ts_distance.py

import numpy as np
from scipy.special import rel_entr
import sklearn 
from sklearn.metrics import mutual_info_score

num_of_ts=72
x_size=480
y_size=480

data=np.load("C:\\Users\\karm_ch\\Documents\\LDA_uncertainty_paper\\time_series_numpy\\labeler_2_ac_2018_03_14.238490133948005_53.76003136277209_0_gmm_labels2.npy")
data_temp1=np.reshape(data, (num_of_ts, (x_size * y_size)))
dataset= data_temp1.T
labels=np.load("C:\\Users\\karm_ch\\Documents\\LDA_uncertainty_paper\\time_series_numpy\\labeler_2_ac_2018_03_14.238490133948005_53.76003136277209_0_change_gmlabels04.npy")

dict={}
for A, B in zip(labels, dataset):
    dict[A]=B


string_1 = dict.get(0)
string_2 = dict.get(1)



def kl_divergence(a, b):
    return sum(a[i] * np.log(a[i]/b[i]) for i in range(len(a)))
  
'''print('KL-divergence(string_1 || string_2): %.3f ' % kl_divergence(string_1,string_2))
print('KL-divergence(string_2 || string_1): %.3f ' % kl_divergence(string_2,string_1))

# D( p || p) =0
print('KL-divergence(string_1 || string_1): %.3f ' % kl_divergence(string_1,string_1))

#print("Using Scipy rel_entr function")
string_1 = np.array(string_1)
string_2 = np.array(string_2)

print('KL-divergence(string_1 || string_2): %.3f ' % sum(rel_entr(string_1,string_2)))
print('KL-divergence(string_2 || string_1): %.3f ' % sum(rel_entr(string_2,string_1)))
print('KL-divergence(string_1 || string_1): %.3f ' % sum(rel_entr(string_1,string_1)))
'''
print(sklearn.metrics.mutual_info_score(list(string_1),list(string_2)))