)KU0)aNTYAcx{5@O7i+FXG|#Kc$i`n-QDu!e9B7o$wKo=_@Pu9YEpQ6NJ&_sk_1H zRr=$~!Or4$&~M^4XSg}iT(jx?&Q5q#bB#tKWad8EYFX_ag%prD%P)?hYu(aYE->qq znCQBZcduDx>sJ;&aCGjFC;IDGP@40zFIS3b3`} zj$<~y9Gx3A+EPF#u_GNu50^H4C$q?s9`0`kUDoM{>2Ma#wyi8OJXaa1$F2me5S{t@wSBhFYr9e}m=y{!WO=MJLt@U|=Xc z-ccun6?NsmYF{)Pi!*!jzHG!v*fTXS_36z))1Pd&U)|}T)C#v2V14-vQ$AnkMD{$z z+vsQMv3cI);Dt9x_S&?=v+9_4TA=t{x&gG!Vp}i3W=|>H?9T2k>JjhZjuVNQ*!yEouj&n#+|#{> zG_Y%b@StkQ?L6tj>sd^%G+FUS)N~_-?1SR@TU5KD-ZKdN4(7Xc&xQ8abYLCzpCQ{? zeed=@2L2}Aen#}{=`5pXvFm!(5sTjKGu#{IyeewwQ5aY2hz3nU_`3gwN2Y{@#e{+J z$30}23WEDC!-Q~@!*}o5H=@5p=lpwOo>9z4Q-P<2qcQZMbyXsS=gFQ z6rPpeY0=*#gn_XGZ?~tOnHvyiU%@7w)`ApY21yOXh?BfGTXX-$BFF%? zE~aSE!LG;daDwt-=w^oPCr<|?ENetVxz)YS^u}`~XIm*eoy1{IB8ZjtMp)0srtspK z4HgCm|XYF#`qYBbL8$8eeY83|0Df!j$2*dX|x4_hM1ei%R>Bv3I33OfP zp#a09JtYU-^YuqM-3(djA^GJu!DBCdqp>W=I5mMRbkukS*W+GE^1VM`B3Pd1e^B)( zo9H!Pme_YM2Lx9>edf<5vh*axZ4m5uMkMX~`4mutF9YTI z;^Kw&_D;@~D>968evezxF;fs6z`so}NuJ}Kn_&f)<$PGb3U!kWbSXFC>USNxJnR>R z?(R_8h&A?c2G!K*{J+=(IPt!%AuuXshC0;@gem9RRMKYiPwKHN z?Wt=h*KzQb+aJ%tNx`Z$3$gnGDoSCF8X2xXL}UyR*vrMv)w_ z=)H10x6t-G zzpr&yq%Y{}t8)!}uo)9xxnvur6tqS4*7emqWy9hwoxFn3RwqZeKaF{;`0-yH z4;~Y41XzUOUGh7z1qC>|NHpw#Oddj{_7O_zCf(nQR;L+f1ZHGsuqw>9q|E*zEZ1SwQaHnZT0M zE}Ii(mW$sim{sskx-CR#8mx%TG}mn0MHS_vm})(63=}>;$K55B$8A#;tX*X%n6ZmWyUMOp6zD^Jwn{$^zMAw@#hkd_3Z>WE=4ZTP5H6GoEj6aDTKa|cv$%q=f{Y>E{HYy z_`kf;CRs(bsWqikzr*@rr&hSPMmGv^LSz3j4L}ZIKxo8K!FCZrF^TwQ`H$x*Msjc;IIjNX*$n@xmKV;uL$bz06S zR+=#_w@O^+jhjKo;ii$(pQwpQDuf6C<8RY%k~nXEDtVb z|Hb|dtv>pBNP?Ro*?(VTYb;b%v6a^)ZT`!L{htjf(Cmer&jK)5r3_$e*O6Si^XmSj zfAv-5$sL6x>LW-PZ=SKO`xBA*o6jUTXyCrq5-`3cEgB0E&@w@&JO3p~2SNkZL$U6n z2}*T$&7I9y@jvma%#}Y9!#{WbuOMqRxY2bl{DKA0*#}6U4v+;ITVL_0_kWF1x(x#* z#@O&aAOsJ}LQXdCs#cjZ-1L=dw@hs7d0Q<1BEDafZiwCY^fkntS2OEQQgU`IZ-S+H zoBVhFNHMlTk0?yEPZfR`yqwLzj1=L6+NXrbz4d6adKAL3HPY>)7}M^K>gx;xvzg?> zSKnJVd!GxjTb^l1*~j7Wg_@g!X*a(iYqt7J!R00y$3vcH%}ow1+sz1Hdi5|HKeg_T zr-!%C8`DihT;fP=qYYZAGMmEu2o)852mhB@kX+QZc!`F~&xJ)svtxTDM~$1+vIUY5ddwu8*N8AS%5ej?*gY74%J82#uwQJ4g{z@2O}U#^wk{IPk3R>FLLxy63R z3Y8`9_epUfO3drJP!b89vh5(K+vS194>)v_p%&{Pq0O#6wmvR&PYG_wP? z6Qh@~_?;J5?w5{2_CwA;S7mKSDO_NFweZPC@1u0+TG%;9b^NXMP{N(~Cu@;w$xJ#> zaP&j;+%btpV8P_TuZ1RuZrJg*3vBFuw=U>PM%dL{26Z~TgOnh&G*}7d{(blDl}p;o z3Es;x^R915pAnye#y)s!o5xJoupQP-jfnG71HC~(*iy=ERu*{+*@u451E0^n)qSVN zW|MkGT|YOT=5EWh46qcShU*{NR}C^`ae5?LnGOk!?W-rLLqUS4v>{kYk`t0+dWMuQ zum3e@8njvozV}!gO+HHO1GVI)Js^I0)8rkERP)!kDxXv2 z+s}K_o|}`_M)ak_Rg3{CxK}dZoq`(l0_~-J3!w_!Lb)^OcNj#_7e)=NUX@%2LLLm>+dTb9_JGhFhDTd7A1MIU}15| zkzvX@Yk^Dmk+QVA?hin2Jo371AZ}&C3OqBt1|(Id%5!4^(D5f_3TH~q)gc_+MCJ~r za+w)LzMFZc4f};VZJ*W`uwz)fC2BCjZ@Zjr1(s406SbF>lnq$S*Mwh#NO3 z*Jg0NG*;em?O+IxRqu%Lr^^y^lR!)Ch*AwpgvYLC#)KgsGQ-+d1g2Lu%2u~)#u%7T zZ_H6F5yS|>eS_}DWv5(6^@Rm)Y$vP|CfKLz)!+bfg{zRKAXR=%mcXaJ$lu>)UawnT znEiOqqA9l@`nyGm#Y|{F653hs7yjng7Yk3ZK z32cOvD?*mjwLt~KneyxS^wDiWi~-hLnUjS5uU%(wQj0U1x(N2ReB8@&C?q~bF3fGN zAvV%RvIE=>5A=bV8meD|L&4uuSKaq{RwO zm&41g2pv>>(aS*e@*Kna0O~#NV7FIgQ*m*)H5N80+x)x@k$CY=t1GNI=N>5&+3&1H z%Bg6o+$uDATGUdxK@ZM-8DEAZHqzB@o_#ANAwxx=f3+$5QC5ZPtP^cUY%rh}+}L@xT@&{fv3cMSew5n6eV&3x6YdQajE-twFJ zr_L7V_@M79#pUt`S0Lw;mD-?yE3Q4YYNn>L6WVK^%Zr1)$5uy!L#*=opG@N8^-`XZ3xO9$ z`rrM6+mOD9HyMU@e{v(9;J2E=3)rYohXnsV|59^8FHp)^2f#hg+tju@;DFf_v{A4t zVuwdAeQ2S9WMALl@91YHmFM1c86`bz0@9dwA0fd9-m|gi`Wnt`!CXm9%P(jV6$RNr zC(no`CS;#hoVG>^dxk{6+c&KR(}|j-Kb$Bw7Aa0puPLHlzZG5nyzD>z5zWq;unsCr z*j!Rm;GDB(EkjmQElc7)YhPf0K%f_G>kMh#qIU|JSn5%x#jr9<;Am!uIYK+4KEgUu z>&!39x#I2|7H-l}#Vtz2JK{Lva^`VnNwak?xNDL3^3tZd=iq(x%t%viv#l212Pm0Tn8`{i8QNSI&a3Oo&AHty3E zPf=@-x$nP4N8o9&_Yn4_L0pv!gfMH&a&2e)!|G$Ojcm(yLHb`qb68GF6)y$de1hZ6 z1no8?kd1bjMWZUDp68bA@&gQ-q2`P7bM7*q&{r4<$3D z6bvP)zo3To1RC2(Sx0Zwkv#Uwn?3sDrqrd+&*X^{i$xf`ai=jH7_V1moe(u<=egeM z%$^gUGQKU!SLi2CX+VsmV4Xu>@8lS+vSV2B;ylTLu10b9-G3i$mTa)ts;bCu0J@St z*R5?C%8P8D@+yOc!miRyTuyryfCX@VV5@iWNSGIyXhDuk$M4mkYcm9($<`rANbwtH z@bL^HS)k!zQ&;5sj=4~u#3}zoyXNtr{lrMVG&30QbdaXj;<-=iI9px=82&7^zc<~I zy2SV`+C63#9dwJYanfTFz4%#nqE1I%vt#4Jp%uR>BL2OhuDAY}>)hmf@s+ta594;i zAVsE!NkM}&yR69gz~3)S_vTIZrEV`*1UiD>2y3~;kY{e|6E!}8$vg}4bIYwebO=bW z;^XyEbEkngW3^I*Ke&$qjUT!{EdQjHjnyip)c$Rc8T0cr$dlxP_#)}{JlqJcL}yNI zQ@th!@5(!9Pn)CFLW~K*$jHwY-fOBmmY_Z*%5y8qah>ZQ`-w+xph9gG2E; z)aJA|Jp&62aCzKJS_y%l<`gpU(CHclq%Ul$|>HN^oU37Y%?1+=?0th z9-cd@`(QWW9JyZyY+ZaR>iAnGy2&q7&x3Qzzs%6q z)V}+{%J$CV8W}0e;Teuu^oSk@cg=yYDO)e!oFbz=o=W;pyYyR!MNa~Tq`qXSSr1sc z#w6IPb$qX8x=!x+C2RGz>`~t?lL^5FQ?JoayXpG8Z@78N9b1?DV?is#%%DxcpjC8$ zUOI`RgC|O>VZh?PUv(6uC3F0F;o{%xoPKxfUt)KBMwZmaWLV}RI&YaOqvv{fkXR~$ zrCJvYc0PS`L#STEkXgKeJ1sgunN7xggzH7SlqRJaFzlxg(6`Zxk2;^Et`-5>7aI{f z+cu98=aJ^?rcfuMa3*K!g=LGgaxKCqEHvh6BfOOsyg-zS8VB;ji{_Jx(>F}KYy|bh zHlcTIEusT}CsDo3`M5~1(cOGB{im3MTHOD{+0f%|=-MPUN@L&`pyiyDdvf?~pUTds zY0?@LBpY5@t1&a5`xjYIijU#1_~UZoEm0GJK0I_J^-+Po?pc zsDlyl1BrC{j#z3>E&Nwx3)k`TdFRqY;yp8qC(`h>iAaMw>8fo>N2d)6Eig+Sx@3nP zy3C*Z*z&iyBRDdWZz{u(NY_Wz#`Xdz^w%wta2o!=Ldil8!5^?5iP2Zi-)W0#b~@_A z>~-Mz|6wLY`&|}>{xN-KxmDtcVw{C?X7|O{CAQ=UT~9jPyweE@2?R_6jJeevqb*gS zB0r@d3NT3OOgo-pZ>e#YbfhlH=YLL&@Nv(i{QrQ&9ou0+`CroKqOJ$s$S48wL>lE&Q zurR@^s#xEbb^V0WzRq8`(W_KhDSOy;8wd{&alDUzy~b02#=bk=Fnw`yO}Umx&$YG0?p1yxATV448-7(RX4@uIR)9D^;|Bnb2tZ5RC=*!!Yy_g|}NRQ&lcv7FEE zo~eg-r+o1AnXyk`==pg6z-J@K-34(9!K{TSuBGQ|w4%c30q1%_r|`*Vth6K~$=p8R zW5?B3d?RPne}puMWfD{&ufh|CcyVW|A}=~mAM7iCS;)`*+L<7eWi=Ni7(9*!u?hZM z3?X8r8*C}MPtSik_hJZN!tsGH`#xDY_N{eTJ)J3 zjLdzy*|3Ek*$(;g1WfggTukMLIwDqYiwE&n@O8;wxEyET}p9kSVcWv>6T zaewyo_L=S{oIit4rqGzcu|{n6x663zE9+TU3v1IQu@SU8kdYk^2amQL2yrQat#thf zqccVM-fLeQxdYPu%aq_zVEMg^_1eoTF&Hpbd7stzJs^m@Iebp)w+DH`2WmqQ?iqbq zk#yDcN0+Y4_dVl4*rnZn@FQ?3v_&rhqS*zZhoi^j7T?ut$Xaz70&K%4V=Bk*nc}{E=a7`4U$Kk74NzOrZ9wp^uaXPUV zU~+J&0FmWzp0-5Eu_}({cZ-BjlXhF$)OGRND$aKVd8eMl_nWG(rT*=2mJuspvm8)n zx(n@ERC2LWg&8|1n;MYCI>d~zdgAIb7%oH`#`J+v{~*rt@}VKjQAUNu7O*PJldqU$ z^*S$JIHog4J{Aj2;_~S>4UqEg8e86qYFONvhz>dR8rXWsiR6gPHS7>I9#ZJ@OV_Um zn|+JFOma4~Z9xoJEqYv%PNX}5I)z1M^YHAOFO509mIZ9J{IkX2GsWC9Lt9v zW9OxfiX8+s=3H%J5AS|Is3WbETR1+^8Y_2ErcW2<*@)Y_X1Pd-K#O=U`Dc%aCl5s6 z4TtX>2AR-4=q3FYRL^a-b@x!`A0IyXmXYCh&4!aBBMs3nXLmYWj0i$G_XXdul;YYl zPOs}NjE!I;Ecg2PBA4(Q6*OUS=HU*+jWl-z%FKm0uc>f!lioYB?{QYaCP&o``=f9W zb3723Q)-==kJe(LJPyn%IRcuD`*hr((dWGag1}r0>YQ6A>K-3NoW}S?sz3DLBWAh! z#h2LorM$=MDEUcigHYR#c6R49;{bkmr+J<2Fp+Iz{BNj8)G5>95@BcT0-(;~vqx-I z*-O)Nn;|S87G3y4|NbmmjeP14e2QeK0&th@3>XnCCelXln(OlEE8ZF#Tln$TK_M@F z@|?49VoJon`6h>#vUJXA4$jNg#(=~Tz-r6A_P%Tc7vLjSB>OwOC+d)Jk!qD>w-pJv zXutm@?s2nY#zruru50edOf^d~pyGiBWO`Yi>XpgNJOI`F-c&7K%;oG+_E;ChqGJOowti(cTr>Db3iG>f;lEO_yrm4Gs2&TRDVTHN$%R3Q{8Dz3B zlBj^bK;o$hUnElyb*Kb=h{`8iEoU6n2|5+Iw20b3xlKp?<-w7LNDy8nb4@|h|N zYv7JdOtZM8eBo#FH_y$>HsPv!9rEJ0ZV`p%ZH&yt#3l0kBcu!l6-AGvckq_j#2IKN^l%Y$uL)zjIjGq4WH zP?San9wn7O-sqS@+VdB^Q+f9E*2cy=s({cRi1mCaA<*kaMNbgyfBE`*;XWO7qu zvoqPmXW~NhKJ)#tsU5rm-DOMaFwtzds|qm(O$upoy!L=){&n0{=W4_>Gu?NYud6Wn zXAWFQIbugv^G-ibjC)E<3hePXyzbxuU{_vK1avzUWuk}vp#)a7~* zDEX?LbS7rjPUF<;m|x2Oi=&>#mWkwcE5)!Ww23C%_;#DwuI|qn+1+>61>YEASBCuJ zG{(=yKekGGT@stJ-O*MM_blID0!;p|aoR(D*2=da$cD!JR`Y`Hs^V3Sf}Wt&#S3oc zch)f@24w5hZkjiVLNE|^`kGjI^VT(+rNzOqMxF0CFS$#RVXY{W6s&)XR7@P#_Ln{G z;eueNN;whcq23J~cr;;Tr^%C6(p7IHB!}+151ITKQ2h;uY$$bzc|yye4@7znqgw=_ zM`Q^vb*WIyD-f}37Lar?ed#*I*>lSY*1F$}|GPaV*dM^A5_P+mnQLXAJN+gH_ZUQU zl-g(EAe($Wr7W3j#F;+lk=rM$KW~3UUv-U9>dDz)`>y|bqb=wLVjn$Ewal7ySy2$> zt@ZYU-p}DGuZHqp$j4;?4|CG4^z(Gbqb7%*z$E!8G}gTt*(vc8#&gNtc4d#m!I5!E zhN7V2F5KyN6WH%HSml;gmNl2Pmo>r|+_XxMOkkJqrzoRj$xcR}6y8$oUmn*FGn!CD z0s+Z?etgo0qc5nM0g%J*r#7NvNXP@?L+aBv2K;3F{G=~lkui~$52|8#F4*tfW#-G& z-U?E)#JIbbn62c^hav?=FJ<>>etPuer26uGVj};OW3J9ym5;Cn_n(632BAa5Y;tnr}ziz4mV@RF01I)x= zQAKJFI};>B$lxAfC$+76^VZG;`#*qFfzC%nKBAWm@YQ(Ip!G1q0fB80AdIWR}Qqy5I*S(@Tpv6JWoeNGI_JlZeqtuYC%) z-5ydCm$Idh79)ZAb#}gaSi`g;q?fy~y?}_pQ_iYo55+8gPm6&>P2P?la?N+G)u7r? zt4I_Tzq-I1VJBKY+SF$|=%T40q!8ro-MBL{Ri}@aM{e3kbvEvI?6#1wh2F2l&yumu zbv18=<{(n9iR7tzZ=&CaIk?Vhg}3ANFI>T*o$#fP$jw_}f2u>CEYPhaHSwuiW+trd zHKs?RB@^{#)-`pdH?DWDhuWwTW2e(boN4^&F+?oYFsbKvLkI3(XvAxt<w4=8_o*q;_{*OD+!7Z-#B31b6Nzw9Gl zq3w1S3o(>XrEh6Kq^Vb5x~e2#+bq`o-32b{=(Bc5!Yrx@<=n}=GU}u$HOc5%v6;%X zhMwHDciZu?&J=5@-W9K2JrCqc5_D$btoeTDQhcS*mAFedDK;}HctG#`tJQ{wz#pC3_$FS>VNyu8Uz1kmzxJz7GG$s@U5kvd4f?v_ zWf6uJ@^; z`ovUme7N5Agd0VNub9p6*X~M=T;3Q%^FMqdnK9DzpZR8Isxx-kN%=rM9*!-0ZkC)t zQUE3&ckeP#&S-)6U)l#?Kw|bHN;#DG)R~?g3|9TAr9vPNX}#UMOnroTzZMb~Q<6#l zaVM9`+xUu3FE!>_XtkS->Rjwq9#_7%WEZ$ra3Xpp=v_XB)-wD+!FEBNGHh?H%)lm< zE-q#KU!fKAS&Raf&US&Lgl!v9bca4YC^&6jP9?21CAk<$LTD!Xn%^!1K;@e}1QxeV zzf-wj7yGl$fMrze2eIGX_h%Xb+usZh0Z-KnTZM=REmOPy67RaZ4bqI>9Az8YO&Q{) zx5vrAoKg_?DErU?=vh_{TCD4R^!`wq&n z*T;Md^{!LBUT3Iw%#hDUyavWDcJqO}jhs2r^6IWeGs2; z8p^RHV+7GXeO@Z92zy3cs_qpSbs;4>d$78f2B(TIyRkKYI{2G7ZW7`H=8Aw?8uwgl|c%DfTFw>>)*qNn@Ib zLzK!|(3*WxWM7BhQl1y%kO7i?SKba_@9A`@!l?S4;95ENb;-4YV-B!iZ}YcZ>Rb_1 zLC}r+YH6S3z^w*+$8^ho%ukIOJ>UI}1PJ?6jg8T=&@%Qi?lKYUGs~DEb}gvE*V|5cF^N-)RKA_F6 z4%OH0q0XmcpagTH{H@$b0~5 zenEa3T^Uy67UjOc2U?%e+gJ^L<=Cg;_ah_n!|-bGFc zJ%EHMrg1=V^$P9GNnA@0d>lbrCU>Pg-C5C43(u$T(#-M=eZ-3Y1|WD$GnVIov4%ox1D_^ zl%9-HbM1?xk^tEiYBTrm;&%?H@(n+xSUeUoZj$^^t` zVam<(ooNRn0U!LK|3(GZVqRWiW9GNF<($Mg9=-NeJSovnxdDZ0Hk%qbfXZ?_gLf%B zA8;V)McMd4^x&lWn^Xb|*F)n?!8K^VG@B&WGG>GL9}7qM+oY<-f3K}YmH$x55``0? zQgL_Ljk$Nd=S)DqAc6l@l&H$w1kRnPYtqi8%PA@f2Zakvmc^RqBT$_Yu;4!a)DvZJ zF{!2KR+cozKNb=Knzlsm_zmh3GLmT_=HW_v^u8hnH3@Q+8`CQ}GeqK4(h2;=eEQ`( zMzA8)Fm|4CjOWYYO4nB=g-BhjDnTGQ)JJGaf1&F76?ENny)$MP>&ZjbjnOTMqe~{A z+C9yng+VJa7cUo)du}S%ZvN@0)a;i%6O3?~MDwzV%v8syyXi`x#iJfLRahdXik^s? zR)6QX-oT37PwJnv|ArOu|NZQ9F|EQ?F`tZtLt!=QE|keb!dgx5 znx>J%O%scNcB@tfx!IV{-B3tO@l1itwl}{AFG?Q0!q6&1)2*wRtFLk@vUYf&@*VQA z!HlRO1#z0v9o9r%%3zN4|A=kQ7oOSNmsb~A>%$d6+#D8xFy(R@jEkVMB#zsAEAk>u zTO^E>!Cq=M!56^24!f?$S3~49MT+xuy|Ilw3uiIP?a@oE#zhrKEk#N^tUBlhB#}Re zH`)qH!xKjg7*8bK)6D&7h$H_iyqR{usrRtp>As=>FuBg?dIJUv>E41<2>yHi2pBmB zu)_trZ;B{z;LuT9Swfg%g`akZaJQ)6{8ww`hQhDiSXaiLFZ;>Bb*9DRt$i zXV^PnyjO@0{yTda=i7(8YW6MT#ekfz=I)u_GB&Oer98LmMkFc4Jg=v{4vI~W$#UB) zu{~GGqUh);<}+z1E}Ag=Vl`VqwKT^QxmxnqY`n4Zen`J+fkXOc~svngq+V|E#fnE z?q@q1Q?B;%nTH_@CN_V|Vh&;E-RHaLkscV~{Jq46(H{JN`BY_gUTx=Y)0Y$T#oJN5 zvRd!l?N9@oFF)4|1HUXgDkP50xNgU6e1*u^ZD)z7d~EWNn8)YlWHn2H8s@w8;ap8A z5^?OlXRz{J%;`&7xq17%be{uHL-ZWd#x}#s`D#48{^TiDM|OW1cH)+~O(b`J@=o_! z{kb9&8ZR3)Gw7JWM?%gWs(EhyUGl5fJSF4{cMLU#RjKM#mY3eO-%6W%Fcd6r{Q<;Z zD;USe4OUt&0Z;+t<_dgS87bh*w$SL<;7(lnzAb;Kn5MRkar)E13#_VWk_y!iDG2Q$ z%>^soPd>Ta3ils==9!c`7_O-hk}Jd*_sNSx zd(u*WvEc~bH?ue@LUtE|u*YxMi$D=Y5YyrJO#vZy|B#BLS3vsYGN4yL>PN+ma%5wO zh=|m1=-5YAvTW__bOh_^&fNEN@q$-l_p`emv`t_ zWup)i8uPw^-+aVUk*FC)Dyj1KmvyO=TQYIk(PxcyP; zXW{IjuKIGG9Htu+Mv`&MH`X!(foA%Gj)l2tYf?SZ`$UgDR*NHxO&YXXCU@PttrP~P zM!aWQCkv0m!Xmb}Io{huZCptxbCn7a=Z1OHl~By=6z&%voykp>X~grzs3pflXVy}) z=VwEr$bUOwHB$WKi@wVJ*tSf3Ghs!M=X~SRlFql{UY(lv294B8NT*5OcAaqzZ zZ-dMEdey4k^G0*NMBragk-(WI8X4cD=w>5c*eiEJ3*4>1b$ST&=AGrh>#H($do*8{ z-dnnocuV3f<1v(}qok|kqyw8}z-#F~KMi=|!aEi)yrj#K9ym9$1!2_M(nq2N4XrQK zn5OO>9m2=rv6up<=&o4t>ulGp)aX6NK>(rj%T@@>M_U)vdiy);vPQX{3z%xl49cp49Cz_p8CeyxYB9rQ<7;_qHJ?R;uG^4Th6m>|IP4R?= z`me+)@snPisR}wZyTF&@jltC5m!#iRv?!Or@bV(Wy->#!kV@7(CfKu5+wvMPZ&gG6M8f3VdVOp?qO&F#X-9K) zCU!y`TJ{1O%jB|>h1x7zOauknL)kMie&6-|?{KKLe=RpA^kKr706WWK8UGVOGveZ$ zaY#PBs(}7i9(<%GxxXgU6uTQ8)GFLtqhMcTZflk-fe8M%xKO=P$Z;OD8{{N@Drc+z zWEY7y1e>8*gXN}#JCDY&h+jo3(1}$g2Z=+dP!=n}R~vJwhp9_;%NZh9^&qk~l@UXm z2A<#8p?=AQv><;HP7J*=(q{xU=w=*dCJ@Q=VeXivO3{)O;^IRK_07O7Q-u=tJI-BB zTiJ2&{;r{Dhx;Ss1?%SGuTU2C4I~uHjC}2X$3>~6&0o&5_pAZ@Z!Gikw3=kbZRebZ z%$*|yaBN#mPQhIoHe4*IoSo?|B(oY);7ux5Fu(07`@uk1G9~Yg%vRZW1dI2ut!(?K7vayAcu0113g-1GY(~N>}O!p+|_` z;LR|K&xCZ&0RY$u*+1HvVhQ38y>8KZyJ|SRWRMTBP$+Y&t#*DoopRG2vZKiz1G8@5 z4@8sdr_2h}@v}szR@6hkK!n*Em@TxSH=FSbLPhU595v`n^Y3@%YJ1Yq`Ts4*wiX~R zO#QaM{eBvE>DcFV#n;?DYF#t$FZ2Q8!`i%qj;P8rNiYpt+1XeNzR}jy@u30%6y!Fv z01-b96p1J`rU!60e@MVp4XFP{5LbU+(=n|QM%(?C& z?-!$xj}evXZ5iqep~~o8_Le`K0q)=WDclklv)kj)a28blZ%l@SO=TafFN8DmoAU*8 ziBi^NN7ZxUYb_-u8W(7mi!_aRfOntIlzw7ppe%#=K-SCbyy;Ka?@O;Z?wIndz(U~f z7XhGor4@)Zp90mjMs>fvQA>ErmlPkT975?akr&Bd9>BfjXh|nmyw|=zURPVlQRj_` z6T=sEfX7N9X4Dd(l%GfS0?L>+RFhK6(<2I4n#rW+^1h(W!cfwRz`~?{$`*6nxl?3# z!1kJW0H#vLYOTJgOR@kX}?vGDWwPR-_!8%LKmvi4troLnYahPJqg@ai|5 z-@?;fDz^^JAR*U|r8qxZ48I+bhA+06dZg5t)0RQm5LPU2F~`NFhhwSzOXDmZ#qmNU zA9^IFe2-F&+2u&+-9&$r;rIVtF|k?wU>&FO?Kym1FGph!v{kaus!5w>lMiOikv5((NSiT{CedbwJd` zRi6UlI7j_7?vS7sI=1|>k3XBp zfEMoK2M1^L4rzapsGUlY3S~~P>rO~MyhJ~vm2+}e?quGU$hHVB#$K0kv1sT8Ajxiymo|qKrks;)ia$ zH?38O+>p7DKazh0HG@x0H|RG4w_fq@(_7HxwxUZ|&^g6>#CQlR6DJ&fu5DC^X%1!d z-H4>R+Xsvm;M!9A>~dyg)RLb{fjSD7{p(_+-a+#9l17z5a{V*fSAEA_AWr*}v!lH! z=rZW>b~jk+TZXwmkL2Kbm2+p*nd6rrb}@!ik4c$~Tw^W&j-7@>5!l19++)hUD;yxd z#Ke`3z{Lb^hTz|1{~-m|JbV&reO{v!0w|jK25M}Owck8xBawvE;XvJ}iA)EMm8wm_ z%~(I-uzaVNYA}`Fe@Xf*S?C;_Dh9imK&so*8*~D>E9oD{hPs+ID!+%m>`R@_V+tfz z_WHCo@G5ko`=+*`dR#yRvTMNBNvwJ$vUoHu!Bsx%~CBIY)6SL6i~$d}9Maxmk5QSg0KUGNPi_na+| z6U~wC3+#Z5hmF~0Zwrbtx!1J`*?-S!w{4#ox3l`~m3QJKp*6bkmAS*Bs|E5}9IX+A zp5ESJz!r_UA0tigH2=@DaUN!*YipGw+tWqB1dSwkXsRLzvicfyw~9}98>E49AL11E zb1}&2YKrIobLe=RcxR&PqM~8VUY#$knTY}vFt^NRDzSwVU=Kx2-UPLxrtL^#F)RU8 z?ih8TV6#6~+B<^UzWVA4H0q+96$ zqeNkXfV4r$5Co(VkdRV3r6r}C_eOv3pB{&Q=$_}kp6foZ^ZZ;6m4a_>tDZ=w#QJ0q z;XAW3g#&4mZWbFb%ae;u#VMo&99~ZdQ5J97o1`7ti{K2mam_)^`&_8KcE;u&TynxtFH?*yM%8ryDtNZC*QIw{1SUToSt<3ky? zo}SW&l!);VkNmYagAgCgx`yXWjU zb8_FwzsrEO#Rh)q`K=3ZTJr*fdeZz9!jKX4d=a*2vb+4uMNdjdo-h@P(cyc{f6tvO zALslTFmS$_53E+Db~h}aMRvq8O&`|S`u`HX|1Ph!bZzS}&!YuWDe5zGmGA~ScPU$+ zYAkK`*Au0Ul+%>zQ4g;-IgwMRFF7Dr6UiFVwmXdCo>?VV%b2lTBq?Ch>`~RFIpN#F^Q{YY7l z$J^*ZER$L*5>1>FqDh{j<(VO`WEu4i@tCHU*W*4SYF#;7YAtMV%Uon!^aas6I)AnLFN(81Ahhr>M!yDZQVy`|{r%iEryLk* zl_{n-U}sR4hvB-q+X~1@n|z=CVE!uaf}1nrXZ>BOtb&FdNxbaNPHuw-P8RYH;Jp`_ zZN}?s%!!)e*h?jRJ@*JuDi$&oGYL*?3?>*o!+UeCa?pQ|BiXT5rC+maWCb(e`z&?M zBDyeNGj()=|M-RDOlxO_`V7O!%B@@FHn(qfbpN(_rCFudR4!LwweWo)?;dZlp1T$r zHUr`U*G;_zE={N|ziGBHowIF&Dc=`*-C>jTdhqqS&1*dQt;y`&gATXap%aspBx_TR zkAs#7)y!8})8-_&gnnJBEjUN(qcRgnBpzG565)+#ehg8YvM6XLkEaCTZffZTccUxH_sMd&&-G!$)WnUJD!JXECZp`J+wY_HL2C8+y(_KC zEc<-#=IcsJ*lEKO;Y>0{cQrMY#$Kc*i8fBH-5>B|Yg_rR7e!5H_v?){XmobM#97?r z-MmWUs$^y>#>U@OkA6*4u~5EYrb{K|=@pSv{U$(3?52y?BvqY<*U$COGD|vZHM(m# zVbGV9*$0dFYd3lC>476FI$rN5pVketd*M%xCe3oV*?0Po z%`Y!@Q3ort<*K%((a5knAqObf@yFg)Ca?=OdAie2k?Uh3J(=?p30e{zyhP135=6vl zE=gAG$Vzw5>=F%fWB}g{OAJHR+(Z2cnwyRv3ZuSUr%hzvyX*dD>2ZG$e>)csly?W} zct?~pNeXL$XNZJ*|BA#g8d-V{=wSea#dFAv``1)vL<+lK!SGUks*y|3Ftc@NLK(#c zf9Jn_;rsb>-}<@IsXYl$XMm}Ye_*SF0It+fD2=#g=?6hCN1k5d=;tkU&o{!AD`fTU z5W&jf%i9)E2WDK5CdF`FuHMbTTi33vqh(S*4Yd>u&zrrG(Ak624-ow_)zqM~Xbl^@ z@-9NS;-y}7jbQJLWrsseh?UhfL7gv+lu@qdGy+%AB>jH=uz_+_#^rJ=lhJ~wLEoX+@& zV;X)a4J>|hAl76s@}8>ysY7Lfk{m&l>5Fzb)OcVG+x!Hhb?jB@>%*?YwsNE#t9Kb4 zGak7+r^DW4WQS~VRQVm&dn>ZG3NT7|Q>Gww{$w4*EA3APFp^ES2JKt^=H|NCFBpCL zeIc*uW9te!RJQa4Dj|ZFpMAL(4!JZp?)8WJM#zy8(#n0R90h%Pj!{*Fn&m0GFR;3*Qe44Xbng<*zP*v!<6 z>&!*nFIRWvT2JI<`Sz`>aAdNbPmi@;?0Y96TbwoIa%p~XX3FOc^x7 zw!l~L((~e-%nEy(2pV=f&cJ=M$T>ldp`?WSTP8kE8cc8p)u_8oCpvtHBsG5*uKGM$#!jJ2}*D?`yWmG{#vtkrvlagjc_ zQY(5Pj8dNyd^N1}vahi+_D&GZt+fA;Mcs0YxMxhH8BZrxb_O4Lm6FrWD>=X+^4tF{ zs`XdqS{M!I!+A)9$@cW|2k=WgT+JmX_|rwp{v{0OlDAk^td?(8WkNu{SPVs@=Vxcr zY6>2InqZnd#Wil|Gx%2TC5)xD25;wE?fKs8T5+Yx2|BYAA|;ZJl#Yb1sDLs>dgnqw^F@;IcDOj@ z`4M^%_Vjcs!xMBrhf5LTT`NlD?3Q<#b$Hs-?i1eVR#S4@DSTDxZk~NKOSu%R&ici_ zd|1?wCb4T1Lt<3MCtBJiYGkeOA#!=60^Tv+=1SJ-Umn3-tBS}AO!faX*M)quw@5k=MSI)Az`_a7^GpD>0o>Z#h*n*XS!-in{;_JsF+bx`PaXNrlS7v=pL%uFlB6K3@|xm&34 zZdh-=5e1Z2ht(aLmhh#^yR^STrrP+C24N_#qIvwDiiy3Dz%0PG{Xn zNhx|q+@8oIdNSU(nLtX}eRH3ajU?1qq(yBwR-;Q1;7?zDZB|kvp?z%_`e8 zRn;yz#4);1dBNlLp|a7mPdF|Z2+NHPi)my~HLwac3 zhht_;co0-u0 z&UFDy!{6R&HD8(Uwav@!JX|6J-FHrLTRTtK?aeMbN*?0Ii|^tNYEOWXF~ZH7Q= zNiV!Tg@Ri#Z|D5expslA>)~6Be1qyCz>7_kS{dun;n=0W3Z*12h-4cY@+fCDrF5`O zGHRzgVj-)d$|K*#+iWW~P5Uq-X%4B^v&Tz1c`6*tegsE7#CNe;MJ|OD+MY%Xe42Ya znzf}inmJlsZV~6@IJ(y45KWVzUmPx?KUk1l;F$e!AgfZQnfm!M6*r1HY@XB2_{ZB` zGsMP|jxI%Kp0}bysX^m5{et%x;L)oDZ6-+Ib}x5yr?FpQtf|_pba}*@CZCG9*)a{i z2_E(`$C^-#r|fM)y#5&1>yC{S9QRgJ@|r)&Ov`R<)$OwqVUe(CR)P2cK-;D8>Wwd% ze4ihFWE+jW8o;`A)q}Z|_7C%ly}D`oOm)lpYdhf>*qO9euW361*Ge#|HusW!RAsd} z{%3(W&?o*eC-u`-_AYSiLZ#<4PfDQknnR;5zO+JQTyBoE=8!#_5*ua&PsVX&O4uo9 zV0=U(^yb5Y5*3h3ynnM|8%m(ntBeflrqdNRW3A<4@G zetnD*cYBQaUMu?{gs@cnO4Gm7VS)ni{3qO`c9J@b9G`mS`+U)>m`wRJ-akIRww9+2 zUk-dPLp5%6BimZ9d}w^?dAN5UT~U}%@3@t=tJhEIm^&`lE^*%=L4HDHO?8Mwv<1oD z6^a(tv}Q`==p~Po#YV=#p5%JO-sMf9y?)hW3PFP=a2RjoMF2{k6Q)R zwiQD$eVnG=v&{V2EW?koph*Kk`Ps29LHiADZW>Q=-YbP>^2R7CFQjNmra2-f-5X*5 z;Z;>rJ(wz0ucfW#HhKvNWB5$XLm)3-k{)$`R8;%rXU2hg8}Gt(E}NxXxQ;yEj@r@I z=CW7W-mHv8D(PhEoo1QR8v+~7A8!IF0Mi}yy%ZVSmcm2_!VQqU{+10d z?*>o2(t4O}%D`vGrS}*PDIdB4d5`Il=p~Mn#?K+gA*DqPZz|r@K)Ej&k#$8XrQt~} zeyi(>Tz(LLfV#J-^I{Bn&=Fnu*)(_faz!6*uajTFS~H3E(|4h#rCrrp9~~tR8p_9K z{YJA!y_bLb=6HPLaB$o_v+kk{*Q@!(oMif}+TDs9C3KOFoka*Orx9fH$)=~sex@rp zM;Kj)?iUuyVqMN&5``X5g>=PAH?DxTArmsg-AkFR4yR#?!zjs9sW6k(yA!zdr7|MF z#aIo|@^m9&VsYP~M$Bnx!PlvY*?QLmR_Kt(z^LbB6T}$}rxth(39h71B^;xG>i<#j zQaLBJpOw%Pg(3OY^a~_cR~1Ylv-`P$UX*8U8q$)8-<++n!3=S)iVZ4wDhz9nzRWqv z*e1|EFImT*?VrH?b!^gEB(YH=@YKsOQd;5V8^YgPp82~l@RDs=2%u!%m7Ef{2xzo5%NcB$t zdPrfvSVRsB=e4b=8k~P?TM0cj{xQpPRrfM63+7|>N8paeQir1($Mt4rzhyYfH=3SC z!9P5q^-5l5G03_ks+)A51s7GgzUIGNEz`}PT4nrf`CFM+O+sE#zS5mb3gT^!7GK9q zrMH70GaKGf#<~XSOrS0|d$(epkke3;L9}7`{fiI>mpuddIg#w;3tOC{*HSpScVI!0 zn-TwYemZOAY}!QlD`aR>1P)vm+PN?aa@eDHsQbH-{>DJoBxko^O*2B)gm&Z;<4P2- z$=49aeTK=be^mKy4=Lc;_HZ2yf%OW2Om{P)DX1YNP4#`sd_i&Vi#2_Up84)qv^u! z$!nTH{Q$aS^-$QQL*&WwT(y>a?6UYEZJKR)q82}qG96)Te(lq{WlJgdj=W2B67*y! zQ8hZbYiMXQ6pMg%k)MzYQ3#P!lb?huy=L1yr0&$O>`2Qa|BHFTm2M6@$pbyF z%!48J<#$O&Wo;}!4eVkJDk{k2_DWZZsvl=X%7ioDUVALY zs{Dhe=j}IF!rlbtT86`J_2esdKL@OIn;!}?<~0eG+i0fi`YzqH(#9K-0O?{4(Iat` zv0Jp6@Fwc%?(O%P3z$!~ELRc1_-F((^$J)nWH$f!#zb?SX~6uw6%LDt*6jQdGA+z5 zPbbG|LehdJTT>NdBh9vWKa61Em=VuV;ktl(B1Zm z>CcgA)a)?X*1Z0P!pC9itx zyNvS#&sdK+(7KfQ$dYKgf{pKmA03BYQ{AEDgjwZ7H1P0-d9lWu>YT!zwhpadg@Lf& z;oFI>(lRsjWEIY{U0D?V{cbY5;OI;3kL?u!UvZK5jLg!iW9&BN)rsFdYRG&^Z#nt$ zdR1s`D|SA^g_A;pn7^f?$_>Z~rg8n_%jrerQ(A3;QX;L6Av>^k(J}TyXjb18^CUTE zXz-GA0Qa4ZROfJtFg$T25%yKw+h#nmIu%Y0cOBZh->)5_^rvP|nOm1@J#hqY?`gL! zx5nr%mN=6_;dR^c1v1bIS@%U%RXU0V%Y?hrJk=_23pt2Da_!y4eV;LY<^inz4paoL zWiA_2ddnA_>)cr-vB=TC!+VK{AYyV~ggpp&l;_?}9=S-)W(q6}F0Mt7P1Gg<`XWN* z@0=_cMgelg| zH10}J=Bcp8kj0T{8E67lK_7X%Y7K`No8zr;C9(+%*QSl|@>u>j7>y>!nUXjqT7?5G z19A0dl)uSd9@i3wh!cpaUS()6sxH#1xVy_qc`;g=<4a5MyEtni@U9CPiRG!uLJ%Gz zqu5dPapPmQq?xT@+S68*hCPi-U&_eXIAbqjnrzUsOfG`v&2={PxCW{rlk~e~&L);Q z+)h7ia;>cM#~*<<^075Xz&04ngvhLa>DQ8;CQ&e&aQBKic=^cq)v*Zbd-dV(4}b0d zUSL&(#b#Z6FkZZM3Hm9HA}ZmWJ8XOB6N^dwT&cDRfNH)W)5S{S z&H3rRHx4*;V&c@a3iG5G>*?D;y=T3mYl{sNa4wDPTJ>ct98Y^PDmV6_TE=N^EnPiR zr%n@E_b|bI5NyADvz+)DXmzU8iV6-OS@+u@IQuLNM}h<)bNZC^61p(QN7^c)yPp`$s03 zpHH@pGye-RKfoqSo9VTg;DTUV7wvTM zNNS!a2QHuyXFXJOSHuP9Ts9WHj)L;7Zfp+E--8w#&c@g$1*tu|Rg$eY6VsJ{#i{&u zj~c6o1g8Yl!ZeagpjlT=Y3uQ$@sNvYOc$A-MU~U-u%BKF<{DLrQ0h=>jfCaF;~Yjh z3RL%$?I#_oBn*+wqSiI$l6!>swO29b1rDW=SCK0gk)CUZNu#emO-?$xcx{n& zHJa}!-R)$EdBS;|K3tX@(vXOgzN0P}`ZlbmDCF>db~Z=ylOgQ zJPT73N7#CY*ac(6@<-W>M4D5I@>Zm~Eu5xh=U2JzJ(?{&fBuXv$r%E1xzWh!R^V~% zADR4Q?*jA4Zth0shbyNduA{v>8bDlBbF(xKuE= ztBJkA(<;G)C-JLl$sY{Eux8i}_%eF!RV8Jr!*HR6hR(UI?}{oKtBMd@EZjtK&@5dp zQBI<@GFxua&T!~D6*Q=_o0}aGwv4=P!({Ue5&xqBMU8^DKtcDs8?x*w{)=1Gf?lI; z&{Uq?rSBqLX+{|_Zq|Z+@{Q8;%iciTyb6Vqdy6dAEWS4e#X1`p{#}-R>(eFPzoj&5 z*QFED`q{6Kv%t1LZx_I&4X&lU)f zTQCyVJ-GdV6n>ybS?Uf=y3<&{P=C9BbAvv;96Z=q<@(YvbUpKr;MHh?%nKV~L@-zK zB{F~W#|>@2vL#m?%Exz64{wxn7$qB|k}2qCS8D4IyigUg=&f)Hamr>>CQEV_?J?GH#mtB=wku#3x|=8K-n2;~HU?t( z>*R1%Z08`sTHD#T34-Uyr6sVz4&YawGhhALtTtOCBUVqX=Ema^ol=*=J*vYS=nP%n0hkm>b2Fq>_Sn|Hz7p~p^^j5Q)=|kIUaT#QCZ~YSYEkHAgF>a*mCy+N*~6D#sW z&Mljfp6iHw7^374f04sVUUhY!&|(UTv0!q*U%a6Jjd`^*E?H{#dg>Yad4kRgRTYii zHS?T$kh=GzC`CbqTza!iFuC>E;Q8R)V6qV55GOh@h5xVuZGMXikj1gLFZJju974!0 zeORC|()z9oaFgFIceGSb+UZHhYL4BpEVGsCDL6>N?2Lu?rrr8ZJ>H`4U_TJ9Ath5y z{ppF{FHiaI)$LzT8dlWH+U(fAew3Zw(_eWnS;rh^Tc!ZqTo;Ad$J2JDB`jjeItL9c zp}>zAHS^-0*2hQ2K3{iKS-cZD<}07$>{yAeheXhFI*rD!%adw2<1Zq5Gv5KJ3(Cup z*GPeu=ixGMG+@Oas@-JEzHoVX9mG+$FtwL=Pv z>HVZf9b$U#A#|#gg(N7SA-4QVRuzoYq$t9oHT-^q$y0Woi}Q0=8E=roNiQ=)xry#^ zwa97sV;fKhyd#`CXG|BAkR&`Lm#@S-xUBg)8#ut4EtC)PLndMuVNIvJ-skf>(5hYr zZv7%T5T^+|1j(Zz@D0}Q=m{PSxOK9*wtFu~}>fir3ce6-CNrX<3{a}$~z@fV{(=osq zB)>M0H^nwBqJPiW%-*~SewDZwfWJ>YYA2dUC29XuKunp*&9T#(G=U@^KLpmE(|uofgqkYNZ3gbq~xlVwaV6IX>%g9UALSI9Vdx{OD57kFQ?%^>U9r;a%hL2#-r^kpd-%}pc{^HJN zsC-n0mAK%4kCF%8f_=`x>nf%;xvEq+f ztN!sSlQ55kI0%`1)|sP#eAJuKieM|u0oC||2G!BmU0|J`%ilQL)T=v|C6ln0LVOq; zDKN8sRCP6*g8y!aRs$ElUzTZ56!H0G{N>E>AgwZY+eQYV_Kyc!ZVm$lf~mh~U zxHjRiKe(lL_f#Y!?-RI8CHdZSDY5iC>+uv=R9qc-mEc2_$Q7a-sk~1mlrP=UWtusG z$?skj?&rjzu zPu88jt`OAxOn^LR(->#~rn|zIi8|`Q%6R!ZJP9X#*({qNAe6CMEcEFK62-(O<5@6S zw|!!bT$AZ25Xv$|3U=&S)#=`2AarHN$&cjUYjM$IxM!h$zU+MSxlST*_Jf1?&f#q) zwFTE8T(>5EAUUCdqJp9v_Zz4Qyfv^3h_yHlc9OAecit6pBH>Z)uujlENL=c8ng19| zp)p&3deXG`goo6eq?rr*U#Hquxnoxx@|q^x9MPii?{#k{*3~t1>yijO1Qgsfxrw?@qubNlnGMw%mIR*u$FM<1E|BuqEa?XFfR$tXuCOEucGh+iESWa1`9{qVoj z&Y*K+5yM0Td^6>L^e{ETK(KU$Q=3y9OZ@&MYEMwbDt~@wP|>=seE91%<{POURRHcyg4+D;UfTehx`2P-Svvi*fzV`^8 zgE_!V7=KxdX%>2HNMEk^0>}5c@zY%&u})Q(q3l3F>?osrG&YzCrmnh@+)W!M z5L#DnB%)?zP#mLo;@JSQSk6tVK@aKL6oVK-jZyE-L^a_)v-iY#Hhxu+8Ff0^%D;3u3g|+zSq^x{LB~G|wx+u` zPWUW5n=Akl4p%tOMnf-ureG0md5EU~fG9nurA&-hgy_Y71b zF02mpNxT&T%Cq5G7)?4&dr_Kwm$Wwt$ccx%RU7=Np?mGb<=xLN@=!bv{w%AB{a1Vn zVlq`C4!7xXf^v&*LRNyvIn=(x=3;=)^@kaph(S=#@w-97Lc0TDm@`Gwd&r^10Q6=2 zj9M2}CJVToUa5{P`#Fv4In~U4bNuXhMUaNS)+7X$w9?dmd;T7!2pfAAIiUW!w?#jN z8%X|ibPF8P?hK8@nI3W)Of6W7yPFMj)wT*E*7f zyUM?Ofpm0hd*9>g)A*r8&MHqT|NSgEWU&rmeI% z)x6$F9pquUj~j>>Uhk4p*;aIKdcnnRy(1T`knNRHhF_TXqMv(Gso6Do7EJ`Fu>CDwf$4<;1ENGcK7>2NMgNQ_}d;4$X8qZ~)HsFq?4HEkk%;K=_ei zX9NnS@!8RjSnnRew@@Wz6g>{2Bn1Ra!`HPE#3Fsbr6w-D8*tk1o)ORQH!+cXlLs5N zKMs1rDYG~o%U=yM7=$)X{g?F4Wnh8DIJ)@g;6RBM#S5Ss1TX2Ac}YcQxs6NeIco0c z0akfgh~m7|UpGH6+kVrO8n>(%bB)g6r(Og4j4cA0_SWZL1v6AMRe2LVUsvfC@)jB3 z3pIRt7^zdsENP~!k}CZ%GKTqHH#%0MAPTa7vZoYpk@mcgB-_3;08M3aBPWN2=)hki zAcvNp6f!T1^T|0}osh$Rrl|O$dYJs<emMq{CzBvMwnt6BHd;Ts+kgm|(}W#hu;R zaZ@+*U%{?xG4#3@tK%}?`D!iwb2a1UZ?`7F8<6ups^~eqqfPPyXhOTvF=?@I`|>b9 znWlm_Lk>(s!dRX52dtAv9*A{w*W* z0%~?BDYsz_Tz zA?7yvHkw~V+a$yYLoAYpW{xAX!+EJiWsVg>yDYFntsnTFH!tL}ek-c*ecBjxCU<;d zq)3Q{Q~0#S$A1f*H2K$r|G6t%IeXb}HbM1x8|p<|13>hnHhpK2ED`5#K6g7Q36Mc6 zOGtph))zLSMt!_mfiVdP^G;sMW3BG%3#WB9Y5O!BpNEDlm z%L#+V7{3)*c};D65~l8?v9|d~$G>_7-1nq)Pozd-#$$wGq}6a9^nD6m)z~sX3U3QE zfAQj6nElO#T4@>@oW4MCy7uZ5ht0#{G`)wer&*kxd9pKR zL4c-XzB@6xu{LXr(3TPu3okpX{0` z+KPBld_x8ChZ$dePm`r=&ftG`0@ZN|;O1)Y$;4zlPr4R!8eb-1n0o3-iwZBf^h-XEwH6_@Di!IAWB_O`acy%~B*))q4@VeUfo zS)`j2IqI1HysHkoD~1(?zH1Vr`4Il+rnt*(z2^E1U3H#F#`GIgq5&HTba_v< zHpfNvxZ6!4Lb^tPI-XN;@S^1lZ4T7963%?Jgal51xZUWK*6BNag^o*yGPQEGbhRZF zFR{V~i3A@CS1gxPM!#xrJ+ErI$qV&~6-Af?>U*?=Us) z*tJ%%@8h0gqi))6QXCkWb+%4_6wR1t606BcGTO#&hYPm3W1~3g$XJ$FSbn2a*l5#ev~H}ll(O)x+V}e3L?e9Y z|3p(?So0}D@<93om6+`HuZcxng2NBM&Kb9We0D>VcWtrA-vUR2b*C?1-+V9AD>!sn z>^Al7a`2K15aSP^eb5?cFP!AEkz@){~zKZacq)K#DFbE|XRC|)jA9A+H^XwvG;2k?}`NTNLl*^S;7?Wa_f)UjgolRKP z-=;~C@1hVC^Rp!VCTu%3MI~qc{1{YcWDXyPG#m%Ft{Yx>3|fxs@m7G)vQvsuYAykm zOz`rdZf{l8oph~LEek6I0Fe9953Vc+4fnM41>m16`m%Xv)H)-;-@RUQj9OlY#TgJ5 zMV<*~smv`-BSo=qq{iJ7Yrtf%C{ALs1;=awE+kLSSH`%XTp2WjDgQ9SZmQ~=@Em@c z_*&4l#$Cz@BinkEB9yk?E=#y_;56K6`U_l+;7pRs)=8sd1vX&@140!J?}vYJ8Zp}g z+{96a&TFUf#<_2_cb_LL57=7`yh+rXSu5HWw;%d?vObc1SlCs{nxZZCxAqX!RZGVQpxupN+;X#vB04Br=U`yHs#svy^l3rRe zG-D%_;v5XHAD9lHj}W^FJopz4sm{hWsSc$>lH=~Rj@6Fo)GYT`sJ)qftNrX)bWFO| zZmMp&Z(uOn5Ea}#@g>wT@lffY!2~K`hXbh6C77<11FJBoIW>IY z+^@U*huhB7(?U$vT3|%J?@f~`H=E7CDhc7COgMnv3#ZwYPZArU6;RrzGXKvb#9rX&t2`lYMe3*{P>d28 z^$7c#cPAiC9Bv$W0={IzwoG~$(`yV zi?d6d?;6=gzJ<&~Y#)kB))8_KEdcooBpoJ0|gt%(oCwRU1p;-!Hi!ExMSa{4RmJD9Ev z=cp&HJUu(gI9bwu>biG)G++bMq|kZL!;xS1&0f6wn-5m_Cd@o9Md;w0!@HYI#6Un> z*F+fnA?TeFRdn(Y8m%y2akCmQTPy=uk4LE0^Gy5?PI>zR`|-{X*v1^;~j&wChfnJhUE;A$tvp4;&~1K!)& z)I&L5k%mD%Xgb|-JZRZw@rXCr5-f8uGKeA)0p;bgy>DJ+3JG93!5;6;C36GyDK8ST zzmz{Mlbe6f;`l7D zt*$P~anLNPci(-XG%sbVYYDa+@!k8(_oyg{OVmwvKQ`bfT$$3m*4jf{=I>&d;9*Sa)zgS{p zwkEnrur$5Q*T{9qh-7L}C=)8XPWCkG5d!~7J#jW16+^0WZC*P|DJj49vCim~zt18| zSg|}73<0{d6@Z{D&q~gWMD3WrTD~vRAU*^9OB6cxkqHM(>bU7Oz>9n)ZH`i_yzBei zx%Lu&_1Ea8(ifX-h|Y4|yS8xJCJr(sv;y6@NS+!uH9YGYrs1PS)1D$d`>no}55>x0 zw^JaBUMNypm*pdU6X!3+gmwKgxc#sw@MRNUo~8`}-6!_#2*Db9f3#d^To60dP7v=5 zD(Dcu9yuI1)6(z^pLlO&H`Dv4SK+(n^^Uz6KUYKPaBSqicj`U^z(|B}zB8C|Dgtp; zgAA66A{4#h3ex4=dCSNhC$lEK{PRO{^)+W(Fodlzdh+$kE?7J}j6z4=VyXPz29DA1 zeldleANPv1B?jW&1OS7_3aF*k>>vp-5C+o)*L)6oMm&hu8Au~@py@kBsi)ry1SSP# z`U^NJonY#v<9<%Vw=Eh)#U?0Q8V)A$b6pUOiZ1g)8O8OmBrPEN8PS&4r*syM0`^%;Gs4;feg_QrW+Fq=!=vM^|KMWg4zq{q?@gPvJA)~*aBo`F+OW2LRA7+qnPwB@ zzwQpbi>NUB)wDlYT>o7>TIAo66t(}+RB*t6 zDJ1(`8lu-f;Z!4uc_p1AX=vDB8qpFuK`-X=M&v>3#dMM!)1LJbGce7m>s`;Hd01== zr^xbgB*Y9~@MA|uU=@fW-Gzj!b>xKEq-XpRNbSS`G4tm{MK|R9SAa* zc*@+gf2oTOubUh=~+IQhWHmGWx~v z&0@UV7H&9%Hf7u`Zup*d7(G8W8?`?Qg_D+2clxhTW%2Xyx753PJGVdxI@!S+>J8Fh zj^8};EwM?!HjBDt8wmS3&5A21zNY4g>m_{94)`JBIZN-0vPz1)Q72YNFvVLu*?|HJ zsrrYGMSBuao`olcT*#@zyy+=#dd6$Vi`&1Ym5u%i2iO#nyo%zw(-9VNn<-w(=dKl! z|1Ba_o5HCH=Ozd(g%kHtyAOp0fhULH<&YT#Hw))D`NOw>=%x?xYX@U<5pdC6@9lnBmF> zwxP1(VRSXgmPxh~MlU9j((B`JRHf7SM@&av=&I#_n0-C6rYog#C&lUa>G~8)-MrI8 zRYa>ww0c*45KWjOhbhp0~)kP9;xykg0kQARIeED)NQ2au2uRQqg-;Y9V0WNzV0jGce zVl}?sb27k7p-1n&T(98c#Cy0?Z;oIj=98&0 zRn|4w8GMF2)|?BZLb>pwjYu@n?8_Y(?=Sm}cBq91wM zXBW@I1cUUxnEPdKd;6adr1J@`ym#(f=8(77^`EprSZlj(UgvP87))Ugw z9Kjv0CJkI@cc(&VeO374Yb^ob_angF^q*VChrl+6>BGA>R{;O#BQW@nYJrw!Ia&-w z!Tp{H7YyVr&8I&dew~GZO)p}d=ApIxsVN?}S_)+Bm>}A0%SR;wZ1xEP zEB_+)u=LnSm-ZK?7c|H((j{`f6N6U4$u575JIlTm-&=^o#)6%#A>5&tsi8{tJ#RKR zmUElDcn`To(f8-BBzUNx$Nwi`-^id6`O?Gj^!SNF%&XW%ABiw$!~DerRgOQ`gT2O{ zs1B-r`&3vlF%q~}Jvj`H-ol|)lTn7S?9%#RamM*hPx+V)>ms+LjEwaK0$T?m9UMAZ zri)dgIjrdi<-;$myBKh;&Wx|DDu;JQ@Y>vCb2fzCqslXb)HoKksA-Hr_~017nI4&K z(f+W*v=`!_NQ7^%*+y1H5LZ0gu>A$ST4GMa6KsEeRM(%65b z^%wNXHd{1Ut_EFFbH^6ty$?hpRm+nW!ST4;G7r}I*Ef6g+1vCnho=c6NA9SKtg)YHOm*Re{N2Fk3r zZ`;7tW=?fuW;fs*&f|2vFgpTUTC6@2Kn`_J#?QvDAe)t;UDIoo|%%!+mxH1hg{j$r|MyMq}Z z*+s>XfHpb9mzc|84XJp(Ec?XksK;YIsq#CWF9}AC2XfgKe)l^qH|o>X=~F}2u#&Rp z;;^mpn`T5XKS`-&Z>-*EO&3cdO;+Jeuut4m(!Q^wV6Gg)9%wOeAaE{sT3M!EVW=#X zWrEVZT9fMxUcNT9Z{(V)^r3NEl`Ph4(uiFs-bYZSp`_Kmvp$H7LCodL&O;`_Xvr4W zInoU6d7AA{I@yh;hVsb`RZEA=DU2bu2+z#a%vYIlFbKEA^EYKhRm3p*DS{OJa}+4Q;Jmfga`KhYq)ZI>w9lTNPf;PfPwGqF z8qUL?_kbi(7Ai^D$SxZlQwju+fpa~i`kZW6+96ChzcQchw<`Enh-4ojAm8zbCO}Ls zO4Peg1r=0{qsh)>7OJV`x&sv2bveQ?p~)|YtVKZWjzu3eZ-F^I$|YCRN}l&kwikoh z4H+4VM~FLIy~Hr1er-mc<6e%v-&b^vp;zPSSx|Gg3(Q}L^I9SxJi%sAhhb4{Cibxo z^j}}D{6lLE;3|(m$y~_?Og-~S)nZ{Br@}HXF_R4c)MBA^hBHkTPu1COF-wxkTs59tplHi~0lUD=kg72vam@w>9oG04F&;)*rc>Zb?Jczf?5Tab+ilI7rh!bue zZVetS*tm#<6`Y8ltd5B}-Um)br>G2bTi^yCinXldF{*e6X5wPDxN{gfJet# zKQ;&yaL*(BCxxDxHT&{jEsUm=+g1u#Fp@x2^RDDJ{EFm?`M_sYtm2T}GXoK;aR!5k zIZ4GWrPABiSOEeuTI3-uR&y}-4v%n?WKXO7>5#+z#!qJH0rG9hZDg4w*|aDZBQeh~ zvaP4#^y?${&-Z)km|jHlhd+rQd>b8EMeTci_%Zyl$Kk zzq6Aj(KX2yvd{fAgiI&Ze_zbE-lw3F7PqG;qwkd|9jj$+0*~siv-j`RC9aq`qXWd`C@5P_VB}Q1v)J3@s)GrAYYt_Fu|(cbyD{ai z^dx{=pDXxk*!DI29()^i6)I*DL4z@>Cew-BzJ_z)zdRKyld@W6uNI{ALNO1P0jHL9 zGAyl(+p|0bNr>|gM9${7YZ*g`(3&Sbv}bi^M|F4TT-*5I-W#UXyCe3ipoj51$)Fj? zy~`-GQUX?V?SYGa!DT$sbH%3apv!p$xT!wW07to`+q_09+$^~J@w3VR~o z0@5I$bR0q@1?g_2MY{VSARsCwl2Rhwjid^a(hW*?AL`$q_uhZJH{N?=IAq+bcs};G z_u6aCHRoKq-><;HDS?g7dh3*`sea$dMf@9FQHRYI^Q}oW%_h6&^C59TAwl&UxXEGz zK2tDsiK2%1HDwN1b+9Idj`wj0R@xT=-bWIK+!FoQBq z@SQuY7rqt0zRSWTbCVMZzVq)VY2L1MDgs`j$|}uP(R1ZczEb9soM6=1$FLxsa*9 z`{_c2kA}(l@zz#G2*ExYmN9VeddloKRcUK}zSGiT==SsJ@nRIugETh zD6gl;dbBh__rTwZs|Gh7h1#9qcdMXDrHz%RhHisx zh#tnqk;8K53j;f1(1x;7nWQZUO`H4)39&<**J_Ozb-P9WNc-Oa1~=pl+b=UT?S}qU zMkcfkxFv}`(AH$($9*ZtkjCpc1C3n&lck))UWnTaQJdb{1VwO_{lpNJu;Zr4^yeb& z5~FJ16|I-2l6G;^Aa)u?3a`JZp4%}AvsK~6Vj>&TiYG`??}`qzO7U7_kYEbrmwXxh z3uQ`9>Gl?HFmacphsbFqDV%6?MilWl4R|s*(WA2uVOA7{-eDYlHR)O=WOk`0PPCS< z0tOJ_;F5D65@CSPT0ftq!1&H#$umUGZL2ww$U+@?RoJ?WSnsCG-1B$WN++d^Efjv;HRsYd@Xls%Taw2v<-C?dUc@U%v zI^iu<;Ku!;%6xd@E^edJ3s^ZwHq6I~+(!pV+}zyNnD|tl2Mh-l9M1Wy>2Dt)ud}S$ z`ny-nl{~Q%`Meo0;)%@Z3gi`SgOd<9zABpc2Ag+d3NWCuZhX|j$R(nr{M_5PzfV+3 zTiq=&RzffTVA#=&3V?Q~#McKB>>Qw(e4kPA=6|gBqb#9n{3L*t3fzZfo4`IemtY8Xu7cY*(90=8AL$AZNBec?=-=c6AYF%>+@pfKsgjcsGoH@LaqpS?;ktUFV;( z16Wul3N$XEZp@YSfL6j^DzT_}hOM{ic(8)y7b9I|8H0Xmm(h?p^|>#w$C@aTmtZ}n zo-S-PxG{EA9!RzfXhqP|xU9BJsIpBQ9iM}RXYCM{Dsih3H)^o##XuJS-Ba-%C^43l z%t*G+lj>-LG&z9U7hzw4#{@+iy1Shnk;HzXHx~q*;1NZ&%vY-h#T#l^LmOd)BOxPM)iOCXq=m#XV=(nw1xHyz29gj4m41tY z7l(ks*oZPC_Nrh*?Gj{bl^QA!I;+RcaK{pDEr6lqC&L$ma>pRjPMw6>dyJGA&*8d2 z*_b$THQWCNqZtviH`Hen1%;$Tg0j%w_fhzZ%Y2&qU3>Sx5AeW`P}=FJXyn@{+DgO} zXB_K=$fav!zsv4qXKckgQn)g13|N>AO%{uhv@A@DRd)vC4gN@H>AY7Nrvk1o zn@n9Q`h@h&X`hA^Mx@822VwDE2<3cZ!jaR`$~|xL1~b7OxOjL>amh%grt5A>zV!V1 znt_CgpW85HHy(ose`ej^r(jpY%(VYQXZ5dloj~3CTMEyYGm0?s%((-*BGKs4xabZ# zAIXx*xYk~wL#~tPWdH*3L$O5HSStKZ1hvOPY?6~H<)xy^iTZ8n#9o2XCcpdBcYxs~ zWteMYeq(q|6cvU8hP<27I?0pCZZ_cz_oi)W{k9zmJ0Q+8o0mWjLA52Xn8uKDkedmu zHj8X9o&R74nx3PXvRDD>-{#4FM%LeS3%)}0dqW>^F{og$gp}?Us%Fk|B-#;t_>|A8 zmjbF6Tl$y%1z1`0H=TXcykOjRcm|1`MLS)2)36}c~OxWgs;Gtam4ve!KwIhiR ztshSPkCmDt?jU}IH+LUIuGWn~(`4Hg@~-Ix19P}{nmv+U8K-N12r)HlU+y|Y(oj_=>phu^A|bs!y|SAjv>*{R zhNsQLd-Dz_YhKS=U;S$PpuW_TimB_+Pa2=6eUzu2l7ET3P8M>tD!#mMt=7M@u>Pb> z_(*;KV1hecZL`^_OuwebO{m?DNqw_sCQ~-(XjOf_TGs-fn(X`y%O=k;_@WYaJ~fDgKD*M>i*;^wA{op|wZAiiVb2-7NJw)~1s{NB3 zwG9^XOqYA&TX?aCc(`w(OMe3)dmDFnaQP|ue8Y45#il!pHjQH=N9OGdr%H(w#` z(z8;6Op`F!L1wcrZ`J08l@`rO+a8fBewH&q?Zjej&yXgPoaSjpp}E1$CDE=RQao`a zFU>z~PQ`+gX&3xiId%8tsJy9^BMr6zZ9Jv4CKh0+|GWt-y4<9otq1y6nyk;FbP2Ea z6I;fOuaVl4em#s4 zm+%9#fUZVsoQ&sg%3{ifIC ze{ogj+`G!$acP}AvQsWJuI0PkCA8>nW~y&C6S?wCcM3=S{Gmh3Qyw^|S=er_hpmSy z$~lB7=H-q>ddzxtZ#SPiO8S9fbP_aG*6@Rv<)e2?D3~4xU-7{nuGVNIjkdr3ng5y1 zc+myklF0y{19Lo$qUEt!zGd_VXEny_I?tFAZWie>;sZ);YSXUF^DfMxnT{?^rV z{-#vz#v%sAzx!cUoh8j*3~nif&oo46wkKgm^6Sj}&o`W)OE{niCja`DY(R7|ux9aU zCuSj__GaA#H|`0g4#lay5r`Tgj5G9-lc3$adF?cq6&Aqj0$kycBk<*$3g~x+`T+)H z1+^Yx<$L5}6A};sXqb4r-``;P3yrP>VPy}j?LH4t=w@=GRwAYhzHC>j1--UY5~UaL zXPtOYw|TwQ-oJv!6elYs)v91QZHR)yVIivYmEF z(*4^kLT;d3+eZSBRAfD@>A%s-sEbDMt&R2HEBF9T2fUFN?hxD`h0l^4`v6 z@AgZlNfoPWZg5$&VmAexkn>A}id{N&40g~*!?;yOcwiFr^Ha@Gi|S#UjnHyK8Fy(c z810fce=Ssr(F$>^`0h5ec!%=}yZ1^3=5!Yu;pQIdZmm;P_#NP|ZUG};12 z!kH?BlrO=f(|Tvgu6Nt&{>Ph#Ollg!g!=j9f_!>O zZ1iS6fwnKJgMA`BFA3eKkr9=ojiVhteV`#X&fRkHTvFNZ725BiHi{nEx{<7&-WC=R zNK^)loRw@jYFD7hFh7pELLP?Y2Bot7G}4-F0E7en5$fALoVD8-rkN2$6FceG_k0cq zGdvvF;KX7FN)>?Bkrotkqrc#{WTgOJh)oKd;)S7q`ZJ0I29eYT_!zc@JmF6iC?)_o zsXFbzk*+2y6ec1kt;j;b{EA7eTWFK#P?lUh{rvHDi;m>gUTnZP2DyO=8WzE!^vpNh zWDGmrq&-46EEwIUXzF{5`;VY5a=KYkL}eNAxAJQ+Xy*df*BMLr7Ho(jT0nO*V&mpO z92>ybR-bt0hpf`zN;;^hyt8ESU0Dzmc)$HThsS_g+d%s77liajPC(w(t`+!ZteBn!| zFsoUutE9%h;6h$<|8qEnh2N7~SSYXAf}%-OC~Si(%B^aTx*9IOy`b%;0|P}lY8=BG z;#J3U9#U?5^f}Xh%^YAEN9}k`})m zT{to26k+~#kNt5#&<-MLvY}?84J*|FI6a-w3>UtMDAgN7z!L7yc?E{%w<_bnL>_c| zAz0Vx40w32oQEGR1Th6$`lvJ}%X^ZG{_ABDO1D87wPQr?k#d3b?3iGxX}l5+2Bu4=ktUy9&?gB(f0h*pv0 z-uopmaV8T#o;$TIg93vH9WZAhVZ#P2yh(KD-g$pO?~E$erM+K@)TnI5w6|SFf~wGz zOZU&IFX#wLUeeiE5#~fQrLdx%6oM8bWhpCkb=p6-4z$Bi4AcMGbZPl9R&9_PJoQIl zWrwz*+uL-q4OsWCQ^0vDt^er)#6P$_H{)zUR@Bzf6P$$MH4wflKbZ}PB>Z(9&uUR1 zYKa4Yb|w6HVH-cdt~b7$m0fQ$$dt}R**<@DzO$oH=^LD-#7_!sQ0;}9UEK+Bc>n3pnOX6 zoky;{{1ajmYjFh_@HU<0LLKNBYI3|-%b=Z5zPKfians*F!gkZ{)^dF!cIIYwWW2** z&v+-FSi7bTag{Bq+g+DX34m;gttPraT#?aA(j<(VA}*aJWCj5P z$Pa$+5O|EYcMsRTQf^?^%;71yD;I24Q>)Vkt3-x(-+RJEW#J(6zh1e6RU1^Ws)!$H zUUSX^%)J6CdN=aS+D`#r(g`HIEteaBNSf;F(O!U4VKUJ^9$EDT;mmDxVS-o{v)9+} zAnT<`!U@7M}Fw2*IRyudSO7*Qy3I>}{s2f(l?EjHvSoswsZ^ zH(?(LNrI(gwTJfQv*~hMM$!}8zggF&59g`~o-bfk>@Bs6M9Uae+El3%@=^`KSg07D zwi&t>JthaFdl`89^fV!5+*a3I^tsz)s7hMz3@-*_{E^Tk+rsEWohEO_LJMB3+;`>e z>!0;7_MW#~_Ns^t0t}_^=h`xAOK^kZECrNLZkzzMy;UM*4TSRSgF=h_nPPw^=xSiQ!I#@g+gssZS?!-HO8XoNvFIA zeIFcz>~(a~dqOF&BT4StzMcTX?ywOqaf&``R?_}YjLyXH!lMbhcDos{QF$xvZAYse zC>|zUi}xLQ+~&ZjIWtyIid>o835l-u5yc2<)G6E1UVFei z!^G#Gvg4^RbicXCcK*$yRZKC3#E6fR$&-Mh)Z2^co*7B>oa`G(x3xyHkJ;V~p{If) z0B##4os3Bta;JGQAU#UA62|5+v=LOy3ba1hw|J=NDDMr_S5Mwub+NqJN{oe*ee-#N zjH_5}L!T|d3Q)El*#v)-CTRsU*bl&a zbH*cN&3S~ag3lcEvSgWv@XX&xN!?(<;JqUe4MwNJ(K`$jyv?KCFVC?|UFM}K3_b8M z#h%bP)mifX?owfqeEXB0P^e82)fm%A+?#mc1V_eMqlQhL;1sj?twP<1w>n zn~eJjij?O}tgPLD9&A5T?{hR#XqrGVk^d zrd~0!?QojCnTG8Lzd?ljPjR*$w5$`IFw2sJr?T?s`XnSSj z1IK51!eAw;6iDqv^8_m)m)zWBXLiBwoh%JpjAfr;gP1EWzSF+qZUA>ajRfld)kX(1X1|wP zes5HFUhIf0j-4C@76Qq7M3=?&V9wqt20pdW;r28XZWV}WKo`yrxt4bfbyh|SH2mJQ zU0nF4@|lOn&^#D|JRZ&g9Pj(hY_!crN8PHe{)Pnh6HgODmUa4qi0^2<5jhxb4A?pN znS|1D_n2;9T12DE@5=JR;lO;Gl5j9iy}P(@Lw;*^jC!Z&T)Pr(p7lXw=MSqQLIEa37K$iC4=#R*$nkmXb#EU zQj6|eU&NRw{^m`9pBGKO4b&FXM_dBF@HXX0v)kX^1_z_2o05S${u6@?G=2#Th_5*$ zMRNuI^H(+i2P%UE`CYN}PN0|u1S@&K2@lxn3)Rdy-kO>M@R@m{_ddK)!Y?=@B1r*BM7jJ=SOV?1e4pjzQ0XsX%qCJU zMn;~WwRtFR@R9S+lc6mDeAA&!j>4dqVYwXd&fk505EZ+AB^o;d(t2EaGh}w)EMFCg z!_cr>>Bv^hF&b`a$m2e3-|t7!xE z5Z#EO%@vBr$f1R<{Q6|^Jn1H-XLPP2T;_`P0P)XXEMuH@pK@>?Ru(>6gA5OTNSxbC za@`HLA+w)ttoz05T^bJcA7n)qf24`}2+V_A#vYP=1FU77KxCCc6p*YguOZWit?DU9 z|73#YD$-x5UE;ldd~VKOLA7I*Ybn2_BxGk*{&}b13D>#dFt`Aq7T|G%o|^?r8$<2U zDC2FHUbQdf|NHpbdf+TgyZOqUbOpAKA1)3}pSq#5r$`(4*8?c>zc_~hCXZuZ5 z4~6{cxBmUx6KUZN!tu}gJ&Hqh6%)TxQ;&9>MuAm?C};LRON-c_2bgpzv8vc0^%$hK zX#k*If@??kKA?e40{WOO@T!Z15scNkm_uaHv=>7UuWCa8+MvnL&(A3=JeG3L*6g!> z&2d3wB!i;uSaDj3nC4wDiLq?KnIcsXm)VnO{u{?SYnQA%W&%@=Y5PAE)W4o_K`H_+ zx;fZxzZ4Tf+MIDO;BEZ{UF|MH=O%5`JdMaf52<`jfDQ8^TL;~ z>a*XeKG_*;)HDr%ZZ`;tK_FNj;ATRj0Hzh)?_u7`O?51XfE@@HPuyA;If|h7ukB|z z2TwE!!d~OuMTUZO0OEMG7=WBgTKyA4Aw#gi&a4%e9y;*?GA_+ww7Ch$NmJZ5DX#uj z>Hc{@geEpypee7ghuha)r5}IU`+$cOiX-LFI_+^^>W)2`PzDp}@A`l3O%~V!A>)YX zM!-SpoKE*E+Rwmh(;YY@O#*j9VIKf6Ivj6K`0JLp)A{bUK|a}#Jl+A|@Sm6ni~xhX z`efEe|C83|i}EUbQ0jj+Pf+eXp6Um2x$k9Ef>SXwoQQ_E)gExe0B`nR4#q!02k|Ww z5uEP3ADQkCoC$y=&s4}`+rS3^gqz{ij-?RKq~2}AYHPBp0+P-A!4mD;qU!k){k09m*o^JD51Jbg zZoXN+QCtC;_=>o#FDHKf1Q_DaNGYxqC8s|n*2&+BsG?+V_u3h2wNWH#LnGi7y>PBb zbLZa}Y%fyJUPk?A$NlSb%4qT?%|nDGNr=9Cjve9_1g6fPg8;G#2>~yUYg(%Mgr<5^ zgdCnhYy(5!fGttKI9_IcSsp_rSZ?Ss^OL)NmB|N+%b;st1ByZbIkY$ckG^bLujTkr zad}GOPtZb&An+6?cRyF-bhR{{2yF~^+h@nfcrO$csg+l>mp2O8D@r`CxVr2+o|^+8 zZ?E8eE9I0>%YS>=e|AE?&35B~O1crbVu24(d!4RL%7eBRSW0X6e({rWFx}u8$S*2D zzjqv1cbSnJI0sztE;O6IMkJZi#0|EoIIeDW52i9c?=zKz>Wno6g=J^ zF_kBsJ_QLw)w|z)vLXxO$Gl)zpDTX&1bbenO$Zh(iBeDmjhTBdpB3P_0JjG#*7qPY zao;LLK;r8EckAQd_L#V;TL(17X+sOJLK9=LcQLb=Y zu`5r3=+Pw1WQl#nigp6rE@`}A>-VNWc*c0iON0S}D0|G`B7}c`+I$-dCowD9?4P*u3J6KE?RZd)zz0Gav_*j4pTs@MSF|HZB1ts`4s58na7IlBhmniwHlEA^c)?FjkPrCz$%GgIM`~v z(^XMnOF}<0&b9cjE0Rp;5{n3iWHJzEzrv^?Q^p24txM6!M(gX%-)|W+z(Ms|D0J!fdU<3s`z~k%?IUx zzwlq)2_yV=$lj|Ljk~=9{eN=9JdRXIng&JGfV1z$o2_WDO82cv-LYcBEzl>Zp?0io zoAo_VP8G0zWn8V1rvj>-h;b;_0Ij@rC%0 zEdjlho@DC}Ys@EYAgwTk0#SOgjaB~#2cT1^2dvnD3R#%=2vp_Kin>?MUHle41`^+& zMt(<6Qy;kI8oy(Uz@l*uG=zNWhaU4xfa=C!fBEY%B+!Ez>mXuX7P)PV`s=C-fLQfk z4lsDAyEl71fa`9+#UWTQO9(km=XL_uu+vWC!K)cCNJU}>Y;Ex(a9aZ)wxK07 zRarTizmsRG@F@q7WjAfT0O^s1Z0cTO!q zdcV$0X$0c;F_@t_0V4Poc#PwE#ybo<4@+4$R z;STi7Q-cch4&pw?hu>dL#{RLV_%FLqJ;>yTD5zI14l^U@o`LHs{r0V!*VmJ5ckh;) zwuhhn3V^uqfP?7(mGo)ulV2e@(K#RoN}eB2IbD`dOiWaRdA#Ef05$=Y9#B~e(TTyT zSAl=H9da<+m}#gn)*4X;q5r?_Lm32)*+{;+L@+%O2FSv!eLw=4;OhTX+t|+%Cbk<0 z2`GRq#FReh9t4PQ6jfZD_r@OX%+39{zl3)V?jQ%yfGY)M^*Ly4{}CXhJ^#jD-ASg; zl1Rb?S-Jk#Pu093VEv;K)NrG}f27#~5ptw3&Mg!|1h%4u6ma8X2eY`45>VU&;n;{? zy|lD+fG27XhQ=T-g%<>fAcvUJ(ZRr%mH=nZ?th-KUwHD8OgIqb0=n%1#j1P;R6fwV z)CFbnXQ#R5t0qYKA#t`$tl#WgXB_({Bm)i57NDtAT7G|b_U20t*_}?_g7Av+@{4kr zB3$4U_1^=^10WRZ9v{e*y*&I9hP(BRqofiX!WoIvc{F{nsRLT9`#<>CK^S~E9_3<{f2D#? zb*~3hB(DwhfJ`13krc9fvoQ}v9tCRuycz#$Jq(dghj^{lL|Cwz!Z;ef9{w*MBx8Sn z%)m9~kE9u#(kb8vYZrnAtELpN8VB^Lrl4IY8i1^DO0U{J*+3TY4Tu^k%?KOBxuE`2#X;SGqWE|Q+cZkTJ%9*n(RKZ4^kk*fp z#OuJGM6m_jFW&R#dL|69jup3}3XSg;zSg(mZ&FDS@gfGCqL zhg`iTatjS~n-Lu%((x#Yfzh2=Ea2%x)&Ojet6uZ4#ecOq%WUB6PH2ALKy#;%ibGod zBk&k*GDqr~Seh_^E+@Y6m>MrO6x{$Jd0&#TkUY{1U%&VW^4EyC2Uf<%0pw&=P?;+P zT%S4E*aq^Co?(-xfB7PlD(+JaC^@QEK}ulSukWfZ9n-4tr&Ki_G9Du^u3D)SU>~}G zXjf$3{pwvpg1MZW-1@*zJ{v3>m{zgDnnE_zrhpijQ#eL)Ghy|RCYZkY{e~}~2wY;{ zF1`&lvuY(c^{aaU!e}9ReNh38C9ZehdIkcFMp}M$hmW$npvniaiMhwlmo|f0a?fw7 zYcJ3x`1ApTapbwuws1v7g`iOmW@dZ%l?YlX&2BNrwd~_T{wY2vf@XpqzPCzfUJ6P#I(Yq&i zLXuUk3&N-$2%*iin?NNC*w)Ms6cFDcn#qn_020*9{e-oSo+OKZ0jVUO<2M4Dse)c! zURVA^JZLFL;4_}->gs;2lhNcBbEO6N-`JM{dEqI0f$x2N%0Ec6%+3Jfw;7Df4N|=D z2~ZIz7*h$^C%xB)Fa@ku=Rb!Aq&Yj+J5x0!g0@2)xEIK?Pds98mZZ5m5CmC{?PrhD z<^?5mp0E}>${J>TM!@!LIlvxSKMHY|Q4fSS`l|_>rK#yNz9?ZI8f?l{WSzBp0-Pzp z9)9@)X2xDuJoYHVe31v1}1AP0!t*)zaO4-n4u@2udc z^T9Qg?%{&+Qvg~B<`lCm-MBWGj3#N4*-#&t^ebeLPFYIQ7hEq9RcO)ql602yKHz`{ z#B77ek8I2OCUyNt5Eo37{>@+V^3f6y<(C6;$)Dsl8KX0tI6$vA7>e{f@K4O%i2N(U zPR1`Gl!nphuyq6p8RJltW6Fk5hoLDPLrHWsRc#6rpil+Ou|pW^Z3j-u4I63|Ob1FHFgD6{vfMiZkR z8VslTV*jxZcrh2sgP@vn@8EMUX6g_aYuo2AYKFvm-V%N`KpO~#|<`l)89j0Vo-}$4;l!0yoE95Zzu`6(~`Ad z(5J`=Nschf+jc+-+tB#2j#L+3K(^%@ewhGsL8c`9#K&MvRBF~16!|{Rm-96W?`g$6 zOPvHj8-jJpGSRS+A-6iqQAWQ_HV0C5HnYRxY&bmg3w-+a!^|Tw2S7$T1=x6g&vYiO z7{8&dlRJpTi#WDix;s&zZ{Gy8v9*;AUouh*w_wD^)^6X=G@q2g8n)x-h~c42-=RF!_JZ_e_&YXMy56_8DNHMrT&DxZj% zLq7zX92ifSL3h8f4`G=WAW;U*QPq}E0Qb?BJkRaAl}*S73c^A2 z+z{%~Pg&P7_b&2O282Dq{k8?*iGu>}I)4);OG?W2b+U`!dkd}LA6pE=mgnPt9$$$? z5e33w4WdQxBqb@zhJG8QE0St!0J7P&eQe9BBkz{wJAZ|bAGq|;n+RtmhwvB5(aQsG z=efVre`8vEz&lgen2R{RA!F17DU0>zH>K z=oTaEafR_I#61y@(KDdLM}d*lEfo}V!_;E_S7+7@fU&;`?jO}3L~d_&E3CYhU}!|V zv9sPApCw*L=TCu|BYv|wO5Z4p=_!KXGnfhU^jEuBl2X44KNXjQ(#asCcm00OM!UT~ z6yYV#Hf!34n4h4wsXe`En>`JH#TH6m!oAkeo`PQ5Le5 zn0bhA3{*Q0m*kEyFg2fG4fd~(m6Uk1-F-XGtpYB$F$JTV2`lP37^_a`dUihElal_D z#{HLLuiX5|o?;vk^+hT@g9@^eLKOiV_Trwk@h~a53(3-q{&#`_N zvT=iaVWGtC8o45QJptuX;7^#ub{W6mcF~I*i}l=gNY0XlO;Je-hsBvm?^n5zsU0xvYPtP_T!w6k8^vl~@9Qq!aw5x^NUMCw&W0Zu|@Nt6Wv zQ`38}ur~R|1!LChvmj~u2+sR0vxc4Z*f$9WEF;e5-;1HM}>N ziT~LM+nfuv$cxpx6((|`-{g+M@4ZgKaH{)7=0pZq3!;~bXzDelq5KZ;d$=C}IoAmL zmK*t#J{MJ@zF_!!mZPHlhJn2f&YmW8E9qi-P?CB)FV6pjM=jLLT{|EA)JmF3H=)aD z#E^_rXI^7Wp@QLZ2q1u*5B|*P5M9*)?V4ixxK>${mr8|M<(XGR;y%Fd;Ho*i?Jeo$ zfHGpJ6JzI-&O+ic3Dg3kpWH7C;UFCqeEWIg&r;u>&3EN{Nq$GYCR0U?KA*ghnCaFAddXWa zG{!=nJ!EFJ41DnjZA1&9Xb&YaEY+m-{s_%{T#_nqZV3>%9}#+4<%%VCg+5t#6&+4I zpZ>;Bf16VsQsAb6_{P#STR`{M2eRe|6u7B@PZnBpE6q~xS42E}(6A z?MU`WMyd!GlrNzo(tdFhvN>PG<<>tD7C3|z)1xk4qHl$QuldA#_0oTO=R)W+t}hD% zy%Y6LpiKX@5p@5!Nr7#9d+kUSe9rB)F=zu$LBrRdJpz6XpTXcKkl*WnGH&tTH&jpj z`t@sBjL>8cxQWb_Q|=8a>MSUb-vj1wHN>DDQ@y#lxlE&;xR4^9W;55^BnUR*S#kb% z3o9!<&`kdf>c7z}Io#_O!b(^raV2?OUENm?IYYoee!^2bHnOTK9gq#8WV2D_v!CBU zYZAO2(Ec=zh_G{EsKjJ~-VWVW(EO&KHQm|dt=D4Qt?qr2--} zWv>aSWIwlro1B8RG(iWN2#|@Fff(*B8<#(2awE%c+%_-G9-w-ATj7M$3icqtK?WE4 ziQI-DIc{@GpV=2+?&eBSmz+5>817*QZ^=w9io2=|wpjGl;C7ZS9A zclwS6Y|p>63x?QX-C;<^z!>?Y-S2&}4Pn5k!nk)|zRxfO)UpuaFjMY3`#FSg;%RW> zT>4>HNz0Xy!9VBDNvExXZ2S+5GI|q}JstxkCZPzszJ~BA%B696<%`X6^}!0cx9yjL z>J7Hkk@KIyVq-vyjo$y@Yq;zbLZSeqDz`q=J@?ZJzoV|*)7TaNo=irQn2>7`{|Ts$ znldu6&Vy4en3@?J)sx<}KCM|e=|+cBOY2LgW@>f>6X4O>t)pkb+~6RA2;acMg!N{R z;L^qAXLArn@R6J2$g*g4Jd|#bIw#x_v=s)rnW^3LF(9!~Xv+I6#r8AsHdtOu%df9* z3z>J}l^Ekw@U|+5j=8}#g&@5|vVgUBTBtX$sOLF~kb7GO`lNH*LN)`f)^E(DUR4_W zeY_!%9MC;{3}4^Cwh$aRa8`f1ys~nB9FLu7v1F=Nh56&jK81<^h9tp!f;+n3=Hsf4Av{jkApKW z5qu7a*U`bKys5)Ghqeim`>w#z`yo{C_U2P}A6hTB40OzPYqP1n?HU@OFVLb&qlmlszwp0b5| zf`aUb)veA)SAV(2{ezMro(E$&z+Wtp%A(5M;ZmO<{2Jp;5%Y-z|0U5=45G^X*pbJ0 zA$Z@^8`$nrLnKGkrMpM*imhdaZ@iv@!l(d%Ee^P3oV%D><@M}}QTcuU6C{!ud45k* z>c69tq4~f5=l3mMPc*=seToYmFG6(6G6V79FEd6EZ{lvW5h|9%5k{iRynV-TcQtC4 z+!*f39)9bQ3@Zx@BV!l~;gg4KneddMC~O3YT+#Tof3>a)#gF#_HFI-Kb2|sUuG`|) zRy9?}lRkbYlRm**uh4NZr3vk&m+=quVV~HCR}~1+H^RnK8OD)1G5&P=MCj6l|MqJm z1bjyu>ZT2@wr&wFPW%$mtYHGDU!gnL#E#idEGc5O`&HD>H-?7{NiSC){rmT}{cf$g z;)r%2RQl3#@xA56^=a(&xF6_=ZeDHgToqqlr(d6?&v^KadAQA;H_c654!|zePL?FE zmS$kryD*Qr%bB?u!wtQ$dgRy60sNMu&rWXF7yZ}M`_2!q=dQNq^yvIsj;l&$_RoUo zPlLAW$C?~Zm?ig_x2|>1E&#OYa}wj^oBj4E{c0^;Z}NJN{@UfJIds{)bKb0V zMLj)_ITL%e8e6>+(m9Z8adZ-JeKJ<*R&{XV+kaxf{E7a&jNZRS}jL%Lk-9JqW5EUKipEdhcW7DIPsS? zt7Xn#Gke(rAmD1pNl51WY94lFvJvn&sInRPvqZ>mpycQ%lbghsRn;qw+3(HHQL#sp z-F70TzBK)Kdbyq7o6m9K`xWAnX9+RfI%Hg+#w{-pb<~F+?sd>HY z+hxBVBDzeySXQ{ckc8P|80OL+)*Z^~b=z_t(ttBWAAd4Zp_Uq+PgPuUTo|zfFJfrl#xF&?kvU{RTs(cK*Mu{d0d2 zMawn}q8tYmAzh`OABs-cs4ijYmyML6^L@G(AskN9VYa7sQf##12MfG)=QA1Me)E#}899-lPrqA7=f2 z_M(06=QamQQv!V**dA97!=!UTVrv;LkDHMfOu9qyL&R?Cu+@IS?^ge|$=hBbBv`4? zQ;BYp;RQ#YE|Ry5`yFP}Zvl!y;kN44vTA(`@>9#RC{Hpt`ozq-R?YghTZxZb%i-ge z5i;v{<@1(l!nx5_^%`3F#LWC%bREmq6d{R@C~k!V>^!S6{&8( zcJi@Dx9DuyFHt*}0oNVRZqfRa=94C_CLX`F$74AfXbED5WT%`1GhmkC4uRNi>-GO^dLeed}0buz>P1Bac)V10PN%Q{Ok{3XzcCKFs4c`*JEWMHM zOJ?un^slKSkwYE&a;&*UDRB5b_;vA=0WHB6s2#MMA*^_P1;D|jI_g%m> zwy8e~g861l(AM)AjMre}lQ}xQ$fkDlsQYeM4+uiF4Zw!X1}*Wx1Jw+ZB)V zZNR!}#c$g+eKdF3Jk#$sgQ(ZH9niFL@WSDKn^-2JhIQGnVYoAith+d`Hnri1mZd)* z@1Cq(L4F4@$ciJ!QtWlO%T8jzS>jZde{dj~Y(|8Cmr7|{7e0R$aThFz$iBqiJ4$~# zE2jUaaUyCP`HPax__B^hIbL$g;Kp&^19EfHR0|_Mj+TD#7!KR!V$TQht0R8gXv#d; zq!iD9f!>q#Nh_8_>6#%g!R6@vHoAaiW_nvcm2)kI8u!aikP#&zF^>YX*~d7*SWDJW?5D8k%{I2=XuaMHJzV|ix8pOnUD{6Y z{Z1$bLx2MkXh<3-++A_zZqki3TQKF1=Q5UOQePRxl-1c|T`$b+_z~l%V6yyK2SaW;U^@(R#EpTCL~!OI(`*u2MG+=!GHbgdwI&$ zX4PGlemqx;++o0$Hwo%6>jFBw!KNL*9Xs3|GB>V)TFxTf+KN?6mp-#Z4b5DF8)VAp z!8f}~8Ei~<`@zjH#gvJmjT_E=ro(ctR;Q;zhc(^QB+CZ_*U10u*9;hyI59-bV#4>( zkh4dp@ze7V2GXcn$rD$}u`Aeb**Ns%rh~K{dj|t)da~Inj;4W83pthBpFVPWdd+_M zeAo|Q61(8Ko?fSY#>d)eLae10AWgNCwhe2h^BzS_lVXZOlDVz%iy!!6@DKhHDUgrT zCp*rQmK-iVlnJQL;4o2W&`NiG7JK{=mrSE-M{~7($E)9G((H4X9?I>Hb{}z(EvhXy zxt-G+xtMTwi77O_^P#j|x58n_(qmO;;^olIwVU^|v4W$}vvD-ByV{EFB>7Fo)VNtS`RGwY=j8cY}?#xu}l{>oW;ydf&VjoGv2I%vi3IQ0*Y2p=>nnYNHG+ z>dx83Cg%wy22ZTjjMiAjH}NxkHN5^KG(Aa7_#z8t0jtS;Pg=XvwqN8WuVGx{O1oLe z90yw)Tv>7S{K%_rT2pqK>~g*fELv zHM8M+jFY#Hr}ws>n!GUVf_d9-n!7&N8~05TU4J}TCUo~{LyAv|)B2o!sFZ6^FmmGg zCXWwGX>3efwb-(u>rL%GzH!2(8?;OwUN{5-;V85k+~(nBA@nlcIf5g6wnGY*WOy8_ z@BPu22Hv3_9af|$P%M3wX?WKE%-p^6zVXt*6JgGy%m>cZ-n{D^=Z=K50Cdn$$T zysNGU{M#G3d+lP$uXfXOEPp)W$@!&mHi6@&V|x|?_VTX~ajT`ir+q}dXyYyiSj~BR zjY24;=bzTVOv7JX#^zxz1J@VPR7=5IK|`{xFr>B0Bc9dFo!+3{^uF6JS{y%R&yrOK zAGm%VI{%iiJvOaiw2x<9@7IbMFDVgf^u~Xr@%>H+EGL$uEb1@{X6dN3v#jpT#^bH1 z8r9E!39A+wXXR)05owLhqCfm%*(l_XdP!YO$xUj{k(L)_&*Zn$kFgv|jqm1yhmzxbYDm+Ij}eEN zPULG!O>!SMN1laz&e_=id?b1C&h7qttog^5EB-Z2*Be=A?MvKWaBC!O%lXeAu|J=NlgzNQxpwb`o;NfSWJ~ zz~a)iupnc?cf;!F7gAjFwsyFiDlk{%BT0l4k%PXs?OF;KK zZwjEplWPfX;bL2FH1$bvhuRk3|ukKka2{cs@PsBqviA!#l6WR>o z`vqD|HkX^3Fh3j70`=6AWptsXV1{DA;W$#39Cw$N>tgYu`o79A1;4 z`=dsVR-hkcSwKS+9k~up{%2hr1>x?~uAyDu}PEZ@tiA zmEv9@dE)oBXyOud_L}>Qd+F0?*a*vhhs`HkY7?;f_TiARxFR3N_C3Jh8?-!NTr1~z z5&hXjV>QggO~mTMt!UYE=J^w3|x6S0i|1; z{Kl<~Z)mOi&tCn`?vz}(7meSQ{WEZ{v#}$|YSx_K3oHMJ2Vpbs@B4NfqZsmD^N}x7 z@0s@Lu9=?Et4xClN@=M^MK4qy?wkp^p_bPX1SA_hnriI4AkX{K(uR@BgEfU#Th(mX zaXtuh)va%<>1tTq?u$sYBj0P<=d&N^_F(U4d3MU?HpMVBd-r6uMZqv z5&BIdO|qRoZoY9NbT2y`@QtVKFs4de%e+2+zw+#cwOz9(K%YGTrt0{e+~bZajjkDY zaNn!VjWtHf6??~ zoaIr|{i^TDr4GUrr)$B)6ZaoYDygO2P$!1*m+8m{+&?Dk#{W#Jr!SKK9hZiAvC_j*HQA9larH)E_JqryPNVG?j5!qa?}_psS#*%p}F zxBFr@4;iT#679-bO)Brc8OvwnB0}`YUEXT@n#{^0WburbkUB8*1gY(wwOj&7OMUJG z_#)jZ*b=W>TIpUjn1lc!fVDT{fI4N4U66LGK=&iMx79D4TvMki=Wka#RI&cl$G7OS zwW{Fq#=T&4Djyh~oDo0l$ik}QKCzHqbA640i=*|6p=@JQyuqS~Kb)a7MgUuRi6BpI zPM)7Eh@ERmC&p(pQR;TRQbgPb8poWt`m?mA{6N=gSQdRnn{E!doRy96lGqvczH?b; zmz7Wmz7auLn|iLf#Kiy6be3^V|KIzUj*%h*q(MQYK}I(SN-9c7jaC>UFv(FvK&7QS z1^Fo54N{{^V04e6G^0lS-+q6O|1EC#fLp%K`<&}s*Yn^h=O@adI57l=d|qpjLDscT zWd>QW&$53PT4`}+W0=P!I4B<5a(>B33a`9aGXJ!w$%IkM=H<6-4JcNs1i^&;t$Y28 z{3Cvw=oyqAu=~Ka$J(~YeAoeVjEk`HJK?1U^rM^P9p?V$K0|J3hT2~?!cDf(u=#vP z7?sJg$KAW^`AabJ1)x)Etz;zU_mA7IDHfa07;RSbV=bc3gWJjYN3p@~awi|snonk1 zz!Kb^`C#)MBT3-u`%XjjQExFQ@cS9c#+?~W#ltuks@&<)))LWoUMP_%a+gUpXb(RZZ_+HHFS_K1j8wtXOLCmEa{c~*ul`jDtMVOb8qf4Ahh%5B zB>z$k>601h;a>X-W83FbXZKwdzeG_=3kH1U^7ODi3V;7{j_8spJ@bzXGcHybcRV^D zytp!e<<_%#v^N_rW8_kLTM*^Ax|T9`n7XmKN;b*%nu{>o|4|XsHhOs}g>@P+g{QiX zT->j+K=65v$)~s!#K1isE;mhh)W!&>*$$Ak3D7pTApW?w`8;lVP&Sxe$LBN@rklqA zyW@XHz1wJvK3%dzVP!We*&CS{6lKpaV`VGOX?Wf0b+EOcjwfBn4Feob6rYpOlfHYW`YKPHHy6?E0*oXWeZyunGF&hl@Fmnr9pjumBroGTNe zDONs4%FcA((~Kn}8&&>TUjD|>CR+aeYG*=6uR>M@7f()C0mlJXOd1;13k}PW7-g{& ztj7c+YEyYRRgsPD5R2x3h?ial4gG0|~9pZj(>kowa_lZyyAA@vn9HtIlF_3!vK zWItZPC)GPP^Jy@JphbB|PzLtni6yanfmepY;XQu-SPFEW$of!7O9$G#*S^=!+dL52$sI{-Cni-`E8gXsF;msr%347l(?Wz;Jk6< z6c_SNkg&m{wJ&ce#$qygA>{klaHtW3Lqs|JWdOVk_JTea{wq(E_1-Pz)IQ&oR)m_o zZn)=o8OkDiVwxDibxOQDPSC;9>NVodZbCpz&bm(h>l3^CZCdsOPWFl38NPLh532~4 zF!hny#!&H-CzVWW0c+t<8bt2Iklw{&&7CnJ@XBCJvINdSgE z!&v3900o%1a&&X%eJ~G(5ijk}#q0jq$jwP+6vnU+S2VCk0vVnU`MFiL%#ltvUxx5E zdw#-GYa?5od)-G9Ye-^{`qz|eRDVc}IQj3utaq_;j)}KM0fmX<#qryY6})SdKxwOB z@A?DwbUtEGr2xK-i-)PD)J=c(ldSF6gSeJlFM9#xo_b9kgvYfphqDR&=49sb>(2M6 z2eW=MS5YM#>BhfF)`vL7%W{775Vw*v_|c$PxE*Nvj7;h!TPg0mWU|#$S{ScGe|bCl zWHyT^n)2tQT$|6J@HjCeXxGQp9JMQnhY6 zr+bYJY^*bFOGB9syx?7=cU|39PRlhRyQ7+oDoog@C2Z_@Q^WR)OUsV$ge2#dAMVyQ zh$;qgyiT@>*lKX!;>9fEBX7=$4yx+-P|56%`OHMbjgB&5%i>S&)-812tT{94mz$gL zw@dzGg+`H;_q=h5;79|ZLa1Nz6)I!%336*4dM)0 zq~p^+loInO3r&V@(0zPc7fDg^lvFd5*I~4Xfsa^=7X(ESGMIpHU`*Wi>nvni^>2Ru zD7IuGy!BVT#=kyEfom`)|?i!?M}`sh%ddC0vVO^Mg&$)Yf#*%_1>+XWt; z<8}_2@S%vf>o-x1H{-{EN79-%8=sg3P=M~edA+GTp?g#H6#mWCIFbox;`?Vg;H}$k zTy(!h2PKNzr-B`xA`AlOXm`M{AxNIc^L1#XsONhl0jPI>vo(FJ2g-lH$a3GbLWZ$m|@dHqdvGnqr= zBgK#k^i|-petvGlH5E=t$78k@$udz*N+TLb2ip}OOgVcZ1F*KZ4)2nsI8c+-T5~|Z zzIw`$b28yIvRkt$zi*5UDNK3iCpFRGoln>c$~T3)$NyEg?)6Qjc%@w2C3AVRAGsN! zRkD{E2~!lpj@cCZP!LN0`{?>Hw?$nc9)+%DTTho^eit;X^+pLA*KvY9gp(}T7 z0hUt`4zJ~uuDLTR7Yv19Qq)$)2A^gW-o%E>7i9hn1R@Bvy7L&DOJ~`JgdygRXpnEH zBQTif`TW;zUu0Jm0q^8<@o+dY3a3JU(w0iLVG__v+D$jC`mwzP*vXdQS;FIi=da^M z9gCV^<&`riCYZ(k#hI=*fqd47XMaD_2tecaMPXl^!8b~bm(l?RImYSgTfB;6zM3=3 zKqM^h{GB(YtTS)%MdSkF7xXXDO~^zS#VXG}9VBV(-g}~MrKT?euW8T*vQJOHMFc__ z0*{RYYqnNHIB+?&#m@uVOUx$TYY|;Z>RUWwy1i`z{p+-oIAVbBP-IcMdO|kSz1nFE z!lc`fpcQ>J^-F5XM`}o`(#}abUUPRKRYb|i#q`)iT*Yn1ZB;sdq3FOEq4&}hDCC!J zPmn}JMhCuy^goRdO=|HBmFZ?&Vs?mAaQ9Oh5kN<7V$hS$mAgJ$YXS^{#q6IWZkfNk zHS4KCV#q~`n!lHL87^vC8`7Ppbd{@(DlJy=vimE}&F4HK$FdIy^ggFwjGOcP4kE1P zTYzQE?y&6NpS7`4u3|<8Wd86}=7&|z^(YYE85k=_+@F2T8*u>&=%`b>_1$wdXq1G( z&_x==R|@~Nt0bPUJwQE`LgM?-yR$F3=DiNBFV9dnwI%} z;0A1r&WdCj=rpi&R(X^VaQ4_J18~&4SfX~@aNk6BEH7rKYmKAvQMxQu&$q*5qRt8Y zew8viJ8X&h*;NcRtERz7nizFDM~!~p{16KeL^caXn$p=W4J^^u_jpnr8>XV^@z%p{ zJEqJhn~&{9{7f$%kCZP%FHK`)x3Qy7CiQ0h&=|Fpy_xU79Fm2mM3@w$K5q6cUjAA- zPV>F;C%^u(Oa_ms!te@T5J*5v}2SKgHFl{ELhT zD~#ivuudJx{n_$Pj&#knWf-7=$PbWjGrURkGp}8P77`sVPp&xv-IaaSSVmoQcNSkw} z#rJPqyUCFnfhne-o(TU`gnfQh3jff`Go8kf=TP>}W8EMmoi|#$v z|8QF_onOgwjI9G-Ch2_+&Sv`mjWTqf#W#W#Z1II$`6G45zA-dK<;xLvMkA!-1=G6( z4d~ZPI%Yo0no+CR!VaxP6vQqwbP0%B)cRp0vS$>4#E&cJ0{p2(cCw3toAsD(9p{vKyl4s&?~6Yh=ICAZ00G2WJM$I*kukkdXv5 zndT0QMl|ZfVOHq4$Vr!nr#TR7dAo3P5qMMPH&vB2Nf;@#?bjV<3?-+eFp3VaN}D80 zru20jR0*3Kgig0aaiZh*%05F(|1c^K{xY#CQ4o>u2O+!LaN&sTDiJ;`Hg`R=zNU^QcT zfGMD|HRy>vAI;LP_}jC8NEW$<$;V?}Fje+xyhhye(6~z88$Fev?4czN*qJ@i2~=#i z)cT>9B3h=r95a;o&05$tdSwP335eInL~~KR)#wpYvs>U>yvh+*kgW10SgVGqieL|! zr)4bIvfpEXAaFbkDNtIUHzY~SONm-(N)ow!W8>G3P+MkDcznZ%Sia|>VsbV)uGZm4 zNVabYb{H}ysrUALbm}A#!eD(?51(f1?!z8RFtPk9*1*Lt8k-_ZXw$8+yu0c_Tul$8 zLl?&ge1n~Vlx<-{#koL0E!>%MUNh!Ec)W0JiV@2yXHr|Pmp9FSHVXs%rvELH$#I;0 zgwfC#<$TY1a9G5Ti=x$_HL}fEA34}?Q-zhZ3w9u3CE}=vVbJY6g5L$H4%2NJJZTHcDN<1c{LT64?6*6)P#ujBGoQx zegA;Ur2ma$=(K=2a7EVTIkOnkAX!nSx&K#!ROi0|7fh%uHTz1Z@h|t5>qlPOrITz; z$T6_SL&d8@7@scl_??fFH8BG~hV7FyB9Ie_y@8RBSx$T%7oG2a^Kg6e^G9`lm!*SQ z*{^qZ$ZLOvb_uS=ZaW04?RdpMS*IoQef0J*30}-`k+8XS`1zp>=oJ{$4vzrz30UY; z$eij@+%*9Fn)@eM*mm%Cz<*-U%7KwGS}H1X_}vuUKY#f%OSmdjXQY>oKO|Z(|MJ@Y zbeTg)U{v*Pj)Zm0;qcCS{oN22%8-Y?tZTJ@vtJ}J9B3X>r*{i@VLr~YZ1Zd4;0e?ljzxf|oB=-9#b9izRS zfRpEh1;jG>zn`T{rMOcd(UWE>+FJmE_hpRm-_wrvllQ=QsgZ|M4MfBr@L)Bx?$+vmnCgZGIk z`^{#ZD}dI_a+E*?@r@Rt*9d(lO3c~GN2D0-6BZ@xJyz_C;_r!B8y!p^blm@`Hx1{h z9TF}zZu#?S=eM@+e=#_E8awBZt6cu=gmc04-X~)%+)g2_erFweYeo9o^LG-IHYV86 zRklSosaAh^F9l9guA_&_LQ2A~@%L!zZ*HCipBC-GM`v$cIJ3Kh_qC?zu|n_S^uj>% z0ocEsbDKAb(NR~M-=s8pBa$DJi3UR3B%Zi(4vu$Dl4gnvr_0UKK^Nc7dyW5ScUtp0 z7lMo6nYPKb%As_TLG*PGtZl3fD&ANJPjX3sKE!mit9MlyvUfbBYqvh={gu|&2o+ZU z)Lmy?o+Iu~M{+a&&kztTz4~dlBE}zfYSwmawg3P|Cls(Mv`$R{2}gh#+=-vc*FPup z8&pv+GTFRF#;3_aC?nWU`cHu7gV?o~Um3W+kgF-9D|Dk{5bL~6y$ZW-3jiG8Y{vm#aZ=CSd*5OA|3tlnqO;)MNFLrlGcm8Ypt8e#& zvwH6*RA>~lJ9fPt;b?3dXMQ)so#Cb3?NrkgCbWdJ)Z+*IFD9ean@;|@E4+Z~iSX0( zw)1q|5gH(Z4C$-E3Y4F9G3WzuhNOUUALWy8A@=duRl(7Las5-j+Oj}c#-8<``b~}# z6&7>@v^S`m3B-i~gAK8f2R-v0&3Ux4-O=UzK;9$J3m6?NB*5p0b|Q{ZB-vKX5SiXR zMp0>VbqWYrvW!chOj8zMUNmXSuhLJWzMAiRR4<)j-%_uyYdtP;3GPgJK}^P)+xGW# zTu@Hbb^gdN%8xOY4HjS$RJp$g2fI6R4uTwnR9P6#_%V#&Q(Bqz*j;N{Ka+Ih(Xw87 zjMzd48RtBBSgoi_I$^VXmGXpIbJBT;H-<;rSVd(l|-X$n2SBq?wZ=R6ODP{C3~o=QW1y_B%Y#FY<~ab$=2Gl-kuG4U=PxECg5QRp*N<7DgCUq z5L|xR;=VBV;%D4ySw`ibS2I1XK9#|fWa64sp2YV|L;-h$=nBe>kOjW}QNtj+imG|< za#x(ryOvPch_XvFlMwrni6P^9Y*_7#J#Mv&SyRO9Yyn8+G-i4&HYjWwmgm?W$kbVi ztyYV6AWeU!N2QtIHk^12#5q}U+OC&+MdpSS0w6n$=GXA!yHxJ?CS+F^?sf6sRw)nv z{KdF$w)?)wdB3YsLoXQH?wX-I;^$eyXKY4re>&|}lHQw5Q*515YC@*zL+N!w4SFrG zy+BUNHI+Vvz3D(HOxCVV zOd6@cJzKi?rx9YpD;9U_091c{Ch4b3kMRt{&J1-8j6b=kt-j!F1xWG)w?lkkZE3x2 zBal(wZd9)M!+H&wsKNWCcwM$lbx6}rR~Bl+>HWh+=KD)COgtTy&E{lnPgKHMU5b3eKjZ-WroSu z6Ni9*d((4z_~qh8FJn0d;cm)7%9H;VsD1#4 zb&1e46aKJ0ZMZ2r#=Zr*jjQ7_p~N-@z05vS#><1>hq8XJtDuOr4rCClZ>ga7fE+!w zJu6bW+JXQpn_IR;wqzadDkFnL-GR9(3II`_gq3lZeIX+ae)HTU=u zI}40W+HC#FBKXw~-5uh76_3>YaRPn~4)5h)-C?aq5qP~j#hJ^t>l)=X%QXz(c^M5x z(a4v#wPXv4*a8Y}o2j&+NrP2kp~^M|SWCa1NxdxhXAjYnk7x;qQI%Zeo`NRr3|y!R zh0P3)*Fm525C5l5v;`m{z;hwqt`tm!&L_lSm_~TsL8kQ z?~9a$-`75(@}OKVJ?{Zn&@=zeqwc8No|>Y)f!zE8_<<8QOcH}ix6fi93|kgf>qI^Ply$yp=BJzE=8Gw?NAD1siDmaBjckjv*5& z@&W;rM%~YPo1&T@ZdaU#`KU^V?|dFt6QvbW$dDAl{(Ur>Dm!LRIq0{+uQfWKycA-! zCiO3%yx@A-<60e$rj^C!U?gVlY~nr~3C{}zK19!cZ-3P|HQigm<#Z}|>Y_kMeQr4A z@^fZCb^lS|WpCiT!g-(mZWgY9|3ll=T-$;I;3(WHtzZ4pGk_*X1$v3J36L9o0fy4| zTbJlrWR6K&Q%sg$;kFmuPo~I~{~3?C+9aRLexB&^Ta^!iNBnmWdId9{-xu(fb+6Gf7VeMr6zz_RbE= zCaSd7{P-G=-Ak)(Qf~k~_Gl#2d_fm9aPK8i^T8f^UZqwr#Bcft@nFN9C?}Y)iw! z&bw)Nj3qaYgqDj28Kr3zgH3YQx%toaB6HXm`0m3SU>Y+b*|w0W; zuEb2iPu1XCcD;>;?LDxt))5Zoa5QCLa>@6lN9^iUO}|teVw!(F9zMhbUAriv>n-jv$2f(+a~V=9+IDJ2{^sW@ld~u6DIWXA#E>0zO8{wexOy` zkuuWkLmrnflGB>%hoRfX-dD{lArX%+P-oAY2Zyfu;0+#$cqC>Z{rc|_?@yJTqd;3KO_5?Ea1=Qa9^4#1K#bJ=Wwx8BNKnk z{*dx9(YNO8Hujc?InS$L>F2-$l(QaJ3Qyv$XMl0Z;JqoPDJ$MgA`V$fYaQemG8aGcf&sk##x@0>%=*3xQ%B_&oCLertf1s zkDH!`I1BN82NTl&r{@}%ZBed85?izsM}ZrBr_~mfIH%sDFz$BqI>}EUNHL*gnabIC zNA8JUWAEN9S^`elSM>th1PcqYYUB#K^SR-!0kFUR6Y=S*)s{(l1y}ZX$DgGm*w-ch zyK3@t(@J;~TkRQw?5fc^{_QaFL0OkaXtq(;OvMjCAl0+XXUO?A*Al>|T<~mngOQ~C z47((zpm>m{jf{xWwDl;0dZ;dWekDZ?o#}_!&D06t{o*<#ccc1d6au`KoTB|YW`k)E zCso-lAfSNgQL0yuhffAIe3YRyQ{itth;n_*K9Khd5v$FK@rN}=l)cIN$I|fxSRPkq zHx?9;e|w1i7nUZ&!|j7s5IL9k14C}Ubtb2X|KHq%#zX&RCViNFTT~NIa*(0P{^8}C z7+&;L6rAGK7%3t^@sSqEJ!or=d5B>J)=_=oZ0=ff>h{Ao>YBEJ+^LIUBK0dw zf6a3_aw8(lZyJyHP8+~g`ANNVsWL~-?6WTZ+6j{E!-S&h3E>h3ipUC2TvA>0x6J?U zr*H$mu`xi>;Xo?fSLp4!-Wjnh5;XzbK!3yfBLrFlUF>yj`19Y7*RfoSL%z>A83#wa zihb@Y3XntOZlBWw7L@1tV17ZP6a;xedX1LNO6NP(Q)Y6(B;N8uy`LAS+(XR(t!=Y_ zyvT(annAbn;RUK{k`ncxQkPxe(_7dI-x+6KCAb+IoGI`r*VN@nq32m@f%{43)S7xC zWtXWgN;@^747}ph@hafZ=pAq@7njgFrQkoR!(IT9p!&JHLf+8t&c&- zhuElg_K|W%MPtH4PafK@pN36Gy&ff08-xDz7@ZX``LlzPq9^ZJV6$%#c$&8rTG>L1 zk)Qiwbl)RGHYwp_=#8dIMo)AVNbRh&N8znN=Zl_}2LEFoMjW9Vb~Ee)Tji8i6oHKqd)MQ)h>Iwzy!&4;)Q(G zU5^bw$AT)N{gV$G>N!sqnq!xgAWtq@Ad%iv=GOgeB(A}K*VD?qRQULsoGNdvyhKf0Vc0MNkTDsR(Z3e{ z#;ja}CGP;Ee66wlMHG#!zXI!Sop68B(6}bVIZhFu`A)Yc0vZidi#4`cb1AT2OS)xL z*1ZUL8fS}S+PFwsYlh~LlqJ0WJI+>3vfsE&6!D%ZG{Y08*XA7_S;g14iJd8qe$fJ_ zqIY&=^ok2ZkUv_u2V@)7ULm11%OxDfmOl1#4{R)k>W4+djE>B(IdxTM!?UbPp|rg9 zqx_20WKZ-%pjH33TdB#}vO%@*bl$TAZ$ZUj?V`*n^1MuR6h|2)T5h%mh=WcVtWFEUCR|<8dCPymF9YKXdRs%EmTYZqDo`VW( zHtLT$N@GV3gBS|A)K$e6A?RQGFXVR;USp%X5sglDD?gjD;_if78heR8&l>6^1fk;; zjVc&eH?uV=c&e5!ZoR|EBr-GHN7g#;d`myDQCFugYMVXJ#ZiGFeHGQHD3i-UE9F1tAdjoKp>l72RMv7HB^=qvXC;9VQ)K2 z6-Tu4hGTpl!+=`NoYUVfo>&$$*R17u3+7>h`A)qQv(blZStF*gZA%tC)N5Ir$ZsJW zgO!ZO3oj$LKJFb`*Dje@SA71topyB&gq+42mS22`hs#Mc1>Gf1;^N-pC z(`9nF18l>(=^g3kX=QI#Vs*2pqYdpf>LRy#Jnaf6-pqV7g?zfa?4d1Gx+EnCuW(Q zd~&HG_1rEO9HKNhpgalIez)yw9g;{|XE>$sRzLJ)WPU1Ti& z7VG;v`Gr7lPrGv9Pp$NS-JP!cX2_np2OGWGAz$VHSR&07VU zk)3L(B0@(Y2@~D&z^T)N0tn|Q>{@pnue8*sg1!6W-(es{<*3|Lk-D7;+s9ep=fJH>i? zYQi0v4wZHF#><_L$MfmUED+2MK!hlxmp|S!g!UH=G^MlK6VS ztHigCk19t0%E>;R>erc`S0ereZs5|@%OcSNT~JW5hC>|rtK+Y`j3VhDgSB+2gkpJl zx4w?3TJb;YInvT|uUu;-Or+P;11AXg25>wp!%x$-8X6s|bP0g`$7$)xJ09caYDLTf z|Hj)Efa<-HEbj=m%drfrHZChPLUrz{a%>Y{-P1n#w0PnYDi+2u_jxk_BxM^-)#Rv? zP`HU@bN7ElaraUL-vswh(mM+P_rPtW%eWl%eohoE0$HWH>6haSWiKlk7R>FlT&(_h z-^)Q$fVML#CvkL3H){hELOjta6)-g2f6= z2Vz{;8yWO*NQ_ZT0kM=}Ia`2sxdVaQDDeM#Gfp9XcuIGFE!R?cNe^e;`s@VOcJPr? z_1dz@p_%zetp}(nSmc1ohN%aWY{hRVdI#CsJlLCY*TyzgLp|W3ct)9bjNiQN%Dn4? zAFc0j_6~w|^d0D3R}avYw~>Kp`ifLKSwUHhU#$}7cM@+%kZyTSv2YcxOZrzv(lT$tK-5O?xh5h2TAoDm=O;& zud(mF0d@3%Hbh-Tl|E!okodcnin|WeZ>Qw2GUjMlp#pSh`<5=d6s{qTqDv|L+RHVA zQs#$`m)Iw~_WicJhw_2vG_F}2*E}FReCSU=+bLYB4|n@wA(+CSzP7eO^7k3Db;G^bvAh|C;ocPU3qgxJsX$m-_1CPY&U6Ji(1g%tgfXXNI6?$kd`yi5$#H7W>R7SEi`No|EV<|(%Rg;9>oh_~fF zjHWeJWbAbLeK}ZOS{^vlwe<++LTh!|06lcd@@uxESt|xB6@+KCdn>c2UUuK-3HD~f zeqrM!-X>r{SLAH@H@v;^Y&F8{<@vwh)2EWwqp(5IV#xdYy%JY{XR~`V|HQ5k8rknn zz-k4d`+k;3OM8}-uPbeu#vPiNs1{-O+(C0y)*kiwPQN3ME6=wa+B?Nde_O!ZX9S5}cCj-xnRCpDm5sJ(vTiW^;O^V%GaQ+3V_pDf?+>ud9!h6dvck zACpuG1cSDsGN;Im%Ra{^A2Pqobf~_^Q@#LKiBSnojh$mMi_? z-GjCO^V)WFJB_*bN~jAJHGlDve;Y~5IpS>z^geNPLEvaFGlN}wiu_5wq5@q zlu`7&FQ@T^qu%)SHi$55(Ra$bm;Ni$T!)cXM2QykKpOR`Y0Bhz5naM|EL7kelC#k8RG8^jv{m^qSQ^ z-^HA2z}X=~L>`}j1kqF9H!B~KehzV-XX8fa3-rh`?q+nCt)!n7^dj=aQYIREn>Fxw z<-k@3m~MH5w-sKC?4S(64sjgQzy`#gY&S{EO0|bK8mYkfs|Yo|p3+<$RfRXp664!{m8a z{F^HlLV$(JNz1vi_Xt#Cbyokv)%?nfj+iJVmRT9`T9h%?77G3C@7}HxCiKRHnc;B< z8b}Z=?o{Rz2>zCFFdB{OliZ{pyN@L;;Jr1rnyO(5D%GM487RB#FRS+3Fpdxx4oolL zedJV56zK6fQFA%DIpOofo8Y+4FLo8u`8r^KK}CwYXn-d^M}M+wHJsC6q0EU=lXKRL ze<}O?-hG?Cywq+}@zbVdRR$u^dX`2l<^KB^61clc5|5I05?o_6v|HKzo?4^o37c5g zM~%RUHh6Ddw#FF%*e?Iak!St<-0q|J1nXOqwP; zGTzU8V5R~=2KPnUmc(3SUEym6^yWy8-b|hx!kMB1v^)j z&egl@)_czKT(li`U>+ps@};c%&Vlg~EuvH2JuS$G&!TZUqe;{>CyJ|x6u^m*^uu4)TtVWB5SQs-ZDA^cXh zlrz7t;Ng0)iEf&=;(Rj2K1Qd`O`0sTbSQ2sleqb0HTzfUuWQ%gT=4d6x21JliO*At z5*Cc#1^O8UAc2E;EP3+q&otwIXPfgb$>oD}F}fYn&xDaxLo7o=TO>_7!*;y&!zKnN zi+jTE)*x(btyikZpCVX;nBXA=B)@|&}Qjyb`&}|7nS6Df=``xMa;^?O6(2a|E$t_ROR;`Iw zYX(EF2C1X0rKuF3S^NuL-1Z+97Gtp<(|}D`ik!rGlmE@a5FnnN zm?~4>P*1bNIB~}nQwk)yRhk%&Ud3=-bj9 z_CwF?zOc}mG{Kb7*yQfV#4I+*d6YiK@zb*QRT0%!mgLjc@5AJ!UHAlWHQ+n+^tccX zGwB2o3&>b56Q8VUPCBiBj9xhlT=hn>`1kO759ugHcEH4lv>4c@j}jYqEwWq)cF`5x zNfK#cA(V;vc>7E{YdhxYe~w}5*Mj6+Rqf6lmdgYddB*VI-G?HXN337_@1+EcQQwfQ zaVb?g#E@K?sB>HS1EH3nNA5)ug#Kt+ag`osKb%^vYEF(~zk~Ml**NVy#lXkepgC@f@jtiAG^VWFbt5~BbzduiC(d7J76pw5K<~avR z8_LgTpO;5o>oG;{;W|_HH~@qDe9=PR3r;32RPCF<5oXSUaCtr;dW3tj^u%C7s!?S+ zl4$+Obb*SLg?%dVIZr-SF%l-*3IJiHskT9G zVr<2h-4B;ze;1$>ndx}8@(NrT2J#71y0tAuarfMFe}J4aj~Dhy5;~|mzt8^T{G)u| z`22G#aL$ABGE6uoa9LIdd=4eT*@{yL`zE71J+qCq^`?u@(Jwsa*Gj{l&@QIi6v$2ktAGBwRGlfNp zS3bn0ay&HNG6vkI9mAs(M-;Lh0Tpr`(t`}dfk&KZw?oN*BNeorZB<29KrF$GUl`E9 z;$4@d-nT(5PlALiCNVMgRSIf+u;6B=fR*^GILK8((Lw)}y~LoHpx*I^LRgsFr_5Ns z3r$CD{eo`3^Q_F}>2f90H$ce2MCmtTcs|<7DXW84>|Oe1)f(oG^ds9SF4x*Q=5hea z4he|k{Oe;IRgq`eYtL3pBD;pH=)39Rcssoj_?CI|jH&0*k+tpZn-4(Lz6qml^R@w5 zLgGQ7NZPO_3Z$m7+&AIEscR|mDRqELfIFhnQteG^gucX5PB)1O^-MXYozF&OyQtZ}b-4u9%eVN|wa8rcLl9tD6YeU&x$r4hxz!u2WiRugJE z*D&j0>#Mu)bfIkz5v0M4&Uxxl)i?OeD-WNI75%BW+K9t_JB zqazFPDpW#vwcc4+9~l!noo>OU+=Mz}(vKFdFxNazY>eY_HYHX5 z4;T-p6-(r1$J7(RksYi!F4AZnt;HGDgbBxW^sKJZE$mq)SFqI?;8E94;}!Aj1-eW# zvL}p71UI!`H#8ymh``3)wUQs?iSj;?NYyv95^FJGQq?CKVI@+$M8TSEHl=-YL+Sg1}a_w-nX@;)~VN>*A3VnE;h6@*E@}q8Xr{t zWWJV@H1X!}BUJ^LF#!RXMTp1jV}-%bK~nJ!CB190H!RImz8wgzdwt38>QO<44sJAV z1yM4ucNyxx`NwJ6eGDQL|F0b+da#UYz_P#0VId68dW9DbCW=3(Fc`OCwJGIB~yG#R^;3t)RluFLXxr_PIT!L+3qxCHPybtN8*0rQfaiihyut z)!A|e;9!U`h~}GG&>6h#`U~BC5dvj4&aD%bO=+X;(DvJqU_G8(Y=*|ow4kNkUD?PfmbY0kqZ^5B& z*CMI&n1pS)Si^qaSzjUMUN5S~z7nOm_o*VCy@;X5-+y?OB%OBMRcCB?J^V4ZoPQ%M zOi?>VzLNu8T)b!8=Ey4DbttA6b#r+mlHHUdIGDx*6P{*q!@MQsLc_P zBuMO2LD9}Prr^0Wk9-%$%QDWeRjC30LQTJ&?+Ti#z$)+IuhNICIi2P`e3Cd4OAPqc ztHobedvH63c_aY6p}wz||p=!0Mu8ulVPfj=EC3{jX{w@(B&x@8Bhc z0kVVUn}eiF>m&A0Hw_XNm+YpHYeDjLT0$bevcInu(gRhq5x@Y~f z%g^zjSREN=^c!J%Px;-5dJvEi#JGaNJX0%?m34ZrxGm+!Z9(IU}q0W#R=}y(Nl!F;;o;^@?mOm2?j_6Cp7Qqbu? z2ej|<-H<*Ud42gK9#xS0uH$j6t57tP*s`HqJU5y41!dCmce6V?a!>hq`M6uvJiqcD z1}P@)Qt^*~r!U;$B4nc;e>9<5j z#t3{w2%GMe2PxuQNaQ+u9!O-;ndi)-3>qB|SqI5{zJ7xUYlfPml3MQbSK7zS_`2*G zW}hg`RI%C(G>p2|v}~dp!KQd<3~IC`i`N)m6xg&>1q4FuxGjDCfj_B`M6(bdy*HCn z$vB?J#FphgScIMTB+JX&4ir+wI^%-oYXlmcZEY!7&X-&zy|~C^ReIK~sl-lmBGN~d!2wT{%bmU^mTTKK@j$Ojs!G$dz?+~3&)IBq9 zr$Z*QZPCuXh^0v=0bnN#FcQ|~Jl=1H{rmu12|n|EPypO{mOGNlRL(S;V&11_#d{Hu zUC0Ux-kYI-sL53^+U1quqgx8~ZFAJfEyN!)J+@@7!anOmP{C(R!VB^x#<}(?8T&d16pXrU5*oFi3Kp}FxqbtIyRz1ybf$VCm&EKl) z(RrGPZ_?t{9KK$rhbNtz{cVzxO@?UUgF5$T0~ztx6@gdP{}4$vWPTQ{Ip^ORh*LCQ zOEbiBNh=2h-$c>(+)YPC64}?eWkBT6)jd*ig$D!Zm$B17e?SAExg%jLYe@}ZZawqk zXnL4S0>?O~?;Abq070$)%EIA8%@fVzJz99{T`D;6PuzGxg(D7+aYBXsI)wOXN_xu3 z>TH27nage!_mlSArL)(E;&qVTS5v^lj@+VPe=H)RlodUrqozsBJgSf1Owe|X@0grs zW7AxtX4tJEwG%cE@LxLV8@QQ@$EG)9^AYHmjjn4kP}Sz!R^0&wHdV zk&4UsQRdP?G)}u6oJo%eHO_Ipl=9sO7&(vPsCw(YO4x+5JY;eszLa9qAo*1ufy=Zg zn8pwyq{HL1B1V`v1OJBXjtVu}bv8m31Lp*t@*Bkx700>P_|rGyX^7@D92W*kXSUxA zIx<9UeN*ePc?+rrH$M{@aG>lL=g+wb@OU~>FOu_W5jP>vjZVvmmxQA8Ni!5qy0w(+yvbS{#r zh%#y1WtKL--u#(b{eD$C0Nkk~PY06J3JR6z0+Wq&%@cRG@GWq$hr|ykWKBx??|Gng z1+0A15FUiiD33vlXe9ygs)yE5&H_V zi&P1lKyTGf6Q ze!y?WZLyEoVgi21f-_S{4}WLjh0otO-b{ovR#%*OmI2Sv)pJHwX(nDQlS@GSu<2Hi z>AKa%)4kZO^*=sK+Ua_U|Btt~j%xb<|G!BosW$>59ZG|u(ltpH0qGtJ2neI3V@L=n zT_V!mAuWvvOcd!FF;a5G7%&Ed>oq@deShctzSp1EIj?ib4ffpg^{mJJe!o@a29JFv z8=h)0UZr4t(7-sx2Ixzi9h4-E15O4AAcGv3Hgq2LJ?DU16`ST5b?n>Zh=b9Gky}6n3!0ET1_rM#w z@fpgT(P+og!!WI_caow(ZbNm9@XuA&wCERI0iFH65FHvQE3piZxq=GW{L{$I5cv%d zboZZggQn1-3Y1{8gh0uIA|0O~PLYT&CbhZ4b+osGA&AH;ORNOTEv3zCGE6hKu%Kyl z04{m&*Jsw2l72CXy>t18-o8eP*tYIg4R}DUK9h*gIGlbs@W&t=` z$mp|9$qEHjV!X2-sX6DLUM82Pw4#^hfkip!o(uIQ&W>7lP%YJhKu#~uAr;_dDvtjn zB-xCozCf0}$=|pT-bK+TC~pj<{VEtrm3OU=>$4%W)?S(-OW~d2$fZK(qrYfhl$8AG z+Rdl-B|6P`j|%)SnHfmygk|&-p)4SZ4jcE z50{iFe!wyLsPihFuNtA1wyS6x46)HQ>ZhoXp-Be0S?g!w-*ZA>`7)>d`Br015QHjQ z>|9?_F-7s#)7ViW!VdSJ?Fm+0K#HT2>jFd6UC)7NEM7;`|wefeopl#yVWNZ^qjvrP|n|bKJ+-&L(dvz;gR&BM${icSMD`{^V z3z+kyHqr~ArLWm)KY=I{$#c2omxGtl??r&w#P zw+T|QUDL=(@aN`ZU|kwV8tqDCL}E7iI5f0f^f&zNlBx*EjEF>J@FZ+3YtS9Y(rJDt zLHIyBpYhjvm5Ba}FVkgQOp7F{OWL9rX4bzc*yqSXF(jrGw+bVf+(oy?R9H(gO-%ha z3)d;d2zr98RJ1rDF?0`Bt@!({gAmTI4q~kD^0)XwD%%SMYh<4xj))i-QW|kwx*y_d zqrH-HxfBTlcVEuRmXywoEFhv*Q-pi2qi0V;U(E}Tyq4DZc!&K)=50+QMM)&NNlMx? zV|Bf3x+T@>FB z>r@Q2)@xniGXVWYt64X^HRvfL%FCMHNf{iwEjf}|7izeb<0-5;g|W0y!HLgA2R+zN z$!=yMF=ud2QPvNE5*cUhNlmE8YT666Al?V>=3DEzzAlj3E&sC6u0JBg?&*QmlNMQR zdk`o@X32bGPH9w@T#^@mROl2XNX4VSN72N@^g~|u3yvQ^T=ebAO z^f6p+35%(}jjcRCakOInNau))c#Gnl+1LqXdEBDozD9T_a!@|X9Y#OrQrX(08%xpH z>a;(8_xUGP{-JzU;c(Ga&(t+s!5eK))&ODG@0oZ>Hb>^ELTA0jE2k#D$-}z7Kc>cr|J$R1?^{Rn0>?c6XUqH_ zU)J^|S_7c+|MfrMnpye3weJ6T4rh1#R_lL%0`ey5|KFOyn)m>GbqqGXV%*(~5(%@m zNmTs5y_w@~Rab-CX8#|}Lg(!GuU9owSIWHyL{b}_TZmt51Yg>p;*&=g5EnxTPfpyX zM%&*EVA_jrT2Cn;@3JiiH=BdROMblZnb6gpax;{ewyyKGz44q&&r{~$L*|9FPIOW6 z8+=f28(?5AJI~|9=6Ox`uv6$`%K#tqzk;v9fK~sqJU-#G@_afNH@z3<({AH4H@Dby z-ots(v)9%s9QnpoGGP>axeoRQU#u{<4MR`r7o-4@?%o}#uaAB3(Xnkk)Gi~=d%%>4 ztk_?Kn@oVG%9U{EKvZ`9JAl6uIoEgMc*VM~slI^Epxz7UIt{4l#9O!ib-g|QBNy5T zT5D;f^Sjmq=f(*GJ?d!bztvixA|jYF{$=s^0Nz62iAPPEAIu1rWR7+0+KKM~mI>O2 z8k*gMI=Bg)#tkW+1ZN>#zf93 zxQE)Sv}rkJ96Kw8bYges7EhY?7;Ajhe9qR>nxO#|<_khSIT<&I8h-2oVZcbdvTrZu z{_oAp%l)g)HbAC4kIjFJME)bvqNz0fC^*(NlyvqH~d@bb^+e2pdk`!dNd znU|THuJH_1y@TB+pQ70U1Km4=OvV)7t!Gel?FabW;vh<GK z^Kv}_RH~KFda7~LKF4XxfEQpY?3xWt%~WO9@KwA#alLBFA}`6IM2*>Ju5GCPK8|1j z=_ujQdjJ*931^Z!&X9c&z6YZ9*FBe;38c#n0`R#$Km(e$jcdzQ;r1WE@VV3r0Q7Ml zht|3BJPkse1{otdj`J>oYP<%VB7GZ&k6s^Q!u=Y)_%-QMDK=Q-*;+s_hmWp^jmCGM z)_{c_s|5gFQ^y@7^Pkpw+f_%O`3oQL{s6iPNbM}+D!>4D*?o5N*{l5Fp0Lk17 zkot5vpl50*YyTZio8;SR4uQaIZ>sq>*ty$j7Gv7iGrTN4N7Y1^zOTC?ECGKt^46mH zG(b>95D{hZVWPV*(F^z+TWUaaB;1<1<{o1y_rVWWtJp2iI_n)}>)-rJXFt?cx%lYR zLv86ofzWxSvCJbn&7%;lpjB4HOsJS8Q6T*$cF3Rjea-dIYaOx|dp=zOR46Y!+gh3B zhu^WBFBdBiSIeB2d2O>pF5cAdcfyB}s_HE9B)eHacG3cw5~hc#+lBje717CHzl;L& zy&Ww6wt7Vy8)@~!<(|lY-T-UfTHp-;lGc?Ty3h?^lG54>AubH^7zMM?pdw1!XmZUf*~&5)O#+lt_MbD@7MB&nm}_6l=42PM|^_{i^P*+Bd4m@VaY-)Fsd zyb61B6cXdS7W~tEajXN#+K0GCANEJDTqC7&A^Qq3U_WMM6U|1vTEQ>O{|;nLiZ2y@ z-C{Av53|lRfzy*Y_^F}R?uAscwO9z1l{gkZdLi7;7W@3p@G+20JMZjKGp$fG#l18EAy!1r$0|Lc-rUvvXs~K$ zw+?fgJ3tj^{~P@6eMOeRjAj4!zMW+2a!eejN2Oa&P;VL4tyoSQv}?oZ1WF0z&q6m( zS)Y^eZQlu_;?9qEHP1M zAQo{7kh@Sm^X7=Brz!f{ju{c8TI6@WJzk+YWS5Q;2e`knP6C&AXJl)i(9hxV#N`?tHS34QVY43psLPP(NpNPZQ|vP(2}!IHet43H{lHJ~ zJs-(D)N^y#wY=Z7vMC_CGH?DdP}az6%OR#<%aog6rda^s-(%TtB$f_L5fmQ|27dz7 zvnr$*x;>Y-=N22c+zg<}h31hymo^;%7J7vI=DKlQ`h`s2HpP6tKK0p!i^}HeM%MZaUtMvLQy1SbXS3-$* z49!uEVJ+4FU7H+}AU4Apv>jYI04496&j_U*B)1qL<8gmtYoOA86|Y#&P5fLEFnL>1 zvo@UCfGDp0INUWn;5#kKe7aYOYKk_L6e7!flUMn%ZmAT9uRf`mH2G53JJXVPW&9qbc-b20lEL`m)cij6^ ztgTWG2)NJSn1ea7eRHpw{LuuT^1l^eKhG$FOn!DT%8488pEeVw0A^H4GLN-q_!4t|F1ukyluK;@0dMp-x5DJ%@eVIXi&N4=DY{^Qk%QGO^2skwK;Eyz?N?TXqPdE z#>=oZ#fP{Ju~8nezK}(*eY# zm<$1}=vzN@M?wx%pztODm_%t`=Au;BbV zk-1I68TfR+trzwz-@Xf5KQ6($RHz!1w`WCt0mgZQ@pW?f^yA70r0Ou@yq>e=19JJ# z$G2ASYh;KcH|xB;kF6zuHfX*jxGZ(XtMZ4Qhs`3*o+QGJ1Xp!E`-nACw^huqW#D$a ze`jz=J5H?XvL5H*PZ}1m)?Yo?y`)X%DO(goLE(OJKgU7oyg z?&V$<-}+kk)Wadj$o%WTzH)zW;H!=ZJfq&FZKHirxwZO<(uL_kJ?i4&#pQ-*-bgAZdh;~=-gNb`#B+6$u3S3;{yxnufcT#c--1Q z6TDxBBFgGfE#)GE_i9A+LD%*JH{|A=Sm*?aD%(RX5E=r{-bn~%$O>+V$Y3nzTA9}@ zQGC2zJ(d1`Qyp+av+Q?j5Zx|8V)HzCl-Uzjp)`Ao+mSDD7wSqCG-GT}n>+XCNYWml z@7vvS`UzbSj&E(=I}E@X`3;3hq{ZI+v#~v`1E2KPAM@V7iyJJs+LB5Oe4=Chn!OX# zC+_{%oNK+v+*k)+Dy`5GDb~Q1$Q#1FdmX!)88r*h3xe3mas z%E`r6RAg>H>mJ3cJpb(bYhNnvAo+Tx2-uG(UzXS}r$RpV5+aVWPn~6Xtsk0P=g+0M zgGHrv&0`Z3w~+MGnB%ulZ1!n;37*(Kx6_N;_xvjjKR%>Fq#-(tJCYUUj%^pjF$Mznj)L6f09bk&=Up2cj ztWfy->Kd%`?B!Qk7Tz&i0ZMjOU`4)ApM%@oi{g!I!N&C+WOkFU^#i+=!94&3?WA3{ zFhI^DM=w)-+^WA;z{Opdlzg2&ZBOGa{8)bdK`9>9vxSe6`$H~EF;s8W_v0)%a(wI( zC4N@!*`W2=O`fJ}ag zP_*T<#gm-byHt{>3S*i!K&Fnz=`>$fLD0lq6iEfbgN;!mY6L$uZsD-03*((@MNr#P z@r3ak`JZ-0ZfJrD*vHJTDwWMna-g+pL@`cmy51gjBM;ueRERp)Zi_qSz&~7{$p?hN z2itO&%qTmC)c2xgoh6-U#+1i%ua&UHi?$WRTaibeaq}_3o^v;g7J#iCzb9U>crHg0 z@SjtP9a2RWj;O1lV0M&j_tkRzp(yz9gK@)9cy9wwfr;Jgm+A!8#)VQOU2!>K z0n6iud5|di)A=_UIsF6kK#m)nB79Nr8z)$_R6j?_jKb)CRcyC^Dp*n1Z&^VcAYS$l zTLH$cXD8C%OB9dGVuLNe$L^MXnqG{yUG_2VQp$#6x2moz-TN40OTN6nDgySoe0#3X zq55yz5)pV?Lx0+mN+s3imGC;JacI#5WY-GO%adN&K%Mp>7WPJ?KmSC6X?d;k`ON{b zPRvUGL#1^=9Y_8RvvE^#?cdU+N*%y}<+56K9FK=5RHZKk;SaZB)&8}w3+14bfknxh z(s6THI{;FVR^Ucms-^f2BBFGDQK(j^Nq43%BG?XZTYlDwfKo&9S3RoWrcgXE~xo-_!?d`~D> zXO_+)yGnu*OBchtq?SujqLYwnrNKPClm{t)qcKCuUqCE>XZVqreSH{HmrHQK0&iR znAb6$wTEUetjo=jHPw~n61Z`=lOlWeL?!oQ6CXp5JDX34Cs85EhJj^2Ji&tAFk4|g zMc<4oyS1$Cr09)pK`b31QSI8ZSE0tFHHmGgG?%l0Px0Pfj+Y`OgwNjQJjp@*ANJLZrrM=z31+D-(O>+_@mwRZNXZt zx1VHR3nO{nYS;3tgjy7bYA4rpmhQI{R&c;o5In?GRNI@$7WiOE$@DR~kgPRc%}UOg zb*TRZFs8Q$^s+9FTpB@9J`z3o)*xwQ!+7h4{Dw?b42nl5BqAD~u$y9OiDviT9lx?; z&ZH^R!cwvdKydFf|D?uh9p48Pj7<)ch5&kQD^QDbTC{ds*OO-hDO@< zaT`-$zjLW%gS7!xoqxU8svLaGy+!4a(yTZvM)ire#uQgdEBnXPF_qmq*dkRiAUCJB zs-FuNVmT7mA16pf)t`YM&z_ZRs*#4h{>cq{d|p?CHq1(ml&=^TWSA@7g==*2>U&M; z{F!w=`NngLwP-jg@^Y}#KU?cRD>w2f;y_6#`PE3@28Y!;;;;h~5vXi&V@BS8=0)Vg z%fW{`XHS|X{STwhK3~{EnK<9;tbG$VTzU?HdvyIoNctS+2?@9-4sS`x7o$*fg&`^g zowbO=dBZ+#t3Vm77D87N!UJXr0Eb&|Va z69ZXp`AnLc>Hm~q^tEWG)YkkgnGeUr&}I*WT+a(i)d>*X5s?uH*s9h+l;s3$Hzz&s za_J3tn!T4XP1EV~xDBpn+HhZFTj|}lijCJ%$Il?c$ktaqr-uBF23NY;ft;lU4}^1C zOKN_mcJ7y3WG8wz{xo5s+)K4um#Tn@*%>O?r4*w}l(E@|fkl%#U8qt?1Dey*bKr(q z$~_G7eRRR2(8SHbuO#9$*$gg*e%%gUch^a!U}_XR`x5EQtC(r|AYzIl62EEg$BQ6= z0d^#efC^-Zvj(*VrQH&Ny+D#t@!TTacmGCe&80{JmUfRB506<(cq2hCo+)Kj%Cp{Pjc5dF{-zfdKAta0gBP9sBkD=acx4YFlND?&>KAJ2b9wHZ#1}=@pclPVU z>Gg7Ew=_Alv;8T9Ike4eJox=;+R;pv31kWoLpB5t6C#0U}Qf2 z)Q_@r3~?+6as+&`B$rz`=r&GU4O-H2itO?Dlm!L2a>mI%dl>4g^4GO>ui2xu$8p<* zw4SQ2(VmvFvg>t@)hNQgz4u2kPn2h`m2x@(`Xi-Qep>IC>TTC;lh!I+ z#L3U2v1uA?BARr~)J=~w$k?(_ClV)0yEPhJrv{IYNqxbw*0(4vGlp4asTuP=))7f< z&Q(1L6{BFA^tLUc8C*%?4C0fr`Md#(ydGHNDsXgjen$q$r{p!}^P>ps`JcJG$1KdX zNuVX47N8rZxc-PI45Wj)eLz?)6&%(Jndr7kHhjg@1qxqsbN(*Y$uK`Fyih-#)9mjO zQ-XX{S|Gcc_%n|WKR(O#B?fYZhFk-c9gm35PKoj{VHZs(+(ZrhN!v0PW;XDZMV@`? zq&-BsHvpA5nTAk6TSg*>hx^h53Q+zT@4WStAQJImJ<&^)2jtoZ_3w@35|cO1rt5(* zhebe`gAt*Nv^wq~JBVNBKG1JlqKE9s1MnvE^t1lJ-8`k3pD1S~6FQ&u$(#%8c>=$R zniI}HPGQCN4a|PS6_F=NWM4CZGqh(-EjN%c`rYQ5$u2(ziaPi7eTJV9Z`_M_(q)3g ziTkW@`YnfV7NM3OF>T%EcxN@2fJ|<8lk}}iYCQRP`P1s8uGk*c``&<#;ZjS^QA$Ez z99b%|Rwg3-Q_U`&Y~4WY;@wI9DTc2-iMk3~r6|D&PisHE2LXHNDxDO;pLcGT0r7xq z$vQ5l%SC}D0P&LqrkZmb9AuesqeKOK9HM4TyBXWQNoSsi+}ui${9Yhf(VD0-ON#1a z-R9pGTUeo)w6TKk*_^ho58R4;FsYxwdKr_Im~%3&xR=aqp1Zf#bKg&Waf;@(eqCoC zWTshAGE*^0Nv15v|A&5hIdr=Q+BYYJ-BMhykKZ=msISL*7-vtyIB79<9-}D;SSIDGcugVLx~5Nprb8 zKjf}@3h|fd++x$UJGS6twB>ykSKZ{DSt@jM$QcdYM&%|%W=_lMUEN|4kU z&L)zg&2&Ga3*{<0B6C?auZ0VI>pz0m1lL(7{@B*iXb9P#pkUc&RKVtVyq1&Tzi)St z_4Xd_)mL3{h%tW~_czjqgT@?#3h%BP7}w-`|HF$SNi4G(`^Qsr8tvsDPN1Tg6oMVg z;rT>DQ4a9F?9cg_GE_UUf@e>oNk+&sj$`7+X6j zHaHm90wXfj9^NWF8R5SeI^h*Q|EWt?JmrP7G<*>tMx}KN1&{-rvK!%l=;0RbM8Q zNi1kwbh{suQlYB$+~Wf*80H_4V&tJC2ezQj7HbMNf-8#=s;)Uipe*!mS1h7TB=Bx0 zb8t$R{I$+$_W|~4qA0iL*~e_)`27$2tK#LXmzuTooL&2fGH@evaJqOTCpz=#L5#%8 zG-i14ht!|2qpK911vmM3FV0sN<@w3iAJ+IiTb0K2T)ePL->`qJKP;SPe|R@kXKlcz zuP4S`deHcSxcW%|#y|-ITHMZGyzo}6IO%?u>Gq8RIj*p+!ii}LS|X4~&@!#*kF(Yt zRkoH(fiv5;NKF%oIa;LY9fzr~Le-$Vw4Lg*Z~~=>`aHQck|F zuoDuT>Uk+|$GFwneiGBym6YQu`~b!l<6XcBzdjIN@fA3{TWt9U=kOXk7;ZdObIdAn z+T5vfZL;f!vO;5G*H*Z3Q!BH0#LxUlmCpAuTO(9p%(stCih0i1FN(F~@R|BWvwVfd z=0G=t=xr064(nHv_AgYPF);FJqmE-M{b!SLyf=JpPqqq z?;NK#ON-%EkK5kKdqMH%?`%N9#HzmpibhwPKAJC)X8%X?71fbVt3|5iBU zMo!W&t59haXa3n6ngP^DE^WxE>>92+gl3@%(sSE`i65Ko zrK85w#*VXWv_(<6if~odH*Sqp{^tSNZv-H!toAH2-vz5$4WKGquM&u1B(}5;VE_uK zBT<2fQCO4AG#c~#THPupi8%eoQ#to?h5IQXt-@)t?tzIv$gd51d~VF>W@nbD40VGU zv_>&`sW>FChlb5ZgcuIyRnZ%5tj7y?G zxYznVHnLC|qX|wg;$JXDDukoB^&4+*faQ4kH?0WM0d#Bm!d6t^Le2hci0I7wz|DdS z%Q3wZo`-2vIL74=bl$10U=Dt3y)`KR4+ZQgSnUrlCQ)E5>MhUyIj}LTs(D(0m6iy% zN{F;ioP?8as{Z)Zfot7T*;DD~_L=2BbsR413Dp!(WyYy0j;qNkJk8cr_->A*xx?zO zIxv%hFXocc?Wk-U6&6AZEmI9~xuK&Gwz8X2@Oa3G&*tZ+b4g#4577!asO`C@XItX% zh)!-G%`h!Uc_}T9_Q!)Jsxvi>M2FojNkge1&1}OisFFu~Ly95m=mJ!!{2hfI4}W2f zqz3A&Do62L!@FpFdo4fgSV|9m8*#5x{F^p7m2)od3Z}v;XiPY3|6Da=Zb<9ID*C(F zYjHnEo8u}EgDZWbuF2DF`D>81mkf>d?P#a7vYH3~-O5R_y8RQX;^xmN0_)nie;iby zw(08M$Eca++6P{3Ug2KrsaaE5rNA8bZ7oAA#+J;_v-V}KLGu{CV)_DKhujVf?zwZJ zY;kLml(v8awN|it`^-PkTQNg+73oGR{Rk2w-lMU6@4!TC9XLV86?0%j*$gM`nT8|p z_c%RHt|`qrxYZZ*B;Yz~oaw>`DjBSw{xfpDV7k4+*$9Rw5D_tqHh){49^%jQ8f}+Fp zhQ%G7_J2CrpIBiclNwPwRH!_q?Rb5{)A4}yH%c6I%*(;;L{O0FMMtT~k#eUmtNf?S zCs1sJwT>Sw046*{G`l0WCffb1V^eV9xpz>URcS%fH2BA(%N9FtP)0o?%Inzc8Dh+8 zEoOKu*e$^}bog+dC!tIK6wQmQ)WQf(*IwP2;X^*6wZN#!8yrHn3)!$8-uVT-TwXQU zHVw|BbKPlJN9g!Ln6gh%nNFGTWfPVx>o&u2Rnyqju{mdIjK2>b^0xi*@RQ7!0ILqZ>oc09FYREd+yFa*e zw&FfuYrut|^qk;_F_wJD5OaCnwglN~BLFdHy)@oTk#IN{O7_+l2WsD&+@{!A=(9YCcRChgjbA@ zNrTx8m^G5?`(}?tG=}z7xFV%PG;jZY$fZX@2!JV1clLndesi~KTs^^Oo>A#HY!BkH zA7nz`jVyilgLdIgcA&WF;EvQrqEI1fM`1bstojn@c4gq+vX)Q)y&#*-Z$5srj7qOs zt}|{EyaB9vi6vne&vw}Zv@X_80U+0G4FPUyI>W)(iV)a8w=}B!jWRB?_ z&T1nSDjSPhjh5(H$TJyyw^+hcnpEoqO4=iS*U7%YB*tVqojDpKS6J%2xoe9=IlWY_ zl?t=#d7o*L-BZ4_EyUp#$uz8#pVJ$n68?CEkt!%%XF@ATm?$LN(WF3@?Qih|Ug{w8 zY|qBfvIONzeqo5b_$OC-*hj_D0Ev;{&B+ilju}PfB1_r#^6Ermu8^>BnTvzoHt=~P z*eOV7_UWFdGTrC%5AAw1evVXF3qGG5=TY??MiCQaX!oN8ZUXvY=N^2i8XuebY?$Pm zkjgCMv3H`d=@`UrDtIgQJfhSZ!4m}>ljv$9m!JKZw(X}^XkI_;#EBS!cTky?SEiAZWoHh3g^wxs zIp`q+GkhvhEaJ5N!IbzuK-eU)@=xL-oTiLlNp>A7t96P_hAyNIP<(<@atnS+%~IN} z?KN8*%!^ZN6=q0p)yYTzi7x%(6<9BOHmffi&34Cl*m0nf{Xgq8t{IW;ca^Vdxq_2} zk*`(h7j zk?(QHerPxNeR{1oju5PThjEREy4@@|?9Ygfp=)5Wy7nwJq~_-T#T#ldXv9#{mr)QxpY6L$sBT=909TCf|i3i0A{ zoz{>67dPVGWiL6d>7$a;P&b-OE#P{V7;L?-Ddcxeniz|4azK$F&d-9>ff+~9yy=QU zReFn?o?Ef*fLj=kx$P0cQq|Y^LF!(KUn}6xzc$Qz2AnN+-#*U^5z~8c@p+tvo>T zdwULUaW2Sge)o=!-O&^We_A=OVvN|R{kknHj3`IdKJ|IWhmryf#+!d`{ZYAuTc%I^ zIC2i_3aSx3blrXu?DhneER}82xvlNGSq7~Lt|jv_AwDnZRH-)3@u^YVuHxMdf3Qg) zcSi{l-3!3gOjFs%H-dC%)}|vp8*-rU!_eLgr$ALu%s5An({557`&RY~J?c)$J+bP& z+pMS}&ap3!>9LdEhLT7p=iCx!h4ebXscf_i(omu(%=x|ofEg>w!R8>XH4dX~xB|{b zNj+>44Hfp2ULRxsx)aEL=E&eBu%KF7w-2Omr4yrxVPz{^RAL~QTby(@<7VI4>J45u z=_<_}!6meZ_KqqjE&xgXhTCRMI4G_IP5NmDG_ZBVS~60vNUxm@vc(J`GKY95WHPtA zXIHg&`vW~zq=+(gZJGHEWZsT{a9B2IYk1!@A~v|EuQ2&ID1kbAR524-HgYXDR{Ce= zYeJ>WU`R2wCA5SVrQ+t+3QAz8BTRolJ2)AMO<*P})a0*2m2pHB@qFF9=|SSSYDg4S zt9Dj(I{~6LW1yS7Uk+AoG@q;>|DH-0LdqBEtx@{6VXOCsgH#ZiyV?f9`_S(bQZP&Z zZ9Np1u&lZ`T^z1G!ec?8mUx#i{};bZX&RTg+~!u*Wr5=R7-^YrdS?b52F4sC_MzbE1Dij?*_&yY-sPZlUxG_kYF*>yD6u)qq2K zcY{FZD`wY?SV}R%=aQSVBxbI1OKFG8m;kX$;ou-9o!dDpng@{=ikE3zY@J`x5Lo^D zS#vt)*T-3bn1pWoUUzafdd*O&m3pC#yhskXWRc(08%11A`-OudkCp1%4;!h*vm2oF z9WzSosklGWw9T6=hF<4dhL+yp1wffLcV`li*fdY}Wf+#n2BN}9hbM}yk(v?uz5=H$ zZt9ZuPs7@OOx||&!#1SrvaamP>w}&Smg6>haGn&jX0oAS--M}LQZ*)H<7RIP2Dg`D zM;Y45O`Uj;MXlakf11j8ef^E zBR_QQRpa>0;zg+lzu_J6&?mx6`Yf}YnOSx9gh6fj)z}-BHZ~+-Ux?8^9~Tk*VVZqc zu5|uRXX0uTFi(imrg=B?huSa%s;d`uO+xkI*Z6l2$ocOGmB_B5Q@fnQv58rMklYhi z_6Zs+XImMfoKtSIjfh4IQOFD*cJ+O@p7yb`3|Y93YTm~}qV0`}{rpS!oFne3vuqaL zQzgi^pN$|2GQI=a36+^luP$dcAJ!J&oO^DE$~v*G39Q)z61U#IK-OHxxJwRntiA8@ ze8&kWJ7!EOolkDy#pXR0ZrRb@D!;hzO4kyJ^>dk!fW9q1g1Jvg`;$kj{q_XiinQH=mw1E+{+u++m$&_$H-c_?~)01N44A2caL0UkL!R*K2iQ+Ar7VL ziGHA}`Z#Qt{+4#yyrzTH`e?*by-sE2ZS*~oGWo4*NgmmYr+EtvoIyuYM5?)LW46x` zDz>RsNpt>Fe)OB!YUEzylHRFftg0fRj`W)iihQ+k?8y;mIwPBqvbFdbkFLOm<&<`7 z$}?kYhJ;#8S-!iQH83SUUes%j^5A+uZ02PwiFzG1gyWxK!}TK>!)8j{#(=eP%f+ml zY-!ETYQ3haJ%@gPLvWA3AwMKPHck<{$wYoz^SVyhxIkQ-JoDM+u^gfqKI(X> z14SkO<&B7vYtH>N&KQiRA)g}bkz3>Pb>&#oXA5gxj)Q24ng*=ClP$|o=4@5lUwe0c zW3#|Tl3wmb%&*NJe_~qu@g|8;<9X*T36_!PfyKr|(}ov!v<;M(4dPy2w>8A+6C`{0 z)htD99m}V$)*U=08amiPcM_3qeU{Qnps?hn!s#gS6(BB=2$y4~oBYALX=<`fR&>&W znCS|?Eefkh%Y5@iy=4>=XVcRTlCKn!a(_PAe1;q-oH;O*_)`vMw7$sFLd#tU$(_k{ z9mye)ykj?%Bc&$1(YewlNUM32fk1{bXuWq}bpoBpudwy_FHXh@o0uzeMtNbt;dR=q zY`tCyE!=-y5{f`O?%bYwHh9f1TOlFs^08E)IBn0zMFTV-(W_4FK}b(6M=U8$WOek= zQtDn~D;YxgE?gFaaqZezURt*NlR!X{I)}P(Lo<7D?L4Uh0X_8zZcfdht`{?44F~k( zHZk%kInV4WEdCzp{?8lohF_!soe1n61J?uExUh{Sk9=wOOsTl9PF8Q#5cEu*ccO}j zO#gtEu0IM36IfC;CEesY3&aLr&ol>t_SD|Oo_41+NW)pz58OTIGc&azcj%s)b7d!J z9dYSW+?%9b%%GH5`ox>k5N*wqa|9ph$g~5+q;jT{-&DP$>B{t^S-t?dWrQT*B+h@V zxMoK99IhdE{Qh$+SK-7PP4DlXo@!&(wP#k|>caO+!Lt?!LEeCw8uLl!GQ;m*Wb4A-!O|Acs;kRy~bYmAntx^P`@F;@93r_ue( zU|tMH7P;d(AB#1ZOL#~96<`m+YvFCf9lyiQEqsYXG^A;^9Ca>Rymy+pU7`Z9@?tCk zNBK(~(Dd?M$wz~@%Af~A_nfpagVR>Ckad)807U1Y0&i;ZHPXw z>`|0%_iNuVD3#JSS0b=0l{6=5jnJnT5YVM6$>+evGAwUC2j=G=kGrOPG6f!=1YJ6n zXIGImy!V&6M+*+`EjAB@&D_06#+DjeZWkkJDTwChYxdxp=e4v)Xsv+P#E|0HO_8{S z&cUu--P!!knC)4?k!H6QFJQ@=Oe0jUxWW*xeO2y)(g&}X6BwO*UYN3PWa#{Ya6OEI zKM#JDyjkJ7^-fV72W9X9v@O(fON=IDoJe-ZatzT|DMSqO-zIN@IAD5YXX^ZoITy=AVnUNaC#pS-G(NblPF za;w2yQJH12-{ z=dV!QWDY`$B*^gx-P~-j%*_%uNw{vIzaw6&uWXqpY+II-jN|bedFa8bRL7;`v7lLd z4@*r%>sZMCq0msLhRqMK_)J*>8k~joH1J{v4ha}Rb^wnHB+b~BYpx;s!DKA-aezov zi>D=Ymy+F5o6~{>SK%F}X=Z049<1rdGqCEcGQHep&PmPu1mniZ$T_ z)=~BTw-4m;@YD^ZtFr~XRcGwW(fo>9*Oe$K0BI|^*kzxKb!Myga&GJ+M(&vMiwKMZ zV!ie~ASnhS*FRYmT4hX$%oR~1AG~k-U)S9nX%ILkUU;+jTCX3!#4>|W_w_CU-wgP! z>E)5UX4w8licpCAkuS}9FY%1#E8^NK(ytWZ^t<8w8D!AnZWiX?Y*!&`b^pP3`lT9R z4I!+xD_!IE}+-%NPzNDYqZ{V6jzkJ3ZW`QdT9tJo4A# zpp?_wrWtyqlgmB?#$vhJwaA-F#L{LN-CMV3f9Uvccd#gvkz%%n;bC!Q$y^$;QnVNa z+<>EBc0n@Qv};0JdeT%NsANF9(I$*(-SCtBMg(1@`hHgFmg_MC-{uXF#Ujt_s>iIu ziG}H;#@mKg>hnsENy96h(W-AL=UdlZ=s2Io&)P{wLDJtrKRah$mBfgWHWg%N5Mh^y zPE1*(#uo8Qv47Y09+Hy%wBu(UxK7^)`?Bibt}U!QaKy&1{pVA!(HTJ!NO9~Fkt3Df zmtN|%Ek8n?L51PvBXgwQDwpvB35$%VQ_L?jC%4~|8XXflEc6-GWypL*n?j^~LYOy* zfLEfpu;+_WzsDz;Q@Bg^ujmK=+z?vQvO2Pz)M(U5B#n}WsJ%+vn`rp+;^CMHtw8`W z`xIHpN6Wd<)7MfYOY+~o(w-%RbQ*nM{?twt4FwAp0)=+D!e@$#19ESm3LrjM(vl zG^V?bqbj!yIa|9=FeG-(Eujw@h~Wq>>5-$c#CWdn!Zq!69mQJ(oHz&GPem0SOtV4j z%r`Bm)SaV}=`CzRK6+E~deMJ&j_ee_!aHGp7~y`H(&a6w^HmeLZ6(`nVfDKEQ$;2P z|7TwgY(^DetR;O#;lBoh|FJO2pb7#17GD*PJF^SAO4JeHBy@=Or5vUxUA>Y?AG`{^ z#MFldmS4B2bu`sY2*3?(Fa!!HT%M@E4Hk1#U($BZzVIMSAqx0&=+#1ZdXNz$mYTuE z2{9oIL|IHkU=-F1cmp?Awhx$XdOU&_P$FW=ZX-N<`U|9`qw!-#%Yycm zte{C(w}vShn889yGgvAOTPL_(?7A{rHGpAN+VFuG=W4)}Tep`Ly(yFgS`xBe&NZe> znpSuI)W012cFgzR>gbpfopHg#>bbO4Pi}0XN>!J@2oT1 zzkQ)Q1T;78qsG6&-lN*y-ZozNc~0| zRo}m|_Zsl06qH@hG=&t4Hc$)%y>3fL_ufP(72!ZE3e&U%nUU+THfVu9JZHWF$DEff zSeIKvtis;`jwnNR6@c0p4=Y4JJ{oisw|yBB_bgEcmNpN|kD4z#`=)1iPYsGHp(1BR zEbE!ag1e~78Rm_*xL9za5MA2v8f8!kJsUoGLF0jhmAM1! z*60=V~gZ>oz7GM{Z9^J!k)r{Y$<-A0WLWqjOqpVWn*`YV9DlD`g~ZkAmWqj(0&E=>v~=2Lvr{kuaR57yl-3$ zzd&i|`Pxvh<&7EgL7`G@wFUMWoe_%xd#y8kzH7qi_Uw9E_M1ld}uctWlFuJY&w`@hxYPiZe%jz@c z5E}n%cKg$;Tv?{wU;VYrp$4SXW0uD~R~?&b;@gXW88&#%+~&i$IR&h?UlA!v=l-h+ ze{!}H25_)|k3F&kg5CF2#6%+%1&}age6g4RE8wH51rwiqrq#+Dst{;7ZL&v5stIO6s!3z2#Mp3b^>mMl%l_9|< zgB`3CNZmD=x_t6{K{s$Ci>cP_e?W6%^k-?=B7;5x@kQOZ=%IfoRf+secJEy_(-z7 zZem*J8i`rIjjTDB;JW{(z4!cTs%!gpAtdxF2q-OpN(U9CSE*70QboFeh=}wK0fGXG z(h-$jLkFqS!GaX&J@gKthtNWMSG?})ex7%XJ@!Aaw<8}i_`txGxmM;p&+~U2Nl{~U z&^g*SpaXo$wJKQo(w~hys|-YY8bbpr_u?F)l1_w>bDgy_x=$@$H2^U|DEx$?hOWd6 zE^T8;z`3n$`+f2)(g4}5K3+$z4k9MyO3SCk5mxh*5|xF!y5<<^@t?p9koYVL)>&F0 zLJ9I7zkD=Daak(C*GI_lu}^Fz(ohaU2nC7eVe~-uXd$-NmSuSj`$yLUWJ2t>Mc^t@ zgUQQ%r6g5}1UKzyH*=^_mgT*}!BT{lh$9V)r9tP#JHxVR@kmEyO@7&bz!zV98HT{4y53+kRw^pOF&$#w8r%qXmD8h`%_9*(4k%xNOgV0Ze<*zp3BkzWufiI=L7fumJUD9(Q4m( z{A^l$#CJia;bR;Vz~6Xv;xfu#ZbLpb>AlO}#Ao;ENtqjJRq{8)S3pVfM_P$hqAz%o zc6&8ON63Z)EoVq<^C`(Ww?s~z)XrSP)UVxeu2|#^X)=PZ0!692!&Mo}rpZ~MIi}Dc ztYSH(_G(G@=cI<`Gxj`e`k<+&9>>uO7G&RF*PRW=>G<~LS^DMBdGF2{?C2+uc$O~3 z7YZ(FE~iiNVfJa~^HFC{yyA&F6HG1q{Yec`OtR)_H-!n_w(6M_fR)ao>5L{FZm<4yhq#PuPwFY*dm6v~(0Vm_9DwG*8-9|D!E{`V zv7ZpiTf}N%Y3{v}PE7X^F63*vEjrp{%vAq2qsKMonD%5(A_F!T=-TIIFs9w|A5|~E zyun%E5b{g+M_SsiVI`^TQm=1gZ3ZnN=W288Egl;1 z?e*Hw%hl?jO!(L>TPRgfUuf0o-Limp$)DfU{}u%)jW{v#+@Zka+w-@^`~hEg5Ird+ zUaHK_m=#(7Gb{!Tg$FaUzbO@p^9}ja(dOQD`<65FsT1Fp%(BH$H5o~-MG!fA8hPm>YAx!0RP7ARACg;;@0oIwXF}$fhr)+4>s;J z1^`=w57j_Vv!SejgOqjnm7%;?Yn}yb%7>rfY+jE~lm9??Ug$XZc1H{mr(a-^-j_$0 z7qebn92UA6X@vAmYA^irrKKp2kMX|DZIiQd>E%yQ9NCdCnTFI`EwY zdq_rjQp#|>R&q=lH%N%Lyz#CoNAiZsl3@Gq|SEY0X!`{8R%TVZ7Cn5Z@d_x)j++@H(RAf$ag&`z1>|-ROb}e%{I9 zyG{(lI@6KIhVsjYl8}fGXz2{?Q^imFP#U))w$Hb9R&GYuBXNDhVtIX$*cOyhDiD&X zkv1>S7#k&z6F_uqIE0S@UQ5NYtpjc)hf@RKW4PPdH#%Uxaeit zp}sx%vz5GJP*(~3ydGPIvuFZHMaU|hOaF_}+GbD923v#vGFng6-drfnHah?dsLA*- zdpDhxsZ$4(?Aa&<=K`r?ffd-OH)0Pg0 zJ0sg&64e%Ip3bgzO1^>*ELSuGz4(WO zYOf{KAXen{!#C#Fu-E3sO_N^7$~&?^08giGOxB}qgyhw3mt$qxET8L%Ehk-ZAne(P8g{A`H#Oy~f?AAYRv!U+x7qskD4fo}8j)%+$TNo;J?m z{h~9tOxd+B8H&y)E^+Brzw}{}xsVOAF6_Gne2rD8EK`O6xoOz9JuoxUJM&^e-z01| zRpGn2*F&)ZAqXzy9#f{m11|C$q(K2BCBJmcVzd_=k%~ zS@PM6mqwk3+)$mwiYs*mHwg4XieDtne=5d(a_f7d(9g_;B;`HG8Pm_$!j;h$Dohq0 z$*hUkD!>zGAF~`=uw#n+J08G2{4^#99FF?pIhwLaK$Em4UFsjPTI?2vrbm3;!+y&sn77ypx<@<2OtgM=3Sx?4`hde=jd3~f74iPUO)Yge`4|-`KKn298Nm*C94pp zku2bKA^sl>Q3bkZ!wFE~laDPMm1)0nwdP;8R&EQyb+2nW3;`*m&*TmQNj1i`CT+?# zyOcpqC?)BhMX`@W4o99(=`>0sml{z0Q9gzjJFD zQcYf%7K~-tFO{8@4)U3C!aG>*HY)xMo4iPtz!^6s{#fo|#jmT$FjMF1ao}(;`D{`J zNi1B=_L1BnKDnm<;v0WkK(ya*_LK9%O~dGULr?4|U2;fuxD*iDJaTg0XYR6`y^>+i z{Z&i?aE?#1sJ-N>Dfl*n;k2Uh4)fmEc$(}5Stsi+YVDRwFrS;Tm)rl`&bePNUxb!0a+lPKJ+W2_!atTjeh4~M!JhF{A&T`yHpJVG6@ZQLHNe@4R zvy7Vl*}TJ8U~6aPL@U6GU%!*ADWYYi4B#79lp)+Pf41)^2JfrqT{L+OpR};*IWV?? zSY0mN8#nx^pQgIhM*MM$jxPb93voGubMm8GnGP1KP?@VwmRh-AVr&Vg8NJFgT>n(1Rl5SeB);A$W;G%2P&MSBTma$9x=7oP$V*z8)x1P z2~_mq)KxgC9!NMGR!u7znvyA+?N3+fJ57agdvv%yzT`X;mY1T!U{d;gopLS7s|;_Y z0}AQS9hSPx4?2@Zk}`c9@BGVjJwH8DD{=6x3=D_>8=dT1n)}V?O~+Vw`Hy zIKnr>8y6cq&P9}K8j=rOyMJDjRA+q$b5bAUbzHrc^%vZav-R7OynzUh1w%Ayb#}xzIMrc`*utUq))5UQ8Hg_ z*ek%pQ-J*yUDbVy%odY+kBmNuM#~Ze^M$Jj0VJ3b0|j$G7!zUBnn#JXq_TvNU8PvI zwMajm<)kGguV2w3g{N+(I@TJO1MD7(9ED=7*<1A+r9u%e(Wa7SGad66v!_k7|BNrW zYKav*3I7ycy(taVEgj{nB7u*y$m1JUMn-lV{*~{F(8t3-Q1Q!YAbCB#An&>L=^$O% z4aSsJDsIp|n1x)5L1Ima13^_SN=;RJ0U(sJoASmtq7KG5$9bnJdVYFz`l*-n{P|w4 z1)_KMpPUq{lR0jqdDRx#wVwE_jfhn6=-DY6fo-*N(O*RQ?%@ium3I! z?wIp%e;nv`2#F=V#{d=;;W#3Po%Dyx4I_MG9;-)km4?kNNxhhTB?}Y$u5ywRc5~c? zomLc2Y%;ablq_*>dgN(rp=f4%Q{9POVi^@35KE=NsQ!*TVIw=Dl&e8T$tY5tz$pMg z5mk!`KK0cyf+#}}8S0?B=^BM4V2o$ibPo3z0l{64hsufutmc`?!T0oyr6ri?zM!H@ zf$`1!W?*j(2$=g+Fbcc<-O+Gt_So8TmV(F)w!12IiSaTrrncS8Yrp+xPiGDbqNA6& z%@6v+X-$P^$&B`;C<4`+HABR(?{&@tgx)suSBnNcnu9qvJ)LQ2R7o6x~T5)e` zgz}MoM@*W}LIhvzDa$Lb4HTg%Y`a$u9=-?+)n*XnwUlbxMt7sdiMK>jeRJ;n3h3u@ zJJ~&@>4J5;Ke{w{b>3k}L4#vX)ob~m6X8BVcsNBq0sDnubBIoD$K~Sw$4le0*B;ND zw0=o4T>iKelpk>qO!s4)iX>UAH3rJ2ko86Xgv>LOzWEn<#EK{vzLrAy^Z>z@55X1R zs^m@PNpZ=PtC^@DdHoB5YXlZqBBB9riqP6d9!klAFDSNdM(6ITvy3$7ir@&I(=Vr8 zABp>-HnmK_}^Mg=LrN*|=V*@4?UvTtGUX5|=_%58}m3-wxkB zZS(!BaOoql#0@50_kK(9bN#?hTgT}J~&Lvi7(%Y-#VR!6FN;fDxo=BLtD1zzrqtuoA(SII1B{lE^ zs4)|mdD=er##GF$AVssQYq!#A;Jcwcb!lpDTw_Z%W9+}ZD-raT3dDb*JR$b`U7O9_ zA}x)O3bNPA&?&T0?F(Z-FZ$d@*Ub2L3SwH88Ow3IL!92kr z(qEr6JP{-r5|DY>#i;`>QfO3po#(ZWrVO#aPz@kW8{w%MVO}dIkiiM&$6Wxa<8wp# zLOFx7d=;x`LONc8Q30-{@Bp6EI}d2F=1W(yG~gW`4r@#t$E_qEm9dQ0B^(?6`yTeP z!PsZ}jfdIl;;$$1nR9dB<-`v@rsP!+^J{S!^xL(Oos~{|QrvGbMow*>4$aG5llttI zv4_#P{9xyX#@`WQ%bfmpduMvx!$~JHr>6PvB}}m!nG`DpjQZb$AE_AeG2vNskGTW^ z?dH$)Uy_0g&yLH)=#O@;+TNJ6{burBzEi||&-V6>`iLNKY-Z2+&t# zw^ezi(2UtVlYxhuRj0bu1@SZ8kPlF*u3pWPYq*lnF^v=*$-DZAe7`HB5f#tW=U4@(FpRKN6_np(3Ch3}>Az%%?c%#OjM%<2x4aZ@{1Nb@M7ACcz;%mdI z4l=wj8uSlNwh7Al)P+%QSJG|{JL#q0d7_Mr2rl*(b3X6fd)d%Mp2FQ*t7 zqU}>xgNdl0m!;1*)lQ>O)crRPv-_BO5fv3p=gk@EbuZZPYj&LZKi$6?NeMfUmPDL| zw8JVZrq#h7WRLHQd0k<%uL8-RZAphH;h7kyCe{~G)lgOkNe8O0x(aeow8S^o9&)Xv zKW^Vs(T37XW*s+>8DuZ%f9?ab+Dq3F#4EA4TIYcW6D>?qq636Z7<+VrU|sGhNVJ5^ zHm=Igaxp@IJ>#+AYVauW1=`=SnhSQxgpi7Q-zy@{N~yISGCqJ zb7RtQMOFeK)?M#d$1ZbjMnng6DAjioww<8)GeowIzoeQs>A9j>TR<-v=wsXO8>lMpIGBTjXWh` zGZs%r&q*J}m7Gp*=$ksuf~Gy3r?V~9<1HMs2fu10Pn`+-B(TN?X6W=kYRg}WKJ#qx z!1{0TDbxpjK?{ECt6a*9Vls14!~@>S4s6b=(HfTK=)x4-y=1+Rw}VN2Swn;eeTCUA z8~KI@xvqEg@|NO(Pb6%rff2EO94&UrcS0>KH)9EI;8psxs2O zT=1^qObZsVBuutxEtr2Dy<6zGRDwW;p;w`Gxp59ABX~h5pGS4BGZUF1 z7bwga{zmhO<6ILqgglDdy-oXD%)(&10yV0ag(CnYtbF;EGbpVz#DD!}DH%-xl}2f0 zhSTno>W3{H36eDE%;*JQmzvw7LBUf8mS$0BF4f1}fx0G1-|Z3gbsg>neWeias0l?| z&=Z;gxg`br5$8v{%d}Um3R7_h*_t?Qh(sTV4ZYEH5kWD$?V;#*ekjlpeNi6Xyt98i zm;!O&i?14gcgMuQQZQ{`yw&T-^BJAEh|_^sTED)X0na&@M!x%R^Km2mKc#9h5~R*; z@(03`+Xcy{#m8<7BO@0HK?HBcZA1rVs&yr*2JIWG4&yqy78_7|nDw!^d3 zZ4cy%t^0F2AtjgQ)W=5130G#wKqDrx?Bz)#z-oMX(hQuuw*QukWztUm*w+Lnliwq} zG4x5))MW>l0^ivNzJ^STkX9-BI(e>&?&}Um9dURKtS$E@n`T5wSrHz9c|jnH&k;xZFlzsJ?r$OMCqI0A5>x|AEVlj~pR^Go0f8@_s0+kI&prf{9RXp{vfb}kRh%qLF|-a>F%VzZb$`}{^LzsLck zyUr5$%Q&e)Z=8Bqpa0B$J)Q=#Wl{6a;!tN}h-ah@w~Q|`iJP{MI&eEY)W2QjHROCC z;6{HVQdwUWoM~lrNE4IjF~mjSmgJP}y;&%?dU?lZ^eKrsC)0T%>&U9Mv!8zD>Qa8Q zN8&V8&8|ZCd*19}O|277xXjmT@_?IZIx zlEW)s?~odNXKKVluLw1L?oC033lK?@M|TtCYq!Cb99UF?H3le|EMCS_+3a^Y(+Eb$$D8RJg$7pqk6er94w+J zZTrhU{i=-WGsJ5bjqM~ki>#Icf6^1uk3amMD+MsYtm$}`*!>=*RQL%ltQZk_e1t6! zyWT;+WvML~T$N$#eLqUS1VrF4s(-uRvfsSyga+woIW0m)QN4`-`gkXw&`d!b0v zu5j5oJ*hiPvEr((i=19p!zBmoYTz2+(pRyBB4yD6*@wgJEP#{tnNKPCi*1f4WBQ|~jr}BUKGyx*0qT<%iTFy>#~Wn% zJHf%iO5k}C`Evu2#+JQy!vG)P@KS!~+u_g!i}5Af-=jA>t00dR86#0|q?Kq{Q3^S= z@z$pWD>vcX9|eq!VfmO#8Pebfahz8r&eU1@$TbStngb=~HC#wmH5Fp~A?IXwKa(p{ zhL^Ca!tx?TBN+sLQ-@d^=Mc!-ElF@ciqZH>w8K}o661m60;{p7p4!R7F+*f2=V=Uy z?OcwqFR9pktNV;>U){>Ckv~ol*-X=0=vSAN-OHCK_uIT~!&wo))a(ADQ==vW>B~8_ zwW&pIHtzS9n1{p4tI$7hOXW~a3v|FHJA(cwCaI+wgvktzc1ZTc-Nzs!SlZOk85dZM zx@jf(=|%s8;TqD6xFkQ$wJPMS^ zS^~Act(U4d3~pxx>)jZl1)^?5sqwEyl+sFH@Hj2WEpr{IoI9|0KsePcb4yZlNa(L`S2V{FsajqI!6PX=Zu|?BWY0|QmY;>T-(a-NzlfjwCZ!2Y;$Q&dMiB? z{mE85zuTm~ZdD!OdVdcqbo5*_sE_CxR~Uq6FtX5bWXQWKySVFhv5_$5miS$3@p19M zlt7ly!@Fs~C33tI(R4>ukl+5_Dn*-WNS__|waa1zugg8LL7`ts^8hCK?^!#U!UpFZS@I^U#>Ar?KM*s0(0OUQlmYRE-! zdg=DB$vIJ%E8Fv@^CFNpd!MJH9mfF-@0UwtT3|0(dRw@WI4dQGf$2Qw@eoGO`z-Wu zEG)oMhMZ&amvcM!LyEB+*=nkmMl`=HyajrHi|4}v)RoXj{$!45SknToFXCH3UkChF z9yX_pt&?BNr%Y$17bwf*m~hM#FfH}6G1~2eYpAYM!B(Pi@;ubF_&B+F+aY-Y?jq7f z<->XwUh_z#qE+Nu`qc0T3;#(17-4PZD>3lQhQ-l~v#di_x7UMu;!g7z`9MLFSj%+L zETwPYwmi)Kd$*W|`sY7T#WUezIW(mUua@<<)L^-x!oGz9F$ax@=HP>v8mSo0sMp)0 zWX{LGtsAh9%u?$Y46tllVDv-y#Zgnm;=`%}{R=R|0&Iy&;-P2WP^n1`C|&mz8QIP&ElXKN z9>HlZ%HlrXTYK--Z=CDwE>l``T+XK}vQ_3ZtbejD4*$Sft+UK`85eoj7U*j+Az)SQ zp>0~2q-j2#9qr#sqXpz=?rCxax4%`GyXMExB;`}(2v6M3IYUW^#43z&$c z76CD5UGSZYaUD;={M_@BSBtTU++;yiE?3D%UFTuWgsQiT>W3K22XK?d$bz0rRHsQLvv#Rv6#_2&RRs#XJ(sN|&$huwSoPQ20@)~E`HAA&t-qk9EVL(^%5bvjl zwdF#E_1~0tfC@+ybr7XD>kC%^B*;(}-$)OAMvT4% z;aye}SP%#6Gno8uV}kTk$z4Qtp-?+YX?ppG$;U7^mvULnQJXddD^OP-Mde{5e+W@6 zLj`c*0eY6}(7FyywoZzWNGs;h+2N4l^@w{KJTXCCP?3ru;X${#0tue8lm26VSSfN8go?C0#+>o<78+x z;P$36*}YvM7c-uFXu^5ROl;$6J(Jd#LzBkCwD|Iy66@_OFoIRwS1C zZ2?MX{1XnVd99G+Ya3>?9c3(keMM6hrN9!)CRrJTvHxAoAq{&qII2@E5J&Br2Om_; zi{m!ulwJ=32npu5)SL=vPNS%e>xH*&nYS=KEaKP#@R~wf7l?2-$@>Pa8K)Bp6t`bC z=_{4m|DBmFi-fd0x8t9F zIk4EQn%ZV3U$r65Y>sxPhfV4N7p@&2|02UJHDIhE&Ag^f)kHkRUF~Yi(wHRslLCyu zwnp8#IU!$w-RvUEaehDmAwO)*I}18{okkgU=G4gfN@)@sxC*);IwD5@;4!B0-3^Qc zI)be%jknV$XS;qz3#VH&EDY?jBy&F2g$FoaF2+atME;|}Fw*Go(V(p#R4@DmfmO0C{4!~~XAcnY2NUPU4xiTVxoRZWK@s+iGNsJC*lmGBqS3K=Evl15ab3@{Rc zOaIeRDB<8RRJjo2;)?3_mRlD&eMln^Gd?L8z%LGuC_P-x6)E#Bw^Ec0*vYEInoJJZ zCrQ;+Rqjl9Y`u^r7GM2vV}_k;O@E>@E{bUuCX9RSznlKeZ7chI*OT>Aj@h!FaX06& zMMz%q4CWNLNWm@8 zcM>(r!(5KToxgn5qFl5RHGv!y#iJF^n`p7VYmgJ(WNdB^-`J}QRRu#ReZ_-v8Xl&% zw#KZWM-BW$SBM_Ivt8b$;WYQ<94UlBIdX{XP2mwVH_{JPF17&x@s|SkIl^Dw05mj9E7JmD6 zK!=7@F1@ZV!`FOyXkUlH=^E3%mk11U@?12B?HZTJlF~|76X^)g@Q^!NxMy&kE3Z1& z>%nde%V1JzJkJAFf{$nI2uLoUi03T|r`~3+U2t1$sPIjJ>%VVQUe6Gd; z_}oQyfDGz(We2i?F2Zuz@D-*%56xpPwK{a!JO*HmB+0*Ib8lwAyQ%At3KpnPHeF(e z(^6DV?sS5-^bp6np8o({jbRnHJzE&EaNyT0l7~3t2 z8FTLLE6$bM|-qd~}h47x~OJsv&&mU+n|qWB8reO!OmO-_J5P1x6MZf-OOMINtT z-8Be^&w#=kp;rTLP`|M>WxWdA#8NeP=#Dc`;B@Gmrra^}Nk>_Fv;yOSTN!RK^!U;T zZ1!Gkxnbc^nCA#1>Yco>yd3|v{SwIJvFVfJE52m!+xGU7UCU9g**i>8DEOwWVnA-i z0wVDP6!Qm|X>3!=g8!+c1gk>S2^ZSb; z&k|!{&(&HAl4^?X#$25kFfSX=qWGsdQ}8 z<+Q2Hu;7e6zf4hdY<05^FGe{R5WpE2>h49~7{g99)pY{-bwcCUD@)C;qz3aHEZ? zG6my1=iwFyc5?}q&>2*;aMk+V@#cYS^x49kZ<QC?3Un30qJ@H_U+ZauJJ=NYK?=Bo>v@tC7)44>g&3|ue!YQZ=MxPM#Z`;m zhSPH+Bo)qy-MYvO&O-yHsR$7*k{is|@8f6IZE$~|x*0Lo&uEJb=#nf>x{QrT@$S>q zsb4Z|Q?%J-;Ypuya6DK(b9B#16?*_99ClpXxQyruS!~UqP*c9gdD-+-Hf5KEf(zG3jH8#$=?&>VK7l$IAPCJU z-X1er%4XmErTejKQM!ds+W!Z8Z|Gev3Azwy@u5iU}-gN)+#9pacVeYMd*Y4vp zwTm0GSJEZAnK+1MnUA=S$&uN;gGCwjc*S{OfOs^%Li?N3#;N9}IH3;qfs_ObeE&AB9*bc{^eM!Ek4H{xBEgYep4UgsB|eySeJpYZNDu0h0`&bfy{>+`OMYv%(gIeMf{i`$KP>za67v!oxGdm{(yGlPGUAXq4 zL4Le6o&&uoP5%WI7oSMo#%oQZuey+DOQwX6u_$jWk+XWbi=S@AZ_bgofZE_DHfXCj zs7q~HK;c^7W8VR1r+h73H2sZ{-<@IbeM)m9YMkh(eP9`^>&@a|Qh=K35;Ue1fekQn z$+=jx=yOmyi|NbUoVItb{O}b^CS$X|ujg~a&N>~-Ev1%*UeVLpxFr7kR*pRuP*Z%g%1MD%SJ4Ts-VV$#yn1>?!>@y-ADoHWfOk3<8P-&?NCvG zEA$x2>?-S<(E2mO8|q4!6z*>dw$p8R#S&!TK+nqVQEl3@%N3 zTP}D2fXY6)?HCyNoqO0@-}Ke)wE5KRzj2YfG~H)o1P3-g4HCZ*W*56Om*ufM>I}B+ z=YZ5?C8f_TSfB?!k(W?3S9ei)-Uk8^PlS}q0Yk$mN7oo)nt(YXmXmhtGAljXVZryg zvXFDHjZ5=SiC!(Nnw<@U>4`Mhp-*u$4A8GJq83ztYpZnPsL>+27!6HHIwy-0%^=#S zh${`Ec(IZ<-cUg?h3T@&A*-P6l%Ua{ERdTXYZ*JzbZEJ#wDDyOTaQ}SUoXxvgE$yS z9h~V6em4Ag2RoFSq$XdIn~xdRO6fE@0yZiusWRXDBfh# zLqp%l$-J@qgT~p+NWT9iKe(8*OOKf6DOtvMYtrj2X4 zv!)EOLXJp(etoOYLfU)9jB9Ox_+AJfXtioX6KFtt@qf&?cab;i^eM&I*g@L5$5G*A z6s2W>LaoSO6cGp=NB~~Q^cyQuP1MYw zvRL&FI`U?*N1E*iya-6h(3^@GlM?7t0;1E5E=bnEY5IKSTOJ#D5zEwM**U80{`sH< zLU=?cCQ54Tp*JWItYpa6bWJ(tiR;@)D2L*gZ-aj41r>9TIw`_cbUHN&vMMM@r2uh9 z1D*l@VXU0*6Lk?wlocfwW_rw^z+?3g$|bCw5#RX2MtlaA-v3Mh64;+7LDlBi^`2$8 zb|ul9l+bLcpkyt@N^}d16OMW`aiEVe3zUtF7jKNb)92I}C)BReVNu8itGxG8C)LRwcXDr}x;%uZQ2s z!o_)n6E2{S@MF_kwu(|u&FX;8%j8lcDOU9|S9%kAZW}ofc$j5dAY-fY5uJNkac#3y zN@Rg-9g+%fqp;{lKUcad9A7GwD~}I3IQ|=Q2+9uoafifyU`O)})+o+Mwq<5{W}r1> zH3CMJL1;_#E3>@ zGgK4sSv9KKLN#OVpcp3*t)|)?Wu*DJV2aXblUGN%dQ=`KAZM)~)rnO;wzcyC$jSd#hoIW@td3I?haPqDt0o&5iPL1xe> z@}GH}MJkXGUPo}JjNe3iWtxF0~h7gV4mAflmQwkAq0HzjC26xkzfh zyC%@ZFQFC8RQezd(E|SNVBZ#sf}N5pTql}%PJs)*>`wxdksHNFc@(jui-q;EfCP(O zPRa7hw_9t`e;{L$6WWt;JzS);+-i*#hl8*#%PnPUgiba6#GnX&xT-WFt3&x!P(%&;_>wV+#o?(*10TzFpa~gDN42eZ*S|bQWCLOD7OQN4>GCC%;~p zgd2D$wNQ)RLEoYh3$ZX(VE;TMzFxs0A-k0|jpYt}qkXdhZkc&G=w3Jn=zBP->j#-a zGEfvWwWJsWaf}Ujj+*nZeo$#WfApU6F7occRVL{Pn#bPGv&{uMkcaF*2h{1q8#Y^{ z%gm^8mFeGa!haQtri>}B)WY`3MAFd-0#J=j=+&6eKMM>LGZieOux7r7l~MA21SB?E zo#BQ``CTX^TK?jAv^T=#_2ME2z_5)`qOQY5jS#v~TdVhd^slfg4-6KRJ>9Hi$=rBg zLH_+Tb@n7x`v4s^)_hpK#XucN%t)HDy^Cjc34#AJ6R1eLNm~AL)KcmPkx&xF6Ne|~ zs~dG+SCia!7_i;bm;MoO_U_KILH2GNBiwe$X`uN1Szlj6DUdmT#GZ=B{-bZ= zguXDDWKaIl+grH_6nSz0OY^5Se^LB=iXU%zb=R#_4aW!#Z@y&#%5F&jY;7$W(84== zac*e{W|^@a_3baWw*`Va znowiOu3$Fu@0DI5{`6)NSBnhuIKeDZbPT^={ivKXB}5#1o~zT^$`eTa9UFOhHbDQ+ zcdaA1?Xef7zbpdEf;1=Vc#uExN+oF#w{@K7Cqg-U~nDS7nz@*ehgLd{iD$m z1Tf#M&Cf+XQU4}Mgvp;(LP2gPz?Z7>gy1A^Z9wPCh)(<<`lJ1o&ezj;X;BvSD z{^Umy=F2;H!)RsYpeNN3`;2Kz&rr99&GF`Lp}&=dpZ>h9EDHw&GPA%~`&>~QhWK%H za?`Zzip6WbWszYAxB%Z}V{rwF7b%+W@O4V(p?gZFRlAYmQkWRl-$JEEK@k{gzH;VSjRQ`=L$^ygNTNe9Y z#;VMnqh`HX<(_uzor8$!LiJe}tbP;M%`UPZJ(sT7IH2;99wY<1(!K;Ydi`>US7Y-u z_?xLb4V6C*J^T_oUt}r$Wcnu$ z{m$*j8m5x3{iw$nyGkB>8Rny<4P)_eOYdp768%bWmoW4P{Z}!~(?UmxN~zdrR#Z{y z%8MKxsy2Ty8UR%Su`%}?F?6at8MbQ6Rk$til2{fMaaXX*U9_tE8Di{I>g}LF>4ERY zRSSEk(7l8)!7`HKPKk)NjfI12HS{f|mh*SGhk!q-C*+@*3_v8t8KKe-<+)mCV_!~B z-1Oo&D^mDH!_9h@e}vs_eu>WUX54xjKN5qOYTy{2;o!630@Rg z9Y4^Bo`5}7Zhul$d2j0*7^8dz7gZceum-BQPWZfb7wFRuJFdEQGH7_o4CX8@_)&tr{q}-$iT7>J$jS4gfC9 z*%v_+tjbpaFKW6|5&@7t-%lQEeDNUW6SNI~(3`~T-FjsRU$amV+F72XjqzCvcPIN9 z{Nr=QR-o*z)L@Wkg1cIHhC+QX^Kc%BWc?S-p!mpV&N^oEu?V-*m$ZSglT^0 zg1`R9!)d&vsl5{~33cEh1*V{OoSZr7c-NeHyyruXYw^Oy2B7EBa6TP0vZJc7%T2wG z_sUYZ{S2I05vz45JN{{DXk}U+Fqg1*4R0o$WMKe5m2B!+(Y{=yGLGDEAx{>+AApHZ ze@R#<*0cy|GBSdH>xA;o6}gUhfBV6P!u`V2B@Zv30iCbuQH$n_gRwgM zgK`L|{`|N#98 zEX=S%t+bU0W}cZe=&It(ocq5TS#N=6w{~`4|Fb@E&OcI*lK)}m-Q;ABbbn{-N}g{1 z1+PB2Xl5 z2o?zU%4Gs+Ho-r=8aLQXXd9W*eP8BPznnaS&D$QgiFV?l@cXrSak?|RRXt|0bx@*M z&YgjKZ{>L2Uatk2yruhh#c=*VTc}_1jpy?nYXoNyG!&KA7vkHnV-~T%G#GsU3)L?x z`mYL#$r4R_Z4QAmatpS_Jz!5)^1on|RA2;JNCFtsI9&Id4LmxOdWxhVmZpA5>HTFH zFMLe}MwoB_2o41<@OEN`_DZ%qV!*FwpILh1K9>){2>3ouaDG?@2sR0 zLuUH-4$6X>AogTXkBW?0FvDm80`{Q+|LB@IoAYd*ky_AZPbTXU&e+7?%@lX_fB$tU z<5ytKU4fwT#nyF4m@wXiGX*kP&lQh~;W^Ka{Aa6C^x;2`@&9fI{_i(!jQ{VBap1SK zG5)8K{qGe{{8=72FU+j#{XFabC=Fz)XxbB2xy5XI&$R;j?^}YOorwRwkEvm+oD>h2;=lI^|FefGFy+z*R*3%tS9)3; literal 0 HcmV?d00001 diff --git a/blogs/deepspeed-fastgen/assets/images/throughput_latency_13B.png b/blogs/deepspeed-fastgen/assets/images/throughput_latency_13B.png new file mode 100644 index 0000000000000000000000000000000000000000..d11c9bd1f6df0384e11e1f9bfee4d1727b346f4c GIT binary patch literal 339033 zcmd>mWmr^y+pS19f^-YgDJ2a9BGRRZlr#t4Bcm= z{?Ge7?|I(u=Q`(ua?M;bd+%SZd)@0^_k?MvJ;leN!nt$j4*oMmd96ElaIo*(LDj~> z0N%-Lv;P46L2=T03c6D=K(hh7K(mrnmA!MPEC%<|1RZ#d?VzaVbmtCj$n8Irn3a8t zJ9lpKp2^GVxEpO}V8Uph*VW7)&Y_}#)Iqy}4<6tsk;pdYFdV;o61!!EPFh!R->>kc z4H@j+Mh+YC*I^JLG~~`k(&5~e$Y82ow|Ti`R^#C0MY`ejATOt9mZX5~i!`B%>w^-v zs%Z)n5Ef$~(SP`_Kzxcd>p#5z`?CcC=FDvW@oPk&5%x#K_y314Zr-QF#`(|pBXS9k zL<{;)PeKF=6?>xK|36$(KH~wC?Ei4d9JRQ(?Eid8EX~hdiW>jZ*tP#(8lI->>R^_A zDGaHEO!@DVky9M|&kW_*y{2(TcC${m_@N@Q0|?5muP&y+HM7Zfr2l>DD}@ol3e5As zYK<@=rmB>iNIPR0akiFK z6NF37qQ7`;x(F>e~ea=>n7Xk?1nlNA;l3)6?qLV+4aa=|>krxz(fFY}WAG<|{ zX!s_;^G#P|SC~-IaT>>1Kvcij?T|-~shj`Z?x*?N!aKY$qtBaZPg9 zG3%1MSp^vA8x#m|ctXWOP?VHgD;7`*B^XNmpA#t1YJXN!y$%=MEco=*k|oX9-+81! zvzs@q4%x$rxI7qCzvjT(_d?hjJ9QMP4wV{9Uu>xRzOl;kc10>nwR_G%)<`XA4g%;e zTX-6_%a=>PC&tfdEf(L;J<-lOPZYSe<*`b43^sJ?7upoqd>@OlyYV^82+qXN`TB8c zu&v`vt>YV#el(^fvlvNNPB}vg(Rb!UR}%Y|YrfZ3-{=SYKf`aX=5M$;cIdD7LY|x$ zy3^(jrmSiH8LPbLYb`ByDvYI{ZG_S@Qs+Z4_0M&d_l-h36bcJMwSHZps0o+Pcx)B7 z-fpSTH=x_YuDO=jrbtAAX2!<$H6@nK;|Hm(okh`4vIWV_%AabuRJAf?3jRTO__&S( zqH7aX<3?$hRYL8`XN`N`al3g{f9j;e(+`PxF)E5iwupxHcyd6b}{K2h*Pv@IDE%;9n-TCvK@xx0qa8H z0W~huQYzEf(KYsVNvQu&U=Je6g}$2mF+yb!vEF~^w6PU;eE^putWx({Bzb5XS~W(G z%_wjC8w?AaKp1pJ>0R;OuLp(`q3^BrzYZx9eY!7kCbaUT^d(bkGFCH_LSz?dztX;c zQvMg`ioQF_N6AY1Bh(mCPiY+GWlj19Gmp%J2D#ktb9D%ODlkg5WoCOtATY_>bh^kA zl@h^EynH?f>;}75fAntN9|5xTHdu{)U1_3*c!#V@oiXay#D8aNbx4az7A~JZbtjo1hTB{1MzO({Y-pUJUR(O}96b!AsN^gwq(7^N&LfT{jO{$a4wCiXvQN2J zC%W4=omo;pc6g0c-U?9Aihi!+b1^+%Z6hTQk(uUjabY>ni52GPuM}kK9Q3_DctW|E zYzjC2;-iMexQX#Qri}19A5A)}BfOP=+?}!u{!)++tE8f(fuE{oi{nMOBM>!-fTE0lf) zXD=f*;W;H4$`zZ=yx{7-5H|W-1bpG5-`5a6{k8#|;M2+K8?gw5A?Lc6gG@{6l#ZN$ z^QT@JBPIhJU{exz9Mdqk^|$oP@{jqjR%{djs{!Y^10}us8j6?xBDjb|N)N6z<%Wlm z)FK1iSFsk~1F;7xOvo&~6$a%2ByV-fk4YKiEQLLqEPv>GxkR5}uTQMRn@DWWW|;I| z&l55(vSFcMF&EF)_m!EqZxvVe42LlsH3o}oR@wKt+MZf!kWc)KPg8ktz~`^F zrH#U8h3;@=y%J$U*bJ@bcG<&HG?|MlWJ+^703Ss!-p^^^Xv5$6;GBMGs8V0vd9S4T`rO<$ z-4dCpO4vL4ZbxFO#=hC+22PAUe=&C=3}#&(MOkCmMZ?fedSDLICX~51`#z_`!j5#~ z{j$$5mrSQ<$Ln@^y%^MnTaL)#b9% zktm6rp$tDUX!1&}PoB)&QOA-}%@r}{!<|IpI`;8*+TG(_Q*N8zKuFzJuG>M6m72pe z&R-Zz;^7?Vg@hN2EHVuS!rpOX_9LX9X)E5jx!Uz@PTJ?i@SYJVcktqeFhxOnslAt( z=$|#+5&{5p2ocCGO{28^EnW3HuCS^NtfDm6Cy>V>%-3hQ^sPAjPSIB65PeBJ}apNU~;&_YP4R%Q$*NFP)(~%wv#ed z!pc~7#gIt9osVN51Brs5^QYg*n__{F8dx>OEH%n^DT+=l*UaIg)}uXZ_**988ZIUq(f<^_R(+uczTH z$k2@z(cxYJ4YHy4%(bP2ZD|h5YI7Bx-|q8u-mhSmGs>vq9D3+kK;VGLoQ)9_a{Cgk zai;AvZtUy*c2;ndh5fUJ5pEL1)fjXvQJDw#IhpgcFYio_4S`=zu1JK6B+kLA)5v#~ zc^ebAHAUIl;B{XrKr7UX-*I{II+73~!y2vz($J;y)|yWshZ2p+3$VfbBp7(6@*_(B z@@{>Q(EXSG4PTIWd1-~DY55FeBYd?6{Ay$G9oMBrf4Lp-(0+3g#it{}` z%_(dLGsHzEZ3P$&IRxhhm3eKH6Fy@u+hDa*E7`N}SCc_(7_y_t<>nscj|3DLs=Zl# z*OmiW-bg#7p7g<(70X@uRAI&b^mZAe{D0F-EaYMYFUx#&u5PJOca2xRdX~{)Iac3- z_)4O7xDm*Y(tA6GwMffEn8;}D!@ZYS1FcBBJ}EJPu_F9&dDv?4Pz2o|C2vvm@5jgkvY| z+SF5i2U?$_LSq-BIi(<{Unsox9p6Hnz8zwXE6};?d>RNpipT1E*B)UEb(cH6uKXeu z+VoXtHzBwARkC)r+QxKR&*u=2OL&OX@(y}snHGhBW#oZ7IkG8rrKn<{=O{a+CJ?>z z{#MX8A~|AS6oXHCF<YqU6^^jeuznEgZrbuwh2xlM$3C{ zP1bQdS&k)p{)`a9n3cCFU;{9;A>jIZ^5nx(dJt4^L1ngRefvfi&Zq#heQObig_h-% z{%j!&Y#X~jF({33F6(lcwcD6~-OYWs2b9hTUMBQCmE&t%7FGSS|H9w4q>cqxg4_sK zJ-kHnSE|XTN3uvI&p{(j1CmRHX)Xoz#aBx7y8)%R^J5ij?8mPLK&f6{8(S$Y_RH+* zxIOVwVaOY6 z`I}^ufdEvI9CTu@NMxvvOSni~VUrI(At7wSgc3RAp?4g$CO) zgrPSr!{7~ry-kjvwnH;4wm1M}5%jrj^?eA>CQI6jE^*^JtF^dbIHWxDG6x0u;zDCL z$dbY8gw^~feQLKdEU0To-TEzUB52O#!`jg|gFx8QSxcYf6bGYy7ED-ujFoKaf0*2A=Hm>hc~h7I+$v2xeWrNe;`O^0w_S zJ`5S8iqE|+ikFOz9W@1+nMZSXY&Cz@89e)8{RpyQ%0dcz47jz98$&auot z@?x)@boet{;dWDC(J=@Hy$ThPH-bUV^+mfmQF|Ijgk`SF_la<91k#yuW!KOLG}#Ox z-sDH!)lk)8AQmaN;BOeBNi6f{hw_!N=V9%g1evXa!fo zlD$?pdM0Hc%SRD5pw4)L&BfgJDeZQO^SGeEO6g4~{31aDdG427)N^;lQS`iSY&|BB zD2M8IRzD&N4VeW6;L;TwU~n+*Kd3YobHF29N*9n^=n_~&2GC1utRys71jy5aPzPbM zsf4*ISL5`$UX(7m1Feiix(RebKwml@F zs9R8c|Ir(~Drk|bJZ#b%t3Rwp)CHiE>^c}DzIjliv_~!YL7@6)Mw6dJf_(^sK;w_T z6^oA3xm*q_^FA!IQ(gZcag})Rd0Ju4Wk#pV$d+iTRHLgSP8TysP#Z}pmj9)t4tMnV zI>n9|^ie-9M2uw(AE9m?ZDKs4GcAUsdt8YY9%=|Jv)Ue=%lkFxnj2cKlHp~ne zX#%<+Ukc-Krh#zk`1U_6eQWKz;*Z20RV^qM>{R4qRy2>nPGWB+eZV`Xlpyly8&M*^ zi=JJVksNgEqgb!)w8Ku_+FKVDMZPsJM$PxJR;N^7c|oCY<@X?eImJ6UC34GoY1fq+ zlY!CQTe_eP=uIzE11~gP!5`}Ja5+r-GLv;XxMX_UN9voe;_F^zaRi6*oHs+MkG>MS zjV({*3S1O)awBNR`{Hnb)AO&kVolIL^qN5(w<{VVX9kDus~^K4OSPtQnbFt~DiAxP zU_O{*Igha6%4S+l4U2J_t5Dy9ofVBP+ibrcN|;zr zhn)Bu8+wgD87%3oVk)1J;0*`A?edf%K%|KjD3VZp<5y%8WWd6I5X!1Nw~!ZW1%vrn zLOLm$xRlp?q1|+HT>@iR8J|A1QUx3zGNiiW;Ag)A8O>^3uP~^4b#rdP z7ve6---JNl4bUab?wqg#AeNLbs^AY9S2pTeWpb7Pe zS&`RSKY0ZT+-$`}2J>tke7E|tFK2DgcMUP~Qa73;8^-KGU!fFpb2Ug8S!MOMCLOV= zj%@VPVKijmodYe8TBN?~(BBa2vb9YbZkB{T%Ha;K^!CB6`U!7>fO{}|CXIUoLF2Cq zwnbS+nq#wyGP|1?&19!6*mkykCS_7@c_c$1X*E;?kKn-OhX(b^aUbpt~J&+p6=bW6Kr6-!^Qp1?dAmsbjj zpp}9yk$~cgyNLi0egIo?J()^U#`RRaZq0fivP3CfS40&b7Qv{F#t?``iiJ%F;I+qKVecXTB*G^PwERQg zX6m~lU!pE=_+wQvWV< zXfp9I1z+TAyd?ZcIh-hyiOkVJ(9(;$3amkqWTBkS26Ge<^Kut!Q5neLAJ>euBD|7; z5=#kMf}b=biEuW0Af3xeoGu?`Qy$}NxD$y*!xOJH2tcVs2mxibD4O1PZ9gI#*h4f# z*H#*dD3%3XR797+U$0*&8HsbEfr%RPCWz`w zmiH*#rQB8?M*5pQo2{dh5Sc{+Ldhb~T|=ZRX=bMg>+A9o;u2V^)?q^kR^ zu+N(1RsKixOoQTQK~_I@8|xF}t0dU}Xsb)j(E^BUllKLXw3siphjrJC{i@$8sW^eH zNutYz>KZ{Fo{b5^S3x$aAsP)d0+uSsVlQ~`PQ$KL1E8ZKaud|bp>iRw8e96N?~j*(zekv2*9iFbb+X0ofmcjgtX=8veOKN9%z4Aw!WlpH8nfh6Oc|H0x1%ZDb=X^ zAQONoK-LX<^PG)IuxSrxedY}52;T!_a@P5tHlPNvJ0N7U!@UeWHuCgmBf8|u#2#CCAgacvfND*GvG742ulJXNi}W=>G<7^-Id{t5uDka3 zozk+dncQNH>E&&n8IU}gSuXvXFZ>}1jMeW$Jb9g?0NpN(ZDkeE{MDMk&!Bz`%lMbb z#X?;&?pKLkBrOZ>6W>e*J_zC7eO)*0pg6;cnG{~bP9Sx<0PIcJ*=kyMO>#nG^*x^6 zC$?0=~lE`EOwfwu# z_xaF!Oxx?9V$^&lQ|T`kF!d?Pka+tosD!ny>rknU5XgRFQ1#RlM)zNa9~Z<=q(BH+ z>gNifd8a7Lgj@MF)FY+Z(Qx#MJ9?+=UY;CdyIhM6QDMdJgR0#XOSkSS~8>9 zz*iqu$q*Addm%-U5=}2Xw?X73{G6N{U5+f$x%{=Ac9Ekge}|t9oQ$`A{efyn;aw0F zDVTPt733F|qSp~ko94x-0=2DsiHV)d*xvV1phn47EU@GN5M3mJopWBCUwdd48?iSL)KCnxHg>voIh44-5If6~B(c{-IaMYL^hk4bihAKrnQx*M zNr12i2R0_KFnQZG^FF}XO%z<{0KL6FibnJi1!x;M_T3_yZ~2JK^B)WF9LSU6)m(B- z$4;Ro;A4ra8`{QFLF2 zoP4S_VMn#$ICNPEKrZ|E8D#nZ*uBWW zCMyXY+*^9|O&_Z&!c7-4$H$PCZ+Bq_1Nwa<1o$dxB)e+jh(xbK8pvJ2 zm9!Ki%aBDAE^3dX%_Kk|){|?8-?<6rESc!BuKUme(T~t*j8D<1PsLe5I-ErqUp`=C zc*I018{}kiiLjb1_pOqdr$stsd&;h~ev(xbeDc0s=G&JPGl=&KZ6g-Yc$2*7U!Bs{ z|84sO8Y-H^o>}Gdx2R&pIm3_MxR8<_hPM%jb6I1Bs2S>DcOB^Hc0bqo>LL`M&t_$;J+Sq`wvbF$v$a(=)D%`3yo0%7 zr#pK0#5_`EsFb&1`_Xvw(BRNK)FK4NlU#oIjt)F}TR9S9_+3VSE~i~`a8SKQ$uf!x zwa|-jY&G6pK-mIh9;NFkyZ|MP-XA>EacfU@fy#q202Z0bH%u`SkIu;CV2n$jlRNaW ziD*Cp;7F>{rt{6^(!~}yulyJhzxjH-d3XG>{?RZ1pv9Y2>;ctEB&lDvtPHF3Q7lA< z8nfTQjrfW1s;KeMA`Bnp)){e4oqPO^5rP;&EC4%6EF;9gQl&&-YjU~gi36fq0Ey`S zd8hddsH4DIjghDEaCm0Py3HD@7tYVYIogsxk;KkHYOmb&(I{4-jyfK@ILdMdj z*Au0}-}^={+N1dXH5Cpe7V^8lrxui)%;tE`a~Htry&P`BDcg06u)h|75$ab@@+?NC zI5@0P*5f}ZL@D`9_9f{W>5UM#UFas6tHBTug36Zz*Tr@ zgPbx~UMfZdmxU|Mycq8}>B zgjM~kT!s3of$7E?Zb4l?J%1YUa*hrHTd(ntg98k*W`BuCj>O|Sl{BBw=!-8O#p#fw zupO(OLw8IF8EOkK8+%NEnjPz>T!8w6hcX`>b*uXfb6X4(6LEny z&=*;`Avw&=KaD1MM=bed`iS4Oagn_LWk$!91Uj>jv-M}YM~u5SlY+S-H=m^H<|J8` z_fgcz5?TDXHn<5`ZWdpmEPk^N=J{7@1sUVRjqj@m%h#H2+7(63+b6~}-a_c`t(y-E z^C*&sJA+DsKrSh68Ospozy69)54p^P8kStd$S?rkZcfA2H1nmcRMWlqaCuOdue*PJssimuV26zCd zxN3?RO}hll3rfC5!J6{9f6@)341Iu497OJY&G;R(lGPR4dJi7I;tUj&nZv&SD0slu zbECFrbTEj98mOj7{#~8#ZcPUnz(!*D%GYv$iYj#`ZlA1UBvExYKoXV^_f&eJHG#e( z03Z)@nMk}}Am!YJ3~s|ry~nH_hI$lUP6kynV*C!+khHb{3dB+ewf)R;f1yTK@{>$* zNqeuShI?Ts8lf=Z(9SlZJV2lyPielis1a17no?p!Z7@AW*!!l5>qoz{nubH=4C}oY z?0pEsgx%hLt>-!YkL0L71;Ac7Q@~t&E$KYiZk5}MjY4{ zL~ky*f-=X~s?o8G6ZNgfgfa<;DCE2?*mBQ$berP6rt$7l2gg&WYe`j#pR=5Y!>$`9 zR(}08GUdE@4WmP*&yVCjD$XI+4tuVn_|)Th^gC;+_O|xcBDJR(jhDz?*?yrI3AySB zo_aQsVrxRijL&Yn0DE~n2PaNlr-ZuSMk-+9&2`5CTZ#>v@n(dFt&R7}Mr9Ikgr{CG zv6G#rYRvywMu0S)PCAq8mjq?QKph6{#+-pATk}^tZrwF}5DVRPk1xz40pl14=Ll07 zRJ^KETwWO8S+x)ZTq<-9#H0d!;adsW7ARHy>ftO^sjg1fxYRXpd{SP;m&@erRrU*@ z*XpKI@o`K~F;0AedT6|4HOuY?d}?84Pdin%E5KP{TJEBJD%oKFg^@}IoX%$gB3>Gs zah(QJPgmKJON@osogOn3&f*B}0;yxJ9Nk*FlbXYgb5}n9hzK#VYr1*DVb*2*AwO8v`X;|e z&{?J6>b{a2^mxnwIWRs_cYd-<4?GuRKMZmpOh79pSZiEn(jI#6ncZIsIwS!y+k0ot zoVgk~%4jbUbxiCJ+=xUybTyZ=K^kh{@hp?fKluA8KW_8N1}F52QIJO=ggtluu0-GKB0_tp65N|F1 zWkcn3R%IhSIKBD5RlckRUcU94bnIvJsI39GtN2-=a2Q5W>ql+KHyZOU{cKEi%c0jS zXN8Y(xOKHucecoHGlxPvccV{IN`Ar<}Yln+zY;eK@W?AKnc7@ zt5?4WeP6<-U(J1`ao?C;x3g11l@zzPh!q%k+(rlBv8quG4mxc2v+OkfMuK}A6%Y;c z%}#SKK}HC8a78Aoy7blI!bFDq&%)pZQvUY56eB9TUhXK7%2$||KpHsl6|6B>^m3RO zZgWt1@UArHF7(|&VO(yNSi#{NW}uALpvb!t2d+3YhF;1jA6azuio1LasuZ#!O+p`qF$qs_VFrzde+6cmOo|A7S)drsaX8* zkqn$dW#Iu{+pw5x+_tRPa{+Zp8^sRh|ah>UkeA{-N^M`TPOd zt7=XYVdmwkUXVuVi0*o=#zN|wuNfi3hAO_QhC-tu`4@5UV-S&aXGk=Y`OgpI3Tgg_ zq+;q~Z~{PPm|O1h0(A;M&xe|U5Fx@uO1uKeoU%rBWN)U4-stw@GqQy^;q(;@2Cp98 z3NW{o$hfD9+H9W{hG)vemZO%!@gbS{*`gagJL+sHMp^`jXntWaO*%VV+7x^}&;^mA z)0eq#NZ4~B*1%0%aLr3F^L^x|z~Gj#HuQ__m%0_rf^GF4J)r=&)~i=d8yb<<+KMNv zw5o<8K<&uQF!^A%6YwBX%*HIcYhS%EqJ;96MYWv3r8Z(ybe#v`*SCfL z#8;S*3vz1rn%I0KllzCygQ*{L|y#N)xYKUP1UW_s?3x)~I3 z$<4g$u& zZcmr~sQP>pkZ7;?Q9=yt0cV-R#wWv5;l*F}K33vqwgX4qcB&EwjxcHU1-d{0 zTk^n7B6MkYSub6B=Ru?*XBb5#)RSaf?|pX8UPa#xy6pR&Hyec1<58p+xQJ*ZPk-0L z-kn_SCoE(9?3HBhE?})wzkY+qhAA;iR!?>sRSuQ@Qgc!Jm)tIE7G@swqA#6l2?plB z&@|C^>Ji|@?;jc_<26bLaLKwMQEsk>I!K1M`v9h(&r!8qAnm*as$`yX73I`T)L&B3 z9(b-3KIA*WT0EL(gBVp*LMeR{w3+0*R%|e&bCryf36gV!$|aeRk~$A6HDms&Up}`Z zNMy|<>3yD=r}5=^0h^QawicB4X~dHVJk38+esufG-;QI_<@T0H5g(GAIzBTVNr9d7 zyA{3v<$T_ZQ?h+)gZL4WFJWxKKI-JdKR2G1P_lg;7<k!UBgz$&2j@o=gp?-X~?_E<#lSN+qBt-lR_Uhu)Ygx_C*OYq#KO z9g_DNX?Xkeq>F%5I`!u44h9>Zd>|Wo@@j|7tzAvshhJo&m~3D1uR9IE8$mM26f1@| zxJ)I#fRI3Zo6m{{gW}$-&jn(q;rI5nru%2Um{igq4+Y~>R|okq2+W%^{Hx`@fE-Y1 zn^ZCU9RPFh!}&kXC#S!DcloGqKn}+`Z~D^I&w}Q!-QqXg<s zw+Q{c^W=N~`oQJ!{nlgv10TtDc`uK$&%au~L#I{&CtFP^U?iz__ysyS+56RoxYZkP zUAsV{@GqAzVVFa$Fhc*YO%2ReGUX$4MI`qz!zm!`O}s}zjW_=D(8_;qJ%qzxX$KgM zBU8UZX6erM_bIyB&0Q;7nD5pl(>+?dp zqs`H=GP9>!!+8-d>u|0|d;%IAJygzub#IQlEpQZ6P&iu29Il>~* zwnFoiQ>H%fn;FQs@66Oypsi887ke`=<+@pC(uP6lgfq_Kl`RS93Czyg1AoUHnpnSP z1m#}f%u1sbloIJ#lSl!7=p=1l(3MnG*^a-8p_8<`xjLx<0Cu|GZ8!7Z`aVK?9x%0v zwdnq;wE4N6pZ2yq&Aw^UvOjet676~8Xq2z$nxxdCrVQ^!pgoD#Fx?@@7*6KysGQ9W z*DX-RnrE@5Z&EEV6cn!a4dtptpXag0Z88M>|K34eAbog+<<|SS1%O$Y8Ut3!?LL9Q zO2B?UaHU3zrAbihs%H||-FA57USwW%FQx?izt=f*jH&pdSifdqLl|5)916_Hov*RM z0ye|Uw|4Grx8{qC3vZxotH-V$H0 zE*MXQbp%{z_((LSOL(L6ddZu$DO;Z7oMTL(C7{YLfsB9c(pTb7p`H-g)38vH+q1P) zpn*Yyss2_f=i!w^P^Z}S&1#g#TF~e=H?AFlcdUZ5D4Vdk0 zfe0QAOyk>~hp}Jl&KEl5Soy35Xwt-7zItK@JoN+Gd$t}LqRKVAthW28nkg~K%@|#_ zyU-et@vT%CP>5LgPCxRS1#hi+9&g(;?9>-}x@)9at6a8N0G5AYi=O*#S@SlZGwgsI zElKtzaGjR*5UcRBgSFcRA~2w8x(3s|A&J5U-(~tXSz#p}vqTY%ESfAeZdSbtwN`QO zSk@z1x|b=yuVqT?KP7Jl@_o-IOF-N z{MWL4o2YMafo)+_m#L?>94)@{joKbI6h$g1DpQG3DcO~raEH6v@>ifb@5@!I?vkWL!NFFbb}YXq+EMd2G|cSKSyMTYa!%mRkb z^bkJaLwLfj+b^y7>gRnfCWWW%6723p_2fK>n7IlVlsukji=w>uLw@OdU_i~BM=y(7 z`uaK6ISxyHC=z)~26z1iB@QCJ*Tm<`cD;8!J@UydP$4Pgw18K?()emq*Y&m#H(!_U zEuMhJ#SwI3J;1s|2w^7!`<*054f)xJ97cdD(VL8O4Ts4nV+xJ$7f0<3=AAwJAzmI%7NJZnpv@B8p|7f&QRV=jvkmO84R;; zv~Dx}lMkp!<&%rz7zW96%332BNVHxi)3nh;iizEeEh&3`M39R!$!3Z?F$=Ktk6Qqk zO1hOPa8U6yjcRv@t}}hkZ78wB!~JUZZ^b=Vpl#@J>9Fy|)cA}5^kj@tzE0L@?IC6+ zC#0iAj$fA3G;}Q3wF`|Mk7zHsKSCJNO9n={dT_;f^#Jh>o-M0Zq20{(iextHiIA+P zxe#vpDwjbmT(lQz3~+gbW~FfDQ#>@An29kw3ci-8jG@*Mx+f1?iIQ5DuK@dK$0QaM zKj+>tVp?Rf(8Zcf-9OZ%Z>CjjBErmko*X6PE=r6c9s~A|m-r;~ycTE7xl2p#^yJWZ zc$}fv{ts6`q7qg22O5B?0EvxjlH=>WDI5_-SOZ+@B?}dboS-o5+X}|7C^QJ9@S#5y z^^T#9x<9NIjKIXu^!}Defnfr?mhcSiKxcA_Y#mnc@AD^#!d;>@O1wp%Gzo0eJIqwn zXp=>(vXJ7pc5s36Of*lR=4D5^viO7CxYFL|AZ;Llp%>rSp%opdLt2C+Ve=7PX5F08 zivxX2vm8vfy-nrMLwnvw|3DMY)6(k;d0VlADfmQre zlFccGU+L+PP+%i@%#ODw--h)!9#1LnNV5o@)6zglA(m~qa!uAy!4u_QupFB9dUVke64#V?g)-7wxJP<36jFZ^_YY*{W zoj!062vjjseqA}&vgVp;L_r$UKvI3qG+T3Eh=H9Wl%0_mfJY>+zc8I?AL21a3dSa2 z^wK|+4IN{d#&Gc^17lx30P}g5lI0XDo@PlI)sybO~@n#xb_G>U?%Urq#{`ZS|L{!FAlP9w!im_ z%x^DxF8E;zbJI?xDtijXCUm8Ccvc+BV$Bnb@C35X5{$3}cGd(rPZLF{)=R#h3hb)J zW|U|G7erxoRijp6Grl1&G3Mnvpe;QF=}=AS31baS!FSwamZo1r-Nb;mBwQ}46&LIo zU!^%y1H-C*(D>lf)F*4(;j0I-CpJT=@YShTA>hYMt$=Oh2js}?8F?}s0<@W|C9hrq z)^PJq8mCD8S0PuK?u*z~ha&Hm71yXFaFaxUP~i#mm1dB)0#G-`z3&Q)V(J-d#`2=M zC}HGX$E*vWjZHMhpNpLR-{|}ZDsqbDD|*$F-vM8fOG13}OjD2w--0=`Y(8DoX@=^g zErhTp2C#Dw=KX0Q(nQ+IU>x`}fg5F#*x4tE?dp;k~D5e8rXhs0p-;TH=twRe+^}7`%E1grIUA4g?gj05`NAJZT zXauu9ImjN>LxVI(!}y;u{6ilDi5|$yQDU3IPJj-eA;_ucc*C4U1EM+&pqzvCu821o z;%?Qq{l7y`>A^s3WZjOHBXH~lDsh9BeIHDm)&^!xCEwwp&UboM`)9AYcp;;rida?s3!>>f5io0%GBc?llv^;8`t#@$>_MspwF>aoN;F;U}a@iWI z^OkT&ZC}5oU1K%&#%n@E_Ws`IhtH1rwuU~1AOQ2jfc7tdml)E_t;ar$>;*F`N^zL5 zEzB0h8wDZJ7X%QLefHWA5z9R=<5h1-*o&ue}t31g_r=*;X!*X(UNS_ z0kskOJYBhdA1mSX7o@d^N_N}&05_SOtmQ?!5Y zDa|h!r7*fO$x@jEU<4ZAySEfxB+JnIa4=JH`fcqmjEhd`Ym-~w+SO^NboK2iK&T6m z>yNf?p*qkg(AdZk_$1sJdBi5&bALDfw%gF99eAg8=S)rRec@Vzs%CX8%YjqUy-fb zE<7K3L9!8XzLIpHtK4&S`J;bJ;Nz;WjzHI6E(9HU>r{uN5>8)LD`|RC)*zWfKf@%#YWJbxvI#As7_r&J8p@jxTH&r-6sj4|eO|MJH3Xyg<&W3S* zF$w9mw{0A#SmbTrOP%)H`~37a0wTT|y?h8*m1<5gcKb8t1K@@q0E>~_F<@Y7NGa;1 z_k{tVHt!_D+0wgE>$tQ%X_e)I0hxhcZ#7@{Qev;V&{1GN=!v4>|K9lR$KySqNL@+` z*nS4~z7gf_0=t$L)leo*C6dGUkP;QtqpJwcOJyr0}+143Wngwi zeqG`#Ho#JEJKb}*?Aam$k!bs3?m!}TKZfUANnBW78>Xa&SY!Y=o7z`qV~L4;VcI~ zxMHUwnWx)?mhM6Jct7x2eowf^o7bH%&lBDjF`Jl1xXDXJVtgWXddvW1{2DR5#_|VU z$D~b!_o6~dJ*&cya+LnQXHj>&foxQ~Gz#ne{Pl5-1v6WZnrIZ-)d&1f@BDA3Pe?)T z)@~}*R(8#R9-RInw@PQz?whd^b)qta{8;FV0;%%Zv%TaEsC?t+ z!RXFJXxzcE*3nUeEj7x52B8L*F6}ca|qu=WyI4=E~D>NC$XmAC%TMvUZbzgKW0h@!e{dsOOcPnb> zpaN;YXkaAT6Cp+>*`eev#cxUr794Y^^pNFyRuW5Rk~Ge~{bv2f=TKE~+fz(S-T(x1|?bC zu-YIfjEx>+1kc=>Iy6cgQJK&X;%=suWMa0kmAcj83A6Oa4$yBW!>9qoj3@P~d+FWG zOt35gBbmLH)29gNe)fPQKrtN@lx9No!lU?0jl)&JpBxoWNF`!!jkWnIA5Bp)@Ep9+ z3mHzUvxSEwE?u5d9$cNRWw|NiNdsku0U07j4PZ~t06%yUYn#cB;RuO>t8XjKs|W7w z;mJ}%5!ufb<$z3m$^^XF$pE8RnWbDB^C%F%MG(NOmXmr5Dz}GWikK3+nxA4J60o%z zx&K1HjkCYRP-oy~?IE9=Z%yC&E?A*vB&=%AIXf=nhE@OxO$Br#$bceWDUuH)pH9th z8W?d?@F}|7ELybVhw;GV+01KR*dC~1HPtpfZh_ZlANOQImuopHM8W=BN5lhr=5AQaPUSv+*XL9iC?s*}%CdmJXijGZth3 zs=#%G6;)N+)cbm9N$tGD{QhpvIrh)vT z?%gAGwV^(J+iTv<$BNHuY&DR+Vfp?B__+b+B)8*9tLYe@L#7%pU9178?uRM0PW%*) z0dpH-cGstd1ZZtQHK#0#;BA_q_$4Es%CC8&;|6g3aby$>L)IF@y z(%p?<(Iqf65)UN?5|RT0!cfxPFu+ivbSS8(fW#n3_W+V30y8usg3^uTcSfJ*ec#{r zkJqIa4)>hP4AwPzX100hNCS}6IBbBt=2Q;+Az%4R8S zePz^I0GsljDu%Q_C3Lm2dy1|DG2dvSywypcm!Zrx;s272uE{T~J?vtniaa^7D8#z* zmDpCyCf5-*IsAg?kL&Z-uCZHP+e)M64)c;o86$u$>Lf7!+$a>^a^aa~`YNbVFZ@-v zVUDX$F{ag5W7>z=5K+OrI=P~Ipl7qzg7aDWa##H!k`eTH|`k9X4$S?*!R2-w>f zDm^tM%4lkrdw1U6*RK8O#oq6F@9hVbX_`;9?#c7{9g3t-ZzKu6H)eF@(}DMuX4E92 zDQ;$m%&1K2RVq*Z1@4eTn1Wm}cU`i{Wh5wF;R_d!4~u>}0lLJ;&Et?7a3BbMUo=St{j9|5%xn zIcgZbm%HqlibF5g+F~o@Cqlz<@9Qv2->&9Az*J^m;L(uAfWY^ODi2`NK%L;Z|2@r6 z-Yp`Mn6ipL=~g?H@`-!xt0dVI=ga#Y$2=n1TXcuONviHORE(6+^7FTUP5_m{eW_!F zr;u$4m<72-QN+Bz_~jSDE0Qlkpr2x-c-TA;8|MnKoQ9>vTo_;1n{dd`Y~h@CeRn5lVZucum%j?m{a81;K7TbPrX{gkoevnh=H*8aO??9u|R+Brjj9V`-JLdY)V zC8MT>cu$HJDTc1m!}=(M{}1CGd}mP$xe@r*3XI%J+MFSHklSA7*b26jFK(`=j1%vO zXN7s@BlZKOA68mBWB5 zyMYBcr3-WO@$|-oZ1Wc7@mOuMeGAY}YW3%X%tz7Z3bhlF zGhqSJ2gf@&{HcwV&wmdMig^IBsrzy8uX_bg^8W-5d=Xzw$Fe&R`~`l(-9K{@jhh@t z_gR)&X0G130IL%T7=V8(7??S-hmiYBOm_|B;~V)-QKjKA;Dlm~m-E5?2Y5ewQ7C)A zOlf?ixlwdM(Rn6Z@pt;IIYzM?aUT-)yHecSx%=OoPs0}#dMllaKAB*mZ;fw8PFj#+ zYNVc}S{EDSl>#S>bx%A^Pc@QQZX*m-t?tB;gcPJ?XUD5JlncOP233Yd>n`@cS(~ga zUu!u(+yCI()RzR=)?PA7Io!N+3-qLXDTpITxPvM&C3sHe0T!RLVU-IqM?rgM>ZPo z5GQi7TdX{N{bbca8E-)M8Ql921pq4o6FnbKT?&Ub6L(ON58j)5*u2en+(sQT2O+Zr zInMecj&n?M$k%-!1e@Ce2fC|%#m%YUNv9FQc7#k=E*ZZNQEwj>I%vRJKUmq+z--{y zTG(^kL>G+QQY>e;rFV7AwD;_qNg>NZYn*I$lV@mSKYC(WHoe8~4JBrIdw)9}hB0xs-WczKW_fXUxrJ(oQ)wUHxS3Enz5W<}_ zfB5{y!+MnfM8@v-xB;a+_<{ds_mS5jG(jif&l)iXY>V?@7R1@!257IF%t$|J zPmRW0F@%y${8VgO1aD3ysm>$|yQdw%{JXYr%bP~p2cx%S3 zWW%Gl00`n0mHO53i> zK=M;P4H96VodMZNtFvw9KwxN=8*X(CSjCL}X>ojis+$`-fbYkFO_z+@{W+31Z^diU z90;h(BGk&@7RoCg0-fAG%iz5SwZcsl|I+1h9y6apiB~V6T#}_H_wKAC8Emo4=rG#N`U=UN~;xmC19*IA(yux=^W0Y@vs z=v@(eBhuuBB-@Vc^dAE^TxwQKVGX&g!LuW!!Lf(8cr`|cX*46YB?Q@Mk%Zle><-jQ zhc8a|Yg<0zqa^4^KeKc8uogH3?NA)ZCZIqD*Ki7S1D`$Y`MZ9lcwteZ-of4Mn)K z7J!0ocl-XJ__Kc=ki~c|!<@%Ic;5)Bkk|5jNNV$dntPZiE@w@@YvF*GlOxB{UZCNw zHffT*g8ya-lBOrw>UHM^up+|$|Ev6pzS5KPuWs#5&z7#J{0U|vvN;EZIAm-;Qxj_- z%__F5^f;+9Fo>XyJPT5~&rsm_qFPVJioX_9G$b(Y?uQb_xUf+NtOzv zozwBu0ao(%D*L`iJcD-=8S#lNuLeQ;j&vWv0k#)Dd62nw;(>CZl3l?Lz+HA9H8{P) zeUk+j%Lx1zC%9%M%$0;E&?rwY_(oH|&j|uYRoppiTF(L;GVq(`0YzG<>4T_INTy0Z z;Z%lIh#bz0!_jTo7r0rKo7;Z}T^E`ssGCq7XZ_H}Lw+T+n z(wA5tQ*Gx?o=xD#R{@mOV`X>MtR;5weAwOgyL$%zUNxz6Mimf>UKx^w!3GHrtyL2k)$U)xmm^*1W!H$>B=0_OkHnAH%(AejOZTFeMXG&t zf~M*jJs?-uP^0uL>^{G~_5wDIE9gd40@{XYchc+W6mh9l!2H=;#&iViul~Fgnjx={ z;pGmmb)9}>211gG9_u@A$w`ev=|*I~6&TK>Enrwd3z(+76n zds6oSIb{%7$OY7@Xm*R3~V zk*tzv{hK0o0zFRG@Wrm~Ie;ba)!yv|sIF6yR|SkYOc`2Vuu){^rBef%J9ob`kKw^1CT^3&~Chp<$7YvVY#P<_CGV(aYkUT-Pcq zv-0rSTHo^9w02<8X8kB2{Ac?Ws?Pte^mIch1?ITyxT4k!W!S~QwI_=Dh{?2BjCVY# zkYWqnEcAy$R=kHvzP*k=UUe9IZ6y`4&d_}Y99NL0c2wRKx*)rhIl*hEm*HUDZD%8; zZu~tH1Qqg9mb;fb2fkZ0p8kHIDMqu~cfv#-Ak_VoiRXjI0vhiE@WYt)8sP+Q{n9I) zw5eGZk*8eSYK{KC|D}sP3m+v188Dq;sbw10#ZNRM#H3=I6Gc?ipWaA&YB`aqPKv1q z_40k{m51H38NKHY`(xz;GiQmo-G+g5Lj#|y7i@>WkiqV4Ad~s^>PF$1%;g{V*>w%9 zcVjMc!~`QOZxVgWz~Cj#*xUjJGqlQpPMxc1W(*-i9AO0EN%4I1h#o= z-q{NA)>rYDLkL^|#8n;N`BJe5^-01^v zy2Pl%%h(|6K*#iL8o)105Hdc!OI$=C@veU?SQ&f5$1s>HH{ws*sqGHxei5#J-L(fL5SJ zL+eL>*HEeq|5ZC(^veThsr8n5RPVH8D~=bE$%`-tgTGt^;>3Ez^b2i( ztTTF=B-Dnl>R8F{a)F!TPZSx5aTvT^t_4M!+b%?odM;#@L*kFL))Fzy8f7||TXthD z&>{l-HFTc|LRxsPUpKv4Hrsh4QLnB8-fLb-jAdAk7&&v@-*+p5hT6+)4Z+AY$ETh= zT(z4F}uxkx+K*EXXoTV%Y)7y>VYQ( z!TTODp^jOO*^c?9V(vi;iL$F8a#tw@A7Oc8Ilf7$C8f#yWtwXcwYfsu5-^#+(2hv& zhr8}vi61Yl6uM64todc*Dm?pZ6vs+E3rx%xS%bG&P&((#fsBvjmB!2gBoGbYy5}_0 zyc4$vD*G;=^3@0GsU@Ifaih9Waku>#*@zL&O2eV+>JCtF5~$CjjQ6$Oc>;A8HlBI2O_04Bt&P{ z!Uxwiz_8CnH2HM*w>U#*!W>uNstumhU~WdpubR%ej|$7W*8hZ$aJx@$9^bF#TlCk7DL%;RcYZZ|&p1xc>D7mNa<%%%=^7%K~e#F!`vvqfq^ z6Xgs$p-&k%G!|$YFXJJiLO%C`E@z$t2jrfqeblJhreM#~`?}sF*0GyOispx>Qy!SD zQseg1DSe&~!i}|%>Q+1-#-A=%QDL@-?`Reu;Nu1K>bq_YJ0m(5q-$&1SNt8H6Od3< z1MXkGnW|)JUg=+d*i{^gWWBgq8Y+GnKH_#rQg%{iE zzY~HJ1_<_rmmQf~qE@C9nzJZ=SK;PHKpD(FEmk<-sMIHzqkn7;FH0JZ4K%Y{)jur4 z^#=|Z8^|uH($rJROm0D84WSr#>pX_C#DOD}GDYUi%1l#(BwgpNq*>Dt==S&C8rZ-O z=Cqq#ZEgsA(#IC{UX`HopW*DNY9@^z70Bi#{)Vg2O6GD!i%G`P3Y%VvzG4*C_>>z$ zDly`p<5J_{KLJAGG#n~={V%D(C?(N(!K1x!+?GH5_Yhjrs@uvR?jvu&*$>*r zNh+AaZ72d?xQ04Pg)Flqi*EAt*j+*M0!k)3>z`M6NkjHK<IRi=agFIKgX@rp#t~wmt9^(ElG^b3gZ*_mSfh=v2?JNz|zPndkZ&>~D z=5n;7vz|l@p6^rkIh?rrZ1lzJmp^UhVl96PX;ZUGbiM|CQ7hlahD!tNWJ#b+#LDng z=-=o~YD6}yT_oO-O;qVASY99zmte!-nG~>UTQl+#)=*Ty1siio!TEG&Qz}lvkLd^M zDZIq%PBZioO=Jd#>hQ@njdeEGWHO_TKleUoea_xYA3T;?Q}!u+{}HF-STH*(iPLr? z+$0;tfz1_yZg2y5>)b~+ol&ncaer1setqsxv-(sA))aGl%SYefj=30$3AB`d(hQlH z5merE7U;QYgZy=vWWyurAzrEuSC>)`3!iHk2%r*Q`NE&%fP1z4iKKAz4&1#b>WV)DKCpgH7!NdIj5>Bt&MQ3jTJn20 zce^}iz~*`CvgKeC3x5cbXk1v*`~M%%nqpYKr(iR~baWenW%mD>Y1WmRLF}71Ny+bbVbw9uCI`|RSx+qkb(!&J z?g4pBkO4h%y}KclckD1=m1DNNj=4PdvCNJVYO6&y=h)86EuVRM)=DPluo!!9^9va! z1T(lgYp)Wd?C_D!(%8li`+-0xPHim=CgV8#LjQ<04W4xsiOD5_${8=H7=o@US&bf- zrms4`-Dqm{o3XJJDvFC)9Er@Tjhs)LK}CsUB?}l<5s$Y!zeqs*>4D}|a>3h?)&EWv z`K(j5&2Q$|V9lYEtu*eAoJ-;cr91;_UEvDt zs7ODXWfygKEgA`x?lz|A>Qv^`~Ip@r|PW5sMHp#uS$3r`10O^q7+iZcf| zXxXpc%*pkm|4QzJzw_Xm-$+FxN@*fuXG3B&4wg6z13kJ;2s*I^ft8Xcp;u&>a{bG3 zjSk-el}b{*Hn^SRB9oH89j2HT1Q*THZ?sBl9`CQ^*>G51%u-3faOiHp5(zOBS&tg+ zdg+Q$FNF#l_I{2yjJ)Rh`aDRJjpoD2Yb*?NI0) ztfYLtJz~<=z=uJt8t=L^8keIub%SY7jEUpqC3cg#I2_yjyLHx&h9h@%#GwFd)An047*to^iR>npo`6Viq+m;6%iq}!( zO!q&zif8L90uLJlZ0A+bECsLi!c%SvioX@eRZ?_d$&XH5(^9L{6DCS7lMgi1>cgHU zkT%E7ca_%9Ww@+ww%=u)Y#IlilKsS7j4m5ooa`egU1QYDHWSF@(!G9KP;dO2yJqh1$84x@Xm!au|Zq_d=4GA|-Qem1-J1S!n0HWZ% z&dL3-!&;z+(2U|NQvlhmva3RE#)|fWqLChEY?dVA?>jcBCHLPL~~CxQn=@IvN|Koo4C3J6ln&(BWtR5_0?MVL!R zg!4qq04{JY4}zCRH=)#|vRyNYa>$3O=`5<+P#h)2FOwL8DoJNr!am;1bR@(` z;kF?*n6ad0grADWC!lO#&Z~&?5csWnVb>ig`{Hkf#K#OdEz4$)^=6?G_%^mLbVZlE zfGv=$6Fg^grK&Yk)D76+{=_uMC@Wt?FR2n%xg;fJOgoPA;!N>5ygdnJLZHV+Q(@o% z$-=Rit_82J{X^Ea7sJyfMaMxc1PjN3d5ux+VN`B`r7{s?-5NIOi}u@M6%f%Qp0s%J z#*P=@6t-N~l6HCw<`Xcx(5S0Ic(#v0D{oGawgvDrU7tvq38f{7UllmdR{2Zy)rkH% zEXQ_lFGo+w z;!ivL^yJ`*Teg047aXLRk?r!JLZSF+5)#S!krc-A)kX~QBIZs{zgd(zb-SrC>CKE( zYuyE2i~f@}Fhh5b9$>Vavlz^MCwu z0P=T6gi>sFm02x;O>#Pny)%4beNjwo8D$7L_8vNqPoCbALUA>0TIa6fTY_M>%GFy@ z2l=i(s90LBWYPxMGf(jRqVF)syv_~-IomqtfGS~BpKE=zB%{o-X+pT;7>~m7%U`J%68Gbb{}0Van_GgSJyYgnzmYaqyS46RBl{174UM+?U#cvF(U*lxcwl2zZWTf!%M3 zd;!Ja=ERwC^|@#7S&{8Vgf}4&xe%_8mly0*$AQt8!0*7-mg27(D* zk5rhKDw435nK(shh_I0_pz8O(_u&1@Xey;=YMx27YSm2mo2?s72}`h&&DnJTP)QeG zL4m;MfPNUj3jsX&h# z%h%UUqydea-9j;eh?u@~Vb6k)8gUyS4HP+SF>S|kFD+~mN5G-n-*mXslAxduTmyPo zzK^&b8&MNR_U*d@h@h9YGX3GTM$6?bHrxFJ zwL>zrAGw{5jq#nf3r}@=^{(O-N3yJtzjY@5_b&dat^AH!SGprI07U&*X2}B;UEY;{ zX^RKz`4APZmFaTa?B`$MGV3fex4C?fSC~E}F7KmLeV%}+#CUiX#5cFF1u{&1!WCJR zwd>8+JSIo+2rL>0u%}n2h{ zatzymx+;XN?B-fQasdS4xD0mL^N}UT-@2@=P1}96)bWv6_f%EKptifpZiQS=={+}3 zbV$-SD@(UGa^3iFTZk()h1XbnH9C=&B;E{>tnU_0T;JgA51^D`_X`7ubU@xGF8`u2 zAdk?Li+tM;y&fQl$S3(l>L!4&1j5)UeR>4`J}Yj(+q({8KMY5idsXjV=KJ|T^-FlL z8ga~H7uV7E%ceIXT7It+)L$v+Yzv9r)QkWyfrlks7Eds6d#u~UOHTB8U{9*G{zS>+m~; z3a%&0&Nwn#c%_p2t?9;cpC5Mn02Uwi!9xps{BSSOukMrgU0;P=@Pu$th)XmA4Fld) z0fo1(H0L~@1tZgZ00n*$Xaf|0n&%h}d>3?_`dL}&l9?~7+d^OB(5uPAh7cYVz_qU5Gy>Qm{Y=eOzcb}8Bm*||H(~d)li#kUGkB$u{tx?Y z;R&9}Z57XstJM05ah=P~A>Jvgmw5?|$+iv`BQf*C3%7hQg|~BzS#`LVI=*!!OxHn7 z2dE=ksxel0Iy!Lw^~}D)wvf@hdnFJ^#A>5ct$C;aSL~w?s5eHOJeMm4Saiu1s3SzP z$}DVRR39a;4IFXW)=VES3`2|k-iblTtG4KoHo1c26khwrevW3TG7=;Y1HT~ zHu5a16>uw2^n|p7fB-kkLcNV^@ZGQ*VnLFLi`dnr@~X#tGOZ&G0t3l>(Sd0~hBFK% zHQ3Ms3H$hgEXi)x9a<8}=i&weToB{uM{hoW}1~-=U<$Zjm3e7CGp~zNK$(tR)t31OL9d*OoqYk z1fP2$e!&)qQPKcR!mMmv=XH#73@RxKgm4iCQpv1@nm|0TKchAz+1;Z|%XurcRI590 zyLD#s(5GUUvs5ZauF_Z^E+$)#WsNgmO@wGJO+bTg;yzjx^!67?0o?Ex}T)(0SfRGj+Nh{gDL@&YM$wnq?g?2*iMU^&&V|65mB$G zp*oF1V7%K$#tFxjr%d{f>7xSrimp-~JwCd8Xtvq^XMSVSlz)~V%SO+^984! zu5ex%&FN}#iVFY7ubw!=KP8z2z=4hT2jIQ1EDF2>loK+TEeF5UZm)Lk*i46ux5fVi z9i*JHTvkQva`Ba~lKYx7o{qb9>bqh%kj{#+B92KGl+g*042N#I!pb?*2p!8^=3Uxd zMm-7o#YMc(A2ch5!T6zdbTtz5V6bpXtdd?`_>$$~Ffk@I{vu_>0&Bs=Y#Nv9J#;?% zz?g7i1_dlS#g|LUd>gjQo)%g(;22RVSk;%+d!65?d6>Y|h0w))kpeYJz7Q;{>2IPa z8zRz*avxv&7W_!Y$+N%HQ*r2Us^gv25yKqr++rgj13$2EuqR3UW4S2gw|FXZk&-b< z+rlhR0!`iRV!2G$XjCA})QQ~yLpD8bAjN)W?ssU+V~VsIpjknBegOQ;6XykFjUL~m zD^y>)<>a{OqcImvU)tfRDvH(mf^?EEGsShY)m##{s7|jgsk(Ir^7S#GKL2x#t>~>` zWBzBBx7QLa?o*#lGe8+G?TyuMytz&Q6+Uuaf9Ds$dCUDpUk*)K9QW}~@27&T7W7$R znRRZs>%h*njewubbynmUrp0HUXvTnKJ~MiBC(`-12D=77d~K@~@2xQ*O{a&g zQ2N|ieMmHm_6Ci}&Y19qkk9bp<60S-U9X=GW6*~vAETuv+kY5_;+Jb(`U5r|%WPV} zg7;;dt(rIkuW3dg!a?FtD@mbo0WudxFco=!xh?-_K{dg&mgSJs|9k2>3&rX7rAcZ) zbR7?Ae5=jgj!pKgEW5+uWt$tj!>bpt5knCreqVrRX3O_o9w%D;6#>7hlFmc3!&0Hf zw_QRm5qFY*Gjypd3-ha{0$mJW0_eIh9pPfc=pHo;!$%chB}=zwGj0!^VLadL@O)C2 zGQeI55F5MO@fci^RY1St4e&)rN~Z~Jngt~Xfx}$M_}EW>q|cImlb+y`aZoS$j2JNIr&)pu|CIh$x|GO(AO!Cp0s@Sme{K_mx1@SbS^; zEwmhh*NHU&VG<}cRu@(h0+oUl&;dyLL%aSkX0@YWC0ssL(4QDVO^F z^-+lAXTTmP}P)A|DS?>6bm$M2v*j&mp>e3mi-Ln^}~FB9tQwfuOo(1Yd7 z`K@jKU}}USqPk>;WEN;w-WA!dynGhy939YqkN59u8P@*IY~F$QRQ^A&#niHAtSOZT zN{S?y@J3P+%O50D5{mP*Hu%MU=nc7al5hz;Z!(h5oCG}X*o!xbgmT?6-hWuf=RvDP zDZYl&Bw1_j|G;?=nh26MDZk6#Eoc41OuO?dN!mtt7N80T(xA1$A*0Xsc1@f3$qI3pJ@3-#|Kj!DuJKLjwwsiqPb?0)mt}*jVZu_SaYdMSQ_Cfskvxtzn z!QVC!c2u|yww1l{fc=A>_gU+z|Fm!eritHvIJ{rSzwFrRH4G&yCuR zyV0RHIoqG|=vTVl-hC#noG;w(!CBwkNAYk=S5-ur3+bm_oJ)(YipY0_eBydcAH>%a zXk;lRvO9v%B|LJ}I5T5bXd*?JpmiTpO#A;1bAKaa#c}hzlg@BT%TK3Wm-o3jw8MKB zbXGsTZrq$oh^0Cr$9x(dG5~a^89Wk`TL*III_5@iGL*^qO(}D^8%NC|GIZ9x^9I51 z10cgpb9+m8IA%e)@o;Whs~ryQKy?C6$(#cL>uj@+;>BqQDK2fXw0le!Ei~+olmCm)vt8`Gnoir9a0_|~lDZ?C z|EE$nVt1c~d8f7&XRqzcp{A9&0VS>>)P_KH87nRClZm6^MkbxYy;hlR7gUn$%@ZSO z8D6rrA!2)V8nJJm(dvEt@z-t%Q6CeBbI!;r%V%aM7s-Lw1UcL0pcY;$WIL!ov}U*N z8d2y$xOOk1rqiIlXMS)26OuH9cxQ=mvYDT|Iq>A!$z4>ed!A!vsq>c%jmVA~H4`I# z)Nt$U5qt_PU>GkXiq<1D9e#!_=7rwykG%^ zue1#Q-<{&AEWVAZ)>F%@)R~|3J~i$N6X&svHno4%ZBXf8W6XuwQR&*`Is?d!|H)s6 z%t8Z`76~W=3#Qq#@v=Y~Dyu>`>a83&=zmHQ@a=r8vl1v*(v#aeH29ySQDX(= z>fhM8`AZJQK88kzTht~B!MxX@W6k%fihw0titT@NP;@ zjF`;y?$WLZ#6RQM|Mc@faTWGj7~HH%IKav9FcF`|Q+?SlrT<-T{9}S_|BDUX{=^j{ zlV^?}N(LrAN&?bjsi!hZ6u!SWgZJYprUZ9v$=P-{1Uac`%U7XHBj;yLA#rMHZz^KEI%2 zb#A*j$s-+9b@W^NUEVr>z zIjjYY_WBK$U@BzdePMNulE3hDYx8BexY4$H-`S&BMkyI7~^}DhwbvdT91pP;=&Hz|O(z(l||DN*p6z0i2M5lb zc;Ov5(j2kb1`Cy^oK>EpP^g7<L|Z2BCA%UqT)OhY@14#6^Q~W_Jk(1YKIFUA z*w*8!bp9=}oKbLQwXgUTuQ0xqMweKnkgbiy9bhf{KomX+U>mJJDh@oISlF~m09WMw zSa&1ZWzn<}XcWd?s*~*DfmAkvM&fUZDHTivZIm$1de9=Hj}8G0s$T~YJ( z*Pp1UA=#ENmmLke@g_?+$GYZFAAp0MHeFvGk01^!kILvRyd4{5w z9+VDT=KU4#!(Mrx#=8wzs;>fxC>~T9G>}_lIQ~5p>Iw#NYgI{eQy7RDx3R=7s&Czw zwr;(!fTiCy5UO%R~4owo=ArLQ1HnA}_!;l!EMC2C(!o zJy@SE0}y`AfXpNojC8TXgG0xwK)?PZ=8z-NB$mbWu8&*eR!kF6|F0jlF2`@b(YxB> zI$dA44Gcb7)&Oo*1E{W(9~#!PJBNMT6$6SS8@p-W9@G%FZQ#k{csr^5plzLT8N9Q{ z1P}-WkN=2|uF|>#?5`UW5_|?{=Mrawa=Lg@HmaBKv@JZf5I^aA4IoLbU^Ld*_PgTz zCo_!i3h?eFd$oig7H{HKQjYTQRQhbG-seQRx9Ek|(gTE&@NCw`gSE+Vi-kxMYM(`w zbG?Swl;aZ5$h-~mmxc;9kXMd(e?ozjz9e48ihMW*uT>Z?Ujqj1>s&AXAQ7K7wyI*9 z{eg1c=Rp4drmwzMc(DoIyidQzSPpT=IYUEspbir`U5d4)F^sS{%JELlS3G1Gu$^y@ z-k)$TuXCWzFOQ+({Do!)9(ShS@BsSI8;CF7g?8Ql-m?s5s~`-12*w1kV1?qR4&Y}V z-;qnGid^>qN|gybEDDzIl?6~%$OIj3+BqvBhgcGxKo%s=V5_~E&jO>+tYuop%5QzN zMUN<!uWlO_MS z%voMD5p;h^7k4}ZBQbtV_|_*DPafSUZ~XQv?xOHw#iVP4oO)diad>Ps>bpct1k1sb z>D1Qd6p#6F^gUWhGyQrkI%b75nF>Nb%;m^T_mZ9c$h6a1P>ZkiDev#8ZcXG;pn5Nd;Uc$Oa=wfm(yu&{jrup?ElFDi9mGkEu`4>aIAoLCvmkU{GDZSeAUYoSJLfIeG_$qGqw=)gacoil6 zoA=PKig!|}m;Z3*;q}n6xE6z8Wv!;sj)rT$lEsy_&N-#tLA_5qJjJddCzzZ|`u}X6(mML86c!l5X3@VGl*rcOh3VyEY zqTPZW3eYcs3>Y)(jtHe5p$Q%~-l68QI8dd~?yc*Hu|33nc@4ozge9#TVNmMhA25pV zGj1~*RNYH#TkMXcrT#e;#ByP0I0Sg3O`IOgZctqGH;UY1BQ|9j;vqfk8yn|rp(P=f zF=>$0Y{cp740kxB1>v443^NPuNi%7)hAN%OT1OCPuUh zA93s#<QY&{_W94t=nXOf-H&W;cNScQO=bBZqaPlq47Tix6RtCiB<$q00jmkWS7 z>VxYt=0Ln|FCQhs}bU%b}eTU}7<$u2w!k(*O(4 zblJgqEX_Kb(z%d^^1=n9%LHvc0c~F%0pn=9c~iA~!W+pDfY_+y6Rr^ax-`VJ9C7cB zo`>0`3C1CxuatY?GqFCcJ4$;R!7sV1jxIsNt6{C;&)HV!W!acOm}4D!*q6!GalP7g z-Q$@M)X|c#@a5g~XS_o_A&jv=R$CXy6A<9s^6DL9=GXW)u^+lRXFy-$RxsXG&ZvY@nor7V-%DB1I9hn@7Xx0Q1?6=+oNcVxD*B%?EsQgO;+Kz<4ul%A< zwj@+=(A_OCdILMCbPNZ+EF-Im8~(|tt4d%b-Siw?c+{;=S-g7IVNtDN+K4-m)_2yP zl_VNbK^(t@uZNx=gyyNTtIYywkQ0!D<(hLb^NV~=+3xVWgbAd1feT~$H-R?y*m>{`M3Nin_JI|o9? z3I5+#HlAJd-ygrF>G@1~G0Cy#W1^95oz>MZXtRdAR8e3q%=I~9CFR7Z+hzKP^F{o| z`;-Ls*sGTXWu}G^j~i`XkaK=(xs94%ymmmHug@Y1-o@Pi?!OBibpCXu!EZ?@it#!w zL0VeA{xgmoj4@k+GlbX7^V$7ib0IPLW|zNGi`u(2?HM$RW zL}A%H>w|p4669V4aLphEVYq{4eivA|YNF9O z9U##6)=*?6)9kH$tvdkg7YkgIMmgkHI)?T42cpEG{WBqxg%_&;Sp=3Ql%t;mt8p|6 z;uvDsUHB*wqQNi^>*T?(1Ad;^;nI9w&WKGcCk%h|WEFdMdgtf}F4REpt8>yZU!s(R zzEjsA+wY^ljis)nFruK130sa(38!TijdvvIERaRKK23prvt>dgcPF^8NN-6*ktp+6 z{H)&0JK=xBGsb&(VM&m)Gu%WisQHm&@vCm~WUj&RR*gq2qNBtWZ0$U zbblRY$a*(~<~uP2YYH;s1ON_Bbqt{<1EMg6TCB(V8)o)GwyLN(GzXvKH_@@ zj})p!Cm3V{DF{u^vNs+z-0W35%OcW0UdUb~qFPIF&ZxRF+Zle?ExsD7>fF;48N$(_ zJHhYI`>Ndei-A0A3WiH9tqo#>Cx#AK>Ll|dA)mMKu*wxKi?C&QD@}96zUq$GNM{dcwPv+mD)Q{ z!0ze>0OfBxx0;maClVHQProg1mo(rgJo#E!d%_k&8bS-ZKA6Z6-*h z(GSy!X8C7cde}MCQD>Q9S$0|G>orlE8;XK+S!pWfy}v@bS}PY_p~wWWx%Oc?mxM~X zMU5Ijy&2D|=mM4($u%FD2ED2%P{jSql1#Am{vt0b>}Kbmh6 zl6mm)OU8~`Ob%_NhN}u*R=`D}LPPr2u#+}4ve3ZKZ(G!z=CXK&8jmVXr#2Z704vl| z%YK3{saZf7x{!8+aljIm+u+tQD9_d?UeL%P$O)v0Y5qT+&cd(B_6_?I5>gvEYNUkJ z2+1*Ol#-%INlHBe(%m^=bax8)h)9WuGy|j=_(&;@fPk>k>3j2g-_QF8>;sGYy3X@F z&f|Mj@<^q=`N>IXgNpWR?o!Y4nNU1!H68C%w6#WBYDqc|SH-9d11lX{*gcV5_Llf-Q7kAqoE}UVi;6+}Xz|;Envn5vSB58$RKt_S z4QWEYO`aD42NEfsTuGHNjn#T!RdHL)WMZQxx|BV2e^NP>6QB4Fdat9|c7+sm@g( zjvN>4tDG;(s4M|I#vjPIejbni6jXqJl{a)a_GHip^YN>rfNsq4jaosuQhJ*8HZatC z38+V3Rh7&e)2zJ#o^a$`CF_hjO|jvUrOr}h8pmL*75m?$PhI7gm0N|Z6<&?N&H}xM zM=hMh7`L=+llP5>69`FQvLVu@S48P;}o&T!pX~@5yONGkU zrb4l=#kVT1C&PZJoR=w-+n1|Jm@?TtqvUnY3C&9LF5Vu*N@FFk*OT^gG@I{;-_LDc z`Wc}O$!@1z!DkIhCFo0lx_AqQGv<25m?sh-Bn?Qz!YACbz*aI8>-A!0Z&dzRSY_e8 z2;@s?&Ille{*U$j28t#IA5U^uG}I{`XR4p~SoZ_rEL(_=yo@)q$zZOMEB2%$*Atf} zd+fE)@97t8?U%^E|Hf=mL{qqF0M9%$8>E=a70=d8hUwT0xxRfxX?7xOCPB zlc8D)C2SKE{3nA>6n%*C6;fW8zav&RtCT7JTD%-T3c*r>G{bSej{<`gHvCQ=vz~ak zO@+@!iqqG->;O?Ufa9-tdSkEwvqqIMmC6WjOp~F!I4d}^%Z-A>!iX^HfnRyHd6zKi zh$b2awkQvV;(1Wo6@}SUuLto|Tm)%^H`lKxH@j;iB@yfa_eIfNFp{HfF8$q!PK@q3 zPWv2JpK+}7)dSLQQ>pcZQj=o(oB5J&&#T2Az@tJemi^t{ zV0i&aHk{o!7@mH`9cZMno`L5pV%uL3ajtxy+UfFRoyKsC)jyqI3~A^~Y{f+E@Qa6y zF0Mpe_p)-#!)-Xh^O)o_+k}hNaNJHTtD=1(0rx6LQ;E_^L(1o1yB{L~l?PZzy z)YqizQTm~zCpZI{qrVm@nKWTW)RhyTL`tE8nyb7WSH*6Z5Yt8rnY~0;j7pvt+a78s zcS*o}Nhy^YS|0y_zQ*WxWNVGgaFTagvG*bp0;to%7Mefa0TK}Ws7W0FfaMO-(j^XR z&^$9VVN?wNUco^87nEdAP>TDJIy4-tyeoM|_>}pF^ z)qD41R+;PfZ28~vwg0-ZsBWL(R^S;v5a#0P83y*muq-f*cL21nxzlND&zK?LWKI@l zn)|uIDGu=I!PqQ7imFJpQKwNhecrnN21!4dmTHUtv8x?jw?W6k$d1G=#NjKSOO{J; zThq)mlbZ&wAspu(DqEow_b9f6Apew!myuUK|3~?<)b*?(ka4`nR|#?dXJ|hup&40M zSxjkJsti6OdE}FrQW@jVm`xMB3UjLTR7_}}UrU8yDzM)AimcfNpDY^r2U!StQVf8w z^`dNgg=R=j<%hvP=|!SW^yI|nGWJ5*`VVO*rnX87IXcp=(h*W7&Cfzl-+92C85DEi zwD{6u0m(n)I=*~i#H(B>nTi|nE$WDlx&qQ;q}*jAx4}s!k@F@(AO5XkCbQ8%2F*WL z;$12ZrtzRv@<_Ej!L^BO#GU-d#-p-Y+KJV4Uec*Nwj^|3LX^hfhCa7f=_qaN|AB@KDEikvSdw}CUkgjsMV$?6+jCmAc#nOFGHeOKf74D!~R9Q9E zDt?xKe8R8XhNXDxw>a(2Z%$IS^+$mU-`}B9ct!h-2Zp3`P zr8Qf?YAXJhPUm@&ozq&g&~V<_SB?!dfz7K~NufBXQRWxwO zd~;hatPI`+UpSK1331I67GwQmfx%MNviGu6=|?FRgEG?B7Oud6<7kprak}A~`{oU{ z^j=4~2I*7XZZEM3FZdU6#z|npkq=xWnG7VpAKiE%$NhS_V5g=++$j*COPdtp zI>ifj?r(~#z%`ZiKmGD+yu0ZC)*9dJfJHbv9+I0FXl{+riG68`iAFnSkTY|w14r$H z7Kap1f)vhJCti+L-kNucmQnt(Tj8bM;U!0t*`R-dSs+(lMEyInx-vi4PwcdlkI{I;&vHL8F$BJC;~jf|%%f$B{2zPYL6he}gOamq z?kXhou$O&CaYH>|^*rs&DQvW|>gK?E&<{F0^oc1cKIO=O#N~B-8mIj*RjPl!1dOL+ zcf!SVDrz=i1?r=0J!e@*H&1kauqqCeN0-p+kCejEuF1*X`XlgV!>kaANS zj)X>xWbTDg-tJK1z-+~`9~1IB!nes@yQ}3PsnGGba1HiJQD@RjF{o&K*XOg&yq5lt zj}(B3yh35M(rye9BW1yveXpqeKP{;R>E9W=RYJUPMW%8O#^Iv2~;y?iD0 zq(dpNCK-wW#|Q4ZeD6TrglwBT82y12gz6vNQF>4uGcVZo$!|%pINmJv7<(gGF3X3s zvG~@bi)Qhbf_1mnhaBxj3H+xg_H;NV%&@yjU?K;&fZ>mULIe-|7Mn?qU-CkJR=JoR z$>LOx{5@x0IE0-i&Q+ugglYB}uojrvulF;pEn)Ab>H7cJo@=^z*aL^~+XeT@mM}d> zfvf$Ps&Uaj9P_`N0iU5@g4ySo#oV7!LtXGEUgm&TsSZUF;)8P2Edm2T#!5e zVrG0_;2%7H&>h$5MpkCrL;~V>$G5otg*syMQ@9fmEaU@hX>Mrr<=% zH?HHcT)L1{5B}5UU%+C#_P1|TxU-?w@Pm^@&EV(so;ivmiWH=hZy#TE# zxP@Ea#HV7&f9A4fBRCuot2+5 zG&%H_Ll=p_XN+Q16?)RrDpJ2b#b86wYvnj+spX^`_P0zBn9N-n2@Em1+`m<|R;5%K zgma%(lT~e9wV~vZm*+rdgYLs9`%FM-gcSlvrV~vvPEi_eW=Uc2k1b62wH$Sn+Ny1~ z=)e+_=Bh7mBNr1@tH8m}d`08RnrOl`yn z41ER*cEA#p!)-VQFW)#0@gGOjkH`)2`)p7tDWV=-B{%Q_cZOYiOwIZZwweM=z5)i5 ztt5I8*=xWtD%Vf=o4KZt@b#n%yK@$dJP>wU^+&GdJ~!1MFbSaR$&*&Oa5DPFyj>Ow zz9(M0u0rJzvj!>xaa>EVEtaM!=ZlZ`SySPXWl<@wdaOt(HT=mRJiXwT%QOPbWJF!|IFFo`<$9$Lb5~vh&xP2t3$v_R*K>5S(i21wYM|M$x2RlffX_ z`)2mvKdkJ6xxWRbO?NdDtRk-}Ts~u zNH;VBZ_QFOzv44m1z7R=flg{iMBCl~?x{u`*gKKqkv=Fm^rtUIL90pU)zfQh~5w$;H=yG7M44Yi9uy>N4qXv11~pp{=0ZDAFE==?XhOb(=>nrLAu2 znRR0mO1EOY6D15(2;(Y-14C5@eT@{zR@}KzFBk9+cSMet@ihvJ9ki?u-h3bZ)f2Uew5?7`$nL z=?e|sgjgku^uID}{!&FrqD2vyRicGv>=*yW9>6!&OfbyA&hjF+?W-k`asbv>91Tn{ z>9#qcR!qvQ%1Y}N7D1KMtQf-q#P!v$1zLi=9B&WGz0AgnKi9-A&AQFzVPmvR5fZvq zYlSJk8ih{YTYq;;-dMO&gcL}3hwAWFjZJq7yq-c9YtapsEqIC!!`@j@+$!()Y>u;^ zVLAWtkt#LIX`kMi{8?*M;@|5hE;I~meX#(8VIGh*1T9|w{+WK{gwpGinXddK76yut zu?Pd9`FAE@H$!@w45_g(?cG^s!nH z5}Q#qcF9u50gg+X11|t%t*~z$TE(}iJi1&+_GmY5cF>nHrM)zKxw}@20CohX-PFdW6bgez3_K5VfX2k{AiA@1#MZ`XtAW9a+MT{hG zX`opkSc#P=6^2EtE~~=&AXGS_w6wkE5Ip?@^JqAR9%%sI0?TMvL$uR0fM$xb#D(Vr z*;BH73kCnreW~ci<#||uog{NhCY&cJSL5?i{}AX8GN0PN;|QapVejz>%t`z20@Y@sw5`XSQM0Zb~ki0HTq)9*=+H0{*B) zL6Y74{CBRjYDmVfC#9Vo5cNeTtL6Y=VF%00xvtS4_Ekb8QJIp?rODb#k&um84`^71 z)sk=Cvysqd-QQ6mf<{30LR@yWU)e#vE;Z|OQeT+i zR~9%n@7L4V?{HCNN(NNwf~&1;pYUCj7=ghznj1IIo6po;6r^ZWaZHwCPHyE z6ytdj^L%^sYuZz~mcY)(hnswr$v23Oo;UiEt}$L7A!kgKkWyRWc$|pGxin@}jQDwS zzd=?b-=E22`bb3L3FcVk%d*LoCac?6kp6pa{r}BQ0U<}~sOd+Lf7+v^;f~^E+22@P^{I|C_Vq0V8_}IR27|d>( zp0}-Eea+k9SH{hjAuo%xCeSR_=It8r!1Oe}sAlxZk{5C9Xv1h)rYB0@xKC`FgsOc| z{&G+%wme1|LQX~XGyY_iU$>mDQ~@@iJfWyMFEv!nHQZ9fWMn$?108PM1{s{N_!LGwdUuT_7Z*d3NcB= z5#ur9+-+1dgzs$J!-N>5AXAaaS|J9;=>Bu|u@vtztMZQWjC!%ns)_19T1xZiIC&M) zzHZ#Fg4e$D49<46!W-%qTd>;iy`O<{sn192#S=>>euSvE)RE*cwQgRcHC|{QsM6<* ztSXs#bbp6$DjgXQkakP|zB@)Ma&L@jY(l^4J{14R)aXxTyyu%6Y$k9kKaRMG6~9N~ z1PtXF>`isxnTqzR3p~&%fLq(WOnsejqI~m}VlTR-N6x9F+v6L}hKJ)fZ0Q&AP(G?H z=WlnHuzv)Mj*xX6!!f>?&MV!LAxZjt)=Q62DGg&9mBVA#sKAOkEGZT%p!vIqL0dNN zLt=KSn>ira{oVYX8Cd0lpJy!TboJ=C+@zf+_KAZ2S%B{)pP`}}%L<2>fV7s_EA027 z=YsOlVKV9$Atn9SafphK`$`+UKT`}_j%e0(QQxCqCHQ{jzvj69pY>z87|@QM5z33b zYpwrx#Ek6%c};ln>Qj${V!5MuHcOqCfAWTD=B7&V;Sp55nXkR%){077?vbIJPai~! z9;p8lV&jFn@L^S2EGO_2rex&d{!T$zvyBAJwdUa_)eoYa#s1-fSSxpj7xX9$a@NAO zPKm-ep@cO5kU=454eVz~3sv{!2x>4XR|F|M*->G$J3KyubU>|;226|%cXbBg!b#PT ztmupb%RzBAeHPHiGxi7yl|G8B%ESm6PI~U(#$XX{Zs*lth&Punm%D#YVK-VE2_BB_ zc|tZd@cAo$FI6x5=4mhaCX!4Qttwm&6?jFg0u78HrJmBxLoxa*lL1pVg0 zgKKgr?P+c|Rk~mfRXwj|e=&3_2jvN9uZ9LKe~#Ho%B$%Nts$D~KLJu(ED9Kxj-*Na-xS%;248FlbK?M(pR?DE|g$1t? zye_3nVy{w}rV?{?nK^$C|@1-C%y#%Cwg5$7GZfa^o|sd+{gT9|!z~QxK7HzuR{J06dZ*N`=#V_tyP+k`_0&EBHKpg5^{%KvBo#brMZM#O;s zlZA=$;Ff%DI&aH9!H?}^Z|j$QF1I*l`QHfr?M-Zmip@H~k2E}2ZtifZVNR0ENYd>N z#Txic{A(OSJMD5Zo5T-wo8vQL-N2gmk{y!$b?PWg8VJA3VM6yRCLOAIuD@J1N z6kM4kOAfKM*$O{ZTUA|9=07weIoIq`J*=eEk|aUv)MK{D$L@w?9vUH>G7E)><;gA} z8^5W9W^?wPqAL5V6aRZAAm&ZWXI<`GR#2rf!ZQYWR>oc~QtlHN1s@(^{Z%$y##Odp zV410upB2|!Az98|c3%4PATqvn>fIv{Q`>!c&}D}7>QO1Hp=f955NoU@W{tn>Np&Ti zZt_XgO%x;Y9F$M7aW9i%QlVtXGqZ~Z6TdjZ>--^tU@xIf<4ctY?c|ykjdR(q(gil9 zt?@9U4}0I4Cb(R^d}B&4Cyqhe#Gq_4T|{VmUb$Rn|ra@*?~uMpo~qLPe*w+=ylTz-v*~U6P~mUM(O_$ z84`Ms8xr&I{!WWfie!lLA~}Mrabl$G`}JN(Ks04-zCZA2KtGTq5&u^er7Ubcsdq)W z41SbrgZd=lP-2V1ED_h{IvZhP#2L6*apDx&hs&;LeKAi?i0vrG1EbCs%-tIh1t*3_ zgT)L8%4sjTA8|qgSIEKlGabica@Ef{mAQktM$NEK zDk;h+>|kQ!oLbzB=OLgHkbkaV9)o-Mp^+adyfEs}Qx6}F+MuG>R_LX|J>kogvmdAy z3xf~YNDUs3`o!0$DH0AEra;obOaUtEhtRVroc}~HWDyI69jQi_?}DSW=$h=$5dc&% z$HAf6ail0T+_M@EirtcgajGG!Y9B{j3>bj-(=c(H3dBN4D`NOmL#ngS}Kwtxx2Hu8F>y`C2p^V-Gj`l>pUjeO{ILO~*8!mLKCh_g&1K5BVh*H14NjcD9OrjkFWOe&24g^PSSM`=nrzX}0RKk@sQX)Fz2KbX~= z_5;DjcN%q%n15axPs9`28a9Wl4bO(B@{lbNSwf9ZTLoi9^vRdoAzh zPWbcp&cIC(ahhSTIR=^V+^t+J#I-NuR|K(}qg!l3BX`%4`r#8%CV62k(jzOkt_`~8m6D!_9Ee%m2`M0mG`DCA+E0`|JJ6LtO>F5U1LR8;?v@N6ay)-NI+bocX>{0L{gKVy4sq zW%6Ql3-SiW#~T{TPd!{LEBKmn61J#nv~pF+`hu7}QGXTrrEL5wk36&e(8mNE+q?ey zB#F~^{qDk-PoyiUs8aRk={7eI`pYt_njch|G}cxMo5?A=f%&%XqCi+Omq;dS$v3%c z%H#C~zp?v1)sM( z4#jpm(;daIX`U18^}cFCmwhHd0Oa;B!N3l~T^zPbR!p9~#D(ZW{r>Y-voWe z#g7VZma?OQ|F|+2v#C`4 zzEx~m3}4GVe~0g%4~5{LwYf7h?MOaW_DMu)F0I$UdCa&DUUeO@r9FJ~CGo+Gu@eMV)L4$pWgHx|~Y0wdk)M zm;fO1Csu!ttxkXhh&cp4Ek#y^sG)ZvSRG{>kOl0WPm2_(p8HR66E6uq22kmjsnUe@ zdh`C>rMo~ZTqx!l7Pmh0wA#KuXm6D!?;R-K^$43 zaK8$B$niU?kHDt$4EZ-;W1u+XddJFdL{QS`KZ<{w<)Xzfs`HG;nt4BgxTm_ksp500 zw2byNO=9Np&1p0$R{Wu=pe=tSr@dxDb-U9CvMf{xDmybozH6@i@Dq)xfKZr7V^QP! z1^M1%b}nd>_aML0x%P8LtJ^g7K8~AuI%hS5?DaI?h|)(2zg8lzCR>~+f~#Ew_=$t( z84aonr#~6eI9L3{LTmQnxbSwvxb>7^EL3ez6R6RsCd>XuEA_pqnOioh*uT`2sp@oK zuW*d@@uv@-cbR|T-~|P&vzuu2-(uE~Mho{ir>?q=v5Q^1mOj#og1y0VS?ULHA|JJ* z?oQ9_xrD8V5LfSHU+f{5?bDgPHRf9OPoaiFw{EQq5_T~b{DS%GGId_1Y_fFH$BbPE zk#k7nsrlXD_gvQKV+lR35X$vgkqi7 z**i#IWHsKkY&1%NM0>DZP!CvrqdOkPAw)Tgh^L13GxNm4ezR0)A4%htFR*%oep3VD z>!f`Ol;bek3Z3&JU(nR2g#Mkr^%AMl(=sf;DnCPd4b*2yl#Z2sup!$b@-R5fBriU# zd9_+0JlRa<6uB70yEK$Dvjb>E!c7$he4Rc!~$IwI^!F1}S;{6O8~ z(h}m1Vq#Nu*NcV{vlCN@nMIdUwrfhm%g^r%XfcsQkJ7#o20h7TkH?lW4aw@WjBB<{ zy}V14Y@>AGY{xOo7C~x50tTg}GhahQpNx0b3b1kne-2rpuyWV1TaWQPRb|*z$+XLc z`2f`I6n=J&g|9;P6L?{i41_qHkNAOA==%G&p4DzOJWg+C`_6}mxbH_wMY0kY&V|v| zC}HHI<%rebGhh(4AgROF)J^V&4zQV&tyVVPQ99JnDGGOSj6GoLI#YLyj$=}rQemE^ zHkp@kb;^_GJU%|tbs>BCTQ(v2bG2c4x>sI+fwXZ5yZ^4dAh5Y3=m_1~CwLpuP`kwC@Zv&U<{pzB zxw`echd;1S~TJt;d_>8sy7TV(!DUjSIoE)aZN6>@9L2Erni?sop| z`g;nbu~Ue7XGDr~_i|EQqh%p~D>O?z>Zyp}=$ugCkX^XTWMRE@@%@tpeCd4LZx^sp zwEQp3>DKaeoABz$sb#VC(wVBKn&FyABs$emD`qjOWs9G{j#DEPzV#d;T^0ocXC1-h z7&@Ytu+y-@-aEi+|1`d;`#ezBabo6(it*jV<&&mLAJIv%96e|r65djB(cDp0IvjX~ z&^t1_8K?o=A7E|@Pk&6v)+55OWxO9C8qF}Hb1x-IsEG8dd1 +ss)_yNdD}EUnOJNGh(pw z&Yy^-bebQR);Uo)*Jv)av>C22SWva#4`4n_;oS+deUMll_tkA*Oa3gu1_2DnxU#@?$-&J%%_z zcs(a?0D^4zChn}vtM;<@E4mdV2!042=42JSd^w%P;59oQ+KsDB2@&v{8uiOA=A?q? zKDhzB5~oGVYIFWQB>Qoh{0rbd7%`H#!Ua_l?t~FGrtA^$!hG=le0g?;+*N;RM^5KJ zOvmKRovIlE?Xmg)Kf@3cebI!dXA?u z^5}6ri0S<3cvVH`WG~wtuxZHI=|s!c@i;8q0I|6rS>0NDm?as!Y-anp2pZ{SAk41 zr+>Zx-w(@O9!2;)`KYd_QTfw&Hz#a|^Y!p-#qzqjZAv8M#0M3+|6{s#BDCRD8BKAE z$Q-t@J8JLoWE8e)_cf$1_Xqs7Zjsm~rl#j#_8}9NB7IG_L1$e5MBT?$5RgQA{lc)= z$f06~2jB+nbhiGvX?K4!Hj{c2N}Z(87I!VbbW2*^$`R5CyHWnyWKF2G%2Ece@kF@?Kn+% z3j)Yjqp_Wd~N3uFiI{^H$FvAI>>V9kQz=Aa^ zaC9n%)(QjO>rrRZEFfpz_YiFfFr%8va7h;D)dR6irj79@P5#)vL+-OI=;TXre>JP} z_8L+?gjcYaaiR(hW((b%=GH8=WZ9S+MtMBh>h!#@V3G}}D}K?9Ws|q-tgBC}RjW0v zW<1j{u_t~eoD1b$;y$(}}{TC+Q=x9~E+!`lLV-}#Hv+*V(v z%2r#c@aN8g>6V&PZ*$TTnt6Hq{z``rw!~v%wTP@8gr-NAzBH`UM}t_)opK`=iANb+b&+BO}<4#^tZ@1w* z@Y_*wc+GZ=_BX-TcnL_T#Fy+~S^?7ehXFiemY9)~ZD7)ydp+ubiR$?gw0H=WG0N$F zi1FkVF6onij@}`i+Fc;+`K$X?wZkq|EAX?PfFnj7uTp$(WkE2ajiyJ=na09sJL6Ylzj1M>;>6(oSPR*x!mTKIj86J|pM zXkwP9=OZ_wM_T1JW z=<9ZBbs5PP*A_o-h_Mcf!l<ZA$m+w9q7C zcD!1qb@(>|deT>auY-&|9$s$&YoigWNj3In^?rtams>h~Nw9jDUmUmZB<9?>xdE9P z>&(=hhoB|Yo~vzgi9m#(+*C$LJj!9FYcXYCpDwv+en(_Hsa&AGx%x9^hC>|s@|ZqZ z%>E;UaxMIf45Pl7An11@6z8x$a`y#=NTSr6=S*CaEeBF49QWo%5#>N3aL{)Zm9g$y z&i>iS`cGYe$oMr^$>I)|PtV7d3G%{2ARE(fAU=~_!+*`=K5`39PSn(mGa!P4WmBT!hJr8Qlh|U2kbBdAePnW9%<)d51DV z>DM3!BtoZi_ou>*i!JEhOkQpxe(=>7?k~YLN-ghm_%g7`mP>H=t6Mdbk zd2yCJs}9{tgvA~aQk2X|94=iW!owf4K}fh=u+I%0P-x|%Hz~*V91r|Dd3dKp&ypPx zcqR!8p0Uat&&`Z3^C!QY6BtspnCvIc!Rp|@2aG7Jp#5XZ;2S4}=GVWY0@qGex8Er8 zc=ABGvqSe@S$Qoov}J7=zZ`O z8_KUqbBIeE;4}Rs!BVJx`n0!g!_eZzs#inv-PlO>RY!^2T1kppOLuLB(ILxg7YjWEL zo2ep`BBW2b{n&$jG89WFF0f&tb6~Gd*Z$8cZ=^ORGmU>i8gH#M9wVw{HtC7=#tN&| zcfFSEv^e8ETWUJ;AE-aoo0jd;1hNRrk3UTW@YeIsS5X<=&P|KU(-M0Od_Dcs%y5wS`PJZcl`v56pA1WZ=LL$_0jaWVO!APn&!jG*`A)P`WCLrg-L4FfZtjTM zDl6jo*GAwgg3O8RzJ-;uczmh(f0{+QhASsjYl5kLSP~mz=|#jNFw64Hp8hk4EGD$^SIJN!TJ=ylO-S5EXcu@N%Zllm$yBUDAHE_rLC>)_uJoro!247aY(fjg_1a88KrlWlg{WsM=8edFZ=OzVcE%TtpTYjCbDWJk{6D#vFqnhIen%k>)M(t|(FQaD; zEKUxsi{z-ca2cnYm$8;Ef-<${C-NR8u9!7Md-DDBvC&aH zwN06U2#rth@A5oSKe`c{&lna{w`4U)M}j}n&lZMxgyN}ekg)|}Qc)h^=7BtRrf~gH z?j*@K6*MnXw7QotOw47Gtt1U~QZ|XRWJ#v+6`+C@586hQe&xt4kc*UVe&etzw;C^; zGi!R{_(+LITyWKTA@PkQO#>AzSNudhbPiWmH)SuWFW=DlgZ9IkSA(mG8_OEkO*;Ey z+ZQMid@<&AOU&M*-5C1@Hw?b@?n1#!aa%dz99PZVY*)3lDeo_bG7G4|S%8xqztVuv zX;0p0M6`446elxgbhhj?&Zaom=6a^tvesYELNR41p2i>Vx#}<{Hh%8SK zzf4N#Q>qt~)Y(#G@9${0Crw}`+mEWgo;*?#vrlOp@8m`d+#r4PZ0FiJ7Sd-L*k~QK z<#|nEdr7ITI<@&Gk`0)rW{9cl^u;9>+ij_>&1|1l+55hla!cT03c8Kzof9ql9Fgx4 z5~_1ml)E)9hQ2K_(b6r)oemBT-=DiDHD$fpOqRZ1(FWQad!{}zdhzk+hn{1erxtGu zjc`0*)FY0c@1{588JrB59p;VS#3kU@Qk04*?*{`+bAD~c_Uu0we=ocKDr{$^W}7zG z^17>T;DBkfmxS1m+SAe^Z=Bh;f_MMYx7g{>Qy$M0a^isOm{yBb>+1&SV^eJ+1Lty@ zQ*9IOaAkwYmjY%|5-!qo+DgPhTQ*56MsL@Rs#KmZZ=_q_As|m53B@H^@!Dq+i>IfO z*}l<6@^&f$?&LNAf|71Ip(VwmMG_g;+|0YIx-72yq#c0qWQ3F37j3vd7KKA8cwo7vrK8vip5`oX-H_ zpAKy&*;i2X%8%}ldK$8}LjTqHp~f>80G1sj(U43WJKLfooKp!~jM@tXVi0*x<-!`+ zB0stK@ewM!c$qb)yKC6qFUbnOKf-O89F39UN+Om>e>gW@NI|ZL<>A=Tj^w?Zj@YRik~o( zr)2WeBF=-ieeGBe&jlpmgB~h!oCX8p87GuOo@AvGhIO*iv(i8I6ZVkjmGJuaKora@ zuj~*G4}5??kLThy{IYFdcHkV&rm&6cK76~oFYrAg{QmpsIk8OjWYZJo1p(cIjY1XE z`*()cG8Ml^gcOlH0-kuwiqO2~-HEC6rAHsd?4guXOO)sG|C&RJ|8_JV(co@QQjxP} zbaXW4(lM9jwy6ocJ$UTOW7o!~@=}sbd3kvo8F)NogeNv<#9d7vZ8`NE{ECxN^(tJJ z4oAuIomPq424zC%k=5+@a8<+kiW;s@iI%;GRrznl#*oAFtAHYa|--d}T8>mq|-ijO!V)b3@) zJ+WPy8*;PUA(lVTf_7#tM2SH=JPPElLmRgi)8{9uNo%6J- z>rABMYcTckGB{F81Da(ysH2=iHj^jBbNjO9j-hyUBeFjYnPj9VcVBzH zG+pBj*;CZiTcJpL-lSAvwe))gjAqtpXI{YiBkzk#ugB`?k2q>%Gg4LWrgta0vVVTc zdld76N8V~=H@QfI?ZLOrkAH7yEBHV!cVE{W&^d7*s1nl0zu>J-16-C&r`nVir?Rw> zWPGWWNcUJ}A~gqtPI7B+bwW0Bbz%bmAC%vo98NsK%_S0aek7sgNjp?W%t+?lU|GvC z^d!^6>~A_sD>XuyH~v|Bwtc;D_E2E;jy5gNpxzAQr#TgBLI=2hBsl z{#wU;w1(~T3+4ct^w0jq#4lUKBi_(&a-2~02+@Mce}N5b*-dkHT}#}=(r0CtDpe+S zyQxQPInv37KPPxmF2v5bxh^ybr}s=DTcxy9@38EQ_^tN^>oR zYIByv@_CJ?=6*O91jPqOMY22i+2m~XgS_>}W<^FCTK;yII5|_R6Y-5K*Pcz{p@87= zBgO6@+S-jDqYdgQAGUqhBqbZ*>1K0(G9NBxm=n5{!Vg}k25@Pocq!O(i@f9AZV+{u zpdX~XEZJx8j3S%&5I$fQNNmrZ=naHZdYjH7#pqS;BDHnFn^-!XyZm(tPNj!)qDgE=$|4i(xkUt9oWgA#*nJ`81 zM%elySt0-^fWSy<(enBd*pgr}wdi?UlqC#VWVkhYB#geDn<5VDgKf#$pPTLTYxLsd z1;OT=1HH)Fv{uVD;a~i2snX5B-Xg}K5gca6{YR5Xb2aTp6}hEw$_DT<7$(fWL<*dq z%U~iZ{>b(p3}ar2EV;*+ly8sY+Q(X%{XTem@);kw7ls5lBs{<^{0)kheODNQ_;-~< zwip8~Hr@dwTil+8Dqn-jYWDU%^%aQC$UTP-oOoi=aBE({_b^do!CNjRx)p!Onig{b`HNT$~kk99`` z&kETJg89B@g$xFI|J1_>PB*Gw6)k;eGjBP&YM(gq4pEYnlw9c3BB_j!5ubbJhjH%9 zc;fYa%O*I4uKK+dw`(4UXy)c4s+ETrd)DVMGl-ktuV$23ewjmaUO5n%VgHikT@pdH z*`My#96X8V3s;Fj-vg0zKShxVq&lhGEmqbv>(MbJs5-q*2qvc7>PP?Cd>byV#Gv*d zt(}yCKU+H{qnnT}#VCHFPUqJ>hG(?OS~tsHllMdCcm6x6FMdk>w)=_7x+EQMxoY7^ zLa1N##XtB3JCh^>+?%QV09*d(i!Y038r!3iZ+MbRh5nAxh0te(J~$y37f{M%N*|lD zW>R5}G~;D^L$0&yn7&Ci58|}$GHek@*(~%*2GBFRyb&qOCNtNAmj~__dgDk}xm0pDlz=Q!lQ_}jIbp_jAngISY)>FQkoKL4 zWxl(5Y2-wIkf%%I&0NG=DVu~!{$!POT5XKt@i(K4m+!p4CVQJ#mBi0ftANw%rzQsE z!LW^h7P&$APzQgJ_Y+S`&Q_v8YF_t&P)>m%@zLN~rG;3e1YZm5p?kZ#sQ-xhVmdE1 zU7^jnwNib@oF#8-8t+%HTA~i(s-aXDnMjk}Fa*&;{zkp!u4IdXi6&v9bEM=(ti8vT z7hq#eSyz z?7yihyxMym_Lwk-S3dpe(20G4%T8h6=F1nRG{a_XVbd>VzvsO)9;O!gDoC11iV(8T zoX~?&@0v#zWD*UNE(nVJ-6a}^O&nVCrW-Aq=nu=8XaXHjoLw}h&}PsHdr4v`oIv`7 zeW!2J_Il3PIXu^=Qpccb?iU-sbECDHo*-+cisgv7+I|r)Gi|#r5zsQV)eC>mY3#qi zw&hI>h61PDj%K&4Vx6E-rpR~6pMTna-6oYcrks!Rt~HJxT{n6J;{(obR^kYTiI+T_ z^X9?B43{WR&Cgm&r1oCs8{L-uh*fxx_)vrPIo@~VHJc{eQX!8ZO=jXz&{sU&rfKJ^ zbJ5KoZSN1ycU+2jL^=RE*(dkEF$S$#TNEidG5c2yE8Ic7=TEd{XJmwFCLSJWtf&f) z_SIuVSo?xk2-`iMob{p;gYJqv-&RRI&o_=sOmW9qNlT0>7_vw`v{*jEQJfNRcNkaQ zTV0yKtf7DM7oA92?eA?Ko~$8q#Nm$8cIMwL~4dkkyg4J5fI6dLFuj$ zhVJHWzVDoS&z(Pb9u;TixA)p>ul25Xy{p?@!t5l{9O@xs}<@3DFM@0%$$mM(!y+a-+<(JCr=u(#p zxR1H8TBy|G$A`=Wf0nmJ1yuK4ec8qaHr6`Y;MoG7-I33n_jN7MM8}1UvcO^T&RR|M6MZB(djd#Y^^Chz`jj z1MlHTUF+kM;wfvR_eDE&%2VjGC~q6is1O9h4_{LGoC-F6)&-u<%lED>QJsxsvy4BU zuZP*QuPtyRj_-KzO0rFUaQya@wbR`@^LVlWotZ{p61gv8PQN|VFAvvgbRNqrX3Ihx zDIf$R2rDw1aV8@jk-k;phZ22ru&fQG6w{{sK1<(N$Paf1;)OQu@FTOP%C_-AL;Ui= zei@&+s;Z6Qjlr|>>4=mH-gWwsAf&?7$79r>DSi1McZ|Pmf6A7eqzP8AJDBrcajkuPmVnGk)Eo&xD*31` z9;{RJa_nfWyfo=pv>pFEEQrAwiLwYjGM}XZ69>x9jeodRRuI3(RMAFY z$i7;ixe%>A-5Mz`IVpCyo&QFxVsq6R6Eo(O&Lew?$HuIoLvJ3bl1NAkS9sYm7}+K^k(JSNHX`&zdYIm6%Wg_;ct1F+wRSl7~CVn zLZ};kKY(rc+3I<3^lncOX5aeZp4ai+o-I5wJm@0r)~Kvuq}xEp7Px&E?g`C)D8#396i7S~u|0HH&f=WUM%vzJJFQ z-njQ{p6O#`S#_w1r@#wd!gmnq!M1ZJ%2K@=J922sH2xevG1&n(?4CH9W_1Z+p&CqH z7X9cK8;R*$-+RY_C)QFf8VonSU4HTt*pZVvSxoE|->Y!OB~Y(bP+?YRk9a50vuH5s z)l6)aX3F1?`+-a2YM9q@wR34Z$M}*I;vzH70*yVgGSa1+3^{9NyD25f9CVMA(Pzt{n0l+* z+WagpNUYZSbX=D$v8!<}V88jNfF7Tu$%hl4EurI1{=oJ-6p%evB8Ku5ol23ca&A&W zFjTUC$^CWKFCQsZNkMV}-nS=WzP>vFh745_sEhd8M=U2Fkg3>Y^}1%1#bn-uRw4dB zE06mfLf%Z0Sz27FKN&m3y5&Q>?dSO|;TZ|0>k=51aPJMwv+bB?RA-r1Ll#ztPcTHi z;yr#?sk%6|dZ{)8ZsWWL9oFcnmwFbnyFSAL-DoT7atU0+is|I6#kT@7^srvYjh{!iCV8eWxkHcxwWhS*fyUj3Y$APbV^Ua{G0#=> zXXR}XvaHbVakQZL4pZt30v1IU#5;W=cISFaTr-z_L$|O7OZ-%w_Gkvw$f}tfbvaA? zpXpzAyYOGiD&JZ}}}$*%t=Q0`vO$!$&3pp`+yC_U#{bvTuEpJ4@d7I%<`2IC6<6Mm`%`gNRAbRoLHl@soU`q9Po#f&5aPb zIeOa<$0h5c-|MDpl+a1U5o27$sgwtPD|4ZJ?s5oB9MiW~OpW<@bhppC0g@hvl~5|! zXLZsq)BMgYl_@`i6_3<)%u6Zb78FY?cXA6QCwi%-|B-@aVgtD-+43_ila$Tx>ul@BC` zpV=zENA{vEr*$9nc>7W0%^q|D;hb9PKBgbvnu~EJGu&_XL7wzlTZ>gV`eeee%NEa{ zuiLzWV-+>+8esRaJdyK@P1XQz{J}`XZr77V$%DE5F$F2Uv&4-bo}c_G+s25SRjI11 z&eSBwMw+92Qh@f@z#)@#qGNaE$;VE=2xOiIXRQ<&ro;Zkkj7-);>V3{Pi2c74oRs|VF&kG5=`foNhfL3VAEqix>{ zJ7XIap~yQ++SBu@3)kb#^9XdxfY|s_%_{jF?X@gH+G!Z0m9_%8E_ogOIDfoBV2^+N ztwFh1+-VLa&828x@!)q#BY4@*E`Wr7-6W_(hT+GN&nq<;Q8lY39}<>rg?9*TVwDm% zU@xS01H+uGADJD72H%-y65l$*D-hUszB+Vo|AmiNX2$OrZMCu^`66T6y`EFLxdfCV&huP^!@9WrIUM?f+&TuHYC})Fb6Yc zVR0k&83TWLtKpIPq`O(C+*1m)!=6yG+`S-t)J76x_Jcw;ofWd*7+-c;WN@gakLjv%c4B4g$|t6j zvQdO)0VxHKgFmm>ZdS+mEiw1(B(V}#IN6$pn1XUau%qoW-mb=~lCjst@~3x@yFTg5 zK^_b4_kLv;lf_zQ;J&jAd=eaiMR#5JWkcgc7x4Mb<3k_4XTU>N1vG~tyDN41ZmFEN z@BsUL|BVJ6FBBBH_-MLqRRAxV#Kq#uZGOrF029Ib+>sXwk0KmZT|#|^9aSmk1$Xtv z4;}i}Ez%r2pDf}>RCaq>_*Q0kI|HLMo2-!XrXPtMshB(~Ht2Y|M~7vO_l$JRe6mui zQIo?oO>-of-K$s((`&sfm>=qHb8?_J?y%0~ zt{ebUgeE^OmtSJdR30&7y4{7#%Ky`zeYP0KD#QO< z=Y^x=*L}l5khoT;6sw*#pse!I1gyP0fd!auG~*-WA%jb=XU=-X%O)mWXrx7=Y6F|; z&)S8l3T6l9UJgumz|!8{xVJwQZuLga`nCn(S%$3BVAS|>|1NPiNlU03kwVDcbPH>F zP{bZL+N0Ef9bc$?PT%3Heuv!gkCgc(w^#MY0a{=1DCVQm7TvXTa$T*JEReA;Xvlmo z?G4E|8B`VRH=|3m>x$p+C5GZ}n}sVd$0Yqor*n#l#0xR?HoK+EW7bgZu)MYKC{SR^ zB{@7-8sln6|BK-*hOxB0Fw1G9@%t{$sQ-C+Uk&YOxZ$F^zfFVO^e&*Bd&8gY7ODDk z7Txw0GbkmoKL$j+Y_8IJrq$`lF@~s7&v&Pk2TzSvFFv*vacR&mi+vWxSnr{`ljn@) z;eg-r~;$?YY;aT2J zC~d0#&L?DX94b|+jH+x5=c`ETvQ4grl@qXk9RjJVTtv{Y7e^w<-O%6gap zbn4whf<7xBTV+lowoB~EXb*;>56Hwr$WOGQMaZhP#l?7Hr>?g|l#=b+0R^67Ht)a% zI!G1&+O5f`u)h641?7Y-)rCiXNtJ$yJvHP#mYpIIKCawWi-P6G`-d+sc&1;k)>=~* z)01Idh$lJ@O0ulR`V-WWoke+=(!+Cqp6*iWWmTt_Uya91k>;2;A!57z6OrNot7Vf= zucVI<&gFl+GB0=!VLpAFF0LvO;=g@8ld(sjiA!R(oKW(p0m;JuD&XPS0!3T05Gv1M zr`4dtJ|bv0W%VX}*5o*CF$CF-@s>Dy`;cre?q0XGsfn7-F;6(M!@ZLg`h~Uj`HA_- zddPjNsN{!}A1W2iXBfk2;LXH)8gzyB>XLt_;ccQs<0|Q@qObWHmf*QVKKjHKfKA~| zD{;G$&JVlSmWsl1+*WkXedk2vhn};!6pd88cCW*eLrSX;m0gD~Eegaq4g6HqYDEUy z`7ijccKLc@l4vZhF1M~S`(^rTFemShMUZK%r^xXhP}h#!6a zZAfh*X31x}6{Dy*Lq_e+hP|@&@|-v2$=ms-F_^@Ri;_u)SL)8pl0SJW+QF1r@gkXh zRyI6rElWwDXU1amYl|Hk`;I6~2eRGXbK5y!FYTbztA(rR6F`a-IW_i0j`y;z#EWnZ z!F1^!aV^|J-9y>sFwsw}2`+i;6hlk|uoV<{dZOTFyjfV}UQ)J{7E*n&w zKkt(3hh8vAT=@~)4Ldb2gmgb>|2^7}=S04C{`B1CP)?Z?A`I_7eM%%1`~=H!B^2;U*5;6OYxUdA+As5RHAy0)i(%a(ucX+?G=PW7F1l4vx%2s zJKt@#o;h7NkU&Yvdi)@t^E?iPhQb>KcMJ~6N3YJudJpLE93UhfIDG~D;&|q^m81uE z)sh(WqOp_q1zuF%>oF9%2hmAL#BY^Hb&A6?QJpW~WZuzZI;l24V!lK4`1xVC55|6j z$pHxvJW}f8(ZGvQNKl0RgTy{BEvmYe3q1wsF?te9iwZv5`MqjI*SbjOToGfN5zO{@87@X^5I4tkZ^+ZBNep^&wm)?%;5 zVnFhiI{7$xx|6nC9&LMziBM$h(DpOc=q62|!|g+^Htq7}5`X!sD0(0QZAnWhnV7k+ z@H~e`{q@)(#HeCgL5oQ`7VE_!&KWDuD+iN9`ksVqJr}rP_o*7u7;m#Mrh~Y!m}^U} zA*bSKuv78G5AwYt^P16%nAKMSC?I6$qg!9V>S>zt6Jw+YcZ5rG;)M)tvSY7nyYo|s zn+82=trBg@avZMWE%pr2d%g>gAzbD399--2iD^$Rt?0|6bcvv`LR6{PIC4G#NsRgN^j(xcX+mY&+hWN4jmH)S`CRXROxQB{SJ56_O&NRXSN=qI9XXVrKw z04&6LSkN{p_C2u5s_fd;C&X#5)hpuTlBH;-|9w)>(2FPP`CL}TCGkLM0z9!S;dCHvbCXlVm3B%m)sIo`=Nt)Pj~DND9|>A z&9>N%K`n-Chwt-p=0=cpc+HewlAJGYGsiAM+DF;&NY`^vC|ZZ^O!flc2BhhNnXqJ9 z4@#U_ZsyH=a~*%|9g9JxMlK)xKHQ4*JgV9p#NefpJX<=@%5%<9aFm^FwSGMIQvIV|5;%wB6p);M*&!P2F6WF>uzOQ# zgbWK(G!N!<=SuFSAVr|R+@i-!164^OE3aM^xLhzzQ}goB`Nu;HLhYye2_{K&^ulQ$ zgj3|RNmJcNActoIhzm-UZ;|oakVYR5uIOjg-zem(QhkrF7RVYA=`Uwah<0^q@};vL zZzd*O96y4DlEmBsFax4kCdUijuTaf|J$P1d;5m}CynXsc6!KUDX`PXhbo`+2^Ni|u zGEDN1tZQna6+Bt?76quq?9Ek?*|V{Jw_bS+DcHF!2n{G!ZREB6PLdVkVaBxbay{-~ zUbNTiuvW?U12ZLj&&`vY7r5Y?Yu6?JXfpF;3S;kiF2W;kzL_M9<4A_1Y_Qe@_54UV zQ8C$B5|)ya9?^ksrp07cfP-(M%yWv#$Yp09&wg|+-Op3fS!;s|lh^>2TIw=%Ri*fd z@YxtB%FswsNF`U&jD$OOJ%MyX`SU3$iGCSL1XYI6R zj*GIV7!l=h79T_}36mVMp@k%IbYJz-d04IW_+e)*>RL2aTtJJJ`xNwFpG0XG}OPWlVF#Ee5`a987UOp7B5>p_5b#SHDJ&)m&U~qg!Ai0=;_q` zWGe$;0o--z5F3BISqgBrD{8TeoYY0evhXNMj=JKE*K~fMd%xti<0yo@R#?u-{qAz{ zvyc~pKN4QVuT`p_w$Q9imb-&iuuzF0?814eDPXvB5w-5-5AuU<<{nYjX3hDX3C0r98$R}_QS(0wvv9Uw%WQ}XG)!o1le}Awz~-v+ZPW? zX`lOxIJ}B@QIH*-**`nG7Nr$jbY3NeJGe9~2h*>u7g@p~g|%D${TLIlp9785iO$|v zN^zSo(BwerVj2Em7G4KO@KdPh;hsZFpqI2HLzN8SFj-8BB#4>#^mQ|D~9UvSE zp+7kKOC8{>TAMGJixZlXIlR1`VaxDq`SJj@@jNv31A$eDDf1|gU-j~OFT*|~MDY$2{x&;j`D(8fqScO`!MGFYHv*JDO9D3jF;?M%qlb*Tb+U|0S z5=AS#bF@(q7l=-`m3UwIyg7A?w3J2R%fuhqWKc`5zX1&<*ig*KuM2$C^Wrmk%f-;- z!dBqb*^=L2c|8H6y+5(5U@x=PFFd`bFpH-xVXK7c!oVO$1h1_!0)geLW?v{$VxF6n zrAVDg5a>)~QofP>T?>+cl9`j%Se7j^v!a0|TS_mqcKRrVb!JCc`{`17y*?Jx`m+1> ze=im-eFaO?5ORl@u(h3<`| z$%WQg4qKa)g`H~E7v`~0OVD=SU_fK$J1mev9GUp^%ypb2bZsT;FCA9gYX@ufZxs+GXVq6neEg4Tn-f!g?Ju5f!bn>1-)yPRz4&LvF@ z7P`TTI1|Ouh1=380DPUz%2cB9cr+`(B3tGMSA;yT%}jOe%r}8(az*! zdv3n=d{+qwBKZq`ypw|8qO4>}$%@!}nT%o##yo(O=!k*85!X?Z)R*fZ)}AlF4Ie4U zm~V5j%84HQ>=HA&8<;d85oE6$TUN@{x%0Rwv2`9ZkaBLFS+4EUVa;KWxv=l1~Vp|$YFBV z_6iFVTB~pWFrQ6$;A-*21)`FJU8d~Z>AU?P@rq$Ip`SA2D^dj3?Pjv#xjA{Ct`;DSu|%cHo%SILq`d9;q+tUNXv@F?7sI_4 zxjS(nw2^8ML0-;OI$BWYO-5{_IvMmi+cvnQlxrO}m`_S=YwCO#r{yubdw>VSSz@Uc z;rW8AN+vtlzxOnydzGWSg`>IkWyYOQP6Fbj2en^vV>`b;ulcIGac9|1TP|6QJ81^D zu2CFAxLQ+#Bu`wMY54B>0_NbVG@W!~|JrLqjAd_ZlVu&hLtycsu{Gcf0tI=p{w?2k z$Pn1JvdXyee)vQ=u8X-vnv3t5IRaY_hQ(lHGvs{yg?}Hp zFJZw0`SoSl42Y(!R?1r$B*OC~^fj;}woX&?w=4ohp}&f)!PcIVf~|dI-caJf58Ym- z51qtW_{mq+X?!9%Fr^B+SDg4Fi+kXh!uB`b{Fe02jyMRNClTq8G5K|6_}uPtpGOCw zFqlxp-kfTH zGBWB#baqr%#vAo4OKvG4t+k&0U0s)sNP-Buh9hANfc00GUMdgHdi6CUEN<0ouH9V(`Zc8}x(l*)SQALQ`HM{}As@wGTv`|jSJzD|rF z*m8Wq1QP2I>AsqvM93IczxNV!-c` zKX}7`CPKc}X#%=i6N%)bKN1}D{c)bubZm;a< zR>7qPYh~XftMnV2Kcc7+mcr4#txCfdQWxuo`m3Jv%}^%6<=_toV-*$%x45tFhyy+v zK^pl;AQlov^IBK0_YTyIm+XY2<%fO?tR|6L5ANsn>#OMK$)tE2E83>C1!YgRg11~C zhF;jOAy6_@BemO1)2?5&C3ngs53j;SdT1%DSsw(fStduqvo%OMQo^j*TJGwrDz$|a z8ED|FoF+cyBpXead<7b!$bb9TOx3$B_d@tX+~ug!cWcHp1aI=&>#pN>U}*Q1kY_o z1%V}67YNinUIs{&w=zqFi$cq<%mPV9ukC9S_@94V4?Mr7f}*|VNa^cVyxmF|e%`i- z6x)w>v!}1AM@M=toL&Zr6#xpVjvhkA`VH&M=KmWL1sE3%Q6x zV4dLxm|lzX9YLVm=X=P&mvGb350DN5K|Zk!loH>cwo6PKD_e94q3<(P2T(qb-+w9+ zczM3G^j!@r$i&Ml;;0qA%SNW}g zu9|`drnBsX-@pz;FvwW(`}5Loi`IFGulH<}moQ+pM`jYJpN1S$$}1dqFn8 zl?P2ed@O+%{Zu1$e^8`<`&URVN@A43`Z~Fqz3-`19|qScBkJAn zTc8VY5;o<@O)-1)`gOx7@8hy0DxG%|XZhP~GUVYfZ8M*BiE@W{t*6}LE8nV9mibJs zQd3%Hz|hq~zQB_{XxUfD{Lti1rqjOcXVNH61^si-d6)*2!(C-p^Z-RsuuAfCiQE^1 z`@STWcKSu51P?m=^LYrMel6AO`66w4^uI0_bG z>1MUO!d2zM7ZA0Lc!BgKn69z=^CdTdsuImYory+0P}`6GjnMmP3(jQV0=<0yZcA3~G;AW@dPpq{WFuo=w`tKI+()tnE%LyMk&{F@@&S&#_1hXbE{e!X{J#ay!Nn zU)_oG;NRscc8)ot)J3p#A-^_n11;{F-(fy03K)8A0im5A)(h;9Ui;$W1@BWG%d&s> zn&p}H%G$SK&)h>R^(A)AV5N){$LW-q$LrdTaIa>E8)pQnu%q@xY^D*fput*G&|K!5 z<)x)62L}f|i&H3CiTb=q^~?f{!KPpy^c_K!bX|^2IMK_n#S@Il)t7Mk01?U&9$LEi zQ2~N>O;w4!v#mBMN{ukOJGLGop6g?lqW*EWslh{B#I&^y6sxY<#8JsawFE)hqtbKD zkwMy;dAt1C*lUc5fKA-Lu6AI0JHGI&>KrWuM5Un;5$rpA?>J36G zTDDtPs#Lx@H2NYwr4r4roxBGQP@he@M$fESCvX1+pv+y{6=eVho5uw4wJM-(K+)J@ z@dcb+QhF_q#XOW@k$;cE^nV z<7)5@wQgkF!=HH}u+t4qU%&PbZ9{lw)Mcb>xb9Oa*lUGR=ijmS&CybWBt*r2VTCT| zJSP-7cf8*1e_u`cRYrE>G`>WJSIO0z0k*?{jNPJT2+<}=2&3VSg5?NpzU0?)IVREg z<6Ts)Ye!TzeVkjpk~#L3p#{HEJkfqcNnr@ac;5_0@VX50Na0Eog=UQg2GF4Y3u14gz+Ba@B);sM-yHd z7nS*M*q?$wluZ=hpAt{@ZkC0cw}A&OeKRJB%MJyrMH(|p81^BXbWH0sJenZuIh zvLM|DwTCIFEhe_6l#~OReTJWpI-h*i>l9KG%iqMz6`3vroL0-sA^(`9F`V|s`;UMLoY6y7RwE2aGc(a8&~^KS!u0_F=~e4Mi0;vceBX24wO)* z5p&VY+HeM$n~1-{ftP;MzFRFl`yj@WxpwfV4u-!6G4I;JTx40V8J~j9Moek$t8drH zR|`R6eEi=riL31jldR^EhC;l zSls5)KVzbe&f%0qACXoE;0TeJ^YDAbd}CLfsQ5I`$ul=r-C~Yh=9Zr7NXP? zJ*lCY`(#+uiAp5~-+R)r+Bu8fC?tm|u|RMigj4!Q3;(lN*FPp%&hvlZ@)A%spYmR_ z!0nClaoXc5aN1q1p^LNJgfo2N#w>)XbeEFyRKEE_`>`VQdvy+YWr5_I zZ&nUCU(kW{F7=$Vi1GJN#Dju)!%PA}anY%0f`n+eNEYx@2cotTQ59m%`(()3utNG4(0osYF`S4@tq&mwPoxZkI@#3Nw8FYTv=JM z^zZBKg=Z4?_4OgzhLJPFSx@hXIU>TM#CX9zac zfaRu%U9u?gmHw0*axNVkR4#NP@ajTaT|LfcWS7JvzAKUfa~;pMiV;2i%Me8;@pR9D zh(s2I2X_0!;gg~B=J|S@l-z{lQpz}rE@)i2YYu(Y6u`z~jEw@LuS);HVY z6Kx*>W{cYmeU_toY^Q}%aNC}9>DO1`U%-mer1#%f_g~Y@LKR3|*aH4_1CK%NU+UT8 zu+sX4yR_hho`UhkIXTf!Irr9U)zOf_9*nTVSnRok!`LJ3SFhNjN{x`^j@iZc+CCd2 zdHA_mlPEDSS4BUD75)xl50UgbAYSJyl`Be9Q)PUhB2TZat=+Plrms4`cdu$u%;R)_ zbpj0XGdY>=F0OM?RwuHzvxCE=4gyCn996S3cYsQj%|Pno)TbVH0vgP-qqWozd#p|d zzy%o2mnk|QTqs8skXt<}pF32bz!FkR7s*l-Ym75G!#2>yn$Dd`w3qPX;b!#_lnfbD zM*FR>srh&zQok0A{^5VvR+g6$s<(xkq^MubPDFBEmDP3Duw7dRr9XcA=6HVo(ZiQ) zZYgZ1v20m>0ujV+N)Fq*eSBL7LpFPy&_@(whW<6?m15A~xpxZ7p}j&q)Sm=iF0+)x zib6T+P&cqA(;U-UN_YU`jT^1TTAzpdv9!n+sLqOVze9FYWsrR85G@dg?cA{JzU|); zldDX$;2DF!D&Ycg@v9a%M1eSX022?vaf0z`aq4v6&>CmA^)8GemJlHlHGJmQRAzO0 z>-twg;zJAIyP5K@py1aYIZ9>%VU-TXmWp?gNMW56;WaZrjBE`X*dHOFTLW~o)5qQM=1Ua{oJ5(boSHNKT@Na4#i|j z`$887Kg!^5{Y*^C)2{1Eb}}2IAScU>>sIO@A__ce@Bc#M2A|8U^DOx;8SB)d-FSHl zTKIX>{cGP0$)Seixzg?Lm+}az?lXd~oIWuPk-jC0b+brWlTj%|%$CG8uhF_ZY8Z4v zU=7c-a|e?2e>QYUOT?oOJc-(SWEC87gVO4A8_v`)Qr0v$lAbldkI8gpcSZudSt-EI zjSdvyro&XW?uA36d_ZvBGJ#C(L*pV%*`MJeDEtGXSZ=5H=wo z?*U>@|FX~vZX;94Fd9Ej_|V8&FqhwA?QpzDPnnl(!~o@{;FqB;Vn}p`bBK>7YIS~Z z^=N1EzSJ*wiC!<&z!mUD`2R)jZ;U<@ngZR7)@fS0bn&CUyf zCG9$gb1UpEq`t$S)1H^LL0&HTlU!LL?J2!}An{+sw*XXn&fO0wJ43?0#H! z*NR8?ZYlXa39j{{3l7BMNM9Q9|DR=I2x1SbW`I7zQ}pPZ5B-|25ORcHhY%3dH@RMy zUTAPj$2CdUQy3ciP}|Yb(ai|@ozN9`E%Wzeq5UKE7(pfV7xgQUur8j_`{Q_duUmrL z6xKhw{k*8AdHMoXS5c<0ZmD=g?Ya%Lg$~bV6|UDO2v|2CNaLbCYP**t%_{)=u*2fx zlEzbIPWnF=nK$%^;3THI;Rvjw=0DkEVP%CNRX=J&8o}rEtf5!?swn<$cEb=+sEa{5 zMC9!f6Z9U|n?fxzf~CZ^lxME4Z`r)YO4sX_=lkV<1!KjSb%gm)qeV%v8XnDm_5q3w zRKiY%q_G9f^$trEgAM7_xNfETHJs49%qh{JRBZBSNf9Uq>$ddXz198niI8t3l$GgTDsS(y|t$!nUyKM;Jw z{k7MQ2O7iJMTdDJLkQDxg(ABzLi?Xw0%c#%&_ajb?P5|vmB4rAM--}$PpNK->0&nt zg~|i$xnYADCJ#?Go$=Wx@%<5BQlS z>}>4jIw&`k%x__w!q!p~#3yD0=%n;}MV&8G1mpx?_bc!wiV_*I?#drF1V5-bx9O9$ z*9VR(|I{L7Zf|Ei{dXEW zV&TR0?gQU;NCcdg^Vl?`{XY863(EO!GB|b>*`0yL&mYwe+1bwPUo5^@%jXpwNfU97 zn^G3dCD3?aM0D0FB>D4FQC@FhYhBkTaOr=JoI8i%q)=(xT2d#LG?gP_=@q}qml-9h zfMl&FoP%}-;HXQK6>%K09v`(Sy)t4MNJ|1;L%)8!GscqIC0YFx@_k;7mL6k&y;6V3 zpurpzh^d^(OLj#=WbN8b0UBmkQ8w)o5Jwx1Y8LA z?ZJ1T&N=B^^>Tr{6c-Y6bp_W80$?nUKU|;^s5CUhRP$J +Z9PsV~cSmoY2K*+CG zt=tP34os#L@U^Xh`3cET?) zvo5o6$20)X3*R$8#+c-l1tfGf^oKOPRg&lB>!|{R?DG0}5v(fx;^R2izUK>ImXuh7 ztgCqr$DeobWImB0Y`Jco;Z_T-8;#ie*x6&(FgkhYeHuwWZ@1E)o6ujoi2u35Y%Zez zchi4r!%d%tFq5Ap+HU zrf7Q>w`s_lzxT-H+BvXgKsIV94&ENuJEjT8{rlh>4^Q4k+2;=MDb(!c=GSN#7u=4I&EL@T#l)g&Eh^Fvmy%}+*6EB!Eq_5z+nfkS!q!H}FB%_`qpVd$!0{lk+;|wgr7;L@=;!P%%Y?QYdcu6};(T z;TMfXOC3wEz9Y+ZoxjKK8ojOpAl!f7+>ttnUB3wB%g_@<%fr8|C7X3+MfCi?pA6J4 zsB)Tqd8fr0q3EzyAIBPFDTRgEDtsJzAhOYYDjbIQ+IxOsK?G}O##bqvhKbh4k|sEL zwY&qsT#Mb^*`X;?v>(PAk=FSCS@hb&#Qu97O6ID;)ml8%9E1}}U(#vp786|mchL!{ z>;>WY%+1w*x_(UTGbn5Y<;mnNT$|Z#{GMO}W^td&>QQ6V{wjDdtml&;^aF?JdNoHe z>mTugA=oJMnmpF?X&SpYIgM49B0IR2|dYC3hlJ|6cRjE&+AOU=}d$_ajZtpA(C3za1$x`dV9C z+yAxBu?Nm_txvO`Lw})Qa{Qg*&A3(4>UFhae;gU0bFJ|EW-&^bH6pmAWWMRoKi{jr zSnTO@|7yp`-;JVEBYV$iUe~*6eC!igh)fw*0Wr?M7SdwuYrzV20SCVK%E-uos)j&tQ6Y0O#1iq`UVpx5_qhEv0y-S<+fZf8g%d*RCy=to|1l`u z7jHP|xD#RshHZ+tmB}p@F5q_EuLm5i@85qP#JH?P+Rl(Xe_6x+rC+f_u8>V59Oyr3 zgMTijXM{eYJ@v}+KO6#8$Fl2Im&WtmK7{dAec*aU_m`X;X6U_hP(v&N%xmW!OrxE% zvudIoaWDa&xciE70syvx@p%4W#;_t*)y#APqki#GlR$sCn$5qytX4*7jLz+o z(3-66(hjP;Be%E6_2>L|)T0n!0}g0#{r~5Cc9nl8Fd1rAx6`M05{7Q@XsAOS^7KXZ zIH^fqzq!}j-+!9)`7@KYj*i{;cX&_N%RAnHD(xDNGng>ey#2gok#_6XH9kb9Dy}|U zp!y0#>?pChoEXW}eUUM88JxC#lu{$mY8A2`Oo!OIto<%}x=}q0w*qj|77D0CuPph8 z5m0_kG6~QF$+#B^!6-p-4N8NTBSpkpK!aOr$!k9gouuTKG|avt$4@LFi}3vc+&5IB|^B<#Dc;G>i}2B&1kLJO=sK*4;w9qEU(< z8t6a8&$=q+%NB1`1|_6kbZ17_;Y9BfF7{{%TsQls{^g7A3bPH9qX%C-9JY)0t)^$9^U-tw8QuChzhB|6e#VVB}ucK7kXw1Wq| zMzBG5(Q1i-<)PEybvLw|jc+P4QY?r0+@4U=i8_A#t0A4E>gWEvWoK2_i2r!WA^`K- z$LiN90GD^+GC9sfo}fs+2!hi*lvzp&aEqrgo7kpMR>-RJ!P*}vaqzkBGjn>0JNXhs zyW%}T{W8e6riJD(7R#BE`5hAG7Z4dDuQzWXcwu702!=Ao?DB>Ny{M?DF97+36Q&%e zO}=TVCze#Qx5scD`D8{cd+)|aa`0%zR|%0MwNd7a@l^4;XTo4^qWiyJ=kUD4~4*Lp(1XIumLv9C_=^=?{ZyqK?k-Ye@=^tMY#QnXk|(dh<@`9M^U`fZy!bX$?RTSAknp zEfobSW?PoJ1M9{G+*d9<{9c=eJo?oNIWMA6K{#}U?IAeUnKgz!fIf8qbW#mtBu@(^ z4~kS*ueD?A-07tqSj8e@VWU8@tyJn_Rkaa#1CQV8@~fcY;wKRQNCMBTfBWLJlS~yz ztDdS?s!9P6{4c1r!B7TsBeBZIC7{A7Jmn(emY#8~VD<(DWn;m6FIMTfF|qV;**}7r z|2O~Wr7bb8ANJrfgm?z{FW6F?Jlrf=>ZC_?t-OOs=@!cODX!yfiij`q>aZ~gNn*tA z-z?ftCjnpCotroJ;voYHz;=!=5`iEN!`iW_$g^?)@X@A((lM)$H+aV zHs40^V8?rFZ#We%O(rUgmp8tCH!0CD3!S+8<+q_4?d2C-WSB)g!zFgL^%*0ktx!s{ zfUPWSjQzlM?QdRQw2uLjDM0382}nVrK-+=UK$!W51dUyJp2h(|k(>EzIt=pvC6wh?*Ih*E z5Zu=7@Ay42U$2W#ynw*D6Yv5)2iP$91L>lg^dM;6uPO6)Ogj*Qu{5uJ`3heq zJ(t-1$&A2i2QwsE43^(w0^nD4-g7r4_J7If*Vt*38qQ3dq|pQLn7m&0g-xV|&AkD~58OsO~myMXIesj_)U)^R5r zc+93jXz&Gbc-DXOOpp|ofQS}gdeRuXQ8-peu6#RdQHi7<-k%@cz4A~NmDiNG|Fb+> z%yBU=ik6G%fU^xwE-oe5JabBn2z5)@TYKJUN*HzX;e`FcdKHot8oUpwFEPHUcL*3C z84!lu3_rkRTZ?k`%NYdSy-o?Y*McF;8qt|PD=E1YM1#|iyZMFqYFEG-BW}hH3mhT*#8e*?;TI||HcnD$Q~JGkL=l_gwnDZd zM>a=hC^`p+j4~@Dj*;wRM(9|DWJ^N#>yXd)ci)fucmD9{51+U9dClv(p5vN3%I^(g zUP#`OT1XBOcN$z6*roe~UwK^$gWqmW(qh-{W~i@_$%`4Xp(~JMaYrIt?LLgluB@wq zE?x#EE9WRYb6+BMb#dtHGMAHX-TvL;Dwu`zuF1_(ZR+r+khuSeYePJUdVXSXVK9hh zrdj7{l|_~F60q{OL2&v0UdVXQr-qN4e}sckv6?$6_toN)>*E^g>OBMpgzkqs85s%f zciqZ?7{~MfN9kr)rOgq`uJyYQnb8D5Kr=kvcS+PblY2uVUYsP9g&EeE!%i#j%m%bY zfXa?+#P?Rv3!?cldQ}%Bs?g*Wsr10jF20o>ss10P3SSkIoVy zQjA6q`iwcCa_kG`27h#O#E?@8vBnrxXT?nIX}nQLBSP_fV3B*G)}VV-(!K)Jyzk z%~^9r5Z-2sj40r~F!}l2O~PI`6Ju4|c0Vz6WtnMi6wh*T=Br&?XK7<>-YOZUVRgf= zR+~SG9HZBh!}%G#qMxfMgD!fnY^k6BW#f+sErp)ypU?fg8-ezbZNxyK{=Vn`zEDg4 zKqFvj10tWc{?DX{N3v{+J4#g;GH~RMi2Kgc<$l* z`(iOOM5=o&V2Jl8M%TGk@o2RwFskkfQw7YqR5nt^Px2WH5BoPqD#B!zM%L@+5{Di% zZLzX~vp4H*Z_tG-v>oO3fv9Jp(X)*JN$=d&Z;K(c&u zyMvxJKQ9wQmzHYx#kq~X8mxv5RnLU|;prMmN5O868<3$y>`H8T{i-1-btEV=UF_7N zreWY!e4;^~&+7lc$PntH%M^T=)nGa_dHrst7$uQ^l)SI$`T~iRBO; zurD~f`MDqtg}&j43Avpzt|{b2LmZgLezY{0Wn)$L>gFBiVXYgEb?`a!&xa2oGzs*OF=Tv;5yUb7eo!_Z<2zqQW9p%xPiHLaX^LDG0s z;WqFbR}P^(z>`)cJ>QEkUfp}bl?VZ@F>~=!(jj~6JvkMcYmc}z;vsO(a=?FK<7)A4 zkENf~q2pFd<|N8W8rzxxT!)~KCxKg41N}03wbyV9p2;#koeBHE$|hFw^iwCq=_RW5 zjr{v83(C`EMws0(NKY%70Mu7XthhbGIC+Uj`_h<9JLX&T0P!p9^M{mOw#+L6>qmdq z#v-Y%7CUvPqYDbIcd|an=~Mk1!s;!V$T01Nk(y5?uBM9) ztc~+66E$wBZP$XRxS`${I0c@UdJMqx&r1Xeb2Fx}krwG>6?CQY@%=hZMDcI1N!^ti|J|@zc$c@> zA01!-C<(XNoX2~D0NMMbRKj;fNELidHbGr1Uqfa;~ve5hpeouSv*iDCnr4k z>IZJ~;)Vugi)pST<$}*ucioN-ep}Xi=i7vA+{AM|g1$Z?8vgPnhnsGb332h{g@ls^ z?>w9T&_VxNQ|&2viXbl#OW8M!_6T=^Yvp)H$$q78YisV?1*I#@Cd1=p7m}LRb$OBq zl=_^pHT|NW$spKXYzTs-&q+1}avl>;$*Ds6RepURYd)$8TaRS_ZZ)%?-n5b6#XV%! zyg%Z&-!2xG4;WaWbI3WP?Q;Pjjr{(qst>!-$^M-CsLcAUA)wgVZi8#r0Ix>%1Ejw54f((KXZO=$U6qiibTX=+60VVJ>(ssg1q$876jJTEAI{DS z^R{8fi=*elJ4XiG6?7vHSjdfleb&|)8{Nh!{7F+Ag+UeEJ6_Wdjln!~U+|ZFW%MEW z1D@7W{%%fCSmf3*Gb}Us8J?K!y3?-!D0~>Ny|)0%9RFF!1^zkr-9O}V6U%Dn zvP$T6^NVp4)4UWB?gP~>;RgrZNyoqU`P$fLwj&v*a|Uj+;usIBR$4;C&*ErTdmS3# z@+VuM@=SrJBj_P^^i&Uh8&^HNv=O^~N^8owz?&ZW#&*{iT-baVYN?!caP&(9*gx%Q zLs)$gfcAJ*NF`MpN#D2^g8q~uZ+RXiTLn?)NIJy|3JMw(&ojF2Etgh=^=^Z#!x$g;Z-Z86df)Pb%fZ0m zDF;sJ+m^JvIuIHP)dL*W3jPTQCoCB!=8~=dXQv>*VJ&a`H9gt&=-*vXp>j}~l<;UyUdN9A8&8P8))5cqjaevX}Fy`hO35sA%%SUZ$kre(oowQFLXaD5X?S za@Ctvow;oZFo z$f}@;rkVxIn*zsVV+AOaw8}2;zk1*CU_=tMj4Gm0yk8_eV!#euJ^SCEelek>J^ZT0 zRg|$}SM3i}o$~1g-Ly}dTI~y2Z%@t|{7>O;6U?X%dc#$1EW8M*+Yj+0LoM#$-WK)% z+y{@#|9cl|40&DYjRZpV5o1N&RZLKXZ93hLiS;A8CqnL3bh)Qlm7213pRHb z3%Mx8z5pv-H)J}6p;c%|d1bXW*Uug|@Sx;y#*^`9enLvRmw_aejQV>#W5+)%%KiMc zSUn7_B8M%RQ_n8ltn{lIej{+CTUlFwU67Tj(9ZByeNtljUiPSXxTc$qsI=6|f5Wzt z>Z0;aN(2Hza0-d=aE z)FXC_-b9Nh5?nbb$Lfqee^_@!kDC`+MWY!In>Tpg4@m&A%jvFuC05Y>y|0?Zsp%aL z7S4y7cx$*-I8G`9mG+JOSN1+0P{8{(o#AObaDYvxVI!VHg#L1lG;qj7eS;8i3skz_ z$kfp0QAf2R_9jMyV22Jk8P(LqFje6>O=h;(RMd=q=orOr4(-cI}Y#*<*v)~UW1lrwy{2+ z{ivds2F0WPFta5!64c6c0OxWa!<(6)k#~To&kTkBi*lm3_=bU6>F9SNTWhFF?(DD{_TI<&B`gc`Q^q~e|2L}mu~kq{ zdPvS{Eg{Q$Hl1=-4GoK|{v9Q6s54i>Fv)YPvRGNUPH$q{U5C;!b9N$1;-{4*V}_bn zr1ZYtxel}=!(QfDae8vi*RvIdy|2XCRfyrg1AD)QqDZ__G}zc_DWS4D*~tGL zJQiRm0kzNJ%}nxn)F&T_t|W7h;Hqk;Cg9kQH}YqGp3VU3!^)lO7_P*W!X`7OzlH1j zU55Fo=X;MejQ>{Eoju|RNb2;(=t7Q`3U_mNKdp*Kz!Ut^p_b)t(RlEk-kgB&&$@BMa}!>@YAyr|G;~(DYg?<_L6sgL-6IQNr@o>$WM0nj1GXon-z{^RRPt2k z(*LclbR|~9V24bm)dQ+1{zL3p;BFnva@Kjo4QpU7e3aQ(rEb{dZaD2)^m;z<;$ z8rj>jun;$=SNkx!u`TLK;KX~+#bTG zA+D~jAp3Zw_bzJcZ2DU}%moqi%4=25y;|uzpnVvh(Yqczx%F^=$JN7QbyV{BLKE)4 z2mvs8g1jQ7l-f1ez|(O=HQJ`+_QgUIL+$&Sb&wLKi^1$Hf!6N%pTy|_&;UD(v>Lm4 zyjG5X+>6Z=X*?s@;5{-77DpE~x<*zne)GCZEpT*w?1pjZT|w@naQ;S%YEtT&@}|qn zNL}23d9%$#rP@S;0*hd87NdIxo-|Jq0U~~VnmwRDoxxY(@|yPA#g_5q@uNA_BlYK? z`B-Y_-TpjQ(#b-Db$M;CnMDJd=OLTW+rg_f`x9eLg}_S6n*GmsCt?XN|HV6cErDoP zkLE(1`)6#K)5%uhVaT@Mj=@&{>g0S3v(w^;$!4Mb6-NZ+(og1X5}~|xOIBJrQ%z`; zK-UCZ{HGt`yQH!95qB22Vc&5+S47_*Nx+8gzs^v*UIi=Q_{EqVuo{YOwbfc-OSY7a zCh8^UCKk2g?adYH%@gUh^F~GCk}aNPwfI|rxs;WW;ZfNKfEPEEAOv)XwOP_lZfjBc zSXVwZPO!cIJwFGkVhy`IDW*8f`rYE6IGnL&#p`DGP;I{`Jlbj~SjuSmaq8xsbFs6>3=0 zP(CVZ<@D)S!8ZPC3VIg$x9JZu!v+E5L(7byd7!xx573S<{>+$Hg%g*Tlwk2`8R#gw z1Zu^j$(-tlK*+2RKP22^QWnn8^n3A{--wAB5FD`?Iys>dRgVI0gEqGBKsVlPV3ViC z4m!#C;fecrZ3JHExuEzD3w~GDs9=0G?`irp&=_zZOwq9yS*DnZOfZ2+@Uf*&G3Qt# zOQar$#Q;g<|GvFSQht=#jEKk{tiF+pl1~Sc&LE;rt>1Vf`zQ&(y<)3v`#W~K*Go(F$mie>oEbTDWD3%4WNCTqv5+kH zK?kz6qoXI{mDQ#oG?}{*i!}sbr7)CQc`?&r8yJOIcm)kT1I&zIbY-Oho`c_ZSjIzE zTUQ6lcVgD{2B2Q{1dk+mfZA)|A>-Yge7u%n{{zw07gQkGe+OvUJHRsAgfGWCf>$r6 z7=b|a7@n-s+Q=KtAdKo(hOM5*ezHPlTNr8vG;j|aX?3?63(`mbVHGi4{gQbD(mgbh z0XC+cYp*c*u!sK_27sHNe*C$I_@9yUG5N@0g}x3+;JkWOBs3yGk?Z)yO=ZDtn~18Q*9IP9c*C_an$vP{!8grpct6q z+#9QML2_Q4R-LjxM_|g&Ea%(L_oz+i(nUSKAme!D zmiJoYFTFg6037SiwwbN%b8-=#h&uqZf76hvf7Rl$wYM5p2xk|~S_s4+ERsGAhXyPEb*23XL_s%1isi zm2B%(vgwR~$WdDL(JdZzBf6*XG5P=F>=x#G(&=9f^++((PB2!|w+u4+q2@gm`C9JU z^|!uSGTV1x%YTr-hwZ|en8})m?mO{ae)%6y#Pxl0>QJ&Bz7`!=X4JbV9N+>{t~tq` zk&)9bHHG#oa~*CZVcCzUQwak@G)K!5R7h;XLv=1$S6-WV!^0vussHN@#~OlwPe1P! zCiDfD}~?^ zymrao@P!6z7QM}r!LU*feLR0P;u^V_+BKJ2*^!^(FI|#(>J_fm}>BY&wF_D-?YDs&Yp^3`O-TBp?3s(^P z-#OP`cOrw=(wzI#Wxe{8-=62DDL3bwx=i_=i0w%X{9u*4g|&OqHug<4jWfd?_~nj7 z1i`cH?CLu$=kORIuC|p|>RDnCc-Ji^>4yn({q&Pzm3TSSno@w}FC0PDwiizzT10QFyX<_O# z+j%IEhw(NWW$&4ygMw&+4Yl0@dJ*dre^5?x&=rg%gictHqT9RRe;LqcPncJ%`ngo< z`ar$ym1|ZwRR%-)D06zA#9f%}_r`43} z`E`7+EzE?u-?41IN-3ce!J$I>Gj`XzSX@{FL)W|>kxK9$(@aAM_Jb0Ov-}7aTQE{; z=C<%D4QkJ+LX98}Sgjdb#uo>HLM3*7x1_Wj3p97kQvhD?+IlAI@ku!U+qTA)0x+FB zB1cDZ%>1pdbCFJxL^~EqhU6sD!ceC1HY;6t24`no zWMum0R2-Mk577?0988BaapS~LzTdz5-+5bW^`uU=JbBqBy|#zyms#hYTl)?4y&F2D zI!po0R!n{c#)i3pcImU9#=AMT;;&kOHV6$@8){uG>si_J*jdso5Lutyzgn)n^*%`fYmSLv}?aYG3$@w5aEzu+<5U;%g})-;CcKSpIs#`D+`@H*m^k)Z{MC#%LBe>vCC@ks8&s)WR@31?q@vTXzk4fV0 z+1>7b)cu%Kgy=R<(fNs-7o|Pa9k;EfbaMGR+G{sK7tJbK0(|VQB!tIF?@Gfy9^qP5 zX*ulei*a^D(ck(JybTJ<&H4el0S%2dgd)r)1D4F`S<1$2X+NiQ^3@DiEoQua% zr_s(;VnB!EYWepTmmU!F-`||5~ zD&-ddci@SMP@yN+3%=p_TW4yOb#OPwHyjmI;%@#ITccBNo!L)cD+5!2^wS_5u|3a@a!Fl$B>>nV-2V@m5iZ!vxU?qo;{+T9$%#mpJT;0(P59`{gu) zjjE!utg?m1-L8SkFUa5)1KyCl%7EJ`n{QhO{u`-5(R8`oMj`1=-a10MONbr(L zUe{@7fNcw4hqaN7s#kQwpG(hXW-8Znk?SfSRAzEyHte3qjt`+W2eTM^E~Ja#DlKY- zdADVx9ofFG^iyQsaV%*B}59)mZ(< z?Xsm|h}hUYH4#?|o)r+Z;Q0v)jtq9m!~ysir(vsa=qwS-i(4->(12lsx1yjqj~Bg` zSyX9wg|1lb)YdiX;xD*D-DkB%XK~rt+1wXXU~?S_U7JE<#cYpp=8V^6X))~h2=;<5Tz|DE2rdHlo(iQBfe6@d8;7yBr} zq~!k!_+=D<*n{uPfe{sqW+=wqjFJO@=DX!4Hvh^rpy+BtL000;RVD^xxJj7p2vLcS>} z0#HM;KYNFKxv=VPzn=X#$QCkor-QxdFGsJKRM}L6JPEJrhVoH>@z$4|HQM!3a&C-C z`3RT=qDA`8%R_%n$)WFO?eK=HTG?J#iR#Pvn3bvMSF$_l)y(ZVB_`pjJtv6v&oh=% zI?xr1Sa>b>0YJNvzhZnW?dzhf@DYx}N*nzB}^;mLv2bFk+1g zc*-%)u;=jcGXnwW%?A|L@8RVgpiAyBuCwOU@d%bo_(W0|+-C~AYp({=7I_6*{EWTgdppzC<%FWw3Y)LnujQmJ2w{qy7joST{c(G2op z?v{Ng+>O8X^|!&{t7l&+$SexV!*Et6FIp&4ZhEy=bi-!lq8`ncEz>fDb1&j&z52V2 zjxRhUd4Yx7xZEpeCq=9&J6>_k8<85rD}EwsZZ+ zHb1o_eSpzbAlj~rwnFFtEqAh5C;b2Ys*bBhaFdEqZM}1;qelQBhqg(BykVZAF*#|$(zYD_I&t7

qA z(5ZEXUBlvqj)5#cRY>@r_26^5;;I)0QMn#=u7-h~7bup@i%Sws^0A|@Oif(02c{NKhn6Sx}hPbzCp?8Oh)8@iTe#1cUx3)|5_VrXA;XXlktyvi?y zo1v+|_s=SpLT+9bx<0j*2#vs%0-D$4*OP?~cx`-$JEG8Fz%Fo=v(eA|HV!P_)Dkn>c`a7gUYqNi_gjYc14^ba82LyC8sLVZO7MKTwEYQgigfu0O#i= z=`fzQ=G~-Sh+yCRt;Y$?Av5WtH!<11val6&RjZlH)y*ySKH);-bbK@wdtOLJz0N$} zr zUU@+a-v?i&#_Xex4|8fM4g>uk`S0E-z8!t_@y4LFsQ>Mt9#l+5yzBCVWfsev}rQ9);pWluV)I#l;PCS_6^d?!*5IOwh^=x zgiWa1v-#jvtYMAPUU|MNWmvy4fgf=_wC|U`+(&7PrX(sGW=lhH%Bg_`RCv$Nlw9tB zqu<|F&a>}*kWD2(yU_GB0_@QjPm;smoB~BT1%Qja@3(~*cz}=L@%c>CUNM0T_o!MMw^jJi9nIN7rJAEd{Hhw1QbO z_2Slzuvz5~zaOrJ_R&A35**oP`@x-baP#fgPU><2$WfNOPr>d?_+R9`70E%f*S0o? zHfud!#9XZNBX*Ls{lrSE{Iml&SjL>VP@fD+%chksm9%u7BpfLMPU56A-k9HO4)(KR3zEC=As|KB_={e?#48oK2IP7gqG^tj0{WoSnh zYIDs{d!-KEepPWFOre>rSeZHJ;T$_wW(>0R?QY&U1=3bxcs*KGm^GwQ+TC3K#-Fca zHntF&ZF8fuY3vIRAiwhD#1aomz)on9rv1eA=p4MK~ z=R4BlLw4(b>$fS|me3Eqd;J}|(dR$+-6}Su*ZB-Yv45UMSI>4FmTv_!jT;4B{dm=E zJs>D-O|iM?AoE(_$E$qaGYeOm`M~)#Q#ck@arshpX{BEyX||d6>N6B2oY>Py?N$t_ zP)CXP$E$*`m?~Ne-XLkr82p2+J4P$0Dr+QDu4brI6dhzG=P_UVyL4NT+CJUWA&n8+qC!id!C-(#)!@buJUdwKa94Ph1SnI+e4 zJQEv8fpU5<-V*xC8#k~#0=GX)saEhQtS=QkA3g3qJ&7?;A-B_su+v{+a`Yeg+~a;{ z{r8tzJ`r{x8BgoJp^BlVr|6yq3L0hVUEAOMXjj*I3Yp1**kcoof0c%v%D(jql@8Zu zW6`kN4KsrP#(|ooqGCx5Zv{GsNlh|3bZYTE7jDE+82b4ww&dRU^Q-Y!dY`;Hu!GQm z_bS^>J_Rl#+{B3Wy1=bPB+&=p7=h1|yKY?PV2I4gI~y1V_~l)zc>dDZi;@y8;IdMj z2qH(v?{+k2pmdHs-xcmJa(COmRR&NWFsG;is%b~%`!?!Sk5zUeJ;p|lJP*B`WySYG zJSS7rsWNC+J~@j#xU53T+~E|v}Vo#<1|jn7y3WoW`}Zir444a zVt-I8^2NUzLKuZt@%g+?d5~X)=<#oVa#2EU{i)Hccng~KkPdOEjG<@u&-C42!XEwU z%p>&pv*}8k!trsS85kA;&vQLIb0c->SM%RSo3P&tj(MQyzg#=%IkYooTk-tw!MFWh zudtUOcD)TUU_-nDCiN@es~~L7CV?U}(o8LgvH?h7wt;hxY+x^*Vv4VvbV)8LtmD-u zDVl8PrJ*9X?*sWiM^0#ASI+6|XiKROj|#$aEzNn;$YWH9P2&^P_<}B-h|Z4DL7M@_ zqZ%k{uU#ycF)zh??P#czonToXH?g9pTdfdeebP;AY1{#Izb^%^0`1AneEwcV+17$F z^&C(GDDD_9C9BW6J@6Z!4ME^1b8AqW19Lw0xFkIFQ+YrY%zmFP5<1na5s`Nfp+jr~Nm)0?;e03=)cXeotbl@A` z#VqXm8P;gwJV_pwY zDu`Hl_6e8oN6u*OEnN~G*ZGer70~sUnqZ4UkAguJFu5n(o4--H<0&OHq?Lq^yKbA` z{CktgG4w;e%`pDP!qrklrW}zi*M<`WlCMsvNq)K&lj}!vTRZ1;%_q%r(va$*FZH)3 zs-Y!iDBNA=3nwW!0O-u9gGsp+DqUlx=U<4k@|*NO;rbxX6rZ&!i#Ydd+DZ{euoOXt zj1vRSf_HN+vMU|!$qP$#J=|sq03H{rhp7M+0D|O=xh}$P;U0hfiZCDh;;(@37aegQ zSWJynEQ8z{6}FlB9EexK!O9-N-|{DAEWFRFQzlD0yupyU>c+a{P8t^QD zF%nUn5K7GmwQL|H%8JOE=@AZ>memq61y`c>{Y8ja6TgThH_$l%UDwIyRncI;`+DVF zHu(Tm*!*kkIu~j=bk>2ePfN0!o`#T0o`d)A?Xb5%ONdqdTLomMuj@YiYS0Y&`v<6D zrLXpjKqF~+A$si-3V~y(95E5xtAj%`uRGaF9TF9ImzI*I37~s(yC1odhCwKltJ6~l zL?&*)qXAlAd-v{L$w$pVM%AVsM5SZ0*(dm`Mj|ErnOe6%>TZm9ZY{@^H|JKkZsUB2T-+5FW z7OKPR{!QBTUv+Afixks5J6`$h!ju!Hb#-8`PV=EeH)GH_$NI0FIPquFP{mnduz_t? zZzjk5(PU>$CDBYkT%d?0iNJrqX=^3)ix(XKAT@v8VtqQZPP?cq2oFRW!`Btm+b=Z)d$0Yufb7;lm=0bjI<4b-&6&x7@Xwg=_=XO4KM z0`KPB-LOZa1*%S&9aXnSj}JEyaLi|BOL4ViT6n#D@L}^FxX9*dcvwKR>)rU2k?K z1VR=0V2VDP-MM}?F|mjIw92Yf`qBi5*|Oyz38kpQ8VQSm9``e++U3t|w>Z*_wbN)@ zrEdwcGe&cY@_zz+GgI5rjDyxk*(jiooV zvije4&jpIk-W3#_9vG*2zgGr*J%(Q1xeI92RY$$2^bM7Sd;!IG63RE#7DvHLl%iAE zLd188>wRJ1QXXd0LsvONAAGqO&AsS#dWzh}j`GV0r@DTFFSUy|(iH|89@3!w?*D(D z8$-}i3#jI@UxJ>HN>$akp+G$txTsAu#QE&DUWawjEH(YHe{;vdVUgtY!MP)*px%yP zN^i>}07e_VZpcw4^h*V))GUj?rvmq?yc-dBrmmnD6#%gW5T4|XZx(fSNLL*U4NylD zIhT}9?pzF3RJ31v1b~|Ej7}@IE#hmZ`@utcz0gN)V?Vvh&8qFq4Ix`k44?Uhxag0<}`KnBBt`yy6I%NbO&vpW2~Si$VT@~Mr!3hKFKKU zG0y^Hh=!E`&e+funJ=|cqJ{6@OI96Vzp&8mySmOET66wRUKkzinEr3yyhGLKmDpa2 zX7Y4PCXGd5afZJ$y_IeV3itN7IFJ54U9hkCfC;Qva&1piOJM}U;5-?vx(VIm5_7d% zIsHAt`U5M#R83B{BmdF;s09T7`0zXq{%>cx^HqDMGeKw^=QA#WsJeuKy)XtB=7*o* zM=hr-=WT~LMuZvJk>*oR8bm3Kq+g~+hdedD{;+FALaJMpkT6S0VbrMmANR28>=RSC zNDbhe$#0TTG$~D@3N``xDkcKY1;4=XSJ3HrlG3$TeN$_$yPD^bhz~lMYX7)dc@H5y ze+F&u{zG@5M>vV@vbKo{&e)t$FGfL`F#Yk6PTjgnQ$MmC}p zR7I_J82s3)dCvnDOtSa89mvmr%cxc*m(@^P*L{AfL6(=~(o>)fLA|-SQClo6RRzfE z|8c9V^Ei;Hb(B6Z6Ave*dtv_^uVPF$E7O7SdU7h{B1<|fMCTHV7Up2;`|qU3A7XIZT3B4YPZprboP26a zqn5d&M5UkH|c#hGCbAw4>RwCvkD7lXGU-JqbfLCKgN4ILRC+ zPTN_QND{GdEVF3N`h4a;^4HYJpr}k4p6wu0o~$@&f1LwEMr$p|Oiz1(%$iV2>LpRj zTq1BD=@_TI^8lqz{<7Lwah`f9UJ7}h>PsUV$-loROf@iDStcsG%d3`bF4A}*oidT@ z!@&6a;vY3Vdoo9^P8i&;ochZbyZZb*h3bG;sA7y2k z7`Nfzzh+eR$W?m=T9jG^GH|hXdy@OES}Cqd17nEU6(EcHw>~CsZra{L8B6L2mQ(&J zwdYzZK^wN$uh&Hjh*v%O@8Es^O5N2$4A7a?jjLtaJ4xdl>;a#KAWONYGV8Mtq(w3jL{@i(k(`9)(%CPGWIz-2r z32>u7g8EZ)%o09kpv*r>Tvih!3gmev(#UVAbq~eHR8ZdHq`Cz{M&L>Zs-&Z3aVo@m z*>ut5R~H$UD7=#OlqD!NLxykp%pQS>TnmWa?t7qxxBCcm{E^_T858yI?>duMvR_K? z=$MK>BQ<{w9b}7+;G}GI0IpBc|F5shL8%M=E;io>68q+&;m+09m(}2}#DSirxGc>k zL;~2<0=Wg^Y(EIM_w1L{*%6|R#zDY?1O#7zOMwjkBex{=JOvPofa^*;ST<~gC#1@r zo_7h{E%s5 zl1Q6~l`i#&!8Ksd8}{jGCcnG`9F!@fJ+guJ4DV?J3}?7B9Boej68M8kz0KzTmMSI9 zSu$;~xeXWE%hV#r#Aj)}?Z&TBNJZ}t2C zW7JHI4)WFvLv&a0qAItt;b~z%e@AC=scwLKou+=C0$l#Xe94ZTMb#o`F1&c{?3X&TYrp?2cM$v7Jb#DfXJ)aMc}*9Ejs+Hj?`PB*Y>VY@fpv-$M&t^JXaYy@N4J8`JqsyS+Ua{a*u zJy)-u9>Cm3i&!DsaZeSLawNa?rkYkH0-^bTubc$qOMlPVao@@nSqsF1WJV6|rS{7xls1Zjd3H5N0l=C* zdv>Q05gX=-xb$w}uWPgQggk~_2Ivr^F_S0Acv@%09n^9_wxfTlO>+-s2#ZKB+KB_# znuI`Qr4I1cMLU|4*2&9AX^)I)lU;)JtFi}Efh9VS!0b>F;WpaF8GU*K1d(IN$HZbV zO^H?nqqAqup!;85T-@it<}=I91w$;`D^F237&q^jQ(SS z_cz!yvwBwT@1a9&4G?X>eh}-K8yNwzf9#WshEV7t;puVl_C}-gU`4o5boGjVJqh+HD3Eb~xfGmfSJf0T zsQE-dQbq=vo14Y=Kf%qb(sTnUr%W6?&Br{}-xa+I`T{IBB3u4wq0}X%o&m3QI9;Bc zezugRWuxLQ|2UQ=^(>T9`M?{t?~8sTi8s?Z*&5bdm?!Y*oj7~UTAwVsUJ-~9w>iwy zz#Q&Jbaaf5XKHjllWo7D#>8sXe=4rAkPRYtgv7#m@|~Gn(H<0rYQ>6aW!yH)r&2tTXOyUh86a!8%9OqlZ_I9^ zX`i39L(0TRYv$-nLqo{un#o20dkAQCYHH3Jc$o9#+&*8XS0{{cb%s(1*;ho|Ov8Ib z*}G?Uv^%!6V(}YdlT7{5@Ca0>RRF;Whf-U=m#sY#5ld1b{@L7&zknPeKk)WJ#~

    =z*vVKZRa35|%PSju(6zXHa3eh# zEepNYWzgtM|71AyfHQK~@p4Q<&uXHrl37PjRX0I)J@?kfgTLAYLtknryf`H(DU9Ny!&KR;0%h-RTLzX^ z@0X8DeFYardN0^GZ*CcmIHE(`-Pd&;)&bAWXYU;m#%+ywNMxht*_`2bPt`GsZ|7heD_sE>Qz*S}M1^HLzYRA(t60 z*aU`M-~jr-iXl#7xOo0S^|n(0j!_E4XtVdL&&Zzc0s%OoXK5To<1^FfNs*1PRc(p0 zfYQec841?LOEJ>{IoY1GkxKNPDCMm$ZzObdaT_ulSGjtUAwW4;SSA7j@@jE)axao- zuZg|P%qtu1Es;`ER!OI20>;=+_Fszyhc$#c+8kRpy}M_CmC-q~=is;=L}U(P*CN=M$$ zTaG6+kO~Ly0*vHB6Qo+ zfP;I%W^Q-3gw#r>{1m;@W&z`FOEYN~Ni*pe>TD!Tnqoug6TCQ!42u(l1Hj2CE_&Z7Ag zI@_Fe=SkKMu|0g71;RD==@Ww)i!?uIzP6FM13XSo3YB4;YLzc}bjq@bdV0vY_b(#d zJQKSfkzLjjYhs8#mtRxy@7VjyHUa`pcT-7a@y4@y;my+f+~B0^30rbclT{HC*{`Yo z(Qh~xo0)HHO&3hf@^^7Pn+n2}NVb*H=9F!bG)v6_G&8ayyDX45Uz0gOJb_zhWu=^X z(-S{BcdEbPVj=$T#=*fwOicDv5ui}U-9ku) z1Xmc;D?(Q{>+*^45>P35JNQt854ttbp823XMTf689EwRy?#L0Pn3n4s`*KBQ7?NOx zGzCEk`4)?4f{GE-4qV>0izD!fUDdSOj})Mwr}*L~1_N8Il_Mr{shD?EDP;8cC;*k5 z1~m{bTmCiAWJ{CnMP%bkIl25U&Vf(vurv}Q2C}yDQ69` zQTBIu%a4sfYih9R-T_;364361+gdcf5hus*juHr6l;noRFGpKC6+6+;X`O@gZv2ig z;PMkPlLag96E7r^rje9hsFT90_f#YCsi2SpxI0fi z78RU~;vhDI#jjjS0~*Fw+FYNu%=YF)6|zm1kA&QJlNm0k(^WK>Sl(-_UV6J;53GCl zuV{!{yuJ?d2|z;*c<*kZbE*2vILh_|;EdFGu#ws+_t%ySpfy}Cty8 z=>6RQhUJD$NW%^SVC!GUm&eEx&z>Zd78M+|lW(A|)~vKg5OU<}2Ud;4mDeXgr;wj# zd0uu<5AatcUr0cdz@aFf*ZBdxTiEPB$ekp}v%Ap?!}VrrRczOj9#T+Ltde-Uy-Zz~ zk9ou_3jsNS)LpG46hZlI~5U5o>rTv#bzA~$pX&d((n#iR4 znA|G8=j@xAeHS(Ipq=2FebcHVC%u+BRZ3yAZT5j4X!nFLD&AdJr6DytxNW&!HNJ8u zxNuAjIGiHRf9pwM3%Aae6YF4qmTmKFT=NPJkoJu9p<^~ep!*5-R}Vp~8*dV$sIL1u5X{UCAiFB7k$8C07Usel7f=_ZQR1ySg;lAayzIOd zL}~rHz}Rifw4}UT=jP2Q@cRH(0+~Oc55ClZ^6+F`&L*^~b%$y&T2#9X3cZZ4uBqYS zN|1l_FG9Nq5Ad9e9p8J_@s-sVI!foz1Uhpk=tC&F?Kkg`fKpDXMx$DGz$57+YuGrb z7SuD0yLb_A^*=Q|&;_E&;xA1ssnk<1Us&i3SaEbc&DN1TU0A7FLmpueUxJN>cQ-ddHp zb=uGf(5C~M^oZcdxi=dHyh4~ADP#4~`l(iIOWD!(ga|O9Nk%$L~THkQjF5Z6o#z0s1rl5ep zK&4%7*Fft~Q71tQr@Ew=WVa-c+8xgGse+Mv3k>q}W zS(yjgaph^IQacRK>c1yRQd@)5CdA=>Wz~^$cjoe3y;e%i)#TzexW(LD~cc)V$IJi zV6sMUl?FnK#!;azP?ZCgs=m_2z;jY`Y6P z{~&r&uH%+tNvi{GUmEXiehO7!?ZdbzLM#SLlf5b;(kQUY52<+kQ+L3 z`JuC`egzJ&ss9g8XB`#w_O)@k zySpU>kr*23l#oy)q`O4~LApb_yG4`+DM7j!x*H^=yW>4`@BO`hsdwGAX68F*&e>=0 z=lMLHAh=;2aH~>55~JaElAgmTG?iD{U9ZeSVE5n4Ky~JEKUc@DiCywRPxwpeAc!&q z&(SfATtRkYEHwXX-(iegHUS}GwR|L7Spv+lqgwqJb^J3g*Gngi|y@{fjEcn`-E z0?Y#QZ>aK^j$t|q%<0Llo)8@DkYzXPb%j5Q+?;(eZPB6(r3Tbnf1}EziFv7xe_wVR zV3gjvg1{UYjm2Z>l*a21VfM)5tdwRA4<`_#G%0(BZ@Y3-8y%&F=ZW zqmLd;pi$PaNdW>Je9%O0V+$_ZP{tJgJ@!A8@=Zu6;+buu>-v)rDL%>N*Ol})DvH&L zh<_Q<*$KO+OYJVoeko{hUaO$(>dO4_!)Cm~SS5`sSJq>HG8X9|yxn@-PLE7~9i~X? zjb)sinkobmA0VTR;5h+_K@!W?L!ZtD3?F#N%>z+7Xv{nR{r!Souf%6`Gr`runqx-_ z55i^h1TqSralEFVx<69rB0b_E`}bo-Yo-jwZ_PVQ<5qquNd@skHTj<^2mG;B>bTE) z+7mYDipZEe@`|cbj^K){EAS%k>+1Ug?i0{?HElP>l;dx>b%4dtO-)_Rl%?<~fy^gC zq)L}%)y%)byx&uSvRx4`#;gs@`{iJkzJK7SFq{l-zyb~M8~kHpt8xR%q|d+w8MlOd zh~W_iw+i@4S<0`(t_wDYGSiue9hRDei2N5>|80vUEt5UXbC;$vsW%4ce|C--eA#hi z60jaaMDzx*N;`EOitA#9JtHPub_ zyw4w)4V{9(n3Cf2dnzrryN0pO(~kXF5NiYwx$UA%xAtEr87o)Ea>rs2+ihQk0~e-e zF@&?pRa@HtwB;f!YT37G9X9#}@xSbC$2@JP7qLr$*bqgkta&ow0&ndzeF{n*-sQ3t zm&iybA&%%Q%-F^G9Q#GpATM&c0`@pK`DmI_fN9eB$8`ZBd=GE#w#3U}Zh;$+W>y44 zjDi%~P!wF6jGq+V0302^vhvFNgb*C&hU4H|F~0{#l?nms7BqDY?%%R+|5vSAbxG!P zdv9Pm8S77I#e1d#j+W2MGg3cC^v%GyPVSpM)~w|20%-?ZY!+fNM;Hj{woq_%3;4L= zHx`JV2|JisrmVL|jtZz$s;SfQF-C!6>)6+hel;Eb@2>Hr15KdI*jJIjPM5J}BCu!PB>xbA<=PR_9{L2&`$VCKLZi9iP?<>;+$7@<4kN*ff6vAJr~4_-Fv~ z$dL)yj(s2JtiSyQM)PUlfQexuf_YkBpo)Q$xlfHr`4Yqp!f1U3iN}DMnTe%Y1$-Cw z02@AdCT;ysH@B7%DNb0gT?Tg}S56zK$Z#-g#$YE)aNf*Z4B@=&;0H_FRP(smtMr-@T8d6@TE%kHEWZ1d1y|-u_;|iH?on zsp24l0w#AfkI6r8G^tK1Gd|X2F0FlnwI-fL1J47Ese1sgBQd_9%|l5VHzAmo#)E`u zbHGzXLz}`tT$aLBJEh$=WbgT#2<+%&FdwPTlqCy6dhz*`kyEi_ql(A+qu@Kxh+9G; zuL;yP6D9&;41_}!@sGP?nN3mkZ#L2A3u!1NEP&*0&avhGAEY(#fKS1gk<2f!U@E^~ z>v2kPJ>K7$o6raZY z=8!Oiw9v(Bazaf~2;9mY?CZm*|61Xa-iyp0-j_9XL+u1cOg=4%{*CrYD63~s6~Br7 zxacx==<^(=$Qkybp>5oG`ggh|$-rCZN6FsnH^fS;@973zt@QFEVz}Ho@#VK+F|y_f z_$Iy)Jnip0VF7jo>`Xc_+RKIyC&^}JvB2-N+sDcnAt(i|yMavqv2=8@cA%YaFfqO| zQ263TD#F;>arC(ax*4h5bPdrctI&DpNW{%KqA2w({GG8 z2?fyZl$D+PX|2iA*;}Gh{qgbsa(6S`QPn@FZsEfhV9xd{uxtZ5(jckvoT#WMvpHfQ zM*L!|#QKEeivb4tQpyVBZmYeQm)DqyyCrDu#8mASs-0+a4!lh6x!vCp{%4Fdt4h}@ zUG5hkwI8qbl?esB%|AwekF;$U@cq?FYjUT4G2rMonH$u%h~l`6Qe*k4x3D(3IuvCF z&v(}>`jn>5a?(l01hxPj%)>G6UtFJNs+Ss6R;Lhrfz6@*7Sj4r^x`pBYEQfL_?MjB!Nj1X&QW!#-E zhkb7XWB=+LD7P`%Ta*cCD;N61G$s?cQ~*1RM!WK=bv(RDw*0!>2u`&P{B95>`wjTeOuNi+*bI> z$QIMO0xGQn80+EJNk98J;Pf+~(;Vy^bugh2_+|(>TU0F+mpAJg-p)>9=n)^SN^aR0Hy(uNw5_+{I8VlMJZh^!q=d`D_(0O z_(-v2asFC)i=$I?_HMemp1&cFHh(v=6^airh|?Sl#;2bFzyO(nhf0b!$FoQ0qGusJ z3)MNP8)VO%$%oduNKOANV+(Z3v>pV~ks$;Em?e!F+MkPy_d5m0zt`527^Q=;M#XOT z;<2RfU=i|(>T+VIsl(e~R51xei%D2-l@b4d6a|((zn0s@Lj$nahlt-9^Yr(^IM5*M z&MfIa{T(8SWLVybpcIxgd^9lOJN<;XwaldFcs%m%@+$N*qBOiWRwgQtZ|Q$&y&!0P zclCY-$J;*R_|fuE6mI?LZkbQ*rI{vwU*Tm|j2zkviET6le&?809(M z{#OVLbQ)hTnc)!v1Bu4wEvnfXzt30w_8xZ6@`#`9v#T{5*gLoe>945`&b%VdTSHz$ z8IO+u17@*0u&53!ZKDl&^yG8lW$RfE(yi%;kK0Z9m0D|BPMc$6!+})uwuDlL-lNEZgm%XNG##` z-YkVnp^frzwKMV(YS-hr4O7{OJ!lqvNS4)R#pxKd%< zVMlZfmgmLS2r591B@=L<>wJ3~G(>v9aTXI{A^kQSQoi3{hqgFmq zGYR?ky-IzQ(}tr#>|{a=L(Hg6=OIww8sNZid|6U~m-oilQ6LS;8gKo{*EF-!(XWy* z>K*TJ0;y3vEgCT zd_ny2`qjKn^y0--xuGNSi!=l_tSq00>npG-Luen4Las)mz0ICK2O$YmAX-2{NvXQ< zH)ux83Yy5pTtz!#MBJOpgwg6LYyJU%|2?#()feFgYfdi$o_Ksh=E(Jgk10^Q$phjE_N^U zAV`>X9v`|YIUsSq>V@Ez)q#=%nl3PG6Q;ax)Q|Q&6|Y^&2rPwv)`LDA1>&q^40mTZ zI!WGQZe;jp2FBpj*Kyg0Wtwxj+%YJUIIfqc#`2PKc(~*UXZx6)JszK^7#38s#T*(V z1hk}HcKobAQ^J4F!qStrVYPV+1on}z{!OhCt;k+w0BwGtb@#66QYksF#R|Y}R&g^z zPz-?Cv>sfB--8~{{01!WRuEu^?aiIk)$vI5t6(AQwOXlopq>P#(SqUftFwmI9~P%69L!oqmX@b5_I-<=pcG|9n0v z2nsJ|DJSaFi!}6J8ZeKRVTUUJtVI?+ucV|Cc{sj%@pKByxlhxblKzg?=nFY=F%INO z(jkhg7lzpb=FCHuyO_H}Eithr=dAd$sHIqo90>zDy^N_*BBzS~9$kG86;3*0vE6-& zak#rmf^MK7s~>W`2rw}v^R!E?01W;lH=4#PB^8w#?=Q zfa|F2>f=Cgu=6YbZ_S6k4|WzAJPh1MQwulVXa16`<3>fi0sTBdHv1%|@b|>)>5b=P zO9Jn6zA#w`v~A86X|F@~FL$?Udk%tRYD%DpM5!^h!|wpJkp(J zvzFqNirhKFm1^XYiC3D&#=U+k35a&wAq?VU`3-8?j6_$ihbSi} z3l~9FoD&09>!#B!fI5SX3(DNggjn#$$W9Ok91V{d8tAN<*R&tHvoUf|Z9e%D@_kW# z*!1{lXR_G7zZdMQmbAl2xc~bZQ4=wjC)lHg*R>WlBy9cq^PXO2Mv5|l7EIajxGb=^ zTsgQtm~E%58qJ&^*01??;9%M_*3sE(X={zKa3JdQ3e0flsV*Kxi)D76d;3c2kgu1&cLA5PZq0_?LVbpgv zX|`lsr}*7(JJV+Emk9^=PEakhJd#q6(^5;bEhJ4>rbzuFqv1W8A&;9DbXzaz*6@wC z-sRYor<=?6q@ahUe#QAsMCO{xOD)IZPKhrSDxF^fO+^&K!2C^_j{cRnpPQ-#Zs`~C zO!ML!1k6#YqL9k^WZ4>JB6n~+f`g@^K1Cn?-rr=$a^hhKL|_7M`!X=N90droF~S<^ zCF(~hrcsigKmVFs(k_CUEP{HBPsm&bb{K;7C_|?HX76nBH|0urtMh`rRA@yI1w($| zY?tE%tK~&HPgm;2%w%-f6V!n2_IOo!-c_ z{Z&4^YwA~SY1p%gG1U0im(H0N5cD7Mx&Q1T+wW4G&-)p2J{FVDa$2NJeoYK2Yfso} z!B@ASGMOb4dFZ-;(_lfc?n#*~ORcMuWkGjU6MK`pWj&I}}n`-@Kfyl^vNL!0+GH|&c zdDHHTBEisR#wO5@jS+IHE@Ju=4bxjwLxamm+xSfbfwGSf*0IjUFjtB8_lOY5Zjh^W>E2C$$+DB#^SOj zh)LR9WTTL+_I{!F>m%DiFtK!Y*#>0 zYOxl~hQ^O97cwHFpqc&rAu&0*E~Tg&Euk>&`YJ>0;mTALWL0TleXVx~T~j?q#%#-i zYnh)l+bo(ncv2r#KYEg)b}Wnk%nmgY-VjsQL{2b+M`j<~;GgE^;-Lst<>DQFP1lTz zXK9fFxvoK4a2@{J8p`e~xeON3Zk#&P!F)7R|KaET{rz_Ry}%DH zT^Ips0$ja1hZ@~k5aG!MI{Ne6Bh#4RYD{5ERFKqbVnT9-(#~KgZXVdxqs(YrS)R=^ zmetkp?`Rl(HJ<>)=13UmWf`w^`VQMn8vfVUpkWoNvsAaWdz+ScyYqJ$2sEZ*j?_+ z{N6DWoq$vZ#|ahpQjNVw!Yomz033#W}2d?XYHpD*^TMF>CW5fSSMElliB z$U8$l!91t3lDSx`v}$q?7~%^`-7c{GWtb)Z+r-ecqLL2bU)Bnb7V!)(SUG%AgeSO3 zk3ER+kLA#`Z96qd!||ZODX6@>SEe&h;6)fy)U?`^awnHiFLNI6th`-e%DIenxOXsZ z!tn>ya8<=PEw=yOjeN&SE&;2dnr=)J9fdy|BPIz4S}IR_nICAGS5nM^N|+1K7eJ9a zfeHTe{ET|cU-}oH&jhXd(a@F2bC}V?0;Nz=Pm#CxSSJQ2h1Whl;Y*>~$K^?=r9PZUav|Dj0j zb%A6|u(T^v)A^L%Af?YA_{lDNco8xUq%+gtYl@==z@EpHw$|f&{rWXTr@#)KweW=R zxJxc9Y87`mzx7h_FqN+DHqcR&M4yzJU53^F5^IA!Zo_WySsCKzEUgt5e}J;tYDyBlhDv$utTnp zsZDxvR6R|TrmiH-NZp9cW4;coz@19Zf16={5`hlQ{`UA(OqvDGR~HSr8NQ=4(3Sx4 z7JI@fe%+!ZnAK|2KKE(yJW^}4m|55FQgKcmAC@0G4uSFO4ip1Dz;x zg}*a86wVREEaDUpU};DoBMCU=nSzq*J$)vJ=A@uTVI0fDr*NT)MxO}5ocm4oM%0=4 z{p6e0{W?VP`|o-ngv!s+o3Q>&W2K+nZ?_y3gaWOk*}j8vtwmCX^O59+71;;#^tr9V zZUX!ofs-vTxX*rGkAolN2M{QPLzp%23gDv=s3QZ=(b1osNKXSa&oP+&&S>3ugI06L zMzW!U2uYQc&`fCWOJglGUW3#{BsNtZk`1YgwlkyiV@ID`JrD87NN*zU!N5Vka$Z^G z@R)ZO*lB(~9wpaEcFWDwux`gO@VL(G?7@isT5cP=-P0B1jx|=YR}W4tn_`KvvBq5o zfM3Dz?AhRh+PBMM+xC0^MLgJLFhAanDi1z2+i9OrpYVY2?qHz5zkPEseLvb7gjZE1 z&$0YsUSPRe;x(}cp$-=wSzBa#W$mXV`=sE1Cl!6WE`;opaX4L`{cL9S`a{SPi?&WU znvB+gq|7La42k0h;=)Aa&BGs(_8$c%KBoK5O?0SH&d-+|gel*?^LPfw{&qGs#)Mde z{NMqr*Ar{z^lQWH0u8z40XuwNydS(FV(_~?VYI3a`k)dDt=8I+k_Pmvt+e@C`|4Bb zq3o;s(c-t|UOo7u=PMq=&QQ=$$}KAD9a=@jrLo&Oo?~fGFJj$cz-MI*<#O&tE72hg z>-xf2@!<)SCo+I<)2oCp`<%kVis50St*9h|Yb$X>a5Ydxa3i5~d$tB$OF%Jvr zJ%v{Y?QX}Zg;bnUeFCkSi^ifCY07AlGsh`CvIohcxPRQ$ZSP#4cD9XNR->^9zS3#< zKFR&-q=Fi5?a#YTuKb|u$+0*vV2CP&YH0Uv+Sw&^`&(Et?~Olr zk>il>=z=h_fZqlDy(W`MR!!4HY32m$(aI%78`9QBtI;t;@YX4|_6XUWvb{+;+{}{U z9*}MI7*D{%u$$oN;_oh~5KbFq6)RzG2BDrMsU6YaiOQz6cwT*iY+gIRc$gXRNsfP~ zMSU6=MHt>1h5POzo7RVHv*j9$3tLr#t)HG5*0oN*`uW`R@!I?7OX#1V=RZzS-#v=K z7TT*0+*c{qSZlZ#R6J3x`t`4lL_)>GosLkSW^wci&bb`7FwO`C^DQOqa%NmlJIbbO4+9?WSF>$D7<9n4^mg&-^(U9;;YPyKSU!3Awu4S$2=jG4rfc0y zc8mQ|ZLCBjQz`D!Uxg0d8nyarob9|ZUV255aIuTM$!HkTTKB7#AAy^Da(ljx?EWS( zWkcIJVxtMty+>QKnK52kzL*bs*GgB{e{u?)0kU(>p9{kmg5hcwzXI3+j<%l{ez_F$ z<&NSB0uqbf`c$2x;o08oxQ+Q8rF)6c>wJ`G9mv*bjC^K9mrDfuX3Js%?V>f385PfY z4H$&Y36U^HT3GTpn-gGVO`rXxnwx}0~%bPXf9NGZQ zGN;7hNaxQs{&&Vn^R_Bx1VsCO%4iH!F^Xz|ggofs?(JQWmUc(&&yw*M z;dbNf2F-+X&|R|Uh)$O##NeEzpcF(PnRvjd+u$?nqNXCH<6wJivLnx@;2G>F`$ybR z^B@_iuo^ZS-B&ie=6>CON115Q)84CKNmM{w%A29;q8)FfM0iEyiC^=~IzllzJgYRTFRPC&UZr1X8w|w7N6PyzQF`U zb!wqK$u(eqtTDTk(qgUtqw`;(H(F`Lt^6#{MhX9I`j^18%x$ek*oBe!TY;0`!!h%~uvOGO=A<7Gnr2J2Z8fS+sq%S@vP#Y20aa_a@ zpK&Ro4YJHIDZt;rGDAwTWZ+|{kA$lfWAx9!l&3{Iev!L}$~4kyV(d=!>KXPq2gPBL zQ>AwIG(h0-NZCg(BI?Wx%osla$uwgPzZ|X)$Wq}aNM&vVdV>AkS_(9uqi>2a!qh5X zh@4ZmC8?t}6&|mzkFK4LC45hrNU`vzk7J%oDi24&NpDw1HLBa6+k{bUf(%nWcJu*Qb6FSUWgF zM1sZ$Tuynr2lCPBVzrU^9jQnSHms|W2-K@yL}HYTLr!``P*1<;Y4^AV3Mmt5NHUX+^FB5b~qk`e-9(!hHWb$Vu7|-t`;W#>hK4Fiy4cYs^h8Uz@8P{>z~|fKwkT{5$aCMzz58uJxpY|>iBVPthw-*e z4l{WllJ0s+Gq1zd{CvdpSA#O|ZVCL!+Zi%5OQLn*O+gj);%R(~DbbOIR+he7H+R>&FrcwmD5_VUotr5TV?5CQn?$3Bx=7x7z2KU2N!G^fXq3;2%Lz##u*%1Zp|6HM#^ zcATGVJoKUOC731gRnvuDx>{bk zU6F~z26uvr{D{iZPMblN*iX3FO}271JVw=voBL?WEbd7(b=5)q>UG-FGrr{R6eRLh z5uXQ7cDO3tv^PCW6Jal8KjS=pJ;(MEMbCFS1W**aF{H$DLS!GG_wx_79ZC-(kjM&z zG^rIZoN{1KX@J&~igo*N4!SkUvV#?s{X}kd1g_u9e{wfs<9PY)Xw8xM^j(|3IMwk? zFDDie!4>(IrQydHZY3+9I9rOZq^W#*w<;MD@6)k`wLZSMJ4ADpT&?t9J|%M3PHA<< zp=STZEUIpi^mKdj=9IL!K2P!v|8vJiEq*@jL2> zGZix5xz4sC6&gR?YP&&CzAX3hCl$kO9)tataPtO;{P*+Z@Z{9W@rJJ%M(nA$&Istt zAr<%3M!W15!5O9po6g+JeoxU(pY zbMqr;v}v2J_#hOZM%fYAStDHXJ=WMNxjleM{PZe_?&<*KG?SXfZdZZ=n$Mf5 zmUF898Q;`*k84D54iN3n1-{4USmByU;jO|AYA@h|)wk*^{J%qx_37M~$zm5E|(rYTeqQohgV977DGsd z%1-*ovTJ`Um?c%~tzNH0t3U*f6|+Gnb}D9Sf>7V#OCCjWH2M61{gFI|H=T93HRe^_ z9mue4r*v%c0|3u?EzDbHNyE-HMIv01J?7pWN3mf2=Btxy@xrTk)yMPIdx-OhM4wWY zuh$yoaz#5ma(hy+)Cw$}RTQw@fG`pSzV~Hag&hTFGBg+-uX~OaDAwK>e6+s`c*btX z#&E>DkWutuv|%M$`L6k^WTwSu8uqK-48Qm}jB-Y#3E5bQLb0d!U-j<2k8L47q4K*e zfa>Ixj>2vjFT=l77;kv)&HlndjK|~>EhSOsCVJ7tl$2#L)XtQo&=4uJD~*dQfm!zA(dlU596~VEjrHbs1NKJg(6Q9VwqOdt9?_S~rMkw^s4zU~qUIZ!}Sk zz4ReJ7ybZDqT0Y1Ir6Q6dzY)!8m=l}@7kVpeqmuL8S?^w$vs?cPfraEjr266YpNO9 zm_qri5>2E6^)Dz`kB7OEA&xkXTVX`VwWiT8NMKQnk$9Uqp>BY!q>YGb`jXXQ%udkb z!p;^;6ohi@OBU8FnW>7Bg(y$6Wizs8dz|lVEV8%bWBd_S|3)?#e=6p6qfM{#(cbFB zD1_@_q;(D5M>$5Z!nT7JdL>O%{5WHu+=G!>fuYrNs@--UATDu&s))X#Pe~!?uG;1- zmDN)P_pL+g$6F$}C$kdYQ-hNBxdkQ=xW5m@^bC1H=DAPmx^mgSb>BD68=&%Jf9w`d zBlp4XEbDO#) z-s-Vrf2ykhDE_L=`XRjA0@=ynbEi&5HKit=>39f&yL$8;y^6 z#r{rEGbR5rRHg7Q{AN)}ppbmd_TSQXxc7tf=j-u))k8npM{-3~*VC`9;${tWd-zgA zxP^p-V4~ZS)JIT3b%TO&oZis#(KCK4M>OilBls8H*3-Z(nn*r$0*8<-F36`N(cn|# z$77Y*pBW?Ch~BPJC!;9TRFL4FT+`t7+p3@Hlch6ixEs8-V1th?8DWqIk6h1^8SPu)*k?!fEYQDD?e0_@uiNZy$~jZCGW z2;r%xy$-0z7ZPs~Wq$9owH2Ig=#xF}{cV(j@xA!^(WNvEqr10Kq|9xuoL*$BS8PH; z!ru6$`@8j(%byqWkyBJ{x4F&Pd~?ow^F7c5?EtidWRg6$t)~+tWzuyi8cZy5>1!q& zHhFRCgv+!wVIAs(iyfj#(cJSR58p@qJCWmMV^~=~!k44RAJM0daLL1BKoflq@c%)OO0} z0mnY?^H-029xr`L75W8|OgEvaQ@%54r7x}NR&S410&sOT85m`?sTtZy4>Ui1*9%lc=R=xkTW~kS13pmp< z-e}&_uPL>=NCWP;Dwr9E?+sTjQ6v5HY^51-8+DH|;!kDMj3*`;Jaf25n)rJAa<4sF zj-x|PgA0)?ZhWE2M$JM<&ve)SNRVN+I|OXLH_Dw3t3rD9WH#_B#ncy_@M*=TL*2cl z%&>?1aj1J25~To#ZW65Hvmcg;L@u&0x)v;CxbX?gr410Wb$@>es-756d+P8(k3(+L zc76C$3Xhb2S?Q2D>rmDB8n7!Y8!d1}#)4}(4N%MQ+TN_=VorSW9B1K^e%Nl!qj`7% zirq?X5p4dtTY+;;j6SM~*egd3qsdwQP8a_$k#-pWGJ;$) z*tp~AXg;b+0LRhCU->QA_!K(=8QB%_DYMqz-d}EcY!(PaMZt)F=;TSL*nF&Lug913P-8t4!PZF0SFGSO4v8@Y zkndY%e}Hl;=OYa}3c8I`_GchVMUv%$_#O8MfWOc+E5;B~WYry^Dz^~GepK;S@rOj8 zy!$QP?%)kHHHe0NS^^!)+q+P|QAG>6$$k)OEY#ZF;|FE&{Y@ztfcTkqLm2Hsj6h-U zFiW1!Gincr5b%4wJ47!G9fX9DxA_Rkys%@ue>E(*cB_EbV>KhdGVC9)e(j&i*q(7K z)BzMtoWJtRg^*6p4vkDG#|^B}I5=IMI!aD9j!eU@wJDHvqODaap$p?K^brvy5*C}f zX6EK;rZQ>IkoJ2wYu(@!knL*R;85{1xhO{=B5VaHMt~@gm7J(4Z4ANbU zvv^5je|ml5uKw-pc+h|11^W3M7&}bZV2HB7Zd)XNx7-m{1{_)Z`&e2bgbnrI>eQ*Bb7(`2GoRE# zr(}D25V@ZRxnpa;Q{JaTaCb+aX?}q&;>u*SHRP6%IJ%-0VDS4Tw}mnd3J!;VeEfO3 znQ5+8x1?8yH|LJ<1d3zRad(t4L4i^Cfsf}O)1IiByu~0=11se6!uI|$Akn1;PKvb1 zbt67bYzN5xanE26pDQKXULN(j7!ias6$|RL`l2RE`|QU2#+PYv05-8 zb~gRfz=}|nPy0x%r4FmNFZd|LJnQS^cIo1cdID(0VG9bhx{N?GBo{mvr19^{{$kDQ zF}z9jSyzL>Zx{AaSKpd8|9wV{KzuE6AtDh){zS(icwtw;Xr|830XxHR`Zc1+1E$?) z0@kHIy_q_GKglYLNytcTLFZG=V`K}H=wb+U@K!}HsNjbEQ>gvq;2y9E%Td1uF z)@^$l7_%u2HA$RL4_Ds#ZOT5%9(|J`#9>UfD75 z5e#9dLKU)Jc+~QJH47dWI7AM=M`YFs`WCFd8c#;RQ<;BhoGjF zyPZC8jkdZF6il?IN%!Nc+^d9aN~B>I|NcqSJh|YIp(I=L`puh^wN72T1=e^Y+p$?? z|A}T;eim3`bKX58YCQ?~JC-Uz;(6omuRsEs?z70P=a3z4^T4fodr3`Y|2yxm>M^H~ z@FvL4bw^9KQ)|R0)8sDLG!L-a1NhKj;`)pFfy-vLtpm~FHK%#n48$u2Uexm^g*g>h zHvkb^QO=^N0+fBrGg1C`PAS7*dRbX@F%CQe(>w~V5!Sqi+dXi+9qznTL#H~Iaoz9M zynEN4Y;`Dyje#SQ!pFxffZ!ndZfLW)*DG#$e2!LD5AW(8y4=>>k!wRLMCIv*K-Nq3XQq=IT$W%Tj6oVxhK zCC}xH9XXC)zhOfYu~+sn6viQ3=vBKkA{$PBULjJ6Y^Y1vs(up%vu$keAWx@z$e`;LVdE{0Cb{HfEu*8DE*_aY#NdOtgOFOpt0c8lH zbJl!alKL!lRVBke%x&YH<9Bb}PNwaSGT$1{DB?}pXq3F`&%xdfE;-*T`rnMx^9*5~ z&V<%JdTPwTwn4^tBRAvb?o!E-j7@s>foHs&DURr`F11DaB(sUco6o6eaM81rxa|GJ z&3Ja+ICYdtTk@hepN%P9fxvVlO$Z3kFdg*UyOYN*{a!IJ1nXkFHT0V7zWTrkbdKP& zJD$JB$u3MmN^fru!rZ@A*^k#JAs2#9hsyRgf3J9AED+du@N1x1V1n@w2y#GZSx4jB z&?&!7jx4<+jV+SA-z%R``3N2?x(=V%{1{V^@sM;Txs-pfVDU2iN!qqNOjYXiz$Rmw zqvLYkUtYm$$%}Hi8$8O0LSLpQPRBX8FpnV>f`{sNYVi;y^&-I8dB=#lc~%NjU2P>i zd2k>gk=2tIxVB*a?)bR;cyGwA;C<+^+yS(e zwvptKek}JTS%qk@k)UxO!RvcxZh>+`toCa#^`t$hmzAy` z>Tjw+;fTisA?i*Eqo$Q#GZi623c_2(Wp|zxdVl|!lo#q7kjG$&XV(5SKX_I)<~$gJ zRU0g+7P=|r%;p2P^(5c*dukDMi^IH?f$4B&k_Y=z zSzepAAS)-F^Ew3jz-_zjde%TZqiwDqjDOSzIukqnB&fVdg~c%NIY;tj?LVB`80~6J zKjGrW-<~C`jW;Scz*kU=m8{?$DXEJ|M6rt;B-C@|$gS`CvAij99PXg6<)+TDR;yP$ z7BigsOj-xsv+qeVkPVc<^m>TzoVtjwhCVe1z^WF5F(Nk*fH>QBuKMEqwQ?Hd`zX$k z_x;T*&X=tW=cv*-^GRN0&Hwaa@^84+eo?db!(l+7)$dCi4fMMe{p&R9CE9*`&r6iy zq?ftpLU7wswIote(1=*R8(N4ZT5I$Zl9j%GRex zIKF(bAHqO2^1wQm-g>Wkn`@<4Uw&;WUQ>g=nZ!!@Jc7)1@@>eQQ86T1To=!eBSZ@J zABJh(&L~^c_j$e}_wO^Y2#Weh117p==3MN?Y$@DB5v79ZdW!@0+y_>mAH#nTn%DtZ z3$h(p3!rBD8298~1*3GzsW}DGfGgY+J>NeZ(~=c--*!2)9V~JuH@#gch@?VQ6}+3j zkSc$*G|k> zZJs`&7@P>#^hr;$Si71zxo|soVk-;flFSj5C;z8c4kVs%E@mHO{gmIW6cNUC`oqha z&Ydj~O`f$)E!}MSX9}4mg&B(^b4Ep|P8c*Z#3B{R`j2UyD25(rYfP=wd0aXiE)`1V zMlg3oL|`Q(CBd{9uD6(P<5${bM3-1%U;Xp*2n|qsvxDW=rRO(hXNgt9;hLwelFEyx zAroiB6u2Gu`gQ#_j<`t*N?TmqJvCAYwljW507^sXDp#UhV8t)$K?lhAH5=nz+%%OH zirJ+E5Vd7q3hHW_o~lHNL6~o*-TQX$JP;n4mw)!&x{G8NKMF39TG_1MN0<%)xn7u- zpi8E>^vi`^?CwaK0|+V8u*unZOgho9C?LIZh?xJbdSQGBf(F%GIy{m6#{r@s6X+iA z5)`rq4gJaWYooRBf=7aK~%3*J9ZOt>7QS1;95|-wc@nM91m%m{aCt*)cOq`v!{&~G6Df%=BF>8Ks^qQ#Hy4m4B`QYgHA`XOb9+>+i>$zXk*DRE#fi&bcsS32gpnn$6+0Mq8N})2m#UMg= zH0tLwG%KX1!pMWd7?(UAFv(O&g=|~Yn@WJ>=QMENATH6Z`Jw)Xr_pu4U7pq}5cD|7 zftvFx@r<^L2;hA*>ZgMJ6875LG8J(@8J1-+N$uZrcf7I7PK0ARL)8F}&z%)-d3YhI zDD@j&o@N`4zg{g|@!LjFaiJw83ByOdX)x`sm#j9rJP<|-KMGkq0u2S}nSk&;D}m2gY{fIgi3* zl?LCGQ%$hFz&OEX7U&E0oFe2Twfa9?gLy*-2vD|jdDa6aDkXxjS_Qbc3_H^uo5P01 z9~UqjiHK)2-~XDOmrMN}>7V-VZ$FQ)l`QP!-Auo_R{UI1*azOz2zV2Zl+;ZLI%`(E z>3OJH-fvW3me}&ENtm1*;!vO^!1nh>m2|}|fSax>7e)(@LTp5<>P?oDbY0D#jbdX| z#x#aHm>QG9ru(}0F#{YFo37S{wvh-XHYpI^vCd1F&~3@%unSND0$}bo?PkvgeM=w` zgoQf7?5lqsZV!7paK-J;RmJ2d_Q9xQWJix{+~v2j<_2=c?;gW5n5nd$BU^&EM#g_kc!C7$1Igrjz^9Bk`#T@ zRB0f|hDwBXkO%z&NH)@tK=@i%rD;zd@K^T^4uWUrX&h7mi=3zjE3~Bb0to%)ts+6J z&rrA{uGm%1@9cJRb{jSZf^Z+rZ>n7YbmGt@@&9WSX@tBZ%J1u}L>op9D`JtwZHWx; zz@E89`d^7LJUXX}RIfreTmq@kA>?AG__%knoL;Y*mcB(7L#YLZWU;<&n4sZEsu6D}pxHQniJ!+&)iL9X)u*0l+2SNkULvk&=01^&=ew_dH{Y`W9asl&cs+ z!7KtV1{9lGJq23VKt5-5d$I3I-Hlw`dkruWRhQzxb~c5jph?$=J+|GW>g(AIk=3{? zKZF3rtE^!6&~oh$i8sJE9OS;$NJc!FE+ z1wV!_%rRy{AuF4P_zG~5LCuJGe|vfPfc)BDtY&3L+9i|DrIt zQ6sJ0BB(zlAdvcKV1=Qq;$qP%8KBLIjzzi2X-!h#rxqH8wC1ZLA?ki&md?|zpl{;^ zHp$2jxf*h0!%F%wg(78Z-y$k@4S=ejkDABa47_lu2|+Q{Pl`QM_HX>3PI4HpH9P1ZrDu2p>H=)e1r<&=XVIx_sO#25KFczM5%##K{3 z#${sLEn%%!Ut13HbIa3#H-1$*l;@i#XyZ2NfJUqO@X-$p;q?2hO@^+g} z{&-XSw63tQTWdi>Eq~>r)X@tQY@iLqTSSfjJxUh8Sb`^dV4t?|uNZ}j4W5T;(+Q)8 z9%2_nKujTc={qkZ(CCF(5`;gWBgjR-6~03Kci$cxd*>N~M~%JX18Gsmq!Mt*c!ppR zB4*>OOkf<2mhU;MbNKyRmN1?9I*{Ra1LG=7)6Gt?cAZ1!B!+eSX}pqdaXejQC}c15 z^Un7xfXuKJ`G=sv&zv%xJiHt2bH@nN(adc9Wy$+zdy+TZ;Ab(gMpos1-;r{FV+qsSU%c${Y}$G6XXgr^a^uKHH|;efV@*@1t70;DQPl)!y)?|N2^~o4br!y zN=P5Xg%K2i`FFaU76qHaEJ#di6|kM=!Ptx}+_$++?ceWn)bE%vcvh%lW|u*`xo5$_ zB8Mo9;06@s&eG~vp(ca_Lj_xG6VEe0vpx3=;@eXm7YAg3^HY+D$?h5Uq* zbI2x9S=Jbyw~^%JFvp;;#V=X;)vA3{bls@7!eU(4;DVfu%~K#x^v#tHnY(Rh77tFx z6~vDIyD2@?bF@tO@9_;bp7UjsoR5v;zU(J{J3PF9!#^OePIw7vb2wo8^PpnjAP5e|f5i`a%^M53)D0Ir=tpisB7cT_zj_0S%VUW; zCkeYxbT31<#BbxT<(*f+b_X&u>)T0)$34MK=%3Hd5Rr@4?zY-qb5PX{ zJ!bfglsZErVIqvTjQW2>U1uQGaoe^Xd+!~I?7ca5g(4E!>mYlRz4t16l#rB>m33^{ zhpg-^Qg*_z-`{!adERg71Lyq5@4oNrzJ~YB_#4rdk-ZX*974FCKnQIliXQI<7EbFi z>`7JMnndaT_oa7e4W3}7{lNGJn;LjNwxz;ae{7+e$39<*cPfwv^hUS^F-opf%bQVt)LHoOS5~iy*lb*bI3KE^T%Gl2b zI3`N8YPVIZj!XU>*D(i?CRnqIRo3wnAs=ujSZSAO#8soDtKKkMMUsJ9W#>|=`}h-L z1gG8w7uW%-Hw~(i9$DQLaG5UnW^GS@bLnfL=7Q`p-(PC2`0(Ec|9)(Ydzkj%J&#$E zn1^U<=V*&}vgXw=g%(?=fTDFBl)W9sM)01!_$-1y&e7=pPJ}Qc&6fv%EcX6`2Q@w2 zpm|A4UB29uXk1T{`Bj`?9C-W6(vQ;*k6px%15NV2>bE= z$>NK0&Jx2az<;=R|GpwP@d<(K1NGZIy>Ho*@8EheU~-UnjgiN~N@ksSQ$(?1;*Rve z%#}n?=D=N2jUgd2k&WS({f1;`b{!E$xqSC-3oc|V*5XtFu>HeX;TYq*w0Xi^Lmg?U zA|sNRAbYANwR@82-wCA@AtjbR!fKGS%WqF>oe9msSbu>T8*XywYh7n#1|g%~Su|<@ zzXi89H`(d=-O+XhK|4~0PCuUbhgt?QC(plqef*yf8$)~#iEE`kIE{;lX7S+2Oe6ojCrP)2hn#$b# zux<&~#Qee* z-y{xnWoUrAcK5aap425!UG@UBQ$Y6c3q>OH4YE(+0B~w4KpQCps;8q~?7zc6^TP+% zYH}Ny(7{&d=cbWgV}V&mXE9uNs6T{*icA4zXO+RJ317((t31z0ZxkgOKb?F}f)a#U zVvUw%ia-WQZmS&W495I><01&ipw_&kx%gk+p$iREaxEm&Z93J*7m(gEE^u!06G?fs z{#d8fsm8(l0siQxH$0yD2-oSgl;U4;arh|3gIVBlx)?5)8@rJ}OA&*j1S2;Df@95q zK(rBnmpe#+kT0zue?AJ11HC=nBZaHFS;OI_yMJ#5#O!XXZ)_nUIM|y^?O@sIK+6vq zYbtmeYdPob(P;JMpvDG#5l*NYzOFWdG-a0fFdg>os{cL^U~D4Ghj0*MJXS-;!whL; z9%1Y+UCpAb4U78{3As~h0<)Hst8J!_N%p}}R3ogNX^K9BsMofKZwZSIaTw_8hQEAC zUT)UD&`cZh4E8YQqT{^N?6q%|Hc_mI_oO?H7;LUW4WX_AUXn?nmX&Ry-_C zBm4IfK-g#Qr{rV4%F>aZN;rhBp|Rc_Gg5?*@$W2dYl69Y@8x|iO*sq1L4jf%-QGv= zJ^Q&<5C}~NeAIek7YXKq3c9L=9mCkz*sVo)!)!H<1%TKSneFZE?>%~?esg_A*3F#x zuk0QaeS~n$?leO(9K|vjKsyG#b}w*@j%T6<-wq=&#LuszM62WF?T*28e)lX+C_NV3 z2Ujf_2lQi6L>?xf{@oR{vF=M^GN^Z?GX*lSTw!+~HhG!&+0X-h2? zU#Sl8jhz$qLznBzvw9EKPwMLGR=}nQ=nQdCZ81ubEcvteRgNgi*eefQi7ZLCTjX{Y zRDZu`1kJY%=2Q&w(!O^XYURA6Y?{piN zkeQH>4ug<(<%?J^3BlN`1t4xgoM)=g-+oicLyoZhYg@~7Q^uf#eyIf9CCS%QbfrVYByg}-_KeNa@8+BUFP z?`8q&rVjkg#u;hzu{cP?Y7wdY&PrV=q=m1bI*FEZAD;j#2a0bqiYYKC{`2)dOx^bw zR~JK!T+P%yo8290WSsF;+bmX*_N2uy1ATp1d~j~A@Hh|-5Yh^wRX;BKZ)cldDt40% zzz)}{#%$FQqXvworiaTwYIkn)qIsK>mT9+wx0!MHWfk3Nb-Wra)s=WrDcb7j|Vh$OB3&28-4Cs;DeE#&X|L4(AG0@gUPpjT>6w$8f z@uDc__j*z!io;2S`7*mU`?gh90@v(m~CGSeytR;`f3Lw>wS6HI$zYkGV!tpPrhTmUy3^6m5mRSo(8 zu$@*RNH&U)&Jz2d98~2_mm1-|I7O%ybXB^5V}6ku+$6lKd__U~m~5{cI+$MADJ!^) z$SN6DE6Eul*9A!$ZcW{S;MS)CE~;$Jva@%52%7)4HveO1KFMYN{hlp(LhPYFrO8Vao=VKi&8?`X zmzGFmaYwN?HvBxq!l#Tgn_?}?x z(3Xk^`KJ~w6i`!xl^9~xCvC+lgG($F-9pBmXOTkHq~xgzFoRkFn+WXgNt|SbYa~#z z8pEDJ)JsKE&!gT5ckTA%l4Lzr`*&chk(O4V-~An0$Xnnlrd>67yGM|%MBeUJdB4Xq z;N8;QS!k<5k5hhDf)T?1rKoGND5B>Kn=R@=}o+)xs z%)qI+hS5LrAb0^6qwf@l92da55$L~Ixgxd>^qyjFY;2A`U0+=3e9@mq`P5!R$7xIZ zP95G@y2YWOuG-*mbQD>)Zi!|mN^a?{K|}TrM~$%Kd~zFwL2BJ#Ov#}+A{AJFm;j}V z5nIf&U!7;0*Nf1f-ur57D(aDJM7*A`#}6FKeTufm!AX8knU!6^j+kSIxBtH_5l2M{Wsl8N>K{n=(ppN{_K;#Q|~*dJLP zJqA(8*0NbOqY@IJ)VIJ3LC7dMN=@0o@qmu(-{G%Z(2T`pfamCs+5dui2Dp`6;6hiQ z6%0fmz;=0sc`WfD9360IyGXSD;*q>+QV^bVzTbD^Em;} zfPW*ev8+ybj`JCR0bo-$<8p}%`Dugb$%xUfDAHRyZ=X0M zpZR+9*WacuH011DYW1Ioq_zcKSpFEv2jvkHA0NNP|K9_`M6I8!pDEoLL)KmYQIsVn z4ijjT@UB9N-;Qw5zTteIKw7BW`7ktB9KGx*rdc31;tk2=X?q7Q3)-6?d|o3Q$5{;R zwUkm$uqCo7295p{Yrk{w|6g{&aA|?{aQ33mQY_Y-^yJNxHBIxO_{uW#&bu`CHSfkF z0&_zzH=ujuK49zN{{W$$)iVl4zBM~GVl_|(6G)^`Ko)R(o2d+IYHC6P*TO32MQKW^ zY=LCzLUBC>&3_ewN2!`?E`iHK$<(Sc*W-6TEvq!$y)6me7`75cvLJTDdw6ZSn|xB#$nCL5TdVVZQs!{KC7SZgM_`bPcT(mzY_W z7KqKOn$5kv>;HS(A!Pb0XhPheo_BE$=L5*2lY{2(3Gw|up zBJ#!!IJXCK<`DQk4)Y6$?;74p(~mrhgQDb+uL|H=%o?UJ;fM%xl;F7u0tbND!}UIP zIk`Zy!Fz4;qW>N?GENoAe2L0Cqo-KrU!Fj3YD2y#LhIlRx95R~WSym>{%Ppq8%%%d zM*c8xE7gg0dwT2dS!jpr_@l#fT1qh1DrTF5 zHuE`*OlRqV5`^saFNVE3 z=Fj)dNntyJ22Y47ZPEQSLQRekNYojUwl8}6WG;k`O$FIyXQ4*up}lN<_kq{vvzZBY z-m>&k)Rr!j*DwtDJ(2+10YQ`?8j$UX_)VHAQ9ZF6pd+t=B?0(63!!E==hH~A07||& zdvnui;2JYh{Dads3wcuoo>s zcm4TzrH3Ze_O;tkBA%Q%_FkcCN7*N7E4IY29kqL*T9E2D=$JS&X{jjYaT*ylWvv~O z?kKd#xqtiL9cSx`=(j8QA!?4?=vZ+<8&D1{y*I5>CCt)qgt2o80o&o{KZS)G)O+Tl z@!7gae&vbcz-*c?<%Nsv|vZ6j|4;F=O^9U zQx)C76DJA~6iHm#&N!D|Z1bU9782<{wO>mE4>O|q(5i^RmVZ?@9duyd(8SWAUQKYq zh$ki0AD~KRBJiDkemZ4d6*RI>dHx-~9NU3F8!#8WW(nTeX`&RNE(QtdBQq;=T&2pP zq;LY7=FYva^anF!g~XE>4KB+`i*TQLK;7{L{uXnY!oN|n_LRqJrE(dWnLPjVw0VaA zyj89Qqcs`l#jyJ#cG0VlkIxJ0NV#S)_0n>v96}?OD%ZhCxl`cXWQAK5cyUb}mYNDh zO_VGtuZcOblf8{P&>+-~y2&;yl2IB5XJRD=Ms*}$*P*YkKerc3iB89+6mCwG{)hEr zSy_c$zgH8p0Eu{`NnG5*>ot|X556SUEqcjsKhC_pva=O)vaQE(L|)BrERKo!3E9fu zQl?$1e0DD3k{EU7`}G0*`cIR@)U4W#4}YKG(f${O9P`&hTuVs|`m%94`(_>;-txjX z4|S2j-`=bgw+5>&e+@Qom`M8pigE;5CrdD4%PDh}Q?8GdNP*5lMT7BW6*%Q4BVla{ zH)mgw?Kc!cD_3fl)|6eztXb}C)NmOtvpy0+3 z8lbR^S9c)4{JYaSdv*D^5>!^alGlP))<-bg(fCB{m#>EV>TyRKJ~8avouSqid1sNT zI>$S=Y2GRJrU>!~tnxMQ-|tLeQ%-sPItc~!Z|D@MSpd~LZi0|iuQ|x2rWLYO3AY+8 zP^1-i%$l{oxFsz;cQO61kJ1Xm0jGUAiJF?;u2*lzOI~xAsHO!N%u{s<--LkvcTnBV z+rSqF9?0wQ_En@`!+L7@X;otfA_lXcd#Ia1hI)1I9f<)Z+f8eD=f$Io%Lb%JufvRX z-uhy5u#71F5`2|ve=5ATKoWC*viz_Tz3(_DgeMAdScx|#HEToc6i);1{Pk_KW78&q z_j--Md}DCaMLlL@5C~a7gA1nc`O0A4^WmYP?(^gAL$Kx82k7&Gt_uAF#zS2->Rghl z|FwjLR57UKAesBz!(OC2 z--fbrjN+e*GxK5Ny|VHoO2({+VU?({^T_a+-PQc{B+}pc+BTv9VU1_ zlNcn3fu}aMGw{SR4!$}@4m=79+Xl`UA)WJA7nd91i733F9|{x58XZ+Yy%5V+ga(nVn;{72Z)YZoedR=y2c{-IdEb2BYuF zDD)oYN@qgYQ(nJ;*3{AL&rI!IMR@x_*`GFuTTRIU-I;-#|nb zUfE{=g)*i<>z~ICI~PE%83UA0_&^%8fgav)@t+Yld>!b{dr+2Rs1w;a@?VeI`#OZ`sq^a=3{CiMMfKl&<%LLWLi50m%;k2n6lxW8iFkB+0^}G zruY^It>|Q1pY{?c=1L^NFrhS7g50%lvLB1A8s|H!Js}^@mO(paC;j-)ZN|P^V7(Z zGVal>dZTE74Ga_%>sJo{tscbLIrs$}4E-iMgkf+6%)7rsxIE1AvV(j2(5pA`#Wo8G zS%2`IO0ucI0Yc3SD^VLD4x3aoG@`SWub1=32*&~Csj)Jw0$T<{SU)Qt)6Aa;dDK(j!{;cI4J;1o2fQ?@Ep82(^Uqrd}wJt7Qp9_4oJdGc3p- z-pKv!2)}d+)^^fMKW&+~=n{j~Bk`-YMQmyRth!fc_f(VDr5$!{zEDYvPcHzfT^EsB z8U5HjWJmd{UAZa`z=#(e7o(BCmT-7;T>0&`TpMnQ7mKqx*No1=QwId0R7`_6ZzYBQ zWv$MSRir6};H?iVrnz45#mcYhxIa1if1JEe+-&_rOPU?GOw?NS@OcbrK2e;e=pUiI z!rRz)^qXp=GnHuB#Ue`AJUi<>ORq_dln&HR)wd)>riRB)?1 z=KHy2yo=e@zI5<>PxS?y6o1U{ZRn>3#$Q}7KU)v8sq&~1dCZenMYvXNFW=5IwoDA< z%NKSBkkZPe?jzMx(5VD#{hg}VRBGS04H>L19)*_w2-)42%FlZ2?ns^@i@z0&Blw-X z;Hw)7l;oU}(}#>zYOfb>4y@VaV9q8c_aN#XSBC>hLuI_VW#MUvehBttcJ>HuKXK^I z)h7*idx*UHuR2ljwzCouJbsg3%w=^MCq;&C-9$PTOFvs-0*ke_$+J2Q?7;bRr_`^s zut)FTpy%$I(O(*q<8JEK@Px8MO=;)43(>fQ;l3mx(sz5aJ`tLoeP2HZWDaEGSB0}? z`pJ~3+vD@jtysqR8NS}GU&L#&=qWPLt6EM`w;Oc8|KxHrZZt>IaF}`j>68%m_?FLD zh*gWjg~g8GdgQj1=i^{-m9H?Q@m@B;Pt_?dF4?yOWow;Y`w&U;TS`8Rr`tCn5RhN0 zC8r7Y6DaxGN(Oo_xJz-1TXpO1$9KWC+mpB~3`ohO>wV+BMTQ*oePg!{E95>?0-cA8 z^or1Nt4~1NLOb%ZX|bd4KwRFoO+AtAp!)iNMI)Tc><#EotXzK7J8b`XyYoUNr~4UY zZc%;JxS}GD^qiB?=Y+zt?nx5jcM1Q!UAP4`Xfw4PBDie?J0`e#X<=SJV)hCC$k5SV z5*3_|L4T?E1dd`uztTbV^5&yR3l|HrlNOQUj;@&rbR~M^EF^#!)eJciW3e%H;ln2| zNjO+&8abFYz_`tH&BUWGS+w$dW3Snc<=Fzqy>QmyP(wpQgIX97%3FsQbnjr*$z?m+ z`magF780qLyKB371i!9U25Tf?eaRy?5f6Xb{vd{aiQ1|MB1Q2D-b`5b^^>KSRwjp?8V8&ugwBG(NRn-{KN%q^N6f)@ejQZB(`|11T(->b7Q9V8915)> zxaH!{u`LH50o@?stk_`~=_ei;-YHw&@4HeqO-rAz4wlQl{^<*rK%R{ezA!!K-Jc7x zn_EBHxbEr0^7;7P>C*p`@p*{YF%tsKPY`MR9?{y_Cja)NyAnS!fSY^s$|77(zq^4n z;n#}ytC!D~Lbz1a!#7*&HLtb*w~GhxU^j&mqm&JFks!4{Ec}tv__79Qy z)@^8VEL4dFSSZZcK+m~1&!G+6pg1UsFv;u-m1L&2PjY_H5e|6k<6=u%>^z&31MDF? zY~P@>WGmlSChI;{^t;1D07`@T0Qq#jb{$D{OPSu&r$tp&q`BfwD2ntUAVjPI?PizE zeoHu*Ilh|y*r>4?(pEAm!{rIJr$+#m5%3}v0#*fp`!;C$V^?hO;P0GWNR~DGgu}?T zCrj1mh1t-yD84==+|D8g%_n$IGn_IEN$X9@JBv?g+pJShQThRA4Y5+5cxUb?LE*9T&m353X+4p_`4$p-ej%T2@d0 z$TR;WTWoPy4RvvKh*R84WQup)X_+S${C()PaY2K?g=a+8)wiV`skUZJgR+BqAZh|N z7uB(Qz;ns!@^@P>d8GwTc3C@GN+nTE^TB>V#QXmt*eUVTw*GhY9z*D?>77!JKiKDt z`w1a(@0)xrs9$+pZK)K@bFuE)72IU+9V*3mcS~TGlr(qQ_2R#Ohprr*lWZ64W8|=w zOKE>lbUnDWzNxgE;Wm&f0k4cH43xBfWo~cJjUx7eagZ9Qw@89+E|%wqaG8D49?2eJ zy$<+&Ulvq7Byjkz$R3R@WbNq$p5-fm6V>@O=tM*otmo>FSg}=dNd6&ML{Ka9v}`;e z0I)qZ#llj}-=g$@HzhA(15)pqS$3!VUe80qLZMM_Qe!8W?O7EWMV_0me%QA{MzB10 z;>ePbenj(T{^L8VkR;!q z)Nzo|+D$I`OgK2aq3+FW>m5Ihr0om^cW8s6k`>n*mcIUC&;6;Z121HypIq2YN%&!v zUoyN|F-2Q=rukcoOMECcfw3`mnf*m}^6x5(5nf*|r<_T$YUQ*l@73cmIm{Z-e%X;n7S?f&@EyjClA+N~ztnqHBK-C|%>`z$Kdu#?(uc z+_(T&gp8Sf>BR^i?(@9#^|_z_GIJf#=ZkjL8oh)3$nro^^wetZW*(^ql|r}(@=hT zSHenK;Xe43?=KPi{LG+;hDk{!x}gAfT3m7!Z^OlTwcGm=uT9#%UrT@8t{+>j@>hT8 z#1^D;PJbSBjmh9KMMvQOi(DdwZ9f^muXjLInZjcn=xt|@}MIl?HI)sy1 zR=UH|xX+!HV(M?-9rPk7Ub^qGazwX~GO+4 zWf22c4i|HT5PXv9N6*rDDS$DxtbUxxVC3?gtn_^ounhXfx&Qjg^&)u0m&kUwy><-e zvvTNHfwb(T;O@-Yn-~W5*H$vp?ylLr&FlVJf2KfpMTTr}UtHm`rQbJ2oOQuB%}>8T z!!1jZwrT{pU00lfZQ00AOHOVjsvy#2_5hGxRk9`I<#OGG>c0kmceBkZ2^~-&M-hRPix1_f zvedqMiVR07d17Ctr2_A3W`Nn;!~;bT0dVP)*lWMJ26oP1nR_41a5)4=M@JQH$Qi5T z)gKBsQXt}^-D=?F?J2HOJIjAVh8?G7CigOS+6J0V>} zg0a-ME)w!fxAPgjvqbPXtEB9Bf>pSOwBz}>-4xhAQwEXwYPdWVO@i%;-`-qAv9c5K z{eZH|$q}LzD0<1{fe?}eG}zq@y|6H=DrKJdvs9+vXad`S`Yf(5Vm zjRn7L&U|x|miMhN_)e;VCNimE((_k*i5%@f!m(Y5lzsHKu^)Q|S9N~s`H}lzSy*Xh zAS2&V?%kL3zUB6jc<*V)SG9cG%;3CZtOpLKUn?$=^H4kOqJgTbKaSvTq79My zw}WN&Zn=Em@0`z>3XA~9x}U#3IMy3{K#^t)xI#X|CTHZk{KUrsZhH1&?x&0E9}JG) zIi8Y~F&r^Mh=SfWbF}5vb{I)Z23#&!1^I*+R@Zh*E9qwf&I`_wc!d$}&caWX+xRx+ zAK#Vg!Q?lAdGw6WNUStkjAWrb=t0N_7^MHy#SemBkvI?;;HCS!sCI}(;h z0xR9miw}de=7SL`+T&|6E1zNX|Ye6|Z^8VAyy({#Fwj#C~&Jr{0Vpx`6DxC{9V*_YZ2sHWN04fDBAN`i zOkWeEIf_^jXj^1(6ZW0Dk#!|7!0spl+kM5^h-Y?^_ijxpQ&1SSIj|8WE%3wH>%V)e zuG>v%0BSi<>2o9Yg4xxRz5rvN#C_lxUmw4MV&J;~jp(U7TgokS*YAR>P9Hves4rqI ziuLgDNX^W&ZQ%L0(z!(ji=vvR_&mUF?8Sw<9$wU9_vA(Vr6IJ~P{N*T)IY<}CBg9h z(dR+SFFWjp`+n1mqWMCvkF_23CmxgR7*CU`K6wYO?^$<*%XDnlJ9I(G-%}D&;}BZp z>6l(0(&YE=6KqowrM}bi;U(L~b*ryCGK>7LdaxOs>YN?Of9C?1bzF6pkVWU=m z2|!~849j?WSg{~BN8Ek67iis4QX(vgA)=7S9hMTEIz=H;9ti&{uay|~kWfq$$0`B& zGZW0%8(KBOH?Cp+gzWLS+1+kqCErSOk6RLo+Le?otL@dJ64_c$SU*Ubs{=A%; zS=ykHyY(eH_Oy01E0b5aMA0?=Y;=g@zjRuy;${r4Wz4kA#)n}Wej}o^bbvNB&>af2 zhI-wxcR=Nir2f#Q^ptlPE4|g9WQS`e<4aCA&gZ9`mD(&$+@+l-J9wV^dxW5eLxGl0 zI#Dv3)LS6Hs&l5=N(iW?=E+jD|5%9$%3z)gb-yJHi&!&9C0jJTm^nVpk!vk2EnON3 z{|8@ct7N@$_8~SIstk1tL}HN)-`zQJL|?0&W&z7%!Nlf4>{1yQ z>AGzX-ud-~a3bKef6}4F!AZkP%Utw8PJXvJh3RA_bT65r%|R>ad(=kqoJLoB;#jNv zcKCIx#yR6IpOhLl-Ji`p@e3#qx#Kwd&qKM^cjtIi6O1i<^ruRy$QlQhCZBE|Smv_A zo#WvkQ5u}1&Ptn;m$1$}Vf|xVh*^f7(3XxY!nW%77n9>SYljD1&>>%`PM)v{~zM-Ui~~9?B3+ynocN*}kV2 zNHlq_^fP>t_$n=RX2u{_xMm_?CH;TfNUaEZS2s@WMry8kCuczqL2t16CEzHQDKdAI zc+Wt<)`c^5e{&Nfdvkq=BV9EV4E>9b@v*dbv7`4^(sHF*K~PH zrc=i^{1R@zR%G56w*(mK&t927y8D}X2c!SO_h;^v(i8j(z`%Gp+igeA3$}LYeZqKu z+0$bxEgsGnH=mzWT8g^hX;+b#-u3p;zFFn%P_v!N$^Ky}GI1%9az(7&(zD>Xl6=ZU z)k}!$vm(e0<7|n1lxG~V6`qwPELnPL+PB%bwBGve3h~v* zwj$JQ)5TT77qKTdW;)JAd&!Jg+?B=y#NbPNhSB%Q@iec6d$O?sGWjk=5k=M#K{4%_xC z4i2H+EWRO|G{YpeJtMezQI65&l| zN$IIFYq!iM@D_Z8#8~2XELjdP7IgNWK$}#N#lxQY<`y{O2eKg(?@3zZ=BDn z8#fgkh?M5W^_mKwX(PsG>9cgCw(&3`%r~#m@MyS`^9ZmbsHmv8czNq?v8u}A2gNSP z$Oepkl0Oun)dwqS%J1U0G5+2?9hObTTr#o^cxacGvmRB}yaOWkGYyS#8lK|~+RX6> zqe?DM6U_By^i~qw@keNnqjKo7vWfMonmsA1et586h?$?-Zi+j9TBn#%9E}$G^mffm zDvT(S!#D7ap`4n(hSvb^(kdTQfAboq4Bm-L!7FAhM*!DasTZK9jCGvCTGmWFV9M=q zbFLSbHvtv$cBKwlN?zv*3*Ui)0RmHbGJF&hLRA^^osD&! zoy)Ort%o9mpo6YHKezh;(IfOXE*$chDjZx?&(E&WCztdT#bi9 zA_ReBPDYHB?Gz@0@nOl=P3ZX!gl(Qg^mY^_<7^__RHue*KwQwI zu8DAhWz>8!xaiz7V~qE#Sce0n$w?5zVF|P_Y~cn~os@NL&FH@+&GUf)`rw1nTOvlq zTe>wpbXb0d&-u$*lN>gHex`k)+-j<<5^W8^TRE*+usi^r6B=QiIaMk6>iAT3Fe%JN zl|LhE5qCDy^)$dEclKdLGcw|qo+h7PU|%L1k>$-S%ygQRaeliaW+AE|bT`}SsKucp zJMjJd!*EO)Zz8X9@QUgp=XT+ zXr<+zIP&^A#0;A_>s&8i=9y3FafyBT(pVXhz`wcT?jP@@TKrV_|Xdb{fm|5@vn?wpdkuR zwvkS#n*m!B5K2Q_jMm?u0^I8v_>^?2q((Umw6#O>@)(V7BaoKPxn2K0OIJ8fr!YO1 z1P16Z-aV!IrFvh5vJ(9+;wtJH{l!NOgcTyaA0kr3NI=;7CBX&r0|la7nCYIvB4Vd) zKOwC8L)G4$YihGb{Qdxh*R5jvQF@YN_CT(YUlQG5hmvSB%@be!87?9zbix zFf&$Jz3}-&Vd>>|+$q?wyqNwae7EC^!84;+EgN`?192}wL$3_{UEN;$x`%U&q`7P)cIflmXWD^X}C;J)_l9L#!alkUtne46bXFGv3;7Pcq_4K0oK z%Z&{JT>jIM zEcK7+#0jTT1`94FuQ7ihyJ-YX`)i785FYho^DtU`4kqL0W-VG` zYXiL9tel)Ui|(jd4R18#&(EoZg@wykleNDTWzpDC7O$uu3X=A|EH`Q(1`!%+`uZ6r zw#b83STu`gNAtfk^}fN?Kxs_ZG4%Rw0oTxi{_}@bR-pzW8k&rd;snJCx|m~xPC|A} z0#5TIoTy+L+JLv&fkbYn*yK*IJbM_m#M<3y!_i>bhT-QETq3@HP#1AUyAgN|vlWw; zc~Eqtw&u1xBLL5;eJb3lL>B)uLeekwHsFqk?=M`R?hWK0on%i1HLs0Hf8|T_`y_PAfv{8gQG>1SUC^x3qL4gkWH7~WA-9Y14jl{0B^+t`db-17>#jDwxfr)9 zdz^YgHLZoWNSyY^+myEg4s_K@mp3*xEC4ZgYK7*Wl~(v9$p^B>R>g*bw{NTdze0<) zbY0CygnfWFS4z*ba3WM!4b#mY)@8E?*8EgEv{|_Xj`KJ>3wV^&7&bP0-X&!T;w;On zn)oEj!2)ksv_ima0wQoaDjF~pF4~Uy=A?r+vTKE?Dcl0g5`QbV35!z2B>+WLP51Zg4Sn~;xLRivZt=o z4s(3U);D#`PBU_PeQ?HNW_*!(%x}C+?iHCU9|)a3x(;g1rwAm5ZFw-ElLYjFqYrAp zB9SCJZ-hTQA>u5%6$Dj=t@zW#v@2lIOUsC)LAKZnWjl0qR`1#d(n3jTel+7G(E*)e z!t@G=!aHCS5^^&egv3XEAljSztl<1QvIB_VmAr6K%g1`B`JIBG{@s(G6YAJs{=ONI zG15tAtH;KD9TDiT8gbft^uO?jvboqJATp(i02ACCMtetTYU?IS@c!Zt`uIe5gL%$l zEeQ`h3k-xsbQ9Mq#Y0zi*X(uVx=FO+uNG0rM`}@76bU!@=34SsX%Ps~()_cxI3E}4 z(t1qI{9=nu7ND@cKrD6$s7Qw01!k(o#>NRm4C3Zzzc+k=v91lieVW(C?LQY0r4y}i;RW_e8` zA4liz9(6zpqOc~5-L_3@X<%rRmScwVn#w9HT#|MBoGW%nNk_bT? zj!~|xloahguKpqObu%5ER6gd-bqS|AYdA>_m*2rk=UrCnf5&oEmA%LZKnzhjTvGRa z;U+MXMhkE@9b-brMSN>($ycmm6R+h$ExL^nt`FiGEOA+dQm=)kuK-KCFWJWg^Wr3( zThQ3vd*J|NbcZwkWBVnS%0MEbaK{<#4 zE6Byl2J=n~hTBROhgIFg-QNdZIrUV1jQr17p_18?OuGap;}RB8omZ^tr^Dht;2^|$ z8Qt7@S7&L2WyOJKaUGtBj+3_`jVJuLQK*W+;zlHtjV3pH>kLNi({koB-YTQ2EGxP} zAK;0|pK76T6t{|&&go>%cIN|QT7zL!l|p$iSx~K*^S*7-$Wf&j;wD~-WI0KW@zpCF z|AxlOfPjYrh?Z|^)YG7y`c1BP6NLs2#a65^SZE*(o%6^S?o9LtIlr5~ z+3v_>g&st_kR~13V`{!I=eia>ba#Mx_6}jfDWcBdYTjhwtQJ3@`UB3Q(e*Y%A?ZA! zj2__0ryk_u0u1>qVGzkje~#ka;@n$I=4MYIIsxjI!NHxYh|Ed6KszpL;Q`=25(-kE z!C)-q4xhw(0{w6H7f}a>OAflB-fNJh%xsZy(KxdTk@AcH+RR{TO3A>q$O)9$4MusY z^~=H0x0Tzb>Kb|$nb2O+BNuaA6>pWaVv07E$r_c0Y{!Mhv7Bp=Kuv*)7RC=tp!+#h zk()?mp#ou7w!b|~zF^bp z2F&@_fUV2tMLJbvS2Bx&M3975j`PHOLzipj{g93XjB3dJ?^_+Q&qaH9{qsQduY=r%~MzCCZe;Od3vm?ntL ztfo3TO7nHr^3feEx(5#)3`_MqVE86sw}O@+a-pge7SG}=`dRFvo+e3)FuVKNm)zfK!?FvKEbp@Iicrx1v zOa@86k?r6~(L3k8pMje_=XWi?4DqZ(^Glvgu@4B{2J+Dq*}GQbPY6?n+aAT48q z|8rOs4n4LR$&Z)7jC4TY)n#*!^EL`tN)BgZ8j|Ixyei1%n-cniJ$0>}W*?<(-7vjF ze~T?b5-sEQIBP#L;agcgY42XBBd;1+!eMuL4j=a+Gi$^asi3nYI4^JC?x7YN69v*N zLAfuF=-4hTVxl?#PUcO{%gZPO^U4-(*FQ<2d& zB?jxTI-TclMKNpc5El7K%|t%JQ5>6Ne0e{`l1WBK`#<*hZeyx6t1U+(sQ8L7_GH;{ zeTeSD`igXU#@kUJkS)57iqYOiIYpqnV#-PHMOR9%{D)o^!>^$g2`EXK@M1+lPCT$k zF1!WErdWb5GPDWI&fe-B^&ga|lsuiEKc8trGiVq|)XL36%XqU-7?zN4PH~KmnFpCt z!-k`BMDy>>q((ejk!U+4ntvN%GCIq)7EU7H0nD|m38D;D5FC*eMh$DgIJ(p0tDG%S z*wK%#>Ofxf4Ko?oNf-wl_K4d=pu}!}YlyGl)w_?RT#GoAB<*gl_k!4-C@%8$r}m{Jl-W8E`stK8{RLf~m*wkJWfbAW#EI0ZipcDU zQ*X)R+@p7d#U&;%WJ+>G*s0Urg<7qQzw7w;gx}J4-_(bNLZ40mk+q6P5hApCP*Q5a zCd~<~g&HC|H0gjSHc5;~cuyW+J3?I%nF&m=OTnV3Y~k_l0OSV(;SUkq?COke5ti@& zvG{MLxC z@Cnh;(TPb;UY>s9dj97pN`Jdc>$~9AH|qu%3G*7>L|{-sPdWUuhvar7ex0Q0?FCj^G#rGFu1uY!YMV`4(fd!3byhL(1d-ZaE|GZPZ6O%7*|9rs zvDpoUaRLA{o?&w`>s$2>v!o-fV1T6H4C5SE$e}lzN*}lqV3(IAXc5Lo3VYs)N9Q?&elc~F~G2)^ccX= z2`K^EMjm=%MoH3vfq^CVxumreh+B1=(2|&SI;;mtvtQ^*?e7tO zX8QZk5wrFpaKMbyO6az5D#cChbLAW?VZg8KS5a&%c&vY0uN-un%*7Fg74q4S?``^A z3<2Q=*sfN;IDYQXY$Xek*&(a!9tnws?e8tM!V`&p&epzXGKT$tdNU=QJi{LK?Qcvq zfqPp3E6S;Y2EQX1yj6PkX9`LHr2}e$Wc04sq(ncW+aZIpw>Yqu2YA`=;AY#6a?~nY!5=kUxsP ztulP0n9A8%Ue3!Pwa^|M{C{MfRaDkn7q98=?v{}5lrEJ<2?6PDX=z@%QIPHs1WD;` zR8qP-q&uX}dfj_}=ZtY~9A0?ipT&wf=kIy+Fz-Qzj=Fx~9l9Uzmag|pJaDT3s#!B; z8hkcDlFZjSf%4Q6(ceu)HaSGXRa+Zovr+Q7fSS50Eo!M0%Vm$oPQ;l5_yq-+h{!WD zOEmpz1f7US^j`xFFPV8E@t7lCHS>~dVYx~zBeDZcm{FQo7L5eV-B;D$eAzJf^L2*%2x|Xw%JJ3rYXruN}`<^Ep z=>K!rHgSQN9q3P`t>@83>cTcH%E?o}y!>I3aZI0}fP&k1ou>^tEv8}QWkXR>@b51X z{#*eZ2E2T%D<#6mkHQs~RBV8haQ%i9pBE0Smj&FX=D`tkV8s-VAQnZdX)l0=IS((d zJ@A#grk>)yNdJ$5rGn{Z);`cOfzgM_o5093%!`7>oqQiaLiiRHH+EE{*1Mkd%kU4l z>Y@G=xhb}~PsVEB1cyFsvHYoWp=rKX?^(Yfdw4pE!HfWS@F5cJzt;swm*-P{Y!pf3 ze%NXS6vGAg+ja1XLR;AZTJ8Z8^PUn~qLl9dP%1bmY#>H0*SJ!<;;q)clB#O`>tA3X zRls5-%U%>sW5%rK1jMFWzkb8&>?#R<1t+IKtRVHehG+;dL5n^WRHWc^X@20O0LtSOk(bAgLmX4{6FNd_9__wD4>My zYS~``dbj)wY-{uJs?n4x?o}5|m+%+Q0-0Sv&U+R6%=^;Q_P@z@;=Jn7@z`l$Wm0~X zR}n8@@Q^`DRj3B6CbsWC;Ys*@^dyZ)Ab;XOtaIR}wmlmwQdbMVEC8i*vTGk1%Kluk zKR)?6j-%@rQ-Hl4l*%ZP2FIkF77X;gZ(N{^DSJfTlWZB^%C*@eNg?we1kKMJAcVi} zXaxiWh#<(=b?Q%BB*ev+A`8A+`<Dp7Uf89=kXQx0h2LCso2kY zFk6E;dPma>BrZN91-!gT=mbPUb1)rMjPw4d{xJ%xFY5w6-MOlI?jTf@^(m+V@sZ;# z*m`GGqV5Fm+G}gWqoQgaiyj@noi9ImL)E4ocq^-Sctz)fy2Ny@OSFLOT87i%jc36iyZV&(Q#xs$ZaKRkQ}ot-WP~7-(_+B&D;Etg>a)+i1W8j^sgeetr*oNl zjtRl~t0|%=YPIRgL`lat>yDlX$@fizft0Y6taG6q(+#{SbtP`x+>qz=eBcn&643GV z=DMB%cWNSnfA=BTt`_l8KQHZ@Ans9T8L_akqCzetqOj`H)WG6aDDfR;RB(?}N0A-=?5{xylpb z)@o?raooBszC!)+?mjqsU2V_S^Gl>O`ToAAT{Zui0_M}w zA28z`>tSC@$!%H^j?eU9K`#9Ed{tMZ!CmbqUuvz5s@I25+B zaaQonuP&HBzE4Pay)#{bMz7Gs?PNr_w!Y`^lKxJ9hx<<_!yKXMf}unK<|t=p-8!TChawrhOGf~ zh0Cxaa9KiVzdvw1xVOx3Mgsd)3^+EY`mXUFxsf4{Z1s)ZKGL8X@CPqg@bGZm2sbb9cwhY}!N4}5<A>rG07r%@K!Q-C!T>Ju~c9!ts zOouFZq3%Z2>00ykdtae%%^{o%oK_ZRBm7>`!0j|5?3N+KcQ-;YGIAY5w-f*a^LYMT zR_YaL{uc1mdl&<(DD;YauyW<5unF{e-`)_9hz<*HQ)B>~mJk4|rGMRI*QxBVfj}p{ zi8%DbHC2E#9SF9#fDhRbbByHLZ^8F!u>nwXxooG{7QkK#Jq$f6aK6H~OK6qp7`xqr z-QNV%VJ;Um!|Q@E2|jTeI@IDlFm^3}Njsa5#bJltUKQhb*G?hsBLr0xT#(Yv;eo;i z!{9=55n3@D4BB)WWZ-Fzt0+``+Y65h0J~~Aw?f4H`$PO-f?s?1;?wSDQ2M}?@@=8| zEBh|I2*IEfnHWAxiD#Demb7=h*qxlIHb;jtGXaW1uz6OSW{Ypaj##HA^nvQ|3XC6r z#vZ(Z;WCB;LGdP3cei4%MS#1!JO!7fUT6X2UdsM?E2GqdQKkL704sqDp-S=f#`XF- zcvy$s90Ben=42DkIn241u%+>rS$0tK=p&lYH`|6Wn;~zRd~p6H^T)iIWmip za&=Hd_>$K)rR}NE!ME{aaIqdEdnUV{0HjFB|5MNccS{#6SQs=y#1$M!Kipw&el8Xd z2!4Ei<~K$&5WZ^0I}&+>F3aSFZ6gZQL76e&E1%L20$L?igA?sw8x}!&93N&7n0ukI za`zs6F$wopcB~WC_9KnzTk9pgCf5?eNUxl&9H)BE6O+MA(P@sef*@ppKgz7T~yQ4YFjnas;v zz3mIPc|hsrr0KJdD#wh=U0|~MQ#h7FxqeP>Fdf?O=$^6>!OCXh6|DdBO1ek2&5*T|~qLi(xABW)z3z zEUT@SDvSH41%*ifF~<{4p=)tjN@k09{%9}W#yuhvj16SDLY-&IRP>%@qj9_J5XXj% zz3S(S~1Fd&o%YHd)-WOA^0}QbS$#=umMogc5&s zCI&F~cW19bZbmG%Ydt|x@*Mp22a5kFL5u>U)EL!3FijUMu(Sw46+p=6_X`Qdh+l^8 zBKjoVaoKcG$@wV=kr}WMG??I@L-g^K(FxK73BfyeIS_pkD07idIxmptBmQCbSiY7R*%laLi z_Jl&K0qO=+%pmA90Ls~8H9A?ggRY(it1&t#i@)w8O@5&6Ghd+AFovp&WK3LC+phl{ z{~V{j{o94M__}Ue!MioA1cImc zL_Vwg*gFe<@BG}gxgDH^$aI4@CNZg~pX}shPvm72H)+0@2-1DIqN|=-P=RI2m9W7_ z&S!LHwpqW0D#`c>RQBiHc8*Awy|c1`){uJwJR_^N^Y~!gijLy4s0A0Ea9uToHegxIibr&zs&Wwkj(e2N)q#|VjDWyW1y=>aqFEZZtulz1UHoT7IQdfs6*S|O> zxq+bB!o`*hc13e>vYG={{x5TDGu zgNQz;c0jZ}{q?_91pFVa&Mcn3*yeMc=T}up}0O$iPiPcK5{b7_USAiX=K-oe?WCD|S z;^gS{y%%04gZ2f5g|jHV-6mD9|wD937Zh2!&xD(Ot;hSVAwV z7`6Ee6l=R5DAXg3ocMe~+p9idX%tb0B_+@4t;v8F>{5~u0T&B;bfH2wo}8m?ER@=f z(?yGYF8nV%E>G~nl@al7(|a#@C%!W-{$Af&EdxG zcoYJIzw4#row!h^_IB5ijR7oi%C0XQTR!ag10AIVGvD=z0@K5QtR^+4pMElA4+%T% zQRr?D%pQ`gH*S$Fc&^82-Q_#=st8g34o=1GPvH5ZRb()lBm=->hRD}ak}tJh*M_^_Um5B8xCIhg+qC-@~Ch?HT^Ki zR(UJS+eb%djY0>OpY%6c;6>-KTAO#k;w8w`s&L08N=@zHa6z#d)>C%$%4yG-zu7Il zpR9=fkz*Y%O*90;d$1oFVRGwl=}6#Cjd4N513F5$~`%fBlha}1&uKw>igmU^;Qkw zRPe3Ql&G*=$OYoA`gi3Xn$kf3oH zDJB>IxFYO;_mZ0@Ir($OJ2XbsEIUz>s_G0zp|=v3Gp5b}6g%JPM$X&7QLnI0`!2tB;}eY|s{|I$`ER;Ssn=F^#P<@cr;%?l8<~83ZnN z^mkl9c?X`hE9DL6`p0Ck)>tJj-p(w^mQ++mOgG3Fj{040^Tn{UTLw<2Rrb!cnoizI zV)C}gExSI7&x0_RV-~punGL7J?ckz!fO;k7SlqY;&k1|2~@i^*>_rD=0zrThdFm+F7J*_eDI~{ zj6>zWCK1n~L@W7R6}1WWyqLfBuf^YXlnMWKk-7$(0&SWiehiw#dv=E#+YGRYTUL1` zrsB&f`ry(nf6rE{gXF@66PSrm@>yeQhsG^04_*DJ9<9_(xArAI^^gx! zhEq_h7=YE;g-P|nYId0Jxu@M)MXvi&=nHy6;k0f@qS&Y0`xFCHN1%jdSL3;AG~>P` zh;;hF0Ph}U6H*0c5}+n)Phb@&(~S@P4th?5?g|tGhRT2Ivu{Z6 z*hNaF_Q3&uwE*uU#JB#7Ze(W29e<98d|$slTn230Yd~bWd1th(ay4`Mj+lGUDmnt+ z09mRJJ-wtECof9UK0c(glZC!z_!Vwn(uofQ=v63%4;$9g5_!5VKvSW#qHV5^%kcg0 z!YPJ4J`Z;09x1stql=YRk1T5{S+| zuY5+Q9+Yi*3^<55l=ze*a;Bqxzvfb=>+9w)Igs&i%`JQu_3t1t5R!E|+6Id0`3&bh zWCvQ=aTbv5z26!^Uh={BeDGQU;KW83-AW|3HycPpzSMpwE$4X03oadcj*YjZZk;Lt zxkfP*tZ!M2)JNCayLU{q+CC*Vc&<8ZRZi7c>`_=1olddM%v1c@n8nT8TS-TdDU4Rb zHv6IY4_|1hA2)-s3pYh_jnl=+mkjCVXcWK_H!hX_*K^S7Bttk0_+T&T{Hbve>FH(D ze25kDzf=-EOa0n&$E8*}f;*Pjkm+xDV_XB9J67OY$la}EVw_Nmrz6{d_ioy5N5LWu zNq6=(^T-Fe#&V=iUOH`~i}Jx>a&l4tudtw?r!O0aox;4N!2TM!#r4fPIfk z5kA+-gV_!tijqLm27;L_9Sob1R!x>v^o;Bw%Xf+t(c|b+O9-Xez43b0G7EOeciN4% z0GBIAepq^4Sxyc?z&^6)S0OIo9!%#q$f>Gux}ca|by$xFtafJ#S5oe-1mhe;gO*MD zO$vcDgj!Qphi30{bR~02NRJZ63BGI)zC2u*zrPWv!AE+hLgJc`{~6As6?rm|__Gea zm%w-m4p4iskb0dKTdGn($wnwJkwrr+CgIGHoA9`4~|D;CXN_NbaQ+CyO=Hm z)f&&PSR;AyQAb!wlcIThjHEw2!6{5$g6sj%r7J-TtPEfoNl8fso;J+BBb2Y#c*sK~Fo9vmF5t z3a^j+jkl68ZF@8NYTmh>Spdaz0Jb;`5vThitnTD)_l3}Q#h1v+qOPHI`az0WH*BRB za2lF{f$)K#=Y6tnfM}N|T!2d3M5T0}gUSV3-F=d~M$p}4X(2Hoa4Tng>VH-1u_84? z4uBIehwI&{c1>?=>($>%xs3$E3*WMt=2<-(%EQHDw4%XQ*mi?k3L;`VS@p>WW-AW= z7R$m`#}3it_1=t|rDqp1^eag(sXG9+ZgQ5We&~R&cgJJ#RG~7tUBUCtWDqf&>kdsS4)6&Ui9?OSZzl@D-r>dLG@-_V!=|1w(b1f|GeNJW*#R{%s% zwfO;RoPy#Oud^R-sQ-40CoBBS9UpW3-Z$zC3xdrdxt(p#Ur+)YDhq}rYv8>`(JgUAKc z*6k*3bx1)1o_`E*Uf17b9SV%>!^yM#70;+pvkbjfgG>)$JY$ywTOsj3BNhOWIY*ZB zu40c4aDQR~`7l0!c4l2zqmt9>%)Ga7G4AY`n7)Eveah4yJ^|k)HJ34S?RoB@jUiBde}5EE)yCZ4tnKP1#DBkrzLFoO@8jfR zd?m%3)SD;A@j?h^__?X6EhvX$GOCpI4kVPD(r|uOK?HdG4Nh2sQeV0jaz+hXhaah0mJqj*<`TVidd(?*RZd2skTpw zch!UVv9}Kkt&!W$4$6nrqjJ0lp7GNLwi?(Mg-W(6vi?5`K9wpbj$4|cD6uXwC; zoPh5fx`;r^t&~p__De@rAy@7_9H3wufmS=P!Ehc!VL*VUTDK*Vc5L3Mo*~WnMEd=JO_TQJKK`mehO$$Vb8ve&3)0^$x zkv4Ll|F-j7VQjtxoIZQI-EpyH=E190*PRMi`f~_vi-Nx^V_Z(Q@dC(PLX6nShJ2$j zilizySC6qS-YyP_F~@0@kpMg<@x| zcG4w!ENJ95be~3%clq04_W=su<;^j7G%S{!kqHL1|8mdK>?R;xcB(35*eCmD&72yy z`l6(WcMpe%3&si|(|`QcYMb}9xD0`WXf9rbmHo9V0i6xDx_zlYb|`4s`+)HfXlDX? zLFpGuyCXT$vla()g8c5+MG|k3$Y5>=JRd>#^=rgo5IRQ(c>RY4VjMb1T{~97)6o0h zIG!}#jPz3ba6HiP5%u-;(Q$BKfwa3j^mMp2f!w##A9DvD3pmh&wy$`Wzo8lHCV$m9 z?RpHfG?U@AmD&J^P~2(l=i!whx5)IR+F#$8Uww?gXPUF|8LM684qbdXd*j*@j(h3C z263JaUDgKEwhHp{Nbk>gjN=ITvOwi;q~5d4=<^0BS(}1s->>emV}&a>6zf!0IA#T0 zLqTPiWgO01ojPZ;FR2{*xM;pdUoS1e&r=f}9%q;U2?q(E6;59ac^JZ{8aFz+x5CqU zU#PXrkz}Mo7XQHIN}g@*OLPswwdl>os`(?cK4&4QWyol@oBM`pJ5?s-0}1W=Z^chzP-^Qh`m!?0XEqHh}e+IO1^i4t!1duB2ajm!rXy8qPW<~_Nf^@ zSEjrBGfxiaw0`L}C`K4}1*1mB*dwM$12&!D3UuEe5?qefb8LNG;X_Ly!H06sClTv%MFp&1jYVVrG64R>+B zYJu0sNcZbXPA~sxn=7Yo?)M-fYv6*MKB)x zQuc>J>Z;)}yv514#KLi(-?!LuP+fxzzvr|*apx`L+qNU!YL}Jbs|u;FmY8NBe*xzD zS1MN8y|(T|uSiKs%sJD1D-0Z-OShKuX^D$%k{!L-hKOea-J|LM<}+i&v(%31e@-N? znIt#mWn1*-Ry(aqm;T?0Of%?2=7$!SZ539Bwu>pfI>-vtstNiuzp`ANMn^lS3$23Y zn%qk7bDWPwVnQB{kG3OtV;t^+a%l4kSFyMG?y7*|K71c*fHMzxD`%*uZh#odE3@9k z>DfbkU;i;fmZ46|u-jcfq6DMU@$!nh_3n!4#rj_`oh*roTsj@E%mNbm6MZ|zK7j}0 zdR`RyJ~-6h7K}m!iewl)SBQC{$*`&@V{NZqsr77QXOxT777h#{PL(6_q@=V>N9a)m zw_~e)+&p@HPO#^}N-OJ+Y=rY0i3vbYnw=u*!2vWf=KnG3UBzr{19;qV z3Wi%|pG-_>!GTFC(C-8x9MpHZeCYB=R8L17ZzuG=`~+|n?a!vm&aI3qpXrBB$|pA; z1_fISN0pRF_4{EIWsHq1rKf4f@OR-*_MHYUj?y=pa6{?a-<s;~QV zb>v9d{N>9(P&s>~X2efhFgHH~77!)X8PyDS5h&#)dp^THV@&17R3zOrp~*AD+kfg* zJ#3?{vI1ecAdv99#bl-Ws@<}(P7hCokEWG^Px73(kos^e1!Y9FYQzD*70nt z2ol>Ctls_Db3b_*C+hlfCHaPQ>6f1s7Dm$LI}p*3?!)37u9y#T?0tLYC)lx50P%<` zkYcZFUY3K8?FCEa{d5>(DY10!;CwMS<5lBS-4(0a}#H9Riv673!N=? zpEj-K#Dwe@3xWqbg~x*PUxq)JmzRNtX-&Q(*!S3vc3yprlB`>FAo}|L3%d)-4E&VYh<4ui(AUqjLT&73* z{d~|PWu(}hAaiJoQ~c8*$?Lhli=*tD`KANcUq0xUY_+^)-TN+1zpGELSo}ljE0cBT`yKrOB=MhXk;F@mL z3EOet$n7hdICnQdl*>n*!qtPbN$4C@q=Zrez8xYj)i>mnZ+RA5aN4wAS?o{C&T=P= zwXaT@W+Mrb{U*D;I=x^2!?uC7b&d1>C=cfiNABBf%};vpnKCWDSIgth@f5v9b< zcU8O1Bk-ItrSQ96t&5@&MNi=Qy!G;p-m|+s!^gSum)$fmA;FobOiC#MM}Ak$ zLhI4Ib{_w7blzaPW3vszp$%?uAF;YLk^zBr4@D&3*mucOuAFIH$z5YfeKH(3qu1#> z8KXUyi@pBB9Td|ze8c<`h?R|M2)Wir`SPwP=U%NwEPlH80(FxFp7lU&O~$=lRaW1hu{YW>;o346yWqv+4&^5lUOTm~4iWzWf##R`QdFALB?+=gZ#M zLGWp@Oybw8I@2?Hj}q19b=!F^y<3*M;}K7?KK32i9R_1c#PWjc$PpgFLk?pD5KGGd z$Jk;2$f#*Zz0oYKGe;M_3Q2C&ao3(8+r^{wH-@bhQEF|a{~%jt0`4%QIEr( zwce<&I66?@1XA@&z&Or|!dDd=!vf}H6h*1HE~;kuET=A1As{6CntKeCO;vl!nIeP$ zSfr&>kn034egf}woV))%4D%*eqQ6EqX3eTJtTOX{+)aByBhpI>hd)%1y$BDc+&8Tr zMkyWmz4K}tjBoO1$fXI4HhRF0)tMxtf;BrCJBo32l0~>w`zbJQWJF6G_t&@n$Y0ir zVe_4vM%A|H$IZX$TWn(>i+5z`+2i-Aj#I3`6fevfO1a@h(RhOBCYG;15PD0?&@z*> zd0e#mA0U42cs7_lRl|g{9OP%lyC`sGmm=IGT<~PPY<v&0Cce?$%_P2wHSc%+DRZp&7wW;i-umJQq5~8 zXRaZWmdp}PeM=b>UW2!C3H7g%QCT3Zg?%#JZ=q;Xw*U*$;vZY+RLhY=t9TCM9sO6C~IYBYzYw@$~C%`_Pn-^Sig=Dk6 zOzIj`|Bbp$o-=gR6|$ZQK_#@lGk-t=*=pYPP(Wd8o)Y&D0Pc7NR4_O|v7#|Fmr zw|DQ4TCdxKC04trn7nQGKeMzWBeXyIWZ%4Tj#g2*I>gPpfQH$}sG%jOhkVs+ap20O zqZ$f=zE74ul#*Z-8MgZgZ6<0ut%eZ4fu~)MWMSs{v;nHY(ky53fA{xbEA}r#1rL}z=A0&77;u39ZAq0Ons-T_c6TQ@G$?tI5Z zEq>nCQ$7vqlEVsn*dyAu{EMg)Z2^EMg9juG7^D-NiN)qBL`7BPSTVd5@kIb$f1NC> zhT>ezh2^=_;sx?45%kW^&RW~1O(>f3KoTHgOAs9yS&hn|gl+t80OU#Rk1T%-Tud7} zVnQH;ffZ8}DIkqh*5)~Dp1caJ5yN-Kv^7+c@JzW(o6N zS~?KzMPVtSJ{uQ&1}FfY1~7RKYH&F`nP`d5EZ+`qV1@7PR~7h zzElzQb={5I=V_kcc7F2eL`d1&a=0Czb7Gvds<6@0*UJ93%6eM&*=@_Y?xhH)*S)ig zH}iwX{m&&l0CG1}9P(NMz~x_lL10z!AFG9Sa6*>g#kVpxN|yLnk}}b;uh|1A;oGb3 z-zd#K>o00ot1s*m@wSt=bLrz3${%rc9J;@Ic1Gz~Vth^EW02j>eO$YSQlIBOCJZJ_ zD~pQr`6`@9*9Z%3Tl5M77%9uG&M{2OjnkoQH3% z9@cK(nKBuo@XYIDHPf)!q<8}aCRL>OSMG~w);Ji&zkyBCV)b)0 zG(bc20D>h>hCc?N=GkUgENbxun-C@RTn8cRUj+aP^EadOmF zG>qe>LS%3|TCk%)uDNW+?Mky-rSv9O_e19K zTu_|!P8opIw~bJcU^s5ycOp1izS(`49%O2;r7?_>G-Wn~7#WDq;3yghHH>d?mzcIb z5fFu6TD+fxjl8K`Lsgs?hE-RWtap&i7MJutQ!i5I;vTy*xVNON5JE)*srDFW^ zA*`H*AW+x1)jLB(gIBv+5KqS!EvU#jmV z=DmM9%oW+c$1cI4k(S0SKt!;9>Mu8@N>uMkC}AmpvabI0_h z`m45Ei4ox|Mvz!ZUmUvh^>gB>AK4$w=e>{uXsp^l#DLqvsd=@ zyY5NN_DU+9rcV0dsd*jGIs%uv+&elZyFxlzybe>u2>_X_r%L11t6&fvp%oOIK7S1# zx66YU2|O`+Qmf|42B9EoHg`ZutPd(pTIKpB2>;SeWsd@$Zz)N~QLQ{h!b(|xv&%=b z6rEtk?W066MZqFVh+-Pup8BV;8;dyui(m(_i^QDyc5Zm($LeQK&DfXXEtoE!7Q$0s zmOC=&m_~nyIhhl!emr2e&nSC%r54g*nBJbCB3LF#tm77oy1+h5<$*vT8v13af!XwZ z)Y|iM?9l#`-jBg1-;IJwk^FEN_6sDPRMZB=g%a*2GbWcNa8)R|E{^vKQI7LJ6mjs? z)^>z_uUN?stT3|uWa}u~8WeWb`fLA|f=%8oV#{P5AYTvgPMe>&r&4Cb2 z(h$ZM(jH@XOIgiZoiYCvn#KB~3v%{Qh_PgVy}x-|eSr%KRzN(0OYEPKWqtG~1?U9x zwLRYo=;`8&Quyb|uF0fvpK=i>=74$KHlbY%$FJ)@aHQodg%?t$U8X%X6@>f;#2mj^ znuJaR+|SA&uK{7lC65=6)8%Z({%mC(iTE!GbXpO(rVy@B1=?lc{VDr7P-)y<*2-nv z68F5XCk_*-Il8ioW<8Jqt@;OzeFTe`513v)=|rZ^0bRTKe{9VtwC{P0w;mO@Uw!N- zIR9Nq)a_cD>!R!#U7u-%efF!5k_hfkj04li&||K0-X*1)(2$S=^Jl!%wNAn%0xk*r zAi>G^%vyHJhwO$5P8Tgk-ez+{57txK2j_y*Vxv}z_L>!^MP%9w`7oz_&I*9rDsmGk zzt*DtE={XvPZ90Q%hr;<<1_0Bbo<(#BG~U8>me`Dw!QI446(}DgqYkYRH%jWzbC+l z1a`WQY}`q?w6Z7^2k)}8{5eJ9vCsjZ(V;%D zagNs_Hq0`6G6e6S?7-md@N|6%=;WY}CImm~;7Cj!*i<+fG^ECR@Ny~B872PD<}OxaxdvDiOm<2f2qynW-}DOSdRdaZaBpMn|7*1q>D54z_0! zU)v7nH5QHp@on$m`Lu+Pj<=N1JCQ%gmDQ0@Y_~e?oMbg{jq9zU5FjBPbszNiV2e{& zn$H-&{*EO$4$H?>Af|OK;_@zFu{n@(Yx8h+ean3K>6}cnUqZ_7pD zaj?(I&ma&CGM-z(#N)V5DG42tnSbJH*dc-*d$xgdXuN1F0&x~A*{Y) z@An!FnK1NOP}mkx3BK|Y=aUEW@ET}ZX(5OwlcN6G3QWx)K(GFODdow9B4fN#;`oRo zF?-&CIfY>9<4)MA`=MW2yX|2MPJz<+;-~4T23^YMOMjA>iF6_(No8Ng2a4OV~6Ee|`85d3jk|#unkX6{! z^t_)W)C?51><7;)v7R{4Uk$ZP&A~nHU8iGdOBR@-AqM;ToG0sLiBbGCdv`X%=s%YZ zUzaGL0yBa^E0i%h++M?U8qrzPT0$>mYzd!L$GuFyEU5MRe7U=^B!<1{em4sE6Z=Dn zEgC$lsoPMvM0cvxP!ye4`eHhECxQDJq_knv1%#_|q?z)*etXG40)h^#kLoCjfiz#G zgAM{oDGFk^YJ#s0<}KlUz7e8^c6J5@!9n}bk*KrgOR?eh&LSTC!`Qm=!TWaP4w=q~*2&VfQt8cnyU+ zY`!EqF7@HKUfbsP?VveF8q zr&us@dx2}p!$XX(25+437p0h`3~0*aNh0B;@~S2E3)}kCW<0(HTWKMkv(g~F>z+d< z7w%jTJ2@qaW`HHAzKCfl4cGX3$d47D02_11)XBkzHi zV189Hg<<4c{Ns0su>Feb?EnrEySIx=H&+quPY!f5(Jk7~)Qg0BDEv|TS`Y)g)q;Y^hsu!TWxw%;)Qpj$3r}@J4Ld(HU^;#@IB<>Kark*~!1i;!6uOg>MjGb+`>y?4@0tp0 z(ZQV}r%d#63f7Ax$wE>mqB6MCPr?Z_A%UCCLuVQco}(=fMbBBK5zZFf{4sT%kD~I2 zzolkV96JiHQ?VI(HrsDwW$oFI2@hiZ2xTB(LPDku(p$WFl)gNlKglk(J+}GfCgizB z?ff`s+e+^${t++COEX6J%}Q}C*A5MSjXE9a-)yJ|xQtbmT2ZP7rQD?*SG4&tC!!YM+D=aynbLXP^A zfP-zU%FoYkBivo?=a6wP2j!IK#!#N@poczR>14C*{NO34KJHedX$8CBPSWmU zwEc4_BkE#F=n&4a+*hq~hLY{O!JcwR-&e;k`Y)p#$XYNqDbCIxu+x1ws9$Ek)8X=3 zZ|DWrtEC@$%RQN4*&aFk?|gkp%6v*r!_8|{Wu+03u#rA|hMx4lFQ94-)#65ra| zG69Myfzj=53?noIXvPu@YZDL=nSgGzCp6{zaCg5sGgf2JsUioJ^e-@5ZAVYxw^}d>d^X@%a*KKL*$-U??D`KH4P$59 zjY5R8NGqpnH5mpv&_nzQ;K4}_&sQj_?g4n%|VU%+5}AE45(}e)h9@yk&1ISKIT-P;OM}u?<2xWWw~_q+V3t_1!>_1BcQBua-ULK@h&?Nl^?>97qVwJ;5_8Y#+;4m^WGh z=$|TJI(9Kh;~j^XNR(%TXfPmx8`b!Oesu_#x20|e$OR{;OBPgLx<>pTT)==rd*@Hi zb~}}84N&bvfO7uVq?co5|L#lY2ts*WUxzXDZHE*KLv@oTE!wb{x$_e<<6!2g1@iHE zgHS`I61pZW?hOMLRa0<(U=>-X-VarIW{B{blwjAHSv?utXPd_)`07&EA7Y!?DZ07sYs}YES$y=(*DOY1^ADA+oRkkNVQ761k{D7R%n}KJs6D{l#ivs@yjX>J< zN%TCoxMDIc+9UI6X}LqjdFO=^okd1G2i@%KEUkfj1NFx+kp8w`k5YQE)kc8+-}3w1 zspIkRf_VDUz~xfxU~QlWbm@4{+mEqbNKXxRo+ zwv_p|z@?;ze5gmhy>TlrKBz5(bl7HUTXU!f&1+v$oy@e-!%Sj*2-pyFnLIVWhSBwO zG;l212FW=vTsvZ$hb)j1b>Fh4$IkmoKQVsXUU`@9NsY#_{l_Tn<(378I7_v&dHV{9 zXR$A&H_gW|Xe6*>7ZTkoFUw~%Q;DRX#s4`%>`YA=wk`WUM{qhZ%%eSab4&PIED+E$ zUms6>Ee`@iJ#d^YM&eQQpN*sT=Pkv<9%161A+_8K#AhA^2TKR%8PIhmCK9kdn34-Q zg$0GjFF9MyR7M|XE*D%zQa|1)lhbT)e^yQpU5}Gk zqDoF3`1`iZL++fQ*K&{RDSYQ`K=q*73LHPG`C}(}F)zQo4AS^X!sB>^&MM=9WdVI_1c;iNf zwX*yjXLUjt`-QNMK^;2c8~a+fHG$yk-?AjWloY&qC`wqEiaxj{uzQR9v!*i|5-aL|Z0qQjzeflV4VAaPmrx zlR7@>%af3s@f<^eYy0tpaDWUK4Dy%L0!>k)=D`m1fWP<24eW_LS7!$DI$!@ok%(YM zBCuTl)H~F7d?(@LY`2$IvlSlV6hpeFhBMIT`OV%U?KL6`*`T4trS}Lb*6%bK*xVhQ zK1K{wQbV)5qn0aw3v%`dmX>CV$!7yTziq^br4##Rj4CrZP0Kr-<+@Bq-n{Ex-F4bd z+eRVe756yKh_CqiL?`gDpZsxksBzkDd9cFMC1!P*iUi+1Z-72fTKnX9;Lc8ulw0R; ziYG!UNUUo+w-h~&TL1Nl=(>#FGk=F(19xh}(SC*DM~NCSf@LcIT*1jiO&NwkWF972 zL`5_hy{_I+gA1vpwNpp8Crt3f(V5ZXAGT-Q0`CNFb-h(ic+oeA&O8NfJFfxW`q1BnD)lA=Fw)?&uX~uBE+)0t zHM9Q^$j4I5Dl!Z_h2+b}bOj-zJDv0ro33_+OubJbV)%w_jKmEejB%~52;fh|Yag@k z7@!_oa>+!+zdwLHCS!Reu_L2c0=1Gx=hct+q&kQ<$uky*{>4T?4|CE#yZ19ruj z$QCYXDzAizomMyL>f?SQgr>u8ri{SZO0U*}WHJxTAdH*?de z)6b(csU$9hTvlyjGp&$fd=^*L9sI|RP|P8f?bR?~=^d>ag-Oh0XA97pF^6;(88t)) z#;WDjeL}%FycVmwvbWX4rEooNw3tW*w-^#T;fn)mf$v2d%V!};9^Nhris9j%#PXSk zLvEPbF*-ZS{GmtW5Vi;J5e1}+!{wXm`d(ZTC-QWziFDh@nn)`pqdLw;yBWAM$70pG z+niP|_7H2-P8b}B5XMqsVhRV}vTJrD2jpAcw&+WA+CQ{PmxWO%)VhdH2XnPW*rIrW z78jC{_ddzEjKnVK68FDVuQxgUtqtTsqjXs}IRPp)_XHpR>wpuGGpF!cVihT+ntc55 z--!IWwbRqwKQDBqGu|{UPNZm|6F9mP!g}0ID)ExGQ81gc{h?>VD}aet5pW%5S=vDe z9`vpF>eRn_!Yoce*{8~6h|AOVGli{mLVfU}_qE4*U?%5$iX*cFMKek6zK=iGiRRKP zQ%}2g5&hyBPQwv{@%EjeS}H9T3<8P=*I!ArhRLc%lVT6{+8oW@eHx_LTUDN=y&0i&crCrfw4mREFHoEgADaG>MqPrcqQ9D``?9XzH#cz^kKz#y- zvGFZ1hE@8*)`xvxxMhV%c;BvWgVPHSCWByvCJn?+7n=oE!>YCMlL~#?$2z_An`ZHO zC5o0>F6+CS?aZG_@#-?AC48Jwx9!(-mA=jOYT-2C76AV5fgRAMB^nG15e2si!tUPQ z*V;<#uTnnTtH3xA;&rw)`z43~y$Hm~%8azI%{qlYa%BUVk)afX>#l!M2S_*7ZF-pz7WX-Y?%^8JD z##z8d#4ypd6_@SD{{KVOTZTpTzF)j_cO%^?(j7wxsGuOyAl)t9GjxYYgOt);(p{1w z(hW*CN+@xj(eLk^|9PX=bt_ zu9=u!#E=?Ule=+vz)R?pgTqWQft(vDjE9NjqNbL<#H7gLgbS1Z2#7WYX*?+N)WF0d zuNPwxPR;~yFJ?A2Gq{V(!u;!pCr|$e^K@gx<-s5pKv1SQG@Wr9E8#gaaDDAEjfgGD zm#g|}75mjz)8N-b9+&nGyoea2jd1bVEbEr7w9UpDAth>3tSQ=z+t`AE`;5&&;ztd_ zjd{6EG5d>1MuSG9`z9Yd8@oh#HuI!!S}7zPB6C#l+zL7~@gBx}WoIgw3z>Y7n26R9@ zmw+hj2Tp+g)Xp0VsvD+})JKEhCWd*gc%y$T59j#X3i_~4e(r+KcNKF%*DawmGCc#D zYIce=QD*SZ1?zl1P7s(c}c+(1a4dCTL8Bp-#RkBu5Mx-$BQfPEh4)sIYz zQ2=fqe8M|8{su{rR>UT#j>^%&U2d|-W?Y<+1m`G(sDTg-DM&z>@`H^XLZO&`Sixye zq)aU=LG9$%iM!C?%@Fqz-$mUr<2zsTqu4_{>SZvM&}#35&F6O4lSuB_&XFFi;+N~A zm+K57y-`@ZYU`K15R;Bm(U+U;Y3=JuB;Yk9CaK#9QSwvQR zQD@8Ft`FR#q9#0!)X7*l1_&|j3oP}6=1rwg_bocslH=cO)rW}W@s&I%!ydll!urq$ z0!FKArYkPJ(DfV?#Xvj?Ui=7^R|YM?CVrPP=Yey36pu_Sn$aymq#ea4l-YaBy=^@f zx$k@&fNWJ)>3}U#eqAT>!*n<>S83maCl%P&xH!cmCG&`#qI2&=*PU0KUjsqrcum;4A24 z&l`OEuWN2*jzbGG+R+;pOlA3O|19eI%?ck==AZ#W=ml1yrC1%m4Wi^mQe_Z`Ryc-K zgbd~()R-5fe$=&7bm9J#jM0bd($n>>1aihqZxT|*%^otidET*H-o=yl$MW@Jxf(A4 zZM<$l{Wpha1kfNMOY%mcs~&{U(QJ23npIr>1xZ&e~HC}#0pup;4 zAWa}~$Fxd{6R&o>c)|PxSSQh)7q*aFzTyAUyEeMe(DnP*_(eBPsKcyH*fiL>E5@N% zjv`o&yD`d=eZ?B?2P;#@@KXpF%k^})j*AVl5>9D}f?ASK-bh8ch>#U4o<`!LdSr-q zYEm^vng)3Eke*9lo7)L+rxO0-!YQP7^|`WU;kP1C_%x=!N9+4YQ^KD1M;KHxxyBz| z>uPFt$F^MP)Qse7PzYo&l+O)a$AyA(F)i;K`)?HXjaA&Qdlnr=6I>LdnuE+@V>tMA z(|aoXt)%Am_8iLvn{3y!g)OQ444(fvU25M-BS~sVUi>5-`~$3SocK`|c1nF-hcS$g(YMx=P+5)BWcu|CyVjSbdQ?f) z=Z2t8JETslp;X=9ywi0rT#B*xR;`(hVONTwFe&pENGy6J%e7oES$)Of#3?A^#mv6B}K>-HRdTmJAuyP$H< zzL;)QwXSwW_1N39T2*+#AjomLbjuYv+H^vqgETEurFzTtXQ&3PS*nb&k1u#i9Rui=Li;ZJHL&Me7)lJM8 z%gK~K5k3FI@N|6YeVTB;IQb%4Ql(Stq43R~c4rS@-h@mdI*p=Yh(H|j=p3Cj@PgVHv;T3)tJ*c?UanA>Dl)u%UkAMb9`kO zUoEdbv%vyyv5SjOv}lndBfE;EQ4PEpoT>LcViGh|UzqHE{U|Iwq%?rmd7gbdr?6D{ zVethjT7h?7PxKE^7A;=M6Q;TUGj03ux&%YJi(2iUb&Ath@l)RatP<9B=RG$t{vszp z^z4$Xkx&cm`)LwerjkHuT{vw zTH|B-vKlwFB=e;(XmJ{yI{MHWNw#OS?~m1MyH6G~7Hf9~D{q(L{;Y@{?j`*BU3D5) zk~288lhWzrcL@9QyFv^`Z{wI#fQf=x#_&5wcJ0sbz(mOLWsSKJTJL4##Jr&zN?H#7 z!Q6U~)JSUU9?ZgmM_)*MArg5oqtbrw6?r|SOvgD&5TLhCo+vT}r#~s&G*y#&H-eF> z{y`cj=-^=98H*4^^;L`Sbvd`?q(x-p-=Clkztv zJp*<`#81;jTuA@xT=t(E$~=I5Qyc#_@I-JJ~;Yd zt3F5Szi&mffppw9&ptez$7RHeV=xQ<{%vL9Q28p}s{>(js1oOUp&*Z*(Sh%c?ptZ~ z_88@3WVE|Mk4A@7@}n&CvFG~m-b^DmccA0!g%B7{{Ia9N(S4jiO4nLvWW2>zr^$iy z(}Bs_svDxJ7)Fdzd*z$5I}93vhhuIQ%hFHDLE<2>$2(==Sq;gdf#K({X|!7`A~1D( z{7g%7zS)(nlw$IKo4-nCenMsHeZ5ymkY_Ikn)j&?q8unH8-sez4PW{VlhkF;v#B7` zF!*4-dq7HpFh5`ElEEm+YR{HdfjE4bf%r7Nj=V5f^;NC@;M&`&QsYqxml$FU zJZaL1=dL4vZT^PE)r1sm>Q|5P zc?&Y?V5dfiXgNoHeg?0N&`F$}Nl`!DPshxR-X(im&RenxbL%0$v&=M0zm^klIE)^{ zcn;8*Lro!hO-_Zo!7Y?%IXZmK131PT3g;~YaFoXRmRMpgzgQ?i&-x;v(pRg&E*lo& z|EP^EF+-yC9to6LX?&Pw-&DM+cuuq)P^@kdgpLg~-J>0NW3eyxH^@cMLj0wi*AS|G zgsR=Sx+E23V43;N>IDWYvMp^VWZIro6ubxBSG-2#sQxi5K!!wn<|WA!Uv(0Np}_c0 z$3E%vz}C?D!ij4OOlGu6GY(S` z(Z>SBj?;EOJ}1zr?YDTXm2NgmB2l*9Vtekltbb&PG5IGKa?(#67*qYcl7ARNgsfUm z$iwA2)le9X6Ef2yA0{fByP8%Z?p0k1T<8kCuzU#S|48SCSca}qGtCMYwb?w+Rr}S= z-Jo{kYC1)ktd(MAV0FTv@Tofr_U8i8_X6S1WouLr3F24Fw4mlhCPS4JImu2_qZfR< z##w1qPQTr>HLKhC%(e`__FFCpdP-t3hhivx0{0jArWLr>tUn8)N#M1k+PBlq-994c z&5MgKGYd!P9Y-GsKpYaCjIkfJHa{t`ynALzVwp0SF^OyPxVzTDvi^d(`j){;)4n3y z=8+X1S+=dBqb`KHB1$NVP>wn~uf|4{#h)`k^7N&>^d@bLD(|g2n}!DR!2&A0kzDv5 zq{hM$8~lurkY^|8+Vp1#@Nv&$}a{?pqv-r36R1-~31JYoud4P*zbIU3ETA>wT+=^KR{)Hu(79$|ig zZ4u`<;|Z#(uk;A_+{OV$_8?HTN(Iy0-7gFPxnVp)W@RMXZ7d7J z7QC(|AZt}#5dX3TgR#qKX#SsT!H4S(@o%te_z3yIkPEF8I~KjnMAAYP9AVm37aw*V z(?!9NBBpRFP0vAQHb+_~v|N>=cPg1b7D~l1s_$l*AE8l-;0Zp})4~PaX1^_Eoy8jw zn%CBRqnttIm4h~{ZUP=7jCoSH9pl&MhM9t&`xEr_)e{zLY}dYj?uuj+&!b;<2*C(d z00ja3a3BpyzaRT(CaRG(Dp-a%lR24sy{d&nefPxpTQX$91!CT|yve%+FreQ61`!2R zK)io4evHoK@4y$ZUZWTfy%`f66(5xdKv7&(xYz~v!Oskb*qNYZN&m*s>7X7O_s}hs z@SrYX)zSRn@INpP_}xBaLM)JL>0GzWGTNqXO5>6CtX(z(HpgnoyJIN9)r*`jvumx5J(|_!YA_S2$r69wL1M!mxc>pLJt1Bbe<}^c=~FuVo{1H@uyPPzKHbh<2w@isOf&yd%%a`8@O%0itY z{oVICu)CFe{{YwUs({pPd<_iD&V8wDp1;a2`Fra~WkgW0ZBcblP~7THNh6yF^gN8_ zXL=OZh2cPD8wLA%2wB%VZz3+PLd+ktX8jy9`gOZGX2fJ0IruqW0kyzcBpsTa7`P#% zFt89lIUvteAkS2JP9x*f3NGt$s2Lub4$6lk5wkW8vXdE*{{4;rel0o{0v8^l#Uk|W z7{o?ctpmMy7E~>zP0(3wJ&q;tu31n)E$Cy43QI*A2KT9z29;qS2Hql>S%MnsNw35 z{QIBHr%^~FBt`WvkhW0NHF|S_mA;MO65NIx(Og(o<-sUxX+c)$B1^6(rofV}zVOO; zXtzt*H6}4NefP2Dyt#nhv-1m3eZEJmdvSS;K2zyV)(c+#EkTGyX%yu3pNE_qNzb>M z5%2`SJNJKANC7{ulBjqXb=b4j%l(>XJ3xM&e%aoyDDx`sz1?`|a|8;flu$;Sw#Be< z@y{Bg9>G~nywcba&(V9#+HgY&vQ-sAo-GtoB)I5m5qwZ~es}%!Lb02AzqCdUPHYY> z%mi4kwfkF{wX>0I(S#A{>n9=;Ka6e|ox4#hgXk|c424|;Go&7Ka&lG!*d%-q(fDb_ zr%Zkkj2vPuMpD+{1Fj4qJCNTM@NP2mb8ghXlSnF=m;UXqQFO#e-SG!$i0>kDeGW|$ zc_@f16|G5d!(2b490aj|Z4@#&02#|Pt+FvL=iCW6^|XV_$ML2m|XwzJC&@r zo}6cXuTWQrSMoswB?e9!dQ1}qg+FP+FB@_jSJ%GvOrUSdB1KJCk-71!%h_&{ zpQPVhmTV-*pUM+;9wIRQ%4{|D%w&14X8bU8rriy?d-z2wt;Q+}!D`%%tgp_p*_)>b zwT8eUZ+=Ms$&*OZ6R>3p?15Wv!)bkr>`5PB|}DG3gS zZU1?f@|pN%UFu&&Javbqhzc-9j|dh}PQ7JCpghlJK}HR(XD0(il$^hm+AEZU72NXI zxPB)vaA2fv_XSfIva*Q5*7_oO*adur!Si3nU2IZLkYDv>+C{ASe^?7~YAlqp^Xv-V zhhYGs1Kv|!)C5WEDE`4OTp2rE?l+savJQ#9>mzd!;29lIdk&=gD;a#>SY7vrq%$i# zf=(lY3>N8I<*XFp-tuB4L=z(zDG_?QW8^fUQoKivHEX#5?L!598!cv?*RSK;_h;6D ztEOSBihVBe5Do{)bn1wNJJaEEBYI^p7J=W`JL|lQZ#{aF(X%y?4?~0Vgk0>=5=7Bv zf?)9^Xani$RApF-SIKAqFkuDpT}MWT9@rcI@i1PMhBn(bG*tS4TwWfkM#fT!Y`~jcmo~NM41-^h3>Bq+?=VemA0_?2EsUA*O@@n|76JG2 zRw0nLaRyG7JYQ2vG~TeLH%YU5VrE$Dt5EKlP(YBrMl*OY*yMQtsZ+6N57~KYZ}Q^2 z;rRiKyBdhp42dber)-Is&gidf~DPht;W00?=GGoOst%KP>NYQ4Blct z=|yh@d(}wDktIBlGDUi0G}61af>b&n&PN7oA;ejss1@ArDmX%RM!=ADqifuQ@*3De zZHeDevT^6Gw&oSJ#;`m7TWBcY4-8aAL(i5WlRmkERK*X`88@Iih|hthls;*}>}y*# zj>&puL+Hf;Q;0Qj%%qlftHRM%u65i@Tg(qoE)~rif%GkQxL_w34Ob3uCTeSI#p`1q z2n`EA2YHGQ02-`jLd3KWLxQHlsp@Y5JCWBN)3Z0qK8mJ&foLINBEZW*@?DS%Yvx_r zp@GFtFYgJFKv%t+D=d9uD|rQotMk;(yN!vXmt2*f+F?Z@+N^i&!R)hWja!5kB#R}8 z<$hY|JgMd~P}C^_doYs?-n>O+D~wZHRW>6%y(naKygAND_%=u4xXsK6`Cd`{eYsI5 zt(cgYa9(rs>>4(OH+P3)M}U?h9LscaHu7k3F9Ogu5W!5&JT#t&^q;n7^6?cZip@iV z;ig{<0tD;=M+avSN&U=1I9BTW7dJbOyIF^fN?BL+!71Kp3$CfjmZ=FXDXC^2L5C?#KQNu zogDbsd4ST^LnV#3Tc68^@y79xztb8A0(c&#AQOz@=>-2|6$dA056E%>QETBfW`hgI ze4&f^_dQe=<)z)6Z6JiFu}ayZ-F~uaJy!LCwERvIxV>Yj#cEw0=B@kHOcafN_=4(0 z0EvtD;c8ayM~K@~X!Z#P8ybn>b8qkF_QJVV&nnfnbV+aijg8+a?T_aK$h|ejrr>kh zr{890%>d*=3x$x1$}UhQ4TUu$9I*(O%6LVDjm-RFw+jZbEF{KZW&U|*|0+szrd-#o zN=)p6_WzrEvYRge7Ma*8x+z##DnEaIF%uPM3X)-(?tbr|f+_YoI3R@`0j8JsUK+5l z;ncGi($^n8eb^U09~C%lUA@Ii(rIZxHgQ)FFXCNk2Ki)vM82+v_%oX20j!EnlXIT` z9h{C5vehO1)+MV8E4=MO56)z!da3rrg<&&Oa{Oeymk)4o{BAbs21nBc#+;(A)pyOW zE|)(BrwF?EN&b4=#^RXYTy-< z&U3{GM|$3liG239LNz}lH=)~$%XFmf3`SRzwthd#SwzxsXv>?rAS79VyC^EO56@AF zYO^$}IT2`+XZpv|$wJ;z7df{nTs|dB*bsfBctM??TphJWcAk8$caLJ-2oJb->ZG)KSb7{lPsGIn>y!pyt|X8{3< zP#ioyB-P-}lcI*`M;&vGKo0e7oG$R@^y2#8pviO6_^!%z1EPnieSrSd6QSQyP^)%Y z6aRx+0xx!5$1tkFo__Hqm=rEYI6YH$Y~e(Er}L0k)FSSb9xZ(U2+OjJSrK%>0b?ac!GT zf5@(8nbv%!cAE>N+n>Gzl3)tA{^*~wl@Yvj9@dRdRTc|{SW&EOy_gv^=sT#qVRm8LmN@v}T?vBN-nNc8FAaEw z5;%bKOU6=Hy;uYojTUv9i|aD)4g6M~SA;PAZMOBobW=&up&PM~JATlMh#0zK6mxiN zbe~w$J62;?GRFx`^}>`b3@{mdSaGN)-YDpCJeHdawGCo|ql|i$=MK75q>r4s{K5YNmpxx_;ixK}@VL zlgoAAUzEhcJx~v4Dn0%8yEz@FkKh^!SbNHiEC8-%`7`jA{#Eiq*E4CnT0ie9o)k-$ zGI1Jfdn_}&Zd@E2WJrXO{^Dc(+-K9cW%@2TIe8{yXW4VLVH(Ls1DM3&IXV8fZWhh( zlrasg(75{Z0N}ZD_$aqv{GVV_9C4t*ZtRzE8Rm>8f{8lB4U0eoC4$amL7CnX8ts71 z;3KHAXDOLVio!u4>D=YlZnx{TX4+t{azQ3nPm{d=@BXt4(L{Vjjq+3&8Fd79L+aS# z_XucyXX-p!DbpXbHbI{*xD<}PnNL0K9o^w3RC^yzkdH2VN*4+xL3v35D9@E|joJrBwi^){w+_^1s&L+H>b*Buu!w>ky$b-UEp~=q^mnwosH>|JQ7%~G*8u*G_w+G3%A`4NW()VwZ+tT%s_=&ytwS;FXaCaMF?>=Tu+sGUg^DkzHb ze4?)PGS`J!Uw9ln;%6hxKiskV_o$3wl;N^fCYWa7!)AzEJrMV2C17hALOlEL5I;s?k=y4-BnZR1 zLAg!ssr|Sa%D%xtef@QCkn81v{zRif6t2yRh;}8d;UKZcTp-mr7{r2caBzSFpW!)h z=;6>wQ^PH0H80nfnq3cDKzNjn`WoPJ9R`r#&>o8Yx9V_UW*nHuVtau^G$=Mp6@^MX zh*T4tixnjI6L06H2P?>ODAxZ^7C40<0$GD=-CZG=U83itMaF6vlOF$@LP7k6)bw#) zkt1kcZ`ELo>xmBQ@-wF=rjE5_;V@f@nHYbdOb+E}Vhq_4dzXrgj*mnof@C)LCr1?cG5m5 z&e3HT*Om-`v#W*V>O`Jif{LBpx%)#Jg_XRl3usHgLV>5ckNXw)q_>$5&gIwj5kx$D zq5E81Tw}*i5uZMLb|}e(S#f3=UTwdHe$zHvR`@1>Yr&M zVryp@BtqO#q=GmQCuDSt;uP-je?MZrOxIJGd-73D-Z(eL0}MVWQPP$ItekoXg5^M| zBg@M$f{AP<0FBTg*HLc--9@H}WcWtsTOuna6&L!&i^K7^LGy(;tcgF}jDK3QVVU#V z^kJmBKd@^N{5KR*3TEu8t>r&&)d(sr<@7a!UG&#;ZJDvVFmREuvarlY1-rYuvvYFF zd3$>oEx$7slE)(Uf5jl{h6>xJEOK9oJfc5Yn(}6E$#fPN`gRTe?mDl0ksTi6p#1}h zMIZXr0pbw(iIL#^Qy3rSdp>(qe;=goTgL6sll7A{J_9{$4dx*wR(}opA*>*?fhJ+| z2+%zcIU@x{wjTkrToip`OVNY&_v$C=dL?LGOu7)wnAL$z! z9#+`Ukx={|8_X&6N(4HUSMI%Cx6|L^K3=1|w=y`I__K&|R8TlGITpU1Ei<8%9d2v5 zjI=Hk+!+%uPB-0Bd;PN{AhBw;{`bPuZ<=Je9p(}b&QkIj0#Wxy@bfz=2kf_RRWF>d z9c2eK>@@6fl@=d08GUb)Ih3KT2xa`{VK|w=^sUi$OyoRnUQ25Rm?8IT-0j-E?IDgl zsp_NKrcFt#LiAi^h?tm47ntsCDGyHRkSCPDJ=HL7-+i^M?@46&T4zd|)rSQ<%73FH z!A^ez2+;dMj7uD*pcT0kS-`Ew;X>%I{nTA-q; z(SAPY;J_Y^Hy9g%G7-Au2R7^c(wmXF@^&6d8-X1gq_b^ z;P$+49?Arn#{Dn6HG+`G0iqU;ri_S{;Z*=KdU5{YqceNct<2piejA7;YFz!{ zX@9!W-}L?MvvTL^hA;VPgW_H%xejed1M_1kfY%wo^WrNwRia=)5*O(J;~VKk{8V{q#3RO;<9-Z9d4oQI8ba&A^j5?Cb0Iy^K~ec6+)GM#eH_iew) z;X7#) zWZsqH+1NBt$aLTlvwkz_f}#n8;nh7^|EAOMiEHzE=)3p;0+5&-O3jXrs!&PH9@p=1 z9~`HV_v(U1YTE7eHLY223nS!W8=Y);__?A#x^5>Shn^a3|uy2%h>iZc^Q${CSi z@>7`eB#&MnDFoSru_O}-8lK3qz9s8uLjy798_G_+uRTwg*)zZ|o~2JTo+xZ!l*RhV zPILoe!yFz5fi$Y@TS@P88Im?j0ED}b14GJ@i`E;O4p0*=a1(uZx`q!d;aPp~mjB{- z)#=Ys*IlNp)!oJR_wRGV7;rUOWy>C?N|xv(j}>vgz*=)ZR*K(uu_q>E6G{v%MDGJ* za`;v%&(+*SV7H7DkKZSW(>>OnY7NZcFwFfHPu2iIu_cV<@0&eVGF^VIF#QABJ-JvO zr<4(`fYB6VxFj7Oz?9}}9m z3p2fGb$r-3vu)}m&u~UH+%6R7VnRtlKo-B*6TZGNwAuDC_BzD-t}(*aYjkzc0MHhC zW`90iW-S-qc0aKj#(CE5JT(xG-)F_N?foos+fMVoZzjyZY23x5&)iR7@0G;u%Votj zjZb5an=&s9bl!Q#BD*N|g=n>h!`d`DX`E&pg{oSMC(qB)dJ}?Nw%l`Q`Dmwl1%Qa) z>hZ4;uL9}COxHd^h4lpyC*jR)omZNn4B{<-iA^G!3Nm`0;!x`Z-|R3N82(aDnpn#i zBF?-W%?O^2mZTGNoB_8Lr(Yud@&4U9eqiw?$HwbZD$Dl+gA%gzs6hpah-L!_W`p82^pArQzjlN7k|-L* z=ttv(SP>#GGnfb|`%{Lzv+CPS{G~7vnkFY-Cx|nhO!2|$(3)8>8OEiC#{L1Ba`7)) z+wA2V*9kVS_6$93ksP)CZ{J!rDO4dQ=Rh@gRO(MWt0e;~vehY9C+vp;@cg2<5&SRrhD*YG0DB`(C)YI;aNhJjgfq!zm!3s#dNRMck(^a z0BqD$ItXTf?m5;`xvptE`IRIouFyqu^U>h+gov1B@xWzhP+yK#r+H{VcaH*X(#Y%b zMW~bZ{UU8@6~hTdkz(v+qqLl+4zBdD4;3b+8w|4-ifBS1%so2biH?gU@8?IcA~HcE zxlzb|`nEVdHg0u+Or_Z2+c)t#ceYaIxo|a>zh^@!GkcIai!5ObCuWq|flOMoklk!S zODVMmo2AJsOU@oUZ0K3=a}*to@3eN$R0M)$1R!-AojT_dNyduWB^t{GI|9XTbLEp! z;=2x(-m5*SuEZPqoQIl5jP>%%phgmE{nx4XSUo6{mT+A5?Ww>vhafi(TBD zn>8rn(G?EQ>2)9Bt0!tG1y6$vgE=uT?DaH27Q4)X?wP>N=NAE{4^2CMetZN%n7Ft` z9LcG7ez+WW-UqaSSV65N5zA2uz+Mw;oP1r@P(5Ku+6!FP^G{**`x~b>Yy6wf`~ntm zWv;&`0r0@K^F}1w2D;1@;?{V!U4Kj)4?(`b$sx|j@UXq6>H_5%I7?=iMPwB2VYlZ; z%dHO6+5@v>Y>_XT8NR00#fwjP{E{`zF1dhO^m zUgOn8?f0q?TcY$$8_UD<{_be!^N8roYZ4eamQ0VxCz+Vqt*94*y}8Bal0v&^F;SmS zQV_6bmKMIdXA|%ti)bww57e7%J*ZrXso5X(Y9r(0E5})YmI<#$Ye&yC$4WhakscL3 z5ZIq-Ms@U4XTap<7YHaGtq(n$v#_R zE`}l_xCIr0@;Pv4TI`n>+>1-N2R_uu^S*J$>EXIWVrdASMjq|NeCuj3Vh@FG#-zZf#g#_^WOuQ8ROeI=SRIAfH_*)AbIbXk&crcbbF zVeDYOf61oyMahLB6e@|9TKCeJT?#ec;ZCLX{ou1FVs9iTNg`1m^Acft;1Yow(R%e+ z;u>%yI)e9?fh(@%SjMz6v+|XH-tu4cqusqP7*ua?8C^#;{8Z(Wf#6)B^`uQ0dM<~u zJWHWxV(O(m|2g$yuqPB-!XoQM;wT(KZFIY*ohB355sN#X>FMUn31Anm^?Bnob#$z& zxaR!$V)}IR%huWn--na*^QJsi7p`(~r;R#-PcmH-x6Sq6A_K^Y&+W0RADl~fBV>1llYicf3@AD87Y?>J4>=1^JU&m*rMDg~{^>Nbn{r0KI`bOS)d-F{1hgL9d)sejic%K9R zcPN$_*psMg%wE7qGS{aufMz4c-D(1VS7A&_;<{TGbI6*j;=SGTWGeCC${(3wjM07d zQ36=;avSGjMkxg&QPT2~1<5FQ2kST6v4WUj?5?1xZ6?SkW`BrUTL;^Ij|@`hXYX#T z;PEi;y^y?aT-KZdk?(%8t$B7L1D8L@Sz==?FLvJOlAKDeFv90H!sFIQvtQ_g=3k=7 zR@>57geqlJsv^O4Q-|GE!%A1c2sYE+);-Zz@!Q8nIX6MrcHRB$X;J77r`O8$b{06BmIXbfu_lvIE&|&tjm!;N z#mP-R>O=Dz;A+vlnR#rbB1Uc#=AM^DOp+EvJ!W6a=VaUcCkiEg_XD@IE#q#1&LOaQ zM2L{x-~dk$S^Q`_;h%AzFIY&yd0xB(SPbb|<7DhR=hqv>0{B%{`W+`1z4@YV4LuNm z6?uP*X8qmI5*^dm_@6GRjObF)&kJJjt&`Le)owfi7Dvl?=wdT6b-|)r4SHFWUFG8% z+iWtHBael0Eqw!wBnSJqv!^2SHXVcA_74~17eC!CW+GQUs41TGl^@>YJk|N(N+Bgq zt*g7Ptop-e_Wja*K+-ndgUqlOcv9V&c2wQlIRQnFwZ9UNi)5evortza{(kU+*$X4` zS8IYnYV6LNQma~6sh@TWmn2%mQzTP#Bupj)#UIf5!Z9SU$91vHB`XdGkx0kIJpq=);sK$!$HE2yq z{sh%5gFcMGCii|pghG56hU=P)$pR-@Pd?C*4IK;D>gk|)L?<>eP-#9Ht9>h%j{FtGLo0?2C#f+`so^F;@>U|m?>iIu~$n7?FgL; zy%7s_hE;EvUm=CG)B0>qZce#$3@(OjD6DKp|AV`V8Pc=XG(yDs4$XwO-;@8)vPy*O4Bkf%l4C>0?_ z7C_%|Y`F*$d_=igQSf|K{UMu|YZP#EAr_N`npywprLV@|cxJOIYPdGueVPbq zhAaBSufjM;jiFtxT6|EuS|C+_P#))9=tgKTm)s1`&F*(Da16(&<|0(@?yu;y;V(8e zhd4{W&*ER!cMVzFiqCtD3!|)xuj0Wv?iQgt26j9G zBe5Z?#Rdhv18agyzIWDUd6-$6=KLI|Dh7i%5!s7;`UFxGUA~iaD`}6*8d;bASUhPN!V$lWsl}^S zD)yfeZaWEvonR~9`koY#1B1DyrY23=Up!9c&PxHTTtTey)k!aYjn6Mz`<1qN$4-E_ z;3Dt?hMNk4M5Jy^VpijyJCj=GbGjm*Um?oE1q}i4S*%2H*&9-n5f3ELqI{)+IB*cq z?!R5#jgfbuC*CkamyJAWk3QNV*Ohwor6)CZ`UuJ7W0%aKUw`Ac<>N_VK%nSvAPvuv zd4+9(OH0a3W{`uPMnGEpy=T=8kW)(~$%6F&gL zA;1!(vAdk{swUol6bn@;_JPIY(4{S~R%5z5373IK_esDx!l*%jBhn8RXYdVgUtizc z_P@tL4y8-f>zX##9$hUN+fIVou`)aT=6%|a4PZxv4S6}CGw9)?dd`-&pu>rY!2cXw zZN*c2`0Z0*#G}fV#m1wxPxBMB0e5P#`@!?n%xU|0TZM7piNH(h^*lxsws*$da*H)t zN}Wa_XBDmxz4AHAP)x>#!$3Z~UH6vHIARO2mTGBd{97hy)4Olpdh+p!QC9J8bG}$Jr4M#+5$)V{ZN=87hVo7B5Pj^&Q8cJ!hwdzSob;@XwOA&GG-^6KoamLh$kd_;-)4{= zKkgdO?#i#$Xnz>L#vL9%ge0$vYb5&9%jqe`B<+8Tm>r%6qY)9hZ>Lcqp_B~C8HlM4 zi-u+w%v@N)!^v34jAm-&=`q9AIIlwR6&*%->3N7Fo_Zj@({?w|_<5)U`+Ri318t38k8a5VS-7voq7MxKRDJ&E5{H6W z;NET8$&YrvyWZXec;FsTdb{kE6fHP_z;}NnE<}*qYYYSq9#ltx!yH9(dqLP_+|^U? z{iM`DmY6Fqh&65f8-M{0Y81lq2Wji3OEvNa&nSLyhlwpeRewt$z1@B`CuH!mp>PW#XD@levOGtB+U9yC*|E%9et&oTN&ry} zz^%z*hfffTG3X*%cu}8#wZFV(GWJh3>LKG7N@_;OAy5+3f%GHQWZ*23b0cxt3b(jA zS=T~=8xl8O1pX#I0xh$?>dD2T(;z$u_b~jpWuIpWXj>Xx=0Q|^ICx+;L7q&4C$o~J zWg#i%YyV%>2C- zeVZbt?tvCsGr^!-ClTh%>})}4DHcE+4}JdJ4OWabcC$S8s~t3e)H7FO7H;?&2;deL z7A*UrCa-s~Xv+W~cOmg$sdqYD&G`A0@m=q${` zn|YoV4-2V#Yb;J6L#;gYMia#hy%j5{C56o4`#kapAWo_sG+xKFx&Vk7MJ8{b{(^h; z8x!78ZpO*pCwS`puCBY%PNf_%q^mW*NpbXj2*)d$F=k_%Fk~_U2m^eVriX@{k-8I^ z5VuzOQjbarqchq>#@}2;=8 zAein3Aztzf1HK9f3JPM~=vec{!ZGqa6v}kD>7~0PSiKq$v9~-`6jc@E`(SNgPp2G}qAw!6|QO`g} zV?6$rimHJt#ockb*VwX1*ngQTs*$1f04g$@)+V|3P;VS1(IC^33w=D}he8k`p`#dT zW)a#vy2Qx3RrQkO3;KG>BRp6Ufq!fO3lR{kR3{^eIIYCsOJ5;4wVk4+aO(yE9suTq z)Ft&D2*6#d__R{H(fRf3OMP$Ngf0Z*fB@;4N<+KLK)7h`_kNXrH$9z^|1ZrOkWkjJ z51_?1TqI{+@DI7;Bq-`ak4#5`k&@m`_%I||IIVF!#$!}4byicxnCJ403g?jDwlBqj$qdPNnW}%UfH?#nel8saoEG z8iKelMAx5hk9}QLy@$&S=pR+n9D0jqyuf)+^HR_I*RYJ22Ht?P4cE48H@PR z3W1UhVYl$3$^}1b(u4RkASpnX)sMLHqURZRv!W4g`CcQc5FNQYxB1^I8~pSLo{%H4 zm`L*SH+T|$+?dX@n2oe#j3{{rKKC`-5A|K*NH3B4J-|x`z>u#dx-Euy1hx$joVF_J zrmxv&?6Bp{;BeU!9bnz=lMdSeO&h&{2UNr}?w!6@_8JMd$5W+92R+*Iqti4|MeVv5_|1(&ho=zsQ;XRE2B~ zsm#4TtKrcYCPQ3kfM+X67Y|Ao0uU@X@Zm`k0U@DDe{f=<=Y~gK zxxBURC{|Fp=#Xtlxp-Q7dSoR zFNZlWIuLJZ@im!+sfRl-iX%ngX=td+g2K=Z zL!*+Cl0yw$5&{xKcUW|zfV4D-q|%MFw3M`TKbz~m?)(3|5MSs=;mkSv?7e<#eOJh{ zw-rO~Zig#s&|V4=6E~!yhoPR+Z{*``(Q`xX$=>=q%$wZ@5mz_TvmiEaC6d7tPL5s0 zHDCHbgsyIv{;sc96i}3A|M)pst7GAs{=J9(ScXfvzwsxyJ30RaIYj~IhMLK>fL{?6 z505^T@a=u5SPIHiD>>BSJewHJK3H#xmnaC7{;2Y9pMLQ!Z2y|bY{856*6>CYr!!5@ zO_IY7nI~O%u*)}nwaOf~`SXbAGYj%@qCD13{Z|>Jq@;Y{MWLE4J+DV72^u9zHC;i8 zdfi_}MZZ#zcGd{fkK4grA;&@Fbs zDIEEscE&f?9ViEbBkJu%?vVo4=!X$gQC`C7dO}feBF;x>_V&#m=#fH{YO<*$r?H9d z%`yL3ITU#F3Vvdv;!3}aI-aROMW31870CawC@xrXu}-A& z`8-e{0roDxu^6Ge+T|8=2(%?%ceP~)`ysB>z+w67)*v@*`EL(Qx*u+7@483$Y9w?; z_T-*#56}{m?mo5+j|tCOQVWA?b$Ogn$axUmJY@s+i+tsTE?_u41ZNmLVBUILOUD4} z;#O8x3~Q)QA98b7e?Sv)oJIhg$UmFPUXJWo(04LpIZ~H04gw-xDc$9zr8VH7?+!NW z`O?a-w@UwcpjpsLF%0x~9mBSZjrFXUmbIWdI7F5hF>j*O>3aHt+~7pr#X+T1qOfwG zj9s*lFCl}%ZVN+Y%B^>Dk$X8gNYk~r(=GE->l@YfU+b$f&kZ*O$92=p-;-mHy)^Lp zPq)lJm-1T}2Bk>F0P<5|%S(1hu zXc@t=jFw(ABjn$g}d7~XkN`Z>3R*EMS- zf6R!AE1b8q{H7ctr%s8W_v>39f=>oW(JWYCdzP#r5K07gAu$i$F(+m zT0|Ue5Qr`8o9Zhj%C)GDm&N62`>y6{l`LyIp1Ur6e21{x;n8j;&y&ZN=7?WwCh`wi zI)ZK}Y!IEdaP`Xn=MEm_Fvt7GPsnNNGAo_~%!Vc&Y+gfXzDuX$U@EPtN@1K*^*D4+> zx39KTGkkX`2pH1i!_%c735+8n<1zDoU5|t2e@=65cfP0&OOZd^ zO$;q_Amrp!Qljjy_V5okL2io`Hw}(nF7pQc_=22#ODfNNRQMmcYVFLK+3r}sjzK0N zI8h=$s)`IYm8)c;BZeTr4Mo`NcU~v9 zB;ikeED}G4&5TVtl-AcI7XQ9nFX_mz#w;5A5~``-Z~Z-yF;n40-$O60nB4FZ#mB{V zzgsof2Mha{Mo!AgF0i`0)L7KMb!U|ATq$&1^)pCZdQXnNpW#(&vf;Af(r<~TSB|?7 z9%s>#bdaF&g;_B$=qlV-s=ZR2p@hm=Y&A>i>hkJHh+$VKQL`em-;m{}bQ!to^GBP3 z7$=mHVPgUVM{(Xbgh?{NGT(p2&lMWS^%rU=R$cVHLQBSzm`Nv*^FTAlkG{Px`fS*1 z+CI2G8eRpzy6;PI9!I@k9;tw+Ae#bn%>540$hi?cUD&DW%I~DatX8hQr>BQq)|3JC zZ0vOc%Bu@E{=Lx$U|zFo0c^B&QqJ5$T|M3jSlcFBQm!x=LV^{_uYkor95fP6%+lX2 zVail~Ab6;WM{pUl7qNlsLh>6RiHEbKPK{T^x^X@j7Y>&Mb_}=F)R29BwO^M|BV?oa zJfPcC{aC|Wy^_&}+^FvRcZE6q1Fa1OWs3mq-C@xPr;~yS=H}e4M`fM~C*)9*haL7f zh(39&l0M|=KZ7ogIv9Uem2%hvB<)v~^O+@wjA0j(oszcSE*8@hB86k}e(~er?AlD% z4UgW%9h8s>xCKqbU=q;TR^WS>g8$a^fac824&}NbDFp7rX=P14g@H^KdTC$+3Dp&VGnsE7ujsY-l6^p?*HdNhK`{1itu>leQS4-0P!H(1ko1&u3_3~I1? z`ap%ZiLTs{4c&q%9qsi_>T#FDAg~#}&J+V8ute6xzVcn@Mrig=?-9SoRJ2|PHF1{B z3{kK1+rO8wt2g(7z4dgLu43JigI$+E+Qo~ZS36ga6_h)qZ|mlqJ1YWCxCgRj7y)n& z6HKwU%gFQGVLQG6sOS|d4#W#u)QwtDql~>DR|@Q1UE24>b0Cbu&{EF|STYvOJCr`A zpq&eYltaD7@+EWATb>*L_31AyI3U!?bt*>5(6nC}%(S}t<`AkXPL#a$JqSC0fM1V@ z>{}4f-2fi0FyQdr>CjVm*NiG4y3)S7-}HfvURr}-^upln&5#?rFa;GKiXIac6)Ah{ zVbX`x^4zpp+6~pt>~8OI#huSmh=gx9{yPOI9vf10b=b43*C=y#ES>p;FCC;7ejirgYnHxhf~8 z{cDw56zkO+6k;qb5*B-oQ%}ft!B9oI+bqQ#3N8FkkIy&FRs8p zy%7;;lGBnG*tv**N-8PhEq&0u>9Fy=yMO->Sd&^0cdH{eW6;bTFcO&4RxTezqm~r2 zXgwzhX>Rr=Bm16HBZ#uLwj6;f=IM#QF8+z3w^x@+un53JbXA@f3%6=`dZkC>yxKe5 zh%U^%+suBuLos7u(C^}BM}|>F!i@T~`O$hn040|=#$5(1dpSYo_-eG(+^pR_cq)B| z)!TH_uzGdrb#EOAor0J|VC5HlPlkn$5jY=+yEyP>x@c9GPCv3&moVH_3DfcdWtRH$?3mcEsa2PTak7 zk3qXO40P?DTKI#kliqnZ7i|D61KHhlT3hoRXkc^2_rhP>eO;79oTiAb_wo-$N%-$0 zcIkRcYnj`F+BJ4OZe^|aJCF(>N zSoGw*CuHYXicjY#Ofy3&X#|2lKVBuU&5@E*K%vQduF3c&b~chsC+t%mNmP*oBwH)w zs+mSA!D*;&yteJc$>nw*bJ?ez!d{~Gb^;|(=5j^s@T_U=G{Yr|yVEA(SADB+YsJAZ zHi^ANT_vRW@Q>CDYIqlaDHTj-{D^b@7zDwIZX^5RtE$ zKJ)+PkZW3EcxAK%iZK{3Z!|5pw_6$KC$JbVviJ{=fhJVLm-$XZdZz20hqi);U@X8S z4N25}DLw`$QtN;Sfp4{P9t5SfL(=gHs2F&ax`Q`Vg+wOkE~kGbtz>%K6F2g}tlQ*V zQQ6ycOpJ?_RX$I5bj@Q1I3dD(s!*X@UQ=ABDc07N&f;26pudQi_;6KdaK96xvkXv0 zwm%qkXz(1jT+liCCAn2ap%{GaUoeMl^7y&7wo_}-$F~cp^wxj>B33Lo(5xCl%vRfY z!erdZFfX4hh7R$eqYt^k=fz<^a9|T$up8f{!|=eB&(q3>@U`cqj}~Lc46EIUqTYP* zcAfO3BAWk#!e#GZ@IJ;YfvCRY*tYf$>W!+HXbC9Z0a*Ux(z96qT-^81Hm z$?-Q3u{*zaN$%bh{kguJYIWuaby$@G}O(MU%TXENZq!K2@ zyLp{1<`^pTu_oyXjBLX0DRdaft!Ll%UyQQ;2)Rso&2YWLdAYpNHJqINaNVYxcy|o zMYZUtdQ`mC=ju(;sT4&T-Bh4yM>piC;spgfK*Qnb0O5_PfQ6XtZfK|dGa^&7C?JD2 zw@V7^q~aFPR15VR|3DeFvyBvPz2T#_jif<@_bB{~;K&hv)V@RP5Z~{Bxqw%vL4k5~ zefps0&Skx^9U1Qmhrrfb{lGVL>ha~^e)hpx&{QOy0YocI);5dMf6&?cXP=8Vuui#8 znk?wtb+u0ZHV9tDGJSTC;X0LeE4-s&%(}enoR`Bi5c{g*%j;L6c3SJcLLJ9K=98CG z#%U>T_>!Y$Qg7e>8T@M~km#Az`R2{_?#p4oIF6@M;FmGy9`Bd5sX~25(q?JrN!vFd6xAe(f zhtIgg(M*SrXq-TV5|*|1XQISf1A@09~-hY8^+)0Zdoz#}w!be89+{N%&tB(D!T_1Jm_nBfNjf8S--43AR0VLFs4XiMNe zxDZu4#6lgqI}5yv0G>O{`o(~+^nT}?`Fk%81XkmWZjlZnA}Ket3)O2ao|lhrOxzu` zj?>k+9Vj6`48g*4)RHZC&kO803VxwyRPU_lANc%8P= zI$5{l{^-D`dE@sL*|qPteY)ZpBE4o}&iVE4!9rp1E8lAAF+m$eJk3Lkbym_*qIbyO z(-v>1ne{j$RVbI1mN}(%9IqF#76vxj(l#);$rI?rp0l(E;0@#A1ln4IC{BfhkEyHc z9M~Wp1O|=)Q~f1w)^Wk5h$IIm3w&j)KrNk!mew**1_PhQaq5d4sO5&>NEKWLe zvj6Y8KP}>v0ehh$#phS;wEE3jZNR>Gd9#^)&R}~9YYuq+`AZyu1$5m@Vc#} zqUq)#o#cT2KJt*GDckv{_s`tA5A~UHNo!t%SNGmWj!3k*0-fz{S7@)QuQ1Lv=iZ8% zPc)_&Jr2fCYDb%+GnaiRBFI`;x_5-rJ6evf(6sEK*0|8wruvG8>NF&lkovDnmws-$om?iKdxnm z2hS4X7WrBVGwh1^SW~zJ=8+V-(io@kLLbD-P9s){gb&z4@x`f^x2lP=_=$Lfdu-hBXieb8 z7Z%pe#3YLF(o%?O8d*5CKnKL%mx^80b@Xa(x96aJ!sU1%-T5ZA~JMgh*F*l3NWSfkv%{iu_r$Y78? zuBgcK*>775fB50lJOyC6*q+Z|~f8ID$V%{xxRZ zwV%xqk@qE};vs9Nby$`C(&lIz0HxrKqsI%Gum~u9W7aCkk(YJ~tFAImdYI6@`d9QR z&7LpH6^?pI1(DRMVf8-gc?$+E*1s1&PoIPs^E&|si4tm5!H)zFax~KJ2AtsKVxC6| zE+M%cZ%H^#^O;=FoGE1rMiRbQ3#&)qJmlYNcFrteI-fKu$6-Em>X@HGp_HB|68hyT z!4(+5Iw>to?$Wx;MLbqzko$~ToFWPhzg<^fl+i#d$Tf?l{pU{|Bj@)=k4BpTl!Y7R zUVeckSi9R{0xeQL9Pf)Belz&n_-XO!GC`I#NZ}U zO7f+p3mbD3cnhJtoFPP+yTjQD?2W>ZNV~MpX=11mFXRWv#YgQVh`5MRG6)=vLM2J(3Q2u@v{kDO1!1@6HEB^_bB_)U5%_=&+m1 z0yhUR%>!hqpqXFoFOD2@xwtghQr^5VM-M?qN57xkkfVgpCT-u>t%(myazDDK;g5}t zHP_N>EGZBNVsAs;A-}LmzIP9LS6NcSB2f88JpQ;^zXMQ@F}aX3?z4nm9^!~3NHr$^jC zHt_?h9bdD1At~11ZPg&AcRd^$Uhx7M1V#0pDK}U5Jwt~!>}B@5QDn;{AHqUi;Y z9G1Huv^8rz?J5r6HTUksj4=OOr-1D1s(@CZ709`E#bnL>Byr zO`YltRyu*pFB7QbKf$Oa2kRgyE&+H~b72W9?L(m!ZR=bVe?7|{@!vdQY_FheuvD(G z>z!^C7_CupAZ>r7ZYXp9A_snM*~PyQj|@Tjo`aWe@%6t`4-mJ->N?{lLnWO4yZ8I z>f7xRyFWDGHhAfWP7Lq5wKMUU#lr(Y;l)zIX?ECOD|S-9^;jnFdj_{%e*l0``1gZ&Ql96rJ)S~8)<@QBtDoo8ia>@tk(68?$ll5V z`d&Kc5!G%QQJ%K9_u2M>jxFcX_M>exP)#w>mUP|=%;ldK@IyrK@*HQz5csqfm3G=( zRn^*+fq`t%GeT)RkS>pyQi7gGni!sPQU3)(f*(LZCKN@sYpUN%qMJb+U5l;f_jTHH z7glBkKubcaPp=Sh3V*fx5_|J~4J2IBHjxyoDjImWDzVqv&&jb~JgLm*y89AO85ZFu z73;|z@K~XH1l;tcL@MozAQpNEsQoK0fmgH)0q5qDz#w<;HkpTS3}t!)OX4B8rnCt% zx1#rhWQkVuP7EWHM3&aA@xs;nvJG*Ixa8CrCGTFEgXcp|`W^(Kn4h6JDq z1(4moEpuIeNayx3ZLfZAc6nGZh`AEpv|7XdEfGZl=bGw0%u~+j17uF{=*k4Hghhs4 zHD+GxlVunWJx?Re)OB4kI7;I{K4cBr6x0{0jDlq!6==0_^0g9WD!J=kqcjHWhSU|q zlsi`EYj?$5%JCSoV<7#OntBv*X=$zs$`ZKCUkV)PLEyhgh&5pJwLCRKBTA=4M$a^9qk0^-H>H4pNshNK|`s1k+X7S zd@k^%ym)*7JTrBF7r&h?#2C4JBd`DU5_j+GeoH4ydzn%F3h>==0E;8TKTvK)VDy*J z7&QwBXxTc?gZTWl^4qaw>=fI`u>28RDt-P_TVwg2L&F<*@eIIf?kX+h^@y5*ym*j0 z?ANt-&tW*0=FEsevcO2wYMX&89FQM*QxG|@*8v2OV%Ra6$zUriKz8XeIH-HW|*%w!}V#RO#j3)mZK`iAH> ztOU2Zc0d29{P`p63@kixbd_vewm9@W%!bw{SfguJKL~+Z8xbsxgDydWgJ#64={UH( zZCA;~6av8{KuQ-u?N4fteT7I+ef5MJ0|QQ_85M00wXg`G&KHCgZ_#Low%1Y%WT+{) zBQyNZ<$*P-0eh6)d;ViP^M~tU|Tg`o`}|k0IU<-Wb=Gp&2Frf_T-hX zD%^{fr!u3~JHwH?={LBr3JSu@nKtFMn9(D$d~71xHgF}&o9OPBEzngn=%S#2W?^ah zCL_bq$ijjJgMcm)05r_k2U1hSJ-L+PXu@Pc%z21sLXSXIMFm1{UbE#Le=Pap#fub) zS3K%sYh?FTPGI{a`}Aq_d;s{k>w=e>eU3^a6DPk@Q&TTw-LC_@p%Z5-0zt8)2-(84 zZa`b%gMOrbsFRGW=wR|d7Uz7Lz(=m7%749YK2~sL+yuS79XAuk>sd@~ouY-F>nWSr z*k|(zs>fV2+10Lp1=m@@Mdu#i<0fpM!3G6Q6qw&rR=WvbbwK9^9G(q`^JhcGoaD7E zL_a|A?6H`e(kc?fhSTk-4GAi^W`Qw|mJD;uzCp+sVwEK_Apv8GKxyxggJo~g)H}Jy z&14T;mKLv^7Y@a~<{G5>)EDdihi}#+%2JimE?iQNt_M$ZR;h;UH}OAD&_7@Nsk$~E z5&++L3Fg?h?pOF09a8kZn?Ffq2A*B(`u>FRw~pi96J zt&!M2jzX~Z85zkQ0K<_~0BF`&={LKrs>Ne~zB@ruO3ykR5E%ZR+;aTD=tJTMw8nkj zfJ0nR0lL@$9P+Wit8Ji_cmJN}$YJ8Ol!lzfv>p6d@1BSyq9R;Ul;OtOyXB}nk!Cvc z7wF@x|7sAljA(Suev>SQoK09w0Kj=BkjHpbenni6ZroDZAuZ^8TJH|F zWM4YS0UIdPG&M_0EgW>+VFE#^S9E|b-TBm?@}CGvYJ4&KwBeOSOF5`z{?@pFEb#Cl zkrh!y*yyV?XHa4F*_eSEn=Qrp+2BS#B2u=#h0`swf|Li)r`J`RG|N&1?W`Zw0l%?}^};^gQ4GD9@+!zCx36~7PQ zfZD`YQ>&{Udx_$Z5wkzgN*H$wmX1b4O(@~{cm1*>AFvN3+sh2N+-d-rA-8sP%B6d9 z!M%2h9_aTm@}T)>sPkeYJK?@ir<*ZQvMZK4AHdo8 zU|IV9u5ZO6FX*i?^32F99qUQ4_@BK=nDCKhktGD^8i8L&-3hL7?bv!asx&t7Qlo-L z#-%iXV8>1VoQc!HoY|KwFw2v^!|;t0FmOK3TDS`mnUZb@bclaL^8u00VLEB)o+EY* zn}F6quh)cP*V3FnfUY_MD`(O$L}$2Aie5;reE*;&x>9_pyl)e3@dz@=YJk=DpD)MS zn+Mw3e;yp{AIv%f%Rp&~`$}6wE)PaFgr@E-YygRgrE913^`hJ{{2Iq!OKZJ^et#zv zlMo}E5K}3w;u0_my9Q3mg1V_hiG)v8bIq08Hq?dZ0pDf_U-{6HSsa6W{hp4fF!yBw z*#`MVHb%I~W)=Sr8 z4H#Ia@$kcJ{&!`tMgu=TM2nt82Gw(b;`$>Xr|x^HEmEE~B9s^$|6py7$sc*U8JI8e zN=K9r=}R!;1so`*o0dh(1YsF9EF(^35F_S1^B|#|T zI8%*yiskvcu4=@Z`pAMm4{hVHMKI%s1&76`RWko%0mr^2zhqDjk3;C@V?WCQpGn_y z+Zx49$^nzPtj^-R-38+#&6@Ln{6A8tILLO?#^z1(i>Em;osthz=~h?d}V*>_%Db{)*#A&*u82EM*$!g z%aOwZse{@FYzM|`i8eDdeN?`NTIhQkF&cUpGq%88x|X8ocKWkQUImx%j%JBM zllfwozb@6;KjL4GyVM|qLM1~~r||JS$y|3r#AJJ8V_Th5Td{6i1S1LvwK+_s?RjLA z5@K{B;i0l?7dA2Bp>jm6YgjJ+hwzGx=j=6TEs;NNMdg<0NUBjmwhin z5$Uiq2tcW41)1zhN%6QU*%T^1l@Ix?mB@9!#}4zOUUyJWH`m*?Jdy5tKZi>?qalHB zP}zokGmEMir;4~c?>1>RA_oXKga5P#P?Si3@~ydz$mk z4tn7U?g-EA_)j{vl5oyhE3tlpo21ML6GT4c%O`-d6L`Les(BnqZ>5(Wf@Up@&3s7# z)^cQAfr8iWgPr!lK&+@#GeIfAdfA%)1!749oHl+gJYj$H8@E@16w7&1WiH8yZmt75dLIuR!^UJ2V1za zhD_SFY4WiVVxl1ZBU2(uJ78SPM<7_n!;+G}BE>JtwZo!N*#CF05c$7{wT>iV#XwOz zoT|Tg(Yl<${G`p%h_ZompoRqAEbWw({5^X)DMlJSQ{Q?1n2P7}Ao={dZ*i6PtLf3C zhqPktnt|3g+}6&yhTD<+^V6<~xg9#zS&Vnx&z=r!v~)hm<2Bo}82;+j=1~rEAD{z&vsB~voes<&xwT*6he*aC1;;jG4AaCR( z`pD_Jh}%|&8Fa&oqT9GX6!66Lo3qSXP?xJ*jG?6b73s;#s`PdKy1MUS|G)JyKbN4$dfgUT)2&T%UDRl4 zazrx$BAC);`99SjcTjHF<^7Y(C#1%b($1C7&NWh3-d?UMBfW`&v;J7WB2g{Q`t}>+ zSJk0bdaPOyQ&@2IKo!Qq-Il`U>E93dXR9UX2}?qbKepFs53lHlEsy)n>M9gHB!Tri zsR_-aD|4#c^UH?kc1DPr`;$$2WO|n|RZ0OT8rKD%Hbu(&a*3_2$1yi6t0ogrLp(FT zFC|K%9UjVjd`-HX3d*eiSv~1aCU{_)CQCaQoSWRe!ubs+05>O)=S@0&XLfX6_X4g& zzSdZ7>Z6JQIj{nakVofy?zMrEkWI3ww*$I<4bg?|FeA`ds8 zPw>Qv$mi~3Cke2CM@O)pW5niRyn>VX!3 zDSbq}6f2O?2Aa?T`#h3B`9HNADO$2ZnkGBN=t+A>(UTw*v6OT~;OdH9d1nnfYekr7 zAYD8Qntfil8WV~1mIFDT2eT`s#5)>Yn*%j98>wo5+&?_E{k8~LnqmR!`bZ#79BG_(!|Os3-wY6wg1sYO|Sn)JH?!CpvW z?^Y?DN;{KfZ|P4;$_syNo8)Q$sh~Ps&Ou?P^eL|fy{a#VhZ2l3?ZT5HmIB?KJG0Q?B-{Ek2B5& zV&dK-FTLS?w&1qO6+NJ}96DKV!Y~8Aq-FfV-l%l~>So+MOxEvC4U_4=JL4MkrdyqjMun?9Pw1gu z4ot}PKxnCryAq`%+}FjDaIbR>(@gn1H)6PHx*fncs4@h)A9P+lIRv3dAPB51tHu{bY8|>Ti&Ori)kWS0>weZj zp12s=6LFz--Fyu81eOcWZP1PFCv{VZ(VQZ-2YP~6-!5e{8o4C_Q8HxHAVMm zt^-9dFb7ubQgxnuQOIaB9fo~@I(od!0{Ro6hO@L=HC3ha7+{T7`&{kTyUkc9zt94$ z9`AOmfN373Fw-mTVj-Qk<-c2mQ7)Q{rHYn5)ih;2Og7ca#Vbb8oUwuQTOXZ`3{Q{F zmG6%J4*Gqlm8Io3D%&zr9_`Z+P5rI--Ye8cY;7AvHz-XFbaC`+5sp9F<3q{OBueCR zZB;*A_-Zq$#4OjYKVAfISFV?CHeE?}^UosEQUy#b#1&u9Bcq3qJ$z0i>c#?b&UhBT z;Q(su>F+yKKm>w;YrCV_{A;>1N(k^6m>wX?*uTRTRox{p(7;dlBJxMNssJm79#-iz zuFhH`PT;Tj;I>U2rXk~6j`JPt}tGgrCuEDivOJAIrp;aKO)4ozI*~qeK!G0Wl zd^V^bB{vS$=-ZZQW4tc%3o(cQDHY#}$_W(i@tjWOt9dl$SFUPptkHmb>^FLF?RvkA z%Yx)1lO{4knd?fEdaGU%kb}i?p#Jo~AcrC9>NX#0(9c!lssIlbO?}+=d+31L{&K-f zPImkH9V(fN;L$!x!~)viWad(DRF+~w7XDI#WskcjnK<7QT=@b=D?}g6`<<38UDgsB zet{{s{Xvv{W>Engik@^5r!cqmDt-RxZE$`dnK4#vu8Kk6}X*Pp@1 z?A%y88b3{uQxsklQ^p$g^w6fffBYi#I-6z-gi^A1F_LZw6aMOszNp6d~ z6t9|3?hKOh9<7U0U|993C8sBEpzxKF^jJ&dAs_&!idX%BIwh4KJw>Qrm*8U=%0^@S zDPfg1QQny*tmS|VYY(9Q?(5L+LwzmX*ylSoGBK29e#G~n*Lvs7GESBD|DC5s`OSHd zv-wW=vM~YE{O#V&*Gms1@dKL5wS=~KWrVy764=U)2MQ#Bofzgo-J2qZNT`zn*B?wC zw5kEVlmnOf0IGNT2cKKpLbp;M=6`Ekw~MLVxBXSS+o(5#WdrtK9kG*9REpY9pJseJ z2*+>hifh}O=TTB)UI3}ZP@z8DZWePbuTKSdkAcdTV-^;HP75O4Ak#A$NZR!KXMBi)G6-ca_HXapNRmMMkQfE?$NL%g<(#p@!Ogws7LM(Uw zpH_uqk3SB;3mZ>G#$3FP)bX?)%)k)WgoAB&CgTh46sYJWnf|VzWu|Vak4Lq0GbwHtfyZ2gati!34#FB$#0EgoK7t`43>8A5 z-6^};^8sY+&r6G*zmW5Ih&KY@Aw`eWF#_SdjqQ4rJDf&Qr+?%4HfA1;#ed8%!Fo=m z!o0LT*cMIRR%PAR_n{3CD?tAE^=SiC4MK)AW|1=4QRA(m@v067pznGMWp3=6umARY zrIDj0*w~FQ!H2WrrpyBA@6$t>t>91GfJ{1P_g@8OrHCyaS;+)B_68+Uc<#sS7EH?s zkE{%pAjQq@_IXZ0j}s%t&#)i4f__vutX+9rzHA*90BtK;Z+a|20avXiM>j%U?d(Q5 zb5SFgZONK!VRu7gFJ~10&4pRDJH5P49TU%IA`fSW1T;TWwaN2+8eN}HBQH7F>g^p`*rn?~832DQ40>I@e%ijyPe_8uZbK^E4~V zf-atFD$2(}G>id$r+B;Spr3aC^c~ws0WL4|)9ep3#GDiUb+bRfU}brV_y`0zx3R=aap?AiJ{G4k1e?JyE= z)2klqGh}#XicIjUvIqL!VcvIC2Y{YiUmzcd?j~@U{uzv_Thb@C@=IHRcIBpyFgUcV z0Dx*=kYFQ>T(&Fwe&hst_vGvEc`1=uN+utuf4T_hpFML5Ypd+gjVzyz{eCGGSuz)K z(Dw~aRyk~Zv4NzE4psX?iYELs-R#s>WXVR4sZbvKqJuT_2P7=G<`FEBdCZZQIGe~n zZ4%Aa&sJF#*CHC@D=PdAe`hW0l3R;dozLOWXwD~B^ITH&a?mPnxTX-4F5GBc%Gq;b zT|f&D(lNxS7H2AaNMg05gf&`!zIbyBhG(CebAV;w_}f@c;R{w5hE%5rpM z1ZG%hGDArZLL;z+rWFig$!HVYc38B)DSiO#P`2f zhE){`-3LJ{T7z488Z0;iRF;BN5CUwUM#qaPPyn=DY=4$QuaZP-EuC0D5Q*>P0;}So z2Qo6jt-n@IEt2{xS;TZMj{%Xj5)N}%Ae^in=l%d-^bkP6r)(V_NWtUynQAsO89igr z53`L@=Y0j^C*TbYRMV}CFL1~8&N~TBPM~`DRt4;ep*6(trLcBtxE%q_?O|uo&2=3}Gg3#a{Hv7VXJq}{KF#u*+S*D%nO z&Ux-oGX9UA+4qm0X{A(P_=ZX2#i$o2$w%$MNa8l4aq^F!#HP})NsVcxf0ZS_jw8ww z`zZyajm^#)-OTn~3C&S0vsJEZT}pA>rN2_p&(v%m9Qhv4>7rOsz#uvwex?RVoLPtG zWIyPcq99(yxCG9t&^ORAZyVwa-;CF)sa8pG=quqqLBVh|1l|v z;eyT_3|-EU$gK5L*5%1mBDoA!Hmy$@(CiIlD<6|%&)iI=lz;t*HvM%rhWa+~Evt)$D-prYU@2xwZwq%6l)+V{g<^%8iNlOX!LtB8Pg zjmVT2)#j7ew*u$xl!io$psCmkGp^xY_ShJ#CAbaM(QU*D0{Hs<7F+xL75`B75PaX&9Wi79cProd&&a~-~>KrN?+|%-#1ds zY@PlcbO|601XSMLW6%a=7Z&U=nc|@6meSYv)EM+ML|WMC{Z|{VnEv?LH>d9BybY(I zrLrEm)j-tpU7Hf|a?3`Ll!h2zE;*IG>VtUKA%|{C=&YflGa_RyB7uIs-z~U0 zJ12S!VY4qm^YP4?=eVioRn@GS$%lGbrWT^n9c^tqo%D@QRFxL3)ZW@e4TFMuxPoQx z#|yp$1LG83O0sW)ZsYnnvqJ0uPHZm{NtN-ZwvX6F@q=QbK z)@i%r14T@|QMjxt1H68*67l@tW-LS0O`17H)8|1zXBvyvP-FGD-;#W)0kpjyQLvh6l`lomN+C!IDn+?x80ORJc`0n12u>tb22?x~EUN*OJkr6xPxWgMXN;U& z$DfdL>EaYx*M_UhskXa6I!|ng>VRfs6KYQKw^BIFZ+LvxbU^ki?E0oF9l9lDeo_4E4fP zp5)**Y60U3sIdC@By(7p5qT(ap2l+>9jhPquTOW{BZ*l{p%5{#8Kc)b89wW&-Zv*x zd>||T8w9Ch1%}#K6C_v4c3X|z2Zg*(U)bpC>?OPXdkBqe!LihA)^F=TJCQvT=sQEf zG($aYMKg*;91sNR;6PlFA*avUA~gVf&;(JuKo@Hv%Z>M2FaZFx=a!n-E(K#V1Dt=e zxRg6IJPIf;12L$AEx99(8`r2i~Y_$lVf#HQ_BOdF4v zM$*-PCm+SXs6#0lIKlWK`%_mm*K7gP*=SEbh_Ubm_6U3qy^;Kh3O z2h1B~J^a#XOdt$d9-MrsV&uM2EKE%@FCZOkm%An#XAwl{sl;MDS^7U3`= z(kjUWN5VxVCmku*>d56Hh2@?_?b_Uanb*DaOL-uExyHLXsYb7@TOxKYK9|>yTlWy2 z^dN-uXDS(`gv2u4(YyAgr%s4%iPfSO%|u;fOnj8^)bbR$w$eu&B*(Pen9oI3jLhNo zILw#qrA!T=L+X21*9u@tbQ}1_zJheqzsexT;Q>=6Fxy|+4>*&Mm}k5I#XS+24Bl<& z`5&%~0f*{8OVuUOskD5fek>wh!D#*)&NftM>sQRn1@}GW3+`Y3ku=qkEF>xnn0ThP zvn))j3b2vxfnz)9tRoZG&^+$)H&k(!4~{6H+Lz1)GW^QpnZc+Gc$rLezgJkGc+RqV z_(Os6h$z?mx@o{q3RFebrS2hE+JWm|a^BfMtun7;4lt_XkssoMch)h!V^U^gaQQlOc5cwitCM_X0CHoqNaR4y}q24JYZVr6!*ouDMa&qOs={^(*NN8^yM0 zRCu8K2lZSg^pHC7oa8N=evQ=ZnMeZ;oLE{Mws_(E++OIVBy6b{I`$pTh5Y;ziMiLA zE(boALy^d{7temxz`i1bw;D&y@5iDWF5OnLssQ~}RQ71v&>^Z$PHQzxUgKchFnkW( zf>0VM>2ZEo(N+uhw>Xr@C}c1pUmtKINBumPs}v^ONljf|1(-1V z?i0HD%b$`>tHw*2L<4XAB<88D;Be3kqey9D1rmyJE=$TXmv1unH11#R=WD!yc7^`3 z?7QKeI_3WsvyM(x%C*Tum&KXRKychWUeQB!MJm2yq7AecOWb0Rxp;2JABT{-{0OK##S%pcuoAq&FuWIkkB~53AuqL+s3WNSTIG+QrNI1;zV|nM zzdf5JwqfRb)Rj(}3plP`yC}krNYa-**yz;taO0~bS=gC#4qND9Dv?|b`F|cPMBV6Ctv0-5ijYfVbX(i0Cd{q5CZ~!U-Nf4=c;GE(^hs+ksA9BP8eM_BFsOlE? z0fp6&4^k{LY@L^E(mNJ72Ak}NKaT(13gBvPcE85a^(jX}Mh_}w>EQ~p{wn#3WuGCOoTwZl zdNR%WK_xKLN2JS+VYlUIBZ4@5&ZwH@&{5Y@tQp-uplgE^z}6lEMhWR(QNMD{FfxVK zu&7l5z_oGDL@gnI4%!N|^*4Sw8$5O`yoeq%OyP6rs>#nod}O0cSodgnF9n7nP?IQd z{E|je^oKKLND@=|r1mF0%lcjQr>JlhmiFMHdS66;CP9pQ`GyzS=oZWIu7uWV>iJH$ zWomX;gQ`HJ@Ktvz@~ei%zjY#TROg{8RhhD2Oa1}jkoi+GBz6rx)WaEPB5Btk2kN82 zs?ToFxfgb8qc5%P(MZ!zUGX zo0yG`Lqa&u&Q{=n9U&;?@VIFcQ~06d>8|6Zo2cESv?UT6L-;R{?9%UgIRb>n5%X$` zU7J3F-ejXT517Sej9dnv-2jf3$=|m>^~v-Up@s~oHvJc&zE@%&$o>TJ9pqr9Paegz zVqP8WP00HBeGI=qw8C9}Yi74WTpgZKBc@)<%=eY5wEt?tRU-ucX`I%a`-~n4`2zQR~!F>E3u1<%i)WEhY$3TwGkmT{5B~M`*yM129{q~ z=gnR>d?Qy-UKrA=CTz+6#Q$Xq)XEIM^Y;`B_YB~=>p?j6iOTYcID91>ox-4V_e0YPMWKNa_yTz|;y>z>t>mAp*PRsCbFkc4W*>fq)x6O#Z}>KZ z7&-I(Yw-7z7MDj}^8Bnn|B$K*?fpJ64{g zR=$yU#-=)g#*??p z0R)$%_E<#3*AWijFFlvtF6!{DOp5EK-LALCyweLiBufrTVond|$CLz87VP&M!mz`w z^ZaH&$IsgW7>GZc9x$7yY*rdKA zHoSr}nj@-&Q%BWw4f2zMb25j^?lop@uu0)j_ocXSc+9v>heW=*|4;v6JY{*NO2KfQ@+D9$jmm@Hlo?@=5UF@}~x%IYOzG zo?x-Mm3MjkU1eK1bn+RC1$Mg6=FH>fXXiAub#=`Fy><>{n-XlP<_P_|j4KH_w76-- z7t@5vNJt0*qV4jP7QL|WVs2LGF6-0r%2*@;i=F$>|C@H?J_x8cL;x(=bkwkT`px(O zawXxKPp>sFs`B$E?)=93nBoHY7DTu7 zGat;9_W)j1LP|=QViE&Qxa7P)65a+g0?1AjT(espyYBnGR8>R0iVFq~t>#7hxWpY+ zDh6(>DoV>hPVhKci81X7Hyui5{t%b;tYME7%JAI`@wKo!6<)o?#BI?-EmMLTmAnbb z1+)~wB4$tFT52?=M&Q#%V7tk>n* zFE-4-^}t6N%NG2pQ+uHJQb8V;A|w^xB-?#P{|4Nh$K^+Dr=q12@u#BKh{2(X@$^DM z&FU!svpVvCvQo*kh-koDwMJ*nrlL#tWQ6ol$>W+s)7pDX&OBSRiUXS-AMrJr88<^D znca+aof5T8H6otp_zw}y55zgNLHRZ~kiW09=KOd($r&P}=%#;h&8%-Mv*^NsIJTSj zB<~8--A4|r{zx=v@F;T|zefeXIw$C#3CtUbywWAmP;0|3hy~>4fvpY~c-zzs&=yEL zruR3F+-uf4)XwvbAL`nAU}}Rjf_F&6b?|B+2QVYZ!?%P?8&8aUG8HPRm^m!pe6FW{ z9IlvceHK$N7nh3Wm~s60kr9UQvQXlbGeYI*@h>g^!{Ke`0$n#Q(yM4n!taQXdpGtf9q)>5m!$h#e}b^@HC zh5eNQsOMV*yr~|U06p!DPS42LFE;cRO7Degz6U1T;)DVN*3}`uth^)eP^K_GdUW7F z^kMd=j)WLdM@A@bG9n~S zrPo2J*EoiXl)*GERU2)rMJ!;B#A%!8+NEJCOe13r76U<<=#bxK_)ueJ!u--}0uSEd za)AMg&7Zfr<@-4T)9jmO%R1OW^m%Sq=~6d~We+SeMgM*4s_+$E)JB}#w&H6>On;tG zB+9MtQh1iL??`{(8}Q$7WBV4_unClhIOsm4@Q`W;Wv9c>Z7u2nI4y&Qra(o!N4I)B zG%=rw#(XaHW==ZSv|Ciu>v6E5*USl@1lQH@<5d`lQG_6*u^Ct$buj6Ck~LGegf1@> zu9b@YVHOSmaTaU;1n6VcdqC9w0mADYOQH1MI}(Kd8nM6|bd8I|GE&RewOvsyWpao; z{$TSwUIg_6AUBr6;#8NTXgm7}hv;9;9WCizvapkT!x37MEA}~HEj z;T}dXfG}Xv1t+|mgo=gLSosH&*uhwLhWB zXa6eU(7pQ86~u3Fa>dQ^kubId*&>t*!eiek1%9!i#hrTmmaz*8%fDkR?AOxM3OC>v zwMr_hG`?F~pR={T*wruS-!Bh)g&?cNglU0$7f%Bswf7-YpuW`=bXpNHIn<-(I{Od^ zs%{~)f3;j4l|e=&VM8L7Y(Uv>T_98L3Y$_X?_~v{|NeZ6LK<4LB+7FCiV}2_phD^G zgm4?bT9XOF;N6LwyVsWe@g3Iv&?N0mn!rLG7Z`!oiVR{j`>~jI&uW^*Dps3LBDsz% zAG9R!W8SRHJOY}~(_7Hv#Q>uCfZ7w77g4ddo=OQX8#IB;b0@IAwzv-!SK1#z zMyBwpme*E-{G+(fF07tx3LrsAK@?GF6dViQ@8<{8{{H%nRR@3g>)wLT)8b3h_rKoA z%&ttn>0N30oX@0?w&H_^OWnr!bh0FjXJoXZY$WB~o@Z2|UecCfFeXX`o$7J>|M{i6 zxgJ3!7oWFq;uqaQ_gUek2EHuxC;2rp1>7=|Jdez2c(=F^LS{oWyM)U718U-5s2}X$ zL;Ht?VzT)#ULi0)^D^7CR=7ESF4DAPud;d-l3~R|=uvkSJD@r5_I9bTTMf`qNqpwi zLjRD2W0y$82TLH;tW`-P=gXpb@yfiWDJW!8di9A`68zrcjhE!~(5lS2r+%dfgQnxSg$l147IptZ?ZLejYDa3CDO9M3MjJ8W}dC?lH;1+}C z2Z!f~QC8A{bg{6nDLHKGRIn1aDT-gRAv!dpHCr&i35y}|wbA!Wl09dx`Y9T3*6ryr zcuO|nFJdQP2K_fM*eozXo(_d@9Np5RKP;o_HDTWz3$G1x1Qjc9C(-7m4N*VqjYMc1 z*!>{Yr70T`S;&Jz>2t^hMfCG+0wO%vJ%3}_cR&i5!qm_$MD9m`uv`QD`F@YSca>oy z*@AdPhE72w;s<55r(mlj!#SXdwrgiJD+_!Sk^g>_%4zPX;~NkM<`B0m=lEjH5a8G4 z=idH$k;!5#*YQJr|J^^{Hw-e)^*<5yn+m5 zyb@COG9Px6T1L89pEgn^gd(OYN8Tc7`^Jb8Pf!6%2CfZ+u%^* zUXK_AdB|L#BxE&eol}`5Ol30&zZk1s4~K5L0gQcrbXqu>gPv^ge?I@n!f62>iGlH| zgd`Q80m4PX?g0h;kDL|3h+bnua1?N4vV!)Mb|*F*n{0s-eKYiu8cOR)q8HBj=&;l| zh@0)L1L)u2Adf?2eCh1L&M}2KU&0X&vTADr+kEWXu3{-XvUAo%4+^EFgVKq~tkREz z$TbB%ga3BUf8M~en1j=_-t4kXY)=!^D#wZ~hk%-on|DNT$Z4?5@9aJp6DBfKj|i@X zQmi0v6)!vx;8^55NoDC(8;elD$tzXd2q=CE+@UK|HnF4X56=tROVzB3tuKfWl}2kh z)zi{}H&{jB>u`wUjl&G9VWpS4^5U8BqMg{`xy{NcO6HcBO7Jj%z6!62DSr`-_p0|z zESpzvG%{^c4peoQi6|;1s;kzNixgB<*8K2YQkM%~=4W}QP5+q*BaNf|1}^Q5M@B0$ z1ghnWfVk4J#OBfOySY(R0rxm;uj`=Xx#&-80HUHOy?!Y}R{RINXvMD|bX9wIk+GF!AteE#u>mWSQsE|d`;eu{}kZ%CX+ z^<^^5t>?AlnBM$^Mm)_2lEkJO(}8u^YSc09{R#KelCe@@EetFV4{JvE8Q1X~w0nj| zr(OsP{-p<=Hooi^+hB;kaK$tlajZ2+5NEI(w=K;H|Dsh|=yQEMrdiC_owwRxzexYH zYKSNK@@SC=tg(8eKF3Je+OilFsgTgr#BFxlH}mh0AyqW3(R4DpU8(ffQuy=yvkS#l zAh+m-$6Tc9`S>fnd0wC1Ty^usl)!ZZ=THz&!^Mjga`v(A!%G z8fuRB*w9`LrW2cU{WM;iG5k(vy+m1OQzk}c<*l9o-fk!3<@fheGKnEEPFw{9wvO^X z0V~U7nLan5D`5$o4|EFcSCNBwVTRVH$2$QqiOu=$P9`L>;xf$EcXkF{Ee0WSub5}` zHAECs&}X2wyY8=XfU5|txx>*(HdtRJe+}|NOI5|96tOuc7%}DvZ4J|GF%f-!(PV3D zTf&~=^)WtN^Hl3CMed)w9O3tTIhM~W{f)m6+IUJ_cfu)p*)}XwCt(H~*tE?@n>GqGcXha=+oYOEn|024vMKeaX9{1@@&xkQk9kwPFd_cTP4cl^xSTBo*}MI z$aIKSxh^mv8UQeLgX3x}_zq`Y)q}=gSec;sttkcX0kWXQ@>Cf61YEprN;skGpk2XDp)W!Zx>E6*J7Rr$=I09_{y4Hv{4wA^eH3 zA!Jq&N}CZWXCgA(rjPj1`v?jKe#U%%mnt-(?Pf@zhsBR*wl-vngFYeZO3+58Ogd;%DWnS3cpu>-u zr^~M*o?%MZS7>nB2Xhrb9S*Iv5+spgqjBzB`V}&poIz&{`TqjqK7krw=soT z)CVs5r3U&Fil@;tJvq#)PPcoS2*G~{Jp4 z4(%aSg3~|a-|1gWBhe$?{AaR0t{_mgve91WCXFWS%^3ezA&w>r#87;s%=oVFI-Eos zO^?;3l+SwCxvO&0?>pGgTmSoGDrdCG=F zBq9qO_~hG~Z<_j@O_QnzZs;H_TdqBkxH|FJ9%G*bsxo}#n6M9N^Wq*Kf`ZPQkC`gZ zIK9esk;-a|vrMvm32*rSq*O3rpf#6imdO>EeA6k9)?*VBThV|0l>a$4)(>Ac5#UHD zKpb(Sb9JR!#rGXjUm){Z-H-CG{`kn4YtsiGbPWp7A}`m$G?xEf(VA$cB%f%vqkpqy zA%wfK^I!~^k}1)~w<9X(V~`ud7CBUfquH688X!@?_kcjH*YF96b8_7i-I1E8FWrc2 z(v6L))cB8wqfTEx3#0b8U&BB-r!%mi;9N!{d+pBhO^?{fvmetWq?MAGz&fh+1!zNZ z_(JYDC#TaLLVg)O`ve}?^XJfS$(LW8@80NKWrpbvTGk+w_4>OZjm&{fYXMo8667>x z2loV3=b(?(|>>XImer(O)0OpJte>WpK@upoQ{ zIpy^*DbePuut}Pi4W;JBuZZxNUKG7dhz+6`uPq2$%z^ze%*B^_D@VPzNxbJY@Umaz zD%ITXgX#N4EBNi3Z?hY^;QB5xx%t4oD{h7mOkT;<-+SGQ)vv-wpn$5u79B#>8;Jj# zs~I!r=i;Dsa77SLN+DB~wK^G6Q&wxFpc-S>5Cp`{Pke|BNJ?0E1-;tQBDZ(O^BlWJ z+wP8A7>ZP-1I-2wDmzx9xi9w!bPY}WzJ&noYaJZD{F~|`X9(;~YjE6@Ochbo8jR55 z9{Tg8H`(B}yCHg^C^f;Hd7!$x^YW-OXf@~|%CZSvK_40K^_DcVKr5C=%~DeuzS)IU zt-bygJ2#Xf7*4NZwu$2`NW%lcKzGP`Z?4Gm=NWRQCr$P0rZ*w%=csW4b!M^G*#vHQs(|KB>^K{*#xy@Hh@1qovWF*4=y)H?UTG*1f zkIPrk&5>dP_V5p29yje{&LrbiIY${P94>9MdqVd z!QiWDza(qK)5}RV7~kV-QvQE?l4)8So%l4=Epmhxc;GN-u&2dtVrw+_LHI^J2 zXlRl)i|>YPu?SHmn)3kf+2FUpONuMQeQbp!*l>ap>Ub-7eAJp8`jQAVY*wdQXvS zjKK|h?Fgdzb{aXL6RGrf;`dX;&s?)kkmBTT>3q)?UD@MIZ(Aj7XJ)miNJcaa-!C4{ ztyFpL)1h%H|GsliO(J>AK_9!sDL%}&`A`Fe829)v;%>9ONJ7P0Jn5pmZ;2!BnPx11 z2{-oQ&v3DlUjfvoHzaStMR=IuS#e16Im|~^<2Q2|ac27w^3P9~PEu(6?9ttb8F*(p zAmD^bkWgQL3=Uc77W}#b${e_s_l*Sf+!=8nDwBnO)a2xY8g8h z1XCjj*1p+iCZ(QrC?Em|@>YF$KF5ODepmBWtp#TwQ8WDlE72FckO8qE2hN1$UG+B@ z2p5RCkz?+`haz+g67;HCn`kB7asGSRCBH^mB*C?YHCRP#Xhl7|L04GZV`YVpL`@=u zSzM`IbaQFA@mim#t;je#&nIaS23V)2ZWS;ifvI}5&G zUG1klZ$khRMy#@Mbk|OjQDl7koTgegZv z^Qt$3F^&kM5Te6^+kmLkEW20E;5$0rUwk1AHl_#F`_yywEE8W}ZOaa7BoEQEu%H3` zlX6U86TS6jBWldBxS-(QlYcE&z*1$7EEEZI@xo$BJx=J##hgaOwqUEXSi;Y#cuUDL zb3MGai1Qsn(2|bJ7yNsT5o_)xe;`IDUMI1%H@?~6XH)~wJN2+pUk0-WK!CVYJby(9-c=Gf41{QjDr9W&A9A{dT(GJTDPI52nr%xQV^G*h1KcR#6+nDrD37#WFgRG5?l!e%5p?nE`u!>|K8* zNYNY{{S=`E-Rsx7tu}+jQ~GfpH(W1xexmB!6A`{^WR?UWnxu^qin0|U?&}v_a)al5 zDmoTq;Miymv9>V{&-%>7#*OPVv(Vzaaz1F2l3@}WScZ>!b>@{X7dpHLnCoK*TrP9o~I^cF9cpPN>s@2N;$|U&_pcdf#(?lY#{JZdMil>U=PP6<-%I`0I=Mnp{_apm$ZQ_J5Dm!~kiL*+w(I z6<1;1>AO4O(nEQSi}o7bbKFI?s|SVyU$!2vn$fq4+VcZq#Hb&S$X+2XYBSvrkJm}( z!Y^uzn6@CR;;Xzh8znGnHkj>9ub7btvS)Q8u((4RC9cDe`~piHeDf=a-(+yFA4?d4 zOE?yJ!7E!zP-usXsq0VJY?hZaUv$599&DM;&2JXrtxI{Td7c0<>CD$b6sF77mEM~R zY89<7=9qKYgBKh0ap)@%joGTJC8XdcK1EQ}goqqeE!zl2YA%{g$Ek}yOh;4s99CO- zp0r(W01!VPc-#e;?Xd*ltVCsv*IY___I(I_@J&etA||WsK$omdVjKSGMHKbl}H z);iHk$VCo@Xue(?O+e2^*+g8J-s@$pn|?SuJF3AHOt?%cjJkL%|KUwb0i}`%Smy_O zj;7DC9(O`C_gbJH!9>=1Bv;e>_~D)@yUCq)%$SR084I1?A&|`g?r3olyNf(7+D%6c zor%F>duSk`^L7{Bynmqs2c;}t@Q5ccT<6nPQvZX^iK?tEu`e($HeF6BkKWvzZJq&S zRfcnKH$(!0cz1mgS69d5rJEZLAY{VIEOKUMg{`II{N3#i%gq__J?jx9TJk*@|1E2k z>_ftl1^Kvs<1F$ij7ge+!KXf_bA$cKX*dH4I#=thJ7`2|%IWA?ll>-o$LYr|OcR}& zh)q9{n_u~xPXzkP*`M23n`WUb$`#_46-c=;Qa2Dy_=Xm(dt@i1Gtp}&zwQp43spju z1o*DdlZqDRHS6 zIqn+f=&F7=4uUj9m z6e@&p29{_LElTVq7Obe~=imyCmDG~&M z>DOKVS~Eu6W@vP^%_=LzenAiH3`z=EpqO5zim2Et;DC@(5qOY#nQQX~_U^hdAZje| z?3~1$UI|lh^w!XSgv-OvBm7{cZDT!>K=z9PG&U3keqW;BsTuzp^RxZ=+EN@y?f_A- z6If|ohk8&px5H*LF@srjD`f2K%AB$7W-D@P>o)#ReSb@XKOM#opLqqp5u1FbmrqTl zu!wPJ40HIP_~N2t;8N#csh_hOG!?gZ{wZgxL1o(S7zt&E%OyDz;)rp+)MPU3;SLO_ z+C>Z|y2$<|6wFm^cnFh2&z9FHWOPF=f|a&lS?uffAZVx!RraEvcE_6DaVx&+!u}Dq zDfXhLCx~JQDq&iqY^dBaR~lp8-*a=imQYkYeXew#`Q#bP?&Gk*v&&(Ka!)VrQ}Gnl ziYveohSPRo=U4we$K5`>Unz)=ZtErSK zpV8`>JgBp8-sg}!L80nUi%4|GpBV#z{h}mjnh|&Tv?Gr$w9T5YGEu3t?UrI@yU|Ax z1&uA`pV49dLRi`3ANtN{V z6#=<1!d`JhQQ9}+US2vfOat%6a{EJPeX%!x=)wVlZGQ8#cm97z1Ahny)3iqV2V`^D zk#O0oH@npQb7;bzZcyh=X{!#FLPWbbAuOHcL- z(jLZuzhr56!g)%LB<(dr(;9m zoK&l{U&U2DWd}ef ztU9wE;QO%tY9SE9Z9T?AQ8^2e7E4n8D!-&@@n`u17k*vv_Zle4+@>KVYB`a|b?O># z+3s)cld>m4f(te&X=J7FhJ~XDIpt#wNFzR~m0e?O zkba_%m3n=vw!ZNUmqSP*nbkP27H5ikNKbr9Qrr(e=7E7R`Pv >;kcnzM)E3AT@- ze^e@YN8S18U0hxN|6*$L%C=I*OL77Ku8t}JeWh9{HyZ?pqjMsWIajHiCxWB-T4WZT zG6^*`Oz-PsHpFR6&CZ!eDvXTiX!NAgx$mx}&Mtsg&rix9DV9kXVNundU1?{)$x_wBc>(p!fL5@TB{LbPG^Z^#&+kf~`w;}btZhVb@NK4aP0CZne0Sa5 z$gs|Z?y?<>l;44=3Nb)DMhP|wA2;=A|44PuAt-Kr$@;MODGpl7_5iPcG>j*T&wC~% zg6>wt;SyvlN}gND4##ho6{#;-^v|AUh&a4{ooS1=KUIt-H9aN&g+^~?!zW?f5`aA? zfuP6pc6vbnwYoslR5;ek14exNywG*^cRH2x#&UI}g-67JB@8=1<(W5_T7m?$tKp|R zGasuqZzWo4Et;7rU*BO;8eTLkR~9nW!!_d>;d@i+uzl#(HP-Jk4#)Dd*ROiKX}t*i z6ssgF^Fv8SjzP)4I;oY+hqkq*dy*W1A(_d)ml|N|{mHf}4=TVneyD@jFAT9vnY_F%Y=5xMe`w&346H0Y*|YI7xM}r+ZgC zworSO>90=932qKApF`E|^uLmr`uv`tm7&e_DEjq*uK*|%$wZK+^Qj=aL7WoD6sco~ z$+(Kia8_quS|CQEp1_z|qG90{!*~Cntz1{k1+T+byUlP@U&Yk>=4)eS?gm*bW_{u_ ziDKXj+TuyL;lj@3=bgF=G@fH?|E8$4%lVkEKE|5xJSr=-!w@L5wqE*Rl-P2^tmxnP zt8-1iF-4%UgRLIpx6^Tq9auBe%n&<7=xrYtQ*fEfqQiIx>2d^mUUq zHBgnXn&*M2v1e5Z*L4nNMd~KN1Z9;CHw@JS_xCktn8U!V{Ide}z}UwWW^DS3Boo)Z zWZi#a5)ezSaDGwR1Q1T#z93uwa72!ex>F2$SL7=v-(`Q&8T!|(4(e9iGXYut!8S9Q6^Hn_rLBG(J9`8UTt*N+xp-?S`XUV@MCtuP2n;&m_C0?s$^N3raI0I=SifyMAb^ZwYm zHlN1*_y;gr7gA378_xF$JWK>guRs1ZN%g7`8Kr=#OWYx9|7q8vc$JjoRqWyeZ+wl; zj3{-ONi))U&p}Z<$^j4Ylpe!2Z+053;5M7YZ_@cyq+7fIYBq7wL7L?R5>f0h=a9fY z3JbC;vx)4|4YKg5oGPccs2ZDyQl(ioH#KyResr^WM|Y4bh^GR41V7F&o6(Mn@ZNqK3sd(38zb$`5{1(|F4r z&BN<|@40!i0!Ce0)}#vfHNc8TQm{5rAf1qw76A?I92WF!ZDXngcYyeX?WMCgJd&EK2Rf6@Cg?DOsBBu7g2OE909_Qz7& zEiVw0@%Nei`K`ceOvx&3FezDWay&-Z<9Vy3>MAfTrI%88cihx=x)g=y= zu&w@PzN_fk8f;%oPL_sdNs)Xb`_H2cL!7-Qr<>v~Y>T!}G%$2T3{pf3^m7do_fJ{( z1!4tUNAw$shE2i&I41=<1A!9mwX5qu`$T{d9dF(nlQ0ZmX2pUt^I1G*WM+A$$evs% z)4k$)sL$9Vb^MugJd%*07a1QVSLk{@FzgvVx+bgVUlB6R3YycYz<196iwSbX3e9r$ zeFq^9FQkii^zs_WI2<1;WHg853w#^(2D>>?&?Bn+{WxbTViKp(S#;&d67rJ~tF7bi zAAj0?4h_9*_bfQuY%1K=BVR1z7_$*2uf%AD6{qAT*%Zd)U&kdD3WAeNrJ$qB5iikx z-kuxq$c`_I8VxwV1gO&ts+ks{?}pZ0`$(+UKxm;nNItA~KhUAWzX!R8qcvbUOccgU zBJ9Cx*u)+z^S=%W&0$j?g{0g>FY&&HfGWR`UId}A?+&H~s)0>G5*MEW z70Q!dB?}FwQw7>K!ujb6aKlC5I5ACpQw*X$=wrrKQ9+ogK=glFjzrYQNJzGbs7O!f z|H|Z!<58H)Z5>NeDp6(J;cwL{A|ori4)!lP!Q#`~5-QI%*d^fYurg#GOlH?ObVI&11Vy!%BJcKEiGshy-_%~)@Iw=CN?p$SEce{RATBZx%WLHf5X_e+6ex-#)I z{p?@fp|wRW`FgHXnIy97xg>w}HNL{>SXd0~Mwo`~V-8?9AkU5G1r$6rwjQBL- z;#72U(VgI)Jt}*1d$=h-Ci(w7+rja$kI9$)y z3a9mTE)kEj>>{>93x)Z-W)RA>N@3!IJGY0oPj&UKCFFKYk$w4gUnHva$2MGu^q?YR zXS~vQOQHM5oj~mI#++NsybnFwsT^Z^k=d+PD>gv$~(bo-P!L&Z|h7a9I%&%&@r$5&TABis#wmVDN zp;{Ds5!vR#I{%_r;y^2~&)$+kKc^cKiT2(P)-KBwZ5SUi72~w2{Njm)`duX(k5&a8&Nw3pVUe1Nx(#EjIIriz*u6r`gtNAU_Pn^=&3kZAb)MnNaYcLd zhf2c0JY6cvNNQj7yjx|1T(^qL`|n0-B35+=w~>%snXF3%Qgr*Ls6(Xb$_(BQaK&%) z`0I3UyAqJG zk7n`9SIrFlkFs ze(`aYAneUJh7R?>8RVso_2pN3eIr4Kh`VRXhw-nl%A~D#y>+Pvyf1P^?fseyr+LA( zE(Vp~t?q%Pj@wcEIZ~{Z8EN24L=it2OIV_jc zFn0LL7O43lt(Iy)`Mu10);4ci!Jok*`nofNbGEbNpIVG72oR?p;(|N-jsnx!C6Xv# z96HGQO}?P``70c&#PSh`E-008;D&ne&5(!J`8Bb_O{eB!7P~WH)d`VsPIdp|vO>RP z)iyD;fBG^<4_>WAIS^E-R_55p6N&H<+IiN5Hd7mIF13?VV{>=7>yD4h;D39*PoI#D z*^mf#)=3@;6n#Elo4}k$QQ0P@o4s?deJgXcfM~a96@eZRpJB`LnwIu_qRy6Yp?+9) zwjykgJ?|U)`4=UHsuT)*^qNAtS7Gtei&5FMOpZ3bO9M zO^>Pt`6_2_fF+ZzFd`$H5_2=N5#AFC<%Q=z2vEO9{oh5(boq&!8Uy$00kTP(o&<}4 z2tRogq|>ih;?IOliH5%guYVVmoYI+_F(X1_*=A2>YMDm*j7h1So|PX`P^n(G(Mi=O z;K~yIUiAH|ZXu6w^pe<*lkbVcB51oPgrV-VWub#!efhD!=SFRF`cus)HRhsWA(5JC zH$bup0qgQ7R@r7e07c!w2p;U29s6gu3 zV(U%A49MwzN(Yws6-{KB%woKIet`s^^X%u6Y3s;|?;e}-`udpqGx@))@i##G%)GbH zprQ>(qKU*~PgvSRm@i3s(=iZbtl?WL;Pr{y;{%IOZs@RZ?w*Q%vrK*W^75EVN#up= zRDsG88;g(mq?2+?=(6Q7<3*SywL%QxH` z2hC@#Z#?am@_g)#FxT`%uf2Mwth2#|lGgI2b+)t3ab@=;yoc=@d-t87`G#|g6IJd( zGScie?|EyZiU{8dopFCs4qM*y+fc3%SlIx4G~D-#pHIeQB6_XORE9Fr@wk3!dsIz@ z5D=A3+sxO%*e_a9)6kG;g)fD&WadRUhgokl!CM!u_Hm!-@2E`N59V_T;IfQraWa`M&3%v;^=&>(Y&%NGHzd%{T@kVmFIXc6XBk!FDYU0+F3#L-n1r~F@6>JQB|7=rC z(bWvj1OYoA{x5+`vitWv!nPhgV=~c&&dY=we{-jrJa4vxrcee zbk{wzJyz}w3B`5Q^7we@R~;$yGC4!SMd`T?t8r{mk>B1_anvmZ|5ODFM&{CzVVQ9U zB^kd%ap5t!)a(MwQOX>A!jxN3Rrnr%syV=%PU(HFD@k+6pQLXQ8 zv+(CR$-nX=Zi=xwtA)~jxzp*el|n{_iM0MAiR^tkJsy-kRo?;CzG zG`G^Cmxoa@%=$hO$ipnH2wzujT8N6zY7S?$ZeA}qpJi;-Ol4DeRkn-$hVAG`nmV5S z;!2geDt~q`ygN+>;iWRX^N_$yD(~KOlD>0pYB&<0!?fU#)6t&Tjx!U}(cxSQs$4yh zMXr>t%weRZ4L(+Dz@?+5)oQ0W*v_dY3wZbFiyD-1{pXTC{$Lk1@6*L%Z1;zz;}2+2 z>{l%uM`Fv*19pphTGjDU+0Sv6TCVp?wmH%~#oE@JIfh!sS1(RZ1W$d9y|rfeeDfYu zSKL>wi+F~F#w3{HMaFB; z5VJYBSyEh_(X2i5@;VP(m%s^d@Ca8_mTEhnNTr*}{GxkNolbblm)S-sey5DME z{RXF|5>q0{NTMjNxu8`=U%a>Br;|@F<6l1qJyz9~Xhwd&*!@L+rycF+&5gT_irC$P zqo8)r2lut@wC!CM@y4EF`$cVl6H&_ceoG4^*U*`AAKL7f!2A^yfleSi(d8L(fd8~lJX|K$eQ_}x=y0&eU@*F*8D05r4i-7KcCjmGa0Ywc7C zy(*KSFH~Fc?jgWx&R95)trzndYOHaOlN#Kf923)W@%}=#J=&rI$-D>@_eZ)KtKQ6( zAO5QXq)uzuK*PK+)lW072~vz~r_3;R3%V{LZm7v_KPDw=C3TwmyX}j%aN%)Ka2s~G z>764k%0HQyGfyWi6@G(WM+S6$b~%LoiXx_G8Lmz-+&bg|!I=_XMR6i= zMlUbV*D^QO@0p4&o?R8@7mL7kP3311_l8)8@f1&dC{TxY?q@+e4w0%lM?rW~p?*_~ zW9$V%ct}W1`=|wXFN`m&;q+>OY(3x>&3UnbA08J4*N7V;(m$59F@j1iRGspy^D8Qo z;g(>^Gp9{nQqVWKhz?)JZa2AD6#{qj3}9&>EB zo7A_%77@;U&mY{2?<>G~kt~`@d<8^UQy{#e*6VQ2LCE5FbO`@n@d*hMbgx2fNlTb$qC zwKJe|9ON;76(%R13}2WYP!KMAOY9ZV-MJco%;zAY-7(go>TEKe+o0yRCP8Kzsn!%QORQuKwp9CHp83=;!*a&T?Kd(Azs=EHx^NCnX&2IHrQT~0__Q3?km zJjX6$gGwFA&9FXSwyl*V&jQ(~t&ZWQ8Wq=_x#OGFAVuGvDZ=tv#*8Fo8}B$Uzobl- zdRu|7sjrk`L5}Q)BHbiXSACR2$P)G{96RQ>RT$f2#YsMucyBj6FwlC+%~%(89#<|9V~A z^r=UEJUupkutH71%IQ#Yoqx`ORf!=YJ#6mfeUTRNFkDcZ*N8phzWfQE>?gcvUzS(y z%ZbZ@zgX~(clgLi&-dsYPww>}|Fm5T-e#RTuR|=V`pb!08QmPs&&8~8Kfp{TEzcuA zr_V2Y%Vfb8=ag^|&~(Rd{nf5r!H{_3VWYn6bD)by7rBbE7s0{oB4upL7T$XavG?dP zYglg?z5mn{V0&LywNWk(sSB}Jz~JDhWpBXbUldesmP8fbEj_i27w>qThjz4dRQio) zJkN?;W_(&7Yb!X7TSS1&mbL)vM?G2i;0>sZM{hs|mZg+F(;qT91!__Uy?@mZsBK94 z6rDNH{CAKU<(c4N5fwytl{*-XdDGmXN+GZ4>Z)Yej8B+{#DolQwd42_y#hC zOvh_!4HHQ|Y+Tbt72;nXHe2YX7$Um+@gr;iMJldnQeH zWQxWHl5~i{Zg@owA(hxA(|)6f6GZk2tM=%OGP?+CE)9I&V4aA}Nm$Hjl`k zs5D4Uju*)%{s1Q=)_{L>gB!^;V*N@Yh%C&jLT@lnYgPS@DYWe7I)hfhc(?auUn_%~ zvbp!?U4&Nli&kdUfn6-J3U`ai@3*+rxJFO94gM!81lDq6QWjHhe!)k6Z|eBNDE6+q z$8MZ>p`b2PKozK zG*4dKCm+VExXi*(KX%6Vs1#cY?u(7(#zq_jKiw1F>UecQXu`jH~N=YOV{A+`u}6=ErY83-so>>I5Zr( z8>G8iLOPW0RO#;SLw86@D-F`!ol18j4H8G{x%vM7PrP_$7~V3&+um>F7dQp!ePBi12aMr94ZZ-0c_u(&q?JMW#($64fA{Dw*+qXTQ=?{X zS9otzCys)!HY1b@&HlU_K-RK&d(rza1@;A(LB1q%?FbU6KE!R zxO~D&{S<>tM8&yPhsQ^3s%l&<1~ug2{6o9h_@Px$rY-I7; zU=a{l_kXb~^RBd_rlGk!D$U!z8e`|5bNyYYQ`7mUjF9{uP@MDZRJUJ&RD@3(z6mMe zq3~f58h^c@VrOLJ5FZqQe)D3OJa7gYlgF(zWYJ{It9layI82tG9b(5f4=!J0s+uS-)Qf z=WgRSmmM#$nFXt|4(MsOko5_c@#ur2U(iHos~Y9aX!o(gD(1QkD9&@E!^_|0ePV`q za{z?MBlq!p3&ARbCZ%cG4;*p=1TAo*`t;znN0?n_XnHi}D1Q;DCx{)&zk2QNW!vE} zv#`6*zxBP{X2$dYnMI#|sKpmvmbDGvoIe$ul~jyE@ijN|WI+DXSe_}5`8N13KC;5? z+}{Ki7BmCB{V}@B%g%H`^N3AuSG_T+$zN2XG`l+;o+DF!5R_Qgd7*bQXSfKpXt7&j z$)s1hH_sC|o&1l2m2V*$^=jaIAR28gW+UL`sd+1$WV~79rF;y}_W!mGR2K*&e3KA@ z8@xrx8m=iv(~m*adwRd`(P+nxMwFe2Tx4H2llQ4?no%!}q5GZI7bWuudU6L5WI(^Qk8D1$RN?5A0Kp5MVG6w=cR}OuZN^(|5#U1U z@?<+_5kEHVI}s!7ymx#ToT6V7B5DYm*-YUes*%xN0>DSjDc4ZFRz(oQ@%N!+KKU98 zf-jHPm)BE`6X_VifH0_7J8mw;*h`5P}<;)N{4d}B236@rvBi(vT+G}nciTnj8& ze-7^8-dp$1o9T5%u08xF&uNa=DZ>mOKc-Lkp3!!PSCDk|VbA7;t=l^O9xKtbn>2NKyO>&{ZQkYI2unjhsH+xxk2BYT7Jqye^AT;m zl=*SJbba!}0Eg5}&_b3UbbForK*)Rmx)r}SnE^ciVMN?knG61PD1&~(CUB!d; zpXuPg#n?}9X9-$5K=yz*{(QA6@G6-ms9^k2u9*6a_sfcCUgH}+SsB-I*iq4*~8Hx z>H|A8qWU-6C2%Lc3AVlNueibbcRa_dk#1dU-2y5EpNixCI*2#3rNyH6txR;_MnKce z>CzJ2og5XVH0qpWLb=^aU;JPpb-VxSZY7b8cY4{OFa} zD)*D=rem<{bnK&Zdi-4$V2>LD_9gMCY|Y&r$3^BHzmAJoE-=JrGe6YDdE3%~<5jFx+?=`mQ^aG)@@jlN-*oeYR~t z$;p8#u%ulAjv;HMkIBczEk*^hK6F+kq#g(a4n-Q3o&2ep)A|BWZ;_y4EG`}iaa8sB z@0{J?mY6NXj=nmW1NwPPHPzc_imGs<;S-Ez*ORrED96pI%|I9$f&q3SWs~WU&zCqp zhw!My`Y)-K9;(Ve&w_K+u!g&akSml#jL_LDez}nYrm%0`v#LG~*5aI1YgPFBHtX?+oUA+430ei#O_7Kn|u!F#R0T_JQbol7}tAWF*?C1{nNpy1@7{2b+R9skOY`IXO33eoZmGMkXU zV{AkW$mWI@sRSEu_zXBS>wHNy=5-S(HUHta8A&$Q*CM&DN84If?ti^arxjGvTLySv zFh&;o%!ozJ;r2m2t~=(FzutdjYTkj8-LJJGok6uGZ2)>L%Qao8BByz@y2hcXbSKV` zj)(N}c|DfWLSKTsSFL=3ptN(->QGfRtu^z*@S=VESAPf_N15Z_LEHwoH3=uIl;3pz^Z`a9&4_%)aees#JN9$WVD67DuNzBk7Qd2TyNb8N zE*ic&x+p!lkmKF@pRVPVk@@%M5^)YqheoljSs)Z`@bG#vnq4;ptd*2K=ils`s^Fil zfjX;{VcO(e{QYu|)bkj!kX{T#Oe#UjLFTF z@ENe@z_WLNo1eXfMIKJ1w|i>MgIuN%UbBeKgaA1kkeCoy6{EsdjWX}nmU9Ldwk@uO zTcK+n7BrR)kW!{Q#wIr!P)d?J5nVp~5eD??uIfdp00ImNFt67WRofZ3&dA&Yxa?hi z3L72`Fz6ZR^(JRq5tKIxmou}SN10xcwy|$!ae*q5lJQEP&6rATTKw@MgvepNz2)&CYYjjn0^HPwc{JNzcm6hs>AF@!xsI{vBeTZ{1Tu`CLUm-=SE<$XbLRt_Cn37Ve0KKj{qaurXpOGTghlRZ1-$-%XjK=M`cO% zU8r60uRjCD(RuwFX=3`j{X>_EK-#YR$`>_PubS-h7=_-vY2|_+*b>s8!qKgUN`V-w zrTSBV;b4+W;n9JscO!`ewf&i_3hpD};DPfoBxlX5HjhV`B-~n2PI~w;nZ5lJ<%;zx z*pyi%kH7e^C9lB>QFfxGH1wDpA^hiWLoT)1E3O<)ptnvEy^<0=Z@*T8w#dbk|F@qQ zWjj^~`JbQUwHnljHhEA{;Yw&h;v-I(>#V; z97NVSTlVuYB30xZBnWb*>eH9c(tqFDP+Tk+v*zy%ZmC*05$7ffIXzv@Zj z`1@nG@e3&1BNts}oT^mtC7k5V7n?=(TsD7Nh9TKy6mZp7mg>Ap{Yrs^oLgP~E=?n0 zFs8HMKrotMcKPo%1<`P~BL3lc$Uo}VW#vq+0bLDYLFjIBY}|L7g(`Z|b6{)n$3X4^ zorI6@R9j069}$!M?PB2Y zh-h^B12hH+Nl6&@lSR|Utv(GyYDov)AKyMOHN=%R#|H}n+d)PSOV)(eNzi(Xe*W%0 z;}_FN)9~$yb#_5PhCBRa9Z;C^X#$<}dEWbkU8&E!6l{W{j=J$X2I+h^u0%!oRrKE3 zmkw8}L|-ygA~T&;5HuLAGUoT?;3_|*iyMGC4R<|h_H_0HNCw{2c2wvjjD4Qkke{Se z1^qfn9I{DCXM0FH&hIz`Q%&Cq4v2t%LP382B;qgkfB_>2BS1k9C>D1UZzzP!Z{Upb z?2S~++tXeZG{a%>JL?NgHXm$ob9K8~-EhPCP3RDONHwqyV^iQ_BoYeHbeL*$ywAw^ zvU~jf2OmG!PYc#)GRTX~(8pa*V= zI_XvdtZ?6ANRMk!HUsne{(*r^Ncz05?pt&W;9{GSERv}evv#e$6|#+ zgI{l`8&1*gU0DimO=5ndeh=kBp}N$h{|fmo84=`^H#Tc=Se56!37Am`F{_cZaVJ#Z zl)OUuH}OZA%YDEQn)v~n5u?4KkI+`^D9kwwpAA(AMuX)roU){zQYKJ3{Ec)3YzPwE z)58i>qlg`wi2>^~{P8~KD0A41uU2*Ax9*!`J0yQC0!N~Ss3;h{M)>8?A7O46F}sC~ zH`tI<1O7Iroxv?2CmOA}qnVxvV8jUC79M3#U63Az>k#BWM-M655zvg(*sT@bh6RclkdlA#a~OQiEKvMY@|^KZZhJ zQ;7!z{-`rHeFViv2X%;#tvF%SI0UiEYw%>qlu9t!;tN4QnmQ?~2BEwAhs_HO@w@xW zOVHnc3O}f4ORT?F_gXDm!CO-(Dc7vmPB z@O#8%*JU`Am{pmk{UYT?uzVn|=e2{`Kq)jIm%GsyA@j}xlPVr;Q^VaKg=hTVx+Dsk z{~Q;RbEt&=Ru=E%Z_n4a02W=NIe!!v53i(xD1C_Qf`d}g!SLS6iKfhSj93=A8+f?R zE5APEdqhqgcqPwR#G|k)AxT;xtTfaX;OR(85q2&R&}_GA``jGHP&I8eq=n>^*A;|p z640FexMmBc&4Tp@?kO+`{-{%RQ8+ECxjk# zmyIv`mGKjy2AhHr8sB_dhT}v@5`vo{9aZF02*;kN z)Tv?<9c8(Yb!%{k%EdHb1Sz@V#GKk8gmxm^SV_}2ZSvJsea*UU8XLcv^w81*zpU16 ze-vrBH3SM1FtQ4TE3WE0)exR&o=A-$V-JqtA65l92TyEP6~lbPO9MEEK#0kN{92?zDJ)GRhO{@<3Fz845UdqOw?({FkS?MhOIR@SVHLaIgyVh&)Lq;C zxHei1o$tkvUq!Y@oTJDv_C#OHb?dDJ_EIz*U(IxYSY%y!VU>6TOaod!`b$!PtlzfLQ2HpfeKWwzm(H zPGwHCh9x+A0Y-RBlW@4Er{b$|)8zbIeZAN~a0RsqlK(SnD*^p*x!bW-;jQc8n+6lj z`u5=OYTgBmyuIrkKCwEnHx_6sinYWy@^0u~3SUKIxb&HTGl6Kqz(x6|j3Fy~x`8uF zCv=~Ld(vidxb}HCAzhQ0OeVBdC%&Az$GJBmvml;O_I|S`M5oiY<+HqeacT@fo5%Se z;J27?h9I|%q(`sR*0sOQr{vV>>^yt^@#mZ|HDbkcTSNbx$oO}dh9iqbPc7vFc9mC0 zx9*|&8GFV-0zWc=u|EdJ#&u;`_E(LXqqT`sLQUGMsrC)*IFlpV5nh;ukmM1GKtg`w^$ooHql zAyux3FRn{V{Z0PM!bmkSXG+sC`GGpA&K!05vu~+4XB(QeT<$@FZj)(^_LomJux3ED zb*kqTPOy?2by9GsNIR@b66RmwY7#%(=@*23RZlFodaKjkc@QOr&_FSAUdGeB+Q!)b z1pjC9;78rCAOCnaPM)*Bu^Qnq6V|Rz$do6bd4!1G9jp4{=fyWy&$p^nDQ@g7P3jz_(0A_%CkJ%#X)NDKe{S{^$K|~-Py$R~Q=)g-M0tF6#Qz~9jBpfWWC||5TqDTH{{!ER+YeM!<-V`+N*An<>T8BKOhAY();fZ?IBeE zMJm>B0JP&y zlY$47d0VYtu$$K0N!C4g(7B&y71_7g1EA4FcM&TynZE|S$MdBsGg$~?q?MybUftfW zPN>T=fJEeZ&EfIq*GMzc{@xx@*m*Dl>czaiLuSM-U^X4ddwI0}d$vXb5Lj8YrImnU zaCUAkmO3>`oY=$1L;jMwkMiX}{DqSp^|=xd3FNEhi49Lo@ZwE-|F9uw_h?UVcL0PU zJZ#C}Bn>;ITI-t)b}_}X4VIL0*!QH)Tm+KPqd~q-tXMTdoT%P>N3h2DkdcWm#%HFf zTtn5a7$&=6YHU#1C3OIOV$!7(?vt+8_gLLuIx>nxFME6Zc*+ZsezFJWWw$|T~1Hv{7Uy#n?AmP zTY3@D+Khnf)V%|A_^NboaQp=s=@{ALV$LEL9@YXU6IlTZfcfUCaBU)<6nXZXGrZA} z3v2HBE#BE0M_;B*RMNri6OBV3r?QWb3+zFxVZGB+mNsy>MAojIAAe3b0r-$dyJV4l zH>gZzlNL6Ag%@7+dDfm1S+CDpY5U}lkDi*V)6XB!z9gv?Tssr-zW6`pHvZ16yLf&S zv906&h-18BFl)t=`$upl<~nopVKbWb+Oem^TCv@*$}WJ5{K&lkkhuidK4Zn(J}+hp zy&EO`3U#~mSZiPS!T5Z(_402mvWt|nKl5$W#)H=*iL>BQD=`{qC=9kf?p46_;_Ca_9=EHK6X3jP0lBe zNHr;VOGd77wo&njt3mM99u!m8_o5ADD;Kg<>ssbsBH&t zVXnN9N2@~Fq@n6T>)=%BrVaZI@d{eArPLuU+-boJNAFr7pLb*sA+d{fKnSI^!=A!x z7yC7r%}N9dAY}*vqk#I71GZznFMl%2lsqw5HtO5Ofjo`Z>AGy-=t15>3D7W2W8QhCZt1lgW>jr}OTG~#pC$YRW{oNcDIPapcBkd^##vDR0`IOPTKB$GgJ$*c zuH~$b$BP?rqV?ymGz|b9<6}}fh-U-$-5wxf#~yy3%6;cWE(;?zaVu*ZP4Sg6*JOEU~4Kag*s{6>+#5V?{O13F&jd%Uys~J5L#bOW|~i z_Jut95^K|{S$G9WJ0S7Ko0(c1aLSZGenrjQ}<|}kdlG2GPgM(p{M@r=r%rNj} zb+N}uMq!!y=z+!-P5CPytM*n(lXYJ>x=yPr6QGf@63ZW-zI1SUVIwWgNFWjefmH!Y zoPpO~>b}FZXXN^yWu)^cX4pN&&?&{A%L9OP7(s^H;%!O*DpoK{D85xb_Uw#&ctKbu zGOK**K+CWsT3zr?A{jV(HUE>Yu+lB%R2p`xCA9!MQLMB3SJO_GF;x3tDJByD=QAW# zf;}m?!(y8iPGLWFFlVk)y2-oiHOy;anr$WrcYqymu*1e+5b=QGzRt}wJTbzTi9LfDAoJI~WON^_>mgAnU6meRCNZG0fz4!pj#Z&Zt)4wX{TtKUZ?Sbd_vi7<-QGnd*am`vI>pkMf6Cs>_Y zZTF1hMSB*eAY4;Ebiw}Z?`$|q^=$YBD#1Rm200pA7bjTHnn6fF556| zS64P8jF&gY2kd-vY-_>#+cWKBHI|A+V8p`2Y|=0DPmlLQwfrT`GzJl?X&Kp0QP}WT zxD!;k$L?#|+*1cxPjpieQz&_{RHkGTZu{P-E{*@%F?WetFyTb~%}E%hx_oTg=P4}NZoHo2FpgsactuP5eRhI&RXMrair##E5aawtAQZtP zXC?`n-#zQ87NnCVY`g!r)*7>Rhn56_v`IR@5b%(PoARAjGzHI zp1l&`zM&!r82jC%hL7N3$k7S_^VD_#aeDs>uOlTphI_Q*tR;`^MF2H2&RQZ+h6Mu> z?6)Mq%#Jq*f|ary2)IF(-^E74UvpJlOwl&(?e;|8f7Z ze%{8~EaVBe2%MmM>0g~dcRCR`$Y96WhFEor&&XQjA82(VOiiy#^ZdEcNyT3t>rqtu zW%2Pv41azCtkJP!lygo)WY*+=uXDf5#z|f`&KFWGoJ+kagd3gqrL^= zPmUd^BsND?h%BjgB-ZZe?PsoWN%C(AxL|XO2odIIG8#sn_m>eZLgZ62KgDnrDhI9& zo!C3W!KK;deLLQyA<4n5UoTr0A#$Jqj-A}>2pT8JX-aF*zJs{IJy{*0-fBn}{7ugu z;l2{Wphwv81&_Z|3C5O}!8&mIck;&+BK>mN9uYQ7bGY4*jgO5C?_u0&1~T7LI7Rjo z9Y%8pLkK1vSRcV8iMxOjo{~Hs3EK4i_UFX=!?@#+IHwX8UwTh=dULBNpRL{>m`^@@2)}nJ9E9W((jRit9_o%dv))LzUGu)VF#(jNFCrzS;^#5 z7_ZzvRIbu@C88U3C0_NK2T&f;{*@e2_z-a3d$oP+am=N=UVtjOVEK~L=;u{7(_W%s zbLq?W{W{a6?-Y3Ktp|rJzK#O61Za@uNV*Lm*RhA>0IXiI3e z0YSTm_4h4j0cy$loSOBJR2R#zn=-|kv+&OoS-kWWAm##Hm|->g|Gxvb7Y$!LZZ^v5 zin`lu7tQ31ZjnL_f^)%y6-2V#_6-M>JQydFOB7+2l7|zl-!Uh2Pz(|!NJ+KAhtIJl zR0SyqsmX@SdWT97i5&B>sIaXU{Ds6ZYz@FHq?JXpv5V7H_B*-QdMzi<>0t~W&?rn+ zpPXMa>jSbK#XP2&R&BtDX061;E}55o6hpRM46`u8?edMNv~cryC4fgd9KRF5R?3nN zU$zFu-kTd!F)@|vba>O%MMRH*gH@MD;R zmT^G<2k80oOrtET+HEb$lSPJ($#U-G6!0u`!@R@J|eb8TP&MC_*cxm^+^8R*GLNCMK}F^7_`hGXi$qM8Yr%B zs`gd$bIZTH;Yp0NclulDd}lCrk~9PP&EArn-!7m40%jJxGq{AyJAda?J@)7Q?jN>= zGqyF1aJAlT0B;k0xx8ArEua(m1)s_)rIrH8rI2D>~&YVoU zq204yen&->15cw5bJ-V|VLv%3GAxK}ah7ybPue6Q8|bXz2;VxLr3TDmB@0mTcWP66M04WTMoe0*3OI82!))#PJM6Um+5gVQg8w1 z7+apH17N=n5K{cZUO^)aNSc%*^rRwsB!uj>@ni7&CP#hof!Ho_yx*3Hpkuv7X!=9` zlD?N(q6t|j`W5(Oh7-6%#5nL4(P+j&wpO>gJ&Nk&c63D1S;sXnc~sGWLXQWHRA9P7 zn;e@k=S^m_F$U8I*QSGnri5>az<@e4eemzmoD-1zl?9Ugu5HL5+41`)>bfu};B~i} z%cQ0E_he}f`TbI)+TPs{%7vqkw-L3;m`S18PsrcML{XCm63jWn6qts?DS~aLJ$DZ{hn8ll!vieAg5GPDxoz z2#}sc(n;%_(A~N4>^%S2T=7J>+WF5-$W{`gmuBJxX>?j^O++H=Q$FT7YS)J19i!brV2 z95Vg3ovAU#tMk(ajeiS?q6Va`A=NF`*N(gP3B|MOJk@IT>6U?>KFb5)8foNI>T;Q=6-83qwc$&fhMPOn5jO$pUx|V2Wc<-GB+I9+>J+t8|uyGM(> zFL@Nk2Fxq%+rS~V2}pUA#eZm4_3#k=G42wD5!?M_IK_IMx&#bzbSygg@!99M_gNNcEY&}Uz z3K$V&x)@qfR5H1W5B2!p0R{PR-32hWzFe}-%LKLldy|KtO*;&d4=2nO59lPiS_bqE z|L0HxWx^M@w>TBHt5;@T*FW*rs{GYIQ%edVXgsE;W9-txwJ6!kZ-uxoXGo}l85hJN zbGmkyudn79gPH1hryoC2HkK|~ayFyCAdazD)rR&RU6*V|_a~7_*eK9?0@RPo|7OP9 zJmADlfNMp_ah!&8Bft8hSQp^%r%--cq;3+XS-R=o5Z#aB2{3*TEoNE@Q?pFd z^Y7i)#>v2S@`@3NcbO7hh*8ZJz(D|=Q+4~7> z;gxznxiAy-QBM60T0~JPZSQA3_~4wQod3`l*!A1-U2l5ZVH~@AzI!UEGx#>tvPx3t zRx5KoRuWoq}=2*mQpaSJKY99YLQt^c0)BAAKj8!DPSq-u} z77c608F!cA!XqxZB@#I2qIodS;ZY9{2h$7xvi<~I`643c#g2|sI(J=bEbROwqw^7} z3Em3Tg7YOOn0s*(U(vnB0*|czJPQz8xJzBVzA9<3Ufen}6tJFuj~1)tVMz^5w_-xSl;Tu+zfuT<%PlP~G% zNk~m&2Z=0Oe7aI;uPQH{;8=hz-iy)Y;!Zu({D^tj%p7c@h$=2D&D@~7+mb3Eu3nok z9NdoB!rU6qkS^%y0J`!GCDMzNTUqbFa{W*k7H6XjuFl}?u=;4Tmtfx zY@1El^}?6R7%M!Q0ltB8#Dhavfj5;$`4LXny)ObINH(u9vwB+ z#jLi;Go;HH&zTi%`m_gei-!FR*;x<8G}T<0y~ovR2IjVQp4UXwuUGg7q8;C zW@U7&NvWgh+w4PbwEi!L`?uo}&TmjPW@~wwQNoRUMc%hJwFTXBX=EcFBRi@Gf`Bd} zK`Ys}(>l>hR&=f%?pOi&?J+kLdMjqTg23n|D#wn1| zC$08X2C0W8uFMZsve;H^qrL_}vfzbbD^3ZYi*QfYGyzh3NV$UzW6g6sY)Ac6qum5} z68K;Heblp-tt6-x1cbklb-AtWu#(&Jrn3nJY&f(8x&eiIfmBJ)K+CWLz>WC|9Dxi6 z0aCYL>hwieBwL4_n|#dxeGtiKz1?f@V**Dzj^@X|W-Cn&6tB<|Vv$3=Vb8r2-qQaJ zy!3)4ACnZ#`bhJbIT*PB=2$myY46iQ~$UcWaEcC!HJkV{}qPm4j zIA&c&db7P~@kO8daRVK~yx2}Hx+K@VV5!Z0CDlFP)R9Yrs6usPPf7RjMz0DKKHf?e zorD>7M+;tJC4F)j`OT9pJ#uU1nSf3&6#blTSI!W zY*TIet4w_|d$!;rA-F@L*PLUvhkHsyQ)%-Mll&TJ1akv}~iS~*cS7yHeW8YMMq zxAMq-sHVx58#(TcR4-jESIJiDTynBz_P-~;!_^j4@4puxyVw_E7P9(GT2wlh^%#*? z`KM4&Snk5@_y-c~yz)F_m+YlEO=udAc~T3C9IXlGmCg!R$29B_jw7Di{2H-6fC0bp z_m%h4+M+28=mYO-+CTh#P*UI5j*6~jvm;o3*KazQurmI*yh$tHBnX1rOLZXqgM1vS zi?mvvjFPE=EpQHz?yxXX09{ERFeHN@QTR?D){lk&yJ|_sUnch? z&nCLVMfuTGPOFGIQ-fYY&rX6l&x?Umr#iaYA3d3)Ns?xp&omJ3B27b?o|&<^7UAku zO2Er>1QY$^=%@(|x$#G!{##hUn*M9;md|E&oekC6R6|Vd8B_`R_5tOeShb$c8>MS3 z1EN1vrfMvrn<4K_UVw@^QupmMdl$Ue1KRz) zxp2Ux*^0jA=!atJB?e6=my6AU(cnt`(~WO$rYnD4hL-l3Z+GezB&5>iRrxaC7GA_> z{R@T9|1S_kUJ}D(1Nje`4%4ryoJvBo6(4CV+xb$DYjhg?r zu684Coqf7Qt$Bh2Al-nEEpYcYBa_lK%z!--t%GjEsa#T~feAV)F=b337U_q;&0bsh zgW}bAiFm$9eQ=hKdE0#!;fiBf^jf4rbVxPYCba08AC! z1fl{lZCD`9sjCt{Wh)KV(qQ9}TFp@@fm17E8rr(O+^y!$4s=| zI)TP#SqJs1gG#>O%p=vG-7gpi@o&Hz`NWdl1Y+)f?DMuMx!Y*EZa!i4dK&*~ef*<_1hJWUR#2E{DhqQ@>ITR?k`{c)Nm(E ze9xd570Oc`8ZQNx1ZyW{{C`$k>YeIshG8YI03Ov&5E41Z4iGLyJqc6;Uh&#z(T~9D zoEKKI*Tx8Vmz|C||16_uSOO|(+CM2BZnBnR5Ae#dDcO7pT)htk|B^%u@BsDajhAH zlDWhU(&{P885Lsl&#%Al4bxxcUV9Tb$Zy2Hx3Nk&BsxN2E^jY@+u9>jbu9Ug^uO;} zN)&QyH3|QGO|hBx0uSUaWQdV@9*_kRV`?=39bx4v@MGjdj3cvG^nDY%3@9N+#BtTBdqxfp4pFCo5+QgZ zOZTAf?YQ@E?@@1h#_*f{SM0#Q)%t}RcBid!)~5(TEpV~v)wDDwQIn8V0g*Ju0wO+} z!C{PUEs5~mN2H3wtH%}Be`dv-z;|z$Lf8_hj>@>{lF4BT3uMv$0iuVbI+?4Qdz;^3 zL12#Jngd>*%^vo74(f4fw*H|V^q5}^0Ty}VlMFzY=9wsJxb}3V-v<$SHUuFYaGR^a ztnNfcMKS$G0Afl0^B_=jdBC-GA@H@MRV4d9mK@Cw#8EknQ6FQPoZFi0{BYODzzt~V z?F$EvtQ3fQri=rxr^N?5rE{Vcmdn6#)~uzE~sn(9KR+RR7gCddwk(3i~3?w>D7Gl5KX{=MuJ_#wod|umOgj~7@erK)0_|69+ zIA$&#_alt{GE?9EMcJLt`cMx7TdIOkq6}x0x)2SvyPT2ZQ>PsEG8h!HarZ4XiTxMb zSHY(JD3M5U{J^wXB$4p@O@)xjC)3Ho=47`kz{ycBe*9;p$=a)HdKEW-x+g1e%FoUZ z>Fo`WJ$e+VeAf(s07{Ky=gN>P6rTaKj(LO5$*x~73PVPjZYqqh`ztBlDw{3LI@>9n zsG8ADbhQQoKZddfN3@+tW%_BR-F{8zm8%$o+r*g$+Y9WRzbg2LlW+HpM}*?7Teh?^ zDoAkfICkzfT+IfOGL_Y1d^^O1teAd9w$cX@eujWZTD*XUQrgn<8c@J0pTCE|=rRqw+oCK5P&xC~bibi? zL^w5Peg|{Ansx517jvrJcN;5MUB775c!qP+!7!~+Y=OrMa4rY1_119Wk0v=tM8p|O zRP^Dv`HTsoDE#&fqR1`m{OT;jROxmb_GD;&k`Z=Vl&;5CVL?1Pe^{Cz7{RWn7E4iM z`i%|LQmu$}+SEW8QRZPdqn7W^81F@Nzs#qHiFeo8s({wW>r8N$$0yZzi>3lOK-~H# z`IZuZ-qa!d5+urlGGGCK_|4V5TIdUgCyHC)gI#v}{sG_yK}~>jHOgch5U@L(NY7GV z`ePHQKu=>*cT=^EnCN%S%TNQhk_Gd&6b(FCn>*MEI~T7r0O6BUZA3jk05nTQnURF~ z#;nD{?$+U~{ND#&tfHBKv`fF@jsRNHpq(103bSqNO1%6y&pJ&a%e36}VII2Nid9mP z*3e`BwRWP>@RxNeMj!WeYb|T|!B}Hv3NNyXVeIfKYb4uZ)P-$k9@|2y9siNOWd~W3 zJ4~jJh46c@#(weWO{&kP@+UCX5|BD>Mg%QSkC~%WyV|uKi?X$*ksy+eBxHVTp|9dd zL&^97hk0nRiw{}>L-|dcZG_ZzzO9m zGXBRhD_qF}G3xU^PQChL`oveb{doJea%eI+FYVy6`N;CX`?}fF_w47^B~tJ~XS*~y zZX+4IPX#VunlsghKgZ0&_7HCe_DHVn%;>qvZLwcP;)ABX9&{@DJX<~0;_L)`g}`jA z_>f#JWq!Im&LnJgf~^{9MxR0uj~-ZJj&)3sg{_!#m>Zlwx=Ya}IKBZyR6IAgQ;1K& z&tK`OwvVs<$kW+Jg$M$g7pmXp`DiTrWf>0owayN;I2p6GPpzM$d29r=b2rBwbgq3M zULCw%y7?J<;A>2+Xgt-`b>|=*k`uJnD%iSI3s>(P`AW}nb*ArnyoyE{G0;0WlK2ml zhLb7cRfl5j+uWR!18DdqA|oRuN>y^Es|EiI{7StpR;pxr>s zHm=JOgbk3Ez*Xw~j++iuuq!4F`)R-=pg}ksd=O;Wc(+86^stvO$KqHQs3&2=5Rb`@ z8qx4A{uGXf0;7Qtg(n$PEoOI?T2DtE;E^c6?c|al+*f=ugNZ;O1{G#7lI}KC_wIaV zMcOHb(b#k?IfkupHKhaH;IO1mGbLanEGcqh(LR`aJ9DjuSVYC>B(MdcNsn`2)m!QI zQ2j(b=^43Mx{UY-6HchLKLaolFDSri6 ze=0^hJ9AH)QrKA2MH+YU2iPF{Ee8p6op1A+pW39)9bdKgkZYik*oqTA5#BB@&Fw`X zFaQ+Ds1={SW2`4(Rxl?N-B&l%^oMRsHXX2-HXDP`{d%xS*#O;WEH+WvV4@s(6ky^R z%B%~PX~p*5VI}vyU9ygj8zNY(_F&w*0>`>E4UJAKHfG=z56Bk&GZ!MRdMe?KICdddc%c00Cx#r)*-ENTS zNHZL(M6x&BX*T7A8ayzD`X0V?AeaJk#VybkI6b$78fs__(|u8{--pv;;qYpf|Rt?xsWFzObFe zy!U$fj^nm0g7Iueo{W0_()C~f`9E@GAs5AnZ{O5i&U$c3#F?oGw>K<0wgYvWHv87D zvs<2ZT^uoGpIj-bmj5#Pqi2SQXL+I|Q19N&%IxR-CA+wL2wd4E)YpZf9tf-WY6Xz9 z#p-Wn(EN@u$)_X)*#d5!coEJ~CQujgdz{2**j|q?I7vNQ3wvPo_Zi5^q{QJ)=x|(WM(F; zEDqzzmz=4{rP}zT=-n#PfJ5CB|x+bL}hR`xkGrq;ciJAdouhkeoAD|8r;V`^3e zS=mO(Q#l8aub%DwryO96V&vch^}LU!Tr-fLsTlz9yf=gt$B?S)^)bhe0|1?M)@2_R z0|Qc3P&Ndf1R2d|4q*DD5%UZ;qs+28?v19>7fn`oJwpH;=R_o|U?;!@vR!NaEy4)K z#~>G3j6}s700WxRy~Mh8d<yyf%U~3)RmA-K|^dfS2_sahiPb~)dhkxNhoyTxng9Z;IA@Tq!ZSfb|yB;h! zFxjk~BR~SA9ya>=i=xJwp${cAY>7O~^`#IiVJcmOjsPTLFN2!}AT_S@No;>n|BlFBdcT&1GDu16_oF39fog%7S|Pd{r^gCL>#Hs(5L_f}WI|^eF9} zbJZ&TwM6`1RJ~l64H&*4FZw_h;%p7`bZ<)-9srIk^|D+ zuy6kFe)n%CT_pC}RnZgZ;`7tLnt%VknBB>Gztr{YpmPKGB>IgMSAVsJebWWEOZ zO~KEVSLw!<10-uP6#N6iTvtG=3ge7?eM`M1H6;v~9}xWyYT)S##5q~N{*%Q=Z^TOm z_KrfjT)SVOWiR1@-i3Mx{R4o!`Sz}E{eQPa$upoQxN2bT{%92^G9WM;rGIy*d@rl(|`C{V!Sisawuos;A`|* zs7yBNmlbFI4MvwHvsE_GX7sb+G-S*tq|5&%^^JqRvCNOJ9?KKh-{BEmI!X?TWC#YI zI4sRZPzs+ZDbWr?P;^j+Cc9KFdsWQzo?x^hsDMA+;WK(ij|(xZ%!)<6MpaRhAC?JF z=s~Xtc3FnTlF73eY81!!lj~6UhWk-7z%8)Rog#o$y8aR|BKY5#(oXrvjDdmKm>2=c z1AM|rQB<7g^F_L(x>D&E*xM_EGn-b3+B@m)(M>8zuTswP_^jDE!L#&8eRR|KjLD6}IyUMv(E90d zd;7T^un~lRCEPhJ!=+1YK${LR8`L%RHJvyRep~9XUhlawoe_Wu_Ingc+w=toV}Ys5ctp}ob3<51l^XiXWiR;Kx8UT$@(N;>rMN27)=-NOnBjq zi%3w5ZIMc$h6(q95nG?Dkl?=mfP0gLf=1o)mv{9-dVm!1&yCTa)MzgYKT`yRvx%#p?+<}DR4!mV*#KsheJ`1| zQtp;rMzw6g@tE{}U_C&M^@DelZRF({7LMx6K32lXi7V5E=)dnpW5^<&yIKMcvtwd7 zj$7u)U1!^)H8!K9Tie?Ob`yEF;#jBvTsU@q-e88OP{8%9&Z$b2l<%55?ZdEAH7lim zDX*HVX>>Vrz6Z>zO6=-=(I&k=d`di_uuFVRB@67rI6V2PD&zB1w`W(%p6AI@)*i$ph{2o)JR5rdpYd ziNreKF(y zkc-PS=C=$B*GcQ##vCVSt1cE!f>YAs%l(qEkfXAWY>=`&AsoVWV45o>Cdj4#Q3t}v zXhBeKQU1qp&RnHlCe?y8E7gwna!$%$=SsO4K7Hi16Y`$@+_q)|dwcs=ME#%k^l&bt zSoD{n$Dej71b13mZ%Mp5#K^oMK2Vi|@KL|j6W@`2$j-P&cBca65`#GTpYp~Zk{-Y@ zF~Hw_ow2cY&AP>bYbund`$2~DY(`=Y=IQCl$xu3$w0{os{adRKNPt-4JZRJXdI#!u({T-dV2+uxZV-Q-;+g;{M29im>p`&)xo73gPv!RWo~ zUB!A0$G$eiLz(c{dGd?rI(8-Ea_}KqjBHSdXk)M2g1c_krUzpE{RH(IxVcbu+i!up zUZ=swe92M{95o8yx%;Mq*WJ)5=zs5-hv2JA6iIOMU61FB9q!y8T|c7lw=nZnv)-Rj z6lN@~J>8me#5Y~Ly5&2tJbi9Ey?hY*-Et3r)Cpd$5fq(3NcU@D1zX5et&OT~kUy~5 zld%Ut-bntp>~oDS>I!6bZA+DS*cwx&MBaCked2XmZ0c(JRA%*I205!C++c zdcgd=8-NL*+CBy>J?W)+xt=*i4{bk>E=%lKQ#MT@%Ujazg`imv9>eEcp-Ajn?fc7I$|3Z;yc{u zo&(icgzamZ03C``#@FM~clN%&f;#!nc?vw}&lhnR)EMZQ@Mr^nsmn&z!sERNt-m8i z;5ql)&A1w+mjgS4zk_x?2cu~sD%I801mZ$pU@&#RKJ89#Yi$P-S*@dY0YRu};MZPo zVaNA8{fQ+eAC#)@_1n%@uNr8=yeu8Q!BEFo%t7<)SZ-NvW(WiSDsrG-L2hirEG}yF zsg0c{prv>MuHS+ajzV!n9k00W&F`^$v0CB#Jq=_H-Ev-@)JT!!AvUIUiAu5LQIFm2 z34~MV(3(ISKTY};VuFQ5Q#w=q*sHfO9O)89jGnACk8xkq_w}XOaL!(#eR6zMREUDV zUKXL&w)`-edh(U!(v>Hb7lYD>L8=zGIS9dSVvz@9##q@-u z8{)IuFh;P#?z)H{R!zW{$Oow1<=!{p%>ZTM987%S-Fd<_bt|a(3`*> zJFB!Ai>&L2Lc=^+^5UcUmR;x<+ z^F9glu1FvK%D!}eRRuoaSmN8#ijd4T&_wdK8?9e3RGj?na=USxd`mrkoll07Yocvo zEPr~tJbHeJEpfe?0w}ct@0ziWiTgtWxITQ8W((MPdfu^m4yZ~UtkC*FyEB#<2JCK3 zc#hfR?gERB`5A0 zd-9!DcokVMQeNJSj{u++H(-7IQmqKdM2M9!8JiqGa_6QJXi_J%9?cEw!jS`-I1L)}#*=p8Ty-55me_^>&l>s{J$iV*eT~Q|C z8*%}uLM_)xdJckO*8~XiPHs`rHvhmez;243{zlG7hX$F$rIVmWMMY&vIjozY#kzj` z=j6DR)3}L;zJMMYc<=z{21mw2UTF3&CxnRns$oHj1LVlyXA;4H_caYw8o?y`YEs^n zHUWHV57~nbzV8i9D>@WZzwZiX$HQRvVQ)H==nYvhfi8kIK7#J2cmXCCK3DxAr+_sZ5$aLx{iu3FsMJ$no`b7!j41t*%X~)< zLI3K#9ou$wsnn*Rdv42c^Yt4iepK~4dZu#z>Chzx;JQIV7@;gN7loXAMQiI%XpE>$ro;)$@|stu8?Lq25SJgS^D0icnF|wMVAH;k-S~#u zQDDc9^W)#c3TT!ll92+)H+G$v67l;*D;?MGa*bC{QS8cvubQib+cfiIo8~aC+tvdi z#%u}R{UNJu!hhfAa}|HZbM_oaS?pS@cZkiVVT$MD1FsyU^lrWg%!-wh$8Lbzj4M)kK~CjdVl7E{z+|^91Z=~ZNqbQwcu}K0#2A-$YbnEOG?~h%UowF2F&ld zE9uclFN#xf1_mewZar8yW7zO7Vt)__BOi?9ijq1!i&ke9EU(1EL>B7n3R%=+Nn;tzzb$9&)Q@(kuNuyov4`pUlbyUnH3 z$B!!wbOu+mI=rj4=tZ+DKU=7=isaYxI#^=0vH1UtANAowdV-FwWu1!+=vjnXl8PpD z_3q`Wg_uaW$i&vUq=Sfr+n#R=PU4&cV`}9X06}M}?(C8`0q_pVNcc6WV6-TzIISr- zP1H+$d_SJ`deT0GF#F6#5=2QFcKccRXl+=%UKR_lnQKX#ZNA7gJ}y`cK5*zHY);81 zKxK_fAsm0m1w#GQ(~$R9X<-%f4K*$6eUa?9axxHmI{Of_Q}pZH*gb#O$6B)SqHP(( zvH$m89PL|q!hmsx)2_1n39fnIE&MADUiixuW^78+*6@f9d@*KM#?Jqx$9M43GhY1+=}CI zLUN?KQ8z78-FlS84xkSMZYJ;-9Uc!Jc@#yI^4r&jY&H!qxO^S117M@z86+Gm>Lxzb0o zcgc+iF6~crfNS1AZYWfmjjj3%18A8CD^H6#BWG+NaG!E0^f^bQtE%&TH@!cmyPr)U zq$$6ehXZFzEBEEQ3ks4`P{;~9ke7628w_0g%n!YPbnh~)=PK7a2Wfn&bIMMU{Ymcif4ez0 zT2f>y{SBg%;7N=%YYE&FW8$`ifh6!Dwt(bF%bs74m|(x3{1@jC(_813D~iOwru2cp z7#vT@5PS1tuptq7v|P*eS%zNutO|s!LvG)L^31hg9{OcS?i@9ORFEr}7A11AfxLxD zXHtpR+<@3&qpJa=U~hS2mc>1J$lX$x;~=Cg4sdxC@Bv?{5>E_cw*oAV#y56s4WluL zY2MA7eN0iyf-5x9F2Y>r!TtO!T{X&Rw@}_f$L7>`?Tdu$T8mbSa7F&>>{)cQH?P6; zcgJxHFax*Md5NFlawqdU^<5_MmBDk8x}P>6tDb;kNjzrbb9?wBDVWH(<#%cQ*@(K@ zueh&jnb3ZdW+i+@0!1c-XyiflA3<4_3AkS^6Y^#&4JyXfa zzs2i8{8XfJ02V#}4juUaDW7m-t?RaQrF^=Ii_H&~$hdkjm{}u(D;LXMF2iS*S=1#Y zIS=@Mo#crx&ywENQT&jX5KVn&xDk!Qw4K~n@bU&tLG6|OKa@*x>A`q9mw}&HdSfOY z<8NRoxFQcd*xbUp+hGOKbb%Mzc37q9u41i!{JXUJ^Bz=6J_n!78^0isVrU9A^EXK5 zJ?4q&5q52dOvkwbDOiWFHMJRgW%XZ)r5Rs268v8bqfTtK%n)D-geK{46To?yxmWJl z7OjlXOXdFGb^?Bz15tohc^uZ0p3*v`wOQK@Yn z{KL9~_>Dgnv=mY1CtA=1_nc`^z@)|Y36-Iq3dLdO+P>z`G!D@gkiRHR&d7HEA~a9@ z3dOi-fuVqPNjSyT4uG4yYv(9yq}Bl+nbkvETXUiI5{u>G;`($98X`#au9r>E8MAEoD{1DMm(!9Nsd%iOvknD$>i0@>jYPT5x zPfIz0MI{{tRWN0VSUr8Y!rD_OWldk!f7b(F75ycge=L=Mzfb&hp!P>yCHz3|V>mW< z8v)ShuAnH(UIE@U0Cu%G2N<)ksHg*Lcj|+CK(JtC%#PJ_;?vrHQZK406g=z-&dF)Q zg!a{>9Y!UrQR%%TWlI;VD?zJ}nfY1OxR|S(9}YI{Yt~JsrQ4e@Gso}cb8=V}xm-z} z1t8%{Y|@n2GrL4>uF z6L&gTH{^I3(4YvkO0`7#Yel%H7vW0-8EG2;D3rNfFabqv-X3C0`svBfs3qrCT(_7n z$F2aMA_aqr8UyG{cLtG%9+Ilp?&del5s(z|7#)mALzslINcL z_p%4zHn<&=3Ij6H+`z|j3il$H2xy|X?$b?CYV4EhVLs1j&m}?bt}q7V?f-8R+cE=J zV53b<`9ox%_o}!icyj4ny!zH_xOyBB=)v5FXMfKeRjo(r#^pvfT1!2pa}+kO)O54h z;M_+V(&|xQW(%4l#(FsODBtDg&`x?7cN{kJ$U>6CA_nUTSOoS~Xe0IBPLm`YvUwsq z{(hTQQ^tNiA|V&*iBNnJUl2wicpIfHIQ7W$Im$XJ9>vms!8`3co|HA!0(XCi`oUU} zuDRbk1(qrkl7Y5g1A zM1j{|39JW61p7o_vLGQI>Up(Mw3NU9)F>~(1Y{kk1n1?(yBeGr=G#Syf*uILnbTTz zu|D|HS4hS7@eByxq4rZfq&KZzMRkNcuJpiwcl%x1iOZDK4;xhu(92gLMan{ck;892+v{dBk+^=lS0dtsNcuO*;n|*HPw=GDh z{Z&1-eB9(l;&JaPABxjUG_Llw@VAG`{p**jJ4sOjBcC$#Z6&JqZ+TRs*TKgDb9-@z zr~*jeZen}HAiLVa>DeEb2zL|GJ+!6~mG3PWiFjKG;1u1|sNQ*N1QhVk&G=cwlOgd_ zU(`hV(CBU2aaUAlkRXV(DQ&r>1J%^;M<1MLFS*~ICC2>T`oh6woncWo!DRic zZVl<7mF$P^9INt8#GW6VG^zb&*;yY26TBUPHlCA|9lQ34t?qq^%@=VX8T80~cD|<3 zNk6hE!Nj=rae4m)FhC@mDp!V^1l+^<-w6WHcfjutdkNg9zi&RwpzShLz5D3!=?bOAyZ#Uj*^-@9aDEO_ z0==nZ-<1LIcU??xPnv@k{Eqe^AVZEfgOX3D`2TDA*1=9|N4fZcb1>cNm$HAnoTdBj+Vs)R56ML~K$7$!DD5 zt(Sn$`3|68Mfj<&D?H@dYoVlpH~93(AnZy%B#in=Mo5zM&Ir&i`!p7I5i$WpV)Umo zvS6CbyUU+q(sV1BV2CvoNfz(rp|0ZB1lZ(GFXzlguZzDBV(2s}cALKgfQtGg9ydi= z2{XWY*BeHOf=k=rUjAO7mR(z2fGv+9;+K)41dOAmp>8j_@*UbAuFIE8)-1N^Z3MfuQ3zN9IfNV@ zWRnah1@)05E?ldjS=>Zq6hb|BJ>&bOo4Mt(1v%(~$~>WLn!@8eqo8+(klT00Q@|Kf z`Eo;b83cNDECgpYHeGg_=j~u1;a?c&ViZ+J*U`<*G_8&|AB@wZ_WiBhHFfsUh8__M z&98zybtv2G!&&TxPz7kVWZ0(<+l(RVXBNzKe$B^)g1{upT5|^jZ6L&6yh)PJ1$%p#Luy(uNI!?+1MMyAZ zC@l9dfdmPJ2b3cmh?Z6ydV7HCIH||Wp*6q32@nhbCGXqtWKD^M3xpDb_*5=Xd;9%S zD{OjX8)#1O9);w!tASxfQ7%B+%(f%jV7<@pfI>~W06RXtXP{O_O-oLRw*Z?1r+FM) zmZOv63wxl#o>%R%7>J>;&6C_ky6Hxz13@)+YSc4$5P@BFjs_PEj}DHl&AT>O3C;D^ zaPrqH)!nzdeq*wccyDH$w^`+(ZM56{6~-L3;l`Zk#IK{qiu zkC_X8)xqA^Ea|M+*v^CN*jjf-|V?@eks%x zBq9+pX4BqIxaZVZD8_HSXuE$o170lB1u~aHh+xluMXGC%Rta^)<;fDz$l4t05MnOP_luwk`mVq(hYRLD@AY$u%FOxJ&=C zl>J*n$B4eRq-Wio5}rTB85hlA7Mo^iN5* zAW{}uwR+c@SCN%@Xqzn)q)hfkgp1w74;K=A5vHhcHW$l@GThO7*du%^`Dhi z=Sm(AQZ`;D`lU>K7I1jb^ZnW+U(h4;Qra2)BY-PWksauZQmcc0e7|>m|A%Gqvdkr7 zZ+H-imqXcJdrb+eigQ^|oCvBIBO|?Rz7<(7W2(o8{&k|LF_Ji0cTl*)(vb>KR>M9kUU{nYp zpnH(=V2oe~hMU?cy%v+-q!qfnL9O-J&O=fE1(W!xZXp|+x>5CefXH>|cR7r!gp4Ei z`5dcQ*8&H{gHZrx?2 z<-jwSmKU`upnI|dXc*Fo^eYR$m9>6wSlier=LO<++-zVIYeZ=vdM;YG0DEU&BS0{q zg1%(Qfly(I1@QwP#$}kyH+nN7j@Py3(4ZgH>mC#keVqoT9sVOf(4lgnzbPp+pw%%K zl1|pV<@Z+26_THOW@ChPr89>(jFY<`KJn&5NihkAvJrH*E!IA%lVY6$r=k_4^xjB-La7_e?R%B>2&ZBBsC$k7`r6%GkV#JfldQLApU?UyHz&BwiV4~f6v*AYRBQfzv z45vMD$xi4`iaF6K1#4}I=ql|-W2=*!l<5qGZ+L}!8OH~?>@Rhdrf(UL4NF+lVtzrf z`~>jsYw?;GDW+Z>|CG2#wGw z6KnAW{kMwrwH3SuJR34hC@EfQO6E?}s8PxZ?c?;8eT;38R`AMDxF; zYpx30Hu)pJnw$q78Cb9>rXo$Gcvr@P(HF?A#YRxf0xd-7fx_xYjxb91jgW3<2gKzb^5Fc z8u9&XuM3VWQ{g%m@@su?L=EoN5w^pta4V;A&pLm;8P-#({sxN%UD3Z>dj7*b|7R!e z)j{&Oy)&{eF~~wX4@N6ydqrQYmGEVg>tX5O4Nwp^ zga)^xE7Q~DM*mB(8Hz@YMlswgWRZHQJ#8j6Bve4(0fsS5Z4Ho(zqnI=`2KWcODhVF zXB5YJHt2@Gj$s1LXfpo&Kp<2(c+wTpYgu#pEm*2uX|&!CGW2e`DK9z zp%+vPj8bjs3m8dvu50Y482Mp-!^6Wnk6n4&YSeZc57eaOz~_b}b0!M%%1q$~Phho+ zt=t;0I_2E5p;*Ps(fKj!hA=FV!zChGOQJXzST;xCELgYHyk?2tG>Q71K$mW!#O$;C zF64w}5i4mjf}0JBWTNo}`(6416EtO;D=ebLz6q6*1A7z5NBQT?L7Q5GmqHL(<;wFs zoG|_ddPWyAO0olmqH-s1@y$h6+1Veu!h;?sZ5m>9vlTNR67Gfet2`n=dv3eA!ufc? zu!Q?Ehk|AWce7{{CJLcz4Oqxr&K#dnC0G_W+hN>+D`*Mhe(NcNd2R;dVXJBE$=d%M zXQ-a%zRsU3v`{1(&bDPmg@TLKleO4@(Q>YBujpoKO=Z9;0Xds00Jd6l&5iGOVTug2 zzyKv0#A`cJQ`^I;(ojO5U=bKEtjG6DRy`Zd(tr5IHIw^1P8cnUiY-X?iKTH}Mq zpj2G$iMIbUOK6f$a&fs!A*=ilZ=Z0eln8z_lE_3jN2##ZtagZbZb*-q8DnzfT!)cN zHGJ>ron9eF2DjK;!O29wS%w7!gYP$6bL*0!?e6fBDE#-&>fWu<-!@I|3mzDETWa|~ zmc|_1mUHf1fkk~)ub7`IPqU}Yz?+%^lVfVP=AK~L#pYBA!aDNU{6uk%1y;~Qk+I8- z(gR}g7$&p&`jN@z`xt)5pSt&N{$@3f28u7RbTxzs+%iv!BsgI&ZPhirlgNz`cuQG+ zZD)k^b@t%*F?Iug6D*{okodIG?Y(NVujz@{O^ol8pfJp@1D1I28f2IED-4hP{KAe|Qr`53rjem$22< zqTJmCeL%7i35s~IpS;Sx=Yxgm1mmSr8-JaWf{$AJ=H%L%NL7Mr7kU%g?;6@UM4;+B zt}id;f8DRY>s}}}N|H{`RiUSGqmwFAd^|D+qdcOFWg?H|<3Bu!Ry*gCAWV}?5|j4r zJt>~j_9A%IRC3wk=w&wb5~_Kepbx||PxGVC7eq>1wX zCaIue8#1^X_>i*`z*U7I#&W#5E#XHiMn^W`wan9Jv#wYDf{<|tkBDGoXCJwsmE0>% z{vQ?^s4jR5T5dZw081ZTAArD$e2JPo+OJdqmmShG-1^WiYL3OirdVx9Hfw&QjzGP( z^k#H<3#;a_bgjwXLZZ1#3pKg9usv(V$oAsft`5I5h9{Ewm*Jn(hNH6QRFUecn)#g2 z?CbcrKtP|sC=uT&Jw+orp}VWd+7Z~fpmE}Pyt!rxsWu=oIcMfQgGsl*l~qc^Xp;}J zH9m`JGDV$>MI!m*vK1FiPLR-&LQ!lj=a+LaU>XAJ5xv;>jLh7a(gQVGI%&q-n?3Mo zg7(Tp!qwpbrj{{b=4pN;HCPe7E^KC*IQr5zh8nK5ll{TY@5`eoQWJgjP0w~OA%pDI zYD`j}N)C(Oq@7L<$+WxxI^&1q-$eFm7IcT-tRZp@7WSqS19owBfgu(?;0K3AvM4L^ zzadFz3s+G$77gm5G99Ev!D)jk5YuT)sh`QrxorC})p(yDvX9mn#9&O!$Mcse-!}bf z^9!ULH4W2KC23xHC3P|cRn4H*!q%)MYjfbrYzLb)+HK@Sws0_R#G4Oitrxv_zoM}} zk9@7-=>^sKVqpTR7DkwhQEklIxn4Z8l^+2bw_^IyetGjZnS>15VsGCIPyZKOVAh^X zqwIX0%9as8hqanz-isL&P)dT)z~vVt^O?q`F9bf!u@9Jdwy3>Zsy?C#9d5*sMD`e% zYGO^JO(aL`OSws$Q3~4>nZDVfem5{cRTCEgHCa-leeKn^;f_wXgGU zfLk`3xCzOF>;T>5KmmRs8^U#Hp|JjMA7Dqrb{Ut7n$7F9nHx7xvOnO6UrtdksIYaZ zB_`Q8Qjxa1K=mz0zNNYhuTWL}{+$ZzpJ;;Jb2EdT86zKaRw?hZm5xBs#YPtwlu&h~ z4{vq-jo!Cu8K2ljRE~ zWgYvt#Q0b6x}-31&Sn~pz%eeHpob`-LTHfwu(Gmrdy#d3p*h^!g0py(ec5?Js4Fhs zya!Z0-TM{==y3j~HK^L>_#^MT?1@*rNhgcIFUBJuq=UvI_zXhiGv>kG|GHg2reujj zGv>$Hc07+rOcbmUwptxYH;NA=c~$BALvpufrrt>)AM6c{;3yy$^CYBF2NtU-!aL$< zuqWI6;vBQXzJCvxxvyfRNoC_KW=qrv1$?9Dc@|)d3}z{1)(FttlO=A*Jmy$dc=((*Uq@Pe8TQ`gA| zLm=8!pf1G z%E&%*d@LtS;ngb7-5) z8P59yUvr*vtk{XpwEG%IMaTdm$OPfBTShW7JpZ3D>q0*3j>{?Jyk+E!5iVz{pI0>| z^aDZK)&~A}R~Q~KZL`r7AK?*Z7?0G!25LTe6P?VEd$aNCP2(3C@#OckDv1>p(g<@U z`zBvHdq+|f^azynC9$5!ZKC2nvn(d?3&1`*7xv~eb$u-**CaAQ0JC}GqKmCR?RCu= z#6e{wqi?b!avalgmJ{ z;Fi$V80*{nT_jK2>%VApe~pv9(jEN9if%Gb0ze(OM5R}Oq&HgjsTXajL&~mO%6vsU z7Ly$dYV(5g_`Oyu*)%@$SS4_OGDtBMmQz!^e*RZ=Xdj4vJs){|fW~G2{_S!6Rk|h; zVwHw!Lgjb#)bwL?4`#4%vbeHrQ_fF94!TK$-&WyQGaYX9e^rOuVxhFw@@jq{_?>Rg z2F$KfYR$J%I6ktWU6g<4QNbLx2O0jYtEO|oKIII^Uj3A0P;JYQ3x%62igb_QT;(la z(N@FJjSPFoAlDO*Q_|H%AU*r_37I$8n>5_N+8+$G6It(vUnteka1-r-TX2Gb~!6m9rE*m~OaWVp>;MDg$KoN8GY6x)>i?lx> z7%$E7cf%=QT^^@Jf!kHobpTz~1qgboBhxxmaust?Qd=-q_%PR$V=3S8wGSCp&&3lk zQ20P_K9vp)DQ${<2gJQyo|Nf561m6j#V48}d2SU&2+oltL)vVETE9T6+8Kk}3T(Tm z``O@PtKZ};Cx?%Q5-}dk<8>NA1E($w zY%BHDO|{y=A+K1f0=4sb^`<3jBgt{RlRO3&6UBB$?Y!%!FEd4xZ&jLPVMS6^iOem- z+TmmjUl`oL17FknG_F8f{Uc`swH-{=n>$?HeBDmXT=I^fg-q9G(7dwy%7G=Mi0411 zTB>?!TP24rd2cx_MDT`@^O@h*&W=q@~m5NYH&l|iGF?Wl`E$R_(b=crHGU^hxo}daECbULTTDj0 zjD(0w9#ed8_;ITLh99n2!QYz8{9vwhs*=+f zlQqgn!nqu9$lm9km};?+e`kvk7>Sqa6lb+?zX@!*wf#HB@AL@yn*Lj69A%%#Rl>>o&caRQ8^^vtSc;3UB<(KmYUUX} z^h^F|$K8M@TN({^_?g(VS8nIiWfe;_T7lFFTg1E$CC+8Zi>NtThw|Xd&7 zFjPHK%Y~&_248SxQSWNj8$He;8CRzsT)Hti-h6H``A|0*t4o7LF-CcG%3_ z@K{quYjU|!Y~0u97D041$3;%ud(u72o51y1fTBQ`u5sJDWBIoHhDYhiHo@2fFBg?0 zd~m>9Jwha;6j(Dom^;mybk=WrYL;$Y`nl z9#5H_A@9EjCl&d|d9{F+Y(^VLn;xfVT++S9`8T>=CkYycn2$g>wu+Pz_V1&SdI=hr zH5U~!wAxtYsy0{cP$C?x&1&h6;_ubdAMGYf;_(9GM1eL?d8Z&_2Oq4L)6TumMqz01 z(+&oQ6n}v$nBBp0zc_YM4eOy&c-;RvM{(Wod-=qj%iIUroYTU9J`^h1ud z@dfp@7}l^_#``DSRtM~B?!mr0ALmge(U=aah?ShvBPmaxEd@{cPp;_2-)L%kAWV`i zk(~QsJ-e&5LAO-h?lI7V;t3Tm^f>gb@OH zj!oBj@%+?E!(*nNpUc~=t#D!A9zPbk1XDP(D7pX#C6TbN21-H?a}6+ zl@T@{Qk9HH-3_K;x6~vB~ zJ?@bZt}y2`jmnA8M>F}V%8kmEFo#crvT!Pqlb`xAT&6r<;i;FJKmTz$=Z`zHQMVwf zS@8&KuEIMITCF-gyNYT5nX;{0rgjvPZ?-;V8(AB^8tB09S;5y#I9}#WrEG|y{sIeu z1Z6~ZJ5F6ogk2R`#9)bjjz95gOmlovw?P*aj>oU0QJ9icq?7)weH+Nds@;^#Xo%wKmTAaeAL&fK9YOb3%rZ?*Qo~nK_9U_{ z!7zCgz8JXZNuj2}-gd?xn#Srf_BrNl0s_yDAQ}|1kp{E`u6NRKTV%S)xjIE?UYT zGVMA%KHh(KxnJY9sQ}p43S8DY27VL@N>c?i9P0=uG61=5@2qpMtyM^xrBKVC8Xnc3 zK`j%!V@~g&-%C8`o*S~+5H6_gEOIuD#Z^Et-OBW9o0f?OuzPkarhy5v?Gd^R*^ zO%%Gn8D+S}hhCg?lftbmi{pJt z?eNf0f}L0D!;^O*z;%6*hq_g0Q*wb=92aU)9cIgUD-|9-H*qPd^lk^Y&8KslIiKYK z9X);q-hWxGe*n^%os9p%TBliu@@Xmj2$MDEFXy{|ln;6~Ha6=(WU69LyqY3V_;wHMi;C_95-V_T8fZ6S%d@#2#iJuFG_PZtU9 zQh#HOB9Z%}= zhyU2$MsHj2#(=v=x5kij_Ti~p_J4sll@E5xYG+{5Eos7vew6n(96#XBdbD278uD$4 zxKha5*ghXU3BzvHj>L{SrSL*Xs46O|6Rmuz;>QGYw{2)3h_!i) zeH21V%6vhQn>i`lZCniQ_+4i|>nfz5K6flLwJ=QoNhnm=$O&T5wct5B3rqOrA~{=W z68Y-_huQNe6pu!$j#!>ZWhPk4QU0wqATJ=Y!SFie0nhiSF0JTRl+2mbdNlTMg|IdtuQ+~S< zcZ~zBF>*h@v!0IZZZ%BG>PN>n-~=&iU1GSzgEWKbpfYch%aKJGexc2zhk~ zX^_Df)>P5{5-}`tvWTh=KG={>Td+?``l8KFyiZwv>%0_*i$Y1&vwLHvHbOi8tsAuMG-wr%zry`9m0B(+F zyMY(h_GX^{5$~o-??%ghWpCYt#8-gsg9PxE3RBQW%&-iq%|m|BZ6$bxy>M#U=C=Dy zW9^+yw8aWk!YxSW_mkIypKso|+k4mqeA$8sbw{BG5>72&Jy}i2UnN*?YksKqVKG4l z?^?;@ad^&K)gJCi4W2u4%b1*tB+O& z>wf3KbEe3&fuNg-py0ik4~@71%-fmQtH`Uq$i7D+t;T0s3Z9M1rWKqh`huXRn%dx0 zw1ZT|bw=zJPk!Tuu@=N9_Np7hx?MSxv3!?wksZ2M=6+n(pyct$z%?$?} z-rb(17QIVfSo5L)x3{orY6}XLmZ6C1|7LL2L>ajr!q%2ya{*wjN#$URDMO2g5ss?A zfGkD8B5M9dijV{-KjeSd#%RU2lO;RlR$*edEkRDFQn_ttV!qO)*LdUM=z9Pnp;tfVhd!E0a{jFOPq{;N7m8REq& zC5lf?+q@4>{4JZjxXA+4_Mc;L6uaaaFiWB)0DCde5sTy3i3`E=^Ne$V9X3lYxaca^ zocR=PAHPZKTxzOfHsr2t%u(dd?(rXz*a{$s?dj3tu|$Fr32w;p)SfdYQ2jDae-M!H(eoE&rCLH0&Hwe8I&0CLI+2_+h+SvYsNzIwNHoq&%#x1R-F#9WW;b zhlb*?L<{KXge7K@gtmn$xKo3OMkb@!WVN)&IR}Zt-( za)y1<8J_0(6tTomf3)QBwm4RL=@qD?)G)xcj^8^fT;Yrc93~~(GRquLRm&u?|I<)& zyien~{>L#%C*p(NNo%)D823DA+=Zr-;=xo)bK(lUfu`D{!-#I;DgyEBhb<7{@Q_8cFSP_6w}7O+~fDI$j2Z?^IiFTFnu8JW5Hh+yf!AG)zL||yuYPeG>^iTlMxfXVo0Cq8XslEB z2XX$Y(+S<51Sf8E@BEGz3fcQJNE__A{u67k@4xWD{T9ZSFEgo*20;3(kZpTsh$+4g zP?<$(zum$nGj>IM>NUku$?X9$u;ZD2Iw=qYXox3;c?nN$_TK!Ufa_nWix=jv<)n7P z#B~o8`+sm{IiH*VBDmxkXrXwtfUNdS-UP+q&+F!*clag;L={@`Uaf!`d5feAG2+nX zZbhT4JY~C<)9ZDD0=y%(cASOYLo~skfNjj{YA?jClOyHXBN%BGmvza40a&Q|oWRMi z;a6~mY~3xXf(7G6kS&+*Lyk34{s=zLo=5Ud%+5pJk6O4DD`g^>o5MQD*g#ix5BDBN zQ-bK|1hwMn&a5VVT3`@OI(HF>PI)}PKCTIZz@ardg_+`M_E^~9S_h6^LB5suNsd(N z^&!Or?qi9kY&Z@PZ3(tR($uwBS@cXCL?1pdC@Ev|d5PX9JJMjG`uzAHJp{v5VA$`C zZGx4`C@Fsq+<+w1hC}w`KAM2jgf!-rYG}1@jhEj0 z+P+c}_a%eKx@Rc@_p>bsRiq^Au&=W;4<0p6&z-uoL31gupkrKrnv zrS^~N3F_2#?`?>$$69?fV8uZVzDW_zB#0}>uh1Ci;UsA%y|7sroGrC?}<<$&JVR6YSAhN0f z%bHSuz5#!D)uCj(3kb~ScSx|s-@iL4>zk;&y%@0@40xZr_b;4LvO%$WJj{tGK=J%> z+y#5>@qwi}F!cC?*<<*;h7>K|Jx z-_qnUc1KH~*RS7V`f0fiq(T5qFc7>NeJeGsy}p&KDOix6qGN6!mBZZZ3oJQLXqMuD zA9;Qu_u}%)p>PPZAy`+bU(FV)ev6+(*$!Rei;;helr)7FRIC!g#teG>o;rklX1gRN zJ&bCnU=P6$`~TtStRI?g+bFIyNQ~|nozmS63R2P`0|}+OL0~iz(lJnwk_PGS1_>G6 zFgk|B$al~C2W-FH-}}DKb~;lIR2*-um;6ZtCUtCT zB~{zuyg)u-dLZb&*Y7_@5k|6mqlc2rxIO*~rk$+?pW#S+<5Q%frZ(jg$api+gS21# z{8?O_4GW!$h^+s#P-Dp4*LjkX!Js&Leor`%0h#&!!-2Yse%#~#eQ!2~lLqAc{QUp< zLN$k}BL5a!lXt~g55~LAYk!tRwgU<#IDkMPR9^uq77+YeK}bt=+Mx=G!PL}o3t|Bzw}knpeTAnB3mb&3x4&id(R#)(Ldoov?m zDzZ7GFaPNu8?nqt$}jH=?L*nM-FNfqmwtA1;fPo<4u(r{SpA?^oU8}kvP7ZTA(zhg z0nJyG@cHBLksfIb@59MAN}xM$l49p?zh5Q+bw!s_S+@^NcNu)%EPZ8&T%CAQ)Fslu z0&O#7VuFPgL4~$_kgl&UCQfCeH~kT!`xt75ayZzUs%|5C^$D4iAz#g4T}F#qdiA_V zPd;cGBw51-0&7GOtaGmHH~#y#lsKZ99Ty+lqDA;q=Sh7^<0 zr;lALA-*1H5o7DyZhSwhpdI|s3bl6fgkDP=?A_b&K0v<*&*AYTNqleE#T^cVRBeNi<~p^gWZLp&GjadC zUCj_vK6|vjejVYc3FTd9u)X>KQ`?(W2n6DHEjqP*mxk@bOI#(x;eOLNoc@}}J!qWK z@NBxt@`Zi!M2_Lksh^R-7esD{AaVM{)Z1@OGpi^*BgLBh_fL{Hld4HD=Ybt{+mscj ztY|=jJVae*kom^NdZM*cp*SH2&0YgAMQl$cLMFs##C#1|l9f^Ohg42Vs&-ZM16%t9 z2hP7@O)fa9mO7SVaqD_CUTnWhJR{e<$HXmOJG_>MTCo`{Q&-9drGtp{cbZErmc)$NE*rjLTj-GOuz7Ynh;-v! zc#tv#=$7W|FVh_drwBL!O6kIp=SqT!CQShHvE zi6(5c{dPk$pPo1rz5EvTrM(WafiNY`UXGW&a-CW=xmTIqi`C8R7FqcN|NQxbZ_FAU zv(C75c&*C0%l+Lpf7x6g9yg3V20|kPK8^;Ik3E8m>djj6jmkzJHx`+W%ayC9lLqtK z=reMOZRGv^#>d7kAIa~o@XS%a%9k$`>Hg68k6`SL%(=G0WY$%M{gP`w6{vVDJ2fHd z`Yf-UnpQzk@8r@x^`3udh*_ZvV4s9w;+qNsnZ`#B|im?qp-Rcw+ zwAKt?y}pf>+yDK#%C6gkJ6UxCW0SywjkS4=C6eV3M?#krFH#FF%oiOqK zF<#~e-;KnT05UG0_&M{r$$@B@9)+sy=>X|5h z;_q4v%1Z{`W{bHM1R=|CWpFQ$+?wMXNm#DOalGK<`H9fAx+qEYE2ZapyQypX`GS20 z|7`})+JJMnD!zGd8N^fzKeS|#KpSGMmf-a(sQJmw`%=# zGTAA}#G_IS^XV$)GIZzHt!!6eG^~ves~hLl>R+y(ymF5l_y_nyKtgck2m%3~ zo%l`KV%+#yKa#ZB1|6ZC-ozOVGuW-yOwJz?A=d3(w#P~_YV-c#RI3~+?2_lAnN4E! zco+Q_pv~NFZhNtXn&Na6_`x{Q$bG9{!!SQ3q_`{H*!ZF~s(st9k`h2}Hat#$U8lHFIK=>ky6+@wC)rpL}+P4lVi4kV`Eaf`H@R zJPp^fyd9Rs(zm~>h8S|&IgC=*NHG=uahme7Gug-(SjXS=VF-S;%d?Z5x4wi>neVi9 zc+z4E>QYl*HWW&gJXAbXK2*Kin2T#K;kuKiIa54*gr%l)8#CcDOL5wy)KQ?nThKt{ zJ8fSQ=YP-<7k7FHx?U?dOL=w@yAp6%gg}CwFp07|vvqXMQ{S*3&Fxa2LXr-?1*o>R z3b(HVW@eu@r88VP#8??FxHogZ_6a_w`= z$&3xqG7v5$Kj_ltwTiQ{lte8B45Zq07?F-W(zD9iXiYu}A=asiue;d9*+!{GnF{nf ze{AGNxKu6ttbvrItNx=#N`bk&O-Rl0YFdM{&u8(!e5DLxhxi^o#@o^HQqtukl2>@s zih{w%`1v@#iMW<*!OrSE)xsZ%=%?2rS;)h;a*Z?-?95y#w)90RQX$ic<^d-SPo{K@ z@{ZuEPo8N^MA!yQzkaI$!tpq$6yJoCKd_(&6|D{>K6&y8$4idcq3JNtp*}`yi}EP> zs`8e-1E31%&L)mJlBs}+PNCGHViwpPvv^-hrsB|4L(ab1b@*SsE>Z(*6}BC~fSlw+ z*s-3R7)8da*3IcPrDg_x z;cxSt1+5FJ;p?4nnOJZQUy#f@Yj(@0L^`~3l>h#bnTX^-jhQ{Rur(HKGuxr6w(If4 zCHN_tDWS$L&rjuK|IH**qSThM9)(#KtH=jo6~>-J)HB1S4Fm!!2_VB9C>I6&3MX!G(kV#cQUBeDQtotGwk_B;Rz2(d%TwRGD?hZ z42Sr&v9~)F(Gi^Z%1POq+BEBLZ>3^%bTe;RF2`sS+Vz$~FJ85K9L}Rh0#EkkhG@ik zbAlfn-gx9o8yoJF#awb4BAX%ibN4#lIJx(b&&w`Uj4OH%AOPm*?P`|-_bXS?Yg>wo zM-KsnJ)Y#%hVN{~>dYZJQ-3+GoEF`owp(m%>n9FlKvGTJ?CWZz8Zjn|Bkb-WX#U)kFxC}1kwXr*>*#tn@}7n!>cQbJqkv-G5mTjW7t>B6 z7T-fB16me;VZv9+**~^ZRlApQ1)P&^6GB*m)@u%MRi&1hUC7blqnX31e?%L(6ffw5 z5}Tmid~Grlb=x+>XQgx6NqXH7RSh@sA*E($sZO5c8R_e@-6ye~sTS3}scI+dArZG5 zF^V>;Cm_q1A-pwtOGo7`xBlY7B;u%{IJc;7&#dpc?C5QFH7?kne_hLO-7sAS`zB`(vm~5HDBTXNJ32jc z@V;BPM8eLLXc=WHcqJV%>U}amb!vJQ(_zdvXaL9#L0#Z445v-`3-!1At^T@9hQzR> zUw~#zT3f5yENl0$$!fx3 z?~n2o50XQr`C7aw0=iP%0@HqFJ-TGWysH_sO6N? z?%NE4_NPQ4q7Kg@;%p+Asj6$U|5D<(5~EFO7Xgn2NaBwLgVA($&VLa)=hlPb2BpxW z@Rxrx%)PVv!#Es&sU^V`q`i?rYTq{hPti79`2Tj=PLa!JUS7ElWY6Yckk&B+CzPeob zcxBOIQFRgoK=6VqRy}+|)OOokzqJ-co`D`Mc;~9HJbNS&?C?Y%j0FU%Hy?cX&EBF% zMbSv%1a0?o9nzw(Ecl#^^-6+-qfZ$F6Qfxcwb zIevTV+?x8FMG)g>D)Kkb-V!53(2~aLDR6Cgn4cam^5q&BkClXX)9+|z0WdKByD_M~EW(CdC~zii(xD z7ylm6kJm0OzS(!r$c!H)=SZCLA<{Kr2Xi0=r=q5&8*WHyatc>4{ayNJZ;I=8i!Axl z@nVFChsqB^xcA{4V~$i7N#6=ft_Q?sd44o`;X?WBrP_|YjKp>{>L&jOmtC4Do+(m^inSyo#GZ;S{husMqxVc zy&~^-5~08JlL-!@GlbM8=xZ_7kCuLK28Kz;Jcc`QXlx^@vOc7UJH;p*$5t5E11!F+D{U>9b%`~nIjQGT-w!&_Uus0{u5h*YP2^_P9L#bMZ_)`4 zi=?Jy&V&8-o@0m=jji%qJcyqb6D;^M}cp zVg^Ea3g&UdH#@Ljx*P6%N@k)}?rqy*!sS-*0E|<(Rz`T;bL(#I{c3bq%u$pRDfJPu zdoJh}NIy2wdE4u_Gi(|1byK%?I{3!CX0h%>l|BoP_VZvf-63ZCcOA_4PzM8AF-eav zs}MIZtzD2~lv2P^w!O30`Ij7(93>{+cnL-a(>7rk-@U`6yuL02#h8^w*_B)ced$oC zY^m1PJ-2y|Ul*PBc`$epc?nv)@Uv@hAYT)PNJQM_TYH#U zQXT0YUhU6)@z0a`;t|8)%=>0a+_-I_%OjySrS2|Z^Me=fjn$hG zT5q^bZ(f@ya9ORjLq|=!q1m5To(iS#0-*Yb;}6n6iH86?5ur8&i673H)tW(4*hJ3F zUhb8pWy2}o)k*~XiE_nfk6U2lr_D?hFGNWt2HH+Pi&)C>fmN&pX}yTn(8}m;7KbW6 z$QnwQNE1;s+qOa99B0mE4Ryb@GL=}=uzvHHpWv^rl!vA)0cU2P`7z6>6~+iTJXYs} zBWXa(|LuigpWbfk0JOXc;62fcNV>$|cTyKh0b4O_-`=zlCc>+DwCZw>_ecD6 zizPpw^BKKQT`EbKdb{9#2P#1wYC-z)1J{uDes$-!byS_f{_UqhShFTI2anSg2BV_*YF_QK!l3tFl;z zy$v5_9_U9gZXoSgBP$igp>17=CyTjX4SRR>kN*y`t_R2)93OlY-b4i%kxz+QIHV8x z5&bBy{VVKZVGsUzid^qR1y`Ahrsm5fw>43USSb~2W^Op=-qTCcbV&-!?uFh{j>*}w z<^kf`0Rpki2fykk@UCT({B@df=Ajtx~khfU^P0t|cy9t+31hUF6)*frnr zjM+|%x=tmH4gbBF3*CvTNtB59aRT`0u|++HpA)!LAYPvZet1sSje%vn5pEZz8@2WW zQ2(mv6jdJk4C6P0OB&hPXQ3!oJ3e0FErss`k5c^ zcU|O=7NdM}f)et`#}Z)|Hv%$C6u zC&h=DK*PO8&10`YZJw*dM}$P~d7gXf#~&ccYT7@UlE15|GRo`F2+K2LDMsSfS|>`+ zH?6JarRzLU zATaT1nvL%eZ_~LifYGrWwwr#u-&T2?J!{7{tWk6wQT{7`q|D2$TF~&Nh=5{5&6FA6Qo$DE8K|f5B$uC>taS1L4w6~PPFh-y*Hpc&g#eQ zJ>cWc9vNail&vpQH0F{y|KGZn$$d6l%Lov{cK&062Q;Zg{`z5YSq(5?`4$iKmk*bP zcTfUh^Hj4ZWkhH=81ad^gs}Q$nA}F-2=CsA#Hn8Ml=7(swJt>U4?5p#U zs&ehqEpN{*ZG!we2~VB4wQ||EghtuD<=Q2-yy9%iv{t1p(Qv7a(n(x6Y`BcRGYWzr4IIP5?rk-pJd?It~tJF>OWX@v)Z&>HsRv&{CKmB?BP9uaiJ^$Bu{?Nnpkxq>#5eC?$@7N{m1mpjzZ7- z7a)v*rYgCSr{7uYIT(9>C`K(I9a12r)QwdWlbNGFB=va_iKpK>!@P-pIfwi%e8!;g zJ>MB(pV_TSYv`3$YpF`Rx@4;wx&t3|;JZohSkb-4m2nK0b8TFX)b8OTiwYuqwZ+31 z2~k5=B#s11k<7bGo3>*R5fud68FD$3gCp<)&_Wr2QP5z62ta{~f~vT8_5lg;o3KB? zLe-qFRHRZ^R=KZV*+(VsI|?o({OJOtsQ{WG<;8@U&S$?qE6n`C*1}Z$K!Z?alUlXs z00R`Fv(Xu8l*B3S&bu3s&?)@@Eie6K3g>1ajCVjzRO-769BaO{$6K|W{QNUnFpy}$ zEXn8Bd0Vr*Qf|t(nI(=)g^*+;J48CW+2M4|OG^{qyq6Me@!GP5mzF9% zeRwOhK0?HEQ==4T&NE%w#&Rgsq~s0R@nf0dXIdr_zv3?R)*CBbUJ^KJtkn?B&oZAX zl1M*z7wz&d%h`1D*Uz2mP%RGV4mn4asbq^V3+p`7ztIWJ=4-Xyy>a!0xiV~O(bu5$ zpo|*&5L{rWanw!I+@pX%ZzmQ4pbms9`~>vXTUE<*M-SuP!Xi1;egKoelb#HsUmpDL@$4Q{adpo1` zPRjWbt;nmxG?1xN8t7!n#%LBP_jf(QAL{GHF^&XR+GJ)m1uNaQH5~ph{Pe#PsC);} zDA(tjb;>!IOimilwf_wF8D@ij<*2cx8a2O_t+s!PP2@uWAY~+Q(KdAgRJ>t&Ef{9oXu0MNU!}Q5JbYp z2sTMMU-Y*5RNDWe(akBy_-LN;)FSFZ_Y!vJ;d1xS_;zFF-%A9gp!6iv;@i%G`5<=tFXA5@i%hGi!b*5?>#b zU{i~5gP2snCn2{llbw7Aee<%HyJ=hE0m;0an(lfZRfiq>ISOoFiiSTzp7b9#x623oChzraSw5Px(UafiQ9evNG zIot!6XAsM@5-J3=+h3pL)VxGhcpQ;*AEFgB3(5kmwnp55>TM~XRtN97+qu+Oo+A%) zXViHUH>n~JV*0vExg2SK_RQ+z;UjWw?0^R(MBLy9ihMKjzJwBCZAqUaZD8CDkGPwQ z$lzth11~NW+su~*-vT^hm(qh>2XhX{txS~~=M%0H9vTal!7J!xA4UJZmPC#3z3FY` zVyfM^ZM$uwX8~gRJK;QU&S#M$$cDV@8eh}e755^!HL*f>m`xR|MJy|XDjJQ_$)q~v zsKFZ#r6;mVAL=Pt&sqgDb$Q#=Sr)(}hLnPZ5c9OrG_z5D3<6oEp$qDeI%6@+$P8KnZ)dC40Ll21RkO^di85<|Q%b-M*7@4AK0Q z1m6ZIW5-Y?V?5#Mdx&hrdi&=xQ;WYIo_b#6BL{;+WmZbt@Rv%XEGf0kfr^lBII^T#hX%Z;E>eTj}=J zOhA-)t8ST~UIE4cKJASDAXSFpj-fvKA*THI>^WxEr@XMJj)Y5)>1478OYLia=vGq_ zflcW8cxP(3kqF1~f%YD3(%d?OMpi~jbb+<&QN&mM0X`;tP+r|zm(0}Xf1D|0y&zVY zBtn{Ai@(ekBR@lvXtSW=m$1lMJZG2NFP<|0>zbzQ1Vuq~TfM$u^Tu^m=8k<%k@}Sv zA@r_(N}+?5leg?_drrVuf5Q{%Gb)wgTfW<}Yc3gd$J=GSar`2&WsYDDkK0;$O6tsh ztrjcOub630YtFDI=NIMVC+ro~M>UB0s(MgN3hG+ST+osKI?iG5A3K!d2VN!RWQ(W& z-u-_qZvJEB;gHX2dZx|ciLN9)BzpUj-yeN=&VH!=_a6>*?cQ<+317OP!x9l|Gc ztfw1HjRR9trRuZ64OhgZ2^WN1*q^^=Gezo_0BVJ|&`ddYUt}Ki{u%qmA6eCW@27K- znii@V4QCb=g#YtE zr=C|#Uge3`a#pSch3g4S4~3Y z(6x$pT=gcW9Z9nb+-;r4kz>|DBFm)pVYGfT+qT4X0y*%C#Vz`eB)9LSNk0T@@M(4nJN%*8ou5EX|av2D>_-O%>DaN|2D0EgJnQF_p;qYP{vkGE< zGqp#>%+f!YWI`wF#_QkjiPiK@Y!mzjY0df(MF)oww;<_YqW-QYpGqK&^JkA?Dw-S& z+-GOOH5*~hyljqT-!#$oTFP#ao@=O@beEo#uPTA|Ufq7n0W?~{^JpQAvVk1jC?0R3tK2F4kDMFr>N!k=ZB|@01%+&ILLn8mj`=8e#S&X-*_3@ zRQ|!iZz_tf{=5HDG5W=jn~hw8iI>QN(Za_)(Jte$4OUK!|87Avd4{cSSzv;hlbbNs+9F1{ae)W?6x^a)q2+DEa4Sa+`YkfrcIhPe;M(|e;rwD6bWHy zpJr%@<{;7+`&B1crR+rz2MiB=2JKa_f`sn;x1WuNy!|qM#ygohQ-;1d2CDR zO0zL$f6c=dmnFJZoMef`UZf8t^s6Yx*`QR~iP}hnk-ub^;@h$Shhd@Vp?6hmXC)}e zW9Tz=D&`Jf%meUa@zUkz=ZN{+1;R?;B!As3gjxvMJm?(qVS~h&xtI3{bA?pWc2h(` zk6_qgPkF6LDnm?egBsI@7{qX9(pZFFnDD*o`&0KNy@GxJ^jyF| z%v0&M(U7+egBo3#R~ozYVM+9;5_jVEN}Lov^BPjXJsDxk_jF1WxZ7{R-nifmBku{v zmtaS%N7ffQd|CCJ)$~Rxz4uBV!@WYu6~SJ}EmPYn@@FDp1OpuBvSY?m-etgKD5h11 zbCupD@*z6E{01wO29#%0VNKo0!=okp_`9j8KTr?Zn7bqK2KRb}q~=7|~^G{01g z0zGyblQ)d#+O?w{)$C||!urLiDX7u&+fhSv`f9wWng4{eUw+Hlpl{RUl?;G#-kV3T zKR8}#U;#`RtKG-7iW&=YC_1_T*fNqT5P7U*^1UAk_zozD)qbzEkSeV5QDWFLmH*uG zV?rdQMDOo0z|AIh63Oa*y_1`Q{8I_)m`J zA05~-A4r2jJ|0LT=FDQB^4}3GpgKPbd0u{cfXUt?{_C^Hmwj-BzIMlB(gAXR=v-+K zHn*MyCVww|v%v#E6`?1Qc{o>eNebQw#XSD-e3v1bE{kNUT@O;b_+D5x9x0t5Lowbv z$*Lw@!ALSfECz5C_4sErq9)Obiv^eWnJL!v6?F6ZGnFAdad8F~ugtHw<*Bp!xswzj zoR5F88_gW+qVK32cOycO6TY#$ljOcafqzQ&f-~)X3(@_5A?K$XD=vUcFHkt*>=#w_El8ty43V5;HwOc z`W59%r_B}8@(Ag-tGnvt^GYCEZ_0w#mL5H<;f74doaHx~k=(li*7tJXznl{+;{Z(v zIT?)DC*$Y~(9v*xO%=~oOFvxqIQN#P1Quk7D!np$HyKC5Tiw;K;K)$KBkPu9bUH)4 zE07*ADsDrT{DEZu{A-#(Jb>-d=wZseU%ZS_shx~bNXb<#_D70!^s(3rN@}L5S3`;m zeTbjEf-yX)c9Kh`gq=X&?ZtuDZ?=s0a`+l`^Uf$_a=|?5S3;kljSZ|T7eue0!ON_h z=+VAFCClL^Z=x_oDGC8g>SZ1UYqWR1a*Dn*9p5)^BmJ4p)o3=Gb3IHmu{JP+Bv4^gEBVX{JJ3ZEAtAN6_LVw0^8Qu z{l;bI%SH~23rO(Y!H6*FD}~1Qo2$qqUPkV+n1s)%qahOHbNHM3b>7=A)-Gq9BkzBh z6B}j`)rqk_CDzd9p}%G7?7L^M>Ua-g7>Y#*(1WNOZSt=|r7;MSEWf&I>gXnls)a{FvP^TBXNW$m8q)6fu5Tp`~Z0KQE4Zzu^K2ejH zr??qKNg!6;O-no1{_kbU2P#FEer0S2+5Ng^JkY{HSWt0+(qq$O5$lSt<%jj6 zHiN7HJGXMmMgR`=$an2kyu3BEZm?y%~z|?&~U)^}jedr}kV9J`F2M zf@4iq693A(ez_#@9-W%S8MLI**WDB-iSGK0BN%)!a-+ z_Co--a`l(!EG*1;2zt(|<>EdYMY)LXmp?53{UlS?|D~&>9|ot1kY1oK#FQHqXdHfj zp}CTN742&!X5CIW$Tx`87aL)Dpf}AFHzm~8L=e~bB{#HyR-{*soc5E%0L40=Q3{3MTe}ty*q>Tk+Gp#J z*1u^XpKw>G9WskAzy8Rb?%e61(}E(MB49ucF2)b!x_iDskqBv^&0n+#IN4PUkA<_u zq6rELR#JYjjVNWD`=vj2mD0NOpBOsMpZP%_^DCH8D#~uFeh+z$N za+&{LZd{o6x+OtY$CApKH2v?{G0~6E+6f}M<|5Kvq59rd3OC9d!LWea`wMDWB!n@`Xh3YU=x)V(C3MtQrlriGYoaeRc!X>_hw zBv{3e`th>rx-YEbWXQO~X+;aqaqB0zsr*5TEgCczLYqhb~;;ZLE!NkWH3XY7jU==eO!s zR-A%jf6%lbaEcx~6oCEm_0eK_`wZ|ye_OOSCGOYC6ub0(sWCt2+6b#AW-U0(SL!U9 zfKcF@mj4s(_2iV6k+ZXv7c)h6*EBUS=H&*nW7D)G&Sq6&6gJ~zUmwtl z3nQvTB^5{_fMBSz8uebt(~B_3LAl|c4_#r>L13N5G5S=ijYJfI&xiv{qEYlZ%^Bw2 z@{BR}hj3fBd5W>#>!eS&Ym*_0LvPbus4kY#-$8sa0H|=O^Y|oQiL`xSk_QL=;fqew z^9L1|0xVQNXkD%k&lJy;xR+7|L8+>*Ge_G+2Pe4H@3&i8A5vF@DdP1Z)r$kgO9SQu z)GR1DU2}ym(lZ~(zKMlej?CWO?u0n>!I7;1%+IeZ?iRYl0QANFX#H*UjEYh;#eD*3 z5oJRihrh}L=>|gChUc{Ras}zIW`Q3Hr2lL4C|6FR4wu5lsTG$yhB{)_^JuwsG(95U zi2D9g&{SQh+GU|UwpOSZV4H-=Xz|5aD7UKp!a@T|&Xv#6DoMD0qXQ5JsKkQCgda(% z5;k*DB&ea9nc(79C}UVT>VEXO1;Z-gD^=R2>H_GYXG2jnC&3e!kS5{_O?cq`&fq)V zlcHuRQi(x|;sd2^7T5fsUVAf@EddmAe#=8FpCvhpG2|=Rv5$TEYeW(QCLINw{J-AUv5}_E(CV<{bB7%OpV|VoGcqw+(rVokUVQ*GQhV&m#M9 z>DrKE>^{bW$6o=kc_W|aH}$1OFr3eI^9Ct@WKB>~&8$p~ zlLgu|-eT6<$Y$NJH@hb+^NME46Dfr$`GGdPP@$VHVNu;rwiB4(VO}20me?h)JL5>p zB&S1WOlLc&&wZgI7vpuPR!^pB-gn)eG;N>ycWhZ+GEBJ+29=g?_iiSi`yFRO#)}`P zaLC@(*a(Kyv@l^V&k-s*t|X$1e0Jx(58OJv-n>OUDg0EjRN^sNTWx#lV$C|EGHk?C z#qsV%s67E2)wC($7MeoA)&;I9ws)H6GR|_XG>W{E3JZ2T7*K@CPc+3eMI`d3-3ji< zxTUC+`V_AmN#8nQCHn*^yKK(6xBXIK z#Eo3o)L*hyevKGwVm77Zje0W zY%YPwUqov6W)+!d$j`3v+f3De5m$J5@r0INg5ro7+$-Lnfs4dpN7Y?x1h*?jNCUE! zN2oSmi8|m$nfJ>aarqIBty%4aYo(GDx^1fk)-u$Mr^BDtyn6C#6e~cdq`~1LIm#qR z;XSh_r=Du=CO>a1^cVGm0odoEk>YdcI*sIf_s_k0rRfdGKOOhWrXX}xsex$8t?lY_(>#mf9_qbZ&W&Sdy>^G;~(*GrQB zpCVqQU0jQo{nc}EM?;F_z7`r%w{aRC_SlyZOphP&V zdxg@L3zoAB%TDZ85CiGDF_83=9ra;vbJV05ci^9fK-H^PR|1nLg=K;# z1Z0+(+i_H;!lbsR*IGrCPj8i9S2G?hH4F!kYkpnoecGXx&MjK9C}XR+P~i*+Dd7vP^wLC@ z)sYI4Q7AhpdDVsz6Qo8H66EEn&lc>m(KNVDTdjC_6XiS69DHHg|A4^uF&A7FF+RN` zzXXU4hr<=cf3!1fFb!a-@W<-p4*C(<#0v~{yjK4*8Uzbu$IK;$jk&xW6|sT965&&V zLzfZFyfR!;Oj3zbB?Ho^J3St8dZ=&)-aKI5$UZ;cJgWZ`4sag*x;u`yb75*7IV^dP za+KU6esvEgGX;sQ&J@2t3+R5r{hq+LP%A_ea9msEjF-k`)rllf^iGw)k-i=l7==(n zN7IiAf`YF;KQ2_P$Bnt1hp~X*=_B zco0erNIC@M`y#N6AP7+_KjE!Yf#FT9Q0RzGYh!Xb-x=v`Li#2 zNOEPwx~AJ#d-Z$YTQ>4dIxAU;(fHE)d3^_^>Nq9|$_d0~6o`cFgazrFKB1IY^c+3l z8=4gOV=*?){oAAg$BgqI?J_$?Jgq&hfoCq^5&vfh;i22y-FJD=XtW{jyl46d6KTBW zbdrI|et!x*N>^c`h~#mND_zmcaB8Lpdzf2r{Ldjk@nkCidD=!(oEjGgU!fD?qNvFA z>PJry3<~3>Onmx-qobU(p!FV$0v(IA-STI-lq|mYY`&~Oj%aV75lAU)%r0w}N8k)p zQ_8ULUeF=y?`KDuyEi;g>?!X{FxhMTcIo^V# z^-3zepWDX3r3A~$B(y)5br=_H3rc~J1S+}OGuH~yzfh}#@@&8zQqWopxg>CNV$b}7 zAR$xHD?;KJg{JoIRxhsKt3JGSK{w{xKa#43PTMe-dBRI-x3?Z~2_lr7V>U<1OaTDh zsh>-Zems_O3D?`~BDUh#fDC1z`2($^p@Al4LO7P|u@Mqt(x<0=Zh`E)=(&DGPDKGU zKN1miUsk1PCkag@yglKd*lxles%x+7xJ|p-EDtG#)jBN_2rx;JPt{_e$FHvH=M|}D zyWK@+QQr+Ins1;M@M2j{(Xb@4JYh@Qlg@1FVq4I0Fy6#DU!u_}q75BJ`ddK=s$N_R?p!@foj-;g1>p$4?VkI(5?fcu3SOf|c0C|ky( zQfn`KR`{ierwiHi(l4aWSFnNodl+sE$tW+T8NynX^*6$=iEb1Ek1CtqKxPoR1PdH| z(XB!onl;#XKbY4q>eRMRn%3Vj*Ar9x@{YCO7V3L};lag@f*T=rM zjPk%y6qVY&HnN>>7d?N3x4odFUOZa8@_(F*U<#u@8C@&KNvqhp8@bG0Wg4(WmhVgd z31HyvtmOhPJp5TK-j014mLS0>E|J_9g<1F_CV{0uOrp}0&acF3j=hY@%%XqNSBT2> z-A^%HUe@$lna4om$ze%}0ncxQN;9AKs4Q`RI0=1z?^;f;99m&#s$w#joHi|J*}936 zYzdJ-AngpHM{4AiyTBxTuYXUfL)i&}$e!+wFgK3KZcEAdzZ6xa2D6LWOoYp`BQRfz zRxC=5AxLr!#e^6>(>9^dAjb*sBrVW-1(WFoMi?F;`~O$csiYJ>Ya!2%e+%FLm(ZtI z5DNNO$!@x%jubCw*81opBlUr#u#?0|(Z<=8gC5q_kGXG&SDppO1IW3J2pgG>ohw~v))hydz3Us$tY%QEjI0q2)wWU znGGwm6$39s*T>S040B|vziK{kEVITNTc!;0)`@#~I9`nIrDyz|6rgtgbYE8Yef2;_iOC4W{z1|ymWbama^BHJk7kAs z7$kNRE{C|C)1vW4yX9vNv*FyS(S%#Ob?U4P`MKgdYRL@r)GofZKn?ae^ooW?IjF=! z0Y5f&MzEV)67Fdm%$^3Q*AOy&Pb;#g#w0K6;*Y4OlGnShHvHY!TjQowWnDc*Jkf$_-8U3-kkQsE)MOeF(dkI+zo43WrLcl6 zpxNNcl=-V`eyBk3a>E;QE^>YRG%R2^HgKi80SDVbz9X!Xd%dBRJP2kod)7+!kG-h^ zO!ga4UKl`@XL0c|C`4Pm;_c z>mCU!s%lrGWe*!t69s_1-JaAzb{8ADuBQV`MCRCwA`Tf%~Q?K z>2CNe)}xUwIU^a}{%d#4bUy5MVb_$F5;W{Zc>0|*ke+18$wc;5PlmATHvu0`QJ_^x zUrj13TA1afWVgrF=EJ6T=}6I*FuV@NNK?$!Z5|BChC08X)Pa5pji!jyPW!aU7KXD^ zL!x0Vx1^K5vVCo>q^Vt3^8C~uU=@m(@I~N)hy|>co*ocj(&e_8Y<28en#;vI2iqzZ&ylH_ic9fGF(^g^1fwM z87V>Tc5!HIAtQ{$nj--rE_?r3j$!5VuCC{sHevjqbuA;(`EE&>gk`-;FiJ9031&!$ zFY06zpeDfHuk)ZrTCeeIj+kKriBCZa-l~3l zI3e9IBC?GIOjyvN7&Fem29|AV6+LJIM}`@#1EebblcZl3v|o zP!?D8yw8`d)vl(?OZm&aC&N-|lR1riONOA0Sb{|!(k@Q4uNK)Oc<)$@0}Wds-LD5B zfg?5FZ7STmpG|%@s9dV-QVIRxXcwXGh4-nq|H+TbG}f?XeOX+>8|^+L-FW`fTem8+ zgr+)`nJm*KUB^#;)f9e{@EQyc4|-||MM{sAMsv%ViETO$NK!QTSXOrY(1IGq?H{6oF;q@$AwhpKkVVLJ8_BUWP8NIB-oD$K z^rEjeT(SiHVyV`Nc3NPKPN;b{;*aaEXGq}4bR*SQt&R)w)FbvP4dI-(UOy35?x6a8|;^OQQmL*SgDGWmEzUt=jb$UZ{Biibt z8$w1MZzXzFQhEuLw2l$?eTb?3eI2uisaKSJLg*mrmI$4}CzXO0n6HWVxdoMRt%>ym zSWvyjCG5Vt@A#%YuIZ(?t?dIw-r_}yMcrN1h&+`!qM@9h_igH0m{$k$+d*{===1NcH~TXYuid#GwG(z;IP&&{7hu~FVQCRnu=^8j*-=c&cdH+=f$`5 z4j-g()&VNTvzHUB!NGmxMn4ACdb|ft1atWt9c>Tp>$b_Z7?|bD9kG3g==06Op~xuj zDl3MQ!dW?#smBTh+FB~)ogDKqhcKFAs*|?2DFQcG1z1zhV0-MZcyIXUQ?$hZ9&FcF zLwJ7^YIlMo#DsFT3%#&pMSJTv;V-&_*jLSEMATEa4w2ieQ_L{*tAZoqhBGzu@cNi$ z=_xUu&n>RYowA411kWx?-S;*` zHnm_g(~HCnMmImfxy76XW+vRm@|p*%AFGjIUW)hX7U*dS8q;#>7t)i+u0JnE5}N5` z*!-*tsS>v>mFX;%h;irZ^EDV9D`kywUPcUeIMy(#{#vYJRnE5|>N#n&{Z*3x9bdjAlsuRIyC`r_#^eJ_}uf|-N{Y+}Fe^=z> z$$w7Vrzf_G6hfFFqH6m!Ot`qSB?P|}Mt2LbnD1-SclZ6geqZG&UFOX<3$$hj#Dv>JJ5fRw%8SPJMyC-A2DBM#1AZ!3JdUIAehEbI8O zd4lGB{b!qeuAA{HQ>k4{i#Hp4CS1@XfdvMP7P4<0CmZKwtzpCnL1VMwB71C*?ryr0 zsj+AVK2v$On?dvIuFhih!$(Ff^&5dWJn#Fo9>z^GQVed)YvR0ZZcSnl6Ih}?I_>h` zw&y(Roe0~g9ob@yuGSGOi|S~%UaeJ?i)WA9AL&l4-AS@*!ND8XUtS~62a#3}Qp?r# z#VhJyirQLV&fyCxOrmZ&Xm61UhUus_ZDLR14@_esV{eI-O;0wF!~sZmeGFsCX044h z-RzC%(?ajL3Ro&il0?}A(-2-_<(!WpEehonnPbkz!M1fOr(@9*$y1^0%-%ltTWa>R z~_ZYmaf!Q-&u>H@J}@8beod#Na1QZVU*G0kX7;cg~ez%jW&x@ z)%rT#?^c1Ze`AuWN*n`5ADb%;e%MzF_oPm@n%428r)^b_%gMw==QoAXf5wz+RVy;g z{S6majh#cDs@hiE=819ar3T>ePudE(=ZnH!meuxR%T6Unocg!M_?)9WzpL0_>ugLi z%&`jvUC-C&o|k%89L0A@zMsjE9hS8C!8=A*5yCeU!Y#M_G9gia)vMU%K{o29>N_qp zj<+{Cij~mF^O^_QJkv|-wLX_X_|l9WOfv{`fWUk9=yN$&)}L5y{Wg{cV_MQz_K0}7 z<96FWz0*TTPT~ABIO!}{{R^8=!X$Ur-)Kdq-7+Lm2R=axZN>q8L2-+i8OYtB?O;lg z5V+)^Oa8KX|1wz9%%ELe3P(cYyj;3}3S9M5(t>kdP^WTzSya*=SwM$c3FvX z`7ES*(j2vXWxOt93l(#Djz=ForSCLWa30T{v;XB%St8bAZm;(zd_SK3Vri*7jl?2+ zN^2yY5yIZw`Wu)EZ)$D5CZOE9cn|Dq_7CbVCVCwBqt)Wxn>V6K_kn1e8-PlSU8z0R4Dw;iB3RD^z$ zBAB z6GV23zBK>{aVyny_Tmjx;}Qyny^G3@mi!fiuwP&dh`;_|N3r37U}snjAPmA@-fFay_x~Y)7A4vduMR72Pdz^f5jR zFEdj&@4EeHyKC`kgosVK zRCLc{AeEf7JHd|pU7GMc&N*BaK*I(hwgbwgBMx#U{Tjk2T@GOWVfR3oGqB?KDNgfk zy?qQ*PGWJs4~sR~Rn?(uz_Qr=!BzL+8_Tt2biZB570@sbtw~yklW@p$!lOcImAE)X zxY)HL*x$j~TEB=23Q;Lu?)&zA{=y)P424J)M=rwXuN-YWE|I&;BqUcLvxlq?nbYKk z-}vC@O<4K|eA?gB#voi zDOEN`b6u>Pmg3zRv+C8ifnhY< zqxh}mA`u6G0*yK2i`$G#&Freh;Vj- z7L1LjWfh$Z->}{X@z}Y=GDj+rNqjh-(D79mz9pA8Y{VX^9B9z}s^3??V@d|2Yi|(U z5p*UPYf+w%3%Z%L*1)u-=Rg#L#^7%shhnHUn=tsK7ZFYfLAKV}9$E6n z=7T9J(w^76+-Skg6>SIaAMQS)@pIoG2a9Ifl9CpZHN4pb%ux&o0YosSFhUKmK~o{_>S3?HIh$sA$---On?sE`O1pO zgw43ASvLHEFy))>-1VJb?l-PU;~OI_46cU+ubBN_s-n_jI>ai^li11%enb&G3fq4z zAG>E8x*n{M*yk7In63A$B$}m@ao|RA}NLV95(wi?8?1BeYTp<+(9Ly)Cz$&R%*>YYGGx;F4w zNy3CMfIH)bTPuU*(t8KDJ{1#&z#rNH#*zblxap#R$Is%EtaOj-w_g;f5vmzPKa9mY zz>L8vP4wEX$6TRIm97lpfki4x6S_g>4*8gGvFx<9#;{Qpgn4f;{ATseCH=1c14&I8 zUe0>oOtG^~9Iuw-Eo}mWAlGOxA}dkaRQBsp%UmH#+miGla-FbFxTWv+b4TBDVhl7J z5@%CiieWv^l1EAMP5T`2_vJjugxi<uo25#V>~ zoPG48qJ-ik(FHoB!i|SpxqE6UO|WCH@VET{U-1g(0Oz*%5pg|j9P5~KnoF-rEb^rH zlL88H5aIbLglSZ>eNks@>;rF$J}OA-EEl`o2!pcDd8KsaVK@X3)u(BfCf%f zHsO{!0T3}iR>28BV7`^`xwaTzGx%#`4WQ-@hf4F?y)fX&gq@2I_8WaOMzJ*Vs7cV_ zH@P+(%bm_p{k-VxmO0QrpmY&v5ZRw_H#(4Y5oqSZfG~%EZP2Bht?XJ(Wyi>l_Iz$E zlCbqO!gJ3IU)RpzhcCz%N{iQ^uwnf9G3H}+55y4WRF!3 z1ih#TAgz|lK@cK_E{wavsG!U2a%@2d8-SC*NBE8E(z(ng{`>`qLMjH9T&<-P26GQ> zslnvH{*5Kr1D-~&NIKVLe{{d(iF%hep?NPb6LDrtJwV_-MCi^{i1vy`&?I1}p(A`R``1yBn-_Eq&=tM4#lCnX!6F#o6lq!>kBH*W)T96-Kzq#)&&B{I&}~i)T6KdR^mnPhtle!PmQnqoC(>HS*^3v?&Bbw7JM~hPc(u4 z!c>N=cz+3Ts1j*ZeCuu#!3()61Tz60+w zxqsTT!y|8gaxmpTHFf}!Fr2$}J64hVPr7|WS=@Be1$KGO4A8zX^8~`#2JKBu+HQRS zK~1dCo)6Hfxcjr*9*(+7mNP2&o#5qUSUAZXL$-Cv3X#$S>xQj_C*laVZ!>?Mv)^Ru z>+SG%9507aAEr74F4c{X&#PKoF_o9bGkq-&#TZj$6dvF}bND_kNlYC0WZZhnOik zfufN<=J^AFX)Fmzw;7x$Y!FUVx|FM}M~ke!N{|X*8Lyq6NwBJrE`P7}L$@n95I0jU z03BixzRqRqJFZkM_21SK!tO_^C`0%PVCKajUBV-C`ZB3wezX{mof%(Tky#DT$hsl= zwrMj&VC5DChI&MH@H}jb61?rNGZ7>9?Xm2Bb$i>FN{9_02r2a=_#fTePv9h%N*OJ9 z>Wq%eHx%W zfSe{g5YyGhSQ^u`Ja~8eEmJYK~yjU?{{h8Dlpg zc{wo}3DLt4FhIkH8;o$T!i&er#8#|TAxoBLL5cFs$;z1xX4&*7pBko}(P_$M3iTtl zCxyuV;|;=a8oH+hUyAE&%kp8Mxzz$k)9L*5chCDHj}6-reyYIbLcY4h-Fp#R<-N72IdnrnBtz$KKWkJDyn3Y4ETK)bN}2yg}n;!M9-(2BF-F^ojbP-UL~>1%LI zWqpd?t#?$HMTaNL%Anays)|LZO;1}M8(*B8{E)~1s#p$avwUa!&eOck7$I(El&UGA z|9}4?=#9Ef^GJc^JchCRSA}KCq$tw`vIMy)l)CkLWL(q# zM)j@eK2-(4Y+T4pG@N)=&~d{mzxdTsa}k}N4c0bXsuf!1 zSFHZH3A=0#`(E@_H#@H@<*(rP;Txy#wQf}?>~Tjfl|T@JHkq0_q;E2Ep8wYK9}w_v{{k+aQm8Rks}kGlBemE$IBX{C-}h_Lu-hZ& zu$1ne@m_8&Qg{t$s{Za~*#cw;!>BQ-SVA>8e^?YemhU!dnT~ZEwU-<_*252*`1%6oGK6k zBb8b!>mesC2&!W8O;4GjCTZcP`50!x(eYxI6OTLTtl)^_ut>S+=41b10MLevY?`p-02olLN%e?O42K z9PwQv->cH3wdYf4uKYNqm_HC7fbb~*SpFZt-*~Fu&BaLs`YgSwCSnom6tUgU| zL`Z;Ct$Q)AbOns^?4_r$cANS-OeL__8av*Fx!M@~~tbzwmdg zUSDA)AQjb5m{t1UU)0i@spN|wQ+>ng^(S+GI*97}Fs4`)z?m;mkSn{1`S69jPFL&b zk+hi3>$0X|yLKf{BHl24ejV$u4=d}+wm+^NbH|Q3s0dLdK0ANIHmQe<%$;X`$|r`D z^yvEogm(o2Xc@zu?2#R7{vM(kX*m44P~dcdMk<^1_u)RD7c-2+1X|njp;60^-792I zI&J6gH1=8I#)0{@tu7YGFLVqI8zbuRN?c75yMSI8D0Pw61?wkSIqRy7y^tlZJ?WLj1@i7 zyOB0A-xMk0p#%jF->7Wd!rB2_p?TF%YZ$qr~ujDd_04j zIwn|iaOS1giAsW}&^|cvwuQLC`h~D$CbL#A*?QVMGFH-Ooq~G%kGBjl&zgccCj`fx z;4HJEsF|Jp#v)ZvK+%p^Aj!AsEo}`aOf2tcHGaIN4j~G=7x2O{y6P>ORuysQM_6lA zM714B0NZ|cUYXdK+gL1u_Lo>{dX#7Av$+F5L&HK6Q%;{vnG$!;w>j@@FVutEXnJCC z3?mMQTd1uhZ~p&XgQ-9WG-6+~_f$@^d<+2dZCT=yT6_PIN}8_+m{2t7)1Co4;K;AS zmV`>HnsS2Q#G9^b{_>TpQ>eJ>WtosoMU3uAb{C_= zDxFo*0)28`()!JXo~|5e`bLPrd$=?)2W8r6A`)PvO?d7y`i|^C8mNy3wvc z`9c%-lIylvb0m@y&h4WAOS8VgYUJ--c59-D#%-ZTDSiDOf0}yflKZ%^^Mya%IBy7aJPsf#UMg1hYS>wvmd8(DG|hTi54NtERMc z)o_ljCUTHQ8Df^mg}NW=BQdpJCO&47h*GKNnJfD5;P6lZYY0N`m~wAQX^6dMkoH?r zU?+mT^q%Lh{KGK@1WN|U%>2FevEgp%-TQjUE)n^TO^O2#O?iwI53tPULmSU5A7F($ z+DVD&t53Qve%T&(D6mO5B#jh*wKvEZyMD`K&&E4R_q%T==v0@z+Snj;f9wiu(+oor7HSi0Ka(2=N z|83SqL{j+K4b|*B!3tk|=o|dd2BdnA1>7sV_IX`8o<1+JtQh+E^QO(fm-aDN7_F&P z_+4!%j;e@{gVMa$$!Wj&cVHgP7eJLUMgRD`$`SphHMwBh&3g>|TDqhReT`>-+-W%4~9@#}kX3MH|I%OXZvT?UDw~yD57l)y~ zfM1!Anz~xHJ|oG=T|rq7oFq;75Sd))FBRs4f;L5=x=X^H&)039elAUR?@dJMNjBIP zf6w;!;_#p+92y*Lw5v+gr7xpmkANSkz=-pc%o3#fdHsuYwFI-p;Si21xl=7H1$Hn9 z_0#eZO8yzq-Uo0fHFdJx_uI-<@)9BGE4w{GWw&j$q2UjZr_ZMhUJ?f4-WKAmmzx!e z+Lak(x~?THt(;#?c=yCOIQS9aQ@X+j99OQrxvT4SuA~$G5SIF4XYYSCiHDe&3*ZkY zSP+9-QTr&9#ko^(A2jAt;MKWh-6>=+4SyMEjsRLBd=6RY3q%zvgPpKB!j_g`c3nVp z1`n0<2M+5)ZbrOO?Js5~U^G=yBsOzVp$$EZc~~1+Za*&3lOkqdlBc)i?cfhN-s)|W zCf1!)Y6;tdQNAIW(Z`8}5UGwwvr*x(vImY_58+ZpB*oJ-`#HTm2x&ZWzORhIP?O3? zI~pUjx;>q{1!r!+1lh(mirP}_w!JV9|DkM{uCA%`eep{*+u12!+4!u|Ys_!KpRQhp z9y3H$$1C6M08{Z`@=Dn&3ZH7BNw+b($*qbc%6HijtL>K?z)PcvF#sp59_x?;YhA(p zOD#8l?V;(gkrN9YyK)y%fggIaCCDgZ0?a~JrH^K`P!8`v_AP&xxL)!SwhKC2li*Ea zVlpQS@R+2l`7UFX>j|2|dO|~=oGjJ3D=rvnmiPH|;-q!n0BU{RqNVFts_37!N6$*U zSSBv=Wxw?|@>XRWIW{sxcIW!#INNIBs{*#_dBd)^+OG_#Xsxu+LnlVMGokH=2Yzj^F-HjS?pCY0PL})X-V*+RQXE+ThN4-*#nl@kLH^?FC8^<3`T2y90*+@=3 zcJyV~-k+j+HyrygF~!ZDMk%xldfwuP*x1~YLJ8P$UPu4*V&L3%)$w1_47`Hc(Y2k9 zF27x-5~GXJHJn*cV#DQrdSbKjI6|Ox@D)r!h{NFDQSIsi9+wWo`p!}B+jb|wR(El! zMIc$sE~N$-b-4ljV3a+S*)&hDY@hv6wCwSVs_j;$bs!;rKZI4$i7^wW<8u_9 zp2FaqGRj3?bPAncX-4cqD{k24`vm`2Z@%OYxUqZz_;*V;kaS*L7@hCX3((QNV+v-X zQyf5`wv45lzq8&zoCd zDwxHHH=QKdVEpSXd0+8(1V*F3DZFZ3l z^CGPjF*||B5}z~-zk6xqH~lFlQ|c5c`6QRj=S-?F{X5KlM7!jjT(e66E!>SKP4TVc|_N(ciT4LD@xmn(|*+-oSf1Vo7!*e!$A&r2xg4?;Au4_3q+ zq6`hkY~7?mAmM}@SGe@=LCGUysREE0=e5T-JqXRwA{kfYPIymq{ubVl_xm~%a<`LH z#QcOOp&4}#W2maZiEH9yn{k7mu`kZ~4}GSoS$74nEDQN^zek>zzFa6EHze3*1$->FG?Ym&slffDnahB!y{lF@AS7K; zlcBk(ViFV#OI2>T_e?Mfb+SxaqPv-Va2|R+c9V;1k^EKHqK~=Q413wcWBV$0`}Z@J z5&j|_=`;JfD_9GW7mo<YO zNwgQx)vZU0*CEJ)9qSNMaMEZddOwoZ-Z;*#+H*-RKawo9$jlbFnyML>%Ae3aa!vOK zsVSm1php1|)jy#q(1$iBjh5P=K7yF5;yWJq2gHs>oHDFiwGf#zC3!MVNJ{4c_w+cKeOyzh39xthZ1DbN8dC6NG1Xxd z)H!z`hTzQOA!c7Uv;X!wXE$Y7UNQkdl z;C>eRgGxsH+mzdk!A)k(5La0^Y&_MddX>`v)pJ0VrGoiJLEyqruQ!C1u>FJ+geTHj zUHq?j<)*6hS2P15u9IBMgnjP^g?&B_Vl8WoAz^^Rz#tKrh+yA?13bpO=iw#(r%h`C zB%L)$onY$j3%Y$*V{kjVx0#0`IUu4jHk2R&(sNvdK>D6KiNCUZlm)1h&@=o{&=j@Dp`#;%MiN`C+A>peuyo~*g1Tfe24bh=byp8sDAjeS@Nh2hTLeKb`tx%vTS_HiR75jVWQ6`-HuS&s z59KaQTITIL@2)o%D<(ojtD6VSJm6f-D7W}gut3-}9WvJ|StQUO9m&fbUl!%C%-Zq9ET;3UaJeH`9s1kD$SP7IXlv8ap*<`?|{f>^NAV zz~{iY<_jy^-$;UYaB@|#!AXSO=k@XIFk_&axx5>TX3~6ie)<{Pn+@wAWxReI@Eiiy zYwqFj@azA%fLyzz7zx07+4H?m>`F+GP(^t{z!Nd@?pliy?=4vGoc?`ZnR{4)J-4L0 zcpW>|y)y9C>=Mp4@R$h({GFH)u&vxI>-zcA$u}8qA5^SKR?m6(@O#FECJP@h_PPNn zBb%Mby3_jDI*qG@7+wu3hPr6To zK3=P8w_T7f+mt9-c&sFwT-vo#tf^VjYG0=+Gk#%burS=5H|FT>fir!?X>T0bXrI97 zGJ0 z6%*DHZIBInh#KV+zb1`3nSBQfU<>-Vpz`k(V;-1+R=W&XZyW|xVtxZcl7L*juC}xm zmrlAOMNs7_n0ryrP5(4^uyRHK$qg_@Q1UP&^g>?`<0xf&PTr~^eahb0#1rngl~)#+ z`4&Fdw~Ktd-v+sB2)RJro5h0>T+2Aek(k}Ym)mJ9zjw#@xBFDE-A~SR57{*DPNW%c zrlC$~U;GwB-&WXmL|<{=E#s-$1_HJVdhy>^$5*T za{VjZUvxAim$|$Fuzz|Al(sff8|B&W3&MTYB{ivJ4?m;>Z;*3$b1SPc4XJhWI?$i4 z0W-*Ahld3cknHNS{jsszOKD~xeA6G8L)W>SeR>DHpx=+mfNKW`+EVsZ;eZ@+6%W4w zz+b@SSZw(skh@_#*|Lw*sEO5*B}UxFJxRA_-*SXv{71ZA?K@o!GhtgzesTdyec2J< zltw2_c6}Th)mbbilk@UNV#UjTs)U-84b&HKsZkKTFl&xdLmJRH&3#JjbQ}nmFUT^9 z{j{cK^TB*#*XnrJz6Y7)Fi_2zsuo7Dh>d5jW7uJP@H(}Q`M$@J`1(& zr~lnk4?+X#JYeJRVnd>7fKKkeWu~W=6Y#!5e|c&QZ5(5bz6K*Poo+r1!#VgGySJE4 zO*dEy)FK*bHB)}HX0v6Sv|F8@qDE8fxbH*QHDFhE*2A@inH~N-ETNJUIhl_yg!a6~ zkbKslXsmNV>2qvwDiag)69#y!&t344nTy=w{E&@`-84eTZ=Z*<+w^`Yj6C~2Q#NB4 zTP>I))KXq~Bir_0OAK4+X`wd)lFOry@BP6cswu1f)>xMR_%C2E9zU$FS}TrN%^oQB zJ3Ah60AmK2^iK4rtjmFRN8r@=_-~PLw?tDY`T4kO_s-7FsTbDFd!GilBjIt%Jqs>u zESVHcJx&{G)UtQV)J@PcsQPsDuz@8;*v9$8NnJ2tiJ__O$Hnj4h(gDNB%Bv|EF)^g z^(22T=ou#j=#@li%YA~aTR7BO@7EKwsEvCpykkQsOC!K3L~_oy?kxO_6eED9SzK|5 zOw(e9Tx`)|>>?WqJSf+O_>`NCw|K<_m4k3K2-{oVRiR*h#u)LyH3N=dW{?37%Ah}MUZ8MYlfx{+B3HHHUK<%J*PelrKT zehstu-)H!(0XP^NyHB_#z6XKVsp}xn<6!q)9MbQwA;V*q^m4Om8zZw{c)Yhd23+h` zdl-8B0qH%N{=7jfpWrVjlA{;c7=0`IidAWonEH4&R-$d*q-!mqu&m#29R2g78w-Dg z7sANLY0F+Cys?5d?&Z4Q&pH`l=b7)+^SV}CQ8u`V+!0DkJAaFz865zNb^$D$jLhIU z932{VzYH2T-OAv2dZWU&@Co}qc%$B%dE1ojl2~^30W(JMmI)%@;*)+-&^Ik&t!@@X zZQPGY_Di+08`<=uJJ=>%P8h@li2FGHG}?!BK~E6fRF7mxehocwuFvL*d%AVKeArOo zkJ{nPZxIF_fw$bmhMOgt2y|4}JB7jjJ)mOFHqr4*A3CX%wUV<75Rq9E!$VC%7O`W& z&k11xZec@US!)4cEFN%yEv*6G1oqO{&aLI@GaTqHEd39gkkyrmeWws2tL`-GTTgR) z!Ms9*xTzLHnf;!DrKZR@62Q0TCzj5OXbbceRkk~SOdH9}9;&cmGT&9c-HY^yPSJ%v z8X0LYzFLD1O0JpB@m!#q+{(!{jKhctSLsBG!i2B5#9~5Cu?R8Qbe525;NR@{rVU%sIgWG zrQCBP#=ZQW=i?D~nTQd?WjHVS2m~3VgHG4X4aC)t+E=YF9zV#R1K0bLut?sL+~Z5O z4BVLSST+=Mwt0V`C$)Ewg*pvSrIe1htuGB(47>GPvO}G6 zI!N$lRRhzZU_ckj8jw3k0|R_+SV|E;Y|P}Kv5L(kw4c-dEwHNy8wiky-8C)b5AtUK z{DClnlV{}ao-k6?o&adZ;H~V%K$)v!JgnCarrvK^C;)YKC<1-8)p>e*W4q~wu-mf) z>-&NK8;nCQLx}Thk6+4ZevG~3M3TWYFt1c*LRy|r!(nr>oCgepa$bL5qzak=L zx>@EOh}$@0Qdy0$E%}yFShrP1q5!rGyqcs!YZ%SY>WiTd>RS>xpVwmo7E?KAjc%#N zyA9?M>9(NM2-N8d!^9@|pD;Ty?bOTe|KtX6RzX;Kk)W|dlVNCHR6`lGS;gay3nj-Z zM?a@<7XGViM$p!l>mkOm);}V`QYiKxWfia;EAav}*)p`e?M2yyqJ$~25g?1p?<= z#TIgp+C&4k&OYCCi;bfeYu6>|O4iENn1*@-ZoaBDQ*0Y6`v8$4UUJ(Xv`(Qy(5#9t zZ#+)(-#Lr>1xVg6+vXNq735AY&NnYkeLUsoUtjv512oVI@aB?MFMgCpvq+7`sj)6c zc{$V`Exv4eROPn5_N?ys-PvA&e<|4HcUg9&SkQ25{DqSha;|}bI`mfS3E0FMG@KZL zJw;`P-O!w(II#9SpP@iqt)?vIt%_5R>WUx{F6m0y6JPbo(RJUKY0KRUSZtzkz{eFs z=G2=h2tAp|8*8cLpm1@-hpHl21p(7o+0>37`Aj5SkouKGv_c!5T61)9@p@Q_LXDRK z59V3Dx`62LJ_cJRP|7preAN zNg3QcR1cUT4m_mP^!313{!mtA9Aoy7fARwHQ7_FdUjlY+qSw^GJ1gHE3~Rm5tx1Wk zMvxMwGQcGeY3d>r)x>04&ml6AY6^-JX4w{nVul1Hl=|@yl}LAmA(Og=RK%=eFyi0N zFU#jPz$<(Q;;!d&-Wh;Plhp#~{%5JFGe_KqEtnX>bptdd_Lmvy*(W5G8Nw!!Fp_;R z5(7lVl7dmhk!5{xsK(oUwJ$rg)M08;uDmpCyD<3dO!<(W}z*} znUQQ|AcArX~Ep{Fo8aRoR7~L4;9Q!#G5Xp`P#m4Ku)I6A2LfgPM&O=ch-L=ZC@Z%sPkMe%2QWfy!Bn}E*<}*s zdibsUtw~bu={FO7g(ga4&OI{gz+I~pXkLED%%_VxWFSW_wPB-uYDhc1*?VA@01`J5 zbJ`<_AB9zXIfkp&trTHUsc4hg=bL&&Jd6yqEPomAA}F#b+hH^zSQ4T1qkclXdB2nbSx0v0LVAt;S>H`3kt-H&Un{o8BrwZG$d|9D?tfrokS`-<~C zuk%W~IBm(TJ$Hi}EJ8_A4er?|cYbKP@fblFcAq49w(;(_r^N##&)>7|J^mQ%{x*hc zouQP!(=`1VzECqN+>en^ElTWUfvL#2u^q^YR%7os#KrswNQH}VVDN~Q8{~+Rsi?E) zWaXC`l_}YxdcMX#M$f$$;xaB93#SmV!2gt*rgWvJNUt}8>_Y-6D;MFXTaJP1dAyn& z0TXbn#KSffM;Gt+U>rzpR{#OKrl42#Pyv1>GMI7vNB-g^NGTOo0OZIOcr*!WFG{E; zUk%|*&4=6ctmiz+jufx^7yNPL>(3a%Snob+RXPmzeH!Un6~CP<;lUoL9o30O4(cd* zvGPz7?d8|A!opcjTK)PwdadlKR-J~n^ABS3*Qm4@(X1yA%n(CJ;TEUScWYJln+Wwd z3amYsaPU9sCZ)b>PfK5~t%3CA`r+aN8+M~m9#kV<266m7VF*p}7$dBD{4a9cJRAtu zhLi>aVCD_3kr+ivOWouX%Xz=#@fR(fOx1`g9~&^>VwiI1ZOX;PwaG0nKY4$a|BmPk z5Xs1SR-Bz}c$Wsly@n{H77+rPW1y=U98oK+oS81MWjfq!sAgrSH{dKR6dpfYe%X;! zTsSrsz_0h@+0MfC)12v#cl>=}vonQe#2dF@rWvE-e-cRAgs7d?o<8$Dp$Aj17+k9L z0{+YDs9-kL#OhPK1|pp(D%yIZpI=@$rY^Z=Ur=tzjLjMW<%Clm&O)o3pq9CL5LFk} zq3L-NWhA6PARwi8*m`L7wXaZJWvGpdI`{ z`dt~u0W%P4tG=YE-FG*KH@@OBo>k~Ok>^iGORPMl-zxI+(1_Bs;-3*7bD6e8?y(HG zcz0(+>|t_rno2lWxtqhUn1(S3?62J&NtTq9d|a4YS2?NgJU;&i?lOfVz4Eo_)I%!H zS%y3-D-ZA+u;6E#|Wk%pvhIQP%GGp7~WtTO$fetG_af;26dMO zq>{O)YYrpJ!@c>{WHwaE??H4^uscrz9)(1mJIjmV!jBgUrL2f^<@XqgIdCzBl@LAk z_xLfVU8MWf2NOFHax>A7%3kbl+2%_z9TZgBbC;d&o|sO`zl?IOm2PiI>x$bj#0dx)+JEXJ=-eV8-n6MgM&`r2; zbfZh@i+0}6V3z4I7$Z`KU8cYys$96EJaDu2$`y+nsO1f=$TAiH+&ykcHga}alY%Gt znRI%_+%RsOQl^G4!B~3L6=AAaFSMh+NXFewgK2*`X158}j%4x4Dd&(JFQda$Y3E!4 zcpR~uaIe?{eY8kMZu)o#PLdc{fJg38UPtsz82s5?#PQ55{=8oaz8RLqFj*ynqeS#d z0BkgAm-EY0{jF|gO4lvYF|fGVssFL|kt?I5-JAf*cQ-5yPZ{wt*{WnK)B;dKHwRG)3hU9g!EXO>0cn=%1_L`e7B4!nAS9I)tTt84~mS|$ZCIl0wxRe{N9)RG%qb3e* z4FKOxIQ^g~=Pp3ycxd`n*ohY7F%BdkDwRbQHW=9yEJ3(r3D9@AOTYZAU4KsnS zp>qz(lo3*as;P@Vc$Rqv)nNP2Ias=CJ$cl0Dco5?hNcRLCeD92gSFI0sIQw8%u;d3 znRumyZ+W(UY#AH^SkrA*7U?r1RZc`>4EDl&pN*2AkaAjGBtPF0({35<_=pORPZ}U> zDAx3y3NnayIZ4c#Xze?~BIQchE$3VLbPJGRci{NVws?WAeWm*2y`8Twd+D^YHBv1S z8Ok&lJ5wl6-{tlgv@LXgv+B(rES0ohKK6iZ{;-b-F_V-ekn8{a*a)oNF6%T;6nlj_ z=ly0-?ICcrz4a*|fX;+$I~JLmcAh8qO%~qx$?Ihe)am!{&RrLKfA}@NBp247)_1Hs zuY}Un#>}Ma<=n++f`4=P?Do#(VBC2v)qQ<7T796m3T^Tr zTRh|Ap@CjB%8{BFSL`?Gjk3C~4Y0UlAE02m(t4v|_htZV^gUKX>X(|Hahdo&_Ixx}3r31kE^Oh~AF(3GK!N3%b6pD2|Cn7J4o{YDmTB0mxbwOJ4Z)dOGwIp4th0j?3 zus|5!FUwkO^9&hOvy|{qS6Z#~II@%f@rXO*xn93xM}V*|C)d?)>{8n`;ySLf~OMke#Y}56mh;WY>6O!gajuJq2+{51^7>uj%6U&}}Ctn_L-&-;B`s1Sc26`sIj^X z`_GG%qgRv^Et6bm9rV564;gI{PGSe4xzzt>M7YOJ-)mXXghLc?6u35^i)Q%V2v;cVuus*e68 zCsWPF1vzgo!4IBCABUq^1G>mni(hn`GI@zT8mR@cSgXo!U?D&tx6-TCZAOCO;5Fnb z15JM9+9AacA$tDXE@-5Zv~%*C-hi~{vZK@J=rdy3Yh+O_1kaw?G!PNKe=FL%CSISS zowXw9RG=D>ROua3eEBSqQ6#=#J+%eZu{F*`0$u{euL9=TPu zGfiZQ8$lyyvmL&(rO_wjVu(1*D7ZbNj11~LR7B@(C8>EFn7;UsJ0)8?AFrFsl9LW{ z%=ARB`FjalELa|ZRUx&;sGepXsDWdsSZkbRv3WLkwe9$+YgQlg$EplyXbzTAvC(oM5PM1p{?YW}72C z^{FBredZNQJ#4p)D4;QBLh_0ED^L-c5XGrKh!dqIel0_^2P_0uk0@Ui79AMPuElzcs ziW#tiLBGq1MAid@qCc!-D$o1ui@tq_D41uI(QB+dTeEfo`!0sTyiA&2#gaPv6M_&h z3Pil^OYYA!9RV!s6X)nH(x>-2r8kD4*^ERZx!);ZL7|*gB*pT~k1$WJX90eM%ns4; z42{M-DgOuLF{2EIiJs@K>O_~R^4{V*R;|RkvN1Q5Fguf_ZVJ|I-9)@LWfusNcuhmx zs25JFU$6_MwO=?gaw(y?;kPcOBxQK2X-g7VVg|OEB}KsD0g)elCj$`C zE6#H<3#oyd$*tuSU*dt6X7nfOKO^b>;i(r0{AvzQUnLc!;>Jg>3hr-9M zCv!kzT>O4rT4!jqec$1Cx1%@}o$7TUIfm5o)miAhWhv@6K}QWd{Y7ciVe| zh9^`%|8gS4GgOcMnWMIhR@~PmWdsAdYemdKos76ydnJ1H^Fy%sx=I3H`ns6Rf>6JS z|5uIwpV)E;__YyG46Oelmi8FbcXmDbG<4JJWOQ-_sPs0n0oH46x0Lnh3d--J^K(v_ z@JoL_{)lLB=~OFmbbX=N-sd^hN@k6jK0fkDil4^5CUQvdo!u>dN-SmJJRbCMKl_MpnIKCiJs&V zOIbc&4JGHSzMB(H^q?H3^p$d%pr0I>^7$!;AO~Bn(qZU$;=K9^50tV4ejL57Qbw4ah)(^HFkfNzixh${2!A5z{*0g3P|9-5cmZ! zaiDT7HO%m+Dt%rQ^n=~A;xN@Tq2El?@X#5FtKU9Q^2-ElDUOYSf|y6EpHhwqkIhXj zTo90~fF5O>5B9%kisP#W$ZG?O3(OaeH~LjTC>= zuXGv!z+3I)0^&^B#_?RR%EOhYSIEPKNT{j+afAKY)}J&W7rvJm#ez>oIHLj+@coqV zW;8g1(wKQ-Z<8AHTmj{Txu|d?W%VO&2}1z`mJVCVru zH#aX~?pb%qZZ(5Vce%Ad!Ni6W;KN?S0K3JmN>#7bY`1DTbZNAdQcQWkl(J7j_n%(~ zV`|%;Z>uJ(zZJ_-F87Mp9YCo1hP{qrD(?U={p215jb1G^to533pIYq6g!+RbFooKe zbuW3Lnq*;qez^Ueh}GQBn0+8?8U^D6OUg7wl(>x@*Qvh}r+Xtb2^$ouPN_`F_!Jz} z4=DqD6B&znQ*F)9U#Fucuc=AyoB*M;U&{@B zzkl(i&;w(ZiSXPDe3@;0hE*5>F;{CZJTYa8-^nlc3jd7nZo<2q_!l0-a4M~WdC7t= zLUr+uWQf<}QePWQ|2gvoBw0dWzxGhRajx~dX<4HEI`_9+kG6MqJW)Cz(Yr+Xj8{6B z)_9&A|O>&a2+62&>w&{Q;K8?}lEG zkZyICHIGg<2VK$N3_3AuePKxE%D4eLndZ)u0271?_`ZzfSGe03jlz8~q*So@=EW8v z-Y>J9_-}}Cui$iGUZ~!&w+apU`cFI?y4k4WOUZCCLFDVz0yRw8))&KfYqj1m0^!7X zoJP#WA&b94bbaFxbwQf(okn|vzqL+_-D)6NE8RbW0+{m(66uoZy-dln`ezFv5#3{f zZGR{2{Mt?fsW|6pyOWxN4=JmvLKghRPh(An~u$1s>xSi=?En zjo{>EM)mi}Oj*Sigwt@W(&Z84wXxUIMZhq^KlF-+UXo(dzaF`%q6}Kv0~FBvJ5X`G z#-&;m(xF5qO4jkkIPu;)ZZVh=0gX9hmk$yCeVHxPD1afY2EOXZeczbXj8mb=u z9lZM2Q=mEmFia>WR^-ptxAh#q3d^ohvEoG*aL@QgUxeKJhvyAAk(ML?C#(Ex4|EQ? z)FJ@H4|auhDAN*)z5Lx|$0+)wBU;^tjdj%q0Qpr!HbBYg)U8MeZZxwwEK7d%81&p~5)zEZ7=GWZBocoWY!)5v)Qa>%HuM#(5~Vi86qxR}sqrJ~E!y zJxCNnk#Q)`kVy6#+nfAv|N3hMpJ4U7+id~|JJ8g&J}?>pc(ClG_eJ$zu>jEAJw>C@ z#o$u~Q4rUK0G#y911^`X9}J^FeKMs?{&&3xkfOU2YydUFtSk|dhc;Q`+Q+c^Z+rsW zN5T~eRdkn8{|vPd0TV;wq#WR*`}ztp+F{WM%86$u$v~mqKzax9Ezw{-zmXD#lL z#+F~PX_V{$kY}XMT`l+&&MZ9u%(u9KaH*~8QQ2kkt4{z(11!w$?+r48+PfCP#@6_% z|Fao6lh}89{-=kqCKodH;@9IN3|C2LM)SV`^u3Ir%H^vQVWrBrLeQRL^xq!~Gj$!NhC6h<(uurRHR28LKQ&%T7-1uz zrkt>!!m5N>LXaD-B3i17o z%eo~3oay}UlC6IpCyb^I!sh`fCPleLM9Lrj>#Yx=v~u;9WQXr13f8NMs33l7p1uKi z_OL}>^@8mA*|K(RA($7{0Y`ro!xotjHAm1o)}i;3_NOH%f7#Xi##QLnKx90~Biina zRY1gT2_k9{G!kv~MH=7)JZ)OK@0eZEbtY@fRaAZjG7$lql}vEdc@hEHAN=>v@gGkM z;A{UW00b}p0V4gE8@C=L0%){r!t2E9?`#aH;7h>7)NjCg=ewNl;4(l(09cZK)BR{ZVQs`BuNb`Rp{GT%C>Yv#w_ z{ApD>ZVtf>R>l2V6K#_qrCI~JySRF@fpi~O2pw4^-+XBTP>rawX-e;WXt$ud(fj`~ zHu8FHwDe#!6oS-l`*lYNHjY;&f4PR%1a#PYvlTpFRs%fnE3D>ZwmXgYf4##{bi5bK z^dCM&kSO0No&p6Q<>3)fQMI_!ekGn;SQr6Qo|X{j``32Ts(Nn#?D}9E1#cWoGzlC` z(6&S{{VDxsSX;AJ(8A*M>w3Zv6qV&qH~l-(B;3-{mTlh+WM?9h_cpA-fj0-ANiJ3hbg!0d zP}KRKufJ(wA-8gU+t|UkyBvyqX?d^b{`?t*$n}fk!P?buktErl%#vulfu1orZsO0yUxgZjag;n4BzzXcsUgVgyC{6aW$pFh`k~_(RLF zT$64R8~YzjjIOqebUdat&rV~hB4u%V0mEbQDg$vOImI!4n1CTM#lvnBG4v@`hKS^U zKJVWHffwcC((DiAj`Qt=+9!VbN&P-x!^!V+;3_`^No>JmRGR@AF4z*e3XrL4Sunu> z$=ho6&CRpc%4Q??p>5Fak4}M?jOC9|y1zvW$kzW*);H)TazS1pk5d=Xuniy1lxIQt zR|^n3E!RxJK4e8fjNs;KQHR$-CCr?W(ho$@?jQgAg9Q(2`I;!!9mvsi(dZav&)uGh zMuW!$_3#FHd>4AlH9iccBO$z5QQFtBFo_0;Cpv+4(8kg~&XBwiXjH8tn8Zs!$Q|)ITQM93l%N?{r4YjpHdb4Y0ttsS z(1EN00az(eSl7k#>6lDMQ~ezTB>x`xU$9eQ}be4>~)*ml6)q>slIv zQx7&6$WK6y3{1k$F#J!y|Fj-0w|_3ZkZ8J^mHc|!IQCCh8t~8A^p^$C`S<>^fc>_6 z#9Q1v%x~@z|A+Gn44BPvDg6S*1EGlIcCfzor!9G@H+vMMOA~K!hSvZbUjR5~-E55% z5O^XBfgG%7!fk~=8Sqq-B>?Sn0;KdPKy9ocwyz^qc9i38;#AiKrT3vf)N%}bxJ6U@ zE$0+cV*iH^0647q0tBtaz++2A{sPnte^;&k5f`^#XfF_-aA`c>FRdL0(k5eODTjQJ zk%sfU_IOW#Q1;c|;kqQzHtR6$MkuuBR_{NBV!hSK_m^U3t+d_~-p9RZfW)BuTXk5e z^rOHO)qi{lF4Nrdg{voELT~mDNW!n-4bJxYM)kPv5W@mSkM@ltDjk+g_;pLZLreD3 z8t({HKYNs|u{SDaJu&uQ6HJgk85%A#2f|c;LQ2rhRd!NHuL@d>mVBGk(uDu@{P_=0 zfDy(=BMeS2S$lw3SCclW8Gl3n70?W4Ks>S2?^~Q;NKO_|PFW@3;YI**{}QlR@9({r z=&>FJ5qzV36U6UQkTyraOz3m@i_;=XUXMm{@gX29c*!j7v^yHQ9P!sX;y?WMXjR0| zN|2?_WTu0_D|C~u&G@QU{#)*B_faJM@xu21HXD-s>k-N6HsfN zl{l^p6ho~AWFFuS2J@_T7kf%d#{TuLVjV#mL#qkq4kSTcn; z=a~@B<0H06i3XtYk&~~LBOxV)!jIR<ecnZNDh>lcXcQTh`2!WBK;Ap*o#wRSnSVST40g?wJ682|y18T0=0bbdQZ$Cn;lOua93jTgQOPD>Pt)xfuzGg z-Sez1LEs<*tX8_qWvoP^mXQHV_WM6S$(UK*m{9e zy`10-yfv+zK@T$$&8mFrgdu^0K{@FBdAgX$kbyO}(dzkEW7XN;*kcPDtAC$SU(YCH z)l|d0QRipkVMA7vL2|3x$i4B53Z)MTctGy<|IV~n_~hBT^Zw~h7v}}3!*buRNKivp zP6siqLRB;*O)RsWT?PIfg>5$~+d-r9474P7U-LIDK(C00(KDQCsmZQR(@g0@rX4Gi zhY%~Fc>~4c_8j8~9^(J~91yfRp+RQ%`M27j4yGuAE;?w`Ek&Mpeg|lb>-5xPZGAKF z5vPe-4Dgi)!|41`e*NKCEK+{n|GsL_Yil#kiR?ah6d&oQdF4J!VG$AS1cJB3RJR0Z zsNFHU!iwXpuf&$D?!6_(DbUX-$-*?*MJJ(#8yzFoieFSx0W?M7m7(U@Lv&2jcD5ZC z&(+rS*36t258F~(Gjd*-R3|0{j3I!Gic6Y4OmH)XnlKn zy%dJsMzmRr2DaV}I{U16`XJq%0~zPP%l=BNzzgvy7Z&q`L!J7ZC_4QKZE$OVI)IM5 zOJJhOy9dD|bjgIW*5EnF_H@6_Y7T~kqUHhBqRFCSrZ5kttGsUM3`T&-_LBXzQO*k* ztX(YE@6GvFjgZ^M%5M-I2AzhC2`hn|4E&|n@o8y!=;gOC11SOm0Q@`B~p}osIyzf@i;|lK~~gZ%{QfQetHR@Vz2%A$j@uG}Us06lE1W^I!Gz$R9sD z-d!3SVi#9gsmb+1|J$ES>>+>ntDetI0C1g=gTn3;>F2rSDU0 zf3Ci+88(P|G%4KVy|7uR^_|W07a*P3k)j`EMBLJ~2Vi7P$}M*hX=z=nj;kaGG6Di! zgEV~%plsN8R5?4|+dJoC!yel8<)EF~_)CX?j<&jb0aE-8>S7A<-5VecP#d%{d=-n2Wr~ z1MX1#JuiB?U#)L>tJQ5lMRm?RUiAF_p1(?MfK5D23gKiRHSdukiUjXAq6I5bN6sw+wl3Uww`> zhcidON6N24sn7EF3Q4s*{sRsZZ$+znIE)Qg8(K|mqGHjuV3!nPd3ykoj6jA#peWgD z+an7t^oyz^!rx;5*IVsfa1OY)fyjo^V6ILJApFMH{wl4k>tb#aQ-3 zO6sKJ#UffWxU9LXF^@kna3{p6jCgzGg?ZY8m3k1Z2^KKFc)R^Gof?VU%5An!URFU* zJYdoO`efQsbn#`$kO}DuK#2Ok>1xlTEw%EC-vE_7lXcNML1RA#BH`#VK-m5I)=CIy zb*ic-!IbdVY^vd#DKO8~Ph(RlzLfZeo_lX0W6K&L2MDzQ;Eo4D%<;5FADN3m`2i?{ z{%Y)lPmm6Wpegn!%hQ}r&yh6h{(!evPY6Fm6F8(($Q>F@hG3QR23?5Ush8yHJ%5kW&Z=hH4@dF-Yi5oD=QO)#MCADQ>~4tB;pN9)uow|bCBb=jF(-# zr|Nx#*)kNMzkk4zxjI$v@0n5n0`Lf^8`jB{g~y)LXfa(1IGoZ z(vQtB_+YZc!~&AI)mA|{O;M#818ZW}G%vOo)cTW2-O$)&+VWE<$^dUI6sYu$ zz)}}r4hdBtpx1Gf^HAP>`W2A>m@auUznXf;7Xq5lqI~)51aFILR2`;OG?m&Cg$s(@ zPu7N!!00PVlf4SZ&!iZb2W%Pi4xltqXEvbpJI1*UiuX*)`?qE5#dv7U@YcjFFX`QH zz(l)n1agsbY0C_Xn78Kx(yRR({Q4(2n5znWt$54g4^i_Pn)ple576-_6~Ah+;;wIQ zXia=buNVN%^}7eIv~>AKruFL)y2S!#C z7XCTI;a5bOZFqr!0P>WD-8UGPNAVd|mETNYDu(TNh*gX;j=!lB z2qE_UhlmD6jp{BjN!X+_C|NuR<)Xw+rP}DDbd3<#&0r@wV30$TLZzH~<=6ahMkak-FCVgvK6r5UD|00%= zFx0bSfX-Wf`Bt*+znm;?y&l;AMkA!-9xbu(3wW>s;EMIj)vm^X^V-1w7AOCe&%-m& zL~OwqTn>qe?Lm1B=HHV@4D@f5&rzR%Abz$BEiVBnLauf~8c@345}bH_y( zT{O+s{xQK|S@s_El1rgNp4zp~28>{n_#a-{g=S+?m(ldIF zY#<7;Q;nmN^CLnJpm1=Pz|lT1egn&)<_T#sy6(zNKab-CYrW#6*~*$xnJCPa1#SRO zFxbxaKE`F^A>1VG2;hqzI!&?C#&ep`ZQg!+?3K0Jb1-KBsv?IAOe|AJH3a0JS=Dt?lU_^Q04 z)QpW(EH)7&`O;u)kHpf4*i$2&CNN4*yL(*+LxLZZ?ofeD=@ZTnMk4-J^FWFKW+*TY zTzH@670i@Cr5UI$fX73lr0MArxyKi_jY}!$Qt+W&Nd3CFxOllLW*e_dAWpe1Q))sI z6aWv~*!GZnA7JzMj!$&##VxL~vZ){j^9{Fi?` zm%#6~q}J3NqK7N8S)tz%Ja78k1dA<6(5W91q~9u|W59K6#h(`_Ci@KTB}X+`y!bLV zF}l8qQpn>&Zwn*NmZ{tWIn5jRO@-+dN+>RxIKmJa6wi0=huKOW2I$suYzHbF zEiC#!zwQ6@h_hU?@0p#Qon;t*Nj{jmdfFI7F1r2rum>M3df47Wh0*i#NlQzYtM)x} zdH#R{6uPE+D}(oM(22flZ@+8t`$Gt5U)qzztd=hr{EmIy{Eo4pZTR&O%b~0ryASD) zvsBwij+vn9*E1Tx{<>OPTA)4y4Yk){4n3iQ#U(%J1+hNzT&1JC#TUiiABM$tgWDuF zbMo#JjPBMe$?*Ayn52L>P9;r2k9^1@3+l7C;m-mTBq{(D31#)yDbmjIKa~%KyTPvn z1<5^7HvzUYT956OFb4Q`853SmC@!+b1S?!B*&EP3slcj5u^13Ral|_Z?lkVGJqose zcviFGuVEl~6IiN=LnYBjQbQOn1ndA;wcw55Za5Wy2X2$X`>~f4;hrigRBh9rAAeyF zqN3B_i$ZMq=5XzL(Z3~b%|z*7g{elvKEqY+L6{N`44>JB3N=*#N(sP3%D9{=5;dzy z$qoF94W62n^+N~91EB2fpv3faN_{2XRusw)_tjvom8|>zcYr4JhT+ePYCI75ZZ$d( zRu7|_eff~6iVzGObd~bLh+8d+>+Xo2QM95qCaNLN5uyS`l)#T6i4ec^lE56I#S7}U zAA!taJD#8C+Q^{GQr7Dx(^8NtsXo-EJr)QJ4cK^(s?B)aB=-qqQg#pf`LnN>yV$HZ zM+|Cq21Mx2ZmuZuOdy(eQ4W$@B)!_=;!>ZZVq(BtmPCF4-=cKHcO-BOQlcQ0y5W$+ zO~dHU#Rh6Du_}3(6XNPey-i~jvz{-S5iMjQ7~Z%o$eqKG12@r)byEhl8~H)4MvBx# zo2(OQbRqvz)5kB+IL;!%^v zg2k8HLz95q96D3s(~bOCxF`m{8*Pe@$M$NIF*~ZnPgns@G_vpw`qrflvI*uYWtRjH z0MLj2u)Da5IOs!=+q~W z496cnEC-Nk|7VPd2z{?78t31C_l5Sh^73;=_HD!BS8wGnuZW0YbBhQjHle5Ia6kNFYo$#(LFCq zNx#kUV`{qBRQas;I@5}a!K}M~g^TF~vx32xz;^aE8=3Pz5?fcI)<`B8uEn0*y27{a zGS_(2?5xJ8{i(@L6$xJk`D5hJ6a$_HeK`kSrv1rEVV!NITwL$|1iX)q>3IFpHmm&l z^*y{#WhX?oy6dK2w7vlWGquy zh}DCTh*@U15^Qj9b+{Lr=Uy2t&AZ?#Vue>64v7P%yl^ zjgdvI2^RpvrJX6#5PJv-XwS=jc+x(r1tOmzaAp^Fk^vvO>@h;9OecHeY>N#$ERZ1a;fjK2@-;C3N34y>fXEqeGqf^dLT$aZtUV$EazYER~j>++&g)sTv1~!kQQfB4L|L= zqGZ;@bhTRL3@FymFZS2*#*)5BQCPiH}tiB4G zz3gtp*PH7Zl{OH*FO;gaA23j>(7szRaw9gn8$$f)!*SkZWV{U7^Uim=HSQbE8z*vI zGJhuQ&7E|CZeXx-aH4nlx(((O3|X|8^RC+jOMfqT#yvZs6{x@`HYgKoz__zSoYGyp z*AIyL^Ks90`?Lk7{9m3|GzdI1Nsq;)>w^(#HSkBKc%w$SFUVj&Rp6A;tN9#;f`-wl zl7YM7ip?^|k4tb*yEN3J4444`H4Z#e-*ga?Dx8n(-Z4CWShTi4TSId*$MY9Tu%JWf znX>q4^;J{*VT$v{7SN%IaWHl*2Fl$doNtMOsAI1iWqI-%mt>7O7elt1PGhU>)9#VA zTIN2SHCmuogyGlo^_{X4`62%PW0Z;Xx|3f$t!?1OoO>ZnR3gI{{-mi4kQ$ZuJ(nL@ z)sLIx23|=Z5G8e@ujKozGiUdzRkP8$0Cg!5K}PCJk6*j@&ARx9BI!4J68-H0i5CwF ze4fW>$Tj+)4oIjW!AylkZh6*6PKi3?n~Eoxf*0EJm+QIdt*vdtbiZHtAHl9En#0c) z)pVJ(^MeVT*KP}Z7uAV*^y?KjseWpy5$_oUVFV#=2=63&{QUhmyN#F&GRoBze43FL zoLVf%>GRX1=7(Oz-7XnJoVyO?8d+3h@dRftySu*gR(GfI&yn=?FJW+`9`m`AITV}byy4T2|oT?7M@YSKn`0bI}9^RX_a6d zl-zJT7Jbu^iaEHnezy55BT-~Sj>qZpx_tsn8{%P0;+4Z4BMs+vxZcMI>9gdH!=VJYeaEj$oT%Y6P!BRTG-Kn&) z1^jf1uo5tNl}mV|3IccFx8oP+92&=>uPx2+mE}1!X4~fBVpoK{x=l^t$+5ll3youF z!{3p`il0XM$9g>;U6`}sCz8VYy5dc}huLsEEd##Kvu-5aDSH)JAH+zw2rF-U^dM1) zY%$U9iby!~DNKsvTSnxqSUH{|#FP4kT{gzx6590aw0@Ze#{2XmkKMJHig=rkuBIha zpg&QbCFphVx6>_!AykgE;NlB?5)E>D;_WfEhT;+Mg*@ZPU5w1>O?5_!KgfWw43KZLoVkI_4J zHI#DEa=A@6ehbLOl?zqJsx4(yEwak;%VqD%4~-e6{d~W=bV)wctD7`oYvho4*SgHS zOf&zRL+Xvfklku|hG_WfM2g?F{i&q*OiT{`a!Hi&a)tLlMp+`|a9(}Yi}c#viOKS4 zrH{Yy`~E)leg3Ha?hB(6g{9!F1eV{yHToY!_6Fh#AAKP&d6U~#I&iV6J1}kYbXwYf zja>4UAW^PAPV76Cp9DJ3QO>)!>lzwP=T z(AUomD82UpW_};COaZ!RExq;YS(b)BENxp18K8A+@(k*1p;o*&>qtRy;gE zdzHM>$d;vbM2G2~*#V|k+_lvLV72Q&zS`w9@@$v`=eENgg)7-4!>{Yp47f7g`W<}u z3m18#pSUg4h5L_R$Ma=)mSr77?rwCqd=VS=RNj5n6h5k9-nYOjY1>hTa9P2FA1d@x zmgoz{2;E9;7MA)c)}zRI-YtK$d{O+f;f75`gg9U$%?~mtJ#&`sM7aM}x^zW%NbW#- zcomS6>p)3So5fi#F=SjXhwHr<=w&(9o#o&+?)-akmgS&qfMKt69$#&-Yk}6U{R8;| z{S9Gk*xVnv52>~h)l#4s&`mEt8G8i#X}p?~j;^rKqfuy9?UqkdY07uS^`>SoV_dC} zyCYz!-lJBWCN?l2mYtX>yL9XRtL%EIEnCU#$mK83I@aO9J=p5phWVZ9c!Z4=d+^=9>yzK?Xn(EyKSP5`}G11Bs<=x2Hs4j+@A(AF$=ehsK@fAy&aR^1y1TyqZ20g#)IW{ zGdi>c=;lxcr|*C-U}gvLBC}2s%$b5r45hUoXSZwiM{HX4!@o}Lr-4QdAh6Y z^Qi#qN7A7)tmKIHa?3b%kO9f=iF^Lqs<2^QRUGs4uA|6>%pda8H^a%BGy8qEepEFj znopQgG;j*1N;;0em+)vBz|j}H5Mr2N+A{9YSu#F?YYYK>U`n#VR|kz-8%JV}GX>=o z-tP=ISWyo%*ttUtpAk@VxE3VORI_aUb*Quy4ZLXYI?+8w?VA!hO7JA++Iygcj z|6U;Q`KR*lkD9jBl+$E~^rm!saO57@61d=rwH>pkOlP`=HG(Vm+rjpu>Gj$-owH^wU#@o){3Uo~;}`tHmLlD^$-*_};a;df zEP1TM0WmIhoG(|R8&4c@DS?8k?IK@*1iBHQX}GyZq(IGwaqMmdnG45Ljv2XoL9c9P z-`mmO{q526lJ)@?qNyo-=S7ctg@LZ+_6LLa!}E9hO0JR}I@A+eUSqf7NYpUOAtrqH zggsOKnp>DL{ia8kPVz`IM=$;OUCX<6Kh+d^X2BMSET-RVY6*jHD;AlVuP)Tbd@$8; zEb!4K4adDi5cMvf2$v_gcWy#<`?~@I7OeTwe<_gtm1vh$WR2rR>&oxww;1IQSg(+w zxu#eKCSg$`47oy$dk3#HdV4TI|J+=9@p6hOCqDBpLj}rU{bvmI*mW;$MbzCl7;42!YuOv8)7T0-Uv9`|QW>Vmf@)K z2Ab)Tql7(C!&B_P!;nJXlWkKy@SV$sOiV?R>n{rgcTXXDh0-iN309W|n<;lLxyDx7 zs-_E$ez>+k=J4^pd9T2;sQdP0OIM=s6ALI+Znr59pIP5PZDf+}WOT*E3K`b+!G{UN z5^C~iYkF%`%=tJO2M;4}itR5>?)p~&^ZrFK284LRhMwiyi#8{+`JAc+iB)1vh>iZ*ojIT9bOk`9?^kN=J6~3=gr4S2? zzH>yFyWDj{tX1Np%x;fJGp-c!Mezcr`KJ9d(o#!;@%>erw8JK`pl%E8NnnycgfzWv z`NM;8G@_jT@-O|-fseAoW+nCSJ7wbX8iEpOmhJBlxIar(p9D;pU&+f5{wcg<$|oLJ z|I-IEZ`R zw%ZAT><3a$*_+Pk&WFNonAk+H{@Xb6aq?$;;_Y{b5(_0B=11j+ilV{vK6TI2T9*h$ z#Sg_7nnlx!<^1I{#ZN66(n9oaQkpX@v5o7I9Io%aU&{V$6URM^`(^I4*HewI8<`HH_6HN&TjFXr^gX0bvuYF% z7N_mX*Njf^PW^U)ev>cTBydqV{;A6U+#~B?TaeJE$KCBGM?6bK_AZt9 zb@;ycaano&jW^QOjrIqVHp&-(s(NAd?4z~G6}*#R|Hcr>#NZ7j`K_l57>Nf*g!TIY z_)vVvhecupiN%(tPDhWXP~lR)e$1)b5Vmw(3dqTS@?0tMrcFa*oMMY?Y5@!x-SmuQ z8WEM${j&uTlqgbnMUU_JumgK9#L+hXr=+Cq-E?|S z?koFw=O$w?b+XF*rJO)_QhWOCC8lw$8=cpfX~?i+%L;>@;sR^mC;~Dr@zbCw-XM~n zD$EnTl(+iSg=D$+^w*$gO*Y)Yl+i6uhq{rt1S3$W$+7E@hfE{*A)5?Rxp*- z=ve2)^wY&exlS?G;$NdKiVQzBhk2|!Z&JUho?2v*$EPCY5R;?ZdLe_QDK470X#RTj zz=XFsWBmIUgu=FQ@1D@MNYvDKl2n~r(TM0kxvn76OTu~GElYS;VloRtMcQo5lNTdp z^t}Y~Z5LRc-SZ~c8}gX)nqL_WEo3O5>&+nFyYJ-lp2wzDudBhw>s2Icj-KH_^Bsbj zCO=->d+Ac6dG|?42E8rxZ1>D0In!fYE>P#)s$$Y zfwm{D;AHIK-uJleraepK`lhl#Nyj5k#*57`y~s(kNZ!Hsd|MemZ&0SGNc0AG?f=Kq zdxo?5zW?L)o>j4D)!rlasw%CbEv-$gYKg6OjG(G^ZLw>#Z+q7YsfxXeSVa+0YQ~E7 zOF!S^|9|8-h{FT-eO>4InyYF5xokgUAknUKw;y{$)2}m6@f&Lfv?lmg&(q3_DI5g? zf0n;GEf`1G*519M^|bP4k5$(7ei?fA`02$NBm*B`**T>aQRs>|!3T-E2HShY4g2{N z88Z1raUZ?(WK>(CVjk@6dk>$M;fUp@|IFDFTu)W!DDY6!d~h^5QgC=78*?9Ujk zrw7hEoCSo@hdA}pK}tFNB7dDH(&co+h9&#7#IiIT z=xWB#3pE)uLwup3^V*ILC;W~ID{|Td`Q_k2R$fpasi+& z&yqsU@#U#eN7Id#*$MI0gTS>>`p6(`DlLDwuUx30y0m7M_s_t;s@+n0QX{0Kud|Y- z-GeQcn>2N11(^eG)CqRK*k#2BJPwO$Z9Iy;6nky{VFn5 zbyIp94?-Y+Eb!+NJfzkX1PU4{e9e^XpZZ-HD&Fb!%Dq%RKqtx@I8MOcw=MU~=f3wVPJl&LHnm={7ppl>)G>*aDZ$4Tmo0jPr#MUZRk-Tp(Nf4_4g{TP7y> zpHoS`8Fiklz^A^K(S#g*aalqoZ!+bOuO_5qPoLKLym2v#^LLp(QFbI(_eu><4HS!! z=@;X{nrJKIq{x4mYu{jvHP1Brb3AZ`h9zQ<4u{U7{wWjkGlO`OWU%QT)e9}E!T7&L zy?z!c1U6K!b$IU5*pCR2J4V+>rCCOPH$l_JN2KfUt8UJ+_O)ZMq?l{F*Y(d2)(!gv#*Mg?<~J`j zD0A59abJ`xw=mT~G0HFMeCjCC7`z)vzmW;6!}FOQ>2{rDjk;#vIxz|3FUhm(xO>1N zuChRDn1p8VdmaIDznfA)@NmV9SB~qeS;+~+xCPnSyELct=3bPGnjMKNWM9DP7+)(5 zcsy#r(ESgRYKc8uM2)vHq1@pvd!7Y`su%8#Lp5&R23vH193uSRXoV`kt`nII} z(dB{8qqu)}T+qIwYTrNSKtI4|%{fi$aH{+oMLrJJcex}q zF(B+>B6zKnqjrno5Rwyplxp_BzT&M2gR@8Y{n*Y(~(QrTN*lpOXU*@!qPy={`O6 zKhNFeKHo%GwU)1Wk>T75+D#<|Q1Bjnotao+F*XV!8-`1yYl!cQtp$uI_EFVAut^1$ zZuW(4w67J8siDmepd?YND!kv7d}+EKe;7=Pp`qs2F;8i&Tif8F76Hsl42b&>?VPa} z0chD3zIhSZ98Z~rC4G%y7EIWhO+349in8N+ zAVE?>QQr}ODJ!BCvYCa-sOLGiMkJ_z-ldT0u>EzE*(z_PJ!ho6nf|DSe|tV16ogqk zg-1z`zc(g@MMv8YZ*yHwCC9GHvwg-?tP%osLG~*av*W2LL_z6HYqlF_3{dlgI7#3~ z$$P0v`UnT?yYGSZcZh0}r@o5#QhiyqmzDpOFgSk3eBhXhE=@CX^)HdUwKBC2k+l`4 z*9mf`mMzBiR@DE-!zdod5BzWOS#u$0@UyPkF3X>u;Gyh92SwMgo`O0tWUfW<2)Bik zTy2V=6~cpbTDpn(3MV7DR;{S13CowrEcCi{$yBpoOUg!L9zQFpexBXoCm78rbr@= zSx0AFA3iKR8eqJC$jMdbq8#l(p|(P#7-Uj-7SNA zLCsOa)x%-E#PDudMF3g?JF#V2c-L;ITVX|Et7nBu<8H!c+OPHJBUT6ZfLD_t&UN;^ zS88}Xjb%H%NHdg;;|-G%q=_%B)=H>~GKEnX!{(oAOZgkdUg|J5Oa_aEQBS)gJ^=sJ z^`j@1lsJOi2bg+B>9;C_Bw@;mQs=%`-oaXXyM+Y#JHozx+yi{r0E`P^n>&%p%f9)?H0p8iGXS|up zB6|^GQwHQM=ZRrMS;9X?I!7_VD8#t!JQ))kLfGF7sExrxe*BnE^P_UQ-UK_*|8)_g`ZT zL5e)3AAkyosQr_q`vysXwzPG;(_>;0S{av(#H9IBQ4g{&raYZ86Pv2)A|` zM>pD|Ss?n2NM-+FbQIpOSwBxX$FXvb1lBT`CMdax{d_ygDV$9Y0b4Z{D@z1X9o_4n zgP%6LdmOgUb=8C{F8-Q@H}1?{d5cT;e{->2KcixJ03Q*Ry0$l1(vd<{isN5ax8+*Y z{dg#=_`YXp%2Q)Mc4ka=&iDd3uPzF2gQn{hyDvun`IY8uH_=Wsj?a(@7ZU{r z^*hLeBl3iGT$$TfYSae5ghoVL$<1&8=8}zco%}c4|7}RQCJ#ue^rg^~Y%|`ye)Q0k zbe2s20T}!UfAcLw{zWocZn#8ATv3HGX7`107^6IaD-Knw%ZZ{c>P4%n}z~uK2)KoJ< zn&S|Z(xKMvZb`od{4R3hY1Y#DagX;a#S!ROLuo8y0KasK&E50|`xt_*p5{qC|LG;v z6|<&Qy8`Ln=l(#Jx6|G`RM@Svzx1IGfaK(2TBg!>FQ%Dr z_(Vl={Ht}eb$+D8*YZj$lUKXt72AgH7AZN^9aDgc>vY}LITZg#FW5Tz0kcRtoUj-D zzCKGllQ|_u@!I!6aeW)n%R|m)4_E8guu&G2A?|T9>1{2#-Ky^J**f3#=h$n-`k!PA zGZVdh2{Zo~BiOzJxcoDesgG*HaHY?avc>Gmp~-*&tgU1-Oq3NvPfYb-gv4?7K_X=w z;o!shqA1?uT~|611?>P_nl=*D&nQpHQuO>8AgyU(CaOtJUu84qo_*1aN2(&7kJj+k z0oQm0MH@&%X=me48qI*U)!AaV)+wh*nrYh04T4z6?C%9}{lZ^46WLdvin%HW2wCQb zvoui6USrBzO>=DK`eMRS4MU6H^EyexCrmcW#dHSD+LqS?oKLTQ`grQrF-vYZGro7i zk9KdnVdDG{r7@Gi5|N2y(;T%V_$*5*01Ny-)>@>H*q_NCH$cviV%H8h_a7Voja%O_M+;@wj=E=ac ziq+Qw+#Lv{*@Vj=9tIo8t(FFO2dAcfU^(Ncz@$xYfi;xyw7^2;BT-%f^KjFmRJd`H z(+4GQs`tvI7N0b`7}P#&;OWaG+$n>C-d=Kuh6{r83))~e9jnFz4ozW!c(qgIP%@Ot z;V(1k{!(hcp}g;XEX8&MjxA*X&%Z6r7k!+cHR(u)$?XoO3vc|aKFG(qz!CR)Js%ok7ua!mkTXzl~BrKTmOO12RS6=<4~E{Q7;}Yd9$R^e_$@h0cC;^ zk;BOG0a1;GWA{E7Ukl#z&7|UKh=&MhXES!cDZ7#eeTRr-Q{u0C{Wlh{^#mazq>Mh) zs2$`XO~TWQCela>dQU4EIQcbrKnR(g#VivE`V+ z!%t`eH2rP>5>p*z5&UhT$OP%^tH(h&qMCSV5DP`j?UN4o!<1t7kl; z;9WANcxJ#WTFJ&|XRIVowbWE+2=j;{IRFsxFFB>uzaK=H`*Dy_ir(vgHZhLpFM9Uy zB8dG0+|LUC@vzgz4t$wKd&tkv+%x_c!drN!!=lk$Y_1md6GW(&&y$*kyiS`&z(|ie zr(gHII|JULZE+O35q#Xc6N3H!h}7ljg43owe7{8S8Qypg=PCwIj{UBlYg4}Nr}Vj2 zOP_m?ZyfwY6C(@Bf9cV_&%Np2Eh|;FWJSC1GJLLZAVK5JiiNgi(dhE$>v@*(8szV9 z@=EK1<@y4$jl5DtF0*?s&y1Y@`f#^1CvH(a;wy%_b-7wUmvba-5;Igsi6KorHbO?TGo`N*L4f)%%=&q9?ld!&op zfCBMz)x?;kdGceZMZ7VRFjcc8!B%*G~?Zm2yRCBNFU9S>|7su(|W ze}7r32s9Ict-*7}E@|*^W{Wz?3PDKZfC_*mAI3NKoH5Dw#7gh6uGknf&xewQ60iK= zWK6anvmYChk|}WFL|GP*13=rw7tjmm?e7d|PV6(@;tOKP(UK{h2}6SN-L8uL(BgmE zfJDa=G8WGTFa8$Aw$&dR=Lq^1TB+-_xYwh}(9x)d_+5^!J2X9#W1l)g4zcJpD8+Wk zRynS_z~1CgDpUa0L&|SX{fF4tt+#^KY^wFq597v3K~!*2K;!PZkM>!W%xWY?^_qSC z{rpo6IIs3K$sDUkpxb9Go)ua5DNqDX+g)X}iNHQoQnYZtT9gDFl-gGB{S}_oKQuFX zdGZ9j-l{g=^M^R+9@*l(I|8iWAl8RUUU{1DIYsObr?-_?e^!@C@OL&hq2wfk3mUAvzA*aksK8ub-96pAs! zGbBY$8bDOmuk(DPw(>hK0mzSD&8R=`ST^*_-=%j9#u{%#^3pTitxaflGq63%SgGIZ5DT~3RHYApwxoEmdgPpV#z zs(XsjiMy^1{H!Bn8^DonmhLHzJ^AA4`??qhR4;A(_mta6&cCwM=${gU>?eoRGjs4D zgFBZ0F}MCWo~iLcA0tZNqq{k0e02%x_2&JOKT*@IOuMl`G4=7)Uf+qV5Zu-86RBU3 zK~Pr=-tnCWAL))iekwGUJy{TlTnEF`Km$`t4~lp%32<%Q1~ty)NMsi~$99Gb&i50@ z5t{eMiK;=dpW)4;0P#SJ2rg(IW1SJ(>gQ}@8uys^>ebH^#MD)Q7O>aq{*PDUv;mK4oP*w=`C)Hv+V!w)UB- z5=mgaN2~?g{dxvd=z|7ya`RG60LS^fFI2nx_2WUlh@Jc5O}_6%pRwi$MEE@=SHcYl zv_Qk^X^EmyRD?eUCh#im#(eor(-(E|FS&`xQ=IVs5xY7sA?XyIe>GMuJ$LkeYhJC< zd)AWXE8Kov-rbEJUICL>tvw`f`9A1p^erRvFn$`WHbfSkNa*tpXEiwN1{$1F>&+wj z>ITQBv4r^jLzq+%CYf-EyTfgO||8FXtdpf?sHxMoS_YzauI)_3$UoL*H>z2H1hnPLP4{? z)%TCstyYI8iFoUH|111sRV6|4nsV%Q_#qYgM57FFR>GOaZ7 zJYNkAw=ZBAvv(`OGK}aL(7TLMmS9DCy+SHqTb)&JF+hQqB@w)IR5T z!tq)>#lpeYN)^?xfHm6-0|fdJY1LFSTqB9pxWlE*bBhr)P}tz%IQo7(+AL^!$9Ask zAVz8zkB4l%8xI=uZO`i3bzc5A^#`(X5M1o$Y<&w}YG#jNC$vEpvG zhP^v8l074DxjFUYZ;6Px?j@zCjho6kR~CKv0=HJyxPM%KD%y3ZYSD(&kLKHmGUes* z@k{06FWJB-#^a!!DPQ1hnH)GajcguL#sps$0>=xQDzjQFtfVrGCOB6ai5vanTq`p zvkP8yExO@~E-1%rA`K*iTdz`P)10jQHE$K!{q8}NmZ7Y0hZj#3oPDMPQgOePAXeYwRv->mL$rYZA@x ztY~@dlx_p=lh)1=5$GsoyJOA7oX$~gTh9xum*|FBh28wKt#CC~_)=a*@=iKyILzws z^N1b(BIY$pSw-v4H(22GR{-0Nqs3VcEYy4bbE`J$Qid(b-7l?aY)(|W`c8yh7^=u{ z{@-fH8z&0I0>|WL?4Tfth(=jKeC1u4d!yLi5U$N6f|O!vlGj|Msw>zx_T=vA^UxT@ z!)z&bSn+PLF>u=DDiaj)u!gQ8F8-Q0_o^j9>AlpX%U0($MPiK3Je5(M{|wx*v^3u8{fB2)X{8tk9Cq zb)ZZQ3Hk0J+*>tW2E{)tvu{oCvedA%UuW=E(gVEE%GSM~BlZ<&&~?DrD@BXHUEm8U zP#zdS1r)@(3%T&lCfqvdXF@tkx>+c#*MDW_;g`d+YBYXTQJskN3|;4NUz-j^4n_cl zCSzwJ+M<=Ith9x~j?xO{xrQCSho%_tpdsM)_}VuG{y!M?{OChay_UB&lnFaX63Ft( zOl+)k#9C<4)&20O+p^xQY}x-?oKRPwo%|$xDOW{H8p&baKcswJ1}A+c4ag2?PmmI$ zjG%||jW}APMbI2XBo~9^+<~A)HRmx`K68f3);2Z#nCAie3NDp3tD#-cu49+3@c$5H zB8ra%+4cXgJ4pTVd}&LgHDB=jH%s@oe2=&J`dSIipXtL6qW<2M(5AuvhOnI^1yI)O zF|ap;n8%hR{B0h-GCOL~oQtHA^jITo_}>NsB|h|+HNk!14TBquiXB>L;}s~Wym(eg z{PoY0XkDr_IzqojfS!0DN*%*Z9HHy;W9BF2yIT#X95WB^y*!#VKU(5=CcaJ})!P=! zNLJm3ua2Zri4rK1+vK9E7|CEz1=WK$uVATPxgdfSd7%}@O4?;aUay#rUI=bh_O)8R z$tdj1fN_gZtC!WjGv>i27j)=#rO$trUMyLHk zMv^N(+!EhVC4I6`N&3!FEOaU!oZDGvyg@iLk=o~;=5>x*qE*e!Rja1nYcmvKVWVFwUOw_&>HDid4Rf}ZvTEnE zQIQl7I7&Gt?e^8M(Zv0B9eNC_>RRrmBWZXtd_;mfSYe&n%zAR8l&kAE3Ey6ZBYIyf zvJ(}NYw&0NUHcfr^#uW2F!HKo)zi`Fq82;cg-49rd%ArJU0ylXaN@7g5%P|GSq+x6 zUO0^9=&~RArhD!Y<9M#v)11$%5?=YCOxn}G5rp=uzDU{f8aRRoo>R)?1;QMg-x0{r~5f)%VT=e5hGyh*grw*X^PrSI-{}Nqf>6yyg;t z@%;*O*3-Zb=4Ns=#|8;AR{#}^yEwe@6luKHOCgGlGwzL?oaRMg^ix0#*0GiYeln}J zG4@9?@nF0n;u(G~&rPMGhz>F;x_r+B-hd8{;wyBP^1YF4?p`uqy+xx*c{*4!@J_h5 z85Pg3{S+IlZB9jch*fJjumIFt!X0?>2SFG_9;%d z%O`h397ckTxw$CiF{0i{7Ma3F{F^yRss}4nhB_0IewuK5*d~8@LS3$Hl{tFaX6@iU z+X<~2y^jxEtJ;fZGJ1217i9igCM4&auOc2u_* zi4*12pP(r<=3~}4d(?NM+eq_e_b&2CFD{pzm!$WCB+XSY2yv_DD~5g9M(4IXuP+ms z0XYLhlJkp8Xy9jGMA;u!^&%^D4LWzqIX2TWxf&XF$-W}NsUorQomvW9JL3(Nx0{6W z!8B^QDuzCKuuUia)qh$s-ay$1X1KUWY1e-!b6NBy*`k!I+?@#) z6}0cHTDjkhI$n8}hf&L3oUrm`Ip)?FL*<#Zjre*=AFbbFc<{63ie%;}M`OJ7u;grK zAmd0RKBDia1?WTvmeB8sUp1|;{#^hG$3x*Q*ROAGFS~Cqm$UG-vjAbo8%`22((tZX z3O7k<018Tfs@hetRlwF*tX*eWM}{9QJXhsno48}Q{M%tB5sjH_Re~tZoy))%9JzR? zGjGq`GsH@D`RuWA@QY)s|G_f~wLymXTDYxdt?-@wP-czcWBgFNhu7e`kH_~?KogUy zHb@cgRtczv8c_}p#2NaDD&qDT06!!Oq;K*xlVF4TsdqSa#y$_PNOsdcrjFDgORlVG z|BqHs5M|q84_){B@HGuKRC~n!J%CJrTAoaanClZU+xH7vWN{*=S#|z^HA7L{HiO@L zjc@%cesA$Dns0E=h4a0b->aC&at_tsYC2?y8-9@Jp}gAOa*Uh{@cUGikyFzaPVrFU zg?4V9Ds?gd%ftYrb00H$BXk{C&3f;(ctS3NC9B?psr!X7aHI5utyI{CQ!Ip3_+CPu zfUpf69_g|@spSi;sQMEXVFtjJ^6=V5y~VV;+qK-x3iS^Fy!OSrgv|p3?&OjBX@~eB zvgM=IP?WgGnY)Z#e;277Y1CVe$F2cjBqyu?c($? z{^4b-J(g!yVFzSBQ%FY!qaGyDYgEWsb%YWAf zQb_5alxB89e>!dysL^D&<{LauPvfT~(!k4Jg zpQVKUTHR(DrKEPp3bWCl5aIkw=#O7ExLhU*T`?S;;V+E7Y#!OyVn{6(YozFSZpTDi z85zQZ8N2DMflg=O7!v0`GDuww?YqR!L)2SgIu;n<+4i?5e?f|wBBj3D16 z&?U+8mmYE4gHy*~{TsB<-eIe9)CJPimZ;VlzV@l&VlNvWwPMik#?1`&NY`jc+_3TY z&fKRCWc5#ll*F6tIv~{MIeF;&^%?+9_W}<^B{_}q!EN~BnASGgmC>u$L0swhu2>BSK=K2}<+&TCQg{NbPL#Q}`f3ZP>aw7gW5K%mTa(-%-UserzExHk zl0ee;B+SSY&yaTOvaD{0t40sI@k!IL75--{B2lwTiJM@dUnU9Uw`6umPj;9p3F6TM z@i+Jhn#n5O!FJ#6yI9~j-76!OsIhrD5%1h6GxhnIKOq8hMyh}PdUq}v%(86!E#mqA zl{s@1&jc)_unJC-(VYDcx{SYfh9}C}@j=hs9rzr$Fr8-${Q3q9WCGE-bJRZ885#HD zt)IWvTTyvAKzVn_g~*erY=m0j8w$9xx~NA3u*wQ6hj7+9E}T= zj`j{0FSZI-jEH2OA3QBwdMKQY$vv>FI?+^@PxYEqK!gD(w-fVcKD?@pO3ON8PGJP> zV(#as+jSLG2?}AMza}VE49&#Rd|V84-UrCYgUJ?uAO6>v=*J}>aoKf!YB)B3WgxKz zrbn^iba}*YaywtuX=hk?P2!Vc zW`}w+Del@QW0*P^6Li97TbrCUzjgZG?PzMf9h_IcnM)o9uyUXx4J6x26ZtkkF5Wyu zDSt&@!%*SZbAh_y15a;oZMTDqlNk8f&>h@d_7PvRlkozqt1?E6Yu zM-Hkes<-vbl)~KPZLV|dVt;-{eo{BRJ+^`D-{O{6z^#3acn@MxA|H}}g{9PUL>f5B z;ZK@!pFVgZYJn08?lbN1m`viHAm5;dyeRmO+G_1QREQTAa$aa#A7{j4r zh;Z_lq-VHiw*s3R=)XQXQ|o%!lG{J=^7wgBj#~pWG&RaQy?P>PA?>ON9r;XUJs?>7 z{e5HUa@c<3jNkXrNi|3A4VSS&WJrNAD9IsjGI@)!`sk!8$f(Omr!9TjbLdq|H2rVx z7Lq}9h6kD%Sp)(mAi!P9;iGFO!#Z@zrHgon=HgO=?z}JCC0=NAlbHIi=O|(M6Vd!>E5B)mxrwlu1zE@qY z_VRdY2zB~pm1({4;zK>muz++u>7w8wCMYVAw7}{v&3ZuS%rf35wKs4tc^A)HR5Y6? z6&yqhEqSSLiWtAG@Y%&9QKOm)JZt;xjwDEYLgkt}w`+oBweA*Zy@#^|J7N8a+tMrl zzT2_hjgY#$;@4?{3Nq(Iw{t;_ZJEM1Pb)^=E<66>hd_!nS15r(DBVseyvXm|3{J

    From 3437a5b1b6e6aa927add2720efceb5cc856694f9 Mon Sep 17 00:00:00 2001 From: Connor Holmes Date: Fri, 3 Nov 2023 17:48:29 -0700 Subject: [PATCH 08/13] Update minor CUDA version compatibility. (#4613) Co-authored-by: Jeff Rasley --- op_builder/builder.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/op_builder/builder.py b/op_builder/builder.py index 163463c28c47..cd2f71ea560c 100644 --- a/op_builder/builder.py +++ b/op_builder/builder.py @@ -70,7 +70,7 @@ def get_default_compute_capabilities(): "10.2", ], 11: ["11.0", "11.1", "11.2", "11.3", "11.4", "11.5", "11.6", "11.7", "11.8"], - 12: ["12.0", "12.1"], + 12: ["12.0", "12.1", "12.2", "12.3"], } From 016b796227692737dba157567af0c55128f7f0cc Mon Sep 17 00:00:00 2001 From: Michael Wyatt Date: Fri, 3 Nov 2023 20:27:47 -0700 Subject: [PATCH 09/13] Update version.txt after 0.12.1 release (#4615) **Auto-generated PR to update version.txt after a DeepSpeed release** Released version - 0.12.1 Author - @jeffra Co-authored-by: jeffra --- version.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/version.txt b/version.txt index 34a83616bb5a..26acbf080be0 100644 --- a/version.txt +++ b/version.txt @@ -1 +1 @@ -0.12.1 +0.12.2 From 4f7dd7214b1d81dbbdff826015a67accc10390d2 Mon Sep 17 00:00:00 2001 From: Jeff Rasley Date: Fri, 3 Nov 2023 20:47:51 -0700 Subject: [PATCH 10/13] allow cuda mismatch exceptions to be triggered --- op_builder/builder.py | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/op_builder/builder.py b/op_builder/builder.py index cd2f71ea560c..cf2d3311667e 100644 --- a/op_builder/builder.py +++ b/op_builder/builder.py @@ -462,11 +462,12 @@ def jit_load(self, verbose=True): raise RuntimeError(f"Unable to JIT load the {self.name} op due to ninja not being installed.") if isinstance(self, CUDAOpBuilder) and not self.is_rocm_pytorch(): - try: - assert_no_cuda_mismatch(self.name) - self.build_for_cpu = False - except BaseException: - self.build_for_cpu = True + #TODO(jeff): need to come back and fix cpu-only builds, this came in on #3085 but is hiding real user env issues (eg. torch cuda != sys cuda) + #try: + assert_no_cuda_mismatch(self.name) + self.build_for_cpu = False + #except BaseException: + # self.build_for_cpu = True self.jit_mode = True from torch.utils.cpp_extension import load From b51d15b37524106a58b2a8d93326e4f6afe81ccf Mon Sep 17 00:00:00 2001 From: Michael Wyatt Date: Fri, 3 Nov 2023 21:09:47 -0700 Subject: [PATCH 11/13] Update version.txt after 0.12.2 release (#4617) **Auto-generated PR to update version.txt after a DeepSpeed release** Released version - 0.12.2 Author - @jeffra Co-authored-by: jeffra --- version.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/version.txt b/version.txt index 26acbf080be0..aa22d3ce39b2 100644 --- a/version.txt +++ b/version.txt @@ -1 +1 @@ -0.12.2 +0.12.3 From d89027be61c0eaf1567332564e78a412d7c3c8eb Mon Sep 17 00:00:00 2001 From: Masahiro Tanaka <81312776+tohtana@users.noreply.github.com> Date: Sun, 5 Nov 2023 10:23:46 -0800 Subject: [PATCH 12/13] Fix figure in FlexGen blog (#4624) Fix the latency-throughput figure for 13B. --- blogs/deepspeed-fastgen/README.md | 2 +- .../assets/images/throughput_latency_13B.png | Bin 339033 -> 0 bytes .../images/throughput_latency_13B_no_arrow.png | Bin 0 -> 176008 bytes 3 files changed, 1 insertion(+), 1 deletion(-) delete mode 100644 blogs/deepspeed-fastgen/assets/images/throughput_latency_13B.png create mode 100644 blogs/deepspeed-fastgen/assets/images/throughput_latency_13B_no_arrow.png diff --git a/blogs/deepspeed-fastgen/README.md b/blogs/deepspeed-fastgen/README.md index 6b440e35b675..b271c112a762 100644 --- a/blogs/deepspeed-fastgen/README.md +++ b/blogs/deepspeed-fastgen/README.md @@ -145,7 +145,7 @@ In this experiment, DeepSpeed-FastGen outperforms vLLM in both throughput and la

    -
    +
    *Figure 3: Throughput and latency of text generation using Llama 2 13B (A100-80GB GPU, no tensor parallelism). A normal distribution was applied to prompt and generation lengths with averages of 1200/2600 and 60/128, respectively, and a 30% variance*
    diff --git a/blogs/deepspeed-fastgen/assets/images/throughput_latency_13B.png b/blogs/deepspeed-fastgen/assets/images/throughput_latency_13B.png deleted file mode 100644 index d11c9bd1f6df0384e11e1f9bfee4d1727b346f4c..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 339033 zcmd>mWmr^y+pS19f^-YgDJ2a9BGRRZlr#t4Bcm= z{?Ge7?|I(u=Q`(ua?M;bd+%SZd)@0^_k?MvJ;leN!nt$j4*oMmd96ElaIo*(LDj~> z0N%-Lv;P46L2=T03c6D=K(hh7K(mrnmA!MPEC%<|1RZ#d?VzaVbmtCj$n8Irn3a8t zJ9lpKp2^GVxEpO}V8Uph*VW7)&Y_}#)Iqy}4<6tsk;pdYFdV;o61!!EPFh!R->>kc z4H@j+Mh+YC*I^JLG~~`k(&5~e$Y82ow|Ti`R^#C0MY`ejATOt9mZX5~i!`B%>w^-v zs%Z)n5Ef$~(SP`_Kzxcd>p#5z`?CcC=FDvW@oPk&5%x#K_y314Zr-QF#`(|pBXS9k zL<{;)PeKF=6?>xK|36$(KH~wC?Ei4d9JRQ(?Eid8EX~hdiW>jZ*tP#(8lI->>R^_A zDGaHEO!@DVky9M|&kW_*y{2(TcC${m_@N@Q0|?5muP&y+HM7Zfr2l>DD}@ol3e5As zYK<@=rmB>iNIPR0akiFK z6NF37qQ7`;x(F>e~ea=>n7Xk?1nlNA;l3)6?qLV+4aa=|>krxz(fFY}WAG<|{ zX!s_;^G#P|SC~-IaT>>1Kvcij?T|-~shj`Z?x*?N!aKY$qtBaZPg9 zG3%1MSp^vA8x#m|ctXWOP?VHgD;7`*B^XNmpA#t1YJXN!y$%=MEco=*k|oX9-+81! zvzs@q4%x$rxI7qCzvjT(_d?hjJ9QMP4wV{9Uu>xRzOl;kc10>nwR_G%)<`XA4g%;e zTX-6_%a=>PC&tfdEf(L;J<-lOPZYSe<*`b43^sJ?7upoqd>@OlyYV^82+qXN`TB8c zu&v`vt>YV#el(^fvlvNNPB}vg(Rb!UR}%Y|YrfZ3-{=SYKf`aX=5M$;cIdD7LY|x$ zy3^(jrmSiH8LPbLYb`ByDvYI{ZG_S@Qs+Z4_0M&d_l-h36bcJMwSHZps0o+Pcx)B7 z-fpSTH=x_YuDO=jrbtAAX2!<$H6@nK;|Hm(okh`4vIWV_%AabuRJAf?3jRTO__&S( zqH7aX<3?$hRYL8`XN`N`al3g{f9j;e(+`PxF)E5iwupxHcyd6b}{K2h*Pv@IDE%;9n-TCvK@xx0qa8H z0W~huQYzEf(KYsVNvQu&U=Je6g}$2mF+yb!vEF~^w6PU;eE^putWx({Bzb5XS~W(G z%_wjC8w?AaKp1pJ>0R;OuLp(`q3^BrzYZx9eY!7kCbaUT^d(bkGFCH_LSz?dztX;c zQvMg`ioQF_N6AY1Bh(mCPiY+GWlj19Gmp%J2D#ktb9D%ODlkg5WoCOtATY_>bh^kA zl@h^EynH?f>;}75fAntN9|5xTHdu{)U1_3*c!#V@oiXay#D8aNbx4az7A~JZbtjo1hTB{1MzO({Y-pUJUR(O}96b!AsN^gwq(7^N&LfT{jO{$a4wCiXvQN2J zC%W4=omo;pc6g0c-U?9Aihi!+b1^+%Z6hTQk(uUjabY>ni52GPuM}kK9Q3_DctW|E zYzjC2;-iMexQX#Qri}19A5A)}BfOP=+?}!u{!)++tE8f(fuE{oi{nMOBM>!-fTE0lf) zXD=f*;W;H4$`zZ=yx{7-5H|W-1bpG5-`5a6{k8#|;M2+K8?gw5A?Lc6gG@{6l#ZN$ z^QT@JBPIhJU{exz9Mdqk^|$oP@{jqjR%{djs{!Y^10}us8j6?xBDjb|N)N6z<%Wlm z)FK1iSFsk~1F;7xOvo&~6$a%2ByV-fk4YKiEQLLqEPv>GxkR5}uTQMRn@DWWW|;I| z&l55(vSFcMF&EF)_m!EqZxvVe42LlsH3o}oR@wKt+MZf!kWc)KPg8ktz~`^F zrH#U8h3;@=y%J$U*bJ@bcG<&HG?|MlWJ+^703Ss!-p^^^Xv5$6;GBMGs8V0vd9S4T`rO<$ z-4dCpO4vL4ZbxFO#=hC+22PAUe=&C=3}#&(MOkCmMZ?fedSDLICX~51`#z_`!j5#~ z{j$$5mrSQ<$Ln@^y%^MnTaL)#b9% zktm6rp$tDUX!1&}PoB)&QOA-}%@r}{!<|IpI`;8*+TG(_Q*N8zKuFzJuG>M6m72pe z&R-Zz;^7?Vg@hN2EHVuS!rpOX_9LX9X)E5jx!Uz@PTJ?i@SYJVcktqeFhxOnslAt( z=$|#+5&{5p2ocCGO{28^EnW3HuCS^NtfDm6Cy>V>%-3hQ^sPAjPSIB65PeBJ}apNU~;&_YP4R%Q$*NFP)(~%wv#ed z!pc~7#gIt9osVN51Brs5^QYg*n__{F8dx>OEH%n^DT+=l*UaIg)}uXZ_**988ZIUq(f<^_R(+uczTH z$k2@z(cxYJ4YHy4%(bP2ZD|h5YI7Bx-|q8u-mhSmGs>vq9D3+kK;VGLoQ)9_a{Cgk zai;AvZtUy*c2;ndh5fUJ5pEL1)fjXvQJDw#IhpgcFYio_4S`=zu1JK6B+kLA)5v#~ zc^ebAHAUIl;B{XrKr7UX-*I{II+73~!y2vz($J;y)|yWshZ2p+3$VfbBp7(6@*_(B z@@{>Q(EXSG4PTIWd1-~DY55FeBYd?6{Ay$G9oMBrf4Lp-(0+3g#it{}` z%_(dLGsHzEZ3P$&IRxhhm3eKH6Fy@u+hDa*E7`N}SCc_(7_y_t<>nscj|3DLs=Zl# z*OmiW-bg#7p7g<(70X@uRAI&b^mZAe{D0F-EaYMYFUx#&u5PJOca2xRdX~{)Iac3- z_)4O7xDm*Y(tA6GwMffEn8;}D!@ZYS1FcBBJ}EJPu_F9&dDv?4Pz2o|C2vvm@5jgkvY| z+SF5i2U?$_LSq-BIi(<{Unsox9p6Hnz8zwXE6};?d>RNpipT1E*B)UEb(cH6uKXeu z+VoXtHzBwARkC)r+QxKR&*u=2OL&OX@(y}snHGhBW#oZ7IkG8rrKn<{=O{a+CJ?>z z{#MX8A~|AS6oXHCF<YqU6^^jeuznEgZrbuwh2xlM$3C{ zP1bQdS&k)p{)`a9n3cCFU;{9;A>jIZ^5nx(dJt4^L1ngRefvfi&Zq#heQObig_h-% z{%j!&Y#X~jF({33F6(lcwcD6~-OYWs2b9hTUMBQCmE&t%7FGSS|H9w4q>cqxg4_sK zJ-kHnSE|XTN3uvI&p{(j1CmRHX)Xoz#aBx7y8)%R^J5ij?8mPLK&f6{8(S$Y_RH+* zxIOVwVaOY6 z`I}^ufdEvI9CTu@NMxvvOSni~VUrI(At7wSgc3RAp?4g$CO) zgrPSr!{7~ry-kjvwnH;4wm1M}5%jrj^?eA>CQI6jE^*^JtF^dbIHWxDG6x0u;zDCL z$dbY8gw^~feQLKdEU0To-TEzUB52O#!`jg|gFx8QSxcYf6bGYy7ED-ujFoKaf0*2A=Hm>hc~h7I+$v2xeWrNe;`O^0w_S zJ`5S8iqE|+ikFOz9W@1+nMZSXY&Cz@89e)8{RpyQ%0dcz47jz98$&auot z@?x)@boet{;dWDC(J=@Hy$ThPH-bUV^+mfmQF|Ijgk`SF_la<91k#yuW!KOLG}#Ox z-sDH!)lk)8AQmaN;BOeBNi6f{hw_!N=V9%g1evXa!fo zlD$?pdM0Hc%SRD5pw4)L&BfgJDeZQO^SGeEO6g4~{31aDdG427)N^;lQS`iSY&|BB zD2M8IRzD&N4VeW6;L;TwU~n+*Kd3YobHF29N*9n^=n_~&2GC1utRys71jy5aPzPbM zsf4*ISL5`$UX(7m1Feiix(RebKwml@F zs9R8c|Ir(~Drk|bJZ#b%t3Rwp)CHiE>^c}DzIjliv_~!YL7@6)Mw6dJf_(^sK;w_T z6^oA3xm*q_^FA!IQ(gZcag})Rd0Ju4Wk#pV$d+iTRHLgSP8TysP#Z}pmj9)t4tMnV zI>n9|^ie-9M2uw(AE9m?ZDKs4GcAUsdt8YY9%=|Jv)Ue=%lkFxnj2cKlHp~ne zX#%<+Ukc-Krh#zk`1U_6eQWKz;*Z20RV^qM>{R4qRy2>nPGWB+eZV`Xlpyly8&M*^ zi=JJVksNgEqgb!)w8Ku_+FKVDMZPsJM$PxJR;N^7c|oCY<@X?eImJ6UC34GoY1fq+ zlY!CQTe_eP=uIzE11~gP!5`}Ja5+r-GLv;XxMX_UN9voe;_F^zaRi6*oHs+MkG>MS zjV({*3S1O)awBNR`{Hnb)AO&kVolIL^qN5(w<{VVX9kDus~^K4OSPtQnbFt~DiAxP zU_O{*Igha6%4S+l4U2J_t5Dy9ofVBP+ibrcN|;zr zhn)Bu8+wgD87%3oVk)1J;0*`A?edf%K%|KjD3VZp<5y%8WWd6I5X!1Nw~!ZW1%vrn zLOLm$xRlp?q1|+HT>@iR8J|A1QUx3zGNiiW;Ag)A8O>^3uP~^4b#rdP z7ve6---JNl4bUab?wqg#AeNLbs^AY9S2pTeWpb7Pe zS&`RSKY0ZT+-$`}2J>tke7E|tFK2DgcMUP~Qa73;8^-KGU!fFpb2Ug8S!MOMCLOV= zj%@VPVKijmodYe8TBN?~(BBa2vb9YbZkB{T%Ha;K^!CB6`U!7>fO{}|CXIUoLF2Cq zwnbS+nq#wyGP|1?&19!6*mkykCS_7@c_c$1X*E;?kKn-OhX(b^aUbpt~J&+p6=bW6Kr6-!^Qp1?dAmsbjj zpp}9yk$~cgyNLi0egIo?J()^U#`RRaZq0fivP3CfS40&b7Qv{F#t?``iiJ%F;I+qKVecXTB*G^PwERQg zX6m~lU!pE=_+wQvWV< zXfp9I1z+TAyd?ZcIh-hyiOkVJ(9(;$3amkqWTBkS26Ge<^Kut!Q5neLAJ>euBD|7; z5=#kMf}b=biEuW0Af3xeoGu?`Qy$}NxD$y*!xOJH2tcVs2mxibD4O1PZ9gI#*h4f# z*H#*dD3%3XR797+U$0*&8HsbEfr%RPCWz`w zmiH*#rQB8?M*5pQo2{dh5Sc{+Ldhb~T|=ZRX=bMg>+A9o;u2V^)?q^kR^ zu+N(1RsKixOoQTQK~_I@8|xF}t0dU}Xsb)j(E^BUllKLXw3siphjrJC{i@$8sW^eH zNutYz>KZ{Fo{b5^S3x$aAsP)d0+uSsVlQ~`PQ$KL1E8ZKaud|bp>iRw8e96N?~j*(zekv2*9iFbb+X0ofmcjgtX=8veOKN9%z4Aw!WlpH8nfh6Oc|H0x1%ZDb=X^ zAQONoK-LX<^PG)IuxSrxedY}52;T!_a@P5tHlPNvJ0N7U!@UeWHuCgmBf8|u#2#CCAgacvfND*GvG742ulJXNi}W=>G<7^-Id{t5uDka3 zozk+dncQNH>E&&n8IU}gSuXvXFZ>}1jMeW$Jb9g?0NpN(ZDkeE{MDMk&!Bz`%lMbb z#X?;&?pKLkBrOZ>6W>e*J_zC7eO)*0pg6;cnG{~bP9Sx<0PIcJ*=kyMO>#nG^*x^6 zC$?0=~lE`EOwfwu# z_xaF!Oxx?9V$^&lQ|T`kF!d?Pka+tosD!ny>rknU5XgRFQ1#RlM)zNa9~Z<=q(BH+ z>gNifd8a7Lgj@MF)FY+Z(Qx#MJ9?+=UY;CdyIhM6QDMdJgR0#XOSkSS~8>9 zz*iqu$q*Addm%-U5=}2Xw?X73{G6N{U5+f$x%{=Ac9Ekge}|t9oQ$`A{efyn;aw0F zDVTPt733F|qSp~ko94x-0=2DsiHV)d*xvV1phn47EU@GN5M3mJopWBCUwdd48?iSL)KCnxHg>voIh44-5If6~B(c{-IaMYL^hk4bihAKrnQx*M zNr12i2R0_KFnQZG^FF}XO%z<{0KL6FibnJi1!x;M_T3_yZ~2JK^B)WF9LSU6)m(B- z$4;Ro;A4ra8`{QFLF2 zoP4S_VMn#$ICNPEKrZ|E8D#nZ*uBWW zCMyXY+*^9|O&_Z&!c7-4$H$PCZ+Bq_1Nwa<1o$dxB)e+jh(xbK8pvJ2 zm9!Ki%aBDAE^3dX%_Kk|){|?8-?<6rESc!BuKUme(T~t*j8D<1PsLe5I-ErqUp`=C zc*I018{}kiiLjb1_pOqdr$stsd&;h~ev(xbeDc0s=G&JPGl=&KZ6g-Yc$2*7U!Bs{ z|84sO8Y-H^o>}Gdx2R&pIm3_MxR8<_hPM%jb6I1Bs2S>DcOB^Hc0bqo>LL`M&t_$;J+Sq`wvbF$v$a(=)D%`3yo0%7 zr#pK0#5_`EsFb&1`_Xvw(BRNK)FK4NlU#oIjt)F}TR9S9_+3VSE~i~`a8SKQ$uf!x zwa|-jY&G6pK-mIh9;NFkyZ|MP-XA>EacfU@fy#q202Z0bH%u`SkIu;CV2n$jlRNaW ziD*Cp;7F>{rt{6^(!~}yulyJhzxjH-d3XG>{?RZ1pv9Y2>;ctEB&lDvtPHF3Q7lA< z8nfTQjrfW1s;KeMA`Bnp)){e4oqPO^5rP;&EC4%6EF;9gQl&&-YjU~gi36fq0Ey`S zd8hddsH4DIjghDEaCm0Py3HD@7tYVYIogsxk;KkHYOmb&(I{4-jyfK@ILdMdj z*Au0}-}^={+N1dXH5Cpe7V^8lrxui)%;tE`a~Htry&P`BDcg06u)h|75$ab@@+?NC zI5@0P*5f}ZL@D`9_9f{W>5UM#UFas6tHBTug36Zz*Tr@ zgPbx~UMfZdmxU|Mycq8}>B zgjM~kT!s3of$7E?Zb4l?J%1YUa*hrHTd(ntg98k*W`BuCj>O|Sl{BBw=!-8O#p#fw zupO(OLw8IF8EOkK8+%NEnjPz>T!8w6hcX`>b*uXfb6X4(6LEny z&=*;`Avw&=KaD1MM=bed`iS4Oagn_LWk$!91Uj>jv-M}YM~u5SlY+S-H=m^H<|J8` z_fgcz5?TDXHn<5`ZWdpmEPk^N=J{7@1sUVRjqj@m%h#H2+7(63+b6~}-a_c`t(y-E z^C*&sJA+DsKrSh68Ospozy69)54p^P8kStd$S?rkZcfA2H1nmcRMWlqaCuOdue*PJssimuV26zCd zxN3?RO}hll3rfC5!J6{9f6@)341Iu497OJY&G;R(lGPR4dJi7I;tUj&nZv&SD0slu zbECFrbTEj98mOj7{#~8#ZcPUnz(!*D%GYv$iYj#`ZlA1UBvExYKoXV^_f&eJHG#e( z03Z)@nMk}}Am!YJ3~s|ry~nH_hI$lUP6kynV*C!+khHb{3dB+ewf)R;f1yTK@{>$* zNqeuShI?Ts8lf=Z(9SlZJV2lyPielis1a17no?p!Z7@AW*!!l5>qoz{nubH=4C}oY z?0pEsgx%hLt>-!YkL0L71;Ac7Q@~t&E$KYiZk5}MjY4{ zL~ky*f-=X~s?o8G6ZNgfgfa<;DCE2?*mBQ$berP6rt$7l2gg&WYe`j#pR=5Y!>$`9 zR(}08GUdE@4WmP*&yVCjD$XI+4tuVn_|)Th^gC;+_O|xcBDJR(jhDz?*?yrI3AySB zo_aQsVrxRijL&Yn0DE~n2PaNlr-ZuSMk-+9&2`5CTZ#>v@n(dFt&R7}Mr9Ikgr{CG zv6G#rYRvywMu0S)PCAq8mjq?QKph6{#+-pATk}^tZrwF}5DVRPk1xz40pl14=Ll07 zRJ^KETwWO8S+x)ZTq<-9#H0d!;adsW7ARHy>ftO^sjg1fxYRXpd{SP;m&@erRrU*@ z*XpKI@o`K~F;0AedT6|4HOuY?d}?84Pdin%E5KP{TJEBJD%oKFg^@}IoX%$gB3>Gs zah(QJPgmKJON@osogOn3&f*B}0;yxJ9Nk*FlbXYgb5}n9hzK#VYr1*DVb*2*AwO8v`X;|e z&{?J6>b{a2^mxnwIWRs_cYd-<4?GuRKMZmpOh79pSZiEn(jI#6ncZIsIwS!y+k0ot zoVgk~%4jbUbxiCJ+=xUybTyZ=K^kh{@hp?fKluA8KW_8N1}F52QIJO=ggtluu0-GKB0_tp65N|F1 zWkcn3R%IhSIKBD5RlckRUcU94bnIvJsI39GtN2-=a2Q5W>ql+KHyZOU{cKEi%c0jS zXN8Y(xOKHucecoHGlxPvccV{IN`Ar<}Yln+zY;eK@W?AKnc7@ zt5?4WeP6<-U(J1`ao?C;x3g11l@zzPh!q%k+(rlBv8quG4mxc2v+OkfMuK}A6%Y;c z%}#SKK}HC8a78Aoy7blI!bFDq&%)pZQvUY56eB9TUhXK7%2$||KpHsl6|6B>^m3RO zZgWt1@UArHF7(|&VO(yNSi#{NW}uALpvb!t2d+3YhF;1jA6azuio1LasuZ#!O+p`qF$qs_VFrzde+6cmOo|A7S)drsaX8* zkqn$dW#Iu{+pw5x+_tRPa{+Zp8^sRh|ah>UkeA{-N^M`TPOd zt7=XYVdmwkUXVuVi0*o=#zN|wuNfi3hAO_QhC-tu`4@5UV-S&aXGk=Y`OgpI3Tgg_ zq+;q~Z~{PPm|O1h0(A;M&xe|U5Fx@uO1uKeoU%rBWN)U4-stw@GqQy^;q(;@2Cp98 z3NW{o$hfD9+H9W{hG)vemZO%!@gbS{*`gagJL+sHMp^`jXntWaO*%VV+7x^}&;^mA z)0eq#NZ4~B*1%0%aLr3F^L^x|z~Gj#HuQ__m%0_rf^GF4J)r=&)~i=d8yb<<+KMNv zw5o<8K<&uQF!^A%6YwBX%*HIcYhS%EqJ;96MYWv3r8Z(ybe#v`*SCfL z#8;S*3vz1rn%I0KllzCygQ*{L|y#N)xYKUP1UW_s?3x)~I3 z$<4g$u& zZcmr~sQP>pkZ7;?Q9=yt0cV-R#wWv5;l*F}K33vqwgX4qcB&EwjxcHU1-d{0 zTk^n7B6MkYSub6B=Ru?*XBb5#)RSaf?|pX8UPa#xy6pR&Hyec1<58p+xQJ*ZPk-0L z-kn_SCoE(9?3HBhE?})wzkY+qhAA;iR!?>sRSuQ@Qgc!Jm)tIE7G@swqA#6l2?plB z&@|C^>Ji|@?;jc_<26bLaLKwMQEsk>I!K1M`v9h(&r!8qAnm*as$`yX73I`T)L&B3 z9(b-3KIA*WT0EL(gBVp*LMeR{w3+0*R%|e&bCryf36gV!$|aeRk~$A6HDms&Up}`Z zNMy|<>3yD=r}5=^0h^QawicB4X~dHVJk38+esufG-;QI_<@T0H5g(GAIzBTVNr9d7 zyA{3v<$T_ZQ?h+)gZL4WFJWxKKI-JdKR2G1P_lg;7<k!UBgz$&2j@o=gp?-X~?_E<#lSN+qBt-lR_Uhu)Ygx_C*OYq#KO z9g_DNX?Xkeq>F%5I`!u44h9>Zd>|Wo@@j|7tzAvshhJo&m~3D1uR9IE8$mM26f1@| zxJ)I#fRI3Zo6m{{gW}$-&jn(q;rI5nru%2Um{igq4+Y~>R|okq2+W%^{Hx`@fE-Y1 zn^ZCU9RPFh!}&kXC#S!DcloGqKn}+`Z~D^I&w}Q!-QqXg<s zw+Q{c^W=N~`oQJ!{nlgv10TtDc`uK$&%au~L#I{&CtFP^U?iz__ysyS+56RoxYZkP zUAsV{@GqAzVVFa$Fhc*YO%2ReGUX$4MI`qz!zm!`O}s}zjW_=D(8_;qJ%qzxX$KgM zBU8UZX6erM_bIyB&0Q;7nD5pl(>+?dp zqs`H=GP9>!!+8-d>u|0|d;%IAJygzub#IQlEpQZ6P&iu29Il>~* zwnFoiQ>H%fn;FQs@66Oypsi887ke`=<+@pC(uP6lgfq_Kl`RS93Czyg1AoUHnpnSP z1m#}f%u1sbloIJ#lSl!7=p=1l(3MnG*^a-8p_8<`xjLx<0Cu|GZ8!7Z`aVK?9x%0v zwdnq;wE4N6pZ2yq&Aw^UvOjet676~8Xq2z$nxxdCrVQ^!pgoD#Fx?@@7*6KysGQ9W z*DX-RnrE@5Z&EEV6cn!a4dtptpXag0Z88M>|K34eAbog+<<|SS1%O$Y8Ut3!?LL9Q zO2B?UaHU3zrAbihs%H||-FA57USwW%FQx?izt=f*jH&pdSifdqLl|5)916_Hov*RM z0ye|Uw|4Grx8{qC3vZxotH-V$H0 zE*MXQbp%{z_((LSOL(L6ddZu$DO;Z7oMTL(C7{YLfsB9c(pTb7p`H-g)38vH+q1P) zpn*Yyss2_f=i!w^P^Z}S&1#g#TF~e=H?AFlcdUZ5D4Vdk0 zfe0QAOyk>~hp}Jl&KEl5Soy35Xwt-7zItK@JoN+Gd$t}LqRKVAthW28nkg~K%@|#_ zyU-et@vT%CP>5LgPCxRS1#hi+9&g(;?9>-}x@)9at6a8N0G5AYi=O*#S@SlZGwgsI zElKtzaGjR*5UcRBgSFcRA~2w8x(3s|A&J5U-(~tXSz#p}vqTY%ESfAeZdSbtwN`QO zSk@z1x|b=yuVqT?KP7Jl@_o-IOF-N z{MWL4o2YMafo)+_m#L?>94)@{joKbI6h$g1DpQG3DcO~raEH6v@>ifb@5@!I?vkWL!NFFbb}YXq+EMd2G|cSKSyMTYa!%mRkb z^bkJaLwLfj+b^y7>gRnfCWWW%6723p_2fK>n7IlVlsukji=w>uLw@OdU_i~BM=y(7 z`uaK6ISxyHC=z)~26z1iB@QCJ*Tm<`cD;8!J@UydP$4Pgw18K?()emq*Y&m#H(!_U zEuMhJ#SwI3J;1s|2w^7!`<*054f)xJ97cdD(VL8O4Ts4nV+xJ$7f0<3=AAwJAzmI%7NJZnpv@B8p|7f&QRV=jvkmO84R;; zv~Dx}lMkp!<&%rz7zW96%332BNVHxi)3nh;iizEeEh&3`M39R!$!3Z?F$=Ktk6Qqk zO1hOPa8U6yjcRv@t}}hkZ78wB!~JUZZ^b=Vpl#@J>9Fy|)cA}5^kj@tzE0L@?IC6+ zC#0iAj$fA3G;}Q3wF`|Mk7zHsKSCJNO9n={dT_;f^#Jh>o-M0Zq20{(iextHiIA+P zxe#vpDwjbmT(lQz3~+gbW~FfDQ#>@An29kw3ci-8jG@*Mx+f1?iIQ5DuK@dK$0QaM zKj+>tVp?Rf(8Zcf-9OZ%Z>CjjBErmko*X6PE=r6c9s~A|m-r;~ycTE7xl2p#^yJWZ zc$}fv{ts6`q7qg22O5B?0EvxjlH=>WDI5_-SOZ+@B?}dboS-o5+X}|7C^QJ9@S#5y z^^T#9x<9NIjKIXu^!}Defnfr?mhcSiKxcA_Y#mnc@AD^#!d;>@O1wp%Gzo0eJIqwn zXp=>(vXJ7pc5s36Of*lR=4D5^viO7CxYFL|AZ;Llp%>rSp%opdLt2C+Ve=7PX5F08 zivxX2vm8vfy-nrMLwnvw|3DMY)6(k;d0VlADfmQre zlFccGU+L+PP+%i@%#ODw--h)!9#1LnNV5o@)6zglA(m~qa!uAy!4u_QupFB9dUVke64#V?g)-7wxJP<36jFZ^_YY*{W zoj!062vjjseqA}&vgVp;L_r$UKvI3qG+T3Eh=H9Wl%0_mfJY>+zc8I?AL21a3dSa2 z^wK|+4IN{d#&Gc^17lx30P}g5lI0XDo@PlI)sybO~@n#xb_G>U?%Urq#{`ZS|L{!FAlP9w!im_ z%x^DxF8E;zbJI?xDtijXCUm8Ccvc+BV$Bnb@C35X5{$3}cGd(rPZLF{)=R#h3hb)J zW|U|G7erxoRijp6Grl1&G3Mnvpe;QF=}=AS31baS!FSwamZo1r-Nb;mBwQ}46&LIo zU!^%y1H-C*(D>lf)F*4(;j0I-CpJT=@YShTA>hYMt$=Oh2js}?8F?}s0<@W|C9hrq z)^PJq8mCD8S0PuK?u*z~ha&Hm71yXFaFaxUP~i#mm1dB)0#G-`z3&Q)V(J-d#`2=M zC}HGX$E*vWjZHMhpNpLR-{|}ZDsqbDD|*$F-vM8fOG13}OjD2w--0=`Y(8DoX@=^g zErhTp2C#Dw=KX0Q(nQ+IU>x`}fg5F#*x4tE?dp;k~D5e8rXhs0p-;TH=twRe+^}7`%E1grIUA4g?gj05`NAJZT zXauu9ImjN>LxVI(!}y;u{6ilDi5|$yQDU3IPJj-eA;_ucc*C4U1EM+&pqzvCu821o z;%?Qq{l7y`>A^s3WZjOHBXH~lDsh9BeIHDm)&^!xCEwwp&UboM`)9AYcp;;rida?s3!>>f5io0%GBc?llv^;8`t#@$>_MspwF>aoN;F;U}a@iWI z^OkT&ZC}5oU1K%&#%n@E_Ws`IhtH1rwuU~1AOQ2jfc7tdml)E_t;ar$>;*F`N^zL5 zEzB0h8wDZJ7X%QLefHWA5z9R=<5h1-*o&ue}t31g_r=*;X!*X(UNS_ z0kskOJYBhdA1mSX7o@d^N_N}&05_SOtmQ?!5Y zDa|h!r7*fO$x@jEU<4ZAySEfxB+JnIa4=JH`fcqmjEhd`Ym-~w+SO^NboK2iK&T6m z>yNf?p*qkg(AdZk_$1sJdBi5&bALDfw%gF99eAg8=S)rRec@Vzs%CX8%YjqUy-fb zE<7K3L9!8XzLIpHtK4&S`J;bJ;Nz;WjzHI6E(9HU>r{uN5>8)LD`|RC)*zWfKf@%#YWJbxvI#As7_r&J8p@jxTH&r-6sj4|eO|MJH3Xyg<&W3S* zF$w9mw{0A#SmbTrOP%)H`~37a0wTT|y?h8*m1<5gcKb8t1K@@q0E>~_F<@Y7NGa;1 z_k{tVHt!_D+0wgE>$tQ%X_e)I0hxhcZ#7@{Qev;V&{1GN=!v4>|K9lR$KySqNL@+` z*nS4~z7gf_0=t$L)leo*C6dGUkP;QtqpJwcOJyr0}+143Wngwi zeqG`#Ho#JEJKb}*?Aam$k!bs3?m!}TKZfUANnBW78>Xa&SY!Y=o7z`qV~L4;VcI~ zxMHUwnWx)?mhM6Jct7x2eowf^o7bH%&lBDjF`Jl1xXDXJVtgWXddvW1{2DR5#_|VU z$D~b!_o6~dJ*&cya+LnQXHj>&foxQ~Gz#ne{Pl5-1v6WZnrIZ-)d&1f@BDA3Pe?)T z)@~}*R(8#R9-RInw@PQz?whd^b)qta{8;FV0;%%Zv%TaEsC?t+ z!RXFJXxzcE*3nUeEj7x52B8L*F6}ca|qu=WyI4=E~D>NC$XmAC%TMvUZbzgKW0h@!e{dsOOcPnb> zpaN;YXkaAT6Cp+>*`eev#cxUr794Y^^pNFyRuW5Rk~Ge~{bv2f=TKE~+fz(S-T(x1|?bC zu-YIfjEx>+1kc=>Iy6cgQJK&X;%=suWMa0kmAcj83A6Oa4$yBW!>9qoj3@P~d+FWG zOt35gBbmLH)29gNe)fPQKrtN@lx9No!lU?0jl)&JpBxoWNF`!!jkWnIA5Bp)@Ep9+ z3mHzUvxSEwE?u5d9$cNRWw|NiNdsku0U07j4PZ~t06%yUYn#cB;RuO>t8XjKs|W7w z;mJ}%5!ufb<$z3m$^^XF$pE8RnWbDB^C%F%MG(NOmXmr5Dz}GWikK3+nxA4J60o%z zx&K1HjkCYRP-oy~?IE9=Z%yC&E?A*vB&=%AIXf=nhE@OxO$Br#$bceWDUuH)pH9th z8W?d?@F}|7ELybVhw;GV+01KR*dC~1HPtpfZh_ZlANOQImuopHM8W=BN5lhr=5AQaPUSv+*XL9iC?s*}%CdmJXijGZth3 zs=#%G6;)N+)cbm9N$tGD{QhpvIrh)vT z?%gAGwV^(J+iTv<$BNHuY&DR+Vfp?B__+b+B)8*9tLYe@L#7%pU9178?uRM0PW%*) z0dpH-cGstd1ZZtQHK#0#;BA_q_$4Es%CC8&;|6g3aby$>L)IF@y z(%p?<(Iqf65)UN?5|RT0!cfxPFu+ivbSS8(fW#n3_W+V30y8usg3^uTcSfJ*ec#{r zkJqIa4)>hP4AwPzX100hNCS}6IBbBt=2Q;+Az%4R8S zePz^I0GsljDu%Q_C3Lm2dy1|DG2dvSywypcm!Zrx;s272uE{T~J?vtniaa^7D8#z* zmDpCyCf5-*IsAg?kL&Z-uCZHP+e)M64)c;o86$u$>Lf7!+$a>^a^aa~`YNbVFZ@-v zVUDX$F{ag5W7>z=5K+OrI=P~Ipl7qzg7aDWa##H!k`eTH|`k9X4$S?*!R2-w>f zDm^tM%4lkrdw1U6*RK8O#oq6F@9hVbX_`;9?#c7{9g3t-ZzKu6H)eF@(}DMuX4E92 zDQ;$m%&1K2RVq*Z1@4eTn1Wm}cU`i{Wh5wF;R_d!4~u>}0lLJ;&Et?7a3BbMUo=St{j9|5%xn zIcgZbm%HqlibF5g+F~o@Cqlz<@9Qv2->&9Az*J^m;L(uAfWY^ODi2`NK%L;Z|2@r6 z-Yp`Mn6ipL=~g?H@`-!xt0dVI=ga#Y$2=n1TXcuONviHORE(6+^7FTUP5_m{eW_!F zr;u$4m<72-QN+Bz_~jSDE0Qlkpr2x-c-TA;8|MnKoQ9>vTo_;1n{dd`Y~h@CeRn5lVZucum%j?m{a81;K7TbPrX{gkoevnh=H*8aO??9u|R+Brjj9V`-JLdY)V zC8MT>cu$HJDTc1m!}=(M{}1CGd}mP$xe@r*3XI%J+MFSHklSA7*b26jFK(`=j1%vO zXN7s@BlZKOA68mBWB5 zyMYBcr3-WO@$|-oZ1Wc7@mOuMeGAY}YW3%X%tz7Z3bhlF zGhqSJ2gf@&{HcwV&wmdMig^IBsrzy8uX_bg^8W-5d=Xzw$Fe&R`~`l(-9K{@jhh@t z_gR)&X0G130IL%T7=V8(7??S-hmiYBOm_|B;~V)-QKjKA;Dlm~m-E5?2Y5ewQ7C)A zOlf?ixlwdM(Rn6Z@pt;IIYzM?aUT-)yHecSx%=OoPs0}#dMllaKAB*mZ;fw8PFj#+ zYNVc}S{EDSl>#S>bx%A^Pc@QQZX*m-t?tB;gcPJ?XUD5JlncOP233Yd>n`@cS(~ga zUu!u(+yCI()RzR=)?PA7Io!N+3-qLXDTpITxPvM&C3sHe0T!RLVU-IqM?rgM>ZPo z5GQi7TdX{N{bbca8E-)M8Ql921pq4o6FnbKT?&Ub6L(ON58j)5*u2en+(sQT2O+Zr zInMecj&n?M$k%-!1e@Ce2fC|%#m%YUNv9FQc7#k=E*ZZNQEwj>I%vRJKUmq+z--{y zTG(^kL>G+QQY>e;rFV7AwD;_qNg>NZYn*I$lV@mSKYC(WHoe8~4JBrIdw)9}hB0xs-WczKW_fXUxrJ(oQ)wUHxS3Enz5W<}_ zfB5{y!+MnfM8@v-xB;a+_<{ds_mS5jG(jif&l)iXY>V?@7R1@!257IF%t$|J zPmRW0F@%y${8VgO1aD3ysm>$|yQdw%{JXYr%bP~p2cx%S3 zWW%Gl00`n0mHO53i> zK=M;P4H96VodMZNtFvw9KwxN=8*X(CSjCL}X>ojis+$`-fbYkFO_z+@{W+31Z^diU z90;h(BGk&@7RoCg0-fAG%iz5SwZcsl|I+1h9y6apiB~V6T#}_H_wKAC8Emo4=rG#N`U=UN~;xmC19*IA(yux=^W0Y@vs z=v@(eBhuuBB-@Vc^dAE^TxwQKVGX&g!LuW!!Lf(8cr`|cX*46YB?Q@Mk%Zle><-jQ zhc8a|Yg<0zqa^4^KeKc8uogH3?NA)ZCZIqD*Ki7S1D`$Y`MZ9lcwteZ-of4Mn)K z7J!0ocl-XJ__Kc=ki~c|!<@%Ic;5)Bkk|5jNNV$dntPZiE@w@@YvF*GlOxB{UZCNw zHffT*g8ya-lBOrw>UHM^up+|$|Ev6pzS5KPuWs#5&z7#J{0U|vvN;EZIAm-;Qxj_- z%__F5^f;+9Fo>XyJPT5~&rsm_qFPVJioX_9G$b(Y?uQb_xUf+NtOzv zozwBu0ao(%D*L`iJcD-=8S#lNuLeQ;j&vWv0k#)Dd62nw;(>CZl3l?Lz+HA9H8{P) zeUk+j%Lx1zC%9%M%$0;E&?rwY_(oH|&j|uYRoppiTF(L;GVq(`0YzG<>4T_INTy0Z z;Z%lIh#bz0!_jTo7r0rKo7;Z}T^E`ssGCq7XZ_H}Lw+T+n z(wA5tQ*Gx?o=xD#R{@mOV`X>MtR;5weAwOgyL$%zUNxz6Mimf>UKx^w!3GHrtyL2k)$U)xmm^*1W!H$>B=0_OkHnAH%(AejOZTFeMXG&t zf~M*jJs?-uP^0uL>^{G~_5wDIE9gd40@{XYchc+W6mh9l!2H=;#&iViul~Fgnjx={ z;pGmmb)9}>211gG9_u@A$w`ev=|*I~6&TK>Enrwd3z(+76n zds6oSIb{%7$OY7@Xm*R3~V zk*tzv{hK0o0zFRG@Wrm~Ie;ba)!yv|sIF6yR|SkYOc`2Vuu){^rBef%J9ob`kKw^1CT^3&~Chp<$7YvVY#P<_CGV(aYkUT-Pcq zv-0rSTHo^9w02<8X8kB2{Ac?Ws?Pte^mIch1?ITyxT4k!W!S~QwI_=Dh{?2BjCVY# zkYWqnEcAy$R=kHvzP*k=UUe9IZ6y`4&d_}Y99NL0c2wRKx*)rhIl*hEm*HUDZD%8; zZu~tH1Qqg9mb;fb2fkZ0p8kHIDMqu~cfv#-Ak_VoiRXjI0vhiE@WYt)8sP+Q{n9I) zw5eGZk*8eSYK{KC|D}sP3m+v188Dq;sbw10#ZNRM#H3=I6Gc?ipWaA&YB`aqPKv1q z_40k{m51H38NKHY`(xz;GiQmo-G+g5Lj#|y7i@>WkiqV4Ad~s^>PF$1%;g{V*>w%9 zcVjMc!~`QOZxVgWz~Cj#*xUjJGqlQpPMxc1W(*-i9AO0EN%4I1h#o= z-q{NA)>rYDLkL^|#8n;N`BJe5^-01^v zy2Pl%%h(|6K*#iL8o)105Hdc!OI$=C@veU?SQ&f5$1s>HH{ws*sqGHxei5#J-L(fL5SJ zL+eL>*HEeq|5ZC(^veThsr8n5RPVH8D~=bE$%`-tgTGt^;>3Ez^b2i( ztTTF=B-Dnl>R8F{a)F!TPZSx5aTvT^t_4M!+b%?odM;#@L*kFL))Fzy8f7||TXthD z&>{l-HFTc|LRxsPUpKv4Hrsh4QLnB8-fLb-jAdAk7&&v@-*+p5hT6+)4Z+AY$ETh= zT(z4F}uxkx+K*EXXoTV%Y)7y>VYQ( z!TTODp^jOO*^c?9V(vi;iL$F8a#tw@A7Oc8Ilf7$C8f#yWtwXcwYfsu5-^#+(2hv& zhr8}vi61Yl6uM64todc*Dm?pZ6vs+E3rx%xS%bG&P&((#fsBvjmB!2gBoGbYy5}_0 zyc4$vD*G;=^3@0GsU@Ifaih9Waku>#*@zL&O2eV+>JCtF5~$CjjQ6$Oc>;A8HlBI2O_04Bt&P{ z!Uxwiz_8CnH2HM*w>U#*!W>uNstumhU~WdpubR%ej|$7W*8hZ$aJx@$9^bF#TlCk7DL%;RcYZZ|&p1xc>D7mNa<%%%=^7%K~e#F!`vvqfq^ z6Xgs$p-&k%G!|$YFXJJiLO%C`E@z$t2jrfqeblJhreM#~`?}sF*0GyOispx>Qy!SD zQseg1DSe&~!i}|%>Q+1-#-A=%QDL@-?`Reu;Nu1K>bq_YJ0m(5q-$&1SNt8H6Od3< z1MXkGnW|)JUg=+d*i{^gWWBgq8Y+GnKH_#rQg%{iE zzY~HJ1_<_rmmQf~qE@C9nzJZ=SK;PHKpD(FEmk<-sMIHzqkn7;FH0JZ4K%Y{)jur4 z^#=|Z8^|uH($rJROm0D84WSr#>pX_C#DOD}GDYUi%1l#(BwgpNq*>Dt==S&C8rZ-O z=Cqq#ZEgsA(#IC{UX`HopW*DNY9@^z70Bi#{)Vg2O6GD!i%G`P3Y%VvzG4*C_>>z$ zDly`p<5J_{KLJAGG#n~={V%D(C?(N(!K1x!+?GH5_Yhjrs@uvR?jvu&*$>*r zNh+AaZ72d?xQ04Pg)Flqi*EAt*j+*M0!k)3>z`M6NkjHK<IRi=agFIKgX@rp#t~wmt9^(ElG^b3gZ*_mSfh=v2?JNz|zPndkZ&>~D z=5n;7vz|l@p6^rkIh?rrZ1lzJmp^UhVl96PX;ZUGbiM|CQ7hlahD!tNWJ#b+#LDng z=-=o~YD6}yT_oO-O;qVASY99zmte!-nG~>UTQl+#)=*Ty1siio!TEG&Qz}lvkLd^M zDZIq%PBZioO=Jd#>hQ@njdeEGWHO_TKleUoea_xYA3T;?Q}!u+{}HF-STH*(iPLr? z+$0;tfz1_yZg2y5>)b~+ol&ncaer1setqsxv-(sA))aGl%SYefj=30$3AB`d(hQlH z5merE7U;QYgZy=vWWyurAzrEuSC>)`3!iHk2%r*Q`NE&%fP1z4iKKAz4&1#b>WV)DKCpgH7!NdIj5>Bt&MQ3jTJn20 zce^}iz~*`CvgKeC3x5cbXk1v*`~M%%nqpYKr(iR~baWenW%mD>Y1WmRLF}71Ny+bbVbw9uCI`|RSx+qkb(!&J z?g4pBkO4h%y}KclckD1=m1DNNj=4PdvCNJVYO6&y=h)86EuVRM)=DPluo!!9^9va! z1T(lgYp)Wd?C_D!(%8li`+-0xPHim=CgV8#LjQ<04W4xsiOD5_${8=H7=o@US&bf- zrms4`-Dqm{o3XJJDvFC)9Er@Tjhs)LK}CsUB?}l<5s$Y!zeqs*>4D}|a>3h?)&EWv z`K(j5&2Q$|V9lYEtu*eAoJ-;cr91;_UEvDt zs7ODXWfygKEgA`x?lz|A>Qv^`~Ip@r|PW5sMHp#uS$3r`10O^q7+iZcf| zXxXpc%*pkm|4QzJzw_Xm-$+FxN@*fuXG3B&4wg6z13kJ;2s*I^ft8Xcp;u&>a{bG3 zjSk-el}b{*Hn^SRB9oH89j2HT1Q*THZ?sBl9`CQ^*>G51%u-3faOiHp5(zOBS&tg+ zdg+Q$FNF#l_I{2yjJ)Rh`aDRJjpoD2Yb*?NI0) ztfYLtJz~<=z=uJt8t=L^8keIub%SY7jEUpqC3cg#I2_yjyLHx&h9h@%#GwFd)An047*to^iR>npo`6Viq+m;6%iq}!( zO!q&zif8L90uLJlZ0A+bECsLi!c%SvioX@eRZ?_d$&XH5(^9L{6DCS7lMgi1>cgHU zkT%E7ca_%9Ww@+ww%=u)Y#IlilKsS7j4m5ooa`egU1QYDHWSF@(!G9KP;dO2yJqh1$84x@Xm!au|Zq_d=4GA|-Qem1-J1S!n0HWZ% z&dL3-!&;z+(2U|NQvlhmva3RE#)|fWqLChEY?dVA?>jcBCHLPL~~CxQn=@IvN|Koo4C3J6ln&(BWtR5_0?MVL!R zg!4qq04{JY4}zCRH=)#|vRyNYa>$3O=`5<+P#h)2FOwL8DoJNr!am;1bR@(` z;kF?*n6ad0grADWC!lO#&Z~&?5csWnVb>ig`{Hkf#K#OdEz4$)^=6?G_%^mLbVZlE zfGv=$6Fg^grK&Yk)D76+{=_uMC@Wt?FR2n%xg;fJOgoPA;!N>5ygdnJLZHV+Q(@o% z$-=Rit_82J{X^Ea7sJyfMaMxc1PjN3d5ux+VN`B`r7{s?-5NIOi}u@M6%f%Qp0s%J z#*P=@6t-N~l6HCw<`Xcx(5S0Ic(#v0D{oGawgvDrU7tvq38f{7UllmdR{2Zy)rkH% zEXQ_lFGo+w z;!ivL^yJ`*Teg047aXLRk?r!JLZSF+5)#S!krc-A)kX~QBIZs{zgd(zb-SrC>CKE( zYuyE2i~f@}Fhh5b9$>Vavlz^MCwu z0P=T6gi>sFm02x;O>#Pny)%4beNjwo8D$7L_8vNqPoCbALUA>0TIa6fTY_M>%GFy@ z2l=i(s90LBWYPxMGf(jRqVF)syv_~-IomqtfGS~BpKE=zB%{o-X+pT;7>~m7%U`J%68Gbb{}0Van_GgSJyYgnzmYaqyS46RBl{174UM+?U#cvF(U*lxcwl2zZWTf!%M3 zd;!Ja=ERwC^|@#7S&{8Vgf}4&xe%_8mly0*$AQt8!0*7-mg27(D* zk5rhKDw435nK(shh_I0_pz8O(_u&1@Xey;=YMx27YSm2mo2?s72}`h&&DnJTP)QeG zL4m;MfPNUj3jsX&h# z%h%UUqydea-9j;eh?u@~Vb6k)8gUyS4HP+SF>S|kFD+~mN5G-n-*mXslAxduTmyPo zzK^&b8&MNR_U*d@h@h9YGX3GTM$6?bHrxFJ zwL>zrAGw{5jq#nf3r}@=^{(O-N3yJtzjY@5_b&dat^AH!SGprI07U&*X2}B;UEY;{ zX^RKz`4APZmFaTa?B`$MGV3fex4C?fSC~E}F7KmLeV%}+#CUiX#5cFF1u{&1!WCJR zwd>8+JSIo+2rL>0u%}n2h{ zatzymx+;XN?B-fQasdS4xD0mL^N}UT-@2@=P1}96)bWv6_f%EKptifpZiQS=={+}3 zbV$-SD@(UGa^3iFTZk()h1XbnH9C=&B;E{>tnU_0T;JgA51^D`_X`7ubU@xGF8`u2 zAdk?Li+tM;y&fQl$S3(l>L!4&1j5)UeR>4`J}Yj(+q({8KMY5idsXjV=KJ|T^-FlL z8ga~H7uV7E%ceIXT7It+)L$v+Yzv9r)QkWyfrlks7Eds6d#u~UOHTB8U{9*G{zS>+m~; z3a%&0&Nwn#c%_p2t?9;cpC5Mn02Uwi!9xps{BSSOukMrgU0;P=@Pu$th)XmA4Fld) z0fo1(H0L~@1tZgZ00n*$Xaf|0n&%h}d>3?_`dL}&l9?~7+d^OB(5uPAh7cYVz_qU5Gy>Qm{Y=eOzcb}8Bm*||H(~d)li#kUGkB$u{tx?Y z;R&9}Z57XstJM05ah=P~A>Jvgmw5?|$+iv`BQf*C3%7hQg|~BzS#`LVI=*!!OxHn7 z2dE=ksxel0Iy!Lw^~}D)wvf@hdnFJ^#A>5ct$C;aSL~w?s5eHOJeMm4Saiu1s3SzP z$}DVRR39a;4IFXW)=VES3`2|k-iblTtG4KoHo1c26khwrevW3TG7=;Y1HT~ zHu5a16>uw2^n|p7fB-kkLcNV^@ZGQ*VnLFLi`dnr@~X#tGOZ&G0t3l>(Sd0~hBFK% zHQ3Ms3H$hgEXi)x9a<8}=i&weToB{uM{hoW}1~-=U<$Zjm3e7CGp~zNK$(tR)t31OL9d*OoqYk z1fP2$e!&)qQPKcR!mMmv=XH#73@RxKgm4iCQpv1@nm|0TKchAz+1;Z|%XurcRI590 zyLD#s(5GUUvs5ZauF_Z^E+$)#WsNgmO@wGJO+bTg;yzjx^!67?0o?Ex}T)(0SfRGj+Nh{gDL@&YM$wnq?g?2*iMU^&&V|65mB$G zp*oF1V7%K$#tFxjr%d{f>7xSrimp-~JwCd8Xtvq^XMSVSlz)~V%SO+^984! zu5ex%&FN}#iVFY7ubw!=KP8z2z=4hT2jIQ1EDF2>loK+TEeF5UZm)Lk*i46ux5fVi z9i*JHTvkQva`Ba~lKYx7o{qb9>bqh%kj{#+B92KGl+g*042N#I!pb?*2p!8^=3Uxd zMm-7o#YMc(A2ch5!T6zdbTtz5V6bpXtdd?`_>$$~Ffk@I{vu_>0&Bs=Y#Nv9J#;?% zz?g7i1_dlS#g|LUd>gjQo)%g(;22RVSk;%+d!65?d6>Y|h0w))kpeYJz7Q;{>2IPa z8zRz*avxv&7W_!Y$+N%HQ*r2Us^gv25yKqr++rgj13$2EuqR3UW4S2gw|FXZk&-b< z+rlhR0!`iRV!2G$XjCA})QQ~yLpD8bAjN)W?ssU+V~VsIpjknBegOQ;6XykFjUL~m zD^y>)<>a{OqcImvU)tfRDvH(mf^?EEGsShY)m##{s7|jgsk(Ir^7S#GKL2x#t>~>` zWBzBBx7QLa?o*#lGe8+G?TyuMytz&Q6+Uuaf9Ds$dCUDpUk*)K9QW}~@27&T7W7$R znRRZs>%h*njewubbynmUrp0HUXvTnKJ~MiBC(`-12D=77d~K@~@2xQ*O{a&g zQ2N|ieMmHm_6Ci}&Y19qkk9bp<60S-U9X=GW6*~vAETuv+kY5_;+Jb(`U5r|%WPV} zg7;;dt(rIkuW3dg!a?FtD@mbo0WudxFco=!xh?-_K{dg&mgSJs|9k2>3&rX7rAcZ) zbR7?Ae5=jgj!pKgEW5+uWt$tj!>bpt5knCreqVrRX3O_o9w%D;6#>7hlFmc3!&0Hf zw_QRm5qFY*Gjypd3-ha{0$mJW0_eIh9pPfc=pHo;!$%chB}=zwGj0!^VLadL@O)C2 zGQeI55F5MO@fci^RY1St4e&)rN~Z~Jngt~Xfx}$M_}EW>q|cImlb+y`aZoS$j2JNIr&)pu|CIh$x|GO(AO!Cp0s@Sme{K_mx1@SbS^; zEwmhh*NHU&VG<}cRu@(h0+oUl&;dyLL%aSkX0@YWC0ssL(4QDVO^F z^-+lAXTTmP}P)A|DS?>6bm$M2v*j&mp>e3mi-Ln^}~FB9tQwfuOo(1Yd7 z`K@jKU}}USqPk>;WEN;w-WA!dynGhy939YqkN59u8P@*IY~F$QRQ^A&#niHAtSOZT zN{S?y@J3P+%O50D5{mP*Hu%MU=nc7al5hz;Z!(h5oCG}X*o!xbgmT?6-hWuf=RvDP zDZYl&Bw1_j|G;?=nh26MDZk6#Eoc41OuO?dN!mtt7N80T(xA1$A*0Xsc1@f3$qI3pJ@3-#|Kj!DuJKLjwwsiqPb?0)mt}*jVZu_SaYdMSQ_Cfskvxtzn z!QVC!c2u|yww1l{fc=A>_gU+z|Fm!eritHvIJ{rSzwFrRH4G&yCuR zyV0RHIoqG|=vTVl-hC#noG;w(!CBwkNAYk=S5-ur3+bm_oJ)(YipY0_eBydcAH>%a zXk;lRvO9v%B|LJ}I5T5bXd*?JpmiTpO#A;1bAKaa#c}hzlg@BT%TK3Wm-o3jw8MKB zbXGsTZrq$oh^0Cr$9x(dG5~a^89Wk`TL*III_5@iGL*^qO(}D^8%NC|GIZ9x^9I51 z10cgpb9+m8IA%e)@o;Whs~ryQKy?C6$(#cL>uj@+;>BqQDK2fXw0le!Ei~+olmCm)vt8`Gnoir9a0_|~lDZ?C z|EE$nVt1c~d8f7&XRqzcp{A9&0VS>>)P_KH87nRClZm6^MkbxYy;hlR7gUn$%@ZSO z8D6rrA!2)V8nJJm(dvEt@z-t%Q6CeBbI!;r%V%aM7s-Lw1UcL0pcY;$WIL!ov}U*N z8d2y$xOOk1rqiIlXMS)26OuH9cxQ=mvYDT|Iq>A!$z4>ed!A!vsq>c%jmVA~H4`I# z)Nt$U5qt_PU>GkXiq<1D9e#!_=7rwykG%^ zue1#Q-<{&AEWVAZ)>F%@)R~|3J~i$N6X&svHno4%ZBXf8W6XuwQR&*`Is?d!|H)s6 z%t8Z`76~W=3#Qq#@v=Y~Dyu>`>a83&=zmHQ@a=r8vl1v*(v#aeH29ySQDX(= z>fhM8`AZJQK88kzTht~B!MxX@W6k%fihw0titT@NP;@ zjF`;y?$WLZ#6RQM|Mc@faTWGj7~HH%IKav9FcF`|Q+?SlrT<-T{9}S_|BDUX{=^j{ zlV^?}N(LrAN&?bjsi!hZ6u!SWgZJYprUZ9v$=P-{1Uac`%U7XHBj;yLA#rMHZz^KEI%2 zb#A*j$s-+9b@W^NUEVr>z zIjjYY_WBK$U@BzdePMNulE3hDYx8BexY4$H-`S&BMkyI7~^}DhwbvdT91pP;=&Hz|O(z(l||DN*p6z0i2M5lb zc;Ov5(j2kb1`Cy^oK>EpP^g7<L|Z2BCA%UqT)OhY@14#6^Q~W_Jk(1YKIFUA z*w*8!bp9=}oKbLQwXgUTuQ0xqMweKnkgbiy9bhf{KomX+U>mJJDh@oISlF~m09WMw zSa&1ZWzn<}XcWd?s*~*DfmAkvM&fUZDHTivZIm$1de9=Hj}8G0s$T~YJ( z*Pp1UA=#ENmmLke@g_?+$GYZFAAp0MHeFvGk01^!kILvRyd4{5w z9+VDT=KU4#!(Mrx#=8wzs;>fxC>~T9G>}_lIQ~5p>Iw#NYgI{eQy7RDx3R=7s&Czw zwr;(!fTiCy5UO%R~4owo=ArLQ1HnA}_!;l!EMC2C(!o zJy@SE0}y`AfXpNojC8TXgG0xwK)?PZ=8z-NB$mbWu8&*eR!kF6|F0jlF2`@b(YxB> zI$dA44Gcb7)&Oo*1E{W(9~#!PJBNMT6$6SS8@p-W9@G%FZQ#k{csr^5plzLT8N9Q{ z1P}-WkN=2|uF|>#?5`UW5_|?{=Mrawa=Lg@HmaBKv@JZf5I^aA4IoLbU^Ld*_PgTz zCo_!i3h?eFd$oig7H{HKQjYTQRQhbG-seQRx9Ek|(gTE&@NCw`gSE+Vi-kxMYM(`w zbG?Swl;aZ5$h-~mmxc;9kXMd(e?ozjz9e48ihMW*uT>Z?Ujqj1>s&AXAQ7K7wyI*9 z{eg1c=Rp4drmwzMc(DoIyidQzSPpT=IYUEspbir`U5d4)F^sS{%JELlS3G1Gu$^y@ z-k)$TuXCWzFOQ+({Do!)9(ShS@BsSI8;CF7g?8Ql-m?s5s~`-12*w1kV1?qR4&Y}V z-;qnGid^>qN|gybEDDzIl?6~%$OIj3+BqvBhgcGxKo%s=V5_~E&jO>+tYuop%5QzN zMUN<!uWlO_MS z%voMD5p;h^7k4}ZBQbtV_|_*DPafSUZ~XQv?xOHw#iVP4oO)diad>Ps>bpct1k1sb z>D1Qd6p#6F^gUWhGyQrkI%b75nF>Nb%;m^T_mZ9c$h6a1P>ZkiDev#8ZcXG;pn5Nd;Uc$Oa=wfm(yu&{jrup?ElFDi9mGkEu`4>aIAoLCvmkU{GDZSeAUYoSJLfIeG_$qGqw=)gacoil6 zoA=PKig!|}m;Z3*;q}n6xE6z8Wv!;sj)rT$lEsy_&N-#tLA_5qJjJddCzzZ|`u}X6(mML86c!l5X3@VGl*rcOh3VyEY zqTPZW3eYcs3>Y)(jtHe5p$Q%~-l68QI8dd~?yc*Hu|33nc@4ozge9#TVNmMhA25pV zGj1~*RNYH#TkMXcrT#e;#ByP0I0Sg3O`IOgZctqGH;UY1BQ|9j;vqfk8yn|rp(P=f zF=>$0Y{cp740kxB1>v443^NPuNi%7)hAN%OT1OCPuUh zA93s#<QY&{_W94t=nXOf-H&W;cNScQO=bBZqaPlq47Tix6RtCiB<$q00jmkWS7 z>VxYt=0Ln|FCQhs}bU%b}eTU}7<$u2w!k(*O(4 zblJgqEX_Kb(z%d^^1=n9%LHvc0c~F%0pn=9c~iA~!W+pDfY_+y6Rr^ax-`VJ9C7cB zo`>0`3C1CxuatY?GqFCcJ4$;R!7sV1jxIsNt6{C;&)HV!W!acOm}4D!*q6!GalP7g z-Q$@M)X|c#@a5g~XS_o_A&jv=R$CXy6A<9s^6DL9=GXW)u^+lRXFy-$RxsXG&ZvY@nor7V-%DB1I9hn@7Xx0Q1?6=+oNcVxD*B%?EsQgO;+Kz<4ul%A< zwj@+=(A_OCdILMCbPNZ+EF-Im8~(|tt4d%b-Siw?c+{;=S-g7IVNtDN+K4-m)_2yP zl_VNbK^(t@uZNx=gyyNTtIYywkQ0!D<(hLb^NV~=+3xVWgbAd1feT~$H-R?y*m>{`M3Nin_JI|o9? z3I5+#HlAJd-ygrF>G@1~G0Cy#W1^95oz>MZXtRdAR8e3q%=I~9CFR7Z+hzKP^F{o| z`;-Ls*sGTXWu}G^j~i`XkaK=(xs94%ymmmHug@Y1-o@Pi?!OBibpCXu!EZ?@it#!w zL0VeA{xgmoj4@k+GlbX7^V$7ib0IPLW|zNGi`u(2?HM$RW zL}A%H>w|p4669V4aLphEVYq{4eivA|YNF9O z9U##6)=*?6)9kH$tvdkg7YkgIMmgkHI)?T42cpEG{WBqxg%_&;Sp=3Ql%t;mt8p|6 z;uvDsUHB*wqQNi^>*T?(1Ad;^;nI9w&WKGcCk%h|WEFdMdgtf}F4REpt8>yZU!s(R zzEjsA+wY^ljis)nFruK130sa(38!TijdvvIERaRKK23prvt>dgcPF^8NN-6*ktp+6 z{H)&0JK=xBGsb&(VM&m)Gu%WisQHm&@vCm~WUj&RR*gq2qNBtWZ0$U zbblRY$a*(~<~uP2YYH;s1ON_Bbqt{<1EMg6TCB(V8)o)GwyLN(GzXvKH_@@ zj})p!Cm3V{DF{u^vNs+z-0W35%OcW0UdUb~qFPIF&ZxRF+Zle?ExsD7>fF;48N$(_ zJHhYI`>Ndei-A0A3WiH9tqo#>Cx#AK>Ll|dA)mMKu*wxKi?C&QD@}96zUq$GNM{dcwPv+mD)Q{ z!0ze>0OfBxx0;maClVHQProg1mo(rgJo#E!d%_k&8bS-ZKA6Z6-*h z(GSy!X8C7cde}MCQD>Q9S$0|G>orlE8;XK+S!pWfy}v@bS}PY_p~wWWx%Oc?mxM~X zMU5Ijy&2D|=mM4($u%FD2ED2%P{jSql1#Am{vt0b>}Kbmh6 zl6mm)OU8~`Ob%_NhN}u*R=`D}LPPr2u#+}4ve3ZKZ(G!z=CXK&8jmVXr#2Z704vl| z%YK3{saZf7x{!8+aljIm+u+tQD9_d?UeL%P$O)v0Y5qT+&cd(B_6_?I5>gvEYNUkJ z2+1*Ol#-%INlHBe(%m^=bax8)h)9WuGy|j=_(&;@fPk>k>3j2g-_QF8>;sGYy3X@F z&f|Mj@<^q=`N>IXgNpWR?o!Y4nNU1!H68C%w6#WBYDqc|SH-9d11lX{*gcV5_Llf-Q7kAqoE}UVi;6+}Xz|;Envn5vSB58$RKt_S z4QWEYO`aD42NEfsTuGHNjn#T!RdHL)WMZQxx|BV2e^NP>6QB4Fdat9|c7+sm@g( zjvN>4tDG;(s4M|I#vjPIejbni6jXqJl{a)a_GHip^YN>rfNsq4jaosuQhJ*8HZatC z38+V3Rh7&e)2zJ#o^a$`CF_hjO|jvUrOr}h8pmL*75m?$PhI7gm0N|Z6<&?N&H}xM zM=hMh7`L=+llP5>69`FQvLVu@S48P;}o&T!pX~@5yONGkU zrb4l=#kVT1C&PZJoR=w-+n1|Jm@?TtqvUnY3C&9LF5Vu*N@FFk*OT^gG@I{;-_LDc z`Wc}O$!@1z!DkIhCFo0lx_AqQGv<25m?sh-Bn?Qz!YACbz*aI8>-A!0Z&dzRSY_e8 z2;@s?&Ille{*U$j28t#IA5U^uG}I{`XR4p~SoZ_rEL(_=yo@)q$zZOMEB2%$*Atf} zd+fE)@97t8?U%^E|Hf=mL{qqF0M9%$8>E=a70=d8hUwT0xxRfxX?7xOCPB zlc8D)C2SKE{3nA>6n%*C6;fW8zav&RtCT7JTD%-T3c*r>G{bSej{<`gHvCQ=vz~ak zO@+@!iqqG->;O?Ufa9-tdSkEwvqqIMmC6WjOp~F!I4d}^%Z-A>!iX^HfnRyHd6zKi zh$b2awkQvV;(1Wo6@}SUuLto|Tm)%^H`lKxH@j;iB@yfa_eIfNFp{HfF8$q!PK@q3 zPWv2JpK+}7)dSLQQ>pcZQj=o(oB5J&&#T2Az@tJemi^t{ zV0i&aHk{o!7@mH`9cZMno`L5pV%uL3ajtxy+UfFRoyKsC)jyqI3~A^~Y{f+E@Qa6y zF0Mpe_p)-#!)-Xh^O)o_+k}hNaNJHTtD=1(0rx6LQ;E_^L(1o1yB{L~l?PZzy z)YqizQTm~zCpZI{qrVm@nKWTW)RhyTL`tE8nyb7WSH*6Z5Yt8rnY~0;j7pvt+a78s zcS*o}Nhy^YS|0y_zQ*WxWNVGgaFTagvG*bp0;to%7Mefa0TK}Ws7W0FfaMO-(j^XR z&^$9VVN?wNUco^87nEdAP>TDJIy4-tyeoM|_>}pF^ z)qD41R+;PfZ28~vwg0-ZsBWL(R^S;v5a#0P83y*muq-f*cL21nxzlND&zK?LWKI@l zn)|uIDGu=I!PqQ7imFJpQKwNhecrnN21!4dmTHUtv8x?jw?W6k$d1G=#NjKSOO{J; zThq)mlbZ&wAspu(DqEow_b9f6Apew!myuUK|3~?<)b*?(ka4`nR|#?dXJ|hup&40M zSxjkJsti6OdE}FrQW@jVm`xMB3UjLTR7_}}UrU8yDzM)AimcfNpDY^r2U!StQVf8w z^`dNgg=R=j<%hvP=|!SW^yI|nGWJ5*`VVO*rnX87IXcp=(h*W7&Cfzl-+92C85DEi zwD{6u0m(n)I=*~i#H(B>nTi|nE$WDlx&qQ;q}*jAx4}s!k@F@(AO5XkCbQ8%2F*WL z;$12ZrtzRv@<_Ej!L^BO#GU-d#-p-Y+KJV4Uec*Nwj^|3LX^hfhCa7f=_qaN|AB@KDEikvSdw}CUkgjsMV$?6+jCmAc#nOFGHeOKf74D!~R9Q9E zDt?xKe8R8XhNXDxw>a(2Z%$IS^+$mU-`}B9ct!h-2Zp3`P zr8Qf?YAXJhPUm@&ozq&g&~V<_SB?!dfz7K~NufBXQRWxwO zd~;hatPI`+UpSK1331I67GwQmfx%MNviGu6=|?FRgEG?B7Oud6<7kprak}A~`{oU{ z^j=4~2I*7XZZEM3FZdU6#z|npkq=xWnG7VpAKiE%$NhS_V5g=++$j*COPdtp zI>ifj?r(~#z%`ZiKmGD+yu0ZC)*9dJfJHbv9+I0FXl{+riG68`iAFnSkTY|w14r$H z7Kap1f)vhJCti+L-kNucmQnt(Tj8bM;U!0t*`R-dSs+(lMEyInx-vi4PwcdlkI{I;&vHL8F$BJC;~jf|%%f$B{2zPYL6he}gOamq z?kXhou$O&CaYH>|^*rs&DQvW|>gK?E&<{F0^oc1cKIO=O#N~B-8mIj*RjPl!1dOL+ zcf!SVDrz=i1?r=0J!e@*H&1kauqqCeN0-p+kCejEuF1*X`XlgV!>kaANS zj)X>xWbTDg-tJK1z-+~`9~1IB!nes@yQ}3PsnGGba1HiJQD@RjF{o&K*XOg&yq5lt zj}(B3yh35M(rye9BW1yveXpqeKP{;R>E9W=RYJUPMW%8O#^Iv2~;y?iD0 zq(dpNCK-wW#|Q4ZeD6TrglwBT82y12gz6vNQF>4uGcVZo$!|%pINmJv7<(gGF3X3s zvG~@bi)Qhbf_1mnhaBxj3H+xg_H;NV%&@yjU?K;&fZ>mULIe-|7Mn?qU-CkJR=JoR z$>LOx{5@x0IE0-i&Q+ugglYB}uojrvulF;pEn)Ab>H7cJo@=^z*aL^~+XeT@mM}d> zfvf$Ps&Uaj9P_`N0iU5@g4ySo#oV7!LtXGEUgm&TsSZUF;)8P2Edm2T#!5e zVrG0_;2%7H&>h$5MpkCrL;~V>$G5otg*syMQ@9fmEaU@hX>Mrr<=% zH?HHcT)L1{5B}5UU%+C#_P1|TxU-?w@Pm^@&EV(so;ivmiWH=hZy#TE# zxP@Ea#HV7&f9A4fBRCuot2+5 zG&%H_Ll=p_XN+Q16?)RrDpJ2b#b86wYvnj+spX^`_P0zBn9N-n2@Em1+`m<|R;5%K zgma%(lT~e9wV~vZm*+rdgYLs9`%FM-gcSlvrV~vvPEi_eW=Uc2k1b62wH$Sn+Ny1~ z=)e+_=Bh7mBNr1@tH8m}d`08RnrOl`yn z41ER*cEA#p!)-VQFW)#0@gGOjkH`)2`)p7tDWV=-B{%Q_cZOYiOwIZZwweM=z5)i5 ztt5I8*=xWtD%Vf=o4KZt@b#n%yK@$dJP>wU^+&GdJ~!1MFbSaR$&*&Oa5DPFyj>Ow zz9(M0u0rJzvj!>xaa>EVEtaM!=ZlZ`SySPXWl<@wdaOt(HT=mRJiXwT%QOPbWJF!|IFFo`<$9$Lb5~vh&xP2t3$v_R*K>5S(i21wYM|M$x2RlffX_ z`)2mvKdkJ6xxWRbO?NdDtRk-}Ts~u zNH;VBZ_QFOzv44m1z7R=flg{iMBCl~?x{u`*gKKqkv=Fm^rtUIL90pU)zfQh~5w$;H=yG7M44Yi9uy>N4qXv11~pp{=0ZDAFE==?XhOb(=>nrLAu2 znRR0mO1EOY6D15(2;(Y-14C5@eT@{zR@}KzFBk9+cSMet@ihvJ9ki?u-h3bZ)f2Uew5?7`$nL z=?e|sgjgku^uID}{!&FrqD2vyRicGv>=*yW9>6!&OfbyA&hjF+?W-k`asbv>91Tn{ z>9#qcR!qvQ%1Y}N7D1KMtQf-q#P!v$1zLi=9B&WGz0AgnKi9-A&AQFzVPmvR5fZvq zYlSJk8ih{YTYq;;-dMO&gcL}3hwAWFjZJq7yq-c9YtapsEqIC!!`@j@+$!()Y>u;^ zVLAWtkt#LIX`kMi{8?*M;@|5hE;I~meX#(8VIGh*1T9|w{+WK{gwpGinXddK76yut zu?Pd9`FAE@H$!@w45_g(?cG^s!nH z5}Q#qcF9u50gg+X11|t%t*~z$TE(}iJi1&+_GmY5cF>nHrM)zKxw}@20CohX-PFdW6bgez3_K5VfX2k{AiA@1#MZ`XtAW9a+MT{hG zX`opkSc#P=6^2EtE~~=&AXGS_w6wkE5Ip?@^JqAR9%%sI0?TMvL$uR0fM$xb#D(Vr z*;BH73kCnreW~ci<#||uog{NhCY&cJSL5?i{}AX8GN0PN;|QapVejz>%t`z20@Y@sw5`XSQM0Zb~ki0HTq)9*=+H0{*B) zL6Y74{CBRjYDmVfC#9Vo5cNeTtL6Y=VF%00xvtS4_Ekb8QJIp?rODb#k&um84`^71 z)sk=Cvysqd-QQ6mf<{30LR@yWU)e#vE;Z|OQeT+i zR~9%n@7L4V?{HCNN(NNwf~&1;pYUCj7=ghznj1IIo6po;6r^ZWaZHwCPHyE z6ytdj^L%^sYuZz~mcY)(hnswr$v23Oo;UiEt}$L7A!kgKkWyRWc$|pGxin@}jQDwS zzd=?b-=E22`bb3L3FcVk%d*LoCac?6kp6pa{r}BQ0U<}~sOd+Lf7+v^;f~^E+22@P^{I|C_Vq0V8_}IR27|d>( zp0}-Eea+k9SH{hjAuo%xCeSR_=It8r!1Oe}sAlxZk{5C9Xv1h)rYB0@xKC`FgsOc| z{&G+%wme1|LQX~XGyY_iU$>mDQ~@@iJfWyMFEv!nHQZ9fWMn$?108PM1{s{N_!LGwdUuT_7Z*d3NcB= z5#ur9+-+1dgzs$J!-N>5AXAaaS|J9;=>Bu|u@vtztMZQWjC!%ns)_19T1xZiIC&M) zzHZ#Fg4e$D49<46!W-%qTd>;iy`O<{sn192#S=>>euSvE)RE*cwQgRcHC|{QsM6<* ztSXs#bbp6$DjgXQkakP|zB@)Ma&L@jY(l^4J{14R)aXxTyyu%6Y$k9kKaRMG6~9N~ z1PtXF>`isxnTqzR3p~&%fLq(WOnsejqI~m}VlTR-N6x9F+v6L}hKJ)fZ0Q&AP(G?H z=WlnHuzv)Mj*xX6!!f>?&MV!LAxZjt)=Q62DGg&9mBVA#sKAOkEGZT%p!vIqL0dNN zLt=KSn>ira{oVYX8Cd0lpJy!TboJ=C+@zf+_KAZ2S%B{)pP`}}%L<2>fV7s_EA027 z=YsOlVKV9$Atn9SafphK`$`+UKT`}_j%e0(QQxCqCHQ{jzvj69pY>z87|@QM5z33b zYpwrx#Ek6%c};ln>Qj${V!5MuHcOqCfAWTD=B7&V;Sp55nXkR%){077?vbIJPai~! z9;p8lV&jFn@L^S2EGO_2rex&d{!T$zvyBAJwdUa_)eoYa#s1-fSSxpj7xX9$a@NAO zPKm-ep@cO5kU=454eVz~3sv{!2x>4XR|F|M*->G$J3KyubU>|;226|%cXbBg!b#PT ztmupb%RzBAeHPHiGxi7yl|G8B%ESm6PI~U(#$XX{Zs*lth&Punm%D#YVK-VE2_BB_ zc|tZd@cAo$FI6x5=4mhaCX!4Qttwm&6?jFg0u78HrJmBxLoxa*lL1pVg0 zgKKgr?P+c|Rk~mfRXwj|e=&3_2jvN9uZ9LKe~#Ho%B$%Nts$D~KLJu(ED9Kxj-*Na-xS%;248FlbK?M(pR?DE|g$1t? zye_3nVy{w}rV?{?nK^$C|@1-C%y#%Cwg5$7GZfa^o|sd+{gT9|!z~QxK7HzuR{J06dZ*N`=#V_tyP+k`_0&EBHKpg5^{%KvBo#brMZM#O;s zlZA=$;Ff%DI&aH9!H?}^Z|j$QF1I*l`QHfr?M-Zmip@H~k2E}2ZtifZVNR0ENYd>N z#Txic{A(OSJMD5Zo5T-wo8vQL-N2gmk{y!$b?PWg8VJA3VM6yRCLOAIuD@J1N z6kM4kOAfKM*$O{ZTUA|9=07weIoIq`J*=eEk|aUv)MK{D$L@w?9vUH>G7E)><;gA} z8^5W9W^?wPqAL5V6aRZAAm&ZWXI<`GR#2rf!ZQYWR>oc~QtlHN1s@(^{Z%$y##Odp zV410upB2|!Az98|c3%4PATqvn>fIv{Q`>!c&}D}7>QO1Hp=f955NoU@W{tn>Np&Ti zZt_XgO%x;Y9F$M7aW9i%QlVtXGqZ~Z6TdjZ>--^tU@xIf<4ctY?c|ykjdR(q(gil9 zt?@9U4}0I4Cb(R^d}B&4Cyqhe#Gq_4T|{VmUb$Rn|ra@*?~uMpo~qLPe*w+=ylTz-v*~U6P~mUM(O_$ z84`Ms8xr&I{!WWfie!lLA~}Mrabl$G`}JN(Ks04-zCZA2KtGTq5&u^er7Ubcsdq)W z41SbrgZd=lP-2V1ED_h{IvZhP#2L6*apDx&hs&;LeKAi?i0vrG1EbCs%-tIh1t*3_ zgT)L8%4sjTA8|qgSIEKlGabica@Ef{mAQktM$NEK zDk;h+>|kQ!oLbzB=OLgHkbkaV9)o-Mp^+adyfEs}Qx6}F+MuG>R_LX|J>kogvmdAy z3xf~YNDUs3`o!0$DH0AEra;obOaUtEhtRVroc}~HWDyI69jQi_?}DSW=$h=$5dc&% z$HAf6ail0T+_M@EirtcgajGG!Y9B{j3>bj-(=c(H3dBN4D`NOmL#ngS}Kwtxx2Hu8F>y`C2p^V-Gj`l>pUjeO{ILO~*8!mLKCh_g&1K5BVh*H14NjcD9OrjkFWOe&24g^PSSM`=nrzX}0RKk@sQX)Fz2KbX~= z_5;DjcN%q%n15axPs9`28a9Wl4bO(B@{lbNSwf9ZTLoi9^vRdoAzh zPWbcp&cIC(ahhSTIR=^V+^t+J#I-NuR|K(}qg!l3BX`%4`r#8%CV62k(jzOkt_`~8m6D!_9Ee%m2`M0mG`DCA+E0`|JJ6LtO>F5U1LR8;?v@N6ay)-NI+bocX>{0L{gKVy4sq zW%6Ql3-SiW#~T{TPd!{LEBKmn61J#nv~pF+`hu7}QGXTrrEL5wk36&e(8mNE+q?ey zB#F~^{qDk-PoyiUs8aRk={7eI`pYt_njch|G}cxMo5?A=f%&%XqCi+Omq;dS$v3%c z%H#C~zp?v1)sM( z4#jpm(;daIX`U18^}cFCmwhHd0Oa;B!N3l~T^zPbR!p9~#D(ZW{r>Y-voWe z#g7VZma?OQ|F|+2v#C`4 zzEx~m3}4GVe~0g%4~5{LwYf7h?MOaW_DMu)F0I$UdCa&DUUeO@r9FJ~CGo+Gu@eMV)L4$pWgHx|~Y0wdk)M zm;fO1Csu!ttxkXhh&cp4Ek#y^sG)ZvSRG{>kOl0WPm2_(p8HR66E6uq22kmjsnUe@ zdh`C>rMo~ZTqx!l7Pmh0wA#KuXm6D!?;R-K^$43 zaK8$B$niU?kHDt$4EZ-;W1u+XddJFdL{QS`KZ<{w<)Xzfs`HG;nt4BgxTm_ksp500 zw2byNO=9Np&1p0$R{Wu=pe=tSr@dxDb-U9CvMf{xDmybozH6@i@Dq)xfKZr7V^QP! z1^M1%b}nd>_aML0x%P8LtJ^g7K8~AuI%hS5?DaI?h|)(2zg8lzCR>~+f~#Ew_=$t( z84aonr#~6eI9L3{LTmQnxbSwvxb>7^EL3ez6R6RsCd>XuEA_pqnOioh*uT`2sp@oK zuW*d@@uv@-cbR|T-~|P&vzuu2-(uE~Mho{ir>?q=v5Q^1mOj#og1y0VS?ULHA|JJ* z?oQ9_xrD8V5LfSHU+f{5?bDgPHRf9OPoaiFw{EQq5_T~b{DS%GGId_1Y_fFH$BbPE zk#k7nsrlXD_gvQKV+lR35X$vgkqi7 z**i#IWHsKkY&1%NM0>DZP!CvrqdOkPAw)Tgh^L13GxNm4ezR0)A4%htFR*%oep3VD z>!f`Ol;bek3Z3&JU(nR2g#Mkr^%AMl(=sf;DnCPd4b*2yl#Z2sup!$b@-R5fBriU# zd9_+0JlRa<6uB70yEK$Dvjb>E!c7$he4Rc!~$IwI^!F1}S;{6O8~ z(h}m1Vq#Nu*NcV{vlCN@nMIdUwrfhm%g^r%XfcsQkJ7#o20h7TkH?lW4aw@WjBB<{ zy}V14Y@>AGY{xOo7C~x50tTg}GhahQpNx0b3b1kne-2rpuyWV1TaWQPRb|*z$+XLc z`2f`I6n=J&g|9;P6L?{i41_qHkNAOA==%G&p4DzOJWg+C`_6}mxbH_wMY0kY&V|v| zC}HHI<%rebGhh(4AgROF)J^V&4zQV&tyVVPQ99JnDGGOSj6GoLI#YLyj$=}rQemE^ zHkp@kb;^_GJU%|tbs>BCTQ(v2bG2c4x>sI+fwXZ5yZ^4dAh5Y3=m_1~CwLpuP`kwC@Zv&U<{pzB zxw`echd;1S~TJt;d_>8sy7TV(!DUjSIoE)aZN6>@9L2Erni?sop| z`g;nbu~Ue7XGDr~_i|EQqh%p~D>O?z>Zyp}=$ugCkX^XTWMRE@@%@tpeCd4LZx^sp zwEQp3>DKaeoABz$sb#VC(wVBKn&FyABs$emD`qjOWs9G{j#DEPzV#d;T^0ocXC1-h z7&@Ytu+y-@-aEi+|1`d;`#ezBabo6(it*jV<&&mLAJIv%96e|r65djB(cDp0IvjX~ z&^t1_8K?o=A7E|@Pk&6v)+55OWxO9C8qF}Hb1x-IsEG8dd1 +ss)_yNdD}EUnOJNGh(pw z&Yy^-bebQR);Uo)*Jv)av>C22SWva#4`4n_;oS+deUMll_tkA*Oa3gu1_2DnxU#@?$-&J%%_z zcs(a?0D^4zChn}vtM;<@E4mdV2!042=42JSd^w%P;59oQ+KsDB2@&v{8uiOA=A?q? zKDhzB5~oGVYIFWQB>Qoh{0rbd7%`H#!Ua_l?t~FGrtA^$!hG=le0g?;+*N;RM^5KJ zOvmKRovIlE?Xmg)Kf@3cebI!dXA?u z^5}6ri0S<3cvVH`WG~wtuxZHI=|s!c@i;8q0I|6rS>0NDm?as!Y-anp2pZ{SAk41 zr+>Zx-w(@O9!2;)`KYd_QTfw&Hz#a|^Y!p-#qzqjZAv8M#0M3+|6{s#BDCRD8BKAE z$Q-t@J8JLoWE8e)_cf$1_Xqs7Zjsm~rl#j#_8}9NB7IG_L1$e5MBT?$5RgQA{lc)= z$f06~2jB+nbhiGvX?K4!Hj{c2N}Z(87I!VbbW2*^$`R5CyHWnyWKF2G%2Ece@kF@?Kn+% z3j)Yjqp_Wd~N3uFiI{^H$FvAI>>V9kQz=Aa^ zaC9n%)(QjO>rrRZEFfpz_YiFfFr%8va7h;D)dR6irj79@P5#)vL+-OI=;TXre>JP} z_8L+?gjcYaaiR(hW((b%=GH8=WZ9S+MtMBh>h!#@V3G}}D}K?9Ws|q-tgBC}RjW0v zW<1j{u_t~eoD1b$;y$(}}{TC+Q=x9~E+!`lLV-}#Hv+*V(v z%2r#c@aN8g>6V&PZ*$TTnt6Hq{z``rw!~v%wTP@8gr-NAzBH`UM}t_)opK`=iANb+b&+BO}<4#^tZ@1w* z@Y_*wc+GZ=_BX-TcnL_T#Fy+~S^?7ehXFiemY9)~ZD7)ydp+ubiR$?gw0H=WG0N$F zi1FkVF6onij@}`i+Fc;+`K$X?wZkq|EAX?PfFnj7uTp$(WkE2ajiyJ=na09sJL6Ylzj1M>;>6(oSPR*x!mTKIj86J|pM zXkwP9=OZ_wM_T1JW z=<9ZBbs5PP*A_o-h_Mcf!l<ZA$m+w9q7C zcD!1qb@(>|deT>auY-&|9$s$&YoigWNj3In^?rtams>h~Nw9jDUmUmZB<9?>xdE9P z>&(=hhoB|Yo~vzgi9m#(+*C$LJj!9FYcXYCpDwv+en(_Hsa&AGx%x9^hC>|s@|ZqZ z%>E;UaxMIf45Pl7An11@6z8x$a`y#=NTSr6=S*CaEeBF49QWo%5#>N3aL{)Zm9g$y z&i>iS`cGYe$oMr^$>I)|PtV7d3G%{2ARE(fAU=~_!+*`=K5`39PSn(mGa!P4WmBT!hJr8Qlh|U2kbBdAePnW9%<)d51DV z>DM3!BtoZi_ou>*i!JEhOkQpxe(=>7?k~YLN-ghm_%g7`mP>H=t6Mdbk zd2yCJs}9{tgvA~aQk2X|94=iW!owf4K}fh=u+I%0P-x|%Hz~*V91r|Dd3dKp&ypPx zcqR!8p0Uat&&`Z3^C!QY6BtspnCvIc!Rp|@2aG7Jp#5XZ;2S4}=GVWY0@qGex8Er8 zc=ABGvqSe@S$Qoov}J7=zZ`O z8_KUqbBIeE;4}Rs!BVJx`n0!g!_eZzs#inv-PlO>RY!^2T1kppOLuLB(ILxg7YjWEL zo2ep`BBW2b{n&$jG89WFF0f&tb6~Gd*Z$8cZ=^ORGmU>i8gH#M9wVw{HtC7=#tN&| zcfFSEv^e8ETWUJ;AE-aoo0jd;1hNRrk3UTW@YeIsS5X<=&P|KU(-M0Od_Dcs%y5wS`PJZcl`v56pA1WZ=LL$_0jaWVO!APn&!jG*`A)P`WCLrg-L4FfZtjTM zDl6jo*GAwgg3O8RzJ-;uczmh(f0{+QhASsjYl5kLSP~mz=|#jNFw64Hp8hk4EGD$^SIJN!TJ=ylO-S5EXcu@N%Zllm$yBUDAHE_rLC>)_uJoro!247aY(fjg_1a88KrlWlg{WsM=8edFZ=OzVcE%TtpTYjCbDWJk{6D#vFqnhIen%k>)M(t|(FQaD; zEKUxsi{z-ca2cnYm$8;Ef-<${C-NR8u9!7Md-DDBvC&aH zwN06U2#rth@A5oSKe`c{&lna{w`4U)M}j}n&lZMxgyN}ekg)|}Qc)h^=7BtRrf~gH z?j*@K6*MnXw7QotOw47Gtt1U~QZ|XRWJ#v+6`+C@586hQe&xt4kc*UVe&etzw;C^; zGi!R{_(+LITyWKTA@PkQO#>AzSNudhbPiWmH)SuWFW=DlgZ9IkSA(mG8_OEkO*;Ey z+ZQMid@<&AOU&M*-5C1@Hw?b@?n1#!aa%dz99PZVY*)3lDeo_bG7G4|S%8xqztVuv zX;0p0M6`446elxgbhhj?&Zaom=6a^tvesYELNR41p2i>Vx#}<{Hh%8SK zzf4N#Q>qt~)Y(#G@9${0Crw}`+mEWgo;*?#vrlOp@8m`d+#r4PZ0FiJ7Sd-L*k~QK z<#|nEdr7ITI<@&Gk`0)rW{9cl^u;9>+ij_>&1|1l+55hla!cT03c8Kzof9ql9Fgx4 z5~_1ml)E)9hQ2K_(b6r)oemBT-=DiDHD$fpOqRZ1(FWQad!{}zdhzk+hn{1erxtGu zjc`0*)FY0c@1{588JrB59p;VS#3kU@Qk04*?*{`+bAD~c_Uu0we=ocKDr{$^W}7zG z^17>T;DBkfmxS1m+SAe^Z=Bh;f_MMYx7g{>Qy$M0a^isOm{yBb>+1&SV^eJ+1Lty@ zQ*9IOaAkwYmjY%|5-!qo+DgPhTQ*56MsL@Rs#KmZZ=_q_As|m53B@H^@!Dq+i>IfO z*}l<6@^&f$?&LNAf|71Ip(VwmMG_g;+|0YIx-72yq#c0qWQ3F37j3vd7KKA8cwo7vrK8vip5`oX-H_ zpAKy&*;i2X%8%}ldK$8}LjTqHp~f>80G1sj(U43WJKLfooKp!~jM@tXVi0*x<-!`+ zB0stK@ewM!c$qb)yKC6qFUbnOKf-O89F39UN+Om>e>gW@NI|ZL<>A=Tj^w?Zj@YRik~o( zr)2WeBF=-ieeGBe&jlpmgB~h!oCX8p87GuOo@AvGhIO*iv(i8I6ZVkjmGJuaKora@ zuj~*G4}5??kLThy{IYFdcHkV&rm&6cK76~oFYrAg{QmpsIk8OjWYZJo1p(cIjY1XE z`*()cG8Ml^gcOlH0-kuwiqO2~-HEC6rAHsd?4guXOO)sG|C&RJ|8_JV(co@QQjxP} zbaXW4(lM9jwy6ocJ$UTOW7o!~@=}sbd3kvo8F)NogeNv<#9d7vZ8`NE{ECxN^(tJJ z4oAuIomPq424zC%k=5+@a8<+kiW;s@iI%;GRrznl#*oAFtAHYa|--d}T8>mq|-ijO!V)b3@) zJ+WPy8*;PUA(lVTf_7#tM2SH=JPPElLmRgi)8{9uNo%6J- z>rABMYcTckGB{F81Da(ysH2=iHj^jBbNjO9j-hyUBeFjYnPj9VcVBzH zG+pBj*;CZiTcJpL-lSAvwe))gjAqtpXI{YiBkzk#ugB`?k2q>%Gg4LWrgta0vVVTc zdld76N8V~=H@QfI?ZLOrkAH7yEBHV!cVE{W&^d7*s1nl0zu>J-16-C&r`nVir?Rw> zWPGWWNcUJ}A~gqtPI7B+bwW0Bbz%bmAC%vo98NsK%_S0aek7sgNjp?W%t+?lU|GvC z^d!^6>~A_sD>XuyH~v|Bwtc;D_E2E;jy5gNpxzAQr#TgBLI=2hBsl z{#wU;w1(~T3+4ct^w0jq#4lUKBi_(&a-2~02+@Mce}N5b*-dkHT}#}=(r0CtDpe+S zyQxQPInv37KPPxmF2v5bxh^ybr}s=DTcxy9@38EQ_^tN^>oR zYIByv@_CJ?=6*O91jPqOMY22i+2m~XgS_>}W<^FCTK;yII5|_R6Y-5K*Pcz{p@87= zBgO6@+S-jDqYdgQAGUqhBqbZ*>1K0(G9NBxm=n5{!Vg}k25@Pocq!O(i@f9AZV+{u zpdX~XEZJx8j3S%&5I$fQNNmrZ=naHZdYjH7#pqS;BDHnFn^-!XyZm(tPNj!)qDgE=$|4i(xkUt9oWgA#*nJ`81 zM%elySt0-^fWSy<(enBd*pgr}wdi?UlqC#VWVkhYB#geDn<5VDgKf#$pPTLTYxLsd z1;OT=1HH)Fv{uVD;a~i2snX5B-Xg}K5gca6{YR5Xb2aTp6}hEw$_DT<7$(fWL<*dq z%U~iZ{>b(p3}ar2EV;*+ly8sY+Q(X%{XTem@);kw7ls5lBs{<^{0)kheODNQ_;-~< zwip8~Hr@dwTil+8Dqn-jYWDU%^%aQC$UTP-oOoi=aBE({_b^do!CNjRx)p!Onig{b`HNT$~kk99`` z&kETJg89B@g$xFI|J1_>PB*Gw6)k;eGjBP&YM(gq4pEYnlw9c3BB_j!5ubbJhjH%9 zc;fYa%O*I4uKK+dw`(4UXy)c4s+ETrd)DVMGl-ktuV$23ewjmaUO5n%VgHikT@pdH z*`My#96X8V3s;Fj-vg0zKShxVq&lhGEmqbv>(MbJs5-q*2qvc7>PP?Cd>byV#Gv*d zt(}yCKU+H{qnnT}#VCHFPUqJ>hG(?OS~tsHllMdCcm6x6FMdk>w)=_7x+EQMxoY7^ zLa1N##XtB3JCh^>+?%QV09*d(i!Y038r!3iZ+MbRh5nAxh0te(J~$y37f{M%N*|lD zW>R5}G~;D^L$0&yn7&Ci58|}$GHek@*(~%*2GBFRyb&qOCNtNAmj~__dgDk}xm0pDlz=Q!lQ_}jIbp_jAngISY)>FQkoKL4 zWxl(5Y2-wIkf%%I&0NG=DVu~!{$!POT5XKt@i(K4m+!p4CVQJ#mBi0ftANw%rzQsE z!LW^h7P&$APzQgJ_Y+S`&Q_v8YF_t&P)>m%@zLN~rG;3e1YZm5p?kZ#sQ-xhVmdE1 zU7^jnwNib@oF#8-8t+%HTA~i(s-aXDnMjk}Fa*&;{zkp!u4IdXi6&v9bEM=(ti8vT z7hq#eSyz z?7yihyxMym_Lwk-S3dpe(20G4%T8h6=F1nRG{a_XVbd>VzvsO)9;O!gDoC11iV(8T zoX~?&@0v#zWD*UNE(nVJ-6a}^O&nVCrW-Aq=nu=8XaXHjoLw}h&}PsHdr4v`oIv`7 zeW!2J_Il3PIXu^=Qpccb?iU-sbECDHo*-+cisgv7+I|r)Gi|#r5zsQV)eC>mY3#qi zw&hI>h61PDj%K&4Vx6E-rpR~6pMTna-6oYcrks!Rt~HJxT{n6J;{(obR^kYTiI+T_ z^X9?B43{WR&Cgm&r1oCs8{L-uh*fxx_)vrPIo@~VHJc{eQX!8ZO=jXz&{sU&rfKJ^ zbJ5KoZSN1ycU+2jL^=RE*(dkEF$S$#TNEidG5c2yE8Ic7=TEd{XJmwFCLSJWtf&f) z_SIuVSo?xk2-`iMob{p;gYJqv-&RRI&o_=sOmW9qNlT0>7_vw`v{*jEQJfNRcNkaQ zTV0yKtf7DM7oA92?eA?Ko~$8q#Nm$8cIMwL~4dkkyg4J5fI6dLFuj$ zhVJHWzVDoS&z(Pb9u;TixA)p>ul25Xy{p?@!t5l{9O@xs}<@3DFM@0%$$mM(!y+a-+<(JCr=u(#p zxR1H8TBy|G$A`=Wf0nmJ1yuK4ec8qaHr6`Y;MoG7-I33n_jN7MM8}1UvcO^T&RR|M6MZB(djd#Y^^Chz`jj z1MlHTUF+kM;wfvR_eDE&%2VjGC~q6is1O9h4_{LGoC-F6)&-u<%lED>QJsxsvy4BU zuZP*QuPtyRj_-KzO0rFUaQya@wbR`@^LVlWotZ{p61gv8PQN|VFAvvgbRNqrX3Ihx zDIf$R2rDw1aV8@jk-k;phZ22ru&fQG6w{{sK1<(N$Paf1;)OQu@FTOP%C_-AL;Ui= zei@&+s;Z6Qjlr|>>4=mH-gWwsAf&?7$79r>DSi1McZ|Pmf6A7eqzP8AJDBrcajkuPmVnGk)Eo&xD*31` z9;{RJa_nfWyfo=pv>pFEEQrAwiLwYjGM}XZ69>x9jeodRRuI3(RMAFY z$i7;ixe%>A-5Mz`IVpCyo&QFxVsq6R6Eo(O&Lew?$HuIoLvJ3bl1NAkS9sYm7}+K^k(JSNHX`&zdYIm6%Wg_;ct1F+wRSl7~CVn zLZ};kKY(rc+3I<3^lncOX5aeZp4ai+o-I5wJm@0r)~Kvuq}xEp7Px&E?g`C)D8#396i7S~u|0HH&f=WUM%vzJJFQ z-njQ{p6O#`S#_w1r@#wd!gmnq!M1ZJ%2K@=J922sH2xevG1&n(?4CH9W_1Z+p&CqH z7X9cK8;R*$-+RY_C)QFf8VonSU4HTt*pZVvSxoE|->Y!OB~Y(bP+?YRk9a50vuH5s z)l6)aX3F1?`+-a2YM9q@wR34Z$M}*I;vzH70*yVgGSa1+3^{9NyD25f9CVMA(Pzt{n0l+* z+WagpNUYZSbX=D$v8!<}V88jNfF7Tu$%hl4EurI1{=oJ-6p%evB8Ku5ol23ca&A&W zFjTUC$^CWKFCQsZNkMV}-nS=WzP>vFh745_sEhd8M=U2Fkg3>Y^}1%1#bn-uRw4dB zE06mfLf%Z0Sz27FKN&m3y5&Q>?dSO|;TZ|0>k=51aPJMwv+bB?RA-r1Ll#ztPcTHi z;yr#?sk%6|dZ{)8ZsWWL9oFcnmwFbnyFSAL-DoT7atU0+is|I6#kT@7^srvYjh{!iCV8eWxkHcxwWhS*fyUj3Y$APbV^Ua{G0#=> zXXR}XvaHbVakQZL4pZt30v1IU#5;W=cISFaTr-z_L$|O7OZ-%w_Gkvw$f}tfbvaA? zpXpzAyYOGiD&JZ}}}$*%t=Q0`vO$!$&3pp`+yC_U#{bvTuEpJ4@d7I%<`2IC6<6Mm`%`gNRAbRoLHl@soU`q9Po#f&5aPb zIeOa<$0h5c-|MDpl+a1U5o27$sgwtPD|4ZJ?s5oB9MiW~OpW<@bhppC0g@hvl~5|! zXLZsq)BMgYl_@`i6_3<)%u6Zb78FY?cXA6QCwi%-|B-@aVgtD-+43_ila$Tx>ul@BC` zpV=zENA{vEr*$9nc>7W0%^q|D;hb9PKBgbvnu~EJGu&_XL7wzlTZ>gV`eeee%NEa{ zuiLzWV-+>+8esRaJdyK@P1XQz{J}`XZr77V$%DE5F$F2Uv&4-bo}c_G+s25SRjI11 z&eSBwMw+92Qh@f@z#)@#qGNaE$;VE=2xOiIXRQ<&ro;Zkkj7-);>V3{Pi2c74oRs|VF&kG5=`foNhfL3VAEqix>{ zJ7XIap~yQ++SBu@3)kb#^9XdxfY|s_%_{jF?X@gH+G!Z0m9_%8E_ogOIDfoBV2^+N ztwFh1+-VLa&828x@!)q#BY4@*E`Wr7-6W_(hT+GN&nq<;Q8lY39}<>rg?9*TVwDm% zU@xS01H+uGADJD72H%-y65l$*D-hUszB+Vo|AmiNX2$OrZMCu^`66T6y`EFLxdfCV&huP^!@9WrIUM?f+&TuHYC})Fb6Yc zVR0k&83TWLtKpIPq`O(C+*1m)!=6yG+`S-t)J76x_Jcw;ofWd*7+-c;WN@gakLjv%c4B4g$|t6j zvQdO)0VxHKgFmm>ZdS+mEiw1(B(V}#IN6$pn1XUau%qoW-mb=~lCjst@~3x@yFTg5 zK^_b4_kLv;lf_zQ;J&jAd=eaiMR#5JWkcgc7x4Mb<3k_4XTU>N1vG~tyDN41ZmFEN z@BsUL|BVJ6FBBBH_-MLqRRAxV#Kq#uZGOrF029Ib+>sXwk0KmZT|#|^9aSmk1$Xtv z4;}i}Ez%r2pDf}>RCaq>_*Q0kI|HLMo2-!XrXPtMshB(~Ht2Y|M~7vO_l$JRe6mui zQIo?oO>-of-K$s((`&sfm>=qHb8?_J?y%0~ zt{ebUgeE^OmtSJdR30&7y4{7#%Ky`zeYP0KD#QO< z=Y^x=*L}l5khoT;6sw*#pse!I1gyP0fd!auG~*-WA%jb=XU=-X%O)mWXrx7=Y6F|; z&)S8l3T6l9UJgumz|!8{xVJwQZuLga`nCn(S%$3BVAS|>|1NPiNlU03kwVDcbPH>F zP{bZL+N0Ef9bc$?PT%3Heuv!gkCgc(w^#MY0a{=1DCVQm7TvXTa$T*JEReA;Xvlmo z?G4E|8B`VRH=|3m>x$p+C5GZ}n}sVd$0Yqor*n#l#0xR?HoK+EW7bgZu)MYKC{SR^ zB{@7-8sln6|BK-*hOxB0Fw1G9@%t{$sQ-C+Uk&YOxZ$F^zfFVO^e&*Bd&8gY7ODDk z7Txw0GbkmoKL$j+Y_8IJrq$`lF@~s7&v&Pk2TzSvFFv*vacR&mi+vWxSnr{`ljn@) z;eg-r~;$?YY;aT2J zC~d0#&L?DX94b|+jH+x5=c`ETvQ4grl@qXk9RjJVTtv{Y7e^w<-O%6gap zbn4whf<7xBTV+lowoB~EXb*;>56Hwr$WOGQMaZhP#l?7Hr>?g|l#=b+0R^67Ht)a% zI!G1&+O5f`u)h641?7Y-)rCiXNtJ$yJvHP#mYpIIKCawWi-P6G`-d+sc&1;k)>=~* z)01Idh$lJ@O0ulR`V-WWoke+=(!+Cqp6*iWWmTt_Uya91k>;2;A!57z6OrNot7Vf= zucVI<&gFl+GB0=!VLpAFF0LvO;=g@8ld(sjiA!R(oKW(p0m;JuD&XPS0!3T05Gv1M zr`4dtJ|bv0W%VX}*5o*CF$CF-@s>Dy`;cre?q0XGsfn7-F;6(M!@ZLg`h~Uj`HA_- zddPjNsN{!}A1W2iXBfk2;LXH)8gzyB>XLt_;ccQs<0|Q@qObWHmf*QVKKjHKfKA~| zD{;G$&JVlSmWsl1+*WkXedk2vhn};!6pd88cCW*eLrSX;m0gD~Eegaq4g6HqYDEUy z`7ijccKLc@l4vZhF1M~S`(^rTFemShMUZK%r^xXhP}h#!6a zZAfh*X31x}6{Dy*Lq_e+hP|@&@|-v2$=ms-F_^@Ri;_u)SL)8pl0SJW+QF1r@gkXh zRyI6rElWwDXU1amYl|Hk`;I6~2eRGXbK5y!FYTbztA(rR6F`a-IW_i0j`y;z#EWnZ z!F1^!aV^|J-9y>sFwsw}2`+i;6hlk|uoV<{dZOTFyjfV}UQ)J{7E*n&w zKkt(3hh8vAT=@~)4Ldb2gmgb>|2^7}=S04C{`B1CP)?Z?A`I_7eM%%1`~=H!B^2;U*5;6OYxUdA+As5RHAy0)i(%a(ucX+?G=PW7F1l4vx%2s zJKt@#o;h7NkU&Yvdi)@t^E?iPhQb>KcMJ~6N3YJudJpLE93UhfIDG~D;&|q^m81uE z)sh(WqOp_q1zuF%>oF9%2hmAL#BY^Hb&A6?QJpW~WZuzZI;l24V!lK4`1xVC55|6j z$pHxvJW}f8(ZGvQNKl0RgTy{BEvmYe3q1wsF?te9iwZv5`MqjI*SbjOToGfN5zO{@87@X^5I4tkZ^+ZBNep^&wm)?%;5 zVnFhiI{7$xx|6nC9&LMziBM$h(DpOc=q62|!|g+^Htq7}5`X!sD0(0QZAnWhnV7k+ z@H~e`{q@)(#HeCgL5oQ`7VE_!&KWDuD+iN9`ksVqJr}rP_o*7u7;m#Mrh~Y!m}^U} zA*bSKuv78G5AwYt^P16%nAKMSC?I6$qg!9V>S>zt6Jw+YcZ5rG;)M)tvSY7nyYo|s zn+82=trBg@avZMWE%pr2d%g>gAzbD399--2iD^$Rt?0|6bcvv`LR6{PIC4G#NsRgN^j(xcX+mY&+hWN4jmH)S`CRXROxQB{SJ56_O&NRXSN=qI9XXVrKw z04&6LSkN{p_C2u5s_fd;C&X#5)hpuTlBH;-|9w)>(2FPP`CL}TCGkLM0z9!S;dCHvbCXlVm3B%m)sIo`=Nt)Pj~DND9|>A z&9>N%K`n-Chwt-p=0=cpc+HewlAJGYGsiAM+DF;&NY`^vC|ZZ^O!flc2BhhNnXqJ9 z4@#U_ZsyH=a~*%|9g9JxMlK)xKHQ4*JgV9p#NefpJX<=@%5%<9aFm^FwSGMIQvIV|5;%wB6p);M*&!P2F6WF>uzOQ# zgbWK(G!N!<=SuFSAVr|R+@i-!164^OE3aM^xLhzzQ}goB`Nu;HLhYye2_{K&^ulQ$ zgj3|RNmJcNActoIhzm-UZ;|oakVYR5uIOjg-zem(QhkrF7RVYA=`Uwah<0^q@};vL zZzd*O96y4DlEmBsFax4kCdUijuTaf|J$P1d;5m}CynXsc6!KUDX`PXhbo`+2^Ni|u zGEDN1tZQna6+Bt?76quq?9Ek?*|V{Jw_bS+DcHF!2n{G!ZREB6PLdVkVaBxbay{-~ zUbNTiuvW?U12ZLj&&`vY7r5Y?Yu6?JXfpF;3S;kiF2W;kzL_M9<4A_1Y_Qe@_54UV zQ8C$B5|)ya9?^ksrp07cfP-(M%yWv#$Yp09&wg|+-Op3fS!;s|lh^>2TIw=%Ri*fd z@YxtB%FswsNF`U&jD$OOJ%MyX`SU3$iGCSL1XYI6R zj*GIV7!l=h79T_}36mVMp@k%IbYJz-d04IW_+e)*>RL2aTtJJJ`xNwFpG0XG}OPWlVF#Ee5`a987UOp7B5>p_5b#SHDJ&)m&U~qg!Ai0=;_q` zWGe$;0o--z5F3BISqgBrD{8TeoYY0evhXNMj=JKE*K~fMd%xti<0yo@R#?u-{qAz{ zvyc~pKN4QVuT`p_w$Q9imb-&iuuzF0?814eDPXvB5w-5-5AuU<<{nYjX3hDX3C0r98$R}_QS(0wvv9Uw%WQ}XG)!o1le}Awz~-v+ZPW? zX`lOxIJ}B@QIH*-**`nG7Nr$jbY3NeJGe9~2h*>u7g@p~g|%D${TLIlp9785iO$|v zN^zSo(BwerVj2Em7G4KO@KdPh;hsZFpqI2HLzN8SFj-8BB#4>#^mQ|D~9UvSE zp+7kKOC8{>TAMGJixZlXIlR1`VaxDq`SJj@@jNv31A$eDDf1|gU-j~OFT*|~MDY$2{x&;j`D(8fqScO`!MGFYHv*JDO9D3jF;?M%qlb*Tb+U|0S z5=AS#bF@(q7l=-`m3UwIyg7A?w3J2R%fuhqWKc`5zX1&<*ig*KuM2$C^Wrmk%f-;- z!dBqb*^=L2c|8H6y+5(5U@x=PFFd`bFpH-xVXK7c!oVO$1h1_!0)geLW?v{$VxF6n zrAVDg5a>)~QofP>T?>+cl9`j%Se7j^v!a0|TS_mqcKRrVb!JCc`{`17y*?Jx`m+1> ze=im-eFaO?5ORl@u(h3<`| z$%WQg4qKa)g`H~E7v`~0OVD=SU_fK$J1mev9GUp^%ypb2bZsT;FCA9gYX@ufZxs+GXVq6neEg4Tn-f!g?Ju5f!bn>1-)yPRz4&LvF@ z7P`TTI1|Ouh1=380DPUz%2cB9cr+`(B3tGMSA;yT%}jOe%r}8(az*! zdv3n=d{+qwBKZq`ypw|8qO4>}$%@!}nT%o##yo(O=!k*85!X?Z)R*fZ)}AlF4Ie4U zm~V5j%84HQ>=HA&8<;d85oE6$TUN@{x%0Rwv2`9ZkaBLFS+4EUVa;KWxv=l1~Vp|$YFBV z_6iFVTB~pWFrQ6$;A-*21)`FJU8d~Z>AU?P@rq$Ip`SA2D^dj3?Pjv#xjA{Ct`;DSu|%cHo%SILq`d9;q+tUNXv@F?7sI_4 zxjS(nw2^8ML0-;OI$BWYO-5{_IvMmi+cvnQlxrO}m`_S=YwCO#r{yubdw>VSSz@Uc z;rW8AN+vtlzxOnydzGWSg`>IkWyYOQP6Fbj2en^vV>`b;ulcIGac9|1TP|6QJ81^D zu2CFAxLQ+#Bu`wMY54B>0_NbVG@W!~|JrLqjAd_ZlVu&hLtycsu{Gcf0tI=p{w?2k z$Pn1JvdXyee)vQ=u8X-vnv3t5IRaY_hQ(lHGvs{yg?}Hp zFJZw0`SoSl42Y(!R?1r$B*OC~^fj;}woX&?w=4ohp}&f)!PcIVf~|dI-caJf58Ym- z51qtW_{mq+X?!9%Fr^B+SDg4Fi+kXh!uB`b{Fe02jyMRNClTq8G5K|6_}uPtpGOCw zFqlxp-kfTH zGBWB#baqr%#vAo4OKvG4t+k&0U0s)sNP-Buh9hANfc00GUMdgHdi6CUEN<0ouH9V(`Zc8}x(l*)SQALQ`HM{}As@wGTv`|jSJzD|rF z*m8Wq1QP2I>AsqvM93IczxNV!-c` zKX}7`CPKc}X#%=i6N%)bKN1}D{c)bubZm;a< zR>7qPYh~XftMnV2Kcc7+mcr4#txCfdQWxuo`m3Jv%}^%6<=_toV-*$%x45tFhyy+v zK^pl;AQlov^IBK0_YTyIm+XY2<%fO?tR|6L5ANsn>#OMK$)tE2E83>C1!YgRg11~C zhF;jOAy6_@BemO1)2?5&C3ngs53j;SdT1%DSsw(fStduqvo%OMQo^j*TJGwrDz$|a z8ED|FoF+cyBpXead<7b!$bb9TOx3$B_d@tX+~ug!cWcHp1aI=&>#pN>U}*Q1kY_o z1%V}67YNinUIs{&w=zqFi$cq<%mPV9ukC9S_@94V4?Mr7f}*|VNa^cVyxmF|e%`i- z6x)w>v!}1AM@M=toL&Zr6#xpVjvhkA`VH&M=KmWL1sE3%Q6x zV4dLxm|lzX9YLVm=X=P&mvGb350DN5K|Zk!loH>cwo6PKD_e94q3<(P2T(qb-+w9+ zczM3G^j!@r$i&Ml;;0qA%SNW}g zu9|`drnBsX-@pz;FvwW(`}5Loi`IFGulH<}moQ+pM`jYJpN1S$$}1dqFn8 zl?P2ed@O+%{Zu1$e^8`<`&URVN@A43`Z~Fqz3-`19|qScBkJAn zTc8VY5;o<@O)-1)`gOx7@8hy0DxG%|XZhP~GUVYfZ8M*BiE@W{t*6}LE8nV9mibJs zQd3%Hz|hq~zQB_{XxUfD{Lti1rqjOcXVNH61^si-d6)*2!(C-p^Z-RsuuAfCiQE^1 z`@STWcKSu51P?m=^LYrMel6AO`66w4^uI0_bG z>1MUO!d2zM7ZA0Lc!BgKn69z=^CdTdsuImYory+0P}`6GjnMmP3(jQV0=<0yZcA3~G;AW@dPpq{WFuo=w`tKI+()tnE%LyMk&{F@@&S&#_1hXbE{e!X{J#ay!Nn zU)_oG;NRscc8)ot)J3p#A-^_n11;{F-(fy03K)8A0im5A)(h;9Ui;$W1@BWG%d&s> zn&p}H%G$SK&)h>R^(A)AV5N){$LW-q$LrdTaIa>E8)pQnu%q@xY^D*fput*G&|K!5 z<)x)62L}f|i&H3CiTb=q^~?f{!KPpy^c_K!bX|^2IMK_n#S@Il)t7Mk01?U&9$LEi zQ2~N>O;w4!v#mBMN{ukOJGLGop6g?lqW*EWslh{B#I&^y6sxY<#8JsawFE)hqtbKD zkwMy;dAt1C*lUc5fKA-Lu6AI0JHGI&>KrWuM5Un;5$rpA?>J36G zTDDtPs#Lx@H2NYwr4r4roxBGQP@he@M$fESCvX1+pv+y{6=eVho5uw4wJM-(K+)J@ z@dcb+QhF_q#XOW@k$;cE^nV z<7)5@wQgkF!=HH}u+t4qU%&PbZ9{lw)Mcb>xb9Oa*lUGR=ijmS&CybWBt*r2VTCT| zJSP-7cf8*1e_u`cRYrE>G`>WJSIO0z0k*?{jNPJT2+<}=2&3VSg5?NpzU0?)IVREg z<6Ts)Ye!TzeVkjpk~#L3p#{HEJkfqcNnr@ac;5_0@VX50Na0Eog=UQg2GF4Y3u14gz+Ba@B);sM-yHd z7nS*M*q?$wluZ=hpAt{@ZkC0cw}A&OeKRJB%MJyrMH(|p81^BXbWH0sJenZuIh zvLM|DwTCIFEhe_6l#~OReTJWpI-h*i>l9KG%iqMz6`3vroL0-sA^(`9F`V|s`;UMLoY6y7RwE2aGc(a8&~^KS!u0_F=~e4Mi0;vceBX24wO)* z5p&VY+HeM$n~1-{ftP;MzFRFl`yj@WxpwfV4u-!6G4I;JTx40V8J~j9Moek$t8drH zR|`R6eEi=riL31jldR^EhC;l zSls5)KVzbe&f%0qACXoE;0TeJ^YDAbd}CLfsQ5I`$ul=r-C~Yh=9Zr7NXP? zJ*lCY`(#+uiAp5~-+R)r+Bu8fC?tm|u|RMigj4!Q3;(lN*FPp%&hvlZ@)A%spYmR_ z!0nClaoXc5aN1q1p^LNJgfo2N#w>)XbeEFyRKEE_`>`VQdvy+YWr5_I zZ&nUCU(kW{F7=$Vi1GJN#Dju)!%PA}anY%0f`n+eNEYx@2cotTQ59m%`(()3utNG4(0osYF`S4@tq&mwPoxZkI@#3Nw8FYTv=JM z^zZBKg=Z4?_4OgzhLJPFSx@hXIU>TM#CX9zac zfaRu%U9u?gmHw0*axNVkR4#NP@ajTaT|LfcWS7JvzAKUfa~;pMiV;2i%Me8;@pR9D zh(s2I2X_0!;gg~B=J|S@l-z{lQpz}rE@)i2YYu(Y6u`z~jEw@LuS);HVY z6Kx*>W{cYmeU_toY^Q}%aNC}9>DO1`U%-mer1#%f_g~Y@LKR3|*aH4_1CK%NU+UT8 zu+sX4yR_hho`UhkIXTf!Irr9U)zOf_9*nTVSnRok!`LJ3SFhNjN{x`^j@iZc+CCd2 zdHA_mlPEDSS4BUD75)xl50UgbAYSJyl`Be9Q)PUhB2TZat=+Plrms4`cdu$u%;R)_ zbpj0XGdY>=F0OM?RwuHzvxCE=4gyCn996S3cYsQj%|Pno)TbVH0vgP-qqWozd#p|d zzy%o2mnk|QTqs8skXt<}pF32bz!FkR7s*l-Ym75G!#2>yn$Dd`w3qPX;b!#_lnfbD zM*FR>srh&zQok0A{^5VvR+g6$s<(xkq^MubPDFBEmDP3Duw7dRr9XcA=6HVo(ZiQ) zZYgZ1v20m>0ujV+N)Fq*eSBL7LpFPy&_@(whW<6?m15A~xpxZ7p}j&q)Sm=iF0+)x zib6T+P&cqA(;U-UN_YU`jT^1TTAzpdv9!n+sLqOVze9FYWsrR85G@dg?cA{JzU|); zldDX$;2DF!D&Ycg@v9a%M1eSX022?vaf0z`aq4v6&>CmA^)8GemJlHlHGJmQRAzO0 z>-twg;zJAIyP5K@py1aYIZ9>%VU-TXmWp?gNMW56;WaZrjBE`X*dHOFTLW~o)5qQM=1Ua{oJ5(boSHNKT@Na4#i|j z`$887Kg!^5{Y*^C)2{1Eb}}2IAScU>>sIO@A__ce@Bc#M2A|8U^DOx;8SB)d-FSHl zTKIX>{cGP0$)Seixzg?Lm+}az?lXd~oIWuPk-jC0b+brWlTj%|%$CG8uhF_ZY8Z4v zU=7c-a|e?2e>QYUOT?oOJc-(SWEC87gVO4A8_v`)Qr0v$lAbldkI8gpcSZudSt-EI zjSdvyro&XW?uA36d_ZvBGJ#C(L*pV%*`MJeDEtGXSZ=5H=wo z?*U>@|FX~vZX;94Fd9Ej_|V8&FqhwA?QpzDPnnl(!~o@{;FqB;Vn}p`bBK>7YIS~Z z^=N1EzSJ*wiC!<&z!mUD`2R)jZ;U<@ngZR7)@fS0bn&CUyf zCG9$gb1UpEq`t$S)1H^LL0&HTlU!LL?J2!}An{+sw*XXn&fO0wJ43?0#H! z*NR8?ZYlXa39j{{3l7BMNM9Q9|DR=I2x1SbW`I7zQ}pPZ5B-|25ORcHhY%3dH@RMy zUTAPj$2CdUQy3ciP}|Yb(ai|@ozN9`E%Wzeq5UKE7(pfV7xgQUur8j_`{Q_duUmrL z6xKhw{k*8AdHMoXS5c<0ZmD=g?Ya%Lg$~bV6|UDO2v|2CNaLbCYP**t%_{)=u*2fx zlEzbIPWnF=nK$%^;3THI;Rvjw=0DkEVP%CNRX=J&8o}rEtf5!?swn<$cEb=+sEa{5 zMC9!f6Z9U|n?fxzf~CZ^lxME4Z`r)YO4sX_=lkV<1!KjSb%gm)qeV%v8XnDm_5q3w zRKiY%q_G9f^$trEgAM7_xNfETHJs49%qh{JRBZBSNf9Uq>$ddXz198niI8t3l$GgTDsS(y|t$!nUyKM;Jw z{k7MQ2O7iJMTdDJLkQDxg(ABzLi?Xw0%c#%&_ajb?P5|vmB4rAM--}$PpNK->0&nt zg~|i$xnYADCJ#?Go$=Wx@%<5BQlS z>}>4jIw&`k%x__w!q!p~#3yD0=%n;}MV&8G1mpx?_bc!wiV_*I?#drF1V5-bx9O9$ z*9VR(|I{L7Zf|Ei{dXEW zV&TR0?gQU;NCcdg^Vl?`{XY863(EO!GB|b>*`0yL&mYwe+1bwPUo5^@%jXpwNfU97 zn^G3dCD3?aM0D0FB>D4FQC@FhYhBkTaOr=JoI8i%q)=(xT2d#LG?gP_=@q}qml-9h zfMl&FoP%}-;HXQK6>%K09v`(Sy)t4MNJ|1;L%)8!GscqIC0YFx@_k;7mL6k&y;6V3 zpurpzh^d^(OLj#=WbN8b0UBmkQ8w)o5Jwx1Y8LA z?ZJ1T&N=B^^>Tr{6c-Y6bp_W80$?nUKU|;^s5CUhRP$J +Z9PsV~cSmoY2K*+CG zt=tP34os#L@U^Xh`3cET?) zvo5o6$20)X3*R$8#+c-l1tfGf^oKOPRg&lB>!|{R?DG0}5v(fx;^R2izUK>ImXuh7 ztgCqr$DeobWImB0Y`Jco;Z_T-8;#ie*x6&(FgkhYeHuwWZ@1E)o6ujoi2u35Y%Zez zchi4r!%d%tFq5Ap+HU zrf7Q>w`s_lzxT-H+BvXgKsIV94&ENuJEjT8{rlh>4^Q4k+2;=MDb(!c=GSN#7u=4I&EL@T#l)g&Eh^Fvmy%}+*6EB!Eq_5z+nfkS!q!H}FB%_`qpVd$!0{lk+;|wgr7;L@=;!P%%Y?QYdcu6};(T z;TMfXOC3wEz9Y+ZoxjKK8ojOpAl!f7+>ttnUB3wB%g_@<%fr8|C7X3+MfCi?pA6J4 zsB)Tqd8fr0q3EzyAIBPFDTRgEDtsJzAhOYYDjbIQ+IxOsK?G}O##bqvhKbh4k|sEL zwY&qsT#Mb^*`X;?v>(PAk=FSCS@hb&#Qu97O6ID;)ml8%9E1}}U(#vp786|mchL!{ z>;>WY%+1w*x_(UTGbn5Y<;mnNT$|Z#{GMO}W^td&>QQ6V{wjDdtml&;^aF?JdNoHe z>mTugA=oJMnmpF?X&SpYIgM49B0IR2|dYC3hlJ|6cRjE&+AOU=}d$_ajZtpA(C3za1$x`dV9C z+yAxBu?Nm_txvO`Lw})Qa{Qg*&A3(4>UFhae;gU0bFJ|EW-&^bH6pmAWWMRoKi{jr zSnTO@|7yp`-;JVEBYV$iUe~*6eC!igh)fw*0Wr?M7SdwuYrzV20SCVK%E-uos)j&tQ6Y0O#1iq`UVpx5_qhEv0y-S<+fZf8g%d*RCy=to|1l`u z7jHP|xD#RshHZ+tmB}p@F5q_EuLm5i@85qP#JH?P+Rl(Xe_6x+rC+f_u8>V59Oyr3 zgMTijXM{eYJ@v}+KO6#8$Fl2Im&WtmK7{dAec*aU_m`X;X6U_hP(v&N%xmW!OrxE% zvudIoaWDa&xciE70syvx@p%4W#;_t*)y#APqki#GlR$sCn$5qytX4*7jLz+o z(3-66(hjP;Be%E6_2>L|)T0n!0}g0#{r~5Cc9nl8Fd1rAx6`M05{7Q@XsAOS^7KXZ zIH^fqzq!}j-+!9)`7@KYj*i{;cX&_N%RAnHD(xDNGng>ey#2gok#_6XH9kb9Dy}|U zp!y0#>?pChoEXW}eUUM88JxC#lu{$mY8A2`Oo!OIto<%}x=}q0w*qj|77D0CuPph8 z5m0_kG6~QF$+#B^!6-p-4N8NTBSpkpK!aOr$!k9gouuTKG|avt$4@LFi}3vc+&5IB|^B<#Dc;G>i}2B&1kLJO=sK*4;w9qEU(< z8t6a8&$=q+%NB1`1|_6kbZ17_;Y9BfF7{{%TsQls{^g7A3bPH9qX%C-9JY)0t)^$9^U-tw8QuChzhB|6e#VVB}ucK7kXw1Wq| zMzBG5(Q1i-<)PEybvLw|jc+P4QY?r0+@4U=i8_A#t0A4E>gWEvWoK2_i2r!WA^`K- z$LiN90GD^+GC9sfo}fs+2!hi*lvzp&aEqrgo7kpMR>-RJ!P*}vaqzkBGjn>0JNXhs zyW%}T{W8e6riJD(7R#BE`5hAG7Z4dDuQzWXcwu702!=Ao?DB>Ny{M?DF97+36Q&%e zO}=TVCze#Qx5scD`D8{cd+)|aa`0%zR|%0MwNd7a@l^4;XTo4^qWiyJ=kUD4~4*Lp(1XIumLv9C_=^=?{ZyqK?k-Ye@=^tMY#QnXk|(dh<@`9M^U`fZy!bX$?RTSAknp zEfobSW?PoJ1M9{G+*d9<{9c=eJo?oNIWMA6K{#}U?IAeUnKgz!fIf8qbW#mtBu@(^ z4~kS*ueD?A-07tqSj8e@VWU8@tyJn_Rkaa#1CQV8@~fcY;wKRQNCMBTfBWLJlS~yz ztDdS?s!9P6{4c1r!B7TsBeBZIC7{A7Jmn(emY#8~VD<(DWn;m6FIMTfF|qV;**}7r z|2O~Wr7bb8ANJrfgm?z{FW6F?Jlrf=>ZC_?t-OOs=@!cODX!yfiij`q>aZ~gNn*tA z-z?ftCjnpCotroJ;voYHz;=!=5`iEN!`iW_$g^?)@X@A((lM)$H+aV zHs40^V8?rFZ#We%O(rUgmp8tCH!0CD3!S+8<+q_4?d2C-WSB)g!zFgL^%*0ktx!s{ zfUPWSjQzlM?QdRQw2uLjDM0382}nVrK-+=UK$!W51dUyJp2h(|k(>EzIt=pvC6wh?*Ih*E z5Zu=7@Ay42U$2W#ynw*D6Yv5)2iP$91L>lg^dM;6uPO6)Ogj*Qu{5uJ`3heq zJ(t-1$&A2i2QwsE43^(w0^nD4-g7r4_J7If*Vt*38qQ3dq|pQLn7m&0g-xV|&AkD~58OsO~myMXIesj_)U)^R5r zc+93jXz&Gbc-DXOOpp|ofQS}gdeRuXQ8-peu6#RdQHi7<-k%@cz4A~NmDiNG|Fb+> z%yBU=ik6G%fU^xwE-oe5JabBn2z5)@TYKJUN*HzX;e`FcdKHot8oUpwFEPHUcL*3C z84!lu3_rkRTZ?k`%NYdSy-o?Y*McF;8qt|PD=E1YM1#|iyZMFqYFEG-BW}hH3mhT*#8e*?;TI||HcnD$Q~JGkL=l_gwnDZd zM>a=hC^`p+j4~@Dj*;wRM(9|DWJ^N#>yXd)ci)fucmD9{51+U9dClv(p5vN3%I^(g zUP#`OT1XBOcN$z6*roe~UwK^$gWqmW(qh-{W~i@_$%`4Xp(~JMaYrIt?LLgluB@wq zE?x#EE9WRYb6+BMb#dtHGMAHX-TvL;Dwu`zuF1_(ZR+r+khuSeYePJUdVXSXVK9hh zrdj7{l|_~F60q{OL2&v0UdVXQr-qN4e}sckv6?$6_toN)>*E^g>OBMpgzkqs85s%f zciqZ?7{~MfN9kr)rOgq`uJyYQnb8D5Kr=kvcS+PblY2uVUYsP9g&EeE!%i#j%m%bY zfXa?+#P?Rv3!?cldQ}%Bs?g*Wsr10jF20o>ss10P3SSkIoVy zQjA6q`iwcCa_kG`27h#O#E?@8vBnrxXT?nIX}nQLBSP_fV3B*G)}VV-(!K)Jyzk z%~^9r5Z-2sj40r~F!}l2O~PI`6Ju4|c0Vz6WtnMi6wh*T=Br&?XK7<>-YOZUVRgf= zR+~SG9HZBh!}%G#qMxfMgD!fnY^k6BW#f+sErp)ypU?fg8-ezbZNxyK{=Vn`zEDg4 zKqFvj10tWc{?DX{N3v{+J4#g;GH~RMi2Kgc<$l* z`(iOOM5=o&V2Jl8M%TGk@o2RwFskkfQw7YqR5nt^Px2WH5BoPqD#B!zM%L@+5{Di% zZLzX~vp4H*Z_tG-v>oO3fv9Jp(X)*JN$=d&Z;K(c&u zyMvxJKQ9wQmzHYx#kq~X8mxv5RnLU|;prMmN5O868<3$y>`H8T{i-1-btEV=UF_7N zreWY!e4;^~&+7lc$PntH%M^T=)nGa_dHrst7$uQ^l)SI$`T~iRBO; zurD~f`MDqtg}&j43Avpzt|{b2LmZgLezY{0Wn)$L>gFBiVXYgEb?`a!&xa2oGzs*OF=Tv;5yUb7eo!_Z<2zqQW9p%xPiHLaX^LDG0s z;WqFbR}P^(z>`)cJ>QEkUfp}bl?VZ@F>~=!(jj~6JvkMcYmc}z;vsO(a=?FK<7)A4 zkENf~q2pFd<|N8W8rzxxT!)~KCxKg41N}03wbyV9p2;#koeBHE$|hFw^iwCq=_RW5 zjr{v83(C`EMws0(NKY%70Mu7XthhbGIC+Uj`_h<9JLX&T0P!p9^M{mOw#+L6>qmdq z#v-Y%7CUvPqYDbIcd|an=~Mk1!s;!V$T01Nk(y5?uBM9) ztc~+66E$wBZP$XRxS`${I0c@UdJMqx&r1Xeb2Fx}krwG>6?CQY@%=hZMDcI1N!^ti|J|@zc$c@> zA01!-C<(XNoX2~D0NMMbRKj;fNELidHbGr1Uqfa;~ve5hpeouSv*iDCnr4k z>IZJ~;)Vugi)pST<$}*ucioN-ep}Xi=i7vA+{AM|g1$Z?8vgPnhnsGb332h{g@ls^ z?>w9T&_VxNQ|&2viXbl#OW8M!_6T=^Yvp)H$$q78YisV?1*I#@Cd1=p7m}LRb$OBq zl=_^pHT|NW$spKXYzTs-&q+1}avl>;$*Ds6RepURYd)$8TaRS_ZZ)%?-n5b6#XV%! zyg%Z&-!2xG4;WaWbI3WP?Q;Pjjr{(qst>!-$^M-CsLcAUA)wgVZi8#r0Ix>%1Ejw54f((KXZO=$U6qiibTX=+60VVJ>(ssg1q$876jJTEAI{DS z^R{8fi=*elJ4XiG6?7vHSjdfleb&|)8{Nh!{7F+Ag+UeEJ6_Wdjln!~U+|ZFW%MEW z1D@7W{%%fCSmf3*Gb}Us8J?K!y3?-!D0~>Ny|)0%9RFF!1^zkr-9O}V6U%Dn zvP$T6^NVp4)4UWB?gP~>;RgrZNyoqU`P$fLwj&v*a|Uj+;usIBR$4;C&*ErTdmS3# z@+VuM@=SrJBj_P^^i&Uh8&^HNv=O^~N^8owz?&ZW#&*{iT-baVYN?!caP&(9*gx%Q zLs)$gfcAJ*NF`MpN#D2^g8q~uZ+RXiTLn?)NIJy|3JMw(&ojF2Etgh=^=^Z#!x$g;Z-Z86df)Pb%fZ0m zDF;sJ+m^JvIuIHP)dL*W3jPTQCoCB!=8~=dXQv>*VJ&a`H9gt&=-*vXp>j}~l<;UyUdN9A8&8P8))5cqjaevX}Fy`hO35sA%%SUZ$kre(oowQFLXaD5X?S za@Ctvow;oZFo z$f}@;rkVxIn*zsVV+AOaw8}2;zk1*CU_=tMj4Gm0yk8_eV!#euJ^SCEelek>J^ZT0 zRg|$}SM3i}o$~1g-Ly}dTI~y2Z%@t|{7>O;6U?X%dc#$1EW8M*+Yj+0LoM#$-WK)% z+y{@#|9cl|40&DYjRZpV5o1N&RZLKXZ93hLiS;A8CqnL3bh)Qlm7213pRHb z3%Mx8z5pv-H)J}6p;c%|d1bXW*Uug|@Sx;y#*^`9enLvRmw_aejQV>#W5+)%%KiMc zSUn7_B8M%RQ_n8ltn{lIej{+CTUlFwU67Tj(9ZByeNtljUiPSXxTc$qsI=6|f5Wzt z>Z0;aN(2Hza0-d=aE z)FXC_-b9Nh5?nbb$Lfqee^_@!kDC`+MWY!In>Tpg4@m&A%jvFuC05Y>y|0?Zsp%aL z7S4y7cx$*-I8G`9mG+JOSN1+0P{8{(o#AObaDYvxVI!VHg#L1lG;qj7eS;8i3skz_ z$kfp0QAf2R_9jMyV22Jk8P(LqFje6>O=h;(RMd=q=orOr4(-cI}Y#*<*v)~UW1lrwy{2+ z{ivds2F0WPFta5!64c6c0OxWa!<(6)k#~To&kTkBi*lm3_=bU6>F9SNTWhFF?(DD{_TI<&B`gc`Q^q~e|2L}mu~kq{ zdPvS{Eg{Q$Hl1=-4GoK|{v9Q6s54i>Fv)YPvRGNUPH$q{U5C;!b9N$1;-{4*V}_bn zr1ZYtxel}=!(QfDae8vi*RvIdy|2XCRfyrg1AD)QqDZ__G}zc_DWS4D*~tGL zJQiRm0kzNJ%}nxn)F&T_t|W7h;Hqk;Cg9kQH}YqGp3VU3!^)lO7_P*W!X`7OzlH1j zU55Fo=X;MejQ>{Eoju|RNb2;(=t7Q`3U_mNKdp*Kz!Ut^p_b)t(RlEk-kgB&&$@BMa}!>@YAyr|G;~(DYg?<_L6sgL-6IQNr@o>$WM0nj1GXon-z{^RRPt2k z(*LclbR|~9V24bm)dQ+1{zL3p;BFnva@Kjo4QpU7e3aQ(rEb{dZaD2)^m;z<;$ z8rj>jun;$=SNkx!u`TLK;KX~+#bTG zA+D~jAp3Zw_bzJcZ2DU}%moqi%4=25y;|uzpnVvh(Yqczx%F^=$JN7QbyV{BLKE)4 z2mvs8g1jQ7l-f1ez|(O=HQJ`+_QgUIL+$&Sb&wLKi^1$Hf!6N%pTy|_&;UD(v>Lm4 zyjG5X+>6Z=X*?s@;5{-77DpE~x<*zne)GCZEpT*w?1pjZT|w@naQ;S%YEtT&@}|qn zNL}23d9%$#rP@S;0*hd87NdIxo-|Jq0U~~VnmwRDoxxY(@|yPA#g_5q@uNA_BlYK? z`B-Y_-TpjQ(#b-Db$M;CnMDJd=OLTW+rg_f`x9eLg}_S6n*GmsCt?XN|HV6cErDoP zkLE(1`)6#K)5%uhVaT@Mj=@&{>g0S3v(w^;$!4Mb6-NZ+(og1X5}~|xOIBJrQ%z`; zK-UCZ{HGt`yQH!95qB22Vc&5+S47_*Nx+8gzs^v*UIi=Q_{EqVuo{YOwbfc-OSY7a zCh8^UCKk2g?adYH%@gUh^F~GCk}aNPwfI|rxs;WW;ZfNKfEPEEAOv)XwOP_lZfjBc zSXVwZPO!cIJwFGkVhy`IDW*8f`rYE6IGnL&#p`DGP;I{`Jlbj~SjuSmaq8xsbFs6>3=0 zP(CVZ<@D)S!8ZPC3VIg$x9JZu!v+E5L(7byd7!xx573S<{>+$Hg%g*Tlwk2`8R#gw z1Zu^j$(-tlK*+2RKP22^QWnn8^n3A{--wAB5FD`?Iys>dRgVI0gEqGBKsVlPV3ViC z4m!#C;fecrZ3JHExuEzD3w~GDs9=0G?`irp&=_zZOwq9yS*DnZOfZ2+@Uf*&G3Qt# zOQar$#Q;g<|GvFSQht=#jEKk{tiF+pl1~Sc&LE;rt>1Vf`zQ&(y<)3v`#W~K*Go(F$mie>oEbTDWD3%4WNCTqv5+kH zK?kz6qoXI{mDQ#oG?}{*i!}sbr7)CQc`?&r8yJOIcm)kT1I&zIbY-Oho`c_ZSjIzE zTUQ6lcVgD{2B2Q{1dk+mfZA)|A>-Yge7u%n{{zw07gQkGe+OvUJHRsAgfGWCf>$r6 z7=b|a7@n-s+Q=KtAdKo(hOM5*ezHPlTNr8vG;j|aX?3?63(`mbVHGi4{gQbD(mgbh z0XC+cYp*c*u!sK_27sHNe*C$I_@9yUG5N@0g}x3+;JkWOBs3yGk?Z)yO=ZDtn~18Q*9IP9c*C_an$vP{!8grpct6q z+#9QML2_Q4R-LjxM_|g&Ea%(L_oz+i(nUSKAme!D zmiJoYFTFg6037SiwwbN%b8-=#h&uqZf76hvf7Rl$wYM5p2xk|~S_s4+ERsGAhXyPEb*23XL_s%1isi zm2B%(vgwR~$WdDL(JdZzBf6*XG5P=F>=x#G(&=9f^++((PB2!|w+u4+q2@gm`C9JU z^|!uSGTV1x%YTr-hwZ|en8})m?mO{ae)%6y#Pxl0>QJ&Bz7`!=X4JbV9N+>{t~tq` zk&)9bHHG#oa~*CZVcCzUQwak@G)K!5R7h;XLv=1$S6-WV!^0vussHN@#~OlwPe1P! zCiDfD}~?^ zymrao@P!6z7QM}r!LU*feLR0P;u^V_+BKJ2*^!^(FI|#(>J_fm}>BY&wF_D-?YDs&Yp^3`O-TBp?3s(^P z-#OP`cOrw=(wzI#Wxe{8-=62DDL3bwx=i_=i0w%X{9u*4g|&OqHug<4jWfd?_~nj7 z1i`cH?CLu$=kORIuC|p|>RDnCc-Ji^>4yn({q&Pzm3TSSno@w}FC0PDwiizzT10QFyX<_O# z+j%IEhw(NWW$&4ygMw&+4Yl0@dJ*dre^5?x&=rg%gictHqT9RRe;LqcPncJ%`ngo< z`ar$ym1|ZwRR%-)D06zA#9f%}_r`43} z`E`7+EzE?u-?41IN-3ce!J$I>Gj`XzSX@{FL)W|>kxK9$(@aAM_Jb0Ov-}7aTQE{; z=C<%D4QkJ+LX98}Sgjdb#uo>HLM3*7x1_Wj3p97kQvhD?+IlAI@ku!U+qTA)0x+FB zB1cDZ%>1pdbCFJxL^~EqhU6sD!ceC1HY;6t24`no zWMum0R2-Mk577?0988BaapS~LzTdz5-+5bW^`uU=JbBqBy|#zyms#hYTl)?4y&F2D zI!po0R!n{c#)i3pcImU9#=AMT;;&kOHV6$@8){uG>si_J*jdso5Lutyzgn)n^*%`fYmSLv}?aYG3$@w5aEzu+<5U;%g})-;CcKSpIs#`D+`@H*m^k)Z{MC#%LBe>vCC@ks8&s)WR@31?q@vTXzk4fV0 z+1>7b)cu%Kgy=R<(fNs-7o|Pa9k;EfbaMGR+G{sK7tJbK0(|VQB!tIF?@Gfy9^qP5 zX*ulei*a^D(ck(JybTJ<&H4el0S%2dgd)r)1D4F`S<1$2X+NiQ^3@DiEoQua% zr_s(;VnB!EYWepTmmU!F-`||5~ zD&-ddci@SMP@yN+3%=p_TW4yOb#OPwHyjmI;%@#ITccBNo!L)cD+5!2^wS_5u|3a@a!Fl$B>>nV-2V@m5iZ!vxU?qo;{+T9$%#mpJT;0(P59`{gu) zjjE!utg?m1-L8SkFUa5)1KyCl%7EJ`n{QhO{u`-5(R8`oMj`1=-a10MONbr(L zUe{@7fNcw4hqaN7s#kQwpG(hXW-8Znk?SfSRAzEyHte3qjt`+W2eTM^E~Ja#DlKY- zdADVx9ofFG^iyQsaV%*B}59)mZ(< z?Xsm|h}hUYH4#?|o)r+Z;Q0v)jtq9m!~ysir(vsa=qwS-i(4->(12lsx1yjqj~Bg` zSyX9wg|1lb)YdiX;xD*D-DkB%XK~rt+1wXXU~?S_U7JE<#cYpp=8V^6X))~h2=;<5Tz|DE2rdHlo(iQBfe6@d8;7yBr} zq~!k!_+=D<*n{uPfe{sqW+=wqjFJO@=DX!4Hvh^rpy+BtL000;RVD^xxJj7p2vLcS>} z0#HM;KYNFKxv=VPzn=X#$QCkor-QxdFGsJKRM}L6JPEJrhVoH>@z$4|HQM!3a&C-C z`3RT=qDA`8%R_%n$)WFO?eK=HTG?J#iR#Pvn3bvMSF$_l)y(ZVB_`pjJtv6v&oh=% zI?xr1Sa>b>0YJNvzhZnW?dzhf@DYx}N*nzB}^;mLv2bFk+1g zc*-%)u;=jcGXnwW%?A|L@8RVgpiAyBuCwOU@d%bo_(W0|+-C~AYp({=7I_6*{EWTgdppzC<%FWw3Y)LnujQmJ2w{qy7joST{c(G2op z?v{Ng+>O8X^|!&{t7l&+$SexV!*Et6FIp&4ZhEy=bi-!lq8`ncEz>fDb1&j&z52V2 zjxRhUd4Yx7xZEpeCq=9&J6>_k8<85rD}EwsZZ+ zHb1o_eSpzbAlj~rwnFFtEqAh5C;b2Ys*bBhaFdEqZM}1;qelQBhqg(BykVZAF*#|$(zYD_I&t7

    qA z(5ZEXUBlvqj)5#cRY>@r_26^5;;I)0QMn#=u7-h~7bup@i%Sws^0A|@Oif(02c{NKhn6Sx}hPbzCp?8Oh)8@iTe#1cUx3)|5_VrXA;XXlktyvi?y zo1v+|_s=SpLT+9bx<0j*2#vs%0-D$4*OP?~cx`-$JEG8Fz%Fo=v(eA|HV!P_)Dkn>c`a7gUYqNi_gjYc14^ba82LyC8sLVZO7MKTwEYQgigfu0O#i= z=`fzQ=G~-Sh+yCRt;Y$?Av5WtH!<11val6&RjZlH)y*ySKH);-bbK@wdtOLJz0N$} zr zUU@+a-v?i&#_Xex4|8fM4g>uk`S0E-z8!t_@y4LFsQ>Mt9#l+5yzBCVWfsev}rQ9);pWluV)I#l;PCS_6^d?!*5IOwh^=x zgiWa1v-#jvtYMAPUU|MNWmvy4fgf=_wC|U`+(&7PrX(sGW=lhH%Bg_`RCv$Nlw9tB zqu<|F&a>}*kWD2(yU_GB0_@QjPm;smoB~BT1%Qja@3(~*cz}=L@%c>CUNM0T_o!MMw^jJi9nIN7rJAEd{Hhw1QbO z_2Slzuvz5~zaOrJ_R&A35**oP`@x-baP#fgPU><2$WfNOPr>d?_+R9`70E%f*S0o? zHfud!#9XZNBX*Ls{lrSE{Iml&SjL>VP@fD+%chksm9%u7BpfLMPU56A-k9HO4)(KR3zEC=As|KB_={e?#48oK2IP7gqG^tj0{WoSnh zYIDs{d!-KEepPWFOre>rSeZHJ;T$_wW(>0R?QY&U1=3bxcs*KGm^GwQ+TC3K#-Fca zHntF&ZF8fuY3vIRAiwhD#1aomz)on9rv1eA=p4MK~ z=R4BlLw4(b>$fS|me3Eqd;J}|(dR$+-6}Su*ZB-Yv45UMSI>4FmTv_!jT;4B{dm=E zJs>D-O|iM?AoE(_$E$qaGYeOm`M~)#Q#ck@arshpX{BEyX||d6>N6B2oY>Py?N$t_ zP)CXP$E$*`m?~Ne-XLkr82p2+J4P$0Dr+QDu4brI6dhzG=P_UVyL4NT+CJUWA&n8+qC!id!C-(#)!@buJUdwKa94Ph1SnI+e4 zJQEv8fpU5<-V*xC8#k~#0=GX)saEhQtS=QkA3g3qJ&7?;A-B_su+v{+a`Yeg+~a;{ z{r8tzJ`r{x8BgoJp^BlVr|6yq3L0hVUEAOMXjj*I3Yp1**kcoof0c%v%D(jql@8Zu zW6`kN4KsrP#(|ooqGCx5Zv{GsNlh|3bZYTE7jDE+82b4ww&dRU^Q-Y!dY`;Hu!GQm z_bS^>J_Rl#+{B3Wy1=bPB+&=p7=h1|yKY?PV2I4gI~y1V_~l)zc>dDZi;@y8;IdMj z2qH(v?{+k2pmdHs-xcmJa(COmRR&NWFsG;is%b~%`!?!Sk5zUeJ;p|lJP*B`WySYG zJSS7rsWNC+J~@j#xU53T+~E|v}Vo#<1|jn7y3WoW`}Zir444a zVt-I8^2NUzLKuZt@%g+?d5~X)=<#oVa#2EU{i)Hccng~KkPdOEjG<@u&-C42!XEwU z%p>&pv*}8k!trsS85kA;&vQLIb0c->SM%RSo3P&tj(MQyzg#=%IkYooTk-tw!MFWh zudtUOcD)TUU_-nDCiN@es~~L7CV?U}(o8LgvH?h7wt;hxY+x^*Vv4VvbV)8LtmD-u zDVl8PrJ*9X?*sWiM^0#ASI+6|XiKROj|#$aEzNn;$YWH9P2&^P_<}B-h|Z4DL7M@_ zqZ%k{uU#ycF)zh??P#czonToXH?g9pTdfdeebP;AY1{#Izb^%^0`1AneEwcV+17$F z^&C(GDDD_9C9BW6J@6Z!4ME^1b8AqW19Lw0xFkIFQ+YrY%zmFP5<1na5s`Nfp+jr~Nm)0?;e03=)cXeotbl@A` z#VqXm8P;gwJV_pwY zDu`Hl_6e8oN6u*OEnN~G*ZGer70~sUnqZ4UkAguJFu5n(o4--H<0&OHq?Lq^yKbA` z{CktgG4w;e%`pDP!qrklrW}zi*M<`WlCMsvNq)K&lj}!vTRZ1;%_q%r(va$*FZH)3 zs-Y!iDBNA=3nwW!0O-u9gGsp+DqUlx=U<4k@|*NO;rbxX6rZ&!i#Ydd+DZ{euoOXt zj1vRSf_HN+vMU|!$qP$#J=|sq03H{rhp7M+0D|O=xh}$P;U0hfiZCDh;;(@37aegQ zSWJynEQ8z{6}FlB9EexK!O9-N-|{DAEWFRFQzlD0yupyU>c+a{P8t^QD zF%nUn5K7GmwQL|H%8JOE=@AZ>memq61y`c>{Y8ja6TgThH_$l%UDwIyRncI;`+DVF zHu(Tm*!*kkIu~j=bk>2ePfN0!o`#T0o`d)A?Xb5%ONdqdTLomMuj@YiYS0Y&`v<6D zrLXpjKqF~+A$si-3V~y(95E5xtAj%`uRGaF9TF9ImzI*I37~s(yC1odhCwKltJ6~l zL?&*)qXAlAd-v{L$w$pVM%AVsM5SZ0*(dm`Mj|ErnOe6%>TZm9ZY{@^H|JKkZsUB2T-+5FW z7OKPR{!QBTUv+Afixks5J6`$h!ju!Hb#-8`PV=EeH)GH_$NI0FIPquFP{mnduz_t? zZzjk5(PU>$CDBYkT%d?0iNJrqX=^3)ix(XKAT@v8VtqQZPP?cq2oFRW!`Btm+b=Z)d$0Yufb7;lm=0bjI<4b-&6&x7@Xwg=_=XO4KM z0`KPB-LOZa1*%S&9aXnSj}JEyaLi|BOL4ViT6n#D@L}^FxX9*dcvwKR>)rU2k?K z1VR=0V2VDP-MM}?F|mjIw92Yf`qBi5*|Oyz38kpQ8VQSm9``e++U3t|w>Z*_wbN)@ zrEdwcGe&cY@_zz+GgI5rjDyxk*(jiooV zvije4&jpIk-W3#_9vG*2zgGr*J%(Q1xeI92RY$$2^bM7Sd;!IG63RE#7DvHLl%iAE zLd188>wRJ1QXXd0LsvONAAGqO&AsS#dWzh}j`GV0r@DTFFSUy|(iH|89@3!w?*D(D z8$-}i3#jI@UxJ>HN>$akp+G$txTsAu#QE&DUWawjEH(YHe{;vdVUgtY!MP)*px%yP zN^i>}07e_VZpcw4^h*V))GUj?rvmq?yc-dBrmmnD6#%gW5T4|XZx(fSNLL*U4NylD zIhT}9?pzF3RJ31v1b~|Ej7}@IE#hmZ`@utcz0gN)V?Vvh&8qFq4Ix`k44?Uhxag0<}`KnBBt`yy6I%NbO&vpW2~Si$VT@~Mr!3hKFKKU zG0y^Hh=!E`&e+funJ=|cqJ{6@OI96Vzp&8mySmOET66wRUKkzinEr3yyhGLKmDpa2 zX7Y4PCXGd5afZJ$y_IeV3itN7IFJ54U9hkCfC;Qva&1piOJM}U;5-?vx(VIm5_7d% zIsHAt`U5M#R83B{BmdF;s09T7`0zXq{%>cx^HqDMGeKw^=QA#WsJeuKy)XtB=7*o* zM=hr-=WT~LMuZvJk>*oR8bm3Kq+g~+hdedD{;+FALaJMpkT6S0VbrMmANR28>=RSC zNDbhe$#0TTG$~D@3N``xDkcKY1;4=XSJ3HrlG3$TeN$_$yPD^bhz~lMYX7)dc@H5y ze+F&u{zG@5M>vV@vbKo{&e)t$FGfL`F#Yk6PTjgnQ$MmC}p zR7I_J82s3)dCvnDOtSa89mvmr%cxc*m(@^P*L{AfL6(=~(o>)fLA|-SQClo6RRzfE z|8c9V^Ei;Hb(B6Z6Ave*dtv_^uVPF$E7O7SdU7h{B1<|fMCTHV7Up2;`|qU3A7XIZT3B4YPZprboP26a zqn5d&M5UkH|c#hGCbAw4>RwCvkD7lXGU-JqbfLCKgN4ILRC+ zPTN_QND{GdEVF3N`h4a;^4HYJpr}k4p6wu0o~$@&f1LwEMr$p|Oiz1(%$iV2>LpRj zTq1BD=@_TI^8lqz{<7Lwah`f9UJ7}h>PsUV$-loROf@iDStcsG%d3`bF4A}*oidT@ z!@&6a;vY3Vdoo9^P8i&;ochZbyZZb*h3bG;sA7y2k z7`Nfzzh+eR$W?m=T9jG^GH|hXdy@OES}Cqd17nEU6(EcHw>~CsZra{L8B6L2mQ(&J zwdYzZK^wN$uh&Hjh*v%O@8Es^O5N2$4A7a?jjLtaJ4xdl>;a#KAWONYGV8Mtq(w3jL{@i(k(`9)(%CPGWIz-2r z32>u7g8EZ)%o09kpv*r>Tvih!3gmev(#UVAbq~eHR8ZdHq`Cz{M&L>Zs-&Z3aVo@m z*>ut5R~H$UD7=#OlqD!NLxykp%pQS>TnmWa?t7qxxBCcm{E^_T858yI?>duMvR_K? z=$MK>BQ<{w9b}7+;G}GI0IpBc|F5shL8%M=E;io>68q+&;m+09m(}2}#DSirxGc>k zL;~2<0=Wg^Y(EIM_w1L{*%6|R#zDY?1O#7zOMwjkBex{=JOvPofa^*;ST<~gC#1@r zo_7h{E%s5 zl1Q6~l`i#&!8Ksd8}{jGCcnG`9F!@fJ+guJ4DV?J3}?7B9Boej68M8kz0KzTmMSI9 zSu$;~xeXWE%hV#r#Aj)}?Z&TBNJZ}t2C zW7JHI4)WFvLv&a0qAItt;b~z%e@AC=scwLKou+=C0$l#Xe94ZTMb#o`F1&c{?3X&TYrp?2cM$v7Jb#DfXJ)aMc}*9Ejs+Hj?`PB*Y>VY@fpv-$M&t^JXaYy@N4J8`JqsyS+Ua{a*u zJy)-u9>Cm3i&!DsaZeSLawNa?rkYkH0-^bTubc$qOMlPVao@@nSqsF1WJV6|rS{7xls1Zjd3H5N0l=C* zdv>Q05gX=-xb$w}uWPgQggk~_2Ivr^F_S0Acv@%09n^9_wxfTlO>+-s2#ZKB+KB_# znuI`Qr4I1cMLU|4*2&9AX^)I)lU;)JtFi}Efh9VS!0b>F;WpaF8GU*K1d(IN$HZbV zO^H?nqqAqup!;85T-@it<}=I91w$;`D^F237&q^jQ(SS z_cz!yvwBwT@1a9&4G?X>eh}-K8yNwzf9#WshEV7t;puVl_C}-gU`4o5boGjVJqh+HD3Eb~xfGmfSJf0T zsQE-dQbq=vo14Y=Kf%qb(sTnUr%W6?&Br{}-xa+I`T{IBB3u4wq0}X%o&m3QI9;Bc zezugRWuxLQ|2UQ=^(>T9`M?{t?~8sTi8s?Z*&5bdm?!Y*oj7~UTAwVsUJ-~9w>iwy zz#Q&Jbaaf5XKHjllWo7D#>8sXe=4rAkPRYtgv7#m@|~Gn(H<0rYQ>6aW!yH)r&2tTXOyUh86a!8%9OqlZ_I9^ zX`i39L(0TRYv$-nLqo{un#o20dkAQCYHH3Jc$o9#+&*8XS0{{cb%s(1*;ho|Ov8Ib z*}G?Uv^%!6V(}YdlT7{5@Ca0>RRF;Whf-U=m#sY#5ld1b{@L7&zknPeKk)WJ#~

      =z*vVKZRa35|%PSju(6zXHa3eh# zEepNYWzgtM|71AyfHQK~@p4Q<&uXHrl37PjRX0I)J@?kfgTLAYLtknryf`H(DU9Ny!&KR;0%h-RTLzX^ z@0X8DeFYardN0^GZ*CcmIHE(`-Pd&;)&bAWXYU;m#%+ywNMxht*_`2bPt`GsZ|7heD_sE>Qz*S}M1^HLzYRA(t60 z*aU`M-~jr-iXl#7xOo0S^|n(0j!_E4XtVdL&&Zzc0s%OoXK5To<1^FfNs*1PRc(p0 zfYQec841?LOEJ>{IoY1GkxKNPDCMm$ZzObdaT_ulSGjtUAwW4;SSA7j@@jE)axao- zuZg|P%qtu1Es;`ER!OI20>;=+_Fszyhc$#c+8kRpy}M_CmC-q~=is;=L}U(P*CN=M$$ zTaG6+kO~Ly0*vHB6Qo+ zfP;I%W^Q-3gw#r>{1m;@W&z`FOEYN~Ni*pe>TD!Tnqoug6TCQ!42u(l1Hj2CE_&Z7Ag zI@_Fe=SkKMu|0g71;RD==@Ww)i!?uIzP6FM13XSo3YB4;YLzc}bjq@bdV0vY_b(#d zJQKSfkzLjjYhs8#mtRxy@7VjyHUa`pcT-7a@y4@y;my+f+~B0^30rbclT{HC*{`Yo z(Qh~xo0)HHO&3hf@^^7Pn+n2}NVb*H=9F!bG)v6_G&8ayyDX45Uz0gOJb_zhWu=^X z(-S{BcdEbPVj=$T#=*fwOicDv5ui}U-9ku) z1Xmc;D?(Q{>+*^45>P35JNQt854ttbp823XMTf689EwRy?#L0Pn3n4s`*KBQ7?NOx zGzCEk`4)?4f{GE-4qV>0izD!fUDdSOj})Mwr}*L~1_N8Il_Mr{shD?EDP;8cC;*k5 z1~m{bTmCiAWJ{CnMP%bkIl25U&Vf(vurv}Q2C}yDQ69` zQTBIu%a4sfYih9R-T_;364361+gdcf5hus*juHr6l;noRFGpKC6+6+;X`O@gZv2ig z;PMkPlLag96E7r^rje9hsFT90_f#YCsi2SpxI0fi z78RU~;vhDI#jjjS0~*Fw+FYNu%=YF)6|zm1kA&QJlNm0k(^WK>Sl(-_UV6J;53GCl zuV{!{yuJ?d2|z;*c<*kZbE*2vILh_|;EdFGu#ws+_t%ySpfy}Cty8 z=>6RQhUJD$NW%^SVC!GUm&eEx&z>Zd78M+|lW(A|)~vKg5OU<}2Ud;4mDeXgr;wj# zd0uu<5AatcUr0cdz@aFf*ZBdxTiEPB$ekp}v%Ap?!}VrrRczOj9#T+Ltde-Uy-Zz~ zk9ou_3jsNS)LpG46hZlI~5U5o>rTv#bzA~$pX&d((n#iR4 znA|G8=j@xAeHS(Ipq=2FebcHVC%u+BRZ3yAZT5j4X!nFLD&AdJr6DytxNW&!HNJ8u zxNuAjIGiHRf9pwM3%Aae6YF4qmTmKFT=NPJkoJu9p<^~ep!*5-R}Vp~8*dV$sIL1u5X{UCAiFB7k$8C07Usel7f=_ZQR1ySg;lAayzIOd zL}~rHz}Rifw4}UT=jP2Q@cRH(0+~Oc55ClZ^6+F`&L*^~b%$y&T2#9X3cZZ4uBqYS zN|1l_FG9Nq5Ad9e9p8J_@s-sVI!foz1Uhpk=tC&F?Kkg`fKpDXMx$DGz$57+YuGrb z7SuD0yLb_A^*=Q|&;_E&;xA1ssnk<1Us&i3SaEbc&DN1TU0A7FLmpueUxJN>cQ-ddHp zb=uGf(5C~M^oZcdxi=dHyh4~ADP#4~`l(iIOWD!(ga|O9Nk%$L~THkQjF5Z6o#z0s1rl5ep zK&4%7*Fft~Q71tQr@Ew=WVa-c+8xgGse+Mv3k>q}W zS(yjgaph^IQacRK>c1yRQd@)5CdA=>Wz~^$cjoe3y;e%i)#TzexW(LD~cc)V$IJi zV6sMUl?FnK#!;azP?ZCgs=m_2z;jY`Y6P z{~&r&uH%+tNvi{GUmEXiehO7!?ZdbzLM#SLlf5b;(kQUY52<+kQ+L3 z`JuC`egzJ&ss9g8XB`#w_O)@k zySpU>kr*23l#oy)q`O4~LApb_yG4`+DM7j!x*H^=yW>4`@BO`hsdwGAX68F*&e>=0 z=lMLHAh=;2aH~>55~JaElAgmTG?iD{U9ZeSVE5n4Ky~JEKUc@DiCywRPxwpeAc!&q z&(SfATtRkYEHwXX-(iegHUS}GwR|L7Spv+lqgwqJb^J3g*Gngi|y@{fjEcn`-E z0?Y#QZ>aK^j$t|q%<0Llo)8@DkYzXPb%j5Q+?;(eZPB6(r3Tbnf1}EziFv7xe_wVR zV3gjvg1{UYjm2Z>l*a21VfM)5tdwRA4<`_#G%0(BZ@Y3-8y%&F=ZW zqmLd;pi$PaNdW>Je9%O0V+$_ZP{tJgJ@!A8@=Zu6;+buu>-v)rDL%>N*Ol})DvH&L zh<_Q<*$KO+OYJVoeko{hUaO$(>dO4_!)Cm~SS5`sSJq>HG8X9|yxn@-PLE7~9i~X? zjb)sinkobmA0VTR;5h+_K@!W?L!ZtD3?F#N%>z+7Xv{nR{r!Souf%6`Gr`runqx-_ z55i^h1TqSralEFVx<69rB0b_E`}bo-Yo-jwZ_PVQ<5qquNd@skHTj<^2mG;B>bTE) z+7mYDipZEe@`|cbj^K){EAS%k>+1Ug?i0{?HElP>l;dx>b%4dtO-)_Rl%?<~fy^gC zq)L}%)y%)byx&uSvRx4`#;gs@`{iJkzJK7SFq{l-zyb~M8~kHpt8xR%q|d+w8MlOd zh~W_iw+i@4S<0`(t_wDYGSiue9hRDei2N5>|80vUEt5UXbC;$vsW%4ce|C--eA#hi z60jaaMDzx*N;`EOitA#9JtHPub_ zyw4w)4V{9(n3Cf2dnzrryN0pO(~kXF5NiYwx$UA%xAtEr87o)Ea>rs2+ihQk0~e-e zF@&?pRa@HtwB;f!YT37G9X9#}@xSbC$2@JP7qLr$*bqgkta&ow0&ndzeF{n*-sQ3t zm&iybA&%%Q%-F^G9Q#GpATM&c0`@pK`DmI_fN9eB$8`ZBd=GE#w#3U}Zh;$+W>y44 zjDi%~P!wF6jGq+V0302^vhvFNgb*C&hU4H|F~0{#l?nms7BqDY?%%R+|5vSAbxG!P zdv9Pm8S77I#e1d#j+W2MGg3cC^v%GyPVSpM)~w|20%-?ZY!+fNM;Hj{woq_%3;4L= zHx`JV2|JisrmVL|jtZz$s;SfQF-C!6>)6+hel;Eb@2>Hr15KdI*jJIjPM5J}BCu!PB>xbA<=PR_9{L2&`$VCKLZi9iP?<>;+$7@<4kN*ff6vAJr~4_-Fv~ z$dL)yj(s2JtiSyQM)PUlfQexuf_YkBpo)Q$xlfHr`4Yqp!f1U3iN}DMnTe%Y1$-Cw z02@AdCT;ysH@B7%DNb0gT?Tg}S56zK$Z#-g#$YE)aNf*Z4B@=&;0H_FRP(smtMr-@T8d6@TE%kHEWZ1d1y|-u_;|iH?on zsp24l0w#AfkI6r8G^tK1Gd|X2F0FlnwI-fL1J47Ese1sgBQd_9%|l5VHzAmo#)E`u zbHGzXLz}`tT$aLBJEh$=WbgT#2<+%&FdwPTlqCy6dhz*`kyEi_ql(A+qu@Kxh+9G; zuL;yP6D9&;41_}!@sGP?nN3mkZ#L2A3u!1NEP&*0&avhGAEY(#fKS1gk<2f!U@E^~ z>v2kPJ>K7$o6raZY z=8!Oiw9v(Bazaf~2;9mY?CZm*|61Xa-iyp0-j_9XL+u1cOg=4%{*CrYD63~s6~Br7 zxacx==<^(=$Qkybp>5oG`ggh|$-rCZN6FsnH^fS;@973zt@QFEVz}Ho@#VK+F|y_f z_$Iy)Jnip0VF7jo>`Xc_+RKIyC&^}JvB2-N+sDcnAt(i|yMavqv2=8@cA%YaFfqO| zQ263TD#F;>arC(ax*4h5bPdrctI&DpNW{%KqA2w({GG8 z2?fyZl$D+PX|2iA*;}Gh{qgbsa(6S`QPn@FZsEfhV9xd{uxtZ5(jckvoT#WMvpHfQ zM*L!|#QKEeivb4tQpyVBZmYeQm)DqyyCrDu#8mASs-0+a4!lh6x!vCp{%4Fdt4h}@ zUG5hkwI8qbl?esB%|AwekF;$U@cq?FYjUT4G2rMonH$u%h~l`6Qe*k4x3D(3IuvCF z&v(}>`jn>5a?(l01hxPj%)>G6UtFJNs+Ss6R;Lhrfz6@*7Sj4r^x`pBYEQfL_?MjB!Nj1X&QW!#-E zhkb7XWB=+LD7P`%Ta*cCD;N61G$s?cQ~*1RM!WK=bv(RDw*0!>2u`&P{B95>`wjTeOuNi+*bI> z$QIMO0xGQn80+EJNk98J;Pf+~(;Vy^bugh2_+|(>TU0F+mpAJg-p)>9=n)^SN^aR0Hy(uNw5_+{I8VlMJZh^!q=d`D_(0O z_(-v2asFC)i=$I?_HMemp1&cFHh(v=6^airh|?Sl#;2bFzyO(nhf0b!$FoQ0qGusJ z3)MNP8)VO%$%oduNKOANV+(Z3v>pV~ks$;Em?e!F+MkPy_d5m0zt`527^Q=;M#XOT z;<2RfU=i|(>T+VIsl(e~R51xei%D2-l@b4d6a|((zn0s@Lj$nahlt-9^Yr(^IM5*M z&MfIa{T(8SWLVybpcIxgd^9lOJN<;XwaldFcs%m%@+$N*qBOiWRwgQtZ|Q$&y&!0P zclCY-$J;*R_|fuE6mI?LZkbQ*rI{vwU*Tm|j2zkviET6le&?809(M z{#OVLbQ)hTnc)!v1Bu4wEvnfXzt30w_8xZ6@`#`9v#T{5*gLoe>945`&b%VdTSHz$ z8IO+u17@*0u&53!ZKDl&^yG8lW$RfE(yi%;kK0Z9m0D|BPMc$6!+})uwuDlL-lNEZgm%XNG##` z-YkVnp^frzwKMV(YS-hr4O7{OJ!lqvNS4)R#pxKd%< zVMlZfmgmLS2r591B@=L<>wJ3~G(>v9aTXI{A^kQSQoi3{hqgFmq zGYR?ky-IzQ(}tr#>|{a=L(Hg6=OIww8sNZid|6U~m-oilQ6LS;8gKo{*EF-!(XWy* z>K*TJ0;y3vEgCT zd_ny2`qjKn^y0--xuGNSi!=l_tSq00>npG-Luen4Las)mz0ICK2O$YmAX-2{NvXQ< zH)ux83Yy5pTtz!#MBJOpgwg6LYyJU%|2?#()feFgYfdi$o_Ksh=E(Jgk10^Q$phjE_N^U zAV`>X9v`|YIUsSq>V@Ez)q#=%nl3PG6Q;ax)Q|Q&6|Y^&2rPwv)`LDA1>&q^40mTZ zI!WGQZe;jp2FBpj*Kyg0Wtwxj+%YJUIIfqc#`2PKc(~*UXZx6)JszK^7#38s#T*(V z1hk}HcKobAQ^J4F!qStrVYPV+1on}z{!OhCt;k+w0BwGtb@#66QYksF#R|Y}R&g^z zPz-?Cv>sfB--8~{{01!WRuEu^?aiIk)$vI5t6(AQwOXlopq>P#(SqUftFwmI9~P%69L!oqmX@b5_I-<=pcG|9n0v z2nsJ|DJSaFi!}6J8ZeKRVTUUJtVI?+ucV|Cc{sj%@pKByxlhxblKzg?=nFY=F%INO z(jkhg7lzpb=FCHuyO_H}Eithr=dAd$sHIqo90>zDy^N_*BBzS~9$kG86;3*0vE6-& zak#rmf^MK7s~>W`2rw}v^R!E?01W;lH=4#PB^8w#?=Q zfa|F2>f=Cgu=6YbZ_S6k4|WzAJPh1MQwulVXa16`<3>fi0sTBdHv1%|@b|>)>5b=P zO9Jn6zA#w`v~A86X|F@~FL$?Udk%tRYD%DpM5!^h!|wpJkp(J zvzFqNirhKFm1^XYiC3D&#=U+k35a&wAq?VU`3-8?j6_$ihbSi} z3l~9FoD&09>!#B!fI5SX3(DNggjn#$$W9Ok91V{d8tAN<*R&tHvoUf|Z9e%D@_kW# z*!1{lXR_G7zZdMQmbAl2xc~bZQ4=wjC)lHg*R>WlBy9cq^PXO2Mv5|l7EIajxGb=^ zTsgQtm~E%58qJ&^*01??;9%M_*3sE(X={zKa3JdQ3e0flsV*Kxi)D76d;3c2kgu1&cLA5PZq0_?LVbpgv zX|`lsr}*7(JJV+Emk9^=PEakhJd#q6(^5;bEhJ4>rbzuFqv1W8A&;9DbXzaz*6@wC z-sRYor<=?6q@ahUe#QAsMCO{xOD)IZPKhrSDxF^fO+^&K!2C^_j{cRnpPQ-#Zs`~C zO!ML!1k6#YqL9k^WZ4>JB6n~+f`g@^K1Cn?-rr=$a^hhKL|_7M`!X=N90droF~S<^ zCF(~hrcsigKmVFs(k_CUEP{HBPsm&bb{K;7C_|?HX76nBH|0urtMh`rRA@yI1w($| zY?tE%tK~&HPgm;2%w%-f6V!n2_IOo!-c_ z{Z&4^YwA~SY1p%gG1U0im(H0N5cD7Mx&Q1T+wW4G&-)p2J{FVDa$2NJeoYK2Yfso} z!B@ASGMOb4dFZ-;(_lfc?n#*~ORcMuWkGjU6MK`pWj&I}}n`-@Kfyl^vNL!0+GH|&c zdDHHTBEisR#wO5@jS+IHE@Ju=4bxjwLxamm+xSfbfwGSf*0IjUFjtB8_lOY5Zjh^W>E2C$$+DB#^SOj zh)LR9WTTL+_I{!F>m%DiFtK!Y*#>0 zYOxl~hQ^O97cwHFpqc&rAu&0*E~Tg&Euk>&`YJ>0;mTALWL0TleXVx~T~j?q#%#-i zYnh)l+bo(ncv2r#KYEg)b}Wnk%nmgY-VjsQL{2b+M`j<~;GgE^;-Lst<>DQFP1lTz zXK9fFxvoK4a2@{J8p`e~xeON3Zk#&P!F)7R|KaET{rz_Ry}%DH zT^Ips0$ja1hZ@~k5aG!MI{Ne6Bh#4RYD{5ERFKqbVnT9-(#~KgZXVdxqs(YrS)R=^ zmetkp?`Rl(HJ<>)=13UmWf`w^`VQMn8vfVUpkWoNvsAaWdz+ScyYqJ$2sEZ*j?_+ z{N6DWoq$vZ#|ahpQjNVw!Yomz033#W}2d?XYHpD*^TMF>CW5fSSMElliB z$U8$l!91t3lDSx`v}$q?7~%^`-7c{GWtb)Z+r-ecqLL2bU)Bnb7V!)(SUG%AgeSO3 zk3ER+kLA#`Z96qd!||ZODX6@>SEe&h;6)fy)U?`^awnHiFLNI6th`-e%DIenxOXsZ z!tn>ya8<=PEw=yOjeN&SE&;2dnr=)J9fdy|BPIz4S}IR_nICAGS5nM^N|+1K7eJ9a zfeHTe{ET|cU-}oH&jhXd(a@F2bC}V?0;Nz=Pm#CxSSJQ2h1Whl;Y*>~$K^?=r9PZUav|Dj0j zb%A6|u(T^v)A^L%Af?YA_{lDNco8xUq%+gtYl@==z@EpHw$|f&{rWXTr@#)KweW=R zxJxc9Y87`mzx7h_FqN+DHqcR&M4yzJU53^F5^IA!Zo_WySsCKzEUgt5e}J;tYDyBlhDv$utTnp zsZDxvR6R|TrmiH-NZp9cW4;coz@19Zf16={5`hlQ{`UA(OqvDGR~HSr8NQ=4(3Sx4 z7JI@fe%+!ZnAK|2KKE(yJW^}4m|55FQgKcmAC@0G4uSFO4ip1Dz;x zg}*a86wVREEaDUpU};DoBMCU=nSzq*J$)vJ=A@uTVI0fDr*NT)MxO}5ocm4oM%0=4 z{p6e0{W?VP`|o-ngv!s+o3Q>&W2K+nZ?_y3gaWOk*}j8vtwmCX^O59+71;;#^tr9V zZUX!ofs-vTxX*rGkAolN2M{QPLzp%23gDv=s3QZ=(b1osNKXSa&oP+&&S>3ugI06L zMzW!U2uYQc&`fCWOJglGUW3#{BsNtZk`1YgwlkyiV@ID`JrD87NN*zU!N5Vka$Z^G z@R)ZO*lB(~9wpaEcFWDwux`gO@VL(G?7@isT5cP=-P0B1jx|=YR}W4tn_`KvvBq5o zfM3Dz?AhRh+PBMM+xC0^MLgJLFhAanDi1z2+i9OrpYVY2?qHz5zkPEseLvb7gjZE1 z&$0YsUSPRe;x(}cp$-=wSzBa#W$mXV`=sE1Cl!6WE`;opaX4L`{cL9S`a{SPi?&WU znvB+gq|7La42k0h;=)Aa&BGs(_8$c%KBoK5O?0SH&d-+|gel*?^LPfw{&qGs#)Mde z{NMqr*Ar{z^lQWH0u8z40XuwNydS(FV(_~?VYI3a`k)dDt=8I+k_Pmvt+e@C`|4Bb zq3o;s(c-t|UOo7u=PMq=&QQ=$$}KAD9a=@jrLo&Oo?~fGFJj$cz-MI*<#O&tE72hg z>-xf2@!<)SCo+I<)2oCp`<%kVis50St*9h|Yb$X>a5Ydxa3i5~d$tB$OF%Jvr zJ%v{Y?QX}Zg;bnUeFCkSi^ifCY07AlGsh`CvIohcxPRQ$ZSP#4cD9XNR->^9zS3#< zKFR&-q=Fi5?a#YTuKb|u$+0*vV2CP&YH0Uv+Sw&^`&(Et?~Olr zk>il>=z=h_fZqlDy(W`MR!!4HY32m$(aI%78`9QBtI;t;@YX4|_6XUWvb{+;+{}{U z9*}MI7*D{%u$$oN;_oh~5KbFq6)RzG2BDrMsU6YaiOQz6cwT*iY+gIRc$gXRNsfP~ zMSU6=MHt>1h5POzo7RVHv*j9$3tLr#t)HG5*0oN*`uW`R@!I?7OX#1V=RZzS-#v=K z7TT*0+*c{qSZlZ#R6J3x`t`4lL_)>GosLkSW^wci&bb`7FwO`C^DQOqa%NmlJIbbO4+9?WSF>$D7<9n4^mg&-^(U9;;YPyKSU!3Awu4S$2=jG4rfc0y zc8mQ|ZLCBjQz`D!Uxg0d8nyarob9|ZUV255aIuTM$!HkTTKB7#AAy^Da(ljx?EWS( zWkcIJVxtMty+>QKnK52kzL*bs*GgB{e{u?)0kU(>p9{kmg5hcwzXI3+j<%l{ez_F$ z<&NSB0uqbf`c$2x;o08oxQ+Q8rF)6c>wJ`G9mv*bjC^K9mrDfuX3Js%?V>f385PfY z4H$&Y36U^HT3GTpn-gGVO`rXxnwx}0~%bPXf9NGZQ zGN;7hNaxQs{&&Vn^R_Bx1VsCO%4iH!F^Xz|ggofs?(JQWmUc(&&yw*M z;dbNf2F-+X&|R|Uh)$O##NeEzpcF(PnRvjd+u$?nqNXCH<6wJivLnx@;2G>F`$ybR z^B@_iuo^ZS-B&ie=6>CON115Q)84CKNmM{w%A29;q8)FfM0iEyiC^=~IzllzJgYRTFRPC&UZr1X8w|w7N6PyzQF`U zb!wqK$u(eqtTDTk(qgUtqw`;(H(F`Lt^6#{MhX9I`j^18%x$ek*oBe!TY;0`!!h%~uvOGO=A<7Gnr2J2Z8fS+sq%S@vP#Y20aa_a@ zpK&Ro4YJHIDZt;rGDAwTWZ+|{kA$lfWAx9!l&3{Iev!L}$~4kyV(d=!>KXPq2gPBL zQ>AwIG(h0-NZCg(BI?Wx%osla$uwgPzZ|X)$Wq}aNM&vVdV>AkS_(9uqi>2a!qh5X zh@4ZmC8?t}6&|mzkFK4LC45hrNU`vzk7J%oDi24&NpDw1HLBa6+k{bUf(%nWcJu*Qb6FSUWgF zM1sZ$Tuynr2lCPBVzrU^9jQnSHms|W2-K@yL}HYTLr!``P*1<;Y4^AV3Mmt5NHUX+^FB5b~qk`e-9(!hHWb$Vu7|-t`;W#>hK4Fiy4cYs^h8Uz@8P{>z~|fKwkT{5$aCMzz58uJxpY|>iBVPthw-*e z4l{WllJ0s+Gq1zd{CvdpSA#O|ZVCL!+Zi%5OQLn*O+gj);%R(~DbbOIR+he7H+R>&FrcwmD5_VUotr5TV?5CQn?$3Bx=7x7z2KU2N!G^fXq3;2%Lz##u*%1Zp|6HM#^ zcATGVJoKUOC731gRnvuDx>{bk zU6F~z26uvr{D{iZPMblN*iX3FO}271JVw=voBL?WEbd7(b=5)q>UG-FGrr{R6eRLh z5uXQ7cDO3tv^PCW6Jal8KjS=pJ;(MEMbCFS1W**aF{H$DLS!GG_wx_79ZC-(kjM&z zG^rIZoN{1KX@J&~igo*N4!SkUvV#?s{X}kd1g_u9e{wfs<9PY)Xw8xM^j(|3IMwk? zFDDie!4>(IrQydHZY3+9I9rOZq^W#*w<;MD@6)k`wLZSMJ4ADpT&?t9J|%M3PHA<< zp=STZEUIpi^mKdj=9IL!K2P!v|8vJiEq*@jL2> zGZix5xz4sC6&gR?YP&&CzAX3hCl$kO9)tataPtO;{P*+Z@Z{9W@rJJ%M(nA$&Istt zAr<%3M!W15!5O9po6g+JeoxU(pY zbMqr;v}v2J_#hOZM%fYAStDHXJ=WMNxjleM{PZe_?&<*KG?SXfZdZZ=n$Mf5 zmUF898Q;`*k84D54iN3n1-{4USmByU;jO|AYA@h|)wk*^{J%qx_37M~$zm5E|(rYTeqQohgV977DGsd z%1-*ovTJ`Um?c%~tzNH0t3U*f6|+Gnb}D9Sf>7V#OCCjWH2M61{gFI|H=T93HRe^_ z9mue4r*v%c0|3u?EzDbHNyE-HMIv01J?7pWN3mf2=Btxy@xrTk)yMPIdx-OhM4wWY zuh$yoaz#5ma(hy+)Cw$}RTQw@fG`pSzV~Hag&hTFGBg+-uX~OaDAwK>e6+s`c*btX z#&E>DkWutuv|%M$`L6k^WTwSu8uqK-48Qm}jB-Y#3E5bQLb0d!U-j<2k8L47q4K*e zfa>Ixj>2vjFT=l77;kv)&HlndjK|~>EhSOsCVJ7tl$2#L)XtQo&=4uJD~*dQfm!zA(dlU596~VEjrHbs1NKJg(6Q9VwqOdt9?_S~rMkw^s4zU~qUIZ!}Sk zz4ReJ7ybZDqT0Y1Ir6Q6dzY)!8m=l}@7kVpeqmuL8S?^w$vs?cPfraEjr266YpNO9 zm_qri5>2E6^)Dz`kB7OEA&xkXTVX`VwWiT8NMKQnk$9Uqp>BY!q>YGb`jXXQ%udkb z!p;^;6ohi@OBU8FnW>7Bg(y$6Wizs8dz|lVEV8%bWBd_S|3)?#e=6p6qfM{#(cbFB zD1_@_q;(D5M>$5Z!nT7JdL>O%{5WHu+=G!>fuYrNs@--UATDu&s))X#Pe~!?uG;1- zmDN)P_pL+g$6F$}C$kdYQ-hNBxdkQ=xW5m@^bC1H=DAPmx^mgSb>BD68=&%Jf9w`d zBlp4XEbDO#) z-s-Vrf2ykhDE_L=`XRjA0@=ynbEi&5HKit=>39f&yL$8;y^6 z#r{rEGbR5rRHg7Q{AN)}ppbmd_TSQXxc7tf=j-u))k8npM{-3~*VC`9;${tWd-zgA zxP^p-V4~ZS)JIT3b%TO&oZis#(KCK4M>OilBls8H*3-Z(nn*r$0*8<-F36`N(cn|# z$77Y*pBW?Ch~BPJC!;9TRFL4FT+`t7+p3@Hlch6ixEs8-V1th?8DWqIk6h1^8SPu)*k?!fEYQDD?e0_@uiNZy$~jZCGW z2;r%xy$-0z7ZPs~Wq$9owH2Ig=#xF}{cV(j@xA!^(WNvEqr10Kq|9xuoL*$BS8PH; z!ru6$`@8j(%byqWkyBJ{x4F&Pd~?ow^F7c5?EtidWRg6$t)~+tWzuyi8cZy5>1!q& zHhFRCgv+!wVIAs(iyfj#(cJSR58p@qJCWmMV^~=~!k44RAJM0daLL1BKoflq@c%)OO0} z0mnY?^H-029xr`L75W8|OgEvaQ@%54r7x}NR&S410&sOT85m`?sTtZy4>Ui1*9%lc=R=xkTW~kS13pmp< z-e}&_uPL>=NCWP;Dwr9E?+sTjQ6v5HY^51-8+DH|;!kDMj3*`;Jaf25n)rJAa<4sF zj-x|PgA0)?ZhWE2M$JM<&ve)SNRVN+I|OXLH_Dw3t3rD9WH#_B#ncy_@M*=TL*2cl z%&>?1aj1J25~To#ZW65Hvmcg;L@u&0x)v;CxbX?gr410Wb$@>es-756d+P8(k3(+L zc76C$3Xhb2S?Q2D>rmDB8n7!Y8!d1}#)4}(4N%MQ+TN_=VorSW9B1K^e%Nl!qj`7% zirq?X5p4dtTY+;;j6SM~*egd3qsdwQP8a_$k#-pWGJ;$) z*tp~AXg;b+0LRhCU->QA_!K(=8QB%_DYMqz-d}EcY!(PaMZt)F=;TSL*nF&Lug913P-8t4!PZF0SFGSO4v8@Y zkndY%e}Hl;=OYa}3c8I`_GchVMUv%$_#O8MfWOc+E5;B~WYry^Dz^~GepK;S@rOj8 zy!$QP?%)kHHHe0NS^^!)+q+P|QAG>6$$k)OEY#ZF;|FE&{Y@ztfcTkqLm2Hsj6h-U zFiW1!Gincr5b%4wJ47!G9fX9DxA_Rkys%@ue>E(*cB_EbV>KhdGVC9)e(j&i*q(7K z)BzMtoWJtRg^*6p4vkDG#|^B}I5=IMI!aD9j!eU@wJDHvqODaap$p?K^brvy5*C}f zX6EK;rZQ>IkoJ2wYu(@!knL*R;85{1xhO{=B5VaHMt~@gm7J(4Z4ANbU zvv^5je|ml5uKw-pc+h|11^W3M7&}bZV2HB7Zd)XNx7-m{1{_)Z`&e2bgbnrI>eQ*Bb7(`2GoRE# zr(}D25V@ZRxnpa;Q{JaTaCb+aX?}q&;>u*SHRP6%IJ%-0VDS4Tw}mnd3J!;VeEfO3 znQ5+8x1?8yH|LJ<1d3zRad(t4L4i^Cfsf}O)1IiByu~0=11se6!uI|$Akn1;PKvb1 zbt67bYzN5xanE26pDQKXULN(j7!ias6$|RL`l2RE`|QU2#+PYv05-8 zb~gRfz=}|nPy0x%r4FmNFZd|LJnQS^cIo1cdID(0VG9bhx{N?GBo{mvr19^{{$kDQ zF}z9jSyzL>Zx{AaSKpd8|9wV{KzuE6AtDh){zS(icwtw;Xr|830XxHR`Zc1+1E$?) z0@kHIy_q_GKglYLNytcTLFZG=V`K}H=wb+U@K!}HsNjbEQ>gvq;2y9E%Td1uF z)@^$l7_%u2HA$RL4_Ds#ZOT5%9(|J`#9>UfD75 z5e#9dLKU)Jc+~QJH47dWI7AM=M`YFs`WCFd8c#;RQ<;BhoGjF zyPZC8jkdZF6il?IN%!Nc+^d9aN~B>I|NcqSJh|YIp(I=L`puh^wN72T1=e^Y+p$?? z|A}T;eim3`bKX58YCQ?~JC-Uz;(6omuRsEs?z70P=a3z4^T4fodr3`Y|2yxm>M^H~ z@FvL4bw^9KQ)|R0)8sDLG!L-a1NhKj;`)pFfy-vLtpm~FHK%#n48$u2Uexm^g*g>h zHvkb^QO=^N0+fBrGg1C`PAS7*dRbX@F%CQe(>w~V5!Sqi+dXi+9qznTL#H~Iaoz9M zynEN4Y;`Dyje#SQ!pFxffZ!ndZfLW)*DG#$e2!LD5AW(8y4=>>k!wRLMCIv*K-Nq3XQq=IT$W%Tj6oVxhK zCC}xH9XXC)zhOfYu~+sn6viQ3=vBKkA{$PBULjJ6Y^Y1vs(up%vu$keAWx@z$e`;LVdE{0Cb{HfEu*8DE*_aY#NdOtgOFOpt0c8lH zbJl!alKL!lRVBke%x&YH<9Bb}PNwaSGT$1{DB?}pXq3F`&%xdfE;-*T`rnMx^9*5~ z&V<%JdTPwTwn4^tBRAvb?o!E-j7@s>foHs&DURr`F11DaB(sUco6o6eaM81rxa|GJ z&3Ja+ICYdtTk@hepN%P9fxvVlO$Z3kFdg*UyOYN*{a!IJ1nXkFHT0V7zWTrkbdKP& zJD$JB$u3MmN^fru!rZ@A*^k#JAs2#9hsyRgf3J9AED+du@N1x1V1n@w2y#GZSx4jB z&?&!7jx4<+jV+SA-z%R``3N2?x(=V%{1{V^@sM;Txs-pfVDU2iN!qqNOjYXiz$Rmw zqvLYkUtYm$$%}Hi8$8O0LSLpQPRBX8FpnV>f`{sNYVi;y^&-I8dB=#lc~%NjU2P>i zd2k>gk=2tIxVB*a?)bR;cyGwA;C<+^+yS(e zwvptKek}JTS%qk@k)UxO!RvcxZh>+`toCa#^`t$hmzAy` z>Tjw+;fTisA?i*Eqo$Q#GZi623c_2(Wp|zxdVl|!lo#q7kjG$&XV(5SKX_I)<~$gJ zRU0g+7P=|r%;p2P^(5c*dukDMi^IH?f$4B&k_Y=z zSzepAAS)-F^Ew3jz-_zjde%TZqiwDqjDOSzIukqnB&fVdg~c%NIY;tj?LVB`80~6J zKjGrW-<~C`jW;Scz*kU=m8{?$DXEJ|M6rt;B-C@|$gS`CvAij99PXg6<)+TDR;yP$ z7BigsOj-xsv+qeVkPVc<^m>TzoVtjwhCVe1z^WF5F(Nk*fH>QBuKMEqwQ?Hd`zX$k z_x;T*&X=tW=cv*-^GRN0&Hwaa@^84+eo?db!(l+7)$dCi4fMMe{p&R9CE9*`&r6iy zq?ftpLU7wswIote(1=*R8(N4ZT5I$Zl9j%GRex zIKF(bAHqO2^1wQm-g>Wkn`@<4Uw&;WUQ>g=nZ!!@Jc7)1@@>eQQ86T1To=!eBSZ@J zABJh(&L~^c_j$e}_wO^Y2#Weh117p==3MN?Y$@DB5v79ZdW!@0+y_>mAH#nTn%DtZ z3$h(p3!rBD8298~1*3GzsW}DGfGgY+J>NeZ(~=c--*!2)9V~JuH@#gch@?VQ6}+3j zkSc$*G|k> zZJs`&7@P>#^hr;$Si71zxo|soVk-;flFSj5C;z8c4kVs%E@mHO{gmIW6cNUC`oqha z&Ydj~O`f$)E!}MSX9}4mg&B(^b4Ep|P8c*Z#3B{R`j2UyD25(rYfP=wd0aXiE)`1V zMlg3oL|`Q(CBd{9uD6(P<5${bM3-1%U;Xp*2n|qsvxDW=rRO(hXNgt9;hLwelFEyx zAroiB6u2Gu`gQ#_j<`t*N?TmqJvCAYwljW507^sXDp#UhV8t)$K?lhAH5=nz+%%OH zirJ+E5Vd7q3hHW_o~lHNL6~o*-TQX$JP;n4mw)!&x{G8NKMF39TG_1MN0<%)xn7u- zpi8E>^vi`^?CwaK0|+V8u*unZOgho9C?LIZh?xJbdSQGBf(F%GIy{m6#{r@s6X+iA z5)`rq4gJaWYooRBf=7aK~%3*J9ZOt>7QS1;95|-wc@nM91m%m{aCt*)cOq`v!{&~G6Df%=BF>8Ks^qQ#Hy4m4B`QYgHA`XOb9+>+i>$zXk*DRE#fi&bcsS32gpnn$6+0Mq8N})2m#UMg= zH0tLwG%KX1!pMWd7?(UAFv(O&g=|~Yn@WJ>=QMENATH6Z`Jw)Xr_pu4U7pq}5cD|7 zftvFx@r<^L2;hA*>ZgMJ6875LG8J(@8J1-+N$uZrcf7I7PK0ARL)8F}&z%)-d3YhI zDD@j&o@N`4zg{g|@!LjFaiJw83ByOdX)x`sm#j9rJP<|-KMGkq0u2S}nSk&;D}m2gY{fIgi3* zl?LCGQ%$hFz&OEX7U&E0oFe2Twfa9?gLy*-2vD|jdDa6aDkXxjS_Qbc3_H^uo5P01 z9~UqjiHK)2-~XDOmrMN}>7V-VZ$FQ)l`QP!-Auo_R{UI1*azOz2zV2Zl+;ZLI%`(E z>3OJH-fvW3me}&ENtm1*;!vO^!1nh>m2|}|fSax>7e)(@LTp5<>P?oDbY0D#jbdX| z#x#aHm>QG9ru(}0F#{YFo37S{wvh-XHYpI^vCd1F&~3@%unSND0$}bo?PkvgeM=w` zgoQf7?5lqsZV!7paK-J;RmJ2d_Q9xQWJix{+~v2j<_2=c?;gW5n5nd$BU^&EM#g_kc!C7$1Igrjz^9Bk`#T@ zRB0f|hDwBXkO%z&NH)@tK=@i%rD;zd@K^T^4uWUrX&h7mi=3zjE3~Bb0to%)ts+6J z&rrA{uGm%1@9cJRb{jSZf^Z+rZ>n7YbmGt@@&9WSX@tBZ%J1u}L>op9D`JtwZHWx; zz@E89`d^7LJUXX}RIfreTmq@kA>?AG__%knoL;Y*mcB(7L#YLZWU;<&n4sZEsu6D}pxHQniJ!+&)iL9X)u*0l+2SNkULvk&=01^&=ew_dH{Y`W9asl&cs+ z!7KtV1{9lGJq23VKt5-5d$I3I-Hlw`dkruWRhQzxb~c5jph?$=J+|GW>g(AIk=3{? zKZF3rtE^!6&~oh$i8sJE9OS;$NJc!FE+ z1wV!_%rRy{AuF4P_zG~5LCuJGe|vfPfc)BDtY&3L+9i|DrIt zQ6sJ0BB(zlAdvcKV1=Qq;$qP%8KBLIjzzi2X-!h#rxqH8wC1ZLA?ki&md?|zpl{;^ zHp$2jxf*h0!%F%wg(78Z-y$k@4S=ejkDABa47_lu2|+Q{Pl`QM_HX>3PI4HpH9P1ZrDu2p>H=)e1r<&=XVIx_sO#25KFczM5%##K{3 z#${sLEn%%!Ut13HbIa3#H-1$*l;@i#XyZ2NfJUqO@X-$p;q?2hO@^+g} z{&-XSw63tQTWdi>Eq~>r)X@tQY@iLqTSSfjJxUh8Sb`^dV4t?|uNZ}j4W5T;(+Q)8 z9%2_nKujTc={qkZ(CCF(5`;gWBgjR-6~03Kci$cxd*>N~M~%JX18Gsmq!Mt*c!ppR zB4*>OOkf<2mhU;MbNKyRmN1?9I*{Ra1LG=7)6Gt?cAZ1!B!+eSX}pqdaXejQC}c15 z^Un7xfXuKJ`G=sv&zv%xJiHt2bH@nN(adc9Wy$+zdy+TZ;Ab(gMpos1-;r{FV+qsSU%c${Y}$G6XXgr^a^uKHH|;efV@*@1t70;DQPl)!y)?|N2^~o4br!y zN=P5Xg%K2i`FFaU76qHaEJ#di6|kM=!Ptx}+_$++?ceWn)bE%vcvh%lW|u*`xo5$_ zB8Mo9;06@s&eG~vp(ca_Lj_xG6VEe0vpx3=;@eXm7YAg3^HY+D$?h5Uq* zbI2x9S=Jbyw~^%JFvp;;#V=X;)vA3{bls@7!eU(4;DVfu%~K#x^v#tHnY(Rh77tFx z6~vDIyD2@?bF@tO@9_;bp7UjsoR5v;zU(J{J3PF9!#^OePIw7vb2wo8^PpnjAP5e|f5i`a%^M53)D0Ir=tpisB7cT_zj_0S%VUW; zCkeYxbT31<#BbxT<(*f+b_X&u>)T0)$34MK=%3Hd5Rr@4?zY-qb5PX{ zJ!bfglsZErVIqvTjQW2>U1uQGaoe^Xd+!~I?7ca5g(4E!>mYlRz4t16l#rB>m33^{ zhpg-^Qg*_z-`{!adERg71Lyq5@4oNrzJ~YB_#4rdk-ZX*974FCKnQIliXQI<7EbFi z>`7JMnndaT_oa7e4W3}7{lNGJn;LjNwxz;ae{7+e$39<*cPfwv^hUS^F-opf%bQVt)LHoOS5~iy*lb*bI3KE^T%Gl2b zI3`N8YPVIZj!XU>*D(i?CRnqIRo3wnAs=ujSZSAO#8soDtKKkMMUsJ9W#>|=`}h-L z1gG8w7uW%-Hw~(i9$DQLaG5UnW^GS@bLnfL=7Q`p-(PC2`0(Ec|9)(Ydzkj%J&#$E zn1^U<=V*&}vgXw=g%(?=fTDFBl)W9sM)01!_$-1y&e7=pPJ}Qc&6fv%EcX6`2Q@w2 zpm|A4UB29uXk1T{`Bj`?9C-W6(vQ;*k6px%15NV2>bE= z$>NK0&Jx2az<;=R|GpwP@d<(K1NGZIy>Ho*@8EheU~-UnjgiN~N@ksSQ$(?1;*Rve z%#}n?=D=N2jUgd2k&WS({f1;`b{!E$xqSC-3oc|V*5XtFu>HeX;TYq*w0Xi^Lmg?U zA|sNRAbYANwR@82-wCA@AtjbR!fKGS%WqF>oe9msSbu>T8*XywYh7n#1|g%~Su|<@ zzXi89H`(d=-O+XhK|4~0PCuUbhgt?QC(plqef*yf8$)~#iEE`kIE{;lX7S+2Oe6ojCrP)2hn#$b# zux<&~#Qee* z-y{xnWoUrAcK5aap425!UG@UBQ$Y6c3q>OH4YE(+0B~w4KpQCps;8q~?7zc6^TP+% zYH}Ny(7{&d=cbWgV}V&mXE9uNs6T{*icA4zXO+RJ317((t31z0ZxkgOKb?F}f)a#U zVvUw%ia-WQZmS&W495I><01&ipw_&kx%gk+p$iREaxEm&Z93J*7m(gEE^u!06G?fs z{#d8fsm8(l0siQxH$0yD2-oSgl;U4;arh|3gIVBlx)?5)8@rJ}OA&*j1S2;Df@95q zK(rBnmpe#+kT0zue?AJ11HC=nBZaHFS;OI_yMJ#5#O!XXZ)_nUIM|y^?O@sIK+6vq zYbtmeYdPob(P;JMpvDG#5l*NYzOFWdG-a0fFdg>os{cL^U~D4Ghj0*MJXS-;!whL; z9%1Y+UCpAb4U78{3As~h0<)Hst8J!_N%p}}R3ogNX^K9BsMofKZwZSIaTw_8hQEAC zUT)UD&`cZh4E8YQqT{^N?6q%|Hc_mI_oO?H7;LUW4WX_AUXn?nmX&Ry-_C zBm4IfK-g#Qr{rV4%F>aZN;rhBp|Rc_Gg5?*@$W2dYl69Y@8x|iO*sq1L4jf%-QGv= zJ^Q&<5C}~NeAIek7YXKq3c9L=9mCkz*sVo)!)!H<1%TKSneFZE?>%~?esg_A*3F#x zuk0QaeS~n$?leO(9K|vjKsyG#b}w*@j%T6<-wq=&#LuszM62WF?T*28e)lX+C_NV3 z2Ujf_2lQi6L>?xf{@oR{vF=M^GN^Z?GX*lSTw!+~HhG!&+0X-h2? zU#Sl8jhz$qLznBzvw9EKPwMLGR=}nQ=nQdCZ81ubEcvteRgNgi*eefQi7ZLCTjX{Y zRDZu`1kJY%=2Q&w(!O^XYURA6Y?{piN zkeQH>4ug<(<%?J^3BlN`1t4xgoM)=g-+oicLyoZhYg@~7Q^uf#eyIf9CCS%QbfrVYByg}-_KeNa@8+BUFP z?`8q&rVjkg#u;hzu{cP?Y7wdY&PrV=q=m1bI*FEZAD;j#2a0bqiYYKC{`2)dOx^bw zR~JK!T+P%yo8290WSsF;+bmX*_N2uy1ATp1d~j~A@Hh|-5Yh^wRX;BKZ)cldDt40% zzz)}{#%$FQqXvworiaTwYIkn)qIsK>mT9+wx0!MHWfk3Nb-Wra)s=WrDcb7j|Vh$OB3&28-4Cs;DeE#&X|L4(AG0@gUPpjT>6w$8f z@uDc__j*z!io;2S`7*mU`?gh90@v(m~CGSeytR;`f3Lw>wS6HI$zYkGV!tpPrhTmUy3^6m5mRSo(8 zu$@*RNH&U)&Jz2d98~2_mm1-|I7O%ybXB^5V}6ku+$6lKd__U~m~5{cI+$MADJ!^) z$SN6DE6Eul*9A!$ZcW{S;MS)CE~;$Jva@%52%7)4HveO1KFMYN{hlp(LhPYFrO8Vao=VKi&8?`X zmzGFmaYwN?HvBxq!l#Tgn_?}?x z(3Xk^`KJ~w6i`!xl^9~xCvC+lgG($F-9pBmXOTkHq~xgzFoRkFn+WXgNt|SbYa~#z z8pEDJ)JsKE&!gT5ckTA%l4Lzr`*&chk(O4V-~An0$Xnnlrd>67yGM|%MBeUJdB4Xq z;N8;QS!k<5k5hhDf)T?1rKoGND5B>Kn=R@=}o+)xs z%)qI+hS5LrAb0^6qwf@l92da55$L~Ixgxd>^qyjFY;2A`U0+=3e9@mq`P5!R$7xIZ zP95G@y2YWOuG-*mbQD>)Zi!|mN^a?{K|}TrM~$%Kd~zFwL2BJ#Ov#}+A{AJFm;j}V z5nIf&U!7;0*Nf1f-ur57D(aDJM7*A`#}6FKeTufm!AX8knU!6^j+kSIxBtH_5l2M{Wsl8N>K{n=(ppN{_K;#Q|~*dJLP zJqA(8*0NbOqY@IJ)VIJ3LC7dMN=@0o@qmu(-{G%Z(2T`pfamCs+5dui2Dp`6;6hiQ z6%0fmz;=0sc`WfD9360IyGXSD;*q>+QV^bVzTbD^Em;} zfPW*ev8+ybj`JCR0bo-$<8p}%`Dugb$%xUfDAHRyZ=X0M zpZR+9*WacuH011DYW1Ioq_zcKSpFEv2jvkHA0NNP|K9_`M6I8!pDEoLL)KmYQIsVn z4ijjT@UB9N-;Qw5zTteIKw7BW`7ktB9KGx*rdc31;tk2=X?q7Q3)-6?d|o3Q$5{;R zwUkm$uqCo7295p{Yrk{w|6g{&aA|?{aQ33mQY_Y-^yJNxHBIxO_{uW#&bu`CHSfkF z0&_zzH=ujuK49zN{{W$$)iVl4zBM~GVl_|(6G)^`Ko)R(o2d+IYHC6P*TO32MQKW^ zY=LCzLUBC>&3_ewN2!`?E`iHK$<(Sc*W-6TEvq!$y)6me7`75cvLJTDdw6ZSn|xB#$nCL5TdVVZQs!{KC7SZgM_`bPcT(mzY_W z7KqKOn$5kv>;HS(A!Pb0XhPheo_BE$=L5*2lY{2(3Gw|up zBJ#!!IJXCK<`DQk4)Y6$?;74p(~mrhgQDb+uL|H=%o?UJ;fM%xl;F7u0tbND!}UIP zIk`Zy!Fz4;qW>N?GENoAe2L0Cqo-KrU!Fj3YD2y#LhIlRx95R~WSym>{%Ppq8%%%d zM*c8xE7gg0dwT2dS!jpr_@l#fT1qh1DrTF5 zHuE`*OlRqV5`^saFNVE3 z=Fj)dNntyJ22Y47ZPEQSLQRekNYojUwl8}6WG;k`O$FIyXQ4*up}lN<_kq{vvzZBY z-m>&k)Rr!j*DwtDJ(2+10YQ`?8j$UX_)VHAQ9ZF6pd+t=B?0(63!!E==hH~A07||& zdvnui;2JYh{Dads3wcuoo>s zcm4TzrH3Ze_O;tkBA%Q%_FkcCN7*N7E4IY29kqL*T9E2D=$JS&X{jjYaT*ylWvv~O z?kKd#xqtiL9cSx`=(j8QA!?4?=vZ+<8&D1{y*I5>CCt)qgt2o80o&o{KZS)G)O+Tl z@!7gae&vbcz-*c?<%Nsv|vZ6j|4;F=O^9U zQx)C76DJA~6iHm#&N!D|Z1bU9782<{wO>mE4>O|q(5i^RmVZ?@9duyd(8SWAUQKYq zh$ki0AD~KRBJiDkemZ4d6*RI>dHx-~9NU3F8!#8WW(nTeX`&RNE(QtdBQq;=T&2pP zq;LY7=FYva^anF!g~XE>4KB+`i*TQLK;7{L{uXnY!oN|n_LRqJrE(dWnLPjVw0VaA zyj89Qqcs`l#jyJ#cG0VlkIxJ0NV#S)_0n>v96}?OD%ZhCxl`cXWQAK5cyUb}mYNDh zO_VGtuZcOblf8{P&>+-~y2&;yl2IB5XJRD=Ms*}$*P*YkKerc3iB89+6mCwG{)hEr zSy_c$zgH8p0Eu{`NnG5*>ot|X556SUEqcjsKhC_pva=O)vaQE(L|)BrERKo!3E9fu zQl?$1e0DD3k{EU7`}G0*`cIR@)U4W#4}YKG(f${O9P`&hTuVs|`m%94`(_>;-txjX z4|S2j-`=bgw+5>&e+@Qom`M8pigE;5CrdD4%PDh}Q?8GdNP*5lMT7BW6*%Q4BVla{ zH)mgw?Kc!cD_3fl)|6eztXb}C)NmOtvpy0+3 z8lbR^S9c)4{JYaSdv*D^5>!^alGlP))<-bg(fCB{m#>EV>TyRKJ~8avouSqid1sNT zI>$S=Y2GRJrU>!~tnxMQ-|tLeQ%-sPItc~!Z|D@MSpd~LZi0|iuQ|x2rWLYO3AY+8 zP^1-i%$l{oxFsz;cQO61kJ1Xm0jGUAiJF?;u2*lzOI~xAsHO!N%u{s<--LkvcTnBV z+rSqF9?0wQ_En@`!+L7@X;otfA_lXcd#Ia1hI)1I9f<)Z+f8eD=f$Io%Lb%JufvRX z-uhy5u#71F5`2|ve=5ATKoWC*viz_Tz3(_DgeMAdScx|#HEToc6i);1{Pk_KW78&q z_j--Md}DCaMLlL@5C~a7gA1nc`O0A4^WmYP?(^gAL$Kx82k7&Gt_uAF#zS2->Rghl z|FwjLR57UKAesBz!(OC2 z--fbrjN+e*GxK5Ny|VHoO2({+VU?({^T_a+-PQc{B+}pc+BTv9VU1_ zlNcn3fu}aMGw{SR4!$}@4m=79+Xl`UA)WJA7nd91i733F9|{x58XZ+Yy%5V+ga(nVn;{72Z)YZoedR=y2c{-IdEb2BYuF zDD)oYN@qgYQ(nJ;*3{AL&rI!IMR@x_*`GFuTTRIU-I;-#|nb zUfE{=g)*i<>z~ICI~PE%83UA0_&^%8fgav)@t+Yld>!b{dr+2Rs1w;a@?VeI`#OZ`sq^a=3{CiMMfKl&<%LLWLi50m%;k2n6lxW8iFkB+0^}G zruY^It>|Q1pY{?c=1L^NFrhS7g50%lvLB1A8s|H!Js}^@mO(paC;j-)ZN|P^V7(Z zGVal>dZTE74Ga_%>sJo{tscbLIrs$}4E-iMgkf+6%)7rsxIE1AvV(j2(5pA`#Wo8G zS%2`IO0ucI0Yc3SD^VLD4x3aoG@`SWub1=32*&~Csj)Jw0$T<{SU)Qt)6Aa;dDK(j!{;cI4J;1o2fQ?@Ep82(^Uqrd}wJt7Qp9_4oJdGc3p- z-pKv!2)}d+)^^fMKW&+~=n{j~Bk`-YMQmyRth!fc_f(VDr5$!{zEDYvPcHzfT^EsB z8U5HjWJmd{UAZa`z=#(e7o(BCmT-7;T>0&`TpMnQ7mKqx*No1=QwId0R7`_6ZzYBQ zWv$MSRir6};H?iVrnz45#mcYhxIa1if1JEe+-&_rOPU?GOw?NS@OcbrK2e;e=pUiI z!rRz)^qXp=GnHuB#Ue`AJUi<>ORq_dln&HR)wd)>riRB)?1 z=KHy2yo=e@zI5<>PxS?y6o1U{ZRn>3#$Q}7KU)v8sq&~1dCZenMYvXNFW=5IwoDA< z%NKSBkkZPe?jzMx(5VD#{hg}VRBGS04H>L19)*_w2-)42%FlZ2?ns^@i@z0&Blw-X z;Hw)7l;oU}(}#>zYOfb>4y@VaV9q8c_aN#XSBC>hLuI_VW#MUvehBttcJ>HuKXK^I z)h7*idx*UHuR2ljwzCouJbsg3%w=^MCq;&C-9$PTOFvs-0*ke_$+J2Q?7;bRr_`^s zut)FTpy%$I(O(*q<8JEK@Px8MO=;)43(>fQ;l3mx(sz5aJ`tLoeP2HZWDaEGSB0}? z`pJ~3+vD@jtysqR8NS}GU&L#&=qWPLt6EM`w;Oc8|KxHrZZt>IaF}`j>68%m_?FLD zh*gWjg~g8GdgQj1=i^{-m9H?Q@m@B;Pt_?dF4?yOWow;Y`w&U;TS`8Rr`tCn5RhN0 zC8r7Y6DaxGN(Oo_xJz-1TXpO1$9KWC+mpB~3`ohO>wV+BMTQ*oePg!{E95>?0-cA8 z^or1Nt4~1NLOb%ZX|bd4KwRFoO+AtAp!)iNMI)Tc><#EotXzK7J8b`XyYoUNr~4UY zZc%;JxS}GD^qiB?=Y+zt?nx5jcM1Q!UAP4`Xfw4PBDie?J0`e#X<=SJV)hCC$k5SV z5*3_|L4T?E1dd`uztTbV^5&yR3l|HrlNOQUj;@&rbR~M^EF^#!)eJciW3e%H;ln2| zNjO+&8abFYz_`tH&BUWGS+w$dW3Snc<=Fzqy>QmyP(wpQgIX97%3FsQbnjr*$z?m+ z`magF780qLyKB371i!9U25Tf?eaRy?5f6Xb{vd{aiQ1|MB1Q2D-b`5b^^>KSRwjp?8V8&ugwBG(NRn-{KN%q^N6f)@ejQZB(`|11T(->b7Q9V8915)> zxaH!{u`LH50o@?stk_`~=_ei;-YHw&@4HeqO-rAz4wlQl{^<*rK%R{ezA!!K-Jc7x zn_EBHxbEr0^7;7P>C*p`@p*{YF%tsKPY`MR9?{y_Cja)NyAnS!fSY^s$|77(zq^4n z;n#}ytC!D~Lbz1a!#7*&HLtb*w~GhxU^j&mqm&JFks!4{Ec}tv__79Qy z)@^8VEL4dFSSZZcK+m~1&!G+6pg1UsFv;u-m1L&2PjY_H5e|6k<6=u%>^z&31MDF? zY~P@>WGmlSChI;{^t;1D07`@T0Qq#jb{$D{OPSu&r$tp&q`BfwD2ntUAVjPI?PizE zeoHu*Ilh|y*r>4?(pEAm!{rIJr$+#m5%3}v0#*fp`!;C$V^?hO;P0GWNR~DGgu}?T zCrj1mh1t-yD84==+|D8g%_n$IGn_IEN$X9@JBv?g+pJShQThRA4Y5+5cxUb?LE*9T&m353X+4p_`4$p-ej%T2@d0 z$TR;WTWoPy4RvvKh*R84WQup)X_+S${C()PaY2K?g=a+8)wiV`skUZJgR+BqAZh|N z7uB(Qz;ns!@^@P>d8GwTc3C@GN+nTE^TB>V#QXmt*eUVTw*GhY9z*D?>77!JKiKDt z`w1a(@0)xrs9$+pZK)K@bFuE)72IU+9V*3mcS~TGlr(qQ_2R#Ohprr*lWZ64W8|=w zOKE>lbUnDWzNxgE;Wm&f0k4cH43xBfWo~cJjUx7eagZ9Qw@89+E|%wqaG8D49?2eJ zy$<+&Ulvq7Byjkz$R3R@WbNq$p5-fm6V>@O=tM*otmo>FSg}=dNd6&ML{Ka9v}`;e z0I)qZ#llj}-=g$@HzhA(15)pqS$3!VUe80qLZMM_Qe!8W?O7EWMV_0me%QA{MzB10 z;>ePbenj(T{^L8VkR;!q z)Nzo|+D$I`OgK2aq3+FW>m5Ihr0om^cW8s6k`>n*mcIUC&;6;Z121HypIq2YN%&!v zUoyN|F-2Q=rukcoOMECcfw3`mnf*m}^6x5(5nf*|r<_T$YUQ*l@73cmIm{Z-e%X;n7S?f&@EyjClA+N~ztnqHBK-C|%>`z$Kdu#?(uc z+_(T&gp8Sf>BR^i?(@9#^|_z_GIJf#=ZkjL8oh)3$nro^^wetZW*(^ql|r}(@=hT zSHenK;Xe43?=KPi{LG+;hDk{!x}gAfT3m7!Z^OlTwcGm=uT9#%UrT@8t{+>j@>hT8 z#1^D;PJbSBjmh9KMMvQOi(DdwZ9f^muXjLInZjcn=xt|@}MIl?HI)sy1 zR=UH|xX+!HV(M?-9rPk7Ub^qGazwX~GO+4 zWf22c4i|HT5PXv9N6*rDDS$DxtbUxxVC3?gtn_^ounhXfx&Qjg^&)u0m&kUwy><-e zvvTNHfwb(T;O@-Yn-~W5*H$vp?ylLr&FlVJf2KfpMTTr}UtHm`rQbJ2oOQuB%}>8T z!!1jZwrT{pU00lfZQ00AOHOVjsvy#2_5hGxRk9`I<#OGG>c0kmceBkZ2^~-&M-hRPix1_f zvedqMiVR07d17Ctr2_A3W`Nn;!~;bT0dVP)*lWMJ26oP1nR_41a5)4=M@JQH$Qi5T z)gKBsQXt}^-D=?F?J2HOJIjAVh8?G7CigOS+6J0V>} zg0a-ME)w!fxAPgjvqbPXtEB9Bf>pSOwBz}>-4xhAQwEXwYPdWVO@i%;-`-qAv9c5K z{eZH|$q}LzD0<1{fe?}eG}zq@y|6H=DrKJdvs9+vXad`S`Yf(5Vm zjRn7L&U|x|miMhN_)e;VCNimE((_k*i5%@f!m(Y5lzsHKu^)Q|S9N~s`H}lzSy*Xh zAS2&V?%kL3zUB6jc<*V)SG9cG%;3CZtOpLKUn?$=^H4kOqJgTbKaSvTq79My zw}WN&Zn=Em@0`z>3XA~9x}U#3IMy3{K#^t)xI#X|CTHZk{KUrsZhH1&?x&0E9}JG) zIi8Y~F&r^Mh=SfWbF}5vb{I)Z23#&!1^I*+R@Zh*E9qwf&I`_wc!d$}&caWX+xRx+ zAK#Vg!Q?lAdGw6WNUStkjAWrb=t0N_7^MHy#SemBkvI?;;HCS!sCI}(;h z0xR9miw}de=7SL`+T&|6E1zNX|Ye6|Z^8VAyy({#Fwj#C~&Jr{0Vpx`6DxC{9V*_YZ2sHWN04fDBAN`i zOkWeEIf_^jXj^1(6ZW0Dk#!|7!0spl+kM5^h-Y?^_ijxpQ&1SSIj|8WE%3wH>%V)e zuG>v%0BSi<>2o9Yg4xxRz5rvN#C_lxUmw4MV&J;~jp(U7TgokS*YAR>P9Hves4rqI ziuLgDNX^W&ZQ%L0(z!(ji=vvR_&mUF?8Sw<9$wU9_vA(Vr6IJ~P{N*T)IY<}CBg9h z(dR+SFFWjp`+n1mqWMCvkF_23CmxgR7*CU`K6wYO?^$<*%XDnlJ9I(G-%}D&;}BZp z>6l(0(&YE=6KqowrM}bi;U(L~b*ryCGK>7LdaxOs>YN?Of9C?1bzF6pkVWU=m z2|!~849j?WSg{~BN8Ek67iis4QX(vgA)=7S9hMTEIz=H;9ti&{uay|~kWfq$$0`B& zGZW0%8(KBOH?Cp+gzWLS+1+kqCErSOk6RLo+Le?otL@dJ64_c$SU*Ubs{=A%; zS=ykHyY(eH_Oy01E0b5aMA0?=Y;=g@zjRuy;${r4Wz4kA#)n}Wej}o^bbvNB&>af2 zhI-wxcR=Nir2f#Q^ptlPE4|g9WQS`e<4aCA&gZ9`mD(&$+@+l-J9wV^dxW5eLxGl0 zI#Dv3)LS6Hs&l5=N(iW?=E+jD|5%9$%3z)gb-yJHi&!&9C0jJTm^nVpk!vk2EnON3 z{|8@ct7N@$_8~SIstk1tL}HN)-`zQJL|?0&W&z7%!Nlf4>{1yQ z>AGzX-ud-~a3bKef6}4F!AZkP%Utw8PJXvJh3RA_bT65r%|R>ad(=kqoJLoB;#jNv zcKCIx#yR6IpOhLl-Ji`p@e3#qx#Kwd&qKM^cjtIi6O1i<^ruRy$QlQhCZBE|Smv_A zo#WvkQ5u}1&Ptn;m$1$}Vf|xVh*^f7(3XxY!nW%77n9>SYljD1&>>%`PM)v{~zM-Ui~~9?B3+ynocN*}kV2 zNHlq_^fP>t_$n=RX2u{_xMm_?CH;TfNUaEZS2s@WMry8kCuczqL2t16CEzHQDKdAI zc+Wt<)`c^5e{&Nfdvkq=BV9EV4E>9b@v*dbv7`4^(sHF*K~PH zrc=i^{1R@zR%G56w*(mK&t927y8D}X2c!SO_h;^v(i8j(z`%Gp+igeA3$}LYeZqKu z+0$bxEgsGnH=mzWT8g^hX;+b#-u3p;zFFn%P_v!N$^Ky}GI1%9az(7&(zD>Xl6=ZU z)k}!$vm(e0<7|n1lxG~V6`qwPELnPL+PB%bwBGve3h~v* zwj$JQ)5TT77qKTdW;)JAd&!Jg+?B=y#NbPNhSB%Q@iec6d$O?sGWjk=5k=M#K{4%_xC z4i2H+EWRO|G{YpeJtMezQI65&l| zN$IIFYq!iM@D_Z8#8~2XELjdP7IgNWK$}#N#lxQY<`y{O2eKg(?@3zZ=BDn z8#fgkh?M5W^_mKwX(PsG>9cgCw(&3`%r~#m@MyS`^9ZmbsHmv8czNq?v8u}A2gNSP z$Oepkl0Oun)dwqS%J1U0G5+2?9hObTTr#o^cxacGvmRB}yaOWkGYyS#8lK|~+RX6> zqe?DM6U_By^i~qw@keNnqjKo7vWfMonmsA1et586h?$?-Zi+j9TBn#%9E}$G^mffm zDvT(S!#D7ap`4n(hSvb^(kdTQfAboq4Bm-L!7FAhM*!DasTZK9jCGvCTGmWFV9M=q zbFLSbHvtv$cBKwlN?zv*3*Ui)0RmHbGJF&hLRA^^osD&! zoy)Ort%o9mpo6YHKezh;(IfOXE*$chDjZx?&(E&WCztdT#bi9 zA_ReBPDYHB?Gz@0@nOl=P3ZX!gl(Qg^mY^_<7^__RHue*KwQwI zu8DAhWz>8!xaiz7V~qE#Sce0n$w?5zVF|P_Y~cn~os@NL&FH@+&GUf)`rw1nTOvlq zTe>wpbXb0d&-u$*lN>gHex`k)+-j<<5^W8^TRE*+usi^r6B=QiIaMk6>iAT3Fe%JN zl|LhE5qCDy^)$dEclKdLGcw|qo+h7PU|%L1k>$-S%ygQRaeliaW+AE|bT`}SsKucp zJMjJd!*EO)Zz8X9@QUgp=XT+ zXr<+zIP&^A#0;A_>s&8i=9y3FafyBT(pVXhz`wcT?jP@@TKrV_|Xdb{fm|5@vn?wpdkuR zwvkS#n*m!B5K2Q_jMm?u0^I8v_>^?2q((Umw6#O>@)(V7BaoKPxn2K0OIJ8fr!YO1 z1P16Z-aV!IrFvh5vJ(9+;wtJH{l!NOgcTyaA0kr3NI=;7CBX&r0|la7nCYIvB4Vd) zKOwC8L)G4$YihGb{Qdxh*R5jvQF@YN_CT(YUlQG5hmvSB%@be!87?9zbix zFf&$Jz3}-&Vd>>|+$q?wyqNwae7EC^!84;+EgN`?192}wL$3_{UEN;$x`%U&q`7P)cIflmXWD^X}C;J)_l9L#!alkUtne46bXFGv3;7Pcq_4K0oK z%Z&{JT>jIM zEcK7+#0jTT1`94FuQ7ihyJ-YX`)i785FYho^DtU`4kqL0W-VG` zYXiL9tel)Ui|(jd4R18#&(EoZg@wykleNDTWzpDC7O$uu3X=A|EH`Q(1`!%+`uZ6r zw#b83STu`gNAtfk^}fN?Kxs_ZG4%Rw0oTxi{_}@bR-pzW8k&rd;snJCx|m~xPC|A} z0#5TIoTy+L+JLv&fkbYn*yK*IJbM_m#M<3y!_i>bhT-QETq3@HP#1AUyAgN|vlWw; zc~Eqtw&u1xBLL5;eJb3lL>B)uLeekwHsFqk?=M`R?hWK0on%i1HLs0Hf8|T_`y_PAfv{8gQG>1SUC^x3qL4gkWH7~WA-9Y14jl{0B^+t`db-17>#jDwxfr)9 zdz^YgHLZoWNSyY^+myEg4s_K@mp3*xEC4ZgYK7*Wl~(v9$p^B>R>g*bw{NTdze0<) zbY0CygnfWFS4z*ba3WM!4b#mY)@8E?*8EgEv{|_Xj`KJ>3wV^&7&bP0-X&!T;w;On zn)oEj!2)ksv_ima0wQoaDjF~pF4~Uy=A?r+vTKE?Dcl0g5`QbV35!z2B>+WLP51Zg4Sn~;xLRivZt=o z4s(3U);D#`PBU_PeQ?HNW_*!(%x}C+?iHCU9|)a3x(;g1rwAm5ZFw-ElLYjFqYrAp zB9SCJZ-hTQA>u5%6$Dj=t@zW#v@2lIOUsC)LAKZnWjl0qR`1#d(n3jTel+7G(E*)e z!t@G=!aHCS5^^&egv3XEAljSztl<1QvIB_VmAr6K%g1`B`JIBG{@s(G6YAJs{=ONI zG15tAtH;KD9TDiT8gbft^uO?jvboqJATp(i02ACCMtetTYU?IS@c!Zt`uIe5gL%$l zEeQ`h3k-xsbQ9Mq#Y0zi*X(uVx=FO+uNG0rM`}@76bU!@=34SsX%Ps~()_cxI3E}4 z(t1qI{9=nu7ND@cKrD6$s7Qw01!k(o#>NRm4C3Zzzc+k=v91lieVW(C?LQY0r4y}i;RW_e8` zA4liz9(6zpqOc~5-L_3@X<%rRmScwVn#w9HT#|MBoGW%nNk_bT? zj!~|xloahguKpqObu%5ER6gd-bqS|AYdA>_m*2rk=UrCnf5&oEmA%LZKnzhjTvGRa z;U+MXMhkE@9b-brMSN>($ycmm6R+h$ExL^nt`FiGEOA+dQm=)kuK-KCFWJWg^Wr3( zThQ3vd*J|NbcZwkWBVnS%0MEbaK{<#4 zE6Byl2J=n~hTBROhgIFg-QNdZIrUV1jQr17p_18?OuGap;}RB8omZ^tr^Dht;2^|$ z8Qt7@S7&L2WyOJKaUGtBj+3_`jVJuLQK*W+;zlHtjV3pH>kLNi({koB-YTQ2EGxP} zAK;0|pK76T6t{|&&go>%cIN|QT7zL!l|p$iSx~K*^S*7-$Wf&j;wD~-WI0KW@zpCF z|AxlOfPjYrh?Z|^)YG7y`c1BP6NLs2#a65^SZE*(o%6^S?o9LtIlr5~ z+3v_>g&st_kR~13V`{!I=eia>ba#Mx_6}jfDWcBdYTjhwtQJ3@`UB3Q(e*Y%A?ZA! zj2__0ryk_u0u1>qVGzkje~#ka;@n$I=4MYIIsxjI!NHxYh|Ed6KszpL;Q`=25(-kE z!C)-q4xhw(0{w6H7f}a>OAflB-fNJh%xsZy(KxdTk@AcH+RR{TO3A>q$O)9$4MusY z^~=H0x0Tzb>Kb|$nb2O+BNuaA6>pWaVv07E$r_c0Y{!Mhv7Bp=Kuv*)7RC=tp!+#h zk()?mp#ou7w!b|~zF^bp z2F&@_fUV2tMLJbvS2Bx&M3975j`PHOLzipj{g93XjB3dJ?^_+Q&qaH9{qsQduY=r%~MzCCZe;Od3vm?ntL ztfo3TO7nHr^3feEx(5#)3`_MqVE86sw}O@+a-pge7SG}=`dRFvo+e3)FuVKNm)zfK!?FvKEbp@Iicrx1v zOa@86k?r6~(L3k8pMje_=XWi?4DqZ(^Glvgu@4B{2J+Dq*}GQbPY6?n+aAT48q z|8rOs4n4LR$&Z)7jC4TY)n#*!^EL`tN)BgZ8j|Ixyei1%n-cniJ$0>}W*?<(-7vjF ze~T?b5-sEQIBP#L;agcgY42XBBd;1+!eMuL4j=a+Gi$^asi3nYI4^JC?x7YN69v*N zLAfuF=-4hTVxl?#PUcO{%gZPO^U4-(*FQ<2d& zB?jxTI-TclMKNpc5El7K%|t%JQ5>6Ne0e{`l1WBK`#<*hZeyx6t1U+(sQ8L7_GH;{ zeTeSD`igXU#@kUJkS)57iqYOiIYpqnV#-PHMOR9%{D)o^!>^$g2`EXK@M1+lPCT$k zF1!WErdWb5GPDWI&fe-B^&ga|lsuiEKc8trGiVq|)XL36%XqU-7?zN4PH~KmnFpCt z!-k`BMDy>>q((ejk!U+4ntvN%GCIq)7EU7H0nD|m38D;D5FC*eMh$DgIJ(p0tDG%S z*wK%#>Ofxf4Ko?oNf-wl_K4d=pu}!}YlyGl)w_?RT#GoAB<*gl_k!4-C@%8$r}m{Jl-W8E`stK8{RLf~m*wkJWfbAW#EI0ZipcDU zQ*X)R+@p7d#U&;%WJ+>G*s0Urg<7qQzw7w;gx}J4-_(bNLZ40mk+q6P5hApCP*Q5a zCd~<~g&HC|H0gjSHc5;~cuyW+J3?I%nF&m=OTnV3Y~k_l0OSV(;SUkq?COke5ti@& zvG{MLxC z@Cnh;(TPb;UY>s9dj97pN`Jdc>$~9AH|qu%3G*7>L|{-sPdWUuhvar7ex0Q0?FCj^G#rGFu1uY!YMV`4(fd!3byhL(1d-ZaE|GZPZ6O%7*|9rs zvDpoUaRLA{o?&w`>s$2>v!o-fV1T6H4C5SE$e}lzN*}lqV3(IAXc5Lo3VYs)N9Q?&elc~F~G2)^ccX= z2`K^EMjm=%MoH3vfq^CVxumreh+B1=(2|&SI;;mtvtQ^*?e7tO zX8QZk5wrFpaKMbyO6az5D#cChbLAW?VZg8KS5a&%c&vY0uN-un%*7Fg74q4S?``^A z3<2Q=*sfN;IDYQXY$Xek*&(a!9tnws?e8tM!V`&p&epzXGKT$tdNU=QJi{LK?Qcvq zfqPp3E6S;Y2EQX1yj6PkX9`LHr2}e$Wc04sq(ncW+aZIpw>Yqu2YA`=;AY#6a?~nY!5=kUxsP ztulP0n9A8%Ue3!Pwa^|M{C{MfRaDkn7q98=?v{}5lrEJ<2?6PDX=z@%QIPHs1WD;` zR8qP-q&uX}dfj_}=ZtY~9A0?ipT&wf=kIy+Fz-Qzj=Fx~9l9Uzmag|pJaDT3s#!B; z8hkcDlFZjSf%4Q6(ceu)HaSGXRa+Zovr+Q7fSS50Eo!M0%Vm$oPQ;l5_yq-+h{!WD zOEmpz1f7US^j`xFFPV8E@t7lCHS>~dVYx~zBeDZcm{FQo7L5eV-B;D$eAzJf^L2*%2x|Xw%JJ3rYXruN}`<^Ep z=>K!rHgSQN9q3P`t>@83>cTcH%E?o}y!>I3aZI0}fP&k1ou>^tEv8}QWkXR>@b51X z{#*eZ2E2T%D<#6mkHQs~RBV8haQ%i9pBE0Smj&FX=D`tkV8s-VAQnZdX)l0=IS((d zJ@A#grk>)yNdJ$5rGn{Z);`cOfzgM_o5093%!`7>oqQiaLiiRHH+EE{*1Mkd%kU4l z>Y@G=xhb}~PsVEB1cyFsvHYoWp=rKX?^(Yfdw4pE!HfWS@F5cJzt;swm*-P{Y!pf3 ze%NXS6vGAg+ja1XLR;AZTJ8Z8^PUn~qLl9dP%1bmY#>H0*SJ!<;;q)clB#O`>tA3X zRls5-%U%>sW5%rK1jMFWzkb8&>?#R<1t+IKtRVHehG+;dL5n^WRHWc^X@20O0LtSOk(bAgLmX4{6FNd_9__wD4>My zYS~``dbj)wY-{uJs?n4x?o}5|m+%+Q0-0Sv&U+R6%=^;Q_P@z@;=Jn7@z`l$Wm0~X zR}n8@@Q^`DRj3B6CbsWC;Ys*@^dyZ)Ab;XOtaIR}wmlmwQdbMVEC8i*vTGk1%Kluk zKR)?6j-%@rQ-Hl4l*%ZP2FIkF77X;gZ(N{^DSJfTlWZB^%C*@eNg?we1kKMJAcVi} zXaxiWh#<(=b?Q%BB*ev+A`8A+`<Dp7Uf89=kXQx0h2LCso2kY zFk6E;dPma>BrZN91-!gT=mbPUb1)rMjPw4d{xJ%xFY5w6-MOlI?jTf@^(m+V@sZ;# z*m`GGqV5Fm+G}gWqoQgaiyj@noi9ImL)E4ocq^-Sctz)fy2Ny@OSFLOT87i%jc36iyZV&(Q#xs$ZaKRkQ}ot-WP~7-(_+B&D;Etg>a)+i1W8j^sgeetr*oNl zjtRl~t0|%=YPIRgL`lat>yDlX$@fizft0Y6taG6q(+#{SbtP`x+>qz=eBcn&643GV z=DMB%cWNSnfA=BTt`_l8KQHZ@Ans9T8L_akqCzetqOj`H)WG6aDDfR;RB(?}N0A-=?5{xylpb z)@o?raooBszC!)+?mjqsU2V_S^Gl>O`ToAAT{Zui0_M}w zA28z`>tSC@$!%H^j?eU9K`#9Ed{tMZ!CmbqUuvz5s@I25+B zaaQonuP&HBzE4Pay)#{bMz7Gs?PNr_w!Y`^lKxJ9hx<<_!yKXMf}unK<|t=p-8!TChawrhOGf~ zh0Cxaa9KiVzdvw1xVOx3Mgsd)3^+EY`mXUFxsf4{Z1s)ZKGL8X@CPqg@bGZm2sbb9cwhY}!N4}5<A>rG07r%@K!Q-C!T>Ju~c9!ts zOouFZq3%Z2>00ykdtae%%^{o%oK_ZRBm7>`!0j|5?3N+KcQ-;YGIAY5w-f*a^LYMT zR_YaL{uc1mdl&<(DD;YauyW<5unF{e-`)_9hz<*HQ)B>~mJk4|rGMRI*QxBVfj}p{ zi8%DbHC2E#9SF9#fDhRbbByHLZ^8F!u>nwXxooG{7QkK#Jq$f6aK6H~OK6qp7`xqr z-QNV%VJ;Um!|Q@E2|jTeI@IDlFm^3}Njsa5#bJltUKQhb*G?hsBLr0xT#(Yv;eo;i z!{9=55n3@D4BB)WWZ-Fzt0+``+Y65h0J~~Aw?f4H`$PO-f?s?1;?wSDQ2M}?@@=8| zEBh|I2*IEfnHWAxiD#Demb7=h*qxlIHb;jtGXaW1uz6OSW{Ypaj##HA^nvQ|3XC6r z#vZ(Z;WCB;LGdP3cei4%MS#1!JO!7fUT6X2UdsM?E2GqdQKkL704sqDp-S=f#`XF- zcvy$s90Ben=42DkIn241u%+>rS$0tK=p&lYH`|6Wn;~zRd~p6H^T)iIWmip za&=Hd_>$K)rR}NE!ME{aaIqdEdnUV{0HjFB|5MNccS{#6SQs=y#1$M!Kipw&el8Xd z2!4Ei<~K$&5WZ^0I}&+>F3aSFZ6gZQL76e&E1%L20$L?igA?sw8x}!&93N&7n0ukI za`zs6F$wopcB~WC_9KnzTk9pgCf5?eNUxl&9H)BE6O+MA(P@sef*@ppKgz7T~yQ4YFjnas;v zz3mIPc|hsrr0KJdD#wh=U0|~MQ#h7FxqeP>Fdf?O=$^6>!OCXh6|DdBO1ek2&5*T|~qLi(xABW)z3z zEUT@SDvSH41%*ifF~<{4p=)tjN@k09{%9}W#yuhvj16SDLY-&IRP>%@qj9_J5XXj% zz3S(S~1Fd&o%YHd)-WOA^0}QbS$#=umMogc5&s zCI&F~cW19bZbmG%Ydt|x@*Mp22a5kFL5u>U)EL!3FijUMu(Sw46+p=6_X`Qdh+l^8 zBKjoVaoKcG$@wV=kr}WMG??I@L-g^K(FxK73BfyeIS_pkD07idIxmptBmQCbSiY7R*%laLi z_Jl&K0qO=+%pmA90Ls~8H9A?ggRY(it1&t#i@)w8O@5&6Ghd+AFovp&WK3LC+phl{ z{~V{j{o94M__}Ue!MioA1cImc zL_Vwg*gFe<@BG}gxgDH^$aI4@CNZg~pX}shPvm72H)+0@2-1DIqN|=-P=RI2m9W7_ z&S!LHwpqW0D#`c>RQBiHc8*Awy|c1`){uJwJR_^N^Y~!gijLy4s0A0Ea9uToHegxIibr&zs&Wwkj(e2N)q#|VjDWyW1y=>aqFEZZtulz1UHoT7IQdfs6*S|O> zxq+bB!o`*hc13e>vYG={{x5TDGu zgNQz;c0jZ}{q?_91pFVa&Mcn3*yeMc=T}up}0O$iPiPcK5{b7_USAiX=K-oe?WCD|S z;^gS{y%%04gZ2f5g|jHV-6mD9|wD937Zh2!&xD(Ot;hSVAwV z7`6Ee6l=R5DAXg3ocMe~+p9idX%tb0B_+@4t;v8F>{5~u0T&B;bfH2wo}8m?ER@=f z(?yGYF8nV%E>G~nl@al7(|a#@C%!W-{$Af&EdxG zcoYJIzw4#row!h^_IB5ijR7oi%C0XQTR!ag10AIVGvD=z0@K5QtR^+4pMElA4+%T% zQRr?D%pQ`gH*S$Fc&^82-Q_#=st8g34o=1GPvH5ZRb()lBm=->hRD}ak}tJh*M_^_Um5B8xCIhg+qC-@~Ch?HT^Ki zR(UJS+eb%djY0>OpY%6c;6>-KTAO#k;w8w`s&L08N=@zHa6z#d)>C%$%4yG-zu7Il zpR9=fkz*Y%O*90;d$1oFVRGwl=}6#Cjd4N513F5$~`%fBlha}1&uKw>igmU^;Qkw zRPe3Ql&G*=$OYoA`gi3Xn$kf3oH zDJB>IxFYO;_mZ0@Ir($OJ2XbsEIUz>s_G0zp|=v3Gp5b}6g%JPM$X&7QLnI0`!2tB;}eY|s{|I$`ER;Ssn=F^#P<@cr;%?l8<~83ZnN z^mkl9c?X`hE9DL6`p0Ck)>tJj-p(w^mQ++mOgG3Fj{040^Tn{UTLw<2Rrb!cnoizI zV)C}gExSI7&x0_RV-~punGL7J?ckz!fO;k7SlqY;&k1|2~@i^*>_rD=0zrThdFm+F7J*_eDI~{ zj6>zWCK1n~L@W7R6}1WWyqLfBuf^YXlnMWKk-7$(0&SWiehiw#dv=E#+YGRYTUL1` zrsB&f`ry(nf6rE{gXF@66PSrm@>yeQhsG^04_*DJ9<9_(xArAI^^gx! zhEq_h7=YE;g-P|nYId0Jxu@M)MXvi&=nHy6;k0f@qS&Y0`xFCHN1%jdSL3;AG~>P` zh;;hF0Ph}U6H*0c5}+n)Phb@&(~S@P4th?5?g|tGhRT2Ivu{Z6 z*hNaF_Q3&uwE*uU#JB#7Ze(W29e<98d|$slTn230Yd~bWd1th(ay4`Mj+lGUDmnt+ z09mRJJ-wtECof9UK0c(glZC!z_!Vwn(uofQ=v63%4;$9g5_!5VKvSW#qHV5^%kcg0 z!YPJ4J`Z;09x1stql=YRk1T5{S+| zuY5+Q9+Yi*3^<55l=ze*a;Bqxzvfb=>+9w)Igs&i%`JQu_3t1t5R!E|+6Id0`3&bh zWCvQ=aTbv5z26!^Uh={BeDGQU;KW83-AW|3HycPpzSMpwE$4X03oadcj*YjZZk;Lt zxkfP*tZ!M2)JNCayLU{q+CC*Vc&<8ZRZi7c>`_=1olddM%v1c@n8nT8TS-TdDU4Rb zHv6IY4_|1hA2)-s3pYh_jnl=+mkjCVXcWK_H!hX_*K^S7Bttk0_+T&T{Hbve>FH(D ze25kDzf=-EOa0n&$E8*}f;*Pjkm+xDV_XB9J67OY$la}EVw_Nmrz6{d_ioy5N5LWu zNq6=(^T-Fe#&V=iUOH`~i}Jx>a&l4tudtw?r!O0aox;4N!2TM!#r4fPIfk z5kA+-gV_!tijqLm27;L_9Sob1R!x>v^o;Bw%Xf+t(c|b+O9-Xez43b0G7EOeciN4% z0GBIAepq^4Sxyc?z&^6)S0OIo9!%#q$f>Gux}ca|by$xFtafJ#S5oe-1mhe;gO*MD zO$vcDgj!Qphi30{bR~02NRJZ63BGI)zC2u*zrPWv!AE+hLgJc`{~6As6?rm|__Gea zm%w-m4p4iskb0dKTdGn($wnwJkwrr+CgIGHoA9`4~|D;CXN_NbaQ+CyO=Hm z)f&&PSR;AyQAb!wlcIThjHEw2!6{5$g6sj%r7J-TtPEfoNl8fso;J+BBb2Y#c*sK~Fo9vmF5t z3a^j+jkl68ZF@8NYTmh>Spdaz0Jb;`5vThitnTD)_l3}Q#h1v+qOPHI`az0WH*BRB za2lF{f$)K#=Y6tnfM}N|T!2d3M5T0}gUSV3-F=d~M$p}4X(2Hoa4Tng>VH-1u_84? z4uBIehwI&{c1>?=>($>%xs3$E3*WMt=2<-(%EQHDw4%XQ*mi?k3L;`VS@p>WW-AW= z7R$m`#}3it_1=t|rDqp1^eag(sXG9+ZgQ5We&~R&cgJJ#RG~7tUBUCtWDqf&>kdsS4)6&Ui9?OSZzl@D-r>dLG@-_V!=|1w(b1f|GeNJW*#R{%s% zwfO;RoPy#Oud^R-sQ-40CoBBS9UpW3-Z$zC3xdrdxt(p#Ur+)YDhq}rYv8>`(JgUAKc z*6k*3bx1)1o_`E*Uf17b9SV%>!^yM#70;+pvkbjfgG>)$JY$ywTOsj3BNhOWIY*ZB zu40c4aDQR~`7l0!c4l2zqmt9>%)Ga7G4AY`n7)Eveah4yJ^|k)HJ34S?RoB@jUiBde}5EE)yCZ4tnKP1#DBkrzLFoO@8jfR zd?m%3)SD;A@j?h^__?X6EhvX$GOCpI4kVPD(r|uOK?HdG4Nh2sQeV0jaz+hXhaah0mJqj*<`TVidd(?*RZd2skTpw zch!UVv9}Kkt&!W$4$6nrqjJ0lp7GNLwi?(Mg-W(6vi?5`K9wpbj$4|cD6uXwC; zoPh5fx`;r^t&~p__De@rAy@7_9H3wufmS=P!Ehc!VL*VUTDK*Vc5L3Mo*~WnMEd=JO_TQJKK`mehO$$Vb8ve&3)0^$x zkv4Ll|F-j7VQjtxoIZQI-EpyH=E190*PRMi`f~_vi-Nx^V_Z(Q@dC(PLX6nShJ2$j zilizySC6qS-YyP_F~@0@kpMg<@x| zcG4w!ENJ95be~3%clq04_W=su<;^j7G%S{!kqHL1|8mdK>?R;xcB(35*eCmD&72yy z`l6(WcMpe%3&si|(|`QcYMb}9xD0`WXf9rbmHo9V0i6xDx_zlYb|`4s`+)HfXlDX? zLFpGuyCXT$vla()g8c5+MG|k3$Y5>=JRd>#^=rgo5IRQ(c>RY4VjMb1T{~97)6o0h zIG!}#jPz3ba6HiP5%u-;(Q$BKfwa3j^mMp2f!w##A9DvD3pmh&wy$`Wzo8lHCV$m9 z?RpHfG?U@AmD&J^P~2(l=i!whx5)IR+F#$8Uww?gXPUF|8LM684qbdXd*j*@j(h3C z263JaUDgKEwhHp{Nbk>gjN=ITvOwi;q~5d4=<^0BS(}1s->>emV}&a>6zf!0IA#T0 zLqTPiWgO01ojPZ;FR2{*xM;pdUoS1e&r=f}9%q;U2?q(E6;59ac^JZ{8aFz+x5CqU zU#PXrkz}Mo7XQHIN}g@*OLPswwdl>os`(?cK4&4QWyol@oBM`pJ5?s-0}1W=Z^chzP-^Qh`m!?0XEqHh}e+IO1^i4t!1duB2ajm!rXy8qPW<~_Nf^@ zSEjrBGfxiaw0`L}C`K4}1*1mB*dwM$12&!D3UuEe5?qefb8LNG;X_Ly!H06sClTv%MFp&1jYVVrG64R>+B zYJu0sNcZbXPA~sxn=7Yo?)M-fYv6*MKB)x zQuc>J>Z;)}yv514#KLi(-?!LuP+fxzzvr|*apx`L+qNU!YL}Jbs|u;FmY8NBe*xzD zS1MN8y|(T|uSiKs%sJD1D-0Z-OShKuX^D$%k{!L-hKOea-J|LM<}+i&v(%31e@-N? znIt#mWn1*-Ry(aqm;T?0Of%?2=7$!SZ539Bwu>pfI>-vtstNiuzp`ANMn^lS3$23Y zn%qk7bDWPwVnQB{kG3OtV;t^+a%l4kSFyMG?y7*|K71c*fHMzxD`%*uZh#odE3@9k z>DfbkU;i;fmZ46|u-jcfq6DMU@$!nh_3n!4#rj_`oh*roTsj@E%mNbm6MZ|zK7j}0 zdR`RyJ~-6h7K}m!iewl)SBQC{$*`&@V{NZqsr77QXOxT777h#{PL(6_q@=V>N9a)m zw_~e)+&p@HPO#^}N-OJ+Y=rY0i3vbYnw=u*!2vWf=KnG3UBzr{19;qV z3Wi%|pG-_>!GTFC(C-8x9MpHZeCYB=R8L17ZzuG=`~+|n?a!vm&aI3qpXrBB$|pA; z1_fISN0pRF_4{EIWsHq1rKf4f@OR-*_MHYUj?y=pa6{?a-<s;~QV zb>v9d{N>9(P&s>~X2efhFgHH~77!)X8PyDS5h&#)dp^THV@&17R3zOrp~*AD+kfg* zJ#3?{vI1ecAdv99#bl-Ws@<}(P7hCokEWG^Px73(kos^e1!Y9FYQzD*70nt z2ol>Ctls_Db3b_*C+hlfCHaPQ>6f1s7Dm$LI}p*3?!)37u9y#T?0tLYC)lx50P%<` zkYcZFUY3K8?FCEa{d5>(DY10!;CwMS<5lBS-4(0a}#H9Riv673!N=? zpEj-K#Dwe@3xWqbg~x*PUxq)JmzRNtX-&Q(*!S3vc3yprlB`>FAo}|L3%d)-4E&VYh<4ui(AUqjLT&73* z{d~|PWu(}hAaiJoQ~c8*$?Lhli=*tD`KANcUq0xUY_+^)-TN+1zpGELSo}ljE0cBT`yKrOB=MhXk;F@mL z3EOet$n7hdICnQdl*>n*!qtPbN$4C@q=Zrez8xYj)i>mnZ+RA5aN4wAS?o{C&T=P= zwXaT@W+Mrb{U*D;I=x^2!?uC7b&d1>C=cfiNABBf%};vpnKCWDSIgth@f5v9b< zcU8O1Bk-ItrSQ96t&5@&MNi=Qy!G;p-m|+s!^gSum)$fmA;FobOiC#MM}Ak$ zLhI4Ib{_w7blzaPW3vszp$%?uAF;YLk^zBr4@D&3*mucOuAFIH$z5YfeKH(3qu1#> z8KXUyi@pBB9Td|ze8c<`h?R|M2)Wir`SPwP=U%NwEPlH80(FxFp7lU&O~$=lRaW1hu{YW>;o346yWqv+4&^5lUOTm~4iWzWf##R`QdFALB?+=gZ#M zLGWp@Oybw8I@2?Hj}q19b=!F^y<3*M;}K7?KK32i9R_1c#PWjc$PpgFLk?pD5KGGd z$Jk;2$f#*Zz0oYKGe;M_3Q2C&ao3(8+r^{wH-@bhQEF|a{~%jt0`4%QIEr( zwce<&I66?@1XA@&z&Or|!dDd=!vf}H6h*1HE~;kuET=A1As{6CntKeCO;vl!nIeP$ zSfr&>kn034egf}woV))%4D%*eqQ6EqX3eTJtTOX{+)aByBhpI>hd)%1y$BDc+&8Tr zMkyWmz4K}tjBoO1$fXI4HhRF0)tMxtf;BrCJBo32l0~>w`zbJQWJF6G_t&@n$Y0ir zVe_4vM%A|H$IZX$TWn(>i+5z`+2i-Aj#I3`6fevfO1a@h(RhOBCYG;15PD0?&@z*> zd0e#mA0U42cs7_lRl|g{9OP%lyC`sGmm=IGT<~PPY<v&0Cce?$%_P2wHSc%+DRZp&7wW;i-umJQq5~8 zXRaZWmdp}PeM=b>UW2!C3H7g%QCT3Zg?%#JZ=q;Xw*U*$;vZY+RLhY=t9TCM9sO6C~IYBYzYw@$~C%`_Pn-^Sig=Dk6 zOzIj`|Bbp$o-=gR6|$ZQK_#@lGk-t=*=pYPP(Wd8o)Y&D0Pc7NR4_O|v7#|Fmr zw|DQ4TCdxKC04trn7nQGKeMzWBeXyIWZ%4Tj#g2*I>gPpfQH$}sG%jOhkVs+ap20O zqZ$f=zE74ul#*Z-8MgZgZ6<0ut%eZ4fu~)MWMSs{v;nHY(ky53fA{xbEA}r#1rL}z=A0&77;u39ZAq0Ons-T_c6TQ@G$?tI5Z zEq>nCQ$7vqlEVsn*dyAu{EMg)Z2^EMg9juG7^D-NiN)qBL`7BPSTVd5@kIb$f1NC> zhT>ezh2^=_;sx?45%kW^&RW~1O(>f3KoTHgOAs9yS&hn|gl+t80OU#Rk1T%-Tud7} zVnQH;ffZ8}DIkqh*5)~Dp1caJ5yN-Kv^7+c@JzW(o6N zS~?KzMPVtSJ{uQ&1}FfY1~7RKYH&F`nP`d5EZ+`qV1@7PR~7h zzElzQb={5I=V_kcc7F2eL`d1&a=0Czb7Gvds<6@0*UJ93%6eM&*=@_Y?xhH)*S)ig zH}iwX{m&&l0CG1}9P(NMz~x_lL10z!AFG9Sa6*>g#kVpxN|yLnk}}b;uh|1A;oGb3 z-zd#K>o00ot1s*m@wSt=bLrz3${%rc9J;@Ic1Gz~Vth^EW02j>eO$YSQlIBOCJZJ_ zD~pQr`6`@9*9Z%3Tl5M77%9uG&M{2OjnkoQH3% z9@cK(nKBuo@XYIDHPf)!q<8}aCRL>OSMG~w);Ji&zkyBCV)b)0 zG(bc20D>h>hCc?N=GkUgENbxun-C@RTn8cRUj+aP^EadOmF zG>qe>LS%3|TCk%)uDNW+?Mky-rSv9O_e19K zTu_|!P8opIw~bJcU^s5ycOp1izS(`49%O2;r7?_>G-Wn~7#WDq;3yghHH>d?mzcIb z5fFu6TD+fxjl8K`Lsgs?hE-RWtap&i7MJutQ!i5I;vTy*xVNON5JE)*srDFW^ zA*`H*AW+x1)jLB(gIBv+5KqS!EvU#jmV z=DmM9%oW+c$1cI4k(S0SKt!;9>Mu8@N>uMkC}AmpvabI0_h z`m45Ei4ox|Mvz!ZUmUvh^>gB>AK4$w=e>{uXsp^l#DLqvsd=@ zyY5NN_DU+9rcV0dsd*jGIs%uv+&elZyFxlzybe>u2>_X_r%L11t6&fvp%oOIK7S1# zx66YU2|O`+Qmf|42B9EoHg`ZutPd(pTIKpB2>;SeWsd@$Zz)N~QLQ{h!b(|xv&%=b z6rEtk?W066MZqFVh+-Pup8BV;8;dyui(m(_i^QDyc5Zm($LeQK&DfXXEtoE!7Q$0s zmOC=&m_~nyIhhl!emr2e&nSC%r54g*nBJbCB3LF#tm77oy1+h5<$*vT8v13af!XwZ z)Y|iM?9l#`-jBg1-;IJwk^FEN_6sDPRMZB=g%a*2GbWcNa8)R|E{^vKQI7LJ6mjs? z)^>z_uUN?stT3|uWa}u~8WeWb`fLA|f=%8oV#{P5AYTvgPMe>&r&4Cb2 z(h$ZM(jH@XOIgiZoiYCvn#KB~3v%{Qh_PgVy}x-|eSr%KRzN(0OYEPKWqtG~1?U9x zwLRYo=;`8&Quyb|uF0fvpK=i>=74$KHlbY%$FJ)@aHQodg%?t$U8X%X6@>f;#2mj^ znuJaR+|SA&uK{7lC65=6)8%Z({%mC(iTE!GbXpO(rVy@B1=?lc{VDr7P-)y<*2-nv z68F5XCk_*-Il8ioW<8Jqt@;OzeFTe`513v)=|rZ^0bRTKe{9VtwC{P0w;mO@Uw!N- zIR9Nq)a_cD>!R!#U7u-%efF!5k_hfkj04li&||K0-X*1)(2$S=^Jl!%wNAn%0xk*r zAi>G^%vyHJhwO$5P8Tgk-ez+{57txK2j_y*Vxv}z_L>!^MP%9w`7oz_&I*9rDsmGk zzt*DtE={XvPZ90Q%hr;<<1_0Bbo<(#BG~U8>me`Dw!QI446(}DgqYkYRH%jWzbC+l z1a`WQY}`q?w6Z7^2k)}8{5eJ9vCsjZ(V;%D zagNs_Hq0`6G6e6S?7-md@N|6%=;WY}CImm~;7Cj!*i<+fG^ECR@Ny~B872PD<}OxaxdvDiOm<2f2qynW-}DOSdRdaZaBpMn|7*1q>D54z_0! zU)v7nH5QHp@on$m`Lu+Pj<=N1JCQ%gmDQ0@Y_~e?oMbg{jq9zU5FjBPbszNiV2e{& zn$H-&{*EO$4$H?>Af|OK;_@zFu{n@(Yx8h+ean3K>6}cnUqZ_7pD zaj?(I&ma&CGM-z(#N)V5DG42tnSbJH*dc-*d$xgdXuN1F0&x~A*{Y) z@An!FnK1NOP}mkx3BK|Y=aUEW@ET}ZX(5OwlcN6G3QWx)K(GFODdow9B4fN#;`oRo zF?-&CIfY>9<4)MA`=MW2yX|2MPJz<+;-~4T23^YMOMjA>iF6_(No8Ng2a4OV~6Ee|`85d3jk|#unkX6{! z^t_)W)C?51><7;)v7R{4Uk$ZP&A~nHU8iGdOBR@-AqM;ToG0sLiBbGCdv`X%=s%YZ zUzaGL0yBa^E0i%h++M?U8qrzPT0$>mYzd!L$GuFyEU5MRe7U=^B!<1{em4sE6Z=Dn zEgC$lsoPMvM0cvxP!ye4`eHhECxQDJq_knv1%#_|q?z)*etXG40)h^#kLoCjfiz#G zgAM{oDGFk^YJ#s0<}KlUz7e8^c6J5@!9n}bk*KrgOR?eh&LSTC!`Qm=!TWaP4w=q~*2&VfQt8cnyU+ zY`!EqF7@HKUfbsP?VveF8q zr&us@dx2}p!$XX(25+437p0h`3~0*aNh0B;@~S2E3)}kCW<0(HTWKMkv(g~F>z+d< z7w%jTJ2@qaW`HHAzKCfl4cGX3$d47D02_11)XBkzHi zV189Hg<<4c{Ns0su>Feb?EnrEySIx=H&+quPY!f5(Jk7~)Qg0BDEv|TS`Y)g)q;Y^hsu!TWxw%;)Qpj$3r}@J4Ld(HU^;#@IB<>Kark*~!1i;!6uOg>MjGb+`>y?4@0tp0 z(ZQV}r%d#63f7Ax$wE>mqB6MCPr?Z_A%UCCLuVQco}(=fMbBBK5zZFf{4sT%kD~I2 zzolkV96JiHQ?VI(HrsDwW$oFI2@hiZ2xTB(LPDku(p$WFl)gNlKglk(J+}GfCgizB z?ff`s+e+^${t++COEX6J%}Q}C*A5MSjXE9a-)yJ|xQtbmT2ZP7rQD?*SG4&tC!!YM+D=aynbLXP^A zfP-zU%FoYkBivo?=a6wP2j!IK#!#N@poczR>14C*{NO34KJHedX$8CBPSWmU zwEc4_BkE#F=n&4a+*hq~hLY{O!JcwR-&e;k`Y)p#$XYNqDbCIxu+x1ws9$Ek)8X=3 zZ|DWrtEC@$%RQN4*&aFk?|gkp%6v*r!_8|{Wu+03u#rA|hMx4lFQ94-)#65ra| zG69Myfzj=53?noIXvPu@YZDL=nSgGzCp6{zaCg5sGgf2JsUioJ^e-@5ZAVYxw^}d>d^X@%a*KKL*$-U??D`KH4P$59 zjY5R8NGqpnH5mpv&_nzQ;K4}_&sQj_?g4n%|VU%+5}AE45(}e)h9@yk&1ISKIT-P;OM}u?<2xWWw~_q+V3t_1!>_1BcQBua-ULK@h&?Nl^?>97qVwJ;5_8Y#+;4m^WGh z=$|TJI(9Kh;~j^XNR(%TXfPmx8`b!Oesu_#x20|e$OR{;OBPgLx<>pTT)==rd*@Hi zb~}}84N&bvfO7uVq?co5|L#lY2ts*WUxzXDZHE*KLv@oTE!wb{x$_e<<6!2g1@iHE zgHS`I61pZW?hOMLRa0<(U=>-X-VarIW{B{blwjAHSv?utXPd_)`07&EA7Y!?DZ07sYs}YES$y=(*DOY1^ADA+oRkkNVQ761k{D7R%n}KJs6D{l#ivs@yjX>J< zN%TCoxMDIc+9UI6X}LqjdFO=^okd1G2i@%KEUkfj1NFx+kp8w`k5YQE)kc8+-}3w1 zspIkRf_VDUz~xfxU~QlWbm@4{+mEqbNKXxRo+ zwv_p|z@?;ze5gmhy>TlrKBz5(bl7HUTXU!f&1+v$oy@e-!%Sj*2-pyFnLIVWhSBwO zG;l212FW=vTsvZ$hb)j1b>Fh4$IkmoKQVsXUU`@9NsY#_{l_Tn<(378I7_v&dHV{9 zXR$A&H_gW|Xe6*>7ZTkoFUw~%Q;DRX#s4`%>`YA=wk`WUM{qhZ%%eSab4&PIED+E$ zUms6>Ee`@iJ#d^YM&eQQpN*sT=Pkv<9%161A+_8K#AhA^2TKR%8PIhmCK9kdn34-Q zg$0GjFF9MyR7M|XE*D%zQa|1)lhbT)e^yQpU5}Gk zqDoF3`1`iZL++fQ*K&{RDSYQ`K=q*73LHPG`C}(}F)zQo4AS^X!sB>^&MM=9WdVI_1c;iNf zwX*yjXLUjt`-QNMK^;2c8~a+fHG$yk-?AjWloY&qC`wqEiaxj{uzQR9v!*i|5-aL|Z0qQjzeflV4VAaPmrx zlR7@>%af3s@f<^eYy0tpaDWUK4Dy%L0!>k)=D`m1fWP<24eW_LS7!$DI$!@ok%(YM zBCuTl)H~F7d?(@LY`2$IvlSlV6hpeFhBMIT`OV%U?KL6`*`T4trS}Lb*6%bK*xVhQ zK1K{wQbV)5qn0aw3v%`dmX>CV$!7yTziq^br4##Rj4CrZP0Kr-<+@Bq-n{Ex-F4bd z+eRVe756yKh_CqiL?`gDpZsxksBzkDd9cFMC1!P*iUi+1Z-72fTKnX9;Lc8ulw0R; ziYG!UNUUo+w-h~&TL1Nl=(>#FGk=F(19xh}(SC*DM~NCSf@LcIT*1jiO&NwkWF972 zL`5_hy{_I+gA1vpwNpp8Crt3f(V5ZXAGT-Q0`CNFb-h(ic+oeA&O8NfJFfxW`q1BnD)lA=Fw)?&uX~uBE+)0t zHM9Q^$j4I5Dl!Z_h2+b}bOj-zJDv0ro33_+OubJbV)%w_jKmEejB%~52;fh|Yag@k z7@!_oa>+!+zdwLHCS!Reu_L2c0=1Gx=hct+q&kQ<$uky*{>4T?4|CE#yZ19ruj z$QCYXDzAizomMyL>f?SQgr>u8ri{SZO0U*}WHJxTAdH*?de z)6b(csU$9hTvlyjGp&$fd=^*L9sI|RP|P8f?bR?~=^d>ag-Oh0XA97pF^6;(88t)) z#;WDjeL}%FycVmwvbWX4rEooNw3tW*w-^#T;fn)mf$v2d%V!};9^Nhris9j%#PXSk zLvEPbF*-ZS{GmtW5Vi;J5e1}+!{wXm`d(ZTC-QWziFDh@nn)`pqdLw;yBWAM$70pG z+niP|_7H2-P8b}B5XMqsVhRV}vTJrD2jpAcw&+WA+CQ{PmxWO%)VhdH2XnPW*rIrW z78jC{_ddzEjKnVK68FDVuQxgUtqtTsqjXs}IRPp)_XHpR>wpuGGpF!cVihT+ntc55 z--!IWwbRqwKQDBqGu|{UPNZm|6F9mP!g}0ID)ExGQ81gc{h?>VD}aet5pW%5S=vDe z9`vpF>eRn_!Yoce*{8~6h|AOVGli{mLVfU}_qE4*U?%5$iX*cFMKek6zK=iGiRRKP zQ%}2g5&hyBPQwv{@%EjeS}H9T3<8P=*I!ArhRLc%lVT6{+8oW@eHx_LTUDN=y&0i&crCrfw4mREFHoEgADaG>MqPrcqQ9D``?9XzH#cz^kKz#y- zvGFZ1hE@8*)`xvxxMhV%c;BvWgVPHSCWByvCJn?+7n=oE!>YCMlL~#?$2z_An`ZHO zC5o0>F6+CS?aZG_@#-?AC48Jwx9!(-mA=jOYT-2C76AV5fgRAMB^nG15e2si!tUPQ z*V;<#uTnnTtH3xA;&rw)`z43~y$Hm~%8azI%{qlYa%BUVk)afX>#l!M2S_*7ZF-pz7WX-Y?%^8JD z##z8d#4ypd6_@SD{{KVOTZTpTzF)j_cO%^?(j7wxsGuOyAl)t9GjxYYgOt);(p{1w z(hW*CN+@xj(eLk^|9PX=bt_ zu9=u!#E=?Ule=+vz)R?pgTqWQft(vDjE9NjqNbL<#H7gLgbS1Z2#7WYX*?+N)WF0d zuNPwxPR;~yFJ?A2Gq{V(!u;!pCr|$e^K@gx<-s5pKv1SQG@Wr9E8#gaaDDAEjfgGD zm#g|}75mjz)8N-b9+&nGyoea2jd1bVEbEr7w9UpDAth>3tSQ=z+t`AE`;5&&;ztd_ zjd{6EG5d>1MuSG9`z9Yd8@oh#HuI!!S}7zPB6C#l+zL7~@gBx}WoIgw3z>Y7n26R9@ zmw+hj2Tp+g)Xp0VsvD+})JKEhCWd*gc%y$T59j#X3i_~4e(r+KcNKF%*DawmGCc#D zYIce=QD*SZ1?zl1P7s(c}c+(1a4dCTL8Bp-#RkBu5Mx-$BQfPEh4)sIYz zQ2=fqe8M|8{su{rR>UT#j>^%&U2d|-W?Y<+1m`G(sDTg-DM&z>@`H^XLZO&`Sixye zq)aU=LG9$%iM!C?%@Fqz-$mUr<2zsTqu4_{>SZvM&}#35&F6O4lSuB_&XFFi;+N~A zm+K57y-`@ZYU`K15R;Bm(U+U;Y3=JuB;Yk9CaK#9QSwvQR zQD@8Ft`FR#q9#0!)X7*l1_&|j3oP}6=1rwg_bocslH=cO)rW}W@s&I%!ydll!urq$ z0!FKArYkPJ(DfV?#Xvj?Ui=7^R|YM?CVrPP=Yey36pu_Sn$aymq#ea4l-YaBy=^@f zx$k@&fNWJ)>3}U#eqAT>!*n<>S83maCl%P&xH!cmCG&`#qI2&=*PU0KUjsqrcum;4A24 z&l`OEuWN2*jzbGG+R+;pOlA3O|19eI%?ck==AZ#W=ml1yrC1%m4Wi^mQe_Z`Ryc-K zgbd~()R-5fe$=&7bm9J#jM0bd($n>>1aihqZxT|*%^otidET*H-o=yl$MW@Jxf(A4 zZM<$l{Wpha1kfNMOY%mcs~&{U(QJ23npIr>1xZ&e~HC}#0pup;4 zAWa}~$Fxd{6R&o>c)|PxSSQh)7q*aFzTyAUyEeMe(DnP*_(eBPsKcyH*fiL>E5@N% zjv`o&yD`d=eZ?B?2P;#@@KXpF%k^})j*AVl5>9D}f?ASK-bh8ch>#U4o<`!LdSr-q zYEm^vng)3Eke*9lo7)L+rxO0-!YQP7^|`WU;kP1C_%x=!N9+4YQ^KD1M;KHxxyBz| z>uPFt$F^MP)Qse7PzYo&l+O)a$AyA(F)i;K`)?HXjaA&Qdlnr=6I>LdnuE+@V>tMA z(|aoXt)%Am_8iLvn{3y!g)OQ444(fvU25M-BS~sVUi>5-`~$3SocK`|c1nF-hcS$g(YMx=P+5)BWcu|CyVjSbdQ?f) z=Z2t8JETslp;X=9ywi0rT#B*xR;`(hVONTwFe&pENGy6J%e7oES$)Of#3?A^#mv6B}K>-HRdTmJAuyP$H< zzL;)QwXSwW_1N39T2*+#AjomLbjuYv+H^vqgETEurFzTtXQ&3PS*nb&k1u#i9Rui=Li;ZJHL&Me7)lJM8 z%gK~K5k3FI@N|6YeVTB;IQb%4Ql(Stq43R~c4rS@-h@mdI*p=Yh(H|j=p3Cj@PgVHv;T3)tJ*c?UanA>Dl)u%UkAMb9`kO zUoEdbv%vyyv5SjOv}lndBfE;EQ4PEpoT>LcViGh|UzqHE{U|Iwq%?rmd7gbdr?6D{ zVethjT7h?7PxKE^7A;=M6Q;TUGj03ux&%YJi(2iUb&Ath@l)RatP<9B=RG$t{vszp z^z4$Xkx&cm`)LwerjkHuT{vw zTH|B-vKlwFB=e;(XmJ{yI{MHWNw#OS?~m1MyH6G~7Hf9~D{q(L{;Y@{?j`*BU3D5) zk~288lhWzrcL@9QyFv^`Z{wI#fQf=x#_&5wcJ0sbz(mOLWsSKJTJL4##Jr&zN?H#7 z!Q6U~)JSUU9?ZgmM_)*MArg5oqtbrw6?r|SOvgD&5TLhCo+vT}r#~s&G*y#&H-eF> z{y`cj=-^=98H*4^^;L`Sbvd`?q(x-p-=Clkztv zJp*<`#81;jTuA@xT=t(E$~=I5Qyc#_@I-JJ~;Yd zt3F5Szi&mffppw9&ptez$7RHeV=xQ<{%vL9Q28p}s{>(js1oOUp&*Z*(Sh%c?ptZ~ z_88@3WVE|Mk4A@7@}n&CvFG~m-b^DmccA0!g%B7{{Ia9N(S4jiO4nLvWW2>zr^$iy z(}Bs_svDxJ7)Fdzd*z$5I}93vhhuIQ%hFHDLE<2>$2(==Sq;gdf#K({X|!7`A~1D( z{7g%7zS)(nlw$IKo4-nCenMsHeZ5ymkY_Ikn)j&?q8unH8-sez4PW{VlhkF;v#B7` zF!*4-dq7HpFh5`ElEEm+YR{HdfjE4bf%r7Nj=V5f^;NC@;M&`&QsYqxml$FU zJZaL1=dL4vZT^PE)r1sm>Q|5P zc?&Y?V5dfiXgNoHeg?0N&`F$}Nl`!DPshxR-X(im&RenxbL%0$v&=M0zm^klIE)^{ zcn;8*Lro!hO-_Zo!7Y?%IXZmK131PT3g;~YaFoXRmRMpgzgQ?i&-x;v(pRg&E*lo& z|EP^EF+-yC9to6LX?&Pw-&DM+cuuq)P^@kdgpLg~-J>0NW3eyxH^@cMLj0wi*AS|G zgsR=Sx+E23V43;N>IDWYvMp^VWZIro6ubxBSG-2#sQxi5K!!wn<|WA!Uv(0Np}_c0 z$3E%vz}C?D!ij4OOlGu6GY(S` z(Z>SBj?;EOJ}1zr?YDTXm2NgmB2l*9Vtekltbb&PG5IGKa?(#67*qYcl7ARNgsfUm z$iwA2)le9X6Ef2yA0{fByP8%Z?p0k1T<8kCuzU#S|48SCSca}qGtCMYwb?w+Rr}S= z-Jo{kYC1)ktd(MAV0FTv@Tofr_U8i8_X6S1WouLr3F24Fw4mlhCPS4JImu2_qZfR< z##w1qPQTr>HLKhC%(e`__FFCpdP-t3hhivx0{0jArWLr>tUn8)N#M1k+PBlq-994c z&5MgKGYd!P9Y-GsKpYaCjIkfJHa{t`ynALzVwp0SF^OyPxVzTDvi^d(`j){;)4n3y z=8+X1S+=dBqb`KHB1$NVP>wn~uf|4{#h)`k^7N&>^d@bLD(|g2n}!DR!2&A0kzDv5 zq{hM$8~lurkY^|8+Vp1#@Nv&$}a{?pqv-r36R1-~31JYoud4P*zbIU3ETA>wT+=^KR{)Hu(79$|ig zZ4u`<;|Z#(uk;A_+{OV$_8?HTN(Iy0-7gFPxnVp)W@RMXZ7d7J z7QC(|AZt}#5dX3TgR#qKX#SsT!H4S(@o%te_z3yIkPEF8I~KjnMAAYP9AVm37aw*V z(?!9NBBpRFP0vAQHb+_~v|N>=cPg1b7D~l1s_$l*AE8l-;0Zp})4~PaX1^_Eoy8jw zn%CBRqnttIm4h~{ZUP=7jCoSH9pl&MhM9t&`xEr_)e{zLY}dYj?uuj+&!b;<2*C(d z00ja3a3BpyzaRT(CaRG(Dp-a%lR24sy{d&nefPxpTQX$91!CT|yve%+FreQ61`!2R zK)io4evHoK@4y$ZUZWTfy%`f66(5xdKv7&(xYz~v!Oskb*qNYZN&m*s>7X7O_s}hs z@SrYX)zSRn@INpP_}xBaLM)JL>0GzWGTNqXO5>6CtX(z(HpgnoyJIN9)r*`jvumx5J(|_!YA_S2$r69wL1M!mxc>pLJt1Bbe<}^c=~FuVo{1H@uyPPzKHbh<2w@isOf&yd%%a`8@O%0itY z{oVICu)CFe{{YwUs({pPd<_iD&V8wDp1;a2`Fra~WkgW0ZBcblP~7THNh6yF^gN8_ zXL=OZh2cPD8wLA%2wB%VZz3+PLd+ktX8jy9`gOZGX2fJ0IruqW0kyzcBpsTa7`P#% zFt89lIUvteAkS2JP9x*f3NGt$s2Lub4$6lk5wkW8vXdE*{{4;rel0o{0v8^l#Uk|W z7{o?ctpmMy7E~>zP0(3wJ&q;tu31n)E$Cy43QI*A2KT9z29;qS2Hql>S%MnsNw35 z{QIBHr%^~FBt`WvkhW0NHF|S_mA;MO65NIx(Og(o<-sUxX+c)$B1^6(rofV}zVOO; zXtzt*H6}4NefP2Dyt#nhv-1m3eZEJmdvSS;K2zyV)(c+#EkTGyX%yu3pNE_qNzb>M z5%2`SJNJKANC7{ulBjqXb=b4j%l(>XJ3xM&e%aoyDDx`sz1?`|a|8;flu$;Sw#Be< z@y{Bg9>G~nywcba&(V9#+HgY&vQ-sAo-GtoB)I5m5qwZ~es}%!Lb02AzqCdUPHYY> z%mi4kwfkF{wX>0I(S#A{>n9=;Ka6e|ox4#hgXk|c424|;Go&7Ka&lG!*d%-q(fDb_ zr%Zkkj2vPuMpD+{1Fj4qJCNTM@NP2mb8ghXlSnF=m;UXqQFO#e-SG!$i0>kDeGW|$ zc_@f16|G5d!(2b490aj|Z4@#&02#|Pt+FvL=iCW6^|XV_$ML2m|XwzJC&@r zo}6cXuTWQrSMoswB?e9!dQ1}qg+FP+FB@_jSJ%GvOrUSdB1KJCk-71!%h_&{ zpQPVhmTV-*pUM+;9wIRQ%4{|D%w&14X8bU8rriy?d-z2wt;Q+}!D`%%tgp_p*_)>b zwT8eUZ+=Ms$&*OZ6R>3p?15Wv!)bkr>`5PB|}DG3gS zZU1?f@|pN%UFu&&Javbqhzc-9j|dh}PQ7JCpghlJK}HR(XD0(il$^hm+AEZU72NXI zxPB)vaA2fv_XSfIva*Q5*7_oO*adur!Si3nU2IZLkYDv>+C{ASe^?7~YAlqp^Xv-V zhhYGs1Kv|!)C5WEDE`4OTp2rE?l+savJQ#9>mzd!;29lIdk&=gD;a#>SY7vrq%$i# zf=(lY3>N8I<*XFp-tuB4L=z(zDG_?QW8^fUQoKivHEX#5?L!598!cv?*RSK;_h;6D ztEOSBihVBe5Do{)bn1wNJJaEEBYI^p7J=W`JL|lQZ#{aF(X%y?4?~0Vgk0>=5=7Bv zf?)9^Xani$RApF-SIKAqFkuDpT}MWT9@rcI@i1PMhBn(bG*tS4TwWfkM#fT!Y`~jcmo~NM41-^h3>Bq+?=VemA0_?2EsUA*O@@n|76JG2 zRw0nLaRyG7JYQ2vG~TeLH%YU5VrE$Dt5EKlP(YBrMl*OY*yMQtsZ+6N57~KYZ}Q^2 z;rRiKyBdhp42dber)-Is&gidf~DPht;W00?=GGoOst%KP>NYQ4Blct z=|yh@d(}wDktIBlGDUi0G}61af>b&n&PN7oA;ejss1@ArDmX%RM!=ADqifuQ@*3De zZHeDevT^6Gw&oSJ#;`m7TWBcY4-8aAL(i5WlRmkERK*X`88@Iih|hthls;*}>}y*# zj>&puL+Hf;Q;0Qj%%qlftHRM%u65i@Tg(qoE)~rif%GkQxL_w34Ob3uCTeSI#p`1q z2n`EA2YHGQ02-`jLd3KWLxQHlsp@Y5JCWBN)3Z0qK8mJ&foLINBEZW*@?DS%Yvx_r zp@GFtFYgJFKv%t+D=d9uD|rQotMk;(yN!vXmt2*f+F?Z@+N^i&!R)hWja!5kB#R}8 z<$hY|JgMd~P}C^_doYs?-n>O+D~wZHRW>6%y(naKygAND_%=u4xXsK6`Cd`{eYsI5 zt(cgYa9(rs>>4(OH+P3)M}U?h9LscaHu7k3F9Ogu5W!5&JT#t&^q;n7^6?cZip@iV z;ig{<0tD;=M+avSN&U=1I9BTW7dJbOyIF^fN?BL+!71Kp3$CfjmZ=FXDXC^2L5C?#KQNu zogDbsd4ST^LnV#3Tc68^@y79xztb8A0(c&#AQOz@=>-2|6$dA056E%>QETBfW`hgI ze4&f^_dQe=<)z)6Z6JiFu}ayZ-F~uaJy!LCwERvIxV>Yj#cEw0=B@kHOcafN_=4(0 z0EvtD;c8ayM~K@~X!Z#P8ybn>b8qkF_QJVV&nnfnbV+aijg8+a?T_aK$h|ejrr>kh zr{890%>d*=3x$x1$}UhQ4TUu$9I*(O%6LVDjm-RFw+jZbEF{KZW&U|*|0+szrd-#o zN=)p6_WzrEvYRge7Ma*8x+z##DnEaIF%uPM3X)-(?tbr|f+_YoI3R@`0j8JsUK+5l z;ncGi($^n8eb^U09~C%lUA@Ii(rIZxHgQ)FFXCNk2Ki)vM82+v_%oX20j!EnlXIT` z9h{C5vehO1)+MV8E4=MO56)z!da3rrg<&&Oa{Oeymk)4o{BAbs21nBc#+;(A)pyOW zE|)(BrwF?EN&b4=#^RXYTy-< z&U3{GM|$3liG239LNz}lH=)~$%XFmf3`SRzwthd#SwzxsXv>?rAS79VyC^EO56@AF zYO^$}IT2`+XZpv|$wJ;z7df{nTs|dB*bsfBctM??TphJWcAk8$caLJ-2oJb->ZG)KSb7{lPsGIn>y!pyt|X8{3< zP#ioyB-P-}lcI*`M;&vGKo0e7oG$R@^y2#8pviO6_^!%z1EPnieSrSd6QSQyP^)%Y z6aRx+0xx!5$1tkFo__Hqm=rEYI6YH$Y~e(Er}L0k)FSSb9xZ(U2+OjJSrK%>0b?ac!GT zf5@(8nbv%!cAE>N+n>Gzl3)tA{^*~wl@Yvj9@dRdRTc|{SW&EOy_gv^=sT#qVRm8LmN@v}T?vBN-nNc8FAaEw z5;%bKOU6=Hy;uYojTUv9i|aD)4g6M~SA;PAZMOBobW=&up&PM~JATlMh#0zK6mxiN zbe~w$J62;?GRFx`^}>`b3@{mdSaGN)-YDpCJeHdawGCo|ql|i$=MK75q>r4s{K5YNmpxx_;ixK}@VL zlgoAAUzEhcJx~v4Dn0%8yEz@FkKh^!SbNHiEC8-%`7`jA{#Eiq*E4CnT0ie9o)k-$ zGI1Jfdn_}&Zd@E2WJrXO{^Dc(+-K9cW%@2TIe8{yXW4VLVH(Ls1DM3&IXV8fZWhh( zlrasg(75{Z0N}ZD_$aqv{GVV_9C4t*ZtRzE8Rm>8f{8lB4U0eoC4$amL7CnX8ts71 z;3KHAXDOLVio!u4>D=YlZnx{TX4+t{azQ3nPm{d=@BXt4(L{Vjjq+3&8Fd79L+aS# z_XucyXX-p!DbpXbHbI{*xD<}PnNL0K9o^w3RC^yzkdH2VN*4+xL3v35D9@E|joJrBwi^){w+_^1s&L+H>b*Buu!w>ky$b-UEp~=q^mnwosH>|JQ7%~G*8u*G_w+G3%A`4NW()VwZ+tT%s_=&ytwS;FXaCaMF?>=Tu+sGUg^DkzHb ze4?)PGS`J!Uw9ln;%6hxKiskV_o$3wl;N^fCYWa7!)AzEJrMV2C17hALOlEL5I;s?k=y4-BnZR1 zLAg!ssr|Sa%D%xtef@QCkn81v{zRif6t2yRh;}8d;UKZcTp-mr7{r2caBzSFpW!)h z=;6>wQ^PH0H80nfnq3cDKzNjn`WoPJ9R`r#&>o8Yx9V_UW*nHuVtau^G$=Mp6@^MX zh*T4tixnjI6L06H2P?>ODAxZ^7C40<0$GD=-CZG=U83itMaF6vlOF$@LP7k6)bw#) zkt1kcZ`ELo>xmBQ@-wF=rjE5_;V@f@nHYbdOb+E}Vhq_4dzXrgj*mnof@C)LCr1?cG5m5 z&e3HT*Om-`v#W*V>O`Jif{LBpx%)#Jg_XRl3usHgLV>5ckNXw)q_>$5&gIwj5kx$D zq5E81Tw}*i5uZMLb|}e(S#f3=UTwdHe$zHvR`@1>Yr&M zVryp@BtqO#q=GmQCuDSt;uP-je?MZrOxIJGd-73D-Z(eL0}MVWQPP$ItekoXg5^M| zBg@M$f{AP<0FBTg*HLc--9@H}WcWtsTOuna6&L!&i^K7^LGy(;tcgF}jDK3QVVU#V z^kJmBKd@^N{5KR*3TEu8t>r&&)d(sr<@7a!UG&#;ZJDvVFmREuvarlY1-rYuvvYFF zd3$>oEx$7slE)(Uf5jl{h6>xJEOK9oJfc5Yn(}6E$#fPN`gRTe?mDl0ksTi6p#1}h zMIZXr0pbw(iIL#^Qy3rSdp>(qe;=goTgL6sll7A{J_9{$4dx*wR(}opA*>*?fhJ+| z2+%zcIU@x{wjTkrToip`OVNY&_v$C=dL?LGOu7)wnAL$z! z9#+`Ukx={|8_X&6N(4HUSMI%Cx6|L^K3=1|w=y`I__K&|R8TlGITpU1Ei<8%9d2v5 zjI=Hk+!+%uPB-0Bd;PN{AhBw;{`bPuZ<=Je9p(}b&QkIj0#Wxy@bfz=2kf_RRWF>d z9c2eK>@@6fl@=d08GUb)Ih3KT2xa`{VK|w=^sUi$OyoRnUQ25Rm?8IT-0j-E?IDgl zsp_NKrcFt#LiAi^h?tm47ntsCDGyHRkSCPDJ=HL7-+i^M?@46&T4zd|)rSQ<%73FH z!A^ez2+;dMj7uD*pcT0kS-`Ew;X>%I{nTA-q; z(SAPY;J_Y^Hy9g%G7-Au2R7^c(wmXF@^&6d8-X1gq_b^ z;P$+49?Arn#{Dn6HG+`G0iqU;ri_S{;Z*=KdU5{YqceNct<2piejA7;YFz!{ zX@9!W-}L?MvvTL^hA;VPgW_H%xejed1M_1kfY%wo^WrNwRia=)5*O(J;~VKk{8V{q#3RO;<9-Z9d4oQI8ba&A^j5?Cb0Iy^K~ec6+)GM#eH_iew) z;X7#) zWZsqH+1NBt$aLTlvwkz_f}#n8;nh7^|EAOMiEHzE=)3p;0+5&-O3jXrs!&PH9@p=1 z9~`HV_v(U1YTE7eHLY223nS!W8=Y);__?A#x^5>Shn^a3|uy2%h>iZc^Q${CSi z@>7`eB#&MnDFoSru_O}-8lK3qz9s8uLjy798_G_+uRTwg*)zZ|o~2JTo+xZ!l*RhV zPILoe!yFz5fi$Y@TS@P88Im?j0ED}b14GJ@i`E;O4p0*=a1(uZx`q!d;aPp~mjB{- z)#=Ys*IlNp)!oJR_wRGV7;rUOWy>C?N|xv(j}>vgz*=)ZR*K(uu_q>E6G{v%MDGJ* za`;v%&(+*SV7H7DkKZSW(>>OnY7NZcFwFfHPu2iIu_cV<@0&eVGF^VIF#QABJ-JvO zr<4(`fYB6VxFj7Oz?9}}9m z3p2fGb$r-3vu)}m&u~UH+%6R7VnRtlKo-B*6TZGNwAuDC_BzD-t}(*aYjkzc0MHhC zW`90iW-S-qc0aKj#(CE5JT(xG-)F_N?foos+fMVoZzjyZY23x5&)iR7@0G;u%Votj zjZb5an=&s9bl!Q#BD*N|g=n>h!`d`DX`E&pg{oSMC(qB)dJ}?Nw%l`Q`Dmwl1%Qa) z>hZ4;uL9}COxHd^h4lpyC*jR)omZNn4B{<-iA^G!3Nm`0;!x`Z-|R3N82(aDnpn#i zBF?-W%?O^2mZTGNoB_8Lr(Yud@&4U9eqiw?$HwbZD$Dl+gA%gzs6hpah-L!_W`p82^pArQzjlN7k|-L* z=ttv(SP>#GGnfb|`%{Lzv+CPS{G~7vnkFY-Cx|nhO!2|$(3)8>8OEiC#{L1Ba`7)) z+wA2V*9kVS_6$93ksP)CZ{J!rDO4dQ=Rh@gRO(MWt0e;~vehY9C+vp;@cg2<5&SRrhD*YG0DB`(C)YI;aNhJjgfq!zm!3s#dNRMck(^a z0BqD$ItXTf?m5;`xvptE`IRIouFyqu^U>h+gov1B@xWzhP+yK#r+H{VcaH*X(#Y%b zMW~bZ{UU8@6~hTdkz(v+qqLl+4zBdD4;3b+8w|4-ifBS1%so2biH?gU@8?IcA~HcE zxlzb|`nEVdHg0u+Or_Z2+c)t#ceYaIxo|a>zh^@!GkcIai!5ObCuWq|flOMoklk!S zODVMmo2AJsOU@oUZ0K3=a}*to@3eN$R0M)$1R!-AojT_dNyduWB^t{GI|9XTbLEp! z;=2x(-m5*SuEZPqoQIl5jP>%%phgmE{nx4XSUo6{mT+A5?Ww>vhafi(TBD zn>8rn(G?EQ>2)9Bt0!tG1y6$vgE=uT?DaH27Q4)X?wP>N=NAE{4^2CMetZN%n7Ft` z9LcG7ez+WW-UqaSSV65N5zA2uz+Mw;oP1r@P(5Ku+6!FP^G{**`x~b>Yy6wf`~ntm zWv;&`0r0@K^F}1w2D;1@;?{V!U4Kj)4?(`b$sx|j@UXq6>H_5%I7?=iMPwB2VYlZ; z%dHO6+5@v>Y>_XT8NR00#fwjP{E{`zF1dhO^m zUgOn8?f0q?TcY$$8_UD<{_be!^N8roYZ4eamQ0VxCz+Vqt*94*y}8Bal0v&^F;SmS zQV_6bmKMIdXA|%ti)bww57e7%J*ZrXso5X(Y9r(0E5})YmI<#$Ye&yC$4WhakscL3 z5ZIq-Ms@U4XTap<7YHaGtq(n$v#_R zE`}l_xCIr0@;Pv4TI`n>+>1-N2R_uu^S*J$>EXIWVrdASMjq|NeCuj3Vh@FG#-zZf#g#_^WOuQ8ROeI=SRIAfH_*)AbIbXk&crcbbF zVeDYOf61oyMahLB6e@|9TKCeJT?#ec;ZCLX{ou1FVs9iTNg`1m^Acft;1Yow(R%e+ z;u>%yI)e9?fh(@%SjMz6v+|XH-tu4cqusqP7*ua?8C^#;{8Z(Wf#6)B^`uQ0dM<~u zJWHWxV(O(m|2g$yuqPB-!XoQM;wT(KZFIY*ohB355sN#X>FMUn31Anm^?Bnob#$z& zxaR!$V)}IR%huWn--na*^QJsi7p`(~r;R#-PcmH-x6Sq6A_K^Y&+W0RADl~fBV>1llYicf3@AD87Y?>J4>=1^JU&m*rMDg~{^>Nbn{r0KI`bOS)d-F{1hgL9d)sejic%K9R zcPN$_*psMg%wE7qGS{aufMz4c-D(1VS7A&_;<{TGbI6*j;=SGTWGeCC${(3wjM07d zQ36=;avSGjMkxg&QPT2~1<5FQ2kST6v4WUj?5?1xZ6?SkW`BrUTL;^Ij|@`hXYX#T z;PEi;y^y?aT-KZdk?(%8t$B7L1D8L@Sz==?FLvJOlAKDeFv90H!sFIQvtQ_g=3k=7 zR@>57geqlJsv^O4Q-|GE!%A1c2sYE+);-Zz@!Q8nIX6MrcHRB$X;J77r`O8$b{06BmIXbfu_lvIE&|&tjm!;N z#mP-R>O=Dz;A+vlnR#rbB1Uc#=AM^DOp+EvJ!W6a=VaUcCkiEg_XD@IE#q#1&LOaQ zM2L{x-~dk$S^Q`_;h%AzFIY&yd0xB(SPbb|<7DhR=hqv>0{B%{`W+`1z4@YV4LuNm z6?uP*X8qmI5*^dm_@6GRjObF)&kJJjt&`Le)owfi7Dvl?=wdT6b-|)r4SHFWUFG8% z+iWtHBael0Eqw!wBnSJqv!^2SHXVcA_74~17eC!CW+GQUs41TGl^@>YJk|N(N+Bgq zt*g7Ptop-e_Wja*K+-ndgUqlOcv9V&c2wQlIRQnFwZ9UNi)5evortza{(kU+*$X4` zS8IYnYV6LNQma~6sh@TWmn2%mQzTP#Bupj)#UIf5!Z9SU$91vHB`XdGkx0kIJpq=);sK$!$HE2yq z{sh%5gFcMGCii|pghG56hU=P)$pR-@Pd?C*4IK;D>gk|)L?<>eP-#9Ht9>h%j{FtGLo0?2C#f+`so^F;@>U|m?>iIu~$n7?FgL; zy%7s_hE;EvUm=CG)B0>qZce#$3@(OjD6DKp|AV`V8Pc=XG(yDs4$XwO-;@8)vPy*O4Bkf%l4C>0?_ z7C_%|Y`F*$d_=igQSf|K{UMu|YZP#EAr_N`npywprLV@|cxJOIYPdGueVPbq zhAaBSufjM;jiFtxT6|EuS|C+_P#))9=tgKTm)s1`&F*(Da16(&<|0(@?yu;y;V(8e zhd4{W&*ER!cMVzFiqCtD3!|)xuj0Wv?iQgt26j9G zBe5Z?#Rdhv18agyzIWDUd6-$6=KLI|Dh7i%5!s7;`UFxGUA~iaD`}6*8d;bASUhPN!V$lWsl}^S zD)yfeZaWEvonR~9`koY#1B1DyrY23=Up!9c&PxHTTtTey)k!aYjn6Mz`<1qN$4-E_ z;3Dt?hMNk4M5Jy^VpijyJCj=GbGjm*Um?oE1q}i4S*%2H*&9-n5f3ELqI{)+IB*cq z?!R5#jgfbuC*CkamyJAWk3QNV*Ohwor6)CZ`UuJ7W0%aKUw`Ac<>N_VK%nSvAPvuv zd4+9(OH0a3W{`uPMnGEpy=T=8kW)(~$%6F&gL zA;1!(vAdk{swUol6bn@;_JPIY(4{S~R%5z5373IK_esDx!l*%jBhn8RXYdVgUtizc z_P@tL4y8-f>zX##9$hUN+fIVou`)aT=6%|a4PZxv4S6}CGw9)?dd`-&pu>rY!2cXw zZN*c2`0Z0*#G}fV#m1wxPxBMB0e5P#`@!?n%xU|0TZM7piNH(h^*lxsws*$da*H)t zN}Wa_XBDmxz4AHAP)x>#!$3Z~UH6vHIARO2mTGBd{97hy)4Olpdh+p!QC9J8bG}$Jr4M#+5$)V{ZN=87hVo7B5Pj^&Q8cJ!hwdzSob;@XwOA&GG-^6KoamLh$kd_;-)4{= zKkgdO?#i#$Xnz>L#vL9%ge0$vYb5&9%jqe`B<+8Tm>r%6qY)9hZ>Lcqp_B~C8HlM4 zi-u+w%v@N)!^v34jAm-&=`q9AIIlwR6&*%->3N7Fo_Zj@({?w|_<5)U`+Ri318t38k8a5VS-7voq7MxKRDJ&E5{H6W z;NET8$&YrvyWZXec;FsTdb{kE6fHP_z;}NnE<}*qYYYSq9#ltx!yH9(dqLP_+|^U? z{iM`DmY6Fqh&65f8-M{0Y81lq2Wji3OEvNa&nSLyhlwpeRewt$z1@B`CuH!mp>PW#XD@levOGtB+U9yC*|E%9et&oTN&ry} zz^%z*hfffTG3X*%cu}8#wZFV(GWJh3>LKG7N@_;OAy5+3f%GHQWZ*23b0cxt3b(jA zS=T~=8xl8O1pX#I0xh$?>dD2T(;z$u_b~jpWuIpWXj>Xx=0Q|^ICx+;L7q&4C$o~J zWg#i%YyV%>2C- zeVZbt?tvCsGr^!-ClTh%>})}4DHcE+4}JdJ4OWabcC$S8s~t3e)H7FO7H;?&2;deL z7A*UrCa-s~Xv+W~cOmg$sdqYD&G`A0@m=q${` zn|YoV4-2V#Yb;J6L#;gYMia#hy%j5{C56o4`#kapAWo_sG+xKFx&Vk7MJ8{b{(^h; z8x!78ZpO*pCwS`puCBY%PNf_%q^mW*NpbXj2*)d$F=k_%Fk~_U2m^eVriX@{k-8I^ z5VuzOQjbarqchq>#@}2;=8 zAein3Aztzf1HK9f3JPM~=vec{!ZGqa6v}kD>7~0PSiKq$v9~-`6jc@E`(SNgPp2G}qAw!6|QO`g} zV?6$rimHJt#ockb*VwX1*ngQTs*$1f04g$@)+V|3P;VS1(IC^33w=D}he8k`p`#dT zW)a#vy2Qx3RrQkO3;KG>BRp6Ufq!fO3lR{kR3{^eIIYCsOJ5;4wVk4+aO(yE9suTq z)Ft&D2*6#d__R{H(fRf3OMP$Ngf0Z*fB@;4N<+KLK)7h`_kNXrH$9z^|1ZrOkWkjJ z51_?1TqI{+@DI7;Bq-`ak4#5`k&@m`_%I||IIVF!#$!}4byicxnCJ403g?jDwlBqj$qdPNnW}%UfH?#nel8saoEG z8iKelMAx5hk9}QLy@$&S=pR+n9D0jqyuf)+^HR_I*RYJ22Ht?P4cE48H@PR z3W1UhVYl$3$^}1b(u4RkASpnX)sMLHqURZRv!W4g`CcQc5FNQYxB1^I8~pSLo{%H4 zm`L*SH+T|$+?dX@n2oe#j3{{rKKC`-5A|K*NH3B4J-|x`z>u#dx-Euy1hx$joVF_J zrmxv&?6Bp{;BeU!9bnz=lMdSeO&h&{2UNr}?w!6@_8JMd$5W+92R+*Iqti4|MeVv5_|1(&ho=zsQ;XRE2B~ zsm#4TtKrcYCPQ3kfM+X67Y|Ao0uU@X@Zm`k0U@DDe{f=<=Y~gK zxxBURC{|Fp=#Xtlxp-Q7dSoR zFNZlWIuLJZ@im!+sfRl-iX%ngX=td+g2K=Z zL!*+Cl0yw$5&{xKcUW|zfV4D-q|%MFw3M`TKbz~m?)(3|5MSs=;mkSv?7e<#eOJh{ zw-rO~Zig#s&|V4=6E~!yhoPR+Z{*``(Q`xX$=>=q%$wZ@5mz_TvmiEaC6d7tPL5s0 zHDCHbgsyIv{;sc96i}3A|M)pst7GAs{=J9(ScXfvzwsxyJ30RaIYj~IhMLK>fL{?6 z505^T@a=u5SPIHiD>>BSJewHJK3H#xmnaC7{;2Y9pMLQ!Z2y|bY{856*6>CYr!!5@ zO_IY7nI~O%u*)}nwaOf~`SXbAGYj%@qCD13{Z|>Jq@;Y{MWLE4J+DV72^u9zHC;i8 zdfi_}MZZ#zcGd{fkK4grA;&@Fbs zDIEEscE&f?9ViEbBkJu%?vVo4=!X$gQC`C7dO}feBF;x>_V&#m=#fH{YO<*$r?H9d z%`yL3ITU#F3Vvdv;!3}aI-aROMW31870CawC@xrXu}-A& z`8-e{0roDxu^6Ge+T|8=2(%?%ceP~)`ysB>z+w67)*v@*`EL(Qx*u+7@483$Y9w?; z_T-*#56}{m?mo5+j|tCOQVWA?b$Ogn$axUmJY@s+i+tsTE?_u41ZNmLVBUILOUD4} z;#O8x3~Q)QA98b7e?Sv)oJIhg$UmFPUXJWo(04LpIZ~H04gw-xDc$9zr8VH7?+!NW z`O?a-w@UwcpjpsLF%0x~9mBSZjrFXUmbIWdI7F5hF>j*O>3aHt+~7pr#X+T1qOfwG zj9s*lFCl}%ZVN+Y%B^>Dk$X8gNYk~r(=GE->l@YfU+b$f&kZ*O$92=p-;-mHy)^Lp zPq)lJm-1T}2Bk>F0P<5|%S(1hu zXc@t=jFw(ABjn$g}d7~XkN`Z>3R*EMS- zf6R!AE1b8q{H7ctr%s8W_v>39f=>oW(JWYCdzP#r5K07gAu$i$F(+m zT0|Ue5Qr`8o9Zhj%C)GDm&N62`>y6{l`LyIp1Ur6e21{x;n8j;&y&ZN=7?WwCh`wi zI)ZK}Y!IEdaP`Xn=MEm_Fvt7GPsnNNGAo_~%!Vc&Y+gfXzDuX$U@EPtN@1K*^*D4+> zx39KTGkkX`2pH1i!_%c735+8n<1zDoU5|t2e@=65cfP0&OOZd^ zO$;q_Amrp!Qljjy_V5okL2io`Hw}(nF7pQc_=22#ODfNNRQMmcYVFLK+3r}sjzK0N zI8h=$s)`IYm8)c;BZeTr4Mo`NcU~v9 zB;ikeED}G4&5TVtl-AcI7XQ9nFX_mz#w;5A5~``-Z~Z-yF;n40-$O60nB4FZ#mB{V zzgsof2Mha{Mo!AgF0i`0)L7KMb!U|ATq$&1^)pCZdQXnNpW#(&vf;Af(r<~TSB|?7 z9%s>#bdaF&g;_B$=qlV-s=ZR2p@hm=Y&A>i>hkJHh+$VKQL`em-;m{}bQ!to^GBP3 z7$=mHVPgUVM{(Xbgh?{NGT(p2&lMWS^%rU=R$cVHLQBSzm`Nv*^FTAlkG{Px`fS*1 z+CI2G8eRpzy6;PI9!I@k9;tw+Ae#bn%>540$hi?cUD&DW%I~DatX8hQr>BQq)|3JC zZ0vOc%Bu@E{=Lx$U|zFo0c^B&QqJ5$T|M3jSlcFBQm!x=LV^{_uYkor95fP6%+lX2 zVail~Ab6;WM{pUl7qNlsLh>6RiHEbKPK{T^x^X@j7Y>&Mb_}=F)R29BwO^M|BV?oa zJfPcC{aC|Wy^_&}+^FvRcZE6q1Fa1OWs3mq-C@xPr;~yS=H}e4M`fM~C*)9*haL7f zh(39&l0M|=KZ7ogIv9Uem2%hvB<)v~^O+@wjA0j(oszcSE*8@hB86k}e(~er?AlD% z4UgW%9h8s>xCKqbU=q;TR^WS>g8$a^fac824&}NbDFp7rX=P14g@H^KdTC$+3Dp&VGnsE7ujsY-l6^p?*HdNhK`{1itu>leQS4-0P!H(1ko1&u3_3~I1? z`ap%ZiLTs{4c&q%9qsi_>T#FDAg~#}&J+V8ute6xzVcn@Mrig=?-9SoRJ2|PHF1{B z3{kK1+rO8wt2g(7z4dgLu43JigI$+E+Qo~ZS36ga6_h)qZ|mlqJ1YWCxCgRj7y)n& z6HKwU%gFQGVLQG6sOS|d4#W#u)QwtDql~>DR|@Q1UE24>b0Cbu&{EF|STYvOJCr`A zpq&eYltaD7@+EWATb>*L_31AyI3U!?bt*>5(6nC}%(S}t<`AkXPL#a$JqSC0fM1V@ z>{}4f-2fi0FyQdr>CjVm*NiG4y3)S7-}HfvURr}-^upln&5#?rFa;GKiXIac6)Ah{ zVbX`x^4zpp+6~pt>~8OI#huSmh=gx9{yPOI9vf10b=b43*C=y#ES>p;FCC;7ejirgYnHxhf~8 z{cDw56zkO+6k;qb5*B-oQ%}ft!B9oI+bqQ#3N8FkkIy&FRs8p zy%7;;lGBnG*tv**N-8PhEq&0u>9Fy=yMO->Sd&^0cdH{eW6;bTFcO&4RxTezqm~r2 zXgwzhX>Rr=Bm16HBZ#uLwj6;f=IM#QF8+z3w^x@+un53JbXA@f3%6=`dZkC>yxKe5 zh%U^%+suBuLos7u(C^}BM}|>F!i@T~`O$hn040|=#$5(1dpSYo_-eG(+^pR_cq)B| z)!TH_uzGdrb#EOAor0J|VC5HlPlkn$5jY=+yEyP>x@c9GPCv3&moVH_3DfcdWtRH$?3mcEsa2PTak7 zk3qXO40P?DTKI#kliqnZ7i|D61KHhlT3hoRXkc^2_rhP>eO;79oTiAb_wo-$N%-$0 zcIkRcYnj`F+BJ4OZe^|aJCF(>N zSoGw*CuHYXicjY#Ofy3&X#|2lKVBuU&5@E*K%vQduF3c&b~chsC+t%mNmP*oBwH)w zs+mSA!D*;&yteJc$>nw*bJ?ez!d{~Gb^;|(=5j^s@T_U=G{Yr|yVEA(SADB+YsJAZ zHi^ANT_vRW@Q>CDYIqlaDHTj-{D^b@7zDwIZX^5RtE$ zKJ)+PkZW3EcxAK%iZK{3Z!|5pw_6$KC$JbVviJ{=fhJVLm-$XZdZz20hqi);U@X8S z4N25}DLw`$QtN;Sfp4{P9t5SfL(=gHs2F&ax`Q`Vg+wOkE~kGbtz>%K6F2g}tlQ*V zQQ6ycOpJ?_RX$I5bj@Q1I3dD(s!*X@UQ=ABDc07N&f;26pudQi_;6KdaK96xvkXv0 zwm%qkXz(1jT+liCCAn2ap%{GaUoeMl^7y&7wo_}-$F~cp^wxj>B33Lo(5xCl%vRfY z!erdZFfX4hh7R$eqYt^k=fz<^a9|T$up8f{!|=eB&(q3>@U`cqj}~Lc46EIUqTYP* zcAfO3BAWk#!e#GZ@IJ;YfvCRY*tYf$>W!+HXbC9Z0a*Ux(z96qT-^81Hm z$?-Q3u{*zaN$%bh{kguJYIWuaby$@G}O(MU%TXENZq!K2@ zyLp{1<`^pTu_oyXjBLX0DRdaft!Ll%UyQQ;2)Rso&2YWLdAYpNHJqINaNVYxcy|o zMYZUtdQ`mC=ju(;sT4&T-Bh4yM>piC;spgfK*Qnb0O5_PfQ6XtZfK|dGa^&7C?JD2 zw@V7^q~aFPR15VR|3DeFvyBvPz2T#_jif<@_bB{~;K&hv)V@RP5Z~{Bxqw%vL4k5~ zefps0&Skx^9U1Qmhrrfb{lGVL>ha~^e)hpx&{QOy0YocI);5dMf6&?cXP=8Vuui#8 znk?wtb+u0ZHV9tDGJSTC;X0LeE4-s&%(}enoR`Bi5c{g*%j;L6c3SJcLLJ9K=98CG z#%U>T_>!Y$Qg7e>8T@M~km#Az`R2{_?#p4oIF6@M;FmGy9`Bd5sX~25(q?JrN!vFd6xAe(f zhtIgg(M*SrXq-TV5|*|1XQISf1A@09~-hY8^+)0Zdoz#}w!be89+{N%&tB(D!T_1Jm_nBfNjf8S--43AR0VLFs4XiMNe zxDZu4#6lgqI}5yv0G>O{`o(~+^nT}?`Fk%81XkmWZjlZnA}Ket3)O2ao|lhrOxzu` zj?>k+9Vj6`48g*4)RHZC&kO803VxwyRPU_lANc%8P= zI$5{l{^-D`dE@sL*|qPteY)ZpBE4o}&iVE4!9rp1E8lAAF+m$eJk3Lkbym_*qIbyO z(-v>1ne{j$RVbI1mN}(%9IqF#76vxj(l#);$rI?rp0l(E;0@#A1ln4IC{BfhkEyHc z9M~Wp1O|=)Q~f1w)^Wk5h$IIm3w&j)KrNk!mew**1_PhQaq5d4sO5&>NEKWLe zvj6Y8KP}>v0ehh$#phS;wEE3jZNR>Gd9#^)&R}~9YYuq+`AZyu1$5m@Vc#} zqUq)#o#cT2KJt*GDckv{_s`tA5A~UHNo!t%SNGmWj!3k*0-fz{S7@)QuQ1Lv=iZ8% zPc)_&Jr2fCYDb%+GnaiRBFI`;x_5-rJ6evf(6sEK*0|8wruvG8>NF&lkovDnmws-$om?iKdxnm z2hS4X7WrBVGwh1^SW~zJ=8+V-(io@kLLbD-P9s){gb&z4@x`f^x2lP=_=$Lfdu-hBXieb8 z7Z%pe#3YLF(o%?O8d*5CKnKL%mx^80b@Xa(x96aJ!sU1%-T5ZA~JMgh*F*l3NWSfkv%{iu_r$Y78? zuBgcK*>775fB50lJOyC6*q+Z|~f8ID$V%{xxRZ zwV%xqk@qE};vs9Nby$`C(&lIz0HxrKqsI%Gum~u9W7aCkk(YJ~tFAImdYI6@`d9QR z&7LpH6^?pI1(DRMVf8-gc?$+E*1s1&PoIPs^E&|si4tm5!H)zFax~KJ2AtsKVxC6| zE+M%cZ%H^#^O;=FoGE1rMiRbQ3#&)qJmlYNcFrteI-fKu$6-Em>X@HGp_HB|68hyT z!4(+5Iw>to?$Wx;MLbqzko$~ToFWPhzg<^fl+i#d$Tf?l{pU{|Bj@)=k4BpTl!Y7R zUVeckSi9R{0xeQL9Pf)Belz&n_-XO!GC`I#NZ}U zO7f+p3mbD3cnhJtoFPP+yTjQD?2W>ZNV~MpX=11mFXRWv#YgQVh`5MRG6)=vLM2J(3Q2u@v{kDO1!1@6HEB^_bB_)U5%_=&+m1 z0yhUR%>!hqpqXFoFOD2@xwtghQr^5VM-M?qN57xkkfVgpCT-u>t%(myazDDK;g5}t zHP_N>EGZBNVsAs;A-}LmzIP9LS6NcSB2f88JpQ;^zXMQ@F}aX3?z4nm9^!~3NHr$^jC zHt_?h9bdD1At~11ZPg&AcRd^$Uhx7M1V#0pDK}U5Jwt~!>}B@5QDn;{AHqUi;Y z9G1Huv^8rz?J5r6HTUksj4=OOr-1D1s(@CZ709`E#bnL>Byr zO`YltRyu*pFB7QbKf$Oa2kRgyE&+H~b72W9?L(m!ZR=bVe?7|{@!vdQY_FheuvD(G z>z!^C7_CupAZ>r7ZYXp9A_snM*~PyQj|@Tjo`aWe@%6t`4-mJ->N?{lLnWO4yZ8I z>f7xRyFWDGHhAfWP7Lq5wKMUU#lr(Y;l)zIX?ECOD|S-9^;jnFdj_{%e*l0``1gZ&Ql96rJ)S~8)<@QBtDoo8ia>@tk(68?$ll5V z`d&Kc5!G%QQJ%K9_u2M>jxFcX_M>exP)#w>mUP|=%;ldK@IyrK@*HQz5csqfm3G=( zRn^*+fq`t%GeT)RkS>pyQi7gGni!sPQU3)(f*(LZCKN@sYpUN%qMJb+U5l;f_jTHH z7glBkKubcaPp=Sh3V*fx5_|J~4J2IBHjxyoDjImWDzVqv&&jb~JgLm*y89AO85ZFu z73;|z@K~XH1l;tcL@MozAQpNEsQoK0fmgH)0q5qDz#w<;HkpTS3}t!)OX4B8rnCt% zx1#rhWQkVuP7EWHM3&aA@xs;nvJG*Ixa8CrCGTFEgXcp|`W^(Kn4h6JDq z1(4moEpuIeNayx3ZLfZAc6nGZh`AEpv|7XdEfGZl=bGw0%u~+j17uF{=*k4Hghhs4 zHD+GxlVunWJx?Re)OB4kI7;I{K4cBr6x0{0jDlq!6==0_^0g9WD!J=kqcjHWhSU|q zlsi`EYj?$5%JCSoV<7#OntBv*X=$zs$`ZKCUkV)PLEyhgh&5pJwLCRKBTA=4M$a^9qk0^-H>H4pNshNK|`s1k+X7S zd@k^%ym)*7JTrBF7r&h?#2C4JBd`DU5_j+GeoH4ydzn%F3h>==0E;8TKTvK)VDy*J z7&QwBXxTc?gZTWl^4qaw>=fI`u>28RDt-P_TVwg2L&F<*@eIIf?kX+h^@y5*ym*j0 z?ANt-&tW*0=FEsevcO2wYMX&89FQM*QxG|@*8v2OV%Ra6$zUriKz8XeIH-HW|*%w!}V#RO#j3)mZK`iAH> ztOU2Zc0d29{P`p63@kixbd_vewm9@W%!bw{SfguJKL~+Z8xbsxgDydWgJ#64={UH( zZCA;~6av8{KuQ-u?N4fteT7I+ef5MJ0|QQ_85M00wXg`G&KHCgZ_#Low%1Y%WT+{) zBQyNZ<$*P-0eh6)d;ViP^M~tU|Tg`o`}|k0IU<-Wb=Gp&2Frf_T-hX zD%^{fr!u3~JHwH?={LBr3JSu@nKtFMn9(D$d~71xHgF}&o9OPBEzngn=%S#2W?^ah zCL_bq$ijjJgMcm)05r_k2U1hSJ-L+PXu@Pc%z21sLXSXIMFm1{UbE#Le=Pap#fub) zS3K%sYh?FTPGI{a`}Aq_d;s{k>w=e>eU3^a6DPk@Q&TTw-LC_@p%Z5-0zt8)2-(84 zZa`b%gMOrbsFRGW=wR|d7Uz7Lz(=m7%749YK2~sL+yuS79XAuk>sd@~ouY-F>nWSr z*k|(zs>fV2+10Lp1=m@@Mdu#i<0fpM!3G6Q6qw&rR=WvbbwK9^9G(q`^JhcGoaD7E zL_a|A?6H`e(kc?fhSTk-4GAi^W`Qw|mJD;uzCp+sVwEK_Apv8GKxyxggJo~g)H}Jy z&14T;mKLv^7Y@a~<{G5>)EDdihi}#+%2JimE?iQNt_M$ZR;h;UH}OAD&_7@Nsk$~E z5&++L3Fg?h?pOF09a8kZn?Ffq2A*B(`u>FRw~pi96J zt&!M2jzX~Z85zkQ0K<_~0BF`&={LKrs>Ne~zB@ruO3ykR5E%ZR+;aTD=tJTMw8nkj zfJ0nR0lL@$9P+Wit8Ji_cmJN}$YJ8Ol!lzfv>p6d@1BSyq9R;Ul;OtOyXB}nk!Cvc z7wF@x|7sAljA(Suev>SQoK09w0Kj=BkjHpbenni6ZroDZAuZ^8TJH|F zWM4YS0UIdPG&M_0EgW>+VFE#^S9E|b-TBm?@}CGvYJ4&KwBeOSOF5`z{?@pFEb#Cl zkrh!y*yyV?XHa4F*_eSEn=Qrp+2BS#B2u=#h0`swf|Li)r`J`RG|N&1?W`Zw0l%?}^};^gQ4GD9@+!zCx36~7PQ zfZD`YQ>&{Udx_$Z5wkzgN*H$wmX1b4O(@~{cm1*>AFvN3+sh2N+-d-rA-8sP%B6d9 z!M%2h9_aTm@}T)>sPkeYJK?@ir<*ZQvMZK4AHdo8 zU|IV9u5ZO6FX*i?^32F99qUQ4_@BK=nDCKhktGD^8i8L&-3hL7?bv!asx&t7Qlo-L z#-%iXV8>1VoQc!HoY|KwFw2v^!|;t0FmOK3TDS`mnUZb@bclaL^8u00VLEB)o+EY* zn}F6quh)cP*V3FnfUY_MD`(O$L}$2Aie5;reE*;&x>9_pyl)e3@dz@=YJk=DpD)MS zn+Mw3e;yp{AIv%f%Rp&~`$}6wE)PaFgr@E-YygRgrE913^`hJ{{2Iq!OKZJ^et#zv zlMo}E5K}3w;u0_my9Q3mg1V_hiG)v8bIq08Hq?dZ0pDf_U-{6HSsa6W{hp4fF!yBw z*#`MVHb%I~W)=Sr8 z4H#Ia@$kcJ{&!`tMgu=TM2nt82Gw(b;`$>Xr|x^HEmEE~B9s^$|6py7$sc*U8JI8e zN=K9r=}R!;1so`*o0dh(1YsF9EF(^35F_S1^B|#|T zI8%*yiskvcu4=@Z`pAMm4{hVHMKI%s1&76`RWko%0mr^2zhqDjk3;C@V?WCQpGn_y z+Zx49$^nzPtj^-R-38+#&6@Ln{6A8tILLO?#^z1(i>Em;osthz=~h?d}V*>_%Db{)*#A&*u82EM*$!g z%aOwZse{@FYzM|`i8eDdeN?`NTIhQkF&cUpGq%88x|X8ocKWkQUImx%j%JBM zllfwozb@6;KjL4GyVM|qLM1~~r||JS$y|3r#AJJ8V_Th5Td{6i1S1LvwK+_s?RjLA z5@K{B;i0l?7dA2Bp>jm6YgjJ+hwzGx=j=6TEs;NNMdg<0NUBjmwhin z5$Uiq2tcW41)1zhN%6QU*%T^1l@Ix?mB@9!#}4zOUUyJWH`m*?Jdy5tKZi>?qalHB zP}zokGmEMir;4~c?>1>RA_oXKga5P#P?Si3@~ydz$mk z4tn7U?g-EA_)j{vl5oyhE3tlpo21ML6GT4c%O`-d6L`Les(BnqZ>5(Wf@Up@&3s7# z)^cQAfr8iWgPr!lK&+@#GeIfAdfA%)1!749oHl+gJYj$H8@E@16w7&1WiH8yZmt75dLIuR!^UJ2V1za zhD_SFY4WiVVxl1ZBU2(uJ78SPM<7_n!;+G}BE>JtwZo!N*#CF05c$7{wT>iV#XwOz zoT|Tg(Yl<${G`p%h_ZompoRqAEbWw({5^X)DMlJSQ{Q?1n2P7}Ao={dZ*i6PtLf3C zhqPktnt|3g+}6&yhTD<+^V6<~xg9#zS&Vnx&z=r!v~)hm<2Bo}82;+j=1~rEAD{z&vsB~voes<&xwT*6he*aC1;;jG4AaCR( z`pD_Jh}%|&8Fa&oqT9GX6!66Lo3qSXP?xJ*jG?6b73s;#s`PdKy1MUS|G)JyKbN4$dfgUT)2&T%UDRl4 zazrx$BAC);`99SjcTjHF<^7Y(C#1%b($1C7&NWh3-d?UMBfW`&v;J7WB2g{Q`t}>+ zSJk0bdaPOyQ&@2IKo!Qq-Il`U>E93dXR9UX2}?qbKepFs53lHlEsy)n>M9gHB!Tri zsR_-aD|4#c^UH?kc1DPr`;$$2WO|n|RZ0OT8rKD%Hbu(&a*3_2$1yi6t0ogrLp(FT zFC|K%9UjVjd`-HX3d*eiSv~1aCU{_)CQCaQoSWRe!ubs+05>O)=S@0&XLfX6_X4g& zzSdZ7>Z6JQIj{nakVofy?zMrEkWI3ww*$I<4bg?|FeA`ds8 zPw>Qv$mi~3Cke2CM@O)pW5niRyn>VX!3 zDSbq}6f2O?2Aa?T`#h3B`9HNADO$2ZnkGBN=t+A>(UTw*v6OT~;OdH9d1nnfYekr7 zAYD8Qntfil8WV~1mIFDT2eT`s#5)>Yn*%j98>wo5+&?_E{k8~LnqmR!`bZ#79BG_(!|Os3-wY6wg1sYO|Sn)JH?!CpvW z?^Y?DN;{KfZ|P4;$_syNo8)Q$sh~Ps&Ou?P^eL|fy{a#VhZ2l3?ZT5HmIB?KJG0Q?B-{Ek2B5& zV&dK-FTLS?w&1qO6+NJ}96DKV!Y~8Aq-FfV-l%l~>So+MOxEvC4U_4=JL4MkrdyqjMun?9Pw1gu z4ot}PKxnCryAq`%+}FjDaIbR>(@gn1H)6PHx*fncs4@h)A9P+lIRv3dAPB51tHu{bY8|>Ti&Ori)kWS0>weZj zp12s=6LFz--Fyu81eOcWZP1PFCv{VZ(VQZ-2YP~6-!5e{8o4C_Q8HxHAVMm zt^-9dFb7ubQgxnuQOIaB9fo~@I(od!0{Ro6hO@L=HC3ha7+{T7`&{kTyUkc9zt94$ z9`AOmfN373Fw-mTVj-Qk<-c2mQ7)Q{rHYn5)ih;2Og7ca#Vbb8oUwuQTOXZ`3{Q{F zmG6%J4*Gqlm8Io3D%&zr9_`Z+P5rI--Ye8cY;7AvHz-XFbaC`+5sp9F<3q{OBueCR zZB;*A_-Zq$#4OjYKVAfISFV?CHeE?}^UosEQUy#b#1&u9Bcq3qJ$z0i>c#?b&UhBT z;Q(su>F+yKKm>w;YrCV_{A;>1N(k^6m>wX?*uTRTRox{p(7;dlBJxMNssJm79#-iz zuFhH`PT;Tj;I>U2rXk~6j`JPt}tGgrCuEDivOJAIrp;aKO)4ozI*~qeK!G0Wl zd^V^bB{vS$=-ZZQW4tc%3o(cQDHY#}$_W(i@tjWOt9dl$SFUPptkHmb>^FLF?RvkA z%Yx)1lO{4knd?fEdaGU%kb}i?p#Jo~AcrC9>NX#0(9c!lssIlbO?}+=d+31L{&K-f zPImkH9V(fN;L$!x!~)viWad(DRF+~w7XDI#WskcjnK<7QT=@b=D?}g6`<<38UDgsB zet{{s{Xvv{W>Engik@^5r!cqmDt-RxZE$`dnK4#vu8Kk6}X*Pp@1 z?A%y88b3{uQxsklQ^p$g^w6fffBYi#I-6z-gi^A1F_LZw6aMOszNp6d~ z6t9|3?hKOh9<7U0U|993C8sBEpzxKF^jJ&dAs_&!idX%BIwh4KJw>Qrm*8U=%0^@S zDPfg1QQny*tmS|VYY(9Q?(5L+LwzmX*ylSoGBK29e#G~n*Lvs7GESBD|DC5s`OSHd zv-wW=vM~YE{O#V&*Gms1@dKL5wS=~KWrVy764=U)2MQ#Bofzgo-J2qZNT`zn*B?wC zw5kEVlmnOf0IGNT2cKKpLbp;M=6`Ekw~MLVxBXSS+o(5#WdrtK9kG*9REpY9pJseJ z2*+>hifh}O=TTB)UI3}ZP@z8DZWePbuTKSdkAcdTV-^;HP75O4Ak#A$NZR!KXMBi)G6-ca_HXapNRmMMkQfE?$NL%g<(#p@!Ogws7LM(Uw zpH_uqk3SB;3mZ>G#$3FP)bX?)%)k)WgoAB&CgTh46sYJWnf|VzWu|Vak4Lq0GbwHtfyZ2gati!34#FB$#0EgoK7t`43>8A5 z-6^};^8sY+&r6G*zmW5Ih&KY@Aw`eWF#_SdjqQ4rJDf&Qr+?%4HfA1;#ed8%!Fo=m z!o0LT*cMIRR%PAR_n{3CD?tAE^=SiC4MK)AW|1=4QRA(m@v067pznGMWp3=6umARY zrIDj0*w~FQ!H2WrrpyBA@6$t>t>91GfJ{1P_g@8OrHCyaS;+)B_68+Uc<#sS7EH?s zkE{%pAjQq@_IXZ0j}s%t&#)i4f__vutX+9rzHA*90BtK;Z+a|20avXiM>j%U?d(Q5 zb5SFgZONK!VRu7gFJ~10&4pRDJH5P49TU%IA`fSW1T;TWwaN2+8eN}HBQH7F>g^p`*rn?~832DQ40>I@e%ijyPe_8uZbK^E4~V zf-atFD$2(}G>id$r+B;Spr3aC^c~ws0WL4|)9ep3#GDiUb+bRfU}brV_y`0zx3R=aap?AiJ{G4k1e?JyE= z)2klqGh}#XicIjUvIqL!VcvIC2Y{YiUmzcd?j~@U{uzv_Thb@C@=IHRcIBpyFgUcV z0Dx*=kYFQ>T(&Fwe&hst_vGvEc`1=uN+utuf4T_hpFML5Ypd+gjVzyz{eCGGSuz)K z(Dw~aRyk~Zv4NzE4psX?iYELs-R#s>WXVR4sZbvKqJuT_2P7=G<`FEBdCZZQIGe~n zZ4%Aa&sJF#*CHC@D=PdAe`hW0l3R;dozLOWXwD~B^ITH&a?mPnxTX-4F5GBc%Gq;b zT|f&D(lNxS7H2AaNMg05gf&`!zIbyBhG(CebAV;w_}f@c;R{w5hE%5rpM z1ZG%hGDArZLL;z+rWFig$!HVYc38B)DSiO#P`2f zhE){`-3LJ{T7z488Z0;iRF;BN5CUwUM#qaPPyn=DY=4$QuaZP-EuC0D5Q*>P0;}So z2Qo6jt-n@IEt2{xS;TZMj{%Xj5)N}%Ae^in=l%d-^bkP6r)(V_NWtUynQAsO89igr z53`L@=Y0j^C*TbYRMV}CFL1~8&N~TBPM~`DRt4;ep*6(trLcBtxE%q_?O|uo&2=3}Gg3#a{Hv7VXJq}{KF#u*+S*D%nO z&Ux-oGX9UA+4qm0X{A(P_=ZX2#i$o2$w%$MNa8l4aq^F!#HP})NsVcxf0ZS_jw8ww z`zZyajm^#)-OTn~3C&S0vsJEZT}pA>rN2_p&(v%m9Qhv4>7rOsz#uvwex?RVoLPtG zWIyPcq99(yxCG9t&^ORAZyVwa-;CF)sa8pG=quqqLBVh|1l|v z;eyT_3|-EU$gK5L*5%1mBDoA!Hmy$@(CiIlD<6|%&)iI=lz;t*HvM%rhWa+~Evt)$D-prYU@2xwZwq%6l)+V{g<^%8iNlOX!LtB8Pg zjmVT2)#j7ew*u$xl!io$psCmkGp^xY_ShJ#CAbaM(QU*D0{Hs<7F+xL75`B75PaX&9Wi79cProd&&a~-~>KrN?+|%-#1ds zY@PlcbO|601XSMLW6%a=7Z&U=nc|@6meSYv)EM+ML|WMC{Z|{VnEv?LH>d9BybY(I zrLrEm)j-tpU7Hf|a?3`Ll!h2zE;*IG>VtUKA%|{C=&YflGa_RyB7uIs-z~U0 zJ12S!VY4qm^YP4?=eVioRn@GS$%lGbrWT^n9c^tqo%D@QRFxL3)ZW@e4TFMuxPoQx z#|yp$1LG83O0sW)ZsYnnvqJ0uPHZm{NtN-ZwvX6F@q=QbK z)@i%r14T@|QMjxt1H68*67l@tW-LS0O`17H)8|1zXBvyvP-FGD-;#W)0kpjyQLvh6l`lomN+C!IDn+?x80ORJc`0n12u>tb22?x~EUN*OJkr6xPxWgMXN;U& z$DfdL>EaYx*M_UhskXa6I!|ng>VRfs6KYQKw^BIFZ+LvxbU^ki?E0oF9l9lDeo_4E4fP zp5)**Y60U3sIdC@By(7p5qT(ap2l+>9jhPquTOW{BZ*l{p%5{#8Kc)b89wW&-Zv*x zd>||T8w9Ch1%}#K6C_v4c3X|z2Zg*(U)bpC>?OPXdkBqe!LihA)^F=TJCQvT=sQEf zG($aYMKg*;91sNR;6PlFA*avUA~gVf&;(JuKo@Hv%Z>M2FaZFx=a!n-E(K#V1Dt=e zxRg6IJPIf;12L$AEx99(8`r2i~Y_$lVf#HQ_BOdF4v zM$*-PCm+SXs6#0lIKlWK`%_mm*K7gP*=SEbh_Ubm_6U3qy^;Kh3O z2h1B~J^a#XOdt$d9-MrsV&uM2EKE%@FCZOkm%An#XAwl{sl;MDS^7U3`= z(kjUWN5VxVCmku*>d56Hh2@?_?b_Uanb*DaOL-uExyHLXsYb7@TOxKYK9|>yTlWy2 z^dN-uXDS(`gv2u4(YyAgr%s4%iPfSO%|u;fOnj8^)bbR$w$eu&B*(Pen9oI3jLhNo zILw#qrA!T=L+X21*9u@tbQ}1_zJheqzsexT;Q>=6Fxy|+4>*&Mm}k5I#XS+24Bl<& z`5&%~0f*{8OVuUOskD5fek>wh!D#*)&NftM>sQRn1@}GW3+`Y3ku=qkEF>xnn0ThP zvn))j3b2vxfnz)9tRoZG&^+$)H&k(!4~{6H+Lz1)GW^QpnZc+Gc$rLezgJkGc+RqV z_(Os6h$z?mx@o{q3RFebrS2hE+JWm|a^BfMtun7;4lt_XkssoMch)h!V^U^gaQQlOc5cwitCM_X0CHoqNaR4y}q24JYZVr6!*ouDMa&qOs={^(*NN8^yM0 zRCu8K2lZSg^pHC7oa8N=evQ=ZnMeZ;oLE{Mws_(E++OIVBy6b{I`$pTh5Y;ziMiLA zE(boALy^d{7temxz`i1bw;D&y@5iDWF5OnLssQ~}RQ71v&>^Z$PHQzxUgKchFnkW( zf>0VM>2ZEo(N+uhw>Xr@C}c1pUmtKINBumPs}v^ONljf|1(-1V z?i0HD%b$`>tHw*2L<4XAB<88D;Be3kqey9D1rmyJE=$TXmv1unH11#R=WD!yc7^`3 z?7QKeI_3WsvyM(x%C*Tum&KXRKychWUeQB!MJm2yq7AecOWb0Rxp;2JABT{-{0OK##S%pcuoAq&FuWIkkB~53AuqL+s3WNSTIG+QrNI1;zV|nM zzdf5JwqfRb)Rj(}3plP`yC}krNYa-**yz;taO0~bS=gC#4qND9Dv?|b`F|cPMBV6Ctv0-5ijYfVbX(i0Cd{q5CZ~!U-Nf4=c;GE(^hs+ksA9BP8eM_BFsOlE? z0fp6&4^k{LY@L^E(mNJ72Ak}NKaT(13gBvPcE85a^(jX}Mh_}w>EQ~p{wn#3WuGCOoTwZl zdNR%WK_xKLN2JS+VYlUIBZ4@5&ZwH@&{5Y@tQp-uplgE^z}6lEMhWR(QNMD{FfxVK zu&7l5z_oGDL@gnI4%!N|^*4Sw8$5O`yoeq%OyP6rs>#nod}O0cSodgnF9n7nP?IQd z{E|je^oKKLND@=|r1mF0%lcjQr>JlhmiFMHdS66;CP9pQ`GyzS=oZWIu7uWV>iJH$ zWomX;gQ`HJ@Ktvz@~ei%zjY#TROg{8RhhD2Oa1}jkoi+GBz6rx)WaEPB5Btk2kN82 zs?ToFxfgb8qc5%P(MZ!zUGX zo0yG`Lqa&u&Q{=n9U&;?@VIFcQ~06d>8|6Zo2cESv?UT6L-;R{?9%UgIRb>n5%X$` zU7J3F-ejXT517Sej9dnv-2jf3$=|m>^~v-Up@s~oHvJc&zE@%&$o>TJ9pqr9Paegz zVqP8WP00HBeGI=qw8C9}Yi74WTpgZKBc@)<%=eY5wEt?tRU-ucX`I%a`-~n4`2zQR~!F>E3u1<%i)WEhY$3TwGkmT{5B~M`*yM129{q~ z=gnR>d?Qy-UKrA=CTz+6#Q$Xq)XEIM^Y;`B_YB~=>p?j6iOTYcID91>ox-4V_e0YPMWKNa_yTz|;y>z>t>mAp*PRsCbFkc4W*>fq)x6O#Z}>KZ z7&-I(Yw-7z7MDj}^8Bnn|B$K*?fpJ64{g zR=$yU#-=)g#*??p z0R)$%_E<#3*AWijFFlvtF6!{DOp5EK-LALCyweLiBufrTVond|$CLz87VP&M!mz`w z^ZaH&$IsgW7>GZc9x$7yY*rdKA zHoSr}nj@-&Q%BWw4f2zMb25j^?lop@uu0)j_ocXSc+9v>heW=*|4;v6JY{*NO2KfQ@+D9$jmm@Hlo?@=5UF@}~x%IYOzG zo?x-Mm3MjkU1eK1bn+RC1$Mg6=FH>fXXiAub#=`Fy><>{n-XlP<_P_|j4KH_w76-- z7t@5vNJt0*qV4jP7QL|WVs2LGF6-0r%2*@;i=F$>|C@H?J_x8cL;x(=bkwkT`px(O zawXxKPp>sFs`B$E?)=93nBoHY7DTu7 zGat;9_W)j1LP|=QViE&Qxa7P)65a+g0?1AjT(espyYBnGR8>R0iVFq~t>#7hxWpY+ zDh6(>DoV>hPVhKci81X7Hyui5{t%b;tYME7%JAI`@wKo!6<)o?#BI?-EmMLTmAnbb z1+)~wB4$tFT52?=M&Q#%V7tk>n* zFE-4-^}t6N%NG2pQ+uHJQb8V;A|w^xB-?#P{|4Nh$K^+Dr=q12@u#BKh{2(X@$^DM z&FU!svpVvCvQo*kh-koDwMJ*nrlL#tWQ6ol$>W+s)7pDX&OBSRiUXS-AMrJr88<^D znca+aof5T8H6otp_zw}y55zgNLHRZ~kiW09=KOd($r&P}=%#;h&8%-Mv*^NsIJTSj zB<~8--A4|r{zx=v@F;T|zefeXIw$C#3CtUbywWAmP;0|3hy~>4fvpY~c-zzs&=yEL zruR3F+-uf4)XwvbAL`nAU}}Rjf_F&6b?|B+2QVYZ!?%P?8&8aUG8HPRm^m!pe6FW{ z9IlvceHK$N7nh3Wm~s60kr9UQvQXlbGeYI*@h>g^!{Ke`0$n#Q(yM4n!taQXdpGtf9q)>5m!$h#e}b^@HC zh5eNQsOMV*yr~|U06p!DPS42LFE;cRO7Degz6U1T;)DVN*3}`uth^)eP^K_GdUW7F z^kMd=j)WLdM@A@bG9n~S zrPo2J*EoiXl)*GERU2)rMJ!;B#A%!8+NEJCOe13r76U<<=#bxK_)ueJ!u--}0uSEd za)AMg&7Zfr<@-4T)9jmO%R1OW^m%Sq=~6d~We+SeMgM*4s_+$E)JB}#w&H6>On;tG zB+9MtQh1iL??`{(8}Q$7WBV4_unClhIOsm4@Q`W;Wv9c>Z7u2nI4y&Qra(o!N4I)B zG%=rw#(XaHW==ZSv|Ciu>v6E5*USl@1lQH@<5d`lQG_6*u^Ct$buj6Ck~LGegf1@> zu9b@YVHOSmaTaU;1n6VcdqC9w0mADYOQH1MI}(Kd8nM6|bd8I|GE&RewOvsyWpao; z{$TSwUIg_6AUBr6;#8NTXgm7}hv;9;9WCizvapkT!x37MEA}~HEj z;T}dXfG}Xv1t+|mgo=gLSosH&*uhwLhWB zXa6eU(7pQ86~u3Fa>dQ^kubId*&>t*!eiek1%9!i#hrTmmaz*8%fDkR?AOxM3OC>v zwMr_hG`?F~pR={T*wruS-!Bh)g&?cNglU0$7f%Bswf7-YpuW`=bXpNHIn<-(I{Od^ zs%{~)f3;j4l|e=&VM8L7Y(Uv>T_98L3Y$_X?_~v{|NeZ6LK<4LB+7FCiV}2_phD^G zgm4?bT9XOF;N6LwyVsWe@g3Iv&?N0mn!rLG7Z`!oiVR{j`>~jI&uW^*Dps3LBDsz% zAG9R!W8SRHJOY}~(_7Hv#Q>uCfZ7w77g4ddo=OQX8#IB;b0@IAwzv-!SK1#z zMyBwpme*E-{G+(fF07tx3LrsAK@?GF6dViQ@8<{8{{H%nRR@3g>)wLT)8b3h_rKoA z%&ttn>0N30oX@0?w&H_^OWnr!bh0FjXJoXZY$WB~o@Z2|UecCfFeXX`o$7J>|M{i6 zxgJ3!7oWFq;uqaQ_gUek2EHuxC;2rp1>7=|Jdez2c(=F^LS{oWyM)U718U-5s2}X$ zL;Ht?VzT)#ULi0)^D^7CR=7ESF4DAPud;d-l3~R|=uvkSJD@r5_I9bTTMf`qNqpwi zLjRD2W0y$82TLH;tW`-P=gXpb@yfiWDJW!8di9A`68zrcjhE!~(5lS2r+%dfgQnxSg$l147IptZ?ZLejYDa3CDO9M3MjJ8W}dC?lH;1+}C z2Z!f~QC8A{bg{6nDLHKGRIn1aDT-gRAv!dpHCr&i35y}|wbA!Wl09dx`Y9T3*6ryr zcuO|nFJdQP2K_fM*eozXo(_d@9Np5RKP;o_HDTWz3$G1x1Qjc9C(-7m4N*VqjYMc1 z*!>{Yr70T`S;&Jz>2t^hMfCG+0wO%vJ%3}_cR&i5!qm_$MD9m`uv`QD`F@YSca>oy z*@AdPhE72w;s<55r(mlj!#SXdwrgiJD+_!Sk^g>_%4zPX;~NkM<`B0m=lEjH5a8G4 z=idH$k;!5#*YQJr|J^^{Hw-e)^*<5yn+m5 zyb@COG9Px6T1L89pEgn^gd(OYN8Tc7`^Jb8Pf!6%2CfZ+u%^* zUXK_AdB|L#BxE&eol}`5Ol30&zZk1s4~K5L0gQcrbXqu>gPv^ge?I@n!f62>iGlH| zgd`Q80m4PX?g0h;kDL|3h+bnua1?N4vV!)Mb|*F*n{0s-eKYiu8cOR)q8HBj=&;l| zh@0)L1L)u2Adf?2eCh1L&M}2KU&0X&vTADr+kEWXu3{-XvUAo%4+^EFgVKq~tkREz z$TbB%ga3BUf8M~en1j=_-t4kXY)=!^D#wZ~hk%-on|DNT$Z4?5@9aJp6DBfKj|i@X zQmi0v6)!vx;8^55NoDC(8;elD$tzXd2q=CE+@UK|HnF4X56=tROVzB3tuKfWl}2kh z)zi{}H&{jB>u`wUjl&G9VWpS4^5U8BqMg{`xy{NcO6HcBO7Jj%z6!62DSr`-_p0|z zESpzvG%{^c4peoQi6|;1s;kzNixgB<*8K2YQkM%~=4W}QP5+q*BaNf|1}^Q5M@B0$ z1ghnWfVk4J#OBfOySY(R0rxm;uj`=Xx#&-80HUHOy?!Y}R{RINXvMD|bX9wIk+GF!AteE#u>mWSQsE|d`;eu{}kZ%CX+ z^<^^5t>?AlnBM$^Mm)_2lEkJO(}8u^YSc09{R#KelCe@@EetFV4{JvE8Q1X~w0nj| zr(OsP{-p<=Hooi^+hB;kaK$tlajZ2+5NEI(w=K;H|Dsh|=yQEMrdiC_owwRxzexYH zYKSNK@@SC=tg(8eKF3Je+OilFsgTgr#BFxlH}mh0AyqW3(R4DpU8(ffQuy=yvkS#l zAh+m-$6Tc9`S>fnd0wC1Ty^usl)!ZZ=THz&!^Mjga`v(A!%G z8fuRB*w9`LrW2cU{WM;iG5k(vy+m1OQzk}c<*l9o-fk!3<@fheGKnEEPFw{9wvO^X z0V~U7nLan5D`5$o4|EFcSCNBwVTRVH$2$QqiOu=$P9`L>;xf$EcXkF{Ee0WSub5}` zHAECs&}X2wyY8=XfU5|txx>*(HdtRJe+}|NOI5|96tOuc7%}DvZ4J|GF%f-!(PV3D zTf&~=^)WtN^Hl3CMed)w9O3tTIhM~W{f)m6+IUJ_cfu)p*)}XwCt(H~*tE?@n>GqGcXha=+oYOEn|024vMKeaX9{1@@&xkQk9kwPFd_cTP4cl^xSTBo*}MI z$aIKSxh^mv8UQeLgX3x}_zq`Y)q}=gSec;sttkcX0kWXQ@>Cf61YEprN;skGpk2XDp)W!Zx>E6*J7Rr$=I09_{y4Hv{4wA^eH3 zA!Jq&N}CZWXCgA(rjPj1`v?jKe#U%%mnt-(?Pf@zhsBR*wl-vngFYeZO3+58Ogd;%DWnS3cpu>-u zr^~M*o?%MZS7>nB2Xhrb9S*Iv5+spgqjBzB`V}&poIz&{`TqjqK7krw=soT z)CVs5r3U&Fil@;tJvq#)PPcoS2*G~{Jp4 z4(%aSg3~|a-|1gWBhe$?{AaR0t{_mgve91WCXFWS%^3ezA&w>r#87;s%=oVFI-Eos zO^?;3l+SwCxvO&0?>pGgTmSoGDrdCG=F zBq9qO_~hG~Z<_j@O_QnzZs;H_TdqBkxH|FJ9%G*bsxo}#n6M9N^Wq*Kf`ZPQkC`gZ zIK9esk;-a|vrMvm32*rSq*O3rpf#6imdO>EeA6k9)?*VBThV|0l>a$4)(>Ac5#UHD zKpb(Sb9JR!#rGXjUm){Z-H-CG{`kn4YtsiGbPWp7A}`m$G?xEf(VA$cB%f%vqkpqy zA%wfK^I!~^k}1)~w<9X(V~`ud7CBUfquH688X!@?_kcjH*YF96b8_7i-I1E8FWrc2 z(v6L))cB8wqfTEx3#0b8U&BB-r!%mi;9N!{d+pBhO^?{fvmetWq?MAGz&fh+1!zNZ z_(JYDC#TaLLVg)O`ve}?^XJfS$(LW8@80NKWrpbvTGk+w_4>OZjm&{fYXMo8667>x z2loV3=b(?(|>>XImer(O)0OpJte>WpK@upoQ{ zIpy^*DbePuut}Pi4W;JBuZZxNUKG7dhz+6`uPq2$%z^ze%*B^_D@VPzNxbJY@Umaz zD%ITXgX#N4EBNi3Z?hY^;QB5xx%t4oD{h7mOkT;<-+SGQ)vv-wpn$5u79B#>8;Jj# zs~I!r=i;Dsa77SLN+DB~wK^G6Q&wxFpc-S>5Cp`{Pke|BNJ?0E1-;tQBDZ(O^BlWJ z+wP8A7>ZP-1I-2wDmzx9xi9w!bPY}WzJ&noYaJZD{F~|`X9(;~YjE6@Ochbo8jR55 z9{Tg8H`(B}yCHg^C^f;Hd7!$x^YW-OXf@~|%CZSvK_40K^_DcVKr5C=%~DeuzS)IU zt-bygJ2#Xf7*4NZwu$2`NW%lcKzGP`Z?4Gm=NWRQCr$P0rZ*w%=csW4b!M^G*#vHQs(|KB>^K{*#xy@Hh@1qovWF*4=y)H?UTG*1f zkIPrk&5>dP_V5p29yje{&LrbiIY${P94>9MdqVd z!QiWDza(qK)5}RV7~kV-QvQE?l4)8So%l4=Epmhxc;GN-u&2dtVrw+_LHI^J2 zXlRl)i|>YPu?SHmn)3kf+2FUpONuMQeQbp!*l>ap>Ub-7eAJp8`jQAVY*wdQXvS zjKK|h?Fgdzb{aXL6RGrf;`dX;&s?)kkmBTT>3q)?UD@MIZ(Aj7XJ)miNJcaa-!C4{ ztyFpL)1h%H|GsliO(J>AK_9!sDL%}&`A`Fe829)v;%>9ONJ7P0Jn5pmZ;2!BnPx11 z2{-oQ&v3DlUjfvoHzaStMR=IuS#e16Im|~^<2Q2|ac27w^3P9~PEu(6?9ttb8F*(p zAmD^bkWgQL3=Uc77W}#b${e_s_l*Sf+!=8nDwBnO)a2xY8g8h z1XCjj*1p+iCZ(QrC?Em|@>YF$KF5ODepmBWtp#TwQ8WDlE72FckO8qE2hN1$UG+B@ z2p5RCkz?+`haz+g67;HCn`kB7asGSRCBH^mB*C?YHCRP#Xhl7|L04GZV`YVpL`@=u zSzM`IbaQFA@mim#t;je#&nIaS23V)2ZWS;ifvI}5&G zUG1klZ$khRMy#@Mbk|OjQDl7koTgegZv z^Qt$3F^&kM5Te6^+kmLkEW20E;5$0rUwk1AHl_#F`_yywEE8W}ZOaa7BoEQEu%H3` zlX6U86TS6jBWldBxS-(QlYcE&z*1$7EEEZI@xo$BJx=J##hgaOwqUEXSi;Y#cuUDL zb3MGai1Qsn(2|bJ7yNsT5o_)xe;`IDUMI1%H@?~6XH)~wJN2+pUk0-WK!CVYJby(9-c=Gf41{QjDr9W&A9A{dT(GJTDPI52nr%xQV^G*h1KcR#6+nDrD37#WFgRG5?l!e%5p?nE`u!>|K8* zNYNY{{S=`E-Rsx7tu}+jQ~GfpH(W1xexmB!6A`{^WR?UWnxu^qin0|U?&}v_a)al5 zDmoTq;Miymv9>V{&-%>7#*OPVv(Vzaaz1F2l3@}WScZ>!b>@{X7dpHLnCoK*TrP9o~I^cF9cpPN>s@2N;$|U&_pcdf#(?lY#{JZdMil>U=PP6<-%I`0I=Mnp{_apm$ZQ_J5Dm!~kiL*+w(I z6<1;1>AO4O(nEQSi}o7bbKFI?s|SVyU$!2vn$fq4+VcZq#Hb&S$X+2XYBSvrkJm}( z!Y^uzn6@CR;;Xzh8znGnHkj>9ub7btvS)Q8u((4RC9cDe`~piHeDf=a-(+yFA4?d4 zOE?yJ!7E!zP-usXsq0VJY?hZaUv$599&DM;&2JXrtxI{Td7c0<>CD$b6sF77mEM~R zY89<7=9qKYgBKh0ap)@%joGTJC8XdcK1EQ}goqqeE!zl2YA%{g$Ek}yOh;4s99CO- zp0r(W01!VPc-#e;?Xd*ltVCsv*IY___I(I_@J&etA||WsK$omdVjKSGMHKbl}H z);iHk$VCo@Xue(?O+e2^*+g8J-s@$pn|?SuJF3AHOt?%cjJkL%|KUwb0i}`%Smy_O zj;7DC9(O`C_gbJH!9>=1Bv;e>_~D)@yUCq)%$SR084I1?A&|`g?r3olyNf(7+D%6c zor%F>duSk`^L7{Bynmqs2c;}t@Q5ccT<6nPQvZX^iK?tEu`e($HeF6BkKWvzZJq&S zRfcnKH$(!0cz1mgS69d5rJEZLAY{VIEOKUMg{`II{N3#i%gq__J?jx9TJk*@|1E2k z>_ftl1^Kvs<1F$ij7ge+!KXf_bA$cKX*dH4I#=thJ7`2|%IWA?ll>-o$LYr|OcR}& zh)q9{n_u~xPXzkP*`M23n`WUb$`#_46-c=;Qa2Dy_=Xm(dt@i1Gtp}&zwQp43spju z1o*DdlZqDRHS6 zIqn+f=&F7=4uUj9m z6e@&p29{_LElTVq7Obe~=imyCmDG~&M z>DOKVS~Eu6W@vP^%_=LzenAiH3`z=EpqO5zim2Et;DC@(5qOY#nQQX~_U^hdAZje| z?3~1$UI|lh^w!XSgv-OvBm7{cZDT!>K=z9PG&U3keqW;BsTuzp^RxZ=+EN@y?f_A- z6If|ohk8&px5H*LF@srjD`f2K%AB$7W-D@P>o)#ReSb@XKOM#opLqqp5u1FbmrqTl zu!wPJ40HIP_~N2t;8N#csh_hOG!?gZ{wZgxL1o(S7zt&E%OyDz;)rp+)MPU3;SLO_ z+C>Z|y2$<|6wFm^cnFh2&z9FHWOPF=f|a&lS?uffAZVx!RraEvcE_6DaVx&+!u}Dq zDfXhLCx~JQDq&iqY^dBaR~lp8-*a=imQYkYeXew#`Q#bP?&Gk*v&&(Ka!)VrQ}Gnl ziYveohSPRo=U4we$K5`>Unz)=ZtErSK zpV8`>JgBp8-sg}!L80nUi%4|GpBV#z{h}mjnh|&Tv?Gr$w9T5YGEu3t?UrI@yU|Ax z1&uA`pV49dLRi`3ANtN{V z6#=<1!d`JhQQ9}+US2vfOat%6a{EJPeX%!x=)wVlZGQ8#cm97z1Ahny)3iqV2V`^D zk#O0oH@npQb7;bzZcyh=X{!#FLPWbbAuOHcL- z(jLZuzhr56!g)%LB<(dr(;9m zoK&l{U&U2DWd}ef ztU9wE;QO%tY9SE9Z9T?AQ8^2e7E4n8D!-&@@n`u17k*vv_Zle4+@>KVYB`a|b?O># z+3s)cld>m4f(te&X=J7FhJ~XDIpt#wNFzR~m0e?O zkba_%m3n=vw!ZNUmqSP*nbkP27H5ikNKbr9Qrr(e=7E7R`Pv >;kcnzM)E3AT@- ze^e@YN8S18U0hxN|6*$L%C=I*OL77Ku8t}JeWh9{HyZ?pqjMsWIajHiCxWB-T4WZT zG6^*`Oz-PsHpFR6&CZ!eDvXTiX!NAgx$mx}&Mtsg&rix9DV9kXVNundU1?{)$x_wBc>(p!fL5@TB{LbPG^Z^#&+kf~`w;}btZhVb@NK4aP0CZne0Sa5 z$gs|Z?y?<>l;44=3Nb)DMhP|wA2;=A|44PuAt-Kr$@;MODGpl7_5iPcG>j*T&wC~% zg6>wt;SyvlN}gND4##ho6{#;-^v|AUh&a4{ooS1=KUIt-H9aN&g+^~?!zW?f5`aA? zfuP6pc6vbnwYoslR5;ek14exNywG*^cRH2x#&UI}g-67JB@8=1<(W5_T7m?$tKp|R zGasuqZzWo4Et;7rU*BO;8eTLkR~9nW!!_d>;d@i+uzl#(HP-Jk4#)Dd*ROiKX}t*i z6ssgF^Fv8SjzP)4I;oY+hqkq*dy*W1A(_d)ml|N|{mHf}4=TVneyD@jFAT9vnY_F%Y=5xMe`w&346H0Y*|YI7xM}r+ZgC zworSO>90=932qKApF`E|^uLmr`uv`tm7&e_DEjq*uK*|%$wZK+^Qj=aL7WoD6sco~ z$+(Kia8_quS|CQEp1_z|qG90{!*~Cntz1{k1+T+byUlP@U&Yk>=4)eS?gm*bW_{u_ ziDKXj+TuyL;lj@3=bgF=G@fH?|E8$4%lVkEKE|5xJSr=-!w@L5wqE*Rl-P2^tmxnP zt8-1iF-4%UgRLIpx6^Tq9auBe%n&<7=xrYtQ*fEfqQiIx>2d^mUUq zHBgnXn&*M2v1e5Z*L4nNMd~KN1Z9;CHw@JS_xCktn8U!V{Ide}z}UwWW^DS3Boo)Z zWZi#a5)ezSaDGwR1Q1T#z93uwa72!ex>F2$SL7=v-(`Q&8T!|(4(e9iGXYut!8S9Q6^Hn_rLBG(J9`8UTt*N+xp-?S`XUV@MCtuP2n;&m_C0?s$^N3raI0I=SifyMAb^ZwYm zHlN1*_y;gr7gA378_xF$JWK>guRs1ZN%g7`8Kr=#OWYx9|7q8vc$JjoRqWyeZ+wl; zj3{-ONi))U&p}Z<$^j4Ylpe!2Z+053;5M7YZ_@cyq+7fIYBq7wL7L?R5>f0h=a9fY z3JbC;vx)4|4YKg5oGPccs2ZDyQl(ioH#KyResr^WM|Y4bh^GR41V7F&o6(Mn@ZNqK3sd(38zb$`5{1(|F4r z&BN<|@40!i0!Ce0)}#vfHNc8TQm{5rAf1qw76A?I92WF!ZDXngcYyeX?WMCgJd&EK2Rf6@Cg?DOsBBu7g2OE909_Qz7& zEiVw0@%Nei`K`ceOvx&3FezDWay&-Z<9Vy3>MAfTrI%88cihx=x)g=y= zu&w@PzN_fk8f;%oPL_sdNs)Xb`_H2cL!7-Qr<>v~Y>T!}G%$2T3{pf3^m7do_fJ{( z1!4tUNAw$shE2i&I41=<1A!9mwX5qu`$T{d9dF(nlQ0ZmX2pUt^I1G*WM+A$$evs% z)4k$)sL$9Vb^MugJd%*07a1QVSLk{@FzgvVx+bgVUlB6R3YycYz<196iwSbX3e9r$ zeFq^9FQkii^zs_WI2<1;WHg853w#^(2D>>?&?Bn+{WxbTViKp(S#;&d67rJ~tF7bi zAAj0?4h_9*_bfQuY%1K=BVR1z7_$*2uf%AD6{qAT*%Zd)U&kdD3WAeNrJ$qB5iikx z-kuxq$c`_I8VxwV1gO&ts+ks{?}pZ0`$(+UKxm;nNItA~KhUAWzX!R8qcvbUOccgU zBJ9Cx*u)+z^S=%W&0$j?g{0g>FY&&HfGWR`UId}A?+&H~s)0>G5*MEW z70Q!dB?}FwQw7>K!ujb6aKlC5I5ACpQw*X$=wrrKQ9+ogK=glFjzrYQNJzGbs7O!f z|H|Z!<58H)Z5>NeDp6(J;cwL{A|ori4)!lP!Q#`~5-QI%*d^fYurg#GOlH?ObVI&11Vy!%BJcKEiGshy-_%~)@Iw=CN?p$SEce{RATBZx%WLHf5X_e+6ex-#)I z{p?@fp|wRW`FgHXnIy97xg>w}HNL{>SXd0~Mwo`~V-8?9AkU5G1r$6rwjQBL- z;#72U(VgI)Jt}*1d$=h-Ci(w7+rja$kI9$)y z3a9mTE)kEj>>{>93x)Z-W)RA>N@3!IJGY0oPj&UKCFFKYk$w4gUnHva$2MGu^q?YR zXS~vQOQHM5oj~mI#++NsybnFwsT^Z^k=d+PD>gv$~(bo-P!L&Z|h7a9I%&%&@r$5&TABis#wmVDN zp;{Ds5!vR#I{%_r;y^2~&)$+kKc^cKiT2(P)-KBwZ5SUi72~w2{Njm)`duX(k5&a8&Nw3pVUe1Nx(#EjIIriz*u6r`gtNAU_Pn^=&3kZAb)MnNaYcLd zhf2c0JY6cvNNQj7yjx|1T(^qL`|n0-B35+=w~>%snXF3%Qgr*Ls6(Xb$_(BQaK&%) z`0I3UyAqJG zk7n`9SIrFlkFs ze(`aYAneUJh7R?>8RVso_2pN3eIr4Kh`VRXhw-nl%A~D#y>+Pvyf1P^?fseyr+LA( zE(Vp~t?q%Pj@wcEIZ~{Z8EN24L=it2OIV_jc zFn0LL7O43lt(Iy)`Mu10);4ci!Jok*`nofNbGEbNpIVG72oR?p;(|N-jsnx!C6Xv# z96HGQO}?P``70c&#PSh`E-008;D&ne&5(!J`8Bb_O{eB!7P~WH)d`VsPIdp|vO>RP z)iyD;fBG^<4_>WAIS^E-R_55p6N&H<+IiN5Hd7mIF13?VV{>=7>yD4h;D39*PoI#D z*^mf#)=3@;6n#Elo4}k$QQ0P@o4s?deJgXcfM~a96@eZRpJB`LnwIu_qRy6Yp?+9) zwjykgJ?|U)`4=UHsuT)*^qNAtS7Gtei&5FMOpZ3bO9M zO^>Pt`6_2_fF+ZzFd`$H5_2=N5#AFC<%Q=z2vEO9{oh5(boq&!8Uy$00kTP(o&<}4 z2tRogq|>ih;?IOliH5%guYVVmoYI+_F(X1_*=A2>YMDm*j7h1So|PX`P^n(G(Mi=O z;K~yIUiAH|ZXu6w^pe<*lkbVcB51oPgrV-VWub#!efhD!=SFRF`cus)HRhsWA(5JC zH$bup0qgQ7R@r7e07c!w2p;U29s6gu3 zV(U%A49MwzN(Yws6-{KB%woKIet`s^^X%u6Y3s;|?;e}-`udpqGx@))@i##G%)GbH zprQ>(qKU*~PgvSRm@i3s(=iZbtl?WL;Pr{y;{%IOZs@RZ?w*Q%vrK*W^75EVN#up= zRDsG88;g(mq?2+?=(6Q7<3*SywL%QxH` z2hC@#Z#?am@_g)#FxT`%uf2Mwth2#|lGgI2b+)t3ab@=;yoc=@d-t87`G#|g6IJd( zGScie?|EyZiU{8dopFCs4qM*y+fc3%SlIx4G~D-#pHIeQB6_XORE9Fr@wk3!dsIz@ z5D=A3+sxO%*e_a9)6kG;g)fD&WadRUhgokl!CM!u_Hm!-@2E`N59V_T;IfQraWa`M&3%v;^=&>(Y&%NGHzd%{T@kVmFIXc6XBk!FDYU0+F3#L-n1r~F@6>JQB|7=rC z(bWvj1OYoA{x5+`vitWv!nPhgV=~c&&dY=we{-jrJa4vxrcee zbk{wzJyz}w3B`5Q^7we@R~;$yGC4!SMd`T?t8r{mk>B1_anvmZ|5ODFM&{CzVVQ9U zB^kd%ap5t!)a(MwQOX>A!jxN3Rrnr%syV=%PU(HFD@k+6pQLXQ8 zv+(CR$-nX=Zi=xwtA)~jxzp*el|n{_iM0MAiR^tkJsy-kRo?;CzG zG`G^Cmxoa@%=$hO$ipnH2wzujT8N6zY7S?$ZeA}qpJi;-Ol4DeRkn-$hVAG`nmV5S z;!2geDt~q`ygN+>;iWRX^N_$yD(~KOlD>0pYB&<0!?fU#)6t&Tjx!U}(cxSQs$4yh zMXr>t%weRZ4L(+Dz@?+5)oQ0W*v_dY3wZbFiyD-1{pXTC{$Lk1@6*L%Z1;zz;}2+2 z>{l%uM`Fv*19pphTGjDU+0Sv6TCVp?wmH%~#oE@JIfh!sS1(RZ1W$d9y|rfeeDfYu zSKL>wi+F~F#w3{HMaFB; z5VJYBSyEh_(X2i5@;VP(m%s^d@Ca8_mTEhnNTr*}{GxkNolbblm)S-sey5DME z{RXF|5>q0{NTMjNxu8`=U%a>Br;|@F<6l1qJyz9~Xhwd&*!@L+rycF+&5gT_irC$P zqo8)r2lut@wC!CM@y4EF`$cVl6H&_ceoG4^*U*`AAKL7f!2A^yfleSi(d8L(fd8~lJX|K$eQ_}x=y0&eU@*F*8D05r4i-7KcCjmGa0Ywc7C zy(*KSFH~Fc?jgWx&R95)trzndYOHaOlN#Kf923)W@%}=#J=&rI$-D>@_eZ)KtKQ6( zAO5QXq)uzuK*PK+)lW072~vz~r_3;R3%V{LZm7v_KPDw=C3TwmyX}j%aN%)Ka2s~G z>764k%0HQyGfyWi6@G(WM+S6$b~%LoiXx_G8Lmz-+&bg|!I=_XMR6i= zMlUbV*D^QO@0p4&o?R8@7mL7kP3311_l8)8@f1&dC{TxY?q@+e4w0%lM?rW~p?*_~ zW9$V%ct}W1`=|wXFN`m&;q+>OY(3x>&3UnbA08J4*N7V;(m$59F@j1iRGspy^D8Qo z;g(>^Gp9{nQqVWKhz?)JZa2AD6#{qj3}9&>EB zo7A_%77@;U&mY{2?<>G~kt~`@d<8^UQy{#e*6VQ2LCE5FbO`@n@d*hMbgx2fNlTb$qC zwKJe|9ON;76(%R13}2WYP!KMAOY9ZV-MJco%;zAY-7(go>TEKe+o0yRCP8Kzsn!%QORQuKwp9CHp83=;!*a&T?Kd(Azs=EHx^NCnX&2IHrQT~0__Q3?km zJjX6$gGwFA&9FXSwyl*V&jQ(~t&ZWQ8Wq=_x#OGFAVuGvDZ=tv#*8Fo8}B$Uzobl- zdRu|7sjrk`L5}Q)BHbiXSACR2$P)G{96RQ>RT$f2#YsMucyBj6FwlC+%~%(89#<|9V~A z^r=UEJUupkutH71%IQ#Yoqx`ORf!=YJ#6mfeUTRNFkDcZ*N8phzWfQE>?gcvUzS(y z%ZbZ@zgX~(clgLi&-dsYPww>}|Fm5T-e#RTuR|=V`pb!08QmPs&&8~8Kfp{TEzcuA zr_V2Y%Vfb8=ag^|&~(Rd{nf5r!H{_3VWYn6bD)by7rBbE7s0{oB4upL7T$XavG?dP zYglg?z5mn{V0&LywNWk(sSB}Jz~JDhWpBXbUldesmP8fbEj_i27w>qThjz4dRQio) zJkN?;W_(&7Yb!X7TSS1&mbL)vM?G2i;0>sZM{hs|mZg+F(;qT91!__Uy?@mZsBK94 z6rDNH{CAKU<(c4N5fwytl{*-XdDGmXN+GZ4>Z)Yej8B+{#DolQwd42_y#hC zOvh_!4HHQ|Y+Tbt72;nXHe2YX7$Um+@gr;iMJldnQeH zWQxWHl5~i{Zg@owA(hxA(|)6f6GZk2tM=%OGP?+CE)9I&V4aA}Nm$Hjl`k zs5D4Uju*)%{s1Q=)_{L>gB!^;V*N@Yh%C&jLT@lnYgPS@DYWe7I)hfhc(?auUn_%~ zvbp!?U4&Nli&kdUfn6-J3U`ai@3*+rxJFO94gM!81lDq6QWjHhe!)k6Z|eBNDE6+q z$8MZ>p`b2PKozK zG*4dKCm+VExXi*(KX%6Vs1#cY?u(7(#zq_jKiw1F>UecQXu`jH~N=YOV{A+`u}6=ErY83-so>>I5Zr( z8>G8iLOPW0RO#;SLw86@D-F`!ol18j4H8G{x%vM7PrP_$7~V3&+um>F7dQp!ePBi12aMr94ZZ-0c_u(&q?JMW#($64fA{Dw*+qXTQ=?{X zS9otzCys)!HY1b@&HlU_K-RK&d(rza1@;A(LB1q%?FbU6KE!R zxO~D&{S<>tM8&yPhsQ^3s%l&<1~ug2{6o9h_@Px$rY-I7; zU=a{l_kXb~^RBd_rlGk!D$U!z8e`|5bNyYYQ`7mUjF9{uP@MDZRJUJ&RD@3(z6mMe zq3~f58h^c@VrOLJ5FZqQe)D3OJa7gYlgF(zWYJ{It9layI82tG9b(5f4=!J0s+uS-)Qf z=WgRSmmM#$nFXt|4(MsOko5_c@#ur2U(iHos~Y9aX!o(gD(1QkD9&@E!^_|0ePV`q za{z?MBlq!p3&ARbCZ%cG4;*p=1TAo*`t;znN0?n_XnHi}D1Q;DCx{)&zk2QNW!vE} zv#`6*zxBP{X2$dYnMI#|sKpmvmbDGvoIe$ul~jyE@ijN|WI+DXSe_}5`8N13KC;5? z+}{Ki7BmCB{V}@B%g%H`^N3AuSG_T+$zN2XG`l+;o+DF!5R_Qgd7*bQXSfKpXt7&j z$)s1hH_sC|o&1l2m2V*$^=jaIAR28gW+UL`sd+1$WV~79rF;y}_W!mGR2K*&e3KA@ z8@xrx8m=iv(~m*adwRd`(P+nxMwFe2Tx4H2llQ4?no%!}q5GZI7bWuudU6L5WI(^Qk8D1$RN?5A0Kp5MVG6w=cR}OuZN^(|5#U1U z@?<+_5kEHVI}s!7ymx#ToT6V7B5DYm*-YUes*%xN0>DSjDc4ZFRz(oQ@%N!+KKU98 zf-jHPm)BE`6X_VifH0_7J8mw;*h`5P}<;)N{4d}B236@rvBi(vT+G}nciTnj8& ze-7^8-dp$1o9T5%u08xF&uNa=DZ>mOKc-Lkp3!!PSCDk|VbA7;t=l^O9xKtbn>2NKyO>&{ZQkYI2unjhsH+xxk2BYT7Jqye^AT;m zl=*SJbba!}0Eg5}&_b3UbbForK*)Rmx)r}SnE^ciVMN?knG61PD1&~(CUB!d; zpXuPg#n?}9X9-$5K=yz*{(QA6@G6-ms9^k2u9*6a_sfcCUgH}+SsB-I*iq4*~8Hx z>H|A8qWU-6C2%Lc3AVlNueibbcRa_dk#1dU-2y5EpNixCI*2#3rNyH6txR;_MnKce z>CzJ2og5XVH0qpWLb=^aU;JPpb-VxSZY7b8cY4{OFa} zD)*D=rem<{bnK&Zdi-4$V2>LD_9gMCY|Y&r$3^BHzmAJoE-=JrGe6YDdE3%~<5jFx+?=`mQ^aG)@@jlN-*oeYR~t z$;p8#u%ulAjv;HMkIBczEk*^hK6F+kq#g(a4n-Q3o&2ep)A|BWZ;_y4EG`}iaa8sB z@0{J?mY6NXj=nmW1NwPPHPzc_imGs<;S-Ez*ORrED96pI%|I9$f&q3SWs~WU&zCqp zhw!My`Y)-K9;(Ve&w_K+u!g&akSml#jL_LDez}nYrm%0`v#LG~*5aI1YgPFBHtX?+oUA+430ei#O_7Kn|u!F#R0T_JQbol7}tAWF*?C1{nNpy1@7{2b+R9skOY`IXO33eoZmGMkXU zV{AkW$mWI@sRSEu_zXBS>wHNy=5-S(HUHta8A&$Q*CM&DN84If?ti^arxjGvTLySv zFh&;o%!ozJ;r2m2t~=(FzutdjYTkj8-LJJGok6uGZ2)>L%Qao8BByz@y2hcXbSKV` zj)(N}c|DfWLSKTsSFL=3ptN(->QGfRtu^z*@S=VESAPf_N15Z_LEHwoH3=uIl;3pz^Z`a9&4_%)aees#JN9$WVD67DuNzBk7Qd2TyNb8N zE*ic&x+p!lkmKF@pRVPVk@@%M5^)YqheoljSs)Z`@bG#vnq4;ptd*2K=ils`s^Fil zfjX;{VcO(e{QYu|)bkj!kX{T#Oe#UjLFTF z@ENe@z_WLNo1eXfMIKJ1w|i>MgIuN%UbBeKgaA1kkeCoy6{EsdjWX}nmU9Ldwk@uO zTcK+n7BrR)kW!{Q#wIr!P)d?J5nVp~5eD??uIfdp00ImNFt67WRofZ3&dA&Yxa?hi z3L72`Fz6ZR^(JRq5tKIxmou}SN10xcwy|$!ae*q5lJQEP&6rATTKw@MgvepNz2)&CYYjjn0^HPwc{JNzcm6hs>AF@!xsI{vBeTZ{1Tu`CLUm-=SE<$XbLRt_Cn37Ve0KKj{qaurXpOGTghlRZ1-$-%XjK=M`cO% zU8r60uRjCD(RuwFX=3`j{X>_EK-#YR$`>_PubS-h7=_-vY2|_+*b>s8!qKgUN`V-w zrTSBV;b4+W;n9JscO!`ewf&i_3hpD};DPfoBxlX5HjhV`B-~n2PI~w;nZ5lJ<%;zx z*pyi%kH7e^C9lB>QFfxGH1wDpA^hiWLoT)1E3O<)ptnvEy^<0=Z@*T8w#dbk|F@qQ zWjj^~`JbQUwHnljHhEA{;Yw&h;v-I(>#V; z97NVSTlVuYB30xZBnWb*>eH9c(tqFDP+Tk+v*zy%ZmC*05$7ffIXzv@Zj z`1@nG@e3&1BNts}oT^mtC7k5V7n?=(TsD7Nh9TKy6mZp7mg>Ap{Yrs^oLgP~E=?n0 zFs8HMKrotMcKPo%1<`P~BL3lc$Uo}VW#vq+0bLDYLFjIBY}|L7g(`Z|b6{)n$3X4^ zorI6@R9j069}$!M?PB2Y zh-h^B12hH+Nl6&@lSR|Utv(GyYDov)AKyMOHN=%R#|H}n+d)PSOV)(eNzi(Xe*W%0 z;}_FN)9~$yb#_5PhCBRa9Z;C^X#$<}dEWbkU8&E!6l{W{j=J$X2I+h^u0%!oRrKE3 zmkw8}L|-ygA~T&;5HuLAGUoT?;3_|*iyMGC4R<|h_H_0HNCw{2c2wvjjD4Qkke{Se z1^qfn9I{DCXM0FH&hIz`Q%&Cq4v2t%LP382B;qgkfB_>2BS1k9C>D1UZzzP!Z{Upb z?2S~++tXeZG{a%>JL?NgHXm$ob9K8~-EhPCP3RDONHwqyV^iQ_BoYeHbeL*$ywAw^ zvU~jf2OmG!PYc#)GRTX~(8pa*V= zI_XvdtZ?6ANRMk!HUsne{(*r^Ncz05?pt&W;9{GSERv}evv#e$6|#+ zgI{l`8&1*gU0DimO=5ndeh=kBp}N$h{|fmo84=`^H#Tc=Se56!37Am`F{_cZaVJ#Z zl)OUuH}OZA%YDEQn)v~n5u?4KkI+`^D9kwwpAA(AMuX)roU){zQYKJ3{Ec)3YzPwE z)58i>qlg`wi2>^~{P8~KD0A41uU2*Ax9*!`J0yQC0!N~Ss3;h{M)>8?A7O46F}sC~ zH`tI<1O7Iroxv?2CmOA}qnVxvV8jUC79M3#U63Az>k#BWM-M655zvg(*sT@bh6RclkdlA#a~OQiEKvMY@|^KZZhJ zQ;7!z{-`rHeFViv2X%;#tvF%SI0UiEYw%>qlu9t!;tN4QnmQ?~2BEwAhs_HO@w@xW zOVHnc3O}f4ORT?F_gXDm!CO-(Dc7vmPB z@O#8%*JU`Am{pmk{UYT?uzVn|=e2{`Kq)jIm%GsyA@j}xlPVr;Q^VaKg=hTVx+Dsk z{~Q;RbEt&=Ru=E%Z_n4a02W=NIe!!v53i(xD1C_Qf`d}g!SLS6iKfhSj93=A8+f?R zE5APEdqhqgcqPwR#G|k)AxT;xtTfaX;OR(85q2&R&}_GA``jGHP&I8eq=n>^*A;|p z640FexMmBc&4Tp@?kO+`{-{%RQ8+ECxjk# zmyIv`mGKjy2AhHr8sB_dhT}v@5`vo{9aZF02*;kN z)Tv?<9c8(Yb!%{k%EdHb1Sz@V#GKk8gmxm^SV_}2ZSvJsea*UU8XLcv^w81*zpU16 ze-vrBH3SM1FtQ4TE3WE0)exR&o=A-$V-JqtA65l92TyEP6~lbPO9MEEK#0kN{92?zDJ)GRhO{@<3Fz845UdqOw?({FkS?MhOIR@SVHLaIgyVh&)Lq;C zxHei1o$tkvUq!Y@oTJDv_C#OHb?dDJ_EIz*U(IxYSY%y!VU>6TOaod!`b$!PtlzfLQ2HpfeKWwzm(H zPGwHCh9x+A0Y-RBlW@4Er{b$|)8zbIeZAN~a0RsqlK(SnD*^p*x!bW-;jQc8n+6lj z`u5=OYTgBmyuIrkKCwEnHx_6sinYWy@^0u~3SUKIxb&HTGl6Kqz(x6|j3Fy~x`8uF zCv=~Ld(vidxb}HCAzhQ0OeVBdC%&Az$GJBmvml;O_I|S`M5oiY<+HqeacT@fo5%Se z;J27?h9I|%q(`sR*0sOQr{vV>>^yt^@#mZ|HDbkcTSNbx$oO}dh9iqbPc7vFc9mC0 zx9*|&8GFV-0zWc=u|EdJ#&u;`_E(LXqqT`sLQUGMsrC)*IFlpV5nh;ukmM1GKtg`w^$ooHql zAyux3FRn{V{Z0PM!bmkSXG+sC`GGpA&K!05vu~+4XB(QeT<$@FZj)(^_LomJux3ED zb*kqTPOy?2by9GsNIR@b66RmwY7#%(=@*23RZlFodaKjkc@QOr&_FSAUdGeB+Q!)b z1pjC9;78rCAOCnaPM)*Bu^Qnq6V|Rz$do6bd4!1G9jp4{=fyWy&$p^nDQ@g7P3jz_(0A_%CkJ%#X)NDKe{S{^$K|~-Py$R~Q=)g-M0tF6#Qz~9jBpfWWC||5TqDTH{{!ER+YeM!<-V`+N*An<>T8BKOhAY();fZ?IBeE zMJm>B0JP&y zlY$47d0VYtu$$K0N!C4g(7B&y71_7g1EA4FcM&TynZE|S$MdBsGg$~?q?MybUftfW zPN>T=fJEeZ&EfIq*GMzc{@xx@*m*Dl>czaiLuSM-U^X4ddwI0}d$vXb5Lj8YrImnU zaCUAkmO3>`oY=$1L;jMwkMiX}{DqSp^|=xd3FNEhi49Lo@ZwE-|F9uw_h?UVcL0PU zJZ#C}Bn>;ITI-t)b}_}X4VIL0*!QH)Tm+KPqd~q-tXMTdoT%P>N3h2DkdcWm#%HFf zTtn5a7$&=6YHU#1C3OIOV$!7(?vt+8_gLLuIx>nxFME6Zc*+ZsezFJWWw$|T~1Hv{7Uy#n?AmP zTY3@D+Khnf)V%|A_^NboaQp=s=@{ALV$LEL9@YXU6IlTZfcfUCaBU)<6nXZXGrZA} z3v2HBE#BE0M_;B*RMNri6OBV3r?QWb3+zFxVZGB+mNsy>MAojIAAe3b0r-$dyJV4l zH>gZzlNL6Ag%@7+dDfm1S+CDpY5U}lkDi*V)6XB!z9gv?Tssr-zW6`pHvZ16yLf&S zv906&h-18BFl)t=`$upl<~nopVKbWb+Oem^TCv@*$}WJ5{K&lkkhuidK4Zn(J}+hp zy&EO`3U#~mSZiPS!T5Z(_402mvWt|nKl5$W#)H=*iL>BQD=`{qC=9kf?p46_;_Ca_9=EHK6X3jP0lBe zNHr;VOGd77wo&njt3mM99u!m8_o5ADD;Kg<>ssbsBH&t zVXnN9N2@~Fq@n6T>)=%BrVaZI@d{eArPLuU+-boJNAFr7pLb*sA+d{fKnSI^!=A!x z7yC7r%}N9dAY}*vqk#I71GZznFMl%2lsqw5HtO5Ofjo`Z>AGy-=t15>3D7W2W8QhCZt1lgW>jr}OTG~#pC$YRW{oNcDIPapcBkd^##vDR0`IOPTKB$GgJ$*c zuH~$b$BP?rqV?ymGz|b9<6}}fh-U-$-5wxf#~yy3%6;cWE(;?zaVu*ZP4Sg6*JOEU~4Kag*s{6>+#5V?{O13F&jd%Uys~J5L#bOW|~i z_Jut95^K|{S$G9WJ0S7Ko0(c1aLSZGenrjQ}<|}kdlG2GPgM(p{M@r=r%rNj} zb+N}uMq!!y=z+!-P5CPytM*n(lXYJ>x=yPr6QGf@63ZW-zI1SUVIwWgNFWjefmH!Y zoPpO~>b}FZXXN^yWu)^cX4pN&&?&{A%L9OP7(s^H;%!O*DpoK{D85xb_Uw#&ctKbu zGOK**K+CWsT3zr?A{jV(HUE>Yu+lB%R2p`xCA9!MQLMB3SJO_GF;x3tDJByD=QAW# zf;}m?!(y8iPGLWFFlVk)y2-oiHOy;anr$WrcYqymu*1e+5b=QGzRt}wJTbzTi9LfDAoJI~WON^_>mgAnU6meRCNZG0fz4!pj#Z&Zt)4wX{TtKUZ?Sbd_vi7<-QGnd*am`vI>pkMf6Cs>_Y zZTF1hMSB*eAY4;Ebiw}Z?`$|q^=$YBD#1Rm200pA7bjTHnn6fF556| zS64P8jF&gY2kd-vY-_>#+cWKBHI|A+V8p`2Y|=0DPmlLQwfrT`GzJl?X&Kp0QP}WT zxD!;k$L?#|+*1cxPjpieQz&_{RHkGTZu{P-E{*@%F?WetFyTb~%}E%hx_oTg=P4}NZoHo2FpgsactuP5eRhI&RXMrair##E5aawtAQZtP zXC?`n-#zQ87NnCVY`g!r)*7>Rhn56_v`IR@5b%(PoARAjGzHI zp1l&`zM&!r82jC%hL7N3$k7S_^VD_#aeDs>uOlTphI_Q*tR;`^MF2H2&RQZ+h6Mu> z?6)Mq%#Jq*f|ary2)IF(-^E74UvpJlOwl&(?e;|8f7Z ze%{8~EaVBe2%MmM>0g~dcRCR`$Y96WhFEor&&XQjA82(VOiiy#^ZdEcNyT3t>rqtu zW%2Pv41azCtkJP!lygo)WY*+=uXDf5#z|f`&KFWGoJ+kagd3gqrL^= zPmUd^BsND?h%BjgB-ZZe?PsoWN%C(AxL|XO2odIIG8#sn_m>eZLgZ62KgDnrDhI9& zo!C3W!KK;deLLQyA<4n5UoTr0A#$Jqj-A}>2pT8JX-aF*zJs{IJy{*0-fBn}{7ugu z;l2{Wphwv81&_Z|3C5O}!8&mIck;&+BK>mN9uYQ7bGY4*jgO5C?_u0&1~T7LI7Rjo z9Y%8pLkK1vSRcV8iMxOjo{~Hs3EK4i_UFX=!?@#+IHwX8UwTh=dULBNpRL{>m`^@@2)}nJ9E9W((jRit9_o%dv))LzUGu)VF#(jNFCrzS;^#5 z7_ZzvRIbu@C88U3C0_NK2T&f;{*@e2_z-a3d$oP+am=N=UVtjOVEK~L=;u{7(_W%s zbLq?W{W{a6?-Y3Ktp|rJzK#O61Za@uNV*Lm*RhA>0IXiI3e z0YSTm_4h4j0cy$loSOBJR2R#zn=-|kv+&OoS-kWWAm##Hm|->g|Gxvb7Y$!LZZ^v5 zin`lu7tQ31ZjnL_f^)%y6-2V#_6-M>JQydFOB7+2l7|zl-!Uh2Pz(|!NJ+KAhtIJl zR0SyqsmX@SdWT97i5&B>sIaXU{Ds6ZYz@FHq?JXpv5V7H_B*-QdMzi<>0t~W&?rn+ zpPXMa>jSbK#XP2&R&BtDX061;E}55o6hpRM46`u8?edMNv~cryC4fgd9KRF5R?3nN zU$zFu-kTd!F)@|vba>O%MMRH*gH@MD;R zmT^G<2k80oOrtET+HEb$lSPJ($#U-G6!0u`!@R@J|eb8TP&MC_*cxm^+^8R*GLNCMK}F^7_`hGXi$qM8Yr%B zs`gd$bIZTH;Yp0NclulDd}lCrk~9PP&EArn-!7m40%jJxGq{AyJAda?J@)7Q?jN>= zGqyF1aJAlT0B;k0xx8ArEua(m1)s_)rIrH8rI2D>~&YVoU zq204yen&->15cw5bJ-V|VLv%3GAxK}ah7ybPue6Q8|bXz2;VxLr3TDmB@0mTcWP66M04WTMoe0*3OI82!))#PJM6Um+5gVQg8w1 z7+apH17N=n5K{cZUO^)aNSc%*^rRwsB!uj>@ni7&CP#hof!Ho_yx*3Hpkuv7X!=9` zlD?N(q6t|j`W5(Oh7-6%#5nL4(P+j&wpO>gJ&Nk&c63D1S;sXnc~sGWLXQWHRA9P7 zn;e@k=S^m_F$U8I*QSGnri5>az<@e4eemzmoD-1zl?9Ugu5HL5+41`)>bfu};B~i} z%cQ0E_he}f`TbI)+TPs{%7vqkw-L3;m`S18PsrcML{XCm63jWn6qts?DS~aLJ$DZ{hn8ll!vieAg5GPDxoz z2#}sc(n;%_(A~N4>^%S2T=7J>+WF5-$W{`gmuBJxX>?j^O++H=Q$FT7YS)J19i!brV2 z95Vg3ovAU#tMk(ajeiS?q6Va`A=NF`*N(gP3B|MOJk@IT>6U?>KFb5)8foNI>T;Q=6-83qwc$&fhMPOn5jO$pUx|V2Wc<-GB+I9+>J+t8|uyGM(> zFL@Nk2Fxq%+rS~V2}pUA#eZm4_3#k=G42wD5!?M_IK_IMx&#bzbSygg@!99M_gNNcEY&}Uz z3K$V&x)@qfR5H1W5B2!p0R{PR-32hWzFe}-%LKLldy|KtO*;&d4=2nO59lPiS_bqE z|L0HxWx^M@w>TBHt5;@T*FW*rs{GYIQ%edVXgsE;W9-txwJ6!kZ-uxoXGo}l85hJN zbGmkyudn79gPH1hryoC2HkK|~ayFyCAdazD)rR&RU6*V|_a~7_*eK9?0@RPo|7OP9 zJmADlfNMp_ah!&8Bft8hSQp^%r%--cq;3+XS-R=o5Z#aB2{3*TEoNE@Q?pFd z^Y7i)#>v2S@`@3NcbO7hh*8ZJz(D|=Q+4~7> z;gxznxiAy-QBM60T0~JPZSQA3_~4wQod3`l*!A1-U2l5ZVH~@AzI!UEGx#>tvPx3t zRx5KoRuWoq}=2*mQpaSJKY99YLQt^c0)BAAKj8!DPSq-u} z77c608F!cA!XqxZB@#I2qIodS;ZY9{2h$7xvi<~I`643c#g2|sI(J=bEbROwqw^7} z3Em3Tg7YOOn0s*(U(vnB0*|czJPQz8xJzBVzA9<3Ufen}6tJFuj~1)tVMz^5w_-xSl;Tu+zfuT<%PlP~G% zNk~m&2Z=0Oe7aI;uPQH{;8=hz-iy)Y;!Zu({D^tj%p7c@h$=2D&D@~7+mb3Eu3nok z9NdoB!rU6qkS^%y0J`!GCDMzNTUqbFa{W*k7H6XjuFl}?u=;4Tmtfx zY@1El^}?6R7%M!Q0ltB8#Dhavfj5;$`4LXny)ObINH(u9vwB+ z#jLi;Go;HH&zTi%`m_gei-!FR*;x<8G}T<0y~ovR2IjVQp4UXwuUGg7q8;C zW@U7&NvWgh+w4PbwEi!L`?uo}&TmjPW@~wwQNoRUMc%hJwFTXBX=EcFBRi@Gf`Bd} zK`Ys}(>l>hR&=f%?pOi&?J+kLdMjqTg23n|D#wn1| zC$08X2C0W8uFMZsve;H^qrL_}vfzbbD^3ZYi*QfYGyzh3NV$UzW6g6sY)Ac6qum5} z68K;Heblp-tt6-x1cbklb-AtWu#(&Jrn3nJY&f(8x&eiIfmBJ)K+CWLz>WC|9Dxi6 z0aCYL>hwieBwL4_n|#dxeGtiKz1?f@V**Dzj^@X|W-Cn&6tB<|Vv$3=Vb8r2-qQaJ zy!3)4ACnZ#`bhJbIT*PB=2$myY46iQ~$UcWaEcC!HJkV{}qPm4j zIA&c&db7P~@kO8daRVK~yx2}Hx+K@VV5!Z0CDlFP)R9Yrs6usPPf7RjMz0DKKHf?e zorD>7M+;tJC4F)j`OT9pJ#uU1nSf3&6#blTSI!W zY*TIet4w_|d$!;rA-F@L*PLUvhkHsyQ)%-Mll&TJ1akv}~iS~*cS7yHeW8YMMq zxAMq-sHVx58#(TcR4-jESIJiDTynBz_P-~;!_^j4@4puxyVw_E7P9(GT2wlh^%#*? z`KM4&Snk5@_y-c~yz)F_m+YlEO=udAc~T3C9IXlGmCg!R$29B_jw7Di{2H-6fC0bp z_m%h4+M+28=mYO-+CTh#P*UI5j*6~jvm;o3*KazQurmI*yh$tHBnX1rOLZXqgM1vS zi?mvvjFPE=EpQHz?yxXX09{ERFeHN@QTR?D){lk&yJ|_sUnch? z&nCLVMfuTGPOFGIQ-fYY&rX6l&x?Umr#iaYA3d3)Ns?xp&omJ3B27b?o|&<^7UAku zO2Er>1QY$^=%@(|x$#G!{##hUn*M9;md|E&oekC6R6|Vd8B_`R_5tOeShb$c8>MS3 z1EN1vrfMvrn<4K_UVw@^QupmMdl$Ue1KRz) zxp2Ux*^0jA=!atJB?e6=my6AU(cnt`(~WO$rYnD4hL-l3Z+GezB&5>iRrxaC7GA_> z{R@T9|1S_kUJ}D(1Nje`4%4ryoJvBo6(4CV+xb$DYjhg?r zu684Coqf7Qt$Bh2Al-nEEpYcYBa_lK%z!--t%GjEsa#T~feAV)F=b337U_q;&0bsh zgW}bAiFm$9eQ=hKdE0#!;fiBf^jf4rbVxPYCba08AC! z1fl{lZCD`9sjCt{Wh)KV(qQ9}TFp@@fm17E8rr(O+^y!$4s=| zI)TP#SqJs1gG#>O%p=vG-7gpi@o&Hz`NWdl1Y+)f?DMuMx!Y*EZa!i4dK&*~ef*<_1hJWUR#2E{DhqQ@>ITR?k`{c)Nm(E ze9xd570Oc`8ZQNx1ZyW{{C`$k>YeIshG8YI03Ov&5E41Z4iGLyJqc6;Uh&#z(T~9D zoEKKI*Tx8Vmz|C||16_uSOO|(+CM2BZnBnR5Ae#dDcO7pT)htk|B^%u@BsDajhAH zlDWhU(&{P885Lsl&#%Al4bxxcUV9Tb$Zy2Hx3Nk&BsxN2E^jY@+u9>jbu9Ug^uO;} zN)&QyH3|QGO|hBx0uSUaWQdV@9*_kRV`?=39bx4v@MGjdj3cvG^nDY%3@9N+#BtTBdqxfp4pFCo5+QgZ zOZTAf?YQ@E?@@1h#_*f{SM0#Q)%t}RcBid!)~5(TEpV~v)wDDwQIn8V0g*Ju0wO+} z!C{PUEs5~mN2H3wtH%}Be`dv-z;|z$Lf8_hj>@>{lF4BT3uMv$0iuVbI+?4Qdz;^3 zL12#Jngd>*%^vo74(f4fw*H|V^q5}^0Ty}VlMFzY=9wsJxb}3V-v<$SHUuFYaGR^a ztnNfcMKS$G0Afl0^B_=jdBC-GA@H@MRV4d9mK@Cw#8EknQ6FQPoZFi0{BYODzzt~V z?F$EvtQ3fQri=rxr^N?5rE{Vcmdn6#)~uzE~sn(9KR+RR7gCddwk(3i~3?w>D7Gl5KX{=MuJ_#wod|umOgj~7@erK)0_|69+ zIA$&#_alt{GE?9EMcJLt`cMx7TdIOkq6}x0x)2SvyPT2ZQ>PsEG8h!HarZ4XiTxMb zSHY(JD3M5U{J^wXB$4p@O@)xjC)3Ho=47`kz{ycBe*9;p$=a)HdKEW-x+g1e%FoUZ z>Fo`WJ$e+VeAf(s07{Ky=gN>P6rTaKj(LO5$*x~73PVPjZYqqh`ztBlDw{3LI@>9n zsG8ADbhQQoKZddfN3@+tW%_BR-F{8zm8%$o+r*g$+Y9WRzbg2LlW+HpM}*?7Teh?^ zDoAkfICkzfT+IfOGL_Y1d^^O1teAd9w$cX@eujWZTD*XUQrgn<8c@J0pTCE|=rRqw+oCK5P&xC~bibi? zL^w5Peg|{Ansx517jvrJcN;5MUB775c!qP+!7!~+Y=OrMa4rY1_119Wk0v=tM8p|O zRP^Dv`HTsoDE#&fqR1`m{OT;jROxmb_GD;&k`Z=Vl&;5CVL?1Pe^{Cz7{RWn7E4iM z`i%|LQmu$}+SEW8QRZPdqn7W^81F@Nzs#qHiFeo8s({wW>r8N$$0yZzi>3lOK-~H# z`IZuZ-qa!d5+urlGGGCK_|4V5TIdUgCyHC)gI#v}{sG_yK}~>jHOgch5U@L(NY7GV z`ePHQKu=>*cT=^EnCN%S%TNQhk_Gd&6b(FCn>*MEI~T7r0O6BUZA3jk05nTQnURF~ z#;nD{?$+U~{ND#&tfHBKv`fF@jsRNHpq(103bSqNO1%6y&pJ&a%e36}VII2Nid9mP z*3e`BwRWP>@RxNeMj!WeYb|T|!B}Hv3NNyXVeIfKYb4uZ)P-$k9@|2y9siNOWd~W3 zJ4~jJh46c@#(weWO{&kP@+UCX5|BD>Mg%QSkC~%WyV|uKi?X$*ksy+eBxHVTp|9dd zL&^97hk0nRiw{}>L-|dcZG_ZzzO9m zGXBRhD_qF}G3xU^PQChL`oveb{doJea%eI+FYVy6`N;CX`?}fF_w47^B~tJ~XS*~y zZX+4IPX#VunlsghKgZ0&_7HCe_DHVn%;>qvZLwcP;)ABX9&{@DJX<~0;_L)`g}`jA z_>f#JWq!Im&LnJgf~^{9MxR0uj~-ZJj&)3sg{_!#m>Zlwx=Ya}IKBZyR6IAgQ;1K& z&tK`OwvVs<$kW+Jg$M$g7pmXp`DiTrWf>0owayN;I2p6GPpzM$d29r=b2rBwbgq3M zULCw%y7?J<;A>2+Xgt-`b>|=*k`uJnD%iSI3s>(P`AW}nb*ArnyoyE{G0;0WlK2ml zhLb7cRfl5j+uWR!18DdqA|oRuN>y^Es|EiI{7StpR;pxr>s zHm=JOgbk3Ez*Xw~j++iuuq!4F`)R-=pg}ksd=O;Wc(+86^stvO$KqHQs3&2=5Rb`@ z8qx4A{uGXf0;7Qtg(n$PEoOI?T2DtE;E^c6?c|al+*f=ugNZ;O1{G#7lI}KC_wIaV zMcOHb(b#k?IfkupHKhaH;IO1mGbLanEGcqh(LR`aJ9DjuSVYC>B(MdcNsn`2)m!QI zQ2j(b=^43Mx{UY-6HchLKLaolFDSri6 ze=0^hJ9AH)QrKA2MH+YU2iPF{Ee8p6op1A+pW39)9bdKgkZYik*oqTA5#BB@&Fw`X zFaQ+Ds1={SW2`4(Rxl?N-B&l%^oMRsHXX2-HXDP`{d%xS*#O;WEH+WvV4@s(6ky^R z%B%~PX~p*5VI}vyU9ygj8zNY(_F&w*0>`>E4UJAKHfG=z56Bk&GZ!MRdMe?KICdddc%c00Cx#r)*-ENTS zNHZL(M6x&BX*T7A8ayzD`X0V?AeaJk#VybkI6b$78fs__(|u8{--pv;;qYpf|Rt?xsWFzObFe zy!U$fj^nm0g7Iueo{W0_()C~f`9E@GAs5AnZ{O5i&U$c3#F?oGw>K<0wgYvWHv87D zvs<2ZT^uoGpIj-bmj5#Pqi2SQXL+I|Q19N&%IxR-CA+wL2wd4E)YpZf9tf-WY6Xz9 z#p-Wn(EN@u$)_X)*#d5!coEJ~CQujgdz{2**j|q?I7vNQ3wvPo_Zi5^q{QJ)=x|(WM(F; zEDqzzmz=4{rP}zT=-n#PfJ5CB|x+bL}hR`xkGrq;ciJAdouhkeoAD|8r;V`^3e zS=mO(Q#l8aub%DwryO96V&vch^}LU!Tr-fLsTlz9yf=gt$B?S)^)bhe0|1?M)@2_R z0|Qc3P&Ndf1R2d|4q*DD5%UZ;qs+28?v19>7fn`oJwpH;=R_o|U?;!@vR!NaEy4)K z#~>G3j6}s700WxRy~Mh8d<yyf%U~3)RmA-K|^dfS2_sahiPb~)dhkxNhoyTxng9Z;IA@Tq!ZSfb|yB;h! zFxjk~BR~SA9ya>=i=xJwp${cAY>7O~^`#IiVJcmOjsPTLFN2!}AT_S@No;>n|BlFBdcT&1GDu16_oF39fog%7S|Pd{r^gCL>#Hs(5L_f}WI|^eF9} zbJZ&TwM6`1RJ~l64H&*4FZw_h;%p7`bZ<)-9srIk^|D+ zuy6kFe)n%CT_pC}RnZgZ;`7tLnt%VknBB>Gztr{YpmPKGB>IgMSAVsJebWWEOZ zO~KEVSLw!<10-uP6#N6iTvtG=3ge7?eM`M1H6;v~9}xWyYT)S##5q~N{*%Q=Z^TOm z_KrfjT)SVOWiR1@-i3Mx{R4o!`Sz}E{eQPa$upoQxN2bT{%92^G9WM;rGIy*d@rl(|`C{V!Sisawuos;A`|* zs7yBNmlbFI4MvwHvsE_GX7sb+G-S*tq|5&%^^JqRvCNOJ9?KKh-{BEmI!X?TWC#YI zI4sRZPzs+ZDbWr?P;^j+Cc9KFdsWQzo?x^hsDMA+;WK(ij|(xZ%!)<6MpaRhAC?JF z=s~Xtc3FnTlF73eY81!!lj~6UhWk-7z%8)Rog#o$y8aR|BKY5#(oXrvjDdmKm>2=c z1AM|rQB<7g^F_L(x>D&E*xM_EGn-b3+B@m)(M>8zuTswP_^jDE!L#&8eRR|KjLD6}IyUMv(E90d zd;7T^un~lRCEPhJ!=+1YK${LR8`L%RHJvyRep~9XUhlawoe_Wu_Ingc+w=toV}Ys5ctp}ob3<51l^XiXWiR;Kx8UT$@(N;>rMN27)=-NOnBjq zi%3w5ZIMc$h6(q95nG?Dkl?=mfP0gLf=1o)mv{9-dVm!1&yCTa)MzgYKT`yRvx%#p?+<}DR4!mV*#KsheJ`1| zQtp;rMzw6g@tE{}U_C&M^@DelZRF({7LMx6K32lXi7V5E=)dnpW5^<&yIKMcvtwd7 zj$7u)U1!^)H8!K9Tie?Ob`yEF;#jBvTsU@q-e88OP{8%9&Z$b2l<%55?ZdEAH7lim zDX*HVX>>Vrz6Z>zO6=-=(I&k=d`di_uuFVRB@67rI6V2PD&zB1w`W(%p6AI@)*i$ph{2o)JR5rdpYd ziNreKF(y zkc-PS=C=$B*GcQ##vCVSt1cE!f>YAs%l(qEkfXAWY>=`&AsoVWV45o>Cdj4#Q3t}v zXhBeKQU1qp&RnHlCe?y8E7gwna!$%$=SsO4K7Hi16Y`$@+_q)|dwcs=ME#%k^l&bt zSoD{n$Dej71b13mZ%Mp5#K^oMK2Vi|@KL|j6W@`2$j-P&cBca65`#GTpYp~Zk{-Y@ zF~Hw_ow2cY&AP>bYbund`$2~DY(`=Y=IQCl$xu3$w0{os{adRKNPt-4JZRJXdI#!u({T-dV2+uxZV-Q-;+g;{M29im>p`&)xo73gPv!RWo~ zUB!A0$G$eiLz(c{dGd?rI(8-Ea_}KqjBHSdXk)M2g1c_krUzpE{RH(IxVcbu+i!up zUZ=swe92M{95o8yx%;Mq*WJ)5=zs5-hv2JA6iIOMU61FB9q!y8T|c7lw=nZnv)-Rj z6lN@~J>8me#5Y~Ly5&2tJbi9Ey?hY*-Et3r)Cpd$5fq(3NcU@D1zX5et&OT~kUy~5 zld%Ut-bntp>~oDS>I!6bZA+DS*cwx&MBaCked2XmZ0c(JRA%*I205!C++c zdcgd=8-NL*+CBy>J?W)+xt=*i4{bk>E=%lKQ#MT@%Ujazg`imv9>eEcp-Ajn?fc7I$|3Z;yc{u zo&(icgzamZ03C``#@FM~clN%&f;#!nc?vw}&lhnR)EMZQ@Mr^nsmn&z!sERNt-m8i z;5ql)&A1w+mjgS4zk_x?2cu~sD%I801mZ$pU@&#RKJ89#Yi$P-S*@dY0YRu};MZPo zVaNA8{fQ+eAC#)@_1n%@uNr8=yeu8Q!BEFo%t7<)SZ-NvW(WiSDsrG-L2hirEG}yF zsg0c{prv>MuHS+ajzV!n9k00W&F`^$v0CB#Jq=_H-Ev-@)JT!!AvUIUiAu5LQIFm2 z34~MV(3(ISKTY};VuFQ5Q#w=q*sHfO9O)89jGnACk8xkq_w}XOaL!(#eR6zMREUDV zUKXL&w)`-edh(U!(v>Hb7lYD>L8=zGIS9dSVvz@9##q@-u z8{)IuFh;P#?z)H{R!zW{$Oow1<=!{p%>ZTM987%S-Fd<_bt|a(3`*> zJFB!Ai>&L2Lc=^+^5UcUmR;x<+ z^F9glu1FvK%D!}eRRuoaSmN8#ijd4T&_wdK8?9e3RGj?na=USxd`mrkoll07Yocvo zEPr~tJbHeJEpfe?0w}ct@0ziWiTgtWxITQ8W((MPdfu^m4yZ~UtkC*FyEB#<2JCK3 zc#hfR?gERB`5A0 zd-9!DcokVMQeNJSj{u++H(-7IQmqKdM2M9!8JiqGa_6QJXi_J%9?cEw!jS`-I1L)}#*=p8Ty-55me_^>&l>s{J$iV*eT~Q|C z8*%}uLM_)xdJckO*8~XiPHs`rHvhmez;243{zlG7hX$F$rIVmWMMY&vIjozY#kzj` z=j6DR)3}L;zJMMYc<=z{21mw2UTF3&CxnRns$oHj1LVlyXA;4H_caYw8o?y`YEs^n zHUWHV57~nbzV8i9D>@WZzwZiX$HQRvVQ)H==nYvhfi8kIK7#J2cmXCCK3DxAr+_sZ5$aLx{iu3FsMJ$no`b7!j41t*%X~)< zLI3K#9ou$wsnn*Rdv42c^Yt4iepK~4dZu#z>Chzx;JQIV7@;gN7loXAMQiI%XpE>$ro;)$@|stu8?Lq25SJgS^D0icnF|wMVAH;k-S~#u zQDDc9^W)#c3TT!ll92+)H+G$v67l;*D;?MGa*bC{QS8cvubQib+cfiIo8~aC+tvdi z#%u}R{UNJu!hhfAa}|HZbM_oaS?pS@cZkiVVT$MD1FsyU^lrWg%!-wh$8Lbzj4M)kK~CjdVl7E{z+|^91Z=~ZNqbQwcu}K0#2A-$YbnEOG?~h%UowF2F&ld zE9uclFN#xf1_mewZar8yW7zO7Vt)__BOi?9ijq1!i&ke9EU(1EL>B7n3R%=+Nn;tzzb$9&)Q@(kuNuyov4`pUlbyUnH3 z$B!!wbOu+mI=rj4=tZ+DKU=7=isaYxI#^=0vH1UtANAowdV-FwWu1!+=vjnXl8PpD z_3q`Wg_uaW$i&vUq=Sfr+n#R=PU4&cV`}9X06}M}?(C8`0q_pVNcc6WV6-TzIISr- zP1H+$d_SJ`deT0GF#F6#5=2QFcKccRXl+=%UKR_lnQKX#ZNA7gJ}y`cK5*zHY);81 zKxK_fAsm0m1w#GQ(~$R9X<-%f4K*$6eUa?9axxHmI{Of_Q}pZH*gb#O$6B)SqHP(( zvH$m89PL|q!hmsx)2_1n39fnIE&MADUiixuW^78+*6@f9d@*KM#?Jqx$9M43GhY1+=}CI zLUN?KQ8z78-FlS84xkSMZYJ;-9Uc!Jc@#yI^4r&jY&H!qxO^S117M@z86+Gm>Lxzb0o zcgc+iF6~crfNS1AZYWfmjjj3%18A8CD^H6#BWG+NaG!E0^f^bQtE%&TH@!cmyPr)U zq$$6ehXZFzEBEEQ3ks4`P{;~9ke7628w_0g%n!YPbnh~)=PK7a2Wfn&bIMMU{Ymcif4ez0 zT2f>y{SBg%;7N=%YYE&FW8$`ifh6!Dwt(bF%bs74m|(x3{1@jC(_813D~iOwru2cp z7#vT@5PS1tuptq7v|P*eS%zNutO|s!LvG)L^31hg9{OcS?i@9ORFEr}7A11AfxLxD zXHtpR+<@3&qpJa=U~hS2mc>1J$lX$x;~=Cg4sdxC@Bv?{5>E_cw*oAV#y56s4WluL zY2MA7eN0iyf-5x9F2Y>r!TtO!T{X&Rw@}_f$L7>`?Tdu$T8mbSa7F&>>{)cQH?P6; zcgJxHFax*Md5NFlawqdU^<5_MmBDk8x}P>6tDb;kNjzrbb9?wBDVWH(<#%cQ*@(K@ zueh&jnb3ZdW+i+@0!1c-XyiflA3<4_3AkS^6Y^#&4JyXfa zzs2i8{8XfJ02V#}4juUaDW7m-t?RaQrF^=Ii_H&~$hdkjm{}u(D;LXMF2iS*S=1#Y zIS=@Mo#crx&ywENQT&jX5KVn&xDk!Qw4K~n@bU&tLG6|OKa@*x>A`q9mw}&HdSfOY z<8NRoxFQcd*xbUp+hGOKbb%Mzc37q9u41i!{JXUJ^Bz=6J_n!78^0isVrU9A^EXK5 zJ?4q&5q52dOvkwbDOiWFHMJRgW%XZ)r5Rs268v8bqfTtK%n)D-geK{46To?yxmWJl z7OjlXOXdFGb^?Bz15tohc^uZ0p3*v`wOQK@Yn z{KL9~_>Dgnv=mY1CtA=1_nc`^z@)|Y36-Iq3dLdO+P>z`G!D@gkiRHR&d7HEA~a9@ z3dOi-fuVqPNjSyT4uG4yYv(9yq}Bl+nbkvETXUiI5{u>G;`($98X`#au9r>E8MAEoD{1DMm(!9Nsd%iOvknD$>i0@>jYPT5x zPfIz0MI{{tRWN0VSUr8Y!rD_OWldk!f7b(F75ycge=L=Mzfb&hp!P>yCHz3|V>mW< z8v)ShuAnH(UIE@U0Cu%G2N<)ksHg*Lcj|+CK(JtC%#PJ_;?vrHQZK406g=z-&dF)Q zg!a{>9Y!UrQR%%TWlI;VD?zJ}nfY1OxR|S(9}YI{Yt~JsrQ4e@Gso}cb8=V}xm-z} z1t8%{Y|@n2GrL4>uF z6L&gTH{^I3(4YvkO0`7#Yel%H7vW0-8EG2;D3rNfFabqv-X3C0`svBfs3qrCT(_7n z$F2aMA_aqr8UyG{cLtG%9+Ilp?&del5s(z|7#)mALzslINcL z_p%4zHn<&=3Ij6H+`z|j3il$H2xy|X?$b?CYV4EhVLs1j&m}?bt}q7V?f-8R+cE=J zV53b<`9ox%_o}!icyj4ny!zH_xOyBB=)v5FXMfKeRjo(r#^pvfT1!2pa}+kO)O54h z;M_+V(&|xQW(%4l#(FsODBtDg&`x?7cN{kJ$U>6CA_nUTSOoS~Xe0IBPLm`YvUwsq z{(hTQQ^tNiA|V&*iBNnJUl2wicpIfHIQ7W$Im$XJ9>vms!8`3co|HA!0(XCi`oUU} zuDRbk1(qrkl7Y5g1A zM1j{|39JW61p7o_vLGQI>Up(Mw3NU9)F>~(1Y{kk1n1?(yBeGr=G#Syf*uILnbTTz zu|D|HS4hS7@eByxq4rZfq&KZzMRkNcuJpiwcl%x1iOZDK4;xhu(92gLMan{ck;892+v{dBk+^=lS0dtsNcuO*;n|*HPw=GDh z{Z&1-eB9(l;&JaPABxjUG_Llw@VAG`{p**jJ4sOjBcC$#Z6&JqZ+TRs*TKgDb9-@z zr~*jeZen}HAiLVa>DeEb2zL|GJ+!6~mG3PWiFjKG;1u1|sNQ*N1QhVk&G=cwlOgd_ zU(`hV(CBU2aaUAlkRXV(DQ&r>1J%^;M<1MLFS*~ICC2>T`oh6woncWo!DRic zZVl<7mF$P^9INt8#GW6VG^zb&*;yY26TBUPHlCA|9lQ34t?qq^%@=VX8T80~cD|<3 zNk6hE!Nj=rae4m)FhC@mDp!V^1l+^<-w6WHcfjutdkNg9zi&RwpzShLz5D3!=?bOAyZ#Uj*^-@9aDEO_ z0==nZ-<1LIcU??xPnv@k{Eqe^AVZEfgOX3D`2TDA*1=9|N4fZcb1>cNm$HAnoTdBj+Vs)R56ML~K$7$!DD5 zt(Sn$`3|68Mfj<&D?H@dYoVlpH~93(AnZy%B#in=Mo5zM&Ir&i`!p7I5i$WpV)Umo zvS6CbyUU+q(sV1BV2CvoNfz(rp|0ZB1lZ(GFXzlguZzDBV(2s}cALKgfQtGg9ydi= z2{XWY*BeHOf=k=rUjAO7mR(z2fGv+9;+K)41dOAmp>8j_@*UbAuFIE8)-1N^Z3MfuQ3zN9IfNV@ zWRnah1@)05E?ldjS=>Zq6hb|BJ>&bOo4Mt(1v%(~$~>WLn!@8eqo8+(klT00Q@|Kf z`Eo;b83cNDECgpYHeGg_=j~u1;a?c&ViZ+J*U`<*G_8&|AB@wZ_WiBhHFfsUh8__M z&98zybtv2G!&&TxPz7kVWZ0(<+l(RVXBNzKe$B^)g1{upT5|^jZ6L&6yh)PJ1$%p#Luy(uNI!?+1MMyAZ zC@l9dfdmPJ2b3cmh?Z6ydV7HCIH||Wp*6q32@nhbCGXqtWKD^M3xpDb_*5=Xd;9%S zD{OjX8)#1O9);w!tASxfQ7%B+%(f%jV7<@pfI>~W06RXtXP{O_O-oLRw*Z?1r+FM) zmZOv63wxl#o>%R%7>J>;&6C_ky6Hxz13@)+YSc4$5P@BFjs_PEj}DHl&AT>O3C;D^ zaPrqH)!nzdeq*wccyDH$w^`+(ZM56{6~-L3;l`Zk#IK{qiu zkC_X8)xqA^Ea|M+*v^CN*jjf-|V?@eks%x zBq9+pX4BqIxaZVZD8_HSXuE$o170lB1u~aHh+xluMXGC%Rta^)<;fDz$l4t05MnOP_luwk`mVq(hYRLD@AY$u%FOxJ&=C zl>J*n$B4eRq-Wio5}rTB85hlA7Mo^iN5* zAW{}uwR+c@SCN%@Xqzn)q)hfkgp1w74;K=A5vHhcHW$l@GThO7*du%^`Dhi z=Sm(AQZ`;D`lU>K7I1jb^ZnW+U(h4;Qra2)BY-PWksauZQmcc0e7|>m|A%Gqvdkr7 zZ+H-imqXcJdrb+eigQ^|oCvBIBO|?Rz7<(7W2(o8{&k|LF_Ji0cTl*)(vb>KR>M9kUU{nYp zpnH(=V2oe~hMU?cy%v+-q!qfnL9O-J&O=fE1(W!xZXp|+x>5CefXH>|cR7r!gp4Ei z`5dcQ*8&H{gHZrx?2 z<-jwSmKU`upnI|dXc*Fo^eYR$m9>6wSlier=LO<++-zVIYeZ=vdM;YG0DEU&BS0{q zg1%(Qfly(I1@QwP#$}kyH+nN7j@Py3(4ZgH>mC#keVqoT9sVOf(4lgnzbPp+pw%%K zl1|pV<@Z+26_THOW@ChPr89>(jFY<`KJn&5NihkAvJrH*E!IA%lVY6$r=k_4^xjB-La7_e?R%B>2&ZBBsC$k7`r6%GkV#JfldQLApU?UyHz&BwiV4~f6v*AYRBQfzv z45vMD$xi4`iaF6K1#4}I=ql|-W2=*!l<5qGZ+L}!8OH~?>@Rhdrf(UL4NF+lVtzrf z`~>jsYw?;GDW+Z>|CG2#wGw z6KnAW{kMwrwH3SuJR34hC@EfQO6E?}s8PxZ?c?;8eT;38R`AMDxF; zYpx30Hu)pJnw$q78Cb9>rXo$Gcvr@P(HF?A#YRxf0xd-7fx_xYjxb91jgW3<2gKzb^5Fc z8u9&XuM3VWQ{g%m@@su?L=EoN5w^pta4V;A&pLm;8P-#({sxN%UD3Z>dj7*b|7R!e z)j{&Oy)&{eF~~wX4@N6ydqrQYmGEVg>tX5O4Nwp^ zga)^xE7Q~DM*mB(8Hz@YMlswgWRZHQJ#8j6Bve4(0fsS5Z4Ho(zqnI=`2KWcODhVF zXB5YJHt2@Gj$s1LXfpo&Kp<2(c+wTpYgu#pEm*2uX|&!CGW2e`DK9z zp%+vPj8bjs3m8dvu50Y482Mp-!^6Wnk6n4&YSeZc57eaOz~_b}b0!M%%1q$~Phho+ zt=t;0I_2E5p;*Ps(fKj!hA=FV!zChGOQJXzST;xCELgYHyk?2tG>Q71K$mW!#O$;C zF64w}5i4mjf}0JBWTNo}`(6416EtO;D=ebLz6q6*1A7z5NBQT?L7Q5GmqHL(<;wFs zoG|_ddPWyAO0olmqH-s1@y$h6+1Veu!h;?sZ5m>9vlTNR67Gfet2`n=dv3eA!ufc? zu!Q?Ehk|AWce7{{CJLcz4Oqxr&K#dnC0G_W+hN>+D`*Mhe(NcNd2R;dVXJBE$=d%M zXQ-a%zRsU3v`{1(&bDPmg@TLKleO4@(Q>YBujpoKO=Z9;0Xds00Jd6l&5iGOVTug2 zzyKv0#A`cJQ`^I;(ojO5U=bKEtjG6DRy`Zd(tr5IHIw^1P8cnUiY-X?iKTH}Mq zpj2G$iMIbUOK6f$a&fs!A*=ilZ=Z0eln8z_lE_3jN2##ZtagZbZb*-q8DnzfT!)cN zHGJ>ron9eF2DjK;!O29wS%w7!gYP$6bL*0!?e6fBDE#-&>fWu<-!@I|3mzDETWa|~ zmc|_1mUHf1fkk~)ub7`IPqU}Yz?+%^lVfVP=AK~L#pYBA!aDNU{6uk%1y;~Qk+I8- z(gR}g7$&p&`jN@z`xt)5pSt&N{$@3f28u7RbTxzs+%iv!BsgI&ZPhirlgNz`cuQG+ zZD)k^b@t%*F?Iug6D*{okodIG?Y(NVujz@{O^ol8pfJp@1D1I28f2IED-4hP{KAe|Qr`53rjem$22< zqTJmCeL%7i35s~IpS;Sx=Yxgm1mmSr8-JaWf{$AJ=H%L%NL7Mr7kU%g?;6@UM4;+B zt}id;f8DRY>s}}}N|H{`RiUSGqmwFAd^|D+qdcOFWg?H|<3Bu!Ry*gCAWV}?5|j4r zJt>~j_9A%IRC3wk=w&wb5~_Kepbx||PxGVC7eq>1wX zCaIue8#1^X_>i*`z*U7I#&W#5E#XHiMn^W`wan9Jv#wYDf{<|tkBDGoXCJwsmE0>% z{vQ?^s4jR5T5dZw081ZTAArD$e2JPo+OJdqmmShG-1^WiYL3OirdVx9Hfw&QjzGP( z^k#H<3#;a_bgjwXLZZ1#3pKg9usv(V$oAsft`5I5h9{Ewm*Jn(hNH6QRFUecn)#g2 z?CbcrKtP|sC=uT&Jw+orp}VWd+7Z~fpmE}Pyt!rxsWu=oIcMfQgGsl*l~qc^Xp;}J zH9m`JGDV$>MI!m*vK1FiPLR-&LQ!lj=a+LaU>XAJ5xv;>jLh7a(gQVGI%&q-n?3Mo zg7(Tp!qwpbrj{{b=4pN;HCPe7E^KC*IQr5zh8nK5ll{TY@5`eoQWJgjP0w~OA%pDI zYD`j}N)C(Oq@7L<$+WxxI^&1q-$eFm7IcT-tRZp@7WSqS19owBfgu(?;0K3AvM4L^ zzadFz3s+G$77gm5G99Ev!D)jk5YuT)sh`QrxorC})p(yDvX9mn#9&O!$Mcse-!}bf z^9!ULH4W2KC23xHC3P|cRn4H*!q%)MYjfbrYzLb)+HK@Sws0_R#G4Oitrxv_zoM}} zk9@7-=>^sKVqpTR7DkwhQEklIxn4Z8l^+2bw_^IyetGjZnS>15VsGCIPyZKOVAh^X zqwIX0%9as8hqanz-isL&P)dT)z~vVt^O?q`F9bf!u@9Jdwy3>Zsy?C#9d5*sMD`e% zYGO^JO(aL`OSws$Q3~4>nZDVfem5{cRTCEgHCa-leeKn^;f_wXgGU zfLk`3xCzOF>;T>5KmmRs8^U#Hp|JjMA7Dqrb{Ut7n$7F9nHx7xvOnO6UrtdksIYaZ zB_`Q8Qjxa1K=mz0zNNYhuTWL}{+$ZzpJ;;Jb2EdT86zKaRw?hZm5xBs#YPtwlu&h~ z4{vq-jo!Cu8K2ljRE~ zWgYvt#Q0b6x}-31&Sn~pz%eeHpob`-LTHfwu(Gmrdy#d3p*h^!g0py(ec5?Js4Fhs zya!Z0-TM{==y3j~HK^L>_#^MT?1@*rNhgcIFUBJuq=UvI_zXhiGv>kG|GHg2reujj zGv>$Hc07+rOcbmUwptxYH;NA=c~$BALvpufrrt>)AM6c{;3yy$^CYBF2NtU-!aL$< zuqWI6;vBQXzJCvxxvyfRNoC_KW=qrv1$?9Dc@|)d3}z{1)(FttlO=A*Jmy$dc=((*Uq@Pe8TQ`gA| zLm=8!pf1G z%E&%*d@LtS;ngb7-5) z8P59yUvr*vtk{XpwEG%IMaTdm$OPfBTShW7JpZ3D>q0*3j>{?Jyk+E!5iVz{pI0>| z^aDZK)&~A}R~Q~KZL`r7AK?*Z7?0G!25LTe6P?VEd$aNCP2(3C@#OckDv1>p(g<@U z`zBvHdq+|f^azynC9$5!ZKC2nvn(d?3&1`*7xv~eb$u-**CaAQ0JC}GqKmCR?RCu= z#6e{wqi?b!avalgmJ{ z;Fi$V80*{nT_jK2>%VApe~pv9(jEN9if%Gb0ze(OM5R}Oq&HgjsTXajL&~mO%6vsU z7Ly$dYV(5g_`Oyu*)%@$SS4_OGDtBMmQz!^e*RZ=Xdj4vJs){|fW~G2{_S!6Rk|h; zVwHw!Lgjb#)bwL?4`#4%vbeHrQ_fF94!TK$-&WyQGaYX9e^rOuVxhFw@@jq{_?>Rg z2F$KfYR$J%I6ktWU6g<4QNbLx2O0jYtEO|oKIII^Uj3A0P;JYQ3x%62igb_QT;(la z(N@FJjSPFoAlDO*Q_|H%AU*r_37I$8n>5_N+8+$G6It(vUnteka1-r-TX2Gb~!6m9rE*m~OaWVp>;MDg$KoN8GY6x)>i?lx> z7%$E7cf%=QT^^@Jf!kHobpTz~1qgboBhxxmaust?Qd=-q_%PR$V=3S8wGSCp&&3lk zQ20P_K9vp)DQ${<2gJQyo|Nf561m6j#V48}d2SU&2+oltL)vVETE9T6+8Kk}3T(Tm z``O@PtKZ};Cx?%Q5-}dk<8>NA1E($w zY%BHDO|{y=A+K1f0=4sb^`<3jBgt{RlRO3&6UBB$?Y!%!FEd4xZ&jLPVMS6^iOem- z+TmmjUl`oL17FknG_F8f{Uc`swH-{=n>$?HeBDmXT=I^fg-q9G(7dwy%7G=Mi0411 zTB>?!TP24rd2cx_MDT`@^O@h*&W=q@~m5NYH&l|iGF?Wl`E$R_(b=crHGU^hxo}daECbULTTDj0 zjD(0w9#ed8_;ITLh99n2!QYz8{9vwhs*=+f zlQqgn!nqu9$lm9km};?+e`kvk7>Sqa6lb+?zX@!*wf#HB@AL@yn*Lj69A%%#Rl>>o&caRQ8^^vtSc;3UB<(KmYUUX} z^h^F|$K8M@TN({^_?g(VS8nIiWfe;_T7lFFTg1E$CC+8Zi>NtThw|Xd&7 zFjPHK%Y~&_248SxQSWNj8$He;8CRzsT)Hti-h6H``A|0*t4o7LF-CcG%3_ z@K{quYjU|!Y~0u97D041$3;%ud(u72o51y1fTBQ`u5sJDWBIoHhDYhiHo@2fFBg?0 zd~m>9Jwha;6j(Dom^;mybk=WrYL;$Y`nl z9#5H_A@9EjCl&d|d9{F+Y(^VLn;xfVT++S9`8T>=CkYycn2$g>wu+Pz_V1&SdI=hr zH5U~!wAxtYsy0{cP$C?x&1&h6;_ubdAMGYf;_(9GM1eL?d8Z&_2Oq4L)6TumMqz01 z(+&oQ6n}v$nBBp0zc_YM4eOy&c-;RvM{(Wod-=qj%iIUroYTU9J`^h1ud z@dfp@7}l^_#``DSRtM~B?!mr0ALmge(U=aah?ShvBPmaxEd@{cPp;_2-)L%kAWV`i zk(~QsJ-e&5LAO-h?lI7V;t3Tm^f>gb@OH zj!oBj@%+?E!(*nNpUc~=t#D!A9zPbk1XDP(D7pX#C6TbN21-H?a}6+ zl@T@{Qk9HH-3_K;x6~vB~ zJ?@bZt}y2`jmnA8M>F}V%8kmEFo#crvT!Pqlb`xAT&6r<;i;FJKmTz$=Z`zHQMVwf zS@8&KuEIMITCF-gyNYT5nX;{0rgjvPZ?-;V8(AB^8tB09S;5y#I9}#WrEG|y{sIeu z1Z6~ZJ5F6ogk2R`#9)bjjz95gOmlovw?P*aj>oU0QJ9icq?7)weH+Nds@;^#Xo%wKmTAaeAL&fK9YOb3%rZ?*Qo~nK_9U_{ z!7zCgz8JXZNuj2}-gd?xn#Srf_BrNl0s_yDAQ}|1kp{E`u6NRKTV%S)xjIE?UYT zGVMA%KHh(KxnJY9sQ}p43S8DY27VL@N>c?i9P0=uG61=5@2qpMtyM^xrBKVC8Xnc3 zK`j%!V@~g&-%C8`o*S~+5H6_gEOIuD#Z^Et-OBW9o0f?OuzPkarhy5v?Gd^R*^ zO%%Gn8D+S}hhCg?lftbmi{pJt z?eNf0f}L0D!;^O*z;%6*hq_g0Q*wb=92aU)9cIgUD-|9-H*qPd^lk^Y&8KslIiKYK z9X);q-hWxGe*n^%os9p%TBliu@@Xmj2$MDEFXy{|ln;6~Ha6=(WU69LyqY3V_;wHMi;C_95-V_T8fZ6S%d@#2#iJuFG_PZtU9 zQh#HOB9Z%}= zhyU2$MsHj2#(=v=x5kij_Ti~p_J4sll@E5xYG+{5Eos7vew6n(96#XBdbD278uD$4 zxKha5*ghXU3BzvHj>L{SrSL*Xs46O|6Rmuz;>QGYw{2)3h_!i) zeH21V%6vhQn>i`lZCniQ_+4i|>nfz5K6flLwJ=QoNhnm=$O&T5wct5B3rqOrA~{=W z68Y-_huQNe6pu!$j#!>ZWhPk4QU0wqATJ=Y!SFie0nhiSF0JTRl+2mbdNlTMg|IdtuQ+~S< zcZ~zBF>*h@v!0IZZZ%BG>PN>n-~=&iU1GSzgEWKbpfYch%aKJGexc2zhk~ zX^_Df)>P5{5-}`tvWTh=KG={>Td+?``l8KFyiZwv>%0_*i$Y1&vwLHvHbOi8tsAuMG-wr%zry`9m0B(+F zyMY(h_GX^{5$~o-??%ghWpCYt#8-gsg9PxE3RBQW%&-iq%|m|BZ6$bxy>M#U=C=Dy zW9^+yw8aWk!YxSW_mkIypKso|+k4mqeA$8sbw{BG5>72&Jy}i2UnN*?YksKqVKG4l z?^?;@ad^&K)gJCi4W2u4%b1*tB+O& z>wf3KbEe3&fuNg-py0ik4~@71%-fmQtH`Uq$i7D+t;T0s3Z9M1rWKqh`huXRn%dx0 zw1ZT|bw=zJPk!Tuu@=N9_Np7hx?MSxv3!?wksZ2M=6+n(pyct$z%?$?} z-rb(17QIVfSo5L)x3{orY6}XLmZ6C1|7LL2L>ajr!q%2ya{*wjN#$URDMO2g5ss?A zfGkD8B5M9dijV{-KjeSd#%RU2lO;RlR$*edEkRDFQn_ttV!qO)*LdUM=z9Pnp;tfVhd!E0a{jFOPq{;N7m8REq& zC5lf?+q@4>{4JZjxXA+4_Mc;L6uaaaFiWB)0DCde5sTy3i3`E=^Ne$V9X3lYxaca^ zocR=PAHPZKTxzOfHsr2t%u(dd?(rXz*a{$s?dj3tu|$Fr32w;p)SfdYQ2jDae-M!H(eoE&rCLH0&Hwe8I&0CLI+2_+h+SvYsNzIwNHoq&%#x1R-F#9WW;b zhlb*?L<{KXge7K@gtmn$xKo3OMkb@!WVN)&IR}Zt-( za)y1<8J_0(6tTomf3)QBwm4RL=@qD?)G)xcj^8^fT;Yrc93~~(GRquLRm&u?|I<)& zyien~{>L#%C*p(NNo%)D823DA+=Zr-;=xo)bK(lUfu`D{!-#I;DgyEBhb<7{@Q_8cFSP_6w}7O+~fDI$j2Z?^IiFTFnu8JW5Hh+yf!AG)zL||yuYPeG>^iTlMxfXVo0Cq8XslEB z2XX$Y(+S<51Sf8E@BEGz3fcQJNE__A{u67k@4xWD{T9ZSFEgo*20;3(kZpTsh$+4g zP?<$(zum$nGj>IM>NUku$?X9$u;ZD2Iw=qYXox3;c?nN$_TK!Ufa_nWix=jv<)n7P z#B~o8`+sm{IiH*VBDmxkXrXwtfUNdS-UP+q&+F!*clag;L={@`Uaf!`d5feAG2+nX zZbhT4JY~C<)9ZDD0=y%(cASOYLo~skfNjj{YA?jClOyHXBN%BGmvza40a&Q|oWRMi z;a6~mY~3xXf(7G6kS&+*Lyk34{s=zLo=5Ud%+5pJk6O4DD`g^>o5MQD*g#ix5BDBN zQ-bK|1hwMn&a5VVT3`@OI(HF>PI)}PKCTIZz@ardg_+`M_E^~9S_h6^LB5suNsd(N z^&!Or?qi9kY&Z@PZ3(tR($uwBS@cXCL?1pdC@Ev|d5PX9JJMjG`uzAHJp{v5VA$`C zZGx4`C@Fsq+<+w1hC}w`KAM2jgf!-rYG}1@jhEj0 z+P+c}_a%eKx@Rc@_p>bsRiq^Au&=W;4<0p6&z-uoL31gupkrKrnv zrS^~N3F_2#?`?>$$69?fV8uZVzDW_zB#0}>uh1Ci;UsA%y|7sroGrC?}<<$&JVR6YSAhN0f z%bHSuz5#!D)uCj(3kb~ScSx|s-@iL4>zk;&y%@0@40xZr_b;4LvO%$WJj{tGK=J%> z+y#5>@qwi}F!cC?*<<*;h7>K|Jx z-_qnUc1KH~*RS7V`f0fiq(T5qFc7>NeJeGsy}p&KDOix6qGN6!mBZZZ3oJQLXqMuD zA9;Qu_u}%)p>PPZAy`+bU(FV)ev6+(*$!Rei;;helr)7FRIC!g#teG>o;rklX1gRN zJ&bCnU=P6$`~TtStRI?g+bFIyNQ~|nozmS63R2P`0|}+OL0~iz(lJnwk_PGS1_>G6 zFgk|B$al~C2W-FH-}}DKb~;lIR2*-um;6ZtCUtCT zB~{zuyg)u-dLZb&*Y7_@5k|6mqlc2rxIO*~rk$+?pW#S+<5Q%frZ(jg$api+gS21# z{8?O_4GW!$h^+s#P-Dp4*LjkX!Js&Leor`%0h#&!!-2Yse%#~#eQ!2~lLqAc{QUp< zLN$k}BL5a!lXt~g55~LAYk!tRwgU<#IDkMPR9^uq77+YeK}bt=+Mx=G!PL}o3t|Bzw}knpeTAnB3mb&3x4&id(R#)(Ldoov?m zDzZ7GFaPNu8?nqt$}jH=?L*nM-FNfqmwtA1;fPo<4u(r{SpA?^oU8}kvP7ZTA(zhg z0nJyG@cHBLksfIb@59MAN}xM$l49p?zh5Q+bw!s_S+@^NcNu)%EPZ8&T%CAQ)Fslu z0&O#7VuFPgL4~$_kgl&UCQfCeH~kT!`xt75ayZzUs%|5C^$D4iAz#g4T}F#qdiA_V zPd;cGBw51-0&7GOtaGmHH~#y#lsKZ99Ty+lqDA;q=Sh7^<0 zr;lALA-*1H5o7DyZhSwhpdI|s3bl6fgkDP=?A_b&K0v<*&*AYTNqleE#T^cVRBeNi<~p^gWZLp&GjadC zUCj_vK6|vjejVYc3FTd9u)X>KQ`?(W2n6DHEjqP*mxk@bOI#(x;eOLNoc@}}J!qWK z@NBxt@`Zi!M2_Lksh^R-7esD{AaVM{)Z1@OGpi^*BgLBh_fL{Hld4HD=Ybt{+mscj ztY|=jJVae*kom^NdZM*cp*SH2&0YgAMQl$cLMFs##C#1|l9f^Ohg42Vs&-ZM16%t9 z2hP7@O)fa9mO7SVaqD_CUTnWhJR{e<$HXmOJG_>MTCo`{Q&-9drGtp{cbZErmc)$NE*rjLTj-GOuz7Ynh;-v! zc#tv#=$7W|FVh_drwBL!O6kIp=SqT!CQShHvE zi6(5c{dPk$pPo1rz5EvTrM(WafiNY`UXGW&a-CW=xmTIqi`C8R7FqcN|NQxbZ_FAU zv(C75c&*C0%l+Lpf7x6g9yg3V20|kPK8^;Ik3E8m>djj6jmkzJHx`+W%ayC9lLqtK z=reMOZRGv^#>d7kAIa~o@XS%a%9k$`>Hg68k6`SL%(=G0WY$%M{gP`w6{vVDJ2fHd z`Yf-UnpQzk@8r@x^`3udh*_ZvV4s9w;+qNsnZ`#B|im?qp-Rcw+ zwAKt?y}pf>+yDK#%C6gkJ6UxCW0SywjkS4=C6eV3M?#krFH#FF%oiOqK zF<#~e-;KnT05UG0_&M{r$$@B@9)+sy=>X|5h z;_q4v%1Z{`W{bHM1R=|CWpFQ$+?wMXNm#DOalGK<`H9fAx+qEYE2ZapyQypX`GS20 z|7`})+JJMnD!zGd8N^fzKeS|#KpSGMmf-a(sQJmw`%=# zGTAA}#G_IS^XV$)GIZzHt!!6eG^~ves~hLl>R+y(ymF5l_y_nyKtgck2m%3~ zo%l`KV%+#yKa#ZB1|6ZC-ozOVGuW-yOwJz?A=d3(w#P~_YV-c#RI3~+?2_lAnN4E! zco+Q_pv~NFZhNtXn&Na6_`x{Q$bG9{!!SQ3q_`{H*!ZF~s(st9k`h2}Hat#$U8lHFIK=>ky6+@wC)rpL}+P4lVi4kV`Eaf`H@R zJPp^fyd9Rs(zm~>h8S|&IgC=*NHG=uahme7Gug-(SjXS=VF-S;%d?Z5x4wi>neVi9 zc+z4E>QYl*HWW&gJXAbXK2*Kin2T#K;kuKiIa54*gr%l)8#CcDOL5wy)KQ?nThKt{ zJ8fSQ=YP-<7k7FHx?U?dOL=w@yAp6%gg}CwFp07|vvqXMQ{S*3&Fxa2LXr-?1*o>R z3b(HVW@eu@r88VP#8??FxHogZ_6a_w`= z$&3xqG7v5$Kj_ltwTiQ{lte8B45Zq07?F-W(zD9iXiYu}A=asiue;d9*+!{GnF{nf ze{AGNxKu6ttbvrItNx=#N`bk&O-Rl0YFdM{&u8(!e5DLxhxi^o#@o^HQqtukl2>@s zih{w%`1v@#iMW<*!OrSE)xsZ%=%?2rS;)h;a*Z?-?95y#w)90RQX$ic<^d-SPo{K@ z@{ZuEPo8N^MA!yQzkaI$!tpq$6yJoCKd_(&6|D{>K6&y8$4idcq3JNtp*}`yi}EP> zs`8e-1E31%&L)mJlBs}+PNCGHViwpPvv^-hrsB|4L(ab1b@*SsE>Z(*6}BC~fSlw+ z*s-3R7)8da*3IcPrDg_x z;cxSt1+5FJ;p?4nnOJZQUy#f@Yj(@0L^`~3l>h#bnTX^-jhQ{Rur(HKGuxr6w(If4 zCHN_tDWS$L&rjuK|IH**qSThM9)(#KtH=jo6~>-J)HB1S4Fm!!2_VB9C>I6&3MX!G(kV#cQUBeDQtotGwk_B;Rz2(d%TwRGD?hZ z42Sr&v9~)F(Gi^Z%1POq+BEBLZ>3^%bTe;RF2`sS+Vz$~FJ85K9L}Rh0#EkkhG@ik zbAlfn-gx9o8yoJF#awb4BAX%ibN4#lIJx(b&&w`Uj4OH%AOPm*?P`|-_bXS?Yg>wo zM-KsnJ)Y#%hVN{~>dYZJQ-3+GoEF`owp(m%>n9FlKvGTJ?CWZz8Zjn|Bkb-WX#U)kFxC}1kwXr*>*#tn@}7n!>cQbJqkv-G5mTjW7t>B6 z7T-fB16me;VZv9+**~^ZRlApQ1)P&^6GB*m)@u%MRi&1hUC7blqnX31e?%L(6ffw5 z5}Tmid~Grlb=x+>XQgx6NqXH7RSh@sA*E($sZO5c8R_e@-6ye~sTS3}scI+dArZG5 zF^V>;Cm_q1A-pwtOGo7`xBlY7B;u%{IJc;7&#dpc?C5QFH7?kne_hLO-7sAS`zB`(vm~5HDBTXNJ32jc z@V;BPM8eLLXc=WHcqJV%>U}amb!vJQ(_zdvXaL9#L0#Z445v-`3-!1At^T@9hQzR> zUw~#zT3f5yENl0$$!fx3 z?~n2o50XQr`C7aw0=iP%0@HqFJ-TGWysH_sO6N? z?%NE4_NPQ4q7Kg@;%p+Asj6$U|5D<(5~EFO7Xgn2NaBwLgVA($&VLa)=hlPb2BpxW z@Rxrx%)PVv!#Es&sU^V`q`i?rYTq{hPti79`2Tj=PLa!JUS7ElWY6Yckk&B+CzPeob zcxBOIQFRgoK=6VqRy}+|)OOokzqJ-co`D`Mc;~9HJbNS&?C?Y%j0FU%Hy?cX&EBF% zMbSv%1a0?o9nzw(Ecl#^^-6+-qfZ$F6Qfxcwb zIevTV+?x8FMG)g>D)Kkb-V!53(2~aLDR6Cgn4cam^5q&BkClXX)9+|z0WdKByD_M~EW(CdC~zii(xD z7ylm6kJm0OzS(!r$c!H)=SZCLA<{Kr2Xi0=r=q5&8*WHyatc>4{ayNJZ;I=8i!Axl z@nVFChsqB^xcA{4V~$i7N#6=ft_Q?sd44o`;X?WBrP_|YjKp>{>L&jOmtC4Do+(m^inSyo#GZ;S{husMqxVc zy&~^-5~08JlL-!@GlbM8=xZ_7kCuLK28Kz;Jcc`QXlx^@vOc7UJH;p*$5t5E11!F+D{U>9b%`~nIjQGT-w!&_Uus0{u5h*YP2^_P9L#bMZ_)`4 zi=?Jy&V&8-o@0m=jji%qJcyqb6D;^M}cp zVg^Ea3g&UdH#@Ljx*P6%N@k)}?rqy*!sS-*0E|<(Rz`T;bL(#I{c3bq%u$pRDfJPu zdoJh}NIy2wdE4u_Gi(|1byK%?I{3!CX0h%>l|BoP_VZvf-63ZCcOA_4PzM8AF-eav zs}MIZtzD2~lv2P^w!O30`Ij7(93>{+cnL-a(>7rk-@U`6yuL02#h8^w*_B)ced$oC zY^m1PJ-2y|Ul*PBc`$epc?nv)@Uv@hAYT)PNJQM_TYH#U zQXT0YUhU6)@z0a`;t|8)%=>0a+_-I_%OjySrS2|Z^Me=fjn$hG zT5q^bZ(f@ya9ORjLq|=!q1m5To(iS#0-*Yb;}6n6iH86?5ur8&i673H)tW(4*hJ3F zUhb8pWy2}o)k*~XiE_nfk6U2lr_D?hFGNWt2HH+Pi&)C>fmN&pX}yTn(8}m;7KbW6 z$QnwQNE1;s+qOa99B0mE4Ryb@GL=}=uzvHHpWv^rl!vA)0cU2P`7z6>6~+iTJXYs} zBWXa(|LuigpWbfk0JOXc;62fcNV>$|cTyKh0b4O_-`=zlCc>+DwCZw>_ecD6 zizPpw^BKKQT`EbKdb{9#2P#1wYC-z)1J{uDes$-!byS_f{_UqhShFTI2anSg2BV_*YF_QK!l3tFl;z zy$v5_9_U9gZXoSgBP$igp>17=CyTjX4SRR>kN*y`t_R2)93OlY-b4i%kxz+QIHV8x z5&bBy{VVKZVGsUzid^qR1y`Ahrsm5fw>43USSb~2W^Op=-qTCcbV&-!?uFh{j>*}w z<^kf`0Rpki2fykk@UCT({B@df=Ajtx~khfU^P0t|cy9t+31hUF6)*frnr zjM+|%x=tmH4gbBF3*CvTNtB59aRT`0u|++HpA)!LAYPvZet1sSje%vn5pEZz8@2WW zQ2(mv6jdJk4C6P0OB&hPXQ3!oJ3e0FErss`k5c^ zcU|O=7NdM}f)et`#}Z)|Hv%$C6u zC&h=DK*PO8&10`YZJw*dM}$P~d7gXf#~&ccYT7@UlE15|GRo`F2+K2LDMsSfS|>`+ zH?6JarRzLU zATaT1nvL%eZ_~LifYGrWwwr#u-&T2?J!{7{tWk6wQT{7`q|D2$TF~&Nh=5{5&6FA6Qo$DE8K|f5B$uC>taS1L4w6~PPFh-y*Hpc&g#eQ zJ>cWc9vNail&vpQH0F{y|KGZn$$d6l%Lov{cK&062Q;Zg{`z5YSq(5?`4$iKmk*bP zcTfUh^Hj4ZWkhH=81ad^gs}Q$nA}F-2=CsA#Hn8Ml=7(swJt>U4?5p#U zs&ehqEpN{*ZG!we2~VB4wQ||EghtuD<=Q2-yy9%iv{t1p(Qv7a(n(x6Y`BcRGYWzr4IIP5?rk-pJd?It~tJF>OWX@v)Z&>HsRv&{CKmB?BP9uaiJ^$Bu{?Nnpkxq>#5eC?$@7N{m1mpjzZ7- z7a)v*rYgCSr{7uYIT(9>C`K(I9a12r)QwdWlbNGFB=va_iKpK>!@P-pIfwi%e8!;g zJ>MB(pV_TSYv`3$YpF`Rx@4;wx&t3|;JZohSkb-4m2nK0b8TFX)b8OTiwYuqwZ+31 z2~k5=B#s11k<7bGo3>*R5fud68FD$3gCp<)&_Wr2QP5z62ta{~f~vT8_5lg;o3KB? zLe-qFRHRZ^R=KZV*+(VsI|?o({OJOtsQ{WG<;8@U&S$?qE6n`C*1}Z$K!Z?alUlXs z00R`Fv(Xu8l*B3S&bu3s&?)@@Eie6K3g>1ajCVjzRO-769BaO{$6K|W{QNUnFpy}$ zEXn8Bd0Vr*Qf|t(nI(=)g^*+;J48CW+2M4|OG^{qyq6Me@!GP5mzF9% zeRwOhK0?HEQ==4T&NE%w#&Rgsq~s0R@nf0dXIdr_zv3?R)*CBbUJ^KJtkn?B&oZAX zl1M*z7wz&d%h`1D*Uz2mP%RGV4mn4asbq^V3+p`7ztIWJ=4-Xyy>a!0xiV~O(bu5$ zpo|*&5L{rWanw!I+@pX%ZzmQ4pbms9`~>vXTUE<*M-SuP!Xi1;egKoelb#HsUmpDL@$4Q{adpo1` zPRjWbt;nmxG?1xN8t7!n#%LBP_jf(QAL{GHF^&XR+GJ)m1uNaQH5~ph{Pe#PsC);} zDA(tjb;>!IOimilwf_wF8D@ij<*2cx8a2O_t+s!PP2@uWAY~+Q(KdAgRJ>t&Ef{9oXu0MNU!}Q5JbYp z2sTMMU-Y*5RNDWe(akBy_-LN;)FSFZ_Y!vJ;d1xS_;zFF-%A9gp!6iv;@i%G`5<=tFXA5@i%hGi!b*5?>#b zU{i~5gP2snCn2{llbw7Aee<%HyJ=hE0m;0an(lfZRfiq>ISOoFiiSTzp7b9#x623oChzraSw5Px(UafiQ9evNG zIot!6XAsM@5-J3=+h3pL)VxGhcpQ;*AEFgB3(5kmwnp55>TM~XRtN97+qu+Oo+A%) zXViHUH>n~JV*0vExg2SK_RQ+z;UjWw?0^R(MBLy9ihMKjzJwBCZAqUaZD8CDkGPwQ z$lzth11~NW+su~*-vT^hm(qh>2XhX{txS~~=M%0H9vTal!7J!xA4UJZmPC#3z3FY` zVyfM^ZM$uwX8~gRJK;QU&S#M$$cDV@8eh}e755^!HL*f>m`xR|MJy|XDjJQ_$)q~v zsKFZ#r6;mVAL=Pt&sqgDb$Q#=Sr)(}hLnPZ5c9OrG_z5D3<6oEp$qDeI%6@+$P8KnZ)dC40Ll21RkO^di85<|Q%b-M*7@4AK0Q z1m6ZIW5-Y?V?5#Mdx&hrdi&=xQ;WYIo_b#6BL{;+WmZbt@Rv%XEGf0kfr^lBII^T#hX%Z;E>eTj}=J zOhA-)t8ST~UIE4cKJASDAXSFpj-fvKA*THI>^WxEr@XMJj)Y5)>1478OYLia=vGq_ zflcW8cxP(3kqF1~f%YD3(%d?OMpi~jbb+<&QN&mM0X`;tP+r|zm(0}Xf1D|0y&zVY zBtn{Ai@(ekBR@lvXtSW=m$1lMJZG2NFP<|0>zbzQ1Vuq~TfM$u^Tu^m=8k<%k@}Sv zA@r_(N}+?5leg?_drrVuf5Q{%Gb)wgTfW<}Yc3gd$J=GSar`2&WsYDDkK0;$O6tsh ztrjcOub630YtFDI=NIMVC+ro~M>UB0s(MgN3hG+ST+osKI?iG5A3K!d2VN!RWQ(W& z-u-_qZvJEB;gHX2dZx|ciLN9)BzpUj-yeN=&VH!=_a6>*?cQ<+317OP!x9l|Gc ztfw1HjRR9trRuZ64OhgZ2^WN1*q^^=Gezo_0BVJ|&`ddYUt}Ki{u%qmA6eCW@27K- znii@V4QCb=g#YtE zr=C|#Uge3`a#pSch3g4S4~3Y z(6x$pT=gcW9Z9nb+-;r4kz>|DBFm)pVYGfT+qT4X0y*%C#Vz`eB)9LSNk0T@@M(4nJN%*8ou5EX|av2D>_-O%>DaN|2D0EgJnQF_p;qYP{vkGE< zGqp#>%+f!YWI`wF#_QkjiPiK@Y!mzjY0df(MF)oww;<_YqW-QYpGqK&^JkA?Dw-S& z+-GOOH5*~hyljqT-!#$oTFP#ao@=O@beEo#uPTA|Ufq7n0W?~{^JpQAvVk1jC?0R3tK2F4kDMFr>N!k=ZB|@01%+&ILLn8mj`=8e#S&X-*_3@ zRQ|!iZz_tf{=5HDG5W=jn~hw8iI>QN(Za_)(Jte$4OUK!|87Avd4{cSSzv;hlbbNs+9F1{ae)W?6x^a)q2+DEa4Sa+`YkfrcIhPe;M(|e;rwD6bWHy zpJr%@<{;7+`&B1crR+rz2MiB=2JKa_f`sn;x1WuNy!|qM#ygohQ-;1d2CDR zO0zL$f6c=dmnFJZoMef`UZf8t^s6Yx*`QR~iP}hnk-ub^;@h$Shhd@Vp?6hmXC)}e zW9Tz=D&`Jf%meUa@zUkz=ZN{+1;R?;B!As3gjxvMJm?(qVS~h&xtI3{bA?pWc2h(` zk6_qgPkF6LDnm?egBsI@7{qX9(pZFFnDD*o`&0KNy@GxJ^jyF| z%v0&M(U7+egBo3#R~ozYVM+9;5_jVEN}Lov^BPjXJsDxk_jF1WxZ7{R-nifmBku{v zmtaS%N7ffQd|CCJ)$~Rxz4uBV!@WYu6~SJ}EmPYn@@FDp1OpuBvSY?m-etgKD5h11 zbCupD@*z6E{01wO29#%0VNKo0!=okp_`9j8KTr?Zn7bqK2KRb}q~=7|~^G{01g z0zGyblQ)d#+O?w{)$C||!urLiDX7u&+fhSv`f9wWng4{eUw+Hlpl{RUl?;G#-kV3T zKR8}#U;#`RtKG-7iW&=YC_1_T*fNqT5P7U*^1UAk_zozD)qbzEkSeV5QDWFLmH*uG zV?rdQMDOo0z|AIh63Oa*y_1`Q{8I_)m`J zA05~-A4r2jJ|0LT=FDQB^4}3GpgKPbd0u{cfXUt?{_C^Hmwj-BzIMlB(gAXR=v-+K zHn*MyCVww|v%v#E6`?1Qc{o>eNebQw#XSD-e3v1bE{kNUT@O;b_+D5x9x0t5Lowbv z$*Lw@!ALSfECz5C_4sErq9)Obiv^eWnJL!v6?F6ZGnFAdad8F~ugtHw<*Bp!xswzj zoR5F88_gW+qVK32cOycO6TY#$ljOcafqzQ&f-~)X3(@_5A?K$XD=vUcFHkt*>=#w_El8ty43V5;HwOc z`W59%r_B}8@(Ag-tGnvt^GYCEZ_0w#mL5H<;f74doaHx~k=(li*7tJXznl{+;{Z(v zIT?)DC*$Y~(9v*xO%=~oOFvxqIQN#P1Quk7D!np$HyKC5Tiw;K;K)$KBkPu9bUH)4 zE07*ADsDrT{DEZu{A-#(Jb>-d=wZseU%ZS_shx~bNXb<#_D70!^s(3rN@}L5S3`;m zeTbjEf-yX)c9Kh`gq=X&?ZtuDZ?=s0a`+l`^Uf$_a=|?5S3;kljSZ|T7eue0!ON_h z=+VAFCClL^Z=x_oDGC8g>SZ1UYqWR1a*Dn*9p5)^BmJ4p)o3=Gb3IHmu{JP+Bv4^gEBVX{JJ3ZEAtAN6_LVw0^8Qu z{l;bI%SH~23rO(Y!H6*FD}~1Qo2$qqUPkV+n1s)%qahOHbNHM3b>7=A)-Gq9BkzBh z6B}j`)rqk_CDzd9p}%G7?7L^M>Ua-g7>Y#*(1WNOZSt=|r7;MSEWf&I>gXnls)a{FvP^TBXNW$m8q)6fu5Tp`~Z0KQE4Zzu^K2ejH zr??qKNg!6;O-no1{_kbU2P#FEer0S2+5Ng^JkY{HSWt0+(qq$O5$lSt<%jj6 zHiN7HJGXMmMgR`=$an2kyu3BEZm?y%~z|?&~U)^}jedr}kV9J`F2M zf@4iq693A(ez_#@9-W%S8MLI**WDB-iSGK0BN%)!a-+ z_Co--a`l(!EG*1;2zt(|<>EdYMY)LXmp?53{UlS?|D~&>9|ot1kY1oK#FQHqXdHfj zp}CTN742&!X5CIW$Tx`87aL)Dpf}AFHzm~8L=e~bB{#HyR-{*soc5E%0L40=Q3{3MTe}ty*q>Tk+Gp#J z*1u^XpKw>G9WskAzy8Rb?%e61(}E(MB49ucF2)b!x_iDskqBv^&0n+#IN4PUkA<_u zq6rELR#JYjjVNWD`=vj2mD0NOpBOsMpZP%_^DCH8D#~uFeh+z$N za+&{LZd{o6x+OtY$CApKH2v?{G0~6E+6f}M<|5Kvq59rd3OC9d!LWea`wMDWB!n@`Xh3YU=x)V(C3MtQrlriGYoaeRc!X>_hw zBv{3e`th>rx-YEbWXQO~X+;aqaqB0zsr*5TEgCczLYqhb~;;ZLE!NkWH3XY7jU==eO!s zR-A%jf6%lbaEcx~6oCEm_0eK_`wZ|ye_OOSCGOYC6ub0(sWCt2+6b#AW-U0(SL!U9 zfKcF@mj4s(_2iV6k+ZXv7c)h6*EBUS=H&*nW7D)G&Sq6&6gJ~zUmwtl z3nQvTB^5{_fMBSz8uebt(~B_3LAl|c4_#r>L13N5G5S=ijYJfI&xiv{qEYlZ%^Bw2 z@{BR}hj3fBd5W>#>!eS&Ym*_0LvPbus4kY#-$8sa0H|=O^Y|oQiL`xSk_QL=;fqew z^9L1|0xVQNXkD%k&lJy;xR+7|L8+>*Ge_G+2Pe4H@3&i8A5vF@DdP1Z)r$kgO9SQu z)GR1DU2}ym(lZ~(zKMlej?CWO?u0n>!I7;1%+IeZ?iRYl0QANFX#H*UjEYh;#eD*3 z5oJRihrh}L=>|gChUc{Ras}zIW`Q3Hr2lL4C|6FR4wu5lsTG$yhB{)_^JuwsG(95U zi2D9g&{SQh+GU|UwpOSZV4H-=Xz|5aD7UKp!a@T|&Xv#6DoMD0qXQ5JsKkQCgda(% z5;k*DB&ea9nc(79C}UVT>VEXO1;Z-gD^=R2>H_GYXG2jnC&3e!kS5{_O?cq`&fq)V zlcHuRQi(x|;sd2^7T5fsUVAf@EddmAe#=8FpCvhpG2|=Rv5$TEYeW(QCLINw{J-AUv5}_E(CV<{bB7%OpV|VoGcqw+(rVokUVQ*GQhV&m#M9 z>DrKE>^{bW$6o=kc_W|aH}$1OFr3eI^9Ct@WKB>~&8$p~ zlLgu|-eT6<$Y$NJH@hb+^NME46Dfr$`GGdPP@$VHVNu;rwiB4(VO}20me?h)JL5>p zB&S1WOlLc&&wZgI7vpuPR!^pB-gn)eG;N>ycWhZ+GEBJ+29=g?_iiSi`yFRO#)}`P zaLC@(*a(Kyv@l^V&k-s*t|X$1e0Jx(58OJv-n>OUDg0EjRN^sNTWx#lV$C|EGHk?C z#qsV%s67E2)wC($7MeoA)&;I9ws)H6GR|_XG>W{E3JZ2T7*K@CPc+3eMI`d3-3ji< zxTUC+`V_AmN#8nQCHn*^yKK(6xBXIK z#Eo3o)L*hyevKGwVm77Zje0W zY%YPwUqov6W)+!d$j`3v+f3De5m$J5@r0INg5ro7+$-Lnfs4dpN7Y?x1h*?jNCUE! zN2oSmi8|m$nfJ>aarqIBty%4aYo(GDx^1fk)-u$Mr^BDtyn6C#6e~cdq`~1LIm#qR z;XSh_r=Du=CO>a1^cVGm0odoEk>YdcI*sIf_s_k0rRfdGKOOhWrXX}xsex$8t?lY_(>#mf9_qbZ&W&Sdy>^G;~(*GrQB zpCVqQU0jQo{nc}EM?;F_z7`r%w{aRC_SlyZOphP&V zdxg@L3zoAB%TDZ85CiGDF_83=9ra;vbJV05ci^9fK-H^PR|1nLg=K;# z1Z0+(+i_H;!lbsR*IGrCPj8i9S2G?hH4F!kYkpnoecGXx&MjK9C}XR+P~i*+Dd7vP^wLC@ z)sYI4Q7AhpdDVsz6Qo8H66EEn&lc>m(KNVDTdjC_6XiS69DHHg|A4^uF&A7FF+RN` zzXXU4hr<=cf3!1fFb!a-@W<-p4*C(<#0v~{yjK4*8Uzbu$IK;$jk&xW6|sT965&&V zLzfZFyfR!;Oj3zbB?Ho^J3St8dZ=&)-aKI5$UZ;cJgWZ`4sag*x;u`yb75*7IV^dP za+KU6esvEgGX;sQ&J@2t3+R5r{hq+LP%A_ea9msEjF-k`)rllf^iGw)k-i=l7==(n zN7IiAf`YF;KQ2_P$Bnt1hp~X*=_B zco0erNIC@M`y#N6AP7+_KjE!Yf#FT9Q0RzGYh!Xb-x=v`Li#2 zNOEPwx~AJ#d-Z$YTQ>4dIxAU;(fHE)d3^_^>Nq9|$_d0~6o`cFgazrFKB1IY^c+3l z8=4gOV=*?){oAAg$BgqI?J_$?Jgq&hfoCq^5&vfh;i22y-FJD=XtW{jyl46d6KTBW zbdrI|et!x*N>^c`h~#mND_zmcaB8Lpdzf2r{Ldjk@nkCidD=!(oEjGgU!fD?qNvFA z>PJry3<~3>Onmx-qobU(p!FV$0v(IA-STI-lq|mYY`&~Oj%aV75lAU)%r0w}N8k)p zQ_8ULUeF=y?`KDuyEi;g>?!X{FxhMTcIo^V# z^-3zepWDX3r3A~$B(y)5br=_H3rc~J1S+}OGuH~yzfh}#@@&8zQqWopxg>CNV$b}7 zAR$xHD?;KJg{JoIRxhsKt3JGSK{w{xKa#43PTMe-dBRI-x3?Z~2_lr7V>U<1OaTDh zsh>-Zems_O3D?`~BDUh#fDC1z`2($^p@Al4LO7P|u@Mqt(x<0=Zh`E)=(&DGPDKGU zKN1miUsk1PCkag@yglKd*lxles%x+7xJ|p-EDtG#)jBN_2rx;JPt{_e$FHvH=M|}D zyWK@+QQr+Ins1;M@M2j{(Xb@4JYh@Qlg@1FVq4I0Fy6#DU!u_}q75BJ`ddK=s$N_R?p!@foj-;g1>p$4?VkI(5?fcu3SOf|c0C|ky( zQfn`KR`{ierwiHi(l4aWSFnNodl+sE$tW+T8NynX^*6$=iEb1Ek1CtqKxPoR1PdH| z(XB!onl;#XKbY4q>eRMRn%3Vj*Ar9x@{YCO7V3L};lag@f*T=rM zjPk%y6qVY&HnN>>7d?N3x4odFUOZa8@_(F*U<#u@8C@&KNvqhp8@bG0Wg4(WmhVgd z31HyvtmOhPJp5TK-j014mLS0>E|J_9g<1F_CV{0uOrp}0&acF3j=hY@%%XqNSBT2> z-A^%HUe@$lna4om$ze%}0ncxQN;9AKs4Q`RI0=1z?^;f;99m&#s$w#joHi|J*}936 zYzdJ-AngpHM{4AiyTBxTuYXUfL)i&}$e!+wFgK3KZcEAdzZ6xa2D6LWOoYp`BQRfz zRxC=5AxLr!#e^6>(>9^dAjb*sBrVW-1(WFoMi?F;`~O$csiYJ>Ya!2%e+%FLm(ZtI z5DNNO$!@x%jubCw*81opBlUr#u#?0|(Z<=8gC5q_kGXG&SDppO1IW3J2pgG>ohw~v))hydz3Us$tY%QEjI0q2)wWU znGGwm6$39s*T>S040B|vziK{kEVITNTc!;0)`@#~I9`nIrDyz|6rgtgbYE8Yef2;_iOC4W{z1|ymWbama^BHJk7kAs z7$kNRE{C|C)1vW4yX9vNv*FyS(S%#Ob?U4P`MKgdYRL@r)GofZKn?ae^ooW?IjF=! z0Y5f&MzEV)67Fdm%$^3Q*AOy&Pb;#g#w0K6;*Y4OlGnShHvHY!TjQowWnDc*Jkf$_-8U3-kkQsE)MOeF(dkI+zo43WrLcl6 zpxNNcl=-V`eyBk3a>E;QE^>YRG%R2^HgKi80SDVbz9X!Xd%dBRJP2kod)7+!kG-h^ zO!ga4UKl`@XL0c|C`4Pm;_c z>mCU!s%lrGWe*!t69s_1-JaAzb{8ADuBQV`MCRCwA`Tf%~Q?K z>2CNe)}xUwIU^a}{%d#4bUy5MVb_$F5;W{Zc>0|*ke+18$wc;5PlmATHvu0`QJ_^x zUrj13TA1afWVgrF=EJ6T=}6I*FuV@NNK?$!Z5|BChC08X)Pa5pji!jyPW!aU7KXD^ zL!x0Vx1^K5vVCo>q^Vt3^8C~uU=@m(@I~N)hy|>co*ocj(&e_8Y<28en#;vI2iqzZ&ylH_ic9fGF(^g^1fwM z87V>Tc5!HIAtQ{$nj--rE_?r3j$!5VuCC{sHevjqbuA;(`EE&>gk`-;FiJ9031&!$ zFY06zpeDfHuk)ZrTCeeIj+kKriBCZa-l~3l zI3e9IBC?GIOjyvN7&Fem29|AV6+LJIM}`@#1EebblcZl3v|o zP!?D8yw8`d)vl(?OZm&aC&N-|lR1riONOA0Sb{|!(k@Q4uNK)Oc<)$@0}Wds-LD5B zfg?5FZ7STmpG|%@s9dV-QVIRxXcwXGh4-nq|H+TbG}f?XeOX+>8|^+L-FW`fTem8+ zgr+)`nJm*KUB^#;)f9e{@EQyc4|-||MM{sAMsv%ViETO$NK!QTSXOrY(1IGq?H{6oF;q@$AwhpKkVVLJ8_BUWP8NIB-oD$K z^rEjeT(SiHVyV`Nc3NPKPN;b{;*aaEXGq}4bR*SQt&R)w)FbvP4dI-(UOy35?x6a8|;^OQQmL*SgDGWmEzUt=jb$UZ{Biibt z8$w1MZzXzFQhEuLw2l$?eTb?3eI2uisaKSJLg*mrmI$4}CzXO0n6HWVxdoMRt%>ym zSWvyjCG5Vt@A#%YuIZ(?t?dIw-r_}yMcrN1h&+`!qM@9h_igH0m{$k$+d*{===1NcH~TXYuid#GwG(z;IP&&{7hu~FVQCRnu=^8j*-=c&cdH+=f$`5 z4j-g()&VNTvzHUB!NGmxMn4ACdb|ft1atWt9c>Tp>$b_Z7?|bD9kG3g==06Op~xuj zDl3MQ!dW?#smBTh+FB~)ogDKqhcKFAs*|?2DFQcG1z1zhV0-MZcyIXUQ?$hZ9&FcF zLwJ7^YIlMo#DsFT3%#&pMSJTv;V-&_*jLSEMATEa4w2ieQ_L{*tAZoqhBGzu@cNi$ z=_xUu&n>RYowA411kWx?-S;*` zHnm_g(~HCnMmImfxy76XW+vRm@|p*%AFGjIUW)hX7U*dS8q;#>7t)i+u0JnE5}N5` z*!-*tsS>v>mFX;%h;irZ^EDV9D`kywUPcUeIMy(#{#vYJRnE5|>N#n&{Z*3x9bdjAlsuRIyC`r_#^eJ_}uf|-N{Y+}Fe^=z> z$$w7Vrzf_G6hfFFqH6m!Ot`qSB?P|}Mt2LbnD1-SclZ6geqZG&UFOX<3$$hj#Dv>JJ5fRw%8SPJMyC-A2DBM#1AZ!3JdUIAehEbI8O zd4lGB{b!qeuAA{HQ>k4{i#Hp4CS1@XfdvMP7P4<0CmZKwtzpCnL1VMwB71C*?ryr0 zsj+AVK2v$On?dvIuFhih!$(Ff^&5dWJn#Fo9>z^GQVed)YvR0ZZcSnl6Ih}?I_>h` zw&y(Roe0~g9ob@yuGSGOi|S~%UaeJ?i)WA9AL&l4-AS@*!ND8XUtS~62a#3}Qp?r# z#VhJyirQLV&fyCxOrmZ&Xm61UhUus_ZDLR14@_esV{eI-O;0wF!~sZmeGFsCX044h z-RzC%(?ajL3Ro&il0?}A(-2-_<(!WpEehonnPbkz!M1fOr(@9*$y1^0%-%ltTWa>R z~_ZYmaf!Q-&u>H@J}@8beod#Na1QZVU*G0kX7;cg~ez%jW&x@ z)%rT#?^c1Ze`AuWN*n`5ADb%;e%MzF_oPm@n%428r)^b_%gMw==QoAXf5wz+RVy;g z{S6majh#cDs@hiE=819ar3T>ePudE(=ZnH!meuxR%T6Unocg!M_?)9WzpL0_>ugLi z%&`jvUC-C&o|k%89L0A@zMsjE9hS8C!8=A*5yCeU!Y#M_G9gia)vMU%K{o29>N_qp zj<+{Cij~mF^O^_QJkv|-wLX_X_|l9WOfv{`fWUk9=yN$&)}L5y{Wg{cV_MQz_K0}7 z<96FWz0*TTPT~ABIO!}{{R^8=!X$Ur-)Kdq-7+Lm2R=axZN>q8L2-+i8OYtB?O;lg z5V+)^Oa8KX|1wz9%%ELe3P(cYyj;3}3S9M5(t>kdP^WTzSya*=SwM$c3FvX z`7ES*(j2vXWxOt93l(#Djz=ForSCLWa30T{v;XB%St8bAZm;(zd_SK3Vri*7jl?2+ zN^2yY5yIZw`Wu)EZ)$D5CZOE9cn|Dq_7CbVCVCwBqt)Wxn>V6K_kn1e8-PlSU8z0R4Dw;iB3RD^z$ zBAB z6GV23zBK>{aVyny_Tmjx;}Qyny^G3@mi!fiuwP&dh`;_|N3r37U}snjAPmA@-fFay_x~Y)7A4vduMR72Pdz^f5jR zFEdj&@4EeHyKC`kgosVK zRCLc{AeEf7JHd|pU7GMc&N*BaK*I(hwgbwgBMx#U{Tjk2T@GOWVfR3oGqB?KDNgfk zy?qQ*PGWJs4~sR~Rn?(uz_Qr=!BzL+8_Tt2biZB570@sbtw~yklW@p$!lOcImAE)X zxY)HL*x$j~TEB=23Q;Lu?)&zA{=y)P424J)M=rwXuN-YWE|I&;BqUcLvxlq?nbYKk z-}vC@O<4K|eA?gB#voi zDOEN`b6u>Pmg3zRv+C8ifnhY< zqxh}mA`u6G0*yK2i`$G#&Freh;Vj- z7L1LjWfh$Z->}{X@z}Y=GDj+rNqjh-(D79mz9pA8Y{VX^9B9z}s^3??V@d|2Yi|(U z5p*UPYf+w%3%Z%L*1)u-=Rg#L#^7%shhnHUn=tsK7ZFYfLAKV}9$E6n z=7T9J(w^76+-Skg6>SIaAMQS)@pIoG2a9Ifl9CpZHN4pb%ux&o0YosSFhUKmK~o{_>S3?HIh$sA$---On?sE`O1pO zgw43ASvLHEFy))>-1VJb?l-PU;~OI_46cU+ubBN_s-n_jI>ai^li11%enb&G3fq4z zAG>E8x*n{M*yk7In63A$B$}m@ao|RA}NLV95(wi?8?1BeYTp<+(9Ly)Cz$&R%*>YYGGx;F4w zNy3CMfIH)bTPuU*(t8KDJ{1#&z#rNH#*zblxap#R$Is%EtaOj-w_g;f5vmzPKa9mY zz>L8vP4wEX$6TRIm97lpfki4x6S_g>4*8gGvFx<9#;{Qpgn4f;{ATseCH=1c14&I8 zUe0>oOtG^~9Iuw-Eo}mWAlGOxA}dkaRQBsp%UmH#+miGla-FbFxTWv+b4TBDVhl7J z5@%CiieWv^l1EAMP5T`2_vJjugxi<uo25#V>~ zoPG48qJ-ik(FHoB!i|SpxqE6UO|WCH@VET{U-1g(0Oz*%5pg|j9P5~KnoF-rEb^rH zlL88H5aIbLglSZ>eNks@>;rF$J}OA-EEl`o2!pcDd8KsaVK@X3)u(BfCf%f zHsO{!0T3}iR>28BV7`^`xwaTzGx%#`4WQ-@hf4F?y)fX&gq@2I_8WaOMzJ*Vs7cV_ zH@P+(%bm_p{k-VxmO0QrpmY&v5ZRw_H#(4Y5oqSZfG~%EZP2Bht?XJ(Wyi>l_Iz$E zlCbqO!gJ3IU)RpzhcCz%N{iQ^uwnf9G3H}+55y4WRF!3 z1ih#TAgz|lK@cK_E{wavsG!U2a%@2d8-SC*NBE8E(z(ng{`>`qLMjH9T&<-P26GQ> zslnvH{*5Kr1D-~&NIKVLe{{d(iF%hep?NPb6LDrtJwV_-MCi^{i1vy`&?I1}p(A`R``1yBn-_Eq&=tM4#lCnX!6F#o6lq!>kBH*W)T96-Kzq#)&&B{I&}~i)T6KdR^mnPhtle!PmQnqoC(>HS*^3v?&Bbw7JM~hPc(u4 z!c>N=cz+3Ts1j*ZeCuu#!3()61Tz60+w zxqsTT!y|8gaxmpTHFf}!Fr2$}J64hVPr7|WS=@Be1$KGO4A8zX^8~`#2JKBu+HQRS zK~1dCo)6Hfxcjr*9*(+7mNP2&o#5qUSUAZXL$-Cv3X#$S>xQj_C*laVZ!>?Mv)^Ru z>+SG%9507aAEr74F4c{X&#PKoF_o9bGkq-&#TZj$6dvF}bND_kNlYC0WZZhnOik zfufN<=J^AFX)Fmzw;7x$Y!FUVx|FM}M~ke!N{|X*8Lyq6NwBJrE`P7}L$@n95I0jU z03BixzRqRqJFZkM_21SK!tO_^C`0%PVCKajUBV-C`ZB3wezX{mof%(Tky#DT$hsl= zwrMj&VC5DChI&MH@H}jb61?rNGZ7>9?Xm2Bb$i>FN{9_02r2a=_#fTePv9h%N*OJ9 z>Wq%eHx%W zfSe{g5YyGhSQ^u`Ja~8eEmJYK~yjU?{{h8Dlpg zc{wo}3DLt4FhIkH8;o$T!i&er#8#|TAxoBLL5cFs$;z1xX4&*7pBko}(P_$M3iTtl zCxyuV;|;=a8oH+hUyAE&%kp8Mxzz$k)9L*5chCDHj}6-reyYIbLcY4h-Fp#R<-N72IdnrnBtz$KKWkJDyn3Y4ETK)bN}2yg}n;!M9-(2BF-F^ojbP-UL~>1%LI zWqpd?t#?$HMTaNL%Anays)|LZO;1}M8(*B8{E)~1s#p$avwUa!&eOck7$I(El&UGA z|9}4?=#9Ef^GJc^JchCRSA}KCq$tw`vIMy)l)CkLWL(q# zM)j@eK2-(4Y+T4pG@N)=&~d{mzxdTsa}k}N4c0bXsuf!1 zSFHZH3A=0#`(E@_H#@H@<*(rP;Txy#wQf}?>~Tjfl|T@JHkq0_q;E2Ep8wYK9}w_v{{k+aQm8Rks}kGlBemE$IBX{C-}h_Lu-hZ& zu$1ne@m_8&Qg{t$s{Za~*#cw;!>BQ-SVA>8e^?YemhU!dnT~ZEwU-<_*252*`1%6oGK6k zBb8b!>mesC2&!W8O;4GjCTZcP`50!x(eYxI6OTLTtl)^_ut>S+=41b10MLevY?`p-02olLN%e?O42K z9PwQv->cH3wdYf4uKYNqm_HC7fbb~*SpFZt-*~Fu&BaLs`YgSwCSnom6tUgU| zL`Z;Ct$Q)AbOns^?4_r$cANS-OeL__8av*Fx!M@~~tbzwmdg zUSDA)AQjb5m{t1UU)0i@spN|wQ+>ng^(S+GI*97}Fs4`)z?m;mkSn{1`S69jPFL&b zk+hi3>$0X|yLKf{BHl24ejV$u4=d}+wm+^NbH|Q3s0dLdK0ANIHmQe<%$;X`$|r`D z^yvEogm(o2Xc@zu?2#R7{vM(kX*m44P~dcdMk<^1_u)RD7c-2+1X|njp;60^-792I zI&J6gH1=8I#)0{@tu7YGFLVqI8zbuRN?c75yMSI8D0Pw61?wkSIqRy7y^tlZJ?WLj1@i7 zyOB0A-xMk0p#%jF->7Wd!rB2_p?TF%YZ$qr~ujDd_04j zIwn|iaOS1giAsW}&^|cvwuQLC`h~D$CbL#A*?QVMGFH-Ooq~G%kGBjl&zgccCj`fx z;4HJEsF|Jp#v)ZvK+%p^Aj!AsEo}`aOf2tcHGaIN4j~G=7x2O{y6P>ORuysQM_6lA zM714B0NZ|cUYXdK+gL1u_Lo>{dX#7Av$+F5L&HK6Q%;{vnG$!;w>j@@FVutEXnJCC z3?mMQTd1uhZ~p&XgQ-9WG-6+~_f$@^d<+2dZCT=yT6_PIN}8_+m{2t7)1Co4;K;AS zmV`>HnsS2Q#G9^b{_>TpQ>eJ>WtosoMU3uAb{C_= zDxFo*0)28`()!JXo~|5e`bLPrd$=?)2W8r6A`)PvO?d7y`i|^C8mNy3wvc z`9c%-lIylvb0m@y&h4WAOS8VgYUJ--c59-D#%-ZTDSiDOf0}yflKZ%^^Mya%IBy7aJPsf#UMg1hYS>wvmd8(DG|hTi54NtERMc z)o_ljCUTHQ8Df^mg}NW=BQdpJCO&47h*GKNnJfD5;P6lZYY0N`m~wAQX^6dMkoH?r zU?+mT^q%Lh{KGK@1WN|U%>2FevEgp%-TQjUE)n^TO^O2#O?iwI53tPULmSU5A7F($ z+DVD&t53Qve%T&(D6mO5B#jh*wKvEZyMD`K&&E4R_q%T==v0@z+Snj;f9wiu(+oor7HSi0Ka(2=N z|83SqL{j+K4b|*B!3tk|=o|dd2BdnA1>7sV_IX`8o<1+JtQh+E^QO(fm-aDN7_F&P z_+4!%j;e@{gVMa$$!Wj&cVHgP7eJLUMgRD`$`SphHMwBh&3g>|TDqhReT`>-+-W%4~9@#}kX3MH|I%OXZvT?UDw~yD57l)y~ zfM1!Anz~xHJ|oG=T|rq7oFq;75Sd))FBRs4f;L5=x=X^H&)039elAUR?@dJMNjBIP zf6w;!;_#p+92y*Lw5v+gr7xpmkANSkz=-pc%o3#fdHsuYwFI-p;Si21xl=7H1$Hn9 z_0#eZO8yzq-Uo0fHFdJx_uI-<@)9BGE4w{GWw&j$q2UjZr_ZMhUJ?f4-WKAmmzx!e z+Lak(x~?THt(;#?c=yCOIQS9aQ@X+j99OQrxvT4SuA~$G5SIF4XYYSCiHDe&3*ZkY zSP+9-QTr&9#ko^(A2jAt;MKWh-6>=+4SyMEjsRLBd=6RY3q%zvgPpKB!j_g`c3nVp z1`n0<2M+5)ZbrOO?Js5~U^G=yBsOzVp$$EZc~~1+Za*&3lOkqdlBc)i?cfhN-s)|W zCf1!)Y6;tdQNAIW(Z`8}5UGwwvr*x(vImY_58+ZpB*oJ-`#HTm2x&ZWzORhIP?O3? zI~pUjx;>q{1!r!+1lh(mirP}_w!JV9|DkM{uCA%`eep{*+u12!+4!u|Ys_!KpRQhp z9y3H$$1C6M08{Z`@=Dn&3ZH7BNw+b($*qbc%6HijtL>K?z)PcvF#sp59_x?;YhA(p zOD#8l?V;(gkrN9YyK)y%fggIaCCDgZ0?a~JrH^K`P!8`v_AP&xxL)!SwhKC2li*Ea zVlpQS@R+2l`7UFX>j|2|dO|~=oGjJ3D=rvnmiPH|;-q!n0BU{RqNVFts_37!N6$*U zSSBv=Wxw?|@>XRWIW{sxcIW!#INNIBs{*#_dBd)^+OG_#Xsxu+LnlVMGokH=2Yzj^F-HjS?pCY0PL})X-V*+RQXE+ThN4-*#nl@kLH^?FC8^<3`T2y90*+@=3 zcJyV~-k+j+HyrygF~!ZDMk%xldfwuP*x1~YLJ8P$UPu4*V&L3%)$w1_47`Hc(Y2k9 zF27x-5~GXJHJn*cV#DQrdSbKjI6|Ox@D)r!h{NFDQSIsi9+wWo`p!}B+jb|wR(El! zMIc$sE~N$-b-4ljV3a+S*)&hDY@hv6wCwSVs_j;$bs!;rKZI4$i7^wW<8u_9 zp2FaqGRj3?bPAncX-4cqD{k24`vm`2Z@%OYxUqZz_;*V;kaS*L7@hCX3((QNV+v-X zQyf5`wv45lzq8&zoCd zDwxHHH=QKdVEpSXd0+8(1V*F3DZFZ3l z^CGPjF*||B5}z~-zk6xqH~lFlQ|c5c`6QRj=S-?F{X5KlM7!jjT(e66E!>SKP4TVc|_N(ciT4LD@xmn(|*+-oSf1Vo7!*e!$A&r2xg4?;Au4_3q+ zq6`hkY~7?mAmM}@SGe@=LCGUysREE0=e5T-JqXRwA{kfYPIymq{ubVl_xm~%a<`LH z#QcOOp&4}#W2maZiEH9yn{k7mu`kZ~4}GSoS$74nEDQN^zek>zzFa6EHze3*1$->FG?Ym&slffDnahB!y{lF@AS7K; zlcBk(ViFV#OI2>T_e?Mfb+SxaqPv-Va2|R+c9V;1k^EKHqK~=Q413wcWBV$0`}Z@J z5&j|_=`;JfD_9GW7mo<YO zNwgQx)vZU0*CEJ)9qSNMaMEZddOwoZ-Z;*#+H*-RKawo9$jlbFnyML>%Ae3aa!vOK zsVSm1php1|)jy#q(1$iBjh5P=K7yF5;yWJq2gHs>oHDFiwGf#zC3!MVNJ{4c_w+cKeOyzh39xthZ1DbN8dC6NG1Xxd z)H!z`hTzQOA!c7Uv;X!wXE$Y7UNQkdl z;C>eRgGxsH+mzdk!A)k(5La0^Y&_MddX>`v)pJ0VrGoiJLEyqruQ!C1u>FJ+geTHj zUHq?j<)*6hS2P15u9IBMgnjP^g?&B_Vl8WoAz^^Rz#tKrh+yA?13bpO=iw#(r%h`C zB%L)$onY$j3%Y$*V{kjVx0#0`IUu4jHk2R&(sNvdK>D6KiNCUZlm)1h&@=o{&=j@Dp`#;%MiN`C+A>peuyo~*g1Tfe24bh=byp8sDAjeS@Nh2hTLeKb`tx%vTS_HiR75jVWQ6`-HuS&s z59KaQTITIL@2)o%D<(ojtD6VSJm6f-D7W}gut3-}9WvJ|StQUO9m&fbUl!%C%-Zq9ET;3UaJeH`9s1kD$SP7IXlv8ap*<`?|{f>^NAV zz~{iY<_jy^-$;UYaB@|#!AXSO=k@XIFk_&axx5>TX3~6ie)<{Pn+@wAWxReI@Eiiy zYwqFj@azA%fLyzz7zx07+4H?m>`F+GP(^t{z!Nd@?pliy?=4vGoc?`ZnR{4)J-4L0 zcpW>|y)y9C>=Mp4@R$h({GFH)u&vxI>-zcA$u}8qA5^SKR?m6(@O#FECJP@h_PPNn zBb%Mby3_jDI*qG@7+wu3hPr6To zK3=P8w_T7f+mt9-c&sFwT-vo#tf^VjYG0=+Gk#%burS=5H|FT>fir!?X>T0bXrI97 zGJ0 z6%*DHZIBInh#KV+zb1`3nSBQfU<>-Vpz`k(V;-1+R=W&XZyW|xVtxZcl7L*juC}xm zmrlAOMNs7_n0ryrP5(4^uyRHK$qg_@Q1UP&^g>?`<0xf&PTr~^eahb0#1rngl~)#+ z`4&Fdw~Ktd-v+sB2)RJro5h0>T+2Aek(k}Ym)mJ9zjw#@xBFDE-A~SR57{*DPNW%c zrlC$~U;GwB-&WXmL|<{=E#s-$1_HJVdhy>^$5*T za{VjZUvxAim$|$Fuzz|Al(sff8|B&W3&MTYB{ivJ4?m;>Z;*3$b1SPc4XJhWI?$i4 z0W-*Ahld3cknHNS{jsszOKD~xeA6G8L)W>SeR>DHpx=+mfNKW`+EVsZ;eZ@+6%W4w zz+b@SSZw(skh@_#*|Lw*sEO5*B}UxFJxRA_-*SXv{71ZA?K@o!GhtgzesTdyec2J< zltw2_c6}Th)mbbilk@UNV#UjTs)U-84b&HKsZkKTFl&xdLmJRH&3#JjbQ}nmFUT^9 z{j{cK^TB*#*XnrJz6Y7)Fi_2zsuo7Dh>d5jW7uJP@H(}Q`M$@J`1(& zr~lnk4?+X#JYeJRVnd>7fKKkeWu~W=6Y#!5e|c&QZ5(5bz6K*Poo+r1!#VgGySJE4 zO*dEy)FK*bHB)}HX0v6Sv|F8@qDE8fxbH*QHDFhE*2A@inH~N-ETNJUIhl_yg!a6~ zkbKslXsmNV>2qvwDiag)69#y!&t344nTy=w{E&@`-84eTZ=Z*<+w^`Yj6C~2Q#NB4 zTP>I))KXq~Bir_0OAK4+X`wd)lFOry@BP6cswu1f)>xMR_%C2E9zU$FS}TrN%^oQB zJ3Ah60AmK2^iK4rtjmFRN8r@=_-~PLw?tDY`T4kO_s-7FsTbDFd!GilBjIt%Jqs>u zESVHcJx&{G)UtQV)J@PcsQPsDuz@8;*v9$8NnJ2tiJ__O$Hnj4h(gDNB%Bv|EF)^g z^(22T=ou#j=#@li%YA~aTR7BO@7EKwsEvCpykkQsOC!K3L~_oy?kxO_6eED9SzK|5 zOw(e9Tx`)|>>?WqJSf+O_>`NCw|K<_m4k3K2-{oVRiR*h#u)LyH3N=dW{?37%Ah}MUZ8MYlfx{+B3HHHUK<%J*PelrKT zehstu-)H!(0XP^NyHB_#z6XKVsp}xn<6!q)9MbQwA;V*q^m4Om8zZw{c)Yhd23+h` zdl-8B0qH%N{=7jfpWrVjlA{;c7=0`IidAWonEH4&R-$d*q-!mqu&m#29R2g78w-Dg z7sANLY0F+Cys?5d?&Z4Q&pH`l=b7)+^SV}CQ8u`V+!0DkJAaFz865zNb^$D$jLhIU z932{VzYH2T-OAv2dZWU&@Co}qc%$B%dE1ojl2~^30W(JMmI)%@;*)+-&^Ik&t!@@X zZQPGY_Di+08`<=uJJ=>%P8h@li2FGHG}?!BK~E6fRF7mxehocwuFvL*d%AVKeArOo zkJ{nPZxIF_fw$bmhMOgt2y|4}JB7jjJ)mOFHqr4*A3CX%wUV<75Rq9E!$VC%7O`W& z&k11xZec@US!)4cEFN%yEv*6G1oqO{&aLI@GaTqHEd39gkkyrmeWws2tL`-GTTgR) z!Ms9*xTzLHnf;!DrKZR@62Q0TCzj5OXbbceRkk~SOdH9}9;&cmGT&9c-HY^yPSJ%v z8X0LYzFLD1O0JpB@m!#q+{(!{jKhctSLsBG!i2B5#9~5Cu?R8Qbe525;NR@{rVU%sIgWG zrQCBP#=ZQW=i?D~nTQd?WjHVS2m~3VgHG4X4aC)t+E=YF9zV#R1K0bLut?sL+~Z5O z4BVLSST+=Mwt0V`C$)Ewg*pvSrIe1htuGB(47>GPvO}G6 zI!N$lRRhzZU_ckj8jw3k0|R_+SV|E;Y|P}Kv5L(kw4c-dEwHNy8wiky-8C)b5AtUK z{DClnlV{}ao-k6?o&adZ;H~V%K$)v!JgnCarrvK^C;)YKC<1-8)p>e*W4q~wu-mf) z>-&NK8;nCQLx}Thk6+4ZevG~3M3TWYFt1c*LRy|r!(nr>oCgepa$bL5qzak=L zx>@EOh}$@0Qdy0$E%}yFShrP1q5!rGyqcs!YZ%SY>WiTd>RS>xpVwmo7E?KAjc%#N zyA9?M>9(NM2-N8d!^9@|pD;Ty?bOTe|KtX6RzX;Kk)W|dlVNCHR6`lGS;gay3nj-Z zM?a@<7XGViM$p!l>mkOm);}V`QYiKxWfia;EAav}*)p`e?M2yyqJ$~25g?1p?<= z#TIgp+C&4k&OYCCi;bfeYu6>|O4iENn1*@-ZoaBDQ*0Y6`v8$4UUJ(Xv`(Qy(5#9t zZ#+)(-#Lr>1xVg6+vXNq735AY&NnYkeLUsoUtjv512oVI@aB?MFMgCpvq+7`sj)6c zc{$V`Exv4eROPn5_N?ys-PvA&e<|4HcUg9&SkQ25{DqSha;|}bI`mfS3E0FMG@KZL zJw;`P-O!w(II#9SpP@iqt)?vIt%_5R>WUx{F6m0y6JPbo(RJUKY0KRUSZtzkz{eFs z=G2=h2tAp|8*8cLpm1@-hpHl21p(7o+0>37`Aj5SkouKGv_c!5T61)9@p@Q_LXDRK z59V3Dx`62LJ_cJRP|7preAN zNg3QcR1cUT4m_mP^!313{!mtA9Aoy7fARwHQ7_FdUjlY+qSw^GJ1gHE3~Rm5tx1Wk zMvxMwGQcGeY3d>r)x>04&ml6AY6^-JX4w{nVul1Hl=|@yl}LAmA(Og=RK%=eFyi0N zFU#jPz$<(Q;;!d&-Wh;Plhp#~{%5JFGe_KqEtnX>bptdd_Lmvy*(W5G8Nw!!Fp_;R z5(7lVl7dmhk!5{xsK(oUwJ$rg)M08;uDmpCyD<3dO!<(W}z*} znUQQ|AcArX~Ep{Fo8aRoR7~L4;9Q!#G5Xp`P#m4Ku)I6A2LfgPM&O=ch-L=ZC@Z%sPkMe%2QWfy!Bn}E*<}*s zdibsUtw~bu={FO7g(ga4&OI{gz+I~pXkLED%%_VxWFSW_wPB-uYDhc1*?VA@01`J5 zbJ`<_AB9zXIfkp&trTHUsc4hg=bL&&Jd6yqEPomAA}F#b+hH^zSQ4T1qkclXdB2nbSx0v0LVAt;S>H`3kt-H&Un{o8BrwZG$d|9D?tfrokS`-<~C zuk%W~IBm(TJ$Hi}EJ8_A4er?|cYbKP@fblFcAq49w(;(_r^N##&)>7|J^mQ%{x*hc zouQP!(=`1VzECqN+>en^ElTWUfvL#2u^q^YR%7os#KrswNQH}VVDN~Q8{~+Rsi?E) zWaXC`l_}YxdcMX#M$f$$;xaB93#SmV!2gt*rgWvJNUt}8>_Y-6D;MFXTaJP1dAyn& z0TXbn#KSffM;Gt+U>rzpR{#OKrl42#Pyv1>GMI7vNB-g^NGTOo0OZIOcr*!WFG{E; zUk%|*&4=6ctmiz+jufx^7yNPL>(3a%Snob+RXPmzeH!Un6~CP<;lUoL9o30O4(cd* zvGPz7?d8|A!opcjTK)PwdadlKR-J~n^ABS3*Qm4@(X1yA%n(CJ;TEUScWYJln+Wwd z3amYsaPU9sCZ)b>PfK5~t%3CA`r+aN8+M~m9#kV<266m7VF*p}7$dBD{4a9cJRAtu zhLi>aVCD_3kr+ivOWouX%Xz=#@fR(fOx1`g9~&^>VwiI1ZOX;PwaG0nKY4$a|BmPk z5Xs1SR-Bz}c$Wsly@n{H77+rPW1y=U98oK+oS81MWjfq!sAgrSH{dKR6dpfYe%X;! zTsSrsz_0h@+0MfC)12v#cl>=}vonQe#2dF@rWvE-e-cRAgs7d?o<8$Dp$Aj17+k9L z0{+YDs9-kL#OhPK1|pp(D%yIZpI=@$rY^Z=Ur=tzjLjMW<%Clm&O)o3pq9CL5LFk} zq3L-NWhA6PARwi8*m`L7wXaZJWvGpdI`{ z`dt~u0W%P4tG=YE-FG*KH@@OBo>k~Ok>^iGORPMl-zxI+(1_Bs;-3*7bD6e8?y(HG zcz0(+>|t_rno2lWxtqhUn1(S3?62J&NtTq9d|a4YS2?NgJU;&i?lOfVz4Eo_)I%!H zS%y3-D-ZA+u;6E#|Wk%pvhIQP%GGp7~WtTO$fetG_af;26dMO zq>{O)YYrpJ!@c>{WHwaE??H4^uscrz9)(1mJIjmV!jBgUrL2f^<@XqgIdCzBl@LAk z_xLfVU8MWf2NOFHax>A7%3kbl+2%_z9TZgBbC;d&o|sO`zl?IOm2PiI>x$bj#0dx)+JEXJ=-eV8-n6MgM&`r2; zbfZh@i+0}6V3z4I7$Z`KU8cYys$96EJaDu2$`y+nsO1f=$TAiH+&ykcHga}alY%Gt znRI%_+%RsOQl^G4!B~3L6=AAaFSMh+NXFewgK2*`X158}j%4x4Dd&(JFQda$Y3E!4 zcpR~uaIe?{eY8kMZu)o#PLdc{fJg38UPtsz82s5?#PQ55{=8oaz8RLqFj*ynqeS#d z0BkgAm-EY0{jF|gO4lvYF|fGVssFL|kt?I5-JAf*cQ-5yPZ{wt*{WnK)B;dKHwRG)3hU9g!EXO>0cn=%1_L`e7B4!nAS9I)tTt84~mS|$ZCIl0wxRe{N9)RG%qb3e* z4FKOxIQ^g~=Pp3ycxd`n*ohY7F%BdkDwRbQHW=9yEJ3(r3D9@AOTYZAU4KsnS zp>qz(lo3*as;P@Vc$Rqv)nNP2Ias=CJ$cl0Dco5?hNcRLCeD92gSFI0sIQw8%u;d3 znRumyZ+W(UY#AH^SkrA*7U?r1RZc`>4EDl&pN*2AkaAjGBtPF0({35<_=pORPZ}U> zDAx3y3NnayIZ4c#Xze?~BIQchE$3VLbPJGRci{NVws?WAeWm*2y`8Twd+D^YHBv1S z8Ok&lJ5wl6-{tlgv@LXgv+B(rES0ohKK6iZ{;-b-F_V-ekn8{a*a)oNF6%T;6nlj_ z=ly0-?ICcrz4a*|fX;+$I~JLmcAh8qO%~qx$?Ihe)am!{&RrLKfA}@NBp247)_1Hs zuY}Un#>}Ma<=n++f`4=P?Do#(VBC2v)qQ<7T796m3T^Tr zTRh|Ap@CjB%8{BFSL`?Gjk3C~4Y0UlAE02m(t4v|_htZV^gUKX>X(|Hahdo&_Ixx}3r31kE^Oh~AF(3GK!N3%b6pD2|Cn7J4o{YDmTB0mxbwOJ4Z)dOGwIp4th0j?3 zus|5!FUwkO^9&hOvy|{qS6Z#~II@%f@rXO*xn93xM}V*|C)d?)>{8n`;ySLf~OMke#Y}56mh;WY>6O!gajuJq2+{51^7>uj%6U&}}Ctn_L-&-;B`s1Sc26`sIj^X z`_GG%qgRv^Et6bm9rV564;gI{PGSe4xzzt>M7YOJ-)mXXghLc?6u35^i)Q%V2v;cVuus*e68 zCsWPF1vzgo!4IBCABUq^1G>mni(hn`GI@zT8mR@cSgXo!U?D&tx6-TCZAOCO;5Fnb z15JM9+9AacA$tDXE@-5Zv~%*C-hi~{vZK@J=rdy3Yh+O_1kaw?G!PNKe=FL%CSISS zowXw9RG=D>ROua3eEBSqQ6#=#J+%eZu{F*`0$u{euL9=TPu zGfiZQ8$lyyvmL&(rO_wjVu(1*D7ZbNj11~LR7B@(C8>EFn7;UsJ0)8?AFrFsl9LW{ z%=ARB`FjalELa|ZRUx&;sGepXsDWdsSZkbRv3WLkwe9$+YgQlg$EplyXbzTAvC(oM5PM1p{?YW}72C z^{FBredZNQJ#4p)D4;QBLh_0ED^L-c5XGrKh!dqIel0_^2P_0uk0@Ui79AMPuElzcs ziW#tiLBGq1MAid@qCc!-D$o1ui@tq_D41uI(QB+dTeEfo`!0sTyiA&2#gaPv6M_&h z3Pil^OYYA!9RV!s6X)nH(x>-2r8kD4*^ERZx!);ZL7|*gB*pT~k1$WJX90eM%ns4; z42{M-DgOuLF{2EIiJs@K>O_~R^4{V*R;|RkvN1Q5Fguf_ZVJ|I-9)@LWfusNcuhmx zs25JFU$6_MwO=?gaw(y?;kPcOBxQK2X-g7VVg|OEB}KsD0g)elCj$`C zE6#H<3#oyd$*tuSU*dt6X7nfOKO^b>;i(r0{AvzQUnLc!;>Jg>3hr-9M zCv!kzT>O4rT4!jqec$1Cx1%@}o$7TUIfm5o)miAhWhv@6K}QWd{Y7ciVe| zh9^`%|8gS4GgOcMnWMIhR@~PmWdsAdYemdKos76ydnJ1H^Fy%sx=I3H`ns6Rf>6JS z|5uIwpV)E;__YyG46Oelmi8FbcXmDbG<4JJWOQ-_sPs0n0oH46x0Lnh3d--J^K(v_ z@JoL_{)lLB=~OFmbbX=N-sd^hN@k6jK0fkDil4^5CUQvdo!u>dN-SmJJRbCMKl_MpnIKCiJs&V zOIbc&4JGHSzMB(H^q?H3^p$d%pr0I>^7$!;AO~Bn(qZU$;=K9^50tV4ejL57Qbw4ah)(^HFkfNzixh${2!A5z{*0g3P|9-5cmZ! zaiDT7HO%m+Dt%rQ^n=~A;xN@Tq2El?@X#5FtKU9Q^2-ElDUOYSf|y6EpHhwqkIhXj zTo90~fF5O>5B9%kisP#W$ZG?O3(OaeH~LjTC>= zuXGv!z+3I)0^&^B#_?RR%EOhYSIEPKNT{j+afAKY)}J&W7rvJm#ez>oIHLj+@coqV zW;8g1(wKQ-Z<8AHTmj{Txu|d?W%VO&2}1z`mJVCVru zH#aX~?pb%qZZ(5Vce%Ad!Ni6W;KN?S0K3JmN>#7bY`1DTbZNAdQcQWkl(J7j_n%(~ zV`|%;Z>uJ(zZJ_-F87Mp9YCo1hP{qrD(?U={p215jb1G^to533pIYq6g!+RbFooKe zbuW3Lnq*;qez^Ueh}GQBn0+8?8U^D6OUg7wl(>x@*Qvh}r+Xtb2^$ouPN_`F_!Jz} z4=DqD6B&znQ*F)9U#Fucuc=AyoB*M;U&{@B zzkl(i&;w(ZiSXPDe3@;0hE*5>F;{CZJTYa8-^nlc3jd7nZo<2q_!l0-a4M~WdC7t= zLUr+uWQf<}QePWQ|2gvoBw0dWzxGhRajx~dX<4HEI`_9+kG6MqJW)Cz(Yr+Xj8{6B z)_9&A|O>&a2+62&>w&{Q;K8?}lEG zkZyICHIGg<2VK$N3_3AuePKxE%D4eLndZ)u0271?_`ZzfSGe03jlz8~q*So@=EW8v z-Y>J9_-}}Cui$iGUZ~!&w+apU`cFI?y4k4WOUZCCLFDVz0yRw8))&KfYqj1m0^!7X zoJP#WA&b94bbaFxbwQf(okn|vzqL+_-D)6NE8RbW0+{m(66uoZy-dln`ezFv5#3{f zZGR{2{Mt?fsW|6pyOWxN4=JmvLKghRPh(An~u$1s>xSi=?En zjo{>EM)mi}Oj*Sigwt@W(&Z84wXxUIMZhq^KlF-+UXo(dzaF`%q6}Kv0~FBvJ5X`G z#-&;m(xF5qO4jkkIPu;)ZZVh=0gX9hmk$yCeVHxPD1afY2EOXZeczbXj8mb=u z9lZM2Q=mEmFia>WR^-ptxAh#q3d^ohvEoG*aL@QgUxeKJhvyAAk(ML?C#(Ex4|EQ? z)FJ@H4|auhDAN*)z5Lx|$0+)wBU;^tjdj%q0Qpr!HbBYg)U8MeZZxwwEK7d%81&p~5)zEZ7=GWZBocoWY!)5v)Qa>%HuM#(5~Vi86qxR}sqrJ~E!y zJxCNnk#Q)`kVy6#+nfAv|N3hMpJ4U7+id~|JJ8g&J}?>pc(ClG_eJ$zu>jEAJw>C@ z#o$u~Q4rUK0G#y911^`X9}J^FeKMs?{&&3xkfOU2YydUFtSk|dhc;Q`+Q+c^Z+rsW zN5T~eRdkn8{|vPd0TV;wq#WR*`}ztp+F{WM%86$u$v~mqKzax9Ezw{-zmXD#lL z#+F~PX_V{$kY}XMT`l+&&MZ9u%(u9KaH*~8QQ2kkt4{z(11!w$?+r48+PfCP#@6_% z|Fao6lh}89{-=kqCKodH;@9IN3|C2LM)SV`^u3Ir%H^vQVWrBrLeQRL^xq!~Gj$!NhC6h<(uurRHR28LKQ&%T7-1uz zrkt>!!m5N>LXaD-B3i17o z%eo~3oay}UlC6IpCyb^I!sh`fCPleLM9Lrj>#Yx=v~u;9WQXr13f8NMs33l7p1uKi z_OL}>^@8mA*|K(RA($7{0Y`ro!xotjHAm1o)}i;3_NOH%f7#Xi##QLnKx90~Biina zRY1gT2_k9{G!kv~MH=7)JZ)OK@0eZEbtY@fRaAZjG7$lql}vEdc@hEHAN=>v@gGkM z;A{UW00b}p0V4gE8@C=L0%){r!t2E9?`#aH;7h>7)NjCg=ewNl;4(l(09cZK)BR{ZVQs`BuNb`Rp{GT%C>Yv#w_ z{ApD>ZVtf>R>l2V6K#_qrCI~JySRF@fpi~O2pw4^-+XBTP>rawX-e;WXt$ud(fj`~ zHu8FHwDe#!6oS-l`*lYNHjY;&f4PR%1a#PYvlTpFRs%fnE3D>ZwmXgYf4##{bi5bK z^dCM&kSO0No&p6Q<>3)fQMI_!ekGn;SQr6Qo|X{j``32Ts(Nn#?D}9E1#cWoGzlC` z(6&S{{VDxsSX;AJ(8A*M>w3Zv6qV&qH~l-(B;3-{mTlh+WM?9h_cpA-fj0-ANiJ3hbg!0d zP}KRKufJ(wA-8gU+t|UkyBvyqX?d^b{`?t*$n}fk!P?buktErl%#vulfu1orZsO0yUxgZjag;n4BzzXcsUgVgyC{6aW$pFh`k~_(RLF zT$64R8~YzjjIOqebUdat&rV~hB4u%V0mEbQDg$vOImI!4n1CTM#lvnBG4v@`hKS^U zKJVWHffwcC((DiAj`Qt=+9!VbN&P-x!^!V+;3_`^No>JmRGR@AF4z*e3XrL4Sunu> z$=ho6&CRpc%4Q??p>5Fak4}M?jOC9|y1zvW$kzW*);H)TazS1pk5d=Xuniy1lxIQt zR|^n3E!RxJK4e8fjNs;KQHR$-CCr?W(ho$@?jQgAg9Q(2`I;!!9mvsi(dZav&)uGh zMuW!$_3#FHd>4AlH9iccBO$z5QQFtBFo_0;Cpv+4(8kg~&XBwiXjH8tn8Zs!$Q|)ITQM93l%N?{r4YjpHdb4Y0ttsS z(1EN00az(eSl7k#>6lDMQ~ezTB>x`xU$9eQ}be4>~)*ml6)q>slIv zQx7&6$WK6y3{1k$F#J!y|Fj-0w|_3ZkZ8J^mHc|!IQCCh8t~8A^p^$C`S<>^fc>_6 z#9Q1v%x~@z|A+Gn44BPvDg6S*1EGlIcCfzor!9G@H+vMMOA~K!hSvZbUjR5~-E55% z5O^XBfgG%7!fk~=8Sqq-B>?Sn0;KdPKy9ocwyz^qc9i38;#AiKrT3vf)N%}bxJ6U@ zE$0+cV*iH^0647q0tBtaz++2A{sPnte^;&k5f`^#XfF_-aA`c>FRdL0(k5eODTjQJ zk%sfU_IOW#Q1;c|;kqQzHtR6$MkuuBR_{NBV!hSK_m^U3t+d_~-p9RZfW)BuTXk5e z^rOHO)qi{lF4Nrdg{voELT~mDNW!n-4bJxYM)kPv5W@mSkM@ltDjk+g_;pLZLreD3 z8t({HKYNs|u{SDaJu&uQ6HJgk85%A#2f|c;LQ2rhRd!NHuL@d>mVBGk(uDu@{P_=0 zfDy(=BMeS2S$lw3SCclW8Gl3n70?W4Ks>S2?^~Q;NKO_|PFW@3;YI**{}QlR@9({r z=&>FJ5qzV36U6UQkTyraOz3m@i_;=XUXMm{@gX29c*!j7v^yHQ9P!sX;y?WMXjR0| zN|2?_WTu0_D|C~u&G@QU{#)*B_faJM@xu21HXD-s>k-N6HsfN zl{l^p6ho~AWFFuS2J@_T7kf%d#{TuLVjV#mL#qkq4kSTcn; z=a~@B<0H06i3XtYk&~~LBOxV)!jIR<ecnZNDh>lcXcQTh`2!WBK;Ap*o#wRSnSVST40g?wJ682|y18T0=0bbdQZ$Cn;lOua93jTgQOPD>Pt)xfuzGg z-Sez1LEs<*tX8_qWvoP^mXQHV_WM6S$(UK*m{9e zy`10-yfv+zK@T$$&8mFrgdu^0K{@FBdAgX$kbyO}(dzkEW7XN;*kcPDtAC$SU(YCH z)l|d0QRipkVMA7vL2|3x$i4B53Z)MTctGy<|IV~n_~hBT^Zw~h7v}}3!*buRNKivp zP6siqLRB;*O)RsWT?PIfg>5$~+d-r9474P7U-LIDK(C00(KDQCsmZQR(@g0@rX4Gi zhY%~Fc>~4c_8j8~9^(J~91yfRp+RQ%`M27j4yGuAE;?w`Ek&Mpeg|lb>-5xPZGAKF z5vPe-4Dgi)!|41`e*NKCEK+{n|GsL_Yil#kiR?ah6d&oQdF4J!VG$AS1cJB3RJR0Z zsNFHU!iwXpuf&$D?!6_(DbUX-$-*?*MJJ(#8yzFoieFSx0W?M7m7(U@Lv&2jcD5ZC z&(+rS*36t258F~(Gjd*-R3|0{j3I!Gic6Y4OmH)XnlKn zy%dJsMzmRr2DaV}I{U16`XJq%0~zPP%l=BNzzgvy7Z&q`L!J7ZC_4QKZE$OVI)IM5 zOJJhOy9dD|bjgIW*5EnF_H@6_Y7T~kqUHhBqRFCSrZ5kttGsUM3`T&-_LBXzQO*k* ztX(YE@6GvFjgZ^M%5M-I2AzhC2`hn|4E&|n@o8y!=;gOC11SOm0Q@`B~p}osIyzf@i;|lK~~gZ%{QfQetHR@Vz2%A$j@uG}Us06lE1W^I!Gz$R9sD z-d!3SVi#9gsmb+1|J$ES>>+>ntDetI0C1g=gTn3;>F2rSDU0 zf3Ci+88(P|G%4KVy|7uR^_|W07a*P3k)j`EMBLJ~2Vi7P$}M*hX=z=nj;kaGG6Di! zgEV~%plsN8R5?4|+dJoC!yel8<)EF~_)CX?j<&jb0aE-8>S7A<-5VecP#d%{d=-n2Wr~ z1MX1#JuiB?U#)L>tJQ5lMRm?RUiAF_p1(?MfK5D23gKiRHSdukiUjXAq6I5bN6sw+wl3Uww`> zhcidON6N24sn7EF3Q4s*{sRsZZ$+znIE)Qg8(K|mqGHjuV3!nPd3ykoj6jA#peWgD z+an7t^oyz^!rx;5*IVsfa1OY)fyjo^V6ILJApFMH{wl4k>tb#aQ-3 zO6sKJ#UffWxU9LXF^@kna3{p6jCgzGg?ZY8m3k1Z2^KKFc)R^Gof?VU%5An!URFU* zJYdoO`efQsbn#`$kO}DuK#2Ok>1xlTEw%EC-vE_7lXcNML1RA#BH`#VK-m5I)=CIy zb*ic-!IbdVY^vd#DKO8~Ph(RlzLfZeo_lX0W6K&L2MDzQ;Eo4D%<;5FADN3m`2i?{ z{%Y)lPmm6Wpegn!%hQ}r&yh6h{(!evPY6Fm6F8(($Q>F@hG3QR23?5Ush8yHJ%5kW&Z=hH4@dF-Yi5oD=QO)#MCADQ>~4tB;pN9)uow|bCBb=jF(-# zr|Nx#*)kNMzkk4zxjI$v@0n5n0`Lf^8`jB{g~y)LXfa(1IGoZ z(vQtB_+YZc!~&AI)mA|{O;M#818ZW}G%vOo)cTW2-O$)&+VWE<$^dUI6sYu$ zz)}}r4hdBtpx1Gf^HAP>`W2A>m@auUznXf;7Xq5lqI~)51aFILR2`;OG?m&Cg$s(@ zPu7N!!00PVlf4SZ&!iZb2W%Pi4xltqXEvbpJI1*UiuX*)`?qE5#dv7U@YcjFFX`QH zz(l)n1agsbY0C_Xn78Kx(yRR({Q4(2n5znWt$54g4^i_Pn)ple576-_6~Ah+;;wIQ zXia=buNVN%^}7eIv~>AKruFL)y2S!#C z7XCTI;a5bOZFqr!0P>WD-8UGPNAVd|mETNYDu(TNh*gX;j=!lB z2qE_UhlmD6jp{BjN!X+_C|NuR<)Xw+rP}DDbd3<#&0r@wV30$TLZzH~<=6ahMkak-FCVgvK6r5UD|00%= zFx0bSfX-Wf`Bt*+znm;?y&l;AMkA!-9xbu(3wW>s;EMIj)vm^X^V-1w7AOCe&%-m& zL~OwqTn>qe?Lm1B=HHV@4D@f5&rzR%Abz$BEiVBnLauf~8c@345}bH_y( zT{O+s{xQK|S@s_El1rgNp4zp~28>{n_#a-{g=S+?m(ldIF zY#<7;Q;nmN^CLnJpm1=Pz|lT1egn&)<_T#sy6(zNKab-CYrW#6*~*$xnJCPa1#SRO zFxbxaKE`F^A>1VG2;hqzI!&?C#&ep`ZQg!+?3K0Jb1-KBsv?IAOe|AJH3a0JS=Dt?lU_^Q04 z)QpW(EH)7&`O;u)kHpf4*i$2&CNN4*yL(*+LxLZZ?ofeD=@ZTnMk4-J^FWFKW+*TY zTzH@670i@Cr5UI$fX73lr0MArxyKi_jY}!$Qt+W&Nd3CFxOllLW*e_dAWpe1Q))sI z6aWv~*!GZnA7JzMj!$&##VxL~vZ){j^9{Fi?` zm%#6~q}J3NqK7N8S)tz%Ja78k1dA<6(5W91q~9u|W59K6#h(`_Ci@KTB}X+`y!bLV zF}l8qQpn>&Zwn*NmZ{tWIn5jRO@-+dN+>RxIKmJa6wi0=huKOW2I$suYzHbF zEiC#!zwQ6@h_hU?@0p#Qon;t*Nj{jmdfFI7F1r2rum>M3df47Wh0*i#NlQzYtM)x} zdH#R{6uPE+D}(oM(22flZ@+8t`$Gt5U)qzztd=hr{EmIy{Eo4pZTR&O%b~0ryASD) zvsBwij+vn9*E1Tx{<>OPTA)4y4Yk){4n3iQ#U(%J1+hNzT&1JC#TUiiABM$tgWDuF zbMo#JjPBMe$?*Ayn52L>P9;r2k9^1@3+l7C;m-mTBq{(D31#)yDbmjIKa~%KyTPvn z1<5^7HvzUYT956OFb4Q`853SmC@!+b1S?!B*&EP3slcj5u^13Ral|_Z?lkVGJqose zcviFGuVEl~6IiN=LnYBjQbQOn1ndA;wcw55Za5Wy2X2$X`>~f4;hrigRBh9rAAeyF zqN3B_i$ZMq=5XzL(Z3~b%|z*7g{elvKEqY+L6{N`44>JB3N=*#N(sP3%D9{=5;dzy z$qoF94W62n^+N~91EB2fpv3faN_{2XRusw)_tjvom8|>zcYr4JhT+ePYCI75ZZ$d( zRu7|_eff~6iVzGObd~bLh+8d+>+Xo2QM95qCaNLN5uyS`l)#T6i4ec^lE56I#S7}U zAA!taJD#8C+Q^{GQr7Dx(^8NtsXo-EJr)QJ4cK^(s?B)aB=-qqQg#pf`LnN>yV$HZ zM+|Cq21Mx2ZmuZuOdy(eQ4W$@B)!_=;!>ZZVq(BtmPCF4-=cKHcO-BOQlcQ0y5W$+ zO~dHU#Rh6Du_}3(6XNPey-i~jvz{-S5iMjQ7~Z%o$eqKG12@r)byEhl8~H)4MvBx# zo2(OQbRqvz)5kB+IL;!%^v zg2k8HLz95q96D3s(~bOCxF`m{8*Pe@$M$NIF*~ZnPgns@G_vpw`qrflvI*uYWtRjH z0MLj2u)Da5IOs!=+q~W z496cnEC-Nk|7VPd2z{?78t31C_l5Sh^73;=_HD!BS8wGnuZW0YbBhQjHle5Ia6kNFYo$#(LFCq zNx#kUV`{qBRQas;I@5}a!K}M~g^TF~vx32xz;^aE8=3Pz5?fcI)<`B8uEn0*y27{a zGS_(2?5xJ8{i(@L6$xJk`D5hJ6a$_HeK`kSrv1rEVV!NITwL$|1iX)q>3IFpHmm&l z^*y{#WhX?oy6dK2w7vlWGquy zh}DCTh*@U15^Qj9b+{Lr=Uy2t&AZ?#Vue>64v7P%yl^ zjgdvI2^RpvrJX6#5PJv-XwS=jc+x(r1tOmzaAp^Fk^vvO>@h;9OecHeY>N#$ERZ1a;fjK2@-;C3N34y>fXEqeGqf^dLT$aZtUV$EazYER~j>++&g)sTv1~!kQQfB4L|L= zqGZ;@bhTRL3@FymFZS2*#*)5BQCPiH}tiB4G zz3gtp*PH7Zl{OH*FO;gaA23j>(7szRaw9gn8$$f)!*SkZWV{U7^Uim=HSQbE8z*vI zGJhuQ&7E|CZeXx-aH4nlx(((O3|X|8^RC+jOMfqT#yvZs6{x@`HYgKoz__zSoYGyp z*AIyL^Ks90`?Lk7{9m3|GzdI1Nsq;)>w^(#HSkBKc%w$SFUVj&Rp6A;tN9#;f`-wl zl7YM7ip?^|k4tb*yEN3J4444`H4Z#e-*ga?Dx8n(-Z4CWShTi4TSId*$MY9Tu%JWf znX>q4^;J{*VT$v{7SN%IaWHl*2Fl$doNtMOsAI1iWqI-%mt>7O7elt1PGhU>)9#VA zTIN2SHCmuogyGlo^_{X4`62%PW0Z;Xx|3f$t!?1OoO>ZnR3gI{{-mi4kQ$ZuJ(nL@ z)sLIx23|=Z5G8e@ujKozGiUdzRkP8$0Cg!5K}PCJk6*j@&ARx9BI!4J68-H0i5CwF ze4fW>$Tj+)4oIjW!AylkZh6*6PKi3?n~Eoxf*0EJm+QIdt*vdtbiZHtAHl9En#0c) z)pVJ(^MeVT*KP}Z7uAV*^y?KjseWpy5$_oUVFV#=2=63&{QUhmyN#F&GRoBze43FL zoLVf%>GRX1=7(Oz-7XnJoVyO?8d+3h@dRftySu*gR(GfI&yn=?FJW+`9`m`AITV}byy4T2|oT?7M@YSKn`0bI}9^RX_a6d zl-zJT7Jbu^iaEHnezy55BT-~Sj>qZpx_tsn8{%P0;+4Z4BMs+vxZcMI>9gdH!=VJYeaEj$oT%Y6P!BRTG-Kn&) z1^jf1uo5tNl}mV|3IccFx8oP+92&=>uPx2+mE}1!X4~fBVpoK{x=l^t$+5ll3youF z!{3p`il0XM$9g>;U6`}sCz8VYy5dc}huLsEEd##Kvu-5aDSH)JAH+zw2rF-U^dM1) zY%$U9iby!~DNKsvTSnxqSUH{|#FP4kT{gzx6590aw0@Ze#{2XmkKMJHig=rkuBIha zpg&QbCFphVx6>_!AykgE;NlB?5)E>D;_WfEhT;+Mg*@ZPU5w1>O?5_!KgfWw43KZLoVkI_4J zHI#DEa=A@6ehbLOl?zqJsx4(yEwak;%VqD%4~-e6{d~W=bV)wctD7`oYvho4*SgHS zOf&zRL+Xvfklku|hG_WfM2g?F{i&q*OiT{`a!Hi&a)tLlMp+`|a9(}Yi}c#viOKS4 zrH{Yy`~E)leg3Ha?hB(6g{9!F1eV{yHToY!_6Fh#AAKP&d6U~#I&iV6J1}kYbXwYf zja>4UAW^PAPV76Cp9DJ3QO>)!>lzwP=T z(AUomD82UpW_};COaZ!RExq;YS(b)BENxp18K8A+@(k*1p;o*&>qtRy;gE zdzHM>$d;vbM2G2~*#V|k+_lvLV72Q&zS`w9@@$v`=eENgg)7-4!>{Yp47f7g`W<}u z3m18#pSUg4h5L_R$Ma=)mSr77?rwCqd=VS=RNj5n6h5k9-nYOjY1>hTa9P2FA1d@x zmgoz{2;E9;7MA)c)}zRI-YtK$d{O+f;f75`gg9U$%?~mtJ#&`sM7aM}x^zW%NbW#- zcomS6>p)3So5fi#F=SjXhwHr<=w&(9o#o&+?)-akmgS&qfMKt69$#&-Yk}6U{R8;| z{S9Gk*xVnv52>~h)l#4s&`mEt8G8i#X}p?~j;^rKqfuy9?UqkdY07uS^`>SoV_dC} zyCYz!-lJBWCN?l2mYtX>yL9XRtL%EIEnCU#$mK83I@aO9J=p5phWVZ9c!Z4=d+^=9>yzK?Xn(EyKSP5`}G11Bs<=x2Hs4j+@A(AF$=ehsK@fAy&aR^1y1TyqZ20g#)IW{ zGdi>c=;lxcr|*C-U}gvLBC}2s%$b5r45hUoXSZwiM{HX4!@o}Lr-4QdAh6Y z^Qi#qN7A7)tmKIHa?3b%kO9f=iF^Lqs<2^QRUGs4uA|6>%pda8H^a%BGy8qEepEFj znopQgG;j*1N;;0em+)vBz|j}H5Mr2N+A{9YSu#F?YYYK>U`n#VR|kz-8%JV}GX>=o z-tP=ISWyo%*ttUtpAk@VxE3VORI_aUb*Quy4ZLXYI?+8w?VA!hO7JA++Iygcj z|6U;Q`KR*lkD9jBl+$E~^rm!saO57@61d=rwH>pkOlP`=HG(Vm+rjpu>Gj$-owH^wU#@o){3Uo~;}`tHmLlD^$-*_};a;df zEP1TM0WmIhoG(|R8&4c@DS?8k?IK@*1iBHQX}GyZq(IGwaqMmdnG45Ljv2XoL9c9P z-`mmO{q526lJ)@?qNyo-=S7ctg@LZ+_6LLa!}E9hO0JR}I@A+eUSqf7NYpUOAtrqH zggsOKnp>DL{ia8kPVz`IM=$;OUCX<6Kh+d^X2BMSET-RVY6*jHD;AlVuP)Tbd@$8; zEb!4K4adDi5cMvf2$v_gcWy#<`?~@I7OeTwe<_gtm1vh$WR2rR>&oxww;1IQSg(+w zxu#eKCSg$`47oy$dk3#HdV4TI|J+=9@p6hOCqDBpLj}rU{bvmI*mW;$MbzCl7;42!YuOv8)7T0-Uv9`|QW>Vmf@)K z2Ab)Tql7(C!&B_P!;nJXlWkKy@SV$sOiV?R>n{rgcTXXDh0-iN309W|n<;lLxyDx7 zs-_E$ez>+k=J4^pd9T2;sQdP0OIM=s6ALI+Znr59pIP5PZDf+}WOT*E3K`b+!G{UN z5^C~iYkF%`%=tJO2M;4}itR5>?)p~&^ZrFK284LRhMwiyi#8{+`JAc+iB)1vh>iZ*ojIT9bOk`9?^kN=J6~3=gr4S2? zzH>yFyWDj{tX1Np%x;fJGp-c!Mezcr`KJ9d(o#!;@%>erw8JK`pl%E8NnnycgfzWv z`NM;8G@_jT@-O|-fseAoW+nCSJ7wbX8iEpOmhJBlxIar(p9D;pU&+f5{wcg<$|oLJ z|I-IEZ`R zw%ZAT><3a$*_+Pk&WFNonAk+H{@Xb6aq?$;;_Y{b5(_0B=11j+ilV{vK6TI2T9*h$ z#Sg_7nnlx!<^1I{#ZN66(n9oaQkpX@v5o7I9Io%aU&{V$6URM^`(^I4*HewI8<`HH_6HN&TjFXr^gX0bvuYF% z7N_mX*Njf^PW^U)ev>cTBydqV{;A6U+#~B?TaeJE$KCBGM?6bK_AZt9 zb@;ycaano&jW^QOjrIqVHp&-(s(NAd?4z~G6}*#R|Hcr>#NZ7j`K_l57>Nf*g!TIY z_)vVvhecupiN%(tPDhWXP~lR)e$1)b5Vmw(3dqTS@?0tMrcFa*oMMY?Y5@!x-SmuQ z8WEM${j&uTlqgbnMUU_JumgK9#L+hXr=+Cq-E?|S z?koFw=O$w?b+XF*rJO)_QhWOCC8lw$8=cpfX~?i+%L;>@;sR^mC;~Dr@zbCw-XM~n zD$EnTl(+iSg=D$+^w*$gO*Y)Yl+i6uhq{rt1S3$W$+7E@hfE{*A)5?Rxp*- z=ve2)^wY&exlS?G;$NdKiVQzBhk2|!Z&JUho?2v*$EPCY5R;?ZdLe_QDK470X#RTj zz=XFsWBmIUgu=FQ@1D@MNYvDKl2n~r(TM0kxvn76OTu~GElYS;VloRtMcQo5lNTdp z^t}Y~Z5LRc-SZ~c8}gX)nqL_WEo3O5>&+nFyYJ-lp2wzDudBhw>s2Icj-KH_^Bsbj zCO=->d+Ac6dG|?42E8rxZ1>D0In!fYE>P#)s$$Y zfwm{D;AHIK-uJleraepK`lhl#Nyj5k#*57`y~s(kNZ!Hsd|MemZ&0SGNc0AG?f=Kq zdxo?5zW?L)o>j4D)!rlasw%CbEv-$gYKg6OjG(G^ZLw>#Z+q7YsfxXeSVa+0YQ~E7 zOF!S^|9|8-h{FT-eO>4InyYF5xokgUAknUKw;y{$)2}m6@f&Lfv?lmg&(q3_DI5g? zf0n;GEf`1G*519M^|bP4k5$(7ei?fA`02$NBm*B`**T>aQRs>|!3T-E2HShY4g2{N z88Z1raUZ?(WK>(CVjk@6dk>$M;fUp@|IFDFTu)W!DDY6!d~h^5QgC=78*?9Ujk zrw7hEoCSo@hdA}pK}tFNB7dDH(&co+h9&#7#IiIT z=xWB#3pE)uLwup3^V*ILC;W~ID{|Td`Q_k2R$fpasi+& z&yqsU@#U#eN7Id#*$MI0gTS>>`p6(`DlLDwuUx30y0m7M_s_t;s@+n0QX{0Kud|Y- z-GeQcn>2N11(^eG)CqRK*k#2BJPwO$Z9Iy;6nky{VFn5 zbyIp94?-Y+Eb!+NJfzkX1PU4{e9e^XpZZ-HD&Fb!%Dq%RKqtx@I8MOcw=MU~=f3wVPJl&LHnm={7ppl>)G>*aDZ$4Tmo0jPr#MUZRk-Tp(Nf4_4g{TP7y> zpHoS`8Fiklz^A^K(S#g*aalqoZ!+bOuO_5qPoLKLym2v#^LLp(QFbI(_eu><4HS!! z=@;X{nrJKIq{x4mYu{jvHP1Brb3AZ`h9zQ<4u{U7{wWjkGlO`OWU%QT)e9}E!T7&L zy?z!c1U6K!b$IU5*pCR2J4V+>rCCOPH$l_JN2KfUt8UJ+_O)ZMq?l{F*Y(d2)(!gv#*Mg?<~J`j zD0A59abJ`xw=mT~G0HFMeCjCC7`z)vzmW;6!}FOQ>2{rDjk;#vIxz|3FUhm(xO>1N zuChRDn1p8VdmaIDznfA)@NmV9SB~qeS;+~+xCPnSyELct=3bPGnjMKNWM9DP7+)(5 zcsy#r(ESgRYKc8uM2)vHq1@pvd!7Y`su%8#Lp5&R23vH193uSRXoV`kt`nII} z(dB{8qqu)}T+qIwYTrNSKtI4|%{fi$aH{+oMLrJJcex}q zF(B+>B6zKnqjrno5Rwyplxp_BzT&M2gR@8Y{n*Y(~(QrTN*lpOXU*@!qPy={`O6 zKhNFeKHo%GwU)1Wk>T75+D#<|Q1Bjnotao+F*XV!8-`1yYl!cQtp$uI_EFVAut^1$ zZuW(4w67J8siDmepd?YND!kv7d}+EKe;7=Pp`qs2F;8i&Tif8F76Hsl42b&>?VPa} z0chD3zIhSZ98Z~rC4G%y7EIWhO+349in8N+ zAVE?>QQr}ODJ!BCvYCa-sOLGiMkJ_z-ldT0u>EzE*(z_PJ!ho6nf|DSe|tV16ogqk zg-1z`zc(g@MMv8YZ*yHwCC9GHvwg-?tP%osLG~*av*W2LL_z6HYqlF_3{dlgI7#3~ z$$P0v`UnT?yYGSZcZh0}r@o5#QhiyqmzDpOFgSk3eBhXhE=@CX^)HdUwKBC2k+l`4 z*9mf`mMzBiR@DE-!zdod5BzWOS#u$0@UyPkF3X>u;Gyh92SwMgo`O0tWUfW<2)Bik zTy2V=6~cpbTDpn(3MV7DR;{S13CowrEcCi{$yBpoOUg!L9zQFpexBXoCm78rbr@= zSx0AFA3iKR8eqJC$jMdbq8#l(p|(P#7-Uj-7SNA zLCsOa)x%-E#PDudMF3g?JF#V2c-L;ITVX|Et7nBu<8H!c+OPHJBUT6ZfLD_t&UN;^ zS88}Xjb%H%NHdg;;|-G%q=_%B)=H>~GKEnX!{(oAOZgkdUg|J5Oa_aEQBS)gJ^=sJ z^`j@1lsJOi2bg+B>9;C_Bw@;mQs=%`-oaXXyM+Y#JHozx+yi{r0E`P^n>&%p%f9)?H0p8iGXS|up zB6|^GQwHQM=ZRrMS;9X?I!7_VD8#t!JQ))kLfGF7sExrxe*BnE^P_UQ-UK_*|8)_g`ZT zL5e)3AAkyosQr_q`vysXwzPG;(_>;0S{av(#H9IBQ4g{&raYZ86Pv2)A|` zM>pD|Ss?n2NM-+FbQIpOSwBxX$FXvb1lBT`CMdax{d_ygDV$9Y0b4Z{D@z1X9o_4n zgP%6LdmOgUb=8C{F8-Q@H}1?{d5cT;e{->2KcixJ03Q*Ry0$l1(vd<{isN5ax8+*Y z{dg#=_`YXp%2Q)Mc4ka=&iDd3uPzF2gQn{hyDvun`IY8uH_=Wsj?a(@7ZU{r z^*hLeBl3iGT$$TfYSae5ghoVL$<1&8=8}zco%}c4|7}RQCJ#ue^rg^~Y%|`ye)Q0k zbe2s20T}!UfAcLw{zWocZn#8ATv3HGX7`107^6IaD-Knw%ZZ{c>P4%n}z~uK2)KoJ< zn&S|Z(xKMvZb`od{4R3hY1Y#DagX;a#S!ROLuo8y0KasK&E50|`xt_*p5{qC|LG;v z6|<&Qy8`Ln=l(#Jx6|G`RM@Svzx1IGfaK(2TBg!>FQ%Dr z_(Vl={Ht}eb$+D8*YZj$lUKXt72AgH7AZN^9aDgc>vY}LITZg#FW5Tz0kcRtoUj-D zzCKGllQ|_u@!I!6aeW)n%R|m)4_E8guu&G2A?|T9>1{2#-Ky^J**f3#=h$n-`k!PA zGZVdh2{Zo~BiOzJxcoDesgG*HaHY?avc>Gmp~-*&tgU1-Oq3NvPfYb-gv4?7K_X=w z;o!shqA1?uT~|611?>P_nl=*D&nQpHQuO>8AgyU(CaOtJUu84qo_*1aN2(&7kJj+k z0oQm0MH@&%X=me48qI*U)!AaV)+wh*nrYh04T4z6?C%9}{lZ^46WLdvin%HW2wCQb zvoui6USrBzO>=DK`eMRS4MU6H^EyexCrmcW#dHSD+LqS?oKLTQ`grQrF-vYZGro7i zk9KdnVdDG{r7@Gi5|N2y(;T%V_$*5*01Ny-)>@>H*q_NCH$cviV%H8h_a7Voja%O_M+;@wj=E=ac ziq+Qw+#Lv{*@Vj=9tIo8t(FFO2dAcfU^(Ncz@$xYfi;xyw7^2;BT-%f^KjFmRJd`H z(+4GQs`tvI7N0b`7}P#&;OWaG+$n>C-d=Kuh6{r83))~e9jnFz4ozW!c(qgIP%@Ot z;V(1k{!(hcp}g;XEX8&MjxA*X&%Z6r7k!+cHR(u)$?XoO3vc|aKFG(qz!CR)Js%ok7ua!mkTXzl~BrKTmOO12RS6=<4~E{Q7;}Yd9$R^e_$@h0cC;^ zk;BOG0a1;GWA{E7Ukl#z&7|UKh=&MhXES!cDZ7#eeTRr-Q{u0C{Wlh{^#mazq>Mh) zs2$`XO~TWQCela>dQU4EIQcbrKnR(g#VivE`V+ z!%t`eH2rP>5>p*z5&UhT$OP%^tH(h&qMCSV5DP`j?UN4o!<1t7kl; z;9WANcxJ#WTFJ&|XRIVowbWE+2=j;{IRFsxFFB>uzaK=H`*Dy_ir(vgHZhLpFM9Uy zB8dG0+|LUC@vzgz4t$wKd&tkv+%x_c!drN!!=lk$Y_1md6GW(&&y$*kyiS`&z(|ie zr(gHII|JULZE+O35q#Xc6N3H!h}7ljg43owe7{8S8Qypg=PCwIj{UBlYg4}Nr}Vj2 zOP_m?ZyfwY6C(@Bf9cV_&%Np2Eh|;FWJSC1GJLLZAVK5JiiNgi(dhE$>v@*(8szV9 z@=EK1<@y4$jl5DtF0*?s&y1Y@`f#^1CvH(a;wy%_b-7wUmvba-5;Igsi6KorHbO?TGo`N*L4f)%%=&q9?ld!&op zfCBMz)x?;kdGceZMZ7VRFjcc8!B%*G~?Zm2yRCBNFU9S>|7su(|W ze}7r32s9Ict-*7}E@|*^W{Wz?3PDKZfC_*mAI3NKoH5Dw#7gh6uGknf&xewQ60iK= zWK6anvmYChk|}WFL|GP*13=rw7tjmm?e7d|PV6(@;tOKP(UK{h2}6SN-L8uL(BgmE zfJDa=G8WGTFa8$Aw$&dR=Lq^1TB+-_xYwh}(9x)d_+5^!J2X9#W1l)g4zcJpD8+Wk zRynS_z~1CgDpUa0L&|SX{fF4tt+#^KY^wFq597v3K~!*2K;!PZkM>!W%xWY?^_qSC z{rpo6IIs3K$sDUkpxb9Go)ua5DNqDX+g)X}iNHQoQnYZtT9gDFl-gGB{S}_oKQuFX zdGZ9j-l{g=^M^R+9@*l(I|8iWAl8RUUU{1DIYsObr?-_?e^!@C@OL&hq2wfk3mUAvzA*aksK8ub-96pAs! zGbBY$8bDOmuk(DPw(>hK0mzSD&8R=`ST^*_-=%j9#u{%#^3pTitxaflGq63%SgGIZ5DT~3RHYApwxoEmdgPpV#z zs(XsjiMy^1{H!Bn8^DonmhLHzJ^AA4`??qhR4;A(_mta6&cCwM=${gU>?eoRGjs4D zgFBZ0F}MCWo~iLcA0tZNqq{k0e02%x_2&JOKT*@IOuMl`G4=7)Uf+qV5Zu-86RBU3 zK~Pr=-tnCWAL))iekwGUJy{TlTnEF`Km$`t4~lp%32<%Q1~ty)NMsi~$99Gb&i50@ z5t{eMiK;=dpW)4;0P#SJ2rg(IW1SJ(>gQ}@8uys^>ebH^#MD)Q7O>aq{*PDUv;mK4oP*w=`C)Hv+V!w)UB- z5=mgaN2~?g{dxvd=z|7ya`RG60LS^fFI2nx_2WUlh@Jc5O}_6%pRwi$MEE@=SHcYl zv_Qk^X^EmyRD?eUCh#im#(eor(-(E|FS&`xQ=IVs5xY7sA?XyIe>GMuJ$LkeYhJC< zd)AWXE8Kov-rbEJUICL>tvw`f`9A1p^erRvFn$`WHbfSkNa*tpXEiwN1{$1F>&+wj z>ITQBv4r^jLzq+%CYf-EyTfgO||8FXtdpf?sHxMoS_YzauI)_3$UoL*H>z2H1hnPLP4{? z)%TCstyYI8iFoUH|111sRV6|4nsV%Q_#qYgM57FFR>GOaZ7 zJYNkAw=ZBAvv(`OGK}aL(7TLMmS9DCy+SHqTb)&JF+hQqB@w)IR5T z!tq)>#lpeYN)^?xfHm6-0|fdJY1LFSTqB9pxWlE*bBhr)P}tz%IQo7(+AL^!$9Ask zAVz8zkB4l%8xI=uZO`i3bzc5A^#`(X5M1o$Y<&w}YG#jNC$vEpvG zhP^v8l074DxjFUYZ;6Px?j@zCjho6kR~CKv0=HJyxPM%KD%y3ZYSD(&kLKHmGUes* z@k{06FWJB-#^a!!DPQ1hnH)GajcguL#sps$0>=xQDzjQFtfVrGCOB6ai5vanTq`p zvkP8yExO@~E-1%rA`K*iTdz`P)10jQHE$K!{q8}NmZ7Y0hZj#3oPDMPQgOePAXeYwRv->mL$rYZA@x ztY~@dlx_p=lh)1=5$GsoyJOA7oX$~gTh9xum*|FBh28wKt#CC~_)=a*@=iKyILzws z^N1b(BIY$pSw-v4H(22GR{-0Nqs3VcEYy4bbE`J$Qid(b-7l?aY)(|W`c8yh7^=u{ z{@-fH8z&0I0>|WL?4Tfth(=jKeC1u4d!yLi5U$N6f|O!vlGj|Msw>zx_T=vA^UxT@ z!)z&bSn+PLF>u=DDiaj)u!gQ8F8-Q0_o^j9>AlpX%U0($MPiK3Je5(M{|wx*v^3u8{fB2)X{8tk9Cq zb)ZZQ3Hk0J+*>tW2E{)tvu{oCvedA%UuW=E(gVEE%GSM~BlZ<&&~?DrD@BXHUEm8U zP#zdS1r)@(3%T&lCfqvdXF@tkx>+c#*MDW_;g`d+YBYXTQJskN3|;4NUz-j^4n_cl zCSzwJ+M<=Ith9x~j?xO{xrQCSho%_tpdsM)_}VuG{y!M?{OChay_UB&lnFaX63Ft( zOl+)k#9C<4)&20O+p^xQY}x-?oKRPwo%|$xDOW{H8p&baKcswJ1}A+c4ag2?PmmI$ zjG%||jW}APMbI2XBo~9^+<~A)HRmx`K68f3);2Z#nCAie3NDp3tD#-cu49+3@c$5H zB8ra%+4cXgJ4pTVd}&LgHDB=jH%s@oe2=&J`dSIipXtL6qW<2M(5AuvhOnI^1yI)O zF|ap;n8%hR{B0h-GCOL~oQtHA^jITo_}>NsB|h|+HNk!14TBquiXB>L;}s~Wym(eg z{PoY0XkDr_IzqojfS!0DN*%*Z9HHy;W9BF2yIT#X95WB^y*!#VKU(5=CcaJ})!P=! zNLJm3ua2Zri4rK1+vK9E7|CEz1=WK$uVATPxgdfSd7%}@O4?;aUay#rUI=bh_O)8R z$tdj1fN_gZtC!WjGv>i27j)=#rO$trUMyLHk zMv^N(+!EhVC4I6`N&3!FEOaU!oZDGvyg@iLk=o~;=5>x*qE*e!Rja1nYcmvKVWVFwUOw_&>HDid4Rf}ZvTEnE zQIQl7I7&Gt?e^8M(Zv0B9eNC_>RRrmBWZXtd_;mfSYe&n%zAR8l&kAE3Ey6ZBYIyf zvJ(}NYw&0NUHcfr^#uW2F!HKo)zi`Fq82;cg-49rd%ArJU0ylXaN@7g5%P|GSq+x6 zUO0^9=&~RArhD!Y<9M#v)11$%5?=YCOxn}G5rp=uzDU{f8aRRoo>R)?1;QMg-x0{r~5f)%VT=e5hGyh*grw*X^PrSI-{}Nqf>6yyg;t z@%;*O*3-Zb=4Ns=#|8;AR{#}^yEwe@6luKHOCgGlGwzL?oaRMg^ix0#*0GiYeln}J zG4@9?@nF0n;u(G~&rPMGhz>F;x_r+B-hd8{;wyBP^1YF4?p`uqy+xx*c{*4!@J_h5 z85Pg3{S+IlZB9jch*fJjumIFt!X0?>2SFG_9;%d z%O`h397ckTxw$CiF{0i{7Ma3F{F^yRss}4nhB_0IewuK5*d~8@LS3$Hl{tFaX6@iU z+X<~2y^jxEtJ;fZGJ1217i9igCM4&auOc2u_* zi4*12pP(r<=3~}4d(?NM+eq_e_b&2CFD{pzm!$WCB+XSY2yv_DD~5g9M(4IXuP+ms z0XYLhlJkp8Xy9jGMA;u!^&%^D4LWzqIX2TWxf&XF$-W}NsUorQomvW9JL3(Nx0{6W z!8B^QDuzCKuuUia)qh$s-ay$1X1KUWY1e-!b6NBy*`k!I+?@#) z6}0cHTDjkhI$n8}hf&L3oUrm`Ip)?FL*<#Zjre*=AFbbFc<{63ie%;}M`OJ7u;grK zAmd0RKBDia1?WTvmeB8sUp1|;{#^hG$3x*Q*ROAGFS~Cqm$UG-vjAbo8%`22((tZX z3O7k<018Tfs@hetRlwF*tX*eWM}{9QJXhsno48}Q{M%tB5sjH_Re~tZoy))%9JzR? zGjGq`GsH@D`RuWA@QY)s|G_f~wLymXTDYxdt?-@wP-czcWBgFNhu7e`kH_~?KogUy zHb@cgRtczv8c_}p#2NaDD&qDT06!!Oq;K*xlVF4TsdqSa#y$_PNOsdcrjFDgORlVG z|BqHs5M|q84_){B@HGuKRC~n!J%CJrTAoaanClZU+xH7vWN{*=S#|z^HA7L{HiO@L zjc@%cesA$Dns0E=h4a0b->aC&at_tsYC2?y8-9@Jp}gAOa*Uh{@cUGikyFzaPVrFU zg?4V9Ds?gd%ftYrb00H$BXk{C&3f;(ctS3NC9B?psr!X7aHI5utyI{CQ!Ip3_+CPu zfUpf69_g|@spSi;sQMEXVFtjJ^6=V5y~VV;+qK-x3iS^Fy!OSrgv|p3?&OjBX@~eB zvgM=IP?WgGnY)Z#e;277Y1CVe$F2cjBqyu?c($? z{^4b-J(g!yVFzSBQ%FY!qaGyDYgEWsb%YWAf zQb_5alxB89e>!dysL^D&<{LauPvfT~(!k4Jg zpQVKUTHR(DrKEPp3bWCl5aIkw=#O7ExLhU*T`?S;;V+E7Y#!OyVn{6(YozFSZpTDi z85zQZ8N2DMflg=O7!v0`GDuww?YqR!L)2SgIu;n<+4i?5e?f|wBBj3D16 z&?U+8mmYE4gHy*~{TsB<-eIe9)CJPimZ;VlzV@l&VlNvWwPMik#?1`&NY`jc+_3TY z&fKRCWc5#ll*F6tIv~{MIeF;&^%?+9_W}<^B{_}q!EN~BnASGgmC>u$L0swhu2>BSK=K2}<+&TCQg{NbPL#Q}`f3ZP>aw7gW5K%mTa(-%-UserzExHk zl0ee;B+SSY&yaTOvaD{0t40sI@k!IL75--{B2lwTiJM@dUnU9Uw`6umPj;9p3F6TM z@i+Jhn#n5O!FJ#6yI9~j-76!OsIhrD5%1h6GxhnIKOq8hMyh}PdUq}v%(86!E#mqA zl{s@1&jc)_unJC-(VYDcx{SYfh9}C}@j=hs9rzr$Fr8-${Q3q9WCGE-bJRZ885#HD zt)IWvTTyvAKzVn_g~*erY=m0j8w$9xx~NA3u*wQ6hj7+9E}T= zj`j{0FSZI-jEH2OA3QBwdMKQY$vv>FI?+^@PxYEqK!gD(w-fVcKD?@pO3ON8PGJP> zV(#as+jSLG2?}AMza}VE49&#Rd|V84-UrCYgUJ?uAO6>v=*J}>aoKf!YB)B3WgxKz zrbn^iba}*YaywtuX=hk?P2!Vc zW`}w+Del@QW0*P^6Li97TbrCUzjgZG?PzMf9h_IcnM)o9uyUXx4J6x26ZtkkF5Wyu zDSt&@!%*SZbAh_y15a;oZMTDqlNk8f&>h@d_7PvRlkozqt1?E6Yu zM-Hkes<-vbl)~KPZLV|dVt;-{eo{BRJ+^`D-{O{6z^#3acn@MxA|H}}g{9PUL>f5B z;ZK@!pFVgZYJn08?lbN1m`viHAm5;dyeRmO+G_1QREQTAa$aa#A7{j4r zh;Z_lq-VHiw*s3R=)XQXQ|o%!lG{J=^7wgBj#~pWG&RaQy?P>PA?>ON9r;XUJs?>7 z{e5HUa@c<3jNkXrNi|3A4VSS&WJrNAD9IsjGI@)!`sk!8$f(Omr!9TjbLdq|H2rVx z7Lq}9h6kD%Sp)(mAi!P9;iGFO!#Z@zrHgon=HgO=?z}JCC0=NAlbHIi=O|(M6Vd!>E5B)mxrwlu1zE@qY z_VRdY2zB~pm1({4;zK>muz++u>7w8wCMYVAw7}{v&3ZuS%rf35wKs4tc^A)HR5Y6? z6&yqhEqSSLiWtAG@Y%&9QKOm)JZt;xjwDEYLgkt}w`+oBweA*Zy@#^|J7N8a+tMrl zzT2_hjgY#$;@4?{3Nq(Iw{t;_ZJEM1Pb)^=E<66>hd_!nS15r(DBVseyvXm|3{J
      FppZ~zPl~1-xFBIT7BH`sG`1kgbw{n zfjq`TRSM{u*46DpsUqlyN^q&EEIp;J9ED@Z_Rw^NQKRV;_;tZ5dp%W9Ps}~5h3?oo zCMjS2k&o)X6z&Bhu1s2djidrwb(1-byi+57owLeXe=CSXJ$(`SW{qgZgi4rpqv{7} zjUrYNvxB_n zQ(h91%N1X~btjQ;rDMN?gxQ-ghrb0_V#_2fGdUt#%CB2TKJ{tXa9_gA=(IDsA720kuy0n>V?!@XlX|+2z5m2dbQhgKCwVE9CaCLksqM zfn{C0e9H?M)tF$DT#%)j$S=hxpuhYeMCQL;G0^T?c#un%@tr9zfHS6urIy6%K1#Yi z<@-b6JetYKc2GvxENg=nxFhZry$!tdoYB%OC6)o=BPVNonUIQDjGkR*sZ~~DC+r;r zNqjP?rGVa!0?y6>&Q@GchPn&bz=y!Eb0FHk)?x8yz5nShWK+8Kl=ChqoJoXK&B>|9gj`UVj26$A9beWp7Z2YSz!es^&q{yx2ok`1Ft|&PF$1 zcABlq?w^XIcPDw3>*Q1HWd_$;xlM~LFq=~O%8}!5DX~@^3UDC*Zz-P?-x8AAe zNKYdsU-LKuLRqUd(Xak56L~93-oHYE2Xo(tl29PTJyyr>JC<8CIHKdQQM^y-`_o1G z0uSd%09NzP13#wTSx)vj9LO73)tA41h25Z819m#%k`SIKS;uP?C`7cmU3(R51c++T`LFL0K-x1ZP>7uu1CgEr*=K(VZf3YS zgM>Ex(u*!$jt2s40?5t+Epy@YYxN@<(Zx}+`T5QKvDO~HLi+o15626}zhppa-=n=i z5I67+(shz|m_H!TWR4o%dp^uag6T`{t8C7>|I*$`-oS zXXl$nuRn+-H2@e?e3(9B?ueTu)dUE!6m@wVXB+ zO&exQeM+AgLb;F;AjWLP&9Ut;-3Wr##Wkc zzMZMj9bDSER=vZ|}wUNu8p&dD}<1bnI;X9)e$Z_3^ z3D$b5gUFC*RT91@b^{C-yd53k&uh77=#OJR)Vu9R$KMGr-LX_-mh!EC^1*u? z)kj42ppa&={!RIOMAR|hgsN1qyP<%qSj%>NRcjfO^jx%0zaQ5S)N}>OZGIIXi zMuZJwqBG65A3NK@xtL{E10i4AGtPq5I-n^>@ zZ1dx`Q|Pc;6y2OMNj#fpU|x2K)g&jQ=)L9lKnZwW`9-;581;hrX;#>r+@~r86my)r z=KpUH#Jiik>~c~v*lbPZ?b>-sj#$&a-mH=Rj{9oFv&c7JEpku){8D{n*=3f8(35d`;hffF$exq@FWY?+^H)tKH)&_o;>)Z)r;Y;?_yUwCw zK4)x~&>!EP{MO9&*(tFUhdRf01OwzHLAGny-Jld{HX}|~joVixHDxu`DjiJFz;L>D zoF~oe664cv1^tA`$<)$lD$?P9NTyrTql?B;2dN!f!JJ?mUHdw_X5 zm5u~y9MnwN7vq;T-O}T{Y^gm#GT%G&-kCr8D-uR>TMkRs%@Yq?FJr`&%+3nayJt@n z1}E;^S{{sHaGEuK4fczvi!&HY`~I!GWr**=iTZV_XOuJo$lzRM-z=pC8$#j_f75(H z?VJzw)!Gt)ODPQ0@%7wx*2D&AuCwi824|e_)Jjd<5rsb4a;sy0}6Qh|T zv#aX0rt!(|79GE97Rhk9*dw{|d0|a6=s0N|XeaeN+e2gGWzIkAuN1Nb^yp|x*#P~k z`Rf2gldd{v@>`p04+3DR^znw_%H<7_cOi10UVSl#NF=Jx#n7@r70(c zZfil^s3-YpD>3)2rTyi>6QdJQokH36$LJ>^!-)SwQgo$=GzM}YVC<6vuhfYufe*LD z`yi7E-oYQV20!Dl>FO85kQueQriGt(2AtF5+!ME}-=Ds)Vt|nCiUT-v)A|H(!OsX_ zRmlDH%Vy}n7h}eWQe?18DRcsWDE71dJF&6MGBbdMw0seF_i-nMM*p~dzO9x)q7y`s zs`!#}u`C2)>Aghot_cQcnT!Sl^bKB)f$p|@MRZ3;!JGk$)*N3}MFjEnEe4J4avDv( zeEEAwwEA@|gP++_n85QRJ3m|zd@b8=W9v%NWsBo%PaLC>W&EwHtTHESLMI397f`gZ zTdYao(NdsP+XXP@Wowc5$?qu}?t2=pU;Mn^@FYXzn?v)HpW!o5@SILvQ_|0H z36m%DEOqY*`@LspJ#~t&P>7VxiqAy4XMLv->Fx+O2NCxeI#$?RcFbIs*s*Gh=Q%me z8d>67+NvLf?BW9cbdY!g?J(YXf8jIoV8(G!awq0Ol>z}YT$(uCe0`loO4cXSl9xus zA*`PCHi8jhMvYriF#>j580EeEH#Bt(1HyQh5E^U7EaEk_-;b-(1I{Sw?iEa(HfvNv z?GYPGwq7-kzEbcq+fO-O#lHtoo(4{!;qM93%$Q&uX4d5p%wkMwvIVYV=k|XafWGpAE?B} z!>670F&VZp#&+O+zHRRekxhrOGy6Ib4H3uj-ikp=0^fL^?dP#B)c|T3K+O0$p7fcW zTaqxkC+y)ydL+;)tHJvKDVPAX);UJk!eq(0Fq(3JFab!UYyv3!PJj}Gy!iQ`7K9n1 z+3MxUsRvlgER)Y-qWA0GR-j+rbL$)3$u`hw+pPnChfs{9Ia67Ep&eP91GIqhnOZuG zZvYP;%)^n3FxB#)i*xGj%elnne!o$1%{J=*Y=qUWyT+LHz%6MjXN$kRYrq!YjN?0h zN<_|con`y$RFhi(ZoH#4hU?vpKtMG-Jmr7 zg30aYC9-@keZlhIRZT3UMMso9*9OfhZcweMY!x4af@_B5$U~fhRTAiS?DQ3F4kd#!T54 z#I~}=ZH=uU_ksRYbld5+>MZGdsNmbRs2Axh%*$Jit=BH1w6Y`Q`~7l<=o3GU0y@Kn)b@nSXrY$6Y`HN}p%`vfovEG4JN$AGTY#5@>pLvY+?fA^X!?5oh{TWXh0l+p0IIzq)Hr73i zUc1NSE#mN7p`Lyx>x}{C80L#Z#S^^Q#cSYW&N=|t5C=1UCtD3SGTAWizk`z=s@KL# zAAdiP_6KzQzIp=^at3=qsH+1;HwA1_I5nIixKUH)tlq>ISJd{3%{O)i<}iM3^TwOj zEroVoj4kI)F?&Ecio8!CHIeQf9Zs;a)yR3a@Dpb{(w*M#W?0qVZR?n6Ic7m(_nKj|7Tzqv7%3%NGA~*ozU&W#sa9zQ?afP5UB&6S{FILXC4#4h1awv6=Yjz4 zsSyH#yGSLD)0xbI@dVgEsJ{`?W>9B4zPq|jYgcYNA*mVOc2^4IXX)vA7DZ=)#bJM8>_bN^O{%x_jFiM{S`RF?c z01RV*epN?6Ek8gG%KkRuzw1U{{1c$SOQgZ6E3Lx?ihd+25#)@m(73E@0a#pCZKiC> zbDy>Hbi3%h_FmzxN8~G8wGWh1U5+`4v{H$h$@&j-E>3&+Th;_y04KsA(712obG z7;2{?FV0WafqTQN1;444rQ5#)JZba#L9BR2!$hY;)5i@9Tw4Sh@Ym%_*m_MtagcMZf2Fg9$&dVZdZ&FmdZ$YzT1RI{_8pWB>Xn}4Z(g4Bc{ZN&qh~y(kNkRDjZeNm z4|ZATK3b7Q1qi;8eejdSH~O+g!<2!lCdyo+IHtD&%IRR3KEy9N6zMw+klt3SeSB5B zDfb>Lnp&1}SUw#E2q8yw1nlcNHb<(;VLrH}SNpK&-JOSKVF9||+QK+>-I-sqxOIOB z@GF@}$e5-$A+js{C4%>@o(M;0H0bx7ec=_(ko*Gqon~8boW_@qpL~ZfubElJ^slB!GVaVrIeal`XnPzwQ>f3byDUQWYL{zHECUA67 z{La1Q8~;=lQQ>E79lYrA$q_I}RV>yVZD&urAp2E`UZHU-r;I11BLVQV{=2=}kA1VF zPPG?fX4VWBiypbJyuf%b#6tV{8h5Rr(9d(>GXx~snvXW_`y9Qumt0ELIXyI4$21-c zWGONSjJ|0*vXx1t*I)v%s4f5+%>gpK?c&@oL(FY?-c;*U{tsYPnFKNevu@r@kr!sy zxF`3ZM~*gv$!NE7Ui8uFtDokUCTMrksjXAzcfXi?TTWKxI!j3=Ix7_?L%pwY`D>i0VzImUsCg~vy)yhOPsaEBk~ zR#GaAI?T2OAV;DR@&qP8y7ruZ&vE>CZ=qi{Q1|%>rz3XxN;o`JbR{r@Vn}uJ7afG53T{`}j@GU%Xz?RPKa(1l-y70TA*U=?>Mk>6fl-K5!fC@&S;Bj3pQ-q$b76$>2hHr#iE2aE3IS zZ<NJ7`9b0gMRbi(o(n}%dyS@m#CvIet7T zCau->tnF@R2rv7v`l2M3u~(wpa}spNa)(cny&u_K*hB}1EQ?iN%Vo=nf(7e^IJ?*Y>iY^7_=bh zxHy}=m}8Q|~pL zX)&i`n8{32I40kAV{V?!V7NSS%=~+b-ex16ai~E3d~-k4wC4<>5${&0Mt6Odvm+n&iqLD?qO@@Gw;Vaocctdp0^7WSe_HxDH4WyvQKS*U3?rodhsVh z=5UMoa4Q**EuKg!dN+Lu_$CwuCwaMLuPK9-+kZiCli{3G8oRBwb+t~u(tLBYl#aGE zlj*^@jJ${O`?b^Sxz&(uqc2lRKt~CMxHU58jp03pA!AbV-_X9 z`q0OsTb$_kmf*FsV7;e``BK-Ht*2>yu547!ugns=m9IoNmX8fdObIzw(jG`oM71zG zev8#>u*^2v2-daMxaeoHk}t@v&)=-t4!-iOjL|lFtYtC^FE6m0*|{6Ea|=7DItx5< z*n5677<3Hw-|po=CUM(arapVVbCr<980a|n56Ij5wfZ0yfRPvp ztt2TXgZP5@~!1 zJU038$<({jB8k$!4uJf=BpH*O6yC_?RgW1W<`LQEtI z>2Y6rHxx5Gq^2gc_5ctNF+N(N)2iNcVh+fcR|Z7`cdDCiY(MZ8G=p{_qxfX2X&0W( zx{s(!D^&?{Dv{7r2>KRVtc|Fy0+ucU5V{Y2p5A7ZDwIAyw3=8r3G zbJ8#tseQ-yq-NgKt~nI4{PmStnaCDvnTS^@E!yezH^(fiZ&?)8^0CdId74)JNVG^s zDd-+kh0Bpri$uP#6(?6WH3eD+N!o55xmY^m0dO!AUlieXyG!USRKCU8Y6ftdqy4Ih zSyJ8Ky7aPsd89wVjJ^p<%ZQE~2f}X1%7I~*mudi z%-G&&lJojwEpcU*@Vqa5!L&ZZI(Hx$aJm@v@g_GrBadG%Z8ELNvV)8%$R^_xWxuw^ zoC1~()i=%hI|5<2#}(jGwS2mZ1<&64f^jep3VE3O48Oveg;`s7)vw>9u_^m_6PUR4FP0uTt3`egs9(=t%1L7&}cZ8F|pfIn6_)U(_c5=%m)cP znT8Cc!^iW(<#z5E4Y$4@Ghb=Y+eC9z*l&1$9B+L|UM+^?ubaQNK8+~lcZP}2W#W{h zvgf7e&VLg>I*vRpcr&D>R@`_Wd_!RSMrL6J_kGc$9+~rDOsfMbIEhFhA2;MDR|?RX znHiU-diQ-NTy71!#rmbe^B72x3%5nNN=YkSew$toCRnK`y_sp<&hT86JdqF8E)Vki z6kB@rz`t~9jbZI0aox|A_01877F&fq*X8qyQjzAuHlp%QcrRA8z=8A>js)HB7M#o5fG;NjdTj1vYOh1@+ z(Sg38Ad`*N@80spKpTA`*vNlMY_^WX$46BA{thNN-Ht2+9KW&X+*!p2h;nHrpwn;ZOa(v!}PkmZ9@!L0Y$_L>}4s;k9i$o zbm62ZMsJY`#Y4o4sC)0cOHh%MOI|kb(&NaI(6|$VQKHKs&tzL5ZueRm_8?NPT>}u* z5>$Qn+8?@r_dBs?3D<5blA`Y&hna)oZrln@HD?;2zp5`wfmnk78gF_tE$>z_kbFhY zFJ|O%7jt#Km=MIG00C}aXw#coSE|B)F!lObr{{RARM+g`zuE8(*x)FvOrXlR_XL=qhQ#{W!7!~mNA@#iQ(Q!=u{OPxjJK7gI?V}Dbz%ksDNzWeK)h@KV ze+8!eS;3xk*F@68Q!!PC!l+r*gVnK={M`>^CbkkOC=QJ9o)P`ZI>vTw?Ca= zlbp<-?dA%vXb`nhi=hpjY{wh~4szX_<8qaAHe|*eg%po;C&7N@1`CyXt*e3im_=f+ z-slvQIHXzuy^% zJ`5uaiL?3yA;DPw`Ay>^wQE`@*4{*-A(?fswwIMxKl#iSZ$cH{qKAAzv1{6LwvP!Q z;)1^8{5W$LL#B8|;BlPHxi@p{_s>2ty?THL;$RsP8q0}BF+l43u8=VN?y0nF-}PjR z?Ugy-qS$;Sf4dDZ^v=b_fM&_F<9Z(17&uRTfOCG-kLTows1A|-0BQ!v^@5=hu$lQ& z6zK;}8-@Xp-y;R4#N$FMkO=HM)9%%!Z-N3Td$O$Yq8XuEcF?@Kq)2nNLr#u2O+GrA z9>zWI>%#Snbuq^jcg<<0cj9IPzt(;yZzW^Xj}!jW`g8C}>j#d8R^X;;PIf5#w%(SP zl=|UZoktocx&7VdDo3yBvms3%!KCDC_sRS3x-DeP`U9&!p5*uf@Vbtio z?WMQ7E>pg8;Fb9?!(MMts?IliW#p9ZR@<<8#^SEA4$m9py9sf+j0TLL=|~Nzr<%dM zz$#+?i>y#y)ryu9!JWOY`k0s;MNPSG?E;e7t)*R=l#NUUl zuVlR*<9H1#aX;y;##IgnR$+GZ&EX%Sqjz6PFp4(Gu5+e_c&g&X6EmFqIS4tr{(w*( zvd#P_L?%5aih>sJC7;>CSiI8vJw5e{rTWJtfB8p!7uIPPIay}5t?T~KFfm$=p4ead z?V^E>t%GLr<;@mAg4>BCT1(xhd8%6H>lk%ER+c?J13jL?sfV!=B|eAye6&!qlCydn z_Wuxf-tTO^@&B(KJ4Oj&uMRb8#aGWGfKAR6ZY%hm;IrgPO`#jqi*_Ft5?v)!83?*QDr@BG8y%+=C%YyLCVOfBc@v>e2x@nZ8DpP}LeB-iNx#t~cM7PAB zgJ+iowxihl&*+7zRaoO)UOLN3tfv>9SARu?u{hKVl*uASKhqQfhkRb|-+Nw{za{He z{&X`Np`6I8Y%kuV$~^Ds0I~K* z^~@?={(8xBR8KF@uAyS~668um@0dyGdFAyTH`AfnJ%cX}#!hTj7zo*zV9e97#1(XkFIt6=5O>c&; zs`!boaLQP9x}U!rlnd?MWMnr(Mfz5u*hbYKm(rPj^#JVUuMjG(7zHDUS0l?-Ojc?y zPYFy{LRUg=G!n>HC|N_LT=)Y{qz?P`jJ8Du1oRLZWrD3Quh2EL2xNS_(id>gw`Kt9 zDi<6#&Ckg2N5x?}Tk5+ZFI#9I!;fa4J&*QApL%YO=-WLx(PRwvEGV}2-TO>lH-_rd z8X{Lau`!DWfM~Xh(@yS82?PNp?96%dJB~wK{86t>c1YhXWtlNK4zg=jT$#K&`JLR5 zG*?_9>8=_1eacvN4*g-@C!deYTLBGA^~>cI(ADqp>e^a1ZDU-n+E=!O{jwi?>)*K* zE>)XX;q>j?tnRCoM(wA)4^4G%T3q=4${NYJea_6gcb+5jnlbSC$X;~VgHhfw)q3Gp z16K>E`!OBHph2%XvhSv*q6x+|@#~-QJI2Q|6tKx&oix?VmXGhi%X|Uczs)CJf;ENs z%xX=X)izUQHzCFYB!?zrzpC~n-G;vfzh`L`WEU_76hkf@MU&@1wlq!UXfE{eW2X%?rw;P4IhSh_XtRTAw$T~oglnur0$KjiQK75#Uij@S)c2=g3g@`O{cJnT8YX3Vg%i64bT6Gmc5{nP&)Psfhev-iw{s zZY$q(qslpVE!eJt`DnEP*r;zAylud^aGiP&qE9lf{gw&N(R7?AFTsBeP@MnS{+|O7{~($}HlOa38#P|$ z&yoX01)SaRXu1_WAV{y2%{x%&w>hs8bl%_uw{r#szM5ZxBZ_gF~f2}bCB-?iWwq69DC3V8wVSe`F@ zNz|T90RIT|3NR%IN+Xa@&-WQ1UtklxV|_MXm;BV$Uzf5e4I)#(H*o&$OHsJKo+IkS z%IMqIUt`|g36L|s%HjG()Vt1YXv3I5$F?=oz2b>&u4 z^o)x=$I>pFG-XGD%GU}LQNXmhx#1N@czc>tdF)sB(=B^2$5rk^tFF*f{&ZQPjzT}# z_BQvhMN-Z0)p%V+2=wJr|096wu_!FSAJf>=q+_gY8dU()u=Fnn3XRsK`mQ|WSJ49k z9j!Y~?H$$7$jm~(bMggt=TpS(-+PkLWM)}?8IL${=kb@hgfgZ16KXPAZ!sv91SQxb z;mUt_u9CB&AtExW@b_N;=}=FMH~rBOY)l#gsu#)Udxd}Z*De5`WMe`@8W4tfX3%ZR znHHNspzb3;r0R3NPC((fANZerFp%=jAbiYHYCWtUdyR*Z$)20f1BBU8T`XpH_rHP*3|1il)Oo6whA;A=kue*5_4RyD-mODKrsXLa=xASmo z`0~rC6))*l(_o(N>Eu?xrZAQ3xo4sFPr+3fzF-Kre^s-s9BWh`+lpe=( ziX-X2h!m_8TIG(H&T`R(e!hv~J`V;WZvXIIfR0;F)CK<%Ld*T80Ks4- z@<;U`D$}NpWxoe}2}pJ}4n6eh<&QlYPvLK`?TIwXlM@xx0f{BV# zOCG%c&uJi*RvrOy=NG7hutU*m)X~noHQ5YsmU*5f+NWLG2X8AZL2_ zVp$@z253}|D)}~DgxSx-kTW%6v;U=sUtgivu(41;YS#K{)inTRS`qFRy-vG zbd}qoFG77|`r~`!^YER|i)f{157YK*Ve`6IZ;=ZgA2E5rWdy$Is@%B`7hF7+Qa=Rp zhT9Ylac}>JPIXuHy5XD~HqW9;#i%zF6e^ z@AGx6&d`u+R2k^R?R)dI^SBD*{4IdDt|vd=PWQ0ODuo_Hk7Uz zd{y@$dly7yNOv}ee7+On0-G`uv0B~gd1Q*?L+fq>JC49kpm;Fj3?HmD%ygb58U-pd zV0&R~jh7gFsP=t?p7g)=G_W_&!IY~{+<55HVAF8AE~7UgD3CN!Pc>8IlGF1}XYFrL zfzADCR}qZ+^Ph+fCq5)|>|{6lqfJ~*$k{ILJlnl53?6ZrNN6I8s1-olRaJHsDBCnN z?wk{!Xj-ImM|S5#Ve=ovA(OVhmU}eD6Yw*vaaK=9v{esQ0ej2zXZEvIwZLI1Ng3IU+_oVdF8TWUan-AMgwBbIm*SaK{Z<$K? zucy(#@iPCOxcqSIcH3`nlL?O#OM}DACprc4v%a}k$|d})Y!0n=pROESPsu3@q;5_S z;|0JBMMDgl!}ejvusMiU{_ei+u3l!E7Ng2j)ch%s#N50Sx*#c7jmqY8Kz$(lxvMHC}Wn zOjkFvFU1JuthzB>Z1oBi!lsT-kp=DNFLNI`-do<-aQ!_5YXH80U4z(h?f_FgLB7 z_J3Z`e=jvU4)_g#@7}%ZKn$Lz3L>U}B9$%BWU8g0Wjk8?pFJ1|^@#s-TmmsU z<`n~I+7pjLTe+VOYY>x>(`***wiX^AUTgt0k_LdlF$qLBT!@G*dVr^9UHUUj{TN8J zYyrf;YBEOUVc#iN(NOEqld-~ov4tOi8Y^el73aT}OWQgLn>80(+#>f2BqbVhF$^u!qBWGW24PcNPuU;?F1kyNq_r8Y=mZRmq7AkaX+4qOVSP z49CzL=xpA0@x)Qd3NXJM>AB?gqla8GCtPrZ-^74PC!?Bs;j`cI#C-ho z45T$cc#H&e!k_>jfqN7{diCHXP#*Z4A^IgX&>!$QfWuNx_Ga<|N7o^vDXnz zyZCaDSZV?G6^{=mBzLb58 znv;MMfKzdv{rvZQJG8aVKOhzM@9qH0UUqlh40K;0rq797k?k`6;hi*n@fx?Prk-C- zD?IOyLDuhFi;RKDy=V~d+XwT_z09&+AKF-d_x=!DE`v!eY@U>|bjJRSp5`R8r=eIgH(C=PIQ?(Zl9LQZfq4V`tOZ)4P z>t~}4j~#U^HHAs4wJ0cj>F~=G<21I|7l7piVQa9ITu1xy#K@b(-;;^9vw#mzWFv zxA%|Qx+hkE@IO#3k=P$^K zLNxQXtNHZFefNPk#FGe;o6v^IW+!(#0-XTs*n1)_i4JeXs^D-s}RYCdUn zIK7+MnG}ZDmz{V(wthp8vbVo z_@sRM_?|Eyo0lI@PH@c1u?lG-m^J|3`DDtQV+b|3Ky4bvybfrap_Sx9i#pMPdLvqR*3+2@bxcOBOT_o}Or~}b< zKF=6h{SRt**%DY_DglDu^Pvr3G5ZKOxhFo7Cv9HzqZ6{6Z>R{2uGGNi_er59ib%92 zs^2}7+b*Qvx=Etq<{M$5rY1bVdl14UtbOM*z_J1s!B@xL&fM#07p*|%g>k*e$;MtQ zSSZl8J^xKK$XnHm<=O2j1ds#!&q^zSBFU$tC!c(`TI}gp*fr^1P3Qa~?MeLrNGYe| ziT^??>h-pq0ua>WQENbw&wnCV6%Fh?zA1Hwxy3b?AUIp;nmhq$m7oP++_kc>7sG() z?<}R5KBxXvf}0~CW7SnzeVV%Vu3}*t9Vqf+<;4SQ*Ssg6n`T2UIrA|U)nDBjthw!H znMI5}wN<6fm6hHo@L!;om{WL?xuQ&zR%aKbTG9X3R5JzjMS2dq!#>f;<)51dq$3MC z9N~#9tA*)CihUcB?jADiMv#5Fhm<#>JfvZ5*=l7v9Iv-GUhujEXd?EK`57xiFs~)38aQH;2cIav6eFnZalNxQGG-tXm zqR2R@x}}w3?}(L+^`#SLcj(3z$9z5kto)YGKNaDAsNmlWm$;Kcw5Cg!v$B?(@WdT~ z32gi#R`5bBldR-h+!uSXtJW>{mPs7aJxYCg$B!W&X(**0?}_R|>?2u?J4pbwE$Yz9v25q@5&^1dTQC9ZL}92SZcL!H$tDQqyMLnT@(kuxcZDIRmsc{7DqNNdPn3<7CT zV}+4|-H%&`S0JmPYPfFIc)Y#Uw@&`c5I`i^o{t5C*5u6!7ey3VM@EXZs6yR=9%j@K ztwhUQe*9*hBJGM+7Y{Bvj2zBV5(P5ExMP&|a?mWhK;2>2N1aP5i^eu@Hj z^^|dAq{`BHlRZ{oi)?>EkU7(t%%@8MX>|57q={{E0=S=kWl2z@NnYKYo(({?KYS**ZC*zA`OOu`u}y-f47;;)ug z^of297G7+JG^Ui7M3cFgxi z?xn5hEhAg2&n%L-?W=O~!{VzDj3ZETIXp$M6t@m*yU4pgEZ~Q&hb@l?tkTvG0aVeI z$l2jcODc>7kR)!k-U8}zYV1{)M{O1yG)ukkP_^c;FvWF{idO!TmyAg=lzpY+3f%U%ZJY|-EA<;M_r6d;M8{>S2+u;81o4-NM0~k^gV$#N4A?_JYewt(v3;V+{l`- zzVDj(w_TpB$DEfskHE*J1G9XWTOiK)K?!@eCMbNFe2i6wbzs|#Y`r%-4v%i=T>mZR zr1+3JA7NF^h<^#tJVPut{{Y70$0xiIZqyapdq7?A;pZ)wQEmTDmOX-{WQ?@_>IqA0 zjQ0y)Y&ay)&dZAOQ`ko0qK?v?zQ?p^f(^|Brz@e; z&sc)!-!E6?5OQ^()Y_dj%@gK2cfhO0 zwKbjh!JXs18NHB?)lDR`k_-$t%mtG|`b&!3EHr03SGO|VUS$tto0q{>;hLiIjk#<( zZ6>IuGY{2c=fkj{o&K&~B?S}Mh91k%PQS9Mjs@*>_H`>`gjk<);dwAOq1SV%xZCTZ z1ITYO-;Rx>5AMC_44BW>h#k3T2l8*oRj@KHVDT;%+p#N^Y`+-rg06uK`o!frp30d4 z&55TPjIMI9SY@tfL0-TDkSEm)i_e+(bo1wJ7NH}IdZ(DJlx1=pM3H%`?}K3w@`krN zMn8Z|l)&H<1dyhV!n!Oc)DCW`j8%=3_zZ@OsJTZ5F15V&E@Bw5bLWcX-_i~fLTgNA zE^2L;Qn~s)<({7R%RRB<#Yp$3%t!4x2ED>~B}_Pw_l%;)1#$f$BO9=#Ve5KY92ifL zj}K||L}9#CZsTBrHM0?s;+9!3hmq0>;~!qXpw}N3xRFGt-k?WoG#i!u-{#pT)`+V9 zZHW)BnK3_p2EA8Ms<4Dj4{;>yX3<6N>a#)_RTHTb4f&WEQX_VR4|>b?u~Jh*Ck8@Ip%7)iRH+5iFzKB`O-VM3@`WnkhfFlu{IAf`qg{ z{>G83#%4T+Z0y@~nn$m;4~O*1f@4C(!FrHE*w-sic2Y(v7L#yTE%+h7gtCqzUIZF7 zAQ~-j5CyXdC2p--dmVFZitapu9F{3}jp%&2wVv3l2ney~esJ<>N^)9KpwaeiZ@LTH znS5J!BF;sdb>9D)1dNVJ5BA~p9da0HTwwQntV!}J-v^Oa98uP7!~UMKD-Y2~Sa>qIZ_C+#zNJHlG*QYcrY!x#>K<9Q63U zVG+Tcerdj8`R?8spsHj%NRA z6q>9eunFsDBZ*)$QO8?dIpZNYqJ@b!UpRypeI1<(6MCK6u4cWo2+}AeVc_Xe>S2m5 zjk%4+0lpfa^Zf;VP95!FwC;58kod1$^Mj^fe}MFISvaN zW+S)$&EXNH^JyXBddnEiFb#|oRAJAJd9WP52Y+`7*QtF630S>#D79hJA+Eco<>o8Sv zS|80Nz6cr(aMu-|-ovXH#tkqbe7Iw`$IKCi0h>U%@YmrtnL;%h0Ww{>oR9~#0Q)@W zt+x7?YnhaoqQS9f z1SDRLVTs+Yh>YnbL>iIoL#NwOQFl_?#yM7uc7#tzxN%-#IvRq-Nv}oMlhhFrry$Pd zXV;l!sK?b-BD|lsoA%^6&BA(+zlw!XRU=#(-mma)jMt-VZXX)6T}G$++Wm#`y?G$R z!(r+oh<`q(-_yX*Dmdck@)Vut-SH+4EfAJz?dWoUiiy74Z{LD>okbKf!EU5OitL1W zkU)LO3u;55fw2BbHEj8_@)=OAA*E5wA%A-skiM)U!$-TN|Dmhs47fEx>;pzeL)~Q| zffWj$zV#%=i)KVl{@TOKI{Qrz32Y-3F8M^bs9#~)jnAG}hJ^<1 z%nxJ3oUJg9AQN&GGB0&q4BCim4b@XcgRU!NH+v zxctm8Q-+^3z1e7J16LWjVD)bnqR`cSZmkI#A!rmC8_Jy zGv&B6k^eQmq=uWTmgez8+g-xUX<($Fmi0{nH=$*yIsxc{pe@S#E9; z)q}FSJ-j&1(VXSItZZH&cGK4VL#?{>$l;kGr3GapdQ##cK<08m)GiD&kl@1?9oxpG zIFl92tzh!Nu}fl%RAV;vaRQ%tZOINkBCyByOU-ckxb}aKujp{%ed<0J&yr<)TS4h7 z$4p$9zb2(d7~d77zfHE?Q~P04$rhAXN-1|K1Yvk3j%O4F1Kp{zhOae-)A^)hGg_MNVpx3?m_kOjB34 z$_zLw95P|dL6qyHu|q_ki~Dgg!_o6a8fOOl)hJp4QUr@Y;uS5ttofs2yso)s6n8}Y zqC+9{&TmL@nc`JiG_RC_#ezX8_+H!*2(Jjgj+ZANWH8zpv4(v>nvKqYG||^#qzaS* z!bID>up?6UE}Y0m`HeT$&=_ zdam$<$=@(hP*EpSjE=3srnihO^5*=oEz}cgvyU!uW__4kqzqMcX)B{_Z$LyYd5spa zuUW<3xwe*3y26TPhf1w7jecA-!yaq4^R>?Qx};M6P?r$df5?0oIYym-`DQLM5jHTK z;I6ihDgDUGpx6`IncCS@$1|0^xFoF+NA4A)*cq7# zGE=$yG{bkvtrl)O5`^@EGxP;(RA^QT3x-goEWqMv5ICzby4HQnRH64oIj~|x9ICXX zxqVCKnnqi_{*-x4gZ6ci`9OrImsmaGzmGC}5Wv@PGWdR~HKMZ@SYP$Y;CX+&y{1-$oXyo(@{0mxlnFT9_7jLR>Ne-0*8KPQeX^3( zqk#rJ{tkuMJw_k(YqU{UJJ7E+&=e%GMH&^t8;L{X)EyZ-PVE<(*Sf%toO9W(Tx~CM+L_AhOC57k6tsd3yERvY0x_37a7x}uDPYc*QML0!M~7U54VZccuA1J z^{bBI2Bna#b33#qCD{0^nC3G2(Pu$2vWT|2W<=OUq@MI7+7SpG>TYk^aW`n(7Bs(syh2g^W`vjAwcNv> z^pe(LdRbkRJ>0uS0^{M5Sc+EsLM2tgwHB@O&_7TM6~cPhdFZ{SwNHi%`>8Q@X=O+i z#mWp#h?K;rt87F#B=+1s~ zby&RKys;p{lXeRe5yR_Yls$6_eSAcC%aytBb9)N04hP7w777 zW?|>L8oT>6RYSZo>@O{f(XI`()C(AwXjcts!M7PmvIbCXhzQU)(T4&IaWGA}ro zl-S@?O^cD8wmXHtf0_TN>~Ic(`pthBwk6;n*U@>zh%Tx&gqngNB@<~2l#BVxZO1?J zy3z9L$W#Gxem%zbk}&De_!hybR92SunwUN3P>en}4(?FPfYV6;lasCRt|tybgKF_$ znF<7FV#+n)FqqCJpV{OiZP0;6-b-7>14C{m?l>FTs|GM$=%dZN<=4`4P+$p5_H7S{ z9xj&WvhS{8Fewzn`GS>!xiI=20btoapN>PlJ>6U|TG}$5WGYoo&xou`Xd%_NtBI;_ zBWi1XDLR+! zw7*r-44c|Q?cAL7%9bHM6~*;rF%JHHJl9qMn8c z6fr4K;>lL-hRvA-7S6Y8bVe!8St2v%#kzL8YKm_}d~V0;8ozbOj~;?W!0yjAzLu6T z(5)*JDw}(TI+aeiFhMs^OuMub7R-e$Nzd?^sSmOaBT>*_6YwRIcz^ER%}7s~{hYL#j1 zxeDB1U0-^?H?`;|k@jH}?xp}xh)Ar6Rgi~Q7h=|XGOHh)Y3_L0oANr7<;b6P(pOrN zFeU+5CDPNTO8_F7$N|ruc?8)3agc*lrs5!7jlzahBhTU}p}J+S_&Sm>t8tPTn>c1V z_fhpU8qyV3&4d|Ox+;Umnm>h0zk@;US&JHc)gGgFl+X9aLXaMqEBi&ID=avpw`qV0 zK={@A$9k5G#u;pMqd~8kPT&Y^1H-{W)Nk|MI_kgR4c<2?ycu%D$MO(@kk2GbWQX(@p=pZ1G7ccVHJpGGHo#FVCH#IOeXn}}jq7-Z)# z;M%g9i&9D9iE;~DKI)h4M41@V zohA8e)-UJ^zo12Hkh2`_j%u59cud1l)_?w`&L- z&e`10stIpi8aV2aY=6X4L)oL6GVo$IpY#AdpXZ+b!Rt!kC;Wo_H$ml@4VH#1tLGWgmntt!`F zdRVNCRCZv|NEhOUud?oo+WQv2Aa$X9${CnoPuI+_6)7kwmh5C30{e?p#iF!_g*I+K zT4ia-a|^&1+LNtg#JDl`AX2*9?6w9U6V??E2b)%adI@K9ntca#ccX)g*E;(T?Yz=Q z=HgkAG@@cEGf2DCh&0wDmHA6F*CHQu-hxFcmQGsqOcyY zFR~T0s{mvmd&r^`yw)~Ds`X}~lwA0h(XG_fp|g_Ryu4-KcAnojfaGl}^ch zsdbd%TuOTY_Q*mk(j*A57=V6=ra+jKezB@l{Z%-w#_&R54lD<&0egtLAo-Ao^UYS6-6@K*cBS%MUU@5l(JG63K1G304 z->p7lsfJM^m5kHvVh(-eKTq-KK^!bfb%J_@)$|*(a#+<;wD`s`{y+mR?7`BrRf>;` zSUS22^H;bS79Zzc(NIcz%j)7*6%6O|=)NGce4oe?r!GPvf+p}~jNN`>nf{%no@*mR zG}lYJh@nxg5%*d;RR#HL^m;~6OqLOVUi;L(M;a7Q2y zt2|3(&tRW$?$A8@;dNu7gL7wg-NZmoQmH&OgWRZ3at)eW$!zbZezIRWV|6^FF4hKW z&Gx*bRMCe!@8_$j95eva+U>dt`UIcm_NHra*>Amu#jESvejIfJ<*`*C*VK%_$$vBH z>gK2R+3)J9~({O6^QkgX&U zJF9)3ey@vq3qH&1mQOZ$>H80&jY^Fo?$A}@4zwk2Owdj9`3NJoOY>aW(S^ZCgNoRK zQtnPOKdkwe4_ytDl-Rd=&Npx!3P_K=XV@AsE75BaPpuCB(}3?iKQ%}B#t^vuvQK}= z2W{N&Zn&56vMVP9-xKftAjFqVEc;hT9qwq>Vy5zE-m_Ijg~|IV>%P?_6FfstoY_y< z!NS;I)q58w54;!31K|*xIJ+x%{IKESAb}EaX?SyS7UwaZqwd<8B^gf?WDSBp3*&c? z#GqWrBcUr)tqtJIkg~JxFdPKy@5~M50o}k{g&cgpTXxZY)YIDn82c^5?I}1IE)EAm zl+D%i52@fUPo;JQ)=N<;cRUY9T5m>fMD!||hF0^^1(oxMNcIp^-xx@iFu4O0&i?G7 z7V@8t>ggwmBF_0f$9-9ZDKMgeSuH*ZU~IS+w->#ayQMem=Qfgsqn-0l%6#N z76JmQr0akR6q18)Ah4A%&A0peJrk#2hlsNDOCNHd>eEV1QDeLIkyc0}5Z$+V|4Bvx zN@v!E>^;QX?V)k9gBY?@5B?*FMKZ5(k!OIJ_l2cfoG}`>k!F@#P+jv5WmXCg8ZK9{ zPh~VXDzzxvB$8v^&&TF0H=)f60jcv9`i*=6HIQ4s;S+012c03w6TiTiV6%%}Yd5z4qS zc-Y~CRV6T@IrrHgK1bz@WVatzi|3LAg$1Kj$0?>4U@(cz6L3*Hx(x3xQe%u~9E|u0 z?AGdbK1C;?>3eICs<3ik??&Gozg2qRqV$yRQn5{HV; z5=0MhsG?aIJ0`}(VTwW}Ryx&t&e}m`J&_NAXW`Z^x)vC}x7g^)f{zV)v4`+pP+or4 z8L(Kt$XRmZsQ&}=de(UjFP>vXV1f==i5jSc znuB+z_WxFXv^w}8&G7#*F&W!PMtzQP=ZOJ90Ip_`$G?3u^slt@D>#FEc{0TIm%P$w zsMy8;Xo6?V>U7+l$|CiY=HclmNpVRJ{bh-ASv+&$~BzrVK_-#ilC{4daw;Q7{T7F!cHcQaP^Ik}-50y@xyc={?Eb zUu!Mdn_expf#lgFn`B%GZ-2>+dNG#+9XWn%d>u-2% z4`+%%AN;1q87_&SH+It09cZC{S}+}NxB}Y~+_C)});CA=Wse1P9LWCEtvpgU!oaX? zl8YSTzdO(WAl~pXS&avo@8|*q(??tFWNr0>^4nWoW*7;T1 zqU)eBv0FReCpuWwC~zzKi__R2xEJTzoxj_Eb+>tf-#aG+#rq{S}8NwtrV|1af5_{w`=n zykkDowUBNhdC2FoaH$_<0AQ^$jPsK#BgF;WK#LR60}?KvKn6Ri-JpJ{QE9NDR%^9Y z75@~73ZfCanI%{9BY-FUI64@`EhYY=^5e|mZay@P98mMAva>QVubgSuequiwG zgyHX#rNHtk7*!x&Oi;mrzkg-uist;3a=dLu^N ziFiwZC<*Wy-8=V`MtxXbZv3VT{MF4qc+R7fgYcRUiLuIqhvvIFC~cp;8_eE)8Ps#& z+jc5@XQB_=5CLS4C}9N6hfK~&ik)K3Hu(%YBS}EXYf6JH^5z+TK;+*0%{@F7BnM6{ zute0&W;2CrT2@%2CV|c9n&6(~nbx?a&z%wPFNjayGTa;ku5O)O3;gweO!SJ}E7nV5 zbG3BEc7GnFve=QiFGTG!+XZx-e^>EAv%lO<(e(~hc32rR+Bro||Z)8e@> z+>`ugH>feUVYROxK(CMAI7pI?R1|q>eRGn|nO!9T8D;x-p@ly#NppwSh@+&4+Pc%J zn@w}V3W=o%`7}Am+7K@o#iNa_=~9@>p2xrnz>tL2!NWC>QiE_GZ`)PkTFsZ z-a*O~dBq^XHI<86DwXyh9`t@RtAtG6!8?NGZ$`zjk2%tb>td&AW72LJgw5HCM&COd zf6pO#u+qtfna99SD61p3pL~}3boj7Nv>;A2Go4=U=&N|ME`yv&fP?l#F+P?~>U2A_ z6gxNaw9XkZ_M__AHSSu%=Eq0-rUFAEkK&ORvRECQ(@Q^Q2F}3g-8>8C{vabPmJOg1 z+|$l&7rQ;;{r2kCL$-SXJoTY^Mm-5Uu>h$s`rj7xojB4p<2=zP+0Ieq4e58|?CmDs zWq`+xSm!K>iKsQS)XIBK@RE(E1|N)qF$8oz~l6uH$U6l z2L!TJ@07#wkqM@kwnL+PsV5F|-E#AT8;FT7mcLF}mdr0U-pd2W&?_&mUF5BMRcWwc zBv(bDMdUwUAoPn{R}Jqnzd#wS>X=G8MJE*8ar}7Q$7o1u-}AGKkNDf)Ct`g~l;(}b zz#M>TWScOpQ_dhiscVHaWhR|UJ>A}`X2%fOycKd>Nq}<3B1cIrKk|1hF9;MTn!#ca zxjYhKZvPTl{lcmD-pkYnFH%IJKnO*vUGuwizc(JCx-a_z8M~9aEPXXK*Ju+x8i?8S zQPHy5tbIuQam8G3#9O?>xNpkIG6NEl~j~c=tPk)&A zH;uvKFHo6u+99}r+{ln0gzL7zue$}1cLlu$vpD}#L7h0z)7Lg;X$_%Rml8P?wr0QhkmLq z1;cDG{eycZG#V4cvT&*~q@rw*&kxo#L&eCm*~SumKesqhkW4TO8_z+PbvB9Jxd%_x z?fd}un`bdr9%1Y-P)vPR<~TRi`KE_&oS23g`fL4~;oqo*6J%Q@JWGDf^IKyiA%Xxv zPra5k>iTA3g5I9vom(G24#a$4&(H|1Q#=^Vt+c)0_QAqRWnhnD5Wq^MvJ{tONIpvf zQ&4FseDF;cps{y*o}lq~D3U?37pN65zW;cO^YpQ{`a}&80flE!+LUzbgwPX>o!8Gm zn6oXzx>RG}8KI-h2oK|8M|OqVKqo?IggK-gvqzo@KE~Q><7qzsl=~ zSPpuTbn4+TP5=)^vZrQK_{T4Rp+$)$l8H1;>)TzTmxrvWI1C~w-iWoo8{%mfQlGV z|N8{Yw;|@*8v+9$N+k3j@roNqyQjpM`Nm9ZkTQa(1-*!^J7yb{eRppnp6XAK5oOwT zzSHUPtdsYm-x)c==y+;#Lw}&BWhtc0t;uoYlg}!QKTqlr7}f{Fy3{9_-YQ?_)JjsI z@bXOtfUJMFrars_p6Y?BvnN2KVI5a}&CqNRqcBEfOa=jn_h;MkqguB!ngQ&26xdD} zplH60;T~w-eDZC>o!A)zGA+vQ@zymP@v~;}bP3Ld7?`>K#?K!}-i7XxqyRzuN+PgLRZGvDlituD z7~(SpAS^e?BnpFmDNqt^{vjhBrObyEbun>`RS z;3CxBXqfCkE3RKr^SlS}_=6K-Ye$udSfIlZ1n==(`gcv4p0l4@lP%vR&S<{$&}vPs zO`g(aC0*MvoA?0M%uuhhc2Y@}->A=Gu`t2}XdI>y8p8on^fi~(Q~+2D1LmSht8yuw z3vV0VtG+mKiskiez)}Ku5n+vFdDB}9iS`E9LVkVGwd21Da{Y@yLtXzdbAjs0t24wq z6Cz&wt$jp+TSe@hEg+XwNG!6*0n|PH3a$b>OP)<}n3=klL1t)&^iO zvh1UOYlV@NSMJ~0zvqR-VGW0|YG+ubA75_KD|k%F1OfaRcK|P|n!gmoYA^z{ zq`tt1Sb%9y$lVzOWPXGFZ zU`lB+d(!HX6Up$9nY&5X@a|%}-uL!Nvk|1;Cv-zrCQll>dCD~q5-`Hy`C)+5Co!<8 zczZaY+pFpx1#ANC^>yZR@V-hGHFx=Ziq`4YvO^O9uZG`vumKwf+k-DMdm2%T|0*I^ zg;4m^`)SZkM}_5Unyb|8aCRIAXzZxp3(K$jkahGvzEExUl{)nz)$6-f8k3?p&)a?0 zSl2SSYmahNf*K^|6>^fX(_=z~PXO?LQmwH|UfCAk+pvJeK^-+mx!qF#h{*D6H3D_5 z==c5PA6d~=!7in*J;rs8W1pP^Xiwi?O-_E_ZGC~UY@z4x2sOCe{E%jXp?r>2u#Zr? zqR+!oA|h=}_I3am)*SzRZCG|+x{9sk&C&LNeB zO7=I6;N@PR)#6Ts{#yZaudTCp?G)Od9|Mi9A4AhXRXcd-BXJ-dZ4)r+g!r{H6%-j( zsacS7lTS0PW(MFUF6R4y-08Rv5krkY7h~w+;!oSZlKl!1;(XyK-?|wgn6k*fOg+Yw z0y~Los94V;6JYL&+Z~R1d%B4|@6c!c#VC;TRY$nwzZZ&U7}4}s48DL%tKBi zX>DgBu!Xhz=|7N7xxSAGtU%JlGk`OtSQ75z{ z*E4&WBcED4s{E^$uf+rBWj}@9gdfhEgKJ|V1fUbOxFz?&;cC_&PkXXs~Y+e(2^5^*%dteZV zb-xs=y0{gUgD`Qn#wJicYqg&njmK*Vb3={x|(Lfc?YomG8X225~3XHtVwDKi5 zhD-r%u%EbdCq+&MPzs#AMD!tmrri>EO;{%x4C6NVA^&(>*CRy(i7CC)hPQ6!`U zh8nuNr9ng#7(zlqngJvwq+tljp-VtQy6f%%{oQ-t=Xw9o$2oImpS|~4d#&&KegFx9 zAHW*06Ac@DG0yKThn)|Ho;K8wPT|-9I>Qu%6M^uED@LGE>DR*FeUjBarJ-XifTZ!2>{XkbQx5C^HFT2hR@m)A z$5_L~pqJ%bw_KFr+K*WuF_yDpCuQ}AHdIpmER8Ams|VgmvYtBHaeEQ~Q>IE7*`rGx z@69ZWE+sj@fbnR_4l0HL_OcNkM`pp9h7%_+t4nj z(xy!JW`C#!N-njRh(y}93`y)*&KryM=b!LarLP-ajN+#`3Pimyon-vJj2t#JMf98qcTIV!5MrQdGcxT5>Vv*SYVy zxm!I5`9Ezf5=9g}yzss70r%L+AFx6=RkMGMs%BOif{>lZ0g1fgf*TR*y_B`v)balL%_T&g`B!mnZ|w z7zto-cOyY$Vl6xtgvfKWNq;=@B0U8%oKBwJn#3Exg?srV>nzG;W40y;YPTG73%4_u ziH=sFYQG7>a;8YN{BL=UItYryg@mvo8_|HK8qlffrNf_vZV9#qV5j{ZbPP!+>;_zi z$I^mnwF^^dGgX)e@SCVvtokfQ|HGEx**Kh!KInN-#%Mm~YS=0n%PpvtV)X3$+J}@J z_r8JA%(5qarwlr_7k%=zj@F=gYb@txPU_6&998p@XVbaua=$!a8}7*&6O4tVM4m#^ zDDhVMHk9}8mvgKK)-~*?SbZQqJhSa>J{st5^;z2;Gc)z9QY=rFw?vKS?=knA#-G7a zz)!s1XoAl=Sv1nhz90osWawf!@U?0R>lOX!nh`rdd-ZCw_Mfu06T-!Ay2bWZ%$|Fq zPXm2Zz{TqI-}TGm3QojpQDT2K`=frF$=noIp`vfz4q;(;bVD2F!-DTQZvx+rP=jmq z*X&3(o1W#em*NsZ?qcE1fnWv3R>E0|!L2TNID@s?mlgn^g&w&cn#Ne^-`&nZ?_n2c z>5D`|gY1h<=-gY4dJVc>wA`WL5DdxpTZ)hL5)6AYv8fMTuW%anW+1t-C1_KMIv7bL zyokOnv$u`gQt%+a)J94NCt`j_r7o* zhhIw;jmj$80-OQT>8D)XKS=k1vf?-e=CM)t?VDmJ`Cb^sGbMwyXE%-wvXh-l(cT+{ zm2)?WXsTL^pNc?!N;8iTj6b^QALsIQ{2QCB)CCK%aMZ1vvE`r!Q$%e2#3t`#9Rln{FprQkj!(Sya@4Cxuf5 z#1WD+3Cp`mbn5r0Jy3$HHAF%!7E1}mYq8ud^${O$>e9WVse9o%IcDi`T&P&)faRYpAv({ zH~`ctkVG)dor`g{d^Qrg`kHLd;Hl_{uOs80&XWK`8{1@5^@VNMUg-1aKu1QI5cUAd z9qh0^ofyjSWg&OY_mjS#B7`OEdUSox_Vfl~lrzN*etf2Tj1N)ni=Fkbr%muJUC?2@m>30#JZBCuoGzs`TxVue7R3u)Fymi2H2o-aqmr@OfFPE(4b? zoMwny`_KoOEmAU}Tp5y}SrxD^EzsMoGZf>{-gwTjVwZn=DB4|-a>)LR95=_<=c3zA zs9rA?GYlH;?>>%nCSL<96fTgj@W+c_>3Lt@`WniApzdb&%MLJm#Bk@gN=Pty%#Y8m zprNfum9k0Ga$&&OQ{bE|=CJ_AZyKsQ9%xU5aV?LM2lhFEt#@aKrE5yf)!7JrNvL;Rlm=FweDHaA|?<6blZBf3PM8(RnbL z5X`rfugX8(qjOQ95W9~e#*F{mGKho z&r%^-KLh$`BF!TP9D%LE7@7z&AR`GVfZ=l5 z4{h54OlydEf!#!a=XgVvJJcXT_{7-OEed62_JtwsY2>%<>G)`jxJ6W|lC9t@z=#h3HQTyun*#mv=qc0n|=wzfeFuGp{$s28PXL2e98H}Gfv+o#P z?Hm1Z!rN4@ynZ+ce&L50Q-SQs>2u(>Cd+JHf3|;_aCm|M7})jus9+-jYG&PQ#UH%^(hes3wuEn^&I( zmk8MgK(xy{(yYpv>^oC8hWNZiHC3tEK>qz3K`cxb)0;wYU5_zy-A?^Ht`BUuKe{A+(qGW`WlNC4j~wlSyu;CdfX%7V`DUU+!$lw%WH` ztqc|DUZO~NrHZ)P-%rRcve{D?9)Syb2ur-0hV}dB;HkCz150a2f@L|5&wKv<;lH0R zi%iS;s8#n7N+>3aQmm~YPq+9ts<$LF{a*95!Opu`rRrOCujZ6YNTPBQQWGD%CS;v-O3Fb?9qH) z=s<8H@{T#=#03$Am`nKgyGx@7oOZGj(WAX6ci{34HKDjAPX876?~{@>eC*zisu$i8 ze5gb-HxN7+YWfen+b#}ZR%o`y{H?E0!ZG=sz#Qg(M8}}6)7<`1OJMZaQ!by3ToL$x zZ#aBWUSecXnU_P)SOYxMA5lM@)5rez3?AO}moSP)`6od4v=an>7fxCJyg8^}<+>aOQ0Lg;)HZ{-q;rrbRp^k2 zj8eORM6{5eG}>?EYObrf-lBvf?^r`DKWN49yU~bX3>QBZSN}IT_vvlz^Ev9}q9E=C zLpD~!pCbp{%I8f2b0q(FYuPp~V^0ic1n#S7ruRPvJWQ=L)ghdA9Qjf^Y zYv2BKVi5bkH_d=xp6u4%gIJbWe~`(vM!2T@eKsL$^Nza7tzhuv*)HI$w$F_T{ykO0 z=QZ97nwE8TOsxvHfwAo6xTZs;q3@xVa*2giC*x%v0@vXrRF7<^uI7Q?T=N~3@^sM{ z9`Xd!DRD#$Bl7#j6m0oFL`UpA)zxzeiHjADWN6cdBfF`~`15}4CW8z78y#WJMAHg` zx2W0)-jB)`-)4b>b9o%S@^Zz6z3OId>g2=weYLEOF7Xj%&9Vmdh0rOU9s`m=h@gb|3y_#Q3L+4-G>5X8m3}lwK z4a6!06)~%(3W-q+v0zHy>VJYPtd}G+f8*a8g<*X)0jS&;7M+wSB$aX5&KyJyN8LY8Gp zwZ84t16c_YyYI9v&e3IRH*D0wG<;~NkKpPF-(t=TP6FlRh*Tsr8th(z)RsENWSA@< zi@Vl{n?VaRrf0)1{W#&03T&6+&(1vzu~q&Jy@ezJR1^c!9a>}#pkoPwYs>k2i(AiP z1W62qI~9m!nvLADM-~+?1XD!DLzwvjF?d;;wGC zD@KeURFv&}!w2qlY6rUc?u4)Rh-K(m6H+&mTk`ySeH3aG_vi) z)r})NRQO{ljKQRo!D;A#K%ZsS5zv46`)Xx?rjg_EKS@BsS0c{b!Dg8_WV0@5NwUZ( zL$bib@}vH|`gANR4deUSFl=+C9d%t$#P<377ni%ppl5qWFW5>=6NfAfhXD_8{VnU69^P@Wx_mV!JLyd0muNuu#%p@1b2#z_U66yuB2IkfvebDF=~tr_)CLls{K0Tl zV#UsjEQt*IYqHGrn|7Taa#(ca=x5&BRs37=^k2gh>H%G5R`S*pHd*R2p*-Dd84oCV zxO5VoR-waYB4yj`dGStbcS4JQk=p-V<2PlhFHrSxuJ~UOXJgR4Br~(+2%qP2Vb7jt zM_Fe5f~DNn5_iL$hMl624VB!pRu!=LcX zER|l7w7&QTMHbBWBzLsft!c0|!`Qz#S%RQEoG#VZ&h7NIV`^hzbJaCcGS~z@CMq=#E;%Au)4cjz8OuKz-in{{Fpv?ZAb> z4tKlgSKL^6h9d*;8v}tnbhU_1PI|b2Mr8c3>yA}s4Cn+y)L%$wzjY(%`Z@e}PEZF9 z7+|YZXtss=r>JcR{(PK{$|4k`IY=9Dq!URN64aB#jp#Z%sezJ9Mz`9Nzr_4aZ&$ot z@AZtPMhP>l)Im&D#51n{3s_lsl7w}5erT;4y-omuZ6z?wl_UCQZ13V; zjyA{L44zo(fFk7K$a^*wBv0;kb^Oq{?YL~PpDmSeb}|jUIEqHay-G}Wh(bO16hG;n zhnkOuCp*5^D$@Jk3#-Y$K{3g~>YKo7?KiC-uCLeK27odEx{oS>V)a7cr%nPA$n+y< zMyk-soGXG`m6ztkLi|VLeSHR-!CZ3l^7$x!v4hu$hq`lAY>5J)cP%MjlZc@3vR)V8 zPtgj<1)jSAJ2sl5mazuTZWX}O4}}J70^{NA3rXt0Mapw5g0 zCWewgV{>7N_5{3bDeH({6(NaT{s}5DWe8kqGBmgB>EJ6s)z70!LibLuX_>jZ!Ij~SJ|1;0IKw{70QNn(|spKUG!fBXyeB9jp zxje0V4vIclHs&4Nn)*u!)0fP@Tds6thkW(LSkET0N(qk2hO*0;RcACbHaUyNGXTZS zExUuclNEpF>TpX~WG~ySN{bOm@l!&{FZQKeG_Bf78<5J>%-ulDM+=?oq#~mv$JT0h z(q}rzRRvb(nOypugJ#mvqXXF6+GtPa*tHuYGH8Q9%Xj4KXuD*?AK;K8n{p?~l+Ne0 zWD?ZgoB`f)Tn_lcp$jXz_8kwL0rg@OC_!dN5{ml!FT*ekW<;BC%w95(`ow%zqFKxg zdp&QANGTN1AIMi(XZ_Ot?GrrGYXH`41Q?;Kfg_sVU_V%&+pmWsHBggU?V2Ylfo%8l zI_bFGfiKaboo64IH>2An-8=C;fOw(Yq^rm*cbMEQXG^H(Z-6&z&o*#NMwNooya#K zmOT9IQc(^7eZU=Cs|Hl{)&UDI?5`W9U|0fUlCOPw*xeKLdVaVNZ%WKH8oM~%q}x6L zT3u5y-xGN!*7A0euGNUGoXLX{8xgP01-0pKm*`<~nT5!XSe6q5g))@dlY1s;Lq02D z6Nsj|dt_4`KkNYYFN0UQPI32uDW5kA)B#&U_s^2dE@~vs59cA~Kbb_X4$l_s2p*4`(DKReyW6B*{_rgcPtR(ixfypIE{{nAOBBA)e??%;bc}d#eHB>*k6=Z+D(apXn zbfDh&@1QOXECzZ@iImbqQ-d2ifo z=gj_Onw<(fm*`0)6X~dR?R-@Cu_tl5BfrfKq|ojIOmU|I!ra>Hd33&WvDP=h4Yt>i z0+!HjCd2t!Qyp)k#ti}pVE~48@cW!aC8_e++ZeZ!foH-j$vLz%6n=FyBQQCx?R&?Wgn54eJ=~!rym@e4CbJzO*K3=V=t6F}IR$NT29{ zhL<`U-JDf2SYlI!oNgM`}2=Py-P_LE(VMAr$t@< zz|hJXmk;+q5p5VS>shKT-Ml_wv)pZrL3=TuStjV1K?I!pu{U zf%>V@uo;5xI5E<_i6Yu26937@w~ZoR8WcB~i%dxEz;`c?arf1N^$^CWO_d74k=Fyb zj~8!vHdDEvm3X|BcBN?Iu-Lq))5^lY)Awn{-SNC2$gKxQzRLiYx#V3#M{34aS@?>` zkD@rO`IpHR93~`-qRQRXbfy)ZdNFiS4T`(E$yrbd6`s{f{-^kAUvAwV=?mP;#76-- zrq_4xR`qe)AB1&09N}O+kI@3R0yO%19sOPbnsc@kb-s)z(fa{5K43k1&#a3&ReJoM z(i{hADHB!6qfB&j%%x*)88bS==(2!DEQWu&jl$T9B*{Jt0Y^sx4|L^6xRFXoP3;HD zfB$~nz-Wn4fl1fL?;&fvKzu_Z3&2MIGs26*C|B>2@LB0S zX-QP^>aswlnmJgy6q`qWD-I$Q)s9gmO}N=$Uw`lWEBHw=OWeXTWp~!a0;Ks+#ZH$p zCNKTF7I3N_vsslCGkhgJOfpaWJ+&L>pw78?Bs)RgI(2_zGNi6K9z#U!S z15&u-pdLxxOB&bMD|G>*u_{vwa=O>si)KnctXd$Gnu65LY6JphhezAAmT9udCtc@+ zRFEP}MFE3kE8uZm;=VVr^T_x{>#9+N5ers6M+@%m@lddo~(0a_j^0) zPuRGIO98A+Q@);fkM_gGb0}m%1%S~(UTN*s4q0*GNuPC1`_Zt8kC2yn4WXD#Qy{)% zW!|HJ>sv2ka9l4=fcN;};FfN+V`%nbG+-DE5kGPUmTRNt*~t{Dy1W|fxHQ}Wy&_9@ zFMgDv>7g+bbsney6fw|bphXB$ArNkD5|l4SG@}9V6m|yd>e9F?S!lN~L)&hmy}g&V zOnDijDqZ4^ocHI*%1@O(pgPY~M%pVPl8eCa6$u+-tqFw8iLiL2!NQW=zanhE3Tg>Y zj6FOaUKy2v>(%4|i_5B$7t=^(6QPC}h)<%+N4H;cMCaAWbh7utGtb~cA}e(jAtk93 z_Of#CpEs9-i>A0(WUQ&h^$P1v`5ev$PyGfug>Q!fLFwC?Qw<$xoB4C#TRaXnB`C4@ zrY~Ja4JRH9?Pm$#;S*xL0m&WbK$#~tt^NnZP&FeF%i>8soZnihEe}_Hm;BnVi^t+y z{|aJ$RIuzmrf&7&o*CoMN+K%0X5#S@U~AvvAgl2?60dV!B!c`yx(6$OWeMZlu`JX1 zX6=1=w_FXB!MxA*UR`{@!-8fS4$A-foTG6sPt5d#VEpFS+ejKyf|aswD5MYqF#Y7+ z^Mzv)8iq=U0|T{;y1lFUseM3NUVrEh9i8y+wI@Fx7rOh7%Rd%MINB)~HGfOWD1UfO zf{$fWPv7bTVWm%TZwb<_kW@UOsKQXc>yrxh%3=;NvxE%rxP4D@ub|9r~5*8^#$S|v#fdX>aJ9I12G z2ldrF2St9*6SNa)cu@IRjywMPkLd({Q};H=JoENEQ_bqdjhnD_rT)R9f_*?s zGwm@$xEEL%?(D%Hg)8S*wxfLY0;{Txv?OD@E`nFPTk9xAdc#)0#@z>tRPSE-mYI&s zIycx~mqIcVhYjbAH-%NxnGf&r=)4$gxK4cDpQVyKXC>h6x{=mh&c9U+2(EG(c&ela4xeNH2o&JPM1QKdp^Pg%&^J2WNK-t zw<)%0`+Jw-%L8FZXI8rfWp8O#xV%Uig`|se_K@U>`DWitNbGSH#^) zAIP30$*(6Du?NI8HMTxAa8H#uSwnccxhM7q6!87W9+MdSJk*#DaOfnjAa+bAtZ@o* zcFe0AHdVNN*?-zR3$^khU;j25&T$ZzYDp@A>L)E*zg_xWYCZhRU_%%|Bql;2+$nG1 z;N!BB%6oD|kJw5T39ku7GrJ0$-omxoP;OYTrob(5{YN+GLH$1`CxCEi1qcRf zZ9D|fyv2F-^^bDPjTG)#Up$KGD;^U=Fj*d}3hfzm(VA8E1w=Zp^=AaTubJK)V?_l- zp1d3r)9IsKY5Zsf7RWijIALP}Hho@XDXue{#ONW{8`taQYUU)uTE=qx-euJT^%d70 zo@lB~W_@H(DpbBmFkDxC#n z!5Lwm8R@vn%S=;;(m)}@EQb(a835_kK%FwB=#P+;%2yVbKT3GK9*xrbmL7rZLco029)VUfzK_DOmjA5Pt4@BoJ4 z#8WAaAL%ElOUv-LpIw`c1Pq@88tWM1;7Kilgek}8ZpBFp>b=X^AWAh<$+XOy_S{w zDPfy-z`hzi(nzsKFlT%ahKxl*>ZV}OTu(+C7aHfhFiq@KY9Uz|v zG}Z}l@BEXAj<`Y<;tFTrtH1uk=QE^}_T^bbSDUv*L_lQhlNY=YauP+Lzay#BZBJ9} zW%`JbE=Hn%Q$a;wVXKX9gVDaTb4fqp#Zb^?eoAdrIJR7c5kaBypA(QdOlp7(Vatw^ zFYgewDE;u8WHX8DT@Ud;T2}dKAq2Zu*j8SId}T6@1+=89()WE1-)*M9ORh`hF5Xe~ z0HxLNrK<><+K)IXY5t~^=|)=3%C|73_2+yIF_D=|_)wZ@^WhT<#4wG#whfyC!SCI< zuP5-msv%egyjzq{{KZ6#Zqc$FEjUd>?wV)%yPfTgtOLX45+I!KTH48?nFXRz8{n3C zv4*xe(6zTK+d2bgvnA-ajU)C#gUP)c}snU48O87u=H}@|?qO-xgPl>Ob0JZ=jsY3kuqW*z(w1f}Tr~93!y<9+V^CS!ed0SNwa7a*tPF(K9hM^3Wmq(~!I$r+Y6%(hkzO7Vse&4wJU@yx$^1bCwfY7MU7{&kM9W7S z>~eoDp1&m}c}e76*e9rqK9j;m`+%#yM9qvp!CDwc;imZ`>zZ|lRN)vfG_x6b?GTv^ zM1_(s@vxL=od(sM0l^*xYU?Yw1vRJJU0#N0{^3vJ*T{yqnRotvx5?BCgvk6H_Zt)6 zv^W|e2wWTmYsPaI=GBuCS08l6^F0N?<2AI9{hsq^WAqn!o?mTrK5GH2_*upccK-Ch zR@ySv=e)|LY1Z~}*Q~Ko9Y|nJ0eun&ox;(mQ8emsYkpqh@*o;Q?bj=i{){0=2`|Yp z%lnaf#$6bwt#4zlH4$*{&aGy~s%%S)^@#6069e^pw6`#5i=CbT;rb#Yjn*PiV^yHV z^gSX=Ni8@m_k4r^5kyIqi{Ua+&K@d9fP_W~z)#CjyJ`KtqvzUf}jIoZ>ON8KvfaOmnGH5!F*&Ow)HtKm3e(U{tP&EVch14@3 zp596YeI@3pp~gV!fT*w<9|IeV4f-vO#)a&IM>3|aDp6j^BYsU{60OAmM{dmDL6;*T zJLIQsGHjN;?#%;w2w?rEX4R}_D}%h2L+tkX{jrhtLGcN)&tG0X5@TZ7kurJ}-?PQ< z!g3nlsz4c{U%ceGuR7rqotV<3WkTt|OY1vUI_c8M`)`?~L%Q2+aV8=_6 z9YG#Hpw{;=^jpNg_27B^cdZQhAR2MDztz8XiV&XX!QE6#*rS{X(MV+Co|B?~tO+g(2Atgdou@O^im~k})J1A|Pogz@GDjv~P^ScMN z|Es1!W!x3`-gncx_fVapOH(=kwyd#*ncgSeMtHZ<9!F(`>#EG#PJ!^@hx2v=#nPCWP0wujbAZwvgwJJu8b#WbkAHx0d5;B+vJG0GaY{2iz%cJQt0$^bFX67#~0;Ps{WgHnrWtp0u^tSK*6DF zqc}YiRce8ZYz#Dle5+HF`9@DfMlM%QU{3$c3%Fn>Y>LTrVkFwVh|RXtfZmpDiMrZ) zWiY3WIzdqbAOJVBmA+qH7xrIXidMK2Qgod*j=6`SXp#qvD1NIBqmYygtn&kFsh#e~ zuJBbI7l;cJ=5qRC2_v7(7@tA;01rHCe`} za>1l?X64JA8K|ZLdHIEWa`za_V%V7lmCZz5qZip#wD@41=@1C^l-s& zN=^KC0aui+7j!M*-HXcMoj_eAtG3LuyaySNx2pPrf~q)>|5CW|zv7)5Q=2w@fKg;O zlS9&njlRZwQ&j}B@>~K3{=bs!3Ks78O?#Yu7ZYF5h1?CY8)4iSRgoh*h`y0c6|&3lRwk!({1q=OF*z#RK## z)qQw?&D^W0&?Ku6KDBr6AWWyam3{fvx`+}ZBuL(@{rZH|^O;-==T7Gxqu>m&KmJ<= zeOt^>F4JiHk_&ur{#ZS|p7|7xp?v0E;O{GvgW6oDR;_a~=ysMFu*7fhKA9gYIg9g^R*#8Lt@ zd`u}?GoZ-Ye1UMGm{HT!rPQ-yp|7l|xn;&CIQHD-^1p+k^KZL(pPpx1SJsGO$V{wD zttN?ivORbgs`7SEve$&~U@9RnDWOBpWdoS+fS6?^xJjVMhzl1l8uyJhhFXVoh|HjY z@eXt(dB2UW;bfx*d#sI4jI_d$t@usQ_#ze8`48fT)gr8oS}aceKh=|0tBnLBBsnkn z#=@6M9p-Tc8k$JT)#UsB_+uwazp9E*Um#BGwijm&L$9!0CUL~KLuz~mlpDT$5=DM` zqetBsR;YKiAsnX}Futp5$1BFXUyVaDO@%k!a&z2;)C|808j0}tTo;_#YD@0x9DAg; zy8F8!oEsBRxL{0wcU=qxwd?N?WUP=zubh4dUgV^FFd3<1W;WK&64uj?*QCzzYR@2> zy8{ilbd`U~r7j%r!RFhrxNEP~p7L$}`eZ%j%#)6NrJBnz686n&cqzPC0O*JwTqP_d zWPg$w{c(EKbSiFYe(QNDTG|woY!Tx#rZ|H8b_c19SaBsqY!QVS z80}fijNAAvrMld+>=jqH+va4#K0NoBnGH^Us>s8P`tbMFRwER)6$E6;TAK%&Ct840D$3A74>FB`t$n7%z zkxRa3$pe=iKm5U`hfX>?w^i8x%6nF#eqw&+!D5BBKDZlVv9d=0tDo)zYyyu{)4G-iU#rR_yP09S2uGGz~(8yqGoqFV^s5;-qOU7n|KAO!DN3k3m0E zI8OlngTJbs$wxG#O_|y*CyANa5B==-CVVawt60P#k|Cm_eA1b}OE2h`_1G=5 zsalIko4?;cos~(#DuCgm_IALPG*lX<<1pSX2HXnS*sQ&&ljW!r9#{{3wFruB4AIB0 zXljyKI2k`-xmr%qn|JP~@%Js_eJ(&z6Z4VgUN^v{c~)#UBPemqRvUkphUeO-gX(9K zryH3{WcLUiH;BqCv;R(B5{yS$F%B&FU_I+#+1>MPm)-b`$GpAmySF;oEtvIotPJBi zG2|)jQt}M0qM~#x^HlQv#v<(TKzwpr(CR%Wb#X+@tdbNmXt26;66^80nMBlQ??z8^SBBpKbJQf^$ffkm(;@whUao z>G8-Y)}e*vlFhJSj7Z!>FJ@QDM>yqo3+O4i>|3HbqPw2Ng6SRV(kUncQ_b(g)_GCMbtz< zPSL{Y9s%noPAz{J^zqehlEcaQJ|YBnZVWD64_|$Qp6!i}y;;f)i^xCTOFhy;h>3M% z=|)Y_fJdC)PT2tpcX^;~oK*HO&HzAOMtX+Sush)ccYtIz@XcBd`m)9qVo(*p$z1f-gj-bLJ6%KK`|{9 zG*8eoGX8KbkNfZ;r{?OVzILrOY`i%fEb*9B50f`fTT=b-$YEvv#&*V@;(d9aI5LfA z7ud3|;ohA|*2Itu_o+@ua-r?3`Bp>s4E#JAKGs6#BN@G!2$D14LZ8Vs)Wn8Sk2`?{|-UT%C${ojdfsi zN~pPevv^a>U@v&lj-dPe%?38HWZM|GUkmp&f)smw$|YA^Ij=^H(TUCx)X`Upt{)5V z7z*JdR zUu^Dh9iBv`st>ZQy~r5y2_}+`roG_w&dj0_9n?BT2*1aPEsM;I8h22g>3f{qTC>b> z;Kai`qVR^Tr5w~o_eG*8u5F41= zow^I$otP971l>M(tk3v~`Ik^_Jj0xyP^3;ub|qS0Q4L8V*DO~%Sy0Z{a@2K)puZ3K z+H#;(?7l@L=>D7mxw6Km+;ztH6mrJVea%Z2z?^~>jW^}W@dPP_EwHS|s>V6vIayhY zDc)ZW+#AYf3YA0ypRO18o#vvF6*ccyo$DA3TQzj+_z9*`>Bd{gHbHbPJ`VDu=*H-^TCnLt-XD3DFqAcAOM-> zAx$#(NiLHZgjfs=Bs4VP&CRkm?}-j-NKFC4;V}C)FrRPo!=(+0kH6EMDA>h3`=@qC z#ghIA%*;7ZGJ347op=Ddyjupdj%mDwLVku0yrNC3?dp!XCCt5?|JW@5+783kE7CKUT%Yi~po%Ct@tEUNA1 zTd8IxdRHVa7YOOXE$AKHiYcTX0in=X&vIGi{%xkwn`xxdh#2V_qvZ3h*Xvh(3O0BmnAEeY3rc^}P^h#f ze&OH7+>%dxqOtiC4`RX8sF8A;L&IRMIdquJ9<)g6aLEFoW8lJqu~_UnTKtxB0bEm3 z%40K!u5`9|e3~)wbxk2G#k9qewLqQ2n>?$|rgVyl5!bFu;gpSzOj7Q(E+Ttq-xED3^8}g#r_9i>+~@w0 z6;YPM`6cNq;9XZ97hdM$33}W%K{3)0v^Nd&ZNGa=RHAoAsq07z41_rEdaRXHZg+C5 zI}fTo(dxYb*-`v@dxak=tnBQiy-#C=fQIB0ygm2rYI_YQ#+e6MPgyqsN7vLx#kung zWc=gB-S{*LU$bppM$2k&aKcDn=F9fI5smdm8%s@@D$|YS1(}B2SX)AI9Pyq+($4Vm z$;Q$^Xkt*x>9az+bLt;k74oJo%h#F{9#9ezGylx#9sL5z1qZW1lzPP)7jABD%Yd#} z9#IQ`x3GgT@V6mAgF!DScpG%PJpkZbiB;q0137<%JhW-CA;}pP9mtmi(awkwvkQHy z(TY!U#cRq2jq`vF8H@|4jR@#4o*R=W%PSr(9{#i-Bk4bC6+wH$wsTM zePOJ@4{IF;qLDw}Ppr{_BUp04JD?RfOu(>RE4gZoOMdmRa&eU>zCVs~w>mrCt6+V9 zp!K4PYcSgLrcV!Y5vR?eYO92^Y~d<7srK~N7aCnqP=Hn&YrWvVpB03Ee&$G;hGUFS z`(rogWN0l!P3%nZ{uVh{8_F~UELA1U6+v2a&g-g<4XPL1W{Q20 z|KGxYFz2sm#c4%&WO{gLeowL(BQtX=r*zI_#RFbmQqbXEK4}XVE;D3VSX|WZZ&^a~ zV0fV2cWP3HAB;aO1i0dxm!XX3j`BF$lU1ej3BZJILX*LRW`Po@7IAs7kd1uww4JPH zR4}G$Z`lU3*~QL3zm&(W)yXho9pb31QHI4v;g@dL$5959?>h2M`Y z#KhqI3<=PbikUN%Hc>?)$f$KPxr}qsu=RUOswbTBv z%yeuNkY^jpQIq@T=Pb;}BJ=!TzV{>=OZ!i=ROu=}8oZwtBk$^TU7OChd0LjUYbYyf#f!xL~5%% z(WUO?t8a`X0~zw`wWytn1aB@cKgv7nl|1QYE?$;wazmlZq(d%i#u*wMHhcP{oUX?2 z3ru@cG&ug;OqmAS(p9CVC;{izmey7fFgLB#SQ*YTg)@Sa{+pUfw2Phib~|Im%=`Vg z!7pXsZe0e(f;X3n`WK}NV^N~lc&#F?TdqyWv19F_-r_f4W(+}8uQz4*E;LGyg!`t9 zP$vtKg(zL)`&9F%LNHZ?CAQQ@j5=g2V9}r7^rv~#?PvSTh99#MT>e|T1?D`U{xN|< zwbg5&AA)qq8J5-kDe08nl7%a1h(WR$im>zFEyz3XdTF8BsVws??t=1jz32)$m4K#n zcPeTtYT!=yA|#w0pP?gThUK))un>MfeG}87gi*E+_xMbD-n=h)(WdSz`3)=5HElY) z;Ei@d&V@Zf_qXVLym7|5!X3K2!@P%|DW&qL6MTLT;4V*4De#PG3fvrhKtV|vjJ8;0 z&RKrmlzBTA=18~|43+mn!_-}Y)duX1wdsHG{2h!QQBxb0s{V|8>G9DE~QhHMnXmDTDn_dX%<)- z1f)Z{-n00BKQKDXD9qUtPu$o2G}Ax9p|t&2^ongyhE4-+M`B-w>KOu{b@RII%3sAa z%4J5)_IWW`7Kkk-1aWn޴XoI$yI{GomG1U$uZYx`0D`M>9`T04NA?c`2)Hn91 z8aF4D8RP6)N231iu`TEClwjNPd_h9!@xS$@g<#>uI*W096}_Lfv9mz6=E4a1%Bh0x zs&ynMHT;%f_=x@UALl75$8oRB^~oJMW(drAfsLoz0s2(!K-;jb8aLTC^UK_q=)@_y zEJ2GxV+Y+vp_H4~8vpTX^sG)4>sA#HNv>z?JnHXZp?t44lp#EHu{qX}4>kSGoNRYB zt%KWdF#}cRjZ%a;t=zshbFkVMJ+C_Q85+#rU_4Bvu5nK)^(J-<94UgDwUPqJ(;>8TsDmJnHmR_k=JZ?Ro4BF}d-AhLnC}+ zvSb*iLhgxU)3uwe>(TX%zn!eNzJ892U(4YFn~n+XfCXg)1IGUd%Eys`8`@30;=Uq zWm57>r3m5|DZ8}=g1ty%K*ga;5XjAnYw`gRW)aHKrkWk0^pU%O>H6=INUz&=z zu09hF^99M0P;O5&qubg*JL@L{v+zlvt+i^L_$dH%+!8?fHr3=)RWh|&98A}au|#3) z7tGUo_2B#W@A~h3Uxke%g}qL>%jdXqWh|CsLlXKWVU+tT+MKpjrBpn4&Ou^uOZ`Od z&4;DL3(^M6kj~m28M0#{ctnne0doL-Wr#tJTQDX(g5v%8X~Z6*QR+A~CVh-<9iP$S zx?!gwvlbg##WN+pgXdl3TnfrM%ogDua&N6Eyl4@v46XiWen#>`xij-e8O8~7H7ON69;||?}Ys$`l>G@tXBogqj zWBnFH)-K>WTL!t%_YYWj?vzAT;xXOHSK?}0fwqqqZhiDqO$J~vX!tv2;~lzm0jWq}^5s_CO~35JX{I zNUDn_UM*W<`lj`~>YxT%6o`fKz(AlyvCh#io)Bo4TO6wugVEs;rLsu*ea$;MUX!3X zHZ<%~Y{RUZn@gv^8zY^LZ;3F;4VZ;Q06x$e*Jgu&mP|fsTckc{;1mDjAO7N=f9!UA zPAhSrZd~OVH%_Fjv`JAq#t{$uHS zbVJT#6obr2v?PF(8h%R+Q_Ynh=?cn}iL-wicRJHf@VqS z8B*(L_@%@=nTL4{3s0YiU_rL?*2VaYtcNKu(`NaleN}csWg~`RD?>N%bB-&h^~)(9xly7Ur&+ zh$A(YF#6O=9nixUXEk7m$Z@$cin$KF_`_Bd7{c1#gC+~P2ejBW2|}GSg>t~Z{Jmxr{#>P zz@pnazEPb?nMwTIV*a;*$f1>O;oZEW^TF8g2)JS)O{#%8u`4)swj2!9(-DvLB2z4CZ+-NfB-Y?`UA-~Te;p`4+xX$M9TRW?vU86dr@ z13f_a3D9+(C8UW;=$Et)X25p^v45$UXy80kj{|Rxa!PK$Z>lHWUkQ9qCIT3j zhU$nHUj%NkevI<*np$D;oWxP`bzohPO_yRytqC-UP%v)wp#BjU zd@oP>JR=Sz35y}@b=1dqrxz=4%k@$zrl`htex}4pRUBXL@#b)BpDfMR@c-TJKi9n) zaOvH`3+NdM$mSDX2_pmr-OU-^3OVd}Fjw)1A_v`8vQx`>Ptx{oizS>+>OG>JGWV8C`4&<_^>>52NZ+LDQXA-R+ zEifxRLze3zs>!bK1C=3#*-b2$PK2UD zG;uWPhcF0}Kpq=Z{;z=5zW3`V9Jf+#h-q{Gz+OA4()IOBGcoMj%d|8VdzTn3Dr?%F zYGwascmKfmXTpWhvXLXV5023QoVAK2SG!0tG526deztvK_WB`-rFbaCS$5!!PZKC9 zYjGm+Ou<7rqZYq_jrUp?ygpixAe0e?j{x-?%m4w(i-GPS|ADcGJF1Bc1F4Oe8s_lJw~6y#7G++~7*G zJTIjo&XDe^_0aEz>zGXKf-y&dYsq)cMKaCe_f~j%yO)Nd-G8@d^ZF_QM;FhkEC=mR zi}q3{3vUieh*xOIi_3$8tGMfh^zz$+;YC_yMh>@FM@}YNlO;O~N*dtbu{C{3rY;K{ z>q(}Ym|fVPf4b&Wbl&_u3QkUAj&vY>z4~ z+3XXkxtj6PYU!Eiy4%Avk;KWbjtfDkGVszj`uJ>HhwzRek)F$#df(GshdFlJ6#>|( zo=Nt*`tQhQSHlyn9-|h1(6(#@W{(cS)EpN&HGQ`8S?$so+j{1wG`E!Hha;yshuAl% zSD2UVXx0=R_?jvS<(o;Ug8Eb~B1 zJy`GBfkHWp`si^^dBgA5|6s3Y<^(qFXn--eDJmMqWmC_KU^m?{O0Ph zG`eo|ARWmH>j_Hw-iqH54R4HOnn}LH1M>z$A4iTNn!z%(v9G8!gRTPZ;G3XAO`@8L>sk2jCY4%nG^7d>7>WYZ{3LR2y6-65t) zCG8YyAIa)PtshIHPX7w(WBEMME~Dw>R&Hu6{(5Qkwhx8{Q9-dBaK;th6eEd(`!2DR z-9kY%1flOcuDDcSO#FzWqwZi?nB#&Sodf-k!Y{;T_4&cUz!SH5s@wA6`A}3r{|KMp zC?`_LT-~EE!#7B4PD6Xru5F#=a774wV__(|L&7L)m3mquU^IUA3$pbr-&tT19SLeY z?c*vxSWg6H%`?5!G0-6xwOMwG$surpi*%$6<0i>2?_yXh{$oU@srGb;4EYxqDWW+B z?~x|4@<)k7PyHH_E_HBvbADkTGkefj@ejk28sgQlN0$>D5^A>_3U|XzrPz)$UUnX= zA21qHq*}+KJzQw8_D7>A(OAqE#PbWWjKvcO{_6t~d4VbShh}QVo@*fZU6#R0!-oH= z$yBp7aWt8!nPU*Ft@?ej;?J)$L;R;(%Hei1=YN7EObqZ&5bjL##9Mo#9zh zvxO6>IzI?hEP1IE4Yldcyf;NWyDAXrAD_?m7SI*_ou2-6%~{FN& zoj~@$kMvJNP(9U-+Vte#-J~!;3S26pUs;=#4Pfqmgx)G&2k(0Wg$fqGK$d{=^QbX5 zcNSCg;(@sz?}01^DVN5R)2qA!xQ;#t4o8Tx2;uNoZQhdkjW6H;7Qry+5DsZ3$oltA zn;m2}RX(2!M|>2JnspPJVRA8AjJ>pOG5!QrJtOAHLd=;oZDF~0iiAS6zY9ZpmU({_ z*0vrn^vJJ0FU6$e&>n@2zZR*~Ds{68%8&isHCeti!0$M2vyD=_wOehqT1K@89hsT*H&E zFFU)A=W+9Xp0UO@dhXbA;Mfq!A-dPIePMcnn)lD&DY&ip>j?q?vJr`NfI^39^Iu(E zo!0;e{#4GfgKTL8-0n3U#bs@sAnkq2Ku8}A)Cf}$L~2YvPTm|Yi5;me2;s5r`tbFN zjI~9VC3m~zzvxjp2I+43b(-M&B$_>5n9DIZB(kVk@TnT8pQx2sLl`yP+1ry3f)>1O zrDe!osv0k((etB4eow{|o1`83{Cs7P^)k((bFpQSg@o{g%)LRhXS-Gp;xk^3X@v40 z^rX#iH?8o&@WnO@^PNGv6=9}!*T>W#`x#_U^4E%s&zOez{se7=4Zb`}T_@9{&tLV+ z+sM=)5VV~cR9d)k&PhUh2-qp~l5cg0$65me#c@(_i-0Tx9js6(-Lt_ZXWEa0wp}D@ zBJGeS5sFyDH%Jp>y8Fi3<6UhI|9%+iwn;v#g$f5-i?TE= z0`wwM6Xx|I9t}92l{+WF7^>e}94v#LaHGe^W9YDA(wrHmW5HA0#%`c#FnZIRVofSJ zr=(LZY6YXUJ0GH&-d4K=!TpZ?&gnwn!pyajqT)<%S9iCr_JbRO%!u+UO-2e2jOwZ%gJs++Ph)?vg(hQVKbQysiM zS(C`3D@@tuSBZwkhL&BPc#y?0iI|*HWJwq9i0p0Gz>PYQ?LsSk>4PqkN!BoBVRDVp zq+4Xf?p8DX*b|JS&`0!awyhd?dwYk|(7p1T8L=AbyPR#^5oIsF`-z2E)t5vsox77w zD}Vxy8Kqz+B>c+Hq=(?!4yBN=?J6D{qE>s!LUqZjCz_lt(-i6Wp)XE2j7DU8Ar4eho)8~={Ja0n@hC9S`^+4tKpXu7ozO+HW4%M@55ut)}Mx~TmJ*FKYT zp~TghQDoo{T7Q?e9%c|n#7xGdrKJHi3C}+0AL_pd@QgxgOHS-&Zg`BEDCZhHs@BI~ ziEqnF)zR)sG z6&tcw7{v|Kr4g`0RKR2;1j8DV`O^%hs)?TAvGdGEfvz5TR9 z@Jh^@iq8xKIRZfZGzNNG9pMM&Z2<9A{Bnx6QN;=hvXV~gIb|yJ_@8_+EC@zzJ9OBwN@z*6ylidd*pJl^p8Gc2N(Qw!S7bV%ULgU1H= zQvYv#M?QnhD&)WXJ4@yw}0K z(62>SSrz=cDCo=Dn<~ewqg#sha6GoVh9WhkJEL+Ax&^nZ{^kM@6EPJojgAG^yJhu| z)Wk2~LdVDdL%yBfK-pPLack%`7Y^QuA;?UfBq;YG=CXa(sG7ic=Bw^DX9N~H2d$g7 zXmnqtCc{_JgmIz>`|T_SYOx~i#?_f9vfZs%nyfbkt-O(nvbElP``4#!Ryn8-0}1?Y zMEA#Dw*5?z(b76j_R6#D>+c7?ulNIc{_#k!gv}a&CfdEHS+6|s%j-}^9)t)i#YEi6 zDsohxEtNK7oG`_-&uZwOLc^<{nA{*Q2$K{0bu~>gL+$J`>l82Mm|oag{DF9_j? zzT5h+?m%%dMj&WR%}rStvM+y~l#vk)YA{QHX{%OaLFU|9jh>UbRy)osKJh@`-tZ4X zp{DK@h2R(vt)|6h*^j6OLqdZKA>b*`?KJCu{3=KxI>_h9V9EUT12cl-a}jbSwH!i~ zR20AK8jf@eFWO{B32x;@s}KiRrsx@bwT`)Z1Y`1ic z@gJ^t2if{js&&1{P(*o+QMftreK_&6ja>F#hTQVu*FxOJgkj;Rbc2UbC#R1h!E?&< z$|*rvcBC#mtprUc>AtYRj4Xy<1MantAGdWywvD|+H}7r0d-hs-lE}u!Mv>)Z8vqb( zCs%A#vc$-yr>CuF>ne0_Pk?dF({Oz=v%IX4e~MgcO3|XAn0m#vGzN(V(ff92felGv z|Jg)+z+ssB(!(tP+Q6xaVr7QnpS5|u9bLWI(OWF34I+7o!QFJf?l<{j%ca56s;(8P zRnE=>M|Ym-iG!`V&aN5YF<`4OQPE4^STlMMMxxdRirfyhphjZ51^-=3GH8pu-+Uo= zmn`u_pI$gj_Gq`a)kS{B^Zs!HbA8=So$wO*38ZewSY=+1J;#wQl0WOov5f znO`rjy}SaoyZ3H}aPnIKs%hK!KPi1eM_f&2N@+(0FFH;_&rV)BFDy zqus}jGjnxYupy}{V~VFCyqGaG>>f=Jl|ZJTbmK%)QqX*};c0p+r!JbrZBtcxF+t{O zk6LXEfm1Jcgz;V;s8}<%ASj8m-$?#r_~a8bfvqfam(CpJG)@E`S`7WQ=?L=RY8ns} z>nuGm(}$vHXvnO?N_|Uy++3+lo{PWa5{B!>IBR79xMIL8ocdVR;hT@`-9sa*@*VeP zy8h#vklT*xTQW68ZN^I`G(tTA&FG57#PDRU82j~E!G`2Be2?W2Om5xYkb$3Vq$AZf z+`#pDWPjZ;_Q3(-u@+E9}3JV2ypv2%lukDvHU<=b64grgRv zT5eR=J!ps@ivqd>ghG%Ra6MwiRfiRn=D(IcCc|gDJQe;$vzxVwow?BSfr!N6Mn>|P zRZ=WXGhHJkxRw>;Xd{1dHxd9>@@`+OIcLB*%c%!VTcxJM-l7O(NxZ20;IjSII% z$BMMvsR%#~h=ZM7&d4Y`t5e8h^S>q#Or8O|m}sEr!UVQ1{P<$^9Xp@~FBI*dg}ASo zLiUep^uQ;vP3(tBvB({Hg<@(6oNS;H@Le<*OKbn`Q-l5(tz?z?BztjviB96SGc)pT z9>KU7v()*1CDRSYw=9kucy)k{bdOmrN>N8Xu}o)+nKgYqGD)~sAe1e_oqa7rMKC;{ zgPF|vu>i87va?C5rpc!DjMAho0>VKRvnlZ(m4X4p_`m=?_bP$ z4_)tRPc?tAm#@=Fdb)84s%ObPdIAEPwZD|LUAE)hLIOPZ0tM-ohGZSIv-alhF0t%v zM+LHEJ4{SWA>E!*XiN-2W8m+fDLNHq;mvcNFO-!l{$W=*>Uf<w_7^7BbqLgw*9Z zoLLbem~k`g(#7e@LV=Ai_B>xN@D8&t#47s!98)Wr*>}+J?*Z8?n0`S5x&WXcM6(ax z%?Wt3<7|{jWGfk?}vb7%s4#xZZiertvu`S^-=nXQD)DMm%F3+{4d4Z$(4p3kolo&wBmyUq9YnMQ@Hk9rTf+_2Ow0u%Pr)pr29e2>f%1zy)G+ z>f}W8#QOf*)7>4(k^F2X{4iw8c?cY~*Ev=G0vXB+VLz4(Jdw}Pnt-n_@-Nr=y89;*q4 z0f~bB`tGZ0#zoD6OJw6Fh8r?!h>&^5%`)yGT9ip1&C90`^qKsBa>A1(oUESx!lWH{ z=gc%{o%bl4AlJ}nb8M{d=# zpn?Dy6@0Izg8Mp0qP*8)RdXoCmot)<2==}QBun3}?3n^c99{|%pZz=?@L_NI@iHnJ zn7Dx0+`zRYZDcY_tR6$(P)7j5)))ts4GnUfV=oOL^?0EWo;+?h?ggS?Xi#P0v$?{% zqkk`I*hcVFu8}V>K9lfE456D%*!PH#!M3jpgC`UT%}Eqz&0w_9 zXPW-Q-!SSVt*dQ@mD#Epc7s&{XztVCXl~amTX)v_znY6bIf7{%K?Oz9x4~kz6iF^1 zLqc&+PGsjU6HJyzO3qz?Fk=~nMV%^y9{W>uF8PvsK$*}9gds7&h4AhF0wtM*TKZ=h zOG%c+jxoYaob1!IiwfR)F{^0jihjRSt)<|z+ySVmx0LE#EE(9z&d~!mnDOEn%6PRB zB&R9%Y{YDvVRBV0#l9=h=?(KJUPyz;sY=9-tZfAor_8^rtBV03Zx?^vGC(8FOaAUx8Rt>S)(9&RqT!N{S&VWp&--muIy0hVEi!D8 z8u^2W_fqw5OSk+n#{#xi)A|FObI6JQdGPcN(Ue4|Zv2ynm`@O65?a6Qk3l(O+bBO4 zM}eYoGSwiTlB7e)(UBVr67xk`v?29f+dE(&NDz?vbwLn*_S2P^z(%wN)b^$dasLT` zQDl{ps?%gNK30XrfHYLt>unsb^2e%yOaZ!QIoO>3m^x-cwsH@v;|y-N%Qtl0;vzA} z)=;LSC9z2~9uB933DitCD?0RI=-=LW=!^94DPxmvKA*QGJ(5QD3Q&=u4Mzi**+)@r z2PXuh7s=HmkN+b8T9a&&p9^(La&l-`_o%}_@N6>i<286A-`O+W{bFdSZ#UPAc~R-b z2k}(zQ@oMa5oef}-()nMSBAI}#$$_RqyBBCdq!=NXG@|n1`mW~!NgqWgxl=h?0-FR zEA=tR&z9MvQRxcICGmuksExC_pO_zwY){85%neipqYoash=%Q(Fs=g3w0Nu!R8ON1 zR(e-JMYI>JP&fb;b+6Y#sU4R(!jW9s0hnKAhWg9u-mL(Ohn(4%YY}BH|9^Wr!#Pc&DlH1Ap-RTMwv7*ut3oI^<_>R=l~NT@dao+nDN41TE9k3>%AB(G_ANXXG{;7 z1`V$_{#Ty_CfM)pm(7P-Y_2ENT=J>t`6$tX2eT)Krn8?Ua47#i8n%rY)7m!3f4!RN zo0+=OV#z!x8lQpC$i^Y@xcBtHW^8ufPD|W%SNHP7kz7xuzDK^o?iEn&i;p|j|0dsh z{@~$5r)Wt1J>X23{i?;e5-bX(n0uiQS4}M;WIUWb^Dd0WL4lI1Q9JZ<3vTj~t7$HT zd(YXCx)aA55L?_{2A^^5hzz5k-2PRTm;DdM&C#ZTnpxBADUooPe11V~a`aH`c4V$# z$U!E)2-G5WMdhxXU4tK(qy(KOPxo<)|i6!w|@45(8EE&+^#BJr({8&#s$eGC6)u7~;C|N^K z60c>sXI=`(aXUO#x0N@HZVpznLMH$zaI`>l^aF%8A9Q-3y?og_`en_Z=YJnbhRF);cFmd{Nvp;} zh4ZM@JzaX(8#^fW@E-T)^K7*z7CflrQuhXYDf=>sNGVYJ(dR_Dlz{CUc9hur^td<{ zcZY8r$`%Dx*L!maGL5J_v$Nv? zO+GX{nZ}P^@a~z(%f{tP=bZFNbJ^9?G}eBmOALG)-X{mgeM3>jYAi7LNosO!EThaf zTo*4IkR%##5tpMh&Q8!hHfFQs_z`aHIN6q_J~wk?hoxt-1Sh*6^%yetsLPBajr|L+ zxob&bAr9+6)W6DXTd)>mg@fui{hysF)b0QD&!x&&-yC`5P7g>Kd)Hn$nEct2aj9pV z&NKS_rmx(zjjvaPQBf8m?*poI#XHgt4=+!4wstTN4-eTnGhd*q#up2oOugO%YYUgCP8K2B}G1Q{3e5_)0# zF`0XpGmkVF*&d9Dz~LyKid0;<`Po*GUe=K~fAepF=pU~1w#*7VAxpuD#8_j9!zX( z*huulkvS#2I~kN1{uQ{;Fg%|q2z$`kFQxn?GPuneae9dQ1yagkEKgc5_>3AwKGj;-+8T03Qk`7dGn2&u!H1dCz>22T90+AV{haLL zbZ?LDpHrC!UN^;i<8On!A)7SJsV#q#OvLjBaLNfx4qmhs>kqDI*JbLiw6`s?&2mE5 zd^5^&p48^gRx}+WiM;c8D%p>l*-yu*pQPwO3C?D$SNC+n8ls@P^RkTWL56nKz6NiM zc%@Wp1vbnotw)%|qhn{({zU;KU%E@J0b)dFof2+n^Ziv2l`twzfOCIM>ZDAWU9eOw zstaAPdX>}B!dfOxaBS<@=~bRyxi|kppXZ9kji zTTSqBfZ7_ne-{eCCn}IbOCT9BJ6V39yG6Fq#J4^FrO!1O$%cQbf|~9bePJds7LWer zJ!@ZYXRuCVB~UX%wBgA&?8gHNdZ0FzzJu!CjAHGTDmoYTDead5f2roe)BoOAKgsvE zLct$A95yjKvXt5NH?tbnEdW-vP-oY})&9PLf$GaYVwKGu$`Clx#Q_9nx=JLQa{o3K zku$mHt{yXnrC5d%`AhPKtwS<%cU>2Nd3l(e7|QfWFjL1sUu@qa@J6wA3mfslMnivy zmQEn>b|f>W1|#*9p?pL1=t|MO(793ziBhn#sh|X0jo36kSKD}S90L{zusUFdIbk~( ze^3sX7738pd4sktA9C=)^)pM#@LMoGu6H*OsQh>V=SM)vivoIM3~&E#m=67KbKsOP zJnrD95RoNEOh$Z{N;z5w{5FfbfSjF3g2ep|SMK3+m8#ZEREcEUop|g9Lo*na8H|qL z$>tkaV!VPw(x-L#DYUBMtMsGEj*bD>N8cWM$_-^pRsaPM0(5{<{Z?XydEM4jIE`Co zVV2DLPuo2p+WYsfJ4jJU0jT)?e|?rsbVwur8j7eQ&g!>1xu2EK zbpon%`28W}e}s=}L|gx$d>*3$d~HzbN16XfKJREa8oT9Rr>Y~KOk;$<8aOXre6_8e zQSN1PvVBYT(s8!1qyHV~S_^g9_5y?a`=1P4BWuXdLHqUq3>h%n9RGeIB91^le_%p6 z1=KATxBeTL59C>n$&|^+fPD0b$1S+fDnE&tl9Ho#K@T37q@*U%Qtu{NH8xv7DS|%c zH=`qhmFpOXP&$n<#d0vilzg$}9eBOCdHoy4e#TOx(utuo`=YzJ%;U=aCj%xkyMS!f zeE730lmkrE*{h~@L?*KdN&O%`N6G>u0DN2rRzw?JDGb&!DxtYz>FZZ9Fz)}i&Q(%c**fkHcl;0Ye!?tXpYHPlfi5fSw8Y|8$ zhhkO55KbknkLMm4Zx*G#a8Y9guuWML_3PRqLsw*21nOv_S{4;WvC7qe%r~i>^X!ad z%Z{LNtV7i{a0xGQR(bYpKRb7KaoABWtqIJ4s@p1U1o^ozLf><01TVbzwGR^hjrjbP zg{aRW#GyWVcqLXnYR?`+tH4W+)kJ~05uPd&?YIPdH#GWTnk+u$FP6Xmtk5!%bLhk} z`J-vFqvw?qTIPG~YG^O_-#b7Cu7Jbc4SdFJjh=o%YAJa=w{=v~pY)S6@=qfvQr%y- z$6Ii7;Qv>vEV9|sP*91P_%C>qAAYL(sd>P<$$;U>z~IZO@Z^;HsHRJ$`mvaap*d)J z@zVkAtjYG7Ws)0wCL3-t-^K=f_~vCbXoeHRvg|UO$!YMb`7|4>UXrN^kzT2lpHqrB zGn6(RKIdpiwMLG0Z~+E_D|yjCc?)_JAcUepqG(`by*ZeScMo$CxThR|?0FfOL0tEq z{NyNG1pc5Ey>|gz2>I4fD{T7x@bj>g_mTdp!O=Z965 ze2*iml44JGKr29=H6k*yePblwW!`6f33S$b!OFzWH~S)Xh328Q6L?^*VNbbfr!#N^ z(QozwR9-%~kLCr&?{P*Hr2Os($9JHh+hJb-%%txn5*G#%+>^S0Cysa=bfRG_zf5{_ zyP??2u_zIrxdPuTGq|bcJIZW@1_`|f<~B{&2*$4|qI>FY^Nfya8BdH#+Aeru<3dZE zTE)(%;$EKpf?^*YH91iJcQL1-p}>GiEG1x1((j=banNJQ?G z6IEzfMD#I0I!py_a#5thG=tXf_WHDh*PQLHge0$?)QbBo3j+YCODWd!I{I*3t+Z)7 zCWeFQ_vIZf$-47jRfA1CgR~UU@ioYhevZfvi#IsN4hO5CYT;=3T;xFzW->s+tFxL1 zS@bzy*lwZ1E-iyu=ALaNbmYX=HVuV5UyyLVE2Vr66Bxr2YFyR3QG{1~9Df+GwUM>R zwylZK)r4g;+a_<@TU`!y^!tegq^ma?ORs~b0AY$+c?Huxc!jE$qeIJuhd=e)r>2f7 zEPTWPU0zusrl7DYV(hdbx9t-uBZX8L0qbEqQBCauK=n9gl9YqVG&5>qGwIHdS>UQ_ z#lf$;)93c1Gmf40wZh#=dYdoOi($qw0G;)7U^-1*LuU3CNnRn=5#0pd% zyLEP25|U<;uqj*wM`nSe%|c6aO(1e9N*q=JoNBLx`gaEbw{61jUVn5g$!!^l%MF<7-}NZuc{Nk<3e>RhHl9D&D`s4oK_~W*PEsBxAjnTM z%7v~ttn{=cFv=jHGLk2pk?f1wf$Knj7UH{GZ?y&TnKq0J9zG@0|Fq}NL`6j(J$_t) z3yq@_M{bQ|qyppO6D-l%1A_VL71AsKm_RT(+p4d;Bfa%DVEYF zhFWMh($G1aocOcdPtt3bbmu2FMYj&VTNTazGn-CFCGUw^;-QjC`qz#v9*ixQi0u55 z0O5snd*321XCdUD`T`lJ28pp6ipDo#rI6>s2XW^OoGe#?y)`@TM@yG)5i1_`X?u=R zzP}PO4J^$J^1tP0qHN7w2KLtatCTyo)J&Z~p+XBs*d##VN`^G>+2XBCqCz}RNK1=+ z7;uIPOkzp0{SN!ZF0<_Co2CJFc+whpdow8*-rok=qqNG{ z?6;5^V#K?UNLlN9%KFd5J~)%>+1q{FlQSg#VC}u1G_%~@c6DCroMdQc*nmf#OGr@^;3Q4ozp0w3;ATg8Ts|e*y98Ik z8=VX@;{=r_$c0X<4>*UB{G%jv##j&rdJEBbG`?$`YPate;2z_FmegCfL$0H*n&z?WmoNz|zkVBF!jukV8 z81&n#YA;{b2JUUzS;kHAHR%jeh997bZ`;w6ELTLAI7dgGsYrBC+0Smv+gAusGJX5< zWg1POG;!#rNh3;M<}Zs{?J`IGnCU6d{s&gbQdj#=Id)TdR}}&us!a0fc3RL`k?~ga$mFyNN0lG8 z5&_}-WN4Xwz#zPE@=^KXTfk%uJq$QGdoTmw1ZqBwLBVBAtJ1{zrvRzzJ3Woq>hu_; zMv`KsMDWN1H0iw`t%{JA8EUfwd@19?bN-lVrX!EYdXKqpd9G(>VmQ^CcGi`~I!@Ghr0uX^U5jp+xqT8J@S}9-K_3r*zRYzzD@4!@&Gt zHtuuKy25j@^zyd$Pa^AF(d?o_v0J}b*o}=<*foWnn}U|J+Cs`CZ|;48S!r}(JKfWW%;7`ApsPx*M#r!GU*&NpYgJd~C84sAxJrDQU{&dv=*`fwe^^QfPbSYL?f1L4)!}9$YdSDQr^T}``3qmf z;~Lr zXsViep%`G|bmdm`X5*&|m=gY0iJEYn_loYot2ee^&h`^NcU>Df9^j|UwnP!32mI1i zMgiwE2%URG)p-4O`*%6M>m4z9Vt;oz>J;qkU} zgl7*qnE{{Z#~O=7HI0fXy{F_t(ot5J{UMAkwdeE;$9Qa!P7b!00&zow1}JkxQ!-k~ z;mKr>sZ<8DjM*3}0sC+a1u}%;Cmv%*p79FT5Z&%5Zma{Fw7x}5DZIQyW~G_UKA=F( z*9z2?I(u;eWT)Fh{V@KSxBIs;xm~s|AQ#rB-XI!Atm?cr&FisZl}mgMl67kmjM<@q zLLAOBY7;VkxS|$QFs^i4{s%&e(E0SYz`FohI>aF2Jx7hfBo*0djI&WaHx5k?& zIa6zwSxgyQFV@L#a%|4K7HHO+_^7ml-N}zJlJhtne_sH1cBK_C;G?CsaH~#{lO@lx zFnWkU!>5j(lf_Jy`1Ygg%6jJHW9LP^@~Y7E;FE#v{o#eYHghEILh_-zBr-A*+-E9z zK(YI*?vi@eZ%?%u>neB)DVoc`Hc?SB)HdE@URUf`7Iu1XsvYp+NFa_+vMjdTISyWG z>*hc?6i*{eh_s!{dTjXW?$e&i76$$gn&H#n6(8zgt#hELP=6o4tAB+oe4#kQUBuPP{~YQNdGo z(-vN48s*X=q+7zlNbbGAjGg@XaFfVjb>iV!Uy=0wfzNRjj*S)6G9~X3iH`WZUj#R5 zo<0iK3Howl9gfl2J#zyiXDeA$^M;GmH+7D5z-g&^%?-VbNa8-^tH+#uqgYc!yIgI< z;pQ#S``eouJq4Z#uftxaUk&8eFnQ1hdNpkf;?Ju*KM0{~tc7AP;BS-Xhcn>|Q)Y}L zB~G$?xDbDgYyt_&_>JH8f8vWl=uppce=xLevLbBnrLC0b>T9x_1)crRposDGyhC~O zqQP4d@75$F9TNs5s0IKVIxBSM-#;Kz(-bJD`l8s!mtM*{I5-?3Q`{#QP9wEW(;U@8*C)mfB()IU{3JM zk!25qksa5ANPIuFAWx8~hH6I;H~A{_={_=v2Ce9YT_KSB+a>lBV#C%&H$I2jcYH5R zqN;YuEy4@u<%ue12*wqhKU`l7Xed@s)y6i)&Uv|z;$@q-hOhFoOhW-CmKOkgC%_vX z?c@Nis^~i(3f0(auDt9QS&m_C7$z$@GP{aLs(-yoe>nTw`o&&Z{316wv{>wYXlka# z(Ann^&K=hfM`pVsSl)?2vtI17FWA_g%8>urluV*AK;-=nD4=17TSI+cW{(?xm&b6_Gy$ z$<9a?0Urxe;T=P6ig28L*{;YDbWuLGIMfSVDXdHYwa2ep(Vsh6JZho;h%4>iIClBa zFD3#BH3Ra0AbM7B%vJ0dzDwW#a&C7oaoNn|+|^`+X(oM)js7zU7~mo!$0gc1{3Go3 z2OFF6qxlctaJUP*Q%zPf`5^xE>^gzNdfN2YYBeS)5iY%$s|b z@6Q^jMabU-=*Qwhq_Vq&3~z_$&qyoyt8e_BqQ*MNVn7%{t3A;-ZRfHRoGP@|qL)R@ zq~d;kBVO+MMfZ9BcTu^-IlB@)AGUGc@Nl5MYvXA}A%TeeN;=+Z4rO{|!e0#XE}`#2 zz!yq92-N*gYqjE$@&1HrC^iwacjLdxogNn2csDJ5|AFGg({&Xr7;KSK0;c+z8JEN{ zvt6#$+N{!Z!q%k z08B7L=o6k>vRZl+(cYE0He)Nz7`#e|O^OV{nN|E!rheHs3;LD>chuVnST(=|qw*AE zgNT`fW>SxvEbr}~AlmBJ(cp&Iv%Wkk0|c{gIh>cm+Z<}^MNvtuna(~7^n*;!VcmAj zg(ZPXz>)x2jPv~ke)HFrTX6!dY3_G?j>B&t&DfIs*L6Yd0On3*iG^QfiiuHY<+kRj z)c$)Y7J7A(0laQw^`gmPX$4QQ_rmmo7J%dN?CBdg?YejL%(qcor~`>DXcs~C5}tlz zQ%N-`5!F%cdoDL$nVyj0v;h1%OywIi{=93Tj-VqW3pZ+hJB>F8WN-9rIewKjLR6A( zt5sVj-g|txy98s^Q%&Wk?5*zgPqFvNZ{r%O{H34bMqgeiKZzlGm)R&}6-Q#;-(Tuh zVyDg4XNwoW1A$B3P7Nj|r?~C3fq;Aju@}%Sy;`v>*i#k;3 zmNV62U2h(%s&m$99R1lgZrIP10!BIQB%h<%SwUNIay3FaJ@N2l&e5w)4s8aXA;~uL zc+{5?s1o}qNLdzeg+E$v+!fLgKYK&0yy4aQ<`;Q}x;Y!AEzvg@g@wKI^ihkBOI^FQEn(BH&ELP} zrZVzjUYhr0FMFL@G3otD&yHR?tqbuQH@f7Lm^-!nY1ljyh*~s3m?^>dS7tZgtYOxH zCKM-7?M3XL-%kD9Gf({e<8{%t`PL8TB!Nj0=?rBk!20`VZWn+8&r^t&izSWspI$bB z)Cx82D5$AMCJndY|7bePfGD3XinDZgm*j%dozko#A)wMohjd6uE&@_ZsYptSMW;)b zbc-yFgn-f_CH2n2|NY3PhuxiL=FYw6{0=7*7NQ=j!nFy=rNd3!e(Ar@OEZg0^fPy^ z5qTgdL4(H1BiF)Cckn>i0r$*@BVa`Og?zz^*_Gg919_GDkp=XNOU#gzb+cYXXfH}U zsCRDOyt(={1p(mWrV{iN6js>&y^6K80w$Bcdth3qYoq9IJ$=gvqtcl9I4yBHh?3Fp3!Iac4nMN!-v$Z@{khMN#sQJN8`sW1qUo*Hn_Wed~$EA3?BRIW$Asm*+!eS zvbMI2HbQ!Tx(|9=%%H}caPll`I&4R$WDBf$oJwxnNKJv(3OMzz7D7hYoU>^W(sala z+$#^V9X+V&8Umc-XpxxrWZ}SsS9>}BOrm#n&@D=`jJ=7G_f<5(d<1`w&2 zF#LLLVfZNF>NQm+TK?F>#rCmf3KxUGd@gvD&5BY$WR&}R6fp#sL3^$U;a84pCGm4} z3iLC!L4Q$=S#EBk^uRYd7jEth#LvRpEter^5?!qv!rWy%AQNG|b{AvbZMvF(Ny@Fd z4($9BqlI@_#jYOr{e3k#Zy#=~cUG5;HA1Y0GhUjUgjA57>Ph1}Dvj=hp~D=`MCW>s zl=|%h6#Q#k6AMjihIG_6h%mSj{4jup_g12XrkqU|F<&#Fv%@FR)>~EpYccJ;o8|$M z9yfN;3}A%Rd@G34jz9y&_GMoEL)Z9g*JB)~R%O&4)WkhFDQ1F9oTK=SpOUy5Z!7p= zTNFKVjdiA9j92er@zEG-^v2!`1-3lnSdY{&d{VT@2~-hFz$n<+8AF9wcei^Sc%_*D z6G^li=XZ|!OVuc1WwZ$mAlaU7PL6J`A+NW>437&F|Kd?!^RNgvon?_;TK$eGdOaTD znABG*_9X|Rl-ZFl!4*Ep3o#P@r(x&TLNaXT-n2&LzVG`Z!O!r%4bfxo`T5I$BJc{!os8Fof_(39&3+-Y)<8($awGhymJbqeBiYrCJ| z^`1=em!geTo}jT`03`3V9)HlGR*|o~pjTte4^XR7B^A2dK?re`%K&tqze zv_BgtYrrS-amblAw#Z3XhYXjCB}`NfS5|xyRdlSS${j^{c)CNh9Z3R4l<|!DdeIX9 z-AzyV^rYR{kEn+-#p;IR$hfEq?ie|KP^ePI-){Zn*74g)GCU z)LMD_k1olkS+L}1eEbI*UVFziFk#Y(uaHF4)d+aCx6g)cc|oaPN5vpR&YeO8vpy(( z82kbPx0yZp%ML8*m`ZL0#LqeX%qy~aZwiCqu8OJhqkm%dGahe9(jT!T1v=jGq~$1L zpV{`v);AWIpg7B})85fTbhP)b*s`&=%`DNogRc9*Lvkn8wtiK?XCIYMB>nwG7N_rHgB776 zPAELB;`K#buEV4T)1(fNZhwbEurLy)tlLXOpZ><)th+JNh^$k^#-V_ji&=FER*!j2 znbT^IK}B!hJ_m~X*@X{!oEs+X%UADXJ6WIB!0Vi)SZ=X3Cgt5X-~2Dvt*J<2e)hv- zsOE~Dp)=Au^|f-7AC~L=-giXQb&xSi;r{#k@dWA7ARA(yZpDG9WdjG;gh3}>@W~UV zc5Eg*d$HcGPR6e&(1`-r5Rb5Bmj_zuZz2d1zL+AcSSJtxCWP@E4I2*;z<@#-@1^{xHx zqF|)e%QgKU&ELK3`C7^2rfKMWP}TH*kIUn>?J|4wa2fRu=U%CVu2Bvw^Ad+BOu~7# zSi(0-1pbJwK1QP!IMCQtZ&^rd^j}hNX5rFE#93(0mzv8x-IyQ&!=JBNGTx0%3P-Na%HAze&p6>$7`t*0XyaY?0X&{!bnWQ9|d?u9lgDuKoXi!6WA71VCF{ zKgp4ap&bQ=vv6{pX%dpCp6avras=iYCXDGr?HX>?R>efPB$$FEWUR*FE9D!a{YwSVviZ_XQtC52jVSkWAkst z=^I)NE0jN`o#8f#IEM=h#>%W-(HkDiA%wygS(H|7cvQ9^VtdL&y0?&lLEc{&mdNTl zclUA7o_ZD-Cw^IsV)7VOr&|Pajc%Yi9&pZAM0*#8uV9Siuh|^9%k4gm`cEl2a*=;) z=gZPM)FCGUh@UWxmti%|=J^P!iYOBA{n8D}1V-Qke?JpNR`$gsWqRjqA6_-&r6KdA zA)^~85VzRLL_VFOb30j_waWG4MLtUQ6px^T9MG;Gz!6ghz9kpRgf$VQAG1*HMl)W}2GbGze;ohS6VDtld=KI97T%;oYpy zW46re7{5+!)inegn*VgUvw}^Z1MCkv zuc5RqcwP5SGSU8`ED0rl<`Mtgm`Uxw%%8lfqSQ19%AZgP`+TY*v#MeHs1&~@E&Ik9iX<1khAXW7T;E+jjoMTsm zKDv|4kiE?q!UhA~_m`8s#`bPv4;lB1s)Utt0RcISt@V8>53*gbEN~_Y1dNvmc|Pv> z=YVO6o+qplMMvD9t`xW-@FRvNOl}9;4TRq$UK%BVi~+Hz~Y)v zv%Bk&>A`nD-uPpxfa$c1Yq49Q4_6%tljIK||L6cMm(}d+%+p#TyvN_(D28(C8W3Mq z@nk@G`Z*34OV!SR05D`+*?$cVubAW_%B&4p8h-H-7Zg0Y+rXzJ&^;s}n<#8-(Ci^P z)JX2Q2eP}Q7D4;21C)JkQqZ|$Y2WH5JL<;&lv2JJs_)JovboA8H4x&*K~nx9kPJk8(IrHDqx~iTk%{6T*pJQ7Yc*)C2Zw^3t$(UN~6^=b4wE7cCS9|Acwn~wtNjk^o;Y&iEj(FioR&JcqFnw(P z&Bq&)qcqxHMaDsmeoC8vKD@g^59iO{^sGYh7E4|b4+LVP!)gwR*a8`JO*XUFls_jQ zEI^2IMAsCsi!ZQw`v4o^Uw8m$jfDYQ&1*jRtl3T&xpMe$w%XTg51t&mH+%UuljXhz zY?uZjQXMs`L;Sjq6@;U-j!DJSo(*Q1uCtII5D-DXjm4`;>CfQ_(7#N{A}NUf7eUrU zef7ov&K$rIwCY+!ssB>)Ohpdjq%tH+?ECW62T`v<;vQqH&J(S;i&C`6Bu;QPzBaUN zsa)HWb07dHjYDYIIyO$ zQaTyuo>_6K(NHpe{{fZ)ffWEt|9&H5AC=}n0GVdNdMaOH&8MNqV{4D;y@t8Jn*Od= zw4dUa3;Wfk9zm{5+Hi`rW74j(!c20gHfe)-H%!dBK~kGZ2ax?iO;dqkdDYlPe?C!A z95H0qO>_VFlRnrhg7-+;IA`Qf=%I_{7#EFBZ1#f{ z+#c-xS4jG2xCE@|O3Yx<|Dgs-+Z_h=V5N#0Cz{=UAwgtn7V*;!+`rqd zsYOV?6mn&ZgDmlE^e5?56rrG0(e$QBQDEsYv8^_TCN)euHq61B>oQFXX7a7!Sv9SV zFU<83Ra1^E$N)-m3R12^oM&cM5uNI*v(S05xcVru>Ni4uy)`}?@#2X9C&PkJqs!;G zrBzNtOS=SGh+m0>8pAF@_RXW|`QRgS^o(;=U4{;*h46`p_TOH`=~<)R6&F+Mao==& zSkG+54GU9eqzwn5=Zv5AVVX zD|g&6Ru2{qRS9@Ayo&hkf zv9v~94-65{yhHr=7gn{~RyJEF;KPI?dMnUsS2VrwbuBQT(gUP}to=~duiKxmzGp#F zJkzyeWwi^hPV1^sV=E=IJ8U0wHmv#XKP3WYqyw!ldl|JtUU4E63UYHQPIK`}50${t z?O8C|RTU3=yqV$*m{7L^Wu1G#R6>pYHwCmKHZE=tnBcDB$-a0lf2s}!-ReW%Ut{lT zJ@h#-t~L)dkp~ZdLB(&G!^@Y5->*^DXAU|_o7|3qVf11FuZpxE<@~TX2Zv1!6}7+- zr2a8RwQyZ>(X^Oz(c zLGc?`VlyzTVs+`>>d;4@IGev;t+3nnb4Az+su(*`_qkLS^N~An9RDGxS_-l}k!wAB zF^ED6w!!@S16=l5alD$=EyiU`Yrp)4O{-RF@CuUBO9J_HtDJH)MqX|%256@-f*c4^ zbR5E?X>^oB<8^#ieX!1Tl3_*v1B2MgGek7MQOuW05`2<8-gSI&R(v&=2h!=|3*4jY zTJ6HO2DV_!>`GO7_Bo&Se-k+&2ebf;MJf;x`BLkX&kpI|^|n2bvI<1Qk$t?t1_VU7 zIBrb`2D6|P2ajz!dK`G#ibZCm><7YB`t+~-cMXC#A6rvD4v`j0Guh>h@@g`!)Q|Vs zL}X|ekLD;kV9n@;k>1I`zr^I3_=CWl+`TQl#>zIx3(;E}q4cUThC!zR!eP$OZ~n|n zjSh}0H7n)Pdf=}D-U!kdjv+1Fd!mF>yR=(P{VPy z?q41+lbxC{i&#elo5!|t$jLE*_-`hYT9+7*Boh#*&Q=kyZy#8Jjw@Ins$`|`C^o6- zd#npwRlETCzr00rpaUW*mWP$*Ws7IZ8EztD)@6TNb8*!j=y)6^8(o%YzO_p)^VP@GLfB^Qy!N^8~s6-Cw`a3G33%7;*<{jaD?DBl}0xb&3s-53`?*% z{_4pI1Tx)Af0k1s33fTuanRX}7ifZ^qIHO=3pW^9$9&+4PtAXDn$sNvJr{4*yHw;z zW6VW+(W9#54D)#3=Q>Je8;h_4?m!*}!m$C41f0|+5|@87HTB{PGuc%aO#9vZ!|T#i z{l2pOj`a@)P|t2mh|7)nL`)=Eg#0RIHIG|c*#ZRWrTx6clD@<=J z6|3Va;8x$pjE=?zv>5eaq+8t?^$~OaM4lY$?Gv)w)Y6;6mb;b?6gSn_R9Ikp$LLNf zxLthw({(c4d@>7y!|ZjruZcfr;R&3-s^UX5?}%{dl222=BWSAAD0PKMUSRY4w{xS~ z5h2di< z8aI0H%>P&5s zdet9|Hm4mjEkJ~Ro71C4QJt{&@&dnBZyC|Icz+|iPE~>tCq;*?;qHj;fci4Bk%ORu zjiIMU-clK^O@ZKQ;~cpKz%A;?@Loalcx?BobMJ^rif6soz9L#?$3a)qc5Cya^xFo4 z9bIt~C1ki!@;Jo7UHBE}z=@HkM+O4UTHS=`B|KQk*%R9{qWYC*$Vra6y(s#n9>@QjC={Wogyo4LJb^!v3GwH5r<&XUoeiGTki*r z`AU$yd9L{?8FKcAXm`q?HH914AtattUI$M}kjQ=eBbk98!dh;3D@lQhcUqF~L35{P z;KQV_Q@qzQcOD`{ZS7%JKl_aMO{ilxT>xP%;7>p0^|e0QJUafQbu~t5)f$L7W_~?? zgQ_do;-G>@2r4ayC_VLdU-Pw85vWLgb%?=ku&&toX>8OEp)r}-_HTB3` z(>hUf3Fq`Po9lXY;=A1=4BMnps;9sOfI0fZx z05gh2jPiD`oJO=$r6$*8+DE~gHCjiT5^`#$tKnGs2S@israqE zb5~Dm{L6&N?0b%~Xf4uahuZV2SJ`vZOMIfdiBsMc&Pa%SDN=p@ zhTTjRMA?L%Et+W%67~TZd~ERHP1WuMV8R+FS3Zv^2hqmp(jQH#EVpdKY!DI@B&DFB z05m5Kor9=H(uHivb8!v&<=P|YIO)u5hI&f8~I5=+gO+}V} zs_U=NpVlLYp@*?_fdG}E6HAgbk0R$lsp;(HHh(v~uuOO+RzgIz02 zYVnh4UV+`D!Z`^^4%9!NSU4%;kAGu|GpyQ|$KXQb#!c#?V#q+FJS>^Bf8^tXiWr(W zz1rtzjEQ=kKUXlo(2NR-VG&Al_V;Dxa#y;Wd~?AXsIJq1hi||6A7f7NVR}HKcI!a8 zm@DKb07;>FI_SzA2|~-Ku72uAz*|)vsUMci+&k-)!c8DHehYqE>RvUWf30dw|FC{ zq=M;i@(k}**qoD1HzrRGG00y=OB5jZLQxwAkm zyo<(Ryh#CtII0olrk0}+OwL+5Y=&Mi52@-?T+(r0(SRBNw#<-?aw+7?rzXLSieQob z(%#$l9IjW;7hlV#^6;VFyExY1!xY7{R%Wd7Af%sC^cHzF0XlNRYNFT}or|a)4n$4v z{y;3h2WFz_Qr%>CH8e!ri$6H|mFxBFodn0!j*MKM-+LWIq=nYsx|>%kTuIwT8NZ8E zl!UE7l=kP~UC@-W&L7Stdgtx8)C{=1=1oqL8>@+}tOcOZ&dvaKg9ul5gb_t=-qcPD zS>rCIuv^xKrD^Q|f$;oL7D3P-g56o~))zNqKc{4KZ>s(LuyK?)|Na2)rSrIZb+kN) z_Ver@k+NLrNv?JD0=`D{sK6Cy)cHPlBup4x`9d~1I;$n8hg{gLU<)MTGvq#CQj=yo z<0_RVO<@^2IF#ZHY-&XfKQMp%>4(d01I{V=mo3R3W14e^bAC?^c3IgXKoZZb*8_8s z{H0BYCP6@PGzo5BCFVUqV_gAcA8G*gOV}-9(`B@9K^j!I6#Czl0}<-!~Q*UD1vh(WfAU3k?f!2F!70qr2rGEeiO$$j<*Ar~;#d z-lxz1Bw_<<+|NON#0Wh3GgJ~nC+pn?6x zM^snQpdorvjUXcQ`0uy-uA3o1U7H64XRE;t?aiba;!mtmIXQXX?SS8T641BNmoQ`V z?jzY{y`e2uFro-%iEhmY5Q7$wsq_69w82jm32&-fwr+5s1Co3w^9z#jT`q{-8nuMS)X&&@k9aegN0~|KtV!}5X*Dw^q% zHZn32tU(OFEs4%f_hcBhRC{|QBA%Q`JyTK!sxzS0pNh1{%j#{`Ye9=i2aOpoGfeWY(Q-I zhnM|5_3XBU8gNoOIRpDS6@YmGxx(*8&IIZ9*cBjL@=yFg-|TkWyt*1V=#l+#CJpGNRzF52LGzYe`ftBYTy*&f{I8*Ct%BQYwZ8~roR#l_Q%j$cE{5mW#6lvZZ zBm)@WZgIeN;DhIrenP)*re-8fKy{;%H^q6mnqCnF;uoxlc<#j0!;g=TOPo736Zo?V zSyng`zs$J)aW`Gu_gvXY6KVZ!*dhOBBVNqv-3N}dc+&>o5&;g&tX9~E>1UU_O?^hm zlKU?8`r##%XnQo!jY)0BOF+ja?Yj7R)fxEMZYBUyeC*5n3RBQ7-%ine!E@j|EiNDG z=_%*P+_k3v?0`-x${blPmWr)GTNR1y(U^<1=+2tQ5DU+(xzmzuP#9dZM(=CY36#W__yT8^KTgQ@|v*U)?uJz>(GR1YEek>(i zw1zh2_*uEOAFZL`LN;GPI~4&)rtNYZzr*8%nkR?MHpV}`6iBmugVQi^ER_(M)c8TY z$q%T%ric>d!f95)G^mO|j#k6VmBGQQV94eVnycG^Mob@M4Yx(mSqEh()8%a3$dSw~ zwQ1})J!9V8LNR_u{-FW8@(=K39ol;;O0ZnI zJwZ6(7tE>X#r7nby^2!HE##+sH;yrHwcmdf?o+k(!*56e{kG8yeV8n8^RP0G9(AHH zsuyqTd`@7nR^yuK{M1}BLuy#ec~8-NE$$b+(3d3xh30{*4pS7Q8|lo@Gmt>aJl%4s2@Gwx$)bzlx!KBw&{D<=?jn}CH5C=v|R#qv-$;*FO_pFZE6+^ z5*3_+{{@c-rf_V50;U`7`7L?u8?}S#G+oHwJp|Om^6#&TpEiPRbTjj{D}tEg3~nyb+#6G2)P#aT>^uZNCFJE_z|h)`_DmeBv*xKs|RS;;n+&u?EG)5qu`rbtV7}3SwkH}gUj`x;b z-#^qgkd5v01Z0Y7plL_c68eU&y4Bu_)6d-Y}z1j zgdactyk(8Qs2DW7+YUXDDl>O=&V*uOJ?^lke3j$f++vzv|3%Hj$di%7n{K*==j)=j zF6G?NW+Auu<*K1Vt!+4h^Etf`XozfeB*nKZWigvv+_BtS!Yxo_=whTgS7-3QY0=GmQVh z^!wSJvlpq<0QWI$P!?h^D$)i;`&=M}LTz0KhHtEXw2>M3#*4NYTr}}v;sqSwDg60e zhSL`d*AB%rElN{_GUxzq6pl%vh+gw$=uYtCbN-%>F(=B_-MaXLX^{P?pSP%PM#%bN zAI{J4TRB_MtXx_~4`7p0xy1kK^8-FsqQ=wGDvG`Cw(y0Qs+ISpArsAN4_QoNJc%&X6j>Ye1pJ9j0j#&;B~@zZG}yV^J#U zisPF-WFGy#<*~QHpeBKN><}S=NM5!8Vcw;y2vm#MW*&s`=uAKCbMvXhp!9aq+cmt0@b{1VhHVJrH$+Q$QBRE7uVVQ z%LxjjN(jk|( znV&q#gBB5~pB`G#OWY!!YuFb*p;5%$sRD~Iq55tb>dm0# z0C`0T3Fd%+MJVaX+R-X@X1;)%JRn zz;9%GjA~JIC_Bk9<8+aB84Yhk))k}(S9BA07jn73Ah&*BPMHPxlLj;249ESOxiozy zYk41-2=3l&fti%Gs%>yY;7i{lG8?u1$NYYa+#Xv+3Z#zh+Z~wo{7pAJ9eHj0)>wg7 zXRJ2IB%zc`yk)7x5Vc44MXT*${1tS{=4Yg0F@o#cr_~LD1#x5?4q9 ztyOs3qJ___>#v%HhPulbQInXFO0|Ssm4~dpr4_thCx3oU^u-@XhzOWHsp;3Lop8BR zf-6Tk<0Ei}ZbRr1?d}9@k~fd>z>*Lzp#l1=-mXx<>?$$9e2i5VNAW4?iA)g^@^%1~E~A#7ub|Z01cP zF{?ZpMUd4jg89R!{S$uqAE1L|4u1OcrJaCj{r=1?A-)I;praDgmd;#kxjZk+Mt%)w z%F^gyAe=?jI2+3G>7hGX`6{Axn3*9sg!+hDKHuIN+Ea*UjpDH3!yB|ZAxA9bR?js( zM)e{hUp74p;CTq4wfM8=%b$5;qz0@4tx5ulYB1A}@W5CHy&Q*^j%b^-xeFSpf#oX% zC<>e49yBq0)pA?1pk4#^czD*&a@EoTt-rJW7)j#`*o@x?hN5Mk$CM;bsZy$(7{hZ(OG@5!fi? zMDuN7|8l-(g%h6u>aa6SMDU3LLx#BL+P(vMI5t>?9v)ULA?F@w$|+jc1P}r)DG2NM zUJ265D^B;M;HD*?$Wolbc`|g`t)eVuy}>X`g91YE(xRAuHqhPB≫ebiyGiKKm^kycDt<3e0}9o-eeEu38+R7T5)MvKjWDXx9p< z+@`SjyOt{#RcCFvhgZKXn{$k2tvVM5LfcZuffAV_PQR|wp8xgj?d(z}r_{3gUI~@V z-u)qO0=de9c8$u@?bay=!}(q{Y(?eZ$Y^ty#z5Q#S8)SCu4vZ z@u@pdqDWz664XsY1&4f!%%4(tbda<@GK9P_Uv+bD9OYhc)V>LUs(rVz;(F<_HB)!z z^5R^?f7cQ{?4G_OilhZF8&Ui5t0&cR4`458G{INeV5s!gtlpZtC$)H@>guie>a9y( zb-tYt^8C_iufJ6vGtL#Hkg#vZ!z95H2YrGi^Gw`hL|JgJ>wGP1+5Xv`-bW8#sS;cZ z>GiYR^N*hH=*Xlw#wB48+49z>@X^rYgC#wn4WL1b>F_O%uX}ceW9LC@$b*~J_N4nM&=(qgRfefD&r#Dk+5OA0KYzp zx0aX(UGppy(XU=%^ru1=g0PAolYopH=6j&oOZ6LnD6oKpYt7$#g6ZHLISu>N6*|S7 z3HnWRwE^37`YD(fOc{1lG=88ohDm*swEbi(ODJOE0}#FPCwBe3SL1R@yD8*GT)oODK|jN!{-7FCpke$Fn)9B zHP<+P#IX)+;XrsPH^wgoU)+)LrH$=Lv2p84Z)+5Zn?-{!dZR{k+h&r(6Rsy~lQouJ z0)gpOB#W*_#T1t2VO`I|Iv%f?r5Q093I0h3 zobr?VY;nC3_h-1tFd2`;jyf8t)Y;m%L86Dt^orSC>HC4hx8ZrJU`Fc5_WzZF^-tTQ z$LG)sg`~B{He?10Zzgt_>D`n2*c{y4N_xqD-Aa4k-w}aK?V016V%4kV5;6HmxPHjw zLTJ~%w8Te{%X2GiB@Z^k0B18y3FCow8J~S4tPJ#9yVs7K&MFg7Q&q{4M88RsKD+|r z&x$)ANp&$oCj4_V8;X8+g+&@9-Mnf4>s}KJE477r!!y6@XOK7G0`pVchDm?g&f!;I zjFf*6#50V=KeWVZAJ&ZIe?2FVoK_yojz~>9%Q%g2bo}oFwH8$O;Jj2j`#S$#IA@f>9_9Tg}Y#|Jl2cH2sJKkxl0Ycbnf3F6$Z zs%pZ*HBpIr5$G4;7`FcM4!Q<{L%yQ^?ZAJbu)dYz8iZm}+ohk+;8%`WTkdmNuDZ6q z;tS(0L+;nUAzGgPjC?wNwGHwSJ2!I|~BV7sqwSlf7$6=O|l` zIi8CsjJG3ywRTy6`8RwZa^`-m9$6}_bdw_i1VpLC+=fFt23#%tEU=lY zBW&}yhW33M17!+f8c{Nf+;A37(Jk>BP)u;p%{z%b7YCG{OsV%2WTeD8h!UyoY zUHcbAl26W+Cak^ZpH-D$$EI8Iy@_778PlS^Ur~GE%oc1QqYTA-4cz5_Y%Q16P;Fg( zUA|jrzwyO6ypRis)~;5Nnq_;$q-;T3k(pcgX&ew>GL{&wv&=?SGnBS^EuFOH(}lHr zz1>xaMez1KUL@r0cdjpYWUEJRI8uR%=Sj<@o9#I>=K{jkVboR)oBUIJ1U|=KFLa@# zW~ro^SlF!eSp@n!0W$UmlW#$4CQwJ0rU>G!AfoqAPN^?P=?THeQ{fTo8=R%mBvVud z%>LPK9LImBl!k7xxj%ms_HvidcD?uhqC|(Ii*Gp%H3ncwV&Q&)qqH;oWUqni(XCxz z6l}T`%PgjC*#a=ZZB(VECGyp$IhP+~+@VPooiw_l=>~7xiM?F}%lG_HByU89xp(ImX4a{abNs=d4S+~dsqC?mb-rIdxpLU`=7x_EFKmrteYzZWafGR@ zwZg&xSz)x2Tw{IXsCj~u=F?C`Mj*DycnexPVdyg_mhjdUq+LeR&{& zWzdsqP%{XJYrMM}o@JN1)qNv75+BDnj<)q}AlMK;d0#04Kv0=#G{Nq&tjuZG?p?O; z*#G)3u!_$qdo(>Y#GUHx+%cV-McYCv2#y;d2t16upSOM+arGi+K_>1Wgboi*<^aS3 zWaM%nk%Av3eQuR|MNtoVsO6hwSC zK&oO`?Va>fz3q^BS<7!XV13>* z8TKiEGJiGPjUXqXpFR-t>}v0&4P?@Gzzm|*F1;k;6%@`W;Tc&}^9CM?+L8c_re!Hv z^KNWrmV1*Vs1I@@{z|QaDca!`sE#L>U|v2=B5Ic2+M|MPu;r z53K(tEv?+keV*kS-SzYO^q}CXE<8FtDJ4(wR3Q5GVaMdg?<@@?U5!a(?VcfZ@46>d+F>G3H7Inl_|Elc5X-;EZm)D|q4V+gN+=(AF$5o=y@R-D2?h(V(R=8sFpit=$9?bH>l(Lp#+ z*?^vpnP~w5XCn_aTm=?oTu;+#&Dk&G3`j(xlC+NBy11t~s$G1#03$PAQxMtN!1J5& zzx9dnN(NH9bQ$&w1+V~!R?!BkcnqfF=>1g(oaJ1e?AQ2wtVFdKzNZj~krcu001y@ss2zT?(kSTsa z2y*9NpeMK@XpmP$QRSsgYHD!``%-zNr{BroFH1(;T^$63b^N=u%guApL=U*s*`^o6&EB8NG=u z@*oTa?S`y)`b&Qkv<8j=@}r6Kx-Jo??h8wixBPBo*`*De8N`SkXg=0ax;za_;E}s6 zwX2u=9X=!~igIjKAI9Jtw8>y3Nt>jL7XEi1n|TA(P&obfmp#$|lL-ud@A+28)UToQ zRRFT$Ia(zHQZ0EEE{<(Ja|6)8k8JkenE(HkBGNu^5k`x($Eg-G9YJKX37$4|lGF{E zZxnM%1vv~oaZku5$Ps_MA`{aZZsuhCsAv@IP>d=n;0OqbEH?HTu$EtadCCh~l^QiZbSr8W$o znxSp96F&@OEo>!vo4eS-g4F6TOg0^AgX2kwvBp+q=w42eDiuXD8@t zi$gf~iCn|?Cx;_wdPd7&_@Mq6|5gi(|=2uhK*(Y}t#w%d=7 zI^PgNnZqg`zAP-ld@5s0H8EYIV5C;b2WT86*uoXl z7ryZ}T!QSV{h{k{PmU3he**lt(XLorhqz0|%mL7m(`N-!@oYT{D^KU@h2Rm}N+_b) z0D|mnb=wd8f@hHlVaxP>H!=Sjq#r)t``2Si;dJ!Dd4lU#@M6_BcyU$iJI{ z_3IqW9<306fck4{(lHDHo=2l9dxUnScA9XT9ETH?`7~O}qxOS*j`%JsJ_w3&tireK zzw*D;OoDFvl1N@0m6i63Om94`MmSM~Z8Y~v(=!ObN>ZivD{T3Q(DAgGIHn;EfRTtUC%t6Fm$xp-`wbngz5!rM8uieML*gjH}cypsqwdj z!|E|{I(&^$;R^3M&yKVWa^Re8Z^Dc2D<3~&R6N1{a_R;m>LIlCdlz?ZK+2j4XM?IV zYtxND!+!qxS;PeU-$e&MW>-~50*#rZw&w_Y*6sSq&Py~%54`{Pe^mAhcCdf}$d*q8 z!@GNA!O%I@fE5aKaNaz{^=6Hk^~HDSqGHer91HcMzE8fzcZsRwjYN~s8gy0TQr2S% z97)YMvV9~YzlEq1yB9BRueU6-!^x?3X zSnf}`Tpu4OK6*HD8+kdx|~;`)VR5%73K z--i5qvUAs?75Nh$@n1#D))y8;)mwH-IC)k_VPOgasPxFzMswuH@vFP$k@=#!_!Xb_ zvi;1~7+dC9tEm`3Qs@uqFBy`F7wt|k=s5kFlvqYUD~1EnR9%^0_+;M(>|`Fq68Ro7 z;yp0LcIEt40+? znePQm@3vUJnFfFXapad(MtT7VNQvK_c5YQA16gECv-dzlt7qttb%lrdX8E%k%~H=& z-yr!BAY`+}FF%@BveEb>%OfB#=+q751yd^~K=#wbIrAPMeM%CIab;i{GDZihiV8Hx{cKYs^Vki)fNs4r@J?W7LgKD%MvnlZEBXcfBgJWmd>%dhf&yD5<=WN7o`0^ds(yMKt6KUjzO~vEWuSVyQ_0 zA@+reRNGV)oVwu2Ib1_>I>USBtw;Z$k1sp5KXj-&U)?QdQl$TlP#ujQ0Re$8TTRPr zzo1Q^V2B1;J@Kr!r+tCL?Ksq-U3lac+0`bgMxx}9vT4_v+6&WKflv^*MfvUrhZQz^ zaew46HZB!&?W1<@*0EGc5Uca^RiJYIA5(7`6;=EF4=de`2+|$WHImXpOLvIW(A^=R zlyr9v-8D#qAP5Ku4Bg!z3WN0jaDRW#TF(n!>0%9YuCuSb_orS7eTAA_cPi`8KjtWk zcxi7~2D$=Q!q$@<8u@c0ZTjpU~E()dk78%M4W&?Grr%>w=cTPHOXK<#MBbpYs+ z!sBoDD8>H#U|3D;wFXk7goft}WzYe}Wf2ql#s9R_;i?sccoIE6G_P|U_;767(E*Wj zq>|7#`ejkpyU*jmM5)7;qq4)=dHKkdvn}D(3iuL?GpVe)=3}uSBQJ^o!51|kfVO-N z&=CFot0taU;y_j=8h{!K4a*-5zug6*78*^wpNIrA@v#(wjo-8WubKn0Z7Q?xzHi@W zMJ0HXT>BX$5?SlPqM%3Fow&>+`54vrNMPPR%KK}%n%=KXio~>?ScXCp)Lw`=@839?LC!7i_h6Lx;iVYc6k%Aza@am zaiafhQD#=)$(h6u@N#$}c;Br(-sgKW)_(<{vyGaeYXZIo#~?X^MZ zoLMl{Q(C%XL(YggF>pYN=z6yo+nqxeyfaUn%tr;z-pXa{`|s$R&nUcwJi}?=9O0Av z>tUW8wK=_)^+RID0rE&WkVZ9oJ3!HEP=xiVR`=e<#9vD`4k`Z_)*F*dJB( zFSeb5Nor{T(FT}x0MVV!8$hVZKuNmk;Ykg-w6wH?!ZK63vBmwx1YqWU243$Q7xbz> z-#Y<#Vk=~}{T2TeheHy>T{5dz3)W5?0~|TavxbrcKi2lKSmpwpK7I3EB2n{`Y%M1& zD#|IXinhmCk^iS}CuFEPo{1cQ$UHK>4MvP3tJPNCZ1pUdCOJ?Xh|_)kgo06VLEjao zL0Ltpx?e6FID~#D=X$vd2_uJOi zYn@ksm~_v3qHS+`;Xq+9DewvxiA5ncX%QC7wGSZy;(ND|h$HZ6PMfs?pVDqbS(X4| z(@MF}9pm``z(6ks4u}SsiASPS_qDa~nCSfG$D|}8y{qRFduxvm-rh;{7vj|MLt7A{ z6f)&k5-BfMSvYSxnm2vf{A;6X<}D9Xr(oPj=r5`R`xlQi4LIEt;iRA|m(rbinA7z# z%+Y$?LHWpGHi*0FVk|UXp>JNLS0kLEy0jbT&+%5uD?Y|Z^Bf6rFlL*d=X(9ooH_+B z{mYi!Tzkp63vhOXMbK4R4B$R0mWJu^8hmpRvi>DT}|5anAy6;K}ZkPE@yf5>t#&W@aQ;Js>A;D-b^UZ`!8z+>%t(+Py$(KOA93Kw_0B~)i>QMknOZ1Hvm zM!sX+mi<}1;mBzyOKE;cXvU_dmYV>)m&ZXB9N_Vd>uqP?_zf}(a-5KZiQB=4vq5CQxV!}0j^QxD&-Lv# zoez3;87w6|{RkL2y)IdU9>=o{d0AA}@e%Jid(WorcBYz9ZzcX4VN`{-Z{AyWP6w zDw4f&sH9%E?cXj-d?^nTh7dff0h7~j=GzqHI?=5t2eg=ZDt1ve5C#Ha9{DA za=#_(ZMv{pQ(sug#YI7y5FV^V8Ck(_BCqd9iws^Om`5rLtiC0M6^KMUS8H-~F>Q`# zjQrN|INj85xCDAPo&T_H`UlNRGsL^4U&X5muN6OtNhwAF~Ue2c-U z>pk?g%PM4r<>%;N$%h7y%_y&|6Dqj*E1tI${PH5l_2%NTl6-aX4$Hi7qvF%W!U*ac z$)W-YI-PAGAM=P37l;vsO59&e@3X`r#{q^%PtM1q*)*F9^}dKtjg5Sq292@+7rGK4 zZIGi_zKb)!98?4l)l~rgtvidNn_DA0J3_27bdgNd!Y1Dr zI9x+bON>x3)sWIXcVc_KMOs5tW+<*To22!LD`bHfal@SrTisQ~5u?S(t(_&BF6MyH zMt1)+Ap$+P?arh4?^=7LH>!Aef_Qu`$6L_CBIO(|Xzt#M+&W~6lgRSnQN9FC`NaPg zx!&$291_r!+>8MhVvy=wuTHN6&5q7T&L|*6XBlz6Kmo8>h*5D@4kjYsOa+jTk#!_G z)B#Sf7s=r}!06&#d^j%o7@$YP_2TTQFe#^Vc{0pB$+)8%iN8m~{)Qy~S63-A8E) zPDPN8YfaaL*tZ6^SyIBLcEYb+n&mp*m8&O^LJuU*#12SA1Ho%O4B)HBt~1oXh1Z6y zc=FYS2VhLab;c5uUjf~>?dOGQ2*9@eKmZOvEsJw4pu@r?^eeKKL zO+lIC02(=MSN6MeJwPm#tli1qT-KGCJrEU3xP0T4<~%NwcS(!}!tH30rkdQ-xXn)K z@fDjwSk1LnKezyLi8m}uAD<%|o0GRpP$U7w=GE9zQNS6@#N=z`e>IEtD~o#Gu?Q_j zj_jw~Ud8`D?;E1A`ct=Nf!_)Uzm$|o-4P>B&Q3i;M04}z!rQf-8XOPeiNZV^G)^}D z_9V2><^<4Pc`NJ^yX_+Fz z2lC8Z7T#gZJ46Z`ql{=6ngiYJcs#!yynLU-JhAoPs*+xUM3x2o(nWK zOl%LuBSFz)ZMP^+=_2@|ZV{iT&P1I5Y3RF|%(=3U!yT_3BY^?AwbYfU*`a@)YI%y% zNSHj$hly#D1FOx^m=y2_DE@Owe@FC8^MMcUvcIzg=LZ<0R_=X~XBO}-am~JCGzZ;> zir971SA$74jVYmcnZofG*psL@LQPDjhm`p%Z{$7hrWYpmDRS&@le|j+16e9HAx*iU zGygTfBx;4^9vXBFoSK~E^JiMWBa08;QUXB}h7kYt7RUX)gjVt`dz%woj{NzH-8#=u1vFrk{RBg#1yOIO@g@ z`Y`l7qJNB-X4!3&)9dckW4vUI;GTlz$==PWehQ#tM))uviuoy4x08vde%`SlBgD~D54Df zrqUw7v$`YpIw>h>xxIor#5-LY!NkN=%ls_&br>418y-kKAY{w4FBJ9kAK}8-`NaUf zv5~PPU>xG!LPSoq{4UofFRSnT$N2ax5STVoVNZq8uB$S_y#D(}Hi{Qae1 zT0-TIg3%UUKJC%8I_Ri+C4VN&J?_EGJ$U-DK;dUA|xF2|{) zKW4Cc3*|+?DDSNx^+@r$b*42ya9ziZ6zJ9A65|(HB+;@?F{$t06|L787rKO`zPylY zelJrpH8nD(&)H)0AyYn1_Nz{1b@k3_@Ehv+U_eq?+bb|$kV{qkg{vJ5>{eiTeur%7 z^n7`}=Mf)5Lua6l)!gO2F(zY?Mfei`dKV04q!yDaw94#&w4oNi!6uyRtA@4)Vi6$4 zx^7P0&h8M=?ic*BnAWgbcI8Q^0@G*yKSqxygM(EE})CnoUan|b{Jax`sk zV@Vi-7L_hj0>SvBj>{KVfO!rxrGxG-fcbK$1aM9s#3^IiTf!V<6`gvV6SS=c8UWeW zRlh1Alsao4{`i@87LfZ4E$q$2x>T`&u%C={^aj^~C0exF3V&8j7%m|K=%PjOf5!)1 zFqrrql&JYUznmwq0?iXx>?L-3vA7UE9GN}3jf;q+R&9l1bGeL*bkF_QFZi*%raqzG zdXF5__=~5iM?nk^1|6Zd(yi9%h(<-|H8c|6un8BoD$zSSVBh`aS?%r9H6WjaIaU~U zcgCi1ITjZcG49DKtJS`ntGzy45?&h6&m()^F>LmDqt^vUn%e=Xb?ypz{5pxnt&0*o zH6ap}1!cTLr}WshHnH1TVZFLn@lToLeTAY&{dPn#m)s_;`ELs}*=9>U`?MN|CmFs7Q}E3u{s%98O0GE0a*6=ZX++Mu{lCW4)V zqiR)|==>#AM|)m_Lsi?$tH^pHE@OpgvfXoDkktBBD&(wdRM1+b#IR*ElicJh>bF2# zbk>EFAAF5fy~E)jC*Pxl7moc9A)0V;+Cq`VCp?yC``8|uXb+mjaOpUKEfr~1?E&gi zPs8kIqIr59*mLZNyQ;eG!vHBya?pYUl>^B}f0G{o9dz7~y<9#9%X4ltFmIB|^(IN$DP#}EiTYkXp9zPuUOBKhj8fz8@B%e#bbg> zJve3)J&te=A73(h?yBgplBIV2<)qYi)(s$Z{_AdCVK}Xw9KT!ax2Z1=%sbq|_6^2L zP8^pKEZ|p-$GmdU!nRu92x?R$Wx8%JG;X~dRrn?kEQcy)hoaAz?;|)Tel&N=Dg9`* zfgGSQr91Cxr9wTlAh-QGQeA>>e>)tl4f+Eew{ltD+sT?PpbGHG3(r=JdVVTwf8UR0 z!`;j`qqR0iNKX2KTmBjI;;rS*9?OzM)K@rN7`k3LV0J+Ksl>V?B~$+k7d&*s=T}93 zditJ&T+I1l9HAHO_wt4}kG~@%Wy=Ox9B3*OTwFFtw%-`((P|Qy08Hzos#_R;{b+~x zS?v=Y+621;d-NAwLQ#7`!dO|(+9$!Gw>OT;<~E@svq5Wt0mM>AgA+oq5!BHuV;O+r6JZAOrF z31Csx)9drrN9M&cp9I8#J?vNF^^c(bOd@$M`f`_#iH5sp!H*qPqzxTB?Xh&d5~<+n zknB)(CZe@}+Vtn~>P%Dm8DEDpS!(>43@`j@Qf8G71Dcz!@R87$zTK9zJkE?1NoCC* z0#dv({S`ltM3I=5o#M?Jn06ZYdv-~YV4v6eI>%_!nIjbmtdz?*Vp3CSXdmeuwSupl zf6$S}zq-MFLfDd>5N%z{DE4NcN)P?EpqQ$Z%gYXj5=`en|LXp4J^OohLWlMXJGGa8 zc9Qc>0g!1LVC)j)QmoNhu;z{ZO(m4_bj4@UL5HZ;mnq7$akzx%xQ(y@!8H+fju9IL zgKdq#Dn=O0gXh@;xc5RuRGBs_$f46r$r$8d3^0b^DYehkMr-jL(~VTOfYUjoedQ^!?_PO0`q>Su)C4^( z%j#B2)s?6N4>LQ)Ud%0CY|V5j;9IpotlW8T7I$kd8#X&Iy9HeKGfS0O^-EKxaFsaM z%=2|CK_%Hri_}GbE+j(GJP?L(YxlJ!pFPEwpaY>{TAl<;i&)P?H!4B_MH)_=WOVU2zw^zuk6r#P0d0x>HU7`!m zcEvf+fXeFmhi@8yF;P|$LQ%jlnG9M$KnH~(6b-Gw-c`S>$?h9|@IM*7;u8QeYZ>fT>N6nT$ROqOJNutv?2djG{T7ALpHs{Y zk_$nm7+diu9b{4ennVUtI(TG!u8zChpP~&P){0BWGH5MPe{Ch8PtvU1=0?+wt^plGZhSLvw zr~Fl7JW~-(zEdI#I2ufiM)Tf!DT-~_P)Gg~&%%nApVC$&iIYoa!%ASci$WKVO4Z*n z_1{F+3Bru5Cw7-{QOEmEc|9Z@A-TB^G#hik+KeD~QNY}PxI{e-o&MPybbI9t zzod&OV1I1laTte*y!z7914vO8E!28NFbqZS(*WDZ2g-cf0eC6Lz)13}Wp4iQ@!yc7 z@jBz4O8MCju0~)haDotHU?9&G0%UYoSEs$#Bxq_yM_;2MV{{ zAZR%&*f)|q)4{E|fo8I>wm6N0&Gjg`v$J58A|#nZ4{b57&eG*^y)Tu5j}h+O($tX6 z(SeoYfU$ohwi(HgGp1+=o{|xg`wiyd;M5&D05cfS_Oq4)Odh z%IPuYB_VK|a=xR-IjE**Psflqg~)y-S{eL4@cb#%Hrw>o9t9kx)%#zswbGv_41)W3 zFhqjq=FjLM`DRL)yw_#?Dmb`ziSS3F+hR?2ZrGg5f}vPbIr>|rk(oO1kz|4!S!NdR zxm%WEdj9 z*CgJfc;}BBnRMmVT*h252eKjm&-tt!0pvW3y2gA+_hUgB+i2eLTL#X$Nk|iO+B^tu zO_98~eLlqcfEoWqQUbWKgkqoFb~1`$WZnG{)_fT0H9zzCO1^)U(HE2izh;F%yUyII zQ>0}JIQapYbe<~5BFUYJQF%pLhpgdMck`6M>g)o7;B`H|CR7*sC%EZk-_2rKWbS3x z<@MfF$JxlS0c7_lt+JxnyR|vuae>Cf0yEdz_w5NLE&~_`jG~V6@p0v#jytJlk}bjSU%9)ZUAS=7+Yo6;11XDPX!O^2+cnrr_tuKJ5d)}Iqur z@TzWqh|uWE&yBD5f^v5FYsf8a7=+6>W@FX`@aOn6({(PMakuo#@C&>oAA)@+1kIJI zBvJ>w^V=po0MHvgOrFp$;jI2Af(IKBl<5`i=+$TZux zdk#lIq7(Hz+@`dN`A81<6WZr=Kx2OX{ONItqIk_8@b9;w0@K;gjicN?V7uG4dFBi} z!kmi)cs19s>mSvwiPUi84`?@UZ$)`+YkvBy-MUQ&?k(Xke{&hpAweIVWp|kxkUF)1 zoV=O}j}NV8WHNFHv>t^C1^&r42NzbFYM89AqNwVJryUrvn)D_fR# zAPf(tluIJ0=^ZURLaeb?<}B$C7N!}4Y#?_&2jJuzOQ!cIukvq|MI_pOf5SGBM>{mv z8m+F7QSzj;Od0+*EHBh!5`s6-Chvuj6LWui6(6L@e>l>oO9MR1rSaavaDopQAGA(L zfTu^Tj*U_{vtTFLm=Nig4Qb0l z1jgeKo0_o#^X$z8gC=NMUealgBpBa-DiZ$_MR$mvX(EDW`fYKjJI~O?o1WXQD`@L# z(rjIP(96$R_6!PW8gZsRGk9*h3|JVm(j+{MrED^3=s6AwG=a@k3`4dBS-%0-r}oav zwSR-c#$8)0eWLSX9oV@&#_pg+i&q414V;8~c`j`;*c*lZ{ogk0B_>3w-eQ7(+pd;S zEUNq40%>U9>W*#2^!vUwsv$JUD|;#sBs9ct(WrE!1z_;<@psWuJQ=q|>@JB4r&ey*YgDIl6c{#2NFfv(#k7lPn zJ`d4ygtJ2SfAzKj(DR*i+UKV+wdelC*@UWL7peW9NyjtSj15cwSd87=-a9~U?QV0HnDZ_X%-5VE@!r zRRQBK_x=RhylSHHA_I1~a3)Z=`qy@RPA9o6>>IK}6a_?6!h7)?)l z9@g-cP~bxD80`rj$%!QX`Q*@lXKpD69nN6Kh`^OLGKHVH9OBoOJbsP}&l5rErD~}U z7c2*okC;;|LRLRyrPmpylIwc79Ont{X_d;9fV6>q z_%3o>%N3C8R1yHKT>#$YI$$xl2T)kNLUN%2WwoA;zwuxoBtm&CLd}>2-A^72|69(A z+ffw3zWnZ*>MZC*%pOebvPq4aIeq!Qt*qk2WQCiyh?A%(I%^6SB#^yHL~>JLn5zsU zN#wT%fBrm)>52!@{H;aXNs#Oy+Vg&zd3vNY9Ub?ZD^aAX`pQQwAbVHQ50K6-|ESqT z>%rcmxV~_O=?sOUVlM~1o#zRy6Ovn&K{rwe*Y*?h$M8D^@Ui5o=kqD{2LGW(sG3I23v9h4SPMB;B);O_pnxes6YkZeNv&tp?Z1_naUp5Q)tD()}vKi5FMm%eo) zLw)mauFGC`v`jM&lUKemb$6)zrN0+_C;hpz5Up5wLpW%`gn%1VQNH$KV%@mk7vyM2 zFbVb}k+0s+Kbk^C*8BFzZy5>M{Wm+Ril*uN6rE5zJ0F*$KK7hXEP?qCK$CGZpoaZY zCbR-Pn#S8;z@*BOP%fqS>JCFl-6Z74Dd4GH#R3AFDzZ0E7Z0`ibaCl@O6i)Hyg{Kh z$r!lt5*zY&Pl}eQotv>O7Ct*5-Ej|%ITaj@oQfBh!=>P$(S*V{q+(Z<>TJ2LWiz_C90HSm*A{%`e_lYZg?3DY);sm zqu3ITdqd=TPhy;yVVZWemz$BBn{Ok}+dH=Rvhh|Z*ET<2&_acT{a2!f97G*J%NW}t z`?OIG*b$VW8;}SdiP0^=H(+%#i9C7Xk&vj<7~r*&(%q;1k2%w&=jDQ`$i_%qX}ig9CPf29BIR5f~H z3VEq7Qy%LlDF@QzLQS@Sn`J^+exiNgE!yY0F83NIbH+J<5nFQ>c;@e9si(YgyY|>^ z>UY32mM^*XX5q=HM6c{51fBZVL0LX^Zj~2p+}*z56A}U(@s$##bw-A1tXY)tX2Cke zBT0|p)q zmGG+e_=ZWu>uAZUSSzLIsUA=taP3Bxk&%?NDPPOmSXJT*_$!JGEdrQ$B@mvW`!us% zBrk4J{SQF$p&*2(L^VuNHf^#f9*o zs&AQ^_@%aRQB_IbS*f|Fre>0R?fnLR@Gq-$lNa~RKp1lNe_sc+x?`l@@W>aWp~YG9 zsWgbxXdYx*cmA~^^QRjnD!gWEq(U8MVqsNb7dLgZs-fGr-joq=eWR}l+Q-a8qNvZ4 z(H9F}@zR5@fKACA`bWR+qWbz@Z|2hGL%z4kJ2s|`0NjRkN=MIB>S^F&|3CCBb!j?? z;K23}$;PnF;N0pOQ6~PtX1j8YQL!hCyqEQ=W%i=R`_D8!4Ylk|-zf%%d@!9Qg{X7_ z4wtNb(7(cFTaDk|7;*iR#U;~f*i%w}i~`#e+R-D3Dqau9kZ;LKDXYo}0~r^3^*2KR z0$mIEpY6lN!;ofHfj;o54i7bTn8#=Fur4q=TEZ8Pq`Z=nI7_!Qp#n{HjcD0}qQ-|_ z7VDNDP>hJd=mu0!MaF-L&qh{0$54Z+H|bo6I2C?dR+l1s0Jtl8gZxLJW{fA4C1*{(JpJExD$7|x?C@iP?FG^eH>4i$on z6cH-sOEHFDzBMYWj6#98slCz{#u@LOy#>_IR?i7Rn9$H19=ew`FR(@8Chb^q8FPIs zci{rbD?nb$+1CcDd$?@+Txx%9?Kz&%(`$MB+rCO65v0f1O9(o43e*{jKPh zozI9yPYlHjuDBm3EKm4=rXpMXgM7hJsX9Iqi>jT&oF`xrJMt(M)?HaVCL*R95A6AM zfhfiL(SPT~H216`Gg|2b9pM-0wNG!X9lH1(H2X)UbJ`Yn$uODX-v`Q=qRP0WZRDGBnTk6uWl-Y~i1u*aDB$Wa6 zzgOcpFKKiAYcI0bz_-W5x%#*<#vL2s1(6;Zx{VuP`0m(n77v{icNZgI_LvYRImAU* z#7Sr@6M@#lL_L));H{*roa^ogAbEbfnKmBJd=14su@2}s*x5IQS|a`{GMr@7=&7Zz zqJ_}>ip*P7mqHYf)zR~bjfd|@OD6HfT80p8&nrm1aH56LLn>aeDA}*Qvpf>GGGILZ z#`G85L8v*mRdUL*czX|pBl>K{613D3$3Y+ViYLFV%S=g|x`YTg8UOZ{(KMyjP>eF6 zK$Sux;!EnhOcR*IV*uKtWefh~`msCaGj03DcTKlIbdqaJWre{>1HolZF`gH>kz)wOzV53lmW1U{P?N)Jfx6fI<-zuA_(A zcG?eTz)IVxbqm(=#!ZwX*(!`c`<7sJi8a#?MIhuN?LBrq{xL2R6{cb@lMo#CcNJa4D{s8;> zU#o438@#T$Uy7{sr46l4szB3=c{33Voil4y9$>sl{{QYoSFFAOIeXO)olG^pYZpsj zrZyv>0n^S6zppkUN>aHM?I#(Xm)z`&X#DI@R%u6=2waq+A64@Wvng#m9Gg`5xf7zt ztd+q0E1j+Ee76uPV3k&qJ2x$i0IO9$3$m zEjP$70To(E>ykvV$Zu0>-xuly_$zBP*`PDbJmXX%M_OF(3wff|jD-oW1!$2OyM*9L z{4-C#AJFPa!2JHdn`Qd0x4W!X?^XjC+`>pm#dMTlsN>CTtyv)b*JEu=|FS1_0#;8x znFpoW(-jryekGxU8ex7;=}Lm3Ht2x=c;#egYFcww6dN3iYY^Kg=Hhw1xn|(HdYBum zYq6bD$*2E`C&KDtp6J!k@K#$x{aG1>?~OmA+5_qQwP=6>4VA2QQu5j~&)v;j&+HSw z@_v)7=huvnnfPV6cm03tCgQRJ{vRHRiyBHoGASAm_XpORYv+W+{9drBUJvsa?7))s zeYZe%{!YO9wv)L?d|;a+tod}soW?ni61p;Q4J;+f8|251#}&#uS8P+|7w0u257IE7 zN5nkqk3yvPps+pEwaHueDEB;edCZwi`R)nc{SZWAvH4jJ-yRMLi`8(jWTgA(e%6J2 z;`@`AD05HI>;Rc18yygsO}BPma^g~ge$AYyVSPFCJ;eRr?dV137TsDP;H64NDKQ&a zZ7*z0;ZgzL3uvOyDhVJ)uQKI3_x$iXt7ouuDIkzL;U-uzLz0c5ojShwwIqR#ENAT_ zPoyQZ;L@wXZWd(05u0d-4DG@b>hpwlWzCG2xBQgQj1J7paQd6H`@LhfGISbC0dRYI zb9WRPAey6`2Brz7BxxBRdXW&<_d6UxNY$#>O=*wI)4o?YY=rwa_Lk`;!7Y$#uTZ%$c;|t{sw&Ka+?E9qTCtT5QTQO9jS;r47VlzPd zR5RYdHXD##Kn5bsQI_anUw%BB(!Yu79iFX0&L!u)U+MBMSt{jmA!2#8I6WPp-(r`_ z!I@xNl$U6~^5gNKLJ#J&a(Gx$OxHHPNw^kva^8%=vpy0w$wT@*dOSpTH)>=zv zLFz6_elV_49t2v(u|N^?&Mo#zXDQ8`N{3lc%>{^S%!^NJv+v!so+oN^N$mXx_(U^Edb{Jbj&WZO8{=BGfn5 zGIlps{%jwXg>KZ7AQVJj6Rfu=F#PE_$=tLz0X!~%9aL~K+Mu~7Q_?k<`RyeMXyv8q zK12Xi4LqF|BXZR0;kHL$M|*pwoFB+sV`LK{>c>iGvC_eShArSQZ;eZBTv}bd&adA8 zM6N$#k-Uz2%C-~-rp~8FH2@HNpN4RGlv{WQ>Slg*OVrQGcX;d)EURGNo))jCV${B8wOAK`J<&@W`(`z#j_p{WfCp%W1iEuXbt&KebKRkFYn0RKsky7S1VYQXuNQd2g6 zhg%TY*OW)vdi0`qzL-C_Ut#Lzs-)!%g^i}sCi^=9mEH5#`HpD=i^9+yFNQS@BSeZg zCUxTMd;kd7Vy^E8rTlhvfB`Ug2ToK;v7_vo-0Hqlk9MqgjB>~<8fdIhiqsCUl1?LA zns#Wq4=0sM(d!erRYeG^wevOkfYbLN0gIU-4QFRin6-}EZw$-r8Py)TjMXVN;V+aC zi?B?u)nN0v0c@`mG00qBcAuw#?HrbxAmw<@--bCtOAp~m!Fd#(&{1OUb$z4yw|+xT zRuhukUGH0AzFr-23FqX+3qCZ`DNM!t^odI89c$C|od7FDWG}~E0Pw~28_Se@{|zu7 z`A$k9Xa}DgAPnZuoyZo!DEjTK6W)a@`ph*YE35MLnFazl8ip2E1`aZ0KDG#pmjU`3 z7GoDl*rhng>^0{LbM-r!yGUIauw?^&qm=8)#iG-WlUo;h2j|H%>{m-tz0id&LY)u$ zpL*zUGCbD$_$~J)2-O3Ln$BEnac`m^Uy*;Bt0u$7h5Hev)7;j_f4u#($M<5I?IvBs z-NapR7vD^HYIS~(3Fg0R9p~xa`v4g`k_uK{rO#%;$rnrPn@Sm z4or}TtXq6+JY_z<`69v5XtZFRy!C|)JtWz=iz!h0Y5`3Y4gkCGfFfRmm_d4Eo= zTcEwm+eK%X16ch~mOLNnwSV&WC~;f*M{7MjPVEQGdV8E{-@W)Y=VgXA9c}>}{nb6x zcbX(rpbJkln-fy zUeM(a+607_pJBNmp5=1!rPd{e5wVtVuy`D*8a;+L+fd$toOP@agU>j3qn2NH?syg3 z7ldRWk`D7P-;ZJbfTP`*7w+cI-2iB5R+Wa01$(hx0tIyXeTWZ}w7`q0x~~G2rsjgo z7~G93gx~)qt|YBCEsd9X`CDvL$kn(X!;%u|(r?atk;Af_b*95yFd!Yj6@3mfG6By^ zS9qP7B+I+0X~hn$RtF(LwSf(~2+AVBLU&EJ&E$;tf5o)>-?zN)oF9EAa?gNy+ZCVc z`n&bV0*Kq#GbW5A9E|L;dDU4DYN6e5d=(nNP`h`Sa@uqWndOnHnG_N9T?-+x=tT+| z_O8J~W;yXUUWXpkw>YD(skXaMJ0#ZRa}+u(%^|b6$|hW*h!3{^LjPVmt@KwVIn}41 z)pXw+?Ko0W$x&gRx2fkqe7j}kVag?Y^*y6|i^E8?%24LRm{4G=xXYiY^&9CULP`}L z9+CLJVhEAY7;4S5sLz%(J|yZq(vBO_NPj|mhv&@j1!$y_jZSpG+cbnEqo__15p%Sr zqx=?->;_hXM*xFiHvjR?CKsSk0%$b0R0e4F8OsuoR9gZNKju++lkD=Zp43UVlr~EM z<8|IazUc%9Shf3z2@c4jjOSt%-TP?j`LYv>NB;XEC7XE;yHwV3%Bzizoz{jqMJwqA zmu{ve>ogw|lX7L2hk72R*vLN@3|1O-^!-FaUuv5=;yzn$J}u;Eq*bn3HPQm{95750 zs#TjA6g*h`3X2t*zr{UbZ^^4h+t2wEX@H%nX^5@^_0R}1&~u!hHL04t{FQ|(-ZHx= zVt*uzn7UC71h|f5j{2!~qYBM47y1=}fmSG0{j-TlB~v4>cXn(bqoqQN?G-IB$MO2K z+T*d|ktGsW4e7SF;gNAgO^Lo_6rr@mZYh;rP8}-*l8!@e*`L@Zo$mMT?7}KSgb^75 zk}h8zeI{O06X)`Uya-LeR=|CKOfT!I)+pJeYPtI}UgXh*cCmX9;~}emERIl3gwmBO zYueOxz0Fo5yIcIQ5w-t^^kEI>P4|hYr&l>t`-p8#s>?UU4o#8`}NkQCngqxhV_6;CnEq z-~fAK3>1^@zF7VKWoxw|DHN-(PoCn3l~)!+rHD7)pOdqWIsrd{@J9Fq*XPOu<$cLN zfneNG5r2uK*{}<0(45~BfgE~DTY`>S*B{ycwE6=Q>ayI6=;8=u?qZYiGtPc6#KU9A z`Ipzm$OkmU$`HSSvX1tArI(dE^sd}W#}-Yldm$V}eK$jQIM3SfJI>gcwRuWb*Gp`Q zPX+yBS~Y=oi-R}`#?wN+FqH?7CVtf|lKH>TU$a`ab)?qb@&upg5GyDzEvkxxF#=u} zv}RZvmSDPx2;pH%R4h<_h2b}%MeirspKn<_fAYQ&M_IOd@f66h>3)-CFtrFk5v z#HcrHhMgl25|!IIEU-L?DYZm`Tj1e`hrWW2SiPWPn|2*2Y?JmbudVGj$OU)qjCnp}>8$)MQgH>AIXYe~q-kWo734zWrdwdLXT4%B|yd+I&S@P*{hi7v3E2+_OE_ zIfi>2V=WZvWO0nC4ivm$ybmR>r+sx4O3#HaM+5J#(UELG+>`iNJ`(foiy#4bTMi?L9;^3AZp5&1L z{8N4QEFhD-dWa9cE1ph0DVAXxjt+fr7#BcAHyZfV-E(*KEIfRQ?N9LIRq#ze%|k(S zG?r>6pB%vLJq30oGq1Pe0MVE#VDw2H$wmPfXmhA6?623e2_nDPB0c}*6={;2If;)e z3x7dh@N<)cL(^BTC|a3VTRUgp&{eTHGi|-4d8LnC@l_>Xw(}1O&DOXDr60T-TAeBj z+3d~cax9#Ni6xG0kV@?&BQrUxB64J?cXT;@MQn=~td7XDH)Kj`iyh8BGn<%DwV9(i zmYZKh&Y#@krZYA7WjKj`WcH<^iY{YGs?sQt+#1t&8`0JY4YzMXIxlwYPO>6jaH< z?Z5)0OOle=Q!?H;4HX#=N0pqu95c%;L*)9~t!?;s#2w%r9HnX!f%=rNzi8`ctW#1& zC0dj)zgTo*Qc~n<^2~TwEZI)#XDj-<)+XXCiC(T9s&`!mB5fxF(3&8?% zf7C@M93&g?L1=JeBOf!K_JlxiO??oVqD*iJZCSu{D1}k-*nH5x_Xn54$4SjvbmDLH zs0*8qvqWJ zw^aVj6`dm=2RKK+ypR{Pg3XJmTn7km0EWM@NGzZZleke_k=}mK8W$-d`|<>X9#8T7 zxpCEef-0`2>;o|Ux|}TxL8F%mZ%psauHm?7p2V)gkYdoNyNEip|(e&p7`sPb>eDzRuRa{he=vvwY}r30vaf$x+Hs9tIE7%spf`U&98DP+_N`4tRgg+2jbpY zNmptflc=e_*8K*9!0wE7cI$*sGOCM`B1cBRm-gBAu@^Jd^qLw_aFvAS6|r{avaaGV zVab9J~hxG#-$1uyDh(#|sMpLobKoZH$te51u-uqj1p8cfKRdk2rcE<>G?lSK z_l>WXMG(i{#7O$KeBxB_+qb+SiN=2~NZ-lC651V<_D>CqS}S-~j}KBJKW1k zTMJf>)iAsILFp9OUq_zE1A34cc*UmE`157lHstSOWNBb#?SWaT-EA*w{wpRHGOfxQm}ZV268Q?NC$}8U)otLRe`AO>PD_o~ zWVXhbt-{LMU)Of>n>iJ~4NFF3l|b?L?nzTcs-y&S`WmSIsBVphb(PnmP5vq3SHcZ~ zT~ogDK77j9aml#WYLvdVTf?BDhHNMc6EsJp5c0BUEyUT>rmKO+HsVMc#qv^$4tj#Nu3~f1U=1 zq2t1PxtO_LJo1C161xTvvG=Rpd&+^;bcYp}trxtIg)Ic&6;K^yotXz=x$<_TIr$*j z2D!7=l>V7vS{W?asD|CWToCGLl$xbo-n=bcQZ|2`nN+8HaDBK;XNG)3V)rxT`jFq^ zE75JBvRTP}{g;@|yBzcjM%r#x<~&<8Q>rs`?g0eOpHt@s8My&^Uj(5()Gr2?owkZb zBbKrqk1Dbp3+1JdnJV?8X{6xmC&6nT5Ls{!h#;oCd+e|B{b*<3>uME-g&hLk6rP)* zwgWfpX(ndJabdCklYm6z^zg+o8-SKo*mu_2`wf(Y#X2xrexvmVTPeQ#pvX(*YsowR z&9;fu+JR)j_k}B7BmK0=cP^JHftNL3r-IBvm6m8qlP=zKxwx&HC$pTu*ywE1w3}7l zlt#U%SyoooYng;~)*-QQa1ie-YybO~v{XUw%bl9SY&Y;QH9xsym)bdVI`f8ooNe~* zGh3D*i$RqxOI9o~Xn_(D%^t&8%AT~STe72A`(c3#by{ah1*kJt?*U_|FSf%|jGRV^ zmPF4?^xkJxFRxB3o%kyYcqJ#;16Sh6o~CA+h> zD!9B@+_%yAP~qnLFAo&8q}}}9VO{2hgIj5AU7JHcplB5|*=?Eo0}fZ7tG;^&Pl9e` z?k*BQioPy}r6rSe@<|g-kPhdV@x#a~l8`zyO4nfE44Hh~(hawxSXS=79&;8cqht7g zG@Vsgn{5|n+u{xdifalKr)Y3@cZwB<0KuIC#T|;fyL)iA;t~iDN^#esEfhNW{+T(; zNltQI@0;h@`@YxO^+J7o_e^PomvGsQ2&j3JD_{a~5j7~UhmF4+1HnG~Y5GFzUFEp^ zl(`Q-+!K`)6ZO!RBI}1w5Jz86RBD!?OO_4kECo_|kQ2UjJ2*W?Zw{7@;@)-;`^Qko zv;IImtD#dn-5$1t^XTSRblRAA@0%r-r&ABIxw3X{X=BI^hIKVmB?YABbrie!~^026QdRTEf|tMB)J&JAPMNz_pH(Hc}7yL=0YM z$_B=CEVzbF&w9)pP+2`DMQGzlhv}b*KLXz7Zt(%2ahIjH9&M9Kx{(RBtfRf^&Uxjj z)7VdJP)JMs`5@6@AuC(lL6h83sRzGw;MbT%G*G~V#Egl629F;0!jkoq{q@c>pZVa2 z0AA-Fv26cFMwRn=3_?OakQRQpj*i7<-h!&VxJnsV&fb0hCA)q5)r^7f*Mr)PIT)Yu z%?b{52uHfjVpq(t8`2GHP+J+pA;@5CS%Q>tvmJoPTZ+wECf#6Fj;ENqjwWUZabp`Z z<2_?`(KDhDE$l<=^#ic2Xog>>G-xTq2RMdAFtZ#Q02t!x#>*;=iP6RSSkVu$C9o}e zgyl_w3HLd;)1m|gXbwLEzFrkaNGCi)bXr8ZzX0%!i0o}r$u%qD3{TxWt_0P92Njk@ zO%SqGTSVX@k@ZbZr3xj&SjUtGSQuUS@`2}EqxZK|7GrO_9(b~UeqmUA@#9lat#o8S z{qy6D00Bf2Y?x*bh3GxHD0%nZ|Dsj>{PmT^pDSQCU%ox6iVaQ6;*1@AGXc}Onoy}I z4iYXO@-n~h-YE>FZ7IsaiZ*$z^n%Ym(J3;aJPCj_zTV-wC>BQwS&zF6jTcwQ3wB4; zhn#8z4#~|*kpyoX$YrA%F{%u|3n71u0$>q@BgIDe$S=`md!g(^1V~s}b%pbfyzos)c6)yu|>X&L{#dx;Z(6BYY?L-d}ukQk>>L-3v zZCme!B@jUcfdDO@3#^U1sBOhA%_e>&ee${0YD&zo(dpYe=T#$Re%wZw`jlPHqr)1! zx7E=%B%K2*B;q@lR@NdcE{(eS`%%S@$)AQ9P;FLz&y|>&jWa$tzZcXk2l6FAp8a!D zY1w_HK{rgr>Io+eIr1Y*#ht{y>^PG?FYoc zmt-a@uRe=0z{fFFR%P8L%kJz;)ox@h(c_aqpV^NpI)o{CZ2`at=@8vhZi> z8oYiHB|`C2YdCNMWNA-L@-DmHO2^P%IK#87vbGIkW3--K-a{$O<{Qq!IgsCuQ?BPT zg59Ch($=1{ZgXpXn({M|7RFw%Rues@T3Yj0!H|YsxG_R#6ZU~~>tr4UQOEW&B2)x&9l zIY$}~N6VA0jWBK^XnFc^=QzxMAvMchpIZ}&dvjrc|FYV9Y#h#do%Pn;)^20SH(SiS z3dUww6+=ncTz)CNYFqcT-Ufuh2eqD_LvLTQdVh41#nq~`{kfm+JXmu-Qm1!uTPPlg z=11hzo-IJOk=O-ejpmC*kxKGZ#C9T&twNrd_71ydyN&*asQ)!f-Rp74xJ%^#llRC< zB(D-0vV~T@N^nu?2)or6-MM^dewpt?1@5DN`?0QeMU}=ZaTt#PK%pe+Em!`X@n}6x zxTQVaf6(w2@u+{jKE!`MD+b0SUraPxm;O>}2`w=NP_dG|)wR*rD{pQg)HNm3)JtFc z*J;yZo5M0`TKY$M-Tr=O*02XLQ&%D^eo2W^Uf0a}R;;9dTALHY3ja`JQmiJvL46y* znwB~pM{L+MjSP!JUiqawVp)Dnq2*Z$uS&Nc3fAWM&`931^SgL>C9oc;!U(eVB8Xku zq8p(J+>0ACDEuY%H%F7=*lIcIg7Jb>#(#lJDHpI{c!Y*Y6}!!P$^>WPUyOviI3mS+ z&i@cPZyPdHb1Le~6odA)_li!V0s)c!%%^YUQfPI2aT_`MA~#1_!#0OE9UrzMQ_^EJ z86cVvylAzO8_Gk9sxR7FL)ou-ntxd2z#T`Jt~qLgYV%^hQw3 z&I0%t5~`+u-UEss2q@do1+~1)Duqin!~hofyBqN#yUSiW9dJL3Cv?c5G+$#L)we3~ zwoL(6d=j)Dp#IIUAV+d+eT@y|_1hY7g*vZhZ^MkOg=2&30vY++_5M=Z_d+d{@@0G( z#pCyv_ckO7RiC!c<>0QQ&QCnj0zZisKL0bWH2u`l8q_ei+zxdUuy1W;`_vg;*T&pX z!oTmeG?tqHPu|q@WjMc^ap%5UdAibAxp`fMN^@-B-m!P&Vc08F+51&HpCI-}Mn~d& zmdx6Ruy^#a{)QctI32bZ9Ab-2=D*LKgK7|a-iIy(rZQK%vTr~$+!E*eoOgtvn~3z!LVusyS!EYhQ;SYII!~CUr%R9^0^kUdH@Uw~B+&-fKj)x<}t!7-%#cUFH2J8eaSot6q&3~G`?9Mkp=D}jmWHN}6mh=~{s zl|386tv|6idNxARJoMd^LJ}S$miQb(>Hn0YU}k`}ch~~>yE6_r>?@M5d2^T)F&MMy zG4AdA8G>7%Fv`_bZWyu!LG4B#zYEqLcP;bIIP&x>g5|#A8%7&7c4NsqkM=a4ukx}9 z+r(uZgU{Mm(Mrd2#MQGc--Z+UE}!?!&(d8bbxD`LRJ4$R$K`=P(WQoiza#e*9B;eH z%3AZ!y%mW{?qm>)dx;>9p<&*$rK%fTt`DrMZkTF!{N~jxz`9? z`Q>`2C+~aHm2+z2xgR%20E+>vS-v;OdT&V_4HLzLl<3AMIUmE13U;VRqS`hB_Wb{M z4`1CNaJ1u+GeU$xKPQS@M{#s)g&Z+?s#g|x@g+NWM^O>W+YmQ*hLtR%&;LDE4?1Zj zdUQls?f&s3vr|%yAO6?aR0D2@9We#>$pDJgYbZp=1IK*F4|Ox}Tkny!=S_njs*1HJmL-D5)1jY>myu!&_JKXezS33)0m{dU#L z3A2vbQ`6rs&&qD8tq?0%*XJSe)cj%oJNTqBC!6ttzkF`C#tzAaR~9s!Yhy||4==8X+_FpWqF|BMRgXA1Mqs6NHav*%?h zbnLq#c5}Nvhd(Z1r>K#Oefiw>ks#_D_KLKv0IpQndsM=yQK1tl>6kyiDVQu1^BMOV zD{$mW*?CPI=Je}6)3ANExM?xY>V5MR9x&@oZPv9@=$dx&qe&xrBAjK0b>T#pm{E?k z%}X_WQA;^Z#eQ;xTVI-kuhXC^t1b;*RG_s8O zq$O9@si5~nfd^=a1D;vyeJ;lklB1O>V*X6e&YVB}ofVlff&Wm$(y|2SV?^c@oGGdh zi^}7R{4c$yqW@<8*Br&O2bR7}FLv90z+jivBsT`X>ybuR<^1l12yC)y$pE+#1tT0+ zpe9Y^Mn*snZN2A0^@)m6CfaY~l=ca<{GMx*_(EnotSMmh(xqg6)qCNfyy^3KTDG&y zam1_`B`C`LguH731fg!Hym7}KvNUH>_w57k+Ud568=txhox(peoGb*TRm(rMoou*< z8&(|J8wDB>r(0Dv@~V!)FqIc|bh4e;RT-?A>jC>O{$+j7eO(b0n4BYCxD^Bs__?5| zv*eLaa{&w0=G9&Kn*y1n05>1q%FTWfL`<*q77i*v9 zuZsL$^2OS`#6ZWdHy^=hAX>rIR-iYgMwD*PkI8BX8KWZq(<=9lV?_3n%gHu7Q3jij z?EMFTa9lC4fQsvxsh*A2!3>YIc|_Z&VI!NedxpnS^D@}c$R*7D^Qyyp-82}>TOQpk zn6xbMr%C&qO6c6T$Ga*GE$&nHm1_n9)hBvQ9^0L*Xy8VA_B+=uCiFt=#a|`L=;@T) z?<xv0&ajnusOLHaAdO?V-(3CHtEXvpu2RGX?S z?hGn#+}tqvEXUYLfPy|QK~S&YP>#Pic1?ne&V8BGs}^U6TIQ0_y?42^L?T zbgKh|bMUJ)E+9(mTqd@QnmPQsENKH1=#nXtTUIhq2@2w$JM zI8@v$wMbbcCNR^R>{2`J!YPAv-ZaEqg#2bJ)jc@nj+3_U9dHfi0Y4OnY|3wyg4&*) zQsCL;*mv(6)gd#G(EA`pOeU1z6u>`={)Gc)?0Gj13IxQe2j0hEmb_q7){}^;2HDue zerkd^Lp`U!1RGBoQl6>>RnL9_^?`XA-EYYdP4>z*{zESp#41kBY38zU=-Nx@G#A#j z%SqXPBsl-o$E4lgEIT4NrA7~hJn@C@LUlZg8hBxs*QC~1J(%6}I(r!;6SZ-H_wFyxXM2gG=gkV@=Nl!&9=M?B;*Jx03|A#BVF@D1JqMV**fp~JixzHwb!y-W) zNBxM7E)I23YTuPthCGP+(;=(6TY)v*tV(nmb;JLZP0n!WNt!k#M-}tF&M17#z?jOa zDO+A+$Xv9pspPo4Jez_{(w67wXM~&S+aa7KM-V5Klc4~$a@?o7u@m_E}=?DMsm-&hDX5d&9D@G0R9?8zb^LD0E0--(8K zs?_pS*Hb!fA2m_QzLPq8;HXb-DbPt!0W@DB_I&N`;(nY7!BG;6UucJbG?v>vKQyiA zM}%IP+v4({bi35|BPLKndE0;d{|gwMkpH(!@!U77`M|Ch=#`J@hN<(7V)ZZ_ahd#} zrfn|8X?@tP71{fi_e7$qE&W5XuxeV4E$BX(`tggY_>AGwr!rz>HAYrGQN=0{;#FZU z6jZ^SG%(?X0f`SbArR_to2mWT5v<1XG<=5dfmA?{I?8$JB^&edSPt}R8_d(tH}BV*}5^lr8wFV8s})chRyhAFaBg5j+0N&wN`l1G1fwK3LQ z%VW$5wL@0pj`#S{`;?jY?wo?b#Vz=_nYjS%5jT@Yb^LljOa&4!0iiKlbo3M8m z_Qd+R%&T^#o>>5_Uv+oNlA3Q=yf@OQS<|ghl?6^)bZtzY!FS9f)53L`Q?)5)nFoX*ZU2|ztV|_HsNmg7HsY9dWgXk_l&v%OtOP)4n3TGHeE|Z)C+}X z28zbK-u94}Us?5cJUcU~pfJTpf1N*utdQz@0F85QgpW!3RhGnQICL))YL?oeTGrN9 ztQEA%C=6sQEQ;O@$ZCg#3ZS|UUfoTg-s3uw*UQ=*&fI3i&}moR@N+~Ul^kM(BLaNx zY@hF7x)(J zsgofC?N8dJRK@8{*l#fI${ z7Qhf_W%$6oW8#U`+L4jOw_Up2@&E-wKKn#nk5Wyptn^4RAD@BplCXY59&hgUBfVML=45@v+pH=xd^l_wo~l zr=3^_J?#oy<8wr79yal0NJ^F!f@D66QkhNT<-)i5OfnFQyT&l7&_m3S;HciN_ppUF z9(ic}qRv=M?EGL`OKJ-GXu>T%Y3V<7hyElfq(j0UgkScv@+kq<8vy10JwiL{NJB@$ zdhZ4Kq8jut#o1pARTBnR47qwEg1!Im*ya_Svg%T1!;;>aq`bC1*$UPUcH?)_yf3HR z#|L&Uzd()!N6*s=S`#r@g&gs6&we=NBu}PiiuNk}w-wAS+i113_LogwOifR!zMV!y z(EEf18u$UVs$*Ko}VD404kmcG!x7hCG?Je0q*&fr3NeG zzg_l;E2Q9?v`P@=uffW->SaR$3PNXWROCy_6(|5 zH?LfO!!SNE(`5z7UATK-LjRw3QR^Oko6?kbwv$nyaGL%qxA4IGv^5LKjeQs+Aq8T` z{05kIB)^*m&2~stM*GZfEmPSr);4{3yW#S0v}lgCE5T%(FYaL1)Hvh6P4iiFC_6G$ zbpC&$8QFJ9UG#eHqxPNEVQMv01T{_eC$zE4!A-&0d<%v6LPyo8_7PK;ldgK&XhvNo zBb!e@AB^>~y9&iTOl}o-f3;Hj;4?AmW#ZTTaLTlFY%t=Si6vh_0F$e1^OMA~th^f5 z!28xVtNWa~L+NvAP9fF2K{e4F*nQk>PAP>>y@sHzs9AE~->;|wCw~eZPvbrww5ND3 zB@GVeM0Sfl=DB91=Wl6=d!10OndZcStKnQkUi0$%RdmAZktz*cbAmKSKI3NP4awgb z^EsPl`L;yWv|I8#PO28OY~Qle>vLZI-Yo;1cUbyB8xV(mF3xeGk4eNbI-o;IHPaA& zXVv+1Fot3OM@iO|_x8W8-_ICjmXo=!AKhe~-%jii^*))7$W&pB^OcLTsA@t!Etev&pXy!SfZ>Vkf}pN+{oJy<30vmwnNKViK1#F_Ln)N_#xgH$%`~z^RyC$J zpSYFkLRH7Btn?T`in&lsx)EF%nebI!N`HoRBIb3QVs3ei()B`x4hM?5weT>)TC+A2 zgL{DlhylKNR8dG8Ob_K|K_Tdq*;JDw-KhZy+ofDF9lxc-mnN#X9jui^1L;;x4I!%7 zwLpq~s34kUAkq+Pi#k_V!^4`5bphoP*3p+ts@spjJhQ|V7gV+Kb&lTefk+}92}naD zJl!ETG1i^4qIYH}6KLrPejzt) zJ~5-JrhX*aSbN=6upo%f?SFrEBz!*K2@%%FCEw|lXhR^h1GhK_2DJ!;aAWjqte$yi z3V@oB+mi;!{XJq~d|LhXm6xpYn@D|M z-w{&>Ta|El+?Ibx!5dZL=lOw4)i{)Z6a=HxuEiPt@%OBE>1Ck4(D-yfEBW|d$8Lya zaEj9tEOT0-xUK2Vi)-zhVr0+SxgtNiMFrL~vfvs)oj_HKm7V8q<=kE zbKJ;}RG0M;9^;2vF1qO0%k1%X70>;X+!0$N%^9WkM+@X}LNZa~lc#Vkh-ix|KS7uR zd<=(1`E)Y_)(fZ$oDqBCb{X0Bb(zYEpl_L|{`bN2ZudmZ{S;um=TbI-dm5)+2`84TxD-7X%Kh< ziPJmYroI)QG$wB?FSTOfddQm$c#yIw8hFN`U;B{Ug(H%IMI;`{5YJZV`8RNB$TC=O zS1gzZ$Cly&=W)AoUmXX0@tU`J(xtyL+ z=cW90SJpeb6&SB1@2PwdMnm1bipuq zY3bF{986rOa)gHRD<;x)$zBLHVG6Lt#jgqbl(r1ehU;^=_;o}GV}{&DkkVYn719oN za$cWS5d9h1fX+hR_VWk7jAT?&AM(f8EBpAuwo98lkWDxsJx`;>Ix3JCh)25IinU&9 z*eK9}=R)m!g61eM7uMz93A*L%LqA~+w8^HdAP-7mL8eH|ee!4wcd>V;a?j&N9MH(U za!5>OGI(Zz(wF`J`Y5tOQqB=ZgP6ePL+o8i{BQzJ;o@qp;-ru!mZsCyv<{3>LS_nS z3l#Y5i0+wf6G+~R@yb5*zp_bwSHvL@!M2?EjG~5!T<t1+T zFd$j%8j^{QFIut?6-}iDdSa&yM7JqMN6WanwH4X(EW4^tTFBw|?-!w?)j*;9R4<{s zTPrF*J2SxhZC}mg0N3h8BZit`(`XxI8N@bYWX<2>Y9~%w+)Vndm8|8H#a&p1P2-&1 z(Feoj0McBf!9JUU?q_G5aSBN%ksX(5^Qn^cAr(}_T$6t_WF#AZBoTniL^UgP_+E5` zH*bPoOFKC&Z2yyUG2u;gwTg99o@ zw3w>X`%hAakWuLrrdWm;MCo;`*b!S%7we8jI*h?f`22TW4YgdF^>lNqRhcVT+V7ZI zKtZ|3#80pHuyF?eL^#n|q8qXd^RsOPI#wTNS_u>6xhH^Y8-sk1&-o}P_8a@FH1&EE zF-9p;^*Uvv8IT3XP{p(HM}s3RG&`EquHLYRzD`UzB{^`uGU?y``2)6M&Pv2a^?R5WOu(Vpy}&0h0D z3AncwIdsA0GX8RKmO)(L%CUZ16QKB>9IW3c{D~hncv>Y+nm5@dRY3XC68nemv*%(X znqfoNfMWf^FfLZTHTQS?AUQ6ES^f*GA54FHS6i6AZEqoD0I>hypv}~56#xr7$=L9_ z#r%N#PY;GvaY~k{`i{cUk-Di)arkx-yu>)2rxj7-(6~a&v@g?B`mM0NIm4XQ>LvGo z919MdPSA?WhQ*M%LkH+03d@{P&2$|1!xj4n-jdj;4?M<%LT5Ez&vHjwgbxdy890j* zn`G4hHL^-1yz?r_;q@B-IVL)Z*x&(6Up>On1CZjXK7W2B+q1)-df{P^3QpEy=%U-75 zd%uu;e=$jQ5wfPp)&QpnkS0bolv=6ovF5#HCd6L6j*Zkh#Hoj$iK=`ECSP4V{1Ngi zDg~()Mca7fxuWT7Zh$j|E?^hDn z3F$Thh9+qRnB`5N3dC@n;W`Jb*s&Ss7`No7BUSOZG!w8Zur~(%-KAMjzmFTl{*<6) z(%by4dMN5$f&|v6t~;`;%p-pRKcb+%L_DjXUQj%flhS4b9zpproo7;%1y$w+yY(@u ze_tL+#JpXQC!RH>b!qNdj0?S%Vo0~Gkse3Bqgpk5I$JRvmh)P=TA*tCx2i_&>p~JQ z)Dt;Iy&$`b=TP?1lF(#;k{<;kiyoZfu{5rBGeWa^u^I#%v*06hIa8pDNTk4mQC z^PXoB#f0BWW$O!oG`tdv8)p%P8-AKnS=Mg* z{TSp!7nj$TX~2xGoCAzEOjCx0yq*0@Jonlt-Y{-B@K118l!)Ga>2k_ZvUE| z<)4%4N5zXeh%1r&Es11I{Tg0zlGA1yRn(eqT;(uEy3HfUjMuBd7&}f%3%1=5=1&!5 z+sF4xkR$^pngF)Aohfn)H1o{fY_Hw6Ixy$DI{W!OvaVbxlL?McD~Je@TZ5-BB_W$O zvt=pdO3(Bqf~8`e)WKv&Jk3)}V?^5>B{KnAR?m{YxU;U_oCT}_f2<476@)&zTHs4e z9P!mNg|Jy7F?BL3{`=lG#W&R!gZh4JW4B5<@V12J7{M7@+?z--8912L;(`&KgdrVY zS|Txeqkx)UV4xuTMFr1<-=2kEX37)K^%9k%yit@bX$__}1?!mCB?fX@JTcBTNTafbdTzfkd zyBM_j?dQL*e(x7sK}%g2FtWX7bFq$#{Yk*dm%%zzMAwTQ7T~zkY zHtH*e-<1N=)L*@>hH6I zRWLb(O-7Ehk8}a5o_?Hm*uwWKZ#nzs_)DEpqlWA{jFqo*dd;TO7ZGd3Dav>Y$J2d! z8TLGycslw^g9`(ZT-21zlGZNB-!6(hd1}o4WS&*~XI~myQ5MXTAWvr)A4hb=@rc^U z(Ee0)mIg{i|9#&cy5KA_iIGbw%-<+K@iHnBmnz-RG@EK*Vv`Nt^#jeYfF`6skhN~@ zicGa9VCBERlQq$Pk6Fl~kC}P9>Szg(Ll#g2yKa)S1ru^(livGPx_Va&p8biqTsSP`u|YuVwC>OY5l6cSMaz<`gVkKj$m?${O#< z6gH~RI&0J8P07?IK7X(-Fdx@6HSBc4M(LPoPl@+)MW!PBChSmR*-TV@BN*ntje>YY zD*sW$84rJqpWUTSnBDa&y$oF_)eSZ19P_f$Ux98%!7_dCZ}`iJwB z(SVSmJ#?{FGI=9(dfgA32d{~c62uE_C~nT9R>Ij=CUB%yN0N#Zz~!qKE;fnMKuD5E)zq90%*u@!h}iV~wy~}C*z_Y) ziBIzQjGpc@i;|E??DiT}CGGn)>stREv7HA=6hCI;d`E%{@_(R=BRJI}^r@ND_BQ(g z*m6lAAXAp1noHoy!$7rKwwszCjIfJURaP9aUd?<53kHZP)xM6op^vni^jvH%^Qz0# z$dR-cC3G(^+Vwm+lzc=f9#ha&{P~9q9ONq-Le_O`*!o^J_MJQ;Nmr~_r?%8}r--mz zDjKyjunKT0hKCb8m>8Eheaf|#p?KRCjZnGwgCGFAdM>^hge3X|MB6CPXoNp4eEPuH zOFO=Aw266g>+6MmDDw>>&=Iuz-NgP7K!pZe?nvC-DkC5r^O&tL=rF~fl%a}TdDTFP zH2bgdLJ|0ONj}S83jghYL9kiT2jdL&ehy5^O+;koB2DD>KV-gtKkEla)mHkE@;?%L zVGBV;zpHR%v<hIlDsz!$t6*R-a?68wu*e+d7s9qy7`#It~Zi5P0EkWOH zJRtRFQ!F>!>VqJ|f|#;Kaa-6(6g0@yD#DNabwbc8kfxE&5{X(mGvS#Ac4fm$DQRMk z@mg+(;7Aq+NDZAP>89qTofSwQ{1rMA!>GcarQ<}-!YPiPZM8HDZwqEWo3%VNQPK$1 zEKX-rUSdiiyn}T|`#sOR@BLfwzV}us`Q=G2`R5burveovkFJEkw{qxdsaGW`K@2vP z+?{z!ghs3%p~!%ec?E$ib1B}GZE=WsT@~KS#6cTQOz(%}MQ|X!$jp5COPliZTo1A) z+>PVgBKc`CO*nxAq6uep*sLwhPFjOD9KncfH0^y!M~QdB8435Y<-+sqscgIOlsPY? zdCf!ydODoOcqEzk)J0;#c*6)JHBtCS{Y-KjU3i1;CAy!w1w<%g6OforuXum?cwzpJ zZm5p%e=H%Ip_Ttds~6P0Xx{KObYBQPP@6=ifNc{#iV6SZK()^+BH`@ZOFm+Hy-B+W zoRr2BB?@@(w4m|;Vp|HvQvsq~ zF2O6CNI|ltQWMT)Fb>t#28M;LxH`t_yUaEQHS~+ERur``>iGb^F*m>DY$BJGuQssf z`1C;t>B!gxz-);Qt$X`1&Lr~HSW|>M zyaM^xQrLgJa`8VC>IgQ7x1jBM{#PIdRP?{1*8G1|_lH1vTA-TXK2i5OmJK(iTo0Rw1Hmln=NFuYcq!=NfIhc^ac`;u?t?`jLC5%EAe<-i6Z)77i}!1T%G@IiGt24 z|8K|lH3d;5`P`TMXT%m+?X_uD-9aUrhSY*#*}EBS!(;L`MT4?>vKGF|n+$mS+?#U& zpV0fw1{HujgTqp$RXi2R!d9^4K$h};ft4H=?yF`lnd;-BJ0E)t+}Cnzg;nTpN*pCJ ztG^WHg^HyA5rJWRWgO0bE|B=>{~jubjYo-Xe*79nAskcPY5=!j9Aat7AjwpX$Iba$ zbvqGOsG3dsSJh9HTrFXBU?L_a%I#*c)k3#fuP(;!94-T1&8(a;N^WJ(DV!i8CVF4f z0UTt1xEoMh$Cpd%goM7DJWpu;{p$;Ul;5nNBz!Kf?28P^?srLTLaXC7!6V*B{EncK zU$9z?M)>`-@wEW-C2Y_4aOdgFk7~+0-($eQ@_U5b;k=6Zvinl(urQVZ zX0D&dXf29hPyFygqiWl?h+b~B(>+MsM<9v3puvVax>O#B77WabO7WH5+4Yh9LH}`i z4rH!{Cp#vT-NK^VukE~wAcaxY&D0^n-Azi1w;(0gyx2pf^pCNQr`nN5EEv+Ahd))Z zhSWO}r!qn~cJ^XZNX`3G?f#3l4Mj!emm8HC)65zLIXoOvex>8{u54Y|7d*GqEYGyB z;rj(k@rtIV`TvR2J&zYhu~}LwMf{Ny*mK33vQaJ5(Ld4$`_75=H=iy$+|ai}qw4o9E^o*5FDMi@`GCZsDcRyYYJey~_l zSVW9H`S9ef(L|wK_wSPK2weF$;^5B`X^bDq)AE)l7&lpjm#Y_CV3%Bkjq-El3iDn} z6wMDYTF`{M%!UJ?QIoQ%3z^hHY-@?ULNve0a$}pMGTk$tX@rDx-nyShayAnl7lz_; zC)u$qKr7Am00s&aiZSDGYC@UeO3jkiaLNkQ;$6hdP6w};taxv4uVZJ`e2I4FjqAJ9 ztv9eyzPR7pY3G1WO+A96q( z*y4t;Rz7?BhUsPwg`FdvZBdoYT`bpRj3v@srq^I0#69Q5^bakO+RlEKCuH~kLE3AD zFFH!phfohHGC)@969s@}2+^5b^dt!gpvvy)ChM2I2{nXdnS^@}QvLggHKOk7)dr%hv- zV%fO=EhQQKyqQSy@HqFWl_)UA&W>HosMoJ1-_Jf_bK%u>F(r1A-0yinmRYQ5K+XUIs$Dt`4&}XsJ0zV01qr=pgr@- zJ};>PHj%T$(qdeP5CeDjaxb@dCqF_-bX(#17qL1JaMA%H4xh>|%IEX1WLjxN0;E
    1. 2bteqpjCLk8u@}Q zTBLqXb!hi2Vb?$epTd?2oO~N9N$d}gkJF|@GI5xZ81zdtW1Grr2X;)uImH zJlu;?%&M1C#cdZPt3RQcShXuZG4XzYgQhHu!8GQRuh@9^_66z>A8#vcaF(HC%2a{` zV62#~HQ7{S)T16;N+ji=qLvum0dRoMAai5jpoLfB_C(|2dfHI371jcARK&((4{nQc*qF-AX+jb-fW+~zv z0RKqo{!6XL#)Zm4a~;eD`C_~Sfc1|Ou*3j?zsCCQ?Ivg@B)6?~{=3GBa9k1E%9uP2 z5ayCV3+{!-q;uJv$~U6c4P>6sZ&e~MKZ)@`27*$x%@$QD$FQ3G%i_0ih(pSxmy1RE zxqDHpGi}0Nz|^sF{CrKjvF+kSq-EzE9+HxVB%!2O5>wcPR1nG7cw(9J{ z3hbu10Koj{FyQu-6!g^e{@%ir$1YsnM-~&kb|9X+fN!P4v_MM+6l@<4z?HJuiG>_@qq%wp3?yXod z*&tZH0VaO+4~S;vToy8E6*(^j5!3w)y26#Aw9{E2H{02`G3#m&nPCBXF7Lbly-F8G zx1pB7=v5f?0T1>rFeC1tyfLx%)pfoYksO3i#Xf!a;>m>;Y)CB;OUJoY5nv_t-O1*X zDFZTv#;lH$r&yET64lB9r;e41cSV;X3y#hn7_+RT7BR_(Jm=&bxiAKfU%}R2C{tw> zSjfx=WDaYk@44e==TpFu^oDE3g6S^#?{XmQ+~A1$_WvF$9OrAam+V2K%2L^6a@-+7 z@Lrf8b7p}$GY@aj#z0V`&`7k)tKTs3N+Is_;^%viN-ys2_;9jOz(@7SbTLhNCS&~- z*;pcrhiIxcR+NTYFt1o|u}X_(wKATRyO?zX_=Bge!T+6%rGy&NpGz1-5nt=FzC9no z9k-lyq?JSr;}A8|rpnr?s-tw~5CGY(G z(Lhj1IU0yWfIp6oJha~qWAtT}@A`Q61S1_3fRB2qS*N>wy1kkjmA;4&kByB15vyTS zb*{X5S2(^4yBL<5nwo6os=2r7{}q96)jA~UMM9v65=tCCst2K`=r^wrxoiv{z=WG9 zK^L=`C-^aMEt7{@`%!qR7O(Z(QP6xt%~YKcr{Hu=usX}041%ex%mrnK?PQO6`9Qw? zO*2Rg$_90OLGL7xS?=noOQr}oq&cUdp`ktMBU*^I6>L{;l1IPo=dcpr!A zyG_Pl(OTvc7s8JAe25G_ybcT`qiw{0G)5Y6W}zvAU#TI@Nt^ejbmCXJLEOF#jwZvt zTKz>{S^otMjusp0&%|6lp{SabA2xkBX)cCPG%;GD=! zL3AY6J+=W3SM|e3y7F)G)SqpyB<~4p8K;Opxme(^*&Z_+#nJtul?&o;X0Mtk-=U5j{b4AYMv<=w}Ef;kj6c+Oe4$BUBZE>u13 zlpbX76DWg`ZiDYXt&G)5b7G2_I!_yU{zitfSUcY4#TX8zIJ@^V7p?%s zN{9IQ0{wK`Q4AKh?)y(d|Azkg@qCk4FkV@O7ZQFVjBm{8tX;T9ZIo0lS-3(HiV3qkXa%|kVLj;{(oc2e;U z>t@m$wQA~9t0Irvd)3Ohe*%>&`lfwGQ0cOBHYQYut9=txS5BU*_{3~n3G5KJo zpw>I^`(OCRo$guOtY%@R?Y^3oAvblnKyeOJOfskusqaB!>YruMPI9Rq^DUVUp zQx3E*$EoGe+li3&R$X8|0br~bT>?vl z1wcYZMiYs-mS83iSn@QK|2WeZ19b|rUKdyRLp@jtEg)WY0`TOamXz_QivQJ^e&HRK z4{Ge;zha#NhiHB-yPh|A&Su&6Hb4ti5_*>r+Sam@ZnUnvs!a3OVYAW9@M}8}O}r+5 zsY)`#zmNI<&ODrYF?uX_8$yiAq+a#w1kTOE|{F%@j<99u7~knmKm-LeKTR|*2BmP5?Ht=Ep9WjsJf zLDvxd^cEDTtVl8`G5`sgXfEbD@DM@}3Df462CSf>+D&KcsrKP^~Tl7Jyfo;-}pc{F=lqbRAj)mWcxDjUS2v}BjymBTzx7Dzl} zNcpdRHAPPMrE(DolZ2WfONB&x!4%PVpOZ>G?I_ZDmq1ZZeZpuT2)$maQGW1bkT<~M z4N1!d5xjOs+RM{#1HdX$=x^hiB8i4kFu8VZ2a=~f@v)@f3oY_0Eh2n3LINAwKdXX1 z4lAx3K`qI`d`#L`O{phmf4Xkj-Y*v@-WFBGFn{?G6;%yNAqW9)`y7xi6pgB%h1N~Y z84xvwxYLxtFXVoL!J5W8Se*0bw3nt`a2cg-;F+0+q!I>>YAIR!9y7dTc=!xlD|hze zc`lPc9c`0x!$BI4l+l9@l3zTOf@P4-+^GbxF3bUyWTxZ$tuc};TM#!5MmKSQbui7_ z0R8q(D@3(Yzs>*0+o*Yf`}H~f69&SB`ju3>xR8hYDp06VX0@p5xv6uIyN_TC2f;yC z{X*tz35cT>vV6k$^D6%JovMbi7TJsQ(k zDrOATA$2=Kc=S5=dV`siY`L*B#%M#uzCoRzXnZPjA#>+2g&+Z$!~&dJvp;-RYp@#1 zj0v~+=9tnaa|o;dyFHd0^TU8NyQW);TSR5AP3AF~I{UYG`QSA0F_9%AjLU{(A?!Uc z!c7QK1p6RuHSt4K<`R3%sd3T3UmMkm^$%beK8yG#iAA@U)r(`9gT+&UPF)T^6PY-0 z5k)%_W$zI7EHK@@B6;p&1ya*rOs1XaqK5~8;kmP5W{6<>jEe)=5OM!wfnL&%`L~oO zJvvnr_(phIX|A;iwUfoNm@4CH`b}IkX>JRLAD>VHR< z18RQ}jU@A&e6;`+*IZrnM<$Fn}e5yOE&AkB5OdHG(`JT|rRr6ZwwWIL=+QD(!Y=`u|Id^Eti=W~^v8YrB;ZAsk2*f#HIBU+D z+1PP8zMjBLa`2&uNZ!H_mQs}Dt}Vx$%X}t?_$n1oriW@T%OcRWW2P1GTh$=Ek)^qt z!Hg2$>2@#jPdgs-T*BM8E==7HrUJwvx#9RVKZ6(cPZ3FmI8}y8TyGtZP2gw!r&<1ep?I(vh5vQQo^qDHAjBV4_*jS+0-w+35 zC8jtK4D^{|g)76QpZVYM@oU24PY*vqnBhrUyu`?PkLGquPsHo_9hdCMzEwOyNcC#a z5KTXb+T1Yr{-Cmn==&})^(2~#Gc!|B;$`NSUylY01%FsNL7 z3)YGud|r6}lRl`V0`Sm#39vz)wcEZef>~lqs^jeoJ?E%nU5_v{o%_3X)yuiVPY|56 zJbzOXHeK4@3rf&Sby<)#*?EPAr5XW?1^?;LPGzONw8K9zj!xU-6hN_RV|V)Bd;JeS z+hs2cR))Mb`%jW|QRg6hviZwg_BEyE8VteIr0(sf`WHbT%QdS#SXqqozjD_$qmo1F{zc%_6O=YugFC0isCJfA zYaf|NQnJy(@?ER0ullOw+tIvNwX|J0R?k+G1gOluoX6K!8z#l2g?ADo3b(Sbz<9&! z4|5h@t*y(HQJ86_3OEG_P(A+x(`8E@nUmO zz5-ID*?Q5byr)rTNI)^5q|pwGJGZA*jm}ARTaA3=i@H5IaAq<|tpz6C=s4QoR2MQ2 zNn{p#(&quUSgSu=sZC6o zXeyiZS!;tzsU%g~{mFr==ET^D{N?-MsJrk#9K%gxu9^y}{~p9*h!|tLJ+e-rb~pa| zWgELen7F*YkYb}li^8uI8&QqzJY%4WVP+KIjdwtBQTZ#<-|3A_{cHEGWN3kAZ>#_; zDUOoz&0AC#*Zu6yH2(UmK*+L{`AyOBjpjO~rL{FhA>i^W_2_MPr3utw^ZwXwL3Zbj zAPVds$f@vxYjV0h)yp&o1P;dcrF>*f(|K<{zb@uCnK*w|M-Cpwr$T#OY037!RRxjU ziH(mUvM`?yR$JaDBu7;xGM&{Wf9iHw98AI4eZVJ<_Rb^G&q53JEFZ_gsiwTTe)jZ> z60?UI_Q2A9sv7sqLzsB>yN3J%AFC;}y_a>SLHh~_ShsyqH58-LgT2zb#?Pr`vcZ<& zZzLt7m5%po3O=6k;VqxEp$aB>8Pj;mkd6P0KW<|(nN2aDXJ&NQu21)v&vRJ<@ALte z-LTM(F`->cF|w4f1M?S+|Han@J{?nlY`wPKsZm@3V3Ek}?ROd8o2fljjACN>!mVta zje36@4eIS*Dl5nCgdtA7RvVqBcH2Nam}1R7NR#8$?Ffo;T>MV1rch*{I%wG7_(eD% zT59uB^AEv|!{Ir2>pY$C1>QstegWG2#D|%2nuk(T6(&xO(SycXwq!7EHsYUntY^_s zIB`u{%HldfV97lH%7Lgj6tabxVDfuWy{?*2B_9#r6 zroLC1h2S0@BCzq;q$dh4U7Xms{Lo*$f~l#FdR4TE(bNK0!2~iOh`y3?o6^WhO>-6K-E;XipF=yO0t6Lasf$dx?%>RF3++Up$PV_F=p;!oV=o zp3$+Iute83u9Ns9kM8$WXP!a*oPS+xm#}tjwxu`-1i}RZ?Hl!=3lNiJ3|K~_dGb=% z3!;M)3z;49A$(Qs#1|{PrbvO4kFB zeDL;Zk@DXk=f+}ypKrx>SJg{hlBn`E5#NBdYMrjviWT=3dbW54at)lps@QuIuAKY0 z3Eao){w2v4ZQeUUx~Qo>lcqQBp3>i&sg4ZIjtEZt zb*Ho6X&LUll9xxRcf{}795VaacgWwFm}Nq!Uq2}Qk)S9dW@z-yG|mw+Tpj@adXY?S zByNYd!SUtS4W#i6#Lb5R^}sA|!=LO%xtI2!xsk1e(#%$yUcMs80Ubl`X=jJUqd6q~ z3ADzfoUIo*hn$JUGBPFoU{knnv$drJb z;{J@bn0CsZ-hIndCR$e96*4}=3oQ8c^_AV_x8}X;6&|@f+bKl?t;#Pd%?YpIjO^1U z;4R<(S}9WdMPi9do|HNs#~K1;|RidQd5i)7Cv$ zZe}DaYsG_;_-P0jp*p}OM9Yetw0z{Jn(`pZ;c%@!!1z)J+&J?}9R>hv{AT!pg$4Kb zL}f>Mkv(_nOhFNZSh83Epa=79X^f~X25m*O(O4aVZ0N5J^VKn)`5r66t1Hj3;6iUY z-v+$q4?wt7&FSy%Svl4X*O{P?Dz|8gj2cLUvIx}TNkZp?gKS{`Sv0}W42oyLg2M6IZ=IBPh9_e8y6fNvr#q|2S#w(AZ6F>Tm|8mT zW{&EWqefFMB{4KGF}jQOi2OJb^n4B%M9fd#iLPdM^~7-4FaDt)xs6ZVdC}XwVH0Q+ zIx~Y7`f)32U)?mvhC)<~29O?kLpp1TQynzsgmb8n%l)V6pJ&g-8r+Fx>3LY3*MENM znO*a%FK%23#vL^65nTRYPIpXzKA_E|k^G~c#kHjopY_@3rXz5EL!r>({qwGA7oIZY zIC`kIm{Ak>aar`4H@=R+w28jV7}U{faPVZVn4hIX&m{M4I-^?den(_l{C5A+#TYc{9oZfix z*TGWX#?6oO3_{xGi}rB!Wf1_qD0%p1HSR;2}LaNU~b zVA{JT#CC^r_9HvtGGFgIvd@|#92Q??n93?sL&XW4ZI#&XDV^&tC79Bjx!MaOsnAWc z*hS{=xHIwpkhqo3Hbf@-iA=rB?B}S%(AZ}lorpCR{`DsmLNK^zT@my}bf4R_#!j{i zCeG5f27q~gZY!VH9+S05r*nKu7@A_>Nq zEKT7P55UkPp{h_X_@Ak2yb!>k5is*4^I811BNFPlt;^^ejQoPY4*!%MAtaM>ZZMT_HLg=jug7uOM`UZozBbl`zU>*MV^r`hCQl95LgK09BaHTNtBoNpCR zOtVLBp@~LL`{$fV z+VV@?h|%oLn%9&t$sW>I=F!l|#f8sEp9!1AKN%nyFV}u5@Zb6WE`yAyPx1c|g7r{f z^PuKa+DdcC6TD zH>h!^J(ebDonNZetH_a?=yh|yT@=>5;nHiWC#4w`Oat)BjDF5Gl(t)|blJkZ8Ou5u zWEtGP==v`RzX}n@RcRk}@t91~xK-j~AG2o1ADVwRl-v5f)ni1ZfGwr!gwFaZCy1MW$GcdwU!qtCZp- zlkJcNF_3uddrD!~z{pA5>@6Bk;e=p43w6(yB3gpH^CcO*6U#n*3!NZFo zF`y4X9DaKr8(tibY^Dg5F2qB+21 z=yg!*04TiJr*jDy=jAa<{?O|Kr z3Y+gv^YRNdy6RyY*q;8Z)O&oieh|ej8`<1&Pl$e`>+;?k?}z*A{49TFbU34L3|q(= zuN_xbqk`t0u&DjXfIl7#$#sXLQ+e92PBZbJGEu6U(#iM6)=#&8Qv;%1lqZA2kG^VV z08B9AgqMmN8rr8_>{MB}3v#Ms0{N^4$u!0*cvQbZv)HVtt4!>@X9h%)&=jr!mJ?J( zKw6=a$`e4Y{cY~!f*e>g9`XjR%SiosN^6cAU#EfB@!XDALk!2}Vd!b)8H)@Z)@NBF zTbbupCS$GnLyAFtJ2VDry7HC#vr;D9LcU9=e3Em2=PBk6zaAOuK90}P=4|@dgI9CV z3kiv>cLbwiC_Kt~^W};Fo~k};foOx73Jd%R&}v!DyXy4`?A?g+Ptq_m-^rdpk0d`= zek7>tQ0k?|t@5;wf5LlWqvMm?;3A_|oZi?d^!c<4E>2WhTw?-1YG@;&qH>_N3zlWj zAA8_2u$Z`-bS|w+7f-Iv_WF0JjvfJ#dl6wOQniq@N)IfmH&u-O*SOtd;&OKqhK@c< zbq5I7ya@{NpLeq#Y8A~b=H*G?TR)=dwEE>~U88Q;k6(#G)K;5h)lauA2fqFLG5HLr zC(QT7u zA=Nh8FqI~*=e5;j10-4@rP5=i!pMz+U3#JgCc~;h)*6KFTm@aFW_Q5dF^$Gikmch6 zRw-fSczA#YIB|C`oe339G|yCqRBG~%b&Y8KzToHdFKj;ec?8mrq^5G;%lIb!C`frm zJl6q7&HM6nY-A2mo0|G^!+=M`B2Y!)4NAF(Kc7$hW|&MSN0nO-zzJF z$>|_W=nn5&Hgr znKc4U0_yUOTFf4n1VL;*%#D<5W`N?5tS}N)6)YTj{Qbfwv6&2Gs6?j{@o#`#eOJ%L zwDs=FZh~|-#XOeVvXM1`JO97+t~v`*sbW#vkkk>B+oE9Zae#p&X1yQA-X-LX9*e9> zfi>hIUM9|gqfA!HSE)TOiWx-FX`fULp@&}lLe1?f+hGq&0pk64=E=#Zyx(EH`JxrUby-gwYIo`^A;L~B}a7v5?0%8+`)+%{dHSJBgm-d!V@nIU# znH$!bf+O&R-z^|q=)`SAzW-U*+>%?Gfc=bV&drQz#FmM&BIX${Q^hNpw0c_;R0OyRvLd)_Vhk0X-%k!Zg~F~pEmnSYlYBZ z92OCs2M6ec4$tt<+~ipv4t|o(Nz3z^f|M8<61Wp1Ojp9?n~?A>a!*47c9vAF7pu5p zcNc3&ZwT04k-)S-=u;=)d_v{heQTW8WY-QdfP-DO{R${0#TAf+#N=bQyw3IEwr7%OvOM$34}Bmh-{HTS4iT# z1sOlzMH9G~bYmX0Qd;f4k0Nylt+t*2AA3e9{RjJQsBv%*4B z14pOD?fCTRt0H?aBEtj)lO%FWiF7SCeicY!q)$q51hvh}6D zf^jXqgSIr&DI&Xbjg}&&K}9ge@Yi)G$I&KQpHaOo%nEFBb+95vu$)n>tf2EJQH$I2 zmAPzl^*bG{_c3Z@k-x=H0*lh&MC?3XSu>8ll25V#s7+Xk9K&=!^Yzif6nW`}&OcG? z-*tc)rXZUM=4^dlpeC#WbmS3rSBG$=>&o!mTE8U*{4YZwg7aI`;^GNq>SjqaJgGzx=|N#3AT@&J)lz`2O&ct22N zzkM+qgfo}Swx2^O0>&_zsbrR(VE^yCDPVXZ3uwo9#Q;;EC0xlVBNwMmQUT7mRKN;5styjxLFq^1qdj~ok#e#nK#9eBb+CVn{tzh;J7sX_Mrj}#4*_7l6eAvqp z@l@nEX@^4f^LxdVjfV{mpC?ky2HNl_-_gHIHJkhz!p@KztTovlO|Oj9B`YQxnIl6& z+NVtdV(RRspZ>O8Xmsuo_T8_W<}&L?C{6sSc5KIwrA24pA1OW$T3eJ zFnkOpKViEa%+#=ad9C6&D|G+CfB=Xo@>{w6uG~KFyMcsAonlW%Z=f9Z6dS2jX%QjA zn_Ii4JNeun+`0yWFk_7N6mObaEHNHh`kqm59s({X&-`H(PpVnn?~0=T7O6}z1O#*> zT_>1tQtfL;ylzU+D2JlvNARA{5pgRPs|8Q%^wV85Jj@}pDh?!TJB83c3{tI=Q zb74DekRhN5E7}-vc)OLbHw*c^e~<{V@F3pTgP-I+at9g94h?#~6?yLwU9uF~j`Jx( zgk-|%1Fui4=})zarQk?AE{2Z;@=j`l5iLnbmd>bwu@W9Ua*L7jM z`!~L+<(aRml!D^^@|XXvGKolMik^pgOF0G)Wz6mFyhf-8%(s2olAOgW&iEnqX}UgE zh_S&HNVAeXwa~XsB!=uIx z5>^ZGGqQibKe8RzIdnrT&66-IXqGItx4Qr&ji!o^E z;Z^FcPlOdj#X#ec29u){|7oSmJpl{Bk9KjO71~hK<*3N;NY%?4E8)jicq~FkKG0Q3 zgg4MxS}#XrmU3COUoHaeD^UNL6_7Dx_4^#4&9|YbTNs~t%u2{ijl_TP ziJS}?lhPHmGk1ZBC>Gp#7{8~`TUvl$9e*9`L@Bzuq$pH+>s-9f5O+2|SEpU7&-5`; zT<(j_j(T(SD=qCQ$p!{KUa;8T$YeHD%lkYqzB zZza&d;TYnz6f1@&o%8Edw8`sjd%iAm1hK(tVyC|}3Zz$3xXml=ra49uio?)Uw7O{J zaO5pO{UP)_^vR$r7JSdu2g+?%Y*8e6u-~)<%1ftAQA+!3K)3A^Q=fs7komqB{&(5p zeXI!A_AX)xD5IOkPZmNI(bq*(7KjEYBBlJy>8}c7`fUMmu*h?E2211*oj2457=9*x zEzyX$eRHl)`Yt0Yi)l5OVumS0FcG}q{#dquB#j?;vH4tTdU(GZ_lZjM=^s7MO#IIZ ztn9txbmSC4!S?N(3r}o{rhmT9X&hCK_SiTanIp(6OWr!1$y^CR96E5lA62cjog6RE zN`E&J6A*ecR!Q~HiplSW``^h#{m1s8cP~tS`yPameGIUEH;G=NQ_=I?onph-Ep!G~ zZKi1g$S|LMPJq%okz2?f#)Od7eLrRtQXE+_4vSxV{f2?~;e@Ts38BZ6DYsY3UixSZV}FJtHq@SfCERaO5+jP^v# z?+Zw5Ei}1~mFn?=`SN^?wAx+x^5=^SAvVKC)x9(0?ML4sG71Nz>4tduB)t)zoU^JJ zjEV+Ul#Q7TB7O{KQ91O8w|z~Cds$|=SS1_CLI9I-mNxO`I8H&RTrX8I5AhobES=Ys zwY4p;uHGJ~5S8%omidu@rP>U(>Yf^BGJBI;W0kL9g=Nc#H$m73e~Ix-b8WGP$Y={L z?3(NSDrXL755Iv94&UML!}QY?d-(~|Cb4na#EmWBjK)g=drR=2= z_Aq%$6kgBmQ;Rn&v=;4`kHO4x_=9I45=~7{f*_3kiW4vXgd{VC0nyeZ&L39&Zz*$D zXLwgA4h-i#Q~go~uA6KETw87cike?UPA@B;tx>`x`m{90e@fmIPc z_;;^5R{S?NM{BV+Lfhpw_jtap$m=~T4aWF1+SuzL|A+JM`V-a}Ap?Iu#caX0rl_{P zKkjGcX=auht-f?hlH)GW?PrjLFhc8vaGTZ9s@STiuJEa&;AL|&r$g<*sw&r7T8XE- zN%@*Il+uMtav_<7w!PH8rPh(^T0;4#v;QtYUeeKE8CbAPQW7P44d2}ys~7@x zJ>wN`1pXPiNKf>I1@l{x1}s?rV1GZ4TF&59PMfsBf^+CjL3whNfDGr~&+aFB9;%De zkk`w+298fDEjCvI$0NAwzbG=6^fTNOcVT(|dre|^$qc3Ic_yZM=dtljoEJ+tLi(ku zXydE-jzT6}Ghu~Csl22ze;*Mwc3BKat#576wPllCmaDE`H@y1f^sPlY8(-b^dce;6 zdhEzx^I-PI;!5Li@ZiIz)E(9~R>|WO;N@1?C?u(WrLYN*v@sT8xjSVI#qzCVTVR$EaITKbaxv}L(J{}D1Q+# zgXG$aS1x#{tB<3~yjcAqzHik!Jf5!*BPmVrTsBFk7W$77vBO%?aunbH;%K325Q+V9 zL>IvE#d;-EVk;$6a`h3W?M?$qJc(=R634|cetTA~Ef-q9bT6Czz}TafTW_Q!sSX2P zJb(0{a>s#IY07S#gzO0J&{LZD5Jx#o*o9D`daUBZ;M@2u-!fCJcn(h=ZahdDT8B~< zeIMySJ;~Sm>`xI6lXEqndoUFKa0|)r@oHpR_7YGqC{XZ#RPq^fxY+Lfv_&}a@^pbx z`$E%vmHjMrXxQnaHHlx+Td*It{r6QTG&I68(le;U`98w?wG;nJ^9CK_nc*_ z&cEfx9@+npr0fQMN7(EL$i^Hp4PX1^YEnDWbO7KOJrO1>`c!1pqM0-V3Gd&36QO{! zx}#o}c|8nHT@F9K&W=gJ)iHyMSq(fC#469RZ;Yz>ZIFTCX(j@5)p)e``GKd@a~{a| zXM(Io+6oB{JiqCi@gF>RsT&~))0oPp6(qm>C%CHh@`83<4o^0S^^Nh6gUR$sXTo%8 zNSg{Z-4>Vd$d|Vqygx_oubo=`V3M`?YimYtAB49_ za@p&`F231RmnrPkKx_P8y^63`c$V4ALlYE)fsIrBxh>$+ptTd>HMNkiWtJVdBfV}n z+(Y#qvt8>?nU;)u{q_v*FuKWnAFTn-sLem$R|au<-Z!!6SXcw@6-Vch{B~0-fLX+= zcd^m;iI>6G+Hq#eZa@hSCNFR8KAM^vSROU>> z!u8C*E8oiV*Nx&1&gBDkf+oq(U!UwFm^8!Q$xwWmb=b2WlrSJh2$N8{S{HSLq&-!c zk~wfp|9?lc+-Q{FhqH2BQfeISGAMGIlnnJGp^e!PL)ri!_Iul5AJ@H$7k<17{MHrJ z`Vd8FL$=8zu|S=X4f5s_izRvdcllV$mfm@igIOxt&lQvhZ)h8-9xdS;thrQL%s#~v z`#`JTV7oE|VHC#qWO>J@Yh^x8S z*ApI7wPBH%7DVWQY@`UEJDOR$0-tKZFbE}8Ol{JbvAKH$?s%(}705sa(oM11GUi{P zu+x}y;wLmVyGc)_E>d9myY#?br5FCBrfKFy6dHNki=hp={acZK$=}YoBX8Tnd!rzD z!%{)xQrvA=BTi70sg|1|p0Qcc6{v~E++x!M#^!!>(LPT)UV7Yl&%du~3X}*ZTOhC- z^L4r_s>*Yf=2jJKn86o&#_4duZrwEY>p{}=R`St}r;Xo@ak&QiTdWrElthy8Bz#mt z_0(6bp85{#JdhuQ-4rGY1aJ63+YJg)KKf0j(3x->JyrKt zaZS|2 z@Um0MXgVBLa^e~0e>vzhZJOde7^rv%m_k&x*AIS?P_CFZwLKE{qeKDHI<|g@8rUIHEbY6+38tFqiXuU2^y~xmuBs$u@ zpaCS30!o}|(MN8ksriUomsMASc%IR?f$a~Xkb)1x-Xcj7-ZZRtLt6~fV?{wqGCS|L z_QRTKF&vYLqZS>755&%AMSsg@Ky-RU8p&ofH2UI7m3dBlp-@st4W;G}!fa#Iv33rw6_a!5J+{X2;mZ){`Fsag3ZPK^_9vmKGW6 zKe!ZdiZh%vf+gKfA-|9_a4JwAr;MNeWTG!=g|7nQM|r_Jpc?N`!sklH(Y)gqSI zBF&ZVUzjqAyFRK&`ow374Q^T=Fo9_R*y#yL?fSxrwL`0`Gv*^TRqepF5=||DXX>tj zDb&aC5P~VS$CsYEoF7F=AWYQaC#o7lED&!=0 zqz2Me9jt8AUNa&wFD4>HHv@$H&h6&^fF?d&&e{)a%gNeX@4fK(!4GlXeJUi)XQZZI z!{BAO2wEvq42k}En1Ed_)3k_i!BUg%E+v!TXCbDHqlq)mV7zMat&Kv6_v$Fu=N`om z#xC;J*E52ne}gK9rp@YOQZ1->`y{?Z4n(bs_qmYLL(aRdd4G0CKe-#SC*shseNbUO za;FJ2hEQ3QyRWMkj+g=;@bVn10(v8`es0A4h$D_6gFF9rf`QWxulAfrjaX-ZGDH+o zonyswQJAKZmMd&iU5BA4eebUVenJ{%c&OL3*gBmP$?#{l3xy zKuMOw1o)CB_Jx!1gjU<-;;8ku7exsdn@Lls4djP9MK;*DoxwYH6(~2)2DOTx0O3;> zmwbX{?#yvdsxITGP4&)Z#XU(xLO)S{HNstqMCY>ri359Y z0UaKZejBxE+lSccA(AIP2prQA%7W+L3#w^wlv++Jt-l$5j~?1=H)VuroA;pfQrqz! z>D@-ZJYjn=ln{uom$!P_<*S}J{=Svsib~BE_pJK|fHn-$L%z*iC%sDM+RfkeBpkjIqKD^lHO73&Shkv9b06PpJ-2L7x>#Z0S5 zq6fF4bkSax=%p0`?PUt@Q7jilRy>;${Vo}NwP&gIDJ_2U$etZFL28Op_7T+B?)l`N z=Bvng`srAz_z{qjcyDTpyK!uz;{yza5Xo?Nn{XRjVV*?C+#e$Tizb3i?6C(qqaCsb z@?C-A6+FDW8&`1lX`UY`%<}nWE06a(%3M@(;-uS{-RtUbnXSgA&G;c_aH#io0z-Ys z{7hiMw>v2&3iE2Wcx$Nv^d}^JCJgrSFzCR+lh-nMWaa&nLbkP@;rlWQho#|PrHE-p zjGwhX7JlBMbjS2@O~aXA@*+@ev)xvhZP3&+TUlZFMf$&f2M<#>Ii9I$QwWxZ|BB?^P84pRg9h0tTW?o>U@yq^eO z8@Ld)MtA?c9wmjdV~d%r)|HPQ$8J@X;~%lAlMU}FI_J=Cil1=5^->fUo&M)>h-QS( zIgUwgNY@?{jtY7DhI2_Nr^b>%>N>Yc0h+gD7!UEkRi^)sq_d2QvTNHgAq>(nz#v0N z4$=)nH`4upfOJWRbT>nXbf*eRN{RH)jY>)kEg&fkemC#;Xa2xi!(!dB_rA{aIF7ib zn=g};*@^Aqm#SP-f`$TyDRy8!pVUxqH>aK(Gmmb4;m@L{4`9ACp*p@i`#soqxIh<2 z2+hMjD&R^FTxnh1+6u<<{>MaEs0O`CPQRuKlzL4P3r5d=Ixr+}TWfuoGM91rQoGwa zYq;L#BGbcuvwy)em2)slBqn30@XWIx&smzs!8{_cVs2#4k9A4y-m0UyA0Jl>oVbCE zeVMUJk7XH5w@@~6epfp8iz1f7wS570@*XtQK{}o&qHLjJ@BgnmBR4vNp~kCOh08Q# zgxLXM;=Mhz6-Q;Tv^kT_l0{BfTz5cssChAE$y6&W9+f{ra$K(Wd^C$&c-r1e2$D$A z-!846K$bXI+3FPzWgQF)@s!N+{KgE;aeOlMbcUlk8mp_zD`T&?F#+HrFvCU~C2;`j z{K+p!aR$Q$T?EEAi#pOEGVI-Y53EE?TX>XL(}nrjv~+xZvyVe7gt8LaMA<~ZX6ybS zrSeHk^4*~KW+#<&>-yV$c;sofUPKYdf=q(@wU<-a&MUThN+ziCkr3<;Mw~N2GaFF8K z@elmf(7a=}Bkro00I5InoSL*=aIrJ~;|NscD}@b*4^_&V?*sw-O~nL55++tUL;ZX_ z1YyR2o1}pcx}WWTrD=-J#4YEJvIHNRMutnArFMDJdbCU`Ok^4jtq71A1zga};rl5T zZ{>BvPw;o|HV>pN7P+jM%js?Y6C(8nGgwrqvX~wGq6+=?HN4Kp2ZGx2yDPNRJrYyD z&^Clo3T39@Bq*ka;8{~N$p@;iAI#4_`-)mJZ&3@dMl3B+xIaB4pPELU`2!Qi zSM^WUD6VroiK8p5=WjzRY#`(8&DC)iTg3=rC_k`wv0Y%iRKhiAYik2?{hq5Iqf?z9 zJ0=w@IA!a*Yg*ZsETl9H4~H4IIupRG_^5sgT&m47-keKxIou-n(q z0=2x+DM9j{`j(T|<$95>)z@FF>dkjmLX93{~n};dCQhY*2}e!q#@+H zpxVd$jKw1s;SMN2vuLFRgR#m+dK1`h?7%_0Eq-z(h_lBGa*bKn%KTRw5*nIBzq9R} zx2k_)LNs#)hGVTw{a#VR-_YbWVB=L7yfV9fqEQvIPjS6Ydsm(Ocl<8{p>Wn zCXv{rWdGTU$)7Oqm1QK0B?{v>m2v5$7HyNX9T`7UvphcImGXFUiV;vE{d52MgNBpZ zi)9BC$CiuC8(ET{&W_OpOTDRSw6T^~x%cGcx3P^qGm>23>*!%H+u#TYlMob5#4c+e zdeB#pwA>U}^LjtIV)VW)D5ln(;X;Pl>|*!yBw|zwe2pbJP!AErwWocDE#-rY@SLd+ z$imwfzOkG&SXeQ|ms(}GYQ@_!0ryU_uI8705#uNR(Lr!?erSmuP!@q&E3%8KxI2Kl zn=dPhAG>ccbh8>ak4AKvg&oM{D|E8LO;Xir|H?n->*e+P03Bq$5E%`)TUuH#o0G3c z*c@2)Z#*R?ip${yQ||ieTv`6iinGpGZ82s!QZma?$=tUOt;>;QG;SZItZ}4tuQ}`I zHM=%5pZ$6&k-pJy)$Q3$Yv6^*!5Xw0T2ZCp{-RNDcF*pr#2WZV9H^Fr7WH$DWYo&0 zn88cs(x(K(Nu2Xo&D~a?Km+s`;7uA)A*$Et%_90qF@vhx-O{K)db!|PutY?EAEBjt zkSJHM#VHMG?NT7O&HX)tHT#6n$pdBl+@yWJfm3n0o+uD)M!p~%BPM)*?2oix$>cmN z#&zWtuULs-Cj@3nLfV>ucrDBM+wbqN9~jO8i%zpunuEZq&3=SQ5A>MvZl*8s)HV>H zwfGPNbk8U8&(B(ZJN;K=?$<4L)G-@~O}l}qa6E1F=4VX~$?p-eJs)~+pbh{O;|Wjd zneHa?0gRXL&0dvyWR=dB&7GaRn+K?X`AhQgA$B@IrMd0<#eNzihQ>SF?z0Yi_8Z6v zA2N=UqZ9j{09T5ICzZq(-y>ZKq5m*#$z#GKR1Co*G-=Tal=TT!vn0wbW=@t+{(3`v z(0g^|3v`4)-~7Fms#uo7D#)DCfH!VL9%c|7#$vvUvo^K~@08PDUDTQNXaIRY!2*D4 zuBUo!(=R)#SSgKE^V9%;gh=n*=XX+Lv1A@tvB0BP?2K(chMYtNX{5*MTdU^T8@tCw zF{29_e|%GXG=8ehv1)8Lpp*eu%Dv0JmT&;l19#D1HoJ7-Kr>dTSG_CveyVUBqnZjE zn7_PY7fyXVsG&4*3j@X;M*hPDxjTPukGiC+Lk@ue1uq6;}_@ zf%a(;i(;^S4xTSuB*-I!TNvwr4wPMD3fP=&-b9#waIO>MCH;EYS|1|>$%(c=~!AI$QtFk{x9h7*vK2Nq2VfxjHsbIK?FXz>P^%6}$Jn0@3J{Vn; z9KNgK?%fh!m9s@s12@6(`nq6?cT)ciLX=;`R^+#|)2z|CS6~n#$xrjg&-3(T;pJS0OPypF(7hY-guShD{nrKkCkld}dvHd4K%*g`Px~MB`W%kfV=0 zF?t;N1 zyCuV);LyOp%5CJ*Pp*E%T930?h%187h;-_lmE1126WsWo0eKNG2|XtHH%5)^iG`4W z5IYw&j?p!Kc+mcRL|mLWtj1#?Ka@U*bMnc5*#&bP@?NQgqg8NfkE8gg%f>IE!>8WL z(`WQI5=Ggb=&JM_@lb#n#gchx7jZn!(x*SJ?-N!%}QH$8B&fZMBNyqe&%VBZ+EP$NfVe@eo4TO<4j7s$=IEmiF z2q@s(3frl{_^M2hd2wOKygi%rMo;FnpPvRkGdtG%t-q-$=%T3JaUDQJoftNYEI}^g zDZu6v2eNenj&BeL<4XDYEHwSYdvSV25A!Jk#CSqPy(mu1`Ztpn^V#!sOZ+zsTrJ#b zh#PtdZfqk_mknt!LQd<6yEjJv-mxoD(U3Qe!y^7PXWw+`Tbx&4@MQA{!|dm#=?=xN z%j8(4ybcZ`CJIa|%Vp1Of2M}!6^0dEsRzf#p32-_aHS)EeEMpa_V(-BFJz;Z^&xn4 zdnQx$Zq$>qQ0`UwYbl(Pb@bSz_0GHTvtpL8W1^>-HqTEb`!;cIsAE{f@=})?*c!9% z^5Prks6=Ic5ogVjS6_M|XPhPK+*Ikte)&QkUJ2ubd=|wB`Ai$uMbX@A#vjBJ(99{E zn$B8F1I3g7s~9{l%|t;LEqH=i=XHx#v-r#G;s>SU<@vzP)$94@7FO>gm#EePnR)6X zK4#Lht#4pmLrWuR(IN$-IGFNMMEIsNOyxC*i95@oQ{*GKp>iZHJ^eqXonKplPoLP% z)|c>(M(}K~FR*FC2wi?ZJ3+INdc50Oy#m#XIBO5ti0cJFNbXFUlG)))YX;%N%g+YI zXf1zr{r;UUV8}XOrzK?#Z$S3FT8N1<7W8ZSCbvE`(9YS7KP-ZOt5H!tNQ>{hLp?7u zNo<937dcklaqf(HRbXe1ez>`*#uzJXptguobd933_3A^t9dkonO!#>X^TuW#xd$+=J$pVUFl(S+)?A{DBozSvzC5UpU?1JG zQ@;-j(z+g+5n5?-zytz`0S17%``p^L+Qe31xc!ei5IIBvq&5K~8DM-LXXo-mPAQH` zi39n+h}<@9Ht@6Mv$0HE4uZ_iwE(*UFI;k3fYU{F<8S`OT+K34cPw(a(&xtda z@5U71jrp%b0mhwlowbPK#}NKsOt^<7W3Em}6A@Bmh+-)v%iu6bDS3SC%quIZ-86ELnxeiS!GiCY8s)v{jT~oE+9mH3dx;(c z1ssgWuh9YzL;vzP1W!`)w!i_hrObFAw zc|-p;PK~9LLl_l8B+(*4KL z+sw(uVf0&QLRvc&?|Dgsf9RFMsXiW#8&$*#TUNIc;w4nxh7DWxIlF!!o8r*4#Tp!3 zk?(}25%}kBoR;H2`v}VNS^+{e7#Go@n4dm8Z7{Zay6m_4U?i7c+jCC>XAdOGZza;O zEjk_gP35q+VA$Z>*5Ly7Ki_rA-8s!24kEEl=JiiT-%Hx)n?x$1Itc%5-%6_7@xi6E zwNEQIA(g}X9@4qt6#Q&=_XXSiftz{Up&_9};Q+{@wSC$saJ=x>DG6BvrsTDlRz_;< zK@CQE75Q_s*CPE=WIYd5IOpU?JRigQ0~g$KfJPT{uacd*mNGe+0^^$c$oU(miZK?!S7n*1fGJZ_hCr+$XJN zicLTh6>i|Wv{~yqOTn7?jS*p%|?`YW`w8l!> z#&dUT*k6>0l8d_nz=vtno%9dY8^t^1E_H!1PQ%X+Pk7!MfbLXvxHNG&>nly_$`TDRK9o!TO%9w<6hD~sE@w~7}px{Yz|Dm}X z1OMoP)ria(Rp8%uzk&-|;V73pX?k1hlZ{gf`L49i56cF?Fpa=P?+qRBa?Aide#GQ* z5gC$V6LbQs-PVJK(MM=;!1jx+DD-HKt%u#RyP{09EgKXj_xl z^(_9MVL}%RcVBIhYmab|9V*Wiq2^^24Y^0U)c*;>91qn-mrX64!*}0vt{48wuY(9+ zV0}+yLLx!3+(BKs)FkKhRmUK_&)4*_mK}SY9ixNWEyT3 z%k^9dR2zIYV94Q-zylpmw@~5VKVk~&G1G6n_~x@SComZ2bK}jBar>V6p_okxz8h;b zEx%N%B5sTOdTrEm(bg@zi}D#5551QQZ`Z{?y9V+TEoY7z%EnjF?q1b_)QGM-YFt;U z?NY}yvvw9}>mB}060_6}@{xFX1BqrJVGG18!)7%wA5n+9&-k$+fZsXhh1Tm?c+9EA zCT3eu5HaKc^Plu1zdTiUExw6Eim4Kto|p59dtBIj{VV}?u_w>hI^`AVP-yW`AuItJ z{tekzXMam__4Y*huF0CC)6@;4Ggk_Il~bIi>$5oKY*uo2zMp!XeF+}54c#7!l7eq` ztf(a1BW1x!l-`gqx}zu*LsU8E^rGLGV>>>s(HJ&So~-oN(>zDi*0SF3bj3*e&P;k) zuO2_RQKHN{m01XR&C|BxHt|_+pl9+YCc74= z1f$I>Ij>&>&rj_qA)?->NTqY^n`I*CLur1tv9rL37ITFwABgUW3mUd0>;9@^OEvzv zzxq!q{_=){fUVnYdYNtSfP||B$kS07Y+PGHaFxz~OI8CLUv_*%qi45#Cz`3;*w{C- zYqIX0zTA$k#+Q!i9jxd_{^b+n>Y@Cao@cc+A(`E23l266#9-kTCzs-oKo1#N#SH<)+fB$2*)+OF4Cdks(+7V1$&~y4d zzRwIYf%=+y$_H2cLB&dVm3~Y&y+iq$8|dguKOZBhff3ich&eyiz^5(7QG&hTJ^R1} zeg#Z}dnvHc?E)lyp8W*~-fb1OPVPlX=K3B1LN{BhN#<{PcAt=tGXg-kfLQJBv|mk_ zKKD+Czgr!eiP(1PDcKa*to}pX4|nrpOZ#P^S4orqRL?Jdj^aMA*XIlG_c<=(M|ShU zH&|wdBL6pHertFW{r(BLO86eH^gN3#;hwh^_8+ zg3kH0DO|DHMRQkw@XSomSSJMMB0MC~{1 zzz5NzQGD;?1L?;SGGQvvgrFf~)#wLy@Daf7)98Z4 zY=$ho-ueD@jUjRJ)|t1}-a&z(xD_>p+A5+=*T?c_ll)k2YE%>+Lam>w>@QwWKf-n9 z>F1$foW#Q<3e&|h-J;B0(K@Uf!u9X?_tEkt|5DtbLQG&x!ppwm9IZ?e_ZnKi-Q{3rYw&#PJW+to4TI!2BGSfNdgfZznKL*8KRdqYg5;20GibGy zO;4TX*N>-;$*_qgB&DKJ32k*ck-lC`P^+*y za+-u+t1FFUom-1yd#o94PZ`bSh|rr?9$vUAHu~e)-Qic-g+lcR8M0rBJ0jdB?AGio zS9uWu%ODUr^eVVD2zTz?q^5g3207^Y)vLSGoE1CH1G}eh5FLS84iJ#U0%T88B|f|%&b-(- zOtiT%d>1Z6;0(Qa?)dX+{KxVN98a0jh1nL+V)dgz zjpOz2WX>f_8)G9G13oFvvnCwi?Y)LHVk5LiTohose~0DmkrA8E*%TbtaAd`aIL`EE zp0RVPte%9akOe)1>+ApSZ$rKF#vf-p!n_6-D^BR17_hDtWZyY>tz8wg_pYmJ&C#|x=`z06MB z#L+ugvTi6DxeDhMi&HN(@+mU1U@6sm9rLY4hvGTE3wiTlg3_$*I0+*e!bWOaPWNv{ z+!82WY00a#!@X@iXD$tMObtee!P7TN(m0kJQ^-jts#$o}cqNoNmqkZ>i|%F~s>g0m?Ix6|k(sL@d zePE+k)OH@j-#-}OSItj`K@6Uixgq~P?OL30ZGSO+8Y+m#c$CfA@8R(e$Q(>XeW9%; zwq>qQz;F)7d^gO_$v8rd3#dU}Yc;{O+GH-?AuRkGj~5+l5BDpO4dHh+o4_ID;8Otk zSOU&Y+~Tle$zn|4J^QUJL`jmZuaNptVC1%qRkP4(r0e8CgZf$v_(lReU_T}bEUrZ@ zjJm`xI1&LzbJl!vCo1+Wxfu6PgnHAyJtfOn{X2YnxdwRnm1m_a1}m<16RY|MspM|~ z#A0WhI&e*rED3>4Zv_Xt%itUD)4#6u z&z8AwF9eJDCkXaBTT5~V2xCMn$0m4`7nBhx%3~twYue^b6Wl#FSw6c@AVc!!I9iw)An#>g}@7WU=t2GsXap=J@NNB zT8GVWaLQq~FX_4JC>$AS6~aKuC7@a>;2)?|gO@ zC-ICd)Z4uq6ZSoFh%)WPixv{1IjHn?_#e4#ZH`gZuxfg&V5Eav;}Kbz)_Bt#A-0yG zWpl=G)5xXMeprq%cr5!RQjLqXam3n)cOOA^SwHA9S^0BCyDl;uyb3C_DY*AU3zNXa zU~p!m$P`=O5eUzlvs>$ykA-N~v!WeNkhtD66>rs_%F%4G?J%$=PB$Jf58Tu15ymW8 zKk>Jl*Q$YM11r&QsOFU|&Ue2f7WBaH=g*QW9DrfvY?DLtGC=@`2+6lQf3YAc5w||R zGBptkZx;UlK0Ftli*ESLsB#}}Nl0))&Tf$X{_F3d1dn2KJ&Lhd_|#^1kVpjqn^Pth z|7R%c*zMb$sp1o!NVXxHwQmWsANy32g8gBtbsuqZ9V%cdU*z&qsdSq-b+?`kNPZLx z$=DO)tqj2;?Qi2}Za~tT|6{|JrrK1Te9$-;UNphK{%HxGQnV%y+YPg%_9icrfnv@U zAvDmPw|)x}Laqjk(Qcg^NFIgGpbdrqKeo3=hwr02hw?jk;7CcmKo2Z z&R*jg#h41R!zd?PxCO`Yr=eBXd&8RZQjBX78^(4cwDt=`w?$trs*W*t^g9M-bW69i z2BYN`FE(ZjcA2Xb9Zy}Tq9j282q=)Hc8RjRr7Yx!+d;oqrfeN%yHha|Q56hH2f^QR zXb+u69t8Cb`%R#`aa9p-2NJbbYn}M1=M27Kd?C#4^z7aw+1R-W84PxzIe*0 z!iEtiXEt9>$+zsBLOHMfunj85VVRWM>~kwMr}#dDkYCaGW_+P!5qA`^r5nCv7#1r_ z+OjSwD>nSlzPJ6&?js+Zdhee0N z#ICwmjMd?z`$LYNMNZk{0}EpfC?Q+v-e98apuX462@QYDWIZ zV^;Yh+k-IDd+D)ZXA>sRi%p7z6l=X}l7>q2f+6NedJek97Uq23i89k?{PT<1P2^Nh zkWTEjoBU$c2=fcwm%hVQ(h6S-Bw?6tnVgLnKLDLuIqfa;Z0Gu2S}FNG{tZWmDUWf! zjA#cZZ|rByA`GvC>=z+Bf5^(!R;{`1Z?(oy%^jR>Qa1We*q)B!o?i5*Fp!58g;BkJ z`LdAee`kqNk2KP3_}h=e#MZ8Pf?@LyZsh6HlLWXhC%i+^WC8=iDOz!Q-oMM5rf!8Uo4qFs-wDny&TWq z;(XZD{c1>P_op!Kr^T!1CisEm<5U0s!ku+^e^wU1zhP(tB?Q;>ov%$DQxJ-wB zX6z|x0Zc@@7$HI7$H`*ea>QdTpMbTP|KjzZVIhVuU(`5MhdT9nm4?vcQpRbfUtaX7 z)#98KS`7xMpZb4Fx>WPvc{Tc7@pV@PhgFP5wrCKu>^qdg=H3;IMveVT842gNmv~xZT9)(`E>3*choG<^=eK2npTpt04 z9+yZXJqKkdZPnq-|E{S2Mt7+{*uERURGppn+nX)}0aUF}0L?esxH+{u@%fItiWKwR{C|S}DdZJET9*+2m1Vb{U@w{qLXeq14HC%j$2Ur4l(&6*7I@3 zM}uPj=J7;eWcd&iT3m0G$T<&7+>o=zvjrz*kJuFvB+wPb1mBZRx40(NtE!GR!NFI4 zBBGrd3Tt-jypKEUjNiqYMUnu{SJZ>)(VY9(yDVf@dAVC#KR(@q!BUM$qxj!5a%f$- zG_sKSmRJtn)t%Tc5j_8em)bqAnci>#S0n4L3rtk)E~xVj^}9ayHJVOJKb$gJUyr6wccFvImML z$nF|vrdvuu#BoBgbrVy8W~PAkcP7wpBijfK&PG*I(i*D+&itwRxN9=3wi7$}jw<^_5})G|e|vgrQKc8CvC!lJc0r&}qFHx!vwkX5x_L=L@MJh| z6U;#e6EG+o70+Zke<*O{35~#J8AXuIIkGLqS2?q2uA*J^SgwZ7FjzP~EllR0yxXCd zb8zFZSr%CHj3@|$2$D{NvEhzhv_9`wuw?vD(0uiF{n^)CJjkJRw-4UyL(;sPHc zbDymXSd#Tp9csyRw0s!50EIsSAAmG2+v!62OZ@QisBRZt6Q6tg#`p8pBlH6l&WhM@ zUPDBVHVc1O5oD3$QW)Ky5nBe27eD)@LJuNLuX5+j582i9ckpbii5VaKN{e9>-4>^A z!p*7bl4fJMR$rIH=+s4t2=ZQ}M|ceP=!I*JJ`|=htqJpsH}?i(u`R;SN%F_s7fa63 z=i?X7ttdV&3RN5;KQ!StVD|jo`S}m9JDrswu*85<2YIurdu|lYI~%I(=p?^{YtuT;(*zJW2ne(vjD zdD>jFRICLR5`o4L9Ls=suS9LF`Z0cr?wG^Nj!fPdvFn5QJ+cpFsuGjfBBuO`%nfir zt8B`Cc5g_Ono!*xJ}+gl_r=I|M1#KU4*AC$fS=6t^Cm4}IxTQ?z5BtD{*)u1d*X{) zpb1Pvs7!3nsZs5@U&20*lGCk+Uf|qG6lSD1aaE*fcvHli+cB^DUI})w;am z3tw|gt4<}4;%5hs_QEKKXbe?acwJ6=Uow=F6Z6PsW$7H1JlGf3%+39Hbl{E?Kmd875$3G$9|I)8W%ib zt@NmCt{zJe8^C^05CFHg>s(X`-2)lS3~VfrnCLXljyWs!LF3L9krbL)5c0jvKE#dq zJAS5}OBlO5oPm$EEMUwD|L_a8RkTxoWPP6dqI-b*SUndsSdZFE5{1)${*0ig8=)k^ zuA(kO5)a$~2~`^^piJ#7Xs!4kkqhT~3_Sh_4Uryr+DgtV@W}4G4zXuL(=LZ@!g}Yo zD0AC(9@K2WZFpeTU9&8&^G7Mk+oHQaouJ@<1XrHzpIS806J#M+I1kuuLBse|sJot=H=QK1ve%(+vQpVgMz&x^D&va1o7l>uNYjgdd1z@ZA zVD$Yf{C;(GAnl(n;_Yrp&lK;T!>q^*2mx(Pa$z&PvDZP0?>H_8 z)wCy?Y)Q`lzLhZIZT62%=d3xQbGqGR9>$uE6WhRl=T@*82a1lzVXbb*-RWp6&^l8K zsU`z($T`p}9sKh-%Qp_h)r2{M2pvC|&h(X?=c|i9>ucLBsL3f&!+PBLFG_(lZQDH0 ziXt`@Q1lRc8A|eA)9f&zai940qIe;Rc7~c;-&V0g{3ZyNJdcxlaPix6Xig~<*9%+J zev*S7S9@iu&6@ar9*9Di>=ZiESSRH)fLBtU3w@2%;0y5bvkM@nXirUDPyVjgLPF|& z*npxyTEP#;5Egr6N|EL{GMP-}eHvH0cS&st)9=ShONNj^?pmQ+1lIwzsyqxCgIf=` zqEiIo4+ik3dF`Y5Dtk_f}G?NSj% z+AKU$Rj|LD3?j#DtiK&_23?SKwsIpZH8Kqx6620K&NR-3JYoX7R+i5AMKMhOXCM3h zd@S(A3ldJ9G=|EP4CM$eqUFy_;2P5H`Gk*61eW-*+9}-F&n=4dY*l&FN6s}RKInw& z8BJ}Mx60pUSduyG-ejqH*@Vn%N}_i@EMor{v$E+WbzRL|Ym|gJ+b*uwAfy8Vj?Jmf z`y_P}63vU%hiifmMji4qf7l#a=0Fg6FC@9p{zvht!PZ>{$F4`qsEWD=K#JO^$a>8u zNyQE{LZvfwsvmNbdf5pu6c=Vwc`7LTa}3+@4Y@;($cqN5crZWZI>{7yQ}bTc>7DPH znyb|0gn65tpvl5ZMbyM&jt!3ET@s5aB*NaWNT(Ps1dnh2Cwo-Vnd}dL@hc2A30INp z8OA=1P7*AT&s4LjBrkY#@!c|+Bh#piaoC9*@dTQ_317KAK4BxRf-5z%@l1%&=>Lhw z2=}R^0O-d??(PQio)%QJGMo$?`HgaT6_6v&H83RJSz6m)BgHG_I3f(4a zKZ7GukqOa?ls5MbsnJ`}+w!&kVp^8yXhh<$pC9B~aOQ>rorWvf^Fo=Libq3>l zJzj(NG3Y_(QQM=5MV+hSisb?xz|txsVhulGrr!LC@$k&k{K26>IPi9Hdx8aKgfq~6 z%406|S$YeZe)}wmnuok9=@+6*ajQ6;%a9Q&Xs z?O3C)Tg6V9JIvy|uDf0yH1KyCo28|`J)|^XQwZBjUf^9ke4E)7Qo^CyGn6W-*kSQf zC=+QV28Qo=jV$owQt@$+%kqXBsSw^J>PD8%a$hRwzKOCF06h`WTC;YjAHyBgyB>N= z6BcpoN=U0kY}Ec;=&LOYkpkaqI#`A%l`&YQqvfsJXF+Dx32xDM*b45nfX5HyG3AK4 zBa!uYg{F1+woy0sw9xk;Unf;O(_}EL_jyEGHs@gGBPhMm?D`6TxQ^v`7sptVPl7;* zR&F~wX!bg9IRuGLli6y0L8rRmHWgv@kUzMgYVmGWb-YsNV>Cm=fGnc zBj;8VP!ao{Se{8>y+aF)Kv|CTM!lF`dz<_sc1%u)ji{i4JN12&zmbMp49vu<HY+jC`%p&G9Wk{mvrELNEpPOwLNqy^NMhg3&QH1KMH#B1phX#bI7gk!EH?b(!_-hFcfgBiJnK_TP$nEyhh@u}hR_*>Y; zJ2TIh?@FKoZQT^3XP<=;KIcAW@%OQLFAC>9PvyB}z0ky7Gpxv7mw_F_@lQBJapp=D z@Sp>FbQKe+DRpxu|JuC|lekRV|Bz)D$Y9u^#0e?tE;_4TW%f`(*Ie^^Ng2)9LXT5( zbXG!iH;a!IK-rksgn-pHR_hq`D{+EHff(Mkl58>n)3|uD78*|0GXgWtWGKHvUiOU& zVv+nA^z&kQ7@zEZ`)2*Yl&#r$DLM)-t2pFenEgV-J5i+xJ1;ktD|eLV=e~`JU5i3t zNgv?X4_K2v0U+>W2%Ka{JX2g~8AR8XH2XCyebyLNhuxph9{waIX(KbaScSLIXR)9Dl8sX{@OoWZu!B&cKuA$S#^=O zN&89|iyWb_J7k!Oy~V1yvPla6lHZ->SN@X8c2CxQ&9_PB%{ z3}`Lr?ex15binhZbMJXj-O6;K1sq6ZOnq4E=BL53wdNa1Ou}qrg_LOS^j%cG2rpvC zE66*N{^vc)TYLyTPcn_17|$&6`;ILyUmn{9#`LkKJYli4xlb!C85NO);=dj4Owj#F zTue5!cqGPx`8D{7Pzc&V4nvG8R1%~p##fp;3R!PGGhR?66v%_Ht3=f!Frq*jU>I;= zCKP*4`rnOxwv>N?Rh71Ja!1gRPp{yRR9_c}8*yc%5|#R4?DKm68~uD(hA*(gzR?AX zK+&UmaX}2Q?mD`#S*!Rta`-s73&^4zyW-j%RBW1Nrp(dVSvDOmw+W@kPPF*!f(ks~ zC=Ksy&vDdhpakMgz9Gi6`=Y2JsypapaqymRG+_>|5jYWNWHr$rYu5Q6{C?#qtSBp| z>MkMB83A1PhD}3e_{qwC=lu8OzRp|q>)i>MAor_IU;NgBGphW!qNv+#yR}>M48G*X zGg@Cl-_D$OgJC$hI^r=CZD(AmeP^R@MW!}zbOJY5p*X0uD6-Tg$P@Mo7thGRxPMn0 zL|%uBHM_2L+@I9 zlB8^l^MoxJqaiI!i}zx&fPJ7C;*(=p$AoyE#-WJSSUqF7VU$2~!yfF>M|9|fiC2b{ zJ|<)Ay{*(tMeNrJ6mv70JW$lwcZu`}fy+7$sjyW%VBW#Ptb6b0zm5b5?ESzUB|c%2XZvLUy*PZErC~ z$u1(!@$@Jk5Bq-O;{P@u`0=Q)8Rhyqrk7#wB=V381m&JdK#$AQyU`6FxOSE662POX z>N%D)_=@)KJn0b~WBoigs@fERp3TKkM;u;+_fKhZ)nwiWq>t_3yy0Icj_@Txp$XD3 zCQX^|A7@0O)FZ+{x*goB=GS+r7i%up5W}+sk+^D`9H_J+d*!Ee%=>)uPggBnHL7+g zIY#=hm6NSKk<$**93hpy$el$J?iaWq5X5~igTCAsN$Lg`g89~U;Hvgbx7m!}@}8!1 z<+Q3;)TmM}g~Ts;NQZ^!LL8QUX1bKjjs71#tTG76Z-z_1!|_ z0=*TTAMMeDQ+X-xhVuOGj2FR$zPBxxEWoEzY^-o5ReEWlu}c41EzK9^f&P@3J2KV3 z#%JaC6MCHv0)%iP_DKV5;%yuDuAsK$S6_oD50iPYU%g0PfZdRQZsDpW|0mr3>N4{m z`&=gjE_F-(4YnUKaM7R#Mf`)dZ+t%oAaoocaeD;?8{INb5T;`&3j-nrGF(%z`aDsb zc0o;6$O*EW%5J^~$2Fgi8Cty7@4=i&QZR|``$Q` zZYb0GmxQh;rZzQguN`UJ#*eXQqW1sXOmQ{G1m3dX8UZWPei_P4)h)&l)fG&E$fK|` zn9#|Wg3F$`Gd#Rs4V|O-i&~K$*x^tS969&oqJn88kQ?rKc4Red<@0cTVR&kIc5?u9 z+K3dXbz4#HWkG#TAJTSIw12agc$D}82Q=)4UPSXx=k~4)tv04?|AlH6DE1HVa;}N?r-xU zxR}T~(uf|{f65GtLQTf@{LZ94$&ycQ^Z4v{W;Mf-rJ5xfG5th}dclVri-M8fgw8^X zE*iCJ*s{}wqDCh22IXyQ7>;zbtgaD4L`V+6IP%zr(aw0ZAV1oK)m|RQb9c5*$Gf;rwRB=)XHa*etWxU1n)|x!MiI~;4_}{ne3K)Gtou<%ZX>GBp z0aAooJyWrWilj`#YsPBf?n&wwgJt7SiqG^>Oy1ZYGhwZ4Lbz$Og;Lh9Kl<9!*c7lk z4=1ynF2CLV!YN{}xsyt`)rgwwJSRQQ?O>uS3Ly0T7sPJuu4StI5jWGOBzrR>v&^>_ z9S@haW0G$afk~NASpm^+=%8hwf?0bLtM+LLRK@bY$so7sC`nPAj+Z66 zwGpj2&iCNcz=n*D9H)|0u2fjV4z!5N?ECL|ZnDUVJW&DL{re0G%v}6)vfMJUOP-Qu z>9yk=Ul@~{j(q0pO#BX?oDIoP;|75LbVLc$O5u71)^cp1?hC2NN9SzHd6UWJHum4$ z;V;u+eM>7XEEQl{@!#jX1A;aKz<)K?^L;Ekx$NU+!>&#TGeDc%cG0{AhnE;8zqNz@ zVX6PUW(3>D+?_pd^FSOOgJRmKb5Ik@X-5=_mj6~$B~`1=qt66aJ;kt|1t*TPur*w3d|N+gu>2H5tMs1vg2 zY?H;dVjUXoQxYG~bYaYXdlgy7-IzpcWyzAgk4RMic0n(IWM_duAX3uPBhJsgm{?gE zd3mSRfW4H{Rka@77~0x%+y1W0lFEqg*bv285;z2kzb(CYypDuOeI{m1QdXEF3?L@Y zBs8PrPL5<|Dh0*0OsfXJREkaaO1Kw$U*Sswe}y0xO_4ZiQcslI?B>JpIJ4`ne0gd@ z=v!wJ_^OCuf0=orLL@~WErM`YEBNZU+bCyu9}%W&morw)BkeZ`Y~JNDQKr0iX#PUQ zSvHj!zKFX44iLSux%fxf?m*pU|4G-|hro=US@f#Qh&rZvj+w_k9n8 zw6t`W(nv~osx(MRN`rKFOLqw(Eg;g}&7~2L?oL6v>pfSW@BjDCpfg;Z`-yYTj%Af0+M9a?4$~BzZK()>^m}P=9M9F@I)BVVdw@6KI<**Tl_c(pa0-Z zX57%~RB>Bpi=ZC%*jro(iqtP2gs;0VmW(syZ!m|79jtKYWYD@-C7VLEmWE)=+J{l- zW=?#VRlRO@xkW=eq^c{L4*+Wh)aL$J0dOt^P))}~ylg~P1YBdeM<3|lPn!CD$i*p1 zQ9F;&><5B=1aQ2>EuNS7>#lyvueXQOLIA>oBT}hUtMX!W*WfX^=A*w>RdGV+Gt3D5 zSrtEKs_l0FDTQ=D;Way_Jv6Q;aOg%nuFOIm6PZcA6=TQ9G^eAv%*vlzz2G&WqaV65 zpV3JjZTOa;#Z31pCuv=$l0dVEc7!4XL)K{&Ht)3Gysr|mmoHO%@@%(VhC)kBRhgK& z+Jv6ieoZu-^Z1L#T>I1Mb&hcPpWRl)zi3M}eyOi5FC_rxb`z9KN!K8E0>sb~x=zu0pfx5~m?rRkCR<8nM~8csv! z^NEQd)C$9?2^TU3r~FRz)7vduACgaL!ZIr2*jvL$`SMj_mS+gFzLwE~m5gwn&iwu# z&(Jl)x-4+SNWzHJ_FqBvU6S=odvFJD$IOh6q7x8)Qj(b}CYfqLG39VjEs*nRxS0FM z?sMnXe7WRl&Mf8%h-KM6efmV3V7W7zosS7S*w_rVX8md~$tV!k^#T>0DOpnATMLLRZ3!^O($C`b&56E@8x91<|2KDMTh3o%NbvnKj zp;-?~M&j1ky0I!RhMMj3qyz}fRX`aG`q!IAos;Zva@>C1e3+pd@5d;ZAF3SO-|c5+ zl=*Hr#m_kl?HL1lQY45p8QvGOguHfH3@FPfKO_OKT$j_I1tLX}d#=)kEn+lcuGeVG zn%&*qpzk31ugs2m$@^6L*;j^A+sW_V4ZJa8b?khz~S5A`$<>2b*R^ zb1i(#!76mHum0Ykl-r?{vz7` zgEHwmzGR;@^ox439p}%2m~|hbntWmLmy9{*F#JL5rq)HA_&X^BZt@}q3nnb6#T4$! zqJOF}v?c=V`SAc#nOidbi)= zn+0y_7wvi-^87T8gZeWyHbUyw2Y$0@(LI>+ZFGD z%fHNWuA|eGutYGKAU`t&iLSKtwiNA;cPLca*96tg41#ez6~+`NtQh{j-aVjdRK^{E zSM%hl`iIC;3+USoEZ6!1ejh zEM*`6Y#a6s>i+2+e7gvI=sqgojuRky5){T`MEr2$!Cg$;JQ8<#>LG7m%9|wu^D#3H zhZCW17>2s>$)YXdTqb(80P4=MZR)gcbPaErXF-4C2&_O`5}=XbR)^u1HBB(iI1O(v zZW8h=c2mRu>29%~W2f80$Kw&b_Y(XO{@Gx`cA0~SI5N`lQyGZ9M%g)oHvQGc-=$q} zn1U->Ic}u0V%Cf?{cm0yt@-H_pao3He?H#|IikoWQPg$0!a^JV&A-a(pB zYzjh{pgMwOS<+V`$8|xZ_H^{T@{i*bB8nE#`rA&uXD0XTwu7&kW!rFn|LXmqEso~W zduyQh7E9$$CcA!nQB+P5cLVpsU^LrM8U;W?OG*?es1z536{;V{H?7%cjt}bm58LWm zjkWS1|Bi&^(~IRigQPECEa8h3f#@qY)`#a)a$lA@Cd`|s7f+>6j zV?9?rOJ|uB)Tyqb@+>|f0kZOwM}djV_^JBo_DFBr!#S&}&k>Xomq;7X`{U9Gq2AwK z@)&A4le^sP{(ML5WnuP%d7-WfQ+WG>p+(4W8iVGwK^G!LP!MkaF}<|0am8JI#8{qe zg+p0FfPhuc%45v=_tfx^>K=w5MKkS7?b@Nl_nY79J~twai~-^6a+ibm*Xif6 zzwJ?lWSnE=)3`r?J+d=hMnO$Y4OAi7-GmA4Wwt+Y>)P2J!H!^a-qHbK2g|IsvtP0v zhbN+Gyw9J*136uRKdrYmBMTCNz1%?Cx=6i5e7aOuUeeh4!N+{0DuyT>R35zXc@~`# zpj70ez5GCPpaSSs4W+Q>s;6(=NJ@J=1yX0p3)|W`A|&E*3qrMD5a8fZyI;Fr_u~cy ztWj>@A>_bxoaEWSJ|3wfN1DHsZoY!G@l_QIoaCG;#nESF!IARZ?%=T{xRdBwnfBni zlMr8}D<$=`>gn6v{e`fzEP(mv0MqEQ`NfsfoSWd|Il~O`@<3}0kddah>q2i2LaAlW zQfcY#?KJ`{Jv);{8o8vQsu*a!GJqnE6=z|X`Q%1OIOiRKg@whI=dVWRk7oz-)0`U0 z7iNGSHigGFGhF2mu+5x2ki(2jJ)4XWl$HL_Wn#?r0wZB%@hrf<~*h0^so_PVJ z97we;#vC~6e%zXlRA99Rjc2uSZ5$A0`^Df1-7bhgP=vR(o0OGS;-?wn-65v>?(PuH zW?n${83g#W<+^)JsrmV3q7^IL6}@at+j}-^;ZSfjR^nL=Veg*ej8Xe?dqr@@?b*aQ zxoU4T`da#`y>ZOQ*peWYz|_7h;Eab>X%J&79@*x z_hP`Pv|4F(9>AobO+Un6Efgl)`bde>R(2PyG{KDSqwf|?kM+^maZwWJA%|R2D>VPy zU2*^S+cclvVM@9_IS|+;g9^Iv-k*)jAHEX;#=b;5YTN@T+ob9ls7c8$g=n_;!@UBB zeWmrG{_SK=jv8QqlQv!B9{O5GI&376wfq6@;Olg%}6TL9gD93Wx|`uJ=|h2*I%zJ;f^-fEB6bG%vSVYC=#Kr?5032owty;g z(2Lt6>Vz+Rlk_T-?hRG=lcep1FKrAe$F||xx7}jfZW{Z9qgqK_$QK!Ah3zLZwRb)z zo_*ina=(9BQcXF6<0~R46KOdu7SV__f=#o1*@r+66maFD&Ms8UR{7TM!x@%_zr2!2 zpq2k?t)idd;K<`E4koi~&Jljw{*ED@77eDpcg-J@wn9m#O;Sy*4KQ5WL1gxIHPvwkL_M&)yxk5gBKKzr*o+SVW-$bF z`5AdhMu1U@ckg?y-Mn?(YIX?uGo-A(L^L@?#ojm>@&;C>B7Q89#YD5)GgX5*F73(@ zlBbIMbPyFQ0Dj2wCL)=RDlI&lS0|NW3hlo~DR~cDkclBF>88$g!~_XbXmQ?0hkUA} znWZ<2X|B-q(zqn#!Y^|U^{3TB0US(rG{Z7133nsZhlk##yAkS&O#TK zs-D-e>An`#xY_i^ez4$VS$Qh2ZS83&j&U0gQsDBN2G`Y^X{+U`$`w&uBxZU>@~uRE z^B?n7>au3#j5020 ztoadD58!k zc(}q4Q1J>M?V&=^SkQuQKgouBzoR;HHuiy5G8NOvx;neB-(3v!|>iBq7cL6vgB97;TXncgGRAy!T3D zZ~1`MCc~x}bgxX%jqR<5Q1n2xW{jpeSIVh6Hqzps;90{W&B4JTVJz4h@}2$2r_M=w zvt(P$_AxLnj`_6LTrOOrnTmq3CJ&CIEBsaO=H;hMAH&R-BTraU&|fUeR$<7pj%Lrq zP*1rUHPp+Epf!P>hlTCBW#z;`Ucu|WK?W`@cHA<;)JvYkw5v%*Dgl%5-fCCGa@D2h zX)tuGNR^O&L;P5zx88;5z(8s|dAs z{agfOPtW&E5EzjVpGiwUiOl}qSC8P`&1_F66c&aT;9dPt3jY@k3|F~t29bhvc}|wf zG6Ivj=nZ(!azv*a zgT<$V$UlE&;SBzwV8#@1DRvJ1JLN>>2sB>_dnviNaE*E**@-lwuz@4Ic>;IJZHt)v zvL=a^@VVwk-2MnUxnzn_F5`n)zk@W#aw`+${9)^mvPaTT)<@z`CU59+zHh$iJ3EZt zUsJjYZ6U+IaJ6W=AoKHwD>)dpB(h?l`%JBRkfqH-*`sc=jHwEn8}J9Vmyb7TD~24c zaXn+;_Ae$uKZ$XRS!FxU-b-rqOnVQf80G{tjjoN!P`om^<9FIIc#rg#WN?~4+{p7D~pB}+y~g|npYH|0WvVzChwZ#iF}{4f%^M@Wq?I6kHzaKKg7 z%*d;hujXv}pNt|lIEUZR=0c7JJ|`9guh2ifbfg(W2Y z+Vw*?qom$flShYy4U1BS`k%$Q6Hb&ivck2cuRDI^A!t@|phZl(c6UbIv-ys%rnK2p z;@j@VSX+3U_@^KiRhL*Qj;4Bnal`2HhT#FYb8YEC*`aW2Gd7j9iDi=HfL?u4XIdq3&yY5%? zu0e^xO$fDU(@Vm?yR)r_1N5NK3#h{oE0+q_P?|eg!hT$=$|k;jsw8S?^y=cG54S8( zs37H8?{;yvp0>#Y0RET!3!SzdRtQrw3f z{9czo)sxd;d7A@-$Q~4$9wSdIM zAx%c8n-oh=pHrk8%^Cxl_~G+B8;R}92~}7`hUVMS5=9Yv>Z6_9c5PeO1X{6|l~aXk zQ1G$`{&$0Fz2Vj*I?sY17@3-``erXW#>m0(tkopXi0Lhhxcs?>{fC+SNIY0_HyF1+ zd@F?oegc1BUN&R7oxCv^qrUJywtA-q^rd!5{JIfN1Thdksf%hiJW1gfAWkrg>eS$r zS=*KZ#3%DgR>Q=>a?8R;m@!gra>!mAbV+f^bi%s+^=4AiT!v|5*fdrz;4g}EQY8I< z@PBa87|GyfD`}=Coc;(()AYq~cou(`%3_6WJv0ChZddsJOjQzi^yCR)&=Qqj@Y~Ta zNRPgU&Qsy|zM^;;O8xU;uG&JnI){SK{ls{GrlRYN?Gezf0=|1mQs28{bG;FgAa4G# zaP$>ug=c)j3+O(&fu**C;AyOY)5eXHcCX_7mCuouLsO}Vyj@t$-oCc(omBtN+Gr&` ze>X1qkj}|<7cL9@2&_wmE8|iL<1L;Y5&_J0nVt`?yCZfq?})|A5nF}b?P_pG3T|0M zUke;_G`GGb(zYJ?@%(s$qjn~;$d8cHlyFqM*5Vh@(e@`layh)969(sJraJ8JIRzJ_ z9V^lL!|3MFn4~>80^$c7GN2}&i#s4NP(G$embGOHboiO;cC1d%lPvR)eW#bdrtU(AoJF$jrk##+Hy~BD zS(F0PSC&Kj?^|9X+oz_-s!X6Gz#d=|;{>{p!Li%LqOM2RbQsV?R|jAp{GN+$GcGG0 zWJUV&>U4+~c?hv-hx)ew6w?G{lQiYz~>YMwHGad3kYp z$wSv@Hj=R=+dfnTfP~aAaeRt8+hv=|Qq8yJt)iHB`5O_6y z4V_B=l4IM1w?lBwhY@p)98!1n#5zJx%r4o@6yhT~VUTtIf+HOXKKV%UBpytCubGFz zZ6!NP2*3i>>>V-Dgi=4-S0NV+&KT0}4p!mQO8lk7LWR)BOmtwT-mxmFb#C z&RUb76XVqExj|Prh|lGvmD%BL!y#HD1dla@t-_+OaVV+}i37S2Lk#*tDbdK+BK$+( zLe7zH!iyjd3-U(Yd8fFQ->t{OxkUy)Rh@MD_bBhs*?rxKm$!#57$9YmM|AwUI=|y@Y2o;V&rvymHiJ{VnGpPxY3URTqe)>Mw zkarQeZGAt;qWq=l-k%@L69D{}M@wC=BUVA{ZKYYTKK-aO^?Nivl64II$T%U>9Nh<|lZZyV*E%#zeG#9+WF`*V zM4PG(xL=@HT2vK!1spW8MB`&x*1J)j(Eh{S61}@2kkr1$(NQHFp>2&ZTDEg6;MyCJ z1FvZnU;4%v>(PL?>E}bIa%g zQ}%d#8s%#1miEfX28nt87mBB&KsIl4=5gWO6EAF%-yHV_NMsn1MCf0=p4o-V(a5~7 z_+i}8*7(e|Kl~~4xhnu@@HA%h)WpPUV^~-ni!t6lCbiVqGz}-~ZkUY;*Zp?}K!2LB z&+P#ZG!o;D!sedrt27xr0#1ZjEW8sJf$^ytZn5y)kwisQfZez}&nnO(Yad8rx;!uO z0D{%>yPcpj((Pj6%KTjpp69!UMH|)VyyCE3SAT#1{GQe##>UYp7$@YNzG{p0RAty5 zX11rcgZOFoCIjc!pSlVTj*$uNIK8?IS4`Q@y^hxiuSgt&+$db5w}`*(cz1GVnW(w*_u%k-=VR7U+R&_Gr<@rquvL=)C(aW7RF zV@D#d0Pgn`T$6^a?}lhO=H6Bd95zR3;nrJ5yiD)J@WWP??qYW; z%@!)$CLHlNHLU2Ic1_$Mvgn0ABPF>^rh1lz)(7cWLtlD5@}0w17KSdmwkw*0NgCp% z_U(hLaF@%N3qArzM_of)HRC}jqJzm9=zsly@(+~c`P+>IMLG8iCX1|hRA;+TnCg_6 z`9LPhUu{YABqmiYh$Vc<=hix5Q?1e#uc(v3cl^6#(G9h@8lrRw``aiJssUL$w$ak` zbi#vLeH1j;tRdrB5%+I(o22{{?o{C_XTP3ubgSa9vN+ecd7xzA!G~*!vG{$56@9He z`?<}`|6jPh)?SJ0`*2MMT114#nd{hN=@98?kkx=Wc&4S*UTa6QQsp(7HkNVu}AyaOJC`)voAG zc`+{9 zdpZCCH4>8xNQ!!U7ipp>@Ucae3K~dk_0jmtvfeidc6V^`o0?Tv(UFDv!?keKq`a#6 zZDugrBAU;u<3w4{s-L>{Sbd7K1rvtqm%awNo6uh`n}1^dTlT=+%+diO-1%{TpkN1R zsF>lbTkT?I0sTc6>qIDvX5Jd-y<8jHe2-+c+0{Pv2zzZ&DW4D6g=173h_U_|Yc9P% zZI3?9H+w}_ZI8DioUdr1K83oN9~=9FY=K5%B9{pB`~-RC&lcq6k_PgkwUAt1XO>uA z)IQCSb{M9K^5*5fTrayGh_(F%p03>;Q3auwBpbhr16AiSA0P}c5V7FjDB^DkAo0H>*&*JBEPbtE+5-&?B zBHScQv$QbuZD0;%|B86#Ot{5?~%jM@x^!*Hs{c9=G3Kr3< zeO^&LV|{e7e;X%{OY_;9KN30HtlCher4`gDNof}5eAHmlNkXbC!V>e*3Pn~3ZWACu zAsVirGNI>wej%k}iT3o=!Q_RAeN4cL3qh(kXJ2whKzb8WmQ<0bq~tZ#^9k$9sD-_x z|GXt%mFY0Yi`m<=U88|(cN_|em`tA=6e#B-bVwcakmrEdgLgAg^=Ks!)9_%fCZ7k- zyB28b!Z9!~07-E?s7Z$cExadvG}uE+0&i=ww(Q}8hHY#cQQU5==g$BcmA`T>r3nB0 z)kR@}Rj`^>u^F67B3;|mZu_q$zr;Mh7*yonv>B6c%cHBvi|zcZNrNbpB&?3n!2_ zHQ3p!>o?l&RE@m+X{jVRTWwsDD-2DnYAP<&f;=>~+yED;4vzW;mWoXNIs$*#S`BTy z>5Y)x-#4FV8RA_)uL<;!znu6*5U}@Oat%JgGErFi094;V*4>r6e*o*=a3;4zpvN*=aR6j?OY(Df`~L7gckdYC;N zII!g`dU^^oKaf4wrpGmk%(!lDd>kpzmYk?fgJaGnkELsG+_;QU1phF*;I!w zgb7_FQ0Wxr^t#WQV>03dHdE{97kWww)xuFFyO>pfIo;-ue}fNB6- z2zFpsFsuwY%GrTuY&=1BrGvon#9Ssr%r5{a0aA7I`JqK zJ}bt$%_o`1|L6QkpPjY1?9cd@m$RMiO`<~i0s(KJwTFtm@U6}wND&QB)BW)G3vwrb zPO2sRQQdTY)OHLU_9Vs5EhXELd)EE6i{2bRr4gJ`?2Bc55z1;BEYF`7H$+Yjk49;562U zatYSIKU9t8&7yT*{rr?Epn8GzJb`;z{qLD6WPz$|=(x<9%+%ZG?n<14ExiUX=1w4# z7tXy&OsTQ=1OS!3T_=7%YlR3kJ%U{8T1t4!W(g^`kSV{_+>xQeOU?_cKYoAp z{a@sSC4F6@S%$8Z%)WE5EGNEkhLgsXkdmDJvN`$MUOqtR5H;}gjA%6J+p1n)CJ)@oXfqH>%Sl*XX0t`k=<6RS&orR#dxAe1b|u?+`^RhHSng$LitHS zy&t}JE|LlF$3Lrqfy`M98mJ73A2WbjLq>yh=5QJhARlREd$kxNbOG(^$O3A&ykMH} zctTZ?3N)C3mVQJ%pKB8uxy`LDLy#=r-l(IK@{U%~Nfr8z4x=C&Xq`VHow`ws{CEwc z@aIL_s*nEK_b&CEMde=V`ol20IwI7wdel(zryHOVVbIT+6^>;2i9>>*Cv?CG`jeLf z%sT1UtYpz4gZ4*J^osvJ4=EchtE!*Gmm|pp&Q9 zaYQJDh|`pki3t^Symf$LNz})(mZ`Inlhbhr^S2VYg!Iht@dVVgzgeo$_5IWjPOzt( z=|@+WMR%}8{jOf8zpOHu%=HK}V<|oV!_agl!dfF}*T{CwN&(6jbyVPV~Bx!XXf*N?2QH)S+S-xikK8qYT zUeyn10_b5S1YFV;Nc3VsZU?;BhhN{`)F0H&vw57E2YveFevwZfl?d!^=B=Mk3?Sn| z2+)RosXro)>60y5g)Mp*Ci*Djk3lf^(Eji8Akhbf!~ZSZ(<{KRXbwU&^R#aM;k!&>@4xM;P$IfNi%lTwS#v1uJ;#$` z5YyDi>Ov6RX0mQLt3OBd z2tPayl{2DzVoXMT_urK^W0i{D|Gg$4p793)#)KV=+}u;h>Z_}(881Sl#l&E@ex^rg zfFt{OGu>u4^C5(amv`nY*)WE|RQ~uCv`Zh_W(G_Q4H(J`r0}`=(AEl=&q*0K8BPm@ zivRaSjG!>`-*p(@7#ika7lU?@C<6#{!9b|K4<_O34V+@#rm5zOvbKkCK$D3J&T8Oe=JrW;u@D5 zP7I(#{_n?Ng6gZf7#2MAhYhW`Ve1Rz)wE86x=cy^TKKT7NW=l0qQF75A*dCy;AnvPQ`hR(qnSXH$e**NA89s?g;$_!Ucb2IVh&C$THyV?Bo@54G0>}_|H-=^X$Zdc&0*SrTUlbsoy(j>fnooP&&;$L zZfl0gTBZy>H{+T362WK;An`n?Rtc}#0~$IP&UJJ}H6BXyt!n@kVe zNzkpVdbj-eiI@Dpy9KW)bu>Qk-|NfztONK{cZE~|p^Sn(crgUvYaHHNp97;oAj?wo z&c2T-1Q#r@_j~0{U5&|0^{7`eKMY3j2ig`}h7c+$a)*HN|BX{Hiwg2AOS-Ww9ygGB$5C^0iM{(hf7C8VL zVxAi{b;q=WVv-a?X)a?!F{Ti}|6~1ibzBx%gS8|H3P1MxYo$C^Yp{!GQg*(JVMPvf z6B58@X70#@6MBJt@Z3DaB`)UiwvCMh z@x0p$=M3>cLSRfUzyNf~*gi~h=Dx!9ue^30Ra|fwWPz+jfM%z*`wIvNF_IEX=53lv z+(5h-jzFNprq{Xn>26!(k-?^MH=ipMbl%)~?@4xqu7VfJV$*!wV>rN2z`XU<@#j{Z z7|$`FXvsT1Bw4%%ht;l z&9U*k1^4ZA2q-q`F;|a@L%^OLzvJFP)C0R<4RDvxg9tX-5Zu5=AA!4iKgpJ2y6hw}kXJFFkb&wI7=wWpBPdsh|$7U7vc zR)^Urt(7~74KMNafgy|l{DQ4lhrW*_)-PV0BlRpd{b^*Hl^H7hK!NgpOaXcYD80$ za}bvEw4nzmCd{LlCD9DlEq!AM^Y(?=1bTJehlQwK6S};dFGdJo1|5d*jSn%B@ z7wAJ8K%zf8JnqPdFxJi(Myz#VipU{;_XQ-c{WCUsYPg=4S%Kea#tslwmL5Z&@4cKD z18-eJ@N86K*W^S`x#vAG}AT%+8Dsr5|K=H`Fv`Irn*F)7vC zQ2Xb7=BQg~ZD6G7jcJB$>-DdJ;S^g$9@5WS$u-k@0tl)fQ4f#{Chbu=ZmPOgpEm@ClfR!YKu}aQr05IRNqve8A9Qe>!ha-uaV?%@FLbKgzC8 z!R@vSNJx^T?8yJYodd1kg>HbtEL{I?jQ|paqmYtVMt>8BxsJG=6}n3k&6`7L+k6}j zHnH7~W&QWuI7M#J7LYEy!TMrdI~xwh4urg8z{j$sND1iK=fD=q9iC}h7Uumj-njhx zIh+`7t!*>Zx`IsVl~j#!uII#9-;3$Eia+tPFW)fL^30j%;?{hR&he2Jb}y!BO9$?( zpT2oBSvwyuV%31O;oUaXp<20O;|lxXr*YrlZNYCsu{6W(%hx|$e=UnW=KOIWLu!Hg z2HQS`A_)w>=X{HSS#Nnu0+z58in4fu#u?4?>Q21Dfq@H8%>BV%p_*Zxw&|^rAjk3M zbUsV_sMD|q{m||DH1qK*T8NnIF>px*o833RK+zlrG4f?lc6yg3sIf3Y{VdcZ+5`KD z`@6VGfn%?G>#L+QiV|j>Y>6H%O8bbfMxW{sz}n%<3orb z`YjD_?Big>TO42m4JB-WLm3UTdymlazu(x3gIr#b8pw^zKMCL%j7v(_61&89p19xO zC&r$?hwqZst1+4s8%^KfAIKEDjO(pWiD*pN6U23%@Q#dR|M|4cP7`5KlDt6N6oVi5 zj2+1u?<)c1y}pCzbFW8H4rnOEMhV2bh!x&x_*dTw?2v=~nW@97EV8Hww&k;X^A#-T ziFC;pCmtH3$Ovw1Cz>arokZA9)FhZ)p@h{mJ#krdkx;*u5qeEasNQMu>Pa750}a20 z>oahZ?|#$oC``odTwuL#`mDnPPH;rySd8JzJ%tk;aEIwoXD7uDri1bizO*25BhRt# zeoMfYWv-B9UvwIO)rJ!A_yMnCnr;Xj;EFH2=7S)i%It|tHOpwYmdEl+2FZ+IO@c*! zYCib&EIP$o8Y_p5=2=x!82_FR@({S5{DiVAj`R5a_gpp%C!RgJp2k1V_nsq%8wQ}z z()*kwWIhjc#Hu#hD=c_91&oQ|l$M#b4-Wiemq2lIpeYz%i@x+mvQ7n_KO97MoGmi! z$aBX@WnM2@K3+gQ;AtA=4q5JFquA-80q+UONb&SGxDXLudLLNXF^$b)CjS{~W03`QD$efmS;DWL=lp z(BcD?GpZsDG%H!~9g! zx4uY=f$Y(+Q-pLkJjjSNJq+GdXQ^2@4e3{C0sN6v?9uS)U{md&gFEUPFgLO4qxio# z@_#{Cq;V(N|1Z4!e;+&jU+@Y(3;a40^x|*SXG66S-yv^>N%((Xr8@v}a^1Uiibs0T zkG8u8f*$!pk*~(_Z^CW8)=74Sa)eqGH$cMpW?BSH$`XvpSa?o0jqUCdyt?$Dys+gXL#RN76u}A=Sn)X+-=zt{AGh2>T6?gD9Nu302PZy;GH+{OxfMVnHoI0I|F&g*Jey z-e|py`vY8oCUfIX4g@4B^PmADGHCniiy|X zB_=leXa;<`u}+nMD)Y@R)*MU(nzXjNO;!l#V9~zYRrFZ)1AuKi`yL`QTqJTg0}l1NPF{BaSm6O`1FO?C14;uc58%4Oi%J`R)I(icRajh# zqXh``UiyGlBmiYg-P8RH`xb5&z@12oMAx;d0Rl6_nn`sgpR-&FJZR*UW`+odg18z& z?(w~TBVINf8u7aUm@2;Fib?8-f$ZoB^t7TXmVx{EY!> z`JWf`H1pYwz8r!D!VWS}IO>HNba*SXCx+r~1qq_)dozAk*O0?wm-UeeP^ehmc!6j! zKG!l%3ZLO`&eybjZ{y$u4~KNx0n%n&9eW5K1Z=A>a*QSpZR%D!)hg zYwT>w%J%bLV`D|(WeTN%IJobxFWVl|;borCz;?U>4&NgDi-SBeY^H%Y7_2gj0Nd2b zoG(o9Ti@|T1nGV^csD9{2e!Vl$}je4NS;h4v`w;i_=Yp&({F(=#}I5t*=^+qNrX@= ztfb_o_z@J^6iW0;FxsRTb4j5F?deyU-QrMD+%tt{qNcz^38OHeRsVwan8oLm-c5#m z7m4G)kzy*vR=xpuCZwCo`a{5aEp7SE^v1KChQH_P+vASFo}B07R9&l@9LT5($u5J~ z<-eh4OYz~no{VvU9?SxJmG=N)ayboGv}>&xZi%XL*BQE3m^dM9DS8oZze$LH-mchobFt@ za3V%0b2B2cQCVj1)%3YR*Hylp6#NgmQ@=_s?#=7mGRNvxV?E`In~<|Jsnc+e5e`Yb zOS%yYy3#RQDe9lXxBGAOh>8g#CSn4cvmVWR=jtk-|Nc_tIqwste3N}9dAUck7Z@kW zzp*{Vo?xYz!TB>@b))~@%aWvlHDNcvz}P@xRGKING_Cj>0!OA`^tocE%=iA|o&ey& zGCUsCf2hoj>l+9iR}d)S1;}9njZu=C`Wx$|=IoX<&OY6qe1ny&hd*EdV`Bn3y_hiM z-G$DB3o>e>3XtK`nI#;T@ubkX0zANp<=Pv#G5G`W`43KdhG8#$9g>m-s+Nurh`a<0z+fE)9e_c3*$aRpr^jFzr3c_48A#xy zW(m2zm|@reKE+yxUNTp?O)gM;`y3@!Vuh@{35xaUdxv0L-ZDdL1bm6gOOPw?Hcs)M z_{tcyGf!(8MV$ciDLc)DxaFj1O_ixC)J8S*S=+Y7d6v*={5miWB%1kxpcFU&@|bKZ zdf`HJrnF!?@C-k;v|84IbCSFNp3R)sNC=L=V6Z{@-Ue7&hp;SrB_=abDBuh%NuEZ6tP04Gm#J@_819n=dZ>rTm}}6b<>v3 zC~{f@RQ}r>Kw08t0RX_>38TNr5nKzU8GuqMxtp4u#r!U9dN~>MyU0lACFtta7aG4RQha$FlHmNwYXnmGhf1#%f24t6j&VUpyA$wAlRz%}u>tsxZ`=al&=;J1 zg?h)7^Sz=n?vec=C};(Z+bz6`rs8~7AEQD4fzWv>fKXrP(j35H@KO$YUw_rf#JE2K zp3yzVapn6E=x^7DdWOOSl7wn~qjwV7Z>>`zM-xaj{w8GZ;5qUeWcas|Kybtxl??(u zk1s{<$7B=$D9S~kz&&fcNoW+Q%#GXNG%Q7+o466yF@QwjrIcaUiBc#S#Xg|Yln%0m zfkW>@&rK)@0&R(INp`gbqrwga=`L@lzZ~YhkoPt|9p+KIANr~Wc?A?Zz2YexDF*KQ zTmc$k8_x-hz%h`aeXjeCLJOR(_@x{&5ddOrf;<)77L?GeqRw*XW(f@oC6tei=)$XGlbbRIQfX7LM z=A)Lr9pC*u@-$M6W|W8Mpa{kwDB~9CK!X8tH-Pe-y)9#-s7M;qn(eL@x}jf;BTKq# z?!pjA#1uMH0)ZE0)}MYdD=Pp6Xj}tq>V?(=#p9KXJE(RfLCrkhFxPk@C_i|X^#p_K zRX`(;FK5)&w8(p*N$S>0%sEQL;@xV+m2F1gVv*WG1Lg#V8{yhxcHCM!}qE{41W4^s!a^#gIbYt zmSA(r!O?S#`?pRp+&K@)Iu89#4+eQuR^N~bm}U~KC0@&~C(-|?nR?G8b#f6Td=esZ zM^$28?jg3gLvxdW7rCeJ)O7qdNbL_MIxMf$-H(`I?s;Y@=gmIi?a*TEq!%7$2x-DBLUyXdORn6%p`NW0doVJmn?CUR?Vl+Nr zTp~dKK$|8&bAJgwC;oY5xtY{3Ldu)6O;v<)wT2Z(jr+Zr)Rz`46U~E{E5vTUKRQv_=_OQ zY^g2lkZAklDg638xNw?rK%@s-NMRcSzGjxD&-LbbiaFYlV;mk*3(?0w12==8k0A8V z=wUSqZR700Dk_dNbBcq4d)Bv47te<{N90bpuCE9HRyKMquI-98)nNFp>uc4b%RIWS z#bjDZQZ*I!p_Aq5{D%;-sk==a&rjSEhn1hcRwR32qgh#vxd4}qCqrs4SyqSiZ8`2^ zadjnn3b%x5$(*7gDbYD4UEDMVfG~_PfA21a068+RU#Gh-tt`|UB9Z>Ou=vak_Wg-Cgnur=u(G+LUgxT0x9DmLhS z6Kgj!t`w=~U$^@3Rc&c6i7NGUpaD2HiJPHm&Mg(v<&86BQozc==xy_eF)u&a_CRc-tx&K{s2`xHtSFM%N*2=tOyax zn!>y=6RNhg?Ida+QcewJ5rqRjT@;JVlIPdWs~6G)d(Ow&lM;XC6-i__4Um{L>bT=Uqk%6*4ndPZ|FPuLCHsb&0d{4-3D?kvc|^ltLcxr{d4ro%<* zcJ`i*VPtOK+#D?F@D^3HTd=`xx}ygNm#R)=M<6z$g!Rl<3E^@)3X>rh?=6~jIh)gT zyJ&u;f&t!3glHq{cAf3V9*$GITtB|itIuR#fSBZV!3L@$Fd(~sjw zKNKg>Rg(8cUeI8kaLw$!eZ=PdwwM-}h)tKoZ6jB)nbeMWZWeu5=r#1Ahwg2e1NR}x zb|2ysGb`PkOiMVvKFDRQ!ejyAV?Ua^6*vSQ=mBw`*fSFRep)c){iqVnq{KlsWJ@-` zJcyKh!ZbvL_A?;q3<)-V%N^v`OKI1PJ3dx$m%u&qMWq`*{dt*Deos)4UlGTws8K|X zrJ>E~S_ z94E0#CxLmh;GY(x?*aasP^^Q#G9KIq0V#7(Iz8dl=*I~GJ_*qb_B=Z6X!npPML(*h z08y>^EklQb3@Oqe zLw8FfAw#!xcPd@dHI%f7APh(&B_#|gphF1=JZt>j_kQ+E*#GLs0q43_o#%0UR}mc+ z*{a+b|9TJhlwoS;`I_HLzX{3)l1%==dqLD+sY#$x0_Qk38NR!fbNWmKypgGtq!oua zkFjrF_H)Sg-whX+laY7kf5f5bZc$ZO6=}Ws>wXiPzm>D3A5I^~eo;KNhMCb+tv{vT z{j=c@$Y&#ped?7)#|2$e>iu4}em_2HjP!GYYi_Gw8}Qqf--~}q0@ulYhYPis;x2u{ z#)fjQE5whL#h2coT}$yU>)yC=n0N!P}r81G@APh79S@x5202NZWZpDBEhN> zIShnx^e5x%!~DISLjtUaCwXYQXwvJ2@Xk{(27l+lSK4e;Bfj^W4i(kn7~6)uv6g{e z%exkdN5VtLcfgAD?X?0nR-QR^%Rt0Dach@gotlq+22(={Nk~wF&)%!L*uyxwBhit2 zheC^?nkj!RSlQ%tOph+}FUsN>JxCd_Dql5Y<%u&gM17Z|c;z0R+*d43v={f6&P{Uu zUOW3jqX?zZBn_KzPGoh+_ZpSVzyXe*ULLjxU&Ut9+aVdR4Vr>T5`OoT-g%p-BWpv} z`gzv8y$fcE#f$~DiN;gIA*{Sd#bH3uh#yUON~S&IdS^68-RN;~~xV7;l>nvX#bp@#)x1s5t>br55YLgfO^(^orOpuW=U@{&vZ7 zY*wtK{^X*qx-akFCuYeSWc!{rvZwr&FHYqugiw1bOPh(-SVw}G18}}Q)YU}V7bauf zvvTq1+K%eKV+j=&WjwTA{X474GVsG}7O@5}0}Y6>I(j;=Cg%9V(78rW-wCLl=wGx#zb= zTI4oNa*$ne;>kVQu-R)|)I<5Y3V!h{HM@NoDN*F;tJqe(=6BbqkZs%)JE%|ZiLm+M zSolD+``)l;z~lJvN&8=g1a-3dXU@C!hwr)9e&VwTZ-EeWZcM+@QvS}D>5r_t4}u?6 zeOzz+QO448f&))uPN32LyWC+#8+B$q^=9vD`WuH^Iy2*jzE}}09(@nZ@6BU&;Ip9C zGg;vx%02rq+gR}D^N=jeBOmK8b5uj++tWG7W=1-Y%C2;Hs4;ngOlx65uf&o~Ul`Xy zf97n%@6V-?SR}dG+W5@wr_Jwzg2%H4Ce_o7`t z78ZU{!Piw8xJ;uZi#H%`l4C}1x*!l*cOn_yH@pAfcr)cq)w0&=n+eJztmovApYyj- zobtDI7{wHo>0<`sR3BFnCW#dHt<=zOXH_o{JfMWqx_+*0d#T2l&quZ+-TTAg;WTx+ zh{n}ed-4pWx%oJk1y2px-r5ZKAwReq!u$t2 z3V$BoIb|XeHa)ZPE{+ah=O!c@-z!MGBtm^58Dg(7iqoYH>5X<1{y8{F8@c`~*87X6 z&fQ&`eJelG5B+Ku3(T(i#Ialz7P^%+A%BfkT@#>&G%Xgmn`!k4e^KJBOAD`J5$~pW zn0i(=4jUWHrotu6M;Z+lckue{f5rATxlNvwCrb8;;9$y1FeK43M^YThl**swBR0j z;p+LsqG=hA4ZLO!e6m>YLM#zk8mrO_$6>EH*~TbKQq-fJZ@!+wHRu!f{lOcvlb!y& zKIVM#BIV3wxTtXvMgeSg>G0Vust~HoD!zfF(UYV1Q`+jdvA>w+rz^{D@VcJ9RPu&` z!(9qvW)BCI2c7_J{7z`%_j$^x+VdGx#F}oq+VakTU^UD%++;TkljJ(oH(AH%F@pBPZ zm@Xm8cu)S^G&f?WUsrzVIp>yXs(5L^UIg@sFio{XWucRw&AkrrQQXJW{&t|Gq&z}( zaErw>)b)xKPSbq*I}rDif;EIMI#kP_Cbqz*%FarnC9F-%^9K*3M}Ojdio>Fsn#i5w?c&n+k|90Kd`)ruu6W_}rl5q$VnVUU~uoAzj>VA(Iv?PAP=RgM@ zP_ap%1E+4}WD_pswL%#!Sk3*2C2FCbPoRHEP1H6CwJjA-utw{Oq-%Rz18K(lfI-n~0znmw5%L3uQ|kT zJlO+zXY%KrSF&AWo>S#*+IhHiZfXX1s;(pbVzguOp537&u$65@-xsWKJ_@%ttlcH; zs>~(5EIid3LdGmkC!IKF+_KQxMK&*elJW+{Y&Kk~Vcu#fg1Z$9$KPPIWfnwmC393o3u z$4D`~)7-b;M+OefZ3%Z)viaiq#9qlW;u3J7zZdQcSJLjdf%@o+1FPy0adAd9?Kf(R z@$&s7H}7YYPaEq62T3{M?u@0s@sMLE_3aDs6g#EdU=gW1(V$cy!GHM4ZW_qr-kYc;=P=UJ_q0KET4)vgOz)_;3KM9wZ3rsR1UWt&LKs79wn;E z?>M3@plpA-0Ws=cWio}!sN8NVC~(%@2$#e^r@R%rm)bFv#pn-602rYO%DBYIXsXmP z1C_%cs&`R9v~hZJOnIKv`rKO-?kzejUC1=u>wmiV)a~}G_}P)QSG!{Cz7bc@%*}$; zFc+jKIn;StwX*l_j=Z?`CS%0)<1SZx(Jw;WA;+C!(;mBv5ul)D-~yn;Qi034(Ey!` zu4WJ6(GO`>F&rE3zj^oBTp7~acj9z;I_^7zL$+n?E&a@pZBV^~84}GoDLhkiW)`YE z{;Iy4Q?yjuL&8e+JRckP8TlHc)dH0*mu7^-G1N8V>hspiV_iT2Bn1iCUU0{Ueyip< zrn~HS4rcj!ySi~?*_~=e*t$g5IEo7Y1v=B-WP9)XOIntviS339Z~66}-PC_By1Umw z?e^5)k*$+jIhMJfR46GsfbTV5#_auHpir!g9tV*Ixxs5YTT>6QvFDP%2_4;vte}#2 z@UMwGa&Pa+7v+=xyYJbp+`cmgz0J{r6Xss{)Zz{ehaoXtRO~~16U=K6oE4H3gqXK? zs0+(;RF;`r##xXk8kheN^vAbf>(-z>H7zI>b%$T%i8CN<+mktZ(_aYRYqqknQw z2qZmtL9Oz=w3bSbx^jxM*Xfs?59enQ)l2pCswF!?Dpc9zcZx&;wNU5FQo)z04BXs~ z&yVEB0wKDUa@DgWIZn=ppf>-06j^tUo8# zd$S%ia!a+h_Kw8%q(u4BF|>ayJ4W-VBU={Zr@zPMPp7I^aO z?4{Ys>+bC^6vIcDRF%o48J7!zIL8p{l9f{c$x)R*mFku2gKXaB9Wjef6fTyA)SOs5TXDGh@|Jo}TVwV^kMQC{UqmbJb{bpkijbg95+(J_ zKR1xy?@=~xOm8W(zxz(ArbRh{{Y^8AW;c^7`x}3Er3b@ze8_P7SIqf^;-lB`cv7^S z3tRKA=R{zO^Y-=AulM~Gbx0y73d(F`PRm^P3I*HPIJd}n{Kfvz>%ZA@q|?4I5_&)D zq}+YvM5n&vFL*yV*&*BFh#+F1bkMyx;M_poT_NO_`EbREbZpa(dLDCj^NF||ufHmN z#YyK%iU`u*R*%f1AIl=+EJo~O?MyiCn9^@dz z&77A7!|glYN>YtDBq`oW{|5NUr!P=Lb9N3>8OKFSv8_^L%ZWC+dgfupG|6GrmYb=^ zm==4fYn6k|-^UTl0DtQPCH54Zbe&~_iaZxMKaT({22~sU=-^(Yx|T-wsH2;ON-vRS zz;Z`tL2agAHjh>_wm4fBgqN2;)_M^5NcEnmn)ss>T~^a=Ic?I>n4d&oMFF+ zQf*hkwTq9MQ(RhMg^s>e8kzUxpaR7H0F*OeN=Y8i@m}IRYTPPC%OukMOcL~V7~~v* z`N+3st|Z`AbgC~(*u7x18IYn|D74`G6EZC|$3$|Bm<`RIC;Wwn#$9scH~nooo1 zurnpIxQjAPZ(?@Ew+u0_J?8E@*NsS%*{T=yr$RR}i}b*In$labC`5`(Yf#IY;jM^& zbO?Prn%%2r@ld_?Jc&OkS+3_Py&~?dtN{i~Lc8cE(-Ja8xd8gzo)mJMaLg!Gr*=%I zWqmnmavsi_VcIr}_~yy!DD%RVVqu znnQv`wjj}V#JMGLtZnRCe>LcU3dFI^mrRF%wvxp=UJMxd5_G)iAD|p0*i>CUSho^F z6;O%Ngo+jgds$Bx92%ys2A%MgXz%9~mW1s03cv6A>RwshY~WK{!cYDa4>Xx&>p>)M zsV7f^Jk^&0{F(yiRtagHUk{~{4mY0c_1fQSYiKO7y4`PPOf8nk-?Dw({97aAJg_VSlG>N6Jfv&v4nm}5kRk3~WxwZ&y z(QOH0sCx{_#Ufqbk~E5a<89!8%G({k6nE`JyX~|kR%CN4Xs1(3o(-x-1Hec4EB_U* zu-HacP*}56Ldj?yRwi!;^Fq71e}$%~{+RddPXWtqu}v!1im{{#owZ-x5CgzGB^ILXEMBw(a>^wqy&3*%+AE0;TisB-(O7UboMIWxUoP=cS>lD zqKI5Nsv8Qq zk@Y)VO6P4zkg-0es()Bgn$#GT9I>EKIq8Vh^V`Ai&grcb)ceV@l%lNvOD##5M6%afx$u>iQAbKE2gR{vhX1V`{u6wx;LA+pkN==U#L$-RW#sXXaqOVvQ%z z0-?967)~oW*IYV}f||5$W_;tYfpzN~-GhfHSl=ydE0^vr5Ze*mFUs3?T_?VbdC7oG z>M5_~43{St=IxkN$g9_o$`5YiUt;GJcqkq34M|Yhz)N~>iDSX|g;AzpH-(nijqczO_Os$b<9=4bsI#E813~LIfKG`?ebI>XC)QinJ5xz04vUbJn`=SM`E+$zcm9mleWg}z#I=a{E!P&z zTs+AmEVgAbMR`HAn7Uu6V$0SVSEjob)uwYWPd2RDO0fQ0)}V{>uab`WwQJ`SsB#ne z%?N0m8S$jr#}Gn{K!yTGWj!J6b_C-<28Ny~_5yf*n zfs>aJcc;k#KN1CgwC)vX_NXk|VCYjgNF5+^(!roUD6ox~|BGr6!(aj!@6CCVq zR8O^QL0A6O9M#7m5Rt&&6a~8aAH4fkFhmI#_yNDBHqz`YUMl6C+2TR1EM9V`0=~fQ z3Wyqc=TQeBK;Y`gcmuTmIRGXyi{PhDC$Qp3J5AMu>DK^g1_dY)tu8S2e0_ZVA z9s=tkPn^sAhx0zb`{>qmppoSg#zna_pJeBVwT8kOk|!MTxKV&((Om+b#`Xgcs3ZZt zJjL$q1_H(w2fQ9D=}dO;H>Xpkn0?&FTu2HANk9SV!N{p`T&~;!LuIN?C-i=30Ekc5f*}>-f5>fD^cvS3U`yTjbo;#8o1yPUQL5_>I-)He06ZwxGyC05WKocjxA=)`i zW=v~KL)k&H%>p{z(1{)B+LwT=Asr0_NKh&U(5guIPE_#)K!rIwy&ag$%2WV6+RqVx z9|Heo@9`lq_%>?M*H;JENbs>@U(NRtm%yLo?DUHj2RyiI_U<|z(40^1fC|BvBqQUdH2R^6Q`)X}x*ieyjIPAv8+xoyWSy?%IdZh<^}?_r}}OYyeei z`i>=hpY+UfKqw~nHN2q5UW6@X|7dSJ2w@=IW8NYdhsha-mpJX#LyFawuq0e2Ew2pMP)+ig=r^7pzCmkN=QiAH+O7GRe#|Lhqq2t-!EFmthE(fYcHL3$ zv?G|F8|WI8(^PuPETZ>ga0ZSSKKbunuQ)|-H;;;U@2=1W}6S?*X0&#MLx|4?pmji zHrFm~D3c4CgNHMsk*HFw>eB?}bosI@aLz7W`3mu?k$jKz%QJOxsb0b9&brfZ3bSNd zs57cwjIqU9^h^7XE0L}hPvh4Nq4vc5$3Czc6W@c+WkpmOOv8?(UV!(=ZvVde(^+Px zrem=dJ;xHUSl9D-xX1bl@5>i+6{^Sf4=W7qO7isI7IjC;RmySesMk&)f?1qf^5Y^4 zM=UDK&U<1rq*Dr}g4+15E$ej`wa`F>p|q-27@PuHE9?4(@<|F@O0CO+n#Cj5l~sol*^ zToX&?eOWogsqKx%ra3XOF6%h(#w%QS#B{$7PX*29&^@R|cUnL?^Z#gwK>j0<7Ooai z6BKqUT=eej?1~?Cb^N7YD2Dq`-DZ(!&BQckQ@xfs9de+aQZ_S!|LSbYn9@qzk18Dw z!W=5&3!VPYC$*&jbK`#z^#30)XU>^5<@vvy!~gLOskVWL^8e-^Z#_jK9InWZ(jg6I zUL=l{;_5;`F^T=Ql|3%{fBv^X#FTT%d?N?ArHyL|NiZF&{{|-<3JhxX8`S0~FzW?B z0RJh#z1-%9vH851k^YZi{|DRu^Zh_mjH~Nc81Iuny38ZbtjaYE1^*Brt3LoO%0mGz zkHIK`oM4t?f^~zN5&($|#8}k`2@$YBQDRov2;W6~&dWuo;!^-4wtoPYP&CFy%mlEu zrYJQav$JflrN&e~cH&@q z>;sgJVHGe1`#^g^y#v7JU)yi{qd;614+d6ziK+Kmoq;uu$ctLIr4>N3yiIiRAy+b@ z9bnpz?mS&;p9QHaFq{_hq}ds?3R~BaY7t0@N4`4zKh+}18k|5f3hkLS+DdT7BR5o_VxkJ(g3sd=)*(` z{AH!0khzUf54d?+*OLMKp1Wl)_2+QHbT8={jFFt$3pkCX3fj-TA%=N!&`z-2_JLhV z#yMwL>JR=|WYmP}pcc+NQW6}tZ8`saLq`wVMxWo~*77^dTIK>YYwjrR^2O3=GzN*v zdA#g*=)Z9aI6=cHS#|11mXA-+HKM`+AW<#NxJVQbsrUS(~UPyK`vCk9daJocC9 z-uWa3uI_Kb=s&{X@u#E+;|1@t6h|@O;gE%T@r?y6F_#`rDD#qoHPq;LFZ~stY*W|@ z`s@SP6~%y!K@A47iUIqAQ`bA`i*m0n2KVYE>&OQ%<+V^WAmhrsz*RCpBGi5eT;`Xc z_fy_|1IK}qWU*z_ z4iLbtNYMp&=os*DYV{nOrlbGaO|R6fFbZxLz;h@iFSk`EH!+g&@!pV5IA#^?#_0_> zZC_Rybq?DCn}KW{!#HG6vlO4&9gLrhMd({XDhEK%{g=QzI2*19!UF|HrVvTjdl@zF z+I6|-ObK31`TG+PU9u;C8lxGhJW@?PwBo>%yO#?1*Ze;{L;-5xqF)g^cjZ9LRq3U+ zwHTJOX%B4No{(KVI8s;Qof>>*X91pz>hB`@%d%PUJx@lrjfO;S)cWteD$h~-I|_Cg z1r~b$8Plp5iL(#vee{2V(s`fIg2UsZS+Fsk-1G!Yw%Zc0f%X6S?>y~4(dCEOWqXa2 zFuQ5;b7lxzS%d^1#}@Yd1UBi5#H)uOHFMG?aAi?OM}IEt<6X{?NWZ+5$rm8IyRV|M z#()+~C+vS5fdRF60*>cUd3o)I4XcEW~_(`0;C<>>iC8u^atrB#%q9~D(!{8 zT%@lg>U@%kfC37<)MX^|eH_kzLoalt?4NSBAjhX8fzd@8?^=XYTPtiP?N&RT29)f0 zFBzjm7w&x<6zqM8M^Eu(rtAk7jqRv<#&T{~!(EQuaQ)MoqFSqe1aN0wk&-XQX?oXv zM4yA6T0m?$hoc@M{XsbK3(gupkM2%Ev6dSjfike$2RwLv zBTyJ(E*=SH*rF*C6tO-U#tb_h{lC8x0=a;bon~40mDr3H!vVX1d6A{Z0@FM4MYM=e zb$mgZ{0y*>xx!=`(8328+0k1VkXj%Tx3}M}G)IRP+zd0=(*VjKrH~POaRRN5T?a5F z^Naw%z7Z9=`JZd%KFb)&`PkQOAs7XOOkN#Oe|??X+~e`2|HcBF2F&r*xZ-cRxq;F> zrGVctrYsTd+zN*tw0XGWZ%(|Hfdr!AmLJgn=E)FO!;{hf*4;l`ZpFE`0;*~dj!kKJ z>zNp;`Kzgf@)8V_r3Bl=?9~T<3y#Z>lFt~qswC@fnH=5hhlks=n@pir!ARqR39p8d zMXq%K_m0Eyo(GtCirs#?{rYDJwhZiJ-y)RaB`br}sn;cnoyEDvo4@l*ZHcXvkfc*T zcas-5D~#cJH?nC~{^P-E7g#cqJ9w^LN1!%)LT=5Khfb`9BlljUWB?b64YgoTQW*p~ zryZ%w7|UJpI(0PGGkyNn1*Q#l5P&#WsVfn%{yb_0oz~JU+0=kR&G@8YZUp{eHJp%< z%LbzgF4wMN3)G#(59Fhcz6$M9UI|sc6dw{VYl z677AFymC2@o76Ohf-5OE4^p@)uN)4YQF)$LK+v$pL}HBpZF=bbGV8f}s^}4%|p&H20at=-z8*N1Z&ak%^E^VmF8>64u$$Ko04~l-9)A=%Q1^ zEkq^VA-4K7GtEt`B8^v-&Mq%pJkHv+TTWj56ZqrFffD05@jGW>uP0H8ugvzvb1)*Y zqQjr73t?v<7_(DBKHuN5f3AcnNDV|3K+*_;qNTgU))S?{$^5eXkhaDVPlLS^qw#+S zboS&~Wfeln8zY<_TqsGuw!fP84M)AZD(>DlRoMh*JZ_A#t(DsiX5yjniMhWe;>>Nd zJNq;H?SK8Hy&?aOmyEyC30B)2Ad1@;BIjr^0#qqxL=J@`41~y55Ct-Vf!n4csPe^z zmWhnb9fHf8rGetb1~7siqrhX)ggqITDK~vWHZYZb9s|yDT@H{5;B;`+f~XQ8s{Wgl zdvT0kfv}txh`$)^WeEgdd^O$i2Gji{6eHzA0f(N3v0%J`CYuJ5NA!~Y2R^tpU=k{_ zfq04p3E^#^h#>mPY8$h<(*o~xq7!HoC^TTHIZnYL7Y^zT+-%@@cXZj(bzEIsJO29@ zc!AR~=z44agIB2-dk@)53HF4kBfv(~t^vOu3S4ZQ=Jrv?J8s`=LXj=a^a`DkQneu_ zV#SGE2W|uxSOce1BQTgcs+!zq6ok&@xhao|0QZu0QHpvP*p?^|(Wi$w@uM&{*u-0K zrX9k_f8NJz0D5OdVaV5`3qsMKiO8Lq?ko4!0j8K#j1R6f`WtMhIx#6pnsY4@8Ge|H zzfr;WcrqvEz{YC`%#6^%VvOtt#W-kB)}1ku8^-=g9ce%4T6nQaTI`EiFyXY0jpImw z9G3Y3H{@M+KspBH-VVU*k}>5Ap>FOWpzJXEFYk{Dr5o6nqru!@Z!#_lLUn3V+3X-O zV8_{cHe>@HSqECl8%*4(U=ja0=J9Xk@X-Tr;t;6bw26Vum8R%K04A&j;|kIEbPAPT zw&gYZ_uQr|Uf#jnHK*~G-DYO$H%5&Cn^xymU@I4&=e&$&4{xSMUjnooI10ebLW25K zrT6sp2e2a<0v{!cvH@O@s<-%7eh3m+D~*^m5@Rp+Zp95A%*5T~#0xfiHb&K=5t^{YqdYIWz257|l#J1?$Zn#(hp%K8MptY? zjW`RxKnvf<&_>CS%0l3<&u{U##?6>6WT>Nb?wdcduXD1BD5Kqsi6$jj?+Aab`-Chm zHGG-N*)!LcTV3~xLUCHHa6&wLllU!LMLFXyg1E+t_EJ2)&8U}TP*(V0>qIpZd`yW^ zd09N;Nm_yzS|SCsm01LCqkrffB5PFYJr<#;jeB z-k*<~IB=Hne4i=k;05f{QBIqBf2%`kkWht#Y{yx(pa|&*B^?{x==&uw{0qdRehoS5 zE!x^3H6NyyM7Df`yAb)yHWo^YoabJ&n#}pgjbfe7n9Abrtf6bG;HLJnsZKnV-&>40 zYv&vM+&Ny2XL%b&XNRlq!P;_F7A7fn&hLxGK&0{t1*^j=(nEY^BGTSo{7IGf|D4I3 zH=-*)9gwargq&miG$Fo&Wcs&u=RZvplAEO|{Kf2~VMy%%&0!I<lSLS%@yAZb>KE3>E!x~VEeZ(%Heuo96zjI_ zVu`FdEx`UPnSQ3DnwWy{V|8q2(d$gEpI69roKy)w=hApr8`{a@lw>4&b`B{&tTI6+ zDm%_OoDw4FbCgV~>*{?XhdJrNgyV{QNMTreGxrlykzQLe0n=U6Tx8!kHU>Q9LleSh4o8wu0H z?UeF146;Zwf+;^7L?I)2Udm2_BpN|C))aCboy$K97vFqhL{5z?tvOq!#EokSkGp}Z zLJr2Z0|({FjlIu;NYhRQ)vGNXq*m7!Lf=%)X2hiik!(}xx-)jth-Jk}=U7R;@W4Q^ z)Ngxdhc5$*PQ|e9uA2CT%5p`GQX%iIr;cqoy(84wqBc(HOEd@PJBOUIJRc7qof>7w zqKYhqp$g`M`pMjF`W?kUZRh!72}R1u=5<|@$7q>6kIn$r?)j|U>X8Xtly+bp2OI<< zTnr1ZJVg5dP^e9(V?4TI9In}O9AijNOhJu!~umx}D|i3|zU75B!HCV+aZ-st(K2RfJSP?bwT5UQ z2Nrz*cWtJ;L7GcJn1-jQ%TEE9+Zcx>l~E`|ECS+ECOyKlA5t2-so(#6M@Ij{&?gio z2?ui@hNAFcVEpa=SIY6LJPC`J?qT9!a2EH@ z`a-iix=M`t{7y7-I^B{?Mk$_~qVl=fw4D{rlNp6x9s1nyL-ow{2fA<^Ce=b9DyN|8vW`3X+pWmrJ`uM}N#XA0ZR z78lefUfqu_DHmf`Aj3c4fU7;iL!~B*Y(A$;B|R;i<-}6rENB}kPcKN#0Zm%PRgGM9 z{#VVbJAYP;7QnJL*&2(Vs^KrcG#z6z)y3YJy8R73`XNH7h-XUE*i?fzn`fS}Z@7TA z$^35E$ukl}noN#`dbegXTAHMI%#OKLQXxY+{H3B@@dtX3S@ZHdyIS;Ag##mJb}G-^ zxMnV6rPLEH74LD=DZ}04`RJ?C9CqB9^94OKQHkhx9m1;MLKFh+^;J;desOl{5W|H~ zILuIur9mnLrpRx6fIY!5vFh0dcwrqs;*PEw^Hk(zIvHh_{Wx zRo`pd#Pg4cKoly=tBJTRaH&0H6vi3u))tJe-^=P13Oph` z)qP5s8H?UKkFr#1jF!4wipb1C0 zQ1_zvv^xu-=)x2)xl^VCn<+!W%;}MBtL(7m>?s|1_t1HeSUJ5&)4{Zrc0w|*cVPs3 zDY>o@e_G|9@G-lioXPg1`S#|L&N9<5JhD-wgVNpN?mY(ugcNL^VA`9> z#7b17&q-~1odYixj`rEY-R{O;kl@SiyM(@exCd%-gI=oj_N+9e{wJ1YBb!q>@=!~N zXKhiu2!|{iiPD|7*uue;15&<{qz}cV+b^uXkZeZ1f}|i5%xY=z!a|+nxd<6?PDK!h z^2QL{_`6n|$x77fdMM=wOVl#@P9^WYc$_(wG{r&$$OxWS9ZZu(K>C7%dtBrT-a9mj zNQxA6k|eV;lHE~%okJ#wYQY(?Pw+CfZ@$Y<<04LnSL1R4Cg^H;eGpUqWV536X_}@^8vrfl$8K=rfN1ZzlC*3>Xf&kP9{5qD*>jxeK}zr*%EfI)a%bXzQ|xzKO>?$*?PiIgOX zmqO{JNeAbP>!R+-wPPzs)#Y@_OTCYDof5tJ_PTe+GHxfdMitUA8>k?FGl3u$CE{wkRnm9rorDaGp3Py8ienjvE{e9W|@e}u&$RQOAT zP9Tnj*zR%M-Hp2O&kVW-)h}_v}MhvOg^(YplrqSBkra2 z;J`5x-Saw|%^PHnx;Bj&s`8}*&pn!AWUIQYn_n0UQ~9M>=x)or-NhxNv8udEm93G_bB`2WCr0s+ zax0q;UMgjsw0JO^>fGEhO)bO^i+eewSXRri@O`6k)z3>v-M`w?}E#F4g|R)ef2 z)}hC6Z{3G!%+N+BD^hSp6E<}sObR#OS7*@uWz(Gqsl{k0iWw5~JNM!A2tV3;HF7}2 zpt44&>a~D^?%9~*yp+Uy3kcB&ZJVs|L+px)v+Ylpi*glXC630b-I@{mbFjSp<$C2e zCQ>aS}4p@EJD`$yvc6GQz+rcr&r&uKF@zbY(!mE z{po(+xn%lJF9u1bE{443jkPLxpdnji8Cr}?0Y(9ffe{jU;igTi0j!F;?RJO- zC_^ZW;E`zJUodg+;_tPJ$;t_QH&nEgP4l#h!qbu~q)MxHg+35wG$3ytjS*s;cwF1a z^j$-4+R#K>2Een>(N1tnk?yt{nXK=08vD(84gq9(q|m8$F4Ih->RwpG00(EVcqzM2 zMcyb@U;j4}vOrYPo`Fi=YN14<1uPT8xRle|@iu4W+fEl;vD0OHp68y(6{N-s=v;3p z@{i|X;s>WpW!@WbTUAmx>b5661^3r8Vnysd7lw(;hfmbj4yK$Ov_gXBv#)pQC+g6n zNNLVbJAAOnjp;{Jd=6ie#uK1bQ%Pdft}>n#$`JySs#TwVe5^Dv4AYET4Vmcw-T(W) zE(Bj8V^NZ~@Y{-oA*%uHt`;N?{^@$14g&V9tIE~SO1P71mU`f}SW?jL3Iu_S_{-xJ z{d3V2mJ+K?kAF%_=k4D4rb_20t(ms^V*0527#mb$XMzc9WAA8S?kHLZI(I$zZhdU0 z8D5>pdYB%Q8CQC zg|VO{x=^E<`2J(NUmYo4evzpDWy!FZ{8CL9x%VcY6Xn)7(j#A(@d%Aok;AsM+t@e1 zv)qvaV-emKO2~3mOAZ{#{x9jV_?3&(o%ub2Bj{-LXhp)ee3*Z%Ne$Q5JC6^muN&r* ziYk><9rYw{waWt3Kfise(~29so@9?Wd9opeIWmMzce_HF`(1cj~&L})p3|&zmIJl?oOhE%@^)P^7=B$rsdP{P36-xsojejhWn$)1doeM8;HUL zYJIJ^BIe0aNBTj?4Oo}B-TfQ6{AF+>B|Cqy)FXB^64#Vnc&5uB&K zT}1A$*afxqiI#B0wF+S>p|d5T{W=e)i@lQ_N5*Mr@S4X;Omlw$Z8}3KK$asSeQPSi zCAdCxyrhEmmGpkk*pG)fuM=S7MQ5Jj^j2!61OIgw@Kpy?*GbT}Ip05d4gbmZ)L2=u zOVb|Y(~n>MzOx?5AT3%vc;kn%^k%I zq}9h(+4t2U_}g#OlHgS}mdHQd%i@8h)u&U69>g5XBVp$3MInv(_Gx8=~=UlC(P0t)JWNp9}h2WHVVaNvI5OXK_6fV2U|;Rcc9$O?_}Dmkb+*5a|`Z`Tg$0 zkq35)?dRv^qW@r1S@Os>yj8VFx^p?)RfN-Tq^mpiX3E> ztah^>+F{6e%l}_yo|)w{p3ORY8)3Fl%eUc1dTG^dk5RM&B?7FwM7-jK zl7pixA8R!OTyuQ;VW{$soVAA1HfkN=Ftl&Oy8f^1Qw0~;95)kvGl{2HhBjeMN~wrE zk(jQkh!5M8XpWJudFAWn>;(HR+NCeb@<OWKxkbw^@$sdy1c6s_(aKAm_0!5}3{hHMR!^+z_&hOT+X8{x(;B)RFV} z+ZrZU`hsD3)vSkPyOruX?sMi*1vROPLzM)*OSA4+VHo=s~S#KM29|E!p%o45xK zP22v9!78`6$4MM#_O)+nW9J*_zn8pYkd`5`nAo-9?>vzovN40_MU;K$Sb5NesiA!# z*Y*BMJfLTja3|Po{lDCB9Q_p*hbGsD#6}How*HFr2$wP!mZgveCo0+E;HtInajkUe zGkB_q-kkN$3ElPR9(=qywMq+1f}N@3cx&siNzoxE)jge;wQoD2=HBIAi$+-RW=PRt zYkI%@P-@9{91G+navsSL-26dS5d@@+T9z>T~esAPzQ6 zijei7>ONzcBqn=so6fbFcGxfwLu)FI+dIBGLm_pX6)Q*VQ2s<@(1TlR44RM7Yl|LB zc~*u52cOnn4=sFclwKaZtJ{7>@sESxLsLxoPmNC4!4HiJH2? zKbg5?eB?~3d7ZXArD0%M{~M{Bx4vDfgeybPCT(;m6Zx{6<`h|g6F$6;%)FB7PmY*`~W14TRx@WDYHX2M{!o4+R z&pHA*tr^lCN%^Z!khY*Gw>azgNi?kUFmq|`S5-6N)Y6uO2kH2!z%AlEkW!h@TS0eZ z+wJIl6%7 zQWyDhR*2QPqIccg$6%AaCB>^+^9l0N4mFA7Uoh06>)9`msBSJK<`6xd1=kZycsywS zl)n{CJWLXsvt-5ASyko<-~XSD2ANN?=AgPqB!fJLSh5TArE4x}HWrKsH5RNI8-O>d zRvKepCid+~sBivdB2``fsjf3Tv+cDmDCa|%OA^$vf!=QH>$p+g z@godB+w<(%5_L$5+}YH2l%?|lFVe7areGM}daBg|j_cAUi`|pHAGU7K-Ar!bs1?ol z>yPQejb_ip{^gALIlF>qr<;&nHXpO;;K(_jwIWEUYX>G(y1{Z`L_y`OwaKMx+x(U4 z+Y`!(V~JPigq<+~#kF&BMQtkWa7}qybMK4mrTng{VYGx}P{dhhzgyW)mYTw=rd?( zUA>6MAS|Izm6W?o=~Am{%XKq!ICGGNrEp}lj_LxF56>q-tCR^lD~gg^b3uJQKRP^+ z`0(QCNG5sT!pTmQWD8}mI~#5WM9p2Vl;x}&{cBFs`$I0=KvzEXfz?j~81;g|>m>{e z(+B8W^xIw!0*qTdZ95XFz5HqI~?g;-_P)U1M2!xl#6#da~J73zcxKjp`CP)SVL7L3I$S9@CEzW zIdyZ#FIn{WS*166`p=g047L8GGruL!20*5AiYOv}$oisfJh3JJ5X?x1HPephIEj!r~OC(uWQ5Z`yd znshkJeBrl>Mdc`kPHcVE_L}HjD#_D5pdRq|gh|==D(<0Sitx|JFF=LI{Yh(?x+tUJ z=#PB(@@>4f4>DA_k4ut$f1B@eelPD&?XqL6|9Gvkg>_>gD#WB?EG%%oxMVllyZyRZ zAw3fh@*$U_;neej1jxr1Jdu^o#F^Sg>HKflUCR7&^n9nN*Ye4Is7ai8|O_VU|Hm)xhu1Suklz^J`D z6ZhzpHN;c-j{Bj}NP?APnY`tFZZyQeOc?|0Wyt{@>Y+a77l3E{8zzQO2|N@BprylQ zKUx(^eytzLS=N5-YP)mfZ)Jh5HI3Sz(5`)|ahx}HKVPC|L=+Xs1!RdzE0q*phvai@ zK7LNlNwE6UjAH^u0vzx$vRa|ck%J+0-oOeY;_)kzYNS4}rq6&=4k?LB8r&3O`u438 zEtV)Mi5!cG7&vR*dwbmC^Ulv&Z9o>6igDdyLyTB__h6dIr<(Pk7<&~0hY5X`rSC<+ zDPIMQr#PWl;y}FS{Z6L!uvxV1$6BQ_@~Lo?_Z3BoEf$ey7P|$RQh!Ad92V~SD)+mk z3In2s1~SLZgbH8fX`F5YYjxAM0$&JhRcGhBY9CBF^p@2-TXk=N(6^`j=Q=2=&F}+m za$H264XRMnwXNSup}A@!I+gfOFY;mDtib{3?FpmGiD#0i;v8mNuB$EwHGRXozhI5I z4W+PchEGbL&6!SGTas(yF8oZYW1Dx=+~#w}g_U>U^WV=R0v}9%^K70?UxsbdC6BCK z@t5;9Cx;{MOnN8z)54||o$rG0E})2J%6nPaL=8@TR=N{!Z0T&{%yycg!za8u3^YiZ z%?kUd5%~u+_$8{0Sn97`bJ7|!bH``+t;%?2vvQ~Z*on2y!*DMsR|PG@>Z6|2uGD4~ z@+6%bO~gIfKipt$S-?=F{`51EKKl4 zxFp^O^&G3b_k{i@xM2=bb@Jt$76-wy(eUx9DtAd7!q)hTl<%(?XHzeNz7k=a1|lia z;Es*?AyThVp>{i|oqFav&nurdA#5c{BMqh>RGQ<+6Wd?_%dB_dF&QblKQxMqV0%CN z)Qn^zn3xHkf36(GwXbd8Pc`;OzMcbLRP!ls$}6?LrX~d>rw&zCHnYyT1yy|T^zv-p zhdkAi)Q+$`M^TtDK#Yp<5($AbY5IN22A(&E)Wuz%$S(Rm81#VlKM_lK_R{x(pj>z7z8n4+BRF2Tc06V$x5>RDj)XPWu!30l34o)_-DIbu6N zql+FDh-BI;Ah|YC*(zKzL=CE5NwpHJ{KYlFJ3tSr&vM7_`bHg)SseA0@pt*){Yj6c zGkt)AG|Nm{8CzIXYf`s7hkL`t)wEYk9(_isFk1VrT7nLc3p}V~*}3^}KbIcrlW{d+ zQW?b@{)wyBU~$uDM>IXCvWl>rI_(yfRUjQ%5`4v6>uxg>J;_`2H|iF)KD|k~Dj-Lq z$Gq3SVuU@dp^2s(R7@|LU#OZ~G8Hgll|1liw17@%m!3W~6F)2We!e73I+Q`Cy^osJ z5Ud2)P$Lnto!w5Vf(|KG)h7wCsRw|72XKp0ZrIhS>p2NazKB;8z2YNk{G4~_H&eAc z&(R5|yok|AsVk2Y(+ovIMlGtozLRt3j$n(%U+1m#Ib_7s!m~8(}!G@GBwV7*n3GYjUtZw6rWeD zqD=m9m!sbpFuNu?Phm$HJaN$nlSuiLN7IMdZyVhVG{!)FVM0iyGDiu~r&mlrptD^} zO&$>}+McCfQO)odw->0Lp8O^sc2UVV3~cjzkChF@1*OIACejz7jS{1YP5+pj;xn=v z{&bD8ak$#YGpu^|3`9fp0Q$s%0k2%h(q~j$=a@gXLU7=zpKlkssZ@PcJ;(EEts|yH zb>&&_!L15qx&LS{$n*(+M>}~PNzCv*Lm^aHi906S$(Gi9(EZZ4XN~=1jypI<9ht)R zuCyPI?!D>oTK!&f z7v@VbtDO=c~tcZRFR`?>N^!mEo-eGaFa7}sFsK`(7n5OB$QqlLo#XPshAa&Ej$(Yzt~`b4TPF0G+o*5EWthJ%cBV)81qJSLd6rT*Ml zYm!C*##T4}X@SZjf4|OU7nJmv>s6ihslVj+TH3)fmSq!CbvMw+!AyIOI zFtdjj^t|0UW-06IRK$g6VT>j(<;crH&kPmx6EK}M!(~VJVElx~Bm#FWkReLk zyD`^VoZvnS3og9P|6@JREtLJw=4$-_@^WCBKo@YmJ*TU!+>dkgL|%;Q7+t!<+$TUv0rZHeDwH!i()_QL`_v$K=VNjy9T48R6zCZN~L`lG8@+F71x#D zyxC^mnYif)tuG6H_PGIPDv{CRzGUEN6s|n!1-(R*T$D1Mn<)QMpEpK%vpp4Nko1bv;TEo^`Zhn*Y=;%f`FW~CLjPP;73;Ut<*Y+z8_V7-&VHN;nFLygxwzHFsLfM&$fl9M*`x*+aBRXhC| zSW`SI$V_Ja!lZ$2_JEVd!ld0SLI}x6Q+WPy(PWbTsb#k71TiCAVIQh6v^m`(>^e2I zS;!K;Y@@nlxolwNdRYJ)HoRxf2s}0XG3z~hB?hUsI@a(4-ZdihDokC{^4vP#z5S{! zJdDAJHm)u2p9uTWV%6+0ZP?gBTeH5b`S0|b>qujbH zNIt;i`f68Ii3lbA2um4Lq)KKSVJT;3xMp{SWyRdMJqIjosZ5qaA^Hsx7%xiA0~N(W zO-}8EwR@2tKdaM>U%3iUt~zPnHrVpyhz62isWTduO8l&FChyp!fPnKP09P8o|6Vm`nX*anbHbcu;CE3kb^`wgN~ zT0n0mnJpF44U#qO|HyFvu+q_^0Ew=ZJHzTQ#a&MhK-_O}GSHnSIAl7dO99|v<6q7k zV23W5c7){$x(DzpnRLrO5FnOH7zreH0YYXFR4sCQ6Wrp1%rFoSNT!qKKG|lFQD$Cq z9cHjdYbVc!AJCJgACVBn$_>Scrr>%Su14?H=c#ac zvg2la;jSDEDu@fTU>f%8Z+US5hkJEBYC&h=EH{2ICV#3uKj2DQX1&Psj3WoCJs=JL zVWwT=n38BE*o+IuMj7!8r`D9Q5XpCU*|JL-=xA5!Mba{E%CKi(hRyV3q^JK>=R*=D)Gj)AJGws{t_W;eJGWj!ww6eU_ro7U^t}~nlN@*$b&#Hr} zkpSh&I8tX3;U;Y-@7_KD*j6T=BsSpFNOgH@>Uzw>tM(+n~Liv-*&>r&hGXo6r1hz&@6s|tX@Vx;QcI$5ot zVCz&X#5O8S@08{Jo`SGQ7!&S7OX42fCDkmoc*w4LaO|<{p#>ui9e@m2E(cuCv1?Jn zi?Vl%0QOWM0PntiRWxBxsQ1XnM#fFm{5_0UQLNs%6nH+y+BGrL6_qf4DlDKn4&%m_ zJg?QDuc>8Wt7M^dDgNs*TZwWG(Oj+4_}joq=XAP9dli!N3*9vO$HzPY2yYZ6zZ;tNwSlPHunt#vkzI z*DIX*d;?l?58C@ec=5W9jMWWhM=ZjdBy~9ETFhLzSy@u0gCa)<(W}IR+wn)(D6zo7 z{D46B1x>XAsCvaSQfdl}ONe-MX{RgdSaymHKi%RgVKRmMEi=}C@`VB|>T=HyLR*2X zp+S;g{pE8sZ6&U5amlRqX_@KARuz%$D``dZ6+|?=u&NTX0>!S7TX3%wJeX!O)u6b- z)2+3Q71zeyf%8i~)~vS%qF_@;PV@`$Z3(JDwU+joVckB8f8c zPT?BAHm+Yl#5xD2$VzY^&EHJS&iS5)kb>*WXWF+;1wD1Bz0mYdwBs?Mt(eWT=}_X~ zZFZu~+KJ*5WM{SmTypDXLCst4Q@ti?U!?k>pn24+WFmlRQ6s}IzdC#w(pmQZdBMKt z#(UyEt{6uhWUPDnQtQ)|qMo0!SlrhmUCP(aeDIyyyC66`V#e>15fk$7{*Lu!dXk6A zaPQAnzb{uWNXK_OjX&X$&fmY6o^o`{$F{9q<}||x`vbG~n_+V>#3W)fYm+Dz?1a!4 zzgxY*7mUyetC1-PQQiHj+(lcn9$-`#r__88g)v;-Fo0i}T}MGeq+`(&iyJ6K`#DgI z^!N^_6}=12Za*D=+@_htxk@TakA=+#D!4Yg4@0Xa>W3x`%%=1Tf;!eiGJbs6#%(9k z1MIHP|DGA039*O=4&+DNb)m^5$V49P-+nF95~Q`zQ0PR?O>WGg7FJgT$~Jb zcEPl{C<)dy-+FA+`9I(F!n~aqV-TzNBzQsx5(JT10d*v7yY6JkL)H2-lBwt&i@^G> zTrxGln(8|@DW_+mNuVO!|SH9qkFa9Bg5PioxyJKKK_=X{1RUj>Jf(?aOoZ3!CLMsmsBbV1U zRYrDh@emb^Ac*5Z$LcSz?S((MldUz{o`&SK(|Y)KiIJ$Hs7If8-xSyd;3l$yBl8II z0VIs1b6nNyYfPB?<^H3$X8t`2`@;KD(QBM1cDl6kruxY9)m<>h#^RM@g=_Dqw@i(G z0E86(Nvgv0T;G{rSYiigOizOIcJ<6onUkJ zj1pWg5;*53D1Tj*u#D}jW?eo6PkUGO36y)cX5eyQ@aYo&d8#R4rQ&nX&J!c0r9yl_hu$!_ zh)Nw|G8lNzQPefjLCfE@qE8Dlzeo5&1^$!9J@Pxz#Yzo4qHzI#EEx=WJASlbf~)4o}8iPGFIo21zNIbjP-<6_w|*T=EzA3WQFI5T(}| zvH$}>p6|&Q@Ns#kL~N^D0EG?yQ9%IWQU29k-cctR>i}jB8nGOdI%>;*>mXr{m=hxI zpxH`%@TVYAzj)Fta#cKJK}_>@fcO0;aT)08yP9?()*IB$yo>9JN@o8CW-Fn1D!l*u zrhtnHej;xCy`F!6`=9p|qiDOojokk_O@N0-4Euk0Q2hV@IWzxfxMIqE2ZM`^ANGNl NzP6E8wT46Fe*uapw_N}L literal 176008 zcmdSAcUTi$*Efm+qJmTrl_mkCgHj?LMS7JY(tDLedT&w$q!UV{cah$Mlz@POh_rwN z2t{fLp-As=2Jid(p67k9^Z&WdA0%eXO!nTh_FBKP=KTu|#alP;-y|R)xTUNlr%ga` zZJdDMs_OM?z!B&n>^1Oq#Z6mLmY{Ng0S9~_vU{ffjDVmfp7hj$82Eg{S;@eSfPmWl z^6$!wEweXp(8@tq-(Ba8gR{DgyM?udy9IEJfWQxA1#xhG;pFhf-3=seLuLaG8VJcnzZf>YVH~8&x82zpl)=zDwOR~H9b^WaA7Q~h+txBB zX#c&Ze(swyQs=hkJ41QT7o7fUxgy_c%iYu>{DS*4=~y+f)x+$R@us;XVDHzfEe5Uo zqNL8|;esynU%fS!b`BTfF)R@bmKBB#wfXrVGw_VJmBiLpnlqPVd1btr-+F@$(-Ztg zGrlq!T~N*U5l-CG{dj+6*lMfUzwJ*aX`jaw&sA(VVFW%MFCmJ4;M0GwHZdZ=qeqra z-+RJBH&Uz_FC$Mz59U4B^)CM|rFgbvc(Bp3Ie>Rl_|IZklkAz(^Tn2`tSAaRh zweww3RB{*)F_-Qu%JVY3{1;Q7uBRqAjw4_|RQPy*z2z_+D>MRg=HI2HK3e+0Q?)zL zsGRhWHq-5v-eR`j(GpzMqw-1zg{aWG|GN$4ELi^z#gt;VR^^VV_pArnWQ&?=N%?|f zn>3EY_n58YUQ}o#lcT6=(v&9#x>bCA^3hR#4%74UCl;MV^uqQ-+V~xm^u%)0J|lzf zsd}fY1INUzCsWnQZm|K#U6g7Y+kw%GZa5#Ma!CZ6xHysEBZ zDi4sca8utB@6NCr-#D_pEnN;ars0O*N09OAc~bt*xt}$2q@&S}6SYONLG6KO^F^se zPP1*%@ZxHx{cpR~0yi;BqE{`_7SB zAhtogqApLGdWe6R7nihCGRxN6ocjW0XN~0mll3fRpV8dRSP1+gh(bLtfAN7wD zc`@cvbsZSL6da-W6X$x&^NbIsy-x8%Cc(WIu#ZcyBdD{7#?zi8~;pA+iQ)24sc z9QbzM!l>8HQeyNuzpRN{+o+;eV=7#YGxHOD)e^N zJj}X@qdRQ6iKSA?(Q$w#upvz+Yp(PjB|n{f|4mFcY*&n&LG)2Jjn<^cx7NZ?J~-DfeXz@O|cwS$y|X5+=J{CaI%rzLt}?w{G_?y9y>a83~}A&X%Ls`f40 z*&O(Xj8RRg%5ZB5wqlN|47Q@__&jHmrDhHk_QmARr$TX6CsrRB_}8czx28!hbysQm z|D=Z7{3=vja<^Ed_I3N@#A}6PUdl(+8Ydvi%O8im-8+~K3d#_5v6N(g(!(6o%3Sq{ zElza)I>r}Yp~{HIf2Glkky%;>QmC2a*j?;2M~ZKvl@zWpY2JwzyWTh(CnKLMBCYdQ zIDA8_`E$Zdop3Wp_icJfZ|764`7eYNF)}IfyN@f|&yRr}gW`Rt%zpu7)1Dj`(+#=#oN6t5EiACxfBQ#BVz}Wc>iw7IK$sq1=!zU~dQys;dsxutM2Li}lZ6M1R zR`KFyu2gnawHnA!L1l^J7@k?irY=4@A!{;(fEGQI(ADsf&PW-alw z(uMB7LdD{;Bm$XguQgdc)v*fcnO5_OTx>DnaHRH+b~=31YuW?O%p7?YBC-C{GVS++ z^hXvUUMfV+ca8nGzHK$aEErR5=+}5~qu1j?76n7E-7>UILz&764VK$(6pTmJb|8a<*UN^ z+sN(FdKHEST3U$${y{))TPSi(Dvwyo^VwP6DlWB2QU|}6;gcCPKuvT8bdE!Tj2=;S zkrR0Az!kAJdvShnsMA~R|5*=pKvb9|A==jIL4J0;w}a-i;&A1ASoJcM$5BlpwL`1R zQ#8(*Q3gm(=K{{Lf%_A7=i(o$a~6GEWr#XOay-w&aLL=E;=#un;^@wRdx);4kMRYP zR*5cgqaqk{KV=w}TFGqZvWB0QSaa`WhkjJg=+5~+rJ(Sc(-!$^C#e9h(2(wRme;)B zWFucMh3ZB%Zzrdg+;+x6ca@9O$O?68mQ=iYd?7NtU0AjN9S4*vG*tL<>lA6LI&mP? zT$y?jc+`yzPLwx8eSZ*RTU9e%EBy{O#139+jVm-WB|%fA!IQj4QYNUq%*>t|j-HU4 zZ16myaW6r6>q=i`X6D&f$-5puT(F=fwMv+Aysz>=pT~64N8u*5ZT~*E&7ss@8x*ud zB*|EY4(BMM++684t@Fhv0>(3K<12yNF-I<*!j$-v${m%^UX=?229<5F!$*r(oHsv#4H z3}XPPybV`v45gI44O;DZD0i#t9{WB}q?0GDLPzbIP_%(N{ZF=vBc%?egwE@U##Hy~ zxy!sJEFu{11R2UUE9zG2zBYo?$e(gqu8x*$72D($>xdJB))1ICz^2qbXu=wvR-8Hy zCvdysiqr1cc4 zO=<>}Mv(Z*mQ*pew~E^GkSS{W`dp5T_lv2DKFrMwvoz!V@k zp_-d0=a@**((g`6gjHJ1`rO)eE~avf(H17s@PlkxSzT)Hi7?7vip`CRm9m>0`PD0V zDP_e9zVz&SU~Z?)DECDx*Cy05@sBz+gB~EzfOtaQN>$xo>C$I?vGn- zO|cio$9jCY<2yJ!fSZx+jjUR9A2Ug0f-K=Iz8}!CPn$A$@x%HSxqGwao0j1(VOZ@k zkGnvCwK9$1;f-un-`q{$?pdzt)sW+3?vldBq_zuHAa;*uR0j%@|&d(JE!`Fd5yQ%)vBJ@IUjAl(Ej?Wws>m$h2dt$4>qZR zSIS7%k~(V_1q(@zd`EQUR&fdm2pwrDaWLfuZ*KE86)l*ila7X!{PZJ7QKkLGQU-() z;tsFc0Zi5xBcN6lxHma?t%u34d=T~GzWtTjqC!8!pDdis9(BGKTAt2iaaM@!nPYOw z+!d3IvIDviSBDukNJfAz*mw}Um z7Dg=f`j6Sg<3cU9*X7118|bQ>S+8S5vA=*Xzg^Mi+sd5#N31E( z)qq7V@RA&+L;^7W$@%TA%h1Y6eQK?ST3L!{YBr+|0(fAbsKJAP(_OiaFE@Vyc2cZAr1abZ>w5MP?7Tp*Jt z%Gjrq!%wA;7u$}EQw!eOd=gNR2VH1LA1xm05S?_08Bo_NZtH($qEZenRO)=%?~%n8 zpQ4@C^c}aC4C#S&(_7{AT7CD(EZUlK8~m07;%`6Q2G;t=>ihA3D!+XE6;!@3{OPiB zQlhtmmgzlbCuisj_!2FS$9IsOvBi1X1I@Pw;tpU~)#3ySYGgSK(OayRMjaHDdB=j6 z`Q|>=e&Agro>u~FYwgAh1T`8Ch$k484btktaXnWC0YuU$!+)SZf9oGvnOM_Ty=QfN z16e!j)XT+tS|2o&vlq6ZO{he1SWPr7H9v=V3AgR?{>svoEdf4z^-8F5{&iD{$^l^)z9xau}1ZVyp6q zW2dbAwHt7EV`n!|G^bjPdLwr6SQ$1lVXM4Y>J-;re;jU%!HQ1keHZ{wXdKUwR|l-V zgSEo@K(7>%wyzbYVgs?`guN$Hx3KB{W?fk8JGrpk3d($H7YZX5Ig#xSlD;@rJuMW5 zlWR~&Y%r?^Ia=E*Io{O$VBU`!k(D{vHj-?YRvRZK7C%q6SA^(tDTcd*2S`2vA>%_S zq(!sqgmDXePW06>c2D^l_-7}al-Ye^aA&S@F>}`zCzidFc9MDqJz+bXkobVYMpb2W ztk5HP`k+gllge=>3jOq0eBRCXq=;&}nu8R}6qnqYa%dyhL?rWNj z^g2o-rF#;h7X@*6R+HJLRUVV#A`=n@o%G#?1k-Q;z6>^~`O4s$$=Kk^IgZV3_9xaF z-UB=VmA%K|?k&st<>Jat1%;lfNi%N%@@amw>mL{E#=AbjeF@F9XbYiPFi2s&dIXfV zdr8HKFypZUws5_xz9Os4T&M`@u7q)Kf|R}pS^jQw5Z)whrGbBM z+*hcI(o@eF5kVW%402^H`Z7FEH-_HHoq_EP8PT#d@q2s-cg*K}9PYK<{KErTs60@t zdu>#4`-A9)Fovc)bZn2TsUbJ{;quR5sxf<^LP1L>5xh_wAQw%G7M4- z+c>apDZ=wN1sc}L)-t*NYN5<`O}rG99-a}lO3DN-ol!;9Z!xOy7d(U~rh{`QywbOM zb;cfZU{i`0si_$JDs8xW&)8#=Ui^4&px{&qlOEi@o0L~b(%8TB>BmdgU$^G#C3^-1 z9k2$?>zwRjjw+y6S@4&rxU#2{T(TA?$@TV1W+O{TFIZpUrWOM3C^L5#0H&txk_W2C z@>em1()Jiwn3k`^-P5Mj0yen1h9)!5A~hd_W+i|qrAGi@aCrj<3Kl0dl>u$&l_BFp zRsKY_%X^pTgaYd-BcPLk%YMq_2oz+(eaO^-y(#|bBtLIiVLbb*-o08eF`1@Y(%9Nt zNf@QUzLzSsMO>Ly$E|RZ{luv^+Vm#gbMI99RWSoai8F8bTheV*d?#|=bQz!U4+_%0 z!wdVUHBAQiC^7Tgk7%|%!QlmnR*6wPx_TCVI7ilLrrU8hrr~XRZVqv~TgCBWk~FBk ztt_A9`bttm8FbAO_vLm~q;_|guQ%|?Y_v()1-oYHd@87Y%2Ny37FTkrYQ?B@dBV`z z$0pY?>KhSDgee|mvXR>bw-aT{t%m|y&DsOE8rcio6eOPacs=Jxk@Z+0t0`*!Tx@ZE z&n}ZH$mnX%2HEhTQE#CRRxDKTTvxo8;h>MSMB)=3<*ZQSROkLf*|it0C-B8TRkh38 zqh0{Re`C2u%@@-0m!71)(DUsS&TJIgop-Y7s>SVsf-z3*0cpZvg(bOUs)amr{&$Tk zYpWa(2S0@dd4I(j(*rlqibe8xrC_UZ+wIF065hH1)R#wrSFzt)Yb!mpyOKRBQb;CQ z0;XLgdz-IKTKfz%NETLMqGH-VgAN!xKH!hi!k~2&mvV08YWQP2WuH{ul@9mUtpNb# zDsi_~t_C*F_)|bl>9C~Fh8B%b3$}k2z3=8dl>1C|!Wh_hqn=j~GJq!49^10f%#(ee zYHf8<&;e6kUmx@Z+*u=NI!tHgKNm0JU{)KOkjkG);gP_mR?Ie1R}qvGs6b4d*B2jB;^JdP!)BP3IrY#p5K{EtIOj zkt36da_LuqN5-tN#;o@{(3wTmBs+6N2g%#0v}$VkhiP07y9r(aZ&e!X1!iBT1AL_ zJFrfc3cF@8W4JY^tD$zz0<=Zl*xLO*L{d*WPkOmzHO1J{VG@QE4chr3enEh~jm?sz zQyC{;&-~bWpy7HCkl6D0P*`n4zv{uoeYa_WFI#nx?bFW+olPlr5*AI10m-IdF=V}) zO_nj59BYUYBfzL_u+rINksjEO3t;+wA#9%c@oq}>Y%Ll#7UI?Zf_-i!%!?(i@kh!i z%$J!y`Q`&dej~aJt|qd59az1|2S<8`77@7KzchN^s1j;U_z>PzsjAIHe+T8cdbqW) zAl65*2b=P0?)L;Rhs`oadoig4At4T&e>PIBrX}WAy#zw4O(M*gLbi| z`OKA>*ND1^4?dchYA!?CW7xhHnRZt$ui9ez>(0j-a5)Wz{h= zzpf_+C4;oDv&g+;5qYlkevWL+MjQ99-BYUw=)3_cp99a1kLsQKGNh%XzCKmPOaL~> zyrxKWt@iLsVzjAW#XlN6=NTi{X&=h8{_(m5Z(@42CMb&mP~mQE6&KxmWd8Kj;fRxm z*rrW?W^)U>24hsRixhutH{D26r0-EXN2YTKGl7p!f$6_tC6ZKMkk974@=Fe-QC51Dg9>>L z8=w#jAHq}@S{XVa_vSUJ3KZkxWccz5JsKxpmmTVS#pVPt1sQe%$&4W=3Q$2A0Tve& zzS(xd59p7;P2)Sioj8Nt{sNX2VEppl8BjBOWx!Z~z&hV?O_ z;wA%P*i7$OTDvTdvG+3aLo{5M>@MFJFsxL?>{rfiGi+v36e-3t0n{#``~bzdoBe8P z6Y#`b22PM{92`jfq(xWi;^HDSm8>rRUpMg+Wn|qAnumvBpmqf5>h(A;S3p?!ascM{q- z?LU}8@GRD*X#4c9_7nAaB|jRNb1t6>2&dn<1(V+J+i0@mQq-tD(4XXT2Qm!6hESMJ zpM-F$f;LW^seMu6+7_X|K*_+w20PriHreVu{;@$-uVea5NYms8hdF(&qF;{kH-gK9 zfeQEdyk0odYWps?O6?O?XinJ^M}zeQ_c#3UG+Dly;t9^3%cYyeIf3Hmw`+h!@6O?l zx6aVB^-Co_IH?2*r|xUi_SB|Bj}d=_uW!09+s#C~srbUb3-R~mQ^)Q)hECx7eCI(F zY^EEqHz)@U?ti4CrR6iNwCPLz257ud*gJF-^x*PK_eB+o2h7Z9tioDtAHQnm)YwyU;p6?MwE6TuyhK7c1)8dJ)m!=tX|nQZSe?@a-}-W;gi?c)&J!b7FldKKW>xR(U-@^%CqoZAKW?#swX|{ ztmq)IZ2E_4bAVWn@uGO+_ROX+P4C{g#JvwU-Q2VN+(lrj80gsgH_lyC@S%;Fl_Q{orN65@B@}D|Mw^Wl?wVNaAhEw8J3^k{8Q+iLqIDU!Jn$ zMQ0_6+g$Z5fJ3kQurW{fuFFXlo^l4rvz^MrYG9K}>% zi|%oJH$33@<^^w8M#W=e7BLy*+0$kXSRp19<|wnojzk(`N@Grttp_~p=m3tePsV>! z_&@uIQdTz-K#|kM!m6@2+bl|4DwVz+tQ5o>YH4Xn79EcRwqB~(GTTp?g&3iEC)M$Q zgXunz4dXt!RY}f4{WpL8U_P0(2$xjUiRK8>&nw5XN<9O-m3$qAO7Q>!X>VsQA>837 z&n-m1G)mZMMiNLUQKKZ-{?EUSt(Jn1Eg*Hq4phO8;z&Leoi+?Dg2`*+@%v7so#g7@ z>uC+l9p=m^FnHG)NKxm4nD*7uvQts>sdfnbBv&|2Z25y=RYPs=hO3DegQj2#hp}6~ zNQ(H3Nv^}l$E{*^cN~iSJeiq$Ay8oO>?ZrPmu%z@`TuTi(6YD-KLRN7AAu#88KT3O zf)JmWdoL&AB58xIHZYBw+5(?kIDfOv+qf(D{Ak_vb#QJD^EJ-^%ua(8u{$mUr@JGX zo|j~J;lZ55h>ODC%-tOJdrc{aS>yI(swZzR7x7Q4dy|qlZwsTf#zNmS%`t8U6*{G4P(gZXVK=BogV3=d#7(ep@ zet^TWlBr-(Ks@ewro`%B^KUVR#EkdsU z`Tw}Y-$3TR?2)+SDVKx$_vQZ=Z6`+0?(i8}*Vzps-jULE>-;s zXXcXFT2*`25He1X-)QN}KDrH-vr#*m2Ce->b+}3Cu~%fkXoV4aALAs0@!#n`Ym=Yr zsQcik`p*(-eb7bp3y@%LZZBszO#(e0Z+5VQjkf*iM-Ds7{X3vmU~^O3-wRXC28a0W zt(L?JUWW>MZ5Ur{N5Kt1et%*LV(oJ%jUUt!YtUzVNMd(Eq8a8ALmKr|ao|KL>k%9H zpO~_!S(eGb$$Ezw>8AZj$MqPYfg!gBf_pN7tpUfl;+eL&Y?8@rZCVZE8vj2hy~q;u z{Rw$!a=@s<#KY4&!1SWm-mMjYfAHr)OIPnH{^2#O_1c^n1JY3YCapXdi zYigKW=G0kY=cf>I=mMB8bopl*VpxIR$K@mU0Qr<~{qUDy!ZXsey)h#v7bkGyEqm6D zob$62qYN+N?m}D*;2tZZq6=Lt>4B%AW-LZ@kV{IoJ z68K+baNs_2RY60gE{^e4n3Ce`*8ut}$O80mpvgz66~Hf^D|r!(}xPE`Oo zs@dfu6maa0vCV*
      FppZ~zPl~1-xFBIT7BH`sG`1kgbw{n zfjq`TRSM{u*46DpsUqlyN^q&EEIp;J9ED@Z_Rw^NQKRV;_;tZ5dp%W9Ps}~5h3?oo zCMjS2k&o)X6z&Bhu1s2djidrwb(1-byi+57owLeXe=CSXJ$(`SW{qgZgi4rpqv{7} zjUrYNvxB_n zQ(h91%N1X~btjQ;rDMN?gxQ-ghrb0_V#_2fGdUt#%CB2TKJ{tXa9_gA=(IDsA720kuy0n>V?!@XlX|+2z5m2dbQhgKCwVE9CaCLksqM zfn{C0e9H?M)tF$DT#%)j$S=hxpuhYeMCQL;G0^T?c#un%@tr9zfHS6urIy6%K1#Yi z<@-b6JetYKc2GvxENg=nxFhZry$!tdoYB%OC6)o=BPVNonUIQDjGkR*sZ~~DC+r;r zNqjP?rGVa!0?y6>&Q@GchPn&bz=y!Eb0FHk)?x8yz5nShWK+8Kl=ChqoJoXK&B>|9gj`UVj26$A9beWp7Z2YSz!es^&q{yx2ok`1Ft|&PF$1 zcABlq?w^XIcPDw3>*Q1HWd_$;xlM~LFq=~O%8}!5DX~@^3UDC*Zz-P?-x8AAe zNKYdsU-LKuLRqUd(Xak56L~93-oHYE2Xo(tl29PTJyyr>JC<8CIHKdQQM^y-`_o1G z0uSd%09NzP13#wTSx)vj9LO73)tA41h25Z819m#%k`SIKS;uP?C`7cmU3(R51c++T`LFL0K-x1ZP>7uu1CgEr*=K(VZf3YS zgM>Ex(u*!$jt2s40?5t+Epy@YYxN@<(Zx}+`T5QKvDO~HLi+o15626}zhppa-=n=i z5I67+(shz|m_H!TWR4o%dp^uag6T`{t8C7>|I*$`-oS zXXl$nuRn+-H2@e?e3(9B?ueTu)dUE!6m@wVXB+ zO&exQeM+AgLb;F;AjWLP&9Ut;-3Wr##Wkc zzMZMj9bDSER=vZ|}wUNu8p&dD}<1bnI;X9)e$Z_3^ z3D$b5gUFC*RT91@b^{C-yd53k&uh77=#OJR)Vu9R$KMGr-LX_-mh!EC^1*u? z)kj42ppa&={!RIOMAR|hgsN1qyP<%qSj%>NRcjfO^jx%0zaQ5S)N}>OZGIIXi zMuZJwqBG65A3NK@xtL{E10i4AGtPq5I-n^>@ zZ1dx`Q|Pc;6y2OMNj#fpU|x2K)g&jQ=)L9lKnZwW`9-;581;hrX;#>r+@~r86my)r z=KpUH#Jiik>~c~v*lbPZ?b>-sj#$&a-mH=Rj{9oFv&c7JEpku){8D{n*=3f8(35d`;hffF$exq@FWY?+^H)tKH)&_o;>)Z)r;Y;?_yUwCw zK4)x~&>!EP{MO9&*(tFUhdRf01OwzHLAGny-Jld{HX}|~joVixHDxu`DjiJFz;L>D zoF~oe664cv1^tA`$<)$lD$?P9NTyrTql?B;2dN!f!JJ?mUHdw_X5 zm5u~y9MnwN7vq;T-O}T{Y^gm#GT%G&-kCr8D-uR>TMkRs%@Yq?FJr`&%+3nayJt@n z1}E;^S{{sHaGEuK4fczvi!&HY`~I!GWr**=iTZV_XOuJo$lzRM-z=pC8$#j_f75(H z?VJzw)!Gt)ODPQ0@%7wx*2D&AuCwi824|e_)Jjd<5rsb4a;sy0}6Qh|T zv#aX0rt!(|79GE97Rhk9*dw{|d0|a6=s0N|XeaeN+e2gGWzIkAuN1Nb^yp|x*#P~k z`Rf2gldd{v@>`p04+3DR^znw_%H<7_cOi10UVSl#NF=Jx#n7@r70(c zZfil^s3-YpD>3)2rTyi>6QdJQokH36$LJ>^!-)SwQgo$=GzM}YVC<6vuhfYufe*LD z`yi7E-oYQV20!Dl>FO85kQueQriGt(2AtF5+!ME}-=Ds)Vt|nCiUT-v)A|H(!OsX_ zRmlDH%Vy}n7h}eWQe?18DRcsWDE71dJF&6MGBbdMw0seF_i-nMM*p~dzO9x)q7y`s zs`!#}u`C2)>Aghot_cQcnT!Sl^bKB)f$p|@MRZ3;!JGk$)*N3}MFjEnEe4J4avDv( zeEEAwwEA@|gP++_n85QRJ3m|zd@b8=W9v%NWsBo%PaLC>W&EwHtTHESLMI397f`gZ zTdYao(NdsP+XXP@Wowc5$?qu}?t2=pU;Mn^@FYXzn?v)HpW!o5@SILvQ_|0H z36m%DEOqY*`@LspJ#~t&P>7VxiqAy4XMLv->Fx+O2NCxeI#$?RcFbIs*s*Gh=Q%me z8d>67+NvLf?BW9cbdY!g?J(YXf8jIoV8(G!awq0Ol>z}YT$(uCe0`loO4cXSl9xus zA*`PCHi8jhMvYriF#>j580EeEH#Bt(1HyQh5E^U7EaEk_-;b-(1I{Sw?iEa(HfvNv z?GYPGwq7-kzEbcq+fO-O#lHtoo(4{!;qM93%$Q&uX4d5p%wkMwvIVYV=k|XafWGpAE?B} z!>670F&VZp#&+O+zHRRekxhrOGy6Ib4H3uj-ikp=0^fL^?dP#B)c|T3K+O0$p7fcW zTaqxkC+y)ydL+;)tHJvKDVPAX);UJk!eq(0Fq(3JFab!UYyv3!PJj}Gy!iQ`7K9n1 z+3MxUsRvlgER)Y-qWA0GR-j+rbL$)3$u`hw+pPnChfs{9Ia67Ep&eP91GIqhnOZuG zZvYP;%)^n3FxB#)i*xGj%elnne!o$1%{J=*Y=qUWyT+LHz%6MjXN$kRYrq!YjN?0h zN<_|con`y$RFhi(ZoH#4hU?vpKtMG-Jmr7 zg30aYC9-@keZlhIRZT3UMMso9*9OfhZcweMY!x4af@_B5$U~fhRTAiS?DQ3F4kd#!T54 z#I~}=ZH=uU_ksRYbld5+>MZGdsNmbRs2Axh%*$Jit=BH1w6Y`Q`~7l<=o3GU0y@Kn)b@nSXrY$6Y`HN}p%`vfovEG4JN$AGTY#5@>pLvY+?fA^X!?5oh{TWXh0l+p0IIzq)Hr73i zUc1NSE#mN7p`Lyx>x}{C80L#Z#S^^Q#cSYW&N=|t5C=1UCtD3SGTAWizk`z=s@KL# zAAdiP_6KzQzIp=^at3=qsH+1;HwA1_I5nIixKUH)tlq>ISJd{3%{O)i<}iM3^TwOj zEroVoj4kI)F?&Ecio8!CHIeQf9Zs;a)yR3a@Dpb{(w*M#W?0qVZR?n6Ic7m(_nKj|7Tzqv7%3%NGA~*ozU&W#sa9zQ?afP5UB&6S{FILXC4#4h1awv6=Yjz4 zsSyH#yGSLD)0xbI@dVgEsJ{`?W>9B4zPq|jYgcYNA*mVOc2^4IXX)vA7DZ=)#bJM8>_bN^O{%x_jFiM{S`RF?c z01RV*epN?6Ek8gG%KkRuzw1U{{1c$SOQgZ6E3Lx?ihd+25#)@m(73E@0a#pCZKiC> zbDy>Hbi3%h_FmzxN8~G8wGWh1U5+`4v{H$h$@&j-E>3&+Th;_y04KsA(712obG z7;2{?FV0WafqTQN1;444rQ5#)JZba#L9BR2!$hY;)5i@9Tw4Sh@Ym%_*m_MtagcMZf2Fg9$&dVZdZ&FmdZ$YzT1RI{_8pWB>Xn}4Z(g4Bc{ZN&qh~y(kNkRDjZeNm z4|ZATK3b7Q1qi;8eejdSH~O+g!<2!lCdyo+IHtD&%IRR3KEy9N6zMw+klt3SeSB5B zDfb>Lnp&1}SUw#E2q8yw1nlcNHb<(;VLrH}SNpK&-JOSKVF9||+QK+>-I-sqxOIOB z@GF@}$e5-$A+js{C4%>@o(M;0H0bx7ec=_(ko*Gqon~8boW_@qpL~ZfubElJ^slB!GVaVrIeal`XnPzwQ>f3byDUQWYL{zHECUA67 z{La1Q8~;=lQQ>E79lYrA$q_I}RV>yVZD&urAp2E`UZHU-r;I11BLVQV{=2=}kA1VF zPPG?fX4VWBiypbJyuf%b#6tV{8h5Rr(9d(>GXx~snvXW_`y9Qumt0ELIXyI4$21-c zWGONSjJ|0*vXx1t*I)v%s4f5+%>gpK?c&@oL(FY?-c;*U{tsYPnFKNevu@r@kr!sy zxF`3ZM~*gv$!NE7Ui8uFtDokUCTMrksjXAzcfXi?TTWKxI!j3=Ix7_?L%pwY`D>i0VzImUsCg~vy)yhOPsaEBk~ zR#GaAI?T2OAV;DR@&qP8y7ruZ&vE>CZ=qi{Q1|%>rz3XxN;o`JbR{r@Vn}uJ7afG53T{`}j@GU%Xz?RPKa(1l-y70TA*U=?>Mk>6fl-K5!fC@&S;Bj3pQ-q$b76$>2hHr#iE2aE3IS zZ<NJ7`9b0gMRbi(o(n}%dyS@m#CvIet7T zCau->tnF@R2rv7v`l2M3u~(wpa}spNa)(cny&u_K*hB}1EQ?iN%Vo=nf(7e^IJ?*Y>iY^7_=bh zxHy}=m}8Q|~pL zX)&i`n8{32I40kAV{V?!V7NSS%=~+b-ex16ai~E3d~-k4wC4<>5${&0Mt6Odvm+n&iqLD?qO@@Gw;Vaocctdp0^7WSe_HxDH4WyvQKS*U3?rodhsVh z=5UMoa4Q**EuKg!dN+Lu_$CwuCwaMLuPK9-+kZiCli{3G8oRBwb+t~u(tLBYl#aGE zlj*^@jJ${O`?b^Sxz&(uqc2lRKt~CMxHU58jp03pA!AbV-_X9 z`q0OsTb$_kmf*FsV7;e``BK-Ht*2>yu547!ugns=m9IoNmX8fdObIzw(jG`oM71zG zev8#>u*^2v2-daMxaeoHk}t@v&)=-t4!-iOjL|lFtYtC^FE6m0*|{6Ea|=7DItx5< z*n5677<3Hw-|po=CUM(arapVVbCr<980a|n56Ij5wfZ0yfRPvp ztt2TXgZP5@~!1 zJU038$<({jB8k$!4uJf=BpH*O6yC_?RgW1W<`LQEtI z>2Y6rHxx5Gq^2gc_5ctNF+N(N)2iNcVh+fcR|Z7`cdDCiY(MZ8G=p{_qxfX2X&0W( zx{s(!D^&?{Dv{7r2>KRVtc|Fy0+ucU5V{Y2p5A7ZDwIAyw3=8r3G zbJ8#tseQ-yq-NgKt~nI4{PmStnaCDvnTS^@E!yezH^(fiZ&?)8^0CdId74)JNVG^s zDd-+kh0Bpri$uP#6(?6WH3eD+N!o55xmY^m0dO!AUlieXyG!USRKCU8Y6ftdqy4Ih zSyJ8Ky7aPsd89wVjJ^p<%ZQE~2f}X1%7I~*mudi z%-G&&lJojwEpcU*@Vqa5!L&ZZI(Hx$aJm@v@g_GrBadG%Z8ELNvV)8%$R^_xWxuw^ zoC1~()i=%hI|5<2#}(jGwS2mZ1<&64f^jep3VE3O48Oveg;`s7)vw>9u_^m_6PUR4FP0uTt3`egs9(=t%1L7&}cZ8F|pfIn6_)U(_c5=%m)cP znT8Cc!^iW(<#z5E4Y$4@Ghb=Y+eC9z*l&1$9B+L|UM+^?ubaQNK8+~lcZP}2W#W{h zvgf7e&VLg>I*vRpcr&D>R@`_Wd_!RSMrL6J_kGc$9+~rDOsfMbIEhFhA2;MDR|?RX znHiU-diQ-NTy71!#rmbe^B72x3%5nNN=YkSew$toCRnK`y_sp<&hT86JdqF8E)Vki z6kB@rz`t~9jbZI0aox|A_01877F&fq*X8qyQjzAuHlp%QcrRA8z=8A>js)HB7M#o5fG;NjdTj1vYOh1@+ z(Sg38Ad`*N@80spKpTA`*vNlMY_^WX$46BA{thNN-Ht2+9KW&X+*!p2h;nHrpwn;ZOa(v!}PkmZ9@!L0Y$_L>}4s;k9i$o zbm62ZMsJY`#Y4o4sC)0cOHh%MOI|kb(&NaI(6|$VQKHKs&tzL5ZueRm_8?NPT>}u* z5>$Qn+8?@r_dBs?3D<5blA`Y&hna)oZrln@HD?;2zp5`wfmnk78gF_tE$>z_kbFhY zFJ|O%7jt#Km=MIG00C}aXw#coSE|B)F!lObr{{RARM+g`zuE8(*x)FvOrXlR_XL=qhQ#{W!7!~mNA@#iQ(Q!=u{OPxjJK7gI?V}Dbz%ksDNzWeK)h@KV ze+8!eS;3xk*F@68Q!!PC!l+r*gVnK={M`>^CbkkOC=QJ9o)P`ZI>vTw?Ca= zlbp<-?dA%vXb`nhi=hpjY{wh~4szX_<8qaAHe|*eg%po;C&7N@1`CyXt*e3im_=f+ z-slvQIHXzuy^% zJ`5uaiL?3yA;DPw`Ay>^wQE`@*4{*-A(?fswwIMxKl#iSZ$cH{qKAAzv1{6LwvP!Q z;)1^8{5W$LL#B8|;BlPHxi@p{_s>2ty?THL;$RsP8q0}BF+l43u8=VN?y0nF-}PjR z?Ugy-qS$;Sf4dDZ^v=b_fM&_F<9Z(17&uRTfOCG-kLTows1A|-0BQ!v^@5=hu$lQ& z6zK;}8-@Xp-y;R4#N$FMkO=HM)9%%!Z-N3Td$O$Yq8XuEcF?@Kq)2nNLr#u2O+GrA z9>zWI>%#Snbuq^jcg<<0cj9IPzt(;yZzW^Xj}!jW`g8C}>j#d8R^X;;PIf5#w%(SP zl=|UZoktocx&7VdDo3yBvms3%!KCDC_sRS3x-DeP`U9&!p5*uf@Vbtio z?WMQ7E>pg8;Fb9?!(MMts?IliW#p9ZR@<<8#^SEA4$m9py9sf+j0TLL=|~Nzr<%dM zz$#+?i>y#y)ryu9!JWOY`k0s;MNPSG?E;e7t)*R=l#NUUl zuVlR*<9H1#aX;y;##IgnR$+GZ&EX%Sqjz6PFp4(Gu5+e_c&g&X6EmFqIS4tr{(w*( zvd#P_L?%5aih>sJC7;>CSiI8vJw5e{rTWJtfB8p!7uIPPIay}5t?T~KFfm$=p4ead z?V^E>t%GLr<;@mAg4>BCT1(xhd8%6H>lk%ER+c?J13jL?sfV!=B|eAye6&!qlCydn z_Wuxf-tTO^@&B(KJ4Oj&uMRb8#aGWGfKAR6ZY%hm;IrgPO`#jqi*_Ft5?v)!83?*QDr@BG8y%+=C%YyLCVOfBc@v>e2x@nZ8DpP}LeB-iNx#t~cM7PAB zgJ+iowxihl&*+7zRaoO)UOLN3tfv>9SARu?u{hKVl*uASKhqQfhkRb|-+Nw{za{He z{&X`Np`6I8Y%kuV$~^Ds0I~K* z^~@?={(8xBR8KF@uAyS~668um@0dyGdFAyTH`AfnJ%cX}#!hTj7zo*zV9e97#1(XkFIt6=5O>c&; zs`!boaLQP9x}U!rlnd?MWMnr(Mfz5u*hbYKm(rPj^#JVUuMjG(7zHDUS0l?-Ojc?y zPYFy{LRUg=G!n>HC|N_LT=)Y{qz?P`jJ8Du1oRLZWrD3Quh2EL2xNS_(id>gw`Kt9 zDi<6#&Ckg2N5x?}Tk5+ZFI#9I!;fa4J&*QApL%YO=-WLx(PRwvEGV}2-TO>lH-_rd z8X{Lau`!DWfM~Xh(@yS82?PNp?96%dJB~wK{86t>c1YhXWtlNK4zg=jT$#K&`JLR5 zG*?_9>8=_1eacvN4*g-@C!deYTLBGA^~>cI(ADqp>e^a1ZDU-n+E=!O{jwi?>)*K* zE>)XX;q>j?tnRCoM(wA)4^4G%T3q=4${NYJea_6gcb+5jnlbSC$X;~VgHhfw)q3Gp z16K>E`!OBHph2%XvhSv*q6x+|@#~-QJI2Q|6tKx&oix?VmXGhi%X|Uczs)CJf;ENs z%xX=X)izUQHzCFYB!?zrzpC~n-G;vfzh`L`WEU_76hkf@MU&@1wlq!UXfE{eW2X%?rw;P4IhSh_XtRTAw$T~oglnur0$KjiQK75#Uij@S)c2=g3g@`O{cJnT8YX3Vg%i64bT6Gmc5{nP&)Psfhev-iw{s zZY$q(qslpVE!eJt`DnEP*r;zAylud^aGiP&qE9lf{gw&N(R7?AFTsBeP@MnS{+|O7{~($}HlOa38#P|$ z&yoX01)SaRXu1_WAV{y2%{x%&w>hs8bl%_uw{r#szM5ZxBZ_gF~f2}bCB-?iWwq69DC3V8wVSe`F@ zNz|T90RIT|3NR%IN+Xa@&-WQ1UtklxV|_MXm;BV$Uzf5e4I)#(H*o&$OHsJKo+IkS z%IMqIUt`|g36L|s%HjG()Vt1YXv3I5$F?=oz2b>&u4 z^o)x=$I>pFG-XGD%GU}LQNXmhx#1N@czc>tdF)sB(=B^2$5rk^tFF*f{&ZQPjzT}# z_BQvhMN-Z0)p%V+2=wJr|096wu_!FSAJf>=q+_gY8dU()u=Fnn3XRsK`mQ|WSJ49k z9j!Y~?H$$7$jm~(bMggt=TpS(-+PkLWM)}?8IL${=kb@hgfgZ16KXPAZ!sv91SQxb z;mUt_u9CB&AtExW@b_N;=}=FMH~rBOY)l#gsu#)Udxd}Z*De5`WMe`@8W4tfX3%ZR znHHNspzb3;r0R3NPC((fANZerFp%=jAbiYHYCWtUdyR*Z$)20f1BBU8T`XpH_rHP*3|1il)Oo6whA;A=kue*5_4RyD-mODKrsXLa=xASmo z`0~rC6))*l(_o(N>Eu?xrZAQ3xo4sFPr+3fzF-Kre^s-s9BWh`+lpe=( ziX-X2h!m_8TIG(H&T`R(e!hv~J`V;WZvXIIfR0;F)CK<%Ld*T80Ks4- z@<;U`D$}NpWxoe}2}pJ}4n6eh<&QlYPvLK`?TIwXlM@xx0f{BV# zOCG%c&uJi*RvrOy=NG7hutU*m)X~noHQ5YsmU*5f+NWLG2X8AZL2_ zVp$@z253}|D)}~DgxSx-kTW%6v;U=sUtgivu(41;YS#K{)inTRS`qFRy-vG zbd}qoFG77|`r~`!^YER|i)f{157YK*Ve`6IZ;=ZgA2E5rWdy$Is@%B`7hF7+Qa=Rp zhT9Ylac}>JPIXuHy5XD~HqW9;#i%zF6e^ z@AGx6&d`u+R2k^R?R)dI^SBD*{4IdDt|vd=PWQ0ODuo_Hk7Uz zd{y@$dly7yNOv}ee7+On0-G`uv0B~gd1Q*?L+fq>JC49kpm;Fj3?HmD%ygb58U-pd zV0&R~jh7gFsP=t?p7g)=G_W_&!IY~{+<55HVAF8AE~7UgD3CN!Pc>8IlGF1}XYFrL zfzADCR}qZ+^Ph+fCq5)|>|{6lqfJ~*$k{ILJlnl53?6ZrNN6I8s1-olRaJHsDBCnN z?wk{!Xj-ImM|S5#Ve=ovA(OVhmU}eD6Yw*vaaK=9v{esQ0ej2zXZEvIwZLI1Ng3IU+_oVdF8TWUan-AMgwBbIm*SaK{Z<$K? zucy(#@iPCOxcqSIcH3`nlL?O#OM}DACprc4v%a}k$|d})Y!0n=pROESPsu3@q;5_S z;|0JBMMDgl!}ejvusMiU{_ei+u3l!E7Ng2j)ch%s#N50Sx*#c7jmqY8Kz$(lxvMHC}Wn zOjkFvFU1JuthzB>Z1oBi!lsT-kp=DNFLNI`-do<-aQ!_5YXH80U4z(h?f_FgLB7 z_J3Z`e=jvU4)_g#@7}%ZKn$Lz3L>U}B9$%BWU8g0Wjk8?pFJ1|^@#s-TmmsU z<`n~I+7pjLTe+VOYY>x>(`***wiX^AUTgt0k_LdlF$qLBT!@G*dVr^9UHUUj{TN8J zYyrf;YBEOUVc#iN(NOEqld-~ov4tOi8Y^el73aT}OWQgLn>80(+#>f2BqbVhF$^u!qBWGW24PcNPuU;?F1kyNq_r8Y=mZRmq7AkaX+4qOVSP z49CzL=xpA0@x)Qd3NXJM>AB?gqla8GCtPrZ-^74PC!?Bs;j`cI#C-ho z45T$cc#H&e!k_>jfqN7{diCHXP#*Z4A^IgX&>!$QfWuNx_Ga<|N7o^vDXnz zyZCaDSZV?G6^{=mBzLb58 znv;MMfKzdv{rvZQJG8aVKOhzM@9qH0UUqlh40K;0rq797k?k`6;hi*n@fx?Prk-C- zD?IOyLDuhFi;RKDy=V~d+XwT_z09&+AKF-d_x=!DE`v!eY@U>|bjJRSp5`R8r=eIgH(C=PIQ?(Zl9LQZfq4V`tOZ)4P z>t~}4j~#U^HHAs4wJ0cj>F~=G<21I|7l7piVQa9ITu1xy#K@b(-;;^9vw#mzWFv zxA%|Qx+hkE@IO#3k=P$^K zLNxQXtNHZFefNPk#FGe;o6v^IW+!(#0-XTs*n1)_i4JeXs^D-s}RYCdUn zIK7+MnG}ZDmz{V(wthp8vbVo z_@sRM_?|Eyo0lI@PH@c1u?lG-m^J|3`DDtQV+b|3Ky4bvybfrap_Sx9i#pMPdLvqR*3+2@bxcOBOT_o}Or~}b< zKF=6h{SRt**%DY_DglDu^Pvr3G5ZKOxhFo7Cv9HzqZ6{6Z>R{2uGGNi_er59ib%92 zs^2}7+b*Qvx=Etq<{M$5rY1bVdl14UtbOM*z_J1s!B@xL&fM#07p*|%g>k*e$;MtQ zSSZl8J^xKK$XnHm<=O2j1ds#!&q^zSBFU$tC!c(`TI}gp*fr^1P3Qa~?MeLrNGYe| ziT^??>h-pq0ua>WQENbw&wnCV6%Fh?zA1Hwxy3b?AUIp;nmhq$m7oP++_kc>7sG() z?<}R5KBxXvf}0~CW7SnzeVV%Vu3}*t9Vqf+<;4SQ*Ssg6n`T2UIrA|U)nDBjthw!H znMI5}wN<6fm6hHo@L!;om{WL?xuQ&zR%aKbTG9X3R5JzjMS2dq!#>f;<)51dq$3MC z9N~#9tA*)CihUcB?jADiMv#5Fhm<#>JfvZ5*=l7v9Iv-GUhujEXd?EK`57xiFs~)38aQH;2cIav6eFnZalNxQGG-tXm zqR2R@x}}w3?}(L+^`#SLcj(3z$9z5kto)YGKNaDAsNmlWm$;Kcw5Cg!v$B?(@WdT~ z32gi#R`5bBldR-h+!uSXtJW>{mPs7aJxYCg$B!W&X(**0?}_R|>?2u?J4pbwE$Yz9v25q@5&^1dTQC9ZL}92SZcL!H$tDQqyMLnT@(kuxcZDIRmsc{7DqNNdPn3<7CT zV}+4|-H%&`S0JmPYPfFIc)Y#Uw@&`c5I`i^o{t5C*5u6!7ey3VM@EXZs6yR=9%j@K ztwhUQe*9*hBJGM+7Y{Bvj2zBV5(P5ExMP&|a?mWhK;2>2N1aP5i^eu@Hj z^^|dAq{`BHlRZ{oi)?>EkU7(t%%@8MX>|57q={{E0=S=kWl2z@NnYKYo(({?KYS**ZC*zA`OOu`u}y-f47;;)ug z^of297G7+JG^Ui7M3cFgxi z?xn5hEhAg2&n%L-?W=O~!{VzDj3ZETIXp$M6t@m*yU4pgEZ~Q&hb@l?tkTvG0aVeI z$l2jcODc>7kR)!k-U8}zYV1{)M{O1yG)ukkP_^c;FvWF{idO!TmyAg=lzpY+3f%U%ZJY|-EA<;M_r6d;M8{>S2+u;81o4-NM0~k^gV$#N4A?_JYewt(v3;V+{l`- zzVDj(w_TpB$DEfskHE*J1G9XWTOiK)K?!@eCMbNFe2i6wbzs|#Y`r%-4v%i=T>mZR zr1+3JA7NF^h<^#tJVPut{{Y70$0xiIZqyapdq7?A;pZ)wQEmTDmOX-{WQ?@_>IqA0 zjQ0y)Y&ay)&dZAOQ`ko0qK?v?zQ?p^f(^|Brz@e; z&sc)!-!E6?5OQ^()Y_dj%@gK2cfhO0 zwKbjh!JXs18NHB?)lDR`k_-$t%mtG|`b&!3EHr03SGO|VUS$tto0q{>;hLiIjk#<( zZ6>IuGY{2c=fkj{o&K&~B?S}Mh91k%PQS9Mjs@*>_H`>`gjk<);dwAOq1SV%xZCTZ z1ITYO-;Rx>5AMC_44BW>h#k3T2l8*oRj@KHVDT;%+p#N^Y`+-rg06uK`o!frp30d4 z&55TPjIMI9SY@tfL0-TDkSEm)i_e+(bo1wJ7NH}IdZ(DJlx1=pM3H%`?}K3w@`krN zMn8Z|l)&H<1dyhV!n!Oc)DCW`j8%=3_zZ@OsJTZ5F15V&E@Bw5bLWcX-_i~fLTgNA zE^2L;Qn~s)<({7R%RRB<#Yp$3%t!4x2ED>~B}_Pw_l%;)1#$f$BO9=#Ve5KY92ifL zj}K||L}9#CZsTBrHM0?s;+9!3hmq0>;~!qXpw}N3xRFGt-k?WoG#i!u-{#pT)`+V9 zZHW)BnK3_p2EA8Ms<4Dj4{;>yX3<6N>a#)_RTHTb4f&WEQX_VR4|>b?u~Jh*Ck8@Ip%7)iRH+5iFzKB`O-VM3@`WnkhfFlu{IAf`qg{ z{>G83#%4T+Z0y@~nn$m;4~O*1f@4C(!FrHE*w-sic2Y(v7L#yTE%+h7gtCqzUIZF7 zAQ~-j5CyXdC2p--dmVFZitapu9F{3}jp%&2wVv3l2ney~esJ<>N^)9KpwaeiZ@LTH znS5J!BF;sdb>9D)1dNVJ5BA~p9da0HTwwQntV!}J-v^Oa98uP7!~UMKD-Y2~Sa>qIZ_C+#zNJHlG*QYcrY!x#>K<9Q63U zVG+Tcerdj8`R?8spsHj%NRA z6q>9eunFsDBZ*)$QO8?dIpZNYqJ@b!UpRypeI1<(6MCK6u4cWo2+}AeVc_Xe>S2m5 zjk%4+0lpfa^Zf;VP95!FwC;58kod1$^Mj^fe}MFISvaN zW+S)$&EXNH^JyXBddnEiFb#|oRAJAJd9WP52Y+`7*QtF630S>#D79hJA+Eco<>o8Sv zS|80Nz6cr(aMu-|-ovXH#tkqbe7Iw`$IKCi0h>U%@YmrtnL;%h0Ww{>oR9~#0Q)@W zt+x7?YnhaoqQS9f z1SDRLVTs+Yh>YnbL>iIoL#NwOQFl_?#yM7uc7#tzxN%-#IvRq-Nv}oMlhhFrry$Pd zXV;l!sK?b-BD|lsoA%^6&BA(+zlw!XRU=#(-mma)jMt-VZXX)6T}G$++Wm#`y?G$R z!(r+oh<`q(-_yX*Dmdck@)Vut-SH+4EfAJz?dWoUiiy74Z{LD>okbKf!EU5OitL1W zkU)LO3u;55fw2BbHEj8_@)=OAA*E5wA%A-skiM)U!$-TN|Dmhs47fEx>;pzeL)~Q| zffWj$zV#%=i)KVl{@TOKI{Qrz32Y-3F8M^bs9#~)jnAG}hJ^<1 z%nxJ3oUJg9AQN&GGB0&q4BCim4b@XcgRU!NH+v zxctm8Q-+^3z1e7J16LWjVD)bnqR`cSZmkI#A!rmC8_Jy zGv&B6k^eQmq=uWTmgez8+g-xUX<($Fmi0{nH=$*yIsxc{pe@S#E9; z)q}FSJ-j&1(VXSItZZH&cGK4VL#?{>$l;kGr3GapdQ##cK<08m)GiD&kl@1?9oxpG zIFl92tzh!Nu}fl%RAV;vaRQ%tZOINkBCyByOU-ckxb}aKujp{%ed<0J&yr<)TS4h7 z$4p$9zb2(d7~d77zfHE?Q~P04$rhAXN-1|K1Yvk3j%O4F1Kp{zhOae-)A^)hGg_MNVpx3?m_kOjB34 z$_zLw95P|dL6qyHu|q_ki~Dgg!_o6a8fOOl)hJp4QUr@Y;uS5ttofs2yso)s6n8}Y zqC+9{&TmL@nc`JiG_RC_#ezX8_+H!*2(Jjgj+ZANWH8zpv4(v>nvKqYG||^#qzaS* z!bID>up?6UE}Y0m`HeT$&=_ zdam$<$=@(hP*EpSjE=3srnihO^5*=oEz}cgvyU!uW__4kqzqMcX)B{_Z$LyYd5spa zuUW<3xwe*3y26TPhf1w7jecA-!yaq4^R>?Qx};M6P?r$df5?0oIYym-`DQLM5jHTK z;I6ihDgDUGpx6`IncCS@$1|0^xFoF+NA4A)*cq7# zGE=$yG{bkvtrl)O5`^@EGxP;(RA^QT3x-goEWqMv5ICzby4HQnRH64oIj~|x9ICXX zxqVCKnnqi_{*-x4gZ6ci`9OrImsmaGzmGC}5Wv@PGWdR~HKMZ@SYP$Y;CX+&y{1-$oXyo(@{0mxlnFT9_7jLR>Ne-0*8KPQeX^3( zqk#rJ{tkuMJw_k(YqU{UJJ7E+&=e%GMH&^t8;L{X)EyZ-PVE<(*Sf%toO9W(Tx~CM+L_AhOC57k6tsd3yERvY0x_37a7x}uDPYc*QML0!M~7U54VZccuA1J z^{bBI2Bna#b33#qCD{0^nC3G2(Pu$2vWT|2W<=OUq@MI7+7SpG>TYk^aW`n(7Bs(syh2g^W`vjAwcNv> z^pe(LdRbkRJ>0uS0^{M5Sc+EsLM2tgwHB@O&_7TM6~cPhdFZ{SwNHi%`>8Q@X=O+i z#mWp#h?K;rt87F#B=+1s~ zby&RKys;p{lXeRe5yR_Yls$6_eSAcC%aytBb9)N04hP7w777 zW?|>L8oT>6RYSZo>@O{f(XI`()C(AwXjcts!M7PmvIbCXhzQU)(T4&IaWGA}ro zl-S@?O^cD8wmXHtf0_TN>~Ic(`pthBwk6;n*U@>zh%Tx&gqngNB@<~2l#BVxZO1?J zy3z9L$W#Gxem%zbk}&De_!hybR92SunwUN3P>en}4(?FPfYV6;lasCRt|tybgKF_$ znF<7FV#+n)FqqCJpV{OiZP0;6-b-7>14C{m?l>FTs|GM$=%dZN<=4`4P+$p5_H7S{ z9xj&WvhS{8Fewzn`GS>!xiI=20btoapN>PlJ>6U|TG}$5WGYoo&xou`Xd%_NtBI;_ zBWi1XDLR+! zw7*r-44c|Q?cAL7%9bHM6~*;rF%JHHJl9qMn8c z6fr4K;>lL-hRvA-7S6Y8bVe!8St2v%#kzL8YKm_}d~V0;8ozbOj~;?W!0yjAzLu6T z(5)*JDw}(TI+aeiFhMs^OuMub7R-e$Nzd?^sSmOaBT>*_6YwRIcz^ER%}7s~{hYL#j1 zxeDB1U0-^?H?`;|k@jH}?xp}xh)Ar6Rgi~Q7h=|XGOHh)Y3_L0oANr7<;b6P(pOrN zFeU+5CDPNTO8_F7$N|ruc?8)3agc*lrs5!7jlzahBhTU}p}J+S_&Sm>t8tPTn>c1V z_fhpU8qyV3&4d|Ox+;Umnm>h0zk@;US&JHc)gGgFl+X9aLXaMqEBi&ID=avpw`qV0 zK={@A$9k5G#u;pMqd~8kPT&Y^1H-{W)Nk|MI_kgR4c<2?ycu%D$MO(@kk2GbWQX(@p=pZ1G7ccVHJpGGHo#FVCH#IOeXn}}jq7-Z)# z;M%g9i&9D9iE;~DKI)h4M41@V zohA8e)-UJ^zo12Hkh2`_j%u59cud1l)_?w`&L- z&e`10stIpi8aV2aY=6X4L)oL6GVo$IpY#AdpXZ+b!Rt!kC;Wo_H$ml@4VH#1tLGWgmntt!`F zdRVNCRCZv|NEhOUud?oo+WQv2Aa$X9${CnoPuI+_6)7kwmh5C30{e?p#iF!_g*I+K zT4ia-a|^&1+LNtg#JDl`AX2*9?6w9U6V??E2b)%adI@K9ntca#ccX)g*E;(T?Yz=Q z=HgkAG@@cEGf2DCh&0wDmHA6F*CHQu-hxFcmQGsqOcyY zFR~T0s{mvmd&r^`yw)~Ds`X}~lwA0h(XG_fp|g_Ryu4-KcAnojfaGl}^ch zsdbd%TuOTY_Q*mk(j*A57=V6=ra+jKezB@l{Z%-w#_&R54lD<&0egtLAo-Ao^UYS6-6@K*cBS%MUU@5l(JG63K1G304 z->p7lsfJM^m5kHvVh(-eKTq-KK^!bfb%J_@)$|*(a#+<;wD`s`{y+mR?7`BrRf>;` zSUS22^H;bS79Zzc(NIcz%j)7*6%6O|=)NGce4oe?r!GPvf+p}~jNN`>nf{%no@*mR zG}lYJh@nxg5%*d;RR#HL^m;~6OqLOVUi;L(M;a7Q2y zt2|3(&tRW$?$A8@;dNu7gL7wg-NZmoQmH&OgWRZ3at)eW$!zbZezIRWV|6^FF4hKW z&Gx*bRMCe!@8_$j95eva+U>dt`UIcm_NHra*>Amu#jESvejIfJ<*`*C*VK%_$$vBH z>gK2R+3)J9~({O6^QkgX&U zJF9)3ey@vq3qH&1mQOZ$>H80&jY^Fo?$A}@4zwk2Owdj9`3NJoOY>aW(S^ZCgNoRK zQtnPOKdkwe4_ytDl-Rd=&Npx!3P_K=XV@AsE75BaPpuCB(}3?iKQ%}B#t^vuvQK}= z2W{N&Zn&56vMVP9-xKftAjFqVEc;hT9qwq>Vy5zE-m_Ijg~|IV>%P?_6FfstoY_y< z!NS;I)q58w54;!31K|*xIJ+x%{IKESAb}EaX?SyS7UwaZqwd<8B^gf?WDSBp3*&c? z#GqWrBcUr)tqtJIkg~JxFdPKy@5~M50o}k{g&cgpTXxZY)YIDn82c^5?I}1IE)EAm zl+D%i52@fUPo;JQ)=N<;cRUY9T5m>fMD!||hF0^^1(oxMNcIp^-xx@iFu4O0&i?G7 z7V@8t>ggwmBF_0f$9-9ZDKMgeSuH*ZU~IS+w->#ayQMem=Qfgsqn-0l%6#N z76JmQr0akR6q18)Ah4A%&A0peJrk#2hlsNDOCNHd>eEV1QDeLIkyc0}5Z$+V|4Bvx zN@v!E>^;QX?V)k9gBY?@5B?*FMKZ5(k!OIJ_l2cfoG}`>k!F@#P+jv5WmXCg8ZK9{ zPh~VXDzzxvB$8v^&&TF0H=)f60jcv9`i*=6HIQ4s;S+012c03w6TiTiV6%%}Yd5z4qS zc-Y~CRV6T@IrrHgK1bz@WVatzi|3LAg$1Kj$0?>4U@(cz6L3*Hx(x3xQe%u~9E|u0 z?AGdbK1C;?>3eICs<3ik??&Gozg2qRqV$yRQn5{HV; z5=0MhsG?aIJ0`}(VTwW}Ryx&t&e}m`J&_NAXW`Z^x)vC}x7g^)f{zV)v4`+pP+or4 z8L(Kt$XRmZsQ&}=de(UjFP>vXV1f==i5jSc znuB+z_WxFXv^w}8&G7#*F&W!PMtzQP=ZOJ90Ip_`$G?3u^slt@D>#FEc{0TIm%P$w zsMy8;Xo6?V>U7+l$|CiY=HclmNpVRJ{bh-ASv+&$~BzrVK_-#ilC{4daw;Q7{T7F!cHcQaP^Ik}-50y@xyc={?Eb zUu!Mdn_expf#lgFn`B%GZ-2>+dNG#+9XWn%d>u-2% z4`+%%AN;1q87_&SH+It09cZC{S}+}NxB}Y~+_C)});CA=Wse1P9LWCEtvpgU!oaX? zl8YSTzdO(WAl~pXS&avo@8|*q(??tFWNr0>^4nWoW*7;T1 zqU)eBv0FReCpuWwC~zzKi__R2xEJTzoxj_Eb+>tf-#aG+#rq{S}8NwtrV|1af5_{w`=n zykkDowUBNhdC2FoaH$_<0AQ^$jPsK#BgF;WK#LR60}?KvKn6Ri-JpJ{QE9NDR%^9Y z75@~73ZfCanI%{9BY-FUI64@`EhYY=^5e|mZay@P98mMAva>QVubgSuequiwG zgyHX#rNHtk7*!x&Oi;mrzkg-uist;3a=dLu^N ziFiwZC<*Wy-8=V`MtxXbZv3VT{MF4qc+R7fgYcRUiLuIqhvvIFC~cp;8_eE)8Ps#& z+jc5@XQB_=5CLS4C}9N6hfK~&ik)K3Hu(%YBS}EXYf6JH^5z+TK;+*0%{@F7BnM6{ zute0&W;2CrT2@%2CV|c9n&6(~nbx?a&z%wPFNjayGTa;ku5O)O3;gweO!SJ}E7nV5 zbG3BEc7GnFve=QiFGTG!+XZx-e^>EAv%lO<(e(~hc32rR+Bro||Z)8e@> z+>`ugH>feUVYROxK(CMAI7pI?R1|q>eRGn|nO!9T8D;x-p@ly#NppwSh@+&4+Pc%J zn@w}V3W=o%`7}Am+7K@o#iNa_=~9@>p2xrnz>tL2!NWC>QiE_GZ`)PkTFsZ z-a*O~dBq^XHI<86DwXyh9`t@RtAtG6!8?NGZ$`zjk2%tb>td&AW72LJgw5HCM&COd zf6pO#u+qtfna99SD61p3pL~}3boj7Nv>;A2Go4=U=&N|ME`yv&fP?l#F+P?~>U2A_ z6gxNaw9XkZ_M__AHSSu%=Eq0-rUFAEkK&ORvRECQ(@Q^Q2F}3g-8>8C{vabPmJOg1 z+|$l&7rQ;;{r2kCL$-SXJoTY^Mm-5Uu>h$s`rj7xojB4p<2=zP+0Ieq4e58|?CmDs zWq`+xSm!K>iKsQS)XIBK@RE(E1|N)qF$8oz~l6uH$U6l z2L!TJ@07#wkqM@kwnL+PsV5F|-E#AT8;FT7mcLF}mdr0U-pd2W&?_&mUF5BMRcWwc zBv(bDMdUwUAoPn{R}Jqnzd#wS>X=G8MJE*8ar}7Q$7o1u-}AGKkNDf)Ct`g~l;(}b zz#M>TWScOpQ_dhiscVHaWhR|UJ>A}`X2%fOycKd>Nq}<3B1cIrKk|1hF9;MTn!#ca zxjYhKZvPTl{lcmD-pkYnFH%IJKnO*vUGuwizc(JCx-a_z8M~9aEPXXK*Ju+x8i?8S zQPHy5tbIuQam8G3#9O?>xNpkIG6NEl~j~c=tPk)&A zH;uvKFHo6u+99}r+{ln0gzL7zue$}1cLlu$vpD}#L7h0z)7Lg;X$_%Rml8P?wr0QhkmLq z1;cDG{eycZG#V4cvT&*~q@rw*&kxo#L&eCm*~SumKesqhkW4TO8_z+PbvB9Jxd%_x z?fd}un`bdr9%1Y-P)vPR<~TRi`KE_&oS23g`fL4~;oqo*6J%Q@JWGDf^IKyiA%Xxv zPra5k>iTA3g5I9vom(G24#a$4&(H|1Q#=^Vt+c)0_QAqRWnhnD5Wq^MvJ{tONIpvf zQ&4FseDF;cps{y*o}lq~D3U?37pN65zW;cO^YpQ{`a}&80flE!+LUzbgwPX>o!8Gm zn6oXzx>RG}8KI-h2oK|8M|OqVKqo?IggK-gvqzo@KE~Q><7qzsl=~ zSPpuTbn4+TP5=)^vZrQK_{T4Rp+$)$l8H1;>)TzTmxrvWI1C~w-iWoo8{%mfQlGV z|N8{Yw;|@*8v+9$N+k3j@roNqyQjpM`Nm9ZkTQa(1-*!^J7yb{eRppnp6XAK5oOwT zzSHUPtdsYm-x)c==y+;#Lw}&BWhtc0t;uoYlg}!QKTqlr7}f{Fy3{9_-YQ?_)JjsI z@bXOtfUJMFrars_p6Y?BvnN2KVI5a}&CqNRqcBEfOa=jn_h;MkqguB!ngQ&26xdD} zplH60;T~w-eDZC>o!A)zGA+vQ@zymP@v~;}bP3Ld7?`>K#?K!}-i7XxqyRzuN+PgLRZGvDlituD z7~(SpAS^e?BnpFmDNqt^{vjhBrObyEbun>`RS z;3CxBXqfCkE3RKr^SlS}_=6K-Ye$udSfIlZ1n==(`gcv4p0l4@lP%vR&S<{$&}vPs zO`g(aC0*MvoA?0M%uuhhc2Y@}->A=Gu`t2}XdI>y8p8on^fi~(Q~+2D1LmSht8yuw z3vV0VtG+mKiskiez)}Ku5n+vFdDB}9iS`E9LVkVGwd21Da{Y@yLtXzdbAjs0t24wq z6Cz&wt$jp+TSe@hEg+XwNG!6*0n|PH3a$b>OP)<}n3=klL1t)&^iO zvh1UOYlV@NSMJ~0zvqR-VGW0|YG+ubA75_KD|k%F1OfaRcK|P|n!gmoYA^z{ zq`tt1Sb%9y$lVzOWPXGFZ zU`lB+d(!HX6Up$9nY&5X@a|%}-uL!Nvk|1;Cv-zrCQll>dCD~q5-`Hy`C)+5Co!<8 zczZaY+pFpx1#ANC^>yZR@V-hGHFx=Ziq`4YvO^O9uZG`vumKwf+k-DMdm2%T|0*I^ zg;4m^`)SZkM}_5Unyb|8aCRIAXzZxp3(K$jkahGvzEExUl{)nz)$6-f8k3?p&)a?0 zSl2SSYmahNf*K^|6>^fX(_=z~PXO?LQmwH|UfCAk+pvJeK^-+mx!qF#h{*D6H3D_5 z==c5PA6d~=!7in*J;rs8W1pP^Xiwi?O-_E_ZGC~UY@z4x2sOCe{E%jXp?r>2u#Zr? zqR+!oA|h=}_I3am)*SzRZCG|+x{9sk&C&LNeB zO7=I6;N@PR)#6Ts{#yZaudTCp?G)Od9|Mi9A4AhXRXcd-BXJ-dZ4)r+g!r{H6%-j( zsacS7lTS0PW(MFUF6R4y-08Rv5krkY7h~w+;!oSZlKl!1;(XyK-?|wgn6k*fOg+Yw z0y~Los94V;6JYL&+Z~R1d%B4|@6c!c#VC;TRY$nwzZZ&U7}4}s48DL%tKBi zX>DgBu!Xhz=|7N7xxSAGtU%JlGk`OtSQ75z{ z*E4&WBcED4s{E^$uf+rBWj}@9gdfhEgKJ|V1fUbOxFz?&;cC_&PkXXs~Y+e(2^5^*%dteZV zb-xs=y0{gUgD`Qn#wJicYqg&njmK*Vb3={x|(Lfc?YomG8X225~3XHtVwDKi5 zhD-r%u%EbdCq+&MPzs#AMD!tmrri>EO;{%x4C6NVA^&(>*CRy(i7CC)hPQ6!`U zh8nuNr9ng#7(zlqngJvwq+tljp-VtQy6f%%{oQ-t=Xw9o$2oImpS|~4d#&&KegFx9 zAHW*06Ac@DG0yKThn)|Ho;K8wPT|-9I>Qu%6M^uED@LGE>DR*FeUjBarJ-XifTZ!2>{XkbQx5C^HFT2hR@m)A z$5_L~pqJ%bw_KFr+K*WuF_yDpCuQ}AHdIpmER8Ams|VgmvYtBHaeEQ~Q>IE7*`rGx z@69ZWE+sj@fbnR_4l0HL_OcNkM`pp9h7%_+t4nj z(xy!JW`C#!N-njRh(y}93`y)*&KryM=b!LarLP-ajN+#`3Pimyon-vJj2t#JMf98qcTIV!5MrQdGcxT5>Vv*SYVy zxm!I5`9Ezf5=9g}yzss70r%L+AFx6=RkMGMs%BOif{>lZ0g1fgf*TR*y_B`v)balL%_T&g`B!mnZ|w z7zto-cOyY$Vl6xtgvfKWNq;=@B0U8%oKBwJn#3Exg?srV>nzG;W40y;YPTG73%4_u ziH=sFYQG7>a;8YN{BL=UItYryg@mvo8_|HK8qlffrNf_vZV9#qV5j{ZbPP!+>;_zi z$I^mnwF^^dGgX)e@SCVvtokfQ|HGEx**Kh!KInN-#%Mm~YS=0n%PpvtV)X3$+J}@J z_r8JA%(5qarwlr_7k%=zj@F=gYb@txPU_6&998p@XVbaua=$!a8}7*&6O4tVM4m#^ zDDhVMHk9}8mvgKK)-~*?SbZQqJhSa>J{st5^;z2;Gc)z9QY=rFw?vKS?=knA#-G7a zz)!s1XoAl=Sv1nhz90osWawf!@U?0R>lOX!nh`rdd-ZCw_Mfu06T-!Ay2bWZ%$|Fq zPXm2Zz{TqI-}TGm3QojpQDT2K`=frF$=noIp`vfz4q;(;bVD2F!-DTQZvx+rP=jmq z*X&3(o1W#em*NsZ?qcE1fnWv3R>E0|!L2TNID@s?mlgn^g&w&cn#Ne^-`&nZ?_n2c z>5D`|gY1h<=-gY4dJVc>wA`WL5DdxpTZ)hL5)6AYv8fMTuW%anW+1t-C1_KMIv7bL zyokOnv$u`gQt%+a)J94NCt`j_r7o* zhhIw;jmj$80-OQT>8D)XKS=k1vf?-e=CM)t?VDmJ`Cb^sGbMwyXE%-wvXh-l(cT+{ zm2)?WXsTL^pNc?!N;8iTj6b^QALsIQ{2QCB)CCK%aMZ1vvE`r!Q$%e2#3t`#9Rln{FprQkj!(Sya@4Cxuf5 z#1WD+3Cp`mbn5r0Jy3$HHAF%!7E1}mYq8ud^${O$>e9WVse9o%IcDi`T&P&)faRYpAv({ zH~`ctkVG)dor`g{d^Qrg`kHLd;Hl_{uOs80&XWK`8{1@5^@VNMUg-1aKu1QI5cUAd z9qh0^ofyjSWg&OY_mjS#B7`OEdUSox_Vfl~lrzN*etf2Tj1N)ni=Fkbr%muJUC?2@m>30#JZBCuoGzs`TxVue7R3u)Fymi2H2o-aqmr@OfFPE(4b? zoMwny`_KoOEmAU}Tp5y}SrxD^EzsMoGZf>{-gwTjVwZn=DB4|-a>)LR95=_<=c3zA zs9rA?GYlH;?>>%nCSL<96fTgj@W+c_>3Lt@`WniApzdb&%MLJm#Bk@gN=Pty%#Y8m zprNfum9k0Ga$&&OQ{bE|=CJ_AZyKsQ9%xU5aV?LM2lhFEt#@aKrE5yf)!7JrNvL;Rlm=FweDHaA|?<6blZBf3PM8(RnbL z5X`rfugX8(qjOQ95W9~e#*F{mGKho z&r%^-KLh$`BF!TP9D%LE7@7z&AR`GVfZ=l5 z4{h54OlydEf!#!a=XgVvJJcXT_{7-OEed62_JtwsY2>%<>G)`jxJ6W|lC9t@z=#h3HQTyun*#mv=qc0n|=wzfeFuGp{$s28PXL2e98H}Gfv+o#P z?Hm1Z!rN4@ynZ+ce&L50Q-SQs>2u(>Cd+JHf3|;_aCm|M7})jus9+-jYG&PQ#UH%^(hes3wuEn^&I( zmk8MgK(xy{(yYpv>^oC8hWNZiHC3tEK>qz3K`cxb)0;wYU5_zy-A?^Ht`BUuKe{A+(qGW`WlNC4j~wlSyu;CdfX%7V`DUU+!$lw%WH` ztqc|DUZO~NrHZ)P-%rRcve{D?9)Syb2ur-0hV}dB;HkCz150a2f@L|5&wKv<;lH0R zi%iS;s8#n7N+>3aQmm~YPq+9ts<$LF{a*95!Opu`rRrOCujZ6YNTPBQQWGD%CS;v-O3Fb?9qH) z=s<8H@{T#=#03$Am`nKgyGx@7oOZGj(WAX6ci{34HKDjAPX876?~{@>eC*zisu$i8 ze5gb-HxN7+YWfen+b#}ZR%o`y{H?E0!ZG=sz#Qg(M8}}6)7<`1OJMZaQ!by3ToL$x zZ#aBWUSecXnU_P)SOYxMA5lM@)5rez3?AO}moSP)`6od4v=an>7fxCJyg8^}<+>aOQ0Lg;)HZ{-q;rrbRp^k2 zj8eORM6{5eG}>?EYObrf-lBvf?^r`DKWN49yU~bX3>QBZSN}IT_vvlz^Ev9}q9E=C zLpD~!pCbp{%I8f2b0q(FYuPp~V^0ic1n#S7ruRPvJWQ=L)ghdA9Qjf^Y zYv2BKVi5bkH_d=xp6u4%gIJbWe~`(vM!2T@eKsL$^Nza7tzhuv*)HI$w$F_T{ykO0 z=QZ97nwE8TOsxvHfwAo6xTZs;q3@xVa*2giC*x%v0@vXrRF7<^uI7Q?T=N~3@^sM{ z9`Xd!DRD#$Bl7#j6m0oFL`UpA)zxzeiHjADWN6cdBfF`~`15}4CW8z78y#WJMAHg` zx2W0)-jB)`-)4b>b9o%S@^Zz6z3OId>g2=weYLEOF7Xj%&9Vmdh0rOU9s`m=h@gb|3y_#Q3L+4-G>5X8m3}lwK z4a6!06)~%(3W-q+v0zHy>VJYPtd}G+f8*a8g<*X)0jS&;7M+wSB$aX5&KyJyN8LY8Gp zwZ84t16c_YyYI9v&e3IRH*D0wG<;~NkKpPF-(t=TP6FlRh*Tsr8th(z)RsENWSA@< zi@Vl{n?VaRrf0)1{W#&03T&6+&(1vzu~q&Jy@ezJR1^c!9a>}#pkoPwYs>k2i(AiP z1W62qI~9m!nvLADM-~+?1XD!DLzwvjF?d;;wGC zD@KeURFv&}!w2qlY6rUc?u4)Rh-K(m6H+&mTk`ySeH3aG_vi) z)r})NRQO{ljKQRo!D;A#K%ZsS5zv46`)Xx?rjg_EKS@BsS0c{b!Dg8_WV0@5NwUZ( zL$bib@}vH|`gANR4deUSFl=+C9d%t$#P<377ni%ppl5qWFW5>=6NfAfhXD_8{VnU69^P@Wx_mV!JLyd0muNuu#%p@1b2#z_U66yuB2IkfvebDF=~tr_)CLls{K0Tl zV#UsjEQt*IYqHGrn|7Taa#(ca=x5&BRs37=^k2gh>H%G5R`S*pHd*R2p*-Dd84oCV zxO5VoR-waYB4yj`dGStbcS4JQk=p-V<2PlhFHrSxuJ~UOXJgR4Br~(+2%qP2Vb7jt zM_Fe5f~DNn5_iL$hMl624VB!pRu!=LcX zER|l7w7&QTMHbBWBzLsft!c0|!`Qz#S%RQEoG#VZ&h7NIV`^hzbJaCcGS~z@CMq=#E;%Au)4cjz8OuKz-in{{Fpv?ZAb> z4tKlgSKL^6h9d*;8v}tnbhU_1PI|b2Mr8c3>yA}s4Cn+y)L%$wzjY(%`Z@e}PEZF9 z7+|YZXtss=r>JcR{(PK{$|4k`IY=9Dq!URN64aB#jp#Z%sezJ9Mz`9Nzr_4aZ&$ot z@AZtPMhP>l)Im&D#51n{3s_lsl7w}5erT;4y-omuZ6z?wl_UCQZ13V; zjyA{L44zo(fFk7K$a^*wBv0;kb^Oq{?YL~PpDmSeb}|jUIEqHay-G}Wh(bO16hG;n zhnkOuCp*5^D$@Jk3#-Y$K{3g~>YKo7?KiC-uCLeK27odEx{oS>V)a7cr%nPA$n+y< zMyk-soGXG`m6ztkLi|VLeSHR-!CZ3l^7$x!v4hu$hq`lAY>5J)cP%MjlZc@3vR)V8 zPtgj<1)jSAJ2sl5mazuTZWX}O4}}J70^{NA3rXt0Mapw5g0 zCWewgV{>7N_5{3bDeH({6(NaT{s}5DWe8kqGBmgB>EJ6s)z70!LibLuX_>jZ!Ij~SJ|1;0IKw{70QNn(|spKUG!fBXyeB9jp zxje0V4vIclHs&4Nn)*u!)0fP@Tds6thkW(LSkET0N(qk2hO*0;RcACbHaUyNGXTZS zExUuclNEpF>TpX~WG~ySN{bOm@l!&{FZQKeG_Bf78<5J>%-ulDM+=?oq#~mv$JT0h z(q}rzRRvb(nOypugJ#mvqXXF6+GtPa*tHuYGH8Q9%Xj4KXuD*?AK;K8n{p?~l+Ne0 zWD?ZgoB`f)Tn_lcp$jXz_8kwL0rg@OC_!dN5{ml!FT*ekW<;BC%w95(`ow%zqFKxg zdp&QANGTN1AIMi(XZ_Ot?GrrGYXH`41Q?;Kfg_sVU_V%&+pmWsHBggU?V2Ylfo%8l zI_bFGfiKaboo64IH>2An-8=C;fOw(Yq^rm*cbMEQXG^H(Z-6&z&o*#NMwNooya#K zmOT9IQc(^7eZU=Cs|Hl{)&UDI?5`W9U|0fUlCOPw*xeKLdVaVNZ%WKH8oM~%q}x6L zT3u5y-xGN!*7A0euGNUGoXLX{8xgP01-0pKm*`<~nT5!XSe6q5g))@dlY1s;Lq02D z6Nsj|dt_4`KkNYYFN0UQPI32uDW5kA)B#&U_s^2dE@~vs59cA~Kbb_X4$l_s2p*4`(DKReyW6B*{_rgcPtR(ixfypIE{{nAOBBA)e??%;bc}d#eHB>*k6=Z+D(apXn zbfDh&@1QOXECzZ@iImbqQ-d2ifo z=gj_Onw<(fm*`0)6X~dR?R-@Cu_tl5BfrfKq|ojIOmU|I!ra>Hd33&WvDP=h4Yt>i z0+!HjCd2t!Qyp)k#ti}pVE~48@cW!aC8_e++ZeZ!foH-j$vLz%6n=FyBQQCx?R&?Wgn54eJ=~!rym@e4CbJzO*K3=V=t6F}IR$NT29{ zhL<`U-JDf2SYlI!oNgM`}2=Py-P_LE(VMAr$t@< zz|hJXmk;+q5p5VS>shKT-Ml_wv)pZrL3=TuStjV1K?I!pu{U zf%>V@uo;5xI5E<_i6Yu26937@w~ZoR8WcB~i%dxEz;`c?arf1N^$^CWO_d74k=Fyb zj~8!vHdDEvm3X|BcBN?Iu-Lq))5^lY)Awn{-SNC2$gKxQzRLiYx#V3#M{34aS@?>` zkD@rO`IpHR93~`-qRQRXbfy)ZdNFiS4T`(E$yrbd6`s{f{-^kAUvAwV=?mP;#76-- zrq_4xR`qe)AB1&09N}O+kI@3R0yO%19sOPbnsc@kb-s)z(fa{5K43k1&#a3&ReJoM z(i{hADHB!6qfB&j%%x*)88bS==(2!DEQWu&jl$T9B*{Jt0Y^sx4|L^6xRFXoP3;HD zfB$~nz-Wn4fl1fL?;&fvKzu_Z3&2MIGs26*C|B>2@LB0S zX-QP^>aswlnmJgy6q`qWD-I$Q)s9gmO}N=$Uw`lWEBHw=OWeXTWp~!a0;Ks+#ZH$p zCNKTF7I3N_vsslCGkhgJOfpaWJ+&L>pw78?Bs)RgI(2_zGNi6K9z#U!S z15&u-pdLxxOB&bMD|G>*u_{vwa=O>si)KnctXd$Gnu65LY6JphhezAAmT9udCtc@+ zRFEP}MFE3kE8uZm;=VVr^T_x{>#9+N5ers6M+@%m@lddo~(0a_j^0) zPuRGIO98A+Q@);fkM_gGb0}m%1%S~(UTN*s4q0*GNuPC1`_Zt8kC2yn4WXD#Qy{)% zW!|HJ>sv2ka9l4=fcN;};FfN+V`%nbG+-DE5kGPUmTRNt*~t{Dy1W|fxHQ}Wy&_9@ zFMgDv>7g+bbsney6fw|bphXB$ArNkD5|l4SG@}9V6m|yd>e9F?S!lN~L)&hmy}g&V zOnDijDqZ4^ocHI*%1@O(pgPY~M%pVPl8eCa6$u+-tqFw8iLiL2!NQW=zanhE3Tg>Y zj6FOaUKy2v>(%4|i_5B$7t=^(6QPC}h)<%+N4H;cMCaAWbh7utGtb~cA}e(jAtk93 z_Of#CpEs9-i>A0(WUQ&h^$P1v`5ev$PyGfug>Q!fLFwC?Qw<$xoB4C#TRaXnB`C4@ zrY~Ja4JRH9?Pm$#;S*xL0m&WbK$#~tt^NnZP&FeF%i>8soZnihEe}_Hm;BnVi^t+y z{|aJ$RIuzmrf&7&o*CoMN+K%0X5#S@U~AvvAgl2?60dV!B!c`yx(6$OWeMZlu`JX1 zX6=1=w_FXB!MxA*UR`{@!-8fS4$A-foTG6sPt5d#VEpFS+ejKyf|aswD5MYqF#Y7+ z^Mzv)8iq=U0|T{;y1lFUseM3NUVrEh9i8y+wI@Fx7rOh7%Rd%MINB)~HGfOWD1UfO zf{$fWPv7bTVWm%TZwb<_kW@UOsKQXc>yrxh%3=;NvxE%rxP4D@ub|9r~5*8^#$S|v#fdX>aJ9I12G z2ldrF2St9*6SNa)cu@IRjywMPkLd({Q};H=JoENEQ_bqdjhnD_rT)R9f_*?s zGwm@$xEEL%?(D%Hg)8S*wxfLY0;{Txv?OD@E`nFPTk9xAdc#)0#@z>tRPSE-mYI&s zIycx~mqIcVhYjbAH-%NxnGf&r=)4$gxK4cDpQVyKXC>h6x{=mh&c9U+2(EG(c&ela4xeNH2o&JPM1QKdp^Pg%&^J2WNK-t zw<)%0`+Jw-%L8FZXI8rfWp8O#xV%Uig`|se_K@U>`DWitNbGSH#^) zAIP30$*(6Du?NI8HMTxAa8H#uSwnccxhM7q6!87W9+MdSJk*#DaOfnjAa+bAtZ@o* zcFe0AHdVNN*?-zR3$^khU;j25&T$ZzYDp@A>L)E*zg_xWYCZhRU_%%|Bql;2+$nG1 z;N!BB%6oD|kJw5T39ku7GrJ0$-omxoP;OYTrob(5{YN+GLH$1`CxCEi1qcRf zZ9D|fyv2F-^^bDPjTG)#Up$KGD;^U=Fj*d}3hfzm(VA8E1w=Zp^=AaTubJK)V?_l- zp1d3r)9IsKY5Zsf7RWijIALP}Hho@XDXue{#ONW{8`taQYUU)uTE=qx-euJT^%d70 zo@lB~W_@H(DpbBmFkDxC#n z!5Lwm8R@vn%S=;;(m)}@EQb(a835_kK%FwB=#P+;%2yVbKT3GK9*xrbmL7rZLco029)VUfzK_DOmjA5Pt4@BoJ4 z#8WAaAL%ElOUv-LpIw`c1Pq@88tWM1;7Kilgek}8ZpBFp>b=X^AWAh<$+XOy_S{w zDPfy-z`hzi(nzsKFlT%ahKxl*>ZV}OTu(+C7aHfhFiq@KY9Uz|v zG}Z}l@BEXAj<`Y<;tFTrtH1uk=QE^}_T^bbSDUv*L_lQhlNY=YauP+Lzay#BZBJ9} zW%`JbE=Hn%Q$a;wVXKX9gVDaTb4fqp#Zb^?eoAdrIJR7c5kaBypA(QdOlp7(Vatw^ zFYgewDE;u8WHX8DT@Ud;T2}dKAq2Zu*j8SId}T6@1+=89()WE1-)*M9ORh`hF5Xe~ z0HxLNrK<><+K)IXY5t~^=|)=3%C|73_2+yIF_D=|_)wZ@^WhT<#4wG#whfyC!SCI< zuP5-msv%egyjzq{{KZ6#Zqc$FEjUd>?wV)%yPfTgtOLX45+I!KTH48?nFXRz8{n3C zv4*xe(6zTK+d2bgvnA-ajU)C#gUP)c}snU48O87u=H}@|?qO-xgPl>Ob0JZ=jsY3kuqW*z(w1f}Tr~93!y<9+V^CS!ed0SNwa7a*tPF(K9hM^3Wmq(~!I$r+Y6%(hkzO7Vse&4wJU@yx$^1bCwfY7MU7{&kM9W7S z>~eoDp1&m}c}e76*e9rqK9j;m`+%#yM9qvp!CDwc;imZ`>zZ|lRN)vfG_x6b?GTv^ zM1_(s@vxL=od(sM0l^*xYU?Yw1vRJJU0#N0{^3vJ*T{yqnRotvx5?BCgvk6H_Zt)6 zv^W|e2wWTmYsPaI=GBuCS08l6^F0N?<2AI9{hsq^WAqn!o?mTrK5GH2_*upccK-Ch zR@ySv=e)|LY1Z~}*Q~Ko9Y|nJ0eun&ox;(mQ8emsYkpqh@*o;Q?bj=i{){0=2`|Yp z%lnaf#$6bwt#4zlH4$*{&aGy~s%%S)^@#6069e^pw6`#5i=CbT;rb#Yjn*PiV^yHV z^gSX=Ni8@m_k4r^5kyIqi{Ua+&K@d9fP_W~z)#CjyJ`KtqvzUf}jIoZ>ON8KvfaOmnGH5!F*&Ow)HtKm3e(U{tP&EVch14@3 zp596YeI@3pp~gV!fT*w<9|IeV4f-vO#)a&IM>3|aDp6j^BYsU{60OAmM{dmDL6;*T zJLIQsGHjN;?#%;w2w?rEX4R}_D}%h2L+tkX{jrhtLGcN)&tG0X5@TZ7kurJ}-?PQ< z!g3nlsz4c{U%ceGuR7rqotV<3WkTt|OY1vUI_c8M`)`?~L%Q2+aV8=_6 z9YG#Hpw{;=^jpNg_27B^cdZQhAR2MDztz8XiV&XX!QE6#*rS{X(MV+Co|B?~tO+g(2Atgdou@O^im~k})J1A|Pogz@GDjv~P^ScMN z|Es1!W!x3`-gncx_fVapOH(=kwyd#*ncgSeMtHZ<9!F(`>#EG#PJ!^@hx2v=#nPCWP0wujbAZwvgwJJu8b#WbkAHx0d5;B+vJG0GaY{2iz%cJQt0$^bFX67#~0;Ps{WgHnrWtp0u^tSK*6DF zqc}YiRce8ZYz#Dle5+HF`9@DfMlM%QU{3$c3%Fn>Y>LTrVkFwVh|RXtfZmpDiMrZ) zWiY3WIzdqbAOJVBmA+qH7xrIXidMK2Qgod*j=6`SXp#qvD1NIBqmYygtn&kFsh#e~ zuJBbI7l;cJ=5qRC2_v7(7@tA;01rHCe`} za>1l?X64JA8K|ZLdHIEWa`za_V%V7lmCZz5qZip#wD@41=@1C^l-s& zN=^KC0aui+7j!M*-HXcMoj_eAtG3LuyaySNx2pPrf~q)>|5CW|zv7)5Q=2w@fKg;O zlS9&njlRZwQ&j}B@>~K3{=bs!3Ks78O?#Yu7ZYF5h1?CY8)4iSRgoh*h`y0c6|&3lRwk!({1q=OF*z#RK## z)qQw?&D^W0&?Ku6KDBr6AWWyam3{fvx`+}ZBuL(@{rZH|^O;-==T7Gxqu>m&KmJ<= zeOt^>F4JiHk_&ur{#ZS|p7|7xp?v0E;O{GvgW6oDR;_a~=ysMFu*7fhKA9gYIg9g^R*#8Lt@ zd`u}?GoZ-Ye1UMGm{HT!rPQ-yp|7l|xn;&CIQHD-^1p+k^KZL(pPpx1SJsGO$V{wD zttN?ivORbgs`7SEve$&~U@9RnDWOBpWdoS+fS6?^xJjVMhzl1l8uyJhhFXVoh|HjY z@eXt(dB2UW;bfx*d#sI4jI_d$t@usQ_#ze8`48fT)gr8oS}aceKh=|0tBnLBBsnkn z#=@6M9p-Tc8k$JT)#UsB_+uwazp9E*Um#BGwijm&L$9!0CUL~KLuz~mlpDT$5=DM` zqetBsR;YKiAsnX}Futp5$1BFXUyVaDO@%k!a&z2;)C|808j0}tTo;_#YD@0x9DAg; zy8F8!oEsBRxL{0wcU=qxwd?N?WUP=zubh4dUgV^FFd3<1W;WK&64uj?*QCzzYR@2> zy8{ilbd`U~r7j%r!RFhrxNEP~p7L$}`eZ%j%#)6NrJBnz686n&cqzPC0O*JwTqP_d zWPg$w{c(EKbSiFYe(QNDTG|woY!Tx#rZ|H8b_c19SaBsqY!QVS z80}fijNAAvrMld+>=jqH+va4#K0NoBnGH^Us>s8P`tbMFRwER)6$E6;TAK%&Ct840D$3A74>FB`t$n7%z zkxRa3$pe=iKm5U`hfX>?w^i8x%6nF#eqw&+!D5BBKDZlVv9d=0tDo)zYyyu{)4G-iU#rR_yP09S2uGGz~(8yqGoqFV^s5;-qOU7n|KAO!DN3k3m0E zI8OlngTJbs$wxG#O_|y*CyANa5B==-CVVawt60P#k|Cm_eA1b}OE2h`_1G=5 zsalIko4?;cos~(#DuCgm_IALPG*lX<<1pSX2HXnS*sQ&&ljW!r9#{{3wFruB4AIB0 zXljyKI2k`-xmr%qn|JP~@%Js_eJ(&z6Z4VgUN^v{c~)#UBPemqRvUkphUeO-gX(9K zryH3{WcLUiH;BqCv;R(B5{yS$F%B&FU_I+#+1>MPm)-b`$GpAmySF;oEtvIotPJBi zG2|)jQt}M0qM~#x^HlQv#v<(TKzwpr(CR%Wb#X+@tdbNmXt26;66^80nMBlQ??z8^SBBpKbJQf^$ffkm(;@whUao z>G8-Y)}e*vlFhJSj7Z!>FJ@QDM>yqo3+O4i>|3HbqPw2Ng6SRV(kUncQ_b(g)_GCMbtz< zPSL{Y9s%noPAz{J^zqehlEcaQJ|YBnZVWD64_|$Qp6!i}y;;f)i^xCTOFhy;h>3M% z=|)Y_fJdC)PT2tpcX^;~oK*HO&HzAOMtX+Sush)ccYtIz@XcBd`m)9qVo(*p$z1f-gj-bLJ6%KK`|{9 zG*8eoGX8KbkNfZ;r{?OVzILrOY`i%fEb*9B50f`fTT=b-$YEvv#&*V@;(d9aI5LfA z7ud3|;ohA|*2Itu_o+@ua-r?3`Bp>s4E#JAKGs6#BN@G!2$D14LZ8Vs)Wn8Sk2`?{|-UT%C${ojdfsi zN~pPevv^a>U@v&lj-dPe%?38HWZM|GUkmp&f)smw$|YA^Ij=^H(TUCx)X`Upt{)5V z7z*JdR zUu^Dh9iBv`st>ZQy~r5y2_}+`roG_w&dj0_9n?BT2*1aPEsM;I8h22g>3f{qTC>b> z;Kai`qVR^Tr5w~o_eG*8u5F41= zow^I$otP971l>M(tk3v~`Ik^_Jj0xyP^3;ub|qS0Q4L8V*DO~%Sy0Z{a@2K)puZ3K z+H#;(?7l@L=>D7mxw6Km+;ztH6mrJVea%Z2z?^~>jW^}W@dPP_EwHS|s>V6vIayhY zDc)ZW+#AYf3YA0ypRO18o#vvF6*ccyo$DA3TQzj+_z9*`>Bd{gHbHbPJ`VDu=*H-^TCnLt-XD3DFqAcAOM-> zAx$#(NiLHZgjfs=Bs4VP&CRkm?}-j-NKFC4;V}C)FrRPo!=(+0kH6EMDA>h3`=@qC z#ghIA%*;7ZGJ347op=Ddyjupdj%mDwLVku0yrNC3?dp!XCCt5?|JW@5+783kE7CKUT%Yi~po%Ct@tEUNA1 zTd8IxdRHVa7YOOXE$AKHiYcTX0in=X&vIGi{%xkwn`xxdh#2V_qvZ3h*Xvh(3O0BmnAEeY3rc^}P^h#f ze&OH7+>%dxqOtiC4`RX8sF8A;L&IRMIdquJ9<)g6aLEFoW8lJqu~_UnTKtxB0bEm3 z%40K!u5`9|e3~)wbxk2G#k9qewLqQ2n>?$|rgVyl5!bFu;gpSzOj7Q(E+Ttq-xED3^8}g#r_9i>+~@w0 z6;YPM`6cNq;9XZ97hdM$33}W%K{3)0v^Nd&ZNGa=RHAoAsq07z41_rEdaRXHZg+C5 zI}fTo(dxYb*-`v@dxak=tnBQiy-#C=fQIB0ygm2rYI_YQ#+e6MPgyqsN7vLx#kung zWc=gB-S{*LU$bppM$2k&aKcDn=F9fI5smdm8%s@@D$|YS1(}B2SX)AI9Pyq+($4Vm z$;Q$^Xkt*x>9az+bLt;k74oJo%h#F{9#9ezGylx#9sL5z1qZW1lzPP)7jABD%Yd#} z9#IQ`x3GgT@V6mAgF!DScpG%PJpkZbiB;q0137<%JhW-CA;}pP9mtmi(awkwvkQHy z(TY!U#cRq2jq`vF8H@|4jR@#4o*R=W%PSr(9{#i-Bk4bC6+wH$wsTM zePOJ@4{IF;qLDw}Ppr{_BUp04JD?RfOu(>RE4gZoOMdmRa&eU>zCVs~w>mrCt6+V9 zp!K4PYcSgLrcV!Y5vR?eYO92^Y~d<7srK~N7aCnqP=Hn&YrWvVpB03Ee&$G;hGUFS z`(rogWN0l!P3%nZ{uVh{8_F~UELA1U6+v2a&g-g<4XPL1W{Q20 z|KGxYFz2sm#c4%&WO{gLeowL(BQtX=r*zI_#RFbmQqbXEK4}XVE;D3VSX|WZZ&^a~ zV0fV2cWP3HAB;aO1i0dxm!XX3j`BF$lU1ej3BZJILX*LRW`Po@7IAs7kd1uww4JPH zR4}G$Z`lU3*~QL3zm&(W)yXho9pb31QHI4v;g@dL$5959?>h2M`Y z#KhqI3<=PbikUN%Hc>?)$f$KPxr}qsu=RUOswbTBv z%yeuNkY^jpQIq@T=Pb;}BJ=!TzV{>=OZ!i=ROu=}8oZwtBk$^TU7OChd0LjUYbYyf#f!xL~5%% z(WUO?t8a`X0~zw`wWytn1aB@cKgv7nl|1QYE?$;wazmlZq(d%i#u*wMHhcP{oUX?2 z3ru@cG&ug;OqmAS(p9CVC;{izmey7fFgLB#SQ*YTg)@Sa{+pUfw2Phib~|Im%=`Vg z!7pXsZe0e(f;X3n`WK}NV^N~lc&#F?TdqyWv19F_-r_f4W(+}8uQz4*E;LGyg!`t9 zP$vtKg(zL)`&9F%LNHZ?CAQQ@j5=g2V9}r7^rv~#?PvSTh99#MT>e|T1?D`U{xN|< zwbg5&AA)qq8J5-kDe08nl7%a1h(WR$im>zFEyz3XdTF8BsVws??t=1jz32)$m4K#n zcPeTtYT!=yA|#w0pP?gThUK))un>MfeG}87gi*E+_xMbD-n=h)(WdSz`3)=5HElY) z;Ei@d&V@Zf_qXVLym7|5!X3K2!@P%|DW&qL6MTLT;4V*4De#PG3fvrhKtV|vjJ8;0 z&RKrmlzBTA=18~|43+mn!_-}Y)duX1wdsHG{2h!QQBxb0s{V|8>G9DE~QhHMnXmDTDn_dX%<)- z1f)Z{-n00BKQKDXD9qUtPu$o2G}Ax9p|t&2^ongyhE4-+M`B-w>KOu{b@RII%3sAa z%4J5)_IWW`7Kkk-1aWn޴XoI$yI{GomG1U$uZYx`0D`M>9`T04NA?c`2)Hn91 z8aF4D8RP6)N231iu`TEClwjNPd_h9!@xS$@g<#>uI*W096}_Lfv9mz6=E4a1%Bh0x zs&ynMHT;%f_=x@UALl75$8oRB^~oJMW(drAfsLoz0s2(!K-;jb8aLTC^UK_q=)@_y zEJ2GxV+Y+vp_H4~8vpTX^sG)4>sA#HNv>z?JnHXZp?t44lp#EHu{qX}4>kSGoNRYB zt%KWdF#}cRjZ%a;t=zshbFkVMJ+C_Q85+#rU_4Bvu5nK)^(J-<94UgDwUPqJ(;>8TsDmJnHmR_k=JZ?Ro4BF}d-AhLnC}+ zvSb*iLhgxU)3uwe>(TX%zn!eNzJ892U(4YFn~n+XfCXg)1IGUd%Eys`8`@30;=Uq zWm57>r3m5|DZ8}=g1ty%K*ga;5XjAnYw`gRW)aHKrkWk0^pU%O>H6=INUz&=z zu09hF^99M0P;O5&qubg*JL@L{v+zlvt+i^L_$dH%+!8?fHr3=)RWh|&98A}au|#3) z7tGUo_2B#W@A~h3Uxke%g}qL>%jdXqWh|CsLlXKWVU+tT+MKpjrBpn4&Ou^uOZ`Od z&4;DL3(^M6kj~m28M0#{ctnne0doL-Wr#tJTQDX(g5v%8X~Z6*QR+A~CVh-<9iP$S zx?!gwvlbg##WN+pgXdl3TnfrM%ogDua&N6Eyl4@v46XiWen#>`xij-e8O8~7H7ON69;||?}Ys$`l>G@tXBogqj zWBnFH)-K>WTL!t%_YYWj?vzAT;xXOHSK?}0fwqqqZhiDqO$J~vX!tv2;~lzm0jWq}^5s_CO~35JX{I zNUDn_UM*W<`lj`~>YxT%6o`fKz(AlyvCh#io)Bo4TO6wugVEs;rLsu*ea$;MUX!3X zHZ<%~Y{RUZn@gv^8zY^LZ;3F;4VZ;Q06x$e*Jgu&mP|fsTckc{;1mDjAO7N=f9!UA zPAhSrZd~OVH%_Fjv`JAq#t{$uHS zbVJT#6obr2v?PF(8h%R+Q_Ynh=?cn}iL-wicRJHf@VqS z8B*(L_@%@=nTL4{3s0YiU_rL?*2VaYtcNKu(`NaleN}csWg~`RD?>N%bB-&h^~)(9xly7Ur&+ zh$A(YF#6O=9nixUXEk7m$Z@$cin$KF_`_Bd7{c1#gC+~P2ejBW2|}GSg>t~Z{Jmxr{#>P zz@pnazEPb?nMwTIV*a;*$f1>O;oZEW^TF8g2)JS)O{#%8u`4)swj2!9(-DvLB2z4CZ+-NfB-Y?`UA-~Te;p`4+xX$M9TRW?vU86dr@ z13f_a3D9+(C8UW;=$Et)X25p^v45$UXy80kj{|Rxa!PK$Z>lHWUkQ9qCIT3j zhU$nHUj%NkevI<*np$D;oWxP`bzohPO_yRytqC-UP%v)wp#BjU zd@oP>JR=Sz35y}@b=1dqrxz=4%k@$zrl`htex}4pRUBXL@#b)BpDfMR@c-TJKi9n) zaOvH`3+NdM$mSDX2_pmr-OU-^3OVd}Fjw)1A_v`8vQx`>Ptx{oizS>+>OG>JGWV8C`4&<_^>>52NZ+LDQXA-R+ zEifxRLze3zs>!bK1C=3#*-b2$PK2UD zG;uWPhcF0}Kpq=Z{;z=5zW3`V9Jf+#h-q{Gz+OA4()IOBGcoMj%d|8VdzTn3Dr?%F zYGwascmKfmXTpWhvXLXV5023QoVAK2SG!0tG526deztvK_WB`-rFbaCS$5!!PZKC9 zYjGm+Ou<7rqZYq_jrUp?ygpixAe0e?j{x-?%m4w(i-GPS|ADcGJF1Bc1F4Oe8s_lJw~6y#7G++~7*G zJTIjo&XDe^_0aEz>zGXKf-y&dYsq)cMKaCe_f~j%yO)Nd-G8@d^ZF_QM;FhkEC=mR zi}q3{3vUieh*xOIi_3$8tGMfh^zz$+;YC_yMh>@FM@}YNlO;O~N*dtbu{C{3rY;K{ z>q(}Ym|fVPf4b&Wbl&_u3QkUAj&vY>z4~ z+3XXkxtj6PYU!Eiy4%Avk;KWbjtfDkGVszj`uJ>HhwzRek)F$#df(GshdFlJ6#>|( zo=Nt*`tQhQSHlyn9-|h1(6(#@W{(cS)EpN&HGQ`8S?$so+j{1wG`E!Hha;yshuAl% zSD2UVXx0=R_?jvS<(o;Ug8Eb~B1 zJy`GBfkHWp`si^^dBgA5|6s3Y<^(qFXn--eDJmMqWmC_KU^m?{O0Ph zG`eo|ARWmH>j_Hw-iqH54R4HOnn}LH1M>z$A4iTNn!z%(v9G8!gRTPZ;G3XAO`@8L>sk2jCY4%nG^7d>7>WYZ{3LR2y6-65t) zCG8YyAIa)PtshIHPX7w(WBEMME~Dw>R&Hu6{(5Qkwhx8{Q9-dBaK;th6eEd(`!2DR z-9kY%1flOcuDDcSO#FzWqwZi?nB#&Sodf-k!Y{;T_4&cUz!SH5s@wA6`A}3r{|KMp zC?`_LT-~EE!#7B4PD6Xru5F#=a774wV__(|L&7L)m3mquU^IUA3$pbr-&tT19SLeY z?c*vxSWg6H%`?5!G0-6xwOMwG$surpi*%$6<0i>2?_yXh{$oU@srGb;4EYxqDWW+B z?~x|4@<)k7PyHH_E_HBvbADkTGkefj@ejk28sgQlN0$>D5^A>_3U|XzrPz)$UUnX= zA21qHq*}+KJzQw8_D7>A(OAqE#PbWWjKvcO{_6t~d4VbShh}QVo@*fZU6#R0!-oH= z$yBp7aWt8!nPU*Ft@?ej;?J)$L;R;(%Hei1=YN7EObqZ&5bjL##9Mo#9zh zvxO6>IzI?hEP1IE4Yldcyf;NWyDAXrAD_?m7SI*_ou2-6%~{FN& zoj~@$kMvJNP(9U-+Vte#-J~!;3S26pUs;=#4Pfqmgx)G&2k(0Wg$fqGK$d{=^QbX5 zcNSCg;(@sz?}01^DVN5R)2qA!xQ;#t4o8Tx2;uNoZQhdkjW6H;7Qry+5DsZ3$oltA zn;m2}RX(2!M|>2JnspPJVRA8AjJ>pOG5!QrJtOAHLd=;oZDF~0iiAS6zY9ZpmU({_ z*0vrn^vJJ0FU6$e&>n@2zZR*~Ds{68%8&isHCeti!0$M2vyD=_wOehqT1K@89hsT*H&E zFFU)A=W+9Xp0UO@dhXbA;Mfq!A-dPIePMcnn)lD&DY&ip>j?q?vJr`NfI^39^Iu(E zo!0;e{#4GfgKTL8-0n3U#bs@sAnkq2Ku8}A)Cf}$L~2YvPTm|Yi5;me2;s5r`tbFN zjI~9VC3m~zzvxjp2I+43b(-M&B$_>5n9DIZB(kVk@TnT8pQx2sLl`yP+1ry3f)>1O zrDe!osv0k((etB4eow{|o1`83{Cs7P^)k((bFpQSg@o{g%)LRhXS-Gp;xk^3X@v40 z^rX#iH?8o&@WnO@^PNGv6=9}!*T>W#`x#_U^4E%s&zOez{se7=4Zb`}T_@9{&tLV+ z+sM=)5VV~cR9d)k&PhUh2-qp~l5cg0$65me#c@(_i-0Tx9js6(-Lt_ZXWEa0wp}D@ zBJGeS5sFyDH%Jp>y8Fi3<6UhI|9%+iwn;v#g$f5-i?TE= z0`wwM6Xx|I9t}92l{+WF7^>e}94v#LaHGe^W9YDA(wrHmW5HA0#%`c#FnZIRVofSJ zr=(LZY6YXUJ0GH&-d4K=!TpZ?&gnwn!pyajqT)<%S9iCr_JbRO%!u+UO-2e2jOwZ%gJs++Ph)?vg(hQVKbQysiM zS(C`3D@@tuSBZwkhL&BPc#y?0iI|*HWJwq9i0p0Gz>PYQ?LsSk>4PqkN!BoBVRDVp zq+4Xf?p8DX*b|JS&`0!awyhd?dwYk|(7p1T8L=AbyPR#^5oIsF`-z2E)t5vsox77w zD}Vxy8Kqz+B>c+Hq=(?!4yBN=?J6D{qE>s!LUqZjCz_lt(-i6Wp)XE2j7DU8Ar4eho)8~={Ja0n@hC9S`^+4tKpXu7ozO+HW4%M@55ut)}Mx~TmJ*FKYT zp~TghQDoo{T7Q?e9%c|n#7xGdrKJHi3C}+0AL_pd@QgxgOHS-&Zg`BEDCZhHs@BI~ ziEqnF)zR)sG z6&tcw7{v|Kr4g`0RKR2;1j8DV`O^%hs)?TAvGdGEfvz5TR9 z@Jh^@iq8xKIRZfZGzNNG9pMM&Z2<9A{Bnx6QN;=hvXV~gIb|yJ_@8_+EC@zzJ9OBwN@z*6ylidd*pJl^p8Gc2N(Qw!S7bV%ULgU1H= zQvYv#M?QnhD&)WXJ4@yw}0K z(62>SSrz=cDCo=Dn<~ewqg#sha6GoVh9WhkJEL+Ax&^nZ{^kM@6EPJojgAG^yJhu| z)Wk2~LdVDdL%yBfK-pPLack%`7Y^QuA;?UfBq;YG=CXa(sG7ic=Bw^DX9N~H2d$g7 zXmnqtCc{_JgmIz>`|T_SYOx~i#?_f9vfZs%nyfbkt-O(nvbElP``4#!Ryn8-0}1?Y zMEA#Dw*5?z(b76j_R6#D>+c7?ulNIc{_#k!gv}a&CfdEHS+6|s%j-}^9)t)i#YEi6 zDsohxEtNK7oG`_-&uZwOLc^<{nA{*Q2$K{0bu~>gL+$J`>l82Mm|oag{DF9_j? zzT5h+?m%%dMj&WR%}rStvM+y~l#vk)YA{QHX{%OaLFU|9jh>UbRy)osKJh@`-tZ4X zp{DK@h2R(vt)|6h*^j6OLqdZKA>b*`?KJCu{3=KxI>_h9V9EUT12cl-a}jbSwH!i~ zR20AK8jf@eFWO{B32x;@s}KiRrsx@bwT`)Z1Y`1ic z@gJ^t2if{js&&1{P(*o+QMftreK_&6ja>F#hTQVu*FxOJgkj;Rbc2UbC#R1h!E?&< z$|*rvcBC#mtprUc>AtYRj4Xy<1MantAGdWywvD|+H}7r0d-hs-lE}u!Mv>)Z8vqb( zCs%A#vc$-yr>CuF>ne0_Pk?dF({Oz=v%IX4e~MgcO3|XAn0m#vGzN(V(ff92felGv z|Jg)+z+ssB(!(tP+Q6xaVr7QnpS5|u9bLWI(OWF34I+7o!QFJf?l<{j%ca56s;(8P zRnE=>M|Ym-iG!`V&aN5YF<`4OQPE4^STlMMMxxdRirfyhphjZ51^-=3GH8pu-+Uo= zmn`u_pI$gj_Gq`a)kS{B^Zs!HbA8=So$wO*38ZewSY=+1J;#wQl0WOov5f znO`rjy}SaoyZ3H}aPnIKs%hK!KPi1eM_f&2N@+(0FFH;_&rV)BFDy zqus}jGjnxYupy}{V~VFCyqGaG>>f=Jl|ZJTbmK%)QqX*};c0p+r!JbrZBtcxF+t{O zk6LXEfm1Jcgz;V;s8}<%ASj8m-$?#r_~a8bfvqfam(CpJG)@E`S`7WQ=?L=RY8ns} z>nuGm(}$vHXvnO?N_|Uy++3+lo{PWa5{B!>IBR79xMIL8ocdVR;hT@`-9sa*@*VeP zy8h#vklT*xTQW68ZN^I`G(tTA&FG57#PDRU82j~E!G`2Be2?W2Om5xYkb$3Vq$AZf z+`#pDWPjZ;_Q3(-u@+E9}3JV2ypv2%lukDvHU<=b64grgRv zT5eR=J!ps@ivqd>ghG%Ra6MwiRfiRn=D(IcCc|gDJQe;$vzxVwow?BSfr!N6Mn>|P zRZ=WXGhHJkxRw>;Xd{1dHxd9>@@`+OIcLB*%c%!VTcxJM-l7O(NxZ20;IjSII% z$BMMvsR%#~h=ZM7&d4Y`t5e8h^S>q#Or8O|m}sEr!UVQ1{P<$^9Xp@~FBI*dg}ASo zLiUep^uQ;vP3(tBvB({Hg<@(6oNS;H@Le<*OKbn`Q-l5(tz?z?BztjviB96SGc)pT z9>KU7v()*1CDRSYw=9kucy)k{bdOmrN>N8Xu}o)+nKgYqGD)~sAe1e_oqa7rMKC;{ zgPF|vu>i87va?C5rpc!DjMAho0>VKRvnlZ(m4X4p_`m=?_bP$ z4_)tRPc?tAm#@=Fdb)84s%ObPdIAEPwZD|LUAE)hLIOPZ0tM-ohGZSIv-alhF0t%v zM+LHEJ4{SWA>E!*XiN-2W8m+fDLNHq;mvcNFO-!l{$W=*>Uf<w_7^7BbqLgw*9Z zoLLbem~k`g(#7e@LV=Ai_B>xN@D8&t#47s!98)Wr*>}+J?*Z8?n0`S5x&WXcM6(ax z%?Wt3<7|{jWGfk?}vb7%s4#xZZiertvu`S^-=nXQD)DMm%F3+{4d4Z$(4p3kolo&wBmyUq9YnMQ@Hk9rTf+_2Ow0u%Pr)pr29e2>f%1zy)G+ z>f}W8#QOf*)7>4(k^F2X{4iw8c?cY~*Ev=G0vXB+VLz4(Jdw}Pnt-n_@-Nr=y89;*q4 z0f~bB`tGZ0#zoD6OJw6Fh8r?!h>&^5%`)yGT9ip1&C90`^qKsBa>A1(oUESx!lWH{ z=gc%{o%bl4AlJ}nb8M{d=# zpn?Dy6@0Izg8Mp0qP*8)RdXoCmot)<2==}QBun3}?3n^c99{|%pZz=?@L_NI@iHnJ zn7Dx0+`zRYZDcY_tR6$(P)7j5)))ts4GnUfV=oOL^?0EWo;+?h?ggS?Xi#P0v$?{% zqkk`I*hcVFu8}V>K9lfE456D%*!PH#!M3jpgC`UT%}Eqz&0w_9 zXPW-Q-!SSVt*dQ@mD#Epc7s&{XztVCXl~amTX)v_znY6bIf7{%K?Oz9x4~kz6iF^1 zLqc&+PGsjU6HJyzO3qz?Fk=~nMV%^y9{W>uF8PvsK$*}9gds7&h4AhF0wtM*TKZ=h zOG%c+jxoYaob1!IiwfR)F{^0jihjRSt)<|z+ySVmx0LE#EE(9z&d~!mnDOEn%6PRB zB&R9%Y{YDvVRBV0#l9=h=?(KJUPyz;sY=9-tZfAor_8^rtBV03Zx?^vGC(8FOaAUx8Rt>S)(9&RqT!N{S&VWp&--muIy0hVEi!D8 z8u^2W_fqw5OSk+n#{#xi)A|FObI6JQdGPcN(Ue4|Zv2ynm`@O65?a6Qk3l(O+bBO4 zM}eYoGSwiTlB7e)(UBVr67xk`v?29f+dE(&NDz?vbwLn*_S2P^z(%wN)b^$dasLT` zQDl{ps?%gNK30XrfHYLt>unsb^2e%yOaZ!QIoO>3m^x-cwsH@v;|y-N%Qtl0;vzA} z)=;LSC9z2~9uB933DitCD?0RI=-=LW=!^94DPxmvKA*QGJ(5QD3Q&=u4Mzi**+)@r z2PXuh7s=HmkN+b8T9a&&p9^(La&l-`_o%}_@N6>i<286A-`O+W{bFdSZ#UPAc~R-b z2k}(zQ@oMa5oef}-()nMSBAI}#$$_RqyBBCdq!=NXG@|n1`mW~!NgqWgxl=h?0-FR zEA=tR&z9MvQRxcICGmuksExC_pO_zwY){85%neipqYoash=%Q(Fs=g3w0Nu!R8ON1 zR(e-JMYI>JP&fb;b+6Y#sU4R(!jW9s0hnKAhWg9u-mL(Ohn(4%YY}BH|9^Wr!#Pc&DlH1Ap-RTMwv7*ut3oI^<_>R=l~NT@dao+nDN41TE9k3>%AB(G_ANXXG{;7 z1`V$_{#Ty_CfM)pm(7P-Y_2ENT=J>t`6$tX2eT)Krn8?Ua47#i8n%rY)7m!3f4!RN zo0+=OV#z!x8lQpC$i^Y@xcBtHW^8ufPD|W%SNHP7kz7xuzDK^o?iEn&i;p|j|0dsh z{@~$5r)Wt1J>X23{i?;e5-bX(n0uiQS4}M;WIUWb^Dd0WL4lI1Q9JZ<3vTj~t7$HT zd(YXCx)aA55L?_{2A^^5hzz5k-2PRTm;DdM&C#ZTnpxBADUooPe11V~a`aH`c4V$# z$U!E)2-G5WMdhxXU4tK(qy(KOPxo<)|i6!w|@45(8EE&+^#BJr({8&#s$eGC6)u7~;C|N^K z60c>sXI=`(aXUO#x0N@HZVpznLMH$zaI`>l^aF%8A9Q-3y?og_`en_Z=YJnbhRF);cFmd{Nvp;} zh4ZM@JzaX(8#^fW@E-T)^K7*z7CflrQuhXYDf=>sNGVYJ(dR_Dlz{CUc9hur^td<{ zcZY8r$`%Dx*L!maGL5J_v$Nv? zO+GX{nZ}P^@a~z(%f{tP=bZFNbJ^9?G}eBmOALG)-X{mgeM3>jYAi7LNosO!EThaf zTo*4IkR%##5tpMh&Q8!hHfFQs_z`aHIN6q_J~wk?hoxt-1Sh*6^%yetsLPBajr|L+ zxob&bAr9+6)W6DXTd)>mg@fui{hysF)b0QD&!x&&-yC`5P7g>Kd)Hn$nEct2aj9pV z&NKS_rmx(zjjvaPQBf8m?*poI#XHgt4=+!4wstTN4-eTnGhd*q#up2oOugO%YYUgCP8K2B}G1Q{3e5_)0# zF`0XpGmkVF*&d9Dz~LyKid0;<`Po*GUe=K~fAepF=pU~1w#*7VAxpuD#8_j9!zX( z*huulkvS#2I~kN1{uQ{;Fg%|q2z$`kFQxn?GPuneae9dQ1yagkEKgc5_>3AwKGj;-+8T03Qk`7dGn2&u!H1dCz>22T90+AV{haLL zbZ?LDpHrC!UN^;i<8On!A)7SJsV#q#OvLjBaLNfx4qmhs>kqDI*JbLiw6`s?&2mE5 zd^5^&p48^gRx}+WiM;c8D%p>l*-yu*pQPwO3C?D$SNC+n8ls@P^RkTWL56nKz6NiM zc%@Wp1vbnotw)%|qhn{({zU;KU%E@J0b)dFof2+n^Ziv2l`twzfOCIM>ZDAWU9eOw zstaAPdX>}B!dfOxaBS<@=~bRyxi|kppXZ9kji zTTSqBfZ7_ne-{eCCn}IbOCT9BJ6V39yG6Fq#J4^FrO!1O$%cQbf|~9bePJds7LWer zJ!@ZYXRuCVB~UX%wBgA&?8gHNdZ0FzzJu!CjAHGTDmoYTDead5f2roe)BoOAKgsvE zLct$A95yjKvXt5NH?tbnEdW-vP-oY})&9PLf$GaYVwKGu$`Clx#Q_9nx=JLQa{o3K zku$mHt{yXnrC5d%`AhPKtwS<%cU>2Nd3l(e7|QfWFjL1sUu@qa@J6wA3mfslMnivy zmQEn>b|f>W1|#*9p?pL1=t|MO(793ziBhn#sh|X0jo36kSKD}S90L{zusUFdIbk~( ze^3sX7738pd4sktA9C=)^)pM#@LMoGu6H*OsQh>V=SM)vivoIM3~&E#m=67KbKsOP zJnrD95RoNEOh$Z{N;z5w{5FfbfSjF3g2ep|SMK3+m8#ZEREcEUop|g9Lo*na8H|qL z$>tkaV!VPw(x-L#DYUBMtMsGEj*bD>N8cWM$_-^pRsaPM0(5{<{Z?XydEM4jIE`Co zVV2DLPuo2p+WYsfJ4jJU0jT)?e|?rsbVwur8j7eQ&g!>1xu2EK zbpon%`28W}e}s=}L|gx$d>*3$d~HzbN16XfKJREa8oT9Rr>Y~KOk;$<8aOXre6_8e zQSN1PvVBYT(s8!1qyHV~S_^g9_5y?a`=1P4BWuXdLHqUq3>h%n9RGeIB91^le_%p6 z1=KATxBeTL59C>n$&|^+fPD0b$1S+fDnE&tl9Ho#K@T37q@*U%Qtu{NH8xv7DS|%c zH=`qhmFpOXP&$n<#d0vilzg$}9eBOCdHoy4e#TOx(utuo`=YzJ%;U=aCj%xkyMS!f zeE730lmkrE*{h~@L?*KdN&O%`N6G>u0DN2rRzw?JDGb&!DxtYz>FZZ9Fz)}i&Q(%c**fkHcl;0Ye!?tXpYHPlfi5fSw8Y|8$ zhhkO55KbknkLMm4Zx*G#a8Y9guuWML_3PRqLsw*21nOv_S{4;WvC7qe%r~i>^X!ad z%Z{LNtV7i{a0xGQR(bYpKRb7KaoABWtqIJ4s@p1U1o^ozLf><01TVbzwGR^hjrjbP zg{aRW#GyWVcqLXnYR?`+tH4W+)kJ~05uPd&?YIPdH#GWTnk+u$FP6Xmtk5!%bLhk} z`J-vFqvw?qTIPG~YG^O_-#b7Cu7Jbc4SdFJjh=o%YAJa=w{=v~pY)S6@=qfvQr%y- z$6Ii7;Qv>vEV9|sP*91P_%C>qAAYL(sd>P<$$;U>z~IZO@Z^;HsHRJ$`mvaap*d)J z@zVkAtjYG7Ws)0wCL3-t-^K=f_~vCbXoeHRvg|UO$!YMb`7|4>UXrN^kzT2lpHqrB zGn6(RKIdpiwMLG0Z~+E_D|yjCc?)_JAcUepqG(`by*ZeScMo$CxThR|?0FfOL0tEq z{NyNG1pc5Ey>|gz2>I4fD{T7x@bj>g_mTdp!O=Z965 ze2*iml44JGKr29=H6k*yePblwW!`6f33S$b!OFzWH~S)Xh328Q6L?^*VNbbfr!#N^ z(QozwR9-%~kLCr&?{P*Hr2Os($9JHh+hJb-%%txn5*G#%+>^S0Cysa=bfRG_zf5{_ zyP??2u_zIrxdPuTGq|bcJIZW@1_`|f<~B{&2*$4|qI>FY^Nfya8BdH#+Aeru<3dZE zTE)(%;$EKpf?^*YH91iJcQL1-p}>GiEG1x1((j=banNJQ?G z6IEzfMD#I0I!py_a#5thG=tXf_WHDh*PQLHge0$?)QbBo3j+YCODWd!I{I*3t+Z)7 zCWeFQ_vIZf$-47jRfA1CgR~UU@ioYhevZfvi#IsN4hO5CYT;=3T;xFzW->s+tFxL1 zS@bzy*lwZ1E-iyu=ALaNbmYX=HVuV5UyyLVE2Vr66Bxr2YFyR3QG{1~9Df+GwUM>R zwylZK)r4g;+a_<@TU`!y^!tegq^ma?ORs~b0AY$+c?Huxc!jE$qeIJuhd=e)r>2f7 zEPTWPU0zusrl7DYV(hdbx9t-uBZX8L0qbEqQBCauK=n9gl9YqVG&5>qGwIHdS>UQ_ z#lf$;)93c1Gmf40wZh#=dYdoOi($qw0G;)7U^-1*LuU3CNnRn=5#0pd% zyLEP25|U<;uqj*wM`nSe%|c6aO(1e9N*q=JoNBLx`gaEbw{61jUVn5g$!!^l%MF<7-}NZuc{Nk<3e>RhHl9D&D`s4oK_~W*PEsBxAjnTM z%7v~ttn{=cFv=jHGLk2pk?f1wf$Knj7UH{GZ?y&TnKq0J9zG@0|Fq}NL`6j(J$_t) z3yq@_M{bQ|qyppO6D-l%1A_VL71AsKm_RT(+p4d;Bfa%DVEYF zhFWMh($G1aocOcdPtt3bbmu2FMYj&VTNTazGn-CFCGUw^;-QjC`qz#v9*ixQi0u55 z0O5snd*321XCdUD`T`lJ28pp6ipDo#rI6>s2XW^OoGe#?y)`@TM@yG)5i1_`X?u=R zzP}PO4J^$J^1tP0qHN7w2KLtatCTyo)J&Z~p+XBs*d##VN`^G>+2XBCqCz}RNK1=+ z7;uIPOkzp0{SN!ZF0<_Co2CJFc+whpdow8*-rok=qqNG{ z?6;5^V#K?UNLlN9%KFd5J~)%>+1q{FlQSg#VC}u1G_%~@c6DCroMdQc*nmf#OGr@^;3Q4ozp0w3;ATg8Ts|e*y98Ik z8=VX@;{=r_$c0X<4>*UB{G%jv##j&rdJEBbG`?$`YPate;2z_FmegCfL$0H*n&z?WmoNz|zkVBF!jukV8 z81&n#YA;{b2JUUzS;kHAHR%jeh997bZ`;w6ELTLAI7dgGsYrBC+0Smv+gAusGJX5< zWg1POG;!#rNh3;M<}Zs{?J`IGnCU6d{s&gbQdj#=Id)TdR}}&us!a0fc3RL`k?~ga$mFyNN0lG8 z5&_}-WN4Xwz#zPE@=^KXTfk%uJq$QGdoTmw1ZqBwLBVBAtJ1{zrvRzzJ3Woq>hu_; zMv`KsMDWN1H0iw`t%{JA8EUfwd@19?bN-lVrX!EYdXKqpd9G(>VmQ^CcGi`~I!@Ghr0uX^U5jp+xqT8J@S}9-K_3r*zRYzzD@4!@&Gt zHtuuKy25j@^zyd$Pa^AF(d?o_v0J}b*o}=<*foWnn}U|J+Cs`CZ|;48S!r}(JKfWW%;7`ApsPx*M#r!GU*&NpYgJd~C84sAxJrDQU{&dv=*`fwe^^QfPbSYL?f1L4)!}9$YdSDQr^T}``3qmf z;~Lr zXsViep%`G|bmdm`X5*&|m=gY0iJEYn_loYot2ee^&h`^NcU>Df9^j|UwnP!32mI1i zMgiwE2%URG)p-4O`*%6M>m4z9Vt;oz>J;qkU} zgl7*qnE{{Z#~O=7HI0fXy{F_t(ot5J{UMAkwdeE;$9Qa!P7b!00&zow1}JkxQ!-k~ z;mKr>sZ<8DjM*3}0sC+a1u}%;Cmv%*p79FT5Z&%5Zma{Fw7x}5DZIQyW~G_UKA=F( z*9z2?I(u;eWT)Fh{V@KSxBIs;xm~s|AQ#rB-XI!Atm?cr&FisZl}mgMl67kmjM<@q zLLAOBY7;VkxS|$QFs^i4{s%&e(E0SYz`FohI>aF2Jx7hfBo*0djI&WaHx5k?& zIa6zwSxgyQFV@L#a%|4K7HHO+_^7ml-N}zJlJhtne_sH1cBK_C;G?CsaH~#{lO@lx zFnWkU!>5j(lf_Jy`1Ygg%6jJHW9LP^@~Y7E;FE#v{o#eYHghEILh_-zBr-A*+-E9z zK(YI*?vi@eZ%?%u>neB)DVoc`Hc?SB)HdE@URUf`7Iu1XsvYp+NFa_+vMjdTISyWG z>*hc?6i*{eh_s!{dTjXW?$e&i76$$gn&H#n6(8zgt#hELP=6o4tAB+oe4#kQUBuPP{~YQNdGo z(-vN48s*X=q+7zlNbbGAjGg@XaFfVjb>iV!Uy=0wfzNRjj*S)6G9~X3iH`WZUj#R5 zo<0iK3Howl9gfl2J#zyiXDeA$^M;GmH+7D5z-g&^%?-VbNa8-^tH+#uqgYc!yIgI< z;pQ#S``eouJq4Z#uftxaUk&8eFnQ1hdNpkf;?Ju*KM0{~tc7AP;BS-Xhcn>|Q)Y}L zB~G$?xDbDgYyt_&_>JH8f8vWl=uppce=xLevLbBnrLC0b>T9x_1)crRposDGyhC~O zqQP4d@75$F9TNs5s0IKVIxBSM-#;Kz(-bJD`l8s!mtM*{I5-?3Q`{#QP9wEW(;U@8*C)mfB()IU{3JM zk!25qksa5ANPIuFAWx8~hH6I;H~A{_={_=v2Ce9YT_KSB+a>lBV#C%&H$I2jcYH5R zqN;YuEy4@u<%ue12*wqhKU`l7Xed@s)y6i)&Uv|z;$@q-hOhFoOhW-CmKOkgC%_vX z?c@Nis^~i(3f0(auDt9QS&m_C7$z$@GP{aLs(-yoe>nTw`o&&Z{316wv{>wYXlka# z(Ann^&K=hfM`pVsSl)?2vtI17FWA_g%8>urluV*AK;-=nD4=17TSI+cW{(?xm&b6_Gy$ z$<9a?0Urxe;T=P6ig28L*{;YDbWuLGIMfSVDXdHYwa2ep(Vsh6JZho;h%4>iIClBa zFD3#BH3Ra0AbM7B%vJ0dzDwW#a&C7oaoNn|+|^`+X(oM)js7zU7~mo!$0gc1{3Go3 z2OFF6qxlctaJUP*Q%zPf`5^xE>^gzNdfN2YYBeS)5iY%$s|b z@6Q^jMabU-=*Qwhq_Vq&3~z_$&qyoyt8e_BqQ*MNVn7%{t3A;-ZRfHRoGP@|qL)R@ zq~d;kBVO+MMfZ9BcTu^-IlB@)AGUGc@Nl5MYvXA}A%TeeN;=+Z4rO{|!e0#XE}`#2 zz!yq92-N*gYqjE$@&1HrC^iwacjLdxogNn2csDJ5|AFGg({&Xr7;KSK0;c+z8JEN{ zvt6#$+N{!Z!q%k z08B7L=o6k>vRZl+(cYE0He)Nz7`#e|O^OV{nN|E!rheHs3;LD>chuVnST(=|qw*AE zgNT`fW>SxvEbr}~AlmBJ(cp&Iv%Wkk0|c{gIh>cm+Z<}^MNvtuna(~7^n*;!VcmAj zg(ZPXz>)x2jPv~ke)HFrTX6!dY3_G?j>B&t&DfIs*L6Yd0On3*iG^QfiiuHY<+kRj z)c$)Y7J7A(0laQw^`gmPX$4QQ_rmmo7J%dN?CBdg?YejL%(qcor~`>DXcs~C5}tlz zQ%N-`5!F%cdoDL$nVyj0v;h1%OywIi{=93Tj-VqW3pZ+hJB>F8WN-9rIewKjLR6A( zt5sVj-g|txy98s^Q%&Wk?5*zgPqFvNZ{r%O{H34bMqgeiKZzlGm)R&}6-Q#;-(Tuh zVyDg4XNwoW1A$B3P7Nj|r?~C3fq;Aju@}%Sy;`v>*i#k;3 zmNV62U2h(%s&m$99R1lgZrIP10!BIQB%h<%SwUNIay3FaJ@N2l&e5w)4s8aXA;~uL zc+{5?s1o}qNLdzeg+E$v+!fLgKYK&0yy4aQ<`;Q}x;Y!AEzvg@g@wKI^ihkBOI^FQEn(BH&ELP} zrZVzjUYhr0FMFL@G3otD&yHR?tqbuQH@f7Lm^-!nY1ljyh*~s3m?^>dS7tZgtYOxH zCKM-7?M3XL-%kD9Gf({e<8{%t`PL8TB!Nj0=?rBk!20`VZWn+8&r^t&izSWspI$bB z)Cx82D5$AMCJndY|7bePfGD3XinDZgm*j%dozko#A)wMohjd6uE&@_ZsYptSMW;)b zbc-yFgn-f_CH2n2|NY3PhuxiL=FYw6{0=7*7NQ=j!nFy=rNd3!e(Ar@OEZg0^fPy^ z5qTgdL4(H1BiF)Cckn>i0r$*@BVa`Og?zz^*_Gg919_GDkp=XNOU#gzb+cYXXfH}U zsCRDOyt(={1p(mWrV{iN6js>&y^6K80w$Bcdth3qYoq9IJ$=gvqtcl9I4yBHh?3Fp3!Iac4nMN!-v$Z@{khMN#sQJN8`sW1qUo*Hn_Wed~$EA3?BRIW$Asm*+!eS zvbMI2HbQ!Tx(|9=%%H}caPll`I&4R$WDBf$oJwxnNKJv(3OMzz7D7hYoU>^W(sala z+$#^V9X+V&8Umc-XpxxrWZ}SsS9>}BOrm#n&@D=`jJ=7G_f<5(d<1`w&2 zF#LLLVfZNF>NQm+TK?F>#rCmf3KxUGd@gvD&5BY$WR&}R6fp#sL3^$U;a84pCGm4} z3iLC!L4Q$=S#EBk^uRYd7jEth#LvRpEter^5?!qv!rWy%AQNG|b{AvbZMvF(Ny@Fd z4($9BqlI@_#jYOr{e3k#Zy#=~cUG5;HA1Y0GhUjUgjA57>Ph1}Dvj=hp~D=`MCW>s zl=|%h6#Q#k6AMjihIG_6h%mSj{4jup_g12XrkqU|F<&#Fv%@FR)>~EpYccJ;o8|$M z9yfN;3}A%Rd@G34jz9y&_GMoEL)Z9g*JB)~R%O&4)WkhFDQ1F9oTK=SpOUy5Z!7p= zTNFKVjdiA9j92er@zEG-^v2!`1-3lnSdY{&d{VT@2~-hFz$n<+8AF9wcei^Sc%_*D z6G^li=XZ|!OVuc1WwZ$mAlaU7PL6J`A+NW>437&F|Kd?!^RNgvon?_;TK$eGdOaTD znABG*_9X|Rl-ZFl!4*Ep3o#P@r(x&TLNaXT-n2&LzVG`Z!O!r%4bfxo`T5I$BJc{!os8Fof_(39&3+-Y)<8($awGhymJbqeBiYrCJ| z^`1=em!geTo}jT`03`3V9)HlGR*|o~pjTte4^XR7B^A2dK?re`%K&tqze zv_BgtYrrS-amblAw#Z3XhYXjCB}`NfS5|xyRdlSS${j^{c)CNh9Z3R4l<|!DdeIX9 z-AzyV^rYR{kEn+-#p;IR$hfEq?ie|KP^ePI-){Zn*74g)GCU z)LMD_k1olkS+L}1eEbI*UVFziFk#Y(uaHF4)d+aCx6g)cc|oaPN5vpR&YeO8vpy(( z82kbPx0yZp%ML8*m`ZL0#LqeX%qy~aZwiCqu8OJhqkm%dGahe9(jT!T1v=jGq~$1L zpV{`v);AWIpg7B})85fTbhP)b*s`&=%`DNogRc9*Lvkn8wtiK?XCIYMB>nwG7N_rHgB776 zPAELB;`K#buEV4T)1(fNZhwbEurLy)tlLXOpZ><)th+JNh^$k^#-V_ji&=FER*!j2 znbT^IK}B!hJ_m~X*@X{!oEs+X%UADXJ6WIB!0Vi)SZ=X3Cgt5X-~2Dvt*J<2e)hv- zsOE~Dp)=Au^|f-7AC~L=-giXQb&xSi;r{#k@dWA7ARA(yZpDG9WdjG;gh3}>@W~UV zc5Eg*d$HcGPR6e&(1`-r5Rb5Bmj_zuZz2d1zL+AcSSJtxCWP@E4I2*;z<@#-@1^{xHx zqF|)e%QgKU&ELK3`C7^2rfKMWP}TH*kIUn>?J|4wa2fRu=U%CVu2Bvw^Ad+BOu~7# zSi(0-1pbJwK1QP!IMCQtZ&^rd^j}hNX5rFE#93(0mzv8x-IyQ&!=JBNGTx0%3P-Na%HAze&p6>$7`t*0XyaY?0X&{!bnWQ9|d?u9lgDuKoXi!6WA71VCF{ zKgp4ap&bQ=vv6{pX%dpCp6avras=iYCXDGr?HX>?R>efPB$$FEWUR*FE9D!a{YwSVviZ_XQtC52jVSkWAkst z=^I)NE0jN`o#8f#IEM=h#>%W-(HkDiA%wygS(H|7cvQ9^VtdL&y0?&lLEc{&mdNTl zclUA7o_ZD-Cw^IsV)7VOr&|Pajc%Yi9&pZAM0*#8uV9Siuh|^9%k4gm`cEl2a*=;) z=gZPM)FCGUh@UWxmti%|=J^P!iYOBA{n8D}1V-Qke?JpNR`$gsWqRjqA6_-&r6KdA zA)^~85VzRLL_VFOb30j_waWG4MLtUQ6px^T9MG;Gz!6ghz9kpRgf$VQAG1*HMl)W}2GbGze;ohS6VDtld=KI97T%;oYpy zW46re7{5+!)inegn*VgUvw}^Z1MCkv zuc5RqcwP5SGSU8`ED0rl<`Mtgm`Uxw%%8lfqSQ19%AZgP`+TY*v#MeHs1&~@E&Ik9iX<1khAXW7T;E+jjoMTsm zKDv|4kiE?q!UhA~_m`8s#`bPv4;lB1s)Utt0RcISt@V8>53*gbEN~_Y1dNvmc|Pv> z=YVO6o+qplMMvD9t`xW-@FRvNOl}9;4TRq$UK%BVi~+Hz~Y)v zv%Bk&>A`nD-uPpxfa$c1Yq49Q4_6%tljIK||L6cMm(}d+%+p#TyvN_(D28(C8W3Mq z@nk@G`Z*34OV!SR05D`+*?$cVubAW_%B&4p8h-H-7Zg0Y+rXzJ&^;s}n<#8-(Ci^P z)JX2Q2eP}Q7D4;21C)JkQqZ|$Y2WH5JL<;&lv2JJs_)JovboA8H4x&*K~nx9kPJk8(IrHDqx~iTk%{6T*pJQ7Yc*)C2Zw^3t$(UN~6^=b4wE7cCS9|Acwn~wtNjk^o;Y&iEj(FioR&JcqFnw(P z&Bq&)qcqxHMaDsmeoC8vKD@g^59iO{^sGYh7E4|b4+LVP!)gwR*a8`JO*XUFls_jQ zEI^2IMAsCsi!ZQw`v4o^Uw8m$jfDYQ&1*jRtl3T&xpMe$w%XTg51t&mH+%UuljXhz zY?uZjQXMs`L;Sjq6@;U-j!DJSo(*Q1uCtII5D-DXjm4`;>CfQ_(7#N{A}NUf7eUrU zef7ov&K$rIwCY+!ssB>)Ohpdjq%tH+?ECW62T`v<;vQqH&J(S;i&C`6Bu;QPzBaUN zsa)HWb07dHjYDYIIyO$ zQaTyuo>_6K(NHpe{{fZ)ffWEt|9&H5AC=}n0GVdNdMaOH&8MNqV{4D;y@t8Jn*Od= zw4dUa3;Wfk9zm{5+Hi`rW74j(!c20gHfe)-H%!dBK~kGZ2ax?iO;dqkdDYlPe?C!A z95H0qO>_VFlRnrhg7-+;IA`Qf=%I_{7#EFBZ1#f{ z+#c-xS4jG2xCE@|O3Yx<|Dgs-+Z_h=V5N#0Cz{=UAwgtn7V*;!+`rqd zsYOV?6mn&ZgDmlE^e5?56rrG0(e$QBQDEsYv8^_TCN)euHq61B>oQFXX7a7!Sv9SV zFU<83Ra1^E$N)-m3R12^oM&cM5uNI*v(S05xcVru>Ni4uy)`}?@#2X9C&PkJqs!;G zrBzNtOS=SGh+m0>8pAF@_RXW|`QRgS^o(;=U4{;*h46`p_TOH`=~<)R6&F+Mao==& zSkG+54GU9eqzwn5=Zv5AVVX zD|g&6Ru2{qRS9@Ayo&hkf zv9v~94-65{yhHr=7gn{~RyJEF;KPI?dMnUsS2VrwbuBQT(gUP}to=~duiKxmzGp#F zJkzyeWwi^hPV1^sV=E=IJ8U0wHmv#XKP3WYqyw!ldl|JtUU4E63UYHQPIK`}50${t z?O8C|RTU3=yqV$*m{7L^Wu1G#R6>pYHwCmKHZE=tnBcDB$-a0lf2s}!-ReW%Ut{lT zJ@h#-t~L)dkp~ZdLB(&G!^@Y5->*^DXAU|_o7|3qVf11FuZpxE<@~TX2Zv1!6}7+- zr2a8RwQyZ>(X^Oz(c zLGc?`VlyzTVs+`>>d;4@IGev;t+3nnb4Az+su(*`_qkLS^N~An9RDGxS_-l}k!wAB zF^ED6w!!@S16=l5alD$=EyiU`Yrp)4O{-RF@CuUBO9J_HtDJH)MqX|%256@-f*c4^ zbR5E?X>^oB<8^#ieX!1Tl3_*v1B2MgGek7MQOuW05`2<8-gSI&R(v&=2h!=|3*4jY zTJ6HO2DV_!>`GO7_Bo&Se-k+&2ebf;MJf;x`BLkX&kpI|^|n2bvI<1Qk$t?t1_VU7 zIBrb`2D6|P2ajz!dK`G#ibZCm><7YB`t+~-cMXC#A6rvD4v`j0Guh>h@@g`!)Q|Vs zL}X|ekLD;kV9n@;k>1I`zr^I3_=CWl+`TQl#>zIx3(;E}q4cUThC!zR!eP$OZ~n|n zjSh}0H7n)Pdf=}D-U!kdjv+1Fd!mF>yR=(P{VPy z?q41+lbxC{i&#elo5!|t$jLE*_-`hYT9+7*Boh#*&Q=kyZy#8Jjw@Ins$`|`C^o6- zd#npwRlETCzr00rpaUW*mWP$*Ws7IZ8EztD)@6TNb8*!j=y)6^8(o%YzO_p)^VP@GLfB^Qy!N^8~s6-Cw`a3G33%7;*<{jaD?DBl}0xb&3s-53`?*% z{_4pI1Tx)Af0k1s33fTuanRX}7ifZ^qIHO=3pW^9$9&+4PtAXDn$sNvJr{4*yHw;z zW6VW+(W9#54D)#3=Q>Je8;h_4?m!*}!m$C41f0|+5|@87HTB{PGuc%aO#9vZ!|T#i z{l2pOj`a@)P|t2mh|7)nL`)=Eg#0RIHIG|c*#ZRWrTx6clD@<=J z6|3Va;8x$pjE=?zv>5eaq+8t?^$~OaM4lY$?Gv)w)Y6;6mb;b?6gSn_R9Ikp$LLNf zxLthw({(c4d@>7y!|ZjruZcfr;R&3-s^UX5?}%{dl222=BWSAAD0PKMUSRY4w{xS~ z5h2di< z8aI0H%>P&5s zdet9|Hm4mjEkJ~Ro71C4QJt{&@&dnBZyC|Icz+|iPE~>tCq;*?;qHj;fci4Bk%ORu zjiIMU-clK^O@ZKQ;~cpKz%A;?@Loalcx?BobMJ^rif6soz9L#?$3a)qc5Cya^xFo4 z9bIt~C1ki!@;Jo7UHBE}z=@HkM+O4UTHS=`B|KQk*%R9{qWYC*$Vra6y(s#n9>@QjC={Wogyo4LJb^!v3GwH5r<&XUoeiGTki*r z`AU$yd9L{?8FKcAXm`q?HH914AtattUI$M}kjQ=eBbk98!dh;3D@lQhcUqF~L35{P z;KQV_Q@qzQcOD`{ZS7%JKl_aMO{ilxT>xP%;7>p0^|e0QJUafQbu~t5)f$L7W_~?? zgQ_do;-G>@2r4ayC_VLdU-Pw85vWLgb%?=ku&&toX>8OEp)r}-_HTB3` z(>hUf3Fq`Po9lXY;=A1=4BMnps;9sOfI0fZx z05gh2jPiD`oJO=$r6$*8+DE~gHCjiT5^`#$tKnGs2S@israqE zb5~Dm{L6&N?0b%~Xf4uahuZV2SJ`vZOMIfdiBsMc&Pa%SDN=p@ zhTTjRMA?L%Et+W%67~TZd~ERHP1WuMV8R+FS3Zv^2hqmp(jQH#EVpdKY!DI@B&DFB z05m5Kor9=H(uHivb8!v&<=P|YIO)u5hI&f8~I5=+gO+}V} zs_U=NpVlLYp@*?_fdG}E6HAgbk0R$lsp;(HHh(v~uuOO+RzgIz02 zYVnh4UV+`D!Z`^^4%9!NSU4%;kAGu|GpyQ|$KXQb#!c#?V#q+FJS>^Bf8^tXiWr(W zz1rtzjEQ=kKUXlo(2NR-VG&Al_V;Dxa#y;Wd~?AXsIJq1hi||6A7f7NVR}HKcI!a8 zm@DKb07;>FI_SzA2|~-Ku72uAz*|)vsUMci+&k-)!c8DHehYqE>RvUWf30dw|FC{ zq=M;i@(k}**qoD1HzrRGG00y=OB5jZLQxwAkm zyo<(Ryh#CtII0olrk0}+OwL+5Y=&Mi52@-?T+(r0(SRBNw#<-?aw+7?rzXLSieQob z(%#$l9IjW;7hlV#^6;VFyExY1!xY7{R%Wd7Af%sC^cHzF0XlNRYNFT}or|a)4n$4v z{y;3h2WFz_Qr%>CH8e!ri$6H|mFxBFodn0!j*MKM-+LWIq=nYsx|>%kTuIwT8NZ8E zl!UE7l=kP~UC@-W&L7Stdgtx8)C{=1=1oqL8>@+}tOcOZ&dvaKg9ul5gb_t=-qcPD zS>rCIuv^xKrD^Q|f$;oL7D3P-g56o~))zNqKc{4KZ>s(LuyK?)|Na2)rSrIZb+kN) z_Ver@k+NLrNv?JD0=`D{sK6Cy)cHPlBup4x`9d~1I;$n8hg{gLU<)MTGvq#CQj=yo z<0_RVO<@^2IF#ZHY-&XfKQMp%>4(d01I{V=mo3R3W14e^bAC?^c3IgXKoZZb*8_8s z{H0BYCP6@PGzo5BCFVUqV_gAcA8G*gOV}-9(`B@9K^j!I6#Czl0}<-!~Q*UD1vh(WfAU3k?f!2F!70qr2rGEeiO$$j<*Ar~;#d z-lxz1Bw_<<+|NON#0Wh3GgJ~nC+pn?6x zM^snQpdorvjUXcQ`0uy-uA3o1U7H64XRE;t?aiba;!mtmIXQXX?SS8T641BNmoQ`V z?jzY{y`e2uFro-%iEhmY5Q7$wsq_69w82jm32&-fwr+5s1Co3w^9z#jT`q{-8nuMS)X&&@k9aegN0~|KtV!}5X*Dw^q% zHZn32tU(OFEs4%f_hcBhRC{|QBA%Q`JyTK!sxzS0pNh1{%j#{`Ye9=i2aOpoGfeWY(Q-I zhnM|5_3XBU8gNoOIRpDS6@YmGxx(*8&IIZ9*cBjL@=yFg-|TkWyt*1V=#l+#CJpGNRzF52LGzYe`ftBYTy*&f{I8*Ct%BQYwZ8~roR#l_Q%j$cE{5mW#6lvZZ zBm)@WZgIeN;DhIrenP)*re-8fKy{;%H^q6mnqCnF;uoxlc<#j0!;g=TOPo736Zo?V zSyng`zs$J)aW`Gu_gvXY6KVZ!*dhOBBVNqv-3N}dc+&>o5&;g&tX9~E>1UU_O?^hm zlKU?8`r##%XnQo!jY)0BOF+ja?Yj7R)fxEMZYBUyeC*5n3RBQ7-%ine!E@j|EiNDG z=_%*P+_k3v?0`-x${blPmWr)GTNR1y(U^<1=+2tQ5DU+(xzmzuP#9dZM(=CY36#W__yT8^KTgQ@|v*U)?uJz>(GR1YEek>(i zw1zh2_*uEOAFZL`LN;GPI~4&)rtNYZzr*8%nkR?MHpV}`6iBmugVQi^ER_(M)c8TY z$q%T%ric>d!f95)G^mO|j#k6VmBGQQV94eVnycG^Mob@M4Yx(mSqEh()8%a3$dSw~ zwQ1})J!9V8LNR_u{-FW8@(=K39ol;;O0ZnI zJwZ6(7tE>X#r7nby^2!HE##+sH;yrHwcmdf?o+k(!*56e{kG8yeV8n8^RP0G9(AHH zsuyqTd`@7nR^yuK{M1}BLuy#ec~8-NE$$b+(3d3xh30{*4pS7Q8|lo@Gmt>aJl%4s2@Gwx$)bzlx!KBw&{D<=?jn}CH5C=v|R#qv-$;*FO_pFZE6+^ z5*3_+{{@c-rf_V50;U`7`7L?u8?}S#G+oHwJp|Om^6#&TpEiPRbTjj{D}tEg3~nyb+#6G2)P#aT>^uZNCFJE_z|h)`_DmeBv*xKs|RS;;n+&u?EG)5qu`rbtV7}3SwkH}gUj`x;b z-#^qgkd5v01Z0Y7plL_c68eU&y4Bu_)6d-Y}z1j zgdactyk(8Qs2DW7+YUXDDl>O=&V*uOJ?^lke3j$f++vzv|3%Hj$di%7n{K*==j)=j zF6G?NW+Auu<*K1Vt!+4h^Etf`XozfeB*nKZWigvv+_BtS!Yxo_=whTgS7-3QY0=GmQVh z^!wSJvlpq<0QWI$P!?h^D$)i;`&=M}LTz0KhHtEXw2>M3#*4NYTr}}v;sqSwDg60e zhSL`d*AB%rElN{_GUxzq6pl%vh+gw$=uYtCbN-%>F(=B_-MaXLX^{P?pSP%PM#%bN zAI{J4TRB_MtXx_~4`7p0xy1kK^8-FsqQ=wGDvG`Cw(y0Qs+ISpArsAN4_QoNJc%&X6j>Ye1pJ9j0j#&;B~@zZG}yV^J#U zisPF-WFGy#<*~QHpeBKN><}S=NM5!8Vcw;y2vm#MW*&s`=uAKCbMvXhp!9aq+cmt0@b{1VhHVJrH$+Q$QBRE7uVVQ z%LxjjN(jk|( znV&q#gBB5~pB`G#OWY!!YuFb*p;5%$sRD~Iq55tb>dm0# z0C`0T3Fd%+MJVaX+R-X@X1;)%JRn zz;9%GjA~JIC_Bk9<8+aB84Yhk))k}(S9BA07jn73Ah&*BPMHPxlLj;249ESOxiozy zYk41-2=3l&fti%Gs%>yY;7i{lG8?u1$NYYa+#Xv+3Z#zh+Z~wo{7pAJ9eHj0)>wg7 zXRJ2IB%zc`yk)7x5Vc44MXT*${1tS{=4Yg0F@o#cr_~LD1#x5?4q9 ztyOs3qJ___>#v%HhPulbQInXFO0|Ssm4~dpr4_thCx3oU^u-@XhzOWHsp;3Lop8BR zf-6Tk<0Ei}ZbRr1?d}9@k~fd>z>*Lzp#l1=-mXx<>?$$9e2i5VNAW4?iA)g^@^%1~E~A#7ub|Z01cP zF{?ZpMUd4jg89R!{S$uqAE1L|4u1OcrJaCj{r=1?A-)I;praDgmd;#kxjZk+Mt%)w z%F^gyAe=?jI2+3G>7hGX`6{Axn3*9sg!+hDKHuIN+Ea*UjpDH3!yB|ZAxA9bR?js( zM)e{hUp74p;CTq4wfM8=%b$5;qz0@4tx5ulYB1A}@W5CHy&Q*^j%b^-xeFSpf#oX% zC<>e49yBq0)pA?1pk4#^czD*&a@EoTt-rJW7)j#`*o@x?hN5Mk$CM;bsZy$(7{hZ(OG@5!fi? zMDuN7|8l-(g%h6u>aa6SMDU3LLx#BL+P(vMI5t>?9v)ULA?F@w$|+jc1P}r)DG2NM zUJ265D^B;M;HD*?$Wolbc`|g`t)eVuy}>X`g91YE(xRAuHqhPB≫ebiyGiKKm^kycDt<3e0}9o-eeEu38+R7T5)MvKjWDXx9p< z+@`SjyOt{#RcCFvhgZKXn{$k2tvVM5LfcZuffAV_PQR|wp8xgj?d(z}r_{3gUI~@V z-u)qO0=de9c8$u@?bay=!}(q{Y(?eZ$Y^ty#z5Q#S8)SCu4vZ z@u@pdqDWz664XsY1&4f!%%4(tbda<@GK9P_Uv+bD9OYhc)V>LUs(rVz;(F<_HB)!z z^5R^?f7cQ{?4G_OilhZF8&Ui5t0&cR4`458G{INeV5s!gtlpZtC$)H@>guie>a9y( zb-tYt^8C_iufJ6vGtL#Hkg#vZ!z95H2YrGi^Gw`hL|JgJ>wGP1+5Xv`-bW8#sS;cZ z>GiYR^N*hH=*Xlw#wB48+49z>@X^rYgC#wn4WL1b>F_O%uX}ceW9LC@$b*~J_N4nM&=(qgRfefD&r#Dk+5OA0KYzp zx0aX(UGppy(XU=%^ru1=g0PAolYopH=6j&oOZ6LnD6oKpYt7$#g6ZHLISu>N6*|S7 z3HnWRwE^37`YD(fOc{1lG=88ohDm*swEbi(ODJOE0}#FPCwBe3SL1R@yD8*GT)oODK|jN!{-7FCpke$Fn)9B zHP<+P#IX)+;XrsPH^wgoU)+)LrH$=Lv2p84Z)+5Zn?-{!dZR{k+h&r(6Rsy~lQouJ z0)gpOB#W*_#T1t2VO`I|Iv%f?r5Q093I0h3 zobr?VY;nC3_h-1tFd2`;jyf8t)Y;m%L86Dt^orSC>HC4hx8ZrJU`Fc5_WzZF^-tTQ z$LG)sg`~B{He?10Zzgt_>D`n2*c{y4N_xqD-Aa4k-w}aK?V016V%4kV5;6HmxPHjw zLTJ~%w8Te{%X2GiB@Z^k0B18y3FCow8J~S4tPJ#9yVs7K&MFg7Q&q{4M88RsKD+|r z&x$)ANp&$oCj4_V8;X8+g+&@9-Mnf4>s}KJE477r!!y6@XOK7G0`pVchDm?g&f!;I zjFf*6#50V=KeWVZAJ&ZIe?2FVoK_yojz~>9%Q%g2bo}oFwH8$O;Jj2j`#S$#IA@f>9_9Tg}Y#|Jl2cH2sJKkxl0Ycbnf3F6$Z zs%pZ*HBpIr5$G4;7`FcM4!Q<{L%yQ^?ZAJbu)dYz8iZm}+ohk+;8%`WTkdmNuDZ6q z;tS(0L+;nUAzGgPjC?wNwGHwSJ2!I|~BV7sqwSlf7$6=O|l` zIi8CsjJG3ywRTy6`8RwZa^`-m9$6}_bdw_i1VpLC+=fFt23#%tEU=lY zBW&}yhW33M17!+f8c{Nf+;A37(Jk>BP)u;p%{z%b7YCG{OsV%2WTeD8h!UyoY zUHcbAl26W+Cak^ZpH-D$$EI8Iy@_778PlS^Ur~GE%oc1QqYTA-4cz5_Y%Q16P;Fg( zUA|jrzwyO6ypRis)~;5Nnq_;$q-;T3k(pcgX&ew>GL{&wv&=?SGnBS^EuFOH(}lHr zz1>xaMez1KUL@r0cdjpYWUEJRI8uR%=Sj<@o9#I>=K{jkVboR)oBUIJ1U|=KFLa@# zW~ro^SlF!eSp@n!0W$UmlW#$4CQwJ0rU>G!AfoqAPN^?P=?THeQ{fTo8=R%mBvVud z%>LPK9LImBl!k7xxj%ms_HvidcD?uhqC|(Ii*Gp%H3ncwV&Q&)qqH;oWUqni(XCxz z6l}T`%PgjC*#a=ZZB(VECGyp$IhP+~+@VPooiw_l=>~7xiM?F}%lG_HByU89xp(ImX4a{abNs=d4S+~dsqC?mb-rIdxpLU`=7x_EFKmrteYzZWafGR@ zwZg&xSz)x2Tw{IXsCj~u=F?C`Mj*DycnexPVdyg_mhjdUq+LeR&{& zWzdsqP%{XJYrMM}o@JN1)qNv75+BDnj<)q}AlMK;d0#04Kv0=#G{Nq&tjuZG?p?O; z*#G)3u!_$qdo(>Y#GUHx+%cV-McYCv2#y;d2t16upSOM+arGi+K_>1Wgboi*<^aS3 zWaM%nk%Av3eQuR|MNtoVsO6hwSC zK&oO`?Va>fz3q^BS<7!XV13>* z8TKiEGJiGPjUXqXpFR-t>}v0&4P?@Gzzm|*F1;k;6%@`W;Tc&}^9CM?+L8c_re!Hv z^KNWrmV1*Vs1I@@{z|QaDca!`sE#L>U|v2=B5Ic2+M|MPu;r z53K(tEv?+keV*kS-SzYO^q}CXE<8FtDJ4(wR3Q5GVaMdg?<@@?U5!a(?VcfZ@46>d+F>G3H7Inl_|Elc5X-;EZm)D|q4V+gN+=(AF$5o=y@R-D2?h(V(R=8sFpit=$9?bH>l(Lp#+ z*?^vpnP~w5XCn_aTm=?oTu;+#&Dk&G3`j(xlC+NBy11t~s$G1#03$PAQxMtN!1J5& zzx9dnN(NH9bQ$&w1+V~!R?!BkcnqfF=>1g(oaJ1e?AQ2wtVFdKzNZj~krcu001y@ss2zT?(kSTsa z2y*9NpeMK@XpmP$QRSsgYHD!``%-zNr{BroFH1(;T^$63b^N=u%guApL=U*s*`^o6&EB8NG=u z@*oTa?S`y)`b&Qkv<8j=@}r6Kx-Jo??h8wixBPBo*`*De8N`SkXg=0ax;za_;E}s6 zwX2u=9X=!~igIjKAI9Jtw8>y3Nt>jL7XEi1n|TA(P&obfmp#$|lL-ud@A+28)UToQ zRRFT$Ia(zHQZ0EEE{<(Ja|6)8k8JkenE(HkBGNu^5k`x($Eg-G9YJKX37$4|lGF{E zZxnM%1vv~oaZku5$Ps_MA`{aZZsuhCsAv@IP>d=n;0OqbEH?HTu$EtadCCh~l^QiZbSr8W$o znxSp96F&@OEo>!vo4eS-g4F6TOg0^AgX2kwvBp+q=w42eDiuXD8@t zi$gf~iCn|?Cx;_wdPd7&_@Mq6|5gi(|=2uhK*(Y}t#w%d=7 zI^PgNnZqg`zAP-ld@5s0H8EYIV5C;b2WT86*uoXl z7ryZ}T!QSV{h{k{PmU3he**lt(XLorhqz0|%mL7m(`N-!@oYT{D^KU@h2Rm}N+_b) z0D|mnb=wd8f@hHlVaxP>H!=Sjq#r)t``2Si;dJ!Dd4lU#@M6_BcyU$iJI{ z_3IqW9<306fck4{(lHDHo=2l9dxUnScA9XT9ETH?`7~O}qxOS*j`%JsJ_w3&tireK zzw*D;OoDFvl1N@0m6i63Om94`MmSM~Z8Y~v(=!ObN>ZivD{T3Q(DAgGIHn;EfRTtUC%t6Fm$xp-`wbngz5!rM8uieML*gjH}cypsqwdj z!|E|{I(&^$;R^3M&yKVWa^Re8Z^Dc2D<3~&R6N1{a_R;m>LIlCdlz?ZK+2j4XM?IV zYtxND!+!qxS;PeU-$e&MW>-~50*#rZw&w_Y*6sSq&Py~%54`{Pe^mAhcCdf}$d*q8 z!@GNA!O%I@fE5aKaNaz{^=6Hk^~HDSqGHer91HcMzE8fzcZsRwjYN~s8gy0TQr2S% z97)YMvV9~YzlEq1yB9BRueU6-!^x?3X zSnf}`Tpu4OK6*HD8+kdx|~;`)VR5%73K z--i5qvUAs?75Nh$@n1#D))y8;)mwH-IC)k_VPOgasPxFzMswuH@vFP$k@=#!_!Xb_ zvi;1~7+dC9tEm`3Qs@uqFBy`F7wt|k=s5kFlvqYUD~1EnR9%^0_+;M(>|`Fq68Ro7 z;yp0LcIEt40+? znePQm@3vUJnFfFXapad(MtT7VNQvK_c5YQA16gECv-dzlt7qttb%lrdX8E%k%~H=& z-yr!BAY`+}FF%@BveEb>%OfB#=+q751yd^~K=#wbIrAPMeM%CIab;i{GDZihiV8Hx{cKYs^Vki)fNs4r@J?W7LgKD%MvnlZEBXcfBgJWmd>%dhf&yD5<=WN7o`0^ds(yMKt6KUjzO~vEWuSVyQ_0 zA@+reRNGV)oVwu2Ib1_>I>USBtw;Z$k1sp5KXj-&U)?QdQl$TlP#ujQ0Re$8TTRPr zzo1Q^V2B1;J@Kr!r+tCL?Ksq-U3lac+0`bgMxx}9vT4_v+6&WKflv^*MfvUrhZQz^ zaew46HZB!&?W1<@*0EGc5Uca^RiJYIA5(7`6;=EF4=de`2+|$WHImXpOLvIW(A^=R zlyr9v-8D#qAP5Ku4Bg!z3WN0jaDRW#TF(n!>0%9YuCuSb_orS7eTAA_cPi`8KjtWk zcxi7~2D$=Q!q$@<8u@c0ZTjpU~E()dk78%M4W&?Grr%>w=cTPHOXK<#MBbpYs+ z!sBoDD8>H#U|3D;wFXk7goft}WzYe}Wf2ql#s9R_;i?sccoIE6G_P|U_;767(E*Wj zq>|7#`ejkpyU*jmM5)7;qq4)=dHKkdvn}D(3iuL?GpVe)=3}uSBQJ^o!51|kfVO-N z&=CFot0taU;y_j=8h{!K4a*-5zug6*78*^wpNIrA@v#(wjo-8WubKn0Z7Q?xzHi@W zMJ0HXT>BX$5?SlPqM%3Fow&>+`54vrNMPPR%KK}%n%=KXio~>?ScXCp)Lw`=@839?LC!7i_h6Lx;iVYc6k%Aza@am zaiafhQD#=)$(h6u@N#$}c;Br(-sgKW)_(<{vyGaeYXZIo#~?X^MZ zoLMl{Q(C%XL(YggF>pYN=z6yo+nqxeyfaUn%tr;z-pXa{`|s$R&nUcwJi}?=9O0Av z>tUW8wK=_)^+RID0rE&WkVZ9oJ3!HEP=xiVR`=e<#9vD`4k`Z_)*F*dJB( zFSeb5Nor{T(FT}x0MVV!8$hVZKuNmk;Ykg-w6wH?!ZK63vBmwx1YqWU243$Q7xbz> z-#Y<#Vk=~}{T2TeheHy>T{5dz3)W5?0~|TavxbrcKi2lKSmpwpK7I3EB2n{`Y%M1& zD#|IXinhmCk^iS}CuFEPo{1cQ$UHK>4MvP3tJPNCZ1pUdCOJ?Xh|_)kgo06VLEjao zL0Ltpx?e6FID~#D=X$vd2_uJOi zYn@ksm~_v3qHS+`;Xq+9DewvxiA5ncX%QC7wGSZy;(ND|h$HZ6PMfs?pVDqbS(X4| z(@MF}9pm``z(6ks4u}SsiASPS_qDa~nCSfG$D|}8y{qRFduxvm-rh;{7vj|MLt7A{ z6f)&k5-BfMSvYSxnm2vf{A;6X<}D9Xr(oPj=r5`R`xlQi4LIEt;iRA|m(rbinA7z# z%+Y$?LHWpGHi*0FVk|UXp>JNLS0kLEy0jbT&+%5uD?Y|Z^Bf6rFlL*d=X(9ooH_+B z{mYi!Tzkp63vhOXMbK4R4B$R0mWJu^8hmpRvi>DT}|5anAy6;K}ZkPE@yf5>t#&W@aQ;Js>A;D-b^UZ`!8z+>%t(+Py$(KOA93Kw_0B~)i>QMknOZ1Hvm zM!sX+mi<}1;mBzyOKE;cXvU_dmYV>)m&ZXB9N_Vd>uqP?_zf}(a-5KZiQB=4vq5CQxV!}0j^QxD&-Lv# zoez3;87w6|{RkL2y)IdU9>=o{d0AA}@e%Jid(WorcBYz9ZzcX4VN`{-Z{AyWP6w zDw4f&sH9%E?cXj-d?^nTh7dff0h7~j=GzqHI?=5t2eg=ZDt1ve5C#Ha9{DA za=#_(ZMv{pQ(sug#YI7y5FV^V8Ck(_BCqd9iws^Om`5rLtiC0M6^KMUS8H-~F>Q`# zjQrN|INj85xCDAPo&T_H`UlNRGsL^4U&X5muN6OtNhwAF~Ue2c-U z>pk?g%PM4r<>%;N$%h7y%_y&|6Dqj*E1tI${PH5l_2%NTl6-aX4$Hi7qvF%W!U*ac z$)W-YI-PAGAM=P37l;vsO59&e@3X`r#{q^%PtM1q*)*F9^}dKtjg5Sq292@+7rGK4 zZIGi_zKb)!98?4l)l~rgtvidNn_DA0J3_27bdgNd!Y1Dr zI9x+bON>x3)sWIXcVc_KMOs5tW+<*To22!LD`bHfal@SrTisQ~5u?S(t(_&BF6MyH zMt1)+Ap$+P?arh4?^=7LH>!Aef_Qu`$6L_CBIO(|Xzt#M+&W~6lgRSnQN9FC`NaPg zx!&$291_r!+>8MhVvy=wuTHN6&5q7T&L|*6XBlz6Kmo8>h*5D@4kjYsOa+jTk#!_G z)B#Sf7s=r}!06&#d^j%o7@$YP_2TTQFe#^Vc{0pB$+)8%iN8m~{)Qy~S63-A8E) zPDPN8YfaaL*tZ6^SyIBLcEYb+n&mp*m8&O^LJuU*#12SA1Ho%O4B)HBt~1oXh1Z6y zc=FYS2VhLab;c5uUjf~>?dOGQ2*9@eKmZOvEsJw4pu@r?^eeKKL zO+lIC02(=MSN6MeJwPm#tli1qT-KGCJrEU3xP0T4<~%NwcS(!}!tH30rkdQ-xXn)K z@fDjwSk1LnKezyLi8m}uAD<%|o0GRpP$U7w=GE9zQNS6@#N=z`e>IEtD~o#Gu?Q_j zj_jw~Ud8`D?;E1A`ct=Nf!_)Uzm$|o-4P>B&Q3i;M04}z!rQf-8XOPeiNZV^G)^}D z_9V2><^<4Pc`NJ^yX_+Fz z2lC8Z7T#gZJ46Z`ql{=6ngiYJcs#!yynLU-JhAoPs*+xUM3x2o(nWK zOl%LuBSFz)ZMP^+=_2@|ZV{iT&P1I5Y3RF|%(=3U!yT_3BY^?AwbYfU*`a@)YI%y% zNSHj$hly#D1FOx^m=y2_DE@Owe@FC8^MMcUvcIzg=LZ<0R_=X~XBO}-am~JCGzZ;> zir971SA$74jVYmcnZofG*psL@LQPDjhm`p%Z{$7hrWYpmDRS&@le|j+16e9HAx*iU zGygTfBx;4^9vXBFoSK~E^JiMWBa08;QUXB}h7kYt7RUX)gjVt`dz%woj{NzH-8#=u1vFrk{RBg#1yOIO@g@ z`Y`l7qJNB-X4!3&)9dckW4vUI;GTlz$==PWehQ#tM))uviuoy4x08vde%`SlBgD~D54Df zrqUw7v$`YpIw>h>xxIor#5-LY!NkN=%ls_&br>418y-kKAY{w4FBJ9kAK}8-`NaUf zv5~PPU>xG!LPSoq{4UofFRSnT$N2ax5STVoVNZq8uB$S_y#D(}Hi{Qae1 zT0-TIg3%UUKJC%8I_Ri+C4VN&J?_EGJ$U-DK;dUA|xF2|{) zKW4Cc3*|+?DDSNx^+@r$b*42ya9ziZ6zJ9A65|(HB+;@?F{$t06|L787rKO`zPylY zelJrpH8nD(&)H)0AyYn1_Nz{1b@k3_@Ehv+U_eq?+bb|$kV{qkg{vJ5>{eiTeur%7 z^n7`}=Mf)5Lua6l)!gO2F(zY?Mfei`dKV04q!yDaw94#&w4oNi!6uyRtA@4)Vi6$4 zx^7P0&h8M=?ic*BnAWgbcI8Q^0@G*yKSqxygM(EE})CnoUan|b{Jax`sk zV@Vi-7L_hj0>SvBj>{KVfO!rxrGxG-fcbK$1aM9s#3^IiTf!V<6`gvV6SS=c8UWeW zRlh1Alsao4{`i@87LfZ4E$q$2x>T`&u%C={^aj^~C0exF3V&8j7%m|K=%PjOf5!)1 zFqrrql&JYUznmwq0?iXx>?L-3vA7UE9GN}3jf;q+R&9l1bGeL*bkF_QFZi*%raqzG zdXF5__=~5iM?nk^1|6Zd(yi9%h(<-|H8c|6un8BoD$zSSVBh`aS?%r9H6WjaIaU~U zcgCi1ITjZcG49DKtJS`ntGzy45?&h6&m()^F>LmDqt^vUn%e=Xb?ypz{5pxnt&0*o zH6ap}1!cTLr}WshHnH1TVZFLn@lToLeTAY&{dPn#m)s_;`ELs}*=9>U`?MN|CmFs7Q}E3u{s%98O0GE0a*6=ZX++Mu{lCW4)V zqiR)|==>#AM|)m_Lsi?$tH^pHE@OpgvfXoDkktBBD&(wdRM1+b#IR*ElicJh>bF2# zbk>EFAAF5fy~E)jC*Pxl7moc9A)0V;+Cq`VCp?yC``8|uXb+mjaOpUKEfr~1?E&gi zPs8kIqIr59*mLZNyQ;eG!vHBya?pYUl>^B}f0G{o9dz7~y<9#9%X4ltFmIB|^(IN$DP#}EiTYkXp9zPuUOBKhj8fz8@B%e#bbg> zJve3)J&te=A73(h?yBgplBIV2<)qYi)(s$Z{_AdCVK}Xw9KT!ax2Z1=%sbq|_6^2L zP8^pKEZ|p-$GmdU!nRu92x?R$Wx8%JG;X~dRrn?kEQcy)hoaAz?;|)Tel&N=Dg9`* zfgGSQr91Cxr9wTlAh-QGQeA>>e>)tl4f+Eew{ltD+sT?PpbGHG3(r=JdVVTwf8UR0 z!`;j`qqR0iNKX2KTmBjI;;rS*9?OzM)K@rN7`k3LV0J+Ksl>V?B~$+k7d&*s=T}93 zditJ&T+I1l9HAHO_wt4}kG~@%Wy=Ox9B3*OTwFFtw%-`((P|Qy08Hzos#_R;{b+~x zS?v=Y+621;d-NAwLQ#7`!dO|(+9$!Gw>OT;<~E@svq5Wt0mM>AgA+oq5!BHuV;O+r6JZAOrF z31Csx)9drrN9M&cp9I8#J?vNF^^c(bOd@$M`f`_#iH5sp!H*qPqzxTB?Xh&d5~<+n zknB)(CZe@}+Vtn~>P%Dm8DEDpS!(>43@`j@Qf8G71Dcz!@R87$zTK9zJkE?1NoCC* z0#dv({S`ltM3I=5o#M?Jn06ZYdv-~YV4v6eI>%_!nIjbmtdz?*Vp3CSXdmeuwSupl zf6$S}zq-MFLfDd>5N%z{DE4NcN)P?EpqQ$Z%gYXj5=`en|LXp4J^OohLWlMXJGGa8 zc9Qc>0g!1LVC)j)QmoNhu;z{ZO(m4_bj4@UL5HZ;mnq7$akzx%xQ(y@!8H+fju9IL zgKdq#Dn=O0gXh@;xc5RuRGBs_$f46r$r$8d3^0b^DYehkMr-jL(~VTOfYUjoedQ^!?_PO0`q>Su)C4^( z%j#B2)s?6N4>LQ)Ud%0CY|V5j;9IpotlW8T7I$kd8#X&Iy9HeKGfS0O^-EKxaFsaM z%=2|CK_%Hri_}GbE+j(GJP?L(YxlJ!pFPEwpaY>{TAl<;i&)P?H!4B_MH)_=WOVU2zw^zuk6r#P0d0x>HU7`!m zcEvf+fXeFmhi@8yF;P|$LQ%jlnG9M$KnH~(6b-Gw-c`S>$?h9|@IM*7;u8QeYZ>fT>N6nT$ROqOJNutv?2djG{T7ALpHs{Y zk_$nm7+diu9b{4ennVUtI(TG!u8zChpP~&P){0BWGH5MPe{Ch8PtvU1=0?+wt^plGZhSLvw zr~Fl7JW~-(zEdI#I2ufiM)Tf!DT-~_P)Gg~&%%nApVC$&iIYoa!%ASci$WKVO4Z*n z_1{F+3Bru5Cw7-{QOEmEc|9Z@A-TB^G#hik+KeD~QNY}PxI{e-o&MPybbI9t zzod&OV1I1laTte*y!z7914vO8E!28NFbqZS(*WDZ2g-cf0eC6Lz)13}Wp4iQ@!yc7 z@jBz4O8MCju0~)haDotHU?9&G0%UYoSEs$#Bxq_yM_;2MV{{ zAZR%&*f)|q)4{E|fo8I>wm6N0&Gjg`v$J58A|#nZ4{b57&eG*^y)Tu5j}h+O($tX6 z(SeoYfU$ohwi(HgGp1+=o{|xg`wiyd;M5&D05cfS_Oq4)Odh z%IPuYB_VK|a=xR-IjE**Psflqg~)y-S{eL4@cb#%Hrw>o9t9kx)%#zswbGv_41)W3 zFhqjq=FjLM`DRL)yw_#?Dmb`ziSS3F+hR?2ZrGg5f}vPbIr>|rk(oO1kz|4!S!NdR zxm%WEdj9 z*CgJfc;}BBnRMmVT*h252eKjm&-tt!0pvW3y2gA+_hUgB+i2eLTL#X$Nk|iO+B^tu zO_98~eLlqcfEoWqQUbWKgkqoFb~1`$WZnG{)_fT0H9zzCO1^)U(HE2izh;F%yUyII zQ>0}JIQapYbe<~5BFUYJQF%pLhpgdMck`6M>g)o7;B`H|CR7*sC%EZk-_2rKWbS3x z<@MfF$JxlS0c7_lt+JxnyR|vuae>Cf0yEdz_w5NLE&~_`jG~V6@p0v#jytJlk}bjSU%9)ZUAS=7+Yo6;11XDPX!O^2+cnrr_tuKJ5d)}Iqur z@TzWqh|uWE&yBD5f^v5FYsf8a7=+6>W@FX`@aOn6({(PMakuo#@C&>oAA)@+1kIJI zBvJ>w^V=po0MHvgOrFp$;jI2Af(IKBl<5`i=+$TZux zdk#lIq7(Hz+@`dN`A81<6WZr=Kx2OX{ONItqIk_8@b9;w0@K;gjicN?V7uG4dFBi} z!kmi)cs19s>mSvwiPUi84`?@UZ$)`+YkvBy-MUQ&?k(Xke{&hpAweIVWp|kxkUF)1 zoV=O}j}NV8WHNFHv>t^C1^&r42NzbFYM89AqNwVJryUrvn)D_fR# zAPf(tluIJ0=^ZURLaeb?<}B$C7N!}4Y#?_&2jJuzOQ!cIukvq|MI_pOf5SGBM>{mv z8m+F7QSzj;Od0+*EHBh!5`s6-Chvuj6LWui6(6L@e>l>oO9MR1rSaavaDopQAGA(L zfTu^Tj*U_{vtTFLm=Nig4Qb0l z1jgeKo0_o#^X$z8gC=NMUealgBpBa-DiZ$_MR$mvX(EDW`fYKjJI~O?o1WXQD`@L# z(rjIP(96$R_6!PW8gZsRGk9*h3|JVm(j+{MrED^3=s6AwG=a@k3`4dBS-%0-r}oav zwSR-c#$8)0eWLSX9oV@&#_pg+i&q414V;8~c`j`;*c*lZ{ogk0B_>3w-eQ7(+pd;S zEUNq40%>U9>W*#2^!vUwsv$JUD|;#sBs9ct(WrE!1z_;<@psWuJQ=q|>@JB4r&ey*YgDIl6c{#2NFfv(#k7lPn zJ`d4ygtJ2SfAzKj(DR*i+UKV+wdelC*@UWL7peW9NyjtSj15cwSd87=-a9~U?QV0HnDZ_X%-5VE@!r zRRQBK_x=RhylSHHA_I1~a3)Z=`qy@RPA9o6>>IK}6a_?6!h7)?)l z9@g-cP~bxD80`rj$%!QX`Q*@lXKpD69nN6Kh`^OLGKHVH9OBoOJbsP}&l5rErD~}U z7c2*okC;;|LRLRyrPmpylIwc79Ont{X_d;9fV6>q z_%3o>%N3C8R1yHKT>#$YI$$xl2T)kNLUN%2WwoA;zwuxoBtm&CLd}>2-A^72|69(A z+ffw3zWnZ*>MZC*%pOebvPq4aIeq!Qt*qk2WQCiyh?A%(I%^6SB#^yHL~>JLn5zsU zN#wT%fBrm)>52!@{H;aXNs#Oy+Vg&zd3vNY9Ub?ZD^aAX`pQQwAbVHQ50K6-|ESqT z>%rcmxV~_O=?sOUVlM~1o#zRy6Ovn&K{rwe*Y*?h$M8D^@Ui5o=kqD{2LGW(sG3I23v9h4SPMB;B);O_pnxes6YkZeNv&tp?Z1_naUp5Q)tD()}vKi5FMm%eo) zLw)mauFGC`v`jM&lUKemb$6)zrN0+_C;hpz5Up5wLpW%`gn%1VQNH$KV%@mk7vyM2 zFbVb}k+0s+Kbk^C*8BFzZy5>M{Wm+Ril*uN6rE5zJ0F*$KK7hXEP?qCK$CGZpoaZY zCbR-Pn#S8;z@*BOP%fqS>JCFl-6Z74Dd4GH#R3AFDzZ0E7Z0`ibaCl@O6i)Hyg{Kh z$r!lt5*zY&Pl}eQotv>O7Ct*5-Ej|%ITaj@oQfBh!=>P$(S*V{q+(Z<>TJ2LWiz_C90HSm*A{%`e_lYZg?3DY);sm zqu3ITdqd=TPhy;yVVZWemz$BBn{Ok}+dH=Rvhh|Z*ET<2&_acT{a2!f97G*J%NW}t z`?OIG*b$VW8;}SdiP0^=H(+%#i9C7Xk&vj<7~r*&(%q;1k2%w&=jDQ`$i_%qX}ig9CPf29BIR5f~H z3VEq7Qy%LlDF@QzLQS@Sn`J^+exiNgE!yY0F83NIbH+J<5nFQ>c;@e9si(YgyY|>^ z>UY32mM^*XX5q=HM6c{51fBZVL0LX^Zj~2p+}*z56A}U(@s$##bw-A1tXY)tX2Cke zBT0|p)q zmGG+e_=ZWu>uAZUSSzLIsUA=taP3Bxk&%?NDPPOmSXJT*_$!JGEdrQ$B@mvW`!us% zBrk4J{SQF$p&*2(L^VuNHf^#f9*o zs&AQ^_@%aRQB_IbS*f|Fre>0R?fnLR@Gq-$lNa~RKp1lNe_sc+x?`l@@W>aWp~YG9 zsWgbxXdYx*cmA~^^QRjnD!gWEq(U8MVqsNb7dLgZs-fGr-joq=eWR}l+Q-a8qNvZ4 z(H9F}@zR5@fKACA`bWR+qWbz@Z|2hGL%z4kJ2s|`0NjRkN=MIB>S^F&|3CCBb!j?? z;K23}$;PnF;N0pOQ6~PtX1j8YQL!hCyqEQ=W%i=R`_D8!4Ylk|-zf%%d@!9Qg{X7_ z4wtNb(7(cFTaDk|7;*iR#U;~f*i%w}i~`#e+R-D3Dqau9kZ;LKDXYo}0~r^3^*2KR z0$mIEpY6lN!;ofHfj;o54i7bTn8#=Fur4q=TEZ8Pq`Z=nI7_!Qp#n{HjcD0}qQ-|_ z7VDNDP>hJd=mu0!MaF-L&qh{0$54Z+H|bo6I2C?dR+l1s0Jtl8gZxLJW{fA4C1*{(JpJExD$7|x?C@iP?FG^eH>4i$on z6cH-sOEHFDzBMYWj6#98slCz{#u@LOy#>_IR?i7Rn9$H19=ew`FR(@8Chb^q8FPIs zci{rbD?nb$+1CcDd$?@+Txx%9?Kz&%(`$MB+rCO65v0f1O9(o43e*{jKPh zozI9yPYlHjuDBm3EKm4=rXpMXgM7hJsX9Iqi>jT&oF`xrJMt(M)?HaVCL*R95A6AM zfhfiL(SPT~H216`Gg|2b9pM-0wNG!X9lH1(H2X)UbJ`Yn$uODX-v`Q=qRP0WZRDGBnTk6uWl-Y~i1u*aDB$Wa6 zzgOcpFKKiAYcI0bz_-W5x%#*<#vL2s1(6;Zx{VuP`0m(n77v{icNZgI_LvYRImAU* z#7Sr@6M@#lL_L));H{*roa^ogAbEbfnKmBJd=14su@2}s*x5IQS|a`{GMr@7=&7Zz zqJ_}>ip*P7mqHYf)zR~bjfd|@OD6HfT80p8&nrm1aH56LLn>aeDA}*Qvpf>GGGILZ z#`G85L8v*mRdUL*czX|pBl>K{613D3$3Y+ViYLFV%S=g|x`YTg8UOZ{(KMyjP>eF6 zK$Sux;!EnhOcR*IV*uKtWefh~`msCaGj03DcTKlIbdqaJWre{>1HolZF`gH>kz)wOzV53lmW1U{P?N)Jfx6fI<-zuA_(A zcG?eTz)IVxbqm(=#!ZwX*(!`c`<7sJi8a#?MIhuN?LBrq{xL2R6{cb@lMo#CcNJa4D{s8;> zU#o438@#T$Uy7{sr46l4szB3=c{33Voil4y9$>sl{{QYoSFFAOIeXO)olG^pYZpsj zrZyv>0n^S6zppkUN>aHM?I#(Xm)z`&X#DI@R%u6=2waq+A64@Wvng#m9Gg`5xf7zt ztd+q0E1j+Ee76uPV3k&qJ2x$i0IO9$3$m zEjP$70To(E>ykvV$Zu0>-xuly_$zBP*`PDbJmXX%M_OF(3wff|jD-oW1!$2OyM*9L z{4-C#AJFPa!2JHdn`Qd0x4W!X?^XjC+`>pm#dMTlsN>CTtyv)b*JEu=|FS1_0#;8x znFpoW(-jryekGxU8ex7;=}Lm3Ht2x=c;#egYFcww6dN3iYY^Kg=Hhw1xn|(HdYBum zYq6bD$*2E`C&KDtp6J!k@K#$x{aG1>?~OmA+5_qQwP=6>4VA2QQu5j~&)v;j&+HSw z@_v)7=huvnnfPV6cm03tCgQRJ{vRHRiyBHoGASAm_XpORYv+W+{9drBUJvsa?7))s zeYZe%{!YO9wv)L?d|;a+tod}soW?ni61p;Q4J;+f8|251#}&#uS8P+|7w0u257IE7 zN5nkqk3yvPps+pEwaHueDEB;edCZwi`R)nc{SZWAvH4jJ-yRMLi`8(jWTgA(e%6J2 z;`@`AD05HI>;Rc18yygsO}BPma^g~ge$AYyVSPFCJ;eRr?dV137TsDP;H64NDKQ&a zZ7*z0;ZgzL3uvOyDhVJ)uQKI3_x$iXt7ouuDIkzL;U-uzLz0c5ojShwwIqR#ENAT_ zPoyQZ;L@wXZWd(05u0d-4DG@b>hpwlWzCG2xBQgQj1J7paQd6H`@LhfGISbC0dRYI zb9WRPAey6`2Brz7BxxBRdXW&<_d6UxNY$#>O=*wI)4o?YY=rwa_Lk`;!7Y$#uTZ%$c;|t{sw&Ka+?E9qTCtT5QTQO9jS;r47VlzPd zR5RYdHXD##Kn5bsQI_anUw%BB(!Yu79iFX0&L!u)U+MBMSt{jmA!2#8I6WPp-(r`_ z!I@xNl$U6~^5gNKLJ#J&a(Gx$OxHHPNw^kva^8%=vpy0w$wT@*dOSpTH)>=zv zLFz6_elV_49t2v(u|N^?&Mo#zXDQ8`N{3lc%>{^S%!^NJv+v!so+oN^N$mXx_(U^Edb{Jbj&WZO8{=BGfn5 zGIlps{%jwXg>KZ7AQVJj6Rfu=F#PE_$=tLz0X!~%9aL~K+Mu~7Q_?k<`RyeMXyv8q zK12Xi4LqF|BXZR0;kHL$M|*pwoFB+sV`LK{>c>iGvC_eShArSQZ;eZBTv}bd&adA8 zM6N$#k-Uz2%C-~-rp~8FH2@HNpN4RGlv{WQ>Slg*OVrQGcX;d)EURGNo))jCV${B8wOAK`J<&@W`(`z#j_p{WfCp%W1iEuXbt&KebKRkFYn0RKsky7S1VYQXuNQd2g6 zhg%TY*OW)vdi0`qzL-C_Ut#Lzs-)!%g^i}sCi^=9mEH5#`HpD=i^9+yFNQS@BSeZg zCUxTMd;kd7Vy^E8rTlhvfB`Ug2ToK;v7_vo-0Hqlk9MqgjB>~<8fdIhiqsCUl1?LA zns#Wq4=0sM(d!erRYeG^wevOkfYbLN0gIU-4QFRin6-}EZw$-r8Py)TjMXVN;V+aC zi?B?u)nN0v0c@`mG00qBcAuw#?HrbxAmw<@--bCtOAp~m!Fd#(&{1OUb$z4yw|+xT zRuhukUGH0AzFr-23FqX+3qCZ`DNM!t^odI89c$C|od7FDWG}~E0Pw~28_Se@{|zu7 z`A$k9Xa}DgAPnZuoyZo!DEjTK6W)a@`ph*YE35MLnFazl8ip2E1`aZ0KDG#pmjU`3 z7GoDl*rhng>^0{LbM-r!yGUIauw?^&qm=8)#iG-WlUo;h2j|H%>{m-tz0id&LY)u$ zpL*zUGCbD$_$~J)2-O3Ln$BEnac`m^Uy*;Bt0u$7h5Hev)7;j_f4u#($M<5I?IvBs z-NapR7vD^HYIS~(3Fg0R9p~xa`v4g`k_uK{rO#%;$rnrPn@Sm z4or}TtXq6+JY_z<`69v5XtZFRy!C|)JtWz=iz!h0Y5`3Y4gkCGfFfRmm_d4Eo= zTcEwm+eK%X16ch~mOLNnwSV&WC~;f*M{7MjPVEQGdV8E{-@W)Y=VgXA9c}>}{nb6x zcbX(rpbJkln-fy zUeM(a+607_pJBNmp5=1!rPd{e5wVtVuy`D*8a;+L+fd$toOP@agU>j3qn2NH?syg3 z7ldRWk`D7P-;ZJbfTP`*7w+cI-2iB5R+Wa01$(hx0tIyXeTWZ}w7`q0x~~G2rsjgo z7~G93gx~)qt|YBCEsd9X`CDvL$kn(X!;%u|(r?atk;Af_b*95yFd!Yj6@3mfG6By^ zS9qP7B+I+0X~hn$RtF(LwSf(~2+AVBLU&EJ&E$;tf5o)>-?zN)oF9EAa?gNy+ZCVc z`n&bV0*Kq#GbW5A9E|L;dDU4DYN6e5d=(nNP`h`Sa@uqWndOnHnG_N9T?-+x=tT+| z_O8J~W;yXUUWXpkw>YD(skXaMJ0#ZRa}+u(%^|b6$|hW*h!3{^LjPVmt@KwVIn}41 z)pXw+?Ko0W$x&gRx2fkqe7j}kVag?Y^*y6|i^E8?%24LRm{4G=xXYiY^&9CULP`}L z9+CLJVhEAY7;4S5sLz%(J|yZq(vBO_NPj|mhv&@j1!$y_jZSpG+cbnEqo__15p%Sr zqx=?->;_hXM*xFiHvjR?CKsSk0%$b0R0e4F8OsuoR9gZNKju++lkD=Zp43UVlr~EM z<8|IazUc%9Shf3z2@c4jjOSt%-TP?j`LYv>NB;XEC7XE;yHwV3%Bzizoz{jqMJwqA zmu{ve>ogw|lX7L2hk72R*vLN@3|1O-^!-FaUuv5=;yzn$J}u;Eq*bn3HPQm{95750 zs#TjA6g*h`3X2t*zr{UbZ^^4h+t2wEX@H%nX^5@^_0R}1&~u!hHL04t{FQ|(-ZHx= zVt*uzn7UC71h|f5j{2!~qYBM47y1=}fmSG0{j-TlB~v4>cXn(bqoqQN?G-IB$MO2K z+T*d|ktGsW4e7SF;gNAgO^Lo_6rr@mZYh;rP8}-*l8!@e*`L@Zo$mMT?7}KSgb^75 zk}h8zeI{O06X)`Uya-LeR=|CKOfT!I)+pJeYPtI}UgXh*cCmX9;~}emERIl3gwmBO zYueOxz0Fo5yIcIQ5w-t^^kEI>P4|hYr&l>t`-p8#s>?UU4o#8`}NkQCngqxhV_6;CnEq z-~fAK3>1^@zF7VKWoxw|DHN-(PoCn3l~)!+rHD7)pOdqWIsrd{@J9Fq*XPOu<$cLN zfneNG5r2uK*{}<0(45~BfgE~DTY`>S*B{ycwE6=Q>ayI6=;8=u?qZYiGtPc6#KU9A z`Ipzm$OkmU$`HSSvX1tArI(dE^sd}W#}-Yldm$V}eK$jQIM3SfJI>gcwRuWb*Gp`Q zPX+yBS~Y=oi-R}`#?wN+FqH?7CVtf|lKH>TU$a`ab)?qb@&upg5GyDzEvkxxF#=u} zv}RZvmSDPx2;pH%R4h<_h2b}%MeirspKn<_fAYQ&M_IOd@f66h>3)-CFtrFk5v z#HcrHhMgl25|!IIEU-L?DYZm`Tj1e`hrWW2SiPWPn|2*2Y?JmbudVGj$OU)qjCnp}>8$)MQgH>AIXYe~q-kWo734zWrdwdLXT4%B|yd+I&S@P*{hi7v3E2+_OE_ zIfi>2V=WZvWO0nC4ivm$ybmR>r+sx4O3#HaM+5J#(UELG+>`iNJ`(foiy#4bTMi?L9;^3AZp5&1L z{8N4QEFhD-dWa9cE1ph0DVAXxjt+fr7#BcAHyZfV-E(*KEIfRQ?N9LIRq#ze%|k(S zG?r>6pB%vLJq30oGq1Pe0MVE#VDw2H$wmPfXmhA6?623e2_nDPB0c}*6={;2If;)e z3x7dh@N<)cL(^BTC|a3VTRUgp&{eTHGi|-4d8LnC@l_>Xw(}1O&DOXDr60T-TAeBj z+3d~cax9#Ni6xG0kV@?&BQrUxB64J?cXT;@MQn=~td7XDH)Kj`iyh8BGn<%DwV9(i zmYZKh&Y#@krZYA7WjKj`WcH<^iY{YGs?sQt+#1t&8`0JY4YzMXIxlwYPO>6jaH< z?Z5)0OOle=Q!?H;4HX#=N0pqu95c%;L*)9~t!?;s#2w%r9HnX!f%=rNzi8`ctW#1& zC0dj)zgTo*Qc~n<^2~TwEZI)#XDj-<)+XXCiC(T9s&`!mB5fxF(3&8?% zf7C@M93&g?L1=JeBOf!K_JlxiO??oVqD*iJZCSu{D1}k-*nH5x_Xn54$4SjvbmDLH zs0*8qvqWJ zw^aVj6`dm=2RKK+ypR{Pg3XJmTn7km0EWM@NGzZZleke_k=}mK8W$-d`|<>X9#8T7 zxpCEef-0`2>;o|Ux|}TxL8F%mZ%psauHm?7p2V)gkYdoNyNEip|(e&p7`sPb>eDzRuRa{he=vvwY}r30vaf$x+Hs9tIE7%spf`U&98DP+_N`4tRgg+2jbpY zNmptflc=e_*8K*9!0wE7cI$*sGOCM`B1cBRm-gBAu@^Jd^qLw_aFvAS6|r{avaaGV zVab9J~hxG#-$1uyDh(#|sMpLobKoZH$te51u-uqj1p8cfKRdk2rcE<>G?lSK z_l>WXMG(i{#7O$KeBxB_+qb+SiN=2~NZ-lC651V<_D>CqS}S-~j}KBJKW1k zTMJf>)iAsILFp9OUq_zE1A34cc*UmE`157lHstSOWNBb#?SWaT-EA*w{wpRHGOfxQm}ZV268Q?NC$}8U)otLRe`AO>PD_o~ zWVXhbt-{LMU)Of>n>iJ~4NFF3l|b?L?nzTcs-y&S`WmSIsBVphb(PnmP5vq3SHcZ~ zT~ogDK77j9aml#WYLvdVTf?BDhHNMc6EsJp5c0BUEyUT>rmKO+HsVMc#qv^$4tj#Nu3~f1U=1 zq2t1PxtO_LJo1C161xTvvG=Rpd&+^;bcYp}trxtIg)Ic&6;K^yotXz=x$<_TIr$*j z2D!7=l>V7vS{W?asD|CWToCGLl$xbo-n=bcQZ|2`nN+8HaDBK;XNG)3V)rxT`jFq^ zE75JBvRTP}{g;@|yBzcjM%r#x<~&<8Q>rs`?g0eOpHt@s8My&^Uj(5()Gr2?owkZb zBbKrqk1Dbp3+1JdnJV?8X{6xmC&6nT5Ls{!h#;oCd+e|B{b*<3>uME-g&hLk6rP)* zwgWfpX(ndJabdCklYm6z^zg+o8-SKo*mu_2`wf(Y#X2xrexvmVTPeQ#pvX(*YsowR z&9;fu+JR)j_k}B7BmK0=cP^JHftNL3r-IBvm6m8qlP=zKxwx&HC$pTu*ywE1w3}7l zlt#U%SyoooYng;~)*-QQa1ie-YybO~v{XUw%bl9SY&Y;QH9xsym)bdVI`f8ooNe~* zGh3D*i$RqxOI9o~Xn_(D%^t&8%AT~STe72A`(c3#by{ah1*kJt?*U_|FSf%|jGRV^ zmPF4?^xkJxFRxB3o%kyYcqJ#;16Sh6o~CA+h> zD!9B@+_%yAP~qnLFAo&8q}}}9VO{2hgIj5AU7JHcplB5|*=?Eo0}fZ7tG;^&Pl9e` z?k*BQioPy}r6rSe@<|g-kPhdV@x#a~l8`zyO4nfE44Hh~(hawxSXS=79&;8cqht7g zG@Vsgn{5|n+u{xdifalKr)Y3@cZwB<0KuIC#T|;fyL)iA;t~iDN^#esEfhNW{+T(; zNltQI@0;h@`@YxO^+J7o_e^PomvGsQ2&j3JD_{a~5j7~UhmF4+1HnG~Y5GFzUFEp^ zl(`Q-+!K`)6ZO!RBI}1w5Jz86RBD!?OO_4kECo_|kQ2UjJ2*W?Zw{7@;@)-;`^Qko zv;IImtD#dn-5$1t^XTSRblRAA@0%r-r&ABIxw3X{X=BI^hIKVmB?YABbrie!~^026QdRTEf|tMB)J&JAPMNz_pH(Hc}7yL=0YM z$_B=CEVzbF&w9)pP+2`DMQGzlhv}b*KLXz7Zt(%2ahIjH9&M9Kx{(RBtfRf^&Uxjj z)7VdJP)JMs`5@6@AuC(lL6h83sRzGw;MbT%G*G~V#Egl629F;0!jkoq{q@c>pZVa2 z0AA-Fv26cFMwRn=3_?OakQRQpj*i7<-h!&VxJnsV&fb0hCA)q5)r^7f*Mr)PIT)Yu z%?b{52uHfjVpq(t8`2GHP+J+pA;@5CS%Q>tvmJoPTZ+wECf#6Fj;ENqjwWUZabp`Z z<2_?`(KDhDE$l<=^#ic2Xog>>G-xTq2RMdAFtZ#Q02t!x#>*;=iP6RSSkVu$C9o}e zgyl_w3HLd;)1m|gXbwLEzFrkaNGCi)bXr8ZzX0%!i0o}r$u%qD3{TxWt_0P92Njk@ zO%SqGTSVX@k@ZbZr3xj&SjUtGSQuUS@`2}EqxZK|7GrO_9(b~UeqmUA@#9lat#o8S z{qy6D00Bf2Y?x*bh3GxHD0%nZ|Dsj>{PmT^pDSQCU%ox6iVaQ6;*1@AGXc}Onoy}I z4iYXO@-n~h-YE>FZ7IsaiZ*$z^n%Ym(J3;aJPCj_zTV-wC>BQwS&zF6jTcwQ3wB4; zhn#8z4#~|*kpyoX$YrA%F{%u|3n71u0$>q@BgIDe$S=`md!g(^1V~s}b%pbfyzos)c6)yu|>X&L{#dx;Z(6BYY?L-d}ukQk>>L-3v zZCme!B@jUcfdDO@3#^U1sBOhA%_e>&ee${0YD&zo(dpYe=T#$Re%wZw`jlPHqr)1! zx7E=%B%K2*B;q@lR@NdcE{(eS`%%S@$)AQ9P;FLz&y|>&jWa$tzZcXk2l6FAp8a!D zY1w_HK{rgr>Io+eIr1Y*#ht{y>^PG?FYoc zmt-a@uRe=0z{fFFR%P8L%kJz;)ox@h(c_aqpV^NpI)o{CZ2`at=@8vhZi> z8oYiHB|`C2YdCNMWNA-L@-DmHO2^P%IK#87vbGIkW3--K-a{$O<{Qq!IgsCuQ?BPT zg59Ch($=1{ZgXpXn({M|7RFw%Rues@T3Yj0!H|YsxG_R#6ZU~~>tr4UQOEW&B2)x&9l zIY$}~N6VA0jWBK^XnFc^=QzxMAvMchpIZ}&dvjrc|FYV9Y#h#do%Pn;)^20SH(SiS z3dUww6+=ncTz)CNYFqcT-Ufuh2eqD_LvLTQdVh41#nq~`{kfm+JXmu-Qm1!uTPPlg z=11hzo-IJOk=O-ejpmC*kxKGZ#C9T&twNrd_71ydyN&*asQ)!f-Rp74xJ%^#llRC< zB(D-0vV~T@N^nu?2)or6-MM^dewpt?1@5DN`?0QeMU}=ZaTt#PK%pe+Em!`X@n}6x zxTQVaf6(w2@u+{jKE!`MD+b0SUraPxm;O>}2`w=NP_dG|)wR*rD{pQg)HNm3)JtFc z*J;yZo5M0`TKY$M-Tr=O*02XLQ&%D^eo2W^Uf0a}R;;9dTALHY3ja`JQmiJvL46y* znwB~pM{L+MjSP!JUiqawVp)Dnq2*Z$uS&Nc3fAWM&`931^SgL>C9oc;!U(eVB8Xku zq8p(J+>0ACDEuY%H%F7=*lIcIg7Jb>#(#lJDHpI{c!Y*Y6}!!P$^>WPUyOviI3mS+ z&i@cPZyPdHb1Le~6odA)_li!V0s)c!%%^YUQfPI2aT_`MA~#1_!#0OE9UrzMQ_^EJ z86cVvylAzO8_Gk9sxR7FL)ou-ntxd2z#T`Jt~qLgYV%^hQw3 z&I0%t5~`+u-UEss2q@do1+~1)Duqin!~hofyBqN#yUSiW9dJL3Cv?c5G+$#L)we3~ zwoL(6d=j)Dp#IIUAV+d+eT@y|_1hY7g*vZhZ^MkOg=2&30vY++_5M=Z_d+d{@@0G( z#pCyv_ckO7RiC!c<>0QQ&QCnj0zZisKL0bWH2u`l8q_ei+zxdUuy1W;`_vg;*T&pX z!oTmeG?tqHPu|q@WjMc^ap%5UdAibAxp`fMN^@-B-m!P&Vc08F+51&HpCI-}Mn~d& zmdx6Ruy^#a{)QctI32bZ9Ab-2=D*LKgK7|a-iIy(rZQK%vTr~$+!E*eoOgtvn~3z!LVusyS!EYhQ;SYII!~CUr%R9^0^kUdH@Uw~B+&-fKj)x<}t!7-%#cUFH2J8eaSot6q&3~G`?9Mkp=D}jmWHN}6mh=~{s zl|386tv|6idNxARJoMd^LJ}S$miQb(>Hn0YU}k`}ch~~>yE6_r>?@M5d2^T)F&MMy zG4AdA8G>7%Fv`_bZWyu!LG4B#zYEqLcP;bIIP&x>g5|#A8%7&7c4NsqkM=a4ukx}9 z+r(uZgU{Mm(Mrd2#MQGc--Z+UE}!?!&(d8bbxD`LRJ4$R$K`=P(WQoiza#e*9B;eH z%3AZ!y%mW{?qm>)dx;>9p<&*$rK%fTt`DrMZkTF!{N~jxz`9? z`Q>`2C+~aHm2+z2xgR%20E+>vS-v;OdT&V_4HLzLl<3AMIUmE13U;VRqS`hB_Wb{M z4`1CNaJ1u+GeU$xKPQS@M{#s)g&Z+?s#g|x@g+NWM^O>W+YmQ*hLtR%&;LDE4?1Zj zdUQls?f&s3vr|%yAO6?aR0D2@9We#>$pDJgYbZp=1IK*F4|Ox}Tkny!=S_njs*1HJmL-D5)1jY>myu!&_JKXezS33)0m{dU#L z3A2vbQ`6rs&&qD8tq?0%*XJSe)cj%oJNTqBC!6ttzkF`C#tzAaR~9s!Yhy||4==8X+_FpWqF|BMRgXA1Mqs6NHav*%?h zbnLq#c5}Nvhd(Z1r>K#Oefiw>ks#_D_KLKv0IpQndsM=yQK1tl>6kyiDVQu1^BMOV zD{$mW*?CPI=Je}6)3ANExM?xY>V5MR9x&@oZPv9@=$dx&qe&xrBAjK0b>T#pm{E?k z%}X_WQA;^Z#eQ;xTVI-kuhXC^t1b;*RG_s8O zq$O9@si5~nfd^=a1D;vyeJ;lklB1O>V*X6e&YVB}ofVlff&Wm$(y|2SV?^c@oGGdh zi^}7R{4c$yqW@<8*Br&O2bR7}FLv90z+jivBsT`X>ybuR<^1l12yC)y$pE+#1tT0+ zpe9Y^Mn*snZN2A0^@)m6CfaY~l=ca<{GMx*_(EnotSMmh(xqg6)qCNfyy^3KTDG&y zam1_`B`C`LguH731fg!Hym7}KvNUH>_w57k+Ud568=txhox(peoGb*TRm(rMoou*< z8&(|J8wDB>r(0Dv@~V!)FqIc|bh4e;RT-?A>jC>O{$+j7eO(b0n4BYCxD^Bs__?5| zv*eLaa{&w0=G9&Kn*y1n05>1q%FTWfL`<*q77i*v9 zuZsL$^2OS`#6ZWdHy^=hAX>rIR-iYgMwD*PkI8BX8KWZq(<=9lV?_3n%gHu7Q3jij z?EMFTa9lC4fQsvxsh*A2!3>YIc|_Z&VI!NedxpnS^D@}c$R*7D^Qyyp-82}>TOQpk zn6xbMr%C&qO6c6T$Ga*GE$&nHm1_n9)hBvQ9^0L*Xy8VA_B+=uCiFt=#a|`L=;@T) z?<xv0&ajnusOLHaAdO?V-(3CHtEXvpu2RGX?S z?hGn#+}tqvEXUYLfPy|QK~S&YP>#Pic1?ne&V8BGs}^U6TIQ0_y?42^L?T zbgKh|bMUJ)E+9(mTqd@QnmPQsENKH1=#nXtTUIhq2@2w$JM zI8@v$wMbbcCNR^R>{2`J!YPAv-ZaEqg#2bJ)jc@nj+3_U9dHfi0Y4OnY|3wyg4&*) zQsCL;*mv(6)gd#G(EA`pOeU1z6u>`={)Gc)?0Gj13IxQe2j0hEmb_q7){}^;2HDue zerkd^Lp`U!1RGBoQl6>>RnL9_^?`XA-EYYdP4>z*{zESp#41kBY38zU=-Nx@G#A#j z%SqXPBsl-o$E4lgEIT4NrA7~hJn@C@LUlZg8hBxs*QC~1J(%6}I(r!;6SZ-H_wFyxXM2gG=gkV@=Nl!&9=M?B;*Jx03|A#BVF@D1JqMV**fp~JixzHwb!y-W) zNBxM7E)I23YTuPthCGP+(;=(6TY)v*tV(nmb;JLZP0n!WNt!k#M-}tF&M17#z?jOa zDO+A+$Xv9pspPo4Jez_{(w67wXM~&S+aa7KM-V5Klc4~$a@?o7u@m_E}=?DMsm-&hDX5d&9D@G0R9?8zb^LD0E0--(8K zs?_pS*Hb!fA2m_QzLPq8;HXb-DbPt!0W@DB_I&N`;(nY7!BG;6UucJbG?v>vKQyiA zM}%IP+v4({bi35|BPLKndE0;d{|gwMkpH(!@!U77`M|Ch=#`J@hN<(7V)ZZ_ahd#} zrfn|8X?@tP71{fi_e7$qE&W5XuxeV4E$BX(`tggY_>AGwr!rz>HAYrGQN=0{;#FZU z6jZ^SG%(?X0f`SbArR_to2mWT5v<1XG<=5dfmA?{I?8$JB^&edSPt}R8_d(tH}BV*}5^lr8wFV8s})chRyhAFaBg5j+0N&wN`l1G1fwK3LQ z%VW$5wL@0pj`#S{`;?jY?wo?b#Vz=_nYjS%5jT@Yb^LljOa&4!0iiKlbo3M8m z_Qd+R%&T^#o>>5_Uv+oNlA3Q=yf@OQS<|ghl?6^)bZtzY!FS9f)53L`Q?)5)nFoX*ZU2|ztV|_HsNmg7HsY9dWgXk_l&v%OtOP)4n3TGHeE|Z)C+}X z28zbK-u94}Us?5cJUcU~pfJTpf1N*utdQz@0F85QgpW!3RhGnQICL))YL?oeTGrN9 ztQEA%C=6sQEQ;O@$ZCg#3ZS|UUfoTg-s3uw*UQ=*&fI3i&}moR@N+~Ul^kM(BLaNx zY@hF7x)(J zsgofC?N8dJRK@8{*l#fI${ z7Qhf_W%$6oW8#U`+L4jOw_Up2@&E-wKKn#nk5Wyptn^4RAD@BplCXY59&hgUBfVML=45@v+pH=xd^l_wo~l zr=3^_J?#oy<8wr79yal0NJ^F!f@D66QkhNT<-)i5OfnFQyT&l7&_m3S;HciN_ppUF z9(ic}qRv=M?EGL`OKJ-GXu>T%Y3V<7hyElfq(j0UgkScv@+kq<8vy10JwiL{NJB@$ zdhZ4Kq8jut#o1pARTBnR47qwEg1!Im*ya_Svg%T1!;;>aq`bC1*$UPUcH?)_yf3HR z#|L&Uzd()!N6*s=S`#r@g&gs6&we=NBu}PiiuNk}w-wAS+i113_LogwOifR!zMV!y z(EEf18u$UVs$*Ko}VD404kmcG!x7hCG?Je0q*&fr3NeG zzg_l;E2Q9?v`P@=uffW->SaR$3PNXWROCy_6(|5 zH?LfO!!SNE(`5z7UATK-LjRw3QR^Oko6?kbwv$nyaGL%qxA4IGv^5LKjeQs+Aq8T` z{05kIB)^*m&2~stM*GZfEmPSr);4{3yW#S0v}lgCE5T%(FYaL1)Hvh6P4iiFC_6G$ zbpC&$8QFJ9UG#eHqxPNEVQMv01T{_eC$zE4!A-&0d<%v6LPyo8_7PK;ldgK&XhvNo zBb!e@AB^>~y9&iTOl}o-f3;Hj;4?AmW#ZTTaLTlFY%t=Si6vh_0F$e1^OMA~th^f5 z!28xVtNWa~L+NvAP9fF2K{e4F*nQk>PAP>>y@sHzs9AE~->;|wCw~eZPvbrww5ND3 zB@GVeM0Sfl=DB91=Wl6=d!10OndZcStKnQkUi0$%RdmAZktz*cbAmKSKI3NP4awgb z^EsPl`L;yWv|I8#PO28OY~Qle>vLZI-Yo;1cUbyB8xV(mF3xeGk4eNbI-o;IHPaA& zXVv+1Fot3OM@iO|_x8W8-_ICjmXo=!AKhe~-%jii^*))7$W&pB^OcLTsA@t!Etev&pXy!SfZ>Vkf}pN+{oJy<30vmwnNKViK1#F_Ln)N_#xgH$%`~z^RyC$J zpSYFkLRH7Btn?T`in&lsx)EF%nebI!N`HoRBIb3QVs3ei()B`x4hM?5weT>)TC+A2 zgL{DlhylKNR8dG8Ob_K|K_Tdq*;JDw-KhZy+ofDF9lxc-mnN#X9jui^1L;;x4I!%7 zwLpq~s34kUAkq+Pi#k_V!^4`5bphoP*3p+ts@spjJhQ|V7gV+Kb&lTefk+}92}naD zJl!ETG1i^4qIYH}6KLrPejzt) zJ~5-JrhX*aSbN=6upo%f?SFrEBz!*K2@%%FCEw|lXhR^h1GhK_2DJ!;aAWjqte$yi z3V@oB+mi;!{XJq~d|LhXm6xpYn@D|M z-w{&>Ta|El+?Ibx!5dZL=lOw4)i{)Z6a=HxuEiPt@%OBE>1Ck4(D-yfEBW|d$8Lya zaEj9tEOT0-xUK2Vi)-zhVr0+SxgtNiMFrL~vfvs)oj_HKm7V8q<=kE zbKJ;}RG0M;9^;2vF1qO0%k1%X70>;X+!0$N%^9WkM+@X}LNZa~lc#Vkh-ix|KS7uR zd<=(1`E)Y_)(fZ$oDqBCb{X0Bb(zYEpl_L|{`bN2ZudmZ{S;um=TbI-dm5)+2`84TxD-7X%Kh< ziPJmYroI)QG$wB?FSTOfddQm$c#yIw8hFN`U;B{Ug(H%IMI;`{5YJZV`8RNB$TC=O zS1gzZ$Cly&=W)AoUmXX0@tU`J(xtyL+ z=cW90SJpeb6&SB1@2PwdMnm1bipuq zY3bF{986rOa)gHRD<;x)$zBLHVG6Lt#jgqbl(r1ehU;^=_;o}GV}{&DkkVYn719oN za$cWS5d9h1fX+hR_VWk7jAT?&AM(f8EBpAuwo98lkWDxsJx`;>Ix3JCh)25IinU&9 z*eK9}=R)m!g61eM7uMz93A*L%LqA~+w8^HdAP-7mL8eH|ee!4wcd>V;a?j&N9MH(U za!5>OGI(Zz(wF`J`Y5tOQqB=ZgP6ePL+o8i{BQzJ;o@qp;-ru!mZsCyv<{3>LS_nS z3l#Y5i0+wf6G+~R@yb5*zp_bwSHvL@!M2?EjG~5!T<t1+T zFd$j%8j^{QFIut?6-}iDdSa&yM7JqMN6WanwH4X(EW4^tTFBw|?-!w?)j*;9R4<{s zTPrF*J2SxhZC}mg0N3h8BZit`(`XxI8N@bYWX<2>Y9~%w+)Vndm8|8H#a&p1P2-&1 z(Feoj0McBf!9JUU?q_G5aSBN%ksX(5^Qn^cAr(}_T$6t_WF#AZBoTniL^UgP_+E5` zH*bPoOFKC&Z2yyUG2u;gwTg99o@ zw3w>X`%hAakWuLrrdWm;MCo;`*b!S%7we8jI*h?f`22TW4YgdF^>lNqRhcVT+V7ZI zKtZ|3#80pHuyF?eL^#n|q8qXd^RsOPI#wTNS_u>6xhH^Y8-sk1&-o}P_8a@FH1&EE zF-9p;^*Uvv8IT3XP{p(HM}s3RG&`EquHLYRzD`UzB{^`uGU?y``2)6M&Pv2a^?R5WOu(Vpy}&0h0D z3AncwIdsA0GX8RKmO)(L%CUZ16QKB>9IW3c{D~hncv>Y+nm5@dRY3XC68nemv*%(X znqfoNfMWf^FfLZTHTQS?AUQ6ES^f*GA54FHS6i6AZEqoD0I>hypv}~56#xr7$=L9_ z#r%N#PY;GvaY~k{`i{cUk-Di)arkx-yu>)2rxj7-(6~a&v@g?B`mM0NIm4XQ>LvGo z919MdPSA?WhQ*M%LkH+03d@{P&2$|1!xj4n-jdj;4?M<%LT5Ez&vHjwgbxdy890j* zn`G4hHL^-1yz?r_;q@B-IVL)Z*x&(6Up>On1CZjXK7W2B+q1)-df{P^3QpEy=%U-75 zd%uu;e=$jQ5wfPp)&QpnkS0bolv=6ovF5#HCd6L6j*Zkh#Hoj$iK=`ECSP4V{1Ngi zDg~()Mca7fxuWT7Zh$j|E?^hDn z3F$Thh9+qRnB`5N3dC@n;W`Jb*s&Ss7`No7BUSOZG!w8Zur~(%-KAMjzmFTl{*<6) z(%by4dMN5$f&|v6t~;`;%p-pRKcb+%L_DjXUQj%flhS4b9zpproo7;%1y$w+yY(@u ze_tL+#JpXQC!RH>b!qNdj0?S%Vo0~Gkse3Bqgpk5I$JRvmh)P=TA*tCx2i_&>p~JQ z)Dt;Iy&$`b=TP?1lF(#;k{<;kiyoZfu{5rBGeWa^u^I#%v*06hIa8pDNTk4mQC z^PXoB#f0BWW$O!oG`tdv8)p%P8-AKnS=Mg* z{TSp!7nj$TX~2xGoCAzEOjCx0yq*0@Jonlt-Y{-B@K118l!)Ga>2k_ZvUE| z<)4%4N5zXeh%1r&Es11I{Tg0zlGA1yRn(eqT;(uEy3HfUjMuBd7&}f%3%1=5=1&!5 z+sF4xkR$^pngF)Aohfn)H1o{fY_Hw6Ixy$DI{W!OvaVbxlL?McD~Je@TZ5-BB_W$O zvt=pdO3(Bqf~8`e)WKv&Jk3)}V?^5>B{KnAR?m{YxU;U_oCT}_f2<476@)&zTHs4e z9P!mNg|Jy7F?BL3{`=lG#W&R!gZh4JW4B5<@V12J7{M7@+?z--8912L;(`&KgdrVY zS|Txeqkx)UV4xuTMFr1<-=2kEX37)K^%9k%yit@bX$__}1?!mCB?fX@JTcBTNTafbdTzfkd zyBM_j?dQL*e(x7sK}%g2FtWX7bFq$#{Yk*dm%%zzMAwTQ7T~zkY zHtH*e-<1N=)L*@>hH6I zRWLb(O-7Ehk8}a5o_?Hm*uwWKZ#nzs_)DEpqlWA{jFqo*dd;TO7ZGd3Dav>Y$J2d! z8TLGycslw^g9`(ZT-21zlGZNB-!6(hd1}o4WS&*~XI~myQ5MXTAWvr)A4hb=@rc^U z(Ee0)mIg{i|9#&cy5KA_iIGbw%-<+K@iHnBmnz-RG@EK*Vv`Nt^#jeYfF`6skhN~@ zicGa9VCBERlQq$Pk6Fl~kC}P9>Szg(Ll#g2yKa)S1ru^(livGPx_Va&p8biqTsSP`u|YuVwC>OY5l6cSMaz<`gVkKj$m?${O#< z6gH~RI&0J8P07?IK7X(-Fdx@6HSBc4M(LPoPl@+)MW!PBChSmR*-TV@BN*ntje>YY zD*sW$84rJqpWUTSnBDa&y$oF_)eSZ19P_f$Ux98%!7_dCZ}`iJwB z(SVSmJ#?{FGI=9(dfgA32d{~c62uE_C~nT9R>Ij=CUB%yN0N#Zz~!qKE;fnMKuD5E)zq90%*u@!h}iV~wy~}C*z_Y) ziBIzQjGpc@i;|E??DiT}CGGn)>stREv7HA=6hCI;d`E%{@_(R=BRJI}^r@ND_BQ(g z*m6lAAXAp1noHoy!$7rKwwszCjIfJURaP9aUd?<53kHZP)xM6op^vni^jvH%^Qz0# z$dR-cC3G(^+Vwm+lzc=f9#ha&{P~9q9ONq-Le_O`*!o^J_MJQ;Nmr~_r?%8}r--mz zDjKyjunKT0hKCb8m>8Eheaf|#p?KRCjZnGwgCGFAdM>^hge3X|MB6CPXoNp4eEPuH zOFO=Aw266g>+6MmDDw>>&=Iuz-NgP7K!pZe?nvC-DkC5r^O&tL=rF~fl%a}TdDTFP zH2bgdLJ|0ONj}S83jghYL9kiT2jdL&ehy5^O+;koB2DD>KV-gtKkEla)mHkE@;?%L zVGBV;zpHR%v<hIlDsz!$t6*R-a?68wu*e+d7s9qy7`#It~Zi5P0EkWOH zJRtRFQ!F>!>VqJ|f|#;Kaa-6(6g0@yD#DNabwbc8kfxE&5{X(mGvS#Ac4fm$DQRMk z@mg+(;7Aq+NDZAP>89qTofSwQ{1rMA!>GcarQ<}-!YPiPZM8HDZwqEWo3%VNQPK$1 zEKX-rUSdiiyn}T|`#sOR@BLfwzV}us`Q=G2`R5burveovkFJEkw{qxdsaGW`K@2vP z+?{z!ghs3%p~!%ec?E$ib1B}GZE=WsT@~KS#6cTQOz(%}MQ|X!$jp5COPliZTo1A) z+>PVgBKc`CO*nxAq6uep*sLwhPFjOD9KncfH0^y!M~QdB8435Y<-+sqscgIOlsPY? zdCf!ydODoOcqEzk)J0;#c*6)JHBtCS{Y-KjU3i1;CAy!w1w<%g6OforuXum?cwzpJ zZm5p%e=H%Ip_Ttds~6P0Xx{KObYBQPP@6=ifNc{#iV6SZK()^+BH`@ZOFm+Hy-B+W zoRr2BB?@@(w4m|;Vp|HvQvsq~ zF2O6CNI|ltQWMT)Fb>t#28M;LxH`t_yUaEQHS~+ERur``>iGb^F*m>DY$BJGuQssf z`1C;t>B!gxz-);Qt$X`1&Lr~HSW|>M zyaM^xQrLgJa`8VC>IgQ7x1jBM{#PIdRP?{1*8G1|_lH1vTA-TXK2i5OmJK(iTo0Rw1Hmln=NFuYcq!=NfIhc^ac`;u?t?`jLC5%EAe<-i6Z)77i}!1T%G@IiGt24 z|8K|lH3d;5`P`TMXT%m+?X_uD-9aUrhSY*#*}EBS!(;L`MT4?>vKGF|n+$mS+?#U& zpV0fw1{HujgTqp$RXi2R!d9^4K$h};ft4H=?yF`lnd;-BJ0E)t+}Cnzg;nTpN*pCJ ztG^WHg^HyA5rJWRWgO0bE|B=>{~jubjYo-Xe*79nAskcPY5=!j9Aat7AjwpX$Iba$ zbvqGOsG3dsSJh9HTrFXBU?L_a%I#*c)k3#fuP(;!94-T1&8(a;N^WJ(DV!i8CVF4f z0UTt1xEoMh$Cpd%goM7DJWpu;{p$;Ul;5nNBz!Kf?28P^?srLTLaXC7!6V*B{EncK zU$9z?M)>`-@wEW-C2Y_4aOdgFk7~+0-($eQ@_U5b;k=6Zvinl(urQVZ zX0D&dXf29hPyFygqiWl?h+b~B(>+MsM<9v3puvVax>O#B77WabO7WH5+4Yh9LH}`i z4rH!{Cp#vT-NK^VukE~wAcaxY&D0^n-Azi1w;(0gyx2pf^pCNQr`nN5EEv+Ahd))Z zhSWO}r!qn~cJ^XZNX`3G?f#3l4Mj!emm8HC)65zLIXoOvex>8{u54Y|7d*GqEYGyB z;rj(k@rtIV`TvR2J&zYhu~}LwMf{Ny*mK33vQaJ5(Ld4$`_75=H=iy$+|ai}qw4o9E^o*5FDMi@`GCZsDcRyYYJey~_l zSVW9H`S9ef(L|wK_wSPK2weF$;^5B`X^bDq)AE)l7&lpjm#Y_CV3%Bkjq-El3iDn} z6wMDYTF`{M%!UJ?QIoQ%3z^hHY-@?ULNve0a$}pMGTk$tX@rDx-nyShayAnl7lz_; zC)u$qKr7Am00s&aiZSDGYC@UeO3jkiaLNkQ;$6hdP6w};taxv4uVZJ`e2I4FjqAJ9 ztv9eyzPR7pY3G1WO+A96q( z*y4t;Rz7?BhUsPwg`FdvZBdoYT`bpRj3v@srq^I0#69Q5^bakO+RlEKCuH~kLE3AD zFFH!phfohHGC)@969s@}2+^5b^dt!gpvvy)ChM2I2{nXdnS^@}QvLggHKOk7)dr%hv- zV%fO=EhQQKyqQSy@HqFWl_)UA&W>HosMoJ1-_Jf_bK%u>F(r1A-0yinmRYQ5K+XUIs$Dt`4&}XsJ0zV01qr=pgr@- zJ};>PHj%T$(qdeP5CeDjaxb@dCqF_-bX(#17qL1JaMA%H4xh>|%IEX1WLjxN0;E

      a!sxeX`R92LKl38@g~F{zaOn>ihm5t zlYKt-^09iQP17kE%JST zS0Y`sk{?ljvGNR-gKuJwMo$KcJaNt8!U$+I-!&Jqt$F542@8T(68O-4z5_xeRl?Yg zl@X$!{G(LSVFb79zoat`o)?lA5&x?MUehKD#puZMukpdEh|I}~)8O~~P&{s<_s-?) z_{eIr#%4Sfp3tAe_xd)%JO(@n5Ip}d-2`?-n95W7LFr94wdu+-`!b4}g}-~9nfr~{ zl0{KE)@RUkWanTTkqy(QPHyzvAI)Xmpa!Q|)Nwt$P^RGr-rHBMVx_$!jJCp=5z^tO z8DNQi-DMA;l24s8;??)f+Ik{+Kyu>4r$Fh-Q*?amR-(a_TdK=}9uPkH`rx~KdCJco zG*9sXZ&g#bwt9w*bH&A7TEh+V+>-+$-QS7IORxXp0peNSiJoqK+39hZem0@#sgHF& zL45ZbQMaE5HUIb9JRXYw10A}at69CF!zi`NvAlba`|%_vDI`hLVp$MU9TGuDfx*`= zEjrP=m$Uh|9BR{Th*q*``sLSX16ZC^q6_d$8j@h`dus3o4d}s-4dQqDQHpSGRe4`Zo-9}vo+1yfjve9RTsG3_ zkhaNyimnXhD3cdG52gncQHmr&SM}@Ng^7(PmF3DJ_%S1d7iK)BX@6T#@hoF>pJ02! zXae=wme(ZPTI0#%_owuaVmZql_P!&Xlb#zy>29FuKA=6AhY^e?2gCs}69MDo1hOM+ z;{cJDs1g_Idy4qo??*gZne?^4b8o)-3nrbg$t$Uz3Ac3TUfJ6}eep)*xWs&q%QB3S z=`BJROrjd$QbKVEy6oawM3WlJ;zxSN@B00$=5L|~g)e)c7$q*Ex+s&g=%YAO+BCSU zsOjSob=i8QENo9(p?FUJE(|osd?_zVs03FVW7k*16L~HK_U7x55C}%?6(5t8t9g)e zukK<}gAomIt&vV^y51p8uP|;8JgLH|Z_^F2-(*I(kr?gq0o$?Ducl24o9nQ6ohMjQ zczb#6`b%aMjzMG6)88T)@#NO(lTe>P* z=A@aZ8k%s>sDe{URYvsCTHj>jq7kofwNF-eu`SsE;q!051f3x?8w9AQMFcJ*qkR#V zr-UBXM`qHo-x7FC?J5TpZ%Gd^#zXskai|=4XK6gd$H~S4Rey#DZqK|n50&qHuZeneprrK8J#F1MNC@7%3cHC&TFv{}*jjBO%z9Iq!~1PD+(qUNpl zaJ5xllHR3SibCg;`#t@el>dM_A05OgK5wAx&q;op3>EZUjpwxx0?xq9*DPB_R;6?V z)HwmdYCA|z7myVcU6UA9{^~GSfyf2CRBl`vw?&K}GtXP~&k+z!gkYpAP&ablp*HAq z_5+E(B}}KCur!b)yN&#_)Dl+HZ(C@2{Kboa%a$2WJ(|$RFw#N%M6k;6>exU+8&B5k z2c3FXY!~bYu1R*HknK&<>!T=TPMlm$x;;0hL}804CNee12{KuM`I7cDdC>^31ntf$ zT~y@xwilHgD5Al9_fq2nKXC<~l;g<~XX|DlV7bk~+dFP(lXU3$U9ilq=uP7$fi|VU z&&c@xi2(H_hs^rW0EKbV@!)`oK69#`h&MuxtBq&>io6_f^E@ai`eMgEMr~}WT8m#| zm`06q7JRvmc5klZ2}L&dKn-UXZHxGzLg^>t!40qH`2TsC=OLc{lNM2am$x}{Yj_iB z_A4y%qUgDfcV%5(%(~;jC8=N}=6wGxiWs+b zGHaVx89ACNc^#!x9>~|ezuxiqeY|qqydt+nq%{8F0|m{8-rt>(A*)$zM$V>Q@qJ0O zDGZ8$j~A6o-K?Fv`)|Ep`IT_y;}}T!1$flu^=vPuH46O~${~$bNYSysD21ka_C!_J z->IdK3(@d*6J9PQ%_Qf@syEt)oaFdtUZ25WSIXlh`SuMhv0EjpAcv@Rg0u-c-Dfp(2EX34&?CW#Go&v&mWLoy@H@EbPV48AYbHT>H*WV{OS$-U zH9XTeG>R3O+%P9(w4B7a=(x!;^veS;s`b3fiB4Pak@O%Lp^(4J8#;7?z$oG3B1OrG zfN(Ps>}L)1H$i|mA38*4IB2xy7CZihfhSW7(Rqpqn!bL(kj&kpl~Eb{+44K*a`?3Z zASSUti{wxn{dDK&xZlh>?*}3BOAvZ)W0F)7Say2E38{bOHvK97i$FsCm4Hf4dhwr| z^)-iPu|MxS+MAtStirdZt-4py*hpR?8h;rX2=oV^o9T!>3I0Inl*V&D=%5a4iR_X_ zj3?ZFv>Zc*0{f;3Sz~eMJ;vfd2s}xP|b8uElGS<+Z%K}i}om3(np0qAFAMHJx&Ud`ffXr6H z;jHfQ$x0ckKYRzJzL{Y$=4^k1DW+i@mmhIaW-G49g_FgynsN=FRoG-c8^e2c;Y@1( z0?cnzdMCYZc_5CUTObdk4_8zTl6FIPcq(9igdvRH1?9+8`O(G|N?)LhICuzpkOHJt}<03L&x8KSBhW((r3e=Nkx_ zlrH-@d+%WSPl?UlXZ?o-c(>MPLO?k?ZP;J{#RdsAXJ5u{laYK0J94Al3_|ExHa)axYVFNlpA7}XSef}3LV4O=Gj z#r`V}okr2^5H$QoK%viy^mk%9nf@>-wq4 z!)bG_(=-~5>y#?01NPk8l(Q>R6(DufYzbiy#^y7O{I&woQF$fsJ%m^X@uvh#MPw(_ zIEI=cMSl@m)D1I6pDQeiIq@s=lCVy&)qr`xhuL?l2}O)P{Eqp#;frS!s*R!!LD_th zx40$KF=F4@@;!2x#)dYKfc2!^t@Eq!jaj#Pn&MVcr43@NSwe3jBBGAdsM*Ery?}bm zL*${Dh3g_7Tyr^0g_F#?V?uOcN*Cz}WZ7>ua;$C%O}N<4=MNbEiNle15rEmuND%P) zcRl`vsLauJ9M&jOgNpnYI+owqNx@u}6l?WY>4SfP2QGcCmr?_+|Aq*Ng}zJKlD)xu z9g2@ZrQj%jIvN~~FZE$Db4)Y|jFsnvP7A?U2+_cKKnliiM4~6MH8`5Ld=Ye?1XHlI z`ML||a4rj88{}^e_7Y2>QEiSo+foqd4~U!z-&-C5_kk|Cs||>{=k8FS?ItZmfg+Ta z35*WK*JC4$afDI$QyOvSoE18)HIJ@g9iQ2$8%y3Ea|vf1u7=CT^PWWv;|jAqR@r(Z z0&|?k2pSmpVS7vijjI|xCr6V2jFDJx`wp&}w<$3b+$0m?lTC#MD84gn`N_Bp9Vk-k z$f~Wi%zpW6{d(j2(h=kg-0tI3p_1BxxIuZx!)N8md~ifXo_Fa#b=wa0+lQikmmYTi zA8^ayd*v2iAu98y?_O?{+X#cJwuXk?Lw_2A(veV3CD2j8)#Q`(wtZ1 z{99zn19&o||K^tu%}YG>(4Qf0x+BWRuP#^fcrxMyhOEL*tl+UjI~ofwgZ#S)F1!cQ zTT*zuZ)QP6c}aMXyYo)>aFYI4q-g8!-vYUNP~0;47__h=X`&k_YW_;W?<%d_Oth$>s~8y{=I{_zLe;Xm5&5Z=bVPPGiQ2d z5_dA*BhkD1f7_Vw*p@$|th5%tkeQqFYU9d0cM}=6$>GBwv$sc{0$03gEkWELqoXV+ zw%&#kgc!XsN8eDsduaC??l{ZpKehr5fzh>P&x*dqx4V%*F?9E+2g~bv4X&Fg8%3BC zEKTk0`4OOomghI$j2yzzjG10l2UJE82Vx>koX-g-y0@`~3U}W2Ij_D^MP`T52HYsn zgn`PCNL!?HlzIAeolA(8wYFr5E~CKA(y%EDXDN%7W0g4Ca6;N2yV`NFh8-EcY~8*Wm9?4vIl2-J%HN-7 z3n7H!;6#7vYx;%T#kfKSkNYnm-3`Ldk*?&gct<%GaKOZ7j%G}R0;#aTM#LA|kQhG{ z>Zm`a2gMw_IPnpojLL6(pvqTAoi z&_B@3L@`Gq>f~O;8t&)1m6gWtjNN}!J*sj+HL5_IUEMiHu=?r9Z;t+gEc@6?2}i^4 z!!y%kEE{Q#mTEG%G_6+~W8eO~D7|V%F;ME!oi%Wa2BcQ_nJr+tC%vfRxW#d->#i%T zy8=}%2m35EQ`(Bl6W5a@)h_*cEgkiHMaI`9d={G1oNL@3@@WN-2-HYL{rKT3`2%t! z)dmU+^iPsM5fD06YcVnw&3*a(QeYAY44phQPI(966Afy^pWgX%Q(D26pA0e#$8q;0 zp2Zzt#a#%^EhnDj{C>h2qc88UFAPdQ*)1p;-|zan`+kaSz6c;FWf`}$Mlb9Y3VR?L zBKC&Nn%*nB>e9MqT(@V-8oMwa_U@0=cIX#1R6~5BCU1;Pt;KNX=Q+)vVl7ViDKN%zNeo8Sy$nToEXY!r`>#Wr;@%ff0^I1(tWaj_%ii4eImvGS&UEvj^T%8 z#LbTwS>`t`8hwvdPXuost3-s+W6(xZD#p;*diC={cb#hNMgfx=Hw4w=1#2~|uE zkou5=t2zpc`?vf`>p4Yje0si5i2Ao~W4B9CTom~&Gz(cg3{m!RWtErf@NI*{ZMheMq%cT*F*@9HYcBynefl4evFH(t_&=>bdmLZkM$i#C@|I zQ^qVT_kFA$sv<>=Gbpda{)|l; z-!qGv7!U3l-%*R6$eY+xSBPe{gaoqL>tqN^Q&NA&Mv_)FJzM9HTBO3pMp`s8m3)0f zbV($wXP>4xsdtn?*SjX^DQ-HB7D7$%jvSL^H$i=cRXKdXSKcO)5LM3Y_(MeGLHdg* zN>UuAGURH7btLKoxjk^@#b576HD6rN`KUwXF&k-N%QwXgR*g?c%;R?bo9+XMJ6 z^^>`vC|fgqg@<^pBPb>vsgx`ic&En!uYz7M&RAFPU3jPA0lDrkf!9;nLJUwKcte)C zTf+3_!$4u&au0xwkP10YxClV*ZdND{=Hw_IWlp#snckS$#UY#=^&qCBw)W5C3M%by zns9H8rhecbn2uLKEzf!jlXZ*>`RiNbfM0xy5OQzLn7mv^u|=aK7#3n+ZjH zJ`+raB>#9+3SqM6z@!xojhdNBp-xHcXT#H`TU2`#NroEZtAm^lnXz0=Fr;0a>Uq7x zC9ExU*d}rLlNLP)@g|fKg@@%j0TwnIGaX1CYjp{hHsI{;=1>@U0``hXRV>=eOy5Pddyn)7-DwV8uoA7+;{eu0%l}1 zWM&^!orPS&u__@0h|G{mrwXCE`LFKWxVW!5NTd7YS*4MDE}>}~(jR(;m%Ho8qF$FO zo|1GEAm)AzWtyd~y*qjKc{sZ?CXvY@nM**og&4-#@2k_gC@0h>jv*LQ-my9p()Gfgie74q)D%GupG-pF z>P37o9y`HzBI?Ihgi>J>Za`6S{Cy19-R9Ha{>%4U;--(|2fa>-Fx#rC{q4jpOst*Ps4m-L}c=UN_MrUg4qfWan$5s4WABV2di%V(FI zVu$elyR1-_`=o-WLw@si51ffshGxbqcz^R7;CISb6Z`X5Uwuh)ceC)P+&^P-o)qp6 z$hzK)(8z*rWfN};B|~lS1z$A3vqZt;ClUUt|JO$b@5lL=cE6 zJDVsRd#cZ#)j^l7E7Ob3ZXTCQUczLdg-81t05Z_5fw3>F$PalMhy-=~HgW)}zEkXh z$LN`!3pPSmtDzp_lE}@MzmD?Bkz)ANmO_IiwJ;x*GnM@qzj)1&6!DxoAuJ+avz*Mm ziilZMF)1((dTAO>6Z zn+p4J!_aEwN4}OKavi73UTlOafY23;gD}V7s{vJX&4bW(kNuID8j#; zA_(-#rZ72Qz4{K!>gxy|J;oIIk`Tsk#{?g8Un+-DKe<{y6870RefflHoSC`fgX@{M zh~H@atkIrqZ95R`V?&F-w$CzrQG^&D?y+HTS|{S{xu(p-%IeXbAR{AFahzp?r0u(l zRY%tP35TE#d!P8an9u`gj>Q=6KXWIi*;NxO(U2p{s>})HD=};Jr(@`Sp7DkomoVZ* zG$d?E4TXn%YLBuD^AVqWvG*TxU%h#Ub^PA0*wq`9& zF_c43bQRb-4X*!;6}0|Fb0_sZOzli!YgP8WTBCfOa{6sO5*c{=4up0DIYfi>->(1MX{sfU3ee1DC8{!4SSlfm=-BxfX& zhm^;vwH@sm{Ocq9EAA5l(mz)ZM7pxPA$fm*IAX@Lv(8^NM-bEqHrf1mlC;I?MEXbZ zQNQB*UNWRHyin%N$(_^9#M*E2RUEnvQeYc6M9uxJ_tck*d`v{vG|5Q%=I-Zf|3OwCifE0NtmKE&#g-&`0 zx4w^3B7*@xnvP#sMt3CuL>Qy>Zzo2a{bYrECp!WPgz8u{0hGPYr=xro56`4tNY@g>%)ir3TvCekB={?L!)i0o)N|Ss|eZU^?5Y#KSqi-f6`sQ`pOs zX3p959#Bql!QLQz*JM~N&3`;TSs;1cKbRN_Ic#H4F)6FaL;Qq8b3;Li0f@ntd5v{# zhLW_;bg=~#TCVrs7arSN6hL#8i1NAfJv|j$SEn^a&E1&z)T57NX{O%dr&oVAy1+(Y zn6SyeVX*d;yR+pBbz9H$3c-B{DLC;?qjBHreXN4JFTbAEzw)Z}2DoBf6>EWUQ~l|; zKfVryugH~N>k{$+?HAFk&4=A@E?0s&N$}sL&^k|@b^@+~41mOZu;kKD)&j&Ufi?r9 z$KK2c28+n;QO`MC>SC$Y%ZuT-&F3^jpcI2ubQ(kOlBgPtZo;RpnipJJ;^>iRLbgeB ziI}vM+}CFrYu`5pGL`!eK0l4Mx1euf9vZP7Oe(FM0duj?YF8#IQC>5svdzM)$%Xs0 zr=kaq{dy#aao`Q%_3QRC)A!de%QW9IPWqe9-~T`g@`O zve0%br;$kQ@)OR26aQ+e?Td+Dbz>RCZ2~^@jKpZ)>8whikm&Nj=UP)kBSSM9Mef6- z<}G0a-*0GRC8Jg}3JqGqA3MbNI5-A5+aFlV7Bk)xwthU#z>jl2(F-UcJ-3Wo1l10<7w40`j5Qld4|vCXH}(y zfY?byIN$mi9@ZDCZC<3_UGMN5i6Jl*5s9H zJodD8q)CFKAM`ZyS^U5UYMv`@6Nzw5<;*YXqJ9miNeBmxD}kfcI@zs+WP}a7tED~K z`XDUBT&_(=!KAYo_%#@*SI0OZt^NKWRA8R%{bm_t; zROY#plVHtSHLspVmTP^EtN1Xy{MDX&7+;^_CgDsX$zd#5o!@ye;_b$j+wBFYzmt() z7CuA{e>vqadE2V;C^7fPohk<>bc3XOpuWyJoPbGsbJ1REKl%#?-_V7?h;$k%|Fc&= z0_Qxmn6@4SC`XwDo~rC*-#^LKC8};>88}dDc?SgGgpQ?eY^R{u=^^jTq!(s(rixk@ zB10czxK4>2Bld^-QJ<^IH!9f$$0js|nbiU2mDve_?`%A|H!#i*;wjH{b6=He1F0n( zTWCYMxU9?L{!n-K`T6M-mKY~LLIoEMCU+){>r?gKL!)ATxouG}@;-EN7)6^Rsp>fwv7>ZSSfcS{05MYCv% zu$RyO%Gqn!GqpsssAa(saOB??!8zX?7XY$8IX%4;;1MB@ocQFL+y|6{9n>swYvN6y zH1S6tHG?8_>c~|2Y_S5K3h3JY^ldkYwwkslOAux~|zx?9Iw&OxYWBbC5{L-Pg zP(B+JdzLJp~+u??rGBK9qZK1|RHpPHRL zu`PXJxn1|-03$?o?e`l)1wr@vEqB+HgsxsDpV8S+Cf{izk3E=d@;q3W7F7~G?%y0q z=XLohT|{)TJP~}x!_1;n0jaZ@YOvZo`A%UjtKa0DQ%f`Mk#&VA>;y!?MhlgS2Huc% z|8cvRw`!fRzrX4i&&ck6;jp4+tH6-?tA^ajWBS!ZiJ$C8_bRHs`?DhP z9T2g2;PgW*phN;@jAB<#wL$=euqxa2@Uh4!F{$S>Bkvr%B}f3e-1q6m^~bO-ow!aP zxlscs=@w+w!z-3&uRs2qosCTh>ZgbRwf8Ee?t9EkC6KcnvU+=67LUU$C#-9KJQJCQ zAm$~$Y6sO{(zuIvC>kNs`}r)Iu74~&m@qqN`>v) zd38Uz=HYU?m-RZ{lgOUufXPD8>YE!FE8`P6Y&&&7pVn!eC>~bgz`<-o#1g4mMr3%S z(-_rq?K{)sRh^EKVbJ#adcPKlRWHe;GtrMcI5Ekf7}6D-yt{DD(z-;h-|gAdH5mQD zd`#o=^MZ8WM;1-v=NiZ@ z%4)j?3D!*38N$LA5o*bCjOy@I1-c`vLz%D}@FEEQrWK9C3=I85M{ggFVI*PVPa+ul zY$U7xT6?#iJdjZJqOih?JN>O$sFt5*;imvml5Rr6x}1?8I+r_m`Nv-fh%tRPQL8>L z`cHh$`*WmAK$C8vfLG%}gLMi}r??PiO=8a6MMr6OYot@h#ZNXjw6An^duJ5)rFn5o zw6y3K8nn{r8MX2uW$xV9ESjMUBskm`c(1Sx@I%!s*~D$A#2j7~3T@R|<_PEKcV2n({|-{Yx-)aMy>^xD-$H`)~=0d?8W; zV+4SD%I~(hR)XS!SZd;#TH@V;+D~KfX0iGs!i_Ws>Y57Zt(>E_$4%Vvr9LwX0auS@FN9E1wNrI(>l$_UXDp~iT~%7nl*SdK z!MrLPrv?tz2U^)paAD%Ta8-TT>NgC)1Qi-pP;?+WceKGSprbRLdlju`>lF8MDwjNB z)~))E&AlG7q(BJN@oamPv_kr^IP7eyV)N1101((~n(^^8a-Wca3yWn~nrM-35zGd8 zi$QoZTFQuF@9{p)?sj|~R`$;zZtJt{Nb73bPm#}`Oi=?oVG0TY#VkuL;}M!gp3ZQU zqsPICy%+Xh-QjQHB;xw7=>#VhRy{3wuC6#w@4@^zKmUWs2JM-)qbYkby(foDoMr8L zvRU)BPe(N?DNPJUOKV4rdJoRNK34|{#=%8qeZICvny8X<{=?H%bC#hEEYEj~#oatp zP{*t#C>CiDY55ytNfpmLVT>{t)SN?49%~*ES%Xv+s$NZ{cP8vHI53(!!d?cESQNCi zE1m<1kWKXd@b!1?FIR$%vlb?>`3wyu6wg5EVJx~#O0&WI@e{&Tjc@lrgEB^S-fi5& zT)b&5jAWO`xkd(rQ8efiy)Pvky}jQJLOU7a_+kB~rDeHz1hng> zdGWL9WwLFgb>0%^2@u%R)NsVb9KOTz%xK7A_8fER;}x8XwEXTPmlpSXX%LGgAfS%B z2#eMBHe4$O`oTUY*Q4Q$9YSd5(jDKB&SU-eZntrcr*=v+!U9h6!Cf-Ep8Io~W?zz^ zl07-v>gG34Ox)ZCp{q9Y^aEdEewXibGk2tP(ZiRW>;cy{&dd8~qk}j!`N-rbyJrU| zLL*IUYY&!!f`pw!T##SvF?|#AJ~Bzo8o~?0ejvreOa{$Y1EYr0xwr@3@cL5xv#cuw zJfbb^3mQH$NebTgJRa_ohqYs3Y={Ur{jAAAG_uDc%6lu*%N?RFCsXuWuvDkgGmo0E z{#!)B^=9Witdb;HC$&Ucq!nw{VvecCex6J8_&c8CukAPCWSa;Kv0~TOm-8_uM6xud zJ=m+*fFQ>C*%r%=QOUWi+>=@HN`W6cb7fdN!}Qo~@}i|g+#wUv0+Xhh)^p>Ky1qbq zVM5$QpN4hI{P_!+#(0IZV=9ji!?%sux%GB#i+NnXb zXP=$iymA9fT|Bjg51(5IGZ9^{*QYF{s3hi6|48(V=gRwKx$=I=gM8Sc+h;~ghuo_o zU$_$`Q0hcSFA5J$xKTbl$uwtk<`VT*dz%l>la(xYDLE6s_V?W z%iT0m`Ng@#pjvMhgBs%Fg)cC8{(=K3SbBLT5}1J`X-OzfSweF1EA~r4wSzUn32~{& z$BNc*o=}RsE5!7WK$9#P49=fRwKpbHoF`=I6@|fx(Y+wF{jG?FeRTdSQBAg}LByLr zhoX{!!h@^}hHw0!1z%JrH8_#AuRF}dcQqJ!5p|)_$@tlheE(6_rXgK2ifT)kXBDUG zh8!ku zrAaD&^1`9w%dLt?p>v4D;sF;WjZ=x}8+$szLhQ&68c~!I)=6P#Sipw4-IKKPP*&NO z$jIWp2)9_gnNG7Dt{4oJOE*jMKMW;D>g_5qNIS?vA4X6AB(pz`^{VO>2QapKkU#r} zMjGfNlfc=G*a8F^Ho3<{YWpP(Ri4(!8aQ`9a{Pr|U%iWVXks`2r5?;+i?xGs!HQZW zhs$Uuf0}EXgHJp*2`n5GyFwZK8D{>^h}6EdIC{`!zbJofslC&F2?PS;mIG2-Tqm${ zIcgk!$?vqTB**Y7$fP7Za1AA#xL;(>W3dz_I*{%0X|z_CuC;u0A#B-AWQ8p-7_t`j zUC8niK}CPz`fC=wN-&+PP_8xa`4074C6A#+a{#I;1zzM@!ukA&1Tw1gezn|$hgn~g zi|R}38bi`G1$E0@((trWt5l1d#Vm1Svh4?(K<|E))?@Y)0~4`{ZqEvZAEw6gub&?+ zw){sB8Nx9!CKVSdRxpjp$SGE@yGx=g)&t1`XUt>Oxv4G)Inz>aDC);-NJ+#l`-@#7 zMP>7$h3+Hre%NPUUa)--myK5P&OBSvK^UD^^WX<{60(h5UvqIYQ>SrkL=@n*AZ%@ZsbH7N_nlAS0 ztM@FuIWgavkVD)e*76cJAs(^UYd?}&jf0Lh2u>~v=g)UGB)5H8+}9XMBa3>+4OK)Z zP~7AD6vO;n)2sb!wZkDK$M8O99$`o zy8Zs{#&-H^@?M=4t-jK`7~vzPY*CL#G%4^tbkU)+|7jp;*d{qHOB>b2;n%~x-1#&E zXcbj1|4P(0u>9Ngv+_N#X{$FZtkKp#l1#FSvOZYP*BsS4$Dqk9q7+%jh=*FFuw%`@ zvhSrc0vGG5zgdWO2dA15SLbrQmV2;3X>Y6&i(jVk{mjt(+zx1xs>qFM<$H$~my$Wk zNSP*($Pl!edvrEe=naf6^AGklOeJVMRj?_Q*JJR@CReV#zq4BYUT2b}m7Z$xZNqE} zs_=32KrfXI2eyhFWq*5pMpiEtK;825L&Wt3?E1{4-5BS_Yx=T-*XSeNZEK6p3UOsG z?w#S1JNVE0eiJJ#!@kBa2P$D-tm@N#O*6<0d|oNFFl*Q+U&cuRFmvu+jQwu7eoWG& z`)#!61xTx>l<+ypZ8gF(T>6MS8r-}#cPiA_udK2aC;|y#X3vf^iQ+G^cBnK&gyI;&p>ZMz9TaUxb30avECeKtGCi3K<+(-7QU)u`S$n3L3>9jTD ze;yzAQ+Qp+;Fmo!_OEuUSQ~Ks%3-=)-5&yp%eO02mEvM0jl8%$m1OQb_%#=T{u>17 zwB^^aCvHPLEA%R#u=8EAh1K+SxDt~}l{V6aszw5cr*=^573MYCE6|yisU5ORN3;_} zh2?LI-;l@?Umw(Gh@OWY=s&|IQhD0^;?vk>381Ha9BkzMhJ)Hc&NsT_5pEO*#X{wa z@NMM3d)9k52{hNV*gLLrL4v(rlM@$RUN8>fF|WPwj1vlrkli(*9ENL{rZTgLi?0+g ze~I8d?2!src{s^C=dm#wnu@iC#-UzsGKYQl{JY;xvkM9S9-&N(oIO?J{kwzmg{M{OxA=U zkYnYog+MX*S=s|9EqsY7CPlM6)6TBeC&llXmRvtde^?ZUDfk{qf;r%Px6x|N93}UM zpwfHnt%nkDedeOdCtIg3{fEEvkIGqM%EO<R*C8OiR@!xLwl zeHR(*C@t=3pILdQo5-6xLL-#F?m)csN3veUcx&cTTg`YB}qV{}tN{x~}E3@(jCZelCO z`zXTJpi|6}cEM3X2>awFHHi*NB%~crC9bolx|H;}N-y#cI)v?>4&(6RbBwTidtY6< z9E_RbW#J|hQ?+z_g`9k7WeV5Gj~KG|DUFMK+lV$pJbswPo@_IWP<)T4Xy2UIgj42igS2};!w{Ziz=x%5S)Qd0FflT|R9BBLO`p+1U?7U; z3+kwIw{e%kKnYq%H>;gt?gEDFByyi$d%=lVqO_u?$}gcysh8*8A^_F;=1ap0cnGUpr}4>$s2?u) z-1yeIINCy4F~7`lK9W$N;4=M=!Z19btJt&CxA!BNWyzW^mognqVBWU2xwkNNrkq>! z6=}Q%vI1g9(f4*Wjo3Rs8}CD1;)y$5vbkdyJfpD^o!)>Sj91^Kb!s4ib&xG7yR9_E zJ$ys`qy$!tj*@R7uRQtY+sdz%94?cEqR?_l64M-82-FVb`-1E)#iex^gbj1gSN~~& z9Iujg4PJC1Vr>s|k7*H!?vKEyJt2>fRPR8tHV!(#hVj0(8n^pa;%>8eHq=Ce7p*HW z)$G9QlA)KA_^T(4*#!>JTs+Uv5UC2*MAL%E!d6*kuA}}R`*W{Kp4w@3IArN0xpmQg35;}5a$_ww{d&jzRwit#wnX;p?Y+3@jFjKsd znDr1-wl!N|s}E)EiK7F<6b4k3Tr@P!b?OU8fJ_f`Ayoya2;2J@`Nb}2X2J4DRz3x=rDF7sl=Tn}B^)Pq`?1S=F)UZ-~r zCj6uil_O5&)Giy>Z!01)ykJhBm_gPvtm+55z zp4ad#1$XW;BR~QS%>DQHV>~%SGoT+t}DtxX}zBP#l{4!X$rw zzKxSvMKT`#f&hEBk!rx;*BjRqT))gO5X1@%jr15rMi%}8s}CdJvxH(&!;GFqe60#3 zW};-d;!so+E;(>6fx7UmeERa6m}dU}mWxUsf$0ZP@F_53(CJ5xsBlmo8zspgD6$z; z{mNHJT-*cAmWIan=+9dAcHiNtz9bMFN`C;bcTN;hj~5JG-3ok8O%WONl|lncE6h5} z*D*74doKUT2PhrJ7aGzzWI0feZ`!ms9t@<1meEcttkkjbsRcnfrUD$4oi*fIr>OXr zj647Fh$7M5hF(hr>5eT5aN@~DJEeyhgC@lgNU=Ees7(WAL*Y-3QVi*s&wY?MstHuy z2Wzswh@$ZW>F6e~D5*y16RszWaE#X~l9Oif!w__X zm_KaPP?7YX4;IP7&ajKB`!m~lw~_IH_2I5r{EzQoz|Ic=lr^M1x5wZ3?afuM&k}FF z1|6|X6engOY!yE1I7s;qiJCMCU$GyLpBu9|NT;dO2z#*8or_5lAsc*!H0UZmXG!?r zx(_DXg`zl*uiQ$ex-~R-o}!)4Fq_KW6f~qRU?#SZ(|M3OEL3diNT&R|09lXWQ>lxP zxYh5tSvr(0;tNXCv&B;LpEN*Z_@isxyw+222lFc5l0s>a9(qrtf`%TVBj>8mWER(M z0|TSo8CIWR`%ZX18}B8Fh_~fiOaHs5b^=-nd8{Y6$dR6@UZ9Hh81{+ga59Zkf zQVVka#%Yg(v;t;zq>WveZ-dl|~a-%q*b&tYmyAH13I{icq;HUbQoeESZ>I zq*!WH?XyxQ=3Mhxi07{vKlV;T$QNh0QMkW@Wx7S5XtQDfE6#%ghOP<*ygFz=H)`^@ zA2~jjO!#s0Et9M=@nG#}0$bZlp`Cn_Kt){IFQ3u0)2n>d^-dx(>*?#L8>BSC&bU2h z?{Q1^TqJ+=PfFn#Qz-5!zujO(cRejyNYzSjTKU^Hi|xmAJ#4?+_3FIIB&qM)u1^X3 zds~H0#yfRW`4lk5aXVoIWYQSkItt>ruEB|Mc1#J(5kgyn)`ex$zX;fU6t4I&>X0n!}otlj8&G8 zEPM*Hv{81cK1F`pBu_R4yq=Yx-=*t$+$(89s~6Le;Ugas)BUq_N(250+O)_4KnNbVibP|o@@Bxjl-Fo|JjC1f;jfL29S{>iGQE~}CQUK9dc5{#$^kEijdzPD zK;^s;IvqBFEIchNkjI!W(d5+?!_^4EvD={2cn=#HsK*IP5Q<6s!2gIz`h?)4)cgXA z9tz2lj8UNU>s(g2#Qop9O#xt034mm5kc}pu&UM|?(bnDt_4iqLmS*U?Q*hnO9=KL` z19)%7Du@JLwp{3;28jeG_LJg4mI|MD9R{((HrRvle%l2PBdJydsa^lP&o7?}agPI0 zNlzf2Ii13xBl;(aee-V|oPsa!1b^4G!*U=Ity=Tvj*`qfmgm>5M888z43ac+53-DO-m=!D{dH7Z3DcC?IanxMR^-0Tg-gG)aGpk<+NT!6ym^?_? zEiS4$TOZ-XKn!oP{r|y%fCJRk+=s%4gp`|GQscZPbx9IID{?YG1MF4-82@Oo{sFeE zKEMtN-QS)XGRtd7r6eU;Pv$FJ1ANifk8CaSJqiV@$KDqxw$`)7**UO>lY#sSdB4$j zZSrPowiE?V^C#ORw(56D1ZFQfb%_Hy>&AG4H~+O$$2`ROisp3$#Jij6BdT`Pf17AE z=Cljl$P{H~XYTw`(ebuO2K01vf^sDF1D{hLX#HSZ1h^m}aH0$|>ojZab9KWjUt+TDBs0M2^=dNT4$Z@7;>&H~9 z^JUAqxPD%k5w&ZeUNs!(#*|-w>eyo~@T&h)6Y}Rj9~zB7q~`oG6X1bWU-YrSV|O6I zz9T!U&Jvww15wQjBPNbFQ6ha8Wn_j=()|gKLDfC@^9t;@m*E#nw|;$q$_ZWF6d`zPk4xFtsX!XbNoZOZ%IN;`jSj4*uUVK;r#0)65`kesQB>J<1lVu`D<~ zH9wdZ;TY(Ex*@O&3quPU{lVEkPt3kG_xp+k(0kH@+iV9OFF9KN`zX9;aa{5H43NFP zW?6^5J`*t52P}T-8Q7i+k90kMFY3Z>E@vA%&RLgShgkJMP6MYH;>V4~v58U!rzx$p z6BbHwE0HG~tsxcS#TvYV{;w+|S7ZG`cMcYiSKqABB4&g?fqk{y+dA{S1R(=e}T5)|>WzxG$;g zljr*xZ%>?RVFEjz~*Cdp{b&oPUsT*p`FZ*+jjM?MJwa`}YC!zS|c2uHNXll;Ot8!oqSr zn5C|*T{xJ?(26ht&zdr@sVT(k$BQHTta&;TL#FVvWC`Dwh?(w{{ohBB$1((7tziXi z^N`k$01y9d8?u!5<%k?xUOi+Sr8#htj+lAw{^?$UKeNr;pO`1L)#9 z-*a15;|>G%wxbYu0xMXg4JP`bQ;-X=D%A7cZ}D$2nZ0^A`+g^ z3VgGb0%ubUcBZ)mEL{tj&g2Dk9t%s$-SHP#(;$V`Y(MTJsR$#^gD@IafsSLjZmlf1 z_MpYfOgIQtU}}m3PSg4C&elY9t_b+EkviH8(h|^EH`J_`aM#mIlZ_!QfmdeXBfTKt za0j>Ca{24~*x=yA&0%LH)TuMz@2UR31Hwb*T&7Nv?tTyI88`rLHw~~(UU0`Nbkv2I zDJJcdi$fGV0}M7CPw9KTQ&bOJmcwGxgy1|JXR#Z_J^B(rSMuymcpm@H3hsxvjfv{O zi4;jZJaM7#+1b6+Bj7_vaGs{!GL=splJMa!AYq7r zK_1if-P!3fE}Wm2b#i}=EMNFrAG>i13!P71^~757LXe<@by0iYjFJ8C#{gUJW;In5| zyF)#`p)za(ezIv`KX()f(-1TsR4lL4NZn%H``s>l85<9Tmj;=9W z-zQib7jxkQefo`#@Iey8K!1&ZcmKchGv)eOJg9>O^#~5C1P3{`0?=?q6LHHbC@4$_ zij3g=&u5!bo^Y_TsyMZurx*qPv4B(5-LEWXc0u{I?-5sE)9eNBr1Ei+^ z_$_L6XeI7fZdGojGFA&lXORNub*e;W%j+dv!jr{u~>l<)XhPd=TqC-B3tL_W~O%SVwjy}&~pEV`2pU{fSNrzqq#c75`MFX28?+LVjN z!+@IeTwYL-_VKgNCh3ar5Ta9`Uf+%7NR# zQebyF7J%Ij4^&X5Uj(4CY4AlCwjk=N#1y+)Jy`;!zgO;ECqwMUcWP%8j?Eh>)OkwJ ztG|!)M+ScDygRSl?IUq2159HTT&eWeSni~Prm-C-$(5w=6UGt*(m__>BdZlmM-UIxj;+pyH-VW9YmN%_QA69Bx4j!cs4=h+N z2EGo{0B&qJ!UwgV2EBRl_g;f}J0Dj=$KVp6(^!E=p}QOGK|bdnblP%{`Yz#b3dQ38 zV!GC#6a!I}@|wzt^)>ac?^|O*cc;4`!T3s`XPvVv!|tnoQmlsU`d#O0P%TgcOydTz z@fqT)(gzg+NX2iEgjWOtX0z0v&W{AIo}Tb(N8B*|NEq?~d9ZP&^EcFKBfo=J9oc%| zcrhOj8Q;~w9ds!?l7Fudbqn+u@}9&jc|yw>wGn9sMPQPjkhor1!;_3@W$qCzhsGt( zo#&T0M-gVXaH=jL02V5J^vs}Y0kG8;F@F0saVB#l`Er7J9P|;|^;~+NmVzQ_JuS*@ zmPlF$3ONc7D!!5HAl)=5|GoZ7Mk5*b_vHx>Fm&#ti`oeyJ$y6V*aYz@ik!VUCy)(z zdVsB`1M);aUDUL}@tu#?BJKdsqz#OJAd|a3aNi$qqqaoMMXG27tgr&I;oS#13p#++*ncq|UK<>Mqof46+sa?58-aRD${E>7s7e$2O z8hJ1;GPrK%8gI&syk?IQ4#~54lab7w|@_Y1;@L_VdMhp zWZZD_lt<|dd4Q_(ciogK3)ID-ZivW*U^Qwt4R87Csb05V;=6t zcoucy!GOEDr@2bF-GMu%XH0TX*in<%bL=IxymLDqO>Z1tO;I@my{a3M6 zf40HX>CT8`gyyuozb@@v%usW&I8FmXJ}GdQ;W1{Xv9DY)z<%zdta|*8<{kfpG2HI$ ziYT5j97P;|8g;#=_D&SpHC|OgLdJuIfBKjp2aANbL>@~N(c04M+~9e1fU6-H%)XO5 zDGtJEQviPr|CL`#L1w<&P1PMb^mBa=KP7#GuE0NdKR8`v*Xi$Hq~9TvNB0J?pF zFDNBef`(9hrCfg6*{l#9HRK4L`IzJ8{w1&(NX#>eYtR|()+lpm9&s%LVB zQAS-Bm$jYzP=j`gPlz8)y0Rahy zD>L1GhPTZvAE z2Q1$qREVR0!<;B`Zqk##4*nr6BCYc#s5je!<1_S(^^)qyx{t`}IpB2stXy+WakTx2 z-CMGLH3VH7WD&D}U0icK{%h-;_lh;F6G2tkXBlpzy52me!`xW#ZBm5;+;rA1hLiUT zdgp8GH&gAKyp<(J;K-+dEr2NG2!<2bg&b3KEs3uA;tBN^K|V2UUpt~*4%j?w{wj^P1#@sgfRgDIjD_^|~K0&p+3 z{%)zs`-ZFZob#{82S8mN097?ps9>$?LV#OURPmenvA;G?4}_dwb=rPD{rfUJbN)&J zCN(S?%bs;!_5SlBc;=<>!j?sU^A_;x@E0F`5P_GF@P#E^_N1lh=l(aiw0>K@<)TaNr;of-=9Ml{ZrLqa9(uN=Ti;Qn>m5_&Up|DEoHp6GJ04 zq%;gA(%lT*p@<+I3KG)Y-8G}qjkJ`6bPt2Hij*|Mh#)B?v9GzGXTST|ANKpnrE7V? z%=M2mj^lT}jLM6nNcb6*fmHou@E<=SQ1unL=^*<3-2g{wHLxBFjyIIc0%npdjk?cT z>iOa_x%>u!f#8PdIFEnMc?X@vzGOcm{X?;Rxl`Et4DhOq-T7SK9E8;a>dT&8pk`5z z&R^d;chR2s)CXC#bDo=q5~7VnHC8|w;m40+Xa#W1y#FUS?cRM-UtIvS4@mPDNCXbM zf%CCiohhK~ko|0Bm@91a=%nDh%L%wwYJtnE(WyN9#eM?sfb*DJtY$|1<)mD9|D1{_x2%~&Nl=)Sdu%P zR*t##-%!mwnUJvhMPz4j0zf8>(NynGV)o<*$piwZpPB)5f*~@M zibSWPsWZRORf%jG%fm<-Be%dWxu25ONBiB23XxNOJtz5}Nfz(}v9I7`K%wvbRMwa8 z0;jz4ow0j8z;!9$d7ZZ->wp^Xk3LEhaHQP416HT{jHbW+ivMq~LK$yghwHPpUOwP* zyY4{fOaV<&yaZ<%&va+;Cq&&_fZm}H$ct)@r6B; zqIL(*c3Q^djQwa=3Ih;#}{^a(o>_+DgNc_-ofIRn(@yzS7L(~3(*XctU)xz&n6^-Ogh0*qbe;HF?IX0l%L*}cczh}&jb;uV1}K2!IdKAp!kBvo{{vDM>*TN z25P$paG`7YdoDJPB|dkZz7u#46e7BvK1_)7LPuZqi(uvhNnnRGPzkSb3;AU>WnFTH zjxW#>r~6miVZeBfUW*W5<e08TA7(| z?Mb8wP;MPR%hBPI6_zM@t;-+~20cUfy<%1}K!nlU_}ultfi^2N!ti(kkR@}tpJb#3 zmvsQAF`-+ZkB9uq$MXzzCQ~_Jg$S(PyAPs3KTe6g=_JYOx&+ke3$!|%fR=|q#IPl+ zBwB2141iHRGuRVYWKGfe?>+JPqG!N`U63INb{FSIzI6t@#}Vdm1q82gPBW@B`#ce% z*{mADz;m9U_m+10ZH2>!`j$Nc0$DzA zWpPlw0>nvgF`iqHq$B~8VBe$Et&L#Jap8a1k{&(+ma?g8C>|Asox=4xX$^0ESJQMR zu@opNO?tQTpNuEsW<8wwbPG5jdrEc|w^>g<0m0yc14NYL2NKO+d4z~}UOoXzw0`Ys zSU%}+7Z?Y{9m)&n4`v<-7zkvgse8Y+KXPZ0VFvbdl5CDuq`JJ6g<`0&>GL;iG>b-* zoAMykZFSe|qyA=jlz^H26v@Cy~aC3~!;eIOqc*v#5fiOfl{)>CKpeIlS~kdDy- z*fXk6CS~fcfWz#Wi>%i*`a*IAP=C)uZP>rDR`~%39JLg>nP7UMJ&BS$SqC1ZP!b6r z1r5`pqyIl|>Vw}wvPM%6rtz%i7fF1d3Y^uo$MtT8lr}-dr*-}VT#?j<;5z5)%W?y>#?LT*->V}P%iquM&h*( z+Fd0wPdDcRXV;|*Z{AjvK<8c)_I^+)PT>7s6s2mwMTuZ=UHkR}PeF!Ab+qkTOJ9(N{8Zl|t?pLrT~E>P5+bGj;1ZpspeC)Jn}m#F zvp*&e#qLfx@S$g0yS-3(Yp(pt!Y-M(S|86lL+C&-c9llTvA* zG^UMnTEv!DoEEZYHw#4ZkAb}K*|4D!SA<^Vk2^F3{5@|TR`m>y@1c4wXa$PFqaQ&P zI|8^BO(}ATRR;J2UOW^`Yy~O}!q8agPZ@~ENdLYBZ{xxka;AMOa3T9_-J@YQw<4!Q+(kz8@}be zj3O_hHW^Mk?>X=0m>(B{C`{VqnP5I6jV8ID>?%6M9?Q(GG_VG=v_sK#%Vf#^R$y~b zJNyMK>F&k63rcS1{q&kU?zPBxjExHUKUJ_YOCVkylO3uowtDXzqQZ`2R>yqzNeu10=GQ@GrG!>d0;`h zEqKFf7PT86f>LIFAdT#T8c)X&YYgCjo(bBSKIEZUu0Yy zye`~}(vWN+oLL%YC@G*j(YeZZrR!lNcNqcv64z!x^r!{J&)2|1n3ts&(OBf7!lpj} zHKfc>{h1>2<2LqQK&Zj&>qimFShxuk*9AJ>-6KpIpVg&}BIKf?o>4$#23SDhLG*Wn z9BH_iE|0IFgGFEq-L?mowM#Hp1IhL&`MnlP3kl8Xh?lsc;}_JALA(QJ@$Q2z|FP@G*HGImYnR) zD*tvOuKRm;#*X+d-$d<+VB8=is1<}O8ujPHuBBr zGC%BO@<-4JiJUzFpWbKQBm$zZJ{mn(K6f4^J*)b)LkEsU(4SwwyG|xcc92T+;){-@ zyIzHVzl#uZ4R@iQHraR4GGk3_v)ZWum%|QgAWjO924C#N6%*ZMTg~y+(Z&yG5^+Fq z7yCG8ExX-eDT=8O{agc;K2k~=XYDIutizZv>rgJ4ZU0d0<IJNYAtzKL+TSXVv$mEdW)OPkJKkJd)Wu=@4Th zr|%yEsDdu|VHYm0NlEj|%X=Rc$5qLR*tNG}yT|p`l$W68me!l*;?kkwu8NyVGrbo* z?Jm!3V!621m1A$2Ox!=>3vvIsh}kvAO`ZB!aKue3S_x+sWSjYZfzHj6 zl$5oXTymNN3EpP+CHrC8<3dauoYWd9!)Rk)w`nFr@4FI~o4y(5W^2pyJg0Km{`xbn z2v-xc`4_64;ot4ZPXHiE;Fb?5o&I{j}gn+#U-1{6NoHl`iT->yzxAD00pZ4g~ z&2ZCq*s!8)R#oggj8w+Ww?aG-E84?9?4v;mLN1dw%=4*+1G=-Nm+=Kv^db3c46=QX zMM+Gplka11OW%EYzRf#XFO`0qw1w6}T#5c2$WzG<(5IdD;H$PTsBG<)PN5ZxU z3yPQAPg;uN*|_GWGbp}QH~y6FA`@h-xuEr?S))+gvYt+h4EN5LZ~P;ffJl=^eJMIh zyyo^4{yhP8gcQF>{=#Ei$UD%d;C1bODik-PPvMAiS*h5W$Hg+Gpx^o$Oob!W(>9Qy zgD$@fzrbIn^_nS?{sbpL`Vv3~I%44bCDOV|0EnpVjNeIomuH{%khQ58|@Kd|>K@I_GtIvRw? z;PmX>J8_y1-7}P?XAC;}75=cmYf_4%{bd)NE?C?Th9l2#Gv=38BG8k2`IO=-Aw?Hu zWm3EOqfyyNzbInq)Tv>C&9q|X0=mxbD<7=!E&YZ<2FRpU}nqr`;|OdghlszSktXP%JA zD!Uz%^;mpgdZoBEb&;NI9W)b02u6woc?0Kym+Vsh!=s6D3DkCwDxKqEl9dCKt#i%@ z5;G)50!TfOD{C@MZp(w0!33@NVNnHeZ=arkc;XBWA+Zim;B(tOJ)XIEW1tKU%VnQ_ zJ!P%8`|cx^eoD}ZT)yTDh|A$bFxOj0%Nl{KwGSFhJmX&a*Fww*qaUn>Fv3F_R5Z$G zgtkeq__i@_CA#BQ_9R<Ksac>3_Jemdzk&+5#KO2rfmQXHTK%$+`pKO1X2TvX~L0j7bfa^Vtoc$tK7 znS800rzNSYFyV1VAb>9TX6PxS>Vcph?Img<2|Pr;*+sBj$~!W~b*Xe94zQ&@e?#~W&=%XJ&;-xY zq~>pt+>BIa7|^QVfBK?|ke1J2Kfz^(8P3(PesovWa@kiS+jokCOPKkyFeKoj$wzrj5+D-G{K zJqu;Q$0pg6J+q|OKc2aZJsm?W7n-d&$Q4=hNs$(B(^k|4!&SAvZU*^@5_rSKnV5bP6@&WG-WyCw#wv5 zv$f}68!I$oQZ6|H0mA~@))*jVyf-(}vRw|gmbMvL-!27aLQce?RW6AZ7UJIVB$_S62uZY8(2zvi5moA6jb zBi4V*WwVYBbK7W8)OL_#I;v#lv$u;pBGXHtC5%)QCntoo9t}NxLRzhB4Ft^ z>~SEdtJiDas=1|RUDnkpA3ec{@nR_+!FaCmY=Uzgjv}^WK$YwDpGwc0B|1rd#Cq?8 zkMd{+3C=TM&Z%If@EPLmm_@_|X~RCpCEWF2yD(RtIz`w$8+(S?A=AafD>lYQPMTkr!DLw}{zpJvb;OKaIij)=W4JP2p-B>0qE z_QQ|#H}$`N+*TLj+`mTXGuU@h-kT3|QPrdgeZzt;V@dvwjbbhLC_H0k89l`eeE67`;94C5bRJDs0}Sa8j|4>{6&!2DJpOJ>9*Q_lCCC5 zV@QnE7x6n{I_cDHqpF0hL)~`5_fIo2wZLr^9&O=YACd$aCW(IP`j0JwZ(K z@oiglOt|O*7@N_kfqd8>&~Z$h!Xw`|-XF9zA{;r?seMGz`EvC>k^M*=ar(*ho%P;g zwH2GBFE_jLHiayBifoVykAVaxXpG49Lg|dsoDdd#kUCU*s4^3ZSv|J?Yn&LJcHg8XJf#_Ne4&CU}e?^h50wN zj-aXOg5~jY{Add$t_PHO$I5tDuL4=XF8Wv^lDb$h*h~3e*1?s(ix>vmoubwt?;ajf zEV%Ap7LYssVxc0YC6Rxk=p{@mfNQtZ%|nR@qZN#bUn7SO5)ByPJs+cO9(XBAsAsFL znqi;3rN>ANIX~@wa3*N08=jmA;HFxPI(Jkq6p|6{-uG@Vw{*mK4MA#1EU5+#9KrBp zBEr|#qTsKTjjc!%0r&xnji~G{$W@ahoxO=C-LO8djD|y9H{FnKymFm|$O%?F13_@f zs1358a^Y2nqIgfp*kjn=(f0Z3Y_kfJqYsG8T64!J!1OLk5fq-WQxiwg2i~=iWU9qw#d5|($a@Tf2HU@!HR2b(x_AYQ zH7sep18$C;5;uHnf~Xj{eabSc;1^=);}&GKmzo{(#`~>Q-wrChXFM_W6oJMx^L#N1 zSy_RHGq!KQrYv9|4YDHwlvNi#egV(+K%2=d^_p4wORrNFF5Ge|lmVWUE4AJ+J46lV z>1Y~89~3h#?og1+ih=9n(I?%rPR{+CE54#pEa#Qso8vl5#Ko}#+1*QA zlPmhd{^|`7SFwI{#ISygIZez@SWdpDV3h^cTOR-E{K`E#WMIy zTxZ+Cy}VcF1$*(yqVh%RBK>NOC9K$XkO*F-DT1^_Phh0e4IoC@<$mII;lwL8_VI}@ zDoL&pz~`dhPHW&ZH5zVM*H!uf%c?;VF!>Tk}On={N&kuUP6WS?O}=EsMeEj zSPG>$^4u&2y44R%*Lr%y@RyO76lAjKB9YvA1Zi+l?bo&7C4S2mCtKSnJ|2uI+&v@7 zQGS~?KVC~neI&Q>s>UZR!~Ch?m{prl(bVO_uD0b|SI9y^3_OfyO5bWj z6>LBIozsR+eocd=(a!m}bwD$C7@KO(-?>xy`ZKdxQd36l-dY-@V8GF_zt!f;h0*uO z-#+PF!HVbjk?m=B6ZKo_hANCMhOQa?5A%Yhrvo0cUa~>^3N{qho`zf6!*gVg2Je0g zV3xUm&hW@^uoab63cJ$8>SG;sVX~)DX)qbk0)Fl_JZi|~{p6Lric zTKT^#0oS3oz3q$c~V;)R{+$hdQO|G57Rqe9dG1W+Vcwu3w}xoiy)qi zbWP8}REKID5BIRh_$Rhq1OU#3HyCV}N4F{F_pT_79st&xgy-+10{#kDPJVJ2(hu=Ms=835-e$_KL1{e&Tt|>cVf^>2U>;Fp}14 zCeR=t^F#;}NPCbECv6XExXb~6V!^tp^&Yqwe6@~&4`|9q+5bjY5lfu=?P!po!bNRkJ|!#xNbcsj{U z-rWOW*_UvfU^1BuouoC$eIqaroyg8QDZ*7^JcIVQ& zALcPI7s*+?V0zRwyzN4Lu^!=HLY087XlAgUOar-_1q@`oYL9s*!piEB+Uok=Kj;HX z3wS2?BIxQKWgM6l9^LwvqY-> z(#niA4$!nDe0lrq!462MOCKtHzx$Wtmn8dVKdM>qDqN@AwD0q#vwPtL%0#Y^oTL*n z2U};Uh4T^*P1jEVfmEt^si@GOq6(Mi!wCb0txloW{S{ZTTZg?Wn0SLy413V`YVXs` zo_PHuRASE=9ytjQ}X|U5)*5tQED6E;o^L_2BsjCrQ2zP`!-sk83DEN4GClXF25}Ek( zp}zUok=P?@JASfNsrt}Y;8Tn^%l^bSX{`*;PlD2NS-2_+?$#`9OL_A4&|jqgZIL?o zXj$XumU%=LciwArXU_d2%|q3usEP{E@W?9Blpd2+ zacIn?Fox87c|$_PA6KF+ZR`jrRO4YS>kEx zIcB+!THrmLm2u<7jZOTUOo=6i<$U3!q$|sq&-sosFD@+7*;rNVO;M4MX?!CF`9p9m zL}9{7t>4g<4qV+!VTb44az7L5hGo>wjr!`0A+r{IrF)poq;pN${YkIBNMeHB`)yIi zkZ82dI;ND`1@Sf8`7FQR#0mIfGsed!xl#A$`&GXkiL4sA;qGH!FY!lJ4M{zLs-@A}z_aw` z%{vY0?&A>hwXQG;D02TyU2S|oXkOySxCOfga-OW!pFJj2uPmRUM8C`Mg9oD+vIVAd zSCrh2H)J}rB6t^h@lh@Au)WI^;uMu$&OmIrckYjlG9k}I+Ho(U+%9&(%naAwNy3o2 zBY7gOLZ#YjDnBLW7;g&6tsCk47r4Mr8JelgB3sRW$zC(4tQlU|Q4APc zE^ZjvEaK9;7*Ni^}XLr&GrlVhwoV%t+Scu!XR2c|5xS)2%mIb0wY`=I{j->A-c0qr5)oLbiB$& zBpKS;O;7GbQ0jzg?0Cr@u+iKKcK>Jv_8*4y%{IKe(lzp=F>u5o8Mn)QmgHwm>an{` znf{6DtjgZ=8mv-^e%ZbA*f3&UsomioYcq{q4de4&7ag^RY@b(RW*s0VtR`O zvu)~w2bAvBQTZ<{ZN9>D(;r`p%vHkd;9vW*H-Qkr#f(;OiABk0b432E@pwgXHnw^z z*KEgIIp1xYbh33HCTXzv3~1h|t6C+Pc4Rz+ms)ziDp4Shu)-woXr!a9q7g{_`n!Hm zb~kWeH6kC9a>Qi<*8C%#=bU#ha2?V?5L^UTiY=d*D#-kvsPxq`qJCwzSS|)$+Q@4$ zSXgS66B@7)*80=k-I7wTmCCJHz*O;I4IVHnyZh|;i?O8U96i`nQmR$)04+)YHQv0x zTCJYUKmbySYfqtE&O2O2^j>Cp?5G<$B+KSOc11;c*u)d$&C8e@1K-5%&`r)@Z2$D2 zCixR*N?AnfKo6~GbQkT?DNISadH!WZzLwCkMtH#0E^N6(i1^(s<8+x#8rx9sY=GG1Nj zE1PiZi_sxT#|&t5YNHnFIuujlt#dXUD!vo;<063^V!)T zCFGO?Zwq&z2muy7>F%AGYzK@&Pfab{7oKP8W;Y~zjcwtG&%BP@R%zFJ(n096TBBBp zw`qjCUbt$P+p?B+#({VJy9hnNuoZnzY6gML{~sa+7FvwZ^oq?3L+Z_xt{-ikS2!p5+jOe(UsD zSJX;B`{weQpMP|@Yta%Ccd^E}Y$-=AaGqwHS5@(MRK-=%z{)oQ7jjIlC2G=LS$ke; z-Z>QIm3__o)(4|4dbm;@S}ylbwn_>pMG8M$j!UhqzhLTr>Z!sNIHF-R@)D2IrV%>N zuxKP{9jx)O3U%_5cc4<7NtE`(UM0{1su`d4CaHM&@x6?Na_sz2A`r&hzEG#VgXN+X zESd%*)dQT!s{Lvtd1u}=3a$udtk8nxWyc|HC|Mw!Oow_x)OV@+v2#A@b}{#0sK<{b zUlKPxAAe_P-5)E4JCVJ;o<)ka_bdiZ_!qa^UAT2#y@C@p#+|oScn%y$;zT`be2gMF z5KaZGJ9C-OTLbp0lnm3o-PFwK34!bOkITn1w)bDGmjyif%wkz(h}RSU24B`;ed~+i zjv=b_Rpm*P-qsvX-cTCg`)?YmP_wH!d^%);oMc&)UM4#uvi!rQJ~*L#F zcHwi%OFv>h;#Udvsn0Lhf?c1tyI&rqny8>lp6mltIDJqKRR= zHe20ByOZU%WrLmE%RzJdT%!{tFwkP=5lSehU8M^GbkB$Wlz5F`{YyE^R}XMQE-*Py@!hl8UQKt5E>Ce2r0lQf_OC%4>)-UsTKE#g^g_FKBI8J?%+_HsO)0_5=n_4t(csy$U{aG7#) z!a+g3fXMQVqQDiLPMFo*r3ChoT+l9jK67y%VF%9kJAiR$IkqQg_!B#&|yiuf@kFvixjANCxytH0*_6 zNshgB4sswdEn7V-VBHE!rCP@2G}i=O57Ngb7FCsC*@Y8-PLju?N+wdfZ?_NV_XWHq zwtvc=kZzKu(Gq#*O;u7J13g{xqQYQx_3S9~Qog2_Ro$R_myIzw7_XfVb3=VnP7e7I zL^ltR5mWZ{`#086SRJZ@mVMv8N6I_i@0Yz+cJs7NRriwxD(~$RtSmHpxKvBVD{H#+ z$P^I9f(Nsw+V(u_mUeOExeXvU7Jv613fdWy`G_1}bgs8Zd~g`sqs%ML6o-_h>=PnB@2Thd!)5tR<8Pnff;&rfv)9)mek5ZGDsV|>z zor6(fHAD-o7stQ-dOK?|RL0ybQ=FTJ3h^mDn$4raBKJ=LEgfKZt^)X_){35wBZ+%k zhTj&AG%HuU%nC@_K@}qP%`|=!ag-A^pd5c#}c7ZAi-57+S=i3L?eq0}rb z+HszgN@yYKSd;{`U>}~=TN8JjR8bn1@G&iJWFF|?!3}R4y~Z1L9pi#a1wz)k`&>Pf z>wnwGkB9_6VX!Os2*?h#JUDySGOW!|CD6!A(zNNt`L>+g_x(@Bzp?+MJAF=5%+ndK zPm`c|tXMasq(t59KvFa}nsKhxXtQ9bC;=C0Z4iQnPGoN_n14cujSzJ-#^F)t!SF^IB$OHpfHfBd$guvenOExaYDD^EQ zZAIoh=^nC`jZWI27IHBz?>NbNS{|UGgW!^4OZ%-{mkho+WDLtEN&Yy;(tq&-d;0J8 zMvC}iCF)kMS@G%VPE9d;$H`7O@kSw*g0QUw`N``8N>dnt{6|+0DmTX@ohCyh;R3mH zIZX=ba9$$M&qV;a+f)0ENpyAcx_^gygzT4r|5Teuq2GsGlt@ogJD%JT_I&iVU?aGz z_{+F@{&s-y*yM{H_?re9B&F1>z)6W@dDipVRAx8($bC%~QS4OftY(n1q!lP*e{4GezQQ-Fup6G#bT7KJJP6iyXW$ztgDk8El&#J<;xbQe zL>%+9*LWV;VTalN(nApRAcQxd(M7_8--$Rl_&D79AY8Pgrc~~VcByo6iR=9=&B0lF zm@64PYFxwd3ii6;Q70+ijDIF(h2PR0K2V# zs8X0B*Rjf5oks}K?4x{wX;AyOXU6rOh1Kav+i9Zi=)yXeN#l1ZR;)$rA7j_joQbAa?%`v6JGXXiIR58p7+C$0yi&z zfr9^>7fmXwGx`{@KO5O^LQC;Z(9+{UxOP!^ene$JB{0A$k-Lvm>&gmX61 zls#XsJUWb3C-38U?BeX(Y+Gg0vKB0h>nk2W9GaVStrj-pOcB6>zbD(I5}RiiY6~U= zEmCNg1d?IX!S8M9kPTgscTs#iR%nO7BU84;HhBoSF2ha(DE00uAh;Rz@MQGq8GP+V zu6c_T@0m^}KU;zfa1xVs`}I1D@sYH-QH0W&H~g0a<@R=j)-$Azzk_@sOf$pR z1ZMrmc+)g#IbMu1aeHvT6Q2APTqpy9?D+uwopWL62)pm=@Nq!6FVxV3&B%wVWIQh= z&moy&Dr%U%!@{*Ke^L_wW9CFMd$H#tg5xEEJ!Q`{4?Yii3ow1iW|MsJFg^l5FlVZT z9skodIm7b`LTSIn+W-TWzA3l0sdXx#-M6hTYMZ3C4rx#y5c^4((_s!6sxQGtjr%61oMFmfPYi!VwTc5SEh#%@ zb;I;6GE*^u1gpYVTbe#Uuety{ zV6Nux?U2=VzMB!|IcmBAJ6AYsFxzAM$9dqi7FIPjo}KSd?0j1ZiQtlod7VA+(GV{z zCc3#k<9J#Mg>mzBhCpth4616cG(P=FUy4?iFtju7uEkD%fW3?Se3mBo9sN&}^XUTG z#xiD8pWEiX4+TcIgY|}tWCkVCGb%As(XkQ8fTkGTTD=l`p#npV%V!aG(-$kCPYG<- zxZw#IkA}<%zeO){hS$BDXRPx{b+-Qu==TgAmlWM4D^LvHC9-tD^y>?Z&`tcc;_#vM+>CdbN5tfPpp7EpL zMYz~8GywJ?yhwD|WkA`hYlC5p+?0hzcTJ>SgrBzPb*<)f7E9qRleZu~N?eaG(8He5 zF-F}Vs*oM7DQnGWOX2$(Lda0Ew84zn$7&0z$A$?}jSVxmA8>qbXW+X<4T zn<|zY%;M9;qdgZL+s8OOf16tZFwFsybVrAU%^Peg#Ox8B za`Zwlo+0ZgO5nKo-j$)u7QsO=zq-GDa5~htq}3wo&SMoCEi_iV0m;@M-``GNBM;54 zc`uuRN={V0hKAYY^AoD;6lp1vEw}7nU74$lG}6z{G)m_zvvLow57O`a(w1CbB$W{s zI$}M1n5E%BR1oe_W-z9uDnp;HKiSFh_VTrEATSSrN^am=qujmaF(%t{TITgRg4%|( z;TF=hW5UNfGk z5Y2J-Xo}xDL&)h2M^Q{_;q7Tfk_B_Tt@d!pwrWSt%8y)L2(_f^qWrZ2^u=7G_U3(I z=jW6%AAie`u%Sc==nndEPpmQ0P&}>g2DDBdSi&SWpOJm(;_7P$ctCHq-u+q--FpTg zj#?=#)EypB5cVu8eZCOPY)UbAxJ7^ySevv*))$H*mw6KE1)uaWTTHNGUXA62Tb`2B zuW;qYxwOar7CCRor{rj~IRtf2Skz?<^)ny6xsnCJ#3(R>!%p&on%+~!2MlA=F>D3k z^HF&*cnpz0m!+@Y2I12gI6BA+wl&a)bRcA*BstUH0Em>i@3!GZe5R`)X@;hJETo`k zy3o5g$cfqNk~B)1v^WAHOnY)+M;!OcwxW@!R*!JG%zrX?XcguyK8d|yKrz24dIwh* zM(j-EnkpSgkB4ogr4IRB=_yF?@_BlG)KRZi8`JnuMj@!@OXxZ6zLZS z*;mr<=1>3ZvaEL-{A!--&j~94la7A4jU#X_!?is2S!DYuCzTPrpE~ooV!xy8&xN(9 zx6i#~sx(+z6y9Wac#mSD^;;p90QQ%q+|WL-EKei;~l*TzfLTGU(FrR0dklBO%sWCo9zNv65v*uV?(l%y@G{?#%Q$=hM z>~jo&25XsW>nC4lJwA+)NMs$&z{s+tiKrKU)~O~8g_*+!FN6C9~ z;@E@?;#N7tN42|6Y6-eL>H@eaT>QEF)!H^PRF+tj4RPrSNe`n!2e1JtHUZO#28y%E zR=#94*`uolyLW;9l!jiPKJlks_2-Nyr#fvgtgWLwfx1bZ+6 z(|Trm&clXL+crt`;j=6$vC?l|hNS=Z;UhmTu!rJ~f9}U^Y>bg2F*!#QEaaw}#ulm5 z@E?c{+ad$flCLvA|5^%)&2bhalDAJ^Uq-Dl#ks^EjIz`>()M^JhT__xidndw{N@Jv133+gkg7|I>f} zO)dO?eaQdtWm2#!{%;fMe-KT`0ZGokXej^lTmS374&ejk?SDR2IV+kkNOJQA6PKxOc038~kXR21C?Oe~V`2c*> z>>U6OOS$wfL|p(tYZEyD_wgeOP=HkfnQj|kpw7lk)I-mvtpP{{AFt5Xyb0i_iNOWR zXwIqs&+BU|Z7aJ$1t-_60O&0XLW8XOwx|Q(GopB-$;GLCFcv>gqD`N>9|~_=T9#gQ zzA2j?vpj2}XwXY0K?_!WhRD^80DF^*7=YP(3Xnro{s6GbAkneqX3{LW-u5<22h(YTjlb|wR3Pf90Kbb z{Xz^`U_eGZV? zisrKc-=NnO=O9K1njY~Fcmr*k*hOPFD6KeNo+}g$14Qa)Xs**NKzd!?nB&O_p7bgZ zG4OsQIeLDF_Oi*L(SLw|+(hoYeQKf+rlEtiMV$e>>iT|+1;^5qY5?=o^$hR<6?kaq zpgq_>OaOb~vnf;i4ApA15}1}hSrXIva1^PT%!E1zezw=ARX7enpDwIM10n)PWL6Aa z0W<*zoy^+{!14DKaHI!tnA*CaSB&0Z2af&gQbK=WL&QAV@7RrY7m@((+Fi6gdb+I? zAa{$^2k6pq0k1&cryqw#XU68=D;SJv=enacuZbT5*jHaC;~x*wtFW&$G5YLH+i@R|_X6H-BY=YS#0TvRC0u9>x9&sRr#`nH@$z(SB=WxS ze2@L2Fyw3(O^s*)x(lzT5)3qJxcffoZEge1@gJ{2r(f3qbK&zA07jPiKs_*3Pecv; zuP=du~>)4-RVq>_Jv<-QU+X1O)tHCqC`TR+WAx_0sa{k-kJX|sF`CC(j zi?$NUFfh*B3FKA$=IsDr)@P6p5K6u|+l_(m?E=ko>90gwQFU-kSiQuu_hTObI`%p`)~_iI2f`8t z@PXa}Vm0v?c&biaOO$>yFg_^rA_aJTl9x#KG|JwwkB!|wOeiIRG}E*bczI)!;@S4v z5Kj4jPW-F;XWPa?%>YQo2#wZwZ3nzao`g1u?w`7ki9etn+_S)dwu|Ny2DE3-lmqnV zlbYI*V@k)2(W&migTBU&unOPcRqbQDXJ3l z&7UO8(~@EI%pLRd~`{GN8r0F?~VI<<}cANWlyPnIFY6LM= zm}THFPM(-H~wGD;ePay&{a9Pv*yMWNCR&AsM{G@vWNnNtx6f zVdgNk+1|??DRW+fgFO83rGqPAVw0UvBmkcJ-Lwm-<^dpgzmHcSYEj+*8(8-(2$G@ zSDdO1LOD6GGP&H%eztiOQ|tbz%N3P(_29~{P|}_Y##Oox0HDZBs@YCkRjqO#BHjH# zN3~?f*X>$4$lGb4rQihZQx;J7o0^{-462)y+)RFC_@X7fhgMQ)~BNYH%P8=wD0yxV5%=&+;f7M*q%dN0@`11S9xes(f2&LSI5e`xSGh}QB zZO_dck!C#1CoawYd+}9WSZN1UadK4YI`*U)>M#OW>!1CeQA_&wq6)=c0t9WhSTcx; zsJqr73T?{2u@j$emqY*VYe&3`Y?c(L182Qh{ZKQXdL1XxwSMnMNDJ0Fd(7XIuLbuf z#OxA)4bh$7Vdao{Z|Q@?O)F=)7~|saxt}tJ-HihyHeM$&5bvcu7uwO2LmP#o1pU`X zTmj90zM0XizNnpfug_krUOzQU1ni89*b6L$l^u6G=V!%VJ{;b3@hd%yUReaN$Xb{` zzJAWAcVIZix$HZoEu5s)f>3|~h8qy2g(O)f(~NRHu_I=mO&)Y@^Jwj4IDri+W-A0iLhk&vT%#YH^iZ0HZJ_F8;;ex%UXjc}Ojhyj3)nGPc!c z=A|iUAHQtrPA709X@x<0E&o(Mm1Fl{KPqQNGroNG8b{uiy*Z%GE4=zzL73m=m}>`F zXXS}sDX6*KyZS`7f_96d`)$u<$aoq4RNVu}SC$^%j{?z_?^r%sya&!4C(Q{3#J|ww zt52sr2B2P0=P;+mS%aTPv>}Zv<1|uuAvsp=u;j^0kI%QK`_w#+S10 z_iHblwIg9N&9ls-4<D2|IP`l&QMpp*#}=nGwSGWlu9=erljdTbdCfp`5p6?u;k zO}-okDGsi?L;?M6+kX&*mm-tX{VliUL@tJq{tAIfN#nM&`t_=54jn)}Y6rrI`4>Ai(+Ktw`sA_$@PE>V=GNDT@q zkt!g)BN9NGsPv{30f7*t_W%|`krt!|C5ZGE`kuVs`|a$`?vMSs{9%Tfa5CpS&w1|q zzFOO|(x>kqj|8ebQxU!If15@pcPcI=6V0*)6Ug<#>Us|u))(xV(ViR+5A7VLgQ+&C znt>Y9R6Fm2>M`ohLNU3N2IYB??Lx7ui>ZNNrpC?+D!DB;NBYlmzyotZry_9fPiRi% zz;K!--&_%mW=GwM;V3r+_9NoJKXGo$Zo#GX^9@z{4I+j6v7+s3mfwwgO7x<`WpopP zuniaMrQe|oar0CT^d4I#VY|l>G?(6taED1|`f{W@pH2D^*sSiRZf6rbdR@tSa#xQFkoDE!k;`Z(F2LW zy|{@V+zAvY)@TmPk|Jvrd|k*-bpcemw*@=ralj>Y*X^jQl2N z0ObKY`#)Ql{bF{s(WmELTb^`^vHU_gPNKI4Z>2G&1q1=kY4-n^+8D4#jeOxWNWFkA z)@LWjGv3EQ;|yci4dcDaYHxcFnzmS4T)QA_pt187n|>_!9?Fh>m3A@`MvZQh8`s`! zS*N^4x!1Qt+R7#6x2_aFC6F1$$Vm$JJ|CrFzOCMgxxD0N^wwP>v6u zMO7iy#e&3RImf*ZcUOMdSHVA#e0j$J56lUlaOKHdahRy5Cp458C*$_$TfOp8xF#qmmnI}pSIv+io-5uosnGSh-;j=gP$xA} z>aNS*S4`L`@y_=S6Aob^hcmN8L~V2f$wSV#(EB3t86ML*lg6j#jbq5JxAYf->ZcB* z>Jbw1j^ch3{!DF76ET`eN*Z1-P5|SjubSP5y^`@L7Jv)EymZe2U`ZD)(l4K&TM&x30! zIH_y>r$%0Wco!O@NK_xZGbz$&wV@l%|B?&$!-FdC=-WeIS^d0xF>JY{qVilLuupRo z-4{N*7~B!p+6m6J4ucp?9)0g&vbF5`1^CleB>^pqGrqRhuE-5$OTRezH&~^OD^Mxz z8Ro)t5XrNX#2wI~Q4DVAGOm7gH54dXF%fHq>S>C5S=d4QWeUmnc-PHyLtAJ(^@+;q zsugrV;MLzdZah>W!MX#V0vC!Gjo=KY0J#Zmlkr1dqnQ&X!K1I*qBW;@w~M=m;pD1) z;iMeBN+OM#7YlYW-N_p$kdHafF~mFWt8X_awa=UEN-x+Rg;$0zbO zt=-@0YXGy1IvyOOAvH)yg>g;`>KY2q_7+o|bc}jT((rGD)NPx3C*ZGicE0fUEaZ8F zw3jUXxdWJYdejdafI0E$3w<{5^s&xK=cQ{W{5eS3xgdkn9mc9hxnDaT=$V$5x}h@o z<4heaT+1##V6^?-^E`1MH?I`u!`q#DnbduGF{4~9(|eMZe@ExLvR5V3t0`p8FBJ!n zd2Z^fmQ$#!7UjV)9Z&HIMc~BC+0f;tFs)R27GaJ1*vtEB#1esk1Z;L>lMs{-Qp~ec z&cci3TR}_Qz!u%k@LA^-wjcWO**9!=zQ_2INk66fwAI6vum-PNhN6KNXnI1R)f$#9 z_rAL!Y0UTtJ!I*hml(Y$S6`e#ArsKf)Ej7jeIM`noCLRoFgLc~-6?81Rc$6%Q#i%4 z%F>rN`wd=24xp$d(^tYBQ}sG2X?j+}ikNIUzBt#qD`uW0Oyrz=a?G->lk`&;)Vo(e zsQKL|&9iOCE6XBT?zhSH5HOx0%hbmK!s60`YmNWTl&IO9>T2uO96K&$@Ho0we_njq zo9k-I5NwJvs`8DRZ_#6Iw(T;}F;oraRpiUM$njjlBHak|X}PZju$7gK7hm3z7j8-d z(ifaH>q=A$xNQfO7j_h|nHDZA&6&|4aEf&Kq-lz<=3bG&soNA=WO66-hA(!LX0o>` zg;iU#H-^xmY2o-~|MiJuSc4$3>mX&w$3tNrT=!L2b*d(nJy^|nNMu*;XsWC|;^_IF z&lH|`C8aCRadl_sa0Y~QzHs6l6KZz;0gP0_BlDyp3g;`oS2F}?#0%323|RZ3rl|yO z7Mgl>H2WITmY#)Cjgp-;K))CQcy#J#_GP)y-IS9Zgq^Hn zUA0r=o?!V_R-7#}t3`U z(y*1EGp-6h=}t1-w8@Niy03!eqD$35dL>a&_yQL55@TYdLlL-_;`7Ep3uQ0kY4 z1Pp`;qBQ2th{WsTVuP{VFM=*H7e=#2HQ9PoL;WQNcGng5Ay1 z=HEr>7^X%P)sV(+{>@YGDHH#`O-2u9a5#mVx+e)E=`b~rR9CC`ez88KBy!OkF-5M4 z8VBT(CD7CkWdbTrwJ#Tc`LM^>(_z=5@p14)*=#{zox&oyWAWG@;dMoUnQ7q~7#UYn zZ(2N5VwIGZ(!#xKCnR4cq#fR1 zZs^Yop`=n02jaH4*nqA7qO!4zSspSBD?H`;?JzIcP1kWAaD0wIIjag7ciF7H`*#@eg4XuFPD0N2Imhmr zUz6|Mt(oi-`Pzwx)?5g^B|wGQmL8Zc&ZO5vGW`{_N^2Ks4Z_`T#-8ZaqXOi zk93FMb`5Njk-O`JI0F_*&uZ}@8@zirLD=ixeT3O2XI}1P7l}!|oFn?5x$5F-v~`TY zTPG6g{EAnII}FtBhC2r-Ekxfu`kM2Ct?!m ze@lUPwt7EHzIu6L*)Gk7HE#U5 zaqW>xRum=}DZm1%x4|cD_toFJMVPidWD<}VHr?d7?5Crau#>g6d7Tqw^2xZ^ zxxHiTPrSE4HH>gWHjm5nqV+`%>(#PV+@>$O46qOROq0;si%A~K8bGC)h%g=yOGkE} z@3OQN#!5X7p#~p|IA!(*2%!Opp8&>1hBK zs<6EhhsQuYqDs!o>EY9Au%zhw>=IKv5mq8z{XAuSEj7`u%i$^=Tp4adKOa(B1jxNo z<&7aTW?l1^5&kh$Plx^bf~sN^`rsK2{o?uLJ$dv*Mbu7&0Xl8|MxcjfT~1B7Vs}TP z%g_&w;1k9KQ(@?c$bzI<~fcF zci4@V*0A_Bo^u!8uMTaVK_`(N5IA3K*h5xOG|pb`Y7hv5_fBs{o3LJPIm;_ZRn0* zR+nWKS26vbCvPtru}*iHx9{3IXaFcgrdL8AkdyIr zA=w#be8r2-xcX^K8c>B=>$$fWJ3bP%^6m6&$w2KUJ+AG|8%eTsPqlO_@vu_;Q?cbH zrIgkJa!-vrDxMr^Otr`GJ-^&XFuN%h00W{Zp zpX4kBdND6?=3a+&)Chnq-C*3+hyvcrJrHe(3*|Jg>G!3JSJQ>3EM6FM7$Ftdl^1SY zP2%Z^JVCHhX1Y0A_)J1}agjvw1LV_qVFbhEL_;l@!7Woh-{hsaTTD({gSLZkQxn3U zOgk>%gGH8)p!tB;FqaU&x@FH88{HKHDO`j0bO44x4YP$*Vn*18GXrGYwnG)>`+YFG zqiR`GlgM<6kdA>2?I7EX{OC%5qfK~{apZB<*XQ;&zq)y=1nE&e~9Xv|83s_{#Y+?oHc`}f9+g{b~bEpT~^;% zxwk}v^PQBqwVl7l?~t3x#L~?Pt35cpX3lB@*hEC<2N}72O6)6gyyehUy>s_q%Yec7 z%O-oEbhgEf=`v4H4&>BXvTI8!yGK5zW!>re8!z1NCxJ5Y(sagWFTKXOvM!qvpVZ%- zRxJ*S*cUELj1tS%OIEZod#GwqLj}mIFNaSEPM>t6x4@ghIWI46nSRqtoib1M$j=Hg zLbHh84E3;`;+ePJStHQ4{jouQ2S(>1{{V1up0wTi`nFjVD%G$vA^2xy)7;q|v24DHQZaqlz5pl7o_kSxsOGl1VyCVwPX0|1w= zekn_zr2R$_QZy`=8ARFR;}idZcO!IpCbK`n;+bP_K{*CJ4Xw@$oNv8 zX290L?al8SLyS9~&mikz?JM%#xCK7C?5C&NQJxp^&sE9P`4I>Eb~D<#*oQ%9{>)KL zSw&sE`24GF%0`Um@9#_`4|(AA@}@!8$$VcdvEcmi@=<&&PS7*oCp0Rjm*}Kg7Iw>J zmOf4@_B^hJLTdN`yvE8aG`x-j-cAR9@OcTLlwY)8$JHW9+sfqhl_C!L2<;SZo7lb){hyzSlXv-5~ZPud0XpPyE|K*>!ph;Cg1%-SS- z$uaboYS;P_s6jTsKdDgWGz$GKpC`VtRmWM}-Q2)GD@fpb_3b648@tOxDF#u*4o^(l z@CPU{E4W@P+c%Ur2#6KPM6}(Gtzyzh%F(HU9)A~%%@b2pVGsw5yuGT-deaE7SFcqs z@sHn0IP|lUGzl!q*plND*b-d4^bg+Q?n0slFiP<#%tVg1QdcQcI~Aj~Rh``Tou%Sx zCMv^_)iCNeK`7;wbR%2?f!-~4>C zZ_a~;&r@rPS{Uu%UN~*D!g&;9V5MeqOCMBG>2O4f0nC_NOO(|HE3j#8CMgE(s`eeC zsc+NWXhX=0dlA$xl&h|?N~JjN4AfYZ+1|X@OcQ#q_hA+%%7&iCNH+@?{ujqo?RuRCy&dp_df$06XV7D(uaLiUcY$71=*>p%Osd!VfQWAx z+Ez4A!hH=!*v-<|5$79R0nUZntc%Z8ojw8@4hFwfc5tIT^brH`Y1<=F-=evIjyki~ zT?1(_l`=>Hhd4&iX(`R2i-o)mM!anHDrUjT(2qi89Po$efs7Oi?CCBn0csCTVY=VX z0slYpJ6=m9=F}|3hnuO<0*C#E{LzM*?WibFX$u@M<;KLH~r7pEv)ZpW{pB=zF!{&&ogwBaXuI}N1ot05fTX>*M zhASUsteBykgrU$B#>n43i-Mol)%9Kko{c)NA!s!W$vx%)+%-N{HN9z(u#@CV8gWH& ztKD*sBJUtERT)~fH{12nxRvsNlnpGj#pd)gKBXWOJTs+9&tbYlQ~aP8L=)a0zeE@8~WYTY=x!1V%)`y(KUl-b?t{rE<~ z%ZabwTBUM@DsGlvY7Lkjm$2ukjVAtigI{@VlfJ8DiKszjPjt@>=ZVHo9^ zqdw-1CDF!Q?8JWaJ_>yLAkS|#Z6|1EuW0ru81;+r!($WNj(V!pP!I|uHktW0vna3& z^RhW=gX@#m*9HR9SRpsjmNuGJg%z59Ta&q0EsInIr} z%a~>`Ce=eqXAj9F=84{f6|MZ=`)d#eE} z@H!5d$@p~!(O!KZvmCw68mYXk*%Pi=4r+Ve;xirMeshWr zm+)DO3nl{CwcM$uA}#r0PNc_l+*OSGf|I~Lnvx+Zne3r0`KME@R%Z#cl2dR~+a1u$ z6j|Q+Y|jf6RS%>B8!mR^UbF0ld|mxy8TCabb{%!0n~&+^>`HnxrE&XYxlTsj8ZSAc zqwjdq%Nj-31i{s}aLKm!c(S(9JR$-r`ktF$+X4?Co{bMX64`=`Hr#+!l8`*{868+&k4bfLuYCuK0ZHP~IsFX6B>$ z9z-=XrAj}s3Q)Ke zI&r`X*WBnuRVzPUltmT;>NOsVbfx;A_^Hv=i2WGBl^V zmyP!OPa`j@rO}gxkTMKKE>ljSgP+a{D)RfmQE}P+ou;Y%~Qn^4U zXDomA`Uxj&eB}iV$P>4Ct~i%z-As`DpSkz!fdPJH`~^(TX-}oSQ;T%tQmU{V_a~-HQS6Wc%Drx~vhIYJ zq@V#3YbOx1fwKcr7jpNv=K>t8*#BX=07QcuQ**&BYy(<=2p`R9nd)pW@^=bKn_FVP z_G`n_4enbu<*KrS8Obb!= zUTPmzrQvkDOVE`p_2X7Y9K%vvROlFg$aEkoksX`J9VwU%35#;=Qt};PZaT6UepF)1 zzOaNIXxSYytstWc)fudee^^J zuJ246Ml8g5XaYk6;VDRbuX2-cf+4*s5}!31?FiH|I}waP$8?XNr{_R9j@utb5JCCNo~q>c&L&OaA0I8zF$YDGuTuL zuN10B>>DM?X%sOT)Gozs@IFBvsqi3>N*k}8`{g>;TGwZ@*%5O6d*2V3*KRPtW!F0& z-P#g17JJ+AXe;>WSudw`I;F6oAHW}yk(H|weSM-1*wJ#+aPeCr<+3XWM`D+2ndbhX z%gC)y?Ykpg;>{KOWp`aJtTJ3&5HfrQ8TvpXmGn+xH0i#fA%!jpmGs;IflG%0t`nx- zzE#L#Rke^91IGlB?+WIc3#fmn@db1|H=iH+ zYU$wWh-P8xQBUrJv7GG@UY&=!qJX5!NXrihuZKVQt1Vn~3?rZyrSesE0=Vq1s=BU!yC0LJ2tj)4N2*&(V=kg?XP% zul(-<+7_2RihK$oiaY0K>koqPkz~XZAS{+Dd>--NqKpFS9csJH}do*H8^%vFID?WUG7J_lF$kcTuOpu8~Wtvk||ZtId(DCuUcLdAp$dGLk`gO8mh4*3#;3$q8bW_PDBVAG8rP zcU%r63D!b=-KNA3yLoFvqUkM{%k_57WE5)hRNhTV$Q}J;fGDVg4b7jGxtQAaRs{L{ zv<8CvoZU=INxb|qQV`TE*>5NzKZ+g;Nrjjeq;(hZ#~Za5<+bicT`DJ%f(#aIepuhS zUVz1{0tzTwJ`ULy1M86OOLDSa9nfjCJtLc>w+ry7J;* zBJvHem~T4mtqzgPvQAeTyB%14x&LMoZ%LIJCEXMXX#45?RlUe_3rMDT#vjsMV=(!F9<*mDn0>Z#hvVk< z2O!6@FqPx(%14ABg{imF^=c3-u#y>JZ~a<9TD3!ki%VR4*=l$ih=> z$|(kPWTf~f;W!!Pc{>Q`XuQVDJa6;m+e5_#j{)Lg$*us?kZ$C#YV(jdNC^?IgW&Z) z@eYk&;%WLULg+luVL1hGGl*>BHzCx$Yh_Wfd|&T;fo+767QD z-xpcSd(w$cr6xL_I>J|k8-5zb2-r8i_L60TlwE#~Z$Vdb@C@*YVS59t7Wy0?68^+3 zzjoKQMI5IO=`@6I41GO?bnjyKa;-J%B?nyZ@4nOWzI4_&>tp=fKi1ltIlpEjbOur81#5DvuW{VU?k11Hw&HAHNx{h-Vt+0 zzk;+WYJaJW)mllWn!~4gepBM>ye?`phjcpR1+3l2F5cViki0gi{QAez3Ao$Q)&$0P zH$!b_iuy!%!?Q5acujSa!(mgK{8ra1dz}ptNt<*BJy?W|1Ynp^U8dRuu@`Q%f+yudn;U=j55w}J3{Wa|FSoZ=ObCW97nz2=G-`p zAEcN{$+&EE+qm|s2-O6ym%YNb4b?tctt`kjkSi#`AHI4(d|GNT0zjYo90J05cz zXy7y8A}f@*A4$b}s1hcF)|>J4At^UDzcD^ZViIzww*`1Oa`P4e%+wIPUEaG_kW`{0 zL7pHi&U12Pe)E%8z(VTI3KAO`B+dG z9~EM(&lFPEUqD&qHN@mHM7+j~G={0%Wiez#3J%=Ec6GkgiU?iF+u55YUax}GK;zgf zwpRSrt{X&7ROt&br%*O^(B4bo-ek;Hg=Iw7xO$x{ex|_@`E!0o)p@Apmz=$lKDf%2 zGnyVkcMB#cI@y@p>jPB5Bud*2YNS8FW1*Pbjywil-9Ub5JH$3nJeyJTB~xE+T7C)- zoZVm~Y6i0pI6@Bo&Kg^@cATqY+PisM-@C{jFo}F{Y{Ij3SIq+eNOmS`|C&(69)SSo z#a;g65uchppsL%W%yLPNwwH}bOO+d3vw(`P3`{IcY?-N5=#d{8Nqy!Ph@HzHNqiDnv<4<%OTn(8E+kKQPxR*TyL zF14PF&5a7{u}&DEx5%(aL#WUs>Z6e!*o)gv z*#rEu)PQoJbg7HITIV!tpHT{jFZRh-qrm;gJR)c%lgL^CYvH6AH{DRTn zi1Rd>HXtVS1FbMb^DJ^{B%JNNG-J%cK9Pz-h&=_X{JN-G1{zz-LNLeH{N|9%Y zq`jDWI`aA@--eCZSoCp0)^T5emnJs9J)4#@=6Jo?!`P-G`KtU0MfvM|L_I)7N25ohbLb{rPq_UAn}4=Kd^kc(+-k6#w{<2<%#5}tRbI^r z4`X1KpCA272>rxf`6cC+x#NdYRAs+OS9kfviaAKiWH0Cz=sM86$WJEYhN**UpPJrF zjLA)<3T9|?$kMzVaaq~%iFSPd>A_sPn(TCLYl28HzhtC)t4vCl?3kjak@pK(F>yFJ zh|Y-0E9Y{G;Fcg+!eG^Mg-yL%Nsen;>8CyQJbDjdsdt{0ISP>HlcU5wnF;25(}E>w z=cCjB-80^nj=y5JiIb%;Vn%ET;O_Zs#c<%Z8iaE`4&eY+J;QgE!)NqX!`$;&oZYChs;>Wr za0-KqhU8CV+Hzb0jF-pG+@aE%X6hPd*c5aZG^D<1tO=Qb1SGD(^vMb9*C|2An09L? zM(bOz0966_AGT*2eGXq{jlp!t`?1zOa<-*R1p?)m&1?u}>Ose~ePzi#lRo>ae0E)I z36#@hG<>)CC@|YDlugrt>D>d*_hB*_nMGt#xjw5o@EMSMQKwDva=eWEvwT%oT=N$6 zUAdCw8?ilFCo;#K#ZEbBbmHBa3Q_uj+f14+>jw6{E1rH8R5g( zbNhJVQ~zX98FTd5WKL>c6kt!a>9*|40zu2s)mJ&kQYiU>o%`&x6B(=-xQ;$&WZT9l~LJ0VHeUdS8xDAgDI52crU~UDb*`^5c&zH!@cX9nO+Mm1T2uws*7v~({X|k2`nG;=iDy}IrXVL^J`L^_s5f&WlgM#5`uuE)SF37_TG;b zxPID`z!gJwVtA4jFRoXobH+?*MqzhVz98fjp<#%$^ZonaXQKyL-T7O@&AA{e)XI;i zLuie&LiO_^xpylLkQ%KZ0x!N-xqENsaJ>kTH{&9dTO)2eEQ@{2__z*;J`A!y9d7un z$h_<6u}n$CTXEuMht+%4epayWfpRTnHHz-cdZhx#Bl&G+*ythNoczOe9 z+5WQ&Dd zZQ-RhOMIdt)X-t{_NT;tlAEvi;NJkPy|U=f(NfR%m^(>z3BV>8h(<4AjMy&;>7z8& z*VXGw50Z1-zUoddPF=gjRx-myxp}&lV`*hDHO|HX!);gF>eR9um~b!O{0&qBTmUJy z+=6vE#HkfBfO{2_HcCSjmUB%olVh3AP4ZuWyhVL4@i)V8)sh-fp?`F>a^bgfI4!OA z6|BAd|CD`r4Ui(Ksv=E)y+(+;J2++x21J-w{JDl^s`>ua<)&g;d0k#BLXT(rtlD~R zk3Na#1MUOD+}%2b$APDs&nP3goLzVVGg`oL^2r+k#6e339ZpxdD8PmP6(-9a=6h}f zhM$rfzRt1|*-N8SfrEbjv*OQ4{>papf17|W4QJR$odyO}?>S)igaLMEYy z@~6B40}R_gZqEq20D{crZH<0_gP?r%e~M&{7-=a9RO|2{y4d1MIrnbIW95uQ`)`Q6 zP+9%`(*W%*9dE+0p@4Kq>bJU^?Q5JOk`NbnFq^F*8eq?m+gIR4lAhwoI`+@vAkHcwyq53I=WdJE-oy0ufVmo{WDtL69 zEa1HN&ZTZ~V0aTz%<8k1+ZMxhuM~37gvbD-fFjJfd&N4uGe^Jvsb1Y~oo8Rp^P=Q= z%}l+Fx_b1nV}v`8@&19Ct8>6S?>n!LQPvv&Vw3<|XDN-ON+w}2!1;8c>BtmUh1qz51dk{!v~2-}LH$Bt4|~KZW}Lb^+4%_

      2. a!sxeX`R92LKl38@g~F{zaOn>ihm5t zlYKt-^09iQP17kE%JST zS0Y`sk{?ljvGNR-gKuJwMo$KcJaNt8!U$+I-!&Jqt$F542@8T(68O-4z5_xeRl?Yg zl@X$!{G(LSVFb79zoat`o)?lA5&x?MUehKD#puZMukpdEh|I}~)8O~~P&{s<_s-?) z_{eIr#%4Sfp3tAe_xd)%JO(@n5Ip}d-2`?-n95W7LFr94wdu+-`!b4}g}-~9nfr~{ zl0{KE)@RUkWanTTkqy(QPHyzvAI)Xmpa!Q|)Nwt$P^RGr-rHBMVx_$!jJCp=5z^tO z8DNQi-DMA;l24s8;??)f+Ik{+Kyu>4r$Fh-Q*?amR-(a_TdK=}9uPkH`rx~KdCJco zG*9sXZ&g#bwt9w*bH&A7TEh+V+>-+$-QS7IORxXp0peNSiJoqK+39hZem0@#sgHF& zL45ZbQMaE5HUIb9JRXYw10A}at69CF!zi`NvAlba`|%_vDI`hLVp$MU9TGuDfx*`= zEjrP=m$Uh|9BR{Th*q*``sLSX16ZC^q6_d$8j@h`dus3o4d}s-4dQqDQHpSGRe4`Zo-9}vo+1yfjve9RTsG3_ zkhaNyimnXhD3cdG52gncQHmr&SM}@Ng^7(PmF3DJ_%S1d7iK)BX@6T#@hoF>pJ02! zXae=wme(ZPTI0#%_owuaVmZql_P!&Xlb#zy>29FuKA=6AhY^e?2gCs}69MDo1hOM+ z;{cJDs1g_Idy4qo??*gZne?^4b8o)-3nrbg$t$Uz3Ac3TUfJ6}eep)*xWs&q%QB3S z=`BJROrjd$QbKVEy6oawM3WlJ;zxSN@B00$=5L|~g)e)c7$q*Ex+s&g=%YAO+BCSU zsOjSob=i8QENo9(p?FUJE(|osd?_zVs03FVW7k*16L~HK_U7x55C}%?6(5t8t9g)e zukK<}gAomIt&vV^y51p8uP|;8JgLH|Z_^F2-(*I(kr?gq0o$?Ducl24o9nQ6ohMjQ zczb#6`b%aMjzMG6)88T)@#NO(lTe>P* z=A@aZ8k%s>sDe{URYvsCTHj>jq7kofwNF-eu`SsE;q!051f3x?8w9AQMFcJ*qkR#V zr-UBXM`qHo-x7FC?J5TpZ%Gd^#zXskai|=4XK6gd$H~S4Rey#DZqK|n50&qHuZeneprrK8J#F1MNC@7%3cHC&TFv{}*jjBO%z9Iq!~1PD+(qUNpl zaJ5xllHR3SibCg;`#t@el>dM_A05OgK5wAx&q;op3>EZUjpwxx0?xq9*DPB_R;6?V z)HwmdYCA|z7myVcU6UA9{^~GSfyf2CRBl`vw?&K}GtXP~&k+z!gkYpAP&ablp*HAq z_5+E(B}}KCur!b)yN&#_)Dl+HZ(C@2{Kboa%a$2WJ(|$RFw#N%M6k;6>exU+8&B5k z2c3FXY!~bYu1R*HknK&<>!T=TPMlm$x;;0hL}804CNee12{KuM`I7cDdC>^31ntf$ zT~y@xwilHgD5Al9_fq2nKXC<~l;g<~XX|DlV7bk~+dFP(lXU3$U9ilq=uP7$fi|VU z&&c@xi2(H_hs^rW0EKbV@!)`oK69#`h&MuxtBq&>io6_f^E@ai`eMgEMr~}WT8m#| zm`06q7JRvmc5klZ2}L&dKn-UXZHxGzLg^>t!40qH`2TsC=OLc{lNM2am$x}{Yj_iB z_A4y%qUgDfcV%5(%(~;jC8=N}=6wGxiWs+b zGHaVx89ACNc^#!x9>~|ezuxiqeY|qqydt+nq%{8F0|m{8-rt>(A*)$zM$V>Q@qJ0O zDGZ8$j~A6o-K?Fv`)|Ep`IT_y;}}T!1$flu^=vPuH46O~${~$bNYSysD21ka_C!_J z->IdK3(@d*6J9PQ%_Qf@syEt)oaFdtUZ25WSIXlh`SuMhv0EjpAcv@Rg0u-c-Dfp(2EX34&?CW#Go&v&mWLoy@H@EbPV48AYbHT>H*WV{OS$-U zH9XTeG>R3O+%P9(w4B7a=(x!;^veS;s`b3fiB4Pak@O%Lp^(4J8#;7?z$oG3B1OrG zfN(Ps>}L)1H$i|mA38*4IB2xy7CZihfhSW7(Rqpqn!bL(kj&kpl~Eb{+44K*a`?3Z zASSUti{wxn{dDK&xZlh>?*}3BOAvZ)W0F)7Say2E38{bOHvK97i$FsCm4Hf4dhwr| z^)-iPu|MxS+MAtStirdZt-4py*hpR?8h;rX2=oV^o9T!>3I0Inl*V&D=%5a4iR_X_ zj3?ZFv>Zc*0{f;3Sz~eMJ;vfd2s}xP|b8uElGS<+Z%K}i}om3(np0qAFAMHJx&Ud`ffXr6H z;jHfQ$x0ckKYRzJzL{Y$=4^k1DW+i@mmhIaW-G49g_FgynsN=FRoG-c8^e2c;Y@1( z0?cnzdMCYZc_5CUTObdk4_8zTl6FIPcq(9igdvRH1?9+8`O(G|N?)LhICuzpkOHJt}<03L&x8KSBhW((r3e=Nkx_ zlrH-@d+%WSPl?UlXZ?o-c(>MPLO?k?ZP;J{#RdsAXJ5u{laYK0J94Al3_|ExHa)axYVFNlpA7}XSef}3LV4O=Gj z#r`V}okr2^5H$QoK%viy^mk%9nf@>-wq4 z!)bG_(=-~5>y#?01NPk8l(Q>R6(DufYzbiy#^y7O{I&woQF$fsJ%m^X@uvh#MPw(_ zIEI=cMSl@m)D1I6pDQeiIq@s=lCVy&)qr`xhuL?l2}O)P{Eqp#;frS!s*R!!LD_th zx40$KF=F4@@;!2x#)dYKfc2!^t@Eq!jaj#Pn&MVcr43@NSwe3jBBGAdsM*Ery?}bm zL*${Dh3g_7Tyr^0g_F#?V?uOcN*Cz}WZ7>ua;$C%O}N<4=MNbEiNle15rEmuND%P) zcRl`vsLauJ9M&jOgNpnYI+owqNx@u}6l?WY>4SfP2QGcCmr?_+|Aq*Ng}zJKlD)xu z9g2@ZrQj%jIvN~~FZE$Db4)Y|jFsnvP7A?U2+_cKKnliiM4~6MH8`5Ld=Ye?1XHlI z`ML||a4rj88{}^e_7Y2>QEiSo+foqd4~U!z-&-C5_kk|Cs||>{=k8FS?ItZmfg+Ta z35*WK*JC4$afDI$QyOvSoE18)HIJ@g9iQ2$8%y3Ea|vf1u7=CT^PWWv;|jAqR@r(Z z0&|?k2pSmpVS7vijjI|xCr6V2jFDJx`wp&}w<$3b+$0m?lTC#MD84gn`N_Bp9Vk-k z$f~Wi%zpW6{d(j2(h=kg-0tI3p_1BxxIuZx!)N8md~ifXo_Fa#b=wa0+lQikmmYTi zA8^ayd*v2iAu98y?_O?{+X#cJwuXk?Lw_2A(veV3CD2j8)#Q`(wtZ1 z{99zn19&o||K^tu%}YG>(4Qf0x+BWRuP#^fcrxMyhOEL*tl+UjI~ofwgZ#S)F1!cQ zTT*zuZ)QP6c}aMXyYo)>aFYI4q-g8!-vYUNP~0;47__h=X`&k_YW_;W?<%d_Oth$>s~8y{=I{_zLe;Xm5&5Z=bVPPGiQ2d z5_dA*BhkD1f7_Vw*p@$|th5%tkeQqFYU9d0cM}=6$>GBwv$sc{0$03gEkWELqoXV+ zw%&#kgc!XsN8eDsduaC??l{ZpKehr5fzh>P&x*dqx4V%*F?9E+2g~bv4X&Fg8%3BC zEKTk0`4OOomghI$j2yzzjG10l2UJE82Vx>koX-g-y0@`~3U}W2Ij_D^MP`T52HYsn zgn`PCNL!?HlzIAeolA(8wYFr5E~CKA(y%EDXDN%7W0g4Ca6;N2yV`NFh8-EcY~8*Wm9?4vIl2-J%HN-7 z3n7H!;6#7vYx;%T#kfKSkNYnm-3`Ldk*?&gct<%GaKOZ7j%G}R0;#aTM#LA|kQhG{ z>Zm`a2gMw_IPnpojLL6(pvqTAoi z&_B@3L@`Gq>f~O;8t&)1m6gWtjNN}!J*sj+HL5_IUEMiHu=?r9Z;t+gEc@6?2}i^4 z!!y%kEE{Q#mTEG%G_6+~W8eO~D7|V%F;ME!oi%Wa2BcQ_nJr+tC%vfRxW#d->#i%T zy8=}%2m35EQ`(Bl6W5a@)h_*cEgkiHMaI`9d={G1oNL@3@@WN-2-HYL{rKT3`2%t! z)dmU+^iPsM5fD06YcVnw&3*a(QeYAY44phQPI(966Afy^pWgX%Q(D26pA0e#$8q;0 zp2Zzt#a#%^EhnDj{C>h2qc88UFAPdQ*)1p;-|zan`+kaSz6c;FWf`}$Mlb9Y3VR?L zBKC&Nn%*nB>e9MqT(@V-8oMwa_U@0=cIX#1R6~5BCU1;Pt;KNX=Q+)vVl7ViDKN%zNeo8Sy$nToEXY!r`>#Wr;@%ff0^I1(tWaj_%ii4eImvGS&UEvj^T%8 z#LbTwS>`t`8hwvdPXuost3-s+W6(xZD#p;*diC={cb#hNMgfx=Hw4w=1#2~|uE zkou5=t2zpc`?vf`>p4Yje0si5i2Ao~W4B9CTom~&Gz(cg3{m!RWtErf@NI*{ZMheMq%cT*F*@9HYcBynefl4evFH(t_&=>bdmLZkM$i#C@|I zQ^qVT_kFA$sv<>=Gbpda{)|l; z-!qGv7!U3l-%*R6$eY+xSBPe{gaoqL>tqN^Q&NA&Mv_)FJzM9HTBO3pMp`s8m3)0f zbV($wXP>4xsdtn?*SjX^DQ-HB7D7$%jvSL^H$i=cRXKdXSKcO)5LM3Y_(MeGLHdg* zN>UuAGURH7btLKoxjk^@#b576HD6rN`KUwXF&k-N%QwXgR*g?c%;R?bo9+XMJ6 z^^>`vC|fgqg@<^pBPb>vsgx`ic&En!uYz7M&RAFPU3jPA0lDrkf!9;nLJUwKcte)C zTf+3_!$4u&au0xwkP10YxClV*ZdND{=Hw_IWlp#snckS$#UY#=^&qCBw)W5C3M%by zns9H8rhecbn2uLKEzf!jlXZ*>`RiNbfM0xy5OQzLn7mv^u|=aK7#3n+ZjH zJ`+raB>#9+3SqM6z@!xojhdNBp-xHcXT#H`TU2`#NroEZtAm^lnXz0=Fr;0a>Uq7x zC9ExU*d}rLlNLP)@g|fKg@@%j0TwnIGaX1CYjp{hHsI{;=1>@U0``hXRV>=eOy5Pddyn)7-DwV8uoA7+;{eu0%l}1 zWM&^!orPS&u__@0h|G{mrwXCE`LFKWxVW!5NTd7YS*4MDE}>}~(jR(;m%Ho8qF$FO zo|1GEAm)AzWtyd~y*qjKc{sZ?CXvY@nM**og&4-#@2k_gC@0h>jv*LQ-my9p()Gfgie74q)D%GupG-pF z>P37o9y`HzBI?Ihgi>J>Za`6S{Cy19-R9Ha{>%4U;--(|2fa>-Fx#rC{q4jpOst*Ps4m-L}c=UN_MrUg4qfWan$5s4WABV2di%V(FI zVu$elyR1-_`=o-WLw@si51ffshGxbqcz^R7;CISb6Z`X5Uwuh)ceC)P+&^P-o)qp6 z$hzK)(8z*rWfN};B|~lS1z$A3vqZt;ClUUt|JO$b@5lL=cE6 zJDVsRd#cZ#)j^l7E7Ob3ZXTCQUczLdg-81t05Z_5fw3>F$PalMhy-=~HgW)}zEkXh z$LN`!3pPSmtDzp_lE}@MzmD?Bkz)ANmO_IiwJ;x*GnM@qzj)1&6!DxoAuJ+avz*Mm ziilZMF)1((dTAO>6Z zn+p4J!_aEwN4}OKavi73UTlOafY23;gD}V7s{vJX&4bW(kNuID8j#; zA_(-#rZ72Qz4{K!>gxy|J;oIIk`Tsk#{?g8Un+-DKe<{y6870RefflHoSC`fgX@{M zh~H@atkIrqZ95R`V?&F-w$CzrQG^&D?y+HTS|{S{xu(p-%IeXbAR{AFahzp?r0u(l zRY%tP35TE#d!P8an9u`gj>Q=6KXWIi*;NxO(U2p{s>})HD=};Jr(@`Sp7DkomoVZ* zG$d?E4TXn%YLBuD^AVqWvG*TxU%h#Ub^PA0*wq`9& zF_c43bQRb-4X*!;6}0|Fb0_sZOzli!YgP8WTBCfOa{6sO5*c{=4up0DIYfi>->(1MX{sfU3ee1DC8{!4SSlfm=-BxfX& zhm^;vwH@sm{Ocq9EAA5l(mz)ZM7pxPA$fm*IAX@Lv(8^NM-bEqHrf1mlC;I?MEXbZ zQNQB*UNWRHyin%N$(_^9#M*E2RUEnvQeYc6M9uxJ_tck*d`v{vG|5Q%=I-Zf|3OwCifE0NtmKE&#g-&`0 zx4w^3B7*@xnvP#sMt3CuL>Qy>Zzo2a{bYrECp!WPgz8u{0hGPYr=xro56`4tNY@g>%)ir3TvCekB={?L!)i0o)N|Ss|eZU^?5Y#KSqi-f6`sQ`pOs zX3p959#Bql!QLQz*JM~N&3`;TSs;1cKbRN_Ic#H4F)6FaL;Qq8b3;Li0f@ntd5v{# zhLW_;bg=~#TCVrs7arSN6hL#8i1NAfJv|j$SEn^a&E1&z)T57NX{O%dr&oVAy1+(Y zn6SyeVX*d;yR+pBbz9H$3c-B{DLC;?qjBHreXN4JFTbAEzw)Z}2DoBf6>EWUQ~l|; zKfVryugH~N>k{$+?HAFk&4=A@E?0s&N$}sL&^k|@b^@+~41mOZu;kKD)&j&Ufi?r9 z$KK2c28+n;QO`MC>SC$Y%ZuT-&F3^jpcI2ubQ(kOlBgPtZo;RpnipJJ;^>iRLbgeB ziI}vM+}CFrYu`5pGL`!eK0l4Mx1euf9vZP7Oe(FM0duj?YF8#IQC>5svdzM)$%Xs0 zr=kaq{dy#aao`Q%_3QRC)A!de%QW9IPWqe9-~T`g@`O zve0%br;$kQ@)OR26aQ+e?Td+Dbz>RCZ2~^@jKpZ)>8whikm&Nj=UP)kBSSM9Mef6- z<}G0a-*0GRC8Jg}3JqGqA3MbNI5-A5+aFlV7Bk)xwthU#z>jl2(F-UcJ-3Wo1l10<7w40`j5Qld4|vCXH}(y zfY?byIN$mi9@ZDCZC<3_UGMN5i6Jl*5s9H zJodD8q)CFKAM`ZyS^U5UYMv`@6Nzw5<;*YXqJ9miNeBmxD}kfcI@zs+WP}a7tED~K z`XDUBT&_(=!KAYo_%#@*SI0OZt^NKWRA8R%{bm_t; zROY#plVHtSHLspVmTP^EtN1Xy{MDX&7+;^_CgDsX$zd#5o!@ye;_b$j+wBFYzmt() z7CuA{e>vqadE2V;C^7fPohk<>bc3XOpuWyJoPbGsbJ1REKl%#?-_V7?h;$k%|Fc&= z0_Qxmn6@4SC`XwDo~rC*-#^LKC8};>88}dDc?SgGgpQ?eY^R{u=^^jTq!(s(rixk@ zB10czxK4>2Bld^-QJ<^IH!9f$$0js|nbiU2mDve_?`%A|H!#i*;wjH{b6=He1F0n( zTWCYMxU9?L{!n-K`T6M-mKY~LLIoEMCU+){>r?gKL!)ATxouG}@;-EN7)6^Rsp>fwv7>ZSSfcS{05MYCv% zu$RyO%Gqn!GqpsssAa(saOB??!8zX?7XY$8IX%4;;1MB@ocQFL+y|6{9n>swYvN6y zH1S6tHG?8_>c~|2Y_S5K3h3JY^ldkYwwkslOAux~|zx?9Iw&OxYWBbC5{L-Pg zP(B+JdzLJp~+u??rGBK9qZK1|RHpPHRL zu`PXJxn1|-03$?o?e`l)1wr@vEqB+HgsxsDpV8S+Cf{izk3E=d@;q3W7F7~G?%y0q z=XLohT|{)TJP~}x!_1;n0jaZ@YOvZo`A%UjtKa0DQ%f`Mk#&VA>;y!?MhlgS2Huc% z|8cvRw`!fRzrX4i&&ck6;jp4+tH6-?tA^ajWBS!ZiJ$C8_bRHs`?DhP z9T2g2;PgW*phN;@jAB<#wL$=euqxa2@Uh4!F{$S>Bkvr%B}f3e-1q6m^~bO-ow!aP zxlscs=@w+w!z-3&uRs2qosCTh>ZgbRwf8Ee?t9EkC6KcnvU+=67LUU$C#-9KJQJCQ zAm$~$Y6sO{(zuIvC>kNs`}r)Iu74~&m@qqN`>v) zd38Uz=HYU?m-RZ{lgOUufXPD8>YE!FE8`P6Y&&&7pVn!eC>~bgz`<-o#1g4mMr3%S z(-_rq?K{)sRh^EKVbJ#adcPKlRWHe;GtrMcI5Ekf7}6D-yt{DD(z-;h-|gAdH5mQD zd`#o=^MZ8WM;1-v=NiZ@ z%4)j?3D!*38N$LA5o*bCjOy@I1-c`vLz%D}@FEEQrWK9C3=I85M{ggFVI*PVPa+ul zY$U7xT6?#iJdjZJqOih?JN>O$sFt5*;imvml5Rr6x}1?8I+r_m`Nv-fh%tRPQL8>L z`cHh$`*WmAK$C8vfLG%}gLMi}r??PiO=8a6MMr6OYot@h#ZNXjw6An^duJ5)rFn5o zw6y3K8nn{r8MX2uW$xV9ESjMUBskm`c(1Sx@I%!s*~D$A#2j7~3T@R|<_PEKcV2n({|-{Yx-)aMy>^xD-$H`)~=0d?8W; zV+4SD%I~(hR)XS!SZd;#TH@V;+D~KfX0iGs!i_Ws>Y57Zt(>E_$4%Vvr9LwX0auS@FN9E1wNrI(>l$_UXDp~iT~%7nl*SdK z!MrLPrv?tz2U^)paAD%Ta8-TT>NgC)1Qi-pP;?+WceKGSprbRLdlju`>lF8MDwjNB z)~))E&AlG7q(BJN@oamPv_kr^IP7eyV)N1101((~n(^^8a-Wca3yWn~nrM-35zGd8 zi$QoZTFQuF@9{p)?sj|~R`$;zZtJt{Nb73bPm#}`Oi=?oVG0TY#VkuL;}M!gp3ZQU zqsPICy%+Xh-QjQHB;xw7=>#VhRy{3wuC6#w@4@^zKmUWs2JM-)qbYkby(foDoMr8L zvRU)BPe(N?DNPJUOKV4rdJoRNK34|{#=%8qeZICvny8X<{=?H%bC#hEEYEj~#oatp zP{*t#C>CiDY55ytNfpmLVT>{t)SN?49%~*ES%Xv+s$NZ{cP8vHI53(!!d?cESQNCi zE1m<1kWKXd@b!1?FIR$%vlb?>`3wyu6wg5EVJx~#O0&WI@e{&Tjc@lrgEB^S-fi5& zT)b&5jAWO`xkd(rQ8efiy)Pvky}jQJLOU7a_+kB~rDeHz1hng> zdGWL9WwLFgb>0%^2@u%R)NsVb9KOTz%xK7A_8fER;}x8XwEXTPmlpSXX%LGgAfS%B z2#eMBHe4$O`oTUY*Q4Q$9YSd5(jDKB&SU-eZntrcr*=v+!U9h6!Cf-Ep8Io~W?zz^ zl07-v>gG34Ox)ZCp{q9Y^aEdEewXibGk2tP(ZiRW>;cy{&dd8~qk}j!`N-rbyJrU| zLL*IUYY&!!f`pw!T##SvF?|#AJ~Bzo8o~?0ejvreOa{$Y1EYr0xwr@3@cL5xv#cuw zJfbb^3mQH$NebTgJRa_ohqYs3Y={Ur{jAAAG_uDc%6lu*%N?RFCsXuWuvDkgGmo0E z{#!)B^=9Witdb;HC$&Ucq!nw{VvecCex6J8_&c8CukAPCWSa;Kv0~TOm-8_uM6xud zJ=m+*fFQ>C*%r%=QOUWi+>=@HN`W6cb7fdN!}Qo~@}i|g+#wUv0+Xhh)^p>Ky1qbq zVM5$QpN4hI{P_!+#(0IZV=9ji!?%sux%GB#i+NnXb zXP=$iymA9fT|Bjg51(5IGZ9^{*QYF{s3hi6|48(V=gRwKx$=I=gM8Sc+h;~ghuo_o zU$_$`Q0hcSFA5J$xKTbl$uwtk<`VT*dz%l>la(xYDLE6s_V?W z%iT0m`Ng@#pjvMhgBs%Fg)cC8{(=K3SbBLT5}1J`X-OzfSweF1EA~r4wSzUn32~{& z$BNc*o=}RsE5!7WK$9#P49=fRwKpbHoF`=I6@|fx(Y+wF{jG?FeRTdSQBAg}LByLr zhoX{!!h@^}hHw0!1z%JrH8_#AuRF}dcQqJ!5p|)_$@tlheE(6_rXgK2ifT)kXBDUG zh8!ku zrAaD&^1`9w%dLt?p>v4D;sF;WjZ=x}8+$szLhQ&68c~!I)=6P#Sipw4-IKKPP*&NO z$jIWp2)9_gnNG7Dt{4oJOE*jMKMW;D>g_5qNIS?vA4X6AB(pz`^{VO>2QapKkU#r} zMjGfNlfc=G*a8F^Ho3<{YWpP(Ri4(!8aQ`9a{Pr|U%iWVXks`2r5?;+i?xGs!HQZW zhs$Uuf0}EXgHJp*2`n5GyFwZK8D{>^h}6EdIC{`!zbJofslC&F2?PS;mIG2-Tqm${ zIcgk!$?vqTB**Y7$fP7Za1AA#xL;(>W3dz_I*{%0X|z_CuC;u0A#B-AWQ8p-7_t`j zUC8niK}CPz`fC=wN-&+PP_8xa`4074C6A#+a{#I;1zzM@!ukA&1Tw1gezn|$hgn~g zi|R}38bi`G1$E0@((trWt5l1d#Vm1Svh4?(K<|E))?@Y)0~4`{ZqEvZAEw6gub&?+ zw){sB8Nx9!CKVSdRxpjp$SGE@yGx=g)&t1`XUt>Oxv4G)Inz>aDC);-NJ+#l`-@#7 zMP>7$h3+Hre%NPUUa)--myK5P&OBSvK^UD^^WX<{60(h5UvqIYQ>SrkL=@n*AZ%@ZsbH7N_nlAS0 ztM@FuIWgavkVD)e*76cJAs(^UYd?}&jf0Lh2u>~v=g)UGB)5H8+}9XMBa3>+4OK)Z zP~7AD6vO;n)2sb!wZkDK$M8O99$`o zy8Zs{#&-H^@?M=4t-jK`7~vzPY*CL#G%4^tbkU)+|7jp;*d{qHOB>b2;n%~x-1#&E zXcbj1|4P(0u>9Ngv+_N#X{$FZtkKp#l1#FSvOZYP*BsS4$Dqk9q7+%jh=*FFuw%`@ zvhSrc0vGG5zgdWO2dA15SLbrQmV2;3X>Y6&i(jVk{mjt(+zx1xs>qFM<$H$~my$Wk zNSP*($Pl!edvrEe=naf6^AGklOeJVMRj?_Q*JJR@CReV#zq4BYUT2b}m7Z$xZNqE} zs_=32KrfXI2eyhFWq*5pMpiEtK;825L&Wt3?E1{4-5BS_Yx=T-*XSeNZEK6p3UOsG z?w#S1JNVE0eiJJ#!@kBa2P$D-tm@N#O*6<0d|oNFFl*Q+U&cuRFmvu+jQwu7eoWG& z`)#!61xTx>l<+ypZ8gF(T>6MS8r-}#cPiA_udK2aC;|y#X3vf^iQ+G^cBnK&gyI;&p>ZMz9TaUxb30avECeKtGCi3K<+(-7QU)u`S$n3L3>9jTD ze;yzAQ+Qp+;Fmo!_OEuUSQ~Ks%3-=)-5&yp%eO02mEvM0jl8%$m1OQb_%#=T{u>17 zwB^^aCvHPLEA%R#u=8EAh1K+SxDt~}l{V6aszw5cr*=^573MYCE6|yisU5ORN3;_} zh2?LI-;l@?Umw(Gh@OWY=s&|IQhD0^;?vk>381Ha9BkzMhJ)Hc&NsT_5pEO*#X{wa z@NMM3d)9k52{hNV*gLLrL4v(rlM@$RUN8>fF|WPwj1vlrkli(*9ENL{rZTgLi?0+g ze~I8d?2!src{s^C=dm#wnu@iC#-UzsGKYQl{JY;xvkM9S9-&N(oIO?J{kwzmg{M{OxA=U zkYnYog+MX*S=s|9EqsY7CPlM6)6TBeC&llXmRvtde^?ZUDfk{qf;r%Px6x|N93}UM zpwfHnt%nkDedeOdCtIg3{fEEvkIGqM%EO<R*C8OiR@!xLwl zeHR(*C@t=3pILdQo5-6xLL-#F?m)csN3veUcx&cTTg`YB}qV{}tN{x~}E3@(jCZelCO z`zXTJpi|6}cEM3X2>awFHHi*NB%~crC9bolx|H;}N-y#cI)v?>4&(6RbBwTidtY6< z9E_RbW#J|hQ?+z_g`9k7WeV5Gj~KG|DUFMK+lV$pJbswPo@_IWP<)T4Xy2UIgj42igS2};!w{Ziz=x%5S)Qd0FflT|R9BBLO`p+1U?7U; z3+kwIw{e%kKnYq%H>;gt?gEDFByyi$d%=lVqO_u?$}gcysh8*8A^_F;=1ap0cnGUpr}4>$s2?u) z-1yeIINCy4F~7`lK9W$N;4=M=!Z19btJt&CxA!BNWyzW^mognqVBWU2xwkNNrkq>! z6=}Q%vI1g9(f4*Wjo3Rs8}CD1;)y$5vbkdyJfpD^o!)>Sj91^Kb!s4ib&xG7yR9_E zJ$ys`qy$!tj*@R7uRQtY+sdz%94?cEqR?_l64M-82-FVb`-1E)#iex^gbj1gSN~~& z9Iujg4PJC1Vr>s|k7*H!?vKEyJt2>fRPR8tHV!(#hVj0(8n^pa;%>8eHq=Ce7p*HW z)$G9QlA)KA_^T(4*#!>JTs+Uv5UC2*MAL%E!d6*kuA}}R`*W{Kp4w@3IArN0xpmQg35;}5a$_ww{d&jzRwit#wnX;p?Y+3@jFjKsd znDr1-wl!N|s}E)EiK7F<6b4k3Tr@P!b?OU8fJ_f`Ayoya2;2J@`Nb}2X2J4DRz3x=rDF7sl=Tn}B^)Pq`?1S=F)UZ-~r zCj6uil_O5&)Giy>Z!01)ykJhBm_gPvtm+55z zp4ad#1$XW;BR~QS%>DQHV>~%SGoT+t}DtxX}zBP#l{4!X$rw zzKxSvMKT`#f&hEBk!rx;*BjRqT))gO5X1@%jr15rMi%}8s}CdJvxH(&!;GFqe60#3 zW};-d;!so+E;(>6fx7UmeERa6m}dU}mWxUsf$0ZP@F_53(CJ5xsBlmo8zspgD6$z; z{mNHJT-*cAmWIan=+9dAcHiNtz9bMFN`C;bcTN;hj~5JG-3ok8O%WONl|lncE6h5} z*D*74doKUT2PhrJ7aGzzWI0feZ`!ms9t@<1meEcttkkjbsRcnfrUD$4oi*fIr>OXr zj647Fh$7M5hF(hr>5eT5aN@~DJEeyhgC@lgNU=Ees7(WAL*Y-3QVi*s&wY?MstHuy z2Wzswh@$ZW>F6e~D5*y16RszWaE#X~l9Oif!w__X zm_KaPP?7YX4;IP7&ajKB`!m~lw~_IH_2I5r{EzQoz|Ic=lr^M1x5wZ3?afuM&k}FF z1|6|X6engOY!yE1I7s;qiJCMCU$GyLpBu9|NT;dO2z#*8or_5lAsc*!H0UZmXG!?r zx(_DXg`zl*uiQ$ex-~R-o}!)4Fq_KW6f~qRU?#SZ(|M3OEL3diNT&R|09lXWQ>lxP zxYh5tSvr(0;tNXCv&B;LpEN*Z_@isxyw+222lFc5l0s>a9(qrtf`%TVBj>8mWER(M z0|TSo8CIWR`%ZX18}B8Fh_~fiOaHs5b^=-nd8{Y6$dR6@UZ9Hh81{+ga59Zkf zQVVka#%Yg(v;t;zq>WveZ-dl|~a-%q*b&tYmyAH13I{icq;HUbQoeESZ>I zq*!WH?XyxQ=3Mhxi07{vKlV;T$QNh0QMkW@Wx7S5XtQDfE6#%ghOP<*ygFz=H)`^@ zA2~jjO!#s0Et9M=@nG#}0$bZlp`Cn_Kt){IFQ3u0)2n>d^-dx(>*?#L8>BSC&bU2h z?{Q1^TqJ+=PfFn#Qz-5!zujO(cRejyNYzSjTKU^Hi|xmAJ#4?+_3FIIB&qM)u1^X3 zds~H0#yfRW`4lk5aXVoIWYQSkItt>ruEB|Mc1#J(5kgyn)`ex$zX;fU6t4I&>X0n!}otlj8&G8 zEPM*Hv{81cK1F`pBu_R4yq=Yx-=*t$+$(89s~6Le;Ugas)BUq_N(250+O)_4KnNbVibP|o@@Bxjl-Fo|JjC1f;jfL29S{>iGQE~}CQUK9dc5{#$^kEijdzPD zK;^s;IvqBFEIchNkjI!W(d5+?!_^4EvD={2cn=#HsK*IP5Q<6s!2gIz`h?)4)cgXA z9tz2lj8UNU>s(g2#Qop9O#xt034mm5kc}pu&UM|?(bnDt_4iqLmS*U?Q*hnO9=KL` z19)%7Du@JLwp{3;28jeG_LJg4mI|MD9R{((HrRvle%l2PBdJydsa^lP&o7?}agPI0 zNlzf2Ii13xBl;(aee-V|oPsa!1b^4G!*U=Ity=Tvj*`qfmgm>5M888z43ac+53-DO-m=!D{dH7Z3DcC?IanxMR^-0Tg-gG)aGpk<+NT!6ym^?_? zEiS4$TOZ-XKn!oP{r|y%fCJRk+=s%4gp`|GQscZPbx9IID{?YG1MF4-82@Oo{sFeE zKEMtN-QS)XGRtd7r6eU;Pv$FJ1ANifk8CaSJqiV@$KDqxw$`)7**UO>lY#sSdB4$j zZSrPowiE?V^C#ORw(56D1ZFQfb%_Hy>&AG4H~+O$$2`ROisp3$#Jij6BdT`Pf17AE z=Cljl$P{H~XYTw`(ebuO2K01vf^sDF1D{hLX#HSZ1h^m}aH0$|>ojZab9KWjUt+TDBs0M2^=dNT4$Z@7;>&H~9 z^JUAqxPD%k5w&ZeUNs!(#*|-w>eyo~@T&h)6Y}Rj9~zB7q~`oG6X1bWU-YrSV|O6I zz9T!U&Jvww15wQjBPNbFQ6ha8Wn_j=()|gKLDfC@^9t;@m*E#nw|;$q$_ZWF6d`zPk4xFtsX!XbNoZOZ%IN;`jSj4*uUVK;r#0)65`kesQB>J<1lVu`D<~ zH9wdZ;TY(Ex*@O&3quPU{lVEkPt3kG_xp+k(0kH@+iV9OFF9KN`zX9;aa{5H43NFP zW?6^5J`*t52P}T-8Q7i+k90kMFY3Z>E@vA%&RLgShgkJMP6MYH;>V4~v58U!rzx$p z6BbHwE0HG~tsxcS#TvYV{;w+|S7ZG`cMcYiSKqABB4&g?fqk{y+dA{S1R(=e}T5)|>WzxG$;g zljr*xZ%>?RVFEjz~*Cdp{b&oPUsT*p`FZ*+jjM?MJwa`}YC!zS|c2uHNXll;Ot8!oqSr zn5C|*T{xJ?(26ht&zdr@sVT(k$BQHTta&;TL#FVvWC`Dwh?(w{{ohBB$1((7tziXi z^N`k$01y9d8?u!5<%k?xUOi+Sr8#htj+lAw{^?$UKeNr;pO`1L)#9 z-*a15;|>G%wxbYu0xMXg4JP`bQ;-X=D%A7cZ}D$2nZ0^A`+g^ z3VgGb0%ubUcBZ)mEL{tj&g2Dk9t%s$-SHP#(;$V`Y(MTJsR$#^gD@IafsSLjZmlf1 z_MpYfOgIQtU}}m3PSg4C&elY9t_b+EkviH8(h|^EH`J_`aM#mIlZ_!QfmdeXBfTKt za0j>Ca{24~*x=yA&0%LH)TuMz@2UR31Hwb*T&7Nv?tTyI88`rLHw~~(UU0`Nbkv2I zDJJcdi$fGV0}M7CPw9KTQ&bOJmcwGxgy1|JXR#Z_J^B(rSMuymcpm@H3hsxvjfv{O zi4;jZJaM7#+1b6+Bj7_vaGs{!GL=splJMa!AYq7r zK_1if-P!3fE}Wm2b#i}=EMNFrAG>i13!P71^~757LXe<@by0iYjFJ8C#{gUJW;In5| zyF)#`p)za(ezIv`KX()f(-1TsR4lL4NZn%H``s>l85<9Tmj;=9W z-zQib7jxkQefo`#@Iey8K!1&ZcmKchGv)eOJg9>O^#~5C1P3{`0?=?q6LHHbC@4$_ zij3g=&u5!bo^Y_TsyMZurx*qPv4B(5-LEWXc0u{I?-5sE)9eNBr1Ei+^ z_$_L6XeI7fZdGojGFA&lXORNub*e;W%j+dv!jr{u~>l<)XhPd=TqC-B3tL_W~O%SVwjy}&~pEV`2pU{fSNrzqq#c75`MFX28?+LVjN z!+@IeTwYL-_VKgNCh3ar5Ta9`Uf+%7NR# zQebyF7J%Ij4^&X5Uj(4CY4AlCwjk=N#1y+)Jy`;!zgO;ECqwMUcWP%8j?Eh>)OkwJ ztG|!)M+ScDygRSl?IUq2159HTT&eWeSni~Prm-C-$(5w=6UGt*(m__>BdZlmM-UIxj;+pyH-VW9YmN%_QA69Bx4j!cs4=h+N z2EGo{0B&qJ!UwgV2EBRl_g;f}J0Dj=$KVp6(^!E=p}QOGK|bdnblP%{`Yz#b3dQ38 zV!GC#6a!I}@|wzt^)>ac?^|O*cc;4`!T3s`XPvVv!|tnoQmlsU`d#O0P%TgcOydTz z@fqT)(gzg+NX2iEgjWOtX0z0v&W{AIo}Tb(N8B*|NEq?~d9ZP&^EcFKBfo=J9oc%| zcrhOj8Q;~w9ds!?l7Fudbqn+u@}9&jc|yw>wGn9sMPQPjkhor1!;_3@W$qCzhsGt( zo#&T0M-gVXaH=jL02V5J^vs}Y0kG8;F@F0saVB#l`Er7J9P|;|^;~+NmVzQ_JuS*@ zmPlF$3ONc7D!!5HAl)=5|GoZ7Mk5*b_vHx>Fm&#ti`oeyJ$y6V*aYz@ik!VUCy)(z zdVsB`1M);aUDUL}@tu#?BJKdsqz#OJAd|a3aNi$qqqaoMMXG27tgr&I;oS#13p#++*ncq|UK<>Mqof46+sa?58-aRD${E>7s7e$2O z8hJ1;GPrK%8gI&syk?IQ4#~54lab7w|@_Y1;@L_VdMhp zWZZD_lt<|dd4Q_(ciogK3)ID-ZivW*U^Qwt4R87Csb05V;=6t zcoucy!GOEDr@2bF-GMu%XH0TX*in<%bL=IxymLDqO>Z1tO;I@my{a3M6 zf40HX>CT8`gyyuozb@@v%usW&I8FmXJ}GdQ;W1{Xv9DY)z<%zdta|*8<{kfpG2HI$ ziYT5j97P;|8g;#=_D&SpHC|OgLdJuIfBKjp2aANbL>@~N(c04M+~9e1fU6-H%)XO5 zDGtJEQviPr|CL`#L1w<&P1PMb^mBa=KP7#GuE0NdKR8`v*Xi$Hq~9TvNB0J?pF zFDNBef`(9hrCfg6*{l#9HRK4L`IzJ8{w1&(NX#>eYtR|()+lpm9&s%LVB zQAS-Bm$jYzP=j`gPlz8)y0Rahy zD>L1GhPTZvAE z2Q1$qREVR0!<;B`Zqk##4*nr6BCYc#s5je!<1_S(^^)qyx{t`}IpB2stXy+WakTx2 z-CMGLH3VH7WD&D}U0icK{%h-;_lh;F6G2tkXBlpzy52me!`xW#ZBm5;+;rA1hLiUT zdgp8GH&gAKyp<(J;K-+dEr2NG2!<2bg&b3KEs3uA;tBN^K|V2UUpt~*4%j?w{wj^P1#@sgfRgDIjD_^|~K0&p+3 z{%)zs`-ZFZob#{82S8mN097?ps9>$?LV#OURPmenvA;G?4}_dwb=rPD{rfUJbN)&J zCN(S?%bs;!_5SlBc;=<>!j?sU^A_;x@E0F`5P_GF@P#E^_N1lh=l(aiw0>K@<)TaNr;of-=9Ml{ZrLqa9(uN=Ti;Qn>m5_&Up|DEoHp6GJ04 zq%;gA(%lT*p@<+I3KG)Y-8G}qjkJ`6bPt2Hij*|Mh#)B?v9GzGXTST|ANKpnrE7V? z%=M2mj^lT}jLM6nNcb6*fmHou@E<=SQ1unL=^*<3-2g{wHLxBFjyIIc0%npdjk?cT z>iOa_x%>u!f#8PdIFEnMc?X@vzGOcm{X?;Rxl`Et4DhOq-T7SK9E8;a>dT&8pk`5z z&R^d;chR2s)CXC#bDo=q5~7VnHC8|w;m40+Xa#W1y#FUS?cRM-UtIvS4@mPDNCXbM zf%CCiohhK~ko|0Bm@91a=%nDh%L%wwYJtnE(WyN9#eM?sfb*DJtY$|1<)mD9|D1{_x2%~&Nl=)Sdu%P zR*t##-%!mwnUJvhMPz4j0zf8>(NynGV)o<*$piwZpPB)5f*~@M zibSWPsWZRORf%jG%fm<-Be%dWxu25ONBiB23XxNOJtz5}Nfz(}v9I7`K%wvbRMwa8 z0;jz4ow0j8z;!9$d7ZZ->wp^Xk3LEhaHQP416HT{jHbW+ivMq~LK$yghwHPpUOwP* zyY4{fOaV<&yaZ<%&va+;Cq&&_fZm}H$ct)@r6B; zqIL(*c3Q^djQwa=3Ih;#}{^a(o>_+DgNc_-ofIRn(@yzS7L(~3(*XctU)xz&n6^-Ogh0*qbe;HF?IX0l%L*}cczh}&jb;uV1}K2!IdKAp!kBvo{{vDM>*TN z25P$paG`7YdoDJPB|dkZz7u#46e7BvK1_)7LPuZqi(uvhNnnRGPzkSb3;AU>WnFTH zjxW#>r~6miVZeBfUW*W5<e08TA7(| z?Mb8wP;MPR%hBPI6_zM@t;-+~20cUfy<%1}K!nlU_}ultfi^2N!ti(kkR@}tpJb#3 zmvsQAF`-+ZkB9uq$MXzzCQ~_Jg$S(PyAPs3KTe6g=_JYOx&+ke3$!|%fR=|q#IPl+ zBwB2141iHRGuRVYWKGfe?>+JPqG!N`U63INb{FSIzI6t@#}Vdm1q82gPBW@B`#ce% z*{mADz;m9U_m+10ZH2>!`j$Nc0$DzA zWpPlw0>nvgF`iqHq$B~8VBe$Et&L#Jap8a1k{&(+ma?g8C>|Asox=4xX$^0ESJQMR zu@opNO?tQTpNuEsW<8wwbPG5jdrEc|w^>g<0m0yc14NYL2NKO+d4z~}UOoXzw0`Ys zSU%}+7Z?Y{9m)&n4`v<-7zkvgse8Y+KXPZ0VFvbdl5CDuq`JJ6g<`0&>GL;iG>b-* zoAMykZFSe|qyA=jlz^H26v@Cy~aC3~!;eIOqc*v#5fiOfl{)>CKpeIlS~kdDy- z*fXk6CS~fcfWz#Wi>%i*`a*IAP=C)uZP>rDR`~%39JLg>nP7UMJ&BS$SqC1ZP!b6r z1r5`pqyIl|>Vw}wvPM%6rtz%i7fF1d3Y^uo$MtT8lr}-dr*-}VT#?j<;5z5)%W?y>#?LT*->V}P%iquM&h*( z+Fd0wPdDcRXV;|*Z{AjvK<8c)_I^+)PT>7s6s2mwMTuZ=UHkR}PeF!Ab+qkTOJ9(N{8Zl|t?pLrT~E>P5+bGj;1ZpspeC)Jn}m#F zvp*&e#qLfx@S$g0yS-3(Yp(pt!Y-M(S|86lL+C&-c9llTvA* zG^UMnTEv!DoEEZYHw#4ZkAb}K*|4D!SA<^Vk2^F3{5@|TR`m>y@1c4wXa$PFqaQ&P zI|8^BO(}ATRR;J2UOW^`Yy~O}!q8agPZ@~ENdLYBZ{xxka;AMOa3T9_-J@YQw<4!Q+(kz8@}be zj3O_hHW^Mk?>X=0m>(B{C`{VqnP5I6jV8ID>?%6M9?Q(GG_VG=v_sK#%Vf#^R$y~b zJNyMK>F&k63rcS1{q&kU?zPBxjExHUKUJ_YOCVkylO3uowtDXzqQZ`2R>yqzNeu10=GQ@GrG!>d0;`h zEqKFf7PT86f>LIFAdT#T8c)X&YYgCjo(bBSKIEZUu0Yy zye`~}(vWN+oLL%YC@G*j(YeZZrR!lNcNqcv64z!x^r!{J&)2|1n3ts&(OBf7!lpj} zHKfc>{h1>2<2LqQK&Zj&>qimFShxuk*9AJ>-6KpIpVg&}BIKf?o>4$#23SDhLG*Wn z9BH_iE|0IFgGFEq-L?mowM#Hp1IhL&`MnlP3kl8Xh?lsc;}_JALA(QJ@$Q2z|FP@G*HGImYnR) zD*tvOuKRm;#*X+d-$d<+VB8=is1<}O8ujPHuBBr zGC%BO@<-4JiJUzFpWbKQBm$zZJ{mn(K6f4^J*)b)LkEsU(4SwwyG|xcc92T+;){-@ zyIzHVzl#uZ4R@iQHraR4GGk3_v)ZWum%|QgAWjO924C#N6%*ZMTg~y+(Z&yG5^+Fq z7yCG8ExX-eDT=8O{agc;K2k~=XYDIutizZv>rgJ4ZU0d0<IJNYAtzKL+TSXVv$mEdW)OPkJKkJd)Wu=@4Th zr|%yEsDdu|VHYm0NlEj|%X=Rc$5qLR*tNG}yT|p`l$W68me!l*;?kkwu8NyVGrbo* z?Jm!3V!621m1A$2Ox!=>3vvIsh}kvAO`ZB!aKue3S_x+sWSjYZfzHj6 zl$5oXTymNN3EpP+CHrC8<3dauoYWd9!)Rk)w`nFr@4FI~o4y(5W^2pyJg0Km{`xbn z2v-xc`4_64;ot4ZPXHiE;Fb?5o&I{j}gn+#U-1{6NoHl`iT->yzxAD00pZ4g~ z&2ZCq*s!8)R#oggj8w+Ww?aG-E84?9?4v;mLN1dw%=4*+1G=-Nm+=Kv^db3c46=QX zMM+Gplka11OW%EYzRf#XFO`0qw1w6}T#5c2$WzG<(5IdD;H$PTsBG<)PN5ZxU z3yPQAPg;uN*|_GWGbp}QH~y6FA`@h-xuEr?S))+gvYt+h4EN5LZ~P;ffJl=^eJMIh zyyo^4{yhP8gcQF>{=#Ei$UD%d;C1bODik-PPvMAiS*h5W$Hg+Gpx^o$Oob!W(>9Qy zgD$@fzrbIn^_nS?{sbpL`Vv3~I%44bCDOV|0EnpVjNeIomuH{%khQ58|@Kd|>K@I_GtIvRw? z;PmX>J8_y1-7}P?XAC;}75=cmYf_4%{bd)NE?C?Th9l2#Gv=38BG8k2`IO=-Aw?Hu zWm3EOqfyyNzbInq)Tv>C&9q|X0=mxbD<7=!E&YZ<2FRpU}nqr`;|OdghlszSktXP%JA zD!Uz%^;mpgdZoBEb&;NI9W)b02u6woc?0Kym+Vsh!=s6D3DkCwDxKqEl9dCKt#i%@ z5;G)50!TfOD{C@MZp(w0!33@NVNnHeZ=arkc;XBWA+Zim;B(tOJ)XIEW1tKU%VnQ_ zJ!P%8`|cx^eoD}ZT)yTDh|A$bFxOj0%Nl{KwGSFhJmX&a*Fww*qaUn>Fv3F_R5Z$G zgtkeq__i@_CA#BQ_9R<Ksac>3_Jemdzk&+5#KO2rfmQXHTK%$+`pKO1X2TvX~L0j7bfa^Vtoc$tK7 znS800rzNSYFyV1VAb>9TX6PxS>Vcph?Img<2|Pr;*+sBj$~!W~b*Xe94zQ&@e?#~W&=%XJ&;-xY zq~>pt+>BIa7|^QVfBK?|ke1J2Kfz^(8P3(PesovWa@kiS+jokCOPKkyFeKoj$wzrj5+D-G{K zJqu;Q$0pg6J+q|OKc2aZJsm?W7n-d&$Q4=hNs$(B(^k|4!&SAvZU*^@5_rSKnV5bP6@&WG-WyCw#wv5 zv$f}68!I$oQZ6|H0mA~@))*jVyf-(}vRw|gmbMvL-!27aLQce?RW6AZ7UJIVB$_S62uZY8(2zvi5moA6jb zBi4V*WwVYBbK7W8)OL_#I;v#lv$u;pBGXHtC5%)QCntoo9t}NxLRzhB4Ft^ z>~SEdtJiDas=1|RUDnkpA3ec{@nR_+!FaCmY=Uzgjv}^WK$YwDpGwc0B|1rd#Cq?8 zkMd{+3C=TM&Z%If@EPLmm_@_|X~RCpCEWF2yD(RtIz`w$8+(S?A=AafD>lYQPMTkr!DLw}{zpJvb;OKaIij)=W4JP2p-B>0qE z_QQ|#H}$`N+*TLj+`mTXGuU@h-kT3|QPrdgeZzt;V@dvwjbbhLC_H0k89l`eeE67`;94C5bRJDs0}Sa8j|4>{6&!2DJpOJ>9*Q_lCCC5 zV@QnE7x6n{I_cDHqpF0hL)~`5_fIo2wZLr^9&O=YACd$aCW(IP`j0JwZ(K z@oiglOt|O*7@N_kfqd8>&~Z$h!Xw`|-XF9zA{;r?seMGz`EvC>k^M*=ar(*ho%P;g zwH2GBFE_jLHiayBifoVykAVaxXpG49Lg|dsoDdd#kUCU*s4^3ZSv|J?Yn&LJcHg8XJf#_Ne4&CU}e?^h50wN zj-aXOg5~jY{Add$t_PHO$I5tDuL4=XF8Wv^lDb$h*h~3e*1?s(ix>vmoubwt?;ajf zEV%Ap7LYssVxc0YC6Rxk=p{@mfNQtZ%|nR@qZN#bUn7SO5)ByPJs+cO9(XBAsAsFL znqi;3rN>ANIX~@wa3*N08=jmA;HFxPI(Jkq6p|6{-uG@Vw{*mK4MA#1EU5+#9KrBp zBEr|#qTsKTjjc!%0r&xnji~G{$W@ahoxO=C-LO8djD|y9H{FnKymFm|$O%?F13_@f zs1358a^Y2nqIgfp*kjn=(f0Z3Y_kfJqYsG8T64!J!1OLk5fq-WQxiwg2i~=iWU9qw#d5|($a@Tf2HU@!HR2b(x_AYQ zH7sep18$C;5;uHnf~Xj{eabSc;1^=);}&GKmzo{(#`~>Q-wrChXFM_W6oJMx^L#N1 zSy_RHGq!KQrYv9|4YDHwlvNi#egV(+K%2=d^_p4wORrNFF5Ge|lmVWUE4AJ+J46lV z>1Y~89~3h#?og1+ih=9n(I?%rPR{+CE54#pEa#Qso8vl5#Ko}#+1*QA zlPmhd{^|`7SFwI{#ISygIZez@SWdpDV3h^cTOR-E{K`E#WMIy zTxZ+Cy}VcF1$*(yqVh%RBK>NOC9K$XkO*F-DT1^_Phh0e4IoC@<$mII;lwL8_VI}@ zDoL&pz~`dhPHW&ZH5zVM*H!uf%c?;VF!>Tk}On={N&kuUP6WS?O}=EsMeEj zSPG>$^4u&2y44R%*Lr%y@RyO76lAjKB9YvA1Zi+l?bo&7C4S2mCtKSnJ|2uI+&v@7 zQGS~?KVC~neI&Q>s>UZR!~Ch?m{prl(bVO_uD0b|SI9y^3_OfyO5bWj z6>LBIozsR+eocd=(a!m}bwD$C7@KO(-?>xy`ZKdxQd36l-dY-@V8GF_zt!f;h0*uO z-#+PF!HVbjk?m=B6ZKo_hANCMhOQa?5A%Yhrvo0cUa~>^3N{qho`zf6!*gVg2Je0g zV3xUm&hW@^uoab63cJ$8>SG;sVX~)DX)qbk0)Fl_JZi|~{p6Lric zTKT^#0oS3oz3q$c~V;)R{+$hdQO|G57Rqe9dG1W+Vcwu3w}xoiy)qi zbWP8}REKID5BIRh_$Rhq1OU#3HyCV}N4F{F_pT_79st&xgy-+10{#kDPJVJ2(hu=Ms=835-e$_KL1{e&Tt|>cVf^>2U>;Fp}14 zCeR=t^F#;}NPCbECv6XExXb~6V!^tp^&Yqwe6@~&4`|9q+5bjY5lfu=?P!po!bNRkJ|!#xNbcsj{U z-rWOW*_UvfU^1BuouoC$eIqaroyg8QDZ*7^JcIVQ& zALcPI7s*+?V0zRwyzN4Lu^!=HLY087XlAgUOar-_1q@`oYL9s*!piEB+Uok=Kj;HX z3wS2?BIxQKWgM6l9^LwvqY-> z(#niA4$!nDe0lrq!462MOCKtHzx$Wtmn8dVKdM>qDqN@AwD0q#vwPtL%0#Y^oTL*n z2U};Uh4T^*P1jEVfmEt^si@GOq6(Mi!wCb0txloW{S{ZTTZg?Wn0SLy413V`YVXs` zo_PHuRASE=9ytjQ}X|U5)*5tQED6E;o^L_2BsjCrQ2zP`!-sk83DEN4GClXF25}Ek( zp}zUok=P?@JASfNsrt}Y;8Tn^%l^bSX{`*;PlD2NS-2_+?$#`9OL_A4&|jqgZIL?o zXj$XumU%=LciwArXU_d2%|q3usEP{E@W?9Blpd2+ zacIn?Fox87c|$_PA6KF+ZR`jrRO4YS>kEx zIcB+!THrmLm2u<7jZOTUOo=6i<$U3!q$|sq&-sosFD@+7*;rNVO;M4MX?!CF`9p9m zL}9{7t>4g<4qV+!VTb44az7L5hGo>wjr!`0A+r{IrF)poq;pN${YkIBNMeHB`)yIi zkZ82dI;ND`1@Sf8`7FQR#0mIfGsed!xl#A$`&GXkiL4sA;qGH!FY!lJ4M{zLs-@A}z_aw` z%{vY0?&A>hwXQG;D02TyU2S|oXkOySxCOfga-OW!pFJj2uPmRUM8C`Mg9oD+vIVAd zSCrh2H)J}rB6t^h@lh@Au)WI^;uMu$&OmIrckYjlG9k}I+Ho(U+%9&(%naAwNy3o2 zBY7gOLZ#YjDnBLW7;g&6tsCk47r4Mr8JelgB3sRW$zC(4tQlU|Q4APc zE^ZjvEaK9;7*Ni^}XLr&GrlVhwoV%t+Scu!XR2c|5xS)2%mIb0wY`=I{j->A-c0qr5)oLbiB$& zBpKS;O;7GbQ0jzg?0Cr@u+iKKcK>Jv_8*4y%{IKe(lzp=F>u5o8Mn)QmgHwm>an{` znf{6DtjgZ=8mv-^e%ZbA*f3&UsomioYcq{q4de4&7ag^RY@b(RW*s0VtR`O zvu)~w2bAvBQTZ<{ZN9>D(;r`p%vHkd;9vW*H-Qkr#f(;OiABk0b432E@pwgXHnw^z z*KEgIIp1xYbh33HCTXzv3~1h|t6C+Pc4Rz+ms)ziDp4Shu)-woXr!a9q7g{_`n!Hm zb~kWeH6kC9a>Qi<*8C%#=bU#ha2?V?5L^UTiY=d*D#-kvsPxq`qJCwzSS|)$+Q@4$ zSXgS66B@7)*80=k-I7wTmCCJHz*O;I4IVHnyZh|;i?O8U96i`nQmR$)04+)YHQv0x zTCJYUKmbySYfqtE&O2O2^j>Cp?5G<$B+KSOc11;c*u)d$&C8e@1K-5%&`r)@Z2$D2 zCixR*N?AnfKo6~GbQkT?DNISadH!WZzLwCkMtH#0E^N6(i1^(s<8+x#8rx9sY=GG1Nj zE1PiZi_sxT#|&t5YNHnFIuujlt#dXUD!vo;<063^V!)T zCFGO?Zwq&z2muy7>F%AGYzK@&Pfab{7oKP8W;Y~zjcwtG&%BP@R%zFJ(n096TBBBp zw`qjCUbt$P+p?B+#({VJy9hnNuoZnzY6gML{~sa+7FvwZ^oq?3L+Z_xt{-ikS2!p5+jOe(UsD zSJX;B`{weQpMP|@Yta%Ccd^E}Y$-=AaGqwHS5@(MRK-=%z{)oQ7jjIlC2G=LS$ke; z-Z>QIm3__o)(4|4dbm;@S}ylbwn_>pMG8M$j!UhqzhLTr>Z!sNIHF-R@)D2IrV%>N zuxKP{9jx)O3U%_5cc4<7NtE`(UM0{1su`d4CaHM&@x6?Na_sz2A`r&hzEG#VgXN+X zESd%*)dQT!s{Lvtd1u}=3a$udtk8nxWyc|HC|Mw!Oow_x)OV@+v2#A@b}{#0sK<{b zUlKPxAAe_P-5)E4JCVJ;o<)ka_bdiZ_!qa^UAT2#y@C@p#+|oScn%y$;zT`be2gMF z5KaZGJ9C-OTLbp0lnm3o-PFwK34!bOkITn1w)bDGmjyif%wkz(h}RSU24B`;ed~+i zjv=b_Rpm*P-qsvX-cTCg`)?YmP_wH!d^%);oMc&)UM4#uvi!rQJ~*L#F zcHwi%OFv>h;#Udvsn0Lhf?c1tyI&rqny8>lp6mltIDJqKRR= zHe20ByOZU%WrLmE%RzJdT%!{tFwkP=5lSehU8M^GbkB$Wlz5F`{YyE^R}XMQE-*Py@!hl8UQKt5E>Ce2r0lQf_OC%4>)-UsTKE#g^g_FKBI8J?%+_HsO)0_5=n_4t(csy$U{aG7#) z!a+g3fXMQVqQDiLPMFo*r3ChoT+l9jK67y%VF%9kJAiR$IkqQg_!B#&|yiuf@kFvixjANCxytH0*_6 zNshgB4sswdEn7V-VBHE!rCP@2G}i=O57Ngb7FCsC*@Y8-PLju?N+wdfZ?_NV_XWHq zwtvc=kZzKu(Gq#*O;u7J13g{xqQYQx_3S9~Qog2_Ro$R_myIzw7_XfVb3=VnP7e7I zL^ltR5mWZ{`#086SRJZ@mVMv8N6I_i@0Yz+cJs7NRriwxD(~$RtSmHpxKvBVD{H#+ z$P^I9f(Nsw+V(u_mUeOExeXvU7Jv613fdWy`G_1}bgs8Zd~g`sqs%ML6o-_h>=PnB@2Thd!)5tR<8Pnff;&rfv)9)mek5ZGDsV|>z zor6(fHAD-o7stQ-dOK?|RL0ybQ=FTJ3h^mDn$4raBKJ=LEgfKZt^)X_){35wBZ+%k zhTj&AG%HuU%nC@_K@}qP%`|=!ag-A^pd5c#}c7ZAi-57+S=i3L?eq0}rb z+HszgN@yYKSd;{`U>}~=TN8JjR8bn1@G&iJWFF|?!3}R4y~Z1L9pi#a1wz)k`&>Pf z>wnwGkB9_6VX!Os2*?h#JUDySGOW!|CD6!A(zNNt`L>+g_x(@Bzp?+MJAF=5%+ndK zPm`c|tXMasq(t59KvFa}nsKhxXtQ9bC;=C0Z4iQnPGoN_n14cujSzJ-#^F)t!SF^IB$OHpfHfBd$guvenOExaYDD^EQ zZAIoh=^nC`jZWI27IHBz?>NbNS{|UGgW!^4OZ%-{mkho+WDLtEN&Yy;(tq&-d;0J8 zMvC}iCF)kMS@G%VPE9d;$H`7O@kSw*g0QUw`N``8N>dnt{6|+0DmTX@ohCyh;R3mH zIZX=ba9$$M&qV;a+f)0ENpyAcx_^gygzT4r|5Teuq2GsGlt@ogJD%JT_I&iVU?aGz z_{+F@{&s-y*yM{H_?re9B&F1>z)6W@dDipVRAx8($bC%~QS4OftY(n1q!lP*e{4GezQQ-Fup6G#bT7KJJP6iyXW$ztgDk8El&#J<;xbQe zL>%+9*LWV;VTalN(nApRAcQxd(M7_8--$Rl_&D79AY8Pgrc~~VcByo6iR=9=&B0lF zm@64PYFxwd3ii6;Q70+ijDIF(h2PR0K2V# zs8X0B*Rjf5oks}K?4x{wX;AyOXU6rOh1Kav+i9Zi=)yXeN#l1ZR;)$rA7j_joQbAa?%`v6JGXXiIR58p7+C$0yi&z zfr9^>7fmXwGx`{@KO5O^LQC;Z(9+{UxOP!^ene$JB{0A$k-Lvm>&gmX61 zls#XsJUWb3C-38U?BeX(Y+Gg0vKB0h>nk2W9GaVStrj-pOcB6>zbD(I5}RiiY6~U= zEmCNg1d?IX!S8M9kPTgscTs#iR%nO7BU84;HhBoSF2ha(DE00uAh;Rz@MQGq8GP+V zu6c_T@0m^}KU;zfa1xVs`}I1D@sYH-QH0W&H~g0a<@R=j)-$Azzk_@sOf$pR z1ZMrmc+)g#IbMu1aeHvT6Q2APTqpy9?D+uwopWL62)pm=@Nq!6FVxV3&B%wVWIQh= z&moy&Dr%U%!@{*Ke^L_wW9CFMd$H#tg5xEEJ!Q`{4?Yii3ow1iW|MsJFg^l5FlVZT z9skodIm7b`LTSIn+W-TWzA3l0sdXx#-M6hTYMZ3C4rx#y5c^4((_s!6sxQGtjr%61oMFmfPYi!VwTc5SEh#%@ zb;I;6GE*^u1gpYVTbe#Uuety{ zV6Nux?U2=VzMB!|IcmBAJ6AYsFxzAM$9dqi7FIPjo}KSd?0j1ZiQtlod7VA+(GV{z zCc3#k<9J#Mg>mzBhCpth4616cG(P=FUy4?iFtju7uEkD%fW3?Se3mBo9sN&}^XUTG z#xiD8pWEiX4+TcIgY|}tWCkVCGb%As(XkQ8fTkGTTD=l`p#npV%V!aG(-$kCPYG<- zxZw#IkA}<%zeO){hS$BDXRPx{b+-Qu==TgAmlWM4D^LvHC9-tD^y>?Z&`tcc;_#vM+>CdbN5tfPpp7EpL zMYz~8GywJ?yhwD|WkA`hYlC5p+?0hzcTJ>SgrBzPb*<)f7E9qRleZu~N?eaG(8He5 zF-F}Vs*oM7DQnGWOX2$(Lda0Ew84zn$7&0z$A$?}jSVxmA8>qbXW+X<4T zn<|zY%;M9;qdgZL+s8OOf16tZFwFsybVrAU%^Peg#Ox8B za`Zwlo+0ZgO5nKo-j$)u7QsO=zq-GDa5~htq}3wo&SMoCEi_iV0m;@M-``GNBM;54 zc`uuRN={V0hKAYY^AoD;6lp1vEw}7nU74$lG}6z{G)m_zvvLow57O`a(w1CbB$W{s zI$}M1n5E%BR1oe_W-z9uDnp;HKiSFh_VTrEATSSrN^am=qujmaF(%t{TITgRg4%|( z;TF=hW5UNfGk z5Y2J-Xo}xDL&)h2M^Q{_;q7Tfk_B_Tt@d!pwrWSt%8y)L2(_f^qWrZ2^u=7G_U3(I z=jW6%AAie`u%Sc==nndEPpmQ0P&}>g2DDBdSi&SWpOJm(;_7P$ctCHq-u+q--FpTg zj#?=#)EypB5cVu8eZCOPY)UbAxJ7^ySevv*))$H*mw6KE1)uaWTTHNGUXA62Tb`2B zuW;qYxwOar7CCRor{rj~IRtf2Skz?<^)ny6xsnCJ#3(R>!%p&on%+~!2MlA=F>D3k z^HF&*cnpz0m!+@Y2I12gI6BA+wl&a)bRcA*BstUH0Em>i@3!GZe5R`)X@;hJETo`k zy3o5g$cfqNk~B)1v^WAHOnY)+M;!OcwxW@!R*!JG%zrX?XcguyK8d|yKrz24dIwh* zM(j-EnkpSgkB4ogr4IRB=_yF?@_BlG)KRZi8`JnuMj@!@OXxZ6zLZS z*;mr<=1>3ZvaEL-{A!--&j~94la7A4jU#X_!?is2S!DYuCzTPrpE~ooV!xy8&xN(9 zx6i#~sx(+z6y9Wac#mSD^;;p90QQ%q+|WL-EKei;~l*TzfLTGU(FrR0dklBO%sWCo9zNv65v*uV?(l%y@G{?#%Q$=hM z>~jo&25XsW>nC4lJwA+)NMs$&z{s+tiKrKU)~O~8g_*+!FN6C9~ z;@E@?;#N7tN42|6Y6-eL>H@eaT>QEF)!H^PRF+tj4RPrSNe`n!2e1JtHUZO#28y%E zR=#94*`uolyLW;9l!jiPKJlks_2-Nyr#fvgtgWLwfx1bZ+6 z(|Trm&clXL+crt`;j=6$vC?l|hNS=Z;UhmTu!rJ~f9}U^Y>bg2F*!#QEaaw}#ulm5 z@E?c{+ad$flCLvA|5^%)&2bhalDAJ^Uq-Dl#ks^EjIz`>()M^JhT__xidndw{N@Jv133+gkg7|I>f} zO)dO?eaQdtWm2#!{%;fMe-KT`0ZGokXej^lTmS374&ejk?SDR2IV+kkNOJQA6PKxOc038~kXR21C?Oe~V`2c*> z>>U6OOS$wfL|p(tYZEyD_wgeOP=HkfnQj|kpw7lk)I-mvtpP{{AFt5Xyb0i_iNOWR zXwIqs&+BU|Z7aJ$1t-_60O&0XLW8XOwx|Q(GopB-$;GLCFcv>gqD`N>9|~_=T9#gQ zzA2j?vpj2}XwXY0K?_!WhRD^80DF^*7=YP(3Xnro{s6GbAkneqX3{LW-u5<22h(YTjlb|wR3Pf90Kbb z{Xz^`U_eGZV? zisrKc-=NnO=O9K1njY~Fcmr*k*hOPFD6KeNo+}g$14Qa)Xs**NKzd!?nB&O_p7bgZ zG4OsQIeLDF_Oi*L(SLw|+(hoYeQKf+rlEtiMV$e>>iT|+1;^5qY5?=o^$hR<6?kaq zpgq_>OaOb~vnf;i4ApA15}1}hSrXIva1^PT%!E1zezw=ARX7enpDwIM10n)PWL6Aa z0W<*zoy^+{!14DKaHI!tnA*CaSB&0Z2af&gQbK=WL&QAV@7RrY7m@((+Fi6gdb+I? zAa{$^2k6pq0k1&cryqw#XU68=D;SJv=enacuZbT5*jHaC;~x*wtFW&$G5YLH+i@R|_X6H-BY=YS#0TvRC0u9>x9&sRr#`nH@$z(SB=WxS ze2@L2Fyw3(O^s*)x(lzT5)3qJxcffoZEge1@gJ{2r(f3qbK&zA07jPiKs_*3Pecv; zuP=du~>)4-RVq>_Jv<-QU+X1O)tHCqC`TR+WAx_0sa{k-kJX|sF`CC(j zi?$NUFfh*B3FKA$=IsDr)@P6p5K6u|+l_(m?E=ko>90gwQFU-kSiQuu_hTObI`%p`)~_iI2f`8t z@PXa}Vm0v?c&biaOO$>yFg_^rA_aJTl9x#KG|JwwkB!|wOeiIRG}E*bczI)!;@S4v z5Kj4jPW-F;XWPa?%>YQo2#wZwZ3nzao`g1u?w`7ki9etn+_S)dwu|Ny2DE3-lmqnV zlbYI*V@k)2(W&migTBU&unOPcRqbQDXJ3l z&7UO8(~@EI%pLRd~`{GN8r0F?~VI<<}cANWlyPnIFY6LM= zm}THFPM(-H~wGD;ePay&{a9Pv*yMWNCR&AsM{G@vWNnNtx6f zVdgNk+1|??DRW+fgFO83rGqPAVw0UvBmkcJ-Lwm-<^dpgzmHcSYEj+*8(8-(2$G@ zSDdO1LOD6GGP&H%eztiOQ|tbz%N3P(_29~{P|}_Y##Oox0HDZBs@YCkRjqO#BHjH# zN3~?f*X>$4$lGb4rQihZQx;J7o0^{-462)y+)RFC_@X7fhgMQ)~BNYH%P8=wD0yxV5%=&+;f7M*q%dN0@`11S9xes(f2&LSI5e`xSGh}QB zZO_dck!C#1CoawYd+}9WSZN1UadK4YI`*U)>M#OW>!1CeQA_&wq6)=c0t9WhSTcx; zsJqr73T?{2u@j$emqY*VYe&3`Y?c(L182Qh{ZKQXdL1XxwSMnMNDJ0Fd(7XIuLbuf z#OxA)4bh$7Vdao{Z|Q@?O)F=)7~|saxt}tJ-HihyHeM$&5bvcu7uwO2LmP#o1pU`X zTmj90zM0XizNnpfug_krUOzQU1ni89*b6L$l^u6G=V!%VJ{;b3@hd%yUReaN$Xb{` zzJAWAcVIZix$HZoEu5s)f>3|~h8qy2g(O)f(~NRHu_I=mO&)Y@^Jwj4IDri+W-A0iLhk&vT%#YH^iZ0HZJ_F8;;ex%UXjc}Ojhyj3)nGPc!c z=A|iUAHQtrPA709X@x<0E&o(Mm1Fl{KPqQNGroNG8b{uiy*Z%GE4=zzL73m=m}>`F zXXS}sDX6*KyZS`7f_96d`)$u<$aoq4RNVu}SC$^%j{?z_?^r%sya&!4C(Q{3#J|ww zt52sr2B2P0=P;+mS%aTPv>}Zv<1|uuAvsp=u;j^0kI%QK`_w#+S10 z_iHblwIg9N&9ls-4<D2|IP`l&QMpp*#}=nGwSGWlu9=erljdTbdCfp`5p6?u;k zO}-okDGsi?L;?M6+kX&*mm-tX{VliUL@tJq{tAIfN#nM&`t_=54jn)}Y6rrI`4>Ai(+Ktw`sA_$@PE>V=GNDT@q zkt!g)BN9NGsPv{30f7*t_W%|`krt!|C5ZGE`kuVs`|a$`?vMSs{9%Tfa5CpS&w1|q zzFOO|(x>kqj|8ebQxU!If15@pcPcI=6V0*)6Ug<#>Us|u))(xV(ViR+5A7VLgQ+&C znt>Y9R6Fm2>M`ohLNU3N2IYB??Lx7ui>ZNNrpC?+D!DB;NBYlmzyotZry_9fPiRi% zz;K!--&_%mW=GwM;V3r+_9NoJKXGo$Zo#GX^9@z{4I+j6v7+s3mfwwgO7x<`WpopP zuniaMrQe|oar0CT^d4I#VY|l>G?(6taED1|`f{W@pH2D^*sSiRZf6rbdR@tSa#xQFkoDE!k;`Z(F2LW zy|{@V+zAvY)@TmPk|Jvrd|k*-bpcemw*@=ralj>Y*X^jQl2N z0ObKY`#)Ql{bF{s(WmELTb^`^vHU_gPNKI4Z>2G&1q1=kY4-n^+8D4#jeOxWNWFkA z)@LWjGv3EQ;|yci4dcDaYHxcFnzmS4T)QA_pt187n|>_!9?Fh>m3A@`MvZQh8`s`! zS*N^4x!1Qt+R7#6x2_aFC6F1$$Vm$JJ|CrFzOCMgxxD0N^wwP>v6u zMO7iy#e&3RImf*ZcUOMdSHVA#e0j$J56lUlaOKHdahRy5Cp458C*$_$TfOp8xF#qmmnI}pSIv+io-5uosnGSh-;j=gP$xA} z>aNS*S4`L`@y_=S6Aob^hcmN8L~V2f$wSV#(EB3t86ML*lg6j#jbq5JxAYf->ZcB* z>Jbw1j^ch3{!DF76ET`eN*Z1-P5|SjubSP5y^`@L7Jv)EymZe2U`ZD)(l4K&TM&x30! zIH_y>r$%0Wco!O@NK_xZGbz$&wV@l%|B?&$!-FdC=-WeIS^d0xF>JY{qVilLuupRo z-4{N*7~B!p+6m6J4ucp?9)0g&vbF5`1^CleB>^pqGrqRhuE-5$OTRezH&~^OD^Mxz z8Ro)t5XrNX#2wI~Q4DVAGOm7gH54dXF%fHq>S>C5S=d4QWeUmnc-PHyLtAJ(^@+;q zsugrV;MLzdZah>W!MX#V0vC!Gjo=KY0J#Zmlkr1dqnQ&X!K1I*qBW;@w~M=m;pD1) z;iMeBN+OM#7YlYW-N_p$kdHafF~mFWt8X_awa=UEN-x+Rg;$0zbO zt=-@0YXGy1IvyOOAvH)yg>g;`>KY2q_7+o|bc}jT((rGD)NPx3C*ZGicE0fUEaZ8F zw3jUXxdWJYdejdafI0E$3w<{5^s&xK=cQ{W{5eS3xgdkn9mc9hxnDaT=$V$5x}h@o z<4heaT+1##V6^?-^E`1MH?I`u!`q#DnbduGF{4~9(|eMZe@ExLvR5V3t0`p8FBJ!n zd2Z^fmQ$#!7UjV)9Z&HIMc~BC+0f;tFs)R27GaJ1*vtEB#1esk1Z;L>lMs{-Qp~ec z&cci3TR}_Qz!u%k@LA^-wjcWO**9!=zQ_2INk66fwAI6vum-PNhN6KNXnI1R)f$#9 z_rAL!Y0UTtJ!I*hml(Y$S6`e#ArsKf)Ej7jeIM`noCLRoFgLc~-6?81Rc$6%Q#i%4 z%F>rN`wd=24xp$d(^tYBQ}sG2X?j+}ikNIUzBt#qD`uW0Oyrz=a?G->lk`&;)Vo(e zsQKL|&9iOCE6XBT?zhSH5HOx0%hbmK!s60`YmNWTl&IO9>T2uO96K&$@Ho0we_njq zo9k-I5NwJvs`8DRZ_#6Iw(T;}F;oraRpiUM$njjlBHak|X}PZju$7gK7hm3z7j8-d z(ifaH>q=A$xNQfO7j_h|nHDZA&6&|4aEf&Kq-lz<=3bG&soNA=WO66-hA(!LX0o>` zg;iU#H-^xmY2o-~|MiJuSc4$3>mX&w$3tNrT=!L2b*d(nJy^|nNMu*;XsWC|;^_IF z&lH|`C8aCRadl_sa0Y~QzHs6l6KZz;0gP0_BlDyp3g;`oS2F}?#0%323|RZ3rl|yO z7Mgl>H2WITmY#)Cjgp-;K))CQcy#J#_GP)y-IS9Zgq^Hn zUA0r=o?!V_R-7#}t3`U z(y*1EGp-6h=}t1-w8@Niy03!eqD$35dL>a&_yQL55@TYdLlL-_;`7Ep3uQ0kY4 z1Pp`;qBQ2th{WsTVuP{VFM=*H7e=#2HQ9PoL;WQNcGng5Ay1 z=HEr>7^X%P)sV(+{>@YGDHH#`O-2u9a5#mVx+e)E=`b~rR9CC`ez88KBy!OkF-5M4 z8VBT(CD7CkWdbTrwJ#Tc`LM^>(_z=5@p14)*=#{zox&oyWAWG@;dMoUnQ7q~7#UYn zZ(2N5VwIGZ(!#xKCnR4cq#fR1 zZs^Yop`=n02jaH4*nqA7qO!4zSspSBD?H`;?JzIcP1kWAaD0wIIjag7ciF7H`*#@eg4XuFPD0N2Imhmr zUz6|Mt(oi-`Pzwx)?5g^B|wGQmL8Zc&ZO5vGW`{_N^2Ks4Z_`T#-8ZaqXOi zk93FMb`5Njk-O`JI0F_*&uZ}@8@zirLD=ixeT3O2XI}1P7l}!|oFn?5x$5F-v~`TY zTPG6g{EAnII}FtBhC2r-Ekxfu`kM2Ct?!m ze@lUPwt7EHzIu6L*)Gk7HE#U5 zaqW>xRum=}DZm1%x4|cD_toFJMVPidWD<}VHr?d7?5Crau#>g6d7Tqw^2xZ^ zxxHiTPrSE4HH>gWHjm5nqV+`%>(#PV+@>$O46qOROq0;si%A~K8bGC)h%g=yOGkE} z@3OQN#!5X7p#~p|IA!(*2%!Opp8&>1hBK zs<6EhhsQuYqDs!o>EY9Au%zhw>=IKv5mq8z{XAuSEj7`u%i$^=Tp4adKOa(B1jxNo z<&7aTW?l1^5&kh$Plx^bf~sN^`rsK2{o?uLJ$dv*Mbu7&0Xl8|MxcjfT~1B7Vs}TP z%g_&w;1k9KQ(@?c$bzI<~fcF zci4@V*0A_Bo^u!8uMTaVK_`(N5IA3K*h5xOG|pb`Y7hv5_fBs{o3LJPIm;_ZRn0* zR+nWKS26vbCvPtru}*iHx9{3IXaFcgrdL8AkdyIr zA=w#be8r2-xcX^K8c>B=>$$fWJ3bP%^6m6&$w2KUJ+AG|8%eTsPqlO_@vu_;Q?cbH zrIgkJa!-vrDxMr^Otr`GJ-^&XFuN%h00W{Zp zpX4kBdND6?=3a+&)Chnq-C*3+hyvcrJrHe(3*|Jg>G!3JSJQ>3EM6FM7$Ftdl^1SY zP2%Z^JVCHhX1Y0A_)J1}agjvw1LV_qVFbhEL_;l@!7Woh-{hsaTTD({gSLZkQxn3U zOgk>%gGH8)p!tB;FqaU&x@FH88{HKHDO`j0bO44x4YP$*Vn*18GXrGYwnG)>`+YFG zqiR`GlgM<6kdA>2?I7EX{OC%5qfK~{apZB<*XQ;&zq)y=1nE&e~9Xv|83s_{#Y+?oHc`}f9+g{b~bEpT~^;% zxwk}v^PQBqwVl7l?~t3x#L~?Pt35cpX3lB@*hEC<2N}72O6)6gyyehUy>s_q%Yec7 z%O-oEbhgEf=`v4H4&>BXvTI8!yGK5zW!>re8!z1NCxJ5Y(sagWFTKXOvM!qvpVZ%- zRxJ*S*cUELj1tS%OIEZod#GwqLj}mIFNaSEPM>t6x4@ghIWI46nSRqtoib1M$j=Hg zLbHh84E3;`;+ePJStHQ4{jouQ2S(>1{{V1up0wTi`nFjVD%G$vA^2xy)7;q|v24DHQZaqlz5pl7o_kSxsOGl1VyCVwPX0|1w= zekn_zr2R$_QZy`=8ARFR;}idZcO!IpCbK`n;+bP_K{*CJ4Xw@$oNv8 zX290L?al8SLyS9~&mikz?JM%#xCK7C?5C&NQJxp^&sE9P`4I>Eb~D<#*oQ%9{>)KL zSw&sE`24GF%0`Um@9#_`4|(AA@}@!8$$VcdvEcmi@=<&&PS7*oCp0Rjm*}Kg7Iw>J zmOf4@_B^hJLTdN`yvE8aG`x-j-cAR9@OcTLlwY)8$JHW9+sfqhl_C!L2<;SZo7lb){hyzSlXv-5~ZPud0XpPyE|K*>!ph;Cg1%-SS- z$uaboYS;P_s6jTsKdDgWGz$GKpC`VtRmWM}-Q2)GD@fpb_3b648@tOxDF#u*4o^(l z@CPU{E4W@P+c%Ur2#6KPM6}(Gtzyzh%F(HU9)A~%%@b2pVGsw5yuGT-deaE7SFcqs z@sHn0IP|lUGzl!q*plND*b-d4^bg+Q?n0slFiP<#%tVg1QdcQcI~Aj~Rh``Tou%Sx zCMv^_)iCNeK`7;wbR%2?f!-~4>C zZ_a~;&r@rPS{Uu%UN~*D!g&;9V5MeqOCMBG>2O4f0nC_NOO(|HE3j#8CMgE(s`eeC zsc+NWXhX=0dlA$xl&h|?N~JjN4AfYZ+1|X@OcQ#q_hA+%%7&iCNH+@?{ujqo?RuRCy&dp_df$06XV7D(uaLiUcY$71=*>p%Osd!VfQWAx z+Ez4A!hH=!*v-<|5$79R0nUZntc%Z8ojw8@4hFwfc5tIT^brH`Y1<=F-=evIjyki~ zT?1(_l`=>Hhd4&iX(`R2i-o)mM!anHDrUjT(2qi89Po$efs7Oi?CCBn0csCTVY=VX z0slYpJ6=m9=F}|3hnuO<0*C#E{LzM*?WibFX$u@M<;KLH~r7pEv)ZpW{pB=zF!{&&ogwBaXuI}N1ot05fTX>*M zhASUsteBykgrU$B#>n43i-Mol)%9Kko{c)NA!s!W$vx%)+%-N{HN9z(u#@CV8gWH& ztKD*sBJUtERT)~fH{12nxRvsNlnpGj#pd)gKBXWOJTs+9&tbYlQ~aP8L=)a0zeE@8~WYTY=x!1V%)`y(KUl-b?t{rE<~ z%ZabwTBUM@DsGlvY7Lkjm$2ukjVAtigI{@VlfJ8DiKszjPjt@>=ZVHo9^ zqdw-1CDF!Q?8JWaJ_>yLAkS|#Z6|1EuW0ru81;+r!($WNj(V!pP!I|uHktW0vna3& z^RhW=gX@#m*9HR9SRpsjmNuGJg%z59Ta&q0EsInIr} z%a~>`Ce=eqXAj9F=84{f6|MZ=`)d#eE} z@H!5d$@p~!(O!KZvmCw68mYXk*%Pi=4r+Ve;xirMeshWr zm+)DO3nl{CwcM$uA}#r0PNc_l+*OSGf|I~Lnvx+Zne3r0`KME@R%Z#cl2dR~+a1u$ z6j|Q+Y|jf6RS%>B8!mR^UbF0ld|mxy8TCabb{%!0n~&+^>`HnxrE&XYxlTsj8ZSAc zqwjdq%Nj-31i{s}aLKm!c(S(9JR$-r`ktF$+X4?Co{bMX64`=`Hr#+!l8`*{868+&k4bfLuYCuK0ZHP~IsFX6B>$ z9z-=XrAj}s3Q)Ke zI&r`X*WBnuRVzPUltmT;>NOsVbfx;A_^Hv=i2WGBl^V zmyP!OPa`j@rO}gxkTMKKE>ljSgP+a{D)RfmQE}P+ou;Y%~Qn^4U zXDomA`Uxj&eB}iV$P>4Ct~i%z-As`DpSkz!fdPJH`~^(TX-}oSQ;T%tQmU{V_a~-HQS6Wc%Drx~vhIYJ zq@V#3YbOx1fwKcr7jpNv=K>t8*#BX=07QcuQ**&BYy(<=2p`R9nd)pW@^=bKn_FVP z_G`n_4enbu<*KrS8Obb!= zUTPmzrQvkDOVE`p_2X7Y9K%vvROlFg$aEkoksX`J9VwU%35#;=Qt};PZaT6UepF)1 zzOaNIXxSYytstWc)fudee^^J zuJ246Ml8g5XaYk6;VDRbuX2-cf+4*s5}!31?FiH|I}waP$8?XNr{_R9j@utb5JCCNo~q>c&L&OaA0I8zF$YDGuTuL zuN10B>>DM?X%sOT)Gozs@IFBvsqi3>N*k}8`{g>;TGwZ@*%5O6d*2V3*KRPtW!F0& z-P#g17JJ+AXe;>WSudw`I;F6oAHW}yk(H|weSM-1*wJ#+aPeCr<+3XWM`D+2ndbhX z%gC)y?Ykpg;>{KOWp`aJtTJ3&5HfrQ8TvpXmGn+xH0i#fA%!jpmGs;IflG%0t`nx- zzE#L#Rke^91IGlB?+WIc3#fmn@db1|H=iH+ zYU$wWh-P8xQBUrJv7GG@UY&=!qJX5!NXrihuZKVQt1Vn~3?rZyrSesE0=Vq1s=BU!yC0LJ2tj)4N2*&(V=kg?XP% zul(-<+7_2RihK$oiaY0K>koqPkz~XZAS{+Dd>--NqKpFS9csJH}do*H8^%vFID?WUG7J_lF$kcTuOpu8~Wtvk||ZtId(DCuUcLdAp$dGLk`gO8mh4*3#;3$q8bW_PDBVAG8rP zcU%r63D!b=-KNA3yLoFvqUkM{%k_57WE5)hRNhTV$Q}J;fGDVg4b7jGxtQAaRs{L{ zv<8CvoZU=INxb|qQV`TE*>5NzKZ+g;Nrjjeq;(hZ#~Za5<+bicT`DJ%f(#aIepuhS zUVz1{0tzTwJ`ULy1M86OOLDSa9nfjCJtLc>w+ry7J;* zBJvHem~T4mtqzgPvQAeTyB%14x&LMoZ%LIJCEXMXX#45?RlUe_3rMDT#vjsMV=(!F9<*mDn0>Z#hvVk< z2O!6@FqPx(%14ABg{imF^=c3-u#y>JZ~a<9TD3!ki%VR4*=l$ih=> z$|(kPWTf~f;W!!Pc{>Q`XuQVDJa6;m+e5_#j{)Lg$*us?kZ$C#YV(jdNC^?IgW&Z) z@eYk&;%WLULg+luVL1hGGl*>BHzCx$Yh_Wfd|&T;fo+767QD z-xpcSd(w$cr6xL_I>J|k8-5zb2-r8i_L60TlwE#~Z$Vdb@C@*YVS59t7Wy0?68^+3 zzjoKQMI5IO=`@6I41GO?bnjyKa;-J%B?nyZ@4nOWzI4_&>tp=fKi1ltIlpEjbOur81#5DvuW{VU?k11Hw&HAHNx{h-Vt+0 zzk;+WYJaJW)mllWn!~4gepBM>ye?`phjcpR1+3l2F5cViki0gi{QAez3Ao$Q)&$0P zH$!b_iuy!%!?Q5acujSa!(mgK{8ra1dz}ptNt<*BJy?W|1Ynp^U8dRuu@`Q%f+yudn;U=j55w}J3{Wa|FSoZ=ObCW97nz2=G-`p zAEcN{$+&EE+qm|s2-O6ym%YNb4b?tctt`kjkSi#`AHI4(d|GNT0zjYo90J05cz zXy7y8A}f@*A4$b}s1hcF)|>J4At^UDzcD^ZViIzww*`1Oa`P4e%+wIPUEaG_kW`{0 zL7pHi&U12Pe)E%8z(VTI3KAO`B+dG z9~EM(&lFPEUqD&qHN@mHM7+j~G={0%Wiez#3J%=Ec6GkgiU?iF+u55YUax}GK;zgf zwpRSrt{X&7ROt&br%*O^(B4bo-ek;Hg=Iw7xO$x{ex|_@`E!0o)p@Apmz=$lKDf%2 zGnyVkcMB#cI@y@p>jPB5Bud*2YNS8FW1*Pbjywil-9Ub5JH$3nJeyJTB~xE+T7C)- zoZVm~Y6i0pI6@Bo&Kg^@cATqY+PisM-@C{jFo}F{Y{Ij3SIq+eNOmS`|C&(69)SSo z#a;g65uchppsL%W%yLPNwwH}bOO+d3vw(`P3`{IcY?-N5=#d{8Nqy!Ph@HzHNqiDnv<4<%OTn(8E+kKQPxR*TyL zF14PF&5a7{u}&DEx5%(aL#WUs>Z6e!*o)gv z*#rEu)PQoJbg7HITIV!tpHT{jFZRh-qrm;gJR)c%lgL^CYvH6AH{DRTn zi1Rd>HXtVS1FbMb^DJ^{B%JNNG-J%cK9Pz-h&=_X{JN-G1{zz-LNLeH{N|9%Y zq`jDWI`aA@--eCZSoCp0)^T5emnJs9J)4#@=6Jo?!`P-G`KtU0MfvM|L_I)7N25ohbLb{rPq_UAn}4=Kd^kc(+-k6#w{<2<%#5}tRbI^r z4`X1KpCA272>rxf`6cC+x#NdYRAs+OS9kfviaAKiWH0Cz=sM86$WJEYhN**UpPJrF zjLA)<3T9|?$kMzVaaq~%iFSPd>A_sPn(TCLYl28HzhtC)t4vCl?3kjak@pK(F>yFJ zh|Y-0E9Y{G;Fcg+!eG^Mg-yL%Nsen;>8CyQJbDjdsdt{0ISP>HlcU5wnF;25(}E>w z=cCjB-80^nj=y5JiIb%;Vn%ET;O_Zs#c<%Z8iaE`4&eY+J;QgE!)NqX!`$;&oZYChs;>Wr za0-KqhU8CV+Hzb0jF-pG+@aE%X6hPd*c5aZG^D<1tO=Qb1SGD(^vMb9*C|2An09L? zM(bOz0966_AGT*2eGXq{jlp!t`?1zOa<-*R1p?)m&1?u}>Ose~ePzi#lRo>ae0E)I z36#@hG<>)CC@|YDlugrt>D>d*_hB*_nMGt#xjw5o@EMSMQKwDva=eWEvwT%oT=N$6 zUAdCw8?ilFCo;#K#ZEbBbmHBa3Q_uj+f14+>jw6{E1rH8R5g( zbNhJVQ~zX98FTd5WKL>c6kt!a>9*|40zu2s)mJ&kQYiU>o%`&x6B(=-xQ;$&WZT9l~LJ0VHeUdS8xDAgDI52crU~UDb*`^5c&zH!@cX9nO+Mm1T2uws*7v~({X|k2`nG;=iDy}IrXVL^J`L^_s5f&WlgM#5`uuE)SF37_TG;b zxPID`z!gJwVtA4jFRoXobH+?*MqzhVz98fjp<#%$^ZonaXQKyL-T7O@&AA{e)XI;i zLuie&LiO_^xpylLkQ%KZ0x!N-xqENsaJ>kTH{&9dTO)2eEQ@{2__z*;J`A!y9d7un z$h_<6u}n$CTXEuMht+%4epayWfpRTnHHz-cdZhx#Bl&G+*ythNoczOe9 z+5WQ&Dd zZQ-RhOMIdt)X-t{_NT;tlAEvi;NJkPy|U=f(NfR%m^(>z3BV>8h(<4AjMy&;>7z8& z*VXGw50Z1-zUoddPF=gjRx-myxp}&lV`*hDHO|HX!);gF>eR9um~b!O{0&qBTmUJy z+=6vE#HkfBfO{2_HcCSjmUB%olVh3AP4ZuWyhVL4@i)V8)sh-fp?`F>a^bgfI4!OA z6|BAd|CD`r4Ui(Ksv=E)y+(+;J2++x21J-w{JDl^s`>ua<)&g;d0k#BLXT(rtlD~R zk3Na#1MUOD+}%2b$APDs&nP3goLzVVGg`oL^2r+k#6e339ZpxdD8PmP6(-9a=6h}f zhM$rfzRt1|*-N8SfrEbjv*OQ4{>papf17|W4QJR$odyO}?>S)igaLMEYy z@~6B40}R_gZqEq20D{crZH<0_gP?r%e~M&{7-=a9RO|2{y4d1MIrnbIW95uQ`)`Q6 zP+9%`(*W%*9dE+0p@4Kq>bJU^?Q5JOk`NbnFq^F*8eq?m+gIR4lAhwoI`+@vAkHcwyq53I=WdJE-oy0ufVmo{WDtL69 zEa1HN&ZTZ~V0aTz%<8k1+ZMxhuM~37gvbD-fFjJfd&N4uGe^Jvsb1Y~oo8Rp^P=Q= z%}l+Fx_b1nV}v`8@&19Ct8>6S?>n!LQPvv&Vw3<|XDN-ON+w}2!1;8c>BtmUh1qz51dk{!v~2-}LH$Bt4|~KZW}Lb^+4%_ Date: Sun, 5 Nov 2023 12:15:32 -0800 Subject: [PATCH 13/13] Fix figure of llama2 13B in DS-FlexGen blog (#4625) Fix figure of llama2 13B in DS-FlexGen blog --- .../throughput_latency_13B_no_arrow.png | Bin 176008 -> 170030 bytes 1 file changed, 0 insertions(+), 0 deletions(-) diff --git a/blogs/deepspeed-fastgen/assets/images/throughput_latency_13B_no_arrow.png b/blogs/deepspeed-fastgen/assets/images/throughput_latency_13B_no_arrow.png index 1f846bfdb274d65c033a6542cb89f56026009b7d..cc7b8ec1ec05f3ad50a41ad8d2c99584a6f69afb 100644 GIT binary patch literal 170030 zcmd42^84+>SN8|NA~3oTx~>|8pt7y>hJoe?E<;n_pPaknF$z0Z(@_dXGHY z|6U@USpIEi{a?#KEr6VoqU`@%QVq7uN%+5ud;kB;;wCIKX_1boHCJPgNZmCZH12Yq zZ;i<+(D>C%n|Lm|NB`H_Yr}vIjHkoKKWsW3-mV|#_G;R%o8M_e!<%f!peZ$N_Q~^& zuwRHE<&n9&y=*-Pez7&BZ#I>y6u0u>5+3^TVhh&V%IlJC*6L3#=xA{B_aFbWuc}Tf zM}gh^TBN-JD_@x0_bPkd{@kydFmi5AIN(WWzb7u=&aRfsO}D6heY5WzFkv1>l@FEh zEtU%>=5eYX5){3?TD24@{`PIPq6M+TIxtk}d)R!rvZ|OKzOO4i%ghi8*)x#L47i0) zv0g&jn#~Eb%_ds;f{E6yZ+v#kNKR9Uf?<(2ry3CZiTu1wf>1mfsQ+7^(q29VWP<3$ zW$T?^#_jNbLWQ4Ee4;;@H>7=}*v62^oVb6tqV?driEi8S>8x#LhJPH?_dsO7X5#gB z{oKdn#>xAW7+od;YOTzYgHXc)SI zceTVv39+&&vxDzEuYKEuNSu8gNVF0sQu2yuGsZA z&_ZPIqs#m!kF-UAOq;-L6U&@^3IA0ZX7ug{<4~+d#;9}tyY?eD_Mm6WtMqHBR3enWZ2v3-wXQ~6Jym6+HTM*lZ|XjII@Sz)40)g_CxO>TFd4o$v`3( z$(JgdBazlv7^ZS?(iIE!;)l4Vi&4RW1DQVi6( zYz)8XLl|0*nohb^-nmdml@*#1&3^?4W8j&Bxzf4Ri>{x?R!n+8oJ?5Pclk40z4pZ^ zWfwQ*!$#uEz6l03UoQE&_!Epurn3d8r-%74%jwN5aGh(K=&hx&T@W1zz&yiSG2&tW z$tsdk$XVVe#6c&M>%_m*cm}z=0@5kp zmw%Kr4{K_i=l1y(v&X2E22 zoVquIJazx?)_}XOJyux3svem|O~)@B#15+Mj#v_e1)XzCp5*9Sk60#d>E}VetVrlw z`2K1pycQ&!7rPVJ{{`e{*M~;qOGGJ61!@eT~YddkA(s!>iq1^jy@z7si?6Aqrn5KX}O6vCNOxxbn?|jNj zz0J?r1oNi?NL~2PL;24W90IFF1yE%vMxELvHFUbY2Af38eFHM2s{_B^Q9H(VLC7I^ zQ*66nN0WUORb~8@X@8kQ!M7x4U>;=r_w_`h{LIAfJ=DIrnFNp?cDycuX1f6k5ta64 zYjZTEE+5Wv%~ujMo(nEK?nbvFA}gO6EUr5HY2D4`MsP>*U_f0xGkP$B=@9UAm^)Zf zxv=+dzEoL7ezz#>rVn|^UOWBUseZ9;(9hIkN|%SIqO*4QF?43nyELNp4taR|qZHEb zAKj3|R?*~|V*B&=UHHg*7QM3+HdD*C!28<;jBE4vmCGz`|I>2`fXi*i*sN~x!~EIAiEPF?4yicPS#{dGMwV$Q2kZ67^n!F(Wkiuh@7-SY zNVmTh_(hf-cYAXDxm&~a!B0yXYYtI_MN7rk!&HazFPUsX;-pf-cz%MrJwW<7vp*YE zF!Nqd_#Lfe%oLC*jz{){gfpUt2#0mH`ekv{F@E7;Q#<6ke+TlbnRq}mqc)>KVJFqi zs#Tv1?%T(F0r!|O5kl(l)WQd)lQ>Pr)&Jd3AX&~l=pMK5P5tdA`w07PcHZuIsu7}& zI9;Z5=8!v9s3p*YO!?m}G`JrUK+GGIKZ!D=lWdNh9EptRe!>Uk4NHHn1ygrq5V1fD zl9_0}MexQyF41cm9{Tj&q{~4=J#QI$soWWPzB^N~R{QqxCiApq%MipmXXpeLUSu!= z_ls|yY}9~@=VF&<0HyX^iyfArZ<=xd@`oVfU6bfD3v~KnT<2e53P!|qudL-Rjdo^P zpuu&6n=EyY{9XirSs;D_>Xqdx651~fBn~okSaWTT#k;PT8*XkPy)(L%k&ttfjF|8+ z;tuu1$9M)ImM7k!I24HoN?B@EJ_QkZbIrY|lt!?f4-%cm`_%o7!X3!BbTd{uSvrS= zuk8``gINq&o-~B&ggsV>>A(nP@mpd-x)}(e?LqwGoZ3!YF(4W9Ra3qKA#Jlp*a>SFYB`QP z)9#peZl$t)J4tvTl9tiMo}o2Uj1>~ifNku5fx{QT>QLP;Gkzav?0lAaKiVVmBd8^O zeu2~nRR;^gX#Bm3{53&^vNGL7op{98vP-fNv*ClhgW64!_M=(S+e;y-lG;bVEu*{q zmFMhoB3DFCW|%`OD0rRaD3B zvw5dQqx)%t{Zpn0s{QjT#F)g6V2(8b5vosyyYxJ3l`g=xs60!;fIP6yK34t#?J-qK zYw<<#-c_-cQPev5ykqsy%IasuHr14!1wL-7veNs=lj^Jby`D!XsjD+VK`IsXH)0ax(>Is^&r-m7u8Hi1Iu9VWvO(0;X))$|?gZjzaVuH&=OSRG;kh0KkknrS}g z%euY+*D0kz$;-X4;-V<*pWkScE)HPPdomxFr*s51+9wyW8n|bmDwgxF2J~f;@ zHJvn3$+I}ax9lexK~-LS(<{X`6~<{73&D1t9?EKCzQ?M_K>eW>pi+5->f&>+_(*xV z6PZh7h;OF5y*5r`PMeYy;}c`UcbimCjYvv3D>RFy*=0Lu++U0++MVDXs1^|IW>1B0utVkV^Oz?>kqR2%P zbk*D#x;^79o$=ARL-+s%suJ_)nv9^3o%Eu`JZ5rc$OF$Y()bZG?4he}3g_I+L=9{K zeQ_Ia7S_HbQ=e_#$IrUUz^t6V8~a|QIsv+1a#YuG7p zA*wy}!@BdWm(13>N2%5f>d~*(hWo9r#i_rR;>d1@v=mNsTdNolc0d1w{l{u+G<)oO%qE1bTv3Wl+l@51Jtfg4%R;IYQVjK!akPo|S#oYgUYu#WvJa0( z;Kc|l-2NEY6v#IQV0^O8aDLjmRDb1$-SVH~>MCu)3|xvQRSsJpXY|edJjgm95e~G@ z+N5y^pcORKO>d3mOsEfZNHCmD^DhZ{ACva`{>Y+jI_YOiD39!FxrxAy;#y+Z+*nf4 zB zra%}S(X{e$RKGZLZ9utr==bp8a9_Qp+^&$z5SUCkp_-T4Yx!f|&em5IP9PG*q2~kv$>cxgJ+~NKS*=B!5_%H7h zQH4aNfMyG*IEZ>2zL`l=&m+bh;5XdW?AGdCg(Q_Zv6@$c`x;Y=5RtkM&#>Xfr2P5R z`s!hC=PkR`%;0X$mTJ7vMB3H$OzW+H|8Va>QwUp3P$@}u zf3Z#W+YeUp#m@a^AV1mTo0lRY`gIv@uF#e0jCObtiPkNs9HJE~RJ^;^cM25NCYaZ1 zWaS>&UIpfXfRt(HCb%XQa%N=N4>?+W6T!jPLxJRdn3Gq_$z`K&>_=neRfA5On=Cey^LZtCu@7^?+`fPaRc(|V^ zY%{z^J5+1!l<-=ivL$>3d!1$4-WnOmOT=>V$%1-^v9(nPy5a%xNQn)&fLt1ua?MG$ z);+t_ZJGi5$!fevoHK)1fWjRNiLT}4_VseTOc7Lm)WtpqNvunQS?3ZI7FP8O zER9QAXZ;!r;ian5eQ`*#&S=%$lpZ(oL=N;*N5+O+=$*08F9Ha!2}=sAO3JXN^1+an zrr=QT=AYDrgdc_K@%`GW-ameNsZW_u2nSGFGi!m;G@apoTSfwO0qhj$+2|FZty)z2= z`oz1;ccp7Z!>3SZp_9!%nvp|?O4HMG4e{w@*Ixkt*B`T@;^b}0+~ub8f|TUZ9D4!* zp)!`gTz0$zNzrSlMM^l=>CKZyQLOf+_S=Bq+0RFoUO9q`ZrH(Z90t2I{;j60Kk*ij*dklSTV-8V3pTs#MpX!s9#fhKw+8 zvUy~bVgq=K+AL%96M}=)dJ@CQD7H#d98bo*2%HOE>JPDG&coDzYOOPSfSpLoPOcU6 z%C(=eoeb*gg~4@{kGWBEdyzv-wlafaymp7*)W^c_{{JzmcHaQse z;|l1hvVVul{h)$)ztw7iFkhT&?YD>%4nU%0P26rR?gr9fyb$YRR+knF7;_!Q7vAhk zqU*OT894^G7Ae)R5DUqc>S&$GU0#z`omu5ParnM5LqO}+n84VOa->Lnzdnc+n)AC} z2}Gs)sFB|L6q#k9FeRZMYw-hONtZ4Zka6RtZ)DA#FNwm@s_CNO%SF!uSfWNmJ4HyC z@S_y_GaNPicZsP>#v#apOn-#dM8U@@j6a+J;OqKsFRiE|rTbda=ITV|H~LN^Z%`K5 zd@1R>nBNPkZWY)&kzhH4yYP$k1)_Gz6T;mbl_XA}N9quzmSbL0g%q3_O~urc8qxxr}7Ba+* z;6aGj*#g6L@y>3@qoU#pU5vTxD{(C-*aXd~O8nB03WBVAu{&D^(2!JB%~wWoVQwUy zJ%0%k{)eT1lp0FFGBoU=a$?JYXsq$hF{h*TF&_cy@dtBKT#_4+?099s*f%Sjnf71| zE=D((q5h6k>-V?Et35q#0E5J_JsqN2KD^1YQs9)ek*4&I1jKO$Ey$wo8kYNZblTvUdXKjoAQ+N8a=LoET(?$WmSceVs zp-pTAf#0*m2N}?o)WgGwKRV~CE;t9Pu?o~*MwomU%}-7)mB@a`*Xi-9P_5DFNQ{nD zOJ;__!W?~F%&Zqa%sZ(>%4*hwOk$b&kbq{E&~5 z*|e6a`qW2<_w4H#+0AZ6ST{J%_Y#6EmCvaTYQOmg&L!J-@uiw1@g}~VFV2!k zyi8fWRi-l1l;Zna8SrAVPpXVP2zXWfFJv`wWtZ6GToQFzKqM(FnHX>uJXd4?fFd`- z^=Yj}LjWaKaMtiIE;wu_Ve#3=cNpPZEdq6Uv4 z`K^K#_7AC%jK~pgGt!kAVa0vx8U=YPC0}}8h5gW(H(xxTO3#51?%G7 zjIL0uXU7#J!NHvpFq{&U3)U_A2)CFmj;q2ynj*7eQkO$`SNf4xkFh%6%Yx48tIBl} z&+}dxgQP+%QciGtQMJkYe6_e?Z9r44qItz*f<+WpjNPBEy@ho5tH2i1}2MLGJKID8R zKlBf&s_lQaJHr)<_7lYvDSlta$uPm@G8^;$%Qk7RekQ9cS-V*Dgc-KVTVs>)K1arDj2$Ytk=rt#H&r)D zSK`|J1b*x1LJ`YFLlFBN;)r$bTxS4k`7MiFH?A^3q;M*6$N|DsG2GfT=F z-yXlTNi@g_^LTGC-yhFKO2|!Z^~SxUQ0pyzYs0(g{dySnywLnGY{k%w2*={@cw7Tk zO%vom>eTOj@5BYjPe6d-EuV_F&50C*>!UpB>vv8zB8P04xDCC;4dV41Yrz zxt;@*Zd4Egz!m{yO~}94*uOMv;w?5HYncMvkh=KsD+Ein*U+x;z>o0xcRx9h ziGfh%7IyPrti!YxM)ED{r`~@$J`Irig>M6jhjO%mi{PwP+?nKeNj;x~y8YxMi!{2ZIQA1TvgFjA-V|$T!1KeN|F9CO4PIf!9_FLNYBSUWpUuvW0 zA;2>i+Hq${Z@~bWkY`fANVEvsrVYGG0sG9y&ikIQlr>^?rsbj*AhjNm7{~YTQYB9bxgqx%qrLFMl5&cO~H(mMA8wmUzQ09X8|?M!<=)_G#iain|nsCLO|=nl*` zTybG1n=GcQiAL{~F4@&8oAgb+ts4tmncYU95=o*L4XaV%^K8n19;F^_deI0TnL{;G zP(H`I7WhU39^_Kps4CLouMD)Ga{&3ykr{@<`U%zIH5o*nJ+uS9!ivvS%kUCR%*#3vQ}D{~JTHu8&seDs0~-mj#3K@~|Iv z-FGN-IpkV-(eo+#O?Zzk(>jhb_67+? zF`IJDNQ2AjIKwk#vH)eqoovi`<#mpHE%YLb`B-2)j}N-$oPrfA6XC3oM7lqpi_vMq zJLiy>8>8Y6XVSwE<@@j9A$hrA_=mMd5ZTFSbds%F zncIkjXUeIWKYZQAgb`{sJm985225D)34Ud)g+H~BQ73Di!W{rp4=ULWvv%JeX9P-I z?X)2qQb%aA`_wdJlZe&l;NSJ-pW+$8Q~O|dK@t#MU9u>G711+!o_~)P@M1N}$!p0T z)Hm^%(h6?D9=G%|53Hw`{bd;~Q8(x!$8BYG-X_A;a<$^aH*J1ID(YVN?#Ftfbp$Th zRtz2b5e*~XC@F>vA95LH9#oe{#m0Mf@48nuZ8MTAJ$zSBr=o|_En*nZ+0Kk)SY&0a z-YS1Ixy10|$(-0=g1~VQu5f(z+~gdS)H72N%c}*bj{U zAyMUYCYT<8TsE6XpH#aM?iRk#fR+B~_E0&e8jgNi(%=trk~aAcH38Cg)f1s^jchVz zo;GU1IYDNK=Rw6R`y+`}>y@P_{=F65q!YP)6p`Eyyf&oaF71y$5a%|7SKtw5`` zkNV$FKoJMzlwlbB|4du|{}TV7_vLuBb$ogv;4sb3Vep-H(E=#yz}oGV*$r!$`)`23 zRJ}fVE>!fzzorv%si}m0qmYQ{2X9l%J43g*0^s4@(KHp)Wkx+tWrnp{fUaLUnaEQ^ z6dfEHxsEl0HxfHv2=1H1?3Cz>rvYfNh%dIi6?Z4UoOEst^4ZLGt|fj6%+=WEI0szM zBl`7Q2r8W3CV8K3bhS=a|6L1Dz6+o2a60om-+?BXpyAa0JNQUI4Ipd3Vk~i(DR-G) z_~DjgF<)--vrr@LB&a7)FsROHzA>^Nf9^AlAbN=SdrjyOMN%`EH@KOFi#z1VsOK;p zLx13z6~^2b!D8jld^o|@jVFxPktNZW_)bzDJCn?r#(?51iaeV@{xlrm4&Mjb=``Sh zc{f`TI!dygc5dDz5E(bv>&!%^{d86=Qqn?Ij@J3O{P?l1G7~^3E-A8=cSI zc2yw#eDSPEFMo6#^4tVHLnR8Q8qLO@_$S!q>OdRQ_$3pM8~!Z{{$6!7XbQ`!&bt`A!)`r?=z0E$hgCyEMb zr2pcGLg>Su0m{%Y_r4RSwwv5x(te(HL<+tZCknjIcg^k$3o`d3{n8u%z&p5I)&`p|CYg*WqHi^&JfF4dGBQ9x)jDhB zq(%CLv>?wA$y#&@pKbTZ-DSWX$vU`BHzZQeqiukzwGv&V__RlZtU zzR8DKMt`C8p3@O28Vp)sKhxa9S(xxcp?(*jGQeZv(zX~+bJJ~JWBdKPGpuD>Py00( z7|wiFBZXED7y8*M(%%WJKa^+dz&->9&z;F4?Z&D2X8_Uufq$G~0wm4Fg*k-WDXEYT zrazG}k#+wBAOaO&Kq-4IaX#r@Qp9010xb*16=iOze+T^0`N876xkw4E+nLPe-c(7F z;O{rV${1*>ry-Ke$2dv+GEB-Cq+ z(vBN@k{L(UTW(GUwshoawkHarVM0Ts-}-=23=Of)6+o7xi#YIXO*>$TMD30WS97k! z#cx?#sCPamQ1q`}wp`3P9AjiiGh;L4H?`4zB`d3g`TH8(4d4?VZ9;uHAHPDwz*VhJ zbq#8l@xS9X1le(@C$|Sb=r1+1p-VmqPI*sS_5tu@2~YIL(+nc4zdHl!tX>!{)p%Jx zB`Kh7Vl#%Nx6wM9oC17lDsD9jt@#X?RDwh z?0u2%eF_LmvSF02p8=KVHI>V+r`yxxUREu*eS05ohQ6|>L!7Mpqf4C@B-bP73vXPm z{uho!p4xQ2(i&{}e5tkpUh`-B5f! zulArv#}(R(Oii+dhuAZIaK^dMt+S2j?{Xl|JmL;V{x*z!cwm#Gs>cyK7V1;t**Jo- zER^yRy|?p{Y5;7s6~Q+=%9vzKfbrGLcV9p)O;Am2zb0<3-Z^taH4AHFB(qi6yX}X+ zbfu+v8z4x$EHkQ~M(kEp=#?4f`KpP$q!>yU(UuLx*1SAetQ{8JHK40_fOtP&=VV!2 z(MT@k-((6tmUC+O6M8JdyWUf%nR5nMX5ONVcH|DFv8W7cb?XepnTQfQCbYZ@s&11R~aB zr>Fq5Euk6JQXp2RRVCq7EKR#=bQCF;t*oi9J~g|Jo-OB?AIVK=6%7x;;n!@ugpB# zo?vQmj$rBfU1_;`ZF4##qz_cAiL(43t}`Z{+kk{R2{4vL3{k@Li-zPBWjleXj>gyU=YbMUA6AEl=uZ&ww3{IkGHRoc=mP)P71= zKQgog(I#JV((|uZs3%~^T@R1cQwv|`F|>XiNEyD@%a_*ob?E-VRFr8x>(;Zl7i#0c z@R$n#J#C<91LgHzD+Os8&T&MiQQsd=7jd&_w&l}aKwn4O?{o_}ig*x9X1*Ywm)*i0#AnbMC|Cb(k(dIp zLqJj1PxG%Ag}<3o4Xc@B&0E9CNxM&_y)VTZr}k#6C(4GR2g$e&x2#oT2kBxYgIv1V zpN!z+24;wPzWd-)5UH0h3QRu5z?AhOI#LfuRV_kqwL3B#oh>yRu>Wb&$_VqGKXD4t zr|)1!(+~`er(xEsc{+kI5;I@uS|v^4Z{%}%PzE&XeBT}1|Ey;xO%=O*FGB`uC8c$m zC0#LZrJ4NTpO};aeJ&`4WpKs(9;7wDiFR@z zf{bs4PX<$gZ8@Ry@G-iJOteMKico&}?%;o(s*lB3(2+$7BS$h_?PFI^+Pm$j2jRrZ zZZ95w5$SzYCT6cWP)B~&otBF4f9P1lD)o1Ff7yMajOL^5!KWs(eQDLLn(vIu%H>LZ z3F;qXo3~9BO5uEIwhZdCF^tXnaT&LF3lCrV@|1EFXE8-x_V0H}y5p!8@{Io1?E zUiz8PY*su6YXCN;pWze-Gx&&^h*=xE!wBEbfS0aD1S=;Jt6pX3O%{Dx?HA2Ns|cPK z)*ZX!^HNDm!X}8bkM5ROHqmjR##AR6FZ|qPuO#UlUzwhY^(DIokueDGp&$$S4G*(r zjZ7=OwX3G~Yxwo6*1-U0PLE|BZ&pb>kW^otz@`7!3JibQ_pIHYKi%(Xj0cLSq2(CZ z{y@im`uCBiYh&d@VZ&hr{Z^Kl@~LeV-RzR;cOZUWUP&MQael$>qtPO6H@sw>zuk7I zwDj{ga*ote>q;UW@sa=J_4ew|T=j#*Zb_D*v_u=l6doc@`GG zpGMX9+NjP3KBFBDY%9H|P|lri<3(u!8qK-l$bLs!|Hrc%@pnz;D@Nn193p@A+~x$m()(!-CngztE15Q@)v(BxuJ8k!*J6OdyvYhTaMmjl2W7&4 z8@b(?r+F76zeF+)%N@hx5LDos(=L^&>jIZ%B^^KFN)8;QiXQ zrR&ORsS0|d07yzED}PO@wH-2!xYBoyi>7bMmN-_SNlT*YHXX>yWT%IcKg9N5SS6MUOP^yHw5n=7>B5*8rEX?pwXnfeyW*9~|NRl2EcKH3l? zeXjn5odW>u*CSt$cuXtGAY$yu_k_!w9ItjA9`9|BViEI;2$oJW5Xufz^yf%u&3OKJ zP}aQ&WHkTE2;Cwy1ULOGd$KO4rdx7c6sC@C;`EcLN2uCs<=<#aFDB6`cj>1RS^(qu z0{vFD`Fu@uu4e#?A2^I3x%zyr=d+Yj^~9{c0}Y)M`Guz;n@g7C3w~hrd_|x6n#SZQ zwtJ**D~E+wu|2j%rJd&MBA?h1o4YIllhkWqE`oMPQI!BJGHQ@1K^fSul2Y^lCq{XJ zzYu|4fuqUyVr;dwJ}v9_PWw>?T!`cJf6@WXq= zIh&E;S?m#yuY+aAEC;e^jp@r;)LpR;;c})Ci=-3$MI0vijhmGRK=yVg+bD$T<_TQH zO(SD#`Y)tu4QV5Nl@M1W?D(A@_7^A3%*7SsVZE+$@oKJhdIjQjSqb@o0Mo_8N((y#BaA<=81 zzySpBR7-o|L%@_ud}``GCP&V17t^VNdcKiC#9Ad6TA9wuX*b4iTT??YzoPuVA16AFK=O3DX7}Q&j0qL z*-LeUyH})7L1Z+~aS#CXma1|2x?=Xtt1_aRJ5A-2RkdgE>TzFv6+X$hxeqaVbDU$^ zU#`Zm?SY-S*q>MEoC3g84$!-c^eaqVDhDAU3joC^heA~-G!|)8YSzVG5&+;3ALRK= z&gifRsLfNZQ1$o7#c_QVxK_dA;ca&TVb*Va+!Nq<$Rxm+>gQSTS^WAeJa7OEBhlHk zY(aGeG)-44;XWHFwiXS2K&#gQ#5hf0hfsdz21#;nGvF|R-NK5OC3))2p$ZbY1l$Kh zNW;n#n-IKDwHjZ@Bs2H@D<~{eUzDQxEqctzDM+}d^tWZ5g-P8=16r?-v1}^-b;8G- zWx$I;PNm{PDtEa-Dfe6cc{~7{cZG7k;x5|G7 zJilK81(`lOENUsdjEg@*#&Q&PrS9O-OJ1uu%|{(iG;O)87f?YKVeiUqBel9pYd7@z z*U+Gu>p{qOKdUQKpNJ-r(cB2VgpbQAq-Y_>fIh3_=m84o1qLo#!GtS(5m7c*68kT05k_h8VKMNVF7*bg}59;Ahk3_h?Qrx;I@ zij!ZYA`BJDHnB@%tV(C$+PQ|i(Ac35rg5s#u&DWSxnP!lhV$hW$;u#@Z|WjsU#fXd za*-x-zv9uxLe>?s-S<%?hD6^0Bc>iOsusJqPk>r%jwirLQCNT2or0>$te@wQNO=gErzUY-U zlO(A#`M@1MnNKXpiRsv83U4e1Q$9AFn9c>)8-jRgs_Q$yHIn6uBYP`o+Q0TzJ;LAY;Xa=orAbukK$XFJaL#0lU);Ypl0a|6*ai zA}3ypagd80(aK71&DR4y^(p8OyUbMqC8@cdDt!I-X!@zFcc^i(UgtY(vrQ!j>Ti$y zqQ`F%IwWd$;@3dJ>SavL-x6~Uw8Ed(haI}rzA5Vz2&g{Fs@Y9|miTx4c@G;jkcrK1 zLB}S|0=khHtH>mVN0*`Yn>F;H*m3{z+ z=XN|>uQO2M>hB)FlTcrnth?@K_f~#dzNWyoMi*zK17P7-4vq9|*)FcNMAFlwziQ^l z+pj0+M-^c;o2EXjd1^X>ZBi*akp+DZ{>v+lA z68`mOE3qf@<&9yJes46W4-)eI2-g{FcW8k&`TKGz+3CzS)7~ZqnH*`7k%FD?zYI5z zp(cxzJ^`TVG-&=k#~hS<97S-Kx8C|>fKU3iO|^|^RaUKp!BywzU{3AO}7Fkk6gWW4D()qUv7h?|MBnM?+Z4tNc-2F)5NPOg<7fi?HM-mgZIIybeBo zFg-3Gy;7f7eqID1pR_z8(y&?o>;2UR2#*{oUk(c5Dga>&YMe^WBNO}Jm>$yBz$#j< zlp-eQb4g%7DKS2%{tWIqboku-K97YOHNn}0pjN%wvmdS*3=?P+LcKu>)ZSJ#os=%G zV#?UEXGzOms6$I>ZjZuTaGqN@WCE-@x$yfp3m7O`k{=x7e22V@u54rrK6->|jydn6 zrP=QYdPpBhUA5z#WgLQSk3EY;xz^q+%YfRBr28ZE1{R-Vcd#2~5ufH4FZKu(?-vAU z4TwCeVZKS5anK)HIwIiMYkW#JWwD`Hd$+|~i@eDm#Ui%T1l+Da=jsK1-pv|Lws`gu z9AK0Wwxb0k>IxOCty{Z%g zC^(O^F$D*JLY!6)Kuidq_RF(Pj2?+~@U^>jITLpOMqK^ix*bkA>2%UrvKlFxFN^~5 zoAFFOIC(u#r}V8q#pKh?#5WJcFUEO~+&W-KuM6yDTK1AAophj+GtQK*V%dW+(g5EL zleZPn{>k4k27!If7E*G~U>eGxJHLfY9@^Ot1%T;NYh7BEo5PkP_!N;p?HNSIWSHJHY>X)1;87k2R;$vdxY$nOZ@XOuBYFc`ITIA+Fv1CP$HxC+CMd-4W%$_{U|3Y$p(5 z?#E-1i!!buj-vyB18BdaVq@;#C)AsZA$u^`ZSnMtIA#ERz^q%8gS@ONMGxhOAQ`C< zJssp*bu4VsWO=51{C3mlj0ig!b+$wY?EQ42HQD16-9GS<6M*zbHuFlh=vxyY7x=S3 z*_0Ql87uyf?PSj|7jN68gBy&7&B^V6X6fkB(NKz+p4Lu3E%(w6h51PETGCDeVNwXR z-~fPlg91EB^PqFSy@4Hr%sBn;Rm}E5(|S#(pa>!k_3_QS#Lzn)tS1c!AplwASJTT- zW9yQZX)xzEaWw~6uONvDKg&_1=bKyFIPAptN#|4GV<2vf^r?@iE!wD4p!YrK*(P=8 zk4ovom+?R4OaKR=NURNjSS%f5J)^>&X1IHS1J~s};zK0@cmVr~s(UvusHo*jj8TV}wY$IsH-h=)f=Fb?2 z=0_cPZ3i2<4P9Q_c(nrPp^_4I)h+Z*8i%uS>sw*)-D&1s1lH;-$Tq_|M_1xaj_&%A zBfvA|BoqP1&QTOt@x54a`CAj8Rd@U8hOFQZ@p=eIjdu)_uT#dF^d~a^lC$<&>3bB! zQCJB3L#kY-J6ex<;XTKu!*4bYHT0AnsIgx5n(p4INX7SMmr(AT=0iU&f16QGv`0z! z&VyOz#KQo=`t_hlgw%i;EtJg$`+jv0{u!22nC%N%OxlNI4^d`7Ld>@+n2<=()#~4U zMiGzkI@j=$KRhox9y+RzX|`7(Gesy@h<=ML-Glny7Egc!h9Ac4Cv}1Pe3B^B(;2x4Q zH!v;af-+$5qAVQgL!cyiwl{5<6>d!*PFpMT!B=M!>Z@XpG4Phe@^9X>`*dLQIgnuB zXe+3a6H+ANBOj*R`3!PkyQv+twYSV_^7_A zl`>D&tf50$0>Jbpi*&FPOj~w=h9)_29BP#5+y_Eta@%b`wK^-81xZeM_}K?*=TaLf z9^Pn?UxKe9G2wF`5(Om{3@MQ~CQW)Zyt$%EVxyPc#!LE37L5|mE3-NS#DB9~j@h&<;{4&D zDapMK=OojC;#CoG#!SFKvoT?7DI-iyB_rLd-UpZ@n-j0U>To%q5KK z!)NFb=~R_vu`iD0PAYJEWB`^Hu}6OH!ch%wi0scSbvmgEE<$T`X2EGtB)=+YxUf|U zCUQ46nmZbdIV3NFxfo97D*!bGlQUKdun5*F`60W1;PInjqBWf0{_$L{dCva?F}Ze` z_venOIY5^;ncEjKDY6y420G2M3e`m;E2X8d+7M~`-P0z9&q|&{B#QG?iQ6iBck=u+ z==E!zr{A)(gm|7|OB@1Ik|%#|r9^9DB}xigD?#z>uX$QKmuowBwk*^E5n$)=`$UyA#`7 z*l6I<(T+vN^9bA!>9&$^d@NGM1He3n1TaD*pKmgb?#yJ`%cVw30$(ad$T|5*lC>Bf zmN|R>Jwl?LOb|I~o*m5`VTIBhW*+O4<|H!aT{N3gZikRrowJNq&p)TO zlXrx8Gv)G!3ZxTcOJ)%KWpv32DLo85F;A?jLNq=ti_1aD3v&mH8@FF<&%+XRqT`2Y?LMb9^u#{{%tz&n0!EGec)OI#iI}-ZIEw6}%@eiBit=n(>nqu6H zyeN~(amAd&xX@5j@#CQ^rRU}h{r4lZL%v)A$HvY8)WzoWkf0N)0S-YAX$q2?w+TK0 z4;Rp{^$~spOJZvi4JBKc9WFKJxb~##dtm`#zU1-|jN)^Eq~d$Cb3LA{kO$`@U}_ue z?f_V&8>X@5{(BL!q@h*#o268rpf^jthh1c9j(~PpYq`|)T6a;vL0Ag$LGAw|>n)?A z`u;y&siB6F8ajm`q+4p}W@u39mM#gUhmb}_N*XCelu!Zb7zU(8kZu`3X++X{#_#X0 zb=Unr&_}w^nSJ)z`yH>(bTe#uO(xJ!1?P4mcCzJaCE>D{avh+3a=d{y0Ga(&un9 zdn>2IRK)M)pE>o>_4`NY$@E`sp8}rLNJ)k2ReOC=x)RrkufVxWCO#b^Vy9f-{33cX ztNW*1k2T=m@YI=U3@J9Upwl_kWVGeUh-GufbU9yfi?<$ap%P#4!4Y@c*3<3aX8%AYeBmDr zQA_&%=*`;`;1<0NC~irs(}2}^-K@YB@O@o)_54;s`&0hWsG)*ejUO74LUm=yT`p+k zSa#iw(tH3hsQ&3cox-L6{OXt+e`WcGu5UUWc&e2VPc~_{_daL&MiUsl^Td9ZZJ&ww zl%cX<5+=|0u#;&O%B|ygChzKw@XObLUO7P-9W`PLwm<}P3FC?nQ1g@meS;sn~ znRxkaxkIUWPwL*4bs{Y{Q1r* zzrQkWEyb?rH~Ev?U}Lk}G`K0|h&qVhX9=J9DR*c0)9_bhv7Fz+mEm>MnNl*6qBcR+ zT+9v1j6sH4e~Hek?rga!yYZ@6i5!!~a(QU3#wFm0#lq7OKYJgf&zd5ro&pS@GF;>x zm&sM5La0~Sy4m3&ejY=>T+n$?tE7$rFx{i_`eL+ zSA=~$ZCaF%EqC~NiUhCQ$*&lj{<7KKr?&T!Remj>4+spJCQB_TV z3qXEi8$MFvU;c&QOvy_B5)nfBn%FffnBGKb|Axs=5weK{C@xxRNC=dh9hbySy^fu# zd^z>;$)9GWi`&i(;6dZ!xq20$0&7t+wXI>BNl;$sDW^uJe^83cj@MnL}9 zL&De^NE=+Nv;D31i8y{~R%hLR;r92>fKuRy(iBvTE&Tlg@{EV#MnWQtnYq?4uJ7~^$dP-1_kqm5sOLS9V2XYJ zmD+1|C5;*^O+=3EsDFL!K*^eWHlv$dPtkZ@jJ+rd9QinPGpf0`6_3E#EYaDtgqgJzCu(37wrnW(+ zAnyf*ml2}Dn68JwWhth}{eQ`LN04UMoqAXthSk_n{4saTFJFV6vMFs0Ry6|6R#b}<%tQx5E@3OohpS~3*= z%)6v)Kwi#h;2c11vCaGm2l6luMNcD`iT$LU-ns>HjgCl;0Zxva;2RQ`$BU7FujMu} zRLe|&!U*=f8NceIVmMYztN8A_;4y=E#`UDIWE96A_pi^bEH>kTYz9>dPkXS=^xHX+ zSFA59?&*xT*p8bU=($+Gy*VA{VvNX!t3tvLG@SJLRfwm7-wID%W9beUylgq ze7yS5^xIw=kwiy8+iLc}y~3+vL@4uPtIbYu(^RELYvZbN70jZ$JI%`Y>h6xvQ{Ex` zO996@-+tx^3VfZ`n&`^tPQK#AFzlG<1!oD|L$fkVf9BNS1fDtNmwC02WU}X@IT!2;Qp>O)Z6w zTGH7ii#i<3(Sx3~FFp!0X?EqytV?oB8Dsa7N~;)f-}kFBg2|qY?5uAVhQ|8ccHg8E zKbzI2d}RlI%0^8ZuHBvPf6ETys1rl=TBi3k5y#wH^M6}W?oN;34eLwZ%<(xPtKg=} z_fFdyOxkLeb#=m&)Z3z#WDv@v2_fJ)|36V#ljZl>kG|+|R+z<-Nh8w~@9#`-Ey4-r zy#-kO)WqgrGyflQU1go6jA6bwEME#x!p!bBi7es>NI4fDL=mzHh9AEiOnH-JXZ{Vv zq`Q2ccJkq!_FIc#b92}K`-3g)-sekLUR-Ir&-uD{*9HzG#fm;U9;x_hkQMPq)qO;= z;`gr#^S7_0yl$_XZjDZMMNiJU>L#!iL0XBZE&p%m*Th1P#MCvD_89tYo5i3&nE*)I+wqAPxlI0;Xu6Q|3}y}nHRjzJ_5N5 zcqStZW1t#+i<+Qc!qEN{0T+1qrobeO@Q(Ov%g;OGs#)XFFFiR=f94H$@w@)3s{Hpt z{cmyBc>ORiaIqXYH-0ps@#s^-l$T?T>YZf6Z@7z}b+0dZJuKh0KfcZL{vWsDzn$)X z->{xngIkA$hu&?*49T~1Z@EwBq^^GDxrkH%38er1k^g>#Xr&Yf_X_f?BGjZ}nx(?P zL&thXSMenY_G@cY!DffAnVM;BQc?qQxqx}T8}1lcBU+LCWWoWbnZTv zujIxu3C$Df?%sD>Y71qQ^?wDBh$P8a5yar*rK4r!<<=6g8*tmIsR##em%TH>0# z#ScDaw~}SFdgr=`ys!r_6fSXj!0QcQ58xE~p%}11${MT2p7ekBPhi2@pux=8l=4iS z^|yLQ&3l=WoIgej-(YiPSUEcgFw8Z=UrWuu69U^+jJm771on|EvB*jZuUv`UfEwq| zeU{(+Mh;rM^1Dre{i9Ow4R16iOl{r%$8cU40DlWw3fv;g0c^tsZW>|4gg-!X|)Eh@cI)TA-^Z~5#{gPGB*j;NFg zAkkO;X*uEnAlYWdle4@AlEj*_rvRAcYu__!Ng$1VQjNlEMInf(B_2&dQvoI-5DrW} zi%U^_{PR0TZ6>$$7Zwq?2w(^s)Lv=XZi$7TAt2^Ed#cU}d8l>(G{dcWGCpi2|ErVR#QjH$+B@Ghck1J&>YlXr zn#QG?k0dgP)8K~_bic?G-A0MwY;MQV+l*%!Dl*Ut_I%+$I3B=c&w(V(DW*Jq=i|J0 za-_$6#zcd&E|Bry;8CQvvPl;OL;l4fayqZnh1*Y_11&X{IeXOFC8L;AVo~UiJUPi> zppHJ*^sL7E^J7snO};W#R^W~FeGe)Om2Hv@isu~JYMYFial8Ee%aQ3@_N1iyB=_~@ z^51D}VEM&c&WD7XNf`;ihxr*OMzLZ;vZDKdYvY&)r1);ry3U;y|GjdR=n@kCGC%g; z!`0nQq1l2xnqcGbWdK%12PkUl2a?YN$AI$QhmL{xYD&s%Z8sOYLEr5Y)VDHUxP`3cUq1E zJeNcE1>-e1S^1jbK_PX$MOL`_^$TIyxhbVHh{2A0c(o{(f)?& z_{i;L0ddfK3Or&BqT8UxHPO7u3?*B$nN@j6O~l{Jb10#|=_za}OJ*f|8wwjSoA%PqzeYxq-j<@Y;fkUSnyE1*dV+E7Bc@ zWY&KDw5`B%MSnJijgrROud<4$(earTIaozSprn2ue+qe6^S>Lz#F^0NpAC?0hg8YwaX=|xna|J?fUwKb zxf=Fxtt~XmTN(mq3lsAuqWxoFvmf0>G<}?EIC(OOtW*!FzZ|joBC;NH;8VLF%S)xQ z;qk7WuIe*2$z3LP%{+D_PBB%-n|rDit>X=B5m1Gb47H6$;Z9@&6whm3N8l~DS7&?R zLHA@74pH4;NBk`xCD32dqo2UzGT}=w#d3T64?TFZ{!s0dWgM!jw$;TWr z!8=ph*dTrLKx}yeM^FJ_MVgfqMVt<&1WtUkV)#X08{%+2@T>mUf}|nJq-EgRP8atR z+cvdH2(Yv4&aDR;ByNR4J?TbKyN-y}4!AsJ7w`--xq@-d`hjH8rQ=D7u*g@iA?yM? z<4AYgemghs+vnhYh~hr*bVtbZPz*Sj+W)>?x+g=v3Sbj&CSrd9MTs!^4x-b3@F#2> zZaM%+Td#*ZCU$!W3;2Le)z9IKM#ktyAa2$_W~gv#S{vY!U5amp?|cV25KYJm>CB3Z zC6j{IU=9|XW;SlBh35(`t43Si0j9p>NMTn?F`(0Zw^amsDQ+7*G9P~AdprVs<|Gp4 zqNl?}YfCKyfd7p`H2eh^viB4_Jv*_kXf#zno1sD7V0-CRJ)1MtzPgK7<}~nrbxRyx znG1>cS8#bTk;eRaHJ(YxJI8ymwW5Xg_!SC5Y>-v5 z>up)EEQ{>h*4UeGcM^ZWKo#j!HgbM9q&Z~%jWOUi>^=hM2P8*423p%NvRkn?!>?2* zlx@50)UVms%Y!*h$}CF6IoNKlwv^KS{W5D6Wz&`woV1+W1D_d#C^@MqW$|PSAD7<@ zCfERppY#eR_|ajl%hA0H_+yp`lNQj-P#`6z@rvN%Irl2{MVwx+KR8sxn|>iAl$vn_xuzAP!gNJo)Bt3$OTTO;e#Z-!ga#C(NkC6nY7 z&(UM;T#Y7%4|A4j0&(gRzH-gHA=*iHqky~_@4VUndt(S>3cZ;tMF!3m9!=bZfbS`1b^6b*t_5CbCm~SsBo5 zQULt`K@wbOFSbKx@nF7Zd5-xVpKD0`cy|r2OP}*txoBue2ZlGLR8-1Uk1s>XvN2Pe zr0va1`Q)B^A+QWK%TKSPQ%qNB>6R?EbHwmuX{u;9m{HOKz2a~4p-`jYQJ_7uZ5a8S zHS%HDIvJ?Pjs#VbH<4$mjay7bLrcx5;-#|Vq zrAKI<>ize$`duKm_@sa!wyQ9D#Xu$&0FtpKw}Yv?seXuFVXHGn+M{W(@qQxpM3qrZ zhVLLKxF9)!+@6s#m+GjoK}0#F`AX z|0w^|m4^`|u$TG-R;?jh(gtyb+qHl)cui6cY&IB?)NxTB4q9bt;?2cC(E>;yJMhrx zkU00a4cwxN4S~i>Qn~Swp4TJ;fL>+Fp4?{~{3cQ|X_pm3VzD-$(^Nvf##y*n)MeQf z~EDrR7pHlhE(~}CVJ=YKiZf79k*Kv>-VnDDm>~k$>p@< zwz;=+I*~%3_2hO1w-0(N*3(Qf=5P0!%SNRn6_BHA(RJw8GA>l8UkGZ0#9h|XV=OC7 zcd!l;lTmJE0&x`%6iK0~RoBh^srPqCS#e;+7l`$sj;VhGY{bj^V1|T+Av-hH6Umet zjOouRK^animV%GthDW7e`FQ=sTWQ2mL;;DS_<1Z0%v`5~yv599_w!g|$fF8UuM80Q zPCM9?z#$_t)r}mZ7rH2d*G?Y3^@FhGm8>>Q4AJODXzg}JHA8L9hT*S5DrD!N{fJ-aRdif#L;{yMm)5 zw9g~YJ}>v@+2+}g0pCcCB;5!_%mh9*F!yupYw9V5qI9AMDbqgwiPjri{K>J(Y`Z8H zoB{?>qE@0LqWcgy&?0+VEpsd8cI@{IHvD&F@nj~HdKI?UNwaF(sN3wSofnVRXunsb zo94bZ_F1v+kCInWAfXMiAVE@t{tl*7ri~}PPZmRH(t-H#om~dmr^Fh`9Kk3uggXrp zSdFL=a=#9MiH8W$6tZ2d5E?-Tc9VlDW$3;7e$oMdqzS<6@bbeT<&+<(hMx_l8j-Ez z#fu|F<`_gY*ry!7+y5X2NrO3&d?w__UZYZ@hE^?#1LcXO_@%G>-a(Uzwn8dPDQ}%c z86{CZo?8g_iycO+++cba#FJR!8ITMv z&Qv;xS*N})nhffLb}YO|T(IC2eNZe_wU|f`ym*SBF*O9_W>Q4`(MSWVIfXMMvUCs1 zi=?_k`fl(qu-O|CSFpV={@wFFVL&{I&RMMr)K|q zt`oeX2Su9TuqJ`CbcaDnF(}iF>TE@uRoTKUJ5CTv7C413t5Ha{z2D9mo?5ru$-1aA zH0EKalOH&^r=qkLvF+MdStSGqw*>FWem3C4=uR9ij<)*96vxby`qe_SN} zfwX}RMGaaIm0>5$iFqoY1uo;7=ZKxoHi(=2`+I)!bcT0AASQ%v<4qqd12Mvts8qUq z=fa}7ZTcZ~;T;8-Labg|TC5&w_~{8ziJ}+-tt^==O)QDO^pR+zPllK$#C+X)GvXt9f|?DP35Z_|?Bli4HAp#saUq2B zV(cN0nhjNQ%XR(1&I8Q}b+L*G6JTo`StdaC{4JGB`p6GNjmt8|Y@LB3o1915okgHlaA^Uy!Adyw87}n{iB*zcByOc+7 z6JkHb5`BUG#1T5tqGdJ9M|zq7qi}SUGElb49aKNZ9Ah+^h{Lm1&lU5Fe&%>tQ^1m(mmpLqtMT5uGFk!>|A_p%YF_z;u0b$L0p7l z9WT?p%}F{#^PwRZqRAu`ujLkkZOam0+yi!sYA6p8I&YP5m93kmlr3GD2G1s+s5*q8 zOdfcbDyvwk!=34ylk z4Np}5fB}!r-jTB7@ORIxK2MeuzL0}bkr%jexZRYhw9ouY#}DD_r-?m%p0&fIRQ)7) zq#E{L=4! z1}QvHwuhwUaJn?y&9YXJKT!?QrHOLt_Uf{)Y(&fKZ^0(Im@-1Pp!yF9)rlQxhM2dM zXUl3IzU*8O7?{Xz;yR0SP=8q(=sd1*a%YaPmz#t_c*2wNd4Q1xbfQ0Gf7W5ViSIh~ z*qHa~k4wp`(iLMR$R7*^(KSRQ5A;Q$gvNn;q$C}u@uNtCzs=n93xG_=pen0+IDGWa zc+Z{VO?)^(MQL3lV(-Q$@(>UP069+p!z4pdac_vmN%w?3g%x5bR+xk#RePY^p+N{s z?-@9CRl2l7s*9?W8m`#k$Rw{SYe;|s2?L==js?oXuu6@{E~4ycK~i)VdJSWbzQ(XY zOU?Sb5vzTiN6YTPeu{Wp;7o z*u+Z`-J)!w%4gN-t!efNfl^rr*narWYC)I63MRpxpfEj*Q|T;rDZVXDQCm?t@(w5d z89LmE?xY8s>#V;S4@YM)1ir=#O7YezG>TxBcjrAJog;MB2UCE-Ml;CFONVJ{@#oDx zbBPhk3T2nR%#8!SqmEK^zXnB-ISgJ%mT{A6-9Qy^Hs*!QbF4RRRJ=YnB@Utl4xL}Y zbJ-tN*--FjMFuMGG;ImfT-h$5TXq_ZX$VXmlAti7VKPnT8SbfuEzYRn`FMJZ?M_~R z3wexHCsr*q&y6E}wOy>0wfw>*HS-6)mHSfD`|^lo$wW1nVB*A#O(ctEk<$amK8-g# zL|LC=I;h&OU*jG5W~58HQHu6{O?4)TxuZjqL}+75RFT&Z*Ol6XsA5|E?hTN2s@0rI zPvAUoN)emPgXvxY{J6xYSW89^dx|X2VcJsh`;rjYr4_}?tdZpTi+?D_l{iKg_sNeI zLeBwsj8XD6q>Zwgk8snWojb(>qIu-Yf+S60W*%_uI%eIx-q!gMdGEh?_ zSuAA|T~Pv*=3CtFS7grwibCyNh@E2tTt$c69otTzT=ta2q3Ce?l`ne)6=aUItjy3X zs0h>^@}Zf&PL)Ai%sHcTb_+a&lKrF0kaXGDU}S#dA5L?32now;k@^)AQ|v=OSNyvq zeL_iGtb(ckIYNvAx(IbS;kytf@zUP`dU$}SGnK;5L&zXYTyM-m)XCa6#{Q*>X^7^8 zK9JE#+l2^LNzZ=l__WI{PMls$gxtbYl|s1^C4!6YfM|+D*ulG?(eh~>@n0DXp2*B* z3UdXt_85^7O1O-!$ix>U$I`WuqnK3aT}B<%tk^jW16Y_b26Ef4Ftv#|W!1VkVv-H= zC@HX{434qTh6h+(rl*vxloU6N10)(KJd}!O^bPmhX3ASG;QWB*oKNX79?V#)|pNX zO{u`r$mE9iWQ%dXHElT^hzyD`)ezimyc6P+M<@Cx)_aB5Q+TM?o(iv1+E`y)kGO35 zjzOlzUmVFn;E_|L8qky%O~PJ=?Grc3vLa<=%SN3@Hfth)}mq# zGJV~*2{l|CCW3R_<7bqDP%<%qsM**f2yPfjq2fpd3zxE6eDJ&*WNo`6#ooreyL{5q zH*hdVy8nf{;eIz^=LC?4q>jxqLodQ|s-^PeQwlWOelv(a-`^w|B8`!c{X^6jGT_YW zjv(t2NE7jd@+na} zZY+hA!Bdl?L_WmVStK_3*x^0E39UWd>y zBje2tJ?QHN%QdYnWAXSZI2wIh!3^ zRknTOxo-9&dK!tj;){Oe9KM6uL#TOPjveSy40t$&%v}D7ag7e_4g^Nle>h9I%q%uk zzIFL!tQ%9c3c_6aiJ}Gb2KVxM8T}>2oihUR?)DEef-NCt<2B<{xvPEk;0y6A5DQo# z22HdvP>ndmIcF&(QMG(r_)}Q!7I|doc)Li7pLap47~$O%UW^CEI4={=DsR>tN3|%i+ zPOqawCfiDXiWCF^KiKF!1TWsQelE5F>nt-E|NCey7ZoZCS|lpWnsu55%eq9C7W<7& z@ybz!i8)uKima{FCe|l?)a>l)LKu;_Y6m*cWgeufWtBvX*Nc^#7PYaygxxFC%q$k$ zuxCn3M{>OM(s*EGn|Cu<4@fHXPIvi-^oHgTmrm`1%rZ6e;GOhK)c2ulC%EDT>Pt$^P|9h)IhjHsx!+NRw_S(=MaUL?Yz#zEJo&)1a7g+)O}u zH?V($j=VBtV{c`9SK~n15>McJBy_lvxPmqgkjEmG8Ia)tU*(@q@cmQRta^xIN|NCz z@Zyn=J;!O3py7z_kl4fMG3GHnRz+$XAqr>Z@w+6OL`$BUR{YL-U3LfymLxEU0jLPb z;SvzPgCUhi^DJ;>Y^i={3BPl4$S4znD<@{AW%9qlyD8JMXYXE#tzu?hYrJFJ4mI7oAWB0yD(TCg5&N( zV=(lBPZ}8=L;-(j)SUM=zU(Og%j>&ZpdPX|^G%Oo`CDaP(S5mn;xe-X7=D z8rE4W5?S-5faGc|P+#z9P`X^gT^FR!tA%&OY;Bp7OG$buockcLcy})7w+4KT+k$s^ z1MUm#u)K)284eAeH|8`8ZFXZ2OE5L6p?Ip~zFMw2O{TJaoW|U7D^d1_ z-}ssGzPHr=I=saQ&S{##I~RDJU)>zNr%<@}HPwCo3M^D&Q0=-ykLPawqt+V3HFdWU zJ>D-ip4q%#hQ~ktNo&F5hlCsO?h=$fb|om?&*B~_=K1i8BWPJ~=l!Gn(g=HqXxmNC z=K>;w!LT=}FPt^i^vikCb}*P$=Tru*?8lGQ!w1oaw)#1mdE)+`jGAYPEPsfN?NPOgr}(L#fK zam?|M`eh*{5fbN)FezVdAw5UCDH^v=aWh|V?so#vZ%`H z$dGnV0A)W_9G<$syNr@HV~SA2Hav0}_Uuf_0U|=2hdD?7BMvQmc-I76jK5-Y+*}|6 zD%O*r{($#|zboV+T8$skb8(j?)(!o6Z(_;JfQ6SYkYb-;;h&NB+)4yBMVt6Q+(jh$ z&LSM|y)=HN<)%Dq8V=4m<2mh?fRiPd6upFgJ6vlIA|G|+2CH24V_+w@PVRQOMMXq) z;}eR48Lc7;)_h}nEn{No{H2{51pR~f%5gTM1S6@K=f~ZN-US1R4vi`im2-QEFksAp zL!MYWZgZ8f)BIpEvq9+h_?zBqYQ;k2H=Hwr5FeMvt9F_!&)GtYwP=TXhIaUhiR{s} z0w=0_IhSibhf6g%wP4AKDjD+jIDfm6b-Akw>t;F>bSLqs{DBx3sjI$9Pb14B;24JU zi)T*=yNMSFJ`0RyjXlHw94lS0YT_Y(c|(K4f}=SDQ0Qo?9)NofX@&ABd1WfG1x0rx z)gg}WgeTn~zui4kD_keCln8-2J3HFy@5}-RC%(ZHBbX=yjm`bW82E%_+m(4v>j_1S z-WrW5+dP={PJ~G*@oBZ_Wm+1@GEsXV&oHkTv@kW!&qDs5_CpsqDu|kueX?W%5#T@BM5L6(%aw-xg zo1iXR`L%->ZgV;K^v|0X*0bk)nuzZ!kfC^KcGmcXNORYxl|Rqrddr`$!--mEX8KiWJ+)Ki*tbJm9V<_Ki@vs;?H^9KLV@{mKpH`@!*(Fp2S3VmJ8c`uVZ%qMj`F zGB7;J-bx8$PYem$lii-sq+J@EG%>tG3SQTwwppi2SIVD@>FC3M#m+MgN$Y<@+g{62 zZ+VxGrZ1S=f>h_PHCJT(lb+szFda{;>`Unk2yteR55qV&FT zwW77gf<}|OZK^4#~BQANS5@^*i3pWFiLJ+j|D6UNs{3$iVFE;{}aYB)YeNm zWR(Bs%jsbZR~u;Njv!wBD%B9NWFib-1cQY`h5;SQ6Tp~L4Ab!US)kAmrc>f!W;yK& z&D)L>B`ht52SGWYx1j~xE}TZZ3)C9?D@B&*Y?1gWvLO;@72+~)(P8+qq~v`w&=813 zeCS6c;44?75A%*D*Z}Im;desxfb9i%v#iK6gWeg^PhF95+9C4Mp83_1o37&5Vv)xU z!amI^+bY^+HpIGGOab~knNRx1HJQFjA%;u#jXs5L5*q5y#s6mS<0#d|2v3P#=UX|Q zkPZEtaZ$1QVrWj;!iF#Wa(?m|uMWNSs{_Ss10`#@6ZrN zH&xYEpMzT$E?K5(*Af-+V%u!k9EQ?#(PU!}WV1`@hITz>CdfP#cQmZbaW)9Uh(~oN zSCa{MUik;^sThnEZ1{4jp255|r$^C&1S*l87dvl&2_c#{SmwExTAkf*nj-~&@vtF) z2U%{DHR7wX_d^^-z2SFMw$8rXDEyToV}4>#jEl-y#X{K-4GxIP1Or99lHm7B0fOsc z=o`cgQ$d?n4k3J@0jH6 z70jrL$wdB0OiX*Wr?=?q6`0vZjy%|>GzZUd-vg;KpX%8Jj>$t{ zEw*#dKdP{)M5?l_-E+nrQc;3Ms#}ch#WqaxY^bNJ`24Vma%^D@8h(7V!DmJvT8;M{ zd+T~ASiq~#z^TtfrRV(ScPSw$_u3n*nCaT*PId zszR(?N=f=;GzSr^DEcYF1V0SPcq9@K52plLdSYUfmD*FTbi0iG&jOC<)p2nk{kZ#5 zJ4|E3DS||wi8dd`_PeyOCebI6#3-ORj<|>%?UbohTU1MxYkFi^x_^P$m-_<3f$IXgoX+5ar1lBOAWZ@PemM=SfBNT zpWsxZcq6Z5QKk}eZ8;AoJRvI*-5Ex*fyG3gWV-SkKbP7h*e8e(&Y%y$O7HSLx46Vf z_%Fz$1jvWzk}N3O1DC@GT~Xmwh1w~pASxcT$7GJ}+QirjUuph;+lOztq76>rS*^Hv zi3g@K&7R9}l|X8cZ0d(cnP-;oXlCEiY^>+0k*o7)fUJ!QlX~--saVh08}DdL@T0o8 zuDE|TWu0Uda~JP2ZJ2J@Zs?}K(T!~;BDdMM%5?2ylv$lY;M1flk}0MmE;&*$#S4KW zW;vP*+5#pDL_(!f@FC`m@f_S}|EMutc>X z!c;qWYVYluPy9e%%Ycm^kLohG`|M?g1KbHi354%E=3XE1Esfw3p$z@Z(#%g2oh{@B zaZ{wbCZ(S_xB~tz(9iVDW{71rD+RJ_ggb?dCX2JR(mV^}P@1oyxZrYDFhH{@z@Qe!LPg zRKQU5N4K+d&lj_ClNlay5X;ur`bzMsLLOr?8$x`+w(qc){OiPGXzYUwZ+OA+igF@{ z_}@B>u{D_srN&#LwPyG;g9T#eUaR{_Hy)p&szn)_aCD^;+^f87k zLWZbdR`KVJ!75pHu#lY8NTdAonkjyCwSrVut2ZTn*M+rOlHXs~B70_zOoiTEel+tJ z`nx5HD=fnZ5>%=b0q$qh)@HDyQiK`=7wvg9FT zSj-u(oG3_(?Um<=ELdTRB}gya5By8{Nr{@h78l2f>&%MG4)RG=#O!E`tc6qnv3l0n z3m8lJNk;L9KCz0s6u%IYsNc1=zJ^4a%&MWy*#B`OcCIv z);R8@F5?d&@4BT@;Xny>ICbbiBUieGPkL_+g!?P(#d;uPI#d*9-JFkr+WO7^ekqAe zTrNGSSiX@m>fz^HhnmyrejlDpfsi$VnNzg`VkB>e!uS?`j$Hw~>&ZJELfz(*0S z^|xdGRI*(u!+73V@Xq(MS2--$AF$eHVQ@d22^ahXETqr&T5HUhhnTwWW#0p+nTJJlnavv-44z*VSB-4 z7Yo5piAu#v{&ra=xI&t>y2A`)V3Bezgs%U`1U?xxnF9PMFrYr&Js)c;q4R$ zn&$LR5fDEG&^Atwzm=&E>is?7rx{Vt8y%HDv1a2}lpXV6jk2W?KC2K*wnI`RxQHCaKD87}=bY6XYfDvd&WsUWG^l@4_^8zpnzVclOyFIh&s#+#& z5Z3*w7I=Ip>vYMGM!^e8!UT%0Gn~Q$nMtRbHfgJh%^+WvJ-j;zxs@BSYT={foMTrK zrcfEe8CC67A7?l{^lns-4uinXU%(L0lUi$=5*NNV^d&$ov_^9-C2VwPK(g9o^vk)d zkjFCkX#5OqSmI;%MD3*BVbIcv=YgEDZl>V#dvJ zN6FTFD84@V()m`DhqQ zmoR_K(z7BhzKp$F6ar^^0PFjJU0n1DzT>SU(IaQRj7>ZC@AFBDMI9!ah+5-T1SN&yDNj?xKJc)6)Vt=Cav1{B?1eq`kPr5>R1gr~ z9ysQpvX5Il_OlH=|H4Q>FAH1s@Xq^eA%(Hf7iMQy(mr?Qm8wDkoVf$ zM#2m`hpQ= zXd)b=9QB#PS+04qWlj8A1TUK%K6X8fi}Up?*d0{y0A$qO-KrK`7EXOAA0bROMPRvLf+!|xb+_wQWe z_qx<9n>a?Dfz!yY3|t#D*9j_K9kAw}hEm?I(z2c={TwougEk>VSRjkSrSfWT+a%xj zxEps?8Z-RD1#04J36tu@At0&OEBCy zwC3l{qp*Fc`+3Iyy+FPx3*OJ=XLwN$z49o9#V(B_4%dNKJy*}J5PJA_J$v0}pxvVX zaehy{Flc%gromLb)b7>v=qECEf+QxSt~<5@m>`&UE$zoS16E?Fu|$pCip3vbNm!L2 zHk%zUaTV*bd3g$dumHqVMIZH5n)@t@Oo0`dvkQy=^*fvxc<%R&F`ZKpRSKlz{plPz01RfUx$rn*}NT-HzP1Haqf z&YWa4F6JZbrf@WpQhhd7q3r9-MjU%f#Z$c&t>8Sf$_QQERMFqB`8Y2r-m~Ja$s(!F z|EAGD!~iH-XxF<%OE@AW4$0gnUrt^zY^U8G2A}h^@#q73=iR&VAtn~=7bQsoZxJKD~^9!rDZFh9xxD^l37rW6Q*D?$ppSMiUO~;Hni5 zW@Wwil(yzh8MC* z|E1dMqh-Z&&qpA?dSb`J#^P(iI#>0Os5#KS$moc}S779XSs~&JcBr17-O?2MX(uf* zBLmeA4G#&2ixuevS5hw7WArC^VK{k^qtwVdZiKe@#B<_rcSyj7GUj&;)#iEldtQpg z9z;TFq`tFT>5iZ#^SkeBVdIQ(9w*bFf*i>T%l)L?|LkUFWHJHlJ%*76nL#(4sj(zL zcG-TiDgLbf(dt48SBD^3i}A&p72W$Bx!P<5=ph2&>Cn<---Q0X=2m03$H}t?W)Gt- zUoryC{WyAtC6m4Zr*31qB4-YZj&&omyhGykTQ=(0V#z}{Wb;fP%=Kl}i4PR@WP1a>EP^LJeDu zqd$pdyJD}SHl=^>&AZDj4-|ytE->X^-vS?Vil}61+>|1k|2mTYZIITV{Cp#x4j!`4 z>iHL?CxeTVD~->`z{`*4wdRiGYQ{dr`cEo)R$DXSmEsJf>ZX`8`3QQ|tCuNBsxzK? z**iH*{M8j|;TBpu{qQbaABg?GjW|rU#Q@ZRH)3gLxuTAN07}vw82vL1Bx`GdzK68X znulIyyoA#nz+wRWf2q~Qx%%{`**lXLygFW=xph!gXS^J@c-yIHGVaCV`lm^RtCe{K z&Z{woR5a~a0O)~U(7hK^1b^_~e)b+GyX4iV-bsR~-Da16+0WxRL|<^2i}L}ftGsa? zD+-ePn#5p!eQ|2?&Zhk*KGw6z-NaL+t?At{e&-dl^xY``o@)NV>yy=>$%FZrJsHdU zB8K;!w@MxqG2iU*eAvw^DtXMoKJ#&2=Ye;g;C15=|8^RWR=L#u`X6J!l$$kx@O}BH zaYX7X0HZqrn7RFs*%!oB*f9%0P`wV|5p?=xPXn<<2Y)?R;DZPFs>;dJuyadEts6pZ zrXs$e9j=kQqFSrdJ?F{9zd)RB4WDK{76Kq&jYVZ+3uQ`Is}$>U%7t=4 zBxKY4l|vywmB_&ocg>x8HwXW&G|fK#UcB3+x$+vADM#vtG8w6U3#n z=K$apS|gk{wH8&V>8HHbuEJcl{?qh$p>^!@Q9Y*ce_7ZO)56beyg?$@K5sWKU_$oU;I1{ zGe1=s8|NM_ASrA8Jxl2NAy6}VL`wIF*L)W`>d zcSTl>mP=ZU3UZ6N##tgesV{*4(4*M~X9?pE9y3A_V>jlg9NUN*ChumBc#6Iib**lI zZf@?5c}+Z<@>UOBTEH+Xb@NK{ac;DoQ-)Gv0T-FO+lj4tT0EV5DSF+~b6LQ1-o59nT;Y~D7bIhY%e52qQ zDf-lkaxOtiXAb3MdiGx9Tis_J*W$a2fmwm0vXh39{MUa5Cf%#FXRLXx{o6eAwIeN zw&S&%Qq=?$j+0BQGfHwu+;6R~nFWsy)d-+|XFm6M-5%qkS*RtmKHJo+lsS0XL@rGD z$*6z)|KaN`qpIw_HeLZq>6UKkMvyM)?(W(&QqqEoQUcQ5o9>j5?vUF;4l3DnBzAEu5Md!p_K|Gs&RC90Ms8em~BS((HD z=X3%*K6^aLuFii|0%x}cC#_I_lKjbP%YK8T$V3lY>Q@%y;q21bJ&+c8-ifY*}$hv74&Z!$6qP$hb*6$y96i#qFQv$ zijRx-L&2G?tRM<&=|G_tVBPMx@jN{#1_7j2imh!l(K0#usEN~x%B^uz(WVgv@tZLm zpvP%l#Om%Zw9Fh~%!KG0OW+AxNrWsnbwL$QaPx2_3l**}9)9HvoX2MX0;1wG>%cUz zS>nug5JNNdrA2gk{*F`uPS8i-$oZz+M`P}G5=x{PhomUwg3dAfSR-w0(%v+#p-E?# z3_pQ@PT1noma(s&^2sH3^g>6lbfzK?uOx61WrtC4ngP4avj<0_zv7LJbsel-&KFs9i3K)(W1Ld9}AGd+|k4xYYN5=A)p6L!(R{cSw1voBx69yHQM08Ot zVdFkllPMk;v&^Z!1r_ah5zK%iSsctBXHOssJPsy;LNM z>bacDmq0H1&ev3cKR^3b)xd$pq;UBnLl8m);aIfKvdk$K=C2fD@l#tS)c_HxfnVs; zMqUfnm!Q3>G)_g3X~oh$B9*w@6}P9@_3~s#9Z9lqC><@0)a!(MCYTHD5;fq>R_~1! z4{_DV|>nT+{J{1E`?VUZ-q*-Hr zqg>0>VjK|JcG|uDDr(;K@V*xqG{AZQ*AZB^n88n`<2=e6IhWSoGRrD&J(y6+o7blEEe168Qm3^n=>5paON1Z_wqAQ})Ix`BEDs0Lb$&k^-p+TO1x zFDOqte&O+qMKX2$vbNdY2dg9RNV2tgBox^0s8cvN%aHy9a%jDifz*ZgXy&I_eq7L@ zTmh!+)augBPN~FYvqH{c7ToKaI~k^OmJ}HM{uCaAAAlXk8{sE(W6@j({-+&!&-FU6 z#Nv^#RAy_@#CidCsmE?+-gM!q0YRh=aRRLUJp=Um{Wh1VmX3X{CzbQ*dpJujH;j5^ z2iAfaU_eCP9nQ*8*?!v$MkEoe*}WN9jpudsAPl`?UsA+yjd>?wjhmIGRvRyJg7Aj> z66NM?;OkerPZ=pWI}Wc8$+*m_2N}}x{gjxtr>6B2cORuAn1EH8c-9e2I7Pq>vH<8P zU%*UR_rpYJV-iED4m%!J5!%Pch=5|I{H{JRP^0S;PTa;3Y5aIe&Q=b%)kgW$+JymB zZFi+HTi>!J{FywBj9+h3=KdDJnqfD6OdKD2CNHWA~V)K8Z9$(Os&(vnfxs>XyG(POkL z+qZ}Fe)Wf}Wb4kpLj8shnxD%o-%jiveq*pZJGfzn4NaL;=7>h zuJGPi22T27;1R{k4Rwsx_k+>>KVU7}9xkK*F?*!*W?;k~0tS3Lq?N~SXwb8nIIxF^ z9s9HGa>|4y;A}fAW)4|IKr5t8v2eT)45HE=gdS7YetQXnSiv$7oUciJko2Jyw&A>f zC>w{H`Vw)z!id%uEhYIB;0&oDV(NOn`4~aT2b_7FJYhx!z(Q^Yc*7jqO<(qM z1m;ArH_L%{0718YJ|Ir0-e>s&c=IJtsJxQph)riy7s z@r;B%uLRXgtnvnX~0w%{aH1I;NKU|GeAGeCh>GQ^OrySrrV35 zOIe%r!OueUOn%Pbpumeaz7NGng0?w<~Wn6KOJ{jSe%$ArO&zT&EoiucjR~4ofxqZofT=MKsGOuqQL2dFO ze6aq*Ky+-(5aQ7{h$A*?-y}S~1|#nRYl&zF-D*AVJ>}otpfC9SH$0RT;8iCJf%WWc z>653PA5Yq1$(!D)rrubOpKlflX}+{=K)n>uljQUlri_ytNl*GpIix01AHh!CZs9vcWjEq99uJpLXwQJ#ML)KXBdsu% z`JLYT#8YZO#HGM{^ucRymM!#6cUA8>2#(p$RPMRlT*~Q1ue5TEzM9X~C|`6?9d#nIZF5e=f3KEuoRnR?yaC~m*Pv>haW($EY~Z~{s?z%mc`#aQ;qHLfzZ9ji3w zJRUD9=jquC1h#UXKu6X$9ypR}LCW<7teB4A2xZPi_Z7r;#Fn|b4`nID4|$7C%dXzuC#iekKM^>uxt+VICoFQF^mYe<_d9 zGJRe#j7Jxr>!=1=_I2fR?duLc>Lu)AOWq(>>*vW7ke0ks3;7XrzTzRp1GBMA#_9c8 z1ap*!f|!PbeReoi3+6!(0as(j;b5TM=iIUng&8#eCLXf|BMzLmYUV!;@+$X5KE{hu zZNJ{&@a#f;vYq`JGdyPohDr|T3?HX|2|B1xV2nPP<2k0-wJXfQ3+5J53KOxU@O+YTsvMwCNwtN+&N_F8DbZE0UfnaKT0uW(OiuR9o5Egn2F%|Zl zh5~D9B!J^ohB(6x=){&o=#p?Bg7i~=hfSB{O@9iK+f#7bTV=m0Ib_!;}7b(FiRYsE)$8r+mW(H1zRQqsv+^^%OWrP}&+v z6rToyvoFb98_gG(M>htU0B5M4srQQKI!X8avc^BK*p26gXspsYdw3K(leYH-5QYw~ zV(BfJyKcI94WL*_3~khG$#G?-Gu6@4PV0xCEK_rwA!{ZE)&#(*1iZ)X2`K|e!HFUs zSh)W_MnX2FI09EoW=p_*xI7x>SWoQ&4%=Y{_4Ww|u`ueGttrlnfPon5iH`bKSN7JX z$N07U-lv1RL!LIw-8bC;s7IR>$@F}dA8_%k<8|*)npewI3&43`A?xig_ksgbc@`Jl ztv$VsooRUzPBZnWC{^N?r}N?l-YZrO+LrkYH#yJM;LB3xxXUW*|1Z_9+oDR1tNOb) zXtW<6l;q&CM&!B*8~n0p3$g<@8bggtSSZhrfy3tW%2l?8d72gbp|IcU{j!z282rm> z|M?H($wW=2xqL1zZ(KESD65u%c@m%>VS-a9F!sosTPRZ{WQ&2Kqxz1d(|z!HHxu%` zrmQt|UidG>jyhCzZ@xOP^|#U~73&K4ObXlEGT_0sGuyll#>osYNT~#S=+lbWrT8Y0 z!B{V2@<;)8ACh}zwJOV^?wd*?z@gWTycE_x+zj|kGr%vT0Dl}tVTcyHeVOj_R+QJQ z^g-4eUg-(Res%DO!%w4<|3D<<2GZ_YQ4 zNfSx73m<<%oJI^NpCBdVp zQdz$GGViURXDX#f($4NA;xljgXWS8^tv=W6V9{_1GF&uAwmPy3vcx?mc5er5)tWXFL^oW2|5~ef#ql&{{ z_^)}GoTLb{U1}T~c+5r;mfANom%a@V)x)ZD{G?{t?=>#;>tyoB;PywhnbJ8{r{9HrKj@f+Lj$Oe0{PXxV*`hU<=*~hzBQ7E4X125Wsu$|3C7Be=>xb zj*$%3(G+63T?%kMmUw3ucOWP3*f3Gz51@W$MFV!dB-+IJdc$aWGa$+EDRMcIMj;bG zPh|55_SJD-9!Hb%vxkzVb6YErk&Pnp#eDhQlP>I2a|;Z^?O|LLp7OHKI_Rf;j!htl z7WyoOGGnKNlgYDc3ELt38&c%gw}z|^g}7Z-TQ7t{XyIzD|73Vma`{}4n@5HQ)__L; zQq%Up9SX}tU@kW$%Hs9Xylol5d7J1IUW;JZw=(_!~ zp?`3?;%&IQk+k$szp=pgL@ytcoNzbn$tGmJ>iGoKbCYW%P9;xF<{cA1`L)Z$x*h17 zFDj8bQK+jdbwXz`FqdILu&v&EZjtBx|2{v?b$fF)GlK?|gtP+VdF-^m*R$q5t#*>*QN7@7NxdD2zWTm;k31!{VSIK!mKOm;2R93kqCs zZo6G)+Wcw)XF$$+$&t3aUZ?4`%&q$fI*UidE^+yI-#>#hHUoJO!UYYB7jr>#itJw1 zXcnqWynJ&m!jtE!$8;%fG2QFhW$=vUds%-rgwTuek*0M?pJ*{u(CVh==09HP^9%1D zt#?YH4po6Qt`!!<7mxGu5o^0cLz$3gIGxi$>@@i#N8rs>x7l0O+gU@!CmD3ce&r_- zrgEV0ml5>YW2azpMpqHI>Y4v?K)XfY+d+o8vw?j$c@veVbQ5{QdNooyNB>8#n&0k+0?qjG46~m~13X$!rQuVm2n;fF!tL;4X~P;QtbPg19ypY+tH<@m&Gk z@gk456+)gxuMip;f&Hrrp zMzLC@Y>^Oc7C9l81KA_~6K3I&To|K(T=soYJzDrk&Zu-R!+SL>vt6T5fPxGek%?`2F$}N#I~mYD)WDqfGotkZ7dqER>O$6y@;4Q)SvitManhOt zl1Cw{jcU79;nN@xD7m$sqO$4tEa6n}+@&l31O-N0gSMJn|9geVIdHZC%%gv}?XV^2 zJ~-&}xqCmIa(vA1oPG;v7~8;xbP9NE1K+3}RkthPz@G=))b!V`R?j^Edk#$J98lJhzK?=WXLR@Y+y6ub0yp0s*x3*mzZJnsdAnIibGGEAK;9z{V;{jX ziN=6K$Mprch7^O0imbzTNmz{woG=2q1a){sW1(1aHps?UB9;A{Fn(ep##Z)xeQBKp zV;3|{0fsf;O#l>uzvIqiBrAn=NXW&2yQRblt9V60g4;RGh7;;D0eHJjDl;IX+CG2L zgI>xrCP3>f1VRIk?VNWqdD;z!g5Ti0mz?;?fr3mBAg((P8BLr`csMA)PaLNINk1GYSR917QFi-5k~FU(^~vZQO$YrG+wa6>EBZXCuyBz zNN=i6c?(@%3Hwd=mdei9WpIi%WdL2kMGmij6U-frpnhta_+J36#A6=I>rlt*uJk=> zj)9T)O@tlkt>BN8`G5Am~Zt9jkEa_2pMvUuw8tw8RF1qjyFC!wX&POs{Q4lr`p1I|)T z`bfjRq1EGA&LkQtu*}?jG*E{B@5+u}EohNn(K+aO8cFRoXHEwucohD-tmGqG^$QmX zxnn7VWXCS1YNy8Rt#bWeIcr;E%3W}nk7-6zXv>@JxrgmX)#9HzO)CALt3ze`$cZC@ ztgynLAb6@QeDv?q`gk!3$C5Uc;%x!|9W~wAk$7`dakhelOEQ(0>319%Csm5|2Fl4W z0Q26D0qCFc5@U-pn$2f>`rl&z9x1^Mx^aTT%K_p}?Yjbw%vhVqkA!2&-m@IUB}f<_ zMQ;U)k|bu9t?(c#p09@%24zYKIt`ltj(?elqi|B%5wo*2B&cteh@WZr-UWc&UYhIg zdAN^PYI8IOih;G07Za)g%Ab#j@`d(@W-Bil@hztrA!k3W$V)+ut_vZRp!?scWU8)n z&-vB^azF=>7-y(y+JZXjW1ozn;^lj<#!~poT`sNVw;X&t_-xpG7YKUAKK6bHd+hS} zc|aJHuEbNXZN`7uC^!i!lIvhV5`0bj(fXoQ({~Bk6wq}7t{7^|78_N0k~jMP_k}4p z1-v_Ku++ziDlTGwJYQB&CUGe^0@4|~278>&O!gANwncKUWRLZzNn{CI7{?QQBd;4c zK-fgaLj&<|1DV;NTWfY9-4$b~rXP5Ak#~S(DEpXww>=(@i5_M5=W4&(i^##mZZ}nlgL@?+G4Q5{!A$oe+|3E3KIQx7w#-%UF=eltLtmcM?Zx zKmrOF_(aLaS34Lp4{BYobh*xrH-Vfms!_S z*%3RnC>meNmkZ{@ZcvFOzH`IeYg{p@fBv6sFq5WD0DIK6p`5q;l;}raM`}+ao|P1Y zC%sseNhY^Bycc8;`vJoU&b;07#*YJ-+bsJ_ihGqDYZYrA&>MZ0)*}F4+x*mT{$w2zAjC)Nsous=0qNWzZc??_AzLPLRd^$$0OA2`q+h zYsX4hSEASDCkx$^7yMGb5~(*n&F;O_WY7ATZS`lG;B|Shp{w-_y_0`zJmiq3*dd5qOgS zDXiOktG>X?xF97;8J}2ROZQ8*hmk^^q+jIEP&8TeOA#(x|R6lKr~iMCGzb^ zN?qoiV!+A7$5ZzSUErv?yIJsY5C4A`6T3|0@kHT=y!f|KK;DdGu`3hPjgrxNt{E*JHB)UzAlzrN=&Vz}=Mu{|Yxq zg}xM^2&k+3-8b3lvm^6w`mS4^2|&6WCi5Hf!@m7GG1o{s&A%1H*wHKYw1xB~$uo8U zLUPn+F@MTDV$49GMmd$tJV>Jx{E7xbBbQd{C^Dwe)Py%QYmPj{rtQ41i~kqoDn}4f zSBtxRVaqpJdWeyy8sobp_h?hWRG1~I$(ZD{F&iK8Z6S?X{J=R-FU4a`Ms*A-8+SUE zww-;%d?pQKe)fj)(z3#2?Wb%z*{}0ObHaYV`EL)A1X~<58pU@$ZE|FSIv}x3y{syr zU_bx#L!8cv)b6-(>Qfn#TP(;?zl8ZV>*?zA-5{m;cH0*f245pMKI-e>Jz@CZ`3BL* zMWk3U>Hl-l5gUe+P8DmFQn$_<^G;bk{;{9OsXggH+;5rJMgPNJtN(LW#F)``_?Nq=Ew6L2noi zOFIf~ZRP5pPba_0za8iI47-d8K;Zh=juz~-Da)EmJUZYe?{2bA3dLX+gqQeu>Rr`Lyzx#Jj7LK*;6c z4?jW@vH*;C-TD$=O6lLSl!6(uOrrd62E@@v_-{2YU?M*5NSkQw-H46W`)7;bSFo&`8=0a)*5W z7p`mry(pbjrvQey3r%yLB+xa=dBol%4rDMDfQUzPkEj3%X^QBSErRsFPvgnUIpsjr z)Jt_-om@Lp8l|s5owxUP+6kf9lLfN;XtE_q&hm~n;6)PekKOExNDLn#z;H``ldTt! zMa5@Hj+z>tk-keF-&a4E2n!U}Om|mBR?Pne7p>SO0S4nM46UjD2X!*In^>>4r`zF= zp^Duu7=lY2aYe*zhQK-A>82&SaGwn<*4u$n6(7O6&ueS(Vi|3+#u^tBTlN!$$Yd0B z)$W9++-5YjAVJ4K9!NT}123nt`u2gDyKY3tANR5JXTBgc#T$725>}aM2FgHK?(-Q* zm5zyzH4Ng=44}(=KrwNG^7qXYbqKj!Kbj=EQ?$tak`Z$-5FFWbX);MT$3WpfpA8LS zZS(xV`}2H|mCZ&QFzf36BxocB4!Zq~F^Dbg-yzl(gjLGNttJwV#V&eaHqZG=$Sx__ z&u|U$oJ#FvMBr|9rL4y}0d2)1(Ds7G216z3k`nKvv0<`oaB}-9jxX#<8^K9y0d18w zs1i-aP=DjCO}t>r6|@5+bprKNDzFfuGbTLuW;}ovn zjXB^LZ+Hi!majvO@ruteUkb~|y-wujBU>xUs5{FEgBB)A&=!H1YVfmgL=eB?->-9uYl#Yns*>p!R0%*AQd%ms>w3c~pB~6J0 zKx@fBQk)#cXV&q+6O{U#_hw3yKA)DmKw!?~`SNk!4fOt0h1}mCbRzT{ERE3Ul#S}n1m=pIu-^mwMjEl-Z*oR{dg4+2-rty`#kT$j|4#c|i%Kjwx8b8H50SN21`2p45K(0M<^U94Xp|wrDL`We^A$^v^Qpzw#4Js!QIlZx`LBO$d`$ydC zkB;^P9*i^==#-*Hpk8j|sH@2*j66*wI7(>f80*Yu19_R$2I@3iXv*z@CM^@J7|rJ} z%lDHrVJRJ%1vn0yT#}32*}rlfs&BQ>hfO;5 zOpUUb^9d6oi)Ay|$)ZJdl|zz?`B&v4b^FLh+?{D2I{5)o$W9RW|t z;coUdk?1;oJTDq3CCPoab0(;|SUF2bEZ-w94Pz}bq!rr+mkH-Yy-a)=xGb|FKPz)R zws2^ec_<|Nf;53cpY*qotxt|UA?Un04Pn}C{c2g>dT3lf3$N@jqieb3Pw#I}bhe41 z1BajDqY0Q#%)E>;PR$HVkHlY1@_K84goXKqlIZt>#shr5$AMcHZSJHAl&614=Kym+ zK+PRg;P-$Je>u$7%_F*)&9Tlv6r_b_I)Q-XG3c$?1@TNVrXp$iET-@La<*4Js;_qU z2ib?7Dm)?0=WBNgmF=0tMYEW_qylPy%TDjp^~Kz1-Tub_T~MR*r0<^mUS$@UrDv55 zsI(~O5ul9&sbV>ZHJCTs0KlLQ*yd_04qD5=t&W$yek<9_g>K~929_)3dp#*gyX}Cz z#)sDoOdG%7#Is-R!6HWX(@s)*&2A7W4SQ%6FcRhiK^1n#B4@;Dy2@(5d+qVJ{a4k= zp4(67&0hCjEz4jievZ0O_6O3*yz`O8Z%(R9RV(0Ee=NOU2&W9Z$bs>f1(Fl$qn3)y z2Q%?*KM-@#jBUykfoamRRK%_3A!*OG;Z&b`9cu7}f~|hP0CDB%)b+uVy?O){0=q@V zq?-4SjpMTF$rw|4Q_(o${KIL(Jt2ub0eSnm*Wrnvh?zRbQ6T?1Zhg^1HzH70J|KF$ zi*WxGq>t@~;s@2CmS-GJ>t_&wqjtBE8k!efml6?1mQj4WKmH6>^oy^G-ThI$-+KS6 zx$i^IswV9J0L+UPG7z{~|4*YBNq+UoBw%0LsbPn~pwN4s^H#ffFLbb3(}{%r@)1vA zW`K@_H`SuNMdYG25u~)74nF_wTz;l+__kwMr~P7`wDym!m{vf z67xC>vjlw&;E^9R@+Fed#$X2NH4?@ojGqF_ocfw}u~fF=3SQJ5KKO|%!lUfN6?A)D zv^?=t=x0^;cQ{l{QL~eP%@o_ZMf=j;U`epr@RnM?0Opb;WDYKxe&=zON+cns!h#9FsgB z9H^>VezrAEKP!r)P^d@+5~`9+jp_Hr1Bb7OHi6t;R|J&C3;x2mxT(Y`z<|Yi2FLRm*gDf$K5elwDV`*(uu_l z`p>>m0w`_t4Rx5*1q9It#Uy*Ak@L7zGXIinS_l`n$GCt zomSV&r_mTSNsf5vw|lG%SrU_5S|;XAX^Y!8dj|9FhHdF3`B65dTOsE&&a3lU-VPk6 z%QCyGuC>QdRaHrN?)7jgi8mK}A7z+(vfp;vZ;$`*({4Qq2fQbSJ>!(c0-*ij#Kf76 zs&{^M=;vq&#WTl&0{lxMUmP+@*h&9*T=HiHxem){k*A^6C5|hli+xCJufRueKKQm`r{zmFPkf?B5Dt5gO~>is~Jf4~U$$Ja6TO{LwQTGYQ-Nz-Zbi`rfa z?X>kbyLx?rXWJ&mxR%;Uw;E|G5a?y0x7uS=WVuD&qBkHv%2V;R8%zI8HkIee9}@!2 zLa_X@M-j(5gNF;6YRuCwf7X6E0b<}?7~UKRicA5Sou#if=xo9E6p%|5RmE;Rj7Z_v z3D)ia5pWySi^BH411&l|h17SlJ5Zzk{&^vD>OS|Zxr&WMlBZo*UjU$CGDT1M^bm;< zbhHIHeHLD=>0;m1n@KePgNHKmy)<44^a;e;4nk3ev=0dPB^G*w3zsJWfn4xm>Cfo=$J?^8qxw;E(|d6tlOHM_MgKHe@38 z9>r;uLunEN7V;gFZ}~!AuV_qQzF5x6t0%6MJ1st!FEwfN{RUV>U7%WKE^7dL8eTtU zV0=fQIKzN3T!3&U462;AIH9$?mFiH@sm#Cf1|}$I=TP$h{i5eRPJ?6id+>8Xjdwxi zKpRtoGQf!=mt)iBgifrLy4i|>1X>^%)L@s}^9!Hofe#ad$&y*cRxW4Lvq|`tY^u5_ z$bg^cg}u5xGp7p$XMkXp8C+?D59!c&+2iq=0!1TjX+7fczH6UI?fgqdWJVhBbBf@1 zu(1z6nF!Ycg-^SCWL0+hR6(l0hrnU{{3j|eBzH9u_G>hO^o(Th9T9p?kaAmcI9qAC z;RSQE>AIG3&Z?}|y52g!ndysi?$Oqm%#t-s=z9?yaG6|a(h2W4nuc(5rF$J%7Gi?lNGc)(a&m_{a`ozy3b1e5 zMNWht;hm7p3=@`A>Y1qH$DSQoV#L7MDzk7IAQyVI(qepA>{AB&Rq1Ov`5Z`oT4*!Y z@B^6! zjg0{*7Zw}u+}4dgvPj-AmS%j^f!)gYJAol)??!zuxaSKDm+#?IQ^gs1XO|>MNEnLs zUI$LneeI_V-d_raKYof01}P4XSZwW-b~ht?-T}pM%aWT;y(U92`LaNb{ic%Hh60uF zGJTovIg&yB3vJRfdTw0er|3-4`NIeaFG|x-?yq4_UM)Ezuzw~NI5+w_ zsTkby3yC6#Okj&FTbJXPipD_IDs$8>(mS)hE@i z%qT@WBJ*-h^nXVCsMyxG8mp*uJ^Y5?$qS z%iyKZG67An_W-<+u#lld_5U?Izp78}i|7aF$GC}BJpK)6% zKslg?@#BD#UulYucTAcBb(6bqS!c3@+TyWzBEk>3(q*IU06!wC5}nrjXHRbV4j2y) zx*K&dmp#LtLVkZ>VNutn1h3sapA<{JY+MBQ}xY)iYMu>)*TUS=+W?y0jj2 z9|fd;Q6OL46(g6tp8}?E&`%^JkEFd3EPN5DC!8~Q1(=vRnmxr|)RaD6t0`dF!^t5l zfPJ{!3eynciw5`|7UQsvfrbh?v$!#FwW3DN{lOvlL=I#^)(5i{p4PpJ3awF-RE*z= z?jK7}kyNa#<;PgKpUXS;lRWw}dGA;#cL7FhnwcO z844R*kfF*5pKnPnPIc%0M7ggkbm#oN#fzkm7F1#@Z0Irh>UK+n0vBdxCw5=O@et1U z_lEpnm-$}K7}a%#JvFt@XV1+UvRFE1f5>KtbbUq|<1E?RC8?W}Bai3Jgz@g*N@SW` zH)NI3^vNb=YZqFJ{+8D+y}Km9&#VU}+kH!Hvq?e8gKzISY$?pgeE1Lnr~K7?V1<>R zp1X_G{qZM=&a@$9&%EQ^e52?U{Wr z?}KMWYzl0Pcr4ItzpZ9APIX)t(wdRq`QJ>Ib)$|++D^=6oie~cq*;7Gt=vYQPdzw{ zz|FJbsZs>`$~p*iA8ch|E&H;#-raO$c*E-&Bc}T>fZS{c*r0-7Zwsf`rUBs_x7m_z zM@2ETGxK5m^!^+57H@Ab0eh~} zCGzsrF>R0K^uOZj>O-kB9Y}umBCcZAM6lq>`$*{SJ|lo5>O+LUL7Wi$FWy{7bt)(+34tI{lGNvak9V!eHpjjTPu#uFUXle z6c?MMx{~*Cf;4j%%lWxjGW8-_K06=A3%qEqU;=EL@7!Z-t&LaHtYdD&n2y6qElt$C z_}fcF!~$xlrJ*YS_6-CEY~~?^7hb8>EiGf}I$UkbJ?)d(t|rWh=U=Uco;~YW{=`nj zhV^00;=BC>jmz8narx6U4DyC3wT~Z9Bgn#{y48xxha#qCJi#$~7rqCJ`?sDdY;IS2 zZEQA&Kl)ilNP&66t6HE+nXOFff()sOufqi|D7LxZoL`?bINfQ}7Y7^6+dA$fW{i1k zf>eQ*r}jxr3d)FHKV6Rv6OezkScN?@J6`y$342b`XKWDms$Czm4NdlE+9X@$3YqIe zaGS1mGw`){tVCIep*})0q94k3r%|_ON9oIuu#grTK)mxB%JTp=GBwWiyh5xQiRr4z zsD|NdgtK`CMpSDkZlKB)Bgk<&8_IeKeOEj?z3PVdd!cwe3}T(-OpoajH5J08QDyDq1n_{yVNu?2@W1zB88{ys*=V`0ti` zS!DvzVyr{*m!+wwQ0x|m6OiMPo-hq2oDGlgz7y;K)NY-JTc0hu6#w38UyniWGx1=E zz}AzrVOF#KQm8~=5+ue2{^J%xOOU%Y!ZUb>a}Jv<-bRk3T*q@_&-WV-TD(r3)su!3PXxgbpsTa$^&)cRx9)A(DSm4$~Jb=l8z zW*3KLu$1_?Q&H!<$g8`*ZxW2OI>CqAKD6VzC{)~-0}H5SH;s_wy9ZWfZti;3yU*?L z$1U*3Z|t*gn80-XC1Fo!iNH`A_y;jrgmk`PbcrfTBOy-b&!0a7qob$J_nVcLn(IUC z6+}Bfn(;AKx7S8y^cZMyIuS$&rMfMvAC&|!h#!!7#DrxS`UoZ9?Jar#a{j)3reJXY zxABT`DjgmDuP|-lB^Xwj**$2>yl#JNR-mEtMx96ncdEjVzVq`g@k!51q%BKiLuhdE zP9o)pUszxdjp-{Kk}n0hH#k7_;4$&pAQ(x1h9^(ff4?JYIGYTHj}ub~!=Xg*mxi=D zO{!I$(`{8-6k{PlMSPoLb;!6tXU&0^kL%G!oGQ}g`MD0Sc>B#6sRo~U5^OSR6?S0H z>Bm^Zx0viYZxKB}^kEm8KER)1Y)5-fiSqC*xGa6d&8)R5IK7%4V{V(T*V_AwDTm{4 z-tT753TQur@c98v>fQA*tO4>oZ`8Ds&!g496xXzmJSW$BXgZFu8vz3YKO)s{!;3 z+)Qj{rpnVt9`%~u!a+UVD`hEZa-TdMlJYI|#l5zLdwcr(DP1hK?eHjn(V7lAy+%eL zGSR!G(+%ip+6AQE?yk3wsy+8~@(T)>D(+W;A(kP;Bh683Ol*6Sjfs`JW%U?pEiKtv z53fcC)ZquI43)}y8(S>`<`=F>O(i3(jgy>O&mc3O1N`comIzCA%4<4jK;b6>KxPT< z^{Jq2V74`uv6^{5oGHdEdtWj6J&H;CS-y;HBOHFDL>GsIpJr#{lKfy@^zCMKfV*{= z>DM2FuaV%MWc-j(;PQE{Bz?MzXynV}dbKb94%7K2Q=n^igeO?&*F#KE;`d>{pYn>A zS#&n-M)Hhxtaz_Qfzf*~P4mVh5>ctN{b8 zSnw^Q0cQ-|!kb6^$|vrWVEtmBpbrc#5J_$dk8l@ZOnJi`<*t#NpE7YK;gb9QrlHKwrZPE zl2(u(PZn^a0R>9=R&XU3ASqBCc;owXvHsDgPkzt1?vI!lhQVPPlZnmr0R^b3PZ7W_ z+$Z9CQ?b)gQur!C6(LRRg#^Chqr|@c`Ahn$%V%c2cn7}biH+*J3MwBA70!)WN}r10 z88Fm!JllHExPr9oir{)W)8RP>*SP#0S64}|x6jwxuA_n{=mhlAS_px?Y97b1!LhS2 zD=^$PD}eBe@G&|dX=khr*SIV-VEFhbjKN;kK&BFcuq{!HFeOMsu#eW<2vJ!+!X`ni zBiq|-*cFAdb0@=I7;#)eEAzvcKgl6--=~);4T;A$D@H~5ME9gc#VOzlUc|5Y#Wxyl z1p~zR@Tc8fU7Qv@C@`z83y>E%1X}F*3Eg*;flVu-p1WFrAavIOTJ9X;jM9-EZm<6g zY|e|pUKqsa$O|FGVOg#Sgd0;7;^P!jF06QV1Y4;$N@=mQ)lFiri+ayfA2_BpMytob zmHCqB-Ezjb50rIPHJ>Y_aw%`y1tt5(x);1j{_N=JY&7fZWBrsK5aOG*+3 z3VPoF+_%<^KoRjsaGiO~fzNsf-*D%9y8Bos&=mxO;>T#~XGMHn*djZSwQ&Hiu?Y_u zeznrxj7G#dK|zMXONxo^i*6@^`JM;kKq9)OhrsV6fek#%$bN!q@_cx~aw&i2V(a7N zrx?UYPU#+-*5oz%M3d=O@Ocm0;h3MQrrFmjuYZsnQHFj)#v^R|pXckiw zi$C3p!H@ zVJ1tx=xG54&{W0q`vA?X;h7S;l0H(t@I3zqJPsb4%`TrxlDLiz@ne2uYGGkA(0o~i z%AAPe4}_(zYXRrq0R2+j!-JQQ!^BbOkx9_aX%i?Z9fHPVJYXQC1`i7t@~elHJ?Q5( zlbY!8x2{e{|18~lWd9-;pIU`}Bi|-jQzuty^<*2rQJ5meKtLM_m^UBqv(t_!oUh<0 z*XPki8c(XSXPv>%zb;{e)EADAt9k4%&YYpNo=hsbO*|)B_*1#q;1p3-#%efZar9#l zwB!>Vf;{UZVBcNC=F@qalCmlFcj{pw(9ddpS^NP{_n{`-&WR=^0H579d)l9F+zEdC z>|wS%d`vB=-d(f)D*V&HulBWtm0ZyT7au5wv0bz_9}Ol%h9e(? zU(7=D5S;5#de?&arht#sz9Hk+*^Vpp(-*s1)%?cl?~oi)M1L{AsFU ztBBPLT}e)^%8e*a7tp~GV3qi^rjp_&``I#muY>QXcnn|5YR76&`Mh*si@Os|#1~!I zhHw8EPbZyU!1^;wAq6f*&lz&FF%r7Z5UXnQr?RuqD@TM+S~S=uZTj(t3WM6d1Qc{E zWP`@1TrN+%jfXhaq%VJ7pZDOYegzGDjqZ9}6_=n#-03H*)pwq50!R+RK^bsujMN|0 z^JtivhdzA%K2QpJCw-%JL|(jjaR_$(1sr34XMar4Jc9!BiVzC=J(R$*e{A-%Ki2Gi zlqnKhR(tqRoAVK# zFZ0=<^fY?xVB3!Fz4*QE6x|xZ@tL?V!z;DCb8G`tNn~nmMjgMjLRfv?zD7{*>m0xV!J86vOs&UDJU2jP~%heSitq zT|$V#Llgs4=L@g`TgK@r>3IHf7UiV~oQR?jSwu2MubR(`L9Mzm{~~frdff<&RomxH z0zp4F3Wfin1Zg@!;_r&JdIus<(q{u@6uvW`-)sL5O26RFrBk}QK^lfm=`N}F9-i;}@BBFDz@EMLtb47i*4n8r5Gd@o%jqV%tyv}!m!C5{lRO^-^fFNTI zO)4-`k2-W6-({jp{BcP50>4S(R$Y>SV1f9(Co_QnM(8#+@AjTuk>_63MITeE#6#cq z#~FK^wgRYA@5qBoSle%#KyYweU71eLDw-)nm2?!-c>B=aup#$$;>$~wY0}83!`-WH zRg)k|#U8nUL%*Juj;?)5f2-nWN(vdpt>c6S-IwJ-7yep@nlgE;*&H#Z@* zOvh4OXzwe$y1=Nz*>roi7V2Cqbxl46(s*+|Hq3mc`QdcK#So)=p8eJ`3&wz#h{WEA zNu=?r879+Nx zi`7^s>{Wm8#@N&3v%T5zC?k`tR^W@Hs9E%4#RFN=6kwqimxb0KLsm3Ed|EcyEF|w zS##S8doSMvU?LSfX?yGg0m*2!n^&1k19|s41P`7nvCkkcoj~fn=ET8z!AJ?5-^C-% zoEbhT<}di|=7N+mg=bNwgtjt1fn>2csi!ob+M@80#gf?dTKvR-xIkFMn5(Ap3~jpd zKxDV{l>6hykB#0?N*VuSA;*)oo5L2GuCA_XlxLp=hqs7JOz!kU%aK)p;f^`g!((`}QPr{2A|%b4G>1m%ln z(Ep44M<*`6s0)*PlHSW;gB{b1#4Ia-s4kE*tKzWcmyit?-`gY)|2#o=H}IWh*HT`~1E0@VnriI*lDcQw8(zl$Gjl->$PysU-!x_e%q zjm>!FIy(9#7Y>)902x<~Tbzb&js0(i4s<-q_cH$`c!y^@`*9&mN>!)l8cd9NaQ8bz z0wh(v0;(N5qJl=U!!pq&P&CX3|v(4n@*#s7`;swr>Aycp1nNFy&az(g6mJ_aZAzHwO*; z(zD4dOvQlo^k0FeuH^sB3SWzspa$7+H4KQXE@Rme5!%|?{MVh!d7_w{OXFLln>&+* z`z<37F+lxJKRhok0VzKaDSuOW%$LSpL=@s|M{8fC78?nVa}vGb+bQgQgTL#1YCe-PxR1+!Gftz=Two<#uXPzW1iEBk>hC6(3(&axGD$bL$XYr3HG%INWVYo{xmmu%w6f-m--^s84 z3Zan~wlWb%{8Jx4Csdih686Umn5j0QTbK;uHN>nM%ix`=aKayGJ$29C-vViubJM{; zW*i#!`C(sU=>0A}%Q0)fmRDJC4!lb!gS&Q~Zf|e%b9eL8#%n=w3@Lb<-4xV0v@1}Y zoSgi3a$l@=jA%PMzf5A+wP2kpQ~WiOsIq-%=ofwDLoRc7a-$E9xX2}XQJId|F?OA* zGH6Y1(#%M{cK9hlOgZ^0ISKk&BZ0U7t*VUXa)t{o5YzZwznA8&EIVNP zzZMJ)fgY9A?-ZobSamfVJVi+SUBb)g=!5Nzr?JIlnze5187>yDfn)FSElTXZ+IIFM zXY*BimO6W8%CepwKuJn{#rKFkKq$PoE=v>`C(n);im%ej7DeJU^b98v)C*`R%;627 z7(bhxcIKgcKJ{Lhh}|4-I^|X$^RYy8E-`ZYf#|yzK*z`WuqCfk?t-f<$d7@Of~EKb zGS*oo$|_*pId>vdc8;CXE2^=D9Wd47L2=RAIoZ!N*piN4MTY#LZd7V&QlLLmj@56e zT(PH4XJh0g&$WI>{8NO}5=_j+&pNGRA8bP^U9bnnCwVm-jedls$?9;YcBEJv+Nm9M z2!=Qje`@vZGMWrKd;rlbgTwk=6{>qzJwhc{ho@wsAlroBvp3cR%v9g)NYzQ=KW4zb z>|-;wn?0mdr3t$h%nep3XS~iJHkq-~WJ)(zzO^$B0lR)SuVd%_`NsYWmSfHOZmgT* z|8T#klTY=gWZi2A$)#>=oYH-tghABb|uBM zp#+UX%ff(9Y1QS_eJZA})8)c}+6uIzn^1#Nz4I-0<9sqpbd`y6Lsb>8$SemFip5$= z*te#;U&lSsOpNIimq}G7`VJR#CXCS!{2|}UWmK(15!-~SW;&vqP%lY6 zUdpa74F?C(Js5B(cpdfJfuW@nE;<1cq0yv)Wx4`KMnGiN}^GPxm$6c00Bu>B8bk5z@*uIcV9oc-WTtm z%?Xm6wFAGsBonCMc_!lknq6i4Sqay(xRaURHp!92f`-)b!*oBch0kAFbzMXWpTE`i zi2B?z0;Jo8V^3JsUUhj&Uhtp35l59Z+}}Kie2?oG1ohm)E(6_w z`tEn1x3Xyh99s7Gitj2RN(j^*2zhqfn_Y3Azt9%}8ENC3oQ~6~UXqFQbF}Yzm5L z&Oe|An) z^Uz`1If5Te2npYrJ@EI4xYF-2?L2&aaW!?B;lILdB7=QTeazFTL?SGGA7V$sVas`R zQv#j~!j0i`wLiIz^mN!$-0cW|3j|{X;aND$aN1&}ut3F;6+B;%J^}iRrD* z$*<)3PMoOF)}s^DUJfhqMGhTQ@h&In251@_YmmPVvG2)rA3llG z@TZns{qnS=nLJNypM_7hLnAg9nz)3AL&vC?7W4+3xIrqOTnO4tUVMI6E9fxRmb_2C zjC5p;C+qkRc3C)c5am!8sVDJzIR9PtxUXh>AJA%M0u4V$<3S;8I43E(CwRK<$gkN| z=ho&A_sL;!PjF92_Etr;P5w<6>EKaC(Z0@(ANV&9!AnlKt2Mn2_bu=Pk3pc#8Ry%Dqo%@YgOSoKE$hJ}{k5tT*IiZ#Gq*s_w1| zN;4&(rv5lFp#n}J6O)fB5Ns}$O`FaoJHM?2-=#F2tGfK{!`iWI#ibTaNIxW+K$-bv zwQ(wE5@|86k3FU8P{ppH?|I_BV6Rp>`SnaBY70HJMV{MkFs0J0mVwiEmr&)vTG#yBt-41P%8hK_=?V?^W~iLx%<>jcS;DO ztxIKQ>zEgxr)>(E>5njgkkv^8u0DZiE%(kC7WyE*1ypup^81uce32X6LPUE&Js)lN z+J(btShvfx6)%!fzsrqc3u3{5cfY8PHt_EFd{VYB{EsZ_3;am%)%bt^Q1C`Ui(B=# zO5RCR5yih_PV&n+*9+5%jM5cWb_qZFWwX*_qpZu}oRrGuyv^V_&$HGn$pulf+ z-kj7~A4hjx8XQy6QsYyo{4Swe>qH1*@l6y(TdZv zr9(CCn~5(Bv^1}TCM zjP@uhJ*3zaviEbduQ0H&UzRG)Ugr1x00&a(^jY~cLs%~36^p@+6uD zSpI2v{}I5bGqqtGgd@VcY#E-LsnzqsYcG=cDJA?GBd_6bYAzPm%fQa3l2eSdtyK~o zs{th611#u0;s#;)Zr}`gx2s#tC9qt~J9Dy)X8*7rr}R>=74d?cJB2agQCJC<}AYjipT795(tu>T=nU zPvQb{m~-TqAbOJzszjn3&8Fis{>d{zh8B3o)poWw_0F_KD4N5LInoms-1&X}Bgtei znu8uQ7E^LJj3V5KjWj-+&)@)_75Y>PP=a}nwMe(Br=^Bq(T!Ua`$5Ng0~}9Of2ERQ z_1+!*+g$1oh`dHJ%&GAOHxHR)dF5x)Z`rwM;Pzz~Uz&4ssbAZogYp z(986obDYxV5Eoo0Snj3>Om}eWD_V=zY=3)E4Cv&U87*C;k1l2)Rvb8Xe=pT5X`ZA zcUqR2KiI@?_IPv$uD^y*w+A31H_E_uaj{qfvuI|;=;0@UIV#Cj;XCh39=MZ#O*km1 z7aI(n%oWuyOKYDW9>!^^iS<)xg5c_LjTj@LwgR`tN4259ipf39)P601cKq)Fawk6j zbfJ%M%b3=>?T)(9y^laeIB~J=Z)Z7{s9$rC%Pl`PYq#Y<$!O}Gb;gZ%8}33(^J^Fk zNcoCNX?WP|2X z2~Fm|`?)0l4u?{ZfE2*&N@C~g-ZV;%J|O!OCif8^bQkGC@N0~HYvh<%jwfqqUP6jU zXqQlC9l&q@p@~(9EYy<-ufS@taZeJ;vIVZ7n(Fw%956p{SoqUWy1;i3U-!4I@_)J` ze_~Q3Fxpea2Nn%U#-v&d23=1# zl{!Yhz8*@yXT=S@HR=;YI2E!x{NF z#GQZVt$B(Za+6IWqlL*{3K{}Q9RBy$TgP{oOE9It-KpXZ0MHKEv@9-KSiZ7`1mm%{77#5)BQ^R&-yCwfMg3bejYW3qSbigY6o1+(`XJtYe~-)0OXIko1z+F(7^ zRA1L8eDM>&OI{ay@Dp)1X;S{}+X}fH&@>PLJNt`+<{Nup_9Vm?Jqjpx^#k0^+;bus zZKtzw;3N0~1Qs6A>ZiPfW#I41;;zcFCLu>~#mdX_63midI2_m|{EMP5+#E!#wl(fT z*2Ej--Tspo8K*HB$uP3yxkZ%$K2$9z4t9FyAh~`{UL#2mAUj83AHiHyV8>_N%2U~L zp0AG|>U9@t$>6_Xn!t$P7H`6=2uhGo;^!}f6Oz$yPriIXCc}5EdV>8ExX)N29BPn) zAED^-s(PS63T`C4vlTl>#U`2Y%q5d;0y~i1%!p;goTy(Yu?$L00?mabr;ru=1BCJ- z1U{4eAdqv-Tp979kJXBcEZ2TOpxUk}Qmq(HKI8yg(QUW@*hyx)HHw}e zgHTKpuos|u-D|o;-~W*LSI{HpAl6l}^RzHP|DEC|ZxKxdWA*@x4t=Ip;VPTt^bYGp zPp`f{4r4ys6bfTvno?(fHJ1Vg)6Tdj8ITc@rqkATG1IE+ACePI1Hx^P{F7JExBy9NVqvJU?1i!5;_rMO|{AExGV`50CWIG=QHpJ0W@aS=N%xh zUy5HXBP*owVuD0eLHM4JLu&Y6pi`qM6WI78gZwiKtE^oBqQQc##Um&&3AeA^0=H(l z(**DXnrxM=GB~)1)l^D7o@*1YGdfsou*+PqVyY=Ia8TRRyr~dm8t?*Li{v0V`XbX# z#qy+@L@}AuQr0@Z8Q}mN7$N@7*YQY4Q`^PA%c!xQuXTc3VC+U(HHvj+RW;$)GKTPk z6lpeHf5;WhV9coe7M)4oqIK>oG6ChDU5s^4orkux*z@NoR3dIwaz)Qxf1Z(F-1M@a z&S--F&*v(y&|>OQX3oEB5^)iPzQ=q{iPU*^oStQzQ3+JI%PFP-vGjqPLr@|WYnezT z=B{;hJ-rM}`}8z{s>MI>-++q+(WncP?5Cr(!i#BA3@b#-tDRA7)>B_E=9($em&VvW#7ksCoS$L37+E8Iv?Qlh zciXt)EhcNe;^c*!yrN*=u}qT~nnoLR1%YsZ`oEg-jd z&|T;qLbVt|vKU4Iaz-#rPJ1fmmZ-~*vGg^50Wlpi@ZQ7&g!eguq>9x*bn3D`Rt|W2 zW9k5qzXb#G3FiGj6`H`wc)*@wN!1%UZh}|@z~BG3)<~7x3~QWaC<6NBY4?3+F)r5d zu>1-2X&aXlGa!8l1OjLA8=R|cCZjcJTKexhnB|}0+gN%i92x`RvNMU>BV=H2s-89n zEgmJ`7ma(ZCfqQ1Sf+#7%-9#|ojyfx$PJY3&`i`BD|UcHiJa`6!Pz!f4mN4bi{3lnO$ zo@80>A`MJ7bc?R@uAzV?+MBooxyN>YjOg|Aqx&2C1GdOoBzW#6T)nf&=eeX&=`jZO z7=oXqV%OY({FZJ2q353dlDRF$>&k}`@Vv#j)BF9R1ZggZSglG$A{`?U(v6=g|Y^C&9o8ovTHzy{FH zKl1YE#KfrJNigj4u`w~rcJOayEZR@5XbspjHA<^L$4se=Zx z=6f;x3FQW^_7)UHTFj!X|4fxOdO`r3=KRQ+FC67D*EpMaX1d*6ZKv-r&_7YR6*N4M zjJ!LQ*fLKepK!EfK`^u%sZFr^@{#zy>+rC;co7N{Ds*}r-KA6OwZfZN{LC3DEMoizjLm@*3u>q`d-8mFJVkDh@$mo>u3Kk8jy&) z81h*ZnLSFy&SmAeW9q17wGncQ5l$37C1u{M;vuY}r_JQ!5~#4R$*@FzIuVKs&zuCF zFiciAwcUAbOx9yg=zT6wrx63b6)sg)jIsHq_rHspk&kIYUJFSvudQ2zV5ofYjtz8-M_z!cSz zfAZ7(?C+o2%`}_Q5`7Vc)R(9687R(04Enb=C5k%;N{lt@adc56Jv_oR>M3dC&$CtxWZ{YZhE9&^*4kRqaL zmq`6J=goE-4s2{Rm`Ct00-S6|lCUUR*o~ zJ+Nt|PE+}xw6#f&WVOaTlYF=@;iZ+)gKDrs(rfJZk=DYED!xEo^`zHmnO2+i5rM!V zirTvi7+^%)*Fd+~-5CZM(1T|Whys!7)%MTqC-2~*PxX+qM4-qG{DF_4+j60i^o;YV z)?MT5rI1liw4@&vsE6E&XwFK1O4UOyWvpl^*JGvp#92$IQvq#m{=?_Hg z@PG-U+pFX1pf_r2aa4A$8;P2$x$#QvaPk?9+W7DRZy0c0587aPt(ho@9@u&GZmpKS z4Oa+y_xE$F?5A!YbZ=ceicM(|C~CY?QEYb1axwEye!KZl&Ljrg!LhNivJcy8xA5W# z#+qnQ%zF&Pfe7fL;;h}(ebJMt`R+pZa1nmr)EIz_6*_`6K!ZHhg86(1Io!E1l9~~? z2L4{aj-u*D)CAhYiC<_bWm4Vks2F6)YYnBr4H@c=2raUi1*!mL3X6Rx_W}=!m7XFx9{KW%)cxz8+mJc ztq60xdc`vN0{#)qyYig$Qp`ylJjS9913U!~UNDlJ7;m>m9wKAkW0@5T+@wjn=2(6C zO}@KXj#cts-$!c|EsVho%)u<&B5h;^{;!k}O}g_piLb{No4jF!(4QI3Sn#wlNQyxy z*pt)CL8%Cj>7P9}XKx$<4{%Q`?(w*#Qew_{NhAA`5Z|VXbo2cL6j{Y=)%K!Vul|Wv z^Su66{Z6OqyS%>W@6n6ybm2cRh5Si&cd|XcNox}h+&d|;F15p=#V)i`u?LU=Vo_N= z_E?A2RdcK|_$jc?>idRP(vxjWp5bZ!V=BAmq4nh}+cP%WAK3iol{{n3gZI~yYP5=! zK)ovt6`GK=Nu*)BnP7u1QWz}8`n^QDXu*Eb z#7^UX%MFi~n>2s}B*!8QlCK!yNBrFTZ<%bB{U~PEHGCM`+Zh6wSI08TAv6KO*!q~U zJDo(IfK@_Z2?w|iMk_WRqE=UzbAS(8JXs&@!|rCBRwR`OQj)>Uvm=nrwz%sIcwN)0 zqZO70j>o6PxM+Nmf0HN~!V@qvELm>F>~N2vb>m<(9%A2IsDG;1!G{8h{aM}g@w4+F>G5DzK;O?T~%CEe{3;FZ)Q%$_An97gbL{4Utv3!zfahX@`$e3g9iu1Ej$y;_H2NJD+2r&%< zyx)fQf0eYhh6K8~N~~s5jam#M_}xuHRhJRCl6J*!9jAx!uut8d|7 z0lUg`M#edzLTBfTlwV5llk-;Bk36P%48=@DL6QWIcOO%#`hng!vHr=-nc?&}mua!C zX3`#e>*hns-C$f}D?>-F(mq$nhXvsjw|a`niGoDdk)Es+zCikt`VWZRkuENJku>;v zKIILK%SZu}?Xc?NI{y~&f#Kn9&=!|@9GkG|)e>w|(mlF+_@cL(gq#Z|RV9C(4xW2J z=u@>AB}kds5Ot_g&Ki*4^AC2)yQUMqRY1v8bUD@7U)-Oy&jl-vAE zZLEN;uC7k2(gGJmJqY-KdU^$V)^evcX@|lNLLCKLVp7t=pvO17#rQ>Q(NLao&A8qO z+Q;t!Gxw5G@-GUI67Vp8OWf`m{Y4YL~F_MW9?<#sHTb<-;MK)YLxZq@T^{=1_{budw*-(QA;e zE>d0f59Crsf}C+L?^ZV)vw1$#ZX8=+dAPpbbb%KVy2%i#gK!1WCuZcuMdEG86wTa} z&$t=ItmpfP=70jZ9IY3PR$*&-EXx zQ=skvKdp^ej2(xj1Uj2PC9~0Le@EvOOQm0zl%`1xeTlO|L+P)G^`r@`P2QR02yAa7 zXTKMW(Sp$lSIJ=M$5^+zrk)cjNR(HCqf4NXMwu!V$QJrU^#B*}fpkQ?9x8WAxW3E# zv8gV=MAvyfggVu{xmjisBA$d|Cfhkc6`w7_bcTTNa3&}9NMG2@38=0FeEE0 zt1Ez{aa)Z&XWQCygx~8`^pu!>)Cz%g^Y}V#1gW(5OnEh!gLI9)5t0yGJ_@`l;jtoo z#5hd^iX~+J#@)DgMaRv@$d!%Adj>||7{=EX$dIc9d*?N~CU&tNd)tZtwyvH(P+)lJ z@)QYO@&Ez%b`c-~Q3dDeOB9f}^4i$=85Y+3ts~p`B9K|S;J{`HHF+$DJ6W=d;eiLS z1p~@_k(=28lOxEw;(Rqc3K?coHKHhg-AM^1shN8PoJJ|fTbp}^T%j^re1TyFQ^kRQ zz6Y)_D)1=Eh&U9>`a%Kcv0F5fDME@bc1Dn=lo<||r>=n-*bVG8A%yhOGrmZ(k>e05 zs>oNL(7xJCmsE@2fGsrYbZ~oYCr#Pyn<=#M*0oS(NDy~hXF!phz>G7Q2Etx$XC}u7 zMBP0E+UGMv{=Yj>Y}qJk?TE@|Y~`~oFZeEP=CJ>sYUaR*OE_*1`S^0Hxu(lKD}~s; z%=)>DaiZh#hD3+FzE#rFA{z#~yg#!8U#DjhI~vc;Fis3cdNBtKwtap<6O3X!v3`ZZ~h#Yr0zCez@4% z+k4g2`B)Ps%8AA;*7P)lV0s2&eR;CuS*)x*ovz0b5!A>!(UqU7_ApqJF!h|XDV(}o z*`?Ec8Pf*1H^sxVc97CnPd5>;x zzK1WTd(ykBY`zX-}mYjYz4pW&fE)aVw5S`p>jcL{5MCt8vVL zK{6X=Z(nL+*@+?80DUi(v>I9`z%at21Nc@_l8XRF^PX=iGA8`Mg4~{w7Dj{S zY!}7HK0p9jPK7nU{a74@phL4AXRKzK^2?;yo3Vh1NN<*4k>p_#$Z8MC<0IcZjAM>6 zF*9R){ra_s9)agRO?#m5W;pZGRdJyk9zt=ZmgneZm*a**! z{`(6dYK-FMK9kCd`HU(`g%R^EETcXmkW^x{t(IIhH<0fh>s8usNkB%){i`xD#yFL>H9?=yH>ZF|u zjqoKCx&qPKCD69prb=p5JXS?#7FmJgeiNe`GumTK7hlX6<5g0|3bN_4RVj}z4^+lb z@=I)+gE%PL3#~Uc(@4tVm>I4lj{Cp=2V?zB+RAc~E84R)O2Dkc6W5~ple+>R8aOzH zr&KM=s>}yS9QK>N(HR*SjKN#rW~R~Lw%_>s@$>A4p9GvGa!9iFBE0RN)sKF)I1JqF z{s;}0X*nDy@8{6r?e7f?#vUlS{OZ$$v5T~|e5UYJ(=xMkQx zJ^&#=7*zm#k_gxqIL)8!qf19)!oxE|^!@(9u}i4T@BZ2nNUO;}BhvjS5{ADru{vD_ zLE04hW!0drpASe(J99#IRdZ8+?VuX{q2lN@T_zPEBfC`=sDZXWE0L0JC*x$Ex2G#Y z57>xC_LMu|&h0L{Li_}CXhv+l8Ah#5JNV_7D(uR#Ih<)TQ08)+sJ`X8ictvHFM+Kt zRm5y)_|y^NzW-+I;P<+M6z8=TJDn$93@@Zt4jF9kR4vfXj}0QEh;eIQ8tg>l zi(erCPLQ2ZDdRmv>xJmA?l=Z{_`a!S_-qenLwta+xO%4xfPWFj9iL}?;k~yR=a?n} ztOXgkaMz*0eD58w`+-z|0HY|~%iCoWfwTF2mO(wpby9F4pz&Q_Ag~GmAKc_=I7pi# zG3}0M0}JSm&&HM4!0%7S&EqfE{aM~BH3ZL2(|-V#rxg5EN*ppS1h|V69J*GQ|9jw_ zF9T+E^_Mj=@TIB&J)hRfJ_a+U`8X*4XCcsnKEMMq@7na_uk)X5Z-$qulPfqmH&Bc0 zXBJCsz$j@!Gi0d~@9aF^TR^;7b|So03JU5uVWDBY+kW_Lr6r7P2dK^XdR$Q7zI|&8 zU?~)EEV&M5OVDJBxOW3>QW)6wK3De{&S9A0M|cASJ|S50M32@bvM$qvydSef2PGcr z7DOzMJjIc8zA5PioNFOG7wBx!ol?_sLn7xKW6QwW4wj|w&^%nB1Mx452;vcd$iD#e zy)&tbHmcPNr~!{aH%u=g0%NSp@>sR00!Wags?X!z zZZL-slI;Xnb`EcYYj~f;2=eAgJ$PwT|^-s-d=Mo1DE38_p1&X zkzc}3++w@}!+rrCi55sE+;9d523~*oP^D;4{@mNbB5%R3i&%3RG#~!~`+cMLqqR93 z@Kbdnf%@ZmA!3Z>?P+yp&{I{(cqNYJd}P~7m&Jmi^d@5Bm;@9yL;Lo69EV_8WHoj< zRwcZ_x_pl2E0cLJ;nc3j$vc95X~4p)punADQWp;Pdk-X)*MdM$E}#IU@R;OB^MUu5 z1Dq`p-iJ-3{x`6tJqB;?3IY0>LOnS5%KQF}5TlXln03?c;&g7hcD;I$yRj6jK!3Pg z{!!9m|I47Rbl=bEzUU2|DQ}UdFwh!B+-h^&cQxD_Wbkv`9R$D@2GZ_#fSh1l^jO24 z2f5!VT*gT7^z-!GlP(U2rgM@RqLKSGKPfkH1msjZ-JMUb{g1u5Ab^X6a4X=D0o<6?16Gj+ z!^JaFns#d#1sem&YrqB7Ezj9>_HzSlH?6_Ftb%sm_ zh)#8a0fYl(&x=@=74?dYs#F&(0LYdz-XFk7|IW|CUhNPwe)Br}8`yy_K?&@ER&4I8QeePk+RD-<4R7K`>mfKWCK^+|4;BzyOe zc^9F|M(V)ZTcUV%95?Ohtn>YJ0rbysJLd&>Pk#pJTSoTw%6{0H=<8;QC?)IQ$~%D; zE&G81X{OPWCoVo7$_>{f20t(8(obsfPoR}FDuvy{en?&qlS_(5N?iG zrzw%Rwe4bGf%#_v)))aiks#+U*bh)1kEom_`cyXEoj8mQRev`jhuod-LX4 zOk)!hgTH;FORqAC?uv;aNH;RgZwKIGIzmg6BFo0hNqy9}3h<8KkfoQaEj=t;#q1j$ zWWg~3I-XT>`E^1*qX8h^0`_J-++ZP4j<-AbekqRqWBr){q2RJN#^DE|B-@#?HeFWm zwvAzI*rK~BA_JF5-)TT@D&s}p1wT3BD-KU+j~13=o~Je28IWw-jNp@_WO$fdLo|yD zK)s+$JS--#GAg*08yXp@k^9xyD0=#gC%eMgC@%a5TSaB%096b$1lS03fCsf&VVNke zQcS^&9bh$paO4a0LqD@jtEaqj->V~_iT0_HcQ|d525Q_LzZbh63BQ%(7ucKERys!h zWZk^ahjqE@^}2hH_*PBquZl2w8(1G_ceyKEG!|ubc^8ceCP=vm%bgAUiu|t2E7!=8 zKqulge18vd_+kFLL=eAs5i6Gs!R=vZ^KeKR?IeYe<;<PJ$|AU61&G5BS>z{1AJbXh2vC~4!inQ>){<89WPJ;9 zz0L98G+|+3aGS_6kS&d)O!{(}_b2E)%}hv0Fq83U0vHefpYDoaHp5=wBZVyB=raaL zI~kvK0=s1`Cy@43ejm(J&bB`sQPu4fYJkHb*?#|k9#WyK6dPG?Dut7cO$vZUyB`cV zUr-?XHcJSeZ~zCNLAd|Q)_X7|W8mY<*xD9J^ZtA0X0Xf+!kM=ZUF<^rS3=h&Ce&}d zdx)A>{l~_~?I?oxQKL{w{qkme{%GK=&Wy*n=r|oG4#oJIR0;tff!o|F_eo-Ni9C+q z(7TuWy9j1xC_wsR^89&1O10wJO{Q6hAT>=jWM;Tj?$+lX1On7<=cOM3SJP_!Zpmr2 zI(`zXWk!|6eDvvA1@AHl_COk*gsGJ6y8{-+6qo_{H&+|`D*SJ8S9amU6^==bUF=Jc zJ2K$53EJM$OeW=A-f9;HY7+?i7>!F=rJGUJjIAEDKu9*)H61DO?}s6NzIXFl+a-JHsaS#Ox}k^Y)%D6Hev7w{oXIf_*yMhH zLU#O|BRW?6^?h(p@1=L=TE|*0{6CEF@bSo)F<8FTpNx*RGQ3;tTqPFI!dy)$tBBS& z<5*ypdqHEE9Ug!a|-p$w57+sK&@Xy;F9J75=`v#4H4kuRAZJYv&ZV$X&;7 z3tVN1%iS$$o$G(?KYk#x4rOnx3Xe$% zk}-1G0L!Mpa84Kl#F^blGmv%=Gi82=vm@haW!^P@s>l5mxQq@vI)?$3W?y=P)S6ZW z!gk%Q?Z@t*)$$w-nq~ZJd7F&X6uc*!7`Zm%Acw6uENgmovJFEMF?r*=>HR`rF z$yB{Pj%-teCjismkK=r{h|RCG{FGSc6lh`wqqK_l4~6etd~~_W%yas8|JA#;yoMid zt6g90L;{&qtWRdj#2!GWFst=>k{-U_*g^ih7VJm0{VCiHH8;L(TSr-ovKr6(3XH>U zRntF`4R-cbfh`d#^j1gC6Fj?{Z@GyHGMS9U;irKhvd%TklWsuIs&$zep7v+gJiZ!& z1fm8rceAX%ne;!&O120lr+tR_cXpg1JEd%nB5Ke?sO>+9T(EbdELz?Ilq^MtEsI&V)(iE&*?Y=sAZ)ikuhx3vEkvL3fI;ncgVVnakgr;- zOgXdICt$Th%}6}wXbQLK-cqYUINN#8`cwJ4OviRi{uOm?1wZ&;R}0n=*^Y%<3};v= z*Sm;$Z-UzC9neMG0c6Yf>F1vFR!RxiaU&EUfY1B2btCYVK#wrzSTAauD z0)=p<+AOaLfc$@{Mj^+8fPWd5%h5|C7+1c0*Hv(KBu_JVAneF zWz+R$y4|woVx8wMGTe$I2mA&M#b$=!ex3Xp&xKA&8*M$IwQR8BN|7#SN1wqCrFc)? zD>h{J;5Fj>0mxMqmTL($7%6K`N4v}3-iRMX_-i0^mbC5Aph2h^PhpPn~8J^tHd*Al?mp}F7li+dNU7?Do5ka;X>)2_BR2@Zd>XnlVN>&at$pz3A}p>iX` z6IzdvV2{xB=(7XP#iD(LD-h2yyIcxXL5od~xi4JRZCE^dMUc5CvSj<4rPgCzheR3^ zjY3>wqD|bU0fhAZ$z^m|9G2N$OKM9*8gx+AueroBUo!x2T5a%;Z(8pYP-gvM^ba)` zgj-&8lRmj}Gq&lTAH0he4&qUbCJtQ%zY62Y z;0;C#D_0<;yu~^r0D#bYm3bl)=-!B&lR-3!oJNS~iyTIRh|{ay>hOuO@eN;DwFhC| z=+)v6q+jk%mnuo$!#lr;`enwFk0@KktGCeOPl1e^ZYm4OG)B;%A~xEdAVcw@sMfk) z9#~E8hbIh0QsfO-OvQr2fFUet1sH;Dywct_Ska*?VwQ5Mqwn0u;HImMJ%Hf+ciO3k z$}$ACBEiURm6A0hWETuAoe3}A#iXTKL!cSS0UL-yOT;yoX2yTWzXc`1FC@HCiq<@~ zZXPu#rI)$76a=)MtARnq0=LX45Iacd`lUQD7wa!s!v^j%JyyfqjDNO{VQ-DDZoK%J zVIrdgmHNob=pZO0mHoK?Yc;+Q~{yYo7JD!Cu9_;?{0a@oW2Vss7mU(Fbnk;xY{^` zSSMxUv*}|gWbS-C(5{^C{G;Ywpi`m;A8 zNV020W@8e&%js}M{8M?a!k`-2Np&!)sku}7c6sLboW6l-*d-3_{`nrFs2~QUv>qT_ zE^CL&-&ZQU=G1G)*g$ zzDSEAo9`H3l;H1Z8|*8&-7He$HKXJ9k5={?gv%xTaSODATIAYQtmR>mz}d5Ci{FU9(AS?I=j)zXzbfIz>Ctej=Av|ELGa!P6C+k96+vhwPrf$QOM%_39 zmuv%W?mH+T7`Do`FG_YXicakr2zlF2%>`&vJ2(iB!CA2eBs*mcm*8MGlRrK_0o3d7 zFAfV^+5YxOhLCG=sGJ7~fY%}ec@&ZKQG!U88BS=!LR*x&6!3Pp2D7-HS@%(R$PTcn zs;&|?T9^hQqgwa@R?yAmdyi|wy^${Jw2U6>KB5^S-{WI@5FcT`8Kbc#I(}P|Qugo4 z6fQihSwhQGa6-Il^lWu#K{WKkY&;`6nzLygOtt4tn#cgB@CrPs-v!Ay(v6eh;c!PTf0b-EmXyVv z>?(Plba;R7J*{Dh4x$(SogsnJT+c-*Abs9-X*-(;F17DCj6dy*a&_QOQ*y~U}8K-g&% zRY@a;w=Eux!JQbt_P;}zLtqazvfp)=ZXNojr29S+NZ$3ha~1d}kNkgZon=&1U;OSF zy1OI|5ka~|VrVHz0qMH?`1{{=*S+h$@GZj3 zKKp#*c|MQ7!!Uq7IGSus9e|vlnc*OaGzRrSl30fxTR)1^K+2w)%e4jJ>}UBmgZ|yE z`u~m_Z3^1p!b&olu*!u)EiXfq6IK_IyUGiyR78p8!-%ZpH3jWz&#A+;ia0u2JUTZN z{@!rY9jou}7C2wr)AF5sS2x?}&g{~0?FGRX&%hS6%5y{A%EpFCMutIFR<@p1azH!D z#Gu|O8Vp2)t>gYJVTx%^%)a0KVuBU9Ks8HJJQ6-&iDcIo=i z&~kYSb1nT$u6EsMr>|hd9msIfW^s9*{iSK_QcR z>DiEPVqw9_`kkfZHNJbB2ogYnb(rIq3|gtv9t;@s5n*o8C4PBX+VH4^zErFAWY*CK zgj+dl#blvC7x+V<3qAwW=mW6gUj{-~Ss8Q{TpO{22mFlgJ}X@lJE5z<~#r!?82VbM=g#2EBfO6juQ7z8yPUH+YN<P=*x zPhiFh&&&<|sR$EjYa`HN5Wwmq(b@Upn=MNPvt)t}m7nk|jRtCbVqd#{+8m&l$1Y|!j>TZe<7yJhoG{M1s;f^m~ z=Z;g_2y+4UbiJV0NB#A7usTx6sj(LEp^@^d zQ&UQG&E~|^3tcEoflUEC+3pA3IccC;kR*a$+CT+}GSQ5O`6vVBtLB14@l#y9Tcen_ zLw4$0Mr5-5w^@1j8Kve8HDMSSa0{k{_rJ;$9JHH%FsM{Ki}$UbXbbD6-i9u>WNx2p z=nZ=M0S9`0beSfT`>PaxckyxjID`FBnM#+Cf0;P7)k3f|{=mLwXrP--aC}I=`w4br zTA|-ksLKr}NwJCc4W^Q^vH;AH37Yxy8V^#%7ABB1XqrjS0EW+r1i0E~fGf)z!1aJ_ zfWhe>dQpy(oe!gT7JoHUW!U;yGQ~vaeC;AU|6H|}1{8<`DNQ7*YdllB6soV?pX*u| ziP{fEudia?HN}FnR>qwAJahXy`B65k3hEq8+GhU^!d7h4%KB|VKLR%|zWeNJB7DLH zo@p7OWXRtyg3@H2|9G_2V`(no7n zt;l%F!SnFpTuQy!Q(26EPLkxLYSQTk^hyGW&x1-Dw6>U8ivjhX^(kgS=Hvc%fhG)9 zG~dANe!cAVQh8BG75_(Tl_}(N2c0zE>N~pm>7vnx z#hGG#nvC_gZ)oVt0i~kZxq=n+lg}`|bD!UV&*GY$a@P)`Yf-_(4b>i>j-b(9VU!BG zK6vW7{AeoCX@K&m&fC+Eu0c)CK{1&*v#5xjR!QJIDJyHY90SOsNyx~;fa2c*R9p)N zk?+wx!zseJSzDoxCJz-?NDa|I-77Lih9<49MMu8w$N8kKvn$D-7x^v(KT7;nbiTrD zU2)q&CWyO;WY~MBw>H)!#P7uoBc)tD(EH%3O)W?Qx6ty9=WA8A{oiBf@)+xH_zathsZ-ml1qFZ<1J2x!Hl(%yMO8) zNcs@amo$~iIXp`qapD*6HrSH%-jgG>^*DI}K(<*DFKd)bfst#bi{iaDH&r7Lwt@Qy z4X7Wy7{J|1!oe0_&_6?%$19MSZ0{fOK$M`yob~foQJ%I`QpT3UnrEK2kJxAh{b);r z3jUP}GKyafcQ5tAE; zS_|ETpl!AMv-EWnOxL`Lpj+Kz>EW}>XFz|F)Ih5wNMOWTDyYgV>YBQ31y+l80=Q9o z%ihYRAo?X;DAe}|<%Gln^-X@2V*CpnE}c8HH4S5g%8OU* z`2$B#fhaM}uU*FDZeFsE+lx}XclxAN%4Ag3o)#OkT_nJ7iZkbq<@Q6f!qQ1UO+#qN zCzgSTU*G}aQU}Hv#mGg zX~o+|?WtARmwDXf{C;=Fur2vgU0oFx;(wTXw zW-FO;WtkZO+_cRx^nKIrv6kG^)sEr5{kzE0c`hwv$rZs@#@_(B_zmiUPQ9y&6HvccV;3Svv8%sGTsx(7sjVX{L zYDV10Pjpi7Vsy*ga?7Xvk3!ILImHR!7_8-Bf_ax#(tWQ4?DS`SUjg}-#+=6h4`jJ~{uSEioV?Yh z6}iPqdjLywrU{Y^LOB^X$B`^V^{f!NG=Vw!EOs;8=Kg?(|HwzwSB7?(hPKL0n8l)W z8p6s6MF$dDz3ex1lkZL2JN;4se;pX%gwzR7eYf7=&aROD1D`5Zm2)sn?%>d)`J+eH zS5wFk#=%++ZGJyHjDD1D1R8f|;%*O$vfAM;%2Ufu2PodD^CtNU_y|G?b_gm7dpZ1fovt^+_yb)7HF2t2seBP>P znQNwXLAel6q}iI6b-*Xnkj)niD>wL3BQOP)f$_k<_%j8|${wKt9;>}m0G3(OB}78C z-b^Vem0}f{QNK~DmBA5H)robj^dhnctbFIU5nCH`cEzjb>(Y*c(lZN$3h<@d^=7e> z5`H{(B~O;Y%=GWho%i5%_M4cB=h~eyVxMCuFvTjJ{3IlCVL+eV5A)jmC{hU&(1#5e z{udAI$I*>jKul(t#!>W1^5w5^1UFgl(iX7geg!_tD6j^*3(4YfadH0ux9m#J?Nyukvy2B?!p4}#9of()!NUcW3vB>qMr!;ax0R$h<0udmzAhGQZyS^i(a1?)OkX|@n<#Rf%)|__v7i%Rl4b$Z*S<4Ue%CbPS7D5(LQwP83i6x>tJM_b83A9snVoU|nq9w4e`}X(u`oOA_R=0lbUxX&vbl*+!^z=MrL`FFa9DXM$omv?N+&< zFNXyvdnO9h^Z;>&G@Ss7Kbzs2_=p#c*?faR{pX_!?}0F{qSjCc&h$;3J}cUhi}1Eh zlQVG4q1_O^=;)uD5{+~r0q-S}=kFZ7I1Yc_;l53EtFoWYwxVff4vk7#z`R8OY<~9w z!P8os3i2h6!bVLURf}k)`O=g;ZFyjvvH)|A>3U~cuv~5XPgzv(E}((GH`ZwzRTVQL zG&r{FEKH&5QiwGeJO!Ki%ils5+KkUhj@Og=*?D;HZE_ypQu8FI;@=-q7_I1w;?Rr1&buZFM1S4f z9mjez|ahhB{pj-a5Rla?AdhgTX!(R>ow1TQ40?4+l=NhCK=ByC{b^6*-i z4bs=ee5S*K%+$*DV7)OT5>Y1~#W@?UVKx0^J$OaYQ#ZW;SHuM!_*&P~WL$3o%;qIT zw_>JwZRKrPO69IpvVz{6t+zv+Yhs{M20>XM29TO1)J`0mT|LwEefpEtY;pooqZq=I zlWsRCXRoz!s21BXL0A6@)ivU)N!HF#UwCpxmllJ2IA|*ecImnyn>>_0_yLTV#Xuu* zF#4Ei#!AP;B)b9fJD$;tJVNGRzG$B|BJ#40ZgLAO#0!bHZQkA?POYb#zTrtYtBqEy z&%Ex`c*Q7c6^+9u(^Y>OLxFfI1Dh!LTBj0{M4sBU0v81K)|>Yi&_|xA6*Fg~BhT7A z`v61lguQBu18800>k?#)dnESm?WMbmFs>Z_OKE(ko66`Pf~|2{y3XY`Cg=%OH4!KFhB_jQ5cXX52}8nP1rrC z#_)VXr=+n{qLfF*wUGQ0gD<)$FEoK zbO#!R7zLfLK_*c*APKPq(^WBm#s*%*5VN1i;``cbMRbeX1yfJuI8_#z`~4oN#=P^l3=SIB=!a=`Fk;>lU`yAU%`!F!e% z3H4oSQ6HR`Dhk>{UC4~|5j*5lNcZkvF0?AB6<1VM=gfe2dBhHbUp@;}-`H=9{rDs1 z?+YLfHJcIiv=cOXtDb^RUOX~BM0=V0RvJ)h4c$j&v6jK?7l;q8o%tMx&AY9%fs%g= z;GHFBOKOCllzIeIW}3*56mti@6-MpGKz^ zrWV$)Nu7HO*sNYpB0H`E-0Mw*W~k!w5(Ae{u@`BLUhGxCIBnKm>xRe&qm8Rw!NCSO z>=vg^24?3U+jJvFEHj#nJE(%`73j+AopJp-T`wF0}E!1HoxOXL|)dU9kuootdNWjG&%BM>(};W<>}Azv`8M6 zj&Fz)aXXr9SEf=M(Eo-O%&V3TU5Qvb!%}2yyaHy>6K(uI+COuXdT;3TN9MR ztcR`TpW9qL^AJ?`_rZ%2v@Rm%8kEC|Z2{u^_U|%xngEBv-y7tur7Z#)e6bhZIZ%|* zK_D1<$9s~{7X(ec5zy|{zICfhWaNh5zU%ZC;4Owg(uV)7fOmBqnyE5JiPZP`ibw}j;(&c?NNj$K>0lEur17P61W~S&V?4D!ii*|ibpMQFsax@=#c}B@v zPRU2u8GFav%ReCC9!SuDcLcFSz41Ikf$?<~@cG(86}*L{ zAxk+pI7I4a0gN4}BP1lZiVF_VCjsH;;u0qcB!W=&(0Ru4;5E7}KY~o7z5V^2x7Ule z%K*Vy*H_SHez^yj+SOTif`K`UmMkS6?~dZJFvCYluo3`y4OEet&Gjh0q3>Q?4*9x(QTuG?XhER4i}3V;_=^q_vk2? zQ(&KLzeF?n6o;XwftbM|811T&kIZmxssSB?f>~Ug&ahrC^W+huL0zxNZj!2NaJI1b zrSk6hY4llDOeYYII{r$A-khLesBg%30vjrUqDJ5+az=eE)xa4hwFU_HAnD1jwj5mi zG8+Fk$hZ2PMUs`X>`FghV)GtW-Tvqu5@$m66-AW9bv<_rD~4h7FXtVI&yt}=t^kFR ztqYs4M6tN^=gI@mUTb9@#YkbM8X#DRj>o`6+5GN4oaWK01pws|!Rj%9uZ#zIxnnKb zp3|4v0tAwRhx@@GfoiPWQ1tS{jEO(s2{;8%fuRb}cml|;2;a%4U4bT5rKrb#aHxZcq3ikMEU`irur1 zyU_@QVrXxF>SjWlWv#ENa%q>Gu43@K#ElDV=hp)ly3qzc|6o5!WKAR1#iPILhjQ$kRr7*bO|xqB*=O=7y; zO!>}C$IgzEYL>)gio=SnY##TH+~5)A^~Lcsk7qYnlGyww>p0&*|0Ys&^wNU7Af4gO z5d1q$Z5igFDw>G!&ml(&QEY6HsbF2J6VW&W*5>0;iB-z*7KEk~6CnLiYqa#$~d z52&It0DLX;e~9EV$*JghJQ(jbnB0%F?psslUMJuYPx@gKYdKi?@9~-)AFacK(DTx+ z!%#!(*-(~wLJ9WMd-_m54<36JHj?Q2v$p2-pQP3qIIrCBwVSzN-oKB+tbhXG=K!R3 z1X0L4ir2n2#_X%`2FNJL%ljHS5C*G28UQ3&PPE|nhxxx~>!;E{phc$Z&Kij#-Li8r zb+)h+q{+Qf4NyHUhQmsyJ~S1ub5|4PV= z6DOtJ#)MuyC-deO8X#r#FSS=f&+P6ZUtSkYnWV9}XhPQ!|3e4Jsz4qmv}-)ER2~vX zI}?v;;sXO&&W8F=?AYB7`w;R}ut++KM|2ad%Rkbc?hbvxDbrh4Fn5Yj)u>dMmK632f){A70-Z+qU^kDkVj)3_!Cyxs6aty&&ZB{87<=Kv8% zA9t(ac{n1qUZ~rqqKQjb@RFIzX*q8#ET;?pvUp3{|^z~5V zQ^729ahuL@GA^riT08-KvfhRDpF*;eugEIz?kngg@A2zCPOI`eHC|fqpMO|o-8X^-=sJ!2(q>!F zQ&3+VbqC!v(WWP7?e@bwlC-_aM1a0v{0}&c^7vZQ$X75WZ7OL(w&j28%(^ErJ~{|F zL%6z`%6*Iz1(JYV%E60Q+-Bd9AhyEjt&*2l9K%XKO1(H`loY}!n9k1oGmh}%(9cF@ zlR}6?#gMry$mb&9=CXKKY)@>-hwQX=%9aP;FA*~a=TkrTm(ady=C=>Wzk5Ke@LlVn zF{uwZz<>g+pcNINz!TO9P8Bv4AZRIx%D*1hZI;tMdY8%~pM7snZj-O7h*> zVf^k}4v1mJZsR3oOGPEnCS55PWd2m7M5bGF<7WJQYy!uRK?|_fQI+3wA=HMU?&c+rD?x9R59dC%U=RTu#e$q$XmXpB>OtE zh0|9gp?1CtM=b=K%%5KqMA{9#Ml=7S8zjm#0ATvDoNqfHr7m6UNtz)=7+EpRBE(rs z#jIH?Q%f|snFy33-v4vKYsKlk&b23E3}S!sS>kirQ0+K9RXWIu?#;dzy z`)XK zMNVf(#5SRe@(G3YRk1zcf z-K~|WMTsg|m^RQ|b)K@-<6=+;q02uIItVqJJr%v-^2hB7{m z3Zxe>5Kph9hM`l&A6fh39G3V!#M9t-7fJQc7v8cj4UtW}!Y9E}Fg`qdcDIcOsM*=w zUFJHC82yD3>hXcV!OHoM$UFPi*BR5H6x4dS{H=SeAr{)G!;0E>xLf9PBhQ5#D&KvI z`XRvb5BuYQrl_&DbHV&k)fk(6C;d0OF;x|}zp}%%GVpiwbIAJnQH!|5c9m#{_71LI zNy8@!I5nY3C)1ZcTH|Cfg6Rv2Do2R~VcK3jyw$>{>2EzmtOnMGP5MHjSWfcpdD=!F z&flL0brjhimZZ)-5P^lI+ zYj6oVWs6oK_-~k!S3qF{yR#iqmk(cXA=3~+%ODVvwN@bu#PF4jma!CnzZisv*qZ#> zJ7j(D8s+akQ7j#f-PYZI|D`w!AY8<*7dEFc`ASz^iZWW?)kq)={|+0Q4>`NHUem&L zXD@P+YPJE(@7M#U>=0YfTx+r21qbih_>(*V&40t3JXQ*-j3tjRDtZ6~7w{$o#(9sbkOcy6&ukF$Hh>FNl$VQCzpTO2xn~|Ekp(&Q3;g40F#zwj*-C2LSO`=?>d5Ki@ zhd){j#L6pdIh>X9-}4VDg`tHwR>0yV3GI+S4LO+N6mkK_Ui=s`^vya}b_0ffG$go4 zcue>e;w}Kd2@MB_zk19+(g;qIq;dKAPC09kux^-n;jqyv89D>Yf9eV{q4_4Y%Qvy?NuBtk&6rO&Ye3T3HZyu^m9%g+sx>3 zN?5w1T35=i^-7CQ&XQdF(A`|aPiF+#4E$dLpOALWDo$O>C8JZ&g= zpC)Q=%kQy=I#)5(bZM;`ZQ2QNgJAo=N=aL48m*(i)2w;>fuf<>wwO=jb{T6a(-q7! zJdP~f4a(5r{jsqsPxE~|MeVj*X}4i<;9-FUU(!M`@XigeUp;74MkoOn#k9Wv zSjuta@3op*LjSJ05YDZNo8-TK>0&bXSu6SDt4?D@v5qa2Q^ls8fTKdUc1~XDPoG?* zIay0VTsS6y$J#9^>5b9h+d}iE3Xn{#b$uzT8=b*W%~su zWD*Pj1pMhHPqxm^&XR}YoFKDFx}DFLqL&xrE=%3j`}oA$BYG-&hTD7~FMbj0pl{{K zn5{XZ#2EjcIg{N<=0)rlC7_7@DEQD(@|V02zcIZmCJLK>w-p6>r@qXxYwTI3NYgvk z5h<*Ig9uvt|6I~7z`k_09~^3%4Fp8kU>9?%%+HZCdvmqv2~9MUb&@vL+tfbT(1b6h z4VZ=mOrQ|Npggh&Atf9TePinBe6Rk~Oo6qe_iwE!))R9{zd`Dy#$O?KrRIPIFYy|O zU7&TFU;)Y*0qRatMLKtM;G?Sc+BGOM_|hK``ua0y7^p>E=+wp6$bo2JtfgmRZDRu| zUUeoWzmA-ysuTR(O=4`D^)iJ=)CYV#Nb#C9?q4_1h)+%q{j5gxbLxWYc|7e~HWfud zb8)*-#!R34^tsI-lY&J^hz#TvJ_3`kosG!`wVtBA^Z2p7`4;DmkP73L&j4q8GQCNC zPQp48>~?ACd6@HeGcOWxgfV?9Y-#?GQcfimgrGI~Je?`dVfF zZKt)U-Lfm>19M@}HO1M6%)IqP!CrM+%WZt5Fwamjyy?Wi?95FUGeYWlzQ-AT0~l~2 z*Y_q0iDnyIO_>+n+fPxES2+Mb07cMxw%$f>S9UB^q;5Ua&;Vq2>6!0NJ5IeKA9om# zg8^*@UmUD|(InPDfyN-}J2S3y+KY-KZbK0=7F`;8gX}b>qQ42$#WX1de1n3fKOIBM z;#~GhTCHk%zX=PXh8C&A$&S)j=^D)sKsx$FZ*dccY6%k?8@Y4tx57;8}Q~`ua6b+$>})1l}(qux%$CmO3?Zb)KQo9(Mp6 z3P?)}HWX&Ge?X>}B1nl$1Wqy96IBsokYxyMlunUumhh|p@db&P_D7%e%u#i%udYt_ zKweiVVlW^T*|b>NK|W>scon(tv%0o6v#n|YTn+Ba8t*--TldGiOCdE_0F}bk^=MSb2IBcW&t$5Nj6FLy+9yadzzKE+0>nkj~rtDuZmy32D4vge@DXU#H-c3(KLbn0=L zON&r`PYFP)`St@|-}dGKK)all->WQ>Dy1Rc9n>mMo{c;3@iwiJLV856VNTXUF`5$a z1bNl{08koQv9d<^c}2EBcm4TLpOABdMN(3uIxn4d_j~b& z3hjwAUH2&lZimu;k?MWW^9wHDF>=j#MZ%0i_;;hH=Adq{lmYqop|lg58@Z9%%El2S zgHEP^GQf>MXxsv2G4KCv)zS_V{E3Qe_v$NMdBiv^OfBBbS->z_7ilk-feeb>eeP9f zA-jB_NtP^+RWs&u&MolM9GT?1MSQ>@BuRdM`x`liW`!7`Od6U3pa{#;>=78Vi?TGT}tp+v7rY-1dE^x@4V zVz;4Xp>VA`l7C3+LxA9D^B2RyZzk?nSHSm&0bH`zQtTCdBSg1t;FX8VmhHm?U=UHX zE{$K2Oq24;rTw6&?tiSzHABTxUZSiSZ_gi&5}MEI-mikHMU~07eGFHsv-dfHb*c># zkN?;w)7f$Q64;Vi>#(h3G9on9khZvCM`l{jzgvk?=Zx3Y)i`Xpg~yxsKW?Ag3taP> zs9ZY<@4YJ-slleW+zPQg^^C?a2IwbA>-=KkbVztw(-+vRnmm zAU5tsPH^! zSY5|7J3mCpta_ph6H}_Q>_LgWLl(tre3tcnZRGUe{FekD+KyfMs#Qwu%Z32}q;9&j zwz;XU_}fvKAApzqQvcbT{R60P2tyA}ubq(YiSmVF7ziYAd~V9TlQjh; z;-+yrTG+(bFDp-t=bLaoZu<=WGa4}5Vf1Y;Rmo|f1yw|3L7TDfTX^!xk@w6D;uQv> zGs0!uPRDY!x8&1fV}hF)Oyo}ZVE*SJ_}+fYe@K4O0)ZY(W6iq>Eo4gi=sQ<9(+6!$ zE*2OSN9M7J``S*yKu=31^CM@XMp?N+LRL1^_o5npc_P0#$%VnJVD&2m4fQ#phFpm@ zDE^K&$Iy(MUsSWOid3VNE~}V0&o@&*NtZ_fFRVy@a!%`xKf&PvDgn%p2b;C3h8Cxl z@Dm+Ic2Y|f-6IhFVR;=|ssC&R89xCa(*TrQw~ia; z;-KAc^M^|2q|n?_FEt_ABHcaaR*KOrx70?`js|yTeqX^V|Ax%_8M0!T!BT!clFvbw zNal;?&wYhJ+oK+#{2#E+J6}ONpBF}Ii=tdR^pcFh3J`MVUN`W^wU`%}`-Ip#)&Jew zxE-GerQiy=v;NeeVff|WL-WiprZQ%K%mA*UK1YT4?r#tgnt`rY*IHB~EoJu~4 zV!)NFf51~p?5cQROQ&z05#jaOZWt?IDYS}AQTL0S>BcaEqcBMZHd&I4?8x%S0zZW5 z_%T5FHVkl~iD67Xm>e?J*1`^d;9U(X<*RyZ_(qdefj>@N=KCv+bJi~9e;wX`cyrxH zne?pzotI}m zi(dyYGn{^Y6QY@Gl#XaizO(9=f?;K!d^DnK4~Kq^UszvpT{8Rvmoe2E1yA)G?|0$N z3((gK58%LS&25aY?Uq>B&^$L#8G`XuvvA!+C{nQI%x~(9TUxWgXh0wQ(wa(2*cYe! z5FZu`02S!Lm26F>OZT!nT${|R4U%ec;3*nVtYr*5B#RQkFu02;cRKCw_2`O}?8y#? zU+=vNVbhn5Djx$MO^XTSqm0O**ogKS{rR5m#wu8m>_W+y*g@AyE;LF|df<7EU3{iKO z{hP2kmVw~#qHk|xEPucIC+7~CGtxzk7?MgswW^6?&MqNSbRrONjRP(EU0?LZ1BInI z&(Mjd-nk_}G~jb)zSM~~j}uywa{nK!&7mvtd4xFJL@=RxJo#oo1vTc)+zG-g^+Sv# zq6;l7Bqb7cNC8EYeyEnxQqG^Yi^19&27p=}*XyW+UBdje_yyp8oWzD!KcHH zrTeX*Q!V^YTzZyhG@+piZsj!mor@Xck-1d`!)jr%Coe5(Q9}^KqBSLVwVf>Fj|$X= za*}OtT2)g&tP}xko^d4S0-;dwJK!VLno6pJhIbrzVLw)zSbeTZl#ExlW^RYk3pqJ< z8azCVoP**kX4{bEj+3~XQ?C0*s*~i|bY^|R$?aLxRk2!%KbJI;?bX||JfQoo^ z@ksleUlt5sI?`mKJ#~2W4;6i#So19`LIl>IO8&wZC2}l_Q3TSl{~1jRvz7sNm2$1> zpXPb=qCpSzt1)@aD@s5-(E`#o!kc2)G}98>Na8L zB!X%eNS@$Ur1CtXA%q8w8sX=AV{HBDBZ^+f-8AX>FA?cwz;~qA^&~QVK^^mryxWa7 zYo*xXlWt}Jto-kEtbEgAv>yn`n}M4!5zlocqet&aGdd6Azdp}MSw$AYLIji+~^?q20Hy+*(txdLVt9#ULc)5 zx4NZP#n3Ko%^*p1SKf-H3;UsTd7K&Y4k^SO@MJI@$h`2BfUS^68th%HWB{50%QfKx{|e)^sSEQQM#=?5$)t0r<1_^ zovYZ=mk$+OT}0InO@AjowdUaXTaonM)nt}l22jS)N!5_jmT~snE}m;0NAFwd;{e|X z_{8ki6pUw_w6A;hzy%M;m!`o4z$_u*oL|tasV(J)0yO+|L5sYB@I=a?>m6#|HfHq3HgP?48!<=>-Hi*m;$hT7$n>Mlifa z1XGWOI=)a~W5Yo5#n*zbb{hqUIQ+L5sG(mMb}w>!uVz|Qr)$hYTb#zxH)qHI7mZr5 zi)X&M$+r~s<+^U(cPSt%^Z|Z-_i6J;2-k7)^NIFmDKwa+;rlt+`xGWP)k30!ZbmNR#^E~;D4DdOkI91HL}^K!8k1L3{)7J!Hufi5-31$AYtOO zjNw}P9N-(l(F9YADOP`b%-5Ol`ZgRVtY+UbRBgp?rq%dM!&&R2yPBE>&)B9)pF_vz zO2a+0)G48wv{xiG^9GH(9jkYCWp`=j%G9i{4L)m8>zp|=<;2Hf2|)Q7wB^Z7KjZ2- zhcO=gGV*U95`ZHuL)V!@*(a3nD@@brhG&``DEMD;@;lQc*xfv2h|k`dCJ|dk7X>$+ z4h{}nDXJk1xY40rDb>gWhY(m7{I3Bbu&dGimm379Z@=6-I(gTx@oyEeg@xI4SqhLs zoJ@=;QBK?I6;H&d?R>F`%gZ7=EQls1OO2OYR$$x)=y`?yv;=q6|55%{3iXCQ9*Z$@ ze-~aAL8Dz18fbyPwqK>NLIGen5k8jS&e2tHE zqJLC?CH*o`Jv_>h4uni0X=1`C!2X#gY){JDxP&?fBvXxGRH267(xars0eV5`LF}+D z!>AAV?PpQIcM2H)uA)`Po(-n&qIdeAm%xLJYcZRFSgHIGXeGLfWkH?8Xvl0^iIAoN6>b2=bg8f_@?)@87?q;O$ zVY`=9Vy7Y6?g+{(9bO(UaCZ9>t>VNza6bEFb7#H(_KV~lJNE>8q_?1Rz|r(Ewu-5c zzysI2Kp1G4!Xp&y#hH+@F5yruWt1V|82uxTfL3HWT-(YjPfd@-t;muC(tFq90fW-sImK242l%VOhWYNev+XR1mD5hTh9B0XED1u1%crGu(Rg7Qt`C1R%9`wN`6Y%4@HH8oC5G0|MWQ+>R>;e1{Te5Wf= zx*tTghcwn;@^DzM1i@3lvldi>&fbsV;oXSxz@M#ZkHD-CjraHJx2wsEhLC)s<{G9M z*8YJ7Y;r++ZO*3d%RFv$&=e(pjKb>Ky4G<(Xy_6x$%J#xlIoP>6Q={9usUNk0yOoeRwO?fZ=-Io*8Ry{()+7#s0 zdqxgEBm-!e@+kM#Z;tcOKfZOfq8LD_xnkhLDKqbVPf%wE!>2*VX8=}mJ~CwMP^P0* za+Hi&r$Ygb`eN{HAf%#Nn5_X~zH0^(06Y5UzN$!ZKeIxs3OM(G(xw;6GydC)p7CLR zuD&*!za%D0D5$v3-3wgDWR$^}J~TD^W{4Oz4Ow9#8zaCpxVwd_XR48zv*Tx1;32r7 z1_+#d#reG>Io0_6q@L5%gPwU5Y6J?7dgwFU2VPS1q(5oshC|ZC(BIVXHT}EUe zaJTu74tyfp@@I^oaEys*KF29S21#30A8FLOUh~?okDfQ=qE&V z52|Gk>U>Mf7Alk0DR0|ftpLFD@&S`cCtt)N1XcV4&^hToURi1$0-@SDy9;f!(5~Hp zzr`f%zfw@fno0L0ZO2W*4ROt?@nFo*4m-j5a|_F9)9NH3#C~oW%(p+oNWac}f$nES z1X2{_yiJUM1CWk2NrDqo=ME1mb-c&gV?Acmd5UXNG`|!SBP&AWTbbsj(wLZ#6rRv9 zLR3t`AK3r!etvH$h;>wj2|DDwk#(nKvDz-Y$aQ8_Za2))6FEkg5c`e>x7@m|J353* ztVAdzWcN#-(Lfq!py(HEvh5xD7p+@^Kv4 z0P0v(u+VnN$o``HV9 z_^ejy56*rs=IZH#GE|(>`((UM0-sP+{#YjoOggJsUM4@A=6(?TJ@yWwOdl*(7Zjcb zJmhzx=J%`HPMXY(R56p!NM+Qpb&|jO4qoT+;F%={%ryicM0%q?|9rF^MZ)Z~;?SFL z&ns^nT+18I#WD~HqRhC35`Y5MF9!lbCm91VAjib5rF{Y=YsU5q*yx_z#48*OEw!c0 zx;gpqp-9v$KFry>`+jB~6`U!sRXQC|lg?$aPR9v+bz_)P5h`i($3HMUKd+!OD-u`z z5R&ooTK}#*4)iUP$mnXVl0J(Nui*B+;50jRGMjqA=v(2K>lr?|K4P>MR`k=wrF2g> z)9X64Kx&cPD|>cw1U2)<^6s~~HSvZ?X}3IS|j0bDa7$p~Oz4!S-Ley<}d>oUwaY&=rxG1FxzQ}KdWp*wVC1Gu_pEoN6bPz|MFBDYl2=kK?bzT@K1E?}&i;N2gp z4yz)zr}0N)z_^&RnLT1d3CrOpKuj15a>>x^UE^1s@d63nOu9F_p;x_%Iv}yToy;JH zM4>(MoNu9F&$$27v`CZM?=6oBmWp^;3!yZ(E6*d_;RnGN>Z!fBIPS`vDG18Mi{(t$ z_OJ3(#^FG!4+sm=MOvNxh05<)&n~trcYk^OpnR`;vj-p^Jpgzc2jq4nJcjjU2jd}N zA`1@84+c<3>qxIXiRHJkeyUtgE1DVNGla*=G${D>j{((v+Y0YB#YewF!v)}18jPQL z804?$&@STXLmv}i#7ooIEf`I_{b#;r(GfF4KJ*NumG5q{@y@LG*?Pnnm~WVqgNv?+ z3d9@;w_4})rl5+eKP3dO3 z61AU7|5niDG2K(p+2!|MtNHp6BW1(?7r|0~Y6u0FQT6K*kflXV=29jcybk0?fM0?~ ziBhTqGAQUKCx3=B0F|Yss#xb?u+v6N)NLk63Lci&=Ki}iIlx+`M_MV-c5|@};kb4- z%ONdnK+Ye@;oDX~;%Ed+3bzH{4fB<*XeCuuAgqLHA5Q+n&F@62UFRYR!btRTR#a7L zaP4%WH;}M0n(X_zR|MYa71Xp(zJt_nm^^pcW2 zH#djZ+Q0`=ZbugMQNC-sIG@eOsr>r#V?H8Hp6#bmHhLT}pV?zR|IoKf-Ppl+mx|(l zQ>CxXG2r}%+Sy}vNlijupyYR=#atYwMSRsOZDegog(zyMKz5d|oiy&IS1rNAH=lSG z7jX^p5j7!4-BM>~mYMUw|E;fqOP;zAgv2!@PbQ>@(Nb0h`G583j0`g>U`oTTQ|*9Q zjzr5{V}4e*H9cly#1W&_FjMcGNm6#Tily?k1nhIOS+obhV$|NxIuI-ysfBFtSJE={ zBhP>oT9u*FGG6?9lCo979r;pfg=8He36EUxcd$R1FJK(N)#hS=dA$+cJb^vxKG+SQ z7~Q_>?t_JLLtgQ7K!OMu?qO=(hEBZdd&H6$&pR~ppMQwO2f#j{yfnLb5e3`P|G0yd z#VW5!F0YxCQ7C6a#zjv~(a(ITBA$P-TFu{+vbrA5gj7j6k&rJDDMWJ~N!VbvGV%Yo zCbslvBlMmtjGh-IpkKqiQv&Up0qu1=+KbzU9ih%zvtV`ufcHg=$oSjPq1#xu-8Aot zYTs#ne{VXBGoji;@U8VNyHa)C@tGvJ=pU2%oMf$ykAz1c z9Wfl2*;wBof}r%g=xPWs>{Wlz4?KfD?U>KcC3zXhTLP>PYqdeRc?h?_oGR<2t4JER zq`KMPjROvM?9@^TBM5F z;oJUFk4y=%An|`*u4mp|bcom~ilcSL`8ue|A? z2M+Z^YHCSJfUDeO;l&mz&k#<0yX`vlj66RG{t3YH#LocRb5)c_W{YO&oeJ+)km<}g zm9Yys>;NAq*eB%&A&3j<22q997rL5NF;P{rXn+~4@Z;IolYig;cLMsg9bS3_J0-jl;( zpa;7rOb*r`7FPy6Ce@=HZlABHU-l><{(XGxlTS1WuVRFj4b!Zgx9T3i$C=FbRR!h> zdXF&RKh>w5xV}c|SafMtIYkPY%GoFT#Q#~5#cfd%`A94- z;v*Q4-*OlnxW;IeVIw$Ewx^JcTqk829FZYj+!5TJU13XRhL{N!6l~JA(T?((05E6Au5q#NEnzyEU|yaygT%$?LG+7$eD0AyHa3ta_6eHwnoc3sH=Qx9kYCFDPlsI!*o`?o_$aB)4=pa$pHjW1^Rzl&TGm2WUbsczS| z@rl*5=T4Ad9M8z0#$Ig!Qya;%Z?*#C+Cta|%;5%+#+rkTE|*7xo0dezUa~t}5{is)zqVLJt?zSXc0WkEHPd9Fe=B*3 z?zR0aP#-&_34pc%{C?%I_ZcQ{jK&CN65#$56*1{n{XWeUpDME64-hE&yJV!Smq>-D z%GR*xk>2+WV1BUPg9xK^Z&1$gbxtA!Ow&%604B_;6?vmE$v2r=X`0oOS=&s7bG*Hh zt*RmIN_rNDk;-I+LdHrRdWu4KJ(m!m6c^w7vc5lv@S80uOwM7Wtn!zejF}bDwKBJC zY&}Ov1C#d97du~L?sV8oty1*SWOwK?b&uEX*oSu0UV_ACFWZSbClhL_RQxw#a$v0lu*5CQQ9q9&EWhUb@0;nDfH~mb z8cCu(aP9tWh41HFy(1(@1CgfUAaKjGHp&+xBAzW6r}2NsTYAyLr^~$gk*4$;LY

      y7WB0NPj%Xx3gMj5a3re~rI{SiIcmm7gOBvu1;yn^P?)D^rRa>1FL*u2O$uL{JG z=4!+NHK?Q2*IO;tm5c8HSf(&_)N(z_u){C5nF#t9O4Q!-Q)X^vC1=d$oV?HB*@sgl4M;o$%2)2-UXeN`)z z)KPhdpNX>gpXtYuQQ$aICzak`6ja5I8p}KhIvi^l`jPg_E?7F2&E=m|Zh za%zLO25Rh0?VfSDbO|$X^yU9TLUWKY3v?@WBD0XBg zfzNMbsRl_jPH|*UE*dy|yzw^;!f)4J<~c33o)TdntdoQmaX!mfTq(#XERuEBh6>Z$ zltf|Mfz`sAzm2rgf=r~7KE)S-eINN61S2HVW|JtDAaUxb8 zOKef0=X}_R#1c(3)YiNC4Pkmx#!cYt^&f2(l)z>&Il7wY2a4HQfnK(@sMdsKyoEoZ z3+eUO#WxnXj5m(w<}-^A0N=Q^Aaq$UDR8RZ`W&O|m3qo4L9t_M-zjK)4Ci*f_rqYo z?umE9@Abffq)85+t*Dd=nf_yi%)r?)Yan+W{gsW%S8H&la?I*k?(hVquQSIMVB7AW zxL8HngSB>)dfw&q7YbA#zr5GmSv+o^d%IX?8My-rO)alP7Ux!GbFlY%v4*C%?|CPO z$a~xgN8WH8CkUZS!dQ3*Zsk4L++4%h!Q9!4xl$;tXUBBD4DAXV{Udr{f1u`nt0o-$ z-fuT~#;1CT!e`ZH)ZW{TRTM|zN8!+1<1+9S@N7a?O~3A+c3+dLBw0!6Xn#CEvd|qq z=Hf^`uM2Mql30cY;XYLK&DV#31-f%2VRH1^E*WteNB!H~J+2&e(9vZ0O`XjyZw0FUpaOgCHQ_94Tp`Y7>v`fj6EZpg3Wg2WxV7CIcxe5U|f zHh=H9Ex1zakT@m?r}Wb%c<>k7zBk)$fx_HB!uF7%`8_w-275B0!qzdM(GUmzT<`av z&F+ehXY~pD+57`vL$eDBC6kr)XUlGhq__9&I7+;!hxQ6~rrR4^$?KcNJ8GMv0RR z$D0fWJ@n5$p!iv*5@i^O?tS<&f#(SJ=vXWn{H!Own4ifd5r-7M>!UnX1LMwQX zp=T4c?W-DpjWTiH2p+WPUWOpgdDTm0l+%|h6mdX@T=*ny`)f{K#dbm3_HEJ2hjyiU z*9Dx%3ztrV)gZaCf)*zTyvg{=902I!-GY9$eFnGaa9FvZ0Wbd*!~{CA$gA zq2ZXzV!IG0XOWtCdF??XBbchg*oe;c8*Bu}b&pDIn~w@@JkCTX;Tn1#5<(O-S z;@c4}1JO&@ue3jhr3vP+>maoP#_+)B?8-b7wdKZZ$&L>{930bA4U5OI`fX3T@kc4G z|9his;|a>s_+?6@eE}~9cH5JbR(lAe|1QV@OFY3w(G2{O^V1EV4!?v)UXmU*ko(0| zW!Dj)P=jrP1_Y}iDr9JF#9C?=l|8pIm;#J>_*GaYpPfU^Gg5+&vr?= z-(L?Bnfw!!6{CI8{#P%KX~I5uFRB%CDOU76XLBP7Xk11{kZYbJ674qngMXPjR(+ZC zVAFH8ANvLZA9dI3Exfz1V@E3A#MBRk;_a2;|2lR&l_Je8Yw+}-bwslEo)S$kMm2cg zw%7ozgcd+oV<$of9IEDMbN7Ewk0@e_gT7o)Etl)*z!Pj?Y1zrEH`{jk6X=_xz71y& zBcW{(NdMn9O;jguIPMMi(^$+sRehdqLN3JmLD`Vx)O|w2U4FZM_7sM(zfHfAFQuz^ z;p;bY{Z2|99ln-xfesDmZtLcyA7|^>7VmP|LajVW8(U{PuORx45rqx(n~|*t)wMmx zeK0Oc#=k1r_onpLYcBX8b;aM59-ei~Ig9s8;5WhWsnX;y^0(8vElH5RfQWiqP*Y+x zaN>z7Dp*)uyhESd2Aj;y4)eC@3pn{(VjPSN%h8+C$+{@pzw~VSa^ZL(s?*)qIcYvZ z*Wqgy4hedUj^xo2h7$ zdIqyd0sxByw^eApR@DfZw6-QWr+Np=FQ8-q53dt#cMv0n& z>)(@ophn2{vWGO=%WKsdODEbxP`rgK?X~>VQcHj;agrVMAKv)41D+}IZH%`~WR+aHEe6`27~`W-#F85ZV1^tERGUB~uM0PBBDTeQ z{I;I?7?GJK#SuCrQsIZ!Ws&EGqDX?DX}P`$0a}F%IQl;|xMFH10qv$$%g}pX4gXE!CUzmt2*(7G$Qe=CT zpBx-dR5Wu~u&vKi6@uF|p2CZa)5#jGc@UOV+Z)tE=oWiE#WlCDidS% zX7$(~vOMziU#Io9w8Q*9gk{oxCI`nfT&|d-HB=CuK0_^S<;qXdxyQwt`Ci{C4aZ2s zcZN6bwWH{P%TT_>kyg2Kj?b6JgY1qY*C(ElJy+BG+9zkZR)I<3J+`SbA=)k2q^I*i zr^u_!ztUkeamJe|?|r@%8am9SM-~ooPil@^JOqVfKJcdK`Y!yB`vh?lA)VnOuyLM^ z!d!z0<-&{4he-3)fKWoZLC(@|WYle;w5Rz}!{JqlgL|p5G6+_g&Yt~(&8Wa%w}A#X zbr*C$ZkIZFt+FoY)PEQrdOy+h@@SSB3(e+7k zy!pF30d;L!^z$n1y{{&juL$f?B}x@{?W(m8yWH;iB)0Bt*D|r!?j#D5wj|s1zC4`9 z*k5>|iToyg1Bq3P&k~1~cy5ZW${4{O6XcHf)k~AXOg3`1Q`9Y#b`~6}Pk6l7!Tgtx zvqnX>1w4AP$I&^)?c^UXzXz;VR@JXHi&DTOfZ;PN8&Z(#=q}SoILo)BFnVVe1N!Z| z>P13m@T@MTsJ;a|u`jvazK?2eh{sM`rLH)**C1U~Hg4eEza2Q1*G+7%gqAvkBGpC> zN9~HIsPH&1UdEzLFDNe90(i_CMhxwaUv2n+WNei7fXb>V#*B74M^KLl2^EH9hi;E{fex5k$C& zR_0wAj?W{051vK(+|%AJcC2ncxVXYo*^XUJr&JSc zpsn((1*(y2T*nNKh)+s45-Idc2$^#JEL&?%vmUGAV%uZ;#?j|KJtT4W5|ZmtM+$cl zLecykd6g0e|D$@EtC1K<$LBLRVLcIOKT;xg|5#?i*^$d9X-q$=bNH2TcT;+@$5j2) zB<6?11aU(oyVO88_#5!_G0%T?S(*>fNT%y%UQ}V%A!Po}+fVO!uzI(r5XS#D+~63C z;0n*3bu?={_`P4L#mxur@7X0wLn0EsM!5c)lWD5|Q%>$2k&a4@-tSSHg63$EXcH1q zUwI}-&!W^`a|(APt~U)LUMs-k{Q7?5v=c6-#52xsTU5fBOqMLNt&N<6Y17O&?Hq+Yrmwto-<^N$j1U9OSOP1^ZV? z0i||C&6tEDu}&h$diSK*S;&|tHrue5i3O#5`lgDCE18jpFusIM{=MgjN^8Ds6qU zRhm=f^-C!ffOx9JK&7A*K`Ivw=pTz-`thSnaFLfQ>eH}#!3-&6-xtLtW*3y}=X>Ld zAWcu~9pXzV(4K%M`KfKeeLTwS32Q0RK`KpODTyHyvx1y(Vjn>}>NhiRNIvhqqjD2? zK<(u~!=4->wPvusd$3^x!|hY*HYIpFMXDb7;vtN%`!b z30g6L8D&%#g-_vSM6=(_)+ssUo0z9KlR^;SH=01Rkkftw_aW}{4s0>bVUz=(UUZ%O z)6ubXldb)CYlFxp?{Ed~-L|C-8B0N&;V&#vADGhr6%D-rC?~+1NEP zWXb>(J5GBJ8*-htH1NE==M_8GXeB_I-&wJl!GPy0Wt(Yg8+ff&Z5t_mFy-;43_U9V0 zeQS#F!$d=#(U)LXRqTY$BJ?im+TV0rTHjT6asXZ-}qbV zFLcx3wvVq~vzlPOX68MP7E&voNAewfBvPMzZ!%HGHCzjI8aB+B)>h2s=q=x zCc9TK<hl__G5pY^MExb#QdLwxC)o*5JsYtpqw>JUcsx-i zoz(e@=zV}3Tmz4#?IbDv{xF!A-b4yoqygo*RL=ik=|1lprKf`arNp}>pgYB4Pq`@G zn`5h1Y~leg%1he+`V4b`L!PES-}nZBF5#6jA2djGz+zn=;O^bLs^$_Gx-2c>RV_wX z#%okq{X#>$Bv`e|pT}09(p_X2^BBKvpOVIV5WOTZpBapVc1R2@1g?^ZBP@uB#gV5` zllkAh|CxbEWh3Q`3YdRvN{E}-r_PD+0+vV~tQtw2Tbhr>aaxf>&q&ND0j043PYLvt z4o7+3GmBu1(4u z3C$^jV=RFp3cEU!eYn~G`EVCT=E&P!rFu1b2Df?Rjg?<>91 z0Pawjre}@5A|lEErYn^H`;lJMC^h^pmo$L-!Qqo|#Al$zB)%Qp`%@Bye05P+#ki{d z3$LbW3HNiK@1n~M#Wn;@b0m$4IuvT@74>;GPO+OJwEn51W#L$xIuKDsD*((dFE~x> zD6w1`UzUs`iUvcp4(XK zTuTkaL*)0<3tI1zz?i`41SW)GM6dGEia&K!^j$>1tWbvUmhisteEnH*_rLn(PLn!O zF25JJ7fRym9rB@~R24;fkzNvt3W_voDqTwGAYEEOlq!OBLJ7Tu zj)W3Q?#}z3^PO|<`tJGjuC>>qB!txo9aM}^zXCa!8hCaA?tFX%VTFP|roZMJ@eEC@JJ|-l_##j0kdwi?Q8pVbgpqO| z;$r4$!d7x!(^KE7)aeqEL~sM=^s43ohF`U5~~{5t+p}KYj9s2 zVsi{pda`8mD|?cY-fn*kzy(Z`NP}AdHCgj8A@el%q#1kC?ApogyW#4Ykos5Zi5PlM zYz*7{^-BTTx%hma)DG9$zafW>!X)o1 zsh$C=*o->|CP0P9u9#Fa!x)j9G!7=O^-P#Kp%zGu96H-askp$N2xSI&9>(}K<(vSR z+%Yals3`hu5OU*ay<}-1 z$@HD{6OUDVNsMU=o67~w#$h1#wUycdn}t+>7;G?!2JX^%@2Hb_#9OypX7^uX;8gjW zlrw{ROe$TxJwW-%MS_7*V7VwrnxVEsssrEbp?^Leh-lkcXHg6N*oIoYR)ZnoR9xX$ zu-56a+xau!y2j~JFv}UhMce}XGBH~u$ptzt7A_AUC-McTD)x1N1Z7`s5--pB_J5)sf zQJhvGXc1Qljj}4kL^_+JJBrMP&BPz*BlbSFGt#`l`<6)dczb=yX~onvElIdoHXSx^ zPm>Vvs|-pzt=?_)(GxQFC#E{28h=Q*{)St23S3B~s$T#e z(PXs1eln$-+>Q)K+ROg=)bI(jVg0cj-UPmkJ(9xkdM6rhfgafeTm$toh_k$>;ax8% zzi~?LtGGq*-3oXCs7!omoZfUDle0=SucH`sno0n`D>lU9-S|uZ(8xJfj`kah#8+UC zqOs%BsrA5guhsg`ht}%EXoD{@tqD;^HCrK$mo{Hkn2!ObIT#8(%kzV25^ASxuG!nl zZ|;=@6M!T_?oc5Irk7>v92nwlX<%2aO+{7JS`)BM6hjVnZOfq{Y2`snV$mV0>X5gf zP6!1NP246r!Tlq1j#WQr0M3sRRm5hR`PO&nqVU+3&O{@G&aqM_W~yXU=q%b!0WA6!@MFc?ovD zbznN+(K3P0mG*Jo0Alk$hwOSN=@F3R52880-)z}*adS$=fauTLkw=WFNS}n%sYQZf zcpq76m&)0izAuR(Fxj-47~E1{_du|~`6^pY|HInMEPdEBK<-ZHlM(=*IIIhSk=x7I z;yUA3Iw(0NT8V9ph>Ae$$;{Xc!va0~4|>{5Q?#BIUBuJJ2Qg zd%%a=(zT@3bNs{6(z*&M^4c~=OzVc#gT(!!SAk{`=h!8&jdBuL!e|{vQs~!3=~+eW zG5n-&tIfQ>fe8t}vj8I3`gu(_uS`MDRZQ-)1#__f3!iSq4-ASpupdpy2qTyV<6WOT zfMhesYu7*>cr4^7`JTw37(rQG~>ezHhxId}JpD(0x-R-c-Wx z!9Z@pS?%p{XlNwi9mIx?A21Y3v6#*OGWZ!tsQ>BLwufOVN5C^E0iUD+#bpWiTp8gQ zD^x~+V_ z(K)Ui5Q93N3wnX=3slqFA!cQcI=rK3B*BqKx1m6EubsuzEPJ~P%h)0o1$Z4&BPW=G zF_Tv~1B`Q#5YLo&H_ecT$jZer1COv{&5wK5tr7pS>X8>f~er<=FD%Ma2L z$Rhd)@Eb2pEw&og%Uoo$SHPS!gR^84Uk*qZT?xN$I$`IGGK_L#VIH6q{wgMo!}?C; zhDN{!R9?rO{~QQ;`FM6vjA7Tt))b(VjuC4bqa~DBVFAy>h{^d-!{2~I7jX;57+|0X zb`fhX^9c+;*MNpmOxPWm@iqa&b#j9MEH>q?Z=jLRRZb6W8s^`MVa8G1@ATkH`;KWY z^r2za?2^J2)`iz>6ko5CDr0|NX1Y7kBoUy4{CGg3yYBd`YKCq8lJe`ciT zFTM1@kC@h(KAaB+>5R)=8|botA8FQWokGyw5~o$Rn-J}RXD`TaQY3)F1KLAn4QIn% z>`gtU&iL_^dYRR)xO{%iBb+eu@Q&^0Iw}#$vLDZe0q5G%W*58C-b?qWVysoXs9)+8 zhp-E%pz7uW>K>>x0f5Oyl2D3Xtjp|ehPnE>5uIAhXu#D#`zcnz%>=zud*Vjs?UvDp zq>)x}Gd{==09~T^DzqV{;Q3?D?22S2OK*SzR2!nNs88bZ z)FrH0hOvc5(-FdFD`)v3G0o5VB^t~=Oo$2vD%>>9s^K05k0PB368_GCnJP2H-^^EX+P`XGc#X_gdf&R1j43&3)h%_Bgw14=ugjLzht zRT-t&G#O#-xSZQ;f@XHV9m)M4d|z<1y*>CTslzeZgGc5uv^V}RpmxqkxZmvK@|nqi zKH-qMi}i}x=8|E^(B*sGZnRA2){P-#moKlrG=M!b?ENB9kr>d|=zCwGc;vNio8xk4 zE&oXu?QZ9mKyFaoCmCPmn|aL}UUhLxdv(yK5jXA(kJqw6=!6s$jM;BG-T%Jcm@fJp zCh^bs2Et}(n{30M^zErlz!Xi^`GDIRp*R~t2^5v|AV26S5M<@6hd{tQZ2`zn)qPH< zS$x~fpthjZ7kHmA+6v3{=uG&n%ehCpvB)M54jJYJSuWKIz(1)>OzVpH1;j+^U*q#R zF$zV2fQR?YsN0UASE1XOtk^rnFmmuR&Bbc?E3xJU?(K0&q?^+!y1t9F0JtczUXFAl zE()nkoGsa$Kc%}&+BV4W>^02&lOV-WuV{t5?ClO~*Hw+{b%K|FUH-&o8~U?KaVIOB z!Lj@jle_e~xdUxv92I}~Q2TFY_rKP?+erfU4gVxluT&AMim$KL}Q;jn)n`*rM=F$0qQ zk-~v7O?F!~ssskM|dW2_Z7QxZgo%djtQR#wbw7IQhA%-;4t zHl4xLZIPpc=jd#ntB1-AXGC(ALsYHnvA(zxr83bd&!1wwMsaNh5y8QEZ*zUG9^VZ2t(TXS?;Znh3`>B@e#tS>iD5 zeq^SC{@q^`{#WN<>Gp|C+co;LdmlS6l{SYxIa{THr`Bj=>pxjPp9;J_{z||n_|!}% zR!=M5L+`C334PZx2U$Kmj3BMl)1ADxi5Xg#)u)h{xeIbO!rBd*5D4TPt- zF;2FmZl@Zi^^QBpUTj%l=;6Pz4JVowRw9+}iZL2Ub2nZ6UZ`48Ty+H!JXw_H}+wXEr4Zb9W+-zzyid8kjTnJ0JY53vIbny!6$d3Vl+IY z0&3kB)z$&p>TaMr!QIB%EtlROeOAJ(H#aInnHJg0X1?;db4K#<#TRj!;CrsR&sQ)_ zrCfe@&pvWFK0(Z2Sb+GuyQ7OVEmk4jT*WJh_Ue)p(%-lHTFPI0?f_+D)%px)>I-}` zj`tORy#i8)+y-n+>$ipt^<91mOOxNUTj7hSu?F0CFLS-T=jqA!ZSX8oqcIJYcD~P! z+-08z(38_#szRCmqhhl1!F%os7#kwqrp}Y5c0Q3$Fp26WZn!j{PmG$sa`GpGREiha zfc6UcA?9O*+>*O4-R=nX%`?&8zD+f{&6PiH9$K?!2c;P@obXc_x4zvgrjuvAZx>Ok z@KeAv=|@%AGr9bpkb|${`_%+R4wR6XC5~^GN3x(Y<{`?S)6nG7~fDb)YTzGfu z(+nh@NwXht3TYHlIqudQ9f`r+n@K94MrljBptwiEH|W}Bj8@Rcsb~iWz7I>qn;^mW zik%jmFV08eOPDc(voDLon^q%gQ|nHQXGm$KMeA>pW5uAzH%<}gtDl-N8UD*3Hw{}Q zls)G2)9b*Yq4w$CChLs2JOsvi0z^YAJeaXpo+af_NHvj=)KCsuTxY!w7_bH%0-NW& zvvT6NLjyoxuM5zB=t9I>%VLGd!f0ePU2`TLitrH6(9tlQ>f0K91|Sm`xl zgKh%0hLg>NQjLprOl2A`tNxN~-XGjj<0Tjj!H$P~8x>GZZ%u4JH~f}pd+znIq&lT~ z0*ue$>?w;O;6Gc94(8fp?x|>8{_-`fs5@`}BYl;D06L(iA}Xe9!O7<2;W`$hsj`zg zE7$PZGAh#YdwEdq?{b?7|)l$LJ-H|W_ z{Bcf_?d_q5j3vg$OO#3MGk-`DRHqw&+3Sp5tXfdFuJz#2n8sAs!CU{$W^FcvFaBf! zFr}O!VE>i~`1H4y%*iAc`sTzU5qo*ob~>0W3St5jz%?kM$y8UE|)D~}Ce`{!hN%ffEsOMaC}TAGIeGGN^Ti|H_2=LP)C zeJyUaosdcroOC3=2_V;fuR06i!wJIFLsYW?)8RFOB3pRQbuy}Bp{zcgFdsP0b>?yY zxMGT6OI|>^0wZ=26ilPdazo(Cg|C7kKdT~p{C`bG0HP6F#^nLVbR7m%Rx(qSpSys|iar88vqcNxVXlTX-j z*TgS_C1&kH=P<`am?{HD6M*}6T^eS<^~5(%PH!F1*;U$RD(Dy}K(1V#TQ^cTzN;3J zTr0=|^WgF!_oB!FEg4yLVB##9jWS645DT)5hQh{!ycc#k(M)6%<%@dH!c4%W@r`>a zuE&8NVtu?F8125L0yeS9Y@YC}_gP;>+KPMjK8-cyGtcj_2MM-nyBovgZ2&sX?dg@y`=iqb;AImtb`y?1U|fCN9;Ht#qsCVNE$4+hzU;Moat*e}}?SH;@RHM_#bMtxt-N+*t^S*CU)%0Kgvr(xk zzxQb*mXP%%Bgtff{Gii#t%{03GwFD7karxs`F!I;WtMG|JRm~$3I02ZmSQAXEiU2U zsqR{_f*7N}Nv7fTTD^&yScFzSt|aqk)x^fDAGc)gW?o}FOA5v-ef(EBv;=(Qe9Cm5 z9o4Vg)Pr6K+@(JbsTZzr*pE|Va5znJ8vT>x>asJ>{^#~__o+DaQfAjHo1%~$Ovw@= z1?_^HS2Mmb&!0Ku=67#tsZNJ7f-L9}T4(4mDP`{{Mt|e@9NdzE_8h*2mlsDhEt2;r zFmAm-+b5Ym#5=1HZ`P-jmz+u-i-{GiH?4?mBhp{v<7)$oBZ_U{$P;?r8;NhKk6Vte zL6vRBf$*}xq9n$xc}lFg$Iax-vZn~qBdoWk zAG)u+aD2Bo6R~vPrGRT|`*2LeyA`%@vh(stBfwdp319fgytw5|8sBS76pIZkx>r%4 z*Bz^D$GJPV@%d&o{I9+VGE2S>bP=HegFgwQ=OvhXuGPm^oR&I_HIn<_++UGX+V95?!|x_k0W3w7!7z^R8M z;krf_7N0X0R<(+--ir{vl=fW&pO$qiMd4zbzl1#X)k~)WSEzpHw&23QYkR2?*P(J9+U>ykzCu zd)3+KCBA_NfrD1wBfs|RDExB;-0*`unD~t#Q*xuK*>hUR=P`_#1)XRrd!o64eW=ae z10R@Mq%ta%b@&y+F;JTP>0zBxb&mA=q>vkxMMFQCILqE+J+rLs?#*B>`%9@bHmzN6 zPP>$*k(_~gvgkstI=#x$TxM(Lq9?T^vU(RewhhX05RzuuK_LgX5$@c|m$R-vhpRmB zIebp(f{+2$dY>}(xIaNiz9<2z4pba|-;^GaaSns_Z&k2*y>l&MZ}!n&J2gOs zN*Ezs^o8=~E-RniF=wvwN``l-J&Y7o|3WYg$@RoVID`D_wWe3&_L2~I7#n;b0c~r7 z7t}g4J9(c%_^s?;@x;p*Y1ldFvMF0_labQmGZ33A#ngW%fw0<$;}2HYaPtR@(h?;G z5q2dP|3IV;I&|VY82kO6))pi06gGhcD%rhsS_^6g#JmmIQchO|zH$sHX+iB~1B2B* zcFtP&I|v;YCbZ~Ej-dW^EKz&==vh>lD4|tlSM50Y^hzt$92M$l8VtRvQXB@#a}r36 zlIpFS=iQrgKBz8C((iHeLetxwuI~*!J01$&(Rb-l1}fj57~jgy_U}%gc2FhCsKgTIX3Gtc-A5W`cy~oc_ErZ@jiC@rHQ(XEEdpIOSP|Ik@wF)D)$~r#f4VR-`}Ml*)rx0A|smdP(W+wcz);qBDpvv zeRTWT$wO$b7wOwXN+p0FxQqhx;r_|i9<=TZT&`rpJRx{9zk z0YiWgYQZ~naE%K8#HNUQsW^iWyA4;`{SKU@_xKW!NWGAI`{sKiV>LP+3j4;qW}t(R znXV{6l?M}(0e4VS`czrODW7Nsv1!UieBK(A*fLOF&fA3g^jid%cxW3cgX`w{$2pE2 zR5ll&{$IejEgq!=`$d<{EE$1yPR#4z)W_{>0>=pngu)c1q(U6(9B0}z&Ht-RHDAuS zUo3UMx3=MQT4g+|gWNl7g|NZT@rAO6Clon)FdFji5n-AMXR8-pTD<2ss^L3g{xP?3 zD$i~2J^<4)xqdjn?vpSxM_rRt115~}z91!rc&&eq;w0XTFkD)cqk+x~^Z7BcYtz4E zp4V*!*^hNQatXv_w9@uBn983@k##jciQ}=44mZiAzVx8=%7Zqd^_y?x#oWE;$*$l> z9y0Qi8b%&|@*X{Jnl4W*-L9h0mrK<6=^>MpP$fhkgQ^TDi={1#zfNUp(Wu{06CQAM zk2h_VpQ1*@uTS{>vyH#xpL$B!vv~-wGS8`%!N)6|`aDjw72ys^u}@0grH%0~RTr~- zT}~G);8xN<@1oh89T=kQH@d;@@8m0NwaZm`iHRM-`(7Vo`Ee>rNnNvmPWn#c06E&< zga;?ApPNo7R16+CzzHjFnm*;A=q}pOrU;O@q8l)nJ~E>=K)*=AfBF7`|K^yAPKCCe zcfJ6o)fe~j<)J9gnZA4t_s4{f(ns*D4~Qzuf|T0ggej)kFpecL#-)rq6h+0Ra7BM+ zhp<*7DF+&vTqlBp@|_?)?rzoKHGO8?Fv^hx_o3q5zJPjVqcYET@O~^;4 z5W8}sfAYPK#lj%x`^JgbuOD09r3NHrJHiq~PRp+6RhI@hG?Hah9C@$Y&aR1;k(EHx1WI}wfW zG`AcoThw;z{{Dd^hxhkYQTk<=3}ngRYLKM^rr9gqBGrhPhW`Q|DIo@S1Ys@p5E?Er zqGbggdJ-F?fVhE<#ukbP-DqJ)CZd%tw2WB*66tJmFQfKKR*nLKOoMol_osKX!rmPn z1ryOmjow{v^&_l2XIY4p>4 z_HABv5;or7Sr%GXX@kY5zX{QQ{S_v|o^>rmAHvH(uBB@2*W#~0Q%~OZoOI;NX`^Y4 zVXZT*G~snQ4P^|<+D*p50>coWgNNg19HjPqb)qaidI+PRD}PclXg2<4bb0;c+X!%K zdgx&J2JzSJ;HTfj*2f6+A5Sc>eBn5(T~-FB3+wf**nB#Qz5VI$c0R0nhu5l}TY#ai zMW%twnHdnpTsF+zM1E&rETD2l8r~)6sAx-y>2a`vrY26ynm}<}!0W$G`G;a^UtWoI z>mFG}?O@7o0OpPjz!ufakX*8TvL?knfE`y3{*NKu)4h_m(%$LfcJ$D9*^ zX#z0@S4UidQP)Ng_!~#imi*n3ipCQs7o*|o-XwP?lCE5Ic9j+~f5q5 zRzWL-b*oBq_;s!G{S$FscF*tyF@yZDNx2hhE`J@vIK~wNp0r_1&2{De7g^1&{$p0Y zF8!uE?_0%TJZa9}ORBw-x46{S_*V$_+#J>BA%Dx77M`ndBuqKs;nyAdb1^SnZYYDJM|C<$H(jnt z+wOaqy?if5uQi5_#-vTwPSgq%)xam+#qb5@+8!dhG%y+|8Y6#)=FNkr!xsikW%u5{ zCez+jv&Q9fkLV(_!GcU5eEdMIM068=vba{Fc-}osjl`aZATZ7*Jr!s8N~EH-yz>zT*F4fN%hJw1h&W$ z7KPwXU*k{DxTbH%^R3i}GJ#(C^AqPDl8qci1fI`ZO!|S!U>T6B8^E5cUE<#7-0YND zRrKo1bLk$}@grpD`{OtBPWRbe)Ua`0th2pP$d+zrgy7Z5SNorHx^5lVq!y(@1^bI^ z(7a!@9`+PPuezdnN5Wo8$Fo%GrwSaNj_PNg2{aBmczgbCBXDwTzTdtw%zhtL-V?hy zfLzRuPAm>tth(~KZ`1mEV7ba;_L`KY>MsNn`68l7F%}o-%@Z!;`Q;79+?r5xI&6CN zKjIruYoM2-f$_;T(_UJ{{`p8~8R6fTZJejg1(LDyU5j0&y;;+}>|M|)VSax{M$%!Y z^z7Fm7bSXt#`5$rv?k_B&}sFI2<{JGS=XV@zw z^;}=+L)NuZ2+}OQ3oO+-wv3vWtD%!GjPNmlBYZp>F)X}o5gi))e3`VDzW>nzr8R+<56F(e zSR=2|9WNwLfL(r>bd8OS-450bj0c!U#5UM_8ffQ>DAjZL#87N?mGm6Svi;o9HldnM z*?o?E*WkxVVS?xCA?8iYeCTj9Zks3g=eC-az#;A3c1DWnqD*YM@zO;6b zh_~2Wtz?7PczPKdU~af6;N&YIb+l8-nE_p~zmm4rC^6xEh9-MCQd?CmbBECCF9?)A zt&*3h!vw_BH07=hr!i+w`*zRl@w`W;oe=~}>pm>h#fo~HD(QN^an@z6u~n5l{juq4 zqV)3<(F7kusi%)*SNY5d(THD_hOdeuS{`i}UMJk;Df6FtOsYKB{)6tSxb_7D3ro{6 zWrn7WkY8oxM`Q*cteqdl66=#{C_+W)9mEt@(h^Vz zTuIQ@oVsWrvoP0VxZsL;s`EMowx?Dy{z~-ug~dMwTFOO%z>)L>#=0VPg;}SI0+i$_ z&@9)3Q%@dzf1U_|fygKP;lUs?68hE2rLc_EUHt;ZGRQLE0&|{3yjYJ&d3Hj{FRtK4 ze$+z*#<jz;=UATE&54Wr^7-peg!ReyE5t77i3;13ImQmx&8Gm{W{u2#;o7Z*|8ui=ma z=Z}s@?psVkY#5aH2ez1F{nXb3d#pWClwti-rnyO;UAh(nEBtjQf>AOLiy>`}?B7`* zU)+swe?~_-r*;Lb!JPuGlbLYzloixPS$4m76{|Yeb(@k^2h$8XeOI>1K{>}Y@c;^L zbz>^M#T2;Jn%d8GT{qW;WF7SAFWHr+Da?Ay<4`|0&Jr^AbMt*!pIgXuV2y1EDBkWN zRITqZ59|Ww%GbiPE;DRC@oyn6N(Q;v*C9%Hku!*HZREWv3)ow`+p!y?K3z)3zeNs* z9yOpc_t(pm;7!9Ok=n45jZ?s%dvMQ$!CSo2>SNb%Fy`e&$GK-bH${MLV8)<%?v1Yj zaK(d7)qip{V9nGdm$bVb4mwUfbwbjqw<}V|v>VUG9L1lSLWQsfwimE8*jAKeS$F>J z!*|E82UoUf3g#$#qj;~oaZ#W@Zrik!HXWs&s-9gBMRxlSvH5+5d!C}2t-l9 zLTa_V4S<#Rxv%Nc+Q0NCdD|*Y5}reGespN}tMjmk>E4ZDDH7vBfz3uFTFWr?VIjite4+_$mJAH9w* z5Y2!q>&GcEy|w-YQPsv({dx^#LW+G&U^nIvV@CW|TEHlOV*SmIP!_iI2lpeQ|9NXpxe zT?yqHT+e=ZtpNTtPwP4+JlzbWaNX7fJGzX~|2iIa(_b{TZeOm@hu*Bhvq(4;Ec zk)T5HJDjl+w{lU)jjBcWUv)%eY02MaFi5yq~fjStZ6Ku-u7nwxXBugjHN_pzQg z@vGNYLLx!9SZ0uF_)T=2tbDYGhqu&3oYYxdOka<4No{eZgUu64<)`c2iF#iSa{C?1 z--~eDrLKMPd@gl1OmwDFd~3r+Inf&{87FX(Z>3yZR^@C-id|5%^LwIyA6MUZe+@>+ zg+?uOew@ETCkp%S{#GuHb@s>F!bJgBx#F%2k=YH;yXznrL*ORa3%*B-V!;f}`{3m9 zo1GJXIPC|jpv-z1Fx#U(Y}!wLD)~>i*6sy$f$u?{Gh>u#`hx5$hVmg*Z*Ax9HCDf= z9gytVGE^SX368s%@f(`9Q8;iRD_X8Q%^?(;hH%NJw?d#8_tc!-Yhmsz?_smcO2VeR zI5q3YSb*`;g9?BozMcB~M4QCcOV9bN4(=J{TJ*&)hqrKeomeM-W!S!z7D&jHO(J z^^R9NH>kLOU;hKi;u3k|rOw}@MHu9GAF*%sQ5C7y)2-EVc2J70QM9Wsswj^hZyFiQ_X{gc%gzj6Yg9TnxMGZ=%bPze-i(r z?|BA3O`j+xrSd5@lAZa(y#JE%fxp3WLvq;pSyaXk{J7D|%Nv~QppBguhrnj6Rvu4+ zz!XQx78-FLFH+HcD(q{;#9u`17;JgWy>sqKzTlrwtUI%}@4%8R^L@fUKP)NH_@qDl zWbG19wHJu$l_}e>XlxNDcNO1*bHDre93emE%27?6Mbq+?WKppzJx(7kOt%q*Zv_ix z&Qq=BM-O8h52t1J9FFrIPBzI6H_ z@1V=|wsNu+z1?zBt6)U@%??hRe*#*0uL17tb41;|*X(xjPsW%HD8VEPC0(f^r})&K zlHj9Zg%@e)=%A)}LbaNbpabrasH8JS9?)eL5PzHd>sateZa$9&cV1t8_wFvT_$eT5k*#X|mGodj*}bq4ut(7?D`TnJjC2ENZBEak z;%1xmvGp{**+29f#)e~O!8ss{(+(WKzX&L^S$HQrx+)_PIo(=Vure)KqVJgRe98 ziHqqKlqqJNT)p5Fe2w+%XFX=xv_XGRXe|r%wGT1vEtfBge4iWfNpo5jLKA%P zrl@{HXj&TTF!;IbrWW=z`iKb^1uOox z%gcBFLfDVfuAbNi$%Hb?Wpyeo-`o^D5r|c)1Y6C{A%TN$7YWP;F! zmCpLlEt%J)T0S&^gh)DJYa|sQb{1~dTJ(Ukgl1(2}9-?nJ<{qE46|326o$=}PN4D(VQV35D8 zuXy43GEmu``a)lL&B?9aK**bAR;_L|y~#b3B51g8`H|4ucn)PC(9AiHnp`KmO_i{^ zPO0#=e;(q_r451g6YzUnY4n+ElX$ZvXk0ygPn+dIV@i-6GX=|L^C18rJ6r=wzk0auBSl&imJBH#tP zH@D#7Na*@oH54$GA<&;`SEL1jgG6H+PAteOJWW&veQONc+z8e9OY7I4hAZ{+bm@1pb$0=GUfi#lSt{^#%9x>iw$s;eH?#U#Rcm zPkTX7|LI#mDdM+yz(s3=iQ@ztCxr^`?^cCv59@ zpLj*CCML@NnRm8!gVyyS~!W6%Zd@AN1S#qta9AC;ZEacLiBHn6QR=ZX{EkOr% zm^yT&COc5RJB(dSMVik|Fl_%>wpZ^hOor>oukD+V4~uGLtkikI)rOf_=8XJzl_LZX zudFqHf>iORhcF08y|uG?NsB19y+ugOU!%BaZ<>e@Y6TY5O`&r=;C4H4ReXc<(( zs2?_H%AS_ZMBZ97A&Oc&WmZKSN`aU>?*-s-+2#ce7G%F)QwTGU7Sr!tmr~45;utP> zF53o_b2i3GaZx`-znO7Qz&4FhbeF4>up^-*FiyxUQ^bD9i?Faf+0ilf(4UEV2GyRu zvC|w?HgNTl^p-s%*CjsAHQ@xWF6Vo^DUSsf4Hx3}|BEKmW^y6Z)K&v5PjhD;wae)d zhUjFWeph5Ta%2!Bi{M3jy(M;~q}1y9$y^G|q2V=#p8nYTbKDjr1FcYHOVt5?3Lz8c z$D$JVYXzT8I*p2m)+6~E^_JmrfGTj z9O+`VIlW@N(iSY)6dEGH6|r9}gtTnLPcq4>LxXuzE++cl2b7EN8T)WTmLi054ROmu zueSFU(;au0(>yjHh9^oI5uUr!# z2Xm$lRgyXH?i(!>(pdJmu-wylRy45Bb-fBhgB{avqYyDDvWNqz%w2q}H(vHE+#bNy&NGrfL{j;sUn2xM$Ry2($2Pu}T9HmjaOr^cqol->DOxrMI#x|kaiN$}@%w5hfKQvXOPZhfB|t=N`r!|s3&2X&ja%7#V1BRBW^d9%~Y&99xa zAweJdHSY=Pt40)pXhFLuQm=nJp;*h=E%$7F+%f6`W~>TsWROoXHk=kgr49N-7F&SL z%P#}j^YXM_kmf|5N!9C-y@LoTO(^Z$0{^T(`V(XrDm`epj3+}{4Wx#AoOj*{qeua* z&&_pYom14MB`&=vggy78vZO{&TCi0=>Grr77PbodN@;DtM31xysLD$F!AdvJeD9hq z4g1aOZqSCl8Mtjb_YtpYIAbBp&L|yD$Cp4Tjd5z@E0DED`SB}Vh(XYf;mlL-gNNpP z$xJ2V;BlK*R4C;hrlQxbO?2xA;F_6RXQ(%ZGHa-~4TajGyWZ>h(G zuFIs8mo{NJ6G_E7IXmhC4F+ziD8q-L@);{PJS020YI>;R==S0}E^Ax3zvXs;==alk zJ=pS($-p|AUhS;MU9az2GkYR9T6+Oy89c0emdKpF_6Z}2yQMk$i&FkM`*$gdCwH%M zIX)HF`e<0bx!m2<#ov}i(YJV8-z3;5@Ul)w$dHMh$r@bGiN{1Bx<<7|cHmL4yW>O+ zcW}9n+G(G_r;|8Fj#Vd(D>2=T6|09ycL=!fYPC@#Me2=8gV`F0bsL`nJZpC6(*Ck+ z=ril``$BK;ng$GNdi96<{Chuv?+r8ca6dK z*K#y%x7lnR>~Bc3$9Mazp6sa*PyeJkk7TBS8EGBWXJ}88q2|eDQebDx?g#02DE^cC zR$<)gE2i?6@!6BH&w3@NLa?Vbg-l0Fr_wF+A8bYsd5>T2G5ggQ4B zc*M=UdQ`s4XgaaZMmB{|DT2#)o1ZA{JmOSY@pUoH(!!`V;O3w{G%CUcap{jIYd*@D zm$md7o)0EpIFGgZFyiBU^(40i(_59%uJ@jPLS(2Zx+TNnZSVjChIY}k$(~e8@E~cC z0rA63131rtrS)n2O2C3ezv8Un_Eu4?yX(Bz7V0!jhUfC-dm7sJ=)EWo`DMkl4IYYl zICkC`IbO6Cr4%wrb)a+k_9zo0X_wDUIMjDQ8ff<8DT&LE2ajXtM zVIGQv1E{-h60dG+e@K7(m(rh~HuXWs4i0;-m9!V|ztO{<-G2tvO8A;Oy7Gz|eIDr- z>6vimZ3oYD@(Gg}eC-E_KPD08_!os}EO>|pSL_uRb)GC!K=s))Bx{I@RJN?8yh>9p zQ5}8yHx-N|Z-4mKwCQsoz?Q%?8Ny{hrUxDo!)K%L#YiU;r-HkB>Lf`NVv61?uKrt zXOUVol`|mgYW7bMC&h8ooi|5nr%vP?hVZ_XTVJW%3|nquKVeNdhn{7hIx>}XYL#TY z^WWjJ(Z68;#&@R@8-%l%6OC`AXC*e6W-(>5F;?&lzZ z`#|{4ZDZm?S}6?%IX;3lIdJ0mQQQ8$;(Lzf5;A@1^dhH8TDiyM_GJ%mwfZ{pqG;#8 z?;vKr-CWlbkdwhq+KL^O)C)G;>jY&*PgE8))(*WU^c$6Vi$L7)PXrP*TEd7;qgG^@ z4TYT*uDHB&8L~jmH+?NXuB}&X=w-gJJ4Ec8kC z?_{k&3akD;Y16)sab|J1INPnM;jH4?z9Uq`Lx30g6(C>!L2OCf{IOOFvNL;puE$nVPZG?H&| z;&ZVNJv*qEZ!RLVuBu|#$C6<+fVnOtq-vOKFwX%W?{}L`g?vt*>vT-& z`_U5f^_|7}AaUAD`vgvu03@e-+JP}rhdi7eFrxJRMkC}Ul;QW7?zz`Di1Y{OMVB8} zsx|Op1Da7@f9D*ep0uqB(O;aW#0UznI^S=lBbLdD^po=@_sc9t8AHLLycE*;dk-j{ z;Yzt$kt-^9?eAX=QX;qCl0s-G45lvimw~I3S_PjnE#@)+yWwBMstTdjVF!1u7YE%I zZVoGFB%qvV**|Y~HDN5cm%F(ZuerFCyDrLr_YBYoG68O;=fl{GlPlRWiUg&28LQkw zuyiZ)@7p`#%^!>G7t4sVeM(}@A_27$p+fDOG^@ov*-_`6#&ac2(T_(hOhnq7^3?(`TUxvbbxQfj41vn5XPTIMm%tH}D}u2z)p1vB3Jg&1>Df zsz$I)1u@So@EksGi}D(wi4EfgQ;Q~qbdib&F6}$^Dkl>MwO2gt1gfVWcqdA%O~N*7 zW&(JC+~ZWq1X)c_>Tn5$m;s~5Pe+3d)33v;oJ9OR*m@_D8HGwI6!Ke}gFDLxQy;u8 z#EaSoXBb+>)k!C^Y)sHT_^4GvKZl4p8kRoD-ZFf82?iG)WH6dnoo8i6YI$LU@nzS`W+=;$(A7 z8Q4s3o{yjjrqiUI{TATWwfp=}qOLfUe@o2zM0}oWugS(uHkkPkRUtH}`xZOpQIC8cW#fG{*E%MYRHX>*zZKVb5^F;rF!)|gUaFXsrB(?*;{npvG$;s zMU&$2$83xD`;Bb*KGNJ6e70vr-(?>NL)#2~EOb%5aMpT5iPp+O@DIhu`0$m)NB3rK z985mI`1dp)so_10GwS(*l4ftZ%~iB{k8;Xc;$O#*s3_n;|57QV*|MzK`8yE>r|w!9 zI}6x(4-8w=s81djMoilpEAQOdYmRP_8AhBg@S?L35L<&rSH92O|%_xZFGN>UVezKNCA=l3Kmp=htD*l?Fuv z@(Q=he2Hg=Zk%tx^gU_{_1c`TV*&1{3B!_yKg(N%q42`mX2@pC&m^b7N=6{Rwjlox zK@8&yTF-xez zf`S48X#uG!NR9MPK!O5_bPyFnhtNTKFBS+OEr>|yCDNpY9@=lm=Q-!h`~3@EW)DNi z3=`Jgd+oLE`?{{r!vT1L5GL_=JZ%Cr!D=>SAlv2Phhxo@pru-hJBA!`bNGcC&jKsV z2eN~`HIR`#$<@ew^phx&BYw-;b3@4f9iDh83>hOomO5;dK|C=uMP0d|w>ZnysISJ_ zK-EX;GxRihV(B8^t>VoWGV2sJ1K~>OGGE)zN*_{{Rx?&rjV9i!NMX*M8GJB)XzX)u zb%eFPm|S@je?>O8btnN98u(=<6fYEtIb91qt(draNKbW5d+2pgbY)^lHRY;_jQ(-a z%_q9?nYOAp)AkT60u)|hEfumIJ%Kt`4vkct%+9xx7F~u?Dy;i=kWzT-2hX!uRL;>@ zZy`2-y4nWM@vrW>*-oO7f`cZH-LZfSTT&3LrXGI;M&WlN#pOGqt@|q0XJ%D~Z4XSOHhLoi` zOj%ft8}LLAr5+h(BB_0n>H-@JR@scFik0(fFJ?GVKcMI2#_CxZAJ5&`;xOtf?=PNm z=drs{%Y7Vp|J@GORX6S(rpsu6n-cVh-TFO2pdI2<@`Iy<=C*e(*+IE0YQ5$)Q4 zZPXJj!GQa>xi#e1O3dLZV0^cbPQHmL{7^D16T!_ z?b1LK;5@$fmhaA}R_o%`s~ z%`W`X&ZO}Y?*}#0+7fqNQ@C!UgN0h%V^fC{ccy6$NIj-fk@vEo$+IMsB$*#pQH}i zuuoVGpbLl`XE-np(r=yIYy`yVE6tM&Az?+3)8lZ}!|;uAzgCxt+GfB`?RNk?60py0 z4puZ~&i3mkVr?q|R+7RNA%1N~{cbXedXLIOhE|nm`|_qCN-eqh%&xbFOS%$bSXoO=LEH?eYGH~iB=d>`vlQP<-;EE1*zi_E`+ zT`Z8OQ7wPKI71XGn6`@sjXw1^;cm76dd+q{JC=gm+aM^Y+u)4wu0x44s4{_x3vaa?q~lY+iyGDtB~bkk`k8NDv)h;TJVoNZIlX9 zI1Xw*2+}K;iJTyTEnJrmSa@S}0nGeB2fmdA#2ya2ye8TR^?whoJlqb)-U&^^Y7sBn zTJ=&-;uyO!x%al@7;eR>-S`?qBHjEAblxwRN;tr(FP(d8=9_<|c5gWyoGagykMPNv z5@7CAhEy^CsgKHIipo(Tv;GlR?W~pIlCbb5p2>e9%u`oL3yeHT3ZN{$pacFzcF z#2HWraoZ^k0YEm&n|=y0X#VcUr;%IDC(F0ceO^H2JS9Y=!}LS&&(>-?fzVD{M1OFT z7Ejc*ih-fQdOt^1X$|+`pmlY6-#-c|ZV4e3GP~_WYu1!9um)jEb~q?+ck?N+sGZqU ztkdCec}%uzb4shO1+k|4bMbRZ;!!?187ncg**7kDQIIpBNGT-O&HJ^)dx}65FV!`1 zW`icy7yny6jCPz`(v&}yKf34uRk^AXsj9#l>~z4+j?n+i#1=VBF_Xw^(E>VWl3MoS z`SCimM@nJ_&Q?!~@5nAplJ@r%jWdjoJaK&=9{5ar(0uEAa?87LXAM}l7PNkTj!oTS zft8S!r$a%o%-Yql=Bp5?5=B21wHfa=#QNV>+gx#hwd4t6J&XT5m-iO<@NT-VWR2{bdvkbOgm zX0a3gWzYf4xRlaA2IOu{60}tC3Hu69DsKO@iL}ji?~`E~xw^*x^I7n~=Sp2N=1p&u z;_XHyAf4m6dUWqjse^J!Pv3-=dx~G%4${J+r&}zsHPQS-uZkp@qQOa)Z#SK(EsR*i zxN*HrfrB_HbQX{?YrJ?v!ZILasIC@|v`udVEIK5&4?^}zY+$h&YGOhU$6s~LiSb>R zo8U6eOb_l9B%=klfA8yVMHs38b5$)L&mwD84rK|yTz%}1dd0Y%H;xZgP&QQ3=@c)& zFR7j%y&LUV&6@>TJ+Ia7f=#_ zxqJ~w?zS^gGvK@MhKp~f$PW8}&j!^yoFbqfZ@=o?{aZl5=y=!%uFXAqK#b(TLI&^R z)&30i5tjtA%eWj8sYNd4yNxMHN=g=2`YtVZ6YazMMR9RBL94x?NXAE!GgO8A8qmFBo1zw2E7q@LAYvR;8&0kIk@P+Qtz$^^(_co*wzmu8&$ax(S zh0cYJvZLSiuO*ETE~E)-8?<>Nrbbgo-Dt5k3Oi_LylKG$Eo|>so@C^^rJS>(NkT#i z8`tg)h5V3zCt_gOyGZF7*bTxIPt1eKGoL4%cP~ktR~%BUe8A%9Lu%+~TIq6xjQI~XtPYRPT*A*1xQ@*(>Cm;_7*%tSeHRjQTUhyWJMYK_E#5D zpBlD1&;75juNdCj*9nN`wdlQF5rUm?v!_T=&#b-yon0JI+?=$&D=xeVY)-Tmj%ESb zz4d~mN0@h&luMv9`6o4X`|f~w#T0w$Ug*cl+jN7{CRld6AF9boS}LrnG}`vp=HpeF_k z)R#n+s?^*-ywxRrj2lI%L`QuD!-9`+srA0e`U!M9YVV>*HOaLzH}v=~n^Nb2Ft5(U zB9)j*$hO#+K4AOBq(SLNoOAq+rkO*gPqV&|Z7x^aJ|L*(BX^;vg#?4vH*6lt7eHAZ z&hL35*k%V1%Vw z2Rixn-yXa5874Bs=54RtRU(9dwxJ(9r?2%eZpR9D_j`59|1{j%a@%p8 zym&Fr0qH`?-Vc1Bj}s*3-7E%tmK!;6mp6b?M*GL}2OA1$yVs z*J_P{$mfNP^n<=z0zq@@i}DahaPZ-TA6W#Dh7ngAB#+^8?v%xNR$ac6{Q}T6c(}^H zXN^a{#sw>WEAjo>7_SXcW*NHXzeLPT6EvgVK{L-fQZrv9L2+?C4fF0@(+=*8!T;A&LyHzv(ng z^moY^vblR|7VqA}YFB3OCo&+1tQT>hPK8hOq-v>z#pz3o46H!J9F9dvbatRzb59tL z5lW=L(HP}u0m9`Z`ML_Uqd6Pz$u56unKe-(!`yiJI^+tF#)>UOJW8>8l<(qQfe=1z zzgT4y%4&C&PuDzQo{(7s+Drg|Wl(vv6))Q?3bJ~w{Vu7s9bX(VvQL82aAvRN%nQiG zn6za_BbDPb`;_rlZ=UwU5iBdw$Gb&$n<~q)HxLrz!k#RYMR8&`)3js=uYGIL>fNLL zD22m;R=c&31HzA=*m^bDDT{+;c1+$Fw$Qs2b4|9L=JHjoE4<4r-o}i8@&-O7#eHfW zME!OnGed!9V@JHPXn~l`246MGbUrl;Eo;I+Q=3FlD*XkX`$Px_Z8F{SYBnh1c)q^9OIvejeZYlBDU9A@$GzzX8STC z=P(wo>QC#hp8iI6ZuYDPD{V6YON!Z=6JIiW<-|yt>KgC@)pW!Vn$W5$B|Fo^&8ksGk}A z?eSb<-R&)Ff*SZ+LSxo7OKlIQr+gH!B{~KbQudtlUs0<1V;!x#%wsPje_XgaW|7j^tL&(eUmj`4bgVUd#B6*D2=qha3tp#wON&Btm! z3HwAgiyPOXSEIk(yv+G)qisNZ{&#-abQeot@!QqB%-K1I@igF`vOWE`L-R{u_7Bw^ zbvxsZOF8u8UfbW;{fH4$&9bd4dQ;{O$WcVNr9&p%_iva-(A1js7O$c~*&`0y15a z##_I?vP~WXXsya-Vi@772<1|;5yZ)<4cMS<3;=VN(TIcB89DPP&5!FA|7nAfihyh}zvgd+iqu__m6j}S~FZ;@tufQt#uC?s3nr6U<(tbLLIhs|baQF(k6wfW#VkoAcHhc@xW(i*12aw@jY zR+yeH>Ksh(mHX8ywEeM{a!Eb!q4l_R&#>(FE{qVmoE<6CoUaE`CvJ#0tdDx3GH{jc zuPhd)zpnV?ZSE$}5?S58QUN_W*j888)<2m8DE(ruy*|~WBvq%nw`C*uFo^|xj_U<8 ztI)-3a%=$wZ==)!!8$LW$jUW5Z&6>}u%fR?=Fa&gC6|Qoa-i|`Aa>R3_Xh`=AV}=0 z4s@K()A$n%-v)Wtr?y1x8_Wo+bMt2pLnY*0hh~<6a6#J z(2)qubPd>L&Tx{e!JXZzXWMLrFSWkV%_oXF78{`a_Q)}Z^(PJ|^GfK*_EO#8-_%b& zfR$LLH5m;&tXJ*AuUflKuMiGN(+&^U$Z~#zJ4=PoWn{ivIx?iiX<03wNQ6jTupI>h z)Vs{38m7Vf=IT%FuE|lN7DHARt`gE3Bje-v1q@-Km`SaUq z4gl_ikv%t~!6QusJ}^QGix{b4VwEMAc`|>)hMv}^h zW!_y|+TF~_)>F+^Tb@XePZKgEuf2z&BCma|-bkHy3r}c}cGicrWMqxlMJyHD{yuvQ z-f{gZUJ0+>1#3ZMjQoF-wX%~G#--@?yn+3cT~3z)5-NVd^COKApp@6*TPHgR_T#O$ z_yZr^55K`R2;U;LdwZXcIo&M3MG|LGPV#mJkl^Y3grQ*@OQ^kal!3k9KeF3vC}H)j#mrAN3JrWTM?(AR9!Gx zM6PM9z!K)$>$?};X+$V?zFB5t>3EYvS%^qxZRSLRL^@Rm7sk z<@m6q#zlVg%LO1*K?(IS*J6F8I{ipRu4ol!O|Rf!OyKgxmA8Z2(OP`06kH-#4t0)d zz}-9|RXYF_*_T_P9aGHmc{yIjyJA)WCU}0M+mgpenggS2A&67b`Kg^L;0od{3MY(+ z;e6x93YxkPR*PUvs#m4(mTE`DGjzx1{#Sjx5%QgV7`QBV;Q*K-Wr`1Tv z;QCJXX;6TIE&IcFC%aV_Q9D+(;IkK* zPE?Qn=i#h(o5xk+_N%;O) zM6it#HSfeXw=Ti^G-G+nwRG(*tD?#w?X$Pmg<|IUJV^ajk7mh#=vapsNd*+KGy;EB zh#Qu*e)J7@%zjdShP6qHkHk*`X=^Smah@?A;Rdjx^%Q8@9G^$&VOraUQ`&rpyHqco zKi5fQ-9xFW4ylIB$saM+LCqu(()#S@GGr=1-mc?G&OVptON^l`Y!0sq7dKC^$-J4P z(Ew5}!0){EjBjpo{DRG!j+Dt`D4&0hGg{YzPP+`}~hpj}mY|Z`wL0rtV4zzVX;=P?poTXsa3I)#z34^^*<3!FAqxU=E0oLPU2DRRL z7R9MLR#UlgL0#2^(l%|sd>uCX=pT)c6z|aS%h2|ILaMLrG4ezkg$APr_RVM>s@S_{ z?g_z4q$*_zRUxJA{i`f(K#^uSQ`#x?h%k*(5KC=xW~nk}OFQhjKa)GwbuS=s<Gn_i{omYj2`YMS!)PxGG}VBxkgKTDm)CcGy0u`mVPx7b4d+PnLhX36CvNq<-Pu; z^Y@k~qr3`I1H5HBE(Z>#PAU^Eh%l-KVI16>RxKQLpON;-Tk`%e z6Q~`hz>r9kh9|w;m);yn?9f$JCq5CmB5iUYKS%P)C@BTkkhXvqsdundL&1 zH}Z;DEhYuO(G?K-D)2eq8%TkZgIkF#ziD#3*;M#_Rrlc8UL$`p=eTjj13*_{5b{+( z`HrE9;;=cP2H5-50P-%pP0X^j-e3JTMumFF6BhmIzTI3c z<)Y+18bF-XRJZ-kp;^`Ht~J7Gb)&VFSHPV2pfwbR`qR6yEy;}xuvock7AwU-b!+{$ zk$#t2dvZ2`iBsGSRtn$CecWc3M_0o%o^bP6DAd>T_|wATHQ1?fv%nJLvsEG36;XK% zYgH1Dwg%_0z2}d2gv!lj_xv@?bIe{)LN#F+P{Q?z})Jb7Y< zHR>~TU%4$TuD6^0{k&#~5GB5BTW#^{Oc&?8OpV@*pX98nyIYn@u;*Sp#Y}yJ_Bl;M zC(TadX_@ZHGJd62jEAoM6B7c2>ps|5rN3_x9~1L_HC)3O8q9L^>sFxzn2Iptg*8+Y z${2d4c0}9vI%sGbo1k)$*Lb~v4wt>*jfBX>#FRVAa=l!tB|Uo;`xZ>)Op#iXV#Tq? zeN+^8vO-D!1s?LnV#q2$f;}U@M2@-TL!b!>cZk!w=d3l*?LlwdBlJxE_p4$)c-d)i zJ&YVBdk`Psk%ML1tOJUC^0g&86D~HeD{kur$YmD!HPlZ!(6808-V&=GuZQtj=RI#r z-z2@0Q5K`8Vn?+GFD0E-W$spzMYXnmMtLqoE<4*;6Y!5HsMwOf97qC?YCwkQTa=`u zc$}d(R{ry;yVtbZJd7WqoPrcz@!wuw%+YV4IDaiRrqWwjn$mW@eTsc6aUjF(18C8& zR?UIA9C{e8WKhbUSrjeiZz=)}(}4c)j~sFIzLVdUrZ4#Lh~CQ-}sByuvo z)f=hAbOQ*ulV!UitQSyp)7nWu<4m3vVet>Q$f$K2cbI|u6dw{d_NYU4Z%I; zL~CVWQf)@{+E!PC4hktOzHY40?3%YmfUXZcm3h0dHp8a7ydIz04i2rU?$!$eHtmNW;OtEL-i&h5XF+6`>NNZ2{jXsxL zFL1?vI9pD|MgyT2y`%a@%1rTSD_=T}B@BB6L0WUgZv3u2UbiPP26`>aW+S0J#KM=S zu{gs{`2xXyyUPX5RB9(0^-wWE>%YsQd*D!J-YZ$8?4&0ja3HGr6;Lwe=MErl;7Ciw z!+NKq;_%&aoCbp`3sZwed3nE{cMHxwsS%lB zjgu}dsT(pq@i^Lk*z?{^0^K!(`J9T@QKOBGn|T{PdTm#28uN4isSwq?SPCn9sjFQ_ zedgsW%koabk^4U)fV4lbi3q28UnI}0d5zQP*f|1;?mNY%Y*x_w({a&e@`m0hdFy>g zk)YRFW0ewCijFOxuj(m+sIr)82`wkgs9S9*v+~_~=~)!FAFsc1!Q5mT`?}?1b9z4e z*|v(&oH_h$=^k9#z`=2Xpw;JyR%n zbh&F$-Zc+T^(0rJoe87*qo2sm+?@}-2okC*VZO|ksPh0*#tdJvpze?dsZ$9&r#mY= z>x1PhC_>?FOTJS@oJ9ujxVZ>l+6a zNXg~;0L?PbBcu|eqV@$yDM33IB45I%wm!h`n<)51!y+Du7g?&bKGWjy#41ZEuOvdj z;FwOg^q@dg*=XuWMSp@1mQn;vX}k!k%F|}C$GvCgctEdBx@c01!VmFhuA-^@HscRJ zP{0af31mw)M|t@fSO^x&684%l8?_3dfNM#mg}D4v(ecnn8?F25wY@9qWf0&G!lMeB*)dGb`NopUG2L(SiIT7GE9@1{eX z>%g-h^Q;_stgxPRD&VGhBOvHrTQ8!}ls-VuN#lS|yeEQQdu+$=gvGUH^xan-{`NqQw_ELN@2~OI#n=9an$fe{IOVqBDw1Me?ix~bP z!~PRt9o#l@CK=iKm|;nsgDH(t&F*SEm@}{JnFQH?+YPQ4DVaU6A7k|WNoT#rHQ!de zrzRC7lLE!daXet84ZVEuW#&s`KRLa22i>_o$?Ga$);T|^=C|RS{aR$_FMcK%n&|dg zKOGkx7v}c%n92Lq$U}z}lj~Ms5-J`h;~YzWsy1;R69a}&bE_rXnykI}3@`7OOk)O8 zc{!=QpAo&r>0!Z#DP*OSYqD(;=M9K$!tCbG-Y6LQ4Yh|PQ3$(@H$+iSKIA>a6QgQm z#uEWfb09x=I$e9fC6b~T(T?s@vZ7=b-U8R&+i|=XYXl4L^Tm3*8#j>e{e-NV7m_H4 zJ=mq8u;8Md5~bJ9VvxT-man&RH6$5Ql55L7n0RD%RFmrK1^l)R`!YBg%)qC*{)Qsq z?Cf}R#vdG5nvOM9-hz8;$`j(-XL@>Du?DK~4CBI~VE*}`#dQ_?u`VOIgn)Qe<>bgF zM~C%%?QqaYiBdvJSDpbwxs$Y@WWHvhesq{OZf^=4;G^cHf_>xZDla z_ir6aTbAo(jIW(`zfM?PxT4V>5Fsf>amuk)-3ArHefC(!)qBg_b(j87{8B1(x>mHGrGq@cc!R}pIe@@nILuHUYO#lm4SLpy3c z@BYCLoy?JPEL71&)d{$~Zlv`^7+W+&O7^5q1{YU&i^Ur?f6@S+Zqg!=x5ka(CQYjb zxUnTyfoE>XoeeMgzO+gpq(#~Z0j5^qT8xpsfj+xcz7UWUHmd~|oOc`7IQ4n!WuKOA z7clpVy>6Rqski)7^sX?m9bs-GaIuEDl^d`agkZ*{%-{OYv4ojF-Gjz&We_Bh)>Al)O$PzVMGtbPfu7bGfk-zuR$i6c-k)1v|10cX_*t1ja62{uwf{yxZ zrv2&C!d5Dq<|w<=Hfvt#hk22ufcWMyAi16r8SjPlT`E|e;SdGv%?M*n!S{wv_SWUQ z5?d~!pyvswx*RXW_ZpVvX1ROzWBNT%_;Nqz^j5?(G4wt6_Z=2tIZtgJYM6K;bQf^} z>sN+!E7Nv+EZkG;4vqmh&?M1M;uaP`pi}I$7I)SN%*%A^4;dd%#V;4C+!d-#>wzJSrFTWUXdSU%W1+Qo% zfzs5QN^ZCPOrL$1^i(_ar`wOX0Ub`MP3Bwc$JCGPZ`LN2kd@|(XveiV$>O|K)4D@2 z3{N%puCCCNe;JLlR=6J@dkIRfu$*;m>28)RL2BwT#fp%{_l;`yt2#J-u_|M!)6{-j zztI3DeQl}bU4T8-Wva!Q;PJs5W_`{;EY?%}yXv^;{EO4emF=|6SljKY)98+i)y6s< zmmmG|j}8vC96k&qjF>3yC1ATv&p5RP*S|lQ;g49>mW&*I7Nb$ZMpG#_uYMi;e{*!^EA(7ZOw&xqs=@ux}lQVmfXihFTo<=yuh)C*ox-POXWccM}-eaLDd z*I8s8X$w56F3pu7%x=cRqNJEG`A4&yLP#Y%2dYAC5mP#7! z1~V=5sA$+QS6=%S_Z>Vc-=jB?*e^gyE2z~{wU2@H+N^8RBOtXb;}{_MTQ|QAB8gU(` zm~cY0b+=M|_4+(sYE&8FB$aP@eUBKTNpED11T7_Nkd~RqtQbGiFmBQjv3YDZz#rkn z&^LE{_h^5Uw~rzspzLCUoh7ovdNK>ZS?zv`gS!T$lp1XnxceCO+Hfj^-;S0%MaGN7 zyRP0xyy|^wYJFN))IJ40kVMff;bgy-o;YONTyx(v>@~jh4ko`sJMum<1iOP@LDeRE zM>$zt<^F zk-fo7?%5H~!9T~I8u&it3&sv3C|FaMD(BJ@af$brlZQinP0gRB1pS%L&%!LD6xUW~ z>bb;$kszENAI~tXt%j<5%c&H40}-4f+w~)8V4)LtYdSu2?TVC;8g&Op+qc}Av(p-} z)*`c*rDN3O-pFO^lriJoiP{WgO1aDX{jF&NEW_IL@)w6l=oI-5CXHHXWqeEc(pFu4 zD~#ruC!VwX*R`2I`z--okMAr69y(Y4eQk*C|>w1@9 zf{2gX2*`C?C7^Yf4*)@r!yT&sHES zN{1(4F=0x_hJe}7Ki}1#SksKCa5QJ&b47X=$N87NvD)`-4Z0eq)|KCK`PzTZk!@>b z)Q&j};&^psFVkwx)6I@{t=Hxo?#G$|N{#Q*^P&528IEwhuVmdIYI^++tnhkCIV)`= z_7|Iv<@znCpt2Jsw%Vz|x>JmD4Ul2O(!QfEP)ULjW{(AzhXa!zKg{6Fwvu6P=^E=v zvbHZ9w5Zs^I&(&&S|9-uwt8KpwAM5ggO$ibpHu2Mjt|6y1Ks~%U-ig4#Smb5ls30w z7*cpSTAZGy@2`IP1gWNIMJED>EvY#E6y*k|b|K7gb8nd*KKVoZ% zbjT?pP--7w8ZOQfS?9h~v?oBQOW+r=tDs0K`pMAq(`F*qZT%~YT*EY@_C64^PLEc{$wIWocS8E^F)OP0q3=1!lj}WJ97#p zlv99|=i`A~hKkzRdy`AtoBrMVqWUK$;2Llr1%xAWI}JejtjAE1Ri^)ZHd1iD9Z~Yl zN6x1|jRRPESJZSEr%i1mta-&$C+!$7ZjCmq68V6z(|JIJtbZIT=#Fl9?ji-qD&i(C zU&t8xeXQ^@3`VU*%!u+x_!nJV7er|}=l`?Fm8?<@!}Wu}`AIMy=NEe5*K7e$9$|jM zOQT}aRtPJZ2U%=d6A-1;EUhbvxdbAyMn-zJ3pg(k{(H;&^nnYQckTNw?LQIZE40l1 zx5Gaz=O>XzPu|V>Wpza4ImS><0#sdrLEC0hi|8v37JD4~qvs{#4*#1=ZAA*{K9Ze* zY)`-0oZMC9u9RT{VqwA#OKeV^5AvnpMF3;%=8d+lYjQ%~@**?$%?>}tbc`6oTW;WP zW*+GV3BK|)jh}Tp>2k9Sb*N;XCEMoor$4!oK_B|ZEwJ*VI>O&NtaQxMOK*V+Ofg>s zrn#nloy3Jxu5Nysi_2C|V|MDEd9%Sif~DK{Po|I8$!69!w_XBM8z*PGHI2!Ibm1s9jf)@gd_M7tXkJZ*rOQnI0J<|zzJe3nTO#?FGpwzf zcTKtjVj3$m?<|-i`<{VtpPea;az%jyQxe1U!uvY@zX4Og)n&%B73zNrRYSy@P7#T) z2y$>$!y8_2D(g?`7p0+MlUHDH0i`4@V)enq_^5%7Y+n~CwI ztp&FM{<(1`lU>=2S!xmA1c;J2X9fRf&ie1O@1GymnPnWQo0u;NS8f24<2q4+R0n;0 zMqKPX7QFFGNnCB{x1O87yFzAtj2xAVpt@DMg5RhK)Q8>DU7eovK z_sKawBdr4fm9XDm+wkST2wQ8xpo*ZWl4bxp8FY#SP6=>K(En@n13rdNgh+s|`G0~R zh;NH?`?IQhBS>2l764t52Td=t-0~ImcB(6-Mj#}!A93z#UFmcQ0vt1GiZaXqB4C^W zxS!c93u1|4xaJu`*^INj>6z_u*vQXx<3nIr+rgC>9k)yT`UEZZe>U}XxBnMA^v{R- zKi|_Xwg;d7S0eF$-u&lf8two7jsG*U|GfPF1He0Zx#^r2pWZv^1Oq#Z6mLmY{Ng0S9~_vU{ffjDVmfp7hj$82Eg{S;@eSfPmWl z^6$!wEweXp(8@tq-(Ba8gR{DgyM?udy9IEJfWQxA1#xhG;pFhf-3=seLuLaG8VJcnzZf>YVH~8&x82zpl)=zDwOR~H9b^WaA7Q~h+txBB zX#c&Ze(swyQs=hkJ41QT7o7fUxgy_c%iYu>{DS*4=~y+f)x+$R@us;XVDHzfEe5Uo zqNL8|;esynU%fS!b`BTfF)R@bmKBB#wfXrVGw_VJmBiLpnlqPVd1btr-+F@$(-Ztg zGrlq!T~N*U5l-CG{dj+6*lMfUzwJ*aX`jaw&sA(VVFW%MFCmJ4;M0GwHZdZ=qeqra z-+RJBH&Uz_FC$Mz59U4B^)CM|rFgbvc(Bp3Ie>Rl_|IZklkAz(^Tn2`tSAaRh zweww3RB{*)F_-Qu%JVY3{1;Q7uBRqAjw4_|RQPy*z2z_+D>MRg=HI2HK3e+0Q?)zL zsGRhWHq-5v-eR`j(GpzMqw-1zg{aWG|GN$4ELi^z#gt;VR^^VV_pArnWQ&?=N%?|f zn>3EY_n58YUQ}o#lcT6=(v&9#x>bCA^3hR#4%74UCl;MV^uqQ-+V~xm^u%)0J|lzf zsd}fY1INUzCsWnQZm|K#U6g7Y+kw%GZa5#Ma!CZ6xHysEBZ zDi4sca8utB@6NCr-#D_pEnN;ars0O*N09OAc~bt*xt}$2q@&S}6SYONLG6KO^F^se zPP1*%@ZxHx{cpR~0yi;BqE{`_7SB zAhtogqApLGdWe6R7nihCGRxN6ocjW0XN~0mll3fRpV8dRSP1+gh(bLtfAN7wD zc`@cvbsZSL6da-W6X$x&^NbIsy-x8%Cc(WIu#ZcyBdD{7#?zi8~;pA+iQ)24sc z9QbzM!l>8HQeyNuzpRN{+o+;eV=7#YGxHOD)e^N zJj}X@qdRQ6iKSA?(Q$w#upvz+Yp(PjB|n{f|4mFcY*&n&LG)2Jjn<^cx7NZ?J~-DfeXz@O|cwS$y|X5+=J{CaI%rzLt}?w{G_?y9y>a83~}A&X%Ls`f40 z*&O(Xj8RRg%5ZB5wqlN|47Q@__&jHmrDhHk_QmARr$TX6CsrRB_}8czx28!hbysQm z|D=Z7{3=vja<^Ed_I3N@#A}6PUdl(+8Ydvi%O8im-8+~K3d#_5v6N(g(!(6o%3Sq{ zElza)I>r}Yp~{HIf2Glkky%;>QmC2a*j?;2M~ZKvl@zWpY2JwzyWTh(CnKLMBCYdQ zIDA8_`E$Zdop3Wp_icJfZ|764`7eYNF)}IfyN@f|&yRr}gW`Rt%zpu7)1Dj`(+#=#oN6t5EiACxfBQ#BVz}Wc>iw7IK$sq1=!zU~dQys;dsxutM2Li}lZ6M1R zR`KFyu2gnawHnA!L1l^J7@k?irY=4@A!{;(fEGQI(ADsf&PW-alw z(uMB7LdD{;Bm$XguQgdc)v*fcnO5_OTx>DnaHRH+b~=31YuW?O%p7?YBC-C{GVS++ z^hXvUUMfV+ca8nGzHK$aEErR5=+}5~qu1j?76n7E-7>UILz&764VK$(6pTmJb|8a<*UN^ z+sN(FdKHEST3U$${y{))TPSi(Dvwyo^VwP6DlWB2QU|}6;gcCPKuvT8bdE!Tj2=;S zkrR0Az!kAJdvShnsMA~R|5*=pKvb9|A==jIL4J0;w}a-i;&A1ASoJcM$5BlpwL`1R zQ#8(*Q3gm(=K{{Lf%_A7=i(o$a~6GEWr#XOay-w&aLL=E;=#un;^@wRdx);4kMRYP zR*5cgqaqk{KV=w}TFGqZvWB0QSaa`WhkjJg=+5~+rJ(Sc(-!$^C#e9h(2(wRme;)B zWFucMh3ZB%Zzrdg+;+x6ca@9O$O?68mQ=iYd?7NtU0AjN9S4*vG*tL<>lA6LI&mP? zT$y?jc+`yzPLwx8eSZ*RTU9e%EBy{O#139+jVm-WB|%fA!IQj4QYNUq%*>t|j-HU4 zZ16myaW6r6>q=i`X6D&f$-5puT(F=fwMv+Aysz>=pT~64N8u*5ZT~*E&7ss@8x*ud zB*|EY4(BMM++684t@Fhv0>(3K<12yNF-I<*!j$-v${m%^UX=?229<5F!$*r(oHsv#4H z3}XPPybV`v45gI44O;DZD0i#t9{WB}q?0GDLPzbIP_%(N{ZF=vBc%?egwE@U##Hy~ zxy!sJEFu{11R2UUE9zG2zBYo?$e(gqu8x*$72D($>xdJB))1ICz^2qbXu=wvR-8Hy zCvdysiqr1cc4 zO=<>}Mv(Z*mQ*pew~E^GkSS{W`dp5T_lv2DKFrMwvoz!V@k zp_-d0=a@**((g`6gjHJ1`rO)eE~avf(H17s@PlkxSzT)Hi7?7vip`CRm9m>0`PD0V zDP_e9zVz&SU~Z?)DECDx*Cy05@sBz+gB~EzfOtaQN>$xo>C$I?vGn- zO|cio$9jCY<2yJ!fSZx+jjUR9A2Ug0f-K=Iz8}!CPn$A$@x%HSxqGwao0j1(VOZ@k zkGnvCwK9$1;f-un-`q{$?pdzt)sW+3?vldBq_zuHAa;*uR0j%@|&d(JE!`Fd5yQ%)vBJ@IUjAl(Ej?Wws>m$h2dt$4>qZR zSIS7%k~(V_1q(@zd`EQUR&fdm2pwrDaWLfuZ*KE86)l*ila7X!{PZJ7QKkLGQU-() z;tsFc0Zi5xBcN6lxHma?t%u34d=T~GzWtTjqC!8!pDdis9(BGKTAt2iaaM@!nPYOw z+!d3IvIDviSBDukNJfAz*mw}Um z7Dg=f`j6Sg<3cU9*X7118|bQ>S+8S5vA=*Xzg^Mi+sd5#N31E( z)qq7V@RA&+L;^7W$@%TA%h1Y6eQK?ST3L!{YBr+|0(fAbsKJAP(_OiaFE@Vyc2cZAr1abZ>w5MP?7Tp*Jt z%Gjrq!%wA;7u$}EQw!eOd=gNR2VH1LA1xm05S?_08Bo_NZtH($qEZenRO)=%?~%n8 zpQ4@C^c}aC4C#S&(_7{AT7CD(EZUlK8~m07;%`6Q2G;t=>ihA3D!+XE6;!@3{OPiB zQlhtmmgzlbCuisj_!2FS$9IsOvBi1X1I@Pw;tpU~)#3ySYGgSK(OayRMjaHDdB=j6 z`Q|>=e&Agro>u~FYwgAh1T`8Ch$k484btktaXnWC0YuU$!+)SZf9oGvnOM_Ty=QfN z16e!j)XT+tS|2o&vlq6ZO{he1SWPr7H9v=V3AgR?{>svoEdf4z^-8F5{&iD{$^l^)z9xau}1ZVyp6q zW2dbAwHt7EV`n!|G^bjPdLwr6SQ$1lVXM4Y>J-;re;jU%!HQ1keHZ{wXdKUwR|l-V zgSEo@K(7>%wyzbYVgs?`guN$Hx3KB{W?fk8JGrpk3d($H7YZX5Ig#xSlD;@rJuMW5 zlWR~&Y%r?^Ia=E*Io{O$VBU`!k(D{vHj-?YRvRZK7C%q6SA^(tDTcd*2S`2vA>%_S zq(!sqgmDXePW06>c2D^l_-7}al-Ye^aA&S@F>}`zCzidFc9MDqJz+bXkobVYMpb2W ztk5HP`k+gllge=>3jOq0eBRCXq=;&}nu8R}6qnqYa%dyhL?rWNj z^g2o-rF#;h7X@*6R+HJLRUVV#A`=n@o%G#?1k-Q;z6>^~`O4s$$=Kk^IgZV3_9xaF z-UB=VmA%K|?k&st<>Jat1%;lfNi%N%@@amw>mL{E#=AbjeF@F9XbYiPFi2s&dIXfV zdr8HKFypZUws5_xz9Os4T&M`@u7q)Kf|R}pS^jQw5Z)whrGbBM z+*hcI(o@eF5kVW%402^H`Z7FEH-_HHoq_EP8PT#d@q2s-cg*K}9PYK<{KErTs60@t zdu>#4`-A9)Fovc)bZn2TsUbJ{;quR5sxf<^LP1L>5xh_wAQw%G7M4- z+c>apDZ=wN1sc}L)-t*NYN5<`O}rG99-a}lO3DN-ol!;9Z!xOy7d(U~rh{`QywbOM zb;cfZU{i`0si_$JDs8xW&)8#=Ui^4&px{&qlOEi@o0L~b(%8TB>BmdgU$^G#C3^-1 z9k2$?>zwRjjw+y6S@4&rxU#2{T(TA?$@TV1W+O{TFIZpUrWOM3C^L5#0H&txk_W2C z@>em1()Jiwn3k`^-P5Mj0yen1h9)!5A~hd_W+i|qrAGi@aCrj<3Kl0dl>u$&l_BFp zRsKY_%X^pTgaYd-BcPLk%YMq_2oz+(eaO^-y(#|bBtLIiVLbb*-o08eF`1@Y(%9Nt zNf@QUzLzSsMO>Ly$E|RZ{luv^+Vm#gbMI99RWSoai8F8bTheV*d?#|=bQz!U4+_%0 z!wdVUHBAQiC^7Tgk7%|%!QlmnR*6wPx_TCVI7ilLrrU8hrr~XRZVqv~TgCBWk~FBk ztt_A9`bttm8FbAO_vLm~q;_|guQ%|?Y_v()1-oYHd@87Y%2Ny37FTkrYQ?B@dBV`z z$0pY?>KhSDgee|mvXR>bw-aT{t%m|y&DsOE8rcio6eOPacs=Jxk@Z+0t0`*!Tx@ZE z&n}ZH$mnX%2HEhTQE#CRRxDKTTvxo8;h>MSMB)=3<*ZQSROkLf*|it0C-B8TRkh38 zqh0{Re`C2u%@@-0m!71)(DUsS&TJIgop-Y7s>SVsf-z3*0cpZvg(bOUs)amr{&$Tk zYpWa(2S0@dd4I(j(*rlqibe8xrC_UZ+wIF065hH1)R#wrSFzt)Yb!mpyOKRBQb;CQ z0;XLgdz-IKTKfz%NETLMqGH-VgAN!xKH!hi!k~2&mvV08YWQP2WuH{ul@9mUtpNb# zDsi_~t_C*F_)|bl>9C~Fh8B%b3$}k2z3=8dl>1C|!Wh_hqn=j~GJq!49^10f%#(ee zYHf8<&;e6kUmx@Z+*u=NI!tHgKNm0JU{)KOkjkG);gP_mR?Ie1R}qvGs6b4d*B2jB;^JdP!)BP3IrY#p5K{EtIOj zkt36da_LuqN5-tN#;o@{(3wTmBs+6N2g%#0v}$VkhiP07y9r(aZ&e!X1!iBT1AL_ zJFrfc3cF@8W4JY^tD$zz0<=Zl*xLO*L{d*WPkOmzHO1J{VG@QE4chr3enEh~jm?sz zQyC{;&-~bWpy7HCkl6D0P*`n4zv{uoeYa_WFI#nx?bFW+olPlr5*AI10m-IdF=V}) zO_nj59BYUYBfzL_u+rINksjEO3t;+wA#9%c@oq}>Y%Ll#7UI?Zf_-i!%!?(i@kh!i z%$J!y`Q`&dej~aJt|qd59az1|2S<8`77@7KzchN^s1j;U_z>PzsjAIHe+T8cdbqW) zAl65*2b=P0?)L;Rhs`oadoig4At4T&e>PIBrX}WAy#zw4O(M*gLbi| z`OKA>*ND1^4?dchYA!?CW7xhHnRZt$ui9ez>(0j-a5)Wz{h= zzpf_+C4;oDv&g+;5qYlkevWL+MjQ99-BYUw=)3_cp99a1kLsQKGNh%XzCKmPOaL~> zyrxKWt@iLsVzjAW#XlN6=NTi{X&=h8{_(m5Z(@42CMb&mP~mQE6&KxmWd8Kj;fRxm z*rrW?W^)U>24hsRixhutH{D26r0-EXN2YTKGl7p!f$6_tC6ZKMkk974@=Fe-QC51Dg9>>L z8=w#jAHq}@S{XVa_vSUJ3KZkxWccz5JsKxpmmTVS#pVPt1sQe%$&4W=3Q$2A0Tve& zzS(xd59p7;P2)Sioj8Nt{sNX2VEppl8BjBOWx!Z~z&hV?O_ z;wA%P*i7$OTDvTdvG+3aLo{5M>@MFJFsxL?>{rfiGi+v36e-3t0n{#``~bzdoBe8P z6Y#`b22PM{92`jfq(xWi;^HDSm8>rRUpMg+Wn|qAnumvBpmqf5>h(A;S3p?!ascM{q- z?LU}8@GRD*X#4c9_7nAaB|jRNb1t6>2&dn<1(V+J+i0@mQq-tD(4XXT2Qm!6hESMJ zpM-F$f;LW^seMu6+7_X|K*_+w20PriHreVu{;@$-uVea5NYms8hdF(&qF;{kH-gK9 zfeQEdyk0odYWps?O6?O?XinJ^M}zeQ_c#3UG+Dly;t9^3%cYyeIf3Hmw`+h!@6O?l zx6aVB^-Co_IH?2*r|xUi_SB|Bj}d=_uW!09+s#C~srbUb3-R~mQ^)Q)hECx7eCI(F zY^EEqHz)@U?ti4CrR6iNwCPLz257ud*gJF-^x*PK_eB+o2h7Z9tioDtAHQnm)YwyU;p6?MwE6TuyhK7c1)8dJ)m!=tX|nQZSe?@a-}-W;gi?c)&J!b7FldKKW>xR(U-@^%CqoZAKW?#swX|{ ztmq)IZ2E_4bAVWn@uGO+_ROX+P4C{g#JvwU-Q2VN+(lrj80gsgH_lyC@S%;Fl_Q{orN65@B@}D|Mw^Wl?wVNaAhEw8J3^k{8Q+iLqIDU!Jn$ zMQ0_6+g$Z5fJ3kQurW{fuFFXlo^l4rvz^MrYG9K}>% zi|%oJH$33@<^^w8M#W=e7BLy*+0$kXSRp19<|wnojzk(`N@Grttp_~p=m3tePsV>! z_&@uIQdTz-K#|kM!m6@2+bl|4DwVz+tQ5o>YH4Xn79EcRwqB~(GTTp?g&3iEC)M$Q zgXunz4dXt!RY}f4{WpL8U_P0(2$xjUiRK8>&nw5XN<9O-m3$qAO7Q>!X>VsQA>837 z&n-m1G)mZMMiNLUQKKZ-{?EUSt(Jn1Eg*Hq4phO8;z&Leoi+?Dg2`*+@%v7so#g7@ z>uC+l9p=m^FnHG)NKxm4nD*7uvQts>sdfnbBv&|2Z25y=RYPs=hO3DegQj2#hp}6~ zNQ(H3Nv^}l$E{*^cN~iSJeiq$Ay8oO>?ZrPmu%z@`TuTi(6YD-KLRN7AAu#88KT3O zf)JmWdoL&AB58xIHZYBw+5(?kIDfOv+qf(D{Ak_vb#QJD^EJ-^%ua(8u{$mUr@JGX zo|j~J;lZ55h>ODC%-tOJdrc{aS>yI(swZzR7x7Q4dy|qlZwsTf#zNmS%`t8U6*{G4P(gZXVK=BogV3=d#7(ep@ zet^TWlBr-(Ks@ewro`%B^KUVR#EkdsU z`Tw}Y-$3TR?2)+SDVKx$_vQZ=Z6`+0?(i8}*Vzps-jULE>-;s zXXcXFT2*`25He1X-)QN}KDrH-vr#*m2Ce->b+}3Cu~%fkXoV4aALAs0@!#n`Ym=Yr zsQcik`p*(-eb7bp3y@%LZZBszO#(e0Z+5VQjkf*iM-Ds7{X3vmU~^O3-wRXC28a0W zt(L?JUWW>MZ5Ur{N5Kt1et%*LV(oJ%jUUt!YtUzVNMd(Eq8a8ALmKr|ao|KL>k%9H zpO~_!S(eGb$$Ezw>8AZj$MqPYfg!gBf_pN7tpUfl;+eL&Y?8@rZCVZE8vj2hy~q;u z{Rw$!a=@s<#KY4&!1SWm-mMjYfAHr)OIPnH{^2#O_1c^n1JY3YCapXdi zYigKW=G0kY=cf>I=mMB8bopl*VpxIR$K@mU0Qr<~{qUDy!ZXsey)h#v7bkGyEqm6D zob$62qYN+N?m}D*;2tZZq6=Lt>4B%AW-LZ@kV{IoJ z68K+baNs_2RY60gE{^e4n3Ce`*8ut}$O80mpvgz66~Hf^D|r!(}xPE`Oo zs@dfu6maa0vCV*

    y7WB0NPj%Xx3gMj5a3re~rI{SiIcmm7gOBvu1;yn^P?)D^rRa>1FL*u2O$uL{JG z=4!+NHK?Q2*IO;tm5c8HSf(&_)N(z_u){C5nF#t9O4Q!-Q)X^vC1=d$oV?HB*@sgl4M;o$%2)2-UXeN`)z z)KPhdpNX>gpXtYuQQ$aICzak`6ja5I8p}KhIvi^l`jPg_E?7F2&E=m|Zh za%zLO25Rh0?VfSDbO|$X^yU9TLUWKY3v?@WBD0XBg zfzNMbsRl_jPH|*UE*dy|yzw^;!f)4J<~c33o)TdntdoQmaX!mfTq(#XERuEBh6>Z$ zltf|Mfz`sAzm2rgf=r~7KE)S-eINN61S2HVW|JtDAaUxb8 zOKef0=X}_R#1c(3)YiNC4Pkmx#!cYt^&f2(l)z>&Il7wY2a4HQfnK(@sMdsKyoEoZ z3+eUO#WxnXj5m(w<}-^A0N=Q^Aaq$UDR8RZ`W&O|m3qo4L9t_M-zjK)4Ci*f_rqYo z?umE9@Abffq)85+t*Dd=nf_yi%)r?)Yan+W{gsW%S8H&la?I*k?(hVquQSIMVB7AW zxL8HngSB>)dfw&q7YbA#zr5GmSv+o^d%IX?8My-rO)alP7Ux!GbFlY%v4*C%?|CPO z$a~xgN8WH8CkUZS!dQ3*Zsk4L++4%h!Q9!4xl$;tXUBBD4DAXV{Udr{f1u`nt0o-$ z-fuT~#;1CT!e`ZH)ZW{TRTM|zN8!+1<1+9S@N7a?O~3A+c3+dLBw0!6Xn#CEvd|qq z=Hf^`uM2Mql30cY;XYLK&DV#31-f%2VRH1^E*WteNB!H~J+2&e(9vZ0O`XjyZw0FUpaOgCHQ_94Tp`Y7>v`fj6EZpg3Wg2WxV7CIcxe5U|f zHh=H9Ex1zakT@m?r}Wb%c<>k7zBk)$fx_HB!uF7%`8_w-275B0!qzdM(GUmzT<`av z&F+ehXY~pD+57`vL$eDBC6kr)XUlGhq__9&I7+;!hxQ6~rrR4^$?KcNJ8GMv0RR z$D0fWJ@n5$p!iv*5@i^O?tS<&f#(SJ=vXWn{H!Own4ifd5r-7M>!UnX1LMwQX zp=T4c?W-DpjWTiH2p+WPUWOpgdDTm0l+%|h6mdX@T=*ny`)f{K#dbm3_HEJ2hjyiU z*9Dx%3ztrV)gZaCf)*zTyvg{=902I!-GY9$eFnGaa9FvZ0Wbd*!~{CA$gA zq2ZXzV!IG0XOWtCdF??XBbchg*oe;c8*Bu}b&pDIn~w@@JkCTX;Tn1#5<(O-S z;@c4}1JO&@ue3jhr3vP+>maoP#_+)B?8-b7wdKZZ$&L>{930bA4U5OI`fX3T@kc4G z|9his;|a>s_+?6@eE}~9cH5JbR(lAe|1QV@OFY3w(G2{O^V1EV4!?v)UXmU*ko(0| zW!Dj)P=jrP1_Y}iDr9JF#9C?=l|8pIm;#J>_*GaYpPfU^Gg5+&vr?= z-(L?Bnfw!!6{CI8{#P%KX~I5uFRB%CDOU76XLBP7Xk11{kZYbJ674qngMXPjR(+ZC zVAFH8ANvLZA9dI3Exfz1V@E3A#MBRk;_a2;|2lR&l_Je8Yw+}-bwslEo)S$kMm2cg zw%7ozgcd+oV<$of9IEDMbN7Ewk0@e_gT7o)Etl)*z!Pj?Y1zrEH`{jk6X=_xz71y& zBcW{(NdMn9O;jguIPMMi(^$+sRehdqLN3JmLD`Vx)O|w2U4FZM_7sM(zfHfAFQuz^ z;p;bY{Z2|99ln-xfesDmZtLcyA7|^>7VmP|LajVW8(U{PuORx45rqx(n~|*t)wMmx zeK0Oc#=k1r_onpLYcBX8b;aM59-ei~Ig9s8;5WhWsnX;y^0(8vElH5RfQWiqP*Y+x zaN>z7Dp*)uyhESd2Aj;y4)eC@3pn{(VjPSN%h8+C$+{@pzw~VSa^ZL(s?*)qIcYvZ z*Wqgy4hedUj^xo2h7$ zdIqyd0sxByw^eApR@DfZw6-QWr+Np=FQ8-q53dt#cMv0n& z>)(@ophn2{vWGO=%WKsdODEbxP`rgK?X~>VQcHj;agrVMAKv)41D+}IZH%`~WR+aHEe6`27~`W-#F85ZV1^tERGUB~uM0PBBDTeQ z{I;I?7?GJK#SuCrQsIZ!Ws&EGqDX?DX}P`$0a}F%IQl;|xMFH10qv$$%g}pX4gXE!CUzmt2*(7G$Qe=CT zpBx-dR5Wu~u&vKi6@uF|p2CZa)5#jGc@UOV+Z)tE=oWiE#WlCDidS% zX7$(~vOMziU#Io9w8Q*9gk{oxCI`nfT&|d-HB=CuK0_^S<;qXdxyQwt`Ci{C4aZ2s zcZN6bwWH{P%TT_>kyg2Kj?b6JgY1qY*C(ElJy+BG+9zkZR)I<3J+`SbA=)k2q^I*i zr^u_!ztUkeamJe|?|r@%8am9SM-~ooPil@^JOqVfKJcdK`Y!yB`vh?lA)VnOuyLM^ z!d!z0<-&{4he-3)fKWoZLC(@|WYle;w5Rz}!{JqlgL|p5G6+_g&Yt~(&8Wa%w}A#X zbr*C$ZkIZFt+FoY)PEQrdOy+h@@SSB3(e+7k zy!pF30d;L!^z$n1y{{&juL$f?B}x@{?W(m8yWH;iB)0Bt*D|r!?j#D5wj|s1zC4`9 z*k5>|iToyg1Bq3P&k~1~cy5ZW${4{O6XcHf)k~AXOg3`1Q`9Y#b`~6}Pk6l7!Tgtx zvqnX>1w4AP$I&^)?c^UXzXz;VR@JXHi&DTOfZ;PN8&Z(#=q}SoILo)BFnVVe1N!Z| z>P13m@T@MTsJ;a|u`jvazK?2eh{sM`rLH)**C1U~Hg4eEza2Q1*G+7%gqAvkBGpC> zN9~HIsPH&1UdEzLFDNe90(i_CMhxwaUv2n+WNei7fXb>V#*B74M^KLl2^EH9hi;E{fex5k$C& zR_0wAj?W{051vK(+|%AJcC2ncxVXYo*^XUJr&JSc zpsn((1*(y2T*nNKh)+s45-Idc2$^#JEL&?%vmUGAV%uZ;#?j|KJtT4W5|ZmtM+$cl zLecykd6g0e|D$@EtC1K<$LBLRVLcIOKT;xg|5#?i*^$d9X-q$=bNH2TcT;+@$5j2) zB<6?11aU(oyVO88_#5!_G0%T?S(*>fNT%y%UQ}V%A!Po}+fVO!uzI(r5XS#D+~63C z;0n*3bu?={_`P4L#mxur@7X0wLn0EsM!5c)lWD5|Q%>$2k&a4@-tSSHg63$EXcH1q zUwI}-&!W^`a|(APt~U)LUMs-k{Q7?5v=c6-#52xsTU5fBOqMLNt&N<6Y17O&?Hq+Yrmwto-<^N$j1U9OSOP1^ZV? z0i||C&6tEDu}&h$diSK*S;&|tHrue5i3O#5`lgDCE18jpFusIM{=MgjN^8Ds6qU zRhm=f^-C!ffOx9JK&7A*K`Ivw=pTz-`thSnaFLfQ>eH}#!3-&6-xtLtW*3y}=X>Ld zAWcu~9pXzV(4K%M`KfKeeLTwS32Q0RK`KpODTyHyvx1y(Vjn>}>NhiRNIvhqqjD2? zK<(u~!=4->wPvusd$3^x!|hY*HYIpFMXDb7;vtN%`!b z30g6L8D&%#g-_vSM6=(_)+ssUo0z9KlR^;SH=01Rkkftw_aW}{4s0>bVUz=(UUZ%O z)6ubXldb)CYlFxp?{Ed~-L|C-8B0N&;V&#vADGhr6%D-rC?~+1NEP zWXb>(J5GBJ8*-htH1NE==M_8GXeB_I-&wJl!GPy0Wt(Yg8+ff&Z5t_mFy-;43_U9V0 zeQS#F!$d=#(U)LXRqTY$BJ?im+TV0rTHjT6asXZ-}qbV zFLcx3wvVq~vzlPOX68MP7E&voNAewfBvPMzZ!%HGHCzjI8aB+B)>h2s=q=x zCc9TK<hl__G5pY^MExb#QdLwxC)o*5JsYtpqw>JUcsx-i zoz(e@=zV}3Tmz4#?IbDv{xF!A-b4yoqygo*RL=ik=|1lprKf`arNp}>pgYB4Pq`@G zn`5h1Y~leg%1he+`V4b`L!PES-}nZBF5#6jA2djGz+zn=;O^bLs^$_Gx-2c>RV_wX z#%okq{X#>$Bv`e|pT}09(p_X2^BBKvpOVIV5WOTZpBapVc1R2@1g?^ZBP@uB#gV5` zllkAh|CxbEWh3Q`3YdRvN{E}-r_PD+0+vV~tQtw2Tbhr>aaxf>&q&ND0j043PYLvt z4o7+3GmBu1(4u z3C$^jV=RFp3cEU!eYn~G`EVCT=E&P!rFu1b2Df?Rjg?<>91 z0Pawjre}@5A|lEErYn^H`;lJMC^h^pmo$L-!Qqo|#Al$zB)%Qp`%@Bye05P+#ki{d z3$LbW3HNiK@1n~M#Wn;@b0m$4IuvT@74>;GPO+OJwEn51W#L$xIuKDsD*((dFE~x> zD6w1`UzUs`iUvcp4(XK zTuTkaL*)0<3tI1zz?i`41SW)GM6dGEia&K!^j$>1tWbvUmhisteEnH*_rLn(PLn!O zF25JJ7fRym9rB@~R24;fkzNvt3W_voDqTwGAYEEOlq!OBLJ7Tu zj)W3Q?#}z3^PO|<`tJGjuC>>qB!txo9aM}^zXCa!8hCaA?tFX%VTFP|roZMJ@eEC@JJ|-l_##j0kdwi?Q8pVbgpqO| z;$r4$!d7x!(^KE7)aeqEL~sM=^s43ohF`U5~~{5t+p}KYj9s2 zVsi{pda`8mD|?cY-fn*kzy(Z`NP}AdHCgj8A@el%q#1kC?ApogyW#4Ykos5Zi5PlM zYz*7{^-BTTx%hma)DG9$zafW>!X)o1 zsh$C=*o->|CP0P9u9#Fa!x)j9G!7=O^-P#Kp%zGu96H-askp$N2xSI&9>(}K<(vSR z+%Yals3`hu5OU*ay<}-1 z$@HD{6OUDVNsMU=o67~w#$h1#wUycdn}t+>7;G?!2JX^%@2Hb_#9OypX7^uX;8gjW zlrw{ROe$TxJwW-%MS_7*V7VwrnxVEsssrEbp?^Leh-lkcXHg6N*oIoYR)ZnoR9xX$ zu-56a+xau!y2j~JFv}UhMce}XGBH~u$ptzt7A_AUC-McTD)x1N1Z7`s5--pB_J5)sf zQJhvGXc1Qljj}4kL^_+JJBrMP&BPz*BlbSFGt#`l`<6)dczb=yX~onvElIdoHXSx^ zPm>Vvs|-pzt=?_)(GxQFC#E{28h=Q*{)St23S3B~s$T#e z(PXs1eln$-+>Q)K+ROg=)bI(jVg0cj-UPmkJ(9xkdM6rhfgafeTm$toh_k$>;ax8% zzi~?LtGGq*-3oXCs7!omoZfUDle0=SucH`sno0n`D>lU9-S|uZ(8xJfj`kah#8+UC zqOs%BsrA5guhsg`ht}%EXoD{@tqD;^HCrK$mo{Hkn2!ObIT#8(%kzV25^ASxuG!nl zZ|;=@6M!T_?oc5Irk7>v92nwlX<%2aO+{7JS`)BM6hjVnZOfq{Y2`snV$mV0>X5gf zP6!1NP246r!Tlq1j#WQr0M3sRRm5hR`PO&nqVU+3&O{@G&aqM_W~yXU=q%b!0WA6!@MFc?ovD zbznN+(K3P0mG*Jo0Alk$hwOSN=@F3R52880-)z}*adS$=fauTLkw=WFNS}n%sYQZf zcpq76m&)0izAuR(Fxj-47~E1{_du|~`6^pY|HInMEPdEBK<-ZHlM(=*IIIhSk=x7I z;yUA3Iw(0NT8V9ph>Ae$$;{Xc!va0~4|>{5Q?#BIUBuJJ2Qg zd%%a=(zT@3bNs{6(z*&M^4c~=OzVc#gT(!!SAk{`=h!8&jdBuL!e|{vQs~!3=~+eW zG5n-&tIfQ>fe8t}vj8I3`gu(_uS`MDRZQ-)1#__f3!iSq4-ASpupdpy2qTyV<6WOT zfMhesYu7*>cr4^7`JTw37(rQG~>ezHhxId}JpD(0x-R-c-Wx z!9Z@pS?%p{XlNwi9mIx?A21Y3v6#*OGWZ!tsQ>BLwufOVN5C^E0iUD+#bpWiTp8gQ zD^x~+V_ z(K)Ui5Q93N3wnX=3slqFA!cQcI=rK3B*BqKx1m6EubsuzEPJ~P%h)0o1$Z4&BPW=G zF_Tv~1B`Q#5YLo&H_ecT$jZer1COv{&5wK5tr7pS>X8>f~er<=FD%Ma2L z$Rhd)@Eb2pEw&og%Uoo$SHPS!gR^84Uk*qZT?xN$I$`IGGK_L#VIH6q{wgMo!}?C; zhDN{!R9?rO{~QQ;`FM6vjA7Tt))b(VjuC4bqa~DBVFAy>h{^d-!{2~I7jX;57+|0X zb`fhX^9c+;*MNpmOxPWm@iqa&b#j9MEH>q?Z=jLRRZb6W8s^`MVa8G1@ATkH`;KWY z^r2za?2^J2)`iz>6ko5CDr0|NX1Y7kBoUy4{CGg3yYBd`YKCq8lJe`ciT zFTM1@kC@h(KAaB+>5R)=8|botA8FQWokGyw5~o$Rn-J}RXD`TaQY3)F1KLAn4QIn% z>`gtU&iL_^dYRR)xO{%iBb+eu@Q&^0Iw}#$vLDZe0q5G%W*58C-b?qWVysoXs9)+8 zhp-E%pz7uW>K>>x0f5Oyl2D3Xtjp|ehPnE>5uIAhXu#D#`zcnz%>=zud*Vjs?UvDp zq>)x}Gd{==09~T^DzqV{;Q3?D?22S2OK*SzR2!nNs88bZ z)FrH0hOvc5(-FdFD`)v3G0o5VB^t~=Oo$2vD%>>9s^K05k0PB368_GCnJP2H-^^EX+P`XGc#X_gdf&R1j43&3)h%_Bgw14=ugjLzht zRT-t&G#O#-xSZQ;f@XHV9m)M4d|z<1y*>CTslzeZgGc5uv^V}RpmxqkxZmvK@|nqi zKH-qMi}i}x=8|E^(B*sGZnRA2){P-#moKlrG=M!b?ENB9kr>d|=zCwGc;vNio8xk4 zE&oXu?QZ9mKyFaoCmCPmn|aL}UUhLxdv(yK5jXA(kJqw6=!6s$jM;BG-T%Jcm@fJp zCh^bs2Et}(n{30M^zErlz!Xi^`GDIRp*R~t2^5v|AV26S5M<@6hd{tQZ2`zn)qPH< zS$x~fpthjZ7kHmA+6v3{=uG&n%ehCpvB)M54jJYJSuWKIz(1)>OzVpH1;j+^U*q#R zF$zV2fQR?YsN0UASE1XOtk^rnFmmuR&Bbc?E3xJU?(K0&q?^+!y1t9F0JtczUXFAl zE()nkoGsa$Kc%}&+BV4W>^02&lOV-WuV{t5?ClO~*Hw+{b%K|FUH-&o8~U?KaVIOB z!Lj@jle_e~xdUxv92I}~Q2TFY_rKP?+erfU4gVxluT&AMim$KL}Q;jn)n`*rM=F$0qQ zk-~v7O?F!~ssskM|dW2_Z7QxZgo%djtQR#wbw7IQhA%-;4t zHl4xLZIPpc=jd#ntB1-AXGC(ALsYHnvA(zxr83bd&!1wwMsaNh5y8QEZ*zUG9^VZ2t(TXS?;Znh3`>B@e#tS>iD5 zeq^SC{@q^`{#WN<>Gp|C+co;LdmlS6l{SYxIa{THr`Bj=>pxjPp9;J_{z||n_|!}% zR!=M5L+`C334PZx2U$Kmj3BMl)1ADxi5Xg#)u)h{xeIbO!rBd*5D4TPt- zF;2FmZl@Zi^^QBpUTj%l=;6Pz4JVowRw9+}iZL2Ub2nZ6UZ`48Ty+H!JXw_H}+wXEr4Zb9W+-zzyid8kjTnJ0JY53vIbny!6$d3Vl+IY z0&3kB)z$&p>TaMr!QIB%EtlROeOAJ(H#aInnHJg0X1?;db4K#<#TRj!;CrsR&sQ)_ zrCfe@&pvWFK0(Z2Sb+GuyQ7OVEmk4jT*WJh_Ue)p(%-lHTFPI0?f_+D)%px)>I-}` zj`tORy#i8)+y-n+>$ipt^<91mOOxNUTj7hSu?F0CFLS-T=jqA!ZSX8oqcIJYcD~P! z+-08z(38_#szRCmqhhl1!F%os7#kwqrp}Y5c0Q3$Fp26WZn!j{PmG$sa`GpGREiha zfc6UcA?9O*+>*O4-R=nX%`?&8zD+f{&6PiH9$K?!2c;P@obXc_x4zvgrjuvAZx>Ok z@KeAv=|@%AGr9bpkb|${`_%+R4wR6XC5~^GN3x(Y<{`?S)6nG7~fDb)YTzGfu z(+nh@NwXht3TYHlIqudQ9f`r+n@K94MrljBptwiEH|W}Bj8@Rcsb~iWz7I>qn;^mW zik%jmFV08eOPDc(voDLon^q%gQ|nHQXGm$KMeA>pW5uAzH%<}gtDl-N8UD*3Hw{}Q zls)G2)9b*Yq4w$CChLs2JOsvi0z^YAJeaXpo+af_NHvj=)KCsuTxY!w7_bH%0-NW& zvvT6NLjyoxuM5zB=t9I>%VLGd!f0ePU2`TLitrH6(9tlQ>f0K91|Sm`xl zgKh%0hLg>NQjLprOl2A`tNxN~-XGjj<0Tjj!H$P~8x>GZZ%u4JH~f}pd+znIq&lT~ z0*ue$>?w;O;6Gc94(8fp?x|>8{_-`fs5@`}BYl;D06L(iA}Xe9!O7<2;W`$hsj`zg zE7$PZGAh#YdwEdq?{b?7|)l$LJ-H|W_ z{Bcf_?d_q5j3vg$OO#3MGk-`DRHqw&+3Sp5tXfdFuJz#2n8sAs!CU{$W^FcvFaBf! zFr}O!VE>i~`1H4y%*iAc`sTzU5qo*ob~>0W3St5jz%?kM$y8UE|)D~}Ce`{!hN%ffEsOMaC}TAGIeGGN^Ti|H_2=LP)C zeJyUaosdcroOC3=2_V;fuR06i!wJIFLsYW?)8RFOB3pRQbuy}Bp{zcgFdsP0b>?yY zxMGT6OI|>^0wZ=26ilPdazo(Cg|C7kKdT~p{C`bG0HP6F#^nLVbR7m%Rx(qSpSys|iar88vqcNxVXlTX-j z*TgS_C1&kH=P<`am?{HD6M*}6T^eS<^~5(%PH!F1*;U$RD(Dy}K(1V#TQ^cTzN;3J zTr0=|^WgF!_oB!FEg4yLVB##9jWS645DT)5hQh{!ycc#k(M)6%<%@dH!c4%W@r`>a zuE&8NVtu?F8125L0yeS9Y@YC}_gP;>+KPMjK8-cyGtcj_2MM-nyBovgZ2&sX?dg@y`=iqb;AImtb`y?1U|fCN9;Ht#qsCVNE$4+hzU;Moat*e}}?SH;@RHM_#bMtxt-N+*t^S*CU)%0Kgvr(xk zzxQb*mXP%%Bgtff{Gii#t%{03GwFD7karxs`F!I;WtMG|JRm~$3I02ZmSQAXEiU2U zsqR{_f*7N}Nv7fTTD^&yScFzSt|aqk)x^fDAGc)gW?o}FOA5v-ef(EBv;=(Qe9Cm5 z9o4Vg)Pr6K+@(JbsTZzr*pE|Va5znJ8vT>x>asJ>{^#~__o+DaQfAjHo1%~$Ovw@= z1?_^HS2Mmb&!0Ku=67#tsZNJ7f-L9}T4(4mDP`{{Mt|e@9NdzE_8h*2mlsDhEt2;r zFmAm-+b5Ym#5=1HZ`P-jmz+u-i-{GiH?4?mBhp{v<7)$oBZ_U{$P;?r8;NhKk6Vte zL6vRBf$*}xq9n$xc}lFg$Iax-vZn~qBdoWk zAG)u+aD2Bo6R~vPrGRT|`*2LeyA`%@vh(stBfwdp319fgytw5|8sBS76pIZkx>r%4 z*Bz^D$GJPV@%d&o{I9+VGE2S>bP=HegFgwQ=OvhXuGPm^oR&I_HIn<_++UGX+V95?!|x_k0W3w7!7z^R8M z;krf_7N0X0R<(+--ir{vl=fW&pO$qiMd4zbzl1#X)k~)WSEzpHw&23QYkR2?*P(J9+U>ykzCu zd)3+KCBA_NfrD1wBfs|RDExB;-0*`unD~t#Q*xuK*>hUR=P`_#1)XRrd!o64eW=ae z10R@Mq%ta%b@&y+F;JTP>0zBxb&mA=q>vkxMMFQCILqE+J+rLs?#*B>`%9@bHmzN6 zPP>$*k(_~gvgkstI=#x$TxM(Lq9?T^vU(RewhhX05RzuuK_LgX5$@c|m$R-vhpRmB zIebp(f{+2$dY>}(xIaNiz9<2z4pba|-;^GaaSns_Z&k2*y>l&MZ}!n&J2gOs zN*Ezs^o8=~E-RniF=wvwN``l-J&Y7o|3WYg$@RoVID`D_wWe3&_L2~I7#n;b0c~r7 z7t}g4J9(c%_^s?;@x;p*Y1ldFvMF0_labQmGZ33A#ngW%fw0<$;}2HYaPtR@(h?;G z5q2dP|3IV;I&|VY82kO6))pi06gGhcD%rhsS_^6g#JmmIQchO|zH$sHX+iB~1B2B* zcFtP&I|v;YCbZ~Ej-dW^EKz&==vh>lD4|tlSM50Y^hzt$92M$l8VtRvQXB@#a}r36 zlIpFS=iQrgKBz8C((iHeLetxwuI~*!J01$&(Rb-l1}fj57~jgy_U}%gc2FhCsKgTIX3Gtc-A5W`cy~oc_ErZ@jiC@rHQ(XEEdpIOSP|Ik@wF)D)$~r#f4VR-`}Ml*)rx0A|smdP(W+wcz);qBDpvv zeRTWT$wO$b7wOwXN+p0FxQqhx;r_|i9<=TZT&`rpJRx{9zk z0YiWgYQZ~naE%K8#HNUQsW^iWyA4;`{SKU@_xKW!NWGAI`{sKiV>LP+3j4;qW}t(R znXV{6l?M}(0e4VS`czrODW7Nsv1!UieBK(A*fLOF&fA3g^jid%cxW3cgX`w{$2pE2 zR5ll&{$IejEgq!=`$d<{EE$1yPR#4z)W_{>0>=pngu)c1q(U6(9B0}z&Ht-RHDAuS zUo3UMx3=MQT4g+|gWNl7g|NZT@rAO6Clon)FdFji5n-AMXR8-pTD<2ss^L3g{xP?3 zD$i~2J^<4)xqdjn?vpSxM_rRt115~}z91!rc&&eq;w0XTFkD)cqk+x~^Z7BcYtz4E zp4V*!*^hNQatXv_w9@uBn983@k##jciQ}=44mZiAzVx8=%7Zqd^_y?x#oWE;$*$l> z9y0Qi8b%&|@*X{Jnl4W*-L9h0mrK<6=^>MpP$fhkgQ^TDi={1#zfNUp(Wu{06CQAM zk2h_VpQ1*@uTS{>vyH#xpL$B!vv~-wGS8`%!N)6|`aDjw72ys^u}@0grH%0~RTr~- zT}~G);8xN<@1oh89T=kQH@d;@@8m0NwaZm`iHRM-`(7Vo`Ee>rNnNvmPWn#c06E&< zga;?ApPNo7R16+CzzHjFnm*;A=q}pOrU;O@q8l)nJ~E>=K)*=AfBF7`|K^yAPKCCe zcfJ6o)fe~j<)J9gnZA4t_s4{f(ns*D4~Qzuf|T0ggej)kFpecL#-)rq6h+0Ra7BM+ zhp<*7DF+&vTqlBp@|_?)?rzoKHGO8?Fv^hx_o3q5zJPjVqcYET@O~^;4 z5W8}sfAYPK#lj%x`^JgbuOD09r3NHrJHiq~PRp+6RhI@hG?Hah9C@$Y&aR1;k(EHx1WI}wfW zG`AcoThw;z{{Dd^hxhkYQTk<=3}ngRYLKM^rr9gqBGrhPhW`Q|DIo@S1Ys@p5E?Er zqGbggdJ-F?fVhE<#ukbP-DqJ)CZd%tw2WB*66tJmFQfKKR*nLKOoMol_osKX!rmPn z1ryOmjow{v^&_l2XIY4p>4 z_HABv5;or7Sr%GXX@kY5zX{QQ{S_v|o^>rmAHvH(uBB@2*W#~0Q%~OZoOI;NX`^Y4 zVXZT*G~snQ4P^|<+D*p50>coWgNNg19HjPqb)qaidI+PRD}PclXg2<4bb0;c+X!%K zdgx&J2JzSJ;HTfj*2f6+A5Sc>eBn5(T~-FB3+wf**nB#Qz5VI$c0R0nhu5l}TY#ai zMW%twnHdnpTsF+zM1E&rETD2l8r~)6sAx-y>2a`vrY26ynm}<}!0W$G`G;a^UtWoI z>mFG}?O@7o0OpPjz!ufakX*8TvL?knfE`y3{*NKu)4h_m(%$LfcJ$D9*^ zX#z0@S4UidQP)Ng_!~#imi*n3ipCQs7o*|o-XwP?lCE5Ic9j+~f5q5 zRzWL-b*oBq_;s!G{S$FscF*tyF@yZDNx2hhE`J@vIK~wNp0r_1&2{De7g^1&{$p0Y zF8!uE?_0%TJZa9}ORBw-x46{S_*V$_+#J>BA%Dx77M`ndBuqKs;nyAdb1^SnZYYDJM|C<$H(jnt z+wOaqy?if5uQi5_#-vTwPSgq%)xam+#qb5@+8!dhG%y+|8Y6#)=FNkr!xsikW%u5{ zCez+jv&Q9fkLV(_!GcU5eEdMIM068=vba{Fc-}osjl`aZATZ7*Jr!s8N~EH-yz>zT*F4fN%hJw1h&W$ z7KPwXU*k{DxTbH%^R3i}GJ#(C^AqPDl8qci1fI`ZO!|S!U>T6B8^E5cUE<#7-0YND zRrKo1bLk$}@grpD`{OtBPWRbe)Ua`0th2pP$d+zrgy7Z5SNorHx^5lVq!y(@1^bI^ z(7a!@9`+PPuezdnN5Wo8$Fo%GrwSaNj_PNg2{aBmczgbCBXDwTzTdtw%zhtL-V?hy zfLzRuPAm>tth(~KZ`1mEV7ba;_L`KY>MsNn`68l7F%}o-%@Z!;`Q;79+?r5xI&6CN zKjIruYoM2-f$_;T(_UJ{{`p8~8R6fTZJejg1(LDyU5j0&y;;+}>|M|)VSax{M$%!Y z^z7Fm7bSXt#`5$rv?k_B&}sFI2<{JGS=XV@zw z^;}=+L)NuZ2+}OQ3oO+-wv3vWtD%!GjPNmlBYZp>F)X}o5gi))e3`VDzW>nzr8R+<56F(e zSR=2|9WNwLfL(r>bd8OS-450bj0c!U#5UM_8ffQ>DAjZL#87N?mGm6Svi;o9HldnM z*?o?E*WkxVVS?xCA?8iYeCTj9Zks3g=eC-az#;A3c1DWnqD*YM@zO;6b zh_~2Wtz?7PczPKdU~af6;N&YIb+l8-nE_p~zmm4rC^6xEh9-MCQd?CmbBECCF9?)A zt&*3h!vw_BH07=hr!i+w`*zRl@w`W;oe=~}>pm>h#fo~HD(QN^an@z6u~n5l{juq4 zqV)3<(F7kusi%)*SNY5d(THD_hOdeuS{`i}UMJk;Df6FtOsYKB{)6tSxb_7D3ro{6 zWrn7WkY8oxM`Q*cteqdl66=#{C_+W)9mEt@(h^Vz zTuIQ@oVsWrvoP0VxZsL;s`EMowx?Dy{z~-ug~dMwTFOO%z>)L>#=0VPg;}SI0+i$_ z&@9)3Q%@dzf1U_|fygKP;lUs?68hE2rLc_EUHt;ZGRQLE0&|{3yjYJ&d3Hj{FRtK4 ze$+z*#<jz;=UATE&54Wr^7-peg!ReyE5t77i3;13ImQmx&8Gm{W{u2#;o7Z*|8ui=ma z=Z}s@?psVkY#5aH2ez1F{nXb3d#pWClwti-rnyO;UAh(nEBtjQf>AOLiy>`}?B7`* zU)+swe?~_-r*;Lb!JPuGlbLYzloixPS$4m76{|Yeb(@k^2h$8XeOI>1K{>}Y@c;^L zbz>^M#T2;Jn%d8GT{qW;WF7SAFWHr+Da?Ay<4`|0&Jr^AbMt*!pIgXuV2y1EDBkWN zRITqZ59|Ww%GbiPE;DRC@oyn6N(Q;v*C9%Hku!*HZREWv3)ow`+p!y?K3z)3zeNs* z9yOpc_t(pm;7!9Ok=n45jZ?s%dvMQ$!CSo2>SNb%Fy`e&$GK-bH${MLV8)<%?v1Yj zaK(d7)qip{V9nGdm$bVb4mwUfbwbjqw<}V|v>VUG9L1lSLWQsfwimE8*jAKeS$F>J z!*|E82UoUf3g#$#qj;~oaZ#W@Zrik!HXWs&s-9gBMRxlSvH5+5d!C}2t-l9 zLTa_V4S<#Rxv%Nc+Q0NCdD|*Y5}reGespN}tMjmk>E4ZDDH7vBfz3uFTFWr?VIjite4+_$mJAH9w* z5Y2!q>&GcEy|w-YQPsv({dx^#LW+G&U^nIvV@CW|TEHlOV*SmIP!_iI2lpeQ|9NXpxe zT?yqHT+e=ZtpNTtPwP4+JlzbWaNX7fJGzX~|2iIa(_b{TZeOm@hu*Bhvq(4;Ec zk)T5HJDjl+w{lU)jjBcWUv)%eY02MaFi5yq~fjStZ6Ku-u7nwxXBugjHN_pzQg z@vGNYLLx!9SZ0uF_)T=2tbDYGhqu&3oYYxdOka<4No{eZgUu64<)`c2iF#iSa{C?1 z--~eDrLKMPd@gl1OmwDFd~3r+Inf&{87FX(Z>3yZR^@C-id|5%^LwIyA6MUZe+@>+ zg+?uOew@ETCkp%S{#GuHb@s>F!bJgBx#F%2k=YH;yXznrL*ORa3%*B-V!;f}`{3m9 zo1GJXIPC|jpv-z1Fx#U(Y}!wLD)~>i*6sy$f$u?{Gh>u#`hx5$hVmg*Z*Ax9HCDf= z9gytVGE^SX368s%@f(`9Q8;iRD_X8Q%^?(;hH%NJw?d#8_tc!-Yhmsz?_smcO2VeR zI5q3YSb*`;g9?BozMcB~M4QCcOV9bN4(=J{TJ*&)hqrKeomeM-W!S!z7D&jHO(J z^^R9NH>kLOU;hKi;u3k|rOw}@MHu9GAF*%sQ5C7y)2-EVc2J70QM9Wsswj^hZyFiQ_X{gc%gzj6Yg9TnxMGZ=%bPze-i(r z?|BA3O`j+xrSd5@lAZa(y#JE%fxp3WLvq;pSyaXk{J7D|%Nv~QppBguhrnj6Rvu4+ zz!XQx78-FLFH+HcD(q{;#9u`17;JgWy>sqKzTlrwtUI%}@4%8R^L@fUKP)NH_@qDl zWbG19wHJu$l_}e>XlxNDcNO1*bHDre93emE%27?6Mbq+?WKppzJx(7kOt%q*Zv_ix z&Qq=BM-O8h52t1J9FFrIPBzI6H_ z@1V=|wsNu+z1?zBt6)U@%??hRe*#*0uL17tb41;|*X(xjPsW%HD8VEPC0(f^r})&K zlHj9Zg%@e)=%A)}LbaNbpabrasH8JS9?)eL5PzHd>sateZa$9&cV1t8_wFvT_$eT5k*#X|mGodj*}bq4ut(7?D`TnJjC2ENZBEak z;%1xmvGp{**+29f#)e~O!8ss{(+(WKzX&L^S$HQrx+)_PIo(=Vure)KqVJgRe98 ziHqqKlqqJNT)p5Fe2w+%XFX=xv_XGRXe|r%wGT1vEtfBge4iWfNpo5jLKA%P zrl@{HXj&TTF!;IbrWW=z`iKb^1uOox z%gcBFLfDVfuAbNi$%Hb?Wpyeo-`o^D5r|c)1Y6C{A%TN$7YWP;F! zmCpLlEt%J)T0S&^gh)DJYa|sQb{1~dTJ(Ukgl1(2}9-?nJ<{qE46|326o$=}PN4D(VQV35D8 zuXy43GEmu``a)lL&B?9aK**bAR;_L|y~#b3B51g8`H|4ucn)PC(9AiHnp`KmO_i{^ zPO0#=e;(q_r451g6YzUnY4n+ElX$ZvXk0ygPn+dIV@i-6GX=|L^C18rJ6r=wzk0auBSl&imJBH#tP zH@D#7Na*@oH54$GA<&;`SEL1jgG6H+PAteOJWW&veQONc+z8e9OY7I4hAZ{+bm@1pb$0=GUfi#lSt{^#%9x>iw$s;eH?#U#Rcm zPkTX7|LI#mDdM+yz(s3=iQ@ztCxr^`?^cCv59@ zpLj*CCML@NnRm8!gVyyS~!W6%Zd@AN1S#qta9AC;ZEacLiBHn6QR=ZX{EkOr% zm^yT&COc5RJB(dSMVik|Fl_%>wpZ^hOor>oukD+V4~uGLtkikI)rOf_=8XJzl_LZX zudFqHf>iORhcF08y|uG?NsB19y+ugOU!%BaZ<>e@Y6TY5O`&r=;C4H4ReXc<(( zs2?_H%AS_ZMBZ97A&Oc&WmZKSN`aU>?*-s-+2#ce7G%F)QwTGU7Sr!tmr~45;utP> zF53o_b2i3GaZx`-znO7Qz&4FhbeF4>up^-*FiyxUQ^bD9i?Faf+0ilf(4UEV2GyRu zvC|w?HgNTl^p-s%*CjsAHQ@xWF6Vo^DUSsf4Hx3}|BEKmW^y6Z)K&v5PjhD;wae)d zhUjFWeph5Ta%2!Bi{M3jy(M;~q}1y9$y^G|q2V=#p8nYTbKDjr1FcYHOVt5?3Lz8c z$D$JVYXzT8I*p2m)+6~E^_JmrfGTj z9O+`VIlW@N(iSY)6dEGH6|r9}gtTnLPcq4>LxXuzE++cl2b7EN8T)WTmLi054ROmu zueSFU(;au0(>yjHh9^oI5uUr!# z2Xm$lRgyXH?i(!>(pdJmu-wylRy45Bb-fBhgB{avqYyDDvWNqz%w2q}H(vHE+#bNy&NGrfL{j;sUn2xM$Ry2($2Pu}T9HmjaOr^cqol->DOxrMI#x|kaiN$}@%w5hfKQvXOPZhfB|t=N`r!|s3&2X&ja%7#V1BRBW^d9%~Y&99xa zAweJdHSY=Pt40)pXhFLuQm=nJp;*h=E%$7F+%f6`W~>TsWROoXHk=kgr49N-7F&SL z%P#}j^YXM_kmf|5N!9C-y@LoTO(^Z$0{^T(`V(XrDm`epj3+}{4Wx#AoOj*{qeua* z&&_pYom14MB`&=vggy78vZO{&TCi0=>Grr77PbodN@;DtM31xysLD$F!AdvJeD9hq z4g1aOZqSCl8Mtjb_YtpYIAbBp&L|yD$Cp4Tjd5z@E0DED`SB}Vh(XYf;mlL-gNNpP z$xJ2V;BlK*R4C;hrlQxbO?2xA;F_6RXQ(%ZGHa-~4TajGyWZ>h(G zuFIs8mo{NJ6G_E7IXmhC4F+ziD8q-L@);{PJS020YI>;R==S0}E^Ax3zvXs;==alk zJ=pS($-p|AUhS;MU9az2GkYR9T6+Oy89c0emdKpF_6Z}2yQMk$i&FkM`*$gdCwH%M zIX)HF`e<0bx!m2<#ov}i(YJV8-z3;5@Ul)w$dHMh$r@bGiN{1Bx<<7|cHmL4yW>O+ zcW}9n+G(G_r;|8Fj#Vd(D>2=T6|09ycL=!fYPC@#Me2=8gV`F0bsL`nJZpC6(*Ck+ z=ril``$BK;ng$GNdi96<{Chuv?+r8ca6dK z*K#y%x7lnR>~Bc3$9Mazp6sa*PyeJkk7TBS8EGBWXJ}88q2|eDQebDx?g#02DE^cC zR$<)gE2i?6@!6BH&w3@NLa?Vbg-l0Fr_wF+A8bYsd5>T2G5ggQ4B zc*M=UdQ`s4XgaaZMmB{|DT2#)o1ZA{JmOSY@pUoH(!!`V;O3w{G%CUcap{jIYd*@D zm$md7o)0EpIFGgZFyiBU^(40i(_59%uJ@jPLS(2Zx+TNnZSVjChIY}k$(~e8@E~cC z0rA63131rtrS)n2O2C3ezv8Un_Eu4?yX(Bz7V0!jhUfC-dm7sJ=)EWo`DMkl4IYYl zICkC`IbO6Cr4%wrb)a+k_9zo0X_wDUIMjDQ8ff<8DT&LE2ajXtM zVIGQv1E{-h60dG+e@K7(m(rh~HuXWs4i0;-m9!V|ztO{<-G2tvO8A;Oy7Gz|eIDr- z>6vimZ3oYD@(Gg}eC-E_KPD08_!os}EO>|pSL_uRb)GC!K=s))Bx{I@RJN?8yh>9p zQ5}8yHx-N|Z-4mKwCQsoz?Q%?8Ny{hrUxDo!)K%L#YiU;r-HkB>Lf`NVv61?uKrt zXOUVol`|mgYW7bMC&h8ooi|5nr%vP?hVZ_XTVJW%3|nquKVeNdhn{7hIx>}XYL#TY z^WWjJ(Z68;#&@R@8-%l%6OC`AXC*e6W-(>5F;?&lzZ z`#|{4ZDZm?S}6?%IX;3lIdJ0mQQQ8$;(Lzf5;A@1^dhH8TDiyM_GJ%mwfZ{pqG;#8 z?;vKr-CWlbkdwhq+KL^O)C)G;>jY&*PgE8))(*WU^c$6Vi$L7)PXrP*TEd7;qgG^@ z4TYT*uDHB&8L~jmH+?NXuB}&X=w-gJJ4Ec8kC z?_{k&3akD;Y16)sab|J1INPnM;jH4?z9Uq`Lx30g6(C>!L2OCf{IOOFvNL;puE$nVPZG?H&| z;&ZVNJv*qEZ!RLVuBu|#$C6<+fVnOtq-vOKFwX%W?{}L`g?vt*>vT-& z`_U5f^_|7}AaUAD`vgvu03@e-+JP}rhdi7eFrxJRMkC}Ul;QW7?zz`Di1Y{OMVB8} zsx|Op1Da7@f9D*ep0uqB(O;aW#0UznI^S=lBbLdD^po=@_sc9t8AHLLycE*;dk-j{ z;Yzt$kt-^9?eAX=QX;qCl0s-G45lvimw~I3S_PjnE#@)+yWwBMstTdjVF!1u7YE%I zZVoGFB%qvV**|Y~HDN5cm%F(ZuerFCyDrLr_YBYoG68O;=fl{GlPlRWiUg&28LQkw zuyiZ)@7p`#%^!>G7t4sVeM(}@A_27$p+fDOG^@ov*-_`6#&ac2(T_(hOhnq7^3?(`TUxvbbxQfj41vn5XPTIMm%tH}D}u2z)p1vB3Jg&1>Df zsz$I)1u@So@EksGi}D(wi4EfgQ;Q~qbdib&F6}$^Dkl>MwO2gt1gfVWcqdA%O~N*7 zW&(JC+~ZWq1X)c_>Tn5$m;s~5Pe+3d)33v;oJ9OR*m@_D8HGwI6!Ke}gFDLxQy;u8 z#EaSoXBb+>)k!C^Y)sHT_^4GvKZl4p8kRoD-ZFf82?iG)WH6dnoo8i6YI$LU@nzS`W+=;$(A7 z8Q4s3o{yjjrqiUI{TATWwfp=}qOLfUe@o2zM0}oWugS(uHkkPkRUtH}`xZOpQIC8cW#fG{*E%MYRHX>*zZKVb5^F;rF!)|gUaFXsrB(?*;{npvG$;s zMU&$2$83xD`;Bb*KGNJ6e70vr-(?>NL)#2~EOb%5aMpT5iPp+O@DIhu`0$m)NB3rK z985mI`1dp)so_10GwS(*l4ftZ%~iB{k8;Xc;$O#*s3_n;|57QV*|MzK`8yE>r|w!9 zI}6x(4-8w=s81djMoilpEAQOdYmRP_8AhBg@S?L35L<&rSH92O|%_xZFGN>UVezKNCA=l3Kmp=htD*l?Fuv z@(Q=he2Hg=Zk%tx^gU_{_1c`TV*&1{3B!_yKg(N%q42`mX2@pC&m^b7N=6{Rwjlox zK@8&yTF-xez zf`S48X#uG!NR9MPK!O5_bPyFnhtNTKFBS+OEr>|yCDNpY9@=lm=Q-!h`~3@EW)DNi z3=`Jgd+oLE`?{{r!vT1L5GL_=JZ%Cr!D=>SAlv2Phhxo@pru-hJBA!`bNGcC&jKsV z2eN~`HIR`#$<@ew^phx&BYw-;b3@4f9iDh83>hOomO5;dK|C=uMP0d|w>ZnysISJ_ zK-EX;GxRihV(B8^t>VoWGV2sJ1K~>OGGE)zN*_{{Rx?&rjV9i!NMX*M8GJB)XzX)u zb%eFPm|S@je?>O8btnN98u(=<6fYEtIb91qt(draNKbW5d+2pgbY)^lHRY;_jQ(-a z%_q9?nYOAp)AkT60u)|hEfumIJ%Kt`4vkct%+9xx7F~u?Dy;i=kWzT-2hX!uRL;>@ zZy`2-y4nWM@vrW>*-oO7f`cZH-LZfSTT&3LrXGI;M&WlN#pOGqt@|q0XJ%D~Z4XSOHhLoi` zOj%ft8}LLAr5+h(BB_0n>H-@JR@scFik0(fFJ?GVKcMI2#_CxZAJ5&`;xOtf?=PNm z=drs{%Y7Vp|J@GORX6S(rpsu6n-cVh-TFO2pdI2<@`Iy<=C*e(*+IE0YQ5$)Q4 zZPXJj!GQa>xi#e1O3dLZV0^cbPQHmL{7^D16T!_ z?b1LK;5@$fmhaA}R_o%`s~ z%`W`X&ZO}Y?*}#0+7fqNQ@C!UgN0h%V^fC{ccy6$NIj-fk@vEo$+IMsB$*#pQH}i zuuoVGpbLl`XE-np(r=yIYy`yVE6tM&Az?+3)8lZ}!|;uAzgCxt+GfB`?RNk?60py0 z4puZ~&i3mkVr?q|R+7RNA%1N~{cbXedXLIOhE|nm`|_qCN-eqh%&xbFOS%$bSXoO=LEH?eYGH~iB=d>`vlQP<-;EE1*zi_E`+ zT`Z8OQ7wPKI71XGn6`@sjXw1^;cm76dd+q{JC=gm+aM^Y+u)4wu0x44s4{_x3vaa?q~lY+iyGDtB~bkk`k8NDv)h;TJVoNZIlX9 zI1Xw*2+}K;iJTyTEnJrmSa@S}0nGeB2fmdA#2ya2ye8TR^?whoJlqb)-U&^^Y7sBn zTJ=&-;uyO!x%al@7;eR>-S`?qBHjEAblxwRN;tr(FP(d8=9_<|c5gWyoGagykMPNv z5@7CAhEy^CsgKHIipo(Tv;GlR?W~pIlCbb5p2>e9%u`oL3yeHT3ZN{$pacFzcF z#2HWraoZ^k0YEm&n|=y0X#VcUr;%IDC(F0ceO^H2JS9Y=!}LS&&(>-?fzVD{M1OFT z7Ejc*ih-fQdOt^1X$|+`pmlY6-#-c|ZV4e3GP~_WYu1!9um)jEb~q?+ck?N+sGZqU ztkdCec}%uzb4shO1+k|4bMbRZ;!!?187ncg**7kDQIIpBNGT-O&HJ^)dx}65FV!`1 zW`icy7yny6jCPz`(v&}yKf34uRk^AXsj9#l>~z4+j?n+i#1=VBF_Xw^(E>VWl3MoS z`SCimM@nJ_&Q?!~@5nAplJ@r%jWdjoJaK&=9{5ar(0uEAa?87LXAM}l7PNkTj!oTS zft8S!r$a%o%-Yql=Bp5?5=B21wHfa=#QNV>+gx#hwd4t6J&XT5m-iO<@NT-VWR2{bdvkbOgm zX0a3gWzYf4xRlaA2IOu{60}tC3Hu69DsKO@iL}ji?~`E~xw^*x^I7n~=Sp2N=1p&u z;_XHyAf4m6dUWqjse^J!Pv3-=dx~G%4${J+r&}zsHPQS-uZkp@qQOa)Z#SK(EsR*i zxN*HrfrB_HbQX{?YrJ?v!ZILasIC@|v`udVEIK5&4?^}zY+$h&YGOhU$6s~LiSb>R zo8U6eOb_l9B%=klfA8yVMHs38b5$)L&mwD84rK|yTz%}1dd0Y%H;xZgP&QQ3=@c)& zFR7j%y&LUV&6@>TJ+Ia7f=#_ zxqJ~w?zS^gGvK@MhKp~f$PW8}&j!^yoFbqfZ@=o?{aZl5=y=!%uFXAqK#b(TLI&^R z)&30i5tjtA%eWj8sYNd4yNxMHN=g=2`YtVZ6YazMMR9RBL94x?NXAE!GgO8A8qmFBo1zw2E7q@LAYvR;8&0kIk@P+Qtz$^^(_co*wzmu8&$ax(S zh0cYJvZLSiuO*ETE~E)-8?<>Nrbbgo-Dt5k3Oi_LylKG$Eo|>so@C^^rJS>(NkT#i z8`tg)h5V3zCt_gOyGZF7*bTxIPt1eKGoL4%cP~ktR~%BUe8A%9Lu%+~TIq6xjQI~XtPYRPT*A*1xQ@*(>Cm;_7*%tSeHRjQTUhyWJMYK_E#5D zpBlD1&;75juNdCj*9nN`wdlQF5rUm?v!_T=&#b-yon0JI+?=$&D=xeVY)-Tmj%ESb zz4d~mN0@h&luMv9`6o4X`|f~w#T0w$Ug*cl+jN7{CRld6AF9boS}LrnG}`vp=HpeF_k z)R#n+s?^*-ywxRrj2lI%L`QuD!-9`+srA0e`U!M9YVV>*HOaLzH}v=~n^Nb2Ft5(U zB9)j*$hO#+K4AOBq(SLNoOAq+rkO*gPqV&|Z7x^aJ|L*(BX^;vg#?4vH*6lt7eHAZ z&hL35*k%V1%Vw z2Rixn-yXa5874Bs=54RtRU(9dwxJ(9r?2%eZpR9D_j`59|1{j%a@%p8 zym&Fr0qH`?-Vc1Bj}s*3-7E%tmK!;6mp6b?M*GL}2OA1$yVs z*J_P{$mfNP^n<=z0zq@@i}DahaPZ-TA6W#Dh7ngAB#+^8?v%xNR$ac6{Q}T6c(}^H zXN^a{#sw>WEAjo>7_SXcW*NHXzeLPT6EvgVK{L-fQZrv9L2+?C4fF0@(+=*8!T;A&LyHzv(ng z^moY^vblR|7VqA}YFB3OCo&+1tQT>hPK8hOq-v>z#pz3o46H!J9F9dvbatRzb59tL z5lW=L(HP}u0m9`Z`ML_Uqd6Pz$u56unKe-(!`yiJI^+tF#)>UOJW8>8l<(qQfe=1z zzgT4y%4&C&PuDzQo{(7s+Drg|Wl(vv6))Q?3bJ~w{Vu7s9bX(VvQL82aAvRN%nQiG zn6za_BbDPb`;_rlZ=UwU5iBdw$Gb&$n<~q)HxLrz!k#RYMR8&`)3js=uYGIL>fNLL zD22m;R=c&31HzA=*m^bDDT{+;c1+$Fw$Qs2b4|9L=JHjoE4<4r-o}i8@&-O7#eHfW zME!OnGed!9V@JHPXn~l`246MGbUrl;Eo;I+Q=3FlD*XkX`$Px_Z8F{SYBnh1c)q^9OIvejeZYlBDU9A@$GzzX8STC z=P(wo>QC#hp8iI6ZuYDPD{V6YON!Z=6JIiW<-|yt>KgC@)pW!Vn$W5$B|Fo^&8ksGk}A z?eSb<-R&)Ff*SZ+LSxo7OKlIQr+gH!B{~KbQudtlUs0<1V;!x#%wsPje_XgaW|7j^tL&(eUmj`4bgVUd#B6*D2=qha3tp#wON&Btm! z3HwAgiyPOXSEIk(yv+G)qisNZ{&#-abQeot@!QqB%-K1I@igF`vOWE`L-R{u_7Bw^ zbvxsZOF8u8UfbW;{fH4$&9bd4dQ;{O$WcVNr9&p%_iva-(A1js7O$c~*&`0y15a z##_I?vP~WXXsya-Vi@772<1|;5yZ)<4cMS<3;=VN(TIcB89DPP&5!FA|7nAfihyh}zvgd+iqu__m6j}S~FZ;@tufQt#uC?s3nr6U<(tbLLIhs|baQF(k6wfW#VkoAcHhc@xW(i*12aw@jY zR+yeH>Ksh(mHX8ywEeM{a!Eb!q4l_R&#>(FE{qVmoE<6CoUaE`CvJ#0tdDx3GH{jc zuPhd)zpnV?ZSE$}5?S58QUN_W*j888)<2m8DE(ruy*|~WBvq%nw`C*uFo^|xj_U<8 ztI)-3a%=$wZ==)!!8$LW$jUW5Z&6>}u%fR?=Fa&gC6|Qoa-i|`Aa>R3_Xh`=AV}=0 z4s@K()A$n%-v)Wtr?y1x8_Wo+bMt2pLnY*0hh~<6a6#J z(2)qubPd>L&Tx{e!JXZzXWMLrFSWkV%_oXF78{`a_Q)}Z^(PJ|^GfK*_EO#8-_%b& zfR$LLH5m;&tXJ*AuUflKuMiGN(+&^U$Z~#zJ4=PoWn{ivIx?iiX<03wNQ6jTupI>h z)Vs{38m7Vf=IT%FuE|lN7DHARt`gE3Bje-v1q@-Km`SaUq z4gl_ikv%t~!6QusJ}^QGix{b4VwEMAc`|>)hMv}^h zW!_y|+TF~_)>F+^Tb@XePZKgEuf2z&BCma|-bkHy3r}c}cGicrWMqxlMJyHD{yuvQ z-f{gZUJ0+>1#3ZMjQoF-wX%~G#--@?yn+3cT~3z)5-NVd^COKApp@6*TPHgR_T#O$ z_yZr^55K`R2;U;LdwZXcIo&M3MG|LGPV#mJkl^Y3grQ*@OQ^kal!3k9KeF3vC}H)j#mrAN3JrWTM?(AR9!Gx zM6PM9z!K)$>$?};X+$V?zFB5t>3EYvS%^qxZRSLRL^@Rm7sk z<@m6q#zlVg%LO1*K?(IS*J6F8I{ipRu4ol!O|Rf!OyKgxmA8Z2(OP`06kH-#4t0)d zz}-9|RXYF_*_T_P9aGHmc{yIjyJA)WCU}0M+mgpenggS2A&67b`Kg^L;0od{3MY(+ z;e6x93YxkPR*PUvs#m4(mTE`DGjzx1{#Sjx5%QgV7`QBV;Q*K-Wr`1Tv z;QCJXX;6TIE&IcFC%aV_Q9D+(;IkK* zPE?Qn=i#h(o5xk+_N%;O) zM6it#HSfeXw=Ti^G-G+nwRG(*tD?#w?X$Pmg<|IUJV^ajk7mh#=vapsNd*+KGy;EB zh#Qu*e)J7@%zjdShP6qHkHk*`X=^Smah@?A;Rdjx^%Q8@9G^$&VOraUQ`&rpyHqco zKi5fQ-9xFW4ylIB$saM+LCqu(()#S@GGr=1-mc?G&OVptON^l`Y!0sq7dKC^$-J4P z(Ew5}!0){EjBjpo{DRG!j+Dt`D4&0hGg{YzPP+`}~hpj}mY|Z`wL0rtV4zzVX;=P?poTXsa3I)#z34^^*<3!FAqxU=E0oLPU2DRRL z7R9MLR#UlgL0#2^(l%|sd>uCX=pT)c6z|aS%h2|ILaMLrG4ezkg$APr_RVM>s@S_{ z?g_z4q$*_zRUxJA{i`f(K#^uSQ`#x?h%k*(5KC=xW~nk}OFQhjKa)GwbuS=s<Gn_i{omYj2`YMS!)PxGG}VBxkgKTDm)CcGy0u`mVPx7b4d+PnLhX36CvNq<-Pu; z^Y@k~qr3`I1H5HBE(Z>#PAU^Eh%l-KVI16>RxKQLpON;-Tk`%e z6Q~`hz>r9kh9|w;m);yn?9f$JCq5CmB5iUYKS%P)C@BTkkhXvqsdundL&1 zH}Z;DEhYuO(G?K-D)2eq8%TkZgIkF#ziD#3*;M#_Rrlc8UL$`p=eTjj13*_{5b{+( z`HrE9;;=cP2H5-50P-%pP0X^j-e3JTMumFF6BhmIzTI3c z<)Y+18bF-XRJZ-kp;^`Ht~J7Gb)&VFSHPV2pfwbR`qR6yEy;}xuvock7AwU-b!+{$ zk$#t2dvZ2`iBsGSRtn$CecWc3M_0o%o^bP6DAd>T_|wATHQ1?fv%nJLvsEG36;XK% zYgH1Dwg%_0z2}d2gv!lj_xv@?bIe{)LN#F+P{Q?z})Jb7Y< zHR>~TU%4$TuD6^0{k&#~5GB5BTW#^{Oc&?8OpV@*pX98nyIYn@u;*Sp#Y}yJ_Bl;M zC(TadX_@ZHGJd62jEAoM6B7c2>ps|5rN3_x9~1L_HC)3O8q9L^>sFxzn2Iptg*8+Y z${2d4c0}9vI%sGbo1k)$*Lb~v4wt>*jfBX>#FRVAa=l!tB|Uo;`xZ>)Op#iXV#Tq? zeN+^8vO-D!1s?LnV#q2$f;}U@M2@-TL!b!>cZk!w=d3l*?LlwdBlJxE_p4$)c-d)i zJ&YVBdk`Psk%ML1tOJUC^0g&86D~HeD{kur$YmD!HPlZ!(6808-V&=GuZQtj=RI#r z-z2@0Q5K`8Vn?+GFD0E-W$spzMYXnmMtLqoE<4*;6Y!5HsMwOf97qC?YCwkQTa=`u zc$}d(R{ry;yVtbZJd7WqoPrcz@!wuw%+YV4IDaiRrqWwjn$mW@eTsc6aUjF(18C8& zR?UIA9C{e8WKhbUSrjeiZz=)}(}4c)j~sFIzLVdUrZ4#Lh~CQ-}sByuvo z)f=hAbOQ*ulV!UitQSyp)7nWu<4m3vVet>Q$f$K2cbI|u6dw{d_NYU4Z%I; zL~CVWQf)@{+E!PC4hktOzHY40?3%YmfUXZcm3h0dHp8a7ydIz04i2rU?$!$eHtmNW;OtEL-i&h5XF+6`>NNZ2{jXsxL zFL1?vI9pD|MgyT2y`%a@%1rTSD_=T}B@BB6L0WUgZv3u2UbiPP26`>aW+S0J#KM=S zu{gs{`2xXyyUPX5RB9(0^-wWE>%YsQd*D!J-YZ$8?4&0ja3HGr6;Lwe=MErl;7Ciw z!+NKq;_%&aoCbp`3sZwed3nE{cMHxwsS%lB zjgu}dsT(pq@i^Lk*z?{^0^K!(`J9T@QKOBGn|T{PdTm#28uN4isSwq?SPCn9sjFQ_ zedgsW%koabk^4U)fV4lbi3q28UnI}0d5zQP*f|1;?mNY%Y*x_w({a&e@`m0hdFy>g zk)YRFW0ewCijFOxuj(m+sIr)82`wkgs9S9*v+~_~=~)!FAFsc1!Q5mT`?}?1b9z4e z*|v(&oH_h$=^k9#z`=2Xpw;JyR%n zbh&F$-Zc+T^(0rJoe87*qo2sm+?@}-2okC*VZO|ksPh0*#tdJvpze?dsZ$9&r#mY= z>x1PhC_>?FOTJS@oJ9ujxVZ>l+6a zNXg~;0L?PbBcu|eqV@$yDM33IB45I%wm!h`n<)51!y+Du7g?&bKGWjy#41ZEuOvdj z;FwOg^q@dg*=XuWMSp@1mQn;vX}k!k%F|}C$GvCgctEdBx@c01!VmFhuA-^@HscRJ zP{0af31mw)M|t@fSO^x&684%l8?_3dfNM#mg}D4v(ecnn8?F25wY@9qWf0&G!lMeB*)dGb`NopUG2L(SiIT7GE9@1{eX z>%g-h^Q;_stgxPRD&VGhBOvHrTQ8!}ls-VuN#lS|yeEQQdu+$=gvGUH^xan-{`NqQw_ELN@2~OI#n=9an$fe{IOVqBDw1Me?ix~bP z!~PRt9o#l@CK=iKm|;nsgDH(t&F*SEm@}{JnFQH?+YPQ4DVaU6A7k|WNoT#rHQ!de zrzRC7lLE!daXet84ZVEuW#&s`KRLa22i>_o$?Ga$);T|^=C|RS{aR$_FMcK%n&|dg zKOGkx7v}c%n92Lq$U}z}lj~Ms5-J`h;~YzWsy1;R69a}&bE_rXnykI}3@`7OOk)O8 zc{!=QpAo&r>0!Z#DP*OSYqD(;=M9K$!tCbG-Y6LQ4Yh|PQ3$(@H$+iSKIA>a6QgQm z#uEWfb09x=I$e9fC6b~T(T?s@vZ7=b-U8R&+i|=XYXl4L^Tm3*8#j>e{e-NV7m_H4 zJ=mq8u;8Md5~bJ9VvxT-man&RH6$5Ql55L7n0RD%RFmrK1^l)R`!YBg%)qC*{)Qsq z?Cf}R#vdG5nvOM9-hz8;$`j(-XL@>Du?DK~4CBI~VE*}`#dQ_?u`VOIgn)Qe<>bgF zM~C%%?QqaYiBdvJSDpbwxs$Y@WWHvhesq{OZf^=4;G^cHf_>xZDla z_ir6aTbAo(jIW(`zfM?PxT4V>5Fsf>amuk)-3ArHefC(!)qBg_b(j87{8B1(x>mHGrGq@cc!R}pIe@@nILuHUYO#lm4SLpy3c z@BYCLoy?JPEL71&)d{$~Zlv`^7+W+&O7^5q1{YU&i^Ur?f6@S+Zqg!=x5ka(CQYjb zxUnTyfoE>XoeeMgzO+gpq(#~Z0j5^qT8xpsfj+xcz7UWUHmd~|oOc`7IQ4n!WuKOA z7clpVy>6Rqski)7^sX?m9bs-GaIuEDl^d`agkZ*{%-{OYv4ojF-Gjz&We_Bh)>Al)O$PzVMGtbPfu7bGfk-zuR$i6c-k)1v|10cX_*t1ja62{uwf{yxZ zrv2&C!d5Dq<|w<=Hfvt#hk22ufcWMyAi16r8SjPlT`E|e;SdGv%?M*n!S{wv_SWUQ z5?d~!pyvswx*RXW_ZpVvX1ROzWBNT%_;Nqz^j5?(G4wt6_Z=2tIZtgJYM6K;bQf^} z>sN+!E7Nv+EZkG;4vqmh&?M1M;uaP`pi}I$7I)SN%*%A^4;dd%#V;4C+!d-#>wzJSrFTWUXdSU%W1+Qo% zfzs5QN^ZCPOrL$1^i(_ar`wOX0Ub`MP3Bwc$JCGPZ`LN2kd@|(XveiV$>O|K)4D@2 z3{N%puCCCNe;JLlR=6J@dkIRfu$*;m>28)RL2BwT#fp%{_l;`yt2#J-u_|M!)6{-j zztI3DeQl}bU4T8-Wva!Q;PJs5W_`{;EY?%}yXv^;{EO4emF=|6SljKY)98+i)y6s< zmmmG|j}8vC96k&qjF>3yC1ATv&p5RP*S|lQ;g49>mW&*I7Nb$ZMpG#_uYMi;e{*!^EA(7ZOw&xqs=@ux}lQVmfXihFTo<=yuh)C*ox-POXWccM}-eaLDd z*I8s8X$w56F3pu7%x=cRqNJEG`A4&yLP#Y%2dYAC5mP#7! z1~V=5sA$+QS6=%S_Z>Vc-=jB?*e^gyE2z~{wU2@H+N^8RBOtXb;}{_MTQ|QAB8gU(` zm~cY0b+=M|_4+(sYE&8FB$aP@eUBKTNpED11T7_Nkd~RqtQbGiFmBQjv3YDZz#rkn z&^LE{_h^5Uw~rzspzLCUoh7ovdNK>ZS?zv`gS!T$lp1XnxceCO+Hfj^-;S0%MaGN7 zyRP0xyy|^wYJFN))IJ40kVMff;bgy-o;YONTyx(v>@~jh4ko`sJMum<1iOP@LDeRE zM>$zt<^F zk-fo7?%5H~!9T~I8u&it3&sv3C|FaMD(BJ@af$brlZQinP0gRB1pS%L&%!LD6xUW~ z>bb;$kszENAI~tXt%j<5%c&H40}-4f+w~)8V4)LtYdSu2?TVC;8g&Op+qc}Av(p-} z)*`c*rDN3O-pFO^lriJoiP{WgO1aDX{jF&NEW_IL@)w6l=oI-5CXHHXWqeEc(pFu4 zD~#ruC!VwX*R`2I`z--okMAr69y(Y4eQk*C|>w1@9 zf{2gX2*`C?C7^Yf4*)@r!yT&sHES zN{1(4F=0x_hJe}7Ki}1#SksKCa5QJ&b47X=$N87NvD)`-4Z0eq)|KCK`PzTZk!@>b z)Q&j};&^psFVkwx)6I@{t=Hxo?#G$|N{#Q*^P&528IEwhuVmdIYI^++tnhkCIV)`= z_7|Iv<@znCpt2Jsw%Vz|x>JmD4Ul2O(!QfEP)ULjW{(AzhXa!zKg{6Fwvu6P=^E=v zvbHZ9w5Zs^I&(&&S|9-uwt8KpwAM5ggO$ibpHu2Mjt|6y1Ks~%U-ig4#Smb5ls30w z7*cpSTAZGy@2`IP1gWNIMJED>EvY#E6y*k|b|K7gb8nd*KKVoZ% zbjT?pP--7w8ZOQfS?9h~v?oBQOW+r=tDs0K`pMAq(`F*qZT%~YT*EY@_C64^PLEc{$wIWocS8E^F)OP0q3=1!lj}WJ97#p zlv99|=i`A~hKkzRdy`AtoBrMVqWUK$;2Llr1%xAWI}JejtjAE1Ri^)ZHd1iD9Z~Yl zN6x1|jRRPESJZSEr%i1mta-&$C+!$7ZjCmq68V6z(|JIJtbZIT=#Fl9?ji-qD&i(C zU&t8xeXQ^@3`VU*%!u+x_!nJV7er|}=l`?Fm8?<@!}Wu}`AIMy=NEe5*K7e$9$|jM zOQT}aRtPJZ2U%=d6A-1;EUhbvxdbAyMn-zJ3pg(k{(H;&^nnYQckTNw?LQIZE40l1 zx5Gaz=O>XzPu|V>Wpza4ImS><0#sdrLEC0hi|8v37JD4~qvs{#4*#1=ZAA*{K9Ze* zY)`-0oZMC9u9RT{VqwA#OKeV^5AvnpMF3;%=8d+lYjQ%~@**?$%?>}tbc`6oTW;WP zW*+GV3BK|)jh}Tp>2k9Sb*N;XCEMoor$4!oK_B|ZEwJ*VI>O&NtaQxMOK*V+Ofg>s zrn#nloy3Jxu5Nysi_2C|V|MDEd9%Sif~DK{Po|I8$!69!w_XBM8z*PGHI2!Ibm1s9jf)@gd_M7tXkJZ*rOQnI0J<|zzJe3nTO#?FGpwzf zcTKtjVj3$m?<|-i`<{VtpPea;az%jyQxe1U!uvY@zX4Og)n&%B73zNrRYSy@P7#T) z2y$>$!y8_2D(g?`7p0+MlUHDH0i`4@V)enq_^5%7Y+n~CwI ztp&FM{<(1`lU>=2S!xmA1c;J2X9fRf&ie1O@1GymnPnWQo0u;NS8f24<2q4+R0n;0 zMqKPX7QFFGNnCB{x1O87yFzAtj2xAVpt@DMg5RhK)Q8>DU7eovK z_sKawBdr4fm9XDm+wkST2wQ8xpo*ZWl4bxp8FY#SP6=>K(En@n13rdNgh+s|`G0~R zh;NH?`?IQhBS>2l764t52Td=t-0~ImcB(6-Mj#}!A93z#UFmcQ0vt1GiZaXqB4C^W zxS!c93u1|4xaJu`*^INj>6z_u*vQXx<3nIr+rgC>9k)yT`UEZZe>U}XxBnMA^v{R- zKi|_Xwg;d7S0eF$-u&lf8two7jsG*U|GfPF1He0Zx#^r2pWZvCvtmig1U8!DdOfI zPQuE*?X6oIQEE!^a3AxnS%R#TvFz*p+%Vn|_#=5`JQAimenwkd#}4n+k~v*XlN|cG zlngZ4486ZhIZ#p0a&|HJRlspl*p=VWQsT)+dsJRvQCaQRc3!Wf-9FsPJ=r{Rv)>8W z8Qty65^O=`e3pAaoye&M-1I@U0zt|DzDykN6}q_#xG9Tj{_o*o(xm?$ghSf@`+}3U z|Ns9#=JHXeCpqZz!x>eU0Gf|aKC4?yR9L`jr|dHS)ckvyzrVum8Rv!6i0)2hNusQK zk{<}am}{E!nszBL|Ne5NLj7ic3blT2jD1{cr+~#WdCv$RPub@>jFlPw#N}$|3y0w-TUqf!314gX{J;JOdh%r z=gCU{@;F3#VW`mhuqsR3W_hDPv3B;m9H-&;%Ff@9#@7}8J1i-?} zSOA>kvuV9ZPQdY$53r8+2z2~@=KKY>w)FN_xR5?^xZ2Z@p2Cmi4ZLv~O3!`Qi_(gP zqVl3_Mm3cGZ9&6>Z`1VEcAe3*`&ihuWZ+&1OJP-_Ytw!gJOj>q^K7B@9=;0ABDa={ z6+wga|BOgn_S=lRhAFCF%<E331;4tZ$?^Ok9?F+wb>T zkzIOWO<6pN_P+-VD>hsQ8Mz%lZ{a++c^MM|Z9 ztnQot^RTyCdr6&QYef?SG|==#Z%YkIZOC927bwkjEbDQMJR zZ2TVA4^sndGCGCNs62xdcydwKt~_p`%ePr^28aVC8Kbynf((yZ;@2Wk%`z)+yo{cn zfsc>-E5#pe6O2@A%e+kRl_4zb-N(Zqa}5x8RggxSKpdAwYRAD!uR&!8YOJO|vGAe6 z71PrS^Y1jAVy-jL+WElq5{s6=m*$C_jv999@OC;ubFirWKot zKbbtmUY+lEV?CaP_F^&^uTS(jDS62<9SPzON+%Ns@t zoPmOe(n%E`|K_p*8<0ueUdSNsx$(;fus`e5y+vWbOVY(%o|aXzO-@-wFj>zBoWvz* zOFf8}+9;U>c1pk7Am!2eTUF+$XD_~ssU$En?S4No|72XnE^A?up4^`P&tYd!gKN{@ zUYPptcMj((MjX_uJ0aQ%+oCy{iBi{GhYHn_{;0}cqnrUV;=Oa_!KQ?P*_;lN1n%zn zC9yg-Zr5tG_?O;bR=xgmnG;IR)D~lh2&&ZLnjf=9~iXf#{oO=%b1r9Rl88d zZ$Y>2F4H}HTBr*cxTA(`I{Y>vZ0)zWRa%E3>}&I9ee{F27iuXeVx8)-gnVu8HLcwRF>aA{8k?HFFtH+6gAsOvlle@Ch4*Eq zxXZiMx14IfsSLf-ZCVOMLe*ZhTzbVai18NdjzrVd;$(Htx-+lVDdlz8ya7(K228czy{isbSodzhMdSTz`#9hT)dz)2NrQZ} zGeo#N!#F5$4?}}nn23Oyc+5!6p zPNwva^LASPMU1N}p8CbU8m(@cfSENm+gl))L;q15OIfy*r}}M6>OCPL33;krTlyk} z$POPhkB9AB>ScEr%NOx}u$O%C{pd-V5BsP0>`|U$ zO3EYQ;V>t^-2VmBOqTXS5z@Jzx284WEW%J)1f5}cWkh3#ZjF+WfWm$JG440KcWSME ze{~cF{FJ6km#x!S4bKQz1aPbA2lm!Gh)owapKOT@>9p9xn?&g26X#o^&wi2r+5w@l z?hBJ212XMMx*A-qiX4B1ThrQYcq5hI!tR;4SJ1z!jI27wkSDo2R9kmOiy@fxGL-U^ zDyzktQu)FN^Pd2*Y92tE6z*r%HBL9w+M5xU7}Jdzc%U=IoSwW} z%R#?%myjLrpV-o2_OnLNKfYOG;@7h&3}2FaUUCQpFN#p#$ENH)EVrGmjiETMb9Q0Q z|FAzDinJyUd?)5axBa_DV@YCo#Z~b|mCg!{cevkLegvkA@kOjtc|CXY<@O7-cgy+g zAF+h_vjOjsi5stBxXA4JNJdr>{f&!2EAzPdE;Uv8i=4K>nFbos!Qak(eq1-paQdbk zp(aVg`>7)z=iTl1Jq!h56(Miz6+?0q;1t5zRNhDqV+^?~p4v>m3!>7r%kt$-;B6&f ziEX*Q+(n|9ygMwe|4xWevyEI|?Ou1~$%l*t>Uk=MJvsWFNPwODdh)&cP0jnG{v^ez zqX%|5HB;#)|HRC=>qni7-E*Jw?QcuJohr20dWF$oY9n)B+dms)d}JMFRSNd&BSrha z40JjV2VpEp+*;~Z{XhZ(1L?qIBegc)ZXk>PW)Vy4A9A(lQ3T&a2W3PDB(Yg>#9k%t z*3uS0nn)M2QYe#Q7vh}}WSBL@j~#Nx5!Rc%XLG=6rKV5Hh~t>1AmY->>wIr$EmFaGUT-qzC5eov%Jrqb6fGx2SU+89q* zd(pIQb2}RwkA>*Hi(mp~8JBDKRL!R?GEOsYnUt#{|T=bk7&ij|Qg-mhNptJRNsiPCGr)zLAO9_9UM! zQUs;9apKR)?RQWwMRLSP$_4y&hNp?{h@hd5L(TEzLhWWJz+SPsc~D4Lg3#D}Q=QsaJ8fdLsHKDS_ij5an=*R!6gu>egjkb-<^c zvWhsgZWzBzd|u7yDFNe0eBLl=N_7e{S@aaS0hUek*=^A&Nj}5r}02g z_Y7D@$$K)6dAIx>tFQ+AK`__pvSgtElVBNbOLgUc^wA>j@_4$!KmG`z`**XVMJ`Oi z)M;(}^Rq0k-;T&X!>VY&q?*QG#XK*}bZvNWg$=smF`LLch<_Brt5l2lknN>DsPP*D zXj1(*#cdaS8aqaYE*hbg{^`N0wfZp&SpzS+`@)ZI9Jyj)6BD_oF zS>~RNGMo3?{a#ufB^34QI8j3&idpt$@?ce6`bNjI+N_tsYB!Gtidg8j#*? zx~5)UbxJ&<3I{1YG_GX*-X%^ssUOzu<-ApAllc5D3=J4c=?})B`1FIu$Q4nJ+K1X^Q5uCQE%Ki%0UcNxM!dd8?%qg_RS zAo}z$mq|Ftv}5LkV=kz1Yf~a?W8_cZ6lC{%HJaWwD71xsU-M@ zmV{O-dwwS}8I&qC1ey5t3g`An$He9Vpb3jNQ-u@n0k+Sb`u1b=b4DwwY56n=e%GM4 zb|EJEN9rIwj=oBo)N9MK>QjCYx8AUlP}lyAKy@De947$3AX24}V9wqcPaf1@xN7st z71uItdnUtB>LJL{ZI#8@WtE>vk3;os(>^PG{$f=Eh|xNQJ1zuZw?_=DnaIR@No)oC z{jXXeeWf<+rkMJgeBLM(ImD>0vi$9O`@x*5!vX2EK{i7svj%scwwiv=$7zDNBW*tb z(L3Z4>y~6GmcXU7r|`zWNH<%_kSdrjMI}@bek43>3t1XKhN>;06Liq8E`M54p?~!J z(@DSG!qsd`O!MCUF4oHQ;0ts;8f5*Rw-KyP*2Y}MMELTaM`rLx!d&9rV-;oUU`lx! zvX{JZymMp%1@an3k__KT5BmMq1XODb${XH@D;223>ASUccy=H4d-jQ$niPRU#;IQG z$AxNVOBL;fZB~|ts)a351XFt3EDgl?dl4p+|I84x=ja>H65X4gAw%gtJRXt|! z>yFMk?^=L)a(t}cRfxbfhA4}vqNevn+oW)44cVL;`|IJfq4FN>%jGKYUQnHGFg$W7 zOunqtlzp;rB;M&wRAViOd#uRS3Qs;1LE53U&}skX(wcqi&N;B3DM3vG_V3h-Z;GX% zNmti@N)Sn!#LL=kHN)w8qT{CYN-6Kp=NAEdH+hAIPVcP}M-8S<1uo7Mbg{obH5e#9 zAhr1ksOdPyf7?EcP=gz}ef-la*dh}K$GcS{d|E3G3*~_~f(mu>+aK+EjPfffQw^k<6Fslg%1+R;pXZhu|{o@Y*+dKN4B)1HUc8O|_H&;smkB zG0RFB@_>quHpZN^j0_s~>F?jHcLd-YiQ)#az8y@1$CHQz$R1>gYekyuwE!YiEmYG3 zRN%*BrEdo2d(dk|iC_`d%jmcQ-5;*m4!djq*oq4ANcaiO?R9Jq=Pu?S#VN8%=X5Eq zOJX&Y41EYcpd{D(E_zdQ@ax4Xr9%*1u{2*Pxsf~)UAGH(O9jH!=%-_Ds(nKi5U*Cc z5F0+}J<1!W32_^xjdyj0KMlSV{)amYon^IR5``2{h6U5^4Y=m<+GaSDLWfo93gHAOw!_pCc~X^*vXSm0uhLR z#`FNVx;V-MrvMWv{Py0bcRkr@7J-t62mC;$3)2jJ91Ba{<$SDbaBpFmvb0QIx6?M6 z{f$C-7CS$?ruF%Ho2qfw)mGzfbLrDkn$4Rla$fSab2^HuN(gpovui%BvQ}>Jo9=$g zVNmkb=4T>Jhfunh*l%iAt)QV%vnGCxEOk{nWPC zxXB0kGQezhggB&pyLMXOcs}U*bXB5-9(MLq2XoLXm~bNrJg5Ib=ad|)#OL(+kEY;P zl~Y=eErLwNS9qR3X%8ne=p;{3o56WF-x}AL3z~EzkT5Uk`JiiM*~Y1+rP z#5{X1&yTm43X`;>fxs;I_jo&V2UB}fOip~c$Y}!qGw5((P5p zAo}_1tCJ-(>TW$ifdpI|){9Sv z69SWgb29~)oMZjAR&0f;OmPJi z|J9adQtg~$??1EGx=;;Q$}1IAzhE5WJ8!tM6L&dZZx(YtUT+?MZ{{_{bw{04lT3l3 znftNzOrgxFIH3)p-5NWZb;4gOPw7Yd_K5{h0H}YkY7Z~kLBWMVD*IPJswt+7u^p<3 z3vM9*1gBy+=~>!I{5@gz@43W9YDwHhl`)H#tjDR0;)gDa1;1YNHIe4ZFQ&Sg|Nw1y3c^;Gm30>~ka z_NT*3mWpVf7rGrYRv?y#36rbWMqSQwJs&G-XXW0Hfk4tsPmW7JX5R{V4#v$AMsyXT zI?vX}J{p0M9iVG>k^tQrQSY6ZbYX;wW{!;HtXD3o_iO+rLX1|lW=~)AAN{`7&5;lb z6kHtuL>c`HyiB?;_Ji6>YLUc75|Q1aFK)pi+> zlZrSxtQi|UH*2|#IcyTYu4C-+#-cC$7lGcwp>*WDvPX>=B=u`&hKFLu=_W_2K>Wfr z@yFxq7iC0$EEgsW&@mh@|0Fofh|4}qp$cM|b?<$M1RPKWpu_)QLz0eM{lyp%whG0b z%V@UoumniC5Y_TZ(`nX(qEbS#E~>B>LqlbsdUyuWXSTd_y*<4T{e#T0fA_9rs=#w; z-=twJwJzp&YuN~%oL&0!r?TG104Kk>P%$CSwMxlL-m6@{GHApg8}W8$eEJ*Kk)Sb zMDhPcLbd-_*#G~{k3Apb$@SO2N&!|@s1*#Xtn_yP&|0}c2OTr@s2lJdW%9+YbEHUP z)aeHLzr5+ue03B8Lcb{7*LYpePv$r!6BZQM^k+l^39#s{7eVLKr$En!rlR@p^Rthg zho?6fre^Ra%<8VR0#zgn&{W~G{+Y;CVG~L_+2&82+i?Sb$ zD(2oZZf{AWU##?`tf%YhwIh4bpQ@L-nplD^stm=QT9Gd=k0uZRG~Wk$8(pJ1Fs9KD zI)XPHi&OVF0RYKAOhf7UxEN*41zlhM>9!9hE^Tmg!8$pg9jrK&a~+rR*3CWn{Iq5j zS{d862@y=&Obf6#`xJ6?vtCBT$7-0v>zxu1^KUvWRfBTpN}TN4 zrKed^p3h?$B?NAO-wnbaJ;ThvkNM(WeFeZTa!8(_Q;jD(x!k3>qJLBJx^G!j%4cuE z<6$-s)Q5rBav(XLu!#ivg%1js7y^KU)-O@2BnrfraGn%;=m(Vk(Z}1<9|GoqVtsUQ zt^W=-YmonZKzPOPIALD-f^_MSSM6D{nuJnpi=6d3rrHAEcK$_v5Ng8l;u_SFBV`UA zE(lUus$F+1O+GzeOMT`V=ei*HjUyy72@XoY8cjdg;SYXk{Qigp_6kzEhIX;-EUy&@* zUe3>7QNPgyRFkLZdtvpEtgicH!e!3T4|tk*TvMrVq$to8{q^X>v+*y3a*ROCSHtNB zT0SGf%abch;x50xLSSUaH#ME=r3Jj(O=}=Sl*h~-@`pll`3+5*3!y?f(7}|s?87jy zNV+e-+ib)@keCjSehHu__43ae=<~C0+OXW`hwoNVD&4)uK)!6zfd%=U>@b}JmEFs4 zF+;Je>%f7lrG3LBsJn~EdNpB_g#pq!Py7yq}-ku4;-L04zL zYNZ!LsoxuX&|axesNRsaeB*lZzdXc;9#9sozY%d{7oC~v(+bvxk&*9+e$3?%?z>!&DvbDe!} z*68Ip*T8*!%4>^wJA_yp$}eT4lLw(gRt%6GVsZYnhus$t;{?go^ty%DB=23GUE7z6 z0PQ->R-$}@d~k}|4Fg^h%{|k+vFSSMH*3tYhO@Zh`0kw7@^Uh69XY;L-)?s6MeMQP1gA zuYG3Jr7h-n{{&y|3%O~6)Js+jjHK?#4Pvhzq}VKtTL?dTn!%}d=M#`jK56QcmF)Cz zwsU^g8@19mc~gm(-0*2%Z1yM$?*+tk{fSaYj_#|jGk{Vf(!bKx z>&I`>O!D|s znp29e0j-1@i!Ue|H`>Q-bFl@(&RmFgf*YfM$TV1Uqtl-a8LPSFGB1t_weV4 za9*<@hgbSCClF~`mfFRac88w-AW&hzShzWRwo^M~Mb0OHEsQHL0#if$9; z6A}`>ho1NkQt%H*#(DBGmpE8oaFx9re^T{`@`sNEJZ)$@Vj)J@Vp8rk<6JKVjJwOD z%vqDYHpNwYi=-&iG$^{wUdA>fbAYm-%o zzhK>Ag*p{6vN(g|q|24M;m7uuO>-Y~av>DZ+E1A~%AxpYU?Gbg74#;5%u>(3^>oR* zMWhQ<;8P}IVBem_M7-C^0(!g!>$SwxecnWM*DmjSO|rzZyjN-|C7+vs?wcPFSoLc!8^jX-@cDmNq+zw`Qhwg>~g#`5pgr^|m!PN$X`?q4{gc8Dtkk0UmNH{V;^QlEEu3GdU}m~4rn-y#$zbaU+G zYTvpgxN!fn&LL|(`f+>%hb7w!8vs2Z_NpZTFba+vqi{0nII?ZQ1|DV}@Q5S6ZZ=8}J(uSct%Nv(a0aV5n4~#(k`Z1V7MLWD^({zvFm&Z>;HY9LLfLF~y z$$NKBR;-%HbFIt>M0UV(wHHyk-&;E}8)aL)3!F-q994W33;gMRw{ZRBKolV3gqvmSw{{HlKjQ)W-T}1 zQrgDZY;V!>r%15qjZJ|2X}25-Y=VzNfK7B+_5rzi?951-p-c(D6AGWGL-kC*`O~er zrxM)89)Y{}>laa~V!Ct9K|r;(epd=hRKC30hE~yVjnGr7Yg}EW~ldk=th~ z%Y8}qHC|F%69Q~n*@N=i0}|-{%64*%Tpc+Q{-m=y{5rXN=;*;aFhSnN=J5!G3oxc| zhcszJ4yf{8ZICBjkZ9CIUk?eK>TLJZohi6{vnOUuU-08*sDRWtbm#@khutPKF2f_F zEpf2Cg-~Ew9n2>&vm+ECR6CVd=15}EhhWkZ&WKPB$yCO=mwKS!+xcW$W7E#(!{!d+bZLMmg`S#S$Y*X8eb`a_0HnB!2XI+ks$9*}M#TH-1e zh&vIjvEX^@xoAAg;C?#evNOjbJ=qeNGgw7>vp_w;fip;pNe6XRmID({MF9wDR7oF_-OY+QUUq`I)krS2=ts4Cx)JGs@cAGT zQ|H!?6sTLBv}?A{tQVj1R^Ta5X$u&{zIIBC%8OXaQiO7+W_*isdb=5HTlpPN=u^x# zK00bpw>*@+Zt1WC#UFCmCu#i*nku8irN0ySQ^r}Mckv}=Qe@wgyvCU2yxZ?A$CqSZ z5(CDTNBW0PnzxD7hB<|-Af4;%asb>85ya8%!X~a%e$IMPn>GkTM{cH^e97{6ar^ET zTW>BIj)&a(2vijt;=8NGV$Tm;UGvPIj34tmg{H!)=nKC4bbvq+%uiAIRB^Vi@px~$ zhrA)a#fvi@7S4lB+Q4B5g*^6k9h5!MG?a(ftXnJQ$#>%R${uuc7$}pa5X0Sv?u2i? z?h+p(g9xDRf`^QkB^f@#9cF-{pcJXSf=>(Pk8pZp857QP2)Yl#b<wC++99y=KKDx#o z%BqQZ8(=p*=@m~;3zN~WZ}SqSpx#BClQcT7tPJ~HpU+?CFq~&;@P?Ocbf+-QFvTq? zK~+;U$y=a0c;*W;ZdWDoJ@PE*;3e%N_<8i>puy$SjF4 zEBNQVz|{6Dy{PsxyB5M8A+v znm+>7=h=x3U05fvI8`uxxn8s$6blUpIEKeYSC{ZI!^#&WF}f7L3;*KZ5l&ev$+)?q zaxtNwvy9@oHoJ3O;>jk6Cg9}sL_^RBJj(@CpHpZ!e~8nY%}^}S8}}OaUQ7cP)m^{f zO|VEcL5D7}9;%Vc(ip(m#O+=522J2k^Qk?f*Ick4$Mih4JUb zttBbT*-#NkbiFp=tCFXA@59abwX#;FZ&Tv@TE zI$RJS^LW$RLaOO)e~I)aVXtxJ5YD$-aeKM$EgJ914N%KJF&7_W;g47d=Ovf(#?uQW z_t?iwQcDd{+Rq!48!|idprck+hJugwOn3+vJrxuh+L6pAbm$yZrx#MQEqH zOYUC_;8hoElTf@YRa}PD_Ik`+`e~g{h#CyOzj44dr~veegc>@Eil=TAa)Rask~`3C)HcJCh^d$aVTYGpFpV?@ny z;bm$&HpEsKM5ia;bq-i1?sOYAxs0hcd;ce)t1*Q4FV=n3YQF*uP-sptp)-zz zSb9I+Pe@W)$oyD&P_^)PL9`p2c-aGu02sN0zj>Wpeb=d`yZ>a4sa}Wslxu3clmC;_ zUrR3+#pLEgmB5M$56ZOy@%J*8zx?$2V&U&~_NhuIizS`|9dBzG_J7xxq)BBI+cgc8 z+_VN6cmuY~1W_vdWJ(M%a285tN`)b4aWn>)jnf*t3%MtWg(PMqu)racCggECZMPXl zA#{$BqgpXN&*)W_OOZ>2vP#Z%w|WE+Ptp`wbtg^|3b?lJtpD+^pw=JHoKt2 zI@M;#rxM{puc5tcIEcJskH?;=bDrq_V)crsFK0C8S3Om}H~5Px;ra$>9~Y zlAS=lDPgG}oAHV;R0=Fu8)F%R^oAZc4OC%e+@0&5KwpGXIa8zGEd4}cUSjT5g3F2I zHt|F+60N79<7ah;%z+tIq_62y2REJ10gDKw+#*1yRS-QYiG43Uhe(i3E!XS9F8c90 z_%LlDDcj_AK0HR#Ab=xA`e4|LoEE!}P)2Eu(aZq2Ik%s8?<9px8=^b@`69VzKR_&z| z;1ZyYD%Ux*R~0iZYdLat@t*&Fj72PoV3W`tH<(gnacmP<+$FEyn!2z1A3Gl>eFUkA|`~V2Ydg9`{%PFc6C~#>73Q`{l_< zaVS1}T-8iuVZN#@5t%kG8R1hx1R;&XJ^ds3PMz|=JU#TegXZYt4~v?ttsN6hX}Y0# z;ax0ba9sLk?mR0qO^A=9Gga)6uB~h+E<~3|U8GhbNXMIpVG80yL^Mgx{Tf#x{eoe7 zpm~VZF7wMwdQp}3(4%yf0%(GKhf;`4kd`ODJU<)Jxgls;rCU8DHJXH1*1?pT{}q-f zk?={$>Z0#)W}W_iMl_Gbf()4j;0-`q_Y=%K9fi^Z!7lHT4FT^?H!hCWDiw)R5vHZa zBano%QsyDE8$S&HD8x~}gGVb0f2}+AL#ZIeDJtqmq*Le#)i28BJ=xh1LWH}+qmuTM zrrhUunSEis?{PzKMf8v|G<&46B`-?d#-ZNiAM3oJWzTCoCxswV$fQa=}fwsB>wLd|7wn8PPy& zO&7rm=Ssx;$=&5oa0*^hxTj=^-nTRXrmWqFsY~9oa~6!;H!`4QC$*LZ6U0liaUzI? zVH3BlC^qYX#*@AgUu}e^fmuf&r3j_I8xv~47|0fzNeFaR55fi1H+-3>&em*)!O|Ow z<1zN>JZ_UBdU9}-dXg4(gpzfSozf{c>ypPby&A-uPQDF{nacct8cKqx%Y_5US92r6 zpc7z{ascVMg+G@nuZd1i^F+>_l2%@CF=CSj4X?m77<&;m{#@U#2N-{(7S%7eK=&{Z zHgUV~Z*;&yy8En%-f$~Qd? z)3g8R(TX|wT6AY%)k8_QrF6GhR5_}SOYIpcA5aSL0TBQ6yc-w{5R2}TZDoxUWkA3U z_GHPPGC0Yo63Vw9+zjsAj1D1z5$ewwK0EQ!JCDgM0wed^V~E2{Ykr{M(4q9=RyB@P zx2Th9e*6`?27t$munu3`vSA*2W9M5>^#GA+qFRn?)!b9L-K2eAJ|y!w@$qNMxTP7c zZ9a=LbyKe?+kcI6&pRB+-NATlx2^CHkCFU@RFe{L=N}?LJI7zeA+M$+n5VdL?=dV{ z*4L|tQbyi-y%*t$%7S{4zV25#A~H)QtbSC|wZGtBtvSGPJ|Ir@2^kAa3LlI0vh`U> zdc4K>DcqeKJpRJ6@SQ)z1!H>f)M`abxo;3TNNbQ);UDV*lgGuG0EWn_?HJSR$R}9> zLHd%4KDXz3MF^P)RiGAG1r-6?PV4^2xn^MUt1At7$HM6hRxYgz$qrd&_{l~te?Olv zcv6ubY;`|nz_Bo{{~<0?6$jCrYM#+8kHq?BcR_Zoo{f*Q+MBmM=!#*F z><)o{`Rq(Gt+Yab!x)ScOkk+apQPfxC|>CzgKqc|9cNb?73Dis+~!h{pUr1>uOM3b zSHP-hu39jT!6FjthnswZXW|S|z<_9B04wQYlRGnz`fsHOH4QYwO|{L()0+Pj{cHP8 zK#kzo$~efylt60cNgTSzjUOtZ*>u-;9ym4|!VO~rtn}{P3A%GNN`**O%Gcnn<-?zz z$}9I$4Dh(&RKju;387l`uOnQFi6bmS%Xo0*6@(hr^rk$3fy)n@ZyKxh9wehZdgN8R z0;~ONV_a5GE3>m0#}#cUg6HS2j=$BNt!9okl>A);vs(nwt{@B|owE0m0Yq3>H*gl*c*yWj8L1Pfkaqn%!FmSDtMlVLGU+GC&zP%iC&<6^y) zugsHLi-ofb$;Iy}Ow>0qThu85K`@^ zNp(yqMEO-{KzA`Y_d_peHt%3|iZF3aN>M_XAs>QqtV8keEAX_vkM8|BIBXkbTlMhl z?|c0Sr?eHSzq-vC-@dVT1z3HD*9*bOEOBvoCC9^^G8V$}UOx&EFl>tO)I+u%xI^2M z3946(hnZqFXn0|h1o$5Yf;Bqv#Se#4B6svXVDF#NA$ls~vNYo9cYk^#m+FSsK0MQ1 z6z)aGsB)KiL64m|u<_410c^ux#n0-3;@w{z+5jqt2a6aP=LdRICyX_@m)@gludr?zAo|vih+2nE= zG?3Xnt8~HY6JMaX2&|WF5-8HS;g9uN(;DkRDuzu0lU{NW&S+qY(O|9p=&hot+%#0BBz{@zK9`lH&A!w$)%hJ}0XGLpIe0>7ormVpj1`Pd0@8mfn3@MfNQSnRUbA$G85W4$5>LYf8uCAUe*STH3 zRhgM(pdgm1eytIe0(nE{f@s!gd}b>`G7;SS3^9;o$6xwWZ4^8VBakjf%~7GE;m1R5 z6c3t<-)6KkF|6V}q#vV$t1qk5GEALa${gg#?91lg?q!GPv1Q$+Uk>2l&mY#juh?EK z;w*tfNoI5$=F&6kJRWX4dcU7~_ja5wAU0URmKmRNx4Wm4tIXpWt-5{>QF1peymjd@ z3w`rV%qR+EOqu8A%qwl4MG;S%Pne=~20e`7jF5Gv=(S}@Cr&fmm#O9-5M?uyI|h@a zqQ%;($315x<-pzzUqInB%#3cbl;gUlMX z#~XWc2lq)}71hh?TuoCaFvW44Z8hn*QQbi7k^9w@3cfdwDch|A4IA>6Qc8A6zJAn? z@D;ZiK3A4r-iu~RZ1x}vI`@PMN)`Tj^CfSGB!;2HImms5`p_-5E|92+@Ew#TC7|{Y z`?jK<9F^B8Qa}DB`WzTeD(nZ^-GyQfgjB4;+N5#jZ6JoI`nTDQ^ur6B&^)%x#;-$< zi{5H<%NL|nID+lN$kbQ+XWu&1QJI1u%R;YIQ{R4Q`qX9%zj!xM>S1|qV!opd2jvr3 z;f@L8i9!B2mb|!Mn;;ZyOSAJr8O#$;KgIxViH4FpQ(MslllCH}-Cwnx;DPZp!F;KT z_md1|%q1cBI}Y()0i`Z$LT+*b?V%_|@BLMoYSL6yXr~uyW`fSOauAPK(?5IAq-)}= z3F~=RJ7rrZHRNp@Wr4^Rj}=ugUENhlQ$AIA8E^Sf6O#uQ2Ylt!ZO13jWpnVrK5T=V z_jt|vv5_qP4?BAJU%iaZT6fb!-yWYpvpQzN7tlHQfv#i6K0iJ>`~l0y3$O+1Qh!k? zuUD*74!cR|%rX7cM~lYwq#?fd@A1tnOeQxli$tPdHLQwA#7DT=pcs+(6vdlVW^N^+ zc?`iDbi4*7Zu!G%h}0dvKAmrj3ZZS5AwbXM3+c{?Qd*knTU-S~hR&H3vYDdjXLX}v ziIowSTuV;l%0mi2lsVp2kT8~R8M)3>3p9jQ@vkCS9J{}O+i=E}z1bN00QL~{`XQxz zC3(Z=mFb^fW_>f=L$WRxBpY0+WcvVy4Ca+`aKna64Iau`jMg~sEwny>TV5{s)+Uu} z&&bB)MgE8+rfG{e&FHAliw=L3P{yGCq*T)?7aH-2;%t%{j|x&5fOk&@?WH@WOiOmI z{iw6L&WRq-Y4>H0`I9bfnB(uTfqVPWIopNL!vMvBBA4{?@0S64*Ty-;pxUZznPiMi zpzS9z7E^HKT=aqXXN|?`$w7bSDeYxXH@VV_d_RWizD-k$u<0LFi*-XdaS|^F?t?CWc2mn16CQLjE8Z4QFewiGbcYBbr!mEJdv*OsdNzh& z`UzXPAdiO}@EOWbo-hP=l}x;B{`BwVT*Ia`7tFifV5Vu3c*86Hp_Z)8xRc?O7^g5d zT0Cq15Y~u*o(b#bCv{#^a;dFVR)M%OT`cdVO|L#t5>}DQ#o(c`{Azhv<`wGsSf|_1 z59*XmuT53Yp4!&~9jJV+LnKorbMq>AMIiC)H}>}e!veS(W*WhtdD%02c~HOl?q}|^ zRPY3Jw((B4TfRqZ%^)v+A(2ZoE{G>TDpVkj$8j=U<^@1D?v=sKp4 zoW5_>H~}}{tl&ei$@si+ZUd5JzPJSCI11zp;Z39O-X@S!|qDn^hdX zQZ+2&v-H~dIa-da*_$}uHI#+u*AEant)+IHw>puU0^mruSY8{E)x||-6OR*2XoWxB zemv@iEq9JZw$}pe;M@_6@}H!rgEeQh$fZf~ymF)T_x!lYNp#RC>`*ryu# zc#Osvme06rD?K$hLx}Z3W=-kT#;v$9?Spe&6og(A)P}9N3z(>s$685GNyO8ix1X0j zvCf>*4lG5M*E177AwH?%ofK~qsZFz`qvl@8Rdr^Z(7f{Ptm$CI~QiZsQlqpVMd2ilt z@yben7MLt!(4Dm!d14!Mzr5wet6IXDhQDPW%hiL1PIl*Evl~(p;}hY%5i!ZI1z(mp z@+LdbH5l96<-`*}};ABBW7jnT;D4!afYt3nlRwaM)rCP(t_1mluPrfP^$ z$2su2lz6;vRU^bHr)OBYsZa|+njEPsZe~?3eZ3v$grtN{gmop)fA+FHZaC=SGbqSl zN&FPiCC)~h;4C_E|6KE4TOi1Wl6<-69>WmCd5XL?>QzLzAnmo*|epX@i0m)s7lp?SgsSJhjlmu^>+J}IZ{AS&VS3Mq79E{ybTWg|A(lAJ zseIiyvT|52o@GDqa5{6>wW}VJRsH=8dl&n55ZBNW9t*<1JPEq?Zn%VmwrzTF8W-ac&&f9>ZK)y z!SeiO?q+`qbI+}*gr<;$KYtRKE~1}5pF7d-einIz^f#FR{&GUr@zP=Lwg1I356pzx zpDtvn|G4V2f9+`w3fM(b_sIKd&tlaik}uhO0;VW_SHo%_G>K`J`8`R~x_G(agH%em zvJHaNXz6Gab8Cz^n#^#_-&qE~0w#{+sGF)}YEw2t zwwL{@9` z%=!LpdTmWdFg4eZN242yLb9W@v6s!^buIj5&Q+ZY^zadD~r;Hd^a$;iDk(BuOmQ& zkh@b490q5`or9a-nilsAu2=4!9_ufD8UVHbCCdMSk97tm7U}TnuqrAvv3922aTiJQ zep#0n>YYV{>x$4e(k->U`Ki6d|7+i)8*VS&W+|Uw8(LMPx=~@j&+cYed&2e4c5I97fY$1bCU0b)OI~^O{d&riq>YT* zNA~WzJme14SGQze{pLrF8zVjQmq{`ObLk-y;HT4J<}zPvGDdp(^L?t<=zllKm~u8- z_F)Mp7t1Fb30GpC+;(_#+7C@n5Y@Acnznd`j~i;tsMkKUCVvipdbOOlk+7JojwK98h4@#KcL zbo?6o%O5jRt;LFKPH%i(47^t^W$iw7Ztw6>j8et*&j%>zi(YySj4!_NTsXtMK3vTE zL7>rZ`MF$NgVS0MJ&~T|m@Lf|ARXtmxo{)dHe&Qj1&vqr8S^!{udVtGSM%r^d40>y zJUa%E>c&g6wU;fYOb@2ve@1_AwqWR1sy-ln9ox(7qwuCGdGGkB^z9Q)5(m2-iWhFg zJH1(0O`T5pZk4of<8o9P%LU**E(jLvnOiY%xgdC0&^kPkI=W)2tL?(z6a2g1SIllF zsO)~16MuyjxmIl+wmTFZ7k@L^q}KgRcD90VHryoXc)PKtlKb!l}5dLv_cSh;G@4fx3WjU*Vf(XZ~ycgpvJ-hM*y9^%pUc6OcH`NDW0NH;)>Qic| zUG%%pq-1ie`KGIMB3IyHww}J zb?H5bnTH|(N^W%gHJO)psotsYPj5YbwX$Umc~NIae%5hqibjClbSp(DN$!$Fl%#2x zcKxRc$59m6k*OSb0JMLyxrFTA>Lj)d`g?{*z6(M~FlKvZ_r-QBKP_u?f)5gv9S#N{ z>u6O`BV&>}Aw<_P3Ar1DPU{$!7Qa1)K5P3pM7p3KB}W+voGC5gYTgMWP< z4D~t7#e{Bvn1)>E&58vGYW(~mLab4rAO4WB?$ z0TJua(~}Q^=DXF8nXwzx%y{}zcgYtv3pUC84;8Tj$SEtptv&+pF&uQ>4dGIQET93t zy;jK9+LTRoy)~7CPdy?mkjqD!!|lK-G7wh`KVTbyfCDp>1R~}<@i9Tc+v8WAi`B~t zFqe^@ThxtI|DFw1{Ik{y8qZBhkgE+nG5w}1HNXT>C|AkyY7l3bl-l6dWTy{o>VKpC z`Luj<>Gm%azjNpId?HROiojUogi;;GT@TbmuX>~Cjwq}*b#&8Q;}PGW4~Y^YYGEw& z=jG;AsBYyGCP8{Q1K!rQ9PN1QV{!0wFbUF2hI7c?@vGyAI}O^glA`!49zr;Q`I>x` z9)c_#;9O^s_gD&&U+F)+K|5X6XP>lrwWx5I+PA`~Dgk)uqJ-5U%j%OeMc>GIK_y~F zvwJ@Luot$1ppHbJp}AN2@O<{5wbl9F-u@a!7UP7%a8Sm|2nwG8Pk5tHUUHNYKQVqj za0*<3$x69GM!DIFCMO~7V71A(ngN2AjMkz;q`X$=Dge{8s`cMDNoz%H<#7IZPDsZF z(sO<(C!Db2@|rs*e)A*u9WZ!f6T86ZezU*Aam`T6R{2%sJu8izgNy@9tJPpi7P@Qo zdtcq{F?d&OenUy?si1vWIbCydUpyvimE+wH1L8&P_Ji*LfJ__c2am@q5Q!KbCWK-_ zkW(DUbmuTGMFkw6+3j}%ZD7@8Iym1b{Fc_NeHI=nBmhW_fGsQ(@fYD4^*w40lqc7V zn-dOTB{v|bK;tdVbUH(@H3~uTrDUv{YZP02;k`~P;>*n_Jl}ZRr6)l&xH*U}*oHrX z`H#_ig|Cj#dP2rgR2~1O21JhxMX#q0yNcC_JYKyLRi^Pk(*+_(Y@ivT;a;tP1H$l4 z9d|qRZExgr2Uk=Qv2slD;=cQs7?#$cS!+tY_sSR*RRgg<`x8adQ<2Uek~s`j+oh`K z!=*tnV8nd>+G?6{*ZFi6Q6saA_dw_$1pC3*o&e1+(#22xuj;Hjw%%Ek>$8jBrD<3& zCwu0P&#5X| ztI_#TVRNTE`;ni)iW6%?7G;&sd4o3`Zlk}18|r_sZ`Sz@g;%l$56Q~iAn%CRPjrM| zLa+r&iC{rFBUF!0fG8XV3}sy^x1Wssq(rF^=2Bmwg%zilM$ri<2ketP-W!R|D&X|8 zf-`H(a-_a#fvYV?t@KlO&E_&TrIaUD^YA@( z#+_`8=HLFD`$_8s7SDfcv=dfasWv-!&%=JelA1ue+63UjtIEzDgn%MIC71*`d{&1Y z*T&)-Q5Y0>ey9KJi=L=fT`574NxPNS=RHP(kC0IG1=EhFwB>6zHL zMnP|L3cd`NZ8{rUeYBqa1teyC&aBO}6KLq_(NTs@f8oYk?b(`Bry4!ovvtyF=%p#%|wHcL7)9@o|-Lrn^NECM8wg@<#2>&3v<2=o9q~R>3@*4_*Gv=b( z=U{H1rD3Jq7~v77HxWTjA87t@!ynnDg4XANdOO6j!U{oh9X8TEpYvk#r-;ON$J zP4pa#H8%Vg%WTKy=filEhE@D1C-|9}9yoq0lWV&cseYfIERLCuuC-E?QhbxSzyJ; zZpwQ7vsc7hcSUlRNqw2Y;v<~Yh3tZ2ERwq(ba?+jK_Ab&#zd@-@e3C$ER{a%QdT!%Cya;JV(&b z=ee}GL|$4a`&{{PB@n>Ftz-G{d;inWkmP0D_*eI{u2Q*8gal#bhxP0Ja@0L7(wYER zw|JqWdj<~0Hn28{|7-&pc-s9`!oSme={6A7caf{PWmH_u@sLazPb58#C2p%EVeTEN z@qzEa(H^&{h#N&%Qu6Ivdofi9Cq2v0$ZW3faST0Sp34L7q;zw)PU4C!ZUJ@V;?9ZwevtXrCDjlso z($(k@cz}Zy6&)8(Oi#3BsT1tB-U#71JPS;8h3X$<1?4=~Kxfeo`mBew0=XbQZb0_z z@)G%_PTXtC{Fdgc;Bskn14~cO>gf4LslP4!ZHaqI@Xpuf z0G=>iF$j<8Rt+%hE5mnqAOmz)^VOk-h*m7l|02+)94>G%1kwLf-ydJ8KtruBVP;uwq&9 z9R0w!u<|v=X`Yx&){wbix{$dLd*{>F-)l5W&pfXi$2!R2$m z4pDn;fJub}Pdi08dV;8yuOdM=&;plVG$A}+gLgpWNxy{`f0W|vN76v|!qqCu;pg$ z%2#*cq;magze}nO)iqdhl18>}@_(VTQ3L%=IO=-|`7#7M0?RC<2#k0x++L8gHCj!r zEtuQbTA(iXEDS5d_u)hKc~b{$ZT|5$|E){f`YFTzO9Nm3FUYL@zv3el^plO$xPek3 zyDai5?5%<{lRlfmr|o?ATA;~fLD8qE*#$3kw+`Geq0Ax|vb#|@D-D&aTPy53CJ08= zX%@Nc(_jtm=*dU!?&gDPeG`KaH8*6%O?cTN*{1>t z5t;bDGUV3nfl?CLf+L@n7hU8mzxwum!WpOQIw>*|sEsRGXV=RR6}5c5Jl#zWBSVpy za0RNaiFA&lY+e+LI|v4eiEYpG_j0uTb?dX)cuD45#m%8di_r#p z`Csbd_DQL6TEWS~f6coWOKq5?2HmWZAv6@6rB!$zZ0cjW3e#~aFHDo}fbeT0s45?orwW6x0xisA7@DY=o?=jaQKV@rQFRZ9Wd zp3nECs$_<~ZM-MQqqFZrM_lmzSh)N0-di9TgHcbKN5*9ei-K&0jw(AI#GQp-5ek8- zBywU~8svtm{}_Cob!({F5w&d)q!T3^|A6{117uj;qVjMv$2lmu(>7;NU{LBvGK9%F zf-J25hQi`wiH_hSsVp~=9o48@Fh%4Q6J>+?Rf<%E!jQt*;)Nff^x{t(;X5V^%{Oek zes~UP57g+L8H>3I?}NZXgs#yNxPN4$U&vt!6wdWnp2`Uk6b>i4Odo_W7I%nkstGpV zm@+Q#G);Azg&<>hs>g&^!sIo@U%&!aI}2UXEZ$kl6KaM`s6$X(Y!5tnqfY&m*?|se zO%#L5?lt$e4MJljMmfa{Eb32%#J*~n*Em!L-dm|V11{ZQ#FlUGcAtWk?UBtJvV)wv z$X`$f8-NGH%Y>2~JGgNu1)AOIy;y^ zk7hS7FgrhOcD-3B;&<8j+RRqTTypIc1b^OD$!P<3c-(z~*-vH21V7#NwE5dnS^cqH z2IaSg6mWE6tiJ${GJGg*+x4U*%yNuvqPVEi=^F{v8?!Sc?jbK-DVf^iS1!Fy%-FPj z05$^jm0Uzwpp@sr9@VFGVZ!s(cr^&eE?amBTGDqUv8$tKi{IBk2)Ag*yTPKl!&05yRhBvyqd6qOx7Y;C&w|MQO$GEm?tc zH$3G|uh77|r>FeSR*}y3JD=zCANgSCuhYQ&yZ{o55f&C6UH^imK1nQT5SFpqVvCd& zgyjM0ubj(En<$utT_LIT?5fho=vPHgX)TytuTOs}Fj|668aXOShpOiS zD!xW(kF;x_3^WgSo&`9 zE?tOYcnD4O1S`r=7C+GeMqp;h7r5vJD$8ILy}4U6FZs3>`|W>u3JN*Q`u7}QXF(2q zoVNd!h6`t}+4Z4X`nOX<2Xx6xCyjg}sE(R|UmN|>uIr)asUZrrj$1f#q32v|I9E(s ztHu6dz2b8W`8RE#GNg*G5V?$&{@ua;O+`gt2(8Qx5ttS320T|hjaN#z zpI<`F-9R$k4&{^5NxK>du$h-bCJkE%r!w5u0;QNsPu8S-Z? zne2i5*RyY;UzO5lK6y5MF?Da)$$Dw+SezK1n%Guq($z#VWl;7h>s2j>b4c3ix4jp3OX zEyl{t7e|N=X^=E1sF%qLYOO5nJ)x(5R|PmhQ+1aIa(kdVhZnF$KPM@fVb+eRB3Q|Pen1IH z;OKRlA-nW}Xz-=^({mE$PW!!~nL{uJC;bck0G0gHcV>E=_iwl_j_1tY=oi96UdDN| z_JqJob2ZS>SUtnyTP+|5P7CM^fDFt z-P7pWSn?r6p`-~Q8gzGVx7B!b2z84Qd@v8-UW=(^n@v?qP~onJ=p&U>?F@$M-530g{mr&MIk}gO)kK*Wv5Jta7!HZk#}S_Z!4{z?NYrO3i>m?Q z?~u-M#PF)~-?;(m94=U5i;r3<0}m5Ur=OJeCFmPfTS|D59i=X<+}zpqS=X9SM9$1_ z{P0yQs@eQiqM#%|Y+Rn~ID$g$q(M@U1;5aCR(cz8xdf|+?jzk%v4=kat4pKE0ZdG7 zDve0fl;^>mgz1Stf@3~@C33mM~eV>~m69M$Q)xz1NHsiH;2#*Byk`pM7D&@utblK-t)^Xx0mm^C$% zH0WxZ9|_{z;A1T(OAErQ|3>#4oooK@1c|7oEen}SN|OFBoR0rhW$X_3{$HogoR%eT z=8`jz!@ZN3{`48!MO|m3t&ti8`v>GdghcZHuZaa)U(x@Me(d7Ay6i~BO=;b&Pu{y8 zBd)`%;sM7Cy4q2~cmiZ89fR4fkOTMa~(g@$9Lb^9pPHElD+7 zCqsK~##bvr31DG#MJ_xE0`~sTef8S7`26yld%LVg>VvJr1je^k zis69qf@pzlcM-WNhU#=p`uEJsQIPb6M{vlh!ST>>x1Ul27}Rj!N{`I~O;Q2nRXZQE z4)&BTocu7%;c~Ug)@th%*op^Y!xug+3jxAU6@||K*;qNxtOal41}dngx}R~OHVlmL7{+p(!>4}P+@Ci`O9*BfMiOlC2?=cK9O42jYvxpV7Lq3#Yg*YS;E&XdYH2xtaiDWe5dF6&e zT#F~Q6d{sawUz|KuncmhnItF#pXc%34}#u79i|ajLT4C(l4E`?O-5JSLv16D?-g%R zb~~4_gt(ogTh)hA!W8PIptxyI&_eSx&@_%jaqFo2!Z=zor)|x6BLTZ|7z71a*97cS z&7C_LhiP$|2v;qETH6CMix6XhD*HAlzQqw4QlQQ5=gP?X8c3Le{rz*pbyojAc>oNjYpdE6&6Le^%Wz1$l(w0)J^qC0nuB2c>2-A?^yK zTP(3=Fz)g?eP?UcMiJCggk&?CS^exgyzwx8prDW~R+D+YAILHz;vJM!k64sHYiWND zlS7H0@Z8lnx!I7Pb=gqp%)G-5jfm#!RsOo`9w!C@;-N#>JEk4v8s} zBqUCu(?lFHv1oeGv&h@|>TrHik%POP0I^oA6~Zx%jHPP@=q`wHb{vqMd01^c&wd&` zk>2_<*9`~0C4hrAsRbq%@S*hTfnRBk5t!o>i9gXvunA2-p5sHeO$O9di~_o;*Qrr?>EeM%mZ0B=Sb>aAX;{Nm(v#WI zBk6tzLg34hg^=>RPv66}9;3|Sqb7 z=F6LVVyMK7msC1JlARiO<(dq&k0ZQl-mFbnp!1uZdsVh)X6WrSgo6-GXDYaDjmKo6 z3~w+i$_XkRBdwDo1vL#PW94A1L|)p)%nk5Quk(fnEX}-Zg|Q9LM?WtE*+wIf_JGF8dD*ak%OD$5k6T~R8}Nw`R&KfOEB!hqsa;3fh9$+il6Rp{)b&qu=)zueF1X06vdB= zjEa1Wy$M;t)^zrqm30@z?$IUufZAT?$Jm`eI$mf#3V><7xEG>A#*~h>o2X4ewvYRu2jh>yF%phdjz7kcf6; z9paS2Q5GwtFq(`Nhhbv!9au%Lp1pXWnGMQ&9XK3a9&g{;Eh5(Nb%+ZGB2t~0GwK4Y|KaP&O zwLJ(6E33sSXaS@~vD^Fz)2jZXPrzLEKyGZ4`BPRx5*h#{)RXT!l`r+XDVV;W))CV+d z2O(cjH&(kUk17976Xbhs{vL&v&o&WtFfcvk{;o(ePn|830nfdBeh7|C!->zxq|Q=lR-p{#+grv22=3~`{v(8g)y85yTsfVzTSjdhaEam$Jxqn zq=LLuRz%14NkX0_2=q7U(G3rz_wJvLP?ZSDA-XgZf_w{_LdU-CY#}EKN7-D`tm7vH zB5lM3)CeVcmK^{z z+elv}zb-)&unH3to8!3LY23(ce3>4>R~P@(2Q!YshX{w8Gxv?16&(Y_sx+wL+M#5$ zg4t`q=q%H>M<71bWL@*yw^xe(LKDAI_TXzz^2WK=948dRlkE<>e-ti6@~|9hxlFW^ z1H{m9lSr@D#i|cpFs>mEq+M~!XFCxGER0i6P`TS+qCxHT8c<{sP}+i+cC^`Un~*xJOR^!d5BWaOK8Gq8^lVtG38EX zSS~g&ERgCd#`5;HmYL5ljY?Zl=He=xFdQ{7MyJE=ERy=Oas{bZe{+X&cKk^8@P_}H zW6xk4S+8}%sDhoU2`WcHCbqW$Rci8{KXpdMC7_JieG%d41U<_?EY!*0*-#@A3!B)i zx`Z&?qknH1N`t%;UQh&i^cw$6qJnXS~Dd<_$h9Znq?+M0WWdhfJ60P ztXm_Ia6zzGZx0K6Q)F1=i>2?r7ifYP9Q=jzbV7mKBcg1Lj$1HH2>wKwg{UUNwkYH1 zJ0-cLAH$e8qW0lfdASAo)+{)S3uFRUyixo~tM{x^flL9Bj#6yLPXK=6DVT{<20L-4 z33XLfW$D1c8n@^D#dMT!)xQ`uFPO&|5ftf?-ii5zcRa*<;U$bmO9ujIK6T z&uvK}nQVQ%1ZTJKH-YZ^GFVoE=JHJZB~j5?HCMYIaZsOLhmW?yP>!1bO;8miiNYXhunv<70g4jDXW*{|%xW&;-Ope7C_ zDy2i5hpRk*jw*(MSlBjY3A#}Rm`>1rih`Mf4Ia}8^UmN|JPH}_cr!$4YHa3$BPOxl zGiY#Ol#$<9>JYk{p*;vCzv1rA?&_bVTCW;#^_wsLpa_kAAg6^&GVV{K($IpnQv=C1 zS;M^GG7&NvDLv^mykF=mLAXl-E==*)t0SQW+8llQFhd4Z9vsps&39!x0Hqb#?WMRm zAK7uo(H<=@Jg?vYo!I8D7Y~|8Fq>apnL>$y!*|wden%WqUz*@*tH66)6c%`~`&1yD zX_CuD#Ca0i=-24;CwyH-ah_}YET~8K)5qvvt6x<|ICnCXoYuTS@61oHia$HeHSjgR zr{Dv3@Q#4ssg!=*Z~}W*Wl`V!H}^N5XEYnG-=?nEKpfc4ntdIZAfoG!)CgM0qv=ZN zZXJDafy!V&<%?>*ioS{}6XO8YVoXimG>1ud@6Ro04*%IE=Om-mkw;--dUR$zuIHldKQEJ*Y*n*{Ct_#u-t8)Mu>DnrFCL;^+aZ9du2x7(1XF0twIw& zXoJ7XuEco%07W)?>L+mu9)iOJ)SABeUt#u9V-tYo&7I2mO{+>_@9J4J!_I@Y!Ace8 zg)wWLsmjpiW*nGwOI`W6F@r`lKFjKvqjKi@ha*hU1(e<}=xN&gj|a*pwEhK-p_L`} zNbCdpDHxiu^ks;Ao?3V4+gh}f2wLec6ev&lD{+6~(aHGT+uQZ%8&8C2hw4} zjJ=IJHSajjM~jaRgEiGt{1oX8p@;HTL=~G4!#xMvmxDZ&lpLHt@WbbjFOUK499=9% z!8n2?tDzvKH`ml=zcvmOBW8i%)Tk_|{j?KwuP~l{p84H^T z1(dx0pMm!qy}RpxiIFCjs^2t1%x@FEH7OV+_Uu2n3=ps!bMIl#eUYAi z^0GZ;{PrqN3k_>RMatymWw74;TAsdnGT_W2RAr`>2PMxYvntWcU1g{p^&I z-QHa$5C$_~y<;8?KqQZ=1X)d&%w6fe^A#kYR!|utE`f=ae$N<{2te6c&DuST_Z@!!Mmf;C znNZ@Y6(yR|;nJ76@Zh8W)q;fJ9eXLooW(?3HdO=Y{489jyN(AOv6JolJ)U$cH{|Gj zZqCmqQ`1l`@#TCKP1WCDyP#630m`gdZm z4IeN}1K|F;SaS1`5?}4H**lG81`-W2g?w8qil(sIyE^_^IY}s%+(3ym3Zdgja0!?Bt$Qdf7NjioSX=ktA! zCJ2pYJqR8o4-;%bxC%u&8l`hKM7BsCa*G|Il4(TBd8SM_LJPf)r`@r#`eNI4PQSjM z}UVR?_44FV-4&qCL@D7NE|Ku_Q(};8! z2g(+@8+6d2$qsdykiW1?Lj3}!0_Y9b*<@lz+j6)!H+v>AL|CMj|31G`I#f-0A-ohh z?6Y(U#jRL@w%3twUpV1#al0}@jIQi?OK)yO(_o^d<8_{Hn=t6plDW?CmOfS^Ouk-P z6A{!%>qev3BCSFxi4|miaHN43n{n|5;-f6i-#*4O|S^P@MeyOFq$ z1lmpS%-bP5Fjet=E0Lf=|9)in)9yo;=|C9b=JA@vId-%Nk3HLhxFjL^D(^87bIs`@ zy$EvCiXZuFX}+s#^PZnzjk!9}rPeJO&P(8Xh{F93?O1=_HCWCx|K1F8}fe z0VBfBJ{!~KjVFEPLn~nKjR1Ueo9M)5wjw-e5WzhOwb^;sWGe;Y_;&!d#-1GvbO;F% zjH^3^sv^it*ElKwNMJ!+QXN6fZ5DoKXG}>FTY2@L`AA=OG>^|{)e#Bw93#jBUghyl zxE`7!8Qh8oGQIRlU-LEeDyg!&xHj4S^J#uoQ)`Q4cvG(LMx>Z#(oRrod*6YRB!AZ4 zUVt{*=T`^xVUS>(cfvkwSvg(hkQDJDzG#8xvhhbF9lDZg_YhMO35Fc-<6~&!)!us@ zvz+~MewRQk6`PpLe&3gLtDUy2#f$Uw_+N1;e7@bE3R zstab$B$5uwjKvrcyytG><)9{QS2JDnsjryk&0Wu!IhpX{$7_1)ff9n94ff7$qSucO z>*+s=n_1Z{f1S$DHEG`Hc2Zj%CDIW>R2>l#FTY z9=#B*4Z?1~i>yJ|v$Z{l-cY-%!um0iDi7rJIErpXH<;Ei3g@nH+7|tNX~vS zw?*iLI@~m=|CNVsj|Jw|P_u|~qojT}Sl1)qg&<3A07VW>%(r7b^^ZmvK*%stmlI8b z7~Piob>5E}Y{5hEjq<)4fLCKgV_%P%5ycT`Y*J=tIz`b9fVy~`A>7JYpe?d~IE%pU z-OAIw)K{!bp{tF2`9Vem<*YlGOUK5!Iz_|dk6l!Joz(X!Ns(jbvw5crt*^bKJ*yZuY)5gbW` z@I5rFF@odAud_-!TqWpvz?sjze2vVD!Bc!Bqo#ZnOUpr+ z^^mjXez~uH`pcu0nCA&w`Wj7~K45fjf))y3*zw)Kk8g^GB+7C>PKsNwN+`IlCh`3Y!x*>GS_+G>bv4L=T7R|---S#59 z5j?%aRriMB^{ZPoj|6ZZhS$tK6O449o!|OT0@TqVrCw_b2(7LH&;T?*Vd5}5`eWb5 zmUbOaVFTosI83O$K-RqhyTu}5gRKH!Pz)s4;@d<*)Kr-b4H?xzGVgnB03aSlJTduZ z130D;ky@zSfVBP7sB!V#7rp8MQadrA{#jsQ$wBuZ=Ci&kF1iFYD9095$6pW!EmZx# z!)PFVr#eUsY}ZE))dZPuF2v89L!C!1#b&`fM7Hr%K(Rd+1KL+%U~JgX2*bobP>F;> zjJzz&yJrQz6N<$rAFIa;;zWbFtzmbf!QXBGSB!wV?fl!1Fx)#m6d8%l{0FencyU(WrbYN_qFO(yO`6bBoC>Ddp zjW7wD{zYG9wKxUo`DtLNBbr?vdg!E{#Yf0@XEMI*sOkm%g06)jnrH%$Oa#za&)m(g zO-!k!*H<~GjOsb~k!McHQL0qh z_Vp6WuT|wigD&J|`MLic9w581XWJ6DTm-z&;b%XoT~$dy9LP=h_SV?xfko9Q-^{b^ z`z0J^vZqs*`b|sK+Usf0w5`2PP>1r1f@@E6?o>WtXV!Dy`&01BUxjm_q&&J?Q*Qg! z?*KbYagX61?M}K~H#=~6kSOXv&Tfap3isN1>(p8FqajHi1 z?UmUAB(loG5*P)LH;YF=G)zLzJA1U*Y>28bbX8{P`%^j|h97BwDHAkCiW~qwFFCdS zOX8qnvjOyk7!CsaZkaBn(%kgNAg62 zL$Xn|ICiLd8#s~3`)Yxt`dKg)WzrG8bm*FI)dTUv!lxD%1Vf2(9tMd$bPF$XxOzEf zZSh;)Gvr;Q$cCkmvx3NBWB^5^bjR%iy%&Gq9Qp<_E%TvtMF5u?=ngZ<(90x3;uP11 zX~~nO5nv%4G4ZPmj?N{8``P;#1+(8w!cDHV2xT>X9?p`yx!@uvN=;NofloA}{e zYljZO3UFXCvPl&9W7q{;!2P>Y#gGZDw{1D!;rB(h0;yKe9_;uRJNG2u`KOVWi9Ce_ zG4PPa9}3`*{J>P&j^I}%a1@K;SPWW#LeFG6K`#tso9E;jn)hRmKZY1T4ZmAk2X(<6 z=+dz8ts+9(a7COV=Om0Tp08~6K7kYF0E(DAXb2qGba3KonvLZD!|N>by`-bv>%A|W zez_t1aYpm@x3<4Z_s)ZY*RU)Grp^GvHXx^C1sr?b&c3qHPmoo&;IL>gy^+Fv$lS|u zy{~5D()#b3Kb{`h!oe%sZ^-_rX+q4cMf0Fp8yzgOn5dLlXN3qGvnG@!H`#%zp8Rl2 z2P=TsW)S{Hep0nZmAotdQs33S*r(g~Q)gkXNHi7w0eWFaPhVmUy>Qz=I9+g&J&fG5 zE(FQVn(X!!9v^c|C-eRU5QGu1Um8KBi3~gQKxj3BCUF#zK#QV1k~l4B5yB8n4|i^t zPXEq`Z-;Gz*WUa-c+7+jNsQ4vx3L9@Y12y$TVF*K;k>3ubj77UG@%@gZyrG-lSk09 z>b~N69F$iE3hJF`DB>U(Sy_O%A&WH0R}=(ZKi!wN{SIHk`TH8A<3j6)CGAY8qJV%o z1P#@Rv@9D;IFfvqydSO$%tOC?n_FxVSxeAFq6q+ZNPw!fcT*~|$(TqYq|o5l$&*G~ z@31rrw(EhrEZsrR9!p;L;)_!bV2E^oy!|B6CBkx9mNu0}iGTJBsf33mjO~nyrTc-Z zEusbMDJiYfgi{ZQRdQD6TLmm+X;^{~Le;1v>^0Pz8)lD3HsNDes-(SJ2L?gLkygER zJsGz2rHckTAo04jgPoxL(D|Txt7<10L$?f-YWreXRtbOvP*`lr9Ep=c6Y-Fo9CYw# zM{v6P!UM=;t<EB)=uqG8s>jc{8AoLw|JbxfKUx%;;Ne6m-~avbm6bVTu5Y;( zenBU?V|0f?dwLlKAWfMJTL$HXDsG&a`BT>|zs{cfBdDL zT#i6B#Mcn#C6DSLg0MFot(WIKuJTp}{lv8J@!xW*cQ};`OpXP93_csI9_I*qG>2~T zq4I%XApY{+R^u+hEuI|?`b$uiXeKHCm%c2UW@=nbT1A54+%7-MZ^2Xism;t9@6{*j z76TTa-pULqfK$I1Ea+sq0yyZ)XxL#JEMF8f@MkRI?-6{44f}WrkPH%F7_lVHCbRyD zJ0ue=XYEnrOzv50WHau#79pHZxFAy?eE|w9 z$&gquYR5wI0%Qh|oxgWwOdo%OeIWhi?yeC)A;Q28O9DqR+mU1kC3YaQ+MrFsn%2vO zIhDW#ur;$ht2W48M6Z$ENhgu2=A_pqk~ksB!az&7Lml_u+u2S(L8U=IO~Jd166jw_ zRa9R&BSj{kfo^^Gd-X#SN<8CMsUeK5hpI{I)MW#&Jy)>dD$-k&Rd#RrISYnS);1Ti zTosZgXsy`4L;HjdHf-r9aJ(-`_+2x4s$$CQ>XHR@O)eCFIb-x1-=T0V@Ju#)x^+lQ zJ&20@M^}Ks>icC-cRC{0;Q710)$LN86HX8@eMz&fiU7q}E$Es~`l@d2@+4wZfF_%R zQAN06$MiN=3@+cq>bgW%=dFJyxL@)%Y!Ogc{MH(YPRQglPL^-)?=FTJLN6 z`ZznK1A%Vu%@^eu7tyIkL)y}>RKpYM6OYvT_P@GRSbsOYZ57Z|y>v#=vfqW?=vIV@ z*-!6{^Jy12YR$cqlOE9U5-%G`dY`nkJs(qI`k%p+n@iuG|JiH4GUt@=BI1AG^|#~y zu6$=@a_Z8yjqOV515K2bfohsBZ~e%+J$!Eo*@u%3W4%A%4hoJn){0sq0t5J%24pU; z0M&2@#~Akiy!4}J-_S^UAYUe|srp^Ca&#!LfEu8r-P+K}b|q>FDD5Z2q9KW#kMsYr z9FgF3kKo=1St6lO0OZ`=T0@!=&o0(x1Hj7tfR>9>YRJoJ9f+q0aN0or2pS*=?L=iR z0hqVEk(+SS<(Ks4L#;%fB>3vicLX9LqP>JX4i=?@MCG1xR?qT4Wi1jBh|h08jy4O( z9cF#nqTd5SeAjQaCSpjdsZtedn z;(_TM=tHt*G?^U`?naoH%?d6~A1{bq(eqwD@p`Gf2YD~nu>z=*1@lbu)>vIY-`kgB zpLq!!fc73+TmqD1Nx$-eGyD*)d@|=-q17Dz)rYO`fUG~PK@en3ePq1v;f|0Mv028^b1iUqBlAYdfw;8p zMFi}+oHjY*m+LL8RvW z3jeLg;qpY6EAoxOg%aPjB~N-M5MPh7PKbLMF35wz0F4{X6~zOE`6n-AyUG*))Hy1MN?XQi!z5R{Fp>Iq8jDCOOZL(|U0!0fzNoLK#&KGz!<#N-xCc! zVGs3{jyV^I0Z_x1Yk$+Z(<%*sptx#9I4O;KIuNGUvnukeDr%N=!7ibGU;-0?d6MtbH0q_ldRhGpl|!Gqoe8y zpcfZepTBzX(0@m8ks0=@=*e^Yap#d+Caj9nextQ)JUtNqB*-Dy-W81Ia>W}}@zDYaNzSIZ>Ae;n-QecjP$=Wvj}J2A=JN4F*ME> zo~TsXLNu)L@P5V!9Kz8opAw6MFqcme+%Zu|$0yz+ZpBT{4m$H1SllCr$ni>O_Ur1q z^F9nx1I|WN@%ei!A0zz-Ri+k)U|Yb6NOu(}8GV@;8fnF z@09lfC8yZ15j-kHSj*p_Ms)-Ylts7dEFLTEGSef5G}g8Jai#>(;2l9Orv(jWLmGc! zY7>O_M?W-I=VbR%uHzqU-tpCWF-uG6d|0eX4%2jjn`;%}>hFAqkq=IQcxt{KKx#ZB z<3$+8Bpe7~P+K=5ZML*7yPx6SJLDU~dhvNnBw&}X+c4&s(21xxkr*^?{u3^AXSS2Z zITr>-W(DWNu#KpRiS$qAG(cO1d4wh)N<-H$0+pxkKy%l@1-qlQ2WtzVipl*FvP>6Z z;Eff-@~m5=+wU3l0db+x@<+t_2|ulvzKARKBHhSbmcFF!4b>T^vx-k_aH( z&u|zkm9^cxZ2J$&n-JV>OFC3Z3E`MaY5u6<5Obn@yQ?wT32C;jTp zeB*Ll^PB%-JbXP{gNVUSb-sqBGEeMes%B`V$tBufx32duryhf2Q=C2roc`wk+oQ=A zm<)8$gyX%zriUQ#CYTZ|pi6~u$H-MM_Wrfwu!DbfvgqcF#&Cs_^XHL+zh~&)?#8Ky z?XUmRjpcG=^SOUC^_%VZ(8sB?QVY+0+L)Ud;u}c@7W(VBGKJJpiSeDoranI&aF@u8 zCs>Wt?mwx+Xx5v6`H1@DM<_3O$r{$VQQQ@mQTzeAf_;iN|}myOYEZrPKr{KQXSEORK|!z`9u@ZefUda zaR5dEqn=O`!+e@joO)AN>Ll?Cs)PdMvR#?|c!}YTc@iYQOj*{xbMXrIpV2(k88Ml^ zA{cS#?P1sYtR%K`bUkVj9InoVQykx=KVaP-DH*OIx{!ouan8pP3cn_>%nb^RR~$Fp zQbx_2MvuXmZqvsl*3VCA_Ewof#DnSMY}V(RI6>x2%RrjzyvKOb^G&`h{i`CXLgxh3 zF&T-105WUxFm7@8R|rv6Ej3>4q7yY>a&xCMw;c#6IQ=hS%iO zrR4T7=V=9N9sB91tb8b9-;Zkp?58ON;xvt=YS{@dr1gw)bu_Isip_9u7bdQ2oOMNk zHan|fs`wNV1b6EU{pu=jUql&(Q&$>U4BgLeU3u9`F2osshk#c(EijbXr7cs&dbYn( z;K!`mL*=`5SBJ3*^jF3RSs^`wB$G^SmRMzfv#ML(7VQu0PUE!vFQ?0d27fQ4sud;_ z_Sh>GuqTVn%QOB?6M2^qk3YUFwuL$*1Lh>^9(5|hPuSCH`MyF|X-E;ErI&fCDlU>Eaa<=RN@QW|UATt?K*| zS&dmMrl8x^9z9fyr+(1#eQ{r=$)55~-@gJxP!TFZ+*&drfokk*NbvIW%M^kXe7T=r z-e~mS$Jjnym4@g3XnA)5k<~>vI~$6&%3y2nDd&V3`u?v4GIW$}W=~P&VFRr?;b5lB zBut|ZL-YyUF4b;uMU6tbh_=}XDl#S@^M=6Elz(*O#4Rj8JyV-mM3GIvgrNFW6&Iv8 z8%1e#$V6bxfkcNFx$b>J9&q6lBwg$Q5nDU=SXfnR<6pHaO_E9ib|C-G$c6R(ks4&| z2Qh0dfsgo#4EzG8eF}iL%_E4%1K$w8WA_>Ryw;TdQ^Zm5baYE7@!!u1aZTl`;AP5E znw-o_ivZkP!XAPJS3kG<97Otu4i8LGz8-JB%%Ane(fj*P1i{=L@&9y11(JsanbSir zefy9y3Nl_Dvq70VGYNaTuiIsF-09(Qxsi37jHMtbC_pw~*R4e#VzWVU{qc>%+64e* zk3I7VxkisIeuA7cNa(l3z9{B+L)yNz)w4j&@)BZe5Gg!OD4HT?N=9W{+AV8xa31{z zfJEqCPB(5;5{O@+k5vHww$$ja>1WT%s5tCb&R?S+J>oj>H#S))zT&@0?%+_{nL%U4YhaSdF9Re|ofSJ5WwC{T^s-}xbeaY#ZKzzevO*+E&sKllq_xL|(tiBJ=*SC!bu`n_R_{fPbOpQji1UE>=)7 z_Eq@CT%h8UvR9S(cU4JA{MK1+QsuL0PH;w00ya&I%Br13(pW%|fxX)manV z*~G86BZvt)Ekugg9KQQXzuJ@Cte;iVE}^y%K4y2Z7vY{28R)Zu+*}ac;5YxHwvjor zLu&M3Tj%je^O#*}Q}x$haPIkS|1o`rmOC~XI9%(jtvQ1L4w6Z47zH1_%Yji;9CVQ` zu2x8jUwYnMKBw9#dFw&`KkdC~IMwg_?_Keg3ZYcSG8Q4T$dK8x%p@dZ#*i^H6`3*= zna4~aLz1KnnaeyZV<_{Kgk)x)*XsNK9eeLbd*8=#Ke_LJ$Kgq}*0MgI&vjkr@P5C} z(DGXUlNzkrxBfm0Q8q!$djDpSR{vEM`Tvm{_P_1H=&{KApKrb9KTyt&czoh4IIs98 zPYoiljIuU4RLe?R9`+!YOst$BWsEA}O>aq%X&|1ca;lgJ z;7nuOAq+GFe@dSIJowdv0jLyrMGC*(^Zvk2`+{}|xn|Qr;L+It>pSj4I)aztfhMRB ziw3XJL9?#}Y%WbuV4CQ^bx#Fcd)ha;T*DvOK%`jG%Kfnc+$#*oE{BRNKmsh7MePd^ z61*(Bhp!n?om&34plt^X9SEW`wauZ*6muOSJLLcckS`c|&tr{3QLEJv7!C?Mf0laP zl~75y5?VhikXmJzM2FH#hPZ09BLgtn&gk4m9Pv(Qd@0Dl34_Y-$WuEA_gs^z7Z@zG zs4!;6$vf5)7}opm;lKgE0dd6KnEU`=lQI;LwSL2~r9vnHsn}g;O*7(!AK;UO#@#if ziK02Mg7$};o7-d+GNL2{7dww551iX2U@geu$OiXhw(H6+#4f4*D((VM2F3L)L8h1I z#EK|zpv+_c3-&26p=03)B7R2f^1)@E66vR)U5@xDnxAFUanPHv0@x~zauQHyVnXRz zlD^O3wA54{Y63C|1jBz6TWx@?U<7NQ7P<%44cad1v=k9?I4|mAx<3)EG~;>c)+TBm4KZq?S4HYp?{p z&bdX`r`0#wXlM5o;lMFb&*691gTsjCL6Q_&3W^NduTk6C+-vBQ5WMiczP> zx|Rxy? znVgN+m*RzrdF}B|Xg7*!ul^bfuWHqA+~6bY1z00h^hGt*uP`V<7mEzo(2CHO(o6aK zAoQC9W`98kB7J|0l`CLzedoG)0!a6@P$=q zE%Ta$Xqhb?DbmDXYB~}(G((W{!2(fhCs3@vc>A@nRK9wC$4?jhO0(`YH`nCiAW0`K zLyl=sd4wX;t*f({Z4R21?!%!e<}&u^nMyID5q~d941v-RFJ0 z1Ujo7ntJ9m!uFLpltE8}4_6#B?mxrU?HcAmY~X&y?2?_;ru5kggJ-hqU8u=})KNfB z5J2#)LG!rVJ@4%1R;f8FpYD$N2<6oTlFE%(R|zs?T}OWi3o@f28CmA+g$-G0KVl~sx3@TwR>cBrXlX+Kt_LpvFH+P;sH9(@Fzqo1p&Y0guvCgv zKHy;G1s0dKMnh$Q1hmWadR52>6&|1B2vi}|3412H@eVKqIZ7{~@x_BM&vUS%^PlHi zYn$I7srvhTABUA|a%c$@G^o#NDBl`SttO*pjQ$6a}Zrvaq;?QbCkXpE0+<-j3T+rF!VfPn;Lth{@5C=1(-~2_=C=DG5jH z`@#bxKs_n`Tm5+Kd;VLDGkCE@QheNi-W3zd*32E+ho@?_Q#XA<*3@SxatC<#==$OL zW7^Ef+l=%-u!BaRG#Pa7FGk@kTYd&2TH)G35m^+8H<)1TBP~q!fp;mdulF0izmzO8Foy00Woq6vb;Vs!&>ikt?ht9cTd3dqBWz7EL+pTrv>;Czm$>|;9i26br z|MF>&LkR0VbZN9yMOabkIMObveP*sh}tbPvvhhboDjqo#QC_N-Un-{`#926`#ClNFlf4ayGk9 zfo&T!P7;sOB~yrRD9D0G(ARd-?vJji$5%wrK==NkHuUX#lwXFa2PeoU$B{pA;P@7r zW*lO4^eBJkrbOfrkRVL~9v^~`KkdmAKOI#bR;3~_hk#kW-@(2kwDP#Cz9z9?E(>f_ z1#8SIgl)++>`pB9W2*~kG!53dQg0d@dS`d4WKo3TTJkOLUL)Jp9JlW#XJ ze>4=lxXkA~XhQMrU_a`5o*`sh4|n}c%wKDV8wX#vas3X%tMxCpf(rbt5?FBR>n8Lw z%#-4W=ab|6=SOtA@?GCv5*5s+4-4SG|NaXbg|ecsV$Vf5Zg+;Mnk$1B4~DP!I>;Qv zSG-<~T7G4*{*xXwy;PtwA^nz15Qs^mP@y_hs0zth)(!Nc_ouCI?0Q<1ha?$^y}a5H zzF}D&aldgnzkRJm z6@x+#I-|h;FO3lYfO8u8Rj>*gmvhKz1@lB2Q05DSPg@ei@dt6$xqqWnBt}6|5<)m# zUZ?r{psIlPpamJG>ym*b!d3KsA{1|2gj3FU7lO<%6r7?ktr2n0kGurcu&~%GvD9dz zNyMx@*k3+~fb5SGk!601&L%sn8n}(Ru7Fh+PaoL)8;59LCW+=isytBtrC3}brRHNX z1eOUx`JkCO02BbH3rbO~*(f9tpur3jPNbdOK|CP?Ap z7zlRUEdG9neD*-hW6*?W4@nbZka;1I7q2pb+BXBF`%Z}|E=|;yvM7fFCY?%YAj00^ zd#@0)+g!zp1gIhNLvcli=%Orw-~cN=0)(ex2nM(`T!#I24}?jGi#o3N_Y=!%dGbDK z^2I*qWe!eIXrn}tAc~|>b+hSa^=b@6^QTbXFx>9l5ZjX5ZzNEHgY_%|mK@d?qLw9; zP4@C96;d4UhS3`dby+ZiCd=6kQqRX25QYoBMO*Khql8z&s&}N9@++f#eQJw|hi{;YeL*luR{4V>!OOUHr$EqVGIl@07wjkVJI$FE88pGIm=(cXk znrK4L-xT6F)zT;kyb`3dQIPp5bii0hlPsDf=I?tK0zL7yayaR`q2Dwh4Kq&+Y2eJT zA(XiN!386Eanr;D0ibSf72Pd+tq|3^fPYa>=s7kj;%FpA7w8RE4|*LuB-5su-$cS< zK}m77q@05Bbdp=KO?-ow)sOXl7E9^BVKfzdR9mX}HF`o(+lY3yLC$+-2N^crk=Ke! z9t!b$5Wj$&VmRBFLuwbG{o=Ujd#%G#C+5AR2d)Jxg$;~RM zrihy>^J{BJ;a-J4u(=__v7dx;Qvg_mp(O}wski}iG37MUWltgp>rts*2yw{~8rk-3 zx4dA@$-gBQup_Zi=^p6^N68$hgpWsqT=_g?B6b3wO^8lO!ge#zLz0aQ6NfD1pq3sR zUed;$=07kE<4!9m?iXGe_6w{)x1Hth2a?B8Wvx4E(6sj~NX%gw8X zZWZz4cSA7H5f!N}3Bce7{B8sbOR;rmP$9%Pq!@DW{c{rD?jIJZ`4jF#cVQ3!&^`2B z?mytF&>=$=Xw=2=h(e%D(-=|ojE4k%WKUjiCq zZ*k!l0*924i+coK3b%EGu8zs7Zkn=}P~diF#0RzM~s{MAuXn$xg3$r!+`a)gb#5%RaQXgjkl z>(g-r)pj$igSfhvwm_bkpGE2qFAii6Z|BUHHLee1wh(driE1OGdP`^?nHV%L5!nWB z`!2-W{?h<3x7m2b_nE}(p1`}`e+x<0D){T+s2~*kfgY(Iep+~U6udxl{m?mq+Frepl_HoC{PM<1N4M2nB1eo)o zXGv=L_k!!MC34-*OpY|^WD;t$aQ;y_FL-d4;s^y@$!;O=rRKjcwy3=F7y~la(?5cy zL|WiO?HyAB;DBm9bF6VE%O3^PlPn#bB3IqX&U;jOKUh9>or1;gbuH2ST8;zMJeon= zlVBl_XatDD9QoKhiiK-nWP?dQ#IZ&~yrcg^BR zeEBCMZ7$Q@U%L~X^FK<5zLkzAX;NRgO|`PyNKtCRnO6@(ME>wdrdwS+DmHSZqpHF$ z%&!=qchEb^Hc;J^{qa_-UU=cq5Yg9ll$n*x=Oi`3&XmGdSis$(Z1dEg_UDYe`r9vT zNeysZrV7nie4#y`P=*6<2GrDLJPpXAoPNo4HCT)wd`;w}{XOVi&K^BX`a(UDK^#w* zbo2iT7N*|2JUf?a*uJ%XgGj>5Q;iU_wT?frqWT5uvxs8&!zA*=O07I}rg!^VDpUWS zfLh2AjZc;Tk28WS(Rc61)%RB1@Cumq79NaifnNQK?a5{Q1G{jewDECZXIbn=sOe5Z z%{~lT*rWpqx}vvA!gH|fD$WHx*+Ffm$7(f$=2j*9}j z*9FbSlY`})nME1MOPn3eD={k?J(iJ>8Uw|XHV~r^7q$#F6VI{jiN{ss?YCc{X_+sU zue|B)eK0z4ZdSv&`D2N@{>o0mjosA5v*+>Kpev-m>%|_e!LfGC`c!FNxT%u&n^as| zm7C&)^=I?Xw%vB!&xaeArNr&rtLgf&&93KD@VKR%G6O~JS62@AB|U9Ax$K0|YD?vL zC%Lx4?@xcbI#(AbNyIEE6(Y}E#NfufOzX!aBR6t?&>0{CwZG zvzL=<_A<@P{9t1WBa^R33m(A}Z9Wk1eLHjI9zaxuCv=yby9^A9rT$;p0orLERc zxjVp}k?oa@VkVn_f48ux|4!Qe*QD+5KJfq5eEvp)@IAvya4_PVr6`X<{R?Ey#3Kg~ z4FxGLp*-9SBu_KoC6IP*Pr+7n8sV6cD=a12d5xyBALWpSJE#eXoGj2GrA_@Cl>DBD z&gv?htj!>|!qeN)Oi9Sc3hVxq_5$nOW783D1yXy!0zNEjS{g~#ro zDmXqqhQimzBoOdai+#ZJgggwI>2En||7ImU3s9AMm>`X2fQfBQMe_pWPQWafS_yU~ z)yl1zr1?c8CQTEEhA8qO7Hg-2a_^}CtY!h)5Iw~!zyic01K#PV{mBtV!q0IKk zzY&%i3?+HUYT2g9)a3!m(*(+tk>?O0(KojTYEms&Q$YUfh>)e|q}6;2F{Xr1GDuVE z^;#Ic5vfw^+&Y^6fF?CS3i_m;>>6hk5r4F**h<47I*jMuY$GzVT+q^>T88K+aTE|lET%?eX zxgFR>r|q*NQ7qDcA>^}lKPCTjB!-CxbHs2jY?g_* zg7Hinqm_@V18^yHV&O7?xYru$>MPJ5Zi0)MR+o%STu4Mb2U9(4OB&BOer61U*oBS) zWM8hsr<1bpaVkuj4XM02RtRyFGqP@UKs3CgfLfg(aw~zzSq|oa_1<0C21;QlQfd1n z>skmyx)jJ?=u2o7lV4M$SE)8;XboYDzQ*z^I=vt711B!v4&}_*6%3z6R&P1feT15*?Rp|>g^lCNxo|m9*#DFy zmo#yr?*o*&&Xjf}DsTU$`@NFHq-&#->{1vhUD193ju%?Qu<2EP0hTbX_Tx;SP2Q=q z#wM3YZ9p|FO*#v(o>$OJx<8_(^4+1ku&^TT7FDn<1vM}zw>>a5yZ0#5hV!GDzkO^Y zuX8B1mgIJKWKYlgH=FA-nV$DvwtNr1z68BLEZzm3=;(@EE2!U!jHSab^YZo&7e&Q3 z8u1?-TiykJmIy_(!kL=}jRS@*eaYJHPvqC;yS*AB50>A!-}x=cxh3`fsBYoHnE1rE z1M!JPBcs)_uVvO1iH>px2IlR_r+F7Lf_ZB~jq8d;SD3 zTQ*kyv-Bj*!EIP>o;RO#v50nM^1OoMQu&lKTkXJbcDxN{rFrzj`@xJ_->t;pv!U|d z{cWr-8l$kjeNX%Co*c06=lHG74NP7vH=vyS_U+9kk>B@eosU-S*LHsegQg#7-q68V zD0BQ-?mWr!Q>Qb$p8LaG#-K~D$|Qoaef?9w4Pq0EU2?WPG_3|zq^{*JK{K2E`*@ss zh`9wE0Qbb8EMuLs(f}YAA%qXAds8MZ!k^ zew@OuUIGK?{5!)xlCFnLJ0OjSr?=iN!9-1@zn=C8Cc;RhG-6d!K;oj6sn##p1O)el z89>*8T{dj?GgAq~a4D`l>0hZDr?~e`(}23BCac7y`gXf@jJzRJlj<1DPnxYrUJK36 ziI2`tGUxvs>dj|VIOvIE0)_vV+|7%Xj@kV}?V)vFGv<~$5;O(!-DH&+&eR6nuHQ?n zzpD9lqQ_!8IXFn(kC!Myg5wM)lk#@@)yliJh4WwCzLbo48uUV(<`F9SGggk?P<>Y> zVbi!_b0^lohWCWLpyg-=*_BI5Kc5bGHHl|dQ=a$6WpDocJ>l@XNN0VnWNO$w$u5=< z!|GO{H|bQ9a*P>BBH=I}ie>ujj~#DhH3L^>)v(S5d`*BG=#Y5ZnFV!<$4o*ARAjRs z1Z5k=5bMWt$$pX11Mv(el@h6|6^>LqhVH>IC#+-O2*b_ zaKLSOCNwis;mvs$1TO4PC$_*O>MnW_yE>SZM}wVP9aC#dHcc>62~@;Vg|m%!Kbi=2 z_gy)ga#VXO)kdaXl3w)6{LLG5YiI1vt)1nitEVChvY~3dJiPjfV^GuTG4b;=RDRQI zmRwtAodsXY7!OE=ac0%Y z#l2s2dH?4Pg8T)ZnJe}b&Szc2h7~~ zHXww^@pp#F;X;owef&k&r7@oyB*kLi#f)n2_V)f#EB9FWRu}BPrc-usv+5m_iv9?v zg>o9@spjyAM!9Ph9vh^lrR^V1AN(;vXc4a48`|i9%WMtO0C8PR%NEs z6XI`9QYQJW?Qh+)n9*RZZgvu4=$s8u>MMO?qq-t-il$|vZtzvXu-xIvf%Ud7#+4@r zTY4a1ls-?li~eyBTYoP!C7OItnKiAyeUjCP_R7{f z*b%YpzG{W~<=orm{F4!1?}Z#D@TI>kyhvy1J^Ru6n;{^ZyF?AlavO0|yIy%giOMyp z$C&5}G7zdplT0J!p90Rf3Mv%BAkS1(Y()wCeZ!figYIxHXD`rSK$L|z1;u(Ze*j?l z5yw}K<^)@NQcLxHxtNw*mo*Ll54zA3pa)dM_6;`AM{_&xuqbkYG%Q8p&B~tzeJr~< zVad0hv5g*j(xIt6AjwA@bg;~8eY(&K+@kY%-3ubCCOJ_;%eS9I=gIEwtfbo4LWiJx z5|$@UTe~v-G^r#0ETnVuDPa)Vi65lT<9t^91Rg4-zR==2;BzO@y%!%B7*l)YX?kGf zme}d2rm3`3KY6Ia2W(m*gb!Z3-?JsNKWe1%uqL{tCgzQwa8y>`s}KA7k*wD3`PFRi z$X>DkS`MvzWadhQfzdSNIP#3CYjGuXVd442j`qlq9;_DOFiqROnnk4qJ4|<<*&xuu?6vD4@~(TF70hkpmdF%*+(T&8rPsys-5Y+7`1M}s zgt??5(F%w(WA&~QzeTtXSB82r9h=KYD$4_F6jCw;VcN<-P zQvquooAiY`w$xIH8@Wbv0dN02A=3$Wh1uOy?GaSZB@vUoj>+QujmfP5mEB}&B}80D zMB!IFhw2UQpyGaY3Hxz~Yq>+|aEYg0=DD?&~a(D$K8XYc2g*x~_s2O-uF9U{%%)27~ z;-d_h6NT`ZS`Xx`P%wsMoZI^#r|R`G-bS=)Iy5UFafqew@;(=TW@LR{Wz_Ha7i)E& zsLscgsb#U`S!)EXl{uQ-Do`wg;mg+_$X<90J6leIOq-G?a^WHV2{M4e_?g0U*Y4Y* z1ASiA-n*TI@&4$=DaH$?@~8Mi+}ptUPJ=)=sEgPxk^j7&#o}T~g^*hmvkTKz8o)@y zcesWST&I=7=|uP*>$I6Ny+y*nUW;+fs)^+Q)>OJzUIbS$njW~xO4&3i!42&5EWt?8 zZaLiS>d+>+aBkDS>8>i~r@0l={XC%68Q&RL{ij21vIZA4c7Rxl!1b;`awDqG`{XP}7@}9bv0G&rk!Y zWY+A>$=&26<71U<*dK#!Nz={&>HsKEQW$%EE8i{b(=WG1C$)X~P2uI|h8X>|xdsOw z+K@#O`T8oe0+vEO!kym))W+s=p)aZzFr^dHb@XR1^DqaqS_MdRI=3dh=pJXi%0vi* z%Zc;H%pQ=w9hwX_zsouLB{KWE_GGd$@2m*tt4YB%`FOFo6QV+2XSUiDojj5xPHgT> z+%5SGM_dKzjL_4AX(p z`^wqkmaV_Jx6R}Kl&G9%Z~bkE{a>jp1dUxsj0ci_EA_OLl3}L>FYlM^6#qMwEgMuh zp}@rE0S?hwSXK7$d>&nayVqFv_(bn#B_Q3Pycmsh(uEdX1}NxCV2hFkSX^&y=Py*= zfKtQ+B<&$E%phkLV6&I@I42hoGZ*ETC;`Sp@gZ%46wJoG;oGlagwL6!hkPEnLcsi= z(oySW4Zt@Sn5gmL{t^wX?Q1EZRI*j&u=JI}r#C`B$P{KqA;C17<=4BB8$yRev=&rr z(Aq;lnz(+m1ylLL=?a`{ha6W5@O|=^lJU*)&KD7@&xWhT#3J84bnZ6nvR7a9P^&zf zS_7MBV@{HNqQtCMly+q9T~O6VfU2(n0W!hKn>Xf2+S2)cEQ`!}5180r2dy&0WBW~t zF#HUyWt?Tn!x-_L2+cH+qXR{>_U}zxU|gekoTs4=SF4B>LGCPW42>c5>Dcg)jES7} z0@jJzdfegB^OV`n?Rjb{NN8zSDe1o ztw?EuCi~&bCR0ykG|ezFBf7v_T@Tn60mqIz^rN))ykKQ(Sq6A2i)RzakY?H6hkPH+ zKR}a{%s{z@Sr=jc;Q^gT%_VP;v}>ij)RKJ}a6PpvNMDscJJmVWV9g_CfsAzdN$rsi zITZ%jkB{f(u2r5j+*wOI@b)~9yF*0JlE>g!wT%=NU|@^4xwIZp+wZM{z0m1u&oG!Z zqY+fWol`XPP^ZW|ZydLZyyfx{bO}WzTM~k?7p#F;qN^VSiu8n{^2-`#9!qEf6mkWk zsTG=BGlKbd1&z=^#=bj6rHH^Sh^j(?X-cn*o;*2DnugWK^0x3=b*1A`i=a)$&=b;N zhcTS*?ZOlekk~^IuZ7mQ*Ini_YqCe zt-NdlKPN6F@vJjz;x>Zo@iZyL=6SOs4`xSj1`OE3_RiTAu{h=dt6wxYoubmbd6NBA zL*f*7$t{{BW+SyVh7C9XPOEBs&CZ!(Pnv^XV0F}%TU4MxNLw1B+lr!x4FK1Eu82-?? zr(L-cV8L90+8u|h{<#&4sT`<=<&EDtM*PhW=4AA~9Y-fn^?KpGY`CsB+yYqg47(UB zreklR_%s;iN;CJVh!P@<9F~1fmPe_5^@Qj}7foUZR7HtEQk6b!i<6|_LHP$8=Dy}* z;nH(>oDTN`4X+e!DaZmmC3m0wg6;a}t10nln4TlxYjZYA(LqU5jm~ABkTm`iq3I#5 zEyW-jas)MePP4|OM;J@riJ}`7spjv*YsCa(?{=edro=`?XZGD2kKBtC5Teb^Hzg#Mgr4e;O|cvE@i1!Zmp$pqiW z=c2oi<8j}XOoOsw3F0p%CVgDqpV*oX}DdB{`DP8k@SJSEDj6*_PL{p^II-io4`MekpN z10%`)LbZo=1kub8cI6C;JM4NtE4!@tJde~-iyjrQ^&rp@B@=ijpdF^#YXX>a1UraO%$!Z7ayvKA2nw`Qbi;^9lsGek=`4QfrNMb_aC0wE%@r+46z{+avt=ZVM zOer{UoZ_EvC}k8;G?djOX*AxHg`N?~b_Jsr6ApM8*~NY*$60ve09EnINNr zY&IdO!orDEnUs}$BJpgD3FT$t_YyP1@y1W?n*8$as4vM_w6%71%2m|Nw5sHzVWwfD z;d;bOBUi^N%H?v$YHHQYoq?IhJMEb(jg}}!!A6c$J0tE*moDFew13NDcS&Ayc!LzP za%Kmg7>xnV9r=3d7nT7EQ4%Md=x5B#RdV_%FY#hS65EJHBs;O+2EY05Js0}sy{66V zLGt2cXQqy`x{;n!E#*~F1x`0Jo(K9V8pR@K{Yh~iM8vxJQcq>SH(V!}&5~&dzBVws+$}2GGbB(8okK5&YlJ+3vP>6XA&a?=VR|d}pbM3WG=DEXVaYNfU zRSQbqThBdP%N(-lj!$o^i06#WF5a!XTvWB?DK{|Gvv4P#B~PPbJRfiZ%@2v5YdFZO z>mGl&OD7b$2NlFq__6C#7bQc`h`OSRnZ(VSgPj@yxw!76-DDeq&;WJ98=TR`zaKZf z;|&X)wXO?mC^bw8)P_sIvFKnXm=JN|g=VU`tawMQcuI&t@z&ruUwNNwXBmY6QGM#q zS}zmEWse3Ws)@4=TAtAsG=#ODo;F}{{Q=oClc=oD(53~R^{1lAmedqe&Ri|Eu6$+A z9Uf!ZeCXuK;)*noK9qfNzz7A-~Ssr{8!|TIf$Ta#Vu`K{bimNjSYStBBVnaFd#p zqc>=QZM745R&3m45!7 zD$d)hJiEY~LfvsYd`P9TSuFZ_dgzC?ov18Nxuwi%$HFg8<2C_8oYtACnwP1+-}yp& zv@A(4(e~T*lpJoIj3&*ripv+V{?1F;_no-w(!K3EDsq>Ryz)cS%y-YEMcbcD&FfbO zYVl$xhk8=$$$MkB?A5YzjN@76&8%FkD)VKiX#{DMxJDSKg@4HV;v%2>_(xSOgY%DeDV3R>0E`&hQB`WG(~_B| z>N~FoJ>JGD5LHT)#VkJ@xG-{$Q+dKGru2pf*DwWvFJ?VV^s|&oCWDTtam=-nX&P)5 z+b@%QB1Ityg!vJ$I5-HV z22(G-u3R}-y)rKf)%*_~=}TyM{qI!b_(~xi;LzGf^^h=%vi=Zh(BzF3>xX@Z_IMp{XE z#6$6Dmeso8@WDVI?f+`?f2A{vg@UN+EOSNSW>9uYZAg6s%b`eBbuj& z6vZMxQfbtzX&#ra@Q}pW*j!sr{YY|SIZ0OG$$kF~PuI{8uDRMSE^OCN_cIL zYI)X#(m3Iw$YZM&XBfb{pI?vDDtd9u_PrOIU}RGBYtHz)Un@0)%HkiSH9OgIF)sS?k7X?%5`{HI$VUc-2SzkqzT%fj zk*SgWFsmf1SIQqV`NfGuzS}uomS6{>&>Z=MruYu5gppte{M*YX8_M<8ursG>FwcML zs!i{e)|K_KAnFDM+0zAP!2};(V*lKaxN+3!vA&b&*)k_!a`J^(sbWQnzzcFOTb|3$ zJypYBvYnEcPEZxYPa^xmCz(mmu=vXm+kD-LK0O~W~x@7v39`zxIk(`IS`98 zT;NR8iVK41JvLI~$(G`6)+|dQw$f(^PU6BN^L%geJyDJj<=iPI4Sx}LhE7X|2M={D zczG|i3SQ;EuJ=x!tJ6d?HKcRE;M6I~759$Ljl2PI&)FP}y6zdpM`71!cbJ zeRlV9ZgZd{+b1AcekWO>Tc6S&6T)t9CwQdleFEqnena2T6n{!Zes=)-HS zksxJCs8#DyyN(CN-!(~X?L5bmE#yotu_9{FtLf}f)zX#PnYZL!vn_f}GUDCCjfMhN z8&Uoj@zip+V-uxICr&oruE;8*D_=Z-{4!a-SWl@|nOXFbz*7ND7lt{5t{3CRMC2x? zB463g=>(9zPv~h?+<>GaT3!u5?Qo-dqY5|oXzw}dRQ)bBix6S6|Zm*aC*E7A^9$-dFEv!R$gE+ zb82ViHn;)i*CEMKzX(FqF&8(11)9>;W7fn!k3HEG87PF{B|*mE{H#OjcFN+FQrWG~ z12{`*#?($%fzs;fH4%KTn%H^cHrbT5#n$r5C$q6j% z;nh<^;sSvJ&%@Gx1B`Qj{QSPjyn+QY%yK$tpAnyOxSWQKTELidO>ajHB?oNdy1ng^H^{Q(`6q*a1#QQtVfR{ZiN7X^s9IWG zJGP{!qf9z0V$i#)PpE|(jUgQuxjF4BR93OAe@p*1_?mo_r(F!mXoN(mUlEYfW&U(g zJSRS0S*Zg$f%zJN9I=R7`8}V60-rhbzbWNqBE8ai>-O@EC<)S-vwJvYX)o#fqGw#l ziLLS%Em%`aIV-9yZE-DuBRG9gu`q+15dmwSgqkCSFPyAb-x#Ka+vw@kdZjZscZe-& z!*?iZ6iGID9d%dBhHoITuay5XS20ivB8ri{PSDdWuAvw?9R z{h&VW-8fcjb{a`Cnv@njh1R>Lt!=6Max$gn>NztiBPk>&&E%UOhrgM(CD-IlRQ1;j zi6FWAdA^dIvSj5lZN-edcv>V+=*NFbM7Xqq=HK%isd>XJ5~Y3DNhODQ4P)_0^Ge>c z3bt9j^sPg+64>tdmnl-n=yCSrqvzUtjH!_ciKW_#e=Iuvdj z8tyP+*3(PwPEx(-*YwJ(r^8LbU0LK=Pav%v8caGAyhJm2GlaZ`a-feS-KntMIlnWe zI6SYOEsU{ATn>nPr$&=hrPCqApLlmz(bl*C@}lost?hT!L*3Iww^d(IG&C92PPyKYLYFXzgi%L-N?)ZM0@nl)Fk7NZ~ z4w>lYfMqZf6Vmk%6K3GNE!$)hwOUu-!R6rYGC zm&^RTMlE--U_y{M$|`yOGtbA6rl@YZKl4wsIv*JXZN)33-VA7-=}JjdCRcm;n2gvp zL4OZqv)Q+_v}D%WRR1)X5!lD--3h>Ksiywek_rmA9#WWdb~Uo9zZ51I9g&U zH|P*U^lY{#y+j~fgRVcCpQ(Me&CoR3mPT0KT=!&0(_Oey9Q*#8E~&Z2Ddw>e*5F!#RBZS+ zs3L2H&w6_tBf@eZ_}*0*3m3~>tef!62ebW_=gAtlmGHEr_R1{oH(P%Y8$)`PYFs}y zQ8B$!gxEbdflkC~LO-_I!QNt#pWrtlcll?=?DVYYT0EsZO);}3)**Gh!Bwd9Ws%m) zyjtT}$>liFwjPm%nt^7U4LzsGU3}o()SN$1IzvlT+J9dXD}~hty{MC6c3Q z{QA<9%~r1VBk!O5!+hmOy%fRC5Y?mMml!GNsYDU|SJzWQf^RhLLCEx3L=M+at>wBE zS5BS5(nnC|V#Dql-I>5HJ8KFogT0TQL+5NR(KA|*@M^YO3Asi;NbMZz4fa&_VePt@ zo58>zT2%%3QEcCh}Yw??JBDIGucmGQq&Y5*N;z-^mxZQkZn-v zWM?pQe-8&~tIC~tJ+-~|ffBm5H?zYDcW+w+Xa-fpX4~r+wek^v;`AXpOZ;xtBX^b# zXKHw{6Zf(Rgq7Xc7cI{SjHm;FU%os+5jt06W2L}l^Y z-DC1^N!t=LPSx+2@!MOlII12g_Lf-SqmOT)ehEdX+`EK=xNAiE71dM2G3hzmO}wpj z>l`MhOJ%7q7_<3}%?3Y6UDP?^8B0-07v4!r%k#a8{EQ=g+g9$p$I0%6McWQ3uV=z5 z-28_a3y(uanpc~-R;!>SZ6D83pK3n7A{N%ZchAvO zHRdl{Ry-oYvvLyWXEU4|D>jFrrBIdGaid>sfyOn0Jl8fn(|6SyO3fD4@a58*W!;Y>~9_qyGUj=`{iX literal 0 HcmV?d00001 diff --git a/docs/index.md b/docs/index.md index fa99aa9ccaf3..60bcf19b84da 100755 --- a/docs/index.md +++ b/docs/index.md @@ -7,6 +7,7 @@ title: "Latest News" --- DeepSpeed empowers ChatGPT-like model training with a single click, offering 15x speedup over SOTA RLHF systems with unprecedented cost reduction at all scales; [learn how](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-chat). +* [2023/11] [DeepSpeed-FastGen: High-throughput Text Generation for LLMs via MII and DeepSpeed-Inference](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-fastgen) * [2023/10] [DeepSpeed-VisualChat: Improve Your Chat Experience with Multi-Round Multi-Image Inputs](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-visualchat/10-03-2023/README.md) [[English](https://github.com/microsoft/DeepSpeed/tree/master/blogs/deepspeed-visualchat/10-03-2023/README.md)] [[中文](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed-visualchat/10-03-2023/README-Chinese.md)] [[日本語](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed-visualchat/10-03-2023/README-Japanese.md)] * [2023/09] Announcing the DeepSpeed4Science Initiative: Enabling large-scale scientific discovery through sophisticated AI system technologies [[DeepSpeed4Science website](https://deepspeed4science.ai/)] [[Tutorials](/deepspeed4science/)] [[White paper](https://arxiv.org/abs/2310.04610)] [[Blog](https://www.microsoft.com/en-us/research/blog/announcing-the-deepspeed4science-initiative-enabling-large-scale-scientific-discovery-through-sophisticated-ai-system-technologies/)] [[中文](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed4science/chinese/README.md)] [[日本語](https://github.com/microsoft/DeepSpeed/blob/master/blogs/deepspeed4science/japanese/README.md)] * [2023/08] [DeepSpeed ZeRO-Inference: 20X faster inference through weight quantization and KV cache offloading](https://github.com/microsoft/DeepSpeedExamples/blob/master/inference/huggingface/zero_inference/README.md) From 09834bbe1ec4cbda2800535bec9de750ee88403e Mon Sep 17 00:00:00 2001 From: Jeff Rasley Date: Fri, 3 Nov 2023 15:34:59 -0700 Subject: [PATCH 05/13] bump to 0.12.0 --- version.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/version.txt b/version.txt index 1a96df19c09a..ac454c6a1fc3 100644 --- a/version.txt +++ b/version.txt @@ -1 +1 @@ -0.11.3 +0.12.0 From 58d3b65a215d344a22e4f96073ff48f9123c8a7b Mon Sep 17 00:00:00 2001 From: Michael Wyatt Date: Fri, 3 Nov 2023 15:38:05 -0700 Subject: [PATCH 06/13] Update version.txt after 0.12.0 release (#4611) Co-authored-by: jeffra --- version.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/version.txt b/version.txt index ac454c6a1fc3..34a83616bb5a 100644 --- a/version.txt +++ b/version.txt @@ -1 +1 @@ -0.12.0 +0.12.1 From ff53c224855888bab6574b59ee9a1a77a40d33d3 Mon Sep 17 00:00:00 2001 From: Masahiro Tanaka <81312776+tohtana@users.noreply.github.com> Date: Fri, 3 Nov 2023 16:57:00 -0700 Subject: [PATCH 07/13] Add number for latency comparison (#4612) This PR adds latency comparison --- blogs/deepspeed-fastgen/README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/blogs/deepspeed-fastgen/README.md b/blogs/deepspeed-fastgen/README.md index ad6bfda16626..6b440e35b675 100644 --- a/blogs/deepspeed-fastgen/README.md +++ b/blogs/deepspeed-fastgen/README.md @@ -165,10 +165,12 @@ When vLLM preempts the ongoing generation of previous requests, the generation l ### D. Token Level Timing Analysis Figure 5 displays the P50, P90, and P95 latencies of the generation processes. Both vLLM and DeepSpeed-FlexGen exhibit similar P50 latencies, but vLLM demonstrates significantly higher latencies for P90 and P95. +Regarding the P95 latencies, DeepSpeed-FlexGen achieved a reduction of 3.7 times. This discrepancy is due to a noticeable spike in vLLM's generation latency when it preempts the ongoing generation to process new prompts. In contrast, DeepSpeed-FastGen typically processes the prompt and generation for previous requests concurrently, leading to much more consistent generation latency. +