Logistic_Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "8eab3464",
   "metadata": {},
   "source": [
    "Importing necessary libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "4dd0a2c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from keras.regularizers import l2\n",
    "from keras.regularizers import l1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abc39f11",
   "metadata": {},
   "source": [
    "Reading Dataset into DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "61fa715d",
   "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": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df = pd.read_csv('diabetes.csv')\n",
    "data_df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ded7847b",
   "metadata": {},
   "source": [
    "checking for 0's in colums of each feature. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "013fea40",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pregnancies                 111\n",
       "Glucose                       5\n",
       "BloodPressure                35\n",
       "SkinThickness               227\n",
       "Insulin                     374\n",
       "BMI                          11\n",
       "DiabetesPedigreeFunction      0\n",
       "Age                           0\n",
       "Outcome                     500\n",
       "dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zeros_data=(data_df[:]==0)\n",
    "zeros_data.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc5223a6",
   "metadata": {},
   "source": [
    "Cleaning Dataset by substituting mean of each column in the palce of it's vlaues where the the value is 0 respectively"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "27a76170",
   "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": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_clean = data_df.drop('Outcome',axis=1)\n",
    "for col in data_clean.columns:\n",
    "    val = data_df[col].mean()\n",
    "    data_df[col] = data_df[col].replace(0, val)\n",
    "data_clean['Outcome']=data_df['Outcome']\n",
    "data_clean.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3534e414",
   "metadata": {},
   "source": [
    "After eleminating 0's from the feature vector we can now see there are no 0's in each column in available 8 features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "2760f325",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pregnancies                 111\n",
       "Glucose                       5\n",
       "BloodPressure                35\n",
       "SkinThickness               227\n",
       "Insulin                     374\n",
       "BMI                          11\n",
       "DiabetesPedigreeFunction      0\n",
       "Age                           0\n",
       "Outcome                     500\n",
       "dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_clean['Outcome'] = data_df['Outcome']\n",
    "(data_clean[:]==0).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcbdfdbf",
   "metadata": {},
   "source": [
    "seperating feature_vector which is independnt varibles and outcome_vector which is dependent varibale "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "a22fcfbd",
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_vector = data_clean.iloc[0:,0:8].values\n",
    "outcome_vector = data_clean.iloc[0:,8].values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9214fa53",
   "metadata": {},
   "source": [
    "using sklearn we will now scale the data using prepocessing method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "b7fd70e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "5e8c9cfc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.88235294, -0.14572864,  0.08196721, -0.41414141, -1.        ,\n",
       "       -0.2071535 , -0.76686593, -0.66666667])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "min_max_scaler =MinMaxScaler(feature_range=(-1,1))\n",
    "feature_vector = min_max_scaler.fit_transform(feature_vector)\n",
    "feature_vector[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "8317d795",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(768, 8) (768,)\n"
     ]
    }
   ],
   "source": [
    "print(feature_vector.shape,outcome_vector.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0830e113",
   "metadata": {},
   "source": [
    "We split data into Training ,Validation and Testing dataset using train_test_split method.\n",
    "From the avilable 100% dataset, we split in such a way that we get\n",
    "60% Training dataset, 20% validation dataset, 20% testing datast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "390d0fdc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Training Datast Feature Vector : (460, 8)    Training Dataset Outcome Vector : (460,)\n",
      "Validation Datast Feature Vector : (154, 8)  Validation Dataset Outcome Vector : (154,)\n",
      "   Testing Datast Feature Vector : (154, 8)     Testing Dataset Outcome Vector : (154,)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "feature_initial_training, feature_test,outcome_initial_training, outcome_test = train_test_split(feature_vector,outcome_vector, test_size=0.2)\n",
    "feature_final_training,feature_valid,outcome_final_training,outcome_valid =train_test_split(feature_initial_training,outcome_initial_training,test_size = 0.25)\n",
    "print(\"  Training Datast Feature Vector :\" ,feature_final_training.shape,\"   Training Dataset Outcome Vector :\",outcome_final_training.shape)\n",
    "print(\"Validation Datast Feature Vector :\",feature_valid.shape,\" Validation Dataset Outcome Vector :\",outcome_valid.shape)\n",
    "print(\"   Testing Datast Feature Vector :\",feature_test.shape,\"    Testing Dataset Outcome Vector :\",outcome_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6444339a",
   "metadata": {},
   "source": [
    "Visualizing the shape of Training,Validation and Tetsing dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "3babb117",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset = {'Training Data Set':len(feature_final_training), 'Validation Data Set':len(feature_valid), 'Test Data Set':len(feature_test)}\n",
    "split_data = list(dataset.keys())\n",
    "split_datasize = list(dataset.values())\n",
    "  \n",
    "fig = plt.figure(figsize = (12, 7))\n",
    " \n",
    "plt.bar(split_data,split_datasize,\n",
    "        width = 0.4,color=('purple', 'purple', 'purple'))\n",
    "plt.xlabel(\"Types of data\")\n",
    "plt.ylabel(\"No. of tupples in data\")\n",
    "plt.title(\"comparission of sample sizes\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "887514a0",
   "metadata": {},
   "source": [
    "Now we will train the model using training dataset and validate the model with validation dataset.\n",
    "we use one input layer , 1st Hidden layer with 12 nuerons and activation function is relu , 2nd hidden layer with 8 neurons with activation function relu.\n",
    "one output layer with one neuron because our output is binary\n",
    "\n",
    "We use accuracy and loss as our metric to evalute how good is our model\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9cb3ace",
   "metadata": {},
   "source": [
    "Now in the first training model we will use L1 Regularization \n",
    "Hyper parameter are\n",
    "Input layer with 8 features as input\n",
    "Hidden layer 1 has 12 nuerons\n",
    "Hidden Layer 2 has 8 nuerons\n",
    "output layer has 1 nueron\n",
    "No. of epochs are 100\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "89f4137e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "15/15 [==============================] - 0s 7ms/step - loss: 3.3994 - accuracy: 0.2870 - val_loss: 3.2989 - val_accuracy: 0.2987\n",
      "Epoch 2/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 3.2309 - accuracy: 0.3435 - val_loss: 3.1465 - val_accuracy: 0.3636\n",
      "Epoch 3/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 3.0699 - accuracy: 0.4587 - val_loss: 3.0013 - val_accuracy: 0.4740\n",
      "Epoch 4/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.9165 - accuracy: 0.5609 - val_loss: 2.8629 - val_accuracy: 0.5390\n",
      "Epoch 5/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.7709 - accuracy: 0.6348 - val_loss: 2.7300 - val_accuracy: 0.5649\n",
      "Epoch 6/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.6294 - accuracy: 0.6565 - val_loss: 2.6013 - val_accuracy: 0.5779\n",
      "Epoch 7/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.4969 - accuracy: 0.6587 - val_loss: 2.4797 - val_accuracy: 0.5779\n",
      "Epoch 8/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.3700 - accuracy: 0.6587 - val_loss: 2.3618 - val_accuracy: 0.5779\n",
      "Epoch 9/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.2492 - accuracy: 0.6587 - val_loss: 2.2469 - val_accuracy: 0.5779\n",
      "Epoch 10/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.1306 - accuracy: 0.6587 - val_loss: 2.1306 - val_accuracy: 0.5779\n",
      "Epoch 11/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 2.0159 - accuracy: 0.6587 - val_loss: 2.0204 - val_accuracy: 0.5779\n",
      "Epoch 12/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.9054 - accuracy: 0.6587 - val_loss: 1.9114 - val_accuracy: 0.5779\n",
      "Epoch 13/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.7984 - accuracy: 0.6587 - val_loss: 1.8059 - val_accuracy: 0.5779\n",
      "Epoch 14/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.6964 - accuracy: 0.6587 - val_loss: 1.7074 - val_accuracy: 0.5779\n",
      "Epoch 15/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.6011 - accuracy: 0.6587 - val_loss: 1.6143 - val_accuracy: 0.5779\n",
      "Epoch 16/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.5103 - accuracy: 0.6587 - val_loss: 1.5201 - val_accuracy: 0.5779\n",
      "Epoch 17/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.4236 - accuracy: 0.6587 - val_loss: 1.4325 - val_accuracy: 0.5779\n",
      "Epoch 18/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.3418 - accuracy: 0.6587 - val_loss: 1.3540 - val_accuracy: 0.5779\n",
      "Epoch 19/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.2622 - accuracy: 0.6609 - val_loss: 1.2757 - val_accuracy: 0.5779\n",
      "Epoch 20/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.1857 - accuracy: 0.6630 - val_loss: 1.1980 - val_accuracy: 0.5779\n",
      "Epoch 21/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.1137 - accuracy: 0.6630 - val_loss: 1.1314 - val_accuracy: 0.5779\n",
      "Epoch 22/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.0499 - accuracy: 0.6630 - val_loss: 1.0711 - val_accuracy: 0.5779\n",
      "Epoch 23/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.9913 - accuracy: 0.6652 - val_loss: 1.0127 - val_accuracy: 0.5779\n",
      "Epoch 24/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.9364 - accuracy: 0.6652 - val_loss: 0.9609 - val_accuracy: 0.5779\n",
      "Epoch 25/100\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 0.8881 - accuracy: 0.6652 - val_loss: 0.9145 - val_accuracy: 0.5844\n",
      "Epoch 26/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.8460 - accuracy: 0.6717 - val_loss: 0.8752 - val_accuracy: 0.5844\n",
      "Epoch 27/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.8080 - accuracy: 0.6674 - val_loss: 0.8406 - val_accuracy: 0.5844\n",
      "Epoch 28/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7758 - accuracy: 0.6587 - val_loss: 0.8105 - val_accuracy: 0.5779\n",
      "Epoch 29/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7478 - accuracy: 0.6587 - val_loss: 0.7812 - val_accuracy: 0.5844\n",
      "Epoch 30/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7236 - accuracy: 0.6609 - val_loss: 0.7541 - val_accuracy: 0.5844\n",
      "Epoch 31/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7003 - accuracy: 0.6587 - val_loss: 0.7318 - val_accuracy: 0.5779\n",
      "Epoch 32/100\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 0.6810 - accuracy: 0.6587 - val_loss: 0.7158 - val_accuracy: 0.5779\n",
      "Epoch 33/100\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 0.6683 - accuracy: 0.6587 - val_loss: 0.7047 - val_accuracy: 0.5779\n",
      "Epoch 34/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6561 - accuracy: 0.6587 - val_loss: 0.6896 - val_accuracy: 0.5779\n",
      "Epoch 35/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6459 - accuracy: 0.6587 - val_loss: 0.6795 - val_accuracy: 0.5779\n",
      "Epoch 36/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6400 - accuracy: 0.6587 - val_loss: 0.6774 - val_accuracy: 0.5779\n",
      "Epoch 37/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6364 - accuracy: 0.6587 - val_loss: 0.6726 - val_accuracy: 0.5779\n",
      "Epoch 38/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6343 - accuracy: 0.6587 - val_loss: 0.6713 - val_accuracy: 0.5779\n",
      "Epoch 39/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6329 - accuracy: 0.6587 - val_loss: 0.6713 - val_accuracy: 0.5779\n",
      "Epoch 40/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6312 - accuracy: 0.6609 - val_loss: 0.6684 - val_accuracy: 0.5844\n",
      "Epoch 41/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6302 - accuracy: 0.6609 - val_loss: 0.6672 - val_accuracy: 0.5844\n",
      "Epoch 42/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6276 - accuracy: 0.6587 - val_loss: 0.6628 - val_accuracy: 0.6039\n",
      "Epoch 43/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6255 - accuracy: 0.6609 - val_loss: 0.6607 - val_accuracy: 0.6039\n",
      "Epoch 44/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6259 - accuracy: 0.6761 - val_loss: 0.6596 - val_accuracy: 0.6104\n",
      "Epoch 45/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6236 - accuracy: 0.6761 - val_loss: 0.6563 - val_accuracy: 0.6169\n",
      "Epoch 46/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6214 - accuracy: 0.6848 - val_loss: 0.6513 - val_accuracy: 0.6429\n",
      "Epoch 47/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6185 - accuracy: 0.6957 - val_loss: 0.6457 - val_accuracy: 0.6494\n",
      "Epoch 48/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6162 - accuracy: 0.7065 - val_loss: 0.6501 - val_accuracy: 0.6429\n",
      "Epoch 49/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6144 - accuracy: 0.6957 - val_loss: 0.6442 - val_accuracy: 0.6494\n",
      "Epoch 50/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6120 - accuracy: 0.7087 - val_loss: 0.6413 - val_accuracy: 0.6558\n",
      "Epoch 51/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6116 - accuracy: 0.7152 - val_loss: 0.6391 - val_accuracy: 0.6688\n",
      "Epoch 52/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6084 - accuracy: 0.7109 - val_loss: 0.6377 - val_accuracy: 0.6688\n",
      "Epoch 53/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6067 - accuracy: 0.7152 - val_loss: 0.6337 - val_accuracy: 0.6948\n",
      "Epoch 54/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6051 - accuracy: 0.7413 - val_loss: 0.6284 - val_accuracy: 0.7013\n",
      "Epoch 55/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6024 - accuracy: 0.7370 - val_loss: 0.6269 - val_accuracy: 0.7013\n",
      "Epoch 56/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6004 - accuracy: 0.7457 - val_loss: 0.6249 - val_accuracy: 0.7078\n",
      "Epoch 57/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6020 - accuracy: 0.7261 - val_loss: 0.6226 - val_accuracy: 0.7208\n",
      "Epoch 58/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5978 - accuracy: 0.7500 - val_loss: 0.6158 - val_accuracy: 0.7403\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 59/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5972 - accuracy: 0.7543 - val_loss: 0.6205 - val_accuracy: 0.7273\n",
      "Epoch 60/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5943 - accuracy: 0.7522 - val_loss: 0.6127 - val_accuracy: 0.7338\n",
      "Epoch 61/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5932 - accuracy: 0.7565 - val_loss: 0.6098 - val_accuracy: 0.7403\n",
      "Epoch 62/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5912 - accuracy: 0.7565 - val_loss: 0.6108 - val_accuracy: 0.7403\n",
      "Epoch 63/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5890 - accuracy: 0.7565 - val_loss: 0.6076 - val_accuracy: 0.7338\n",
      "Epoch 64/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5873 - accuracy: 0.7565 - val_loss: 0.6081 - val_accuracy: 0.7338\n",
      "Epoch 65/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5852 - accuracy: 0.7630 - val_loss: 0.6026 - val_accuracy: 0.7532\n",
      "Epoch 66/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5837 - accuracy: 0.7522 - val_loss: 0.5991 - val_accuracy: 0.7857\n",
      "Epoch 67/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5823 - accuracy: 0.7522 - val_loss: 0.6016 - val_accuracy: 0.7532\n",
      "Epoch 68/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5814 - accuracy: 0.7609 - val_loss: 0.6002 - val_accuracy: 0.7597\n",
      "Epoch 69/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5814 - accuracy: 0.7457 - val_loss: 0.6032 - val_accuracy: 0.7727\n",
      "Epoch 70/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5821 - accuracy: 0.7587 - val_loss: 0.5988 - val_accuracy: 0.7792\n",
      "Epoch 71/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5782 - accuracy: 0.7543 - val_loss: 0.5966 - val_accuracy: 0.7857\n",
      "Epoch 72/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5771 - accuracy: 0.7522 - val_loss: 0.5922 - val_accuracy: 0.7857\n",
      "Epoch 73/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5758 - accuracy: 0.7500 - val_loss: 0.5948 - val_accuracy: 0.7792\n",
      "Epoch 74/100\n",
      "15/15 [==============================] - 0s 1ms/step - loss: 0.5741 - accuracy: 0.7478 - val_loss: 0.5915 - val_accuracy: 0.7792\n",
      "Epoch 75/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5724 - accuracy: 0.7500 - val_loss: 0.5898 - val_accuracy: 0.7857\n",
      "Epoch 76/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5734 - accuracy: 0.7565 - val_loss: 0.5898 - val_accuracy: 0.7792\n",
      "Epoch 77/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5700 - accuracy: 0.7500 - val_loss: 0.5868 - val_accuracy: 0.7792\n",
      "Epoch 78/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5699 - accuracy: 0.7522 - val_loss: 0.5869 - val_accuracy: 0.7662\n",
      "Epoch 79/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5680 - accuracy: 0.7478 - val_loss: 0.5866 - val_accuracy: 0.7662\n",
      "Epoch 80/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5671 - accuracy: 0.7500 - val_loss: 0.5860 - val_accuracy: 0.7662\n",
      "Epoch 81/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5659 - accuracy: 0.7478 - val_loss: 0.5861 - val_accuracy: 0.7662\n",
      "Epoch 82/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5660 - accuracy: 0.7457 - val_loss: 0.5886 - val_accuracy: 0.7727\n",
      "Epoch 83/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5642 - accuracy: 0.7500 - val_loss: 0.5843 - val_accuracy: 0.7662\n",
      "Epoch 84/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5648 - accuracy: 0.7500 - val_loss: 0.5848 - val_accuracy: 0.7662\n",
      "Epoch 85/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5626 - accuracy: 0.7522 - val_loss: 0.5821 - val_accuracy: 0.7727\n",
      "Epoch 86/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5602 - accuracy: 0.7500 - val_loss: 0.5791 - val_accuracy: 0.7792\n",
      "Epoch 87/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5611 - accuracy: 0.7522 - val_loss: 0.5786 - val_accuracy: 0.7792\n",
      "Epoch 88/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5605 - accuracy: 0.7500 - val_loss: 0.5771 - val_accuracy: 0.7922\n",
      "Epoch 89/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5585 - accuracy: 0.7500 - val_loss: 0.5769 - val_accuracy: 0.7792\n",
      "Epoch 90/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5579 - accuracy: 0.7500 - val_loss: 0.5849 - val_accuracy: 0.7662\n",
      "Epoch 91/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5586 - accuracy: 0.7565 - val_loss: 0.5792 - val_accuracy: 0.7857\n",
      "Epoch 92/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5549 - accuracy: 0.7522 - val_loss: 0.5799 - val_accuracy: 0.7857\n",
      "Epoch 93/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5549 - accuracy: 0.7500 - val_loss: 0.5834 - val_accuracy: 0.7792\n",
      "Epoch 94/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5530 - accuracy: 0.7565 - val_loss: 0.5786 - val_accuracy: 0.7857\n",
      "Epoch 95/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5517 - accuracy: 0.7543 - val_loss: 0.5739 - val_accuracy: 0.7792\n",
      "Epoch 96/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5503 - accuracy: 0.7587 - val_loss: 0.5758 - val_accuracy: 0.7857\n",
      "Epoch 97/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5501 - accuracy: 0.7565 - val_loss: 0.5796 - val_accuracy: 0.7857\n",
      "Epoch 98/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5530 - accuracy: 0.7609 - val_loss: 0.5729 - val_accuracy: 0.7727\n",
      "Epoch 99/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5543 - accuracy: 0.7630 - val_loss: 0.5816 - val_accuracy: 0.7857\n",
      "Epoch 100/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5526 - accuracy: 0.7630 - val_loss: 0.5762 - val_accuracy: 0.7857\n"
     ]
    }
   ],
   "source": [
    "model_L1 = Sequential()\n",
    "model_L1.add(Dense(12, input_dim = len(feature_final_training[0, :]), activation = 'relu',kernel_regularizer=l1(0.1)))\n",
    "model_L1.add(Dense(8,activation = 'relu'))\n",
    "model_L1.add(Dense(1, activation = 'sigmoid'))\n",
    "model_L1.compile(loss = 'binary_crossentropy', optimizer = 'adam', metrics =['accuracy'])\n",
    "metrics_L1 = model_L1.fit(x = feature_final_training, y = outcome_final_training, epochs = 100 , validation_data=(feature_valid, outcome_valid),verbose = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d810b70",
   "metadata": {},
   "source": [
    "Evaluating the model one in which we used L1 regularisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "13e73761",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4/4 [==============================] - 0s 667us/step - loss: 0.5469 - accuracy: 0.7609\n",
      "2/2 [==============================] - 0s 999us/step - loss: 0.5762 - accuracy: 0.7857\n",
      "2/2 [==============================] - 0s 1ms/step - loss: 0.6057 - accuracy: 0.7597\n"
     ]
    }
   ],
   "source": [
    "results_train_L1 = model_L1.evaluate(feature_final_training, outcome_final_training, batch_size=128)\n",
    "results_validation_L1 =model_L1.evaluate(feature_valid, outcome_valid, batch_size=128)\n",
    "results_test_L1 = model_L1.evaluate(feature_test, outcome_test, batch_size=128)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "daae0b66",
   "metadata": {},
   "source": [
    "From model in which we used L1 Regualarization,\n",
    "Visualzing Training Acuuracy Vs Validation Accuracy\n",
    "Visualizing Training Loss Vs Validation Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "6304a612",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy and loss of L1 regualisation model\n",
    "\n",
    "# Accuracy\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(metrics_L1.history['accuracy'])\n",
    "plt.plot(metrics_L1.history['val_accuracy'])\n",
    "plt.title('Training and Validation Accuracy')\n",
    "plt.ylabel('Acurray ')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Training Accuracy', 'Validation Accuracy'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "# Loss\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(metrics_L1.history['loss'])\n",
    "plt.plot(metrics_L1.history['val_loss'])\n",
    "plt.title('Training and Validation Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Training Loss', 'Validation Loss'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c4aae14",
   "metadata": {},
   "source": [
    "Visualizing Acuuracy obtained in Training,Validatin,Testing datasets  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "03d05135",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_avail = {'Accuracy_Training':metrics_L1.history['accuracy'][-1]*100, 'Accuracy_Validation':metrics_L1.history['val_accuracy'][-1]*100, 'Accuracy_Test':results_test_L1[-1]*100}\n",
    "split_data_avail = list(data_avail.keys())\n",
    "split_accuracy_cal = list(data_avail.values())\n",
    "  \n",
    "fig = plt.figure(figsize = (10, 5))\n",
    " \n",
    "plt.bar(split_data_avail, split_accuracy_cal,\n",
    "        width = 0.5,color=('green', 'green', 'green'))\n",
    "plt.xlabel(\"Datasets Partition\")\n",
    "plt.ylabel(\"Accuracy %\")\n",
    "plt.title(\"Obtained Accuracy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8346dc27",
   "metadata": {},
   "source": [
    "Now in the second training model we will use L2 Regularisation \n",
    "Hyper parameter are\n",
    "Input layer with 8 features as input\n",
    "1st Hidden layer has 12 nuerons and activation fucntion is relu\n",
    "2nd Hidden layer has 8 nuerons and activation function is relu\n",
    "output layer has 1 nueron because the output is binary\n",
    "No. of epochs are 100\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "a0708708",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "15/15 [==============================] - 0s 7ms/step - loss: 1.6927 - accuracy: 0.3522 - val_loss: 1.6312 - val_accuracy: 0.4610\n",
      "Epoch 2/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.5950 - accuracy: 0.4261 - val_loss: 1.5469 - val_accuracy: 0.5390\n",
      "Epoch 3/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.5052 - accuracy: 0.6761 - val_loss: 1.4699 - val_accuracy: 0.5844\n",
      "Epoch 4/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.4243 - accuracy: 0.6739 - val_loss: 1.4006 - val_accuracy: 0.5649\n",
      "Epoch 5/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.3508 - accuracy: 0.6652 - val_loss: 1.3377 - val_accuracy: 0.5714\n",
      "Epoch 6/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.2836 - accuracy: 0.6630 - val_loss: 1.2808 - val_accuracy: 0.5779\n",
      "Epoch 7/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.2234 - accuracy: 0.6609 - val_loss: 1.2284 - val_accuracy: 0.5779\n",
      "Epoch 8/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.1687 - accuracy: 0.6609 - val_loss: 1.1808 - val_accuracy: 0.5779\n",
      "Epoch 9/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.1182 - accuracy: 0.6609 - val_loss: 1.1361 - val_accuracy: 0.5779\n",
      "Epoch 10/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.0718 - accuracy: 0.6609 - val_loss: 1.0942 - val_accuracy: 0.5779\n",
      "Epoch 11/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 1.0290 - accuracy: 0.6609 - val_loss: 1.0544 - val_accuracy: 0.5844\n",
      "Epoch 12/100\n",
      "15/15 [==============================] - ETA: 0s - loss: 1.0818 - accuracy: 0.50 - 0s 2ms/step - loss: 0.9887 - accuracy: 0.6674 - val_loss: 1.0184 - val_accuracy: 0.5844\n",
      "Epoch 13/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.9517 - accuracy: 0.6696 - val_loss: 0.9841 - val_accuracy: 0.5909\n",
      "Epoch 14/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.9167 - accuracy: 0.6761 - val_loss: 0.9526 - val_accuracy: 0.5974\n",
      "Epoch 15/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.8849 - accuracy: 0.6870 - val_loss: 0.9211 - val_accuracy: 0.5909\n",
      "Epoch 16/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.8553 - accuracy: 0.6891 - val_loss: 0.8943 - val_accuracy: 0.6039\n",
      "Epoch 17/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.8278 - accuracy: 0.7000 - val_loss: 0.8637 - val_accuracy: 0.6364\n",
      "Epoch 18/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.8018 - accuracy: 0.7109 - val_loss: 0.8389 - val_accuracy: 0.6494\n",
      "Epoch 19/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7777 - accuracy: 0.7152 - val_loss: 0.8159 - val_accuracy: 0.6429\n",
      "Epoch 20/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7551 - accuracy: 0.7152 - val_loss: 0.7947 - val_accuracy: 0.6558\n",
      "Epoch 21/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.7325 - accuracy: 0.7217 - val_loss: 0.7703 - val_accuracy: 0.6558\n",
      "Epoch 22/100\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 0.7130 - accuracy: 0.7283 - val_loss: 0.7456 - val_accuracy: 0.6558\n",
      "Epoch 23/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6946 - accuracy: 0.7283 - val_loss: 0.7380 - val_accuracy: 0.6623\n",
      "Epoch 24/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6758 - accuracy: 0.7391 - val_loss: 0.7099 - val_accuracy: 0.6753\n",
      "Epoch 25/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6577 - accuracy: 0.7391 - val_loss: 0.6998 - val_accuracy: 0.6688\n",
      "Epoch 26/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6420 - accuracy: 0.7370 - val_loss: 0.6827 - val_accuracy: 0.6688\n",
      "Epoch 27/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6273 - accuracy: 0.7391 - val_loss: 0.6665 - val_accuracy: 0.6753\n",
      "Epoch 28/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6142 - accuracy: 0.7478 - val_loss: 0.6557 - val_accuracy: 0.6753\n",
      "Epoch 29/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.6016 - accuracy: 0.7457 - val_loss: 0.6462 - val_accuracy: 0.6818\n",
      "Epoch 30/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5924 - accuracy: 0.7391 - val_loss: 0.6390 - val_accuracy: 0.6883\n",
      "Epoch 31/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5820 - accuracy: 0.7478 - val_loss: 0.6285 - val_accuracy: 0.7078\n",
      "Epoch 32/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5717 - accuracy: 0.7413 - val_loss: 0.6127 - val_accuracy: 0.7468\n",
      "Epoch 33/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5640 - accuracy: 0.7478 - val_loss: 0.6149 - val_accuracy: 0.7208\n",
      "Epoch 34/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5566 - accuracy: 0.7522 - val_loss: 0.6036 - val_accuracy: 0.7468\n",
      "Epoch 35/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5506 - accuracy: 0.7543 - val_loss: 0.5958 - val_accuracy: 0.7403\n",
      "Epoch 36/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5450 - accuracy: 0.7500 - val_loss: 0.5902 - val_accuracy: 0.7338\n",
      "Epoch 37/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5391 - accuracy: 0.7522 - val_loss: 0.5911 - val_accuracy: 0.7468\n",
      "Epoch 38/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5333 - accuracy: 0.7522 - val_loss: 0.5788 - val_accuracy: 0.7403\n",
      "Epoch 39/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5300 - accuracy: 0.7609 - val_loss: 0.5768 - val_accuracy: 0.7338\n",
      "Epoch 40/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5246 - accuracy: 0.7630 - val_loss: 0.5710 - val_accuracy: 0.7532\n",
      "Epoch 41/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5216 - accuracy: 0.7652 - val_loss: 0.5740 - val_accuracy: 0.7468\n",
      "Epoch 42/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5189 - accuracy: 0.7609 - val_loss: 0.5758 - val_accuracy: 0.7532\n",
      "Epoch 43/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5152 - accuracy: 0.7630 - val_loss: 0.5582 - val_accuracy: 0.7597\n",
      "Epoch 44/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5138 - accuracy: 0.7587 - val_loss: 0.5665 - val_accuracy: 0.7597\n",
      "Epoch 45/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5114 - accuracy: 0.7696 - val_loss: 0.5525 - val_accuracy: 0.7597\n",
      "Epoch 46/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5059 - accuracy: 0.7696 - val_loss: 0.5609 - val_accuracy: 0.7597\n",
      "Epoch 47/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5035 - accuracy: 0.7630 - val_loss: 0.5534 - val_accuracy: 0.7662\n",
      "Epoch 48/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5023 - accuracy: 0.7652 - val_loss: 0.5564 - val_accuracy: 0.7597\n",
      "Epoch 49/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.5002 - accuracy: 0.7609 - val_loss: 0.5572 - val_accuracy: 0.7597\n",
      "Epoch 50/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4971 - accuracy: 0.7717 - val_loss: 0.5466 - val_accuracy: 0.7727\n",
      "Epoch 51/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4957 - accuracy: 0.7739 - val_loss: 0.5493 - val_accuracy: 0.7727\n",
      "Epoch 52/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4933 - accuracy: 0.7717 - val_loss: 0.5489 - val_accuracy: 0.7727\n",
      "Epoch 53/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4935 - accuracy: 0.7652 - val_loss: 0.5465 - val_accuracy: 0.7727\n",
      "Epoch 54/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4911 - accuracy: 0.7739 - val_loss: 0.5531 - val_accuracy: 0.7662\n",
      "Epoch 55/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4892 - accuracy: 0.7696 - val_loss: 0.5409 - val_accuracy: 0.7662\n",
      "Epoch 56/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4883 - accuracy: 0.7696 - val_loss: 0.5458 - val_accuracy: 0.7727\n",
      "Epoch 57/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4876 - accuracy: 0.7652 - val_loss: 0.5389 - val_accuracy: 0.7662\n",
      "Epoch 58/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4878 - accuracy: 0.7674 - val_loss: 0.5405 - val_accuracy: 0.7662\n",
      "Epoch 59/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4897 - accuracy: 0.7587 - val_loss: 0.5419 - val_accuracy: 0.7727\n",
      "Epoch 60/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4892 - accuracy: 0.7739 - val_loss: 0.5385 - val_accuracy: 0.7662\n",
      "Epoch 61/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4865 - accuracy: 0.7674 - val_loss: 0.5421 - val_accuracy: 0.7792\n",
      "Epoch 62/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4849 - accuracy: 0.7609 - val_loss: 0.5368 - val_accuracy: 0.7727\n",
      "Epoch 63/100\n",
      "15/15 [==============================] - 0s 1ms/step - loss: 0.4822 - accuracy: 0.7696 - val_loss: 0.5395 - val_accuracy: 0.7792\n",
      "Epoch 64/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4804 - accuracy: 0.7761 - val_loss: 0.5412 - val_accuracy: 0.7792\n",
      "Epoch 65/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4786 - accuracy: 0.7674 - val_loss: 0.5396 - val_accuracy: 0.7727\n",
      "Epoch 66/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4784 - accuracy: 0.7674 - val_loss: 0.5396 - val_accuracy: 0.7792\n",
      "Epoch 67/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4802 - accuracy: 0.7739 - val_loss: 0.5360 - val_accuracy: 0.7727\n",
      "Epoch 68/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4784 - accuracy: 0.7652 - val_loss: 0.5327 - val_accuracy: 0.7662\n",
      "Epoch 69/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4760 - accuracy: 0.7761 - val_loss: 0.5331 - val_accuracy: 0.7662\n",
      "Epoch 70/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4752 - accuracy: 0.7717 - val_loss: 0.5297 - val_accuracy: 0.7727\n",
      "Epoch 71/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4797 - accuracy: 0.7739 - val_loss: 0.5306 - val_accuracy: 0.7662\n",
      "Epoch 72/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4747 - accuracy: 0.7652 - val_loss: 0.5291 - val_accuracy: 0.7727\n",
      "Epoch 73/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4760 - accuracy: 0.7783 - val_loss: 0.5339 - val_accuracy: 0.7792\n",
      "Epoch 74/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4734 - accuracy: 0.7696 - val_loss: 0.5312 - val_accuracy: 0.7727\n",
      "Epoch 75/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4722 - accuracy: 0.7739 - val_loss: 0.5241 - val_accuracy: 0.7727\n",
      "Epoch 76/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4726 - accuracy: 0.7739 - val_loss: 0.5293 - val_accuracy: 0.7727\n",
      "Epoch 77/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4720 - accuracy: 0.7717 - val_loss: 0.5352 - val_accuracy: 0.7792\n",
      "Epoch 78/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4702 - accuracy: 0.7717 - val_loss: 0.5287 - val_accuracy: 0.7662\n",
      "Epoch 79/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4696 - accuracy: 0.7739 - val_loss: 0.5408 - val_accuracy: 0.7792\n",
      "Epoch 80/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4711 - accuracy: 0.7717 - val_loss: 0.5269 - val_accuracy: 0.7662\n",
      "Epoch 81/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4695 - accuracy: 0.7804 - val_loss: 0.5297 - val_accuracy: 0.7727\n",
      "Epoch 82/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4691 - accuracy: 0.7717 - val_loss: 0.5348 - val_accuracy: 0.7792\n",
      "Epoch 83/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4698 - accuracy: 0.7761 - val_loss: 0.5297 - val_accuracy: 0.7662\n",
      "Epoch 84/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4691 - accuracy: 0.7739 - val_loss: 0.5344 - val_accuracy: 0.7792\n",
      "Epoch 85/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4687 - accuracy: 0.7652 - val_loss: 0.5313 - val_accuracy: 0.7727\n",
      "Epoch 86/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4667 - accuracy: 0.7739 - val_loss: 0.5284 - val_accuracy: 0.7727\n",
      "Epoch 87/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4669 - accuracy: 0.7783 - val_loss: 0.5288 - val_accuracy: 0.7727\n",
      "Epoch 88/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4658 - accuracy: 0.7826 - val_loss: 0.5260 - val_accuracy: 0.7727\n",
      "Epoch 89/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4708 - accuracy: 0.7674 - val_loss: 0.5228 - val_accuracy: 0.7727\n",
      "Epoch 90/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4674 - accuracy: 0.7761 - val_loss: 0.5249 - val_accuracy: 0.7662\n",
      "Epoch 91/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4662 - accuracy: 0.7761 - val_loss: 0.5243 - val_accuracy: 0.7727\n",
      "Epoch 92/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4665 - accuracy: 0.7717 - val_loss: 0.5314 - val_accuracy: 0.7792\n",
      "Epoch 93/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4643 - accuracy: 0.7804 - val_loss: 0.5196 - val_accuracy: 0.7727\n",
      "Epoch 94/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4645 - accuracy: 0.7696 - val_loss: 0.5251 - val_accuracy: 0.7662\n",
      "Epoch 95/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4627 - accuracy: 0.7804 - val_loss: 0.5322 - val_accuracy: 0.7792\n",
      "Epoch 96/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4656 - accuracy: 0.7739 - val_loss: 0.5292 - val_accuracy: 0.7792\n",
      "Epoch 97/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4629 - accuracy: 0.7783 - val_loss: 0.5267 - val_accuracy: 0.7727\n",
      "Epoch 98/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4625 - accuracy: 0.7804 - val_loss: 0.5266 - val_accuracy: 0.7727\n",
      "Epoch 99/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4638 - accuracy: 0.7674 - val_loss: 0.5250 - val_accuracy: 0.7792\n",
      "Epoch 100/100\n",
      "15/15 [==============================] - 0s 2ms/step - loss: 0.4620 - accuracy: 0.7717 - val_loss: 0.5270 - val_accuracy: 0.7727\n"
     ]
    }
   ],
   "source": [
    "model_L2 = Sequential()\n",
    "model_L2.add(Dense(12, input_dim = len(feature_final_training[0, :]), activation = 'relu',kernel_regularizer=l2(0.1)))\n",
    "model_L2.add(Dense(8,activation = 'relu'))\n",
    "model_L2.add(Dense(1, activation = 'sigmoid'))\n",
    "model_L2.compile(loss = 'binary_crossentropy', optimizer = 'adam', metrics =['accuracy'])\n",
    "metrics_L2 = model_L2.fit(x = feature_final_training, y = outcome_final_training, epochs = 100 , validation_data=(feature_valid, outcome_valid),verbose = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "01826bb2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "58/58 [==============================] - 0s 511us/step - loss: 0.4595 - accuracy: 0.7848\n",
      "20/20 [==============================] - 0s 877us/step - loss: 0.5270 - accuracy: 0.7727\n",
      "20/20 [==============================] - 0s 735us/step - loss: 0.5524 - accuracy: 0.7662\n"
     ]
    }
   ],
   "source": [
    "results_train_L2 = model_L2.evaluate(feature_final_training, outcome_final_training, batch_size=8)\n",
    "results_validation_L2 =model_L2.evaluate(feature_valid, outcome_valid, batch_size=8)\n",
    "results_test_L2 = model_L2.evaluate(feature_test, outcome_test, batch_size=8)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fad59a12",
   "metadata": {},
   "source": [
    "From model in which we used L2 Regualarization\n",
    "Visualzing Training Acuuracy Vs Validation Accuracy\n",
    "Visualizing Training Loss Vs Validation Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "354e536e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy and loss.\n",
    "\n",
    "# Accuracy\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(metrics_L2.history['accuracy'])\n",
    "plt.plot(metrics_L2.history['val_accuracy'])\n",
    "plt.title('Training and Validation Accurracy')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Training Accuracy', 'Validation Accuracy'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "# Loss\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(metrics_L2.history['loss'])\n",
    "plt.plot(metrics_L2.history['val_loss'])\n",
    "plt.title('Training and Validation Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Training Loss', 'Validation Loss'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b58ee5c4",
   "metadata": {},
   "source": [
    "Visualizing Acuuracy obtained in Training,Validatin,Testing datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "263124ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_avail = {'Accuracy_Training':metrics_L2.history['accuracy'][-1]*100, 'Accuracy_Validation':metrics_L2.history['val_accuracy'][-1]*100, 'Accuracy_Test':results_test_L2[-1]*100}\n",
    "split_data_avail = list(data_avail.keys())\n",
    "split_accuracy_cal = list(data_avail.values())\n",
    "  \n",
    "fig = plt.figure(figsize = (10, 5))\n",
    " \n",
    "plt.bar(split_data_avail, split_accuracy_cal,\n",
    "        width = 0.5,color=('green', 'green', 'green'))\n",
    "plt.xlabel(\"Datasets Partition\")\n",
    "plt.ylabel(\"Accuracy %\")\n",
    "plt.title(\"Obtained Accuracy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a4ddcf8",
   "metadata": {},
   "source": [
    "--------BONOUS PART-----------\n",
    "Building a Neural Network Model for Logistic Regression Classification which uses Dropout regularization"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f77a29e9",
   "metadata": {},
   "source": [
    "We have 1 inputvlayer which passes 8 feature values as inputs\n",
    "1st hidden layer with 12 nuerons and activation fucnction is relu\n",
    "we have dropout in 2nd hidden layer where we will drop 20% nuerons of it's previous hidden layer\n",
    "we have 3rd hidden layer  with 8 nuerons and activation function is relu\n",
    "we have output layer with 1 nueron\n",
    "No of epochs are 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "29dec849",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "0e752b3f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "58/58 [==============================] - 0s 2ms/step - loss: 0.6978 - accuracy: 0.5522 - val_loss: 0.6766 - val_accuracy: 0.5844\n",
      "Epoch 2/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.6585 - accuracy: 0.6652 - val_loss: 0.6666 - val_accuracy: 0.5974\n",
      "Epoch 3/100\n",
      "58/58 [==============================] - 0s 989us/step - loss: 0.6368 - accuracy: 0.6804 - val_loss: 0.6586 - val_accuracy: 0.5844\n",
      "Epoch 4/100\n",
      "58/58 [==============================] - 0s 912us/step - loss: 0.6186 - accuracy: 0.6717 - val_loss: 0.6501 - val_accuracy: 0.5844\n",
      "Epoch 5/100\n",
      "58/58 [==============================] - 0s 964us/step - loss: 0.6034 - accuracy: 0.6913 - val_loss: 0.6390 - val_accuracy: 0.5974\n",
      "Epoch 6/100\n",
      "58/58 [==============================] - 0s 918us/step - loss: 0.5895 - accuracy: 0.6935 - val_loss: 0.6242 - val_accuracy: 0.6234\n",
      "Epoch 7/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.5729 - accuracy: 0.7109 - val_loss: 0.5974 - val_accuracy: 0.6688\n",
      "Epoch 8/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.5454 - accuracy: 0.7283 - val_loss: 0.5731 - val_accuracy: 0.7078\n",
      "Epoch 9/100\n",
      "58/58 [==============================] - 0s 756us/step - loss: 0.5103 - accuracy: 0.7500 - val_loss: 0.5433 - val_accuracy: 0.7597\n",
      "Epoch 10/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.5168 - accuracy: 0.7522 - val_loss: 0.5545 - val_accuracy: 0.7532\n",
      "Epoch 11/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.5169 - accuracy: 0.7457 - val_loss: 0.5421 - val_accuracy: 0.7597\n",
      "Epoch 12/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4984 - accuracy: 0.7565 - val_loss: 0.5387 - val_accuracy: 0.7597\n",
      "Epoch 13/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.5085 - accuracy: 0.7457 - val_loss: 0.5348 - val_accuracy: 0.7662\n",
      "Epoch 14/100\n",
      "58/58 [==============================] - 0s 916us/step - loss: 0.4949 - accuracy: 0.7761 - val_loss: 0.5231 - val_accuracy: 0.7792\n",
      "Epoch 15/100\n",
      "58/58 [==============================] - 0s 895us/step - loss: 0.4810 - accuracy: 0.7500 - val_loss: 0.5113 - val_accuracy: 0.7792\n",
      "Epoch 16/100\n",
      "58/58 [==============================] - 0s 962us/step - loss: 0.4749 - accuracy: 0.7739 - val_loss: 0.5103 - val_accuracy: 0.7792\n",
      "Epoch 17/100\n",
      "58/58 [==============================] - 0s 913us/step - loss: 0.4909 - accuracy: 0.7674 - val_loss: 0.5146 - val_accuracy: 0.7857\n",
      "Epoch 18/100\n",
      "58/58 [==============================] - 0s 908us/step - loss: 0.4739 - accuracy: 0.7652 - val_loss: 0.5084 - val_accuracy: 0.7922\n",
      "Epoch 19/100\n",
      "58/58 [==============================] - 0s 942us/step - loss: 0.4732 - accuracy: 0.7652 - val_loss: 0.5110 - val_accuracy: 0.7857\n",
      "Epoch 20/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4607 - accuracy: 0.7804 - val_loss: 0.4996 - val_accuracy: 0.7922\n",
      "Epoch 21/100\n",
      "58/58 [==============================] - 0s 934us/step - loss: 0.4508 - accuracy: 0.7848 - val_loss: 0.4999 - val_accuracy: 0.7922\n",
      "Epoch 22/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4502 - accuracy: 0.8000 - val_loss: 0.5080 - val_accuracy: 0.7922\n",
      "Epoch 23/100\n",
      "58/58 [==============================] - 0s 929us/step - loss: 0.4745 - accuracy: 0.7565 - val_loss: 0.5005 - val_accuracy: 0.7922\n",
      "Epoch 24/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4544 - accuracy: 0.7739 - val_loss: 0.5088 - val_accuracy: 0.7922\n",
      "Epoch 25/100\n",
      "58/58 [==============================] - 0s 997us/step - loss: 0.4535 - accuracy: 0.7826 - val_loss: 0.4998 - val_accuracy: 0.7922\n",
      "Epoch 26/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4490 - accuracy: 0.7739 - val_loss: 0.4984 - val_accuracy: 0.7922\n",
      "Epoch 27/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4744 - accuracy: 0.7435 - val_loss: 0.5061 - val_accuracy: 0.7857\n",
      "Epoch 28/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4582 - accuracy: 0.7674 - val_loss: 0.4976 - val_accuracy: 0.7987\n",
      "Epoch 29/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4484 - accuracy: 0.7739 - val_loss: 0.5087 - val_accuracy: 0.7792\n",
      "Epoch 30/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4398 - accuracy: 0.7739 - val_loss: 0.4991 - val_accuracy: 0.7922\n",
      "Epoch 31/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4478 - accuracy: 0.7804 - val_loss: 0.4981 - val_accuracy: 0.7922\n",
      "Epoch 32/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4489 - accuracy: 0.7761 - val_loss: 0.4971 - val_accuracy: 0.7987\n",
      "Epoch 33/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4559 - accuracy: 0.7783 - val_loss: 0.4984 - val_accuracy: 0.7987\n",
      "Epoch 34/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4679 - accuracy: 0.7717 - val_loss: 0.5068 - val_accuracy: 0.7857\n",
      "Epoch 35/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4536 - accuracy: 0.7696 - val_loss: 0.5059 - val_accuracy: 0.7857\n",
      "Epoch 36/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4370 - accuracy: 0.8109 - val_loss: 0.5006 - val_accuracy: 0.7857\n",
      "Epoch 37/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4574 - accuracy: 0.7652 - val_loss: 0.5094 - val_accuracy: 0.7792\n",
      "Epoch 38/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4489 - accuracy: 0.7674 - val_loss: 0.5005 - val_accuracy: 0.7922\n",
      "Epoch 39/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4398 - accuracy: 0.8109 - val_loss: 0.5039 - val_accuracy: 0.7857\n",
      "Epoch 40/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4455 - accuracy: 0.8022 - val_loss: 0.5103 - val_accuracy: 0.7792\n",
      "Epoch 41/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4404 - accuracy: 0.7652 - val_loss: 0.5077 - val_accuracy: 0.7857\n",
      "Epoch 42/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4374 - accuracy: 0.7717 - val_loss: 0.5044 - val_accuracy: 0.7792\n",
      "Epoch 43/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4263 - accuracy: 0.8043 - val_loss: 0.5100 - val_accuracy: 0.7857\n",
      "Epoch 44/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4304 - accuracy: 0.7783 - val_loss: 0.5123 - val_accuracy: 0.7857\n",
      "Epoch 45/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4202 - accuracy: 0.7891 - val_loss: 0.5081 - val_accuracy: 0.7792\n",
      "Epoch 46/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4305 - accuracy: 0.7935 - val_loss: 0.5113 - val_accuracy: 0.7792\n",
      "Epoch 47/100\n",
      "58/58 [==============================] - 0s 974us/step - loss: 0.4385 - accuracy: 0.7935 - val_loss: 0.5103 - val_accuracy: 0.7662\n",
      "Epoch 48/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4335 - accuracy: 0.7913 - val_loss: 0.5159 - val_accuracy: 0.7727\n",
      "Epoch 49/100\n",
      "58/58 [==============================] - 0s 954us/step - loss: 0.4291 - accuracy: 0.7935 - val_loss: 0.5111 - val_accuracy: 0.7662\n",
      "Epoch 50/100\n",
      "58/58 [==============================] - 0s 959us/step - loss: 0.4255 - accuracy: 0.7891 - val_loss: 0.5085 - val_accuracy: 0.7662\n",
      "Epoch 51/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4268 - accuracy: 0.8087 - val_loss: 0.5153 - val_accuracy: 0.7727\n",
      "Epoch 52/100\n",
      "58/58 [==============================] - 0s 989us/step - loss: 0.4178 - accuracy: 0.8022 - val_loss: 0.5121 - val_accuracy: 0.7662\n",
      "Epoch 53/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4060 - accuracy: 0.7978 - val_loss: 0.5195 - val_accuracy: 0.7662\n",
      "Epoch 54/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4152 - accuracy: 0.7978 - val_loss: 0.5255 - val_accuracy: 0.7727\n",
      "Epoch 55/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4277 - accuracy: 0.7978 - val_loss: 0.5118 - val_accuracy: 0.7597\n",
      "Epoch 56/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4291 - accuracy: 0.7935 - val_loss: 0.5158 - val_accuracy: 0.7597\n",
      "Epoch 57/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4333 - accuracy: 0.7913 - val_loss: 0.5108 - val_accuracy: 0.7727\n",
      "Epoch 58/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4257 - accuracy: 0.7848 - val_loss: 0.5212 - val_accuracy: 0.7597\n",
      "Epoch 59/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4158 - accuracy: 0.7891 - val_loss: 0.5179 - val_accuracy: 0.7532\n",
      "Epoch 60/100\n",
      "58/58 [==============================] - 0s 911us/step - loss: 0.4133 - accuracy: 0.8000 - val_loss: 0.5111 - val_accuracy: 0.7662\n",
      "Epoch 61/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4336 - accuracy: 0.8043 - val_loss: 0.5173 - val_accuracy: 0.7597\n",
      "Epoch 62/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4179 - accuracy: 0.8043 - val_loss: 0.5187 - val_accuracy: 0.7662\n",
      "Epoch 63/100\n",
      "58/58 [==============================] - 0s 906us/step - loss: 0.4121 - accuracy: 0.7913 - val_loss: 0.5181 - val_accuracy: 0.7727\n",
      "Epoch 64/100\n",
      "58/58 [==============================] - 0s 907us/step - loss: 0.4095 - accuracy: 0.8196 - val_loss: 0.5259 - val_accuracy: 0.7597\n",
      "Epoch 65/100\n",
      "58/58 [==============================] - 0s 916us/step - loss: 0.4094 - accuracy: 0.7913 - val_loss: 0.5255 - val_accuracy: 0.7727\n",
      "Epoch 66/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4308 - accuracy: 0.7826 - val_loss: 0.5331 - val_accuracy: 0.7597\n",
      "Epoch 67/100\n",
      "58/58 [==============================] - 0s 939us/step - loss: 0.4070 - accuracy: 0.7978 - val_loss: 0.5224 - val_accuracy: 0.7662\n",
      "Epoch 68/100\n",
      "58/58 [==============================] - 0s 966us/step - loss: 0.4172 - accuracy: 0.7978 - val_loss: 0.5394 - val_accuracy: 0.7532\n",
      "Epoch 69/100\n",
      "58/58 [==============================] - 0s 740us/step - loss: 0.4143 - accuracy: 0.8000 - val_loss: 0.5323 - val_accuracy: 0.7597\n",
      "Epoch 70/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4210 - accuracy: 0.7935 - val_loss: 0.5251 - val_accuracy: 0.7662\n",
      "Epoch 71/100\n",
      "58/58 [==============================] - 0s 886us/step - loss: 0.4118 - accuracy: 0.8000 - val_loss: 0.5322 - val_accuracy: 0.7532\n",
      "Epoch 72/100\n",
      "58/58 [==============================] - 0s 830us/step - loss: 0.4215 - accuracy: 0.7957 - val_loss: 0.5308 - val_accuracy: 0.7597\n",
      "Epoch 73/100\n",
      "58/58 [==============================] - 0s 934us/step - loss: 0.4102 - accuracy: 0.8087 - val_loss: 0.5301 - val_accuracy: 0.7597\n",
      "Epoch 74/100\n",
      "58/58 [==============================] - 0s 867us/step - loss: 0.3988 - accuracy: 0.8065 - val_loss: 0.5315 - val_accuracy: 0.7597\n",
      "Epoch 75/100\n",
      "58/58 [==============================] - 0s 818us/step - loss: 0.4166 - accuracy: 0.8130 - val_loss: 0.5406 - val_accuracy: 0.7532\n",
      "Epoch 76/100\n",
      "58/58 [==============================] - 0s 822us/step - loss: 0.4099 - accuracy: 0.7957 - val_loss: 0.5388 - val_accuracy: 0.7532\n",
      "Epoch 77/100\n",
      "58/58 [==============================] - 0s 843us/step - loss: 0.4057 - accuracy: 0.7978 - val_loss: 0.5412 - val_accuracy: 0.7532\n",
      "Epoch 78/100\n",
      "58/58 [==============================] - 0s 828us/step - loss: 0.3981 - accuracy: 0.7957 - val_loss: 0.5276 - val_accuracy: 0.7532\n",
      "Epoch 79/100\n",
      "58/58 [==============================] - 0s 845us/step - loss: 0.4094 - accuracy: 0.8239 - val_loss: 0.5387 - val_accuracy: 0.7403\n",
      "Epoch 80/100\n",
      "58/58 [==============================] - 0s 843us/step - loss: 0.4051 - accuracy: 0.7978 - val_loss: 0.5381 - val_accuracy: 0.7468\n",
      "Epoch 81/100\n",
      "58/58 [==============================] - 0s 841us/step - loss: 0.3949 - accuracy: 0.8130 - val_loss: 0.5490 - val_accuracy: 0.7403\n",
      "Epoch 82/100\n",
      "58/58 [==============================] - 0s 864us/step - loss: 0.4125 - accuracy: 0.8022 - val_loss: 0.5365 - val_accuracy: 0.7597\n",
      "Epoch 83/100\n",
      "58/58 [==============================] - 0s 854us/step - loss: 0.4072 - accuracy: 0.8087 - val_loss: 0.5415 - val_accuracy: 0.7468\n",
      "Epoch 84/100\n",
      "58/58 [==============================] - 0s 824us/step - loss: 0.3758 - accuracy: 0.8196 - val_loss: 0.5413 - val_accuracy: 0.7597\n",
      "Epoch 85/100\n",
      "58/58 [==============================] - 0s 897us/step - loss: 0.4105 - accuracy: 0.7870 - val_loss: 0.5356 - val_accuracy: 0.7597\n",
      "Epoch 86/100\n",
      "58/58 [==============================] - 0s 915us/step - loss: 0.4063 - accuracy: 0.8109 - val_loss: 0.5353 - val_accuracy: 0.7662\n",
      "Epoch 87/100\n",
      "58/58 [==============================] - 0s 848us/step - loss: 0.3967 - accuracy: 0.8109 - val_loss: 0.5375 - val_accuracy: 0.7727\n",
      "Epoch 88/100\n",
      "58/58 [==============================] - 0s 897us/step - loss: 0.4026 - accuracy: 0.8109 - val_loss: 0.5425 - val_accuracy: 0.7727\n",
      "Epoch 89/100\n",
      "58/58 [==============================] - 0s 847us/step - loss: 0.4017 - accuracy: 0.8022 - val_loss: 0.5363 - val_accuracy: 0.7597\n",
      "Epoch 90/100\n",
      "58/58 [==============================] - 0s 874us/step - loss: 0.4051 - accuracy: 0.7957 - val_loss: 0.5443 - val_accuracy: 0.7597\n",
      "Epoch 91/100\n",
      "58/58 [==============================] - 0s 863us/step - loss: 0.4019 - accuracy: 0.8000 - val_loss: 0.5463 - val_accuracy: 0.7532\n",
      "Epoch 92/100\n",
      "58/58 [==============================] - 0s 955us/step - loss: 0.4201 - accuracy: 0.8000 - val_loss: 0.5483 - val_accuracy: 0.7532\n",
      "Epoch 93/100\n",
      "58/58 [==============================] - 0s 849us/step - loss: 0.4082 - accuracy: 0.7870 - val_loss: 0.5437 - val_accuracy: 0.7532\n",
      "Epoch 94/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4050 - accuracy: 0.7848 - val_loss: 0.5560 - val_accuracy: 0.7597\n",
      "Epoch 95/100\n",
      "58/58 [==============================] - 0s 826us/step - loss: 0.4076 - accuracy: 0.7957 - val_loss: 0.5486 - val_accuracy: 0.7597\n",
      "Epoch 96/100\n",
      "58/58 [==============================] - 0s 879us/step - loss: 0.3986 - accuracy: 0.7935 - val_loss: 0.5495 - val_accuracy: 0.7597\n",
      "Epoch 97/100\n",
      "58/58 [==============================] - 0s 912us/step - loss: 0.4146 - accuracy: 0.7978 - val_loss: 0.5495 - val_accuracy: 0.7532\n",
      "Epoch 98/100\n",
      "58/58 [==============================] - 0s 1ms/step - loss: 0.4201 - accuracy: 0.7935 - val_loss: 0.5594 - val_accuracy: 0.7403\n",
      "Epoch 99/100\n",
      "58/58 [==============================] - 0s 839us/step - loss: 0.4014 - accuracy: 0.7957 - val_loss: 0.5549 - val_accuracy: 0.7597\n",
      "Epoch 100/100\n",
      "58/58 [==============================] - 0s 808us/step - loss: 0.4024 - accuracy: 0.8022 - val_loss: 0.5555 - val_accuracy: 0.7532\n"
     ]
    }
   ],
   "source": [
    "model_Dropout = Sequential()\n",
    "model_Dropout.add(Dense(12, input_dim = len(feature_final_training[0, :]), activation = 'relu'))\n",
    "model_Dropout.add(Dropout(0.2))\n",
    "model_Dropout.add(Dense(8,activation = 'relu'))\n",
    "model_Dropout.add(Dense(1, activation = 'sigmoid'))\n",
    "model_Dropout.compile(loss = 'binary_crossentropy', optimizer = 'adam', metrics =['accuracy'])\n",
    "metrics_Dropout = model_Dropout.fit(x = feature_final_training, y = outcome_final_training, epochs = 100, validation_data=(feature_valid, outcome_valid),batch_size = 8,verbose = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0bd0b62",
   "metadata": {},
   "source": [
    "Evaluating the model in which we used Dropout regularization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "9fea90f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4/4 [==============================] - 0s 985us/step - loss: 0.5469 - accuracy: 0.7609\n",
      "2/2 [==============================] - 0s 1ms/step - loss: 0.5762 - accuracy: 0.7857\n",
      "2/2 [==============================] - 0s 1ms/step - loss: 0.6057 - accuracy: 0.7597\n"
     ]
    }
   ],
   "source": [
    "results_train_Dropout = model_L1.evaluate(feature_final_training, outcome_final_training, batch_size=128)\n",
    "results_validation_Dropout =model_L1.evaluate(feature_valid, outcome_valid, batch_size=128)\n",
    "results_test_Dropout = model_L1.evaluate(feature_test, outcome_test, batch_size=128)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c891383b",
   "metadata": {},
   "source": [
    "From model in which we used Dropout Regualarization\n",
    "Visualzing Training Acuuracy Vs Validation Accuracy\n",
    "Visualizing Training Loss Vs Validation Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "cc5b0f44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the accuracy and loss.\n",
    "\n",
    "# Accuracy\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(metrics_L2.history['accuracy'])\n",
    "plt.plot(metrics_L2.history['val_accuracy'])\n",
    "plt.title('Training and Validation acurracy')\n",
    "plt.ylabel('Error')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Training Accuracy', 'Validation Accuracy'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "# Loss\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(metrics_L2.history['loss'])\n",
    "plt.plot(metrics_L2.history['val_loss'])\n",
    "plt.title('Training and Validation Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Training Loss', 'Validation Loss'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62dc48c1",
   "metadata": {},
   "source": [
    "Visualizing Acuuracy obtained in Training,Validatin,Testing datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "fe77f237",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl4AAAFOCAYAAABXHxOAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAik0lEQVR4nO3dd7gmZX3/8fcHFqSDyIo0WQ0olgSUlUhsKBjRqKASEAuLcmU1MRpRoyQ//YnG5KeJJQZjWRurIqAgYgsRNyDGgiy9GhBBIJQFQZog5fv7Y+6jj+spz8I+cwrv13Wd65lyz8x3DmfYz7nnPjOpKiRJkjR6a0x3AZIkSQ8UBi9JkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JN0vSQ5N8oXVuL+PJ3nH6trfSvuuJNuNYt+SNAyDl6RJJTkwyblJbk9yTZKPJdlkFba/LMkew7avqtdW1T/cp2JXgySHJ7k7yRbTVYOkucvgJWlCSd4MvA/4W2Bj4MnAtsCJSdaeztpGIcn6wEuAXwKv6PnY8/o8nqTpYfCSNK4kGwHvAl5fVSdU1V1VdRmwL7CA3w0m6yQ5OsktSc5IsmPbx+eBhwNfT3Jrkre25V9uvWe/THJKkscNHPfwJO9p07sluTLJm5Ncl+TqJK8aaPugJO9P8vMk17bblOsOrP/bts3/Jnn1EKf9EuAm4N3AopW+H5sm+Wzb141Jvjqwbq8kZyW5OclPk+zZlv9Ob9/gbdkkC9qtz4OS/Bz4ryG+N+sm+UCSy9v6/27Lvpnk9SvVe06SFw1xzpJ6ZPCSNJE/AdYBvjK4sKpuBb4FPHtg8V7Al4FNgS8CX02yVlW9Evg58IKq2qCq/rm1/w9ge+ChwBnAEZPU8TC63ratgIOAf0/y4LbuvcCjgJ2A7Vqb/wvQws9bWp3bA8Pc7lwEHAkcBeyQZOeBdZ8H1gMe1+r+UDvOLsDn6HoFNwGeDlw2xLHGPAN4DPCcNj/Z9+b9wM50/202Bd4K3AssZSAIt+C7FfDNVahDUg8MXpImshlwfVXdPc66q9v6MadX1TFVdRfwQbrA9uSJdlxVn6mqW6rqTuBQYMckG0/Q/C7g3a3H7VvArcCjkwRYDBxcVb+oqluAfwJe2rbbF/hsVZ1XVbe140woycOBZwJfrKprgWXAAW3dFsBzgddW1Y2tlu+2TQ8CPlNVJ1bVvVV1VVVdNNmxVnJoVd1WVb+a7HuTZA3g1cDftGPcU1U/aO2+BjwqyfZtn68Ejq6qX69CHZJ6YPCSNJHrgc0mGHu0RVs/5oqxiaq6F7gS2HK8nSZZM8l72y25m/lt79Bm47UHblgp/N0ObADMp+uBOj3JTUluAk5oy2nHv2Jgu8sn2P+YVwIXVtVZbf4I4GVJ1gK2AX5RVTeOs902wE+n2PdkflPjFN+bzegC7e8dq6ruAI4GXtEC2v50PXSSZhiDl6SJ/BC4E3jx4MIkG9D1/iwbWLzNwPo1gK2B/22LaqX9vozu1uQedLcQF4xtuor1XQ/8CnhcVW3Svjauqg3a+qsH66IbazaZA4BHtvFV19D13G0GPI8uHG06wV9zXgH8wQT7vI0uHI552DhtBr8/k31vrgfumORYS4GXA7sDt1fVDydoJ2kaGbwkjauqfkk3uP6wJHsmWSvJAuBLdD1agz0qOyd5cesdeyNdYPtRW3ct8MiBthu29TfQhZJ/uo/13Qt8EvhQkocCJNkqydhYqS8BByZ5bJL1gHdOtK8ku9IFml3oxovtBDyebrzaAVV1Nd3Yq48meXD7Xjy9bf5p4FVJdk+yRqthh7buLOClrf1CYJ8pTmvC7007388AH0yyZesd2zXJg9r6H9KN9/oA9nZJM5bBS9KE2mD4v6cb1H0zcCpdD8/ubWzRmOOB/YAb6W7ZvbiN9wL4f8Db2+3At9ANRL8cuAq4gN8GtPvibcAlwI/arbnvAI9utf8H8K90fy14SfucyCLg+Ko6t6quGfsCPgw8P8mm7bzuAi4CrqMLmFTVj4FX0Q22/yXwXbpHbgC8gy7Q3UgXYr84xflM9b15C3AucBrwC7pHfayx0vZ/CKy2B9pKWr1StfJdAEnSbJTkAGBxVT11umuRND57vCRpDmi3U/8KWDLdtUiamMFLkma5Nq5tBd14uqluZ0qaRt5qlCRJ6ok9XpIkST0xeEmSJPVkvCdSzzibbbZZLViwYLrLkCRJmtLpp59+fVXNH2/drAheCxYsYPny5dNdhiRJ0pSSTPiKMm81SpIk9cTgJUmS1BODlyRJUk8MXpIkST0xeEmSJPXE4CVJktQTg5ckSVJPDF6SJEk9GWnwSnJwkvOTnJfkyCTrJHlEklOTXJLk6CRrj7IGSZKkmWJkwSvJVsAbgIVV9XhgTeClwPuAD1XVdsCNwEGjqkGSJGkmGfWtxnnAuknmAesBVwPPAo5p65cCe4+4BkmSpBlhZO9qrKqrkrwf+DnwK+DbwOnATVV1d2t2JbDVqGpYFXlXprsEjaPeWdNdgiRJq80obzU+GNgLeASwJbA+sOcqbL84yfIky1esWDGiKiVJkvozyluNewA/q6oVVXUX8BXgKcAm7dYjwNbAVeNtXFVLqmphVS2cP3/+CMuUJEnqxyiD18+BJydZL0mA3YELgJOAfVqbRcDxI6xBkiRpxhhZ8KqqU+kG0Z8BnNuOtQR4G/CmJJcADwE+PaoaJEmSZpKRDa4HqKp3Au9cafGlwC6jPK4kSdJM5JPrJUmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKknBi9JkqSejPQBqpIkPVDlXZnuErSSemdNdwn2eEmSJPXFHi9Jq8Tf4memmfCbvKSp2eMlSZLUE4OXJElSTwxekiRJPTF4SZIk9cTgJUmS1BODlyRJUk8MXpIkST0xeEmSJPXE4CVJktQTg5ckSVJPDF6SJEk9MXhJkiT1ZGTBK8mjk5w18HVzkjcm2TTJiUkubp8PHlUNkiRJM8nIgldV/aSqdqqqnYCdgduB44BDgGVVtT2wrM1LkiTNeX3datwd+GlVXQ7sBSxty5cCe/dUgyRJ0rTqK3i9FDiyTW9eVVe36WuAzXuqQZIkaVqNPHglWRt4IfDllddVVQE1wXaLkyxPsnzFihUjrlKSJGn0+ujxei5wRlVd2+avTbIFQPu8bryNqmpJVS2sqoXz58/voUxJkqTR6iN47c9vbzMCfA1Y1KYXAcf3UIMkSdK0G2nwSrI+8GzgKwOL3ws8O8nFwB5tXpIkac6bN8qdV9VtwENWWnYD3V85SpIkPaD45HpJkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JEmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKknBi9JkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JEmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSerJSINXkk2SHJPkoiQXJtk1yaZJTkxycft88ChrkCRJmilG3eP1YeCEqtoB2BG4EDgEWFZV2wPL2rwkSdKcN7LglWRj4OnApwGq6tdVdROwF7C0NVsK7D2qGiRJkmaSUfZ4PQJYAXw2yZlJPpVkfWDzqrq6tbkG2HyENUiSJM0Yowxe84AnAh+rqicAt7HSbcWqKqDG2zjJ4iTLkyxfsWLFCMuUJEnqxyiD15XAlVV1aps/hi6IXZtkC4D2ed14G1fVkqpaWFUL58+fP8IyJUmS+jGy4FVV1wBXJHl0W7Q7cAHwNWBRW7YIOH5UNUiSJM0k80a8/9cDRyRZG7gUeBVd2PtSkoOAy4F9R1yDJEnSjDDS4FVVZwELx1m1+yiPK0mSNBP55HpJkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JEmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKknBi9JkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JEmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKkn80a58ySXAbcA9wB3V9XCJJsCRwMLgMuAfavqxlHWIUmSNBMM3eOVZLskX0hybJJdV+EYz6yqnapqYZs/BFhWVdsDy9q8JEnSnDdhj1eSdarqjoFF/wC8tU1/HdjpPh5zL2C3Nr0UOBl4233clyRJ0qwxWY/X15McMDB/F93twW3pbh0Oo4BvJzk9yeK2bPOqurpNXwNsvgr1SpIkzVqTBa89gY2SnJDk6cBbgOcALwJePuT+n1pVTwSeC7yu7ec3qqrowtnvSbI4yfIky1esWDHk4SRJkmauCYNXVd1TVR8B9gNeCHwY+GxVvbmqLhpm51V1Vfu8DjgO2AW4NskWAO3zugm2XVJVC6tq4fz581flnCRJkmakCYNXkj9OcgzwMeBw4O3APyb5QJJNptpxkvWTbDg2DfwpcB7wNWBRa7YIOP7+nIAkSdJsMdnjJD4BPA/YgK6n6ynAS5M8g+5xEM+ZYt+bA8clGTvOF6vqhCSnAV9KchBwObDv/TwHSZKkWWGy4HU33WD69YFfjy2squ8C351qx1V1KbDjOMtvAHZf1UIlSZJmu8mC18uA19CFrgMmaSdJkqQhTBi8qup/gDf3WIskSdKc5rsaJUmSemLwkiRJ6smUwSvJC5IY0CRJku6nYQLVfsDFSf45yQ6jLkiSJGmumjJ4VdUrgCcAPwUOT/LD9jqfDUdenSRJ0hwy1C3EqroZOAY4CtiC7n2NZyR5/QhrkyRJmlOGGeP1wiTHAScDawG7VNVz6R6O6uMmJEmShjTZA1THvAT4UFWdMriwqm5vr/2RJEnSEIYJXocCV4/NJFkX2LyqLquqZaMqTJIkaa4ZZozXl4F7B+bvacskSZK0CoYJXvOqavAl2b8G1h5dSZIkSXPTMMFrRZIXjs0k2Qu4fnQlSZIkzU3DjPF6LXBEko8AAa4ADhhpVZIkSXPQlMGrqn4KPDnJBm3+1pFXJUmSNAcN0+NFkj8DHgeskwSAqnr3COuSJEmac4Z5gOrH6d7X+Hq6W41/Dmw74rokSZLmnGEG1/9JVR0A3FhV7wJ2BR412rIkSZLmnmGC1x3t8/YkWwJ30b2vUZIkSatgmDFeX0+yCfAvwBlAAZ8cZVGSJElz0aTBK8kawLKqugk4Nsk3gHWq6pd9FCdJkjSXTHqrsaruBf59YP5OQ5ckSdJ9M8wYr2VJXpKx50isoiRrJjmz9ZaR5BFJTk1ySZKjk/j6IUmS9IAwTPB6Dd1Lse9McnOSW5LcvArH+BvgwoH59wEfqqrtgBuBg1ZhX5IkSbPWlMGrqjasqjWqau2q2qjNbzTMzpNsDfwZ8Kk2H+BZwDGtyVJg7/tUuSRJ0iwz5V81Jnn6eMur6pQh9v+vwFuBDdv8Q4CbquruNn8lsNUQ+5EkSZr1hnmcxN8OTK8D7AKcTtdzNaEkzweuq6rTk+y2qoUlWQwsBnj4wx++qptLkiTNOMO8JPsFg/NJtqHryZrKU4AXJnkeXWDbCPgwsEmSea3Xa2vgqgmOuwRYArBw4cIa4niSJEkz2jCD61d2JfCYqRpV1d9V1dZVtQB4KfBfVfVy4CRgn9ZsEXD8fahBkiRp1hlmjNdhdE+rhy6o7UT3BPv76m3AUUneA5wJfPp+7EuSJGnWGGaM1/KB6buBI6vq+6tykKo6GTi5TV9KN05MkiTpAWWY4HUMcEdV3QO/eSDqelV1+2hLkyRJmluGenI9sO7A/LrAd0ZTjiRJ0tw1TPBap6puHZtp0+uNriRJkqS5aZjgdVuSJ47NJNkZ+NXoSpIkSZqbhhnj9Ubgy0n+FwjwMGC/URYlSZI0Fw3zANXTkuwAPLot+klV3TXasiRJkuaeKW81JnkdsH5VnVdV5wEbJPmr0ZcmSZI0twwzxusvquqmsZmquhH4i5FVJEmSNEcNE7zWTJKxmSRrAmuPriRJkqS5aZjB9ScARyf5RJt/TVsmSZKkVTBM8HobsBj4yzZ/IvDJkVUkSZI0R015q7Gq7q2qj1fVPlW1D3ABcNjoS5MkSZpbhunxIskTgP2BfYGfAV8ZZVGSJElz0YTBK8mj6MLW/sD1wNFAquqZPdUmSZI0p0zW43UR8D3g+VV1CUCSg3upSpIkaQ6abIzXi4GrgZOSfDLJ7nSvDJIkSdJ9MGHwqqqvVtVLgR2Ak+je2fjQJB9L8qc91SdJkjRnDPNXjbdV1Rer6gXA1sCZdI+YkCRJ0ioY5sn1v1FVN1bVkqrafVQFSZIkzVWrFLwkSZJ03xm8JEmSemLwkiRJ6onBS5IkqScjC15J1kny4yRnJzk/ybva8kckOTXJJUmOTrL2qGqQJEmaSUbZ43Un8Kyq2hHYCdgzyZOB9wEfqqrtgBuBg0ZYgyRJ0owxsuBVnVvb7Frtq4BnAce05UuBvUdVgyRJ0kwy0jFeSdZMchZwHXAi8FPgpqq6uzW5EthqlDVIkiTNFCMNXlV1T1XtRPfE+13oXj80lCSLkyxPsnzFihWjKlGSJKk3vfxVY1XdRPe+x12BTZLMa6u2Bq6aYJslVbWwqhbOnz+/jzIlSZJGapR/1Tg/ySZtel3g2cCFdAFsn9ZsEXD8qGqQJEmaSeZN3eQ+2wJYmmRNuoD3par6RpILgKOSvIfuhdufHmENkiRJM8bIgldVnQM8YZzll9KN95IkSXpA8cn1kiRJPTF4SZIk9cTgJUmS1BODlyRJUk8MXpIkST0xeEmSJPXE4CVJktQTg5ckSVJPDF6SJEk9MXhJkiT1xOAlSZLUE4OXJElSTwxekiRJPTF4SZIk9cTgJUmS1BODlyRJUk8MXpIkST0xeEmSJPXE4CVJktQTg5ckSVJPDF6SJEk9MXhJkiT1xOAlSZLUk5EFryTbJDkpyQVJzk/yN235pklOTHJx+3zwqGqQJEmaSUbZ43U38OaqeizwZOB1SR4LHAIsq6rtgWVtXpIkac4bWfCqqqur6ow2fQtwIbAVsBewtDVbCuw9qhokSZJmkl7GeCVZADwBOBXYvKqubquuATbvowZJkqTpNvLglWQD4FjgjVV18+C6qiqgJthucZLlSZavWLFi1GVKkiSN3EiDV5K16ELXEVX1lbb42iRbtPVbANeNt21VLamqhVW1cP78+aMsU5IkqRej/KvGAJ8GLqyqDw6s+hqwqE0vAo4fVQ2SJEkzybwR7vspwCuBc5Oc1Zb9PfBe4EtJDgIuB/YdYQ2SJEkzxsiCV1X9N5AJVu8+quNKkiTNVD65XpIkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKknBi9JkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JEmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKknBi9JkqSeGLwkSZJ6YvCSJEnqicFLkiSpJwYvSZKknhi8JEmSemLwkiRJ6snIgleSzyS5Lsl5A8s2TXJikovb54NHdXxJkqSZZpQ9XocDe6607BBgWVVtDyxr85IkSQ8IIwteVXUK8IuVFu8FLG3TS4G9R3V8SZKkmabvMV6bV9XVbfoaYPOejy9JkjRtpm1wfVUVUBOtT7I4yfIky1esWNFjZZIkSaPRd/C6NskWAO3zuokaVtWSqlpYVQvnz5/fW4GSJEmj0nfw+hqwqE0vAo7v+fiSJEnTZpSPkzgS+CHw6CRXJjkIeC/w7CQXA3u0eUmSpAeEeaPacVXtP8Gq3Ud1TEmSpJnMJ9dLkiT1xOAlSZLUE4OXJElSTwxekiRJPTF4SZIk9cTgJUmS1BODlyRJUk8MXpIkST0xeEmSJPXE4CVJktQTg5ckSVJPDF6SJEk9MXhJkiT1xOAlSZLUE4OXJElSTwxekiRJPTF4SZIk9cTgJUmS1BODlyRJUk8MXpIkST0xeEmSJPXE4CVJktQTg5ckSVJPpiV4JdkzyU+SXJLkkOmoQZIkqW+9B68kawL/DjwXeCywf5LH9l2HJElS36ajx2sX4JKqurSqfg0cBew1DXVIkiT1ajqC11bAFQPzV7ZlkiRJc9q86S5gIkkWA4vb7K1JfjKd9cwymwHXT3cRq0MOzXSXoLnNa0Uazpy4Vnq8TradaMV0BK+rgG0G5rduy35HVS0BlvRV1FySZHlVLZzuOqSZzmtFGo7XyuozHbcaTwO2T/KIJGsDLwW+Ng11SJIk9ar3Hq+qujvJXwP/CawJfKaqzu+7DkmSpL5NyxivqvoW8K3pOPYDhLdopeF4rUjD8VpZTVJV012DJEnSA4KvDJIkSeqJwUuSJKknBq8hJdk7SSXZYbprWVVJnpPkrPZ1a3tP5llJPjfk9q9NcsAUbRYm+bfVU7Fmi1l+XayX5IYkG620/KtJ9ptku1vb55ZJjpmgzclJJv3T+yRvTLLewPy3kmyySiehB4RZfp3dr39/2j4OTLLlKOvsk2O8hpTkaGBL4L+q6p0jOsaaVXXPKPY9cIyTgbdU1fK+j625Z7ZfF0m+CPxnVS1t8xsDPwUeXlW3T7DNrVW1wRT7PZlxrrOV2lwGLKyqWf9QSo3WbL/OBo5xMlNcF6tzu5nKHq8hJNkAeCpwEN1zx0iyZpL3JzkvyTlJXt+WPynJD5KcneTHSTZsaf0jA/v7RpLd2vStST6Q5Gxg1yT/N8lpbb9LkqS12y7Jd9p+z0jyB0k+l2Tvgf0ekWTo914muSzJ+5KcAfx5kr9oxz47ybFjv40nOTTJW9r0yW2bHyf5nyRPa8t3S/KNgfafaW0vTfKGgWO+o/3G899Jjhzbr2afOXJdHDlWe/MiukfdrJFkWdvnueNtn2RBkvPa9LpJjkpyYZLjgHUH2n0syfIk5yd5V1v2Brp/SE9KclJbdlmSzdr0m9q5npfkjQPHuzDJJ9u+vp1kXTSnzZHrbLzzekWr8awkn2jntGaSw9vxz01ycJJ9gIXAEa3t7P+Zryq/pvgCXg58uk3/ANgZ+EvgGGBeW74psDZwKfCktmwjukd2HAh8ZGB/3wB2a9MF7DuwbtOB6c8DL2jTpwIvatPrAOsBzwC+2pZtDPxsrJ5JzuVkut+yAS4D3jqw7iED0+8BXt+mD6X7bWNs+w+06ecB32nTuwHfGGj/A+BBdK+ZuAFYC3gScFarf0Pg4rH9+jX7vubCddFqu3bsZx84AXh+q2+jtmwz4BJ+e4fg1va5ADivTb+J7pmEAH8E3D1wnW3aPtds188ftfnLgM0GarmsHWtn4FxgfWAD4HzgCe14dwM7tfZfAl4x3T8HfnmdTXWdDezzZLoQ9Rjg68BabflHgQPauZ040H6Twe2m+7/F6vqyx2s4+wNHtemj2vwewCeq6m6AqvoF8Gjg6qo6rS27eWz9JO4Bjh2Yf2aSU5OcCzwLeFySDYGtquq4tt87qur2qvou3VsA5reajh3ieCs7emD68Um+1479cuBxE2zzlfZ5Ot0/BuP5ZlXdWd1tlOuAzYGnAMe3+m+hu/A0e83666Kqfk335ox9Wm/TE+h6vAL8U5JzgO8AW9H9DE/k6cAX2j7PAc4ZWLdvul7lM+muqcdOce5PBY6rqtuq6la66+1pbd3PquqsNj3Z9ae5Y9ZfZ+PYnS5knZbkrDb/SLrg+MgkhyXZE7h5yP3NKjP2JdkzRZJN6X4A/zBJ0f3WWnSvPhrW3fzubd11BqbvqHZfPck6dMl/YVVdkeTQldqO53PAK+i6oF+1CjWNuW1g+nBg76o6O8mBdL1Y47mzfd7DxD9Ddw5MT9ZOs9Acuy6OBN5BF7aOr6q72s//fGDnNn/ZEMf8PUkeAbyFrhfixiSH35f9DFj5upr9t100oTl2nQ0KsLSq/u73ViQ7As8BXgvsC7x6FfY7K9jjNbV9gM9X1bZVtaCqtqHrUj0beE2SefCbC+QnwBZJntSWbdjWXwbslGSNJNsAu0xwrLEf8uvbff19AFrv0JVj99OTPCi//Wuow4E3tnYX3M9z3RC4OsladD1eq9v3gRckWaed3/NHcAz1Yy5dFycD2wOvowth0N06ua6FrmcC206xj1OAl7U6Hk93uxG62z23Ab9Msjnw3IFtbqG75lb2PWDvdH91uT7duLPvTXF8zU1z6TobtIyul/mhY/Un2bb1Oq9RVccCbwee2NpPdK3MSgavqe0PHLfSsmOBLYCfA+e0gYkva7ct9gMOa8tOpPth/j7dxXIB8G/AGeMdqKpuAj4JnEd3u2Pwt5pXAm9otz5+ADysbXMtcCHw2ft7onS/9Z/a6r1oNezvd7Qu8K/R3Yb5D7pxLL9c3cdRL+bMdVFV99KNl3kI8N22+AhgYbvlcgBTXw8fAzZIciHwbrrbgFTV2XS3GC8CvtjOecwS4IS0wfUD9ZxB9w/aj+mux09V1ZlTnYfmpDlzna10rAvogtW32z5PbOe0FXByu/34BWCsR+xw4ONzZXC9j5OY5dpvHucCT6yqGR9ikmxQVbe2uk8BFrd/aKTVZrZdF9Js5HV239jjNYsl2YPut43DZtEP/ZL228wZdIMxDV1arWbpdSHNKl5n9509XnNMkucA71tp8c+q6kXTUY80E3hdSKPndTYcg5ckSVJPvNUoSZLUE4OXJElSTwxeklabJPe0P/k+P9173d6cZNL/z6R7B+HLRlDLGweeNzRM+0OTXNXqPy/JC+/P8ZJ8K8km7euvBpZvmeSYVdm3pLnDMV6SVpskt1bVBm36obRnV1XVOyfZZje6d3au1gfqtqfNL2yvrRqm/aF072F8f5LH0D209KHtOV9Tbbsm8NPxjpdkAd17TB+/amcgaS6yx0vSSFTVdcBi4K/TWZDuXaBntK8/aU3fCzyt9TQdPFG7JFskOWWgR+ppbfmfJvlha/vlJBskeQOwJXBSkpOSrJnk8LbduUkOnqL2C+letbJZkq8mOb314i0ea5Pk1iQfaA+r/D+Dx2vrL2tP4n4v8Aet7n9p53dea7NOks+2ms5M95R8khyY5CtJTkhycZJ/Xk3/WSRNM9+fJ2lkqurS1hv0ULqXpT+7qu5Isj3d63kWAocw0OPVbteN1+5lwH9W1T+2fa7Xgs3bgT2q6rYkbwPeVFXvTvIm4JlVdX2Snele9Pv4doxNJqs7yR8D9wIrgFdX1S/aE7NPS3JsVd0ArA+cWlVvbtu8eux4K+3uEODxVbVTa7dgYN3rum9T/WGSHeie5P2otm4nupd23wn8JMlhVXXF1N91STOZwUtSX9YCPpJkJ7oXPD9qFdudBnwm3btEv1pVZyV5BvBY4PtJANYGfjjOPi8FHpnkMOCbwLcnOPbBSV5B9264/aqqkrwhydhziLahe6/jDa22Y4c684k9FTgMoKouSnI5vz3fZWMPpkxyAd37Ig1e0ixn8JI0MkkeSRdQrgPeCVwL7Eg3zOGOCTY7eLx2VXVKkqcDfwYcnuSDwI3AiVW1/2R1VNWNSXYEngO8FtgXePU4TT9UVe8fqH83YA9g16q6PcnJ/PZlwndU1T2THfd+unNg+h78/7U0JzjGS9JIJJkPfBz4SHV/xbMxcHUbrP5KYM3W9BZgw4FNx22XZFvg2qr6JPAp4InAj4CnJNmutVl/4Fbdb/bbbkmuUVXH0t2afOKQp7ExcGMLXTsAT56k7crnMdVy6Abwv7zV+Cjg4cBPhqxN0ixk8JK0Oq3bBpGfD3yH7pbeu9q6jwKL2mD0HYDb2vJzgHvSPX7i4Ena7QacneRMYD/gw1W1AjgQODLJOXS3GXdo7ZcAJ7TB7lsBJ6d7T+gXgL8b8nxOAOYluZBukPyPJmk7eLzfaOPBvt8G9v/LStt8FFgjybnA0cCBVXUnkuYsHychSZLUE3u8JEmSemLwkiRJ6onBS5IkqScGL0mSpJ4YvCRJknpi8JIkSeqJwUuSJKknBi9JkqSe/H9vlbg4UfrtaAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_avail = {'Accuracy_Training':metrics_Dropout.history['accuracy'][-1]*100, 'Accuracy_Validation':metrics_Dropout.history['val_accuracy'][-1]*100, 'Accuracy_Test':results_test_Dropout[-1]*100}\n",
    "split_data_avail = list(data_avail.keys())\n",
    "split_accuracy_cal = list(data_avail.values())\n",
    "  \n",
    "fig = plt.figure(figsize = (10, 5))\n",
    " \n",
    "plt.bar(split_data_avail, split_accuracy_cal,\n",
    "        width = 0.5,color=('green', 'green', 'green'))\n",
    "plt.xlabel(\"Datasets Partition\")\n",
    "plt.ylabel(\"Accuracy %\")\n",
    "plt.title(\"Obtained Accuracy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c08ac00",
   "metadata": {},
   "source": [
    "comparision between accuracies obtained in model 1 in which we used l1 regularization , model 2\n",
    "in which we used l2 regularization and model 3 in which we used dropout\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "84de9ded",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_avail = {'Accuracy_L1':metrics_L1.history['accuracy'][-1]*100, 'Accuracy_L2':metrics_L2.history['accuracy'][-1]*100, 'Accuracy_Dropout':metrics_Dropout.history['accuracy'][-1]*100}\n",
    "split_data_avail = list(data_avail.keys())\n",
    "split_accuracy_cal = list(data_avail.values())\n",
    "  \n",
    "fig = plt.figure(figsize = (10, 5))\n",
    " \n",
    "plt.bar(split_data_avail, split_accuracy_cal,width = 0.5,color=('blue','blue','blue'))\n",
    "plt.xlabel(\"Accuracies in each model\")\n",
    "plt.ylabel(\"Accuracy %\")\n",
    "plt.title(\"Obtained Accuracy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "69ae4e1f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}