ML-assignment 1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "spoken-hybrid",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "dangerous-remedy",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Import dataset\n",
    "data= pd.read_csv(\"diabetes.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "shared-harvest",
   "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>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "      <th>Outcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>148</td>\n",
       "      <td>72</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>33.6</td>\n",
       "      <td>0.627</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>66</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>26.6</td>\n",
       "      <td>0.351</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>183</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23.3</td>\n",
       "      <td>0.672</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>66</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>28.1</td>\n",
       "      <td>0.167</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>168</td>\n",
       "      <td>43.1</td>\n",
       "      <td>2.288</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pregnancies  Glucose  BloodPressure  SkinThickness  Insulin   BMI  \\\n",
       "0            6      148             72             35        0  33.6   \n",
       "1            1       85             66             29        0  26.6   \n",
       "2            8      183             64              0        0  23.3   \n",
       "3            1       89             66             23       94  28.1   \n",
       "4            0      137             40             35      168  43.1   \n",
       "\n",
       "   DiabetesPedigreeFunction  Age  Outcome  \n",
       "0                     0.627   50        1  \n",
       "1                     0.351   31        0  \n",
       "2                     0.672   32        1  \n",
       "3                     0.167   21        0  \n",
       "4                     2.288   33        1  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "concerned-faith",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pre-processing block"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "communist-expert",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Train, test and validation spilt\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "data_set_values=np.array(data)\n",
    "number_of_rows,number_of_cols=data.shape\n",
    "train, test = train_test_split(data_set_values, test_size = 0.4,random_state=57)\n",
    "test, validate = train_test_split(test, test_size = 0.5,random_state=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "metallic-poison",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Spilt all sets into input and output sets\n",
    "train_x=train[:,:number_of_cols-1]\n",
    "train_y=train[:,[number_of_cols-1]]\n",
    "\n",
    "test_x=test[:,:number_of_cols-1]\n",
    "test_y=test[:,[number_of_cols-1]]\n",
    "\n",
    "validate_x=validate[:,:number_of_cols-1]\n",
    "validate_y=validate[:,[number_of_cols-1]]\n",
    "\n",
    "#perform normalization on the input data set\n",
    "input_data= np.vstack((train_x,test_x,validate_x))\n",
    "mean_data = np.mean(input_data, axis=0)\n",
    "std_data = np.std(input_data, axis=0)\n",
    "train_x = (train_x - mean_data)/std_data\n",
    "test_x = (test_x - mean_data)/std_data\n",
    "validate_x = (validate_x - mean_data)/std_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "macro-assets",
   "metadata": {},
   "outputs": [],
   "source": [
    "#End of Pre-processing block"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "entitled-concert",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Start of Logistic Regression - Part 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "prescription-namibia",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Logistic Regression - Model Definition\n",
    "class logisticRegression:\n",
    "    def __init__(self,total_iterations,learning_rate):\n",
    "        self.learning_rate=learning_rate\n",
    "        self.total_iterations=total_iterations\n",
    "        self.bias=0\n",
    "        \n",
    "    #Activation function\n",
    "    def sigmoidFunction(self,X): \n",
    "        return 1/(1+np.exp(-X))\n",
    "#         return 1/(1+scipy.special.expit(-X))\n",
    "    \n",
    "    #Training function\n",
    "    def train(self,input,Y): \n",
    "\n",
    "        rows,cols=input.shape\n",
    "        self.m=rows\n",
    "        self.w=np.zeros((cols,1))\n",
    "        costList=[]\n",
    "        iterationList=[]\n",
    "        for _ in range(self.total_iterations+1):\n",
    "            # y= x * w + b\n",
    "            output=np.dot(input,self.w)+self.bias\n",
    "            \n",
    "            #apply Activating function\n",
    "            Z=self.sigmoidFunction(output)\n",
    "            \n",
    "            #calculating cost\n",
    "            cost=(-1/self.m)*np.sum((Y*np.log(Z))+(1-Y)*np.log(1-Z))\n",
    "            \n",
    "            if _%1000==0:\n",
    "#                 print(\"loss during \",_,\" iteration =\",cost)\n",
    "                iterationList.append(_)\n",
    "                costList.append(cost)\n",
    "            \n",
    "            #Gradient Descent\n",
    "            dw=(1/self.m)*np.dot(input.T,np.subtract(Z,Y))\n",
    "            db=(1/self.m)*np.sum(np.subtract(Z,Y))\n",
    "            \n",
    "            #updating weights and bais (Learning Part)\n",
    "            self.w = self.w-self.learning_rate*dw\n",
    "            self.bias = self.bias - self.learning_rate*db\n",
    "        plt.plot(iterationList, costList)\n",
    "        plt.title(\"Cost Relation\")\n",
    "        plt.xlabel(\"Iterations\")\n",
    "        plt.ylabel(\"Cost\")\n",
    "        plt.show()\n",
    "        return \n",
    "\n",
    "    #Prediction function\n",
    "    def predict(self,input):\n",
    "        output=np.dot(input,self.w)+self.bias\n",
    "        Z=self.sigmoidFunction(output)\n",
    "        # if output more than 0.5 then yes else no. (Binary Classification)\n",
    "        prediction=[]\n",
    "        for i in Z:\n",
    "            if i>=0.5:\n",
    "                prediction.append(1)\n",
    "            else:\n",
    "                prediction.append(0)\n",
    "        return prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "postal-spider",
   "metadata": {},
   "outputs": [],
   "source": [
    "model=logisticRegression(total_iterations=10000,learning_rate=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "introductory-showcase",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.train(train_x,train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "accepting-canal",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "test_prediction = model.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "immediate-republican",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy =  0.8116883116883117\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy = \", accuracy_score(test_y,test_prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "front-voluntary",
   "metadata": {},
   "outputs": [],
   "source": [
    "#End of Logistic Regression - Part 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "covered-tamil",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Start of Neural Net using L2 regularization - Part 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "widespread-graduate",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_3 (Dense)              (None, 512)               4608      \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 4)                 2052      \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 1)                 5         \n",
      "=================================================================\n",
      "Total params: 6,665\n",
      "Trainable params: 6,665\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "tf.random.set_seed(221)\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras import regularizers\n",
    "from keras.regularizers import l2\n",
    "from tensorflow.keras import optimizers\n",
    "\n",
    "input_shape = train_x.shape\n",
    "neuralNetwork = Sequential()\n",
    "neuralNetwork.add(Dense(512, input_dim=input_shape[1], activation='relu',kernel_regularizer=regularizers.L1L2( l2=0.1)))\n",
    "neuralNetwork.add(Dense(4, activation='relu'))\n",
    "neuralNetwork.add(Dense(1, activation='sigmoid' ))\n",
    "# Compile the model\n",
    "neuralNetwork.compile(loss='binary_crossentropy',  \n",
    "                     optimizer=optimizers.Adam(learning_rate=0.005),\n",
    "                     metrics=['accuracy'])\n",
    "neuralNetwork.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "artificial-sullivan",
   "metadata": {},
   "outputs": [],
   "source": [
    "#train the model\n",
    "history = neuralNetwork.fit(train_x,\n",
    "                        train_y,\n",
    "                        validation_data=(validate_x, validate_y),\n",
    "                        epochs=100,\n",
    "                        batch_size=20,\n",
    "                        verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "ranking-collect",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.title('Model Accuracy Graph')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train data', 'validation data'], loc='center right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "faced-advertising",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model Loss Graph')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train data', 'validation data'], loc='center right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "incident-onion",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy : 0.8246753215789795\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "score = neuralNetwork.evaluate( \n",
    "                x=test_x,\n",
    "                y=test_y,\n",
    "                verbose=0\n",
    "                )\n",
    "print(\"Accuracy :\",score[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "settled-times",
   "metadata": {},
   "outputs": [],
   "source": [
    "#End of Neural Net using L2 regularization - part 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "artistic-legislation",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Start of part 3 - NN using L2 regularization and NN using Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "first-culture",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_5\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_12 (Dense)             (None, 256)               2304      \n",
      "_________________________________________________________________\n",
      "dense_13 (Dense)             (None, 1)                 257       \n",
      "=================================================================\n",
      "Total params: 2,561\n",
      "Trainable params: 2,561\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# NN using L2 regularization\n",
    "import tensorflow as tf\n",
    "tf.random.set_seed(221)\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras import regularizers\n",
    "from keras.regularizers import l2\n",
    "from tensorflow.keras import optimizers\n",
    "from keras.layers import Dropout\n",
    "\n",
    "input_shape = train_x.shape\n",
    "nn1 = Sequential()\n",
    "# L2 Regularizer\n",
    "nn1.add(Dense(256, input_dim=input_shape[1], activation='relu',kernel_regularizer=regularizers.L1L2( l2=0.1)))\n",
    "nn1.add(Dense(1, activation='sigmoid' ))\n",
    "# Compile the model\n",
    "nn1.compile(loss='binary_crossentropy',  \n",
    "                     optimizer=optimizers.Adam(learning_rate=0.005),\n",
    "                     metrics=['accuracy'])\n",
    "nn1.summary()\n",
    "history1 = nn1.fit(train_x,\n",
    "                        train_y,\n",
    "                        validation_data=(validate_x, validate_y),\n",
    "                        epochs=100,\n",
    "                        batch_size=20,\n",
    "                        verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "sound-president",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_6\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_14 (Dense)             (None, 256)               2304      \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_15 (Dense)             (None, 1)                 257       \n",
      "=================================================================\n",
      "Total params: 2,561\n",
      "Trainable params: 2,561\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# NN using Dropout\n",
    "import tensorflow as tf\n",
    "tf.random.set_seed(221)\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras import regularizers\n",
    "from keras.regularizers import l2\n",
    "from tensorflow.keras import optimizers\n",
    "from keras.layers import Dropout\n",
    "\n",
    "input_shape = train_x.shape\n",
    "\n",
    "nn2 = Sequential()\n",
    "# Dropout model\n",
    "nn2.add(Dense(256, input_dim=input_shape[1], activation='relu'))\n",
    "nn2.add(Dropout(0.2))\n",
    "nn2.add(Dense(1, activation='sigmoid' ))\n",
    "# Compile the model\n",
    "nn2.compile(loss='binary_crossentropy',  \n",
    "                     optimizer=optimizers.Adam(learning_rate=0.005),\n",
    "                     metrics=['accuracy'])\n",
    "nn2.summary()\n",
    "history2 = nn2.fit(train_x,\n",
    "                        train_y,\n",
    "                        validation_data=(validate_x, validate_y),\n",
    "                        epochs=100,\n",
    "                        batch_size=20,\n",
    "                        verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "running-scottish",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history1.history['accuracy'])\n",
    "plt.plot(history2.history['accuracy'])\n",
    "plt.title('Accuracy comparison of Neural Network using \\n Dropout and Neural Network using L2 regularizer')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['L2', 'Dropout'], loc='center right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "fuzzy-mustang",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history1.history['loss'])\n",
    "plt.plot(history2.history['loss'])\n",
    "plt.title('Loss comparison of Neural Network using \\n Dropout and Neural Network using L2 regularizer')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['L2', 'Dropout'], loc='center right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "innovative-masters",
   "metadata": {},
   "outputs": [],
   "source": [
    "#End of part 3 "
   ]
  }
 ],
 "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": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}