pytorch多gpu数据并行

import torch
import torch.nn as nn
from torch.utils.data import DataLoader, Dataset

input_size = 5#模型输入和输出维度
output_size = 2

batch_size = 40 #批量
data_size = 1000 #样本数量

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

class RandomDataset(Dataset):#随机数据集

    def __init__(self, size, length):
        self.len = length
        self.data = torch.randn(length, size)

    def __getitem__(self, index):
        return self.data[index]

    def __len__(self):
        return self.len
#随机数据加载器
rand_loader = DataLoader(dataset=RandomDataset(input_size, data_size),batch_size=batch_size, shuffle=True)

class Model(nn.Module):#模型
    # Our model

    def __init__(self, input_size, output_size):
        super(Model, self).__init__()
        self.fc = nn.Linear(input_size, output_size)

    def forward(self, input):
        output = self.fc(input)
        print("\tIn Model: input size", input.size(),
              "output size", output.size())

        return output

model = Model(input_size, output_size)#实例化模型
if torch.cuda.device_count() > 1:
  print("Let's use", torch.cuda.device_count(), "GPUs!")
  # dim = 0 [30, xxx] -> [10, ...], [10, ...], [10, ...] on 3 GPUs
  model = nn.DataParallel(model) #数据并行

model.to(device)#模型加载到gpu

for data in rand_loader:
    input = data.to(device)
    output = model(input)
    print("Outside: input size", input.size(),
          "output_size", output.size())

有4个gpu情况下:

数据被划分成4份，分别在4个gpu上计算