credit_default.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "06ffbf7d-879f-450c-9e7d-5a0fa699f0c1",
   "metadata": {},
   "source": [
    "# Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "70d180ee-f194-442d-ae5d-5e3eff853aa4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "import math\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import random\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics import confusion_matrix, f1_score, accuracy_score, precision_score, recall_score\n",
    "from sklearn import tree\n",
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19c73813-ae98-4bd0-b795-07fe57274d4e",
   "metadata": {},
   "source": [
    "# Auxillary Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "38e8ee61-231c-476c-8a87-fd8a4370886b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def path_loader(path, view = False):\n",
    "    '''\n",
    "        This function takes the path to dataset and returns\n",
    "        the path for training and test data, and feature\n",
    "        descriptions.\n",
    "        \n",
    "        view = False : do not view the paths\n",
    "        view = True : view the paths\n",
    "        \n",
    "    '''\n",
    "    for directory_path, _ , directory_files in os.walk(path):\n",
    "        path_all = []\n",
    "        for files in directory_files:\n",
    "            path_all.append(os.path.join(directory_path, files))\n",
    "        for i, path in enumerate(path_all):\n",
    "            if 'test' in path:\n",
    "                path_testing = path\n",
    "            elif 'training' in path:\n",
    "                path_training = path\n",
    "            elif 'Dictionary' in path:\n",
    "                path_help = path\n",
    "    \n",
    "    if view == True:\n",
    "        print('Training Data Path: {}\\nTest Data Path: {}\\nData Dictionary Path: {}'.format(path_training, path_testing, path_help))\n",
    "        \n",
    "    return path_training, path_testing, path_help"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9aecf9a9-1d4b-4428-b891-9e67bf00e8d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def col_name_changer(dataframe, name_dictionary):\n",
    "    '''\n",
    "        This function takes a name_dictionary and a\n",
    "        dataframe, it changes the names of the columns\n",
    "        according to the key-values given in the dictionary.\n",
    "        \n",
    "    '''\n",
    "    dataframe = dataframe.rename(columns = name_dictionary)\n",
    "    \n",
    "    return dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ecd07c15-a4c8-465f-9464-0dd05ed693ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "def count_visualizer(dataset, col_name):\n",
    "    '''\n",
    "        This is used to visualize the count of unique\n",
    "        values in a given pandas column\n",
    "        \n",
    "    '''\n",
    "    data = dataset\n",
    "    fig, ax = plt.subplots()\n",
    "    ax.hist(data[col_name])\n",
    "    ax.set_title('Frequency of Unique Values')\n",
    "    ax.set_xlabel('Unique Value Label')\n",
    "    ax.set_ylabel('Count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d5ca9add-023c-456d-b7d1-04e827ee15dc",
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_evaluation(y_pred, y_test):\n",
    "    '''\n",
    "        This function is used to evaluate the models based on\n",
    "        y_test and the predicted y_pred.\n",
    "    \n",
    "    '''\n",
    "    cm = confusion_matrix(y_pred, y_test)\n",
    "    \n",
    "    print('MAE: ', metrics.mean_absolute_error(y_test, y_pred))\n",
    "    print('MSE: ', metrics.mean_squared_error(y_test, y_pred))\n",
    "    print('RMSE: ', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))\n",
    "    print(\"Precision Score: \",precision_score(y_test, y_pred, average = 'weighted'))\n",
    "    print(\"Recall Score: \",recall_score(y_test, y_pred, average = 'weighted'))\n",
    "    print(\"f1_score: \",f1_score(y_test, y_pred, average = 'weighted'))\n",
    "    print('\\n\\033[1m' +'Confusion Matrix : '+ '\\033[0m')\n",
    "    sns.heatmap(cm, cmap = 'OrRd', annot = True, fmt='d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4a2fc883-b6f3-409b-8a22-f20f5183842b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def detect_outliers(dataset, features, sensitivity_factor):\n",
    "    '''\n",
    "        This function takes a given dataset and features for\n",
    "        outlier detection. The average deviation is first\n",
    "        calculated which is then multiplied by the factor\n",
    "        to give a suitable range for outlier detecting.\n",
    "        \n",
    "        sensitivity_factor > 3 would result in less number of\n",
    "        outliers.\n",
    "        \n",
    "        sensitivity_factor suggested range = [> 1 and < 3]\n",
    "        \n",
    "    '''\n",
    "    outlier_index = []\n",
    "    outlier_list = list(np.zeros(len(features), dtype = int))\n",
    "    \n",
    "    for i, col in enumerate(features):\n",
    "        Q1 = dataset.describe()[col][4]\n",
    "        Q2 = dataset.describe()[col][5]\n",
    "        Q3 = dataset.describe()[col][6]\n",
    "        \n",
    "        avg_deviation = ((Q3 - Q2) + (Q2 - Q1))/2\n",
    "        \n",
    "        outlier_range = avg_deviation * sensitivity_factor\n",
    "        outlier_list[i] = dataset[(dataset[col] > Q2 + outlier_range) | (dataset[col] < Q2 - outlier_range)].index\n",
    "    \n",
    "    outlier_index.extend(np.intersect1d(outlier_list[0], outlier_list[1]))\n",
    "    print('Number of data: {}\\nNumber of outliers found: {}\\nNumber of remaining data: {}'.format(len(dataset), \n",
    "                                                                                                  len(outlier_index), \n",
    "                                                                                                  len(dataset) - len(outlier_index)))\n",
    "        \n",
    "    return outlier_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "af617d96-f91c-4531-9ebd-f7aa192522cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_pipeline(dataset, test_size, knn_args, dt_args, view_data_shape = False):\n",
    "    '''\n",
    "        This function takes in a dataset and divides it according to a \n",
    "        given test_size. These are fit into the ML models.\n",
    "        \n",
    "        set view_data_shape = True to see the dataset shape after division.\n",
    "        \n",
    "        Models : Decision Tree and K-Nearest-Neighbour Classifier.\n",
    "        \n",
    "        knn_args:\n",
    "            \"algorithm\", \"leaf_size\", \"metric\", \"n_jobs\", \"n_neighbors\", \"p\", \"weights\"\n",
    "            \n",
    "        dt_args:\n",
    "            \"criterion\", \"min_samples_leaf\", \"min_samples_split\", \"random_state\", \"splitter\"\n",
    "    \n",
    "    '''\n",
    "    \n",
    "    x_train, x_test, y_train, y_test = train_test_split(dataset[features], dataset[target], test_size = test_size, \n",
    "                                                        random_state = 39, shuffle = True)\n",
    "    \n",
    "    if view_data_shape == True:\n",
    "        print(\"Shape of x_train: {}\\nShape of y_train: {}\\nShape of x_test: {}\\nShape of y_test: {}\\n\".format(x_train.shape,\n",
    "                                                                                                              y_train.shape,\n",
    "                                                                                                              x_test.shape,\n",
    "                                                                                                              y_test.shape))\n",
    "        \n",
    "    # Call the KNN model and evaluate the accuracy on test data\n",
    "    KNN_model = KNeighborsClassifier(**knn_args)\n",
    "\n",
    "    KNN_model.fit(x_train, y_train)\n",
    "    y_pred_KNN = KNN_model.predict(x_test)\n",
    "    accuracy = accuracy_score(y_test, y_pred_KNN)\n",
    "    print(f\"Accuracy for KNN Model is: {accuracy * 100}\")\n",
    "        \n",
    "    # Call the Decision Tree model and evaluate the accuracy on test data\n",
    "    DT_model = tree.DecisionTreeClassifier(**dt_args)\n",
    "\n",
    "    DT_model.fit(x_train, y_train)\n",
    "    y_pred_DT = DT_model.predict(x_test)\n",
    "    accuracy = accuracy_score(y_test, y_pred_DT)\n",
    "    print(f\"Accuracy for Decision Tree is: {accuracy * 100}\")\n",
    "        \n",
    "    return y_pred_KNN, y_pred_DT, y_test    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5169f70-ade0-4f17-bc6e-9e383ec3d987",
   "metadata": {},
   "source": [
    "# Dataset Procurement and Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ca98b7b-9714-4bf1-9505-0d75ca919372",
   "metadata": {},
   "source": [
    "### Procurement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "463e2fb7-546f-4521-b057-dfbbcb7ee85d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Data Path: Dataset\\.ipynb_checkpoints\\cs-training-checkpoint.csv\n",
      "Test Data Path: Dataset\\cs-test.csv\n",
      "Data Dictionary Path: Dataset\\.ipynb_checkpoints\\Data-Dictionary-checkpoint.csv\n"
     ]
    }
   ],
   "source": [
    "path = \"Dataset\"\n",
    "\n",
    "path_training, path_testing, path_help = path_loader(path, view = True)\n",
    "\n",
    "dataset_description = pd.read_csv(path_help, header = 1)\n",
    "train = pd.read_csv(path_training)\n",
    "test = pd.read_csv(path_testing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "346465d4-b8c5-4588-850d-2de1518ca457",
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "    The columns chosen for dropping were chosen based on\n",
    "    specific application based testing and information requirement\n",
    "    for decent accuracy\n",
    "\n",
    "''' \n",
    "\n",
    "columns_to_drop = [0, 5, 6, 9, 11]\n",
    "train = train.drop(train.columns[columns_to_drop], axis = 1)\n",
    "test = test.drop(test.columns[columns_to_drop], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "dea2f447-f008-4f1a-98d9-8d463b3f590d",
   "metadata": {},
   "outputs": [],
   "source": [
    "col_dictionary = {'SeriousDlqin2yrs': 'target',\n",
    "                  'RevolvingUtilizationOfUnsecuredLines': 'balance-limit-ratio',\n",
    "                  'NumberOfTime30-59DaysPastDueNotWorse': 'late30-59',\n",
    "                  'NumberOfOpenCreditLinesAndLoans': 'purchase_count',\n",
    "                  'NumberOfTimes90DaysLate': 'late>90',\n",
    "                  'NumberOfTime60-89DaysPastDueNotWorse': 'late60-89'}\n",
    "\n",
    "train = col_name_changer(train, col_dictionary)\n",
    "test = col_name_changer(test, col_dictionary)\n",
    "\n",
    "columns = train.columns\n",
    "features = columns[1:]\n",
    "target = columns[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa27dc15-3eec-492d-8ea4-ac63a71d9d01",
   "metadata": {},
   "source": [
    "### Data Analysis and Processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "88c095fa-0aea-4bd8-8bf1-ea46a9000d47",
   "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>target</th>\n",
       "      <th>balance-limit-ratio</th>\n",
       "      <th>age</th>\n",
       "      <th>late30-59</th>\n",
       "      <th>purchase_count</th>\n",
       "      <th>late&gt;90</th>\n",
       "      <th>late60-89</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.066840</td>\n",
       "      <td>6.048438</td>\n",
       "      <td>52.295207</td>\n",
       "      <td>0.421033</td>\n",
       "      <td>8.452760</td>\n",
       "      <td>0.265973</td>\n",
       "      <td>0.240387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.249746</td>\n",
       "      <td>249.755371</td>\n",
       "      <td>14.771866</td>\n",
       "      <td>4.192781</td>\n",
       "      <td>5.145951</td>\n",
       "      <td>4.169304</td>\n",
       "      <td>4.155179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.029867</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.154181</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.559046</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>50708.000000</td>\n",
       "      <td>109.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>98.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              target  balance-limit-ratio            age      late30-59   \n",
       "count  150000.000000        150000.000000  150000.000000  150000.000000  \\\n",
       "mean        0.066840             6.048438      52.295207       0.421033   \n",
       "std         0.249746           249.755371      14.771866       4.192781   \n",
       "min         0.000000             0.000000       0.000000       0.000000   \n",
       "25%         0.000000             0.029867      41.000000       0.000000   \n",
       "50%         0.000000             0.154181      52.000000       0.000000   \n",
       "75%         0.000000             0.559046      63.000000       0.000000   \n",
       "max         1.000000         50708.000000     109.000000      98.000000   \n",
       "\n",
       "       purchase_count        late>90      late60-89  \n",
       "count   150000.000000  150000.000000  150000.000000  \n",
       "mean         8.452760       0.265973       0.240387  \n",
       "std          5.145951       4.169304       4.155179  \n",
       "min          0.000000       0.000000       0.000000  \n",
       "25%          5.000000       0.000000       0.000000  \n",
       "50%          8.000000       0.000000       0.000000  \n",
       "75%         11.000000       0.000000       0.000000  \n",
       "max         58.000000      98.000000      98.000000  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "ea797aa0-f459-4398-bab7-f321a4387fd1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of features: 6\n",
      "Number of features without missing/zero data in train dataset: 6\n",
      "Number of features without missing/zero data in test dataset: 6\n"
     ]
    }
   ],
   "source": [
    "#Check for missing/zero data:\n",
    "\n",
    "missing_count_train = (train[features].isnull().sum())/(len(train[features]))\n",
    "missing_count_test = (test[features].isnull().sum())/(len(test[features]))\n",
    "\n",
    "print(f\"Number of features: {len(features)}\")\n",
    "print(f\"Number of features without missing/zero data in train dataset: {missing_count_train.isnull().count()}\")\n",
    "print(f\"Number of features without missing/zero data in test dataset: {missing_count_test.isnull().count()}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "a59234f3-a2ae-44bc-9dcb-1cb772030cf0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the distribution of the unique target values\n",
    "\n",
    "count_visualizer(train, target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "6285c63e-e999-446d-aa17-1d6134455df8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# There are too few target = 1 cases, so we can duplicate a few rows\n",
    "\n",
    "multiplier_factor = 3\n",
    "\n",
    "for i in range(multiplier_factor):\n",
    "    indexes = train[(train[target] == 1)].index\n",
    "    indexes = random.sample(list(indexes), len(train[(train[target] == 1)].index))\n",
    "    indexes = pd.Index(indexes, dtype = 'Int64')\n",
    "    rows_to_duplicate = train.loc[indexes]\n",
    "    train = pd.concat([train, rows_to_duplicate], ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a3ecdb3a-bd96-416e-8e6e-2ee3b6826818",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of data: 150000\n",
      "Number of outliers found: 2784\n",
      "Number of remaining data: 147216\n"
     ]
    }
   ],
   "source": [
    "# Drop rows with outlier or artifact values\n",
    "\n",
    "outlier_features = [features[0], features[3]]\n",
    "outliers_to_drop = detect_outliers(train, outlier_features, sensitivity_factor = 2.5)\n",
    "\n",
    "train = train.drop(outliers_to_drop, axis = 0).reset_index(drop = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d05db826-dc54-4162-8b98-51d65915e852",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "    Check Correlation between the features using heatmap\n",
    "    High correlation has been seen between target and the late submissions\n",
    "\n",
    "''' \n",
    "\n",
    "data_correlation = train.corr()\n",
    "sns.heatmap(data_correlation, cmap = 'YlGnBu', fmt = \".2f\", annot = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb278800-6ffd-44f3-922c-033dbf1ca30f",
   "metadata": {},
   "source": [
    "# Credit Default Prediction Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "3fe6005f-236c-4eac-9513-f5958f28507b",
   "metadata": {},
   "outputs": [],
   "source": [
    "params_knn = {\n",
    "    \"algorithm\" : 'kd_tree',\n",
    "    \"leaf_size\" : 150,\n",
    "    \"metric\" : 'euclidean',\n",
    "    \"n_jobs\" : -1,\n",
    "    \"n_neighbors\" : 3,\n",
    "    \"p\" : 2,\n",
    "    \"weights\" : 'distance'\n",
    "}\n",
    "\n",
    "params_dt = {\n",
    "    \"criterion\" : 'entropy',\n",
    "    \"min_samples_leaf\" : 3,\n",
    "    \"min_samples_split\" : 4,\n",
    "    \"random_state\" : 32,\n",
    "    \"splitter\" : 'best'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "fa9595f9-8b88-4d28-b2ec-3673329914cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of x_train: (176145, 6)\n",
      "Shape of y_train: (176145,)\n",
      "Shape of x_test: (44037, 6)\n",
      "Shape of y_test: (44037,)\n",
      "\n",
      "Accuracy for KNN Model is: 92.9695483343552\n",
      "Accuracy for Decision Tree is: 93.09444330903558\n"
     ]
    }
   ],
   "source": [
    "pred_k, pred_d, y_test = model_pipeline(train, 0.20, params_knn, params_dt, view_data_shape = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "7d71f0e6-70df-408d-9cbf-5e2d309519db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE:  0.07030451665644799\n",
      "MSE:  0.07030451665644799\n",
      "RMSE:  0.2651499889806673\n",
      "Precision Score:  0.9364000634722753\n",
      "Recall Score:  0.929695483343552\n",
      "f1_score:  0.9305497097911181\n",
      "\n",
      "\u001b[1mConfusion Matrix : \u001b[0m\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# KNN model evaluation:\n",
    "\n",
    "model_evaluation(pred_k, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "f4559f4b-8a2c-49a5-86a3-bd229056ddb0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE:  0.06905556690964416\n",
      "MSE:  0.06905556690964416\n",
      "RMSE:  0.2627842592501388\n",
      "Precision Score:  0.9372139477638242\n",
      "Recall Score:  0.9309444330903558\n",
      "f1_score:  0.9317595427008306\n",
      "\n",
      "\u001b[1mConfusion Matrix : \u001b[0m\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# DT model evaluation:\n",
    "\n",
    "model_evaluation(pred_d, y_test)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "credit",
   "language": "python",
   "name": "credit"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}