KDEyCS

class mlquantify.matching.KDEyCS(estimator=None, bandwidth=0.1, kernel='gaussian', solver='slsqp', cv=None, stratified=True, shuffle=False, random_state=None)

KDEy Cauchy-Schwarz (KDEy-CS) quantifier.

Targets prior probability shift. A multiclass distribution-matching method that models each class with a Gaussian KDE over the posterior vectors and selects the prevalence minimising the Cauchy-Schwarz divergence, which has a closed form. The most efficient KDEy variant, since the expensive train-train kernel terms are precomputed once at fit time.

Parameters:

estimatorestimator, optional

A probabilistic classifier with fit and predict_proba methods.

bandwidthfloat, default=0.1

Bandwidth of the Gaussian kernel density estimator; controls smoothness.

kernelstr, default=’gaussian’

Must be 'gaussian'; the closed-form divergence is Gaussian-specific.

solverstr, default=’slsqp’

Constrained optimizer over the simplex (see mlquantify.solvers).

cvint or None, default=None

Cross-validation folds for computing training scores.

stratifiedbool, default=True

Whether to stratify CV splits.

shufflebool, default=False

Whether to shuffle data before splitting.

random_stateint or None, default=None

Random seed for reproducibility.

Attributes:

estimator_estimator

The fitted underlying classifier.

classes_ndarray of shape (n_classes,)

Class labels seen during fit.

See also

KDEyML

Maximum-likelihood KDE variant.

KDEyHD

Hellinger-distance KDE variant.

Notes

The closed form makes KDEy-CS the fastest KDEy variant: the train-train kernel matrices are computed once at training time and reused for any test bag.

References

References

[1]

Moreo, A., González, P., & del Coz, J. J. (2024). Kernel Density Estimation for Multiclass Quantification. Machine Learning, 113, 3075–3107.

Examples

>>> from mlquantify.matching import KDEyCS
>>> from sklearn.linear_model import LogisticRegression
>>> from sklearn.datasets import make_classification
>>> X, y = make_classification(n_samples=200, random_state=42)
>>> q = KDEyCS(estimator=LogisticRegression()).fit(X, y)
>>> q.predict(X)
{0: ..., 1: ...}

get_distance(dist_train, dist_test, distance='hellinger')

Compute a distance between two normalized representations.

get_metadata_routing()

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:

routingMetadataRequest

A MetadataRequest encapsulating routing information.

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.

save_quantifier(path: str | None = None) → None

Save the quantifier instance to a file.

set_fit_request(*, cv_prediction: bool | None | str = '$UNCHANGED$', estimator_fitted: bool | None | str = '$UNCHANGED$', sample_weight: bool | None | str = '$UNCHANGED$') → KDEyCS

Configure whether metadata should be requested to be passed to the fit method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

cv_predictionstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for cv_prediction parameter in fit.

estimator_fittedstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for estimator_fitted parameter in fit.

sample_weightstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for sample_weight parameter in fit.

Returns:

selfobject

The updated object.

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.