XGBoost should not throw error when tree_method='AUTO' and grow_policy='lossguide'

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.

Apache License 2.0

6.87k stars 2k forks source link

Doing a grid search with grow_policy as hyperparameter and tree_method not set (default='AUTO'). Training fails with: {noformat} ModelBuilderErrorV3 (water.exceptions.H2OModelBuilderIllegalArgumentException): stacktrace = water.exceptions.H2OModelBuilderIllegalArgumentException: Illegal argument(s) for XGBoost model: xgboost_1_regression_small. Details: ERRR on field: _grow_policy: must use tree_method=hist for grow_policy=lossguide water.exceptions.H2OModelBuilderIllegalArgumentException.makeFromBuilder(H2OModelBuilderIllegalArgumentException.java:20) hex.ModelBuilder.trainModel(ModelBuilder.java:240) water.api.ModelBuilderHandler.handle(ModelBuilderHandler.java:64) water.api.ModelBuilderHandler.handle(ModelBuilderHandler.java:17) {noformat}

Should we not handle this case on backend when validating params and automatically set {{tree_method}} for the current model to {{hist}} if it was set to {{AUTO}}?

h2oai / h2o-3

XGBoost should not throw error when tree_method='AUTO' and grow_policy='lossguide' #12735