check_classifier_params

skactiveml.utils.check_classifier_params(classes, missing_label, cost_matrix=None)

Check whether the general classifier are compatible with each other.

Parameters:

classesarray-like of shape (n_classes,) or a list of such array-likes, default=None

The classes labels (single output setting), or a list of arrays of class labels (multioutput setting).

missing_labelscalar or string or np.nan or None, default=np.nan

Value to represent a missing label. In the case of a multioutput setting, we expect that the missing label is identical across all tasks.

cost_matrixarray-like of shape (n_classes, n_classes), default=None

Cost matrix to quantify costs of misclassifications.

• Checked only for single output.

• Must be None for multioutput.