MaxLossReductionMaxConfidence

class skactiveml.pool.MaxLossReductionMaxConfidence(missing_label=nan, random_state=None, target_type='auto')

Bases: SingleAnnotatorPoolQueryStrategy

Maximum Loss Reduction with Maximal Confidence (MMC)

This class implements the query strategy Maximum Loss Reduction with Maximal Confidence (MMC) [1] that selects the samples with the largest loss reduction under their most confident label assignment. That label assignment combines a multi-label classifier’s label predictions with the number of positive labels predicted by a label-cardinality discriminator. This strategy is multi-label-only: y must be two-dimensional and each row must be either fully labeled or fully unlabeled.

Parameters:

missing_labelscalar or string or np.nan or None, default=np.nan

Value to represent a missing label.

random_stateint or np.random.RandomState, default=None

Random state for candidate selection.

target_type“auto” or “multi-label”, default=”auto”

Declared target type. A fitted classifier’s target specification is authoritative when available. This strategy supports only multi-label classification with a single annotator.

References

[1]

Yang, B., Sun, J.-T., Wang, T., & Chen, Z. (2009). Effective Multi-Label Active Learning for Text Classification. In Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 917-926).

Methods

query(X, y, discriminator, clf[, fit_clf, ...])	Determines for which candidate samples labels are to be queried.
get_metadata_routing()	Get metadata routing of this object.
get_params([deep])	Get parameters for this estimator.
set_params(**params)	Set the parameters of this estimator.

MaxLossReductionMaxConfidence.query(X, y, discriminator, clf, fit_clf=True, candidates=None, batch_size=1, return_utilities=False)

Determines for which candidate samples labels are to be queried.

Parameters:

Xarray-like of shape (n_samples, n_features)

Training data set, usually complete, i.e., including the labeled and unlabeled samples.

yarray-like of shape (n_samples, n_outputs)

Labels of the training data set (possibly including unlabeled rows indicated by self.missing_label). Each row must either contain only observed labels or only missing_label values, i.e., no mixing within a row. This strategy supports multi-label data only. predict_proba must return either shape (n_samples, n_outputs) or a list of binary probability matrices with shape (n_samples, 2) per output.

discriminatorskactiveml.base.SkactivemlClassifier

Model implementing the methods fit and predict. It must support single-output classification with a single annotator and declare target_type=”auto” or target_type=”single-output”. It predicts a candidate sample’s number of positive labels, i.e., its label cardinality. The parameters classes and missing_label will be internally redefined. Class-dependent cost_matrix and class_prior parameters must already match the resulting label cardinality classes.

clfskactiveml.base.SkactivemlClassifier

Classifier implementing the methods fit and predict_proba.

fit_clfbool, default=True

Defines whether the classifier clf should be fitted on X and y.

candidatesNone or array-like of shape (n_candidates), dtype=int or array-like of shape (n_candidates, n_features), default=None

• If candidates is None, the unlabeled samples from (X, y) are considered as candidates.

• If candidates is of shape (n_candidates,) and of type int, candidates is considered as the indices of the samples in (X, y).

• If candidates is of shape (n_candidates, n_features), the candidate samples are directly given in candidates (not necessarily contained in X).

A given candidates is authoritative, i.e., an index array is taken as given, such that labeled samples remain candidates, e.g., to relabel them or to recompute their utilities.

batch_sizeint, default=1

The number of samples to be selected in one AL cycle.

return_utilitiesbool, default=False

If True, also return the utilities based on the query strategy.

Returns:

query_indicesnumpy.ndarray of shape (batch_size,)

The query_indices indicate for which candidate sample a label is to be queried, e.g., query_indices[0] indicates the index of the first selected sample. If candidates is None or of shape (n_candidates,), the indexing refers to samples in X. If candidates is of shape (n_candidates, n_features), the indexing refers to samples in candidates.

utilitiesnumpy.ndarray of shape (batch_size, n_samples) or numpy.ndarray of shape (batch_size, n_candidates)

The utilities of samples after each selected sample of the batch, e.g., utilities[0] indicates the utilities used for selecting the first sample (with index query_indices[0]) of the batch. Utilities for samples that are no candidates will be set to np.nan. If candidates is None or of shape (n_candidates,), the indexing refers to samples in X. If candidates is of shape (n_candidates, n_features), the indexing refers to samples in candidates.

Notes

An exhausted candidate pool, i.e., a fully labeled (X, y) queried with candidates=None or an empty candidates, is a valid acquisition state. It is answered with an empty batch of batch_size zero and a warning naming the exhaustion.

MaxLossReductionMaxConfidence.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.

MaxLossReductionMaxConfidence.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.

MaxLossReductionMaxConfidence.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.

Examples using skactiveml.pool.MaxLossReductionMaxConfidence

Maximum Loss Reduction with Maximal Confidence (MMC)

Maximum Loss Reduction with Maximal Confidence (MMC)