dtw(method=’itakura’)

pyts.metrics.dtw(x=None, y=None, dist='square', method='itakura', options={'max_slope': 2.0}, precomputed_cost=None, return_cost=False, return_accumulated=False, return_path=False)

Dynamic Time Warping distance with Itakura parallelogram constraint.

The Itakura constraint region is a parallelogram. Contrary to the Sakoe-Chiba band, whose width is constant, the Itakura parallelogram has a varying width, allowing for larger time shifts in the middle than at the first and last time points.

See also

For documentation for the rest of the parameters, see dtw().

Options:

max_slope : float (default = 2.)

Maximum slope of the parallelogram.

References

[1]	F. Itakura, “Minimum prediction residual principle applied to speech recognition”. IEEE Transactions on Acoustics, Speech, and Signal Processing, 23(1), 67–72 (1975).