LogisticRegression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "import random\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from sklearn import metrics\n",
    "from sklearn.linear_model import LogisticRegression, SGDClassifier\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis\n",
    "Below we show histograms of each feature. As you can see, the data is highly imbalanced for stroke = 0 (no stroke) and stroke = 1 (yes stroke). \\\n",
    "We also show a histogram, which we added to further show the imbalanced data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = pd.read_csv('./stroke_data.csv')\n",
    "data.hist(figsize=(20, 20))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# heatmap - visualize correlation\n",
    "plt.figure(figsize=(15, 10))\n",
    "heatMap = data.corr()\n",
    "sns.heatmap(heatMap, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Engineering \n",
    "Clean the data to be a working dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#find and fill null values with averages (BMI has 201 null values)\n",
    "avg = data['bmi'].mean()\n",
    "data.bmi = (data.bmi.fillna(avg))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Give numerical values to categorical variables\n",
    "\n",
    "<table>\n",
    "    <thead>\n",
    "        <tr>\n",
    "            <th colspan=2>work_type</th>\n",
    "            <th colspan=2>smoking_status</th>\n",
    "            <th colspan=2>gender</th>\n",
    "            <th colspan=2>Residence_type</th>\n",
    "            <th colspan=2>ever_married</th>\n",
    "        </tr>\n",
    "    </thead>\n",
    "    <tbody>\n",
    "        <tr>\n",
    "            <td>Private</td>\n",
    "            <td>0</td>\n",
    "            <td>formerly smoked</td>\n",
    "            <td>0</td>\n",
    "            <td>Male</td>\n",
    "            <td>0</td>\n",
    "            <td>Urban</td>\n",
    "            <td>0</td>\n",
    "            <td>Yes</td>\n",
    "            <td>0</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <td>Self-employed</td>\n",
    "            <td>1</td>\n",
    "            <td>never smoked</td>\n",
    "            <td>1</td>\n",
    "            <td>Female</td>\n",
    "            <td>1</td>\n",
    "            <td>Rural</td>\n",
    "            <td>1</td>\n",
    "            <td>No</td>\n",
    "            <td>1</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <td>Govt_job</td>\n",
    "            <td>2</td>\n",
    "            <td>smokes</td>\n",
    "            <td>2</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <td>children</td>\n",
    "            <td>3</td>\n",
    "            <td>Unknown</td>\n",
    "            <td>3</td>\n",
    "        </tr>\n",
    "        <tr>\n",
    "            <td>Never_worked</td>\n",
    "            <td>4</td>\n",
    "        </tr>\n",
    "    </tbody>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['work_type'] = data['work_type'].map(\n",
    "    {'Private': 0, 'Self-employed': 1, 'Govt_job': 2, 'children': 3, 'Never_worked': 4})\n",
    "data['gender'] = data['gender'].map({'Male': 0, 'Female': 1})\n",
    "data['Residence_type'] = data['Residence_type'].map({'Urban': 0, 'Rural': 1})\n",
    "data['smoking_status'] = data['smoking_status'].map(\n",
    "    {'formerly smoked': 0, 'never smoked': 1, 'smokes': 2, 'Unknown': 3})\n",
    "data['ever_married'] = data['ever_married'].map({'Yes': 0, 'No': 1})\n",
    "\n",
    "#divide dataset into features and labels\n",
    "#drop ID because it's not necessary for analysis\n",
    "X = data.iloc[:, 1:-1]\n",
    "y = data[['stroke']]\n",
    "\n",
    "#replace null values again (1 in gender)\n",
    "X.gender = (X.gender.fillna(1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Heatmap with new Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# heatmap - visualize correlation\n",
    "plt.figure(figsize=(15, 10))\n",
    "heatMap = data.corr()\n",
    "sns.heatmap(heatMap, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Handle Imbalanced Data\n",
    "Now, the distribution of stroke and no stroke is equal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "smote = SMOTE()\n",
    "X, y = smote.fit_resample(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y.hist(figsize=(5, 5))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Standardize data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def standardize(X):\n",
    "    #Standardizes the data using the mean and standard deviation\n",
    "    mean = np.mean(X, axis=0)\n",
    "    std = np.std(X, axis=0, ddof=1)\n",
    "    s_X = (X - mean) / std\n",
    "\n",
    "    #add bias feature\n",
    "    bias = (np.ones((s_X.shape[0], 1)))\n",
    "    s_X = np.append(s_X, bias, axis=1)\n",
    "    return s_X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "s_X = standardize(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Functions used:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sigmoid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sigmoid(x, thetas):\n",
    "    return 1 / (1 + math.e ** -(x @ thetas))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Gradient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def gradient(X_train, y_train, thetas):\n",
    "    sig = sigmoid(X_train, thetas)\n",
    "    gradient = (X_train.T @ (sig - y_train))\n",
    "    return gradient"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Batch Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "def logisticBGD(X_train, y_train, X_test, y_test, n, m, kfold):\n",
    "    count_num = 0\n",
    "    new_cost = 0\n",
    "    percent_change = 1\n",
    "    thetas = np.random.uniform(-1, 1, 11).reshape(11, 1)\n",
    "\n",
    "    #Terminate when absolute value change of loss on the data is\n",
    "    #less than 2−23, or after 1500 iterations have passed (whichever occurs first).\n",
    "    while percent_change > (2 ** -23):\n",
    "        if count_num == 1500:\n",
    "            break\n",
    "\n",
    "        #batch gradient descent\n",
    "        thetas = thetas - ((n / m) * gradient(X_train, y_train, thetas))\n",
    "\n",
    "        #cost\n",
    "        total_cost = -np.sum((1 / m) * ((y_train * np.log(sigmoid(X_train, thetas) + epsilon)) + (1 - y_train) * np.log(\n",
    "            1 - sigmoid(X_train, thetas) + epsilon)))\n",
    "\n",
    "        #percent change\n",
    "        percent_change = abs((new_cost - total_cost) / total_cost)\n",
    "        new_cost = total_cost\n",
    "\n",
    "        count_num = count_num + 1\n",
    "\n",
    "    #print(\"Thetas from batch gradient descent that minimize the loss function: \", thetas, \"\\n\")\n",
    "    if kfold == \"kfold\":\n",
    "        accuracy, precision, recall, F1, TP, FP, TN, FN = prediction(y_test, X_test, thetas, kfold)\n",
    "        return accuracy, precision, recall, F1, TP, FP, TN, FN\n",
    "    if kfold == \"no\":\n",
    "        prediction(y_test, X_test, thetas, kfold)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Stochastic Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "def logisticSGD(X_train, y_train, y_test, X_test, n, m, kfold):\n",
    "    stoch_X_train = X_train\n",
    "    stoch_y_train = y_train\n",
    "    thetas = np.random.uniform(-1, 1, 11).reshape(11, 1)\n",
    "    new_cost = 1\n",
    "    percent_change = 1\n",
    "    count_num = 1\n",
    "\n",
    "    #Terminate when absolute value change of loss on the data is less than 2−23, or after 1500 iterations have passed.\n",
    "    while percent_change > (2 ** -23):\n",
    "        if count_num == 1500:\n",
    "            break\n",
    "\n",
    "        #stochastic gradient descent\n",
    "        index = np.random.permutation(stoch_X_train.shape[0])\n",
    "        stoch_X_train = np.take(stoch_X_train, index, axis=0)\n",
    "        stoch_y_train = np.take(stoch_y_train, index, axis=0)\n",
    "        thetas = thetas - ((n / m) * gradient(stoch_X_train, stoch_y_train, thetas))\n",
    "\n",
    "        #cost\n",
    "        total_cost = -np.sum(\n",
    "            (1 / m) * (stoch_y_train * np.log(sigmoid(stoch_X_train, thetas) + epsilon)) + (1 - stoch_y_train) * np.log(\n",
    "                1 - sigmoid(stoch_X_train, thetas) + epsilon))\n",
    "\n",
    "        #percent change\n",
    "        percent_change = abs((new_cost - total_cost) / total_cost)\n",
    "        new_cost = total_cost\n",
    "        count_num = count_num + 1\n",
    "\n",
    "    #print(\"Thetas from stochastic gradient descent that minimize the loss function: \", thetas, \"\\n\")\n",
    "    if kfold == \"kfold\":\n",
    "        accuracy, precision, recall, F1, TP, FP, TN, FN = prediction(y_test, X_test, thetas, kfold)\n",
    "        return accuracy, precision, recall, F1, TP, FP, TN, FN\n",
    "    if kfold == \"no\":\n",
    "        prediction(y_test, X_test, thetas, kfold)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### K-Folds split the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cv_split(data, numfolds):\n",
    "    dataset = data\n",
    "    data_folds = []\n",
    "    fold_size = int(dataset.shape[0] / numfolds)\n",
    "\n",
    "    for i in range(numfolds):\n",
    "        fold = []\n",
    "\n",
    "        #add random data to the folds\n",
    "        while len(fold) < fold_size:\n",
    "            index = random.randrange(dataset.shape[0])\n",
    "            # save data at index to fold \n",
    "            fold.append(dataset.iloc[index].values.tolist())\n",
    "            # delete data line from dataset\n",
    "            dataset = dataset.drop(index, axis=0)\n",
    "            dataset.reset_index(drop=True, inplace=True)\n",
    "        data_folds.append(np.asarray(fold))\n",
    "\n",
    "    return data_folds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Cross Validation K-Folds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "def kfold(data, model, kfold, f=5):\n",
    "    data = cv_split(data, f)\n",
    "    result = []\n",
    "    accuracy = []\n",
    "    precision = []\n",
    "    recall = []\n",
    "    F1 = []\n",
    "    TP = []\n",
    "    FP = []\n",
    "    TN = []\n",
    "    FN = []\n",
    "    # determine training and test sets \n",
    "    for i in range(f):\n",
    "        r = list(range(f))\n",
    "        testi = data[i]\n",
    "        r.pop(i)\n",
    "        for j in r:\n",
    "            if j == r[0]:\n",
    "                traini = data[j]\n",
    "            else:\n",
    "                traini = np.concatenate((traini, data[j]), axis=0)\n",
    "\n",
    "        k_train_x = traini[:, 0:-1]\n",
    "        k_train_y = traini[:, -1]\n",
    "        k_train_y = k_train_y.reshape(k_train_y.shape[0], 1)\n",
    "        k_test_x = testi[:, 0:-1]\n",
    "        k_test_y = pd.DataFrame(testi[:, -1])\n",
    "\n",
    "        if model == \"batch\":\n",
    "            accuracy1, precision1, recall1, F11, TP1, FP1, TN1, FN1 = logisticBGD(k_train_x, k_train_y, k_test_x,\n",
    "                                                                                  k_test_y, n, m, kfold)\n",
    "            accuracy.append(accuracy1)\n",
    "            precision.append(precision1)\n",
    "            recall.append(recall1)\n",
    "            F1.append(F11)\n",
    "            TP.append(TP1)\n",
    "            FP.append(FP1)\n",
    "            TN.append(TN1)\n",
    "            FN.append(FN1)\n",
    "        if model == \"stochastic\":\n",
    "            accuracy1, precision1, recall1, F11, TP1, FP1, TN1, FN1 = logisticSGD(k_train_x, k_train_y, k_test_y,\n",
    "                                                                                  k_test_x, n, m, kfold)\n",
    "            accuracy.append(accuracy1)\n",
    "            precision.append(precision1)\n",
    "            recall.append(recall1)\n",
    "            F1.append(F11)\n",
    "            TP.append(TP1)\n",
    "            FP.append(FP1)\n",
    "            TN.append(TN1)\n",
    "            FN.append(FN1)\n",
    "    print(\"Accuracies: \", np.array(accuracy) * 100)\n",
    "    print(\"Precisions: \", np.array(precision) * 100)\n",
    "    print(\"Recalls: \", np.array(recall) * 100)\n",
    "    print(\"F-measures: \", np.array(F1) * 100)\n",
    "    print(\"TP: \", TP, \"\\nFP: \", FP, \"\\nTN: \", TN, \"\\nFN: \", FN)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prediction(y_test, X_test, thetas, kfold):\n",
    "    size_test = y_test.shape[0]\n",
    "    predictions = np.zeros(size_test)\n",
    "    TP = FP = TN = FN = accuracy = 0\n",
    "\n",
    "    #find probability for 1 and 0\n",
    "    for a in range(size_test):\n",
    "        p_1 = 1 / (1 + math.e ** -(X_test[a] @ thetas))\n",
    "        p_0 = 1 - p_1\n",
    "        if p_1 > p_0:\n",
    "            predictions[a] = 1\n",
    "        else:\n",
    "            predictions[a] = 0\n",
    "\n",
    "    #find TP, FP, TN, FN\n",
    "    for a in range(size_test):\n",
    "        if predictions[a] == 1:\n",
    "            if predictions[a] == y_test.iloc[a][0]:\n",
    "                TP = TP + 1\n",
    "                accuracy = accuracy + 1\n",
    "            else:\n",
    "                FP = FP + 1\n",
    "        else:\n",
    "            if predictions[a] == y_test.iloc[a][0]:\n",
    "                TN = TN + 1\n",
    "                accuracy = accuracy + 1\n",
    "            else:\n",
    "                FN = FN + 1\n",
    "\n",
    "    #precision, recall, f measure, accuracy\n",
    "    precision = TP / (TP + FP)\n",
    "    recall = TP / (TP + FN)\n",
    "    F1 = (2 * precision * recall) / (precision + recall)\n",
    "    accuracy = accuracy / size_test\n",
    "\n",
    "    if kfold == \"kfold\":\n",
    "        return accuracy, precision, recall, F1, TP, FP, TN, FN\n",
    "\n",
    "    else:\n",
    "        print(\"Accuracy: \", accuracy * 100)\n",
    "        print(\"Precision: \", precision * 100)\n",
    "        print(\"Recall: \", recall * 100)\n",
    "        print(\"F-measure: \", F1 * 100)\n",
    "        print(\"TP: \", TP, \"; FP: \", FP, \"; TN: \", TN, \"; FN: \", FN)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results Without K-Folds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split the dataset into train and test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(s_X, y, test_size=0.33, random_state=0)\n",
    "y_train = y_train.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Common Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "random.seed(0)\n",
    "n = 0.01  #learning rate\n",
    "count = pd.DataFrame()\n",
    "percent_change = 1\n",
    "count_num = 1\n",
    "cost = 1\n",
    "m = X_train.shape[0]\n",
    "#y_train = y_train.to_numpy()\n",
    "epsilon = 1e-5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Logistic Regression using Batch Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  81.11561234029293\n",
      "Precision:  78.09894385769873\n",
      "Recall:  86.88930117501546\n",
      "F-measure:  82.25995316159252\n",
      "TP:  1405 ; FP:  394 ; TN:  1198 ; FN:  212\n"
     ]
    }
   ],
   "source": [
    "logisticBGD(X_train, y_train, X_test, y_test, n, m, \"no\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Logistic Regression using Stochastic Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  81.52072296665628\n",
      "Precision:  78.10098792535675\n",
      "Recall:  88.00247371675944\n",
      "F-measure:  82.75661529514393\n",
      "TP:  1423 ; FP:  399 ; TN:  1193 ; FN:  194\n"
     ]
    }
   ],
   "source": [
    "logisticSGD(X_train, y_train, y_test, X_test, n, m, \"no\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Built-in function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  81.738859457775\n",
      "Precision:  79.98836532867946\n",
      "Recall:  85.03401360544217\n",
      "F-measure:  82.43405275779376\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(s_X, y, test_size=0.33, random_state=0)\n",
    "y_train = y_train.values.ravel()\n",
    "\n",
    "#y_train = y_train.ravel()\n",
    "logistic_regression = LogisticRegression()\n",
    "logistic_regression.fit(X_train, y_train)\n",
    "y_pred = logistic_regression.predict(X_test)\n",
    "y_test = y_test.values\n",
    "y_test = y_test.reshape(3209, )\n",
    "confusion_matrix = pd.crosstab(y_test, y_pred, rownames=['Actual'], colnames=['Predicted'])\n",
    "sns.heatmap(confusion_matrix, annot=True)\n",
    "print('Accuracy: ', metrics.accuracy_score(y_test, y_pred) * 100)\n",
    "print('Precision: ', metrics.precision_score(y_test, y_pred, zero_division=0) * 100)\n",
    "print('Recall: ', metrics.recall_score(y_test, y_pred) * 100)\n",
    "print('F-measure: ', metrics.f1_score(y_test, y_pred, zero_division=0) * 100)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "SGDClassifier(loss='log')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results With K-Folds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = np.append(s_X, y, axis=1)\n",
    "data = pd.DataFrame(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Logistic Regression using Batch Gradient Descent, k-folds = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracies:  [79.42386831 79.9382716  81.06995885 82.81893004 79.83539095 81.27572016\n",
      " 81.79012346 82.61316872 78.60082305 81.99588477]\n",
      "Precisions:  [77.31397459 76.68593449 76.12208259 80.11257036 78.47619048 77.99642218\n",
      " 79.73977695 77.56183746 73.58834244 79.58412098]\n",
      "Recalls:  [85.02994012 84.3220339  89.26315789 87.5        83.23232323 88.08080808\n",
      " 86.31790744 91.26819127 86.50963597 86.2704918 ]\n",
      "F-measures:  [80.98859316 80.32290616 82.17054264 83.64348678 80.78431373 82.73244782\n",
      " 82.89855072 83.85864374 79.52755906 82.79252704]\n",
      "TP:  [426, 398, 424, 427, 412, 436, 429, 439, 404, 421] \n",
      "FP:  [125, 121, 133, 106, 113, 123, 109, 127, 145, 108] \n",
      "TN:  [346, 379, 364, 378, 364, 354, 366, 364, 360, 376] \n",
      "FN:  [75, 74, 51, 61, 83, 59, 68, 42, 63, 67]\n"
     ]
    }
   ],
   "source": [
    "kfold(data, \"batch\", \"kfold\", 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Logistic Regression using Stochastic Gradient Descent, k-folds = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracies:  [82.61316872 80.4526749  82.51028807 80.76131687 83.43621399 80.14403292\n",
      " 80.76131687 82.09876543 80.65843621 80.96707819]\n",
      "Precisions:  [79.81818182 75.84369449 80.03731343 75.64102564 82.08955224 77.87934186\n",
      " 77.02205882 77.37881508 78.32422587 75.65055762]\n",
      "Recalls:  [88.32997988 88.77338877 87.19512195 88.43683084 87.12871287 85.54216867\n",
      " 87.11018711 89.97912317 86.17234469 88.28633406]\n",
      "F-measures:  [83.85864374 81.80076628 83.46303502 81.53998026 84.53410183 81.53110048\n",
      " 81.75609756 83.2046332  82.0610687  81.48148148]\n",
      "TP:  [439, 427, 429, 413, 440, 426, 419, 431, 430, 407] \n",
      "FP:  [111, 136, 107, 133, 96, 121, 125, 126, 119, 131] \n",
      "TN:  [364, 355, 373, 372, 371, 353, 366, 367, 354, 380] \n",
      "FN:  [58, 54, 63, 54, 65, 72, 62, 48, 69, 54]\n"
     ]
    }
   ],
   "source": [
    "kfold(data, \"stochastic\", \"kfold\", 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": 4
}