IntegrationTest_v01.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\gensim\\utils.py:855: UserWarning: detected Windows; aliasing chunkize to chunkize_serial\n",
      "  warnings.warn(\"detected Windows; aliasing chunkize to chunkize_serial\")\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\funcy\\decorators.py:56: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  spec = inspect.getargspec(func)\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\funcy\\decorators.py:56: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  spec = inspect.getargspec(func)\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\funcy\\decorators.py:56: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  spec = inspect.getargspec(func)\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\funcy\\decorators.py:56: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  spec = inspect.getargspec(func)\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\funcy\\decorators.py:56: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  spec = inspect.getargspec(func)\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\funcy\\decorators.py:56: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  spec = inspect.getargspec(func)\n"
     ]
    }
   ],
   "source": [
    "from dariah_topics import preprocessing as pre\n",
    "from dariah_topics import visualization as visual\n",
    "from dariah_topics import mallet as mal\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": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:92: DeprecationWarning: DisplayFormatter._ipython_display_formatter_default is deprecated: use @default decorator instead.\n",
      "  def _ipython_display_formatter_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:98: DeprecationWarning: DisplayFormatter._formatters_default is deprecated: use @default decorator instead.\n",
      "  def _formatters_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:677: DeprecationWarning: PlainTextFormatter._deferred_printers_default is deprecated: use @default decorator instead.\n",
      "  def _deferred_printers_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:669: DeprecationWarning: PlainTextFormatter._singleton_printers_default is deprecated: use @default decorator instead.\n",
      "  def _singleton_printers_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:672: DeprecationWarning: PlainTextFormatter._type_printers_default is deprecated: use @default decorator instead.\n",
      "  def _type_printers_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:669: DeprecationWarning: PlainTextFormatter._singleton_printers_default is deprecated: use @default decorator instead.\n",
      "  def _singleton_printers_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:672: DeprecationWarning: PlainTextFormatter._type_printers_default is deprecated: use @default decorator instead.\n",
      "  def _type_printers_default(self):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\IPython\\core\\formatters.py:677: DeprecationWarning: PlainTextFormatter._deferred_printers_default is deprecated: use @default decorator instead.\n",
      "  def _deferred_printers_default(self):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['corpus_txt\\\\Doyle_AScandalinBohemia.txt',\n",
       " 'corpus_txt\\\\Doyle_AStudyinScarlet.txt',\n",
       " 'corpus_txt\\\\Doyle_TheHoundoftheBaskervilles.txt',\n",
       " 'corpus_txt\\\\Doyle_TheSignoftheFour.txt',\n",
       " 'corpus_txt\\\\Howard_GodsoftheNorth.txt']"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path_txt = \"corpus_txt\"\n",
    "#path_txt = \"grenzbote_plain/*/\"\n",
    "\n",
    "doclist_txt = pre.create_document_list(path_txt)\n",
    "assert doclist_txt, \"No documents found\"\n",
    "doclist_txt[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['corpus_csv\\\\Doyle_AStudyinScarlet.txt.csv',\n",
       " 'corpus_csv\\\\Doyle_TheHoundoftheBaskervilles.txt.csv',\n",
       " 'corpus_csv\\\\Doyle_TheSignoftheFour.txt.csv',\n",
       " 'corpus_csv\\\\Howard_GodsoftheNorth.txt.csv',\n",
       " 'corpus_csv\\\\Howard_SchadowsinZamboula.txt.csv']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path_csv = \"corpus_csv\"\n",
    "\n",
    "doclist_csv = pre.create_document_list(path_csv, 'csv')\n",
    "doclist_csv[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['corpus_tei\\\\Schnitzler_Amerika.xml',\n",
       " 'corpus_tei\\\\Schnitzler_DerAndere.xml',\n",
       " 'corpus_tei\\\\Schnitzler_DerWitwer.xml',\n",
       " 'corpus_tei\\\\Schnitzler_DieFremde.xml',\n",
       " 'corpus_tei\\\\Schnitzler_GeschichteeinesGenies.xml']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path_tei = \"corpus_tei\"\n",
    "\n",
    "doclist_tei = pre.create_document_list(path_tei, 'xml')\n",
    "doclist_tei[: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": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['corpus_txt\\\\Doyle_AScandalinBohemia.txt',\n",
       " 'corpus_txt\\\\Doyle_AStudyinScarlet.txt',\n",
       " 'corpus_txt\\\\Doyle_TheHoundoftheBaskervilles.txt',\n",
       " 'corpus_txt\\\\Doyle_TheSignoftheFour.txt',\n",
       " 'corpus_txt\\\\Howard_GodsoftheNorth.txt']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "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": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "corpus_txt = pre.read_from_txt(doclist_txt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "corpus_csv = pre.read_from_csv(doclist_csv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "corpus_tei = pre.read_from_tei(doclist_tei)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Tokenisieren"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "37    typographical\n",
       "56          textual\n",
       "59           square\n",
       "72              old\n",
       "75             such\n",
       "Name: Lemma, dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_tokens = pre.filter_POS_tags(corpus_csv)\n",
    "list(doc_tokens)[0][:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['scandal', 'in', 'bohemia', 'conan', 'doyle']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_tokens = [list(pre.tokenize(txt)) for txt in list(corpus_txt)]\n",
    "doc_tokens[0][:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['arthur', 'schnitzler', 'amerika', 'das', 'schiff']"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_tokens = [list(pre.tokenize(tei)) for tei in list(corpus_tei)]\n",
    "doc_tokens[0][:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Create Dictionaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "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": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sparse_bow = pre.create_mm(doc_labels, doc_tokens, id_types, doc_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>4</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1367</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0\n",
       "doc_id token_id    \n",
       "1      4          1\n",
       "       6          1\n",
       "       7          1\n",
       "       1367       2\n",
       "       13        14"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sparse_bow[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Convert to Doc-Topic Matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "source": [
    "'''import pandas as pd\n",
    "\n",
    "doc2id = {value : key for key, value in doc_ids.items()}\n",
    "type2id = {value : key for key, value in id_types.items()}\n",
    "\n",
    "cols = [doc2id[key] for key in set(sparse_bow.index.get_level_values(\"doc_id\"))]\n",
    "#idx = [type2id[key] for key in set(sparse_bow.index.get_level_values(\"token_id\"))]\n",
    "\n",
    "set(sparse_bow.index.get_level_values(\"token_id\"))\n",
    "\n",
    "#doctopic_matrix = pd.DataFrame(columns=cols, index=idx)'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Remove Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import os.path\n",
    "basepath = os.path.abspath('.')\n",
    "\n",
    "with open(os.path.join(basepath, \"tutorial_supplementals\", \"stopwords\", \"en.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": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sparse_df_stopwords_removed = pre.remove_features(sparse_bow, id_types, stopword_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3649\n",
      "3605\n"
     ]
    }
   ],
   "source": [
    "print(len(sparse_bow))\n",
    "print(len(sparse_df_stopwords_removed))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Create Feature Remove Lists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stopword_list = pre.find_stopwords(sparse_bow, id_types, 100)\n",
    "len(stopword_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1715"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hapax_list = pre.find_hapax(sparse_bow, id_types)\n",
    "len(hapax_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "##### Remove Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "feature_list = set(stopword_list).union(hapax_list)\n",
    "clean_term_frequency = pre.remove_features(sparse_bow, id_types, feature_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1576"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(clean_term_frequency)"
   ]
  },
  {
   "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": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#num_docs = max(sparse_bow.index.get_level_values(\"doc_id\"))\n",
    "#num_types = max(sparse_bow.index.get_level_values(\"token_id\"))\n",
    "#sum_counts = sum(sparse_bow[0])\n",
    "\n",
    "#header_string = str(num_docs) + \" \" + str(num_types) + \" \" + str(sum_counts) + \"\\n\"\n",
    "\n",
    "#with open(\"gb_plain.mm\", 'w', encoding = \"utf-8\") as f:\n",
    "#    pass\n",
    "\n",
    "#with open(\"gb_plain.mm\", 'a', encoding = \"utf-8\") as f:\n",
    "#    f.write(\"%%MatrixMarket matrix coordinate real general\\n\")\n",
    "#    f.write(header_string)\n",
    "#    sparse_bow.to_csv( f, sep = ' ', header = None)\n",
    "\n",
    "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": 24,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from gensim.models import LdaModel\n",
    "from gensim.corpora import MmCorpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "mm = MmCorpus(\"gensim_txt.mm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#in case you're only loading the corpus - build dict first\n",
    "\n",
    "#import pickle\n",
    "\n",
    "#pickle.dump( id_types, open( \"gb_plain.dictionary\", \"wb\" ) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#id_types = pickle.load(open(\"gb_plain.dictionary\", 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#mm = gensim.corpora.MmCorpus(\"gb_all.mm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#id_types = pickle.load(open(\"gb_all.dictionary\", 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "doc2id = {value : key for key, value in doc_ids.items()}\n",
    "type2id = {value : key for key, value in id_types.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "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=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['grauen',\n",
       " 'wahre',\n",
       " 'asche',\n",
       " 'gespielt',\n",
       " 'treiben',\n",
       " 'schweigend',\n",
       " 'ruhigere',\n",
       " 'scherzte',\n",
       " 'gesucht',\n",
       " 'boden']"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "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": 33,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'n = 0\\nfor key, value in type2id.items():\\n    if value == topicTerms[n]:\\n        print(key)'"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Anzeige der id des ersten keywords aus dem topic\n",
    "'''n = 0\n",
    "for key, value in type2id.items():\n",
    "    if value == topicTerms[n]:\n",
    "        print(key)'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'n = 0\\ntopicN = model.get_topic_terms(n)\\ntopicN'"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Gensim-Funktion -> Anzeige des word_id-probability-mappings eines topics n\n",
    "'''n = 0\n",
    "topicN = model.get_topic_terms(n)\n",
    "topicN'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0,\n",
       "  '0.021*\"grauen\" + 0.017*\"asche\" + 0.016*\"treiben\" + 0.014*\"wahre\" + 0.014*\"schweigend\" + 0.012*\"gespielt\" + 0.011*\"ruhigere\" + 0.010*\"lehnen\" + 0.009*\"uhr\" + 0.009*\"gesucht\"'),\n",
       " (1,\n",
       "  '0.038*\"grauen\" + 0.027*\"wahre\" + 0.019*\"asche\" + 0.017*\"gespielt\" + 0.015*\"treiben\" + 0.013*\"schweigend\" + 0.012*\"ruhigere\" + 0.011*\"scherzte\" + 0.010*\"gesucht\" + 0.010*\"boden\"')]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Topic probabilities\n",
    "topics = model.show_topics(num_topics = 10)\n",
    "topics[:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "source": [
    "from scipy.io import mmread\n",
    "test = mmread(\"gb_plain.mm\")\n",
    "import pandas as pd\n",
    "pd.DataFrame(test.data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Visualisierung "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "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": 38,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "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": 39,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\ipykernel\\pylab\\config.py:66: DeprecationWarning: metadata {'config': True} was set from the constructor.  Metadata should be set using the .tag() method, e.g., Int().tag(key1='value1', key2='value2')\n",
      "  inline backend.\"\"\"\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\ipykernel\\pylab\\config.py:71: DeprecationWarning: metadata {'config': True} was set from the constructor.  Metadata should be set using the .tag() method, e.g., Int().tag(key1='value1', key2='value2')\n",
      "  'retina', 'jpeg', 'svg', 'pdf'.\"\"\")\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\ipykernel\\pylab\\config.py:85: DeprecationWarning: metadata {'config': True} was set from the constructor.  Metadata should be set using the .tag() method, e.g., Int().tag(key1='value1', key2='value2')\n",
      "  use `figure_formats` instead)\"\"\")\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\ipykernel\\pylab\\config.py:95: DeprecationWarning: metadata {'config': True} was set from the constructor.  Metadata should be set using the .tag() method, e.g., Int().tag(key1='value1', key2='value2')\n",
      "  \"\"\"\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\ipykernel\\pylab\\config.py:114: DeprecationWarning: metadata {'config': True} was set from the constructor.  Metadata should be set using the .tag() method, e.g., Int().tag(key1='value1', key2='value2')\n",
      "  \"\"\")\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\ipykernel\\pylab\\config.py:44: DeprecationWarning: InlineBackend._config_changed is deprecated: use @observe and @unobserve instead.\n",
      "  def _config_changed(self, name, old, new):\n",
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\traitlets\\traitlets.py:770: DeprecationWarning: A parent of InlineBackend._config_changed has adopted the new @observe(change) API\n",
      "  clsname, change_or_name), DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'c:\\\\users\\\\philip\\\\appdata\\\\local\\\\programs\\\\python\\\\python35-32\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaYAAAEXCAYAAADm5+DTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa68d770>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "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": 40,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEaCAYAAABEsMO+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa72f730>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visual.plot_doc_topics(doc_topic, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#heatmap = visual.make_heatmap()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#visual.save_heatmap(\"./visualizations/heatmap\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Interactive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#vis = collection.Visualization(lda_model, corpus, dictionary, doc_labels, interactive=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#vis.make_interactive()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#vis.save_interactive(\"./visualizations/interactive\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('grauen', 'asche'),\n",
       " ('grauen', 'treiben'),\n",
       " ('grauen', 'wahre'),\n",
       " ('grauen', 'schweigend'),\n",
       " ('grauen', 'gespielt')]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "segmented_topics = evaluation.topic_segmenter(model, type2id, 10)\n",
    "segmented_topics[0][:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "grauen           {1, 3}\n",
       "boden         {1, 3, 4}\n",
       "woche               {2}\n",
       "berührt             {1}\n",
       "fanden              {0}\n",
       "stück               {0}\n",
       "wahre               {0}\n",
       "uhr                 {3}\n",
       "schweigend          {3}\n",
       "klatschten          {4}\n",
       "lehnen              {1}\n",
       "gespielt            {3}\n",
       "scherzte            {3}\n",
       "treiben             {2}\n",
       "ruhigere            {1}\n",
       "müsse               {0}\n",
       "asche               {1}\n",
       "gesucht             {4}\n",
       "dtype: object"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "umass_probability = evaluation.token_probability(doc_tokens, segmented_topics)\n",
    "umass_probability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.040750318343069615"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "umass = evaluation.calculate_umass(segmented_topics, umass_probability, doc_tokens, 10)\n",
    "umass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\philip\\appdata\\local\\programs\\python\\python35-32\\lib\\site-packages\\bs4\\builder\\_lxml.py:250: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead\n",
      "  self.parser.feed(markup)\n"
     ]
    }
   ],
   "source": [
    "wiki_list = evaluation.wikipedia_table_crawler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "wiki_corpus = evaluation.wikipedia_crawler(wiki_list, size=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "uci_probability = evaluation.token_probability(wiki_corpus, segmented_topics)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "pmi = evaluation.calculate_pointwise_mutual_information(segmented_topics, wiki_corpus, uci_probability)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uci = evaluation.calculate_uci(pmi, wiki_corpus, 10)\n",
    "uci"
   ]
  },
  {
   "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
}