PyTorch绘制训练过程的accuracy和loss曲线

Loss_list = []
Accuracy_list = []

Loss_list.append(train_loss / (len(train_dataset)))
Accuracy_list.append(100 * train_acc / (len(train_dataset)))

#我这里迭代了100次，所以x的取值范围为(0，100)，然后再将每次相对应的准确率以及损失率附在x上
x1 = range(0, 100)
x2 = range(0, 100)
y1 = Accuracy_list
y2 = Loss_list
plt.subplot(2, 1, 1)
plt.plot(x1, y1, 'o-')
plt.title('Test accuracy vs. epoches')
plt.ylabel('Test accuracy')
plt.subplot(2, 1, 2)
plt.plot(x2, y2, '.-')
plt.xlabel('Test loss vs. epoches')
plt.ylabel('Test loss')
plt.show()
plt.savefig("accuracy_loss.jpg")