Databricks - Machine Learning Associate Study Guide

3: Model Development

This chapter builds the model-development core of the Databricks Machine Learning Associate exam: matching an algorithm and an evaluation metric to a stated problem, correcting class imbalance, distinguishing estimators from transformers inside scikit-learn and Spark ML pipelines, and tuning hyperparameters with Hyperopt at scale. It closes with the mechanics of cross-validation, the parallelization of single-node tuning, and the arithmetic of how many models a grid search actually trains.