pytorch/hub

404 for pre-trained ResNeXt models

eric-mitchell opened this issue · 1 comments

eric-mitchell commented

I'm currently getting a 404 when following the instructions here for loading the Instagram pre-trained ResNeXt models. It looks like the source repo changed their master branch to be main, so instead of

model = torch.hub.load('facebookresearch/WSL-Images', 'resnext101_32x8d_wsl')

one simply needs

model = torch.hub.load('facebookresearch/WSL-Images:main', 'resnext101_32x8d_wsl')

and all is well.

NicolasHug commented

Thanks for the report @eric-mitchell

I believe you're relying on an old torch version (or more accurately, not the latest one), because this is something that we fixed on the latest torch version. Running model = torch.hub.load('facebookresearch/WSL-Images', 'resnext101_32x8d_wsl') should work properly in torch >= 1.10, as the default branch assumption was updated to "if main exists use main, else use master".

  • 👍1