Neural_Network_Weight_Initialisation_Tricks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4beca8f0",
   "metadata": {},
   "source": [
    "# Artificial Neural Network"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15d65011",
   "metadata": {},
   "source": [
    "### Part 1 - Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b58eec44",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the libraries\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "85547a2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the dataset\n",
    "dataset = pd.read_csv('https://raw.githubusercontent.com/krishnaik06/Complete-Deep-Learning/master/ANN/Churn_Modelling.csv')\n",
    "X = dataset.iloc[:, 3:13]\n",
    "y = dataset.iloc[:, 13]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "536936a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create dummy variables\n",
    "geography=pd.get_dummies(X[\"Geography\"],drop_first=True)\n",
    "gender=pd.get_dummies(X['Gender'],drop_first=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb87230b",
   "metadata": {},
   "source": [
    "### Concatenate the Data Frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "395dcf0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "X=pd.concat([X,geography,gender],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "32fae833",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Drop Unnecessary columns\n",
    "X=X.drop(['Geography','Gender'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "37c6ab41",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training set and Test set\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, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "181afed5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature Scaling\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "085b1b07",
   "metadata": {},
   "source": [
    "### Part 2 - Now let's make the ANN!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "25216a7f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the Keras libraries and packages\n",
    "import keras\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LeakyReLU,PReLU,ELU\n",
    "from keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7b7683ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialising the ANN\n",
    "classifier = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1944859e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding the input layer and the first hidden layer\n",
    "classifier.add(Dense(units= 6, kernel_initializer= 'he_uniform',activation='relu',input_dim = 11))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "f7631188",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding the second hidden layer\n",
    "classifier.add(Dense(units = 6, kernel_initializer = 'he_uniform',activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7b6927ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding the output layer\n",
    "classifier.add(Dense(units = 1, kernel_initializer = 'glorot_uniform', activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e15c5a14",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compiling the ANN\n",
    "classifier.compile(optimizer = 'Adamax', loss = 'binary_crossentropy', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f0fc033e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.4970 - accuracy: 0.7962 - val_loss: 0.4907 - val_accuracy: 0.7955\n",
      "Epoch 2/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4768 - accuracy: 0.7962 - val_loss: 0.4789 - val_accuracy: 0.7955\n",
      "Epoch 3/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4657 - accuracy: 0.7962 - val_loss: 0.4700 - val_accuracy: 0.7955\n",
      "Epoch 4/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4574 - accuracy: 0.7962 - val_loss: 0.4636 - val_accuracy: 0.7955\n",
      "Epoch 5/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4511 - accuracy: 0.7962 - val_loss: 0.4582 - val_accuracy: 0.7955\n",
      "Epoch 6/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.4459 - accuracy: 0.7962 - val_loss: 0.4538 - val_accuracy: 0.7955\n",
      "Epoch 7/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.4416 - accuracy: 0.7962 - val_loss: 0.4503 - val_accuracy: 0.7955\n",
      "Epoch 8/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4380 - accuracy: 0.7962 - val_loss: 0.4476 - val_accuracy: 0.7955\n",
      "Epoch 9/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4350 - accuracy: 0.7974 - val_loss: 0.4451 - val_accuracy: 0.7967\n",
      "Epoch 10/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4324 - accuracy: 0.7985 - val_loss: 0.4433 - val_accuracy: 0.7970\n",
      "Epoch 11/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4301 - accuracy: 0.8003 - val_loss: 0.4416 - val_accuracy: 0.7978\n",
      "Epoch 12/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4282 - accuracy: 0.8015 - val_loss: 0.4399 - val_accuracy: 0.8001\n",
      "Epoch 13/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4265 - accuracy: 0.8044 - val_loss: 0.4386 - val_accuracy: 0.8023\n",
      "Epoch 14/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4250 - accuracy: 0.8061 - val_loss: 0.4372 - val_accuracy: 0.8054\n",
      "Epoch 15/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4235 - accuracy: 0.8080 - val_loss: 0.4359 - val_accuracy: 0.8042\n",
      "Epoch 16/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.4221 - accuracy: 0.8067 - val_loss: 0.4349 - val_accuracy: 0.8039\n",
      "Epoch 17/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4208 - accuracy: 0.8087 - val_loss: 0.4334 - val_accuracy: 0.8058\n",
      "Epoch 18/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.4195 - accuracy: 0.8078 - val_loss: 0.4321 - val_accuracy: 0.8058\n",
      "Epoch 19/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4182 - accuracy: 0.8099 - val_loss: 0.4308 - val_accuracy: 0.8058\n",
      "Epoch 20/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4169 - accuracy: 0.8097 - val_loss: 0.4295 - val_accuracy: 0.8065\n",
      "Epoch 21/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4156 - accuracy: 0.8100 - val_loss: 0.4282 - val_accuracy: 0.8084\n",
      "Epoch 22/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.4144 - accuracy: 0.8113 - val_loss: 0.4270 - val_accuracy: 0.8069\n",
      "Epoch 23/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4131 - accuracy: 0.8108 - val_loss: 0.4255 - val_accuracy: 0.8088\n",
      "Epoch 24/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.4117 - accuracy: 0.8123 - val_loss: 0.4244 - val_accuracy: 0.8084\n",
      "Epoch 25/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.4104 - accuracy: 0.8125 - val_loss: 0.4232 - val_accuracy: 0.8076\n",
      "Epoch 26/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4092 - accuracy: 0.8115 - val_loss: 0.4218 - val_accuracy: 0.8084\n",
      "Epoch 27/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.4078 - accuracy: 0.8134 - val_loss: 0.4206 - val_accuracy: 0.8107\n",
      "Epoch 28/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4063 - accuracy: 0.8130 - val_loss: 0.4194 - val_accuracy: 0.8114\n",
      "Epoch 29/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.4050 - accuracy: 0.8134 - val_loss: 0.4178 - val_accuracy: 0.8122\n",
      "Epoch 30/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.4034 - accuracy: 0.8141 - val_loss: 0.4164 - val_accuracy: 0.8118\n",
      "Epoch 31/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.4019 - accuracy: 0.8149 - val_loss: 0.4150 - val_accuracy: 0.8126\n",
      "Epoch 32/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.4004 - accuracy: 0.8149 - val_loss: 0.4136 - val_accuracy: 0.8133\n",
      "Epoch 33/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3988 - accuracy: 0.8166 - val_loss: 0.4121 - val_accuracy: 0.8152\n",
      "Epoch 34/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3974 - accuracy: 0.8179 - val_loss: 0.4109 - val_accuracy: 0.8137\n",
      "Epoch 35/100\n",
      "536/536 [==============================] - 2s 5ms/step - loss: 0.3958 - accuracy: 0.8186 - val_loss: 0.4093 - val_accuracy: 0.8145\n",
      "Epoch 36/100\n",
      "536/536 [==============================] - 2s 4ms/step - loss: 0.3941 - accuracy: 0.8210 - val_loss: 0.4080 - val_accuracy: 0.8148\n",
      "Epoch 37/100\n",
      "536/536 [==============================] - 2s 4ms/step - loss: 0.3926 - accuracy: 0.8203 - val_loss: 0.4069 - val_accuracy: 0.8160\n",
      "Epoch 38/100\n",
      "536/536 [==============================] - 2s 4ms/step - loss: 0.3910 - accuracy: 0.8224 - val_loss: 0.4052 - val_accuracy: 0.8186\n",
      "Epoch 39/100\n",
      "536/536 [==============================] - 2s 4ms/step - loss: 0.3893 - accuracy: 0.8233 - val_loss: 0.4042 - val_accuracy: 0.8175\n",
      "Epoch 40/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3877 - accuracy: 0.8240 - val_loss: 0.4025 - val_accuracy: 0.8190\n",
      "Epoch 41/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3862 - accuracy: 0.8252 - val_loss: 0.4013 - val_accuracy: 0.8205\n",
      "Epoch 42/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3847 - accuracy: 0.8266 - val_loss: 0.4000 - val_accuracy: 0.8209\n",
      "Epoch 43/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3831 - accuracy: 0.8281 - val_loss: 0.3987 - val_accuracy: 0.8205\n",
      "Epoch 44/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3815 - accuracy: 0.8291 - val_loss: 0.3976 - val_accuracy: 0.8217\n",
      "Epoch 45/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3801 - accuracy: 0.8293 - val_loss: 0.3963 - val_accuracy: 0.8220\n",
      "Epoch 46/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3786 - accuracy: 0.8306 - val_loss: 0.3950 - val_accuracy: 0.8224\n",
      "Epoch 47/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3769 - accuracy: 0.8321 - val_loss: 0.3944 - val_accuracy: 0.8251\n",
      "Epoch 48/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3757 - accuracy: 0.8339 - val_loss: 0.3927 - val_accuracy: 0.8262\n",
      "Epoch 49/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3741 - accuracy: 0.8332 - val_loss: 0.3915 - val_accuracy: 0.8292\n",
      "Epoch 50/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3727 - accuracy: 0.8349 - val_loss: 0.3902 - val_accuracy: 0.8300\n",
      "Epoch 51/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3715 - accuracy: 0.8365 - val_loss: 0.3893 - val_accuracy: 0.8304\n",
      "Epoch 52/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3702 - accuracy: 0.8388 - val_loss: 0.3883 - val_accuracy: 0.8311\n",
      "Epoch 53/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3692 - accuracy: 0.8373 - val_loss: 0.3873 - val_accuracy: 0.8334\n",
      "Epoch 54/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3679 - accuracy: 0.8416 - val_loss: 0.3866 - val_accuracy: 0.8338\n",
      "Epoch 55/100\n",
      "536/536 [==============================] - 2s 4ms/step - loss: 0.3668 - accuracy: 0.8425 - val_loss: 0.3859 - val_accuracy: 0.8357\n",
      "Epoch 56/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3658 - accuracy: 0.8434 - val_loss: 0.3854 - val_accuracy: 0.8360\n",
      "Epoch 57/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3648 - accuracy: 0.8425 - val_loss: 0.3848 - val_accuracy: 0.8364\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 58/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3641 - accuracy: 0.8433 - val_loss: 0.3837 - val_accuracy: 0.8398\n",
      "Epoch 59/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3632 - accuracy: 0.8447 - val_loss: 0.3829 - val_accuracy: 0.8413\n",
      "Epoch 60/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3623 - accuracy: 0.8455 - val_loss: 0.3823 - val_accuracy: 0.8421\n",
      "Epoch 61/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3613 - accuracy: 0.8485 - val_loss: 0.3817 - val_accuracy: 0.8413\n",
      "Epoch 62/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3605 - accuracy: 0.8487 - val_loss: 0.3815 - val_accuracy: 0.8402\n",
      "Epoch 63/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3599 - accuracy: 0.8483 - val_loss: 0.3810 - val_accuracy: 0.8410\n",
      "Epoch 64/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3593 - accuracy: 0.8481 - val_loss: 0.3808 - val_accuracy: 0.8436\n",
      "Epoch 65/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3586 - accuracy: 0.8487 - val_loss: 0.3799 - val_accuracy: 0.8425\n",
      "Epoch 66/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3580 - accuracy: 0.8487 - val_loss: 0.3795 - val_accuracy: 0.8444\n",
      "Epoch 67/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3574 - accuracy: 0.8494 - val_loss: 0.3791 - val_accuracy: 0.8440\n",
      "Epoch 68/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3568 - accuracy: 0.8505 - val_loss: 0.3787 - val_accuracy: 0.8451\n",
      "Epoch 69/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3562 - accuracy: 0.8515 - val_loss: 0.3785 - val_accuracy: 0.8455\n",
      "Epoch 70/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3558 - accuracy: 0.8520 - val_loss: 0.3782 - val_accuracy: 0.8455\n",
      "Epoch 71/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3551 - accuracy: 0.8524 - val_loss: 0.3782 - val_accuracy: 0.8451\n",
      "Epoch 72/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3548 - accuracy: 0.8524 - val_loss: 0.3774 - val_accuracy: 0.8455\n",
      "Epoch 73/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3543 - accuracy: 0.8543 - val_loss: 0.3770 - val_accuracy: 0.8459\n",
      "Epoch 74/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3537 - accuracy: 0.8550 - val_loss: 0.3768 - val_accuracy: 0.8466\n",
      "Epoch 75/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3532 - accuracy: 0.8545 - val_loss: 0.3763 - val_accuracy: 0.8459\n",
      "Epoch 76/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3528 - accuracy: 0.8563 - val_loss: 0.3758 - val_accuracy: 0.8459\n",
      "Epoch 77/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3523 - accuracy: 0.8574 - val_loss: 0.3757 - val_accuracy: 0.8466\n",
      "Epoch 78/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3520 - accuracy: 0.8567 - val_loss: 0.3756 - val_accuracy: 0.8459\n",
      "Epoch 79/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3516 - accuracy: 0.8584 - val_loss: 0.3751 - val_accuracy: 0.8478\n",
      "Epoch 80/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3513 - accuracy: 0.8582 - val_loss: 0.3749 - val_accuracy: 0.8459\n",
      "Epoch 81/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3506 - accuracy: 0.8574 - val_loss: 0.3755 - val_accuracy: 0.8455\n",
      "Epoch 82/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3505 - accuracy: 0.8578 - val_loss: 0.3752 - val_accuracy: 0.8448\n",
      "Epoch 83/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3502 - accuracy: 0.8584 - val_loss: 0.3744 - val_accuracy: 0.8459\n",
      "Epoch 84/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3496 - accuracy: 0.8591 - val_loss: 0.3741 - val_accuracy: 0.8493\n",
      "Epoch 85/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3491 - accuracy: 0.8574 - val_loss: 0.3736 - val_accuracy: 0.8508\n",
      "Epoch 86/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3487 - accuracy: 0.8576 - val_loss: 0.3740 - val_accuracy: 0.8474\n",
      "Epoch 87/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3484 - accuracy: 0.8574 - val_loss: 0.3736 - val_accuracy: 0.8463\n",
      "Epoch 88/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3480 - accuracy: 0.8600 - val_loss: 0.3733 - val_accuracy: 0.8451\n",
      "Epoch 89/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3475 - accuracy: 0.8597 - val_loss: 0.3731 - val_accuracy: 0.8523\n",
      "Epoch 90/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3474 - accuracy: 0.8595 - val_loss: 0.3727 - val_accuracy: 0.8463\n",
      "Epoch 91/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3470 - accuracy: 0.8591 - val_loss: 0.3724 - val_accuracy: 0.8466\n",
      "Epoch 92/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3468 - accuracy: 0.8600 - val_loss: 0.3722 - val_accuracy: 0.8485\n",
      "Epoch 93/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3463 - accuracy: 0.8587 - val_loss: 0.3720 - val_accuracy: 0.8478\n",
      "Epoch 94/100\n",
      "536/536 [==============================] - 1s 3ms/step - loss: 0.3462 - accuracy: 0.8593 - val_loss: 0.3718 - val_accuracy: 0.8478\n",
      "Epoch 95/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3458 - accuracy: 0.8597 - val_loss: 0.3717 - val_accuracy: 0.8504\n",
      "Epoch 96/100\n",
      "536/536 [==============================] - 2s 5ms/step - loss: 0.3456 - accuracy: 0.8591 - val_loss: 0.3715 - val_accuracy: 0.8482\n",
      "Epoch 97/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3453 - accuracy: 0.8597 - val_loss: 0.3715 - val_accuracy: 0.8478\n",
      "Epoch 98/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3450 - accuracy: 0.8604 - val_loss: 0.3714 - val_accuracy: 0.8478\n",
      "Epoch 99/100\n",
      "536/536 [==============================] - 1s 2ms/step - loss: 0.3448 - accuracy: 0.8599 - val_loss: 0.3709 - val_accuracy: 0.8482\n",
      "Epoch 100/100\n",
      "536/536 [==============================] - 2s 3ms/step - loss: 0.3445 - accuracy: 0.8608 - val_loss: 0.3710 - val_accuracy: 0.8489\n"
     ]
    }
   ],
   "source": [
    "# Fitting the ANN to the Training set\n",
    "model_history=classifier.fit(X_train, y_train,validation_split=0.33, batch_size = 10, epochs = 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a8446191",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_history.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "545376ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# list all data in history\n",
    "print(model_history.history.keys())\n",
    "# summarize history for accuracy\n",
    "plt.figure(figsize=(14,8))\n",
    "plt.plot(model_history.history['accuracy'])\n",
    "plt.plot(model_history.history['val_accuracy'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "bc8269ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# summarize history for loss\n",
    "plt.figure(figsize=(14,8))\n",
    "plt.plot(model_history.history['loss'])\n",
    "plt.plot(model_history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ff61a31",
   "metadata": {},
   "source": [
    "### Part 3 - Making the predictions and evaluating the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "44a34940",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63/63 [==============================] - 0s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "# Predicting the Test set results\n",
    "y_pred = classifier.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "08c96ec6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Making the Confusion Matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "97b63b31",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1504,   91],\n",
       "       [ 197,  208]], dtype=int64)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "5d67873a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the Accuracy\n",
    "from sklearn.metrics import accuracy_score\n",
    "score=accuracy_score(y_pred,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "3cc44781",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.856"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}