xhs-note-crawling/python/py/Lib/site-packages/torch/nn/modules/lazy.py

# mypy: allow-untyped-defs
import itertools
from typing import Any, Protocol

import torch
from torch.nn.parameter import is_lazy


__all__ = ["LazyModuleMixin"]


class _LazyProtocol(Protocol):
    """This class is used to avoid errors with mypy checks for the attributes in a mixin.

    https://mypy.readthedocs.io/en/latest/more_types.html#mixin-classes
    """

    def _register_load_state_dict_pre_hook(self, hook): ...

    def register_forward_pre_hook(self, hook, *, prepend=False, with_kwargs=False): ...

    def _lazy_load_hook(
        self,
        state_dict,
        prefix,
        local_metadata,
        strict,
        missing_keys,
        unexpected_keys,
        error_msgs,
    ): ...

    def _get_name(self): ...

    def _infer_parameters(self, module, input): ...

    @property
    def _parameters(self): ...

    @property
    def _buffers(self): ...

    @property
    def _non_persistent_buffers_set(self): ...

    @property
    def _load_hook(self): ...

    @property
    def _initialize_hook(self): ...


class LazyModuleMixin:
    r"""A mixin for modules that lazily initialize parameters, also known as "lazy modules".

    .. warning:
        Lazy modules are an experimental new feature under active development,
        and their API is likely to change.

    Modules that lazily initialize parameters, or "lazy modules",
    derive the shapes of their parameters from the first input(s)
    to their forward method. Until that first forward they contain
    :class:`torch.nn.UninitializedParameter` s that should not be accessed
    or used, and afterward they contain regular :class:`torch.nn.Parameter` s.
    Lazy modules are convenient since they don't require computing some
    module arguments, like the :attr:`in_features` argument of a
    typical :class:`torch.nn.Linear`.

    After construction, networks with lazy modules should first
    be converted to the desired dtype and placed on the expected device.
    This is because lazy modules only perform shape inference so the usual dtype
    and device placement behavior applies.
    The lazy modules should then perform "dry runs" to initialize all the components in the module.
    These "dry runs" send inputs of the correct size, dtype, and device through
    the network and to each one of its lazy modules. After this the network can be used as usual.

    >>> # xdoctest: +SKIP
    >>> class LazyMLP(torch.nn.Module):
    ...     def __init__(self) -> None:
    ...         super().__init__()
    ...         self.fc1 = torch.nn.LazyLinear(10)
    ...         self.relu1 = torch.nn.ReLU()
    ...         self.fc2 = torch.nn.LazyLinear(1)
    ...         self.relu2 = torch.nn.ReLU()
    ...
    ...     def forward(self, input):
    ...         x = self.relu1(self.fc1(input))
    ...         y = self.relu2(self.fc2(x))
    ...         return y
    >>> # constructs a network with lazy modules
    >>> lazy_mlp = LazyMLP()
    >>> # transforms the network's device and dtype
    >>> # NOTE: these transforms can and should be applied after construction and before any 'dry runs'
    >>> lazy_mlp = lazy_mlp.cuda()
    >>> lazy_mlp
    LazyMLP( (fc1): LazyLinear(in_features=0, out_features=10, bias=True)
      (relu1): ReLU()
      (fc2): LazyLinear(in_features=0, out_features=1, bias=True)
      (relu2): ReLU()
    )
    >>> # performs a dry run to initialize the network's lazy modules
    >>> lazy_mlp(torch.ones(10, 10).cuda())
    >>> # after initialization, LazyLinear modules become regular Linear modules
    >>> lazy_mlp
    LazyMLP(
      (fc1): Linear(in_features=10, out_features=10, bias=True)
      (relu1): ReLU()
      (fc2): Linear(in_features=10, out_features=1, bias=True)
      (relu2): ReLU()
    )
    >>> # attaches an optimizer, since parameters can now be used as usual
    >>> optim = torch.optim.SGD(lazy_mlp.parameters(), lr=0.01)

    A final caveat when using lazy modules is that the order of initialization of a network's
    parameters may change, since the lazy modules are always initialized after other modules.
    For example, if the LazyMLP class defined above had a :class:`torch.nn.LazyLinear` module
    first and then a regular :class:`torch.nn.Linear` second, the second module would be
    initialized on construction and the first module would be initialized during the first dry run.
    This can cause the parameters of a network using lazy modules to be initialized differently
    than the parameters of a network without lazy modules as the order of parameter initializations,
    which often depends on a stateful random number generator, is different.
    Check :doc:`/notes/randomness` for more details.

    Lazy modules can be serialized with a state dict like other modules. For example:

    >>> lazy_mlp = LazyMLP()
    >>> # The state dict shows the uninitialized parameters
    >>> lazy_mlp.state_dict()
    OrderedDict({'fc1.weight': <UninitializedParameter>,
                 'fc1.bias': <UninitializedParameter>,
                 'fc2.weight': <UninitializedParameter>,
                 'fc2.bias': <UninitializedParameter>})

    Lazy modules can load regular :class:`torch.nn.Parameter` s (i.e. you can serialize/deserialize
    initialized LazyModules and they will remain initialized)


    >>> full_mlp = LazyMLP()
    >>> # Dry run to initialize another module
    >>> full_mlp.forward(torch.ones(10, 1))
    >>> # Load an initialized state into a lazy module
    >>> lazy_mlp.load_state_dict(full_mlp.state_dict())
    >>> # The state dict now holds valid values
    >>> lazy_mlp.state_dict()
    OrderedDict([('fc1.weight',
                  tensor([[-0.3837],
                          [ 0.0907],
                          [ 0.6708],
                          [-0.5223],
                          [-0.9028],
                          [ 0.2851],
                          [-0.4537],
                          [ 0.6813],
                          [ 0.5766],
                          [-0.8678]])),
                 ('fc1.bias',
                  tensor([-1.8832e+25,  4.5636e-41, -1.8832e+25,  4.5636e-41, -6.1598e-30,
                           4.5637e-41, -1.8788e+22,  4.5636e-41, -2.0042e-31,  4.5637e-41])),
                 ('fc2.weight',
                  tensor([[ 0.1320,  0.2938,  0.0679,  0.2793,  0.1088, -0.1795, -0.2301,  0.2807,
                            0.2479,  0.1091]])),
                 ('fc2.bias', tensor([0.0019]))])

    Note, however, that the loaded parameters will not be replaced when doing a "dry run" if they are initialized
    when the state is loaded. This prevents using initialized modules in different contexts.
    """

    # modules inheriting from this will change their __class__ to the specified
    # one after they are fully initialized
    cls_to_become: type[Any] | None = None

    def __init__(self: _LazyProtocol, *args, **kwargs):
        # Mypy doesn't like this super call in a mixin
        super().__init__(*args, **kwargs)  # type: ignore[misc]
        # pyrefly: ignore [read-only]
        self._load_hook = self._register_load_state_dict_pre_hook(self._lazy_load_hook)
        # pyrefly: ignore [read-only]
        self._initialize_hook = self.register_forward_pre_hook(
            self._infer_parameters, with_kwargs=True
        )

    def _save_to_state_dict(self: _LazyProtocol, destination, prefix, keep_vars):
        # This should be ideally implemented as a hook,
        # but we should override `detach` in the UninitializedParameter to return itself
        # which is not clean
        for name, param in self._parameters.items():
            if param is not None:
                if not (is_lazy(param) or keep_vars):
                    param = param.detach()
                destination[prefix + name] = param
        for name, buf in self._buffers.items():
            if buf is not None and name not in self._non_persistent_buffers_set:
                if not (is_lazy(buf) or keep_vars):
                    buf = buf.detach()
                destination[prefix + name] = buf

    def _lazy_load_hook(
        self: _LazyProtocol,
        state_dict,
        prefix,
        local_metadata,
        strict,
        missing_keys,
        unexpected_keys,
        error_msgs,
    ):
        """load_state_dict pre-hook function for lazy buffers and parameters.

        The purpose of this hook is to adjust the current state and/or
        ``state_dict`` being loaded so that a module instance serialized in
        both un/initialized state can be deserialized onto both un/initialized
        module instance.
        See comment in ``torch.nn.Module._register_load_state_dict_pre_hook``
        for the details of the hook specification.
        """
        for name, param in itertools.chain(
            self._parameters.items(), self._buffers.items()
        ):
            key = prefix + name
            if key in state_dict and param is not None:
                input_param = state_dict[key]
                if is_lazy(param):
                    # The current parameter is not initialized but the one being loaded one is
                    # create a new parameter based on the uninitialized one
                    if not is_lazy(input_param):
                        with torch.no_grad():
                            param.materialize(input_param.shape)

    def initialize_parameters(self: _LazyProtocol, *args, **kwargs):
        r"""Initialize parameters according to the input batch properties.

        This adds an interface to isolate parameter initialization from the
        forward pass when doing parameter shape inference.
        """
        raise NotImplementedError(
            f"initialize_parameters is not implemented for {self.__class__.__name__}"
        )

    def has_uninitialized_params(self: _LazyProtocol):
        r"""Check if a module has parameters that are not initialized."""
        # This is to avoid the JIT to track this parameter and force
        # custom modules __setstate__ to add it
        params = self._parameters.values()
        buffers = self._buffers.values()
        for param in itertools.chain(params, buffers):
            if is_lazy(param):
                return True
        return False

    # torchrec tests the code consistency with the following code
    # fmt: off
    def _infer_parameters(self: _LazyProtocol, module, args, kwargs=None):
        r"""Infers the size and initializes the parameters according to the provided input batch.

        Given a module that contains parameters that were declared inferable
        using :class:`torch.nn.parameter.ParameterMode.Infer`, runs a forward pass
        in the complete module using the provided input to initialize all the parameters
        as needed.
        The module is set into evaluation mode before running the forward pass in order
        to avoid saving statistics or calculating gradients
        """
        kwargs = kwargs if kwargs else {}
        module.initialize_parameters(*args, **kwargs)
        if module.has_uninitialized_params():
            raise RuntimeError(f'module {self._get_name()} has not been fully initialized')
        module._initialize_hook.remove()
        module._load_hook.remove()
        delattr(module, '_initialize_hook')
        delattr(module, '_load_hook')
        if module.cls_to_become is not None:
            module.__class__ = module.cls_to_become
    # fmt: on

    def _replicate_for_data_parallel(self: _LazyProtocol):
        raise RuntimeError(
            "Modules with uninitialized parameters can't be used with `DataParallel`. "
            "Run a dummy forward pass to correctly initialize the modules"
        )