import numpy as np
import re
import itertools
from collections import Counter
import cPickle as pickle
import os
def clean_str(string):
"""
Tokenization/string cleaning for all datasets except for SST.
"""
string = re.sub(r"[^A-Za-z0-9:(),!?\'\`]", " ", string)
string = re.sub(r" : ", ":", string)
string = re.sub(r"\'s", " \'s", string)
string = re.sub(r"\'ve", " \'ve", string)
string = re.sub(r"n\'t", " n\'t", string)
string = re.sub(r"\'re", " \'re", string)
string = re.sub(r"\'d", " \'d", string)
string = re.sub(r"\'ll", " \'ll", string)
string = re.sub(r",", " , ", string)
string = re.sub(r"!", " ! ", string)
string = re.sub(r"\(", " \( ", string)
string = re.sub(r"\)", " \) ", string)
string = re.sub(r"\?", " \? ", string)
string = re.sub(r"\s{2,}", " ", string)
return string.strip().lower()
def clean_str_vn(string):
"""
Tokenization/string cleaning for all datasets except for SST.
"""
string = re.sub(r"[~`@#$%^&*-+]", " ", string)
def sharp(str):
b = re.sub('\s[A-Za-z]\s\.', ' .', ' '+str)
while (b.find('. . ')>=0): b = re.sub(r'\.\s\.\s', '. ', b)
b = re.sub(r'\s\.\s', ' # ', b)
return b
string = sharp(string)
string = re.sub(r" : ", ":", string)
string = re.sub(r",", " , ", string)
string = re.sub(r"!", " ! ", string)
string = re.sub(r"\(", " \( ", string)
string = re.sub(r"\)", " \) ", string)
string = re.sub(r"\?", " \? ", string)
string = re.sub(r"\s{2,}", " ", string)
return string.strip().lower()
def load_data_and_labels(vn):
"""
Loads data from files, splits the data into words and generates labels.
Returns split sentences and labels.
"""
# Load data from files
folder_prefix = 'data' + int(vn)*'vn' + '/'
x_train = list(open(folder_prefix+"train").readlines())
x_test = list(open(folder_prefix+"test").readlines())
test_size = len(x_test)
x_text = x_train + x_test
# Split by words
if not vn:
clean_func = clean_str
else:
clean_func = clean_str_vn
x_text = [clean_func(sent) for sent in x_text]
y = [s.split(' ')[0].split(':')[0] for s in x_text]
x_text = [s.split(" ")[1:] for s in x_text]
# Generate labels
all_label = dict()
for label in y:
if not label in all_label:
all_label[label] = len(all_label) + 1
one_hot = np.identity(len(all_label))
y = [one_hot[ all_label[label]-1 ] for label in y]
return [x_text, y, test_size]
def load_trained_vecs(vn, vn_file, en_file, vocabulary):
folder_prefix = 'data' + int(vn)*'vn' + '/'
if not os.path.exists(folder_prefix + 'trained_vecs.PICKLE'):
binfile = int(vn)*vn_file + (1-int(vn))*en_file
trained_vecs = load_bin_vec(folder_prefix + binfile, vocabulary)
with open(folder_prefix + 'trained_vecs.PICKLE', 'wb') as f:
pickle.dump([trained_vecs],f,protocol=-1)
else:
with open(folder_prefix + 'trained_vecs.PICKLE', 'rb') as f:
trained_vecs = pickle.load(f)[0]
return trained_vecs
def pad_sentences(sentences, padding_word="<PAD/>"):
"""
Pads all sentences to the same length. The length is defined by the longest sentence.
Returns padded sentences.
"""
sequence_length = max(len(x) for x in sentences)
padded_sentences = []
for i in range(len(sentences)):
sentence = sentences[i]
num_padding = sequence_length - len(sentence)
new_sentence = sentence + [padding_word] * num_padding
padded_sentences.append(new_sentence)
return padded_sentences
def build_vocab(sentences):
"""
Builds a vocabulary mapping from word to index based on the sentences.
Returns vocabulary mapping and inverse vocabulary mapping.
"""
# Build vocabulary
word_counts = Counter(itertools.chain(*sentences))
# Mapping from index to word
# vocabulary_inv=['<PAD/>', 'the', ....]
vocabulary_inv = [x[0] for x in word_counts.most_common()]
# Mapping from word to index
# vocabulary = {'<PAD/>': 0, 'the': 1, ',': 2, 'a': 3, 'and': 4, ..}
vocabulary = {x: i for i, x in enumerate(vocabulary_inv)}
return [vocabulary, vocabulary_inv]
def build_input_data(sentences, labels, vocabulary):
"""
Maps sentences and labels to vectors based on a vocabulary.
"""
x = np.array([[vocabulary[word] for word in sentence] for sentence in sentences])
y = np.array(labels)
return [x, y]
def load_data(vn):
"""
Loads and preprocessed data
Returns input vectors, labels, vocabulary, and inverse vocabulary.
"""
# Load and preprocess data
sentences, labels, test_size = load_data_and_labels(vn)
sentences_padded = pad_sentences(sentences)
vocabulary, vocabulary_inv = build_vocab(sentences_padded)
x, y = build_input_data(sentences_padded, labels, vocabulary)
return [x, y, vocabulary, vocabulary_inv, test_size]
def load_bin_vec(fname, vocab):
"""
Loads 300x1 word vecs from Google (Mikolov) word2vec
"""
word_vecs = {}
with open(fname, "rb") as f:
header = f.readline()
vocab_size, layer1_size = map(int, header.split())
binary_len = np.dtype('float32').itemsize * layer1_size
for line in xrange(vocab_size):
word = []
while True:
ch = f.read(1)
if ch == ' ':
word = ''.join(word)
break
if ch != '\n':
word.append(ch)
if word in vocab:
word_vecs[word] = np.fromstring(f.read(binary_len), dtype='float32')
else:
f.read(binary_len)
return word_vecs
def add_unknown_words(word_vecs, vocab, min_df=0, k=300):
"""
0.25 is chosen so the unknown vectors have (approximately) same variance as pre-trained ones
"""
count = 0
for word in vocab:
if word not in word_vecs and vocab[word] >= min_df:
word_vecs[word] = np.random.uniform(-0.25,0.25,k)
else:
count += 1
return count
def batch_iter(data, batch_size, num_epochs):
"""
Generates a batch iterator for a dataset.
"""
data = np.array(data)
data_size = len(data)
num_batches_per_epoch = int(len(data)/batch_size) + 1
for epoch in range(num_epochs):
# Shuffle the data at each epoch
shuffle_indices = np.random.permutation(np.arange(data_size))
shuffled_data = data[shuffle_indices]
for batch_num in range(num_batches_per_epoch):
start_index = batch_num * batch_size
end_index = (batch_num + 1) * batch_size
if end_index > data_size:
end_index = data_size
start_index = end_index - batch_size
yield shuffled_data[start_index:end_index]
