evaluate_feature_set_to_feature

evaluate_feature_set_to_feature#

mlproject.corr_analysis.dependency_graph.evaluate_feature_set_to_feature(X_source, Y_targets, n_splits=5, scoring=None)[source]#

Train RandomForestRegressor models to predict each column in Y_targets from X_source using K-Fold cross-validation.

Parameters:

  • X_source (pd.DataFrame) – Feature matrix.

  • Y_targets (pd.DataFrame) – DataFrame of target features.

  • n_splits (int) – Number of folds for CV.

  • scoring (dict) – Dictionary of scoring functions for cross_validate.

Returns:

Per-fold metrics for each target feature. summary_df (pd.DataFrame): Mean ± std metrics for each target feature.

Return type:

metrics_df (pd.DataFrame)

Contents