[{"data":1,"prerenderedAt":12},["ShallowReactive",2],{"stack-tr-en-lightgbm":3},{"tagline":4,"description":5,"strengths":6,"useCase":11},"Microsoft's gradient-boosting framework, built for training speed on very large datasets.","LightGBM trains gradient-boosted decision trees using a histogram-based approach: continuous values are bucketed into bins, which sharply cuts memory use and compute time. Leaf-wise tree growth, GOSS and EFB techniques for sampling gradients and bundling sparse features, and distributed learning make it a very fast, memory-efficient booster. It's our choice as soon as the data exceeds a million rows or hundreds of features and iterations need to stay short.",[7,8,9,10],"Histogram-based approach: features bucketed into bins, lower memory and compute.","Leaf-wise tree growth: fewer iterations to reach the same loss.","Native categorical feature support, no one-hot encoding required.","Distributed and GPU training, Python, R and C++ APIs, scikit-learn compatible.","Large-scale scoring, ranking and recommendation, forecasting, anomaly detection.",1792765050404]