Description
Machine-learning models can be trained with a fast gradient boosting framework based on decision trees.
This package is useful for ranking, classification, regression, and other tabular-data tasks where boosted tree models are appropriate. It is a framework and command-line/library tool, not a no-code analytics application.
Model quality depends on data, labels, validation, and tuning. Check privacy, bias, leakage, and reproducibility before using results in decisions.