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.

Sign up for GitHub

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

Jump to bottom

Retina Net #6

Closed
8 tasks done
oke-aditya opened this issue Nov 6, 2020 · 4 comments · Fixed by #46
Closed
8 tasks done

Retina Net #6

oke-aditya opened this issue Nov 6, 2020 · 4 comments · Fixed by #46
Assignees
@oke-aditya @ramaneswaran
Labels
enhancement New feature or request good first issue Good for newcomers High Priority Should be addressed ASAP. Model to Implement This model should be implmented

Comments

@oke-aditya
Copy link
Owner

oke-aditya commented Nov 6, 2020 •
edited
Loading

🚀 Feature

Our second detection model

The following needs to be done.

  • Support all torchvision backbones.
  • FPN Support for Resnet Backbones.
  • Support for negative samples etc.
  • Mixed Precision Training.
  • train_step, val_step, fit APIs
  • Sanity Fit API
  • IoU, MaP calculations while training and testing.
  • Lightning Trainer.
@oke-aditya oke-aditya added enhancement New feature or request good first issue Good for newcomers High Priority Should be addressed ASAP. labels Nov 6, 2020
@oke-aditya oke-aditya added the Model to Implement This model should be implmented label Nov 7, 2020
@oke-aditya
Copy link
Owner Author

@ramaneswaran up for it ?
I have never trained RetinaNet yet, so it's new for both of us. But should be very similar to FRCNN.

@oke-aditya
Copy link
Owner Author

@ramaneswaran I will create the backbones and model_factory I will leave the engine to you :)

@oke-aditya oke-aditya mentioned this issue Nov 11, 2020
Merged
This was referenced Nov 12, 2020
Merged
Merged
@ramaneswaran
Copy link
Contributor

Looks like most of the work here is done
I will give sanity fit functions a shot

@oke-aditya
Copy link
Owner Author

Nice. We need slight refactor with detection models, but that will not affect sanity fit API.

@oke-aditya oke-aditya mentioned this issue Nov 23, 2020
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@oke-aditya oke-aditya

@ramaneswaran ramaneswaran

Labels
enhancement New feature or request good first issue Good for newcomers High Priority Should be addressed ASAP. Model to Implement This model should be implmented
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Complete sanity functions for faster_rcnn and RetinaNet ramaneswaran/quickvision
2 participants
@oke-aditya @ramaneswaran