[人工智能]Error:all tensors must be on devices[0] 初学者解决方案

一、问题简介

• 作为Pytorch初学者，我跟随教程实现图像分类数据集 FashionMinist 的 softmax 多分类项目。过程中最后一步调用d2lzh_pytorch包中的train_ch3函数，产生如下报错提示。
• RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking arugment for argument mat1 in method wrapper_addmm)
  
• 错误原因在于，我们初学者的计算机大部分都是单gpu的，而目前封装好的函数大部分是默认支持多gpu同步训练。因此我们首先需要确定本次实验所要使用的设备，然后将所有的网络模型以及tensor全部加入到该设备上去。
• 参考学习链接：https://tangshusen.me/Dive-into-DL-PyTorch/#/

二、解决方法

# 1.规定唯一设备（如果cuda可用则使用cuda:0，否则使用cpu）
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# 2.将网络结构net加入到device中去
net = net.to(device)
# 3.多个函数中每次从迭代器中读出tensor都需要加入到device中去
for X, y in data_iter:
    X = X.to(device)
    y = y.to(device)

三、完整代码

import torch
from torch import nn
from torch.nn import init
import numpy as np
import sys
sys.path.append("..") 
import d2lzh_pytorch as d2l

# 1.设定批次大小，生成读取数据迭代器
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)

# 2.构建网络结构
num_inputs = 784
num_outputs = 10

# 2.1扁平层，将二维图像压为一维
class FlattenLayer(nn.Module):
    def __init__(self):
        super(FlattenLayer, self).__init__()
    def forward(self, x): # x shape: (batch, *, *, ...)
        return x.view(x.shape[0], -1)
        
from collections import OrderedDict

net = nn.Sequential(
    OrderedDict([
        ('flatten', FlattenLayer()),
        ('linear', nn.Linear(num_inputs, num_outputs))
    ])
)

# 3.初始化待训练参数
init.normal_(net.linear.weight, mean=0, std=0.01)
init.constant_(net.linear.bias, val=0) 

# 4.定义损失函数以及优化方法
loss = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(net.parameters(), lr=0.1)

# 5.规定设备，并将net放入设备上
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
net = net.to(device)

# 6.定义准确度函数、根据测试集计算准确度函数
def accuracy(y_hat, y):
    return (y_hat.argmax(dim=1) == y).float().mean().item()

def evaluate_accuracy(data_iter, net):
    acc_sum, n = 0.0, 0
    for X, y in data_iter:
        X = X.to(device)
        y = y.to(device)
        acc_sum += (net(X).argmax(dim=1) == y).float().sum().item()
        n += y.shape[0]
    return acc_sum / n

# 7.重写train_ch3函数，在读取tensor时将tensor加入到device上
def train_ch3_my(net, train_iter, test_iter, loss, num_epochs, batch_size,
              params=None, lr=None, optimizer=None):
    for epoch in range(num_epochs):
        train_l_sum, train_acc_sum, n = 0.0, 0.0, 0
        for X, y in train_iter:
            X = X.to(device)
            y = y.to(device)
            y_hat = net(X)
            l = loss(y_hat, y).sum()

            if optimizer is not None:
                optimizer.zero_grad()
            elif params is not None and params[0].grad is not None:
                for param in params:
                    param.grad.data.zero_()

            l.backward()
            if optimizer is None:
                d2l.sgd(params, lr, batch_size)
            else:
                optimizer.step()


            train_l_sum += l.item()
            train_acc_sum += (y_hat.argmax(dim=1) == y).sum().item()
            n += y.shape[0]
        test_acc = evaluate_accuracy(test_iter, net)
        print('epoch %d, loss %.4f, train acc %.3f, test acc %.3f'
              % (epoch + 1, train_l_sum / n, train_acc_sum / n, test_acc))

# 8.进行模型训练
num_epochs = 5
train_ch3_my(net, train_iter, test_iter, loss, num_epochs, batch_size, None, None, optimizer)