MatchingHistogramQuantifier#

class mlquantify.matching.MatchingHistogramQuantifier(bins_size, distance='hellinger', solver='auto', strategy='ovr', histogram_features=None, bin_strategy=None, laplace_smoothing=False)[source]#

Abstract base class for histogram-based distribution matching.

Subclasses learn class-conditional histogram representations from training data and estimate the test prevalence by finding the mixture of those histograms that best matches the test histogram.

This is a binary-only method. When applied to multiclass problems, a one-vs-rest (OvR) strategy is applied automatically.

Parameters:

bins_sizeint or array-like: Number of histogram bins, or array of bin counts to sweep over.
distancestr, default=’hellinger’: Distance function used to compare histograms.
solverstr, default=’auto’: Optimization solver; 'auto' selects based on the distance.
strategy{‘ovr’, ‘ovo’}, default=’ovr’: Multiclass decomposition strategy.
histogram_featuresint or None, default=None: Number of score columns used to build histograms.
bin_strategystr or None, default=None: Aggregation strategy across bin sizes ('median' or 'mean').

Attributes:

classes_ndarray of shape (n_classes,): Class labels seen during fit.

Examples

>>> from mlquantify.matching._histogram import MatchingHistogramQuantifier
>>> from sklearn.datasets import make_classification
>>> import numpy as np
>>> class MyHistQ(MatchingHistogramQuantifier):
...     def __init__(self):
...         super().__init__(bins_size=10)
...     def fit(self, X, y):
...         self.classes_ = np.unique(y)
...         return self._fit(X, y)
...     def predict(self, X):
...         return self._predict(X)
>>> X, y = make_classification(n_samples=200, random_state=42)
>>> MyHistQ().fit(X, y).predict(X)
{0: 0.5, 1: 0.5}

fit_predict(X, y, X_test)[source]#: Fit and predict class prevalences without storing models.

get_distance(dist_train, dist_test, distance='hellinger')[source]#: Compute a distance between two normalized representations.

get_metadata_routing()[source]#

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)[source]#

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[source]#: Save the quantifier instance to a file.

set_params(**params)[source]#

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.