tensor保存图像

def denorm(x):
    """Convert the range from [-1, 1] to [0, 1]."""
    out = (x + 1) / 2
    return out.clamp_(0, 1)


def tensor2img(x_data, fname, nrow, paddoing=0):
    save_image(denorm(x_data.cpu()), fname, nrow=nrow, padding=0)

---

方法2：

n_show_images = 4

def imshow(img: torch.Tensor):
    """
    Display a single image.
    """

    img = img / 2 + 0.5     # unnormalize
    npimg = img.numpy()
    plt.imshow(np.transpose(npimg, (1, 2, 0)))
    plt.show()

dataiter = iter(labeled_trainloader)
images, labels = dataiter.next()
imshow(torchvision.utils.make_grid(images[:n_show_images]))
print(' '.join(f'{sup_classes[labels[j]]:5s}' for j in range(n_show_images)))