Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "\n",
    "import lda\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import lda.datasets\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "from scipy.spatial.distance import pdist, squareform\n",
    "from scipy.cluster.hierarchy import linkage, dendrogram\n",
    "from scipy.cluster.hierarchy import ward, dendrogram\n",
    "\n",
    "#from sklearn.lda import \n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "from sklearn.decomposition import TruncatedSVD \n",
    "from sklearn.preprocessing import normalize "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>resume_text</th>\n",
       "      <th>resume_stopped</th>\n",
       "      <th>resume_nouns</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Petros Gazazyan North Hollywood, CA Werkervari...</td>\n",
       "      <td>petros gazazyan werkervaring engineer structur...</td>\n",
       "      <td>petros gazazyan engineer structural ttg engine...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Travis London Java Software Engineer Tucson, A...</td>\n",
       "      <td>java software engineer bereid overal naartoe t...</td>\n",
       "      <td>java software engineer bereid overal naartoe t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Stephen A. Kraft Mechanical Engineer Seattle, ...</td>\n",
       "      <td>kraft mechanical engineer bereid overal naarto...</td>\n",
       "      <td>mechanical engineer bereid overal naartoe te v...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         resume_text  \\\n",
       "0  Petros Gazazyan North Hollywood, CA Werkervari...   \n",
       "1  Travis London Java Software Engineer Tucson, A...   \n",
       "2  Stephen A. Kraft Mechanical Engineer Seattle, ...   \n",
       "\n",
       "                                      resume_stopped  \\\n",
       "0  petros gazazyan werkervaring engineer structur...   \n",
       "1  java software engineer bereid overal naartoe t...   \n",
       "2  kraft mechanical engineer bereid overal naarto...   \n",
       "\n",
       "                                        resume_nouns  \n",
       "0  petros gazazyan engineer structural ttg engine...  \n",
       "1  java software engineer bereid overal naartoe t...  \n",
       "2  mechanical engineer bereid overal naartoe te v...  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_pickle('pkl/df_stop_noun.pkl')\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_samples = 2000\n",
    "n_features = 10000\n",
    "n_topics = 10\n",
    "n_top_words = 20\n",
    "\n",
    "tfidf_vec = TfidfVectorizer(ngram_range=(1, 3), max_df=0.9, min_df=2, \n",
    "                max_features=n_features, norm='l2', use_idf=True, smooth_idf=True, sublinear_tf=False)\n",
    "\n",
    "tfidf_vec_prep = tfidf_vec.fit_transform(df['resume_nouns'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17049, 10000)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tfidf_vec_prep.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(17049, 2000)\n"
     ]
    }
   ],
   "source": [
    "svd = TruncatedSVD(n_components=2000)\n",
    "embedded_coords = normalize(svd.fit_transform(tfidf_vec_prep), norm='l1')\n",
    "print(embedded_coords.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17049, 2000)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embedded_coords.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.74921872546547919"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(svd.explained_variance_ratio_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x14259ae10>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwQAAAHfCAYAAADunR8nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10488c470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(np.cumsum(svd.explained_variance_ratio_)).plot(figsize=(13, 8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Word 2 Vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# word2vec takes a list of words\n",
    "texts = df['resume_nouns'].str.split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gensim\n",
    "\n",
    "model = gensim.models.Word2Vec(texts, size=100, window=5, min_count=1, workers=-1, sg=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.00046616 -0.00418129 -0.00266784 -0.00060634 -0.00478803  0.00332045\n",
      "  0.00164161  0.00238047 -0.00488407  0.00336936 -0.00064776 -0.00372381\n",
      "  0.00214476 -0.00424208 -0.00217381 -0.00432206 -0.00415413  0.00291412\n",
      "  0.00478131  0.00156394 -0.00073102  0.00439307 -0.0045449  -0.00241699\n",
      "  0.00254593 -0.0001029  -0.00351394  0.00298304 -0.00365746 -0.00060026\n",
      "  0.00429216 -0.00484251 -0.00101826  0.00227789  0.00143506 -0.00073767\n",
      "  0.00092654  0.00333487  0.00149107 -0.00402783  0.00219245  0.00346284\n",
      "  0.00248748  0.00209765  0.00345457 -0.00402564  0.00036272  0.00179107\n",
      "  0.00393706 -0.00213186 -0.00318811  0.00485853 -0.0039414   0.00055312\n",
      " -0.00190455 -0.00336747  0.00056317  0.00464614  0.00405323 -0.00077426\n",
      " -0.00246418 -0.0007037   0.00062034 -0.00181774 -0.0045054  -0.00428172\n",
      "  0.00429121  0.00327288  0.0030564  -0.00235024 -0.00413837 -0.00255522\n",
      "  0.00024547 -0.00368606 -0.00478979  0.00429033 -0.00267319  0.00277202\n",
      "  0.00456404 -0.00482699 -0.00391602 -0.00229604 -0.0042844  -0.00256562\n",
      "  0.00179273 -0.00011742  0.00230063  0.0020994  -0.00189446  0.002338\n",
      "  0.00314394 -0.00033031  0.00392425 -0.00235475  0.00477209 -0.00333553\n",
      " -0.00292476 -0.00371102  0.00367263 -0.0049001 ]\n"
     ]
    }
   ],
   "source": [
    "print(model['computer'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import nltk\n",
    "\n",
    "class MySentences2(object):\n",
    "     def __init__(self, dirname):\n",
    "            self.dirname = dirname\n",
    " \n",
    "     def __iter__(self):\n",
    "         for fname in os.listdir(self.dirname):\n",
    "                for line in open(os.path.join(self.dirname, fname), encoding='latin-1'):\n",
    "                    word=line.lower().split()\n",
    "                    yield word\n",
    "                    \n",
    "sentences = MySentences2('/Users/bryant/nltk_data/corpora/gutenberg') \n",
    "model = gensim.models.Word2Vec(sentences, size=100, window=5, min_count=1, workers=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dist = 1 - cosine_similarity(tfidf_vec_prep)\n",
    "\n",
    "linkage_matrix = ward(dist) #define the linkage_matrix using ward clustering pre-computed distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAWZCAYAAACL8OWEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBgAAAPDsCTAAAACAZ0+AAQAAADx7/xwJ1+yI4wqUUgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x162c10588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(15, 20)) # set size\n",
    "ax = dendrogram(linkage_matrix, orientation=\"right\");\n",
    "\n",
    "plt.tick_params(\\\n",
    "    axis= 'x',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom='off',      # ticks along the bottom edge are off\n",
    "    top='off',         # ticks along the top edge are off\n",
    "    labelbottom='off')\n",
    "\n",
    "plt.tight_layout() #show plot with tight layout\n",
    "\n",
    "#uncomment below to save figure\n",
    "plt.savefig('ward_clusters.png', dpi=200) #save figure as ward_clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics.pairwise import cosine_distances\n",
    "from sklearn.metrics.pairwise import paired_euclidean_distances\n",
    "\n",
    "print cosine_distances(X[0], X[1])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [default]",
   "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}