07_Create_Corpus_LDA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from gensim import corpora, models, similarities\n",
    "from collections import defaultdict\n",
    "import re\n",
    "import pandas as pd\n",
    "import logging\n",
    "\n",
    "logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "documents = pd.read_pickle('gensim/preprocessed_text_v3.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "dictionary = corpora.Dictionary.load('gensim/dictionary.dict')\n",
    "corpus_tfidf = corpora.MmCorpus('gensim/corpus_tfidf.mm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:gensim.models.ldamodel:too few updates, training might not converge; consider increasing the number of passes or iterations to improve accuracy\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2min 40s, sys: 5.55 s, total: 2min 46s\n",
      "Wall time: 2min 44s\n"
     ]
    }
   ],
   "source": [
    "# lda model on tfidf\n",
    "%time lda = models.LdaModel(corpus_tfidf, num_topics = 100, id2word=dictionary)\n",
    "lda.save('gensim/model.lda')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:gensim.similarities.docsim:scanning corpus to determine the number of features (consider setting `num_features` explicitly)\n"
     ]
    }
   ],
   "source": [
    "# calculate similarities\n",
    "index = similarities.MatrixSimilarity(lda[corpus_tfidf])\n",
    "index.save('gensim/tfidf_lda_similarities.index')"
   ]
  },
  {
   "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": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}