def main():
pos_file = './data/train_pos.txt'
neg_file = './data/train_neg.txt'
validation = './data/test_data.txt'
stopwords = './data/stopwords.txt'
vocab_file = 'vocab.dat'
inv_vocab_file = 'inv_vocab.dat'
cooc_file = 'cooc.dat'
embeddings_file = 'embeddings.dat'
label_file = 'labels.dat'
submission_file = 'submission.csv'
glove_seed = 1234
kmeans_seed = 4321
xgb_seed = 1337
sampler_seed = 7331
build_vocab([pos_file, neg_file],
stopwords,
vocab_file,
inv_vocab_file,
cutoff=5)
vocab = load_pickled(vocab_file)
inv_vocab = load_pickled(inv_vocab_file)
build_cooc([pos_file, neg_file], vocab, cooc_file)
train_glove(cooc_file, embeddings_file, glove_seed)
train_kmeans(embeddings_file, label_file, kmeans_seed)
train_xgb(vocab_file, pos_file, neg_file, label_file, validation,
submission_file, xgb_seed, sampler_seed)
def train_glove(sentences, emb_dim=50):
glove.logger.setLevel(logging.INFO)
vocab = glove.build_vocab(sentences)
cooccur = glove.build_cooccur(vocab, sentences, window_size=10)
id2word = evaluate.make_id2word(vocab)
def evaluate_word(W):
words = ['good', 'movie', 'bad', 'worth', 'dog']
for word in words:
print evaluate.most_similar(W, vocab, id2word, word)
def save_per(W,i):
if i % 100 == 0 and i >= 100:
filename = "log/glove_%d_iter%d.model" % (emb_dim, i)
W = evaluate.merge_main_context(W)
glove.save_model(W, filename)
evaluate_word(W)
W = glove.train_glove(vocab, cooccur, vector_size=emb_dim, iterations=3000, iter_callback=save_per)
def main(args):
print 80 * "="
print "INITIALIZING"
print 80 * "="
dataset = StanfordSentiment()
print "Done, read total %d windows" % dataset.word_count()
print 80 * "="
print "TRAINING"
print 80 * "="
print "Training %s word vectors" % args.model
if not os.path.exists(args.vector_path):
os.makedirs(args.vector_path)
if args.model == 'word2vec':
word_vectors = word2vec_model(args, dataset)
else:
# glove model
vocab = dataset.tokens()
word_freq = dataset.tokenfreq()
cooccur = build_cooccur(vocab, word_freq, dataset, window_size=10)
word_vectors = train_glove(vocab, cooccur, args.vector_size, args.vector_path, iterations=args.iterations)
def train_glove(sentences, emb_dim=50):
glove.logger.setLevel(logging.INFO)
vocab = glove.build_vocab(sentences)
cooccur = glove.build_cooccur(vocab, sentences, window_size=10)
id2word = evaluate.make_id2word(vocab)
def evaluate_word(W):
words = ['good', 'movie', 'bad', 'worth', 'dog']
for word in words:
print evaluate.most_similar(W, vocab, id2word, word)
def save_per(W, i):
if i % 100 == 0 and i >= 100:
filename = "log/glove_%d_iter%d.model" % (emb_dim, i)
W = evaluate.merge_main_context(W)
glove.save_model(W, filename)
evaluate_word(W)
W = glove.train_glove(vocab,
cooccur,
vector_size=emb_dim,
iterations=3000,
iter_callback=save_per)
test_corpus = ("""human interface computer
survey user computer system response time
eps user interface system
system human system eps
user response time
trees
graph trees
graph minors trees
graph minors survey
I like graph and stuff
I like trees and stuff
Sometimes I build a graph
Sometimes I build trees""").split("\n")
glove.logger.setLevel(logging.ERROR)
vocab = glove.build_vocab(test_corpus)
cooccur = glove.build_cooccur(vocab, test_corpus, window_size=10)
id2word = evaluate.make_id2word(vocab)
W = glove.train_glove(vocab, cooccur, vector_size=10, iterations=500)
# Merge and normalize word vectors
W = evaluate.merge_main_context(W)
def test_similarity():
similar = evaluate.most_similar(W, vocab, id2word, 'graph')
logging.debug(similar)
assert_equal('trees', similar[0])
vocab = glove.build_vocab(test_corpus, synonyms, antonyms)
synonyms = glove.build_syncab(synonyms, vocab)
antonyms = glove.build_antcab(antonyms, vocab)
print("Building Cooccur")
cooccur = glove.build_cooccur(vocab, test_corpus, window_size=1)
id2word = evaluate.make_id2word(vocab)
print("Training vectors...")
W = glove.train_glove(vocab,
synonyms,
antonyms,
cooccur,
vector_size=100,
iterations=25)
print("Evaluation:")
# Merge and normalize word vectors
W = evaluate.merge_main_context(W)
def test_similarity():
similar = evaluate.most_similar(W, vocab, id2word, 'trees')
print(similar)
logger.info(similar)
#assert_equal('trees', similar[0])
test_corpus = ("""human interface computer
survey user computer system response time
eps user interface system
system human system eps
user response time
trees
graph trees
graph minors trees
graph minors survey
I like graph and stuff
I like trees and stuff
Sometimes I build a graph
Sometimes I build trees""").split("\n")
glove.logger.setLevel(logging.ERROR)
vocab = glove.build_vocab(test_corpus)
cooccur = glove.build_cooccur(vocab, test_corpus, window_size=10)
id2word = evaluate.make_id2word(vocab)
W = glove.train_glove(vocab, cooccur, vector_size=10, iterations=500)
# Merge and normalize word vectors
W = evaluate.merge_main_context(W)
def test_similarity():
similar = evaluate.most_similar(W, vocab, id2word, 'graph')
logging.debug(similar)
assert_equal('trees', similar[0])
