5_word_embeddings.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "D7tqLMoKF6uq"
   },
   "source": [
    "This notebook is based on this TensorFlow [tutorial](https://www.tensorflow.org/versions/r0.10/tutorials/word2vec/index.html#vector-representations-of-words). It's been modified to use a SummaryWriter so we can track the training process using TensorBoard. For a nice getting started with TensorBoard tutorial, check this [out](https://www.tensorflow.org/versions/r0.10/how_tos/summaries_and_tensorboard/index.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "collapsed": true,
    "id": "0K1ZyLn04QZf"
   },
   "outputs": [],
   "source": [
    "# These are all the modules we'll be using later. Make sure you can import them\n",
    "# before proceeding further.\n",
    "%matplotlib inline\n",
    "from __future__ import absolute_import\n",
    "from __future__ import division\n",
    "from __future__ import print_function\n",
    "\n",
    "import collections\n",
    "import math\n",
    "import os\n",
    "import random\n",
    "import time\n",
    "import zipfile\n",
    "\n",
    "import numpy as np\n",
    "from six.moves import urllib\n",
    "from six.moves import xrange  # pylint: disable=redefined-builtin\n",
    "import tensorflow as tf\n",
    "\n",
    "from sklearn.manifold import TSNE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "aCjPJE944bkV"
   },
   "source": [
    "Download the text corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "output_extras": [
      {
       "item_id": 1
      }
     ]
    },
    "colab_type": "code",
    "collapsed": false,
    "executionInfo": {
     "elapsed": 14640,
     "status": "ok",
     "timestamp": 1445964482948,
     "user": {
      "color": "#1FA15D",
      "displayName": "Vincent Vanhoucke",
      "isAnonymous": false,
      "isMe": true,
      "permissionId": "05076109866853157986",
      "photoUrl": "//lh6.googleusercontent.com/-cCJa7dTDcgQ/AAAAAAAAAAI/AAAAAAAACgw/r2EZ_8oYer4/s50-c-k-no/photo.jpg",
      "sessionId": "2f1ffade4c9f20de",
      "userId": "102167687554210253930"
     },
     "user_tz": 420
    },
    "id": "RJ-o3UBUFtCw",
    "outputId": "c4ec222c-80b5-4298-e635-93ca9f79c3b7"
   },
   "outputs": [],
   "source": [
    "url = 'http://mattmahoney.net/dc/'\n",
    "\n",
    "def maybe_download(filename, expected_bytes):\n",
    "  \"\"\"Download a file if not present, and make sure it's the right size.\"\"\"\n",
    "  if not os.path.exists(filename):\n",
    "    filename, _ = urllib.request.urlretrieve(url + filename, filename)\n",
    "  statinfo = os.stat(filename)\n",
    "  if statinfo.st_size == expected_bytes:\n",
    "    print('Found and verified %s' % filename)\n",
    "  else:\n",
    "    print(statinfo.st_size)\n",
    "    raise Exception(\n",
    "      'Failed to verify ' + filename + '. Can you get to it with a browser?')\n",
    "  return filename\n",
    "\n",
    "filename = maybe_download('text8.zip', 31344016)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Zqz3XiqI4mZT"
   },
   "source": [
    "Read the data into a string."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "output_extras": [
      {
       "item_id": 1
      }
     ]
    },
    "colab_type": "code",
    "collapsed": false,
    "executionInfo": {
     "elapsed": 28844,
     "status": "ok",
     "timestamp": 1445964497165,
     "user": {
      "color": "#1FA15D",
      "displayName": "Vincent Vanhoucke",
      "isAnonymous": false,
      "isMe": true,
      "permissionId": "05076109866853157986",
      "photoUrl": "//lh6.googleusercontent.com/-cCJa7dTDcgQ/AAAAAAAAAAI/AAAAAAAACgw/r2EZ_8oYer4/s50-c-k-no/photo.jpg",
      "sessionId": "2f1ffade4c9f20de",
      "userId": "102167687554210253930"
     },
     "user_tz": 420
    },
    "id": "Mvf09fjugFU_",
    "outputId": "e3a928b4-1645-4fe8-be17-fcf47de5716d"
   },
   "outputs": [],
   "source": [
    "def read_data(filename):\n",
    "  \"\"\"Extract the first file enclosed in a zip file as a list of words\"\"\"\n",
    "  with zipfile.ZipFile(filename) as f:\n",
    "    data = tf.compat.as_str(f.read(f.namelist()[0])).split()\n",
    "  return data\n",
    "  \n",
    "words = read_data(filename)\n",
    "print('Data size %d' % len(words))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Zdw6i4F8glpp"
   },
   "source": [
    "Build the dictionary and replace rare words with UNK token."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "output_extras": [
      {
       "item_id": 1
      }
     ]
    },
    "colab_type": "code",
    "collapsed": false,
    "executionInfo": {
     "elapsed": 28849,
     "status": "ok",
     "timestamp": 1445964497178,
     "user": {
      "color": "#1FA15D",
      "displayName": "Vincent Vanhoucke",
      "isAnonymous": false,
      "isMe": true,
      "permissionId": "05076109866853157986",
      "photoUrl": "//lh6.googleusercontent.com/-cCJa7dTDcgQ/AAAAAAAAAAI/AAAAAAAACgw/r2EZ_8oYer4/s50-c-k-no/photo.jpg",
      "sessionId": "2f1ffade4c9f20de",
      "userId": "102167687554210253930"
     },
     "user_tz": 420
    },
    "id": "gAL1EECXeZsD",
    "outputId": "3fb4ecd1-df67-44b6-a2dc-2291730970b2"
   },
   "outputs": [],
   "source": [
    "vocabulary_size = 50000\n",
    "\n",
    "def build_dataset(words):\n",
    "  count = [['UNK', -1]]\n",
    "  count.extend(collections.Counter(words).most_common(vocabulary_size - 1))\n",
    "  dictionary = dict()\n",
    "  for word, _ in count:\n",
    "    dictionary[word] = len(dictionary)\n",
    "  data = list()\n",
    "  unk_count = 0\n",
    "  for word in words:\n",
    "    if word in dictionary:\n",
    "      index = dictionary[word]\n",
    "    else:\n",
    "      index = 0  # dictionary['UNK']\n",
    "      unk_count = unk_count + 1\n",
    "    data.append(index)\n",
    "  count[0][1] = unk_count\n",
    "  reverse_dictionary = dict(zip(dictionary.values(), dictionary.keys())) \n",
    "  return data, count, dictionary, reverse_dictionary\n",
    "\n",
    "data, count, dictionary, reverse_dictionary = build_dataset(words)\n",
    "print('Most common words (+UNK)', count[:5])\n",
    "print('Sample data', data[:10], [reverse_dictionary[i] for i in data[:10]])\n",
    "del words  # Hint to reduce memory."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "lFwoyygOmWsL"
   },
   "source": [
    "Function to generate a training batch for the skip-gram model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "output_extras": [
      {
       "item_id": 1
      }
     ]
    },
    "colab_type": "code",
    "collapsed": false,
    "executionInfo": {
     "elapsed": 113,
     "status": "ok",
     "timestamp": 1445964901989,
     "user": {
      "color": "#1FA15D",
      "displayName": "Vincent Vanhoucke",
      "isAnonymous": false,
      "isMe": true,
      "permissionId": "05076109866853157986",
      "photoUrl": "//lh6.googleusercontent.com/-cCJa7dTDcgQ/AAAAAAAAAAI/AAAAAAAACgw/r2EZ_8oYer4/s50-c-k-no/photo.jpg",
      "sessionId": "2f1ffade4c9f20de",
      "userId": "102167687554210253930"
     },
     "user_tz": 420
    },
    "id": "w9APjA-zmfjV",
    "outputId": "67cccb02-cdaf-4e47-d489-43bcc8d57bb8"
   },
   "outputs": [],
   "source": [
    "data_index = 0\n",
    "\n",
    "def generate_batch(batch_size, num_skips, skip_window):\n",
    "  global data_index\n",
    "  assert batch_size % num_skips == 0\n",
    "  assert num_skips <= 2 * skip_window\n",
    "  batch = np.ndarray(shape=(batch_size), dtype=np.int32)\n",
    "  labels = np.ndarray(shape=(batch_size, 1), dtype=np.int32)\n",
    "  span = 2 * skip_window + 1  # [ skip_window target skip_window ]\n",
    "  buffer = collections.deque(maxlen=span)\n",
    "  for _ in range(span):\n",
    "    buffer.append(data[data_index])\n",
    "    data_index = (data_index + 1) % len(data)\n",
    "  for i in range(batch_size // num_skips):\n",
    "    target = skip_window  # target label at the center of the buffer\n",
    "    targets_to_avoid = [skip_window]\n",
    "    for j in range(num_skips):\n",
    "      while target in targets_to_avoid:\n",
    "        target = random.randint(0, span - 1)\n",
    "      targets_to_avoid.append(target)\n",
    "      batch[i * num_skips + j] = buffer[skip_window]\n",
    "      labels[i * num_skips + j, 0] = buffer[target]\n",
    "    buffer.append(data[data_index])\n",
    "    data_index = (data_index + 1) % len(data)\n",
    "  return batch, labels\n",
    "\n",
    "print('data:', [reverse_dictionary[di] for di in data[:8]])\n",
    "\n",
    "for num_skips, skip_window in [(2, 1), (4, 2)]:\n",
    "    data_index = 0\n",
    "    batch, labels = generate_batch(batch_size=8, num_skips=num_skips, skip_window=skip_window)\n",
    "    print('\\nwith num_skips = %d and skip_window = %d:' % (num_skips, skip_window))\n",
    "    print('    batch:', [reverse_dictionary[bi] for bi in batch])\n",
    "    print('    labels:', [reverse_dictionary[li] for li in labels.reshape(8)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Ofd1MbBuwiva"
   },
   "source": [
    "Build and train a skip-gram model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "collapsed": false,
    "id": "8pQKsV4Vwlzy"
   },
   "outputs": [],
   "source": [
    "batch_size = 128\n",
    "embedding_size = 128  # Dimension of the embedding vector.\n",
    "skip_window = 1       # How many words to consider left and right.\n",
    "num_skips = 2         # How many times to reuse an input to generate a label.\n",
    "\n",
    "# We pick a random validation set to sample nearest neighbors. Here we limit the\n",
    "# validation samples to the words that have a low numeric ID, which by\n",
    "# construction are also the most frequent.\n",
    "valid_size = 16     # Random set of words to evaluate similarity on.\n",
    "valid_window = 100  # Only pick dev samples in the head of the distribution.\n",
    "valid_examples = np.random.choice(valid_window, valid_size, replace=False)\n",
    "num_sampled = 64    # Number of negative examples to sample.\n",
    "\n",
    "graph = tf.Graph()\n",
    "\n",
    "with graph.as_default():\n",
    "\n",
    "  # Input data.\n",
    "  train_inputs = tf.placeholder(tf.int32, shape=[batch_size])\n",
    "  train_labels = tf.placeholder(tf.int32, shape=[batch_size, 1])\n",
    "  valid_dataset = tf.constant(valid_examples, dtype=tf.int32)\n",
    "\n",
    "  # Ops and variables pinned to the CPU because of missing GPU implementation\n",
    "  with tf.device('/cpu:0'):\n",
    "    # Look up embeddings for inputs.\n",
    "    embeddings = tf.Variable(\n",
    "        tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0))\n",
    "    embed = tf.nn.embedding_lookup(embeddings, train_inputs)\n",
    "\n",
    "    # Construct the variables for the NCE loss\n",
    "    nce_weights = tf.Variable(\n",
    "        tf.truncated_normal([vocabulary_size, embedding_size],\n",
    "                            stddev=1.0 / math.sqrt(embedding_size)))\n",
    "    nce_biases = tf.Variable(tf.zeros([vocabulary_size]))\n",
    "\n",
    "  # Compute the average NCE loss for the batch.\n",
    "  # tf.nce_loss automatically draws a new sample of the negative labels each\n",
    "  # time we evaluate the loss.\n",
    "  loss = tf.reduce_mean(\n",
    "      tf.nn.nce_loss(nce_weights, nce_biases, embed, train_labels,\n",
    "                     num_sampled, vocabulary_size))\n",
    "\n",
    "  # Construct the SGD optimizer using a learning rate of 1.0.\n",
    "  optimizer = tf.train.GradientDescentOptimizer(1.0).minimize(loss)\n",
    "\n",
    "  # Compute the cosine similarity between minibatch examples and all embeddings.\n",
    "  norm = tf.sqrt(tf.reduce_sum(tf.square(embeddings), 1, keep_dims=True))\n",
    "  normalized_embeddings = embeddings / norm\n",
    "  valid_embeddings = tf.nn.embedding_lookup(\n",
    "      normalized_embeddings, valid_dataset)\n",
    "  similarity = tf.matmul(\n",
    "      valid_embeddings, normalized_embeddings, transpose_b=True)\n",
    "\n",
    "  # Define info to be used by the SummaryWriter. This will let TensorBoard\n",
    "  # plot loss values during the training process.\n",
    "  loss_summary = tf.scalar_summary(\"loss\", loss)\n",
    "  train_summary_op = tf.merge_summary([loss_summary])\n",
    "\n",
    "  # Add variable initializer.\n",
    "  init = tf.initialize_all_variables()\n",
    "  print(\"finished building graph.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "output_extras": [
      {
       "item_id": 23
      },
      {
       "item_id": 48
      },
      {
       "item_id": 61
      }
     ]
    },
    "colab_type": "code",
    "collapsed": false,
    "executionInfo": {
     "elapsed": 436189,
     "status": "ok",
     "timestamp": 1445965429787,
     "user": {
      "color": "#1FA15D",
      "displayName": "Vincent Vanhoucke",
      "isAnonymous": false,
      "isMe": true,
      "permissionId": "05076109866853157986",
      "photoUrl": "//lh6.googleusercontent.com/-cCJa7dTDcgQ/AAAAAAAAAAI/AAAAAAAACgw/r2EZ_8oYer4/s50-c-k-no/photo.jpg",
      "sessionId": "2f1ffade4c9f20de",
      "userId": "102167687554210253930"
     },
     "user_tz": 420
    },
    "id": "1bQFGceBxrWW",
    "outputId": "5ebd6d9a-33c6-4bcd-bf6d-252b0b6055e4"
   },
   "outputs": [],
   "source": [
    "# Begin training.\n",
    "num_steps = 100001\n",
    "\n",
    "session = tf.InteractiveSession(graph=graph)\n",
    "\n",
    "# We must initialize all variables before we use them.\n",
    "init.run()\n",
    "print(\"Initialized\")\n",
    "\n",
    "# Directory in which to write summary information.\n",
    "# You can point TensorBoard to this directory via:\n",
    "# $ tensorboard --logdir=/tmp/word2vec_basic/summaries\n",
    "# Tensorflow assumes this directory already exists, so we need to create it.\n",
    "timestamp = str(int(time.time()))\n",
    "if not os.path.exists(os.path.join(\"/tmp/word2vec_basic\",\n",
    "                                   \"summaries\", timestamp)):\n",
    "  os.makedirs(os.path.join(\"/tmp/word2vec_basic\", \"summaries\", timestamp))\n",
    "# Create the SummaryWriter\n",
    "train_summary_writer = tf.train.SummaryWriter(\n",
    "    os.path.join(\n",
    "        \"/tmp/word2vec_basic\", \"summaries\", timestamp), session.graph)\n",
    "\n",
    "average_loss = 0\n",
    "for step in xrange(num_steps):\n",
    "  batch_inputs, batch_labels = generate_batch(\n",
    "      batch_size, num_skips, skip_window)\n",
    "  feed_dict = {train_inputs: batch_inputs, train_labels: batch_labels}\n",
    "\n",
    "  # We perform one update step by evaluating the optimizer op (including it\n",
    "  # in the list of returned values for session.run()\n",
    "  # Also evaluate the training summary op.\n",
    "  _, loss_val, tsummary = session.run(\n",
    "      [optimizer, loss, train_summary_op],\n",
    "      feed_dict=feed_dict)\n",
    "  average_loss += loss_val\n",
    "  # Write the evaluated summary info to the SummaryWriter. This info will\n",
    "  # then show up in the TensorBoard events.\n",
    "  train_summary_writer.add_summary(tsummary, step)\n",
    "\n",
    "  if step % 2000 == 0:\n",
    "    if step > 0:\n",
    "      average_loss /= 2000\n",
    "    # The average loss is an estimate of the loss over the last 2000 batches.\n",
    "    print(\"Average loss at step \", step, \": \", average_loss)\n",
    "    average_loss = 0\n",
    "\n",
    "  # Note that this is expensive (~20% slowdown if computed every 500 steps)\n",
    "  if step % 10000 == 0:\n",
    "    sim = similarity.eval()\n",
    "    for i in xrange(valid_size):\n",
    "      valid_word = reverse_dictionary[valid_examples[i]]\n",
    "      top_k = 8  # number of nearest neighbors\n",
    "      nearest = (-sim[i, :]).argsort()[1:top_k + 1]\n",
    "      log_str = \"Nearest to %s:\" % valid_word\n",
    "      for k in xrange(top_k):\n",
    "        close_word = reverse_dictionary[nearest[k]]\n",
    "        log_str = \"%s %s,\" % (log_str, close_word)\n",
    "      print(log_str)\n",
    "final_embeddings = normalized_embeddings.eval()\n",
    "print(\"finished training.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Start TensorBoard\n",
    "\n",
    "Start TensorBoard while the training is running (or after it's done) by pointing it to the directory in which the summaries were written. The script is configured to write them to this location.\n",
    "\n",
    "```sh\n",
    "$ tensorboard --logdir=/tmp/word2vec_basic/summaries\n",
    "```\n",
    "\n",
    "Open a browser to `http://localhost:6006/`.\n",
    "\n",
    "![TensorBoard](images/tensorboard1.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "collapsed": true,
    "id": "jjJXYA_XzV79"
   },
   "outputs": [],
   "source": [
    "# Visualize the embeddings.\n",
    "def plot_with_labels(low_dim_embs, labels, filename='tsne.png'):\n",
    "  assert low_dim_embs.shape[0] >= len(labels), \"More labels than embeddings\"\n",
    "  plt.figure(figsize=(18, 18))  # in inches\n",
    "  for i, label in enumerate(labels):\n",
    "    x, y = low_dim_embs[i, :]\n",
    "    plt.scatter(x, y)\n",
    "    plt.annotate(label,\n",
    "                 xy=(x, y),\n",
    "                 xytext=(5, 2),\n",
    "                 textcoords='offset points',\n",
    "                 ha='right',\n",
    "                 va='bottom')\n",
    "\n",
    "  plt.savefig(filename)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "both",
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "output_extras": [
      {
       "item_id": 1
      }
     ]
    },
    "colab_type": "code",
    "collapsed": false,
    "executionInfo": {
     "elapsed": 4763,
     "status": "ok",
     "timestamp": 1445965465525,
     "user": {
      "color": "#1FA15D",
      "displayName": "Vincent Vanhoucke",
      "isAnonymous": false,
      "isMe": true,
      "permissionId": "05076109866853157986",
      "photoUrl": "//lh6.googleusercontent.com/-cCJa7dTDcgQ/AAAAAAAAAAI/AAAAAAAACgw/r2EZ_8oYer4/s50-c-k-no/photo.jpg",
      "sessionId": "2f1ffade4c9f20de",
      "userId": "102167687554210253930"
     },
     "user_tz": 420
    },
    "id": "o_e0D_UezcDe",
    "outputId": "df22e4a5-e8ec-4e5e-d384-c6cf37c68c34",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "try:\n",
    "  from sklearn.manifold import TSNE\n",
    "  import matplotlib.pyplot as plt\n",
    "\n",
    "  tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)\n",
    "  plot_only = 500\n",
    "  low_dim_embs = tsne.fit_transform(final_embeddings[:plot_only, :])\n",
    "  labels = [reverse_dictionary[i] for i in xrange(plot_only)]\n",
    "  plot_with_labels(low_dim_embs, labels)\n",
    "\n",
    "except ImportError:\n",
    "  print(\"Please install sklearn and matplotlib to visualize embeddings.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## How-to find the 'nearby' words for a specific given word\n",
    "\n",
    "Here's a function to find the 'nearby' words for a specific word.\n",
    "\n",
    "E.g., picking \"six\" as the word may give a result like this (after about 100K training steps, for higher accuracy, try running for 500k):\n",
    "\n",
    "```\n",
    "\n",
    "Found word six at index 22\n",
    "Nearest to six:\n",
    "seven\n",
    "four\n",
    "eight\n",
    "five\n",
    "nine\n",
    "three\n",
    "two\n",
    "zero\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "test_word = 'six'\n",
    "test_word_idx = dictionary[test_word]\n",
    "print(\"Found word {} at index {}\".format(test_word, test_word_idx))\n",
    "\n",
    "test_embeddings = tf.nn.embedding_lookup(normalized_embeddings, [test_word_idx])\n",
    "test_similarity = tf.matmul(test_embeddings, normalized_embeddings, transpose_b=True)\n",
    "\n",
    "top_k = 8  # number of nearest neighbors\n",
    "\n",
    "# Extra: eval the 'test word' similarity\n",
    "sim = test_similarity.eval()\n",
    "nearest = (-sim[0, :]).argsort()[1:top_k + 1]\n",
    "\n",
    "print(\"Nearest to {}:\".format(test_word))\n",
    "for k in xrange(top_k):\n",
    "  close_word = reverse_dictionary[nearest[k]]\n",
    "  print (close_word)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "default_view": {},
   "name": "5_word2vec.ipynb",
   "provenance": [],
   "version": "0.3.2",
   "views": {}
  },
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}