bluecast.ml_modelling.catboost_regression

Module Contents

Classes

CatboostModelRegression

Train and/or tune a CatBoost regression model.

class bluecast.ml_modelling.catboost_regression.CatboostModelRegression(class_problem: Literal[regression], conf_training: bluecast.config.training_config.TrainingConfig | None = None, conf_catboost: bluecast.config.training_config.CatboostTuneParamsRegressionConfig | None = None, conf_params_catboost: bluecast.config.training_config.CatboostRegressionFinalParamConfig | None = None, experiment_tracker: bluecast.experimentation.tracking.ExperimentTracker | None = None, custom_in_fold_preprocessor: bluecast.preprocessing.custom.CustomPreprocessing | None = None, cat_columns: List[str | float | int] | None = None, single_fold_eval_metric_func: bluecast.evaluation.eval_metrics.RegressionEvalWrapper | None = None)

Bases: bluecast.ml_modelling.base_classes.CatboostBaseModel

Train and/or tune a CatBoost regression model.

fit(x_train: pandas.DataFrame, x_test: pandas.DataFrame, y_train: pandas.Series, y_test: pandas.Series) → catboost.CatBoost

Train the CatBoost regressor. Includes hyperparameter tuning by default, then trains on full or partial data as specified.

autotune(x_train: pandas.DataFrame, x_test: pandas.DataFrame, y_train: pandas.Series, y_test: pandas.Series) → None

train_single_fold_model(train_pool: catboost.Pool, test_pool: catboost.Pool, y_test: pandas.Series, params: Dict[str, Any]) → float

Single-fold training approach. Trains a quick model and returns the metric from the single_fold_eval_metric_func.

_fine_tune_precise(tuned_params: Dict[str, Any], x_train: pandas.DataFrame, y_train: pandas.Series, x_test: pandas.DataFrame, y_test: pandas.Series) → float

fine_tune(x_train: pandas.DataFrame, x_test: pandas.DataFrame, y_train: pandas.Series, y_test: pandas.Series) → None

predict(df: pandas.DataFrame) → numpy.ndarray

Predict on unseen data using the trained CatBoost regressor. Returns numeric predictions.