mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
a1a2f548a9
To make sure `fully_shard` and `replicate` can work together, we need to check for each other in the implementation. This change adds the check in `replicate()`. Pull Request resolved: https://github.com/pytorch/pytorch/pull/90711 Approved by: https://github.com/mrshenli
96 lines
2.9 KiB
Python
96 lines
2.9 KiB
Python
from typing import List, Tuple
|
|
|
|
import torch
|
|
import torch.nn as nn
|
|
|
|
from . import _ddp
|
|
from .contract import contract, _get_registry
|
|
|
|
|
|
@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::
|
|
>>> module = nn.Linear(3, 3)
|
|
>>> replicate(module)
|
|
"""
|
|
_ReplicateState().mark_modules(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.modules: List[nn.Module] = []
|
|
self.has_initialized: bool = False
|
|
self._param_list: nn.ParameterList = nn.ParameterList()
|
|
self.kwargs: dict = {}
|
|
|
|
def mark_modules(self, *modules: nn.Module, **kwargs) -> None:
|
|
for module in modules:
|
|
if not _can_compose(module):
|
|
raise AssertionError("Cannot apply `replicate()` on a Module already managed by `fully_shard`")
|
|
self.modules.append(module)
|
|
replicate.state(module)._distributed_state = self
|
|
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
|
|
if hasattr(replicate.state(module), "_params_collected"):
|
|
if replicate.state(module)._params_collected:
|
|
return
|
|
replicate.state(module)._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
|
|
for module in self.modules:
|
|
self._recursive_collect_params(module)
|
|
|
|
self._ddp = _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)
|