Typical workflow for decision trees and decision forests - 5.2 English - 68552

Decision trees

1. Initialize a aoclda.decision_tree.decision_tree() object with options set in the class constructor.

2. Fit the model using aoclda.decision_tree.decision_tree.fit().

3. Evaluate prediction accuracy on test data using aoclda.decision_tree.decision_tree.score().

4. Make predictions using the fitted model using aoclda.decision_tree.decision_tree.predict().

Decision forests

1. Initialize a aoclda.decision_forest.decision_forest() object with options set in the class constructor.

2. Fit the model using the aoclda.decision_forest.decision_forest.fit().

3. Evaluate prediction accuracy on test data using aoclda.decision_forest.decision_forest.score().

4. Make predictions using the fitted model using aoclda.decision_forest.decision_forest.predict().

Decision trees

1. Initialize a da_handle with da_handle_type da_handle_decision_tree.

2. Pass data to the handle using da_tree_set_training_data_?.

3. Set optional parameters, such as maximum depth, using da_options_set_? (see options section).

4. Fit the model using da_tree_fit_?.

5. Evaluate prediction accuracy on test data using da_tree_score_?.

6. Make predictions using the fitted model using da_tree_predict_?.

Decision forests

1. Initialize a da_handle with da_handle_type da_handle_decision_forest.

2. Pass data to the handle using da_forest_set_training_data_?.

3. Set optional parameters, such as maximum depth, using da_options_set_? (see options section).

4. Fit the model using da_forest_fit_?.

5. Evaluate prediction accuracy on test data using da_forest_score_?.

6. Make predictions using the fitted model using da_forest_predict_?.