RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the

1.问题

今天练习使用pytorch的时候，准备上GPU，结果出现了以下错误：

RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same

我百度了几个解决办法，但问题刚好和我的相反，他们的问题是：

RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor)

我的是输入类型（Input）不是GPU类型的，百度的是权重（weight）不是GPU类型的，所以解决办法不适用。

2.代码（已调整好，可以正确运行）

import torch.optim
import torchvision.datasets

# 准备数据集
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
from time import time

print(torch.cuda.is_available())

train_data = torchvision.datasets.CIFAR10(root="./dataset",
                                          train=True,
                                          transform=torchvision.transforms.ToTensor(),
                                          download=True)

test_data = torchvision.datasets.CIFAR10(root="./dataset",
                                         train=False,
                                         transform=torchvision.transforms.ToTensor(),
                                         download=True)
print("训练集数据长度：%d" % len(train_data))
print("测试集数据长度：%d" % len(test_data))

# 利用DataLoader加载数据
train_data_loader = DataLoader(train_data, batch_size=64)
test_data_loader = DataLoader(test_data, batch_size=64)


# 搭建神经网络(单独放在一个.py文件)
class Net(nn.Module):
    def __init__(self) -> None:
        super(Net, self).__init__()
        self.model = nn.Sequential(
            nn.Conv2d(3, 32, 5, 1, 2),
            nn.MaxPool2d(2, 2),
            nn.Conv2d(32, 32, 5, 1, 2),
            nn.MaxPool2d(2, 2),
            nn.Conv2d(32, 64, 5, 1, 2),
            nn.MaxPool2d(2, 2),
            nn.Flatten(),
            nn.Linear(1024, 64),
            nn.Linear(64, 10)

        )

    def forward(self, x):
        x = self.model(x)
        return x


# 创建网络模型
net = Net()
# 只有模型、数据、损失函数可以运行在GPU上
# if torch.cuda.is_available():
#     net = net.cuda()
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
net.to(device)


# 损失函数
loss_fn = nn.CrossEntropyLoss()
# if torch.cuda.is_available():
#     loss_fn = loss_fn.cuda()
loss_fn.to(device)

# 优化器
learning_rate = 1e-2  # 0.01
optimizer = torch.optim.SGD(net.parameters(), lr=learning_rate)

# 设置训练网络的参数
# 记录训练次数
total_train_step = 0
# 记录训练次数
total_test_step = 0
# 训练轮数
epoch = 10

start_time = time()
writer = SummaryWriter("./logs/train")
for i in range(epoch):
    print("------第%d轮训练------" % (i + 1))

    # 训练步骤
    for data in train_data_loader:
        imgs, targets = data
        if torch.cuda.is_available():
            imgs, targets = imgs.cuda(), targets.cuda()
        # imgs.to(device)
        # targets.to(device)
        outputs = net(imgs)
        loss = loss_fn(outputs, targets)

        # 优化器优化模型
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_train_step += 1
        if total_train_step % 100 == 0:
            end_time = time()
            print(end_time - start_time)
            print("训练次数：{}，loss：{}".format(total_train_step, loss.item()))  # .item()能把tensor类型转化为数字
            writer.add_scalar("train_loss", loss.item(), total_train_step)

    # 测试步骤
    total_test_loss = 0
    total_accuracy = 0
    with torch.no_grad():  # 测试时，将梯度归零。不需要调整梯度进行优化
        for data in test_data_loader:
            imgs, targets = data
            if torch.cuda.is_available():
                imgs, targets = imgs.cuda(), targets.cuda()
            # imgs.to(device)
            # targets.to(device)
            outputs = net(imgs)
            loss = loss_fn(outputs, targets)

            total_test_loss += loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_accuracy += accuracy
    print("整体测试集上的loss: {}".format(total_test_loss))
    print("整体测试集上的正确率: {}".format(total_accuracy / len(test_data)))

    writer.add_scalar("test_loss", total_test_loss, total_test_step)
    writer.add_scalar("test_accuracy", total_accuracy / len(test_data), total_test_step)
    total_test_step += 1

    # 保存模型
    # torch.save(net.state_dict(), "model_{}.pth".format(i))
    # print("第{}轮训练模型已保存".format(i))

writer.close()

3.解决办法

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
for data in train_data_loader:
   imgs, targets = data
    if torch.cuda.is_available():
        imgs, targets = imgs.cuda(), targets.cuda()
    # imgs.to(device)
    # targets.to(device)
    outputs = net(imgs)

在上述代码中，imgs和targets不能用.to(device)的形式，这样使用后就会出现Input类型（torch.FloatTensor）不是GPU类型的，只能用另外一种方式：

if torch.cuda.is_available():
    imgs, targets = imgs.cuda(), targets.cuda()

这样可以解决输入和权重类型不匹配的问题。

4.参考

https://stackoverflow.com/questions/59013109/runtimeerror-input-type-torch-floattensor-and-weight-type-torch-cuda-floatte