dcase_models.util.evaluation_setup

dcase_models.util.evaluation_setup(fold_test, folds, evaluation_mode, use_validate_set=True)[source]

Return a evaluation setup given by the evaluation_mode.

Return fold list for training, validatin and testing the model.

Each evaluation_mode return different lists.

Parameters:
fold_test : str

Fold used for model testing.

folds : list of str

Fold list.

evaluation_mode : str

Evaluation mode (‘cross-validation’, ‘train-validate-test’, ‘cross-validation-with-test’, ‘train-test’)

use_validate_set : bool

If not, the validation set is the same as the train set.

Returns:
list

List of folds for training

list

List of folds for validating

list

List of folds for testing