Evaluation of a Classification Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluation of the Model\n",
    "\n",
    "### Three Ways to Evaluate Scikit Learn Model / Estimator.\n",
    "\n",
    "**A.Estimator Score Method**\n",
    "\n",
    "**B.The Scoring Parameter**\n",
    "\n",
    "**C.Problem Specific Metric Function**\n",
    "\n",
    "## Evaluation of a Classification Model\n",
    "\n",
    "### Classification Model Evaluation Metrics\n",
    "\n",
    "**1.Accuracy**\n",
    "\n",
    "**2.Confusion Matrix**\n",
    "\n",
    "**3.Classification Report**\n",
    "\n",
    "**4.Area Under ROC**\n",
    "\n",
    "\n",
    "**Import Libraries**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "warnings.simplefilter('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Import Data Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>cp</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>chol</th>\n",
       "      <th>fbs</th>\n",
       "      <th>restecg</th>\n",
       "      <th>thalach</th>\n",
       "      <th>exang</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>slope</th>\n",
       "      <th>ca</th>\n",
       "      <th>thal</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>145</td>\n",
       "      <td>233</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>37</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>130</td>\n",
       "      <td>250</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>187</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>130</td>\n",
       "      <td>204</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>172</td>\n",
       "      <td>0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>56</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>120</td>\n",
       "      <td>236</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>178</td>\n",
       "      <td>0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>120</td>\n",
       "      <td>354</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>163</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  slope  \\\n",
       "0   63    1   3       145   233    1        0      150      0      2.3      0   \n",
       "1   37    1   2       130   250    0        1      187      0      3.5      0   \n",
       "2   41    0   1       130   204    0        0      172      0      1.4      2   \n",
       "3   56    1   1       120   236    0        1      178      0      0.8      2   \n",
       "4   57    0   0       120   354    0        1      163      1      0.6      2   \n",
       "\n",
       "   ca  thal  target  \n",
       "0   0     1       1  \n",
       "1   0     2       1  \n",
       "2   0     2       1  \n",
       "3   0     2       1  \n",
       "4   0     2       1  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Heart = pd.read_csv('Heart Disease.csv')\n",
    "\n",
    "Heart.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Checking for Information and whether there is any Categorical Column**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 303 entries, 0 to 302\n",
      "Data columns (total 14 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   age       303 non-null    int64  \n",
      " 1   sex       303 non-null    int64  \n",
      " 2   cp        303 non-null    int64  \n",
      " 3   trestbps  303 non-null    int64  \n",
      " 4   chol      303 non-null    int64  \n",
      " 5   fbs       303 non-null    int64  \n",
      " 6   restecg   303 non-null    int64  \n",
      " 7   thalach   303 non-null    int64  \n",
      " 8   exang     303 non-null    int64  \n",
      " 9   oldpeak   303 non-null    float64\n",
      " 10  slope     303 non-null    int64  \n",
      " 11  ca        303 non-null    int64  \n",
      " 12  thal      303 non-null    int64  \n",
      " 13  target    303 non-null    int64  \n",
      "dtypes: float64(1), int64(13)\n",
      "memory usage: 33.3 KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(Heart.info())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Splitting the Data Set into Independent Features and Dependent Feature Labels**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = Heart.drop('target',axis = 'columns')\n",
    "\n",
    "Y = Heart['target']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Splitting the Data Set into Training and Testing Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_Train, X_Test, Y_Train, Y_Test = train_test_split(X, Y, test_size = 0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**X_Train : The Set of Data to be Trained.**\n",
    "\n",
    "**Y_Train : The Set of Labels (Target Variable) of the X_Train Data, This Set of Data is also Trained with X_Train.**\n",
    "\n",
    "**X_Test : The Set of Data on which the Trained Model is to be Tested to Check the Accuracy of Model.**\n",
    "\n",
    "**Y_Test : The Set of Data which will be Compared with the Results of Trained Model to Check Accuracy as well as other Evaluation Metrics.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "70% of Train Set : (212, 13) and Labels of Train Set : (212,)\n",
      "30% of Test Set : (91, 13) and Labels of Test Set : (91,)\n"
     ]
    }
   ],
   "source": [
    "print(f'70% of Train Set : {X_Train.shape} and Labels of Train Set : {Y_Train.shape}')\n",
    "\n",
    "print(f'30% of Test Set : {X_Test.shape} and Labels of Test Set : {Y_Test.shape}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### According to Scikit Learn Algorithm Map\n",
    "\n",
    "**First we will Try Linear Support Vector Classifier**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import LinearSVC"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Instantiate Linear Support vector Classifier Object**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "SVC = LinearSVC()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fit the Data to the Classifier to Find Patterns, in short Train the Model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearSVC()"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SVC.fit(X_Train, Y_Train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A. Estimator Score Method\n",
    "\n",
    "**Estimators have a Score Method provided as a Default Evaluation**\n",
    "\n",
    "**For Classifiers it's Mean Accuracy**\n",
    "\n",
    "**Check Mean Accuracy / Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Linear Support Vector Classifier Mean Accuracy : 49.45%\n"
     ]
    }
   ],
   "source": [
    "print(f'Linear Support Vector Classifier Mean Accuracy : {SVC.score(X_Test, Y_Test)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prediction Made by the Linear Support Vector Classifier on X_Test Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "SVC_Prediction = SVC.predict(X_Test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation of Linear Support Vector Classifier :\n",
    "\n",
    "**Linear Support Vector Classifier (Classification Model) Evaluation Metrics**\n",
    "\n",
    "**Accuracy**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Accuracy of Model on Train Set : 54.25%\n",
      "\n",
      "Mean Accuracy of Model on Test Set : 49.45%\n"
     ]
    }
   ],
   "source": [
    "print(f'Mean Accuracy of Model on Train Set : {SVC.score(X_Train, Y_Train)*100:.2f}%')\n",
    "print()\n",
    "print(f'Mean Accuracy of Model on Test Set : {SVC.score(X_Test, Y_Test)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Confusion Matrix**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[41  0]\n",
      " [46  4]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "print(f'{confusion_matrix(Y_Test, SVC_Prediction)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Visualize Confusion Matrix**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted Labels   0  1\n",
      "Actual Labels          \n",
      "0                 41  0\n",
      "1                 46  4\n"
     ]
    }
   ],
   "source": [
    "print(pd.crosstab(Y_Test, SVC_Prediction,\n",
    "            rownames = ['Actual Labels'],\n",
    "            colnames = ['Predicted Labels']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Classification Report**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.47      1.00      0.64        41\n",
      "           1       1.00      0.08      0.15        50\n",
      "\n",
      "    accuracy                           0.49        91\n",
      "   macro avg       0.74      0.54      0.39        91\n",
      "weighted avg       0.76      0.49      0.37        91\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "print(f'{classification_report(Y_Test, SVC_Prediction)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Accuracy can be Improved further using other Model \n",
    "\n",
    "**According to the Scikit Learn Algorithm Map We will Try Random Forest Classifier**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Instantiate Random Forest Classifier Object**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "RFC = RandomForestClassifier(n_estimators = 100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fit the Data to Random Forest Classifier to Find the Patterns between X_Train and Y_train Sets**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier()"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RFC.fit(X_Train, Y_Train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A. Estimator Score Method\n",
    "\n",
    "**Mean Accuracy of Random Forest Classifier**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Classifier Mean Accuracy on Train Set : 100.00%\n",
      "\n",
      "Random Forest Classifier Mean Accuracy on Test Set : 82.42%\n"
     ]
    }
   ],
   "source": [
    "print(f'Random Forest Classifier Mean Accuracy on Train Set : {RFC.score(X_Train, Y_Train)*100:.2f}%')\n",
    "print()\n",
    "print(f'Random Forest Classifier Mean Accuracy on Test Set : {RFC.score(X_Test, Y_Test)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Prediction made by Random Forest Classifier on X_Test Data Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "RFC_Prediction = RFC.predict(X_Test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Probability of Prediction**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "RFC_Probability = RFC.predict_proba(X_Test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.92 0.08]\n",
      " [0.34 0.66]\n",
      " [0.44 0.56]\n",
      " [0.81 0.19]\n",
      " [0.27 0.73]]\n"
     ]
    }
   ],
   "source": [
    "print(np.array(RFC.predict_proba(X_Test)[:5]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Compare X_Test Data Set with Y_Test Labels**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>cp</th>\n",
       "      <th>trestbps</th>\n",
       "      <th>chol</th>\n",
       "      <th>fbs</th>\n",
       "      <th>restecg</th>\n",
       "      <th>thalach</th>\n",
       "      <th>exang</th>\n",
       "      <th>oldpeak</th>\n",
       "      <th>slope</th>\n",
       "      <th>ca</th>\n",
       "      <th>thal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>57</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>276</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112</td>\n",
       "      <td>1</td>\n",
       "      <td>0.6</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>228</th>\n",
       "      <td>59</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>170</td>\n",
       "      <td>288</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>159</td>\n",
       "      <td>0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>57</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>150</td>\n",
       "      <td>126</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>173</td>\n",
       "      <td>0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>56</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>134</td>\n",
       "      <td>409</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>150</td>\n",
       "      <td>1</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>71</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>110</td>\n",
       "      <td>265</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>130</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     age  sex  cp  trestbps  chol  fbs  restecg  thalach  exang  oldpeak  \\\n",
       "179   57    1   0       150   276    0        0      112      1      0.6   \n",
       "228   59    1   3       170   288    0        0      159      0      0.2   \n",
       "111   57    1   2       150   126    1        1      173      0      0.2   \n",
       "246   56    0   0       134   409    0        0      150      1      1.9   \n",
       "60    71    0   2       110   265    1        0      130      0      0.0   \n",
       "\n",
       "     slope  ca  thal  \n",
       "179      1   1     1  \n",
       "228      1   0     3  \n",
       "111      2   1     3  \n",
       "246      1   2     3  \n",
       "60       2   1     2  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_Test[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 0 1 0 1]\n"
     ]
    }
   ],
   "source": [
    "print(f'{np.array(Y_Test[:5])}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluation of Random Forest Classifier :\n",
    "\n",
    "**Random Forest Classifier (Classification Model) Evaluation Metrics**\n",
    "\n",
    "**1.Accuracy**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Classifier Mean Accuracy on Test Data Set : 82.42%\n"
     ]
    }
   ],
   "source": [
    "print(f'Random Forest Classifier Mean Accuracy on Test Data Set : {RFC.score(X_Test, Y_Test)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**2.Confusion Matrix**\n",
    "\n",
    "**Compare the Labels that a Model Predicts and the Actual Labels it was supposed to Predict.**\n",
    "\n",
    "**Simply its a Comparison between The Actual and Predicted Labels, i.e Y_Test with Prediction.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix of Random Forest Classifier :\n",
      "\n",
      "[[33  8]\n",
      " [ 8 42]]\n",
      "\n",
      "Predicted Labels   0   1\n",
      "Actual Labels           \n",
      "0                 33   8\n",
      "1                  8  42\n",
      "\n",
      "Total : 91\n",
      "\n",
      "Total Labels : 91\n"
     ]
    }
   ],
   "source": [
    "print(f'Confusion Matrix of Random Forest Classifier :')\n",
    "\n",
    "print()\n",
    "\n",
    "print(f'{confusion_matrix(Y_Test, RFC_Prediction)}')\n",
    "\n",
    "print()\n",
    "\n",
    "print(pd.crosstab(Y_Test, RFC_Prediction,\n",
    "            rownames = ['Actual Labels'],\n",
    "            colnames = ['Predicted Labels']))\n",
    "\n",
    "print()\n",
    "\n",
    "print(f'Total : {35 + 41 + 9 + 6}')\n",
    "\n",
    "print()\n",
    "\n",
    "print(f'Total Labels : {len(RFC_Prediction)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src = 'Confusion Matrix Anatomy.png'>\n",
    "\n",
    "**1.True Positive : Model Predicts 1 when Truth (Actual) is 1**\n",
    "\n",
    "**2.False Positive : Model Predicts 1 when Truth (Actual) is 0**\n",
    "\n",
    "**3.True Negative : Model Predicts 0 when Truth (Actual) is 0**\n",
    "\n",
    "**4.False Negative : Model Predicts 0 when Truth (Actual) is 1**\n",
    "\n",
    "<img src ='Confusion Matrix.png'>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Heat Map of Confusion Matrix**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set Font Scale\n",
    "sns.set(font_scale=1.5)\n",
    "\n",
    "# Create Confusion Matrix \n",
    "Confusion_Matrix = confusion_matrix(Y_Test, RFC_Prediction)\n",
    "\n",
    "# Plot the Heatmap\n",
    "sns.heatmap(Confusion_Matrix);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**For More Intuitive Heatmap**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def Plot_Confusion_Matrix(Confusion_Matrix) :\n",
    "    '''\n",
    "    \n",
    "    Plot Confusion Matrix using Seaborn's Heatmap\n",
    "    \n",
    "    Parameter\n",
    "    ---------\n",
    "    \n",
    "    Confusion Matrix.\n",
    "    \n",
    "    '''\n",
    "    \n",
    "    Fig, Axes = plt.subplots(figsize=(3,3))\n",
    "    \n",
    "    Axes = sns.heatmap(Confusion_Matrix, annot=True, cbar = False)\n",
    "    \n",
    "    plt.xlabel('True Labels')\n",
    "    \n",
    "    plt.ylabel('Predicted Labels')\n",
    "    \n",
    "    # Fix the Brok en Annotations\n",
    "    \n",
    "    Bottom, Top = Axes.get_ylim()\n",
    "    \n",
    "    Axes.set_ylim(Bottom, Top)\n",
    "    \n",
    "    plt.show()\n",
    "    \n",
    "Plot_Confusion_Matrix(Confusion_Matrix)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**3.Classification Report**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report of Random Forest Classifier :\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.80      0.80      0.80        41\n",
      "           1       0.84      0.84      0.84        50\n",
      "\n",
      "    accuracy                           0.82        91\n",
      "   macro avg       0.82      0.82      0.82        91\n",
      "weighted avg       0.82      0.82      0.82        91\n",
      "\n",
      "\n",
      "Classification Report in form of Data Frame : \n",
      "                   0      1  accuracy  macro avg  weighted avg\n",
      "precision   0.804878   0.84  0.824176   0.822439      0.824176\n",
      "recall      0.804878   0.84  0.824176   0.822439      0.824176\n",
      "f1-score    0.804878   0.84  0.824176   0.822439      0.824176\n",
      "support    41.000000  50.00  0.824176  91.000000     91.000000\n"
     ]
    }
   ],
   "source": [
    "print(f'Classification Report of Random Forest Classifier :')\n",
    "\n",
    "print(f'\\n{classification_report(Y_Test, RFC_Prediction)}')\n",
    "\n",
    "print(f'\\nClassification Report in form of Data Frame : ')\n",
    "\n",
    "print(pd.DataFrame(classification_report(Y_Test, RFC_Prediction, output_dict=True)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Precision : Positive Predicted Value**\n",
    "\n",
    "**Recall : True Positive Rate**\n",
    "\n",
    "**F1 Score : Combination of Precision and Recall (Average)**\n",
    "\n",
    "**Support : Number of Samples Each Metric is Calculated on.**\n",
    "\n",
    "**Accuracy : Accuracy of Model, Perfect Accuracy is 1.0 or 100%**\n",
    "\n",
    "**Macro Average : Average between Classes of Labels (0 and 1), Classes must be Balanced (Almost Equal Number of 0's and 1's )**\n",
    "\n",
    "**[Metrics and Scoring](https://scikit-learn.org/stable/modules/model_evaluation.html)**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### To Summarize Classification Metrics :\n",
    "\n",
    "**Accuracy : A Good Measure to Start with if the Classes are Balanced (Same Amount of 0 and 1 Labels.**\n",
    "\n",
    "**Precision and Recall becomes more Important when classes are Imbalanced.**\n",
    "\n",
    "**If False Positive Predictions are Worse than False Negatives, Aim for High Precisions.**\n",
    "\n",
    "**If False Negative Predictions are Worse than False positives, Aim for High Recall.**\n",
    "\n",
    "**F1 Score is a Combination of Precision and Recall.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**4.Area under Curve or ROC - Receiver Under Characteristic Curve (AUC/ROC)**\n",
    "\n",
    "**AUC/ROC curves are comparison of a Model's True Positive Rate (TPR) and False Positive Rate (FPR)**\n",
    "\n",
    "**1.True Positive : Model Predicts 1 when Truth (Actual) is 1**\n",
    "\n",
    "**2.False Positive : Model Predicts 1 when Truth (Actual) is 0**\n",
    "\n",
    "**3.True Negative : Model Predicts 0 when Truth (Actual) is 0**\n",
    "\n",
    "**4.False Negative : Model Predicts 0 when Truth (Actual) is 1**\n",
    "\n",
    "<img src ='Confusion Matrix.png'>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Make Predictions with Probabilities**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "Positive_Probability = RFC_Probability[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.08 0.66 0.56 0.19 0.73 0.86 0.7  0.06 0.04 0.57]\n"
     ]
    }
   ],
   "source": [
    "print(Positive_Probability[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Calculate FPR, TPR and Thresholds**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "FPR, TPR, Thresholds = roc_curve(Y_Test, Positive_Probability) \n",
    "\n",
    "# Y_True : Y_Test Labels \n",
    "\n",
    "# Y_Score : Probability Estimates of the Positive Class i.e. (Label : 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Check the False Positive Rate**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.         0.         0.         0.         0.         0.\n",
      " 0.         0.02439024 0.02439024 0.02439024]\n"
     ]
    }
   ],
   "source": [
    "print(FPR[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Check the True Positive Rate**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.   0.02 0.04 0.08 0.12 0.16 0.18 0.18 0.22 0.26]\n"
     ]
    }
   ],
   "source": [
    "print(TPR[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Creating a Function to Plot ROC Curve**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def Plot_ROC_Curve(FPR, TPR) :\n",
    "    ''' \n",
    "    Plots Receiver Under Characteristic Curve (ROC Curve)\n",
    "    \n",
    "    Parameters \n",
    "    ----------\n",
    "    \n",
    "    FPR : False Positive Rate.\n",
    "    \n",
    "    TPR : True Positive Rate.\n",
    "    \n",
    "    '''\n",
    "    \n",
    "    # Theme Style\n",
    "    \n",
    "    plt.style.use('seaborn-whitegrid')\n",
    "    \n",
    "    # Plot ROC Curve\n",
    "    \n",
    "    plt.plot(FPR, TPR, color = 'orange', label = 'ROC')\n",
    "    \n",
    "    # Plot Line with no Predictive Powers (Base Line)\n",
    "    \n",
    "    plt.plot([0,1], [0,1], color = 'blue', linestyle = '--', label = 'Guessing')\n",
    "    \n",
    "    # Customize the Plot\n",
    "    \n",
    "    plt.xlabel('False Positive Rate (FPR)')\n",
    "    \n",
    "    plt.ylabel('True Positive Rate (TPR)')\n",
    "    \n",
    "    plt.title('Receiver Under Characteristic Curve (ROC Curve)')\n",
    "    \n",
    "    plt.legend()\n",
    "    \n",
    "    plt.show()\n",
    "    \n",
    "Plot_ROC_Curve(FPR, TPR)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**ROC AUC Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC Score : 90.68%\n"
     ]
    }
   ],
   "source": [
    "print(f'ROC AUC Score : {roc_auc_score(Y_Test, Positive_Probability)*100:.2f}%') # Max AUC Score we can get is 100%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plot Perfect ROC Curve**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "FPR, TPR, Thresholds = roc_curve(Y_Test, Y_Test) # Pass Both Parameters  Y_Test Labels.\n",
    "\n",
    "Plot_ROC_Curve(FPR, TPR)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Perfect AUC Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Perfect AUC Score : 100.00%\n"
     ]
    }
   ],
   "source": [
    "print(f'Perfect AUC Score : {roc_auc_score(Y_Test, Y_Test)*100:.2f}%') # Pass Both Parameters  Y_Test Labels."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Perfect ROC Curve and AUC Score mean Perfect Model with only True Positives and No False Positives.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using the Scoring Parameter"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Let's Try Evaluation of Random Forest Classifier using K Fold Cross Validation**\n",
    "\n",
    "<img src = 'Cross Validation.png'>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**K Fold Cross Validation can help us to Find Best Estimator we can get for our Data Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "np.random.seed(42)\n",
    "\n",
    "CVS = cross_val_score(RFC, X, Y, cv=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Model is Trained on 5 Different Versions of Training Data Set, and Evaluated on 5 Different Versions of Test Data Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Score : [0.81967213 0.90163934 0.83606557 0.78333333 0.78333333]\n"
     ]
    }
   ],
   "source": [
    "print(f'Cross Validation Score : {CVS}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**K Fold Cross Validation Mean Score for Random Forest Classifier :**\n",
    "\n",
    "## B.The Scoring Parameter :\n",
    "\n",
    "**Model Evaluation Tools using Cross Value Score or Grid Search CV rely on Scoring Parameters.**\n",
    "\n",
    "**Default Scoring Parameter = None (Accuracy)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Mean Score : 82.48% \n"
     ]
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "CVS = cross_val_score(RFC, X, Y, cv=5, scoring=None)\n",
    "\n",
    "print(f'Cross Validation Mean Score : {np.mean(CVS)*100:.2f}% ')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Scoring Parameter = Accuracy**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Mean Score : 82.48%\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "CVS_Accuracy = cross_val_score(RFC,X, Y, cv = 5, scoring='accuracy')\n",
    "\n",
    "print(f'Cross Validation Mean Score : {np.mean(CVS_Accuracy)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Scoring Parameter = Precision**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Mean Score : 83.30%\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "CVS_Precision = cross_val_score(RFC, X, Y, cv = 5, scoring = 'precision')\n",
    "\n",
    "print(f'Cross Validation Mean Score : {np.mean(CVS_Precision)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " **Scoring Parameter = Recall**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Mean Score : 85.45%\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "CVS_Recall = cross_val_score(RFC, X, Y,cv = 5, scoring = 'recall')\n",
    "\n",
    "print(f'Cross Validation Mean Score : {np.mean(CVS_Recall)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Scoring Parameter = F1 Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Mean Score : 84.27%\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "\n",
    "CVS_F1 = cross_val_score(RFC, X, Y, scoring = 'f1')\n",
    "\n",
    "print(f'Cross Validation Mean Score : {np.mean(CVS_F1)*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## C. Problem Specific Metric Function\n",
    "\n",
    "**Classification Evaluation Functions**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score : 82.42%\n",
      "\n",
      "Precision Score : 84.00%\n",
      "\n",
      "Recall Score : 84.00%\n",
      "\n",
      "F1 Score : 84.00%\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy Score : {accuracy_score(Y_Test, RFC_Prediction)*100:.2f}%')\n",
    "\n",
    "print(f'\\nPrecision Score : {precision_score(Y_Test, RFC_Prediction)*100:.2f}%')\n",
    "\n",
    "print(f'\\nRecall Score : {recall_score(Y_Test, RFC_Prediction)*100:.2f}%')\n",
    "\n",
    "print(f'\\nF1 Score : {f1_score(Y_Test, RFC_Prediction)*100:.2f}%')"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}