Skip to content
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

Allow checkpoint_path to be an url (Required for transfer learning) #1138

Merged
merged 1 commit into from
Jun 6, 2023

Conversation

BloodAxe
Copy link
Collaborator

@BloodAxe BloodAxe commented Jun 6, 2023

  • Allow checkpoint_path to be an url (Required for transfer learning)
  • Added synchronization primitives to prevent race conditions when downloading weights

One can now load checkpoints like so:

models.get(Models.YOLO_NAS_S, num_classes=80, checkpoint_path="https://sghub.deci.ai/models/yolo_nas_s_coco.pth")

Why we want this to have: For training YOLO-NAS-POSE we want to use YOLO-NAS-OD pretrained weights like so:

checkpoint_params:
  checkpoint_path: "https://sghub.deci.ai/models/yolo_nas_s_coco.pth"

This PR enables us to do this.

Added synchronization primitives to prevent race conditions when downloading weights
Copy link
Contributor

@Louis-Dupont Louis-Dupont left a comment

Choose a reason for hiding this comment

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

LGTM, just small question below

@BloodAxe BloodAxe merged commit bc82b29 into master Jun 6, 2023
1 check passed
@BloodAxe BloodAxe deleted the feature/SG-000-allow-url-in-checkpoint-path branch June 6, 2023 09:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants