Explore_vector_similarity_interwindows.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from gensim import corpora, models, similarities\n",
    "from scipy.spatial.distance import cosine"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## INPUT PARAMETERS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# parameters to select the topic2vec model to load\n",
    "n_docs = 11314\n",
    "n_window_t2v = 2\n",
    "random_seeds_partition = 154\n",
    "n_topics = 10\n",
    "\n",
    "i_window_sel = 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DIR WHERE THE MODEL IS SAVED"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "cwd =  os.getcwd() # Prints the working directory\n",
    "results_dir_path = cwd + '/results/20NG_lemmatiz_win20_n_topics' + str (n_topics) +'_n_doc' + str(n_docs) + '_n_win' + str(n_window_t2v)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "topic_list = []\n",
    "for i in xrange(n_topics):\n",
    "    topic_list.append('topic_' + str(i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "fname_win1 = results_dir_path + '/t2v_20NG_partSEED' + str(random_seeds_partition) + '_win' + str(i_window_sel) + '.model'\n",
    "model_win1 = models.Doc2Vec.load(fname_win1)\n",
    "        \n",
    "fname_win2 = results_dir_path + '/t2v_20NG_partSEED' + str(random_seeds_partition) + '_win' + str(i_window_sel+1) + '.model'\n",
    "model_win2 = models.Doc2Vec.load(fname_win2)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "1 - cosine(model_win1['home'], model_win2['home'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "1 - cosine(model_win1.docvecs['topic_2'], model_win1.docvecs['topic_7'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model_win1.docvecs.most_similar('topic_6')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model_win2.docvecs.most_similar('topic_6')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SHOW THE WORDS CLOSEST TO THE TOPIC SELECTED"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dist = []\n",
    "t2v_voc = []\n",
    "\n",
    "for wordVec in model_win1.syn0:\n",
    "    dist.append(cosine(wordVec, model_win1.docvecs['topic_6']))\n",
    " \n",
    "index_dist_sorted = sorted(range(len(dist)), key=lambda k: dist[k])\n",
    "for i in xrange(20):\n",
    "    t2v_voc.append(model_win1.index2word[index_dist_sorted[i]])\n",
    "    print(model_win1.index2word[index_dist_sorted[i]])\n",
    "#print(t2v_voc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dist = []\n",
    "t2v_voc = []\n",
    "\n",
    "for wordVec in model_win2.syn0:\n",
    "    dist.append(cosine(wordVec, model_win2.docvecs['topic_6']))\n",
    " \n",
    "index_dist_sorted2 = sorted(range(len(dist)), key=lambda k: dist[k])\n",
    "for i in xrange(20):\n",
    "    t2v_voc.append(model_win2.index2word[index_dist_sorted2[i]])\n",
    "    print(model_win2.index2word[index_dist_sorted2[i]])\n",
    "#print(t2v_voc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model_win1['home']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "model_win2.most_similar('jesus')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model_win1.vector_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model_win1.dbow_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}