#!/usr/bin/env python
# coding: utf-8

# Deep Learning
# =============
# 
# Assignment 5
# ------------
# 
# The goal of this assignment is to train a skip-gram model over [Text8](http://mattmahoney.net/dc/textdata) data.

# In[1]:


# These are all the modules we'll be using later. Make sure you can import them
# before proceeding further.
from __future__ import print_function
import collections
import math
import numpy as np
import os
import random
import tensorflow as tf
import zipfile
from matplotlib import pylab
from six.moves import range
from six.moves.urllib.request import urlretrieve
from sklearn.manifold import TSNE


# Download the data from the source website if necessary.

# In[2]:


url = 'http://mattmahoney.net/dc/'

def maybe_download(filename, expected_bytes):
  """Download a file if not present, and make sure it's the right size."""
  if not os.path.exists(filename):
    filename, _ = urlretrieve(url + filename, filename)
  statinfo = os.stat(filename)
  if statinfo.st_size == expected_bytes:
    print('Found and verified %s' % filename)
  else:
    print(statinfo.st_size)
    raise Exception(
      'Failed to verify ' + filename + '. Can you get to it with a browser?')
  return filename

filename = maybe_download('text8.zip', 31344016)


# Read the data into a string.

# In[3]:


def read_data(filename):
  f = zipfile.ZipFile(filename)
  for name in f.namelist():
    return tf.compat.as_str(f.read(name)).split()
  f.close()
  
words = read_data(filename)
print('Data size %d' % len(words))


# Build the dictionary and replace rare words with UNK token.

# In[4]:


vocabulary_size = 50000

def build_dataset(words):
  count = [['UNK', -1]]
  count.extend(collections.Counter(words).most_common(vocabulary_size - 1))
  dictionary = dict()
  for word, _ in count:
    dictionary[word] = len(dictionary)
  data = list()
  unk_count = 0
  for word in words:
    if word in dictionary:
      index = dictionary[word]
    else:
      index = 0  # dictionary['UNK']
      unk_count = unk_count + 1
    data.append(index)
  count[0][1] = unk_count
  reverse_dictionary = dict(zip(dictionary.values(), dictionary.keys())) 
  return data, count, dictionary, reverse_dictionary

data, count, dictionary, reverse_dictionary = build_dataset(words)
print('Most common words (+UNK)', count[:5])
print('Sample data', data[:10])
del words  # Hint to reduce memory.


# Function to generate a training batch for the skip-gram model.

# In[5]:


data_index = 0

def generate_batch(batch_size, num_skips, skip_window):
  global data_index
  assert batch_size % num_skips == 0
  assert num_skips <= 2 * skip_window
  batch = np.ndarray(shape=(batch_size), dtype=np.int32)
  labels = np.ndarray(shape=(batch_size, 1), dtype=np.int32)
  span = 2 * skip_window + 1 # [ skip_window target skip_window ]
  buffer = collections.deque(maxlen=span)
  for _ in range(span):
    buffer.append(data[data_index])
    data_index = (data_index + 1) % len(data)
  for i in range(batch_size // num_skips):
    target = skip_window  # target label at the center of the buffer
    targets_to_avoid = [ skip_window ]
    for j in range(num_skips):
      while target in targets_to_avoid:
        target = random.randint(0, span - 1)
      targets_to_avoid.append(target)
      batch[i * num_skips + j] = buffer[skip_window]
      labels[i * num_skips + j, 0] = buffer[target]
    buffer.append(data[data_index])
    data_index = (data_index + 1) % len(data)
  return batch, labels

print('data:', [reverse_dictionary[di] for di in data[:8]])

for num_skips, skip_window in [(2, 1), (4, 2)]:
    data_index = 0
    batch, labels = generate_batch(batch_size=8, num_skips=num_skips, skip_window=skip_window)
    print('\nwith num_skips = %d and skip_window = %d:' % (num_skips, skip_window))
    print('    batch:', [reverse_dictionary[bi] for bi in batch])
    print('    labels:', [reverse_dictionary[li] for li in labels.reshape(8)])


# Train a skip-gram model.

# In[6]:


batch_size = 128
embedding_size = 128 # Dimension of the embedding vector.
skip_window = 1 # How many words to consider left and right.
num_skips = 2 # How many times to reuse an input to generate a label.
# We pick a random validation set to sample nearest neighbors. here we limit the
# validation samples to the words that have a low numeric ID, which by
# construction are also the most frequent. 
valid_size = 16 # Random set of words to evaluate similarity on.
valid_window = 100 # Only pick dev samples in the head of the distribution.
valid_examples = np.array(random.sample(range(valid_window), valid_size))
num_sampled = 64 # Number of negative examples to sample.

graph = tf.Graph()

with graph.as_default(), tf.device('/cpu:0'):

  # Input data.
  train_dataset = tf.placeholder(tf.int32, shape=[batch_size])
  train_labels = tf.placeholder(tf.int32, shape=[batch_size, 1])
  valid_dataset = tf.constant(valid_examples, dtype=tf.int32)
  
  # Variables.
  embeddings = tf.Variable(
    tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0))
  softmax_weights = tf.Variable(
    tf.truncated_normal([vocabulary_size, embedding_size],
                         stddev=1.0 / math.sqrt(embedding_size)))
  softmax_biases = tf.Variable(tf.zeros([vocabulary_size]))
  
  # Model.
  # Look up embeddings for inputs.
  embed = tf.nn.embedding_lookup(embeddings, train_dataset)
  # Compute the softmax loss, using a sample of the negative labels each time.
  loss = tf.reduce_mean(
    tf.nn.sampled_softmax_loss(softmax_weights, softmax_biases, embed,
                               train_labels, num_sampled, vocabulary_size))

  # Optimizer.
  optimizer = tf.train.AdagradOptimizer(1.0).minimize(loss)
  
  # Compute the similarity between minibatch examples and all embeddings.
  # We use the cosine distance:
  norm = tf.sqrt(tf.reduce_sum(tf.square(embeddings), 1, keep_dims=True))
  normalized_embeddings = embeddings / norm
  valid_embeddings = tf.nn.embedding_lookup(
    normalized_embeddings, valid_dataset)
  similarity = tf.matmul(valid_embeddings, tf.transpose(normalized_embeddings))


# In[7]:


num_steps = 100001

with tf.Session(graph=graph) as session:
  tf.initialize_all_variables().run()
  print('Initialized')
  average_loss = 0
  for step in range(num_steps):
    batch_data, batch_labels = generate_batch(
      batch_size, num_skips, skip_window)
    feed_dict = {train_dataset : batch_data, train_labels : batch_labels}
    _, l = session.run([optimizer, loss], feed_dict=feed_dict)
    average_loss += l
    if step % 2000 == 0:
      if step > 0:
        average_loss = average_loss / 2000
      # The average loss is an estimate of the loss over the last 2000 batches.
      print('Average loss at step %d: %f' % (step, average_loss))
      average_loss = 0
    # note that this is expensive (~20% slowdown if computed every 500 steps)
    if step % 10000 == 0:
      sim = similarity.eval()
      for i in xrange(valid_size):
        valid_word = reverse_dictionary[valid_examples[i]]
        top_k = 8 # number of nearest neighbors
        nearest = (-sim[i, :]).argsort()[1:top_k+1]
        log = 'Nearest to %s:' % valid_word
        for k in xrange(top_k):
          close_word = reverse_dictionary[nearest[k]]
          log = '%s %s,' % (log, close_word)
        print(log)
  final_embeddings = normalized_embeddings.eval()


# In[8]:


num_points = 400

tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
two_d_embeddings = tsne.fit_transform(final_embeddings[1:num_points+1, :])


# In[9]:


get_ipython().run_line_magic('matplotlib', 'inline')

def plot(embeddings, labels):
  assert embeddings.shape[0] >= len(labels), 'More labels than embeddings'
  pylab.figure(figsize=(15,15))  # in inches
  for i, label in enumerate(labels):
    x, y = embeddings[i,:]
    pylab.scatter(x, y)
    pylab.annotate(label, xy=(x, y), xytext=(5, 2), textcoords='offset points',
                   ha='right', va='bottom')
  pylab.show()

words = [reverse_dictionary[i] for i in range(1, num_points+1)]
plot(two_d_embeddings, words)


# ---
# 
# Problem
# -------
# 
# An alternative to Word2Vec is called [CBOW](http://arxiv.org/abs/1301.3781) (Continuous Bag of Words). In the CBOW model, instead of predicting a context word from a word vector, you predict a word from the sum of all the word vectors in its context. Implement and evaluate a CBOW model trained on the text8 dataset.

# The main difference between skip-gram and CBOW is the inputs and outputs are reversed - skip-gram predicts the context given the word, while CBOW predicts the word given the context.  The second difference is that unlike skip-gram where each context word is a separate, label, in CBOW the context words are summed together in the input vector (so instead of having a single bit "activated" we'll have 2 or more bits turned on).  This [blog post](https://iksinc.wordpress.com/tag/continuous-bag-of-words-cbow/) explains it pretty well, although the exact implementation used here isn't as complicated as the one suggested in the post.
# 
# The code is very similar to skip-gram, with the exception of the batch generation function.
# 
# Full disclousure: these code updates came from other course participants through discussions about the assignment.  There are multiple ways to implement this so there is no single right answer.

# In[10]:


data_index = 0

def generate_batch_cbow(batch_size, skip_window):
  global data_index
  context_window = 2 * skip_window
  assert batch_size % context_window == 0
  num_labels = batch_size / context_window
  batch = np.ndarray(shape=(batch_size), dtype=np.int32)
  labels = np.ndarray(shape=(num_labels, 1), dtype=np.int32)
  span = 2 * skip_window + 1 # [ skip_window target skip_window ]
  buffer = collections.deque(maxlen=span)
  for _ in range(span):
    buffer.append(data[data_index])
    data_index = (data_index + 1) % len(data)
  for i in range(num_labels):
    target = skip_window  # target label at the center of the buffer
    labels[i, 0] = buffer[target]
    targets_to_avoid = [ skip_window ]
    for j in range(context_window):
      while target in targets_to_avoid:
        target = random.randint(0, span - 1)
      targets_to_avoid.append(target)
      batch[i * context_window + j] = buffer[target]
    buffer.append(data[data_index])
    data_index = (data_index + 1) % len(data)
  return batch, labels


# In[11]:


batch_size = 128
embedding_size = 128 # Dimension of the embedding vector.
skip_window = 1 # How many words to consider left and right.
context_window = 2 * skip_window
num_labels = batch_size / context_window
# num_skips = 2 # How many times to reuse an input to generate a label.
# We pick a random validation set to sample nearest neighbors. here we limit the
# validation samples to the words that have a low numeric ID, which by
# construction are also the most frequent. 
valid_size = 16 # Random set of words to evaluate similarity on.
valid_window = 100 # Only pick dev samples in the head of the distribution.
valid_examples = np.array(random.sample(range(valid_window), valid_size))
num_sampled = 32 # Number of negative examples to sample.

graph = tf.Graph()

with graph.as_default(), tf.device('/cpu:0'):

  # Input data.
  train_dataset = tf.placeholder(tf.int32, shape=[batch_size])
  train_labels = tf.placeholder(tf.int32, shape=[num_labels, 1])
  valid_dataset = tf.constant(valid_examples, dtype=tf.int32)
  
  # Variables.
  embeddings = tf.Variable(
    tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0))
  softmax_weights = tf.Variable(
    tf.truncated_normal([vocabulary_size, embedding_size],
                         stddev=1.0 / math.sqrt(embedding_size)))
  softmax_biases = tf.Variable(tf.zeros([vocabulary_size]))
  
  # Model.
  # Look up embeddings for inputs.
  embed = tf.nn.embedding_lookup(embeddings, train_dataset)

  # Create a mask to apply to the embed tensor to essentially "sum" the context words
  seq_ids = np.zeros(batch_size, dtype=np.int32)
  cur_id = -1
  for i in range(batch_size):
    if i % context_window == 0:
      cur_id = cur_id + 1
    seq_ids[i] = cur_id
  
  # Use segment_sum to add together the related words and reduce the output to be num_labels in size.
  embed_sum = tf.segment_sum(embed, seq_ids)
  
  # Compute the softmax loss, using a sample of the negative labels each time.
  loss = tf.reduce_mean(
    tf.nn.sampled_softmax_loss(softmax_weights, softmax_biases, embed_sum,
                               train_labels, num_sampled, vocabulary_size))

  # Optimizer.
  optimizer = tf.train.AdagradOptimizer(1.0).minimize(loss)
  
  # Compute the similarity between minibatch examples and all embeddings.
  # We use the cosine distance:
  norm = tf.sqrt(tf.reduce_sum(tf.square(embeddings), 1, keep_dims=True))
  normalized_embeddings = embeddings / norm
  valid_embeddings = tf.nn.embedding_lookup(
    normalized_embeddings, valid_dataset)
  similarity = tf.matmul(valid_embeddings, tf.transpose(normalized_embeddings))


# In[12]:


num_steps = 100001

with tf.Session(graph=graph) as session:
  tf.initialize_all_variables().run()
  print('Initialized')
  average_loss = 0
  for step in range(num_steps):
    batch_data, batch_labels = generate_batch_cbow(batch_size, skip_window)
    feed_dict = {train_dataset : batch_data, train_labels : batch_labels}
    _, l = session.run([optimizer, loss], feed_dict=feed_dict)
    average_loss += l
    if step % 2000 == 0:
      if step > 0:
        average_loss = average_loss / 2000
      # The average loss is an estimate of the loss over the last 2000 batches.
      print('Average loss at step %d: %f' % (step, average_loss))
      average_loss = 0
    # note that this is expensive (~20% slowdown if computed every 500 steps)
    if step % 10000 == 0:
      sim = similarity.eval()
      for i in xrange(valid_size):
        valid_word = reverse_dictionary[valid_examples[i]]
        top_k = 8 # number of nearest neighbors
        nearest = (-sim[i, :]).argsort()[1:top_k+1]
        log = 'Nearest to %s:' % valid_word
        for k in xrange(top_k):
          close_word = reverse_dictionary[nearest[k]]
          log = '%s %s,' % (log, close_word)
        print(log)
  final_embeddings = normalized_embeddings.eval()


# In[13]:


num_points = 400

tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
two_d_embeddings = tsne.fit_transform(final_embeddings[1:num_points+1, :])


# In[14]:


get_ipython().run_line_magic('matplotlib', 'inline')

def plot(embeddings, labels):
  assert embeddings.shape[0] >= len(labels), 'More labels than embeddings'
  pylab.figure(figsize=(15,15))  # in inches
  for i, label in enumerate(labels):
    x, y = embeddings[i,:]
    pylab.scatter(x, y)
    pylab.annotate(label, xy=(x, y), xytext=(5, 2), textcoords='offset points',
                   ha='right', va='bottom')
  pylab.show()

words = [reverse_dictionary[i] for i in range(1, num_points+1)]
plot(two_d_embeddings, words)