Tackling_MNSIT_Dataset_Using_Callbacks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "nd9GI4PzUjuR"
   },
   "source": [
    "## Handwritten Digit Recogonition (MNSIT Dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "DjsYPH0WVrrM"
   },
   "source": [
    " ![MNSIT.jpg](data:image/jpeg;base64,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)\n",
    "    ![image.png](data:image/png;base64,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)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Np2T2-4hWb3H"
   },
   "source": [
    "**History of Handwritten Digit dataset**\n",
    "\n",
    "Modified National Institute of Standards and Technology database (MNIST dataset) is a large dataset of handwritten digits which is widely used in image processing and machine learning. The set of images in the MNIST database is a combination of two of NIST's databases: Special Database 1 and Special Database 3. Special Database 1 and Special Database 3 consist of digits written by high school students and employees of the United States Census Bureau, respectively.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "OpyRaz3XaI7m"
   },
   "source": [
    "**Why to start with MNIST dataset?**\n",
    "\n",
    "The MNIST dataset is a well-known dataset for image classification. Tensorflow and Keras also provide MNIST dataset directly through their APIs. For the learning purpose, the MNIST dataset is easy to use and experiment with different machine learning techniques."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Ft12-eaOaVYF"
   },
   "source": [
    "**About the Notebook**\n",
    "\n",
    "\n",
    "*   In this notebook, I have covered the necessary steps to approach any Machine Learning Classification Problem.\n",
    "*   Included Image Visualization for better understanding.\n",
    "*   Quick Links to the functions I have used to explore it in depth.\n",
    "*   Basic techniques such as k-folds, using Callbacks, etc.\n",
    "\n",
    "I have tried to make this notebook as simple as possible, along with covering the basic approach to tackle any classification task.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "cGZpD73RbJBg"
   },
   "source": [
    "**Task**\n",
    "\n",
    "The task is to classify the images in 10 class, i.e., [0-9], inclusively."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Il7Mch-vbvYI"
   },
   "source": [
    "## Required Imports\n",
    "\n",
    "**Imports:**\n",
    "\n",
    "\n",
    "1.   pandas : For handeling csv dataset\n",
    "2.   numpy : Support for Pandas and calculations\n",
    "3.   Matplotlib - For visualization (Plotting graphs)\n",
    "4.   keras - Prediction Models\n",
    "5.   tensorflow - Prediction Models\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "jzaHWbYlaHiL"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import keras\n",
    "import os\n",
    "\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "kEQ8mVVGefWf"
   },
   "source": [
    "## Loading and Visualizing Dataset\n",
    "\n",
    "\n",
    "---\n",
    "\n",
    "\n",
    "**About Dataset**\n",
    "\n",
    "MNIST dataset has the following features:\n",
    "\n",
    "\n",
    "*   Dataset size 60,000 samples of handwritten images.\n",
    "*   The size of each image is 28x28 pixels.\n",
    "*   Each image has only 1 color channel, i.e., grayscale image.\n",
    "*   Each pixel has value in the range of [0,255] where 0 represents black, and 255 represents white.\n",
    "*   Each image has labeled from 0-9.\n",
    "\n",
    "**Loading Training and testing Data**\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "id": "6D3y4kkieRxS",
    "outputId": "4de0c5d8-ecc7-4cdc-8e22-2d3c1121ec47"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
      "11493376/11490434 [==============================] - 0s 0us/step\n"
     ]
    }
   ],
   "source": [
    "(X_train_full, y_train_full), (X_test, y_test) = keras.datasets.mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "id": "F_cULG-2gNDA",
    "outputId": "2886561f-2e13-4619-8ca7-77d311ad13d1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The shape of training dataset - (60000, 28, 28)\n",
      "Each pixel intensity is also represented as a byte(0-255) - uint8\n"
     ]
    }
   ],
   "source": [
    "print(\"The shape of training dataset -\", X_train_full.shape)\n",
    "print(\"Each pixel intensity is also represented as a byte(0-255) -\", X_train_full.dtype)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "yyHLKaofheP6"
   },
   "source": [
    "Prepairing **Validation data** and **Normalizing the data**\n",
    "\n",
    "**It requires a few steps:**\n",
    "\n",
    "\n",
    "*   Assuming the validation set size. I am taking it 10% of the training set. \n",
    "*   Splitting training set into a training set (90% original training set) and validation set (10% original training set) from the training dataset.\n",
    "*   scaling down the pixel intensities down to the 0-1 range and convert them to floats, by dividing by 255\n",
    "*   Splitting the labels for both training set and validation set.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_MUp5d3-gxMZ"
   },
   "outputs": [],
   "source": [
    "X_valid, X_train = X_train_full[:5000] / 255., X_train_full[5000:] / 255.\n",
    "y_valid, y_train = y_train_full[:5000], y_train_full[5000:]\n",
    "X_test = X_test / 255."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8_OrhEgrisKK"
   },
   "source": [
    "Shape of training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "id": "7J59svO1hMwz",
    "outputId": "57295472-782a-4bc7-e3a5-e0f9a3191591"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((55000, 28, 28), (55000,))"
      ]
     },
     "execution_count": 5,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, y_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "RZLqxXCKjJUQ"
   },
   "source": [
    "## Visualize Digits dataset\n",
    "\n",
    "\n",
    "---\n",
    "\n",
    "**Visualizing the digits by plotting Images**\n",
    "\n",
    "This will plot the first 40 images of diits with the label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 308
    },
    "colab_type": "code",
    "id": "3NLkuz79i7V9",
    "outputId": "75adf8fa-619d-4f7a-eed1-7a35aa37e02c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqIAAAEjCAYAAADpBWMTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdebyN1ffA8c8KJUMDRaUMlZKhNHybxferQZOSkgakSSlNmmk0Nc8lDSJKo1QqzYV+DTTQSBqElCFlCIX9++Ox9nPOvede1x3OPsN6v15ernPPvXff7Rn2s/baa4tzDmOMMcYYY9Jtg9ANMMYYY4wx+ckGosYYY4wxJggbiBpjjDHGmCBsIGqMMcYYY4KwgagxxhhjjAnCBqLGGGOMMSYIG4gaY4wxxpgg0j4QFZGlBf6sFpF7092OTCIiI0VkrogsFpHpInJm6DZlAhFpLCIrRGRk6LaEJiKdReRbEVkmIj+ISKvQbQpFRM4XkckislJEhoVuT2gi0lBEXhWRRSLym4jcJyKVQ7crFDs+ChORWiLywtrrx0wROTl0m0ISkV1E5B0R+UtEZohIh9BtCin0NSTtA1HnXA39A2wFLAeeTXc7MswgoKFzbhOgPdBfRPYM3KZMcD8wKXQjQhORQ4Cbge5ATeAg4MegjQrrV6A/MDR0QzLEA8A8YGugJdAa6Bm0RWHZ8VHY/cA/QF3gFGCwiDQL26Qw1g6wXgTGArWAs4GRIrJT0IaFFfQaEnpqviPRLz8hcDuCcs597Zxbqf9c+2eHgE0KTkQ6A38Cb4duSwa4AbjROfeRc26Nc26Oc25O6EaF4pwb7ZwbAywM3ZYM0Qh4xjm3wjn3GzAOyMtBBtjxUZCIVCe6117jnFvqnJsIvAR0CduyYJoA2wB3OudWO+feAT4gf/sDAl9DQg9EuwGPO9tnFBF5QET+Br4D5gKvBm5SMCKyCXAjcEnotoQmIpWAvYAt104hzV47bbJx6LaZjHEX0FlEqolIPeBwohuJMQA7Aaucc9MTXptCHj+spCBA89CNCCjoNSTYQFREGhCFf4eHakMmcc71JJp2bQWMBlYW/xU5rR/wqHNuduiGZIC6QBXgeKJjoyWwO9A3ZKNMRhlPNKhYDMwGJgNjgrbIZJIaRMdGor+I7jf5aBrRTOxlIlJFRA4lGotUC9usoIJeQ0JGRLsAE51zPwVsQ0ZZO00wEdgWODd0e0IQkZbAwcCdoduSIZav/fte59xc59wC4A7giIBtMhlCRDYgilyMBqoDWwCbE+UUGwOwFNikwGubAEsCtCU459y/wLHAkcBvQG/gGaIBWN7JhGtIyIFoVywaWpTK5G+OaBugIfCLiPwGXAp0FJHPQjYqFOfcIqILZGL6St6nshivFlAfuM85t9I5txB4DHtQMbHpQGURaZzw2m7A14HaE5xzbqpzrrVzrrZz7jBge+CT0O0KJPg1JMhAVET2B+phq+URkTprS/PUEJFKInIYcBL5u0jnIaJBeMu1fx4EXgEOC9mowB4Deq09VjYHLiZa8ZmXRKSyiFQFKgGVRKRqvpYrWhsh/wk4d22/bEaUez81bMvCseMjmXNuGVG060YRqS4iBwDHACPCtiwcEdl17XFRTUQuJVotPixws4LIhGtIqIhoN2C0cy4vpwYKcETT8LOBRcBtwEXOuZeCtioQ59zfzrnf9A/RtNIK59z80G0LqB9RGavpwLfA58CAoC0Kqy9RysKVwKlrP87nnNnjgHbAfGAG8C/Rw0q+suOjsJ7AxkS5kaOAc51zeRsRJUoNnEvUH22BQxIq1+SjoNcQsQXrxhhjjDEmhNDlm4wxxhhjTJ6ygagxxhhjjAnCBqLGGGOMMSYIG4gaY4wxxpggbCBqjDHGGGOCWFdttWxfUi/l/P2sP5JZfySz/ijM+iSZ9Ucy649k1h/JrD+S5WR/WETUGGOMMcYEYQNRY4wxxhgThA1EjTHGGGNMEHm7/64x2WrNmjX07t0bgPvuuw+ADz/8EIC99torWLuMMcaY9WURUWOMMcYYE4RFRI3JEvPmzQPgmmuu4aGHHkr63E8//QTkX0T0rLPOAmDkyJF88MEHAOyxxx4hm2Qy0I033shTTz0FwNixYwHYfvvtQzYprb755hsA7rrrLgAefvhhevToAcCDDz4YrF0mvHnz5jFlyhQAXnzxRQDGjx/PV199BUD37t0B2GGHHQDo3bs3G220UdL3+OOPP6hVq1ap22ARUWOMMcYYE4RFRDPAzJkzgegpFWDAgAGIROW2nIvKhu2yyy4A9O/fn+OOOy5AK00oc+fOBeCWW24BSIqGtmrVCoB99tkn/Q3LAA0aNABgxYoVfP/994BFRAEmTpzIkCFDgChaXJAeN3ot6dq1a5kiGplq4cKFQHRtnT17NgCfffYZkD8R0eHDh3PNNdcA+D4QEV599dWU7x85ciTHHHMMADVr1kxPI03aPfLIIwAMHDjQj0GUc86PQYYNG5b0uY033piLL7446bWTTjqJ119/vdRtsYFoIPPnzwdg0KBBPPHEEwAsWLAAiC4SehCoadOmAVFY/KCDDgJgiy22SFdzK8w///wDQNu2bYHoBqo222wzAKZOncp2222X/sZlgFWrVjFgwAAA7r//fv/6eeedB8Add9wBwIYbbpj+xmUAHYhCdMMFOPHEE0M1J5hVq1YBcP311wPRsfLXX38BFLqWAEyYMAGIz7cvvvii0A0nF+gxoQOwfPDvv/8C+IHB2Wef7V8rzuDBgwG44IILaNSoEQD9+vUDcuuc+uGHH3yKgqbzfPvttz5FoVu3bsHalg466Bw4cGDSvyEaZALUqFHDXzd0XLJmzRoALr30UjbddFMATj/9dAB+/fXXMrXJpuaNMcYYY0wQaYuIPvbYY0D0dF67dm0gegoB2G+//fxUUa7r378/gJ8qERE//a5PIPXr12fLLbdM+jp9Kvn55599RFQT0LORRkLPOOMMIDkSeuyxxwJw5ZVXArDNNtsU+71+//13AOrWrVvu7QztqquuSoqEAvTo0cOXbTKxfI0KA/Tp0weAW2+9FUieWivooIMO4v3330967Y033mDJkiVAbk3Hvvfee6GbkHY6S3LVVVcV+Z4mTZpw4YUXJr2m95jVq1czY8YMAM455xz/+WyNimo0+OmnnwaiiKdeK/S8mTx5ct5ERPUaoZHQDTfckBNOOAHAT7nvvvvu/v3PPPMMADfddBMAU6ZMYcWKFUnfc1336HWxiKgxxhhjjAlivSOiTz75JACff/45AEOHDi3R1/3555/xD60c/ViNilWtWpVq1aoBsOuuuwLxKLxgZDDbaXkEjVYkRi2aNm0KRE/xBfM/NaerdevWPl80m91+++1A4YUU5513HrfddhsQHRfr0rt3bx9tv/baawG46KKLyrOpQVx33XUAvi8Azj//fCCOeBh44YUX/McnnXRSwJakn+aF9unTp9AxUb16dS655BIAOnToAEQzLQCbbLKJz+3S/PQtttjCX5dzgc6waA5gPtDIn5biSUVz7R966CEOPPDAdX5PzTPu0aMHkydPBuKIWqbT8YXOPupiz2bNmnHnnXcCcMghhwBRDvGsWbOA+F6r+ZK5VhJv1KhRSf8+8MADefzxx4t8f6dOnQCoU6cOEK/nSKSL20qrxFcevajdfffdQJy4Whp6gKgVK1b4UK9Opeg0wKhRo3JiylXTEL777jsgvilsueWWftCpN5O+ffty9dVXJ71PUxd0Gh/i1dNnn312RTe/XH311Vc+CV7pdOBdd91VohvipEmTgGhF36JFi8q/kYF89NFHANx7773+Na33p+feBhvYRIY+CL/yyitANJBq3759yCalnQ4iEwcGO++8MxA9yLdo0aLIry2YxrDjjjv6G28u+OOPP5L+znWrV6/2x4HWS02k6VzPP/88gE+PS3TkkUcCUU3iESNG+O8LsHjxYpo1a1b+Da8gK1eu5MwzzwTiYIeeD8OGDStUWWPbbbf19yD9PbVSzZtvvpmWNqeLnhMaBCvp/2vjxo2BKAWuefPmSZ8ry3gQbGreGGOMMcYEUuKI6LPPPgvEI1+dQi/qKfqAAw4A4oUnxXnrrbd8aPjnn38G4N133wWi6TZNMs7maXp9utJInkZBE6fgNcL50EMP+SinRkRHjx4NJJd2ytZ6ojfddBPLly8HoEqVKgC89NJLACWeHtQp6z/++MNHd0pyrGU6TS/QKO/RRx/tp5YsEhrTWRX9e4MNNsipiF5J6OIB5xwtW7YEYNy4cUDqhXt///03EC3a0Klrvf7o9SWXbbXVVkAU/co1kyZNom/fvik/t//++/Pyyy8DxS9E0yjh0KFD/WI23bEtW6xcuRKIUps0EqpjFS1npcdBQTrGmTNnDhDPGixbtozq1atXXKPTTFN1NE3w6aef9uWsUtGUjMsvvxyApUuX+pKCGmkv673J7mzGGGOMMSaIEkdE3377bQC//6gm+ZZHqY9WrVr5kgmap6K5lO+++66Plvbu3bvMPyu0Jk2aFPk5jU7svPPOPodHk6oTox8aGc7Wgvaffvqp/7hdu3YAtGnTxr+meUkFc4khKkYMJJWf6dixIwANGzYs76am3Zdffpn077POOot69eoFak3m0lw3E82S6PUhMRKqs1dffPEFAKeeeioQXVs111yvt7lGr5uJNDK27777prs5FUZzOTVClWj//fcHont3wb3Bc5VGfm+++WY/m6izBEVFQlXigmqIN1TJpWgo4KOf06dPB6LNcrTUl5ZvGj9+vD+m9J67bNky/z10xvr//u//APwMZ2lZRNQYY4wxxgRR4ojoTjvtlPR3edN9f3U1tRZYhTgamAsRUTV+/Hggik5oZFPzSKdNm+b3Dp83bx4Qr3CrU6cOr732WrqbW2E0p0d98sknPtepJKsVt9pqK19hIJuNHTsWgN9++w2I83+POuqoYG3KZHPnzg3dhIyipVUSaSQ0VfkZnYlItcI6F6Ta7CMXcsiVRqn02qd5jRDn7Wl0cH2jod9//31S9Atg00039ffoTLRw4UIALrvsMiDaolIL1G+99dbr/Pq5c+fy3HPPVVwDM4hGirVEYOfOnX1pK/27uA0x9t57bw477DAgXknfo0ePMo3PcqdwXJbReqwPPfRQoZ2VnHN+AKqf0+n4Xr16FSo9kW2uuOIKunfvDsQh/v/9739ANOW+PqUgzjrrrEKlJLJRwcUixx9/PJB6n/DirFmzxhY15QmdOoR48LHbbrsB0Q2i4I1VByS9evXixhtvBEpWqzdX5FIagqYjJQ5AldbTLW3a3IMPPujvP6pevXr+GMtEWu9UFzvvvvvuHH744UW+X9O/hg0bBkT7rv/4448V2sZMoUGwktajbt26NYDfzW+HHXYo91QPu2MZY4wxxpggMiYi+sADDwBxqYBEmgiri1z23HPP9DWsgiVGvFJ9rE+h+vSS7dFQgF9++cV/rLuBaGQU4sUEWmZizpw53HPPPSm/V67selGw8HaqgtOpfPjhhwB+Gmr27Nm+DEmtWrXKsYWZ459//ilUVqa4RYC56tFHHwWgefPmfipVFw988MEHhaLpeg6dddZZaWxl+o0YMcJHyFSNGjWoVKlSoBaVr2eeecYv5lXVq1dnv/32A0of+dW0IC0jmKise4mn26xZs/x1sGBZt5deesnv3KjHScOGDbniiiuAaKETrHtxU7YZM2YMEJcI1IXnqTjn/PVCd/QrTuJGO6VhEVFjjDHGGBNE2iKiurhg5MiRKUtrFLf4QJ/2NY+w4NNuNjr55JMBmDlzJgsWLADiklVLly7179NcrlyIhKrTTz+90BaDqnPnzn4/ZI1gDBo0qND7dJ/kI444ooJamT6LFi3y5dFKYtmyZX5WQCODiaWudDtezX/KNcuWLSu0h/jBBx8cqDXpp8XoNc+8qGiEvq6LdHI9Eqrldx599NFCiyAvvvjinCmD9vPPPxcqbde8eXPeeOONMn3fhx9+GEgu06O5gBotzFSNGjUC4gU4N9xwg98jPRW9x+ji6HPOOcfvNa8RUS1/lQvmzZvHhRdeCOB/T50x2Wijjfz2yFr0/6+//qJatWol/v7ru5ahoAobiL711ltAPJ0+ZMgQoGw7NZx++ullb1iG0Cn3xARwHYj26dPHh9F1JZqulM/W2qGJtt12W6688soSvz9VHbcLLrgAKPlOTJls1apVSQ8fRRk1ahQQrWycNm1ake/LhQe14qR6aNVV4Lnqxx9/9Nc/raGrF//Em8Dee+8NRHV5dS/6d955B4irUGgN6FyjA9HEGsM6kNphhx2CtCldjjnmmFJ/rT6w6AKeRJom1bZt21J//3TQc+D6668HoGnTpv4eqnSqvVOnTilryWpVAN2lTGsVF7VjVTbQQeduu+3m7wu6iE1/r9NPP92ngvXs2ROIUr206sJpp50GFL970rnnnlumdtrUvDHGGGOMCaJcw0nff/89EIW59Sk8lQYNGgCw+eab+9c0RK7lRDRBNjHyky0J0/PnzwfikkslpQsunn/+eV96QneF0H1zL7roovJqZtZIfBLTj3fcccdQzSl31apVY+eddwYoFOlcvHgxTz/9NABnn312ib5fru+5rtcKiOus5lLqSiJdcNG1a9dC081qn3328QtUNKJRq1YtPzWpC/p0ai5Vjc1ckGp3F73H6M59ueqAAw4o9de+8sorQJwGlkjT4bJNp06dip2aT2XJkiVAvHC0pAtGM1n//v2BaJZMU1N0EVKqurq6aPynn37ipZdeAuIUIN2ZLRW97pSWRUSNMcYYY0wQ5RIR1cVHWvD0xx9/pEaNGkC0IwPEe5hus802PglYI6Op6NdBnNOQDbvMjB8/3ud1aoRT9wNeH7pjhiYPF5cTmOsSy4kceuihQFSwOFdUr17dHyv6/3zNNdcAUZK5FmkuiZYtW/q9hHNV4sIujXjlSmkeped9165dgWgHMi1gr3um6/7Q//3vf1Mu/tNcNy3XMnDgQCDavUxzSXOJRnwT6Q4wue7aa69NKoG3LgsWLPDlv3SBTyLNqe3SpUv5NDAL6EymlhfU8oHZ7MUXX/Qfa2RTF/oW55hjjvGL33TP+eIiomVlEVFjjDHGGBNEuUREtai2bpHVvn17HxVc323BdH/kmTNn+td05aPuxZ6J9GmqR48e1K1bFyhdJBSi8hk9evQAyl4oNpvpKr/Fixf713I1R1b/v3Wl4ieffFKir9PVolqap1+/fin3Hc8Fv//+OxBvgpDLpkyZAuDzQhs0aOBXvZc0P1pL/Hz88cdAVJ0h8e9codfeRYsW+dc0t1Fn6XLd3Llz/XafqcpUaZRPKykMHjyY2bNnF/n9tEJHw4YNy7mlmeu9995L+ncuVKjR8YNzbr02OOnUqZOf6dbtXvU+vMkmm5RzK8tpIKq7uuiUUVnKHcyYMQOIbzqQHTUCX3jhBSCaWm3Tpk2pvse3334LRPsI6xStDjTycecYHYzNnDnTTz3m6m5BujhNB5G6y0lRdD9prUebDWkrZaWLtbRMD8S/f67SG8nxxx+/Xgv0Fi9ezPHHHw/EZZtylU5JJ+7KpzUQtbzbqlWrcqLUG0TT5bqA8fPPPwdg+vTpfvCd6hq5cOFCIL6/pqKpcp07d6Z58+bl2uZsUHB3u1ygKRYLFizg9ttvB+KUnuKuJ5UqVfL3XL3e6lS9XlcSvf7662VKg7GpeWOMMcYYE0S5PCLqE1h5FH7VaX612Wab+eLlmaxVq1ZAFMHQgspacmmXXXbxO+EoTT2YMGECo0ePBuK9YJ1zPhKqU9GpEvFzXa9evfzHuvjtP//5T6jmBNG9e3e/6OSMM84AohJWuV6iKZFOIermGBDPkuTqYpTddtsNiMvZJU4x9+nTB8AvXoI44qUzKSeffLKfjtVrSdOmTYHcWuhXlLFjxwJxKbNrrrkmZXmibLT11lv7e63OCKxcudKXTyypKlWqAHHKm0ZZtZScyX660cHHH3/sd9rTknAa9U51Db377rt9apymKBx99NFF/pxLL73UIqLGGGOMMSb7ZEzSTIsWLYB4m0t16KGHst9++4Vo0nrRp8rjjjvORza19IqIFCq4rdGKBQsW+DywxK369Ik3G6LBFSWxgLdGiPKFFh3u2bNnzpUmWl+aLK+LMSAuUF7WPY4zlUYXbr31ViC6DmiO19ChQ4HkhaC68YWeM4mzKvvssw8Q7yWea9F0nZHTkn+JW9xq1C9X9plXWlpIZ9q++eabpNzpdWnatKkv23TCCSeUfwNzgK55yWa6CPauu+7y11HdTloXMerfiRKvH3ru6KLxVMo6U5kxA1GtlagrOvWikm2rpB988EE/yExMnteP9T83cfCpifU6mL3qqqs47rjj0tbmbJAvg7FU+6ibZK1ataJ9+/ahm5EWek1o0qSJH2joMZJYI7CgJk2acMoppwBw+eWXA6SsNZoLNE1D0xe6dOni01m0ektF1kAMaeLEiQD8+uuvvk6k7pGuA4xBgwYVun6ecMIJxdbxNtC4cePQTSgzTd+ZNGmSfxDVQNlXX31V5Ne1bt3aT+vrdaQ4+nBcWjY1b4wxxhhjgpB11KlMSxHLUaNG+SfW6tWrA/DII48ArPd+sQWU97xdifpjwYIFQLw7DsCQIUOAqDQTJNco04VIaSjRFKQ/SqtRo0ZAFC3XaI4u1NDdYsooq/ojDSpintv6JFmp+0NL2hVcFPrWW2/52sU6k6JR0AqQMf2RIaw/kmVtf9x2220AXHbZZUCU7gBlrl+etf1RQVL2h0VEjTHGGGNMEEFzRHWHlFtuucVHvLRYahkjoUFptHPw4MH+tcSPTclo+aZ+/fr5/LgNNrBnJ5OfNOqpuV7GmPKnOwfVrFkzcEvyh93VjTHGGGNMEEFzRHWF/J133ulXOR5yyCHl+SMsPyOZ9Ucy649kliNamB0jyaw/kll/JLP+SGb9kSxlf2TEYqUKZAdBMuuPZNYfyWwgWpgdI8msP5JZfySz/khm/ZHMFisZY4wxxpjMsa6IqDHGGGOMMRXCIqLGGGOMMSYIG4gaY4wxxpggbCBqjDHGGGOCsIGoMcYYY4wJwgaixhhjjDEmCBuIGmOMMcaYIGwgaowxxhhjgrCBqDHGGGOMCSLIQFREaonICyKyTERmisjJIdqRKURkpIjMFZHFIjJdRM4M3aaQROR8EZksIitFZFjo9oQkIhuJyKNrz5MlIvKFiBweul2hiMjSAn9Wi8i9odsVkl1Pk9k5U5iI7CIi74jIXyIyQ0Q6hG5TaCLSWUS+XXve/CAirUK3KZTQ15DK6fxhCe4H/gHqAi2BV0RkinPu60DtCW0QcIZzbqWINAHeE5HPnXOfhm5YIL8C/YHDgI0DtyW0ysAsoDXwC3AE8IyItHDO/RyyYSE452roxyJSA/gNeDZcizKCXU+T2TmTQEQqAy8CDwKHEPXLyyKyu3NuetDGBSIihwA3AycCnwBbh21RcEGvIWnf4lNEqgOLgOZ6EojICGCOc+7KtDYmA4nIzsB7wIXOuWcCNycoEekPbOucOy10WzKJiEwFbnDOPR+6LSGJSDfgOmAHl6d7Fdv1tGTy+ZwRkebAR0BNPU9E5A3gY+fcNUEbF4iI/B/wqHPu0dBtCS0TriEhpuZ3AlYVeBKbAjQL0JaMISIPiMjfwHfAXODVwE0yGUhE6hKdQ/ka7UrUDXg8Xweha9n1dB3snElJgOahGxGCiFQC9gK2XJumMFtE7hORfJ19C34NCTEQrQEsLvDaX0DNAG3JGM65nkR90AoYDawM2yKTaUSkCvAEMNw5913o9oQkIg2IphiHh25LYHY9LYadMwBMA+YBl4lIFRE5lOjcqRa2WcHUBaoAxxPdb1sCuwN9QzYqoODXkBAD0aXAJgVe2wRYEqAtGcU5t9o5NxHYFjg3dHtM5hCRDYARRHk85wduTiboAkx0zv0UuiGB2fW0CHbORJxz/wLHAkcS5VT3Bp4BZodsV0DL1/59r3NurnNuAXAHUS5xPgp+DQkxEJ0OVBaRxgmv7YZNmySqDOwQuhEmM4iIAI8SPcl3XHtjyXddsWgo2PU0JTtnkjnnpjrnWjvnajvnDgO2J1qkk3ecc4uIBuGJKT35nN4T/BqS9oGoc24Z0dTzjSJSXUQOAI4henLNOyJSZ20ZiRoiUklEDgNOAt4O3bZQRKSyiFQFKgGVRKTq2pWf+WowsAtwtHNu+brenOtEZH+gHrZa3q6nRbNzJoGI7Lr2OlpNRC4lWiU+LHCzQnoM6LX2/rs5cDEwNnCbgsiEa0iogvY9icryzANGAefmcakRRzQNP5to5dptwEXOuZeCtiqsvkTTJ1cCp679OC/zd9bmQvYgymP6LaF+5imBmxZSN2C0cy7vp5/XsutpAjtnUupCtAh2HtAWOMQ5l8/rEPoBk4iigd8CnwMDgrYorKDXkLSXbzLGGGOMMQZsi09jjDHGGBOIDUSNMcYYY0wQNhA1xhhjjDFB2EDUGGOMMcYEYQNRY4wxxhgTxLpqM2b7knop5+9n/ZHM+iOZ9Udh1ifJrD+SWX8ks/5IZv2RLCf7wyKixhhjjDEmCBuIGmOMMcaYIGwgaowxxhhjgsjn/buNMcbkmDVr1vDzzz8nvTZs2DBatmwJwH777QfA1ltvne6mmSzQt2+0m/SCBQsA6N69O/vss0/IJuU8i4gaY4wxxpggLCKaZpMnTwbg22+/BeD3339n2rRpAIwfPx6A6dOns+222wJw7bXXAnDWWWelu6nB9OrVC4D7778fgHfeeYc2bdoEbJEx2UEjgS+//DKjR48G4L333gNApPCC1XfffReA1q1bp6V9FWnSpEkA3HLLLTz//POFPu9ctOC4Tp06AP49Bx54YJpaaDLVlClT/D126tSpAKxcudL/rdH0jTbaKEwD0+SOO+4AoE2bNn7GIB0zBxYRNcYYY4wxQVR4RFSfQp966ikAbrjhBh8BTGXnnXcG4O233wagbt26VK6c/YHbsWPHAtChQwcAVq1aBSRHKbSvRIQ5c+YAcP755ye9/9xzz01PgwPSPtG/33jjjZyPiP72228AvPbaa0AcMf/mm2949dVXAejduzcARxxxBKYqtdAAACAASURBVLvssgsAG2+8MQCbbropAKtXr+bxxx8HYNmyZQD06NGDKlWqpOPXMIHocXP11VcDcVQHCp9PiY499lggigjVr1+/optZrpYvXw7AqaeeCsDrr78OwN9//+3fc+SRRwJRVGfJkiUAPP300wAcc8wxAMyePdufRya/XHXVVUA0PimYV6yGDRvm37fTTjulq2kVTsdYd999N1OmTAFg1qxZAGy22WY++tugQQMAPvroowprS4WN8NasWQPE06sXXHCB/9wGG0SB2OrVqwPRIEsvKjpI1anp5s2b89ZbbwHRoDRb6TTQ6tWrgfimULNmTfbaa6+k9+66664sXboUgJEjRwIwatQoAM4888y8G1R89dVX/PvvvwA5+bsPHz6c7t27A6kHC/ra7bffDsTTJwDbb789gB98TpgwwV80VevWrWnRokX5N9wE9c8//wDR8aAD0FTHT3H++usvAO677z5uueWW8m1gBdOB5Pvvvw/ED+lHHXUU+++/PxBPpVaqVMnfk/Qa/NxzzwHR737ZZZelr+FpotfMWbNmccMNNwDxdaI4F1xwAddddx0Am2++ObD+x1UmW7p0qU9beeCBBwBYvHhxke9v1qwZm2yySVralg4LFy4E4JJLLgGi+2tBel0A+PPPPwH8OfXkk0/SsGHDcm2TTc0bY4wxxpggRKeDi1Dq7aQeeughIJoWTFS5cmX/tKVlEn755Rf/ND5kyBAgnoqGKCoK8MEHHwCsz9NJxmyvpRHOI444Aoiju3feeaeP/qZy+eWXA3DbbbcB0dN7z549S9uMjOmP4mj0XKPpzjn/hFazZs3y/FFB++PXX38FoEWLFixatChqUIrIg06N6NRRcdEJ55z//BZbbAFEUyqNGjUqSZOCb/E5YsQIAD7++ONS/0CdXXnsscf8axoNK4WMO2f0mn3zzTcD0KdPn6S0nqLerwsfAfr165f0uYYNG/LKK68A+LSPImRMf+i1QKOZib9fcXTBqKb79OnTp9AswnrImP7QY1yvE4cffjgA33//fakbo2kMJ5xwQkm/JGP6oyg9e/Zk8ODB63xfvXr1gOh+pPfhUsiY/tDZ1QcffBCADz/8sNB79JzafPPNWbFiBQDz5s1Les9FF13kZ+c0WrrZZpuVtBm2xacxxhhjjMkcFZIjunr1al8ypKArr7zSR0JV/fr1ue+++4C4jMiFF14IwNy5c30OgyahZ2O+Ro0aNYD499IIVXHR0MSvUy+88EJZIqImg2iyuD5VQrx45Prrr/evaWRz/vz5/v2nnXYaADNnziz0fWvVqgXE0YwSRkMzwsSJEwF45JFH/GvFRfsS31Pw8/rvHXfcsbybGYQuKNBZI/07keZutW/f3i+MPOigg5Le8/333/uIqJo5cya//PILsM6IaMZZ3wUkmpOv+dW54ssvvwRg9913L/Q5za1PvHc2adIEiAu3699//vmnz6O96aabADjkkEPWJ+qVkfT80cWfRdHxSZcuXYDcWKD04osv0rVrV6D46+iLL74IROOwuXPnAvGCP+2/8ePH+4WzWnLykUceYbfddit1+ypkIDpv3jy/uEY1a9YMiBbbFEenAO68804A3xm5omPHjmX6+qJW9pnskzg9pAv39CaiK4AB/vOf/wBxncSXX3455QBU6cNNNlYa0IVY/fv395U2/vjjD6D4C+j8+fP9wgOlD3033nhjRTQ1rZxzxQ5ANeVn0KBBAHmxOE3Tv1IttiiO3jy/++67cm9TuumAcfr06XTu3Dnle3bddVe/WEkXeKWi6TD9+vXzg7XPP/8ciNJA9NjKFjoGOeecc4B48Zam7iSqWrWqP6+0CoMuqs5mOh3ftWtXikrDPPXUU1MuYtP6oY0bNwbgiy++AKLUlk8//TTpve3bty/2nrQu2d/TxhhjjDEmK1VIRHTMmDH+4w033BDAL0bShRfr8uSTTwLRvsBaY3H48OEAXHrppVSqVKnc2puJNJH4hRdeSHo916aT8plOmU6aNMkvZituwUWqKeqqVasCUcoLROeZTqG8+eabQDStli00Mly9enW/w1ZJvPnmmz4iqtOPWp6kYHpLNkks0ZQqEgrRTkFapzifnHTSSaX6Oo2MJS6IzTZ6vTj77LOBuE53ovPOOw+IFrxut912RX4vrTesi9/WNXWdDZ566imfwlZcaSbtl8suu8xPXeeCoUOHAvGsUOI9Q8dgTzzxBJA6lSORTrlrubOC3w+iOtiaTrWuWe9ULCJqjDHGGGOCKNeIqO5ckVhwW5PnNYeppPTrunXr5p/UNOpz7LHH+h2Ycokuxho7dqyPjGmBf43yFFzoZbKXLgSYN28ew4YNA0pWOLpBgwb+KfbSSy8F4mLDS5cu9aU1dLedbIqIlpYm2UO8GGVdCwGzgeZd9enTp9DntIC7RsVMyRScZcpGunYiMRKqs486E6DHR3HRUMDvsZ44k6l0gVLt2rXL2OL0ePnllwE45ZRTSlSyTUt51alTp0LblW4anUzcZWybbbYB4NlnnwUotJFOUXTR1jXXXOO/j5Z5mj59OhDNLmiUvjQsImqMMcYYY4Io14io5jPNmDGj3L5n06ZNC702ZMiQpKhrNtI++uCDD/y+4uPGjQOS94lW+nTbqlWrNLXQpMu11167XvluzZs39ysai6PHVT64//77fTT5wAMPDNya8qMrvBNXvGqOl+YAlqbckn6/xO+7js1Nsp6Wp9LZB9WyZcsArSm95cuXc/TRRye91qxZMz9zWNLZR70HaVQw0WGHHQbEVRgyvY+0csKJJ54IFL+BxVFHHcWjjz4KxKXxUtEqJYmVe/bYYw8gs2db/v77b58HnUhnTkoaCVUaDddygI0bN/aR0MRKDbry/qKLLlrvNlfYXvNKdyfIZ3/88Qd77rknEO+mU3Df46Lo9Oqhhx5agS00ITVs2LDc9u795ptv/MfZVg+yLETED0RzYV9srRmrJYoSfyctcVfa/9/+/fsX6qM2bdoUqjeaa7QMmC7OadeuHQAHH3xwsDaVxqhRo/xAQKfj+/Xrt17pb2PHjvVTron7iitNg8v0AajS+2iq0kxK62EOHTrUL/LUBZ26i18iHaAnDkS1P7p27cr5558PxDVaM0WPHj347LPPkl475phjSrzzWEHVqlUD4Pjjj/ev6fGXSBfJloZNzRtjjDHGmCDKNSKqxVMTde/evTx/RFZasmRJqYu9auQiF4rrllTBacNcnzIsDzqN9Oqrr1K3bl0gTufIZTp1nWjhwoVAFBmAaApSp+70fOrbt6+PaGQijcYkTptqMfLEXbfWxxlnnAEkb5agLrnkEh/5yEUTJ04stLPMrbfeCmReRGtdEqNRmp6hO7KtyxVXXAHAww8/nDISClC3bl2aN29exlamly5SSkUXa2r5oeeee84v5nn//ffX6+doUfcvvviCtm3bAtGGAZlA0xW1LBPEC9VGjx5dIT8z8d5clnTJ/BndGGOMMcaYjFKuEdGffvqpPL9dzqhdu7ZP6p0zZw4Q56tstdVW/n1auP/BBx/0W3lqro7SJPJclirXT7dp1CiGiWhO1FFHHQVET6h6TOnWbNnozz//9NEHze3TZPlEb7zxRqHX7rvvvkKvtW7dGogjR1qyJlOlimDoIqXSRi4nTpwIxPmnEPdLNi+CnDdvHgAvvfQSEEX7CuYKTp8+nZUrVwLxdeX5558HotIz2ZILWdCOO+64zvcsXrzYlzd7+OGHgej8KspTTz2VNeWalJaZSkVni7Tk47x581ixYkVa2pVOX375JZB831xXsfrSmD59OldffXWhn1UWFb5YyUQ7u+hOUSVx5plnFqrtpsnUhxxySF5N0yutyGAiWrO3W7duACxYsACILgz6kJONPvnkEyCaOn/77beB1DtKFUcHV4kD0lTVNzKZThkmTn3deeedpfpeurjp+++/L/Q5ndrddNNNS/W9Q1m+fDnXXXcdAPfeey+AH2gWRa+busBH918fNGiQX4WuD3RdunTxC0p1IDN79mwg7s9MoAPL9u3b+8HY//3f/wHxVPQnn3xSooUkunhrn332qYimVqhUe6UrHXQXN/gWEX8OaMpCtqWE6c6TIkKLFi0AfHWA8vTAAw/w448/Jr1Wp04dX7GgNPJvRGOMMcYYYzJChUVEdc/o+vXrl/v3zsVdlRLVqlXLTxvpjjm6l/Rzzz1Hp06dgrXNZAbdUSVxRyGIyotolCsb6T7Xb731lp921AiW/juxlFn//v2BqMSKnhep9t3ONmUtRbVs2TJfz0+vJYnfS3dJSSzJkg100dmZZ57po+eagqLXyjZt2hRaJLvVVlsxZMgQII7+aprPF1984ftI//71119ZtGgREEeVNOITOiK6ww47+I81feWggw7yu++VtoyO7i6kpY3ygV4ra9So4dMRBg4cCBQfQd19993ZcsstK76BpdSzZ0+gbDti/f7770CcytOvXz8g2pmp4HWpatWqJaptXRSLiBpjjDHGmCAqLCKqlf0XL15cqq/XXTBuu+22Qp8L/US6LrrnauXKUfeW5QlT83Y++ugjIMpnsohoftIFGGeccYaPHKpmzZoBUYSwLE+moenvcdFFF/kncJ1dSWXw4MFAvNAv32k+4+WXX56ytBVEm4xkW1m9d999F4gjWN999x2nnXYaEEc29bqredMQ74Dzyiuv+Lw5pQX8Z82a5Qubn3nmmQD06dPHR0779u0LQO/evcv3lyql008/3ecQ69qD4hYK165d2/+uWsLqzTff5IEHHkh6X67PNKaiG8Y450q02Fojgd26dcvo6+xuu+1Wpq//4osvfN60bsJTnPbt25fp51lE1BhjjDHGBFGuEdHEwq66ylnzLQrujbsup556KhCXJAD8XrqZvMJz/vz5fuXlySefDMCFF164Xt/j33//9flMBVfb61O/yX66yvG1114rtrDyf/7zHwA+/vhjIF4hn0hL15TXVqGh6GxHaWY9si3Kt740f7Fjx46FPqfXGJ1JKioaClFUTEtBZQv93b/77jsAmjdvzjbbbAPEUXQtybN48WL/+2mh8+IKtG+33XZ+u0u9dj/00EMZW9qqUqVKfrZAV/Z//fXXfvZRcxsPP/xwIDovtKi7uummm/zH2lfZfP7oZg06I1BSBVd/F2WjjTYC4JxzzgHW/56eDomr/DVPfurUqUBUru7TTz8F4qhu4j1H7x/vvfde0nvW9XO0vKSO80qrXAeiuvOHHhQQ1wAsqUGDBgHxTRegSZMmQLxTSqVKlcrUzor05Zdf+iR6TRqfP39+sSf5Cy+8ACTXS9RpgoKla+6+++6KabhJG619mXhMFFeiSBeqJb5HL4z6wJLtA9D1pTdb3TcccmuRxYEHHggkl1zSclQ6Vab/9998802xx49+Tq+turtONilYQ/arr77yC5eULuK56aab/IChpHR3JV2wo9PxmapRo0ZAHKj4448/fDqclrFKtVBYa8kmLmjab7/9gOSa1tlGHyA0GKYDsPKw++67M27cOCA+PjJR4gLHe+65B4jTA/v37+8fVPR9qXbWKukiyfJ+eLGpeWOMMcYYE0S5RkQ1ItG8eXP/tKqhb93X+ZJLLmH77bcv9LVvvfUWgC9SrE93TZo08XsjZ/KUvNpqq638Lh1aWmPgwIEMGDAAiJ80UkUwUr228cYbA1G/AYWmWHLN6tWrcz79QJ+uUz11lqRcj4iw5557AlGpmnyk59bMmTMDt6RinHvuuUBczmrevHl+ur3gtHviMZP4sU5Zn3LKKUB8DclGGqG8+OKLgei413JFuqBTU6H09XxSq1atEr3vlltuAUjaWahXr14V0qZ0qlevHhCfL6+88gqXXnopEG/+UZwqVaqwxx57JL2mM7Dt2rXL6Eio0iilpudAtGsYRJsxrO/GIEpn3/Se06pVKx8JLa/d+ywiaowxxhhjgpB1bGNVqj2ufv/9dw4++GCAQnk8jRs39sVW1fDhw/nhhx+Awk8v999/f6H3r4fy2Qg1VqL+mDx5MhAXgB03blyxW1TqE4rmOP3888/+KfW4444D4pyxMgrSH+tjzpw5hXKbNtxwQ7/oQI+rcpL2/liyZAn/+9//APjss8/iLyzB02qq9+iCDc2J2nzzzde3zYnKuz+gAo4RiJPqtS832mgjHynUxV3lJOg5o4sIOnToUPw3XXts1KxZE4hy5UaOHAlQ3guTgvSHzpDpNpv16tXzGx0ElvHXVIjzinUGZcaMGT7PVPNGy6kcUcb0h+bWa2k3vadWq1bNL3xWVatW9Quky1na+kO3ex04cKA/53UR3tSpU/124RoN10WvlStX9vcRXfy2wQYb+P7SNTpHHHFEebQ/ZX9UyEAU4n1udT/fggPSouy0004Afjq+fv36ZdlbPSNOigkTJvibgq6UPuyww4BooKm/37HHHgvA9OnTfRi8nGVEfxRnyZIlvr6dHgPXXXedX8FaztLeH1OmTCk0BQRFD0SPPvpoP/jW99xzzz2FVnvOnTsXKHMyfdYNRNu2bQtEe8knVtgoR0HPGV2UNXLkyGJX6uqxobsAVeAK6Iy/hqRZVvSHTsnrKmeIB6C6I1U5yYr+SKOM64/p06cDcRpDzZo1kxaYV7CU/WFT88YYY4wxJogKi4gqTZbVfUuHDBnChAkTgOT6bKeffjoQ74ShZQfKKOOeRgKz/kiW9v749ddf/aKRZ5991r9erVo1AK699log3h2mVq1ahc6Fv/76y5do+frrr4F4OrpGjRplaX/WRkS7devG0KFDK+JH2TmTzPojWcb3x5w5c3wKi5YDa9eunU93KudyiBnfH2lm/ZHMIqLGGGOMMSZzVNhe8/4HrI3maHmFG2+8saJ/pDEZa5tttvG7Xujf6yuxjFk2lBWpCLopgCrrXsfG5KpZs2YlbYwA0Lp164zeGMbkF4uIGmOMMcaYICo8ImqMMeVNN89o0aIFEFecMMYk23fffX1ZHmMyUYUvVgrMEoWTWX8ks/5IljWLldLIjpFk1h/JrD+SWX8ks/5IZouVjDHGGGNM5lhXRNQYY4wxxpgKYRFRY4wxxhgThA1EjTHGGGNMEDYQNcYYY4wxQdhA1BhjjDHGBGEDUWOMMcYYE4QNRI0xxhhjTBA2EDXGGGOMMUHYQNQYY4wxxgRhA1FjjDHGGBNE2geiInK+iEwWkZUiMizdPz9TiUhnEflWRJaJyA8i0ip0m0IQkaUF/qwWkXtDtyskO2eSicguIvKOiPwlIjNEpEPoNoVk50xqdk2Nich7IrIi4RiZFrpNIVl/JBORhiLyqogsEpHfROQ+Eamcrp8fIiL6K9AfGBrgZ2ckETkEuBnoDtQEDgJ+DNqoQJxzNfQPsBWwHHg2cLNCs3NmrbUXxxeBsUAt4GxgpIjsFLRhAdk5U5hdU1M6P+FY2Tl0YzKA9UfsAWAesDXQEmgN9EzXD0/7QNQ5N9o5NwZYmO6fncFuAG50zn3knFvjnJvjnJsTulEZoCPRyTEhdENCsnMmSRNgG+BO59xq59w7wAdAl7DNyhh2zkTsmmpMyTUCnnHOrXDO/QaMA5ql64dbjmhgIlIJ2AvYcu004+y1YfGNQ7ctA3QDHnfOudANMRlNgOahG5Eh8v6csWtqkQaJyAIR+UBE2oRuTAaw/ojdBXQWkWoiUg84nGgwmhY2EA2vLlAFOB5oRRQW3x3oG7JRoYlIA6LpgeGh22IyyjSiiN9lIlJFRA4lOk6qhW1WeHbOeHZNLewKYHugHvAQ8LKI7BC2SUFZfyQbTxQBXQzMBiYDY9L1w20gGt7ytX/f65yb65xbANwBHBGwTZmgCzDROfdT6IaYzOGc+xc4FjgS+A3oDTxDdPHMd3bOROyaWoBz7mPn3BLn3Ern3HCidBbrD+sPRGQDoujnaKA6sAWwOVGOdVrYQDQw59wiopto4lRa3k6rJeiKRXZMCs65qc651s652s65w4giG5+EblcGsHMGu6aWkCNKaTGRfO6PWkB94L61A/OFwGOkcWAeonxTZRGpClQCKolI1XSWCchQjwG9RKSOiGwOXEy0Kjgvicj+RFMmeb3yV9k5k0xEdl3bB9VE5FKilZ7DAjcrKDtnCrFr6loispmIHKbXDRE5haiKQNpyADOJ9UeytTMGPwHnru2PzYhyzaemqw0hIqJ9iaZOrgROXftxPufuAPQDJgHTgW+Bz4EBQVsUVjdgtHNuSeiGZAg7Z5J1AeYS5Yq2BQ5xzq0M26Tg7JxJZtfUWBWi8m/zgQVAL+BY59z0oK0Kx/qjsOOAdkR9MgP4l+jhLS0kjxdXGmOMMcaYgCxH1BhjjDHGBGEDUWOMMcYYE4QNRI0xxhhjTBA2EDXGGGOMMUGsqwRMtq9kKu+6YNYfyaw/kll/FGZ9ksz6I5n1RzLrj2TWH8lysj8sImqMMcYYY4KwgagxxhhjjAnCBqIZ6JtvvqFWrVrUqlWLnj170rNnT5xzWM1XY4wxxuQSG4gaY4wxxpgg1rWzUraH4LIqUXj58uUAnHfeeTz22GNJn/vnn38AqFKlSll+RFb1RxpYfySzxUqF2TGSLKv64+OPPwZgxIgRjB8/HoAVK1YAcOihh/q/DzvsMAA22mij9f0RWdUfaWD9kcz6I5ktVjLGGGOMMZkj7RHRpUuXMnDgQAC6dOkCwC677FLeP0Zl1dPIxIkTAWjVqpV/bauttgJg1qxZAFSuvK6KW8XKqv5Ig6zsj2effZYTTzwRgGeeeQaA448/vjy+tUVEC8vKY6QCZUV/TJ48GYCjjjoKgPnz5/sce5HCv8Jpp50GwKOPPrq+Pyor+iONrD+SWX8ks4ioMcYYY4zJHGUKr5XGp59+yu233w7gI6P5bunSpQDcc889hT530kknAWWOhJoc0q9fv5RRHWNMlAN63HHHAVEkFGDvvffm5JNPBqBz584APg//ueeeY9iwYUCcI/rAAw+ks8nG5LUgoxtdeDN8+HAAunXrFqIZGWPcuHFANOWqGjVqBMC5554bpE0V6eGHH/a/s/5+Bx98cLFfM3v2bADefvttID+PmVGjRgHw/fffB25Jet13330A9OrVC4AmTZpQu3ZtIE5nyScffvghAAcccAAAe+65Jy+++CIA22yzTbB2hbZs2TIgmmafM2cOAJttthkAAwYM4H//+1/S+y+//HIAunfvTvv27QF47bXXAPjzzz/915rIt99+m/TvCkypKxdLliwB4oDX9ttvD8CXX37p3/PGG28AULVqVaZMmVLk9+rRowcAd999N1CqRW0Z69NPP/UfDxgwAIAxY8b4VBb9f95yyy39vy+88MKkz5WVTc0bY4wxxpgggs73rlq1KuSPzwjLli3jtttuK/T6U089BUDjxo3T3aQK8+qrrwJwySWX+HSEd955B4Add9wRgA4dOlCvXj0gjoQBLF68GIBff/0VgEMOOQTIrwjQtGnTgHhGIdcNGjQIgL59+wLxIr5///2Xr7/+GoBzzjkHgOuvv94v7MsXmp7x2WefsdNOOwFw5JFH+s/vscceABx00EEA/j0aTc41mr6kCzsBdtttN4BC0dBEW265Ja+88goAv/32G1DmMnkZS6OaXbt2ZdKkSet8/+jRo4HoXPzuu++SPnfVVVcBcPXVV5dzK8vHm2++CcDNN99covcXl+70+eefA/D7778DUL9+/TK2LhxNV9Hr61133eV/91QL+vS+o///EydO9FFU/b/v0KFDmdpkEVFjjDHGGBNE2iOiGsmAOEf0jDPOSHczglu9ejUQRTC06LISETbZZJMQzapQmqNTp04dHxH9888/gbjciv69LnfeeScAt956a3k3M2PdeOONQPFP7rlEZwU0Qj5ixAgAGjRowBFHHAHAkCFDAGjWrJnPIc1Hf//9NxAtvFH6sUY5yrnUV8YZOnQoAJ988ol/TaPB61KrVq2kv3ONRsF0Ede0adP8tVavJ1rwf8yYMf7jxEhZwaiZzlRkakS0IG3/nnvu6aN8Z599NhCdEz/99BOAzy/u2LGj/9o6deoAUK1atbS1t7zpMaC/S8H/z8SPmzRpQvXq1VN+n++++84fO3369AGiWZaSnmuppH0gWrNmTf+xJr/mo7/++guA999/37+24YYbAtEgq0mTJkHaVZH0d+rYsWOhAaQe9I0aNaJGjRoAfPTRR+ltYIZbR83fnDJu3DimTp0KwJNPPglEA1C16667AvHiknySuLhgffTr1w+IFjltvfXW5dmkjKCLUkTELzQ677zzQjYpY3Tt2hWIp1lFhL333tt/DMnTsvpa4kOvfqwLd7KNpmkkPqgk2meffdLZnLTTqfiC/7cdOnTwA0rVpEmTIgfdAwcO9A8hejw98sgj/vsl1kEvKZuaN8YYY4wxQaQ9IqoLVgA/vZaPEhfiKH1C7dmzZ7qbk1bXX3+93+9ZS1Zp8vfTTz9N1apVgXgaScvVQBw5zfU+SiXxSXaLLbYAYNtttw3ZpAqTWMpM9wFPdNNNNwHxQoR33303b6bmddpUI1ht2rTxi/50Md/TTz/t33/JJZcAcdmamTNn5lREVCNcOqUqIj5ql8+zbuqcc87h9ddfB1JPw6b6t55zughFp7CzyZgxY1K+/v777/vz5OWXXwaS01X0Ppxr11ZNO9H/Z51Kf/755wu999tvv2XmzJlA3I+aBiUihY6dESNGMHLkyKTvr/fvVN+/IIuIGmOMMcaYIIKWb9Li5Pm0WEmfxBIjPhtvvDEAZ555ZpA2pVu1atX8LlKaw6VR0AYNGvDVV18BcSmVRFqiRQv+5wMtopxIS/Hsu+++6W5OhdLSVFOnTuXwww8Hil9Ast9++wHxeZUPCuZ4zZ07139Oy5ldfPHF/rXevXsnvT/XFNyhr06dOkkLTfJV//79AXjhhRf8/71GwVIVIj/rrLP8x1r6K1utWbOGX375Jem1b775BoB27dqxcuXKpM/pD8QFrAAAIABJREFUwkiIFj5CPNuSKzO3+nvpsaDlmBI3KtBzacyYMX6DiILXm1R5wwU/BmjatGmJ22YRUWOMMcYYE0TaI6I6KgeYMWNGun98cFogWKN+EJfByMdtK3feeedCr/34448AvpwGQN26dYF4m8t8sXDhQh588MFCr3fp0iVAayqebss3efJkv21lcTbddFMgLgOWj6ZPn57y9YLnSosWLYDka3Au+OGHH5L+fd5557HnnnsGak04Wp5HZ1A0uuWco3Xr1gC89957QdqWbu+//77PpVYl3QhES0xqbnXbtm1zYktPzfc99thjgTj3s2nTpikrJxTMA008pzR/VL+uX79+PidU1y+sj7QPREuztD+XJNZRVVon0UTuuOOOQq9puZ7//ve/6W5OUKNGjSo00GjZsiVHH310oBZVrJdeesl/3LBhwyLfp4nx48aNA6Kp6NNOOw2Agw8+GIBTTz21YhoZmE6bap3QogZdiQtDAS666CIguYReNtPzQtMy9MaZj/eY+fPn+ylkLe+VOFVa1p1vso3WzE2lbdu2Pu1HFyZBvAhQA0N6fGnN72ynQTCdki9uWl1EePzxx4F4ij0xXUN3tNPB56GHHlqmttnUvDHGGGOMCSLoYqV8ok8hiUnRECXWd+rUqciv04Vcum/uSy+9RMuWLSuoleG9+eabKQsO59sTvZowYYKP9OjfjRs3zqnyO4kOOOAA/7H+n2v0QhdoAYXSFXSnLcAXM8/ViOgVV1wBxJHfVBHRL774wkeX9bjRjSJyhUZCNS2juMVYM2bM8ItjtbTVgQceCMQ7lmWzgw46yBcXT1WWSRev6Xnz6aefZvUuQeuy8cYb+99PF2jpTo6bbbYZVapUKfQ1en3RiKh65JFHuOCCCyqyuRVu5MiRftHivHnzgOJ3VurQoYMv4ZWqBFqqdLGysIioMcYYY4wJwiKiafLZZ58BhRdVtGnTxpdvUqtWrfIFmXX/ZNWhQ4ekRTy5QovWP/zwwyxfvrzQ53X7U42CaJmafFAw0nPttdcGaknF07zQTp06+Tyv+++/v8j3awS1UaNGvmC7FnfPdcUtyBk/fjxLly4FcrdsU0m89dZbQLStsC6E0/6YMGECEC0c1fy5bKOld6ZNm5Yyz6/gxxo1HT16dM7OGEC0zkDvoakWxK6PVPejbKHHde/evVmwYAEQ7zWv+Z1nnXUWAwYMAKJSXxAtZNItlVOt2ShvQQeiOl09bdq0Mh8s2SpVku9TTz1VaACq1qxZU9FNSquJEycC8TSsniwFaY1VnU7TOqxbbLGFry2aL3SleC7SB46nn37aT6Hq/3Vi7T8dsOrxANGOXRDvujRlyhQg3mM6HyxcuBCAwYMH+9e0j/JpoZ8O0DS1aenSpX5XNl3op8dXLlRcOOWUU/z9VBdrXX311UC0kEkH3To4GzhwYE4PRKHsA9BspjVANc1g3rx5/mFExxyJ1wjd/Shx4KrpTnre9OvXr8Laa1PzxhhjjDEmiKChJJ0q0b9z2eabbw7EEZ9Vq1YBUYRTp480Cqr/TmXx4sV+h4j12bkgU+lUe1GR0II0gqqLdW6//XZf7y2X6IKtjz/+2L921FFHAcXvNJSLTjjhhPV6v0ZOdbFOPkVEtXZoYskvnW3QaEfiDjq5aMCAAb7u4+zZs/3rWvJLF63kAt0hacSIEUW+Z4sttvDX18Qpeo0ap9plyWQ3nWLXVAwR8dHRbbfdFohTUxLLnel0/cSJE7nrrruSvpdFRI0xxhhjTM5Je0S0du3avqByPkRCVcECuhrZO/vss9fr+7Ro0SInIqFKo3wffPABkFy+pyRefPHFnIyIavHxWbNm+de0/EbBxW0mtaJ2HMpFOsPy+uuvA1EZFr3OJu47n0vatGkDxHmPmvOWakbpqaee8tfe888/H4hL1Wh5n1ymu93o3wsWLPA7MVlENKKzc7ng4YcfBuJj/JRTTvElqEq685F+bTrWpVhE1BhjjDHGBJH2iGjz5s39lpa6yi+fnHjiiUAcES0pjYLptn65QosOa97KRhttlLQ6GqK8lZNOOinl15dmX9tMplG8uXPnAsnFhnW/aJNaLuwHXVqaB6rbeopITpf5StSnTx8gLrKdqlzVxhtv7N+nka+qVasC0K5du3Q0s1xpvq/m9K2LRj21b0Qkp2bWyoOWMCqoc+fOaW5J+dH/77PPPnu975X6tRtsUPHxyiCLlfTE14Ho7Nmz2WuvvUI0Je26dOkCwBNPPAHARx99VOz7W7RoAcDll18OxDXAck39+vUBuOSSSxg0aFDS55o2bcrxxx8follppyWHfvnlFyC6GOTyTlrlSY8RLVuTL6ZPn84rr7wCxDePtm3b0qtXr5DNShsNbBTnpJNO4u+//wbiPtL0l3333bfiGldBdK9v/Z3WVYqpf//+QLyrzjnnnJNzD/FlMWHCBMaOHZv0WteuXYH43pRNateuDcSBDE3DWBct2fTEE0/43zuxzFNFsal5Y4wxxhgTRJCI6LHHHgvgywPcdNNNtG3bFoinaitVqhSiaRVOi5Hr7i8LFy70C5m++uorINpLXKfVdOoll/cFNsXTKLopXuPGjYF4QeCcOXNCNidtUpVVOfjgg32puHyhkdFUi04Sd8fRWaWOHTump2EVQKOZjzzyCBDtslXUoqP+/ftz8803A/HvnuslvCCOAmppK52JTUxJePfdd4HoWNCUML1Hn3zyyUB27kyWmIKxPnSR04IFC/yYI9Ve8+XNIqLGGGOMMSaIIBHR/fffH4C6desCUdFufZr79NNPkz6Xq3Tx0bbbbuv3yDapjRkzxudzae6LMUXRKIBGOxYsWJCT+XC6AcYTTzzhc8GuvPJKIM4pzye6CcSwYcP8whONhLZo0YInn3wSgE022QSIF0hmI70e6qLXpk2b+nuoHgtazHznnXf2Wz7qdq977LFHWtsbguY23nrrrQBJZf60hJeu1fjrr7/853SNQqrtt7PFKaecAsAbb7wBwN133+33jt9zzz2BOGL8+uuv+3xYPXZEhNtvvx2AJk2aVHh7gwxEq1SpAsQrPNu1a+cvDrk+ADXF23///f1q1hUrVgBRyoJeSPNtINqtW7e8qHNYnrQ2re5K9eyzz3LuueeGbFK50vrLd999NxDdNHSg3bNnz2DtCk13W7vqqqu46qqrAremYmn9aR18jhw50g88EwcTEA1INQ0hHQtPMsV7770HwD///APEA9HRo0f7xaCqRo0aPsVlfWt7ZyLdLUmvCxMmTODII48EYLvttgPi3QxnzpxZaAq/Y8eOJa7IUB5sat4YY4wxxgQhiXUKUyj2k1mgvLOMrT+SVUh/7LPPPkA81QbRUxuUeymNrOiPNKqIrPy094nuLqSLEw4++GD/Wilq4mXcMaLTsvfee69/bdKkSUBaplwzrj8CC9oful/8wIEDi6zLfcwxx/h9xtMgY46P7t27AzB8+PBCn9NZWa1PfdFFF1VUmbyM6A8R8VHPghFz55xfxKZT+ldffXVFpTOl7A+LiBpjjDHGmCCC5IgaUxzdAeWYY44J3BKTjf773/8C0KlTJwCeeeYZXnjhBSC7S/YozYVUgwcPzovFJ6YwzRHVEkUmpjMHujB4xowZQNRnOuumJZpy3bhx4/wirPHjxwNxRLRHjx6+nFeo64hFRI0xxhhjTBCWI7p+rD+SWX8ks/4ozPokmfVHMuuPZNYfyaw/kuVkf1hE1BhjjDHGBGEDUWOMMcYYE8S6puaNMcYYY4ypEBYRNcYYY4wxQdhA1BhjjDHGBGEDUWOMMcYYE4QNRI0xxhhjTBA2EDXGGGOMMUHYQNQYY4wxxgRhA1FjjDHGGBOEDUSNMcYYY0wQQQaiItJQRF4VkUUi8puI3CcilUO0JROIyEgRmSsii0VkuoicGbpNmUBEGovIChEZGbotmcD6I2LXj2QisrTAn9Uicm/odoVk19TCRKSziHwrIstE5AcRaRW6TSHZ9TRZyOMjVET0AWAesDXQEmgN9AzUlkwwCGjonNsEaA/0F5E9A7cpE9wPTArdiAxi/RGx60cC51wN/QNsBSwHng3crNDsmppARA4Bbga6AzWBg4AfgzYqPLuerhX6+Ag1EG0EPOOcW+Gc+w0YBzQL1JbgnHNfO+dW6j/X/tkhYJOCE5HOwJ/A26HbkgmsP5LY9aNoHYkG6RNCNyQku6YWcgNwo3PuI+fcGufcHOfcnNCNCsWup4UEPT5CDUTvAjqLSDURqQccTnQzyVsi8oCI/A18B8wFXg3cpGBEZBPgRuCS0G3JBNYfhdj1o2jdgMedcy50Q0Kza2pERCoBewFbisgMEZm9Np1l49BtC8Gup8ky4fgINRAdTxTBWAzMBiYDYwK1JSM453oShcRbAaOBlcV/RU7rBzzqnJsduiEZwvojmV0/UhCRBkRpCsNDtyUT2DXVqwtUAY4n6ouWwO5A35CNCsiup8mCHx9pH4iKyAZE0YvRQHVgC2BzovyEvOacW+2cmwhsC5wbuj0hiEhL4GDgztBtyQTWH8ns+lGsLsBE59xPoRuSKeyaCkQ5wwD3OufmOucWAHcARwRsUxB2PU0p+PERYqVpLaA+cN/aHJ6VIvIY0B+4PEB7MlFl8jefqQ3QEPhFRABqAP/f3r1HWT3vfxx/foszg6RoKhHlWgiRtaoVxiXlEkJWLVTuHJxyr0XqnC7KXS7lJyW3EkIr91shlwoVp5hmnVxSKIyaUpL9+2N7f/beM3v27JnZe3+/+zuvx1rWaN/msz/z3d/9+b4/n8/73dDzvAMikchhPrbLL8WoP+Lp/FG1/sBYvxsRUPX2nBqJRH71PG8l0XWy7ma/2uOzYnQ+TRCE4yPnEdG/R9srgMs9z9vG87wmRNc1Lcl1W4LA87zmf6dNaOR5XkPP83oA/ai/i6j/j+gXxqF//zcReAno4WejfKT+iKPzR3Ke53UFdkO75XVOTW4KcNXffdMUuBqY7XOb/KDzaXK+Hh9+rRE9A+gJrAFKgS1E33h9FCE6ZbQS+BW4AxgciURm+doqn0QikY2RSOQH+w8oBzZFIpE1frfND+qPpHT+qGwAMDMSiaz3uyEBoHNqZSOJpioqAZYBnwGjfW2RD3Q+rZKvx4enzZUiIiIi4geV+BQRERERX2ggKiIiIiK+0EBURERERHyhgaiIiIiI+KK6PKL5vpPJy/DrqT8SqT8SqT8qU58kUn8kUn8kUn8kUn8kCmV/KCIqIiIiIr7QQFREREREfKGBqIiIiIj4QgNREREREfGFBqIiIiICwNdff02fPn3o06cPLVu2pGXLlixevNjvZonPli9fzvLly7n77rtp1aoVrVq1om3btrRt25Z+/frV6bU1EBURERERX1SXvkmybOrUqTz33HMAzJ49G4BIJILnJc/6MGzYMC666CIAmjdvDkBBQUEOWlp79l7sZ0FBAR999BEAhxxyiG/tEhGRqC+++AKAnj17smrVKiD6XQQwffp0navrmdLSUgAeeughAB5//HEAfvzxx0qP3bRpE2vWrAGgqKioxr9LEVERERER8YVnVzxVyEry1GOOOQaAOXPmuNuGDx8OwIgRIzL5qwKbTHbq1KkA3HLLLaxcuTLxl6SIiMbf9+yzzwLQu3fvdH+tL/3RoEH0eqdhw4butpNPPhmAF154IcNNqpHAHh8+ybuE9jNmzADg1ltvZdGiRWk/b+DAgUyZMiWdhwbiGCkqKqJ///4A3HnnnRltUA0Foj8ANm/eDMBnn30GwPvvvw/AvHnz3IzLDz/8UOl5du654447AGjXrl1tmwAB6o/aevnllwHcTFt8n9n44IEHHuCf//xnOi+X9/2RYXnVH1u3bgXgkUce4brrrgNg/fr1ADRr1gyALl26cPjhh0cb8/fx8eijjzJ37lwA9txzz1S/Iml/+DIQrWqQFe+dd94BoLi4uE6/qi5PTiJj/bFgwQIAOnfu7G7bb7/9gOTT1SUlJQAsWrTI9Z89bu7cuey4447p/Fpf+uPjjz8GEkP8PXv2BGDmzJkAbLvttilf448//gBg8ODBQOxL5B//+AfbbFPrFSaBPT5+++03ANq2bcuhhx4KwNtvv53Wc22KrW3btgDssMMO6f7aQA9Ely5dCkSPo2nTpgGwbt06IHZ8pMvzPDewq2ZAGohjpHnz5qxduxbADbgPPvjgzLUqfYHoD4AhQ4YAMG7cuFo9384bCxYscJ+xWghMf9TUpEmTABg6dCgAP//8MwAtWrTgP//5DxA9vwKce+65CYGEFALXH5s2bQKgV69eQDT4c+SRR9b1ZdMVuP5Ixgag5513HgDTpk1zf/tjjz0WgLvvvhtIfuE2YcIEzj33XIDqxiKqrCQiIiIiwRGIiGhxcXHCNH28d955py5R0cBejZSVlQEwevRo9tlnHwD69u0LwE477VTp8RYeP/bYY/n0008B6NOnDxBdSJ4mX/tj2LBhAIwdO9bdZpHNQYMGpXzu+eefD8ATTzyRcPv06dM588wza9KMeIE9Puzqctq0aS5SPm/ePAB23nnnKp9XWlrqHm+zCkcffXS6vzbQEdE99tgDoNJSlrr666+/Ut0diGMkPiLao0cPAJ555hkAGjVqlKGmpSUQ/QHw6quvAnDiiScCuKhm586dXbTzwgsvdI+/9tprgcozC9OmTXPn3loITH/UxOzZs12E0L6PbUr1rbfeYq+99qrtSweuP0aNGgXEvn8mT57svk9SsSUKt99+u1symObMY7zA9UdFa9eudf1hG6YhtuzJxhkZooioiIiIiARHINI3HX300S56Y5FR29B0zDHHuIiobWiq47rRQGjSpAkQvdpKh12JdezYkU8++QSAr776CohGS2txpZY3FixY4N5z2D388MMALqUXwN577w2kjoTaGklb2wXRBeRQo4hooNi6JYtofP/992k9b/fddwdiEY0///wzC63zj0UCLcJja7fqm+OPPx6AFStWALHPR+PGjWv0OrvuumtmGxZg7777LhBbLwmx9fmWnqcO0dBAsTWv999/PwBt2rQBqDYaaucNO75KS0vdLMQJJ5yQjab6qkePHm6W1daFzps3j06dOuWsDYEYiMazQaYtGRgxYgT//ve/gdggtZrlBKHy+eefA7gdaZMmTXJTKbvttptv7coFG4gsWLCAZcuWJdxn0wV2sgiLSy+9FIhNl7Vv356JEydW+zybRnnyySfdbU2bNs1CC3Pn999/B2D8+PFA4ue+devWAG4n77777uvus/y6CxcuBKLLX+xLKZ7tAs1XEyZMAKBDhw5ccMEFaT+vpKTEPf6cc84B4PLLL898A7PMpt9tgJHKxx9/7DaIGhtw7b///hlvW9AsX74ciGWl8TzPDUDtwqZbt26+tC1bXnvtNSCW9/KWW25J+Xi70LWsCv/973+B6JKgMA5Ar7/+eiC6+dHOmbb8y5YL5oqm5kVERETEF4GLiFY0YsSITOcWDbyNGze6SI/l2bTNSvFuvvlmoFYLqPOCVfdItpHJpuGSbezKR5aOqKKZM2e6qeZkLGoYPz1rkdB8jHLFs7RTlt8wfhnLrFmzgMRUZxYxtcjH6NGjq3xtz/Pc5ycfRCIRtwltwIABAIwcORKIbsix84NNOzZu3JgtW7YAsfRpFiEeN26cm360tHD5fqxUZEtVbEZp0KBBlc6hlraoZcuWuW2cD+677z4gMXf3W2+9BYQvEmpsaVK6bHZg8eLFQGzJQqrzSD5asmQJENsoDLFza64joUYRURERERHxRSAiorb+sToWGQ1DhNSuzi2pMMQ2qpSVlSWt52psbdhhhx2WxRb656mnngJia3zCrqSkhCuvvBKIRfVsXaRFwapiybytukzTpk1d1CN+3WQ+ss9Iuhv6LAqWTgSjqKiIf/3rX7VvXI55nucS2FskzzaxDRgwgKuvvhqAe+65B4DCwkKXlsrWBybToUOHrLXZT1dccQWQeH411n8WaQ+7UaNGue+WgoICIHp+CWsk1Pz0009pP3bcuHF88MEHCbd17doViKXSy3dWJKViusObbrqJs88+248mOYqIioiIiIgvAhERnTNnTqUop0VJkyW6Ly4uzrsUTlYH2Upa2vuzSFa8+Hry22+/PQDdu3cHoutCrc5rGFhy+44dOwLw7bffujWhVr4xnu2QrW4HZD5ZtWqVK3Bgf3fb5W1JyyG2i7NLly4usmG75e15w4YNC02Uy9Y+WynH+EIIFtWyOtkQ+4ylo7y83H0G8yW91UEHHZTwb4tiNGjQgJtuugmIRT/jzyHJtG/fHoitHQyb119/vdJtRUVFAG72IewsBdzYsWPZvHkzgKsffvHFF/vWLr/YDMgvv/ziUjrZ9/Hnn39eqbBF2DKylJaWJvy0rDtDhgyhsLDQt3aBDwPRqiooWYqmiix3KMRSO+XbIBTg6aefBuDBBx8EYlOwVX1ZWC48q4Pdu3fvbDcx66zySbNmzVyVGPt53HHHpfUatoDc0k3ksw0bNgBwww03VLovPh9oRQ0bNnR1n2062k6aYfqStc+GXaTYe966davLO2znjYULF6a9xAdgzJgxeTMABfjyyy+r3JR41llnuWlES981efJkvv76awA3CDFNmjRh6tSpABxwwAFZarG/bPOeVVYqKytjzZo1AC4dWqrPWBjYZp0NGzbQs2dPIPzvORWrwBWfxsvyZjZu3NgFA4xtlgyDDRs28Oabbybc9thjjwE5r8yWlKbmRURERMQXOas1X7FiUlUs0pGhqGdg6rxaxMJSqVi/d+3alY0bNwKxjSkzZsxg4MCBQDSykUGB6I9u3bq5fqgp24CzdOnSWj2/Al/7wxbTx1d2SRUpt6mUSCTiUltZ5NzS1KRK9ZSGQNeat+kjiwLXhC1nsc0rrVu3TlmpKk4gPjM19eOPP7pIsqVqMmPHjk0ahU9TXvVHeXk5EF2CYJvYLK3Vhx9+CNR502fg+sOWudlsQaNGjXjvvfeAxPODFQSwKn8ZEpj+sI2cVoHM/u7bbLON+z62ZT9btmzhtNNOA2JpnCxi2KBBneJ1geiPsrIyt5zJzqPVVWSzKfyHHnoIiG10GzJkSF2iqKo1LyIiIiLBkbM1ovFrQ+PXfULi+tB8XP+ZDlscbnWvTzrpJABuvPFGd5utARsxYoRbRN25c2cAnn/+eSAcdZGnT5/uSqbZGtFff/0ViF6923u0Ep//+9//fGhl9m233XZANImwXX0aSyx82mmnuei4RfD69u3rIqJ2Xx0joYH03XffAbFjpKalfS3Z/SGHHOLWzsYnwA8jS/t21FFHsXr16oT77BxUh2ho3rHIzdChQ10pS6u3bpvbwpYGzxLVmz///NMVLLBSydtuu63bnGMJzsNWMvrGG28EYrMhVvyjoKCgUslOO48C3HnnnUCdI6GBUlZW5tYMW9GLZOw7d9iwYW5DV8VCEG3atMl46rOcb1YaPnx4qPKBpsv++KkOAtOuXTv3IZg/fz4QC4+Hoc923313N7VuOzttSUbHjh3dBgMbiFhN6LCxC4+vvvoqrcdbla25c+e6L88xY8Zkp3E+sWn3ZcuW0adPH4BKg/Sq2BTj6aefDsQ+M7bBLczsYnbUqFFANDetLe+wvIFVbQitL+wzYwNRyxiQT/lkU7GMCRU33WzevNkNQC0X7fz5890mNquxHraBqEm1+90G4c8//zytW7cGYlPQYVJeXs6mTZsAOPXUUyvdbxewVrUtVQ7vL774IuPtC8+QX0RERETyStYiojYVb1fhNh0ftql3u6q0q9GKuf5qyzYu7b///gCcfPLJGXndoLFoTcVqD1KZLbr3PM/lA7Tp/Xy2detWF5mwDQaWHzVdTZo0cVH1sE+/V7R69Wo37R6fV9XOIddffz0QjmU9dfHll18m/NuiQOXl5YFIYVNXNqVs+YZNYWGhqyu+cuVKAE455RQ3dW2Vgyy9zx577JGT9gaBbdhZt24dZ5xxBpDxzVuB8Mknn7j/X7FiRcJ9a9eudamt4vOa24yUzVK9+OKLWWufIqIiIiIi4ousR0TtpyWPDltE1GqC25Xmvffem5HXrVgn3CJERxxxREZeX/KHbaooKSlxt7Vq1cqv5mTcwIEDXSL22mrUqFG9i4Sa+fPnJ0RCIVo8wm6r75FQiFbQqbiJx9L5lJWVhSIiahsWK27qGzVqVNLNJRYRtdm8RYsWAfUrImoV7ACuuOIKH1uSXZauCuCDDz4AcBvYevTo4SKhVrnwtttucwn9LVJs/7bnZZIioiIiIiLiC19rzcfv4rT1XfnGog1W+3zLli2ujGdtrV+/nkceeQSIRcHi13jUZ5bmyaIb6ZYGzWdWqtF2Rrdu3ToUpRn79esH1Hw9aOfOnfnoo48Sbvv+++9d6cbLLrssMw0MONsFa+873quvvkqLFi1y3aTAsGjnoEGDAHj44Yfd58fYuvSwpD6z2TnLlmB//zPPPNPtZbCZO8tMArHE9lbusj6w1FXxx0SYSnqmsnDhQiCaRhGikfD27dsDMHv2bCB6TrYMPXbsWIYW27eSSTkbiManHUpWXSlfp+wtBY/9sSZOnOhq2d58880AdOnSpVJtdEtNtHr1andAWHh8zpw57mSy/fbbA3DNNddk823kDcspaZszHnvsMTdNPXjwYCB6UXDrrbcCsS8kM2vWrFw1NSN+//33ShUwxo8fT1FRkU8tyhw77pNVkUrFphLjRSKRSn/rsFq3bh0QrTEP8MYbb7j77Lb6OAi1QcX06dO56667gMTNF8a+f26//fbcNc4HVlHrhBNOcDmIk1W069WrF4CrR18f2Ma1V155BYimtWrZsqWfTcqqnXfemXbt2gGx937JJZcA0UG55Su34ICljQTo378/AFdddVXW2qepeRERERHxRc4josXFxQlVliB/p+UBunfvDsQWA3/44Yd8+umnAPTu3RuAZs2aVYoCP/NroXp8AAADlUlEQVTMM0DyaJDnee52u2o55ZRTstD6YLPEwskqD1lEuXv37q6vLCL25ptvuuTV+V4dY9y4cS6q07RpUwBXE7m+it9gYFq0aOHSjYSd1ce2dDue57H33nsDsfrYYWUzIk888YS7zQpC2HIN23QTr6CgwKU8s+8im5IOi5122gmI9ZFJNoMAsQ3EkyZNym7D8kCbNm3c+TWMdtllFzcb2KFDByCxYtKUKVMqPcfSv40cORLIbmGQ/P6WFhEREZG8lbNLQtuYFL9BKQxJ7m0N5+TJk4FoEtiKJbB+/vlnnn322bRf8+yzz3brHutzWT5bV/vUU0+58pa2IL+8vByIrguz9XDxC/At+Xu+Rj0s4jtjxgwX8bX3FBZW7tZqINdGYWEhAO3btw/1Gi+IRUAnTJgAJM6mXHrppUA4ChwkYxv2bJYpWdQznkW3OnXqBETLeWZjk0WQ2AbXit+nhYWFnHPOOQA0btwYiG5gik/pI+FnKSEff/xxIDrOqMgS23fq1Mlt+sxFqkCvYs6xClLemYpNv6famJSDKfma7YKoXrX98dNPP7ld8y+99BKAm6oHOPzww6Mv9He/t2rVipNOOgmIVU/K4i7OnPdHplkONKtRP378eF5//XUg+bFWjcD2hx0Tr732mtvNunjxYiA2ZZIFme4PSNEndsHWq1cv97ezC46ysjK3ETDZ4Mp2Qx944IFA8vrJGRKYY8Qudu2Cy6bKxowZ4zYz5mApii/9kU5deKuIc8EFFzB06FAguiwqywJzfAREXvTH0qVLgcTzR5YqB+VFf+RQ0v7Q1LyIiIiI+CJrEVFjkQ6LkA4fPjwhlVOW6WokkfojUeD6Y8OGDQB069YNgCVLlrhptRxsRMlpRDSZVatWAdHlGLYR0OeKSYE5Rqw2ui1BsLx+Fv3LEV/6w5aq2Iajb775xk0Z2uY9ywua4ypJgTk+AiIv+kMRUd8oIioiIiIiwZH1iKjPdDWSSP2RKHD9YesmLQq43Xbb8d577wHQsWPHur58dXyPiAZQ4I4Rn6k/Eqk/EuVFf9jsQt++fYFoIZosbWbLi/7IIUVERURERCQ4FBGtGfVHIvVHIvVHZeqTROqPROqPROqPROqPRKHsD0VERURERMQXGoiKiIiIiC+qm5oXEREREckKRURFRERExBcaiIqIiIiILzQQFRERERFfaCAqIiIiIr7QQFREREREfKGBqIiIiIj44v8BpNYsCO1bctsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x345.6 with 40 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_rows = 4\n",
    "n_cols = 10\n",
    "plt.figure(figsize=(n_cols * 1.2, n_rows * 1.2))\n",
    "for row in range(n_rows):\n",
    "  for col in range(n_cols):\n",
    "    index = n_cols * row + col\n",
    "    plt.subplot(n_rows, n_cols, index + 1)\n",
    "    plt.imshow(X_train[index], cmap=\"binary\", interpolation=\"nearest\")\n",
    "    plt.axis(\"off\")\n",
    "    plt.title(y_train[index], fontsize=12)\n",
    "plt.subplots_adjust(wspace=0.2, hspace=0.5)    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pcihrWrhrVou"
   },
   "source": [
    "**Visualising one digit**\n",
    "\n",
    "\n",
    "---\n",
    "\n",
    "This will plot an image of the first digit in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 267
    },
    "colab_type": "code",
    "id": "cldO4opXk8Ne",
    "outputId": "8710fae6-5fb4-47fa-e805-d9ef251647e5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7\n"
     ]
    }
   ],
   "source": [
    "plt.imshow(X_train[0], cmap=\"binary\")\n",
    "plt.axis(\"off\")\n",
    "plt.show()\n",
    "print(y_train[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "sdHL16RQspHe"
   },
   "source": [
    "## **Building Model**\n",
    "\n",
    "\n",
    "---\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "9KLaSpAAqA9q"
   },
   "source": [
    "**Model Using Keras**\n",
    "\n",
    "There are two different ways of defining the Model in Keras:\n",
    "\n",
    "\n",
    "*   Sequential Model\n",
    "*   Function API\n",
    "\n",
    "Functional API is used to build a more complicated Model such as for multi-output Models, directed acyclic graphs, or models with shared layers. I am using the Sequential Model in this notebook to keep things simple.\n",
    "In Sequential Model, you can add each layer sequentially.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pQdXYCEBeuhB"
   },
   "source": [
    "Let's build a simple dense network and find the optimal learning rate. We will need a callback to grow the learning rate at each iteration. It will also record the learning rate and the loss at each iteration:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zgi-eEN5r6w4"
   },
   "outputs": [],
   "source": [
    "K = keras.backend\n",
    "\n",
    "class ExponentialLearningRate(keras.callbacks.Callback):\n",
    "  def __init__(self, factor):\n",
    "    self.factor = factor\n",
    "    self.rates = []\n",
    "    self.losses = []\n",
    "  def on_batch_end(self, batch, logs):\n",
    "    self.rates.append(K.get_value(self.model.optimizer.lr))\n",
    "    self.losses.append(logs[\"loss\"])\n",
    "    K.set_value(self.model.optimizer.lr, self.model.optimizer.lr * self.factor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4wDES7egg2UV"
   },
   "outputs": [],
   "source": [
    "keras.backend.clear_session()\n",
    "np.random.seed(42)\n",
    "tf.random.set_seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "m_ZpNNhOhMmb"
   },
   "outputs": [],
   "source": [
    "model = keras.models.Sequential([\n",
    "    keras.layers.Flatten(input_shape=[28, 28]),\n",
    "    keras.layers.Dense(300, activation=\"relu\"),\n",
    "    keras.layers.Dense(100, activation=\"relu\"),\n",
    "    keras.layers.Dense(10, activation=\"softmax\")\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "m7naGk57hdqY"
   },
   "source": [
    "We will start with a small learning rate of 1e-3, and grow it by 0.5% at each iteration:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QJ0MHUhGhXOU"
   },
   "outputs": [],
   "source": [
    "model.compile(loss=\"sparse_categorical_crossentropy\",\n",
    "              optimizer=keras.optimizers.SGD(lr=1e-3),\n",
    "              metrics=[\"accuracy\"])\n",
    "expon_lr = ExponentialLearningRate(factor=1.005) # increasing by 0.5% factor at each iterton\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "U9cxa7gIh7rs"
   },
   "source": [
    "Now let's train the model for just 1 epoch:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 56
    },
    "colab_type": "code",
    "id": "oD0O6USzh5b9",
    "outputId": "7e1e93b8-3ddb-4d24-cba9-1352d00e1e34"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1719/1719 [==============================] - 6s 3ms/step - loss: 38081929542287021309952.0000 - accuracy: 0.5762 - val_loss: 2.3911 - val_accuracy: 0.1126\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, epochs=1,\n",
    "                    validation_data=(X_valid, y_valid),\n",
    "                    callbacks=[expon_lr])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "1cHAlelQj5ei"
   },
   "source": [
    "We can now plot the loss as a functionof the learning rate:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 302
    },
    "colab_type": "code",
    "id": "xBrmMGYEid8X",
    "outputId": "1e2f25f3-0858-4640-a3c7-23b59d963306"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Loss')"
      ]
     },
     "execution_count": 13,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(expon_lr.rates, expon_lr.losses)\n",
    "plt.gca().set_xscale('log')\n",
    "plt.hlines(min(expon_lr.losses), min(expon_lr.rates), max(expon_lr.rates))\n",
    "plt.axis([min(expon_lr.rates), max(expon_lr.rates), 0, expon_lr.losses[0]])\n",
    "plt.xlabel(\"Learning rate\")\n",
    "plt.ylabel(\"Loss\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "I65NysJplXh-"
   },
   "source": [
    "The loss starts shooting back up violently around 3e-1, so let's try using 2e-1 as our learning rate:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "w33d6v93lFgo"
   },
   "outputs": [],
   "source": [
    "keras.backend.clear_session()\n",
    "np.random.seed(42)\n",
    "tf.random.set_seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "q23s9cz2law0"
   },
   "outputs": [],
   "source": [
    "model = keras.models.Sequential([\n",
    "    keras.layers.Flatten(input_shape=[28, 28]),\n",
    "    keras.layers.Dense(300, activation=\"relu\"),\n",
    "    keras.layers.Dense(100, activation=\"relu\"),\n",
    "    keras.layers.Dense(10, activation=\"softmax\")\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5tTeJdIlljXE"
   },
   "outputs": [],
   "source": [
    "model.compile(loss=\"sparse_categorical_crossentropy\",\n",
    "              optimizer=keras.optimizers.SGD(lr=2e-1),\n",
    "              metrics=[\"accuracy\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "colab_type": "code",
    "id": "n5HSL1qXlpDd",
    "outputId": "8b62fd6d-5a30-4cbd-e7e2-aef547db008d"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "string"
      },
      "text/plain": [
       "'./my_mnsit_logs/run_001'"
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "run_index = 1 # increment this at every run\n",
    "run_logdir = os.path.join(os.curdir, \"my_mnsit_logs\", \"run_{:03d}\".format(run_index))\n",
    "run_logdir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "zyUCXplYmuq1",
    "outputId": "137502ee-c601-495f-e883-72f2744d4e52"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "   1/1719 [..............................] - ETA: 0s - loss: 2.5649 - accuracy: 0.0625WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/summary_ops_v2.py:1277: stop (from tensorflow.python.eager.profiler) is deprecated and will be removed after 2020-07-01.\n",
      "Instructions for updating:\n",
      "use `tf.profiler.experimental.stop` instead.\n",
      "WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0083s vs `on_train_batch_end` time: 0.0302s). Check your callbacks.\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.2375 - accuracy: 0.9271 - val_loss: 0.1052 - val_accuracy: 0.9674\n",
      "Epoch 2/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.0935 - accuracy: 0.9713 - val_loss: 0.0917 - val_accuracy: 0.9752\n",
      "Epoch 3/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.0647 - accuracy: 0.9791 - val_loss: 0.0735 - val_accuracy: 0.9784\n",
      "Epoch 4/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 0.0463 - accuracy: 0.9846 - val_loss: 0.0791 - val_accuracy: 0.9780\n",
      "Epoch 5/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.0335 - accuracy: 0.9892 - val_loss: 0.0728 - val_accuracy: 0.9794\n",
      "Epoch 6/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 0.0258 - accuracy: 0.9916 - val_loss: 0.0655 - val_accuracy: 0.9838\n",
      "Epoch 7/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 0.0191 - accuracy: 0.9941 - val_loss: 0.0773 - val_accuracy: 0.9806\n",
      "Epoch 8/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.0116 - accuracy: 0.9966 - val_loss: 0.0736 - val_accuracy: 0.9826\n",
      "Epoch 9/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 0.0082 - accuracy: 0.9975 - val_loss: 0.0749 - val_accuracy: 0.9820\n",
      "Epoch 10/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 0.0068 - accuracy: 0.9981 - val_loss: 0.0717 - val_accuracy: 0.9840\n",
      "Epoch 11/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.0038 - accuracy: 0.9993 - val_loss: 0.0741 - val_accuracy: 0.9846\n",
      "Epoch 12/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 0.0013 - accuracy: 0.9998 - val_loss: 0.0729 - val_accuracy: 0.9852\n",
      "Epoch 13/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 6.1827e-04 - accuracy: 0.9999 - val_loss: 0.0761 - val_accuracy: 0.9840\n",
      "Epoch 14/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 4.1232e-04 - accuracy: 1.0000 - val_loss: 0.0754 - val_accuracy: 0.9848\n",
      "Epoch 15/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 2.8165e-04 - accuracy: 1.0000 - val_loss: 0.0773 - val_accuracy: 0.9852\n",
      "Epoch 16/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 2.1609e-04 - accuracy: 1.0000 - val_loss: 0.0777 - val_accuracy: 0.9854\n",
      "Epoch 17/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 1.8939e-04 - accuracy: 1.0000 - val_loss: 0.0787 - val_accuracy: 0.9858\n",
      "Epoch 18/100\n",
      "1719/1719 [==============================] - 6s 3ms/step - loss: 1.6742e-04 - accuracy: 1.0000 - val_loss: 0.0791 - val_accuracy: 0.9854\n",
      "Epoch 19/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.5348e-04 - accuracy: 1.0000 - val_loss: 0.0795 - val_accuracy: 0.9856\n",
      "Epoch 20/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.4121e-04 - accuracy: 1.0000 - val_loss: 0.0803 - val_accuracy: 0.9856\n",
      "Epoch 21/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.3225e-04 - accuracy: 1.0000 - val_loss: 0.0806 - val_accuracy: 0.9860\n",
      "Epoch 22/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.2311e-04 - accuracy: 1.0000 - val_loss: 0.0812 - val_accuracy: 0.9854\n",
      "Epoch 23/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.1569e-04 - accuracy: 1.0000 - val_loss: 0.0819 - val_accuracy: 0.9854\n",
      "Epoch 24/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.0869e-04 - accuracy: 1.0000 - val_loss: 0.0821 - val_accuracy: 0.9852\n",
      "Epoch 25/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 1.0342e-04 - accuracy: 1.0000 - val_loss: 0.0826 - val_accuracy: 0.9856\n",
      "Epoch 26/100\n",
      "1719/1719 [==============================] - 5s 3ms/step - loss: 9.8508e-05 - accuracy: 1.0000 - val_loss: 0.0827 - val_accuracy: 0.9852\n"
     ]
    }
   ],
   "source": [
    "early_stopping_cb = keras.callbacks.EarlyStopping(patience=20)\n",
    "checkpoint_cb = keras.callbacks.ModelCheckpoint(\"my_mnsit_model.h5\", save_best_only=True)\n",
    "tenorboard_cb = keras.callbacks.TensorBoard(run_logdir)\n",
    "\n",
    "history = model.fit(X_train, y_train, epochs=100,\n",
    "                    validation_data=(X_valid, y_valid),\n",
    "                    callbacks=[early_stopping_cb, checkpoint_cb, tenorboard_cb])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "-HDHXrGbonS3"
   },
   "source": [
    "See, our model compilation stopped at around 23-29 epochs that means our best model is somewhere here and there"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "colab_type": "code",
    "id": "oEcY2OFhoDkn",
    "outputId": "65eb1b01-33fd-4490-debb-b2d1f7813573"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 2ms/step - loss: 0.0616 - accuracy: 0.0977\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.061620958149433136, 0.09769999980926514]"
      ]
     },
     "execution_count": 21,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = keras.models.load_model(\"my_mnsit_model.h5\") # rolling back to best model\n",
    "model.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "RlcCtSWdpsR1"
   },
   "source": [
    "We got over 97% accuracy. Finally 😃"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Th1aM_2WrbWz"
   },
   "outputs": [],
   "source": [
    "valid_p = np.argmax(model.predict(X_valid), axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "tctIaRAz5Lho"
   },
   "source": [
    "## Visualizing Result\n",
    "\n",
    "\n",
    "---\n",
    "\n",
    "#### All Errors in the Validation set\n",
    "\n",
    "Let's see all the errors in the validation set. It seems that in most of the cases, the recognition of digits is difficult for even humans. So we can say that our model is performing well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "rFEdIfzEwRdG",
    "outputId": "3503a545-4264-4b8f-c9ea-ab4bbd5074bf"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 81 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rows = 10\n",
    "cols = 10\n",
    "\n",
    "f = plt.figure(figsize=(2*cols,2*rows))\n",
    "sub_plot = 1\n",
    "for i in range(X_valid.shape[0]):\n",
    "    if y_valid[i]!=valid_p[i]:\n",
    "        f.add_subplot(rows,cols,sub_plot) \n",
    "        sub_plot+=1\n",
    "        plt.imshow(X_valid[i].reshape([28,28]),cmap=\"binary\", interpolation=\"nearest\")\n",
    "        plt.axis(\"off\")\n",
    "        plt.title(\"T: \"+str(y_valid[i])+\" P:\"+str(valid_p[i]), y=-0.15,color=\"Red\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "L0fQCvEb5niB"
   },
   "source": [
    "## Predict on Testset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "If__hnv25tNZ"
   },
   "outputs": [],
   "source": [
    "y_test = np.argmax(model.predict(X_test), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 590
    },
    "colab_type": "code",
    "id": "pUvHYXqU56zB",
    "outputId": "9e14ec93-6737-45d8-9965-bace157312c4"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABGkAAAI+CAYAAAAPT1feAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdebxV0//H8c9S0SQkJVJRmiSVDPWVkqEoRSOajJFvxggpNCgZ8s3wlQy/IkSpVGQqUcaixJdKNAhpkCZp3L8/bpbP2u65nXPvOWfvs8/r+Xj0+L53a91z1te+a99zd/uzlvE8TwAAAAAAABCs/YIeAAAAAAAAALhJAwAAAAAAEArcpAEAAAAAAAgBbtIAAAAAAACEADdpAAAAAAAAQoCbNAAAAAAAACHATRoAAAAAAIAQiNRNGmPMFt+f3caYR4MeFxJjjDnAGPOMMWaFMWazMWaBMebcoMeFxBhjehlj5hljthtjRgc9HuSPMaa0MWaSMWbr3jl5SdBjQv4YY441xvxpjBkb9FiQGK6n0cJczFzGmJrGmJnGmI3GmKXGmAuDHhMSZ4yZtXcO/vU74+Kgx4TERH0uRuomjed5Jf/6IyKHi8g2ERkf8LCQuMIi8qOINBGRg0Skn4i8YoypHOCYkLifRWSwiDwb9EBQII+LyA4RKScinUXkCWPMccEOCfn0uIjMDXoQyBeup9HCXMxAxpjCIvKaiEwTkdIi0kNExhpjqgU6MORXL/W7Y/WgB4P4ZcNcjNRNGp92IrJGRGYHPRAkxvO8rZ7n3eN53nLP8/Z4njdNRJaJyIlBjw3x8zxvoud5k0VkfdBjQf4YY0pIzrW0v+d5WzzPmyMiU0Ska7AjQ6KMMReJyO8iMiPosSBxXE+jg7mY0WqIyBEi8rDnebs9z5spIh8KPxOBdIv8XIzyTZruIvKc53le0ANBwRhjyolINRH5X9BjAbJMNRHZ5XneEvV3X4oIT9JkEGNMKREZKCI3Bz0WIJsxFyPJiEjtoAeBfBlqjFlnjPnQGNM06MGgwCI1FyN5k8YYU0lySmXGBD0WFIwxpoiIvCAiYzzPWxT0eIAsU1JENvn+bqOIHBjAWJB/g0TkGc/zVgU9ECDLMRcz22LJeUr/VmNMEWPMOZLz+0bxYIeFfLhNRI4RkSNFZJSITDXGVAl2SEhA5OdiJG/SSM6jTnM8z1sW9ECQf8aY/UTkeclZD6NXwMMBstEWESnl+7tSIrI5gLEgH4wxdUXkLBF5OOixANmMuZj5PM/bKSIXiEhLEVktIr1F5BUR4aZbhvE871PP8zZ7nrfd87wxklMqc17Q40J8smEuFg56ACnSTUTuC3oQyD9jjBGRZyRnsdLz9k5GAOm1REQKG2OO9Tzvu71/d4JQephJmopIZRFZmXNZlZIiUsgYU8vzvPoBjgvINk2FuZjxPM9bKDn/Yi8iIsaYj4Qn96PAk5xyGWSIqM/FyD1JY4xpJDmPrrGrU2Z7QkRqisj5nudtC3owSJwxprAxpqiIFJKcD6JF967Gjgzhed5WEZkoIgONMSWMMf8SkTaS84QbMsMoEakiInX3/hkpIq+LSPMgB4XEcD2NBOZiBBhj6uydf8WNMbeISHkRGR3wsJAAY8zBxpjmf11HjTGdReR0EXkz6LEhflGfi5G7SSM5CwZP9DyPx/Ez1N41ha6WnA8xq40xW/b+6Rzw0JCYfiKyTURuF5Eue3O/QEeE/LhWRIpJTu3vSyLS0/M8nqTJEJ7n/eF53uq//khOCdufnuetDXpsSAjX0wzHXIyMriLyi+T8TDxTRM72PG97sENCgoqIyGARWSsi60TkOhG5wLdJAsIv0nPRsPkRAAAAAABA8KL4JA0AAAAAAEDG4SYNAAAAAABACHCTBgAAAAAAIAS4SQMAAAAAABAC3KQBAAAAAAAIgcL7aGfrp2CYJL8e5zEYyTyPnMNgMBejgbmY+ZiL0cBczHzMxWhgLmY+5mI05HoeeZIGAAAAAAAgBLhJAwAAAAAAEALcpAEAAAAAAAgBbtIAAAAAAACEADdpAAAAAAAAQoCbNAAAAAAAACHATRoAAAAAAIAQ4CYNAAAAAABACHCTBgAAAAAAIAQKBz0AZI8HH3zQ5m3bttm8cOFCp9+ECRNy/fqePXs6xw0bNrS5a9euyRgiAAAAAACB4UkaAAAAAACAEOAmDQAAAAAAQAgYz/Pyas+zESljkvx6gZzHTp06Ocfjx49P6utXrVrV5nfffdfmihUrJvV9CiCZ5zGSc3HJkiU2V69e3eZHHnnE6XfdddelbUw+kZiL8dq6datzfOutt9o8cuRImxs0aOD003O7UqVKKRpdgTAXM19WzcUIYy5mPuZiNDAXE7BhwwabV65cGdfX+D8PPfzwwzbXrl3baatWrZrNJ5xwQrzDYi5GQ67nkSdpAAAAAAAAQoCbNAAAAAAAACHATRoAAAAAAIAQYAtuJJVehybeNWhq1KjhHLdo0cLmH374weYpU6Y4/ZYuXWrz2LFjbe7bt298g0Xg5s+fb/N++/19z/jII48MYjhZ7+eff3aOn3rqKZsLFSpk87x585x+U6dOtblXr14pGh20L774wua2bdvavHz58pS+79tvv21zzZo1bT7qqKNS+r6Ij56LrVu3tvnRRx91+vXs2dNmPbexb2vWrHGOO3bsaHOjRo1s7tGjh9OvcuXKKR3XXzZu3Ogcf/DBBzbrz1ciIkWKFEnLmIBMMG3aNOdYX09nzZpl83fffRfX6+m1FkXcn8/bt2+P+XV79uyJ6/URbTxJAwAAAAAAEALcpAEAAAAAAAgByp1QIP6yh0mTJsXsq7eb06VLZcqUcfqVLFnS5h07dth8yimnOP2+/PJLm9evXx/niBEmCxYssFmfd12+gdRau3atzd27dw9wJEjEW2+9ZXNej00nm752P/vsszaPGzcubWPA3/w/+3QZk3bdddc5x1dccYXNxYoVS/7AIkZvv3vcccc5bbq8qFy5cjanq7zJP4b69es7bevWrbPZ/5nt2GOPTe3AMtSmTZtsvv322522//3vfza/++67NlM6Fl7ff/+9c/z444/bPGrUKJu3bdvm9PO8gu1IvXjx4gJ9PbIbT9IAAAAAAACEADdpAAAAAAAAQiCt5U4TJkywWe8aIiJyxBFH2Fy0aFGnrXPnzjYffvjhNletWjXZQ0SCfvnlF+dYPxqoy5tE3Mfzy5cvH9frP/jggzZ/++23Mfu1atUqrtdDsL766ivnWO840q1bt3QPJ2s98sgjNk+ePNnmuXPn5uv1Zs+ebbO+BpxwwglOv9NPPz1frw+RXbt2OcdvvPFGIONo0KCBzcOHD7d569atTr8SJUqkbUzZTO/cIyLy008/5drv4osvdo79n7Pg0iVCIu4OTv4Ss3//+982+3fRSpfBgwfbvGzZMqdNl3NQ3hSb3iW0X79+Nq9cuTLm1+iyqEMPPTQ1A0OBrVq1yjn+z3/+k7L30jvW+n8PQnLpXX71Ndu/9IbemUvvJHvNNdc4/fTufGG4VvIkDQAAAAAAQAhwkwYAAAAAACAEuEkDAAAAAAAQAmYf24sVbO8xn6OPPtrm5cuX5+s1SpUqZXOtWrUKOqS4HXXUUTb36dPHadM1+klikvx6ST2PeVmxYoXNBx54oNNWunTphF9Pr2nhX89EmzFjhs1nnHFGwu+TIsk8j2k7h6mk16USEenQoYPNuma0SZMm6RrSvmTsXMyLrsktVKhQwl+/e/du5zjWa1SsWNE5fuWVV2w+8cQTE37fAsj4ufjOO+84xy1atLD5tttus3nIkCEpHYdeh+bWW2+1efXq1U6/ww47LNlvHcm5mB96y3VdQy8i8sUXX+T6Nf41jM4999zkDyw+GTEX3377bedYzze/X3/91eYUfN/H9PXXX9t8/PHH23zhhRc6/caMGWOz/3NZPkViLvrXKalXr57Nen0LY2L/373oootsfuyxx5y2/HzmTbOMmIt++tz415Y57bTTbNZz9uOPP3b6nXfeeTaXLFnS5i1btjj9mjdvbrNeX+aUU05x+unvnWLFitmchrXZIjEX86J/99Nbp4uITJw40ea1a9cW+L2KFClic/Xq1W3W31ciIiNGjLB5//33L/D7SozzyJM0AAAAAAAAIcBNGgAAAAAAgBBI6xbcTz/9tM1ffvml06ZLl7755hunbf78+TbrkohPPvnE6acfrc9ryzxNP9pUpkwZp01vL63fS5c+iaSk3CljVapUqcCv8cADD9i8ZMmSmP3044b+Rw8RTvfff79zXLlyZZuZR6mjH+0VcbfJ9pcuxcN/rdSP9OqSR/9WsCeddJLNe/bsSfh9s41+zFc/Vi8iUrVqVZv79u2btjFNmTIlbe+F3C1cuNDmWOVNIiKFC//9ES/A8qaMsWbNGptfffXVmP2effZZ5zhdJU66vElE5Oyzz861X9u2bZ3jJJU4Rc6DDz7oHPu3Vo/HuHHjbJ4+fbrTprfxvu6662xOUnlE1ti6datzrL/v/b9LTp48OdfXaNiwoXOsf6/Un0P9vztWqFDBZl0mjuTSP9NE3LKml19+2eaNGzfGfA19rho3buy06XOsf8f0l91/+umnNuvrgb9cWC/F4d/GO5n4jgMAAAAAAAgBbtIAAAAAAACEADdpAAAAAAAAQiCta9KceeaZuWa/vLY63LBhg826plDEXdNi7ty5cY3pgAMOsFlvtyUiUqNGDZt/++03m6tUqRLXayM+06ZNc47vuusum/VWo+XKlXP63XfffTYXL148RaNDQS1fvtxm/7zUcy4NWxVmlffff9/mRYsWOW16S9F4t+DWdbfnnHOO03bQQQfZPHPmTJvvvffemK/3xBNP2NyzZ8+4xpBt9H+/P/74w2kbO3aszXoL0WTTP/tE3O+rvLamRerobUfzEmvNEuSud+/eNuv5JSJSv359mzt06JC2MWlz5sxxjvW295dddpnNXbp0SduYMo1eM+3//u//YvbTa074P3u+8847uX6Nf70MveZN586dbT788MPjG2wW27Fjh82XXHKJ06bXofGvx3bWWWfF9fp6jRJNr22K1Lr66qttnjRpktMWaztt//k9/vjjbR4yZIjNRYsWjfm+ejt2/TlUxL2OLliwwGb/nL322mttbteundOWzDXKeJIGAAAAAAAgBLhJAwAAAAAAEAJpLXdKhkMOOcTmZs2axeyXVzlVLP4tF3VpVZ06dWz2b4WKgpk3b55zrEuctE6dOjnHTZo0SdmYkDy6PMIvXVuXZgNdVibiXqfWrVsX12v4H/Vt3769zXfffbfNeZUXVqpUyeYnn3zSadPj6NOnj81//vmn069Xr142FylSZF/DjowJEyY4x3rbR73ltoi7nXkqDR482DnWJU5Nmza1+eCDD07LeJD3NVVv76sf/8a+6e9tfynfkUceaXOqt1Detm2bzfoc6m1pRdwx+rcFR+50CcOmTZucttNPP91mPcf8P59efPFFm4cOHWrz0qVLnX66HK1NmzY2+7fqLl26dFxjj7otW7bYrL/vp06d6vTTnxtvvfVWp42lD8JFz53777/faXvqqads9jzPaStbtqzNuhzef77zs0yC3lp7165dTtuAAQNsbt68uc3+z9fpwpM0AAAAAAAAIcBNGgAAAAAAgBDIuHKnZFuzZo3NerVmEffxK73jEI8mFtwFF1xg81tvvRWzX/fu3W32P3aPzLBw4cKYbbrkBQWzc+dO5zjeEif9iPfLL7/stJUpUybhcehyJ//OCzfffLPNW7dutdn/fdC6dWubs2k3vfHjxzvH+r9ROnfA0o/26kf7RUQKF/77Y0O/fv1szqaytCB89NFHNuvdKfz04/5169ZN6Ziyid6F0r+7nS71y888nTVrVszjTz75JObXBbXLVCbT5fT+krabbrop16/x7xRz+eWX26xLVL///nunn/4dQs/LVJfLZarJkyfbrHdv1Z8pRERmz55ts95ZEuGjr2UPPPCA06bnhy4nFXF3MDz55JMTft/du3c7xz/++KPN3bp1s7lly5ZOP73MSV66du1qcypLvXmSBgAAAAAAIAS4SQMAAAAAABAC3KQBAAAAAAAIgaxfk0ZvaajXpxFx68yqV6+etjFF1S+//GKzrq/3b7mtt9fTax6ULFkyhaNDMuk1E/7v//7P5nr16jn9zj777LSNCX/T2zfr85OfNWjyoteWERF54YUXbP7ss8+S+l6ZauPGjTbntf6Ef820VBo1apTNa9euddpq1aplc7NmzdI2pmw3d+7cuPqlc+2iqLnhhhtsnjlzptP2888/2+zfAl2vrfDaa68l/L7+7Wf9a6X8xb8+F1usJ+6ll16K2fb666/brNdNzMu8efPi6nfqqafazGfZ3OnfCzT/58YKFSqkYzhIAr3FdaFChWL2869p9+mnn9qs131atGhRzNcoVqyYzd9++63Tpo/159zVq1fHfD2tXLlyznG61uPjSRoAAAAAAIAQ4CYNAAAAAABACGRludOcOXNs1tu8+enHVmvXrp3SMWWDtm3b2pzX9sCdO3e2OZu2342SGTNm2Ky3tGvRooXTz7+1JZLHvwWhph8lTSX/Y/x79uzJtc0/1rvvvtvmsWPHpmh04aDLPVetWuW0XXzxxekejoj8cytZjZ+FwYhV7uTf/jOdZXFRc+KJJ9r81VdfOW0LFiyw+c0333Ta7r//fpvLli1rc/fu3eN6X72dq4hInTp1cu3XqFEj55jPR4nT11R/aZqeY7qswv+9MGnSJJv15xv/XNRtuoTUf751CWk202Ut2vTp053jAQMG2OwvqfaXRiFYZ555ps1nnHGG0/bOO+/YvGLFCqft+uuvj+v1Cxf++zaGLq3KS14lTvvt9/ezK/p31kceecTpV758+bjeq6B4kgYAAAAAACAEuEkDAAAAAAAQAsb/OLpPno2Zqm/fvjYPHTrU5rPOOsvp98Ybb9icytWbc5H70v75F8h5nDJlinPcsWNHm3fs2GFz06ZNY35dhq+Cn8zzmFFzsUOHDjbrR1gnTpzo9LvwwgvTNqZ8ypi52Lt3b+fY/3imtnPnzlQNw/Hoo486xzfffLPNusTJv+q/ftQ8SY/0h3Yubtu2zebTTjvNadOP77733ntOW+nSpZM5DGd3w8MPPzxmP/191atXr6SOYR8yZi4mgy7LFhE5/fTTbdaf2ypVquT0W758eUrHlQShnYtB+eGHH5xjfc2rW7euzW+//bbTT++EmWYZOxd/++03m/0/W/ROe3qOxdptS8TdoVLvFisi0qpVK5uXLFlic48ePZx+I0eO3NewUyVUc1H/d87rv7nm/+xwzTXX2HzKKafY/OOPPzr9qlatavNxxx0X8/X/97//2dywYUObQ7TDVMbOxd9//91m/9IjH374oc2HHnqozRUrVnT66XLxL7/80ub8lvTr3RH17nn+UsYUyPU88iQNAAAAAABACHCTBgAAAAAAIAS4SQMAAAAAABACWbEFt675F3G3TzzggANs1tu6iaR9HZpIWL9+vc26nk/EXYdG0zXXIhm/Dk1W8m9pN3v2bJtr1KhhcwasQZOxpk2bFth7r1271uZvvvnGZv81IJYyZco4x9l07S1WrJjNuk5exF3PqWXLlk6bXt8nXl9//bXN/m229RaYea0HoLeoROron6Ui/9zO/i96TQxkpoEDBzrHev7p7b0DXIMmMvRaXuPHj3fa2rdvb3Os9WlE3O2Bhw0bZnPRokWdfnoLX73+5VtvveX009fibN5W/ZZbbrH5oYceiutr9Np2Iu66QP41ggqqbNmyNvvX0hw3blxS3ysb6HVe/GvS5Ee3bt1szmtNmlKlStk8fPhwp+3SSy+12b/eURD4tAUAAAAAABAC3KQBAAAAAAAIgawod3rggQec4/nz59t87rnn2tyoUaO0jSmq9COKn332Wcx+F1xwgc3+R32ReUaPHu0c//rrrzbrOYZouvfee22O9xHjypUr2zxmzBinzb/NYra45557nGP9mL2/nO2iiy5K+PV1uYS/pGndunVxvcZll12W8Psicf5SDE0/Ju7fzheZQZ9f//VPP46vt59Fcp111lnOsS4vffHFF232b7+rP7P6S5y0/v372/ztt9/a/Nprr8V8Pf/3QjbRJS8dO3a0uXPnzk6/nTt32rxq1SqnzV/+lExr1qyx2X99rl27ts39+vVL2Rjg0uWg8ZacPfHEEzZfcsklSR9TMvEkDQAAAAAAQAhwkwYAAAAAACAEIlvupB8NHzRokNN20EEH2awfR0TB+VfKjkWXRLCbU+bTO8P4HXLIIWkcCdLhvPPOc44XLVqU8GvUqlXL5saNGxd4TFFQs2ZN5/iVV16xWZfpivxzd6Z46N1L/Lp3727z2LFjY/bTu1EhufSj+7rcwq9ChQo2n3TSSSkdE1Jj+vTpMdv0Tm7169dPx3AgbvmTvxQqP/S1slOnTjb7y53ee+89m3/77Teb9U5U2UDvpqOva0uWLIn5NTNmzHCOdSmULh/Oa/mF/PDv+PX5558n9fWRu6effto5Hjx4sM363PvpcrR27dolf2ApwpM0AAAAAAAAIcBNGgAAAAAAgBDgJg0AAAAAAEAIRGpNmvXr19t8/fXX27xr1y6nn15PoWHDhqkfGP5Bn6siRYrk6zX02kL6Nfx1iRs3bsz16zds2OAcP/zww3G9r66bHTZsmNNWvHjxuF4jaqZOnRqzrVWrVmkcSfby10jntRVlrPUQrrrqKuf4559/juu9/Ns5x8O/pTTyVq9evTyPC+qYY46Jq99XX31l8/HHH5/UMWS7jz76yGb/HNPatGmTjuEghfQ1uESJEk7bLbfcku7hIMX0ltJTpkxx2vTWwY899pjNd911V+oHluHOPPPMmG0LFiyw2b8mjf6d4bLLLrPZ/xlI/16Q1zphSB197nr37u20bd68OdevOfDAA51jve32AQcckMTRpRZP0gAAAAAAAIQAN2kAAAAAAABCIKPLnfyP87do0cLmZcuW2Vy1alWnn39LbqRfnTp1Cvwa+vHR8uXL2/zrr786/fSjpMlWrlw557hfv34pe6+wmT17ts3+/+ZIv549ezrHffr0idlXb/Gqy/f8YrX5r715vYZ2zTXXxNUP6afLa/IqtaHEKXV0GbBfmTJlbL7xxhvTMRwk2ciRI21evXq1zf7PEWy7HT377ff3v4n7fzZPnjzZZr1t9EUXXeT0q1atWmoGF1HnnHOOzX379nXa9LIIo0aNsvm7775z+s2aNSuu9zryyCPzMULEQy+nsGnTppj9dNmov6TwtNNOS/7A0oAnaQAAAAAAAEKAmzQAAAAAAAAhkNHlTt9//71zPG/evFz7DR8+3DmuUqVKysaU7fTOWfoRzlR45ZVXEv4avaK7fvzUr3Xr1s5xgwYNcu2XqY/QJcOkSZNs9u+gpneeadKkSdrGlM3atm3rHN9///02r1u3LqXvrUsxatasafNTTz3l9NNliQgXvUNXfnbrQsG99dZbMduOOuoom/XOhsgcutxJzzH9uclP717i35GyYsWKSRwd0qVu3brOsV6CQe/sdccddzj9xo4da3OxYsVSNLro0J9FOnXq5LS9/PLLuX7Ne++9F/P1Chf++1dmXTIu8s+dXlEw+rqnP8vmpUuXLjY3bdo02UMKBE/SAAAAAAAAhAA3aQAAAAAAAEKAmzQAAAAAAAAhkHFr0qxYscJmvb2a34MPPmhzq1atUjom/G3ixIk2++sId+zYEddrfPPNNzbHu332FVdcYXOlSpVi9mvXrp3Nul4V8fnjjz9snj59esx+HTp0sDne7ZlRMP7ve11z7V8f6j//+U9S3/vOO++0uVevXkl9baTHn3/+mevfs/ZBaumtYJcuXRqzX9GiRW3Wa6sh8+m1LkTctUcefvhhm2vXru30GzNmTGoHhrTo1q2bzU8++aTN+vO0iLs9dJ06dVI/sAynf3b5P/PoNU8+//xzm3/99VenX+XKlW3W50lvlY6C27Jli3Osfz/L63fHE044weZkf64NA56kAQAAAAAACAFu0gAAAAAAAISA8Twvr/Y8G4PQt29fm4cOHRqz39y5c22OtX1yiCV7/9PQnccskczzGIpzqB/NP/30020uV66c0+/FF1+0uXjx4qkfWOpEci6++eabNo8aNcrmqVOnOv3OP/98m6+++mqb/T83atWqZXNIt4WN3FxMtsMPP9xmPc/vuusup98NN9yQtjH5RHIu7t692+Yrr7zS5tGjRzv99KP2GV7mkrVzUW+9vHDhQpv911O9Pbf+nujfv7/TT2/LnmaRnIthsHLlSpv9JcwXX3yxzfozVgFk7VzUnn/+eZs//vhjp02XNZUtWzZdQ0pEJObilClTnOM2bdrE9XUzZsywuVmzZkkdU5rleh55kgYAAAAAACAEuEkDAAAAAAAQAtykAQAAAAAACIGMWJNm9uzZNrds2dJmvYWaH2vSOEJxHrMQ9b6Zj7kYDczFfdDrD9100002h6jOO/Jz8eeff7a5X79+Tlv9+vVtzvBt7rN2LurPsnfffbfNen03EZGePXvafMghh9i8//77p3B0CYn8XAyDc845xzn+6KOPbP7ss8+cNr0uXAKydi5GSCTmot5KW8Rds0vr06ePczxs2LCUjSnNWJMGAAAAAAAgrLhJAwAAAAAAEAIZUe6kt9rWW3D7Va1a1Wa9lWyNGjVSM7DUicTja+BR0ghgLkYDczHzMRejgbmY+ZiLabBp0ybnWJeEjBgxwmlr3bp1ft6CuZj5IjEXjzrqKOd41apVNuutzxcsWOD0K1++fGoHlj6UOwEAAAAAAIQVN2kAAAAAAABCoHDQAyiIunXrOsczZsywuXTp0ukeDgAAAAAUSKlSpZzjZcuWBTQSILVuvvnmmMf9+/e3OULlTXHhSRoAAAAAAIAQ4CYNAAAAAABACHCTBgAAAAAAIAQyYgvuLBSJLdXA9oYRwFyMBuZi5mMuRgNzMfMxF6OBuZj5mIvRwBbcAAAAAAAAYcVNGgAAAAAAgBDYV7kTAAAAAAAA0oAnaQAAAAAAAEKAmzQAAAAAAAAhwE0aAAAAAACAEIjcTRpjTGljzCx5On4AACAASURBVCRjzFZjzApjzCVBjwn5Y4y5yBjz7d5z+b0xpnHQY0L8jDE1jTEzjTEbjTFLjTEXBj0mJI5rauYzxvQyxswzxmw3xowOejzIH2PMLGPMn8aYLXv/LA56TEgMPxczn5p/f/3ZbYx5NOhxITHGmLHGmF+MMZuMMUuMMVcGPSbkjzHm2L0/G8cGPZZkitxNGhF5XER2iEg5EeksIk8YY44LdkhIlDHmbBEZJiKXiciBInK6iPwQ6KAQN2NMYRF5TUSmiUhpEekhImONMdUCHRjyg2tq5vtZRAaLyLNBDwQF1svzvJJ7/1QPejCIHz8Xo0HNv5IicriIbBOR8QEPC4kbKiKVPc8rJSKtRWSwMebEgMeE/HlcROYGPYhki9RNGmNMCRFpJyL9Pc/b4nneHBGZIiJdgx0Z8mGAiAz0PO8Tz/P2eJ73k+d5PwU9KMSthogcISIPe5632/O8mSLyoTAXMwrX1GjwPG+i53mTRWR90GMBshg/F6OnnYisEZHZQQ8EifE873+e523/63DvnyoBDgn5YIy5SER+F5EZQY8l2SJ1k0ZEqonILs/zlqi/+1JE+FffDGKMKSQiDUTksL2PA68yxjxmjCkW9NhQIEZEagc9CCSEayoQLkONMeuMMR8aY5oGPRgUGD8XM1t3EXnO8zwv6IEgccaY/xpj/hCRRSLyi4i8EfCQkABjTCkRGSgiNwc9llSI2k2akiKyyfd3GyWnXAaZo5yIFBGR9iLSWETqikg9EekX5KCQkMWS869LtxpjihhjzhGRJiJSPNhhIUFcU4HwuE1EjhGRI0VklIhMNcbwL7+Zg5+LEWKMqSQ5529M0GNB/nied63kfJ5pLCITRWR73l+BkBkkIs94nrcq6IGkQtRu0mwRkVK+vyslIpsDGAvyb9ve/33U87xfPM9bJyLDReS8AMeEBHiet1NELhCRliKyWkR6i8grIhLJC2mEcU0FQsLzvE89z9vsed52z/PGSE6pDD8XMwQ/FyOnq4jM8TxvWdADQf7tLT2cIyIVRKRn0ONBfIwxdUXkLBF5OOixpErhoAeQZEtEpLAx5ljP877b+3cniMj/AhwTEuR53gZjzCrJqQ+1fx3UeJA/nuctlJx/ZRIREWPMR8K/OGUarqlAeHmSUy6DDMHPxUjpJiL3BT0IJE1hYU2aTNJURCqLyEpjjEjOk9+FjDG1PM+rH+C4kiZST9J4nrdVch5XG2iMKWGM+ZeItBGR54MdGfLh/0TkOmNMWWPMISJyk+TsiIAMYYypY4wpaowpboy5RUTKi8jogIeFBHBNjQZjTGFjTFERKSQ5H2KK7t1pBhnCGHOwMab5X+fOGNNZcnY9fDPosSF+/FyMBmNMI8kpO2RXpwy093eLi4wxJY0xhYwxzUXkYong4rMRNkpybqrV3ftnpIi8LiLNgxxUMkXqJs1e14pIMcmp+31JRHp6nse/+maeQZKzndoSEflWROaLyL2BjgiJ6io5C7GtEZEzReRstZI+MgfX1MzXT3LKSG8XkS57M2t8ZZYikrON+loRWSci14nIBb5FvRF+/FyMhu4iMtHzPEp/M5MnOaVNq0Rkg4g8KCI3ep43JdBRIW6e5/3hed7qv/5ITnn+n57nrQ16bMliWJAcAAAAAAAgeFF8kgYAAAAAACDjcJMGAAAAAAAgBLhJAwAAAAAAEALcpAEAAAAAAAgBbtIAAAAAAACEQOF9tLP1UzBMkl+P8xiMZJ5HzmEwmIvRwFzMfMzFaGAuZj7mYjQwFzMfczEacj2PPEkDAAAAAAAQAtykAQAAAAAACAFu0gAAAAAAAIQAN2kAAAAAAABCgJs0AAAAAAAAIcBNGgAAAAAAgBDgJg0AAAAAAEAIcJMGAAAAAAAgBAoHPQBkj+3bt9vcqFEjm+fPn+/0a926tc2TJ09O/cAAAAAAAAgBnqQBAAAAAAAIAW7SAAAAAAAAhEDGlTvNnj3bZl0yIyKyePFim6dNm2bz66+/7vRr2bJlrq/dsGFD57hx48b5Hifc8iYRkZtuusnmBQsW2GyMcfqdeOKJqR0YAGSJe+65x+YBAwbY3LRpU6ffe++9l6YRIR6ff/65czxp0iSbX331VZv15x4REc/zbNY/W/0/V2vWrGnzHXfckevfA0AQtmzZYvOPP/7otD3xxBMxv+7yyy+3uW7duskfGJBGPEkDAAAAAAAQAtykAQAAAAAACAFu0gAAAAAAAISA0fXLucizMVU2bdrkHHfu3NnmGTNm2FysWDGn386dO23evHlzwu/rf70SJUrYrGsg27dvn/BrJ8jsu0tCAjmPDzzwgHN8++2323zmmWfaPHDgQKffqaeemtqBpU8yz2Mg5xDRmIvx2rBhg3M8f/58m998802b/XNbr33RoUMHmytVquT06927t83lypUr2GATk7Vz8YwzzrB51qxZMfvpNWn869WERMbOxVGjRtm8aNEip02vs6f516TRcyzWujMiIj169LD5wgsvtPmcc85JYMQplbVzMUIydi7CEaq5qNeh0Z8xBg0aFPdrFC7891KrnTp1snnEiBFOv9KlS+dniGHEXCyAiy66yDlu1aqVzV26dEnnUHI9jzxJAwAAAAAAEALcpAEAAAAAAAiBUG7BfdtttznHejttbdu2bc6x3jqybNmyNpcqVSrme+3Zs8dm/1bd+vWvuOIKm6tVq+b0q1OnTszXz2a//PJLzLazzjrL5giVNwEZQZeGPvTQQzY/9thjTr9Yc9hfYqGPJ0yYEPN9161bZ/Ozzz4b32BRIHmVOMXqF9Jyp4x19dVX2+yfO8WLF7dZf4a58cYbnX41atSwuUyZMja3bds2aeNEwel5NHHiRKdNXxv1tbVevXpOv44dO9qsy8SBKBsyZIjN9913X75eY9euXTa/8MILNuulMkRERo8ebXOISkGRBvr3/pkzZzpttWrVSvdw8sSTNAAAAAAAACHATRoAAAAAAIAQCE2509dff21zXo/LH3XUUTY/99xzTlvVqlVtPvjgg20uWbJkzNfTjz35dxnSK4rrHafuuecep98zzzxj8yGHHBLzvbKNXqldRGT//fe3WZc7IfPpXYBERPr372/zG2+8YbN/N7lYuwLde++9Tr/y5cvbrHeh0buEifxzhzbk7sknn7T5zjvvTPjr/eUw77//flxfN2bMGJspdwoX/881JI8uSZo8ebLTpkuc5s6dm7YxIf9Wr17tHOtdtD777DOb/T/v9OfX6tWr2/zjjz86/fQ1We+Qd/HFF+dzxNGjP1eIuOdgx44dcb2G//NCmzZtcu3n36XwhhtusPnTTz+1WZchioicdtppcY0DOY4++uhc/95fItqrVy+bjzvuOKdNn/u77rrLZv+c1edaL7HRp08fp58uR0U06N9X1q5dG+BI9o0naQAAAAAAAEKAmzQAAAAAAAAhwE0aAAAAAACAEDD+mlmfPBuT6ZNPPrG5YcOGTpuuR3zkkUds1nWJqdC3b1+bH3zwQZv19rUiIlOnTrW5VatWyXhrs+8uCUnbefz5559trlChgtPWqFEjm+fMmZOuIQUpmecxbecwL/p7X69Dcumllzr9Ym3dnNeaNFrXrl2d45UrV9qstzj1r0vVpUuXXF+vADJ2Lmp6zS8Rdy0fvS12XoYNG2azrskXcWu/77///rheb/fu3XH1S5LIzcV4xZpjfvv4LBAGGTsXdd37ySef7LRt3brV5nnz5tlcsWLF1A8sGBk5F/V10r9l74IFC2zW65fotb9ERE455RSbDzroIJv9a9K0bt3aZr1Oh3+9xvHjx9ust/E+9thjnX7xXgMSEPhcHDlypHPcs2fPpA1mX/T6inoNlP32c//dW59vveaef5vfypUr26zXKkqDUM3F8847z+Y333zT5k6dOjn9Xnrppbheb/bs2TbrdcFERNavX5/r13Tu3Nk51mvnFSlSJK73TbPA52IyLFmyxDm+5ZZbbH700Udt9q8PlR+ff/65zQ0aNHDaJk6caLNe5yoNcj2PPEkDAAAAAAAQAtykAQAAAAAACIHQbMG9ffv2mG26lCLVJU7akCFDbB43bpzNy5Ytc/rpx6OSVO6UsQYPHhz0EOTjjz92jletWpVrvxNOOME5rlatWsrGFBVffPGFzc2bN4/Z74gjjrD5scceszmv7QxXrFgRs991111n8wEHHGCz3pobLl3ipEs3RdzyC/0ovP9R0ilTptisH9H2P9Y9cOBAm/Ujovqxff/71qlTx+aFCxfG+H+Bgrr77rttHjBgQMx+egtutuNOrsMOO8zmq666ymnr16+fzbqkJsLlThnpgQcesFmXN4mIHHnkkTYvXrzYZl0Wkxe9NbeIW9akf975t52OtSW3LqET+edW01FwxRVXOMe6FGXp0qU25zWPtm3b5hzrn3d5+fbbb21es2aNzXv27HH66c+i/s+lWtGiRW3WW0Dndb2OounTp9usP5foLekT0bhxY5tfe+01p+2OO+6wWZdFvfDCC04/XQY8evRomwsXDs2vz5GglzwRcZcR6d69u83JKHfS1wc/fS0PA56kAQAAAAAACAFu0gAAAAAAAIQAN2kAAAAAAABCIDRFdf3794/ZprexC0qLFi1sfuKJJ5w2fy1dNnv99ddjtl155ZVJfS+95aJ+3w0bNjj9/vjjj1y/vlSpUs7xzTffbHNe34/ZxL91s3+Nkb+cddZZzvHQoUNtrl+/flzvpbdvb9OmjdP2+++/26xrtvVW0nDNnz/f5mnTpjltus5a1/L/+9//dvrVrl07rvfSr6G3GPZvzf7QQw/Z/NVXX9nco0cPp9+oUaPiel/sW7ataxB2/nUr9Fz85ptvcv37vNSsWdM5zmvdL8RPr0MoIjJ8+HCbDz30UKdNr1ES7zo0ealSpYrN+nuia9euMb/mggsusFmvcRJV/u2Q/WvU5If+DJgX/bPrnXfeidlPbxU9b968mP3+/PNPm0eMGBFzPHrb9ijSnyNnzJhhc8mSJQv82o0aNXKO77//fpv11t/+3x9efPFFm/Xn344dOxZ4TPjbzJkzY7Yle52YJ5980uaDDz7YaYv395V04UkaAAAAAACAEOAmDQAAAAAAQAgEVu70ww8/OMc//fSTzf7Hj44//vi0jCkvzZo1s9lf7pTN/KVEO3futLlChQpOm7/0IZZdu3bZrLd81o/zioisXr3aZv1ouN7uVMR9hFK/3sqVK51++hG4bt26OW3J2PYtE/m3VNdbKOvt5nUZi4jIsccem/B76dIqfZ78dOkhYou1naVf06ZNbe7du3dSx3DffffFHJN+ZHzu3LlJfV8gTPR185lnnnHa9NzUW436y510P92mt7wXEencubPNbdu2zeeIsXDhQud49+7dNh933HFOWzLKMWLxf46K5cADD7Q5r+s9Ck7/TpLX7yfXXnutzfp3HP/PxaefftrmjRs32uz/XDVw4MDEB5tBdOmmLnfKi/5vJ+KWJ1199dVxvcYll1xi8+OPPx6z35IlS+J6PcRn8+bNNvvPd6dOnWzWJfTJoH/H3G8/91mVsG2tzpM0AAAAAAAAIcBNGgAAAAAAgBAI7LmesWPHOse6/Kl9+/ZOm39VboSH/1HDX3/91eZ4HzXUu/qIuDu7DBo0KObX6RW/9a4H+hFTkdiPC/t3KtI7RP3yyy9OWzaVO1111VU2v/LKK06bfqxbP7Kbn/ImEbc8Tu8I5X/UX5fkNGnSJF/vFXXr1693jj/99NO4vi6vHUOSTb+X3qULiBJd3iQicvrpp9u8YsUKp+3EE0+0WT/uf9ppp8V8/aeeespmf2noxIkTbdZlL/6SQv1e7Aj1T99//33MtnReu9566y2b9S5Afh06dEjHcJAAvcuW3rHrtttuc/rpz9F619F4lwiIigYNGuT69/7SQz0PevXq5bTt2LHD5lmzZiVvcOKWqtaoUcNpO/vss22O+i5cyaJ3rlu1apXTpkuc/CVJ+aF3iNW78Z1zzjkFfu1U4kkaAAAAAACAEOAmDQAAAAAAQAhwkwYAAAAAACAEAluT5qWXXnKO9bbbN9xwQ7qHg3yaP39+zLZ41ynxb/M8cuRIm3VN/Zlnnun0Gz58uM21a9eO6720qlWrJvw12WDevHk2+7fyLFGihM21atVK+LX1GjQiIv3797f5gw8+iPm+d911V8LvlW0+//xz53j58uUx++o1Mlq2bJmqIcVN1wuLuGtClS9fPt3DAQpk0aJFzvHixYttbteundM2fvz4hF+/R48eNq9bt85p0+v9TZ482eaTTjrJ6aev33oMeq2abPPHH3/YPGnSpJj99Hp4qaDX1ejbt6/N27dvd/rpbbfz2goa4fLaa6/FbNu0aZPNEyZMcNqivo7bBRdcYPNzzz1nc7NmzZx+eu1Lve6PiDt3kk2vJ9axY0enTa/rpdcMa9OmTcx+2W7OnDkx2/Q6lMnw8ssv26x/ZurPwmHEkzQAAAAAAAAhwE0aAAAAAACAEAis3MlPb2eW19aTCBf/9tnxWrJkic3jxo2L2U8/1j1ixAinbf/998/Xe8eit0KtX79+Ul87m+mym//+979O20MPPZTr1xxxxBHOcd26dZM+rqjRZWr7MmDAAJsPOeSQVAwnIStXrnSOv/76a5spd0qPe+65J+ghREbjxo2d4z179qTsvcqUKeMc33jjjbnmUaNGOf30I/lNmjSxefr06U4//XMxm+zatStt7+UvA545c6bNeW0Ffvnll9tcqVKl5A8MSfPDDz/YnNe1Vm/ffNVVV6VySKGjtx/v0qVLzH4lS5a0WZd3irilm7/99pvNr7/+ejKGGJMulezcubPN/jLEF154web8LNOQyfzlmo8//rjNpUuXdtr075a6ny51E3H/u7///vsx39vzvFz/ftu2bXmMOHg8SQMAAAAAABAC3KQBAAAAAAAIgbSWO23dutXmdD5KitTRK9GLuI+UxXq8TETk0Ucftdm/s4t+VPCJJ54o6BBj2rJli3NcuPDf0yHZpVSZRO/usXDhQqdNPz5ar169uF5v7dq1NvvL4/y7OP3Fv5OX3v0NudOPfYrkPf90eUNQ8hofgOTSpcMiIm3btrU5r93edImq/poo0p8BKleu7LTpst23337baTvhhBMSfi+9g93zzz/vtN1+++1xvcall16a8PsiGFOnTrXZ/9lT0yVOYShFDrtWrVrFPN69e7fNmzdvjvkauoTG/5m0bNmyuX7N3Xff7Rw/++yzNuvfdb/66iunX+/evW0eNmyY0xb1sv4///zTOV62bFnMvueff77N++339/Mk/l1l9XX6vPPOi/l67777bq7juPPOO51+uny4W7duMV8vXXiSBgAAAAAAIAS4SQMAAAAAABAC3KQBAAAAAAAIgbSuSfPyyy/bvHTpUqfNv41k2EyZMiVmW5EiRdI4knDx12/q41jrjYi4a5P4++V3W+946Nd++umnnbZ27dql7H0zyTPPPGOzv45Xb2PoX68mHv55pGvxJ0yYYPM111yT8GtnO/8W3HnNvzCI91oBIPn0Z66RI0farNdMEHGvxStXrnTa9BbfUaDXovvggw+cNr0WQp8+fZw2vUaN/hzxzTffOP30z1P9+v5tZfU2zHrNPv8220cddVQu/y8QFt99953N/fr1i9mvRIkSNl9xxRUpHVPUrFu3zjlesmSJzY0aNbI5r3UN87Pm4YgRI5zjTp062dyzZ0+b/WvSvPPOOzbrtVZERKZPn57wODLJAQcc4BxXq1bN5jVr1jhtffv2tbl79+42x1ojaF8qVqxo848//miz//d3/bOQNWkAAAAAAAAgItykAQAAAAAACIW0ljtlms8//9xmvX2e37333puO4UTKqFGjbP7oo4+cNn08ZMgQm6+++mqn36GHHprw++otRIsXL+60+R/zzlbFihWz2f99P2vWLJv95TWafjRcb4t37bXXOv3Gjx9vc/Xq1W2uUqVK/ANGxjvwwAOd4/zMbQD5o7fg9j9yr9v8PyOjVu6kVahQwTkeO3aszf7PfDNmzMg16/IpEZGjjz7a5qZNm9p8ySWXOP30NsK6FLRZs2ZOv9KlS8ccP9LPX3pz66232pzXttuDBg2yuUaNGskfWMToz6U33HCD06a3th83bpzNbdq0SemYdGnVnDlzbK5fv77T7/vvv7f5448/dtrefPNNm1u0aJHsIQauaNGizvHcuXNt3rVrl9NW0GvbTz/95Bxv2LDBZr3V+ZgxY5x++vefMOBJGgAAAAAAgBDgJg0AAAAAAEAIUO6k6PImEZGHHnrIZr3C/mmnneb0i+JjaXnROyTpRwsTocsZvvjiC6etdevWNvfv39/mt956y+k3bdo0m3W5hP57EZHBgwfbPH/+fJv9q+2feuqpcY09m+lHtHWOl145XcR9lPukk06y+bDDDkv4tRF+zz33XK5/f8899zjH/keEkX96nupyRT99DvznA9nDv9Nm48aNbV60aFG6hxMa+nPJueee67T5Pzv+xV/uFOu6pnekERHZvn17rv3at2+/z3EiOEOHDnWOX3vttVz7HXPMMc6xv2QHedO7pPl/B9FzRy9voEuQREQaNmyYotG5v4+8+OKLTpsui9q0aZPTNmzYMJuz4ffKUqVKpey1demYiFtu2LJlS5vr1KmTsjEkA0/SAAAAAAAAhAA3aQAAAAAAAEKAmzQAAAAAAAAhkNY1aSpXrmxzKmvRErF7926bH3zwQadNb9+mt2P09ytcOLuW9jniiCNsrlatmtO2YsUKm2fOnOm06S209fbX5cuXd/rpbdn0+jI1a9Z0+ul1gvTWoE8//bTTT7+XXodGr3eD1Fm+fHnMNl27G+XtXNPhvvvuc44XLFhg89q1a522yy+/3OZnn302tQNT9DjKli1r8zXXXJO2MQCI7dtvv3WOJ0+ebHOtWrXSPZxQKlKkiHNc0PXsVq1aFVc/1s0LH/17wsMPPxyzX8mSJW3Wc0pEZL/9+PfyROgt6/UamSIiffr0sdnzPJv173rptHDhQud4z549MfuGfX2UTKK33PY744wz0jiSguHKAAAAAAAAEALcpAEAAAAAAAiBtNbpNGvWzGZdMiMisnHjRpvXrVvntPm3hEyU/3Gz//73vzbr7Z91mY3f2LFjbT7llFMKNJ4oeeaZZ5xjvbXZ66+/7rSdc845Nt988802+8udtE8//dTmIUOGxGzTjzVWr17d6ae/7sILL4z5XkiNgQMHxmxr1aqVzWy7XDB169Z1jh944AGbu3fv7rS98sorNvfq1cvmZJ+Dq666yjn+9ddfbe7YsaPNRYsWTer7ZjP/Ntt5bbuN9NBlEIcddpjT1qVLl3QP5x90mfKdd97ptG3dutXm999/P21jyiYTJkwIeghIgJ4Huoxffw71Gz16tM3HH398SsaVjXr06OEcT58+3eb33nvP5m7dujn9mjZtavPtt99us38Jh3iNGDHCZr3kwtKlS51+eX2PID3233//oIcQN56kAQAAAAAACAFu0gAAAAAAAIRAaLYl0jsKNG/e3GnLqxwmHrosRuSf5VR/8T+GfP7559t80kknFWgMUaV3vRIRefPNN232r6D98ccf29yhQ4eYr6kfBzTGxDWOyy67zOb777/faTv00EPjeg0kz9dff23zxIkTY/Zr0aJFOoaTlf71r3/ZrHdDEBF58cUXbdaPbiej3Env6uY/9+XKlbP5rrvuKvB74Z8GDBgQ9BCynv/7Xu8+qMsjRJJf7qR3UJs0aVLMfrpNl337Pwc9//zzNteoUSMZQ4SIrFy50uaXXnopZr8mTZrYHJZdUbON3klUxC3T3rJlS8yv06XErVu3Tv7A8I858dprr9msd0v65ZdfnH66/Exf4/K709bOnTsT/pqTTz7ZOeYzEfx4kgYAAAAAACAEuEkDAAAAAAAQAtykAQAAAAAACIHA1qTxb6c8aNAgm3V9dCromkO9XoneFlrE3ZYN8dHrB33yySdO28svv2yz3pbuqaeecvpdccUVNudVH6r7USsfLvPnz7d506ZNNvvXGGLr5dQ55phjbB48eLDT9uGHH9qs1zDR61mI/PM6/ZclS5Y4x5999pnN+jrqr+W/5ZZbbK5Vq1bMsSMxepvteLfc1tuTirhbkiK59DprTz75pNP26quv2ty2bdtcv0ZEZNGiRTbrzy2TJ0+O+V76eut/vZo1a9rcuXNnm/v27ev0K1OmjCD59GegjRs3xuzXpk0bmwsXDs0ykpG3Z88em8eMGeO0xVqHpkGDBs7x8OHDbS5SpEgSR4dYSpYsafMPP/xgs/8cjhs3zuavvvrK5p9//jmp49FrA4q4a65eddVVThvrZybPRx99FLNt8eLFNjdu3Dgdw8k3nqQBAAAAAAAIAW7SAAAAAAAAhIDxPwLrk2djMulHzPzb8upH0fKjR48eznG9evVsvuaaawr02ikS377T8UvbeYQjmecxo86hfsz31ltvtbl27dpOvy+//DJtY8qnSM5FvR2lvgbq7bhFRI4++uhc+/m3ily3bl2u73P++ec7xw899JDNVapUSWDEBRbpuahLnM444wyn7e6777b5nnvuSdOIUiJj5+Jbb71ls788SdPbYq9Zs8Zp0+WBugRJly352y688MKY76VLhIsXLx6zXwpEei7GS5d/X3zxxU6bPh+6FKpQoUKpH1h8MnYuxkuXS/hLVmIZP368c9y+ffukjikFmIsisnr1aps3b97stOnlGHRJ8Lx585x+1apVs/nEE0+0uWLFik6/Aw44oEBjzUXk52J+nH322c7xu+++a/OECRNsbteuXdrGtA+5nkeepAEAAAAAAAgBbtIAAAAAAACEADdpAAAAAAAAQiA0a9LAQY1hNGRtva9e92nhwoU2Dxs2zOmnt2QOqcjPRb3mgd6aUERk0KBBNr/xxhs253XedI1v/fr1nbYAt5DN2rkY1EC9uQAAIABJREFUIZGfi1mCuSgiHTt2tFmvkSAicsopp9j88ccfp21MCYjkXNy0aZPNlStXtnnDhg0xv0Zv4Ttz5kynLQO2TGcuZr5IzsWC0usfiojMnj3b5hdffNHmNK/HlhfWpAEAAAAAAAgrbtIAAAAAAACEQOifxQOQefS2sLrcCeFz0EEH2XzyySc7bVOnTk33cAAg8vR2zca4T7rrcmGkj96mN94Sp5deesnmDChvArJC79698zzOFDxJAwAAAAAAEALcpAEAAAAAAAgBns0DkHTnnnuuzT/88IPNJ510UhDDAQAgNPaxsyoCcNxxx9l8+OGH21ytWjWn3wsvvGDzkUcemfqBAchKPEkDAAAAAAAQAtykAQAAAAAACAFu0gAAAAAAAISA2UddLEWzwTD77pIQzmMwknkeOYfBYC5GA3Mx8zEXo4G5mPmYi9HAXMx8zMVoyPU88iQNAAAAAABACHCTBgAAAAAAIAT2Ve4EAAAAAACANOBJGgAAAAAAgBDgJg0AAAAAAEAIcJMGAAAAAAAgBCJ3k8YYU9MYM9MYs9EYs9QYc2HQY0L+GGMuMsZ8a4zZaoz53hjTOOgxITGcw8xnjJlljPnTGLNl75/FQY8J+WOMOXbvuRwb9FiQOGNMaWPMpL3X0xXGmEuCHhPip66hf/3ZbYx5NOhxIX+4nmY2rqeZzxjTyxgzzxiz3RgzOujxJFvhoAeQTMaYwiLymoiMFJGzRaSJiEw1xtTzPG9JoINDQowxZ4vIMBHpJCKfiUj5YEeERHEOI6WX53lPBz0IFNjjIjI36EEg3x4XkR0iUk5E6orI68aYLz3P+1+ww0I8PM8r+Vc2xpQUkdUiMj64EaGAuJ5mNq6nme9nERksIs1FpFjAY0m6qD1JU0NEjhCRhz3P2+153kwR+VBEugY7LOTDABEZ6HneJ57n7fE87yfP834KelBICOcQCAljzEUi8ruIzAh6LEicMaaEiLQTkf6e523xPG+OiEwRPt9kqnYiskZEZgc9ECSO62lm43oaDZ7nTfQ8b7KIrA96LKkQtZs0uTEiUjvoQSB+xphCItJARA7bW7K2yhjzmDEmcndJo4pzGDlDjTHrjDEfGmOaBj0YJMYYU0pEBorIzUGPBflWTUR2+Z4K/lJEjgtoPCiY7iLynOd5XtADQWK4nkYC11OEXtRu0iyWnH+ZuNUYU8QYc47klDwVD3ZYSFA5ESkiIu1FpLHkPIZYT0T6BTkoJIRzGB23icgxInKkiIySnBLSKsEOCQkaJCLPeJ63KuiBIN9Kisgm399tFJEDAxgLCsAYU0lyPpuOCXosyBeup5mP6ylCL1I3aTzP2ykiF4hIS8mp9e0tIq+ICBfSzLJt7/8+6nneL57nrROR4SJyXoBjQmI4hxHhed6nnudt9jxvu+d5YySnhJTzmCGMMXVF5CwReTjosaBAtohIKd/flRKRzQGMBQXTVUTmeJ63LOiBIDFcTyOD6ylCL1ILB4uIeJ63UHL+hUJERIwxHwn/WpFRPM/bYIxZJSL6MWAeCc4gnMNI8ySnjBSZoamIVBaRlcYYkZx/QSxkjKnleV79AMeFxCwRkcLGmGM9z/tu79+dICIscpl5uonIfUEPAvnSVLieRgHXU4RepJ6kERExxtQxxhQ1xhQ3xtwiOTvKjA54WEjc/4nIdcaYssaYQ0TkJhGZFvCYkBjOYYYzxhxsjGm+95pa2BjTWUROF5E3gx4b4jZKRKpITslhXcnZ/fB1ydkNARnC87ytIjJRRAYaY0oYY/4lIm1E5PlgR4ZEGGMaSU7pKLs6ZSaupxHA9TQa9n4uLSoihSTnZmnRvTs9R0Jk/o8oXUXkSslZD2O2iJzted72YIeEfBgkImUk5273n5JTtnZvoCNCojiHma+I5GxvWENEdovIIhG5wLfYHkLM87w/ROSPv46NMVtE5E/P89YGNyrk07Ui8qzkrL23XkR6sl1sxukuIhM9z6OsIgNxPY0UrqeZr5+I3K2Ou0jOzrL3BDKaJDMsLA8AAAAAABC8yJU7AQAAAAAAZCJu0gAAAAAAAIQAN2kAAAAAAABCgJs0AAAAAAAAIcBNGgAAAAAAgBDY1xbcbP0UDJPk1+M8BiOZ55FzGAzmYjQwFzMfczEamIuZj7kYDczFzMdcjIZczyNP0gAAAAAAAIQAN2kAAAAAAABCgJs0AAAAAAAAIcBNGgAAAAAAgBDgJg0AAAAAAEAIcJMGAAAAAAAgBLhJAwAAAAAAEALcpAEAAAAAAAgBbtIAAAAAAACEQOGgBwDMnz/fOe7fv7/Nb7zxhs3Fixd3+n3wwQc2169fP0Wjy27Dhw93jnv37m3zRx99ZHPDhg3TNiYAAAAAiCqepAEAAAAAAAgBbtIAAAAAAACEgPE8L6/2PBuRMibJrxfq89iiRQvn+J133sm132GHHeYcn3XWWTaPHTs2+QMruGSex0DOYcWKFZ3jH3/80eYsKXfKqrkYYRk/F5Hdc3H37t02T5061eaHHnrI6Xf99dfbXKRIkbheu1mzZjaXKlUqv0OMF3Mx82X1XIwQ5mLmi/xcrFatms1Lly512jZv3mxziRIl0jamFMj1PPIkDQAAAAAAQAhwkwYAAAAAACAEuEkDAAAAAAAQAmzBjUDMnDnT5i+++CJmv1tvvdXmyy+/3Gn77bffkj8wOOvO6Czirn8Q4XVokCC9jtSrr75q8yuvvOL027BhQ8Kvvd9+7r8lfPrppzY3aNAg4deLIj0XH3jgAafttNNOS/dwkAI7duywuW3btjH7ffjhhwm/do8ePWweOXJkwl8PRNkxxxzjHP/73/+2uXfv3ukeDpTly5c7x0cffXRSX/+yyy6zuUKFCjbXqlXL6depUyebjUn2MjHZTf/39P+3nTRpks1dunRJ25jShSdpAAAAAAAAQoCbNAAAAAAAACGQFeVOK1ascI4feeQRm+fNm2fz448/7vSrXbt2ageWZdavX29zhw4dbP7999+dfueff77NgwcPtrlw4az4dg3c+PHjY7bpxz0RfS+//LLNetvfN954w+mn57Dn/b2D47HHHuv0u/LKK20+5ZRTbPZfa3VZ3VNPPeW06XKqbC530mWiX331lc2lS5cOYjiI09atW53jzz//3Ob999/f5lNPPTVtY3rmmWds9pcovv766zZT4opsoeflsmXLnDb/caLWrFnjHJ9wwgk26/KaIUOGFOh9slGyS41Gjx4dV79169bZfO2119rsL9dG4g455JCYbd27d7eZcicAAAAAAACkBDdpAAAAAAAAQiBS9SNLliyx+bHHHrP5ueeec/pt3Lgx169v0aKFczxt2jSb9S43lSpVcvrVqVMn8cFmoY8//thmf4mTdvvtt9tMiVP65VXuxOPu0dOnTx+bH330Uadt+/btNusypurVqzv9mjdvbvNNN91kc7169Zx+RYoUiWtMJ598ss3fffed06ZLILPJnj17nOPbbrvNZl0mc9hhh+Xr9e+44w6bdRlZu3bt8vV6yF3//v2d4//85z82H3jggTY/++yzTr82bdrY3LVrV5uff/75Ao9p9+7dNvt/Nus2IFv4d8nTCrqDkP9a/uuvv9qc126n+Cd/ea++vuqlLWL93pcs119/vc2tW7e2uWLFiil932ygP5sMHDjQadOl3qtXr7b58MMPT/3A0oAnaQAAAAAAAEKAmzQAAAAAAAAhwE0aAAAAAACAEMi4BT90Lec333zjtJ199tk269q0eP3000/OcZMmTWzetGmTzf51OebMmWMz263F9v7779us17e44IILnH7p3HoU7npLIiKffPKJzUcddZTT5j9G5hszZozNf/75p9PWsWNHm2+55Rab9ZahIu6aKMlwxhln2OzfVrFQoUJJfa9M4d8aedWqVTZ//fXXNud3TRq9XtC4ceNsZk2axPm32dbrJPz3v/+N+XWbN2+2edasWU6bPg8XXXSRzQsWLHD66e8F/XM2v6655ppcXzuK1q5da3Mi80if70mTJtk8ceJEp9/kyZNt1ufGv21wrDb/98Tpp58e9xixb3rb7ZkzZ8bsd8wxxxTofd59990CfT3+VqpUKed4wIABNp977rk2N2rUyOmnP7NceOGFNvvPzfr16xMe0xtvvGGzvn4if/R6bH76M6peryavn7OZhDsKAAAAAAAAIcBNGgAAAAAAgBDIiHIn/Qiq3iJ20KBBcX39wQcf7Bzr0iX/Vnix+mmLFi1yjvVrUO70tzVr1jjH06dPt1k/wtuzZ8+0jQn/9PDDD8dsa9++fUrfW2/L7i+70nQJVocOHWxmS/CC+9e//mWzflRfRKRly5Y2n3TSSWkbU5UqVdL2Xpni22+/dY67d+9u8xFHHFHg169WrZrNuiwDibv99tud48cffzzh15g3b55z/MMPP9isH+PXWUTk0ksvtfm5556z+cYbb4z5+rpk22/58uU2T5061Wk7//zzY35dJho6dKjNw4cPd9r051D/dVJvo7548WKby5Qp4/S7+uqrY7Zp9957r836s5J/XlLuVDC6vFDE/Wyhz7e+1oq45TH54d9mW5e38ZmmYHbu3GnzkCFDYvZr1qyZzS+99JLN/mUv9LVbf1bevn17zNceMWKEzRdffLHTdtBBB8X8OuzbypUrneNdu3bZ/MILL9hMuRMAAAAAAACShps0wP+3d+fxU87rH8evWymV0iZRWpxSEVKUpcWShPqdCKHtWNroRLZWLYrs29EpkSjHFhVCWSMHR4usydZeWo7Sqs39+yMu1+c+35lmvt+Zue+Z7+v5eHg83rfPPTOf0/jMTPe5r88FAAAAAEAEcJEGAAAAAAAgArJiT5pBgwZpfvTRR2OeZ1uq2ZrAmjVrOucNGzZMs93rIh7bjvGll15yxooWzYo/xoyz9fAibst02zavQoUKGZsT/pdt5RuU6nbodg8aEZEOHTpojrcnjWXrgj/88ENnjHruxHz77beaZ8yYoTn4WVnQ2nsUjN0PJFhjbb8XU23x4sWagzXg1apVS9vrZpvt27drtq1fH3nkkQI/93/+8x/n+IcfftAcrwWw/e3Tu3dvzccdd5xznt0z7oILLtA8Z84c57xt27ZpDu6Jkgt70nTu3FlzqVKlNAdb57733nuaf/75Z2esYcOGmu3eP927d8/XnOweJXZfjeD+cZ06dcpzDkhM8O8T9vPW7gWU6r353n33XefYvlZB23sXNnYPGhF3r9Lp06drrl69unNerL9LVqlSxTm268/u4xWvjbrdl8p+R4iwJ01BPf7442FPIaO4kwYAAAAAACACuEgDAAAAAAAQAZGp07FtrIO3FtryItvi+thjj3XOe+yxxzS/+eabmoOtJ4MttBNhbyWlpCIxtrwpyJZVcJtu4RG8XTtWidO9997rHNs1Z0ukrr/+eue8YDkV8jZ27FjN9nbc1q1bO+eVLl06Y3PC/3rhhRc0H3jggc5Yx44dU/pa9nt2z549moPlkJQ7/WnmzJma77rrrpQ+9zHHHOMcV61aNaHH2dvpTzjhhJjn2dv6betS24o9yN7GL+KWxQVLJaNqypQpzrEt4dq6datmW4IiIjJw4EDN3bp1c8ZSvSbsa9n5BX9T2Vbg/I5KzC+//KI53pq135Ft2rRJ65ysTL5WLgi2zB45cmSe5wXXbLCsKRH9+vXTPHv2bGcsVkvu4OfN1VdfnfTr4k/BEvzPP/88pJlkBnfSAAAAAAAARAAXaQAAAAAAACIgMuVODz30kGZ7C2dQ3bp1Nffv398Za9q0qeZff/21wHOyt/2moltDYfP666/HHOvVq1cGZ4Iw2RKkyZMnxzzvueee03zxxRfHPM92nIr3fIgt2HHgD/FKHZB+wU5KtlPFTTfd5IzZjoP5ESz7teVO1157reZTTjmlQK+Ty3bu3JnS56tYsaLmZ5991hmrV69eSl/LSvR/R7Cb3hdffKE5W8qdgiVD9jelLYmwHa9E3Pcm3UqWLKm5RIkSmm3XJxGR9evXZ2xOucKuI9vhTMQtD2zfvn1KX/fHH3/UbMsEg8qVK5fS18118X4DNmnSRPPll19e4Nc688wzNdu/b4qIvP3223k+Jt62D0ieXUeFAXfSAAAAAAAARAAXaQAAAAAAACKAizQAAAAAAAARENqeNLt27XKO77zzzoQeZ+voL7nkkpjnlS9fXvPf//53Z+ytt97S/O9//zvmc1xxxRWaq1evntD8EJutp3755Zc1f//99855tobztddey/PxIm6LTPv+DBkyxDmvS5cumosUKZLstFFAwbbbVqL70CTq+eefT+nz5apXX31Vs22z3a5duzCmU6jZdtePP/64M2Y/81LduvPQQw91jitUqKC5aNHIbFcXaR06dNAcbNmcHy1bttSczj1oguz/jlw3ePDguMdRY/87mDdvXogzyR62HXLnzp2dsdWrV8d83GOPPabZfh6mwubNmzXbNuBInm27/PDDDztjBxxwgGbbjjv4fVdQV155pXMca08apNYbb7wRc8yuseA+sj169EjbnNKJO2kAAAAAAAAigIs0AAAAAAAAERDaPc377edeH7LtG3/66aeYj7PtCIsXL+6M9e7dW/P111+vefny5c55sUqrbGtfEdpEp5q9Hdy2547Xqts+5uijj3bGbFmUbV171VVXOefZNpXBNrYQqVq1asyxFStW5Os57ZqzLRKDa4ySpMzYsmWLc7xp0ybNderUiXnel19+mdDz16pVS7O93Rj7Zm/RHT58uDN23nnnabYlvKlw0EEHOcenn356Sp8fiTn//PM1B2/dB2yr30mTJoU4k+yxYcMGzS+88ELM84Ilirbc8Oyzz9Z84403xnyOgw8+WPNhhx0W87wXX3wx5usiOWPHjtUc/PvdWWedpdm2zEZueOmll5zjW265RfPMmTM1221NRCh3AgAAAAAAQAFwkQYAAAAAACACQit3CnbZsV18pk+f7ozZThMNGjTQXLdu3ZjPb2/bHzZsmDP266+/aradTZ588knnvDJlysR8fhSM/XM/+eSTnTHbjalixYqamzdv7pz3/vvvax43bpzmKVOmOOf1799fc40aNTRfdNFFSc46N/Xt29c5tt2YbrjhBmfMlhHGE3zcH1LxZ25vbz388MOdMcqn8vbhhx86x7a7xNy5czUfc8wx+Xr+448/XvOAAQM0t23b1jmPUqj/Zct2W7Ro4Yx98sknmvv16+eM2RIzWzJjPzOTceKJJ2q2nRF69uyZ0OODazFY0oy96tev7xzbDnep7qplS4JtKYeIWxKyc+fOhJ6va9euzrEtx0P6USaTGPv3i+C2CPb3f5Atjf/Xv/6l+amnnnLOs+9DpUqVNJ9yyinOeT/++KPmeNs4BL8n4Qp2w7LdYYM6deqU7ukgRPZ3ioj7m8l2fvroo48yNqd04lcUAAAAAABABHCRBgAAAAAAIAK4SAMAAAAAABABoe1JE1S2bFnNqagptO3upk6dGvO8Dh06aD7yyCML/Lr409/+9jfn2LY+t3/udv+DZNhWe7a1c7Bt8Lfffqt56dKl+XqtXBbcS8L+WX788cfO2H333ac50f1prHjtvuN5/vnn85zTvffem6/nK2xq1qzpHNv2orYV81/+8peYz2Hbsc+fP98Z+/TTTzXbfYEuvfRS57zHH39cM/vT7FWiRAnNM2bMcMbsd9c333zjjE2cOFHzXXfdpTnen6vv+5qD+1ts3LhRs32v7T5eIu7+Q/a9Du5tFdwLAnsF/9wLug/N+PHjneN58+Zp/uCDDzTH2xMjngMPPFCz3S9O5H/3FkTqzZ49W7NdvyIip556aqankxVsW+xg2/JXXnlFc/B7LNZ+Nd9//33M11q7dq3madOmOWOxPm9LlizpnNe7d++Yzw+R3bt3O8erVq0KaSZAZnEnDQAAAAAAQARwkQYAAAAAACACIlPulAo///yz5nhlENWqVdM8evTotM6pMKtQoULMMdtaNhVsS++mTZs6Y7bcCftmyxZsWZqI21rblh0FSx0mT56c53MH263HYsubRERuvPFGzbY8izbqialdu7ZzHCyrSdb27dudY9vq99Zbb9X8zDPPOOfVq1dP8y233FKgOeSiYKlSsFzMGj58uOZt27Zp3rVrl3OevR1/zpw5MZ/v4Ycf1mxbni5YsMA5z66//fffP+bzIW/BtRPvPbHs+23fE1umJuL+t5AKthXx6aefntLnxr7ZMsdgqdxRRx2V6elknQsvvDDucSKCJVO2vNeypaAibmmV/U1kf6+KuKX7yC116tQJewrIYtxJAwAAAAAAEAFcpAEAAAAAAIiAnCp3atOmjeYvvvgi5nlDhgzRXKxYsbTOqTAL7mBvd7q3u7Xv2LHDOS8/XUHs7acvv/xyzNfFvtmOLbbLi4jIAw88oNnevhurvCnoo48+inn8wgsvJPR8H374oeZgZypkhu1IJCJSuXJlzUuWLIn5ONt1A6kT/Ky1bPeuYNmbNX36dM22m0n58uWd8yhx+pPtRGnLguIJdopp0qRJSudUUM2aNXOOGzRoENJMENSoUSPnuGHDhiHNpHDp3Llz3ONYbHc1W6pWrly51EwM/2PhwoVhT8HpkpnofytAXriTBgAAAAAAIAK4SAMAAAAAABABXKQBAAAAAACIgKzek+bHH390jr/88ss8z7N71YiI/O1vf0vXlGBcffXVzrFtNTpx4kTNf//7353zHnzwQc3BvS+sZcuWab7mmms0r1+/3jnP1gKzJ0Zyrr/+eufYtry+//7788zxBFt6x3LSSSc5x7YlN/vQhO+9995zjvv06aP5888/1xzc36Jjx47pnRhS4rLLLtNctmzZEGcSbe3bt9ec6J40UVGxYkXN9erV0/zcc88559n9ppAZ69at02x/z3Tr1i2M6SCfNmzYkOe/b9u2bYZnUnhMmDBBc/fu3TXXqFEjpa8T6++bIiKnn366Zr4/02vSpEl5/vvgXqd2f82qVaumdU6pxJ00AAAAAAAAEcBFGgAAAAAAgAjIunKnlStXaj7zzDOdsc2bN2uuVq2a5tGjRzvnFSlSJE2zQzz33Xef5rfeekvz+PHjYz7mwgsv1Lx161ZnzJZJrV69WvNhhx3mnHf55Zdr7tq1axIzRpAtNbLvp80i7vpbvnx5zOfr27evZlviZNuAI7127typuVixYs7Ytm3bNA8fPlxz8DPVrk37Pgb/uyhdunTBJou0OffcczU//vjjIc4ke9SqVUtz/fr1Nce7FT4swbIle5t48LcUwjVv3jzNS5cu1WxL1BB99n1EZqxZs0ZzkyZNNNvfmiIi/fv31/zaa69pXrVqVcznfuyxxzQvWLDAGatdu7bmAQMGJDFjFESs35Q///yzc2z/zplNW55wJw0AAAAAAEAEcJEGAAAAAAAgArKu3Gn+/PmalyxZEvO8K664QrMtvUB4ypUrp/mll17S/Ne//tU5z5Y/2ez7vnOe7drUsmVLzaNGjXLOa9iwYT5njPyyZWrxOj/ZHddfeOEFzcEOTieffHIKZ1f42FJQEff23h9++EGzLRsUEZk+fbpm+3l7wAEHOOcNGzZMs+0IRnlTdrLlOsEuikcccUSmpxNZtsTJfn6dccYZznnxbqFPNVuyaNdfsPuU7UCCaLFl2fZ3DqLtgw8+cI6Dn51/4PdMcuzfHUTc9fHkk0/GfJztkjZ06FBnzHaR3bhxo+ZgV6BE9e7dW3PNmjXz9RxIXrt27TTbDsK5gjtpAAAAAAAAIoCLNAAAAAAAABHARRoAAAAAAIAIyIo9aT755BPNXbp0iXle8eLFNdt2oogeu0/MK6+84ozdcsstmu3eGaeddppz3jnnnKP52muv1RxsI4zMC7Ze/oPdtyHItm6uWrVqyueU6/bs2eMc33vvvZoHDx7sjNWtW1fzt99+qzlYj73ffn9ex7fvzz//+U/nvOOPPz4fM0aUtG7dWnPnzp01r1y50jmPPWnyduSRR2p+++23nbGLLrpIc6rbcwe/F9u0aaPZ7g+F7LF27VrNlSpV0ty9e/cwpoMELV68OOaY3VuIz9Dk2N8hIu5nXrw9aaxdu3Y5x7ZVd37ccccdzvHVV19doOcD8sKdNAAAAAAAABHARRoAAAAAAIAI8IJtjQPiDqbL1q1bnWPbznfGjBkxH1e5cmXN77zzjuZ69eqlcHYZkeqei6G8j0jp+8h7GI6sWYs333yzc3z33Xcn9Lj9999fc7Bsafjw4ZptOUwWYi3uw+7duzXbctTg97Ft2Z5hWbMWg2yJ06xZs5yxPn365PmY8847zznu1q1bnuc1atTIOa5SpUo+ZphRrMWAhQsXOsdHH3205h49emgeM2ZMxua0D1m7FtNp3rx5zvEJJ5yg2ZYYz58/3zmvRIkS6Z1YbFm5Fn/77TfNixYtcsbs2km1UaNGab7pppucsWBJVgYV6rVov0/tVhmff/65c967776r2f6+iZA830fupAEAAAAAAIgALtIAAAAAAABEABdpAAAAAAAAIiCSLbjHjRvnHMfah8buQSMi8vrrr2vOwn1oACDfTj31VOd4+fLlmpcuXeqM2Zb17dq101y8ePE0zQ5RV7Tonz8HbKvff//732FMJ6fUr18/zywi0rt370xPBxHz4osvOsd2r8hYexEheoL7Q5111lma7fdsiHvQ5AS7/8uhhx7qjM2ZM0fzd999p9nuryfi7r935plnau7SpUvM17XfkbalOsJj27HPnj07vImkCXfSAAAAAAAARAAXaQAAAAAAACIgki24H3roIed46NChmvv27as5eBto8La3LFaoW6rlkKxsbwgHazE3sBazH2sxN7AWRWTdunWaGzdu7Ixt27ZNsy3fqFatWvonlhjWYm5gLWY/1mJuoAU3AAAAAABAVHGRBgAAAAAAIAIiWe4Ebl/LEdxKmv1Yi7mBtZj9WIu5gbUoIvPmzdNCfgMqAAAgAElEQVR84oknOmNnn322Ztu1NEJYi7mBtZj9WIu5gXInAAAAAACAqOIiDQAAAAAAQARwkQYAAAAAACACioY9AQAAAKAw8jx3O4JJkyaFNBMAQFRwJw0AAAAAAEAEcJEGAAAAAAAgAvbVghsAAAAAAAAZwJ00AAAAAAAAEcBFGgAAAAAAgAjgIg0AAAAAAEAE5NxFGs/znvI8b7XneZs8z/vW87yrwp4TkuN5Xm/P8+Z6nrfD87wnwp4P8oe1mDs8z6vted6vnuc9FfZckBzP87YE/tnjed4/wp4XkuN5XnHP88Z7nrfU87zNnuct8DzvnLDnhcSxFnMDv1GzH5+nuSHX12LRsCeQBqNE5Erf93d4nldXRGZ5nvep7/vzwp4YErZKREaKyNkiUiLkuSD/WIu5Y7SIzAl7Ekie7/sH/pE9zztQRH4SkcnhzQj5VFRElotICxFZJiLnisjznucd4/v+kjAnhsSwFnMGv1GzH5+nuSGn12LO3Unj+/5Xvu/v+OPw93/+EuKUkCTf96f4vj9NRP4b9lyQf6zF3OB53iUislFE3g57Liiw9iKyVkRmhz0RJMf3/a2+7w/zfX+J7/u/+b4/XUQWi0ijsOeGfGEtZil+o2Y/Pk9zQ66vxZy7SCMi4nnePz3P2yYi34jIahF5LeQpAYUSazG7eZ5XRkRuFZHrw54LUqKriEz0fd8PeyIoGM/zDhGRI0Xkq7DngnxhLQIRwecpoignL9L4vn+1iJQWkWYiMkVEdsR/BIB0YC1mvREiMt73/RVhTwQF43leddl7a/eTYc8FBeN53v4i8i8RedL3/W/Cng+Sw1oEooPPU0RVTl6kERHxfX+P7/sfiEhVEekV9nyAwoq1mJ08z2sgIi1F5P6w54KU6CwiH/i+vzjsiSD/PM/bT0QmichOEekd8nSQP6xFIAL4PEWU5eLGwUFFhX0wgChgLWaX00Skhogs8zxPRORAESnied5Rvu83DHFeyJ8uInJH2JNA/nl7F+J4ETlERM71fX9XyFNC/rAWgZDxeYqoy6k7aTzPq+R53iWe5x3oeV4Rz/POFpFLhQ0vs4rneUU9zztARIrI3r8UHuB5XmG4oJgzWIs5YZzsvajW4Pd/xorIq7J3F31kEc/zThGRKkInmWw3RkTqiUhb3/e3hz0ZJI+1mP34jZoz+DzNcrm+FnPqIo3s7R7TS0RWiMgGEblHRK7zff/lUGeFZA0Wke0i0l9EOv2eB4c6IySLtZjlfN/f5vv+T3/8IyJbRORX3/fXhT03JK2riEzxfX9z2BNB/vy+j0kP2XvB9CfP87b8/k/HkKeG5LAWsx+/UbMcn6c5I6fXosfG8gAAAAAAAOHLtTtpAAAAAAAAshIXaQAAAAAAACKAizQAAAAAAAARwEUaAAAAAACACOAiDQAAAAAAQATsq5c4rZ/C4aX4+Xgfw5HK95H3MBysxdzAWsx+rMXcwFrMfqzF3MBazH6sxdyQ5/vInTQAAAAAAAARwEUaAAAAAACACOAiDQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQAVykAQAAAAAAiAAu0gAAAAAAAEQAF2kAAAAAAAAigIs0AAAAAAAAEcBFGgAAAAAAgAjgIg0AAAAAAEAEFA17AgCAzNq4caPmMmXKaN5vP67bA1Hw/fffO8fPPPOM5nfeeUfzypUrnfO+++67hJ7frnv7fI0aNUpqngCQbebPn6/5jjvu0Dx58mTnvNmzZ2tu2rRp+ieGfLPfY1deeaUz9v7772s+/PDDMzanguIXOQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQAZ7v+/HG4w4ibbwUP1+k38cTTzzROS5ZsqTmSZMmaa5WrVrG5pQiqXwfM/Yejh8/XvNVV12Vr+eoXbt2ns/Rtm1b57x69erl6/kzKCfX4llnnaW5VKlSmoPvd5s2bTI2p1jWrl3rHJcvX15z0aIJb6uWlWsRjpxci5bdD+bLL790xnbt2pW217Xr6KCDDnLG1q1bl+qXYy1mv5xfi5s3b9Yc3KepRIkSmv/xj39obt68efonllo5vRbtvl7dunVzxv7zn/9o3r59e8znaNeuneapU6emcHYpk/NrMZ6lS5dqPu200zQvWbLEOW/YsGGab775Zs12LYcsz/eRO2kAAAAAAAAigIs0AAAAAAAAERBaC+7gLbRPPfWU5mnTpjljtnWW5/15R1CwVCvW2AUXXOCc17Fjx5hjCJ9teffoo49qHjFiRBjTyXnBEqQ333yzwM9p28D269dP808//eScd9999xX4tZC8hg0bar7rrrs0t2jRIozpxPXAAw84x7bs4+677870dNKqatWqznG5cuU0Dxw40Bm79NJLU/ra8+bN0/zGG2/EPG/MmDGaV6xYoblSpUrOeW+99Zbm+vXrp2KKOeHZZ591jnv06KF527Ztmvfs2eOcZ9uGtm/fXnPnzp2d82ypaTy2lPiaa67R/N///tc578EHH9R87bXXJvTcQLYrXry45uDn8nvvvaf5tttu05yF5U5ZL/g5+fbbb2u+8MILNdvyNRGRihUrai5durTmYHn1jh07UjJPpMcPP/ygOVjiZNlyp88++0zzlClT0jGtlOFOGgAAAAAAgAjgIg0AAAAAAEAEhFbu1KVLF+d45syZmm3ZUvA4OBbrPCtYPmVv5T7qqKM0161bN86MkS7BXdfnzp2ref369ZmeTqEwa9Yszfb2UBH39s5jjz3WGatevXqezzdo0CDn2HYmsR2DHn74Yec8W3bTqVOnfcwaqWJLJ6LIltwFS+Lsf5+5Vu70zDPPOMe2HPdvf/ubM9arV68CvVawXNiWkSV6i7f9zg2WMH/66aeaC3u505NPPql5yJAhzpi9Db9Dhw6aGzdu7Jx35ZVXai5TpkyB52TL5f75z39q/uqrr5zzdu7cWeDXyha21PqJJ55wxmwXkFNPPdUZO+CAAzRPnz5d8+OPP56vedSpUyfP1wp2IrFlalnQKTGrFCtWTPPBBx8c87xly5ZpDq4V+xxInTVr1mju2rWrM2b/LnnggQdqfuyxx5zzWrdurfnFF1/UTElndrnnnnvCnkJacScNAAAAAABABHCRBgAAAAAAIAK4SAMAAAAAABABoe1JE9xrxNbHB+s/7b4V559/vmbbnjlo6dKlMV9ry5Ytmm17SdtaFOGJt+8QUuOYY47R3K5dO2ds06ZNmoNr7NBDD03o+e1zWHbfCxGRDRs2JPR8SC27B0UUvfvuu5qD+6PY74Nc06xZM+d48uTJmu+8805nLF6b7EQE96Thczd1nnrqKef4uuuu0/zLL784Y7ZNrN1jKdj2N9XKli2r2f53F9yTJtetWrVKc58+fTQvWLDAOc+uj3ifn3Z/muBvWbvXzLfffhvzOez34tNPP605+Flo26gvWrRIc+XKlWM+N1LL/rl/9NFHzliLFi0yPZ2cFPw73DnnnKN54cKFztj48eM1231nDjvssHy9do0aNfL1OESX/e6LOu6kAQAAAAAAiAAu0gAAAAAAAERAaOVOAwcOdI7trdvBlsyxbm/v3r27c2xve+vZs6fmDz74IOY8aLsdvuAt/vY2/EceeUQz5WipU6FCBc32dup023///Z3jVLSSxb7Zlugi7i3+UfTWW2/FHBs6dGgGZxKu0047TfOJJ57ojK1YsSLPxwRLDb/++mvNJ510UtJzaNOmjXP8448/5nlekyZNnONq1aol/VrZzLZb7tGjhzNWs2ZNzbaUT0Tk6KOP1hz8fET63XjjjZo/++yzfD2Hbatuy4cbNGiQ/4n9zpZd/d///Z8zZj8D3nvvPc22lTuQ7YLlTvbvfrZcVESkYsWKKX3tm266KaXPh3DYduz2Mz/quJMGAAAAAAAgArhIAwAAAAAAEAFcpAEAAAAAAIiA0Paksa208zpOxNatW53j9u3ba7b70wRbizZq1Ejztddem/TrIrXq1avnHNv3y+YpU6Y4511wwQXpnRgStnnzZufYtg62rr76aue4a9euaZsT/hRsDRpsA/yHUqVKZWI6/yPYWta2ai9RooQz1rx584zMKWqC741t5xtPcC+bREyfPl3z6tWrY55Xv359za+++qozVq5cuaRfN5vZvYD27NnjjNnPvVTsU5IK27dv12z3GbK1+yIiV155ZcbmlC67d+/WfPHFFztjr7zySp6PCa63/v37a+7Vq5czZv9b32+/1P5/n/a/l3ifzy1btkzp6+JPTZs2dY7t7xu7h2Jw30RacKdGcO/QdO4levDBBzvHdj8xhC+4J96nn36a0OPs/n5HHXVUKqeUVtxJAwAAAAAAEAFcpAEAAAAAAIiA0Mqd8mvkyJGag62DFy1apDlWyYzI/7b/RrTY9nrjxo3TPGrUKOc8yp3CZUtoWrdu7YwF2wD/IdimF+mzZcsWzffee2/M82ypqV17mfTSSy85x7btbHBOZcuWzcicCrPPP/9csy2LCSpZsqTmwlbeFNSpUyfNZ555pjOWztvz8+vDDz/U/MYbb2guU6aMc1758uUzNqd0GTZsmOZp06bFPK9t27aa7W9NEZFjjjkm5fNKhF2Ly5cvD2UOhV3wvQ/+nQLZZ8mSJZptmVqwpTei5dFHH3WOf/rpp4QeV7ly5XRMJ+24kwYAAAAAACACuEgDAAAAAAAQAVlR7jRv3jzNQ4YM0Wx3VRdxb0G0Yz169HDOC+7UjujittJw7dy50zm2t4X269dPc7A7j1WlShXNUelsUhj07dtXsy0FDRo6dGgmphPX448/HvYUCr0nn3xS8+23357QYypVqpSu6WSdihUr5pmjYuXKlc7xpZdeqrlYsWKag2XFuaBo0T9/6p566qnOmO3UdNlll2VsTokaNGiQ5mBHU/vbtkKFChmbU2Fj//sJHttOhF9//bVznn2/wuqciLzZrRQOOuggzYl+9yFztm3bpjnYqTRRV111Vaqmk1HcSQMAAAAAABABXKQBAAAAAACIAC7SAAAAAAAAREBW7ElTr149zUcddZTmYP1nrP1Lpk6d6hzPmDFDc7du3TQH22TS4jkcds8gWzdqWwqLuHWKthUsCmbp0qWaTz75ZGds9erVST+f3Qvh3HPPdcbs/geXXHJJ0s+NP7388svO8eTJk2OeW6NGDc116tRJ15Ti+uWXXzSvWbMmlDkUZps3b3aO+/Tpozle2227h1FYLduRmC+//FLzfffd54ytX79e88CBAzXbPVpyhW3BbXNU/frrr5qnT5+uObg3Srt27TI2p8IsuI/lkUceqfmrr77KM4u4v1HZkyZc9vNORGTChAmaO3TooLls2bIZmxMS8+OPP2p+7733QpxJ5nEnDQAAAAAAQARwkQYAAAAAACACsqLcyZay2Nt3R44c6Zw3fvx4zUuWLNG8du1a5zzbnnvw4MF5/nsRtyzq7LPPTnLWyC9bZnbHHXdoDpa3ffPNN5obNmyY/okVErt379acn/KmeOy6FHFbnt55552a7a2oIrTujmXTpk2aR4wY4YzZcqIgWwJ6wAEHpH5iCVi8eLHmBQsWxDzvyiuvzMR0CgVbRhH8TguWk/5h//33d47btGmjuXLlyimcHVLBtv21JWyzZs1yzmvUqJHmXCxxymZXXHFFnv/+oosuco5bt26diekAWe/WW291jm25L+so95QvX945LlOmTEgzKRjupAEAAAAAAIgALtIAAAAAAABEQFaUO8ViS5VERHr27Kl52bJlmoPdnV588UXNixYtivn8tvMM5U6ZY8vbSpQooTlYjvb+++9rptwpdYoXL6450TKjW265xTmOdWvh7bff7hy/++67mm3JS9u2bZ3zpk2bptnepl/Y2VLOuXPnxjzv/PPPd46POeaYtM0p1SpUqBD2FLKaLXFq2bKl5o8//tg5L1Z3xIceesg55rM2Wmx5k0jsEif7uS4iMmTIEM1VqlRJz+SQkO+++845njJlSp7n9e/fPxPTwT7Y36LB36WIjo0bN2r+z3/+44xdd911mil3yj3BvyfYLtHZhDtpAAAAAAAAIoCLNAAAAAAAABHARRoAAAAAAIAIyOo9aYIqVqyYZw7W0Nua7RtuuEHzpEmTnPPsnifjxo3T3L1794JPFgmxdYTz5s1zxuLtJ4T8q1q1quZPP/00pc9t9xsScfe7eO655zSvWLHCOc+2ZX/77bc116pVK6XzywYzZ87UPHDgwJjn1a5dW/Po0aOdsSJFimi2NfXB/S1iCbZl3rVrV57nlSpVyjmOte9JkN1D54gjjkjoMdjLthYVcfdTs/vQxNtLoWvXrpr5voueWG22RUQmTJig+YQTTtAc3A/M7k+EcH3zzTfO8c6dOzVfeOGFmu1nOsJjv8diZYSvd+/emlevXu2MdevWLSNzCP6mevDBBzW/8MILzpj97D7uuOPSO7Ecd9VVV4U9hZTgThoAAAAAAIAI4CINAAAAAABABORUuVN+LFy4UHPwVkV7nK3tu7Jd06ZNNQfL0ZB9TjnlFOe4SZMmmrdv36755Zdfds5btmyZZtt2ujCWO7300kua58+fH/O8HTt2aL7ttttinrdnzx7NY8eOTWgOxx9/vHMcqyzuX//6l3NsW6u/8cYbMZ+/bNmymrmFfN82bNigOVjaZluPxvuztGMnnXRSCmeHVIhV4mRvkRdxy1VHjBihmfKmaPnvf/+ruUuXLs6YLUWcPHlyxuYEZDv7+8j+/hg6dKhzXvXq1VP6ur/88ovmN998U/PgwYOd83788UfNV199tTP2l7/8JaVzKmxOPPFEzeedd16IM0kd7qQBAAAAAACIAC7SAAAAAAAAREChLHeyu2vbjkHBbhcTJ07U3KxZs/RPDHEFb9W3pWrITrbLkL1N1XZzEhGZOnWq5o4dO2oOlswUhu4XL774YkLn2RKxYAlMQW3ZssU5rlatmuYyZcpotu+ViNttZt26dTGfv1evXgWdYqHy+uuvaw7e1p0o22GGjlqpZdeiLS8MevjhhzXbsk4RkTVr1mh+6623NNvypuDYkUcemfxkkTa//fabZrtmbamESP5K7WvWrKn5tddey+8UkYD69etr/uqrr0KcCWxZt4jIsGHDNFepUkVz8LdIfqxfv17zPffc44w98sgjmjdu3JjnHETcUqgWLVoUeE6FwcEHH6y5bt26zpj93fLFF19ofvbZZ53zrrzyyjTNLr24kwYAAAAAACACuEgDAAAAAAAQAVykAQAAAAAAiICc3ZPG7ncwatQoZ+yBBx7QbGt/bd2biEjz5s3TNDvkR3DPoPfffz+kmSDdWrVq5RzbPWmWLFmiedGiRc55hWFPmltuuUXz3//+95jn2RaThxxyiDNWqlSpPB9zxhlnOMe2paHVuHFj53jbtm2aK1asqDm4Rv/xj39onjt3ruYGDRo457GXxr7Zttv2zzUeu5dCsEbbtv/cbz/+/5tk2X2a7L4DIiKdO3fWbNdKKqxYscI57t27t+Ybb7xR83HHHeecF/xM+ENwL5wBAwZobteunTPWtm3b5CZbyNg22yLue/Pcc88l9Bx2z4XgvnxWvDGk1pdffhn2FPA7uweNiMiCBQs0v/3225oT/W04Z84c57hfv36a33333ZiPO/nkkzWff/75mm+66aaEXhex7d69W/Ovv/4a8zw7Fvx8ZU8aAAAAAAAA5BsXaQAAAAAAACIgp8qdbEtm28I3WBJhy2ZsO6+vv/46jbNDQQVv5+X23txi1+/IkSNDnEm09ezZU3OTJk1innfooYdqLl++vDNWsmTJlM6pXLlyef77s846yzm2bdYte6uwiMhBBx2UmonlkJ9//tk5vuKKKzR/8sknMR+3//77a7alK5dcckkKZ1c4TZs2TfO9996r+d///neBn7tYsWLOsW1t37JlS81TpkxxzrMtuG22jxcRueyyyzTbsqWZM2c659mSx8MOOyyhuWOvc8891zkOllLEYj837Xvdvn175zz7OVmkSJH8TBH5YP8OESzDR/rZ7SyeeOIJZ+ycc87RbMu3bZm8iFuS+uKLL2oOljTZ30pt2rTRHFyLnTp10ly0aE791Tp0BxxwgObC9tuQO2kAAAAAAAAigIs0AAAAAAAAEZDV92TZjgki7q3HW7du1Rwsi7E7b0+aNClNs0MqNGvWTDO3lWYn2/Vk3rx5zpgtf5k8ebLmlStXxny+smXLaq5QoUIqpphV7K20sbovRdXQoUM19+3bV7MtzcKfbAenp59+2hl75ZVXEnqOa665RjMlTgWzY8cO59j+ee7cubPAz3/ggQdqvvPOO52xXr165fmYhx9+2DkeOHCgZvvZu2nTJue8sWPHarYlAxdffLFzni2vRHJs2b2I24nmjjvu0PzFF184540ePVpzrVq10jM55Jv9O4TdJoES/Myw6+Onn35yxmwZkl1v9vNORGTNmjWa7W+qYIdL202zadOm+ZswCsR2dwp+B+c67qQBAAAAAACIAC7SAAAAAAAARAAXaQAAAAAAACIgMnvS2D1kgnvNWHbfmeAeJbYetFq1aprvv/9+5zxbT4poq1evnmZacEeLbZkdXIsPPPCA5m+++Ubz7Nmz8/VatWvX1mzbcwdbNyPaDj744Dwz8ta9e3fNwVbLsVSsWNE57t27d0rnVJjNnz/fOf7tt9+Sfo7ixYs7x9ddd53m66+/XnOi6yP4/nbo0EGz3WvmnXfeifkco0aN0tygQYOEXhf71q9fP+fY/n59//33NdetW9c5j31ooo091DLL7kkiIjJx4sSY51511VUJPWeLFi002328WrVqleTskG727xf5+c7NZtxJAwAAAAAAEAFcpAEAAAAAAIiAyJQ72dttbVteEfdWp3glLrb12pgxYzQHb/9GdmrXrp1zHOvW4ebNm2dsTrloz549mletWuWMDRkyRLO95TQVtyBWqlRJ84gRI5yxSy+9VHPp0qUL/FpAFG3bts05/vzzz5N+jk6dOjnHNWvWLNCc8KdgeeXw4cM128/NuXPnOufZst1gCUzZsmVTOUWnTOqmm27KMyMc3377rWb7W/a4444LYzpAVpg3b55zvHjx4pjn2hbaF154oebg92CzZs00lypVqqBTRBpVrlxZc48ePZyxG264Ic/H2Pc3m3EnDQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQAZHZk2bdunWag+18S5YsqdnWdg8YMMA574ILLkjT7BAFgwYNco7t3kW2zTN70hTM5s2bNT/33HPO2IoVKzQnug9NlSpVNNt2syIiRYoU0dy3b9+k5gnkmuB+bPFq763q1atrtm27kV7B3yBAfjRq1CjsKSAJl1xyieZHHnlEs/2tIyJSrly5jM0plzVp0sQ5Dv4dEYXH9ddfH/c413AnDQAAAAAAQARwkQYAAAAAACACIlPuZNtlLVq0yBlr1aqV5oEDB2ZsToiW4C3BtuUpUse2hL3xxhudseAxgNSxreZF3Fb08T7v7rrrLs116tRJ/cQAACIiUr58ec2fffZZiDMBkMu4kwYAAAAAACACuEgDAAAAAAAQAd4+dslmC+1weCl+Pt7HcKTyfeQ9DAdrMTewFrMfazE3FNq1OGPGDM2jRo3SPGHCBOe8I444ImNzyifWYm4otGsxh7AWc0Oe7yN30gAAAAAAAEQAF2kAAAAAAAAigIs0AAAAAAAAEcCeNNFEjWFuoN43+7EWcwNrMfuxFnMDazH7sRZzA2sx+7EWcwN70gAAAAAAAEQVF2kAAAAAAAAiYF/lTgAAAAAAAMgA7qQBAAAAAACIAC7SAAAAAAAARAAXaQAAAAAAACIg5y7SeJ7X2/O8uZ7n7fA874mw54PkeZ63JfDPHs/z/hH2vJA/nufV9jzvV8/zngp7LkiO53nFPc8b73neUs/zNnuet8DzvHPCnheS43neU57nrfY8b5Pned96nndV2HNC8jzPm/X7Z+kf342Lwp4TksNv1OzH92Ju4Tdq9svV9zDnLtKIyCoRGSkij4c9EeSP7/sH/vGPiFQWke0iMjnkaSH/RovInLAngXwpKiLLRaSFiBwkIoNF5HnP82qEOCckb5SI1PB9v4yI/J+IjPQ8r1HIc0L+9DbfkXXCngySxm/U7Mf3Ym7hN2r2y8n3MOcu0vi+P8X3/Wki8t+w54KUaC8ia0VkdtgTQfI8z7tERDaKyNthzwXJ831/q+/7w3zfX+L7/m++708XkcUiwl/ws4jv+1/5vr/jj8Pf//lLiFMCCiV+o2Y/vhdzB79Rs18uv4c5d5EGOaeriEz06RWfdTzPKyMit4rI9WHPBanhed4hInKkiHwV9lyQHM/z/ul53jYR+UZEVovIayFPCfkzyvO89Z7n/dvzvNPCngxQ2PG9mJ34jZr9cv095CINIsvzvOqy93bSJ8OeC/JlhIiM931/RdgTQcF5nre/iPxLRJ70ff+bsOeD5Pi+f7WIlBaRZiIyRUR2xH8EIqifiBwhIlVEZJyIvOJ5HndEASHhezGr8Rs1++X0e8hFGkRZZxH5wPf9xWFPBMnxPK+BiLQUkfvDngsKzvO8/URkkojsFJHeIU8H+eT7/h7f9z8Qkaoi0ivs+SA5vu//x/f9zb7v7/B9/0kR+beInBv2vIDCiO/F7MVv1OxXGN7DomFPAIiji4jcEfYkkC+niUgNEVnmeZ6IyIEiUsTzvKN8328Y4ryQJG/vGzheRA4RkXN9398V8pRQcEWFPWlygS8iXtiTAAobvhez3mnCb9Rsd5rk+HuYcxdpPM8rKnv/dxWRvW/WASKy2/f93eHODMnwPO8U2XtLN12dstM4EXnWHN8oez9M+X/vs88YEaknIi19398e9mSQHM/zKonIGSIyXfZ2ymspIpf+/g+yhOd5ZUWkiYi8JyK7RaSDiDQXkWvDnBeSw2/UnMH3YnbjN2r2y/n3MBfLnQbL3h+i/UWk0+95cKgzQgMobocAABkHSURBVH50FZEpvu9vDnsiSJ7v+9t83//pj39EZIuI/Or7/rqw54bE/b4vVA8RaSAiP3met+X3fzqGPDUkzpe9P1pWiMgGEblHRK7zff/lUGeFZO0ve1s3rxOR9SLydxFp5/v+t6HOCsniN2qW43sx+/EbNfsVhvfQo2kOAAAAAABA+HLxThoAAAAAAICsw0UaAAAAAACACOAiDQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQAUX3MU7rp3B4KX4+3sdwpPJ95D0MB2sxN7AWsx9rMTewFrMfazE3sBazH2sxN+T5PnInDQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQAVykAQAAAAAAiAAu0gAAAAAAAEQAF2kAAAAAAAAiYF8tuAEAAAAgZ0yfPt05fvzxxzXPnTtX8/Lly53zWrVqpXnkyJGaTzzxxFRPEUAhxp00AAAAAAAAEcBFGgAAAAAAgAjwfN+PNx53EGnjpfj5eB/Dkcr3kfcwHKzF3MBazH6sxdzAWsx+ObEWa9as6Rzfc889mg866CDNO3fudM574IEHNH/00UeaBwwY4Jw3cODAlMwzjViL2S8n1iLyfh+5kwYAAAAAACACuEgDAAAAAAAQATnV3alr166ab731Vs3Vq1cPYzoAkDMWLlyoefbs2THP6969eyamU+isWbPGOR42bJjmsWPHxnxcp06dNI8YMUJzjRo1UjY3oDDr1auX5kceeURzhw4dnPNs96ASJUqkf2KIy36Gioi0bt1ac6lSpWI+7swzz9Q8evTomM936KGHar788svzOUsAhRV30gAAAAAAAEQAF2kAAAAAAAAigIs0AAAAAAAAEZBTLbjr1Kmj+YwzztA8ZsyYMKZTELRUyw20N8x+rMXfTZkyRfOgQYOcsUWLFmnu2LGj5kmTJqV/YonJyrVo96Fp2bKlM/bVV18l/XxvvPFGzOfLAoVqLXbu3Nk5XrZsmeZ69eppbtq0qXOeHUtUxYoVNWdgD7+sXIvW6tWrneNTTjlFs32fgt58803N9jdqFipUazGeHTt2aLb704iI/OMf/9BsW3VXrlw5/RNLTNavRWTPWrz//vud40cffVTz119/na6X/R9bt27V/M033zhjjRo1ytg8AmjBDQAAAAAAEFVcpAEAAAAAAIiAnCp3GjhwoOYJEyZoDt6amgWy5vY1xJWVt5KOGjVKs11TIiKXXnqp5qeffjqlr2tLMc4++2xnrE2bNppfeeWVlL7uPrAW87B06VLnuEmTJppta9k5c+Y459myigzLyrXYrVs3zc8//7wzZtdiuXLlNAdvKd65c6fms846S/PMmTNTNs8Myfm1uG7dOs2NGzd2xuya87w//yiCv+Fijdl/Hxxr3ry55j59+jjnXXDBBQnNPQlZuRbjOfnkkzV/8sknMc+zJU62fLR06dLpmVj65PxazI+ffvrJOW7WrJnmE044QfMzzzyTsTntQ9avxd27dzvHtuTMlgvbP38RkfPPP19z8LMxlj179mi2f8cUEXnrrbc0f/bZZ5o/+OAD57wKFSok9FpJyJq1eMghhzjH69ev12z/bNNt3rx5moPfs5MnT9achu++eCh3AgAAAAAAiCou0gAAAAAAAERA0bAnkEonnXSS5uCtaIiWtWvXan711VedMXu72euvvx7zOWrWrKn5xhtv1NyjRw/nvCJFiuR7noXRtm3bYo6l87bs77//PuaYLYWaP3++5oYNG6ZtPogt2AHm8MMP17xw4ULN9nZWkVDLnbJSmTJlNI8fP94Zu/DCC/N8zMqVK51j+3lqO5HYMigRkWLFiuV7nkgN2xkoWFI4duxYzd27d9dsS6RERKZOnaq5bt26moNdLCzbEer22293xo466qg8nw9/st2d4pU7vfPOO5pt2e5ll12Wnokho4Jdmx5++GHNPXv21Pzbb7855+23H/9/eX7ZzzsRkb59+yb0uM2bN2suVapUzPNsOdVtt92mefjw4Qm9zoYNG5zjNJQ7ZQ379z6RxMvM0ilYLmz/e8pwuVOe+GQAAAAAAACIAC7SAAAAAAAARAAXaQAAAAAAACIgp/akOfDAAzXbdl7bt293zrMtYpFaK1as0PzYY485Y7aF7JIlSzQH3x/rgAMOyDOLiCxevFjzNddco9n+dyAi0qVLl33MGpbdwyKoQYMGaXvdH374IeaYXbNZ2K4059h9Z0Tc/S5sHS97WBTMvffem/RjHn30UefYfta+//77mj/99FPnPNtGHeFLtF7/4IMPdo7tfjWWbbMdz8CBA51j1vC+nX322ZptC+Bdu3bFfMzcuXM1sydNbqpSpYrmVatWabbt10Vi7y+GvH388cea+/Tpk9BjDjroIOc40c9X+/eMRPehsYJ7bl577bVJP0euCP6Z2+Pgb0q7T1om52Rbpts9FcPaT5E7aQAAAAAAACKAizQAAAAAAAARkFPlTmeccYZm25byiy++cM5r3LhxxuaUi2yL15EjRzpjzz77rOaNGzfGfI4aNWpoLl++vDNmb0vs37+/5qOPPto5r2XLlpptuYVtmYd927Rpk3Mcr/wseGt9QdnSqqeeeirmeYceeqjm2rVrp3QOSF6wPfDWrVs1DxgwINPTgREsY7K37yJ7BFuDZkqzZs1Ced1s1qpVK822bDDe2rO/lXr06OGM1alTJ4WzQ1jq16+v+cwzz9RMuVPB3HrrrZrXrFmT0GO6du3qHJcsWTKlc4ol0fkVBvG+02bPnu0cZ6rcKTgnWx6+bNkyzZQ7AQAAAAAAFGJcpAEAAAAAAIgALtIAAAAAAABEQE7tSYPM2LBhg+YJEyY4Yzt27NBcqVIlZ+y9997TbPcYCbbGS1S/fv00X3755Zo3b96cr+crrL788kvn2NZhBh155JEFeq1ff/3VObbtgteuXRvzccH268g82yIxWN991FFHac5ULTHyFm8dHXvssZpr1aqViekgCXaNJdoiFtEyePBgza1bt455nt2rIrgnSXAfRWS/77//XvMJJ5wQ4kyyz6xZs5zjefPmJf0cF198cYpmk5wrrrgilNeNongtuMMShTnEw500AAAAAAAAEcBFGgAAAAAAgAig3AlJs20Fhw4d6owdf/zxmoNtJGvWrJnSecRqiTZ9+nTn+Nprr03p6xZmBW1/ffPNNzvHb775ZkKP69ChQ4FeF/ljW2sPGjRIc4kSJZzzgrcjIzzjxo2LOXbIIYdorlChQiamgyTY1rxhteBGwZx66qmay5cv74z9/PPPeT7ml19+cY43bdqkuUyZMimcHTLJfn/Sijk59s9uyJAhzti6deuSfr4jjjgi5tju3bs1T5061Rm7++67k34t5C2K32nBOcWa49y5c53jl156SfOIESNSP7HfcScNAAAAAABABHCRBgAAAAAAIAIKRbnTokWLnOPGjRuHNJPcM2DAgNBeO9atwwUtySlsnnrqqbQ+//DhwzWPGTMmoceULVvWOWaH/HCMGjVKs729s2PHjs55sUoPkRm2c0jwtlzLllV8/PHHztgxxxyjuVSpUimcHRI1bdo0zcGuE3Yt2q548djv5wsuuKCAs0MiSpYsqfn66693xmznJ2vFihXOse2E2bZt2xTODpn0ySefaLYlbGeddVYY08kqq1at0vzBBx8U+PleeeUV59j+xhw5cqTmVHRWa9WqlebDDz+8wM+XK+J1d8pkV9B4XRTr1q2r2ZY+3XDDDc55tqNpOnEnDQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQAYViT5pgu1jkhmXLluX57/v06ZPhmWS3PXv2pPw57T43d9xxh2bb6jCek08+2TmuVKlSaiaGuIKtLW+77TbNLVq00Dxx4sSMzQn7Zvc7WLt2bczz7B4Jp5xyijPWtWtXzTfffLPmTNaKF0Z2P4R4LUoTbV+6fv16ze3bt9c8Y8YM57yzzz470Skin4L7GLz22muaP/zww5iPs/9N2HVaoUKFFM4OYbFt2pG3RPcvTFSPHj1ijtnP1uAeJfnRr18/zcWLFy/w8+WKeN9hdp8YEZFmzZqlbR52j6PgnOw8TjzxRM2NGjVyzktn222LO2kAAAAAAAAigIs0AAAAAAAAEZBT5U5Lly7N899zu1lu2Llzp3NsWwKff/75muvUqZOxOeWCBg0aOMdlypTRbMsoRNw1ZlvVrVy50jmvZ8+emn/99dek51SjRo2kH4P8sSVO5557rjNmy8zuu+++jM0JySldurTmYHvX+fPna/7vf/8b8zmefPJJzXadT58+3TnPthhG8oK3dd95552a7a32wXbNtow3Xst7W+5k168tZxMReffddzVT0pYexYoVi3lsb7MP3nI/Z84czatXr9ZMuVN62ZLAHTt2aF6zZo1znm1nH28tPvroo5rPOecczbVr1y7QPAuDWrVqhT2FpJx55pmamzZtGuJMoiteC+50s9+78VpwW7bN9uuvv+6MxVv3qcSdNAAAAAAAABHARRoAAAAAAIAIyKlyp88++yzPf3/IIYdkeCZIhyeeeMI5njt3rmZb7pTJW+hyQa9evZzjjz/+WHOwi8/QoUM127KK6667zjlv69atSc9jv/3+vGbcrl27pB+P/HnwwQc1z5s3zxkbO3as5oYNG2oOlpbaEot43n//fc12nQZv91+0aJHm4G2lAwcO1EzpzV729vmZM2c6Y/bW3gULFmh+4IEHnPNsicWsWbM0BzuR2Nt+K1eunL8JF2J2vYm4n5W249Ktt96ar+e368Wuldtvv905b/bs2Zopd8qM6tWra473O8WOvfLKK5rr16+fnokVIva76qabbnLGXnzxRc27du2K+Rw33nij5kGDBmkOflY+++yzmidMmKCZ36j7FizbjTq7rcb+++8f4kyiK9ixyX4HBbtvLV++XHOifx+YOnWqZruWRdzflPG6edkx+/efTJU3BXEnDQAAAAAAQARwkQYAAAAAACACuEgDAAAAAAAQATm1J41lW08ed9xxIc4Eyfrmm28027aF5cqVi/kYWw+6ePFiZ6xmzZopnF3u69y5s+ZffvnFGZs8ebLm559/PuZzlChRQvNf//pXzbZGO6hRo0aaW7VqldhkkS9TpkzRbPeqCNbnjho1SrNtJ7ps2TLnPFvnn2i9b7w9aeKN2dbvnTp1EsRn9xuxOdhu/aSTTtJs67eDe73t3Lkz1VMsVLp16+Yc2//WR4wYkdLXsnu12bWMcNi9FWzL+3jsHlPBPVSKFs3Zn/Bpc+mll2r+/vvvnbFPP/1U81/+8hfNwf1p7L5Sw4cP17xlyxbnvAYNGmi+7LLL8jnjwsnuURivlX3wN4BdL+vWrUvotX777TfN+dlPUYR9aBJx3333OcfnnXee5rVr1zpj9nfpbbfdpjkVvynj/Xv7+9LmsHAnDQAAAAAAQARwkQYAAAAAACACcupeyfnz52u2t8rZUhhkl9WrV2tesmRJzPNsS0R7+6mIW0bTv39/zbbdKf7UsmXLPLOIyPjx4zW//PLLmm1rURGRa6+9VvOrr76qOV65U+PGjZOfLFTwNl1bNmhvFxURmTZtmubgbaGWXXO2hC1eWUb37t33Oddk2PI7EfdWc8qd8u+ggw5yju37i/SpVq2aczxmzJi0vZZteR9vnSMzbImhLT38+uuvYz7Gtql94IEHnDH7uwd5W7FihXP88ccfa77nnnucMVviZAVLWeyfu33MBRdc4Jxnv5M3b96suXz58vuadqFn/1w/+OADZ+yEE05I6WvZEt7g3wvee++9hJ7jhhtuSOmccpH9u5iIyD//+U/NPXv2dMZildAHlSxZUrP9TA22+7alS8F235b9u4x97rBwJw0AAAAAAEAEcJEGAAAAAAAgAnKq3Onbb78NewpIAXtb2jXXXKM5uDO4ZUvagt2cZs2apXnNmjWa491ijLxdeeWVeeZ4JkyYkNB58bp3IW8jR47U/PTTTztjtkNPvJ3umzdvrtl2gxFxbxm16zKTt4FOmjTJObZlXEjOqlWrND/yyCPO2MKFC/N8TLAEoFSpUqmfWI6z3dRslzQRkddffz1tr2vLGmN1t0Dm2G5MtstX3759E3r89OnTnWPKnfbthx9+cI5tB6b8lrzbUm9bjhvsBmPLnWyp24wZM5zzypYtm695FBapLm8Kst2ijj32WGcs0XKnDRs2pHROhYEtDwy+x7bcKR77WzTRbky2tCr4vRj8DRw27qQBAAAAAACIAC7SAAAAAAAARAAXaQAAAAAAACIgp/akeffddzVXqFAhxJmgIPbs2aP5iy++0Hzcccc55915552aK1WqpNm27RYRufvuuzWffPLJKZsnEtO2bVvNCxYscMZq1aql2bZHR2y2jjdeK21baxts+2tr4hOt442KbJtv2F577TXNw4YN0zx37tyYj7H70NjHi/Ddmh+jRo3SXLFixbS+1rp16zTbFtzB2vtgi1JkVp06dZJ+zGeffeYcL126VLNtHYs/HXbYYTHHfv75Z+e4Ro0amm2r7u+++845z7bwtX/uM2fOdM6Ltf9Nq1atnPNsa/VTTjkl5nyRHtu3b9ec3z3C7B5v9jcvEhP8jRo8TqXu3btrHjdunDNm94yz54WFO2kAAAAAAAAigIs0AAAAAAAAEZBT5U4Wt/Jmr1tuuUXzm2++qfmOO+5wzovVPvH44493jm3rQ2Tel19+GXOsRIkSmmntm5hE2+oOHjxYc58+fZyxdJdcILNsm/uBAwc6Yxs3btS8Y8eOmM9x8cUXax4xYoTm2rVrp2KKhYotORJx24na1svpeC37fWc/H+zngYhIvXr1UjoPJKd169aaGzdu7IzNmTMnz8ds2rTJOX7ooYc033vvvSmcXe6oWbOmc9yiRQvNp59+ujNm2/muWbNGc7CU+JJLLtF81113aT788MNjzsP+lp04caIzZsufmjZtqvmGG25wzrOfxbY0CwWzePFizd9//32+niPee49omTp1qmZacAMAAAAAAGCfuEgDAAAAAAAQAVld7hS8dfu3337TfM4552R6Osgnezu+iMisWbM0d+jQQfNNN92UqSkhheJ1g7ElFkjM2LFj8/z3wRJPyhly2xNPPKG5V69emnfu3JnQ421ZafC4aNGs/mkQuoMPPtg5Ll++vOa1a9cW+PkXLlyoedCgQc7YvHnzNDdq1EhzsOQR0XHeeec5x7HKnYI++uijdEwnpwQ/y55//nnNwRIx28WpXbt2mm1HRRG3TLtIkSIJzcN2JbztttucMVu6NHr0aM0XXXSRc94bb7yR52NQMLb0ML86duyYgpkgE+x3cLDcKWrdK7mTBgAAAAAAIAK4SAMAAAAAABABXKQBAAAAAACIgKwuPP/ss8+cY9uK0tZ/Itpsu1cRkZUrV2oeN26c5v3245piNlq2bFnMMVvbjcR079497CkgAmbOnKk50X1obBvmIUOGOGOJ7q2A5B199NGax48f74xVqlRJs23/aduEiojMnj1b87Rp0zRv3brVOa99+/aax4wZo7lixYrJThsZEtzPYtiwYQk9LrhXCvbNrrc777wzlDkE98Gw3+l8v2eG/Rxevny55uB7g9xj3+Pg+x2195+/9QIAAAAAAEQAF2kAAAAAAAAiIKfKnawjjzwygzNBsmyb0KeeesoZGzp0qOb69etnbE5Ij1S0nAXgOvXUUzU/99xzmqtWreqc99Zbb2muVauWZspHM2fAgAGaZ8yY4Yz17NlTc48ePTQHb7v2fV+zLXMJlspQApN9Dj/8cOf4/vvv12zbNW/bts0574wzzkjvxIActWbNmgI9/pBDDol7jOhq166dZls6LOJ+z0YBv9IAAAAAAAAigIs0AAAAAAAAEcBFGgAAAAAAgAjw9lF/Fa3irMIj1T3AIvc+du3aVfM777zjjH3++eeay5Url7E5pUEq38fIvYeJat68uWbbRlZE5J577tF8ww03ZGxOScj5tVhIsBazX06sxWXLljnHjz76qGb7+WjbcYuINGvWTHPdunU1lyxZMtVTTDfWYvbLibWIwrsWt2zZorl06dKaE23BPHnyZOe4ffv2qZlY8liLSbJ7e3Xu3NkZu+666zTb79wMyPN95E4aAAAAAACACOAiDQAAAAAAQARQ7hRNOX/7mm0Ta1uQiogMHjw409NJl0J7K6m1cuVKzcF2sW3bttVMuRPSiLWY/ViLuYG1mP1Yi7mBtShuiVOw3KlLly6aR48erbl48eLOeUWLFk3T7PaJtZgbKHcCAAAAAACIKi7SAAAAAAAARADlTtGU87evXXLJJZonTpzojBUrVizT00kXbiXNfjm/FgsJ1mL2Yy3mBtZi9mMt5gbWYvZjLeYGyp0AAAAAAACiios0AAAAAAAAEcBFGgAAAAAAgAhgT5poosYwN1Dvm/1Yi7mBtZj9WIu5gbWY/ViLuYG1mP1Yi7mBPWkAAAAAAACiios0AAAAAAAAEbCvcicAAAAAAABkAHfSAAAAAAAARAAXaQAAAAAAACKAizQAAAAAAAARwEUaAAAAAACACOAiDQAAAAAAQARwkQYAAAAAACAC/h9WlJvLIy//tQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x720 with 50 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rows = 5\n",
    "cols = 10\n",
    "\n",
    "f = plt.figure(figsize=(2*cols,2*rows))\n",
    "\n",
    "for i in range(rows*cols):\n",
    "    f.add_subplot(rows,cols,i+1)\n",
    "    plt.imshow(X_test[i].reshape([28,28]),cmap=\"binary\")\n",
    "    plt.axis(\"off\")\n",
    "    plt.title(str(y_test[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "kjqNktGJ_5tA"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Tackling MNSIT Dataset Using Callbacks.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}