Face Recognition/Face_recognition_deep_learning.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Face_recognition.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "mount_file_id": "14vpbPYwu12sip_ZKAzlxuTy0nCNEGjed",
      "authorship_tag": "ABX9TyNZsqL/arSVwR2Os5MgYg/J"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JI9BSNizPcQh",
        "colab_type": "text"
      },
      "source": [
        "Face Recognition using face_recognition library"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3IzIh3fFPmVY",
        "colab_type": "text"
      },
      "source": [
        "Dataset used contains ~750 faces images, there are total 5 faces and each face has ~150 images in the dataset.\n",
        "* Make Dataset.\n",
        "* Sample result.\n",
        "* Make Embeddings.\n",
        "* Predict Result."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8GapoBTG4kg9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 336
        },
        "outputId": "0889f307-333f-4822-9820-4ee4110746b4"
      },
      "source": [
        "!pip3 install face_recognition"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting face_recognition\n",
            "  Downloading https://files.pythonhosted.org/packages/1e/95/f6c9330f54ab07bfa032bf3715c12455a381083125d8880c43cbe76bb3d0/face_recognition-1.3.0-py2.py3-none-any.whl\n",
            "Requirement already satisfied: Pillow in /usr/local/lib/python3.6/dist-packages (from face_recognition) (7.0.0)\n",
            "Collecting face-recognition-models>=0.3.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/cf/3b/4fd8c534f6c0d1b80ce0973d01331525538045084c73c153ee6df20224cf/face_recognition_models-0.3.0.tar.gz (100.1MB)\n",
            "\u001b[K     |████████████████████████████████| 100.2MB 41kB/s \n",
            "\u001b[?25hRequirement already satisfied: dlib>=19.7 in /usr/local/lib/python3.6/dist-packages (from face_recognition) (19.18.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from face_recognition) (1.18.5)\n",
            "Requirement already satisfied: Click>=6.0 in /usr/local/lib/python3.6/dist-packages (from face_recognition) (7.1.2)\n",
            "Building wheels for collected packages: face-recognition-models\n",
            "  Building wheel for face-recognition-models (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for face-recognition-models: filename=face_recognition_models-0.3.0-py2.py3-none-any.whl size=100566172 sha256=5cd4ce27b9f7405d6d671317f43ee50719dda3b8f268277dbe1204210ae2cd41\n",
            "  Stored in directory: /root/.cache/pip/wheels/d2/99/18/59c6c8f01e39810415c0e63f5bede7d83dfb0ffc039865465f\n",
            "Successfully built face-recognition-models\n",
            "Installing collected packages: face-recognition-models, face-recognition\n",
            "Successfully installed face-recognition-1.3.0 face-recognition-models-0.3.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A9qDbjvb4bvH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import dlib\n",
        "import face_recognition\n",
        "import os\n",
        "import re\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AEDbX1IM6iNq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "rars = ['/content/drive/My Drive/Datasets/gates.rar','/content/drive/My Drive/Datasets/jack.rar',\n",
        " '/content/drive/My Drive/Datasets/modi.rar', '/content/drive/My Drive/Datasets/musk.rar',\n",
        " '/content/drive/My Drive/Datasets/trump.rar']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9y7dwZXu-Yyc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 131
        },
        "outputId": "ce3c719b-948f-46d5-bb1d-18e2e727c90f"
      },
      "source": [
        "!pip install patool"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting patool\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/43/94/52243ddff508780dd2d8110964320ab4851134a55ab102285b46e740f76a/patool-1.12-py2.py3-none-any.whl (77kB)\n",
            "\r\u001b[K     |████▎                           | 10kB 18.0MB/s eta 0:00:01\r\u001b[K     |████████▌                       | 20kB 1.7MB/s eta 0:00:01\r\u001b[K     |████████████▊                   | 30kB 2.3MB/s eta 0:00:01\r\u001b[K     |█████████████████               | 40kB 2.6MB/s eta 0:00:01\r\u001b[K     |█████████████████████▏          | 51kB 2.0MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▍      | 61kB 2.3MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▋  | 71kB 2.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 81kB 2.3MB/s \n",
            "\u001b[?25hInstalling collected packages: patool\n",
            "Successfully installed patool-1.12\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "390GhLGP7Jcz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "outputId": "3116f110-da43-4438-a2d0-024c48e76e5c"
      },
      "source": [
        "import patoolib\n",
        "for rar in rars:\n",
        "  patoolib.extract_archive(rar, \n",
        "                         outdir=\"/content/\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "patool: Extracting /content/drive/My Drive/Datasets/gates.rar ...\n",
            "patool: running /usr/bin/unrar x -- \"/content/drive/My Drive/Datasets/gates.rar\"\n",
            "patool:     with cwd='/content/'\n",
            "patool: ... /content/drive/My Drive/Datasets/gates.rar extracted to `/content/'.\n",
            "patool: Extracting /content/drive/My Drive/Datasets/jack.rar ...\n",
            "patool: running /usr/bin/unrar x -- \"/content/drive/My Drive/Datasets/jack.rar\"\n",
            "patool:     with cwd='/content/'\n",
            "patool: ... /content/drive/My Drive/Datasets/jack.rar extracted to `/content/'.\n",
            "patool: Extracting /content/drive/My Drive/Datasets/modi.rar ...\n",
            "patool: running /usr/bin/unrar x -- \"/content/drive/My Drive/Datasets/modi.rar\"\n",
            "patool:     with cwd='/content/'\n",
            "patool: ... /content/drive/My Drive/Datasets/modi.rar extracted to `/content/'.\n",
            "patool: Extracting /content/drive/My Drive/Datasets/musk.rar ...\n",
            "patool: running /usr/bin/unrar x -- \"/content/drive/My Drive/Datasets/musk.rar\"\n",
            "patool:     with cwd='/content/'\n",
            "patool: ... /content/drive/My Drive/Datasets/musk.rar extracted to `/content/'.\n",
            "patool: Extracting /content/drive/My Drive/Datasets/trump.rar ...\n",
            "patool: running /usr/bin/unrar x -- \"/content/drive/My Drive/Datasets/trump.rar\"\n",
            "patool:     with cwd='/content/'\n",
            "patool: ... /content/drive/My Drive/Datasets/trump.rar extracted to `/content/'.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6EUjDOQXBVmk",
        "colab_type": "text"
      },
      "source": [
        "Sample output "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yunk_rhXBAyc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 203
        },
        "outputId": "679b69fb-fe86-4e99-ce02-d346e5b56ac4"
      },
      "source": [
        "face1 = face_recognition.load_image_file('/content/jack/jack10.jpg')\n",
        "face2 = face_recognition.load_image_file('/content/gates/gates1.jpg')\n",
        "face3 = face_recognition.load_image_file('/content/jack/jack1.jpg')\n",
        "\n",
        "face_encd1 = face_recognition.face_encodings(face1)[0]\n",
        "face_encd2 = face_recognition.face_encodings(face2)[0]\n",
        "face_encd3 = face_recognition.face_encodings(face3)[0]\n",
        "\n",
        "faces = {'Encoding1': face1, 'Encoding2': face2, 'Predited': face3}\n",
        "encd = face_recognition.compare_faces([face_encd1, face_encd2], face_encd3)\n",
        "f, axs = plt.subplots(1,3)\n",
        "plt.tight_layout()\n",
        "for i, key in enumerate(faces.keys()):\n",
        "  axs[i].imshow(faces[key])\n",
        "  axs[i].set_title(key)\n",
        "  if i ==2: \n",
        "     axs[i].set_title(encd)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaMAAAC6CAYAAAADOyITAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9ebxkVXnv/X3W2lNVnaFPD/RAMw8qouLQIA5BiSgmApoocYiKScwnb8x9k9x7M90kN9fcDN7cmzeDRo0YxRm8aJzQBDUKGkURASOIMkNDN9DDGatqD2s97x9r7zp1Tp/TjNrdWr9PV5+qvdZea+29q9aznuf5Pc8SVWWEEUYYYYQRDiTMgR7ACCOMMMIII4yE0QgjjDDCCAccI2E0wggjjDDCAcdIGI0wwggjjHDAMRJGI4wwwggjHHCMhNEII4wwwggHHCNhNMIII4ywCkRERWRBRP78QI/l0UJEviwiv/Io23hzfT9URKLHamwwEkYjjDDCCA+Gp6jqH4rIc0Vkvn41E/L80OvIH9WAROQOEekt63/LD7tfVf0T4Ik/jLYfU8k2wggjjPDjClX9CjAGICJHA7cDa1S1Wl5XRKKVjj/GOEdVv/BD7uNHhpFmNMIII4zwKCEi/0NELhWRD4rILHCBiFwkIn82VOd5IrJ96PMWEfmYiDwgIreLyP/7KMcwJSKfqdvbW7/fukrd40XkChGZEZFdInLJUNnjReTzIrJHRL4vIuc/mnE9VIyE0QgjjDDCY4PzgEuBNcCH9ldRRAzwaeB64HDgp4HfEpEXPYr+DfBe4CjgSKAHvG2Vuv8TuByYArYCb63H1QE+D3wYOAx4JfB2ETnpUYzrIWEkjEYYYYQRHht8XVU/oapeVXsPUncbsEFV/1RVC1W9DbiQMPk/VHxCRKbr1ydUdbeqfkxVu6o6B/w5cMYq55YEobVFVfuq+tX6+EuAO1T1vapaqeq1wMeAVzyMcT0ijHxGI4wwwgiPDe5+GHWPAraIyPTQMQt85WG08dJhn5GItIG/Ac4maDwA4yJiVdUtO/d3CdrRN0VkL/DXqvqeelynLRtXBHzgYYzrEWEkjEYYYYQRHhss3wJhAWgPfd409P5u4HZVPeEx7P+/AI8DTlPVnSJyCnAtIPsMVHUn8EYAEXkO8AURubIe1xWqetZjOK6HhJGZboQRDjCG4z9E5DUicvmBHtMIjwmuA35GRNaKyCbgt4bKvgnMicjviUhLRKyInCwi22BAdni4+/uME/xE0yKyFviT1SqKyCuGyA17CYLUA58BThSR14pIXL+2icgTHuZYHjZGwmiEn3isErOxmuP3hwpV/ZCqvvCxaEtE/o+I3CwicyJyk4i87rFod4SHjA8QCAp3EMgCA8ZabTZ7CXAKgSK+C3g3MFlXOQL42sPs72+BVt3WVcC/7KfuNuAbIjIPfAr4TVW9rfY1vZDgu7oX2An8LyB9mGN52JDR5noj/KRDRO4AfuVAxWyIyJeBD6rqux/jdt8MfAT4AWHy+RfgZ1X14U5yP7EQkT6QA3+vqn/8I+z33cD/VdV//VH1+VAgIn8C/GeCcOqs4It6xBhpRiOMsApE5AIR+WqtYeytY0FePFS+VkTeKyL31uWfGCp7o4jcUsdqfGo4Ol5Ezqo1lZlaA5PlfQ59VhH5tVrDmRaRfxARqcusiPx1HSdyu4j8hgylaVHVP1HVm2p21zcIzvHTf6g37ccMqpqp6uSPUhDV/f7KwSaIAFT1zfX9yB5LQQQjYTTCCA+G04DvA+uBvwL+qREGBDNMm5Ae5TACkwkRORP4S+B8YDNwJ3BxXbYe+DjwR3WbtwLPfpAxvISg2Ty5brOJRXkj8GKCqedpwEtXa0BEWnUbNzykqz7IISJn1wGZt4jI7x/o8Yzw6DEy043wE4/aTLceGE7f8juEWIw/UtXj63ptAkNqM0GbuQdYp6p7l7X3T8BuVf3d+vMYwUl8AvBTwK+r6jPrMiEwmP6Hqr5bRC4gmAyfU5cr8NwmDkREPgp8W1XfIiL/Blyiqv9Yl72AELAYL09FIyLvAzYCL9ZD/EcvIpZgejwL2A5cDbxKVW88oAMb4VFhpBmNMELAS1V1zdDrwvr4zqaCqnbrt2MEB/Oe5YKoxhaCNtScNw/sJkTab2EoHqUWDA8Wn7Jz6H237r/pZ/jcFdsRkf8NnAycf6gLohqnArfUDveCoHWed4DHNMKjxCjOaIQRHhnuBtaKyBpVnV5Wdi8heBAYpFhZR9CkdhAEWVMmw58fJnYQUrk02KedmsTwYuAMVZ19hP0cbDicpYJ3O8GcugQi8qvAr9Yfn27EPmjDjaSWJce0PrY8XCeUNMeXS3kZOnc1yFDN5W00ZUtb2LfGg2Olka1eJiscfXhtrHDqUHVVv0tVNyyvNhJGI4zwCKCqO0Tkc4S8XW8C5oHTVfVKAoPtIyLyYeB7wF8A31DVO2oq7dtE5OcIlNo3sTQY8uHgo8BvishlBPPh7w0XisgfAK8mmPl2P8I+Dlmo6ruAdwFYE2krXbO/ukv+igiqDhGtjwnGxEN1HBDKAl9E9jk/fK4YnqxDfRn8beo2x4YV13AMVE39mXpMTTsWVRmqu1QoLPbThBA1sCwKQDc0JgHMkrZWvg6D1II91Fs5OfnyNhr0y/k7V6o/MtONMELAp5fFGf3zQzjntQS/0k3A/dRBjTVF/I8JOb12AMdR5xxT1V2EPF9vIZjuTgD+/RGO+UJC/Mp3CJH2nyXMDA3L6S8ICTNvGbqu//YI+zqYcA9LtcCt9bH9opnsV3t575cJpXDcGIMxZlBveG5dPtE2giW8D5+b8+saq45rtbHuD8OCbKV2VytbUQdbMnYZEkSL92Sl9w82voeKEYFhhBF+TFDTzt+pqkc9aOVDGDV1/QeETNf3EAgMr1bVVZmC1kSaJZP7rNaXT7jDUPWAEhJsA8jQ5OprraXRcGQgqETqMhTU1MJLhgRccx1LrmpoHEGTadoNmsziOYvtyNDx5WYzX7e3kkBY2dQ3rGUN35Pm9MWxydB5OtDW9kWtfUndp4Zj/XL+GlV9xvLaIzPdCCMcoqjp2s8naEcbCelfHopGd0hDVSsR+Q3gXwkz9Xv2J4iWY1FLGbQ3eL90IpbBa1+hNTz7Dk/oCqK1IGqODZ+/r3AYFpChzLEo3JbP8oua0HDRvnWHBUkwvy321QjZ5viiJjes8Qxf82LbQhCUbtDLarrPYt9+4F3bH0aa0QgHBUTkbODvCJPLu1X1LQd4SAc9aqr5FcDjCTnJLiOkdflxISo8Zmg0I3hopqOVBNRKZSucieqwVrNMUC1rb6WJP2hdrj5uEIlWNOMtN9Et1WSG41ENix4ZXVbW+I9WuJLVrlM8S3xQA01P8N6zXOscCCMNPrC8WhhpRiMcnKjjRv6BobgREfnUKG5k/6ip5tsO9DgOFawqBIbm3OXst+XO/P35YRrCwbC2sbzv5W2tMlICSaB5v2/9YZ/OSmNd3v9KxIil5reV/V8rCks1g/GF4ka4ybJ7TN3Jyvy85RgJoxEOBgziRgBEpIkbGQmjEX4oGDabBTqzEP75VesOv9+fttSYAVdy8u9fgAwLCrti3YeO/XHTLEt9V6v7zVYS4EOltUBqmHWwj2aoDe093OP9XcJIGI1wMOBB40aGY0astU8fHx9n8bex+qpr6e9Hlx1fmYEUfkSeLZsPI00t9z+wl/mF/grtrd73cPtLx1lHn6jifXAIZ1mLKI4pXEVVlgjKujUTTEx0iK1ZnCy80p2eweddxFd184oxFkXDWrheri6ZOPZzbPhj049XrZ3zICYM23mlWyidLCKSUChGuHlPvmLMyKGCcMmLAmi5T2c1YbI/IbOaD2r5+2FT22qC5sFNekPPbchEttJ1rjb25TTz4TqraW/L6emLx1YS3rVm9CDK0UgYjXBIYDhmZGpqrZ5xxvOGy1BVnHNYa5fQYhs79nDdhsLbHG8+h5dDvbB53Rj/8w/fRNb2fOwzX+fif/4CIgZXuSWTxzAdeNl4B300/TjX+AEWz62qin6/T2QMxxx/PJu3HsVd9+zktu130Vkfc87Zz+acM09j/Zo1RCIYjdhzyz3cesVnYfdNJKXDRoYcsCpIPZayLGloydZaiqIYHIOwehcJAZVGIIkMxoT5otvPqVSoKohjoXKOub7j+rtztj3uMNbFBUYMcWx40Qd/cOdj9pB/qFh00jd06+CPUbx3OFfVZqVhjUKHCGECslR4D2rpfvxBy4R+c3wlgbAS9hV69biGxtAsfFYmHKyOfQkQ+7a97IxVx9qQHwZjqHWhKIqIohhE8c5TVS4EQ6yAkTAa4WDAw4obEWEgdLz3GGPw3hNFSx29zQ/DWrtEIC13+i43lRijnHnG6bTTYLRZWOgiImRZRmwj9k5PDybz4fiTpr3hfobrLF91GmNIkoQoiqic446bb+XOm2/liOOP4TmnPpn7du/lvZdcwWcvv4pX/ewLOOu5z2BiMmXtCVtZv/WXuePaa7jrmi/T6d5H5IKQpO7Pe4+rjzVCadjhvXgMTH2ONUErMtZC5UEVr4o1BsRjjB1cl4kseohFKQpCHBviOELEDJ6D9xEiFuccXk0gD4gLJjsPYEEak1Rgog0EU61JiwjWRrWJTuv7r4gxtQYsA8HUPIvBuJZN/Kq+7ieM2pjFBVZQNHSpoiH1GLCDOvvTcBaPwSKzbiWtf3h8dhXf0nJCROMvc4iBJItpZUmtYSu9fsFCf2bFfkbCaISDAVcDJ4jIMQQh9EpC5oBVIAO7fKOZNBOwqmKtHWghwJJJuBEOMLRCNQaDIk4pNeJJJx3N85/7VEQqFuZzvn39DXhxTI4l/MYb38DfvO1Cdk13GV6NNu3tM9L99cuiacXaCNMy4B2333ord9x6O0cddwynnvIktj+wi//9gc9w6Re/yhtefjbPP+1JdDLDcadvY/MxR3LDlz5DcfuNZMaDREHVGZovmvsyMMUQzIORCbb8MKxgdhOCehQbizcOVcLY8EQWwKBSa2CP6QYCP1xYa5hcM84wq7v53kRRyKzgnKOsHEVZ4FxJVVZ4dVhriOMYG4XvWFmUOBcmcUWIIkOSxLTbbZIkGWicrqroFyVFUVKWBa7ygwl/6cSuQ8IHFF8LnKWaRhByhn21nlpTFz8ktJYy+VbSlBb7bgTqvvcttCcDATfcfnP+4DxRlLImOFiMxGRJQqvVIklijLe0M8+u3Tv37YiRMBrhIIA+iriRRkMaNp01Amm5OWS51mJtcBJ7dSAWNYbHH3s4v/Kac+mkHoj55Ge+wI77ZjBEjKcZxx91GK942Yt450WXImKXtPdgYRKNQFrJtCfqESNgIlq1Jnf7Lbdy5623cewJx3DKyU/inh338ua/u4jLn/YEfv3V53HiEVtINm3mqef/Erdd/U3uvupLrMt3YXxJUU8eTV9VVS0KRWnGA6gPTmYJggkRjAiKBLOnRBhja61IBwv8sMJ/8FxvBwviJGLDhrUURUVZFlRVRRRFtNotsizDGoNXT+VL8n5Jr5fTW+iT532iyJCmMVEc4b2jLGP6vYqiKDEG2u2Uzlib8fHx0JYN96WqKoqiotfvMTs7x8J8j6p0SxYG0PipNAghZECiENMIoRIji4JBlyR6AxkszBpGAuiQ2rqaT2l/psKVhNf+mITUI1ds0LJQRD2RMbTTjFaWEpmIKMn4/q0rP6ORMBrhoICqfpaQzuZBMWymEwmTZiOAYKkfp/nxNGXLhZMxBpxHPDzjlMfx6vPPZrId8n598SvXcdnnr8LYBCrP855zGqmpePJJx9HOYnr5vsyrFa5raNz7mk6W04WbMmstrVaLsiy55Xs3odFtnHzKKRx/9NFcd+MN/Kc//lt+6RdewjlnnUYniznhtG1sOmIrP/i3z1LtvB20t6qjW4ygvvGnNStuxdrapDg0tiaVjQzYUOEcj+4TPHowI7IRG9ZPkecFC70+eV6QJQmdVoc0S4miCBFwrqIoK3rdPnPJHN1+ijFCEgu2NgOXZYlIgfOB6JGkHcbGOox3WqRJiyi2WGOpqorSOTplizRLsPEsM9NzuKIA74MGW/uidJDfrQ4PlWAyDRpP7bpSxeOCiU4bU6EidfwOPqoXZcuv3tbaTV1fFpl0i8GwIFKbIOtxDDQuVmYUNuct0tCb0xTB47TCKyRRTCtJSLKE8YmJ1Z/RQ36aI4xwkGGYieScW+K/GQgaWOIvgmaFCeFHpMRRxNnPfzbnvehUkjSnUvjiV6/hwg9+Eo1TBM8pJz+OM5/7NER7VEWxxNyyPL6jaT8c02AWqzWx0nloBE9jqG98SwqqQbA2E1Ecx0R2Elfl/MfV32DT5k2c8czTufH2u/jriz7F1TfezJtecw7HbVlPZ+tGnvzyC7jty59n1/VfxepC0LjUDCY40eWspsbJDtbYut+gLBkBK4odmP1k4KdTY1hthXwwwhrDeKdNmsRESUJZORJrSaKYJEmIo6D5APR7fSyC4IjTFDGGWBxSa4JFWeJVyIuSqlTSJKOdtmilGVmcEaXBd1RVFVFZksQRxgZfiq8q5mZKPH6gZXrvcd4NTHWKDrQddX4gfBrzH42W5JVFJqAg9XcHwJhFsow1CpjFGKoljLcwLtWGcrB6roTlZrrF441PFJrMDCB4hX6RI0AaJ4yNjTE1ObnqMxoJoxEOQSzVdoZJDCvRZRsh1dTxjR1chVZqOP+lL+TZT3sckVRU3vLpf72Sj3zii2icYrEcf9QWfu2ClzIWeSosn//q1+mXFUszFzdoBBCLP/76s6CILxv6FUYWfTuqZlArrCwXHcpiwMYJqbHcf98e/vWyyzhl2zY2PX0bV19/Hb996zt40+vO4cxnPoW4ZXjiC87izvXr+cGVl9Hp7Q2ZykSwNBmcF9l0ThWDgPdo/cI7vII1EIuirsL5xsmttTnr0NKMRIQsTojEBEd/ZjBxhDpHEls6nRYTExOoKtPTMzhf4HyCWk8cJ0RDDrLIWiqnzNkurnLENiKNYhITkcQxNo4GGqVRqDSY37TlKNo98l6Xfh4ygldVFe754Hug2ChaFDiN2XTwZVisF/7Wz0AH/wFLF2DhNxGETnhmK1GvYUA+0JpFOAhN0CVtLzfVLQqn5YsTpSgK8jxHjNBOM9LaP7cSRsJohEMScRwv+cEZY3DOEcfxQEtyzuG9X2LSG667fs04F/zieTzhuI0YX9ArlY988nI+d/nXkXgcQTnp+KP59Te8jPWTBqfKlVd/n09f/jUgDsKEfc1gWRJjTfDXVI0mUeWIOiCYg8TU52kQWrbWkiqx9QSzmHxymH2XpW2qquKbX/l3jjhyC89+xja+fftO/vStF/ODW7fz2vNfzFQasekZz6C9bh3fu+zjlDPboQhEBYMZMLqatsOquPkbppQg2Jsxam2+W4yPEgOVO3QYDMHkaBHxWGPBWqIkQdXTiiPGxzqsnVpD5SrKqqDbW6DX62NMoCcnxhCJRRQiMfTKqr6HYSuJYYd+5RziG/+cxXhFK0dV5LiqwFohSSP6vT7e11tC1Kw2aw3GCM4v5o9rtFEAXaKymMECAYYXP0NQcN4BrvYFWkLy1poBKIsLpvpOLRE2q5EaVqaO70ua8M7T6+eUpcP7iqoqVn1GI2E0wiGIRX/GsCbUvG+o3CJCVQVbfPjRNnaRimOP2cwvvfqlbDksRb1n70LFP77vUq6+/ibiZAyL8NQnHc8bf/FcpjqWUuDfv/1d3vruj6LOEtlFTQsYkCHUe8SHmBUDjCUxsY2YyNoYXxKJoap9XHEUEScJRVFQlJ7puR7dSulVnm4xZJZZtAmi1mFMRMtOcM89O9h1/xc4/adOZ1fnRN7/qW9yx907+J1ffTmb100wfszRPPHlr+O6j38If9+t4b4hiBGsHV4pM5iLhuOijKnvcWOe8YuxU0YM3h86wsirUmHwCK4WAJG1GBOTWEsqFueK4AuzljhOMdbi8yrcDysYazEKqMVIhDUx4HB4nECFx7iCqoLSKWIjUgNlWTA/P8+u6T3MdGfx+MDSiw1ia2IBQQjFcUxRFDS56cJ33Azo96o6WGQN07+HNRlYXGSILGaX8CjSWAUCYyXU2Sf1UHjfaFP7ajx1r41AkhAGEATYEHFCBZyw0C1Y6Jd0e/Oh7ioYCaMRDjmoMrDJN/FFzV8IP5KyDJF1Dc3boHgb4dWz7Wkn86qfewFTbfAacffOXbztwg9x9/ZdZGkbrOHMn9rG+T97BhOxUqF87sqrueiDn8R5S2wTxLjg8NWw+rUC68YyKHrENqLVShhvt4gBi2diLEWrHlPjKZOTk1hrSVptWq1WmOTiiNmFnLwvPLBrjq9dewN33L+HXr19QGPLD4QrjzFKlnUoi4J/+/yXeMopT+XUpz6Fq274Lv/1z97JH/3mazjh2CMY37iOU89/Ld/42MW4HTeS4MCYur2gkTWrcHzN06rNOorijan9Qy6Y7TCgtZA/hMx0zgc/nPFgXCBxGARrTIiKcY5+vx8m1MoRiSG2Eap9yrIkjmKcgTiKSWJDuzJ0Oj2qKnzHGrNcFFkSYyk95GXJ9OwMC/NzTM/MUDpHmqZLFk5lWdLv9/G+wtpAehhmXVprKWsG3sBftwLxBRZ9qLBopluuwSwxyyl4reOpavalEVP7paTW0BdjkgayZ4lFQAlbaNWZyolDeEB9jqL0+30WFhaYmVGKYuWN+GAkjEY4BNFQlocZco3prSlr6jWvSkMam3Ne+Cxe9LxTyayjUOG6G+/kXe/5INMLjjhpMdZK+PnzXsTznnkSCV0Kb/nop67go5+5AmuzYF6TEGDo61gTKzAmsCmrWDclLPQK1q3pgPbJUstxxx6BNY7uvEEkJo6VNDUkmaEo50mjFKMxHenRyjzrjozZuv4UrrvlXr76nZvZswDOy+IidYh1F8UxiHDdtddyxFFbecZTT+E//uMGfufNF/Lff/e1PP2JR5NNbeCZ57+Wb3/8Ity93ycxMiBIhJXxsm0VmtU0ixNgWFwHzc/XWaUPHVEEVgR1nsJXePVY59HKhXigWuPw3qPO46twbXEcBS1bg0aRmYixpEUnNqRZQeX6IBWtVkIUWaIoIk0SkjgiLx17p/ewa8/9zM6GJOrGRlgbptxhxmcchyVLURRYa4njOGjP3tdlYWEVgmGXxtGtJpAay8AiJaH53Swj3KgbuJtEBIwNxjZZ3F126Is3uJ86TILQZsEU/l/UyBiwD2dn54go6bfyVZ/RSBiNcMhhpViHht497EdqTE7GGMZaHc4/90y2Pflo0JLCR3zuK1dz8f/9DMZbWknGxrXjvO5V5/KE47cgvmKhzHjPhz7O5Vd8izRu085S8ryLSIUNTh/EKIlRDmtHpJqTYDjq+GMpy4Lu3AxTU+vIKwu+Ynquj436tFoZpVa0jWFiYoJ+v0+WZRS50M9LKp+TJSnbnnQMY52Ea67/Ptt3z5PbDnnlwFVLBG6gJQs77rqb7kKPp556Gv/xvR/wh391Ef/9v76BU08+gXhtm2f8/Gu59uMfodr5A6Ka9dQEuhp8I+cG5kFTJ+s0xtSmQsU5j5dgItJDSDMSAS36+LLAVY4KhyksdCu6KOgEHcZBIvq5Y36+x0K/B14xEvwupVaUWpJoQiJCJ0nJ04zERhik9guV+LLk3l33c+e92+n356kqTxTFaFXWpriMqnKDVE0A/b4jjjPSNApZL5xHgF6vR5omiEBRFDhXEcyrgvcMnv1wDJFzjqjO2uDVD/ySQN1nNLAc6HBePgJF3IjBiGKkhFoTHtQQU39HmuDbJjNFKG80qMZ/ZmqmXbfXQ6yyUI3MdCP8GEFk6Up+pVgiAXzlKDBsOWwtr/n5czh66xqUHr1K+PCll3H5l75OlrURCyefcDSvf+V5bFgb4X3OrnnHO99zKd+87gZsOsbkWMrLX34un7jkElpasXY8IYljQFHniKTEiqFXwY77djE1tYaNW49E0pj5suD+nfcxPz1NO4nYsMEyOdGhv1BQVTNU3uMwtNpt0nabhW4XExksyiknHcv6Tsb37ryfH+yYYefeORaqhqpbs98I98MkLWb37uXrV3yJZzzr2dx65w7+9K/exx/81it4zlOfQDw+ydPOeTXf/sQHcLtvJ8GFVD9a1avusJqt18R1LKxHRFnol5SVBq0IwWARf+gIoxD4W1KVJUXpKEroFyWUFd47ennOVFlhooSZmRlmZmYoihzvwoRfitCTPuKFKnaU6vHOI4TcfXlZ4B3MTO9l995d7J6Zpl8VoEF4lGVgSSJCv18SxwkQAmNDAG6K955erw8GkiQjz3OiOCKyljzPB7kXqyr4vOI45H0TEfr9/sC0N2yuE5Ulgqi5F8N/hy15Sxd0TXjtopYkQp2RIsHaxbgm5xxlWQ1MiiKCWEOcxsRJTJok2DjG7yccYCSMHmOIyJeBD6rqu0XkNcDrVfWFB3hYP1ZoVu/GmCWsuuGMA4jgxXDicYfzuvPPYV3H4Kl4YLbkwosu5sYbb2OivQbB8cxTn8orzzuLdlbhpOLuB/bw1n+8mNvvniZptWkbz5OO2UT3ruvZaBfwvXnSqk0rHSeOM/p9R1F4ClcxMT7O+OQklXd4ILaW6ftnmJ2Zp5V0yOKIqnAU/QpjoNVuU5Q9XOmoTMGevXtRVdJ2RpwY0tRw1FGHIzYhUsOG8Ra37djD/dMLVEvuSaDjio3o9XKu+uoVnP7cM7n1DuHP/r9L+IP/9IuccdpxZGs7PPklr+LaT3+AaO9dxJRU0kTvB3ZfY6SrvMcKWJuSJCnGBHOdsRHW+wGb8FCA9548z8nznF7h6NUkAHEOBOa6Pea6fdJWEAK9fo/KlXgnSFUSOaEqPS735FGEt0pRBAHhXEWlFWXeY3Z2mrmFuaCpoxhLbc4tB2QzV0FZVgP/UEO4KYo8MB6tpapCPsQkjoNQrL/jcRwPvudpmmHq4NpGQ2oguhjyIAQNGBgQIMLvRwamv2EM6hiLmKUxR1EU0W636Yy1Q7b52mTYz3P6vR69XkG/X9T1MjpjIcNFZC2qLEnTtRw/dsJIRO4gbME8fNUXqepv/KjHoqofAj70WLQlIucDvwWcAnxTVXfAZ18AACAASURBVJ/3WLR7KCKszuzgtXyl1/w9bduTedmLz6CdKJ6SO7fv4u3vvpgdD0zTbk+QWcPPnP18zjrjdGKzQIXju7ds5x8vvJg9MwVJEtGKhcdtnCDKp7njli4QsW7jZtIYkjilyCvyXg8TWY7YegQ2shRljvcVDmVh7x6sgS2bNlLlBdb7YKarSoqqIi8rcB4tPVWaoqVnYmKC0jmyOGZyvEUUJaRJRGLh1ju3k0br6fZy9ubFkhiUxskdRRFlXvCNr3yJU5/9PG6+C/7qHZeQdX6RZ558AvFh6zn5xS/nO598P9n0TgKl2GKsqWOwPM7XrLvILAYP18QGY0xgZf3In/wjh/eehYU+eV7SLSt6ZRkmaxSDxyPMdfskaUqaJhhrSeMY5xeDUrt5SO+TWIMYKL3iRCmqnKpfsLAwx/z8HJ6q9rloSLTQJPK1Ec4HirN3HjEGYwWvFXkeqNeV91CF88UIZZ4jtVkujuKB1mGMqU1twWQbcuKFBVlRFCHDRrOs0EDuCaZsaNiosOhjWgzUHjJzq2BE8VohGlJCJUnK+PgYExMd2u12IFjkBZFtvIzh+9HKUibG2mRZEvL6GUtROnL9yTPTnaOqXzjQg3iMsQf4W8IW02ce4LEcUIgYsiyr862FeJiGiWStJTaG5z73mbzwjGcQm5zSw7U33sx73nMJrkpppRmtdsorzn0hp287CeP7OODfv3UT7/vAJyjKiDhKSaKSxx+xkbUplP0+VoTMQsdURDbYwss8Z+3UBEkaIZTMz80RJwntTpv7dz1AWeaMdzohVgdfT4pdrImY7/dYM74GX1bgodft0WplgKHf6yLGkCYJXuexUcyaDWMcJYfR2dNjplCuuXl7oJLX/p5A1Q33yNqMKs+55mtf4pRnPoub7sj5i7+5lP/13y7gScdsZGzzVk484yXc9NmP0qnmgEX2k9TpfzAspgAiTMiLm1c/tG0KDhZ4YL4sWej1qZzD66I5S71SOo9Xj7ERnU5Eq5US2Qg0mNLKypMXnrLo0+t18eowkcEDRdGnuzDHQjccD7lmg6/G2ghfx715H0x71tT9onhfDejV3g8Fq9YaE4DBUjkPOGy8uJFd2PrCD1IZVVVJHMdkWQosMgQD2y/4mBoiwqIpb/EZNiSgRuB571FcSB/lPCKWKLK0soRWGjPWykiShDwK2lFROqLSEScRY62MTislSeJBX1ElFPtZwRw6Rt9HCRG5QES+KiL/R0T2isjtIvLiofK1IvJeEbm3Lv/EUNkbReQWEdkjIp8SkS1DZWeJyE0iMiMib2Po6TZ9Dn1WEfk1EblZRKZF5B+k/kWLiBWRvxaRXfXYfqOuHwGo6hdU9aPAvT/cO3XwQ0RI05QkSUIql9iiGlLWR9by8+f+DC/56VNJI4czlsu/fBXveNclOOkQpRnr1k3xxgtezbO3nYSholLDJz/3Nd7z/s9QagJRRLuVcPopT+LEYzYixtPvz1MUC7TaCUkrIWnFdPMurU6Lsuwzs3sPux/YHbwpJtjRkyxlas06YhsjCs6VdLvz3H/fLvbuncaiFHmJiSIcjjiNwQhlVVBVBa4sybs9Fma7uMIx3moxPj7G+rVtjjt8Dak1GDFkcUhmiliM2OBwVg/G0u07vv2Nqzl261Z2544//7v3c8/9szijbH78Ezj81DPpmQ4R4BHEUjsHDJGxwQyjnsIVQDDbhcwAHi+HTpwRQIXW5hJfT6YpWRIRx8Gv0coy2llCEhkiC0lkyZKYTitjrJ0y3k5I0pjceWYXenR7fXrzsyzMT9Prz+N9geIXg60jS1UTBRoyTWMCG1YrG01nICBqIV9VFVVtzlsM5K4oy5KqKgdmuDiOB+eXZRkSwLZaA6HSmPAaCnrzOcuywWdrbb33UHilaTrQpoJpLSRy9VoRRxGTnXHWTq5hw7r1bNq0iTWTa2i1WqRZyvjYGJ12mzRNQzqrOuRCh/5fCT8xwqjGacD3gfXAXwH/1AgD4ANAG3gicBjwNwAicibwl8D5wGbgTuDiumw98HHgj+o2bwWe/SBjeAmwDXhy3eaL6uNvBF5MMMM9DXjpo7rSAwgReY+I3C8i3x06tlZEPl8L4s+LyFR9XETk72th/x0RedpD6aP5cQa7t9Z7A8X8wi+8lG3bnoChpO88H/no57j0418gaa3B2pjNh03x//zyK3nicRsxOHqlctEln+WfP3sF3qaoxGzZtI5feuW5HLlujPt23MP3v3cDVdGlkwjjHQFxzM52MQLd3jxl6bBZxuS69SSdcUo1zMx0md2zwMLMAt25HmVeUeWOPC+BEDsiBKptt9utHcAleZ5TVRXj4+O0Wq0QTOuU+T0zzO+dIZOIxDtMb4YNLc9kOyJOsxCgOkTs0MDVxVpLd2GBm77zbZ50wnHcvGOWv3z7B9nbL8ljy1HPOoPW8SdTNIGQNZoMDNQTpXNKs+2riA/yasUMZgcnjDFMTUxy2Pr1rF+3jvFOh7F2m06rTWesw+T4OJOT43RaLbIkJhIhEoitkESGVmJpt4WxsYTxyTadsXFEhLzfJe/3BjFv3gVtxRpLVVYoGkxz9d/wvbUY2wiIaOAzcoMgYh1ouaoyoF03WktD0IkG2cFLyrLE10KwLEvm5+dxVUWWZgNBFF5mQESI40WhkyTJ4D41U2KShHx9xtbmRF9RljmRDZm4x9rtcA/bHVrtFu1WRrvVot0KgihNE5IkxtqIqvLkRUX+Exhn9AkRGb7q3yGQ9e9U1QsBROR9wNuBjbVAejGwTlX31udcUf99DWFLg2/X5/0BsFdEjgZ+CrhBVS+ty/4W+C8PMra3qOo0MC0iXyIIn38hCKa/U9XtdVtvAX76EV7/gcZFwNuA9w8d+33gi6r6FhH5/frz7xHu+wn16zTgHSzbcnw5RCCKLCB0u/O1zTzmFS87h5MftxWVPnu7Oe//4Me54Xt305mYwhjPsVs28/pXnsu6qQiPY89Ml3e+/1JuvPH2wGbSiqee/DjOf/mLkGKWa+/Mmd0zRxa1oTLMz81SZoJXi1dLkimVqxDJUAxz/YJ+vkBVlbTTFEpH5A3tLCaLIjppm8imtQMZqionTQ1VTXdNkoR+v09RFIyNjZGkKfNz87RsBoR4jaSVMp6mbJrMOGHLFDfcuYu9eb3dszH4qtlQb8iHZGNmd+/ivltu4ClPeAJXfe9W3vXBT/Pbv3weWWI4+fln87V772Zq4e4Bi7dJDFSVJSoWVymiYTdYMYpWi8k8DwVE1nLYYetxKnjvyLtdqrwIE7g06aXcwL9ira1pm1Xwy0lC7CNwfapWhnjYtXuGhbyPc65muIX7nsRxiMfyfmhbhyDEbWSwJsL7IOialFBJZKl88Ms5DUm9rcQYMUHLlZDLzighca0KXj3eO7yvqdQaskeURUHlysF4BBMaVCW2BiclxkQkcfABNUy8oigQEVqtVk2KMFhrwk7Ecb35IB7nSsoypyxz4jKmKIKg0aLCSKCrx2LIbIxNIuZ6FTPdiuk9s0MCd4Vn9CP6Lvyo8dLlPiMRuQAY7Oqkqt16BTAGrAX2DAmiYWwBvj103ryI7AYOr8vuHipTEbl73yaWYHhnqW7df9PP8LkP1s5BC1W9shbWwzgPeF79/n3AlwnC6Dzg/Rq88FeJyBoR2ayqO1Zr3xhDmiYhC4B4BOWcs8/k5MdvBSruvW8373jXh5mdr2iPTWKs8vgTtvKKl72I8VbYBuDuHTt5x7suZvv9e2i3xxBjOOs5p3LeC04H22fX/Tnb77mLufnpOlN/yd7pGdpjHZKWpSgLduxeYK6bUzmwUcLk5CQTE5McuXUroo7YmuDsViW1Ea0spTM+zt49uxGU7nwfV/SoqoIHHtjF1NopqrIka7WY6faI0gwjFo0s7SxGNAExpPEESStlzbq1HLVhipvv3c13751mrgDnwRChTlDv6sBDj4lT7rz7HjqTazn2qK388+e/weOPP5Lzzng6rakpTjzrXO667N10yj4QI1IEEoYJqW+8WEQqEjFYtUhsiA8xYTQ1GeKIjBGqsmB2ZpZur4cxwURVFAWuKhZNXq6k9EqapmF7DYQoijEmZ767l+mZ3VRlRVmbzEQWcyaGNFQKasJ5ItjEYq0QGyWODVHtj0MhkYhuWdDHUSoUpUOMR3AYr2SxJRJoG6FbeuZdMKuqC6Y69SF4VRoSRJ0dpCz7RMYOhKwYg1Mhy1K81+ATEiFJksGeV1EU0el06Pf7gCdNF82Izjl27rqfsU6b8ckJqrKimzsWejkLvYJCHWkaY0xEbBMSm5GnFbnbw0Jvvo6TWuUZ/ai+DAc57gbWisiaWmsZxr3AUc0HEekA6wg7ku5gaLvsWsM6gkeGHYTtths80nYOVmwcEjA7CYxHCEJ9WPBur48tEUYi8qvArwJMTk4O4jOq0nH0EZt56lNOQii5f88e/v6t7yEvU1qtNr4oeebTT+ZnX/hckjhkTLjxljt5+z99iNl5R5yN0W61+flzX8AzTzkGQ0k3d3zzW9dy+223g1fSKGJuvodGMUQpD+yZY3q+FyY2mzDWaTHZaXPYxsPYsH4DPu+RJBGJFdaMjxEbiwHaWYaqJ0bYufM+xMT0C89ct4/zjl5ekaYpjojJNWuIbEqSRlgRWq2UJA7tVJUjSxPaWcp4krJxwySdsXu5d7rPfXvmmek6uj5oSg1LCkCM5cYbvsuzTt/G3BHH8NYPf5qTH38kx2+aYuPjTmLv7c+hd/2XickpJGwp0Nh0QhxXnZXaQvAfHTp8OmOEsVYGJtxPOhlRYolmY6piMbGuejNgaBalZ2Y2p902rJmKsbW/o9/vs3v3Tsqij3MyxFAzdZqqENwKAupoRWHDRCMhiW7LCllkQzomE7K1L3jBUpF5hbwii2OMUYwoiTFMtjISEcTE7JqdxcbKfO7qbCCC05CNo+E4tq0NueHUYKwJqYhiSxRHVL2q1tgqitLVu9ZCmqYDoaMKrlK8QhynFEV/4NNa6Pe5e+cOWlmLscM7tMYysvmEhTnB9xxFXlAZQ+lLIhsxlqRsXj+F8Tm9Xo+Z2ftWfEYjYQSo6g4R+RzwdhF5EzAPnK6qVwIfAT4iIh8Gvgf8BfANVb1DROaBt4nIzwGfAt4EbHqEw/go8JsichmwQNAaBpCwX0FMeGZGRDLAqWr5CPs7YKg1yIc1k6nqu4B3ARxxxBHaarWYmZkhSxJOecpJJBGot3zsY/9CUSbYOMEAp207hXPOPp00Chmxv3bNf/D+D38CVxk6rQ4Tky1ed/45PP6YTXgc3crwuS9+nauuvJZEU8ZaCUrO3HxFkmY8MD1Ht18Rpx067YzIKJNjY6ybWsPk5Dg+X6ATR4y3MzZuWEdiw6q40+oQGWF+fp4yb7PQmaRXtVjod8GkgGNuPgdJSCtIszaCJYlbpIkljgRrhVYUE7ctlTr2TM9QZhWHx4bJyQluuuMeriuDptbrl4OVcsOpCYwu+Na3ruenzv4Zrr5uJ29732X8xX++gCw2nPDsn+aqO24jfeAHiHjUh8wLjfM9ioKjPzIhpxl+9VXuwQdBIhvMkMYSq2M8iZB2i3lynA8+sbj2QzoRMDHdYgFv+4yXMZERev0u99+3k+5CgXNaJzlVqH1u6hXxSmYCCWSiHdFKYhKJSeKwQGkZJY0FjRK8g/mygtzRjmL6TrGxkBiDNYaxOGPtWEra9pSVsHOmy0SnRb8s8T5sHV86ULUgnr73TKYRnShCHHgMfSoiE76HcSREJqJyjqryOBe2sOhVFe1Oh7TTwTnH3NwsUWSROoi21WqR53kw4XlHXlXsnZtjzdwcY2smw/cySdB+ST/PyYyQxyAJpGJYM9ZC3Rrm5ix3rvKEflyF0adFllB9Pg988kHOeS2BtHATkABfAq5U1S+IyB8DHwOmgK8BrwRQ1V0i8grg74H3EkgQ//4Ix3whcCLwHWC2bvN5LMZLvbbuo0GPYO664BH296PGfY35TUQ2A/fXx+9hqRa4tT62KoQQDCj1qu3oo47AGM8De2a5+da7yNpTgOfYIzfzkrOfSxKXOFG+cOXXueTjl2MkI41jDt9yGBe8+hy2rG/hqNjbVT78oUv53g03E7uY8SRioh0z3ytJ05jp6Rm6hZKkGRvWrSEWpRULrcQQm7A5XsvA+k6bzRsPo9NKiST4ENBgKtIspd+u6K0Zg36PEkemDueULEnp5zlGIqqiotQ+zhpMNobznnarhZHAfopNwtSkkFeee3fOkI2Pc8Ixm5nvVsz0H2D3TH9xUwAJK+SGQtwrHd+/5iqeeOITuOr67/OvV1zHS1+wjWhsnKNOfz43X3YPrapPpQ7vozqmyIYUSHHtVHdar/4PDWidqqasSirxeF+BLxHCxnsDwoYqNrIkUU3vzzIqV2GNUlYFu/bsYtfuXSEtUu0P8nXi3jhKcOrIEsO6SFiThcwdoiFmzAJZbOmkMSYSZosCTIvu7Byt2GOiCLqOdifD4EiMYW07YeP6Nk4dM7MVx01N0i9KpquYoixpxwl56VAMpQMrjg3tFuOJkNiYuX7OXOkoK2jHEc6VRN5Rlq7eqTbsteRcCDmgZqqqak0LD8K5Sf/T6/UwUUQ/z9mxYwcxhsNcxcJ8lzwvqMqc0jnmUbBhHy1X1Zp1HaqwGn7shJGqHr2f4ouW1ZWh93uA16/S5juBd65S9i8EIbJS2UXDfQ73V3++YOh9Bfx2/aKmnd9b+1L2aesQxKcI9/ct9d9PDh3/DRG5mEBcmNmfvwhCfEUcxYGSG1nWToatjG+9486gERlhTafDeS85izQGiLj+hpv45Ge+SBqPY8Ww9cjNvP7VL2XTpKH0jvv2LHDh+z7GHTffzVgSY3QeEc/MXA+MYW6+z8x8n1INWSsjjSxa5uCExKSMtzIidWyYHGPzYeuZ7LQxhCwFBkAgy1IqVzE2psz2eyRlj8hAlrYoipKiKMn7OWVR0V1YoD0xji9yRDuoeiJjQ/oeY5A6a/nExASzvR533buTOEs5autmdvcM23fN0K+GMzebQUCjEc9dd2/n8GNPYNOWzbz3Y5fxjKecyBHrWxx+0sls/86JuNu+C7iQENbXwZsoahT1wQ8lh47LiKqqmJubQ6zBe8GUPUQ9/cLRLz2Vh8pVqDqs9yQKaSK1iSrCFX1m+n3uve8+ekVex3cFs1UWCYmVkKFCYU0r5tipDuNxCqq0swRrPGNJTCetY3OqEmMce+YWsAKZCJVWrO8ktKKYVishwbFlcozx8ZjZuT5xK0alYqaq0CymaGfEUcx8N8dGCQtFgSJsbAubJ8cRDLu7kPaVfs+RRobKC6UI3ghOPU6bwNfwTBcWFgKzz4U0SQ0VPY51sIVF3u8TxREtG9Pr9Zienmahl9cpi0ryoqCsXMj4Xgs6EcW5xYD0lfBjJ4wOVYhIC3g+cDnBn/InwD8f0EE9QojIRwha3XoR2U64lrcAHxWRXybQ48+vq38W+BngFgKh4w0P1n4TE5EkCZNjLdqtFBXPHdvvJU7bRNbyhBOPY+O6cYw6ZroVn77sS6TpJFVRcfwxh/Pa17yMqbZFteLu++d4xz9+gF17F4jSDkb7FPN72LNQoV7pl8r07AKlA6cOIyZExlcVSRIormVvnsnxNmvG23SylCQyGMAYS6vdae5L2DoiCVHpsTqoKuKoRa9XDtLD5P0+3VkljyIi5yj7OWmaoM4NqNYQqL8z812itE3WGWeh2ydrGTZMJbRTQ7/yg+h6GT7PedSkXPutq3nW85/HVdft5L2f/Dy/84ZzSaKYE57101xz5y1krsB7xdWO7YbJ7Wtmlqm34T4UUFUVO+67jzRrYTFo2QOtyCvIS6VyHq+CUCJeMc7jkwjFIC4wGfdM72Vubi6EEqiAUTqpZW0WIUYo8oJOHHP4xBhHrWkTx8JcUTGRJLQSYbzdJrEh27yb7pNYS4UHV5JGKS1xrMkSWvU+V4k4Nq2bxKkiE0LUy+mXypo1YxTzJRNpxkQ7ZTKNyb3BRtCKYetUxsbxDpV3GBM0rH4KTsNWGkUZkr+CZ14BDHEcfF3OOfJ+kw7IIBI2UyzLcrBNRrM77ZyHdrtN3M8oqgrnPd4pVeHolgWuUqqiImulpGmIN0rSdNVn9EMRRiJyNvB3gAXerapv+WH082MGAd4MXEIwwV0G/PcDOqJHCFV91SpF+1DVa83vTQ+n/SgOWxdba5mcGMNaQ+4cD+yaJslaxMZw/HFHYero8W9f911m53OMjdi0cYzXv/I8pjqBGn7L9lne9u4PMztXkiQp3pV0Z2ep5ueYLwpKdfQrS69XkmaBEWeNsDA/TzuOSGxMd6HLWCtm3cQYE2PBN2QkbFsgJlCJoyjG+bATbaIw0WpRtlvs2jMb4lKswVeBsu6qCnX11gU2bC3QTuNAlbOevN8jToIJSbo5c/OzpFmbufkSryVjLcOGqXHm8llKtzT9ymJyTEPe63PbTd/lhONO4HNXXsMLnvMUtj3uCKaOPp7xYx5HfvM1iLg6IWYwZxkDZVVhvC7Slg8BVM6x8777ieOUxKQYCUzMsgr7DhVFFXx0icGlMVVRkiQJRgzOe+a7C9y7YwfOVYgqJhLGspiNrZjJNGG2KIjjiC1jGcduGGNtFpF7T8c6No536MQRURIRJT5k27AxplQ8FRPtDFeWrB0fo2WUViRk7YTIGIz4kCPQBibfRNyiXxR0ciHuGNaNpThN2bNQ0IpaTGQxR6wdo5NE9Mucfr8gHU/ZPd9FrSWqM4zPdkOGjRJPUXmMFRKbkOd9VF3QfIfCUJv4pTiOodaWCinZOzuLqzNNFIWjLHzIvZeXzBSz5HlBu93m/2fvzYMlu+77vs/Z7tbbW+bNhlkwBECsJMWdFCVK1m7LkRxXokqixOVy4lQWKaUq56/kj8T/JKqUkyrF5SQlR1S8xFIsWVLoiBJFmYwEUhJJcRNIgiAIzALM/vbe7r1nyx/ndvd7AAaiLZAGUjqoN2/Q87pf913Ob/suo9GQtTVDWZb3PEeveTDqBu1/D/hBEjLqs0KID8cYv/pa/67/P60Y44xEhv3z9acsZx1Nk8QjB6MhUQjaxrOzc0AMkrwybG6tJSiqUDx/+UWKvI8Lju9+/3vZGJb42HJ7f8ov/MNfpmk8eWY6tnkioc7HE4yEIBXjeYsSEo1ns19STycEIxnmGbiAyQznT59la+sMZZ6Td0rLSalAIJVEG4VwSTRTBs96r6Sp+xi9zbytyY1GBoP3jrLI04aXGSgLXEjoKq1SMHDWYW1STs7zjDIv8Y0nywoOxzu09YyecuTK03QTx0V/eCkqKwMxCK5dvsZ3X3qcnIJf/vDHefxn/n0qI3jofd/FF658hdzPEUEByWpc687/x/sjr/r6X8EHJuMZUGNUSVlqlBI4H5g3c2bTBuc8WZ4x7FeEskg+hkLivGN7d5u6mSJk0rMrc8nJQnO6zNBSYVtBkWvuG+SsZxGjwVnBfYMeo54mFyK1ujLDtA4UvYpKCE5Hwbxt0VowKjSFEWyNekSRINhlrihiZDKLDAuDNoLtQ08uAqc3+4wqDUIyKCxNbckzzdagojCSvTH0MoclEkVJ3bQMioy+lmwrhZ8K6tASfeoAyI4E27bJGvyoxYVSmul0hnMOYwzWJqLteDLGO48xGd6vRGAX6t+zWXpOZjSDykD+7W3TvQf4RozxeYBuFvDjwJ8Hoz9fr8kKwdO2LSEE1tdT0Dk4nOEdjEZDjIE8L4hS4wM0ttPXCp777z8PIhLJePKTv09tXcdZSux4EQNt3TKdtWitaKKj9ZFTG0OGhcJ7i/dQlAWVyagyxcX7TnHu7OkEWJArFQQp5VItGVga2kkpkSEw6JX0q4Ld8TZKG3Qnz59pSWkMwTraeU1v0CPXBhUhtA4lBM5apDKdU+wUqZJC9GzWMB5PyFQgiy0KTSCh4pZVEUd69zHyzFOf5+Klh/n8nzzPl56+zPvf8iAbFx6kf/Fxmme/gCEQ5ML6WmFMQmO9kVaMadaVZYY812SFxhiF9yR1AQ/jwynT6Rwlk8trmqV4xpMJu3t7S1keoyRruWY90/QzRWM9oyJnozRsVAVGaYwWZEKw1jPojMQRygQoTS08SkWcc6z3c+S45dypk+RGIoRnY1gynTcUgyGZlLjGkZuKrIDWJ16PGJSc21qnX0isD0g5ozaCqizpVTmFFjgraBpL7VuENOQS1ssMHw1aGFplmPmDZIERJd56EBGtk0JIh+hASrVsMS+Ckda6a9fFbr4Uu2B01OhvYUnhkVKgOkTfvda3Ihi9Em/kZYz6o7yRzGTvPH3qVIf1T1wQqRKqf6lr1B2Y7tnLA3R8LQgVK4FHKcVyQJfuQt95jKQfnM2m7O7uUuWCqsowy98X0nsRGqU1UmqEVCQFpfRiq3eT/ljG/Ch4lWPe8TYkMR4VK1w9Z6VjtfpMMZnLdL9PLH/p4vmL93A0W018uld4I93FooRDCsdCb1cIeOqrl7djjFuv8u7/tS8hxHJjOHXqJEppdrb3KYo+Sinadp7OueyMxxBoo6hbh8kUiEjTRL72zGW0KQg+QudQ6eZT5tMZQmY4D23TsLExoqcFgoBtPZsbW/SN5vTmOudPjjhzcoPcaIJrCUHgpe5kV9K1kuwBjmSbnU16VeQMe1UapM+mqfrRGgUYqVKraDCkzPKEHFQaG2yCb4VI8J6yLMmyDOUiUgmsTSivE+s9pk1kentC7RZ6CnT3QHccEQRhuHPnDhfvv4RRJf/Xb/4+73r0frTOuPTOD/CVb3wVgyeIhBiMC+VutVLzfkMsIShyQ6/ME2eryDA6+evMVUdMlYp57dBCdZYeCh894/GYup7jfSRGSd8oRllGL8sxUhEErBU5wxyKTGGkxgSB1kn1PBeKMsvJMoELiiKTRC3J5gVSCU6Phpw5uZYUD6QizzRlz5FXJUjJfDLFtS29omR/LAXbrwAAIABJREFUPufkqEKv9VMyVFXsTscMBxXrPYUMlp6W6Zr2DYWK5CpDR8G+benlBhECsdRMbWQ/z5lZj28FwTuCSJwroyRCpccEqTUrCCgJtm1YoldEmsdlmSZGTwgrNfB0oaUKWhuBKTS9fnHPU/SvDcBwlDdy8cLF+Lf/679NEBFXz5lNp5S9Ib3hkGowSPITriHEltz0iVGS52UXHNISAqSMIBZRWWCMod8vyXKDlAEpIdg9pNtJgQ/BF77wx/yTX/4l3v1AyTvedoGtU2sQoKlrDsdzlF5jsHmKariFztcRsoSOkZ6QoB5BCiyBVT8emRjZsvtv9T4FCEVrC3zIEMqgVN65aioinqZpkl6ZSxtL7KC1QuilDpvQiRmvVJak6I8ILS6+pJTEztP+KIolhICMloHZpad3k/ZVtEgZefjdf+NeNIDXzfKdPXQMnvvOnELKnOs3bjIa5Mxrh4oR12mAaSHZPLHO3e09MplRzxrC1ojGzWjsHBcSkXRxzKKtcfWELJeMpzWjUcmgytBGYpD0hjmb6z0ubG3y5vNnGJhAJjzRWWyMRKUgzsBbqrLClCXIJKVishwbAsppdK7IRMbZU6eZTWvu7OwglcZoQ1WWGJN0xEqj0AiQEkeyBYhdcqWEwihNv9/n7qQmmJxqNAIp2N2+zeaopNyf08qIa1fw5pQ5dQkISX7mq888w8WHHuSLT1/hy89e552PnGfr/kuweRa/cwWBQGkBKKQMOJlsF94oS0lFr9enV+ZUVUleJFRbjAFt5BIUU/YiiogSsWv/OibTQ1qbwByZEowqQy8zlFphAJNJ+rmkV2i0lmglMEZ1383SSlx3ituZFoz3JwzzDCEDa2WfzdGAvMgRHUUAkZTGoxTovqCZz+lVPVSRo6XCW0vshFKHomIzz2mnDSpadHRM2xYRPb0ip3GO3EaKhLbGGMUAwbqzbNYZ43mNFR4vVzB1qUTn/LqCf6/UvBcVUwIzmCLNcGP33Liswo9aVSTtyKL49gajf3neiBDIIsN7h9QJbdQf9DEdJl1rTdPOqNtDqmIDrbLUdjny/LTikaph8Uj6c8lC7yL2kpnePR5CwHVwRokmhgSHTTr6migUQchuqCfxAWZ1w43rN/nGs1e5dvUmjbWc2Fpna2uTot/jzNkzXDh9giqXHRRWdrLxnqO+IWm2cPzkdW+2I8CLrsJZBRpgKRR6rzIsVQcpGC1mBYsLitjpYHmfvFVCUmV+I6wYIlJGPvjB97O2vo7znucuX8EHQfAeJehY8Aqk5IE3PcCzz16lKApu3b7DA5dOMxqNOH/+PF/48rOYTqxSSkmIybmzMIZBv2TQK5Ei0O/1UD5yanONUyfWePTNb6JnJGE2Tu6WoqaqqiS34j3OJg4LSpFXvVTN2RatNMpoIOIDnNraJM9zzh2Omc2mxBgpigKl0+toEcm1IniLFKnVgZSgZCKm+pYiM6yvjZg7R1lV6K4lwt1Dzg4H1Hf28CFAl5gIFu6fXfYqBHv7e5xuanTW459//A95y5tPkxWGc299Jy984hYqtsvrRgiBFDJZWr8GSwhxnqRjeCq9IX4+xvhzQogNEqDnfuAK8BMxxr1O6eTnSCjMGfDXF9qR91paa0ajEZmWVFVJfzBg0CW66qATJSWSRYWWgugtIQb2D3eZzyc4Z0EIjIrkMpJLgSFiBFS5RkuP6WR3IgEhk/TPYtMm/Q2hIsSAc55BpjBZxiDPqApDXubYtk3iuEiElkQpiXlBpjVlUbJWJH5dPZsRAGkUaxFsa2l8ILSWYB1GCYa9EhcFbjzBENmoSoQISKEoVGSznzOeOw7Lhlk7wUrwHR9NKUWIKwSocwkFaK1d7LJLkz5rLUSREvKQKqHlPrPs7CzQl9/eNt1ngYeEEJdIQejfAf69V3vCohuVKcOkOaSpG5rZjP29Xfob6+RZQQyWdlbT5mN0bx14pTbdolLyxyqEZPwpu0ifYIlCJDl5H0ErgY4Qncc7j9DJbiuGAFIglUIID9HTtA3PfO1pPveZr/D005e5+uJdvI9J7FKplHTGQNmvkFLzlice4kf/0gd45OFLaONAKIiqMzGLnW10h1wRMWWvAYRQydiqC1LEtJFIKVn1HGHZooNjgeqlS8qU9qRXS9+D9zhviS4SQrusKl/va2Njjf/kP/0PObG1DhGuXb3B3sGY/mANqRTBWead+COd7H5iykOvV6VzFCJN2xCJtG3bmZMJhDbpmpAiMdBj2mR6ZY4mUOaSN509zaDKUQRiViC8Qx1tJ8e4VOBWeYEJyVaAEJARPBGpFcSIUIKTJzbYGA2wbbOUognR07YNWmukSMEr3coRH7v2sdS0tqUqCqrGkWcZ1lrm8xlrm+uM+j3WhkPm7Zy745aZDTRdazh2M5+FVrcQklvXX+RNDz7CHz31HJfvHvDQ6U3OPvJ2nvnkk0Q5I3nohKSzphXCv2bJiwP+Vozx80KIAfA5IcTHSITu10RcV8qkUqGkpFdmbKwNGY1GtLbFOUsz81jpcUqSGwNOYW3DZDLG+khEImMkV4pMaCoh0DIiJWRSIVREyJQ4IBSEkCra7n501hGEAy8INhFQda6pck0msgROISY7lBA6FfZ0p9bWJ4uLfg4CvLVJKy9CpjuNRimRZc40WnCeopAon+DWrdUIadGJsYyIqVXd05FRqVmrCnZqxyQk0VeJAh9Q3T7hXTgmCRViXHVsYgpNgeQCLKRcVt0pKKmkDoEkhlR53Wu95sEoxuiEED8FfJQUHT4UY/zKqz1HdDOgPM8JvQHOedrWAqkSae2U1h7StGOm832Kai2xwY++xks26MVjq805tSYS87qL3iLJmigRyYxMF4RIJlQu+KSISyAISYyG6y/c4SO/8Qme+epldg5bZk2ktRalklChj4kHUpYlRM9savncHz/N1SvXeMdbH+YHf+j9nL9wCkQk+EDoWnxCJHl50b3nROlYzJC6DFYcl/hHrMhqRyujlwWjbn62LA1h+freB2xoiS4Qgn3DBKOiyDl16gQxwu7+If/HP/zHvHjjOidPODKdTL1OnDhBlJ6723s8+eSniEgef8uDvO2tjyKDxbWWyXiKVoamndM0aeM3RU6vVyGxKcsNgsxoVAz0ipzTWyfo90qUBBUAJSHKVGsEj1RJ/ysSE9vfWmzbdMiklEiILmCFmBIdqRQieKTRXbs1iSyXWXLS9IGlinbwnjR1FwSZDN5MJhmNBhy0lq99/Vl2d3YQyrPZK+kVkQcvnMK9sMf87n4KbF3WmlopAiEVMkbG+2NKpZi18PFPf5lLP/Y99NZHDC4+SLxxAy3SvCHgie61a9J1JOeb3d/HQoinSbPn10xcVwiS/bXWaa6jFVpJEBkqz1F5Ruw0AgMRoxTzucOHZB+eDAcDRWbIpUJ3bXFtFEprtIRMpZrTCIGWkugjOI80iYbgvSV4CDaSSSjzjDxXGJWRujoBLSUurronQiSwi5SCZjZNnY6Q5uoxBIJtIUZyo5m7FqREqaRRl6mMtm0pixwfYhJmjSK1/wXoKBhWhk1Xsj23HLYtrYtp9hMTHUFJsZQ9WvgnZcpQ182y+nHegU2cOhnF8nin87kAMaSAvJidvtL6lsyMYowfIZEZv6mltKLqV0gpGWxsYMoS7x3GZAgZmc4PmE6nbGyeo98/gVBm0T0HVkHnFauCuIrgMcRUkXTsaUQgeouWgjxTnS1BatWEzlRKeA9BcfPqPr/6S79FPXa0E09wAh88WqslWdE537XhAoGA0hneRW7divze4VM88/XL/Ac/+aM89tZH0s/E5NIJC7TJy9uJix5s+rQr8y1xxHcEVj4kqx7tIis7fkyWsySvCEHgnCW6RLiMb5A2nXWObzx3hT/5ky/zmx/5KFeuXqMsS7QynNjY4gPf9UE2NzeJ0fHUl7+GtYH5fM729g43b93mvpNbZFrwwe98H7/+kX/BdJygqNZajI4MhwPE7BCVZwm2qpPO3dpwSJnnaK266pou8CycVkOqanVKrBL8tU2ZbgeRjWKVDEhtlgKdAoFWutOA00iTWodCJKZ8JL2/7irAI4gB6Jxei7zHcDBiOFzn+eefYzSo2LNw484BN+4csLt3uGyvCCFgWYEloMzCvnz7zl1GgyGf+uxT/Fs/+F2cyCynH3yIL33y91BSImJK4IJzSb7mNV5CiPuBtwOf5jUU111YZC+gygsXVNeBQbTWGKNpap8G8kp3PKMlagghIkZCqZLpYOL+pPrBoNCpkYKOHhElUWiEDZhe8gQS0eC8x8g0l8oE6M6SQwRHO0/EUmUyPAkYoKRKAqkxIFx6XQS4GLFNi4whIdq0SpVZrkjux6ndHEIgNxkhB1s3KcuhS4Ql9DPBWqlYrzSHc4W3giYEvEjJU0gIhcXxXFqXL5xrobNl71p2cbEzx5WbbIzJiqSuaw4P753CvG7gMCGGpWVvVhbkVYGLDfP5DgLYWL+P9fX7MfkoyaEIXjEALfbgxeMhpCokBaLUmgrep6yiYxwrEVBGdlLssZNjF4SYSsvt67v83keeRNuMGAQbG+ucPrHOWq9AC0mWFYSY5EScT4ZaxOQB09qGgKdF8Ny1u3zoH/3fPP3M5a7VAgsk3TL3jYsceDGzSrpQi/nRqrGyQvW99Dgc/TmxfP1VcIpAIA35W9vS2gbrknz+G2Fdu/YCP/1f/C3+zv/4d3nmmeeScVfjCB7y3PDWtz3SHRPFpUsX+Y63P8He3l2++MWv8N/97M9xc2cfZOTtTzzAxXOnljfMwrwsz/O0OYfQBYSUVWdSk2uDjOBai23bFGw6UMnia9FnXzhwHp3XLfgX1rbH2quLuV5KFhQh0GWkdOd+dY47CbXUnw+R6DwSQVFUnDp1FqUyGhsY145bB3NuHrTMvST91CohkS9JaARw48WrrA8GvHjjgG9cu4PDcfLC/VCMXvYZXuslhOiTNCB/JsZ4ePTf4nE47Te1Yow/H2N8V4zxXXmWdwoBPu0FPkGrvbVgPVkUlEp36hoO71rGkwMa23SeQT610WSqepQUGJ3kmSQCJVNnRQmBUnIZrBSghaDMMjJjkplfVVEWGUYnp16lFU3TLBNmZ23imQkBwSNFxLct0aW9S5DeQ5lnaCGQRIK1tE2bZnrdNSsFFHkGMaK62z+IuLzWog9oAaXR9DLNelWwVhZJLFYk36Sw6PhAl4CtTP6SKeBiNtTtU6zm0ouqSAhB27ZMZzMmk8k9z9frIhjFGPHO0tRT6vmM4B3et7TNmKaekWUDhqOtNLi997z+yFr9QOzkLrqEIN28wRFcyoBaH2hdy/bEcjj2BAeRgAuONkYO9wV//IffQISMrFeS9/v0B2usr22yOdrg9KnTlNWIqAzSGEyWSmLnPU1rsc5R13N27tzBOc/2bsOHfvHXuHN3mxD98oSley0QiAnGjT4y8DvuqSmERiTx7rRJcQQNdgTRl/jHR46JkMQODBFixPqmC0SWECzRvzEEwK11TGc1SpnUYhIaKRR13XLr9i3+/v/+93nyk38AOO6/eJYf+eEPcunSOfb2xriY8ZGPfZwQI/1C82N/6ft59JEHKPKCLEscES0l2hha65aQ3yLLl2gg27S0dZ2Gxk3S4SJGjDbkWbHc5Is8T5YQXYa6kuVRSKmW1Q4kftHC1C2ETv8tpERWSoVSyQ5aSZXahlqiOskhQkCKSJ7lFEXBcJhmISI3jK2nQZC20kVlfXRXPxrkInU9Q+NRWcUXvvx1ghD01ob0t850zwuoroqLr6ECgxDCkALR/xlj/LXu4dsiieoi/oziujFG6rZhPp0znzXUtaVpLN55XOtomwbbNgTnCC6JhO4d7FE39codFYEWKiHeBF0lHBINJAaCSPeYzDKQoLVA5ybRS2KaL5lMkxcFVa9PXhboLKcoSqr+kCyvyIoqXYtaJ7kg2y7tG/yiwu6UwUUHMFjQQDKdYVSG1Gk2FL1Ps8Hu/UmdeG9Sa5TSEAXWB4J3DIqcQZ4xKAz9Iseo9BljlzQvdOpiWLVn0/Wc5tuhU+SQYoWmWxQFIYSEEJ7OmUzm9zxHr4tgtFC8dR0foigKqmrI2ug0w9EpimqAUPpYhvjSL3jp3GixVmS/EAM+WLxv8cGlasx5DqeRZ69NuH53RusFwUu8i0ym8KWnDtjZ07y4YzmYCw5bw24rOWwhH25SR4PMKvqDIYP+ECFUAjR43yFLVifNOc/h4Zhnv/4iv/qr/0+XDYVVG3HxtcIJvgyynYiH6fvys4vjZfHxQ/CSWROrwJWIaqErsxeGYK//JYRESZM2YykQEnxwTOdTJrOGy1e32TuoCR5ECBgBf+XH/zKIwHRW8/nPP8Vzz18hxsCFkyPe8fgDDIuSB++/yI//xR/gfe95J73hACEVRmt6RUEmFUVV4FxL28xp53Paec28bhnP5p1LpkJJQ6kNuZQYkTJjYsTbRBC0NiS+Cqny8j4lK9Y5XIg0zqWERCS0o/OOpmlpmwacI5OSTCdLCSE9PjicneNjg9aRja0NTp06yaDfZ1oHbt0ZM5vVRFILOR6dMQqO8YUSqhNm4z02N9Z46mvP0XqBUp71C/fhIzjhkYGk5P0aEV87dNwvAE/HGP+nI/+0ENeFl4vr/jWR1vv4psR1A+PJhL29Pbbv7rC3d8BkPGMymTIejzk4OODw8JD55BDbNsxbS2tX7akVglUcO14hBHx0BBmxIRBRS+WCLMvIyhyVadCdT1L3vEXikWU5qgsOiwo9LDdzgZIqmeOFlASE1hKdJ1pP9EmiZ0EsXeyPsWubJQdah1sEIyHIVEpolOzg5lJhEPS0oG8kpdYMyyKBL45UNovPnFDHx5OrBQo4hnjsWB1NtNu2ZT6rmU3re56j14lQamqpaJWh1AL91mWCJgepUu+yG6AtB/ewSvHuUS4lEu0iWAVCdPhgkSRUU/COupXsTmBqPTZaRBC4puL6Nc+80UxCwwE5swOLNjlBRKTweA/lYJ08BPJGMZ2OuzLb4bpAtJglQexaOhrvPH/4qc/wjre/m/e89zuXGcHL8swF+IAFjOF4y+1oMFo+ZfGc5X7T9XGPDKxXePZuIB4cwTtkfGOw6o1WDPt9bBtofUIBSimx1jKZTCiLjCef/CTvePsjnDm5iYiOh+4/xxOPPsQff/7LZNk6N27e5cHzp0F4vvt9b+GtT7yZeTPj6gvXCUUOpqDIktmY1gajNZkU4CzO+iRC2et17bgMshwbIdg09M6MoWlbotQ0rSPLAlkuaF1C+aXNQBEjTMZTirJEG03TNLSd2na65ANKhkSgjJFMa0RUGGVomzopIiOwszkhV2gF9104z3Tcx93cJ0aZtO6ObRKr1spiSJmuJUkMiju3bnPpiVNcu3mV3fGMYlhw6vw5nv2SIViPCwHv3JKQ/hqsD5AsUp4SQnyxe+y/4jUU1/UhsLOzi7AOIzTT2ZzRaIBSkslkwmQyprUtc2uR0iXdNpfmd8vqVaa2Gl2iIEWSSYoyUTNilESR7Mq1WViXd3sXApNn3aYtlvNGlRmkUrimXbbA2qbB6ORIq5XCzueE4MmzvCNyJ6i6VBLrLJAsyG0naOtDwHdBw3uf2pNEVBRdo3Z1zmOIGCnRvqWfKSZG00ZPlRnmPhGcO9z7anWzxsU8m25GtDguiyXEKiC11jKZRLLsde70GrynmdfEEHA+4m2HGolJxH6hDqyMXpKmQteeCovpSne8hEjzlkhHVA+JDNv5VhKD6PrdLmWsIXnMm+65RIHwcDiRXLljl1nB1voaj77zEg+c32J7f5/rt+6gYmRjbROhBQ5HO5sxns7Z3j3gzs4dXrx5izt7Y1qX3BqlhsbViTuA5MlPfYq3v+vdy6xDCEkICmL3njufe4Q8gnRLKMBlRUQKLys2vEjtTMTSBXIVjElHLKbfTwCcw2MhtHj+7MFIfBs4I6dObfFv/viP8JWvPMuN23eYzKbdTZfmAPt7eygR+N2P/x4/8Vf/DUoVyVXkzNYGRisOxnP+wS9/mMtXnme03uPFF2/w9W9c4fKVmxzsz5ARtvqaE1VB9GE5A/JNjTYlRkToeEl5nhN8ABeZSYcoDVWvx9wH0AXTw0P2dnYgBqqiQApBVVXkWU6ZJd+Y2aym9QGdGbx1tG6SKqMQCDFS9CoyoyE4lEzozTIv8Aisd+DB25Y2jIlBUfV7eN9QlR0iVSq+uXFLmkXOp1Mqo5nbwK3dfc6MTrN15gzPZTky1mm2C8uOw591xRg/yctbGov1mojrBh/Z2x0jlUNHgZpIJrMJWZZhraNuZnjvmFvwftJZdBzlAgZklCSQfQQkxKQvEIXChBSo6hiR3pOpnKAUQhQIkVCapsxQQqXKUoDqZV27XeGbligF0UWkT7u4Cy1ay05ORxIJyJgAF8ElEEkgoIxKrf0Yk8uqc3gXcCHiYoDgMVFgMgMhIjutRdvNz0wU+HmLUYJ+Jplb6JmMxjZoFNYHXLTpk8dVQnP0O6TqzONQ8nhYWQS9uW1ow71HAa+PYBQ88/kMWJV1UnaSFjH10hHgiiwhXBYb7ytUQ/Eef+/Gvt3vC90QzxOcRcVAoSS5Sr/fBcP16xPqiWe4NsTGyGOXzvL973qY+8+uszsZ07g3E32DJyKkIUbJ/GCPpq4ZTyZcvbHG0wPDne09Wh/Z2dtnsDYkCHj+5g5BGq4+d5Xd23c4c+7VHMa7QNIFoaM9/pf95KJFJ448twvaq4vmSGUUF8cidAZqr8lQ+lvOGSmLjH/7r/4QP/rDf4GvP3eVz3z2i3z56WfZP5xQ11Nm0zlFnvOZP/o8f+GD38WZUyMikfe9911M5paDwxlKCm7cmvL8Cze4cuUGL17fJaIwZY8sBqpcQ5xBsLTB4vwEYQuEVVDkeBFpvSNax/7+GKRkGqBG4rzHeo8yOTEIDnbuouyMYZUx7A/YGAzZGA2xw366lnPNwWzG/O6YpmnYn7dMZjP2xxPmdQt4tJAokgOpygzrox6jQY+trROMNtYRqsVHT+0E0+k+OlPcuLNN4wJeBNRLOvLxWAa7SlYiAmsT36XfG7F9dwfxptOsrfWpBjmeGToIYocCfKOsEJNoJ8KjRYI4T+bzJPMlJDEmwnvroW1t1y6vadsEUEkJ4mq21t01y7tzGgOTtqaUkSA0eZYAWc4kkqiOntg4isIgsyQxZvJiOU9RRoNOaL+ma5mbPLXwsixf6sIFIDiPiJFAQCiBiL6b2YRl+8w6hz+Cglu8dx8jQnUOvp0uHyHS71VYOyfTEiMjmZJIGQnWdZ2qrssU5fI1V626xVHu+jdikfAf3atSAHfu3gnM6yIYya6/uqhCQodiSkNetRScjF1/++gA7ZtZCxzOUWu7BbEwBochMsg1hU4XV+sVd3ZqMmk4PLjLuYv389hD9zMaVDgvKIsRsfUQMrywSZrGeYLRRAtrlaG4eIqNoaauaybTGS/c3Ga0cYJ564g2cGV3TuPh2gsvcPb8hdX75GjQWMXb1GrrJkDildQnFs96SctOHM9kjj4npBq669YcnVX9q69vB2cEQHhHL4+8/bFLvPXRB7h1a5vPfP5LfOqzX+DK1RfY3t0ny04wq+f4WCGARx+9nzc9eBYIiBiQJO2tw7rmF//Bb/AvPvFppAJBjgXaZkawEU1H8tOaoA2tkMxdy872Ps9fu8v+eE5mEnpqOBxQZAVSaWrrKHsDdvYm7O7uUTcN/SLnXY89QK8sUHVDnhdIoTBKcHV3j2vbE67c2mN7Z4fCSHqFYXM4oBpUjPo98jxH5zmmLNg9OOT5P/kGu+NDLj72COsnh0zGM5ppQ68acuv2Nq0PHRw7HOvnv3Ql9YmE5lr0+EfDIdu7hwQkVVUw6Pdw022SC8IbZ8a4WEpLmtrRetvdZrEzflMIGbqGdtr0Y4wEGuDIPXYEMOAXU7gYEC7jsPUEAaVzzJxkvx5TKMVoULIRQKqcXFq8g7xXUeUlEYNQHqld6nJISZSBIANeQ1mWKJ0hO0i6CgGEwncBzNkWSUQGgbcNbTtfIvKS3BmrOdWilQYIrQjW0dgW2dESlBYUwtASMNKjVbK598GRAGCLRPX4HiEWaJhuDzme+K5WjCts773W6yIYESMh2A6NobHW0+/3yfKSuAg+izlRXPUiWV0jqUIBpEhacCJGVIRO4YmIZEH+FB0fJIQAwSGkItcCo0AgOZzD1GYgBZujdc6fPc9wuE5UOdOgsT4NmREKoQuIDS7O8RhEUQGeUkQ2+n1cqekXkmHVY944fFnw+P2nuD2+ileK6zduLnknizaJ6FQXjp1UsTjpC7j3UeDGSpTzyBNe8v+Ll0kXZeJPJCRQJJmlvdqF8q+yvlWckc2NdZ574SYXz57oRC88Z04O+PEf/m6+5wPv4Q8/90V++2Mf5/kr1/hv/tv/njc/dBHvI3XdIIJnY22Nv/JjP8KF+zYw0bNWGH74+z7IJ//gs4SguFgZRtJydc9jm5pnvvECX0KxPsgZlRV7B4fc2ptSSM33vedtvPn8OUScI0WCZO83ntZ7lBTIdsb59R4Xt0Y0jePZKzf4rd/7DM8/+mYundtifW3E4eEhEcnXr97hzt0dzm6tc+Hh+/HWMa9nRG2YBY+bHdALOUPZo+pJttYyqmKTnZ0x/9sv/jPKtQEX1irecukibEUOD6cEoRA+gvjTq96jA+n5fE6eV9zeG+OEJtOBfr9HnKQNPBK+JfDub9VSQjLsD5nqmul0SlvXiJCksDxNl/RFjm2Ji/0lRkQHlIki4GOG9QKrQIuM7ZnFusCgzPECJt6yO3FEL8gPp5zYyzk4mHFi2KMqM6q6pZ3XVFXa36KAjJQczpo5dWORSjOZzjDao7IcG1vmHTpXaIUj4l2L7HhSrnHUrUO5lkoLiD7NtWIiZVsSGk4jUAiMSC03iQQRUCqSS0XhFJkMCNGil8AEATHZlyRlmgWX6Mhh6n6GC5QeAAAgAElEQVQPSpG0Do5s0IAQkcirJy+vj2AkANHpNqFQRrFQhRVHfmQRiNLnjMcPxtGXIlVBXkakSFOlRFoWHdfAkfZfT/QOFwVIhclyhBDUFurYx+vAaPMEp0+epCwqXADvBd4LoifBilU3VIwS7Ulw11CQ3BNbQrAoIRHRIaJHS83JE0POnRxx+dYeu9u7eBeQRh0p3Val7dFMNh77lMcP4ALjf/QiWGAbYjxeIa0Qd4s+b2L1y9dgZrR8Ry/hjBz7HDFGIf7l5B7iEWHdzROb8e/9/D9ia3ON97z7bXzHWx5jrSyQ0THqa37oe9/H+9/9Dj7x+5/mV3711/nYxz6DC6CNYGtzgxMbgX/2G7/Nz/xnP0kgzSav37yNUoaNLPLBM4rzGyd4/uwaX73yAs/d2uPqnQlfv3VIYx02RqSI/MT3vJULWzkyHDCu4fLNPa7dOeDGYY0sklX42RMDer6lpzUnTm1x6b5TiEzzSx/7Ayww6pXM5zW5NqxXOR94+6MoIjdvbeMDTIVmfz6nbsZUouGB0+tsDAPbt/e47+wWmQ686fQW2ituXz/ktKk4W1UoYHpwQOikX1YH8mVHtutgJ6hwJICIuPmc4XCNg3GTYMRCklUlje9aVFHg3jhdOqQSjIYF/b5h1ss42D9kOk5agGI5BznSwuZIVyLGZUdi8Q/pnpLUrcWHlkFVUhpofct4HhjXljaAnga2DxTXd/ZZryrObg65b61kmCuqskQZhdSSwuRIpZhMJ0ymM5TOk0W4jdQhMmsdh/M5IoqEvjMa7x2F0WgpyLTB2gZhG9osVdpEga0tRmtCAOs9QihkTG15SMT9EANSgEGQSejlmtwbBMmLC7/KhZcrvmTfFaIDdLxyEkz36Kt1s14XwUgISdnrdSdbpxPfCXwSj2/Ii3WsMoqRTloQESKxK7lViEjUMmgt1RFCTCfD2yTv7yN144lBEEmolfXRGoczy4mNdU6sDylyg1Aaj0B0bGfnHJJk3BVjTDPNKHEhtReD0um7C3jfEqMnuCQ/dH5rxO2dPa5eu7p0UFwgUhbrKGHy6LG693EUi7nq0UeP3VTHD2Anlho9otPrey3Wq3FGYow3/6ycEYGgbjy3tyf8yq/9Lr/5kU/wvne/he/74PtZGwwgWAYl/OUfeT/vfc9b+Sf/9MN89Hd+HxdaprMpw2GLVDrNDaXka89e5sO/9VGkyhChZnd7j9Nrazzx8EVkppjMZgwKgxaCiY1c3Z1wMKs5vTkgeri5O+Hq1W0mjWdQDcldpJWCwuScObmF394hzGfcvXWbPDOcPrFJlef0cs3pYYUf9tivPWe21shlx1eSGSYzOCeow5iJt2xuDCjLgqaegxHMW8upjU02NirWewWPP3aev/jB97HJjE9+7RqHziYV5ldov65a3SzRUmkrTq2ntqnp93q07QSCJwqBKUvmUaSURXQgmDfIEgKqIsNIzbBXMRr02Nkdc3A4YTadpXuY2Im/puMlZRJEXqqgdPtsjCzVMYgwKgsKrWgby2ze0LiEnBLRI6XGusCej8ztFKTE2oaNXs5wFihyRVkofOZQRjOfz5nXNbVrmLcwnlva4GlDZG865XAyI9cZRZaTZxmjYY9+lVHbFhkgenAh4n3DqCoQru2M80xHI+lGFF071ouY/Js8xBAoTaDKPKXVLDlxnVxZ6GgoK+2btBbgsMUxu8cZIHV97n2OXhfBKKXwCx6RXKLE4GiNsPi++OCrUC0iKAR9rWnaGptBLiRrQTA+nBCGfaKPabPtuAEixKTAEBJmflZ3JNUAWgbWhgrrJScGffplUg+PnfCmlApiMuDSwjOvx8wnhwiVSGS+kwKxEVrbJhkXEZFS0NpkRLU57HFibcjEthxV8H75oVn1eo9DuI+c1UUHTxw9YvfOUBYrLo5n/NN+8ptf3wRn5Gd5OWfkp0QyYXwv3wRnZKGW0LYNCMO8DXziDz7HJz/9Ob73A9/J9373exhUCoFla6PkP/+P/l2MEfzzj/wueV6SZZof+IEfYNYEPvbxJ/no7/6/7E9c16oUXL21y2Q652FXk+mM+05ukIUa0TqE1jzxwHk+9aVnMVLyteevsbu3S5VXvOfNj7I2GDJuHUHnZFox7JXUZ84wPtjDkYiHtbWMMsMHHr/Em+87SVO3/Pann+LEqM9jD72J4Bx+y1OUFVlvk5m1jA8PGBjF2rBH2S+5s32DvDAQFUIpTgz6fM87H2GtCDz1pef4yKefZey+WRTdK55HsswwmyRVjohA6AwXOjniNxB4IS1JpksKY5A6MBj2qHo9RuMZ+/sTxuMpdV1j2xbf0TJCEMQggZC4bLGDwXX3m49JygwhqOcW7wJllnFqVDLqFSRb88j+ZMrBrMFbx8HhFNsqZo1nvUpyPBvDMnEthaBpWubzhjpqdieWuklJrAOUSLywum1o2xalDa1r0WqYADeyE2EWmmk9o6kPWO9lhBDREUKQWASKkFTIpUjyQl03xUhNpKEwYEREq8VOu5hXH1lCHNublRQctSk/yv1cJcHy9V8ZpV3xHpvnEaTGAr2ScBmr0kgi6GnD1tRx+cYLcOEMVTmE3V32X3ye8m2PQF6lZ4ZI9InaHkKaNanoaVpoWofyFX0KjPAMBgOGoz5GJ726ubVEKTDKIHSGLnKkBmUtupkhICGRsiK1OmxDLCpE9IhZjQ8Nbd0ShMEYTZ4XoEt0xze513B5cRzk8gSznBktId4iTcdWgfpPC0bpORFS3/m1C0jfcs7I4v3PZvN0bkLAZAVzBx/+nU/y5B/+ET/2o9/Pe7/jbWgRUbQ88cRD/M7Hf58L587xN//GT1L1NH/3f/kQX3/uFtaBDxKBQxu478L9+PmM7Tv7+EwzWt+A6R4YR5lJRqXm7HqFkZLgHaPBGo8/dI614RqZLjhZlhTDtaT511rafsH6Rp95PWc+n2L3PVmE9739LTx88QzWer5xe4fDeY3Rgl7VxxOIQtArJFu9deTZtU6HTlCUOSaX7O5vk5sM6yy9zDAsM7xtuT1pmNgMGSNx2Xo9fnbvuSV0Wc1C+kV2hpLJ3DHZCgSSZFZ4QwWkNDcT0qONASUZSUmRafqDPrNpImROphOapqFpWpxtsE4QnE8zMrHYeCVIaLxHRo/zHq0yTqz12BhmjEzOoMzoD0u0yqldy9645dbuhDt3d6inNeMokg2FUDgrCRRE2SGHo2AymTObtdgmKX3ETjnhTKmRVYYQmiAVKjP4NhkBojxGSjId2Rz2ce0s+RJJQeOTGaWLEi0BFFILfJu0DdN5VojYkilFaUKnWu+SdockRV9ip0DxkuspOgTZEnkHxwPREjzxKnvc6yMYvWQdfbtx1WNDCo73brslhaBAMr29w+7N22xeOIesI3vXt9m7vs3gcZB5JGIJ0RFi0n5alJxKa2JioXFenKY6dZG9yR2+eHkXHwWZFIldLQ0R2BtPOJzMMEVFWfYZFhVmcAI/OUSKSFZUZFlGlVfMJn2mkz3auA32ECGSJ7xHURQFG1unO9faVzkeL0tLjjx+r+N3JFittKKOXhBxBc38Zk/MN7G+HZyRGCPWOqRUtNaTZck2XAhJbjLmTeBXfv13kUHznnc+xtQGnvzk53nTpYv89H/819gYVPwP//P/ytUXdrE2gE/+SACjXo9z633sfJuiVzCOEbKC6VwSm4agSyop2BoU2DZyYbOk8TllVqA6Loita4LYIzrPfDqhrmtm8zmD4YDBsODpK7e4cPYED587xeZogPWe73nv2/jHv/5xospopzW1b5BRIvOI9HUigGuFDR6pPYOqoJ3ltPWU23tTijIpiwsko8EQwotpDssRiajjM+W0uuwuocR8l/Il5XGI5HmBQHXYGYmIgeSUHPBvEPkoYAmhNlKgOs1ArTOUyqhKaIsCO/TMmwFtZ/9xcDBmd38H5hYXEsdHRTAqmTbKmPg+eZaR6ZzgPDt3d2iKnGY4wPpAlVmkaDldFdy3dpqdU2tcv3vA3uE4qSO4VOVrKRhUBbmSBOuZ1p5GJdWGfj+nKnOqKqPSSfYpz0pklhGkxKIZT+YE12LrKU5BofOULFiXRFpFsrfxIQEdfJB4AW1IauNadJqcIVV/C4WIRPpNrcskVRWOgKXSxbPoyIgoj1VER9fxCumV1+syGL10LaVuIkk3TKnE8ehuMyUUqvVJ/y3XCB3x4zGT7TsEAbnKqZrIfD6m0S71ToPvUEHJ8z0zBmEMYSppqgH3n+/z3K0DCuFR3mK1oXUBFQSDvIfwiru7O3zta18hR7LZH7JWRnr9HtZ6ZKbYbzzXbu1y6/ZNDvf3kNFRZgWtHSONQkjJfefOJwsKXrpPLE70vQEMK/TT4tJYDWCPwrrvdSEssxZ4SRf49b4SU33Rt19AcRdZvNYSqWA6a9mtJ3zoQ/+Undv7/PRP/XXOnR7yuc/9CTdvbuMsHXQ1zc4MgjOjilEWCCq5su7e3Wbz/AVQMpmXNQ39QaCfKba3Dzh3/xqbSKJvGVQFk8mc8eSA+YszrLW0vkVnmvvuu4/BcMQsRu7cusP7nngITcS3lrIqeeTSBU4OB4zHUx65eJbx+ICdOzvcORgj78Ql9aHq9+mP+oRezrCXM60Fz197ljedP4sSkGlDYbIu8MRj18hLDmFai1zvCA9NkGxQWmvp9TOkiuBkJx/lsc6ipMb7N841431gPG4RQROERZuIVsmNVYmIyhS5VuRFAgZ4XzHo9/DB0jQTokubrhSdGKpIBoNSGRSRw8MJO1Fho6exY6KccnKwxplhzpn1nJ44pGcaBr2K+06NCKLFThpApRmhVuRaY6RArI/wwmDymrZpEAiqXBFdzf7BDKUaqjLQ6weqKmdQ5QyzivEM5jhcG5J9ujKJFhSTE7E2OVJJlJdMW/C+IUqFi4FMddB+SNJDUi6rnwXJfkkriUmAWsqjrq/HjfNeab951c4Pb5BgtFgCwUZWIce7bGPxeQFCkCuBu7PN7Z0XKB5+kFJK2sMD9vd3EaeH5EFht3e5ff3r9B/uJ9q0twn5JiS6UJgeVBraakAmM070B5zbWCeLSf0guaFGfNOgspzcCNYHfbxdo967y87ty8zLkpPyPuz8NlaWXL5yhRs3b7K3f0j0kSKH0aggyzRSFYQw5vyF8wjZhRK5aq2tzuWq37pU8X7JkLB7lAX+ML4Kl2ShpwUSGTtLABk6xNQbBaq7+AxiGXRDCDhrybKKGD2nT53mwpvO8rN/5xd48YXr/Jc/9Te5cKqHCI6IQYoMpQIuulUjK0QkES0F6IJ6VjMbT1nzPkFWlaG1lno65uJGxWev7MKbz7A5LIlBcOvmdUIQZL0eudMM1vr0Bj3KPA2bZ3XD5Rdv4a3noTObhKbG5zn4wCBTvOPxx/jKl7/Gdzz+IINByYn1DfZ3DpjNZtR1vaxi3bwmG+QoGdk7HHP52nW+/7veRb29S+/UaXbHhwTFcSHTY/NEjmQ2HWxhUTWTEpmiKLDWcWKjBDxCatqmXm5YySvwtaypv7XLOcfN7btMmyEDl1MVGWVWUJgMJVUahGlQUXVVUyRXmoNBxd1dSfCCIBOASCiHVA6ixkbJZDLnsPHst56oJP8fdW8eY+l1nvn9zvZtd6mtq6v3latISpREyqJESZZs2bEdxw7GWyZwRsAExtiYGEESIAMkQIIgEyT5ZzLJDGbgzCTQzMCxDdvj8ciW5UWLKVMUJZLiIopsrs1mVy+13vXbzpI/zr1V1U2KkiLFQ57G7aq6dbf6vvOd97zP+7zPs1PWDErLkZDx8MXXOd3LuGe1z7FuzkK3QueaHMHUtnifkKooxBul0BxCeYRsEcrGvLZtGNae8WTCdlURSDFpQ5EnLHZyFrIBWaIROiE3CbVooLHUZYVTkjyNenTCz5R3hSU3UZ/OOoltBY2NrgMiSHTwUcHfz1jJKsx8imZKJM7v1bn32myCObAZfjPC2bfZFB0Y74hgJGbFMo0grRzXX36NyWJOcnQNGSBXmkZ6ShPoLy2QpjkuH0E3pbO8TChrtq5cYXN3QEHnQOd0PGxaa9LUoKXGJyVKBuqNK9xyeInOQhpNs3RKEIpSOnbWr/PypU22leTCkw9ztCtYKBIOn7ifhZUVLj3zKpc3XuHo0ROcO32ejZ1NtnevMhkMaMsSpKFJJa4aYKqSvZM3Q+P2+o4O0Nf3P+0+DDc/NnsUnxmc970tEfP88p2zy41HR+Gsmwl9xskutca6wHvuuo2f/olP8A/+t/+DZ569yJ13nOb86cME3xAQ3H7LKU6dWOOl17dmfRMNuCjptDEqEYs9irSLVznnT0b17fFW9MCR3qGAfiE5vtzhG998je77bmW5m3P+3Gm0VniZYGfwoWAmd+UcG6MJX/7a0zx41+2kIlBXFVmeY9sEbz133HqUx771Db7+xPN89IfeQzfJWcxSqqbChch6IsTZO2lb1jd2eeLZ57nr/HH6HU3iM6SH7WG5JyMV4TVx4xpxU4w6mHGHEJt3s17BqK5Z7a4SgsKHlrYa4aVHiwzrHMK8I5YPIGa+u7u7lGXJpCzodQt6RZciSWNAUgKlJWg12xSCFpJOt4MxBussXkhkAO1jwV5JQRCOWkomvsEGz9GFBVoJuzZwZWMrWuIsLTNOU569ss2ZlT5H15ZR+Ljp8Q4jBYkx8X1sTfCeLEkwsmVcVdRVzc6kYVhZLo8mvDbaZSpSjh89wnjjAqcXMk4uFSxmiiO9Lt1uRnAWrRXOOeomoJUkSQVuJsdT1w2tU4RgSXRk8TW2QQlwSCoHtYubVoXar5epmAHN2Xh7aw9zBOaNqM38d99pvC1n08Haxt6NWKCf7Oyyc20Tl62SCIFCknjBxDtUkZGlGaqFaTVmahuWuws0uwN2r1/D57FwGa28I8XbzcRXtRYQFI3YwhuBaxtOrhyiFYFMSfIspQ2CVOaYwysUvUM8d2md51poakGbpiwurrK4uMQL9ZRhPWZ4ZYP6yoi/fPjzHCo0xw8t0TWKYCy5hzQEVtcO8UYS/zyi3LhU3Dz2j8/BZ4oDRIbvPLyPSs3vnD3uvpmX9x4ho1eKlBLbNHzgve/hF//Gj/P4E1/jyae/hcn7TMuK8bSlyHOkCPQKw9/51M/x2FPPc3VzB1LNFz7/CINdy+XtCZMTfRaLDsYLur0MrR2aGkEs6sbufM355ZwnX1jn4ackJ4+ucOLoKkcOLZOZ2NcxmE7x1lLWDa9vbvHlR57g7NoR7jhzAqMdwcWag5aCoDW9XsEDH3gvX/v6U2xc2+CO286wvJCTzQzh6qpmc2fA9e0Bl69epbaWu991C8eWF+l0c4q0YGd7TAhR33BPwxAO7nduGAfV4gGC8EhpSNKMyUZU7w4+WqY3kylJYjBBg22ilMw7aDjnKMuSum0ZjytGRUWR5+RFRidLMEqik30DvkREodIw2+gJIFEa7UF7EaniAmTwtCIjMylrWUKhC8aTKaKFQ70uHzp3nCNLS7z06jqTyS5pntHpdZiUQ/LE0MlTsjwnMQlVJQnO0y0KHCmT3ZJKNGyMKq6NLSExVMqy29QMLl8C14JRTMKUQjmu7Yw53M85vNTDzIRsgwh4ESgrS9vOXaY9Oo36lCI4JH6PDGSDpLRQtz5KBoVI2AhiRnef6TUCew3Bc8V3bqrr36gq89bj7RGMBCD87IPvKwzAfoQNQmBDoHENtWtIvCCgUAHKZsxw6xreOzSC0eY2o+vXcGVJXnSx05IQHGl/AS9iKhw8eBzC1UgEXWnQStA4i5Q1LhF4E/3im9ZRdFTceQtPkztkPeFwEvjo3bfj7YRjZ89z9PAaQQV0L2W56pJ2DpPmC1R33I7B4e2UTsfQ7/fwleVd99zD0tHjM2hE7GVD88be73QKDzbhzRl18wMapMDLmd/9jBA/1557I9HunZQVxTFfQJ21s81Z4NTJY/z8z3ycjrS8cOEVhEyxwbE7GvP5h77C3/jZT2KCRYrA4kLKj37sPi5vTflnn/6X7GxvgzRsVQ0XN3c5sXaMxcUMaRRltUMmBY3WKGa7wxAo0pT7776NiztDXrrwEs889RxBCLyXJMmsHQCB9ZaynnLs8Ar333GWUE1Q/ZxER+8kGQQEQZqmnDt1glQorl3d5VtPPo+bNSVLKSmKgjwzLC/1ufX8B8gyQyIstDEod7oZbRuoawu8BSlmtmDMIds98z8CQjiMiSww5x3Ly4t4LKFx1OMpyQxOVkrwTrGpn4/59eJaS9l6mrJmnE7QiSHPMvIselYliYnyOwKqeop3bmZVItEqkBpDnipSI8lMRr/ISMcNovGsKMXqoSVO9boIpTi+0me1p8lVQ3FiGWdzOrmgqmsSJegXCUWng9QKL6OHmRZRg7DIDYcO9dke75KlBcuu5tDRJdjYprm6TROglya87/ZzLGjBK69c4uqwjf1f3rO80CNLTPybZ/CqR5KmJq6BhD1Fd2MUeEHTtFgPlWtoaKlsQAaN9/Fx1nq89eCiFrnzDqSI6gwiIG9oQJpPQMec6/u2Z9MFHyDIfUWF2ee9IdULAS0lWSeh081nNtkB7QXlcJdysEtwniRAuTtgsruLEIrMpKBqTGbQi33ARTkgon0y1pI4T1+AwdLYmrrZ5sSxM4ynQ5SGxrdkPmBUgSgycmXwWLJkmdPHVvDCk/QWyIucaVvSKXpM0wFHDmdMRmNuOXaI6XhMr7eCNmCdZ9hMuf/BB5FK7ZEP9gLwDW1k+1+/7YmcB6P9O+ICvQfLCILYJyvsedocwAB/UNp0fx3j4CI6V3fHtXzyYx+kMBHPzvMOQs5tu+Gzf/ZlBqMhH7zvbhZ6C4ynI771zZf44pefYGcwJk1zWmdpAry8MeT+6ZCit4CWAq0SjElRqkEFh5aSRAvyToe2CZxf7nDL6gKNl9ReMJ7UVFWNUhI5s4Ww1ZjFfsFksIVVkn7vBJ0iR0sVQUep0dJjlOLY2gqFttgqhQBpnqPVfBGBrMhIEoNQsb8lT1KCSRFCo03g6vXreDEzO9tr2JzPrb2juAfQzpGW6FAcRTONycjSlLWVJRyeelxSjYekIeAksd9OvC2Wj+9qCNiHk2ZKC/NMiaqinJYoFa22kyRBa43WCoIlWmv4mBUqQZEYOpmJ2+YQkEqyWCSMqRiGhn5QLCeaXlFQEHDTkumkRBtDp2PwvqWuahKtydIEIWbiuk1DU7fUdfT8kTJhYbHDXbedZflaxXQ8hVTSSw5zSBkwCUeWFzm91KUr4OS50+yMR5gEsBaFiUrjUmKyBCHAthYlIhwbvKeybq/FAy+QUmMbR9NYrPNR8JcQ20+IzGUnwfl5Ji0Q1qB1TAzcDRvdg+tJ/IV/C0PGt8Vs2tra4umnn+SWW24hy4o30JDndswER4ultTX19iaZO43WBUoatJSkRU6RZjgtGLYl/bXDhMZR7gyYTCd0U0MI9gA0EbMx09YsqpRMOpAVw/YKd6zcg/COkYueMW3rCMqBLhBek0mHCC3r66/SXTlMp68RzpFYKFSHTBqmO+scPnScQq8ijx2nKAxbO9d47cpVTH+JpcNHsW9GdDqQFb7Z74S4kfs2952fPeIHfHbefmMejOb+LwFJIuHEkTVCAIfn/vvfzWf/4hGmVY2UCSD4q4ef5KGHHo1ulFKgZIKSCUVnIdLcq4raBy4PKi5fv8K5To5S0YtFaEWiE4RroxqzsxgjSYyhu7aGzlKESjBpRlm2DIcjJpMpo+mIwSDQWzrBqZOHKfKCcjSMC50Ak0YhTJqWRAU6mSGfdeKPBldY7nUxSYIUkGpNIO7MlTZ4AczYUGmSIaQm63UIJsGHcu9YvamCyd5tPyARYta3srqMSgyrywssFCmBmivrG9TlFJmImapD9CB7p4zYPiV4k0sKYI9mDVGXT0qJFAohZj5BDgqtyKQg14pMS4xS4KLnUSfxZCowaRyTckquDBPr8ZkmSxISpRHKYa3H2UBdtSitUFrTNA2ZjeqasRYnkSKQZQlJUVBPKvrSYhKJ1IYlk3Hm9gVMosFZUuGRQZAUmpMrh0i7XZpZW4HwlqxI6fQ7TKdTnJbRdt072tZHZrK1BDH3LpKE4GMG5ObrzUxUOASUECRKRIjZRw0OLy1Ba1qiGvo+QDNTswj7uplvVTp6WwSjspzym7/5f3P7HXfwoQ99lDNnzpOYdA+LjCPgfGA66z72zhNsgzA5upNBYkgyjdCCNvFMheX42lF83dJMpgymEw73e4Qw3ZPEQIANgrossbVCBocQluvDV3nyxYdZSk7RSIPWXQbDMUwblFHYcszw2hVevvAcvaVlTt3xHsxsgahHU4w0JLrL1tVrlLtT1g4fxyhNW9VU0woQrJ0+Q9BmjwN34AzObnO2yjwz5AYJpO84Dqjk7sO2BwPWnD21z6J6JwWy/YAELjhaZxmMRqytdVDBcfLoIX79V/8Wf/GFh7i+uUMInqoqGE8qggClTCQ+zBuhZ4u294H1yZSpTAlBkOcZ5dWS4CO7ygWLCx7ro0dOURgQDjHTiAw4kC0mEdTbFdeuXkOZhKLIyYuCLE3xdfTQ8QJUYqIdA9GNUycarVN6reP55y+wOxhw9vRpFvs9ijxFGx39iVR8jjGGupyiC8e0qpiKwLixuOBQbxEs9lmV+0MgMEpz8sxptsdDzp84TKECNiguPP/yGxbyH5Sf0V/H2J/ZIbJR99sXuXEHD4RoOumJqi3eeZQU9FIT+5QkpErS6+So4HE+EgSWi4KJDUyrElzAty1BK0gUSZ7EbEp4dssSGwJpltDtZkgjUKnBz1KUxOSEEEi1Rmc5iyuHcGXA2ZK6aVGGaHfu2ljn0gnGJHSyHl0jCUHS70hCP6VpSqx3JFpCanBtDHo+BGrZUlftbO7LPdQkzNaKuplJpCmw3pFriVECLzWFECQEVnodDmWGqTO8vFUyaEJkHXqPFwfUcuYs4bc4R2+LYEQIVGtkl2oAACAASURBVNMpTz3xBC+9+DJ33fNuHvihD3PixEm0mXlvBIFQmry7xMqxE+xsrENbkYmMxk1px2OEMzTjAbtXX0M3NUuHDqOqGuda5GIfU3Spp1cjLTFEH3kHbNcNF7YnHCo9a0Hhfc2TL/wlxxbez8nVO0l6OY1vaCcthEA93uLyxVcxxQInb30f61sT8A2Lhw4T6oqNzets7uyyvTtgfXIVfEF/oaVyFRs7u1TOs7C6jEeBl6Bm7Lm9mtmbj3kI8TfB9TcKs89EHYNAejWD6vZ7SOYzLtIxPcy8jGSAH5Q23V/nmGe51ju++NCXOX/u54jKgA333n6Yu2/9RRqruPDqRT79r353D98OWHxQs4twrjYQIb+pDby+M+S243YP1oiiMNESunGBamdAt+jhA7jxiDQvSLICoQyj4ZD1qxts74w4vHaYxYUltjY2GH5rh9SkLHZyci1RiSHpdME6KmtppaQOgno0ZTSZsLK2xvrFizz3/AXOnj7J6qEl8jQhyTK8j2rPTVMzmU7JehYvBIPaMZkZVYaZK/JBgsIbbCT29nrxmzxLOXrsBE9ffI27b30f+JbgBM8+8xzvUuKGHPydFIzmY97wfTOh46168YQArSTZ3tyJ/Vyd1GCEo7UOZvWehEC/0LFVIsRNT6cwGBXlx6ppjbMOLwN5akgNZB2DSBTUkenmnSJyDzxGSZaWF1HCIPQWw9EucvZZZYhtCEZ5siSgVCRkJUqghESalLRI8QTqskZLjTISFzytCyDj/HfW0wRPS8AGh3UthEhycEESHGRaRiKXlhRGs2RgrZex3MkQraNuLMcWl3jopS22rceEgJUCNfM/Cjdvrt9kvD2CEfsXyXQy4muPPMILz13gfe9/P/e+9z2srq6SJimhtcgkoVhdYrrzOsVohFlZYro1YaW3yOrZM4yGYybr64g2YJDUzZSSluVjR5HK3HBhSi8plERrxWYp2G3gUPAID8G3TOsx49rSSyyJVjRVyXA0IdGKzvJhTp29ncXV05hpxWDzOq9euMBwsMG3nn6O8c4OSgaOHjtC0ILBeMS0nbIzGICJbqCRMLcPuc1rNvsLx4Hjwz7U8IaNHN97TnOgEvU9PvPtMeZNr34m7+SD5Ctfe4ZOp8MnP/4hlnopTd1w6comf/mVx3nk609hfeyfiXpcETMXMnr9BBG1xzyAh/WNEY2tyLTh0MISTdNQV9PoC+QEzgtG05pJ1YKA3ClE6agay8a1TSZVxenzZ1lbPYQIgaqpeebZ59naHdHPNbecOs4pmVD6q0ijsSLghGJrd8jOzi6HlhY4dvQIy/0+r7z4Ci+8cpFpNWap36HX72GyAg8MJ8OoHo+nSBOGdUB6D15E2vq8YfwmQtDBMXezlVJw6NAKOi1IgPNHDmFxDAYT1i9d4q5TGc5HvqYP7ywLifn4trDlmwZWR7y6JEmS7ZFSpBCkJipjF8Yg8pS57pqfqRiIIMhMQmoShIRpOWE8mdIiscKRG0muFToITNCIpok9jAoU0RXatjVJqUmKgtXDi5jEYNZhOpmQGAMiGv4ZFVl/MfhJVJJgkoQkNSjJLEuWtJVkOplEmnqIELcQIurrtVEuaL4chOCjIV8IZEGwnCiKJCFXkp7R9POM2lpefH2DuvaYAEtLnqN9w852SUDvKYMfJAW/7YNR7GwWzGlkAsdwd5svfeHzPP741zhz5gzvufc93HH2ThwGbeHQyhGOHz3HRBlMd4GVY2cJWqDqS6wcOsuhvAOtpRztsDHY4tT5U4RZlGYG07WADI4UgWgdvi0JvodAIZWgbKe0QTEZT8kXekjhyPOUhYVF0ryPzHvUwpIWGQsLPaqhwneXOHvuNq6++ipNW5H3eiysrjDY3GRaV1RNzfLiIlma7p+keTSZ4Wl7/P0DI9z0/41kuFmP0bcJLG9+od2YTb3ThnNuprog92SNEJLPffGrfPGvHqXfyxmNRlGY0gpMkpKlGc7FHpH5cY5EiKgv5pyLDCEB22WFUBm+bQhAG/Z3dkopnPLsjMdkWYYxCbZsKOsJG9u76Mxw9rZbOLa6hpaxx2VpaZGz586SXN/i+vUNvvHyOuM60F/oI7VG55H5572n3+9RFDlZmtIrOug046UXLnDx8jXKaokVGyh6MYA2rSfLUqq2JXUtz79ymcG0xusE4Q+0Rhy4HdQLiwy6OBeUUpw5d5ZLl1/n/PFl+llscHzluRdpRmMCaaw3hDBrIHjnjTcPygdWy70RuWdzCCvWkKK3mpQS4R0iOBIdm9ijW2zAOos0gkRpJAIVAq11SC3xQjGYlHgcK/0ei92CTBpsZfFiuvee3kdZJymAKvaMZUWfpV5OfuIoW5tblGU558buG5IGQVoUZN0OSZ4hQ8ysrLNYPzMA9Z62tQilZ3PhRtNFJaMyjBQyqlMAeRAspwZtElLhSWRgWpbkWtLLClLd0k5b2tGQlf4Ccjdmd2IezMOcKfzW2fTbIhhBDEYHuelCRChtuLvL008+yYXnnuO+ez/Ar/3H/wk67WGOLBC8phnsopSkVYF2MkIqzdr5uwnCMrm+zuZrl5iMRpjFBRTRWA/vIDg0Fu8slfWYoHA1CC+J22XFpJ7SSkddtZBldJdWSbyMvj9uwAvffILNjQ3qaaDT6VCVFZPxLoOdq5EoIRzOh+glMktUE21YObqGFBov4oIigpp1NIu3SHGi2bEIEiVULDwfkHCZU+fmPVnzGDe/7+ZJMIcf5q/jg3unMXWZq3cLIWKT5wzmrBpHuT2O80hlJEpilEEKTfACF/we6SMQoY35Ll8AVgoubW1x+eoGxzopk6pkNJ2iZgFJK4VLBNOqxkuF9oLGtuwMRhTdPufPn2Sh36PxbnZhC7Is4+jaKkmiKDoFr16+xjdfu8zKoQqdJhij6fc6HF1dodftkudRXcQ5h04N5+64Hfec4PrmJmlWEHQJCpIsA6nZ2Nrl9WHD733+EWoEwVuEMm96zG68I/7VIXi0MZw6e5YXX3+d+x68E+UarJd849HH0d7tQXlz8s+8H+cdPfYOh5ghD/vXlBDqAMwZoTodAs4G2rbFN5raNEijSYRCz2CpzJgZySggBbT1iHraMhiXTCpLJzMUiaST5aRZgcgCMk3wrcO7KNNTj2qCsLEPSuT4tkRpRbFgyHpHmE5ryknJeDxBSUlRdEilIc1TdKYx6ezchyhyqp1mOrY4H6I4qq6xrcdLR5DxelACWu9wUiF1RkJFEWClI+nlCdJLOlmC9I5eN2d1IUHa2Pzsasdoe0iSGfAJXgbk3oYlNsrGufcOCEYwh+pmGGPkmqJmdY6mrKhGY4xUpHlONRrhdhqa6RCTprgQaKcTlJCEPFBOJ5S2psGTdztIrcHPWQD7X72AVgq8FtRSYGeBCmDSjJlMS2RWsDOdopOMLO/ivKNxY/K0oZd2ZzCRohwMGOzucPn6NuPJiJXVHmdvK1CJwLkWgScYwYnzJxAqeirN6dd7yc23PThw44n8NuDcwbu/SwTue2lM+05DCJEBfwmkxPn1uyGE/04IcRb4LWAFeAz45RBCI4RIgX8BvB/YAn4xhPDqW35eAntJ7uwYBhFi35gnBgAAATrRs8+l9uoF84wgzGi5Ao8UAUK06JZesTV1/PHXnuXBO06wuTNkOJ1wYjkjFRKlBHma0TiPtYGmbZjWlk6vx+raKsG1jCYjjFA4JcnyHKQhLzocM4as6FJ0O7y2fo1hWZILQbfIOLyyxMJCn4WFBaRRUU3dedpyzGgyYXlthdFkzKX1a5zLT6CUQhqNCy2PP/86n/vGC1zZneBkQggiwm/q2/cbzTNwIUBKxfETR8mXehTXNLedOUUbWqaDilcuXECINloKxNQAiUG8g4PRDT16840cN2/YYA+NiMsR1vt4E4LaOox1GGvJshQJKKP3Mg3vPGXTsDUo2R6VSK1JU01iFHmakHUMSZ7haSPjU1qiCK1EkOCsZDSoSU1LkaXY4AnOo6QmT1M6RZfe4iIgSLIUoQ1qpjPnvNvLmPws43ZzBEGKqOqBwvqAdR4XBA5F61ukFGgJUliOdAuOdRK08mQ6oW9SVlcXKQpFLi31cEI3N2RLC1y2NVUr0d4ShJ8pVkQ5tW9LYzww3jbBaG7K9AZF7tkU8T7Q0Sm+tbS2pWoqrK0ZT4bkNsUPd5BSk2YFTAa0wwGubmkIFAt9hIrdxmH2Xj5IgpMUqeLYssELjxaO4AQeRxl6bI6PEV6ZcNe9a7i2YlhOWdACLwTLq0dYXV7G3L2GaFOGL7/O9TxjMFpiKZNUvqJ/qMNSN6UaDxiOdplORyydWmXxcB+E29N7+p7DwByK/S4kNr7t+P+vVFQDnwghjGcme18WQnwW+C+AfxBC+C0hxD8F/jbwT2Zfd0IItwghfgn4X4BffOuPLmJdDzf/CeFVpJ6qWGGO4sN+j5EpCOwLh0Zs3Vq73/A52+1HW0tHieKRi9u8ujWhqSvOH1lgbTHDANYFdBL7UYKXeA9Lacri8jJBeDZ2ptS2YTHL6CSarJyQFR1CCCip6HVSEB4tjzNuapQQ9LOUbpaRJkmUOBIz6f+mYVCWbO9sY9uaw0dW2bpyFSEkSiVoneCd5/nXr7O+O8SLqLZ9Mz/yBlA23LipCQGSVHPf/ffRVp4P3Hs7RQbCCl771tP46QAp5V4fjpQxp/y+5t+/g/FtWyZuQunelMwgoLYtPpW0wTNuWrpJAkHiXTTaFEohQ6SJCyGYTqeMhxOuDSqCiYGo8S4KM+cZ6dIivUPHcFVFM5ngWo/RCSQerzzKKiYTy87ODo2zZEWBSXVUgQAQEmMi6UYJjdQJ0UJckIhY34qU9Za2tRBizbRxkbbdWkfrPDbEXrwgFAGHty3CNRxKDasLKQt5wnBc0oaalbU+p48t4GyFm1YoJTnU6yF1wrDXYbITsFKDt4CeTUU3o3mrt9zzvi2CkRACjZzx3OfFezkjf822wD5eyE3VUpdjyumQ8WREWZdM68jlz4ouw3JMPRziGo93DluVhGSuMOwRaBDRv0Mo6HcyTq/mOO8olMU6w/b4MOsbywwnKWW9yS32bJTq8I60jZz8Is2QQeJdicw18kQHTZ+84zjZO4lrJlGIs6159bXXmdYNDZ777rkTpIxipjdkO/N6UFxg51fIfHcWEDDDpWMD68xjZu9xfiagGheiwI0X3d5bhf3v/Yzx84P065xZQoxnP5rZLQCfAP7m7P5PA/89MRj9zOx7gN8F/pEQQoS3AJcXF/s8+OB9PPH4U5w7f4amaXjt4lWqqubI0VUCcP3aNRaXltnd3d2rDe1R+gnoGe15XnuC2SIkNEZKtFTUQvHasMYhWWg8tfXI4DBZhklS0pn5mkCwstAj6xSUTc3m1oTXrm2ykfVY7nXIE8HSQkuiFVIqlEpI0CxmkGjAB0wI0FqC9zRti3SC2rZs7e5w+dI6vqlYWeiwstil0MfI0hwXYrOsbR2XR1MalcyKxg4p5l5Ebzg/N9+DVHD06GHOnz/LtY0rvP+udyOco6ksFx57mCI0NCIyr4SISgTq+90M/TsYN5M54p0z4pAIe5fijQSi/cxo6gTOQ9vU+FTRNJZKCDKdEAw4HE3bonQ8JzvbQ6qmZlR7CqORs5pJ1Vpq57C1xdUNUhqaJtZXOv0FbNthMhqipKepSsq6RjcN1rf0gEQlKK2jzXfEFmNNprWITINWUXm8avAEqqrC2xZrW6y3cZPmwNctrnVYTKRhe0djA7UXyKBY6fRZ7SQs9DoMxi1XN0as9gccHnRIjSRYS5FoFheXGI8r0m6XCy+u44NACk1cjWY1o1m97a3gn7dHMGKOKsaPL+UszWUmJDHrtBdaIbWh9paqKmnqBiU1xqRk3Q6eSFs0C8t0kox2sMVgNzAdT2jrElPkBCFnQqmCvumTr3ZYvecU4+k1tttdLm6scWXjEGUraFtL2cJwalnoCKRrKStJks38j5IMrRKwjtQElg4VdLswGjjGu47JdMLG9jYbwx2CENx9370sH1pgJjgdjRnhANNkXgfabxK7OVjdQHiYHb2Itsx3/28sxt6gyMycFh7wN7/mD2iI2A35GHAL8I+Bl4DdEIKdPeR14Pjs++PApdnntEKIARHK27zpNX8F+BWAY8eO0O+nTKdTPvzB91LWLS+++HuA5/Y7zpCmOZ/97CUeeOD9bFy9zqOPPQFS7OkQ/sxP/ySPP/E4ly5dxc7gi09+9IO8fPE1nn32FZRSLC/0+NGPf4w//tyfcG1QsjEYMG5ypAedW3IkShpc8DN5GEO3k2MSxcnDK1y6dI1vvPIKLsByL2Opm9JJDXkSvY+6WYHJE3SaEGygsp7xtOXq9i5Ows5wyuWrG5TjEYf7Ce+7505WF7ukeUav06Esq+i4GmBjd8zm7nRG0JlJLYuI2EspQegZ4SDEeRsOeFgJgTGaD33kQa7vbHF8bYnlToLznpeff5HR5jrSz9oF9mqczBaXd1YwOjhu8PaaB+iD7NW97Gj2V4oQZXIaRyoCVeMYySYqdFQNntgMKr0DG5g2DaO6wXlP1bbYcWA48owbQZEZ+ttjlNhgtD1EK0M1nZKkgl6njvPJxEzMhcBgOCHNU3JtGA+HCBRJVqCNIEnSfThORtjWS4FrLLZpsN7hbIuzlrZtImMuQNM6qqrBCUFrW5AK5wOtC7Q+kKcJi1nKkV5GkaXsdBJGQjIdNexuTlheziiKAllZnLOo0PD8wLE+mEKQhOjGx1zIOZYh3trT6G0RjGC/oP6GnQuzCSPAK0FIIvNIdnJEiJ3BSZYjkhTvPJ3eApIk2ozXU0xS0Fy9wuWnn8QsLWGnm3RkysljZzi+fB7tE6QzDIfbfOGpx3np+phJIyjLGmclSqd89cmX+Kkf+wDD669RNTVdm1OXDb1+n47JMUKhVWyI88Jjm5KqrdkaDdgYRCHOO+65gzvvvYUgIn4r5yvJgUgQfMRz3+IoMe+Yn7N65vdK8f8tv5nDUzeCOt/fCCE44F4hxCLwr4E7fgCv+RvAbwC8++47Q1nWOO9xbcXGxg6NiwSMzc0Ngoskj7ra5sd/7EGeeuabTEuLx3PHbae4955z/PEf/1u8i41+3W7Oj3z8AY4+v8Dzz7+CNgmT8Ta3nl/l13/tP+J//d8/zXBac23ckM+Ya7b1pGlGXddA2KOa5yoKZn7ywQ/yru0x37hwkWsbm9QYRFAIDI2DSdsipxVCaXqLy5jlJfqHFhkNd7l48SKPfv1pjvaX+NH3vZsjR/JorFbkKGNQASbVFKUNFhBFykc/9kOU1mOd5dHHnqaqPUcOL3Nk7TBXrqyzvTueoQz7JAQhBFLAmfNnOXPuLBde+Cb//oOfIDhLbQNf+9KX0DjEDPoLe5Tu6OvzTqN2vxW9/eBj9hbMsF/DZmY00jpB5RWjxiOlx5gAdTTfy4QkVQKpHVXb4qWgtdFv68rOiN1pzeXdkqOHFnF2kWkDnTRFOsektFyZTNkaDFnr59xzeo1uYphUlrJscC5gdMJ0OGbkBxTdgDQNdCO7EyGomoY0pJGQY2u8bajriqauaJqGtm1prWVaWUrraLWibhwKjw+CFknlPS440kSy1E/oZppESlY7GYMspWMShAetDWmisQ0Mx2M2G8fnn79M48JsXdoPQt/teNsEIykVUs7gpfluJYRoc+w8UknGZYkNPqaoSYLJLAqJMYYgFDrL6XT7hEkN2tDmS6yunSZtPRuvX+XapWucPX2Gu2/9AN38ENoXVOWA0XSHF154nlcubTNts1lQ0AQUZWu59MI677q35OyRk6y/8hw7OwOKJGVrMKC/0KNjuoimQTQlg50d1l+/xvWdDbYnQ4KR3P+hD3Di1tME3XKA4zZjc83HAW24PbjgpoMU9k7x7PfzjCeSPOZJUYh33vCcOSHk4GtF6io3fIYf5Agh7AohvgA8ACwKIfQsOzoBXJ497DJwEnhdxFVvgUhk+PZD7CuVR5fXWPvRUqGVYjiZApCZhMOHutx/3z188aHHSVPNT/7YD5NIOyvsxqN+1+238vXHHuf++++m38ko6walJNNywv333sHf/dTP84//2e/w2o7l8JEcV9a0XUu326NpGhCCxlum0wmpSTBGsbSQkcrAyf7dlFXg6tUNdnYHlM6ytHqIztICdVMjk5Tu8jKDwS5hNGXw8lUWS8kvf+BjHDmzxtKSAmoQEmkSbAhMqil1GzXpticlt917N09tPsrGy+s8+MMfZmN3BxUEv/53fpVHHvk6t/7cT/KZz36erz76DELeqEDY7Rb8xE/8ONNyxH13n2Oxo3HecuEbT7P10nMc7UOY1SF88LRtS8Qx3gzye3uP/evlgA4mBzIk9gPVvFVgPuaXkxfQCsGktSgpKNKAdtGsEyHQnQwbjdnpJBlNM0UpRdk4NscVrVZcGU0RrmVrOKWTJghreeHqDpeHNaUX9FVgOracWe7gbSDNNAsLy5hEonXNZDJBSIXJMmzr4kbYGIIITKclzjpaXxKspbYt06qirhuc85RVQ9V6audxQiF1zHSdJ9aPvI/ai0qSZxqpIUk0h8US4xVHqlpSLUnzFOdblFRMLXzu2StsjKqbVrT9/+OS7t/+ckACseeT4SSEOWAX5rRbQDqaukJIj0xSEpOCcRFvVwZhcrpLh9EC6BVgPSoozKHjdE3G+X6H/vETGJUSAtjGMtq+wsXXnuORZ55gY1xRigKZZrS2pG4Dgpw6RN/43//MI/xP/8OvgXK89q3nuH51naaeIkyCUim5BOMqtjY3Wb+yzqgaEfKCe+9+DyfPnY54sVeAR8iAFpq5Src/wO7b1+N4E+xsBq8IouJ4rBmpSIuPJbED5aZwQ32IcOPiEQIILCHE+pUPcwOt7/NcCrEKtLNAlAOfJJISvgD8HJFR97eAfzN7yh/Ofv7K7Peff6t60XzYtkVKSZKkNE0LzP4GpaOIroe6Dnz10Uf4sU9+mKe++U1uufUOdra3WFuKDcfeebRRvOv2d/HlL3+Bj33kA5w7e5xrO7tImbC53fBP/8/f4e/+p3+TsrT89u98hjES17SMRxN6vS5KSYzWjCYTlFS0qiZLJZ3FFYpOl91mm7yQ3H37KYRTjEZjhuMJze6YxHu08vidy3R9S7ff5d3HT9Jd6JMs50jpyURDCFA2DWVdU9tI4NHGgBRcurrB8XcLXn7xMk88d5FLGwOsbfi5n/4Er738HP/P7/w+d91/Dyu9HjLMDPFk9KnRWvHBD97P8uoS25tXeM+77sA6S1U2PPzZP2JJe3JtECJSlAnzfi7w3r31CXobjoPL5EH+3M3EhoMSZHN1jjgUQkaCi/OK1jrKukEqkErgjKL1Fi2iGzVKkkroGMWZo6usra5QW0fpPIPxlEs7JZIJEk/aMXzs9HE6ac7uYJfJeMLrTc3aUp9zZ26h0+kQpCVf6DOtN6iqKSEIvGtItKERQBI/p/fRHK91USarrqJwb91aauepvCegCI4o3OwVTeVobYMIYISkaxRdY8hSQ54n5AnceesJdjd3WOhlpIXG7U4xKufZ7SlPro+w3hBkmFesmTM19ynd/i0zpbdFMDpI+ZFzdFsI2hCobTNboDMCEtfUKGmQJiPNBfiAThL6q0dJuosI7/DW4euWECSeQCc9Qre/hJR9BqMd6nKHK9ubPPOtZ3j6hRe5eO06h4+dQMpAKiVl5XHBYLKM0cTThpQrOxN++w+/xH/+n/0iuexw6Ymvsjtsok6YVmxcX2dj/XUGgwEhBLI8w/QP8+SL15m2gQcfuB1lAkhFakBpj5czLNWJG7OZA2OuGLFnMcHNF82bWUuHPQLDzS95wzo/e+0wK97+gOrRR4FPz+pGEvidEMJnhBDPAr8lhPgfgSeAfz57/D8H/qUQ4kVgG/il7+ZNhG8iPi/mvWPgZSAxeZTBdx6RGD7zpw9z153v4af/vR/h9etDXnjxJW6/5QQQ5X1uPXsSRM3Z8+fZ2tzi3nvv4OHHnon1Rzx/+egTqFTxa5/6JYwJlNvbXHz863STkmI8JE8yBAJvPdOywmcZVT1B6YQ07ZBnCdPJhIndoci7HD6yxInkKFqnEKIwZvA+Eg60xiYSZxROgrVjVGJomoBtIubfOo9rPVppmtaysVsjsdx6y1nO3XU7Tz/1As988zmube7wiY88wI//yAM89PVneDGI2V+sIDiUFJw6cZQHPvYhLq9f5IF330ouA/jAo5//CvX1Kyx14gZFC48SCuFtvBZnUjU/qAnz1zX2ITgfyVGzhtY3EOdu+LuiUDMC7KwX0EiFCgEXPNPWomqJNDKqWzeCpJNFzx8gSwwd79FCsSAS2rYhSEHZ0YxqB75isZdx+7EjrK4sk+qEsrL4AGlmyIymkxiCixuuJElYWFhkOh7jnI/9iqEhBI9r1F7vnfMOD1RNS1m3NK2jtlEGyFkXCVxCRm8j52msp/HggojNvEqhZSBRGls7Uq3paonqZayuLUSFCGkYBMlDL60zmfUz3awQE/x8dxyP5feVGQkhThL7QNaIb/MbIYR/KIRYBn4bOAO8CvxCCGFHxDP5D4GfBKbAp0IIj3+n95kPOSuINN7TekvTNnS7Rzh7+r3cc98HEN4grcfoFI8gWEe3v0RnYYVWaIK3hNDShpbae4JO8EIyqMe4yZDNrcs88ew3+fzTr/Dq1Wt4ndHNVkinnn7SgJyQqRwhPdN2wHTSonRGYgxf+OKj/OSPfYw7P/IxzGKXl/7qy7TDDdKFLqk+Qjm4ihI51ltM0uO2932Yz/7V13n8D77M2Hk++uFbOLqcc2RlgXGZMqyJcvY3bTL3SA1zLsIe9LYXjm58/Js0Moa9fwcKtDc8/gBcE+Y/ff+8uhDCU8B73+T+l4EPvMn9FfDz3+O70O33qa3ntfXXefd77+SJp75F03jW1tZ4/LEngUAnL9jeGfHnX/oKP/LxB/njP/lHCAYe1wAAIABJREFUvOvWW1Do2UVjufe9d/H7f/Bv2dwasrv5Ov/hz/wUnVQyHjUxgxCCP/2LJ1jqdfjlX/g4n/vc02yUFiVH6Czn8GKKcxVSJ4ymFV4ofNuQTqaxbikkOjFUVUsz2mVSj9EqxZgMYzRSSZy1KK0RqEg2kAkhKJJEozVUjSUIhXUtVdUiEZjE4JVkfWuI1wGk59BSD2crgvT80Z8+xNJij0996j/gQ/e/h7//D/6v2IIoLODp9DN+9hd+lqvXN1jtdzm3tgSuZv3yVb7xxT9hSYW9RvQQwh4TbJ5xv9MMGYEDxA01Qwb2GzH3d/BvvJ6UUgSinQJBogIkJi7klXOkrUMrQRvYo3k7FzPIOkhaIgtRK0VmCjp5wvJSH6MVrm3xbcvyQg+RGIKzdAqB0gaNItgW11qkSWbsSEGn20GJuJEheGxtAUHQbg+yrq2lbhqqtqFpW1rr8Aik0iRCUtcxuAUb8C7sUbw9oEIgkQIpok25URl5mlCOhyz3UlZXV9jZ2SXoDo+/do1Xdy0BjfFhhmzNj/fNM0S+yX3747vJjCzwX4YQHhdC9IDHhBB/BnwK+IsQwv8shPh7wN8D/mvgJ4BbZ7cfItJ3f+it3iAwE/8UIL1HMmtaDJYiX+H47R/j1Ilz3HbyMEG2WNsQgsS2FVolJJ0+LgRsNaGtKtq6pi6jurdzVaQ0VlM2r13k8Qsv8BdPv8KwFpTWo1EQCuqxZ8NX9JOKTjZBG83OpMZ7jVZLSJnTVo5/8a/+gP/qv/lVzrz3fpbWTvDsF/+MKxceZzrYBSQ66+Oaljs//FOcuO0u7Bf+iiubu/zR577OKy9e5lf+9ic4e2qFson6Zn7vkj5YvJkFC3egQOxjtijmNeigZhBDOPCc+XdvPOHzDGv/5iE4JCHuzJFzofC3/bh+fZsLL6wjlOZf/5s/54EP3s/Hf/ijTCc1X33kq0zGFcsrC+RZwaHVo3zxoUdZv7rF1c0JH/nIccqm5djx4+SdPqfPHMcHiUdxfWOXbifjvvfdzZ9/6WGOHj1Bf3GZq9c3+L0/fIheZ5FpUFwejknI0WqElopuakiShGlVY30g0dC0lkRGgoJQGqSlbR1VVeNdQ/AjECHaCEiJzlJMmpAWOZkCnScopbHW4bygbh2TqsFbT5F3cEiGdcO1wRiL5pGvPMvWZESaZhw9fJiTp07x6d/8HF9/7Fn+/n/7a7z7zjM8/MTLiCBIEs1P/NQnUZmi2R3x4IcfQLY1lVP82e9/hqIekSagdOzin1davPd7a8mcPPOOGTfr88Xvbn4IEWmIX4VQe20BEgguKqkoITACKqkIwlPZCtPkCKNAeFwIeK2YNC0T53FBketoC2GEoggKPbW0vqa1NWlqaG0bA5ywWAeyETN1eRBpOss6PIIEjyfJUkJdYhuHNJK2sYgWCLE1oGqaiBDYNtb5RJQXSqTA2ciwCyJqMXoRVWdEsGgCRkoUgmChchUykwxGDdo1LC8dwdoG0TZcLz1ffXGT2gn+X+reLFiy6zrT+9be+ww53KnmKswgAAIkCIITKE6SSEpNk5RISd0irZA7wgqHo90tO+xwOOyQn/ziCPeL290RHh6kFqXWSGpyi5LYbEmkRBEUCRIgBqJIzFNNt+rOOZxh7738sE8OVRgIhYoysCNu3bx582Zlnjxnr7X+9f//MihedJ73zhwoFimvztsSL7e+bzBS1XPAue72gYicJtFxPwn8aPewXwO+TApGnwR+vcP9/1ZE1kXkZPc8L3emEDsYKjpL9DVN2xDMgOtv/iHaCG++/RjHjg4RBwejA5rRFEtkeLjHRFvkYJemqmmrmrpu8B2TZHSwy+6553jiuWd54JmzXNir2G3axAhSkFAT25o2K2hjhnoYjwOhnZA7ocwCJuzjnCK541v3P8rffOnb/NhH301x4gR3fPzTmOG1PPjXn6OX9eivnOCGd32Y6268kf3RLkeOnUAkZ3+/5fR3t/nX/+ZP+G9+8ZMcOXpjsgrq2EmoSRZInSL8xVna4uL//kPwlokQl4+VvvIh6ei/+NZreY3GE75z+kkQqCr4i7/826UMPmV3m5e2+eVf/a0Eo6py/wOniQqf/f0/5Q//yNA0njwveOqJpxmNG4wKm9s1//Jf/TLT6ZTQBj7/779IaBNUVYfIr/7unzJYXyVOA21f2N0bkRvBr/YpYkCjUo8nrA17EAUxOUEj1kLRyxEXCIEkMMTQajeMzNiUCRc9XFbgsgKjaSQ5KD4EplWF954yc+ACzuVc3JxQ9ksyU3LDdUdxF4R/9JEf5+FHH+ETH/8Yo60R41HDtFEmdZMsWkR457veyjU33cbp7z3Kz3zk/fQKkAB/9aUvc/6xRzhqA7lz5HPBsCFXmwJQDMl6KZjXydly+dLFbjm7Y/kbYBKrl0XpN6ue2tDiMXgVEEdhk41ZEz17kxGrvTXUuQR3meRa3ZoOBjtowSfz2okLaNbSXxnQMyUuCto2TPIKGThsnhPwVAQcBlVPhsMZ21FHhLwogEhTjfBeCTHp5QJKGzwBxZMICcYYXJZEsK33RFWccyiGaKFpuhlvXYWYGehnBqOBfpFTGCVkDokWDS2TS7tMxfGlx5/m+WmNdmHkxbvSrOqcMev0pR40X3+nnpGI3EiCYL4OHF8KMOdJMB4s6Ua6NdOUXBaMlnUjzmVpnIJ2c0TallZLjlx3J0889iSf/vlP48cXWVk7RWgUMxhQTydoXVPv7VCbLF28raetG5rW0/opu3sXeeyp0zz0vTNsTQu29kAjCGnjkBihtQQZ430gyzIqr7TeJzeHaPFBKOtAnwrRnGk95Qt/+gXuuOs2Tt6wTlaW3HT3PcSo+P1Nrr/1TbQbt8LeU7TBs7J2iN5gg8n0gMz2+O7Tu/zS//LrvOfd7+SH3/dejp84POvwLD7OjkV4Oby2CE5CalJ+/+xUuTIGvfRjXqqWem2vGZQ5s5CaaUKiJhfuGZMj+ISBzoY0NgHa7m/bdsJoNJ4/5+bmJZ59dsxkOqaeNmxtnU6ZnKYsr24a6ks7lKo00TAwge2DfYKFYQiJIOADeZETO1JOmefUjSezjjwrCJ31irgsTantcPQscxRFkSaMCiQnQ4cPyQet7QgbpihobaTvCs5f2KZcKfidP/oSq8fXOXLdUbb3L3H6e0+g/Ac+8vF3MxwO+Z3f+3Mefuw5DJHbb72RO+56Bw/e/yAfeOftvOHQCoGGJ556nq/+0Z9yOIu4Titrljz/rCSKsom+I8X4153odXldpjOa/ywkKGmBks9jFoKPBh8dPhiMsRQmItYxamKaS+UsraRzoFYPLmN/3HKwXaF1i28DEoXVPKc0ho3VirLzu8tdTrmS0/eG/gqYrMU4C9YyDVO8FOSmpOiYSk0T5gMmvW+IUYkitKFlUlcEEQJKkMW17ZxLJB/pxqWITftt12dK+kdhWGYcXR+wNhym0SltYH86wbcNB9UBLi/426e2+NbZETGmfVuXjusyPf4lZTovs151MBKRIfD7wH+nqvvLT6qqKn/HYTjLupGy6Gld151uQTCx5cR1b2MUlLIouPutbyfsPMW0rqirlhg90cDeZA852GJr6wJN1SQDS+uY7k/Y3Nnku08+zQubu5yfCLFQjDRoWzHDo5TE349eiKHF2SF1t3lFTdz9OsCeNNjpFLD4EHjssUf5zoPf4fD6Pbi+Un3ld+Ghe1ndfIKLT97J2k/8M4JtaNvA2soGq+uHGI13mDQjijJj+8Dwxb/4Wx5+4Hu87e438673vJ2TJ49ibYfHq5tXR5epwcWT3PoSK2/RTzJXVD4daXyWiFxRZS3YLd1Gq0Ig8HpyGpsbmy71NZZ/njt5Lz1el7LgeUt1aTNK2aFBsWlsN2EOXc5U+iZaWnFsTmvWjyQDy72DihBTY1u9x46nnDysFIWjVxYMepbGR5J9iyUg5EUf4zonbSAETYxSVSIRl2dEk+GbwHgyQUQSfZduuuzaBiFf49zp57j/wWepulEWGIvYnM3th7n3a/cjCG3IsUa47tpjfOwTP8GFrW1uv+k4H3j7HZg4ZW+/4rO/8tvk9S5FIWQmuXOEmCjKMQYUTSw+jeBePJjv9bCuPFdgsXnO97P5xrqMLqQVMHiSoFONYk3yKcydSdN/NRFRiiKJjHcmFRd3x+xvjRk3nnGb5hVJVEpj6F3aTJCYCIXNGZSWU2t9ThzZYHU1Y2WjT7HqsLlDgaqqmcbUz8ns8jj5dG56DVR1g28DPnpclmGzDCMGay0iluAjVoQoCkZofejsi0wyVNXItYc3OLY6JHMRqxmxCKh1tFWgNcJD5/b40neeY9xmKQGb97IXx/Tljv8rrVcVjDqPsd8HflNV/6C7+8IMfhORk8Bmd/9MNzJby5qSl1xKgqvECBICZuUE/Y3jPPnQ17jhunfSG67jesc5/9wzrA/7TEZ7HOztMzrYYzzaZ2/ngHPnz7NXBRqXs3tpi93JiJ1xQ1b0aX3E1xOyPEd9Eiem7VdBA+qnRFrGI09ZDpKvmSqhM0yNRJJ1QEDEUE1rHrr/ft711jeRffdbHH3sizxR3I7hKfLH/4rnv7DG0Q98EoswGAzZOHyUc2efxXsPdaDIS1of2dza58tf+Sbf/PZ3uPttd/L+D/wQx44dSpUbEV2C6zoztkR6n1G3ZRFULnNZkIUd0LwjtRSI5reXHIpj9/+9HtbcafwlqsPZe5uPqueK93zZd5kfy9nmG0IgBE/UGatk2aIq/e8R2KsqPH0KmzGtW1QMw34fVdgb1eyPDqg2ehRFQb/skecp2ORZjliDcxniLLPPzvtI0BQEkDRhtBWYVBOqqk6jyV0a4LbRGyBFzlvueQ/3ffNbXHuk5JnznjaCMYqYNk2PDQaLwbqGY0cO8U9+5lNMmpbcej72wffitKZqhc9+5neZvPAs670GR55YqLM5UWLwUTE24LISCYlqrqpX3bW7Y2B+Ezijqj9xNc11r4SHlhmol2Xxcvk1s7yBelWmXpNBrBGscxS5xWtgWvlEEgiRomeZTiPjvcDe3oTt0ZSLTWQvenqZYVhmTA2cr/284h3YGh0r+dY+126OuOlwyU3HVzh8tGH12Dp5WWIsVCFgPEgUcEK0QrTJZy4lCEkn5LKCPE/VdzITMFTTluS2ntw5vMZEcIiBNkKjgbwwrB8e0Os7ptOIRostMgamZksLHjo75T88/CznK9tV9d3xuiLIv5Ko+OXWq2HTCYl+e1pV//elX830If8bL9aN/Nci8jsk4sLeK/eLuoSus7Tw0XP8+Bs5d+kM9f4ExHPm/AV++EN389UnH2PvwjbrayUaPVXdsDfyeLPCC/tnuObmO7jvga/T+oZJrUxiIO5t0yv7NM0YfJFmtEtHqVZLFIjqIbRU03HSB5gsZQ+zee6Lf5hVHRfPvcDW2ef55md+hY+st9y2tknV9+xVFVunv8bqTW+kOHoD6xsb3HTbrTz++PeI7biDIFucy5j6QOUj4zbyl3/9Le5/6HF+6L1v530/9A42VnrMTJJgdsHMMtUI1sw3ZFnakOcZ/ozv/xIf/mzkgnaZ7/xvv9/J8FpZejkhY3ldbnx6+e9eSkcye67YjZXQjiY+64nMglaKQ4sG7HC4wqHDx6kunQVp8W1L06aUZdI0nNva4cjqgCIfYGyDc1mSLcQEeYkmnVNS+CfYRCT50c0gyKppORiPUvPdzC7+BDWT93jLO+7GDlcpYs2J9RXObO8DDlXX5S6pMXr8yDF+9uc/TeUco4tn+akffy99CyHA5//wz3j0/vs47BoyASvJXQFVfNsSxaXrwATE2O55f2Bnyn8LnAZWu5//JVfJXHc5YVteLwcjLZ8rs+pVyZi0hhCAaDHWYp3ggksJTADEYvIMfzCldDnHNw5Ruh79cUPtPTccO8raSskoTmnbiqqqWN/Y4NylEcHlXLi4w1N7kySS3W9YPTfh8NEdTp5cZ22toOz1MTHJMWJIYy6sc/jgsQ5KyYl551YvaTTIour3CVUS8ChNiFQh0ig0CN5HLk0avvrgM7ztxhNslDmF1rS1ZaeOfP3slL968hyXaouKYDSSdplFZTQ7flcgZy95rK9cr6Yyeh/wT4GHReTb3X3/MykIfVZE/gvgWeBT3e/+lETrfoJE7f6FV/F/JKFWCPjeGmjOmae/w8aR64lhzFNPPc3t97yZez7xT/jGH3+eB77zIKHeZTIZUbU5Z7Yrbn/3B3nooUe4dHGTwaBAW8X6hhhbqkmkKAvaZoKzvcVJaSyCptkvMSKhpZ7uk/XXkTg7gCY1wZmV+ekiv3ThPI889AjT3oBHnznHO+4aMNl4C/nBBTbHkXCx4o5TOdZZyrLPYGWd8W6DIxJ8RDXMq5fCGAKGza0Rf/JnX+HB+x/hIx9+P3fddSduBuAvcWlnePbiRne/akf/1vmfLLC6K5Yu31w+nV4f68qg83LQwPJjZreX77/ycUkLIfPkaLFmPPt0Hlxz/c2sHlqj2dmkEKVtPNNphWQF1lqe3x6z0t/H5X1WVej3QXxAYpMG8sVUYYlIylS7z05NIjUEYzgYT5hUDdEY8iyNQDEm0qphsHaM/qHjvOUd7+b+v/5zjq/2CaFha1TTqsEai7PCiZMn+Lmf+zmq4Dl//lk+9WPv42g/DT/78n/8K778J3/MqvFkJkFHuRWSe2Oipc/gzQXDLPXoZmMJrtYSkWuBjwP/K/Dfd0nwVTPXJb38Ky+ZV8zm5YokVNVQtdD42JGPFuLZkMavJUKDT2akWa70csNgpeBNwz7D6FjLemgw7AUleoOXEnMg3D7YIOv3qPprqddDQ+1rRiFy8aCi5RLXxTWO9xLrktZjA6gGbCQRHQxgkot8lmeEGOeMzKqqCW0aKyGaxLxtiNQ+0ESlaZRLTeDRF7aY1srXnt/ltiOrHBn00WB4cmePx3drJgFEw1xo702WoM1XURVdebyvXK+GTfc3V3x+y+vDL/F4BX7x+z3vlStZUkSK/mGiWWX18LX08sNI8OxdOs9jjzzD7W+6jbs+9mmuuf0uHvjyX/D8dx5hZf0od99xE9/49tc5/ei3iRjGk4YsS865MxfjEALGutS46y78KCAqWCxtSBBHM9nBFT3UCEKaejgLTGk8QUsMyualTUZ727zxA/8Z9UNf4MyF0/SvW08TO0eOS7sjcAVl0aMwjpXVdaYHO+n1xFTOp+e0aEwbUBTQNvD8mUv89uf+jAcf/i4/+iP3cNMN16NimQ2DE9UkJhNJI7R1Wfg6m+/TZd0vA70lCyGDaiQSsCpJ8/Q6WEo6V2ReJtC9z6XJpcpcrDgPz0ubKixl+DrTjHfQZ+I+dD8vhaHu8Igq115/PcN+zoWgFHmGhkgb07wZJ4Y6Or537iIms1wngpc0FkIwNAi29UDyFBSTyA4qilgBZ5n4yMWdXZrWY5xBUQprWemVUAzoHzqFYvnA+3+Ue//yi0RpuObIgMIJmwcNGOG2W2/hH3/6Z9k6GHFp8wyf/tB7OLZmCApf/cp9/OFv/RarNBQCuTMUNgk3DaT+gbWJ9ks6hqZz7YbZhNyr2jf6P4D/EVjpfj7MVTTXTQjgHICj+9sXvYjlqnkBc1uiCA5ogmcUYLVJgSZrQFXwCE0MeF8znULdSpoRpB5jBRMMvWGPjfVDODFkB/uIWSXvpb1mYAyNb2lixFgl0iKlJe8nYkPIBVWPy4SQGVzIoA0dw0/TPAJj0aiUvRwljXGwuMQcjhmCJ4YAMXSjT2wySPUNGoSze1P26oBIzpkJbD6/h7UHqVAQO0+eVcxSu2CRFH4/I9Tvlyu8NhwYFHxIjbSBWydKxvqx25keHGB7gd2dSzz/xHNY22f92HFWj9/Bj/zcLZz89rf53qMP8Y0H7+P0Iw/gjNB4AVlcKLNMqGkastwuegmqCZNgNmW249yHGj/ZwfZXU2YUu2YfkPCbmGzZQ+TQ4Q3Wjh2Cd36EZz//GNfFNVwYcO0bbmV3/TAqjuHaGrfd/kZeuHCRzfMvoDFgzOWN05nuRyLJE0+Vg3HNfQ8+zgvnNvkX/+znObSximrOfMNdzvT18gb+DNNPw+Iu9xG7vJJYymb4u5ka/v+7uvcSFz2v+W90EUxguXfUwS8L3s/85FC1qAaENB7B2JSEsDTdFEkNbEdgkFlOHVln2FsDcRAjzma0PqAaqDsoeLfy3P/kGfbqwA0nGwZFErSayYSyLCmKLJFumpBqU2sRA5VvGdXJwgURsryHFY+JAYMgaydgJWnQb3vj7axtHGHqtxmUGSePrFP2Kk7ddDOf+Mc/y5MvnOXg4IBPffzH2OiN0SB89SsP87u/+hkGvqJfOgon9KziLGRLzXxLpz/TxDLTmDRpSZn+/WGXV7tE5CeATVX9loj86FV5UmCZJOVsrktpyUu+9st7RDLvtSwndEGFqYcqCJM2kQGUiFelVmEahczHJBL2NZkVfGg50EBeQ76nFNZ1msGMppkCgT3jcBlkZSDrOVxRIE5wuaHXzwhWaUWTOFoMYpUm1oQQaIIny5M3nSsz8qIkqEkOL6TeUKqF0jDRoJKG6iWpIT4K42DY2k8kLWsyxDi8iXgUZInGv8DjZgftRb21K4/v1YTpfuBLSZTutm1pmoj1E9SUaNHQasvWpXM8/+R3idpyzbiiOnQI5zxnzu/QXz2K9Hp4EUIzxTUVIUtBx1qbGtNtwORFojVelhUtDo6zQhPBZQY/2aEo+mA64ZbtBKZqZhw0QvCMmpZe0TKdbnNjcZ6bq222+oajN72Z4vB1HIxrbG4p+z1OnroWl6XhgMaSMhTpgkPXeBTSxNjUr0owwPPnd3n4kaf44fe/IwlctROSLTdgl/BwnW+wXT9kGaZbCkQv4vzPcb/X/rLGkDthY2OVvd0K7wMrqwPGowMOHTpK27YcHBzQxBarOdYJqMcH3/UnLcF3FwiS6CmqWLFkJiMYT0uCumL082NlRLEow36PYa9Hf3WDvL8K9Q7OWlyEOgQCnd+fMWxXgcnT59kdTTiy2k9O74MBTGpUQ9fATsmRzXKCy4mmEyAa6JcFeV7gvJJl0GY9No7fSGtLrIkM11Z4w61v5PRD9zIsLRpbjg6Eo3nD/d/8Or1DR/mnP/l+cqmI3vGVL3+L3/v1X6evDQMLPaOUFnKbKn9jk6/arKqExWbiW4/VgJlPzb1q633AJ0TkY0BJ6hn9a66muS6LRO3lzvOX21Bn1aCQEpJxHan7jkkTErfVpg1/3AZGTWQgqbp0ohRZ6iuFJnB+OubCZIS2Aa+KxVBmyuqwoLc24FB/jd4gJSlZ7sgym/RAKuBbtMjIjMVrGg/uQ6DtKudEcFGMswTSd4symUxoQ6D2qVcUAR+hDdB288zaANvThmlQnMuS3Zo4RP1iX2FZwPpK8OaVQOirX6+ZYDRtKxpfU9f79IJBaXCux97uFlo0XDhfYMsCDY693T0mo23Ob57DuYaH7/sbJjtncDajIUAbEKGjys7IEd1F5GSRPceFwgexWHVEUZrYJneHok+0LVZmG3uXJXXQ2HR/B2cGyLP3YTMYDa7nqf0R1WiPY7esMp5OqBvBSI9Dhw9TlivUvkatojHgY4sT29Gw06muHVwmqumkNRlPPvMC73/f2zG6yPIVj5Kli6traEKCDLpBNi/OTmJHaoipakrPMTvZ4uskFMHGxgZrq33++X/1C/zf/9e/ZTye8Iv/4j/nM5/5FT760Q/z7DPPMp6MybKcv/7yfdz5pttR9Tz33Bk+8pGPsL9/QJY5vvrVr/LhD38Il6XJsNWk4fN/8ie85z1vZzqZ0jQNX/v6fRw9epSVlSGnT5+myCyDXkGeGbLekP7aIZqLe8nZwCZWYhtI3nIGJMsY14HvntklvzhiUG7TK0ucc1iXzEqzLEvq+MyRFzn9YZ/cOcqyoN8rKaXFogTXh8M3YI7cjMdS0oAx3HzLrTx0/70py3WCFceZFy7AuS1uvvl6vv6lMTfcfD2PPPQkf/Z7f8BAKwZOGBpL3ym5CTixnRlq1yeVNFcstmlshIjQti25m/XV0mDCq7FU9ZeAXwLoKqP/QVV/XkQ+x1U01+2en5nr+GLPXMCNCXWSBQTcEZxUIBqD8cLEK20LtRGsRLKYILm28dR1TKJRKxRFRnSpkFzZKLpejaHxQt0quXNsDHNWSmEwKOn1euRZBkScTfCo9y0SDa2PGCw+NabwvqaJLU1XLYNC22KNAQ1EX9N6n8aLR0kJkjGo6WZSWen0RUoIwsXJJDEEjUnPQReAEotnBt4mBOdF42o6ZEfkst+lyckxtQOY+QT9PXpG/xArxqRhCCFQj7aIaw2UDiuOSiMH+zv4KETJmY4resMVXjjzDL088s17v8Bk9yzGgEiWmsK6mOoZgk8VSUjq4hlsk/Bvs1R2JhgiqiI2o54ekBUrCIaoKRtfzqw0JnFsjwl7o4vkKwOm1SUm29s0x99DYTJ6R46xs79HUFhbHXD40GHOjHdAPWIEQxISZi4nasRonLO1Zky54Fuef+EF2ralKPJ56OCy70trmZiQWmPzSujy4qcLw/OKKnI1pa8/SJru/v4eo4MpvmkQa8hchqFhPGpo6ykH+/ughk9+8scZH+yxs7dLXTd84pMf5YEHHuSBBx7mLXe9BRX4whe+yPs/8E62trZ47pmLfOyjH6INhs/82r/jUz/1UY4e+iD3fvMh/vl/+Wn+z3/z/7B1YYfVwRArjrIo6R+9Btl9Du/bpDMylti2+AiCQwRckSX7fh+oDhrs1NPrFayuDRkWfTRz3WbfYEOkEEvhMoqiIJekulfnGBWHuObmdxCyXgpOUmA0cOzYcYScOgils7QhVdmZeJ59/EmeeewJxCQniY2VHhItuYnk+OQ/ZlIvy0oSRSopIKkEGlUCgmicJ02KdM4hP3Ct0f/E1TTXvbKn8ZJJ/KIa0mXSCkKabxSpfGCWGI+VAAAgAElEQVQSlVKTwejMDLQOnqr1ZAK9XobLHENR+lnGSq9kpcwYDgaIyQgKxiXigURN9k8u7TEhBKKGzvA0ouqRTstEm1wVqjowaVpijGTREOoaZ4W6jYTQYoOj9YGmCqi1eE02RWEJGQoxUEVlvxV2py3W5BiTsXB76b7LLFnmJQLR7HEvtXd0e0xHdkh+RS+/XhMGU0lnkUrC5uA8bb2F9xUaPVlZUNdjRvtbbG8+w9OPP8Rjj36L3UtPct9X/ojp7vOEkLD1RM8lQVxzzUjysstcgTHZvE8jIknX1N22xpHnJdY4nHW0zRjfjjGaRKYaDWhq3M0U/820QicjrvcPcWNes1rv0W926PccGOiVK2ysH0a1op87jh0/jhoHXV/CWrfoG3VBUpgF0RlbSbl48eJl47NfxRF9UY/o8iC1+Dk1JWe/uarQy4ymO1szmu4twA6JngtLNF3gX3WPe8V16NAGxirWKW+45Q3cdtsbKcsiCfqsxbdpDMJv/Obv8DM//dPc+aY3I2I4dc0pXnjhedbX1zhx4gghRi5u72Cz5A323PMv8Obb38CFs88TsVzc2eFtd99OpoH7vvI1bNNgVBlurOGJiESOnLqGUAywWYFzCT4snVBYIRchk+RjVjphtVewsTLk8NoKRzdWObaxwZH1dTZWVsiNoVfkHF5fo8gzenlOLgbf1PimYV+GrN16D2bl2Pw4RAwxCv3BkCzv0XYb42wcfVBJ5pdR8G2kbQPeOEKxStvbYFKuMyrWqPIVQpYjziCZELWGOMVFgSDEmDRF84SMBDO/ymLk77RU9cuq+hPd7adU9R5VvUVVf1ZV6+7+qvv5lu73T/1d/5/5HrB0+0WPWQpMBoNRTR6DCHttwKvHB8X7hLQ0GqhCS93UVHWNMUKRCZlTMhvBQqMtQSssFc43WG1xVhH1ED2hrRNNabYnokQJRAk0oaZtKoKPNEGogtISab1HQ+oBTZuWuvVM2oYqtERnEjtPFTWJ9BRj6s97H5hG5VKltFpgJMeQI/MgbLuv1KcyxswlCJd/aAaZP3bxu2WYM/XeDGka8Uuv10QwmkVQ1Ujb7DPZO0uoK0KscHlONAXT6ZSqmTKp9zjz7CM8ffrr1KOLybTPODSC+oCZM2A06Yc0Enya8okRxJo071tcynwk/SwmqZStdVjrMBia6biDr5KRoIoBaxGTmpZ7+yPC1vM8duSn2HanuBSPEU3GQe8QRWkwztLvD+n1+oiBw0eOJH1AjIgo1kgSQc4IFSEke46QnMc1RojKZFpz9vwWic9j5lVf+gqX94zonAY0MNMYzI7xcr8ozkkMiymvVysULdF0f7n7WUg03d/rHvJrwE91tz/Z/Uz3+w/L92lIuCyjKHqowre++S0efPBB2jbpc8qyAMA4y2NPPcdv/fbn+MD73kdTTTl39iwnT57EtzXvf8+7mExHBBGcc7RtcjvevLRF2R9gjXDixAmef/Y5wmiXr3/lK2zv7hJjZH1tDUFpmgmnbrgJX66DS8amhUnVSWmFwgi5KLkEcomUFkoHK2XGWr9gWOQM84xMlczAxtqQYb/EGcEYhbrC+4aLvkCueSu9Ezd3G6jMCSdNiKhxlP1h8igLiqohxOSBFzBEsSgZdTSMgjKuPbtTz4Uqcq4xbIaC87HPWd9jux4waldodEhFTkPql84GEM+nC79+2C7ztUzcSddJnH+nG8ue2JSzdv+yO8Ps/SbI6aCuqUNMY25i7LRHClYBw7RRmsqjYghR8VGRYLCNwVRKHHti2+mEsiRmnh1S1Uj0AQ0eISEWIYYOPfK0bUVoa0Lw1I2nadpu7pvStBUhKtPQMmorGg14EtU8hkjdBKZ1S+Vr2hiY+sDWwQSxObab6EsH2S94G1cktlzRJ5J0HBeMIEMSi5uOxQhJyJ1hJHvZz+c1EYxSE7/bhIOyc+E0062nCJMxIiUrR2/G9NbY293m3PPf42DrabQdpZk1xmHEJO1OVDSE+ckVQ5vgOBW8bzvbqS4AiUW7L8R2WiLpLDQKXNajqSuitog1SXUtSbmcTAkj57a2aHYvcOf4QY40z5H5A1azhsH+00nkaAWXl6xtHMXlOSurfayz3Xx45psKmj4Io5qIShpJ6sLkFoxYHv3uEymrUENyZmB+MS2gtqWLrWPozQKVwmXBaM64604gubxE+vuuGU13Vsa9apouMKPpvuy6cH6T1ivffvBx8rxP0evzyKNPU5QDHnz4MfZGFTYrKMoVHnjwO/zqr/0mewdTPvf7/y83v+E2Pvxj/4i/ufcbjMc1Zd5jMOhz6PAREOE3Pvd5Vg8f5Sc/+mHWyoy//o9/gY0tk3EKXMFYVoZrgKWta/Jej1vv/iFGtSfPIuIsmcvIXU6eOcrM0HOG0gqlhX7uWOmVrA4G9PsFmQMNFb3cstLvUViXHJPbhtFoh1EjcPzNHLn1bRg76zukEyaSfOuMQH+40s1WTJ+pj+BDp90LngAdkyrBM14jLVBHZa+JbDbw7FT47tRwus74btvnaZ+xIznkJYiQyOgR0wU8fUnI5nWwpCODLTmQ0FWTMvsdymUpkXQ9FE29j6o27FUkM1IiISaHlBATwbIlUPuW2EY0CKGJtE3Ex0RUwTlMWWB7PWyvpBwMcXmBdW7exxZrCDF0KEmn+wIa36ZEG0k9ny5Q1U1NXdfUTRqsJ8bifSI6+BAIbWDapAF7U+85aAOXJmne0WI4TUBlRtdeDjIv3hukO1gLScQsOQ7Mgvs8GJGqSox/0fPM1muiZwSpGZo2SQF/wNa5hwjNmNVjN5OXK+TDdZrpDjT7ECpCt3EaYzo2WufgEAJipNvM04VrrCUqc4PAuV1OEMTN9DsmNf1Nyn0kK5DgaSYjeqv9eZakcdZ/EfYubSEH5zhhLkIMtFlGmRcUW09iU06KGks5WOPQxoS19SGD/pD9adX1qGZ7dXruZNG/CB6z9xdCy8WLFwkhdBYsi2CSyBg674VZYxYnx9JK1dLlXaFl+vtC2Pj3Wz8omu6yZiTLMra2d/i9P/j8/DX/xm/9ASLC2fPJlWpBN7V868FH5xvL7//BH7OAMS11O+VX/u3vIgg+Rna39vj3n/0sfRu5rxoRfLL0j+JALVhHMRgSVfBty2Q65bo33smzj3+HyYXTrJQ9fAtq0uYUo86b0QCZS6PRM5dTlgVtUyF4MuPIHWhs0LZhOjpAXI675hZuuvMeXFmCWWDzEUBj6jMBw9V1Ll0waSOQxMIyogixU8qn0QfGOJomDWMzXTBJzE0AxZBYpSNvcGIwKti1I1TOcNZ7DolnKIpKi5XXpz/dMoydUsJZcqjd5vtKf9uBGOrYrWCltGS0GDzOCAdTpbSBnkSmwVJEgIiRSJYFshjxGnBFjitysAY6BlzbJblN8IQYsM4R2jCXpqQeEsm6JyQpCiaJXQOpMopqCLFFSWiR94l8EXySi4xaz6RJTu570fLCQcDjuvd+RfRlibG7ZK81W6o6N1+dS1SWqqg5qzfdk75eYY95zQSj2YuOJkB0SPDs7zxG1ANWN95AVvZxvRWa6RYhWJBUScQYUqMek1g/dK7NsWV20omkimfGRBMx3YXYfZ81KxOAjMsMIfeEZkozOaC3cgQxGVdmCNs724y3Nzlz9B2YyRkuZMfp7Y8ZTRtox4R2iHUR4woOHT7MyVMnWVtdY7S3D6HFWEFDB5XNMGybYLg0JEsxkjKjCxc2mVYVw0HyP2PpwpClRuzMTSHdpR1HYRZAl0Wfi6bRTHsRr0Iw4gdE013WjPR6fV00UrvXfkUvbfnCSYF++c0t/lZEUS+IekptsdFjYktsGqL3SY/RNe2VJEotyzKJVDUyGo3J+ivc8d4P8bU/PEPP12Q2Vb/Bp4mbajqX5O4lzPovmbGpug+KzSCzQlXV7O/uMJ16Nm6+g2vf+SFsfzWNmSB2UEo3kyYEqmmDiGM4WO1eJ2lGrabPM6UmmnQt3TFJCU5S48+A2tn1Z7tDakhzbpLIUTholEu1oWeEQRbJ7JBe5kjtv9fHWt5L0yUkV2yOwmwTvnJdCUtGIqPasjeB9Z7FSotEOKg8hYGsTLqzqBNWewW1tkzamlPZYdpRTa/IcT7NIjJZ2oPqpk4CWWPnPe82tDR1Q920KbkRSxMiPqQ+UNBIJjZVUAgxJKZt2hcl+S0q1E1gUrdUIVC1nkkT2Z4qrR3QHwpEpW2TGD9oQERx1pFlGVGTa0wIYZ7gg5I5R7/Xx+UO65KMJgYIbRrs17RNN56+I8XEVw43r4lgtIxFph2x7QgJjoP9SwRf0+8fpRweQtZOMT3Yw9djQjsF0kkVZs9hAPVAi1EDkmFcjssLxGaIFKl0NIJVSQ68AiqJ0mg0Igh5VuLzHD/do612KVauSdogtIPKoBrvYkabHFmdIOzzpuoC+0VFe/AYk+0L9NZOEes9XHYU0+tz7PiNXH/Njbxw9hwmGtAmMU3oNqsZli2pOa2iKAm+vLC5zbnNTW658ZrkkqvZvOejsYP7utaTiqBGO7iyYwB2sB0dvId6zCxIJbXtMkj89/ks/0FouiFcnnHNKjtjZkH6yqdQEhNsQTEViRjxZDrFxojtXAV88Kh21vrIzBwHSIHEZS5BvaRm8MH+HseuuZ433/Mhnvjan3FsIGTGUHs/n5CamI2GEGKCjEk9Qw2BtmkYlH1Ca9m6uM/Fi1us3/pmrvngf0IxOIR07hvWpDxexUCMtI2nbTzW5fT6A3yI+GgoDCnIprecjlDUhBgA1rqkv/Oh8zWU+fYbYN6Pko7anFCBNGJgTM4keGyjmOZ11jeaSzQWmfyyOwckGPIlyQxXyCRELBocW6MaWzhWTUaQgDeB7dqgtAydMh0FNAYG/ZxaG0ahoQTixOOyFptbaBQrDh9DcmLxHiMCztCGNs2yaiNg8d7TxkAdPI0PiAi+Tc7f2KwjltCZ/YJXoQ2B0bRm0iax/qSN7EcD5SrX9nsd0StQVzWT8ZhRNcEYoV/2KMsMJdLUyng8SclVjGTOsjZcYWNtjd6wwGTp9XsfiLWnqmt2D8aMxxXe169Ycc7WayIYwYs3D2stvV6PSdUwOdijGo8o9jZZ27gR7a0w8lM0LJp+6CzTSbRrRIg4rOtj8gGSDYiSYUwBLo2HIOq8GqITKib4TBBXkhUDWt9QjXborx5HjRADcyaejYGBes72jrHWOPrTs4i0nMmOcenhe7nhhruIxtJUe2S9FQYra9x59zv5xgMPENtJavTNePyqaZPoCAiWiJfEqmt9TdSWZ555hltuODUrhJegqGVYSuYHZF4Ezb51UN1lwtfZ3xK7scY/sHXVaLoptsYr7knfZwXSi/aTjmk5m03koqcwgUwDEiMh+o6FOdvB06jpDgzp8uWAcZ1FjrHzBEbbltHumNvf+V6ag0tsP/4wZR7o5TZpQ0JMmxemO8aKsUrrK0aTA+q6pg2Rixcv8sKF8/SOX89dH/xpspWTeGMpreKMJJjOdKZQAfABYyI4hytyRAONV5wYnO1qYxUUCzEkWF8MaMSITe8tpawvoq+ophHS6fIwEE1yIokB6XzQrir38h9qySzYzq6i9O+CpGBYnE8vXnONIi1qkhPDxb0Ws96j7yxIoBLlUlUzzSzOWSYHFUeDsL5Scml/xGqeY0PEuhbbWKKJEF2qYlCMUYyBJihV46m6KklRfEhwvveJyWetJbddumTjHAWIMeCj0gSlaT3T1jNtlWmITKLiyxXWV47gXA/fekLwNE3D7l6G7qY5W6vDPmWZIQamVUOIabKBiKHIc1ZWe2xsDOgP+tgso4kpOaIJtCExoUV22d8P3b75yh/NayYYXbmMMWxsDOnXU3a2a6y1GALjg6cwrmRlMGRiLa2v09hsTbimb5sOmjBgS7LeKlm5BlmJmBxrXLJ7EUGCT5yZrjISTb0mMRbxliwfkLc1zXiLeryL7a3NCk6MCC2e3YMJsVpHgyfka7TZBc61hzCPfI3jb/8ogxvekgwPRweIW+OaG2+lP1hlPN0DDPPhvJoaplZImigSgSHE0MXNwGOPPc6HPvCeuTPEyzd5ZhvqzK2hCzjKZQHoMk31bMrsVVyq+mXSBGA6+u09L/GYCvjZv+MTv+w8o0X/7XJIVdUgGrCmwWlLQcSpot05sPR6lv4f5v9HgjNTcJlR/JF0njpnCb5mb5Jx5wc/yelyg93Tf0VmhZhbqBNcIbZzUQaaJs0fauqWpvFsb+0wPdgh3zjJWz/2C8iRNxAk6wbt6fx1LCGyIJIcE5xLIlpjuv5PRDoW3PJ70iUoc0ZXhlQxLej+i3NgXnyG2AWyrmIKyY3i9TZ6fPbRvhx7dLmLNLtj2XVgRjlK54IjSnJRmXjhwn7N0RVHP0uu7JGMaaOYOmIyQ7U3YRyhX2TsZ5CpotqQlRliIcaGpm2TX6YowUeqRqnaSDSKden+xisERUIEY1PgmiUJ1swTyhhDMkJtA40P1CHSqKHxirqC1fXj9IZr2CzH+0DTNNRNBR0q472yMigoexnWQF44Gt8ynTbEoORFwWBQMhjmrA4HuDyjahvGUuGyPCV8ZYEqtK1nPJ6QRP0vH+hfM8FobmDaLecch9ZW8I1lZ28KohgLRW7JXLK076+vUMcNqgBtO8XvbRK1JlqHBHDlELdyhKy3miiF4i7nyduIiQE0JFWygGrHcVOD0Yw86xGykupgk9XiCEHCDLAhROH8pQl3bn2TrAw00TH2Y94RHuS5acX5h+7l1hO30BsM8PtjppMtcpfR669yYC0SOojJQIyC2AyiJ80p9IimgVhFP6dpas6eP0/jPUXW0ZeRFIRRZuJWkdgZniYnh1RFLFVRmgKbskRnJaKBpflGr591maD3SseJObgGGTWZtLjQYjTNpVKSFqf7i0QggcuyZplBdJKMc5NWzWGsm6OaxlhcZhBaJo3hze//Ec5srPPsA1+hbPZwqskd2beIGNqmYTIeM5kI+/tjpqMx1k4YXnMrb/3of8rg2HWIzchMMlI1sqTv6BJ41dgNd3MYazpsPul/ojXEaBZBZuk4LK85bXmJNLMcsLQ7RjPwYGakm+qHZabU62R1Zrp0IzJ0cThhGWXoHi5XBKLZPdL9KF0wj8CothAbjg0VsgKbWYwTnAjOGizCfqvUooyDJzcWG0naIhTvmyRPwSaILXYwWwhEk1wcjCqtJrlHYWyHrC9GRFgXyY2HIDQxGfcGFSov+GgRwAQlG5QMVw8xGA4ohwNUhfFoxHg8TsScoLQ+MOhnlHlObh2Zb9mfBNRMQD0uc/R6/WQE3etRuAxsEv7aYMjFkbsWXfeMxvtM6lEKoK9Qdb5mgtFyIErZpsPajEkTMJIEgb28oMiTRb+zStA2zeXUksr02AlVqn4QJF8jX7+JbO0EedEnBMUIWF1kgml2jUdDm/pMHaUTMQkj1xZb9MjCGtODHTRMEi1TFW0VHw3nxi2NGbJOzaQYMlbBVCdp7B47Z57imjAmtkPy4ToHly5RTw7oleUCHpjVWl02ryESY+is+03qgYhQFCUHByOm0yllVs43zRiTfmoRbPSyBPeyftysmOre+2wY17xqukoMhn+ItfyeIO0v0vXLUlUkZAScehzJ0inptmI3GyhBcrNtOjGRuqabdI7LpK0mdjCuaKQoc8Q5sMkheRYsrCvI8gKMowmOU29/D2snTvHkvX/O9MLT9GWKNUqQQIyWZlIzaRTqBjGWjRtu460//ins8VvAOiwBi8OK5UrVe+pDBTKbeloEQaMntwYnjhgNdedlNx9y0MFToouKWjp2aZfOdO92Xg6mxy9i9WW4nI+vz+Rl+U3MRa3dP8vX0Oxhi97S5X+7qFZTkuDygv5gwKjaogkjfJvRs44sdxQmx2ZZclhwBozDA20HqwVVxPaSa7/OT0HomMHSTcCe2QXGCE0XlJg5qashNpLOL694AkGVSNKcCQIhJv/F3gq9YZ+1tTXWDh9GxHDp0iVCCEyrmqzTPZVlQb8sKbMc0zSURRry6IMHSRrJIk9fmctTv4oMW6Tga0JBv5+kE3Z3j+BfGXl5zQWj2eZpjGH/YMxo0tLvDRkWOZnrKhsF3zWava+Tj1x5hP7hN6FhSnWwT+/QzZjBKcr+EXKX40ODSKfjIYnIIELwKA0aG5DYNXotQQNqW1Q9Ll/FZVOq6Salu56gbbp8Q2C7aXiyvJ7MbnM+vxmdPsPYrVCZwMF4ShMcUu1BMWSweohR5SnKYn6Sxxg6RWH3mromegghzaiPneO4c9R1zXg8Zm24dlkAmgtZYzctV1LmrHEJZNBE353Dc7ELZDFpJGZDB19Pa7ZJKGY+NN2pJ4stuQkUWgNCUFLWqWkybLwMKlBCWFjbzPKRpfQYEFRS47/sD1JCYmxq8HcZsyAY67BZjjE5DYb+tW/k7o9fw6P3fpGLT3wTpntJXI0nNJ7cN9RWOPrGd3HXj/wk2cYpTJaTO8E5M4feZGljTDeSb9ysMosx9Qwya3HGdueEx8+lAClIi75c8JD5cVkQLrQjL6R3GDR5o82OiSL49up40/1DreVrZdl14UroaHGcZ2iJzkuoOTW8e6yqcnR9hTvvuh0rnu8+uEesJhw0LZXLyYMhq2ty18FdLlk95VmGzSxZt5+lhCMRRehMe0SEzKYR5kGVYJVMTaJxC6QEyXZJg6Capld7SUQdJSUNMUacJJaeRkO5coi8l9Pv9+n1eoDQ6/XolT32zahLShNtuywK8iynBYztxqDMv0iJfQzUbWBv2jKeBop1g7WgrSXPsjS/yzkC1St+Pq+pYDRrwDubOPK7uyMy51gdlGRGsB0eGkKg8TGxh3zAa53s74tTjCcv4IpDZMOjqCtwLsNleTeLSEljhBRD8oFT8WAs0RtQn8pewFlD1IIQPSa3FKuHqEZb9FbpxIURTM6kEa7be5TDazlDv0fNLmdlm9rtwgRcVBprkOCZTCtW1tY5cvwoj31nqVOhCxAgSmqa+xAQG5MJYky4MwT29vc4eexkx3tYiFyZm6jODA2ZlUEzHKL7vlCew7I4dmYj8/pYcymACkLABY8zkdJEnA0Y9Uj0nfuA6UaUz6qEywP4i2i7qvOAtdw/CCrkZR+MI8tzYjfZVVW75CYFLOMcpc2IWGT1CHd95FNs3nYH3/vGlxideZJSfBqCtrrBbW+5h1N3/zAyOEae98gyg7OpIp7Zrxiz2DjFGHQGyenC9irLO39D47r3mM6X2Ua6HIhmVfHsOLzc6PbAgqEYlPkAwK4sZwnRe+0vgU7ZPb9rQUaYQd2CLhNWLusZzjZ8ZpglKKwNC979jjvY3NrimScfpx9r8qygjco0eJpayE2kbQLWCLW1jKXBipA5oXAZpcvIup7QkuogjQtRTUpaugCjgYguPhuVxQguE6Cb8BxjonVHjThjybpnCHlGb7CClTQgJPqIb1skaqpwijwlb7MpsjZLgxpdSZ6XFEU5R1+0Oye9RtromTYtvvVkdoDNlDaENBFWul5rV9G/3HrNBKPEaXc4kzHoDyh7ZQeBLISqyWW281WKSRiGpvETvh6DqzFYXK9PBFyWyAppLzbdht+RVyXpjWLXi0oMpTQ8jw62sjYH59NUFzPE1FPqegSuBAMOz5N6gkc2fpS+eQzX7+HqlnO9dxHli0y0YNzUFMMVqjomq45pS5bnAISQlM8Sdd4PUAI+tLS+TcFX0zGIQZHMcu7seW6/9Y2LE6Jjwi0aq8tLX0I8dKULQwf3oUR9ffQA5u9bAk6EnEjPeDIJGElvOagQpAAxqY/C/9fem0fbdlXnnb+11m7PObe/r2/VIIEAEUDCssExjRvAbdnY4MRO4vIoJ6OcGslIMmI79Udq1KiMckY8bFOV2C7KdmKniLEDbihsYygwAdMKCYwE6qUn9Prm9qfZzVqz/lhrn+a++x4CPem9i8+n8XTv3eecfefZd58115zzm99UaBWFpWiUZwdgjADh8GkT50AZ4xlkw+58TZrmmCghinNqKYOPVxPnU0r7zVOsERUhKufgHXexcPAYX/ncJ6m3LpGlGUduvYnlY7dhswWUTolUjI40KmbMCTG8V0c87UZEt0acnzoamwRxMuwfQjHc3CnnO+MZ66lrXC00DqmpgIATgx2qgAS1hVCbNCqEC0oNJYJ2CyYYmE1abpgoZ8IDNZHJaJO8rSk8OKVDB/bwile8mLOX1jh35hnUVo9Mp8waTRULa4MBZVVSG01kNVFVh/UMTAV9VZJEEYkZtSWg/VpglMYIGHF+U6W8Pp6opiXF4VRNrHw9SJTyO2VrQ1SlfKMzECFYJ6g4BmPA+oipLkrP5ixKFJBmKWhD0S98HRtP0GmlKUvLc3S3Ntjc0MPsDng74zRluSP0lEI5hdTe8VRVRVXWQQ095NKvgBvCGRlj2LO0RBxFQ6FSY0ZsoIbKOL6b9d3ItnkCUpfoVFBxDnGOiCOOErSJQ2jpezNqOy5HEejUzqCMxlEgtYAOqUIVE6kcqxSutkRJRtnr0lpoUZShG75YYWnjYdJ2Qd8eJIrP0irOcUGlbBae5BDplD4DOrN72NjYZG11C6XCWOCw6/JzbaCq+34xcX6hEe2dqLUVSM7JZ86F8FgPc/7jXyHUApQKA7y2FfUDa9AXxRVuWDfQKH1D3A5fExpomZrIVUQGjAqRLoIVT7+3TnDKj9iwDrSOws5S/Aej0e1rSAFN2hQ17PFSId0xfIpWtFottI4wJkUiT7PVxl83H2U5Iq0xUeQj8kgjUnPixEk+9BcfQqqa+dkZNi+u8pVz9/Md8Ty33b6MSWKcTvxE32gkRdOUMJqMoDi/+VLN58A5v/ts/nZBlsWJQ6wlDhRsPX5vjH22mjSchPS1E6isX/AMXlne3yvNZ2Jkk9pl3mh7GnqShXk5xh/zDsuvHapJzSrF8tISrVaHu44d46ZjB7n3U3/F6UeeRHW7tDUknZSNQbfRL8wAACAASURBVEG3stRi/V1qQ5pQe+mnqLZEkfKUfAn+BE+wMFpjnEMrhdE+bRcpr1SvIg3K+SGI2k/wbTalptmMGE/lVyJIFCFpSmm9jJSrBtRxjHOOsiwpy4KqqsJGw4s3C17BJktgfjalu9BG4zBJ6qXNwu9MlGI+j1G1QeqKuoKqrimDaGxd2+ZCXhE3xOpjrd06c/7cI8/tLGtwYVK8d/25nXAnLAMXeyujA5+9BD/8G1cx/S9eec1++Spw8qn7eN/7/tPXeuoy20Ywfx049g2+7gWDUY4OZWhQDlFdSDNKE6k0UYDzH1Q9tvBM0OJFdvx8DGnM485cKVqzM6EYHUFIdw2jGEYLfrNBWFm5xF/+5ce493NfYmuzpPI5Mxw1Shxf+Mp/4s47XsSb3/I93PaSl/ix4zr2CssNzcWNahtD4om1vtaF+AVHK7TBC6xifS2yiZ6RYYbhsv4yv6WmRlNaYWNjk5lOyzuv8TQVDDeBfuOjr7qQ37iYZGLs5KAue4X469dk6PxZ/LHZ2VmUSjA65ZZbbufI8Vt49CsP8Oi9n+Ps008QlZbM5Gz1atbKIFwqzkcdDk9+sQ7tFInSJKJRKP8cnFdkUNBOY/I09nO0tB8P7z2OQYdBkLULEZEOjfyhjcA07EExVMqw1StwrsYohxONs45ut0uv32Nja8u3xxhNXdcUg4GvYZKA9RmsNEkQVBBerVBRgVV4fdG6xNkgylqVbHV7vlnWWppI/Eq4IZwR8IiI3HW9jfhaUEp9fmrn9YcS8amnKGI0/dbPnfIUZ6+yPnz+9gWmWVVkotMqLNRX2e0raM/MoiMvf6Ik8otzqPEo7Rdw5yx10efTn/0Mf/3lJ1nfrIjyfeS6IKkrKldRSY04Ra2ELzx+hi//H7/NK152Cz/yYz/A0WM3U4mX/YGQaW2yAuK8EkioB6C9iK+Jo7BQjmKeZnfrAmFF69FMr9H79Q53szfg0nqP2U7LOzU36plv6mpqwqEzcui7AsL4EL0mPflsSDuqSW0O2ZYOZRQznZTl5WWUNmhliHRKK5/hthe/jIPLezj95BM8/MD9bJ4+R2pKoj6s9gr6WlGKgNQ+glUqRERe0VKrGKUVM4lhrpPSyVtksSbSQqTCALwoKDYIOBsiNW08UUGgRFG7yr9nAakVZanQScbGxiaDEvplyUxhMUqxsbpGr9elX5S4qsQ66PcNSaQxxhLZFmXtSTFB9I6iN0CLF2utSj9FuSp8k37pCjZ7PVbWt+h2+wi1z05c5fN1ozijKaZ41hCUX4TxBAYJJA/vCJp05WRj7HjK14+fZ4I0JyEl5wDMWAQxrAMp4iihM7MImCDXYgLLztfatAKNo+xt8nvvezdfeOAJ9hy4ndn5/ehEE9d9bFVT1zWDQcGgLkEaIV/Fp+97jM/e+z/z42//Qd7y1h8iztrUGD9pWIG1zo9IEc8m9UQLTZ7lpHESCt7+/RhRQ4HWJgVnm/ph6A+yzlKj6PUGbGz0abdnyGKfNvZadpPvf/i1uZ67iX0ZVDWgSblNOtKdorxJRyUhVeuJT1orlvcsMjffoXFyviUlYX5hDxGKLGuxsP8Qzzz2CCcf+TLm0iXiKGG9V9GrgyNxo3RxGGhBYoT5Ts5yJ2W2FZGkkMQxURQTB11BcY7a+p4gZ6yvC5k4WAoxUFa+PcQ5r/kZG8XtL7+D+x59hrX1dbpbA1Y2uqRRTFWUWOvnJ8XGK3YMbIUZaJyCpPD3nK092amyjm5/QFFbsiolrbyKuB8saekOStY2uqytbVCUA+JYk0Z+XM7G1s5/oqkzmmLXYtioOfZz05MzTuEdkhMCbV4hwx6ZUQHbLwg60LhHTCt/Tq01cRQxNzeHCQoecRwPv/dyToayLHjP772H//bJjxGnHZSKQSDtLKHjlEEhFFXhUyy1Z6iJtaANpTi6F9b4tV/+P7n/U5/mp/7Rz3L8RXdgNSCCMZ6q4qd9+1qOiCWOk8sW04aB1xxvGHPOOZyOcAKFtWxu9inKmoW5GfIkxs/BCn5GRov0OBXaBRkttZsUGIZM0yaSnoyKtmvPjb2wOQrKUtcVgsNgOHBgL3ESbWNmQpp1SPYlbLVydJKQz85w5KZbOXviKZ5+/FE6Fy+wsb7B2kBRVX4whyAk2pCbhE4rYSbRLGURyzM5SWbQSUZR2aCnObqfRWpQ1tcpJUKUbzzt15UXGiTytSqtmWnn7N+/zLFSuPj5NYqiT99uEJmY2XabvJURxQaphaqoGCiL1ZpuUdPdGmCdo6hrqjqohw8qdGXpFwVRFPmav/LMy82tktXVLr1uQRJHzMymZCZHK82pczv/iW4UZ/Su623As8TUzhsEwwUjREj+51EPRFPL2V6k7vf7pGlCHKnJ8zRQ244FFplSijTLmJubo+HfRlE0phbui8Lvfd/7+MznPoPUUNlNtpQjzzukrXmEDBPnVL0uVT2g4XcqpbxCeFGiTMLNL7mTPgm/9Eu/zBvf+Aa+5/t/gFa742fahBRjUwMChg6x8SC+nqEnHMnwvSqFwRe0s0iRdDI/yTVSmEioxydDB8JEc44RvZzdFRXBREpR6dHYiKaGpJqWCOzIMUkUNisWbXz6tUnrRlHCoSOH0XHmFf0FP9PIgFJClPgIqZPPsrG5xmBmjoW9+zh+2+2cPvEYTzzyMNmpC5RbPQZ1BbEiixNyIIs0S+2cvQs5C4stUJ4okEYmTPL1NtTWYq0nSMXNZkP55le0QsUxyipsr8As7OGO7/ge8n0HOLVZUivoliVFWdNONVEckeUpkdY449AKImJEKcqqYqvfZaNboBVkiWcpN6nOjY0BTmC+nZGmMVY5XJiaoDW08hYzeU47SYnMlV3ODeGMxI8HuOExtfPGgDBSCvAss0A9Dgyi8fTc6DU+mWKSmK1+nzzLSGMDItjQIBj2y8Co36TZGCul/NTedouiKBCtg1CqQiuhV/Z4/wf+nEef/Cq3vvhVPPrIgwx6q5T9Hpur52jP7GMum2NQb2I0WBQ6MiARta2pyi5pYnj53fewd88i2mjKos/9f/0wT5w4wdt+9G0cv/lWTFh0SqAZCqmj2PctNQr0rqaZ7tw4IDXmZBuFD8T306nIp4gQRdToHhKitkBkGBMJCuxyx3ZliBsd/nZoHPa247I9cefTlIJDnB/RrQ2+XxFYWphn755l0iREx0ajIr9Ia+03CKI1qhMxl6dkRZ9qUNBZXGT/zcd59evfyOrFFR689zM88dCDtK3XTNRakSQJS62MPUsz5J0UcWY0z8j6Jm0vB+XdqBfGDbRJE6GMpqh75CVUWYv9r/5WXnrPm3Czi5w8e5o0y1hZXaWuKtIkY67TJksTtPb1KAWoNAbRvi4VGMj9/gCtoZ11aLcy4jjGWkdVOQZFRZLEtFsZGIjjhLKyVFVNK8uZaee005wouvKk1xvCGU3xzQWl1AlgE58oqEXkLqXUIvD7wHHgBPBjIrKqvMd4J/BWoAf8AxG5/2v9jqb8IeApqKH2MqHDN7LHLz+KMJ9F6A4KIEUrCRNLA9NnuChtfz3krRZZliECdV0SRRowDMqaP//oJ/nq+TVuevHLEWc4UlmeeuIhbDlg0OvS21phfvkoZVUCBEHLFK2gHKywtNjiZbe/msOH9rM4N4vW/sO/urbK+uY6f/wnf8Yr73wpr/u212G0GUU+w4hlvH9o5ITY5pChSVcGQVn/5OE5TLiGQ4KHNHU2f52b8zkrvn6yW3DFQG4UaXif1LAYBcHrRDrld/lKRTRNwrOdNnkSkRjlU1vaby50FA0jVX99/QyqTM8xO2tIkphOu8Ps3CxpGvOdP/zDPPHAA/zV//tHdE88SEqFEphNMxIVY0gRLImJEG2osFRSh9lDFXVVkuY5WeLbCNCRF0e1CTOHj3DnG7+XQy9+BT0bcfr8eVztyHM/NmJ1dZU9i4u085RWGnsSTGjuNcagnSbCq97YtmWjV6GVMDc7y9zMDGmahOb8mKKqWZzJ6GQpJtYMSktRCd1+QZYmzOQprdYMcTx1RlO88HiDiIzTy38e+IiI/KJS6ufDzz8HvAV4Ufj3LcCvh69XhKCwIc0iqlEG8I8w5oyGSguMogIRIYpiUhVjcRPZpqYBeHspu5EdStOUOEm8M+qvUVtDUZX81Sc/R69bc+uttxNnGXmcc3DvErYccOqZZ6jLLq7uUjpPWKhr3/hY45DacuzgPl71qpdx9MhR9u1ZpJWlGKOxtZ8LgxjOnz/HVx58kBNPnuS7vvONzM/PI8KE8xhKPYnDqMkU5XjqUSlNFE2mL8cbgCdlchpGn2csNj02ViTUJHYPJMxn2h4DhUcDyTIwDJUGaqzUjDfLamNCAOIV97XW6MgQJTFRHKNNRDRUwVDEJiGOU6IoIU5inyIO6SqlI/K5Nne+9jt50Z138cRnPsZjn/hz+ueeoRp06W31sJUibin/+5QijiKqSlNFmqIQdJKQxSl5kqKNUIqg8xavvue7efHrv4tk8RDd7oDB6ipYS4wmz1LyVkZ9ruL8xQskkeal87eRxzHWacrKgnLoKCJShsgonM2Zn7XERrMwN8PcTIc0y6kForyFWMi1IssMUWwYVJbNfsnK2jraGPKsRauVkSTpFf8+190ZKaXejN8ZG+A3ReQXr5MdR4DfBfbh79Z3icg7r/WO/hraa4DPA6dE5PuUUjfhB9ctAfcBPykipVIqDe/r1fgJqm8XkRMvlJ1j+EHg9eH738GPlvi5cPx3xa98n1FKzSulDojImaudTMZSREPV8uFj4JBtzd5q4qtXHAjpuKEzg+HGePyVITKKQ5pBRwlrm6sMBjUf/+Rnqazltd92D0maYoFEGfrdLu005kMf+hAXz/XQUtDbOIerfZTmB/gVzLTafOs993DLLUfZu3cPWRKRRL5BuyxL5ufniJRmeXGepaU9PPrww3zowx/h7rvvZnZuhs5MB1s7lLjh+1WM8vmNrJF3Iw2dmdAHNSIkNMxB64JOt/JUZueE2vqkqB+OHK5f4wR3EYTJuU0jZz32R5cgo0TpJ5x6tgiNAozP9Pk+oLr2863SLKHTbjMzM0PeapPoaDiPrdWZ8U5qjI3YNCvb2o8jV5Ehn5vlju/8fo6++A6+/OH388wXP0e5eZHaDEhURjtPvdi4UxQaFA5RMXESkZmIPEtQJmF2zyFuuedNHHjpK6mTjG6vZKvfpVf0fJSWpGR5xtLSIieeepJa4OTZM8SR5uUvexlVDWvrPUyimOtEJFGEcZYs8aMiYmPotHJmZzvEaU7lIGm10FaIUETGoqIIVVS0WxmtPEVFESZOSZKEOL5Ba0ZhQf0PwHcBJ4F7lVLvF5GvXAdzauCfi8j9SqkZ4D6l1IeBf8A12tFfY/wT4CH8WG+Afwv8ioi8Ryn1G8BPB5t+GlgVkVuVUu8Iz3v782ybAB9SfhbG/xVqWPvGHMxZvNMHOAQ8M/bak+HYhDNSSv0M8DMAcWQmtdXGdvISWHFNomXcoFHTnR9eODzWvL45MlYnGh9nnrZniKIYFSX0K+H0k0/T6xV8xxtey7HDRxAd0R+UKOvo5Tkoxd13380H//SrnHz6SZZKYatfcez2Ozl/fo08Sbj71S/n5puPs7xniTzPMcon26x1tFutkJevSdOIQ/v3sGdxkbNnz3Dh0nnPfjIGW3mqr1G+ED/Sgp8kLzjre5N83cNLOoxEgPwbt24ULdbWMwmVIqg4aIaTUZXyfU67CNt7ii53pr5GhKoRCmDElGzIIk0KS0IDcJIkw3lSWZbR6XTI0oxWnpNmma/nDQkRjZ6coK0dppRFfKrYGUXr2G3c+RP/kIXb7+TRP30fdu0kkTFowVP6bY1yFuUsqUAuMUbnJEv7OfK3XsORl9+Dmt1LqRXdwYDu5pbXnkNhjFe3SZKEmZlOIJB6ivmTz5xEIsO+vYc4dWaFNEuZyduYDMRGRMaS5z5KiuOYNE1JsgxjhQSLOE8Jj7VgrVBUNWma0mq3kHpECrnaDKzrHRm9Bnhc/OA1lFLvwe+UX3BnFBbKM+H7TaXUQ/hF8Zru6K8FlFKHge8F/g3wz0KU9kbg74zZ+b/gndEPhu8B3gv8e6WUkud3W/s6ETmllNoLfFgp9fD4gyIiwVE9awSH9i6AVpoOzR+vD+1ULxod9306atjsOr4zvux3TfbVAAh0Oh20NqRZRpq1yfI2r/22b+XA3mVaWYIycXAOzsu1KMXi8n4WFvextnIWqf2okjSOcJVl6cACt9/+IubmO+R5NuyPquuaJPYLHASaNlFICxmOHD3EkaMHvDKyGy1oxhgY6hyOhHCVUjgFvXJAt9sjSVKyNAuq4OKbiMcWC6310AlrrYepwOF1CRHXs2kYvREx+Xcfc9pKUKoMmpdeIkvrcUJM8wpP4y7KgjiJh+oeWhviJCVJ/RhuZcwk6zAQSRQhKpWQYibILuG1EfOkw63f+kb2HDrGg3/6XrpPfAHlhDgySGiUxVkKJ+Szezh21+s5/Jp7aO07TC0GW/Sotmq6vS6DYuCbpIfv0W8qOjOdkA0YTSF+4qkTrG/0yPNlojgjiiOU9rPOUIo0iYmG18M32SpncWWFtRaTZ6AjXF1RlKVXjmi1KAfFUDexrsfl2CZxvZ3RTrviFzK62BFKqePAK4HP8hx39M8TfhX4l8BM+HkJWBOR5i/d2DJhp4jUSqn18PxvVC7oa0JEToWv55VSf4TfdJxrnLVS6gBwPjz9FHBk7OWHw7Ernz8sBkMnFP43WmDUMPPSPHe883unmtCO34coi0B+yPPcL87acPMtt7Jv7wFs0SOPNbHRKKOROKbCa4n1+30eePDL9PoFIkISt9h77HZMusDMUkWvP6Ddymg3EU5dY13lB7LF0UTNS4W0op9ErDHG9xiVISpq2HJGj6cvvfnWeqXnLG+hTURRVHT7A6KQ3jEIUWSG6Tkbdu3NPCMFzeBgJGj6uebkuwjjfWeeKWfRGlSIFP19EsghjCLjpqmZ8BrnHN3uBitr6zhn0fh0prV+QqufnRaBqOF9Ol7D9OcKjmEbs0IDaD/OfvH4i3jl23+Kh/88Z/0r92GrAbawWCfYfJ5bXnkXL3vD99M+dBuioChqympAvzdg0BtQDEp80jbGGAemRkQjFrIswSEYiT3dWkFdOy5dOs+xoy0O7juINhJGmBfUzvlpClXBwDm2egUqzukXFUWvh7aCrYW63aYclHS3NrGDAYkSJNZUztc/h3qiO+B6O6MbDkqpDvA+4J+KyMa2Zrive0d/raGU+j7gvIjcp5R6/fW0ZScopdqADtFlG/hu4H8F3g/8feAXw9c/CS95P/CPQ1T8LcD6s4kunTSLM4HqOkqheNrySGXZDqMFddkC2jQrXil94MdAa8RokjhFo4nwNF7VSujavh+lIA4dKLaVrYg03Pu5z3Fp9SLdzS5pZiBrkXTaDIoBOkrpbvR5/PHHOHLokO8HKQckISfvNJTWiwNHRnuNMu1HBuD8jtvL/HuGkwr1Df88z/JD9JB8YZRCnJCajCRPqeuasqrobnWJY01bZ160N6Sm/LkMSrng233NRIITql09UX/ZDRg5A++M4tj4Oox/1DsfO6ohubFosJkP1twnm5tbnDt3gY31TeZnF3DO0e/1yFttjDFkWTbBqhu34dmgibRmlvdz5/f/OE/NzfPMZz+KSRIWjt/Bzd/6ZpZuewmSthCnqXtblGVFXddUZQk44sSnVbVSqFgzKItwblhcXCSKIqT2ZAUR8RqHSjh/7hmOHVhCtRIqB2VZU9SOsupTFX3iCGR1lcoJRVnT720Qo1CDiKSoKPp9Nte2EFtRlDXWwaBwVEHz7kq43s7o694VP59QSsV4R/RuEfnDcPia7eivEV4L/IBS6q1Ahq8ZvROYV0pFIToat6Wx86RSKgLm8ESG5wv7gD8KH7oI+C8i8kGl1L3AHyilfhp4Gvix8Pw/w5NAHscTQX7q2f4iIdQxlJqIlGCMGScSaMxX2KFe6dxDanSISrQmy7JRfcn5ekocheFn1mIiwWiD0Y5L65ucOHme1a2S2YX9FIMzKHwPSW/gRSlbM/M88NBT/O1vfwMqNFvGIT0nCBE6kDN8E6I2CsTXK5xz1A35YuxNN2SEyUUwTG3Vo/cdRb52UFQxW91NYMBspz106k1qrxaLEtDKYII9ICijLoswb3QM//7aEce+H0jEUY+lOoOvmrhOzT3UpD1FhKJwnHzmFBcurHD08FGc9YSTwWBAFEWUZenlgeJ4SJ3f6Z7byTmN1zqV1tR5i+LgQTh+hNtfejcH7riHaGavV2p3iqoqqcM0V60UcZKQJpFX4AZP1w5D+GxdBymjZebn57h0/vxwLpWvCWqqGr788JO8+hUZJsnoDir6ZeWjvqogizVFaen2BlTOUdV9jChqhMhsgrVUxQBcTa+qqKxQRxDF5qpK79fbGd0LvCgwwU4B72BU93hBEeouvwU8JCK/PPbQNd3RP1eIyC8AvxBsfj3wL0Tk7yql/ivwNjyjbrudfx/4dHj8o89nvSjU/16xw/FLwJt2OC7Az34jv2t8wR3/flyduol8JunKk+e42rldeI0TIc9boT4wIt9p3Sgiy/DnyEQ8/vgJeiWYfB4tEcXgvB+6VlVUVYWI0K0dF9cLTp09z/7lOR9daU8dVoG+pgC0Q6mmHTewBxEcIwdLoCNrPaZ3RsOaG0kkbd+lmyimPTNHbBS1cyNdAvGinE6aVJ0OpAW/VBpGTL3dAmU82SWKYnxLmWBFwkRk5XORodbmGwhG94zW2jfANrVKZzh79gJnzpxmcMftJHmOdTVVVfp6UlH4+pvRaAxDtudECWnne8/fXzWC49LmJh/9xAf48oMf5sXH9nIyWmfz7JfZV/ZZmN2PjvzMLRvSyVopsiTx9is9YgNWtY/axaIjxXJnhpuPH+P8+dNEyvi6kRWgBqVZ21rnwUcf4/jxW6hqx0a3T1WUGCwuiRBVYDe2fKO1FCiJqEoBt+6VxWONdZbuoKCsSvrG0IrznVn1AdfVGYUaxj8G/gJP7f5tEfnydTLntcBPAg8opb4Yjv0rvBO6pjv65wk/B7xHKfW/AV/AO1bC1/+slHocWME7/F2PcWezPRUCXOakGge1/RzbMf66phalRaNNSpK3MUqjsZ4mrMBEUfgQe+ow+JH0Dz3+GJWC2dk51jbPIWgIkzOryqIwOGtZ3+rxwEOPsPRtr6KVRiRpUCLHjdatwBjUgSU3VhkLRXHPiHPia2NDRzqmtj3+nsdTkkZ5KSC/lmkQTW1dkLjxsZBzMtzRquB8nYp3VclIK0WSxOHa+GM+netdfOPkhz1FTY1sKGQ6ek3D2uz3+pw8eYq1tXU6s3OIeMq+tXb0PBdUHLTesca2E+FGSQ1iWO2t8/sfeBdPPn4fBw+0iGcT1uUZtjYvcnbjcfYtHOHo/rswehlFf2LjVdd+xEgzR6iua+qqpqq9GGoSxRw7eoT7v/h5nPMpS2+zJRRbOXfhAsrEzM0uUPYtthYsUBZVYF56J66Nn8M2GPiUdafToqUzEBgMSra6WxgTMUjsVaPp6x0ZISJ/hl/Yr7cdf8XEvmUC13RHf60gIh/Ds/uaiOQ1OzxnAPzoC2rY84wmBee/l4kFdieHsz2FdzVsj6A880ljTESWtxlOLQq1JF/c9guYtRaFYnVljZNnzqN0ROQc/a0VjFIkqZ9erLQfVuesxSI8+cRTfMc9ryRNk0Abtp5AHZhro+Vy0s6hU1JB0UFGQrHjDglGYyMue7/+3fqpwgBi/CITakOjyzk2EXXslbsGaqSu3tSNxp3yOMtrSBxxY5JH0jQUhyhY+Qm7z3z1NGfPnGPfgQPeGdU1ZVmSpunQKQ0JEDT1KIZ27HhfimKtt8r7/ux3eObk/bSymvm5GayyiEQ4pSCuOL3+FGubfW4//m0kJqXRTKzrmqIohrZUla8lbW1teVmh2iLiiOKYw4cP8/SJZ3x7hDGIa5iUgojl7Nmz9HoDZvIZ4jSnqBz9oiQyhtgokshAZChKx6BfeJamMcSmGRdhKAuLBNUIuUo7wHV3RlNM8Y1gPDIaF0Tdvuhup36Pf21wGY0bv3grM+I6pWk6ol8r7x48Bbr5eTSd9ZGHHsZpT51eO3OCur+FyTPSPKcoikC1VUTaIOJYu7RCXVVDKre3uTGOy1Ib43TzYPlESnIonLrDexuPJpsd/4hhNnJkw0UXJpzabkVDbhndA24UvQzvEfHpqrH33aBhEjbeWZRvob24usZTX32Gm269mYWFZfyARxmeu1GKvxJBpkmxNn8LQRhUNX/45+/m0Sc/hXIVlYLNzQEq1mRpTRJqmSIxm4PTPPjwh3nZi16LdjOIMlTOMigKisGAQX9AVRaUxYDe5iaurrBOqK1gy5Jbjh1hdWWNqhq1AkDDxNRY51hfX6Po92l3ZjBxgriKNG0x02nRylLEaAyKunTMtlrMz7Q8W080VVmzvhlR9AcwcNs2NJOYOqMpdh2GkQCTtSJgIh03XjtpMO6Qtqevxr/6HwhpMEiShCRJ0WaMBKEEnELEJ9AQEGv58gMPonSKFuhurIIrSeIOTjRihaqsQGnSxHfNOydsbmyhD+5FSfNrxxxjWCjNsPDApI0ItRvrJRlep8t33U3Kctg7wyhtdVlRI1xP5xoFhqDi4FfPr0kCuRHROAUbWHO+qddN9FI1zLlxB64UYVw9oaijsUC/Knn86ad5+fkL7Nuznyht+5HeY8575NhHgrUT916IQhBFLcJffPQD3HvfR4mSHpDgbMVsp4vJhZmZWWIVYWuLc4rNzXWeWX2asqp55e1vwtqEoizp9nsMen0G3R6uLBgMelRF4YfvxRl9LGmccHjfHi4dP86jT5wI6cHwtw29RCiw4uiXBeW6JYoi8iyj1U5YWpijk7fQWkijBJNkzLZyZjoRWZYgziAaLq6vUBaDXtknhQAAF6hJREFUr3m/TJ3RFLsO46mj0bFtk0jHPvBXWpQvP+92h9Q4HsGYGK39NFWl8MP4xLPKfBOOxjlha2uTJ586gZo7Sm9jg7K7htKCiVOyVoe1zSIwAEHhmJubxRZdLq2sjiRjGItOIIxUF6+tEHKEozRigFbBcfojw9lNV7gmQ8KFCvWSkLpyzVhzJlN7fsFkuJjKNXZGSql54DeBl+GX/P8eeIRrKa47lq4d36hsH8feYJIEY4c3XpMiba79U099lVOnTnPs2DGymVnqOh6SVCZrm2EWVMAoWhJE1QgJDz/6EP/lv76LuSXHYKugKku0Ei7oi0T5ER74wjkeffgEvd6AAweWufnWA2AqLjz4aWLV4fZb78aWlqI/oBgMKMsSV3pSRRzHZHmG6Jher2L//v1oBpDO8NRXT4Vx6KNswyhCbHywf6xb9Dm9ch4Ri52fZ262RbuTE7dmaacReQuSJMbVmlIc7Xab7kYXsZdvDscxdUZT7ELIhMMYd0RXu9m3R0LbF+fLIiVxEObcGKPHcv8hItPeEYnyPTzOOU4+c5KNzS6RO8P6xbNQb6F0RGtxL7VjuCjUdUWcxFRVSTvPuXD+PAg4aydiE+8qgoMYpiNHjsLboYeR4k7vedypND1E1gaJFu3n/G135k1ENJ4SvFqK5RrgncAHReRtSqkEaOEJRNdGXHcHBzr+/fY05JWo2CG+mTjW7xc8/PCjvOylL2Fx7z5Pu68ryrIkjuOhqsW4HU1KdoSY9c0tfut3fo2NrUvML+2lKgtEvCzT1qblc594ki98+gmKnieSfEk/ydLeFnfedTNxJqye+SMOH7gFqfLhyAkRr2afRn7kQxR7OautrQGHDx9m0FuBbHbMxrGm1PA3V0oNm6mttWgMg27Jif5Zvnr2HEmkaKctZlszzM3PsbDUYX5+DhPHpElCp9NhJVqhctVVK41TZzTFLsRk+gSuMMNo28Jz2QL7bH6TaiT1ff/PMEUXqL9OGgqA/x2PPvYY/f4A1ztF2VtFS4nOOiSteWygSTvn/GIkEqbFai5eWvHd6QriKAx6C1GKU4A4JMxd2n4pFD4wGjrU8P6uROneTn23rqEy+7qIZwkapBJGeUGvMGCCUkMzkuNaQCk1B/xtvA4kIlICpVLq2klxiQzn8owOecLBqGa2c12tSbFJoxQfUrJNlGid8Nijj3HmmZMcPHAYl+XUNqGsC9I6JU0bpepRAXD8/EopKmt53/t+n3s//ynSdsWlixcx2lAUA2pbcG695okvX6SuI3+vWYs4xbnTBfd96kle/qrD9PQFPvu5j/Dtr/sR1AqAI0oMWidEJiLLc6+lF6eUZUWWai6cryhMRZYlVGFu05AODj5dqb3CRCON5NsZLAqNOOjXikGxxdpWn+jSRaIThiiOyPOMOI4oS0sUaary6tuZ3TUda4optmHYyLhtIdmemtqew99+jvHXjh/XnquAiZNJYoCAEgVoHMYrhTvHU08+QVX3sNUAdESSz9KZP0AUz+Os4GyJVhaFJTIGbImIsLrWpTeocDTSRf7cIgqsr01ZJ8M5To1D9GUeRaQMfrSG8qKbePUIF1KKw9eMpZqaor5v7NVYsZSuxCqLlQq0Q/Ad+j4NqPEz+RpndM0YdTcBF4D/qJT6glLqN5VX7/h6pbiuCOscg8GAoiiG7LKmttPcG9s3L+MEh+331dARWYu4mpXNTe7/0pdYX1vxwqDO07rruh5tgLZdrnHSyRe/+CU+9pcfJsdhKtASkeYZoqDdWuLsqS2qQo8U1gGRCGcNayt9nnriHBvdDT7+qU/Q3ThPnvq5QXEc02q1mJ2bY25ujrm5eWZmZpmfnydLU4w2KK2ZnZ2l3W5PRML+fV6eTWjeU8MW9GxSR21LaltTlCW9Xp+VlVXOnbvApUsXESxJGhHFV75nps5oil2J0Qfy8seuVMTfHiE0x66UvvG1m/DvCn0i4xgUA06dOoXWMTrOaS8cYHbfbczvvQ0T5wz6W8MFzqfJ/O+Ioohev093a4vLGQpjdYbtrLqx1U1pNZSv2YmM0bz3JioaLsDjJ1UKJRqtIkT8JNtaFI4wSTfY7XYgSzxHRMCrgF8XkVcCXXxKbtz+8TDtWUEp9TNKqc8rpT7vnGVjY4PNzU16vR6DwWAYAYTzb6vvqInvw/maMw9f56NPR43ir7/yEGfPnKLf7eIqiy1LrK1H9alt1jfnP3nyJL/7f/8aqS7J0hbFllD0Lb1+gVWOtbU1ulvFkFo+3sPkpPa07c2KJJ3h4Uef5oMf+AAL83NEkR+M55mgOXlQEo8iP96ithYTRVRlRbvdZv/+AyOK/zZjR3XFyWs1uh98Y7at66Gzcg6/eRGHUg6tIY53TifD1BlNsUsxWiS271gnC/Tj31+N0HC11zjnBUivRHpoCtwb6xusr68TxxlJa56otUi6uId4dobK1mj8rtJaO0wZGmOoqto3wK6v+0UfGe46x3fuV7I1HPA2aj22Yx0RE7Zfo8Yx6YnR4T4V5RcRhRONFbDijzdpv2ZK7DUkMJwETorIZ8PP78U7p3PKS3ChvgEpLhF5l4jcJSJ3qSAbpZQfgZAkyRXvle0R5M7w0jmKwMKzNVurm5w9fZ6trRX6ZZ9+ZSnLcWadhIgaEAvKUlUVH//4R6iLNW67+Wb27z1ApFPOfnWV2KYsz82hdURV2SHLTazDVjXO1Yjykk11WdDbsKxe2OCjH/0oG2tnmZudR6uEyJhhillrEOV8z5C1xJGhlpqlzgyLCwvEcepFXgmKE0ho2MY3datAdBlrqBYruNohduS4ra2p6wqRhtbu9RKvNqp+6oym2JUQkfDhGP08+iBcvuOFy6neX4vwIDAc1BelydUjASdcunCeygImIYoz8rxDFOXYyk9e1SYhilK0jhHRpGkLJxqlIwYVrG+soNyINj10RljCegBu3OYmWHBYZxEV2HG6WTy87RKKSk75hIoNKTZjDMY0TkYhzo820NozCLWCSGuUSDiuR+KfaMRdm8hIRM4Czyilbg+H3oQfI9NIWcHlEld/T3ncw7OQ4tIhFbWwsECe58MZRFft/2EHpz802kflniTg/za1ddz/xS8htaO3sYEOIxMGg8FEZOTvO8/APHHiCWbncm665RZedde38LYffQdvf/vf4fve+kPMd5bp5LPEUUxdO2xthjUd5xxSF0hVUFdCVQpbq1vsX96DFcv/9+EPMr8wSxRHaKWHBIXmcxLFXhIpb+eICAf37ePAwYPMzHSGCYAmBQyBiRjIPEpPknma+8eFtGRdVxRFwWDQp6pKv8Fyzb06ZdNN8U0GFSjJzaLRNL7Wdb3j7nanFF2D7emYnZAkydWsQUS4cO481omfaZO3SJIsLPAaJw6lDXGcAH53XtWWPPZ9RmVl2epuNRYNz+mcxWFRKr58gVTK9z1ZRVkWniU3JHJ4BWYvRzN8o74mxWh3C97RiGjqqh4uWhPU8tA13zgin2L0o8uvIf4n4N2BSfckXl5Lc42kuLTWxHF82d94PHU5jivVFbfXJ5VSiAOHptaaLz3yKOdX1shn5il6W2StFmVZDlUZxptgV1ZW2dxa5fjxW0h1wt79Bzh203Fe8be6bG51+dKD9/P5v/4UinVsLbi6wLp66CyMVl5OD0dLRxzbf5AjR49w6eIlNjY2uHjpNAuLeyh65Ta7fUSeZhmdvAWPPMHRI0dZK7rs27/M+Qtnsc4GAgN+YyMjD+VTwm7iswcEFqGvxVlrtzVf4+uPV4l/ps5oil2HYe7a/zBcbLcz6q60q91Obtjp8WEaS/sZA8ZEE+eefL6PuM6eO+vZSklGkuboKIHQ0d/sJJ1zfqKrMSSBZdUsUHVdD6sREywuGdPfG7d97HllWQ7p5V7U0+vZjTviYb0hpIsYV1xwI3bc9uvX/N7xyNJauwM9+RuHiHwRuGuHh66ZFFdDZNnuWMZHi09uTBr2WyMB1ByRiWJl0x/mm0Mr/uCP/4Rf+nf/OxfOXSTrzJAkMWVZYkyEcxZjFFVVcO7sM+zds5eZmSWOHb6ZsiqZm99D0sswUcqxo7fQK/o8/fQlXCXUtqAWh3Z+sF1hvZJ7JhXH9+/hnrtfw+z8HFuHu6TGcOH8GfYs7UNlCXVdkyY+BedcDRpirYl0xFzeYm5phqy7yb49+3g4egRrHc7Z0OhMuBZNg3S4V1RTNxoNZUySZNg60Gxqxh3S1caOTJ3RFLsOSjU7QoaLxjihwX8Awo7e5xrYiYc8nqLZacesdKitKEVkDEYbtDL+dYqwsPjoxSq4tLKCw9DqzCEmBm3wkvwVsVJ+t6mbvqCIZuS3tY5Bv++ZTcp4GSKnwge9xoSBftoY1DaCghFwRiirciRmqpTPL4b3OHr3goimsp5C7kQCEco7Pa20n2lDkwYlvIphFAX+uip2Jo/cqBgvuo9Psf1a8A5qeBaGb3z8dmkWaABRfOUrD/P0iae5cPYCkdakSeT7e5T26U8Tc/r0V5nJ2+zdc5A0zYe1PoUhzdoMipKF+WUO7T/Ky1/6Uj75oc+yul6jtLC4sAgq4vylS+AEHRluue3FHL/5VvI8Z21tjVa7hTER58+f4ejRm+n2CmpriTCAT8UaAe0UeRShtaIVp+xbXqbd6lAUNaPLo2CsWdeJ51OCmti0Nfdlq9Uasu0mndFIamknTJ3RFLsORVVdfPzk6S7P47Ta7Xjg0Sd49//zH5/Vc5965FPNt8t8HTb+60//Dv/66zftWuDrsnMbjl1LQ55P7KRbuD2lO47JPiNgqEJxOcFh/LllUfPe9/4Rf+8nfozzF84wuzBLnPkoWGlNr7/F5sYmL3nxS0iyDAK5oometYqIogRjYtrtOQ4eOsSeYzOsPriKEcXswjz9XkGMAVXTmsm5+5572H/wKMZE6Din025TliX9ouD8xXPs23uQwaBExAwzCrau0Sal3W4Dmig2tDs5e/Yusbq+iq1dcDZ+AyXSqFgwEoCVyzMSwLAm1zj/8XrVlTB1RlPsOojIHqXU50Vkp7TODYPdYCPsHjufK7b3m8EoJbs9RTfugPzjzWBBQCQ0/o7OG74L3xu++IUv8Q//h59gz54lzp8/R95qk7cUaZqxcukSi4tLvoYkgrO1JxQM7dF02jMUgwFzs3Ns9la46SXLbG2sc+n0FmkrZ3VljTRydBYTDh9b4szFM3z7/rey1e2yaBLiOGY20qxvrLC+sY5Smv37DjLol+go9WncUNPJ8hyt/FykNIs5cuQgTzz5BHW9c+1MxupF3rGNRpdMXo/J634llZAGU2c0xRRT/I3AlUgK48cnyAmNNE5obm6kcnwpL0zPbf5TKtTpfPPx+YuXuP+zD/Dd3/8WTp86zealVVITEecpUgnt1gyVE7QWimIQUrc+etDapxA7MzmVzIDSRApe95ZX88BnHiOWjCSq+Z4fuYtoMWZjs+KLDz3AW1dXSGZmiSIhy3K0NphIs7G+weraGgo4dOAQ/VKwzvcFVUhg3CnausWepWWWFi7QabUpirVg0/BNh0gp8vVDN+Z4xNcohfG05mSd7mrq5TCldk8xxRR/Q7ATrX+8hgST9H9Bj1TN1ZgWoPJU5vFzTpxD/Jyq9/7x+4l0xKGDh3AKBkXho55OBxHo9XrDBbqRJRrZ56fF9vtdtrrrtNoptSuAhCiJOLAn51ve9EpueslxetUWa5srFEUPcY5Wq8NMZ448bzHTmWNhYYnZmXlWVta4ePESeZ5Q1xVO/JA8rTVZmhEnMfPz8ywuLNJudyaIB4xFPRPknrHr2TBZR9dl0vk3daQrYeqMptiteNf1NuBZYDfYCLvHzueM8VTc+MI6zqAc1okkNGs2k3cnakX+fI37aJhjngtdI1rx6FNP8xd/9kH2LC+z/+BBREFVlSwtLflZQ4PJiGh731scx/QHm6DX2btvkUhn1LamcgWHludQubCxcYlEO/YtzNBqp2ijydM21nml+U5nlj179rG8vIf52UXOnD5Lt7eFMQybUvv9PlVVEUcRSZKQZRntdmvCnvHE23aWpUz8GzVqjyuUb6+17YRpmm6KXQkRueEX0N1gI+weO58TtvWSjdeIRJpaTfPUpm4U5jgphVLNJN3mhGpihR6PapTSSC3UyvHu33svb/7et7CwNM+e5b2gNYNBSVVWJGkSNNn9JFSl0nAeQWuDjjUDe4mlxRZRMkdRxZ5/WQsze+c809IIC4sLpFUbnWSB9OZIk5wqNNzGcczM3BzWOYq65KknHydPc6rSMhgMqMuSKo5IkrbXsmtnLOxbRD+mx2jvo3qPV/IOFPfhIFzvfJwE24eiquN1pPHrdzmmkdEUU0zxTY/x/fgkS+5yxe7xOUMNg65pEWhSdOOR1UThPrQaNP00T534Kh/7bx+jGAwQ64giP/xOnEOrER1apA6Ob2gl/WKLbv8S8/MzLC3NE0WKSMeIVZjMYGtHmrUpKsfK6gZVVRBFmv6gDwhZ5sdGDAYDtra2iOKYmblZer0+jz7yMForyqJEa0MxKL2yd5aRxDFHDx0iz/PhezNjabmG7u4Z7qEXa9g/MNaXxXhqVCau1U6YOqMpppjimx6+Z+vyekXTkNxQmBs0KbvtGFcdGDqxHRbYRiEEgY9//BMAIwVv7WnccZzQvNxaFyKvpq3W0euv4eiTtzLmF2ZJEoNSEcZEJGmEk8q3hSlHr7tOv7tKnBhKa5Gg8pG3WszPz4c6lRBHMUuLe9i7dy/Ly/NYKzgL/V6BrbyCRxRFLLdn2bdv30QP0Shl5/vYFIqhMkPzntWoTw0auaTRVmBcRuiya3a1HN4UU9xoUEq9GT+IzQC/KSK/eB1tOQL8Ln60gQDvEpF3KqUWuYYTSq+RrQb4PHBKRL5PKXUT8B5gCbgP+EkRKZVSaXhPrwYuAW8XkRMvhI3PJ5RSm/jJsTc6nkvP1wuF52rjMRHZs/3g1BlNsWsQFtRHge/CKz3fC/y4iHzlOtlzADggIvcrpWbwi/oP4YfErchoQumCiPycUuqteA22t+Ink75TRK46ofQa2vrP8HI7s8EZ/QHwhyLyHqXUbwB/LSK/rpT6H4E7ReQfKaXeAfx3IvL2F8LG5xO7pZdqN9j5fNk4TdNNsZvwGuBxEXlS/DTQ9+Cnfl4XiMiZJrIRkU3gIfyQtx/ETyYlfP2h8P1wQqmIfAaYDw7teYVS6jDwvcBvhp8V8Eb8qIadbGxsfy/wJnW1RP8UU1wjTJ3RFLsJX/eEzxcKSqnjwCuBz3INJ5ReI/wq8C/xSnTgU3NrItLM4R63Y2hjeHw9PH+KKZ5XTJ3RFFM8RyilOsD7gH8qIhvjj8lO3X8vIJRS3wecF5H7rpcNNwh2C319N9j5vNg47TOaYjfhWU34fCGhlIrxjujdIvKH4fA5pdQBETmjvoEJpdcYrwV+INSrMmAWT6KYV0pFIfoZt6Ox8aRSKgLm8ESGXY3d0ku1G+x8vmycRkZT7CbcC7xIKXWT8kPY3oGf+nldEGopvwU8JCK/PPbQNZtQ+lwhIr8gIodF5Dj+en1URP4u8JfA265gY2P728LzpyynKZ53TNl0U+wqhB3+r+Kp3b8tIv/mOtryOuATwAOM6jH/Cl83+gPgKGFCqYisBOf174E3EyaUisjnX0B7Xw/8i8CmuxlPAFkEvgD8hIgUSqkM+M/4+tcK8A4RefKFsnGKv7mYOqMpppjimxbTvrRv2NYXvC9tmqabYoopvikRFtT/ALwFuAP4caXUHdfRpBr45yJyB3AP8LPBnp8HPiIiLwI+En4Gb/eLwr+fAX79BbT1n+BbFRr8W+BXRORWYBX46XD8p4HVcPxXwvO+IUyd0RRTTPHNimlf2jeA69WXNnVGU0wxxTcrpn1p3xiuS1/a1BlNMcUUU7yAmPal7Yxpn9EUU0zxzYppX9rXj+vWlzaNjKaYYopvVkz70r5OXM++tCm1e4oppvimxbQv7TnZ+3pewL60qTOaYoopppjiumOapptiiimmmOK6Y+qMpphiiimmuO6YOqMppphiiimuO6bOaIoppphiiuuOqTOaYoopppjiumPqjKaYYoopprjumDqjKaaYYooprjv+f/3oH5MkhNwtAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BMJFyiCpQXCP",
        "colab_type": "text"
      },
      "source": [
        "Make Embeddings\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e5L1YoFHHuF7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 73
        },
        "outputId": "08646ea2-5cdc-42a3-cf22-f905dba6d3d7"
      },
      "source": [
        "# create 5 encodings(1 per subject) and compare with random faces from dataset\n",
        "\n",
        "face1 = face_recognition.load_image_file('/content/jack/jack10.jpg')\n",
        "face2 = face_recognition.load_image_file('/content/gates/gates1.jpg')\n",
        "face3 = face_recognition.load_image_file('/content/modi/modi103.jpg')\n",
        "face4 = face_recognition.load_image_file('/content/musk/musk104.jpg')\n",
        "face5 = face_recognition.load_image_file('/content/trump/donald trump speech106.jpg')\n",
        "\n",
        "random_face1 = face_recognition.load_image_file('/content/jack/jack109.jpg')\n",
        "random_face2 = face_recognition.load_image_file('/content/gates/gates123.jpg')\n",
        "random_face3 = face_recognition.load_image_file('/content/musk/musk121.jpg')\n",
        "\n",
        "face_encd1 = face_recognition.face_encodings(face1)[0]\n",
        "face_encd2 = face_recognition.face_encodings(face2)[0]\n",
        "face_encd3 = face_recognition.face_encodings(face3)[0]\n",
        "face_encd4 = face_recognition.face_encodings(face4)[0]\n",
        "face_encd5 = face_recognition.face_encodings(face5)[0]\n",
        "\n",
        "rnd_encd1 = face_recognition.face_encodings(random_face1)[0]\n",
        "rnd_encd2 = face_recognition.face_encodings(random_face2)[0]\n",
        "rnd_encd3 = face_recognition.face_encodings(random_face3)[0]\n",
        "\n",
        "faces = [face_encd1,face_encd2,face_encd3,face_encd4,face_encd5]\n",
        "\n",
        "encd1 = face_recognition.compare_faces(faces, rnd_encd1)\n",
        "encd2 = face_recognition.compare_faces(faces, rnd_encd2)\n",
        "encd3 = face_recognition.compare_faces(faces, rnd_encd3)\n",
        "\n",
        "print('actual 1',encd1)\n",
        "print('actual 2',encd2)\n",
        "print('actual 4',encd3)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "actual 1 [True, False, False, False, False]\n",
            "actual 2 [False, True, False, False, False]\n",
            "actual 4 [False, False, False, True, False]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cNUsDOYJNDlw",
        "colab_type": "text"
      },
      "source": [
        "Wow! just legit. With only single face encodings the results were accurate."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MCvbgr_eEpCU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 129
        },
        "outputId": "f02b54f3-105b-4e69-da5c-c2dbd1469374"
      },
      "source": [
        "flds = ['/content/gates/', '/content/jack/', '/content/modi/', '/content/trump/', '/content/musk/']\n",
        "print('Number of images per subject')\n",
        "print('Bill Gates',len(os.listdir(flds[0])))\n",
        "print('Jack Ma',len(os.listdir(flds[1])))\n",
        "print('Narendra Modi',len(os.listdir(flds[2])))\n",
        "print('Donald Trump',len(os.listdir(flds[3])))\n",
        "print('Elon Musk',len(os.listdir(flds[4])))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of images per subject\n",
            "Bill Gates 150\n",
            "Jack Ma 150\n",
            "Narendra Modi 150\n",
            "Donald Trump 151\n",
            "Elon Musk 150\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xB6cRrVP7JUK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 111
        },
        "outputId": "6edc9732-8310-4837-f576-e419f7cd7638"
      },
      "source": [
        "# create face embeddings\n",
        "\n",
        "embeddings = [] # store all embeddings\n",
        "cl_lm = [] # per class number dist\n",
        "pat = r'(\\w+)\\.jpg' # could've been easier with try catch\n",
        "\n",
        "for fld in flds:\n",
        "  for img in os.listdir(fld):\n",
        "    res = re.search(pat, img)\n",
        "    if res is not None:\n",
        "      try:\n",
        "        image = face_recognition.load_image_file(fld+img)\n",
        "        # it was giving an error becuase some images do have 2 faces in it. so we only pick one\n",
        "        face_encodings = face_recognition.face_encodings(image)[0] # to 128 encodings of single face\n",
        "      except:\n",
        "        continue\n",
        "      embeddings.append(face_encodings)\n",
        "  print(len(embeddings)) # count how many faces we added per class\n",
        "  cl_lm.append(len(embeddings)) # make not for making Y"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "145\n",
            "279\n",
            "418\n",
            "562\n",
            "710\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M7n_ten0lpsp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# make y using cl_lm\n",
        "\n",
        "Y = np.zeros(len(embeddings))\n",
        "Y[:cl_lm[0]] = 1\n",
        "Y[cl_lm[0]:cl_lm[1]] = 2\n",
        "Y[cl_lm[1]:cl_lm[2]] = 3\n",
        "Y[cl_lm[2]:cl_lm[3]] = 4\n",
        "Y[cl_lm[3]:cl_lm[4]] = 5"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ps0AnrNa7JNu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "be08fbb3-4647-49e3-f646-ed9a482c7c3b"
      },
      "source": [
        "print(len(embeddings)) # for some reason, maybe its the output from face_encoding is causing problem\n",
        "# number of faces are more than 1 in some images so embedding are also more than one\n",
        "print(len(Y)) # this is correct"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "757\n",
            "751\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "23bPxDFNd6Sx",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "cca6d502-e12f-4c20-863e-2f184545faec"
      },
      "source": [
        "# new embeds\n",
        "\n",
        "print(len(embeddings))\n",
        "print(len(Y)) "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "710\n",
            "710\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pauf1RzESSF_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X_train, X_test, Y_train, Y_test = train_test_split(embeddings, Y, test_size=0.2)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VaXPjxKrQcQj",
        "colab_type": "text"
      },
      "source": [
        "Predict Result"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_-rs5hFZ7JKB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "e2cd1e48-1729-435f-d3ae-8792c750e5b6"
      },
      "source": [
        "# predict using all embeddings(710)\n",
        "\n",
        "yhats = []\n",
        "correct = 0\n",
        "yhat = 0\n",
        "\n",
        "def predict_results(result):\n",
        "  ids = [i for i, val in enumerate(result) if val] \n",
        "  values, counts = np.unique(Y[ids], return_counts=True)\n",
        "  yhat = values[np.argmax(counts)]\n",
        "  return yhat\n",
        "\n",
        "for test_embedding, y in zip(X_test, Y_test):\n",
        "\n",
        "  result = face_recognition.compare_faces(embeddings, test_embedding) # a big bool array of 710\n",
        "  yhat = predict_results(result)\n",
        "  yhats.append(yhat)\n",
        "  if yhat == y:\n",
        "    correct+=1\n",
        "print('Using all embeddings')\n",
        "print('{0} correct out of {1}'.format(correct, len(Y_test)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using all embeddings\n",
            "139 correct out of 142\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P5_lzGtuGur6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "outputId": "589725dd-bf2f-4279-8e02-e56ba62e9c49"
      },
      "source": [
        "len(faces) # reminder"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wytTwtp0KJzD",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "ffe02504-5838-421c-c8c9-0899b5338205"
      },
      "source": [
        "# predict with only 5 embeddings\n",
        "\n",
        "correct = 0\n",
        "yhat = 0\n",
        "\n",
        "for test_embedding, y in zip(X_test, Y_test):\n",
        "\n",
        "  result = face_recognition.compare_faces(faces, test_embedding)\n",
        "  yhat = [i for i,val in enumerate(result) if val]\n",
        "  if yhat:\n",
        "    if yhat == y:\n",
        "      correct+=1\n",
        "print('using 5 faces embedding')\n",
        "print('{0} correct out of {1}'.format(correct, len(Y_test)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "using 5 faces embedding\n",
            "56 correct out of 142\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}