train_eval_fold

train_eval_fold#

mlproject.training.fold_trainer.train_eval_fold(model_type, fold_ind, X_train, X_test, y_train, y_test, target_name=None, n_jobs=1, save_model=True, **kwargs)[source]#

Training/evaluation for MODNet, GA-SISSO, and RandomForest.

Parameters:

  • model_type (str) – Type of model to train (“modnet”, “ga_sisso”, “rf”).

  • fold_ind (int) – Fold index for cross-validation.

  • X_train (pd.DataFrame) – Training features.

  • X_test (pd.DataFrame) – Testing features.

  • y_train (pd.DataFrame) – Training target.

  • y_test (pd.DataFrame) – Testing target.

  • target_name (str, optional) – Name of the target variable (required for MODNet).

  • n_jobs (int, optional) – Number of parallel jobs (default is 1).

  • save_model (bool, optional) – Whether to save the trained model (default is True).

  • **kwargs – Additional keyword arguments for the specific model training function.

Returns:

Dictionary containing training and testing regression metrics.

Return type:

dict

Contents