A possible bug in torch.nn.functional.conv1d/2d

Pytorch 1.1.0, windows

import torch
import torch.nn as nn
import torch.nn.functional as nnF
x=torch.rand((16, 3, 28, 28))
y=nnF.conv2d(x, a.weight, padding_mode='zeros')

TypeError: conv2d() got an unexpected keyword argument ‘padding_mode’

F.conv2d does not have the optional argument padding_mode.


This document for F.conv2d is wrong.

It looks like the circular padding is indeed applied before calling the functional conv API.
Could you open an issue so that the docs can be fixed?