import torch
from importlib.util import find_spec


def amp_definitely_not_available():
    return not (torch.cuda.is_available() or find_spec('torch_xla'))

            

Reported by Pylint.

              import torch
from importlib.util import find_spec


def amp_definitely_not_available():
    return not (torch.cuda.is_available() or find_spec('torch_xla'))

            

Reported by Pylint.

              from importlib.util import find_spec


def amp_definitely_not_available():
    return not (torch.cuda.is_available() or find_spec('torch_xla'))

            

Reported by Pylint.

              # !/usr/bin/env python3

# Copyright (c) Facebook, Inc. and its affiliates.
# All rights reserved.
#
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree.

# Taken and modified from original source:

            

Reported by Pylint.

              logger = logging.getLogger(__name__)


def get_libc():
    if IS_WINDOWS or IS_MACOS:
        logger.warning(
            "NOTE: Redirects are currently not supported in Windows or MacOs."
        )
        return None

            

Reported by Pylint.

              

def get_libc():
    if IS_WINDOWS or IS_MACOS:
        logger.warning(
            "NOTE: Redirects are currently not supported in Windows or MacOs."
        )
        return None
    else:

            

Reported by Pylint.

              # This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree.

from .cycling_iterator import CyclingIterator  # noqa: F401
from .elastic_distributed_sampler import ElasticDistributedSampler  # noqa: F401

            

Reported by Pylint.

              # LICENSE file in the root directory of this source tree.

from .cycling_iterator import CyclingIterator  # noqa: F401
from .elastic_distributed_sampler import ElasticDistributedSampler  # noqa: F401

            

Reported by Pylint.

              #!/usr/bin/env python3

# Copyright (c) Facebook, Inc. and its affiliates.
# All rights reserved.
#
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree.

from .cycling_iterator import CyclingIterator  # noqa: F401

            

Reported by Pylint.

              logger = logging.getLogger(__name__)


def parse_args(args):
    parser = get_args_parser()
    parser.add_argument(
        "--use_env",
        default=False,
        action="store_true",

            

Reported by Pylint.

                  return parser.parse_args(args)


def launch(args):
    if args.no_python and not args.use_env:
        raise ValueError(
            "When using the '--no_python' flag,"
            " you must also set the '--use_env' flag."
        )

            

Reported by Pylint.

                  run(args)


def main(args=None):
    warnings.warn(
        "The module torch.distributed.launch is deprecated\n"
        "and will be removed in future. Use torch.distributed.run.\n"
        "Note that --use_env is set by default in torch.distributed.run.\n"
        "If your script expects `--local_rank` argument to be set, please\n"

            

Reported by Pylint.

              from .api.remote_module import RemoteModule
from .functional import *  # noqa: F403

            

Reported by Pylint.

              from .api.remote_module import RemoteModule
from .functional import *  # noqa: F403

            

Reported by Pylint.

              from .api.remote_module import RemoteModule
from .functional import *  # noqa: F403

            

Reported by Pylint.

              # This template may cause typing error (the mismatch between ``Tuple[()]`` and ``Tuple[Any]``)
# even if the code is only used for instaniation but not execution.
# Therefore, only include handling moving CPU tensors to a cuda device if necessary.
# TODO: Merge these two templates together in the future once TorchScript syntax is improved.
_REMOTE_FORWARD_TEMPLATE_ENABLE_MOVING_CPU_TENSORS_TO_CUDA = """
def _remote_forward(
    module_rref: RRef[module_interface_cls], device: str, is_device_map_set: bool, {arg_types}){arrow_and_return_type}:
    module = module_rref.local_value()
    device = torch.device(device)

            

Reported by Pylint.

              #!/usr/bin/python3


def get_remote_module_template(enable_moving_cpu_tensors_to_cuda: bool):
    return _TEMPLATE_PREFIX + (
        _REMOTE_FORWARD_TEMPLATE_ENABLE_MOVING_CPU_TENSORS_TO_CUDA
        if enable_moving_cpu_tensors_to_cuda
        else _REMOTE_FORWARD_TEMPLATE
    )

            

Reported by Pylint.

              #!/usr/bin/python3


def get_remote_module_template(enable_moving_cpu_tensors_to_cuda: bool):
    return _TEMPLATE_PREFIX + (
        _REMOTE_FORWARD_TEMPLATE_ENABLE_MOVING_CPU_TENSORS_TO_CUDA
        if enable_moving_cpu_tensors_to_cuda
        else _REMOTE_FORWARD_TEMPLATE
    )

            

Reported by Pylint.

              # This source code is licensed under the BSD license found in the
# LICENSE file in the root directory of this source tree.
"""A Pipe implementation in PyTorch."""
from .checkpoint import is_checkpointing, is_recomputing
from .pipe import Pipe
from .microbatch import NoChunk

__all__ = ["Pipe", "is_checkpointing", "is_recomputing"]

            

Reported by Pylint.

              # LICENSE file in the root directory of this source tree.
"""A Pipe implementation in PyTorch."""
from .checkpoint import is_checkpointing, is_recomputing
from .pipe import Pipe
from .microbatch import NoChunk

__all__ = ["Pipe", "is_checkpointing", "is_recomputing"]

            

Reported by Pylint.

              """A Pipe implementation in PyTorch."""
from .checkpoint import is_checkpointing, is_recomputing
from .pipe import Pipe
from .microbatch import NoChunk

__all__ = ["Pipe", "is_checkpointing", "is_recomputing"]

            

Reported by Pylint.

                  balanced_pipe = []
    for num_layers in balance:
        layers = []
        for i in range(num_layers):
            layers.append(module[pipe_idx])
            pipe_idx += 1
        device = device_idx if devices is None else devices[device_idx]
        balanced_pipe.append(nn.Sequential(*layers).to(device))
        device_idx += 1

            

Reported by Pylint.

              from torch import nn
from typing import List

def partition_model(
        module: nn.Sequential,
        balance: List[int],
        devices: List[int] = None):
    """
    Given an :class:`nn.Sequential <torch.nn.Sequential>` module, partitions

            

Reported by Pylint.

              from torch import nn
from typing import List

def partition_model(
        module: nn.Sequential,
        balance: List[int],
        devices: List[int] = None):
    """
    Given an :class:`nn.Sequential <torch.nn.Sequential>` module, partitions

            

Reported by Pylint.

                  def wrapper(*args, **kwargs):
        return fn(*args, **kwargs)
    # Can't declare and use attributes of function objects (mypy#2087)
    wrapper._wrapped_async_rpc_function = fn  # type: ignore[attr-defined]
    return wrapper

            

Reported by Pylint.

              import functools


def async_execution(fn):
    r"""
    A decorator for a function indicating that the return value of the function
    is guaranteed to be a :class:`~torch.futures.Future` object and this
    function can run asynchronously on the RPC callee. More specifically, the
    callee extracts the :class:`~torch.futures.Future` returned by the wrapped

            

Reported by Pylint.

              import functools


def async_execution(fn):
    r"""
    A decorator for a function indicating that the return value of the function
    is guaranteed to be a :class:`~torch.futures.Future` object and this
    function can run asynchronously on the RPC callee. More specifically, the
    callee extracts the :class:`~torch.futures.Future` returned by the wrapped

            

Reported by Pylint.

              if "__torch_package__" in dir():

    def is_from_package():
        return True


else:

    def is_from_package():

            

Reported by Pylint.

              if "__torch_package__" in dir():

    def is_from_package():
        return True


else:

    def is_from_package():

            

Reported by Pylint.

              
else:

    def is_from_package():
        return False

            

Reported by Pylint.