def _predict_pos_tags(self, session, model, words, data_path):
'''
Define prediction function of POS Tagging
return tuples [(word, tag)]
'''
word_data = pos_reader.sentence_to_word_ids(data_path, words)
tag_data = [0] * len(word_data)
state = session.run(model.initial_state)
predict_id = []
for step, (x, y) in enumerate(
pos_reader.iterator(word_data, tag_data, model.batch_size,
model.num_steps)):
#print ("Current Step" + str(step))
fetches = [model.cost, model.final_state, model.logits]
feed_dict = {}
feed_dict[model.input_data] = x
feed_dict[model.targets] = y
for i, (c, h) in enumerate(model.initial_state):
feed_dict[c] = state[i].c
feed_dict[h] = state[i].h
_, _, logits = session.run(fetches, feed_dict)
predict_id.append(int(np.argmax(logits)))
#print (logits)
predict_tag = pos_reader.word_ids_to_sentence(data_path, predict_id)
return zip(words, predict_tag)
def run_epoch(session, model, word_data, tag_data, eval_op, verbose=False):
"""Runs the model on the given data."""
epoch_size = ((len(word_data) // model.batch_size) - 1) // model.num_steps
start_time = time.time()
costs = 0.0
iters = 0
state = session.run(model.initial_state)
for step, (x, y) in enumerate(
reader.iterator(word_data, tag_data, model.batch_size,
model.num_steps)):
fetches = [model.cost, model.final_state, eval_op]
feed_dict = {}
feed_dict[model.input_data] = x
feed_dict[model.targets] = y
for i, (c, h) in enumerate(model.initial_state):
feed_dict[c] = state[i].c
feed_dict[h] = state[i].h
cost, state, _ = session.run(fetches, feed_dict)
costs += cost
iters += model.num_steps
if verbose and step % (epoch_size // 10) == 10:
print("%.3f perplexity: %.3f speed: %.0f wps" %
(step * 1.0 / epoch_size, np.exp(costs / iters),
iters * model.batch_size / (time.time() - start_time)))
# Save Model to CheckPoint when is_training is True
if model.is_training:
if step % (epoch_size // 10) == 10:
checkpoint_path = os.path.join(FLAGS.pos_train_dir, "pos.ckpt")
model.saver.save(session, checkpoint_path)
print("Model Saved... at time step " + str(step))
return np.exp(costs / iters)
def run_epoch(session, model, word_data, tag_data, eval_op, verbose=False):
"""Runs the model on the given data."""
epoch_size = ((len(word_data) // model.batch_size) - 1) // model.num_steps
start_time = time.time()
costs = 0.0
iters = 0
correct_labels = 0 #prediction accuracy
total_labels = 0
for step, (x, y) in enumerate(
reader.iterator(word_data, tag_data, model.batch_size,
model.num_steps)):
feed_dict = {}
feed_dict[model.input_data] = x
feed_dict[model.targets] = y
fetches = []
if (model.crf_layer): # model has the CRF decoding layer
fetches = [
model.cost, model.logits, model.transition_params, eval_op
]
cost, logits, transition_params, _ = session.run(
fetches, feed_dict)
# iterate over batches [batch_size, num_steps, target_num], [batch_size, target_num]
for unary_score_, y_ in zip(
logits, y
): # unary_score_ :[num_steps, target_num], y_: [num_steps]
viterbi_prediction = tf.contrib.crf.viterbi_decode(
unary_score_, transition_params)
# viterbi_prediction: tuple (list[id], value)
# y_: tuple
correct_labels += np.sum(np.equal(
viterbi_prediction[0],
y_)) # compare prediction sequence with golden sequence
total_labels += len(y_)
#print ("step %d:" % step)
#print ("correct_labels %d" % correct_labels)
#print ("viterbi_prediction")
#print (viterbi_prediction)
else:
fetches = [model.cost, model.logits, eval_op]
cost, logits, _ = session.run(fetches, feed_dict)
costs += cost
iters += model.num_steps
if verbose and step % (epoch_size // 10) == 10:
print("%.3f perplexity: %.3f speed: %.0f wps" %
(step * 1.0 / epoch_size, np.exp(costs / iters),
iters * model.batch_size / (time.time() - start_time)))
# Accuracy
if verbose and step % (epoch_size // 10) == 10:
accuracy = 100.0 * correct_labels / float(total_labels)
print("Accuracy: %.2f%%" % accuracy)
# Save Model to CheckPoint when is_training is True
if model.is_training:
if step % (epoch_size // 10) == 10:
checkpoint_path = os.path.join(FLAGS.pos_train_dir,
"pos_bilstm_crf.ckpt")
model.saver.save(session, checkpoint_path)
print("Model Saved... at time step " + str(step))
return np.exp(costs / iters)