Source code for torch.distributed
import torch
[docs]def is_available():
"""
Returns ``True`` if the distributed package is available. Otherwise,
``torch.distributed`` does not expose any other APIs. Currently,
``torch.distributed`` is available on Linux, MacOS and Windows. Set
``USE_DISTRIBUTED=1`` to enable it when building PyTorch from source.
Currently, the default value is ``USE_DISTRIBUTED=1`` for Linux and Windows,
``USE_DISTRIBUTED=0`` for MacOS.
"""
return hasattr(torch._C, "_c10d_init")
if is_available() and not torch._C._c10d_init():
raise RuntimeError("Failed to initialize torch.distributed")
if is_available():
from .distributed_c10d import *
# Variables prefixed with underscore are not auto imported
# See the comment in `distributed_c10d.py` above `_backend` on why we expose
# this.
from .distributed_c10d import _backend