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

Merge the latest loss functions from upstream #5

Open

sadamov opened this issue Jan 26, 2024 · 0 comments

Labels

Collaborator

sadamov commented Jan 26, 2024

Joel mentioned that he added new loss functions and that the negative log likelihood performs better than mse/mae.

Merge the latest commit from uptream into a feature_branch
This will give you a new metrics.py file that contains many additional loss functions
Start a training with loss = negative log likelihood for 40 epochs (validation every 20 epochs)
Compare training loss with previous training loss curves on wandb

sadamov added the enhancement label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment