OptimizedLinear updates (#5791)

This is a refresh of of `OptimizedLinear` with the following features to
improve performance and usability:
 * More efficient sharing of base weights using `all_gather_into_tensor`
 * Flattened sharded weights
 * Selectively offload frozen weights to cpu
* `deepspeed.linear.Init` that allows injecting OptimizedLinear during
model construction (similar to zero.Init)
* Support for load state dict directly in OptimizedLinear, this allows
loading HF model weights correctly into sharded params
 * Various bug fixes for the LoRA implementation introduced previously
 * Several new unit tests
 
Builds on-top of @RezaYazdaniAminabadi's previous FP8 updates (#5764) to
support dense model fp8 quantization.

Example usage of this to fine-tune llama-3.1-405B on a single node:
https://github.com/Snowflake-Labs/snowflake-arctic/tree/main/training/llama3.1

---------

Co-authored-by: Reza Yazdani <[email protected]>
Co-authored-by: Reza Yazdani <[email protected]>

deepspeed/linear/__init__.py

@@ -5,3 +5,4 @@
 
 from .optimized_linear import OptimizedLinear
 from .config import LoRAConfig, QuantizationConfig
+from .context_manager import Init, init_lora

deepspeed/linear/config.py

@@ -3,7 +3,8 @@
 
 # DeepSpeed Team
 
-from dataclasses import dataclass
+from dataclasses import dataclass, field
+from typing import List
 
 
 @dataclass
@@ -17,10 +18,19 @@ class LoRAConfig:
         base_weight_sharding (int): The degree to which the base weights are sharded,
             should typically be set to the data-parallel world size to maximize the memory
             reduction benefits. Defaults to 1, which means this feature is disabled.
+        offload (bool): offload frozen parameters to cpu when not in use
+        offload_ratio (float): ratio of parameters to offload to cpu when not in use
+        delay_lora_init (bool): initialize lora parameters at time of model init or allow manual init later
+        target_mods (str): target module names to apply LoRA to, defaults to llama-3.1 arch
     """
     lora_r: int = 64
     lora_alpha: float = 16.
     base_weight_sharding: int = 1
+    offload: bool = False
+    offload_ratio: float = 0.0
+    delay_lora_init: bool = False
+    target_mods: List[str] = field(
+        default_factory=lambda: ['q_proj', 'k_proj', 'v_proj', 'o_proj', 'gate_proj', 'up_proj', 'down_proj'])
 
 
 @dataclass

deepspeed/linear/context_manager.py

@@ -0,0 +1,90 @@
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+from .optimized_linear import LoRAOptimizedLinear, OptimizedLinear
+
+import torch
+
+try:
+    import transformers
+except ImportError:
+    transformers = None
+
+
+def init_lora(model):
+    model.requires_grad_(False)
+    for m in model.modules():
+        if isinstance(m, LoRAOptimizedLinear):
+            m.init_lora()
+
+
+class Init(object):
+    """
+    Init context wrapper similar in style to zero.Init. Allows for injecting OptimizedLinear during model
+    construction which will shard base weights and reduce overall memory usage during model init. Primarily
+    useful when initializing a model via transformers.AutoModelForCausalLM.
+
+    Example usage:
+        lora_config = deepspeed.linear.LoRAConfig(..)
+        quant_config = deepspeed.linear.QuantizationConfig(..)
+        with deepspeed.linear.Init(lora_config=lora_config, quant_config=quant_config):
+            model = AutoModelForCausalLM.from_pretrained("meta-llama/Meta-Llama-3.1-405B")
+
+    """
+
+    def __init__(self, lora_config=None, quant_config=None):
+        self._orig_nn_linear = torch.nn.Linear
+        self._orig_causallm_pretrained = None
+        if transformers != None:
+            self._orig_causallm_pretrained = transformers.AutoModelForCausalLM.from_pretrained
+            self._orig_causallm_config = transformers.AutoModelForCausalLM.from_config
+        self.lora_config = lora_config
+        self.quant_config = quant_config
+        self._post_init_complete = False
+
+    def __enter__(self):
+
+        class OptLinearWrapper:
+            _orig_nn_linear = self._orig_nn_linear
+            _lora_config = self.lora_config
+            _quant_config = self.quant_config
+
+            def __new__(self, *args, **kwargs):
+                self._lora_config.delay_lora_init = True
+                kwargs['lora_config'] = self._lora_config
+                kwargs['quantization_config'] = self._quant_config
+                kwargs['linear_cls'] = self._orig_nn_linear
+                return OptimizedLinear(*args, **kwargs)
+
+        def _model_init(model):
+            if self.lora_config != None:
+                init_lora(model)
+            self._post_init_complete = True
+            return model
+
+        # ensures non-lora params are frozen and lora weights are initialized
+        def from_pretrained(*args, **kwargs):
+            model = self._orig_causallm_pretrained(*args, **kwargs)
+            return _model_init(model)
+
+        def from_config(*args, **kwargs):
+            model = self._orig_causallm_config(*args, **kwargs)
+            return _model_init(model)
+
+        torch.nn.Linear = OptLinearWrapper
+        if transformers != None:
+            transformers.AutoModelForCausalLM.from_pretrained = from_pretrained
+            transformers.AutoModelForCausalLM.from_config = from_config
+
+    def __exit__(self, *args, **kwargs):
+        torch.nn.Linear = self._orig_nn_linear
+        if not self._post_init_complete:
+            print('WARNING: For some reason LoRA modules are not initialized, this is usually done automatically '
+                  'if using transformers via (AutoModelForCausalLM from_pretrained/from_config). '
+                  'You must call `init_lora` on each module in order to use DeepSpeed LoRA, otherwise '
+                  'you will error out during runtime.')
+        else:
+            transformers.AutoModelForCausalLM.from_pretrained = self._orig_causallm_pretrained
+            transformers.AutoModelForCausalLM.from_config = self._orig_causallm_config

deepspeed/linear/optimized_linear.py

@@ -40,15 +40,17 @@ def __new__(self,
                 bias: bool = False,
                 lora_config: LoRAConfig = None,
                 quantization_config: QuantizationConfig = None,
-                dtype=torch.bfloat16):
+                device=None,
+                dtype=torch.bfloat16,
+                linear_cls=nn.Linear):
 
         if quantization_config is not None and not is_dataclass(quantization_config):
             raise ValueError(f"Expecting QuantizationConfig but received {type(quantization_config)}")
         if lora_config is not None and not is_dataclass(lora_config):
             raise ValueError(f"Expecting LoRAConfig but received {type(lora_config)}")
         if lora_config is None and quantization_config is None:
             # Everything disabled, fall back to normal nn.Linear
-            self = nn.Linear(input_dim, output_dim, bias=bias, dtype=dtype)
+            self = linear_cls(input_dim, output_dim, bias=bias, dtype=dtype, device=device)
 
         elif lora_config:
             # lora enabled, quantization may or may not be
@@ -57,7 +59,9 @@ def __new__(self,
                                        bias=bias,
                                        lora_config=lora_config,
                                        quantization_config=quantization_config,
-                                       dtype=dtype)
+                                       dtype=dtype,
+                                       device=device,
+                                       linear_cls=linear_cls)
 
         elif quantization_config:
             # only quantization enabled, no lora
@@ -78,72 +82,140 @@ def __init__(self,
                  lora_config: LoRAConfig = None,
                  quantization_config: QuantizationConfig = None,
                  device=None,
-                 dtype=torch.bfloat16):
+                 dtype=torch.bfloat16,
+                 linear_cls=nn.Linear):
         super().__init__()
         self.input_dim = input_dim
         self.output_dim = output_dim
         self.bias = bias
         self.lora_config = lora_config
         self.quantization_config = quantization_config
-        device = get_accelerator().current_device_name() if device is None else device
+        self.device = get_accelerator().current_device_name() if device is None else device
+        self.linear_cls = linear_cls
+        self.dtype = dtype
         assert self.lora_config is not None, "DSOptimizedLinear requires a LoRA config"
-
+        assert not self.bias, "bias=True is not supported by LoRAOptimizedLinear"
         self.zero_shards = self.lora_config.base_weight_sharding
         self.sharded_weight_size = int(float(self.input_dim) // self.zero_shards)
-        w = torch.nn.Parameter(torch.empty((self.output_dim, self.sharded_weight_size), dtype=dtype))
-        torch.nn.init.xavier_uniform_(w)
+        if self.zero_shards > 1:
+            assert self.zero_shards == dist.get_world_size(
+            ), "base weight sharding is only supported across world size"
+            w = torch.nn.Parameter(torch.empty(self.output_dim * self.sharded_weight_size, dtype=dtype),
+                                   requires_grad=False)
+        else:
+            w = torch.nn.Parameter(torch.empty((self.output_dim, self.input_dim), dtype=dtype), requires_grad=False)
+        torch.nn.init.xavier_uniform_(w.reshape(self.sharded_weight_size, self.output_dim))
 
         if self.quantization_config is not None:
             assert dtype == torch.bfloat16, "only bfloat16 is supported when using quantization"
-            self.base_weight = QuantizedParameter(w, quantization_config=quantization_config)
+            self.weight = QuantizedParameter(w, quantization_config=quantization_config)
         else:
-            self.base_weight = w
+            self.weight = w
+
+        self.disabled = False
+        self._initialized = False
+        if not self.lora_config.delay_lora_init:
+            self.init_lora()
+
+    def disable(self):
+        self.disabled = True
+        self.weight = torch.nn.Parameter(torch.empty((self.output_dim, self.input_dim), dtype=self.dtype),
+                                         requires_grad=False)
+
+    def init_lora(self):
+        if self.disabled:
+            return
+
+        if self.quantization_config is not None:
+            # ensure quant-param wasn't stripped, in some cases transformers will do this during model init
+            if not isinstance(self.weight, QuantizedParameter):
+                self.weight = QuantizedParameter(self.weight, quantization_config=self.quantization_config)
+
+        self._initialized = True
+        self.weight.requires_grad = False
 
-        self.base_weight.requires_grad = False
+        # Mark base weight to prevent broadcast and ensure proper offload behavior
+        self.weight.ds_optim_param = True
+
+        self.lora_scaling_factor = self.lora_config.lora_alpha / self.lora_config.lora_r
 
-        # Use RS lora for now.
-        self.lora_scaling_factor = self.lora_config.lora_alpha / math.sqrt(self.lora_config.lora_r)
         # Keeping lora weights in bf16 precision for ease of training.
-        self.lora_weight_1 = nn.Linear(self.input_dim,
-                                       self.lora_config.lora_r,
-                                       bias=self.bias,
-                                       device=device,
-                                       dtype=dtype)
-        self.lora_weight_2 = nn.Linear(self.lora_config.lora_r,
-                                       self.output_dim,
-                                       bias=self.bias,
-                                       device=device,
-                                       dtype=dtype)
+        self.lora_weight_1 = self.linear_cls(self.input_dim,
+                                             self.lora_config.lora_r,
+                                             bias=self.bias,
+                                             device=self.device,
+                                             dtype=self.dtype)
+        self.lora_weight_2 = self.linear_cls(self.lora_config.lora_r,
+                                             self.output_dim,
+                                             bias=self.bias,
+                                             device=self.device,
+                                             dtype=self.dtype)
+
+        # initialize "A" with kaiming uniform and "B" with zeros following this
+        # https://github.com/huggingface/peft/blob/62122b5add8d6892f70c82eaef2147a6ba33b90b/src/peft/tuners/lora/layer.py#L155
+        nn.init.kaiming_uniform_(self.lora_weight_1.weight, a=math.sqrt(5))
+        nn.init.zeros_(self.lora_weight_2.weight)
         self.lora_weight_1.weight.requires_grad = True
         self.lora_weight_2.weight.requires_grad = True
 
+    def _load_from_state_dict(self, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys,
+                              error_msgs):
+        if not any([target in prefix for target in self.lora_config.target_mods]):
+            # module does not match any target_mods, we must revert to normal nn.Linear via disable
+            self.disable()
+            return super()._load_from_state_dict(state_dict, prefix, local_metadata, strict, missing_keys,
+                                                 unexpected_keys, error_msgs)
+
+        if self.zero_shards > 1:
+            if not dist.is_initialized():
+                raise RuntimeError(
+                    "attempting to use optimized linear base weight sharding but torch-distributed is not initialized, please init first."
+                )
+            rank = dist.get_rank()
+            shape_local = self.output_dim * self.sharded_weight_size
+            base_weight_name = f"{prefix}weight"
+            incoming_param = state_dict[base_weight_name]
+            state_dict[base_weight_name] = incoming_param.flatten().narrow(0, rank * shape_local, shape_local)
+
+        return super()._load_from_state_dict(state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys,
+                                             error_msgs)
+
     def full_weight(self):
-        # This assumes weights are evenly sharded across gpus. which might not be correct.
-        # in that case, we should flatten before all_gather.
-        local_weight = self.base_weight.dequantized() if isinstance(self.base_weight,
-                                                                    QuantizedParameter) else self.base_weight
-        tensor_list = [
-            torch.zeros_like(local_weight, device=local_weight.device, dtype=local_weight.dtype)
-            for _ in range(self.zero_shards)
-        ]
-        dist.all_gather(tensor_list, local_weight)
-        weight = nn.Parameter(torch.cat([tensor for tensor in tensor_list], dim=1))
-        return weight
+        base_weight = self.weight
+        if getattr(base_weight, 'ds_offload', False):
+            # move to gpu so we can dequant and all-gather
+            assert base_weight.device == torch.device('cpu'), \
+                f"expected base weight on cpu but found {base_weight.device}"
+            base_weight.offload(revert=True)
+            local_weight = base_weight.dequantized() if isinstance(base_weight, QuantizedParameter) else base_weight
+            base_weight.offload()
+        else:
+            local_weight = base_weight.dequantized() if isinstance(base_weight, QuantizedParameter) else base_weight
+
+        tensor_out = torch.empty(self.output_dim * self.input_dim,
+                                 dtype=local_weight.dtype,
+                                 device=local_weight.device)
+        dist.all_gather_into_tensor(tensor_out, local_weight)
+        return tensor_out.reshape(self.output_dim, self.input_dim)
 
     def linear_without_F_linear(self, input, weight):
         output = torch.mm(input.reshape(-1, input.shape[-1]), weight)
         output = output.view(*input.shape[:-1], weight.shape[1])
         return output
 
     def forward(self, input_tensor):
+        if self.disabled:
+            return F.linear(input_tensor, self.weight)
+        assert self._initialized, "init_lora was never called, please initialize before proceeding"
+
         # Gather the sharded base weight
         if self.zero_shards > 1:
             with torch.no_grad():
                 base_weight = self.full_weight()
         elif self.quantization_config:
-            base_weight = self.base_weight.dequantized()
+            base_weight = self.weight.dequantized()
         else:
-            base_weight = self.base_weight
+            base_weight = self.weight
 
         base_weight_output = F.linear(input_tensor, base_weight)
         lora_output = self.lora_weight_2(self.lora_weight_1(input_tensor))

deepspeed/linear/quantization.py

@@ -75,6 +75,13 @@ def dequantized(self) -> torch.Tensor:
                                                  q_mantisa_bits=self.quantization_config.mantissa_bits)
         return self.data
 
+    def offload(self, revert=False):
+        if getattr(self, 'ds_offload', False):
+            if revert:
+                self.data = self.to(get_accelerator().current_device_name())
+            else:
+                self.data = self.to('cpu')
+
     def __getstate__(self):
         state = self.__dict__
         state["data"] = self.data
@@ -104,14 +111,17 @@ def __copy__(self):
         return new_instance
 
     def cuda(self, device=None, non_blocking=False):
-        return self.to(device="cuda" if device is None else device, non_blocking=non_blocking)
+        device = "cuda" if device is None else device
+        self.quantizer.to(device, non_blocking=non_blocking)
+        return self.to(device, non_blocking=non_blocking)
 
     def to(self, *args, **kwargs):
         """
         Move the parameter to the given device. Then, if the device is a cuda device,
         quantize it.
         """
         tensor = super().to(*args, **kwargs)
+        self.quantizer.to(*args, **kwargs)
         self._ensure_quantized(tensor)
         return tensor
 

deepspeed/ops/fp_quantizer/quantize.py

@@ -91,6 +91,13 @@ def quantize(self,
 
         return out
 
+    def to(self, *args, **kwargs):
+        # Intermediate tensors may need to be moved to different devices
+        if hasattr(self, 'input_q'):
+            self.input_q = self.input_q.to(*args, **kwargs)
+        if hasattr(self, 'scale'):
+            self.scale = self.scale.to(*args, **kwargs)
+
     def get_scales(self):
         return fp_quant_module.get_scales(self.scale, self.num_groups)