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

Get HPO working again #141

Merged
merged 2 commits into from
Jan 10, 2025
Merged

Get HPO working again #141

merged 2 commits into from
Jan 10, 2025

Conversation

dinogun
Copy link
Contributor

@dinogun dinogun commented Jan 9, 2025

  1. Use EmailMessage as cgi is deprecated
  2. Dockerfile should use python3 instead of python39
  3. Add missing source script
  4. Updated OS image
  5. Use quay.io repo instead of DockerHub

1.
Updated OS image
2. Use quay.io repo instead of DockerHub

Signed-off-by: Dinakar Guniguntala <[email protected]>
Signed-off-by: Dinakar Guniguntala <[email protected]>
@bharathappali
Copy link
Member

@dinogun I see scripts scripts/cluster-helpers.sh & scripts/openshift-helpers.sh are shown as empty files in the commits. Is it because of file permission changes?

@bharathappali
Copy link
Member

Also will it be a good idea to use a lightweight framework like fastapi instead of using a basic http server for rest_service.py?

Fast API :

@kusumachalasani
Copy link
Contributor

Works as expected across all environments. Tested on native,docker,openshift.

@kusumachalasani
Copy link
Contributor

There are some future warnings with HPO related to tunables. It would be good to address them as well later to prevent potential issues down the line.

hpo/hpo/src/bayes_optuna/optuna_hpo.py:389: FutureWarning: suggest_discrete_uniform has been deprecated in v3.0.0. This feature will be removed in v6.0.0. See https://github.com/optuna/optuna/releases/tag/v3.0.0. Use suggest_float(..., step=...) instead.
  tunable_value = trial.suggest_discrete_uniform(
hpo/hpo/src/bayes_optuna/optuna_hpo.py:393: FutureWarning: suggest_int() got {'step'} as positional arguments but they were expected to be given as keyword arguments.
  tunable_value = trial.suggest_int(

Copy link
Contributor

@kusumachalasani kusumachalasani left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!

@dinogun
Copy link
Contributor Author

dinogun commented Jan 10, 2025

@bharathappali Reg fastapi, we can do those as a separate PR if needed. Right now just want to fix what is broken

@dinogun dinogun merged commit 1d601e9 into kruize:main Jan 10, 2025
4 checks passed
@bharathappali
Copy link
Member

Reg fastapi, we can do those as a separate PR if needed

I completely agree, Will raise an issue to keep track of it. Thanks @dinogun

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.

3 participants