A1-T1-cnn-muliti_run.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "9PUXkk5s8Cdp"
   },
   "outputs": [],
   "source": [
    "import keras\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "fashion_mnist = keras.datasets.fashion_mnist\n",
    "(X_train_full, y_train_full), (X_test, y_test) = fashion_mnist.load_data()\n",
    "#create the validation set\n",
    "X_valid, X_train = X_train_full[:6000] / 255.0, X_train_full[6000:] / 255.0\n",
    "y_valid, y_train = y_train_full[:6000], y_train_full[6000:]\n",
    "#add label for class\n",
    "class_names = [\"T-shirt/top\", \"Trouser\", \"Pullover\", \"Dress\", \"Coat\",\n",
    "                   \"Sandal\", \"Shirt\", \"Sneaker\", \"Bag\", \"Ankle boot\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "kinits=['he_normal','he_uniform','glorot_normal','glorot_uniform']\n",
    "#lrs=[0.05,0.01,0.005,0.001,0.0005,0.0001]\n",
    "drs=[0.1,0.2,0.3,0.4,0.5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "YuzQ5ZdHY6TQ",
    "outputId": "4bfe269e-2ee9-4540-e4f6-b075853e358e",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 27ms/step - accuracy: 0.7346 - loss: 0.7437 - val_accuracy: 0.8923 - val_loss: 0.3004\n",
      "Epoch 2/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.8853 - loss: 0.3171 - val_accuracy: 0.9033 - val_loss: 0.2634\n",
      "Epoch 3/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9054 - loss: 0.2584 - val_accuracy: 0.9087 - val_loss: 0.2484\n",
      "Epoch 4/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9170 - loss: 0.2298 - val_accuracy: 0.9163 - val_loss: 0.2267\n",
      "Epoch 5/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9258 - loss: 0.2035 - val_accuracy: 0.9080 - val_loss: 0.2436\n",
      "Epoch 6/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9309 - loss: 0.1861 - val_accuracy: 0.9168 - val_loss: 0.2377\n",
      "Epoch 7/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9380 - loss: 0.1645 - val_accuracy: 0.9152 - val_loss: 0.2326\n",
      "Epoch 8/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9440 - loss: 0.1492 - val_accuracy: 0.9187 - val_loss: 0.2464\n",
      "Epoch 9/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9493 - loss: 0.1325 - val_accuracy: 0.9128 - val_loss: 0.2860\n",
      "Epoch 10/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9514 - loss: 0.1281 - val_accuracy: 0.9200 - val_loss: 0.2347\n",
      "Epoch 11/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9554 - loss: 0.1207 - val_accuracy: 0.9182 - val_loss: 0.2715\n",
      "Epoch 12/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9609 - loss: 0.1043 - val_accuracy: 0.9178 - val_loss: 0.2643\n",
      "Epoch 13/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9634 - loss: 0.1033 - val_accuracy: 0.9198 - val_loss: 0.2854\n",
      "Epoch 14/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9641 - loss: 0.1021 - val_accuracy: 0.9243 - val_loss: 0.2881\n",
      "Epoch 15/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9647 - loss: 0.0958 - val_accuracy: 0.9200 - val_loss: 0.3015\n",
      "Epoch 16/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9660 - loss: 0.0977 - val_accuracy: 0.9182 - val_loss: 0.3004\n",
      "Epoch 17/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9698 - loss: 0.0819 - val_accuracy: 0.9122 - val_loss: 0.3713\n",
      "Epoch 18/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9704 - loss: 0.0850 - val_accuracy: 0.9192 - val_loss: 0.3059\n",
      "Epoch 19/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9716 - loss: 0.0814 - val_accuracy: 0.9148 - val_loss: 0.3564\n",
      "Epoch 20/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9743 - loss: 0.0726 - val_accuracy: 0.9163 - val_loss: 0.3568\n",
      "Epoch 21/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9741 - loss: 0.0747 - val_accuracy: 0.9230 - val_loss: 0.3572\n",
      "Epoch 22/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9721 - loss: 0.0807 - val_accuracy: 0.9232 - val_loss: 0.3449\n",
      "Epoch 23/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9748 - loss: 0.0756 - val_accuracy: 0.9232 - val_loss: 0.3675\n",
      "Epoch 24/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9765 - loss: 0.0709 - val_accuracy: 0.9220 - val_loss: 0.3178\n",
      "Epoch 25/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9772 - loss: 0.0661 - val_accuracy: 0.9225 - val_loss: 0.3973\n",
      "Epoch 26/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9809 - loss: 0.0555 - val_accuracy: 0.9145 - val_loss: 0.4254\n",
      "Epoch 27/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9743 - loss: 0.0784 - val_accuracy: 0.9200 - val_loss: 0.3584\n",
      "Epoch 28/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9806 - loss: 0.0612 - val_accuracy: 0.9140 - val_loss: 0.3800\n",
      "Epoch 29/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m52s\u001b[0m 31ms/step - accuracy: 0.9762 - loss: 0.0737 - val_accuracy: 0.9220 - val_loss: 0.3850\n",
      "Epoch 30/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 27ms/step - accuracy: 0.9808 - loss: 0.0605 - val_accuracy: 0.9147 - val_loss: 0.5035\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/caroldong/anaconda3/lib/python3.11/site-packages/keras/src/layers/convolutional/base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.7291 - loss: 0.7365 - val_accuracy: 0.8862 - val_loss: 0.3137\n",
      "Epoch 2/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.8793 - loss: 0.3263 - val_accuracy: 0.8968 - val_loss: 0.2803\n",
      "Epoch 3/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9010 - loss: 0.2695 - val_accuracy: 0.9017 - val_loss: 0.2578\n",
      "Epoch 4/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9144 - loss: 0.2349 - val_accuracy: 0.9030 - val_loss: 0.2479\n",
      "Epoch 5/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9241 - loss: 0.2084 - val_accuracy: 0.9123 - val_loss: 0.2457\n",
      "Epoch 6/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9334 - loss: 0.1854 - val_accuracy: 0.9115 - val_loss: 0.2380\n",
      "Epoch 7/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9339 - loss: 0.1776 - val_accuracy: 0.9142 - val_loss: 0.2355\n",
      "Epoch 8/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9424 - loss: 0.1547 - val_accuracy: 0.9182 - val_loss: 0.2490\n",
      "Epoch 9/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9454 - loss: 0.1496 - val_accuracy: 0.9138 - val_loss: 0.2662\n",
      "Epoch 10/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9517 - loss: 0.1293 - val_accuracy: 0.9200 - val_loss: 0.2729\n",
      "Epoch 11/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9565 - loss: 0.1172 - val_accuracy: 0.9168 - val_loss: 0.2965\n",
      "Epoch 12/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9589 - loss: 0.1090 - val_accuracy: 0.9228 - val_loss: 0.2696\n",
      "Epoch 13/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9634 - loss: 0.1010 - val_accuracy: 0.9178 - val_loss: 0.3117\n",
      "Epoch 14/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9668 - loss: 0.0913 - val_accuracy: 0.9170 - val_loss: 0.3034\n",
      "Epoch 15/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9693 - loss: 0.0859 - val_accuracy: 0.9135 - val_loss: 0.3954\n",
      "Epoch 16/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9684 - loss: 0.0894 - val_accuracy: 0.9195 - val_loss: 0.3058\n",
      "Epoch 17/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9691 - loss: 0.0847 - val_accuracy: 0.9175 - val_loss: 0.3311\n",
      "Epoch 18/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9715 - loss: 0.0755 - val_accuracy: 0.9183 - val_loss: 0.3204\n",
      "Epoch 19/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 25ms/step - accuracy: 0.9730 - loss: 0.0738 - val_accuracy: 0.9133 - val_loss: 0.3786\n",
      "Epoch 20/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9746 - loss: 0.0717 - val_accuracy: 0.9120 - val_loss: 0.3640\n",
      "Epoch 21/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9776 - loss: 0.0649 - val_accuracy: 0.9140 - val_loss: 0.3902\n",
      "Epoch 22/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9764 - loss: 0.0697 - val_accuracy: 0.9118 - val_loss: 0.4196\n",
      "Epoch 23/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9765 - loss: 0.0679 - val_accuracy: 0.9158 - val_loss: 0.3827\n",
      "Epoch 24/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9791 - loss: 0.0602 - val_accuracy: 0.9198 - val_loss: 0.3737\n",
      "Epoch 25/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9831 - loss: 0.0496 - val_accuracy: 0.9183 - val_loss: 0.4227\n",
      "Epoch 26/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9800 - loss: 0.0576 - val_accuracy: 0.9238 - val_loss: 0.3381\n",
      "Epoch 27/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9825 - loss: 0.0513 - val_accuracy: 0.9163 - val_loss: 0.4415\n",
      "Epoch 28/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9809 - loss: 0.0577 - val_accuracy: 0.9140 - val_loss: 0.4763\n",
      "Epoch 29/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9784 - loss: 0.0686 - val_accuracy: 0.9185 - val_loss: 0.4094\n",
      "Epoch 30/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m43s\u001b[0m 26ms/step - accuracy: 0.9848 - loss: 0.0491 - val_accuracy: 0.9157 - val_loss: 0.4541\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/caroldong/anaconda3/lib/python3.11/site-packages/keras/src/layers/convolutional/base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m47s\u001b[0m 27ms/step - accuracy: 0.7307 - loss: 0.7310 - val_accuracy: 0.8742 - val_loss: 0.3242\n",
      "Epoch 2/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.8834 - loss: 0.3252 - val_accuracy: 0.8907 - val_loss: 0.2875\n",
      "Epoch 3/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.8981 - loss: 0.2738 - val_accuracy: 0.9027 - val_loss: 0.2703\n",
      "Epoch 4/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9113 - loss: 0.2418 - val_accuracy: 0.9057 - val_loss: 0.2453\n",
      "Epoch 5/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9210 - loss: 0.2136 - val_accuracy: 0.9105 - val_loss: 0.2475\n",
      "Epoch 6/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9264 - loss: 0.1960 - val_accuracy: 0.9200 - val_loss: 0.2278\n",
      "Epoch 7/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9337 - loss: 0.1723 - val_accuracy: 0.9195 - val_loss: 0.2321\n",
      "Epoch 8/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9406 - loss: 0.1588 - val_accuracy: 0.9188 - val_loss: 0.2287\n",
      "Epoch 9/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9447 - loss: 0.1481 - val_accuracy: 0.9195 - val_loss: 0.2442\n",
      "Epoch 10/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9489 - loss: 0.1332 - val_accuracy: 0.9215 - val_loss: 0.2419\n",
      "Epoch 11/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9553 - loss: 0.1220 - val_accuracy: 0.9185 - val_loss: 0.2734\n",
      "Epoch 12/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9572 - loss: 0.1150 - val_accuracy: 0.9150 - val_loss: 0.3035\n",
      "Epoch 13/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9626 - loss: 0.0994 - val_accuracy: 0.9207 - val_loss: 0.2847\n",
      "Epoch 14/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9647 - loss: 0.0967 - val_accuracy: 0.9222 - val_loss: 0.3199\n",
      "Epoch 15/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9651 - loss: 0.0975 - val_accuracy: 0.9218 - val_loss: 0.3259\n",
      "Epoch 16/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9695 - loss: 0.0872 - val_accuracy: 0.9175 - val_loss: 0.3641\n",
      "Epoch 17/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9703 - loss: 0.0798 - val_accuracy: 0.9155 - val_loss: 0.3639\n",
      "Epoch 18/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9696 - loss: 0.0867 - val_accuracy: 0.9262 - val_loss: 0.3578\n",
      "Epoch 19/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9726 - loss: 0.0798 - val_accuracy: 0.9230 - val_loss: 0.3069\n",
      "Epoch 20/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9753 - loss: 0.0684 - val_accuracy: 0.9207 - val_loss: 0.3456\n",
      "Epoch 21/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 26ms/step - accuracy: 0.9743 - loss: 0.0754 - val_accuracy: 0.9203 - val_loss: 0.3410\n",
      "Epoch 22/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9760 - loss: 0.0694 - val_accuracy: 0.9172 - val_loss: 0.3877\n",
      "Epoch 23/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9773 - loss: 0.0672 - val_accuracy: 0.9213 - val_loss: 0.3962\n",
      "Epoch 24/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9798 - loss: 0.0625 - val_accuracy: 0.9202 - val_loss: 0.3383\n",
      "Epoch 25/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 27ms/step - accuracy: 0.9770 - loss: 0.0675 - val_accuracy: 0.9252 - val_loss: 0.3364\n",
      "Epoch 26/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9785 - loss: 0.0650 - val_accuracy: 0.9210 - val_loss: 0.4549\n",
      "Epoch 27/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9811 - loss: 0.0561 - val_accuracy: 0.9135 - val_loss: 0.5036\n",
      "Epoch 28/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9802 - loss: 0.0604 - val_accuracy: 0.9250 - val_loss: 0.3941\n",
      "Epoch 29/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9823 - loss: 0.0570 - val_accuracy: 0.9193 - val_loss: 0.4281\n",
      "Epoch 30/30\n",
      "\u001b[1m1688/1688\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 26ms/step - accuracy: 0.9796 - loss: 0.0716 - val_accuracy: 0.9202 - val_loss: 0.4124\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAp8AAAGyCAYAAACiMq99AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAACH7ElEQVR4nOzdd3gUdeLH8fdsTe8FEkKvKgKiIBZEQFE8fljPgor1bNjQU7lT0bNgP+yevYF69gKiiA0F8QRRkSLSQktCCunJtvn9MclCIEACyW5IPq/nmWdmp+x8d4eET2a+xTBN00REREREJARs4S6AiIiIiLQdCp8iIiIiEjIKnyIiIiISMgqfIiIiIhIyCp8iIiIiEjIKnyIiIiISMgqfIiIiIhIyCp8iIiIiEjIKnyIiIiISMgqfIiIiIhIyjQ6f3377LWPGjCEjIwPDMPjggw/2eMzXX3/NIYccgtvtpnv37rz88st7UVQRERER2d81OnyWl5fTr18/nnzyyQbtv2bNGk466SSOPfZYFi9ezHXXXccll1zCZ5991ujCioiIiMj+zTBN09zrgw2D999/n5NPPnmX+9x8883MmDGDJUuWBNedddZZbN26lVmzZu3tqUVERERkP+Ro7hPMnz+fkSNH1lk3atQorrvuul0eU11dTXV1dfB1IBCgsLCQ5ORkDMNorqKKiIiIyF4yTZPS0lIyMjKw2Xb9cL3Zw2dOTg7p6el11qWnp1NSUkJlZSWRkZE7HTNlyhTuvPPO5i6aiIiIiDSx9evX06FDh11ub/bwuTcmTZrExIkTg6+Li4vp2LEja9asITY2ttnP7/V6+eqrrzj22GNxOp3Nfj7Zma5B+OkahJ+uQcug6xB+ugbh15BrUFpaSpcuXfaY1Zo9fLZr147c3Nw663Jzc4mLi6v3rieA2+3G7XbvtD4pKYm4uLhmKef2vF4vUVFRJCcn6x95mOgahJ+uQfjpGrQMug7hp2sQfg25BrXr91RFstn7+RwyZAhz5syps2727NkMGTKkuU8tIiIiIi1Mo8NnWVkZixcvZvHixYDVldLixYvJzs4GrEfm559/fnD/yy+/nNWrV3PTTTexfPlynnrqKf773/9y/fXXN80nEBEREZH9RqPD508//cSAAQMYMGAAABMnTmTAgAHcfvvtAGzevDkYRAG6dOnCjBkzmD17Nv369ePhhx/m+eefZ9SoUU30EURERERkf9HoOp/Dhg1jd12D1jd60bBhw/j5558beyoRERGRFsM0TXwBE5/fxBcI4PObeAMBAAyseo6GAQbb6j0awXU19SCN3e8TCJ4jgD9g4q1Z3vG82+bb9vf6TfyBuvt0T4thYKek0H5Re9AiW7uLiIjIzgIBk0qvn3KPj0qPn0qvH6fdhstuw+204bbbcTlsuBw27Lb9r19s0zSp9gWo8Pgpr/ZR7vFRXu2juKKa3woNjCU5mIYNr9/E6w/UTDXLvprXATO47PHXhjJrP0/Nsq/mmO1D2/Zh0u/fLvTVrKsNevub84d0UvgUERFprXz+AFW+ANVe/7a5N0C1b/u5n/JqPxUenxWyPH4qPT7KPX4qqq111lS73Qqa5dVW2Gwoh83AXRNEXQ4bbkdNMK0Jqtbcbs0d1uSwG9htNhw2A7vNCM63Lduw26izj72efWuDb6XHT1nNZyqv9gWXrbmPsmrrM2+/bdcBzw4rfm2Cq9T0DAP2frzIXav9Lp02A4fd+s4ddgOHrfZaGThrlh01+9htBs7afWwGPdObv4vKxlL4FBGR/UogYFJU4WFLWTX5pR7yy6rJL6tmS2m1ta7MQ1mVF8MwsBnWo00DsBkGNpv1aNMwrNe1cysrWfPt1xsGmAGT9ZtsvJu/kGqfdWeuyuvHUzOv3m4eqjtjhgHRLgcRTusuYLXPKs/2p/cFTHw14XZ/FOG0EeN2EOVyEOW0UVleSmpyIm6nHYfNZt3xdRh1lp12m/XaYeDaYdlpt0Ka024LhjPndkHOWU+4q11Xe2zwmNp9a8Lh9l0L1VZNNE0wt1tnsi2gmph1wqq1r7XCwCqD3TCw7Yd3rxtC4VNERMLOXxMo87cLlFtKa0JlTaCsfV1Y7sEf8sefNigsaNQRwUfhDjsRThsRTjtuhzWPdjuIctqJctuJdjmIctmJcjmIdtuJdNVdF+W2E7XDuginrd6+FH3+ANW+AB7f9nMrGHv8Aaq9tXN/nde1+9XWGaydfAETf8AK1YHg623zbfsFdlofME2iXQ6i3TWTy16zXDMPbrPXXa7Ztn21Aa/Xy8yZMxk9elCL7+czWI+zzuVpnSFybyl8iojsR0zTqvNXVu2jvLqmXlywblzdR5vbr699zOnzN11oM03wm3VDSG1jicCOAcXcti64bbt1eyMp2kVKjIuUGDcpMW5SY901yy7iIp3W3aSaO04B0yRQ+9q0Xu80324/auYB08Tn9/PHsqUM7H8w0RGuYIDcPkzWNw/HXSuH3YbDbiN653FaRFoMhU8RkX3gD5hU1Aa/msYRlZ66d5VqH4luf0dq+7tSO96lCu7j9ZNbYGfqH99RUVNPsNzja5a6ZS3F7gJl7XJqrJukaBdOe7OPkwLU3HUr+p3Rh2S2+LtuIvsDhU8RaTNqg+L2DToqa+rEVVRbDT62b2FbeydxpwYS221vTAOQvWNAecXOa2vq/O30yLKex5wxbuvRbYzbgcPWdIGttl5kbb03W00dOJth1VmzGdsaotQub984xWaz6rXVrouNcIQsUIpI+Ch8ikiL5/MHKCz31Kn7t7XCQ6XHT4V3uxbCXn9Nq2AflV4rYFbWtBau8Fh3FJuL3WYQ7bICXoTTalXsdtpx22tbGu84t+/htQ07Jr8uXsQxRw4mPioiGDBj3A4infY9jp8sItISKXyKSFj4AyYF5VbjEqvVcnWdRiZ1GphUeJr0UbPdZhDlrGnYURPk6jaCsBp11N4xDN5N3EWDiSiXVcevqcOg1+vFv85kUOckPe4VkVZD4VNE9kmV109plY+SKq81r/TucrmgfFugLChvXKC0GZAUva3uX2KUy2oZ7KxpAey2W62HXY6aULltm9WC2BEMnM0RFEVEpGEUPkUEsPpOLCj3kFtSRV5pFZuLKpi3weC3z/6g3OOnpNIKmCVVPkorrXlJlXefHmUbBiRHu3bbsKS24UlStGu/HLFFRETqUvgUaeUCAZPCitpQWU1eSRW5JdU7vc4vq66nyxs7rF+7x3MYBsS4HcRFOImNcBAX6SQuwklczXJshLUtKdpFSqyb1Bg3KbEukqJcONTARESkTVH4FNkP1XbIXVi+87Sl1AqWuTXBcktpfaGyfoYBKTFu0uPcpMa4qCzK48AeXUiIctcJkbHbh8pIJzEuR6sdiUNERJqWwqdIC1Dp8VNY4aGwzGPNy6spKPMEA2btckFNwCyu9DaqvqT1eNsKlWmxbtLjIkiLi6h5bc3T4yJIjt52JzI4osgJvdTYRUREmozCp0gz8PoDFJV7yC/zUFATJGsb2RSU1bzebnlv+4pMiHKSFOUiKXrblBrrtoJlMGRadSbVf6KIiLQECp8iDWSaVoOcdQUV5JVU1QmPBeVW10AFNQFza4W30e/vtBskRbtIjHKRHOMiKdpNUpTTmse4giEzOcbaJzHKqfqSIiKy31H4FNmO1x9gY1El2YUVrCusILug3FouqGB9YQXlnobfobS6BnKRHO0mOcZFcoy7pmX3tuXaeVKMi1i3Q93/iIhIq6fwKW1OaZWXdQUVZBdWBINldqEVMjdtrcK/m8Y5hgHt4yJonxAZDJEptSEyxgqatevjI53qGkhERGQHCp/SqpRV+6wuhEqqySvdNs8pqbbCZkE5RXt4JB7htNExKYqOSdF0TIqiU3IUHZOj6JgURYfESNwOe4g+jYiISOuj8CktnmmalFT52FITJnNr57UBs6ZLobzSaioa+Fg8JcZVEzCj6Ji8LWR2SooiNdatx98iIiLNROFTWoQqr591BRWs2lLGqrwyVuaW8tsaOw8tn8uWsmqqvA0fRSfW7SA1zk16rNXSu7ZroQ6JVsDMSooixq1/+iJ7yxvw8mfRn2ws20i0M5o4VxyxrlhiXbHEuGJw2tQ1l4jsmv4HlpCpbS2+ekt5MGSu2lLGqi3lbCiqYOeqlgZQGXwVF+Go0zdlWk2XQsF+K2PdpMW5iXLpn7VIUzFNkw2lG/gt/zd+y/+NJflLWFa4jGp/9S6PiXREEuuKrRNKY12xxDpj610f544j1hlLtDOaKGcUEfYIPX1owQJmgK3VW9lSsYUtlVt2ntcsl3nKiHPHkRSRRGJE4ra525pvvy7RnUiUMyrcH61BfAEffxT9wc95P/NL3i8sK1xGpCOS9Kh00qPTaRfdjnbR7UiPSg/OXXZXuIvdouh/aWlyXn+A7MIKVuWVsTq/vE7ILK7cdX3L2AgH3VJj6JYaQ5fkSPLXLueEYw4nIzGGtDg3Ec62VdfSF/BR5auiyl9Fpa+SKt+2+fbral9X+Cqo8lXhtDm33YVyxhDjiiHOFRdcjnXF4ra7w/3x9lqFt4Kc8hy2Vm+l2l+NN+Cl2l+Nx+/B4/dsWw7s8Lp2ebv1Xr83+B4ANsOGgYFhGLtexobNsIFBcHnH/SIdkSRFJJEUmWTNI5JIjkgOrmvJ339hVSFL8pewJH8Jv+b/ypL8JRRXF++0X6wzli7xXaj0V1LqKaWkuoQKXwUAlb5KKn2V5FXk7VUZar/DKGcUUY6o4Lx2XaQjss76+ta5HW4Maq6JYWxbbsB1rN1ee1zAH8BjejAbM7JDEwuYAUo9pRRWFVJYVUhRVRGFVYWUeEqwG3acNqc12Z3blmteO2yOnbbXWbfd+nJv+U5hMr8yn7yKvOC8oLIAn+lrULlLvaVsLNvYoH0jHZEkuhPrBtOa13HOONZ715NbkUtGXIZ17UKkuLqYX7f8aoXNLb/wW/5vVPoqd9pvWeGyXb5HckSyFUyj2gXD6fYBNTUqtUmfGJimid/04zf9GBgtLvwqfMpeq/T4WbWljJV5pazMLePPmpC5rqBil8M5GgZkJkQGQ2a3tGi6pkQTE1PGlqrVrCj6lWWFy/iocDn5jnxmLnBgM6z/GOyGPbhsw4bNZgv+p2G32TEwsBt2DMOos+/26+w2e3Bd7VR7fO324Lrttjtsjm3rbHZM0yRgBoI/3P6Av87rgBnAF/BtW1ez3WfWsy7go8pfFQySlb5KKv2V+AIN++W+N/YYUJ2xRNmjWF29mugN0SRFJRHvjifOFUe8O77ZfpFV+irJLc8lpyKHnPKcOss55TnkVuRS6iltlnOHUrQzOhhKt5+SI5ODy4kRicTZ4/CbezcAQUNU+ipZVrAseEfzt/zf6g0KTpuT3km9OSjlIPqm9KVvSl86xnXcKQD4Aj7KPGVWGPWWUOoprTOVeHa9rsRTEvwP3cSkwlcRDLMtxb1v3lv3ju12d3B3ustbz3a3fVt9cn/AT7GnOBgii6qKgsuFVYUUVRfV2ba1emuz/lvYG0kRSaRGppISlUJaZBopkSmkRqVay1EpxLpiKaku2fb5qovqBOft556AJ/hHy6byTbs850sfvESkI5KOsR3pFNeJTnGd6Bzf2ZrHdSbeHb9Pn8k0TdaVrGPxlsUszrOmVcWrdtov1hnLwWkHMyB1AH1T+uIzfXV+T9Uu55Tn4Al4KKgqoKCqgKUFS+s9r4FBamQq6dHpxLmsn3tfwBf8v8Jn+qzXAX9wm8/c9tob8AaXa/evdXbvs/nH4H/s0/fS1BQ+ZY9Kqrz8mVcWnFbmlrIyr4yNWyt3OcRjpNNO19ToOiGzW2oMWUkuNpavY3nhcpYXLmdO0QqeXL18l4HC62t8Z+2tTe1doAhHhDW3RxDhiAi+3n6d2+7GG/BS6imlzFtmBQFvaTAQlHvLMTHxBrzB/+T25L1v39tpXaQjkjhXHHHuOOJd8XWCae3y9tvi3fHEOGMori7e6Rdz8HVFTr132OoT64wlKTIJl92F2+bGZXcFJ7e95rWt/nXB5e1eO+3WHYfaPypMTGuZQN359tvqWQ6YAQJmgEpfZfD7LagqoLBy27Iv4KPcW065t5z1pesb9HmnvDUleM0j7BH1/nuoXbfT9pp1tfMNZRuCYXNl0cp6A02X+C70TekbDJs9E3s26A8Oh81BQkQCCREJDfpcOwqYAap81l38Cm8Flb7K4PL280pfZb3rKr3b9q/yVwXfs/bamKZZZ7n2utZZrmd77XfkM31WKKwu2qvPV/tHn2maFHuKCZgNr8teK9YZW+fOYJwrLvgz7Qv48Pq9eAPbTTu+rlnnM+vuW8vAIDky2QqVkSmkRdWEyshUUqNSg/PkyOQmu1NnmtYfG7sKpkXVReRX5PNH7h9sNbdS6atkRdEKVhSt2Om9EtwJ20JpXOfgcse4jkQ6Infav8pXxe8Fv1tBc8tifsn7pd7r2ymuE/1S+9E/rT8DUgfQNaFrg+6+mqZJUXWR9Yd0ze+5HQNqbkUuvoCPvMo88ir37mnB7vgDLeuPFlD4lO0UlXtYmbftTuaqLWWszC0jp6Rql8ckRbvonhZDj7QYutdM3VJjaBcXQbmvjBWFK1hR9CMLCpfz6roV/Ln1zzq/6Go5bA66J3SnV2Iveif1pntcd5b9uIxhxw7DZrcF/1Ovb/KbfkzM4J3EAAECgZr5dvsEzEDwL8Od7kw24m5l7WOMOndM67uLWt+dVlv9+0U4IohyRNUJCrXBwWVzNVn9t4AZoNxbTpmnjBJPSb0BtXa5pLqE1RtX4453U+IpCU614arSV0luRW6TlGt7UY6oOo+jglNUu2B9qmhndJOfNxRM06TMWxYMpoWVNeG09vV26wurCtlavRUT07oz7q+CXVez3GupkanW3cxUK2wemHwgsa7Ypj9RA9gMm/XY3BkFO+eEsPF4PHww8wOOOPYIKgOVu76T67WqH9TZ5rW2BcxA8I++7cW54urc7a6t/7j96+SI5OD62j+UmpJpmsEw6rK7cNhCGw0MwyDaGU20M5qs2Kx69/F6vcycOZPjTjiOLdVbWFeyjrXFa1lXss5aLllLbkUuW6u3snXLVn7Z8stO79Euul0wlDptTn7d8itLC5fu9JTJZXNxYMqB9E/rT//U/vRL7UdyZPJef7ba69snuU+9+wTMAIVVhcGAWuYtw2FzYLfZcRrOOk/mHIYjuM1hOILbHDZHndd2m1UVw27YW9wjd1D4bJP8AZM/88pYlF3E75uKg0Ezv8yzy2PS49z0SIsNBszasBkXZaOgsoCCygJyKlbze9FK3vt5BcsLl7OhbEO97xXrjKVXkhUya+fd4rvV+aXq9XrJtefSIaYDTqdazjYVm2ELPgZsT/vd7lv7y370qNHBaxAwA5R5yyiuLqbEU2LNq0vqvC72FO+83VNMtb+aCHuEFSyj0+uGy+2CZawzttU2NjEMI/j9d4rrtMf9K6sreW/mexw17Ch8hm9bvd9d1AOub/v266p8VaREpgQfnR+UchDp0ekh+OT7N8MwcBtu0qPS9+r3Ue2dvVJPKcXVxcFAEu+ObxE9AxiGgdNwtoiy7InT5gzezRzaYWidbRXeCtaXrmdtSd1QurZ4LSWekuCdxgWbF9Q5LjkimQFpA6ywmdafPkl9QhrYbIaNlMgUUiJTODDlwJCdN5wUPtuAonIPi9dvZVF2ET9nb2Xx+q2UVddfnzAzIZLu6dF0SbGTluglPraKiIhyKvyb2VL5KwWVBXxbsoX38vLJr8jf4+On9tHtgwGzd2Jveif3JiM6o9WGi9bOZtisR+quuEYf6/F7cNqcuvaN4LA5iLHFkBGToT/C9mPb39lrF90u3MVptaKcUfRK6kWvpF47bdtatbVOKC33lnNQykH0T+tPh5gO+r0UYgqfrYzPH2BFbik/Z9eGzSLWFBZh2Kow7JUYtipwVBEd7aFDkkFKgp+IiHKwlVJlbqWwKp9lVQUsKqyEPVcHBMBu2EmOTCYlMoVu8d3oldSLPkl96JXUa58rf0vr0RIf/YhI25AQkUD/COvOpoSfwud+pNJXye/5v5NfmR+sl5dXvpU1hQVsKikkv7yYEk8ZAaMSw1aJYa+ClGpiU+tvFbQJ2FQOlNd/vmhnNKmRqXUqn9dOwfVRqSS4E0La7YWIiIjsvxQ+W7CtVVtZlLeIn/N+ZlHuIpYWLN1932p2MCKhvt4wHYYj2MdjjHPbPM4dt1PArH29v3T4KyIiIvsPhc8WZHPZZhbmLWRR7iIW5S6qt2+xgDeOgCcZMxAB/ghMM4KkiHg6xCfRNTmZ3mlpdE9JIc69bUSRGFeMRgwRERGRFkHhM0wCZoDVW1ezKG8RC3MXsihvETnlOTvt569Ow1/R2ZoqOxNtS2VQpyQO6ZjIIZ0S6JeVQFyEGiKIiIjI/kHhM0S8AS9LC5byc+7PLMxbyM95P9fTobYNsyoTb3kn/BVd8Fd2wvTH0DcznqEDUxjaI5VDOiXitKt+pYiIiOyfFD6b0YbSDXyx7gvmbpzLr1t+DY64UcuOG8PTifLirJqwmQWmm5QYF0N7pDK0ZypH9UghJabljgMtIiIi0hgKn01sbfFavsj+gs/Xfs6ywmV1trltMbi83SgszMRT1plAVSZgx2EzOKxzIkN7pjK0RyoHtI/DZlP9TBEREWl9FD6bwKqtq/h83efMXjeblUUrg+ttho3O0QeTvb4rxYWdKPWkAtYj845JURzT37q7OaRbMjFuXQoRERFp/ZR49oJpmvxR9Aefr/ucL9Z9weri1cFtDsPB4PaDGdlpJBnOQ7ng+WV4/AGiXHaO6JMcvLvZOWX/HJtaREREZF8ofDaQaZosLVgaDJzZpdnBbU6bkyMyjmBkp5Ecm3Us8e54yqt9jHn8Ozz+ACP7pPHkuENwO+rrgVNERESk7VD43I2AGWBx3mJmr5vNF+u+YFP5puA2t93NkRlHclzn4zimwzHEumLrHDv5o99ZnV9Ou7gIHjy9n4KniIiICAqfO/EH/CzMW8gnFZ/w2AePkVeZF9wW6Yjk6MyjOa7zcQzNHLrLEYA+XLyRdxZuwGbA1LP6kxitMa1FREREQOFzJ//L/R+XfnFp8HW0M5pjOhzD8Z2O54jMI4h0RO72+HUF5fzz/SUATBjeg8O7JjdreUVERET2JwqfOxiYPpAOMR1IrU5l/BHjOSrrKNz2hvWz6fEFuPqNnymr9jGocxLXDO/ezKUVERER2b9oqJwdOG1OPhzzIadFn8bQzKENDp4AD32+gl83FBMf6WTqWf1xaCQiERERkTqUjuphGI3v4P3rFXk8+63V5dIDpx9MRsLuH8+LiIiItEUKn00gr7SKG9/+BYDzDu/EqAPbhblEIiIiIi2Twuc+CgRMbvjvL+SXeejdLpZ/ntQn3EUSERERabEUPvfRs3NXM3dlPhFOG0+cM4AIp/rzFBEREdkVhc998HN2EQ99tgKAO8YcSPe02D0cISIiItK2KXzupZIqL9e8+TO+gMlJfdtz5mFZ4S6SiIiISIun8LkXTNPkn+8vYX1hJZkJkdx7at+9aiEvIiIi0tYofO6Ft3/awMe/bMJuM3js7AHERzrDXSQRERGR/YLCZyP9mVfK5I9+B2DicT0Z2CkxzCUSERER2X8ofDZCldfPhOk/U+n1c2T3ZK44plu4iyQiIiKyX1H4bIQpM5exPKeU5GgX//5rf2w21fMUERERaQyFzwb6/PccXpm/DoCH/tqPtLiIMJdIREREZP+j8NkAm4sruendXwG45KguHNsrLcwlEhEREdk/KXzugT9gcu2bi9la4aVvZjw3ndA73EUSERER2W8pfO7B41+u5Mc1hUS77Dx+9gBcDn1lIiIiIntLSWo3flxbyGNzVgJw9ykH0TklOswlEhEREdm/KXzuQrkXbnj7NwImnHpIJqcM6BDuIomIiIjs9xQ+62GaJm+sspFTUk2XlGjuGntQuIskIiIi0ioofNZj2o/r+a3IhtNu8PjZA4h2O8JdJBEREZFWQeFzB0s3lTBl1h8A/P34nhyUGR/mEomIiIi0Hrqlt4NIl53uqdFQWcwFQzqGuzgiIiIirYrufO6gS0o0//3bYM7rEcAwNHymiIiISFPaq/D55JNP0rlzZyIiIhg8eDA//vjjbvefOnUqvXr1IjIykqysLK6//nqqqqr2qsCh4HbYiNI9YREREZEm1+jw+dZbbzFx4kQmT57MokWL6NevH6NGjSIvL6/e/adPn84tt9zC5MmTWbZsGS+88AJvvfUW//jHP/a58CIiIiKyf2l0+HzkkUe49NJLufDCCznggAN45plniIqK4sUXX6x3/3nz5nHkkUdyzjnn0LlzZ44//njOPvvsPd4tFREREZHWp1EPlz0eDwsXLmTSpEnBdTabjZEjRzJ//vx6jzniiCN4/fXX+fHHHxk0aBCrV69m5syZnHfeebs8T3V1NdXV1cHXJSUlAHi9Xrxeb2OKvFdqzxGKc0n9dA3CT9cg/HQNWgZdh/DTNQi/hlyDhl4fwzRNs6En3rRpE5mZmcybN48hQ4YE199000188803LFiwoN7jHnvsMW688UZM08Tn83H55Zfz9NNP7/I8d9xxB3feeedO66dPn05UVFRDiysiIiIiIVJRUcE555xDcXExcXFxu9yv2ZvVfP3119x777089dRTDB48mD///JNrr72Wu+66i9tuu63eYyZNmsTEiRODr0tKSsjKyuL444/f7YdpKl6vl9mzZ3PcccfhdDqb/XyyM12D8NM1CD9dg5ZB1yH8dA3CryHXoPZJ9Z40KnympKRgt9vJzc2tsz43N5d27drVe8xtt93GeeedxyWXXAJA3759KS8v529/+xv//Oc/sdl2rnbqdrtxu907rXc6nSH9Rxfq88nOdA3CT9cg/HQNWgZdh/DTNQi/3V2Dhl6bRjU4crlcDBw4kDlz5gTXBQIB5syZU+cx/PYqKip2Cph2ux2wxlAXERERkbaj0Y/dJ06cyPjx4zn00EMZNGgQU6dOpby8nAsvvBCA888/n8zMTKZMmQLAmDFjeOSRRxgwYEDwsfttt93GmDFjgiFURERERNqGRofPM888ky1btnD77beTk5ND//79mTVrFunp6QBkZ2fXudN56623YhgGt956Kxs3biQ1NZUxY8Zwzz33NN2nEBEREZH9wl41OJowYQITJkyod9vXX39d9wQOB5MnT2by5Ml7cyoRERERaUU0truIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIzCp4iIiIiEjMKniIiIiISMwqeIiIiIhIwj3AUQkdbPNE18Ph9+vz/cRdmveL1eHA4HVVVVbeK7s9vtOBwODMMId1FEpBkpfIpIs/J4PGzevJmKiopwF2W/Y5om7dq1Y/369W0mkEVFRdG+fXtcLle4iyIizWSvwueTTz7Jgw8+SE5ODv369ePxxx9n0KBBu9x/69at/POf/+S9996jsLCQTp06MXXqVEaPHr3XBReRli8QCLBmzRrsdjsZGRm4XK42E6KaQiAQoKysjJiYGGy21l1LyjRNPB4PW7ZsYc2aNfTo0aPVf2aRtqrR4fOtt95i4sSJPPPMMwwePJipU6cyatQoVqxYQVpa2k77ezwejjvuONLS0njnnXfIzMxk3bp1JCQkNEX5RaQF83g8BAIBsrKyiIqKCndx9juBQACPx0NERESbCGKRkZE4nU7WrVsX/Nwi0vo0Onw+8sgjXHrppVx44YUAPPPMM8yYMYMXX3yRW265Zaf9X3zxRQoLC5k3bx5OpxOAzp0771upRWS/0haCkzQN/VsRaf0aFT49Hg8LFy5k0qRJwXU2m42RI0cyf/78eo/56KOPGDJkCFdddRUffvghqampnHPOOdx8883Y7fZ6j6murqa6ujr4uqSkBLAq33u93sYUea/UniMU55L66RqEX1NcA6/Xi2maBAIBAoFAUxWtzTBNMzhvK99fIBDANE28Xu8u/48INf0+Cj9dg/BryDVo6PVpVPjMz8/H7/eTnp5eZ316ejrLly+v95jVq1fz5ZdfMm7cOGbOnMmff/7JlVdeidfrZfLkyfUeM2XKFO68886d1n/++echfXQ3e/bskJ1L6qdrEH77cg0cDgft2rWjrKwMj8fThKVqW0pLS8NdhJDxeDxUVlby7bff4vP5wl2cOvT7KPx0DcJvd9egoQ1Lm721eyAQIC0tjWeffRa73c7AgQPZuHEjDz744C7D56RJk5g4cWLwdUlJCVlZWRx//PHExcU1d5Hxer3Mnj2b4447LlhVQEJL1yD8muIaVFVVsX79emJiYlR/by+YpklpaSmxsbFtpqFWVVUVkZGRDB06tMX8m9Hvo/DTNQi/hlyD2ifVe9Ko8JmSkoLdbic3N7fO+tzcXNq1a1fvMe3bt8fpdNZ5fNKnTx9ycnLweDz1dqfhdrtxu907rXc6nSH9Rxfq88nOdA3Cb1+ugd/vxzAMbDab6vLthdpH7bXfodfrbfU/DzabDcMwWuTPfkssU1ujaxB+u7sGDb02jfrfwOVyMXDgQObMmRNcFwgEmDNnDkOGDKn3mCOPPJI///yzTn2lP/74Q/24iUiLN2vWLI466igSEhJITk7mL3/5C6tWrQpu37BhA2effTZJSUlER0dz6KGHsmDBguD2jz/+mMMOO4yIiAhSUlI45ZRTgtsMw+CDDz6oc76EhARefvllANauXYvdbue9997j2GOPJSIigmnTplFQUMDZZ59NZmYmUVFR9O3blzfeeKPO+wQCAR544AG6d++O2+2mY8eO3HPPPQAMHz6cCRMm1Nl/y5YtuFyuOr/bRUSaS6NvRUycOJHnnnuOV155hWXLlnHFFVdQXl4ebP1+/vnn12mQdMUVV1BYWMi1117LH3/8wYwZM7j33nu56qqrmu5TiMh+wzRNKjy+sEy1DXgaqry8nIkTJ/LTTz8xZ84cbDYbp5xySrD/zWOOOYaNGzfy0Ucf8csvv3DTTTcF/9CeMWMGp5xyCqNHj+bnn39mzpw5u+0PeVfuvPNOrr76apYtW8aoUaOoqqpi4MCBzJgxgyVLlvC3v/2N8847jx9//DF4zKRJk7jvvvu47bbbWLp0KdOnTw/W1b/kkkuYPn16nUadr7/+OpmZmQwfPrzR5RMRaaxG1/k888wz2bJlC7fffjs5OTn079+fWbNmBX+xZWdn13m8lpWVxWeffcb111/PwQcfTGZmJtdeey0333xz030KEdlvVHr9HHD7Z2E599J/jSLK1fBfe6eddlqd1y+++CKpqaksXbqUefPmsWXLFv73v/+RlJQEQPfu3YP73nPPPZx11ll1Gk/269ev0WW+4oorOPXUU+v8Xr3xxhuDy1dffTWfffYZ//3vfxk0aBClpaU8+uijPPHEE4wfPx6Abt26cdRRRwFw6qmnMmHCBD788EP++te/AvDyyy9zwQUXtJl6pSISXnvV4GjChAk7Pbap9fXXX++0bsiQIfzwww97cyoRkbBZuXIlt99+OwsWLCA/Pz94VzM7O5vFixczYMCAYPDc0eLFi7n00kv3uQz9+/ev89rv93Pvvffy3//+l40bN+LxeKiurg72BLJs2TKqq6sZMWJEve8XERHBeeedx4svvshf//pXFi1axJIlS/joo4/2uawiIg2hsd1FJKQinXaW/mtU2M7dGGPGjKFTp04899xzZGRkEAgEOOigg/B4PERGRu7+XHvYbhjGTtUA6usjLzo6us7rBx98kEcffZSpU6fSt29foqOjue6664JdWe3pvGA9eu/fvz8bNmzgpZdeYvjw4XTq1GmPx4mINAU1PxWRkDIMgyiXIyxTYx4rFxQUsGLFCm699VZGjBhBnz59KCoqCm4/+OCDWbx4MYWFhfUef/DBB++2AU9qaiqbN28Ovl65cmWD+sj7/vvvGTt2LOeeey79+vWja9eu/PHHH8HtPXr0IDIycrfn7tu3L4ceeijPPfcc06dP56KLLtrjeUVEmorCp4hIPRITE0lOTubZZ5/lzz//5Msvv6zT//DZZ59Nu3btOPnkk/n+++9ZvXo17777bnC0t8mTJ/PGG28wefJkli1bxm+//cb9998fPH748OE88cQT/Pzzz/z0009cfvnlDeqmpEePHsyePZt58+axbNkyLrvssjrd30VERHDzzTdz00038eqrr7Jq1Sp++OEHXnjhhTrvc8kll3DfffdhmmadVvgiIs1N4VNEpB42m40333yThQsXctBBB3H99dfz4IMPBre7XC4+//xz0tLSGD16NH379uW+++4L9mk8bNgw3n77bT766CP69+/P8OHD67RIf/jhh8nKyuLoo4/mnHPO4cYbb2zQCG633norhxxyCKNGjWLYsGHBALy92267jRtuuIHbb7+dPn36cOaZZ5KXl1dnn7PPPhuHw8HZZ5/dYjpzF5G2QXU+RUR2YeTIkSxdurTOuu3raXbq1Il33nlnl8efeuqpnHrqqfVuy8jI4LPP6rb637p1a3C5c+fO+P3+nUYMSUpK2ql/0B3ZbDb++c9/8s9//nOX++Tn51NVVcXFF1+82/cSEWlqCp8iIm2I1+uloKCAW2+9lcMPP5xDDjkk3EUSkTZGj91FRNqQ77//nvbt2/O///2PZ555JtzFEZE2SHc+RUTakGHDhjV6pCcRkaakO58iIiIiEjIKnyIiIiISMgqfIiIiIhIyCp8iIiIiEjIKnyIiIiISMgqfIiIiIhIyCp8iIvUYNmwY1113XbiLISLS6ih8ioiIiEjIKHyKiIiISMgofIqI7EFRURHnn38+iYmJREVFceKJJ7Jy5crg9nXr1jFmzBgSExOJjo7mwAMPZObMmcFjx40bR2pqKpGRkfTo0YOXXnopXB9FRCTsNLymiISWaYK3IjzndkaBYTT6sAsuuICVK1fy0UcfERcXx80338zo0aNZunQpTqeTq666Co/Hw7fffkt0dDRLly4lJiYGgNtuu42lS5fy6aefkpKSwp9//kllZWVTfzIRkf2GwqeIhJa3Au7NCM+5/7EJXNGNOqQ2dH7//fccccQRAEybNo2srCw++OADzjjjDLKzsznttNPo27cvAF27dg0en52dzYABAzj00EMB6Ny5c9N8FhGR/ZQeu4uI7MayZctwOBwMHjw4uC45OZlevXqxbNkyAK655hruvvtujjzySCZPnsyvv/4a3PeKK67gzTffpH///tx0003Mmzcv5J9BRKQl0Z1PEQktZ5R1BzJc524Gl1xyCaNGjWLGjBl8/vnnTJkyhYcffpirr76aE088kXXr1jFz5kxmz57NiBEjuOqqq3jooYeapSwiIi2d7nyKSGgZhvXoOxzTXtT37NOnDz6fjwULFgTXFRQUsGLFCg444IDguqysLC6//HLee+89brjhBp577rngttTUVMaPH8/rr7/O1KlTefbZZ/ftOxQR2Y/pzqeIyG706NGDsWPHcumll/Kf//yH2NhYbrnlFjIzMxk7diwA1113HSeeeCI9e/akqKiIr776ij59+gBw++23M3DgQA488ECqq6v55JNPgttERNoi3fkUEdmDl156iYEDB/KXv/yFIUOGYJomM2fOxOl0AuD3+7nqqqvo06cPJ5xwAj179uSpp54CwOVyMWnSJA4++GCGDh2K3W7nzTffDOfHEREJK935FBGpx9dffx1cTkxM5NVXX93lvo8//vgut916663ceuutTVk0EZH9mu58ioiIiEjIKHyKiIiISMgofIqIiIhIyCh8ioiISJtRMns2uffdT6C6OtxFabPU4GhHnnKMP7+iY8G3wOhwl0ZERESaSNGbb5Jzx50AONu3I2n8+DCXqG3Snc8d5SzB8d9xHLjxDQj4w10aERERaQKFr7wSDJ4ABS+/gun1hrFEbZfC544yB2K643D5yzFyfgl3aURERGQf5T/zH3Kn3AdA0oUXYk9Jwbd5MyWffhrmkrVNCp87sjswOw8FwFj9VZgLIyIiInvLNE22PPYYW6ZOBSDl6gmk3fR3ks49F4CC51/ANM0wlrBtUvish9l1GKDwKSIisr8yTZO8Bx8i/6mnAUi78QZSr7oKwzBIPPssjKgoqv/4g/LvvgtzSdsehc96BLoeC4Cx8SeoKglzaURkf9S5c2em1txtEZHQMgMBcu+6m8IXXwQg/Z//JPmSS4Lb7fHxJJ5xBmDd/ZTQUvisT0InytzpGAEfrNVfRCIiIvsL0+9n8+23UzR9OhgG7f51J0nnnbvTfknjzweHg4oFC6j87bcwlLTtUvjchbzYvtbCqi/DWxARkRDz+/0EAoFwF0Ok0Uyfj023TKL4nXfBZiPjvikk/vWv9e7rzMgg/iSrS8WCF14MZTHbPIXPXdgSd5C1sGpOeAsiIiH37LPPkpGRsVMAGzt2LBdddBGrVq1i7NixpKenExMTw2GHHcYXX3yx1+d75JFH6Nu3L9HR0WRlZXHllVdSVlZWZ5/vv/+eYcOGERUVRWJiIqNGjaKoqAiAQCDAAw88QPfu3XG73XTs2JF77rkHgK+//hrDMNi6dWvwvRYvXoxhGKxduxaAl19+mYSEBD766CMOOOAA3G432dnZ/O9//+O4444jJSWF+Ph4jjnmGBYtWlSnXFu3buWyyy4jPT2diIgIDjroID755BPKy8uJi4vjnXfeqbP/Bx98QHR0NKWlpXv9fYnUx/R42DjxBko+/hgcDjIfeZj4sWN3e0zSRRcDUPr553jWrQtFMQWFz13Kj+mDaXNA4WooXBPu4oi0GqZpUuGtCMvU0FatZ5xxBgUFBXz11bZGh4WFhcyaNYtx48ZRVlbG6NGjmTNnDj///DMnnHACY8aMITs7e6++E5vNxmOPPcbvv//OK6+8wpdffslNN90U3L548WJGjBjBAQccwPz58/nuu+8YM2YMfr/VF/GkSZO47777uO2221i6dCnTp08nPT29UWWoqKjg/vvv5/nnn+f3338nLS2N0tJSxo8fz3fffccPP/xAjx49GD16dDA4BgIBTjzxRL7//ntef/11li5dyn333Yfdbic6OpqzzjqLl156qc55XnrpJU4//XRiY2P36rsSqU+gupoN11xL6eefYziddHjsUeJOOGGPx0X06kn00KMhEKDg5Zebv6ACaISjXfLZIzEzD8VY/wOs/gqSuoS7SCKtQqWvksHTB4fl3AvOWUCUM2qP+yUmJnLiiScyffp0RowYAcA777xDSkoKxx57LDabjX79+gX3v+uuu3j//ff56KOPmDBhQqPLdd111wWXO3fuzN13383ll1/OE088AcCDDz7IoYceylNPPRXc78ADDwSgtLSURx99lCeeeILxNaO1dOvWjaOOOqpRZfB6vTz11FN1Ptfw4cPr7PPss8+SkJDAN998w1/+8he++OILfvzxR5YtW0bPnj0B6Nq1a3D/Sy65hCOOOILNmzfTvn178vLymDlz5j7dJRbZUaCykg1XTaB83jwMt5sOTzxBzNEN//effPEllH87l+L33id1wgQcycnNWFoB3fncLbOm1bvqfYq0PePGjePdd9+lumb852nTpnHWWWdhs9koKyvjxhtvpE+fPiQkJBATE8OyZcv2+s7nF198wYgRI8jMzCQ2NpbzzjuPgoICKioqAPjll1+CIXhHy5Yto7q6epfbG8rlcnHwwQfXWZebm8ull15Kjx49iI+PJy4ujrKysuDnXLx4MR06dAgGzx0NGjSIAw88kFdeeQWA119/nU6dOjF06NB9KqtILX9ZOesv/ZsVPKOiyHr22UYFT4CoQYcRcfDBmNXVFE2b1kwlle3pzudumF2OhW+mwOpvwe8Du74ukX0V6YhkwTkLwnbuhhozZgymaTJjxgwOO+ww5s6dy7///W8AbrzxRmbPns1DDz1E9+7diYyM5PTTT8fj8TS6TGvXruUvf/kLV1xxBffccw9JSUl89913XHzxxXg8Hmw2G5GRuy737raB9UgfqFPlwFvPkIKRkZEYhlFn3fjx4ykoKODRRx+lU6dOuN1uhgwZEvycezo3WHc/n3zySW655RZeeuklLrzwwp3OI7I3/CUlrL/0b1T+8gu2mBiynn2WqEMGNPp9DMMg+eKL2XjttRROm07yxRdji45uhhJLLd353A2zfT+ITITqYti0aM8HiMgeGYZBlDMqLFNjQk9ERASnnnoq06ZN44033qBXr14ccsghgNX454ILLuCUU06hb9++tGvXLth4p7EWLlxIIBDg4Ycf5vDDD6dnz55s2rSpzj59+/Zlzpz6Gz/26NGDyMjIXW5PTU0FYPPmzcF1ixcvblDZvv/+e6655hpGjx7NgQceiNvtJj8/P7j94IMPZsOGDfzxxx+7fI9zzz2XdevW8dhjj7F06dJg1QCRfeErKiL7ggut4BkfT8eXXtqr4FkrduQInJ06EiguZuu77zVhSaU+Cp+7Y7NDzWhH/KlW7yJtzbhx45gxYwYvvvgi48aNC67v0aMH7733HosXL+aXX37hnHPO2euuibp3747X6+Xxxx9n9erVvPbaazzzzDN19rnlllv43//+x5VXXsmvv/7K8uXLefrpp8nPzyciIoKbb76Zm266iVdffZVVq1bxww8/8MILLwTfPysrizvuuIOVK1cyY8YMHn744QaVrUePHrz22mssW7aMBQsWMG7cuDp3O4855hiGDh3KaaedxuzZs1mzZg2ffvops2bNCu6TmJjIqaeeyt///neOP/54OnTosFffk0gtX34+2eePp2rpUuxJSXR69RUi+x60T+9p2O0kX3gRAAUvv4RZz9MBaToKn3vSrabCvep9irQ5w4cPJykpiRUrVnDOOecE1z/yyCMkJiZyxBFHMGbMGEaNGhW8K9pY/fr145FHHuH+++/noIMOYtq0aUyZMqXOPj179uTzzz/nl19+YdCgQQwZMoQPP/wQh8OqCnTbbbdxww03cPvtt9OnTx/OPPNM8vLyAHA6nbzxxhssX76cgw8+mPvvv5+77767QWV74YUXKCoq4pBDDuG8887jmmuuIS0trc4+7777Locddhhnn302BxxwADfddFOwFX6t2ioEF1100V59RyK1fLm5rDvvfKpXrsSRmkqn114lolevJnnv+JPHYk9OxrdpMyWzPmuS95T6GWZD+x4Jo5KSEuLj4ykuLiYuLq7Zz+f1epk5cyajR4/GWZ4DUw8CwwY3rYHIhGY/v+xwDZzOcBenTWqKa1BVVcWaNWvo0qULERERTVzC1i8QCFBSUkJcXFyw7ub+6LXXXuP6669n06ZNuFyu3e7bEv/N6PdR+Hm9Xj6fNo2er0/Dt2EDjoz2dHrpJVydOjXpefKfeYYtUx/F3bs3Xd5/T/WTt9OQn4OG5rX997dZqCRkQUpPMAOw5ttwl0ZEZL9RUVHBqlWruO+++7jsssv2GDxFdsWTnU3W08/g27ABZ1YWnV97rcmDJ0DiWWdhREVRvXw55d/Pa/L3F4vCZ0N0q+nCRI/eRaSRpk2bRkxMTL1TbV+drdUDDzxA7969adeuHZMmTQp3cWQ/5SssZOOFF+EsLsbZuTOdXn8NZ2Zms5zLnpBA4hmnA1DwwvPNcg5RV0sN0204LHjaGmrTNEG34UWkgf7v//6PwYPr71S/tT/CveOOO7jjjjvCXQzZzxVNfwN/Xh6e1FQ6v/QSzkaO3tVYSePHU/j6NCrm/0Dlkt+JPKh1/5EYDgqfDdH5SLA5YWu2Ndxmcrdwl0hE9hOxsbEaSlJkL5keD0VvvQlAwciROFKaf/QhZ0YGcSeNpuSjjyl88QUyH3mk2c/Z1uixe0O4oqHj4dayHr2LiIiERMnns/FvyceemkrpPnan1BjJF19snX/WZ3jWrw/ZedsKhc+GUpdLIiIiIVX0+usAxJ9xBtjtITtvRK9eRB99NAQCFL70csjO21YofDZU95pGR2u+Bb86nxUREWlOlUt+p3LxYnA6iatpBBRKtXc/t773Hr7CwpCfvzVT+Gyo9L4QlQKeMlj/Y7hLIyIi0qrV3vWMO+EEHCkpIT9/1OBBRBx0EGZVFUXTpof8/K2ZwmdD2WzQ7VhrWY/eRUREmo2voICSGTMASDp33B72bh6GYZB8iXX3s2jaNAIVFWEpR2uk8NkYqvcpIiLS7La+/Tam10tE375E9usXtnLEHncczqws/Fu3svW998NWjtZG4bMxutbc+dz0M1So/oeI7Frnzp2ZOnVqg/Y1DIMPPvigWcsjsr8wvV6K3rC6VwrXXc9aht1O8kUXAlD40kuYPl9Yy9NaKHw2Rlx7SDsQMGH11+EujYiISKtTOmcOvtxc7MnJxJ54YriLQ/wpp2BPSsK7cSMln33WbOcxTZOSzz+n4PnnqVq6FNM0m+1c4abw2Viq9ykiItJsCmsaGiX89QxsLleYSwO2iAgSa+7AFrzwQrOEQm9uHusvv5yN11xL3kMPs+bU0/hzxAhy7r6H8vnzMb2tq5cdhc/G2r7eZyv+q0SkuZimSaCiIixTQ//TePbZZ8nIyCAQCNRZP3bsWC666CJWrVrF2LFjSU9PJyYmhsMOO4wvvviiyb6j3377jeHDhxMdHU3Xrl257LLLKCsrC27/+uuvGTRoENHR0SQkJHDkkUeybt06AH755ReOPfZYYmNjiYuLY+DAgfz0009NVjaR5lS1fDmVPy0Eh4PEs84Kd3GCEs8+GyMykuqly6iYP7/J3tc0TYo//JDVY8ZQ/s23GC4X0UcdhRERgW/TZopef53sCy/ijyOPYuPfb6Jk1iz8ZeVNdv5w0fCajdXpCLC7oWQj5P8Bqb3CXSKR/YpZWcmKQwaG5dy9Fi3EiIra435nnHEGV199NV999RUjRlh9/BYWFjJr1ixmzpxJWVkZo0eP5p577sHtdvPqq68yZswYVqxYQceOHfepjOXl5YwaNYohQ4awYMEC1q5dy3XXXceECRN4+eWX8fl8nHzyyVx66aW88cYbeDwefvzxRwzDAGDcuHEMGDCAp59+GrvdzuLFi1v9GPLSetTe9Yw9bmSzj+HeGI7ERBJOP52i116j4PkXiD7iiH1+T9+WLWyefAdlX1pPUiP69iVjyr24u3cnUFlJ+fz5lM6ZQ9lXX+MvLKTk448p+fhjDKeTqMMPJ3bECGKGH4szLW2fyxJqCp+N5Yy0Aujqr6y7nwqfIq1OYmIiJ554ItOnTw+Gz3feeYeUlBSOPfZYbDYb/bZrgXvXXXfx/vvv89FHHzFhwoR9Ovf06dOpqqri1VdfJTIyko4dO/LYY48xduxY7r//fpxOJ8XFxfzlL3+hW7duAPTp0yd4fHZ2Nn//+9/p3bs3AD169Nin8oiEiq+oiJKPPwEg6bzzwlyanSWNH0/R9OmUz5tH1dKlRBxwwF69j2malMyYSe5dd+EvLgank9SrriL5kosxHFYss0VGEjt8OLHDh2P6/VT+8osVRL+Yg2fdOsrnzqV87ly44w4iDj7Y2nfkCFzdugX/EG3JFD73RvcR28Ln4VeEuzQi+xUjMpJeixaG7dwNNW7cOC699FKeeuop3G4306ZN46yzzsJms1FWVsYdd9zBjBkz2Lx5Mz6fj8rKSrKzs/e5jMuWLaNfv35ER0cHH/sfeeSRBAIBVqxYwdChQ7ngggsYNWoUxx13HCNHjuSvf/0r7du3B2DixIlccsklvPbaa4wcOZIzzjgjGFJFWrLid9/FrK7GfUAfIgcMCHdxduLqkEnciSdS8sknFLzwIpkPP9To9/AVFJBzx52Uzp4NgPuAPmRMuY+IXj13eYxhtxN1yCFEHXIIaTfeiGf1akrnfEnZnDlU/vILVb/+StWvv7Jl6lScnToSO3wEsSOGEzlgAEYIhyRtDNX53Bu19T7Xfge+6vCWRWQ/YxgGtqiosEyNuSMwZswYTNNkxowZrF+/nrlz5zJunNXo4MYbb+T999/n3nvvZe7cuSxevJi+ffvi8Xia62ur46WXXmL+/PkcccQRvPXWW/Ts2ZMffvgBgDvuuIPff/+dk046iS+//JIDDjiA999X/4TSspl+P0XT3wAgady5LfbuXfLFFwFQMmsWng0bG3VsyazPWP2XMVbwdDhIuXoCXd56a7fBc0eGYeDu1o2Uv11K57fepPu339DuzjuJPmYohsuFd102hS+9xLpzz2Pl0UPZ9I9/Uvbd940qZygofO6NtAMgJh28FZD9Q7hLIyLNICIiglNPPZVp06bxxhtv0KtXLw455BAAvv/+ey644AJOOeUU+vbtS7t27Vi7dm2TnLdPnz788ssvlJdva1Tw/fffY7PZ6NVrWzWfAQMGMGnSJObNm8dBBx3E9Onbhv/r2bMn119/PZ9//jmnnnoqL730UpOUTaS5lH31Fd5Nm7AnJBB30uhwF2eXIvr0IfrII8Hvp/CVVxp0jK+oiI0TJ7LxuuvwFxXh7tWLLm//l9SrrsLYx/rYzrQ0Es/8Kx3/8x96zp9H5qOPEvd/Y7DFx+MvLKT4vfco+eSTfTpHc9ir8Pnkk0/SuXNnIiIiGDx4MD/+2LCxzt98800Mw+Dkk0/em9O2HIah0Y5E2oBx48YxY8YMXnzxxeBdT7DqUb733nssXryYX375hXPOOWenlvH7cs6IiAjGjx/PkiVLmDt3Ltdeey3nnXce6enprFmzhkmTJjF//nzWrVvH559/zsqVK+nTpw+VlZVMmDCBr7/+mnXr1vH999/zv//9r06dUJGWqPD1aQAknHEGtoiIMJdm92qH3Nz6zjv4iop2u2/pF1+w+i9jKJn5KdjtpFx5BV3e/i8RzfAzaYuOJm7U8WQ+8AA9v5tLx5dfJvH884gb85cmP9e+anT4fOutt5g4cSKTJ09m0aJF9OvXj1GjRpGXl7fb49auXcuNN97I0UcfvdeFbVEUPkVaveHDh5OUlMSKFSs455xzgusfeeQREhMTOeKIIxgzZgyjRo0K3hXdV1FRUXz22WcUFhYyePBgxo8fz/Dhw3niiSeC25cvX85pp51Gz549+dvf/sZVV13FZZddht1up6CggPPPP5+ePXvy17/+lRNPPJE777yzScom0hyqV66k4ocfwGYj8eyW073SrkQdfjgRBxyAWVlJ0Rtv1LuPf+tWNv79JjZMuBp/QQHuHt3p/NZbpF5zDUYI+i41nE6iDx9Mu3/8g5gjj2z28zVWoxscPfLII1x66aVceKE13NQzzzwTvDNwyy231HuM3+9n3Lhx3HnnncydO5etW7fuU6FbhNqhNnN+hbItEJMa3vKISJOz2Wxs2rRpp/WdO3fmyy/r/uF51VVX1XndmMfwO/Y/2rdvX7788ksCgQAlJSXExcVhs1n3CtLT03dZh9PlcvHGLv4zFGmpCqdZdz1jR4zAmZER5tLsmWEYJF9yMRsn3kDRa6+TfNFFde7Wln71FZtvvx3/lnyw2Ui++GJSrp7QIjrMbykaFT49Hg8LFy5k0qRJwXU2m42RI0cyfzedrv7rX/8iLS2Niy++mLlz5+7xPNXV1VRXb2vIU1JSAoDX68Ubgl7+a8+x23O5E3Ck98XI/Q3fyi8wDzq92cvVljToGkizaopr4PV6rU7lA4EmeyzdltSG0trvsC0IBAKYponX68XeQlrq6vdR8/GXlFD84YcAxJ591i6/45Z2DSKOPRZHZia+jRspfOcd4s88E39JCfn3P0DpRx8B4OzShfS77yLi4IPxA/4WUva91ZBr0NDr06jwmZ+fj9/vJ32Hjl/T09NZvnx5vcd89913vPDCCyxevLjB55kyZUq9j4k+//xzohrQQXRTmV3TFcKuHGB2pAe/sWnua/ycHbpytSV7ugbS/PblGjgcDtq1a0dZWVnIWoK3NP/973+ZOHFivduysrJ2+4d7rdLS0qYuVovl8XiorKzk22+/xefzhbs4dbSo30d+P87CQhwlJVR27gwtJKg3VsLcuaRVVlHdLp2v8/Jg5szd7t+SrkH8oYeSvnEjG596msUbNpL23ns4S0owDYOio4+m4Pjj+H3DBtiwIdxFbVK7uwYVFRUNeo9m7eeztLSU8847j+eee46UlJQGHzdp0qQ6v6xLSkrIysri+OOPJy4urjmKWofX62X27Nkcd9xxux0ZxFgbA9NmkFW9kvYnnmg1RJIm0dBrIM2nKa5BVVUV69evJyYmhogW3oiguZx55pkMGzas3m1Op3O3v9NM06S0tJTY2NgW2/VMU6uqqiIyMpKhQ4e2mH8z4fx9ZPr9eDdswPPnKjyr/qyZr8K7di1mzR90MSecQPr992HY9q8ObEy/n3WPP4EP6PC3yzjwpJN2uW9L/D8hcOyxrP3mG1yFhWS+/DIAzk6dSLv7Lnr07x/WsjWHhlyD2ifVe9Ko8JmSkoLdbic3N7fO+tzcXNq1a7fT/qtWrWLt2rWMGTMmuK720ZHD4WDFihX1dn7sdrtxu907rXc6nSH9R7fH83U5ChyRGOV5OItWQvqBIStbWxHqay4725dr4Pf7rX49bbZgncW2Jj4+nvj4+L06tvb3Ze132BbYbDYMw2iRP/vNWSYzEMC7YQPVf/5J9co/rfmff+JZvRqzuv7+pI2ICEyvl7JZs3BndSDthhuapWzNpfS77/Bt2IAtLo6kk8dia8B326L+XTidJJ17LvlPPAGGQdL555F63XXYGjGYxf5od9egodemUeHT5XIxcOBA5syZE+wuKRAIMGfOnHqHlOvduze//fZbnXW33norpaWlPProo2RlZTXm9C2Pww2dj4I/Z1ut3hU+Req1Y4MakV1p7f9WzEAA76ZNVK9caYXL2rC5ejVmVVW9xxhuN65uXXF37467ew9r3qM7zsxMSj7+mE0330LBc8/jzMwk8ayW31q8VlFt90qnnYYthFXqmlLy3y7FFh1NZL9+RB3S8kZlaqka/dh94sSJjB8/nkMPPZRBgwYxdepUysvLg63fzz//fDIzM5kyZQoREREcdNBBdY5PSEgA2Gn9fqv7iG3h84irw10akRal9q/giooKIlv53QBpGrV1xlrM3a0m5M3NZcMVV1K1dGm92w2XC1fX2pBpBUx39+44O3TY5TCJ8WPH4tm4kfzHHifnX3fhaNeO2F1U82hJqlevofz778EwSDzn7HAXZ6/ZXC6SL7wg3MXY7zQ6fJ555pls2bKF22+/nZycHPr378+sWbOCjZCys7PbzOMhYFt/n+vmgbcSnPoPVqSW3W4nISEh2A9wVCOHuGzrAoEAHo+HqqqqVv971TRNKioqyMvLIyEhocW0dG8qnrVryb7oYrybNoHTibtzZ9w9euDu0R1XTdh0ZWVhOBrfFCPliivwbtxI8bvvsXHiDXR69VUiD2rZT+KKarpXihk2DNf+/hRUGm2vGhxNmDCh3sfsAF9//fVuj325plJuq5HSE+IyoWSjFUC7jwh3iURalNr64HsaiEJ2ZpomlZWVREZGtpnQnpCQUG8bgv1Z1dKlZF/6N/wFBbg6dSLrhRdwdchssvc3DIP2d9yBb3MO5fPmsf6Ky+ny5ps4M5vuHE3JX1ZGcU1ftYnnjtvD3tIaNWtr9zbBMKDbsfDz69ajd4VPkToMw6B9+/akpaW1mD769hder5dvv/2WoUOHtsrH0DtyOp2t7o5nxU8/sf7yKwiUleHu04eOzz2LoxG9vzSU4XSS+dijrBt3LtUrVpB92WV0nj4dewh6iGms4vc/IFBRgatrV6KPOCLcxZEwUPhsCt2G14TPr8JdEpEWy263t7pg0dzsdjs+n4+IiIg2ET5bm9KvvmLjdddjVlcTeehAsp5+GntsbLOdzx4TQ9Z/nmHtmWfh+XMVGyZcTdbzz7WokXXMQCD4yD1x3Dlt5o6+1NW6KxGFStdjAQPyfofSnHCXRkREwqz4o4/YMOFqzOpqYoYNo+Pzzzdr8KzlbNeOrP88gy06mooff2Tzrbe2qB4Eyr+fh2ftWmzR0cSPPTncxZEwUfhsClFJkFHTxYLufoqItGmFr77GpptuBr+fuP8bQ4fHH6sz9ndzi+jdm8xHHwW7nZKPPmbLY4+F7Nx7UvT66wDEn3oq9pjoMJdGwkXhs6nUtnpfNSe85RARkbAwTZMtjz9B7r33ApB43nlk3HcfRhiqTMQcdSTt/2UNU13w9DMUvf12yMuwI092NmXffguwX3evJPtO4bOpBMPnV1AzKomIiLQNZiBA7t33kP/kkwCkXHM16f+YFNYhLxNOO42UK68AIOeOOymb+13YygJQNG06mCbRRx+Nu0uXsJZFwksNjppKh8PAFQMV+ZD7G7TvF+4Sich+zPT7KXjqKbq98iqr7rjTCjGGYU02GwYEl4PrDTCM7fczMDDqHGePi8PZvj3OjAycmRk4apczMrAnJKgByF4wvV423TKJkhkzwDBIv/WfJI1rGV0IpVx9tdUH6IcfsfHaa+k07XUi+vQJeTkC5eVsfe89AJLOOzfk55eWReGzqThc0GUorJhpdbmk8Ckie8m3ZQsbb/w7FQsWYAfMmqkpeIGqJUvq3WZERW0LptsF1NplR1raXnWC3poFKivZcN11lH/zLTgcZNx3H/F/OSncxQoyDIP2d92FNyeXigULWH/Z5XR+602c7duHtBzFH39MoLQUV6dORB91VEjPLS2Pfos0pW7Dt4XPo64Pd2lEZD9UvuBHNt54A/4t+RiRkWz6vzEcPv4CHA47mCZmIFCTRE2rio9pWq2Zt5uC+5hm3f0CJv6iQrwbN+HdvBnvpk3WtHkz/vx8zIoKPKtW4Vm1qv7C2e040tOCd0ojevYkcdw4bG106FR/SQnrr7iSyoULMSIi6PDoVGKOOSbcxdqJ4XLR4fHHWDduHNUr/2T93y6j0/RpIWl9D1Zd2MKahkaJ484Ja1UEaRkUPptSbb3P7B/AUw4uteQTkYYxAwEKnn3OapkcCODu0YP0hx5ixfJlOLM6NHs/n4GqKrybN+PbPpRu2m45Jwd8PnybNuPbtJlKFlIClM+bR4enngppa+6WwLdlC9mXXEr1ihXYYmPJeuZpogYODHexdskeF0fWf/7D2jPPonrlSjZeey1Z//lPSBpDVfzwA54/V2FERRF/yinNfj5p+RQ+m1JSV0joCFuzYe330PP4cJdIRPYDvqIiNt10M+Vz5wIQf8optLv9NvwOByxfFpIy2CIicHfpssuGIKbfjy8/f1sYXb+egmefo3zefDZceRUdnnqyzQRQz4YN1jjt2dnYU1Lo+PxzRPTuHe5i7ZEzI8PqhP7c8yifN5/Nt0+m/b33NHs938LXrU7lE04eG7K7rdKy6d53UzKM7Vq9fxnesojIfqFi0c+sOeVUyufOxYiIoP0995Ax5d4W9yjbsNtxpqcTNWAA8SedRMrll5P17H8woqIonzePDROuJlBdHe5iNruqP/5g3dnn4M3OxpmZSedpr+8XwbNWxAEH0OHfj4DdTvH775P/5FPNej7Pho2UfWX1f53YQhphSfgpfDa1bjVjuyt8ishumKZJwUsvs+788/Hl5ODq0oXOb71FwmmnhrtoDRZ16KF0/M8zGJGRlH/3HRuuvpqAxxPuYjWbysW/sO688/Ft2YK7Rw86TZ+Oq1OncBer0WKOOYZ2t98OQP4TT7D1/Q+a7VxFb0yHQIDoI4bg7tat2c4j+xc9dm9qXYaCYYP8FVC8AeI7hLtEItLC+IuL2fSPf1I2xxqUIm70aNr961/75YgvUYcdRtYzz7D+ssso/3YuG6++hszHHwvbeOKmx8OWJ56k/IcfsCcm4EhOwZGcjCM1BXtysvU6JRlHSgq2+PgGP3KOWvEHm+64A7Oyish+/cj6zzPYExKa98M0o8Qz/4p3wwYKnnuOzbfdhjM9jegjjmjScwQqK9n6zrvW+c5V90qyjcJnU4tMgMxDYcOP1t3PQ84Pd4lEpAWp/G0JG6+/Hu+GDRhOJ+n/mETCWWft1/1rRg8eRNYzT7P+sssp++YbNl57HR0enYoR4gDq3bSJDddfT9UvvzbsAKcTR1ISjpQU7Cm1wdQKp8GgmppC+a+/kfnKK5h+P9FHHmkNlxkV1bwfJgRSr78O76ZNlMyYwYZrrqXTtGlE9OrZZO9f/MknBIqLcWZmtsheACR8FD6bQ7fhCp8iUodpmhS98QZ5U+7D9HpxduhA5tSpRB50YLiL1iSiDz+crKefYv0VV1L21VdsuH4iHf79SMgCaNnc79j097/j37oVW1wcaRMnYjjs+PLz8eUX4CvIx59fYL0uKCBQUgJeL77cXHy5uXt8fwOIGTWKDg8+EPJQ3VwMm432U+7Fl5tLxU8/sf6yy6w+QNPT9/m9TdOkqKahUeI552DY7fv8ntJ6KHw2h27D4Zv7YPXXEPCDTT90Im2Zv6ycnNtvo2TmpwDEjBxBxr33Yo+LC3PJmlb0EUfQ4ckn2XDllZTNmcPGG24g85FHmrU7H9PvJ//Jp8h/+mkwTSIOPJDMR6fi6rD7Kk8Bjwd/TRD15efjr5kHg+qWmm0FBQQqKykaNIhu99/XaoJnLZvLRYcnHmft2efgWbOGP48ZhuF0YkRGYnO7rXlEBEZkBDZ3zTyiZl1ExLZtEZEYEW5rW2QEvsIiqleswIiI2K/qMUtoKHw2h8yB4I6HyiLYvNh6LSJtUtWKFWy89jo8a9eCw0HajTeQNH78fv2YfXdijjrSCqBXXUXp7C/YeMONZD78ULMEUF9hIZtu/Dvl8+YBkHDWmaRPmoTN7d7jsTaXC1tNZ/l74qmu5tNZs1rt3Tt7QgJZzz1L9sUX412Xjen1Ynq9BJrgvePHjNmv68ZK81D4bA52B3QdCss+th69K3yKtDmmaVL83nvk/OsuzOpqHO3akfnvR4gaMCDcRWt2MUcfRYcnHmfDVRMo/fxzNv79JjIferBJh+as+PlnNl4/EV9ODkZkJO3vmEz82LFN9v7bawsj8rg6dKDbzJn4S0owq6oIVFZhVlUSqKqyXldVEaisxKyqJlBVuW2famturavZVmntb3O7Sbni8nB/NGmBFD6bS7fhNeHzKxj693CXRkRCKFBRQc6/7qL4gw8AiB56NBn3348jMTG8BQuhmKFDyXz8MTZcfQ2ls2axyWaQ8cAD+xxATdOk6LXXyH3gQfD5cHXpQuajU4no2XQNZdoqw25vU/9GJXxa/59z4VLb2fz6BVBVEt6yiEhI+MvKqFj0M2vPPNMKnjYbqddfT9Yzz7TJ/9Rjhw2jw6OPgtNJycxP2XTzLZg+316/n7+sjI3XTyT33ing8xF74gl0fvttBU+R/YzufO7ANE1ybrqJ+IhIzBNP3Ps3SuxsDbdZuBrWfge9RzdZGUUk9EyfD9+WLXg3b9425vnmTfg2bbbWbd5MoLQ0uL89NYXMhx8metCgMJY6/GKHH0uHqf9mw7XXUTJjBthsZNw3pdH1J6tW/MHGa6+16s46naTfdBOJ545rtXVnRVozhc8dlH7xBWWfziIdyCkrJfPuu7HHx+/dm3UbYYXPVV8qfIq0cP6ycrybNuKrCZLeYKi0xjL35eaB37/H97EnJBB12GG0m3w7jpSUEJS85YsdMYLMfz/CxusnUvLxxxg2g/b33tvgAFr84YdsnnwHZlUVjnbt6DD130T279+8hRaRZqPwuYPYESNIvvEG8v89lfLZX7B6ye9kPvQgUQP3otFQt+Hwv+c01KZIC2WaJkXTp5P/1NP4Cwr2fIDDgbNdO5zt2+PMaI8jI8Nabp+BM6M9znbtsEXvf6MUhULcccfBww+zceJEij/8CAwb7e+5e7cBNFBdTe4997L1v/8FIPrII8l46ME2WYVBpDVR+NyBYbOROH48i6qr6f7hR3izs1l33vmkTLiKlMsua9yjos5Hgc0BhaugaK31KF5EWgRvbh6b//lPyr/7LrjOFh+PMxgorYDpbN8eR/v2ODMycaQkt9rudkIhbtTx8PBDbLzhxmCd2PZ331Vva3LPhg1svOZaqpYuBcMg5aqrSLnicn3/Iq2AwucuVHfoQNZ/3yL/3nsp+ehj8h97nIr5P5Dx4AM427Vr2JtExEHWYFj3PbxzEZz1BsTu+8gRIrJvSmZ9Rs7kyfiLizHcbtJuuIH4U0/dL8dW39/EnXACBAJsvPHvFL/3HobdRrs776wTQEu//IpNt9xCoKQEe0ICGQ8+SMzRR4Wx1CLSlNTafTds0dFkPvAAGfffhy0qior//Y81Y0+mdM6chr/J8XdBZCJsXAjPj4Dcpc1XYBHZLX9pKZtuvoWN112Hv7iYiAMOoMt775J0/nkKniEUN3o0GQ88ADYbW99+h5w77sQMBDB9PvIefoQNV15JoKSEyH796PL+ewqeIq2MwmcDxI8dS5f33iXiwAPxFxez4aoJ5Nx1N4Hq6j0fnDkQLpkDSd2geD28cDz8+UXzF1pE6qj947H4ww/BZiP5ssvo/OYbuLt1C3fR2qT4v5xExv33WQH0v/9l8223kX3RxRQ89xwAieedR6fXXsXZvn2YSyoiTU3hs4FcnTvT+Y3pJF14IQBF06ax9q9nUr1q1Z4PTu4Gl3wBnY4CTylM+yv87/lmLrGIgDWGd95DD7Hu/PF4N23C2aEDnV5/jbTrr2t143Tvb+LHjCFjyr1gGBS/+x4VP/6ILSqKzH8/Qrt//kPXR6SVUvhsBMPlIv3mm8h67lnsyclUr1jBmtNOp+jttzFNc/cHRyXBee9Dv3PA9MOMG2DWPyCw565bRPZ33k2b2PLkk2x95x38ZWUhO2/VH3+w9q9nUvD8C2CaxJ92Kl0++ICoQw4JWRlk9+LHjqX9vfeC3Y67R3c6v/M2cfvSx7KItHhqcLQXYo4+mq4fvM+mm2+hfN48cm67nfLv59H+X3dij4vb9YEOF5z8FCR3hS/vhh+ehKI1cOpz4I4J3QcQCZGAx0Phiy+R/5//YFZWApBz9z3EHncc8SePJfrww5ul9bIZCFD46qtseeTfmB4P9sRE2t/1L2JHjmzyc8m+SzjlZGKGHYM9Lk6t2UXaAN353EuO1FSynn+OtL/fCA4HpbNmsebkU6hY9PPuDzQMa6z3018EuxtWzISXToSSTaEpuEiIlM2dy5ox/8eWqVMxKyuJ7NcPV5cumFVVlHz8MesvvoQ/R4wk75F/U716TZOd15uTQ/bFF5N33/2YHg/Rxwyl60cfKni2cI7ERAVPkTZC4XMfGDYbyRdfTOfp03BmZeHdtIl1551H/jPPYO5pJJSDToMLPoGoFMj5FZ4bAZt/DU3BRZqRZ8NG1l81gfWX/g3PunXYU1PIePABOr35Bl1nzqDzW2+ScPZZ2OLi8OXkUPDss6wePZq1Z55F0Ztv4S8p2etzF38yg9X/N5aK+T9gREbS7o7J1rjqqalN+AlFRGRfKHw2gciDD6bL++8Rd9JJ4PezZeqjZF90Md7c3N0fmDUILp0DKb2gdBO8eAKsmBWaQos0sUB1NVuefJLVJ51E2Zw5YLeTdMEFdPv0U+LHjMEwDAzDILJfP9pPnkyPud+S+e9HiD5mKNhsVP7yCzl33MHKo45m48SJlH37LabP16Bz+4uL2XjDjWy68UYCJSVEHHwwXd57l8SzztLY3yIiLYzCZxOxx8SQ8dCDtJ8yBSMqiooFC6w+Qb/6avcHJnaGiz+HrsPAWw5vng0/PA17asAk0oKUfvkVq/8yhvzHn8CsriZq0CC6fvA+6bfcjD2m/vrMNrebuBNPpON//kP3r78i7e9/x92jB6bHQ8nMT1n/t8tYeeyx5D74INUrV+7y3OU//MDqsSdTMmMG2O2kTJhA52mv4+7Spbk+roiI7AOFzyZkGAYJp5xMl3fewX1AH/xbt7LhiivJuetuvDk5uz4wMgHGvQOHjAczALNugZl/B3/D7vqIhIsnO5v1l13OhiuvxLt+PY60NDIefoiOr7yMu0ePBr+PMy2N5IsvostHH9L53XdIPPdc7AkJ+LfkU/jCi6we83+sOf0MCl+fhq+oCKgZ93vKfWRfcCG+nBxcnTrRefo0UidcheF0NtdHFhGRfaTW7s3A3bULnd98ky0PP0LhK69QNG0aRdOmETlwIHEnnEDsqONxpqXVPcjuhDGPQnJ3mH07/O85azz401+0hukUaUEClZUUPPccBc+/gOnxgMNB8gXjSbniCmzRez9SkGEYRB54IJEHHkj6TX+n9JtvKP7gQ8q++YaqJUuoWrKE3PvvJ3bYMDxr1wbviCaceSbpN9+ELSqqqT6iiIg0E4XPZmJzuUifdAvRRx1J/jP/oXLhwuCUe++9RB12GHGjTyT2uONwJCdbBxkGHHkNJHWBdy+FP2db9UDPeQsSssL7gUQA0zQpmzOH3Hun4N1k9dAQfcQQ0m+9FXfXrk16LsPlIu6444g77jh8hYWUfPIJWz/4gOqlyyidPRsAe3Iy7e+5m9hhw5r03CIi0nwUPptZzNFHE3P00XhzciiZNYvST2dR+csvVPz4IxU//kjOXXcTPXgQsSeeSOzIkTgSE6HPGLhwBrxxNuT9Ds8Nh3PetIbqFAmT6jVryL13CuVz5wLgaN+e9FtuIfb445q9UY8jKYmk888n6fzzqVqxguIPPiRQWUHqNdfgSEpq1nOLiEjTUvgMEWe7diRfcAHJF1yAd+NGSmbNouTTWVQtWUL5vPmUz5tPzp3/InrIEOJOPJHYkSOwXzIHpp9pBdCXToJTn4UD/i/cH0XamEBFBXkvvEjhSy9her0YTidJF11EymV/C8tj7ohevYi4+aaQn1dERJqGwmcYODMzSb74YpIvvhhPdjYln86i5NNPqV6+nPK5cymfO5ecyU6ijzqKuJHXEuN+E/v6L+G/58HIO+HIa61H9CLNyDRNYn79jex/T8VX02Au+uijSf/HJLUkFxGRvabwGWaujh1JuexvpFz2N6pXr6Fk1qeUfvop1Sv/pOyrryj76isMl4uYXgOIi11GzKw7sOX8CifcBzFpez6ByF7wZGez6Y47yZg3Dx/gzMgg/R+TiBkxQv1miojIPlH4bEHcXbuQeuWVpF55JdUrVwbviHrWrKH0t1xKScKwm8Qt+oKUnw/DdfJtMPBCsGlIOmkapsdDwYsvkv/0M5jV1QQcDpIvvpi0yy/DFhkZ7uKJiEgroPDZQrl79CC1Rw9Srp5A9YoVwSDqzc6meHU0xWtM4hfdRcqwV3Cd9zhk9A93kWU/V/HTT2yefAeeVasAiBw8mKVDj6bn+edjU7+ZIiLSRNTJfAtnGAYRvXuTdv11dPtsFp2mTyP6qKPANCheE8Wql7ew6aKT8bxyJVQVh7u4sh/yb93KpltvZd255+FZtQp7UhIZDz5AxnPP4k1JCXfxRESklVH43I8YhkHUIYfQ8fnn6PzWm0QfOdgKoWsjWXXfl2w8fRDVszQ0pzSMaZoUf/QRq0afRPE77wKQcMYZdJs5IzgWu4iISFNT+NxPRfbrR8cXXqbz2/8lZtBBYBqU/Glj9fWPsvH0wVQv3MOY8tKmVa9ZQ/ZFF7HpppvxFxbi6t6NTtNep/1d/8KekBDu4omISCum8Lmfi+zbl6xX36bzW9OJ6dfRCqG/l7J63BVsPPcEqpf/Hu4iSgsS8HjY8tRTrBl7MhXzf8Bwu0m9/nq6vvceUQM1iIGIiDQ/hc9WIrLfALLe+owuLz1OTK9YwKDkp3WsPvl0NlxyDlV//BHuIkqYlf/4I2vGnkz+Y49jejxEH3kkXT/+iJTL/obhcoW7eCIi0kYofLYyEUNGkvXBAro8fA2xna11pd/9zJr/G8uGKy+jaoVCaFvjKypi06R/kH3+eDxr1mBPSSHj4YfIev45XB07hrt4IiLSxih8tkaGQcRJV9Dhwx/pctMIYrOqACj98lvWjB3LhglXU7V8eZOe0vR6CZSXN+l7yr4xTZOt73/A6hNHU/z++wAknHWm1aDopJPUoEhERMJC/Xy2Zu5YIi56gg6jL6fq5Qnkz1lL6foISr/4gtIvviBm5AgSzzobDDArKwlUVBCoqCRQWUmgorxmXe3rCgKVFZgVtcvbr68Er9c6Ze/eJJx+OvFj/oI9Pj7MX0DbVb16DTl33EHFjz8C4O7Zk3Z33kHUgAFhLpmIiLR1Cp9tQbuDiLjpSzocP43qtyeTv9BPSXYkZV/MoeyLOU16qurly8m9+27yHnyQ2OOPJ+GM04k67DDdZduB6fNR8dNCSr+cQ9WS3zFcLmxRUdgiI615VBS2aGtuREZii4retj5qu32iojAirXWGzUagupqCZ5+j4NlnMb1ejIgIUidcRdL48RjqKF5ERFoAhc+2wmaDQ87D3Ws0mV9MJuWbNyhYGkNFQQS2uGRsyZnYYmJrgkxNCIqMxBZdE34itws+NQGpNvTU7m/6/ZTM/JStb79tjcr08ceUfPwxzk4dSTj9dBJOPhlHamq4v4mw8ZeVU/7dXErnfEnZt98SKG7aQQGMyEgMwyBQUQFA9NCjaXf77bg6dGjS84iIiOwLhc+2JjoZxj6Be8C5ZHwyEfJ+B3LAlQ2DLoHDr4KYvQ+ISeeOI3HcOVQtWcLWt9+h5JNP8K7LZsvDj7Bl6qPEHDuMhNNPJ+boozHsrX9Mem9uLmVffknpl19R8cMPmDXVEwDsiYnEHHss0UOGgGEQqCgnUFGxrQpEeUVNVYiKbVUcKiq27VdTBaJ2UAGzshITsKem0O6f/yR21CjdcRYRkRZH4bOt6ng4XD4Xfn8f5j5ihdDv/g0/PAMDx8MR10B85l69tWEYRPbtS2TfvqTffBMlsz5j69tvU7l4cfBRvyM9nYTTTiX+1NNwddi787REpmlS/cdKyr6cQ+mcL6lasqTOdlenTsSMGEHsiOFE9u+/zwHcNE3Mqqo6IdXVsSO2iIh9el8REZHmovDZltns0Pd0OPBU+GMWzH0INi6EBc/A/16A/mfDUddDUte9P0V0NAmnnUrCaadSvXIlW995l+IPP8SXm0v+U0+T//QzRA8ZQsIZpxMzYgS2ZuhvMuDxECgpCdafbOq7gdvX3yz78iu8GzZs22gYRPbrR8yI4cQOH46ra9cmPb9hGDXVIiIhObnJ3ldERKS5KHyKVR+092jodSKs/hrmPgxr58KiV+Hn1+Gg0+DoGyCtzz6dxt2jB+mTbiH1homUzZnD1rffoXzevOBkT0ggfuxYEs44HVunTrt8H9PrxVdUhL+oCH9BAb6CQvxFhda8sBBfYaG1vqgQf0EhgbKybQc7ndjj4oKTLT4Oe3yC9To+DltcHPa4eOzxcdjj463X8fHY4+IwIiKCwTFYf/PLLyn7pm79TcPtJnrIECtwHnssjpSUffreREREWhOFT9nGMKDbsdaUvcC6E7ryc/jtbWvq/RcYeiNk7Ft3PTaXi7gTTyTuxBPxbNjA1nffpfjd9/Dl5VH4yisUvvIKEf36kZyURN7ChZhFW4NB0ldYuG8Ndbxe/AUF+AsKGn2o4XRii4/HHhuLd8OGnetvDhtG7IjhRB9xBLaoqL0vo4iISCum8Cn16zgYxr0Nm3+x7oQu/QiWf2JN3UZYIbTTEft8GleHDqRdey2pV11F2XffsfWddyj76muqfvmFZKBkVwfabNgTE3EkJWJPSsaRnIQ9MQl7chKOpGTsSYk4kpOxJyXhSErCFhuLWVmJv6TEmoqLCZSU4C+2lv0l270uKbFeB7eVgN+P6fXiz8/Hn59vlb22/ubwY4kcMKBNNKASERHZVwqfsnvt+8FfX4UtK6yGSb+9DavmWFPHIXD0jdB9hHXXdB8YDgexw4YRO2wYvi1bKHz/fVbNm0+X/v1wpabiSEraFjKTkrDHxzc67BnR0diio3G2b9+o40zTtFqelxTXBNcSHKkpuLp0UWtyERGRRlL4lIZJ7QWn/geG3QLfPwqLp0H2fJh2GrTvb90J7XWSVX90HzlSU0m88ELy0tM5dPRonGHuHN0wDOwx0dhjonFmZIS1LCIiIvs7je0ujZPUBcZMhWt/sfoEdUbB5sXw1rnw9BD46SWoatrO00VERKT1UPiUvROXASfcC9ctsR69u+Ngy3L45Dp4qBe8dxmsmQuBQLhLKiIiIi2Iwqfsm+hkGHEbXL8Ejr8bUnuDrxJ+fRNe+Qs8PgC+eQC2rg93SUVERKQFUPiUphERD0dcDVf+AJfMgYEXgCsWitbCV/fA1L7w2imw5F3wVoW7tCIiIhImexU+n3zySTp37kxERASDBw/mxx9/3OW+zz33HEcffTSJiYkkJiYycuTI3e4v+znDgA6HwphH4cY/4JRnofPRgAmrvoR3LoKHe8HMv1vdOImIiEib0ujw+dZbbzFx4kQmT57MokWL6NevH6NGjSIvL6/e/b/++mvOPvtsvvrqK+bPn09WVhbHH388Gzdu3OfCSwvnioJ+Z8IFn8A1i2HoTRDXAaq2wo/Pwn+GwjNHwYL/QEVhuEsrIiIiIdDo8PnII49w6aWXcuGFF3LAAQfwzDPPEBUVxYsvvljv/tOmTePKK6+kf//+9O7dm+eff55AIMCcOXP2ufCyH0nqAsP/Cdf9Cue+Z40nb3dBzm/w6U3W3dC3L4CVX0DAH+7SioiISDNpVD+fHo+HhQsXMmnSpOA6m83GyJEjmT9/foPeo6KiAq/XS1JS0i73qa6uprq6Ovi6pMQa58br9eLdbkjD5lJ7jlCcq03qNNSaKouwLXkX2y/TMHJ/g9/fh9/fx4zNgIPOIKq6g65BGOnnIPx0DVoGXYfw0zUIv4Zcg4ZeH8M0TbOhJ960aROZmZnMmzePIUOGBNffdNNNfPPNNyxYsGCP73HllVfy2Wef8fvvvxMREVHvPnfccQd33nnnTuunT59OlMbMbpXiKtbRsXAuWYXf4/KXB9fnx/QmO2komxIOw293h7GEIiIisjsVFRWcc845FBcXExcXt8v9QjrC0X333cebb77J119/vcvgCTBp0iQmTpwYfF1SUhKsK7q7D9NUvF4vs2fP5rjjjgv76DptyxXgq8a3chbGz69hW/MNKWXLSSlbzoCc6ZgHnEyg3zjMzEP3eThP2TP9HISfrkHLoOsQfroG4deQa1D7pHpPGhU+U1JSsNvt5Obm1lmfm5tLu3btdnvsQw89xH333ccXX3zBwQcfvNt93W43bvfOd7mcTmdI/9GF+nwCOJ1w8Ol4+4zliw9eY2TKFuy/voFRtAZj8evYFr8OKT2h/zjodzbEpoe7xK2efg7CT9egZdB1CD9dg/Db3TVo6LVpVIMjl8vFwIED6zQWqm08tP1j+B098MAD3HXXXcyaNYtDDz20MaeUNqzKlUzgqIlwzc9wwUzod441nGf+H/DFZHikD0w/C5Z9An7VAxIREdkfNPqx+8SJExk/fjyHHnoogwYNYurUqZSXl3PhhRcCcP7555OZmcmUKVMAuP/++7n99tuZPn06nTt3JicnB4CYmBhiYmKa8KNIq2UY0PlIazrxfqth0uJpsH4B/PGpNUWlQL+zYMC5kNYn3CUWERGRXWh0+DzzzDPZsmULt99+Ozk5OfTv359Zs2aRnm49/szOzsZm23ZD9emnn8bj8XD66afXeZ/Jkydzxx137Fvppe2JiIOB461pyx+w+HX45U0oy4X5T1hT5kArhB50mjXykoiIiLQYe9XgaMKECUyYMKHebV9//XWd12vXrt2bU4jsWWpPOO5fMPx2+PML+Pk1+GMWbFxoTbMmQZ//s4Jo56PBptFkRUREwi2krd1FmoXdAb1OsKayLfDbf2HRa7BlmbX8238hoSP0HgM9joNOR4BD3TaJiIiEg8KntC4xqTDkKjj8Sti0CH5+HX57B7Zmww9PWpMzGroOs4Joj+MgvkO4Sy0iItJmKHxK62QYVt3PzIEw6l5Y+XnNNNuqH7pihjUBpB1YE0SPh6xBYFc3HiIiIs1F4VNaP2ckHDDWmgIByP1tWxDd8D/I+92avp8K7njodqwVRLuPVD+iIiIiTUzhU9oWmw3a97OmoX+HikJY9aUVRv/8AioKYOkH1gTQvr8VRHscD5mHgM0exsKLiIjs/xQ+pW2LSoK+p1tTwA+bft72iH7Tz7B5sTV9+wBEJll3Q3scb9UZjUkNc+FFRET2PwqfIrVsduhwqDUd+w8oy7Puhq78HP78EioLt7WeB0jqBlmDoeNgyDrcGvZT3TmJiIjslsKnyK7EpEH/c6zJ77Xqh9bWFc1dAoWrrOmX6db+EQlWg6WswdDxcMg4BFxRYf0IIiIiLY3Cp0hD2J1W/6CdjoCRd0BlEaz/H6z/Adb/CBt+gqqt2x7ZA9gc0O5gK4hmDbLujsa1D+enEBERCTuFT5G9EZkIPY+3JrDujOb8agXR7B+scedLN1t9jW5aBD88Ze2X0NEKoVmDrFCadoAaMYmISJui8CnSFOzObf2KHn4FmCYUr4fsBTV3RxdA7u9WZ/dbs7fVG3XFQtZh1vCfXY6BjP4KoyIi0qopfIo0B8Ow7nImdISDz7DWVZXAxp+23R3d8BN4Sq2unlZ9ae3jjofOR0GXodD1GEjtbb2XiIhIK6HwKRIqEXHQbbg1gdW1U95SWDcP1nwLa+ZCdXHd0Zei07YF0S5DIbFz2IovIiLSFBQ+RcLFZod2fa1p8GVWGN282Aqiq7+x7o6W58GSd6wJIKFTTRgdZj2q1whMIiKyn1H4FGkpbPZt9UaPuh581Vb3TrVhdONPsHUd/PyaNQGk9tl2V7TTkRCZENaPICIisicKnyItlcNt1f/sfJTV6X11qXU3dPXXViDN+Q22LLOmBc+AYbOGA+10hNWavsNhEJcR7k8hIiJSh8KnyP7CHQs9jrMmsMalXzvXuiu65lsoWLmta6f5NcfEZdaM2lQTRtv3A2dE2D6CiIiIwqfI/ioqCQ4Ya00AxRutMLr+R9jwo9W1U8lGWLoRln5o7WNzWnVMOxxWc3f0UKseqVrUi4hIiCh8irQW8ZnQ7yxrAqgusxow1Y7AtOFHKN+y7e7oj/+x9otOtcJo7ZQxANwxYfsYIiLSuil8irRW7phtdUbB6vh+a7bViKl22vyrFUhXzLQmsOqOph2ILeMQOhY4YXMGtDtQ49SLiEiTUPgUaSsMAxI7WVPf06113irY/Mt2gfQnKNkAub9hz/2NAQAvPg8YkNQV0vpA+oHWPO1Aa51dv0ZERKTh9L+GSFvmjICOg62pVskm2PA//NkLKFzyJSmBPIyKAihcZU3LP9m2r90NqT2tILp9MI3LVD1SERGpl8KniNQVlwEHjCXQYzTzqgcz+sQTcVYXWaMx1U65S2HLcvBWWF0+5fxW9z3c8TVh9ABIq536WI2kRESkTVP4FJHdMwxrJKXYdOh27Lb1gQBsXQt5y6wwWhtM81daw4Su/8GatpfYGbIGWy3tswZbodRmD+WnERGRMFP4FJG9Y7NZdT6TukLvk7at91VbATRvGeT9vi2cFmdD0Vpr+vUta19XjDWiU9Zga+pwqEZpEhFp5RQ+RaRpOdzQ7iBr4oxt6yu3wsaFVtdP6xdYjZs8pbDmG2uqldp7253RrMGQ3F31R0VEWhGFTxEJjcgE6D7CmgACfqve6PoF2wJp4Wpr3ZblsOjVmuOStg0XmjUYMg8BV3TYPoaIiOwbhU8RCQ+b3Wodn34gHHqRta5si9XlU20g3bQIKgvhj1nWBGDYrVGaMg+BpG6Q3M2aJ3YGhytsH0dERBpG4VNEWo6YVOg92poAfB6rJf36BdYITdkLoHSTNXLT5sV1jzVsEJ+1LYxuP0/oCHZnqD+NiIjUQ+FTRFouhws6DLQmrrTWFW+A7B+slvUFNX2PFqwGbzlsXWdNq76s+z42hxVA64TSrtY8oaNa3IuIhJDCp4jsX+I7bBuhqZZpQlnudmF0u1BauBp8lda8cDX8ObvusTanNepTRDw4IsFZO0VZnfA7o7atc2y/bcd9I8FRs39UktXwSkREdqLwKSL7P8OA2HbW1PnIutsCASjdXDNC0+qaYLrd3F8NBX82cXls1h3V5B6Q0hNSulvz5B4Qk6bW+yLSpil8ikjrZrNBfKY1dRlad1sgACUbrb5HPeXWiE3eSutOqbd2qgBv1bZt3grwVW23rXKHqQJM/7Y+TXe80+qO3y6M1sxTelj9pepuqYi0AQqfItJ22WyQkGVNTcU0oXwL5P9hdbafvxIKVlqvt2Zboz9tXGhN2zNskNBpWxhN6YGR0AW3dyuYgaYrn4hImCl8iog0JcOwHq3HpEHno+pu81bVPPKvCaP5f24LqZ5SKFpjTSs/A6xf0CcA5u/Xb3vPmPTt5ukQnbptOSYN3LF6rC8iLZrCp4hIqDgjIP0Aa9pebYOpHe6Wmlv+gOL1GKbfqrdaunnP53BE1h9Sa8Orzbld1YHtqhQEqxJU1lP1YPvX2+3v91jvZ3dZXVnVmbvA7thu2Vl3ub7j0vpAl2Oa9k60iLQ4Cp8iIuG2fYOp7eql+rxePp3xEScefRjO6gIoy7NCalnudsu18y3W3VNf5bYup0Ih4LPO2ZSSulohtOsx0HkoRCc37fuLSFgpfIqItGCm4YC49uDsuOedPeU1YXQ3IdUM1NN1VOSu1wVf13QjVdudlDPCumPp91oB1O+pmbx7WPbWv95bYY1utXHRtm6xFr4EGNaIVl2PgS7DoNMQDa8qsp9T+BQRaS1c0ZDUxZr2V1XFsPZ7WPMNrP4GtiyDnF+tad7j1uP6DofVhNFjoMOhGr1KZD+j8CkiIi1HRHzdIVZLc2HNt7Dma1j9LRRnQ/Y8a/p6Cjijrb5dax/Tpx1o9WLQWAG/dee4tsstT1nN6wqMqlJSS5ZCThbEpUNUsnXnV0T2isKniIi0XLHpcPAZ1mSaVm8Aq7/Zdme0shBWfm5NYAXDLkMh/SDwVe8UJIPLO67fTb1VB3AEwKoHtq10RlvnikqqmW8/7bguCSKTrOFiRUThU0RE9hOGYTVGSuoKh15oDRKQu2RbEF03DyoK4Pf3rWmvzmEDV4xVr9UVDa5oAnY3ZQWbibV7MCoLrTqu3nIoLrfuxDaUO84KojHtrKoRiTVVJJK6WstRSeHrJisQsL47XxXYHDWTvWZybJsMm7rykn2m8CkiIvsnmw3aH2xNR1wNPo/Vef+ab6wO/bcLkMHJWbscZYVMV3TNfjHWOkfETuHK7/Xy1cyZjB49GqfDAdUlVlCrKKyZ7zjtuL4QMK3jqkuska/W/7Dz53HHQWLn7YJp123LcZl7V53ANKGyCEpzarrrqm+eA2U5VqhuCGOHQGqz7RxQbQ7ru4xtZ40uFtehZp4J8R2suSuq8Z9HWgWFTxERaR0cLqs1fKchzXcOw7DqpUbEW+GwIQJ+qyFVRQGU59cM6boGCtfWzNdA6SYrmNY2rtqR3WWNgLVjMI3LqBsuSzbvHC791Q39cNZ5TP/ug6jpB7+/Ye+b9/uut0Um7hBKdwipcZmqqtBKKXyKiIg0J5u9ph5okjV0an28lVC0blsYLVpT0+XUGusurt9jjYxVsHLvyhCZBLHta/qTbb+tX9nY9ttex6Rt6znANK1uuQK+mslfd27661+//bK3Ako2WWG7eEPNfKM195RZobmyCHJ/23W5o9MgPhN7bAYDtpRgmzmnppuvmsEJHO6dBzCwu+sOYOBw7bzdEWGNBuaOtboSC0dVgkDA+oOjqhiqtlrzyq3Wd+uOte6Eu+Mgombuim41VR4UPkVERMLNGQlpva1pRwG/Fd7qBNOaeclmq1FTnVC5wzwmvfGt8w2j5vG6HXA3yUcMMk0raAXD6IZtoXT7kOqvhvI8KM/Dxs90BCic27RlAetzBsNe7C6m+rbVrHNFW3881AmRRdteV27dOWBWFVvB0ww0opy2mvPGW/PaULr9cnBd7XIsxGdBYqem/972gcKniIhIS2azW+EhsRN0HRbu0uw7w4DIBGtKP7D+fUzTqqZQE0b9RetZ/tsievfoit30bRugwFddz4AGNZPPs8NgBtXbDWpQCdWlgGndaazaak3hYHdb30VEglWdw+awylZdDFUl1rLpt4JqVbE1NcYh4+H/HmuOku81hU8RERFpWQwDolOsKaM/Aa+XP/Pa0/Oo0didTTSogGlaXW1Vl243lezwunRbQ7F615dCdZnVaC0iHiLjt4XIYKDc8XXtcs2+e7orbZpWFYbaIFr7qL72/FUlOywX110f36Fpvq8mpPApIiIibY9hgDvGmmgf7tLsmmFs662hJZezEfai3wYRERERkb2j8CkiIiIiIaPwKSIiIiIho/ApIiIiIiGj8CkiIiIiIaPwKSIiIiIho/ApIiIiIiGj8CkiIiIiIaPwKSIiIiIho/ApIiIiIiGj8CkiIiIiIaPwKSIiIiIho/ApIiIiIiGj8CkiIiIiIaPwKSIiIiIho/ApIiIiIiGj8CkiIiIiIbNX4fPJJ5+kc+fOREREMHjwYH788cfd7v/222/Tu3dvIiIi6Nu3LzNnztyrwoqIiIjI/q3R4fOtt95i4sSJTJ48mUWLFtGvXz9GjRpFXl5evfvPmzePs88+m4svvpiff/6Zk08+mZNPPpklS5bsc+FFREREZP/S6PD5yCOPcOmll3LhhRdywAEH8MwzzxAVFcWLL75Y7/6PPvooJ5xwAn//+9/p06cPd911F4cccghPPPHEPhdeRERERPYvjsbs7PF4WLhwIZMmTQqus9lsjBw5kvnz59d7zPz585k4cWKddaNGjeKDDz7Y5Xmqq6uprq4Ovi4uLgagsLAQr9fbmCLvFa/XS0VFBQUFBTidzmY/n+xM1yD8dA3CT9egZdB1CD9dg/BryDUoLS0FwDTN3b5Xo8Jnfn4+fr+f9PT0OuvT09NZvnx5vcfk5OTUu39OTs4uzzNlyhTuvPPOndZ36dKlMcUVERERkRArLS0lPj5+l9sbFT5DZdKkSXXulgYCAQoLC0lOTsYwjGY/f0lJCVlZWaxfv564uLhmP5/sTNcg/HQNwk/XoGXQdQg/XYPwa8g1ME2T0tJSMjIydvtejQqfKSkp2O12cnNz66zPzc2lXbt29R7Trl27Ru0P4Ha7cbvdddYlJCQ0pqhNIi4uTv/Iw0zXIPx0DcJP16Bl0HUIP12D8NvTNdjdHc9ajWpw5HK5GDhwIHPmzAmuCwQCzJkzhyFDhtR7zJAhQ+rsDzB79uxd7i8iIiIirVejH7tPnDiR8ePHc+ihhzJo0CCmTp1KeXk5F154IQDnn38+mZmZTJkyBYBrr72WY445hocffpiTTjqJN998k59++olnn322aT+JiIiIiLR4jQ6fZ555Jlu2bOH2228nJyeH/v37M2vWrGCjouzsbGy2bTdUjzjiCKZPn86tt97KP/7xD3r06MEHH3zAQQcd1HSfoom53W4mT56806N/CR1dg/DTNQg/XYOWQdch/HQNwq8pr4Fh7qk9vIiIiIhIE9HY7iIiIiISMgqfIiIiIhIyCp8iIiIiEjIKnyIiIiISMgqfO3jyySfp3LkzERERDB48mB9//DHcRWpT7rjjDgzDqDP17t073MVq1b799lvGjBlDRkYGhmHwwQcf1Nlumia333477du3JzIykpEjR7Jy5crwFLaV2tM1uOCCC3b6uTjhhBPCU9hWasqUKRx22GHExsaSlpbGySefzIoVK+rsU1VVxVVXXUVycjIxMTGcdtppOw2iInuvIddg2LBhO/0sXH755WEqcevz9NNPc/DBBwc7kh8yZAiffvppcHtT/QwofG7nrbfeYuLEiUyePJlFixbRr18/Ro0aRV5eXriL1qYceOCBbN68OTh999134S5Sq1ZeXk6/fv148skn693+wAMP8Nhjj/HMM8+wYMECoqOjGTVqFFVVVSEuaeu1p2sAcMIJJ9T5uXjjjTdCWMLW75tvvuGqq67ihx9+YPbs2Xi9Xo4//njKy8uD+1x//fV8/PHHvP3223zzzTds2rSJU089NYylbl0acg0ALr300jo/Cw888ECYStz6dOjQgfvuu4+FCxfy008/MXz4cMaOHcvvv/8ONOHPgClBgwYNMq+66qrga7/fb2ZkZJhTpkwJY6nalsmTJ5v9+vULdzHaLMB8//33g68DgYDZrl0788EHHwyu27p1q+l2u8033ngjDCVs/Xa8BqZpmuPHjzfHjh0blvK0VXl5eSZgfvPNN6ZpWv/unU6n+fbbbwf3WbZsmQmY8+fPD1cxW7Udr4FpmuYxxxxjXnvtteErVBuUmJhoPv/88036M6A7nzU8Hg8LFy5k5MiRwXU2m42RI0cyf/78MJas7Vm5ciUZGRl07dqVcePGkZ2dHe4itVlr1qwhJyenzs9FfHw8gwcP1s9FiH399dekpaXRq1cvrrjiCgoKCsJdpFatuLgYgKSkJAAWLlyI1+ut87PQu3dvOnbsqJ+FZrLjNag1bdo0UlJSOOigg5g0aRIVFRXhKF6r5/f7efPNNykvL2fIkCFN+jPQ6BGOWqv8/Hz8fn9wpKZa6enpLF++PEylansGDx7Myy+/TK9evdi8eTN33nknRx99NEuWLCE2NjbcxWtzcnJyAOr9uajdJs3vhBNO4NRTT6VLly6sWrWKf/zjH5x44onMnz8fu90e7uK1OoFAgOuuu44jjzwyOBpfTk4OLpeLhISEOvvqZ6F51HcNAM455xw6depERkYGv/76KzfffDMrVqzgvffeC2NpW5fffvuNIUOGUFVVRUxMDO+//z4HHHAAixcvbrKfAYVPaVH+v537eUlljcMA/lzEkYKyTEkrFNNqlS0Mq40boTAIihbWSkpqkQYVFbSQCIK2RX9AOxcRRNCmwLRd7STaCEog0Y9FkGVFLXzv4p4rN07ndOB6Zu6dng8Iw8yAD74+8MWZ0e/3l7ZdLhe6urpgs9mwvb2NUCikYDIi5YyMjJS229vb4XK54HA4kEwm4fP5FEymTuFwGOfn57zfXEE/WoPJycnSdnt7OywWC3w+H7LZLBwOh9wxVamtrQ2pVAr5fB47OzsIBoM4Pj4u63vwsvs3RqMRGo3mu6e2bm9vYTabFUpFNTU1aG1tRSaTUTrKl/T3d5+9+G9pbm6G0WhkL36DSCSC/f19JBIJNDU1lfabzWa8vb3h/v7+3fnsQvn9aA0+0tXVBQDsQhlJkgSn0wm32421tTV0dHRgY2OjrB3g8PmNJElwu92Ix+OlfcViEfF4HD09PQom+9oKhQKy2SwsFovSUb4ku90Os9n8rhcPDw84PT1lLxR0eXmJu7s79qKMhBCIRCLY3d3F0dER7Hb7u+NutxtarfZdF9LpNHK5HLtQJp+twUdSqRQAsAu/UbFYxOvra1k7wMvu/zA3N4dgMIjOzk54PB6sr6/j6ekJY2NjSkf7Mubn5zEwMACbzYarqyssLy9Do9FgdHRU6WiqVSgU3v1qcHFxgVQqBYPBAKvVipmZGayurqKlpQV2ux3RaBQNDQ0YHBxULrTK/GwNDAYDVlZWMDw8DLPZjGw2i8XFRTidTvT19SmYWl3C4TBisRj29vZQVVVVuodNr9ejoqICer0eoVAIc3NzMBgMqK6uxvT0NHp6etDd3a1wenX4bA2y2SxisRj6+/tRV1eHs7MzzM7Owuv1wuVyKZxeHZaWluD3+2G1WvH4+IhYLIZkMomDg4PydqC8D+T//21ubgqr1SokSRIej0ecnJwoHelLCQQCwmKxCEmSRGNjowgEAiKTySgdS9USiYQA8N0rGAwKIf76u6VoNCrq6+uFTqcTPp9PpNNpZUOrzM/W4Pn5WfT29gqTySS0Wq2w2WxiYmJC3NzcKB1bVT76/AGIra2t0jkvLy9iampK1NbWisrKSjE0NCSur6+VC60yn61BLpcTXq9XGAwGodPphNPpFAsLCyKfzysbXEXGx8eFzWYTkiQJk8kkfD6fODw8LB0vVwf+EEKIfzspExERERH9Ct7zSURERESy4fBJRERERLLh8ElEREREsuHwSURERESy4fBJRERERLLh8ElEREREsuHwSURERESy4fBJRERERLLh8ElEREREsuHwSURERESy4fBJRERERLLh8ElEREREsvkTUqOdlPZ37jkAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for kinit in kinits:\n",
    "#for kinit in ['glorot_normal']:\n",
    "    #for lr in lrs:\n",
    "    for lr in [0.001]:\n",
    "        #for dr in drs:\n",
    "        for dr in [0.1]:\n",
    "            #\n",
    "            #a convolutional neural network described in Ch. 14, p. 447.\n",
    "            model = keras.models.Sequential([\n",
    "                    keras.layers.Conv2D(64, 7, activation=\"relu\",kernel_initializer=kinit, padding=\"same\",\n",
    "                                        input_shape=[28, 28, 1]),\n",
    "                    keras.layers.MaxPooling2D(2),\n",
    "                    keras.layers.Conv2D(128, 3, activation=\"relu\",kernel_initializer=kinit, padding=\"same\"),\n",
    "                    keras.layers.Conv2D(128, 3, activation=\"relu\",kernel_initializer=kinit, padding=\"same\"),\n",
    "                    keras.layers.MaxPooling2D(2),\n",
    "                    keras.layers.Conv2D(256, 3, activation=\"relu\",kernel_initializer=kinit, padding=\"same\"),\n",
    "                    keras.layers.Conv2D(256, 3, activation=\"relu\",kernel_initializer=kinit, padding=\"same\"),\n",
    "                    keras.layers.MaxPooling2D(2),\n",
    "                    keras.layers.Flatten(),\n",
    "                    keras.layers.Dense(128, activation=\"relu\",kernel_initializer=kinit),\n",
    "                    keras.layers.Dropout(dr),\n",
    "                    keras.layers.Dense(64, activation=\"relu\",kernel_initializer=kinit),\n",
    "                    keras.layers.Dropout(dr),\n",
    "                    keras.layers.Dense(10, activation=\"softmax\")\n",
    "            ])\n",
    "            opt=keras.optimizers.Adam(learning_rate=lr)\n",
    "\n",
    "            model.compile(loss=\"sparse_categorical_crossentropy\",\n",
    "                              optimizer=opt,\n",
    "                              metrics=[\"accuracy\"])\n",
    "\n",
    "            history = model.fit(X_train, y_train, epochs=30,\n",
    "                                validation_data=(X_valid, y_valid))\n",
    "            df=pd.DataFrame(history.history)\n",
    "            df.to_csv(f'cnn/{kinit}_relu_Adam_{lr}_{dr}.csv')\n",
    "            df.plot(figsize=(8, 5))\n",
    "            plt.grid(True)\n",
    "            plt.gca().set_ylim(0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "\n",
    "files=os.listdir('cnn/')\n",
    "saving=[]\n",
    "for filename in files:\n",
    "    if '.csv' in filename:\n",
    "        data=pd.read_csv(f'cnn/{filename}')[-1:]\n",
    "        kinit=filename.split('_')[0]+'_'+filename.split('_')[1]\n",
    "        act=filename.split('_')[2]\n",
    "        opt=filename.split('_')[3]\n",
    "        reg='drop'\n",
    "        drop=float(filename.split('_')[5][:-4])\n",
    "        lr=float(filename.split('_')[4])\n",
    "        acc=np.array(data['accuracy'])[0]\n",
    "        loss=np.array(data['loss'])[0]\n",
    "        val_acc=np.array(data['val_accuracy'])[0]\n",
    "        val_loss=np.array(data['val_loss'])[0]\n",
    "        read_out=[kinit,act,opt,reg,drop,lr,acc,loss,val_acc,val_loss]\n",
    "        saving.append(read_out)\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.DataFrame(saving,\n",
    "                columns=['kernal_init','activation','optimizer','regularization','regularization_value','learning_rate',\n",
    "                         'accuracy','loss','val_accuracy','val_loss'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>kernal_init</th>\n",
       "      <th>activation</th>\n",
       "      <th>optimizer</th>\n",
       "      <th>regularization</th>\n",
       "      <th>regularization_value</th>\n",
       "      <th>learning_rate</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>loss</th>\n",
       "      <th>val_accuracy</th>\n",
       "      <th>val_loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.982611</td>\n",
       "      <td>0.056435</td>\n",
       "      <td>0.915667</td>\n",
       "      <td>0.454133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.980833</td>\n",
       "      <td>0.060969</td>\n",
       "      <td>0.920167</td>\n",
       "      <td>0.412444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>he_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.979722</td>\n",
       "      <td>0.062843</td>\n",
       "      <td>0.914667</td>\n",
       "      <td>0.503487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.975519</td>\n",
       "      <td>0.078017</td>\n",
       "      <td>0.913500</td>\n",
       "      <td>0.392765</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>he_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.973926</td>\n",
       "      <td>0.082477</td>\n",
       "      <td>0.920833</td>\n",
       "      <td>0.433561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.973907</td>\n",
       "      <td>0.078267</td>\n",
       "      <td>0.914667</td>\n",
       "      <td>0.467344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.973296</td>\n",
       "      <td>0.083463</td>\n",
       "      <td>0.919667</td>\n",
       "      <td>0.378618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.971556</td>\n",
       "      <td>0.086238</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>0.408118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.971093</td>\n",
       "      <td>0.091530</td>\n",
       "      <td>0.916000</td>\n",
       "      <td>0.442338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.966815</td>\n",
       "      <td>0.104892</td>\n",
       "      <td>0.917667</td>\n",
       "      <td>0.372663</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       kernal_init activation optimizer regularization  regularization_value  \\\n",
       "0    glorot_normal       relu      Adam           drop                   0.1   \n",
       "7   glorot_uniform       relu      Adam           drop                   0.1   \n",
       "13      he_uniform       relu      Adam           drop                   0.1   \n",
       "2    glorot_normal       relu      Adam           drop                   0.2   \n",
       "11       he_normal       relu      Adam           drop                   0.1   \n",
       "6   glorot_uniform       relu      Adam           drop                   0.3   \n",
       "5   glorot_uniform       relu      Adam           drop                   0.2   \n",
       "1    glorot_normal       relu      Adam           drop                   0.3   \n",
       "8   glorot_uniform       relu      Adam           drop                   0.4   \n",
       "4    glorot_normal       relu      Adam           drop                   0.4   \n",
       "\n",
       "    learning_rate  accuracy      loss  val_accuracy  val_loss  \n",
       "0           0.001  0.982611  0.056435      0.915667  0.454133  \n",
       "7           0.001  0.980833  0.060969      0.920167  0.412444  \n",
       "13          0.001  0.979722  0.062843      0.914667  0.503487  \n",
       "2           0.001  0.975519  0.078017      0.913500  0.392765  \n",
       "11          0.001  0.973926  0.082477      0.920833  0.433561  \n",
       "6           0.001  0.973907  0.078267      0.914667  0.467344  \n",
       "5           0.001  0.973296  0.083463      0.919667  0.378618  \n",
       "1           0.001  0.971556  0.086238      0.916667  0.408118  \n",
       "8           0.001  0.971093  0.091530      0.916000  0.442338  \n",
       "4           0.001  0.966815  0.104892      0.917667  0.372663  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nlargest(10, 'accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>kernal_init</th>\n",
       "      <th>activation</th>\n",
       "      <th>optimizer</th>\n",
       "      <th>regularization</th>\n",
       "      <th>regularization_value</th>\n",
       "      <th>learning_rate</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>loss</th>\n",
       "      <th>val_accuracy</th>\n",
       "      <th>val_loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>he_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.973926</td>\n",
       "      <td>0.082477</td>\n",
       "      <td>0.920833</td>\n",
       "      <td>0.433561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.980833</td>\n",
       "      <td>0.060969</td>\n",
       "      <td>0.920167</td>\n",
       "      <td>0.412444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.973296</td>\n",
       "      <td>0.083463</td>\n",
       "      <td>0.919667</td>\n",
       "      <td>0.378618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>he_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.951019</td>\n",
       "      <td>0.156630</td>\n",
       "      <td>0.918500</td>\n",
       "      <td>0.288058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.955648</td>\n",
       "      <td>0.135355</td>\n",
       "      <td>0.917833</td>\n",
       "      <td>0.325780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.966815</td>\n",
       "      <td>0.104892</td>\n",
       "      <td>0.917667</td>\n",
       "      <td>0.372663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.971556</td>\n",
       "      <td>0.086238</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>0.408118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.971093</td>\n",
       "      <td>0.091530</td>\n",
       "      <td>0.916000</td>\n",
       "      <td>0.442338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>glorot_normal</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.982611</td>\n",
       "      <td>0.056435</td>\n",
       "      <td>0.915667</td>\n",
       "      <td>0.454133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>glorot_uniform</td>\n",
       "      <td>relu</td>\n",
       "      <td>Adam</td>\n",
       "      <td>drop</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.973907</td>\n",
       "      <td>0.078267</td>\n",
       "      <td>0.914667</td>\n",
       "      <td>0.467344</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       kernal_init activation optimizer regularization  regularization_value  \\\n",
       "11       he_normal       relu      Adam           drop                   0.1   \n",
       "7   glorot_uniform       relu      Adam           drop                   0.1   \n",
       "5   glorot_uniform       relu      Adam           drop                   0.2   \n",
       "10      he_uniform       relu      Adam           drop                   0.5   \n",
       "3    glorot_normal       relu      Adam           drop                   0.5   \n",
       "4    glorot_normal       relu      Adam           drop                   0.4   \n",
       "1    glorot_normal       relu      Adam           drop                   0.3   \n",
       "8   glorot_uniform       relu      Adam           drop                   0.4   \n",
       "0    glorot_normal       relu      Adam           drop                   0.1   \n",
       "6   glorot_uniform       relu      Adam           drop                   0.3   \n",
       "\n",
       "    learning_rate  accuracy      loss  val_accuracy  val_loss  \n",
       "11          0.001  0.973926  0.082477      0.920833  0.433561  \n",
       "7           0.001  0.980833  0.060969      0.920167  0.412444  \n",
       "5           0.001  0.973296  0.083463      0.919667  0.378618  \n",
       "10          0.001  0.951019  0.156630      0.918500  0.288058  \n",
       "3           0.001  0.955648  0.135355      0.917833  0.325780  \n",
       "4           0.001  0.966815  0.104892      0.917667  0.372663  \n",
       "1           0.001  0.971556  0.086238      0.916667  0.408118  \n",
       "8           0.001  0.971093  0.091530      0.916000  0.442338  \n",
       "0           0.001  0.982611  0.056435      0.915667  0.454133  \n",
       "6           0.001  0.973907  0.078267      0.914667  0.467344  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nlargest(10, 'val_accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}