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

Test Brightness Augmentation #31

Closed
mattmotoki opened this issue Oct 29, 2018 · 1 comment
Closed

Test Brightness Augmentation #31

mattmotoki opened this issue Oct 29, 2018 · 1 comment
Assignees

Comments

@mattmotoki
Copy link
Contributor

mattmotoki commented Oct 29, 2018

Determine the effectiveness of brightness augmentation.

  • Take photos of a single plant at multiple times of the day (different brightness levels)
  • Train a model with/without brightness augmentation
  • Plot performance vs. brightness level for both models.
  • Calculate average percent improvement
@mattmotoki
Copy link
Contributor Author

Fine-tuning results show that mobilenetv1 performs the best.

           model   top_1_accuracy   top_3_accuracy   top_5_accuracy
mobilenetv1-1.00           0.6450           0.8559           0.9219 
mobilenetv2-1.00           0.5521           0.7786           0.8698 
mobilenetv2-1.30           0.5842           0.8290           0.8993 
mobilenetv2-1.40           0.6068           0.8490           0.9149 
    nasnetmobile           0.5590           0.8056           0.8854 

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

No branches or pull requests

3 participants