sklvq.distances.DistanceBaseClass¶
-
class
sklvq.distances.DistanceBaseClass[source]¶ Distance base class
Abstract class for implementing distance functions. It provides abstract methods with expected call signatures.
Custom distance function ‘__init__’ should accept any parameters as key-value pairs.
-
abstract
__call__(data: numpy.ndarray, model: LVQBaseClass) → numpy.ndarray[source]¶ Should implement a distance function.
- Parameters
- datandarray with shape (n_samples, n_features)
The samples for which the distance to the prototypes of the model need to be computed.
- modelLVQBaseClass
Any class extending the LVQBaseClass or depending on the type of distance function is implemented a class that provides the required attributes.
- Returns
- ndarray with shape (n_samples, n_prototypes)
Evaluation of the distance between each sample and prototype of the model.
-
abstract
gradient(data: numpy.ndarray, model: LVQBaseClass, i_prototype: int) → numpy.ndarray[source]¶ The distance gradient method.
- Parameters
- datandarray with shape (n_samples, n_features)
The data for which the distance gradient to the prototypes of the model need to be computed.
- modelLVQBaseClass
Any class extending the LVQBaseClass or depending on the type of distance function is implemented, a class that provides the required attributes.
- i_prototypeint
The index of the prototype for which the gradient needs to be computed.
- Returns
- ndarray with shape (n_samples, n_features)
The gradient with respect to the prototype (i_prototype) and every sample in X.
-
abstract
