svm_with_old_score.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random as rd\n",
    "from sklearn.svm import SVC\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# convert to numeric, drop nas and other [currently] irrelevant features\n",
    "def preprocess(filename):\n",
    "    df = pd.read_csv(filename+'.csv', sep=\",\")\n",
    "    # removing unnecessary columns. keeping only numbers atm\n",
    "    unnecessary = ['Body','ClosedDate','CommunityOwnedDate','CreationDate','Id','LastActivityDate',\n",
    "              'LastEditDate','LastEditorUserId','LastEditorDisplayName','OwnerDisplayName',\n",
    "              'OwnerUserId','ParentId','Tags','Title','Clean_Text','AcceptedAnswerId','Score',\n",
    "              'AnswerCount', 'CommentCount', 'FavoriteCount', 'PostTypeId','ViewCount']\n",
    "    droppable = np.intersect1d(df.columns,unnecessary)\n",
    "    df = df.drop(droppable, 1)\n",
    "    df = df.dropna()\n",
    "    if df.shape[0]>30000:\n",
    "        df = df[:30000]\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Split Data Into Train and Test Data\n",
    "def splitIntoTestAndTrain(dataFrame,splitBy):\n",
    "    rand = list(range(len(dataFrame)))\n",
    "    div = int(len(dataFrame) * splitBy)\n",
    "    rd.shuffle(rand)\n",
    "    train_df_index = rand[:div]\n",
    "    test_df_index = rand[div:]\n",
    "    train_df = dataFrame.iloc[train_df_index, :]\n",
    "    test_df = dataFrame.iloc[test_df_index, :]\n",
    "    return train_df,test_df;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Best SVM Classifier\n",
    "from sklearn import svm, datasets\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn import svm, grid_search\n",
    "def svc_param_selection(X, y, nfolds):\n",
    "    param_grid = [{'C': [1, 10, 100, 1000], 'kernel': ['linear']},{'C': [1, 10, 100, 1000], 'gamma': [0.001, 0.0001], 'kernel': ['rbf']}]\n",
    "    grid_search = GridSearchCV(svm.SVC(probability=True), param_grid, cv=nfolds)\n",
    "    grid_search.fit(X, y)\n",
    "    return grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def reportCard(y_true, y_pred, y_proba,name):\n",
    "    from sklearn import metrics\n",
    "    import matplotlib.pyplot as plt\n",
    "    %matplotlib inline\n",
    "    print('\\nReport for:',name)\n",
    "    target_names = ['low', 'high']\n",
    "    print(metrics.classification_report(y_true, y_pred, target_names=target_names))\n",
    "    print(\"Confusion Matrix\\n\",metrics.confusion_matrix(y_true,y_pred))\n",
    "    print(\"F1 Score:\",metrics.f1_score(y_true,y_pred))\n",
    "    print(\"Accuracy:\",metrics.accuracy_score(y_true,y_pred))\n",
    "    print(\"Log Loss:\",metrics.log_loss(y_true,y_pred))\n",
    "    print(\"AUC Score:\",metrics.roc_auc_score(y_true,y_proba[:,1]))\n",
    "    fpr,tpr,thresholds = metrics.roc_curve(y_true,y_proba[:,1])\n",
    "    plt.figure(1)\n",
    "    plt.plot([0, 1], [0, 1], 'k--')\n",
    "    plt.plot(fpr, tpr)\n",
    "    plt.xlabel('False positive rate')\n",
    "    plt.ylabel('True positive rate')\n",
    "    plt.title('ROC curve for {}'.format(name))\n",
    "    plt.legend(loc='best')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def svm_run(filename):\n",
    "    print(\"\\nRunning for\",filename)\n",
    "    df = preprocess(filename)\n",
    "    print(\"dataset size is:\",df.shape)\n",
    "    print(\"columns:\",df.columns)\n",
    "    train,test=splitIntoTestAndTrain(df,0.8)\n",
    "    print(\"\\nSVM\")\n",
    "    best_svm = svc_param_selection(train.drop('ScoreLabel', 1),train['ScoreLabel'],5)\n",
    "    grid_svm_pred = best_svm.predict(test.drop('ScoreLabel', 1))\n",
    "    grid_svm_pred_prob = best_svm.predict_proba(test.drop('ScoreLabel', 1))\n",
    "    print(\"*****BOF: {}*****\".format(filename))\n",
    "    reportCard(test['ScoreLabel'], grid_svm_pred, grid_svm_pred_prob,'GridSearched SVM')\n",
    "    print(\"*****EOF: {}*****\".format(filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running for ai_posts_with_readibility_measures\n",
      "dataset size is: (1680, 10)\n",
      "columns: Index(['Flesch_Reading_Ease_Value', 'Coleman_Liau_Index_Value',\n",
      "       'Dale_Chall_Readability_Score', 'Code_Count', 'Latex_Count',\n",
      "       'Text_Length', 'Punc_Rate', 'Polarity', 'Subjectivity', 'ScoreLabel'],\n",
      "      dtype='object')\n",
      "\n",
      "SVM\n",
      "*****BOF: ai_posts_with_readibility_measures*****\n",
      "\n",
      "Report for: GridSearched SVM\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "        low       0.61      0.96      0.75       205\n",
      "       high       0.44      0.05      0.10       131\n",
      "\n",
      "avg / total       0.54      0.60      0.49       336\n",
      "\n",
      "Confusion Matrix\n",
      " [[196   9]\n",
      " [124   7]]\n",
      "F1 Score: 0.0952380952381\n",
      "Accuracy: 0.604166666667\n",
      "Log Loss: 13.6716204074\n",
      "AUC Score: 0.523291752001\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/amanmahato/anaconda3/lib/python3.6/site-packages/matplotlib/axes/_axes.py:545: UserWarning: No labelled objects found. Use label='...' kwarg on individual plots.\n",
      "  warnings.warn(\"No labelled objects found. \"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11a2910f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*****EOF: ai_posts_with_readibility_measures*****\n"
     ]
    }
   ],
   "source": [
    "svm_run(\"ai_posts_with_readibility_measures\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running for iot_posts_with_readibility_measures\n",
      "dataset size is: (1048, 10)\n",
      "columns: Index(['Flesch_Reading_Ease_Value', 'Coleman_Liau_Index_Value',\n",
      "       'Dale_Chall_Readability_Score', 'Code_Count', 'Latex_Count',\n",
      "       'Text_Length', 'Punc_Rate', 'Polarity', 'Subjectivity', 'ScoreLabel'],\n",
      "      dtype='object')\n",
      "\n",
      "SVM\n",
      "*****BOF: iot_posts_with_readibility_measures*****\n",
      "\n",
      "Report for: GridSearched SVM\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "        low       0.61      0.48      0.54        95\n",
      "       high       0.64      0.75      0.69       115\n",
      "\n",
      "avg / total       0.63      0.63      0.62       210\n",
      "\n",
      "Confusion Matrix\n",
      " [[46 49]\n",
      " [29 86]]\n",
      "F1 Score: 0.688\n",
      "Accuracy: 0.628571428571\n",
      "Log Loss: 12.828874948\n",
      "AUC Score: 0.61528604119\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/amanmahato/anaconda3/lib/python3.6/site-packages/matplotlib/axes/_axes.py:545: UserWarning: No labelled objects found. Use label='...' kwarg on individual plots.\n",
      "  warnings.warn(\"No labelled objects found. \"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMcZk+n/5nTfvTyFCJAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11a790198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*****EOF: iot_posts_with_readibility_measures*****\n"
     ]
    }
   ],
   "source": [
    "svm_run(\"iot_posts_with_readibility_measures\")"
   ]
  }
 ],
 "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}