import os
import cfg
import statistics
wordSet = []
sentensesData = []
cwd = cfg.get_path()
with open(cwd + 'words', 'r') as words:
for word in words:
# print(word)
word = word.split('\t')
word[0] = word[0].replace('\ufeff', '')
word[1] = word[1].replace('\n', '')
wordSet.append(word)
with open(cwd + 'sentenses', 'r') as sentenses:
for sentense in sentenses:
sentense = sentense.replace('\n', '')
sentensesData.append(sentense)
wordSet.sort(key=lambda s: len(s[0]), reverse=True)
# print(wordSet)
newSentensesData = []
max_sentense = 0
for sentense in sentensesData:
if len(sentense) > max_sentense:
max_sentense = len(sentense)
for sentense in sentensesData:
points = 0
import os
import cfg
items = []
with open(cfg.get_path() + '131161_sentense', 'r') as sentenses_file:
for sentense in sentenses_file:
sentense = sentense.replace('\n', '')
sentense = sentense.replace('\ufeff', '')
sentense = sentense.split('-')
items.append(sentense)
# Check
print(items[0])
print(len(items))
stns_file = open(cfg.get_path() + 'sentenses', 'a')
answ_file = open(cfg.get_path() + 'answer', 'a')
comparator = ['บวก', 'ลบ']
for item in items:
ans = 2
if comparator[0] in item[1]:
ans = 1
elif comparator[1] in item[1]:
ans = 0
else:
ans = 2
answ_file.write('{}\n'.format(ans))
stns_file.write('{}\n'.format(item[0]))
import numpy as np
import os
import cfg
f = open(cfg.get_path() + 'words', 'w')
words = []
chk = []
cwd = cfg.get_path()
first = True
with open(cwd + 'normal-word.txt', 'r') as norWord:
for item in norWord:
item = item.replace('\n', '')
if not item in chk:
words.append([item, 100 * np.random.uniform(0, 0.05)])
chk.append(item)
first = True
with open(cwd + 'negative-word', 'r') as negWord:
for item in negWord:
# print(item[0:len(item)-1])
item = item.replace('\n', '')
if not item in chk:
words.append([item, (-100 * np.random.uniform(0.5, 0.01))])
chk.append(item)
first = True
with open(cwd + 'positive-sentiment-words.txt', 'r') as posWord:
for item in posWord:
item = item.replace('\n', '')
if not item in chk:
# for data in data_file:
# if '\n' in data:
# data = data[:-1]
# tmp = data.split('\t')
# for i in range(len(tmp)-1):
# tmp[i] = float(tmp[i])
# tmp[len(tmp)-1] = int(tmp[len(tmp)-1])
# dataset.append(tmp)
# print(dataset)
# ans = []
# with open('answer', 'r') as ans_file:
# for data in ans_file:
# tmp = 0
# if '\n' in data:
# tmp = data[:-1]
# ans.append(int(tmp))
# print(ans)
import json
import cfg
tmp = []
with open(cfg.get_path() + 'info', 'r') as fl:
tmp = json.load(fl)
print(tmp[0][0]['MLP-BP'][0])
print(tmp[1])