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.
The data is already provided for you. You'll be using a subset of the original dataset. It consists of only the scenes in Moe's Tavern. This doesn't include other versions of the tavern, like "Moe's Cavern", "Flaming Moe's", "Uncle Moe's Family Feed-Bag", etc..
In [1]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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 [2]:
view_sentence_range = (0, 10)
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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]]))
The first thing to do to any dataset is preprocessing. Implement the following preprocessing functions below:
To create a word embedding, you first need to transform the words to ids. In this function, create two dictionaries:
vocab_to_intint_to_vocabReturn these dictionaries in the following tuple (vocab_to_int, int_to_vocab)
In [3]:
import numpy as np
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)
"""
# TODO: Implement Function
from collections import Counter
vocab = Counter(text)
#sorted inverse to give lower weight to most repeated values
# this is to have less big values that can cause overflow or very small weights
vocab = sorted(vocab, key=vocab.get, reverse=True)
#enumerate should start at 0 to avoid issues when picking the words
vocab_to_int = { word: number for number,word in enumerate(vocab) }
int_to_vocab = { number: word for number,word in enumerate(vocab) }
#enumerate starts at 1 to avoid a word related to value 0
#vocab_to_int = { word: number for number,word in enumerate(vocab,1) }
#int_to_vocab = { number: word for number,word in enumerate(vocab,1) }
return (vocab_to_int, int_to_vocab)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
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 [4]:
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
"""
# TODO: Implement Function
token_dict = { ".":"||Period||", \
",":"||Comma||" , \
"\"":"||Quotation_mark||", \
";":"||Semicolon||", \
"!":"||Exclamation_mark||",\
"?":"||Question_mark||", \
"(":"||Left_parentheses||",\
")":"||Right_parentheses||",\
"--":"||Dash||", \
"\n":"||Return||" \
}
return token_dict
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_tokenize(token_lookup)
In [5]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# Preprocess Training, Validation, and Testing Data
helper.preprocess_and_save_data(data_dir, token_lookup, create_lookup_tables)
In [6]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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 [7]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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()))
Implement the get_inputs() function to create TF Placeholders for the Neural Network. It should create the following placeholders:
name parameter.Return the placeholders in the following tuple (Input, Targets, LearningRate)
In [8]:
def get_inputs():
"""
Create TF Placeholders for input, targets, and learning rate.
:return: Tuple (input, targets, learning rate)
"""
# TODO: Implement Function
Input = tf.placeholder(tf.int32, shape=[None,None], name="input")
Targets = tf.placeholder(tf.int32, shape=[None,None], name="targets")
Learning_rate = tf.placeholder(tf.float32, shape=None, name="learning_rate")
return (Input, Targets, Learning_rate)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_inputs(get_inputs)
Stack one or more BasicLSTMCells in a MultiRNNCell.
rnn_sizezero_state() functiontf.identity()Return the cell and initial state in the following tuple (Cell, InitialState)
In [9]:
def get_init_cell(batch_size, rnn_size, lstm_size=128):
#def get_init_cell(batch_size, rnn_size): #original declaration
"""
Create an RNN Cell and initialize it.
:param batch_size: Size of batches
:param rnn_size: Size of RNNs
:param lstm_size: Size of Basic LSTM cells (added)
:return: Tuple (cell, initialize state)
"""
# TODO: Implement Function
# Check TensorFlow version and based on that do the MultiRNNCell
if tf.__version__ == '1.1.0':
#version 1.1.0
cell = tf.contrib.rnn.MultiRNNCell([tf.contrib.rnn.BasicLSTMCell(lstm_size) for i in range(rnn_size)])
else:
# version < 1.1.0 (examples: 1.0.0, 1.0.1)
lstm = tf.contrib.rnn.BasicLSTMCell(lstm_size)
cell = tf.contrib.rnn.MultiRNNCell([lstm] * rnn_size)
initial_state = cell.zero_state(batch_size, tf.float32)
initial_state = tf.identity(initial_state, name="initial_state")
return (cell, initial_state)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_init_cell(get_init_cell)
In [10]:
def get_embed(input_data, vocab_size, embed_dim):
"""
Create embedding for <input_data>.
:param input_data: TF placeholder for text input.
:param vocab_size: Number of words in vocabulary.
:param embed_dim: Number of embedding dimensions
:return: Embedded input.
"""
# TODO: Implement Function
#embedding size (voca_size, 1st hidden layer size)
embedding = tf.Variable(tf.random_uniform((vocab_size, embed_dim),-1,1))
embed_input = tf.nn.embedding_lookup(embedding, input_data)
return embed_input
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_embed(get_embed)
You created a RNN Cell in the get_init_cell() function. Time to use the cell to create a RNN.
tf.nn.dynamic_rnn()tf.identity()Return the outputs and final_state state in the following tuple (Outputs, FinalState)
In [11]:
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)
"""
# TODO: Implement Function
output, final_state = tf.nn.dynamic_rnn(cell, inputs, dtype=tf.float32)
final_state = tf.identity(final_state,name="final_state")
return (output, final_state)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
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 [12]:
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)
"""
# TODO: Implement Function
embed = get_embed(input_data, vocab_size, embed_dim)
output, final_state = build_rnn(cell, embed)
#Linear activation in tf.contrib.layers.fully_connected(...), parameter 'activation_fn' must be set to None
logits = tf.contrib.layers.fully_connected(output[:,:], vocab_size, activation_fn=None)
return logits, final_state
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_build_nn(build_nn)
Implement get_batches to create batches of input and targets using int_text. The batches should be a Numpy array with the shape (number of batches, 2, batch size, sequence length). Each batch contains two elements:
[batch size, sequence length][batch size, sequence length]If you can't fill the last batch with enough data, drop the last batch.
For exmple, get_batches([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20], 3, 2) would return a Numpy array of the following:
[
# First Batch
[
# Batch of Input
[[ 1 2], [ 7 8], [13 14]]
# Batch of targets
[[ 2 3], [ 8 9], [14 15]]
]
# Second Batch
[
# Batch of Input
[[ 3 4], [ 9 10], [15 16]]
# Batch of targets
[[ 4 5], [10 11], [16 17]]
]
# Third Batch
[
# Batch of Input
[[ 5 6], [11 12], [17 18]]
# Batch of targets
[[ 6 7], [12 13], [18 1]]
]
]Notice that the last target value in the last batch is the first input value of the first batch. In this case, 1. This is a common technique used when creating sequence batches, although it is rather unintuitive.
In [13]:
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
"""
# TODO: Implement Function
#batch_actual_possible_size = batch_size - (batch_size%seq_length)
#max_batch_num = len(int_text)//batch_actual_possible_size
#batch_block = np.array(int_text[:(max_batch_num * batch_actual_possible_size)],dtype=np.int32)
#print (seq_length, batch_size)
#print (max_batch_num, batch_actual_possible_size)
#print(batch_block.shape)
#batch_blocks = batch_block.reshape(max_batch_num, batch_actual_possible_size)
#---------------------------
max_batch_set_num = len(int_text)//(batch_size * seq_length)
# For debug, size of all variables that determine our batches
#print (len(int_text), seq_length, batch_size, max_batch_set_num, max_batch_set_num * batch_size * seq_length)
batch_block = np.array(int_text[:(max_batch_set_num * batch_size * seq_length)],dtype=np.int32)
#print(batch_block.shape, batch_block.dtype)
# Reshape the whole sliced array that fits our seq_length and batch size for exact batches
# New shape defined by the batch size, the max_number_of_batches*seq_length =
# batch_size,number of elements in inputs(x) if it was only one batch
batch_blocks = batch_block.reshape(batch_size,max_batch_set_num*seq_length)
#print(batch_blocks.shape, batch_blocks.dtype)
batches = np.zeros((max_batch_set_num,2,batch_size,seq_length),dtype=np.int32)
# Loop for all the batches except the last one, since last one needs a special condition for
# the last value in each target (y) array to loop around and assign first element of the
# first sequence to its last element in y.
#
# EXAMPLE NOTE: if we have an input of [0...99] array with seq_length = 4 and batch_size = 5
# Meaning every x and y each has 5*4 elements, so we have arrays of size 20,
# meaning 100/20 = 5 big groups from which to do our sequences: [0-19],[20-39],[40-59],[60-79],[80-99]
# Said that, the last batch will have targets (y) of:
# [[17,18,19, --20--] , [37,38,39 --40--] , [57,58,59 --60--] , [77,78,79, --80--] , [97,98,99, --0--]]
#
# For the reasoning above the for loop stops one seq_length before
idx = 0
for n in range(0,(max_batch_set_num*seq_length)-seq_length,seq_length):
x = batch_blocks[:,n:n+seq_length]
y = np.zeros_like(x)
y[:,:-1],y[:,-1] = batch_blocks[:,n+1:n+seq_length],batch_blocks[:,n+seq_length]
batches[idx] = x,y
idx +=1
# Last batch x,y from the input
n = (max_batch_set_num*seq_length)-seq_length
x = batch_blocks[:,n:n+seq_length]
y = np.zeros_like(x)
# Special codition for the last batch of targets(y)
y[:,:-1],y[0:-1,-1],y[-1,-1] = x[:,1:],batch_blocks[1:,0],batch_blocks[0,0]
batches[idx] = x,y
# DEBUG PRINT OF BATCHES
#print_batches(batch_block, batch_size, seq_length, batches)
return batches
def print_batches(batch_block, batch_size, seq_length, batches):
print (batch_block)
print (batch_size, seq_length)
print (batches)
print ("_________________________________________________________")
# DEBUG CASES
#get_batches([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], 2, 3)
#get_batches([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20], 3, 2)
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_batches(get_batches)
Tune the following parameters:
num_epochs to the number of epochs.batch_size to the batch size.rnn_size to the size of the RNNs.embed_dim to the size of the embedding.seq_length to the length of sequence.learning_rate to the learning rate.show_every_n_batches to the number of batches the neural network should print progress.
In [14]:
# Number of Epochs
num_epochs = 100 #10
# Batch Size
batch_size = 256 #64 #128
# RNN Size
rnn_size = 2 #128
# Embedding Dimension Size
embed_dim = 512 #50 #128
# Sequence Length
seq_length = 20 #15 #4
# Learning Rate
learning_rate = 0.01 #0.001
# Show stats for every n number of batches
show_every_n_batches = 10 #2
# ADDED LSTM SIZE
lstm_size = 512
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
save_dir = './save'
In [15]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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, lstm_size)
#cell, initial_state = get_init_cell(input_data_shape[0], rnn_size) #original
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 [16]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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 [17]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# Save parameters for checkpoint
helper.save_params((seq_length, save_dir))
In [18]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
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 [19]:
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)
"""
# TODO: Implement Function
input_tensor = loaded_graph.get_tensor_by_name("input:0")
initial_state_tensor = loaded_graph.get_tensor_by_name("initial_state:0")
final_state_tensor = loaded_graph.get_tensor_by_name("final_state:0")
probabilities_tensor = loaded_graph.get_tensor_by_name("probs:0")
return input_tensor, initial_state_tensor, final_state_tensor, probabilities_tensor
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_get_tensors(get_tensors)
In [20]:
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
"""
# TODO: Implement Function
nums = np.array(list(int_to_vocab.keys()),dtype=np.int32)
next_word_id = np.random.choice(nums,p=probabilities)
#UNFORTUNATELY MY DICT WAS INDEXED STARTING AT 1, SO WE NEED TO ACCOUNT FOR THAT IN HERE
# THUS SUBSTRACT 1 TO THE INDEX OF PREDICTED WORD
#next_word_id -=1
#if next_word_id < 0:
# next_word_id = 0
next_word = int_to_vocab[next_word_id]
#list_key = list(int_to_vocab.keys())
#list_value = list(int_to_vocab.values())
#print ("+++++++++++++++++++++++")
#print (probabilities[next_word_id])
#print (next_word_id, next_word)
#print (next_word_id, int_to_vocab[nums[next_word_id-1]], nums[next_word_id])
#print (next_word_id, list_value[next_word_id-1], list_key[next_word_id])
#print ("+++++++++++++++++++++++")
return next_word
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_pick_word(pick_word)
In [21]:
gen_length = 200
# homer_simpson, moe_szyslak, or Barney_Gumble
prime_word = 'moe_szyslak'
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
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.