wehub-resource-sync
213 lines
8.5 KiB
Python
213 lines
8.5 KiB
Python
# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# 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.
|
|
|
|
import os
|
|
|
|
import paddle
|
|
|
|
from ...trainer.argparser import strtobool
|
|
from ...trainer.trainer import logger
|
|
|
|
|
|
@paddle.no_grad()
|
|
def _move_param(src, device=None, blocking=True):
|
|
"""
|
|
Move parameters from the source device to the target device and return the parameters on the target device.
|
|
If the target device is not specified, the current device is used.
|
|
|
|
Args:
|
|
src (Tensor): The tensor of parameters to be moved.
|
|
device (Optional[Union[str, paddle.Device]], optional): The target device. Can be a string or paddle.Device object.
|
|
Defaults to None, which means using the current device.
|
|
blocking (bool, optional): Whether to block until the operation is complete. Defaults to True.
|
|
|
|
Returns:
|
|
Tensor: The tensor of parameters on the target device.
|
|
"""
|
|
if isinstance(device, str):
|
|
device = paddle.device._convert_to_place(device)
|
|
dst = src._copy_to(device, blocking)
|
|
dst_tensor = dst.value().get_tensor()
|
|
src_tensor = src.value().get_tensor()
|
|
src_tensor._clear()
|
|
src_tensor._share_data_with(dst_tensor)
|
|
|
|
|
|
def offload_tensor_to_cpu(tensors):
|
|
"""
|
|
Migrate the given tensors to CPU. This function has no effect if CUDA managed memory is used.
|
|
|
|
Args:
|
|
tensors (tuple, list): A tuple or list containing two elements. The first element is the model or optimizer,
|
|
and the second element is a string indicating whether it is a model or optimizer.
|
|
|
|
Returns:
|
|
None: No return value, modifies the original tensors directly.
|
|
|
|
Raises:
|
|
None: Does not raise any exceptions.
|
|
"""
|
|
if strtobool(os.getenv("FLAGS_use_cuda_managed_memory", "False")):
|
|
logger.warning("FLAGS_use_cuda_managed_memory has been set to True, offloading strategy is ineffective.")
|
|
return
|
|
|
|
pin_device = paddle.CUDAPinnedPlace()
|
|
|
|
def clear_main_grad(model):
|
|
for param in model.parameters():
|
|
if hasattr(param, "main_grad") and param.main_grad is not None:
|
|
param.main_grad._clear_data()
|
|
param.main_grad = None
|
|
|
|
# optimizer
|
|
if "optimizer" in tensors[1]:
|
|
optimizer = tensors[0]
|
|
# offload moment1
|
|
for key, value in optimizer._accumulators[optimizer._moment1_acc_str].items():
|
|
if value._is_initialized() and not isinstance(value.place, paddle.CUDAPinnedPlace):
|
|
optimizer._accumulators[optimizer._moment1_acc_str][key] = value.pin_memory()
|
|
|
|
# offload moment2
|
|
for key, value in optimizer._accumulators[optimizer._moment2_acc_str].items():
|
|
if value._is_initialized() and not isinstance(value.place, paddle.CUDAPinnedPlace):
|
|
optimizer._accumulators[optimizer._moment2_acc_str][key] = value.pin_memory()
|
|
|
|
# offload master_weight
|
|
for key, value in optimizer._master_weights.items():
|
|
if value._is_initialized() and not isinstance(value.place, paddle.CUDAPinnedPlace):
|
|
optimizer._master_weights[key] = value.pin_memory()
|
|
# model
|
|
elif "model" in tensors[1]:
|
|
model = tensors[0]
|
|
clear_main_grad(model)
|
|
for name, src in model.named_parameters():
|
|
if src._is_initialized() and not isinstance(src.place, paddle.CUDAPinnedPlace):
|
|
_move_param(src, pin_device)
|
|
|
|
elif "tensor" in tensors[1]:
|
|
src = tensors[0]
|
|
if src._is_initialized() and not isinstance(src.place, paddle.CUDAPinnedPlace):
|
|
_move_param(src, pin_device)
|
|
else:
|
|
logger.debug(f"[offload_tensor_to_cpu]Can't parse for type {tensors[1]}")
|
|
|
|
|
|
def reload_tensor_to_gpu(tensors):
|
|
"""
|
|
Transfer the given tensors from CPU to GPU and return new tensors. This function has no effect if the environment variable
|
|
FLAGS_use_cuda_managed_memory is not set to True.
|
|
|
|
Args:
|
|
tensors (List[Tuple[Any, str]]): A list containing tuples. Each tuple has two elements: the tensor to be transferred
|
|
to GPU and a string indicating the tensor type ("optimizer" or "model").
|
|
|
|
Returns:
|
|
List[Tuple[Any, str]]: The same list as the input, but all tensors have been transferred to GPU.
|
|
|
|
Raises:
|
|
None.
|
|
"""
|
|
if strtobool(os.getenv("FLAGS_use_cuda_managed_memory", "False")):
|
|
logger.warning("FLAGS_use_cuda_managed_memory has been set to True, offloading strategy is ineffective.")
|
|
return
|
|
|
|
# optimizer
|
|
if "optimizer" in tensors[1]:
|
|
optimizer = tensors[0]
|
|
# offload moment1
|
|
for key, value in optimizer._accumulators[optimizer._moment1_acc_str].items():
|
|
if value._is_initialized() and not isinstance(value.place, paddle.CUDAPlace):
|
|
optimizer._accumulators[optimizer._moment1_acc_str][key] = value.cuda()
|
|
|
|
# offload moment2
|
|
for key, value in optimizer._accumulators[optimizer._moment2_acc_str].items():
|
|
if value._is_initialized() and not isinstance(value.place, paddle.CUDAPlace):
|
|
optimizer._accumulators[optimizer._moment2_acc_str][key] = value.cuda()
|
|
|
|
# offload master_weight
|
|
for key, value in optimizer._master_weights.items():
|
|
if value._is_initialized() and not isinstance(value.place, paddle.CUDAPlace):
|
|
optimizer._master_weights[key] = value.cuda()
|
|
# model
|
|
elif "model" in tensors[1]:
|
|
model = tensors[0]
|
|
device = paddle.device.get_device()
|
|
for name, src in model.named_parameters():
|
|
if src._is_initialized() and not isinstance(src.place, paddle.CUDAPlace):
|
|
_move_param(src, device)
|
|
else:
|
|
logger.debug(f"[reload_tensor_to_gpu]Can't parse for type {tensors[1]}")
|
|
|
|
|
|
class OffloadController:
|
|
def __init__(self, objs):
|
|
self.objs = objs
|
|
|
|
def __enter__(self):
|
|
for obj in self.objs:
|
|
if hasattr(obj[0], "enable"):
|
|
obj[0].enable()
|
|
else:
|
|
if obj[1] != "":
|
|
reload_tensor_to_gpu(obj)
|
|
# offload_tensor_to_cpu/reload_tensor_to_gpu use non-blocking copy
|
|
# maybe overlap with compute later
|
|
if len(self.objs) > 0:
|
|
paddle.device.synchronize()
|
|
|
|
def __exit__(self, *args):
|
|
for obj in self.objs:
|
|
if hasattr(obj[0], "disable"):
|
|
obj[0].disable()
|
|
else:
|
|
if obj[1] != "":
|
|
offload_tensor_to_cpu(obj)
|
|
# offload_tensor_to_cpu/reload_tensor_to_gpu use non-blocking copy
|
|
# maybe overlap with compute later
|
|
if len(self.objs) > 0:
|
|
paddle.device.synchronize()
|
|
|
|
|
|
def reload_and_offload_scope(trainer, *args):
|
|
offload_map = {
|
|
trainer.actor_model: "train_model",
|
|
trainer.reference_model: "freeze_model",
|
|
**(
|
|
{trainer.reward_model: "freeze_model"}
|
|
if not trainer.args.use_rm_server and not trainer.args.use_rule_reward
|
|
else {}
|
|
),
|
|
trainer.actor_trainer.optimizer: "optimizer",
|
|
}
|
|
|
|
if trainer.args.rl_algorithm == "ppo":
|
|
offload_map.update(
|
|
{
|
|
trainer.critic_model: "train_model",
|
|
trainer.critic_trainer.optimizer: "optimizer",
|
|
}
|
|
)
|
|
|
|
objs = [(arg, offload_map.get(arg, "")) for arg in args if offload_map.get(arg, "") in trainer.args.offload_level]
|
|
if trainer.actor_model not in [i for i, _ in objs]:
|
|
if getattr(trainer.actor_trainer, "_inner_eval_model", None) is not None:
|
|
# NOTE(gongenlei): for export_evaluate_model
|
|
objs.append((trainer.actor_model, offload_map.get(trainer.actor_model, "")))
|
|
if trainer.args.rl_algorithm == "ppo":
|
|
if trainer.critic_model not in [i for i, _ in objs]:
|
|
if getattr(trainer.critic_trainer, "_inner_eval_model", None) is not None:
|
|
# NOTE(gongenlei): for export_evaluate_model
|
|
objs.append((trainer.critic_model, offload_map.get(trainer.critic_model, "")))
|
|
return OffloadController(objs)
|