Confusion Matrix (Image Testing HAAR & MAR).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ImgNo</th>\n",
       "      <th>ActualSmile</th>\n",
       "      <th>PredictedSmile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10007</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10009</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10011</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ImgNo  ActualSmile  PredictedSmile\n",
       "0    100            1               0\n",
       "1   1000            1               0\n",
       "2  10007            1               0\n",
       "3  10009            1               0\n",
       "4  10011            1               0"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sn\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "img_haar = pd.read_csv(\"Haar_Image_Actual_Predicted.csv\") \n",
    "img_mar = pd.read_csv(\"Mar_Image_Actual_Predicted.csv\") \n",
    "\n",
    "img_haar.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ImgNo</th>\n",
       "      <th>ActualSmile</th>\n",
       "      <th>PredictedSmile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10007</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10009</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10011</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ImgNo  ActualSmile  PredictedSmile\n",
       "0    100            1               0\n",
       "1   1000            1               0\n",
       "2  10007            1               0\n",
       "3  10009            1               0\n",
       "4  10011            1               0"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img_mar.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted     0    1\n",
      "Actual              \n",
      "0          9459   16\n",
      "1          3586  104\n"
     ]
    }
   ],
   "source": [
    "haar_confusion_matrix = pd.crosstab(img_haar['ActualSmile'], img_haar['PredictedSmile'], rownames=['Actual'], colnames=['Predicted'])\n",
    "print (haar_confusion_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "haar_TN = 9459\n",
    "haar_FN = 3586\n",
    "haar_FP = 16\n",
    "haar_TP = 104"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted     0     1\n",
      "Actual               \n",
      "0          5033  4442\n",
      "1          2536  1154\n"
     ]
    }
   ],
   "source": [
    "mar_confusion_matrix = pd.crosstab(img_mar['ActualSmile'], img_mar['PredictedSmile'], rownames=['Actual'], colnames=['Predicted'])\n",
    "print (mar_confusion_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "mar_TN = 5033\n",
    "mar_FN = 2536\n",
    "mar_FP = 4442\n",
    "mar_TP = 1154"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.heatmap(haar_confusion_matrix, annot=True, fmt='d')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.heatmap(mar_confusion_matrix, annot=True, fmt='d')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "population = 13165 #total positive n negative dataset\n",
    "\n",
    "haar_accuracy = (haar_TP + haar_TN) / population\n",
    "haar_precision = haar_TP / (haar_TP + haar_FP)\n",
    "haar_recall = haar_TP / (haar_TP + haar_FN)\n",
    "if (haar_precision + haar_recall) == 0:\n",
    "    haar_f1_score = 0.0\n",
    "else:\n",
    "    haar_f1_score = 2 * (haar_precision * haar_recall) / (haar_precision + haar_recall)\n",
    "\n",
    "\n",
    "mar_accuracy = (mar_TP + mar_TN) / population\n",
    "mar_precision = mar_TP / (mar_TP + mar_FP)\n",
    "mar_recall = mar_TP / (mar_TP + mar_FN)\n",
    "mar_f1_score = 2 * (mar_precision * mar_recall) / (mar_precision + mar_recall)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy HAAR (Image): 0.7263957462969997\n",
      "Precision HAAR (Image): 0.8666666666666667\n",
      "Recall HAAR (Image): 0.028184281842818428\n",
      "F1-Score HAAR (Image): 0.05459317585301837\n",
      " \n",
      "Accuracy MAR (Image): 0.4699582225598177\n",
      "Precision MAR (Image): 0.20621872766261615\n",
      "Recall MAR (Image): 0.31273712737127374\n",
      "F1-Score MAR (Image): 0.24854619857850527\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy HAAR (Image): \" + str(haar_accuracy))\n",
    "print(\"Precision HAAR (Image): \" + str(haar_precision))\n",
    "print(\"Recall HAAR (Image): \" + str(haar_recall))\n",
    "print(\"F1-Score HAAR (Image): \" + str(haar_f1_score))\n",
    "print(\" \")\n",
    "print(\"Accuracy MAR (Image): \" + str(mar_accuracy))\n",
    "print(\"Precision MAR (Image): \" + str(mar_precision))\n",
    "print(\"Recall MAR (Image): \" + str(mar_recall))\n",
    "print(\"F1-Score MAR (Image): \" + str(mar_f1_score))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "population: 13165\n",
      "P: 3690\n",
      "N: 9475\n",
      "PositiveTest: 120\n",
      "NegativeTest: 13045\n",
      "TP: 104\n",
      "TN: 9459\n",
      "FP: 16\n",
      "FN: 3586\n",
      "TPR: 0.028184281842818428\n",
      "TNR: 0.998311345646438\n",
      "PPV: 0.8666666666666667\n",
      "NPV: 0.7251054043694902\n",
      "FPR: 0.0016886543535620053\n",
      "FDR: 0.13333333333333333\n",
      "FNR: 0.9718157181571816\n",
      "ACC: 0.7263957462969997\n",
      "F1_score: 0.05459317585301837\n",
      "MCC: 0.125217300532793\n",
      "informedness: 0.02649562748925649\n",
      "markedness: 0.5917720710361569\n",
      "prevalence: 0.28028864413216864\n",
      "LRP: 16.690379403794037\n",
      "LRN: 0.9734595548725337\n",
      "DOR: 17.14542665923034\n",
      "FOR: 0.2748945956305098\n"
     ]
    }
   ],
   "source": [
    "from pandas_ml import ConfusionMatrix\n",
    "\n",
    "Haar_Confusion_Matrix = ConfusionMatrix(img_haar['ActualSmile'], img_haar['PredictedSmile'])\n",
    "Haar_Confusion_Matrix.print_stats()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "population: 13165\n",
      "P: 3690\n",
      "N: 9475\n",
      "PositiveTest: 5596\n",
      "NegativeTest: 7569\n",
      "TP: 1154\n",
      "TN: 5033\n",
      "FP: 4442\n",
      "FN: 2536\n",
      "TPR: 0.31273712737127374\n",
      "TNR: 0.5311873350923483\n",
      "PPV: 0.20621872766261615\n",
      "NPV: 0.6649491346280882\n",
      "FPR: 0.4688126649076517\n",
      "FDR: 0.7937812723373838\n",
      "FNR: 0.6872628726287263\n",
      "ACC: 0.4699582225598177\n",
      "F1_score: 0.24854619857850527\n",
      "MCC: -0.1418010759653783\n",
      "informedness: -0.15607553753637804\n",
      "markedness: -0.12883213770929558\n",
      "prevalence: 0.28028864413216864\n",
      "LRP: 0.6670833592622284\n",
      "LRN: 1.2938239058528078\n",
      "DOR: 0.515590534573195\n",
      "FOR: 0.33505086537191175\n"
     ]
    }
   ],
   "source": [
    "Mar_Confusion_Matrix = ConfusionMatrix(img_mar['ActualSmile'], img_mar['PredictedSmile'])\n",
    "Mar_Confusion_Matrix.print_stats()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Methods')"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "haar_vs_maar_img = pd.DataFrame({\n",
    "    \"Haar\":[haar_accuracy, haar_precision, haar_recall, haar_f1_score],\n",
    "    \"MAR\":[mar_accuracy, mar_precision, mar_recall, mar_f1_score],\n",
    "    }, \n",
    "    index=[\"Accuracy\", \"Precision\", \"Recall\", \"F1-Score\"]\n",
    ")\n",
    "\n",
    "haar_vs_maar_img.plot(kind=\"bar\")\n",
    "plt.title(\"Comparison of Evaluation Metrics (Image Testing: Haar Vs MAR)\")\n",
    "plt.xlabel(\"Methods\")"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}