Learning how to use various Python hyperparameter optimization platforms for machine learning models. We implement:
- Grid Search
- Random Search
- Bayesian Optimization
- Hyperopt
- Optuna
Bayesian, Hyperopt, and Optuna all provide similar peformance. These should be preferred over something like a simple grid search. In general, experimental with the model first to figure out what scope you want your parameters to be before running an n number of models. This will speed the process as the search space should be as tight as possible.
Download the data from the following location: https://www.kaggle.com/datasets/iabhishekofficial/mobile-price-classification