ProximityTree

Proximity Tree class.

Quickstart

from sktime.classification.distance_based import ProximityTree

estimator = ProximityTree(random_state=None, distance_measure=None, max_depth=inf, is_leaf=<function pure>, verbosity=0, n_jobs=1, n_stump_evaluations=5)

Parameters(7)

random_state: int or np.RandomState, default=0

random seed for the random number generator

distance_measure: ``None`` (default) or str; if str, one of

euclidean, dtw, ddtw, wdtw, wddtw, msm, lcss, erp distance measure to use if None, selects distances randomly from the list of available distances

max_depth: int or math.inf, default=math.inf

maximum depth of the tree

is_leaffunction, default=pure

decide when to mark a node as a leaf node

verbosity: 0 or 1

number reflecting the verbosity of logging 0 = no logging, 1 = verbose logging

n_jobs: int or None, default=1

number of parallel threads to use while building

n_stump_evaluations: number of stump evaluations to do if find_stump method is None

Examples

>>> from sktime.classification.distance_based import ProximityTree
>>> from sktime.datasets import load_unit_test
>>> X_train, y_train = load_unit_test (split = "train", return_X_y = True)
>>> X_test, y_test = load_unit_test (split = "test", return_X_y = True)
>>> clf = ProximityTree (max_depth = 2, n_stump_evaluations = 1)
>>> clf. fit (X_train, y_train) ProximityTree(
... )
>>> y_pred = clf. predict (X_test)