google/yggdrasil-decision-forests

A library to train, evaluate, interpret, and productionize decision forest models such as Random Forest and Gradient Boosted Decision Trees.

What it solves

It provides a fast and extensible library for working with decision forest models, simplifying the process of training, evaluating, interpreting, and serving these models.

How it works

The library implements several decision forest algorithms, including Random Forest, Gradient Boosted Decision Trees, CART, and Isolation Forests. It offers both Python and C++ APIs to handle the full machine learning lifecycle, from loading datasets (e.g., via Pandas) to training, generating predictions, and benchmarking inference speed.

Who it’s for

Data scientists and machine learning engineers who need a high-performance library for training and deploying decision forest models.

Highlights

  • Support for multiple model types: Random Forest, Gradient Boosted Trees, CART, and Isolation Forests.
  • Integrated tools for model analysis, such as partial dependence plots and variable importance.
  • Built-in evaluation metrics including ROC, accuracy, and confusion matrices.
  • Dual API support for Python and C++.

Related

  • Project
    catboost/catboost

    A high-performance gradient boosting library for decision trees that provides native support for categorical features and fast prediction speeds.

  • 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
    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
    uxlfoundation/oneDAL

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

  • Project
    BehaviorTree/BehaviorTree.CPP

    A C++ 17 library for creating Behavior Trees to manage complex decision-making in robotics and game AI, offering a reactive and flexible alternative to Finite State Machines.