Skip to content

Avoid weight download on test_detection_model_trainable_backbone_layers test #4659

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from

Conversation

datumbox
Copy link
Contributor

@datumbox datumbox commented Oct 19, 2021

The tests shouldn't download the backbone weights.

Copy link
Contributor

@prabhat00155 prabhat00155 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @datumbox!

@@ -783,7 +783,7 @@ def test_detection_model_trainable_backbone_layers(model_name):
n_trainable_params = []
for trainable_layers in range(0, max_trainable + 1):
model = torchvision.models.detection.__dict__[model_name](
pretrained=False, pretrained_backbone=True, trainable_backbone_layers=trainable_layers
pretrained=False, pretrained_backbone=False, trainable_backbone_layers=trainable_layers
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Related test failures? I see a bunch of detection model tests failing.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the flag. Yes they are related.

The value of the boolean unfortunately affects the number of unfrozen weights... I think we need to rethink how we test this cause we shouldn't be downloading any weights without having setup a caching mechanism.

I think I'm going to close the PR and review how to solve this on a follow up PR.

cc @jdsgomes

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

New issue created at #4660

@prabhat00155 prabhat00155 self-requested a review October 19, 2021 18:53
@datumbox datumbox closed this Oct 19, 2021
@datumbox datumbox deleted the datumbox-patch-1 branch October 19, 2021 18:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants