ertk.classification.cross_validate

ertk.classification.cross_validate(clf_lib: str, clf, x: ndarray, y: ndarray | None = None, *, groups: ndarray | None = None, cv: BaseCrossValidator | None = None, scoring: str | List[str] | Dict[str, Callable[[ndarray, ndarray], float]] | Callable[[...], float] = 'accuracy', verbose: int = 0, n_jobs: int = 1, fit_params: Dict[str, Any] = {})

Cross validate a classifier.

Parameters:

clf_lib: str: Classifier library to use. Must be one of “sk”, “tf”, or “pt”.
clf:: Classifier to cross validate.
x: np.ndarray: Input data.
y: np.ndarray, optional: Target labels.
groups: np.ndarray, optional: Groups to split data into for cross validation.
cv: BaseCrossValidator, optional: Cross validation strategy. If None, use StratifiedKFold.
scoring: str, list, dict, or callable, optional: Scoring metric(s) to use. If a string, must be a valid scikit-learn scorer.
verbose: int, optional: Verbosity level.
n_jobs: int, optional: Number of jobs to run in parallel. Only used for scikit-learn.
fit_params: dict, optional: Parameters to pass to the classifier’s fit method.

Returns: