catboost/catboost

A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.

What it solves

CatBoost is a high-performance machine learning library designed to implement gradient boosting on decision trees. It addresses the common challenge of handling categorical features and improving prediction speed and model quality across various datasets.

How it works

It utilizes a gradient boosting method over decision trees, providing native support for both numerical and categorical features. To scale training, it offers fast GPU and multi-GPU support, as well as distributed training capabilities via Apache Spark and a command-line interface.

Who it’s for

It is intended for data scientists and machine learning engineers who need a fast, high-quality gradient boosting library that can handle categorical data efficiently.

Highlights

  • Superior quality compared to other GBDT libraries on many datasets.
  • Best-in-class prediction speed.
  • Native support for categorical features.
  • Fast GPU, multi-GPU, and distributed training (Apache Spark/CLI).
  • Built-in visualization tools.

Related

  • Project
    dmlc/xgboost

    XGBoost is an optimized distributed gradient boosting library that provides fast and accurate parallel tree boosting for large-scale data science problems.

  • Project
    google/yggdrasil-decision-forests

    A fast and extensible library for training, evaluating, interpreting, and serving decision forest models like Random Forest and Gradient Boosted Trees.

  • Project
    fabsig/GPBoost

    A software library that combines tree-boosting with Gaussian processes and mixed-effects models to improve prediction accuracy for spatial data and high-cardinality categorical variables.

  • Project
    lightgbm-org/LightGBM

    A highly efficient gradient boosting framework that uses tree-based learning algorithms to handle large-scale data with high speed and low memory usage.

  • Project
    uxlfoundation/oneDAL

    A C++ and DPC++ library that provides hardware-accelerated machine learning routines for tabular data across CPUs, GPUs, and distributed systems.