Created
March 11, 2014 11:14
-
-
Save bistaumanga/9483799 to your computer and use it in GitHub Desktop.
Alignment of tokens in 2 sentence to generate pattern using Dynamic time warp.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# -*- coding: utf-8 -*- | |
""" | |
Created on Tue Mar 11 11:05:05 2014 | |
@author: logpoint | |
""" | |
import numpy as np | |
def align_DTW(seq, ref_seq, window = 5): | |
# create the cost metric | |
d = lambda x, y : not x == y | |
# convert to numpy- array | |
seq, ref_seq = np.array(seq), np.array(ref_seq) | |
M, N = len(seq), len(ref_seq) | |
cost = float('inf') * np.ones((M, N)) | |
# initialize the first row and column | |
cost[0, 0] = d(seq[0], ref_seq[0]) | |
for i in range(1, M): | |
cost[i, 0] = cost[i-1, 0] + d(seq[i], ref_seq[0]) | |
for j in range(1, N): | |
cost[0, j] = cost[0, j-1] + d(seq[0], ref_seq[j]) | |
# fill in the rest of the matrix | |
for i in range(1, M): | |
for j in range(max(1, i - window), min(N, i + window)): | |
choices = cost[i - 1, j - 1], cost[i, j-1], cost[i-1, j] | |
is_different = d(seq[i], ref_seq[j]) | |
cost[i, j] = min(choices) + d(seq[i], ref_seq[j]) + is_different | |
# find the optimal path | |
n, m = N - 1, M - 1 | |
# path = [] | |
path_cost = [] | |
alignment = [] | |
while (m, n) != (0, 0): | |
# path.append((seq[m], ref_seq[n])) | |
if not d(seq[m], ref_seq[n]): | |
alignment.append(ref_seq[n]) | |
else: | |
alignment.append('*') | |
path_cost.append(cost[m, n]) | |
m, n = min((m - 1, n), (m, n - 1), (m - 1, n - 1), key = lambda x: cost[x[0], x[1]]) | |
if not d(seq[0], ref_seq[0]): | |
alignment.append(ref_seq[0]) | |
else: | |
alignment.append('*') | |
# path.append((seq[0],ref_seq[0])) | |
path_cost.append(0) | |
temp = " ".join(alignment[::-1]) | |
import re | |
return re.sub(r"((\s)*\*(\s)*){2,}"," * ", temp) | |
def align_simple(seq, ref_seq): | |
pattern = [] | |
for word1, word2 in zip(ref_seq, seq): | |
try: | |
assert word1 == word2 | |
pattern.append(word1) | |
except AssertionError: | |
pattern.append("*") | |
return " ".join(pattern) | |
if __name__ == "__main__": | |
text = ["user umanga logged in from 173.21.11.33 with mozilla browser", | |
"user puneet khanal logged in from localhost with chrome browser", | |
"user basanta logged out from 110.44.114.70", | |
"user ajay logged out from remote"] | |
# print align_DTW(text[0].split(), text[1].split()) | |
pattern = {} | |
for line in iter(text): | |
if not len(pattern): | |
pattern[1] = line | |
else: | |
log_tokens = line.split() | |
pattern_tokens = pattern[1].split() | |
pattern[1] = align_DTW(log_tokens, pattern_tokens) | |
print line, '||', pattern[1] | |
print pattern[1] |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment