8627 perceptual loss errors out after hitting the maximum number of downloads #8652
Conversation
…um-number-of-downloads' of github.com:virginiafdez/MONAI into 8627-perceptual-loss-errors-out-after-hitting-the-maximum-number-of-downloads
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the ✨ Finishing touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
…-the-maximum-number-of-downloads
for more information, see https://pre-commit.ci
| def __init__( | ||
| self, net: str = "medicalnet_resnet10_23datasets", verbose: bool = False, channel_wise: bool = False | ||
| self, | ||
| net: str = "medicalnet_resnet_10_23datasets", |
There was a problem hiding this comment.
What we should do to improve security is to move "medicalnet_resnet10_23datasets" and any other valid name into constants then check net is one of them.
| super().__init__() | ||
| torch.hub._validate_not_a_forked_repo = lambda a, b, c: True | ||
| self.model = torch.hub.load("warvito/MedicalNet-models", model=net, verbose=verbose, trust_repo=True) | ||
| self.model = torch.hub.load( |
There was a problem hiding this comment.
Perhaps this should have trust_repo=True as well since we do trust this repo and it's name is hardcoded here.
| Defaults to ``False``. | ||
| """ | ||
|
|
||
| def __init__( |
There was a problem hiding this comment.
We might need to check that huggingface_hub is installed as well, the torch.hub.load call will fail if so.
2 participants
Fixes #8627.
Moves the perceptual loss code to MONAI repository https://github.com/Project-MONAI/perceptual-models and the checkpoints to Huggingface.
Description
This PR changes and simplifies the torch.hub loading process and gets the models from Huggingface lirbary.
A few sentences describing the changes proposed in this pull request.
Types of changes
./runtests.sh -f -u --net --coverage../runtests.sh --quick --unittests --disttests.make htmlcommand in thedocs/folder.