pytorch/torch/nn/parallel/data_parallel.py
2017-02-26 20:37:43 +01:00

109 lines
3.8 KiB
Python

import torch
from torch.autograd import Variable
from ..modules import Module
from .scatter_gather import scatter, gather
from .replicate import replicate
from .parallel_apply import parallel_apply
class DataParallel(Module):
"""Implements data parallelism at the module level.
This container parallelizes the application of the given module by
splitting the input across the specified devices by chunking in the batch
dimension. In the forward pass, the module is replicated on each device,
and each replica handles a portion of the input. During the backwards
pass, gradients from each replica are summed into the original module.
The batch size should be larger than the number of GPUs used. It should
also be an integer multiple of the number of GPUs so that each chunk is the
same size (so that each GPU processes the same number of samples).
See also: :ref:`cuda-nn-dataparallel-instead`
Args:
module: module to be parallelized
device_ids: CUDA devices (default: all devices)
output_device: device location of output (default: device_ids[0])
Example::
>>> device_ids=[0, 1, 2]
>>> net = torch.nn.DataParallel(model, device_ids=device_ids)
>>> input_var.size(0) % len(device_ids)
0
>>> output = net(input_var)
"""
# TODO: update notes/cuda.rst when this class handles 8+ GPUs well
def __init__(self, module, device_ids=None, output_device=None):
super(DataParallel, self).__init__()
if device_ids is None:
device_ids = list(range(torch.cuda.device_count()))
if output_device is None:
output_device = device_ids[0]
self.module = module
self.device_ids = device_ids
self.output_device = output_device
if len(self.device_ids) == 1:
self.module.cuda(device_ids[0])
def forward(self, *inputs):
def _to_cuda(obj):
if isinstance(obj, Variable):
return obj.cuda()
return tuple((map(_to_cuda, obj)))
if len(self.device_ids) == 1:
with torch.cuda.device(self.device_ids[0]):
inputs_cuda = _to_cuda(inputs)
return self.module(*inputs_cuda)
replicas = self.replicate(self.module, self.device_ids)
scattered = self.scatter(inputs, self.device_ids)
replicas = replicas[:len(scattered)]
outputs = self.parallel_apply(replicas, scattered)
return self.gather(outputs, self.output_device)
def replicate(self, module, device_ids):
return replicate(module, device_ids)
def scatter(self, input, device_ids):
return scatter(input, device_ids)
def parallel_apply(self, replicas, inputs):
return parallel_apply(replicas, inputs)
def gather(self, outputs, output_device):
return gather(outputs, output_device)
def data_parallel(module, inputs, device_ids, output_device=None):
"""Evaluates module(input) in parallel across the GPUs given in device_ids.
This is the functional version of the DataParallel module.
Args:
module: the module to evaluate in parallel
inputs: inputs to the module
device_ids: GPU ids on which to replicate module
output_device: GPU location of the output Use -1 to indicate the CPU.
(default: device_ids[0])
Returns:
a Variable containing the result of module(input) located on
output_device
"""
if not isinstance(inputs, tuple):
inputs = (inputs,)
if not device_ids:
return module(*inputs)
if output_device is None:
output_device = device_ids[0]
replicas = replicate(module, device_ids)
scattered = scatter(inputs, device_ids)
replicas = replicas[:len(scattered)]
outputs = parallel_apply(replicas, scattered)
return gather(outputs, output_device)