使用异步调用 cuda() 会导致 SyntaxError [英] Calling cuda() with async results in SyntaxError

问题描述

我正在尝试运行此 PyTorch 代码:

I'm trying to run this PyTorch code:

for i, (input, target) in enumerate(train_loader):

    input = input.float().cuda(async=True)
    target = target.cuda(async=True)
    input_var = torch.autograd.Variable(input)
    target_var = torch.autograd.Variable(target)

    output = model(input_var)

但是当我尝试时，我收到此错误消息:

But when I try I am getting this error message:

input = input.float().cuda(async=True)
                               ^
SyntaxError: invalid syntax
Process finished with exit code 1

我做错了什么?我已经安装了 cuda.

What am I doing wrong? I already installed cuda.

推荐答案

您的代码不起作用，因为:

• async 是 python 中的保留关键字，不能以这种方式使用，这就是为什么你得到 SyntaxError

• async is a reserved keyword in python which cannot be used in that way, that is why you get the SyntaxError

cuda() 不再 有一个参数 async.构造函数看起来像这样:

cuda() no longer has an argument async. The constructor looks like this:

cuda(device=None, non_blocking=False) → 张量

• 以前有一个参数 async 但这被 non_blocking 代替，因为 async 成为 Python 3.7 中的保留关键字.
  https://github.com/pluskid/fit-random-labels/拉/5
• Previously there was an argument async but this replaced by non_blocking as async became a reserved keyword in Python 3.7.
  https://github.com/pluskid/fitting-random-labels/pull/5

参数 non_blocking 与之前的 async 具有相同的效果:

The argument non_blocking has the same effect as async previously had:

• non_blocking (bool):
  如果 True 并且源在固定内存中，则复制将与主机异步.否则，论证没有效果.默认值:False.
  
  https://pytorch.org/docs/stable/tensors.html#torch.Tensor.cuda

• non_blocking (bool):
  If True and the source is in pinned memory, the copy will be asynchronous with respect to the host. Otherwise, the argument has no effect. Default: False.
  
  https://pytorch.org/docs/stable/tensors.html#torch.Tensor.cuda


<小时>

作为附加组件:如果您对 async 的实际用途感兴趣，可以查看这里:https://www.python.org/dev/peps/pep-0492/#新语法




---

As an add-on: If you are interested in what async is actually used for you can take a look here: https://www.python.org/dev/peps/pep-0492/#new-syntax

这篇关于使用异步调用 cuda() 会导致 SyntaxError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！