{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "CNNs.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.2"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-WXqA_23DfeC"
      },
      "source": [
        "Modelo de clasificación de imágenes con transferencia de aprendizaje\n",
        "===================================================================="
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0A098b6jtQa-"
      },
      "source": [
        "## Introducción"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3ujlNiDRtQa_"
      },
      "source": [
        "La clasificación de objetos es la tarea por la cual un modelo de aprendizaje automático asigna una o varias etiquetas a diferentes imágenes dependiendo del contenido de la misma. Esta tarea se volvió muy popular en 2012 en el contexto de aprendizaje automático ya que fué la primera tarea en alcanzar una performance sustancialmente mayor que con técnicas clásicas. En esta ocación utilizaremos un modelo basado en Transfer Learning el cual operará como extractor de features para nuestro modelo."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TghjzJl0wKId"
      },
      "source": [
        "### Preparación del ambiente"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ez9d-jEjDfeV"
      },
      "source": [
        "Intalamos las librerias necesarias"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mrs7T2yDDfeW"
      },
      "source": [
        "!wget https://raw.githubusercontent.com/santiagxf/M72109/refs/heads/master/docs/vision/neural/cnn/classification/code/cnn_class_transfer_learning.txt \\\n",
        "    --quiet --no-clobber\n",
        "!pip install -r cnn_class_transfer_learning.txt --quiet"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WMPAGXJStQbB"
      },
      "source": [
        "### Sobre el conjunto de datos de este ejemplo"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5PaXjD9ztQbC"
      },
      "source": [
        "Para ejemplificar esta técnica utilizaremos un conjunto de datos muy popular llamado CIFAR-10. CIFAR-10 es un conjunto de datos que consiste en 60.000 imagenes a color de 32x32 agrupadas en 10 clases, con 6000 imagenes cada una. Hay alrededor de 50000 imagenes para entrenamiento y 10000 para testing.\n",
        "\n",
        "Las categorias son:\n",
        " - airplane\n",
        " - automobile\n",
        " - bird\n",
        " - cat\n",
        " - deer\n",
        " - dog\n",
        " - frog\n",
        " - horse\n",
        " - ship\n",
        " - truck"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7bEodstwtQbD"
      },
      "source": [
        "Podemos cargar este conjunto de datos facilmente utilizando `tensorflow-datasets`:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-ToQd6ybDfef"
      },
      "source": [
        "import tensorflow as tf\n",
        "import tensorflow.keras as keras\n",
        "import tensorflow_datasets as tfds\n",
        "\n",
        "ds, info = tfds.load('cifar10', split='train', shuffle_files=True, with_info=True)"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pl6tmETmDfei",
        "outputId": "02c4b839-2e3c-474e-8c7a-6ee01a1dfead"
      },
      "source": [
        "print(\"Numero de clases: \", info.features[\"label\"].num_classes)\n",
        "print(\"Clases disponibles: \", info.features[\"label\"].names)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Numero de clases:  10\n",
            "Clases disponibles:  ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "StbfaIWWDfel"
      },
      "source": [
        "Veamos algunos ejemplos:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "tnIfBNDhDfem",
        "outputId": "dcdeffc0-51cf-430d-c53e-00c323daaf60"
      },
      "source": [
        "tfds.show_examples(ds, info)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 900x900 with 9 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Figure size 900x900 with 9 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAswAAALcCAYAAADzB+aBAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAA2GpJREFUeJzs/XecZFd5J/4/N1aurs49OUkzGoWRRhIKgBAmSQKzwmAMRlgGhMFrBMYsMrAYY8As+7MJXtYL2Ea2kH9gwGBWwoQVRoFRznFGGk3OPZ278o3fP4Zpq+p+nqtuISyj/rx56fVinjl969a955x7prqe5xhxHMdCRERERESQ+VyfABERERHRf2ZcMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUtjzaRRFkRw6dEhKpZIYhvHLPiei51wcx1KtVmXp0qVimovz35Uc97TYLPZxzzFPi81Cxvy8FsyHDh2SFStWPCsnR/SrZP/+/bJ8+fLn+jSeExz3tFgt1nHPMU+L1XzG/LwWzKVSSUREPv7Gl0jW6fyRSPlWhxfiYxmWA+N+cxbGw9okjFuG8rrNViLWbiRjIiK24H9Bt5oNGI+Vf3HbdhbGD016MP6oEq+bFj6+g+OtMIDxdhvHzQhfs4FsBsY3FHH7ou3DuG1FiVjs47aOgfuBIcljiIi0QnwNGvjwkivhDSxz+GUlDjrvrRfF8v/fPTXX9xej4+/9X298UArFzutQb+MxVfPw/Zut4Qmh3mjDuOvi+718ST+Mo/tqRvjYwwM9MJ618DlWG9MwfngMX4Ojk3jcNEN8bewMnobbyrVs+Hj+8GI8PwXKXBkr7c0Yt8ezigiaVuIIjz9TeS7EylwcGcpGtDG+Nlpc29A26oq3m3X54n99xaId98ff9+WXvVVc1+34u9DDk20c4JvqKs8VJ4sn4azynCsYOO7Xa8lzEXwusdLvROnrho3Pse7htUFLuTa5Mp5rCqUyjIcBHmW2i88nEPy6MzW8nmr7+Pwt04Vx2yrAeH9/bzJoKmMeRkUMZX4LA3wcy8HXQDu+beF5tdXufC6022357F//r3mN+XktmI//aibr2JJ157dg1iZGbcFsBfhUAmUQ2cpDwPBBewu3tZUFcKy0VxfMSntH+XjfUo5jKu9J+zWBtgDWfpVmKnHtHx/a+btK3EYPThO/pqOci6E9OJWJzVdGi2vhQafF48Txjw3mxfxryePvvVAsSbFrwSzK5BW18SQYKE+sSPBx3Izy8Ow+j5/DC2b8ECgpD6usrTxslYV0oYnnrHwbn7soCwttAWE6yvl4eMFsPmsLZnz+gSiLzudgwRz/khbMxy3WcX/8fbuum1wwK/coVm5q98/PxTO4v2ccPJ6yyoLZAv9w1BbM0QIXzKayYA6UfzYqvVEyyjXIZvA/JgILv1dHWTD7yvor4+PXjQ18/tqC2bGV88+CDwlNfBXUBbOhLJiVZ7Tl4HNRF8z2vJa3Tzmfpx/zi+9LWkREREREC8AFMxERERFRigV9Zj1jDki766P7voEB2HbFsqUw3la+09qaHsUv2piCYUP5NWOj3kzElN+6qb/C9GZmYHzsyCF8Lsp3tUpl/MJLilUYP3wUv9fhoUEYz5Xxr6dna3UYn1SOb7Txr3gP1JLXUkRksIx/PbSsJ/l9p2yEf8WUUf6tpv3aN2wpv3pSfsVUzCrfydJ+ldv1/evF+QtZzJRAzK5fRfot3IfNCF+5vPLr1kj52sGaFXj+GOjLw3ijOp08doTnmu3bHoXx6uwYjIcG/q7ynv3J1xQRWX3C+TBezON5QvuqV6T8mtRUfj3rKn3b176qEeDrE4Z43Gu/Eg3Br68j5TW1312byrjX5m7tqxSx9rUR7RscXdcsVN7jomMYx/57Cu2rgfkSHpOitNe+XhB6eJyZtvJVSPB1v5bytSclfUBs5SsQsXLuIfoqgoiYyldNszn8jLaVsa18Q0Q8JW+hHWo5IMrxla+sGAZuXy7h72Cj9Vfbw+cSh/g1bSUpIuPkYNzN4GsfKvNYoHwfvN31HWavjc8b4SfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUYkFVMnYfnBa3KyO6EeGMxjUnj8B4Tlmi/+yJPbh9tg/GN595JowXQRWORx7bBttGSurs2tPOhvGDrbthvH+oCOOrl1Zg/JXKvgZbfnYvjEcFsKuOiKxefyKMN5SqFzNTuLJBC1QWEREZnzgK4xOH9sL4FMhQHiniDGpL2QwijHGXdHO4n8V1nFk9MIyzk8O2Uhmgq7K9FYQiMg7bLjb16rRI1Jlx/MiDeCxMzeIdptau2wjjjQbuq8ZSnJ3dnMX3228m+3ag7Mp1y003wPg9994O48tX47lsZNl6GG/VcTUaR+nDjouzv11RNmxQNgioFPGuXNk8jh8+fBDGH9p6Pz7+MK5ckqkMJWKuUh0gVspkmFoZC0WkVRZRqnMoxVsSZ2OxPI6IHNvkqnujq1wRP+dcZUc1ZcM2Ea06VQuP7TijbEJlg/Hh4udNECpVVZQNTZQ9wcQx8Xs1lc1VIqVChLqRjrLDSvfz6emOM6NU+tIqWfSU8Q6qhjIHuU5yzmq0cHUurUqG1kG8Nl6n5JXaVYZSHchTNngKu86n+89p+AkzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGlWFCVjB2794ttdmZN7jl4BLbtH1kG44USzp6ensIZ7blBnFkeeXif8PrMdCK25d/+H2wLtkMXEZF1v/lGGO8v42oVuZ5BGM8OLoHx5TmcbfxfXrUCxnfP4GoNA8vxNTbcDIxr2clK4rrsObAfxr9+7TUwfvhIsqpG6bQTYNt82YHxIMQ3pbeMM/RzFs7i3XASvvaeUiXD8zuvTbPtidy1A7ZdlLq6zuTkGGx20803wvjhQwdg/Owzz4HxYhbfV8fGmdWFnmQliJ07cBUIz8PZ+JbSl+67D1ev2dDA57JvL64Ucuqm02D8BeecBeMSKxOUMnH1DuBqGJmMC+OThydhfOe22/Dr7izD8OnnvCx5LivxuLccPAfZJn5PSgK8xEpVgpaPM96VsARR5z2MTK20w+JiGIYYXVUyXBf3o9DDc2qg9N92E48/R6lkYQT4WY8+7rMzuOKM5eBzaTVxJYXudc5xmQi395VqFWGIq1L4IT6+5yvn4+ClmqGUdentxWuVWh1XsqjW8Porm8H3NpNJDihXWXe0lEpITaUqit/GgzVS5j1bqZjTbOLqX93aSjUNhJ8wExERERGl4IKZiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRigVVyTgy0xSzK1MxCHB25R0P7ITxolJpYv9evPf5wQNTMH73PY/C+NTkoUSsMYuPYSsZ2/9y/XUwXopxhY/qY7th/MRNJ8H4kmWrYHygD2ehx2vxNWuEOHu42sSZuU0lns/hSiRi4Mxfx8WZyLGTvD4HZ3Gae1zAr7lsxRoYz9r4GpSV44xswFn6gal0eTPf8cd6oyEi1+C2i0y93ZbY7qxqMgmq0YiI7DuwF8a9APe9U07eCOO+hzOce8p9MH4IVOG4+757YNvxyWkYD5WKDJ5SkWdibALGR4aHcPujySoyIiKO4KzwSi+eD7SyNjlceEaqM7iiSW0CV8ExQtw+9nHG/Kqh5Ljvy+EqAHkXn2Q2iyswOEplBl+p+DM5i/vNbB1n5De9zmtvKFVYFptKT49kMp1VDyLl/reUagdNZdyYSlUDV4l7AR4fjpv8vC9SnonaGI60KizKX9gRPk42h5+J7e7yQnPwcSwbf4Zpmvg4kSiVRZRnd/f67bh8Bj9HXaU6h+8nz1+r8NGs4TEZKRWrIqWiTS6Dq3Bo97xYwFWDarVqZ0ArxQPwE2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkqxoCoZJ51xtthd2fK+kmDoOzjrcrKJMxpDuwjjVSUz95CSoT5TTWZqlnI9sG0U4mMf2IGPvaoXv6dCBmewjj/8IIy3d++A8WWv/w0YHzhjM4w//NhWGJ+cmobxo4cOw/jS3gqM+93ZpD/XX8LXszaTzIbdN4n3rq8buOuVB/C/4WIXZ/1aJu5P47NKP8vg7H3D7byHDV/Lbl58xianJNfqHFcHDx+BbQeGB2G8WsNVcG644ccwfvTQKIxffMklML5j55OJmBfgPqOEpX9gGMarjWkYHxrC7V9zyavxC5hKRnuM+2SEC4tIGOI08qhVw3Eftw8bSva6j+9VpJQRyYMxuLIfV/iwlCoDhnJtxMbn3vRx+5yFz9FT4nFXpaTY5rgXEbENQ5yuqhVTNTyXNxq43xnKHG+aeI5vK8/jdglXOxBQUSIOlMoLStWLvl5cdWd6SqmspXzEGCuVGnr78PGV7iu1hjLoDXz+2QyuOjOmnL9r4moebhavbRzl/aJ7aCvzWMbA59hWqgN5Pr4GUYCrcGhVLmwLn4/ZVVmk+89p+AkzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGlWFCVjPPWr5Ns137es1VcSSFQMqpjZW/1Wk2palDD2Z4rhkswPl5JHr/VxNm9QUPJWu/BGaP9lTyMx+0GjPfVlKz1UVwBwGzh4zTbOGv0yJExGPc8fC1zyv7y9b17YTwO8HGGBGefTpvJeEvZ/73g4Cx6J8THtmKc/WwaLowHAe5nLQMfR6LOPtJo4woCi9GOrY9IJtuZXZ218L+1X3zuOTB+220/g/GeEh7H1RrOiH5i+24YX736hERs1+6DsO3KFethfNkynNE++yM8zsJIqT6hZF0vX7YcxltKhr1Xw33QjPHxHSWT3jbw8Xc8gav1TB3Fc3pTmeduu/GmRGzNklWwrZPD41XLdG8rVQNaHr4GrQZ+r36A75UYXf2YRTJERKRWnRW/3Tl3t9st2FarehEp4yBQnkNadQQ3xH3GziSfx7Gy7sgVcaUNW6kOERl43WEqlRfqyrO738WVdCylo83O4rGXK+Dz7O6+xxVzeK2ScXC8XVeq1yj3EN5ypdiEqbzXWJnHAqXSiWHg45jKRdAqo3QfRjksfq35NyUiIiIiWny4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJRiQVUyTrRjydudmY2RUjki4yrZ0EpG40x9BsanJ4/CeOTgPdGdwWWJWH0GH3vskFKtorcHxvO9RRivt3Hm7Okja/DxH9sK476Nb8fs6ASMt8cmYbyQxXu3T+0/AuN7H3kUxitZfA+Lszh7fzWokpEfHIJt+8r9+NjKfvFeDVc6yeXxNaseOQTjTQtnzmbLnVU7/BbOCF+MpsfHxHU7M+bP2nQGbLtjxzYYz5j4PkUhrl6w6Qx8/FWrcPUFG4wdrVrMeUolj9HRnTBeLisZ9jYeZ/UG7qs9vb0w3lSqA9hKRnsGjDMRvVpPfQbPcytWKFU7vPNhPIzwfGBJ8jrMzMzitjaeWyOl4kirjcdrQyl2E8RKyrvS/xKp/QtJmX8es0xLrK6KEN1/Pg6NPRGRhodvklY9I1QuvdoexNysVg0D992MUn2i1IurOHnVaRjXqlgUleocjRYe8xZ8VyIS4uditYrnGjHxxYwEz4m2q8wpyvlMTSXXHqZS6UargKNXscDn7ij9rN3C65H5Hl97PYSfMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkqxoKS/TS89S0r5zkSUtrJtc3UWJ320lPa9fTiJL1y3BB9H+TJ6DL7sHrbwF+ZXTeFkwJaPvxjfauL42heeC+Or166E8fFlS2E8s3EDjB/ZtQfGcUqQiKFsA2s0cSJbycXbVw8q21ofqeEkohNHkvdqZCVOLKqO42tvjeLtxAtKUpB3ECeFHlX+KWgWcD+rlTsTOlu+klW0CJ111lmS69pqdbaKE06npvCWsloCxu7d22H8iSfug/H+AXz/2iBJ07Zxf48F97ED+5+EcVESZbLK+Jgcx9cg9JXkJSU50VJe14vwfCYhfr+BEh8aGYTxgUE8P7kZnLBXKiePUyzgtmGAk4uUdCGJcXOJlDk6VLbnFQPfK7freRFaC3okPm9ls9lE/9ae9bUaHk+iJKdqW2nbSjybURL8QdKtqyQmmsp6wVaSwC0lbmdw/3BsJanQVRL5Z6ZhPK8l38X4/AMlOV1LKixV8DV2XHyNo0jZat4H863y+WtOKQDRbOJkPS2JVHuGeJ4yHyrmm8yK8BNmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKsaCU4P4zN0m5VOqItZVszEqgZVfijEbDx1mgVogzI7UqBgGI28p2qaayLW97YhrGv/O1b8L4vgfx1tJrlezLtefirWf//tYtMD5x6CCMv1I5Thjja9Pf92IYN+t4S13Hw1msTaUKQGAl//1llHAtj97BPhgvKpUEckrGtaGk1xtKdnKkZCH7mc6M67qSEb4Y1ZpHJZTOrV+f3In7fGRUYdwLcZ+p1nDf++GP/xnGH3gQjxG0ZfboUVxBJYpxBR+J8LiZmcJb05cLSjUJJWO+Nosrw7R9fA08C2fA28q2t7aJ57NaHb/unr17YLxSwu/rhBPXwHixnKyIERt4nHnKVtdBjJ8LE1V87UNR5vRMEcZtE98Tp2vOCi1ujS1yrCJB2FWVQNtCWItb2pytVDLJKJUaSln8DDFA5aQ4wPN2pFSZaNWU6hkmXr9IJlmZQ0QkUtYpU8rcoc0FWsEGX1lPZRx8Ptq1DJRtqk2l3wchfl0HPEfNCH/+qvWPOMYdQauGYSoXR6tyoR0n6LqW3X9Ow0+YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoxYKqZJi+KabfucZ2YpylaQrOXLTB/u8iImNT4zC+Y8dOGA+VChflUjkRK+Zxlm2pXILxQy1cHWL7QVytYv/eIzB+w9bHYDy7ZAjGH3/sARg/d22yAoCIyOtf/mswvu6kU2BcHJwpLkqliVCpGuAolSw8kFHbNvHBtWPEAb6vSqETsZS/8CKc+doIcTZ+d8Jurd4Q+Sv8movNHXf+WBy3c9zOVnHVi7ZXh/F8Uamao9yn6Wk8pibG98H43r2PJ2K2g+egqYkDML50eBmMxyHuq1k3B+NLRoZhfP++HTC+Y9dWGN9wwkYYX71qLYyHSgWiTAZnzK9cieeV0MdztGVpJWmSlQmiCF/7WKluMTOLqwncccdPYTxUyuPkir0wvmw5vpYrl6/u+LOtVBhajLrvlJvB/d32laoGajUX5QWVigzNllItRpLj0hR8DLX6hFI1KZPNwHgQ4LVBNofH2GwVVw2qN5X3ZOPzCZT1jvZgLJbw2qbh4TlC4ylrIVQdrVjArxkq91WrmOZ5uNKJUmxDrZKhVeH4RfATZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCm4YCYiIiIiSrGgKhkS+cf+e4o4VLKKlazOWnUWxnc/+QSM792xC8a9Nn5d20m+pVJPD2zbN9QP409uexQf28OZrRsH8HFu34fP/fG9T8K4pZWraODX3XbvfTA+M3EYxodPOAHGB1eugfG2UgUgb+Asej8A98TBqa15G2chi6n8G87FXdUxlPRnJaU2VJp3Z3TPzuLs5sWo7Tck6poqqrVJ2NYw8LgPQpxtbRo4gzqfxTfKMnD/cJ3k2NEy4+s1XOHjiNI5li5dAeONOn5PrUYNxqfreFzeuuVHMG5GeNwvHRmA8WIhWSFIRCSTwe+rqWTq+1omvYXfbxO0DwLtsxilmsAsvjY7d9wL41NKVQ0x8Lxy3nn4vZ6wYqTjz4GB2y02cRhK1PUMt5QqS0rxDIlCpdqBUsnC93Dc8/Gz3gLPesvCz6Y4xsdut/Ez17Dw8yOOcb8OlEohDeXZrT6flOpisTLvxUq1mChSKpcox0nWRDkmUOYCI0rO86FS3SII8Llo1TAccF9F9CoZvtI/NEbXgbr/nIafMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUYoFVcnwrVg8q3N/bl/JDp2cwVn0d9x2B4z3KpUsCkWcgrtr924Yr9eTWamnnLYRtl11As5+7y/jPdH7TZyBe+LatTC+oz4D4/v21WF8pZLlvnHpUhgfWjoI47Nj+Nrv2XULjK85pwHjcW8Fxp0Qd5t/u+nmRGzGx8d+y+suhfGNG07E56Jk8UZKhquW+aplCRtd1TkMUymxsAi1GqEEXdVOalWtSg0+RoCT1KXZxFnYllL9pKjMBxk3+cJaH3BsXKkhlynAeF8PHmcH9h+E8T3DO2G8fwRfnA3rl8P44EAexk0DX/sIZK6LiAQhvg6lci+Mjx3F1X0eeuQRGJ+aSc5zpqVU2MnhikKjo/g167OHYDzycBUbQ6mSMXlkB4zXpjqrczQaeG5ebPwwFLOrSka7iStcTVdxVZisjftdMYvvkaHM8VaoVLIwk3FfqQ4RRVolBa1KBn5OeB4+x2YL98dQqSLmZpRqHsr5xzEMq6II/4CllA4KlNfV5tBcLjm+gwDPP9q52zZeR7gursai0SqCaMc3u571oVLRDf7s/E+LiIiIiGjx4YKZiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRigVVyTCMTCIL2VSW3A/e/yiMb3tsO4y//GUXwrg7jE/xgQe2wvjokYlEbMmSKdjWUSoh9Fb6YNxQ9pFv+LgEQFPJSPUjfJw15SEYX92HzydbwNn+ywaXwPj3/vl6GJ/wcZaoOzIA41OHp2H829/7fiK2fXQ/bOvVcPWMP/uTD8F4TqmOEJk4BVfbHt40lOxhI/3Pi9nk+JhYdmdfjpTM9UZbqeAg+LrHMc4Wb/ptGLcDfJwIVILIZfD4mJ7BWf0jJ54C46tXnwTjsYH7tmnicz9y8DCM9yvj2/eTFX9ERB574mHcPsTXstrAlUgGByswPjmK5+6HHsIVjur15Fju7R2BbQMfn+PRsX0wXps5il+z0YTxOMJz7mPN22G80FWxwWvja7XYRD//31O123jOlhBfM9NUqiCYSvWaAq4K47fweDJA+QVDmbjDEM8b7TY+dm8fruaiVZNoKFUy3IxSbUlZe4QBPr4oFUS0ShBaNQytHkSolDFayPF9ZR3kOLjqRXe1in9vj+cI7RxjpQyHFu8+d+08EH7CTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKRZUJUMMI7G3uG3hQ+zcuRPG3QzeR95SskZjJb5yxXIYr4F97U2lZMLY2DiMez7O9K8rOaYP79gG47MzuDqHq1yzpcuWwnhjqg7jR3bhLP3ZbLJSiIhIO8Dva8UgroYx3caZ6OUcznpdf+K6ROzJo7gywM0/uxPGn9i+C8Y3n3UqjGv7wGtZwlGMs5CjyOj6M86wXYxMI5LuxHbbwv/WDpUKME2tqoHg+2EoU5Pn4fsStJIZ1F4Nv6bfxq85MIDnlEyuorTHGfa5PJ7jJg/hagIFpTrAnn2TMD46jsfI5CyeJ0ITVwtZu3YFjNfGcCWj7jFyXBbM6Y3ZGdh2egpXWvCCWRjXni+xj+NtpUqLJdMw/shDndUzgkCrI7C4RBInKtu4GVy9wHXwmLeUyjhhjKsdGErlndjA8SBK3uuMiceeZqHPiXoT999GC4/tUPA108o42QY+H0Op7hQolSO052KsvF+1qoZyHHgMZa3meQurPKPeE6VCiVpVY4Frg/ngJ8xERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJRiYVUy4jixP7eh7MOdy+VgfOtjW2E84+BTGezrg3E3g9tXekuJ2MzsNGw7O5usqCEi0t9ThvHhdatgfM/2HTDuVnEmrGvjjNdML37dTKUA41UPZ4RHRRiWU156PowvWbMGxvfsxZVOJms4e79USN5zbZ/28QmcRX/gEK6qcfrmk2F8ofvIGzG+J92tWSPj3/mBKVHceR+DEGdnBziRWSxbqXrRVrLLPXygrIOz4E9Ye2IitnI5Hq+mgau8VPpxtZiWj99rbOLjbHtiH4zv3ImrW3hK1Q7lEkux0g/jAyN4HJf6BvGBlOsQOy0YX7Z8E4wvHepJxJ7c+ihsO370cRhva9dAHd94XvE9nBkf4rcqfrvzIrNKxjFeECSe7YUcHnu2VsHBV8a2co39CHd4S6nIE6NKRsrEbSnVVmwbPw88D1fAqdbxc0ubD/0q7tfNBo6Xi7jig6Odp1LRq6FUJTKUeThbwGuM7qpox9ngOJHSDxoNPJ+02/ga57VzUdYS2rPeV66NVlVjPvgJMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpVhQlQzDMMTsyprU9lx/wQtegA+iVS9QsuszLq620dvbqxw+mYG7a9ce2LbVwFmaTj9OqX7hK38NxvPFPIwfuBln1LaUUgIH61Mwfu5L8LVs+VUYj/pxmYwlq1bDuK1kyxtK9xg7Ogbjvp/MzB0cwFVOmrM4g7percO4ti+8lsW7UN2Ztlrm7WLUaolYVlfQ7A4c42ZwJn0pj+NTY7jP20pW+8tf9koYv+TiVydia1evhW3bPu5LMy0l2zpU5izl84bRI7Mwns3dC+O7d+OqGq0WzvLOFZJVKURECsURGLctnHVuWVkY71+G59ZyL25fGUzOf+W+adjWzR2E8enqURj3lXtlxLh/ODZ+XtRq+FpmnM55SJtnFpvA8xK9O7LxmI+V6ghWhOdmI8bHMV0cj1A1DBGxneTxLaWahB/iZ66pVLrxlUoeLfCMExEJlLlAKdoiUYDXHmLh9+okJuBjQuV9aRUi/FCpTKGcaKUHzx02mP+r9QZsq42pKFLuifJMt7QqGYKP7/m4OkdOOp9FsSiLT3Ru825JRERERLQIccFMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoxYKqZJimKWZXpmIY4z3UTzzxRBgfHhyC8ZlJXFHiwfsfgHE/wFUWVq9ZmYhNTyv7v7dxdmVtGlefWDaIKz6UekowPlLG2eyDg0tgvGnhf7/MRPg8cy6uPOAp/wyabeEs1mgKVyq4975HYXz84CEYP/P8ZDWPuIgrdtz60ztgPPTwfY2VaixxjDNqF1rlors1a2T8uyVLVoptd04V1YZSoUW5crGS6T7QvxzGf/O/vBHG3/hGHO/vH0zEAiWD/OBhXOXl1pt/AuN33o3nIEMcGDcFV5NoNvFc6eGwhDgxXiaq0zA+28YZ/Lk8znR3c7i6j13E7Sdm8PudmU4+RqZncP8IlSoApTKeW6vTeHx7TTyXWRaeEwf6l8H42rXrOv7s+748+vAO2HYxsX7+31NFPp6bDeW5FSmVJkTwPbVt3L9CZU6xrORxGi1cZcmw8TkaSmWOTB6PYeXU1aoagVIpRJTKIr4yTy70+acU4pKGUnlHe1+9SpWMGFS4iJT5VnsWW0rlD8fB/SDWni3KtXEcpapLV3vt5xF+wkxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCkWVCVDDElkUyrbfqsZkAUlA7tWxdmte/bugfENG9bAeLmcrFixYsUK2NaxcjDebuBs8/0HDuD2bZx5OtLbD+OehbNAD0b4OFomrGvi44wfnYTxxgzOLN91/+MwfsNPbobxl73kAhh/6csuTMSiHM5af+LeR2A8l8HvSetohpbeq3VMTXfzBf7481m+WEpkLntKYrGWnW0ofX7pEK4YEwruNz+9BVdXcbPJig/aucxU8TjYtmMvjNfauIxFqLxAW6lG47Vx+zBSqmeELRiPQIa6iIhl4+ncmFIqyWhjRKm+4yqVBk4+MVmZaHZyGrYNDRfGly5bC+PVLJ7L2j6eo0eGcX9arhy/UKp0/LnVaonIv8C2i0oYiBF23m8TVKUQEclncUUJz8RVNZpKtQ3LwcfPZnF/RJUmmkENNnUc3HetPJ6Xeodw1ZbCkSMwfnR0FMbDGI9JJ4Pjga98hmlqFR9w5Yhj/TjJU6pQlYr4Hlo2fl0frHl85dimgd9TVuk32trR9/E86Th4TglDXLmkOx4qVcgQfsJMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpFlYlQ6Kf//fvYkNJRVeObCV2qD/GC3DWcyaD2xdLyWoYIiKGkWw/MDgA246NHoXxVg1n2vpNXMXiwJEJGLfLPfg49So+zt6dMP7E7uUwvn7NahhfMpTMWhcRadRx9v49DzwI473lZOUBEZHzLjwXt1+SrAqyYvkgbHvqSbhySW8vrqISKJUBHKUygLbvfKjsG292VSKJlOzjxajeqIvtdF7nIMDXx3bxuBTBmcyjY7g6zo/+7WYYj5TyJZabzLg2bZwBr8UtF/f33oFhGM9kcHtDmeOUpG2JlCo4rXYbt1eqalgm/vwjVvq8xnHxmIqUe57LlJOvWV4G29oW7gdBgLP6BwZw1YsVK5X4itUwPn50FsaPTnTO9e02Po/FJgh86e5OEe7WYtlKdYuMUt0i0kpr4QGSyeHxCh710msn+6KIiK08J9wM7o+eUoWlkMeVHaIArw18JR6GeKyGSqUJR6kyFARKhR3lOJaN329GqUQSRXjMo8oX3c+I49o+vgaONg8rVTK0uKEsNtVqQspx5oOfMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUYoFVcmI4yiRca1VI0D7vIvoVQr6lUoWF11yEYy3WziDHGVqZrI4S7PRwlUjfGWve9/DWbxGNgfjy1bg6hazs5Mwvj6H//2SL+PKEQPLcCb6QF8fjE9P4tddsgJXrGjM4moexZ4KjJtuMut1xcqlsO1ZZ58K4wVlT/tQy3hVqlks9F+Ccddx1H69COWztjhO5711HFwhIoxxPIpxH9YqRIQxHrOxUiWjBZLFQ5xALrGhlKto4oodmaaSvZ/Bc5ClVIJwHBy3bdzn3Sy+lt199ThTKT6gVZJxlWoYBSXbXZTXtaxk+94yrlYQRXg+COvjuL2P78msh7P6t+3GlY+as3iur3ZVPvI9fE8XGyfviOt29tfYUDqYUiXDdnHcUcptNJUKJS0ft3eNZL8zBY/VrKtVgcCTRLWGn335HB6rAwP4mTs+PoNfV1kftZv4GjQC3H8DZZLTKkE4oLqFiIgNxrCIXmlC0DpOeWR2PzvmmivzSRgqFbHwYVTaNTC7yr+gih8afsJMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpFlQlwzTNRIahoWTOGkrmYffPH5fL40oTPZUKjE9O4YoPKPNycnoaH7sXH7s0jDNhtz++C7fv64XxE04/BcbtDL4GJzdwRrhp4tvUU8KZ6Fp2aKkHt1+6HFfzeOTBrTB+6AjOaD8NVDwYGh6EbdeffCKM54q4Hyj52fQfYN/unYk+5Uc4ZbnSvxrGewdwFZwgxn3VVqvsKCcJMqsNpQKHoYwPw8IH1+Y4L8AZ+VaMM9dbbRx3bFwRpLtKwXFRhF9XyzrX4o4yF094OFPf9/D8tG7dmkRs6coTYNvAxxnw7QIe97M1XGVgbAafS7WG483pKRifme08fhjgCkmLTaW/R7KZzsoSrvIcyhVwBZx6YxbGvQCPA0Ppj/U6rhCBilxFSpWToI3va0YZY+Ir56hU3ukpFWG8Xsfnk8vjNYYo6yZfOZ/ZGXyNZ2ZwlQ+t2pQo1UXiCJ8PmhNDpeKIpVXJUJ7qlqXM20oZjkjpT7aJ5/ko7Hyv2nyK8BNmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKsaAqGbZli213/oiWga1llosWj5S91QOcQZ7L52F869ZkZYeJSVxR47RTT4XxvJINPD2NM7Af3b4DxttKpn9poALjmQBn2lpapqq2d7uyL7xt42zVhpJBfGgcX7f7H3oExl9wwTmJ2KqVQ7Dt8nWrYdxxlWugVDawbRyf7z7yx3X3V0vJsF2MDu49kLg+nlKuYnYWZ4UXC30wni30wLivZC4bWr0UcFst7fMAZeDolViUv4lxxrmWSW8pfc9U5r7Iw3PfQqtheB4e37UWrobRmBqD8YnxwzB++OCTidi+PdthW18pQqGdu53F1TNiOwPjvo+v2dgoruwzPn6k489RqNy8RSbjxpLJdPbvoInHdj3A96LpKZUmtEo3Sh/Q1gbNdrL/WsrBbQM/+xwDV8ko5PB7khwe89Umnq+yWXz8wT5csapQxOsarQJFtYqfl6OjuLrM4SOH8PGDJoy32tr7Sl6f2NSqTeB5z3GU5aeBjxMpVY+UohriadWEuiqFxCGrZBARERERPSu4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJRiQVUyEK0ahlaNwFQyyzWV3l4Y37tvL4zfeffdidhLLnwJbLts+TIYnx7FWeIjS5fA+L2PPgbjKItXRCRXKMC4p2Sta1UytOxhLRs/k8GZv5aLu0H/kkEY33foIIw/8UQyW/6UU9fDtn3BUhjXuqRe3WL+e92nHac7bln8t+RxoRcnitvEgsfx1DjOwn780btgPJPD2eJ+iI8fKRUrLCeZjW45uL8bSgUV7fMDrTJHHOF4pExxpoUz9bUKEaEyvrU+rNEqR/gtnBnvNWdhvNmYhvHJif2J2LRSaSOKlGtv4WoY+VIFxk0bVx+wbDx/5JVqG8tXru74cxh4MrbjNnyOi4ibjcXNdva/VrOBGyv31M3g/lvJ48oOQbCwCiWGlTyOoZRMyDh47NmGUk3JUAaxUoEqN4uP31PC/XTJQAnGM1l8nEiUa1nBa4lSpQLjDWVs1+v43kbKHOSA+VarwqVVsnJd3N73cTUWRzlOqDyLRIkbXZVUuv+chqsCIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKRaU9BfHcSJBZaFbY5takpaJ2+cLeKvII6NHYLxcSW61u+6EE2BbZadFMZSkmv7+AdweH0ZqU3h7yrySsGK6ypfXleSiWPlCvq1c41YDf7HfUr58/+bfeTOM796zG8ZnZsH71bYkzeCkB3VLdcUCm9MzYNmZxH0xDGXbZiU5ZWZqFMajyQkcF5zgoXR5EdDnTZCYcqytkuijJOXZJp4m9cRSfBzXxYlnlqsk4ypJNIaBz0fbDl5L6s1YeG6Nsvh9FYr4fPL55Pn39vXDthLjY4QWTl4ylC2wLeXaa4nNpnJtMtnOaxAoCUeLjZVN5mH2ujhZT5SEuoyL+1cc4Lkja2dhPNK2gkfbRStFBTxl6+dmG5+LpaxHLAufo5PD7fv68TUo9SjJ7UqiWrVRw+ejJDbni7i/Dy/Ba5g4xNe4pye5nhLBiceNRh22zSqJjFpyn5b8qSUJionPPVKuZaYrMTjWEpHRS827JRERERHRIsQFMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohQLqpIRGcmCDdquzVoFh1DdchJnKgYhbh4olSNO3XR6IlbsUbbfbXv42Mp76hvAGaaVQbyF9EwNZ406SpWMMMJvNlLKecTKtbGV7Tv3PZnculpEZNlKvEX4iy44H8YrvThzdnYmWSXD85TtfS2c3auVHImVyiWhtp2xUgkhUv+NaM6z3eIzsmJVogJDu12FbX0lw9lW7rdh4ngUL6xKRgyqZBhK1Qsx8bG16jja9uuWMs5MpaqGqVW30LbwVo6jluV5tsS4YoVp4bl7YKA3ESuV8JwbKJOrL3hObDRbMN5fwVV2igV87kfGcTWWuKuiSWwok+pi4xjH/nuKLKiGIiISKlOlZ+P+Eiod2NSqHSh7zfut5PPbVcZGJofngsDCVTIipQqLVlEqVo6TLePKOEEWv9eso2y9rVUcUZ5TjUk8bsrK+YTKQksp4COtVrLiVhDi1wxC/J5aLa0iDb42QYivsTZvh6iKioi4uc5roFVhQbgqICIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFAuqkiFGfOy/p4aUjEbRqmEIzsbszsI/rlrFe8CPHR2H8U2nn5qIZTI41TNWqlLkyzgjNVvB8Q2nnwLju3fthnE/xK+bUTJwgwBfS9PBmeWTE1Mw/sST22H8rLOSlUVERPp6cSb68BCuClKvJquCeCCTWUTEyeAs5ChW3qu50NIAWr/E8e6kaCVJelHKFIpi2Z1TRaBUE4g9nJlsGrivGgYem5apVLhQqp9YdrK9paR4o4oaIiJegLO8tSx9U6nCoZ67aJVbtLGgVA0ItTGiVfPAr6tX88D3ys3g4zu5SiIWWzgbX5ly1apHoTKMgwDPK56P35NaccTuujYRvlaLjZVxxe56boZKfzRs5XM3pepF7CvVM5R+bSiTsVsAfUw5lZbyzA3zyrykHGi2hueIpvKMNmx87n6Mj5MNcf9zu/vp8eMr17KhVOiq9OAqMrO1ZIUrEZFsXqloAuYUyy7CtrksXkf09OBqW56Hx3azhdeCpnJttGd999wRKnMJfK15tyQiIiIiWoS4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJRiQVUy4uqUxNK5n3cY4P29NZGPMxItE2e0Tx85COPNqaMwPlBMZr0a7Spsm4lxRr/jKqnZynFWjlRg/JF7D8P49OG9MD4yPADjQRvvue5mcPbpnicegXE7xFmmJ64axq9bxZVIosY0jE+B91Udx/evR6k4Eiv3BFVBENEzZE2lvaFU22i3O/tlu1qD7RajII4ljjrHhJLQLlGslRdRsrwNPAVZFp4PstksjueS2d/ZPM7aDpVSDdX6NIx7yvizHXwuto3PXbsGoVIlQ6sQYSoVYEyl+oeljBGtmkesxK0MrigQCGgf4HNRiglIoFQx0Cp/tDz83IkFVx+Ilc+Gwq6qPN1/Xqza9UCMroohhoGvuatcW8tR+pdagQgfB1VkEJFE5R4RkUipjOApHU8p/CGxMmYiC/dTt4zbB8r6aEKpZNVW1kdlZS5z0NgTkbaNn6MZpeqF31KqG7m4fR5U6MooFUeUSyyOMu9lbK1yiXKODn6GxA6+J3m3s7qKqVV5Qa8175ZERERERIsQF8xERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhSLKhKxt5H75FivjPD0PNx5qK2F3ugtG+0cHboEzt2wbjtTcP4kZ1bE7GDET62rWRHWpaSSRrh9qMTeB/26YM7YPzx+2+H8ZkRXCXDD/A1i5R95/duewLGlyzBx9/1yN0w3mjgqhpjY9Mw3ppMVi7Z/sAdsG1fP67wkVWycjNKJqzj4oxaR8nQjpRsfM/vzGau1fF7X4ziMJa4K9fZxJdRrZIRa/80V6ogmEqmtFb5oqdSScTy+WTlDBERX8lcN0w8zupKZRXLzsG44+J4GOHjBEo5DCNSKjYoqf1aNQFbiRvKtY9wc7EzSrUCONbwsQ0Tn7ulvVel6oFWySNWKpTEMe6wRnd/1Us4LCpmJitmpvNaBkp1mVC5p6aJ76lhKZUd2krlE+V1vWo9EbOVSjSRUv2k3cIVcAolfJzCQBkfP8Dn2Ko1YLwvU4HxRgu3bytVYdpKv8714ypULaXSiVXE92TWw9Wi3Dh5fQytsojg9ZcZ49fMZnD1oZHKIIy3Q3z8UFlrRl5nf11IYRx+wkxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCkWVCVjfP9OaeY6M2dDJcPQ95SsVCUTdnJaqTQxNQnjgbLn+jZQlcFSqmFkXPz2Mxkcjwyc1Xl0EmeSDvfgbE+/OoaP08bv1bK1rGJ8jatj+2E8ZySzikVEnBBf+1DJXDfbOCt69XBvIma0Z2Hb2dEpGG9Y+DVdJQPXVapkaCylYoCb7bxXXqO1oOM+n0VhJIZ03pdogRUcNIZSkcZWxqaTwWMhl0tWR3AcfK9jwZnlWjUJx9bOBfc928FxL8DXJlQy3U3tWiqFHBxljDjK/BEr1T+UYkDq9cTXDR9ErVahVBCJDaV/KO/JtJRHmo/7q2V2nntsKiVCFhkvE4qZ6bxXcYz7ta88XwMfV6BoNZXnjXKPXBc/RwX0GdPE91+rfGW4+H6bBj5Oo4XXL8pQklir7mTha1bO4/MJlYVWEOB4qFTeiSN8/pkCPp+whY8fg0kiMJRngnJxlFOUXA+ubmS5eMy3mriilR8qlUWidtef8VoS4SfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUYkFVMkQ86U7R1jJbA6VKxvQkzpB1lUz09atX4DPxcBUDGFeyN7Wsb8PE6Zum0n5Jfw+MD5RxtmdfCceVwgB65neM45XcRtxc8N7qmaz27yac3WqVcdayCbpTxlKupZKMnsniTFhDOZfu6g1zxzfwe1ISukXCIP3Pi9ixDO3OjPlIydrWamRoWeSmiW+IltWujVkTtTeUvrfAjwm0yiqOg/uqrVVuUeahWPncwlAqRGgsC5+PpcytWrWNSJn/tCoiJri52njVql5EStUAM1LmaOU4apEWtX2c+ufFyqvWxAw6qwcYgu9FoPTfVoCf0bVGFcbL+RKM5yzleWMn71VWqXrhB7higihVI5RiTWLUFzaWJMQVGCwb/0AzxJVFtP6by+XxyyolKCKlckmsjG0rq1RGaSafj9r8oFUBckQ5trJOabbxtTRcfG0KWXxtMrnO1221lGsO8BNmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKsaAqGdlCTrK5TGfQUjIgtWoHDs54LeRyMF4q4PZalYwQZL3GsVJJQf3ngpLNbirZ7zY+R99T9m1XKgAoCb6ip+Dia9xTKsN4bOB7ZZg4Hkc40zafw8dHWfqhUm1Ce01HycoNlXPRqiyYSlZxJqtkXHcdx3Dw/vSLURQm71UULaxKhijtLWVs6mMW31cUV5K2BbwdERGJIuUvlHelV25R4kpfNZRxqfVhbTrQro02e8Ta+9LuiZLxb8bJMaVWm1Ay5g215+B4EOB5JTbwnBsp85DfdZ6BVk1hkckZtmSNzuWB18bXpljE1S20ikSFTAbGs0o1DCvEBwqCZH+cqeMKHFqlG1/pR6GyBsiZ+NxFqQZlKJ9Jekplhkg5H8tRKkr4eExqn4UaIV7y2UoFFG0etsBcYzpKVS1lmRnU8TWotZR5RqlK1I7xWjCXwVUyglbnNfZa86+IxU+YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpZhX0t/xBI56M/kl7VZLSxjD8abyhW4tNUVLzNOT/pJf4H72kv6UxBTlKmpJf4GS9OcvOOkPX+MwUO7JgpP+8KtGypfsLSt57UMlw0pN+lOSw/SkP6XfKAlTnrJlaHfSX61xLOlvMW+Ve/y946Q/pc9oB1P6gZaMFSrJV4GS5OJ7IIFESTDzPfyaga+85gLPRQwlyUXpe9rrout+7Pg4rO4LrWwvHStbh8fKFt5aEiW69nGMT1Kdm5TJRrvGhtb/lEugJQkaZuebCn6+dfBiHffH33cLPKc95XlmWzh5qwWS8kT0fifKc3EhSX9tbStqLelPmX8sZQ1gGMo2ykrSXyi4vaecp3r+Sl6atl25mvSnPNMtpb02DlAyvxnic1FeUqSF+5MZKs90Zf5sx8ocEStbuXetWds/7+vzGfNGPI9WBw4ckBUrVjztwYieb/bv3y/Lly9/rk/jOcFxT4vVYh33HPO0WM1nzM9rwRxFkRw6dEhKpZIYWqkjoueROI6lWq3K0qVL1dJ1z3cc97TYLPZxzzFPi81Cxvy8FsxERERERIvV4vsnNBERERHRAnDBTERERESUggtmIiIiIqIUXDATEREREaXggnmeXvrSl8r73//+5/o05m1iYkKGhoZkz5498/6Zr3zlK/La1772l3dSRM9zb3vb2+R1r3tdapvVq1fLX/3VX/3Cr/VMxviPf/xjOeOMMyRS6p0T0dN7rtcDfL4/N7hgfp769Kc/LZdeeqmsXr1aRESuueYaMQwD/nf06FEREXnHO94h999/v2zZsuU5PHOi57d77rlH3vWud/3Cx+ke48eP/fKXv1wqlYr09vbKRRddJA899NDc31988cXiOI58/etf/4Vfn4ieG2jsHzcxMSHLly8XwzBkenp6Ls7n+y+OC+bnkOdpux7+YhqNhlx99dVyxRVXzMXe9KY3yeHDhzv+u+iii+TCCy+UoaEhERFxXVfe8pa3yBe/+MVfynkRkcjg4KDk8/lf6BhojNdqNbn44otl5cqVctddd8mtt94qpVJJLrroIvGfspvg2972No5xov/ktPUBGvtPdcUVV8imTZsScT7ff3FcMC9AFEXyx3/8x9LX1ycjIyPyZ3/2Zx1/v2/fPrn00kulWCxKuVyW3/qt35LR0dG5v/+zP/szOeOMM+SrX/2qrFmzRrLZrIiIfOc735HTTjtNcrmc9Pf3yyte8Qqp1+tzP/fVr35VNm7cKNlsVk466ST50pe+lHqeP/zhDyWTych55503F8vlcjIyMjL3n2VZcuONNyYG3Wtf+1q5/vrrpdlsPtPLRPS89nTjVUTks5/9rCxZskT6+/vlPe95T8eCtfsrGYZhyJe//GW55JJLJJfLydq1a+U73/lO6jmgMf7444/L5OSkfPKTn5QNGzbIKaecIh//+MdldHRU9u7dO9futa99rdx7772yc+fOX/BKED3/1et1ufzyy6VYLMqSJUvkc5/7XKJNu92WD37wg7Js2TIpFApy7rnnys0339zR5tZbb5ULLrhAcrmcrFixQt73vvd1zBurV6+WT33qU3L55ZdLuVxWfwuFxv5xX/7yl2V6elo++MEPwp/l8/0XwwXzAnzta1+TQqEgd911l/zFX/yFfPKTn5Sf/OQnInJsMX3ppZfK5OSk3HLLLfKTn/xEdu3aJW9605s6jrFjxw757ne/K//yL/8iDz74oBw+fFh++7d/W97xjnfItm3b5Oabb5bXv/71c/uaf/3rX5c//dM/lU9/+tOybds2+R//43/Ixz72Mfna176mnueWLVvkrLPOSn0v1157reTzefnN3/zNjvjZZ58tQRDIXXfd9UwuEdHz2tONVxGRm266SXbu3Ck33XSTfO1rX5NrrrlGrrnmmtTjfuxjH5M3vOEN8tBDD8lll10mb37zm2Xbtm1qezTGN2zYIP39/XL11VeL53nSbDbl6quvlo0bN3b86nblypUyPDzMX80SzcNVV10lt9xyi1x33XVyww03yM033yz3339/R5srr7xS7rjjDvnmN78pDz/8sLzxjW+Uiy++WJ588kkREdm5c6dcfPHF8oY3vEEefvhh+da3viW33nqrXHnllR3H+exnPyunn366PPDAA/Kxj30Mno/2fN+6dat88pOflGuvvVbdsY7P919QTPNy4YUXxi9+8Ys7Yi94wQviD33oQ3Ecx/ENN9wQW5YV79u3b+7vH3vssVhE4rvvvjuO4zj++Mc/HjuOEx89enSuzX333ReLSLxnzx74uuvWrYu/8Y1vdMQ+9alPxeeff756rpdeemn8jne8I/X9bNy4Mf6v//W/wr/r7e2Nr7nmmtSfJ1qMnm68/u7v/m68atWqOAiCudgb3/jG+E1vetPcn1etWhV/4QtfmPuziMS///u/33Gcc889Vx2fcayP8UceeSRet25dbJpmbJpmvGHDBniumzdvjv/sz/5MPT4RxXG1Wo1d142//e1vz8UmJibiXC4X/+Ef/mEcx3G8d+/e2LKs+ODBgx0/+/KXvzz+yEc+EsdxHF9xxRXxu971ro6/37JlS2yaZtxsNuM4PjYvvO51r3vac0Jjv9VqxZs2bYr/8R//MY7jOL7ppptiEYmnpqYSP8/n+zPHT5gXoPt7QUuWLJlLmNu2bZusWLFCVqxYMff3J598slQqlY5PilatWiWDg4Nzfz799NPl5S9/uZx22mnyxje+Uf7u7/5OpqamROTYr4J27twpV1xxhRSLxbn//vzP/zz116nNZnPu6x7IHXfcIdu2bVO/A5XL5aTRaKRcCaLFKW28HnfKKaeIZVlzf37qPKE5//zzE39O+4QZjfFmsylXXHGFvOhFL5I777xTbrvtNjn11FPlNa95TeJXsBzjRE9v586d4nmenHvuuXOxvr4+2bBhw9yfH3nkEQnDUNavX9/xnL7lllvmntMPPfSQXHPNNR1/f9FFF0kURbJ79+65Y5199tlPe05o7H/kIx+RjRs3ylvf+tan/XmO/WfOfq5P4FeJ4zgdfzYMY8HlmQqFQsefLcuSn/zkJ3L77bfLDTfcIP/7f/9v+ehHPyp33XXXXGLQ3/3d33UM2OM/pxkYGEg8xJ/qq1/9qpxxxhnq1zYmJyc7FvVEdEzaeF2zZo2IPDvzxNNBY/wb3/iG7NmzR+644465X8l+4xvfkN7eXrnuuuvkzW9+81xbjnGiZ0etVhPLsuS+++5LPJeLxeJcm3e/+93yvve9L/HzK1eunPv/3esDBI39G2+8UR555JG53If4518RGxgYkI9+9KPyiU98Yq4tx/4zx0+YnyUbN26U/fv3y/79++diW7dulenpaTn55JNTf9YwDHnRi14kn/jEJ+SBBx4Q13Xle9/7ngwPD8vSpUtl165dcsIJJ3T8d/zhjGzevFm2bt0K/65Wq8m3v/1t9dPlnTt3SqvVks2bN8/jXRMtPtp4/UXceeediT9v3LhRbY/GeKPRENM0xTCMudjxPz91wd5qtWTnzp0c40RPY926deI4Tsd3fqempmT79u1zf968ebOEYShHjx5NPKdHRkZEROTMM8+UrVu3Jv7+hBNOENd1F3ROaOx/97vflYceekgefPBBefDBB+WrX/2qiBz7vvN73vOeuXZ8vv9i+Anzs+QVr3iFnHbaaXLZZZfJX/3VX0kQBPIHf/AHcuGFF6b+muWuu+6Sn/70p/KqV71KhoaG5K677pKxsbG5h+UnPvEJed/73ic9PT1y8cUXS7vdlnvvvVempqbkAx/4ADzmRRddJB/5yEdkampKent7O/7uW9/6lgRBoP7qZsuWLbJ27VpZt27dM7wSRM9fTzden6l//ud/lrPPPlte/OIXy9e//nW5++675eqrr1bbozH+yle+Uq666ip5z3veI+9973sliiL5n//zf4pt2/Jrv/Zrcz975513SiaTSXwNhIg6FYtFueKKK+Sqq66S/v5+GRoako9+9KMdSXXr16+Xyy67TC6//HL53Oc+J5s3b5axsTH56U9/Kps2bZLXvOY18qEPfUjOO+88ufLKK+Wd73ynFAoF2bp1q/zkJz+Rv/7rv17QOaGx3/28Hh8fF5FjH+RVKpW5OJ/vvxh+wvwsMQxDrrvuOunt7ZWXvOQl8opXvELWrl0r3/rWt1J/rlwuy89+9jN59atfLevXr5c/+ZM/kc997nNyySWXiIjIO9/5TvnqV78q//AP/yCnnXaaXHjhhXLNNdekfsJ82mmnyZlnninf/va3E3939dVXy+tf//qOQfRU//RP/yS/93u/N/83TrSIPN14faY+8YlPyDe/+U3ZtGmTXHvttfJP//RPqb+ZQmP8pJNOku9///vy8MMPy/nnny8XXHCBHDp0SH784x/LkiVL5tr90z/9k1x22WW/cC1oosXgL//yL+WCCy6Q1772tfKKV7xCXvziFye+zvgP//APcvnll8t/+2//TTZs2CCve93r5J577pn7usWmTZvklltuke3bt8sFF1wgmzdvlj/90z+VpUuXLvh80p7vT4fP91+MEcdPqYdEzxs/+MEP5KqrrpJHH31ULTHT7bHHHpOXvexlsn37dunp6fklnyERiRz7x/b3vve9p91Su9szGePj4+OyYcMGuffee1P/0U1E/3nx+f7c4Fcynqde85rXyJNPPikHDx7sqNyR5vDhw3LttddyMBH9CngmY3zPnj3ypS99iYtlol9hfL4/N/gJMxHRc+iZfsJMRET/cfgJMxHRc4ifWRAR/efHpD8iIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElMKeT6MoiuTQoUNSKpXEMIxf9jkRPefiOJZqtSpLly4V01yc/67kuKfFZrGPe455WmwWMubntWA+dOiQrFix4lk5OaJfJfv375fly5c/16fxnOC4p8VqsY57jnlarOYz5ue1YC6VSiIicvrpZ4tldf5ILuvCn8nnszCey+ZhvNHyYbxcHoTx81/4azB+xuZzEjHXLcC2+j+gY9zeVH5ACysvEMf4+LaN/3VzZN9OGL/vrttg/KUvewWMZ8oDMP6zm/4fjJ9zzgtgPNczBONBkHxfRhTBtqKEQ3xpRDuMH4Yw3vbxgdotpX3XuTcbNXn3775gru8vRsffe19/Scyuvu95bfgzmQwe9339vTDuK8epN5owns1m8LkWk/epVqvDtqHSZ4IggPFWE5+jNvALRXwNHMeCcdvB03Auh4/j+/g8HRfPH46NX7e3D9+TZrMF45MT0zBuGMnjuy5+T9onONp7CgJ8r1pN3D9CZaIwDQfG47jzHkZRJPv3HVm04/74+163fkQsq/NeLVkxDH+mPj0N4+3Ig/G8i+9FcwavAXK9eJzl3eTaY3pUmU9CZT4p4H7ansb9qKE8V2IDty9ncH+PIhw3imUYL5bwOqvp4XEzNTYF434VX5+ScvyKMm+3vOT7XbVuPWwrDj5HCfA9OWGZ9g82fO3vvX8rjO/ctQ+fjtU5X0VRLIcOTs1rzM9rwXx84WdZdmLBbNv4EI6NB4XjKHHlmjoOvpG5HF54FwrJN53JFGHbX5UFc6GAzz+rLE7U9mBRISKSzeYWdJy8MqgXtmDG1yCM8TULlfa+8kC1lYnNsnB7S2m/mH8tefy9m6aRWDBr16W73XHdD9/jQiWuHUdbcFlWctGmtdXGn9ZeHffKwNeOYyrvVbs26D2JHFvULeg4yoLZURbqvo/ba8c3jGRcO3ftGkTK+FbvlXIc5d/bYoJzPHZ8fA8X67j/92e9mbjftvYPPuW5FSiLQq291r9sG98LdBx1DCjpWvrYg2ExlU90YkN73sx/zIiIGMoL28oYtpRxo42Phc7P6uuCx6i2thO8hBMx8PyTyWg/oK2blLlm3s+QY/PpfMb84vuSFhERERHRAnDBTERERESUYl5fyZhjCPgtpPIxtvK7sVj5dWKr2YDxMBqF8Xvv2wLjjpM8nzPOSH6vWUQkn8dfObAs/KsF7dco2veX9I/4tV8cYitXrYHxnU88DuNPPPYYjJ97/otg3Inx1xQy4DtiIiK29vsq9OtT5ZpFyq+wtH/BxZHyqyTl1zqu8p5i5Xue3S8cBtqvhRYf3/cSfTmTwd8lzmbx14S0X9Frv8bL5/HXhFylTwZB8vuP9Tr+DrM2LrW4/mtb3Pe0r6m5ync3Y6Wvtlr4u8RRhNv74CtRIiKVCv4qlufh75hqv8rUvgqC3m+1iq+95+Fz116z0ovP3VDmD+34cax8d1ppv9hlMk6y3wf4GV1QxrzfwN+XLRfx3JEx8ZiPskp+U0/yq4FWjNvWDozDeBzgeWnpUnwu4zX8nnJ5PLYLSj8t9i2B8Zq2JFPeV5+Dr+X04TEYN0G+gYiIVhxibPQo/gsnmRc2dnQCNp2YPADjw4P4q5322nX4NWM8/2hrSo3R9RUOQ1lbIPyEmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUiyoSoYhZqIAvJbd7Pk4A9tWMsszSgZ5pGSHVmdx1uudd96ciDXqs7Dt6ZvOhPH+PryTnVpVw1lYEX2NvqEJzoQ9ddMZMP7/fvB9GB/u74PxjFL0otnAme7ZEt4xEGWux8pV0PrN0aOHYTxUstwrFXyvtL0mtGxgp7tIv1IZYTFyHDdxv3p6emBbrZJCo4Ez7GMt81kZC+02zlIPg+RxtA1ENPoOgDiuVs1Rjq9Vz0A75YnoOxVqhfq1gv+B8r6mp2dgXNtNUasigu65IfgcY2WjhbaP76vXxs+RbA6fYxgqx1GqYXTv7qj138UmW84k+lm2rGyk08L3worws9tVNspylU3KplrTMN5oJPtGoaQ8o+1JGM9k8LqjtxdXcKj6eEw2ld1AlyytwLjEeJc7L8Y7uFV6R2DcmpmG8ZIyF1R9fE8GlWo0nlJ5Zxzsyrhr+xOwrW3hMRUpzxBT2/QIRvU1hq2sKfPlzvcahpGI4Koi8z0HIiIiIiISLpiJiIiIiFJxwUxERERElIILZiIiIiKiFFwwExERERGlWFCVDMsyE/vLa9nsoZJZXlcyvwuF5N7kIiLtAGd1hkoVjrHRQ4nYA0qmdWN2GsaHh/E+75Xeftx+BLdfunQpjGtZoNq11PLuV69bA+ObX3A2jN94809hvL+/AuOTE7gSSd/QChiPwD23Tfxefa8F4/fctQXGK0rlkhXLcT8bGsRZxabg9lZXFQgWyfh3ruskKk40mzjLu7vqwHGWjS9o93xynFZRQhsjAZgntKoOC6n2cOxccB/WqmdESlWKY9nYSZkMfq+ukuVtKZWGCoW8cnx8HN/H96rVwmNTlMozrpvMyI+VbP+4iefiwMfXTKs+kFMqLWiVUcIQvy5hYRSIdFU0cVxcScF08LM7VirjRCau4ODaOD66uwrjQ/3J180OZPG5KH03p1Rbybr4OKaF31OjhfuvGPg9FSv4hFohnldt5XlZVEpcZTP4fdXqeBz0VvB4ipXPVFvhVCIWKsuXvIOvZSGH5yVR5uHIxHHD0dZT+PDteuc9jJR5GeGygIiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUiyoSkYsyXoNvoerWFiWlqGOj621L7g4AzdS1voGyIz02zjTes/uHTBenZ2A8VYLH8cPcZbmySefDOPlMt6nPopwWmfbw69b7OmD8bxScWR8dgbG773vThifbeF7O7QUV+coliqJmNfA2b1bfoYrdowe2gPj+SzO+s3YSj9TKosoSfRixJ1/YWsNFyHfD8TsqiKiV6vA2eJZJTNe8NARS6kk4/u4T6Kxox1Dq5KhVeYoFnF1gGq1ho+v9B2t2oZ2Pvk8HsfaeYahUpkiXlh77Xy0SkZhkLz2cYz7ge/j+cBxcMZ8FOBrGUX4PZlKVR4tZb67H+uVihaXdiMWy+qsHlDID8K2fgtXdoi1ijmxMreq1x73x+45SUTEU/qXp1SuqdeV9QsusiS5LJ7HbBc/o21Tqfbj4/jh3fharlmJr1lOGZOWMtfYFh5n2prEyeEKF0EhWW0jP6FUt6jicynm8bUMlTneULqNVuWiWO6B8birOlBosEoGEREREdGzggtmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKRZUJSMKI4mkM6MwinGGYegrGa9K9qaWnVzI4CxNT6nOYWbBnujKsavVWRiPBGf3ej7eh73VwBm4Y2NHYFyrFNJWqnAcHR+H8UwBZ+8XlYzXw/v3wvienU/CuPK2pFwZgPEVS5aDY++EbZ94Yis+xtpVML5kyVIY7x/oh3EvwP3SVCoYxLHR1U65SYtQJpNJXDdt/Nm29m9wHPc93MlcB1//wMf3FVbnUKYgrQ9oGe3FXF45FzxP5At4ztIqt7RbOLM/l8fHiZX5Kau0r9UbMO41cLxULOLXVcZErVlPtg3xe83mcLUb08TZ+6gCh4hIrZp8TRGRTE6rroKvWXe/0SoVLTaFYiVR1aWnMgTbTh7ZD+NhC49tI1LmVmXqCHylqk2c7O9FBz8THRM/Q33lmR4rD2nLUqo1OXgMuxncrwMLV4jI91RgPFdSKuYoRWG0SjpOAZ+P5eDzKQ8Ow/hEUE3EChk8drQKOBltnlTKYQTKMydo434WefhZ0Wx0zh1RxCoZRERERETPCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpVhQlYxj6+vONbaW0ShK5qGvZLk3m7hCRKmA29tKprhlJdNGgwifY6icY7WO93Ov1pKZoSIieaWSR6hUAGi18PHrDRyfrePX9WencXwfziadmpyEcS/C7Q8rVT5u+NH1MD5U7kvEbBOn8a5YuxbGN206G7dfcxKMB6JkXGv90sCZvN1J0dqPL0aGYSXGuaNUsWg2cR/O53GWt1LARo1ns/g4LTB/GA7ue5Zyc/MZUGFHRDI2zox3XRzPKseJYqW6RRYfx7CUKgNKtQrbxtnojnKeolTlySsZ8xM1XJkiDpPzqK1Uvci4+JHTUiqFaDwf9zPTwnOxXh3H7/ozq2SIiJT7e8V2Ou9Vo16DbX0P9yO/je9REOJ+bdlKZZw2fj7FoEKXZeH7rPW7XBb3FzFwe6+Nn8WO8hxyXDwO4gwek0PLlsG4m8NjUpRr4yrVaCpZfJ4r12yA8VVrT4HxejN5nY/WDsK2mQw+d1OZZ0LlkR4r83altxfG9+4/DONRV8WchVTG4bKAiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhSLLBKhkh3PqGp7LkuoFqFiIipZJOGIc4gjyIlo1bJhkVHd5UMWSvGx2h5eH/5QNl3vqq01zKztb3LHQefZzaHM3mb0zM43mzAuKecv1Zoot7EWfFHxidg/IXnvzQRe+mFr4Bt3RyudpAvVWBc2RZeJMInrxfJUNp3hZVCBItSq9VK9OWskl3uujjz2bJw3w6VjHlt7GhxA8w3lo3PRRt/WqexHZzprs0TgoefmBbuVJaSFm5p7fHUKhLhagUZG1/7QqkHxrVCEc0mrmQRh8kfCH18kLpSmcNQniOFglIVRamqUa8rlTyUN5XsT6ySISIyOT4llt3Z0arTY7Bt4OHqGTNVfI+OTk7DeO9gBca1ezc9m7zX5RKuUDO4Avd11ynCeFvpvy2lkpWpzB2mMgfZyrynVZXSKnqFBj6Om8XHyedxFY6XvuINMO57+Do0brwtEZsY1yY+POaHl1RgvKnMq4aP32u+jO95qa8E4zNjs10RVskgIiIiInpWcMFMRERERJSCC2YiIiIiohRcMBMRERERpVhQ0l8sIt05U9qugobyRX1tW9cgULaRVhJNCsrWjw5KcFGyZNQEMOVL4FqCo6MkNJnqFra4vZZsMzHd/SX1Y7TEF0/70rxy/rEoCQVKYsLRqWkY3z+e3Ho70zcM22azZRj3A7zVp5aYqCXnRdoW2FoyYNcLaNeKjtGSsbStsdVEO0WlUoHx6ekpGEdJhbG2Xa2SxOeFShKf0me0RB9t3Gid2Fbmj8Rk+3N+CydIt3ycdKPNlUXtGitbIGvX0w+S5xOCmEjatcFqNXwuWj/TkkK1e57p2qJ4of30+arZqCeS/oIMTt6KlIT9fAkn1IUW7qfTk3jODpTEs9GjyediJo/npfLSPIzXGvg9HT2KE+ojJRG3px8nnk0piY+hkgjnDuHn4kwVX7NY2/JdmZsyTgXGDWUpeN3//TaM33PPvYlYXTnHXB4nPg6twucyNoGLCvTn8D2cHFWSUZUkQcvo7NeGMf8xz0+YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoxYKqZIQSJ0o5aNUEtGxj18LZykqyp1rxIaNsI10sJDMytYzRSPn3QhThrPKpSZy9WSrh7M1CEWfsajsxzsziahgNZbvXdlvJWlauvZZBriTjK7UzRFo+ft3b774zETvvxS+HbTefcT4+uHKvHFs59xhnwsYxPnutv3aHtWoai1E2m030nf7+fth2bEzJWFaqJmh90lIq27guHpuWmczQdixlu1qlZ5uC+5Jj4Xghg+cgx8Vx28Z9z1Xax1rFBuU4jUi5xlrFF6WTB2Cr67T2sZG8V4aJj2FZSiUEpX9o879G20ZZmxO7q2SEIatkiIicfOpwoh97Pq4cEXta1Rl8T30fV0KKq7j98DB+jgZgG2mrnIVtnQyeT6rKNt31ulL5o0d5rz5uPzo7DuOSwVVeepR1TdjA46DexM/iVhO3zxXxXDYxiasP7dm9B8ZnZ5JrkjBQ1nx5/JpavZyxSbwOKg7ia9OcxdeyPoWrdhhdpd0irdQbwGUBEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGlWFCVjPjn/+tgKlUyAiVbWaleYCnVM8JQybBU4hbIjLccfOxIyaj2ApxhWioWYbyv0gvjuRzeX77dxvvLTyqZttp71TLCNY5yHYIYv66WL24p93x0dH8i9tADd8G2Z59xFowbSjUMU9nvPRalAkCMj6MVDJCu45jKe6RjWi3ch103WaVGZOF9dWYGZ+RrVRNQn8woHwfksvgcTQNn75sRzsLOmnhc5jO4koftaBVacN8u5HBm/9DwchjfsfcgjFe1THplvqk3cHa5Vm0oAFnmSpGMBdPmLK0/afH5PkciVskQEZHTNyyXbNc42T+KKz60lKoXgYnH8OwormRh4WJTsu7kE2G82UqOp6OTo7BtzsBjfqCnBON5A89vofKslAC3Ny38XjMFHO+xlCpAOTx3HJzC1S1WrlsF48tWrofxjRs2wvhZZ50N43fdfW8yqFSiyeXx+shU1oI9RbxuipXqQL5SjcVU1gDSNa1q8y885rxbEhEREREtQlwwExERERGl4IKZiIiIiCgFF8xERERERCm4YCYiIiIiSrGgKhmWaYtldf6IbeNDREpWcqBWt8BZo1qxgkjJyGyBChQ5E/+7IJPFr6lliQ/09cN4b6UC4+0Wzk6Plb3LHeVaZpXznK0n93MXWXhmuRHhbHytToRj4b9pNZPVBLY+ArJpRaQ2cxTGB4dHYFzZpr67Zsu/x+NfrMqFUhBgUWo2W4mqIVNKdrbWV/N5nAI/OTkJ44ZSzsSycF91wETRW8TnMjjQA+PFAh5/WlfoyeL35Dj4OI6rVJnw8TxRVqpkFLP42piiVIJQRnK9iTP72x6uehBZ2uMieXytykwuh6sSNMHcIaJXYzGVOV2btbTqLc1G5+tqz5XFptZqid/Vn3LFMmybdWowHsa4qtRwBcebNTwO2jU8y7tgks5HyhibxP0r8PD9NkLcj4pK5asYVOcSEXGU6l+5DD7OQAm3tyJcCaJSwtUwVp2yCcbXrcTxWg1XNDnjTNx+1Zrk6z65/UnYds3aIRgfWVGAcWU6FF9ZT7V9PEeMLF8C44VC5xwUBIEcHb0bv2gXLguIiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCjFgqpkmFaySkasrLkjpUpBoNQ18JTqGbGS+W3Z+HWnasnKEbGJM0/zBZyx3deD9z73A5yp6vk4q7zp4ezNlpIGarv4dpQLOBt/ehpXKgiV0hGRkllumzgbPwpxBrGptM9kMonYvv17YNudu7fD+NLlK/G5tJWKI0p/ig18jnrxjM5rY7BMxpxCvpCoShAoZUsiPFwlEDx2wkip+KBUZMgr1TOG+pMZ/CuW9OGTUc7FtZSBE+M3lS/gPqL1nEDJ5j5x9VIY7x3E89D9Dz4G4z1ZXAkiVqb5g2PjMB7F+B1YyjtzQeWSwMfXTCtCESn9QIsrBX/EUir4eErlj+73FCkVjBab8amGuJnO56arVZWaxtWaLAvf7N4i7qdmiO9RzsAVJcpgfESzuBpGbCvPsjJ+TthlfI5+jM/RDPDYqM1Ow7hS9EKaM7hCV+8wvga9pWEY95XqH40ZfK8+c83/gPGzXnAajG88eV0idnR0FLbtGU6uC0RE4vwYjE+P4mt/ZNcRGF+qzJM9QwMwbpid19JX1m8IVwVERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJRiQVUyjJ//76liJV1ZqyhhKlnuWqUJ08THbylZzz4og5DN4goLLaXygqVUSAiU1/RCnJmrXYNAuzZKFYuSUiVjsL8fxmcb+HwaWjaokrluaCUllERy205WI5mtzsK2W269BcbPOvM8GHccnGkbxkq6scJQ+l+ceK9qOY1FZ+nSZWLbnVPFkSM4Y7lWq8G4YytVL/JFGPfauA+XizhTf8XSZEWMkQFcBcdxcGZ8GOHxEQa4ukUxh99TLlOAcRtUkxARGRzG1Ty8GA/Mc885FcZna7gyxX2P7oHxnftxRr5h4yz1KMDHR+/L9/C4nJ3B/UMbl6bgaxYo414rchH6uL3jds8rHPciIpZTEMvpnM+tAM/BpjJu/FYVxvt6cPWCkfWnw3js47lg9MD+RKxSwPOD4eB4NosraLWU6lztlgfjhUFcxSJXxMcv9Q7CeFzFrzs6fRjGpyI8D2eb0zBuC54T/Toel/fdcjN+3Xbynp96+mrYNpfB116rGmQHeMznlQo4g33JCkkiIr29+NlSbXTeQ22eRfgJMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpVhQlQzf9ySKOjMK1SoZSkWG7iobc+1tHM/lcZapr2Rst7xk5rdW9SIK8TEcE18Wrb0X4gzsUGlvKdnymmwWZ5kODeJMW2NyCsb9GVyxwg+VCiVaRQmlTEYEUtSjCF+DO++8Hca3vepRGD9z87kwHnhaiY+FZbt3N1/gjz+vNRqNefdZbdw7Ns6wN0J8/0pKBYqVK3CGfbmUbG9ZuO8VCjij3bJwdQjbxlVqSkoGvCH4OMPDIzAexDjzPqzWYTzj4tctlnB1jgPjeNyH9+N7lc3g6+N5+DyR7qoqx0UhHlha5SBtrskkqlv8/PhK9QzTwdes3fWeIuV5ttg0Z6cldDvvoe3gai5mG/f36lE8tics3K9H92+F8anJURgvZ5L3tKdXqTijzOdTE9MwHiqVE0IL96/p1gyM9w/jMWkaeOw1QzxWh5f0wnjf0jUwbkS4v9eO4moYp61eB+P33HUrPn4pOf5WrFoG20qEq/E0RnE8Y+FzX7YOz/1eGx/nwEHcb3K5znP3lQo6CD9hJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFAuqktFsNsWyOn9Eq5KhZVR3//xxbSW73lf2FQ8jnMU6BSpBTExOwLZ5ZY/zSk8PjDvKuWt51VplgUwGZ3hrmeWFHM7SdzP4GjSU/e5n600YD5RrrxQ0UauFhKDigePg9zQxOQbjN930ExjfdNoZMB5HSiUPE98VLQmeVTF0Y2NjYpqd/7Z2lKoD3VV0jtOqZ7jKfRoo40oNWQe3t8FQMw3ctpDHWf3FIh5ntlLBR6teE8V4fIcGfk9i4ePk8kUYN2M8rzTbeFw+8cQ+/LqysGo92lyP7rnrKNdGqcizkAocIvrcGiv9z1Rmab9rLtPe42LjHWxI3DV3V72jsG1+CFdqyGVxhYjdu3fj9oUSjK9YgqvLLB2oJGKtVgu2PXgUV0yo9OEx5hT7Yfz//vBmGN+75wg+PqgmISIyOIirXrzqklfB+K//+mtgfMnIWhj/t3/7KYw/+PD3YXxNHp/PiSeeCuPbx3cmYo0ZXG2i3IPnvVKxDON2hK9ZVlmtzk7iz32zBTwfnnBC5zVrtT35rtyBD96FnzATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKwQUzEREREVGKBVXJ8D1fQqszC7k7e/5pKeUIggBnN2vVM1ptnFVdrSX3qbct/JphgLMoXVfJos/jrN+FVsPQsuvVa4kvjcQBzkp1bFzBwDYXllkeK5nl2vtF9UIMpVKBofSDO+7Ee9e/+pLXwfi6E06B8UDJxo9jpapG3HkNtGoPixG6Fu12G7ZVqxfEuK8O9ePs7L4SPk5tehLGs0Yy47qv2AfbauO4t4yztqMYzxOWjTPsK/3LYTy28LxiZfB4bVTxNb7nzrth/I6774PxJw/g6gaZLD5/re6PVkECVUDRhw+e4/J5XKGk2cSVfbT+l83hOTdS5kq7q79GrJIhIiI7Dx0R2+q6Vz24/566BI+boXwNxhvj+F4UrEEYzxXweJ1tJTtZGOIlje3iylcDA0th/MwLLoDxsfEpGD9j/QoYv+/h7TDu5PE1e/Nb3gXjq5bh41//r9fB+OHRwzDuKWuPg/VpGI8zeDxUZxqJWL4Pz5OnnLkKxnftT1Y0ExHxfFzpZDBzIoy7PfhZPxPi6mgHD3euET1PqRIG8BNmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKsaAqGWEUKvnTSVpGdaBUprAEVy/wfRxvNnEmZRAmM3BtpWqEYSoVE5R3mVMyuTMZnD2ccXHGtlY9w1eyNZvNZEaqiF4pBL8rEUt5v6YWt3GlAsfB3cYDpxP4+H5rpT+Ojh2B8ZtvuQnG16zFVTK0KhfzLX4RhayScZzjOIkKLmGI76uhVHoxDdy3czhpW8p5ZUxlcGWHLKg04SpVKWwTx+MYn3tLqdRQKOE+4ipVL0r9QzAeKlV8tu98CMaf2L0HxieqMzDuK5Vqsi4ex5GP5xVtTsd9QesHSlyZg7S5Mghwf0Lzv4hIoFRb6n4GGBz2IiKyduMKcbr6h5/Hz9xm/QCMhwMnwXi5gl+zaOP+qPWNVi35XLSVZ1ZOqZ6SLeEqPY0Wfq99BTy2+zO4wkf2fLxmOPeCi2F8wwnrYPy6674P4w88hOeIN7/1Mhhff/J6GP+7//PXMD7S2w/jq09Zk4hN4Usm+w9Pw/ihA2Mw3pPD13jCw+1jA8/npourq0xXu6pkKHMDPOa8WxIRERERLUJcMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkqxsCoZYSjdidILy5wWCaI2jnfvW/9zRowzJptKFmt3Nv+xc8Gpz16MsyPbLZwlLlr1jCzOwNWqVRhKhYgwxK9ba9RxXKkU4kf42ptKNr7t4MxiU7knjoPvCbo+SveQINCy2XH8ttt+BuOvetWlMN4/vER5XXyNuzOxtSobi5Ef+IlxZSid27bwlGIYSh+zcR8rlnB2eamAM59z2WS5DVvp74bSr6vVGox7LVwlw1O6SLaCjxM6uCTIrt27Yfyu22+HcdfB72vN6uUwPv7YkzDut/H857fxGIxirZIFuPYOzlyP8KHFQyV2RB+H2nNHIhxHzwURkWxXv4miSGRqFh97ESn25BPVXrwmHpOxicfk1GH83KofOATjhuDnVrmEK+NUKskKDiUwD4iImEp/acX4Gao9W1shfvaFbdy+XOqB8bPPPhPG68qz/v4H7ofxQg5f+1IRx2eruNLE0hUDMO5mczDuxcl71fBxNa9IqYy2ceVKGJ+YxPNSo4XnAsPA116bz/2uqhi+suZA+AkzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGlWFCVjCAIxDQ7s00NJV2+OxNxLh4qGYkZnFVdb+EMS23P+GI2mR0aKhUZbBMfo1Kp4HgPjmddfBmbTZw16ilVLFpNnI2vVcmYbeDjGya+J9337jhLqRoQK3U+IjUTPXk9LQtntkZK5RLLxsfeu28HjN9zD64kcMmv/xaMa90v7Kpc4vusknFcHMeJqgS2rVVK0eAxEkR4DNourjyTy+H2q1cmK0SMj0/CtjNNPJ6MCM9ZOQu/12yuBONOBlcTuP/+h2D80QdxvD5bhfE1q3B2ed/QIIzHJp5b73toO4xHgTJPOPge2nby+JaJ2/ohrpKkVd7Rqi3V6zgz3lKeC9ocV+iqwKC93mLTnhTpLlDlG/galor4nprhBI5n8dy6bNkIjDsOngu82eTrHj40DdsqXVqaDn7mDq8YgvFcLz6XQ7tw9YlVQyfAeP9AssKHiMj4+BEYn52dhvFiGc9Bs9P42h86sB/GlSIyYhr4whlgTVXBQ0+iJp5X7UIFxgtZ3H5idAofJ4PvSaxU9Ym6FgHdf07DT5iJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCjFgqpk+H6YqLQQxzjjVauSYVg4ldJQMnC1bPxyEWeiZzPJveRjpapDqYD3Wx8ZGobxQrEM414LV6toezj7sqFUw6jXcfa+r+zFLkoVC+1aqtdYqQIQRlqFCyVzFqTaatnyoVI9IxZ87GYDX7Obbv4JjL/gvJfAeE6pbBCFnece4O67OMU//+8pQqVP2hl8v/N5PNZ8D/eDeg2PqaUjFRiv9A0kYkeOzsK2PSU8jkMfV6WYGT8K4/lenEl/5513wfiuPQdhPFIqM6xfvw7GT1iDq2TEFp4PXnjeZhhvNHGlia3bcEUaLZU+8JPn73t4HC+0CkVPTw+Ma8+XVhv3GxE8x/ld2f6hUr1nscn1GOK6nf3J8XD/EgdXyWiaOC4F5blSwnN8aON+arSTc02gPIsdF8/7rRYe8wcP7IJx28H92laq9xQqeJ0yOn4IxmeV59wJp66H8cEKrozjNfG1P7hvH4zXGzP4OAG+nqgCWE+hF7bNZPAYnpnF77XdwufuKtUwtAo4kVLhw7G6+s0Chjw/YSYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohQLqpIRR6FEXenyWtazpVTDcLPJKhYiIrkcziZVEr/FsXHGJDqOVu0hVyjCuO24+EWVqhQeyBIXEdGKW7RaHow3mziuVRbJK9csUN6vmPg4LR9X89DeV6Jkws8Z4GUjpUJJpJyj7+NrYCoZ+tsefxTGf3zD92H81Re/AcaNrqEQzH97+ec9I47FiDvvY6mIq15ofdJUqqXYBu6TWmWYQOlPu/cdSMRmqriSQs8AHt+5PM6kz4zg17zvvvtgfHQaZ/WLieesSgnPQyNLcAa8YeLOefQoruYROXjOveCFuHpGIYvvyZO7cGb/xEzy/aKKOSIirouvfaOBs/FnZnD2vmHguTiO8b3S4t3zitJs0XGKIm5Xd/XH8XhqNGsw7ka431k27r+HD03DeCR4LijFyT5WreGx51g4bs4oC4wYt+8p4so4Ji7CIrGB5719+/fAuK9UHVt/4hoYH8jhChRHxg7D+NTsKIx7Pq7QFcR4vBqgMkXvQB9sKxG+xoGy7rCUuSNfxP0p8HBVjdBRKnR1XePuZ1safsJMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpFlYlI44TacQZbX9vJdMxNnDcUOKBjzNzW8p+47l8MkvfsnEKq6FUjdCqUkj3HuQ/ly/i7HrtvYZKUmbLw9nAGaWyiCiZ4lpVEC/E17Lp4WxVrWqHVhkF0aphhCGOBwE+R4m1ago4u/6222+E8VNPPgvGly1d33l+LJMxJ593E+N5SMmIzmp9VenzGQPfb0sZa/U6bn9kNFklY3R0FratKVnVy5fiahVrlldgfGh4GB/fH8fxJj53W3mvkZIxH8V4/OVzOKPdUuZo08b36sXn4jGSyeDKKLfc8UAi5ilzmesqc7Eyl2nVM2Ll2mjVMLTqHFo1p8XO94LEPXEH8D1ybHwNozbudxLguDeFK6K4BdxPozDZl/w27ndZG1fviSL8jPOVQjczwRQ+fqEM455SbaPdxnPQ2CSudOPGuOrFXY8/hl9XGdu2jcdNvY6vQ30aj7/BQVTBBz8zHQu/ZrFfqaikzAWmUvyrOqtU4grx+eSLuY4/e56y5kDnMO+WRERERESLEBfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUiyoSoZpWWJ2ZctrWcZxhDMXfSUjsRE3YdxStnr3fZwNG4HXtZVqGLHggxsGbu84SkWQTA7GQ20PdSXbs7jAzHLbwXFL2UN9bGICH9/Bmbzd9/o4X6lcgjLUTeXaa1UToghfA8PAP2Db+BqjqgkiIvsO7IbxwcFlHX/2gjo+wUVo2dJBsbsy4SsVXFEil8NjoVLGWeSGlo7u4/nA83B/aoMuGSqdbP9BnInuOLj98AA+9w0bNsB4zcPjZurJPTBeVypBVGdxHyzlKjA+1N8P41oOuO/h92tlcUWJszafBuNHQXWDBx99AraNYzw3aXONVmVHm0AWWvWi2ezsZ/rrLS6ZZVXJZDvvVbGMn3/NWfwcalfx3Jw3cQWHfAX3uyjAxzfBXDCyZgi29abxcyXO4P5iZnE/MCN8DdZuPAfGQwPPNc0IV9vYt387bl/D7ccO7sOvG2trA6XaVEupLhPh46DCJVPT+L2evHoljA/2VmB8qqpUN5pVKqOV8bMoX8FVOJpR57xq4MNC/ISZiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRigUl/bkZJ5HEpSX3Bcq2hL6vbWmKX7OYV7Z1tZS1PtiOWsvjCLU9qpVEtWxO2cpRSR7UXldLKrQMfDu0bYJPWH8ijDsZnDxx+x13wPjRcZxQIMr5ZDLKVr4gYU9LWJxWvsCv9QNtK1ytn2VNfE9cLenBb3T9GSedLUbr1q1M3EdtC2Jta+y+Ck6cEx+3b83iPhwbuP3U9FgiVu7BW9YfHsX9fccOnEDTW8T9/aT1+Pgb1q+H8dkq7lO+sk1utYbbTzl47jN68dbVpUovjMe+si1wiOO2MqZeeH5yK+2G8p527MDJuNq415IBRdk+t0e558UivjZOV+J0GIayd89B/JqLSG+/LdlcZ78PzUnYdjCPE+2mQjyGW1Xcvxpt3N8zNu6/Dng+aYmJcS/uj/ke5VmsJAD3FE6F8d4B3O927X8UxvfuGoXxA0fw+BgYwtfgxHOGYfzwHnz8w/twgnFfZQDGlw3g45vZZCrx/iPjsG3Tw1ueTyhFDpomvld+Bo9538fzs5Od35bZsbKlN8JPmImIiIiIUnDBTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKMWCqmT4vkh30rKa3Wzj6gh9xR4YbzbxdoiGpWyVbCsZkOjfACZ+m4YSz2XxVouFopLpH2vbOeOsTm276FwWZ3IvWb4Wxk/fnMxOFxFxXZydPDC0BsYffmwbjBsWPk5fpQLj25/Ymojddc8W2DaIcca1k1OusfJvu56ePhg/e9PZMD7Ui+9t5M12/ZlbYx8X+YZEXX1Z2x49n8V9vqZsd+q18dbY/b04OztStmpdunR5IuZY+BwztlJlQvA4np7BfWF8Emd/5/I4Y/70U/BW2uPjOLu8rlTVmKziTHdDqY4TCL7GGRdXFHCUezg7NQ3jg6XkdugXX3gebHubez+M7zlwBManavi9ull8jbPKtbeVaj2FQuc1CIL5Z8w/n5XMEcmandfswe37Ydv1a3G1pvVrcbWYR+97BMa9Kh5PgyfgNYM3k1wDjI3jeaZ3Ba6yZOXwGDMcXKmhbe6G8anZ5LNPRCRuV2F8ZnoCv66ytXuhH49JK4ePXxrGc9nuJ/FcUKko25iXk2NbRMRyk/dk+RLctm3WYLzawFtpOwaex5wMnq+iCI9Zz8UVlQyvq9pZiJ8TCD9hJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFAuqkrHp1BeL7XRmMA4N4X3k+wdwZqtj4WzlH/zg2zAexjjDslDA2dAuyIzM53FlhHIJV2So9OLKC1p7U8mur5r4vRbLOF7pxfu5r1m7EcZ7+/G1V4pzyKmn9ePjn3gqjFu2kkVv4W5z0vqTEjGlWIrccfct+DWV/rH+RFxhYEjZ6/7EVbiyiMQ4o7be6MzQbjZxdv5iFASBmF2dylbuk0Q4y7sd4Oxs38Njp6fSC+OmqVSCAFVN2nU8d5QK+NwdJas6n8fZ36ZSZWd2ZhrGczl8nGIhD+NRgCsBeRG+Zo5SOaKtXGNTqSJiu0qmfkHJmLeSn7ssGcBz5atedj6M7z88BuNP7NoH44eO4Co7Row/A3KV6gNuV6UQrYLRYrN6pCz5QudYq8/iqhcrevDzY1jp10dKOL5EqeKUH8FjfhzcqkYNj5lKL34oBhHuL9VpXG2jmMX9cWwKV9Uo5pfBeF8FP1snx3bg448fhnFfmeOKZXyNs0V8LcXEVUSWrsZrjKnZqURsoA/PPxNNXIkkVOalnIPnYc/Dz+SmUqUsUopfuEZnHwmM+VfG4SfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUYkFVMl7+8tdJNlvoiK1auRK27akUYHxqAu8ffvttN8N4GONsz75eXPHBspOZ3Hklu7u/H1el6OmpwHhRqcwR+bgCgFHCmbBDy1bj8xlcAuO5HM44j5WMcDFw3ATZ7CIihYKWGY4zi0MPZ5X29iTP88IXXwDb9vfiyiWttgfjw0P42rgOvsamgyshzNZw9nPc6Kyy0GrhzN7FqNxTkIzbeT2jEPelUqkC476Hr+eMNw7jk5M4binVDjxQJcNVqj1kMrhvZLJKVZsi7qsaR+l7mQzuq1pc+zxjchZnxreUcWkEeExlsvj6OEqWermM20dRMsPea+M5MZ/B2fgnrcPje82qERjfsw9XDdi1E8erDZwyX6t29psgmH/G/PPZzoMHJJvr7Me9Rfz8G8zh58T42EEYLw3iKlR9FfycqytVZ3rLyXHTO4Tvs2XjY3h1PDbaVTyGoxm8ZJocxWPDsGZg3DHxGNPm1ckx3C9bU0qlGx+fZ28Jn2dzBo9XN4PXcbli8n3tn8AVRBqBUiXDx3NBS6k6Nq1UBPHa2ue++NpkM53XxvOUchoAP2EmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUggtmIiIiIqIUC6qSUS73SC7XmTGOd24XCSP8N5kszroc6MfZ0EGMqxqUixX8wkbyLRWUvesH+gdhvKeMj+0oFRkC5b0OKccfWboGxg0bZ86GykWOY/wXC/1XUKjsxW4orxsqmeR1kMXqtdqw7YqluLpKS2k/M4MzZAMHZ9o6OXzykzMTMO75nZmyWpb/YtTf3yvZjNsRGxubhm21KgNhgPtYpdSDXzTC7YMQZ7XnssmxaSjHyGRdGC8o1XRME4+oej1ZmUNEpNnEWeGtFu5TpRKuwuEqVTuiKq5KUK01YHyoT6nuE+HscEMZ+KgahohIHCfjjqNU6oFRETHxvSoq96Rn42oYX7sczyuPPL4Lxm+/7+GOP4chPo/FZseBSXGznc/SoZ4qbNvCw0AiB4+zwMDPUcPH8ZpSxaAZJ/u7EeL5pzaK1xGeUpUpbCvP+llc9WLsEB7z7Qg/t9ZsxGuAoWW4+le7js8nygzDuKfMNaNHR2F8dgZfh1vuuRHGe4eSc9PomNIR8HQlfhOfo2XhMegrq82ZKn5dF6wFRUS8Yuec4ivVhRB+wkxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCkWVCXDtI/991SR4AzD2MCZjlrGeU+lAuN+gFMss1mcPZ3JJjPOR0ZwtYrhYVyZI+Pi7N5Y+fdFT/9SGB8YxnHTUjJecRK6KJdMTAtfG629oWSZWkoaq6mcTwCy4kVEYlCVwHVxpn8Y4fvXaOKMaNvG90Tb695XSovMVnE2c7XRmQHue7hax2KUz+Uk21WFolLBfaBaxVnhpRzu88U8jmsVJSancJb6+NGxRCxQMuAd14LxJUuGYLyvrwLjroOnT22O0yowaNU2LG0eMrSqHfia2cO9MD45fhTGS2VctcO28XULUVeIcFtD8P3WKv6EyjhUChOJpVTn2LTpRBjPFDrPp+158uhj2/HBFxHTisW0Om9sFQ8naXnT+BhKf6m1cDWoShn339jDY35qFj0r8P0PfPyMa/n4+eQolaOCAD8/wiye93r7lGe9gy+m38DzqjeGq3xo7WttfJ6abBHfq90H98L4vkPJ9raNn8XFIr4GjTa+J34b329bubezM/i9Kt1PbLNzfvM9rdZbEj9hJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFAurkmG5YnZXKzC1Kgh4Le77uApCpVKG8UYLt3eUShMDAwOJ2OAgzn7PuPgYto0vS6GAz7FnYDk+TiYP43GMs0OVLdfFUu6Sq2T7G0qWvmEo1TCU8xGl0kTo48ooGVARo6enAtu6OXxfxcT3ZHAYVwzQ7snRiWkYn67hbOOG35kVHcVKiu0i5HmBmGbn9TCVy1Ms4Qz43jKOa1d5egpnPo+NjuNzbCWz2k3l84AoxPE4wgOt0cAZ7ZkKnvssCx9fq55hWvgqBCEeI5bS3lDi2TyuSLPuxLUwHmvlepQqOxknOTa9Nj73KFTmPhMfOwrwXGMp1zIycFUN28HXZsO6zkpGzRar44iIOG5JXLdzPFRnG7CtGVVh3FAqlsw2cHvfw3N8X28JxrP5ZJWr2dou2DZfwWM4ruF+ESlFJqwc7o+9dnLdISJiZ/C8NzOLr0FbqRJlKdXIGgf2w7hj4ePEFj6fqIivsWnguAGmvnwPPsdyGc8/7aP4WT9bH4XxotKfBnO4qk9OeV2n0NmPfQefN8JPmImIiIiIUnDBTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKMWCqmSIYR377yliwZniQaBUWIhwfHBwGMYPH8F7tJeKuDpCpScZN5X6E1rWuqNUtyiUcSas5eBsTy/A+9EbWj2MGF+b2MNZ65GSRW8YC8zSV15XlGR5T6l0IkayO5kWzlTVMvqVpGIpl3tgvFDAGbLtEB+/r4YzvcUqdP58W0mTXoTCKJCwq7JEVqnQknGVzGQLdybtX+xFpbKDrfwEKr5QLuMM72IRH7u3gtsrXVWyGZzV32y1YDyKlXGszIlND2dvt5R4JqtU7bDxG4gjPD9pFS5sBz8uXFCBIoqVKhkGPhdTmWzCUKvYgedQR5n7ogDfE8fsfE+Bia/JYlOq9Egm19mfGjPKQIjxNQsjXJmifxD3jWwG3+vpKo4HQXKOtq2VsK0R4nPp6TkI420T969QKjDu1WFYRncfgfFImQ97RnphvK5UlYqW4/ZhjKu9eG08PgplPLY3nKxUz8gmn7uWjddBhlIZp9KD5+GJXuV+1/GzW6tulB/Ba8RmfazrAMr6B+AnzEREREREKbhgJiIiIiJKwQUzEREREVEKLpiJiIiIiFLMK+kv/vmXopvNZAJepGyjqm0/6zfxt+NbLZxk5Xn4y+vtNk7iQMdRdoQW01CS7JStot0M3s4yUP7doX2XfKFJf6aSIGApiQla0p+htFd2pVWT/rRrX28k+4fv42SLhofjraayta2SuKRt19sE5yIi0mrhxIHuJD/v53+OF5AQ8Hxz/L2328l7ZSv/1I6VBLZggUl/LfCaIiKesi27D7ZQ1tq2PZxAo72mlvTntPD0qR1HG5dKbrC0lOTadhu/L9PC177RxOM1MLSkP2UMKieKcgcbyjjW3quhTDaBMv9byt7sXoSvTaQc3+y6Bse3xl6s435uzIMtmj2l38EOICJRqPRTZStiI1LuaQuPgyBIxi3lmS5aEqrSX7yWlvSHj6N0U/F97drg/ugr19j38HEC5fixkmAMLlnq8dst5R6CrbotS1l3RPhatpX5U7sGgXKOsXYPleRlv915nOPvfT5j3ojn0erAgQOyYsWKpz0Y0fPN/v37Zfny5c/1aTwnOO5psVqs455jnhar+Yz5eS2YoyiSQ4cOSalUEkP7uJboeSSOY6lWq7J06VK1HN/zHcc9LTaLfdxzzNNis5AxP68FMxERERHRYrX4/glNRERERLQAXDATEREREaXggpmIiIiIKAUXzEREREREKbhgfha99KUvlfe///3P2etPTEzI0NCQ7NmzZ94/85WvfEVe+9rX/vJOiuh5Jo5jede73iV9fX1iGIY8+OCDz8l5vOQlL5FvfOMb824/Pj4uQ0NDcuDAgV/iWREtLv+Rz/2rr75aXvWqVy3oZ8477zz57ne/+0s6o8WFC+bnkU9/+tNy6aWXyurVqxN/NzExIcuXLxfDMGR6enou/o53vEPuv/9+2bJly3/ciRL9Cvvxj38s11xzjfzrv/6rHD58WE499dT/8HO4/vrrZXR0VN785jd3xO+44w552cteJoVCQcrlsrzkJS+RZvPYJjwDAwNy+eWXy8c//vH/8PMlol9Mq9WSj33sYx3j95prrhHDMDr+y2azHT/3J3/yJ/LhD39Y3WSO5o8L5l8xnrJDXqPRkKuvvlquuOIK+PdXXHGFbNq0KRF3XVfe8pa3yBe/+MVn9TyJnq927twpS5YskRe+8IUyMjIitp3csUobp8+WL37xi/L2t7+9o27oHXfcIRdffLG86lWvkrvvvlvuueceufLKKzvavP3tb5evf/3rMjk5+Us9PyJ6dn3nO9+RcrksL3rRizri5XJZDh8+PPff3r17O/7+kksukWq1Kj/60Y/+I0/3eYkL5meoXq/L5ZdfLsViUZYsWSKf+9znEm3a7bZ88IMflGXLlkmhUJBzzz1Xbr755o42t956q1xwwQWSy+VkxYoV8r73vU/q9X/fPnz16tXyqU99Si6//HIpl8vyrne9C57PD3/4Q8lkMnLeeecl/u7LX/6yTE9Pywc/+EH4s6997Wvl+uuvn/skioiwt73tbfLe975X9u3bJ4ZhzP0256UvfalceeWV8v73v18GBgbkoosuEhGRW265Rc455xzJZDKyZMkS+fCHPyzBU7bxrlarctlll0mhUJAlS5bIF77whaf9Fe/Y2JjceOONia9S/dEf/ZG8733vkw9/+MNyyimnyIYNG+S3fuu3JJPJzLU55ZRTZOnSpfK9733v2bsoRIvEfJ77U1NTcvnll0tvb6/k83m55JJL5Mknn+xo83d/93eyYsUKyefz8hu/8Rvy+c9/XiqVSuprf/Ob34RfnzQMQ0ZGRub+Gx4e7vh7y7Lk1a9+tXzzm99c+BumDlwwP0NXXXWV3HLLLXLdddfJDTfcIDfffLPcf//9HW2uvPJKueOOO+Sb3/ymPPzww/LGN75RLr744rnBs3PnTrn44ovlDW94gzz88MPyrW99S2699Va58sorO47z2c9+Vk4//XR54IEH5GMf+xg8ny1btshZZ52ViG/dulU++clPyrXXXqvuYnP22WdLEARy1113PZNLQbRo/K//9b/kk5/8pCxfvlwOHz4s99xzz9zffe1rXxPXdeW2226Tr3zlK3Lw4EF59atfLS94wQvkoYceki9/+cty9dVXy5//+Z/P/cwHPvABue222+T666+Xn/zkJ7Jly5bEPNLt1ltvlXw+Lxs3bpyLHT16VO666y4ZGhqSF77whTI8PCwXXnih3HrrrYmfP+ecc/gVLKJnYD7P/be97W1y7733yvXXXy933HGHxHEsr371q8X3fRERue222+T3f//35Q//8A/lwQcflFe+8pXy6U9/+mlf+9Zbb5Wzzz47Ea/VarJq1SpZsWKFXHrppfLYY48l2nDMP0tiWrBqtRq7rht/+9vfnotNTEzEuVwu/sM//MM4juN47969sWVZ8cGDBzt+9uUvf3n8kY98JI7jOL7iiivid73rXR1/v2XLltg0zbjZbMZxHMerVq2KX/e61z3tOV166aXxO97xjo5Yq9WKN23aFP/jP/5jHMdxfNNNN8UiEk9NTSV+vre3N77mmmue9nWIFrsvfOEL8apVqzpiF154Ybx58+aO2H//7/893rBhQxxF0Vzs//yf/xMXi8U4DMN4dnY2dhwn/ud//ue5v5+eno7z+fzcPKK9/tq1aztid9xxRywicV9fX/z3f//38f333x+///3vj13Xjbdv397R9o/+6I/il770pQt810SL23ye+9u3b49FJL7tttvm2oyPj8e5XG7u5970pjfFr3nNazqOfdlll8U9PT3qa09NTcUiEv/sZz/riN9+++3x1772tfiBBx6Ib7755vjXf/3X43K5HO/fv7+j3XXXXRebphmHYfhM3jr9XPLLd/S0du7cKZ7nybnnnjsX6+vrkw0bNsz9+ZFHHpEwDGX9+vUdP9tut6W/v19ERB566CF5+OGH5etf//rc38dxLFEUye7du+c+QUL/quzWbDYTX/b/yEc+Ihs3bpS3vvWtT/vzuVxOGo3G07YjIqz7Nzzbtm2T888/XwzDmIu96EUvklqtJgcOHJCpqSnxfV/OOeecub/v6enpmEcQNNaPJ/S8+93vlre//e0iIrJ582b56U9/Kn//938vn/nMZ+bacqwTLdx8nvvbtm0T27Y72vT398uGDRtk27ZtIiLyxBNPyG/8xm90HPucc86Rf/3Xf1Vf+/jXJbvH/fnnny/nn3/+3J9f+MIXysaNG+Vv/uZv5FOf+tRcPJfLSRRF0m63JZfLLeRt01NwwfxLUqvVxLIsue+++8SyrI6/KxaLc23e/e53y/ve977Ez69cuXLu/xcKhad9vYGBAZmamuqI3XjjjfLII4/Id77zHRE5thg/3vajH/2ofOITn5hrOzk5KYODg/N8d0TUbT7j9NmAxvqSJUtEROTkk0/uiG/cuFH27dvXEeNYJ/rV0t/fL4ZhJMZ9N8dxZPPmzbJjx46O+OTkpBQKBS6Wf0H8DvMzsG7dOnEcp+M7v1NTU7J9+/a5P2/evFnCMJSjR4/KCSec0PHfyMiIiIiceeaZsnXr1sTfn3DCCeK67oLOafPmzbJ169aO2He/+1156KGH5MEHH5QHH3xQvvrVr4rIse87v+c975lrt3PnTmm1WrJ58+YFXwsiwjZu3Dj3HcbjbrvtNimVSrJ8+XJZu3atOI7T8T3omZmZjnkE2bx5sxw5cqTj4bl69WpZunSpPPHEEx1tt2/fLqtWreqIPfrooxzrRAs0n+f+xo0bE/lAExMT8sQTT8z9Y3bDhg0dY15EEn/u5rqunHzyyYlnfLcwDOWRRx6Z+wf0cRzzzw4umJ+BYrEoV1xxhVx11VVy4403yqOPPipve9vbOpLq1q9fL5dddplcfvnl8i//8i+ye/duufvuu+Uzn/mM/OAHPxARkQ996ENy++23y5VXXikPPvigPPnkk3Ldddclkv7m46KLLpLHHnus4yG6bt06OfXUU+f+W7NmjYgcG9RDQ0Nz7bZs2SJr166VdevWPdNLQkRd/uAP/kD2798v733ve+Xxxx+X6667Tj7+8Y/LBz7wATFNU0qlkvzu7/6uXHXVVXLTTTfJY489JldccYWYptnxNY5umzdvloGBAbntttvmYoZhyFVXXSVf/OIX5Tvf+Y7s2LFDPvaxj8njjz/eUWqy0WjIfffdt+DND4gWu/k890888US59NJL5fd+7/fk1ltvlYceekje+ta3yrJly+TSSy8VEZH3vve98sMf/lA+//nPy5NPPil/8zd/Iz/60Y9Sx7zIsWd8dxLvJz/5Sbnhhhtk165dcv/998tb3/pW2bt3r7zzne/saLdlyxaO+WfDc/wd6l9Z1Wo1futb3xrn8/l4eHg4/ou/+Iv4wgsv7EjW8Twv/tM//dN49erVseM48ZIlS+Lf+I3fiB9++OG5NnfffXf8yle+Mi4Wi3GhUIg3bdoUf/rTn577+1WrVsVf+MIX5nVO55xzTvyVr3xF/Xst6e9Vr3pV/JnPfGZer0G02GlJfyhR7+abb45f8IIXxK7rxiMjI/GHPvSh2Pf9ub+fnZ2N3/KWt8T5fD4eGRmJP//5z8fnnHNO/OEPfzj1HP74j/84fvOb35yIf+Yzn4mXL18e5/P5+Pzzz4+3bNnS8fff+MY34g0bNsz/zRLRnPk89ycnJ+Pf+Z3fiXt6euJcLhdfdNFFicTbv/3bv42XLVsW53K5+HWve13853/+5/HIyEjqaz/22GNxLpeLp6en52Lvf//745UrV8au68bDw8Pxq1/96vj+++/v+LkDBw7EjuMkEgFp4Yw4fsrvC+lX2g9+8AO56qqr5NFHH1VLyHV77LHH5GUve5ls375denp6fslnSERp6vW6LFu2TD73uc+pmxCJiBw5ckROOeUUuf/++xNfuUhz3nnnyfve9z55y1ve8mycLhE9C37v935PHn/88act/fbGN75RzjzzTPnIRz4y72N/6EMfkqmpKfnbv/3bX/Q0Fz0m/T2PvOY1r5Enn3xSDh48KCtWrJjXzxw+fFiuvfZaLpaJngMPPPCAPP7443LOOefIzMyMfPKTnxQRmfv1rWZkZESuvvpq2bdv37wXzOPj4/L6179efvu3f/sXPm8ieuY++9nPyitf+UopFAryox/9SL72ta/Jl770paf9ub/8y7+U73//+wt6raGhIfnABz7wTE+VnoKfMBMRPUceeOABeec73ylPPPGEuK4rZ511lnz+85+X00477bk+NSL6Jfmt3/otufnmm6VarcratWvlve99r/z+7//+c31a9DS4YCYiIiIiSsEqGUREREREKbhgJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKwQUzEREREVEKLpiJiIiIiFJwwUxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCm4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohRcMBMRERERpeCCmYiIiIgohT2fRlEUyaFDh6RUKolhGL/scyJ6zsVxLNVqVZYuXSqmuTj/XclxT4vNYh/3HPO02CxkzM9rwXzo0CFZsWLFs3JyRL9K9u/fL8uXL3+uT+M5wXFPi9ViHfcc87RYzWfMz2vBXCqVRETk//fFd0s253b8nRuW8IEdF8ZrYRXGQ7FgPJIQxnNuFsbjINm+4dXxa4YxjBczvTDem1kK483aJIw37DZuP43/FRMFDoz7lo+PXz8M42auAePFUh7G85keGPcCfP6mibtNDO6h327Btn4Ln2MhW4TxcqkC4w0PH39megrGc24Oxj0z6Phzq+XJZz78t3N9fzE6/t737Nkj5XK54+/CEI/L5yM8S4iIiT+BC+qzMD49egTGHRuPSzuH+6qbwfOEZeNxaWifmijnb1hK3MBztMC5W/t0Urua6lX+pbKsznOfnZ2V1atXL9pxf/x9v+uKs8R1O/sTeLSKiEhs4L/I53C/cxzcj0ylzzhGBONuJrnG8PGjUrwAx4MYn6Np4XNse/hc4hj334yLx2SkXEttVtXm22YTP6MD5dnt2Ph9ZdU5RXm/4EyDAF+DvLJWy7n42IUSnveKLu4fy4bx/FnIZGB8crpzzdBo+vL2P/zRvMb8vBbMx381k825kst3noQb4IvhuHjBHIYejscLXDBntAVzcmREFh4t2oI5l8XHzmfwjZRQORcb3+C4tbAFc/ekPveyIe4QZha/32wOt88pHctUJpmFLJgtE19jS/DBc9o55pVrrDzD2y18nKz2Xk18oMX8a8nj771cLnPBjGgLZgv/RFivwbjjFHB8oQtmB8e5YNZpc+tiHffH37fr2pLJdM7zymNUYuVaZTILWzBbC10wZ5PPIeV2iqkspP0FLphFOZdnb8GMr0EYanF8INPEcW3B3H2vj9MXzMnzsZR5T7sG2Qw+dg7cVxGRnNKfCnk87xWUebLVxtdmPmN+8X1Ji4iIiIhoAbhgJiIiIiJKMa+vZBy3rG+15AudvyasN/HvaTzBvwMpmPj7spalfP8uwvHAwMefbSa/P2jE+KP5AWcAxpfa62F8at8ojFdH8XeYS8tOgvG+ymoY175H1PLw11j2N3bB+KG9j8F41RmD8WUn4O8AuT34V8KtBv5+VA78Srjh43NvKV/NcWP865JaA38PvdHG34m3wPfbRESCAL+udH/jw8S/LlqMLMtSf3Xd7fn4q2zlt61iWfjzhrFD+LvKP/6nf4Txcg5/JcMtlmE8r3zXTvsOnvaVjGJvH4yv3XQ6jOfKeO6OjOQcrf2a3nyOvpKh/cq8u1/Pt58/39mWLXbXMzmXVb5mqcypudxCvsIjIuDrlMdeF68BMrnk1/QC5df//gzOmTGUr0BkXPzVPUf7ulWInxeej+ONJr5mprIOarfxesdTvpytjaaMi+cCw1C+mqLlloXJ9so3UCWOla/UKPlEhoH7WamE50lR1pq1On7dJ3d25je1Wsp3jQB+wkxERERElIILZiIiIiKiFFwwExERERGl4IKZiIiIiCgFF8xERERERCkWVCXDCyKxu7I+6x4uxi+OVvUCZ106ShFrWymo3azjLFOUud6bXwnbLrHW4tds4XOxlKLltoWz2W0bV+FwC3gTjkIBZ4FWLJw16itbL81O4KodO54ch/E927fCeO9aXD2jMoKzW/t7K4lYMY+vjeTxv9Um2riSh93C/aBg4Z0BHRNXHGm0cIUPv9rZn9otpZoGiYhedeBXG35P6lsNcB8+vGsPjO9/7FEY7wfZ/iIis0pmvJPD4zJrK31+Vqkkk8fzzbmv+XUYv+DVOC7Z5PmE2mYpyjU2tIv8LFVdeX72118ey7ITlauU/WzUTcocW4mbOC4G3rW1YON7Z1vJ/h7ZSmWnrHIMpRhSVnlPyv5W0mrhsRoGWv/FL9xWjhP4ykYk4BqIiIgy/kxlPAWeMvcplYBsMNdktGumbKJiKufuKBWuTGUMz0zhKheGssFay+98Ty1lHofnMO+WRERERESLEBfMREREREQpuGAmIiIiIkrBBTMRERERUQoumImIiIiIUiysSkYs0p2wminjjMbabB3Gp5WM7UwOZ0wa1gyMhzHOjKyUS4mY2cT7wnsezkiNBVdIMJR91ddt3ICP0zcI47MNfO6TE7MwnsvhbPYgxBmvlovbWzbOrh/buR3GD+3eB+NLThqB8dwLk9n+9fo0bJvJ4HPM2304Xu6B8VrtKIwHDXwt4xBn7DYmOqu9tJUKBfR8hseToWaW4z6yd+cuGFcS7EWUPhk28RyaVyoQBY0GjPtT0zDenJiA8Tuv/xcYLyvVfU5/xUWJmKlVQtA+o3mWqmHQs8MIYzHCzoe9YeDn1uAQrgZlu/i526zhfp3N4X6dUbqMAypiRAbudxkLP9PNGFeriJWqFLbSr7M27r+tGFdligIcjwN8jUt5XP3DspVrZuHzHxnEz9HZKp47ZpV7hSqUeE38nooFfM0qBdw/HFepnuHg9Us7wNfAdvCMe/IJSzv+3GjOvyIWP2EmIiIiIkrBBTMRERERUQoumImIiIiIUnDBTERERESUggtmIiIiIqIUC6uSIYHY0pkZ3mjgzEg/whnkrouzPds1vE94VtlX3DJw+/EdzUSsqFS9GK/vhvFlFZwNPqBkA1cLuGqEnx+C8baHK4VUpw/AeMHDGa+tVgvGPV/JzI2UexXi61Or1mB85907YdwNk1mpmy84Aba1lH7gZnDmrCU4ezhnKZUN8vg4sdJvMnax48+tFr5WtPhoVTLabWVemZzEB7Jxn2zFShUOnMAvpuC/aNTxeI2U8yzY+PMSexpXntnyf/8Zxt3eZGWik857EWwbxzgDXgx+dvOfSda1Jet2Lg+KRaXySYT7Y+TjfpfP4uoFeRePAzPAc3bgJV+3HeJqD4Zyjq6FzyWM8Gs2tfWOj59PcYyP09+THDMiIhkHj484VCYDA8fPPmUljJ9+yloYV96W7Nx9EMZHJ5JzXK2OD2I5eP1ixbh/+HUct0vKNasUYdwy8T0pd5VdqTXmP/dwliIiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJSCC2YiIiIiohQLqpJRzBclX+isbpCpK1mmyj7ypolf0lL2aHeV4+x4/DCM7334SCI23IszTw8cSbYVEXFPWg3j2TVrYNxevgrG9+8aw/HtT+Dj1KZhfPngchjXMnanpnCW/swsjscRzmLVMoUDJbv1sQd3JWK5cgG2Pefik2A8k8P7xZu4m0mpD/enQMn6tWwlS7/U+QLNBq5AQr9KcP/Fufhaa5EoVjLsc3huOu3ss2D8B09ug/FDY+MwnnFxpzeV8hm+8s4MG4+RrOBxP1zCY9Au4upBN/5LsnpGvogz11dt3AzjsYnPPVJulqneRUxt3X3TtU6wyLiuJW5XlYy8UnlIeUxIpFTKKpdxhaSMpVSDqicrX4mItINkFQRPqTjjgbYiInG4sNkgVqoy5Qr42hQd/F5zjvJAC/CzNVCqNvWX8evao4/DeC1IPqNFRHpPxXPWBS88FcabjXoiFinP1skp/Cy9/fYtMO4p1b9sE/ePfA6/biajzHtdS81wAR8b8xNmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKbhgJiIiIiJKsaAqGbZpi91V5cLM4mzowMeZrY6JMxoNpQxCVqmScXgvzixHGbVBZgK2HT2yF8ZvaU/D+MAUfk89K47CuNOLq1vs3YGz5VtH8L7tpbNw1noQ49vXbOHzrNVn8HECnM0cRjgb31Tuoe8nM3kfeQC/1xPOwHvdrz6pH8anZg7BeKxU7Cjk8L7zTk7p8nFn//OVa0v/OcUGympX+i8OS2zgzw985WMFy8F/cdZLLoTxHfv3wfg3/vlbMB62cd+261UcV6pe5Nt4fK/K4gz+IcFzbj7Cc3R9TzLz/rZrr4Vts2/Fxx4+ZSOMh0pVgu7xepylfAZkKPc8UT5Da7fIeL4vptk5puoN3LbSiysh+b5SmSLG/VGrZGFk8D11jORxmjV87CDAVS9sC8/zxRJ+T1ZWqW6h1GHxmvhZbCrVPNasGobxJYNlGA/HcLWwIw/j52XLxcfxa7MwHvXieK6QrGoWm7iKTnUUH2OgB5/LaIjvVbWG5722X4PxUgnPb1HUee6NFp5nEX7CTERERESUggtmIiIiIqIUXDATEREREaXggpmIiIiIKAUXzEREREREKRZUCqBWHZMo7MyEjH2c7RmFOOM1W+6DcUPJUG9UcZbp+GFc+aK/kMy8bDRxlma7ndwPXUSkdgjHdz6+E8bD+EYYD9wKjFenRmG8p4CzTFcuPxHGi2VcUaLZwO+32VSyTD2c/hxG+B76AU4ld0FFk+oYPvbubfthfMlaXN1CYtw/tCR6pfCHTE3jSiFOV1WMZjtZ8YP+84pBlnpk4OnNVPpMpGS6a4UTTMHZ3JkCrhz0klddDONGXwXGf/Cv18P47t07YDyfwRUoqspg8JW5NV/EWeONA7gakOslq3Mc2YvP8ZYf/l8Yf0kG36vhE/DcF8CqKCKR8hGQrVU6weFFz488MburJDXx88AL8Byfz+Gr6xi4n4qyZsi4+DiulewDxQzuFwMVXJEhMnDFJ1N5zXoLrw0CpcKHZeLKNb3lHhg/aR2ukuHG+LmlFNuQ5Ze8AMaNDF5jhMokF4X43ppusgLF5EG8rtl317/BeH54FYz7Jh6sE9PTML5sOb6WjtLNDNvu+vP8S+PwE2YiIiIiohRcMBMRERERpeCCmYiIiIgoBRfMREREREQpFpT0ZxmuWEbntoJ+gJOjcso2ibPK9sxuDn9xe3oMf8m+UhmE8aHe5JfmdylbUZfKFRifmJyGcUvJMCvn8LfLd4ItY0VEfEluKyki0gJJDCIiDz50J4wvW7EOxqdncELkbA1f+5aPk4K0rYKzebzlZBglzz9W9iGeOorPpaYkIhnKcbJ2L4w3GjhZITBxfw2kcyvUVpt75P4qQSPTjPF4Cn2c1BYrW8FnbJwAZCrb6ir5aLJ82TIY/y9LXgfjvUqi0pf++n/D+JFRPO7LBZxofVBJojHG8HHO6RuCcQuM++XrV8O2tcYUjN92/XUw/srX/zaMl9athfGWqVx8JWx1T3HMAhQREdMMEtem3cLPCcvGz79AGU9tX9mmWhmvRuImHVMsJF+3r4LHZKmEE3HbTZyUV1UKApSzyphXkge1LbPzNj5Ou4m3tA5qR2A85yjPKRe/X8vJw3jUnIZxr6nMfbmBRKw2uh22zeWUZ24ZX4NBM3lsEZEjh/bBuCiJlX3K8U2381lvKfM4/Nl5tyQiIiIiWoS4YCYiIiIiSsEFMxERERFRCi6YiYiIiIhScMFMRERERJRiQVUyYsNIVE4ogK2oRUQMA2cu1tstGLcEV0colnH26Wol43z748nKFDv3H4RtJ6fxVtEzyvbJOWXr2cFhvJ1lbz++vLM1/F5tE2fvz07h83eU/V69UDlOE2f+xiG+V66NM2pHBpfC+Ey9logdPoK3wJ5UqmRMH8HbevcNV2C8HeFrfPjwARjvHcBVNcrFriotDu6n9B8jVjLmtbgRJ7emnTq4G7Y9pFSvcZSt4HuUra7tMs7mdnN4TnSyuLqM24Pbv/JlL4dxr4n75uf/6kswvufQGIznSnh875jBY7A3wBnzSyrJOXpdBY+zFSPKvH3P4zC+5Xu4esZL3/Y7MJ4dwvckVCo2dG+lzdo4x/SXMpLrqghRa+KKD6FoW2bjY0dqNRpcPcpR5vh6PVl9oWegAFqK9PX0w7g7hMf20QlclcKx8LmXSiUYLxZwPGPjixPVcJWMuo+rPhkuvjZ+YxrGYxNfYx88u0VEHAvPEbEk10JKARHJrcHro0kbV88YzuH2hnLtq7N4Pq8X8PErSzqf9aZyXISfMBMRERERpeCCmYiIiIgoBRfMREREREQpuGAmIiIiIkrBBTMRERERUYoFVcloTs+KtDsrMNhFnH0a2zijtpTDWax2Rqn44OHjPLHtURj/wb/dmDy2i1+zXMLnHir7vwdKFn0suJJHTxlne0bxFI5HkzDe21uBcdvE16zh+TDu2Pg820r7bCEL44P9QzCezyUzakeP4GoVUYgzU4Mp3CWrPs54tUrKXvcmzk5u15XqLWHntW818evRc8swcJ/fvf3JROzfvnstbNuu4QotJcF9o2TgPha6uI9ZGRwvVHAFh5POOw/GV552Koz/l9e+DsZzGVxt48//4i9hfN/oYRgv2vj93rUPZ6NvCAeT57IDj/vBvuUwftopp8D4ffc/BOO3fPsbMH7J5W+HcTOPnwHSXXVFqcKy2BQzseQyndciiHG/qDfx8yNj46owYYyfQ3UfP19zHp6Ll/Qnn9+rV54I27pZ/CwLlYpSg724OkRPD35PWRdXn4hiXBHLDHGlm8npaRhvTuDKNfkCrlhlZXA/njm4B8Z3PD4K46tOwWuhxpF7E7Eje/ExTliD73cmi++3rVQWCWN8r0br+BzbLXzto6DzdeNAKecC8BNmIiIiIqIUXDATEREREaXggvn/a+/eo/Sqy3uBP3u/77vf+2Xul8wkk4QQQiAQuQaUCAL2qJW2a9kiHptGKm0FG2FVoCjCclHAtoCxiNoDUqjG1uVanno81ooaMQQIwYRL7vd7JjOZ63u/7d/5AzPlnf19NhmJcGS+H5d/zJOdPfvde/9++5dhnu8mIiIiIvLBBTMRERERkQ8umImIiIiIfEwpJaNUNmLZjd2XkVQebmsr74sPCO4yLZfw9sorzmXvYdyFbQe83aotzTitwrZx9ybulRfJV5Suy9xxWG8OervHRUSGq3g/x0Zx17qxccerZeHO2fEcTtuo1/Ens5S0jShIvRARKRRxh2+p5K3H4wm4ra0kD9i29x31IiLhCL5vagHcQd3Th5M8wmHcLV0pTDqOIP6M9NawLNz5XKngTukNG1/x1IouvtbNXU2wfvBVnMiQLON7LGz6Yd0oaQIH6niclfJZWO+cdxqsB2J4TF19+VJYdw0+/pWPfBXW9+7eB+vZKP5crwwMeGrWSzvgtkkXj+PzLngXrDe1p2F9+y+9aUgiIu0pPO4XfeBDsB5pbUwWsZWEpOmmmC+K1Buvd76Mnx+xBH5OOCH8fK1U8POsuwXfG+fOPwPWQ0HvOKgoiQeJEE7yqLvK+iWAn9F1JbGjUsNLqXoZP0fKY3jN4GqJVUrKx/ChI7AesPF9PHhkFNb37sdz0K5j2/F+BtZ7avPOwOudbCu+ruEYTsNoasLJHxVl/TUWwuc+X8DbJyYtKosF/ExB+BNmIiIiIiIfXDATEREREfnggpmIiIiIyAcXzEREREREPrhgJiIiIiLyMaWUjEgkKJFI418ZHfV2SIuIBC3cGZlM4XooiNMRsmO4e9MS3IEbBfsJO3jbUgV31JYnRyb8WqWK63ZuDNZPz7TC+uEyTrEYyw3CuuviTtuQ8v76EG5ml2oVd4O6RnkXu9KxO6K8775U9p6fWBR39GfH8Wc6fGAI1ptacad0JIw/bCbSDOt55f3yY+ON575UxN3Q9BpLTRPQ6jgFxyj1gI2v6+joCKznSt7rdenVH4bbZlozsL5rzwFYN8P7YT1Zw/fIwChOr+kfw5E/RaWzf9b5F8D6We+5DNaVsBu5SknPCCh5QN/4X4/D+vot3iQSEZxWcHgYz4nrtuGu+2oAJyeElbk4jad02fPCL2HddfD9dP41H2z4ulZUYpmmmXyxInW38ZzFkvjZ7SgPHCXoRma04WfCGUqyUXtrF6yPj3iTJkwF33dSx8+DQAg/Q+t1XK/l8f4LWbxOMcqzNdGM018SMTynZA/jNYOrjA87iK9J0OA1gGXhpeC69XhOnN3lTe1IJ/F6YSynJFn1pGA9GcdpG6MBfCxOEB/72Ah+1kdHG5NF8gV83Ah/wkxERERE5IMLZiIiIiIiH1wwExERERH54IKZiIiIiMgHF8xERERERD6mlJIRC8clFmnslK1WcZd7NBHHOzF4jT58FHc09u/DXal2AHex5krebtVUTXkvvMGd2a6Lu8frdVyvKd3yXTHcyt0exR2sxuD91Ot4P46Fz2UohBNHtOAH1ygJBkpHrbFwV2ndeLv9S2WchhFUOlvLyvvf68U8rHd14Y5aCwcPSCmL91PKNX7fUunk3y//Tmd+/f9GSgu80hXuqv80x2NQLHxP7t+7B9ZHh7xpPeUCTjzYfxRf26IWL1PGHfAHs/je3jqI025Gs/h4LGW8bl6/HtYXvefdsK6cSQkF8Vx5xXvfB+sVZU4fWfllWB/o7/fUxpUUi1fHcaqS+yqenBak8HOkr68D1gMuTlFZ/1/fg/XM3O6Gr3M5PD9MN2WxxZr087SYoywXlOdHU9ybpCAisuiM02E9HsfJEaUKHq8By5tGE4riZ7RRntF2OAbrUSXhylTw9qaOR180hse2lnCVPepN/hARcS0811gOnlgTrThxpBZpwd/3lW2w3h7Cz/rzzp7lqUUTyvypzG9uqAnWKyDxSESkXlWuYQBPNq6yPrImrWsssG7R8CfMREREREQ+uGAmIiIiIvLBBTMRERERkQ8umImIiIiIfHDBTERERETkY0opGaViWaxJ3fGOwd2eVhF3qxZq+N3n5QLusIwpXazNSfwectSrms3jLvdkSnmXeTIK6+447tbdNnQE1puVY88pqR0R5b32jtKdrLwuXkoVpcvUxd3MISWxIqDcHfU63n+u4O1aLiodzu3NuAs5FMbnoJAfhfWR4/g+Cydwh3ZdSe1oiTV2DxctJVKERESkbimZDMo/wY2SnmEp8RlKBodYSqd0rv+Qp/bqL0fhtvkQHt8B5b6uuLhTvJRUur9zuGs7qAQw5JUO/oJS11J8AgE8IdhKPWTjc3/++efD+rKPfxzWv/Wv/+qp7di1C24bieM5cdfQEN5eSUtJl3G3v6nja1XM4XNZHG9MLinm8fNp2gkGPA+YqpJw1ZTA4+ms02fDekeHN2FBRKSQU9JZBjbDejTgfd6Usvh50zxDSe2ytWwZ/FkLwzi1q5bDa4xaFW9/cI83WUZEpDAyCuvdvThRQmw8U9YMHvObd+NxdmAvTq85ez4eZ9Wwd44bKeNzmUzg9JOKja/J2PgxWC9VcMpQzeD5MBXH83A83rjGMOrTxos/YSYiIiIi8sEFMxERERGRDy6YiYiIiIh8cMFMREREROSDC2YiIiIiIh9TSsmIOHGJhhs7YoNKwsJ4bhTWi0qnox3EnbaxOE5NSKeTePuYt/NydAx3qmrd42EHd1fGIjh5YayIO6v/68CrsO4qaRV1mPEhki/hVI1yDXcEW8q/g4JBrWsUd65PTkQ5oVLF3xfVQ8q5nDmzF9ZTaSV5IIw7YccL+H6KK++RLxbxuQyHGj9rTUljmI6MeNNn6ja+HpbB97Zt8DwRsvD9IXV8j3Ur4z4j3us1tG0H3Lbi4H24ymeylTSJOWctmNL2NeWzJjs6YP2cC86D9XBYGSNKekZZGa+Osp90JgPrPb14zL5n6VJPbc/efXDbkIXn0BElleaX+/B+wmCeFxE5O4M78iOJDKz3zGxMcsjm8Hwy3VRrIvakx2NUO+cLz4H17t65sJ4fw8/LwUM4WSVm4QSKkb3e8W07+P6KNzXDerhag/VyHqcpbfvl87DuBPF+rDrez9EjI7DeNxenR+WHvYkgIiK1Oh7DR4eOw/qxIzglY0avkgwWx/P25n7v52rpxuc41tUJ67U6flaUy3i95iopRm5dSTpx8TWpVquTvsbbIfwJMxERERGRDy6YiYiIiIh8cMFMREREROSDC2YiIiIiIh9cMBMRERER+ZhSSka8NSKxeGMXakFJcAi6eC1uCrgz0hXcqRhK4GNJNOGEi7Z27zvXh3fijtTxcdyNGQwox25wN2ZA64qv48/kKvtRQjKkrnR7irKfsJKGEQ7ic+YoSSe1WhXWbQunZ8Rj3rSNtNK13j2jHdYDIXwsFXzbSGcLThgIKokEyqUSWyZt7+JzNR2hlAxL8AUJGSVpQrmA5fFhWB/ctxPWd254AdYD494xHi7g7npTwJ3rOWUOqo3j/USVz3RmpgXWpYLHTT2Jx0hPZzesj+fw8WzZvBnWX375ZVg/ff58WG9tb4P1n/78Z7C+fYc3raCmJAEFgjj1qBbG9fEyTq3oLygpKkk8J86cMRPWM21dDV/bYZxGMN0YY4sxjXNgKobv63gUz+WjIzjVYN/mjbDe3qykOKXx+Mgd9o6nkMEJHCObX4J1W0leyGfx/bV780FYnzErBesdnThZZM5CnDiTbsNJE1Ubp1js2HwM1je9tBXWZ3bjBdXCd8+D9Y178fxcA2k3ThwfezSB7xtj8PqiXsfPEKMkJ7l1PK8WC/japlKN+6nWmJJBRERERHRKcMFMREREROSDC2YiIiIiIh9cMBMRERER+eCCmYiIiIjIx5RSMgqVIZFgY3eka/CaOx7Fuw63447JUgVvf/RAP6zXDe6AnNHb5akdOTYAt7UEJyFYSlyFpaRDaOkZtQruAnWUjnCjdJZbRumuV7pJXSUKAu9FxFX2o3arKp+3qdWbUHLGGbPgtulmfDQtHUlYd5XEkWIRd/EmYxlYb25uhfXApKSQQgGnv0xH5tf/ez1buVfHDuAu8sObtuD67l2wPnhwH6yXRodgPVDyzgfpCO5QD+XxtR3O49SckIvv98rhQVh3CjhpIaTMWY4yMn/xk6dgfV8Vd4vv2YXPZX8/nkODjpIkoyTV7Nq9G9ZHRrwJJSFlH7ajJRDhubK3rw/WF110KayfPRenYcxfeA6sh+KN802orkQVTTPJZEIikcZruHPHfrjt3h37YL056U1SEBE5cy5OU0imcTqLq6Q7xWac5qlVhvG9Xs3itJVSEY+BkpOB9ZZFOA1j9vxOWE834XFQFTyGqxW8Ntj86lFYP3AQz0FdSfx9Fy3C6RxtZyyA9YP1I7DekvI+6xMpvLYLhnDCh6WMNdvBz3pL+fFuTVsfKeu74qSEnWKBKRlERERERKcEF8xERERERD64YCYiIiIi8sEFMxERERGRDy6YiYiIiIh8TCklY3D4mERL4YZaJIa7OmMW7oR1bdwZOVrC3a3FCu5ot13cARmPhj21nhm4g3XgmLe7W0TEsvAxBoO4616Ln6gpyQ6ukjKhNHtKIIAv0+Tkggk2PjdGqZdrOA1DCduQWh0nJIRBMkqiCXchp5vwO+0zTTglwwnhYy9Vs7AuLk4kMDV8DfPFxv0UiyW832nI+vX/Xq9eLMBtn/vBj2B9/6+ehfWgkhxhKckUyvQhAeO9J13B956U8T0TVMZr1MFd3m42D+vGweOplsL3sATx9k/953/C+rYsTvMIh/Bc7CqJJsUynltrWuqPcn7iCe9YtgN48nCVz9oSw/NBezN+jvRn8dx9+tKPwvppc3AKwOREITfInyGJiDiRoIQjjeNndASPm6GxUVjv6z0d1juU57EbxOM1Es/AesDxpjIUUjhpI+LgeX9oBKdV1Ir4Xj+nB6dMBJUkjwBYj4iIDPTvhfU9B3Ca0NYDx2G9pM1xShpU0/wLYT3a3gfrbTPx89iAuWZy0tQJtnJu6oLnAmXqkHBEWWsmcBpL0I3hHU1+tijPGoSzAxERERGRDy6YiYiIiIh8cMFMREREROSDC2YiIiIiIh9cMBMRERER+ZhSSkY0lZRorLHrM6wkO4yP447amvJ+70odd0CGlI5zN4U7vDMBb1dnZgi//31kGL9fvqgkAFjKZ9XCKiwb/4VyGScDaN3sVakq31j5vkr3qVE+QF35vpUa7hSORXH3qQW6ZMdK+DrFy/gcl0dx97ATxf+2yyhd0RZITRARGSkN4e0nXcRiHV+j6Sjw6/+/XnYUJ0Q8+8qrsO4ePwbrvRE8BVlGSUWp4fSScsl7vUoVfA+U8+OwXrOVcaOEW9RsPD7sIP6+AQffwzWD93N0cADWx6t4+0wan0stlceycYd9vYLv/UpZ+bwB7wkKKHNiTOmk7+nEyQnnLjwb1je+8hKs7z+E77N5886B9cmHqSUVTTdurS71WuPJCDr4fmlq8aZViIicvfhiWG9pwckL9QK+dtUinrMr4Dldzyvrjji+dys1fMHb2vBnSirpTuUsfs4FwzjlxYriOaK/tB/WizGcepE1eD50ojg5QmL4cwWiTbCeasJrpFrVe/yWjT+TW8PP+lJNWdfU8Lk0Lk40MS6+hraSjFKuNB5PxT35nxvzJ8xERERERD64YCYiIiIi8sEFMxERERGRDy6YiYiIiIh8cMFMRERERORjSikZETso0Umd1bbSQu5EcPdmSOmeDrtRWC+l8Zr+uIu79I/u2+WpuRbeRySCO0lz2TFYDyrvug87+BwYCyd/VCu427OsdL/X67iupXC4ylWtlfE1SUTw++4rNj5v8SROycik055aQOnE15IB8hXcUes6OOkkqySORG3cIVuv4A7cUqGxu7pUwteIXlNWEij2lfH5DSn3cGIMd0qjLmwRkYLB4z5QAB3zFTzZjJZw53epgo8l1IrvYSuF70lbmeMCVTye6koKRyKJ0wSiFXwuo3G8fTCEJ4TBwUFYz2bxGKwoYy0Q9H6A1lbc1W8LnrMGBvCxdF3ZDevzTpsH6zNaZ+Dvq3XST7pYthZ5NM2EnKA44cb73rLw2E4p92mTkoZhasOwXhzdA+u7t++D9WrB+zxrUp7poqSzbNmLn/VdPfi+qxzBz8q0cr9H2nFKRiCMn6HJtl68/zq+f50iTvxJB7TECmV+tvHY7mjBz1G0VCkU8dyczeJnaTiM50MnhL+nbSnPdBenq5RcvP6yJ01XtQCeT+HfPektiYiIiIimIS6YiYiIiIh8cMFMREREROSDC2YiIiIiIh9cMBMRERER+ZhSSkZwOCTBYmPn7NHQQbhtRHmPd8TGXabi4O7kUALvx6plYL3U2+GpjSSVlAmlO7I5MxPW02FvCoSIyHgFd9qOF5V6Bne2Vio4xaKQw/ViEXf1lyp1WHdd3Dnb3dIF6+EY7jiuRvF56+lt99SsIL6uxsX1QC0B65LHHa8VJdkgEsf1uNKBa01KBLEs3DVMr7Et3LVtlHpdSR/IF5V7XkkpcVJ4yoqCLvigkiYRy7TAeqmGj9HVxlMeH2MohTvgxcLH7irTsDJERJRxbFx8nKaOfy7i1vD56ZuJO/X7+/thPZ/3pmqo94dy7CMjI7A+rNQvXXIJrM+fvwDW63VlHrIa68YwJUNEJBwJSDjSGCeQacnAbYs5PFf+8pdrYf20Npw6kx8cgPWN6w7Aenur93gCWqKNkq4zPIC/p1XDSTFNAe/6QkTkOKyKbD38M1jPBnCqRiKB08K6OtpgPTeCI3YyDh7z+w4fhfV0axOsB5TUo2jMm/7R1Dwbbpsbxc9cNbEjjLfPVXC6USyO5+Gai89Ntdp4beuC1woIf8JMREREROSDC2YiIiIiIh9cMBMRERER+eCCmYiIiIjIBxfMREREREQ+ppSS8X/+9zMSCjV2ofZdhLs3Ey24+71Sxh2NzXYG1uMhnJoQT+BO9NaaN8miWsJpFcqrzKVPeY/8rA7cBfp/1/4E1qtl3JmbTODTnkx14v0oSRDjY7g7uVjC3e819AJ4ESkZfE3On78Yb2/hz1W3vPuPRpJw22oJd7zW67jTP5jG3c/hgNJxPZaF9XIc1yOxVMPXtoU7bOkEnHZgGSWpQYl8GM7i6xGO43FvQBqGiEi16h3jLS2489tqw+NsVJmbakU8zmo2PpbjStpGtor3k3VxvRrASRMtGZzWYynJFMNDg7DelMb7+Z/XXQfrQ8NDsP7d737XU7MtfA6UQ5RqFc9xySSeP+bMmYP3o8yVgaCSnjDpgLRzON2Eo2GJRBvPWTCsnMMgTlPasGkXrKMkKxGRs06bD+vzzsH7P3roiKfWP46fcZkkTl5INuG0ikAcbz9o8Lx0bC9+JlYCeJEx90ycxBUM4Tkl5eBxUI4qqUTKXHNk+BisR3fvhfV0BKeatYS958FRnpkRJV0sGFY+Ux0f+5jyTI9FMrAesHHiSKnQuJ+A4Hkf4U+YiYiIiIh8cMFMREREROSDC2YiIiIiIh9cMBMRERER+eCCmYiIiIjIx5RSMgIhR4KTUjLsCu5EHDyEO6ojYaVz0cFd9xLGXc9WHW9fynm7swM27iRtasUdr2fOwd262eM4+aNax525Lc0ZWDcuPvamFu/72UVE7ADuPo0n8fvuK2V8zuo2/r6HDuNrNTCK69e8/3/A+vaBbZ5atjQCt43HcYKBHcTd9WGlezgewh3UVQtvH4rh7mdbGq9hTeluno5qhYLUJqVT7N29E25bLRdhHfdai4SieD5wUilYr7r4nhfXm85h2jJw091KB/nOgQFYb03heWJWdxfe/+F+WD+Sw13eroPv1b75Z8B6Mo3niZCSIDI+Pg7riQT+XJEw3s/smb2w3gfqR44chtsKHt7S3Iw/U2tLC6zbypxereI5rl7BnfCO4zRup6T0TDepeFJiscZzk8/i+ygTw3NlbgZO0Hr5AH6uHDiOn1tze/C90TLX+5yOKfdubw9Opeiu42MPRfFzxdj4/uioKDe2jfdfU+agfBmnR+WVebWgJHGVlP0HIw6sD43jJDGx8BxRGfbOcU4OH0tAWb8EA/haVWr4HGfH8TkYzuL5VpT0r8ikdVCpgO87hKsCIiIiIiIfXDATEREREfnggpmIiIiIyAcXzEREREREPqbU9NfV3STh8KRGgAJuwkmm8esv25UmjkAV/9K8W8K/AL7v4H5YL1S9vzQfi+Bf4O9owa/IXbDgPFjfuOFXsK41D7a04MYlS3DDSjKGXxUZDOHXgKcTyit167g50RjcEOMor409nhuGddvgxoElZ17uqe3r9zYCioiMuLjxw4rgf8MFleYMo7zuW2setJR/IoaCjW1pNeX1xtPRQP9RKeYbx9WOrVvhtskEbqpMBfAraJui+LWpWeXV2yFl7FiOt60wG8TNJscMnvaGkxlYd8N4P2kHj5tSCn+mjpmzYT2ewI2PixacCeshpaHVBY2Pr9XxvVwu43mipjTIGaUhbkaXd64/cvgg3DaTycD6GfNxo3WX0lg5NKzMTcq5MQbfN+Fw432Ty+Gmq+kmlUxKPN54bkpFfL+0xJTX1SsNcqaKJ+Ht2w7B+q69+Fnf2+ltHG9VGu1zymub2zvwGsDN4zHQrKxfMkk8hkeGj8N6pYbXTUHlnInB9ZQyd8TwY1HGs7hxLt3SA+uZDG7Oz+e990JWCQmwAnj+qSufqao8e1OZdlivKA3AA4P4dd9lt/HaFkvKyQL4E2YiIiIiIh9cMBMRERER+eCCmYiIiIjIBxfMREREREQ+uGAmIiIiIvIxpZSMwdqQOHZjZ3jUxYkJyRbcXVkZx6+HDSqviw4pne7tTXj/It5kiqiD0yf6ehfBeiSIu9zbmmfAek/XLLyfND72cFBJglASAMIRnDyQrONzX6niztmg8grerhm4+3TsGO4U3nsAdzN317yfKxPF5+bY4CCs20qXsItPjVghfN+kMziVIZNQUhlGRhu+ZkjGf8uOj4mZlMBQUV7Vmk7ie7U5ihMlQi5+Jauj7Cdo447msdFRT82E8D5iKXxvhCzvPkREonE8blq68HzQPBMnPnT3nQ7rqQjefyqOXyieL+BX0GoJD1pKhvLWXomE8byi6ZvlHeMHDhyA23b04G78888/H9Y7lRQDo3ymSh3PWaEQPpfupOfO5K+nq7GxIalWG8dsQEk7EIPHZHsUP/9SffhV18EAnuSPD+I5IgYSIpo78fxuOfi6Do0dg/VSGQ+OgTGcbpEdxfsJWPh+TKeVZ7Ry/6Uz+DXjRpkP3RA+l8Yo3zeI5yAnghO6gmHvmsrO4rVdsYLnq3oFP0PG8ng/I6M4Fazi4ldbl0rK68EnPcvKTMkgIiIiIjo1uGAmIiIiIvLBBTMRERERkQ8umImIiIiIfHDBTERERETkY0opGc2ppITDjZ2zjpLUEKritINaAXdGmhjuYrZiuLu+LZ2B9aAd8R5LMAG3TaW6YT0L3pMuIjJ6HHd7tmVwt3zFwdtHvIf4Wj2JL0dzCqdYlAq4Y7daxd+3RXkXe7GIr1WwjLuTy1XclXqo/7CnNqOzF3/PIdwNXK/gzyQBXI8nte533C09PjIO60MDQw1fax2201EolRAn2TiG3KCSvKDck6J0RNezo7CeqCudy8r3jbV3eGq7cvg+HTd4fHS2teBvWcX3Qnsnnj96++bAuth4LjPK/otFfM5cgzvgozE8z2mJErEYThEJh/GYCiid9E1N3tSDGT143Ld1dMH6rFl9sB5VEoK05A8tDcNx8HMqEGhMctDSAqabXC4rbn1SIlYI379lJaUgGMTXorsDp2Qk4vhaj3ZmYN0Cw6CrAydiRaI47SFbwPNMR8w7n4iIuMqSaUdhANYLBSW5ZhTPQdEQ/hlmJKWlW+BzJnV8rVpbpzY+XKPM8wHv8zukpKIUcVCI1Ot4e9fF5zhfUebz7BCsBwQfezg4Kf3FViK4AP6EmYiIiIjIBxfMREREREQ+uGAmIiIiIvLBBTMRERERkQ8umImIiIiIfEypJbg32inRSGOXZQWHHUg8gDsgqw7uSs2O4VbKY/uPw3r3PNzF2jPP2+leGMXdktu3boX1rh7c/T6Yw8cST+PYi+Zm3LUeVtIwyuVRWK+WlOQB5epFwHveRURCyjUZruDv29yK318fquOO3ZGxEW8xgP9N5jj44AdGwD5EJBzD17A5k4F1y8L3WaGI69akfztO/no6a+3qkFQq1VA77+IL4bbtEdydPbR3F6wfGcFjKg8SV0REggl87w2CZIdNQ6NwWzuhpMjUlGQUpc37+MgwrCcyePzZNr6ntGSGsIM74ENK2oa2n4gSyxNXUgmiUXyOtfQM9Lm0bZ0Q3rcI7lSfnGJxglG697Vjt1CkAth/paK09E8z8WBYYpNSMcplnNrihPEiIOTg+9StjcJ6IoLHR6ob379lkJpg2/h+cQIpWG9K4DFjO/g+yhWzsJ6IK2lhguvJKD43TcrzrKkVr0mi0U5Yz40eg/VC8SisJzJK2kYNJ4YJuOS2q6QAKWtEUcbw5CS2E4bH8bl38CUX28VjfizfOM+XSkoiE9rnSW9JRERERDQNccFMREREROSDC2YiIiIiIh9cMBMRERER+eCCmYiIiIjIx5RSMq6+/COSjDcmP+zbuQlue3wcd7+7Ydw12qx0OhbHR2E93oH3U6l7O93HxvD73KOhLljPjuHvWXHHYX1Gcy+sByK4C7Rex93GRaUDNy+4U7VUx92dsTDuCDY13AFed3E9P34E1jvb58F6S7zJ+z1tfIzRFL5+3U2tsG7F8DGGgvjGyR7D16qjcwasu4lYw9eFgtIdPA3VS2WpO43no729HW4bWXQ2rFfx5ZZtg/2wnje4w9lJxmB9z4B3vimX8T0TjuEEh+PHlRQcG0+TgSDu5q7XcFt4OIa7/VPpNKwnExm8nzDu4HccfJK1lAwtyUJLpphaHV8/21LOZQDXteQPLXFEO8aT3V47h9NN1dSl6jaes1IFpzWFo/icxaL4WoRs5TlUxeMmEsT3O0qnqhp8r2dzY7BuXG/ShohIQBkbYimpT8qPHu0wHgctTTiVIp3Bc0EqiZ+LoRDePjeG0zCCyjgwVfy8Czj4PBTA3OoKPvehIJ6z8y5OKyqW8DpoXFmXlYs43SioXJNYovGcWXV8DyD8CTMRERERkQ8umImIiIiIfHDBTERERETkgwtmIiIiIiIfXDATEREREfmYUkrGSxtflli0sWP15RfWwW27+3ACRd9CnChhh3CiRFB5H30whg99KDvi3bfS8RpW3jUeCuJ660zc2Wo7+FgGj+AEgJI7BOtN7UpCRCgD62GjdIcqV7Wcx52w2vvu3Rje/8AI7sAtF7xd1H1tCbCliGXhDupKCX9PU8XXxI3gjms7iDtfi2YU1uPRxs7ZoOG/JU8YHjgmlUJjR/OhI/vhtscGcLLK/uFRWK/GM7Beb8LnvxDC942Jeu9td3QYblur4XvDUpId2ntwskpnRzest7a2wXpaScNoavKmy4iIRKN4vgkq6RxaEsRUkyY02vaobln4XAZsPNdM9di1/U/1MxFmhSyxnMZz7JZwioVt42e3E8LXSH2m23hcOmGctBSOZjy1SrkOt62H8b7rrrLuCOLPGgjitYSWyJDMdMB6LIETJbRkn3wBPxfdym5YDyjpUeFoEtZrZZwQFI3iOc4EvOfBVdImjDJW6waf45qS2BFWPlNFGfOWkibkBhrPvbGViDaAswsRERERkQ8umImIiIiIfHDBTERERETkgwtmIiIiIiIfXDATEREREfmYUkrGt//1EQkFGjsKs7lxuO2cwbNhfc/+AVifNRt3Y/adgTvOyxX8vnFT83Zk2koL6/A47qKf2Yy/Z6KWgfWq602HEBFp7W6G9UIdd8KaKu4aPX5wDNbz7iist89sh/VMDB9PNo+P38RwB3HcwV2vxQFvJ++RMZyakMrgLuHScA7WwwZ3vJbCSod2C+7EHhvG7683kcb7uljEnbrTUS6XFRF3Ug2fx5zS5V2r4zEYi+DkCDuJr1/NxV3wLWnvfTA6WoTb5pVjbMpkYP1d73oXrM+dcxreTzNOvUgkcGKM4+DkiEBgStOznkwRwF3g2vYaLYEC7UdN1LBOPmljqt+TTp1o9LX/v16thK+FKeO5oDSK59BIG07QSqTxHC+C0xciUe/cEYvg50pTUHl+1HH6hGXwmCxV8GdKJvH+W5rwZ9WSuI4c2QXrdWsU1t0anuNC+PBFqvjzOsabLiYiUhrD820kOd9TK4syZ9dwKlithvddrSrfM4znw6CDkz9KeDfiWq7v1374E2YiIiIiIh9cMBMRERER+eCCmYiIiIjIBxfMREREREQ+uGAmIiIiIvIxpTbsS9/9HomEG98h/vzaNXDbwYHDsF6o4lSDkTHc7Tk2PgrrmU68n6rr7ag1SkrGWA63UbbWZ8B6OIhbT7MFnNgRT+OueFPFHbVFJT0jGKzBeqyKu1LtvNIt36IkFaTwfupV3F0fdeK43uPtUB4b1j6T9130IiJz5uLkgYDy3vlCFt83lSo+B2HB17Aw1NjpXSzhbuLpKBQOSSjceI+kEkpnspJi4YL0GhGRZAjfS+UyPv/VKu6Yb614k21Sza143za+l84+Cyf7XHbBxbDepuw/pLSoB5W0Cj0hQkmC0MpKcsSpSpo4FSkZ2sGfqiQPjTH4eUGK3IiI27g8CFRwmtJ4DqcgKI8niQfx+KuU8TMhGMH3gGu8y5dAFKdSOGGcEGXV8EGGHZx0E3fxfWQrc4o2hHM5nEoRUM6Z1PEaI5XGnyufHYT1cAAfZ9DFz+njhzfBetfpcz215mb87B4dw2lkpRKey8tFXC8paSzaSUsmcQJTJNr47CoW8GdH+BNmIiIiIiIfXDATEREREfnggpmIiIiIyAcXzEREREREPk6q6e9Ew0Sp4v3l6Gpdeb2h8lrCag3/Qnelin/xuqQ0/xRL+Jfva663Qc4or6Esl5VGsiJuJCsrr8UsFpXXTRbw6S3XlM9UVM5BSXndcE3bXts/Pv6K8gv/deVV3VLH+6kb70XXfrE/4CqNhi7et9b0pzXnVY3SKFJXmkUn3WcnzuF0bhY68dnzeW+zRV55nXqhgMdOURlTpRK+3nrTH26ARfNHBcxXIiIV5VWoJWV85HL4de2REG5SOlVNf2rD2ylr+sP70Zyapj/l9bbKa8D1/UyNNo6DwcbvOz4+7rv9O92Jz10oescZqomIlEp4PBm8ueQK+A+MweMjWMc3atV4x3dAedV1xeD5p1rDx1J28NiuK01/uRyeO9Smvzw+znwePy+rIaUxzcb7KeTx9lWlOdEFYQkiIvkCXshlweet2viZkFc+a0Fptisq91m5otxQAXySrSD+TGbSfVP8dZPhyYx5y5zEVocOHZLe3t433BnRO83Bgwelp6fn7T6MtwXHPU1X03Xcc8zTdHUyY/6kFsyu68qRI0ckmUyespgfov+fGWMkm81Kd3f3Kfsp1+8ajnuabqb7uOeYp+lmKmP+pBbMRERERETT1fT7JzQRERER0RRwwUxERERE5IMLZiIiIiIiH1wwExERERH54IL5TXrve98rn/nMZ3y36evrky9/+ctT3vfdd98t55577htud+edd8oNN9xw0vutVCrS19cnL7744pSPiYh+e/bt2yeWZclLL700pb9XqVTktNNOk2efffak/86WLVukp6cH5mwTUSNjjNxwww3S3Nz8G43RU+Wyyy6TVatWnfT2x48fl/b2djl06NBv8aimBy6Y3wLr16+f0oJ2Kvr7+2XlypXyuc99bqJ23333yQUXXCDJZFLa29vlD/7gD2T79u0Tf+44jvzN3/yN3Hbbbb+VYyJ6JzmZfxS/3b7+9a/L7Nmz5ZJLLpmobdiwQa666irJZDLS0tIiN9xwQ8NLWM4880y5+OKL5cEHH3w7Dpnod8qPf/xj+Zd/+Rf54Q9/KEePHpWzzjrrLT+GH/zgB3Ls2DG59tprG+rPPfecXHHFFRKPxyWVSslll1028bKo1tZW+dM//VO566673vLjfafhgvkt0NbWJrFYTP3zahW/keZkPProo3LJJZfIrFmzJmpPP/203HjjjfL888/LU089JdVqVa6++uqGnyR97GMfk2eeeUY2b978G39vInqNMUZqylvD3orv/fDDD8v1118/UTty5IhceeWVctppp8m6devkxz/+sWzevFn+7M/+rOHvLl++XL72ta+9bcdO9Lti9+7d0tXVJZdccol0dnZ63hIpor9d9FT5yle+IsuXL2/IC37uuefk937v9+Tqq6+WF154QdavXy833XRTwzbLly+Xb3/72zI8PPxbPb53PENvytKlS82NN95obrzxRpNKpUxLS4v5/Oc/b1zXndhm1qxZ5qGHHpr4WkTMI488Yn7/93/fxGIxc9dddxljjLnvvvtMe3u7SSQS5hOf+IS57bbbzDnnnOP7/RcuXGgefvhh320GBgaMiJinn366oX755Zebz3/+81P6vETTybJly4yINPx/7969ZvXq1UZEzI9+9CPzrne9y4RCIbN69WqzbNkyc8011zTsY8WKFWbp0qUTX9frdfOlL33JzJ071ziOY3p7e80999xjjDFm7969RkTMxo0bjTHG1Go1s3z5cjN//nyzf/9+eIzr1683tm2b8fHxido3vvEN097ebur1+kTtlVdeMSJidu7cOVErl8smHA6bn/70p2/yTBG9c02eB2bNmmWM+e/n/4oVK0xLS4t573vfa4wx5he/+IW54IILjOM4prOz09x2222mWq1O7G98fNxcd911JhaLmc7OTvPggw+apUuXmhUrVqjHMDAwYCzLMps2bWqoX3TRRSf1HJ89e7Z59NFHp/7haQJ/wnwKPPHEExIMBuWFF16QlStXyoMPPiiPPvqo79+5++675Q//8A/l1VdflU984hPy3e9+V+6++26599575cUXX5Suri555JFHfPcxPDwsW7ZskfPPP993u7GxMRERaW5ubqhfeOGFsmbNmpP4hETT08qVK2XJkiXyyU9+Uo4ePSpHjx5teHXw7bffLvfff79s3bpVFi1adFL7/Nu//Vu5//775c4775QtW7bIqlWrpKOjw7NduVyWj3zkI/LSSy/JmjVrZObMmXB/a9askdNPP12SyWTD33Ucp+GnTNFoVEREnnnmmYma4zhy7rnnch4g8rFy5Ur54he/KD09PXL06FFZv379xJ898cQT4jiOrF27Vr7+9a/L4cOH5QMf+IBccMEF8vLLL8vXvvY1eeyxx+See+6Z+Du33HKLrF27Vn7wgx/IU089JWvWrJENGzb4HsMzzzwjsVhMFixYMFEbGBiQdevWSXt7u1xyySXS0dEhS5cubRjjJ/B5/+Z5/5sCTVlvb6889NBDYlmWzJ8/X1599VV56KGH5JOf/KT6d6677jpZvnz5xNfXXnutXH/99RP/WfWee+6Rn/70p1IqldR9HDhwQIwx0t3drW7juq585jOfkUsvvdTzO1fd3d2yf//+k/2YRNNOOp0Wx3EkFotJZ2en58+/+MUvylVXXXXS+8tms7Jy5Up5+OGHZdmyZSIiMnfuXHn3u9/dsF0ul5MPfvCDUi6XZfXq1ZJOp9V97t+/3zMHXHHFFXLLLbfIP/zDP8iKFSskn8/L7bffLiIiR48ebdiW8wCRv3Q6LclkUgKBgGcemDdvnvz93//9xNef+9znpLe3Vx5++GGxLEvOOOMMOXLkiNx2223yhS98QfL5vDzxxBOyatUqed/73iciIo8//rjvc1zktXHe0dHR8I/gPXv2iMhrP4D7x3/8Rzn33HPlySeflPe9732yadMmmTdv3sS23d3dsnHjxjd9LqYz/oT5FLj44ovFsqyJr5csWSI7d+6Uer2u/p3JPxXeunWrXHTRRQ21JUuW+H7fE7/UH4lE1G1uvPFG2bRpk/zbv/2b58+i0agUCgXf70FEujf6rzuTbd26Vcrl8sSDUvPRj35U8vm8/OQnP/FdLIu8Ng9MngMWLlwoTzzxhDzwwAMTi/3Zs2d7HrginAeI3ozzzjuv4eutW7fKkiVLGtYEl156qeRyOTl06JDs2bNHqtWqXHjhhRN/nk6nZf78+b7fB41z13VFROQv/uIvZPny5bJ48WJ56KGHZP78+fLNb36zYVuO8zePC+a3STwef9P7aG1tFRGRkZER+Oc33XST/PCHP5TVq1dLT0+P58+Hh4elra3tTR8H0XQ1eRzbti3GmIba65t6T/xaxBv5wAc+IK+88oo899xzb7hta2srnAOuu+466e/vl8OHD8vQ0JDcfffdMjg4KHPmzGnYjvMA0W/uVDzLTwYa511dXSLyWuLN6y1YsEAOHDjQUOM4f/O4YD4F1q1b1/D1888/L/PmzZNAIHDS+1iwYAHcj5+5c+dKKpWSLVu2NNSNMXLTTTfJ97//ffn5z38us2fPhn9/06ZNsnjx4pM+RqLpyHEc3/9a9HptbW2eX3l4fV7rvHnzJBqNys9+9jPf/fzVX/2V3H///fLhD39Ynn76ad9tFy9eLNu2bfMs1E/o6OiQRCIh//7v/y6RSMTzKyScB4hOnQULFshzzz3XMB7Xrl0ryWRSenp6ZM6cORIKhRp+D3psbEx27Njhu9/FixdLf39/w6K5r69Puru7G2JjRUR27NjRkJwlwnF+KnDBfAocOHBAbrnlFtm+fbt85zvfkX/6p3+SFStWTGkfK1askG9+85vy+OOPy44dO+Suu+56w8g327blyiuv9PyC/4033ijf+ta3ZNWqVZJMJqW/v1/6+/snfoXjhDVr1sjVV189peMkmm76+vpk3bp1sm/fPjl+/PjEfwZFrrjiCnnxxRflySeflJ07d8pdd90lmzZtmvjzSCQit912m9x6663y5JNPyu7du+X555+Xxx57zLOvT3/603LPPffIhz70IdjEc8Lll18uuVzOM188/PDDsmHDBtmxY4d89atflZtuuknuu+8+yWQyE9vs27dPDh8+LFdeeeUUzggRaT71qU/JwYMH5dOf/rRs27ZN/uM//kPuuusuueWWW8S2bUkmk7Js2TL57Gc/K6tXr5bNmzfL9ddfL7ZtN/wax2SLFy+W1tZWWbt27UTNsiz57Gc/K1/5ylfke9/7nuzatUvuvPNO2bZtW0PMZKFQkF/96ld83r9Zb2tGxzvA0qVLzac+9Snzl3/5lyaVSpmmpiZzxx13vGGs3Pe//33Pvv7u7/7OtLa2mkQiYZYtW2ZuvfXWN4yV+9GPfmRmzJjREB8lk2KwTvz/8ccfn9jm2WefNZlMxhQKhd/0oxNNC9u3bzcXX3yxiUajnli5kZERz/Zf+MIXTEdHh0mn0+bmm282N910kydW7p577jGzZs0yoVDIzJw509x7773GGG+snDHGPPDAAyaZTJq1a9eqx/jHf/zH5vbbb2+offzjHzfNzc3GcRyzaNEi8+STT3r+3r333mve//73T+2EEE1DDz300ESc3AlaFNxvEit34YUXesbwZLfeequ59tprPfX77rvP9PT0mFgsZpYsWWLWrFnT8OerVq0y8+fPP/kPS5BljPLf8eh3gjFGLrroIrn55pvlox/96En/vT/5kz+Rc845R+64447f4tER0VvhlVdekauuukp2794tiUTipP5OpVKRefPmyapVq+TSSy/9LR8hEWny+bzMmDFDHnjggYafDE/W398vCxculA0bNnh+5cLPxRdfLH/9138t11133ak43GmLv5LxO86yLPnnf/7nKb2pq1KpyNlnny0333zzb/HIiOitsmjRIvnSl74ke/fuPem/c+DAAbnjjju4WCZ6i23cuFG+853vyO7du2XDhg3ysY99TERErrnmGt+/19nZKY899pinoc/P8ePH5Y/+6I+m9AM1wvgTZiIiIqK3yMaNG+XP//zPZfv27eI4jpx33nny4IMPytlnn/12Hxr54IKZiIiIiMgHfyWDiIiIiMgHF8xERERERD64YCYiIiIi8sEFMxERERGRDy6YiYiIiIh8cMFMREREROSDC2YiIiIiIh9cMBMRERER+fh/lj9P64SGxQYAAAAASUVORK5CYII=\n"
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2U9-305aDfeq"
      },
      "source": [
        "## Utilizando transferencia de aprendizaje en nuestro modelo"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VqyhAe1NtQbH"
      },
      "source": [
        "### ¿Qué es Kaggle Models (anteriormente TensorFlow Hub)?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vmoBcM2AtQbH"
      },
      "source": [
        "Un concepto que es esencial en el desarrollo de software es la idea de reutilizar el código que está disponible a través de librerías. Las librerías hacen que el desarrollo sea más rápido y generan más eficiencia. Para los cientificos de datos que trabajan en visión artificial, entrenar arquitecturas de redes neuronales complejas desde cero resulta altamente complejo. Kaggle Models (anteriormente TensorFlow Hub) es una biblioteca que permite publicar y reutilizar componentes de aprendizaje automático prediseñados. Con Kaggle Models, resulta sencillo volver a entrenar la capa superior de un modelo previamente entrenado para reconocer las clases en un nuevo conjunto de datos. Kaggle Models también distribuye modelos sin la capa de clasificación. Estos se pueden utilizar para realizar fácilmente el aprendizaje por transferencia. Puede descargar cualquier [modelo para imágenes compatible con Tensorflow 2 de Kaggle Models](https://www.kaggle.com/models?task=16686). La única condición es asegurarse de que la forma de las características de la imagen en nuestro conjunto de datos preparado coincida con la forma de entrada esperada del modelo que desea reutilizar."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qKXKSWBotQbI"
      },
      "source": [
        "### Antes de comenzar"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VxbNqO1ntQbI"
      },
      "source": [
        "Antes de comenzar necesitaremos verificar que tenemos el runtime correcto en nuestro ambiente. Esta tarea se beneficiará mucho de una GPU."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ABQ6LdkQDfeZ",
        "outputId": "0a12e761-5b11-4e27-e5d1-97b3ff4d15a1"
      },
      "source": [
        "import tensorflow as tf\n",
        "print(\"GPUs disponibles: \", len(tf.config.experimental.list_physical_devices('GPU')))"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "GPUs disponibles:  1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cfrHfmyPDfey"
      },
      "source": [
        "Adicionalmente, configuraremos nuestro conjunto de datos para una tarea supervisada. Esto lo hacemos especificando el parametro *as_supervised=True* lo cual nos da la posibilidad de acceder a las anotaciones del conjunto de datos."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rF2k-8vvDfey"
      },
      "source": [
        "(training_set, validation_set) = tfds.load('cifar10', split=['train[:10%]', 'test[:10%]'], as_supervised=True)"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_E_jYnBLtQbJ"
      },
      "source": [
        "### Ajustando el tamaño de nuestras imagenes"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eW5VvqgHdfuF"
      },
      "source": [
        "El modelo que utilizaremos más adelante será ResNet. Veremos más en detalle este modelo luego, pero lo importante aquí es que es un modelo que se entrenó utilizando imagenes de tamaño 224x224."
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_url = \"https://www.kaggle.com/models/tensorflow/resnet-50/TensorFlow2/feature-vector/1\""
      ],
      "metadata": {
        "id": "AjamZs5z-NyE"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oUFqywzse-QZ"
      },
      "source": [
        "> *Nota: Para averiguar las propiedades de la imagen, puede directamente dirijirse a la página de Kaggle Models y verificar la documentación del modelo que está utilizando.*"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S0W4XWuRtQbJ"
      },
      "source": [
        "Sin embargo, nuestras imagenes podrían no tener este tamaño e incluso podrían no tener el mismo tamañao. ¿Como hacemos para ajustar el tamaño? En general existen 2 técnicas:\n",
        "\n",
        "#### Padding\n",
        "\n",
        "Podemos utilizar una una técnica similar a cuando ajustamos la longitud de las secuencias en NLP, **Padding**. Padding agregará ceros en en la imagen hasta completar el tamaño especificado:\n",
        "\n",
        "<img src=\"https://th.bing.com/th/id/OIP.8D8BXN3aHLCyTNTE2GssmQAAAA?pid=Api&rs=1\" />"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fbVI1vxrtQbK"
      },
      "source": [
        "En TensorFlow, esto lo podemos realizar utilizando el metodo `padded_batch` el cual agrupa las imagenes en batches para entrenamiento y al mismo tiempo ajusta el tamaño de los tensores al necesitado:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_DSM53YdtQbK"
      },
      "source": [
        "BATCH_SIZE=64\n",
        "EPOCHS = 3\n",
        "\n",
        "training_set_batch = training_set.padded_batch(batch_size=BATCH_SIZE, padded_shapes=([224, 224, 3], [])).cache().repeat()\n",
        "validation_set_batch = validation_set.padded_batch(batch_size=BATCH_SIZE, padded_shapes=([224, 224, 3], [])).cache().repeat()\n",
        "train_size = info.splits['train[:10%]'].num_examples\n",
        "test_size = info.splits['test[:10%]'].num_examples"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CBovAgmkxm4S"
      },
      "source": [
        "> **Imporante:** Notará que en este ejemplo estaremos utilizando el 10% del conjunto de entrenamiento. El motivo de esto es simplemente para reducir el tiempo de entrenamiento del modelo. ResNet es un modelo computacionalmente costoso y por lo tanto requeriremos de bastante hardware para ejecutarlo. Más adelante utilizaremos un modelo con menos performance pero conmucho menos requerimientos de cómputo."
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Resizing\n",
        "\n",
        "Otra alternativa es cambiar el tamaño de las imagenes hasta que alcanzen el tamaño que necesitamos. Cuando las imágenes que trabajamos son más grandes que las que nuestro modelo opera, en general utilizamos esta técnica. Esto se puede realizar o bien como parte del preprocesamiento o como parte de la red utilizando una capa de reescalado. En este notebook utilizaremos esta técnica más adelante.\n",
        "\n",
        "```\n",
        "        tf.keras.layers.Resizing(\n",
        "            224, 224, interpolation=\"bilinear\", crop_to_aspect_ratio=False\n",
        "        ),\n",
        "```"
      ],
      "metadata": {
        "id": "-0KFsmFGa5vc"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PziBXrSoDfeq"
      },
      "source": [
        "### Construimos un modelo basado en ResNet50"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-vp215uiDfer"
      },
      "source": [
        "Construiremos nuestro modelo basado en ResNet50. ResNet-50 es una CNN de 50 capas de profundidad que utiliza el concepto de skip-connection, de una forma similar a la que se utiliza por el sistema de visión humano. Esta implementación le permite a la red reutilizar las activaciones de capas anteriores y asi no solo mapear conceptos que se aprendieron en otra jerarquia de la red sino que también evitar el problema de desaparición de los gradientes."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ABF9i2EzDfev"
      },
      "source": [
        "Instanciamos nuestro modelo y verificamos su arquitectura.\n",
        "\n",
        "El namespace `tensorflow.keras.applications` dispone de capas listas para utilizar en modelos más complejos. En nuestro caso el modelo es un `ResNet50`:"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.applications import ResNet50\n",
        "\n",
        "base_model = ResNet50(include_top=False, weights='imagenet')"
      ],
      "metadata": {
        "id": "c9pfiOhw0m5h"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Usaremos al modelo base como un *feature extractor* por lo cual no queremos entrenar los pesos especificos de este modelo:"
      ],
      "metadata": {
        "id": "LLiwIzc304Bi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "base_model.trainable = False"
      ],
      "metadata": {
        "id": "_29P1EDz00hB"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Construimos el modelo:"
      ],
      "metadata": {
        "id": "nOwwb0Lf1U3F"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4QC17MxkDfes"
      },
      "source": [
        "import tensorflow_hub as hub\n",
        "from tensorflow.keras import preprocessing\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.InputLayer(shape=(224, 224, 3)),\n",
        "    tf.keras.layers.Resizing(224, 224, interpolation=\"bilinear\", crop_to_aspect_ratio=False),\n",
        "    tf.keras.layers.Rescaling(1 / 255.0),\n",
        "    base_model,\n",
        "    tf.keras.layers.GlobalAveragePooling2D(),\n",
        "    tf.keras.layers.Dense(128, activation='relu'),\n",
        "    tf.keras.layers.Dense(10, activation='softmax')\n",
        "])\n",
        "\n",
        "model.compile(optimizer='adam', loss=tf.keras.losses.SparseCategoricalCrossentropy(), metrics=['accuracy'])"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 346
        },
        "id": "X46DBEQnDfev",
        "outputId": "3d322ea4-b163-4074-c7d1-7a22f3fcff8f"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential_4\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_4\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
              "│ resizing_4 (\u001b[38;5;33mResizing\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ rescaling_4 (\u001b[38;5;33mRescaling\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m3\u001b[0m)    │             \u001b[38;5;34m0\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ resnet50 (\u001b[38;5;33mFunctional\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m2048\u001b[0m)     │    \u001b[38;5;34m23,587,712\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ global_average_pooling2d_3      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2048\u001b[0m)           │             \u001b[38;5;34m0\u001b[0m │\n",
              "│ (\u001b[38;5;33mGlobalAveragePooling2D\u001b[0m)        │                        │               │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense_8 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)            │       \u001b[38;5;34m262,272\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense_9 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │         \u001b[38;5;34m1,290\u001b[0m │\n",
              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
              "│ resizing_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Resizing</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ rescaling_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)    │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ resnet50 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Functional</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)     │    <span style=\"color: #00af00; text-decoration-color: #00af00\">23,587,712</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ global_average_pooling2d_3      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePooling2D</span>)        │                        │               │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">262,272</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,290</span> │\n",
              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m23,851,274\u001b[0m (90.99 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">23,851,274</span> (90.99 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m263,562\u001b[0m (1.01 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">263,562</span> (1.01 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m23,587,712\u001b[0m (89.98 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">23,587,712</span> (89.98 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FsdjZHdmPSO5"
      },
      "source": [
        "#### ¿Porque hay una capa de `Rescaling`?\n",
        "\n",
        "Las imágenes de entrada están codificadas como arreglos de pixeles que contienen la luminosidad en cada uno de los canales. Esta luminosidad puede ir desde 0 hasta 255. Sin embargo, `RestNet` fue entrenado utilizando imágenes que fueron normalizadas. Es decir, cuyos valores están entre `[0, 1]`. Como regla, debera siempre aplicar el mismo preprocesamiento que el modelo realizó originalmente sobre los datos de entrada. Keras dispone de una capa que nos permite realizar esto.\n",
        "\n",
        "```\n",
        "preprocessing.Rescaling(scale=1./255.)\n",
        "```"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wUr_78wVDfe4"
      },
      "source": [
        "Comenzamos el entrenamiento"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GKNXfxsGDfe7",
        "scrolled": true,
        "outputId": "d1cf0beb-f046-4536-c6c2-f43215db82c1"
      },
      "source": [
        "import math\n",
        "\n",
        "steps_per_epoch = math.ceil(train_size / BATCH_SIZE)\n",
        "validation_steps = math.ceil(test_size / BATCH_SIZE)\n",
        "\n",
        "history = model.fit(training_set_batch,\n",
        "                    epochs=EPOCHS,\n",
        "                    steps_per_epoch=steps_per_epoch,\n",
        "                    validation_data=validation_set_batch,\n",
        "                    validation_steps=validation_steps)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "\u001b[1m79/79\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 289ms/step - accuracy: 0.0955 - loss: 2.3551 - val_accuracy: 0.1030 - val_loss: 2.3346\n",
            "Epoch 2/3\n",
            "\u001b[1m79/79\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 165ms/step - accuracy: 0.0908 - loss: 2.3269 - val_accuracy: 0.1030 - val_loss: 2.3149\n",
            "Epoch 3/3\n",
            "\u001b[1m79/79\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 167ms/step - accuracy: 0.1030 - loss: 2.3197 - val_accuracy: 0.0990 - val_loss: 2.3075\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "> ¿Que le parece que está sucediendo con este modelo? ¿Por que la pérformance del mismo no es la que esperamos?"
      ],
      "metadata": {
        "id": "rtyFaC5ttkD2"
      }
    }
  ]
}