bluecast.ml_modelling.parameter_tuning_utils

Module Contents

Functions

update_params_based_on_tree_method(→ Dict[str, Any])	Update parameters based on tree method.
get_params_based_on_device_xgboost(→ Dict[str, Any])	Get parameters based on available or chosen device.
get_params_based_on_device_catboost(→ Dict[str, Any])	Get parameters based on available or chosen device.
update_params_with_best_params(→ Dict[str, Any])	Update parameters based on best parameters after tuning.
sample_data(→ Tuple[pandas.DataFrame, ...)

bluecast.ml_modelling.parameter_tuning_utils.update_params_based_on_tree_method(param: Dict[str, Any], trial: optuna.Trial, xgboost_params: bluecast.config.training_config.XgboostTuneParamsConfig | bluecast.config.training_config.XgboostTuneParamsRegressionConfig) → Dict[str, Any]

Update parameters based on tree method.

bluecast.ml_modelling.parameter_tuning_utils.get_params_based_on_device_xgboost(conf_training: bluecast.config.training_config.TrainingConfig, conf_params_xgboost: bluecast.config.training_config.XgboostFinalParamConfig | bluecast.config.training_config.XgboostRegressionFinalParamConfig, conf_xgboost: bluecast.config.training_config.XgboostTuneParamsConfig | bluecast.config.training_config.XgboostTuneParamsRegressionConfig) → Dict[str, Any]

Get parameters based on available or chosen device.

bluecast.ml_modelling.parameter_tuning_utils.get_params_based_on_device_catboost(conf_training: bluecast.config.training_config.TrainingConfig, conf_params_catboost: bluecast.config.training_config.CatboostFinalParamConfig | bluecast.config.training_config.CatboostRegressionFinalParamConfig, conf_catboost: bluecast.config.training_config.CatboostTuneParamsConfig | bluecast.config.training_config.CatboostTuneParamsRegressionConfig) → Dict[str, Any]

Get parameters based on available or chosen device.

bluecast.ml_modelling.parameter_tuning_utils.update_params_with_best_params(param: Dict[str, Any], best_params: Dict[str, Any], model_type: str = 'xgboost') → Dict[str, Any]

Update parameters based on best parameters after tuning.

bluecast.ml_modelling.parameter_tuning_utils.sample_data(x_train: pandas.DataFrame, x_test: pandas.DataFrame, y_train: pandas.Series, y_test: pandas.Series, conf_training: bluecast.config.training_config.TrainingConfig) → Tuple[pandas.DataFrame, pandas.DataFrame, pandas.Series, pandas.Series]