TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first. Error occured using Pytorch

Question

我在 Whosebug 上阅读过有关相同错误的问题，但不幸的是它们不起作用。我有一个定义的函数：

def plot_loss(train_loss, validation_loss, title):
    plt.grid(True)
    plt.xlabel("subsequent epochs")
    plt.ylabel('average loss')
    plt.plot(range(1, len(train_loss)+1), train_loss, 'o-', label='training')
    plt.plot(range(1, len(validation_loss)+1), validation_loss, 'o-', label='validation')
    plt.legend()
    plt.title(title)
    plt.show()

问题是在行

    plt.plot(range(1, len(validation_loss)+1), validation_loss, 'o-', label='validation')

发生此错误。然后流经 pyplot.py 中的 gca().plot(...)，最终流向 tensor.py

中的 return self.numpy()

在测试阶段我定义了以下函数：

with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device) 
            output = model(data)
            # calculate and sum up batch loss
            test_loss += F.nll_loss(output, target, reduction='mean') 
            # get the index of class with the max log-probability 
            prediction = output.argmax(dim=1)
            # item() returns value of the given tensor
            correct += prediction.eq(target).sum().item()
    test_loss /= len(test_loader)
    return test_loss

我试过换行

  prediction = output.argmax(dim=1)

正如在另一个关于相同错误的问题中所描述的那样，但不幸的是它没有帮助。

我已经尝试运行在 Google Colab 以及我的带有 GPU 的本地机器上运行这段代码（cuda 可用），但不幸的是发生了同样的错误。

编辑我设法在这个 link 中找到了解决方案。它似乎与 CUDA 和 CPU 之间的移动数据有关。我调用了 .cpu() 并解决了：

    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device) 
            output = model(data)
            # calculate and sum up batch loss
            test_loss += F.nll_loss(output, target, reduction='mean') 
            # get the index of class with the max log-probability 
            prediction = output.argmax(dim=1)
            # item() returns value of the given tensor
            correct += prediction.eq(target).sum().item()
    test_loss /= len(test_loader)
    return test_loss.cpu()

Answer 1

我设法在这个 link 中找到了解决方案。它似乎与 CUDA 和 CPU 之间的移动数据有关。我调用了 .cpu() 并解决了：

 with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device) 
            output = model(data)
            # calculate and sum up batch loss
            test_loss += F.nll_loss(output, target, reduction='mean') 
            # get the index of class with the max log-probability 
            prediction = output.argmax(dim=1)
            # item() returns value of the given tensor
            correct += prediction.eq(target).sum().item()
    test_loss /= len(test_loader)
    return test_loss.cpu()

TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first. Error occured using Pytorch

TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first. Error occured using Pytorch

python

matplotlib

deep-learning

neural-network

pytorch