nlp/word2vec.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Glove and Convert to W2V"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:36:04.469421Z",
     "start_time": "2020-07-31T18:32:48.213405Z"
    }
   },
   "outputs": [],
   "source": [
    "import gensim\n",
    "from gensim.models import KeyedVectors\n",
    "\n",
    "from gensim.models import Word2Vec\n",
    "from gensim.scripts.glove2word2vec import glove2word2vec\n",
    "from gensim.test.utils import datapath, get_tmpfile\n",
    "\n",
    "import os \n",
    "\n",
    "\n",
    "path_glove = os.path.abspath('glove.twitter.27B.200d.txt')\n",
    "path_w2v = os.path.abspath('glove.twitter.27B.200d_w2v.txt')\n",
    "\n",
    "\n",
    "glove_file = datapath(path_glove)\n",
    "tmp_file = get_tmpfile(path_w2v)\n",
    "\n",
    "_ = glove2word2vec(glove_file, tmp_file)\n",
    "\n",
    "\n",
    "path = os.path.abspath('glove.twitter.27B.200d_w2v.txt')\n",
    "\n",
    "model = KeyedVectors.load_word2vec_format(path, binary=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Find Most Similar Words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:39:05.893031Z",
     "start_time": "2020-07-31T18:39:03.425598Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('you', 0.8460860848426819),\n",
       " ('much', 0.7890046834945679),\n",
       " ('always', 0.7601684927940369),\n",
       " ('know', 0.75980544090271),\n",
       " ('my', 0.7519949674606323),\n",
       " ('and', 0.7513090372085571),\n",
       " ('loves', 0.7512385845184326),\n",
       " ('life', 0.7443932890892029),\n",
       " ('it', 0.7426839470863342),\n",
       " (\"n't\", 0.7408117055892944)]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.most_similar(positive=['love'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:39:08.517270Z",
     "start_time": "2020-07-31T18:39:08.463332Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('boy', 0.8434209823608398),\n",
       " ('girls', 0.8288908004760742),\n",
       " ('she', 0.8030765056610107),\n",
       " ('guy', 0.787306010723114),\n",
       " ('woman', 0.7817049622535706),\n",
       " ('chick', 0.7750228643417358),\n",
       " ('friend', 0.7702169418334961),\n",
       " ('bitch', 0.7611054182052612),\n",
       " ('that', 0.7493616938591003),\n",
       " ('pretty', 0.7465851306915283)]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.most_similar(positive=['girl'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:39:09.146095Z",
     "start_time": "2020-07-31T18:39:09.099773Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('girl', 0.8434209823608398),\n",
       " ('guy', 0.7662351131439209),\n",
       " ('kid', 0.7430558204650879),\n",
       " ('baby', 0.7402151226997375),\n",
       " ('dude', 0.7393356561660767),\n",
       " ('boys', 0.726765513420105),\n",
       " ('man', 0.7174534797668457),\n",
       " ('boi', 0.7088537216186523),\n",
       " ('nigga', 0.704058825969696),\n",
       " ('brother', 0.7035762071609497)]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.most_similar(positive=['boy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:39:09.743254Z",
     "start_time": "2020-07-31T18:39:09.694600Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('dude', 0.7253455519676208),\n",
       " ('boy', 0.7174534797668457),\n",
       " ('guy', 0.6998012065887451),\n",
       " ('shit', 0.6852256655693054),\n",
       " ('was', 0.6779443621635437),\n",
       " (\"'s\", 0.6762615442276001),\n",
       " ('bad', 0.6734011769294739),\n",
       " ('men', 0.6714873909950256),\n",
       " ('that', 0.6709873080253601),\n",
       " ('lol', 0.6646672487258911)]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.most_similar(positive=['man'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Solving Analogies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:39:25.614549Z",
     "start_time": "2020-07-31T18:39:25.222916Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           0             1\n",
      "0  australia        hotdog\n",
      "1     canada       poutine\n",
      "2    germany       hamburg\n",
      "3    ireland  cheeseburger\n",
      "4      italy       bologna\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd \n",
    "# pretty print function\n",
    "def pp(obj):\n",
    "    print(pd.DataFrame(obj))\n",
    "    \n",
    "def analogy(worda, wordb, wordc):\n",
    "    result = model.most_similar(negative=[worda], \n",
    "                                positive=[wordb, wordc])\n",
    "    return result[0][0]\n",
    "\n",
    "countries = ['australia', 'canada', 'germany', 'ireland', 'italy']\n",
    "foods = [analogy('us', 'hamburger', country) for country in countries]\n",
    "pp(zip(countries,foods))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-07-31T18:40:02.075169Z",
     "start_time": "2020-07-31T18:40:01.744837Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "def plot_data(orig_data, labels):\n",
    "    pca = PCA(n_components=2)\n",
    "    data = pca.fit_transform(orig_data)\n",
    "    plt.figure(figsize=(7, 5), dpi=100)\n",
    "    plt.plot(data[:,0], data[:,1], '.')\n",
    "    for i in range(len(data)):\n",
    "        plt.annotate(labels[i], xy = data[i])\n",
    "    for i in range(len(data)//2):\n",
    "        plt.annotate(\"\",\n",
    "                xy=data[i],\n",
    "                xytext=data[i+len(data)//2],\n",
    "                arrowprops=dict(arrowstyle=\"->\",\n",
    "                                connectionstyle=\"arc3\")\n",
    "        )\n",
    "    plt.show()\n",
    "    \n",
    "labels = countries + foods\n",
    "data = [model[w] for w in labels]\n",
    "plot_data(data, labels)"
   ]
  },
  {
   "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": "3.8.3"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autoclose": false,
   "autocomplete": true,
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}