In this project, you'll generate your own Simpsons TV scripts using RNNs. You'll be using part of the Simpsons dataset of scripts from 27 seasons. The Neural Network you'll build will generate a new TV script for a scene at Moe's Tavern.
In [1]:
import helper
data_dir = './data/simpsons/moes_tavern_lines.txt'
text = helper.load_data(data_dir)
# Ignore notice, since we don't use it for analysing the data
text = text[81:]
In [4]:
view_sentence_range = (20, 30)
import numpy as np
print('Dataset Stats')
print('Roughly the number of unique words: {}'.format(len({word: None for word in text.split()})))
scenes = text.split('\n\n')
print('Number of scenes: {}'.format(len(scenes)))
sentence_count_scene = [scene.count('\n') for scene in scenes]
print('Average number of sentences in each scene: {}'.format(np.average(sentence_count_scene)))
sentences = [sentence for scene in scenes for sentence in scene.split('\n')]
print('Number of lines: {}'.format(len(sentences)))
word_count_sentence = [len(sentence.split()) for sentence in sentences]
print('Average number of words in each line: {}'.format(np.average(word_count_sentence)))
print()
print('The sentences {} to {}:'.format(*view_sentence_range))
print('\n'.join(text.split('\n')[view_sentence_range[0]:view_sentence_range[1]]))
In [5]:
import numpy as np
import common
import problem_unittests as tests
def create_lookup_tables(text):
"""
Create lookup tables for vocabulary
:param text: The text of tv scripts split into words
:return: A tuple of dicts (vocab_to_int, int_to_vocab)
"""
vocab_to_int, int_to_vocab = common.create_lookup_tables(text)
return vocab_to_int, int_to_vocab
tests.test_create_lookup_tables(create_lookup_tables)
We'll be splitting the script into a word array using spaces as delimiters. However, punctuations like periods and exclamation marks make it hard for the neural network to distinguish between the word "bye" and "bye!".
Implement the function token_lookup to return a dict that will be used to tokenize symbols like "!" into "||Exclamation_Mark||". Create a dictionary for the following symbols where the symbol is the key and value is the token:
This dictionary will be used to token the symbols and add the delimiter (space) around it. This separates the symbols as it's own word, making it easier for the neural network to predict on the next word. Make sure you don't use a token that could be confused as a word. Instead of using the token "dash", try using something like "||dash||".
In [6]:
def token_lookup():
"""
Generate a dict to turn punctuation into a token.
:return: Tokenize dictionary where the key is the punctuation and the value is the token
"""
x = {
'.' : '||period||',
',' : '||comma||',
'"' : '||quote||',
'(' : '||left_bracket||',
'?' : '||question_mark||',
'!' : '||exclamation||',
'\n' : '||new_line||',
')' : '||right_bracket||',
';' : '||semi-colon||' ,
'--' : '||dash_dash||'
}
return x
tests.test_tokenize(token_lookup)
In [7]:
# Preprocess Training, Validation, and Testing Data
helper.preprocess_and_save_data(data_dir, token_lookup, create_lookup_tables)
In [8]:
import helper
import numpy as np
import problem_unittests as tests
int_text, vocab_to_int, int_to_vocab, token_dict = helper.load_preprocess()
In [9]:
from distutils.version import LooseVersion
import warnings
import tensorflow as tf
# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer'
print('TensorFlow Version: {}'.format(tf.__version__))
# Check for a GPU
if not tf.test.gpu_device_name():
warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
In [42]:
def get_inputs():
"""
Create TF Placeholders for input, targets, and learning rate.
:return: Tuple (input, targets, learning rate)
"""
input_ = tf.placeholder(tf.int32, [None, None], name='input')
targets = tf.placeholder(tf.int32, [None, None], name ='targets')
learning_rate = tf.placeholder(tf.float32, name='learning_rate')
return input_, targets, learning_rate
tests.test_get_inputs(get_inputs)
In [25]:
def get_embed(input_, n_vocab, n_embedding):
"""
Create embedding for <input_data>.
:param input_: TF placeholder for text input.
:param n_vocab: Number of words in vocabulary.
:param n_embedding: Number of embedding dimensions
:return: Embedded input.
"""
embedding = tf.Variable( tf.random_uniform( (n_vocab, n_embedding)))
embed = tf.nn.embedding_lookup( embedding, input_)
return embed
tests.test_get_embed(get_embed)
In [12]:
def get_init_cell(batch_size, rnn_size):
"""
Create an RNN Cell and initialize it.
:param batch_size: Size of batches
:param rnn_size: Size of RNNs
:return: Tuple (cell, initialize state)
"""
lstms = [tf.contrib.rnn.BasicLSTMCell(rnn_size)]
cell = tf.contrib.rnn.MultiRNNCell( lstms )
initial_state = cell.zero_state(batch_size, tf.float32)
initial_state = tf.identity(initial_state, name="initial_state")
return cell, initial_state
tests.test_get_init_cell(get_init_cell)
In [35]:
def build_rnn(cell, inputs):
"""
Create a RNN using a RNN Cell
:param cell: RNN Cell
:param inputs: Input text data
:return: Tuple (Outputs, Final State)
"""
outputs, state = tf.nn.dynamic_rnn(cell, inputs, dtype=tf.float32)
state = tf.identity(state, name="final_state")
return outputs, state
tests.test_build_rnn(build_rnn)
Apply the functions you implemented above to:
input_data using your get_embed(input_data, vocab_size, embed_dim) function.cell and your build_rnn(cell, inputs) function.vocab_size as the number of outputs.Return the logits and final state in the following tuple (Logits, FinalState)
In [37]:
def build_nn(cell, rnn_size, input_data, vocab_size, embed_dim):
"""
Build part of the neural network
:param cell: RNN cell
:param rnn_size: Size of rnns
:param input_data: Input data
:param vocab_size: Vocabulary size
:param embed_dim: Number of embedding dimensions
:return: Tuple (Logits, FinalState)
"""
embed = get_embed(input_data, vocab_size, embed_dim)
outputs, state = build_rnn(cell, embed)
logits = tf.contrib.layers.fully_connected(outputs, vocab_size, activation_fn=None,
weights_initializer = tf.truncated_normal_initializer(
mean = 0.0, stddev = .1),
biases_initializer=tf.zeros_initializer() )
return logits, state
tests.test_build_nn(build_nn)
In [77]:
def get_batches(int_text, batch_size, seq_length):
"""
Return batches of input and target
:param int_text: Text with the words replaced by their ids
:param batch_size: The size of batch
:param seq_length: The length of sequence
:return: Batches as a Numpy array
"""
int_text = int_text[0:(len(int_text) - len(int_text) % (batch_size * seq_length))]
targets = np.zeros(len(int_text)).astype(int)
targets[:-1] = int_text[1:]
targets[-1] = int_text[0]
elements_per_batch = batch_size * seq_length
num_batches = int(len(int_text) / elements_per_batch)
# Build the batched data
batches = np.zeros((num_batches, 2, batch_size, seq_length)).astype(int)
for sequence in range(batch_size):
for batch in range(num_batches):
start_at = (batch * seq_length) + ( sequence * seq_length * num_batches )
# Append to inputs
batches[batch, 0, sequence, :] = int_text[start_at:(start_at + seq_length)]
# Append to targets
batches[batch, 1, sequence, :] = targets[start_at:(start_at + seq_length)]
return batches
tests.test_get_batches(get_batches)
In [91]:
num_epochs = 200
batch_size = 128
rnn_size = 256
embed_dim = 200
seq_length = 15
learning_rate = .001
show_every_n_batches = 100
save_dir = './save'
In [92]:
from tensorflow.contrib import seq2seq
train_graph = tf.Graph()
with train_graph.as_default():
vocab_size = len(int_to_vocab)
input_text, targets, lr = get_inputs()
input_data_shape = tf.shape(input_text)
cell, initial_state = get_init_cell(input_data_shape[0], rnn_size)
logits, final_state = build_nn(cell, rnn_size, input_text, vocab_size, embed_dim)
# Probabilities for generating words
probs = tf.nn.softmax(logits, name='probs')
# Loss function
cost = seq2seq.sequence_loss(
logits,
targets,
tf.ones([input_data_shape[0], input_data_shape[1]]))
# Optimizer
optimizer = tf.train.AdamOptimizer(lr)
# Gradient Clipping
gradients = optimizer.compute_gradients(cost)
capped_gradients = [(tf.clip_by_value(grad, -1., 1.), var) for grad, var in gradients if grad is not None]
train_op = optimizer.apply_gradients(capped_gradients)
Train the neural network on the preprocessed data. If you have a hard time getting a good loss, check the forms to see if anyone is having the same problem.
In [93]:
batches = get_batches(int_text, batch_size, seq_length)
with tf.Session(graph=train_graph) as sess:
sess.run(tf.global_variables_initializer())
for epoch_i in range(num_epochs):
state = sess.run(initial_state, {input_text: batches[0][0]})
for batch_i, (x, y) in enumerate(batches):
feed = {
input_text: x,
targets: y,
initial_state: state,
lr: learning_rate}
train_loss, state, _ = sess.run([cost, final_state, train_op], feed)
# Show every <show_every_n_batches> batches
if (epoch_i * len(batches) + batch_i) % show_every_n_batches == 0:
print('Epoch {:>3} Batch {:>4}/{} train_loss = {:.3f}'.format(
epoch_i,
batch_i,
len(batches),
train_loss))
# Save Model
saver = tf.train.Saver()
saver.save(sess, save_dir)
print('Model Trained and Saved')
In [94]:
# Save parameters for checkpoint
helper.save_params((seq_length, save_dir))
In [95]:
import tensorflow as tf
import numpy as np
import helper
import problem_unittests as tests
_, vocab_to_int, int_to_vocab, token_dict = helper.load_preprocess()
seq_length, load_dir = helper.load_params()
Get tensors from loaded_graph using the function get_tensor_by_name(). Get the tensors using the following names:
Return the tensors in the following tuple (InputTensor, InitialStateTensor, FinalStateTensor, ProbsTensor)
In [96]:
def get_tensors(loaded_graph):
"""
Get input, initial state, final state, and probabilities tensor from <loaded_graph>
:param loaded_graph: TensorFlow graph loaded from file
:return: Tuple (InputTensor, InitialStateTensor, FinalStateTensor, ProbsTensor)
"""
tensors = ["input:0", "initial_state:0", "final_state:0", "probs:0"]
output = []
for t in tensors:
output.append( loaded_graph.get_tensor_by_name(t) )
return output[0], output[1], output[2], output[3]
tests.test_get_tensors(get_tensors)
In [97]:
def pick_word(probabilities, int_to_vocab):
"""
Pick the next word in the generated text
:param probabilities: Probabilites of the next word
:param int_to_vocab: Dictionary of word ids as the keys and words as the values
:return: String of the predicted word
"""
return int_to_vocab[np.argmax(probabilities)]
tests.test_pick_word(pick_word)
In [98]:
gen_length = 200
# homer_simpson, moe_szyslak, or Barney_Gumble
prime_word = 'homer_simpson'
loaded_graph = tf.Graph()
with tf.Session(graph=loaded_graph) as sess:
# Load saved model
loader = tf.train.import_meta_graph(load_dir + '.meta')
loader.restore(sess, load_dir)
# Get Tensors from loaded model
input_text, initial_state, final_state, probs = get_tensors(loaded_graph)
# Sentences generation setup
gen_sentences = [prime_word + ':']
prev_state = sess.run(initial_state, {input_text: np.array([[1]])})
# Generate sentences
for n in range(gen_length):
# Dynamic Input
dyn_input = [[vocab_to_int[word] for word in gen_sentences[-seq_length:]]]
dyn_seq_length = len(dyn_input[0])
# Get Prediction
probabilities, prev_state = sess.run(
[probs, final_state],
{input_text: dyn_input, initial_state: prev_state})
pred_word = pick_word(probabilities[dyn_seq_length-1], int_to_vocab)
gen_sentences.append(pred_word)
# Remove tokens
tv_script = ' '.join(gen_sentences)
for key, token in token_dict.items():
ending = ' ' if key in ['\n', '(', '"'] else ''
tv_script = tv_script.replace(' ' + token.lower(), key)
tv_script = tv_script.replace('\n ', '\n')
tv_script = tv_script.replace('( ', '(')
print(tv_script)
It's ok if the TV script doesn't make any sense. We trained on less than a megabyte of text. In order to get good results, you'll have to use a smaller vocabulary or get more data. Luckly there's more data! As we mentioned in the begging of this project, this is a subset of another dataset. We didn't have you train on all the data, because that would take too long. However, you are free to train your neural network on all the data. After you complete the project, of course.
When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_tv_script_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.
In [ ]: