pytorch/torch/distributed/_composable/contract.py
Andrew Gu b27695791e [PT-D] Relaxed contract to allow Sequence[nn.Module] (#127773)
This PR relaxes `@contract` to allow the 1st argument to be `Sequence[nn.Module]` instead of strictly `nn.Module`. This is required for the next PR, which allows `fully_shard` to take in `List[nn.Module]`.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/127773
Approved by: https://github.com/weifengpy
2024-07-15 23:54:10 +00:00

223 lines
9.3 KiB
Python

# mypy: allow-untyped-defs
import uuid
from collections import OrderedDict
from functools import wraps
from typing import Callable, Dict, List, Optional, Sequence, Type, Union
import torch
import torch.nn as nn
from torch.distributed._composable_state import _State
from torch.distributed.utils import _get_root_modules
def generate_state_key(string="__composable_api_state_key"):
return f"{string}_{str(uuid.uuid4())}"
STATE_KEY = generate_state_key()
REGISTRY_KEY = generate_state_key()
# TODO: we can add additional info to RegistryItem to share across APIs. E.g.,
# we can add args and kwargs here, and then we can detect whether fully_shard
# is combined with reentrant activation checkpointing and error out with a clear
# message.
class RegistryItem:
pass
def contract(state_cls: Type[_State] = _State):
r"""
Decorate a function as a composable distributed API, where the first
argument of the function must be an :class:`nn.Module` instance. The
decorator verifies that the wrapped function does not modify parameter,
buffer or sub-module fully-qualified names (FQN).
When a function ``func`` is decorated by ``@contract()``, a
``.state(module: nn.Module)`` method will be installed to the decorated
function. Then you can retrieve and modify the state on a module by calling
``func.state(module)``.
Example::
>>> # xdoctest: +SKIP
>>> import torch.nn as nn
>>>
>>> class MyModel(nn.Module):
>>> def __init__(self):
>>> super().__init__()
>>> self.l1 = nn.Linear(10, 10)
>>> self.l2 = nn.Linear(10, 10)
>>>
>>> def forward(self, x):
>>> return self.l2(self.l1(x))
>>>
>>> @contract()
>>> def my_feature(module: nn.Module) -> nn.Module:
>>> my_feature.state(module).some_state = "any value"
>>> return module
>>>
>>> model = MyModel()
>>> my_feature(model.l1)
>>> assert my_feature.state(model.l1).some_state == "any value"
>>> my_feature(model.l2)
>>> model(torch.randn(2, 10)).sum().backward()
"""
# wraps will make functions decorated with contract() pickleable - needed for integration with torch.package
@wraps(state_cls)
def inner(func):
@wraps(func)
def wrapper(
module: Union[nn.Module, Sequence[nn.Module]], *args, **kwargs
) -> Optional[nn.Module]:
inp_module = module
if isinstance(module, nn.Module):
modules = [module]
else:
# Special case: only FSDP permits a sequence of modules, in
# which case we only want to insert the state object on the
# root modules (i.e. those without a parent) with respect to
# the passed-in modules.
modules = _get_root_modules(list(module))
state = state_cls() # shared across all modules
registry_item = RegistryItem() # shared across all modules
module_to_orig_named_params: Dict[nn.Module, Dict[str, nn.Parameter]] = {}
module_to_orig_named_buffers: Dict[nn.Module, Dict[str, torch.Tensor]] = {}
module_to_orig_named_modules: Dict[nn.Module, Dict[str, nn.Module]] = {}
for module in modules:
default_all_state: Dict[Callable, _State] = OrderedDict()
default_registry: Dict[str, RegistryItem] = OrderedDict()
all_state: Dict[Callable, _State] = module.__dict__.setdefault( # type: ignore[call-overload]
STATE_KEY, default_all_state
)
assert isinstance(
all_state, dict
), "Distributed composable API states corrupted"
registry: Dict[str, RegistryItem] = module.__dict__.setdefault( # type: ignore[call-overload]
REGISTRY_KEY, default_registry
)
assert isinstance(
registry, dict
), "Distributed composable API registry corrupted"
# Make sure that func has not been applied to the module yet
assert func not in all_state and func.__name__ not in registry, (
"Each distinct composable distributed API can only be applied to a "
f"module once. {func.__name__} has already been applied to the "
f"following module:\n{module}"
)
all_state.setdefault(func, state)
registry.setdefault(func.__name__, registry_item)
module_to_orig_named_params[module] = OrderedDict(
module.named_parameters()
)
module_to_orig_named_buffers[module] = OrderedDict(
module.named_buffers(remove_duplicate=False)
)
module_to_orig_named_modules[module] = OrderedDict(
module.named_modules(remove_duplicate=False)
)
# `func` should return the same type as the input module/modules
updated = func(inp_module, *args, **kwargs)
if updated is None:
updated = inp_module
if isinstance(updated, nn.Module):
updated_modules = [updated]
else:
updated_modules = _get_root_modules(list(inp_module))
module_to_new_named_params: Dict[nn.Module, Dict[str, nn.Parameter]] = {}
module_to_new_named_buffers: Dict[nn.Module, Dict[str, torch.Tensor]] = {}
module_to_new_named_modules: Dict[nn.Module, Dict[str, nn.Module]] = {}
for module in updated_modules:
module_to_new_named_params[module] = OrderedDict(
module.named_parameters()
)
module_to_new_named_buffers[module] = OrderedDict(
module.named_buffers(remove_duplicate=False)
)
module_to_new_named_modules[module] = OrderedDict(
module.named_modules(remove_duplicate=False)
)
def check_fqn(orig_fqns: List[str], new_fqns: List[str], check_key: str):
if orig_fqns == new_fqns:
return
orig_fqn_set, new_fqn_set = set(orig_fqns), set(new_fqns)
orig_only = orig_fqn_set - new_fqn_set
new_only = new_fqn_set - orig_fqn_set
if len(orig_only) or len(new_only):
raise RuntimeError(
f"{check_key}"
"Composable distributed API implementations cannot modify "
"FQNs.\n"
f"Only in original FQNs: {orig_only},\n"
f"Only in new FQNs: {new_only}"
)
else:
raise RuntimeError(
f"{check_key}"
"Composable distributed API implementations cannot modify "
"the order of FQNs.\n"
f"Original FQNs: {orig_only}\n"
f"New FQNs: {new_only}"
)
if set(module_to_new_named_modules.keys()) != set(
module_to_orig_named_modules.keys()
):
raise RuntimeError(
f"{func.__name__} should not change the module structure.\n"
f"Before: {[str(type(m)) for m in module_to_orig_named_modules]}\n"
f"After: {[str(type(m)) for m in module_to_new_named_modules]}"
)
for module in module_to_new_named_modules:
check_fqn(
list(module_to_orig_named_params[module].keys()),
list(module_to_new_named_params[module].keys()),
"Check parameters, ",
)
check_fqn(
list(module_to_orig_named_buffers[module].keys()),
list(module_to_new_named_buffers[module].keys()),
"Check buffers, ",
)
check_fqn(
list(module_to_orig_named_modules[module].keys()),
list(module_to_new_named_modules[module].keys()),
"Check modules, ",
)
# TODO: a stricter verification should also reject changing module
# types and monkey-patching forward() method implementations.
# TODO: verify that installed distributed paradigms are compatible with
# each other.
return updated
def get_state(module: nn.Module) -> Optional[_State]:
return module.__dict__.setdefault( # type: ignore[call-overload]
STATE_KEY,
{}, # TODO(@yhcharles): this is a temporary fix, need a better way
).get(
func
) # type: ignore[call-overload]
wrapper.state = get_state # type: ignore[attr-defined]
return wrapper
return inner
def _get_registry(module: nn.Module) -> Optional[Dict[str, RegistryItem]]:
r"""
Get an ``OrderedDict`` of composable APIs that have been applied to the
``module``, indexed by the API name. If no API has been applied, then this
returns ``None``.
"""
return getattr(module, REGISTRY_KEY, None)