torch.nn.parallel.deprecated.distributed.DistributedDataParallel Class Reference

Inheritance diagram for torch.nn.parallel.deprecated.distributed.DistributedDataParallel:

Public Member Functions
def	__init__ (self, module, device_ids=None, output_device=None, dim=0, broadcast_buffers=True)
def	__getstate__ (self)
def	__setstate__ (self, state)
def	forward (self, inputs, kwargs)
def	scatter (self, inputs, kwargs, device_ids)
def	parallel_apply (self, replicas, inputs, kwargs)
def	gather (self, outputs, output_device)
def	train (self, mode=True)

Public Attributes
	dim
	module
	device_ids
	output_device
	broadcast_buffers
	need_reduction
	broadcast_bucket_size
	nccl_reduce_bucket_size
	bucket_sizes
	bucket_map
	buckets
	bucket_events
	reduced
	dispatch_lock
	nccl_reduction_group_id

Detailed Description

Definition at line 27 of file distributed.py.

---

The documentation for this class was generated from the following file:

• torch/nn/parallel/deprecated/distributed.py