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

Support training with only normal images (no evaluation) #277

Closed
2 tasks
djdameln opened this issue Apr 25, 2022 · 1 comment · Fixed by #572
Closed
2 tasks

Support training with only normal images (no evaluation) #277

djdameln opened this issue Apr 25, 2022 · 1 comment · Fixed by #572
Assignees
@djdameln
Labels
Enhancement New feature or request Pipeline

Comments

@djdameln
Copy link
Contributor

djdameln commented Apr 25, 2022 •
edited by samet-akcay
Loading

Some users would like to train models with only normal images available, see #192 for example. Our current data classes do not support this and always expect a folder of anomalous images to be provided for evaluation purposes. I feel it would be good to add support for training without evaluation.

  • Modify FolderDataModule to handle null paths.
  • Modify Trainer configs not to run validation
@djdameln djdameln added Enhancement New feature or request Pipeline labels May 13, 2022
@samet-akcay samet-akcay moved this to 📝 To Do in Anomalib Sep 9, 2022
@djdameln djdameln mentioned this issue Sep 14, 2022
Closed
12 tasks
@djdameln djdameln moved this from 📝 To Do to 🚀 PR Created in Anomalib Sep 16, 2022
@samet-akcay samet-akcay mentioned this issue Oct 18, 2022
Merged
This was referenced Nov 7, 2022
Closed
Closed
@ashwinvaidya17 ashwinvaidya17 moved this from 🚀 PR Created to Done in Anomalib Nov 14, 2022
@djdameln
Copy link
Contributor Author

djdameln commented Dec 5, 2022

This has been addressed on the datamodules feature branch.

@djdameln djdameln closed this as completed Dec 5, 2022
@samet-akcay samet-akcay removed this from Anomalib Feb 8, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@djdameln djdameln

Labels
Enhancement New feature or request Pipeline
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

New datamodules design openvinotoolkit/anomalib
Refactor data modules openvinotoolkit/anomalib
1 participant
@djdameln