为什么在推理中使用 Variable()? [英] Why use Variable() in inference?

问题描述

我正在学习 PyTorch 用于图像分类任务，我遇到了有人在其函数中使用 PyTorch Variable() 进行预测的代码:

I am learning PyTorch for an image classification task, and I ran into code where someone used a PyTorch Variable() in their function for prediction:

def predict_image(image):
    image_tensor = test_transforms(image).float()
    image_tensor = image_tensor.unsqueeze_(0)
    input = Variable(image_tensor)
    input = input.to(device)
    output = model(input)
    index = output.data.cpu().numpy().argmax()
    return index

为什么他们在这里使用 Variable()?(即使没有它也能正常工作.)

Why do they use Variable() here? (even though it works fine without it.)

推荐答案

您可以放心地省略它.变量是 PyTorch 的遗留组件，现已弃用，过去是 autograd 所必需的:

You can safely omit it. Variables are a legacy component of PyTorch, now deprecated, that used to be required for autograd:

警告

Variable API 已被弃用:不再需要变量来将 autograd 与张量一起使用.Autograd 自动支持 requires_grad 设置为 True 的张量.请在下方找到有关更改内容的快速指南:

The Variable API has been deprecated: Variables are no longer necessary to use autograd with tensors. Autograd automatically supports Tensors with requires_grad set to True. Below please find a quick guide on what has changed:

• Variable(tensor) 和 Variable(tensor, requires_grad) 仍然按预期工作，但它们返回的是张量而不是变量.

• Variable(tensor) and Variable(tensor, requires_grad) still work as expected, but they return Tensors instead of Variables.

这篇关于为什么在推理中使用 Variable()?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！