main.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loan Approval Prediction Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 614 entries, 0 to 613\n",
      "Data columns (total 13 columns):\n",
      "Loan_ID              614 non-null object\n",
      "Gender               601 non-null object\n",
      "Married              611 non-null object\n",
      "Dependents           599 non-null object\n",
      "Education            614 non-null object\n",
      "Self_Employed        582 non-null object\n",
      "ApplicantIncome      614 non-null int64\n",
      "CoapplicantIncome    614 non-null float64\n",
      "LoanAmount           592 non-null float64\n",
      "Loan_Amount_Term     600 non-null float64\n",
      "Credit_History       564 non-null float64\n",
      "Property_Area        614 non-null object\n",
      "Loan_Status          614 non-null object\n",
      "dtypes: float64(4), int64(1), object(8)\n",
      "memory usage: 62.4+ KB\n"
     ]
    }
   ],
   "source": [
    "train_df  = pd.read_csv('./Dataset.csv')\n",
    "train_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Droping Loan_ID \n",
    "train_df = train_df.drop(columns=['Loan_ID'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Gender', 'Married', 'Dependents', 'Education', 'Self_Employed', 'Property_Area', 'Credit_History', 'Loan_Amount_Term']\n"
     ]
    }
   ],
   "source": [
    "categorical_columns = ['Gender', 'Married', 'Dependents', 'Education', 'Self_Employed', 'Property_Area','Credit_History','Loan_Amount_Term']\n",
    "print(categorical_columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['ApplicantIncome', 'CoapplicantIncome', 'LoanAmount']\n"
     ]
    }
   ],
   "source": [
    "numerical_columns = ['ApplicantIncome','CoapplicantIncome','LoanAmount']\n",
    "print(numerical_columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data Visualization Part\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x1080 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Now, let's analyze Categorical Columns:\n",
    "fig,axes = plt.subplots(4,2,figsize=(12,15))\n",
    "for idx,cat_col in enumerate(categorical_columns):\n",
    "    row,col = idx//2,idx%2\n",
    "    sns.countplot(x=cat_col,data=train_df,hue='Loan_Status',ax=axes[row,col])\n",
    "\n",
    "\n",
    "plt.subplots_adjust(hspace=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       ApplicantIncome  CoapplicantIncome  LoanAmount\n",
      "count       614.000000         614.000000  592.000000\n",
      "mean       5403.459283        1621.245798  146.412162\n",
      "std        6109.041673        2926.248369   85.587325\n",
      "min         150.000000           0.000000    9.000000\n",
      "25%        2877.500000           0.000000  100.000000\n",
      "50%        3812.500000        1188.500000  128.000000\n",
      "75%        5795.000000        2297.250000  168.000000\n",
      "max       81000.000000       41667.000000  700.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1224x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Now, let's also analyze Numerical Columns:\n",
    "fig,axes = plt.subplots(3,1,figsize=(17,5))\n",
    "for idx,cat_col in enumerate(numerical_columns):\n",
    "    sns.boxplot(y=cat_col,data=train_df,x='Loan_Status',ax=axes[idx])\n",
    "    \n",
    "print(train_df[numerical_columns].describe())\n",
    "plt.subplots_adjust(hspace=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Processing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>ApplicantIncome</th>\n",
       "      <th>CoapplicantIncome</th>\n",
       "      <th>LoanAmount</th>\n",
       "      <th>Loan_Amount_Term</th>\n",
       "      <th>Credit_History</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>Married_Yes</th>\n",
       "      <th>Dependents_1</th>\n",
       "      <th>Dependents_2</th>\n",
       "      <th>Dependents_3+</th>\n",
       "      <th>Education_Not Graduate</th>\n",
       "      <th>Self_Employed_Yes</th>\n",
       "      <th>Property_Area_Semiurban</th>\n",
       "      <th>Property_Area_Urban</th>\n",
       "      <th>Loan_Status_Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5849</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>4583</td>\n",
       "      <td>1508.0</td>\n",
       "      <td>128.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2583</td>\n",
       "      <td>2358.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>141.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ApplicantIncome  CoapplicantIncome  LoanAmount  Loan_Amount_Term  \\\n",
       "0             5849                0.0         NaN             360.0   \n",
       "1             4583             1508.0       128.0             360.0   \n",
       "2             3000                0.0        66.0             360.0   \n",
       "3             2583             2358.0       120.0             360.0   \n",
       "4             6000                0.0       141.0             360.0   \n",
       "\n",
       "   Credit_History  Gender_Male  Married_Yes  Dependents_1  Dependents_2  \\\n",
       "0             1.0            1            0             0             0   \n",
       "1             1.0            1            1             1             0   \n",
       "2             1.0            1            1             0             0   \n",
       "3             1.0            1            1             0             0   \n",
       "4             1.0            1            0             0             0   \n",
       "\n",
       "   Dependents_3+  Education_Not Graduate  Self_Employed_Yes  \\\n",
       "0              0                       0                  0   \n",
       "1              0                       0                  0   \n",
       "2              0                       0                  1   \n",
       "3              0                       1                  0   \n",
       "4              0                       0                  0   \n",
       "\n",
       "   Property_Area_Semiurban  Property_Area_Urban  Loan_Status_Y  \n",
       "0                        0                    1              1  \n",
       "1                        0                    0              0  \n",
       "2                        0                    1              1  \n",
       "3                        0                    1              1  \n",
       "4                        0                    1              1  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Encoding Categorical data \n",
    "train_df_encoded = pd.get_dummies(train_df,drop_first=True) # Add dummy variable into dataframe\n",
    "train_df_encoded.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Split features and  Target variable \n",
    "X = train_df_encoded.drop(columns='Loan_Status_Y') # remove column Loan_Status_Y\n",
    "y = train_df_encoded['Loan_Status_Y']\n",
    "#print(X)\n",
    "#print(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Splitting into Train -Test Data\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.2,stratify=y,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Handling/Imputing Missing values\n",
    "\n",
    "from sklearn.impute import SimpleImputer # it helps to fill missing value using other data mean \n",
    "imp = SimpleImputer(strategy='mean')\n",
    "imp_train = imp.fit(X_train)\n",
    "X_train = imp_train.transform(X_train)\n",
    "X_test_imp = imp_train.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Data Set accuracy :-  1.0\n",
      "Training Data F1 Score :- 1.0\n",
      "Validation F1 Mean Score:-  0.6451230699949734\n",
      "Validation Accuracy  Score:-  0.6986473189713248\n"
     ]
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import accuracy_score , f1_score\n",
    "\n",
    "\n",
    "tree_clf= DecisionTreeClassifier()\n",
    "tree_clf.fit(X_train,y_train)\n",
    "y_pred = tree_clf.predict(X_train)\n",
    "\n",
    "print(\"Training Data Set accuracy :- \",accuracy_score(y_train,y_pred))\n",
    "print(\"Training Data F1 Score :-\",f1_score(y_train,y_pred))\n",
    "print(\"Validation F1 Mean Score:- \",cross_val_score(tree_clf,X_train,y_train,cv=5,scoring='f1_macro').mean())\n",
    "print(\"Validation Accuracy  Score:- \",cross_val_score(tree_clf,X_train,y_train,cv=5,scoring='accuracy').mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--> Overfitting Problem\n",
    "We can see from above metrics that Training Accuracy > Test Accuracy with default settings of Decision Tree classifier. Hence, model is overfit. We will try some Hyper-parameter tuning and see if it helps."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First let's try tuning 'Max_Depth' of tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0feced88d0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "training_accuracy = []\n",
    "val_accuracy = []\n",
    "training_f1 = []\n",
    "val_f1 = []\n",
    "tree_depths = []\n",
    "\n",
    "for depth in range(1,20):\n",
    "    tree_clf = DecisionTreeClassifier(max_depth=depth)\n",
    "    tree_clf.fit(X_train,y_train)\n",
    "    y_training_pred = tree_clf.predict(X_train)\n",
    "\n",
    "    training_acc = accuracy_score(y_train,y_training_pred)\n",
    "    train_f1 = f1_score(y_train,y_training_pred)\n",
    "    val_mean_f1 = cross_val_score(tree_clf,X_train,y_train,cv=5,scoring='f1_macro').mean()\n",
    "    val_mean_accuracy = cross_val_score(tree_clf,X_train,y_train,cv=5,scoring='accuracy').mean()\n",
    "    \n",
    "    training_accuracy.append(training_acc)\n",
    "    val_accuracy.append(val_mean_accuracy)\n",
    "    training_f1.append(train_f1)\n",
    "    val_f1.append(val_mean_f1)\n",
    "    tree_depths.append(depth)\n",
    "    \n",
    "\n",
    "Tuning_Max_depth = {\"Training Accuracy\": training_accuracy, \"Validation Accuracy\": val_accuracy, \"Training F1\": training_f1, \"Validation F1\":val_f1, \"Max_Depth\": tree_depths }\n",
    "Tuning_Max_depth_df = pd.DataFrame.from_dict(Tuning_Max_depth)\n",
    "\n",
    "plot_df = Tuning_Max_depth_df.melt('Max_Depth',var_name='Metrics',value_name=\"Values\")\n",
    "fig,ax = plt.subplots(figsize=(15,5))\n",
    "sns.pointplot(x=\"Max_Depth\", y=\"Values\",hue=\"Metrics\", data=plot_df,ax=ax)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From above graph, we can conclude that keeping 'Max_Depth' = 3 will yield optimum Test accuracy and F1 score Optimum Test Accuracy ~ 0.805; Optimum F1 Score: ~0.7"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visulazing Decision Tree with Max Depth = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: graphviz in ./anaconda3/lib/python3.7/site-packages (0.14.1)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install graphviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n<!-- Generated by graphviz version 2.40.1 (20161225.0304)\n -->\n<!-- Title: Tree Pages: 1 -->\n<svg width=\"734pt\" height=\"373pt\"\n viewBox=\"0.00 0.00 733.50 373.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 369)\">\n<title>Tree</title>\n<polygon fill=\"#ffffff\" stroke=\"transparent\" points=\"-4,4 -4,-369 729.5,-369 729.5,4 -4,4\"/>\n<!-- 0 -->\n<g id=\"node1\" class=\"node\">\n<title>0</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"408.5,-365 256.5,-365 256.5,-297 408.5,-297 408.5,-365\"/>\n<text text-anchor=\"middle\" x=\"332.5\" y=\"-349.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">Credit_History &lt;= 0.425</text>\n<text text-anchor=\"middle\" x=\"332.5\" y=\"-334.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.431</text>\n<text text-anchor=\"middle\" x=\"332.5\" y=\"-319.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 491</text>\n<text text-anchor=\"middle\" x=\"332.5\" y=\"-304.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [154, 337]</text>\n</g>\n<!-- 1 -->\n<g id=\"node2\" class=\"node\">\n<title>1</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"313.5,-261 169.5,-261 169.5,-193 313.5,-193 313.5,-261\"/>\n<text text-anchor=\"middle\" x=\"241.5\" y=\"-245.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">LoanAmount &lt;= 488.0</text>\n<text text-anchor=\"middle\" x=\"241.5\" y=\"-230.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.163</text>\n<text text-anchor=\"middle\" x=\"241.5\" y=\"-215.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 67</text>\n<text text-anchor=\"middle\" x=\"241.5\" y=\"-200.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [61, 6]</text>\n</g>\n<!-- 0&#45;&gt;1 -->\n<g id=\"edge1\" class=\"edge\">\n<title>0&#45;&gt;1</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M302.7032,-296.9465C294.847,-287.968 286.2788,-278.1758 278.1087,-268.8385\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"280.5829,-266.351 271.3637,-261.13 275.3148,-270.9606 280.5829,-266.351\"/>\n<text text-anchor=\"middle\" x=\"269.7008\" y=\"-282.3746\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">True</text>\n</g>\n<!-- 6 -->\n<g id=\"node7\" class=\"node\">\n<title>6</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"517,-261 332,-261 332,-193 517,-193 517,-261\"/>\n<text text-anchor=\"middle\" x=\"424.5\" y=\"-245.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">CoapplicantIncome &lt;= 9537.0</text>\n<text text-anchor=\"middle\" x=\"424.5\" y=\"-230.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.342</text>\n<text text-anchor=\"middle\" x=\"424.5\" y=\"-215.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 424</text>\n<text text-anchor=\"middle\" x=\"424.5\" y=\"-200.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [93, 331]</text>\n</g>\n<!-- 0&#45;&gt;6 -->\n<g id=\"edge6\" class=\"edge\">\n<title>0&#45;&gt;6</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M362.6242,-296.9465C370.5668,-287.968 379.2291,-278.1758 387.489,-268.8385\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"390.3038,-270.939 394.3081,-261.13 385.0608,-266.301 390.3038,-270.939\"/>\n<text text-anchor=\"middle\" x=\"395.8358\" y=\"-282.3833\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">False</text>\n</g>\n<!-- 2 -->\n<g id=\"node3\" class=\"node\">\n<title>2</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"185,-157 0,-157 0,-89 185,-89 185,-157\"/>\n<text text-anchor=\"middle\" x=\"92.5\" y=\"-141.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">CoapplicantIncome &lt;= 8115.0</text>\n<text text-anchor=\"middle\" x=\"92.5\" y=\"-126.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.14</text>\n<text text-anchor=\"middle\" x=\"92.5\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 66</text>\n<text text-anchor=\"middle\" x=\"92.5\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [61, 5]</text>\n</g>\n<!-- 1&#45;&gt;2 -->\n<g id=\"edge2\" class=\"edge\">\n<title>1&#45;&gt;2</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M192.7118,-192.9465C178.9479,-183.3395 163.8494,-172.8009 149.6426,-162.8848\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"151.6011,-159.9835 141.3978,-157.13 147.5946,-165.7236 151.6011,-159.9835\"/>\n</g>\n<!-- 5 -->\n<g id=\"node6\" class=\"node\">\n<title>5</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"295.5,-149.5 203.5,-149.5 203.5,-96.5 295.5,-96.5 295.5,-149.5\"/>\n<text text-anchor=\"middle\" x=\"249.5\" y=\"-134.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n<text text-anchor=\"middle\" x=\"249.5\" y=\"-119.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1</text>\n<text text-anchor=\"middle\" x=\"249.5\" y=\"-104.3\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [0, 1]</text>\n</g>\n<!-- 1&#45;&gt;5 -->\n<g id=\"edge5\" class=\"edge\">\n<title>1&#45;&gt;5</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M244.1195,-192.9465C244.9414,-182.2621 245.8519,-170.4254 246.6866,-159.5742\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"250.1809,-159.7812 247.4583,-149.5422 243.2015,-159.2442 250.1809,-159.7812\"/>\n</g>\n<!-- 3 -->\n<g id=\"node4\" class=\"node\">\n<title>3</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"114.5,-53 16.5,-53 16.5,0 114.5,0 114.5,-53\"/>\n<text text-anchor=\"middle\" x=\"65.5\" y=\"-37.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.116</text>\n<text text-anchor=\"middle\" x=\"65.5\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 65</text>\n<text text-anchor=\"middle\" x=\"65.5\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [61, 4]</text>\n</g>\n<!-- 2&#45;&gt;3 -->\n<g id=\"edge3\" class=\"edge\">\n<title>2&#45;&gt;3</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M82.9808,-88.9777C80.6237,-80.5533 78.0888,-71.4934 75.7007,-62.9579\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"79.0491,-61.9356 72.984,-53.2485 72.308,-63.8217 79.0491,-61.9356\"/>\n</g>\n<!-- 4 -->\n<g id=\"node5\" class=\"node\">\n<title>4</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"224.5,-53 132.5,-53 132.5,0 224.5,0 224.5,-53\"/>\n<text text-anchor=\"middle\" x=\"178.5\" y=\"-37.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n<text text-anchor=\"middle\" x=\"178.5\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1</text>\n<text text-anchor=\"middle\" x=\"178.5\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [0, 1]</text>\n</g>\n<!-- 2&#45;&gt;4 -->\n<g id=\"edge4\" class=\"edge\">\n<title>2&#45;&gt;4</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M122.8204,-88.9777C130.981,-79.8207 139.8108,-69.9129 147.9821,-60.744\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"150.6217,-63.0427 154.662,-53.2485 145.3958,-58.3854 150.6217,-63.0427\"/>\n</g>\n<!-- 7 -->\n<g id=\"node8\" class=\"node\">\n<title>7</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"503,-157 330,-157 330,-89 503,-89 503,-157\"/>\n<text text-anchor=\"middle\" x=\"416.5\" y=\"-141.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">ApplicantIncome &lt;= 3357.5</text>\n<text text-anchor=\"middle\" x=\"416.5\" y=\"-126.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.337</text>\n<text text-anchor=\"middle\" x=\"416.5\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 420</text>\n<text text-anchor=\"middle\" x=\"416.5\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [90, 330]</text>\n</g>\n<!-- 6&#45;&gt;7 -->\n<g id=\"edge7\" class=\"edge\">\n<title>6&#45;&gt;7</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M421.8805,-192.9465C421.252,-184.776 420.5717,-175.9318 419.9131,-167.3697\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"423.3821,-166.8321 419.1254,-157.13 416.4027,-167.369 423.3821,-166.8321\"/>\n</g>\n<!-- 10 -->\n<g id=\"node11\" class=\"node\">\n<title>10</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"665.5,-157 521.5,-157 521.5,-89 665.5,-89 665.5,-157\"/>\n<text text-anchor=\"middle\" x=\"593.5\" y=\"-141.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">LoanAmount &lt;= 171.0</text>\n<text text-anchor=\"middle\" x=\"593.5\" y=\"-126.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.375</text>\n<text text-anchor=\"middle\" x=\"593.5\" y=\"-111.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 4</text>\n<text text-anchor=\"middle\" x=\"593.5\" y=\"-96.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [3, 1]</text>\n</g>\n<!-- 6&#45;&gt;10 -->\n<g id=\"edge10\" class=\"edge\">\n<title>6&#45;&gt;10</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M479.8369,-192.9465C495.5942,-183.2497 512.8937,-172.6039 529.1387,-162.6069\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"531.3565,-165.3518 538.0388,-157.13 527.6878,-159.3902 531.3565,-165.3518\"/>\n</g>\n<!-- 8 -->\n<g id=\"node9\" class=\"node\">\n<title>8</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"368.5,-53 256.5,-53 256.5,0 368.5,0 368.5,-53\"/>\n<text text-anchor=\"middle\" x=\"312.5\" y=\"-37.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.253</text>\n<text text-anchor=\"middle\" x=\"312.5\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 155</text>\n<text text-anchor=\"middle\" x=\"312.5\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [23, 132]</text>\n</g>\n<!-- 7&#45;&gt;8 -->\n<g id=\"edge8\" class=\"edge\">\n<title>7&#45;&gt;8</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M379.8335,-88.9777C369.7675,-79.6376 358.8596,-69.5163 348.8135,-60.1947\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"351.0385,-57.4847 341.3274,-53.2485 346.2772,-62.616 351.0385,-57.4847\"/>\n</g>\n<!-- 9 -->\n<g id=\"node10\" class=\"node\">\n<title>9</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"498.5,-53 386.5,-53 386.5,0 498.5,0 498.5,-53\"/>\n<text text-anchor=\"middle\" x=\"442.5\" y=\"-37.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.378</text>\n<text text-anchor=\"middle\" x=\"442.5\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 265</text>\n<text text-anchor=\"middle\" x=\"442.5\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [67, 198]</text>\n</g>\n<!-- 7&#45;&gt;9 -->\n<g id=\"edge9\" class=\"edge\">\n<title>7&#45;&gt;9</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M425.6666,-88.9777C427.9364,-80.5533 430.3774,-71.4934 432.6771,-62.9579\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"436.071,-63.8147 435.2932,-53.2485 429.3121,-61.9936 436.071,-63.8147\"/>\n</g>\n<!-- 11 -->\n<g id=\"node12\" class=\"node\">\n<title>11</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"615.5,-53 523.5,-53 523.5,0 615.5,0 615.5,-53\"/>\n<text text-anchor=\"middle\" x=\"569.5\" y=\"-37.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n<text text-anchor=\"middle\" x=\"569.5\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1</text>\n<text text-anchor=\"middle\" x=\"569.5\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [0, 1]</text>\n</g>\n<!-- 10&#45;&gt;11 -->\n<g id=\"edge11\" class=\"edge\">\n<title>10&#45;&gt;11</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M585.0385,-88.9777C582.9433,-80.5533 580.6901,-71.4934 578.5673,-62.9579\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"581.9626,-62.1081 576.1525,-53.2485 575.1695,-63.7976 581.9626,-62.1081\"/>\n</g>\n<!-- 12 -->\n<g id=\"node13\" class=\"node\">\n<title>12</title>\n<polygon fill=\"none\" stroke=\"#000000\" points=\"725.5,-53 633.5,-53 633.5,0 725.5,0 725.5,-53\"/>\n<text text-anchor=\"middle\" x=\"679.5\" y=\"-37.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n<text text-anchor=\"middle\" x=\"679.5\" y=\"-22.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">samples = 3</text>\n<text text-anchor=\"middle\" x=\"679.5\" y=\"-7.8\" font-family=\"Times,serif\" font-size=\"14.00\" fill=\"#000000\">value = [3, 0]</text>\n</g>\n<!-- 10&#45;&gt;12 -->\n<g id=\"edge12\" class=\"edge\">\n<title>10&#45;&gt;12</title>\n<path fill=\"none\" stroke=\"#000000\" d=\"M623.8204,-88.9777C631.981,-79.8207 640.8108,-69.9129 648.9821,-60.744\"/>\n<polygon fill=\"#000000\" stroke=\"#000000\" points=\"651.6217,-63.0427 655.662,-53.2485 646.3958,-58.3854 651.6217,-63.0427\"/>\n</g>\n</g>\n</svg>\n",
      "text/plain": [
       "<graphviz.files.Source at 0x7f0fecd24410>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import graphviz \n",
    "from sklearn import tree\n",
    "\n",
    "tree_clf = tree.DecisionTreeClassifier(max_depth = 3)\n",
    "tree_clf.fit(X_train,y_train)\n",
    "dot_data = tree.export_graphviz(tree_clf,feature_names = X.columns.tolist())\n",
    "graph = graphviz.Source(dot_data)\n",
    "graph"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From above tree, we could see that some of the leafs have less than 5 samples hence our classifier might overfit. We can sweep hyper-parameter 'min_samples_leaf' to further improve test accuracy by keeping max_depth to 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0fee058b10>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "training_accuracy = []\n",
    "val_accuracy = []\n",
    "training_f1 = []\n",
    "val_f1 = []\n",
    "min_samples_leaf = []\n",
    "\n",
    "import numpy as np # import numpy for numeric calculation\n",
    "for samples_leaf in range(1,80,3): #Sweeping from 1% samples to 10% samples per leaf \n",
    "    tree_clf = DecisionTreeClassifier(max_depth=3,min_samples_leaf = samples_leaf)\n",
    "    tree_clf.fit(X_train,y_train)\n",
    "    y_training_pred = tree_clf.predict(X_train)\n",
    "\n",
    "    training_acc = accuracy_score(y_train,y_training_pred)\n",
    "    train_f1 = f1_score(y_train,y_training_pred)\n",
    "    val_mean_f1 = cross_val_score(tree_clf,X_train,y_train,cv=5,scoring='f1_macro').mean()\n",
    "    val_mean_accuracy = cross_val_score(tree_clf,X_train,y_train,cv=5,scoring='accuracy').mean()\n",
    "    \n",
    "    training_accuracy.append(training_acc)\n",
    "    val_accuracy.append(val_mean_accuracy)\n",
    "    training_f1.append(train_f1)\n",
    "    val_f1.append(val_mean_f1)\n",
    "    min_samples_leaf.append(samples_leaf)\n",
    "    \n",
    "\n",
    "Tuning_min_samples_leaf = {\"Training Accuracy\": training_accuracy, \"Validation Accuracy\": val_accuracy, \"Training F1\": training_f1, \"Validation F1\":val_f1, \"Min_Samples_leaf\": min_samples_leaf }\n",
    "Tuning_min_samples_leaf_df = pd.DataFrame.from_dict(Tuning_min_samples_leaf)\n",
    "\n",
    "plot_df = Tuning_min_samples_leaf_df.melt('Min_Samples_leaf',var_name='Metrics',value_name=\"Values\")\n",
    "fig,ax = plt.subplots(figsize=(15,5))\n",
    "sns.pointplot(x=\"Min_Samples_leaf\", y=\"Values\",hue=\"Metrics\", data=plot_df,ax=ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From above plot, we will choose Min_Samples_leaf to 35 to improve test accuracy.\n",
    "\n",
    "Let's use this Decision Tree classifier on unseen test data and evaluate Test Accuracy, F1 Score and Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy:  0.8536585365853658\n",
      "Test F1 Score:  0.903225806451613\n",
      "Confusion Matrix on Test Data\n"
     ]
    },
    {
     "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>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>84</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>22</td>\n",
       "      <td>101</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0    1  All\n",
       "True                   \n",
       "0          21   17   38\n",
       "1           1   84   85\n",
       "All        22  101  123"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "tree_clf = DecisionTreeClassifier(max_depth=3,min_samples_leaf = 35)\n",
    "tree_clf.fit(X_train,y_train)\n",
    "y_pred = tree_clf.predict(X_test_imp)\n",
    "print(\"Test Accuracy: \",accuracy_score(y_test,y_pred))\n",
    "print(\"Test F1 Score: \",f1_score(y_test,y_pred))\n",
    "print(\"Confusion Matrix on Test Data\")\n",
    "pd.crosstab(y_test, y_pred, rownames=['True'], colnames=['Predicted'], margins=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be seen that majority of the misclassifications are happening because of Loan Reject applicants being classified as Accept."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train F1 Score:-  0.8699080157687253\n",
      "Train Accuracy :-  0.7983706720977597\n",
      "Validation Mean F1 Score:-  0.7104522093151318\n",
      "Validation Mean F1 Score:-  0.7922194807319991\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf_clf = RandomForestClassifier(n_estimators = 100,max_depth = 3 ,min_samples_leaf=10)\n",
    "rf_clf.fit(X_train,y_train)\n",
    "y_pred = rf_clf.predict(X_train)\n",
    "\n",
    "print(\"Train F1 Score:- \",f1_score(y_train,y_pred))\n",
    "print(\"Train Accuracy :- \",accuracy_score(y_train,y_pred))\n",
    "print(\"Validation Mean F1 Score:- \",cross_val_score(rf_clf,X_train,y_train,cv=5,scoring='f1_macro').mean())\n",
    "print(\"Validation Mean F1 Score:- \",cross_val_score(rf_clf,X_train,y_train,cv=5,scoring='accuracy').mean())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest: Test Data Evaluation\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score accuracy :-  0.8536585365853658\n",
      "Test F1 score:-  0.903225806451613\n",
      "Confusion metrix on test data:-\n"
     ]
    },
    {
     "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>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>84</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>22</td>\n",
       "      <td>101</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0    1  All\n",
       "True                   \n",
       "0          21   17   38\n",
       "1           1   84   85\n",
       "All        22  101  123"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = rf_clf.predict(X_test_imp)\n",
    "print(\"Test score accuracy :- \",accuracy_score(y_test,y_pred))\n",
    "print(\"Test F1 score:- \",f1_score(y_test,y_pred))\n",
    "print(\"Confusion metrix on test data:-\")\n",
    "pd.crosstab(y_test,y_pred,rownames=[\"True\"],colnames=[\"Predicted\"],margins= True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be clearly seen that same result found in this model so we will  try other model which is called Logistic regression and try to get better accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression Model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0fec7549d0>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression \n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "train_accuracies = []\n",
    "train_f1_scores =[]\n",
    "test_accuracies = []\n",
    "test_f1_scores =[]\n",
    "thresholds = []\n",
    "\n",
    "#for thresh in np.linspace(0.1,0.9,8): ## Sweeping from threshold of 0.1 to 0.9\n",
    "for thresh in np.arange(0.1,0.9,0.1):## Sweeping from threshold of 0.1 to 0.9\n",
    "    \n",
    "    lr_clf = LogisticRegression(solver='liblinear')\n",
    "    lr_clf.fit(X_train,y_train)\n",
    "    \n",
    "    y_pred_train_thresh = lr_clf.predict_proba(X_train)[:,1]\n",
    "    y_pred_train = (y_pred_train_thresh > thresh).astype(int)\n",
    "    \n",
    "    train_acc = accuracy_score(y_train,y_pred_train)\n",
    "    train_f1 = f1_score(y_train,y_pred_train)\n",
    "    \n",
    "    y_pred_test_thresh = lr_clf.predict_proba(X_test_imp)[:,1]\n",
    "    y_pred_test = (y_pred_test_thresh > thresh ).astype(int)\n",
    "    \n",
    "    test_acc = accuracy_score(y_test,y_pred_test)\n",
    "    test_f1 = f1_score(y_test,y_pred_test)\n",
    "    \n",
    "    train_accuracies.append(train_acc)\n",
    "    train_f1_scores.append(train_f1)\n",
    "    test_accuracies.append(test_acc)\n",
    "    test_f1_scores.append(test_f1)\n",
    "    thresholds.append(thresh)\n",
    "    \n",
    "    \n",
    "Threshold_logreg = {\"Training Accuracy\": train_accuracies, \"Test Accuracy\": test_accuracies, \"Training F1\": train_f1_scores, \"Test F1\":test_f1_scores, \"Decision Threshold\": thresholds }\n",
    "Threshold_logreg_df = pd.DataFrame.from_dict(Threshold_logreg)\n",
    "\n",
    "plot_df = Threshold_logreg_df.melt('Decision Threshold',var_name='Metrics',value_name=\"Values\")\n",
    "fig,ax = plt.subplots(figsize=(15,5))\n",
    "sns.pointplot(x=\"Decision Threshold\", y=\"Values\",hue=\"Metrics\", data=plot_df,ax=ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic Regression does slightly better than Decision Tree and Random Forest.\n",
    "Based on the above Test/Train curves, we can keep threshold to 0.4.\n",
    "Now Finally let's look at Logistic Regression Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy:  0.8617886178861789\n",
      "Test F1 Score:  0.9081081081081082\n",
      "Confusion Matrix on Test Data\n"
     ]
    },
    {
     "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>Predicted</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22</td>\n",
       "      <td>16</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>84</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>23</td>\n",
       "      <td>100</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Predicted   0    1  All\n",
       "True                   \n",
       "0          22   16   38\n",
       "1           1   84   85\n",
       "All        23  100  123"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thresh = 0.4 ### Threshold chosen from above Curves\n",
    "\n",
    "y_pred_test_thresh = lr_clf.predict_proba(X_test_imp)[:,1]\n",
    "y_pred = (y_pred_test_thresh > thresh).astype(int) \n",
    "\n",
    "print(\"Test Accuracy: \",accuracy_score(y_test,y_pred))\n",
    "print(\"Test F1 Score: \",f1_score(y_test,y_pred))\n",
    "print(\"Confusion Matrix on Test Data\")\n",
    "\n",
    "pd.crosstab(y_test, y_pred, rownames=['True'], colnames=['Predicted'], margins=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic Regression Confusion matrix is very similar to Decision Tree and Random Forest Classifier. In this analysis, we did extensive analysis of input data and were able to achieve Test Accuracy of 86 %"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}