Word2VecFinal.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "#from nltk.tokenize import word_tokenize \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Vocab size:  22\n",
      "Words: {'old', 'of', 'english', 'top', 'manchester', 'professional', 'football', 'is', 'a', 'tier', 'league', 'in', 'that', 'competes', 'the', 'united', 'liverpool', 'trafford', 'based', 'england', 'club', 'premier'}\n"
     ]
    }
   ],
   "source": [
    "corpus_raw = 'Manchester United Football Club is a professional football club based in Old Trafford England. Liverpool Football Club is a professional football club in Liverpool, England, that competes in the Premier League, the top tier of English football.'\n",
    "# convert to lower case\n",
    "corpus_raw = corpus_raw.lower()\n",
    "corpus_raw = corpus_raw.replace(\",\", \"\")\n",
    "words = []\n",
    "for word in corpus_raw.split():\n",
    "    if word != '.': # because we don't want to treat . as a word\n",
    "        words.append(word.replace(\".\", \"\"))\n",
    "words = set(words) # so that all duplicate words are removed\n",
    "\n",
    "# words = word_tokenize(corpus_raw)         \n",
    "word2int = {}\n",
    "int2word = {}\n",
    "vocab_size = len(words) # gives the total number of unique words\n",
    "print('Vocab size: ', vocab_size)\n",
    "print('Words:', words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "index of \"manchester\" is : 19\n",
      "word in 1 index is : of\n"
     ]
    }
   ],
   "source": [
    "for i,word in enumerate(words):\n",
    "    word2int[word] = i\n",
    "    int2word[i] = word\n",
    "print('index of \"manchester\" is :',word2int['england'])\n",
    "print('word in 1 index is :', int2word[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of sentenses are: 3\n",
      "\n",
      " Original sentence::  manchester united football club is a professional football club based in old trafford england. liverpool football club is a professional football club in liverpool england that competes in the premier league the top tier of english football.\n",
      "\n",
      " ************************Input is getting prepared ***********************\n",
      "0 : ['manchester', 'united']\n",
      "1 : ['manchester', 'football']\n",
      "2 : ['united', 'manchester']\n",
      "3 : ['united', 'football']\n",
      "4 : ['united', 'club']\n",
      "5 : ['football', 'manchester']\n",
      "6 : ['football', 'united']\n",
      "7 : ['football', 'club']\n",
      "8 : ['football', 'is']\n",
      "9 : ['club', 'united']\n",
      "10 : ['club', 'football']\n",
      "11 : ['club', 'is']\n",
      "12 : ['club', 'a']\n",
      "13 : ['is', 'football']\n",
      "14 : ['is', 'club']\n",
      "15 : ['is', 'a']\n",
      "16 : ['is', 'professional']\n",
      "17 : ['a', 'club']\n",
      "18 : ['a', 'is']\n",
      "19 : ['a', 'professional']\n",
      "20 : ['a', 'football']\n",
      "21 : ['professional', 'is']\n",
      "22 : ['professional', 'a']\n",
      "23 : ['professional', 'football']\n",
      "24 : ['professional', 'club']\n",
      "25 : ['football', 'a']\n",
      "26 : ['football', 'professional']\n",
      "27 : ['football', 'club']\n",
      "28 : ['football', 'based']\n",
      "29 : ['club', 'professional']\n",
      "30 : ['club', 'football']\n",
      "31 : ['club', 'based']\n",
      "32 : ['club', 'in']\n",
      "33 : ['based', 'football']\n",
      "34 : ['based', 'club']\n",
      "35 : ['based', 'in']\n",
      "36 : ['based', 'old']\n",
      "37 : ['in', 'club']\n",
      "38 : ['in', 'based']\n",
      "39 : ['in', 'old']\n",
      "40 : ['in', 'trafford']\n",
      "41 : ['old', 'based']\n",
      "42 : ['old', 'in']\n",
      "43 : ['old', 'trafford']\n",
      "44 : ['old', 'england']\n",
      "45 : ['trafford', 'in']\n",
      "46 : ['trafford', 'old']\n",
      "47 : ['trafford', 'england']\n",
      "48 : ['england', 'old']\n",
      "49 : ['england', 'trafford']\n",
      "50 : ['liverpool', 'football']\n",
      "51 : ['liverpool', 'club']\n",
      "52 : ['football', 'liverpool']\n",
      "53 : ['football', 'club']\n",
      "54 : ['football', 'is']\n",
      "55 : ['club', 'liverpool']\n",
      "56 : ['club', 'football']\n",
      "57 : ['club', 'is']\n",
      "58 : ['club', 'a']\n",
      "59 : ['is', 'football']\n",
      "60 : ['is', 'club']\n",
      "61 : ['is', 'a']\n",
      "62 : ['is', 'professional']\n",
      "63 : ['a', 'club']\n",
      "64 : ['a', 'is']\n",
      "65 : ['a', 'professional']\n",
      "66 : ['a', 'football']\n",
      "67 : ['professional', 'is']\n",
      "68 : ['professional', 'a']\n",
      "69 : ['professional', 'football']\n",
      "70 : ['professional', 'club']\n",
      "71 : ['football', 'a']\n",
      "72 : ['football', 'professional']\n",
      "73 : ['football', 'club']\n",
      "74 : ['football', 'in']\n",
      "75 : ['club', 'professional']\n",
      "76 : ['club', 'football']\n",
      "77 : ['club', 'in']\n",
      "78 : ['club', 'liverpool']\n",
      "79 : ['in', 'football']\n",
      "80 : ['in', 'club']\n",
      "81 : ['in', 'liverpool']\n",
      "82 : ['in', 'england']\n",
      "83 : ['liverpool', 'club']\n",
      "84 : ['liverpool', 'in']\n",
      "85 : ['liverpool', 'england']\n",
      "86 : ['liverpool', 'that']\n",
      "87 : ['england', 'in']\n",
      "88 : ['england', 'liverpool']\n",
      "89 : ['england', 'that']\n",
      "90 : ['england', 'competes']\n",
      "91 : ['that', 'liverpool']\n",
      "92 : ['that', 'england']\n",
      "93 : ['that', 'competes']\n",
      "94 : ['that', 'in']\n",
      "95 : ['competes', 'england']\n",
      "96 : ['competes', 'that']\n",
      "97 : ['competes', 'in']\n",
      "98 : ['competes', 'the']\n",
      "99 : ['in', 'that']\n",
      "100 : ['in', 'competes']\n",
      "101 : ['in', 'the']\n",
      "102 : ['in', 'premier']\n",
      "103 : ['the', 'competes']\n",
      "104 : ['the', 'in']\n",
      "105 : ['the', 'premier']\n",
      "106 : ['the', 'league']\n",
      "107 : ['premier', 'in']\n",
      "108 : ['premier', 'the']\n",
      "109 : ['premier', 'league']\n",
      "110 : ['premier', 'the']\n",
      "111 : ['league', 'the']\n",
      "112 : ['league', 'premier']\n",
      "113 : ['league', 'the']\n",
      "114 : ['league', 'top']\n",
      "115 : ['the', 'premier']\n",
      "116 : ['the', 'league']\n",
      "117 : ['the', 'top']\n",
      "118 : ['the', 'tier']\n",
      "119 : ['top', 'league']\n",
      "120 : ['top', 'the']\n",
      "121 : ['top', 'tier']\n",
      "122 : ['top', 'of']\n",
      "123 : ['tier', 'the']\n",
      "124 : ['tier', 'top']\n",
      "125 : ['tier', 'of']\n",
      "126 : ['tier', 'english']\n",
      "127 : ['of', 'top']\n",
      "128 : ['of', 'tier']\n",
      "129 : ['of', 'english']\n",
      "130 : ['of', 'football']\n",
      "131 : ['english', 'tier']\n",
      "132 : ['english', 'of']\n",
      "133 : ['english', 'football']\n",
      "134 : ['football', 'of']\n",
      "135 : ['football', 'english']\n"
     ]
    }
   ],
   "source": [
    "# raw sentences is a list of sentences.\n",
    "raw_sentences = corpus_raw.split('.')\n",
    "sentences = []\n",
    "\n",
    "print( 'Total number of sentenses are:', len(raw_sentences))\n",
    "for i, sentence in enumerate(raw_sentences):\n",
    "        sentences.append(sentence.split())\n",
    "#        print(i , ':', sentence.split())\n",
    "\n",
    "#using fixed window size, pairs of word are created for input\n",
    "WINDOW_SIZE = 2\n",
    "\n",
    "data = []\n",
    "for sentence in sentences:\n",
    "    for word_index, word in enumerate(sentence):\n",
    "        for nb_word in sentence[max(word_index - WINDOW_SIZE, 0) : min(word_index + WINDOW_SIZE, len(sentence)) + 1] : \n",
    "            if nb_word != word:\n",
    "                data.append([word, nb_word])\n",
    "print('\\n Original sentence:: ', corpus_raw)\n",
    "print('\\n ************************Input is getting prepared ***********************')  \n",
    "for i, eachpair in enumerate(data):\n",
    "    print(i, ':', eachpair)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'manchester'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-269-cb01960ac99e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata_word\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 11\u001b[0;31m     \u001b[0mx_train\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mto_one_hot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mword2int\u001b[0m\u001b[0;34m[\u001b[0m \u001b[0mdata_word\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvocab_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     12\u001b[0m     \u001b[0my_train\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mto_one_hot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mword2int\u001b[0m\u001b[0;34m[\u001b[0m \u001b[0mdata_word\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvocab_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'manchester'"
     ]
    }
   ],
   "source": [
    "# #function to convert numbers to one hot vectors\n",
    "# def to_one_hot(data_point_index, vocab_size):\n",
    "#     temp = np.zeros(vocab_size)\n",
    "#     temp[data_point_index] = 1\n",
    "#     return temp\n",
    "\n",
    "# x_train = [] # input word\n",
    "# y_train = [] # output word\n",
    "\n",
    "# for i, data_word in enumerate(data):\n",
    "#     x_train.append(to_one_hot(word2int[ data_word[0] ], vocab_size))\n",
    "#     y_train.append(to_one_hot(word2int[ data_word[1] ], vocab_size))\n",
    "\n",
    "# # convert them to numpy arrays\n",
    "# x_train = np.asarray(x_train)\n",
    "# y_train = np.asarray(y_train)\n",
    "# print('words: ', words)\n",
    "# print('X : \\n', x_train)\n",
    "# print('\\nY: \\n', y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "words:  {'old', 'of', 'english', 'top', 'manchester', 'professional', 'football', 'is', 'a', 'tier', 'league', 'in', 'that', 'competes', 'the', 'united', 'liverpool', 'trafford', 'based', 'england', 'club', 'premier'}\n",
      "X : \n",
      " [ 4  4 15 15 15  6  6  6  6 20 20 20 20  7  7  7  7  8  8  8  8  5  5  5\n",
      "  5  6  6  6  6 20 20 20 20 18 18 18 18 11 11 11 11  0  0  0  0 17 17 17\n",
      " 19 19 16 16  6  6  6 20 20 20 20  7  7  7  7  8  8  8  8  5  5  5  5  6\n",
      "  6  6  6 20 20 20 20 11 11 11 11 16 16 16 16 19 19 19 19 12 12 12 12 13\n",
      " 13 13 13 11 11 11 11 14 14 14 14 21 21 21 21 10 10 10 10 14 14 14 14  3\n",
      "  3  3  3  9  9  9  9  1  1  1  1  2  2  2  6  6]\n",
      "\n",
      "Y: \n",
      " [15  6  4  6 20  4 15 20  7 15  6  7  8  6 20  8  5 20  7  5  6  7  8  6\n",
      " 20  8  5 20 18  5  6 18 11  6 20 11  0 20 18  0 17 18 11 17 19 11  0 19\n",
      "  0 17  6 20 16 20  7 16  6  7  8  6 20  8  5 20  7  5  6  7  8  6 20  8\n",
      "  5 20 11  5  6 11 16  6 20 16 19 20 11 19 12 11 16 12 13 16 19 13 11 19\n",
      " 12 11 14 12 13 14 21 13 11 21 10 11 14 10 14 14 21 14  3 21 10  3  9 10\n",
      " 14  9  1 14  3  1  2  3  9  2  6  9  1  6  1  2]\n"
     ]
    }
   ],
   "source": [
    "# function to convert numbers to one hot vectors\n",
    "def to_one_hot(data_point_index, vocab_size):\n",
    "    temp = np.zeros(vocab_size)\n",
    "    temp[data_point_index] = 1\n",
    "    return temp\n",
    "\n",
    "x_train = [] # input word\n",
    "y_train = [] # output word\n",
    "labelss = []\n",
    "for i, data_word in enumerate(data):\n",
    "    \n",
    "    x_train.append((word2int[ data_word[0] ]))\n",
    "    if data_word[0] not in labelss:\n",
    "        labelss.append(data_word[0] )\n",
    "    y_train.append((word2int[ data_word[1] ]))\n",
    "\n",
    "# convert them to numpy arrays\n",
    "x_train = np.asarray(x_train)\n",
    "y_train = np.asarray(y_train)\n",
    "print('words: ', words)\n",
    "print('X : \\n', x_train)\n",
    "print('\\nY: \\n', y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {},
   "outputs": [],
   "source": [
    "#labelss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = y_train.reshape((-1, 1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {},
   "outputs": [],
   "source": [
    "#x_train, y_train\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batches (x, y) ([6, 8, 6], [[11], [5], [20]])\n"
     ]
    }
   ],
   "source": [
    "def get_batch(size):\n",
    "    assert size<len(data)\n",
    "    X=[]\n",
    "    Y=[]\n",
    "    rdm = np.random.choice(range(len(data)), size, replace=False)\n",
    "    \n",
    "    for r in rdm:\n",
    "        X.append(word2int[data[r][0]])\n",
    "        Y.append([word2int[data[r][1]]])\n",
    "    return X, Y\n",
    "\n",
    "print('Batches (x, y)', get_batch(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [],
   "source": [
    "BATCH_SIZE = 10\n",
    "VOCAB_SIZE = vocab_size #12\n",
    "EMBED_SIZE = 5\n",
    "NUM_SAMPLED= 6\n",
    "LEARNING_RATE =1.0 # 1e-1\n",
    "X = tf.placeholder(tf.int32, shape=[BATCH_SIZE])\n",
    "Y = tf.placeholder(tf.int32, shape=[BATCH_SIZE, 1])\n",
    "\n",
    "with tf.device(\"/cpu:0\"):\n",
    "    embed_matrix = tf.Variable(tf.random_uniform([VOCAB_SIZE, EMBED_SIZE], -1.0, 1.0)) #12,5 \n",
    "    embed = tf.nn.embedding_lookup(embed_matrix, X) #50 , 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [],
   "source": [
    "#X.shape, Y.shape, embed_matrix.shape, embed.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tf.nn.nce_loss(weights, biases, labels, inputs, num_sampled, num_classes, num_true=1,\n",
    "# sampled_values=None, remove_accidental_hits=False, partition_strategy='mod',name='nce_loss')\n",
    "nce_weight = tf.Variable(tf.random_uniform([VOCAB_SIZE, EMBED_SIZE],-1.0,1.0)) # (12, 5)\n",
    "nce_bias = tf.Variable(tf.zeros([VOCAB_SIZE]))#12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [],
   "source": [
    "#nce_weight, nce_bias\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss = tf.reduce_mean(tf.nn.nce_loss(weights=nce_weight,\n",
    "                                     biases=nce_bias,\n",
    "                                     labels=Y,\n",
    "                                     inputs=embed,\n",
    "                                     num_sampled=NUM_SAMPLED,\n",
    "                                     num_classes=VOCAB_SIZE\n",
    "                                    ))\n",
    "#print (loss)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = tf.train.AdamOptimizer(1e-1).minimize(loss)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss at 0 8.878108\n",
      "Loss at 1000 2.1282516\n",
      "Loss at 2000 1.3175844\n",
      "Loss at 3000 2.2427027\n",
      "Loss at 4000 1.2110066\n",
      "Loss at 5000 1.5972496\n",
      "Loss at 6000 2.8704581\n",
      "Loss at 7000 2.8053105\n",
      "Loss at 8000 2.0491073\n",
      "Loss at 9000 1.7230179\n"
     ]
    }
   ],
   "source": [
    "epochs = 10000\n",
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    \n",
    "    for epoch in range (epochs):\n",
    "        batch_inputs, batch_labels = get_batch(BATCH_SIZE)\n",
    "        _,loss_val = sess.run([optimizer,loss],feed_dict={X: batch_inputs, Y: batch_labels})\n",
    "    \n",
    "        if epoch % 1000 == 0:\n",
    "            print(\"Loss at\", epoch, loss_val )\n",
    "        \n",
    "    temp = embed_matrix.eval()\n",
    "# words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.6944479 , -0.25988418, -0.6709179 ,  1.0951345 ,  0.9476659 ],\n",
       "       [-0.83016324, -0.19688472,  2.0873039 , -1.7829826 , -0.06266162],\n",
       "       [ 1.0977699 , -0.141658  ,  1.4091423 , -2.64497   ,  0.5532724 ],\n",
       "       [-0.22961949, -0.9885367 ,  0.91091657, -0.755999  ,  1.8764786 ],\n",
       "       [ 2.0185442 ,  1.2253134 ,  0.1301832 , -1.807558  , -0.11586981],\n",
       "       [-1.1000371 ,  0.54688513, -0.01973297, -1.0040284 , -1.1031016 ],\n",
       "       [ 0.56303626, -1.8714243 , -0.29543772, -0.53828394, -0.6383572 ],\n",
       "       [-0.13107562, -0.05864046,  0.48632798, -0.22897488, -0.99308646],\n",
       "       [-0.12486245, -0.33267194, -0.4034236 , -0.74628454, -1.0242493 ],\n",
       "       [-0.7966839 , -1.5956622 ,  1.3138496 , -0.883736  ,  0.40056163],\n",
       "       [-1.8473268 , -0.7873239 ,  0.9229829 , -0.29741883,  1.1911665 ],\n",
       "       [-0.6692132 , -0.56363994, -0.24505115,  0.77865577, -0.02694191],\n",
       "       [ 0.96776   , -0.07424176,  0.424195  ,  0.979758  ,  0.7956037 ],\n",
       "       [ 0.41498128, -0.24653174, -1.3398798 ,  0.02364797,  2.2288456 ],\n",
       "       [-0.05130024, -1.2013073 ,  1.7001089 ,  0.6638979 ,  0.7655233 ],\n",
       "       [ 0.21324031, -0.5775961 , -0.90290797, -2.5316107 ,  0.82906777],\n",
       "       [ 0.28896362, -0.17394818, -1.7002294 , -0.0509948 ,  0.7434674 ],\n",
       "       [ 0.06547707,  1.5542159 , -1.0368454 ,  0.8679077 ,  1.2674894 ],\n",
       "       [ 0.3802185 ,  1.4344517 , -1.1640915 , -0.02721157,  0.5312348 ],\n",
       "       [ 0.45143968,  0.25331858, -0.09044404,  0.7782993 ,  0.17989238],\n",
       "       [ 1.0224817 ,  0.45860085,  0.05999306,  0.2832582 , -0.95989096],\n",
       "       [ 1.1815623 , -0.07145299, -0.0567335 , -0.05053306,  2.1189413 ]],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fit a 2d PCA model to the vectors\n",
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=2)\n",
    "trained_embeedings = pca.fit_transform(temp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.5520433 , -0.6990878 ],\n",
       "       [ 2.5119753 ,  0.10410953],\n",
       "       [ 1.8522092 ,  1.3372521 ],\n",
       "       [ 0.78593045, -1.1874233 ],\n",
       "       [-0.00537155,  2.5185468 ],\n",
       "       [ 0.81831336,  0.8820145 ],\n",
       "       [ 0.63966703,  0.02931802],\n",
       "       [ 0.60480636,  0.3896371 ],\n",
       "       [ 0.44223282,  0.7894859 ],\n",
       "       [ 1.8720802 , -1.0829864 ],\n",
       "       [ 1.0956815 , -1.740187  ],\n",
       "       [-0.35508478, -0.9126523 ],\n",
       "       [-0.89743704, -0.6277718 ],\n",
       "       [-1.6190529 , -0.6100271 ],\n",
       "       [ 0.7840615 , -1.7202698 ],\n",
       "       [ 0.6591576 ,  1.12912   ],\n",
       "       [-1.4080267 ,  0.16863525],\n",
       "       [-2.1986866 , -0.02471079],\n",
       "       [-1.6216176 ,  0.92647344],\n",
       "       [-0.91669124, -0.17399745],\n",
       "       [-0.4759838 ,  0.9261311 ],\n",
       "       [-1.0161183 , -0.42160872]], dtype=float32)"
      ]
     },
     "execution_count": 264,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trained_embeedings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['manchester',\n",
       " 'united',\n",
       " 'football',\n",
       " 'club',\n",
       " 'is',\n",
       " 'a',\n",
       " 'professional',\n",
       " 'based',\n",
       " 'in',\n",
       " 'old',\n",
       " 'trafford',\n",
       " 'england',\n",
       " 'liverpool',\n",
       " 'that',\n",
       " 'competes',\n",
       " 'the',\n",
       " 'premier',\n",
       " 'league',\n",
       " 'top',\n",
       " 'tier',\n",
       " 'of',\n",
       " 'english']"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labelss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "#show word2vec if dim is 2\n",
    "if trained_embeedings.shape[1] == 2:\n",
    "    #labels = data[:10] #Show top 10 words\n",
    "#     plt.xlim(-2.5, 2.4)\n",
    "#     plt.ylim(-2.0, 2.2)\n",
    "    for i, label in enumerate(labelss):\n",
    "        x,y = trained_embeedings[i,:]\n",
    "        plt.scatter(x,y)\n",
    "        plt.annotate(label, xy=(x,y), xytext=(9,3),textcoords='offset points', ha='right', va='bottom')\n",
    "        #plt.savefig('word2vev.png')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}