def main():
dataloader = IMDB()
params = {
'vocab_size': len(dataloader.word2idx),
'word2idx': dataloader.word2idx,
'idx2word': dataloader.idx2word,
}
model = VRAE(params)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
print("Loading trained model ...")
tf.train.Saver().restore(sess, './saved/vrae.ckpt')
# lowercase, no punctuation, please
model.customized_reconstruct(
sess, 'i love this firm and it is beyond my expectation')
model.customized_reconstruct(
sess, 'i want to watch this movie again because it is so interesting')
model.customized_reconstruct(
sess, 'the time taken to develop the characters is quite long')
model.customized_reconstruct(
sess, 'is there any point to make a bad movie like this')
model.customized_reconstruct(
sess, 'sorry but there is no point to watch this movie again')
model.customized_reconstruct(
sess, 'to be honest this movie is not worth my time and money')
def main():
dataloader = IMDB()
model = Model(dataloader.params)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
n_batch = len(dataloader.enc_inp) // args.batch_size
if not os.path.isfile(model.model_path + '.meta'):
for epoch in range(args.stage_1_num_epochs):
dataloader.update_word_dropout()
print("Word Dropout")
dataloader.shuffle()
print("Data Shuffled")
print()
model.prior_inference(sess)
model.post_inference(sess,
'i love this film it is so good to watch')
for i, (enc_inp, dec_inp, dec_out,
_) in enumerate(dataloader.next_batch()):
log = model.train_vae_session(sess, enc_inp, dec_inp, dec_out)
if i % args.stage_1_display_step == 0:
print("Step %d | Train VAE | [%d/%d] | [%d/%d]" %
(log['step'], epoch + 1, args.stage_1_num_epochs, i,
n_batch),
end='')
print(" | nll_loss:%.1f | kl_w:%.2f | kl_loss:%.2f" %
(log['nll_loss'], log['kl_w'], log['kl_loss']))
print()
save_path = model.saver.save(sess, model.model_path)
print("Model saved in file: %s" % save_path)
else:
print("Stage 1 Completed")
def main():
dataloader = IMDB()
model = VRAE(dataloader.word2idx, dataloader.idx2word)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
for epoch in range(args.num_epoch):
dataloader.update_word_dropout()
print("\nWord Dropout")
dataloader.shuffle()
print("Data Shuffled", end='\n\n')
for i, (enc_inp, dec_inp, dec_out) in enumerate(dataloader.next_batch()):
log = model.train_session(sess, enc_inp, dec_inp, dec_out)
if i % args.display_loss_step == 0:
print("Step %d | [%d/%d] | [%d/%d]" % (log['step'], epoch+1, args.num_epoch, i, len(dataloader.enc_inp)//args.batch_size), end='')
print(" | nll_loss:%.1f | kl_w:%.3f | kl_loss:%.2f \n" % (log['nll_loss'], log['kl_w'], log['kl_loss']))
model.reconstruct(sess, enc_inp[-1], dec_inp[-1])
#model.generate(sess)
model.customized_reconstruct(sess, 'i love this film and i think it is one of the best films')
model.customized_reconstruct(sess, 'this movie is a waste of time and there is no point to watch it')
save_path = model.saver.save(sess, model.model_path)
print("Model saved in file: %s" % save_path)
def main():
dataloader = IMDB()
model = VRAE(dataloader.word2idx)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
for epoch in range(args.num_epoch):
dataloader.update_word_dropout()
dataloader.shuffle()
print("\nData Shuffled", end='\n\n')
for i, (seq, seq_dropped,
seq_len) in enumerate(dataloader.next_batch()):
log = model.train_session(sess, seq, seq_dropped, seq_len)
if i % args.display_loss_step == 0:
print("Step %d | [%d/%d] | [%d/%d]" %
(log['step'], epoch + 1, args.num_epoch, i,
len(dataloader._X) // args.batch_size),
end='')
"""
if args.mutinfo_loss:
print("\n\tnll_loss:%.1f | kl_w:%.3f | kl_loss:%.2f | temper:%.2f | mutinfo_loss:%.2f \n" % (
log['nll_loss'], log['kl_w'], log['kl_loss'], log['temperature'], log['mutinfo_loss']))
else:
"""
print(" | nll_loss:%.1f | kl_w:%.3f | kl_loss:%.2f \n" %
(log['nll_loss'], log['kl_w'], log['kl_loss']))
model.reconstruct(sess, seq[-1], seq_dropped[-1])
model.generate(sess)
save_path = model.saver.save(sess, model.model_path)
print("Model saved in file: %s" % save_path)
def main():
dataloader = IMDB()
model = Model(dataloader.params)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
n_batch = len(dataloader.enc_inp) // args.batch_size
print("Loading trained model ...")
model.saver.restore(sess, model.model_path)
for epoch in range(args.stage_2_num_epochs):
dataloader.update_word_dropout()
print("Word Dropout")
dataloader.shuffle()
print("Data Shuffled")
print()
for i, (enc_inp, dec_inp, dec_out, labels) in enumerate(dataloader.next_batch()):
log = model.train_discriminator_session(sess, enc_inp, dec_inp, dec_out, labels)
if i % args.stage_2_display_step == 0:
print("------------")
print("Step %d | Train Discriminator | [%d/%d] | [%d/%d]" % (
log['step'], epoch+1, args.stage_2_num_epochs, i, n_batch))
print("\t| clf_loss:%.2f | clf_acc:%.2f | L_u: %.2f" % (
log['clf_loss'], log['clf_acc'], log['L_u']))
print()
log = model.train_encoder_session(sess, enc_inp, dec_inp, dec_out)
if i % args.stage_2_display_step == 0:
print("Step %d | Train Encoder | [%d/%d] | [%d/%d]" % (
log['step'], epoch+1, args.stage_2_num_epochs, i, n_batch))
print("\t| seq_loss:%.1f | kl_w:%.2f | kl_loss:%.2f" % (
log['nll_loss'], log['kl_w'], log['kl_loss']))
print()
log = model.train_generator_session(sess, enc_inp, dec_inp, dec_out)
if i % args.stage_2_display_step == 0:
print("Step %d | Train Generator | [%d/%d] | [%d/%d]" % (
log['step'], epoch+1, args.stage_2_num_epochs, i, n_batch))
print("\t| seq_loss:%.1f | kl_w:%.2f | kl_loss:%.2f" % (
log['nll_loss'], log['kl_w'], log['kl_loss']))
print("\t| temperature:%.2f | l_attr_z:%.2f | l_attr_c:%.2f" % (
log['temperature'], log['l_attr_z'], log['l_attr_c']))
print("------------")
if i % (5 * args.stage_2_display_step) == 0:
model.post_inference(sess, 'i love this film it is so good to watch')
model.post_inference(sess, 'this movie is horrible and waste my time')
save_path = model.saver.save(sess, './saved_temp/model.ckpt')
print("Model saved in file: %s" % save_path)
def main():
dataloader = IMDB()
model = Model(dataloader.params)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
print("Loading trained model ...")
model.saver.restore(sess, './saved_temp/model.ckpt')
# lowercase, no punctuation, please
model.post_inference(sess, "i love this film it is one of the best")
model.post_inference(sess, "this film is awful and the acting is bad")
model.post_inference(
sess, "i hate this boring movie and there is no point to watch")
def main():
dataloader = IMDB()
model = VRAE(dataloader.word2idx)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
print("Loading trained model ...")
model.saver.restore(sess, model.model_path)
# lowercase, no punctuation, please
model.customized_reconstruct(sess,
'i love this firm it is one of the best')
model.customized_reconstruct(
sess, 'i want to see this movie it seems interesting')
def main():
dataloader = IMDB()
params = {
'vocab_size': len(dataloader.word2idx),
'word2idx': dataloader.word2idx,
'idx2word': dataloader.idx2word,
}
print('Vocab Size:', params['vocab_size'])
model = VRAE(params)
saver = tf.train.Saver()
config = tf.ConfigProto(allow_soft_placement=True)
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
sess.run(tf.global_variables_initializer())
for epoch in range(args.num_epoch):
dataloader.update_word_dropout()
print("\nWord Dropout")
dataloader.shuffle()
print("Data Shuffled", end='\n\n')
for i, (enc_inp, dec_inp,
dec_out) in enumerate(dataloader.next_batch()):
log = model.train_session(sess, enc_inp, dec_inp, dec_out)
if i % args.display_loss_step == 0:
print("Step %d | [%d/%d] | [%d/%d]" %
(log['step'], epoch + 1, args.num_epoch, i,
len(dataloader.enc_inp) // args.batch_size),
end='')
print(" | nll_loss:%.1f | kl_w:%.3f | kl_loss:%.2f \n" %
(log['nll_loss'], log['kl_w'], log['kl_loss']))
model.generate(sess)
model.reconstruct(sess, enc_inp[-1], dec_inp[-1])
model.customized_reconstruct(
sess, 'i love this film and i think it is one of the best films')
model.customized_reconstruct(
sess,
'this movie is a waste of time and there is no point to watch it')
save_path = saver.save(sess, './saved/vrae.ckpt')
print("Model saved in file: %s" % save_path)
def main():
dataloader = IMDB()
model = VRAE(dataloader.word2idx)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
for epoch in range(args.num_epoch):
dataloader.update_word_dropout()
dataloader.shuffle()
print("\nData Shuffled", end='\n\n')
for i, (seq, seq_dropped, seq_len) in enumerate(dataloader.next_batch()):
log = model.train_session(sess, seq, seq_dropped, seq_len)
if i % args.display_loss_step == 0:
bar = '[%d/%d] | [%d/%d] | nll_loss: %.1f | kl_w: %.3f | kl_loss: %.1f'
vars = (epoch+1, args.num_epoch, i+1, len(dataloader._X)//args.batch_size, log['nll_loss'],
log['kl_w'], log['kl_loss'])
print(bar % vars)
if i % args.display_text_step == 0:
model.reconstruct(sess, seq[-1], seq_dropped[-1])
model.generate(sess)
# CONSTANT
VOCAB_SIZE = 5000
MAX_SEQ_LEN = 100
SEED = 1234
gpu = tf.config.experimental.list_physical_devices('GPU')
tf.config.experimental.set_memory_growth(gpu[0], True) # gpu name: /GPU:0
model = RNN(MAX_SEQ_LEN, VOCAB_SIZE)
model.summary()
model.compile(loss='binary_crossentropy',
optimizer=RMSprop(),
metrics=['accuracy'])
imdb = IMDB(MAX_SEQ_LEN, VOCAB_SIZE)
if __name__ == '__main__':
# people can find their devices using tf.config.list_physical_devices("GPU") # "CPU"
with tf.device("/GPU:0"):
# Model Training
model.fit(
imdb.X_train,
imdb.Y_train,
batch_size=512,
epochs=10,
validation_split=0.2,
callbacks=[EarlyStopping(patience=2, verbose=1)],
)
# Run model on test set