def run_task(data_dir, task_id):
"""
Train and test for each task
"""
print("Train and test for task %d ..." % task_id)
# Parse data
train_files = glob.glob('%s/qa%d_*_train.txt' % (data_dir, task_id))
test_files = glob.glob('%s/qa%d_*_test.txt' % (data_dir, task_id))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_files, dictionary, False)
test_story, test_questions, test_qstory = parse_babi_task(
test_files, dictionary, False)
general_config = BabiConfig(train_story, train_questions, dictionary)
memory, model, loss = build_model(general_config)
if general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory, memory,
model, loss, general_config)
else:
train(train_story, train_questions, train_qstory, memory, model, loss,
general_config)
test(test_story, test_questions, test_qstory, memory, model, loss,
general_config)
def train(self):
"""
Train MemN2N model using training data for tasks.
"""
np.random.seed(42) # for reproducing
assert self.data_dir is not None, "data_dir is not specified."
print("Reading data from %s ..." % self.data_dir)
# Parse training data
train_data_path = glob.glob('%s/qa*_*_train.txt' % self.data_dir)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(train_data_path, dictionary, False)
# Parse test data just to expand the dictionary so that it covers all words in the test data too
test_data_path = glob.glob('%s/qa*_*_test.txt' % self.data_dir)
parse_babi_task(test_data_path, dictionary, False)
# Get reversed dictionary mapping index to word
self.reversed_dict = dict((ix, w) for w, ix in dictionary.items())
# Construct model
self.general_config = BabiConfigJoint(train_story, train_questions, dictionary)
self.memory, self.model, self.loss = build_model(self.general_config)
# Train model
if self.general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory,
self.memory, self.model, self.loss, self.general_config)
else:
train(train_story, train_questions, train_qstory,
self.memory, self.model, self.loss, self.general_config)
# Save model
self.save_model()
def train(self):
"""
Train MemN2N model using training data for tasks.
"""
np.random.seed(42) # for reproducing
assert self.data_dir is not None, "data_dir is not specified."
print("Reading data from %s ..." % self.data_dir)
# Parse training data
train_data_path = glob.glob('%s/qa*_*_train.txt' % self.data_dir)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(train_data_path, dictionary, False)
# Parse test data just to expand the dictionary so that it covers all words in the test data too
test_data_path = glob.glob('%s/qa*_*_test.txt' % self.data_dir)
parse_babi_task(test_data_path, dictionary, False)
# Get reversed dictionary mapping index to word
self.reversed_dict = dict((ix, w) for w, ix in dictionary.items())
# Construct model
self.general_config = BabiConfigJoint(train_story, train_questions, dictionary)
self.memory, self.model, self.loss = build_model(self.general_config)
# Train model
if self.general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory,
self.memory, self.model, self.loss, self.general_config)
else:
train(train_story, train_questions, train_qstory,
self.memory, self.model, self.loss, self.general_config)
# Save model
self.save_model()
def run_task(data_dir, task_id):
"""
Train and test for each task
"""
print("Train and test for task %d ..." % task_id)
print("We are going to use this")
# Parse data
train_files = glob.glob('%s/qa3_*_train.txt' % (data_dir, task_id))
test_files = glob.glob('%s/qa3_*_test.txt' % (data_dir, task_id))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_files, dictionary, False)
test_story, test_questions, test_qstory = parse_babi_task(
test_files, dictionary, False)
general_config = BabiConfig(train_story, train_questions, dictionary)
# #### R: this line build a empty model to train
# memory, model, loss = build_model(general_config)
# if general_config.linear_start:
# train_linear_start(train_story, train_questions, train_qstory, memory, model, loss, general_config)
# else:
# train(train_story, train_questions, train_qstory, memory, model, loss, general_config)
# memory, model, loss = build_model(general_config)
# this line
test(test_story, test_questions, test_qstory, memory, model, loss,
general_config)
def train(self):
"""
Train MemN2N model using training data for tasks.
"""
np.random.seed(120393) # for reproducing
train_data_arg = None
test_data_arg = None
if type(self.data_dir) is tuple:
assert self.data_dir[
0] is not None, "training data_dir is not specified."
assert self.data_dir[
1] is not None, "test data_dir is not specified."
print("Reading training data from %s ..." % self.data_dir[0])
print("Reading test data from %s ..." % self.data_dir[1])
train_data_arg = '%s/qa*_train.txt' % self.data_dir[0]
test_data_arg = '%s/qa*_valid.txt' % self.data_dir[1]
else:
assert self.data_dir is not None, "data_dir is not specified."
print("Reading data from %s ..." % self.data_dir)
train_data_arg = '%s/qa*_*_train.txt' % self.data_dir
test_data_arg = '%s/qa*_*_test.txt' % self.data_dir
assert train_data_arg is not None and test_data_arg is not None
# Parse training data
train_data_path = glob.glob(train_data_arg)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = \
parse_babi_task(train_data_path, dictionary, False, dt=DataType.PYTORCH)
# Parse test data just to expand the dictionary so that it covers all words in the test data too
test_data_path = glob.glob(test_data_arg)
parse_babi_task(test_data_path, dictionary, False)
# Get reversed dictionary mapping index to word
self.reversed_dict = dict((ix, w) for w, ix in dictionary.items())
# Construct model
self.general_config = BabiConfigJointPytorch(train_story,
train_questions,
dictionary)
# self.memory, self.model, self.loss = build_model(self.general_config)
self.memory, self.model, self.loss = build_model_pytorch(
self.general_config)
self.init_weights()
# Train model
if self.general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory,
self.memory, self.model, self.loss,
self.general_config)
else:
train(train_story, train_questions, train_qstory, self.memory,
self.model, self.loss, self.general_config)
# Save model
self.save_model()
def run_test(data_dir, task_id, memn2n):
print("Test for task %d ..." % task_id)
test_files = None
if type(data_dir) is tuple:
test_files = glob.glob('%s/qa%d_*_valid.txt' % (data_dir[1], task_id))
else:
test_files = glob.glob('%s/qa%d_*_test.txt' % (data_dir, task_id))
test_story, test_questions, test_qstory = parse_babi_task(
test_files,
memn2n.general_config.dictionary,
False,
dt=DataType.PYTORCH)
"""
reversed_dict = None
memory = None
model = None
loss = None
general_config = None
with gzip.open(model_file, "rb") as f:
self.reversed_dict, self.memory, self.model, self.loss, self.general_config = pickle.load(f)
"""
test(test_story, test_questions, test_qstory, memn2n.memory, memn2n.model,
memn2n.loss, memn2n.general_config)
def run_joint_tasks(data_dir):
"""
Train and test for all tasks but the trained model is built using training data from all tasks.
"""
print("Jointly train and test for all tasks ...")
tasks = range(20)
# Parse training data
train_data_path = []
for t in tasks:
train_data_path += glob.glob('%s/qa%d_*_train.txt' % (data_dir, t + 1))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_data_path, dictionary, False, dt=DataType.PYTORCH)
# Parse test data for each task so that the dictionary covers all words before training
for t in tasks:
test_data_path = glob.glob('%s/qa%d_*_test.txt' % (data_dir, t + 1))
parse_babi_task(test_data_path, dictionary,
False) # ignore output for now
general_config = BabiConfigJointPytorch(train_story, train_questions,
dictionary)
memory, model, loss = build_model_pytorch(general_config)
if general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory, memory,
model, loss, general_config)
else:
train(train_story, train_questions, train_qstory, memory, model, loss,
general_config)
# Test on each task
for t in tasks:
print("Testing for task %d ..." % (t + 1))
test_data_path = glob.glob('%s/qa%d_*_test.txt' % (data_dir, t + 1))
dc = len(dictionary)
test_story, test_questions, test_qstory = parse_babi_task(
test_data_path, dictionary, False)
assert dc == len(
dictionary
) # make sure that the dictionary already covers all words
test(test_story, test_questions, test_qstory, memory, model, loss,
general_config)
def run_console_demo(data_dir, model_file):
"""
Console-based demo
"""
memn2n = MemN2N(data_dir, model_file)
# Try to load model
memn2n.load_model()
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob("%s/qa*_*_test.txt" % memn2n.data_dir)
test_story, test_questions, test_qstory = parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
while True:
# Pick a random question
question_idx = np.random.randint(test_questions.shape[1])
story_idx = test_questions[0, question_idx]
last_sentence_idx = test_questions[1, question_idx]
# Get story and question
story_txt, question_txt, correct_answer = memn2n.get_story_texts(
test_story, test_questions, test_qstory, question_idx, story_idx, last_sentence_idx
)
print("* Story:")
print("\n\t".join(story_txt))
print("\n* Suggested question:\n\t%s?" % question_txt)
while True:
user_question = raw_input("Your question (press Enter to use the suggested question):\n\t")
pred_answer_idx, pred_prob, memory_probs = memn2n.predict_answer(
test_story, test_questions, test_qstory, question_idx, story_idx, last_sentence_idx, user_question
)
pred_answer = memn2n.reversed_dict[pred_answer_idx]
print("* Answer: '%s', confidence score = %.2f%%" % (pred_answer, 100.0 * pred_prob))
if user_question == "":
if pred_answer == correct_answer:
print(" Correct!")
else:
print(" Wrong. The correct answer is '%s'" % correct_answer)
print("\n* Explanation:")
print("\t".join(["Memory %d" % (i + 1) for i in range(len(memory_probs))]) + "\tText")
for sent_idx, sent_txt in enumerate(story_txt):
prob_output = "\t".join(["%.3f" % mem_prob for mem_prob in memory_probs[:, sent_idx]])
print("%s\t%s" % (prob_output, sent_txt))
asking_another_question = raw_input("\nDo you want to ask another question? [y/N] ")
if asking_another_question == "" or asking_another_question.lower() == "n":
break
will_continue = raw_input("Do you want to continue? [Y/n] ")
if will_continue != "" and will_continue.lower() != "y":
break
print("=" * 70)
def run_console_demo(data_dir, model_file):
"""
Console-based demo
"""
memn2n = MemN2N(data_dir, model_file)
# Try to load model
memn2n.load_model()
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
while True:
# Pick a random question
question_idx = np.random.randint(test_questions.shape[1])
story_idx = test_questions[0, question_idx]
last_sentence_idx = test_questions[1, question_idx]
# Get story and question
story_txt, question_txt, correct_answer = memn2n.get_story_texts(test_story, test_questions, test_qstory,
question_idx, story_idx, last_sentence_idx)
print("* Story:")
print("\n\t".join(story_txt))
print("\n* Suggested question:\n\t%s?" % question_txt)
while True:
user_question = raw_input("Your question (press Enter to use the suggested question):\n\t")
pred_answer_idx, pred_prob, memory_probs = \
memn2n.predict_answer(test_story, test_questions, test_qstory,
question_idx, story_idx, last_sentence_idx,
user_question)
pred_answer = memn2n.reversed_dict[pred_answer_idx]
print("* Answer: '%s', confidence score = %.2f%%" % (pred_answer, 100. * pred_prob))
if user_question == '':
if pred_answer == correct_answer:
print(" Correct!")
else:
print(" Wrong. The correct answer is '%s'" % correct_answer)
print("\n* Explanation:")
print("\t".join(["Memory %d" % (i + 1) for i in range(len(memory_probs))]) + "\tText")
for sent_idx, sent_txt in enumerate(story_txt):
prob_output = "\t".join(["%.3f" % mem_prob for mem_prob in memory_probs[:, sent_idx]])
print("%s\t%s" % (prob_output, sent_txt))
asking_another_question = raw_input("\nDo you want to ask another question? [y/N] ")
if asking_another_question == '' or asking_another_question.lower() == 'n': break
will_continue = raw_input("Do you want to continue? [Y/n] ")
if will_continue != '' and will_continue.lower() != 'y': break
print("=" * 70)
def save_answers_to_file(data_dir, model_file):
"""
Console-based demo
"""
memn2n = MemN2N(data_dir, model_file)
# Try to load model
memn2n.load_model()
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa8_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
story_dict = construct_story_dict(test_data_path)
curr_story_idx, question_idx_in_story = -1, 1
f = open('setqa_answers7_2.txt', 'w')
print >> f, 'textID,sortedAnswerList'
for question_idx in xrange(test_questions.shape[1]):
# Pick a random question
story_idx = test_questions[0, question_idx]
last_sentence_idx = test_questions[1, question_idx]
# Get story and question
story_txt, question_txt, correct_answer = memn2n.get_story_texts(
test_story, test_questions, test_qstory, question_idx, story_idx,
last_sentence_idx)
pred_answer_idx, pred_prob, memory_probs = \
memn2n.predict_answer(test_story, test_questions, test_qstory,
question_idx, story_idx, last_sentence_idx)
pred_answer = memn2n.reversed_dict[pred_answer_idx]
if story_idx != curr_story_idx:
curr_story_idx = story_idx
question_idx_in_story = 0
question_idx_in_story += 1
pred_answer_idx_in_story = [
story_dict[story_idx][x] for x in pred_answer.split(",")
]
pred_answer_idx_in_story = [
str(x) for x in sorted(pred_answer_idx_in_story)
]
print >> f, '{}_{},{}'.format(story_idx + 1, question_idx_in_story,
" ".join(pred_answer_idx_in_story))
# print '{}_{},{}'.format(story_idx+1, question_idx_in_story, " ".join(pred_answer_idx_in_story))
f.close()
def init(data_dir, model_file):
""" Initialize web app """
global memn2n, test_story, test_questions, test_qstory
# Try to load model
memn2n = MemN2N(data_dir, model_file)
memn2n.load_model()
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
def test_model(data_dir, model_file, log_path, rnd_seed):
memn2n = MemN2N(data_dir, model_file, log_path, rnd_seed)
memn2n.load_model()
#_, _, memn2n.loss = build_model(memn2n.general_config)
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path,
memn2n.general_config.dictionary,
False)
test(test_story, test_questions, test_qstory, memn2n.memory, memn2n.model,
memn2n.loss, memn2n.general_config)
def init(data_dir, model_file):
""" Initialize web app """
global memn2n, test_story, test_questions, test_qstory
# Try to load model
memn2n = MemN2N(data_dir, model_file)
memn2n.load_model()
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
def run_task(data_dir, task_id):
train_files = glob.glob('%s/qa%d_*_train.txt' % (data_dir, task_id))
test_files = glob.glob('%s/qa%d_*_test.txt' % (data_dir, task_id))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_files, dictionary, False)
test_story, test_questions, test_qstory = parse_babi_task(
test_files, dictionary, False)
FLAGS.dictionary = dictionary
with tf.Session() as sess:
model = MemN2N_QA_Basic(FLAGS, sess,
(train_story, train_questions, train_qstory),
(test_story, test_questions, test_qstory))
# only qa_model in this repository yet
model.qa_model(task_id)
model.optimization()
model.saver = tf.train.Saver()
ckpt = tf.train.get_checkpoint_state(FLAGS.checkpoint)
if ckpt and ckpt.model_checkpoint_path:
model.saver.restore(model.sess, ckpt.model_checkpoint_path)
else:
print(" [!] Not found checkpoint")
sess.run(tf.global_variables_initializer())
if FLAGS.is_test:
model.test()
else:
model.train()
tf.reset_default_graph()
sess.close()
def run_joint_tasks(data_dir):
"""
Train and test for all tasks but the trained model is built using training data from all tasks.
"""
print("Jointly train and test for all tasks ...")
tasks = range(20)
# Parse training data
train_data_path = []
for t in tasks:
train_data_path += glob.glob('%s/qa%d_*_train.txt' % (data_dir, t + 1))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(train_data_path, dictionary, False)
# Parse test data for each task so that the dictionary covers all words before training
for t in tasks:
test_data_path = glob.glob('%s/qa%d_*_test.txt' % (data_dir, t + 1))
parse_babi_task(test_data_path, dictionary, False) # ignore output for now
general_config = BabiConfigJoint(train_story, train_questions, dictionary)
memory, model, loss = build_model(general_config)
if general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory, memory, model, loss, general_config)
else:
train(train_story, train_questions, train_qstory, memory, model, loss, general_config)
# Test on each task
for t in tasks:
print("Testing for task %d ..." % (t + 1))
test_data_path = glob.glob('%s/qa%d_*_test.txt' % (data_dir, t + 1))
dc = len(dictionary)
test_story, test_questions, test_qstory = parse_babi_task(test_data_path, dictionary, False)
assert dc == len(dictionary) # make sure that the dictionary already covers all words
test(test_story, test_questions, test_qstory, memory, model, loss, general_config)
def run_task(data_dir, task_id):
"""
Train and test for each task
"""
print("Train and test for task %d ..." % task_id)
# Parse data
train_files = glob.glob('%s/qa%d_*_train.txt' % (data_dir, task_id))
test_files = glob.glob('%s/qa%d_*_test.txt' % (data_dir, task_id))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(train_files, dictionary, False)
test_story, test_questions, test_qstory = parse_babi_task(test_files, dictionary, False)
general_config = BabiConfig(train_story, train_questions, dictionary)
memory, model, loss = build_model(general_config)
if general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory, memory, model, loss, general_config)
else:
train(train_story, train_questions, train_qstory, memory, model, loss, general_config)
test(test_story, test_questions, test_qstory, memory, model, loss, general_config)
def run_tableQA(data_path, model_file):
"""
Train and test for table QA
"""
# Parse data
train_files = glob.glob(data_path.format('train'))
test_files = glob.glob(data_path.format('test'))
# SV: init dict with pre-trained vectors, e.g. from fastText
# dictionary = fasttext.load_model(EMBEDDINGS_MODEL_PATH)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_files, dictionary, False)
test_story, test_questions, test_qstory = parse_babi_task(
test_files, dictionary, False)
# print test_questions
print 'Dictionary:', len(dictionary)
general_config = BabiConfig(train_story, train_questions, dictionary)
memory, model, loss = build_model(general_config)
if general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory, memory,
model, loss, general_config)
else:
train(train_story, train_questions, train_qstory, memory, model, loss,
general_config)
test(test_story, test_questions, test_qstory, memory, model, loss,
general_config)
# save_model
with gzip.open(model_file, "wb") as f:
print("Saving model to file %s ..." % model_file)
reversed_dict = dict((ix, w) for w, ix in dictionary.items())
pickle.dump((reversed_dict, memory, model, loss, general_config), f)
def test_model(data_dir, model_file, log_path, rnd_seed):
memn2n = MemN2N(data_dir, model_file, log_path, rnd_seed)
memn2n.load_model()
#_, _, memn2n.loss = build_model(memn2n.general_config)
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path,
memn2n.general_config.dictionary,
False)
test(test_story,
test_questions,
test_qstory,
memn2n.memory,
memn2n.model,
memn2n.loss,
memn2n.general_config)
def predict_d():
word_idx = pickle.load(open(model_dir + 'word_idx.pkl', 'rb'))
import glob
import numpy as np
from demo.qa import MemN2N
from util import parse_babi_task
""" Initialize app """
global memn2n, test_story, test_questions, test_qstory
model_file = 'trained_model/memn2n_model.pklz'
# Try to load model
memn2n = MemN2N(data_dir, model_file)
memn2n.load_model()
test_story, test_questions, test_qstory = None, None, None
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
question_idx = np.random.randint(test_questions.shape[1])
story_idx = test_questions[0, question_idx]
last_sentence_idx = test_questions[1, question_idx]
story_txt, question_txt, correct_answer = memn2n.get_story_texts(
test_story, test_questions, test_qstory, question_idx, story_idx,
last_sentence_idx)
# Format text
story_txt = "\n".join(story_txt)
question_txt += "?"
print('Case study:')
print('Story:', story_txt, '\nQuestion:', question_txt, '\nAnswer:',
correct_answer)
model = tf.keras.models.load_model(model_dir)
predict_answer = model.predict([
vectorize(story_txt, word_idx, 552),
vectorize(question_txt, word_idx, 5)
])
print(list(word_idx.keys())[np.argmax(predict_answer, axis=1)[0] - 1])
def run_task(data_dir, task_id):
"""
Train and test for each task
"""
print("Train and test for task %d ..." % task_id)
# Parse data
train_files = glob.glob('%s/qa%d_*_train.txt' % (data_dir, task_id))
test_files = glob.glob('%s/qa%d_*_test.txt' % (data_dir, task_id))
#train_files = glob.glob('%s/qa%d_*train.txt' % (data_dir, task_id))
#test_files = glob.glob('%s/qa%d_*test.txt' % (data_dir, task_id))
# #### empty dictionary
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(train_files, dictionary, False)
test_story, test_questions, test_qstory = parse_babi_task(test_files, dictionary, False)
general_config = BabiConfig(train_story, train_questions, dictionary)
memory, model, loss = build_model(general_config)
if general_config.linear_start:
train_linear_start(train_story, train_questions, train_qstory, memory, model, loss, general_config)
else:
train(train_story, train_questions, train_qstory, memory, model, loss, general_config)
with open('R_trained.txt', 'a') as outfile:
json.dump(general_config.dictionary, outfile, indent=2)
print("######## trained dictionary")
print(general_config.dictionary)
ans_index = test(test_story, test_questions, test_qstory, memory, model, loss, general_config)
####R this line load model
memn2n = MemN2N(args.data_dir, args.model_file)
#Try to load model
memn2n.load_model()
dictionary2 = {"nil": 0}
train_story2, train_questions2, train_qstory2 = parse_babi_task(train_files, memn2n.general_config.dictionary, False)
test_story2, test_questions2, test_qstory2 = parse_babi_task(test_files, memn2n.general_config.dictionary, False)
#print(len(test_questions2))
#general_config2 = BabiConfig(train_story2, train_questions2,memn2n.general_config.dictionary)
with open('R_loaded.txt', 'a') as outfile2:
json.dump(memn2n.general_config.dictionary, outfile2, indent=2)
print("???????? loaded dictionary")
print(memn2n.general_config.dictionary)
ans_index = test(test_story2, test_questions2, test_qstory2, memn2n.memory, memn2n.model, memn2n.loss, memn2n.general_config)
def train(self):
"""
Train MemN2N model using training data for tasks.
"""
#np.random.seed(42) # for reproducing
np.random.seed(self.rnd_seed) # for reproducing
print("np.random.seed: %d" % self.rnd_seed)
assert self.data_dir is not None, "data_dir is not specified."
print("Reading data from %s ..." % self.data_dir)
# Parse training data
train_data_path = glob.glob('%s/qa*_*_train.txt' % self.data_dir)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(train_data_path, dictionary, False)
# Parse test data just to expand the dictionary so that it covers all words in the test data too
test_data_path = glob.glob('%s/qa*_*_test.txt' % self.data_dir)
parse_babi_task(test_data_path, dictionary, False)
# Get reversed dictionary mapping index to word
self.reversed_dict = dict((ix, w) for w, ix in dictionary.items())
# Construct model
#self.general_config = Babi10kConfigJoint(train_story, train_questions, dictionary)
self.general_config = BabiConfigJoint(train_story, train_questions, dictionary)
self.memory, self.model, self.loss = build_model(self.general_config)
# Train model
if self.general_config.linear_start:
print('We will use LS training')
self.best_model, self.best_memory = \
train_linear_start(train_story,
train_questions,
train_qstory,
self.memory,
self.model,
self.loss,
self.general_config,
self.log_path)
else:
train_logger = open(os.path.join(self.log_path, 'train.log'), 'w')
train_logger.write('epoch batch_iter lr loss err\n')
train_logger.flush()
val_logger = open(os.path.join(self.log_path, 'val.log'), 'w')
val_logger.write('epoch batch_iter lr loss err\n')
val_logger.flush()
global_batch_iter = 0
train_logger, val_logger, self.best_model, self.best_memory, _ = \
train(train_story,
train_questions,
train_qstory,
self.memory,
self.model,
self.loss,
self.general_config,
train_logger,
val_logger,
global_batch_iter)
train_logger.close()
val_logger.close()
# Save model
self.save_model()
def train(self):
"""
Train MemN2N model using training data for tasks.
"""
#np.random.seed(42) # for reproducing
np.random.seed(self.rnd_seed) # for reproducing
print("np.random.seed: %d" % self.rnd_seed)
assert self.data_dir is not None, "data_dir is not specified."
print("Reading data from %s ..." % self.data_dir)
# Parse training data
train_data_path = glob.glob('%s/qa*_*_train.txt' % self.data_dir)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_data_path, dictionary, False)
# Parse test data just to expand the dictionary so that it covers all words in the test data too
test_data_path = glob.glob('%s/qa*_*_test.txt' % self.data_dir)
parse_babi_task(test_data_path, dictionary, False)
# Get reversed dictionary mapping index to word
self.reversed_dict = dict((ix, w) for w, ix in dictionary.items())
# Construct model
#self.general_config = Babi10kConfigJoint(train_story, train_questions, dictionary)
self.general_config = BabiConfigJoint(train_story, train_questions,
dictionary)
self.memory, self.model, self.loss = build_model(self.general_config)
# Train model
if self.general_config.linear_start:
print('We will use LS training')
self.best_model, self.best_memory = \
train_linear_start(train_story,
train_questions,
train_qstory,
self.memory,
self.model,
self.loss,
self.general_config,
self.log_path)
else:
train_logger = open(os.path.join(self.log_path, 'train.log'), 'w')
train_logger.write('epoch batch_iter lr loss err\n')
train_logger.flush()
val_logger = open(os.path.join(self.log_path, 'val.log'), 'w')
val_logger.write('epoch batch_iter lr loss err\n')
val_logger.flush()
global_batch_iter = 0
train_logger, val_logger, self.best_model, self.best_memory, _ = \
train(train_story,
train_questions,
train_qstory,
self.memory,
self.model,
self.loss,
self.general_config,
train_logger,
val_logger,
global_batch_iter)
train_logger.close()
val_logger.close()
# Save model
self.save_model()
test_story, test_questions, test_qstory = None, None, None
if __name__ == '__main__':
# Create the Window
window = sg.Window('babi数据集算法演示', layout)
# Event Loop to process "events" and get the "values" of the inputs
# Try to load model
memn2n = MemN2N(data_dir, model_file)
memn2n.load_model()
# Read test data
print("Reading test data from %s ..." % memn2n.data_dir)
test_data_path = glob.glob('%s/qa*_*_test.txt' % memn2n.data_dir)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path, memn2n.general_config.dictionary, False)
while True:
event, values = window.read()
if event == sg.WIN_CLOSED:
break
if event == 'getstory':
question_idx = np.random.randint(test_questions.shape[1])
story_idx = test_questions[0, question_idx]
last_sentence_idx = test_questions[1, question_idx]
story_txt, question_txt, correct_answer = memn2n.get_story_texts(
test_story, test_questions, test_qstory, question_idx,
story_idx, last_sentence_idx)
# Format text
def run_task(data_dir, task_id, model_file, log_path):
"""
Train and test for each task
"""
print("Train and test for task %d ..." % task_id)
train_files = glob.glob('%s/qa%d_*_train.txt' % (data_dir, task_id))
test_files = glob.glob('%s/qa%d_*_test.txt' % (data_dir, task_id))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = \
parse_babi_task(train_files, dictionary, False)
test_story, test_questions, test_qstory = \
parse_babi_task(test_files, dictionary, False)
# Get reversed dictionary mapping index to word
# NOTE: this needed to real-time testing
reversed_dict = dict((ix, w) for w, ix in dictionary.items())
general_config = BabiConfig(train_story, train_questions, dictionary)
memory, model, loss_func = build_model(general_config)
if general_config.linear_start:
print('We will use LS training')
best_model, best_memory = \
train_linear_start(train_story,
train_questions,
train_qstory,
memory,
model,
loss_func,
general_config,
self.log_path)
else:
train_logger = open(os.path.join(self.log_path, 'train.log'), 'w')
train_logger.write('epoch batch_iter lr loss err\n')
train_logger.flush()
val_logger = open(os.path.join(self.log_path, 'val.log'), 'w')
val_logger.write('epoch batch_iter lr loss err\n')
val_logger.flush()
global_batch_iter = 0
train_logger, val_logger, _, _, _ = \
train(train_story,
train_questions,
train_qstory,
memory,
model,
loss_func,
general_config,
train_logger,
val_logger,
global_batch_iter)
train_logger.close()
val_logger.close()
model_file = os.path.join(log_path, model_file)
with gzip.open(model_file, 'wb') as f:
print('Saving model to file %s ...' % model_file)
pickle.dump((reversed_dict, memory, model, loss_func, general_config),
f)
print('Start to testing')
test(test_story, test_questions, test_qstory, memory, model, loss_func,
general_config)
def train_and_test(self, seed=None):
"""
Train MemN2N model using training data for tasks.
"""
if seed is None:
np.random.seed(42) # for reproducing
else:
np.random.seed(seed)
train_data_arg = None
test_data_arg = None
if type(self.data_dir) is tuple:
assert self.data_dir[
0] is not None, "training data_dir is not specified."
assert self.data_dir[
1] is not None, "test data_dir is not specified."
print("Reading training data from %s ..." % self.data_dir[0])
print("Reading test data from %s ..." % self.data_dir[1])
train_data_arg = '%s/qa*_train.txt' % self.data_dir[0]
test_data_arg = '%s/qa*_valid.txt' % self.data_dir[1]
else:
assert self.data_dir is not None, "data_dir is not specified."
print("Reading data from %s ..." % self.data_dir)
train_data_arg = '%s/qa*_*_train.txt' % self.data_dir
test_data_arg = '%s/qa*_*_test.txt' % self.data_dir
assert train_data_arg is not None and test_data_arg is not None
# Parse training data
train_data_path = glob.glob(train_data_arg)
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = parse_babi_task(
train_data_path, dictionary, False)
# Parse test data just to expand the dictionary so that it covers all words in the test data too
test_data_path = glob.glob(test_data_arg)
test_story, test_questions, test_qstory = parse_babi_task(
test_data_path, dictionary, False)
# Get reversed dictionary mapping index to word
self.reversed_dict = dict((ix, w) for w, ix in dictionary.items())
# Construct model
self.general_config = BabiConfigJoint(train_story, train_questions,
dictionary)
#check for config switches format [initial learning rate, linear start option, hops]
if self.config_switches is not None:
self.general_config.train_config[
"init_lrate"] = self.config_switches[0]
#linear start option is passed to babi config constructor function so no need to set here
# want equal of number of epochs for linear start and non linear start runs
if self.general_config.linear_start is True:
self.general_config.nepochs = 40
self.general_config.ls_nepochs = 20
else:
self.general_config.nepochs = 60
self.general_config.nhops = self.config_switches[2]
self.memory, self.model, self.loss = build_model(self.general_config)
# Train model
train_val_results = []
if self.general_config.linear_start:
train_val_results += train_linear_start(train_story,
train_questions,
train_qstory, self.memory,
self.model, self.loss,
self.general_config)
else:
train_val_results += train(train_story, train_questions,
train_qstory, self.memory, self.model,
self.loss, self.general_config)
test_error = test(test_story, test_questions, test_qstory, self.memory,
self.model, self.loss, self.general_config)
model_test_accuracy = (1.0 - test_error) * 100.0
train_val_file = self.model_file + 'train_val_accuracy.csv'
with open(train_val_file, 'w') as f:
f.write('epoch, TrainAccuracy, ValAccuracy\n')
epoch = 1
for item in train_val_results:
line = '{}, {:.3f}, {:.3f}\n'.format(epoch, item[0], item[1])
f.write(line)
epoch += 1
self.model_file += '_TestAcc{:.1f}percent_.pickle'.format(
model_test_accuracy)
# Save model
self.save_model()
def run_joint_tasks(data_dir, model_file, log_path):
"""
Train and test for all tasks but the trained model is built using training data from all tasks.
"""
print("Jointly train and test for all tasks ...")
tasks = range(20)
# Parse training data
train_data_path = []
for t in tasks:
train_data_path += glob.glob('%s/qa%d_*_train.txt' % (data_dir, t + 1))
dictionary = {"nil": 0}
train_story, train_questions, train_qstory = \
parse_babi_task(train_data_path,
dictionary,
False)
# Parse test data for each task so that the dictionary covers all words before training
for t in tasks:
test_data_path = glob.glob('%s/qa%d_*_test.txt' % (data_dir, t + 1))
parse_babi_task(test_data_path, dictionary,
False) # ignore output for now
# Get reversed dictionary mapping index to word
# NOTE: this needed to real-time testing
reversed_dict = dict((ix, w) for w, ix in dictionary.items())
general_config = BabiConfigJoint(train_story, train_questions, dictionary)
memory, model, loss_func = build_model(general_config)
if general_config.linear_start:
print('We will use LS training')
train_linear_start(train_story, train_questions, train_qstory, memory,
model, loss_func, general_config, log_path)
else:
train_logger = open(os.path.join(log_file, 'train.log'), 'w')
train_logger.write('epoch batch_iter lr loss err\n')
train_logger.flush()
val_logger = open(os.path.join(log_file, 'val.log'), 'w')
val_logger.write('epoch batch_iter lr loss err\n')
val_logger.flush()
train_logger, val_logger, best_model, best_memory = \
train(train_story,
train_questions,
train_qstory,
memory,
model,
loss_func,
general_config,
train_logger,
val_logger)
train_logger.close()
val_logger.close()
model_file = os.path.join(log_path, model_file)
with gzip.open(model_file, 'wb') as f:
print('Saving model to file %s ...' % model_file)
pickle.dump((reversed_dict, memory, model, loss_func, general_config),
f)
# Test on each task
print('Start to testing')
for t in tasks:
print("Testing for task %d ..." % (t + 1))
test_data_path = glob.glob('%s/qa%d_*_test.txt' % (data_dir, t + 1))
dc = len(dictionary)
test_story, test_questions, test_qstory = \
parse_babi_task(test_data_path,
dictionary,
False)
assert dc == len(
dictionary
) # make sure that the dictionary already covers all words
test(test_story, test_questions, test_qstory, memory, model, loss_func,
general_config)
