This example shows time based early stopping for non-model trainer function Antiselect.
- Define hyperparameter tuning for Antiselect.
- Define the non-model trainer function parameter space.
>>> non_model_trainer_params = {"data":data, "exclude":( ['petal_length', 'petal_width'], ['sepal_length', 'sepal_width', 'petal_length'], ['id', 'sepal_length', 'sepal_width'], ['petal_width'], ['petal_width', 'species'], ['sepal_width', 'petal_length', 'petal_width', 'species'], ['sepal_width', 'petal_length', 'petal_width', 'species', 'id'], ['sepal_length', 'sepal_width',]) }
- Import non-model trainer function and optimizer.
>>> from teradataml import Antiselect, GridSearch
- Initialize the GridSearch optimizer with non-model trainer function and parameter space required for non-model training.
>>> as_obj = GridSearch(func=Antiselect, params=non_model_trainer_params)
- Define the non-model trainer function parameter space.
- Run Antiselect with max_time set to 5 seconds.
>>> as_obj.fit(max_time=5, verbose=1)
Computing: |⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾⫾| 50% - 4/8
- View the result of hyperparameter tunning using models property.
>>> as_obj.models
MODEL_ID PARAMETERS STATUS 0 ANTISELECT_3 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... PASS 1 ANTISELECT_4 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... SKIP 2 ANTISELECT_5 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... SKIP 3 ANTISELECT_6 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... SKIP 4 ANTISELECT_7 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... SKIP 5 ANTISELECT_0 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... PASS 6 ANTISELECT_2 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... PASS 7 ANTISELECT_1 {'data': '"AUTOML_USER"."ml__td_sqlmr_out__170... PASS