I encountered the following error while training with the ScanNetV2 dataset,what should I do?

[FAkor1]

Traceback (most recent call last):
File "tools/train.py", line 135, in
main()
File "tools/train.py", line 131, in main
runner.train()
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train
model = self.train_loop.run() # type: ignore
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmengine/runner/loops.py", line 98, in run
self.run_epoch()
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmengine/runner/loops.py", line 115, in run_epoch
self.run_iter(idx, data_batch)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmengine/runner/loops.py", line 131, in run_iter
outputs = self.runner.model.train_step(
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmengine/model/base_model/base_model.py", line 114, in train_step
losses = self._run_forward(data, mode='loss') # type: ignore
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmengine/model/base_model/base_model.py", line 361, in _run_forward
results = self(**data, mode=mode)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/mmdet3d/models/detectors/base.py", line 75, in forward
return self.loss(inputs, data_samples, **kwargs)
File "/root/autodl-tmp/oneformer3d-main/oneformer3d/oneformer3d.py", line 369, in loss
x = self.decoder(x, queries)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/root/autodl-tmp/oneformer3d-main/oneformer3d/query_decoder.py", line 340, in forward
return self.forward_iter_pred(x, queries)
File "/root/autodl-tmp/oneformer3d-main/oneformer3d/query_decoder.py", line 457, in forward_iter_pred
queries = self.cross_attn_layers[i](inst_feats, queries, attn_mask)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/root/autodl-tmp/oneformer3d-main/oneformer3d/query_decoder.py", line 52, in forward
output, _ = self.attn(queries[i], k, v, attn_mask=attn_mask)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/torch/nn/modules/activation.py", line 1003, in forward
attn_output, attn_output_weights = F.multi_head_attention_forward(
File "/root/miniconda3/envs/tr3d/lib/python3.8/site-packages/torch/nn/functional.py", line 4967, in multi_head_attention_forward
tgt_len, bsz, embed_dim = query.shape
ValueError: not enough values to unpack (expected 3, got 2)

[filaPro]

Please provide more info, e.g. the full output log of this run.

[FAkor1]

2024/10/24 10:02:26 - mmengine - INFO -

System environment:
sys.platform: linux
Python: 3.8.16 (default, Mar 2 2023, 03:21:46) [GCC 11.2.0]
CUDA available: True
MUSA available: False
numpy_random_seed: 1509538789
GPU 0: NVIDIA GeForce RTX 3090
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 11.3, V11.3.109
GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
PyTorch: 1.10.0
PyTorch compiling details: PyTorch built with:

• GCC 7.3

• C++ Version: 201402

• Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications

• Intel(R) MKL-DNN v2.2.3 (Git Hash 7336ca9f055cf1bfa13efb658fe15dc9b41f0740)

• OpenMP 201511 (a.k.a. OpenMP 4.5)

• LAPACK is enabled (usually provided by MKL)

• NNPACK is enabled

• CPU capability usage: AVX512

• CUDA Runtime 11.3

• NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37

• CuDNN 8.2

• Magma 2.5.2

• Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.10.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,

  TorchVision: 0.11.0
  OpenCV: 4.6.0
  MMEngine: 0.10.5

Runtime environment:
cudnn_benchmark: False
mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0}
dist_cfg: {'backend': 'nccl'}
seed: 1509538789
Distributed launcher: none
Distributed training: False
GPU number: 1

2024/10/24 10:02:27 - mmengine - INFO - Config:
backend_args = None
class_names = [
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
'unlabeled',
]
custom_hooks = [
dict(after_iter=True, type='EmptyCacheHook'),
]
custom_imports = dict(imports=[
'oneformer3d',
])
data_prefix = dict(
pts='points',
pts_instance_mask='instance_mask',
pts_semantic_mask='semantic_mask',
sp_pts_mask='super_points')
data_root = 'data/scannet/'
dataset_type = 'ScanNetSegDataset_'
default_hooks = dict(
checkpoint=dict(
scope='mmdet3d',
interval=1,
max_keep_ckpts=16,
type='CheckpointHook'),
logger=dict(scope='mmdet3d', interval=50, type='LoggerHook'),
param_scheduler=dict(scope='mmdet3d', type='ParamSchedulerHook'),
sampler_seed=dict(scope='mmdet3d', type='DistSamplerSeedHook'),
timer=dict(scope='mmdet3d', type='IterTimerHook'),
visualization=dict(scope='mmdet3d', type='Det3DVisualizationHook'))
default_scope = 'mmdet3d'
env_cfg = dict(
cudnn_benchmark=False,
dist_cfg=dict(backend='nccl'),
mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0))
eval_pipeline = [
dict(
scope='mmdet3d',
backend_args=None,
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(scope='mmdet3d', color_mean=None, type='NormalizePointsColor'),
dict(scope='mmdet3d', keys=[
'points',
], type='Pack3DDetInputs'),
]
input_modality = dict(use_camera=False, use_lidar=True)
inst_mapping = [
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
14,
16,
24,
28,
33,
34,
36,
39,
]
label2cat = dict({
0: 'wall',
1: 'floor',
10: 'picture',
11: 'counter',
12: 'desk',
13: 'curtain',
14: 'refrigerator',
15: 'showercurtrain',
16: 'toilet',
17: 'sink',
18: 'bathtub',
19: 'otherfurniture',
2: 'cabinet',
20: 'unlabeled',
3: 'bed',
4: 'chair',
5: 'sofa',
6: 'table',
7: 'door',
8: 'window',
9: 'bookshelf'
})
launcher = 'none'
load_from = 'work_dirs/tmp/sstnet_scannet.pth'
log_level = 'INFO'
log_processor = dict(
scope='mmdet3d', by_epoch=True, type='LogProcessor', window_size=50)
metainfo = dict(
classes=(
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
))
metric_meta = dict(
classes=[
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
'unlabeled',
],
dataset_name='ScanNet',
ignore_index=[
20,
],
label2cat=dict({
0: 'wall',
1: 'floor',
10: 'picture',
11: 'counter',
12: 'desk',
13: 'curtain',
14: 'refrigerator',
15: 'showercurtrain',
16: 'toilet',
17: 'sink',
18: 'bathtub',
19: 'otherfurniture',
2: 'cabinet',
20: 'unlabeled',
3: 'bed',
4: 'chair',
5: 'sofa',
6: 'table',
7: 'door',
8: 'window',
9: 'bookshelf'
}))
model = dict(
backbone=dict(
num_planes=[
32,
64,
96,
128,
160,
],
return_blocks=True,
type='SpConvUNet'),
criterion=dict(
inst_criterion=dict(
fix_dice_loss_weight=True,
fix_mean_loss=True,
iter_matcher=True,
loss_weight=[
0.5,
1.0,
1.0,
0.5,
],
matcher=dict(
costs=[
dict(type='QueryClassificationCost', weight=0.5),
dict(type='MaskBCECost', weight=1.0),
dict(type='MaskDiceCost', weight=1.0),
],
topk=1,
type='SparseMatcher'),
non_object_weight=0.1,
num_classes=18,
type='InstanceCriterion'),
num_semantic_classes=20,
sem_criterion=dict(
ignore_index=20, loss_weight=0.2, type='ScanNetSemanticCriterion'),
type='ScanNetUnifiedCriterion'),
data_preprocessor=dict(type='Det3DDataPreprocessor_'),
decoder=dict(
activation_fn='gelu',
attn_mask=True,
d_model=256,
dropout=0.0,
fix_attention=True,
hidden_dim=1024,
in_channels=32,
iter_pred=True,
num_heads=8,
num_instance_classes=18,
num_instance_queries=0,
num_layers=6,
num_semantic_classes=20,
num_semantic_linears=1,
num_semantic_queries=0,
objectness_flag=False,
type='ScanNetQueryDecoder'),
in_channels=6,
min_spatial_shape=128,
num_channels=32,
num_classes=18,
query_thr=0.5,
test_cfg=dict(
inst_score_thr=0.0,
matrix_nms_kernel='linear',
nms=True,
npoint_thr=100,
obj_normalization=True,
pan_score_thr=0.5,
sp_score_thr=0.4,
stuff_classes=[
0,
1,
],
topk_insts=600),
train_cfg=dict(),
type='ScanNetOneFormer3D',
voxel_size=0.02)
num_channels = 32
num_instance_classes = 18
num_points = 8192
num_semantic_classes = 20
optim_wrapper = dict(
clip_grad=dict(max_norm=10, norm_type=2),
optimizer=dict(lr=0.0001, type='AdamW', weight_decay=0.05),
type='OptimWrapper')
param_scheduler = dict(begin=0, end=512, power=0.9, type='PolyLR')
resume = False
sem_mapping = [
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
14,
16,
24,
28,
33,
34,
36,
39,
]
test_cfg = dict(type='TestLoop')
test_dataloader = dict(
batch_size=1,
dataset=dict(
scope='mmdet3d',
ann_file='scannet_oneformer3d_infos_val.pkl',
backend_args=None,
data_prefix=dict(
pts='points',
pts_instance_mask='instance_mask',
pts_semantic_mask='semantic_mask',
sp_pts_mask='super_points'),
data_root='data/scannet/',
ignore_index=20,
metainfo=dict(
classes=(
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
)),
modality=dict(use_camera=False, use_lidar=True),
pipeline=[
dict(
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(
type='LoadAnnotations3D_',
with_bbox_3d=False,
with_label_3d=False,
with_mask_3d=True,
with_seg_3d=True,
with_sp_mask_3d=True),
dict(type='PointSegClassMapping'),
dict(
flip=False,
img_scale=(
1333,
800,
),
pts_scale_ratio=1,
transforms=[
dict(
color_mean=[
127.5,
127.5,
127.5,
],
type='NormalizePointsColor_'),
dict(
merge_non_stuff_cls=False,
num_classes=20,
stuff_classes=[
0,
1,
],
type='AddSuperPointAnnotations'),
],
type='MultiScaleFlipAug3D'),
dict(keys=[
'points',
'sp_pts_mask',
], type='Pack3DDetInputs_'),
],
test_mode=True,
type='ScanNetSegDataset_'),
drop_last=False,
num_workers=1,
persistent_workers=True,
sampler=dict(scope='mmdet3d', shuffle=False, type='DefaultSampler'))
test_evaluator = dict(
scope='mmdet3d',
id_offset=65536,
inst_mapping=[
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
14,
16,
24,
28,
33,
34,
36,
39,
],
metric_meta=dict(
classes=[
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
'unlabeled',
],
dataset_name='ScanNet',
ignore_index=[
20,
],
label2cat=dict({
0: 'wall',
1: 'floor',
10: 'picture',
11: 'counter',
12: 'desk',
13: 'curtain',
14: 'refrigerator',
15: 'showercurtrain',
16: 'toilet',
17: 'sink',
18: 'bathtub',
19: 'otherfurniture',
2: 'cabinet',
20: 'unlabeled',
3: 'bed',
4: 'chair',
5: 'sofa',
6: 'table',
7: 'door',
8: 'window',
9: 'bookshelf'
})),
min_num_points=1,
sem_mapping=[
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
14,
16,
24,
28,
33,
34,
36,
39,
],
stuff_class_inds=[
0,
1,
],
thing_class_inds=[
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
],
type='UnifiedSegMetric')
test_pipeline = [
dict(
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(
type='LoadAnnotations3D_',
with_bbox_3d=False,
with_label_3d=False,
with_mask_3d=True,
with_seg_3d=True,
with_sp_mask_3d=True),
dict(type='PointSegClassMapping'),
dict(
flip=False,
img_scale=(
1333,
800,
),
pts_scale_ratio=1,
transforms=[
dict(
color_mean=[
127.5,
127.5,
127.5,
],
type='NormalizePointsColor_'),
dict(
merge_non_stuff_cls=False,
num_classes=20,
stuff_classes=[
0,
1,
],
type='AddSuperPointAnnotations'),
],
type='MultiScaleFlipAug3D'),
dict(keys=[
'points',
'sp_pts_mask',
], type='Pack3DDetInputs_'),
]
train_cfg = dict(
dynamic_intervals=[
(
1,
16,
),
(
496,
1,
),
],
max_epochs=512,
type='EpochBasedTrainLoop')
train_dataloader = dict(
batch_size=4,
dataset=dict(
scope='mmdet3d',
ann_file='scannet_oneformer3d_infos_train.pkl',
backend_args=None,
data_prefix=dict(
pts='points',
pts_instance_mask='instance_mask',
pts_semantic_mask='semantic_mask',
sp_pts_mask='super_points'),
data_root='data/scannet/',
ignore_index=20,
metainfo=dict(
classes=(
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
)),
modality=dict(use_camera=False, use_lidar=True),
pipeline=[
dict(
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(
type='LoadAnnotations3D_',
with_bbox_3d=False,
with_label_3d=False,
with_mask_3d=True,
with_seg_3d=True,
with_sp_mask_3d=True),
dict(type='PointSegClassMapping'),
dict(
flip_ratio_bev_horizontal=0.5,
flip_ratio_bev_vertical=0.5,
sync_2d=False,
type='RandomFlip3D'),
dict(
rot_range=[
-3.14,
3.14,
],
scale_ratio_range=[
0.8,
1.2,
],
shift_height=False,
translation_std=[
0.1,
0.1,
0.1,
],
type='GlobalRotScaleTrans'),
dict(
color_mean=[
127.5,
127.5,
127.5,
],
type='NormalizePointsColor_'),
dict(
merge_non_stuff_cls=False,
num_classes=20,
stuff_classes=[
0,
1,
],
type='AddSuperPointAnnotations'),
dict(
gran=[
6,
20,
],
mag=[
40,
160,
],
p=0.5,
type='ElasticTransfrom',
voxel_size=0.02),
dict(
keys=[
'points',
'gt_labels_3d',
'pts_semantic_mask',
'pts_instance_mask',
'sp_pts_mask',
'gt_sp_masks',
'elastic_coords',
],
type='Pack3DDetInputs_'),
],
scene_idxs=None,
test_mode=False,
type='ScanNetSegDataset_'),
num_workers=6,
persistent_workers=True,
sampler=dict(scope='mmdet3d', shuffle=True, type='DefaultSampler'))
train_pipeline = [
dict(
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(
type='LoadAnnotations3D_',
with_bbox_3d=False,
with_label_3d=False,
with_mask_3d=True,
with_seg_3d=True,
with_sp_mask_3d=True),
dict(type='PointSegClassMapping'),
dict(
flip_ratio_bev_horizontal=0.5,
flip_ratio_bev_vertical=0.5,
sync_2d=False,
type='RandomFlip3D'),
dict(
rot_range=[
-3.14,
3.14,
],
scale_ratio_range=[
0.8,
1.2,
],
shift_height=False,
translation_std=[
0.1,
0.1,
0.1,
],
type='GlobalRotScaleTrans'),
dict(color_mean=[
127.5,
127.5,
127.5,
], type='NormalizePointsColor_'),
dict(
merge_non_stuff_cls=False,
num_classes=20,
stuff_classes=[
0,
1,
],
type='AddSuperPointAnnotations'),
dict(
gran=[
6,
20,
],
mag=[
40,
160,
],
p=0.5,
type='ElasticTransfrom',
voxel_size=0.02),
dict(
keys=[
'points',
'gt_labels_3d',
'pts_semantic_mask',
'pts_instance_mask',
'sp_pts_mask',
'gt_sp_masks',
'elastic_coords',
],
type='Pack3DDetInputs_'),
]
tta_model = dict(scope='mmdet3d', type='Seg3DTTAModel')
tta_pipeline = [
dict(
scope='mmdet3d',
backend_args=None,
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(
scope='mmdet3d',
backend_args=None,
type='LoadAnnotations3D',
with_bbox_3d=False,
with_label_3d=False,
with_mask_3d=False,
with_seg_3d=True),
dict(scope='mmdet3d', color_mean=None, type='NormalizePointsColor'),
dict(
scope='mmdet3d',
transforms=[
[
dict(
flip_ratio_bev_horizontal=0.0,
flip_ratio_bev_vertical=0.0,
sync_2d=False,
type='RandomFlip3D'),
],
[
dict(keys=[
'points',
], type='Pack3DDetInputs'),
],
],
type='TestTimeAug'),
]
val_cfg = dict(type='ValLoop')
val_dataloader = dict(
batch_size=1,
dataset=dict(
scope='mmdet3d',
ann_file='scannet_oneformer3d_infos_val.pkl',
backend_args=None,
data_prefix=dict(
pts='points',
pts_instance_mask='instance_mask',
pts_semantic_mask='semantic_mask',
sp_pts_mask='super_points'),
data_root='data/scannet/',
ignore_index=20,
metainfo=dict(
classes=(
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
)),
modality=dict(use_camera=False, use_lidar=True),
pipeline=[
dict(
coord_type='DEPTH',
load_dim=6,
shift_height=False,
type='LoadPointsFromFile',
use_color=True,
use_dim=[
0,
1,
2,
3,
4,
5,
]),
dict(
type='LoadAnnotations3D_',
with_bbox_3d=False,
with_label_3d=False,
with_mask_3d=True,
with_seg_3d=True,
with_sp_mask_3d=True),
dict(type='PointSegClassMapping'),
dict(
flip=False,
img_scale=(
1333,
800,
),
pts_scale_ratio=1,
transforms=[
dict(
color_mean=[
127.5,
127.5,
127.5,
],
type='NormalizePointsColor_'),
dict(
merge_non_stuff_cls=False,
num_classes=20,
stuff_classes=[
0,
1,
],
type='AddSuperPointAnnotations'),
],
type='MultiScaleFlipAug3D'),
dict(keys=[
'points',
'sp_pts_mask',
], type='Pack3DDetInputs_'),
],
test_mode=True,
type='ScanNetSegDataset_'),
drop_last=False,
num_workers=1,
persistent_workers=True,
sampler=dict(scope='mmdet3d', shuffle=False, type='DefaultSampler'))
val_evaluator = dict(
scope='mmdet3d',
id_offset=65536,
inst_mapping=[
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
14,
16,
24,
28,
33,
34,
36,
39,
],
metric_meta=dict(
classes=[
'wall',
'floor',
'cabinet',
'bed',
'chair',
'sofa',
'table',
'door',
'window',
'bookshelf',
'picture',
'counter',
'desk',
'curtain',
'refrigerator',
'showercurtrain',
'toilet',
'sink',
'bathtub',
'otherfurniture',
'unlabeled',
],
dataset_name='ScanNet',
ignore_index=[
20,
],
label2cat=dict({
0: 'wall',
1: 'floor',
10: 'picture',
11: 'counter',
12: 'desk',
13: 'curtain',
14: 'refrigerator',
15: 'showercurtrain',
16: 'toilet',
17: 'sink',
18: 'bathtub',
19: 'otherfurniture',
2: 'cabinet',
20: 'unlabeled',
3: 'bed',
4: 'chair',
5: 'sofa',
6: 'table',
7: 'door',
8: 'window',
9: 'bookshelf'
})),
min_num_points=1,
sem_mapping=[
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
14,
16,
24,
28,
33,
34,
36,
39,
],
stuff_class_inds=[
0,
1,
],
thing_class_inds=[
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
],
type='UnifiedSegMetric')
vis_backends = [
dict(scope='mmdet3d', type='LocalVisBackend'),
]
visualizer = dict(
scope='mmdet3d',
name='visualizer',
type='Det3DLocalVisualizer',
vis_backends=[
dict(type='LocalVisBackend'),
])
work_dir = './work_dirs/oneformer3d_1xb4_scannet'

2024/10/24 10:02:32 - mmengine - INFO - Distributed training is not used, all SyncBatchNorm (SyncBN) layers in the model will be automatically reverted to BatchNormXd layers if they are used.
2024/10/24 10:02:32 - mmengine - INFO - Hooks will be executed in the following order:
before_run:
(VERY_HIGH ) RuntimeInfoHook
(BELOW_NORMAL) LoggerHook

before_train:
(VERY_HIGH ) RuntimeInfoHook
(NORMAL ) IterTimerHook
(VERY_LOW ) CheckpointHook

before_train_epoch:
(VERY_HIGH ) RuntimeInfoHook
(NORMAL ) IterTimerHook
(NORMAL ) DistSamplerSeedHook
(NORMAL ) EmptyCacheHook

before_train_iter:
(VERY_HIGH ) RuntimeInfoHook
(NORMAL ) IterTimerHook

after_train_iter:
(VERY_HIGH ) RuntimeInfoHook
(NORMAL ) IterTimerHook
(NORMAL ) EmptyCacheHook
(BELOW_NORMAL) LoggerHook
(LOW ) ParamSchedulerHook
(VERY_LOW ) CheckpointHook

after_train_epoch:
(NORMAL ) IterTimerHook
(NORMAL ) EmptyCacheHook
(LOW ) ParamSchedulerHook
(VERY_LOW ) CheckpointHook

before_val:
(VERY_HIGH ) RuntimeInfoHook

before_val_epoch:
(NORMAL ) IterTimerHook
(NORMAL ) EmptyCacheHook

before_val_iter:
(NORMAL ) IterTimerHook

after_val_iter:
(NORMAL ) IterTimerHook
(NORMAL ) Det3DVisualizationHook
(NORMAL ) EmptyCacheHook
(BELOW_NORMAL) LoggerHook

after_val_epoch:
(VERY_HIGH ) RuntimeInfoHook
(NORMAL ) IterTimerHook
(NORMAL ) EmptyCacheHook
(BELOW_NORMAL) LoggerHook
(LOW ) ParamSchedulerHook
(VERY_LOW ) CheckpointHook

after_val:
(VERY_HIGH ) RuntimeInfoHook

after_train:
(VERY_HIGH ) RuntimeInfoHook
(VERY_LOW ) CheckpointHook

before_test:
(VERY_HIGH ) RuntimeInfoHook

before_test_epoch:
(NORMAL ) IterTimerHook
(NORMAL ) EmptyCacheHook

before_test_iter:
(NORMAL ) IterTimerHook

after_test_iter:
(NORMAL ) IterTimerHook
(NORMAL ) Det3DVisualizationHook
(NORMAL ) EmptyCacheHook
(BELOW_NORMAL) LoggerHook

after_test_epoch:
(VERY_HIGH ) RuntimeInfoHook
(NORMAL ) IterTimerHook
(NORMAL ) EmptyCacheHook
(BELOW_NORMAL) LoggerHook

after_test:
(VERY_HIGH ) RuntimeInfoHook

after_run:
(BELOW_NORMAL) LoggerHook

2024/10/24 10:02:33 - mmengine - WARNING - The prefix is not set in metric class UnifiedSegMetric.
Name of parameter - Initialization information

unet.blocks.block0.conv_branch.0.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block0.conv_branch.0.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block0.conv_branch.2.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block0.conv_branch.3.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block0.conv_branch.3.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block0.conv_branch.5.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block1.conv_branch.0.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block1.conv_branch.0.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block1.conv_branch.2.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block1.conv_branch.3.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block1.conv_branch.3.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks.block1.conv_branch.5.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.conv.0.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.conv.0.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.conv.2.weight - torch.Size([64, 2, 2, 2, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block0.conv_branch.0.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block0.conv_branch.0.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block0.conv_branch.2.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block0.conv_branch.3.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block0.conv_branch.3.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block0.conv_branch.5.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block1.conv_branch.0.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block1.conv_branch.0.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block1.conv_branch.2.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block1.conv_branch.3.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block1.conv_branch.3.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks.block1.conv_branch.5.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.conv.0.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.conv.0.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.conv.2.weight - torch.Size([96, 2, 2, 2, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block0.conv_branch.0.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block0.conv_branch.0.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block0.conv_branch.2.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block0.conv_branch.3.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block0.conv_branch.3.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block0.conv_branch.5.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block1.conv_branch.0.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block1.conv_branch.0.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block1.conv_branch.2.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block1.conv_branch.3.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block1.conv_branch.3.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks.block1.conv_branch.5.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.conv.0.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.conv.0.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.conv.2.weight - torch.Size([128, 2, 2, 2, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block0.conv_branch.0.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block0.conv_branch.0.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block0.conv_branch.2.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block0.conv_branch.3.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block0.conv_branch.3.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block0.conv_branch.5.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block1.conv_branch.0.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block1.conv_branch.0.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block1.conv_branch.2.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block1.conv_branch.3.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block1.conv_branch.3.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks.block1.conv_branch.5.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.conv.0.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.conv.0.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.conv.2.weight - torch.Size([160, 2, 2, 2, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block0.conv_branch.0.weight - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block0.conv_branch.0.bias - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block0.conv_branch.2.weight - torch.Size([160, 3, 3, 3, 160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block0.conv_branch.3.weight - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block0.conv_branch.3.bias - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block0.conv_branch.5.weight - torch.Size([160, 3, 3, 3, 160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block1.conv_branch.0.weight - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block1.conv_branch.0.bias - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block1.conv_branch.2.weight - torch.Size([160, 3, 3, 3, 160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block1.conv_branch.3.weight - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block1.conv_branch.3.bias - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.u.blocks.block1.conv_branch.5.weight - torch.Size([160, 3, 3, 3, 160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.deconv.0.weight - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.deconv.0.bias - torch.Size([160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.deconv.2.weight - torch.Size([128, 2, 2, 2, 160]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.i_branch.0.weight - torch.Size([128, 1, 1, 1, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.conv_branch.0.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.conv_branch.0.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.conv_branch.2.weight - torch.Size([128, 3, 3, 3, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.conv_branch.3.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.conv_branch.3.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block0.conv_branch.5.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block1.conv_branch.0.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block1.conv_branch.0.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block1.conv_branch.2.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block1.conv_branch.3.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block1.conv_branch.3.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.u.blocks_tail.block1.conv_branch.5.weight - torch.Size([128, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.deconv.0.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.deconv.0.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.deconv.2.weight - torch.Size([96, 2, 2, 2, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.i_branch.0.weight - torch.Size([96, 1, 1, 1, 192]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.conv_branch.0.weight - torch.Size([192]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.conv_branch.0.bias - torch.Size([192]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.conv_branch.2.weight - torch.Size([96, 3, 3, 3, 192]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.conv_branch.3.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.conv_branch.3.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block0.conv_branch.5.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block1.conv_branch.0.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block1.conv_branch.0.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block1.conv_branch.2.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block1.conv_branch.3.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block1.conv_branch.3.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.u.blocks_tail.block1.conv_branch.5.weight - torch.Size([96, 3, 3, 3, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.deconv.0.weight - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.deconv.0.bias - torch.Size([96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.deconv.2.weight - torch.Size([64, 2, 2, 2, 96]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.i_branch.0.weight - torch.Size([64, 1, 1, 1, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.conv_branch.0.weight - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.conv_branch.0.bias - torch.Size([128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.conv_branch.2.weight - torch.Size([64, 3, 3, 3, 128]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.conv_branch.3.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.conv_branch.3.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block0.conv_branch.5.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block1.conv_branch.0.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block1.conv_branch.0.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block1.conv_branch.2.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block1.conv_branch.3.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block1.conv_branch.3.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.u.blocks_tail.block1.conv_branch.5.weight - torch.Size([64, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.deconv.0.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.deconv.0.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.deconv.2.weight - torch.Size([32, 2, 2, 2, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.i_branch.0.weight - torch.Size([32, 1, 1, 1, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.conv_branch.0.weight - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.conv_branch.0.bias - torch.Size([64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.conv_branch.2.weight - torch.Size([32, 3, 3, 3, 64]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.conv_branch.3.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.conv_branch.3.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block0.conv_branch.5.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block1.conv_branch.0.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block1.conv_branch.0.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block1.conv_branch.2.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block1.conv_branch.3.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block1.conv_branch.3.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

unet.blocks_tail.block1.conv_branch.5.weight - torch.Size([32, 3, 3, 3, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.input_proj.0.weight - torch.Size([256, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.input_proj.0.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.input_proj.1.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.input_proj.1.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.query_proj.0.weight - torch.Size([256, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.query_proj.0.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.query_proj.2.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.query_proj.2.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.0.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.0.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.0.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.0.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.0.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.0.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.1.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.1.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.1.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.1.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.1.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.1.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.2.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.2.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.2.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.2.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.2.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.2.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.3.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.3.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.3.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.3.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.3.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.3.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.4.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.4.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.4.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.4.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.4.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.4.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.5.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.5.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.5.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.5.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.5.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.cross_attn_layers.5.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.0.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.0.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.0.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.0.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.0.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.0.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.1.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.1.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.1.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.1.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.1.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.1.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.2.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.2.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.2.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.2.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.2.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.2.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.3.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.3.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.3.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.3.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.3.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.3.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.4.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.4.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.4.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.4.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.4.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.4.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.5.attn.in_proj_weight - torch.Size([768, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.5.attn.in_proj_bias - torch.Size([768]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.5.attn.out_proj.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.5.attn.out_proj.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.5.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.self_attn_layers.5.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.0.net.0.weight - torch.Size([1024, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.0.net.0.bias - torch.Size([1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.0.net.3.weight - torch.Size([256, 1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.0.net.3.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.0.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.0.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.1.net.0.weight - torch.Size([1024, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.1.net.0.bias - torch.Size([1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.1.net.3.weight - torch.Size([256, 1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.1.net.3.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.1.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.1.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.2.net.0.weight - torch.Size([1024, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.2.net.0.bias - torch.Size([1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.2.net.3.weight - torch.Size([256, 1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.2.net.3.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.2.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.2.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.3.net.0.weight - torch.Size([1024, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.3.net.0.bias - torch.Size([1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.3.net.3.weight - torch.Size([256, 1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.3.net.3.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.3.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.3.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.4.net.0.weight - torch.Size([1024, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.4.net.0.bias - torch.Size([1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.4.net.3.weight - torch.Size([256, 1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.4.net.3.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.4.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.4.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.5.net.0.weight - torch.Size([1024, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.5.net.0.bias - torch.Size([1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.5.net.3.weight - torch.Size([256, 1024]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.5.net.3.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.5.norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.ffn_layers.5.norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_norm.weight - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_norm.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_cls.0.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_cls.0.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_cls.2.weight - torch.Size([19, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_cls.2.bias - torch.Size([19]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.x_mask.0.weight - torch.Size([256, 32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.x_mask.0.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.x_mask.2.weight - torch.Size([256, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.x_mask.2.bias - torch.Size([256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_sem.weight - torch.Size([21, 256]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

decoder.out_sem.bias - torch.Size([21]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

input_conv.0.weight - torch.Size([32, 3, 3, 3, 6]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

output_layer.0.weight - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D

output_layer.0.bias - torch.Size([32]):
The value is the same before and after calling init_weights of ScanNetOneFormer3D
2024/10/24 10:02:34 - mmengine - INFO - Load checkpoint from work_dirs/tmp/sstnet_scannet.pth
2024/10/24 10:02:34 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io
2024/10/24 10:02:34 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future.
2024/10/24 10:02:34 - mmengine - INFO - Checkpoints will be saved to /root/autodl-tmp/oneformer3d-main/work_dirs/oneformer3d_1xb4_scannet.