mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
51ff9ce997
Simplifies the code, such as making self.modules not a list and only a single module. Differential Revision: [D44899281](https://our.internmc.facebook.com/intern/diff/D44899281/) Pull Request resolved: https://github.com/pytorch/pytorch/pull/98889 Approved by: https://github.com/mrshenli, https://github.com/yhcharles
99 lines
3.1 KiB
Python
99 lines
3.1 KiB
Python
from typing import Optional, Tuple
|
|
|
|
import torch
|
|
import torch.nn as nn
|
|
|
|
from torch.nn.parallel import DistributedDataParallel
|
|
|
|
from .contract import _get_registry, contract
|
|
|
|
|
|
@contract()
|
|
def replicate(
|
|
module: nn.Module, # NOTE: contract now supports single module only
|
|
**kwargs,
|
|
) -> nn.Module:
|
|
r"""Replicates a module
|
|
|
|
Args:
|
|
module (torch.nn.Module): module to replicate
|
|
|
|
Example::
|
|
>>> # xdoctest: +REQUIRES(module:torch._C._distributed_c10d)
|
|
>>> module = nn.Linear(3, 3)
|
|
>>> replicate(module)
|
|
"""
|
|
torch._C._log_api_usage_once("torch.distributed.replicate")
|
|
_ReplicateState().mark_module(module, **kwargs)
|
|
return module
|
|
|
|
|
|
def _can_compose(module: nn.Module) -> bool:
|
|
r"""Check if module is composable for `replicate` API."""
|
|
return "fully_shard" not in _get_registry(module)
|
|
|
|
|
|
class _ReplicateState:
|
|
def __init__(self) -> None:
|
|
self.module: Optional[nn.Module] = None
|
|
self.has_initialized: bool = False
|
|
self._param_list: nn.ParameterList = nn.ParameterList()
|
|
self.kwargs: dict = {}
|
|
|
|
def mark_module(self, module: nn.Module, **kwargs) -> None:
|
|
if not _can_compose(module):
|
|
raise AssertionError(
|
|
"Cannot apply `replicate()` on a Module already managed by `fully_shard`"
|
|
)
|
|
self.module = module
|
|
replicate.state(module)._params_collected = False
|
|
module.register_forward_pre_hook(self.forward_pre_hook)
|
|
# TODO(@yhcharles): fix type error
|
|
module.register_forward_hook(self.forward_post_hook) # type: ignore[arg-type]
|
|
self.kwargs = kwargs
|
|
|
|
def _recursive_collect_params(self, module: nn.Module) -> None:
|
|
# skip if managed by other APIs
|
|
if not _can_compose(module):
|
|
return
|
|
|
|
# skip if module parameters already collected
|
|
replicate_state = replicate.state(module)
|
|
# if replicate_state is None, `module` is a child module that has not been explicitly
|
|
# tagged as replicate().
|
|
if replicate_state is not None:
|
|
if hasattr(replicate_state, "_params_collected"):
|
|
if replicate_state._params_collected:
|
|
return
|
|
replicate_state._params_collected = True
|
|
|
|
self._param_list.extend(
|
|
param for param in module.parameters(recurse=False) if param.requires_grad
|
|
)
|
|
for child in module.children():
|
|
self._recursive_collect_params(child)
|
|
|
|
def init_helper(self) -> None:
|
|
if self.has_initialized:
|
|
return
|
|
|
|
self.has_initialized = True
|
|
|
|
self._recursive_collect_params(self.module) # type: ignore[arg-type]
|
|
|
|
self._ddp = DistributedDataParallel(self._param_list, **self.kwargs)
|
|
|
|
def forward_pre_hook(
|
|
self, module: nn.Module, input: Tuple[torch.Tensor, ...]
|
|
) -> None:
|
|
self.init_helper()
|
|
self._ddp._pre_forward()
|
|
|
|
def forward_post_hook(
|
|
self,
|
|
module: nn.Module,
|
|
input: Tuple[torch.Tensor],
|
|
output: torch.Tensor,
|
|
) -> torch.Tensor:
|
|
return self._ddp._post_forward(output)
|