ShapeDTW

ShapeDTW classifier.

Quickstart

from sktime.classification.distance_based import ShapeDTW

estimator = ShapeDTW(n_neighbors=1, subsequence_length=30, shape_descriptor_function='raw', shape_descriptor_functions=None, metric_params=None, n_splits=10)

Parameters(7)

n_neighborsint, int, set k for knn (default =1).

subsequence_lengthint, defines the length of the

subsequences(default=sqrt(n_timepoints)).

shape_descriptor_functionstring, defines the function to describe

the set of subsequences (default = ‘raw’).

The possible shape descriptor functions are as follows:

• ‘raw’use the raw subsequence as the

  shape descriptor function.

  • params = None

shape_descriptor_functionsstring list, only applicable when the

shape_descriptor_function is set to ‘compound’. Use a list of shape descriptor functions at the same time. (default = [‘raw’,’derivative’])

metric_paramsdictionary for metric parameters

(default = None).

n_splitsint, number of splits for cross-validation

(default = 10). Used for finding the weighting_factor if ‘shape_descriptor_function’ is set to ‘compound’ and weighting_factor is not given in ‘metric_params’.

Examples

>>> from sktime.classification.distance_based import ShapeDTW
>>> 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 = ShapeDTW (n_neighbors = 1,
... subsequence_length = 30,
... shape_descriptor_function = "raw",
... shape_descriptor_functions = None,
... metric_params = None,
... )
>>> clf. fit (X_train, y_train) ShapeDTW(
... )
>>> y_pred = clf. predict (X_test)