Classifier

BaggingClassifier

Multivariate Predict proba Missing values

Bagging ensemble of time series classifiers.

Fits n_estimators clones of a classifier on datasets which are instance sub-samples and/or variable sub-samples.

On predict_proba, the mean average of probabilistic predictions is returned. For a deterministic classifier, this results in majority vote for predict.

The estimator allows to choose sample sizes for instances, variables, and whether sampling is with or without replacement.

Direct generalization of sklearn’s BaggingClassifier to the time series classification task.

Note: if n_features=1, BaggingClassifier turns a univariate classifier into a multivariate classifier, because slices seen by estimator are all univariate. This can be used to give a univariate classifier multivariate capabilities.

Schnellstart

python

from sktime.classification.ensemble import BaggingClassifier

estimator = BaggingClassifier(estimator, n_estimators=10, n_samples=1.0, n_features=1.0, bootstrap=True, bootstrap_features=False, random_state=None)

Parameter(7)

estimatorsktime classifier, descendant of BaseClassifier: classifier to use in the bagging estimator
n_estimatorsint, default=10: number of estimators in the sample for bagging
n_samplesint or float, default=1.0: The number of instances drawn from X in fit to train each clone If int, then indicates number of instances precisely If float, interpreted as a fraction, and rounded by ceil
n_featuresint or float, default=1.0: The number of features/variables drawn from X in fit to train each clone If int, then indicates number of instances precisely If float, interpreted as a fraction, and rounded by ceil Note: if n_features=1, BaggingClassifier turns a univariate classifier into a multivariate classifier (as slices seen by estimator are all univariate).
bootstrapboolean, default=True: whether samples/instances are drawn with replacement (True) or not (False)
bootstrap_featuresboolean, default=False: whether features/variables are drawn with replacement (True) or not (False)
random_stateint, RandomState instance or None, optional (default=None): 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.

Beispiele

>>> from sktime.classification.ensemble import BaggingClassifier
>>> from sktime.classification.kernel_based import RocketClassifier
>>> from sktime.datasets import load_unit_test
>>> X_train, y_train = load_unit_test (split = "train")
>>> X_test, y_test = load_unit_test (split = "test")
>>> clf = BaggingClassifier (
... RocketClassifier (num_kernels = 100),
... n_estimators = 10,
... )
>>> clf. fit (X_train, y_train) BaggingClassifier(
... )
>>> y_pred = clf. predict (X_test)