IntegrationTest_txt_gensim.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:35.412321. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "from dariah_topics import preprocessing as pre\n",
    "from dariah_topics import visualization as visual\n",
    "from dariah_topics import evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Liste mit Dateinamen erzeugen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['corpus_txt/Lovecraft_AttheMountainofMadness.txt',\n",
       " 'corpus_txt/Howard_TheDevilinIron.txt',\n",
       " 'corpus_txt/Poe_ThePurloinedLetter.txt',\n",
       " 'corpus_txt/Lovecraft_TheShunnedHouse.txt',\n",
       " 'corpus_txt/Poe_TheMasqueoftheRedDeath.txt']"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.243248. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "path_txt = \"corpus_txt\"\n",
    "#path_txt = \"grenzbote_plain/*/\"\n",
    "#path_txt = \"wiki/\"\n",
    "\n",
    "doclist_txt = pre.create_document_list(path_txt)\n",
    "assert doclist_txt, \"No documents found\"\n",
    "doclist_txt[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#####  Liste mit Dokumentenlabels erzeugen - (Funktion wird durch Thorsten's generischere Funktion ersetzt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Lovecraft_AttheMountainofMadness',\n",
       " 'Howard_TheDevilinIron',\n",
       " 'Poe_ThePurloinedLetter',\n",
       " 'Lovecraft_TheShunnedHouse',\n",
       " 'Poe_TheMasqueoftheRedDeath']"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.252114. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "doc_labels = list(pre.get_labels(doclist_txt))\n",
    "doc_labels[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Corpus laden"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.267213. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "corpus_txt = pre.read_from_txt(doclist_txt)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Tokenisieren"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.275856. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "doc_tokens = [list(pre.tokenize(txt)) for txt in list(corpus_txt)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Create Dictionaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.560498. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "id_types, doc_ids = pre.create_dictionaries(doc_labels, doc_tokens)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Sparse BOW Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.622170. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "sparse_bow = pre.create_mm(doc_labels, doc_tokens, id_types, doc_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>doc_id</th>\n",
       "      <th>token_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">1</th>\n",
       "      <th>16384</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16385</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16387</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16388</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16389</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0\n",
       "doc_id token_id   \n",
       "1      16384     1\n",
       "       16385     8\n",
       "       16387     3\n",
       "       16388     1\n",
       "       16389     1"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:36.990760. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "sparse_bow[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Topic Modeling with Gensim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Saving Sparse BOW"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.007429. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "pre.save_bow_mm(sparse_bow, \"gensim_txt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Import Market Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.136271. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "from gensim.models import LdaModel\n",
    "from gensim.corpora import MmCorpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.141019. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "mm = MmCorpus(\"gensim_txt.mm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.158584. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "doc2id = {value : key for key, value in doc_ids.items()}\n",
    "type2id = {value : key for key, value in id_types.items()}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Remove Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "type2id = {value : key for key, value in id_types.items()}\n",
    "sparse_bow_collapsed = sparse_bow.groupby(sparse_bow.index.get_level_values('token_id')).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "sparse_bow_hapax = sparse_bow_collapsed.loc[sparse_bow_collapsed[0] == 1]\n",
    "hapax = [type2id[key] for key in sparse_bow_hapax.index.get_level_values('token_id')]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "len(hapax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.172883. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "import os.path\n",
    "basepath = os.path.abspath('.')\n",
    "\n",
    "with open(os.path.join(basepath, \"tutorial_supplementals\", \"stopwords\", \"de.txt\"), 'r', encoding = 'utf-8') as f: \n",
    "    stopword_list = f.read().split('\\n')\n",
    "    \n",
    "stopword_list = set(stopword_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.184775. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "hapax_from_remove = pre.find_hapax(sparse_bow, id_types)\n",
    "stopwords_from_remove = pre.find_stopwords(sparse_bow, id_types, mfw=75)\n",
    "\n",
    "#features_to_be_removed = set(hapax_from_remove + stopwords_from_remove)\n",
    "features_to_be_removed = stopwords_from_remove\n",
    "\n",
    "sparse_bow_short = pre.remove_features(sparse_bow, id_types, features_to_be_removed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "75"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.328330. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "len(features_to_be_removed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.334179. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "pre.save_bow_mm(sparse_bow_short, \"gensim_txt\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.421797. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "mm = MmCorpus(\"gensim_txt.mm\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Convert sparse_bow to list of (doc, tokens) tuples (like doc2bow)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.425956. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "doc2bow_list = []\n",
    "\n",
    "for doc in sparse_bow_short.index.groupby(sparse_bow_short.index.get_level_values('doc_id')):\n",
    "    temp = [(token, count) for token, count in zip(sparse_bow_short.loc[doc].index, sparse_bow_short.loc[doc][0])]\n",
    "    doc2bow_list.append(temp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.510798. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "len(doc2bow_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:46:37.517427. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "#model = LdaModel(corpus=mm, id2word=type2id, num_topics=60, alpha = \"symmetric\", passes = 10) #import momentan in visual \n",
    "# -> da ich mir noch nicht sicher bin, welche Funktionen in das tm_gensim.py sollen\n",
    "model = LdaModel(corpus=mm, id2word=type2id, num_topics=20, passes = 10, iterations = 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(1, 0.034670551185678529),\n",
       " (2, 0.033904241123165951),\n",
       " (3, 0.11535209760158363),\n",
       " (4, 0.063366544282656062),\n",
       " (5, 0.035226801353724177),\n",
       " (6, 0.01702470134459878),\n",
       " (7, 0.040311905089352403),\n",
       " (8, 0.24180638719670738),\n",
       " (10, 0.23580797164637937),\n",
       " (15, 0.094649854885207579),\n",
       " (18, 0.087808289556959185)]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.590690. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "model.get_document_topics(doc2bow_list[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['converged',\n",
       " 'cellar',\n",
       " 'letzten',\n",
       " 'one-dash',\n",
       " 'inconceivable',\n",
       " 'guts-of',\n",
       " 'immemorial',\n",
       " 'seventy',\n",
       " 'rescuers',\n",
       " 'refolded']"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.603001. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "#Anzeige der keywords für topic n\n",
    "n = 1\n",
    "topic_nr_x = model.get_topic_terms(n)\n",
    "\n",
    "topicTerms = [type2id[i[0]] for i in topic_nr_x]\n",
    "topicTerms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0,\n",
       "  '0.000*\"supererogatory\" + 0.000*\"cannon-ball\" + 0.000*\"truce\" + 0.000*\"blaze\" + 0.000*\"resemblance\" + 0.000*\"lull\" + 0.000*\"craftily\" + 0.000*\"motley\" + 0.000*\"near\" + 0.000*\"chimney-piece\"'),\n",
       " (1,\n",
       "  '0.005*\"converged\" + 0.005*\"cellar\" + 0.005*\"letzten\" + 0.004*\"one-dash\" + 0.004*\"inconceivable\" + 0.003*\"guts-of\" + 0.003*\"immemorial\" + 0.003*\"seventy\" + 0.003*\"rescuers\" + 0.003*\"refolded\"'),\n",
       " (2,\n",
       "  '0.010*\"flattened\" + 0.005*\"near\" + 0.005*\"guessed\" + 0.005*\"clubbed\" + 0.004*\"acquainted\" + 0.004*\"converged\" + 0.004*\"hurrying\" + 0.004*\"bachelor\" + 0.004*\"mitra\" + 0.003*\"throat-muscles\"'),\n",
       " (3,\n",
       "  '0.007*\"flattened\" + 0.006*\"celebrated\" + 0.004*\"debris\" + 0.004*\"near\" + 0.004*\"blaze\" + 0.004*\"badly-trimmed\" + 0.003*\"clubbed\" + 0.003*\"presumed\" + 0.003*\"vaulted\" + 0.003*\"monthly\"'),\n",
       " (4,\n",
       "  '0.008*\"generated\" + 0.005*\"bristling\" + 0.005*\"inky\" + 0.005*\"badly-trimmed\" + 0.005*\"fernandez\" + 0.005*\"incrustation\" + 0.005*\"alice\" + 0.004*\"angell\" + 0.004*\"varnished\" + 0.004*\"humiliation\"'),\n",
       " (5,\n",
       "  '0.006*\"hoards\" + 0.005*\"monkey\" + 0.004*\"near\" + 0.004*\"chosen\" + 0.004*\"aaarh\" + 0.004*\"clubbed\" + 0.004*\"raj\" + 0.004*\"ts\" + 0.003*\"cæsar’s\" + 0.003*\"blaze\"'),\n",
       " (6,\n",
       "  '0.009*\"motley\" + 0.005*\"waiter\" + 0.004*\"chimney-piece\" + 0.004*\"visage\" + 0.004*\"frankland\" + 0.004*\"mitral\" + 0.004*\"lull\" + 0.004*\"resemblance\" + 0.003*\"grouped\" + 0.003*\"forty-pounder\"'),\n",
       " (7,\n",
       "  '0.015*\"instructive\" + 0.011*\"pips\" + 0.010*\"amuses\" + 0.008*\"supererogatory\" + 0.008*\"tortuously\" + 0.008*\"bandanna\" + 0.008*\"sssso\" + 0.007*\"lull\" + 0.007*\"blaze\" + 0.006*\"puncheons\"'),\n",
       " (8,\n",
       "  '0.004*\"elkanah\" + 0.004*\"cannon-ball\" + 0.004*\"converged\" + 0.004*\"throat-muscles\" + 0.004*\"crete\" + 0.003*\"egg\" + 0.003*\"yorkshire\" + 0.003*\"labyrinths\" + 0.003*\"near\" + 0.003*\"blaze\"'),\n",
       " (9,\n",
       "  '0.000*\"badly-trimmed\" + 0.000*\"near\" + 0.000*\"supererogatory\" + 0.000*\"converged\" + 0.000*\"throat-muscles\" + 0.000*\"humiliation\" + 0.000*\"blaze\" + 0.000*\"varnished\" + 0.000*\"millet\" + 0.000*\"craftily\"'),\n",
       " (10,\n",
       "  '0.004*\"motley\" + 0.004*\"craftily\" + 0.004*\"truce\" + 0.004*\"millet\" + 0.003*\"raj\" + 0.003*\"cannon-ball\" + 0.003*\"guts-of\" + 0.003*\"bridges\" + 0.003*\"waved\" + 0.003*\"mustache\"'),\n",
       " (11,\n",
       "  '0.001*\"sake\" + 0.001*\"opposable\" + 0.001*\"wiser\" + 0.001*\"motley\" + 0.000*\"supererogatory\" + 0.000*\"blaze\" + 0.000*\"grouped\" + 0.000*\"humiliation\" + 0.000*\"raj\" + 0.000*\"varnished\"'),\n",
       " (12,\n",
       "  '0.000*\"badly-trimmed\" + 0.000*\"near\" + 0.000*\"fernandez\" + 0.000*\"truce\" + 0.000*\"neglected\" + 0.000*\"resemblance\" + 0.000*\"varnished\" + 0.000*\"blaze\" + 0.000*\"supererogatory\" + 0.000*\"bridges\"'),\n",
       " (13,\n",
       "  '0.000*\"neglected\" + 0.000*\"chewing\" + 0.000*\"supererogatory\" + 0.000*\"blaze\" + 0.000*\"near\" + 0.000*\"cannon-ball\" + 0.000*\"humiliation\" + 0.000*\"waiter\" + 0.000*\"millet\" + 0.000*\"raj\"'),\n",
       " (14,\n",
       "  '0.000*\"cannon-ball\" + 0.000*\"neglected\" + 0.000*\"blaze\" + 0.000*\"inconceivable\" + 0.000*\"converged\" + 0.000*\"craftily\" + 0.000*\"bridges\" + 0.000*\"truce\" + 0.000*\"motley\" + 0.000*\"undertook\"'),\n",
       " (15,\n",
       "  '0.009*\"converged\" + 0.006*\"inconceivable\" + 0.005*\"cannon-ball\" + 0.005*\"undertook\" + 0.005*\"clinical\" + 0.005*\"eastern\" + 0.004*\"watchful\" + 0.004*\"visage\" + 0.004*\"wented\" + 0.004*\"recognizing\"'),\n",
       " (16,\n",
       "  '0.000*\"blaze\" + 0.000*\"raj\" + 0.000*\"bridges\" + 0.000*\"millet\" + 0.000*\"grouped\" + 0.000*\"converged\" + 0.000*\"labyrinths\" + 0.000*\"motley\" + 0.000*\"craftily\" + 0.000*\"badly-trimmed\"'),\n",
       " (17,\n",
       "  '0.000*\"truce\" + 0.000*\"blaze\" + 0.000*\"craftily\" + 0.000*\"grouped\" + 0.000*\"alice\" + 0.000*\"varnished\" + 0.000*\"badly-trimmed\" + 0.000*\"near\" + 0.000*\"supererogatory\" + 0.000*\"raj\"'),\n",
       " (18,\n",
       "  '0.013*\"wiser\" + 0.007*\"motley\" + 0.005*\"sea-fog\" + 0.005*\"asserting\" + 0.005*\"indicated\" + 0.004*\"resemblance\" + 0.004*\"fodder\" + 0.004*\"millet\" + 0.004*\"blaze\" + 0.004*\"approach\"'),\n",
       " (19,\n",
       "  '0.000*\"cannon-ball\" + 0.000*\"converged\" + 0.000*\"egg\" + 0.000*\"grouped\" + 0.000*\"inconceivable\" + 0.000*\"truce\" + 0.000*\"elkanah\" + 0.000*\"motley\" + 0.000*\"neglected\" + 0.000*\"humiliation\"')]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.622398. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "#Topic probabilities\n",
    "topics = model.show_topics(num_topics = 40)\n",
    "topics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.642119. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import regex as re\n",
    "def gensim2dataframe(model):\n",
    "    num_topics = model.num_topics\n",
    "    topics_df = pd.DataFrame(index = range(num_topics), columns= range(10))\n",
    "\n",
    "    topics = model.show_topics(num_topics)\n",
    "    \n",
    "    for topic_dist in topics:    \n",
    "        idx = topic_dist[0]\n",
    "        temp = re.findall(r'\\\"(.+?)\\\"', topics[idx][1])\n",
    "        topics_df.loc[idx] = temp\n",
    "    \n",
    "    return topics_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.656176. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "topics_df = gensim2dataframe(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>supererogatory</td>\n",
       "      <td>converged</td>\n",
       "      <td>flattened</td>\n",
       "      <td>flattened</td>\n",
       "      <td>generated</td>\n",
       "      <td>hoards</td>\n",
       "      <td>motley</td>\n",
       "      <td>instructive</td>\n",
       "      <td>elkanah</td>\n",
       "      <td>badly-trimmed</td>\n",
       "      <td>motley</td>\n",
       "      <td>sake</td>\n",
       "      <td>badly-trimmed</td>\n",
       "      <td>neglected</td>\n",
       "      <td>cannon-ball</td>\n",
       "      <td>converged</td>\n",
       "      <td>blaze</td>\n",
       "      <td>truce</td>\n",
       "      <td>wiser</td>\n",
       "      <td>cannon-ball</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>cannon-ball</td>\n",
       "      <td>cellar</td>\n",
       "      <td>near</td>\n",
       "      <td>celebrated</td>\n",
       "      <td>bristling</td>\n",
       "      <td>monkey</td>\n",
       "      <td>waiter</td>\n",
       "      <td>pips</td>\n",
       "      <td>cannon-ball</td>\n",
       "      <td>near</td>\n",
       "      <td>craftily</td>\n",
       "      <td>opposable</td>\n",
       "      <td>near</td>\n",
       "      <td>chewing</td>\n",
       "      <td>neglected</td>\n",
       "      <td>inconceivable</td>\n",
       "      <td>raj</td>\n",
       "      <td>blaze</td>\n",
       "      <td>motley</td>\n",
       "      <td>converged</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>truce</td>\n",
       "      <td>letzten</td>\n",
       "      <td>guessed</td>\n",
       "      <td>debris</td>\n",
       "      <td>inky</td>\n",
       "      <td>near</td>\n",
       "      <td>chimney-piece</td>\n",
       "      <td>amuses</td>\n",
       "      <td>converged</td>\n",
       "      <td>supererogatory</td>\n",
       "      <td>truce</td>\n",
       "      <td>wiser</td>\n",
       "      <td>fernandez</td>\n",
       "      <td>supererogatory</td>\n",
       "      <td>blaze</td>\n",
       "      <td>cannon-ball</td>\n",
       "      <td>bridges</td>\n",
       "      <td>craftily</td>\n",
       "      <td>sea-fog</td>\n",
       "      <td>egg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>blaze</td>\n",
       "      <td>one-dash</td>\n",
       "      <td>clubbed</td>\n",
       "      <td>near</td>\n",
       "      <td>badly-trimmed</td>\n",
       "      <td>chosen</td>\n",
       "      <td>visage</td>\n",
       "      <td>supererogatory</td>\n",
       "      <td>throat-muscles</td>\n",
       "      <td>converged</td>\n",
       "      <td>millet</td>\n",
       "      <td>motley</td>\n",
       "      <td>truce</td>\n",
       "      <td>blaze</td>\n",
       "      <td>inconceivable</td>\n",
       "      <td>undertook</td>\n",
       "      <td>millet</td>\n",
       "      <td>grouped</td>\n",
       "      <td>asserting</td>\n",
       "      <td>grouped</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>resemblance</td>\n",
       "      <td>inconceivable</td>\n",
       "      <td>acquainted</td>\n",
       "      <td>blaze</td>\n",
       "      <td>fernandez</td>\n",
       "      <td>aaarh</td>\n",
       "      <td>frankland</td>\n",
       "      <td>tortuously</td>\n",
       "      <td>crete</td>\n",
       "      <td>throat-muscles</td>\n",
       "      <td>raj</td>\n",
       "      <td>supererogatory</td>\n",
       "      <td>neglected</td>\n",
       "      <td>near</td>\n",
       "      <td>converged</td>\n",
       "      <td>clinical</td>\n",
       "      <td>grouped</td>\n",
       "      <td>alice</td>\n",
       "      <td>indicated</td>\n",
       "      <td>inconceivable</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>lull</td>\n",
       "      <td>guts-of</td>\n",
       "      <td>converged</td>\n",
       "      <td>badly-trimmed</td>\n",
       "      <td>incrustation</td>\n",
       "      <td>clubbed</td>\n",
       "      <td>mitral</td>\n",
       "      <td>bandanna</td>\n",
       "      <td>egg</td>\n",
       "      <td>humiliation</td>\n",
       "      <td>cannon-ball</td>\n",
       "      <td>blaze</td>\n",
       "      <td>resemblance</td>\n",
       "      <td>cannon-ball</td>\n",
       "      <td>craftily</td>\n",
       "      <td>eastern</td>\n",
       "      <td>converged</td>\n",
       "      <td>varnished</td>\n",
       "      <td>resemblance</td>\n",
       "      <td>truce</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>craftily</td>\n",
       "      <td>immemorial</td>\n",
       "      <td>hurrying</td>\n",
       "      <td>clubbed</td>\n",
       "      <td>alice</td>\n",
       "      <td>raj</td>\n",
       "      <td>lull</td>\n",
       "      <td>sssso</td>\n",
       "      <td>yorkshire</td>\n",
       "      <td>blaze</td>\n",
       "      <td>guts-of</td>\n",
       "      <td>grouped</td>\n",
       "      <td>varnished</td>\n",
       "      <td>humiliation</td>\n",
       "      <td>bridges</td>\n",
       "      <td>watchful</td>\n",
       "      <td>labyrinths</td>\n",
       "      <td>badly-trimmed</td>\n",
       "      <td>fodder</td>\n",
       "      <td>elkanah</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>motley</td>\n",
       "      <td>seventy</td>\n",
       "      <td>bachelor</td>\n",
       "      <td>presumed</td>\n",
       "      <td>angell</td>\n",
       "      <td>ts</td>\n",
       "      <td>resemblance</td>\n",
       "      <td>lull</td>\n",
       "      <td>labyrinths</td>\n",
       "      <td>varnished</td>\n",
       "      <td>bridges</td>\n",
       "      <td>humiliation</td>\n",
       "      <td>blaze</td>\n",
       "      <td>waiter</td>\n",
       "      <td>truce</td>\n",
       "      <td>visage</td>\n",
       "      <td>motley</td>\n",
       "      <td>near</td>\n",
       "      <td>millet</td>\n",
       "      <td>motley</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>near</td>\n",
       "      <td>rescuers</td>\n",
       "      <td>mitra</td>\n",
       "      <td>vaulted</td>\n",
       "      <td>varnished</td>\n",
       "      <td>cæsar’s</td>\n",
       "      <td>grouped</td>\n",
       "      <td>blaze</td>\n",
       "      <td>near</td>\n",
       "      <td>millet</td>\n",
       "      <td>waved</td>\n",
       "      <td>raj</td>\n",
       "      <td>supererogatory</td>\n",
       "      <td>millet</td>\n",
       "      <td>motley</td>\n",
       "      <td>wented</td>\n",
       "      <td>craftily</td>\n",
       "      <td>supererogatory</td>\n",
       "      <td>blaze</td>\n",
       "      <td>neglected</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>chimney-piece</td>\n",
       "      <td>refolded</td>\n",
       "      <td>throat-muscles</td>\n",
       "      <td>monthly</td>\n",
       "      <td>humiliation</td>\n",
       "      <td>blaze</td>\n",
       "      <td>forty-pounder</td>\n",
       "      <td>puncheons</td>\n",
       "      <td>blaze</td>\n",
       "      <td>craftily</td>\n",
       "      <td>mustache</td>\n",
       "      <td>varnished</td>\n",
       "      <td>bridges</td>\n",
       "      <td>raj</td>\n",
       "      <td>undertook</td>\n",
       "      <td>recognizing</td>\n",
       "      <td>badly-trimmed</td>\n",
       "      <td>raj</td>\n",
       "      <td>approach</td>\n",
       "      <td>humiliation</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               0              1               2              3   \\\n",
       "0  supererogatory      converged       flattened      flattened   \n",
       "1     cannon-ball         cellar            near     celebrated   \n",
       "2           truce        letzten         guessed         debris   \n",
       "3           blaze       one-dash         clubbed           near   \n",
       "4     resemblance  inconceivable      acquainted          blaze   \n",
       "5            lull        guts-of       converged  badly-trimmed   \n",
       "6        craftily     immemorial        hurrying        clubbed   \n",
       "7          motley        seventy        bachelor       presumed   \n",
       "8            near       rescuers           mitra        vaulted   \n",
       "9   chimney-piece       refolded  throat-muscles        monthly   \n",
       "\n",
       "              4        5              6               7               8   \\\n",
       "0      generated   hoards         motley     instructive         elkanah   \n",
       "1      bristling   monkey         waiter            pips     cannon-ball   \n",
       "2           inky     near  chimney-piece          amuses       converged   \n",
       "3  badly-trimmed   chosen         visage  supererogatory  throat-muscles   \n",
       "4      fernandez    aaarh      frankland      tortuously           crete   \n",
       "5   incrustation  clubbed         mitral        bandanna             egg   \n",
       "6          alice      raj           lull           sssso       yorkshire   \n",
       "7         angell       ts    resemblance            lull      labyrinths   \n",
       "8      varnished  cæsar’s        grouped           blaze            near   \n",
       "9    humiliation    blaze  forty-pounder       puncheons           blaze   \n",
       "\n",
       "               9            10              11              12  \\\n",
       "0   badly-trimmed       motley            sake   badly-trimmed   \n",
       "1            near     craftily       opposable            near   \n",
       "2  supererogatory        truce           wiser       fernandez   \n",
       "3       converged       millet          motley           truce   \n",
       "4  throat-muscles          raj  supererogatory       neglected   \n",
       "5     humiliation  cannon-ball           blaze     resemblance   \n",
       "6           blaze      guts-of         grouped       varnished   \n",
       "7       varnished      bridges     humiliation           blaze   \n",
       "8          millet        waved             raj  supererogatory   \n",
       "9        craftily     mustache       varnished         bridges   \n",
       "\n",
       "               13             14             15             16  \\\n",
       "0       neglected    cannon-ball      converged          blaze   \n",
       "1         chewing      neglected  inconceivable            raj   \n",
       "2  supererogatory          blaze    cannon-ball        bridges   \n",
       "3           blaze  inconceivable      undertook         millet   \n",
       "4            near      converged       clinical        grouped   \n",
       "5     cannon-ball       craftily        eastern      converged   \n",
       "6     humiliation        bridges       watchful     labyrinths   \n",
       "7          waiter          truce         visage         motley   \n",
       "8          millet         motley         wented       craftily   \n",
       "9             raj      undertook    recognizing  badly-trimmed   \n",
       "\n",
       "               17           18             19  \n",
       "0           truce        wiser    cannon-ball  \n",
       "1           blaze       motley      converged  \n",
       "2        craftily      sea-fog            egg  \n",
       "3         grouped    asserting        grouped  \n",
       "4           alice    indicated  inconceivable  \n",
       "5       varnished  resemblance          truce  \n",
       "6   badly-trimmed       fodder        elkanah  \n",
       "7            near       millet         motley  \n",
       "8  supererogatory        blaze      neglected  \n",
       "9             raj     approach    humiliation  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.681619. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "doc_topic = topics_df.T\n",
    "doc_topic"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Visualisierung "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"lda_model = 'out_easy/corpus.lda'\\ncorpus = 'out_easy/corpus.mm'\\ndictionary = 'out_easy/corpus.dict'\\ndoc_labels = 'out_easy/corpus_doclabels.txt'\\ninteractive  = False\\n\\nvis = visual.Visualization(lda_model, corpus, dictionary, doc_labels, interactive)\""
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.715104. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "'''lda_model = 'out_easy/corpus.lda'\n",
    "corpus = 'out_easy/corpus.mm'\n",
    "dictionary = 'out_easy/corpus.dict'\n",
    "doc_labels = 'out_easy/corpus_doclabels.txt'\n",
    "interactive  = False\n",
    "\n",
    "vis = visual.Visualization(lda_model, corpus, dictionary, doc_labels, interactive)'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Create a document-topic matrix (that is a pandas data frame actually)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.725233. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "doc_topic = visual.create_doc_topic(mm, model, doc_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Visualize document-topic matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from '/home/sina/.local/lib/python3.5/site-packages/matplotlib/pyplot.py'>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEaCAYAAABEsMO+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f69b2358c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:21.967807. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "visual.doc_topic_heatmap(doc_topic)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Visualize topic distribution in a single document"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from '/home/sina/.local/lib/python3.5/site-packages/matplotlib/pyplot.py'>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noCRJXTKgJEldMqAkSV36/+lxMLOJGmETAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f69b2e1ca20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3.5/json/encoder.py:198: DeprecationWarning: Interpreting naive datetime as local 2017-04-02 13:47:22.305142. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "visual.plot_doc_topics(doc_topic, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "![success](http://cdn2.hubspot.net/hub/128506/file-446943132-jpg/images/computer_woman_success.jpg)"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}