tslearn.utils.from_stumpy_dataset

tslearn.utils.from_stumpy_dataset(X)

Transform a stumpy-compatible dataset into a tslearn dataset.

Parameters:

X: list of arrays of shapes (d, sz_i) if d > 1 or (sz_i, ) otherwise

stumpy-formatted dataset.

Returns:

array, shape=(n_ts, sz, d), where sz is the maximum of all array lengths

tslearn-formatted dataset

Examples

>>> stumpy_arr = [numpy.random.randn(10), numpy.random.randn(10)]
>>> tslearn_arr = from_stumpy_dataset(stumpy_arr)
>>> tslearn_arr.shape
(2, 10, 1)
>>> stumpy_arr = [numpy.random.randn(3, 10), numpy.random.randn(3, 5)]
>>> tslearn_arr = from_stumpy_dataset(stumpy_arr)
>>> tslearn_arr.shape
(2, 10, 3)