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

Corrected SAHP. #38

Merged
merged 1 commit into from
Sep 10, 2024
Merged

Corrected SAHP. #38

merged 1 commit into from
Sep 10, 2024

Conversation

ajboyd2
Copy link
Contributor

@ajboyd2 ajboyd2 commented Sep 3, 2024

Problems that were fixed:

  1. Mu, eta, and gamma were missing their activation functions.
  2. Clipping the time_delta is wrong and should be avoided. This was only needed in the first place because the gamma component was allowed to be negative due to a missing activation function.
  3. Mu, eta, and gamma were not parameters and thus not learned. This is no longer the case since they are now represented as a module with linear layers containing the (now learnable) weights.

Problems that were fixed:
1. Mu, eta, and gamma were missing their activation functions.
2. Clipping the time_delta is wrong and should be avoided. This was only needed in the first place because the gamma component was allowed to be negative due to a missing activation function.
3. Mu, eta, and gamma were not parameters and thus not learned.
@iLampard
Copy link
Collaborator

iLampard commented Sep 3, 2024

Thanks for the pr

@iLampard iLampard closed this Sep 3, 2024
@iLampard iLampard reopened this Sep 10, 2024
@iLampard iLampard merged commit a2e254d into ant-research:main Sep 10, 2024
1 check failed
@iLampard
Copy link
Collaborator

sorry, i forgot to merge it. now done. thanks for the pr.

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.

2 participants