sklvq.discriminants.DiscriminantBaseClass
- class sklvq.discriminants.DiscriminantBaseClass[source]
Discriminant base class
Abstract class for implementing discriminant functions. Provides abstract methods with expected call signatures.
Custom discriminative function ‘__init__’ should accept any parameters as key-value pairs.
See also
- abstract __call__(dist_same: ndarray, dist_diff: ndarray) ndarray[source]
Should implement a discriminant function
- Parameters:
- dist_samendarray of shape (n_samples, 1), with n_samples >= 1
Shortest distance of n_samples to a prototype with the same label.
- dist_diffndarray of shape (n_samples, 1), with n_samples >= 1
Shortest distance of n_samples to a prototype with a different label.
- Returns:
- ndarraywith shape (n_samples, 1)
Should perform a elementwise evaluation of a discriminant function.
- abstract gradient(dist_same: ndarray, dist_diff: ndarray, *, same_label: bool) ndarray[source]
Should implement the discriminant function’s gradient
- Parameters:
- dist_samendarray with shape (n_samples, 1), with n_samples >= 1
Shortest distance of n_samples to a prototype with the same label.
- dist_diffndarray with shape (n_samples, 1), with n_samples >= 1
Shortest distance of n_samples to a prototype with a different label.
- same_labelbool
Indicates if the gradient with respect to a prototype with the same label (True) or with respect to a prototype with a different label (False) needs to be computed.
- Returns:
- ndarray with shape (n_sampeles, 1)
Should perform a elementwise evaluation of a discriminant function’s gradient.
