test_sampling.py

import numpy
import pylab

from sampler import DTWSampler

__author__ = 'Romain Tavenard romain.tavenard[at]univ-rennes2.fr'

ref_dim = 0

data = []
data.append(numpy.loadtxt("data/Xi_ref.txt"))
data.append(numpy.loadtxt("data/Xi_0.txt"))
data.append(numpy.loadtxt("data/Xi_1.txt"))

d = data[0].shape[1]

max_sz = max([ts.shape[0] for ts in data])
npy_arr = numpy.zeros((len(data), max_sz, d)) + numpy.nan
for idx, ts in enumerate(data):
    sz = ts.shape[0]
    npy_arr[idx, :sz] = ts
npy_arr = npy_arr.reshape((len(data), -1))

s = DTWSampler(scaling_col_idx=ref_dim, reference_idx=0, d=d, interp_kind="linear")
transformed_array = s.fit_transform(npy_arr)

pylab.figure()
for i in range(d):
    pylab.subplot(2, 3, i + 1)  # Original data
    if i == ref_dim:
        pylab.title("Reference dimension")
    else:
        pylab.title("Dimension %d" % i)
    if i == 0:
        pylab.ylabel("Original data")
    for ts in npy_arr.reshape((len(data), -1, d)):
        sz = ts.shape[0]
        pylab.plot(numpy.arange(sz), ts[:, i])

    pylab.subplot(2, 3, i + d + 1)  # Transformed data
    if i == 0:
        pylab.ylabel("Resampled data")
    for ts in transformed_array.reshape((len(data), -1, d)):
        sz = ts.shape[0]
        pylab.plot(numpy.arange(sz), ts[:, i])
pylab.show()