[{"data":1,"prerenderedAt":12},["ShallowReactive",2],{"stack-tr-en-xgboost":3},{"tagline":4,"description":5,"strengths":6,"useCase":11},"The reference gradient-boosting library for tabular data.","XGBoost trains ensembles of decision trees with regularized gradient boosting, handles missing values natively, and parallelizes on CPU with GPU acceleration. scikit-learn compatible and available for Python, R and the JVM, it's our default choice when you need the best possible score on tabular data at a reasonable training cost.",[7,8,9,10],"State of the art on tabular data: scoring, churn, forecasting.","Regularized (L1\u002FL2) boosting and native missing-value handling.","Parallel training, GPU acceleration, large datasets.","Python, R and JVM APIs, compatible with scikit-learn pipelines and validation.","Credit scoring, churn, fraud detection and demand forecasting.",1792675979235]