A comparison between ExtraTreesClassifier and RandomForestClassifier for Statlog Shuttle data set from UCI archive
The dataset consists of 2 CSV files viz., 'shuttle_training.csv' and 'shuttle_test.csv'.
The accompanying Python code starts off with using base models of ExtraTreesClassifier and RandomForestClassifier, and measures model metrics such as accuracy, precision and recall, followed with using 5-fold Cross-Validation.
Finally, hyper parameter tuning is done for the best performing model using- 1.) 'hyperopt' using Bayesian Optimization 2.) 'RandomizedSearchCV' 3.) 'GridSearchCV'
Each of the parameter tuning, builds on top of the previous model.