Overview of CatBoost

Python package

Training parameters

get_cat_feature_count

get_feature_count

get_feature_names

get_num_feature_count

get_object_count

MetricVisualizer

get_cat_feature_indices

set_feature_names

set_group_weight

set_pairs_weight

set_subgroup_id

Pool initialization

Pool initialization

get_gpu_device_count

select_threshold

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

Command-line version

Cross-validation

Calculate metrics

Calculate feature importance

Calculate object importance

Metadata manipulation

Objectives and metrics

Multiclassification

Feature importance

Feature analysis graphs

Feature interaction

Object importance

Parameter tuning

Speeding up the training

Data visualization

Jupyter Notebook

Educational materials

Reference papers

Development and contributions

Algorithm details

How training is performed

Preliminary calculation of splits

Transforming categorical features to numerical features

Choosing the tree structure

Bootstrap options

Unbiased boosting

Overfitting detector

Recovering training after an interruption

Missing values processing

Score functions

Educational materials

Tutorials

Tutorials

The following sources with tutorials are available:

The CatBoost tutorials repository.
The ClickHouse repository contains a tutorial on applying a CatBoost model in ClickHouse.
The How to Win a Data Science Competition: Learn from Top Kagglers specialization on Coursera has two introductory lectures (part 1 and part 2) and a programming assignment on CatBoost.

The assignment helps to explore all basic functions of the CatBoost Python package.

The results can be submitted to Coursera for further validation.