Notebooks/04_SLP.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c7c0a93f",
   "metadata": {},
   "source": [
    "## <span style=\"color : green\"> Single Layer Perceptron </span>\n",
    "\n",
    "# <center> Table of Contents </center>\n",
    "\n",
    "1. Train a single layer perceptron model and determine the minimum error below 0.2\n",
    "2. To compute the final weight of each attribute\n",
    "1. To plot the epooch vs error plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "12bd5ee1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from tensorflow import keras\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import tensorflow as tf\n",
    "from sklearn.preprocessing import MinMaxScaler, LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from warnings import filterwarnings \n",
    "\n",
    "filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6f3b4714",
   "metadata": {
    "scrolled": true
   },
   "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>island</th>\n",
       "      <th>bill_length_mm</th>\n",
       "      <th>bill_depth_mm</th>\n",
       "      <th>flipper_length_mm</th>\n",
       "      <th>body_mass_g</th>\n",
       "      <th>sex</th>\n",
       "      <th>year</th>\n",
       "      <th>species</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rowid</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Torgersen</td>\n",
       "      <td>39.1</td>\n",
       "      <td>18.7</td>\n",
       "      <td>181.0</td>\n",
       "      <td>3750.0</td>\n",
       "      <td>male</td>\n",
       "      <td>2007</td>\n",
       "      <td>Adelie</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Torgersen</td>\n",
       "      <td>39.5</td>\n",
       "      <td>17.4</td>\n",
       "      <td>186.0</td>\n",
       "      <td>3800.0</td>\n",
       "      <td>female</td>\n",
       "      <td>2007</td>\n",
       "      <td>Adelie</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Torgersen</td>\n",
       "      <td>40.3</td>\n",
       "      <td>18.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>3250.0</td>\n",
       "      <td>female</td>\n",
       "      <td>2007</td>\n",
       "      <td>Adelie</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Torgersen</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2007</td>\n",
       "      <td>Adelie</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Torgersen</td>\n",
       "      <td>36.7</td>\n",
       "      <td>19.3</td>\n",
       "      <td>193.0</td>\n",
       "      <td>3450.0</td>\n",
       "      <td>female</td>\n",
       "      <td>2007</td>\n",
       "      <td>Adelie</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          island  bill_length_mm  bill_depth_mm  flipper_length_mm  \\\n",
       "rowid                                                                \n",
       "1      Torgersen            39.1           18.7              181.0   \n",
       "2      Torgersen            39.5           17.4              186.0   \n",
       "3      Torgersen            40.3           18.0              195.0   \n",
       "4      Torgersen             NaN            NaN                NaN   \n",
       "5      Torgersen            36.7           19.3              193.0   \n",
       "\n",
       "       body_mass_g     sex  year species  \n",
       "rowid                                     \n",
       "1           3750.0    male  2007  Adelie  \n",
       "2           3800.0  female  2007  Adelie  \n",
       "3           3250.0  female  2007  Adelie  \n",
       "4              NaN     NaN  2007  Adelie  \n",
       "5           3450.0  female  2007  Adelie  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('../Datasets/penguins.csv', index_col=0)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "9e103bf1",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "fd4fb4ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = df.loc[:,['bill_length_mm','flipper_length_mm','body_mass_g']]\n",
    "features = features.values\n",
    "label = df.iloc[:,-1:].values\n",
    "encoder = LabelEncoder()\n",
    "label = encoder.fit_transform(label)\n",
    "\n",
    "xtrain, xtest, ytrain, ytest = train_test_split(features,\n",
    "                                                    label,\n",
    "                                                    test_size=0.30,\n",
    "                                                    random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3279dcd3",
   "metadata": {},
   "source": [
    "### Train a single layer perceptron model and determine the minimum error below 0.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "3a723ad5",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(1, activation='sigmoid', input_shape=(3,)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "4a6dc549",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_8\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_8 (Dense)              (None, 1)                 4         \n",
      "=================================================================\n",
      "Total params: 4\n",
      "Trainable params: 4\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "a9c7c82e",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer=Adam(learning_rate=0.4), \\\n",
    "              loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "2ae83e56",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 63 samples\n",
      "Epoch 1/100\n",
      "63/63 [==============================] - 0s 455us/sample - loss: 13.6814 - accuracy: 0.7619\n",
      "Epoch 2/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 12.5720 - accuracy: 0.7619\n",
      "Epoch 3/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 16.6654 - accuracy: 0.6667\n",
      "Epoch 4/100\n",
      "63/63 [==============================] - 0s 65us/sample - loss: 14.7126 - accuracy: 0.7619\n",
      "Epoch 5/100\n",
      "63/63 [==============================] - 0s 78us/sample - loss: 25.4146 - accuracy: 0.6349\n",
      "Epoch 6/100\n",
      "63/63 [==============================] - 0s 60us/sample - loss: 15.9394 - accuracy: 0.7619\n",
      "Epoch 7/100\n",
      "63/63 [==============================] - 0s 63us/sample - loss: 9.7335 - accuracy: 0.7937\n",
      "Epoch 8/100\n",
      "63/63 [==============================] - 0s 61us/sample - loss: 9.1156 - accuracy: 0.7937\n",
      "Epoch 9/100\n",
      "63/63 [==============================] - 0s 92us/sample - loss: 9.7424 - accuracy: 0.8095\n",
      "Epoch 10/100\n",
      "63/63 [==============================] - 0s 57us/sample - loss: 10.7336 - accuracy: 0.7619\n",
      "Epoch 11/100\n",
      "63/63 [==============================] - 0s 59us/sample - loss: 10.5196 - accuracy: 0.7778\n",
      "Epoch 12/100\n",
      "63/63 [==============================] - 0s 58us/sample - loss: 29.7729 - accuracy: 0.6667\n",
      "Epoch 13/100\n",
      "63/63 [==============================] - 0s 89us/sample - loss: 18.1306 - accuracy: 0.6825\n",
      "Epoch 14/100\n",
      "63/63 [==============================] - 0s 91us/sample - loss: 21.2234 - accuracy: 0.7302\n",
      "Epoch 15/100\n",
      "63/63 [==============================] - 0s 67us/sample - loss: 17.4282 - accuracy: 0.6508\n",
      "Epoch 16/100\n",
      "63/63 [==============================] - 0s 94us/sample - loss: 72.3874 - accuracy: 0.5397\n",
      "Epoch 17/100\n",
      "63/63 [==============================] - 0s 102us/sample - loss: 34.3401 - accuracy: 0.7460\n",
      "Epoch 18/100\n",
      "63/63 [==============================] - 0s 83us/sample - loss: 60.7388 - accuracy: 0.6032\n",
      "Epoch 19/100\n",
      "63/63 [==============================] - 0s 78us/sample - loss: 63.1940 - accuracy: 0.5397\n",
      "Epoch 20/100\n",
      "63/63 [==============================] - 0s 82us/sample - loss: 33.6840 - accuracy: 0.5714\n",
      "Epoch 21/100\n",
      "63/63 [==============================] - 0s 68us/sample - loss: 49.4446 - accuracy: 0.5873\n",
      "Epoch 22/100\n",
      "63/63 [==============================] - 0s 88us/sample - loss: 34.7672 - accuracy: 0.6825\n",
      "Epoch 23/100\n",
      "63/63 [==============================] - 0s 78us/sample - loss: 31.3867 - accuracy: 0.6667\n",
      "Epoch 24/100\n",
      "63/63 [==============================] - 0s 76us/sample - loss: 19.8753 - accuracy: 0.7778\n",
      "Epoch 25/100\n",
      "63/63 [==============================] - 0s 109us/sample - loss: 32.6204 - accuracy: 0.6190\n",
      "Epoch 26/100\n",
      "63/63 [==============================] - 0s 58us/sample - loss: 29.7681 - accuracy: 0.7302\n",
      "Epoch 27/100\n",
      "63/63 [==============================] - 0s 90us/sample - loss: 26.1171 - accuracy: 0.6825\n",
      "Epoch 28/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 24.1677 - accuracy: 0.7302\n",
      "Epoch 29/100\n",
      "63/63 [==============================] - 0s 81us/sample - loss: 19.9371 - accuracy: 0.6825\n",
      "Epoch 30/100\n",
      "63/63 [==============================] - 0s 75us/sample - loss: 26.3754 - accuracy: 0.7143\n",
      "Epoch 31/100\n",
      "63/63 [==============================] - 0s 68us/sample - loss: 13.9192 - accuracy: 0.7460\n",
      "Epoch 32/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 14.3150 - accuracy: 0.7302\n",
      "Epoch 33/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 7.4930 - accuracy: 0.7619\n",
      "Epoch 34/100\n",
      "63/63 [==============================] - 0s 93us/sample - loss: 16.9926 - accuracy: 0.7778\n",
      "Epoch 35/100\n",
      "63/63 [==============================] - 0s 91us/sample - loss: 16.1084 - accuracy: 0.7460\n",
      "Epoch 36/100\n",
      "63/63 [==============================] - 0s 93us/sample - loss: 15.3941 - accuracy: 0.7460\n",
      "Epoch 37/100\n",
      "63/63 [==============================] - 0s 61us/sample - loss: 11.2238 - accuracy: 0.7937\n",
      "Epoch 38/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 10.4237 - accuracy: 0.7778\n",
      "Epoch 39/100\n",
      "63/63 [==============================] - 0s 91us/sample - loss: 11.5960 - accuracy: 0.7460\n",
      "Epoch 40/100\n",
      "63/63 [==============================] - 0s 64us/sample - loss: 11.8496 - accuracy: 0.7778\n",
      "Epoch 41/100\n",
      "63/63 [==============================] - 0s 94us/sample - loss: 8.2977 - accuracy: 0.8571\n",
      "Epoch 42/100\n",
      "63/63 [==============================] - 0s 65us/sample - loss: 11.9161 - accuracy: 0.8254\n",
      "Epoch 43/100\n",
      "63/63 [==============================] - 0s 50us/sample - loss: 9.9178 - accuracy: 0.7937\n",
      "Epoch 44/100\n",
      "63/63 [==============================] - 0s 59us/sample - loss: 10.2292 - accuracy: 0.8095\n",
      "Epoch 45/100\n",
      "63/63 [==============================] - 0s 77us/sample - loss: 8.1335 - accuracy: 0.8254\n",
      "Epoch 46/100\n",
      "63/63 [==============================] - 0s 80us/sample - loss: 8.3830 - accuracy: 0.7937\n",
      "Epoch 47/100\n",
      "63/63 [==============================] - 0s 57us/sample - loss: 14.1788 - accuracy: 0.7778\n",
      "Epoch 48/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 9.9173 - accuracy: 0.7619\n",
      "Epoch 49/100\n",
      "63/63 [==============================] - 0s 61us/sample - loss: 12.6059 - accuracy: 0.7460\n",
      "Epoch 50/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 9.2994 - accuracy: 0.8095\n",
      "Epoch 51/100\n",
      "63/63 [==============================] - 0s 81us/sample - loss: 10.6496 - accuracy: 0.7619\n",
      "Epoch 52/100\n",
      "63/63 [==============================] - 0s 81us/sample - loss: 8.5535 - accuracy: 0.8095\n",
      "Epoch 53/100\n",
      "63/63 [==============================] - 0s 70us/sample - loss: 8.8595 - accuracy: 0.7937\n",
      "Epoch 54/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 12.5553 - accuracy: 0.7937\n",
      "Epoch 55/100\n",
      "63/63 [==============================] - 0s 57us/sample - loss: 9.1274 - accuracy: 0.8254\n",
      "Epoch 56/100\n",
      "63/63 [==============================] - 0s 54us/sample - loss: 10.3936 - accuracy: 0.7619\n",
      "Epoch 57/100\n",
      "63/63 [==============================] - 0s 80us/sample - loss: 12.9181 - accuracy: 0.7778\n",
      "Epoch 58/100\n",
      "63/63 [==============================] - 0s 75us/sample - loss: 9.4187 - accuracy: 0.7937\n",
      "Epoch 59/100\n",
      "63/63 [==============================] - 0s 64us/sample - loss: 8.3266 - accuracy: 0.7937\n",
      "Epoch 60/100\n",
      "63/63 [==============================] - 0s 64us/sample - loss: 9.2216 - accuracy: 0.7937\n",
      "Epoch 61/100\n",
      "63/63 [==============================] - 0s 53us/sample - loss: 8.8174 - accuracy: 0.8095\n",
      "Epoch 62/100\n",
      "63/63 [==============================] - 0s 73us/sample - loss: 8.4217 - accuracy: 0.8254\n",
      "Epoch 63/100\n",
      "63/63 [==============================] - 0s 61us/sample - loss: 13.8495 - accuracy: 0.8095\n",
      "Epoch 64/100\n",
      "63/63 [==============================] - 0s 68us/sample - loss: 7.7079 - accuracy: 0.8254\n",
      "Epoch 65/100\n",
      "63/63 [==============================] - 0s 76us/sample - loss: 7.7148 - accuracy: 0.8254\n",
      "Epoch 66/100\n",
      "63/63 [==============================] - 0s 66us/sample - loss: 10.8513 - accuracy: 0.7619\n",
      "Epoch 67/100\n",
      "63/63 [==============================] - 0s 58us/sample - loss: 9.4752 - accuracy: 0.7460\n",
      "Epoch 68/100\n",
      "63/63 [==============================] - 0s 55us/sample - loss: 7.8689 - accuracy: 0.8095\n",
      "Epoch 69/100\n",
      "63/63 [==============================] - 0s 57us/sample - loss: 11.1713 - accuracy: 0.8095\n",
      "Epoch 70/100\n",
      "63/63 [==============================] - 0s 82us/sample - loss: 49.5217 - accuracy: 0.5079\n",
      "Epoch 71/100\n",
      "63/63 [==============================] - 0s 67us/sample - loss: 26.5380 - accuracy: 0.6984\n",
      "Epoch 72/100\n",
      "63/63 [==============================] - 0s 76us/sample - loss: 66.6131 - accuracy: 0.4286\n",
      "Epoch 73/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 49.1741 - accuracy: 0.5714\n",
      "Epoch 74/100\n",
      "63/63 [==============================] - 0s 61us/sample - loss: 35.6421 - accuracy: 0.6349\n",
      "Epoch 75/100\n",
      "63/63 [==============================] - 0s 57us/sample - loss: 36.4312 - accuracy: 0.6032\n",
      "Epoch 76/100\n",
      "63/63 [==============================] - 0s 83us/sample - loss: 97.4111 - accuracy: 0.4921\n",
      "Epoch 77/100\n",
      "63/63 [==============================] - 0s 61us/sample - loss: 96.8895 - accuracy: 0.5397\n",
      "Epoch 78/100\n",
      "63/63 [==============================] - 0s 54us/sample - loss: 36.9256 - accuracy: 0.6984\n",
      "Epoch 79/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63/63 [==============================] - 0s 49us/sample - loss: 70.8585 - accuracy: 0.6349\n",
      "Epoch 80/100\n",
      "63/63 [==============================] - 0s 59us/sample - loss: 47.4620 - accuracy: 0.6190\n",
      "Epoch 81/100\n",
      "63/63 [==============================] - 0s 60us/sample - loss: 44.0646 - accuracy: 0.6667\n",
      "Epoch 82/100\n",
      "63/63 [==============================] - 0s 77us/sample - loss: 22.9720 - accuracy: 0.7143\n",
      "Epoch 83/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 17.7974 - accuracy: 0.8254\n",
      "Epoch 84/100\n",
      "63/63 [==============================] - 0s 75us/sample - loss: 42.8863 - accuracy: 0.5397\n",
      "Epoch 85/100\n",
      "63/63 [==============================] - 0s 50us/sample - loss: 32.9654 - accuracy: 0.6984\n",
      "Epoch 86/100\n",
      "63/63 [==============================] - 0s 78us/sample - loss: 22.7783 - accuracy: 0.7619\n",
      "Epoch 87/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 24.7782 - accuracy: 0.7302\n",
      "Epoch 88/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 11.8858 - accuracy: 0.7619\n",
      "Epoch 89/100\n",
      "63/63 [==============================] - 0s 71us/sample - loss: 9.1682 - accuracy: 0.7778\n",
      "Epoch 90/100\n",
      "63/63 [==============================] - 0s 78us/sample - loss: 9.3264 - accuracy: 0.8095\n",
      "Epoch 91/100\n",
      "63/63 [==============================] - 0s 77us/sample - loss: 12.1154 - accuracy: 0.8095\n",
      "Epoch 92/100\n",
      "63/63 [==============================] - 0s 64us/sample - loss: 11.7218 - accuracy: 0.7619\n",
      "Epoch 93/100\n",
      "63/63 [==============================] - 0s 67us/sample - loss: 12.5729 - accuracy: 0.7778\n",
      "Epoch 94/100\n",
      "63/63 [==============================] - 0s 82us/sample - loss: 12.5825 - accuracy: 0.7619\n",
      "Epoch 95/100\n",
      "63/63 [==============================] - 0s 70us/sample - loss: 12.2950 - accuracy: 0.7937\n",
      "Epoch 96/100\n",
      "63/63 [==============================] - 0s 68us/sample - loss: 14.9263 - accuracy: 0.7460\n",
      "Epoch 97/100\n",
      "63/63 [==============================] - 0s 74us/sample - loss: 11.7456 - accuracy: 0.7937\n",
      "Epoch 98/100\n",
      "63/63 [==============================] - 0s 58us/sample - loss: 8.8374 - accuracy: 0.8413\n",
      "Epoch 99/100\n",
      "63/63 [==============================] - 0s 73us/sample - loss: 9.1457 - accuracy: 0.7937\n",
      "Epoch 100/100\n",
      "63/63 [==============================] - 0s 60us/sample - loss: 8.4557 - accuracy: 0.8413\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(xtrain, ytrain, epochs=100, verbose=1, )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "1e3e1e2e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Loss: 24.991989135742188, Accuracy: 0.6666666865348816\n"
     ]
    }
   ],
   "source": [
    "loss, accuracy = model.evaluate(xtest, ytest, verbose = 0)\n",
    "print(f'\\n\\nLoss: {loss}, Accuracy: {accuracy}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "929aac0c",
   "metadata": {},
   "outputs": [],
   "source": [
    "ypred = model.predict(xtest)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09b95c5f",
   "metadata": {},
   "source": [
    "### To compute the final weight of each attribute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "bc205e45",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[array([[-2.1627753 ],\n",
      "       [-3.3834507 ],\n",
      "       [ 0.16183503]], dtype=float32), array([-6.940952], dtype=float32)]\n"
     ]
    }
   ],
   "source": [
    "print(model.get_weights())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "985150bc",
   "metadata": {},
   "source": [
    "### To plot the epooch vs error plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "0ed3bd1d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAABI2klEQVR4nO2deXzcdZ3/n++5cydN0jZN2qalLeVuISCXWEAUAQUVUdYDb9ZlRfEEXc9Vd9ddL1bdn3isCCICyoLIIZeUG1roBQVK77Rpmzb3MZnr8/vj+/1OZjIzOZq5kryfj0cemfnOd2Y+32Tm+/q+bzHGoCiKoigArkIvQFEURSkeVBQURVGUOCoKiqIoShwVBUVRFCWOioKiKIoSR0VBURRFiaOioCgTQESaRcSIiGcc+35YRJ7Ix7oUJVuoKCjTFhHZISIhEakbsX2dfWJvLtDSJiQuipJPVBSU6c524HLnjogcB5QUbjmKUtyoKCjTnZuADyXcvwL4XeIOIlIlIr8TkXYR2Ski/yIiLvsxt4j8l4gcFJFtwIVpnvtrEWkTkT0i8h0RcU9mwSIyT0TuFpEOEXldRD6R8NgpIrJGRHpEZL+I/NDeHhCRm0XkkIh0icjzIjJnMutQZiYqCsp05xmgUkSOsk/W7wVuHrHPfwNVwGLgTVgi8hH7sU8AFwErgRbg0hHPvRGIAEvsfd4CfHySa/4D0ArMs9/veyJyrv3YT4CfGGMqgSOA2+ztV9jHMB+oBf4RGJzkOpQZiIqCMhNwrIXzgFeAPc4DCUJxnTGm1xizA/gB8EF7l8uAHxtjdhtjOoB/S3juHOBtwGeNMf3GmAPAj4D3He5CRWQ+cCbwZWNM0BizDvhVwnrCwBIRqTPG9BljnknYXgssMcZEjTFrjTE9h7sOZeaioqDMBG4C/gH4MCNcR0Ad4AN2JmzbCTTat+cBu0c85rAQ8AJttsumC/gFMHsSa50HdBhjejOs52PAMuAV20V0kb39JuAB4FYR2Ssi3xcR7yTWocxQVBSUaY8xZidWwPkC4M8jHj6IdZW9MGHbAoatiTYsl0ziYw67gSGgzhhTbf9UGmOOmcRy9wKzRKQi3XqMMVuMMZdjCc9/AHeISJkxJmyM+ZYx5mjgdCyX14dQlAmioqDMFD4GnGOM6U/caIyJYvnlvysiFSKyEPgcw3GH24CrRaRJRGqAaxOe2wb8DfiBiFSKiEtEjhCRN01gXX47SBwQkQDWyf8p4N/sbcfba/89gIh8QETqjTExoMt+jaiInC0ix9nusB4soYtOYB2KAqgoKDMEY8xWY8yaDA9/GugHtgFPALcAv7Ef+yWWW2Y98AKplsaHsNxPLwOdwB1AwwSW1ocVEHZ+zsFKoW3GshruBL5hjHnQ3v984CUR6cMKOr/PGBME5trv3QNsBh4jNaCuKGMiOmRHURRFcVBLQVEURYmjoqAoiqLEUVFQFEVR4qgoKIqiKHGmdIfGuro609zcXOhlKIqiTCnWrl170BhTn+6xnImCiPwGq4DmgDHmWHvbLOCPWOl2O4DLjDGd9mPXYeVjR4GrjTEPjPUezc3NrFmTKctQURRFSYeI7Mz0WC7dR7/FyqlO5FrgYWPMUuBh+z4icjRWv5hj7Of8fLKdJhVFUZSJkzNRMMasBjpGbL4Yq6sk9u9LErbfaowZMsZsB14HTsnV2hRFUZT05DvQPMduDeC0CHAahzWS3HSsleEGYIqiKEqeKJZAs6TZlrbUWkQ+CXwSYMGCBel2URRlmhMOh2ltbSUYDBZ6KUVNIBCgqakJr3f8DXPzLQr7RaTBGNMmIg3AAXt7K8mdKJuw+r6kYIy5AbgBoKWlRXt0KMoMpLW1lYqKCpqbmxFJd02pGGM4dOgQra2tLFq0aNzPy7f76G6sCVHYv+9K2P4+EfGLyCJgKfBcntemKMoUIRgMUltbq4IwCiJCbW3thK2pXKak/gFYBdSJSCvwDeDfgdtE5GPALuA9AMaYl0TkNqxOkxHgKrulsaIoSlpUEMbmcP5GORMFexBIOs5Nt9EY813gu7laj6Ioxc/rB3p5YWcX5x09h5oyX6GXMyPRNheKohQN//P3bXzpTxs45XsP8amb1/LMtkOFXlIKhw4dYsWKFaxYsYK5c+fS2NgYvx8KhUZ97po1a7j66qvHfI/TTz89W8udMMWSfaQoikIwEmVOpZ8Lj5vHnS+28tTWQ6z/xlsKvawkamtrWbduHQDf/OY3KS8v5wtf+EL88UgkgseT/tTa0tJCS0vLmO/x1FNPZWWth4NaCoqiFA3hSIyaUh9ff/vRvPfkBQyGp0Zo8cMf/jCf+9znOPvss/nyl7/Mc889x+mnn87KlSs5/fTTefXVVwH4+9//zkUXXQRYgvLRj36UVatWsXjxYq6//vr465WXl8f3X7VqFZdeeinLly/n/e9/P85gtHvvvZfly5dz5plncvXVV8dfd7KopaAoStEQjsbwuq1rVZ9bCEdjGGNGDZh+6y8v8fLenqyu4+h5lXzj7cdM6DmvvfYaDz30EG63m56eHlavXo3H4+Ghhx7iK1/5Cn/6059SnvPKK6/w6KOP0tvby5FHHsmnPvWplJqCF198kZdeeol58+Zxxhln8OSTT9LS0sKVV17J6tWrWbRoEZdfnimEO3FUFBRFKRrCUYPXbQmA1+3CGIjGDB538Wcavec978Httlq2dXd3c8UVV7BlyxZEhHA4nPY5F154IX6/H7/fz+zZs9m/fz9NTU1J+5xyyinxbStWrGDHjh2Ul5ezePHieP3B5Zdfzg033JCV41BRUBSlaAglWApej/U7HDV4RmmPOdEr+lxRVlYWv/21r32Ns88+mzvvvJMdO3awatWqtM/x+/3x2263m0gkMq59HBdSLtCYgqIoRUMkGsNni4EjDqForJBLOiy6u7tpbLTat/32t7/N+usvX76cbdu2sWPHDgD++Mc/Zu21VRQURSkaLPfRcEzB2jb1ROFLX/oS1113HWeccQbRaPaD5SUlJfz85z/n/PPP58wzz2TOnDlUVVVl5bUll2ZIrmlpaTE6ZEdRpg/n/3g1C2tL+cUHW7j1uV1c++eNPH3dOTRUlSTtt3nzZo466qgCrbI46Ovro7y8HGMMV111FUuXLuWaa65J2S/d30pE1hpj0ubGqqWgKErREIrG8LiT3UfhyNS9cM0lv/zlL1mxYgXHHHMM3d3dXHnllVl5XQ00K4pSNISjMXwjAs1TMaaQD6655pq0lsFkUUtBUZSiIRwZTkkdK6YwlV3f+eJw/kYqCoqiFA2JxWtx91EaUQgEAhw6dEiFYRSceQqBQGBCz1P3kaIoRcN4RaGpqYnW1lba29vzur6phjN5bSKoKCiKUjSEoya1TiFNoNnr9U5ompgyftR9pChK0WBZCnZMwTN16xSmMioKiqIUBbGYIRIz43IfKblDRUFRlKIgHLNO/ioKhUVFQVGUoiActWIHiV1SAUJRzTDKJyoKiqIUBeHISEtBkrYr+UFFQVGUosBxE6n7qLCoKCiKUhSEY5abyKeiUFBUFBRFKQri7iOP0+ZCYwqFQEVBUZSiIMV9pHUKBUFFQVGUoiCUKaaggea8oqKgKEpRMDIl1eNSS6EQqCgoilIUjHQfiQg+tysegFbyg4qCoihFwcg6Beu2qPsoz6goKIpSFDgWQZIoeFzqPsozKgqKohQFjkXgS7IUXJqSmmdUFBRFKQriMQU7FRUsgVBLIb+oKCiKUhSMTEm1bouKQp5RUVAUpShwUlJHuo9UFPKLioKiKEWBc/L3uIfdR163K+04TiV3qCgoilIUjKxTAM0+KgQFEQURuUZEXhKRTSLyBxEJiMgsEXlQRLbYv2sKsTZFUQpDKE2dgk9jCnkn76IgIo3A1UCLMeZYwA28D7gWeNgYsxR42L6vKMoMIRLTmEIxUCj3kQcoEREPUArsBS4GbrQfvxG4pDBLUxSlEAxXNI+IKWidQl7JuygYY/YA/wXsAtqAbmPM34A5xpg2e582YHa+16YoSuEIR2OIgNuVLAra5iK/FMJ9VINlFSwC5gFlIvKBCTz/kyKyRkTWtLe352qZiqLkmVDU4HW7EEkoXvNoTCHfFMJ99GZguzGm3RgTBv4MnA7sF5EGAPv3gXRPNsbcYIxpMca01NfX523RiqLklnA0lhRPAI0pFIJCiMIu4FQRKRXrkuBcYDNwN3CFvc8VwF0FWJuiKAUiHI0l1SiAIwoaU8gnnny/oTHmWRG5A3gBiAAvAjcA5cBtIvIxLOF4T77XpihK4QhHY0npqOAEmtVSyCd5FwUAY8w3gG+M2DyEZTUoijIDCUdNivtI6xTyj1Y0K4pSFFiWQhr3kWYf5RUVBUVRioK07iOPxhTyjYqCoihFQShiMsYUjFFhyBcqCoqiFAXhaAyvJzWmAMMtMJTco6KgKEpRYNUppMYUnMeU/KCioChKURCOxvC4Ut1HAGGdqZA3VBQURSkKQlGT4j5y7mutQv5QUVAUpSiIpHEfOffVfZQ/VBQURSkK0qWkOu4kFYX8oaKgKEpREI6mSUn1qCjkGxWFKcZfN7TRPRgu9DIUJeuEIqmWguM+CmmgOW+oKEwhDvYNcdUtL/CX9XsLvRRFyTrhaAyfR1NSC42KwhRiMBQFoH8oUuCVKEr2GTUlVUUhb6goTCGctLxgWL8gyvQjbUwhLgrD7iNjDMFwNK9rm0moKEwhnKulQf1CKNMQq83FiJRUT2pK6oMv76flOw/RG9TYWi5QUZhChCKOpaCioEw/Mo3jdB5z2NUxQN9QhPbeobyub6agojCFcL4YQxEVBWV6EY0ZYoZR3EfDojBgx9Z6gxpbywUqClOIIdtScALOijJdcE76mUQhlBBTcEShR91HOUFFYQrhBNs00KxMN0JxURjZ5sJpiDf8mXfcp2op5AYVhSmEE1PQQLMy3XBO+r6UhnipgeaBkCUGGmjODSoKU4hwVAPNyvTEsYLHU6cwaFvKPYNqKeQCFYUpRDz7SAeZK9OMcAb3UbqYwqBaCjlFRWEKES9e00CzMs1wRGGk+8iX1lJwAs1qKeQCFYUpxLCloKKgTA3+7d7NPPrKgTH3c9xHqdlHdkwhoimp+UJFYQoRr2hWS0GZItz8zE7u3dg25n6ZUlLdLkFkhKWgKak5RUVhCqEVzcpUIxiJ0dEfGnO/TCmpIoLX7UqOKcRTUlUUcoGKwhQinn2kgWZlChCOxojGDB0DY4tCPCXVnXpK8rldaS0FdR/lBhWFKYRjKYQi1pdNUYoZx6Idj6UQjyl4Uk9JXreo+yiPqChMIRJNaO1/pBQ7TuX9+ETB2tfjkpTHvAmWgjGGAa1ozikqClOIUCT1aklRipXEdhShMVyemQLNzjZnHGc4aojGDC6xXtcYtZizjYrCFCLRhB4trrCna1DnOCsFJ9Ga7RojruC4j0bWKTjbRmbe1ZX7icZMPD1VyR4qClOIUJqmYCPp7A9xwU8e5z/ufyVfy1KUtCQ2bjw0hgtpdEthOKbgZB7NqQwA6kLKBSoKU4h0wbaR/Pih1+geDHNQB5AoBSbxwmWsuEKmlFRr27Cl4DTDm1PpBzQtNReoKEwhhhJEIV2g+fUDvdz87C4ANauVgpNoKYwlCvE2F5liCrZ7ybEUZtuWgra6yD6eQi9AGT/JPeVTYwrf+etmSn1uFswqpT+kXxalsCReuIwpCpHM7iOf2xV/3LGQ51Q4oqCWQrYpiKUgItUicoeIvCIim0XkNBGZJSIPisgW+3dNIdZWzISiMZyMvZHuo0dfPcDfX23nM+cupammhIEhtRSUwjIxS2GUOgVPakxhdtx9pBc/2aZQ7qOfAPcbY5YDJwCbgWuBh40xS4GH7ftKAuFojIqAF0htivf9+1+lubaUD53WTJnPo5aCUnAOJ6YwVp2C4xbVmELuyLsoiEglcBbwawBjTMgY0wVcDNxo73YjcEm+11bshCIxKkssj1+ipWCMYcv+Xt52XAM+j4tSv1tjCkrBcS5cyv2eMVtdRDJ0SXW2xWMK9ud6tuM+0kE7WacQlsJioB34XxF5UUR+JSJlwBxjTBuA/Xt2uieLyCdFZI2IrGlvb8/fqouAUNRQGbcUEoPOMSIxQ0XAEowyn4f+If2yKIXFcR81VAXo6Bs70Ox2Ce40lkJi7yPHfTSrzIfHJWop5IBCiIIHOBH4H2PMSqCfCbiKjDE3GGNajDEt9fX1uVpjURKKxOKiMJRgmvfZAlDut0Sh1OexhCI6ehWpouQSx300r7qEzjGL12Jp01EhuU7BsYBLfW4qAh6NKeSAQohCK9BqjHnWvn8HlkjsF5EGAPv32JM5ZhihSDRuDSS6j/qCyaJQ5ncD0K8uJKWADIWjiFiWwljFa6FoLK3rCOyYwoi28SU+N5UlXs0+ygF5FwVjzD5gt4gcaW86F3gZuBu4wt52BXBXvtdW7ISjhlKfG49LkgLNjqVQlmApwHChj6IUgmAkht/jYlaZj87+0Kh9isLRWNoaBbAykpyYwkAogkssl5JaCrmhUHUKnwZ+LyI+YBvwESyBuk1EPgbsAt5ToLUVLaFIDJ/HRcDrZjA07BpyRKFipKWgaalKAQmGowS8bmaV+YjEDD3BCFUl3rT7hiMms6XgSkhJDcUo9XkQESr8Xo0p5ICCiIIxZh3Qkuahc/O8lClF2DaxA153kqXQP8JSKFNLQSkCguEoAY8lCmD15cooCtEYXk+mmEJioDlCwGtd9FSWeNh5aCAHK5/ZaJuLKcSwpeBKygGPB5rteEOpWgpKERAMxwh4XdTYojBaXCEcM3hdmd1HkYSU1FKf9fmuCHjp0W7AWUdFYQoRsv2uAa87rShUqKWgFBGO+6g2wVLIRDgyeqA5FI1ZA3ZCUUq8jihoTCEXqChMEYwxlih4XJR43UktBJzsozLNPlKKiGAkht/rpqbUEoXRqppHcx/57FTVSMwwGI5SkmAp9IUixHQ0bVYZlyiISJmIuOzby0TkHSKS3jmo5IRozGAMdkzBlZySOhRBhLhZHc8+0gI2pYBYMQUXteVju4/GSkkFSzgGEyyFyoAHY6BXP+dZZbyWwmogICKNWH2JPgL8NleLUlJxesM42UcjU1LL7YwMGHYf9emXRSkgQ5EYAa+bEq8bv8c1agFbeDyiELHcR87Fj1PIqRlI2WW8oiDGmAHgXcB/G2PeCRydu2UpIwlHhnvDBNK4j5wgMwwHmrX/kVJIhsJRAl4XIkJtmY9Do7S6CEfNqHUKYF0YBZPcR9ZnXuMK2WXcoiAipwHvB/5qb9NZDHlkKGqd4OOWQkKguT8UiccTwBIOn8elnVKVghIMR/F7rBP4rHLfOCyF0WMK4WgsKdBcaae3agZSdhmvKHwWuA640xjzkogsBh7N2aqUFOKDzd1CyYiU1N5gJN7iwqHM59aZCkpBcVJSAWpKfaPHFCIxPOOJKYQTU1LVUsgF47raN8Y8BjwGYAecDxpjrs7lwpRkQpHkmMJgOLl4rSKQ/K8s1ZkKSoEJRqLxQrPaMt+ohWaR2CjuoxGB5kBC9hFA75BaCtlkvNlHt4hIpd3i+mXgVRH5Ym6XpiTiVHR63U5KanKg2QkuO5T51VJQCotTpwBQU+YbOyU1Y5dU6zQ1GIoRisYo9VqfdbUUcsN43UdHG2N6sAbf3AssAD6Yq0UpqcQtBbcLvx1odhqMjQw0g1oKSmExxljuIztIXFvmo28okjS3OZHRitd8dv2C0xG1xGft54iCxhSyy3hFwWvXJVwC3GWMCQNaMZJHnJRUr93mAqyUP7BTUkfGFHT6mlJAnM+mP8FSAOjsT38CD0VN2vnMMGwpOCf/Etsq9nusVFe1FLLLeEXhF8AOoAxYLSILgZ5cLUpJxbEU/Lb7CCzz3BiTXhR0+ppSQIbslOnEmAJkrmoetXW2vb3bEQX7NcHuf6SikFXGG2i+Hrg+YdNOETk7N0tS0hFOshSsL8WgnfIXMySlpIJ1X91HSqFwiisTs49gdFEYK6bgiIKTfQRWp1QdtJNdxhtorhKRHzqzkUXkB1hWg5InEmMKw5ZCLJ55kRpT0ECzUjicRIiAXafgtLroyFCrMFpFs2NBDMcUki0FdR9ll/G6j34D9AKX2T89wP/malFKKonZR87VVzAcjbfHLve7k/ZXS0EpJMER7qNZZX4AOvqGUvY1xhCOmsx1CnagOZ37qDLg0TYXWWa8VclHGGPenXD/WyKyLgfrUTIwlFCn4E9wHzl95sv9yf0JS31WhlI0ZnC70pvlipItDvQEqS33xz9rcUvBvoCpKvEiAh0DqSfwSGy4MDMdw4Fm6yInyX0U8LK3azBLR6HA+C2FQRE507kjImcA+p/II8MVzcmB5rj7KE2gGZJnKvzL/23kM7e+mI/lKjOIvqEIZ/3no9y1bk9827AoWJ9Vt0uoKfXR0Z9qKSRawenwjRpo1pkK2Wa8lsI/Ar8TkSr7fidwRW6WpKRjZEUzWF88+/uUIgqJTfGcys/1u7s1KKdknZ7BMMFwjNbO4evEYMRxHw2f6GvLfBzsTY0pJDZ7TId31JiCikK2GW/20XrgBBGptO/3iMhngQ05XJuSwPDVlCQFmp1ioJGBZsdSSExLPdQ3ROdAGGNMvM22okwWp+VKom9/yN7mNMQDmFMZYH9vMOX5iTU46XCykjKlpA6Go6MGqpWJMaG/ojGmx65sBvhcDtajZCDZUhgONPfZgeayNIFmGG6fbYzhYF+IwXBUJ7IpWcVxFTk+f0hvKcyu9HOgJ7P7KGNMwTMypjB8AVSprS6yzmSkVS8180goKftoONDsjOKsGBFoLrNNbGfQTk8wEn+N9t7UL6aiHC5OplFiY7pgJkuhJ5gyPnO8MQWnojlRaBzXqLa6yB6TEQVtc5FHEusUAgnuo/6hCC5J/qIAlPqTA80HE1IBD/SkmvCKcriksxSGRgSaAeZWBojETEqtwlii4GwPRWOUeN1Jrs9lcyoAeHxL+2QPQ7EZVRREpFdEetL89ALz8rRGBesL4XEJLpeMcB9ZLS5GxggcS8GpY0icetWeJldcUQ6XYJqYwnCdwvApZk6lVauwrzv5osTJrMtU0ex2CU5WdWI6KsCxjZUc11jFTc/sjDeIVCbHqKJgjKkwxlSm+akwxujktTwSjsTw2b5Vn9uFS+yU1GAkbkInMpqloO4jJZs4ApDYg2hkSipY7iOAA70jRWF0SyHxscTXAxARPnDqAl7b38fzOzoP9xCUBDRcP0UIJWRXiEh8JGf/UCQlyAyplkKS+0hFQckig3H3UYKlEInidknSid4Rhf0jgs3jEQUnrjDSUgB4xwmNVAQ83PzMzsM8AiURFYUpQjg6bCkA8elr6TqkwnCGRtxS6B1CBGZX+NVSULLKsPsoEnfhJM5ScKiv8COS6j4KjVGnAMMZSCVpRKHE5+bSk5q4b1ObfrazgIrCFGEoktxauMQetNM3FEnpkApW6qrP7Yqnnx7sDzGr1EdDVUAtBSWrOKIQisbi7VgSp645eN0uasv8Gd1HzjCddDjxhhJvqigAvP8NCwlHDbet2X14B6HEUVGYIoSjJslS8HtdcUth5Hxmh1K/m4GhYUuhrtxPvVoKSpZJHA3rVB0Hw7EUUQCYW+VPE2gef0whnaUAsGR2OacfUcstz+4iGtOA82RQUZgihCLRFEthyIkp+NKLQpnPEy9uO9g3RG25j/qKgIqCklWcQDMMp6UGI1H83tTTy5yKQNZjCg7vO2UBe7oGWbe7a9xrV1JRUZgihKMm3kIYsAPNsbTzmR1Kfe6E7KNQ3FI41D9EJBpL+xxFmSiJloKTljoUjsZnKSQy2y5gS2Q4JXUcloI3c9Lj4jprxMshTbmeFCoKU4RQmpjCQChCXyhCRZqYAlhpqU5M4VDfsPvImMwTsBRlogwmuY9sSyEcSymoBKuA7VB/KF6MCcl9vTLhXBCV+DKfsqpKrNTs7jGqm3/x2Fa2H+wfdZ+ZjIrCFCE0ouFXwOuym9uljuJ0KPNZMYXBkNXvqK7Cx+wKq4BIg81Ktkh2HzkxhdRAMwwXsCUGm8fjPvK4HPdRZkuhchyi0BMM82/3vcIfn9eAdCZUFKYIoUhsRKDZzUH7xJ7ZfWRZCk6NQl2ZZSmAFrAp2SMYicbrYpzGdEORGP40XU/nVKXWKoTG4T7yZSheS6TC70Fk9D5ITq+wnYfUUsiEisIUIRxNdR/12plF6eoUwOqcOhCKDItCgqWgoqBki2Aoymy7MG04+yiDpVDhiEKCpZDQ1ysTjvtotECzyyVUBryjWgpOg8gdhwYy7jPTKZgoiIhbRF4UkXvs+7NE5EER2WL/rinU2oqRkZZCor82syh46B+KctDue1RX7qeuPNV8V5TJEIxEmVXmw+2SeKA5GEkvCnOr0ohCfJ7CaHUKTqA5syiAFVcYTRSc9e081K+9kjJQSEvhM8DmhPvXAg8bY5YCD9v3FZuRQ0QSvxyjxhQSLYVyPwGvm8qARy0FJWs4QeWKgGc4JTVDoLmm1IvP7WJfOlGYRJ2Cw9iiYK1vIBTVxpAZKIgoiEgTcCHwq4TNFwM32rdvBC7J87KKmlRLYfjLkclSKPV5GAhF44NNZpX5ACstUL8QSrYYDEUp8bqpDHiHLYVwNGmWgoOIpAzbcWIKHldmS8GXJUuhL2ES4U51IaWlUJbCj4EvAYnJ8nOMMW0A9u/Z6Z4oIp8UkTUisqa9feb0UA9FzYjso+Q5telwGuXt7hygIuCJP6e+PP0ELEU5HKxCNTeVJZ54SupQhopmsBrjJVY1R+x42WgjYp101dFiCjAOUUjo5LpD01LTkndREJGLgAPGmLWH83xjzA3GmBZjTEt9fX2WV1e8hCLRpGyOwDjcR0763q5DA9TbsQSwxiKqpaBki6FwjIDHTYXfS89gmGjMEIqmdx+BlZa6f0RKqmeUGgUYv/uossRL92Dm0ZyJYzvVUkhPISyFM4B3iMgO4FbgHBG5GdgvIg0A9u8DBVhb0RKOmqTinvEFmq0v0M6OfmrLffHtjqWggTYlGwyGo5T4XFSWeOgNRhiKpM5SSGROZSDJUg2PsILTEe+SOob7qLLEQ89gOONnu3coggjMn1XCDk1LTUveRcEYc50xpskY0wy8D3jEGPMB4G7gCnu3K4C78r22YiY0onW28+XwuCRtPjgMWwr7e4biWUdgtTAeDEfj1c6KMhmCdkuLioCXnmB4eOpahs/lnMoAfUORuH9/ZGFmOoZ7H40+26uqxEsoGksqqEukLxih3OdhUV25WgoZKKY6hX8HzhORLcB59n0FiMYM0Vj6mEJ5IHUUp0OiBVE3wn0EWqugTB5jTLwmwQo0R9JOXUtkbmVyWmo4EsM3pvto9NbZDmO1uugNhikPeGiuLWWHpqWmpaCiYIz5uzHmIvv2IWPMucaYpfbvjkKurZgY7jefWqeQqUMqJAflkiyFcnssYo/WKiiTIxSNETOWr78i4KFvKBJvwphJFJyLkv12sDkcjcXdQ5mYSEoqZBYFZyjVwtoyeoMROgdG75M0EykmS0HJQCiaWvHpfOEyZR5BcgA6KabgVDVrsFmZJI6bxu9xxXsPtfdaxZKZAs1xS6HXEYVxxBSyKAoVtqUAaFwhDSoKUwCno2S6OoVMQWbIbCloqwslWwwluIoq7QsU52LDP0qgGWBft7XfYDg6pigc31TFyc01lE7afRShPOBlYa3VZlt7IKUyetRGKQrSVXw6vtVM6aiQ7Fqqrxi2FKpLvXjdop1SlUnjtM0u8brjn0XHLZkpAaLM76HC7+HFXZ18/MY1PPLKAS46vmHU9zn3qDmce9ScMdcznphCY3UJ82eVIALbD2qweSQqClOAUJqGYYmB5kyU+tNbCiJCfbmO5VQmTzzTyC5eg2ELdLSOprMr/fzt5f2U+z18+fzlfOSM5qysZ7zuI7/HzbyqErUU0qCiMAUYbhiWGmguHyXQ7HO78LiESMxQmyAKYMUV1FJQJstwppGLyoATU7BFIU2bC4cPnrqQHYcGuOrsJfEYVzaoCIwhCsFI3OXaXFeq3VLToKIwBRhKYymUjMNSEBHK/B6GEvrdO9RX+NndMZiD1SoziUT3UVwU+hxLIXOc4MNnLMrJetwusRvzpYpCNGboD0Xj35mFtWXcu7EtJ+uYymigeQrgzLD1jZjRDKPHFMDqlFpX7k+pZVgxv5pX9/fqWEJlUjiWgt/rjmfCjcd9lEsy9T/qGzF/pLm2lK6BMF0DOpo2ERWFKcBwTGH4SxbwuvmXC4/ikhXzRn1uqd+TFE9wuOzk+Xhcwu+f2ZndxSoziuGYgisuCgeKXBQci2Y4A0ldSImoKEwBHFEYOdj8429czOL68lGfu6S+nKPnVaZsn10R4Pxj53L72lYGtd2FcpgEE9xHHreLMp+bjv7R6xRyTUZRsJvhOe6jZlsUtFYhGRWFKUC6iubx8v8+eBLfveTYtI998NSFdA+G+cuGvRmfH4tpGwAlMyNbWjiBXhg90JxLMomCM+vBcR8tdArYNC01CRWFKcBQZOzJVKORqTfSKYtmsWxOOTdncCHdv6mNk77zYFLve0VJZKQoOGmpPrcL1yhDc3JJRlEYSrYUAl43cysD7OxQSyERFYUpgGMpZCoGOlxEhA+eupANrd2s392V8vjGPd10DoT5xeqtWX1fZfowaMcUSkZYCv4CuY5gbPdRZULG3oLaUnZpTCEJFYUpQGiSlsJoXLKykTKfm5vSWAtttoVwy7O7ONCr1oKSSjz7yL5gqUy4Ci8UlSVeQpFYfG0Ow9lHwy6uhbNK2dmhopCIisIUYDIxhbGoCHi58PgGHti0L6WN8L7uIPOqAoSjMX71+Pasv7cy9QlGovg8w64ix1IoVJAZMlc1x2MKCZbCwtpS2nuH4p1dFRWFKUEoTe+jbLJsTgW9Q5GUL9G+7iArF9Rw8YpGbnp6J4e0q6oygmAomjTjwIkpFCrIDJlFoS9oTV1LbKq3wM5A2qXWQhwVhSlAui6p2aSppgSA1s7hCmdjDG3dQeZWBbjq7CUEI1F+/YRaC0oywXDyLOZhS6H4RKHXnqWQGABv1gykFFQUCsiW/b1c9fsX4jNtM5FunkI2aay2vhh7uoZFoWcwwmA4SkNVgCWzy7no+Hnc+NQOunUoiZJAMBJNEgCnMCzbSRETIS4KA6mWQsWIDgALZzmWgmYgOagoFJDHtxzkrxvb2Hpg9A9kOOK0uciRKNiWwp4ES2Fvt3W7ocp67MqzFtMfinL3+j05WYMyNRnM5D4qRkshGEnpFVZV6qWqxKtVzQmoKBQQp+dK4hV6OkLRKG6X4M5R3ndNqZcSrztpHU5twtwqayDKsY1VHN1QyW1rWnOyBmVqEozEkobpFHOg2Wqb7U3Zf2FtqcYUElBRKCBd9od27xiiYI0rzF0hkIjQWFNCa+fwF8NJR22wRQHgspYmNu7p5uW9PTlbizK1CIajBBIsWCclNdPUtXxQOUZMYSQLZpWqpZCAikIBcYaGj2kpRGI5iyc4NFaXjLAUBnEJSb3uL1nZiM/t4rY1u3O6FmXqEAxHk+Ymxy2FAmYfuV1Chd+TNiU1Xav5hbWl7OkajKd+z3RUFApI3H3UOZb7KJazeIJDY01J0jrauoPUV/iT0mCrS3285Zg5/N+6PWMGx5WZgWUpDAtAVTymUNhTS2WJN2WmQrpAM1jB5mjMjGmxzxRUFApIl20ptBaJpdA5EI4X8ezrCTLXDjInclnLfLoGwjz08oGcrkeZGhRjSiqkb3XhjOIcidMYT11IFioKBaTTthTGjinEkkZx5oKmERlIbd1BGioDKfudsaSOeVWBcbmQNu3p1rbc05zBEe6jyiIINEOqKERjhoFQNKnFhcPwXAVNSwUVhYLi5FG39w6l9GlJJF+WAgxbLfvswrWRuF3CpSc1sXpL+6hfovbeIS7+2ZPc8tyu3CxYKQqC4Sh+T+LwJxcfOHUBq46cXcBVpYrCyFkKicyu8OP3uNRSsFFRGCcH+4ayetUbjsboHYrEKyrbRmlPHY7GctbiwiGxVqE3GKZvKJKUeZTI5W9YQLnPw+dvW08kQ3Bu3e4uojGjV1/TnKFwLMlVJCJ855LjOLl5VgFXlSoKvUPW7XQxBZdLrAwkTUsFVBTGzbt+/hQ/fPDVrL2eE084Zl4VMHqweSiS+0Dz7IoAHpewp2swXqPQUJ0aUwCroO077zyWNTs7+emjr6fd58VdncDoYqdMbaIxQygaSypeKxaqSkdYCnaH1HQxBbBrFdRSAFQUxkX/UIRdHQNs2pO9/PzuQSue4IzKHC2uEI7m3n3kdgkN1QH2dA6yN02NwkguXtHIO1c2cv3DW1izoyPl8Rd3dQHQ1q0ZHdOV4QE7xXcaqSrxMpTQPrt3FPcRwIJZZezqGMAYgzGGb979Ejc9vSNfyy0qiu+/WYTstou6th/MnivEqVE4qqECl4yegRTKg6UAw7UK++wT+dw0geZEvn3xMTTWlPCZW9fRE0wO6m1o7QKgrUsthenKyKlrxYRTwOakpcZjCmncRwDNdaUMhqO09w5x25rd/PapHfxlQ1t+FltkqCiMA8es3NcTzFrfdcd9VFfuZ05lYFT3Ua4rmh2aakrZ0zkYd/nMGUMUKgJefnTZCvZ0DXLnC8M9kbYc6KU/FGVxXRmH+kOjBtGVqcug/X8tSvfRiKrm3jHcRwtmWbG9x7cc5Ft/eRlgxo6hVVEYB7sTTtjZarHrpKPWlPrsK/TMr5tPS2F/b5BdHQPUlfvH9Z4tzbNYOrucezcOX1U5rqMLjmsAYH/PzPxyTXeC9ijOQo7ezMRIUXAshXS9j2A4LfWr/7cRt0t4xwnz2NcTTBk8NRMovv9mEbI7ISshWy4kp5q5utTLvOoS9o7iZslH9hFYGUjGwLpdXaPGE0ZywXENPLejIz6yc92uLqpLvZy6uBZg1GNTpi7F7D5yRMGxyONT1zK4jxqrS3CJJXTfueRYVi6oJhSJxd28MwkVhXGwu2Mgnjq6/WBfVl6zayCMxyWU+z001pTQ1j1ILJb+qiQf2UcATXa20baD/WlrFDJxwXENGAMPvLQfgBd3d7JifjXzqq3X0GDz9CRYxO6jRXVluATW27GtviF76pov/Vp9HhcnzK/m0pOauHhFY/yiaCZ+dlUUxsGujgGWzalgbmWAbVmyFDoHwlSXeq0OpdUlhKOGA73px13mI/sIhmsVYPTMo5Esm1POEfVl3Luhjd5gmC0H+lg5vyY+i0HTUqcnjvuoWC2FE+ZX8/iWg4A9S8HvQSRzbO5P/3g6/3np8QDxFi8zMa6gojAGxhhaOweZP6uURXVl7MiSKHQPhuImbrxwLENcIR8N8cCqP3C+MxOxFESEC49r4Nnth3jklQMYAysXVFPic1Nd6p2RV1szgcEiTkkFeOOSOja0dtE9YBVjVmaIJzi4XBIXjWFLQUUh54jIfBF5VEQ2i8hLIvIZe/ssEXlQRLbYv2vyvbZ0HOwLMRiOsmBWKYvqy7IWU+jsD1NT6gOGW0zssX3vm/Z0c+K/PsimPd2AFWjOR0zB53Ex226VPRFLAeBtxzUQM/CDv70GwAnzq+3XKdG01GlKMbuPAM5cWk/MwNPbDlptszPEE9JRV+7H7RK1FPJEBPi8MeYo4FTgKhE5GrgWeNgYsxR42L5fcJyJTPNnlbC4rozOgTCd/aFJv27nQIjqkaJgZzn9z2Nb6egP8Qe7b1A4T5ZC4lrmVqavZs7E8rkVLK6zCoCOqC+LW0HzqgLxYjhlelHMgWawrNUyn5vHtxykbyh1FOdouF3C7Ao/+2Zg5lzeRcEY02aMecG+3QtsBhqBi4Eb7d1uBC7J99rS4Uwjm19TSrOdtrY9C/18ugetmAJAmd9DdamXPV0D7O4Y4L6Nbfg8Lv6yfi/BcNSuU8iTKNRYAXUnSDxeRIS3HTcXgJULho28uVUBdR9NUxxRKMaUVACv28VpR9TyxOsHrVkKExAFsD67ainkGRFpBlYCzwJzjDFtYAkHkLbNooh8UkTWiMia9vb2nK/RKVxrqrHcR0BW4gqdAyFqSod9nPOqrCE3v35iO26X8J1LjqUnGOH+TfsA8OfJUlg4qxS3S8YsXEvHRcfPA+Dk5mFRmFddQtdAWFtoT0OcQHOxuo8AzlxSx85DA7x+oG9C7iOwXKgz8YKmYKIgIuXAn4DPGmPG3VTIGHODMabFGNNSX1+fuwXa7O4coL7CT4nPzfwa64Q52bhCMBwlGI7F3UdgBZtf29/HbWt2844TGnn3iU3MqfTHXUj5qGgG+MgZzfzuo6cclkvgqIZK/nr1mbz7xKb4tpmc2jfdKXb3EVhxBYD+UHTilkJlCW3dM6+ArSCiICJeLEH4vTHmz/bm/SLSYD/eABTFaK9dHQPxEnifx8X8mpJJp6U6BTXVCZaC03doIBTlE2ctwu0SLlnZyLPbrWZz+UhJBagt93PGkrrDfv4x86rwJKxV01KnL4PhKG6X5M21eTgcUV8WvzCZqKUwt8rPQCgab5ExUyhE9pEAvwY2G2N+mPDQ3cAV9u0rgLvyvbZ07O4YZH5C/n5zXRnb2ycnCoktLhycyWdvXFrH8rlW59R3rRy+4s715LVc4Xwhdf7t9CMYLs622YmICGfaFzmZWlxkwqlV2D/DLmgKcaY5A/ggcI6IrLN/LgD+HThPRLYA59n3C0o4GqOtezBuKYBVKbnjUP+kTMq4pVAy/CFdbMcrrjzriPi2I+dWcGyjJRD5shSyjVPvMBMDdtOdYCRatDUKibxxmeVCOpyYAsw8K3dif6UsYIx5AsjkID83n2sZi71dg8QMNCWIwuK6MgZCUQ70Dh1WMBYS+x4NWwqrls3mb9ecxbI5FUn7vvvEJjbteTlvKanZJuB1U1vm07TUaUgwlDyKs1g5a2kd82eVsLyhYuydE3Bax8+0C5q8i8JUwqlRSLYUygHY1t5/+KJgd26sKRu2FFwuSREEgHeubOSJLQc5oan6sN6rGBhvWuqdL7ay89AAV551RNIweKU4CUaiU+L/VF3q4/EvnTPh5znf75lmKUzNy88c0RsM88bvP8LNz+wErHgCwPwEUWiucxrjHX5cwYkpVJf4xtjT+kD/+sMn01xXdtjvV2jGW9X880e38uOHtvCWHz/GY6/lPt1YmRzBcGxKuI8OF5/HRV25j309MyseNn3/o4fBgy/vZ3fHIN/+y8tsbuthd+cAXrckTSCbV1WCz+OaVLfUroEwfo9rSlxlZYN51WNbCtGYYeehAVYdWY/X7eKK3zzHVb9/ga3t2elKq2SfwVCUwBRwH02GmVjApu6jBO7Z0MacSj8xA1f/4UUW1pbSWF2C2zUcAnG5hMV1Zby6fzKiEEpKR53uNFSV0BOM0D8UoSxDsG9v1yChaIzzj5nLO09s5H/+vpVfPLaN+za1ccnKRq5587Iki00pPMFIdMLB26nG3MqSeFeDmYJaCjZdAyEe39LOxSsa+cF7TmDLgT4e2nwg7Ylo5YIaXtzVmXH+wVh0DoST0lGnO+MpYHPccYvqyvB73Hz2zct4/Mtn87EzF/HXDW1c9ouniURjeVmvMj6C4diUCDRPhoaqQF76Hx3sGyqaIjkVBZsHXtpHOGq46PgGzlpWz8fPXASQVhROWlhDbzDC64fp2ugeCM8wS8GpVcj85YqLQv1w7KSu3M9XLzyaH162grbuIM/ZhXxKcRAMT42U1MkwtyqQ8zYt63d38YbvPcwda1tz9h4TYXr/RyfAPRvaWDCrlOMaqwD44vlH8q6VjVxozxlO5MQF1QC8sLPzsN6rcyA0riDzdGGe3Xn1qa2H+OkjW/iHXz7DXev2JO2z/WA/ZT439eX+lOefs3w2JV439yTMgS4ET209yNqdKkwOwXC06IvXJks8LTWH1sL3H3iFaMxww+ptRWEtzEhRCEdj/OjB1+izy9cP9g3x5OsHefsJDfEhG36Pmx++d0Xalg+L6sqoKfWy9rBFIZyUjjrdmV3pxyXw/x7byn/97TXW7Ojk9jXJV0XbDvazqL4s7WSsEp+bc4+azQOb9hXUhfTF2zfwxds3FMUX93C5d2Mb/3rPy1l5LctSmN6i0JDj4ssnthzkydcPcUrzLLYc6ItPiiskM1IUnt/RwU8ffZ1LfvYkW9v7uG/TPmJmuMvnWIgIJy2sYe2uiYuCMcaeujZzLAW/x82P3ruCH152Amv+5c1c2tLEhtaupJPrjoP98RqQdFx0fAOH+kPxXlD5prVzgD1dg2w72M9rk0gyKDS/eWI7v3lyO92Dkx9IPzhD3EdATtJSjTH8x/2v0Fhdwm8+cjL1FX5+/cT2rL/PRJne/9EMnH5EHTd97BQ6+kNc8tMn+dXj21gyu5zlc8df8Xjiwhq2tfdPeOBOf8iaj1Azg2IKABevaORdJzZRV+7n+MYqeoIRdtptyYciUVo7B1hUmzm7aNWRsyn1ublnQ2FcSM/vGBaje0e4sWIxc9hJB+MhFInxhdvXH7Zl6tA/FGHd7i6MgXW7uyb1WsaYKdH7aLLMzUKrC2MM0ZghHI0lWbr3bdrHxj3dXHPeMsr9Hj506kIee62d1w/0Tnrdk2FGigJYwvCXT5/Jovoydh4a4KLjG0Yd6j2SE+1BMi/untgXtStNM7yZxvF2dfb61i4AdncMEDPJQeaRBLxu3nzUHB54qTAupOe2d1AZ8HBycw33bUoWha/cuZGL/vuJwxKGV/f18qb/fJTXD2S2Pm56Zid3rG3lJw9vmfDrJ/L8jg4i9honKzBDEet/4J/molDq81AZ8ByW+6hrIMRX7tzI0q/exxFfuZelX72PZf9yH2/50WN84fb1/Pt9r7BsTjnvXNkIwD+8YQE+j4v/fXJHlo9iYsxYUQCrXfVtV57G9999PB9/4+IJPfeEpmrcLpnwl8tphlc1wyyFRJbOKcfvcbGx1ZpBvf2gZTGM5j4CuPD4Bjr6QzyzLbMLqbM/xFNbD3LPhr0pJ+lINMYtz+7iUN/QhNf87PYOWppncdHx83htf1/8au7Vfb38cc1uXm7r4YnXJ+4P/t3TO9h5aIDfPpXebdA9EOa/H9mCz+Pi8S3t7O44/Jz5p7cewusWjqgvm3TAfCrMUsgWDVUlY1oKv31yOyu//Tc+dfNabn1uF394bhfn/OAx/vj8bi49qYnPnbeML7xlGf+0agmN1SU8+soBdnUMcO3blsfroGrL/bxrZSN/eqE1KyN/D5fpXXkyDgJeN5edPH/CzyvxuTlmXiUv7Oya0PMcUZjJloLX7eKYeZVsiIuCdZW8qHb0Vh5vWlZPmc/NXzfu5cylyQkAd77Yyn/e/2pS471ffNDFW4+ZG7//wEv7+cqdG7n/pX3c+JGTx20ZHuwbYlt7P5e1zOf8Y+fyjbtf4r6N+/j0uRX84G+vUu7z4PO4uPGpHZy1bPyDn4LhKHev34sI3PnCHq5921EpxWA/+/vrdA+G+fUVLXz8xjXc+vwuvvjW5eN+j0Se2nqIlQtqWD63gj+tbSUSjSXNvpgIU2HqWrZYOqecx7ccpCcYpjJN++3eYJgfP7yF6hIvL+7q4j57WuJJC2v414uP5eh5lSnPMcbQNxRJaef9kTMWcevzu7n5mZ18+tyluTmgMZjRlsJkOXFBDet2d03InRHvezSDLQWwXEib9nYTjRm2H+xnVplvTOsp4HVz3tFzuGvdXn739A4i0RixmOH797/CNX9cz5yqAF+5YDm/++gpzKn0c6s9tc7hj2t243ULq19r53dP7xz3Wp+3g9unLJrFnMoALQtruHfTPtbv7uJvL+/nE2ct5h/esIBHXj0QH986Hh54aR+9wQifP28Z/aFoSpru7o4BfvvkDi49sYlzls/h7CNnc9uaVsKH4T7rHgjz0t5uTj+ilpMW1tAfivLq/vS+a2MMB8ZIwewNWhc30z3QDFY7++7BML/JEAS+8akddA2Euf7ylTx93Tk88NmzuOljp3D7laelFQSwklXSzXc4cm4F5yyfzW+e3M5AqDDDfab/fzSHnLiwhsFwlFf2ZQ4M9QbDfOmO9bR85yHe9fMn+ZX9wVJRqGIgFGVrex/bD/azaJwN/77w1iM5oamar9/1Em/7yeN87Mbn+fnft3L5KfP54ydP45NnHcFZy+p5z0nzeey19vhwnz1dgzy+pZ1PvekIVh1Zz/fu3TzugN6z2zsIeF0cO8+qYXnbcQ1sbuvhy3/awKwyHx89cxHvf8NCXCLc/Oyw2Ny3sY0v3bGeoUj6wqc71rbSWF3CP61awlENldz8zK6kjKz/fOBVXC74/FuOBODyUxbQ3jvEw5snPpTw2e2HiBkrlubEwzK5Pm9+ZienfO9hfvf0joyvd8+GNkSGY2vTmeOaqnjrMXP49ePb4zFBh95gmF8+vp1zl8/m+KZqRIQj51bwxqX1uFyHN0L3qrOX0DkQ5pZnd429cw5QUZgEJy20vhAvZEhNXbuzkwuuf5w71rbyhkWz8HlctHVZQ3tmzWD3EViiALChtZvtB/tpHsN15NBUU8otn3gDv/jgSYSiMR57rZ2vXXQ033vncUkzJ9578nxiBm5bsxuAO9a0Ygy8p2U+37/0eMr8Hj77x3XcsbaVr9y5kYt/+gT3bNib9j2f297BiQtq4q9//rGWS+qVfb186k1HUO73MLcqwPnHzOWPz+9mMBTl9jW7+adbXuC2Na389JHXU15zT9cgT7x+kHef1ITLJXzg1AVsbuvhRTsr6Jert3H3+r184o2L4xkwq46sp6EqEJ/bPRGe2nqIgNfFivnVNNWUMLvCn1YUojHDL1Zvw+MSvn7XS2mvjkORGLc8t4tVy+qndPfeiXDNecvoC0W4YfW2pO03PrWD7sEwn33zsqy910kLazhtcS03rN4Wj93kkxkfU5gM86oCzKn08+cX9nDq4tr4PIRdhwb49RPbuPnZXTRUBbjtytNoaZ4Vf54xZkKZTtORRXXllPncPLPtEPt7huKT58aDiPDWY+ay6sh6DvQMpW1FMn9WKW9cWsfta1r557OXcPva3ZyxpDa+77+96ziuvGktX7h9PRV+D6V+N9f9eSMnLayJz5UG6AmG2byvh88k+Hcbq0s4cUE1e7oG+eBpC+PbP3TaQv66sY2rbnmBR145wBuX1lFT6uPnf9/KW46ey3G2EAL8ea0tUidZI1cvXtHI9/66mZuf3sn9m/Zxw+ptXHh8A/98zpL4czxuF5e1zOf6R7awu2NgQg0Cn956iJObZ8WFraW5Jq0o/O2lfbR2DnL95Su5d0Mb377nZaIxwyfOGk7EuG9TG+29Q3zo9OZxv/9UZ/ncSt5+/Dz+98kdfPTMRdSV++NWwpuPmp30v80G/3zOEt7/q2e5Y20rHzh14dhPyCIqCpNARLjq7CV8797NvOVHq3nj0jrK/R4eeGkfbpdwWct8vnLB8hTf4UwXBAC3Szi2sYoH7KDceN1Hifg97lFPjO87eQFX3fIC33/gVVo7B/niW4+MP/bWY+Zy+z+eRlWJlyX15ezqGOD8n6zmuj9v5H8/PByEXrujE2OseEIiP3v/iYQjJin75pRFs1g+t4JHXjnAm4+azU//4USGwjGe3X6IL9y+nrs/fQZ+jxtjDHe80Mqpi2fF11/u9/DOExu5+RnLCrjitIV84+3HpLgg3nvyfP77kS185c6N/PTyE+NxmKFIlJue3klvMMKqI+s5oak6/tz23iFe3d/LJXbqI1hun3s37mN/TzBpWNSvntjO/FklXHhcA287di6fvXUd3713M2V+D//whgUA/O7pnTTXlvKmpeMPqk8HPvPmpdyzYS+funktNaU+trb3Zd1KcDj9iFpWzK/m/z22lUtPauL1A31s3NPNsjkVcQ9FrlBRmCQfOq2Ztx8/j1ue28WNT+0gGI5y5ZuO4IrTmuNmv5Ke45uq4hXK43UfTYQ3Hz2bWWU+bli9jaoSb1ImEsDJCdZbc10ZXz5/Od/6y8vcvraVy1qsjLRnt3fgdQsr5yd/EROtCQcR4dsXH8vq19q5+tyl+DwuAl43//au4/job9fw7b+8zKK6Mp7b3sHOQwNcfU5ydskVpzVz97q9fPKsxVx19pK0Fw/zqkv47juP4+t3beLtP32C//nAiQyGolz75428fqAPEfjJw1uoLfNx1rJ6Vh1ZT49dvXz6EbXx13FOLGt3dnKB3d/rhV2drN3ZyTfefjRul+BG+Mn7VtAfivC1uzbRUB2gvtxyO33toqMP22c+VTmivpwrTm/mjjWtzK0K0FhTygdPXcixjdm1EsD6LP3z2Uv4+O/WcNw3HyAcHY41nbGkls+cuyzlQiVr7z2V+7i0tLSYNWvWFHoZcaIxgzHmsNP8Zhp/Wb+XT//hRQBe/vZbKfVl/xrlu399mV8+vp0rTlvIty4+dtR9YzHD+254hs1tPVx7wXK2tffzl/V7aaop4c//dMak1vH529bzpxesfk+N1SWctayOb7z9mJQ8/1jMjOtku3ZnJ//0+7V09ocJRWM0Vpfw3XceywlN1aze0s6jrxxg9ZaDdNj57hV+Dy9+/bz4ZzMUiXHcNx/gA6cu5GsXHQ3AVbe8wOrX2nn6unOTUmP7hiK89xdPs+NgP8c1VbF+dzfPfOVcqkpmdrJErjHG8K/3bMZgWLmghmPmVfLI5gP8YvVWDvaFePeJTfzgshMO67VFZK0xpiXdY2opZBGrCGVmXT1NBifY3FAVyIkggGXJPbu9Y1z+b5dL+P6lx/O2nzzOV+/cRMDrYtmcCj551sQKG9Px3XceyztWzOOouRXMHmW293ivvk9aWMM9n34jX79rE43VJVxz3rL4AKOLVzRy8YpGYjHDxj3dPPZaOwtmlSZdrPg8Lk5oqubxLe3cv2kfg+EI921s4xNnLU6plSj3e/jNh0/mnT97kme2dfD+NyxQQcgDIsLX33500rYj6sv5wKkL+cNzu6irSO0onJX3VUtBKRTGGFZ8+0GObqjkD588tdDLibPr0ACRWIyFtWVJU/emGz968LWk1hk+j4u/f2FVvNX5SF7d18u/37eZb73jWBaM0qdKKX5GsxRUFJSCcvua3dRX+Fl15OxCL2XGEY7GeM0uYBOEWWU+jYPNENR9pBQt72mZeIsRJTtY7UayHyRVpjYaEVUURVHiqCgoiqIocVQUFEVRlDgqCoqiKEocFQVFURQljoqCoiiKEkdFQVEURYmjoqAoiqLEmdIVzSLSDox/rmIqdcDEp61PbWbiMcPMPG495pnDRI97oTEmbe/zKS0Kk0VE1mQq9Z6uzMRjhpl53HrMM4dsHre6jxRFUZQ4KgqKoihKnJkuCjcUegEFYCYeM8zM49Zjnjlk7bhndExBURRFSWamWwqKoihKAioKiqIoSpwZKQoicr6IvCoir4vItYVeTy4Qkfki8qiIbBaRl0TkM/b2WSLyoIhssX/XFHqtuUBE3CLyoojcY9+f1sctItUicoeIvGL/z0+b7scMICLX2J/vTSLyBxEJTMfjFpHfiMgBEdmUsC3jcYrIdfb57VUReetE3mvGiYKIuIGfAW8DjgYuF5GjR3/WlCQCfN4YcxRwKnCVfZzXAg8bY5YCD9v3pyOfATYn3J/ux/0T4H5jzHLgBKxjn9bHLCKNwNVAizHmWMANvI/pedy/Bc4fsS3tcdrf8/cBx9jP+bl93hsXM04UgFOA140x24wxIeBW4OICrynrGGPajDEv2Ld7sU4SjVjHeqO9243AJQVZYA4RkSbgQuBXCZun7XGLSCVwFvBrAGNMyBjTxTQ+5gQ8QImIeIBSYC/T8LiNMauBjhGbMx3nxcCtxpghY8x24HWs8964mImi0AjsTrjfam+btohIM7ASeBaYY4xpA0s4gNkFXFqu+DHwJSCWsG06H/dioB34X9tl9isRKWN6HzPGmD3AfwG7gDag2xjzN6b5cSeQ6TgndY6biaIgabZN27xcESkH/gR81hjTU+j15BoRuQg4YIxZW+i15BEPcCLwP8aYlUA/08NlMiq2D/1iYBEwDygTkQ8UdlVFwaTOcTNRFFqB+Qn3m7BMzmmHiHixBOH3xpg/25v3i0iD/XgDcKBQ68sRZwDvEJEdWK7Bc0TkZqb3cbcCrcaYZ+37d2CJxHQ+ZoA3A9uNMe3GmDDwZ+B0pv9xO2Q6zkmd42aiKDwPLBWRRSLiwwrI3F3gNWUdEREsH/NmY8wPEx66G7jCvn0FcFe+15ZLjDHXGWOajDHNWP/bR4wxH2AaH7cxZh+wW0SOtDedC7zMND5mm13AqSJSan/ez8WKnU3343bIdJx3A+8TEb+ILAKWAs+N+1WNMTPuB7gAeA3YCny10OvJ0TGeiWUybgDW2T8XALVYmQpb7N+zCr3WHP4NVgH32Len9XEDK4A19v/7/4Ca6X7M9nF/C3gF2ATcBPin43EDf8CKm4SxLIGPjXacwFft89urwNsm8l7a5kJRFEWJMxPdR4qiKEoGVBQURVGUOCoKiqIoShwVBUVRFCWOioKiKIoSR0VBUcZARKIisi7hJ2vVwiLSnNj5UlEKjafQC1CUKcCgMWZFoRehKPlALQVFOUxEZIeI/IeIPGf/LLG3LxSRh0Vkg/17gb19jojcKSLr7Z/T7Zdyi8gv7bkAfxORkoIdlDLjUVFQlLEpGeE+em/CYz3GmFOAn2J1Z8W+/TtjzPHA74Hr7e3XA48ZY07A6k30kr19KfAzY8wxQBfw7pwejaKMglY0K8oYiEifMaY8zfYdwDnGmG1288F9xphaETkINBhjwvb2NmNMnYi0A03GmKGE12gGHjTWoBRE5MuA1xjznTwcmqKkoJaCokwOk+F2pn3SMZRwO4rG+pQCoqKgKJPjvQm/n7ZvP4XVoRXg/cAT9u2HgU9BfIZ0Zb4WqSjjRa9IFGVsSkRkXcL9+40xTlqqX0SexbrAutzedjXwGxH5ItZEtI/Y2z8D3CAiH8OyCD6F1flSUYoGjSkoymFixxRajDEHC70WRckW6j5SFEVR4qiloCiKosRRS0FRFEWJo6KgKIqixFFRUBRFUeKoKCiKoihxVBQURVGUOP8fwRBHmB7yCPAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Loss or Error\n",
    "\n",
    "plt.plot(history.history['loss'])\n",
    "plt.title('Model Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend(['Training'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "62ca6189",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['accuracy'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend(['Training'])\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}