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

[air output] Improve layout of config table at start of training #34784

Closed
Tracked by #34783
krfricke opened this issue Apr 26, 2023 · 0 comments · Fixed by #34788
Closed
Tracked by #34783

[air output] Improve layout of config table at start of training #34784

krfricke opened this issue Apr 26, 2023 · 0 comments · Fixed by #34788
Assignees

Comments

@krfricke
Copy link
Contributor

No description provided.

@krfricke krfricke changed the title Improve layout of config table at start of training [air output] Improve layout of config table at start of training Apr 26, 2023
@krfricke krfricke added this to the Tune Console Output milestone Apr 26, 2023
@krfricke krfricke self-assigned this Apr 26, 2023
krfricke added a commit that referenced this issue May 2, 2023
This PR makes a number of improvements to tackle issues uncovered in dogfooding:

1. Instead of just a print, we render a table for configs and results at the start of training (closes #34784)
2. We round float results to 5 significant numbers after the decimal point (closes #34785)
3. We track the last printed result and only print the result at the end of training if it hasn't been printed before (closes #34786)
4. We divide the results by "automatic" results and trainer-specific results (closes #34787)

Signed-off-by: Kai Fricke <[email protected]>
architkulkarni pushed a commit to architkulkarni/ray that referenced this issue May 16, 2023
…t#34788)

This PR makes a number of improvements to tackle issues uncovered in dogfooding:

1. Instead of just a print, we render a table for configs and results at the start of training (closes ray-project#34784)
2. We round float results to 5 significant numbers after the decimal point (closes ray-project#34785)
3. We track the last printed result and only print the result at the end of training if it hasn't been printed before (closes ray-project#34786)
4. We divide the results by "automatic" results and trainer-specific results (closes ray-project#34787)

Signed-off-by: Kai Fricke <[email protected]>
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 a pull request may close this issue.

1 participant