273 lines
8.8 KiB
Python
273 lines
8.8 KiB
Python
import random
|
|
import re
|
|
from _collections import defaultdict
|
|
|
|
def define_vocabulary(file_to_learn_new_words):
|
|
word_counts = {'count': defaultdict(int)}
|
|
with open(file_to_learn_new_words, encoding='utf-8') as in_file:
|
|
for line in in_file:
|
|
text, timestamp = line.rstrip('\n').split('\t')
|
|
tokens = text.lower().split(' ')
|
|
for token in tokens:
|
|
word_counts['count'][token] += 1
|
|
in_file.close()
|
|
return word_counts
|
|
|
|
def tokenize_list(string_input):
|
|
words=[]
|
|
string=string_input.replace('\\n',' ')
|
|
text=re.sub(r'\w+:\/{2}[\d\w-]+(\.[\d\w-]+)*(?:(?:\/[^\s/]*))*', '', string)
|
|
text = re.sub(r'\\n+', " ", text)
|
|
text = re.sub(r'http\S+', " ", text)
|
|
text = re.sub(r'\/[a-z]\/', " ", text)
|
|
text = re.sub(r'[^a-z]', " ", text)
|
|
text = re.sub(r'\s{2,}', " ", text)
|
|
text = re.sub(r'\W\w{1,3}\W|\A\w{1,3}\W', " ", text)
|
|
text = re.sub(r'^\s', "", text)
|
|
string=''
|
|
for word in text:
|
|
string+=word
|
|
words=re.split(';+|,+|\*+|\n+| +|\_+|\%+|\t+|\[+|\]+|\.+|\(+|\)+|\++|\\+|\/+|[0-9]+|\#+|\'+|\"+|\-+|\=+|\&+|\:+|\?+|\!+|\^+|\·+',string)
|
|
regex=re.compile(r'http|^[a-zA-Z]$|org')
|
|
filtered_values=[
|
|
word
|
|
for word in words if not regex.match(word)
|
|
]
|
|
filtered_values[:] = (
|
|
value.lower()
|
|
for value in filtered_values if len(value)!=0
|
|
)
|
|
return filtered_values
|
|
|
|
def read_words(input_path):
|
|
vocabulary = {'count':defaultdict(int)}
|
|
index=0
|
|
with open(input_path,encoding='utf-8') as infile:
|
|
for line in infile:
|
|
index+=1
|
|
tokens = tokenize_list(line)
|
|
for token in tokens:
|
|
if token not in vocabulary:
|
|
vocabulary['vocabulary'][token]+=1
|
|
infile.close()
|
|
return vocabulary
|
|
|
|
def train(vocabulary,input_train,expected_train):
|
|
learning_rate=0.00001
|
|
learning_precision=0.00001
|
|
words_vocabulary={}
|
|
with open(input_train,encoding='utf-8') as input_file, open(expected_train,encoding='utf-8') as expected_file:
|
|
for line, exp in zip(input_file,expected_file):
|
|
words_vocabulary[line]=int(exp)
|
|
weights={}
|
|
weight={}
|
|
delta=1
|
|
iteration=0
|
|
loss_sum=0.0
|
|
error=10.0
|
|
max_iteration=len(vocabulary)
|
|
for i in vocabulary['count'].keys():
|
|
weights[i]=random.uniform(-0.01,0.01)
|
|
while delta>learning_precision and iteration<max_iteration:
|
|
d,y = random.choice(list(words_vocabulary.items()))
|
|
y_hat=0
|
|
tokens=tokenize_list(d)
|
|
for token in tokens:
|
|
if token in vocabulary['count'].keys():
|
|
y_hat += weights[token] * tokens.count(token)
|
|
delta=(y_hat-y) * learning_rate
|
|
for word in tokens:
|
|
if word in words_vocabulary:
|
|
weights[word] -= (tokens.count(word)) * delta
|
|
loss = (y_hat - y)**2.0
|
|
loss_sum += loss
|
|
if iteration%1000 == 0:
|
|
if (error>(loss_sum/1000)):
|
|
weight=weights
|
|
error=loss_sum/1000
|
|
loss_sum=0.0
|
|
iteration += 1
|
|
input_file.close()
|
|
expected_file.close()
|
|
return weight, vocabulary
|
|
|
|
def prediction(input,output,weights,vocabulary):
|
|
with open(input,encoding='utf-8') as input_file, open(output,'w+',encoding='utf-8') as output:
|
|
for line in input_file:
|
|
y_hat=0
|
|
tokens=tokenize_list(line)
|
|
for token in tokens:
|
|
if token in vocabulary['count'].keys():
|
|
y_hat += weights[token] * (token.count(token))
|
|
if y_hat>0.0:
|
|
output.write('1\n')
|
|
else:
|
|
output.write('0\n')
|
|
output.close()
|
|
input_file.close()
|
|
|
|
def main():
|
|
vocabulary=define_vocabulary('train/in.tsv');
|
|
weights, words = train(vocabulary,'train/in.tsv','train/expected.tsv')
|
|
prediction('dev-0/in.tsv','dev-0/out.tsv',weights,words)
|
|
prediction('test-A/in.tsv','test-A/out.tsv',weights,words)
|
|
|
|
main()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# from collections import defaultdict
|
|
# import math
|
|
# import pickle
|
|
# import re
|
|
#
|
|
# from pip._vendor.msgpack.fallback import xrange
|
|
# import random
|
|
#
|
|
# vocabulary = []
|
|
#
|
|
# file_to_save = open("test.tsv", "w", encoding='utf-8')
|
|
#
|
|
#
|
|
# def define_vocabulary(file_to_learn_new_words):
|
|
# word_counts = {'count': defaultdict(int)}
|
|
# with open(file_to_learn_new_words, encoding='utf-8') as in_file:
|
|
# for line in in_file:
|
|
# text, timestamp = line.rstrip('\n').split('\t')
|
|
# tokens = text.lower().split(' ')
|
|
# for token in tokens:
|
|
# word_counts['count'][token] += 1
|
|
# return word_counts
|
|
#
|
|
#
|
|
# def read_input(file_path):
|
|
# read_word_counts = {'count': defaultdict(int)}
|
|
# with open(file_path, encoding='utf-8') as in_file:
|
|
# for line in in_file:
|
|
# text, timestamp = line.rstrip('\n').split('\t')
|
|
# tokens = text.lower().split(' ')
|
|
# for token in tokens:
|
|
# read_word_counts['count'][token] += 1
|
|
# return read_word_counts
|
|
#
|
|
#
|
|
# def training(vocabulary, read_input, expected):
|
|
# file_to_write = open(expected, 'w+', encoding='utf8')
|
|
# file_to_write2 = open('out_y_hat.tsv', 'w+', encoding='utf8')
|
|
# learning_rate = 0.00001
|
|
# learning_precision = 0.0001
|
|
# weights = []
|
|
# iteration = 0
|
|
# loss_sum = 0.0
|
|
# ix = 1
|
|
# readed_words_values = []
|
|
# for word in read_input['count']:
|
|
# if word not in vocabulary['count']:
|
|
# read_input['count'][word] = 0
|
|
# readed_words_values.append(read_input['count'][word])
|
|
# for ix in range(0, len(vocabulary['count']) + 1):
|
|
# weights.append(random.uniform(-0.001, 0.001))
|
|
# # max_iteration=len(vocabulary['count'])+1
|
|
# max_iteration = 10000
|
|
# delta = 1
|
|
# while delta>learning_precision and iteration<max_iteration:
|
|
# d, y = random.choice(list(read_input['count'].items())) # d-word, y-value of
|
|
# y_hat = weights[0]
|
|
# i = 0
|
|
# for word_d in d:
|
|
# if word_d in vocabulary['count'].keys():
|
|
# # print(vocabulary['count'][d])
|
|
# y_hat += weights[vocabulary['count'][word_d]] * readed_words_values[i]
|
|
# i += 1
|
|
# print(f'Y_hat: {y_hat}')
|
|
# file_to_write2.write(f'Y_hat: {y_hat}\n')
|
|
# if y_hat > 0.5:
|
|
# file_to_write.write('1\n')
|
|
# else:
|
|
# file_to_write.write('0\n')
|
|
# i = 0
|
|
# delta = (y_hat - y) * learning_rate
|
|
# weights[0] = weights[0] - delta
|
|
# for word_w in d:
|
|
# if word_w in vocabulary['count'].keys():
|
|
# weights[vocabulary['count'][word_w]] -= readed_words_values[i] * delta
|
|
# i += 1
|
|
# # print(weights)
|
|
# #print(f'Y: {y}')
|
|
# loss = (y_hat - y) ** 2.0
|
|
# # loss=(y_hat-y)*(y_hat-y)
|
|
# loss_sum += loss
|
|
# if (iteration % 1000 == 0):
|
|
# #print(loss_sum / 1000)
|
|
# iteration = 0
|
|
# loss_sum = 0.0
|
|
# iteration += 1
|
|
# file_to_write.close
|
|
#
|
|
# def main():
|
|
# vocabulary = define_vocabulary('train/in.tsv')
|
|
# readed_words = read_input('dev-0/in.tsv')
|
|
# readed_words_test_a = read_input('test-A/in.tsv/in.tsv')
|
|
# training(vocabulary, readed_words, 'dev-0/out.tsv')
|
|
# training(vocabulary, readed_words_test_a, 'test-A/out.tsv')
|
|
#
|
|
#
|
|
# # def cost_function(y_hat,y):
|
|
# # loss=(y_hat-y)**2.0
|
|
# # loss_sum+=loss
|
|
# # if loss_counter%1000==0:
|
|
# # print(loss_sum/1000)
|
|
# # loss_counter=0
|
|
# # loss_sum=0.0
|
|
#
|
|
#
|
|
# # def main():
|
|
# # --------------- initialization ---------------------------------
|
|
# # vocabulary = define_vocabulary('train/in.tsv')
|
|
# # readed_words=read_input('dev-0/in.tsv')
|
|
# # i=1;
|
|
# # weights=[]
|
|
# # readed_words_values=[]
|
|
# # rangeVocabulary=len(vocabulary['count'])+1
|
|
# # for i in range(rangeVocabulary):
|
|
# # weights.append(random.randrange(0,len(vocabulary['count'])+1))
|
|
# # for word in readed_words['count']:
|
|
# # if word not in vocabulary['count']:
|
|
# # readed_words['count'][word]=0
|
|
# # readed_words_values.append(readed_words['count'][word])
|
|
# # precision=0.00001
|
|
# # learning_rate=0.00001
|
|
# # delta=1
|
|
# # max_iterations=len(vocabulary['count'])+1
|
|
# # current_iteration=0
|
|
# # rangeReadedValues=len(readed_words['count'])+1
|
|
# # --------------- prediction -------------------------------------
|
|
# # while (delta>precision and current_iteration<max_iterations):
|
|
# # y=random.choice(readed_words_values)
|
|
# # y_hat=weights[0]
|
|
# # i=0
|
|
# # j=0
|
|
# # for i in range(rangeReadedValues):
|
|
# # y_hat+=weights[i]*y
|
|
# # delta=abs(y_hat-y)*learning_rate
|
|
# # weights[0]=weights[0]-delta
|
|
# # for j in range(rangeVocabulary):
|
|
# # weights[j]-=y*delta
|
|
# # print(delta)
|
|
# # current_iteration+=1
|
|
#
|
|
#
|
|
# main()
|