{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook I benchmark a few text categorization models to test whehter word embeddings like word2vec can improve text classification accuracy. The the notebook proceeds as follows:\n",
    "1. downloading the datasets\n",
    "2. construction of the training set\n",
    "3. definitions of models\n",
    "4. benchmarking models\n",
    "5. plotting results\n",
    "\n",
    "This notebook was converted to Python 3 by <a href='https://github.com/marcelobeckmann'>@marcelobeckmann</a>\n",
    "\n",
    "Tested with Python 3.5.3. The results are the same from the notebook for Python 2."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Downloading datasets and pretrained wector embeddings. Especially the embeddings can take a while to download. You might want run these in the terminal instead to see wget's progress bar. If you're on Windows (and not in cygwin) %%bash cell magic won't work and you'll have to do all this manually (or with %%cmd magic I guess). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "# download Reuters' text categorization benchmarks\n",
    "wget http://www.cs.umb.edu/~smimarog/textmining/datasets/r8-train-no-stop.txt\n",
    "wget http://www.cs.umb.edu/~smimarog/textmining/datasets/r8-test-no-stop.txt\n",
    "# concatenate train and test files, we'll make our own train-test splits\n",
    "cat r8-*-no-stop.txt > r8-no-stop.txt\n",
    "wget http://www.cs.umb.edu/~smimarog/textmining/datasets/r52-train-no-stop.txt\n",
    "wget http://www.cs.umb.edu/~smimarog/textmining/datasets/r52-test-no-stop.txt\n",
    "cat r52-*-no-stop.txt > r52-no-stop.txt\n",
    "wget http://www.cs.umb.edu/~smimarog/textmining/datasets/20ng-test-no-stop.txt\n",
    "wget http://www.cs.umb.edu/~smimarog/textmining/datasets/20ng-train-no-stop.txt\n",
    "cat 20ng-*-no-stop.txt > 20ng-no-stop.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import zipfile\n",
    "# download GloVe word vector representations\n",
    "# bunch of small embeddings - trained on 6B tokens - 822 MB download, 2GB unzipped\n",
    "wget http://nlp.stanford.edu/data/glove.6B.zip\n",
    "zip = zipfile.ZipFile('glove.6B.zip')\n",
    "zip.extractall()\n",
    "\n",
    "# and a single behemoth - trained on 840B tokens - 2GB compressed, 5GB unzipped\n",
    "wget http://nlp.stanford.edu/data/glove.840B.300d.zip\n",
    "zip = zipfile.ZipFile('glove.840B.300d.zip')\n",
    "zip.extractall()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\gensim\\utils.py:862: UserWarning: detected Windows; aliasing chunkize to chunkize_serial\n",
      "  warnings.warn(\"detected Windows; aliasing chunkize to chunkize_serial\")\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from tabulate import tabulate\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from gensim.models.word2vec import Word2Vec\n",
    "from collections import Counter, defaultdict\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.naive_bayes import BernoulliNB, MultinomialNB\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.cross_validation import StratifiedShuffleSplit\n",
    "\n",
    "# TRAIN_SET_PATH = \"20ng-no-stop.txt\"\n",
    "# TRAIN_SET_PATH = \"r52-all-terms.txt\"\n",
    "TRAIN_SET_PATH = \"r8-no-stop.txt\"\n",
    "\n",
    "GLOVE_6B_50D_PATH = \"glove.6B.50d.txt\"\n",
    "GLOVE_840B_300D_PATH = \"glove.840B.300d.txt\"\n",
    "encoding=\"utf-8\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total examples 7674\n"
     ]
    }
   ],
   "source": [
    "X, y = [], []\n",
    "with open(TRAIN_SET_PATH, \"r\") as infile:\n",
    "    for line in infile:\n",
    "        label, text = line.split(\"\\t\")\n",
    "        # texts are already tokenized, just split on space\n",
    "        # in a real case we would use e.g. spaCy for tokenization\n",
    "        # and maybe remove stopwords etc.\n",
    "        X.append(text.split())\n",
    "        y.append(label)\n",
    "X, y = np.array(X), np.array(y)\n",
    "print (\"total examples %s\" % len(y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Prepare word embeddings - both the downloaded pretrained ones and train a new one from scratch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "with open(GLOVE_6B_50D_PATH, \"rb\") as lines:\n",
    "    wvec = {line.split()[0].decode(encoding): np.array(line.split()[1:],dtype=np.float32)\n",
    "               for line in lines}\n",
    "    \n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# reading glove files, this may take a while\n",
    "# we're reading line by line and only saving vectors\n",
    "# that correspond to words from our training set\n",
    "# if you wan't to play around with the vectors and have \n",
    "# enough RAM - remove the 'if' line and load everything\n",
    "\n",
    "import struct \n",
    "\n",
    "glove_small = {}\n",
    "all_words = set(w for words in X for w in words)\n",
    "with open(GLOVE_6B_50D_PATH, \"rb\") as infile:\n",
    "    for line in infile:\n",
    "        parts = line.split()\n",
    "        word = parts[0].decode(encoding)\n",
    "        if (word in all_words):\n",
    "            nums=np.array(parts[1:], dtype=np.float32)\n",
    "            glove_small[word] = nums\n",
    "\n",
    "            \n",
    "glove_big = {}\n",
    "with open(GLOVE_840B_300D_PATH, \"rb\") as infile:\n",
    "    for line in infile:\n",
    "        parts = line.split()\n",
    "        word = parts[0].decode(encoding)\n",
    "        if word in all_words:\n",
    "            nums=np.array(parts[1:], dtype=np.float32)\n",
    "            glove_big[word] = nums\n",
    "            \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# train word2vec on all the texts - both training and test set\n",
    "# we're not using test labels, just texts so this is fine\n",
    "model = Word2Vec(X, size=100, window=5, min_count=5, workers=2)\n",
    "w2v = {w: vec for w, vec in zip(model.wv.index2word, model.wv.syn0)}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22931\n"
     ]
    }
   ],
   "source": [
    "print(len(all_words))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Time for model definitions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# start with the classics - naive bayes of the multinomial and bernoulli varieties\n",
    "# with either pure counts or tfidf features\n",
    "mult_nb = Pipeline([(\"count_vectorizer\", CountVectorizer(analyzer=lambda x: x)), (\"multinomial nb\", MultinomialNB())])\n",
    "bern_nb = Pipeline([(\"count_vectorizer\", CountVectorizer(analyzer=lambda x: x)), (\"bernoulli nb\", BernoulliNB())])\n",
    "mult_nb_tfidf = Pipeline([(\"tfidf_vectorizer\", TfidfVectorizer(analyzer=lambda x: x)), (\"multinomial nb\", MultinomialNB())])\n",
    "bern_nb_tfidf = Pipeline([(\"tfidf_vectorizer\", TfidfVectorizer(analyzer=lambda x: x)), (\"bernoulli nb\", BernoulliNB())])\n",
    "# SVM - which is supposed to be more or less state of the art \n",
    "# http://www.cs.cornell.edu/people/tj/publications/joachims_98a.pdf\n",
    "svc = Pipeline([(\"count_vectorizer\", CountVectorizer(analyzer=lambda x: x)), (\"linear svc\", SVC(kernel=\"linear\"))])\n",
    "svc_tfidf = Pipeline([(\"tfidf_vectorizer\", TfidfVectorizer(analyzer=lambda x: x)), (\"linear svc\", SVC(kernel=\"linear\"))])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "Now the meat - classifiers using vector embeddings. We will implement an embedding vectorizer - a counterpart of CountVectorizer and TfidfVectorizer - that is given a word -> vector mapping and vectorizes texts by taking the mean of all the vectors corresponding to individual words."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class MeanEmbeddingVectorizer(object):\n",
    "    def __init__(self, word2vec):\n",
    "        self.word2vec = word2vec\n",
    "        if len(word2vec)>0:\n",
    "            self.dim=len(word2vec[next(iter(glove_small))])\n",
    "        else:\n",
    "            self.dim=0\n",
    "            \n",
    "    def fit(self, X, y):\n",
    "        return self \n",
    "\n",
    "    def transform(self, X):\n",
    "        return np.array([\n",
    "            np.mean([self.word2vec[w] for w in words if w in self.word2vec] \n",
    "                    or [np.zeros(self.dim)], axis=0)\n",
    "            for words in X\n",
    "        ])\n",
    "\n",
    "    \n",
    "# and a tf-idf version of the same\n",
    "class TfidfEmbeddingVectorizer(object):\n",
    "    def __init__(self, word2vec):\n",
    "        self.word2vec = word2vec\n",
    "        self.word2weight = None\n",
    "        if len(word2vec)>0:\n",
    "            self.dim=len(word2vec[next(iter(glove_small))])\n",
    "        else:\n",
    "            self.dim=0\n",
    "        \n",
    "    def fit(self, X, y):\n",
    "        tfidf = TfidfVectorizer(analyzer=lambda x: x)\n",
    "        tfidf.fit(X)\n",
    "        # if a word was never seen - it must be at least as infrequent\n",
    "        # as any of the known words - so the default idf is the max of \n",
    "        # known idf's\n",
    "        max_idf = max(tfidf.idf_)\n",
    "        self.word2weight = defaultdict(\n",
    "            lambda: max_idf, \n",
    "            [(w, tfidf.idf_[i]) for w, i in tfidf.vocabulary_.items()])\n",
    "    \n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        return np.array([\n",
    "                np.mean([self.word2vec[w] * self.word2weight[w]\n",
    "                         for w in words if w in self.word2vec] or\n",
    "                        [np.zeros(self.dim)], axis=0)\n",
    "                for words in X\n",
    "            ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Extra Trees classifier is almost universally great, let's stack it with our embeddings\n",
    "etree_glove_small = Pipeline([(\"glove vectorizer\", MeanEmbeddingVectorizer(glove_small)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_small_tfidf = Pipeline([(\"glove vectorizer\", TfidfEmbeddingVectorizer(glove_small)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_big = Pipeline([(\"glove vectorizer\", MeanEmbeddingVectorizer(glove_big)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_big_tfidf = Pipeline([(\"glove vectorizer\", TfidfEmbeddingVectorizer(glove_big)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "\n",
    "etree_w2v = Pipeline([(\"word2vec vectorizer\", MeanEmbeddingVectorizer(w2v)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_w2v_tfidf = Pipeline([(\"word2vec vectorizer\", TfidfEmbeddingVectorizer(w2v)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "benchmark all the things!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model                score\n",
      "-----------------  -------\n",
      "svc_tfidf           0.9656\n",
      "svc                 0.9562\n",
      "w2v_tfidf           0.9540\n",
      "w2v                 0.9524\n",
      "mult_nb             0.9467\n",
      "glove_big           0.9283\n",
      "glove_small         0.9230\n",
      "glove_small_tfidf   0.9082\n",
      "glove_big_tfidf     0.9058\n",
      "mult_nb_tfidf       0.8615\n",
      "bern_nb             0.7954\n",
      "bern_nb_tfidf       0.7954\n"
     ]
    }
   ],
   "source": [
    "all_models = [\n",
    "    (\"mult_nb\", mult_nb),\n",
    "    (\"mult_nb_tfidf\", mult_nb_tfidf),\n",
    "    (\"bern_nb\", bern_nb),\n",
    "    (\"bern_nb_tfidf\", bern_nb_tfidf),\n",
    "    (\"svc\", svc),\n",
    "    (\"svc_tfidf\", svc_tfidf),\n",
    "    (\"w2v\", etree_w2v),\n",
    "    (\"w2v_tfidf\", etree_w2v_tfidf),\n",
    "    (\"glove_small\", etree_glove_small),\n",
    "    (\"glove_small_tfidf\", etree_glove_small_tfidf),\n",
    "    (\"glove_big\", etree_glove_big),\n",
    "    (\"glove_big_tfidf\", etree_glove_big_tfidf),\n",
    "\n",
    "]\n",
    "\n",
    "\n",
    "unsorted_scores = [(name, cross_val_score(model, X, y, cv=5).mean()) for name, model in all_models]\n",
    "scores = sorted(unsorted_scores, key=lambda x: -x[1])\n",
    "\n",
    "\n",
    "print (tabulate(scores, floatfmt=\".4f\", headers=(\"model\", 'score')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1c9d5c8e7b8>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2gAAAFkCAYAAABVSxhvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABQCAENAACgEAIaAABAIf4/dh/FXNVG74UAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c9d5c8ed30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 6))\n",
    "sns.barplot(x=[name for name, _ in scores], y=[score for _, score in scores])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ok, this is how it is. Let's see how the ranking depends on the amount of training data. Word embedding models which are semi-supervised should shine when there is very little labeled training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def benchmark(model, X, y, n):\n",
    "    test_size = 1 - (n / float(len(y)))\n",
    "    scores = []\n",
    "    for train, test in StratifiedShuffleSplit(y, n_iter=5, test_size=test_size):\n",
    "        X_train, X_test = X[train], X[test]\n",
    "        y_train, y_test = y[train], y[test]\n",
    "        scores.append(accuracy_score(model.fit(X_train, y_train).predict(X_test), y_test))\n",
    "    return np.mean(scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "train_sizes = [10, 40, 160, 640, 3200, 6400]\n",
    "table = []\n",
    "for name, model in all_models:\n",
    "    for n in train_sizes:\n",
    "        table.append({'model': name, \n",
    "                      'accuracy': benchmark(model, X, y, n), \n",
    "                      'train_size': n})\n",
    "df = pd.DataFrame(table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.text.Text at 0x1c9d5df06a0>]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3wAAAGGCAYAAAAggoN0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIiIiIpKmVPhERERERETS1F9nj4l/P9PtEgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c9d787b940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 6))\n",
    "fig = sns.pointplot(x='train_size', y='accuracy', hue='model', \n",
    "                    data=df[df.model.map(lambda x: x in [\"mult_nb\", \"svc_tfidf\", \"w2v_tfidf\", \n",
    "                                                         \"glove_small_tfidf\", \"glove_big_tfidf\", \n",
    "                                                        ])])\n",
    "sns.set_context(\"notebook\", font_scale=1.5)\n",
    "fig.set(ylabel=\"accuracy\")\n",
    "fig.set(xlabel=\"labeled training examples\")\n",
    "fig.set(title=\"R8 benchmark\")\n",
    "fig.set(ylabel=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's be lazy and instead of refactoring, just copy the above code changing the input path to reuters 52"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total examples 9100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n",
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\cross_validation.py:552: Warning: The least populated class in y has only 3 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=5.\n",
      "  % (min_labels, self.n_folds)), Warning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model                score\n",
      "-----------------  -------\n",
      "svc_tfidf           0.9384\n",
      "svc                 0.9204\n",
      "w2v_tfidf           0.8920\n",
      "w2v                 0.8832\n",
      "mult_nb             0.8740\n",
      "glove_big           0.8611\n",
      "glove_small         0.8563\n",
      "glove_small_tfidf   0.8414\n",
      "glove_big_tfidf     0.8346\n",
      "mult_nb_tfidf       0.7257\n",
      "bern_nb             0.6668\n",
      "bern_nb_tfidf       0.6668\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.text.Text at 0x1c9c68aeb38>]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4oAAAGQCAYAAADoX54sAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c9c67e4ac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "TRAIN_SET_PATH = \"r52-no-stop.txt\"\n",
    "\n",
    "X, y = [], []\n",
    "with open(TRAIN_SET_PATH, \"r\") as infile:\n",
    "    for line in infile:\n",
    "        label, text = line.split(\"\\t\")\n",
    "        # texts are already tokenized, just split on space\n",
    "        # in a real case we would use e.g. spaCy for tokenization\n",
    "        # and maybe remove stopwords etc.\n",
    "        X.append(text.split())\n",
    "        y.append(label)\n",
    "X, y = np.array(X), np.array(y)\n",
    "print (\"total examples %s\" % len(y))\n",
    "\n",
    "\n",
    "# reading glove files, this may take a while\n",
    "# we're reading line by line and only saving vectors\n",
    "# that correspond to words from our training set\n",
    "# if you wan't to play around with the vectors and have \n",
    "# enough RAM - remove the 'if' line and load everything\n",
    "\n",
    "glove_small = {}\n",
    "all_words = set(w for words in X for w in words)\n",
    "with open(GLOVE_6B_50D_PATH, \"rb\") as infile:\n",
    "    for line in infile:\n",
    "        parts = line.split()\n",
    "        word = parts[0].decode(encoding)\n",
    "        if (word in all_words):\n",
    "            nums=np.array(parts[1:], dtype=np.float32)\n",
    "            glove_small[word] = nums\n",
    "            \n",
    "glove_big = {}\n",
    "with open(GLOVE_840B_300D_PATH, \"rb\") as infile:\n",
    "    for line in infile:\n",
    "        parts = line.split()\n",
    "        word = parts[0].decode(encoding)\n",
    "        if (word in all_words):\n",
    "            nums=np.array(parts[1:], dtype=np.float32)\n",
    "            glove_big[word] = nums\n",
    "            \n",
    "            \n",
    "            \n",
    "# train word2vec on all the texts - both training and test set\n",
    "# we're not using test labels, just texts so this is fine\n",
    "model = Word2Vec(X, size=100, window=5, min_count=5, workers=2)\n",
    "w2v = {w: vec for w, vec in zip(model.wv.index2word, model.wv.syn0)}\n",
    "\n",
    "\n",
    "\n",
    "# Extra Trees classifier is almost universally great, let's stack it with our embeddings\n",
    "etree_glove_small = Pipeline([(\"glove vectorizer\", MeanEmbeddingVectorizer(glove_small)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_small_tfidf = Pipeline([(\"glove vectorizer\", TfidfEmbeddingVectorizer(glove_small)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_big = Pipeline([(\"glove vectorizer\", MeanEmbeddingVectorizer(glove_big)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_big_tfidf = Pipeline([(\"glove vectorizer\", TfidfEmbeddingVectorizer(glove_big)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "\n",
    "etree_w2v = Pipeline([(\"word2vec vectorizer\", MeanEmbeddingVectorizer(w2v)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_w2v_tfidf = Pipeline([(\"word2vec vectorizer\", TfidfEmbeddingVectorizer(w2v)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "all_models = [\n",
    "    (\"mult_nb\", mult_nb),\n",
    "    (\"mult_nb_tfidf\", mult_nb_tfidf),\n",
    "    (\"bern_nb\", bern_nb),\n",
    "    (\"bern_nb_tfidf\", bern_nb_tfidf),\n",
    "    (\"svc\", svc),\n",
    "    (\"svc_tfidf\", svc_tfidf),\n",
    "    (\"glove_small\", etree_glove_small), \n",
    "    (\"glove_small_tfidf\", etree_glove_small_tfidf),\n",
    "    (\"glove_big\", etree_glove_big), \n",
    "    (\"glove_big_tfidf\", etree_glove_big_tfidf),\n",
    "    (\"w2v\", etree_w2v),\n",
    "    (\"w2v_tfidf\", etree_w2v_tfidf),\n",
    "]\n",
    "\n",
    "unsorted_scores = [(name, cross_val_score(model, X, y, cv=5).mean()) for name, model in all_models]\n",
    "scores = sorted(unsorted_scores, key=lambda x: -x[1])\n",
    "\n",
    "print(tabulate(scores, floatfmt=\".4f\", headers=(\"model\", 'score')))\n",
    "\n",
    "\n",
    "train_sizes = [100, 400, 1600, 3200, 6400, 8000]\n",
    "table = []\n",
    "for name, model in all_models:\n",
    "    for n in train_sizes:\n",
    "        table.append({'model': name, \n",
    "                      'accuracy': benchmark(model, X, y, n), \n",
    "                      'train_size': n})\n",
    "df = pd.DataFrame(table)\n",
    "\n",
    "\n",
    "plt.figure(figsize=(15, 6))\n",
    "fig = sns.pointplot(x='train_size', y='accuracy', hue='model', \n",
    "                    data=df[df.model.map(lambda x: x in [\"mult_nb\", \"svc_tfidf\", \"w2v_tfidf\", \n",
    "                                                         \"glove_small_tfidf\", \"glove_big_tfidf\", \n",
    "                                                        ])])\n",
    "sns.set_context(\"notebook\", font_scale=1.5)\n",
    "fig.set(ylabel=\"accuracy\")\n",
    "fig.set(xlabel=\"labeled training examples\")\n",
    "fig.set(title=\"R52 benchmark\")\n",
    "fig.set(ylabel=\"accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And now the 20 news group dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total examples 11293\n",
      "model                score\n",
      "-----------------  -------\n",
      "svc_tfidf           0.8861\n",
      "mult_nb             0.8544\n",
      "mult_nb_tfidf       0.8505\n",
      "svc                 0.7889\n",
      "glove_big           0.7494\n",
      "glove_big_tfidf     0.7489\n",
      "bern_nb             0.6878\n",
      "bern_nb_tfidf       0.6878\n",
      "glove_small         0.6686\n",
      "glove_small_tfidf   0.6633\n",
      "w2v_tfidf           0.5011\n",
      "w2v                 0.4579\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.text.Text at 0x1c9c70bee48>]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4IAAAGQCAYAAAD7iN7YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1c9c76c1ac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "TRAIN_SET_PATH = \"20ng-no-stop.txt\"\n",
    "\n",
    "X, y = [], []\n",
    "with open(TRAIN_SET_PATH, \"r\") as infile:\n",
    "    for line in infile:\n",
    "        label, text = line.split(\"\\t\")\n",
    "        # texts are already tokenized, just split on space\n",
    "        # in a real case we would use e.g. spaCy for tokenization\n",
    "        # and maybe remove stopwords etc.\n",
    "        X.append(text.split())\n",
    "        y.append(label)\n",
    "X, y = np.array(X), np.array(y)\n",
    "print (\"total examples %s\" % len(y))\n",
    "\n",
    "\n",
    "# reading glove files, this may take a while\n",
    "# we're reading line by line and only saving vectors\n",
    "# that correspond to words from our training set\n",
    "# if you wan't to play around with the vectors and have \n",
    "# enough RAM - remove the 'if' line and load everything\n",
    "\n",
    "glove_small = {}\n",
    "all_words = set(w for words in X for w in words)\n",
    "with open(GLOVE_6B_50D_PATH, \"rb\") as infile:\n",
    "    for line in infile:\n",
    "        parts = line.split()\n",
    "        word = parts[0].decode(encoding)\n",
    "        if (word in all_words):\n",
    "            nums=np.array(parts[1:], dtype=np.float32)\n",
    "            glove_small[word] = nums\n",
    "            \n",
    "glove_big = {}\n",
    "with open(GLOVE_840B_300D_PATH, \"rb\") as infile:\n",
    "    for line in infile:\n",
    "        parts = line.split()\n",
    "        word = parts[0].decode(encoding)\n",
    "        if (word in all_words):\n",
    "            nums=np.array(parts[1:], dtype=np.float32)\n",
    "            glove_big[word] = nums\n",
    "            \n",
    "            \n",
    "            \n",
    "# train word2vec on all the texts - both training and test set\n",
    "# we're not using test labels, just texts so this is fine\n",
    "model = Word2Vec(X, size=100, window=5, min_count=5, workers=2)\n",
    "w2v = {w: vec for w, vec in zip(model.wv.index2word, model.wv.syn0)}\n",
    "\n",
    "\n",
    "\n",
    "# Extra Trees classifier is almost universally great, let's stack it with our embeddings\n",
    "etree_glove_small = Pipeline([(\"glove vectorizer\", MeanEmbeddingVectorizer(glove_small)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_small_tfidf = Pipeline([(\"glove vectorizer\", TfidfEmbeddingVectorizer(glove_small)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_big = Pipeline([(\"glove vectorizer\", MeanEmbeddingVectorizer(glove_big)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_glove_big_tfidf = Pipeline([(\"glove vectorizer\", TfidfEmbeddingVectorizer(glove_big)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "\n",
    "etree_w2v = Pipeline([(\"word2vec vectorizer\", MeanEmbeddingVectorizer(w2v)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "etree_w2v_tfidf = Pipeline([(\"word2vec vectorizer\", TfidfEmbeddingVectorizer(w2v)), \n",
    "                        (\"extra trees\", ExtraTreesClassifier(n_estimators=200))])\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "all_models = [\n",
    "    (\"mult_nb\", mult_nb),\n",
    "    (\"mult_nb_tfidf\", mult_nb_tfidf),\n",
    "    (\"bern_nb\", bern_nb),\n",
    "    (\"bern_nb_tfidf\", bern_nb_tfidf),\n",
    "    (\"svc\", svc),\n",
    "    (\"svc_tfidf\", svc_tfidf),\n",
    "    (\"glove_small\", etree_glove_small), \n",
    "    (\"glove_small_tfidf\", etree_glove_small_tfidf),\n",
    "    (\"glove_big\", etree_glove_big), \n",
    "    (\"glove_big_tfidf\", etree_glove_big_tfidf),\n",
    "    (\"w2v\", etree_w2v),\n",
    "    (\"w2v_tfidf\", etree_w2v_tfidf),\n",
    "]\n",
    "\n",
    "    \n",
    "unsorted_scores = [(name, cross_val_score(model, X, y, cv=5).mean()) for name, model in all_models]\n",
    "scores = sorted(unsorted_scores, key=lambda x: -x[1])\n",
    "\n",
    "\n",
    "print (tabulate(scores, floatfmt=\".4f\", headers=(\"model\", 'score')))\n",
    "\n",
    "\n",
    "\n",
    "train_sizes = [40, 160, 640, 3200, 6400, 9000]\n",
    "table = []\n",
    "for name, model in all_models:\n",
    "    for n in train_sizes:\n",
    "        table.append({'model': name, \n",
    "                      'accuracy': benchmark(model, X, y, n), \n",
    "                      'train_size': n})\n",
    "df = pd.DataFrame(table)\n",
    "\n",
    "\n",
    "plt.figure(figsize=(15, 6))\n",
    "fig = sns.pointplot(x='train_size', y='accuracy', hue='model', \n",
    "                    data=df[df.model.map(lambda x: x in [\"mult_nb\", \"svc_tfidf\", \"w2v_tfidf\", \n",
    "                                                         \"glove_small_tfidf\", \"glove_big_tfidf\", \n",
    "                                                        ])])\n",
    "sns.set_context(\"notebook\", font_scale=1.5)\n",
    "fig.set(ylabel=\"accuracy\")\n",
    "fig.set(xlabel=\"labeled training examples\")\n",
    "fig.set(title=\"20 news groups benchmark\")\n",
    "fig.set(ylabel=\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}