{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Wikipedia2Vec: Word and Entity Embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import NumPy\n",
    "import numpy as np\n",
    "\n",
    "# Import PCA from scikit-learn\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# Import PCA from scikit-learn\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib notebook\n",
    "%matplotlib inline\n",
    "\n",
    "# Import Gensim for the word embeddings\n",
    "import gensim.downloader as api\n",
    "from gensim.models import KeyedVectors\n",
    "from gensim.test.utils import datapath\n",
    "\n",
    "# Load Word2Vec trained on Google News data\n",
    "# wv = api.load('word2vec-google-news-300')\n",
    "\n",
    "# Load Wiki2Vec trained on Wikipedia\n",
    "wv = KeyedVectors.load_word2vec_format(datapath(\"enwiki\"), binary=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Word Vectors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the\n",
      "in\n",
      "of\n",
      "a\n",
      "and\n",
      "is\n",
      "to\n",
      "was\n",
      "by\n",
      "for\n"
     ]
    }
   ],
   "source": [
    "for i, word in enumerate(wv.vocab):\n",
    "    if i == 10:\n",
    "        break\n",
    "    print(word)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 6.290e-02  2.230e-02  4.510e-01 -1.620e-02  3.985e-01 -6.590e-02\n",
      "  4.334e-01  9.800e-02 -6.105e-01  1.680e-02 -6.200e-02  1.167e-01\n",
      "  6.125e-01 -7.804e-01  1.250e-01 -1.001e-01 -1.234e-01 -2.259e-01\n",
      "  2.815e-01  5.150e-02 -2.869e-01 -1.962e-01 -2.072e-01  1.118e-01\n",
      " -7.739e-01 -1.102e-01  3.452e-01 -5.316e-01  1.792e-01  5.650e-01\n",
      "  1.162e-01 -9.440e-02 -2.202e-01  4.636e-01  3.135e-01 -1.856e-01\n",
      " -2.852e-01 -1.421e-01  2.247e-01 -6.170e-02 -1.201e-01 -3.090e-02\n",
      "  3.743e-01  1.347e-01  2.950e-02  2.166e-01 -6.869e-01  6.377e-01\n",
      "  2.905e-01  6.650e-02  2.993e-01  2.747e-01 -1.418e-01 -1.034e-01\n",
      " -3.268e-01 -1.560e-01 -2.698e-01  8.620e-02 -6.706e-01 -5.927e-01\n",
      "  3.726e-01 -3.735e-01  2.200e-02  4.840e-02  2.411e-01 -1.792e-01\n",
      "  1.400e-01  6.690e-02 -1.725e-01 -1.923e-01 -3.746e-01  5.210e-02\n",
      "  1.635e-01  5.229e-01 -2.109e-01 -5.000e-04 -1.261e-01  3.470e-02\n",
      "  3.800e-02 -1.784e-01  2.044e-01 -5.451e-01  3.420e-01 -1.630e-02\n",
      "  1.474e-01  5.746e-01  2.365e-01  2.205e-01  4.692e-01  1.333e-01\n",
      "  9.830e-02  4.086e-01 -1.533e-01  1.069e-01 -1.708e-01 -7.374e-01\n",
      " -5.872e-01  2.633e-01 -6.793e-01 -3.180e-01]\n"
     ]
    }
   ],
   "source": [
    "vec_king = wv['king']\n",
    "print(vec_king)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Words Similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'car'\t'minivan'\t0.75\n",
      "'car'\t'bicycle'\t0.69\n",
      "'car'\t'airplane'\t0.57\n",
      "'car'\t'cereal'\t0.29\n",
      "'car'\t'communism'\t0.20\n"
     ]
    }
   ],
   "source": [
    "pairs = [\n",
    "    ('car', 'minivan'),   # a minivan is a kind of car\n",
    "    ('car', 'bicycle'),   # still a wheeled vehicle\n",
    "    ('car', 'airplane'),  # ok, no wheels, but still a vehicle\n",
    "    ('car', 'cereal'),    # ... and so on\n",
    "    ('car', 'communism'),\n",
    "]\n",
    "for w1, w2 in pairs:\n",
    "    print('%r\\t%r\\t%.2f' % (w1, w2, wv.similarity(w1, w2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('batmobile', 0.8302738666534424), ('batcycle', 0.7926667928695679), ('ENTITY/Batmobile', 0.7804111838340759), ('batblade', 0.7716271877288818), ('batpod', 0.7674230337142944), ('jokermobile', 0.7542746067047119), ('batcave', 0.7528234720230103), ('batman', 0.7526453137397766), ('ENTITY/Batcave', 0.7499290108680725), ('ENTITY/List_of_Cars_characters#Strip_\"The_King\"_Weathers', 0.7435188293457031)]\n"
     ]
    }
   ],
   "source": [
    "print(wv.most_similar(positive=['car', 'ENTITY/Batman'], topn=10)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Word Analogies\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def analogy(x1, x2, y1):\n",
    "    result = wv.most_similar(positive=[y1, x2], negative=[x1])\n",
    "    return result[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'brazilian'"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analogy('japan', 'japanese', 'brazil')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'man'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analogy('queen', 'king', 'woman')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'strange'"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analogy('good', 'fantastic', 'bad')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def display_pca_scatterplot(model, words):\n",
    "\n",
    "    word_vectors = np.array([model[w] for w in words])\n",
    "\n",
    "    twodim = PCA().fit_transform(word_vectors)[:,:2]\n",
    "    \n",
    "    plt.figure(figsize=(6,6))\n",
    "    plt.scatter(twodim[:,0], twodim[:,1], edgecolors='k', c='r')\n",
    "    for word, (x,y) in zip(words, twodim):\n",
    "        plt.text(x+0.05, y+0.05, word)\n",
    "        \n",
    "def display_closestwords(model, word):\n",
    "    \n",
    "    # get close words\n",
    "    a = model.similar_by_word(word)\n",
    "    close_words = [w for (w,s) in a]\n",
    "    close_words.append(word)\n",
    "    display_pca_scatterplot(model, close_words)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "display_pca_scatterplot(wv, \n",
    "                        ['coffee', 'tea', 'beer', 'wine', 'brandy', 'rum', 'champagne', 'water',\n",
    "                         'spaghetti', 'borscht', 'hamburger', 'pizza', 'falafel', 'sushi', 'meatballs',\n",
    "                         'dog', 'horse', 'cat', 'monkey', 'parrot', 'koala', 'lizard',\n",
    "                         'frog', 'toad', 'monkey', 'ape', 'kangaroo', 'wombat', 'wolf',\n",
    "                         'france', 'germany', 'hungary', 'france', 'australia', 'fiji', 'china',\n",
    "                         'homework', 'assignment', 'problem', 'exam', 'test', 'class',\n",
    "                         'school', 'college', 'university', 'institute'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAFlCAYAAADWN/6SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3deXyU1d3//9dJRGxANsFqBRKoyJJ9Q2QNIoLLDYJwK0YEWVJQ0fb250NqxIUal5/0VnEpNxYENeKCVbBapRYQqCgJEJBVQMKiFJFNIcSS8Pn+kTBNQgJK5spMhvfz8ZhHZs515jrnGjHvnOu6zhlnZoiIiPhbWKA7ICIioUkBIyIinlDAiIiIJxQwIiLiCQWMiIh4QgEjIiKeOCvQHahK06ZNLSoqKtDdEBGpVZYvX/6dmTULdD8giAMmKiqK3NzcQHdDRKRWcc5tC3QfjtMpMhER8YQCRkREPKGAkZDy0EMPMWnSpEB345TS0tJ0ClhCngJGREQ8oYCRWi8rK4u2bdtyxRVXsHHjRgDy8vLo1KkTcXFxDBgwgP379/Ptt9+SnJwMwKpVq3DOsX37dgB+/etfU1BQwPDhw7nzzjvp3LkzrVu3Zvbs2QDcdtttzJ07F4ABAwYwYsQIAKZNm8b9998PwHXXXUdycjLR0dFMnToVgOLiYoYPH05MTAyxsbE89dRTvn6/9dZbdOzYkUsuuYTFixfXwCclUrMUMFKrLV++nNdff52VK1fyl7/8hZycHABuueUWnnjiCVavXk1sbCwPP/ww559/PoWFhXz//fcsXryYlJQUFi9ezLZt2zj//POJiIgAYNeuXSxZsoS//vWvjB8/HoDu3bv7QuDrr79m3bp1ACxZsoRu3boBMH36dJYvX05ubi6TJ09m79695OXl8fXXX7NmzRq++OILbr31Vl/fi4qKWLZsGU8//TQPP/xwjX1mIjVFASO12uLFixkwYAARERE0aNCAfv36cfjwYQ4cOECPHj0AGDZsGIsWLQKgc+fO/POf/2TRokXcd999LFq0iMWLF/tCAkpGImFhYXTo0IHdu3cD0K1bNxYvXsy6devo0KEDv/zlL9m1axdLly6lc+fOAEyePJn4+Hg6derEjh072LRpE61bt+arr75i3LhxfPjhhzRo0MDXzsCBAwFITk4mPz+/Jj4ukRqlgJFazzn3k+seD4pt27bRv39/Vq1axZIlS+jevbuvTt26dX3Pj39f0kUXXcT+/fv58MMP6d69O926dePNN9+kfv36nHvuuSxcuJCPP/6YpUuXsmrVKhITEyksLKRx48asWrWKtLQ0nn/+eUaNGnVCO+Hh4RQVFVX3YxAJOgoYqdW6d+/OO++8w5EjR/jhhx947733qFevHo0bN/ad0nrllVd8o5nu3bvz6quv0qZNG8LCwmjSpAkffPABXbp0OWVbl112GU8//bQvYCZNmuQb+Rw8eJDGjRsTERHBhg0b+OyzzwD47rvvOHbsGNdffz1/+MMfWLFihUefhEjwCdqZ/CI/RVJSEjfccAMJCQlERkb6fuHPnDmTMWPGUFBQQOvWrXnppZeAkhUiAN+IpWvXruzcuZPGjRufsq1u3boxb948Lr74YiIjI9m3b5+vvb59+zJlyhTi4uJo27YtnTp1Akqu19x6660cO3YMgMcee8yvxy8SzFywfmVySkqKaZ6AHDcrO5uszEzWb99O+5YtyczKYkh6eqC7JRJ0nHPLzSwl0P0AjWCkFpiVnU1mRgbTCgroCizZto2RGRkAChmRIKZrMBL0sjIzmVZQQE+gDtATmFZQQFZmZoB7JiIno4CRoLd++3a6VijrWlouIsFLASNBr33LliypULaktFxEgpcCRoJeZlYWIyMiWAAcBRYAIyMiyMzKCnDPRORkdJFfgt7xC/njytxFlqW7yESCnm5TFhEJIcF0m7JOkYmIiCcUMCIi4gkFjIiIeEIBIyIinlDAiIiIJxQwIiLiCQWMiIh4QgEjIiKeUMCIiIgnFDAiIuIJBYyIiHhCASMiIp5QwIiIiCcUMCIi4olqB4xzroVzboFzbr1zbq1z7q5K6qQ55w465/JKHw9Ut10REQlu/vjCsSLgbjNb4Zw7F1junPu7ma2rUG+xmV3rh/ZERKQWqPYIxsx2mdmK0uc/AOuBi6q7XxERqd38eg3GORcFJAKfV7L5MufcKufc35xz0VW8P8M5l+ucy92zZ48/uyYiIjXMbwHjnKsPvA381sy+r7B5BRBpZvHAs8C7le3DzKaaWYqZpTRr1sxfXRMRkQDwS8A45+pQEi7ZZvaXitvN7HszO1T6/AOgjnOuqT/aFhGR4OSPu8gcMA1Yb2b/W0WdC0rr4ZzrWNru3uq2LSIiwcsfd5F1AYYCXzjn8krL7gNaApjZFGAQMNY5VwQcAW40M/ND2yIiEqSqHTBmtgRwp6jzHPBcddsSEZHaQzP5RUTEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU9UO2Cccy2ccwucc+udc2udc3dVUsc55yY75zY751Y755Kq266IiAS3s/ywjyLgbjNb4Zw7F1junPu7ma0rU+cqoE3p41LgT6U/RUQkRFV7BGNmu8xsRenzH4D1wEUVqvUHXrYSnwGNnHMXVrdtEREJXn69BuOciwISgc8rbLoI2FHm9U5ODCEREQkhfgsY51x94G3gt2b2fcXNlbzFKtlHhnMu1zmXu2fPHn91TUREAsAvAeOcq0NJuGSb2V8qqbITaFHmdXPgm4qVzGyqmaWYWUqzZs380TUREQkQf9xF5oBpwHoz+98qqs0Fbim9m6wTcNDMdlW3bRERCV7+uIusCzAU+MI5l1dadh/QEsDMpgAfAFcDm4EC4FY/tCsiIkGs2gFjZkuo/BpL2ToG3F7dtkREpPbQTH4REfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEE34JGOfcdOfct865NVVsT3POHXTO5ZU+HvBHuyIiErzO8tN+ZgDPAS+fpM5iM7vWT+2JiEiQ88sIxswWAfv8sS8REQkNNXkN5jLn3Crn3N+cc9E12K6IiASAv06RncoKINLMDjnnrgbeBdpUrOScywAyAFq2bFlDXRMRES/UyAjGzL43s0Olzz8A6jjnmlZSb6qZpZhZSrNmzWqiayIi4pEaCRjn3AXOOVf6vGNpu3trom0REQkMv5wic87NAtKAps65ncCDQB0AM5sCDALGOueKgCPAjWZm/mhbRESCk18CxsyGnGL7c5TcxiwiImcIzeQXERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERT/glYJxz051z3zrn1lSx3TnnJjvnNjvnVjvnkvzRroiIBC9/jWBmAH1Psv0qoE3pIwP4k5/aFRGRIOWXgDGzRcC+k1TpD7xsJT4DGjnnLvRH2yIiEpxq6hrMRcCOMq93lpaV45zLcM7lOudy9+zZU0NdExERL9RUwLhKyuyEArOpZpZiZinNmjWrgW6JiIhXaipgdgItyrxuDnxTQ22LiEgA1FTAzAVuKb2brBNw0Mx21VDbIiISAGf5YyfOuVlAGtDUObcTeBCoA2BmU4APgKuBzUABcKs/2hURkeDll4AxsyGn2G7A7f5oS0REagfN5BcREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKmACbO3cujz/+eKXb6tevf8r3d+7c2d9dEhHxC1fyVS3BJyUlxXJzcwPdjYCqX78+hw4d+tnvKy4uJjw83IMeiUiwc84tN7OUQPcDNILxVH5+Pu3atWPUqFHExMSQnp7Oxx9/TJcuXWjTpg3Lli1jxowZ3HHHHQBs3bqVyy67jNTUVCZMmODbz6FDh+jVqxdJSUnExsYyZ84c37bjo5yFCxfSs2dPbrrpJmJjY2v2QEVEKqGA8djmzZu56667WL16NRs2bOC1115jyZIlTJo0iUcffbRc3bvuuouxY8eSk5PDBRdc4Cs/55xzeOedd1ixYgULFizg7rvvprKR57Jly8jKymLdunWeH5eIyKkoYDzWqlUrYmNjCQsLIzo6ml69euGcIzY2lvz8/HJ1//nPfzJkSMm3Tw8dOtRXbmbcd999xMXFccUVV/D111+ze/fuE9rq2LEjrVq18vR4RER+qrMC3YFQV7duXd/zsLAw3+uwsDCKiopOqO+cO6EsOzubPXv2sHz5curUqUNUVBSFhYUn1KtXr54fey4iUj0awQSRLl268PrrrwMloXLcwYMHOf/886lTpw4LFixg27ZtgeqiiMhPpoAJIs888wzPP/88qampHDx40Feenp5Obm4uKSkpZGdn065duwD2UkTkp9FtyiIiIUS3KYeYWdnZxERFER4WRkxUFLPKnN4SETlT6SJ/Nc3KziYzI4NpBQV0BZZs28bIjAwAhqSnB7ZzIiIBpBFMNWVlZjKtoICeQB2gJzCtoICszMwA90xEJLAUMNW0fvt2ulYo61paLiJyJlPAVFP7li1ZUqFsSWm5iMiZTAFTTZlZWYyMiGABcBRYAIyMiCAzKyvAPRMRCSxd5K+m4xfyx2Vmsn77dtq3bElWVpYu8IvIGU/zYEREQojmwYiISMjzS8A45/o65zY65zY758ZXsn24c26Pcy6v9DHKH+2KiEjwqvY1GOdcOPA80BvYCeQ45+aaWcUvJXnDzO6obnsiIlI7+GME0xHYbGZfmdm/gdeB/n7Yr4iI1GL+CJiLgB1lXu8sLavoeufcaufcbOdcCz+0KyIiQcwfAXPiN2RBxVvT3gOizCwO+BiYWemOnMtwzuU653L37Nnjh66JiEig+CNgdgJlRyTNgW/KVjCzvWb2Y+nLF4HkynZkZlPNLMXMUpo1a+aHromISKD4I2BygDbOuVbOubOBG4G5ZSs45y4s87IfsN4P7YqISBCr9l1kZlbknLsD+AgIB6ab2Vrn3EQg18zmAnc65/oBRcA+YHh12xURkeCmmfwiIiFEM/lFRCTkKWBERMQTChgREfGEAkZERDyhgBEREU8oYERExBMKGBER8YQCRkREPKGAERERTyhgRETEEwoYERHxhAJGzghpaWlobTuRmqWAERERTyhgJKTk5+fTrl07hg0bRlxcHIMGDaKgoKBcnbFjx5KSkkJ0dDQPPvigrzwqKooHH3yQpKQkYmNj2bBhAwCHDx9mxIgRpKamkpiYyJw5c2r0mERqKwWMhJyNGzeSkZHB6tWradCgAS+88EK57VlZWeTm5rJ69Wo++eQTVq9e7dvWtGlTVqxYwdixY5k0aZKv/uWXX05OTg4LFizgnnvu4fDhwzV6TCK1kQJGQk6LFi3o0qULADfffDNLliwpt/3NN98kKSmJxMRE1q5dy7p163zbBg4cCEBycjL5+fkAzJs3j8cff5yEhATS0tIoLCxk+/btNXMwIrVYtb/RUiTYOOeqfL1161YmTZpETk4OjRs3Zvjw4RQWFvq2161bF4Dw8HCKiooAMDPefvtt2rZtWwO9FwkdGsFIyNm+fTtLly4FYNasWXTt2tW37fvvv6devXo0bNiQ3bt387e//e2U++vTpw/PPvssx7/9deXKld50XCTEKGAk5LRv356ZM2cSFxfHvn37GDt2rG9bfHw8iYmJREdHM2LECN+ptJOZMGECR48eJS4ujpiYGCZMmOBl90VChjv+V1mwSUlJMc1bkJ8rPz+fa6+9ljVr1gS6KyIB4ZxbbmYpge4HaAQjtdSs7GxioqIIDwsjJiqKWdnZge6ShID8/HxiYmLO+D74iy7yS60zKzubzIwMphUU0BVYsm0bIzMyABiSnq7RiwREUVERZ52lX6llaQQjtU5WZibTCgroCdQBegLTCgrIyswMcM8kFBQXFzN69Giio6O58sorOXLkCHl5eXTq1Im4uDgGDBjA/v37gZIliO677z569OjBM888w7Zt2+jVqxdxcXH06tXLdzv7W2+9RUxMDPHx8XTv3h2AGTNm0L9/f/r27Uvbtm15+OGHT9oHgC1bttC3b1+Sk5Pp1q2bbzLw8OHD+Z//+R969uwJ0Nw5V885N905l+OcW+mc619zn2AZZhaUj+TkZBOpTJhz9m8wK/P4N1iYc4HumtRyW7dutfDwcFu5cqWZmQ0ePNheeeUVi42NtYULF5qZ2YQJE+yuu+4yM7MePXrY2LFjfe+/9tprbcaMGWZmNm3aNOvfv7+ZmcXExNjOnTvNzGz//v1mZvbSSy/ZBRdcYN99950VFBRYdHS05eTkVNkHM7PLL7/cvvzySzMz++yzz6xnz55mZjZs2DC75pprrKioyIBc4FHgZiu5xt4I+BKoZzX8e1wjGKl12rdsyZIKZUtKy0Wqq1WrViQkJAAlE263bNnCgQMH6NGjBwDDhg1j0aJFvvo33HCD7/nSpUu56aabABg6dKhvkm+XLl0YPnw4L774IsXFxb76vXv35rzzzuMXv/gFAwcO9NWv2If8/HwOHTrEp59+yuDBg0lISOA3v/kNu3bt8u1r8ODBhIeHH395JTDeOZcHLATOAWr8fxCdMJRaJzMri5Flr8EAIyMiyMrKCnTXJAQcn2wLJRNuDxw4cNL69erVq3Lb8Um+U6ZM4fPPP+f9998nISGBvLy8ctsr1q/YhyNHjnDs2DEaNWrke+8p+uGA681s40k77zGNYKTWGZKeTtbUqYyLjOQc5xgXGUnW1KkMSU8PdNckBDVs2JDGjRuzePFiAF555RXfaKaizp078/rrrwOQnZ3tm+S7ZcsWLr30UiZOnEjTpk3ZsWMHAH//+9/Zt28fR44c4d133z3pvKwGDRrQqlUr3nrrLaDk8saqVauqqv4RMM6VJpZzLvHnHrc/aAQjtdKQ9HQFitSYmTNnMmbMGAoKCmjdujUvvfRSpfUmT57MiBEjePLJJ2nWrJmv3j333MOmTZswM3r16kV8fDx5eXl07dqVoUOHsnnzZm666SZSUlJ8a+BVJjs7m7Fjx/LII49w9OhRbrzxRuLj4yur+gfgaWB1acjkA9dW60M4DZpoKSISADNmzCA3N5fnnnvOr/vVREsRkQDSRN2aoRGMiJxRTpioS+lNIiFyHU8jGBGRANFE3Zrjl4BxzvV1zm10zm12zo2vZHtd59wbpds/d85F+aNdEZGfa/327XStUNa1tFz8q9oB45wLB54HrgI6AEOccx0qVBsJ7Dezi4GngCeq266IyOnQRN2a448RTEdgs5l9ZWb/Bl4HKq570x+YWfp8NtDLVZxhJPITpKWloWtzUh2ZWVmMjIhgAXAUWEDJNZhMTdT1O3/Mg7kI2FHm9U7g0qrqmFmRc+4gcB7wnR/aFxH5yY5fyB+Xmcn67dtp37IlWVlZIXGBP9j4YwRT2Uik4q1pP6UOzrkM51yucy53z549fuia1Fb5+fm0a9eOYcOGERcXx6BBgygoKChXZ+zYsaSkpBAdHc2DDz7oK4+KiuLBBx8kKSmJ2NhY34qzhw8fZsSIEaSmppKYmMicOXMAWLt2LR07diQhIYG4uDg2bdpUcwcqATEkPZ01+fkUHzvGmvx8hYtH/BEwO4EWZV43B76pqo5z7iygIbCv4o7MbKqZpZhZSrNmzfzQNanNNm7cSEZGBqtXr6ZBgwa88MIL5bZnZWWRm5vL6tWr+eSTT1i9erVvW9OmTVmxYgVjx45l0qRJvvqXX345OTk5LFiwgHvuuYfDhw8zZcoU7rrrLvLy8sjNzaV58+Y1epwiocofAZMDtHHOtXLOnQ3cCMytUGcuMKz0+SBgvgXrBBwJGi1atPCtzXTzzTf7Vpo97s033yQpKYnExETWrl3LunXrfNsGDhwI/GclWoB58+bx+OOPk5CQQFpaGoWFhWzfvp3LLruMRx99lCeeeIJt27bxi1/8omYOUCTEVfsaTOk1lTsoWVwtHJhuZmudcxOBXDObC0wDXnHObaZk5HJjdduV0FfVSrMAW7duZdKkSeTk5NC4cWOGDx9OYWGhb/vx1WjDw8MpKioCShYHfPvtt2nbtm25/bZv355LL72U999/nz59+vDnP/+Zyy+/3KvDEjlj+GUejJl9YGaXmNmvzSyrtOyB0nDBzArNbLCZXWxmHc3sK3+0K6Ft+/btLF26FIBZs2b5VqYF+P7776lXrx4NGzZk9+7d/O1vfzvl/vr06cOzzz7L8cHzypUrAfjqq69o3bo1d955J/369St3qk1ETp9m8kvQat++PTNnziQuLo59+/YxduxY37b4+HgSExOJjo5mxIgRJ13m/LgJEyZw9OhR4uLiiImJYcKECQC88cYbxMTEkJCQwIYNG7jllls8OyaRM4nWIpOAmpWdTVaZ20UzS28Xzc/P59prr2XNmjWB7qJIrXI6a5E5534FTDazQf7si74PRgLmhEUHt21jZEYGAJf9hBGJiPiHmX1DyQ1YfqVTZBIwJ1t0MCoqSqMXkVJVzQubOHEiqampxMTEkJGR4bu+6Jxb6Jx72jn3qXNujXOuY2n5Q865V5xz851zm5xzo0vLo5xza8o8X+ycW1H66Fxanla639nOuQ3OuexTrciigJGA0aKDIj9dZfPC7rjjDnJyclizZg1Hjhzhr3/9a9m31DOzzsBtwPQy5XHANcBlwAOlp8fK+hbobWZJwA3A5DLbEoHfUrLuZGvgpKcaFDASMFp0UOSnq2xe2IIFC7j00kuJjY1l/vz5rF27tuxbZgGY2SKggXOuUWn5HDM7YmbfUbIUW8cKTdUBXnTOfQG8RUmYHLfMzHaa2TEgD4g6WZ8VMBIwWnRQ5KerbF7YbbfdxuzZs/niiy8YPXp0ublgnLgcl52i/LjfAbuBeCAFOLvMth/LPC/mFNfxFTASMEPS08maOpVxkZGc4xzjIiND5lsFRfytqnlhTZs25dChQ8yePbviW24AcM51BQ6a2cHS8v7OuXOcc+cBaZSsxlJWQ2BX6ShlKCUT6E+L7iKTgBqSnq5AEfkJjs8L+81vfkObNm0YO3Ys+/fvJzY2lqioKFJTUyu+Zb9z7lOgATCiTPky4H2gJfAHM/umwpdAvgC87ZwbTMmJhcOn22fNgxERCXI/Z16Yc245cAj4/8wst8K2h4BDZjbJk45WoFNkIiJBYFZ2NjFRUYSHhRETFcWs7OxAd6nadIpMRCTATjbpeEh6+s+eF2ZmaVWUP+SH7v5kGsGIiATYySYd12YKGBGRAAvVSccKGBGRAAvVScdnfMBcffXVHDhw4Ge9Z/jw4ZXdcy4iclpCddLxGX+R/4MPPjihzMwwM8LCzvj8FZEacHwu2LgyX12RVfrVFbXZGfUb9LrrriM5OZno6GimTp0KQFRUFN999x35+fm0b9+e2267jaSkJHbs2EH9+vW5++67SUpKolevXuzZs+eEfVa1mmlaWhr33nsvHTt25JJLLmHx4sUAFBcXc88995CamkpcXBz/93//V3MfgIgErSHp6azJz6f42DHW5OfX+nCBMyxgpk+fzvLly8nNzWXy5Mns3bu33PaNGzdyyy23sHLlSiIjIzl8+DBJSUmsWLGCHj168PDDD5+wz5OtZlpUVMSyZct4+umnfe+dNm0aDRs2JCcnh5ycHF588UW2bt3q7YGLiATAGRUwkydPJj4+nk6dOrFjxw42bdpUbntkZCSdOnXyvQ4LC+OGG24A/rN6aUUnW8104MCBACQnJ5Ofnw/AvHnzePnll0lISODSSy9l7969J/RDRCQUnDHXYBYuXMjHH3/M0qVLiYiIIC0treLKo9SrV++k+6i4mmlhYSG33XYbubm5tGjRgoceeqjcPuvWrQtAeHg4RUVFQMn1nWeffZY+ffr447BERILWGTOCOXjwII0bNyYiIoINGzbw2WefnfI9x44d890t9tprr/lWLz3ueJicZDXTE/Tp04c//elPHD16FIAvv/ySw4dPey05EZGgdcaMYPr27cuUKVOIi4ujbdu25U6FVaVevXqsXbuW5ORkGjZsyBtvvFFue6NGjRg9evTJVjM9wahRo8jPzycpKQkzo1mzZrz77runfVwiIsFKqymfRP369Tl06FBA+yAi8nM455abWUqg+wEheIosFFckFRGpjULqFNmpViT9uTR6ERE5fSE1ggnVFUlFRGqjkAqYUF2RVESkNgqpgAnVFUlFRGqjkAqYUF2RVESkNgqpi/yhuiKpiEhtpHkwIiIhRPNgREQk5FUrYJxzTZxzf3fObSr92biKesXOubzSx9zqtCkiIrVDdUcw44F/mFkb4B+lrytzxMwSSh/9qtmmiIjUAtUNmP7AzNLnM4Hrqrk/EREJEdUNmF+a2S6A0p/nV1HvHOdcrnPuM+dclSHknMsorZdb2dcTS/X17duXRo0ace2115YrNzMyMzO55JJLaN++PZMnT/aV33nnnVx88cXExcWxYsUK33tmzpxJmzZtaNOmDRlE1W4AAA/lSURBVDNnlvydUVBQwDXXXEO7du2Ijo5m/PiqBrUiEvLM7KQP4GNgTSWP/sCBCnX3V7GPX5X+bA3kA78+VbvJyckm/vfxxx/b3Llz7ZprrilXPn36dBs6dKgVFxebmdnu3bvNzOz999+3vn372rFjx2zp0qXWsWNHMzPbu3evtWrVyvbu3Wv79u2zVq1a2b59++zw4cM2f/58MzP78ccfrWvXrvbBBx/U4BGKnNmAXDvF79eaepxyBGNmV5hZTCWPOcBu59yFAKU/v61iH9+U/vwKWAgknl4cSkX5+fm0b9+e0aNHEx0dzZVXXsmRI0fYvHkzV1xxBfHx8SQlJbFlyxYAevXqxbnnnnvCfv70pz/xwAMPEBZW8k/i/PNLBqNz5szhlltuwTlHp06dOHDgALt27eKjjz6id+/eNGnShMaNG9O7d28+/PBDIiIi6NmzJwBnn302SUlJ7Ny5s4Y+DREJJtU9RTYXGFb6fBgwp2IF51xj51zd0udNgS7Aumq2K2Vs2rSJ22+/nbVr19KoUSPefvtt0tPTuf3221m1ahWffvopF1544Un3sWXLFt544w1SUlK46qqr2LRpEwBff/01LVq08NVr3rw5X3/9dZXlZR04cID33nuPXr16+fFoRaS2qG7APA70ds5tAnqXvsY5l+Kc+3NpnfZArnNuFSWrtzxuZgoYP2rVqhUJCQkAJCcns3XrVr7++msGDBgAwDnnnENERMRJ9/Hjjz9yzjnnkJuby+jRoxkxYgTA8VOc5Tjnqiw/rqioiCFDhnDnnXfSunXr0z42Eam9qhUwZrbXzHqZWZvSn/tKy3PNbFTp80/NLNbM4kt/TvNHx+U/6tat63seHh7O/v37f/Y+mjdvzvXXXw/AgAEDWL16ta98x44dvno7d+7kV7/6VZXlx2VkZNCmTRt++9vf/uy+HHfgwAFeeOGF03rv8OHDmT179mm3/VM88MADfPzxx562IVKbaSZ/CGrQoAHNmzfn3XffBUpGJwUFBSd9z3XXXcf8+fMB+OSTT7jkkksA6NevHy+//DJmxmeffUbDhg258MIL6dOnD/PmzWP//v3s37+fefPm0adPHwDuv/9+Dh48yNNPP12t46hOwNSEiRMncsUVV5xQXlxcHIDeiAQfBUyIeuWVV5g8eTJxcXF07tyZf/3rXwB069aNwYMH849//IPmzZvz0UcfATB+/HjefvttYmNj+f3vf8+f/1xyhvPqq6+mdevWXHzxxYwePdr3C79JkyZMmDCB1NRUUlNTeeCBB2jSpAk7d+4kKyuLdevWkZSUREJCgm9fP9f48ePZsmULCQkJ3HPPPTz55JOkpqYSFxfHgw8+6Kv38ssvExcXR3x8PEOHDvWVL1q0iM6dO9O6dWvfaGbhwoWkpaUxaNAg2rVrR3p6uu903z/+8Q8SExOJjY1lxIgR/Pjjj0BJkKSmphITE0NGRoavftlRUlRUFBMnTqRr16689dZbp3W8IiEn0LexVfXQbcqydetWi46ONjOzjz76yEaPHm3Hjh2z4uJiu+aaa+yTTz6xNWvW2CWXXGJ79uwxs5Lbp83Mhg0bZoMGDbLi4mJbu3at/frXvzYzswULFliDBg1sx44dVlxcbJ06dbLFixfbkSNHrHnz5rZx40YzMxs6dKg99dRT5fZpZnbzzTfb3LlzfW289dZbZmYWGRlpTzzxRA18Kv8xYcIE+/vf/+7Z/rdu3WrZ2dknrbNgwYITbnk3M5szZ4499thjXnVNToLadJuyBIdZ2dnEREURHhZGTFQUs7KzA92lGjVv3jzmzZtHYmIiSUlJbNiwgU2bNjF//nwGDRpE06ZNgZKR1XHXXXcdYWFhdOjQgd27d/vKO3bsSPPmzQkLCyMhIYH8/Hw2btxIq1atfKcGhw0bxqJFiwBYsGABl156KbGxscyfP5+1a9dW2scbbrjBq8OvVFWn6PwlPz+f11577bTe269fP02yFQVMbTArO5vMjAye3baNQjOe3baNzIyMMypkzIzf//735OXlkZeXx+bNmxk5ciRmVu7utbLK3vxgZe56q3hTRFFRUaV3xQEUFhZy2223MXv2bL744gtGjx5NYWFhpXXr1at3Ood2SlXNdSp7im758uX06NGD5ORk+vTpw65duwBIS0vjd7/7Hd27d6d9+/bk5OQwcOBA2rRpw/333w/AhAkTeOaZZ3ztZWZmMnnyZMaPH8/ixYtJSEjgqaeeYtSoUSQkJJCQkECzZs14+OGHy/UzJyeHxMREvvrqK2bMmMEdd9zhyechtYcCphbIysxkWkEBPYE6QE9gWkEBWZmZAe6Zt84991x++OEHAPr06cP06dM5dOgQUDI/59tvv6VXr168+eab7N27F4B9+/adVlvt2rUjPz+fzZs3AyXXsHr06OELk6ZNm3Lo0CHP70yrSmVznY47evQo48aNY/bs2SxfvpwRI0aQWebfxtlnn82iRYsYM2YM/fv35/nnn2fNmjXMmDGDvXv3MnLkSN9SP8eOHeP1118nPT2dxx9/nG7dupGXl8fvfvc7/vznP5OXl8ecOXM477zzGD58uK+NTz/9lDFjxjBnzhzdli4+IfWNlqFq/fbtdK1Q1rW0PJSdd955dOnShZiYGK666ipuuukmLrvsMgDq16/Pq6++SnR0NJmZmfTo0YPw8HASExOZMWPGz27rnHPO4aWXXmLw4MEUFRWRmprKmDFjqFu3LqNHjyY2NpaoqChSU1P9fJQ/TcW5Tvn5+b5tGzduZM2aNfTu3RsouYut7MTafv1KFjCPjY0lOjrat61169bs2LGDhIQEzjvvPFauXMnu3btJTEzkvPPOq7QfhYWFDB48mOeee47IyEi2bt3K+vXrycjIYN68eeVuVRdRwNQC7Vu2ZMm2bfQsU7aktDwUzMrOJqvM11xnlvma64rXAO66664T3j9s2DCGDRtWrqxiyBwf+aSlpZGWluYrf+6553zPe/XqxcqVK0/Y/yOPPMIjjzxyQnnZNsr+wvdCxdN6R44c8b02M6Kjo1m6dOlJ3xsWFlZuP2FhYRQVFQEwatQoZsyYwb/+9S/fJNvKjBkzhoEDB5a79nPhhRdSWFjIypUrFTBSjk6R1QKZWVmMjIhgAXCUkuUQRkZEkJmVVeN9mTBhAnFxcSQkJHDllVfyzTffAPDkk0/6zs/HxMQQHh7uO131zDPPEBMTQ3R09AlzY2ZlZ3PHrbeydts2dp6h15eqq23btuzZs8cXMEePHq3yRoSqDBgwgA8//JCcnBzffKaypygBnn/+eX744YcTLt43atSI999/n/vuu4+FCxdW72AkpChgaoEh6elkTZ3KuMhIznGOcZGRZE2d6vsrvybdc889rF69mry8PK699lomTpzoKz9+Af6xxx6jR48eNGnShDVr1vDiiy+ybNkyVq1axV//+lffOmcAD917L62PHqUlZ9b1JX86++yzmT17Nvfeey/x8fEkJCTw6aef/ux99OzZk//+7/8mPDwcgLi4OM466yzi4+N56qmnmDRpEl988YXvD4kpU6b43v/LX/6S9957j9tvv53PP//cr8cntVig75Ou6qF5MIG1detWa9eunY0aNco6dOhgvXv3toKCgnJ1Hn30URszZswJ7x0yZIhNnTrVzMzefPNNGzlypG/bxIkTy80XASwHLBJsD5iB3Q8GWI8ePaxVq1b2zDPPeHSUclxxcbHFx8fbl19+GeiuSDWheTBSG1R151JmZiYtWrQgOzvbN4I5rqCggA8//NC3rllMTAyLFi1i7969FBQU8MEHH/jWMJs7dy5Nzj2XH8o3y3bgF3Xr8tFHH7Fs2TIefvhhjh496vXhBlQg5zmtW7eOiy++mF69etGmTZsaa1dCnwJGqlTVnUtZWVns2LGD9PT0chfJAd577z26dOnim/DYvn177r33Xnr37k3fvn2Jj4/nrLPOoqCggKysLP7/P/6RkRERFPKf60tzzzqL/+rXj7p169K0aVPOP//8chMlQ02g5zl16NCBr776ij/+8Y810p6cORQwUqXKJiSWddNNN5WbjwHw+uuvM2TIkHJlI0eOZMWKFSxatIgmTZrQpk0btmzZwtatW/lDVhY/RESwG/gVMKZ5cy7/r/8itWPHk7YdSs7UeU4S+hQw8rOUvUA/d+5c2rVr53t98OBBPvnkE/r371/uPd9+W/JFp9u3b+cvf/kLQ4YMITY2lm+//Zb8/Hz27NlDZGQke/bsYeOOHcTGxdXMwQSJM3Wek4Q+zYORn2X8+PFs3LiRsLAwIiMjy91J9M4773DllVeesGTK9ddfz969e6lTpw7PP/88jRs3ruluB7VQn+ckZ7BA32VQ1UN3kXnvtVdftejISAtzzqIjI+21V18NdJfOSK+9+qq1ioiw+WD/BpsP1ioiQv895LQQRHeRaQRzhjp+YXlaQQFdgSXbtjEyIwMgIPNrzmTHP+9xZVYzyCqzmoFIbeVKAi/4pKSkWG5ubqC7EbJioqJ4tsJpmQXAuMhI1ni87ImIeMc5t9zMUgLdD9BF/jOWLiyLiNcUMGeo9i1bsqRCmS4si4g/KWDOUMG0gKaIhCZd5D9D6cKyiHhNF/lFREKILvKLiEjIU8CIiIgnFDAiIuIJBYyIiHhCASMiIp5QwIiIiCcUMCIi4gkFjIiIeEIBIyIinlDAiIiIJxQwIiLiiaBdi8w5twfYFuh+BJGmwHeB7kQQ0+dTNX02VQvFzybSzJoFuhMQxAEj5TnncoNlAbtgpM+navpsqqbPxls6RSYiIp5QwIiIiCcUMLXH1EB3IMjp86maPpuq6bPxkK7BiIiIJzSCERERTyhgahHn3GDn3Frn3DHnnO58AZxzfZ1zG51zm51z4wPdn2DinJvunPvWObcm0H0JNs65Fs65Bc659aX/T90V6D6FIgVM7bIGGAgsCnRHgoFzLhx4HrgK6AAMcc51CGyvgsoMoG+gOxGkioC7zaw90Am4Xf92/E8BU4uY2Xoz2xjofgSRjsBmM/vKzP4NvA70D3CfgoaZLQL2BbofwcjMdpnZitLnPwDrgYsC26vQo4CR2uwiYEeZ1zvRLwn5mZxzUUAi8HlgexJ6zgp0B6Q859zHwAWVbMo0szk13Z8g5yop022R8pM55+oDbwO/NbPvA92fUKOACTJmdkWg+1CL7ARalHndHPgmQH2RWsY5V4eScMk2s78Euj+hSKfIpDbLAdo451o5584GbgTmBrhPUgs45xwwDVhvZv8b6P6EKgVMLeKcG+Cc2wlcBrzvnPso0H0KJDMrAu4APqLkIu2bZrY2sL0KHs65WcBSoK1zbqdzbmSg+xREugBDgcudc3mlj6sD3alQo5n8IiLiCY1gRETEEwoYERHxhAJGREQ8oYARERFPKGBERMQTChgREfGEAkZERDyhgBEREU/8P47wtmHtmaiSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "a = display_closestwords(wv, 'plane')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "WebSearch1920",
   "language": "python",
   "name": "websearch1920"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}