tmp = line.split()
if len(tmp) == 3:
i = int(tmp[0])
w = tmp[2]
l = int(tmp[1])
if i == prev:
wtol[w].append(l)
else:
wtol[w] = [l]
itow[i] = w
prev = i
elif len(tmp) == 2:
itow[int(tmp[0])] = tmp[1]
fwl.close()
wtolu = sm.readwl("/home/ec2-user/git/statresult/wordslist_dsw_top1000.txt")
tmp = []
bk = {}
fbk = open(sys.argv[1],"r")
for line in fbk:
if len(line) < 2:
for term in tmp:
bk[term] = tmp
tmp =[]
else:
tmp.append(itow[int(line.split()[0])])
fbk.close()
otype = int(sys.argv[3])
otype = 0
if sys.argv[6] == 'stdout':
otype = 1
def vecof(lines, a, wtol, kk):
vec = np.zeros(kk)
for line in lines:
line = line.strip('\n')
vec = vec + a[:, wtol[line]]
return vec
stype = int(sys.argv[5])
zipf = float(sys.argv[4])
wtola = sm.readwl("/home/ec2-user/git/statresult/wordslist_dsw.txt")
#zipf = crand.zipf_init(len(wtola))
if stype == 3:
wtolu = sm.readwl(
"/home/ec2-user/git/statresult/wordslist_top10000_dsw.txt")
else:
wtolu = wtola
a = np.load('/home/ec2-user/data/classinfo/vt.npy') #lsa result
b = a
ukk = a.shape[0]
s = None
akk = ukk
if stype == 3:
# b = np.load('/home/ec2-user/git/statresult/lda-30-2000-phi.npy')
# s = np.load('/home/ec2-user/git/statresult/lda-30-2000-pz.npy')
b = np.load('/home/ec2-user/git/statresult/lda-32-2000-top10000-phi.npy')
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import os
import sys
import math
from os import path
sys.path.append( path.dirname( path.dirname( path.abspath(__file__) ) ) )
from pythonlib import semantic as sm
num = 0
tin = 0
wtol = sm.readwl("/home/ec2-user/git/statresult/wordslist_dsw_top1000.txt")
for root, dirs, files in os.walk('/home/ec2-user/data/topics/'):
for name in files:
filename = root + '/' + name
if name.isdigit():
fin = open(filename+'.txt','r')
temp = fin.readlines()
fin.close()
for i in temp:
i = i.strip('\n')
num = num +1
if i in wtol:
tin = tin + 1
print num,tin