search

fabsig / GPBoost

Combining tree-boosting with Gaussian process and mixed effects models
Other
530 stars 42 forks source link

Get complete hyperparameter tuning history #138

Closed diggi2395 closed 1 month ago

diggi2395 commented 1 month ago

I am using gpb.grid.search.tune.parameters() which returns the best hyperparameter combinations identified. However, I'd like to get the results of each combination. Is such a functionality already implemented?

fabsig commented 1 month ago

Not yet. Thanks for the suggestion. Will get back to you

fabsig commented 1 month ago

This feature is now implemented (version 1.5.0 or latter, it is on CRAN & PyPI). You just need to set return_all_combinations = True in the gpb.grid.search.tune.parameters function. Then the full parameter combination that has been tried is returned together with the best combination.

Thank you for using GPBoost and for raising this issue!