##Dynamic Time Warping Project
Installation
pip install cdtw
This module implements:
Distance functions:
- manhattan
- euclidean
- squared euclidean
Local constraints(step patterns, step functions):
- well known step patterns dp1, dp2, dp3
- local constraints classified by Sakoe-Chiba impo Global constraints(windows):
- Itakura parallelogram
- Sakoe-chiba band, Palival adjustment window
import numpy as np
from cdtw import pydtw
r = np.array([1,2,3,4])
q = np.array([2,3,4,5])
d = pydtw.dtw(r,q,pydtw.Settings(step = 'p0sym', #Sakoe-Chiba symmetric step with slope constraint p = 0
window = 'palival', #type of the window
param = 2.0, #window parameter
norm = False, #normalization
compute_path = True))
d.get_dist()
#2.0
d.get_cost()
#array([[ 1., 3., 6., inf],
# [ 1., 2., 4., 7.],
# [ 2., 1., 2., 4.],
# [ inf, 2., 1., 2.]])
d.get_path()
#[(0, 0), (1, 0), (2, 1), (3, 2), (3, 3)]