skactiveml.pool.multiannotator.IntervalEstimationAnnotModel#

class skactiveml.pool.multiannotator.IntervalEstimationAnnotModel(classes=None, missing_label=nan, alpha=0.05, mode='upper', random_state=None)[source]#

Bases: BaseEstimator, AnnotatorModelMixin

This annotator model relies on ‘Interval Estimation Learning’ (IELearning) for estimating the annotation performances, i.e., labeling accuracies, of multiple annotators [1]. Therefore, it computes the mean accuracy and the lower as well as the upper bound of the labeling accuracy per annotator. (Weighted) majority vote is used as estimated ground truth.

Parameters

classesarray-like, shape (n_classes), optional (default=None): Holds the label for each class.
missing_labelscalar or string or np.nan or None, optional
(default=np.nan): Value to represent a missing label.
alphafloat, interval=(0, 1), optional (default=0.05): Half of the confidence level for student’s t-distribution.
mode‘lower’ or ‘mean’ or ‘upper’, optional (default=’upper’): Mode of the estimated annotation performance.
random_stateNone|int|numpy.random.RandomState, optional (default=None): The random state used for deciding on majority vote labels in case of ties.

References

[1] Donmez, Pinar, Jaime G. Carbonell, and Jeff Schneider.: “Efficiently learning the accuracy of labeling sources for selective sampling.” 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 259-268. 2009.

Attributes

n_annotators_: int: Number of annotators.
A_perf_ndarray, shape (n_annotators, 3): Estimated annotation performances (i.e., labeling accuracies), where A_cand[i, 0] indicates the lower bound, A_cand[i, 1] indicates the mean, and A_cand[i, 2] indicates the upper bound of the estimation labeling accuracy.

Methods

`fit`(X, y[, sample_weight])	Fit annotator model for given samples.
`get_metadata_routing`()	Get metadata routing of this object.
`get_params`([deep])	Get parameters for this estimator.
`predict_annotator_perf`(X)	Calculates the probability that an annotator provides the true label for a given sample.
`set_fit_request`(*[, sample_weight])	Request metadata passed to the `fit` method.
`set_params`(**params)	Set the parameters of this estimator.

fit(X, y, sample_weight=None)[source]#

Fit annotator model for given samples.

Parameters

Xarray-like, shape (n_samples, n_features): Test samples.
yarray-like, shape (n_samples, n_annotators): Class labels of annotators.
sample_weightarray-like, shape (n_samples, n_annotators),
optional (default=None): Sample weight for each label and annotator.

Returns

selfIntervalEstimationAnnotModel object: The fitted annotator model.

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.

predict_annotator_perf(X)[source]#

Calculates the probability that an annotator provides the true label for a given sample.

Parameters

Xarray-like, shape (n_samples, n_features): Test samples.

Returns

P_annotnumpy.ndarray, shape (n_samples, n_annotators): P_annot[i,l] is the probability, that annotator l provides the correct class label for sample X[i].

set_fit_request(*, sample_weight: Union[bool, None, str] = '$UNCHANGED$') → IntervalEstimationAnnotModel#

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see 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.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters

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.