Developed by Yandex researchers and engineers, CatBoost (which stands for categorical boosting) is a gradient boosting algorithm, based on decision trees, which is optimized in handling categorical features without much preprocessing (non-numeric features expressing a quality, such as a color, a brand, or a type). Since in most databases the majority of features are categorical, CatBoost can really boost your results on prediction:

CatBoost requires msgpack, which can be easily installed by using the pip install msgpack command.

Get Python Data Science Essentials - Third Edition now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.