interSet = len(genreValue1) + len(genreValue2) - unionSet;
#[unionSet.append(obj) for obj in (genreValue1 + genreValue2) if obj not in unionSet];
nominator = min( len(genreValue1), len( genreValue2 ));
return float(interSet)/ float(nominator);
#return float(interSet)/float(unionSet)
genreMap = GenreMap.GetMap();
# weekly aggregation.
print 'Merging weekly data...'
widx = 0;
logArr1 = DailyLog.mergeLogs( [DailyLog.createFromFile(dataDir + '/' + logfiles[fidx]) \
for fidx in range(widx * 7 + 0, widx * 7 + 7)]);
#logArr1 = DailyLog.createFromFile('./data/20131122.tsv');
print 'Computing distance...'
kvPairs = logArr1.genreDic.items();
pairwiseComp = dict([ ( (kvPairs[i][0], kvPairs[j][0]), \
genreMergeValue (kvPairs[i][1], kvPairs[j][1])) \
for i in range(0, len(kvPairs)) for j in range(0, len(kvPairs)) if i<j ]);
#print pairwiseComp
print 'Rank and output results...'
sortComp = sorted(pairwiseComp.iteritems(), key=operator.itemgetter(1), reverse = True);
# Run and get some statistics.
# Created: Jiayu Zhou, Jan 20, 2014.
from DailyLog import DailyLog;
from DailyLog import GenreMap;
import csv;
import time;
from os import listdir;
from jiayuUtils import *;
genreMap = GenreMap.GetMap();
log1 = DailyLog.createFromFile('./data/20131122.tsv');
# display genre dictionary.
for key,value in log1.genreDic.items():
print "["+key+"]", genreMap[key], ":", len(value);
print 'Number of total entries: ' + str(log1.length());
print 'Number of entries unique: ' + str(len(log1.progDic));
dataDir = './data';
logfiles = sorted(listdir(dataDir));
## compute the overlap week by week.
for widx in range(0, len(logfiles)/7 - 1):
print '---Weekly----'
if ('logArr2' in locals()):
