search

h2oai / h2o-3

H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Means, PCA, Generalized Additive Models (GAM), RuleFit, Support Vector Machine (SVM), Stacked Ensembles, Automatic Machine Learning (AutoML), etc.
http://h2o.ai
Apache License 2.0
6.88k stars 1.99k forks source link

implement hyperparameter search (grid search) in h2o-dev #13480

Open exalate-issue-sync[bot] opened 1 year ago

exalate-issue-sync[bot] commented 1 year ago

  1. Write down the architecture we agreed on (1-pager).

  2. Build container / generator class for the back end.

  3. Get brute-force grid search working through the Java API. Cartesian product, no manually-maintained new parameters class (an annotation on the griddable parameters iff we want there to be non-griddable). Generated parameters class whose griddable fields are instances of GridParamSpec or some such.

In parallel:

3a. REST API endpoint. Again, no new manually-managed classes.

3b. Special handling of lambda in GLM, ntrees in tree models.

  1. R binding.

  2. Python binding.

  3. Sparkling Water.

exalate-issue-sync[bot] commented 1 year ago

Raymond Peck commented: Note: This needs to be split into tickets and won't be completed until Sprint 8 at the earliest.

I don't expect Cliff to do all the work, but he's the overall owner. Child tickets will be assigned to others.

exalate-issue-sync[bot] commented 1 year ago

Raymond Peck commented: I believe this is all done except for special handling of hyperparameters with cheap builds (lambda for GLM, ntrees for DRF, GBM).

DinukaH2O commented 1 year ago

JIRA Issue Migration Info

Jira Issue: PUBDEV-488 Assignee: Michal Malohlava Reporter: Raymond Peck State: Open Fix Version: N/A Attachments: N/A Development PRs: N/A