CatBoost

Contents

Overview of CatBoost

Python package

R package

Command-line version

Applying models

Objectives and metrics

Data format description

Parameter tuning

Speeding up the training

Data visualization

Educational materials

Development and contributions

Algorithm details

Overview of CatBoost

Python package installation

Build from source on Linux and macOS

Build from source on Windows

Build a wheel package

Additional packages for data visualization support

R package installation

Install the released version

Build from source

Install from a local copy on Linux and macOS

Install from a local copy on Windows

Command-line version binary

Build the binary from a local copy on Linux and macOS

Build the binary from a local copy on Windows

Build the binary with make on Linux (CPU only)

Build the binary with MPI support from a local copy (GPU only)

Python package

Training parameters

calc_leaf_indexes

calc_feature_statistics

get_best_iteration

get_evals_result

get_feature_importance

get_object_importance

get_scale_and_bias

plot_predictions

randomized_search

select_features

set_scale_and_bias

set_feature_names

virtual_ensembles_predict

CatBoostClassifier

calc_leaf_indexes

calc_feature_statistics

get_best_iteration

get_evals_result

get_feature_importance

get_object_importance

get_scale_and_bias

plot_predictions

randomized_search

select_features

set_feature_names

set_scale_and_bias

staged_predict_proba

CatBoostRegressor

calc_leaf_indexes

calc_feature_statistics

get_best_iteration

get_evals_result

get_feature_importance

get_object_importance

get_scale_and_bias

plot_predictions

randomized_search

select_features

set_feature_names

set_scale_and_bias

rotten_tomatoes

get_cat_feature_count

get_feature_count

get_feature_names

get_num_feature_count

get_object_count

MetricVisualizer

get_cat_feature_indices

get_text_feature_indices

save_quantization_borders

set_feature_names

set_group_weight

set_pairs_weight

set_subgroup_id

Pool initialization

Text processing

unknown_token_id

end_of_sentence_token_id

min_unused_token_id

get_confusion_matrix

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

Select features

Applying models

C/C++

Evaluation library

Standalone evaluator

getPredictionDimension

getUsedCategoricFeatureCount

getUsedNumericFeatureCount

CatBoostPredictions

copyRowMajorPredictions

copyObjectPredictions

getPredictionDimension

Models exported as code

Objectives and metrics

Variables used in formulas

Multiregression

Multiclassification

Feature importance

Feature analysis charts

Feature interaction

Object importance

Data format description

Columns description

Dataset description in delimiter-separated values format

Dataset description in extended libsvm format

Pairs description

Custom quantization borders and missing value modes

Feature analysis

Feature importance

Feature interaction strength

Features selection result

Frequency Based Dictionary

Objects strength

Metrics and time information

Profiler information

Time information

Custom quantization borders and missing value modes

ROC curve points

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

Transforming text features to numerical features

Choosing the tree structure

Bootstrap options

Unbiased boosting

Overfitting detector

Recovering training after an interruption

Missing values processing

Score functions

Overview of CatBoost

Overview of CatBoost

CatBoost is a machine learning algorithm that uses gradient boosting on decision trees. It is available as an open source library.

Key features:

	Training Training Training on GPU Python train function Cross-validation Overfitting detector Pre-trained data Categorical features Text features		Applying models Regular prediction С and C++ Java Rust Calculate metrics Staged prediction Applying the model in ClickHouse
	Model analysis Feature importances Object importances		Metrics Implemented metrics User-defined metrics
	Recovery Snapshots		Visualization tools Jupyter Notebook TensorBoard
	Exporting models CoreML Python or C++ JSON ONNX PMML		Educational materials Tutorials Reference papers Videos

	Training Training Training on GPU Python train function Cross-validation Overfitting detector Pre-trained data Categorical features Text features		Applying models Regular prediction С and C++ Java Rust Calculate metrics Staged prediction Applying the model in ClickHouse
	Model analysis Feature importances Object importances		Metrics Implemented metrics User-defined metrics
	Recovery Snapshots		Visualization tools Jupyter Notebook TensorBoard
	Exporting models CoreML Python or C++ JSON ONNX PMML		Educational materials Tutorials Reference papers Videos