fixed issue #20921 (#20922)
Summary:
For tensor creation ops like `torch.zeros` and `torch.ones`, the docs [0], [1] use `sizes` as the first argument to the function call while the correct argument is `size`. This is tested for pytorch 1.1 installed using pip on ubuntu 19.04
An example
```
>>> torch.zeros(2, 3)
tensor([[0., 0., 0.],
[0., 0., 0.]])
>>> torch.zeros(sizes = (2, 3))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: zeros() missing 1 required positional arguments: "size"
>>> torch.zeros(size = (2, 3))
tensor([[0., 0., 0.],
[0., 0., 0.]])
>>> torch.ones(sizes = (2, 3))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: ones() missing 1 required positional arguments: "size"
>>> torch.ones(size = (2, 3))
tensor([[1., 1., 1.],
[1., 1., 1.]])
```
[0]: https://pytorch.org/docs/master/torch.html#torch.zeros
[1]: https://pytorch.org/docs/master/torch.html#torch.ones
Pull Request resolved: https://github.com/pytorch/pytorch/pull/20922
Differential Revision: D15498741
Pulled By: mrshenli
fbshipit-source-id: 963324ffa004d62ca77ce30ed6f0c3932b5b79b7
