An Overview of LightGBM in Python
In the previous chapter, we looked at ensemble learning methods for decision trees. Both bootstrap aggregation (bagging) and gradient boosting were discussed in detail, with practical examples of how to apply the techniques in scikit-learn. We also showed how gradient-boosted decision trees (GBDTs) are slow to train and may underperform on some problems.
This chapter introduces LightGBM, a gradient-boosting framework that uses tree-based learners. We look at the innovations and optimizations LightGBM makes to the ensemble learning methods. Further details and examples are given for using LightGBM practically via Python. Finally, the chapter includes a modeling example using LightGBM, incorporating more advanced techniques for model validation and parameter optimization.
By the end of the chapter, you will have a thorough understanding of the theoretical and practical properties of LightGBM, allowing us to dive deeper into using LightGBM for data...
