mlresearch.active_learning.AugmentationAL

class mlresearch.active_learning.AugmentationAL(classifier: BaseEstimator | ClassifierMixin = None, generator: BaseOverSampler = None, param_grid: dict = None, cv=None, acquisition_func=None, n_init: int | float = None, budget: int | float = None, max_iter: int = None, evaluation_metric=None, continue_training: bool = False, random_state: int = None)

Active Learning with pipelined Data Augmentation. This method is implemented and analysed in a working paper.

Parameters:

classifierclassifier object, default=None

Classifier or pipeline to be trained in the iterative process. If None, defaults to sklearn’s RandomForestClassifier with default parameters and uses the random_state passed in the Active Learning model.

generatorgenerator estimator, default=None

Generator to be used for artificial data generation within Active Learning iterations.

param_griddict or list of dictionaries

Used to optimize the classifier and generator hyperparameters at each iteration via cross-validated grid-search. If None, parameter tuning is skipped. Dictionary with parameters names (str) as keys and lists of parameter settings to try as values, or a list of such dictionaries, in which case the grids spanned by each dictionary in the list are explored. This enables searching over any sequence of parameter settings.

cvint, cross-validation generator or an iterable, default=None

Determines the cross-validation splitting strategy. Used to optimize the classifier and generator hyperparameters at each iteration. Possible inputs for cv are:

• None, to use the default 5-fold cross validation,

• integer, to specify the number of folds in a (Stratified)KFold,

• CV splitter.

For integer/None inputs, if the estimator is a classifier and y is either binary or multiclass, StratifiedKFold is used. In all other cases, KFold is used. These splitters are instantiated with shuffle=False so the splits will be the same across calls.

acquisition_funcfunction or {‘entropy’, ‘breaking_ties’, ‘random’}, default=None

Method used to quantify the prediction’s uncertainty level. All predefined functions are set up so that a higher value means higher uncertainty (higher likelihood of selection) and vice-versa. The uncertainty estimate is used to select the instances to be added to the labeled/training dataset. Acquisition functions may be added or changed in the UNCERTAINTY_FUNCTIONS dictionary. If None, defaults to “random”.

n_initint or float, default=None

Number of observations to include in the initial training dataset. If n_init < 1, then the corresponding percentage of the original dataset will be used as the initial training set. If None, defaults to 2% of the size of the original dataset.

budgetint or float, default=None

Number of observations to be added to the training dataset at each iteration. If budget < 1, then the corresponding percentage of the original dataset will be used as the initial training set. If None, defaults to 2% of the size of the original dataset.

max_iterint, default=None

Maximum number of iterations allowed. If None, the experiment will run until 100% of the dataset is added to the training set.

evaluation_metricstring, default=’accuracy’

Metric used to calculate the test scores. See mlresearch.metrics for info on available performance metrics.

continue_trainingbool, default=False

If False, fit a new classifier at each iteration. If True, the classifier fitted in the previous iteration is used for further training in subsequent iterations.

random_stateint, RandomState instance, default=None

Control the randomization of the algorithm.

• If int, random_state is the seed used by the random number generator;

• If RandomState instance, random_state is the random number generator;

• If None, the random number generator is the RandomState instance used by np.random.

Attributes:

acquisition_func_function

Method used to calculate the classification uncertainty at each iteration.

evaluation_metric_scorer

Metric used to estimate the performance of the AL classifier at each iteration.

classifier_estimator object

The classifier used in the iterative process. It is the classifier fitted in the last iteration.

metadata_dict

Contains the performance estimations, classifiers, labeled pool mask and original dataset.

n_init_int

Number of observations included in the initial training dataset.

budget_int

Number of observations to be added to the training set per iteration.

max_iter_int

Maximum number of iterations allowed.

labeled_pool_array-like of shape (n_samples,)

Mask that filters the labeled observations from the original dataset.

fit(X, y, **kwargs)

Fit an Active Learning model from training set (X, y).

Parameters:

X{array-like, sparse matrix} of shape (n_samples, n_features)

The training input samples.

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

The target values (class labels) as integers or strings.

Returns:

selfActive Learning Classifier

Fitted Active Learning 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.

initialization(X, y, initial_selection=None, **kwargs)

iteration(X, y, **kwargs)

predict(X)

Predict class or regression value for X.

For a classification model, the predicted class for each sample in X is returned. For a regression model, the predicted value based on X is returned.

Parameters:

X{array-like, sparse matrix} of shape (n_samples, n_features)

The test input samples.

Returns:

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

The predicted classes, or the predict values.

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) w.r.t. y.

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_score_request(*, sample_weight: bool | None | str = '$UNCHANGED$') → AugmentationAL

Request metadata passed to the score 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 score 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 score.

• 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 score.

Returns:

selfobject

The updated object.