I got an error when loading deeplabv3_resnet101.

[tuxxon]

🐛 Bug

I got the error when running model = torch.hub.load('pytorch/vision:v0.10.0', 'deeplabv3_resnet101', pretrained=True)

To Reproduce

Steps to reproduce the behavior:

1. Click the url below:
   https://colab.research.google.com/github/pytorch/pytorch.github.io/blob/master/assets/hub/pytorch_vision_deeplabv3_resnet101.ipynb
2. Click the first cell

import torch
model = torch.hub.load('pytorch/vision:v0.10.0', 'deeplabv3_resnet101', pretrained=True)
model.eval()

Expected behavior

Normal works.

but, I got this error:

---------------------------------------------------------------------------
HTTPError                                 Traceback (most recent call last)
<ipython-input-2-90f9efc58017> in <module>()
      1 import torch
----> 2 model = torch.hub.load('pytorch/vision:v0.10.0', 'deeplabv3_resnet101', pretrained=True)
      3 model.eval()

8 frames
/usr/lib/python3.7/urllib/request.py in http_error_default(self, req, fp, code, msg, hdrs)
    647 class HTTPDefaultErrorHandler(BaseHandler):
    648     def http_error_default(self, req, fp, code, msg, hdrs):
--> 649         raise HTTPError(req.full_url, code, msg, hdrs, fp)
    650 
    651 class HTTPRedirectHandler(BaseHandler):

HTTPError: HTTP Error 403: rate limit exceeded

Environment

Jupyter notebook with Python3

[NicolasHug]

Thanks for the report, this is a known issue that should be fixed in the next version. I'll close as a duplicate of #4156 and pytorch/pytorch#61755

In your specific case, could you try to load the model directly from torchvision instead:

model = torchvision.models.deeplabv3_resnet101(pretrained=True, progress=True)

This will only be equivalent if you have torchvision 0.10 though.

Otherwise you can also set

torch.hub._validate_not_a_forked_repo=lambda a,b,c: True

as described in pytorch/pytorch#61755 (comment)