Solving Real-World Data Science Problems with LightGBM
With the preceding chapters, we have slowly been building out a toolset for us to be able to solve machine learning problems. We’ve seen examples of examining our data, addressing data issues, and creating models. This chapter formally defines and applies the data science process to two case studies.
The chapter gives a detailed overview of the data science life cycle and all the steps it encompasses. The concepts of problem definition, data exploration, data cleaning, modeling, and reporting are discussed in a regression and classification problem context. We also look at preparing data for modeling and building optimized LightGBM models using our learned techniques. Finally, we look deeper at utilizing a trained model as an introduction to machine learning operations (MLOps).
The main topics of this chapter are as follows:
- The data science life cycle
- Predicting wind turbine power generation with LightGBM...
