Overview of CatBoost

Python package

get_cat_feature_indices

set_feature_names

set_group_weight

set_subgroup_id

set_pairs_weight

Pool initialization

get_object_count

get_num_feature_count

get_cat_feature_count

get_feature_count

get_feature_names

MetricVisualizer

get_gpu_device_count

select_threshold

Training parameters

Pool initialization

Missing values processing

R package

catboost.load_pool

catboost.save_pool

catboost.load_model

catboost.save_model

catboost.predict

catboost.shrink

catboost.staged_predict

catboost.get_feature_importance

catboost.get_object_importance

catboost.get_model_params

Training parameters

Missing values processing

Command-line version

Cross-validation

Calculate metrics

Calculate feature importance

Calculate object importance

Metadata manipulation

Missing values processing

Objectives and metrics

Multiclassification

Feature importance

Object importance

Parameter tuning

Speeding up the training

Data visualization

Jupyter Notebook

CatBoost Viewer

Educational materials

Reference papers

Development and contributions

Reference

How training is performed

Preliminary calculation of splits

Transforming categorical features to numerical features

Choosing the tree structure

Unbiased boosting

Overfitting detector

Recovering training after an interruption

Applying the model in ClickHouse

Applying the model in ClickHouse

A CatBoost model can be applied in ClickHouse.

A tutorial is available in the ClickHouse repository.