sklvq.models.LVQBaseClass

class sklvq.models.LVQBaseClass(distance_type: Union[str, type] = 'squared-euclidean', distance_params: Optional[dict] = None, valid_distances: Optional[List[str]] = None, solver_type: Union[str, type] = 'steepest-gradient-descent', solver_params: Optional[dict] = None, valid_solvers: Optional[List[str]] = None, prototype_init: Union[str, numpy.ndarray] = 'class-conditional-mean', prototype_n_per_class: int = 1, random_state: Optional[Union[int, numpy.random.mtrand.RandomState]] = None, force_all_finite: Union[str, bool] = True)[source]

Learning Vector Quantization base class

Abstract class for implementing LVQ models. It provides abstract methods with expected call signatures.

Provides a common interface to the solver and other function that require access to the models. Additionally, it implements a number of functions shared by the currently implemented LVQ variations.

See also

GLVQ, GMLVQ, LGMLVQ
__init__(distance_type: Union[str, type] = 'squared-euclidean', distance_params: Optional[dict] = None, valid_distances: Optional[List[str]] = None, solver_type: Union[str, type] = 'steepest-gradient-descent', solver_params: Optional[dict] = None, valid_solvers: Optional[List[str]] = None, prototype_init: Union[str, numpy.ndarray] = 'class-conditional-mean', prototype_n_per_class: int = 1, random_state: Optional[Union[int, numpy.random.mtrand.RandomState]] = None, force_all_finite: Union[str, bool] = True)[source]

Initialize self. See help(type(self)) for accurate signature.

abstract add_partial_gradient(gradient, partial_gradient, i_prototype) → None[source]

To increase performance, the distance gradient methods (should) return only the relevant values. I.e., the gradient of the prototype i_prototype and potentially other parameters linked to this prototype. This partial gradient needs to added (overwrite) to the correct parts of the actual gradient and this is what this function should do.

Parameters
gradientndarray

Same shape as the get_variables() would return.

partial_gradientndarray

1d array containing the partial gradient.

i_prototypeint

The index of the prototype to which the partial gradient was computed.

decision_function(X: numpy.ndarray)[source]

Evaluates the decision function for the samples in X. Shape for binary class is (n_observations,) with the decision values for the “greater” class. In the multiclass case it returns decision values for each class and therefore has the shape (n_observations, n_classes).

Parameters
Xndarray

The data.

Returns
decision_valuesndarray

Binary case shape is (n_observations,) and the multiclass case (n_observations, n_classes)

fit(X: numpy.ndarray, y: numpy.ndarray)[source]
Fit function that provides the general implementation of the LVQ algorithms. It checks the data, calls

before_fit method, calls the solve method of the solver, and the after_fit method.

Parameters
Xndarray of shape (number of observations, number of dimensions)
yndarray of size (number of observations)
Returns
self

The trained model

abstract get_model_params() → Union[tuple, numpy.ndarray][source]

Should return a view or tuple of views (in correct shape) of the model’s parameters. Implementation depends on specific model as model parameters may differ per model.

Returns
ndarray or tuple

View or tuple of views of the model’s parameters.

get_params(deep=True)

Get parameters for this estimator.

Parameters
deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns
paramsdict

Parameter names mapped to their values.

get_prototypes() → numpy.ndarray[source]

Return a view into self._variables of the the shape of the prototypes (n_prototypes, n_features). At the moment only consistency function, does not actually create the shape and only works after self.prototypes_ has been set.

Returns
prototypesndarray of shape (n_prototypes, n_features)

View into self._variables with shape specified above.

get_variables() → numpy.ndarray[source]

Returns the self._variables array that owns the memory allocated for the model parameters.

Returns
_variablesndarray

returns the model’s _variables array.

abstract mul_step_size(step_size: Union[float, numpy.ndarray], gradient: numpy.ndarray) → None[source]

Should multiply the provided gradient with the provided step size and overwrite the values in gradient. Depending on the step_size being a float or array different step sizes are used for different model parameters (which also depends on the model if there are more then only prototypes)

Parameters
step_sizefloat or ndarray

The scalar or list of values containing the step sizes.

gradientndarray

Same shape as the get_variables() would return.

abstract normalize_variables(var_buffer: numpy.ndarray) → None[source]

Should modify the var_buffer as if it was the variables array provided by get_variables().

Parameters
var_bufferndarray

Array with the same size as the model’s variables array as returned by get_variables().

Returns
ndarray or tuple

Same shape and size as input, but normalized. How to normalize depends on model implementation.

predict(X: numpy.ndarray)[source]

Predict function

The decision is made for the label of the prototype with the minimum decision value, as provided by the decision_function().

Parameters
Xndarray

The data.

Returns
ndarray of shape (n_observations)

Returns the predicted labels.

predict_proba(X: numpy.ndarray)[source]
Parameters
Xndarray

The data.

Returns
confidence_scoresndarray of shape (n_observations, n_classes)
score(X, y, sample_weight=None)

Return the mean accuracy on the given test data and labels.

In multi-label classification, this is the subset accuracy which is a harsh metric since you require for each sample that each label set be correctly predicted.

Parameters
Xarray-like of shape (n_samples, n_features)

Test samples.

yarray-like of shape (n_samples,) or (n_samples, n_outputs)

True labels for X.

sample_weightarray-like of shape (n_samples,), default=None

Sample weights.

Returns
scorefloat

Mean accuracy of self.predict(X) wrt. y.

abstract set_model_params(new_model_params: Union[tuple, numpy.ndarray])[source]

Should modify the self._variables array. Accepts the new_model_params in the shape of the model’s parameters, e.g., prototypes or (prototypes, relevance_matrix).

Always needs to be all the model parameters, can not be used for partial updates.

Parameters
new_model_paramsndarray or tuple

Array or tuple of arrays of the new model’s parameters.

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters
**paramsdict

Estimator parameters.

Returns
selfestimator instance

Estimator instance.

set_prototypes(new_prototypes: numpy.ndarray) → None[source]

Accepts a new_prototypes array with the same shape as self.prototypes_ and overwrites the self._variables array by copying the values of the new_prototypes.

Parameters
new_prototypesndarray of shape (n_prototypes, n_features)

The new prototypes the model should store.

set_variables(new_variables: numpy.ndarray) → None[source]

Modifies the self._variables by copying the values of new_variables into the memory of self._variables.

Parameters
new_variablesndarray

1d numpy array that contains all the model parameters in continuous memory

abstract to_model_params_view(var_buffer: numpy.ndarray) → Union[tuple, numpy.ndarray][source]

Should return a single view into the var_buffer or a tuple of views. This depends on the model and its parameters.

Parameters
var_bufferndarray

Array with the same size as the model’s variables array as returned by get_variables().

Returns
ndarray or tuple

Should return a view or tuple of views of the model parameters in appropriate shapes.

abstract to_prototypes_view(var_buffer: numpy.ndarray) → numpy.ndarray[source]

Should return the prototypes from the provided var_buffer. I.e., it selects/views the appropriate part of memory and reshapes it.

Parameters
var_bufferndarray

Array with the same size as the model’s variables array as returned by get_variables().

Returns
ndarray of shape (n_prototypes, n_features)

View into the var_buffer.