CatBoost

CatBoost is an open-source machine learning library for developers and data scientists building prediction models with numerical and categorical data. Its built-in handling of categories reduces the work needed to turn non-numeric inputs into usable features. It runs computations on your own CPU or GPU and uses the Apache 2.0 license.

The library trains decision-tree models for classification, regression and ranking. That makes it relevant to applications such as recommendation systems, search ranking and weather prediction. Its default settings aim to reduce the need for extensive parameter tuning, while its boosting approach addresses overfitting during training.

CatBoost supports multiple GPUs for larger datasets, plus distributed training through Apache Spark and a command-line interface. The distributed training supports reproducible results. Built-in visualization tools help you inspect training, and cross-validation support helps you assess model performance.

Prediction speed is another focus, particularly for applications where response time matters. You can use trained models within applications through its model APIs. CatBoost has a C++ core and supports Python, R, Java and C++, so teams can train models in a data science environment and use them in production software.