TypeError: can‘t convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to

报错提示图中的86行出现错误，通过查阅找资料更改如下，在.numpy之前加上.cpu()

            print(('训练周期: {} [{}/{} ({:.0f}%)]\t训练数据Loss: {:.6f},正确率: {:.2f}%\t校验数据Loss:' +
                  '{:.6f},正确率:{:.2f}%').format(epoch, batch_idx * batch_size, len(train_loader.dataset),
                100. * batch_idx / len(train_loader), np.mean(train_loss), 
                                               100. * train_r[0].cpu().numpy() / train_r[1], 
                                               np.mean(val_loss), 
                                               100. * val_r[0].cpu().numpy() / val_r[1]))
            #记录中间的数据
            statistics.append({'loss':np.mean(train_loss),'train': 100. * train_r[0] / train_r[1],
                               'valid':100. * val_r[0] / val_r[1]})

再次点击运行，即可正常运行了。

Initialized!
训练周期: 0 [3904/60000 (7%)]	训练数据Loss: 2.337841,正确率: 26.01%	校验数据Loss:1.323054,正确率:52.42%
训练周期: 0 [10304/60000 (17%)]	训练数据Loss: 1.676903,正确率: 43.34%	校验数据Loss:0.803967,正确率:61.62%
训练周期: 0 [16704/60000 (28%)]	训练数据Loss: 1.427475,正确率: 50.69%	校验数据Loss:0.662364,正确率:80.64%
训练周期: 0 [23104/60000 (38%)]	训练数据Loss: 1.278206,正确率: 55.34%	校验数据Loss:0.627918,正确率:81.10%
训练周期: 0 [29504/60000 (49%)]	训练数据Loss: 1.171616,正确率: 58.69%	校验数据Loss:0.409239,正确率:90.22%
训练周期: 0 [35904/60000 (60%)]	训练数据Loss: 1.091466,正确率: 61.30%	校验数据Loss:0.477059,正确率:80.98%
训练周期: 0 [42304/60000 (70%)]	训练数据Loss: 1.030028,正确率: 63.30%	校验数据Loss:0.351982,正确率:100.00%
训练周期: 0 [48704/60000 (81%)]	训练数据Loss: 0.977565,正确率: 65.11%	校验数据Loss:0.424329,正确率:79.98%
训练周期: 0 [55104/60000 (92%)]	训练数据Loss: 0.933881,正确率: 66.55%	校验数据Loss:0.267361,正确率:100.00%
训练周期: 1 [1472/60000 (2%)]	训练数据Loss: 0.606931,正确率: 76.50%	校验数据Loss:0.280753,正确率:100.00%
训练周期: 1 [7872/60000 (13%)]	训练数据Loss: 0.564598,正确率: 78.79%	校验数据Loss:0.212235,正确率:100.00%
训练周期: 1 [14272/60000 (24%)]	训练数据Loss: 0.563004,正确率: 78.83%	校验数据Loss:0.236143,正确率:100.00%
训练周期: 1 [20672/60000 (34%)]	训练数据Loss: 0.547791,正确率: 79.65%	校验数据Loss:0.238229,正确率:100.00%
训练周期: 1 [27072/60000 (45%)]	训练数据Loss: 0.535284,正确率: 80.38%	校验数据Loss:0.169436,正确率:100.00%
训练周期: 1 [33472/60000 (56%)]	训练数据Loss: 0.519554,正确率: 81.09%	校验数据Loss:0.154972,正确率:100.00%
训练周期: 1 [39872/60000 (66%)]	训练数据Loss: 0.513823,正确率: 81.36%	校验数据Loss:0.167662,正确率:100.00%
训练周期: 1 [46272/60000 (77%)]	训练数据Loss: 0.504795,正确率: 81.76%	校验数据Loss:0.168867,正确率:100.00%
训练周期: 1 [52672/60000 (88%)]	训练数据Loss: 0.488250,正确率: 82.51%	校验数据Loss:0.125301,正确率:100.00%
训练周期: 1 [59072/60000 (98%)]	训练数据Loss: 0.478585,正确率: 82.89%	校验数据Loss:0.116586,正确率:100.00%
训练周期: 2 [5440/60000 (9%)]	训练数据Loss: 0.467881,正确率: 83.47%	校验数据Loss:0.120325,正确率:100.00%
训练周期: 2 [11840/60000 (20%)]	训练数据Loss: 0.417171,正确率: 85.33%	校验数据Loss:0.146612,正确率:100.00%
训练周期: 2 [18240/60000 (30%)]	训练数据Loss: 0.437972,正确率: 84.47%	校验数据Loss:0.141501,正确率:100.00%
训练周期: 2 [24640/60000 (41%)]	训练数据Loss: 0.426656,正确率: 85.06%	校验数据Loss:0.130396,正确率:100.00%
训练周期: 2 [31040/60000 (52%)]	训练数据Loss: 0.409198,正确率: 85.82%	校验数据Loss:0.126598,正确率:100.00%
训练周期: 2 [37440/60000 (62%)]	训练数据Loss: 0.414902,正确率: 85.55%	校验数据Loss:0.124264,正确率:100.00%
训练周期: 2 [43840/60000 (73%)]	训练数据Loss: 0.401590,正确率: 86.15%	校验数据Loss:0.081980,正确率:100.00%
训练周期: 2 [50240/60000 (84%)]	训练数据Loss: 0.393084,正确率: 86.52%	校验数据Loss:0.114253,正确率:100.00%
训练周期: 2 [56640/60000 (94%)]	训练数据Loss: 0.396160,正确率: 86.31%	校验数据Loss:0.121534,正确率:100.00%