Experiments_report.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "Experiments report.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "mount_file_id": "1M8oj-_xbaxPjZ7EL36L9PCAhCu7bHQCn",
      "authorship_tag": "ABX9TyM3t76nQVBci7XrothUkJDd",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/AndreyBocharnikov/sincnet/blob/main/Experiments_report.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gDzChMgrRI-H"
      },
      "source": [
        "# Задача 1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EiCtDFSfhRv7"
      },
      "source": [
        "## Подготовка к экспериментам"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tDN_qky5hdYo"
      },
      "source": [
        "Поставим нужные библиотеки, скопируем датасет на локальную машину, залогинимся в wandb"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NFdl1BVbE6K4"
      },
      "source": [
        "!git clone git@github.com:AndreyBocharnikov/sincnet.git"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0ClvUy3cE7I5"
      },
      "source": [
        "!pip install wandb\n",
        "!pip install -U PyYAML\n",
        "!pip install torchaudio"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2Q99t2P9FDhy",
        "outputId": "6219912d-ea0c-4c14-eba7-6ca7be103ae7"
      },
      "source": [
        "%cd sincnet"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/sincnet\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XjPeBaWhqUGs"
      },
      "source": [
        "!bins/timit_lowercase.sh"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pRK6EWpWtWzL",
        "outputId": "a3b5db7e-e289-463e-a03d-5f679c0f475e"
      },
      "source": [
        "%cd src"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/sincnet/src\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XNUjFzQ7TpXY",
        "outputId": "1e21825b-3310-4b99-ff5f-287caa216491"
      },
      "source": [
        "!wandb login"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[34m\u001b[1mwandb\u001b[0m: You can find your API key in your browser here: https://wandb.ai/authorize\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Paste an API key from your profile and hit enter: \n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Appending key for api.wandb.ai to your netrc file: /root/.netrc\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yLZx_h7f3UjV"
      },
      "source": [
        "from IPython.display import display, Image\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "from matplotlib import rcParams\n",
        "\n",
        "%matplotlib inline\n",
        "rcParams['figure.figsize'] = 33, 27\n",
        "\n",
        "def draw_three_images(path1: str, path2: str, path3: str):\n",
        "    img1, img2, img3 = mpimg.imread(path1), mpimg.imread(path2), mpimg.imread(path3)\n",
        "\n",
        "    fig, ax = plt.subplots(1,3)\n",
        "    ax[0].axis('off')\n",
        "    ax[0].imshow(img1);\n",
        "    ax[1].axis('off')\n",
        "    ax[1].imshow(img2);\n",
        "    ax[2].axis('off')\n",
        "    ax[2].imshow(img3);"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C51M42Oehnrh"
      },
      "source": [
        "## Ээээээксперименты!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0jwuinxfipmb"
      },
      "source": [
        "Я воспроизводил подход и результаты из статьи \"Speaker recognition from raw waveform with Sincnet\".  \n",
        "\n",
        "Решаемая задача - распознавание спикера, затем, выученная на этой задаче модель, используется в задаче speaker verification.  \n",
        "\n",
        "Основная идея подхода заключается в том, чтобы заменить стандартные свёрточные слои в первом слое модели на полосовые фильтры. Авторы мотивируют это тем, что первому слою довольно сложно выучить осмысленные фильтры, из-за того, что ему приходится работать с высокоразмерным входом и из-за затухания градиента. \n",
        "Поэтому авторы вводят полосовые фильтры, в которых нужно выучивать всего 2 параметра (верхнюю и нижнюю границы частот, которые параметризуют sinc функции, которыми и задаются полосовые фильтры), а не все элементы фильтра, как в стандартных свёртках. Хотя эти фильтры и обладают меньшей выразительной способностью по сравнению со стандартными свёртками, благодаря bias'у заложенному в них, они выучивают представления более полезные чем те, которые выучиваются случайно инициализированными стандартными свёртками.  \n",
        "\n",
        "Авторы утверждают, что используя такие фильтры, модель быстрее сходится и демонстрирует более высокие показатели метрик."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VibeFMjZL9NB"
      },
      "source": [
        "Все эксперименты ставились на датасете TIMIT, так как авторы учили на нём одну модель 24 часа, учить модели на librispeech-100, который в 10 раз больше не представляется возможным."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QreOm7yp9GKJ"
      },
      "source": [
        "### Задача Speaker recognition"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8nAVkpTy9Nzq"
      },
      "source": [
        "#### Получающиеся метрики"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gpAXJH-ixpN0"
      },
      "source": [
        "Проверим утверждения авторов о том, что их модель показывает более высокие метрики качества, для этого обучим baseline, в котором в качестве первой свёртки используется стандартная 1d свёртка, а затем заменим её на предлагаемую авторами свёртку."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "jQkvWoPzB4jX",
        "outputId": "78f5a952-0fd9-422f-cd0e-f8684501a723"
      },
      "source": [
        "!python main.py # baseline run"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "id zkoioxdj\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mbch\u001b[0m (use `wandb login --relogin` to force relogin)\n",
            "2021-06-10 22:36:29.355297: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.10.32\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mlively-elevator-23\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet/runs/zkoioxdj\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in /content/sincnet/wandb/run-20210610_223628-zkoioxdj\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run `wandb offline` to turn off syncing.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 247
        },
        "id": "rCWpRVe13zvk",
        "outputId": "a8e2a0c5-b92a-4d8e-efba-522020348156"
      },
      "source": [
        "draw_three_images('sincnet/report/baseline metrics/train&test loss.png', 'sincnet/report/baseline metrics/train&test frame accuracy.png',\n",
        "                  'sincnet/report/baseline metrics/test speaker accuracy.png')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABz8AAAEdCAYAAACSd6rVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3gc1fXw8e9sX2nVq2VJltXc5CL3XsAUGxdKSCAhdEjyppLyI4RACKGZXk0zvYWE6op7x1WyjTFusmz1uqrbd2fm/WNl4Y5tTIid83keI6HZuffOzN3ds/fMvavouo4QQgghhBBCCCGEEEIIIYQQQpztDN93A4QQQgghhBBCCCGEEEIIIYQQ4kyQ5KcQQgghhBBCCCGEEEIIIYQQ4pwgyU8hhBBCCCGEEEIIIYQQQgghxDlBkp9CCCGEEEIIIYQQQgghhBBCiHOCJD+FEEIIIYQQQgghhBBCCCGEEOcESX4KIYQQQgghhBBCCCGEEEIIIc4Jpm/Yrv9HWiGEEEIIAcr33QBxSiROFEIIIcR/isSJZxeJE4UQQgjxn3LMOPGbkp9CiLOcruvouo6iKCjK2fF58dA2AyfVbl3XO3+eTccqhBBCiHOXruvh0d8zHJ+cTqwkDnfwHAISOwohhBBCCCHEOUaWvRXie6brOnV19VRWVXUOwJxWGfX1VFRWHrWtsrKKP95+BytWrvq2TT3lNrW0tLKvtPSUj6uqupo/3n4Hy5avOKX9lq9cyR9vv4PKqqpT2k8IIYQQ5zZd12lqbmb/gQPfKt5qbGykvLzilMr44MOPuPOuv9HS0npa9R5LY6OTP//lr8ydt+CMlXk20nUdVVXZW1KC2+0+5f3fee99/nbvfbhOY18hhBBCiP8Wuq5TXV1DbW3daZehaRoVFZU0NDaewZYJIcT3R5KfQnzPKior+dXvbqN4y9bOvx28E/3QO9KP9feD26qqqvn1b3/Pps1FR5Xv9njYXFRMbV3daZd/on2Op62tnT/efgcLPlt02N+PVceR5ZpNJlJSkomMjDzhvkeqq61jc1ExHo/nlI/n0Jmjp3qsQgghhPjv1tTczG1/+D+WLVvR+bdTjRFq6+r49W1/4PP1649Zx/H2KyuvYMvWbQSCgdOKL45Vps/vo2jL1pO64et04rtvEw+dTKz3bes41ILPFvKnP99JS+vXyeVvasPBumJioklOSsSgGI6778ke4zftfzadUyGEEEKcXfaW7OOXv/0dX+3a1fm3U40XvtzxFb/67W2Ulu4/7XZ821jkdOMcibGEEMciy94K8T3y+XzMnjuPqqpq1qz9nPSuXenXt4C1n69j6bLlmMwmLpw4kSGDB6EoBrZu28ZnixbR3NRMekY606dOITUlhdlz51JVVcXn69aTmZHBoIGFx1y6S9d1XC4XCxYuYsvWbURFObhw4kQGFg5AURR27trF3PkLqK9voEtqKlMumUR+Xh5+v58Fny1kU1Ex6DoDBvTnkkmTiIiwH7MeVVVZuGgxe0tKMBgMrFi5ivz8PJavWEluTg6fLVrMeePHkZ+fxyefzmZf6X4iIyIYP24sY0aPwmw2k9alCw5HJKFQiHnzF5CQkEBLSwufr99Abk4OV/7gcqIcjuMuUabrOjU1tcyZN599paVkpKcz9ZLJdOuWiaZprFu/gWXLV+ByucjJyWb6tKmkJCfT0NDIx5/OpmTfPqIcDsaPG8uokSMwGo1n/PoLIYQQ4rsXCoVYsGAhpfv343BEkr82j5EjhrNpcxELFy1G03XOGz+u8/3+q527mDd/AfUN4Xho6iWTycrqxtx5Cygvr2DDxk10z8pi6JDBnXGIruuUlZcze848KioryczM4NKpU0lP7wqE76TfsGETxVu2YI+I4EdXXkG3zEx27d7Nps1FTLr4IhITEtiydRu7du9m6pRLcLtcLF66nAH9+7Fq9Wqqa2qZeN4Exo4ZfdTxLV66jNbWVi664ALi4+M6twWDIZavXMmaNWvx+Xzk5+dx2aXTiY+Lw+PxsGDhIoqKirHZ7Zw3fhwjRwzHYDBQUrKP2fPmUVtbR052NpdOm0JKSgorVq7C2dTEtCmXYDKZWLh4CZqqcfFFF7D9yy/ZW7KP2NhY1qz9nOuu+Qk+v5+58+ZT39BAclISUy6ZRK+ePdF1naLiYhYuWoLL5aJf375MmTKZ3bv3sHvPHi6ZPIm42Fja2tqZO28++fl5x4xvK6uqWLBwMU1NTXz8yWyuuPxSioq3EGG3c6CsjLb2dm66/jo2bS5i5eo1eDwecnNyuPzS6SQmJhAXG4vP58NoNLDjq51s+2I7fXr3YsFnCwkEAlw6fSp9CwoOq1fXdb7Yvp35CxbibGqiS2oq06ZOITcnG03T2LS5iMVLluL2eBjQrx9TLplEZGQkra2tzJk3ny+/3EFMbCwXX3gBA/r3Y9fu3Wwu2sLkiy8iISGeLVu3smv3XqZNmUx7eztLl6/ojJ8njBtLz549+PiT2ewrLSUiIoLx48YwdvRoDAYDFRWVfDpnLuUVFWSkpzN92hRMJhNLli5n2NAh9MjPQ9O0juumcvFFF2IyyZCAEEIIcTZzu93MnjOX2to6lq9YSXJSIj3y81mxcjUrV63Cbrcz6eKLGNC/HwCbNm9m8ZJltLa1kdWtG5dOm4rD4WD2nLnUNzSwaPES4uJiycvNRVEU/H4/c+bNJyM9naFDBrP/wAFWr1nL+HFjyczI4PN166mtq2PyxRexuai4c6wtq3sWl0+fjsFoYOGiJQweNJBePXug6zqLly4jEAgw6ZBYRNd1Svfv59PZc6muqSE+Lo5JF1/IgP79Afjqq53MXbAAp7OJHvn5TJ82hYT4eLxeL58tWsymzUVYrVYmjB/L6JEjqauvZ/GSpYweNZKc7GxK9+9n9ZrPueiCidjtNubOX0BuTg7Llq+gZ88enD9hAnPmzWP7lzswGAwMGTyIyRdfhMVioampmdlz57Jz124S4uOZdPFFdOuWybz5C+ielcXwYUMBWLd+AxUVlUydMpmIiIjvpT8IIb4mMz+F+B6pqkp9XT26Hl5Grbm5meUrVnL/Qw8TFxeH3Wbnnn/cz6bNRdTW1nLPvffhcXsoLBzAF19sZ9Yrr+H1eqmtq0fTdRqdTpqamo5bXzAY5NEnnuLNt98lJzubQCDInXf9jdVr1tLc3MI9/7if2to6Bg0spGRfKc+/+DJuj4fPFi1m5osv0z2rGxkZGbz3/r9ZvmLlcevR9fAyvKqq0traSn19A5VVVbzw0iwemPEwO3fuRNd13nn3n6zfuInCAf3xeL3c/9AMysrKaXQ6mfXKa2z7YjvBUIj3P/iQRx9/kn2l+4mOjubV199gztx5Jzy3jY1O/vzXu1i7bh19C/rw1c6d/OnPf6G8ooK9e0u49/4HsVgs9O1bwKrVa3jn3X8SDAZ5ZubzLFuxgsIB/dF1nZkvvMSBA2VyV5cQQghxljoYl2iaRnNzCw2NjWzctJl77r0Pq81GfFwcD8x4hBUrV9Hc3Mw9/7iPuvr6jnhoH8+/9DLtLhf19eGYzelswul0HlZHfX0Dd9x5N9t37KB/v75s2LCJ+x96uHM51XaXi9Vr15KZmcnyFSt5+tmZhEIhvtzxFS+/8hoNDQ0AbNy0mdfeeIv29naqq2t4adYsnn72OSIjI6mtreXBGY9QXlHRWa+qqXz48Sc8+viTBIMhoqOjDmvXzl07eXnWq8THx5Ofn8/7//qAd9/7J6qq8sLLs3jltTfIzs4GXee+B2ewZes2yisquP0vf+XAgTL69S1g2fIVPPLYk3i9XhYuXsK7772P3x9A0zRmz5nLx5/ORlVVNhdt4ZnnZvLsc89TVlaOP+Bn5gsvUt/QwKDCQrZ/uYOHHn4Uj8fDxk2bufPuvxNSVXJzc3j73fd4/Y238Pl8vPjyK6xbvwFd1ynesoUXXp6Fz+875rX1eDy0tLSgqSqVVVV4PB4+/mQ2Mx55jNlz59HS3ML+A2W88NIsoqOi6N2zJx99/Amvv/UWuq6zfMVK3n7nPbxeL19s387M51/gjbfepktqKsVbt/HQw4/hcrkOq7OpuZlnZ75Au8vFwMIBbNy0mUcff4JAIMC69Ru465570YGc7O688dbbvPnOu/j9fh5+9HE+/PgTevXq1dnP9u8/wPbtO5j16ms0NH7dB15/8y3a211UVdeE4+eHHuarjvj53ffeZ/2GDRT274/P5+P+Bx9m/4EDNDY2csdf72bbF1/Qv19fNhcVc/+DM/D7/Xz08Sf88/1/oWkajY2NzHz+RfaW7MNgkOEAIYQQ4mwXCoU649S6+npaWlqZv+AzHn38CVJTU0GBv/7t72z/cgdlZeXc848H0HSdAf37sWHjJl5/8y28Xi8NjY3ouk5tXR3t7V/HP0ajkfXrN/Dm2+8SDAZZuGgxM194iaXLVuD3+3ntjTfZuHETlVVVzHzhJaxWKwUFBcyf/xkvzpqF2WRmzrx5vP3ue2iaRkNjI8/OfJ59paWHxSIH48D9Bw4waGAh+w+Ucd+DM2hpaWH3nr3cfudfaWhopE/vXsyZN4+nn51JIBDg5Vdf46VZr9A9qxtGo4EHHnyYzUXFVFZV8dKsV9lbsg+AkpJ9vPzKq1TX1NDS0sorr73BPx54kKLiLYRCIRYtXsync+bSs0c+sTExPPHUM6zbsDEcbz00g3kLPqOgT29qasLjs/X19axYuZqXX3kVn8+Hz+dj1quv8fn69ZjM5v94PxBCHE1u8xTiexQREcFVP7qSFatWM33aVMaNHcMf/u/PxMfHMWrkiM4ZigsWLuKaq6/C5XbjbGrCYrHwi5/dQvfu3YmNjeXHV/2QFStXMfWSyVww8fzj1rdvXykrV63m5huv5ydXX4XH4+WXv/kd//rgQ7pnZdHW1k5zRAuKonDDdT8lK6sbEXY7TU1NBAJ+GhudDBk8iLFjRpPdvftx6zEajfzohz9g7vwFjB41gh9eeQWbiorQNI3xY8fy81tvxmq1UjigP+UVFdTV1WM0GPB4PDibnMTExBxVZkJCAj+/9WbaXS7WrFnLgbLyE57bVavXcOBAGU89/igD+vdj9KiR3PqLXzF/wUIGDSzsDOx69+7FH277HTnZ2QA4nU48bg8ul4vzz5vAdT+9hvT0rsedYSqEEEKI/24mk4mrfnglCz5byITxY7l02lT+1pH4HDtmFGaTmY2bNjNvwWfk5+eH46Hm5o546FqysroRFxvLVT+8kkVLljLp4guZdPFFneXrus7GTZuoqq7mmScfo1/fvowbOwa324PVagXAZrPx81tuJqt7Frv37KG0dD9+f+Ab265pOmPHjOaaH19NclIS9z/0MA0NjXTpkgrA+g0bqaqqZsrkSVz9oyuPmsXXq2dPHn7ofhobnez46isUg0JFZRVOp5Nly1Yw8bwJ3HTDdfh8PvaW7KNbZgZz5y+gpbWFhx+8n5ycbMaMGkUwFMR8EoM4oZDKL35+K+PHjsFsNnP3nXdQXVNLeXkFRqOR+vp63G4P8z9bSHRUFLf95ldERUUxauRIIiMjiI2NJTMjg+UrVjLxvAmsWrOGtC5d6FfQ95ixWH5eHuPGjuHfH9Tz61/+gqTERAAsFgtPP/4YqakpKAZDxzloZOeu3SgGA5WVVWiadvT51nWuvOJyhg8bSlNLM/MXLKTd5SIq6uukclxsLP+4525qaus4cKCs47ga8Hi9zFvwGfFxcdz2m18RGRnJ6JEjcUQ5qKisZN2Gjdx0w3X85OqraGtro6y8gpSUlJPoAxpjx4zm//38VqxWKwMLB1BeUUldXR2GHR3xs7OJPXtLKK+o4KnHH6FwwADGjxuLy+Wia1oao0aNYPXqtTidTWzZug23x815E8ZLfCuEEEKcA6Kjo7nyB1fw+foNXHXlDxgyeDC/+u3v6NIlleHDhuLz+fh83QYWLVnCRRdcEI4dGp30K+jDb3/9S7KzuxMbE8Nll06nqHgL117zk84V4iA8xjdy5AhemvUqVVXVbN22HbvdzrYvvmDMqJEcKCtn+tQpdM/KYsaD99HQ0MiePXtRDAaqqqpxOCI5b/x4Zs+ZS01tLcVbtuJ2e7jg/PMPi0VsNhu3//H3VFZVU1VVhdFooLm5hXaXi6XLlxMMBLntt78mrUsXRo0cgUEx0NbWzpKlyxg/biw333gDfr+faVOmkJmZwd6SkhOfOF2nR34+99x1J3a7nUAwSEFBHxobnTQ2OlFVlerqGkr376eoqJjf/uZXXDZ9GlMmT6aqupquaWlceMH5PDvzBUr2lWI0GjlwoIzbfvcbzLKyhhD/FeSZKMT36NA3eQUIhkI0Op3U1NTyyONPoACKQUHXNNIz0rnphuv4dM5cnn52JgaDwsgRI7jj9j917H10mUdqdDoJBoOkdw0vwWa320hNSeZAWRlxcXH8/Jab+NcHH/H8iy+j6zoDCwdw1513cMnkSRwoK2f1mrXMW7CA6OhobrjuWq64/DKOVdvx2qAoCr169sRut6OqKm+98x4LFi4kq1s3HJGRgMLxJlg6HJGYzWbMZjMmk4lg4MQDhjW1tZjNZpKTklAUhYT4eByOSOob6ino04cfX/UjFi9ZysZNmzGbzVx0wUR+95tfcctNN/DSrFd5/98f4vf7yUhP547b/0Tfgj4yQCSEEEKchQ5//1ZQVZXGxkaampp44slnUBQFTdNQFIXEhHh+dstNfPDhxzz/4sug6xQWDuDuO//CkUHPoeU2NDoxGAxER0ejKEpnrHWQxWLB4XCgdPweCARQNfWotuocHggpSjjZpigKNpsNCK/kcdC+faWYzWYanc6jVqnQdZ3de/Zw/4MzQFHo1aMHZrOl42sQ3Lg9HuLj41AUBbvdTr++BeFjaWjEbLYQFRWFoih065bZWd6R5R/5N6vVQo+8PGw2G61tbTz6+JPs3ruXnj16YLFY0NFRNRWn00lkZAQ2mw1FUejZI7+zjPMmjOPfH37Ezl27Kd6ylQsnTiQqynHUuTqR5ORkUlNTMJvN7Nm7l3vvexBVVenVswcWy/GTuCajsfMmvAibHTUUIhQKHfaY5uYWHpjxCBUVlfTq2QOr1YLH60FTVZxNTTgckVit1vBx9ewBQPGWLQQCAeLj44HwIGXfgj7HbIOmHdkHFHr3+jp+fue995m3YEE4fu74CoiDq8iE+2AMiqLQNS2ts4wLJ57P/AUL2bR5M2vXrSMnO5v8vFyJbYUQQohzwJHv54FggKamZpqam5nx6GMogM1mJRQMkZebw7XX/IQFny3ksSefxmQyMWHcWP70h9uOW6aiKAwaWIiuaaxYtZrKqiounTaV5StXsmLVKswmE4WFA6iorOTv/7gfj8dLr149sVos4TE+RWHi+RP48ONPWLV6Des3bKR3r57k5mQfVo/X6+Xp556neMtW8vNyv14OV9NpbGjEYrUQGRGJoiidExgqKivxuD2HxcsHY9ojHbWimwJ5uTk4HOE4c8mCz5j12uskJyWRkZ4enpWq6zQ3txBSVeLjwnFzXFwscXGxAIwaOYLX33iLVavXYLGG4/1hh3w1hhDi+yXr3AjxX0LVNMwmE8mJiWRmpPP8M0/x/LPPcMtNN/LTa36Mz+slIz2d++/9Oy88+zTjx45l9Zq1lJWVdZahayf+4u2U5GQsFgv7Svej6zrtLheVVVWkdUkjFAqRmJjIXXf+mZeef5apl0xm46bN7N69JzwLcsJ4XnrhOR66/z7sdjvzF3xGMBDA4/HQ3u46bp2H3lWvAFZbeDDI2dTE3PnzGTZ0KI8/MoPRo0ehKMoZ+27NjPSuBIMBKiorw9//WVtLW1s7mekZtLva6dWzB48/MoNnn3qcfn0LWLx0GTU1tQDccN21vPrSC/z217+kqrqalatWn5E2CSGEEOL7pWkaRqOR5KQkkpISeeqJR3np+ee49eYbufnGGwiFQiQnJXHXnXfw4sxnuWTyJDZt3szuvXsOKwMOH0DpkpqCqqrUdXydQVHxFl57402am1uAjrzpIWMgesd/TCYTuq7j9frQNP2o5XS/yehRI/nlz3/G+g0b2bBx01Hx2IKFi2hqauaBf/yd3/zq/2G32zAZTURHRxEV5aCqqhpN02htbeWlWa+ybdsXdOmSit/vo76hAV3XWfv5Ot56511cLhdmk4lgKEQoFMTn99Pa2nZYfUajCYvFAsDOnbtYu249N153Lff9/W/hpcgMRswmE6kpKbS0tNLW1o6m63w6ey6fzp5DKBRiwvhxaKrGK6+9jsvlZsL4caiqSltbG4Hj3Pymo6MfkjS02aydy6gtXrKMuro6/nHP3dz2u98QGRl5wnhTOeSXI6NbXdfZsm0bRcVb+OUvfsY9f7uLtLQ0jEYjJpOZ1OTkzhkKmqbx0SefMmfufKKjo7HZrFRWVXV+L/2LL8+iZN8+TCYTmqbh8/k6l4I7rD0KncnUpuZm5sybx9DBg3n8kRmMHT0KhfCMjNSUFDRNpa6uLtzOrdt49fU3aWpupkePHuTl5vLBRx+zZcs2zpswvjOZLoQQQohzh6ZpWCwWEhISyMvN4aWZz/HsU09yy403cNWPrsTt9pCbk83DD93PzGeeYuiQwSxbsZLqmppwDKTraJp21Nhil9RUcnNz+ODDj8ITCC68AE3T+PiT2fTs2YOU5GRWrFzFgbIy7rrzDm7/4++JjonGaDSgAFndujGgfz8++vhTdny1k4svvLAzZjxo/4EDLF+xkh9cfhkzHriPXj17YugYJ+zSJRWv10tTcxO6rrN02XL++f6/MRlNREVFUVNTi6ZptLW18/Irr1G8ZSsmoxGFcFJV7/iqsMNj5fANgAB+f4CPPvmU5KRknnzsEa64/LJw3SYjiQkJmM1mKquq0XWd8ooKXnz5FQ6UlZGclMTw4UNZunw5i5csZejQISR2rEQihPj+ycxPIb5nkRGRWCwWFi1eQkJ8PNOmTuGBGQ/z2JNPYzAY2LhpE3+5/f8wm8089cxzxMREM3BgIQfKy8nqlkmX1C64PW6sVgtLli0jMSmRcWNGH7OurKwsLpx4Pv/+8COcTeEZpi2tbdz2218TCAR4/qXwjM+Rw4ezr7SULqmpZGZmsGHjJma+8CKjRo4kNiYGt9vN2DGjMRqNPPHUM5TsK+XJxx7uvFsKwGqx4nBEsmHjZubMnU9KSjLw9YBShD2CpMQkNhcV8ezMF9hcVIymaXh93lO+u/9YRo8exdz5n/HYk08xeuQIioq3kJHelUmTLqKuro6HH3uC7O5Z9OyRT01NLT3y84iJjeHVN95k0+YiLjj/PJpbWrBYLPTrd+y7xoQQQghxdrDZbERERrL283VkZKRzyeRJbNi4iceeeIroqChWr13Lb371S+Lj4pj5wksoisKI4cPYV7qfLqld6JaR2TlDcvmKlaSmpHDehPFA+G74oUOHkJebw1PPPMvwYcNY+/k6EhMTuPKKK07YruzuWZjNZl5+5VW6d+9OcfHWY66qcfz9uzPlkkksW7GS1958i/79+xF9yBKtGV3T8Xi9vPb6mwSCQRobGslITyc2NpYpkyfxwUefYDSZcDqdfPnlDgYPGsi4MWOYPWcujz7+BAMLC1m+ciV5ublcecUV9OiRz/KVq3jiqWfQdB1nUxORkZHHbFtCQjyRkRHMmTefXbv3sHbdOgLBIMFgiKlTLmHdho3c9+BDpHftypKly5g65RIMBgMZ6Rn079+PVavXMGhgIdndsygp2ccdf72bm2+6gUsmXXxYPfFxcfh8Pt54622u+tEPO/769VlM75qGP+Dn9bfeDn+PVW0tcbGxx1z29mQkJyVitVr58ONP2LhpM8VbtmAymVDVENOmTmFTURH3PfAQXVJTWbJ0GVdcfikXXTiR88aP5+NPZtPe1s6+0lKqqqu5YOL5ZGd3x2Qy8dKsV8nq1o2t27YdXWnH7IEIu53kpCSKireE4+fiYjRdx+vzMmTwIHrk9+Dp52ayuaiYz9etJy4ulit/cDk2q5ULJp7H408+TUxMNKNHjZQZCUIIIcQ5JCrKgclkYs68+TgcDqZPm8ITTz3DE08/g8/nY9sXX3Dv3+4mEAjw2BNP0aVLKn0LCqisqiI3J5ukxETaWttQDAY+/PgT7HYb/fv16yzfbDYzYvhwNhcVM3b0KLpndaN7VhYbN21m9KiRmEwmuqaloaoab73zLjabjfKycjIyMjoTspMuupC77rmX5OQkhg8belQsEhsTS3R0NIuXLqWhoYE1n3+Oqqn4A34uOP98Fi5awoMzHqF3r14sXb6cwYMGcfll05l6yWT++a9/M+PRx2lpaWHbti8Y0L8fGenpxMXF8a8PPmT/gQMUFW3BcFT8E/5/k8lI165d2bBxI8+/9DIlJfsIhkJ4vV66d89izOiR/PP9f9HQUM/OXbtxNjVxyeSLMRgMXHjBRBYuWoKu6/zhd7+RGEuI/yLGe+6550TbT7hRCPHtORyRJCcnE/AHyOrWjVEjR9Cnd2+am5ux2+1cf+1PGTF8GHGxsQwZMhhV1WhyNtG3oIBbb76RlORkHJGRpKSkdJSRSWZGRuebbTAUxOvxUjhgAJkZ6QweNJCMjHTq6urpmtaFn996CwV9ehMZGcmwoUMwGo00NDaSl5vLz265icyMDPJyc8nPy6OpuYlQx4DVFZdOx2KxUFNTw569e5k86eLD7tqyWMxkpKcTCAZJSUmmW2YmwVCQIYMHkZiYiMVipl9BAaqqYrfbuPKKy7HZbHTLzCQ1NQWP10PhgAF0TUujtbWV7Oxs+hb0Qdd1Wlvb6NmzBz3y8w8LKjxeLxazmWFDh5KUmMiokSOIjIykoaGRQYWF/OyWm0lOTiI5KYmBhYV4vT5aW9sYPnwoN15/HfHxcQweOJDExATq6+s7lvf9KcOGDjnsS9iFEN+Zv3/fDRCn5J7vuwFCnCybzUrXtDQCgQBpaV0YPXIEA/r3p62tFYPRwDU/vprx48cRHRXV8b5vpNHZSH5eHj+75SYyMtKJiIigS5dUAv4AGRnpZHXr1hmH2G02RgwbhtFoxNnUxIjhQ7n5huuJjY2h3eUiJiaGoUMGY7VaaWtrIzExkSGDB5Gamkp29yw8Hi+ZGbDan74AACAASURBVOlcdMFEIiIiGDpkMAaDEX8gwOBBA0lNScHn96NpGkOHDCYmOgaXy0W/fn3Jy80lI70rzc0tZKSnk5CQAISTstnZ2SQkxOP2eBkzaiR9CwqIiIigf79+FA7oT2ZGBnV19SQnJXHLTTfQr29foqIcjBg2FFXTaWlpYfzYsVx37TU4HJHk5mQTHx+Hx+tl2JAh9Ovbl5SUFPr364vX6yXCbmfokCHY7TbiYmPJy83F4/GQkZHOlMmTMBmN9O7Vi7y8XAYNHEh7u4tgMMj06VO5bPo0zGYzBoMBt8fD5+vWc83VV1NQ0Ad/wE/xli3k5uTQIz//sGubltYFm8WKpusUFPRG01SyMjPpW9AHg8FAVlY3kpKScLvdjBg+jMLCAdhtdvr364vH6yUhIZ7BgwYTDAaxWCwMHTqEKIcDl9tDZEQEw4YOIaJjVoCiKCQmJpKdnYXb4yE3J4eLLpiI0WikoE8fcnNzGFhYSFtbO6FQiMsvu5TpU6dgtVoZNLCQpKQk6urq6J6Vxc9vuZmc7GySkhLpnhXuA926ZXLhxPOJiIhg2NDBGBQD/kCQIYMHkpSYiNkcjp9DqorNdnj8nJeby4jhwzCbTDQ6nQwbOoSbbry+cxm4CHsECxYuYvDAQqZPmyqxrTgbSZx4drnn+26AEP9LoqOjSUhIIOAPkJuTw+iRI+jZoweNjU5ioqO56YbrGVg4gMSEBAYPGkggEKS5pZnBhQO5+cYbiI+PJy4unri4WIKBIPn5+SQnJ3XGuoqiEBsbSygY5PzzJ5CTnY3NaiU6OpopkyfjcDjIzMigS5dU3G43gwYWMmzYUKxWK/369sVqtZKUmEgwGGDc2DEUDuh/VJLQEeWgV88e+Hw+kpOTuHT6NIxGIz3y88nJyWbY0CH4/QFcLhcXXTCRn1z9I+x2O3369CErqxt1dXUkJiRw8403UDigP44oB7179cDvDxAbE8v0aeGYbMjgQTgckbjcrs4kqcFgoG9Bn44VOXSmT5tKUmICyUlJ9Ordi6GDBxEXF0ddXT15eXn8/NabO8deoxwOli5fTkJ8Ajdefx1Wq/X76AJC/K87ZpyoHG+pyg4n3CiEODMOfR4e/N6eQx0aEBz52NP5+/Eee7LlHNrOouItLFy0mNv/9IfO9fiPt8+J2nEyjtzvZNp+KsdzMtuEEN8pebKdXSROFGeV7zLeOtH2E9V7rHYcy4ked6wyj9embyrzVM/BN7XhVGJBRVFwu91s/3IHb7/7HuXlFbw489nOQbQnn36WaVMvoW/B0StyHK9NJ9p+It/mnB7ruI6336nExKd7TlVV5audO1m2fCUffvwJf7/7r4wfN1biW3E2kk57dpE4UYj/sJONdU8n/vum8k8mTvmmOPqb9j9R+85EjHW8x58opi/Zt49Nm4t4adYrXHvNT7jhumslxhLi+3HMJ54seyvEf4Ej3xhP9EZ5vG2n8vczWb7b4+GG66495ncnfdMb/ukGBGfiWE/nHAghhBDi7PVdxlsn2v5N9Z5s3HG6scuZ3namYs5jbW9ra2fWa69jNBi57be/JjU1BQC3282okSPo3avXKbXpZLef6j6nW9/p9qtvU7emaXz08aeU7t/PdT/9CSNHDJdYVwghhDgHnWyMeaZiwzM13ngqj/kuY6xvevyxtq1YuZqVq1Yz+eKLufyyS0+pLiHEd09mfgohTtvJ3LUlhBCnQF5Izi4SJwohzihd1wkEghgMCiaT6RtX/RDfTNd1QqEQuq5jNpsBOY/irCUd9+wicaIQ4pym6zqqqqKqKmazGUVRJMYS4vtzzCefJD+FEEII8d9CPimcXSROFEIIIcR/isSJZxeJE4UQQgjxn3LMONHwn26FEEIIIYQQQgghhBBCCCGEEEJ8FyT5KYQQQgghhBBCCCGEEEIIIYQ4J0jyUwghhBBCCCGEEEIIIYQQQghxTpDkpxBCCCGEEEIIIYQQQgghhBDinCDJTyGEEEIIIYQQQgghhBBCCCHEOUGSn0IIIYQQQgghhBBCCCGEEEKIc4IkP4UQQgghhBBCCCGEEEIIIYQQ5wRJforvVVVrkE3lbnRdP2NlqprOqn0u2v0q26q9lDUHvlVZ7T71lNqn6zptR+yj6TqrS120etXTbos4M3Rdw/PlatQ25xktV21vwvvVWvRgAM+2ZWgB32m2T0fze9GDJ9dvgw0V+PZt+cY+qqshPF8sR/N7TqtdQgghzk1BVWfZ3na8Qe2MllvWHOCLai+tXpU1pS6004z1dF2n3a+iaqcWi/mCGr6TOKby5gDFlZ7TjkU1LRzjtfkkxhNhansTni9Xoetn7jml6zq+vZsJNdUQqC3FX/blafdZXdfQPG0ntb+uaXi+XIXqbj2tuoQQQgghhBDiULquE1R1atqC1LcHUTX9jOaGDiXJT/G9Kmn0M39n2xktU9V0/rW1mWaPij+oEVRP/8mz7oCbD7a1nNI+te0hnlhZT+iQ8Q5Vgw+3tdDoDp12W8QZouu41vybUFPNGS021FxL+9qPwgNKPjdwmv1O12mZ8wzB+gMn93g1iH4SiVZdDdG24j00T/vptUsIIcQ5yR/SeKeomXb/mU3e7az1saKkHU3XcQe0035b9AR1Ziytw+U/tUTSKxuc7Kz75vfHnXU+Fu8+/fdGVYcPJMYThwg11dC+6l+gndkbCtybPyNYsw9CAfSg/7TLCdaU0jz3OTjJAQbd7wFNkvtCCCHEuUjXv7ukw39z3f/NdF0npOl4AhrtfhVvUCMQCo9vhzQdVdPROs7dmTx/uh4uN6iGP7/5Qxrad5iUOl4bQqp+RpNh+iHn6vRvHtQ7z4+qhdsYVHUCIQ1vUMMdUGnzqbR4QzjdIRpcQWrbglS0BKhsCVDXHsTpDtHiDeE6zjX9Lp8PB9seUsM3Cbd6Verag5Q6/Wyr9lJc6WFXvY+K5gANriCt3nAbg+qx23a88+EPabgDGm0+lXa/ijug4jvkWA8ery+k8+oGJ3fOq+buz2qY/WUrp3Cv8ykxfTfFinOZrus0e1W2VHoBGJwRAcCeBj8GBepcIfp3sZEWY2Z7jQ+72cB+px+H1cCgjAispsNz7v6Qzpr9boKqzuCMCKKsBvwhnW3VXupdIdKizfRPswOwvcZLVWuQbvEWeqfYMCjh2aPba3xEWQ0MTI/g0OIjrQZsJoXq1iCN7hBtPhVPUGNwRgRxdiP+kM6WKi/NnhC9U+10izOjKAoAvqDGugNuqlqDlDUHSIs2s7XaS4MrRI9kG9kJFjQddtT4qGgJkBEXbtOmcg876/xsrfIyOMPeWd5Bqqazu95PqdNPWoyZvl3smAzhGRJf1fqIthkZmB6B3aywvynAzjofcXYjhekR2EzKUeX9r9A1jWDNPgIVOzElpmPt3o9gdQkoCsGGchSjGVuPoWAw4i8pxhARTbC2FEt6T8xpuUedt5CzimD9AcwpWVgyeoFiQG2uwV+6DV3XseUUYoxLRfO04du7CYJBrLmFGGNTQFPxH9hOqKECc1oelvQeneUqioLBEQcoeHdvxBgVR6ByN6b4NKzd+4HBiNragL+kCMVsxZY3GMUe1dm+UGMFvr1FGKMSMNijCdbtR/N7MCdlYIpPw7evGK29GXNaLpaMXihmGwa7A83TRqBqN4rBSKipBmt2IaaEtGP2F10N4S/7klB9GebUbCyZvcPnsbqEQMUujLHJ2HIKwWgmWL2HQOUejHGp2LIHoJgt3+l1FkIIcXJCqs6OOh+VLQFyEq3kJVrZVu0lympkX6Of+Ihw7OAOaOztiNHqXSH6pdlIizYf9v6g6zp7G/xsLvfQK8VGVrwFHdjvDLC73ofFZGBwhp0Ym5F6V4itVV7MRoWB6RHE2MJxW3GVlxZPiD6pdjLjzJ1lGw0KsXYjQU1nS7mHaJuR/U1+chKt5CdZUYCy5iBf1XqJizBR2NWOtSPe0XWdHTVedtb5WV/mZmJ+FDVth8d9drNCo1tlS6UHg0FhYLo93J5KLwqQm2QlymrsPM7q1iBf1PiIthkYlB6OYX0hjbX73Xg7YsQoq5EN5W4GdsReO2p9xEeYUHWdFq9Kizf8IW5wZgQOi7HzWNt8KkUVHoZkRuA4pM4mj8rWKi++kEbvg+dXhz2NfvY1+kmPsVDQxQbA7gY/+51+MmMt9E61saPWR7LDRGq0mdr2IHVtIXKTwtc6pOrE2o3kJVnZVuWlwR0iPcZM3zQ7RiV8Q94X1V7s5nAMvqvOR9cYMylRJnwhnS2VHoZkRmI2/m/GlofSdR3d78VXUoTmbcOWOwijIx5fSRGGyBiC1SVY0ntg7ppHsGovEF59QzEaseUPxWCLPLxANYRv90Y0bzu2/MEYHPGghjrix3KMsSnYcgeCyXx0/GWyoLma8e3dDIAtbwgGR2xn0YrFjkEHzdVMsHY/uhZCbW3Eljfo6zi1bEc4zuuSgyWjJ4qhoz9qKt6dnxM4sJ1A+Q7QNXRNQ22px95ndDhGLfsyHFfnh+s1RMaC0YSvdBsGWwTB6n0YIqLCMazZetg5DDWU4z/wZfi85A3GEJUAoSD+/dsINddi7VaAKSULQgF8pVtRW+qxdu+HKSEd3+71WLMLUax2ApW7MFgjUUwWgo0VaO1NWLr1wWCLxFdSjO5zY+nWB3NqNuj6YZ8RLJm98e3eGK7faifU1kiooQJrTuH/7OcoIYQ4mxw6wK7q4TG0kBa+Yc8b1PAGdVx+lTZfODlzMhRFwaiE41KjAYxKx0+DguGQ380dcWu0zYDZqKB07Hs67dd08IXCSQGT8ZD6FTiyyCPj8mOXGy7z4GoqB/dRFDra2VHWofsQ/swQ1HS8AR1XR1Ki2ROiyaPS6A513sBX0MVOz2QbadEm7BbDaR37kccf0r5OpLkDGi1elWaPitMTrtfpDq+alxBpIjHSRHyEkfgIE7F2I5FWA5FmAxaTgsmgYDjFppzudVN1Otvc6g2fI5dfw3BEvzn89/BPkwEsRgNWk4LVpGAxKh19Ltz+I49B1cNj4u0+lSaPSk1bkKrWcMKsui1IgyvU2X/MxnCZNlN4nNtmDtdjNxuwmw1EWgzE2Y3E2I1E28J9OMJsIMJiwGoyYDLQeR5DGgRCGu0d16TBFaKuPURNWzA8E88VosWrYjMpdIk2kx5rpluchfRYCymO8PWxmQ2d4+/fJr7SdR1Vh2aPyp4GH5srPHxV68NqMpDoMJHiMJESZSLJYSIh0kSMzYjDEu4XBz/DhFSdgKqHXxsCKm1ejSZviCb3133cG9QwG8PXxGoK7281KVgPuV5mY7ivhcv6+vXm4O++oI4vFP7pD2n4Q/phyctQ5+uVjqrRkTQMH6fBAGaDEr6WhnBdh11TswG7WSHGZiQuwtR5LWNsBqJtRiIs4etpM4f7nk54BaKgBgFVI9DRFr96cAWk8DlxukM4PeHnWrM3/Nw/mBvxBPRjrs5kMijYzOF2RVoMRNmMRFuNRNvD/x8I6Z3nxh/S8IU66gzp4XZ0JEMV5ejnh9lowGwM98GddT5UTcdiVHC6QwzrFklajPmo9nxbkvz8H6PrOp7gsTs3hN8kIywGDCd44Wr1qfx9YS390uydS6VdNTCOB5bUMqlXNGajwr3bW3hkWldm72ilvj3EhDwHn+1qo6o1yA/6xx72writ2kt+kpV2v8byvS7+ckEKr6x34gmGB4hmbXByVWEcQVVnyZ52RnePZNZ6Jz8eGEd8hImnVzcwLsdBSaOfVaUufjMmqbPsJXvayU+yYTYqvLyukWkFMdS7QqwscXHXhak8v7YRVdfJSbTyxMp6bh6eQEGqDUVRUBSwdLwYGRSFVzc20eQJ0SfVxnNrGrh6YBy+oM6cHa2My3Hw+kYnl/WNxWIKvwFaTcdIOgELd7exaFc7E/IcfLCthS+qvUzMj+LhZfVc1DOK1aUudtT6uLhXNI8ur+fintEsK3Gxq97PDUPjT/vaf590NYQe8B5/1ofJjGK2HvcNU9d1fLvW0bb0TSIGXkTb8neIcFYRaijHt7eIyCGT8e0tIlC5i6gxP6Tpg4ex9x6FKSkD5zv3kHjjDMyJ6V+XFwrh2jCHiP7n0TJ3JlFjfoilaz7Od+8lYsD5aJ5WnOtnk3TTIzR//DimhK4YrBE437uPpOsfxLVhNoGKndjyh9C6cBaRgydhSkgDQAv6aVv8Opb0HrQtehVDVDy2/CG0zHmGmItuwZyahfOfD2DvPRLN68JdvJiEH9+FYg0PwCpGU/if1U6ouRbnu/cSPf7HGB1xNH34KMaYJMwJaTR9MIP4K/9MyFmNv3QLjuGX0vTuvUQOnYJisuB8+26Sb30SJSLqyLNJ++p/49u7mYiCMbQufAV737FYuxXQ/PHjRA6ZjHvDXIJ1B7B270fL7GeIHDIZ1+cfE2qsxDHyMhk4EkKIb+HgEi/+E6xMYTUZsJwgIaXrOh9tb2FzuYeR3SN5ZlUDNwxL4IOtLai6zqjukbyxuZUmr0pWvIUHltQyuVc0JoPCJ1+28Oi0rp0JQQBXQGPx7nYKuth5aGkdvx+fjDeo8coGJxf2iOaLai9rSl3cNj6ZB5fUMSIrkiaPyup9Lv7v/BRmrmkEIDvRwuMr67l1REJn2Q2uEO8VN5ObaOXJVfUMTI8gO8HKg0vquOeiVNr9GrPWO5mQ62DdATfry9z8ekwSB8Mos/FgXGVgb6Of59Y0dsZ9q0td/L9RSTy0tI5BGXZcfo0VJe38YlQipo59Dh1kKHUGeHhZHeflRVFc6eGLai/d460UV3rJTbRS7wqxutTNHyck89amJvKSrNhMJhbubmdAmh2/qvHWpmamFkRT0xpizX43d5yfAkC7T+OtzfX0SrFhM399J167X+O+xbUMSo/AZFT4x+JaHrokjR21Pj7a3sr5eQ7e2ORkUq9odB3m7WxjQq6DVzY4md43ho1lHoZnRZIabWZvg5/V+1xcNySeGUvrmNonmoJUOy9+3oiqQY9kKy+uc/LTwXFkxlm4f3EdY3MiqWoJsqnCQ4rDxMZyDz8fmcAX1V6W7nUxrNsRSbuzkK7r4VUw1OPMvlVAsUZ0JgCPSQ3S9OHDGKyRmBLScL59D/FX3k7zh49i6zkMc0oWznfvJfG6+/FsX4lv5zoih07GV7IFf9lXxE75f4fFR4HqvfhLt4LJjHvTPBKuvQ/3hrnh+LHncNpXvU/IWYW1e7+O+OsS3BvmEKovI6LwApzv3os1ewAoCs737iXhx3/rLNtXUozaUoe95wic79yDY8Sl6GoI58a5JN3yGO4Nc/GVbsXeczhtS14nov95RAy6uLN9BrMVjCYUk4X21f8i1FJP5MAL8ZftoG3Rq0QOnkSgugTPtmUkXPN32pa/Q9z03+Ja/ylqSz0RA87DvWk+IWc1jtE/6LxRIVhTQvOHjxI5aBKBuv24ixeRdP2DtC56NXys2QNwvn8/8T+4HXfRAtQ2J9ZuBTjf+wfxP/g/2pa+RUJKd4xWO55tyzEnpmOIjKH5o8eInvATtPYmmj96HGv3vuG4/J17SLpxBsGaUtqWvU3EwAtpW/42EQXj8O3ZCArY+4zBvWEuBlsk1pzCM9HdhBDie6PrOjrhJFhI0/GHwgPxLn944Ds+wkiUzYjFGE5wnE7yR9ch2DHjTNPpTAaYjeExMgOHJNmOKP/g/poOATXctoMzgA7ePBaeRRRuv6p/nSDQ9K+Hi0IdA/YtPpV2n4YroOLyhwfWD84cAjAZCI9dnsxhHixfB52vz+PBnwfbD2C3GEh2mMiItZCTYCU7wUJGrIVEh4lISzjhc/C9Tye80psnEE621LaFKG8OUNYcoLw5QE1bOOlyMEl1MIERaTHgsIb/RVqMODombgRCOt5QOKngOyL54guFEw2BjuM/mHwzHJLYNRyR4PUGw8mtVl94Fpc/FD7XZoOCxaQQbTMQZTUSVHWWl7gIhHSSHCbyk630T7PTJ9VGeowFh9VwWJ867NiDGu2+cJKppi1EVWuQqtYAje5wAq3Np3XWDeE+ZTMpRNuMRFkNKEp4Mk2rT8UXDF9fHTqTQtE2A7F2I4kOE9ZjfD4yKB2JxUOO325SsHec64MJQLs5nPQKJ7rCPz0BDafn8MRfXXuoc/KML3gw0dzRzQ5Wf7DvHNKPDnakcJJH6Uy02Q+pP3zdw8k7d0Cjui1IfXuIVp9KQNUxGxWirAaSHKbOiUBWo4Kq05l884a0jr6h0+pV8Yd0/Gp4mzsQ7h8h7eB1DidjbWaFKKuRKJsBu8lAi0/F6Q7hCoRfO4wGsJsNxHec57wkK4mRJjwBjfLmAJvKPSzY2YY3GH5slNVIssNEeqyZzDgLmXEWEiNNJEQYibIasZu/Tvge83UCcPs1DjQFKK7yUlThYW+jH29QIzXKRHaClZCqU9YUYFuVh3Z/+Lg0TcdiCl/TuAgjcRFGdJ2v+3hAw6+GX4cOJhajrOFEsNGggA5qx2fxgzcFhF9Twq97mhZum0Gh4/qF+4nFpGDtSJrazQpxEUbsZnNngvtgnzrsd1M4yXnw8FVND1+/Q5/Xh1xLX1Cj3R/uEy6/htv/9bVUNTAbw4lDm1kh0mzonD15cGbmwcdpuh6+2aPjNcBsVHBYwwnUWJuRbvEW4iPC1yohMpzIdlgNGBUFX1DrvEGiteN1u9UXvgmg1adS3RbEF9I6Xz/CSdvwDSO2KDN2c7i/W03KoU+VzpsgAiod51uj0a2GZ/jqYFDC1+B0vybnm0jy83+M06Pyx0+raD3O9xJZTQr3T04jJ9F6zO26rrO9xkeMzcj1Q+JRddhY5gYgMdLEtYPDybn1B9y0eFXQ4cIeUVzSO5poq5F1HY891OD0CH46OJ6gqvP7T6soawrQM8WGw2qgxaNiUqCyJUCMzUi7T8VsVLhpWAJdY8y8v6UZhyV8R1ZqlJk1pS5q2oLHbHt+kpWrB8bR4Apx57waylsCrCtzc0W/WIxK+Mm6ZE87Bam2jnNhICfRismoEGExsKKknal9YlBQSHaYWbirnaGZEbT7VYwGhRuGJpAWbabdr5IYaaJXiu3o2Yaqzmc727l5eAJ9u9gYnBHJXQuqGZQRLscX1JncK4Zou4GQGr6bzR/SmNonhljbCQZs/su5N86jZcGLx11ey5pVQOL1D4DxeHd46Lg3f0bU+B9jLxiLvfdIQs11BOvLsReMxTHqCsxdcmhb9jboGgaLnZgLrsfgiMdfUozaVH1Y8lMxmYm+4Hqs2QMwxXfBte4TLOk9cAyfhiEylmDAS6ilDs3vQfO60NytWLv3x5pTiK6ruDfOI2LA+aDrmJPScW+cS8xFNx/dbAWixvwQa3Z/NFcLgeq9hBorOtuomK14ti4hWLMPa1ZfAIxxqeHZADkD0XUdU3ImUeOuAl0n1FyLMToRtbkWVBW1pe6w6ozRSURPvB4A71efo7qbMRyR/NQDPjxbl5Bw9V2YkrthyexN0wePYE7OQvO0o4eCRI39IYaoeNS2RjRvO7oaInr81RijE0/uggshhDihF9c5Wbb3+EutXtwzmltHJBx30Mof0llR4uL345PJSbDQJ9UGKOjoTO0T8//Zu+84uarz4OO/c8v0srO9aLWSkIRQA4EoAtEx1RhMYlyxE9uxndiO4zhOffPa8esSEreUN2/sFIwd2xgbG4wNpiMwoiOEQAV11FbbZ3b6zL3n/ePOzM5KK7FqoBXP9/PZz+7Ozs49t8ydc+9znudw/qwwGtjYX2BGo4+WiMVNSxvRwJPbMyRzzrjgZ9Rv8rFlzXTErNrAtkvnRrluYQO+yg2KtSMlSmVNuuDdZDu7J0TYZ9CfLvP0axl+t9KfigfM2uCzfflMg5uWNtIetVizJ8eOkRIrt2VoCpuYhqIzbnPnmhHed3qCtqiXnTqr2U9D0OS0riA/fH6IqN+7sGyPWvxsS5pdyRLpgndT7IxpIc6fFaEtatMetVncGSTsG8vAfGxzmnNnhLnxtAZGCy6v9OYZyTmc1hXkvacnGEiX+dt791AsH/jCa16bn/csSdCbKvOF3+yhULl4/qfH+zilNcANixuw9hnWfemcKC0Ri50jJUbzLkNZh/s2jPKBMxKcNT3E0u4QgxmHHzw3xIfObGRJV5DTp4VIH6QUcVPI5D1LEgRsg9GCQyxgMpR1alVRdo6UWNwZ4L1LEmRLmhd3ZemK2/zDw32kCy4rNqe5aHbkkEfRH490IUvfdz9LuX/HxE8wDJpu/GuCC8474GuUh/ZQHthJ6yf+BeXzY3fOBsNA2X5il34Is6GVwpbVlAd3gdYE559H5Lzfxdc1l+T9t4B2QY29p3zT5hG/8qNgmAzc8lcUX1uL3Todu20Gbj6Nsv2U9m7HN21epf9VJHrBuzGiTeQ3PoebHsEIxwFwUgMUtq2ZsN1WUxexSz+ILhfoW/8kTnKA9NN3E1x4AWgXq6Wb9NO/IrTkbV7A0zDxTZuHtekFbx2ByNKrCC29ivLebUTOuwHlC6CGeykP7BwfUNaa8NIrCZ95DUYwRn79k+Paoiw/kXNv8ALNqQHKA7twRofIb3iG5g/fjBlvwTftZHS5SH7TC7R+9BsY0Ubsrrkoe//zRW1b9iwgcv670IUsodMuwYy3UB7YiZvP4KSG6q4Rzid4yjKckT6McJzc6kcIzD6DwpZVNL7rL2TwnhDiTacrUwE8vzPHUKY8FqQyxmcFmobCqPxcdDTDWYehSubOUK7McNa7EZ0ueMHFavDQZyqawiYdMS8gMaPRC9q1Rb0sooBlYBrezehqZttw1mFvuszuZImdI0V2Jr1SjUNZB0drApUgUTUDKOyvfPcZRP1eICdoG4zkygxkvIDKUNYr/5guuBQcF9f1gkd+qy4gQjUwUh+88m7WW4Yi5POCFp1xm4ZgoJKNOZYFFfF5AaVDqV7hai9TyqnLoKw+Vv09X9bsGCmxZaDA1qEi96xLMpxzcF2I+A3aYzYzxAUTfwAAIABJREFUGn3MbPRRLGu2DxfZMVKiL+0FLDR4wauwxbQGH0u6QkT8BkWn2od2GC24ZIsuO5MlskXv52KlP2kaCsuk0v+ubHvbC4x6+9DaL2vUrcuOrQaUXe1lfYV9hheYilSCHZWsymjAC7pWs91c16tUsnmwwJrdedbsyfG9Z4bIFF2vykizn8WdQXoafQxny+wcKbEr5R0rgxmHdNHbRsFKUKolYpEIWsxu9oJoTZXAWDzoBcfCPqOWbELleCw6uhJI9YIvg9nxmXupnIuj9bh9p2v7cOwxR+taRl7tqxIYgvGZst5rQND2grHNYS/L8LSuIB0xr1pKU8ja7zirHTvjjilvH2Qq+zhd9ILCo4Wx7+miS3+mQK6oiQUMOuM2S7tDdMVtOmI2zWGLWMCovVe99o4FnGvnkrofqst2XGoZi5miFzxL1g08GMk5DOfKZIsus5v9XDArQnvMoi1ikwh576+gPT5gV11WseydLwYzXpD4tRGvQuLOkRIv7sqRyruUXY2pvAqMiaBVqVxj0RHz1q01YuG3FOv3FnhmR5ZXenMMZR0aAibz2wP8wTlNLOoI0hW3CdheA5xKVmOuqEkVvHUYyJTpT5fpS3vfDQUzG73ga2O47hivHmeVY7w6NNWF8ceO1pXjh1rQ2DLGDyKoBtUnyq6u30eHaqJ96rjUjtvavix6+zJZ2Y+pgoPPVIRsoxbkD1YyR4OVbOlq4DZoj73XTGOs7ZNp877tqw5WqR9sMNntsW9Ge66k+coDvTyyKY3fUlwxL0p77OhnfYIEP99yTAWNIe8G00SqH3wHkym4RGujfmBag4/RglPryHijDLwTB8obNaWUwjC8N8m+YkETQ4FVGSGRKbo8tjmNz1Kc0hagJeodpufPimCbimd3ZNnYX+CGxXFGCy7xoFnpfMHHljWTCE18WAdtr2yDUUkPz5e8N2wsYGKbiuUzw3Qc4I1WLHsjR+IBL7V/aXeQlog3GsVQ8OxrWW5bVeDaBTFOr5RQm0h1BJw3ukkRsr2RJ60Ri08vb+Gp7Vke3thPT6OPPz6/hT86r4WnX8vwwKujzG7289kLW7GnYAxUBcOYsaYDBj+rZWIPSGvcfAYjHPeOpXAcs1wENEYg5GXqGubYfESW7X0pwDDQ+85TZJgYoZj352AUXcp78zM98XMCc5dWMj2DKNMkce0nyb3yW0ZX/Bg3M0Lidz6PdkpeW3xBfN3zCcw7d/+6JQDKwPBXbuaYJjhl3Oyot2x/CAXEL/8IVmPHAVfdCITBMHBzabLP34fV1Ilv2sleIHKfzalsP8o0vQNNwUQF07XroF0HFQh72zIQAaeM3T6Thmv+kNy6J0k/9UsCs08nfsVHiV/1MfLrniS98hcE551D/OpPTLyuQgghJi0e8Eo6HUg0YBzwb+Bd1FdvaAC0RmwyRRelvAscr7SXqvW7qn00x9UTfjz4LW+gF3g3bHanNJv6C9y7fpQzpnl9HtPwLmb/5MIWntyW4ZZnBrENxU1LGzGVIlrpT50/y+tP7RjZfzCaVwrK+wwxlcLR3kCv1ojtjYT3GXxsWXOtZOx4mtGCdwOm/rmdcZvPXNDKk9syfP/ZITTwd1e2T7jdRgturSRvyDZoj1qM5BxClT6i6d2Nq43krirXZelW+5NKVUZ6V246LGgPsqEvT3+6THvUql30jeQcfr02xYKOADMbfUT9Bm5l9HYsYKKUIhGyvJsVZZeo33usMWTWLiirc6o4ldHIAAHbu+FXcjQrNqcJ2Qbz2gK0RCzQ3rrGKwPnApais3LjoSls8URlWodTO/efnmFKUgZmtMmbm3IihonyHzi4Bt7gMGX5UZYFKOzmbm/nWvZYyf9qn1KBClT6+6rS/9ynj2uEomBYoBQqEELnM+Q3r8IZHfLKvDZ2oB0Hu3M28Wv+kPy6lZX+1xnetAXBCEalKkjs4vfj6ziJ3JoV+6+6L+DdLa7c0tFOCV0uYlb7qV0nE5hz5oH7bkrVqoSUereQfuLnBOaeiRlv2X9QolIY/rF+977961LfdkafuIPgyWdhxZpQth9dLqG1xvB5U5hYjR2UR/pQKK/t4FVPqb1WZTvWBV2NYBRQuNkU6ad+iX/6fOyOkzDDDaBd3EIGIxyrXCM0oJ0yVlsPoyt+TG7dSoxIAjMx8TlBCCHeCFprknmXRzaN8vPVI7w2Utqvwse+WYne/1GrSBb2eQGrhoBJe9RmfluApkqZ0Iagd4+vL11m+1CB7cMlXtiZ5f71o+RKXrnOeOX/2qIWowWX3tESw1mHTNHFUF6/qDFs0R61OLM7RGfcxjYV+bKuDTJLVwIqQ9kyO0bGSkBq7QUGq+UaF3bYNIW8QEQiaNYCl9WATjXYqSplQA2lan0roHbfzjAmd1P9aFvSNVa2NVdyGc457E6W2DpUZMtAgW3DRZ7Znq0kYXhZcm+bG6U74fW1GkOWF5g9SHfeK1lZCVpVyqyWXY1VyRqsBimMSlW6ySa41r8+HML2MxTtMYO2qMW5M8I4LozkHbYNFXl5T46Xduf42eoRknkv6JIImbRFLea1BuhusJnW4KM9ZtEYsmoJKuNvO6uD3kYyDUXA9u7PtteN3x8/x+DYetUH/+p7YLryQH3JXS+7D0rO2DFbLdcZtA2aK6VUw36v3fVvzSMt57qvWsBU61rm7mSXVf93VfeDUfdICGgIjr+O2rcdmvHHxesuF7wSqLZBU9hibmtgrESw600fksq7DOe8YGTvaJk9yRK9oyVe7s3z+JYM6YIXHAUI+wxmN/u5bmGc07pCzKhcH6kJ2mKYYJsmYR80R6zK+lTX4vW300TGNvmbfw000T711tn7Lf46+7L6n8fqtLhv++rbdiSvBRC04X9d3s67lxSwTcXsZj+H+dKvS4KfbzENQZN/fEfXAU8T3ofrgf9fKe+AvPPlEXaMeOnO//xYPzctPfxyrC/tzrEz6Y0syxa9GzVbBgt88coOQrbBwxtH6Un4uHd9ipKj+dTyFn69Nsn6vQUWdgRYsTnNoo4AqbzLz18a4dSu4KSW642o8QKnizsD3P1KCtNQLKh7jqEglXcJWIqWiIXPUiztDnH/hhSFsmbLYJHhrMMfLW/hgQ0p1u0tcMa0EIWyNyqjKWSOe4PbpuKkZh+PbEqTCFk8sTVDU9iid7TMr9am+PT5LexKhvnX3w6wZk+e+zek+OTyFrYNFfnukwOUXT0l52UKnXopoUUXHfgJyjh4QE0Z+HsWkFuzArt1BtlV91Me3A3m4Z3CdKlA7uXHMaONZFc/hL9nEcWdG7Dbeohe+F4Km57HySRxCzlG7v0usYvfT3jpVQz8zxdwc2ns1h7QmsC8c8i98ltKvVu8+TwnwdezgPzGZ/H3LERrl9TDP5iwDJczOlgZ8e9tFzc1iJMaoPHGv8Qt5Eg9+iO0W+ZQP7ANXxC7qZPcmhWElryN7EuPYLX1UNyxnuzLj9Fw9ScovraW1CM/JP/qM+Q3PEv8qo9R2LKa9BN3gHao7y4IIYQ4dB9YmuD9ZyQm7I9Vb/4c7MItYBl0N9is3JrhkjkRbn12iO6Gw5+TeTjr8NyOLKd2BnlyW4aLZkd49rUsZ08P8fYFce5ZmyRX0gyky3z/uSE+tbyFy0+O8cX79hCwvZJdUb/Bwo4Ad72cwjLUpD6dFLCoI8i6vXlO6wqyd7TEPWtTnFNXilXhXVgPZx0Wtgd4fEuaxR0BknmHn7+U5KRmH997ZpA/PK+Zy+dF+cJv9jCS9wJUQ9mxeXIAFnYEeODVUc6bGWFDX55fr0tx3oz9y75Wb8xtGypSLGvW9eVrc9xPxG8qrl8Y58ltFrc8M8jnL26rDVbbNlTEbynef3qCrUNFBrNekOfkVj+Pb0nTEbO4+5UUmaLLnGbvsZaIyS9eSgJewHnTQIHFnUGefS273ziy0YJ3Y+pLV3XgMxX3bxjl5BbNgvYAP1s9wmVzo7w2XOInq4b58tUdXkndpwZZPitC1H9ifJ4rX4DmD335gIPsvDtDB19XM9GGLuUpbl+LmWhj6Kc3E7v0psNuU3HHekp7NoNhUO7fgdXSTeHBW0lc/ydYLd0M/+JbmKEYhS0vkqvvfz36I4KLLkA/8yt8XXNR/iCpB2/15qefBMMXxO6YhXbLBOadQ379k5R2byIw7+y6zaHQleomlQcAyG9+Ef/s04ks/12yqx5A59NeRuskFba8iH/mqUSWv4vcmkdxc6MYwTBmpIHcq88QOGkJw7/8F4ILlqOCEfIbn8ffs4CRO79N6IwrwTQp7tmCTxkUtr6E3dYz7vVL/TsARezSD1Ie3EV5pA+0HneNkHnhPpzhXhqu/TS+noUk7/9vGq76uHetIYQQb6DqTeqBjMN9G1LctSbJYLbMeTMifO7iNrobxg8wceqyj8Z+9oJhYb9Rm6euft7AA2XXVDM70wUvU2t3qsT24RLbh4r0jpaI+g3O6QkzLW7TGfcysqqDyw40z2V1fWplPivBpXJlYJavUh53LKtu6t23qqq23VR4JUr9Jt0NPs6a7vUFy64XFDWUNxjNPMj+OOAy8AIJ1c/g4FFOeDrcrV9dB8v07pk2hy3OmBbErQysG8yUaxm/fksd9Fg8GsYFXya1iOPnuJtom5gKvEuEN66d+7bjaCy5/jiJmN57pH6uxup5qOzilVIteMHRQlnT3eAjETIrc98eemu8fzl+9vMbaSqfV/flDdpWLO6cXAznSEjw8y2mOsn3kZjV5OOGxQ38+8oBDIVXvizmTX6s8E5BPQkf/soo81hl9H7Mb9K1z8S1Ub/B3BY/tzw9SNHRfOLcZmY0+XjHwjjfWTlAY8j05kgyFJfOifLfTw/yDw/vJeI3uOmMRloiFsM5h2+t6MdQ8LaTo8T8Bj0JHz5T0Ra1iQdNLAM6Ksu2DJjR6CMeNPnMBS3c/uII965L0RGzefv82Lj2ndIW4P4No7ywK8dnLmjlxy8M8cjGtFfi98xGDAW3POO1Kewz+MDSBC1hi/aozR2rR/iDZU21m5jTEz6CtuLDZzXx/eeGuPmhvcQDJp8+v8UbLdbs55uP9mEZ8OGzvJJn6/vyfOORPqxKqd/ABPOITgVKqcMOVIJ3TEXPv5Hkg99j6CdfwYw1E7v8w+ReerSSNQrKH8RqmQ6Ghd0+qzavk9U8zcturGN3nISbGWHoJ1/DaukmesG7cQsZRnasY+gnX8PuOIng3LNAKSLLrvfK6QKBuWcROOk0rOYuUg98j/yGZzAica/UbLmE3dKNMkzsthko08Ju7UHZ3oncjDWB4xCYu5Ty0B6G7/wWKIPgwgu8v9VW1iC4YDmjj/+U6EXvw27pBhRW8zSCiy5k6I6vYyXaCC68EFwXI5LASnSgfAHsthne1lJgt80cy1So7AO7bQbKF6Dh2k+TfOB75Dc8jRlvpeGaP8IIRilseZGh27+Gsv3Er/o4vq45FLa/wtBPvoqyA8Sv+piXxSCEEOKwqcro9iO5ZjOUV+3i+88O8dyOLDMafVw+L8qe0VItG7QhaNIWtQlYXr+oOmp8RqMPX11/wlCK+e0B1uzO8dCroyzqCHDBSRGmNfj4/rNDvNpfYHaznzktfmIBkwtPivBvTwxgGvA7ixuY2+LnMxe08NMXR/j12hQdcZsl0+Ks25unNerNhTI94cMyFDOb/LVAZGfcJuo3OWt6mHTR5euPeH2ga+bHCdpj7QvaBqd2BvnRC8N8ankLIzmHb67ox6z0+7obfFw6N8p3nxzEUHDtgjjdDT7OnRHm3nUpTmkLMD3hQymvykdvqsQ3H+0jaBt89OwmL1Mz5n22WYZiRsJPwDJ43xkJfrZ6hETQ5NSOILGASclRtSohtqmY0eTDMrw+XsA2uH5RnG+t6GfTQIF5rd5c5qd1BXl6e4Z/eLiPrgab5TPDlFzN+05v5NZnB7n5oT5aIxYfqszrfuuzQ9z8UB/tMYsPndlEKu/wn08N8u0V/XQ3eFkYtqVq/e7GkMU18+P8+xMDNIa9bA3T8Abr7Rwp8u0V/fgtxUfOacJvee1xtObC2eP7RlOZUor6krOHwwjFabj2U6Qe+wk4ZUJLLsPumovdPnOsT9nUhRGMYsZba1mZyh/Eau0Zd0fO8Afxz1zM6GO34ebSxC77EHbHbGIXvY/Uwz/ACDfg656Hm03hn7GIwuZVdf2vj+GfsYjoBe9h5Nf/BkBgzplYjZ2VjNAoplMGrVH+EHbrdKrBXbttJsoXIHHtp70+84+/jBGKEbvsg9SfcKzmaSg7QObpu7EaOzGC3rEQWXYdyd/8B0M/vRlf11x8PQvQTqnSn/VjNXVihGKV7RXFauoatw0jZ1/LyD3/ztBP/x5f51xvYKCGxHV/TPKh75N9/jf4ZiwitPAC7LYZpB7+HzLP3I1/1mkE552Nsn2MPnY7ZjjuzesZSWAEIliNnQD4exbg65rD0E++itU8jdDii9BOqXKNcGvtGiF++UdAKUKLLiT38uP4T1pyQt0sEkIc33Qle3NPqsSv16b41doU+ZLLxbMj3LC4gdnN/sOaj3Oyqv3M6rxz+2VqaSplZ8eeP9nXBcb1Ya0pODD/SFS3gV3JRnurqN5Hbqhk8QrxeqrnIZ8JPtPL/O6MH5uSpkK8HjVxymzNsZlpVEx5ulKaoT5T1NHUAqvVn11dLWXhlcPV1Y5WpdNQfayq+jddqVdevVFX/b9qjWlTjWVF1D9337ZUl++1eeK2Vmu8G9XJ0g+yntXSZvV1svdtU+01oTaSpVpLvLp+1Zri1TIeE63HgR57y6qMHMItgzIr28cFFMowvM58daO6TnUYXWUOJoWqjPgee56qPV+pyv9rt3IwmGPPofI8vMer+8W7qnHG2gL7L39cOypF9Kojz6vlvQxzv/3qtdHxnls5OGvLrax/5Zle+6qvW7+8ys/j5geoPlb9XVd/9w7esdc3QO27zLHHhDiG5ACbWqSf+Caqjqh1XWolkw7U75qoj1b/+VC9EVZ9rfq+S7U/s+9rwP79nPr+VP3y3QnaUd/Wg/V3JlrPCftLE7TJcffv3+27rGoGw4H6s2M3B6v/P3F/06hfL8b3U6vtN4yxEnaVT+/a48YEzzX22Q/Vvmj9tjxYv3nfbZUva9b35fnpiyN84Yp2/JZkw9Xbry8IE/YpvT5dfcd/gj5XNWNS79N/dJ3Kjtq33zZR/2usrzh++TDW93PHslon6ucps9IPHv8e8NaT2sE6cV94n/5sXZ9aV59X1zc84P9Xt4N29+lLuwfePtV3nmLccrR269rjrcBYP7xcW1+dT5NZ9SDOcC/xqz8h/dcTh+zIqeUt1090XG8OyLteTnL/hlEUcMW8KNctjDO9wVfrGwghhBDiqJvwA1aCn0IIIYQ4XsjdgKlF+olCTBFaa57fmeOHzw/xoTObOLUzIDdgxQlJa036tz8lv/F5Eu/8U6xE25vdJHH0yElrajkh+4n1c93lyi7DWYfe0RI7hku8tCfHE1szhH0G18yP8fb5cdpj1n6D7IUQQghx1EnwUwghhBDHNbkzMLVIP1GIKcStZIhahmSeiBNXrcpLXdUWccKQnTm1nBD9RMfVjBYcBjMOe1Iltg8X2TZUZMdIkd5UmVTeoawhZHvTLl1+cpTLT47RHJbzjxBCCPEGkuCnEEIIIY5rcodgapF+ohBCCCHeKNJPnFqmfD9Ra83dr6T47pMDpIteOfOY36A1YjOtwaan0UdPwkdn3KY5bBH1G/hMJUFPIYQQ4o034Yev9Ua3QgghhBBCCCGEEEKcmKpzuDrpYZRhYoQbUIbMMyymDq01W4eKfOfJAU7vCnL1/DgdMYtE0CLkM6ifNluCnUIIIcTxSYKfQgghhBhHa42jwVRyMS+EEEKIt44DVcaS/tCh0OCUGH38ZxS2vYSyfATnn0doyWUow3yzGyfEpOTLmv/72wFifpPPXNBKk5SxFUIIIaYcCX4KIYQQokZrTe9omVufHeQDZzTSFbflQl8IIcQxM1Gw6a3+uVPbJlqjy0V0IYubz8DBp6zxKIWy/d6XaaFMC0wLlKK+GlR1G4/f/tpbhuOg3bL33SmhnbKXtadMMAxvGSjvNRXe69Zev/IaWgPae/3alwuui9bOuNfGKaOdsreu5SK65H2nXALLxgjFMIIRDH8I5Q9h2H4wbZRleW2qrMtkjpv6bYtTQpcKuPk0TiaJmxqinOrHSfbjpAZRlg8z0oAR9r7MSBwjGMUIRFD+EMr2oUwblOGtm3bRrjv+Z9eprGMJ7Tp124q67bjPNlSqtt+UMirb3PACh4ZReQ4cXhVYfdBfD+l1nDK6PLYN3XwGN5PEySQp791K+sm70G4ZZdqU9m7HP3MxVmPH4S5QiDeMV+42yQs7s3zl6g4JfAohhBBTlAQ/hRBCCAF4F/qpvMtXH+jl2R1ZNvYX+crVHbRFLbngF0KIfWitoVxEuw7K9nvBiSlwrhwX/KkGqup/HxdY2Tcwc2iBSV0LeDleQKuUx81ncbNJnMwI7ugQTmoQJzWIm89g+ENekCsYxQhGUMEoRiCM8gdrgS9lB7zAkFJeQM4wvaCQMiYMIh1qm/ffTpWAnVPyAnKu6wUltTsW0Kt8rwW7tOsFHi0fyqgEsUwLzGpbFVBpr9Ze8CiXxhkdxBnupTywi9LADsoDu3BG9uKMDqFLhck33jRRlg/DF0IFQpXtGsUIRSuBxChGMIp2y7i5NDqXxs2N4ubSuPk0upDDLWTRxRy6mPfWR6nKNlZ1gTgvGDe2/Q3Qjre9XGcs0Ol627B6vNWOi8q2qwUMlap7rbF9pp0yVLYppoVhB1DBCEYgghGKYoYbMGPNXpB3X8qoHCeV4K3r4CT7KQ/34qQGcNPDuPkMulxCGSbKF8AIxzFCMXCdynbJoEsFb/8rUKbtbd/KNlWW7QUByyW0U4L6n12nsn4OuLouXjkWAK39Xvc3r82WF2C16r7s6lcA5QtMOpNS1wefXacuEO0ds7WftTu5Y0xrL9iZTaHzWbRT9ILkykTZPjBMnGS/93qWb2w/CnGc01qzaaDALU8Pct3COGf3hKfEZ7sQQggh9qcOVNalYspPUC6EEEKIycmXXG5+eC8rt2X4xLJm/uf5IVqjNv/nqg6aQm/IiGe5szC1SD9RvGVUAwduPoMzuJvizvUUtr5E4bV16GIOq6UbX8ds7K452G0zsBraMIJRsGzgyDMZ981Wc0sFdG4Ut5CrBF0K6GIBXcp7QZrqVzGPWw1g1T9el12ny4XxmXamhbLsWoAHy677fewxUF7QpC7ot3/DXZzRIZzRYdxcCp3PeG2pBFkNO+AF54KVAJJT9gJu9W2tva7ywrG+gNcGqAsqWl77bP9YgMjyVYKmYS+gGoh4ATN/CBUIewHBQAjlCwLgZkcrWWsjuOlhr93pYdxMEjczgpsb9dq+X+CoLohXCyhTy95Tll3JUrTHAlg+L3Bl+ALeMTXSh5MZ8ba/YWKE41jxVqzmLqyWbqzmboxghEl9TGrHCzDnRnGzqbGvanAzl0YXvGAehomy/ZVtEfa2SyAyFigNRmrbZ/xudSrZmg64XtZmLehbDUabViXwa9a+1wLVRuUYq+6n6n4z7bF9apiVPFLtHceVrEI3m/T2U3rY2y+ZJE42hS5kawHXsf1TF5ym8rMyMEJxzIYWrIZ2zMZ2rIY2zIYWzEgjRiiG8gW9NqC9Y7JURBcytWCfkx72jpHUEM7ooBc4tX1jGbe23zu2bT9U1/EQBw3octE73mpfOdxS9eex97oXVJ4EpcYySMdlk1aC1kbd45N8PSMYw4w1YUYbMSIJzHBDbcAChsHQ7X9PdvXDKDtA7KL30vD2P6ps18Mm/cSpZUr2E7NFl7/81W5Gcg7/9M4uGoKS9SmEEEJMARN+WEvwUwghhBCUHc1/Pj3IT1YN81eXtnHZyVHW9ub561/v4aRmH1+4op144Jhf/MudhalF+oniDTN2zVIJLNVnLFaCULWMuqOV9Vcu4owOUerdSmH7yxS2raG0exNuehhl+72A57R5GKEYpb3bKPdtp5zsR5eLGP4QVmMHdvtM7M45+DpOwmqZjvIFqDTKCzRUs+kMY1xmnS4WKkGWIZyRPspDeygP7cEZ2kM51Y87OlTLVqtlzNUFNFAGqpL9pywbKoElZdUF4Ww/yvLXgjbjgh6aWqajLpe8DNfKz9UypdX1qGW81i+7rkSq4Q9hxhoxos21QIkZbsAIx71gmy/oZYpVQ12uM36ZxbyXhVjIekG9fMYL/BQLlcBt9atYF9z1vruFHLpQDf7mcIv5sexN16kLrAKMBStVoBIwDca8sqeVwI4RbqgEM61KgM8L7KlKYG+/46hc8rIoi7nK96zXlkLOC2YVc96x0tztBTkbO7EaWioBuAAY1pEdw+MfGCvH6jjoctE7Ziy77r0zPuMSqsm0ap+XOvTT/9HsP+xXqtd1x84H1Y+mynliLCBd+VLKC7Iah1Yu9+BtOLAjWe+DLmPfzO3Xb8k+vSxV9+3I9s2BjhFdzFPcuR5l2fi65o4NWjjCRYkpY8r1E12t+fELw/zX04Pc/PZOlnaHJPAphBBCTA0S/BRCCCHE/lytuWtNkn96rJ+PntPEe09PYBoKrTXP78zxt/fs4dSuIH9zWRsR/zEt6yh3F6YW6SeKo6Z2TVIpM1ke3E1p90bKw3srQaL8WFZjLYuxGgArekEd0/YCab4AyhfE8HvfvZ+DKF/ICyhpd3x5Sqc89nMlGKbLRUq7XqU0sAucEmasGbtrDv4Zi/BNn4/dOgMzkqhldnpzMxZwM0nKw72UerdS2r2JUu9mSv07cdPDY8GlylvHC1bWZcNVS6KiKqVHvUw2r7xmBDPS6GWrJTowGzuxEm0YkYQ3/2Etq9DL0KSaqVkpC7tvhtfkAx963LexIFI122z8/IV1LzqmbhnH6vNj/2taPdZm7VQyFCtzL5ZLXgZFAYbzAAAgAElEQVRd9Vgq5gHtZYYGwl4GZDVovM9cmUfcfq33OXHqo/faQpz45E0ytUypfqLWmg19BT7zi51ctzDOx89txjTkkBNCCCGmCAl+CiGEEGI8rTVPbM3whd/0cvX8GJ9e3ozPMsb9feW2DH93Xy/nzQzz+YvbCNrqWN2klTsMU4v0E08A48qpul55R7eY88pkZkZwkgO4udHJvZhhYkYTmNEmjFC8UjKzUqK0klFWPXdUy1HqfMYLFu7eROG1tRR3rKPU9xo6N4ryj5VDVZYPbJ9XStLnH5vvzg54Qb3KOuhSoVLmNYdbyI+VUC3XlXVVRq20pjLrynPWlXpVtg+rqQv/jEXYnbOxEu1e+U81ufNf/XyRupjDGR3GSfZ7bXFKUC7XZVaOZVVSKqLdMmYkgZlox4y3YEYSXmDO56/MaSjBMiHEW5Kc9KaWKdVPTBcc/vzu3eTLLt+6fhrxwOTm0xVCCCHEcUGCn0IIIcRkaa1JFVxMBWHfMc12fNNorVnfV+Dzv9zFgvYA//vydsL+/S/0Xa15ZGOarz7Yy+Unx/jMBS34rWMSAD3xNvKJTfqJU0R9VqUXiBuiPLDTK6Oa7PfKqia9UqpOZgQ3l/ZKjmq3Vh51cgtya3MiKtPy5r0LxTHCDZjxZqx4M2a8FTOSoDS4i+Jrayn1bvWyIi0bq7ET37S5+HsWYk+bh908zQs4GkalLKcxlkU4QTbhAUvjum5lLsBKmdN9SsSq+mzI2llITTrQKYQQ4g0hJ+SpZcr0E12tufXZIX74/DDfeEcnizuD8vkvhBBCTC0TfnAf0WzzQgghxIlIa81oweVL9/USDxj8+aVtBKwT6wJYa83uVIkv3d9Le9Tm85e0EfIZEz7XUIqL50TIl1v5+iN9BGzFH57bjG1K5pEQxxtdmddOlwq4mRHKQ3so7d1KaddGir1bKA94JVi162L4gxihWC046ZuxCKuhFbOhtTY3oxGszD04mfe66+LmM7jZJE5q0PtK9nnB1dQAhS2rcTIpdD6NEY5jd8wmev7v4uuej902AyPaWJsT7nDOLaq+7Gr9v5tyx1wIIYQQ+9Na88qePD96fpj3LmlgkQQ+hRBCiBOGBD+FEEKIfeTLmm+t6OOFnVlcDd0JHx9c2nhCzfuSzDt89YG9FMuar13TRlPIPOiFvqEUV86LkSu5/OvjAwQsgw+f3YQtFaGEOOa01l7J1tJYiVS3kEUXsl6wMZ/GzY7i5kZxM0lKezZT6tuOkxpAlwoYviBmQxt263QC51yH3Tkbu6UbM9rkzYlp71+W9nCZsaZ9G19JwnS9DMzKXIvVuTglu1IIIYQQb5Z0weWfH+/npGYfNy5JYEifRAghhDhhSPBTCCGEqFNyNP/51CCPbU7zN29rZ8dwke8/O8T0Bh+XzIlM+Zv0WmvyZc23V/SzebDA167ppCfhm9R6mYbi+kUNZIua/3p6kICtuGh2FNtQtEWtEyo4LMSbTVfm4Czt3kz62V+TX/8Ubj7jlZUtFbxgovdEMEyU7fe+fAGsxg6CC87H1zkbu20GZqIdIxTzgpy8wcFGpSoVak2vnZYNwegbt3whhBBCiAm4rubHq4bZPlzk29d3ETlAFRwhhBBCTE0S/BRCCHHUFB2NZTBlR8w6rua2VcP8/KUR/viCFi6ZE6HkaHaMFPnGo320xyzmtwWmdADUceGWpwdZsTnNX1/Wxmldh1bayTIU7z09Qbbk8p2Vg/z300OEbMUHz2zkxtMSEgAV4ghprXFzo+Q3PE36yTspbFqFEWkgdOol2C3dqGAUIxjB8IdQ/pD33fZ7c3NaPrBsL8CovBt4U/l8JYQQQghxLGitWb07x+2rRvi9sxqZN8Wv8YQQQgixP6X1QecgnzITlAshhHhzaa355ctJTu0KMqPR/2Y355C5WnPP2hTfeLSPDy5t5KYzG7EMhdaakZzDX/16N6MFl2+8o4u2qDUlL45drfnFS0n+5fF+/mBZE+9ZcvjBylU7s/zhz3ZQdsC2oCNm839/p5uOmH0kTZx6G/WtTfqJR4k3T6dLuX8nmRfuI/PsPTjDvfhmLCKy7DqC88/DCDdMyfOOEEIIcZTIh+DUclz3E5M5hz+5cycRv8HNb+8iJFmfQgghxFQ2YT9RMj+FEEIcFbmS5o6XkgxkHD589uTKqB4vtNY8sTXDtx/r59oFcd5/RgKrEhRUStEQNPnLS9v407t28Y+P7OXvruwg7DMOeR211uRKGttUWMYbl5GltSZbcnl4Y5p/e6Kf6xbFjzhL06nezlBeD8NxNY57XN/jEOK4o7VGF3MUNr9I+qm7yK1bibL8hE69hMjZ1+KbdrKXyTmFzqdCCCGEEMczrTW3rx5mT6rEP98wjaAt/SwhhBDiRCTBTyGEEEdMa83a3jwbBwqUHM27lzQQ8ZtvdrMmRWvNS3vy/P2De1nWE+IT5zbjs8aP/FVK0ZPw8VeXtvE39+zhP54a5FPLW7AnuYpaa1wNz+3I8p0nBzizO8RHzm7CNhXHMqahtSaZd3l00yg/fynJa8NFLp4T4WPLmrHNI1vwKW0Bzp8V4fEtaWxTccmcKG3RI8r6FOItQ2uNzqXJvHAfo7/9GaXerfg6Z9Pw9k8RWnwRZkMLSkkGghBCCCHE0eZqeGVPnqXdIWY3+2WQmRBCCHGCkuCnEEKII6aB36xPEbQUO5NFVu/Oce6M8HF/Iam1ZvNgkS/d18tJzX4+d3HbAUf+KqU4ozvEJ5e38O0VfXQ3+Lhhcfx15zfVWtOfKXPrM0P8el2KjqjNT1aNUHQ0H1/WjN9SR307aa3pS5e5f8Mod72cZDhbZtmMMJ+9sIUFHUHsoxBTCdmKL17ZwdreHLapmN8WwJJYjRAHVc30zK5+hNRDt1Ie3E1w4QUk3vlZ/DMWo3wy35QQQgghxLGkgUzRpSNmSy1lIYQQ4gQmwU8hhJik/nSZqN8gcDQiRyeY/nSZldsy3LgkwTPbM9y7LsU5PWGOMLnwmNJa0zta5kv39RLxG/z1ZW3EAwcvZWsoxTWnxNgxXOTfVw4wLW5zdk9owv/RWlNyNY9tzvCdlQNkii4fX9bMNfNj3Lc+xb89MUDJ0XxyectRKbWkK8vcOVLiV2uT3LsuRcnRXDInyvWL4sxq8tdK+R4NSinCPsWZ08NH7TWFOFFprdHlIvl1T5J84BZKu14luPB8mt7/Ra+0rTr0MtpCCCGEEOLQua4mX3Zlnk8hhBDiBCfBTyHElKH12HyCb/RNYldrvvfMIGf3hDl/1vGf0TgZZUejFEc07yN4+2Xl1gxFR3PVvBiJoMl/PT3InlSJaQ2+o9Tao0trzUjO4csP9JItufzDtZ20Ra1J7VfLVPz+2U3sGCly88N7+cZ1XcxsHD/HqdaaHSMl/uPJAR7bkmHZjDAfX9bEjMrzbljcgG0q/vmxfsqu5o/PbyVoH14GqNYaR8PmgQK/WJPkkU2jBC2DaxfEufqUGJ1xb0TziXDMCjHVaK3BKZPf/AKp+28hv2UVgdlLaf3Df8E/6zQwTHlvCiGEEEK8gcouFMqasE8GnwkhhBAnMgl+CiGmjHxZc8/aFJfMidAQfGNvGO9Jlnjw1VH2pEosmxGe9FyPxyutNQ9tHCXkM1g+88iCuUVH85v1KU6fFqSrweYCK8ItzwyyYnOa952eOC4vKLMll68/0se2oSJ/f03nfsHL1xOyFZ+7uJXP/3I3X3twLzdf20ki6B0U1eP0lmcG8VmKv7q0jUvmRMbNI2oaimsXxPGZim882kfRgT+9sOWQL8BdV7NxoMD/PD/Myq1p2qI2v39WE5fNjdIUkqCKEG8WrTW4DsXXXiH5wK3k1q3E3z2Plg//I8F5Z4N1aOccIYQQQghxdJRdTaGsJfNTCCGEOMFJ8FMIMSVorXl8S5qvP9rHr9Ym+dOLWlnQHnjd+RaP1rLv2zBKMu/w/M4cG/sLnNLmn9I3rrMlzY9fGCZgG5zZHSJwBGVXNw0U2NBX4AtXtGMqaItYnNMT5v4No9ywuOGolHQ9EtWM4bLrze0ykC5z++phnnktyxeuaGdhx6HPsaeUoiVs8deXtfG5u3bxz4/18xeXtrF9qMj/e2KA1XtyXDkvxu+d1UhbZOKMUtNQXHlKDMtU/MPDfZQczecvbiXqf/0AqNaaPakyt60a5p51KdqjFn92cRvLZ4aJTOL/hRDHjtaa0p5NpB64lezqh7Bauml63/8mtPgilC8o708hhBBCiDdR2dUUK5mfQgghhDhxSfBTCDEljBZcfvDcEGdMC+Jq+Nxdu/jw2U1cvyiO3zy8cqGTlcy73LMuxZXzYqzdm+fuV5LMa2tlqt6+9srUpnm1v4BS8OyO7GFnf2qtuX/DKM0Ri9O7q3Nfaq46Jcaf372bV3rznDHt8G/2VwOXGvbb3geaZ9PVXhmj4VyZ3ckyW4cKbBoosG2oyJ5UiVTexW8p/uTCFs49gqxXpRRzW/x8/uJWvnR/L0NZh/V9edqjNn9/TSdLp4cw1cHLzRpKcdncKLah+NpDe/nqg3v5y0sPPPeo1ppk3uGXL6f46ephFIoPn93I2+fHJxU0FUIcW9p1ya15lMEffxkjECbxzs8SPuNKVDAq708hhBBCiONAydEUHVeCn0IIIcQJToKfQojjntaa+9an2JMq8zeXtdPVYPOD54b47pMDrNmd41Pnt9A+yfkaD2fZK7emGcqWee/pCZ7anuG2VSPctLSRzrh91Jf3RsiXNbe/OMLZPSFyJc1tq4Y5a3oIv3Xo228k5/DY5jSXzY0S83sXj0opFnUE6W6wa+VwD3fP5Eqa7z45wEjOwTIVluF92Sa1ny1DYVV+H8k5bBkssmOkyGDGoexqIj6D9phNT6OPi2ZHmNXkZ1rcpiNmH3HmsFKK82dF+P2zm/jBs0P8zuIG3r0kccDg5UQMpbhwdgTLVHz5gV6+/EAvf3NZ27jSzlpr8mXNo5vSfP+5IfrTZa6ZH+M9SxLH7NgXQhwa7ZRJP3UXwz//JsH555H43c9jxprl/SmEEEIIcRwplDVaQ8CW4KcQQghxIpPgpxDiuKa1ZjDj8ONVw1x+cpQ5LX4MBR9f1szijiDfXtHPp3++k89e2MI5PWFM4+jeZC6UNXe+nGRpd4iTmvxE/Sa3vTDCA6+m+ODSxil3U1trzTOvZdk0UODr7+giV3L5m3v28NyOLOfOOLQsSK01z+7IMpJzuGzu+KymoK24/OQYt60apj9dpi166IFiL6s0xR0vjTA94dsvUKm1xtHevJeOBsfV+C2D7gabS2ZHOanZx/SEj9aIRdhvUr22Pdr7zDAUN57awIUnReiKH15A1VCK5TPDfPGKDr50fy9fvK+Xv728naaQSdmFVbuy3PLMEGt785w3M8wXr2ivvBem1vEnxIlIaw3lIsmHvk/yvv8ics51JK77Y5Q/NOU+I4QQQgghTnT5kouGN316FiGEEEIcWxL8FEIc936xZoR8SfPeJYlacNNUcN7MMDObfPzr4wP87b17eNepCW5amiDsOzrlP7XWvLQnx6t9Bb56TQemAe0xiwtnR/j12hTvXNRALGAe8XLeSEVHc/uqYU7tDLKoM4Drwvz2ALetGmZp96Flfzou3LsuxYL2ALOa/OP+ppTiotkRfvDcEI9vyfA7i+OHvE8GMg4/eG6IK+fF+LOLWycMfrqA1uBWyt1ahqqUQT76Qc6D8VkG3Q2+I3oNpRTn9IT4uyvb+bv7evnCb/bwe2c2ctfLSX67JcPcVj9///ZOlnaHsIw3dv2EEAemC1mG7/pn0k/dSfzyjxC/7ENg+eQ9KoQQQghxHMqXvWvHoGR+CiGEECc0+aQXQhzXdiZL/GJNkusXxZnWMD57UClFZ8zmC1e08wfnNPOLNSP8xd272TxYrM0VeSQcDXeuSTKzyceSaV4GjwLesSBGf7rMk9syR2U5bxStNS/szPHK3jzvXpLANhR+S/HeJQnW7M6zamfukNZn+3CR1btzXHVKDGuCT5OuuM3S7hD3rU9RdA5tO7la89PVw+RKmpuWNuK3DGxTjfvyWQYByyBoG4R9JlG/SdA2MIxjOwfssaSU4szuEF++qoMdwyX+5Be7eLW/wOcubuXb10/jnJ4Q9jGe41YIMTlaa5zRIQZ/+EUyz/yKxhv+jPjlH0bZfnmPCiGEEEIcp/JlF4DgRBexQgghhDhhyCe9EOK45bia214YJmAp3rmoYcKbyUopArbBu5c08PV3dJEuunzmFzu5Z12K0iEG3Opprdk+VOTp7VmuWxgnUMmIVEoxpyXAkq4gd65JHnJQ781UcjQ/WTXMgraANw+n8oJoZ00PcXKbnx+vGp70NtNa8/DGUcI+g7N7Ji6Xayi4+pQYG/sLbOgrTDqwqrVmy2CRO9ckeddpDXQ3TM25VQ+XUorTuoJ87e0dfHJ5M//+rm7ePj9G6ChlNAshjpzWmvLQbvr/+8/Jv/oczR/8P0TO+x2UKUVVhBBCCCEmorUmnc6QyWQnvDbUWpPL5UmlRnEc55gNNM4WNZYBtinXVkIIIcSJTIKfQgjAu9A4nrIYtdZsHChw34ZR3rMkQXP44OVlDaVY1BHgW9d3ccGsCP/4SB/ffXKAsnv46/SrtUniAYMLToqMCzpZBly/qIF1e/O80ps/7O2mtcZx35jt7pXwzfPi7hw3Lkngq7vQq2Z/rt6V48Xdk8v+zBRdHto4yvmzIjSGJt43SilOmxakLWZx3/oUk13Lsgvfe2aQppDF9QsPvVzuiUApxfy2AO89PUFjyHpLbgMhjldaa0q7NtL/nT+hPLiblo9+neDii1GGdKuFEEIIISbiui4rVz7Dbbf9nB/96Gc8/fTz4647tda8+upmbr31Nu6441f87Ge/pFAoHJO2ZIuOV0lIgp9CCCHECe2w7tJorXFdF9d1j6tgiRDi8Lha89DGUV7YmXuzm1LjuPA/zw3RFrW48pTYpII/SikSQZM/vaiVTyxr5merR1i59fBK0w5kHB58dZQrT4mRCI4P7imlOKM7xPSEzV0vJznc+Gqm6PKPj/Tx2JYMZefYBkHLLvxk1TBzmv2cNT00bnsqpTi7J8zcVj+3vTBM6XVWSGvNi7ty7EmVuWJejIPtmYjP4LK5UR7bkmYo67xuO7XWPLcjy2+3Zvi9sxppCE6tOVWPpmpmrhDi+KG1prDxOfq+8xlA0frxb+Offbq8V4UQQgghDmJ4OMmLL77Mu9/9Tm688Xqef341yWSq9nfX1Tz++JNccslyPvCBd5HPF9iyZfsxaUu25NamUhFCCCHEieuQg59aazZv3saPfnQHP/zhz1i/fqMEQIWYwlyteejVUb7ywF5+vmYE5wgyJY8WrTWrd+d4YmuGm5Y2EvVP/lSllHcR887Fcc6ZEeZfH++nL10+pPOU1ppHNo5SKGuunDdx4DVkK96+IM7KrRl2jpQm/dpVjqv54fPD3P1Kki/cu4ebH97L3kNs52RprVm3N89zO7LceFpDrYRvvYClePdpCV7YmWPN7oNns7oafrM+xexmPye3vP7cdpfOiZItujy1/fUD0bmS5r+fHmRhe4AL98m4FUKIN5N2XbIvPkT/f30eq6mLlo99C7tzjpynhBBCCCFex969fTQ0xIlEwsRiUWKxKH19A7W/G4aiubmJHTt209u7l1wuTzQaOert0FqTKUrwUwghhHgrOOSJifL5Avfe+yBXX/02/H4fd9xxN11dHcRi0WPRPiHEMeRqzcMb09z8cB/NYYtnX8uyO1Wiu8H3prar6Gi+9+wgJ7f6DzsA5jMVnzyvmU/dsZPvrBzkry5rw55kEmG26PLLV5IsnxWhKz7xfJNKKS6eHeUHzw1x7/oUH1/WNOl2aq15cluGn7w4zMeXNdEUtvjPpwZ5YWeWjy9r5qLZEWzz6GX9ORpuf3GYnoSPc2dOPD+nUoplM8Kc1OzntlXDLO4MHnB79aZKPPtalo+c3YR/gkDqvq87vcHHaV0hfrMuxeUnRw9YXkhrzf0bUmwaKPDN67pe97WFEOKNorWm+NorDP7wiwTnnUPju/8aI5KQwKcQQgghxCTk8wX8/rH7DLZtkc+PL2vb3t7K008/z86du9HaJRI59OCn47iUy+UD/l0Do7kyPkNRLhUpaJm2QAghhHiz+Hz2Mb2vcsjBz2rWTigUxO/3oZTCdd2j3jAhxLHlas2jm9Lc/NBeLpgV5gNLG/nUHTt5ZGOam5a+eTd0tdY8sTXDmt15vnpNx4RZipOhlKIrbvNHy5v56oN7OasnxBUnR193vbTWPLsjy85kic9f0oZpHPj5zWGTt82Nct/6FDee1kBj6PVPqVprdiVLfGtFH2f3hLnxtAR+S7GkK8R/PjXA1x7ay2Nb0nxsWTPdDUf+AaC1ZmN/gZXbMnzuolZC9oEv7oK24j1LGvjKg3t5eU+O07qC+y1fa82KLWmUguWzJheYNg246pQYX3mgly0DRea1BSZ83kCmzA+eG+KyuVEWde6/bCGEeNO4LqlHfoSVaKPxPf8LI/zWnI9YCCGEEOJw+Hz2uKBkuexg22PXz+l0hqeeeo6bbno3iUScFSueYOXKp7n66rcdUp9LKTAOOg+7JlvShHwGlmliHOR6XwghhBBT2yEHPwMBPyefPJsf/egOTNNg+vRph1WKQutjO7+dEOLANPD4lgxfe3Av584I8dkLWwjZBufOCHH/hhQ3LI4Tst+ci4BM0eX7zw2xtDvIGdOCR3yuuOikME9vj/D/nhhgfpufaQfI5Kwqu/CLNUkWtgeY1+J73cEdV50S466Xk6zYlOa6hbHXbU++pPn2Y/0YCj59XjM+0zsftkdN/vLSVs6bGebfVw7wyTt28JGzm7hiXhT/EZTjcTX89MVh2qMW588Mv+72XNYTYkbCx22rhlnQHsAyxj83X9bcv36Us6aHaI2Ykx78csa0AImgyQOvppjT4ttvnlCt4fYXR8gWXd5/egIDjXsclGAWkyPzk4oTmdaa4q5Xyb28gsT1n5XApxBCCCHEIWppaWJ4OEmxWMR1NanUKM3NTbVr0+p1qmVZKKXw+/0kk6OHvBzDMA4a/NRaky97wU+/zzroYGchhBBCTG2HHPzcu7efDRs28cEP3ohlWdx++11s27aDWbN6DulGUKlUPmgpCiHEsaE1/HZbhq8/NsyynhCfWhbHdEsUCnDpSUEeeHWU57YmOWt68E1p26/XjvLaUJFPndOKUyriHPp0mvv54JIoa/bk+adH9/K/Lm3Cbx34YuiVvQVW78ryFxc1TWr57UHN0mlB7nxpmAt6bIIHyax0teaHLyRZtTPLFy5rJm475PPOuOec02Uz55pWbnspzT891s8jryb56JlxZjX5OZx77ZsGiqzYnObjZzdguiXy+YM/XwE3LIzwzccGeWF7ksUd47M0X9ydZ+tQgd8/I0axUJj4RSbg03DBzCAPvTrK784PEQ+Or6m7aaDAnWtGuHFxjJaAu18JJHH8UkoRCPjf7GYIcexol9EVP8aMtxBacmjZB0IIIYQQApqbm+jq6uD22+9Ca83MmdNpakqwbdsOtm7dzoUXnsfixQv46U/vorm5kd7ePq677qpj0pZM0SVoG4d1fS2EEEKIqeOQg5/JZIpoNEJTUyNKKVpbmxkcHGLWrJ5Deh2fz8bnO3gGlhBif1prXA3VAYqHchO2WlL2m78d4ZwZYf78kjbCPqP2GqdN9zOrKcnDWwucPzfxho+CHMqWueOVDBfPibKkJ3bUlj89qPn0+fC/793DI9tKvHPRxFk7rtY8sDlJZ9zH8jkNhPyvP0mo1pobTk3w53fv5tVhOKcnMOFrV8vp3r4mzfvO+P/s3XmYXNV95//3ubf26qred21ISEICsYMFiH0xm1m8O7GzOMkkdiYT25nETjyZ/DyO4yS2+SWeSTJP7OxxbGIwGLOLzRgsNgEChAQSi9bet+rq2u5y5o9qCQEC1FKrq6X+vJ5HT0vd1fd+q/qq7u3zud9zmlhzbMM7TrGTSlm+cGGac5dk+JtHBviDu4f4pdObuPaEepLRA++wC63lri05mtMRLlvZRDp5YIueXrQ8xk3P5/nx5iKnL2ogMtl5GlrLT1/PMa8+xumLsgf0+uxhLVx+vOHHL+bZOGS5eOkbr5MXWG58YYSWdJQPn9JM+gCmDxYRmQnWWryeVylseICGqz6Dk66vdUkiIiIiRxzHcbjyykvo7x8ADG1trTiOQ1tbC+l0CscxXHjhGlatGqZYLNLS0kwqNf1LoVgLBc/SmHTfNhuRiIiIHF2mvLJ3d3cnhUKRRx55jCeeeJpdu3qm3PUpIofmP54e4YafDvDc7hIlLzygaWGttfz89Qm+uraXM+an3hZ8AiQihsuPy/L4tgl6x6eh5XIKrLX8+IUxcqWAXzytkenMXY0xnL0ozTUn1PPdx4Z4ZbCy39ds95jHw6/kufr4LHWxA3t7NMZwUleSpa1xbn1+lGA/s8Baa+nL+3zroX5O6k7xiVPefU1VYwwRx7B6YYr//cF5XLUiy3cfG+QPfrKL53tKBAc4Hez2EY/7t4zzwVUN1CcO/O0+HXP42CkNPPb6BJv6S3tfq6GJgEdfy3PJ8gypA3x93nhOsKQ5zvEdCe7alNv7Ou0JhR99dYJfObOJhgMMaEVEZoS1jP/sP3HSDaRPv0LXuyIiIiIHwRhDJBKhq6uTrq4OIpHq733pdIq2thaMMXvD0IUL55NOpw7LdVdgoeSFpKf4+6yIiIgceaZ8tq+rS/MLv/AhIpEInufxsY9dT3Nz0+GoTUT2Y6wUcsvzo9yzOcfnbt3Jb920g++tH2HbSAU/3P96jtZa1m0r8NV7ezl9XoovXtRO3VuCT6j+QnLekjocY3hoa37G1uW11tKT87l5wyhXH1/PoqbYtP+i4zqGXzmjiZZ0hG//bICi9+bnZq3l3pfGiTiGi4q1nfAAACAASURBVJdmprT/eMRw7Qn1PLmjwKtD5be9bpXA8n9+NkDFt3zuvNbJ7s333q4xhsZUhM+c08I3r+nGD+Hzt+7krx4eoHfce9efT2gttz4/Sl3M4dLlU3s+xhjOX1LHvIYoNz4zQmAnj6HXJyj5lguPrTuon0/Eqa6R+uyuAttGKgAUvJB/fHyIVZ0Jzl9ycNsVETkcrLX4/duYePoeMms+jFPXWOuSREREROQQhNZS8hV+ioiIzAVTPtsbY2hoqOfss89kzZrVtLY2a7BaZIZYa3li+wQjhYC//EA337imm+WtcW58doRf/8F2vvST3ax9eZzhgk9oq0Fo9XsKfPWeHk6dl+KLF7dTF3978LlHe12EsxaluGfz+NsCwsP2vIAbnx3BGPjISQ04h+k9pSHp8rvntbKxt8RNG0YI9wkPx0ohd27KcfGyDG11U5t21RjDmsVpGpMut7+YY99XLbSWW54b49HXJvjcea3Mb4hO+T3TdQwndye54dpufntNKz97Jc9nfriDH78wRqHy9s5fay27xzzu2TzOtavqaUpNvZsyHXP46MmN/Py1CV7qL+GFlrs35zilO8m8htiUtwfV12n1wjTpmMsDW8ax1nLP5nFeGSzz6dXNxCM6l4jIbGIZf+QmnHiK9JlX6XpXRERE5AgXhFD0rMJPERGROUBne5EjiB/CXZtynNCZ4PiOBKfPr4aZ//SJhXzhgjb80PIX9/fzq9/fzjcf7OeZXUV+/voEX7mnl5O7U3zp4nYy7xJ8AjiO4YoV9WwbqfBCb/Gwdn9aa/EDyws9Je58McdHT26kPXP41ns0xnBKd5JPnNrAvz01wsbe0t6A+NHX8owUfK5emT2obTckXN5/XJb7t4wzkPeB6vPbsLvIPzw+xEdOamDN4oPvbDTGkIo5XL+qnr/7yHzOPibN//7ZAL932y6e2VV821S4t20cI+rCFcdlD2qfxhguWFJHZ32U/3x2lC39ZTb1lbhiRRb3EMb/G1Mu5y1Jc9/L47w2XOHfnxrmkmUZTuyc/vVcRGTmWWsJgoAgCN5xJoIwDPH9/X99NvEHdzLx5B3Unf1B3GxLrcsRERERkUMUWEvZD6e8jIuIiIgceQ5fyiAi0+714Qobdhf5/QvbiUxeqzvG0JKOcPlxGS5dlmH7aIWfbs1z/5Zx7nwxR2gtZy1K86VL3jv43OOEzgSLmmLctSnH6fNSBzRF64HYM9Bd8i3bRyo8taPAutcn2Nxfpq0uwtUrDy6omwrHMXz8lEae2lHkrx8e4IZru4m5hlufH+OMBWmOaY4fdA2XH5flpg2jPLAlz8dPaWCoEPCtB/tZ1hbnl85owp2GhUyNMXRkInzh/DYuXZbhO48N8d9v28X7l2f51OlNdGYj9I373PlijmtOqKd1il2s+6qLO3z05Ab++uEBxooBTakIp807tLVXHGN4/3FZbn8xx5+t7aPkWz51etO0rvEqIrVhrWXz5i08/vh6AM4++0yWLl289z3DWktvbz/33/9TKhWPhQvnc/75ZxOJzL7LUWtD8o/+CNwodauv0c0ZIiIiIkeBim/xA0jtZxkgERERObrMvtEmEdkvay33vZwjG3d538K3B1DGGCIuLG6Oc0xTjI+d0sjG3hIv9hW59oQGsgcYfAIkIob3H5fhX58cpi/v05mNHlLd1sJYKeClgTKPvT7BUzsK7Bz1SMUcVrYn+MzZLbxvYYqG5NSnZz0Y6ZjDfzu3lc/fupPvrR/hlO4kWwbK/Nrq5oPuajTG0F0f5dzFddy+cYz3H5fh7x4dZKwU8JXLO6d1Wh1jDK6Bk7qSfPOabu59Kcc/PzHMz1+f4FOnN9I/7mMtXL2y/pB+oTPGcOGxGX7wzCiPvDbBL5/RRDZx6M9jWWucZa1xntlZ5LPntBzUVMAiMvvk83keeOBnfPCDVxMEAbfddjddXR3U1aWB6vlg7dqHOP7441i6dDE/+MEtLFgwj6VLF9e48rcLhnvJP/4T6s66DrehvdbliIiIiMg0qAQWP7Skour8FBEROdop/BQ5QuTKIfdvyXP+sXU0vkdIWJ0i1XD6/CSnz0/u/dyBMsZw/pIM//LkMA+/kuejJzdMOZyy1tI77vP0zmp353O7S4yVAjoyEU6dl+K/nNXCyvYEjSkXx0ytvkNljOG49ji/fGYT331siIdfybOkJc5JXYc29apj4JoT6rn/5XG+8UA/67ZN8IcXt7O4OXZYnp8xhmTUcM3x9axemOZ760f4u0cHKXiWXzi1kY7sob/FZ+IOHzmpgW8/PMClyzLT8jxiruHy47LkyyHXrTq0gFZEZo+enn4ymTra29sASyqVpK9vYG/4CeC61YEmYwyO40zbzALTyVpLft2tYC11Z1+v9ygRERGRo0TFnww/Ne2tiIjIUU/hp8gRwFrL+h0FhiZ83r/8wKeGPZQB245MhNUL09y9uTp9ajJ64Nuy1rK5v8yXbt/NRCXk2JY4HzyxnjMWpFjUFCc1ua1aDig7xnD9qgae3F7g0dcm+OPLOkhEDq0eYwwr2uKc0JngvpfH+fipjVy8dHoCw/fab1tdhN89r5VLlmX4z2dHuG5VPc407NcYw8VLM7w+XGFxS3waqq1u87zFdcxriM5Yt6+IHH4TExMkk8nJQNOQSMQpFAp7v26MYdWqldx11/088cR6jDF0dnZMeT9haAnDYPoKf4tgbIDxdbeSPP1KqG/H9/3Dti8REZGjm8F1Nb2ozB6VICQIUfgpIiIyByj8FDkCBBbu3JRjRXuCY1tiM7JPx8CVK7L8wU92s7G3xGnzDqwr0lrL4ITP1+/rozUd4a+ub2defYyIU9uwc38SEcPvnNtKybesOSY9LfVFXcN1qxrwQ8uvv6+ZyMHOoztFe6bCXdWZYGVH50FP37s/2YTDb53dwnTODNScdmlOH9r6oSIyu7iui7Xh3n+HocVx3njjKBZLPPTQo3z849fT3t7Gvfc+yOOPr+fCC9dM6b3AWovvH6bw01rG1/0Y65VJrr6OIAjf+3tERERkv4wxe2d9EJkNip7FGA75xmcRERGZ/RR+ihwBto9UeGZngc+f30Z0BsO0VZ1JFjRGuWtTjlPnJXmvPVtrKfmWG346wFgp4IZru1nUeHimfJ0OxhiOaYrxvy7vnLYORGMM5xyT5viOxLSsj3kw+5/u3+P2TKM83dsUkaNLU1MjY2M5giDAWhgfz9PU1LD365VKhTAMaW1tIR6P0d3dyauvbpvyflzXwXWnpxP9rfzcIOUnbqPujCtJdS3GOBqwFRERETlaFL0Q14GYwk8REZGjnsJPkVnOWsv9W8ZJxxzOWjQ93YkHKhk1XLY8y388PUL/uE9HNvqujw8t/OuTwzy+bYKvXN552Na6nE7GGFrrpvetMBl1SE5nm6SIyBGgvb2VRCLB2rUPEYYh2Wwdra0t9PT00dPTx4knrqS7u4s77ljL/PldPP30c1x66fm1Lnsvay0TT95JWMxTd+5HFHyKiIiIHGUKXohjDHF1JIuIiBz1dLYXmeXylZD7Xx7nvCV1NKZmdn1EYwwXHltHEFoefjWPtfYdH2ut5b6Xx/nBMyN8+sxmzpnhoFZERGorEonwoQ9dTUdHG93dnVx33ZW4rkssFiWdTuG6LtdcczkrVy4D4Prrr+LYYxfPmnNFODFK/mc/JH3qZUTbFtW6HBERERGZZsVKtfMzrs5PERGRo546P0VmMWstz+ws0jvuc/lxWZwaDBB3ZqO8b2GauzfnuOb4ehLRt9dgreXF3hJ/9fAAFy3N8JGTG3Ac/TIhIjKXGGNIpVKccsqJb/p8c3MTzc1NAMTjMY4//rhalPeurLVMrL+HID9M5ryPwiwJZEVERERk+hS8kKhjZmw5IREREamdmnR+BvkRcg/8O/5oXy12L3LECC3ctSnH8tY4S1sPz/pm78UxcOWKLK8MVtjUV3pb96e1lv68z9fv72N+Q5T/uqaFmH6REBGRI0hYyDH+8I2kTryIaOeSWdONKiIiIiLTZ6ISkog6aNZbERGRo19NTvdhfpTRO/4Or+fVWuxe5B1Za991ateZtnPM46kdBa5Yka1ZoGiM4cSuJPMaoty1KcdbX52iZ7nhoX7y5ZA/vLidhqSrQWMRETliWGspPHsfwUgfmQs+AUajYSIiIiJHo4lySCJicDVmISIictSryeiOSaQxkShBfqQWuxd5R33jPnduyhGEtQ9ArbU8uGWceMRwzjF1NQ0UU1HDZcszPPLaBAN5f+/n/dDyL08O8dSOAn9wURuLmmIKPkVE5IhiS3nGf/oDkqvOI9a9TOcxERERkaOQtbba+RlxcLVMj4iIyFGvJuGnE0tiYgnC8eFZ1WUnc5u1llufH+MvH+jnwa35mh+bhUrIvS+Nc+7iOprTbk1rMcZw4bEZvMDyyGsTeztk1740zo3PjvJrq5tZvSitAWMRETmiWGspblqHP7CDzPmfAKe251sREREROXwKXkgiatBKPSIiIke/2szrFYniJDME40M12b3IW1lr6cn5/OTFMeoTLt96sJ8Nu4s1C0CttWzoKbF7zOPy47I4syBU7K6PcsaCFHdvylEOLBt7S/z1w/1cuizDh09smBU1ioiITFVp63oi7YuIzVuum3hEREREjlKW6k3mqZgDuuQTERE56tVm2ls3ipPKEuQUfsrB80PLht3FaZui9tYXRjHAN67pYmlrnD9d28f2Ea8mAWho4a5NOY5tibO8LT7j+98fx8CVK7JsGSjz8Ct5vn5/H4ua4vz2mlaium1SRESORKGPt/NlYl1LMdHZcb4VERERkelnLRQ8SzrmKPsUERGZA2rT+ekYnLpGwvwI2LAmJciRLbSW2zeO8Qe37eL5nkPv0OzJ+dy+Mcc1J9SzrDXOH17STjxi+NraXoYLwYwHoD05jye2TXD5iizxyOy4LDfGcHJ3ks5slD9b20ehEvKli9uoTzjqlBERkSNSWBjHG9xBbMGKWpciIiIiIodRaKFYCUlFazMUKiIiIjOrRmd8g5tpIsiPQKjwU6bGWsujr03wfx4ZxA8t31k3RMk/+HAytJZbnh/FdeCa4+sxxtCRifDHl3bQM+7xrYf6KXoHt/3qdLoeL/aW8EN7QCGqtZaHtuZxHcO5i2fXOprpmMOlyzM4DnzxonYWNsZmVX0iIiJT4Q/3EBbGiXUv0/lMRERE5ChmraXoTU57KyIiIke9mp3x3UwTYSGHDfxalSBHIGura03++f19nL0ozf+6opNN/SXufSl3UN2Z1lp6xjzueLHa9dmeiQDVLsflbXG+eFE7T2wv8J3HBvGCqW3fDywPbs3z2zfv4DM37eDP7+tj20iF8D3qLHqWe14a55xj0rTWRab8nA4nYwyXLs/w3y9o48yFKQ0Ui4jIEc3r2YqJxYm0zK91KSIiIiJyGPkhlIPJaW81liEiInLUq0myYozBzTQTlgvYSgniyVqUIUcYay3bRzy+em8vi5pifP78VrIJl8uWZ/mXJ4ZZvTBNW11kShexFrjl+TEi+3R97mGM4exj0nz2nBa+/bMB2jNRPnpSA47z7tu31jJcCPiHx4e4c1OOc45Js7I9wU0bRnnktTwfOrGBD57YQFPKfVut1lpe6C2yY6TC585rnZXrUMyrjzKvPqpfFkRE5IhmraWyfRORpi7cuoZalyMiIiIih5EXWip+NfwUERGRo1/N2sqcTBO2XMBWikBjrcqQI4S1lqFCwFfX9hJ1DV++pIOGpAvAp05v5Oev5fmP9SP8t/NacaeQye0e87hjU44Pn9hA2366LB1juOb4enrHPf5+3SDtmQgXLKl7x+AvCC3rdxb464cHGC0EfO68Vq5YkSXmVjsmf/jsKD/cMMq9L43zS2c0cfHSDMmo2bs9a+GuTTmOaY6xsiMxKwPG2ViTiIjIlIUBlV0vE+s6FiKxWlcjIiIiIoeRH1gqgaa9FRERmStqN+1tuh4bBoTF8VqVIEeQiUrINx7oYzDv88eXddCZrXZ4GmPoykb55GlN3P7iGBt7Swc8/W11rc8xoq7hA8dn3zHUi7iGXz2zmXMX1/GNB/p5YT/7sNaSKwX8/bpBvnT7blrSEb79wXlce0I98Uh1SpW2uiifOaeFv/3wfFa0J/jWg/18/tadPL69gBdU1wPty/use32Cy4/LkogoZBQRETlcwuI43sB2YvNX1roUERERETnMvMBSCdT5KSIiMlfU7IzvpOoxGMJCrlYlyBGi7If87aODPLuryJcvbWdZa/xt09NeuTLLMc1xvvPYIGX/vcNPay27xjzufHGM606of8+1NRMRw+fOa2VhY4w/vbeXnaPe3gA0nFyH9Pd+vItbnh/jV89s5utXdbG4Ofa2QNUxhiXNMf7nZR385TVdOMbwpZ/s5iv39LB1sMxDW/MYDOe9S3epiIiIHDp/uIewME6se5nOuSIiIiJHuXJgCS0kowo/RURE5oLahZ/JOnBcgvxorUqQI4AfWr63foS7N+f4wgVtnD4/td8BynTM4TdWN/NCT4n7Xh5/z+5PC/zouVFiEYer36Xrcw9jDA1Jlz+6tB0MfO2+PkaLAYVKyPfWj/D5H+8C4IZru/nF0xpJxZx33KYxhohrOGN+ihuu6+b3L2pjy0CZz960k396YoizFqVoz9RsRmoREZGjnrUWb/dWTDROpHVercsRERERkcOs5IUAJBR+ioiIzAk1SVistRBLYuIpgvGhWpQgR4DQWu7alOPfnhrm11c3c8myzLsGiqfNS3Hx0gz//MQw71uYpiXtvuPjd4563LVpnI+d0kBr+sD+GxhjmFcf5X9c0sEXb9/N1+7rwwssz+8u8qGTGvjU6U1k4u8ceu5ve8mo4coVWc5alOaW58f48fNjXLEii/pPREREDq/Kjk1Emjpx6rT2vIiIiMjRruRVb5LXEkMiIiJzQ01udxqcCPjaT8d4zcsQjg8f8BqNMndYa1n3+gTffniA61bV89GTG3Gdd79AdR345TObKPoh3396hHc6rEJr+dFzoyQihqtX1k9pqjtjDCd0Jvj9C9t4akeBwbzP16/u4rfObplS8PnWbTalInz6zCa+87H5nNS9/+5WERERmSZhQGXnS8S6jsVEYrWuRkREREQOs5Jf7fzUtLciIiJzQ43O+JZ1Oyq8bjoIxodrU4LMWtZaXuwr8fX7+li9KMVvrG4h6r53GLinM/MXT23kto1jbOov7TdY3znqcdfmHNefWE9L2p1yfcYYzl9Sx19c3cVff3AeZy5I4TrmkANLYwztmSixA3iuIiIicvDC4jjewA5i81fUuhQRERERmQEFL8Q1aMxFRERkjqhJ+FkXd2lMRdkd7Zyc9ladn1IVWsurQxW+em8vi5pifOGCNpLRqXVmfuCEeuY3RPnuY0OUgzcfW6G13PzcKKmow1Urptb1uS/HMZy5IEVz6p2n1hUREZHZyR/pJSzkiM1brvO4iIiIyBxQqITEXOeAbq4XERGRI19Nws+Ya+jIRuiJdhHkxyAMa1GGzBLWWvzAsrmvxLce7Oe/3ryTmGv4o0s7aEhMPVysizn8xuoWntlV5KEt+b3dn9Zadox43L0pxwdPbKD5ILo+92XMoXd7ioiIyMzzdm/FRGNEWufXuhQRERERmQETlZBoxCj8FBERmSMitdipY2BBQ4zHo+1UihPYwMO4NSlFashaS9GzPLOrwC3Pj7F+R4GWdISPntLAlSuytNVFDnoNzTMWpDh/SR3/+MQQpy9I0ZKOYIGbnxslHXO4ckVWwaWIiMgcZK2lsmMTkaZOnHRDrcsRERERkcPMWjvZ+WmIOBoLEhERmQtqljguaIpxd7SRfN7H+h7EkrUqRWaYtZbhQsBPX8lz2wtjvD5cYXlbnN+/qJ1zFqXJJpxDDiYjjuFXz2ziszft5MZnRvjMOS1sH6lwz+Ycnzq9iabUoXV9ioiIyBEqDKjsfIlY17GYaKzW1YiIiIjIDJiohMQjhkhN5sATERGRmVaT8NMYw/z6KMVIHaN+DFsuQCpbi1JkhlhrCS3sHPW4a1OOu1/KMV4KWb0wxW+vaWFVZ5J4ZPqmkTUGFjbG+OjJDXxv/QgXLs1w14tjpOMOV6jrU0REZM4Ki3n8ge2kTrwA0PWAiIiIyFxQqIQko4d+s72IiIgcGWrW+dmRjeJGE/STJSzmobFWlRzZ9kwdG3Wrd69N10XcnnUyp2N71lrGSiF/88gAD23Nk4gaLl2W5aqVWRY1xXDM9NW9L2MM15/YwANb8vzF/X3sHvP4lTPV9SkiIjKX+SM9BIUc0e7lGvwSERERmSMmKiGpqINmvRUREZkbahZ+NiRdsskIu00LYXG8VmUc0fZMH/vVtb3EXcMvndHEcW2JQwoTrbXkyiGPvJpnVWeS+Q3RQx4YLAeWb/9sgEdezfPp9zVzybIMLWl3RgYcs3GHX1/dzB/evpuObERdnyIiInOYtRav5xVMJEa0dX6tyxERERGZE6y1VCoe27btqM7UtXA+0eibx5vCMKSvb4DBwWE6OtpoaWmavhv8gYK3p/NzWjYpIiIis1zNws9k1KE1E2V3pIMgP1qrMo5Y1lpGiwF/uraXl/vLZOIOv/OjnVy2LMsnTmtkfkMUZwpXdNZaxssh9788zn8+O8rrI9V1OP/8qi7aM5GDvuAMQsv314/wwJZxvnRxO+9fnpnR8NEYw1mL0ly0LMPK9gSNSXV9ioiIzGWVHZuINHbg1GnaEREREZGZEAQBt912F67rEgQhzz+/iWuvvQLXrY7RWGvZsOEF1q/fwKJFC3jooUe4/vqr6O7unJ4xJFud9ra1LqJFD0REROaIKYef1lqKxRLPPbeR8fEJVq5cRldXx5QvRiIOzKuPsj3ajp8fxVqrjrwDZK0lVwr5s/v62DJQ5mtXdbK4Kc6dm3Lc+MwIP30lz/Un1vPBVQ00v0eH5Z7Q84Et49z47Ch9OY81i+v4tdXN/O2jA/zp2l6+ekUnDcmpd2paa3lgyzj/+tQwnzytkUuXzWzwuUfEgd85t0VrO4iIiMx1YUBl50tEu47FROO1rkZERERkTujvH2R4eIRPf/oXsdbyj//4HwwODtPe3gqA53k88cQzXH/9lTQ1NbJo0XzCMJy2/QcWip4lFXOmbZsiIiIyu005/Nxzt1ZTUyOtrc3cfPPtfOpTH6GxsWHKO1/YGONpt5nCRA/ZKX/33GStJV8J+csH+niht8RXLu/g5K4kxhg+fkoDFy2t4+bnRrl5wyj3bM7xC6c2cdlxGepibw7+rLXkyyEPbs1z47Mj9IxVQ8//cWk7y9sSuAZa0i5fvqOHv3ygny9f2k46duDhobWWF3pL3PDQAOcvqeOTpzXh1mhhBWMMbXXRmuxbREREZo+wNIHfv53kCefVuhQRERGROWNoaJj6+noikeowZCZTx9DQG+HnxESBYrHEE088zcDAEK2tLVxyyfkHdRP+/gShpeyHpCfDz3d6nIiIiMysw9msNuXwc3BwiImJAh/4wOWEYUhTUyOxWGzKOzbGML8xRj5az9jE63RMeQtzU8EL+eaD/Ty1s8CfXNbBGfNTew8QYwztmSi/dXYLV6zI8h/rR/ibRwe4beMYv3xGE+cckybmGvKVkIe25rnxmRF25zzWHFPHH13yRui5Z3sndSX5o0vb+f/u7uV//2yAz5/fRiL63gejtZaenM+fre1jfmOU3z2vlURUd9eJiIhIbQUjvQSFMWLzlms2CBEREZEZ4vs+kcgbyxC5rkMQBHv/7Xk+g4NDXHjhGi64YA133rmWp556lnPOOXNK12xhGFIuV972+YIXUqgExAgpFkuH9mRERETkkBljSCQO74xcUw4/h4ZGyOcnuPXWOygWy7S0NHHVVZdOecfWWrrrI/iRFAMFWI5FN169u6IX8lc/HWDdaxN8+dJ2Vi9MA2+/Y80AixpjfPGidj5wfD3/+uQwX7mnl5O6kpyxIMU9m3PsHvNYszjNH17SzvLW+Ju6Mvfd3lkL03zhgla+8UA/2YTLb6xuJuq++4XneLnamVr2Q/7s4uqUubqrTkTk6KYgSY4ElZ5XMJEY0db5tS5FREREZM5IpZKUSmWgOuZULldIJhN7vx6LRWloyHLssceQSiVZvvxYXnrplSnvx3Gc/Q6keiakEkB9OnbYB1pFRERkdphy+On7PoVCkauvfj+JRJx///cf8uqr21i2bMmUBj49zydlfBLxCDuKUSbyExjHfe9vnKNKnuX/Pj7Kg68W+MKaRk7rcCkWi+/5fUvq4csXNvD0rgQ/eC7PPz8xxNkLU3z+nHqWtsRxHUu59O53va2ZH2X4jHq+88QoCRPwkROzuO/QyFkJLH+7bpSNvSX+58VNtCYCCoX3rlNERI5Mxpg3DVyIzFbWWirbXyTS0IFT11TrckRERETmjPb2NkZHxxgaGiEMQ8bH87S1teL7Pr7vU1eXpr6+ntde28bSpYt59dVtdHd3Tnk/xpj9jk36YYgXWtIxB8fRzGQiIiJzwZTDz7q6NE1NjdTVpXEch+bmJnK58SnvOBqN0NaYpjkdo6eYIZWIYSJTnz53LqgEln99apAHXinwBxe1c+nyDFPpr0kBFx2X4qwljQxN+HTWR3mP5s23+dhpSUqhw7+vH6G1PsHVK+t56/WktXDbsyOs3VLgCxe0cc7S7JTqFBERETlswoDKzpeIdi3BRHXNKSIiIjJT6uuznH32mdx66x0AnHfeWWQydWzd+hovv7yVK6+8lKuuupS1ax/i8cefpqmpgdNOO3HaZpcp+5YghFRMwaeIiMhcMeXws6OjHd/3eemlrdTXZ9m1q4fVq0+f8gWJMYa4C53ZCDvzGazv4UQ19cS+rLV4geU764b4ycYxfu+CNi5dnsE5yIu/VMyQOoj1WQFcBz51ehO5csi3Hx4kk3C5YEnd3p+7tZZHXpvgO48N8dFTToFdxwAAIABJREFUGrhyRfag6xQRERGZbmEpjz+wneTxa0C3Z4mIiIjMGGMMp5yyihNOWAFUGyKMMSxevJBFi6rLEXR2tvPJT34E3/eJRqPTuqxGJQgJQ0syqvBTRERkrphy+JlKJbnuuiv5+c+fwPM8LrxwDe3trQe1c2NgQWOMJ3bWUfF8ksmD2sxRyw/hn58c5ubnRvmdc1u5/LjaBYrGQCzi8JtntTBWDPiL+/vIxF1Om1f9ob08UOYbD/axemGaXzmjmchUW0tFRETkiGatxfN8du7chTGGefO6iEajb3vMwMAQw8MjtLW10NjYMGPrxQYjfQQTo8TmLdcatSIiIiIzzBhDLPbma0PXdXFd9x3/PV3KviWwkFL4KSIiMmdMOfw0xtDV1cGHP3zNmz53sBY0xrg7SDBRsSj7rLLWUvIt3396hO8/PcJvnd3CtSfU4zq1H6hLRg1fuKCNr9zTy1fv7eXPr+6iOe3yZ2v7aElH+MIFbSSjta9TREREZlYQhNx++z0EQUAQhGzYsJEPfOD9ewewrLU8//yLPP74eubP72bt2of40IeuprOz47CHkdZaKj2vYNwo0dYFh3VfIiIiIjK7lHyLBRIarxIREZkzphx+wqGFnW/dzvx6lyIxhoshLdOy1SOTtZbQQt+4z8Ov5rl3c47Xhiv82upmPnRSw6wIPqH6M8vEHf7wkna+fMdu/uSuHlrqIoyWAr51bTfNKVfdFCIiInPQwMAg/f2DfPrTv4i1ln/6p+8xNDRMW1t1hhDP83nssae45poraG1t5phjFlKpeDNWX2XHJiKNHTiZphnbp4iIiIjUXqESEpmc0UxERETmhoMKP6dTe8rgOIbeosOyWhczwyyAtUxUQp7rKXH3phyPb5/AMYazFqb57DmtnDwvSWSWBJ97GGNoTrn8z/d38KWf7GZTX4mvXdnJkuaYgk8REZE5anBwiIaGeqLR6uVlJpNhaGhkb/g5MTFBsVji6aefo79/gI6ONi68cM3MXDuEAZUdm4l2LsFED279cxERERE5MhW9ENcxxLREk4iIyJxR8/CzPuGQcXx2jAVYa+dEeGatJQhhx2iFB7bkuW/LOD05j2Nb4vzG6hbWHJOmLRPBMH1dttPNGENXNspXrujklcEyqxelZ22tIiIicvj5vk8ksu+aTQ6+7+/9t+f5DAwMcd55Z7Fmzfu44461rF+/gbPOOmNK1xBBEEy5YzQs5PD6txFZ+j5KpcqUvldEREQOjjGGeFw3SUvtFSohEccQj+hYFBERmStqHn4mI9ASrbBt1H/vBx/hrLXkKyGPbytw16Ycz+4qko45nLckzWXL21nWFifumiPmFwNjDMc0xTimSb/MiIiIzHXJZJJyuRosWguVikcikdj79VgsSn19lmXLjiWVSrJy5TJeeumVKe/HcRxiseiUvscbGCIsjJFatHLK3ysiIiIHS+MEUnvWWgpeSMSFqDo/RURE5oyah5+RaISuSJ4dYx6BhaP5JqySb/na2l4e21bghI4EX7igldUL0zSm3Fnd5flujsSaRUREZPq1t7cyMjLK6OgYYRiSy43T1taC7wcEQUA6naa+Psv27TtZsuQYXn99B11d7VPejzEG13Xf+4GTrLX4fa9h3Bix9oVT+l4REREROfJNVEISEQdXY1giIiJzRs3DT+O4zIsVeHHco+SF1MWPzgGp0Fpu2jDKE9sK/MllHZyzuI6oo/BQREREjg4NDfW8732ncdNNPwHgnHPOJJvNsHXra7z88lauvPJSrrjiYtau/Snr1j1JNpvhtNNOmpFrocqOzUQa23EyTYd9XyIiIiIyu0xUQhJRg+vUuhIRERGZKTUPPzGGefESY2M+Y6XgqAw/rbU8t7vIvzw5zEdObuC8JXW4jkJPEREROXoYYzj99JM58cSVAMRi1WnxjzlmIQsWzANg3rwufumXPorn+TO3BlgYUNm5mWjnEkw0fvj3JyIiIiKzykS52vnpqAFBRERkzpgF9zwZOhM+XqVM//jRue7nSDHghocGWNYa55OnNyn4FBERkaOSMYZ4PE48Ht8bbEYi7t6g0xhDJBIhmUzgOM6MhJ9haQK/fxux+SvQ2mMiIiIic0/BC0lGHXV+ioiIzCE1P+0bY2hOOsS9CXaOeVhra13StPIDy3cfG2Jwwudz57dSF6v5Sy4iIiIyZwSjfQQTo8TmLddyAyIiIiJzjAUKlZBk1Og2OBERkTlkViRxmbok9f4o20YqtS5lWllreXDrOHduyvGbZzeztCWuQTcRERGRGeT1vIpxo0TbFtS6FBERERGZYdZC0QtJqxlBRERkTpkVZ/5EOk1bMMj2kQpHS9+ntZbtox7/55FBLlhSxxXHZRV8ioiIiMwgay2VHZuINLTj1DXVuhwRERERmWHh3vDTrXUpIiIiMoNmRfjppuvpDPrYPebhBUdH/Fn0LN9+uJ9k1PDZc1qIugo+RURERGZUGFDZuZlo5xJMNF7rakRERERkhoXWUvQsKXV+ioiIzCmz4szvprJ0BwMMTfjky2GtyzlkobX8cMMIz+wq8rnz22iti6jrU0RERGSGheUJvL5txBasAF2LiYiIiMw5QQhlX9PeioiIzDWz4sxv4mk6zSiFksdQwa91OYfEWsuGXUX+7akRPn5KI2fMTyn4FBEREamBYKSPYGKUWPdyXY+JiIiIzEFeYKkEllTM6HpQRERkDpkV4acTT9LqFHC8Ij1jXq3LOSQjhYAbftrPcW1xfuHURlxHF1YiIiIiM81ai9f7KsaNEGlbUOtyRERERKQGvNDiBVZrfoqIiMwxsyL8JBKjPg5pP8f2UQ9rj8x1P73A8p3HhhguBHzu/DZNqSEiIiJSQ5Udm3Hr23AzzbUuRURERERqwAuq4afW/BQREZlbZsWZ37gRUskEzf4g20YqtS7noFhreWDLOHdtzvGZs1tY0hzTdBoiIiIitRKGVHZsJta5BBOL17oaEREREamBim/xQ0hGNUYnIiIyl8yK8BPHJZrO0FHpZcdohSCsdUFTY61l+4jH3zwyyMVLM7z/uKyCTxEREZEaCssTeP2vE1uwAtB1mYiIiMhcVPRDDJCIzI4hUBEREZkZs+bM72aa6PJ20z/uU/SPnPTTWkuuHPL//7SfurjDb53dgq6nRERERGorGO0jyI8Sm7dcN6WJiIiIzFFFL8QYSKjzU0REZE6ZNTGdm22ms7KbsYLPWDGodTkHxFrLjlGPL9+xm5cHyvzeBW20pF0NsImIiIjUmNfzKsZxibQtrHUpIiIiIlIjJc8C6vwUERGZa2bNmd+ta6LNH8D3yvSN+7Uu5z2F1rJ+Z5Ev/HgXI4WAv/xAF6fOSyr4FBEREakxay2VHZtxG9pwM021LkdEREREaqTohTgG4hGN14mIiMwlsyL8NMbgZJposHkSQZGdYx7W2lqX9Y68wPLjF8b4ozt2c0xTjG9d283xHQkFnyIiIiKzgQ3xel8h2r4IE03UuhoRERERqZFCJSTmOkRdjdmJiIjMJZFaF7CHW9dI2hZpsONsH6nUupz9stYyUQn57mND/PiFMT54YgOffl8Tqaij4FNERERktghDgtwgsUWrQNdoIiIiInPWRCUkFjFEHF0TioiIzCWzJvx0UhmiEYeOSIHtIxUsMJsuS6y19OR8vvFgHy/2lvj8+W1cuSKL66DgU0RERGQWsV6ZsJDDzbbUuhQRERERqaGJSkhc4aeIiMicM3vCz2QGx43Q7Y7zXM6j4lsS0dlxYWKtZcPuEn9+fy+hha9f3cXJ3UkchZ4iIiIis471SoTFPJH6Vt2kJiIiIjJH7ZnBLe4aIrNi4S8RERGZKbPm1G9iSUwswbxInqEJn3wlrHVJAPih5Y4Xc3zx9l201UW54dpuTlHwKSIiIjJrhcU81ivjZJpqXYqIiIiI1FChEpKIOjjq/BQREZlTZk3np4nGcJIZutwcRc8yNOHTkq5deXvuDvuXJ4e5acMoV63M8ptntVAX1/qeIiIiIrNZWMhhwxC3rrHWpYiIiIhIjVig4IWkYs6sWlpLREREDr/ZE35GYjipDM02R8SB3TmP5W2JGa/DWktoYcPuIn+/boitg2U+e04L161qIKL1PUVERERmvSA/gjEGJ1Vf61JEREREpIYKlZBU1EHDeSIiInPLrAk/MQ5uXSMZb4xs0rBjxMNaO6Nho7WWnpzPvz41zL2bcyxvS/DNa7o4sUvT3IqIiIgcKYLcECaexEmkal2KiIiIiNSItdXwsyMbVeeniIjIHDOLwk+DW9dEtOc12ltdto1UZmzX1loKXsidL+b4t6dGcAz813Nbufy4LMmoUbeniIiIyBHCWkuQG8SJpzFxhZ8iIiIic1VooehVOz9FRERkbpk94SfgZJoxr25gXsawfbRCEELEPXz7s9YSWHhmZ4G/XzfEK0Nlrjguy6dOb6I9E1HoKSIiInIECnKDOHUNGDda61JEREREpEaC0FLyLem4ozE+ERGROWbWhJ/GGNxsE7aUZ0Gd5alen6IXknEPT/pprWXnmMe/PDnM/S+Ps6I9wQ3XdrOqM4ljtLaniIiIyJEqGBvEqWsE5zDeRSciIiIis5ofWsq+JR1T56eIiMhcM2vCTwA300xYLjA/HZIrhoyWAjKJ6R20staSr4T8ZOMY/7F+hKhr+N3zWrlsuaa4FRERETnihQFhfphI2yLQdZ2IiIjInBWEUPZDhZ8iIiJz0KwKP510AwQ+7dESoY3RN+4zvyE2bdu31jI4EfAnd/ewub/ElSuyfPI0TXErIiIicrSwvkcwMUq8vqXWpYiIiIhIDfmhpRJYrfkpIiIyB82q8NNN12OtpYECyViCnaMVTpuXnLZgslAJ+eaDfewc9fjGB7o5uVtT3IqIiIgcTaxXJiyM42Zba12KiIiIiDDZjDA4xJNPPoMxhjPPPJWmpsa3jcdZa9mxYxc9PX2ceeaphzxeV/YtQQhJdX6KiIjMOQd99rfWsnXra+zYsWvaijHJDMZxSfrjtKRcto1Upm3blSDk/64b4umdRf7oknZOnZfEdTTNrYiIiMjRJCxNYL0ybrZZ13kiIiIis0C5XOaWW+6gq6uD1tYWbrnlDiqVt4/5FQpFfvSj23nppa3Ts18/JLSWZFTXhCIiInPNQYWf1lqGh0f4/vdv5vXXt09fMfEkJp7EKQzTVR9l+4iHnYbtBqHlh8+OcvvGMX7n3FbOXJjSYJiIiIjIUcgWx7F+BTfTVOtSRERERATo6ekjEolw4onHc8opqzDG0Nc38KbHhGHIww+vo62tZdrG7Eq+JbRo2lsREZE56KCmvfV9nwcffITm5ukdVDKxBE48hR0fZkFjjEdfm6DiWxKHcIeWtZYHt47zD48N8QunNnLlyiyOgk8RERGRo1KQH8E4Lk4qW+tSRERERAQYG8uRydRhTHUGtlQqRS43vvfr1lpefvkVisUiq1atZMOGFw5qP9a+uYWi5IVYa0lEzNu+JiIiIrV3OJsUpxx+Wmt55pnnqa/PkkolD3rH+7voMG4UJ5UlHB9m4YIYt20cI18OiUcO/gXYsLvINx/s56KlGT51ehOu2f++RURE5OBpRgWZLYLxYXCjOKlMrUsRERE5Yux3jOYgr++stZR8y0v9ZWIuHNsaJ+aq807e2fh4nnXrnuSqqy6jr6+fMLQEQYjrOlM6DoMgpFwu7/33aL6E6xhCr0KhEB6O0kVEROQgGGNIJhOHdR9TCj+ttfT3D7J58xY+9KEP8PDDP8fa6uenelHseT6e5735k2GITWQoj/TTHA8oeSE7h/IkTWxK295jx6jHn947xOKmKL92WpqgUqIwfcuIioiIzHkzcbEiMhVBbhAnkcKJpWpdioiIvMVbA7ZDvXlqOgO7qe7HQnU8BHAM7G+v01nLvjVYwA8sBxrlWFtdDsgLqqFk2Q8p+5aJSvjGn3JI0Q9xjCHigOsYXPOWj5N/jziGRNSQijnEIw7xiCHuGmIRh4gDfmj5i/v7eeiVPImI4SMnN/BfVrcQcXWz3FyVzWaZmChgrcVaKBQKZDJ1e7++fftOenv7uemm28jnJ8jlxlm37gnWrFk9pf24rvOmRo3ACYi6hmwmSSp1UJPfiYiIyBFqymf+9eufpa9vgBtvvIWenj6i0Qjd3R0sXrxoShf20WiEaPTtuy/UtxCM9rKgOUnUdRgqu5x0EB2mw4WAbz0ySDzq8MeXddGR1UWOiIiIyOFirSWXG2fDho0YAyeddMLe6c3e+ri+vgH6+wdZtWrFtA8MB7lBnFQ9RA/u5jmRo8GekGRPOAOT4cw0/H+z1lL2LY6BiGswHNx2rbV4oWW8FNKT8xic8Im6hphriLkOsYjZ59+ToctkIOMYcMyej9X9GybDp7eEUNZCSDX4CUKLHzL50VLxLeWg+nljwMFUP05u/61/dyefc8yt1uZOft5M7rNWsyDs+/PGQmir4VM5qD7HoheSr4RMlANy5ZCxYsB4KaA+6bKwMUZHNkp9wiUeMdN2nOxbW2BhvBSwc9RjU3+Jl/vLxKOG1nSElsk/DUmX+qRLMuqQiBgibrUWgDAEL7SUvGpIN1oMGC0GDEz4DOZ9BiZ8yp4lm3CpTzrUJ6rbyiZc6mIOdXGHVNQhFnGIutXt+aElmNxu9TiohoGVyWCw6Ff3V/RCit6ev1dfy6IfUpr8e9mvvvbxiCEZNSQiDsnonj+GxL4fIw6J6BvHeDTyxvG97zHuGIMXVH9m46WQXClgpBgwUggYLvh7/z5eDvCCA5vVyg8tRa/6/PzJ47963O8TdDrmbQGunfwZ7hv0Vm+Ah8BawrdsI+JUn4vjwM5RjzAEg+XezeNcc3w98xp0bp6rOjraKBZLbNy4mSAICYKA9vZWcrlxxsfzLF9+LAsXzgdgy5ZX2LBhI2ecceqU9/PW969CJdx7XGqmGhERkbllyongueeu5rTTTgbgZz9bRzKZpLu7c8o73t9Fh7WWSLYZb9fL1Ect2YTDjtHKOz5+f6ytXtTf8FA/veMe3/hANx3ZiC5yRERERA6jSsXj5ptvZ9myJQSBzy233MEnPvEhYrHomx5XLpf50Y9up64uzapVK6a9jmBsECfdgHF145vMPvt2bgWTAUzJt0xUAvLlkJHJUCcM7RvhnvNG0Gf2CfwcYyYDjWooU6iEFLyw+rHyRmhT9KrLiKzqTHJSV5JFTTHq4s4BhVx76vVCGMh7vNhb4skdBTb2lIi4hq76KAsaoixojNFdH6W1LkpDwiEeddi3wcsCFd8yWgzYNebxylCZl/vLvDpUpifnU/De3L/21hDTUA1Wom41XHH36Upz9ulKizqGiMvkYwzB5Otb8kK8oNr1Vpn86IXV7qNwMsjZ81LsL0DdwzEQj1TDuUTUIRVz9oZrmbhLJuGQjVeDt3SsOsVnZJ+6o64h6vLG85gMT0P79oDprQqT3XlF740/E5U3/9wrk6HdSCFgohJQmPz5+4Hdu92oa/aGi+XJr0dcQ1PKpbs+yqKmGIub4yxojNGeiewNRd/tSNlzHO05XiqBZWgi4NWhMi/0ltjYW+K1oTLj5ZC6mENnfRRrLY8XQ/KVgNJkgOgYSMcc6uIujUmXplSEILQMFfxqYFuuPu89P69U1CEdd2hIuERcw47RCvnJ16jk2b3BoJk8PpIRQzzi4Fu793jwg8nXffJAtZPHQMStBnpRd89xZYhHquFmYjKgzcQd6hPV5x5O3hQwVPApeZbSZGelP3mseUE1eN+zv3c7xl0HSl41oNzzXKuhqqEu7u4Nihc1xQ54iaCoW/3euphDOlZ93dKxaiAbnwxtq+HQm79vb8hpq88xtNXwOLTV/0vlyXB9z/vOxGRAvXPUY/fYKKGZfE0n37Nk7komE1x//ZU8/vjTGGO47rqriMfj9PcPsm3bDrq6OshkqteM7e1tLFtWJh6PHdJYnrXV98g9N62IiIjI3DKlUSFjDJlMhkwmg7WW+fO7SSYTJBLTN92dk2kmLOSIm4COTJRtI1Obp9YP4buPDfL49gJfubyD5W1xBZ8iIiIih1lvbx9hGHLWWadjLbz88qv09w8wb17X3seEYcjPf/4kyWTi8FyfhSHh+DButpm3jeDKrPbWTkU/rA6shyF7A4JqZ1318Xs7/Sa/ZuGNwflwT0cS+4QOb3x0zGR4Zsxkp9Kbg7Q94eK+YZuBvV1o73Ts7nkO4eT0kiW/Oug6VgoYLgQMTvgMTfgMTvgMTgSMFPz/x96dBslx3nee/z6ZWXdV3wAa900QvElRlCjqIE2JFE2ZsmVK9tjjsT0eO+yd2cMxMTsbGzET3t13Gxt7OGLDa489sj0e27JsU5QsUqQk6uIpkgAJgABB4r6BvrvuI59nX2RVdTcIkGiwgeru+n0iOqpQVah6KjMr61/5e54nmaxEoWcrqIlGfc1+1rlhhGsmNK59L80pKg3xIApnWlNQpgJDKu4xkPapNhzf2DfJV386Rk/S54YVCe5al+a21UnW90dhaGvkomsGHJPlkMNjVV47WWL3qTJHx2tY59jQF+eu9Wk8Aycna/zkSJGx4hSVhiPmGXqSHqtyMdb1xVjfFyd0jkMjVY6N17hQaFALHdm4x5reGDesTPKFmxNsGYwzmAkwNEcsNqL1X2u0ApYoSGpN19kOrmaHSnYmaGqE0b8Dj+bIuyisunhEXjKItoFLrs/m8m4HPi4KrgrN0ZP5ShTGFarR5bl8g3IziKw0ohFxrf/ncO3Rcq2RdBCNjjOzXmvmPjDNdd8KyYA5I17j/sy6boVyqeY0pDtWJuhNzoRkPQmfXCIKaVNxj4QfhXm1hmO02ODUZJ1j4zWOjVfZfarMM2/nKdctMd8wkA4YzgXtbawV/M4EgaYdCIcW3hmpcOB8lbPTdULnWJkN2L4iya/dPcDOVUnW9cXIJXwA6uHMqNTJcvSZGC2GjBTqjBajz0zgGTYPxBnKBKzIzowS7Uv5pOPR+gy8aFm19x2zRm8WqtFnbLoSfRaLtZC4Hy2raLuIwr/ERcszaAaegWeItUYce9F68WaN9r30NhOtc9scZVxvbZ/N0Zazt+1LbeP10JFLePQ1R6/mEjPbcCyIQvPLbLZX5FoeH5kdgIfW8czbeZIxwxdv6WVVTp2SupkxhlWrVvLYY5+fc9v69WtZv37tnO1yzZph1qwZXpBttVS3JPzL7+tFRERk+fpQ1ec998x/Cor3YwA/14+rlTH1Muv7YhwZq9GwEPM/+P9b53hi7yRP7J3iv/vUCu7dlFHwKSIiInIdTE5O0dOTxfOi0U7ZbJqpqel2+Omc4/jxk4yMjHLXXbezd+9bV/U6lzrvWvs+2yAsjBNbvwMw7/tYub5a56cr1x3T1ZDpSsh0czrHyeYUjpPN65PlkEI1pFiLQq7WiKFoWtKZ0Y+tKUmNiYYWtcKFVijWsHNH07VG1zFrelOYHWI0pzZtBqOJwJBuju5rBRA9ySiM6E1G17MJD+dgrBRyIV/nQqHBSKHBWCl6j4VqSGij95+KRc/X2wxu1vfFuTMbMJTxGUwH9Kej507FvEsvw2b7LTOhnO9Foc17R0ReFJw5KNYsJyZr7DlTZtepMn+za5w/rlj60z47VyW5a12KDf1xDo/WeO1kibcvVChULSuyAbeuTvLzt/Zyy3CKVbmgPdrMuijkKFRDxoohZ6brnJyscXyixomJGq+eLGGAdX0x7tmY5oYVCTYNJBjOBWQTUWh1rba3lmv9a3D2NKCtbbAxO5CdNd3uxddrzRGZs0fz+t5MqD17G/cNzYA7GqE3e30HzdHB7zdq9ZLiMJCOwvDWe6mH0XkgR4sNTk/WOT5RY6TQaI8kniiHVPMNqg3bDKNndSwA1vbFuHt9ipuHB9g2FGdFNiAZ8y7ZpsAzpGI+A2mfDX1zZwloLdfZnR4+kGmeizIwRHsLgNj7/Y8F1Wqr11oRXuvWxeV6fDfGfcO/e2Alv3xnPzHfsLY3pvBJLnl87kpvu1rFmiUV8z5UhwERERFZmq46/LwmoaIx+NkBXKOGq5bY0N/Dy8dLzV6n759+Ouf48eECf/ziKF+5o4+fu7lX06qIiIiIXCfReb9mhzYGa2cOsJbLZX7yk5d46KGfYWxsvB3kwPzqyjC0VKvVS7ehnKdRnCZM9lIqla/qfSxnzkG1YRkphhyfqJOv2jnnNPQueT0KE1qjoeJ+6zI6d14raJs5T52jVLftEVznipZz+ZBzhZDzheh8dYWqxbqZ0ZXpmCEd98nGo2kkh7MePYPRvy812aZjZkRd688QdZacOX8ezfMyQqx9Djpz0fKYHUjNDqqiaV5LdUe+ZilUHYWaY6RQ5+h4rT3laKUeBVmO6Hx/PUmfvqTPQNrjtlUxBtMJhtLRyMv+VBSUpmJeO6i89IHYsPk335Ub/TcXQoPo72IesCkHm3Yk+cINSaYrIcfGa+w7X2PvhTr/+ZVxCtWQXMJn+4o4j9+c4ZbhOBv64mTiXjOAahDWGpQumpwnbSCdhfVZn4+tSeFIRaNf69FnPBXz8Nu7BwuuRq0C85vjZ2kwRHFbzAB+8+99H3057jLXm9uHjf4sC78cE8DaNKxN+9yzJjWnFa3gfXaHgtbnCCAb9wja00uGuHpIub7ADZQlY3XaXLvZHkSuQBR+RrWKiIiIdJdFN++Il+3F2RBbzrOhf6jdG7wnOfdXY6sXd75qOTZe4/WTJb7+xiSf2pLlN+4ZnPWDS0RERESutVwuS6lUagealUqFbDbTvv/w4eOcOHGGb33rO+TzBSYnp3j55de4996Pzut1fN8jnU5d8r6wlmeiWiI1tPqyj+kmoYV8NTrH4tsXKuw7W+GdkSrn8nWspV0vzx4F5GZdaV1vTSfrGdrTLXrN8ysmLppqNV+Ngs9KI3p8Ou4xkA5YmQ24Y22KNT0xVvfE6E83zz2X8Noj2WKemTVqanFpTUlab1pvAAAgAElEQVRqHXPOHVlunqsym4imz4wHzVGpnW3uFclmYM0gfGI7hM1pbi/kG6zKRdOKLsR66PnwTyEiInLVijVLUiM/RUREutLiCz+TOYwfEBYmWb0qhgPO5+usb06DU647Tk3V2HOmwusnSxw4X2GyHNKX9vnklgz/+pMrmtPciIiIiMj1Mjy8kkKhyNGjJ7DWUqlUWLVqBaVSmVKpzPbtW/jd3/11AA4dOsq+fW9z2203z/t13m/0iK0UcGEdPzdwXUeZtDrlFaqWI2NV3jhdJvAMG/rjrO2NMZjxSTdH/MHVz6By8VSFDtojFlvnkSvVLEfHa7x1rsJb58ocG68xVbGkY4b1fXHuXp/mltVJtg0l2tOrunbQ6Wi9RGuEF9B+7krdtUc9lhuWci0a5VmqRX91G53fb3VPjOFcjIFMcxrXIDp3XutdL8URQK0m+9Bej8tJYGAoE51PUUREZDlwQKlmSce8K58+W0RERJaNRffr1ktlMH4MW5hgYEt0oGj36TJjpZDXTpbYe7bM+XyDVOCxbUWCX7itlzvWptk8ECeb8Jrn2VBVIyIiInI9ZTJpHn30c7z00msYA1/4wsOkUkmOHj3BkSPHePDBT5NMDgJQKpUpFktkMukFrdtscRJw+Jm+BXvOy3HOUbdwIV9n79kKrxwvsudMVLOmYtF7Kjen/MwmPFZmA9b3xdk0EP2t7Y0xlInOndgKMOuhoxo6qvXWufSic+yV65ZizZKvhuSrlnyleVmdCR5LdUutEYWUoXUMZgK2DcX5yh393LQqyYb+OH0pn8BTrSwiIiLLn3VQqVv6Utfv3L8iIiKyeCy68NPEkphkhjA/TibuMZQJ+OpPx0nHot7z923K8pH1KW5YkaQ/7eMbHcARERER6TRjDBs3rmfjxvVzbtu8eQObN2+Y89j169eyfv3aBW9DWJjAeD5eeuEn27x4dOerJ0u8eqLEkbEazsHmwTgP39jDRzek2TQQB2CiFHJmus6JiRrHxmucnKyz92yZyXKIdZCKGdJxrz2Faj10M6Mwm5eeiUYaxn1DPDCkYh6ZuEc27tGf8tnQFyOX9OlJeGQTPj1Jj3V9cYZzAem4OgaKiIhId7Iu6kCWiXsf/GARERFZdhZf+BnE8NI5wulRYr7hF2/vJRF43LwqycpcQPxDThcmIiIiItfGpeqzK71tIYRTo5hEBpNIL9hzOueoNhx7z5Z5+sA0bzRnJBlI+9y6OsUXb+nlttUpVuWC90xrO5AO2DKYgM3RCNDQQrlhmSqHnMs3ODlZo1C1UQga80jFPVKBRzJmSAYeiZgh4XsEPsQ8QzzwiHngmcufG1M1soiIiAhYG83CofBTRESkOy268BM/wEv3EubHMcBjN/cCOpAjIiIiIpfnnCOcHsVLZTHx5Id+LoCpiuWFowWe3DfFOyNVNg/MjO7cOpggl/zgkZXRXdH9gQ853yeX8FnXPP+miIiIiCy80EGlYUkr/BQREelKiy/8NB5+bgBbmABnMZ7f6RaJiIiIyBIQTo/iZ/ow/tWVuM5F086enq7z7Nt5nn57mslyyD0b0vzWxwa5fU2KRKBZSEREREQWu3rzlAKZuKe6TUREpAstvvAT8HMD1M8fBxuCwk8RERER+SCtkZ+5AfDm18PfOUfdwtvnK3zzrSl+cqRAwvf47I4cj+7sYeNAXOeZFxEREVlC6qGjYR3pmEZ+ioiIdKNFG366ch7XqGOCeKebIyIiIiKLnQ2x+QniG2+hNc3sB3HOUWk4Xjxa5Bv7pth3tsyG/ji//fFB7t+aYzDjAwZlniIiIiJLSy201EOnaW9FRES61KILP40xeLlBbLWIq1Ugmel0k0RERERkkXONGrY0hd87dOX/B/jLV8f5610T3L0+zf/6yGo+si5NKmY0ylNERERkCas1HA2Lwk8REZEutejCT2iO/KxVsLUymvRWRERERD6Iq1Ww5QJ+z5WFn8453hmp8vdvTvLP7+7nNz46SOBpalsRERGR5aDScAAkA9V2IiIi3WhRhp9euhfnHLY03emmiIiIiMgSYCsFXKOGnxu8ogCzFjr+9OUxVuUCvnx7PzFfB8ZEREREloty3eIZSAQa+SkiItKNFmUF4GV6ALDFqQ63RERERESWAlvK48IGfm7gAx/rnONHhwq8dqLEb987RG9yUZbEIiIiInKVyjWLMRr5KSIi0q0W5ZEeL5nFBDFsYbLTTRERERGRJSAsjGO8AC+d+8DHjpdC/vSVMT65JcMnNmU01a2IiIjIMlOqWzxjSMRU54mIiHSjRRl+mngCL5EhzI91uikiIiIissg55winxzCxBF4y+76Ptc7xt7snyFcs//Jj0Xk+RURERGR5KdUdMc8Q8xR+ioiIdKNFebjHxJJ4ySj8dM51ujkiIiIissjZ6TFMMoOJJy/7GOccBy9UeWLvFL98Vx+bB+Ia9SkiIiKyDJVqlkRgCBR+ioiIdKXFGX56Pl6mlzA/3ummiIiIiMgS0Jgawc/0YoLYZR9TCx3/6aVR1vbG+IVb+xR8ioiIiCxTxVpIIjD4Cj9FRES60qIMP/E8/Gx/FH5q5KeIiIiIfIBwehQv2w9ecMn7nXP84N0Cu06X+e2PD5JLLM4yWEREREQ+HOccxZolEXg6xYGIiEiXWqQlgMHLDWALk2DDTjdGRERERBYzG2IL4/g9Q3CZ0ZyjxZA/e2WMz2zN8rGNGY36FBEREVnGSjVLMqaRnyIiIt1qUYafxhj8nkFseRoX1jvdHBERERFZxFzYICxM4vcOXfJ+ax1/s2uCYs3yL+8Z1AgAERERkWXMAaWaIx1T0SciItKtFm0V4OcGseUCrl7rdFNEREREZBFztQq2XMDvGeLivv3OOfafr/DNt6b41Y/0s6E/plGfIiIiIsuYc1CqW9Jx73KTgoiIiMgyt2jDTy/bj6tXcdVip5siIiIiIouYq5Zw1dIlp72tNhz/6eUxNvbHeeyWXgWfIiIiIstcNPLTkol77+kYJyIiIt1h0YaffrYfF9ax5UKnmyIiIiIii5gt53FhAz/XP+d25xzfeyfPnjNlfufeQbLxRVv6ioiIiMgCcQ7Kdatpb0VERLrYoq0CvEwPOLClfKebIiIiIiKLWFiYwBiDl+5t3+acY6TY4Ks/HefB7Tk+sj6tUZ8iIiIiXSC0jkrDNae9Vf0nIiLSjRZv+JnIYGJxwsJ4p5siIiIiIouYLUyAH+ClczO3Ofjr1yeoNiy/cc8Avo57iYiIiHSFhnXUGtG0tyIiItKdFm0VYJIZTCxBWJjAOdfp5oiIiIjIIhVOjeClsph4CohGfe4/X+Fbb03xa3cPsLY3pl7/IiIiIl2iYR3V0Cn8FBER6WKLtgowQRwvmcXmxzrdFBERERFZpJxzNKZH8VI5vFiieRt8bfcEGwfifOGmHgWfIiIiIl2kHkKtOe2tiIiIdKdFWwUYP8DL9BJOK/wUERERkctx2OkxvGw/eAEAlYbj3dEqH9uQ0UEvERERkS5TCy2hU/gpIiLSzRZvFeD7+Jlewvw4oGlvRUREROQSnCOcHsXPDYIXlbaT5ZCxYsi2oYRGfYqIiIh0mUrd4RykgsV72FNERESurUVcBRi83AC2OAnWdroxIiIiIrIYhQ3CwiR+z2D7ptNTNRqhY0N/vIMNExEREZFOqDQszkEypk5wIiIi3WoRh5/g5wYJC1O4sNHppoiIiIjIIuTqNWxpGr93BcYYnHMcGauRS3qsyAadbp6IiIiIXGfVhsMBqdiiPuwpIiIi19CirgL83ACuUsDVq51uioiIiIgsQrZWxlYK+D1D7dveHa2ypidGRud5EhEREek65brFM5AINPJTRESkWy3aI0LGGPzcILZWxtXKnW6OiIiIiCxCtpyHsIGfGwCgbuHoWI3Ngwl0micRERGR7lOsWWK+IeYr/BQREelWi/qQkJftj6YyqxQ73RQRERERWYRsaRpnQ/xsPwCFasi56TrbhxIdbpmIiIiItDjnaDQaNBoNnHOXvD8MQ2q1OtbaSz7mSpWa4Wdc4aeIiEjXmveJkFrFyIULo4ShZeXKIeLxGMYsfEHhpXswxmCLUwv+3CIiIiKy9Nn8OMaPYVI5AC7kGxRqli2D8WtSn4qIiIjI/Djn2LfvAK++uhvn4N5772bnzhvatZpzjmPHTvCjH72ItZb+/j4efvhnSKdTV/VaGvkpIiIi8w4/G42QJ598imq1ShDEKJfL/NIv/Typ1PwLkg/ipXNgPIWfIiIiIvIezjnC6TFMPIWXzABwbKJGIjCs7ol1uHUiIiIiAjA9nedHP3qRr3zl57E25B/+4Z9Yv34tuVwWgHq9wTPPPMeDD36GdevW8O1vP8uuXW9y330fu6rObKWaJeEbAk/hp4iISLea97S34+Pj5PMFvvzlL/L44z8HwPHjpxa8YQBeIo2JJwgL49fk+UVERERk4TjnqFarVKu1y05nVq/XKZcrhOGHm86sJZwexUtlMfEEzjkOjVRZkQnoS/kf+rlFRERE5MM7d+4Cvb09rFgxyMqVK8hmM5w/P9K+3/MMDzzwKTZv3kAymaCnp4dyuXLVr1eqW1IxD2WfIiIi3WveIz8HBwf5yld+nlgsRqVSpVgsXdU0FFfCxBJ4ySzh9Ng1eX4RERERWRjOOXbtepPdu/fiHNx99x3cccctc6Yze+edw/zkJy/heR7pdJovfOEhstnMh3rdxtQIfrYP4wdYB4fHqmwciBMPdLRLREREZDEolUqkUkkAjDEkEnHK5XL7/iAI2LFjG845Tp48zYEDB/nlX/7SvEd9OucIraVQDUkGBucsYfjhO9uJiIjIwvM875qermje4WcQ+Ph+ilqtxtNPf49169awdu3qeb9wvV6nXm+8/4NCC8ks1QunKJXK7/9YERERue6MMSSTCZ1bUZiamuall17jV3/1ccLQ8rWvPcHWrZvo6YnOxVmv13nuuR/zyCOfY82aYZ566ru8/vqbfPrT91799uMcdnoMLzsAXkClZjk5Wednd/agLVJERERkcQiCgDC07X9ba/H9ubN0OOc4dOgozzzzHD/7s59j5cqheb+OtZZarU6+3CARQKPewKooFBERWXRanaGupXmHn845isUSTz75ND09OR5++AE8b96z5+L7AZ73QdORxUntvJfiT/+JoDKN37ti3q8jIiIi144yT2k5d+4CfX299PX14hzkclnOnx9ph5+e5/PII59l/fo1eJ5HMpmg0fiAjnAfxDYICxPEN+wEYLwcMl5qsG1IgbyIiIjIYjEw0M/U1BSNRohzjunpAgMDfe37nXPs33+QH/zgeb74xZ9l3brVV1XL+b5PMulRtYZswiedUk0oIiLSreYdftZqNf7+77/JmjXDfOYzn8DzPJxz8y4mPM/AFfTJz3385yg8//dU3vweuft/RUWLiIiIyCJUKpVJJhNAFIrH4zEqlZlzNQWBz6ZNG3DOceTIcd555zC/8iu/eFXTmVkbjRywtSphYQKTHcCGlhPjVayDNTmfMAwX7s2JiIjIVbnW05nJ0hCd5zPL009/F2sdAwN9DA0Ncvr0WU6fPsvOnTfwjW88xbp1azhw4CAHDhxk06YNbN++Zd7bj3VQrjuGMvMfqCEiIiLLx7zDz6NHT3DkyDGKxRJHjhzHGMNDD93P5s0br0X7CFZsIH3b/RRefILsxx/DpHLX5HVERERE5OrFYsGcwDEMLb4/t9S01nHgQNSr/7HHHmFwcGDer9OazgzAFvLYShEy/dTqdd45X6Y34dEbd+3HiIiISGdcj+nMZGkIAp8vfelRDh8+hjGGrVs34fs+6XSKoaEBfN/jkUc+2+7gBpBOp67qtax1VOqWTFzhp4iISDebd/i5ZctG/u2//ddzbrvaguSKGI/sfb/IhV3PUt7/Ium7HlKvQREREZFFZnBwgMnJKer1RnM6szyDg/0459qPefPNfbz00qs8/vhjrFp1dacz8H2fVCo6dUI934B6heTgMMlkguNTIev64vT3pAg81YsiIiIii4ExhmQyyc033zjn9v7+Pvr7o+lv77rrtgV5rdBBpaHwU0REpNvNO/yMx+PE49ev554xhsSGnSS2fYT8T75O+rb7IZa4bq8vIiIiIh9s5cohBgYGePLJp7DWsnr1SoaGBjh58gwnTpzk1ltv4p/+6RnWrBnmlVdeAwzbtm3mppt2XHXHNluaxtkQP9tPLXQcG69x65oUvnJPERERka4Ujfx0ZBIKP0VERLrZvMPPjvBj5D71ZUa++u+pHt1DYvvdGv0pIiIisoj4vs8Xv/h5Tp48gzGwfv1afN+nr68Hz9tAPB7n8ccfmzOd2dVMeztbWBjHeAFeuoepquVcvs6Xhvo+7FsRERERkSWqbqEWOtIxhZ8iIiLdbEmEn8YYkjvuITa8hfzzXyex9S7w/U43S0RERESajDHE43G2bt005/aenhw9PdE523fuvGFBX9PmxzFBDC+V43y+Trnu2DwQVyc5ERERkS5Va1hC60jHPdWEIiIiXWzJdIMy8RS5+x6n/NYL1M8d6XRzRERERKTDwqlRvFQOE0twdKxGMmYY7ol1ulkiIiIi0iG10NGwGvkpIiLS7ZZMJWCMIX37A3jZPgovfQPn7Af/JxERERFZlpxzNKZH8NI9mFiCQ6NVVmVj9CSXTHkrIiIiIgus2nA0LKTjqglFRES62ZKqBLxsP9mPPUbx9e8QTlzodHNEREREpFOcw06P4eX6CfE4PFZj00CcuK/pzURERES6VbXhcA6SwZI65CkiIiILbElVAsYYMvc8Cjak+PrTOOc63SQRERER6QRnCfNj+D1DlBuOU5M1tg3FO90qEREREemgct3ie5AI1CFORESkmy2p8BMgGFhD+o7PUnjpG9jSdKebIyIiIiId4MI6YWEKv2eIsZJlshyydSiBMTrQJSIiItKtSnWLZ4zCTxERkS635MJP43lk7/sS4eQI5b0/0uhPERERkS7kalVsOY/fM8TJyRoGWNenkZ8iIiIi3axUa438XHKHPEVERGQBLclKIL72BpI77iH//N/j6tVON0dERERErjNXK+OqJfyeIQ6PVulL+wym/U43S0REREQ6qFy3BJ7ReeBFRES63JIMP/F8cp/6CrXTB6keel2jP0VERES6jC1NgQ0x2X4OjVZZ3xcnFVuapa2IiIiILIxizZIIDL7KQhERka62JEsBYwyJbXcSX3cj+Z98HWzY6SaJiIiIyHUUFqdwzhEmezk2XmPrYBxPHfxFREREulqxZkkGHr4KQxERka62JMNPABNLkvvUl6kcfIXa6Xc0+lNERESki9j8OCaWIG9SXCg02DaUwBgd5BIRERHpZsWaJRnTyE8REZFut2RLAWMMqZs/hd+3isIL/wgKP0VERES6gnOOcGoUL5HmfDVGteHYNJDodLNEREREpIOcc5RqlkTg4atTnIiISFdbsuEngJfuIXvvz1N643s0xs90ujkiIiIicp2E06N46RzHpg3puMeqXNDpJomIiIhIBzmgVLOkYwZlnyIiIt1tSYefxhgyd38e/IDiT7+tqW9FREREukQ4PYqX7efQeIPhXEAusaTLWhERERH5kJyDUt2SiasuFBER6XZLvhrw+1aS+cjDFF75JrYw0enmiIiIiMi15izh9Bgu08+RiQabBuLEfHXvFxEREelmDijXLem4hypDERGR7rbkw0+DIXvvL2CLU5T2/ECjP0VERESWu7BBWJygmhzk9FSD7UM636eIiIhIt3MOSjVHWiM/RUREut7SrwaMITa8mdRNnyT/469h82MKQEVERESWMdeoY4tTTPh9TFdDtg4lMDqxk4iIiEhXC62j2tC0tyIiIrIcwk8A49Hz4D8nLEww8qf/jsbICQWgIiIiIsuUrZZw1TKnwh58A2t7Y51ukoiIiIh0WN06amE08lMd40RERLrbsgg/jTHEN9zMyt/5vwhL01z4//57qkf3KAAVERERWYZspYCtVzlWyzKQDuhPB51ukoiIiIh0WCN01BpOIz9FRERkeYSfMCsA/d3/B793BSN/8vuU3nwOZ22nmyYiIiIiC8gWp3DWcqSSYn1/nGSgnv0iIiIi3a418jMT9zvdFBEREemwZRN+QhSABoNrWfFb/zvJGz/O2H/5D+R/9Le4Rr3TTRMRERGRBWLz49RMjFPVNNsG43jKPkVERES6Xq3haFhHKqbiUEREpNstuznCjDF4mT4G/9l/YKp/FRPf/EMaE2fpe/T3MPGU5vwXERERWeLC/DjTNslYmGTbioTqOxERERGh0nA4IBVbVmM9RERE5Cosu/ATogDUJFL0feG/we8fZvLJPyScOMfAl/8nvNyADpCJiIiILGHh1AjngxWEsSQb++Odbo6IiIiILAKVenTqq2Sg8FNERKTbLetqwPgxcp/8MkP/4n+jcuh1Rv7039IYOYFzrtNNExEREZGr4JwjnB7lZGwNuVSMldll2ZdPREREROap0oiO9yU17a2IiEjXW9bhJ4DxPFK3PcDK3/m/CQuTXPij/5bq0TcVgIqIiLwP51z7T2RRcY7G9BjHExtZ3RMnm/A73SIREZGuojpRFqty3WLQyE8REZFOWUx1YldUA8YY4ptuZeXv/iF+zxAjf/L7TH/3q9QvHMeFjUWxIkRERBaTAwfe4eWXX+t0M0Tey4bUClOciq9n82AcHdsSERG5vlQnymJVqlliviHma+SniIhIJyymOrFr5gkzxhCsWM+Kf/V/MPlP/y/T3/8vTD37n0luv5vMRx8lecPdeJm+9mNFRESWios78RhjLtmx51K3t77zLr59amqa0dGxOffp+1EWAxfWKZTrjMRX8thQotPNERERWdRUJ0o3KdUscd8Q08QgIiIiH2i514ldE35CtJD93AADv/Q/0/vwv6L81vMUX/8OY3/1H/Gy/aRve4DMRx4mtvYGTCyh4l1ERBY95xxnzpzjpZdepVars2PHNu644xYOHHiHs2fPMzExhecZPvWpe1m5coizZ8/z4ouvUq1W2bp1E3fffQcAr732BocPHyOdTvGZz9wHwIULozzxxLepVKrcd989bNiwTt+N0nGuVma0EaeU6WHLoOo1ERGRy1GdKN2mWLPEA0PgaVsUERF5P91QJ3blRGHG8wkGVpP95OOs+jd/xPDvf5XM3Y9Q3v8C5//wdzj/h79D/kd/Q330NM6GmhZXREQWrVKpzDe/+TS33rqTz3720+zZs5933z3C+Pgk+/Yd4FOf+jhbtmzkySefZno6zz/+4z+xY8c2Hnrofg4ceJc339zHW2+9zf79B3nooftZtWolP/rRCzjnmJyc5t57P8rOndt59tkfYK3t9NsVwVZKnLL9xJNJ1vTGOt0cERGRRUt1onQT51xz5KenaW9FREQ+QDfUiV018vNixhgI4sTW7aBv3Q56P/cbVI+8SfG1p5j6zp8y9fSfEN94C8ltd5HYfDux4U146V7wfPVoFBGRRWF0dIzz50d49dU3MMaQz+c5deoM8Xic22+/heHhlQwM9PP8869w/PhJgiDgpptuwPM87rrrNvbvP0g8HuP2229haGiQgYF+yuUKe/fuZ9u2zQwPrySZTPDii69ircX3NYeUdJYtTXHcX81QNk5fStujiIjI5ahOlG5TrFuSMYOnY3YiIiLvqxvqxK4OP1taQaZJ5UjedB/JnZ8gnLpAef8LlPf+mPyP/47Jp/4YL5UlvnY7iS13kthyO7HVW/Gz/eD5c55HRETkevE8j4GBfh599HMEQUCpVCaZTPDmm28RhiEA1kaX8Xgcay3WWjzPo1arEYsFBEFAvV4HoFqtsn//weZj5n6vaSIEWQwaxWlO+KvZ2B8jEaj2EhERuRzVidJtSjVLOuahw3MiIiLvrxvqRIWfFzHGgDEE/cNkP/Elsvf+PLZcoHHhONWje6gc2kXh5W8w9eyf4SUyxFZvIbHlDpJb78TvW4VJpPASaUwQBz/A+DHwDGBmnl9ERGSBrFw5RDqd4pVXXmft2tW8+OKrPPjgp4Fo3v2hoUFOnTrD2rWr2bRpPel0iuee+wnDw6v46U938dBD9+P7Ad/5zvdJp1McOXKceDxOX19Ph9+ZyKWVCgXOxIZ5dCiBqioREZHLU50o3cQRhZ+pmNed5/gSERGZh26oE80HnM/ymmSyzjlKpQrpdHJJhYHOOXAWWynSuHCC6rG9VA+9TvXkAcLpUQwGfB/jxzCJNF4yi5fK4KV78TJ9+Jno0kvn8BIZTCKFiacw8SRePImJJTGxmdDUBNEfng/GW1LL6ko456hUqu1eAtK96vUGYRiSSMSX3XYuV85aS6VSJZVaWt8Nneaco1gssW/fAYrFEps3b2TTpvU8//wrnD9/geHhVaRSSW65ZSeJRJxSqcy+fQcolUps3bqF9evXAHDy5GkOHTpKX18vt9xyIxcujFIul9m2bQvVapV9+97mzjtvvdbTVGjFLy3Xve+ec44zb+/j934Av/+5TTywPXe9myAd1mg0aDRUM3Qr5xzlcoVkMoHn6dB2N6rV6jjniMdj2gdcAdWJ0kEdqRO/+cYIRyYt/8P9q7SP6EKqE7ub6kRRnTg/3VAnzjv8dM5hrWNsbBznHENDA3je/II5ay3j45MMDPQt6Z1RFIY6XKVIY+IctjBBWJrGlqaxxUnCwiS2OIktTWGL09hyHlst4eoVXL0OYX3WAo6uGc/HBPFm8JnApLL42X783ABebhC/dwg/N4CfHcDL9uGlsnjJLCaWwPgB+AGYmWW6UB/0S28nbmbMspm7Hq/kdZ1zTE1Nk0qlSCTiC9BKWarK5Qq1Wp2enqy+nLpYGIZMTk7T39+7pL8bFosf//glSqUSDz/8M0vpc7VkGirAVRzUik58P0W5XGFwcGDeP0qcc1wYneR/fGac//j5dWwdSsy3CbLEVSpVqtUqPT25pQy7UIYAABaLSURBVLRvkwXinGN8fJLe3h6CQOcW7EbFYglrHdlsWvuAD0F1olwH171OtNZxdmSC509ZvnLX4FLatmWBqE7sbqoTRXXiwlhOdeK8h9s55/jBD37CqVNnMMYwNDTA5z//YFee2L41Ra5J54inLz3yoBWQ4izYENeo4Rp1XFiPLutVXK0S/dUr2GoZVy1hq0VsuYDNjxPmx2hMnseeficKVqvlKDgNQ0wQRCFpMo2XyuGle6IRptl+/Gx/87IvGnGayuElUjjbbEvYiC5tA8IQZ0OwNmq3bWCLU7haOWpTrTLrerONtTKuXgVHFMKme/AyvXjpXvx0D16mJ7otmcEkMnipbDQNsCEKS42HazRw9SrWWIznEZ2YwbQWHjjbbld0Oet62MCF9WYA62YC2tYyZ/YlEMTxkulmuBz94XnNtlx6WmI3+3laYXfYaL52I7qv+f+j7aH5XM3rpn29+b7al7Tf5xLaiYjIErJhw9r2vPvXinMObAMw4Pnan8n7cs7x1ltv8+KLP6Wnp4darcaXv/wYqVRq3s/VnwoYymjWCBERkatxPepEkflYiDoxdI5Kw5GJqyOtiIjI1VpOdeK8jxpNTk7z9tvv8pu/+Sv4vs9f/MXfMDIyxvDwymvRviWvHX7hRdPZxq58hMLMaMtm6NaoQ6OOrZWiYLSUxxbGCfMThIXxKChtjjZtjJ2JgtJaM7hs1JvZm9fsf+ea3fDcrP54DjBRCOn5GM8HPxYFrH4sGl0aS2LiCUxzml4vmYn+Z61KY/QU9uSBaHRrrRqNcG3UZ5aDH0TT+GLA8zBegPfZ36F0+FXM2YPN6X4D8GPRczZq0KjPCRtdWJ8JHp2dvbDa7+C976m1LprvK57ES6Qw8fRMaJvONacn7sWLJbHVErZSxFUK0fupFGduq1fbf9ETe83g1mu+L3/mevu2IFpurWUYay3D6NKLJaN/N6dBjh6XwIsl2tdNLA5BHNM6lywGXDgTZrtZoXZrGYWNKCi+KBB2FwfErevRgqId1Davm4sCXNNcjiaewsTi7dHKBLFmwD1rmUC0rtqva6MAO2yAbeDCENuw2NBRL49EQXxYb677eruzQPvfYWPOddeot4Nw195eout4fnNdN8/F27reWv7t6aYTGN8HL4jWX3v799v/bgXaM5/JWdvX7HDc2ZnPku+D8ZvLbFbQbhbox1hrec5+7da6NCba/prb4JyOBVf25GBbz29xzjWDrYsDfeb13JcLxy4eXX7x/m9OR4bWdtxaBhevj9nb36zODdfH3DZyiTY611o/Fy3Pea+jS732zOtuXLe6+d1Rm/uZb5vbEQNjZr36rDZ53qXb5RylN79P/gd/jUmk6X34t0hsvVMBqFxWo9Hg+edf4ed+7mGGh1fx5JNPceDAu9x5561Xvt04h7Mha3pjOrAlIiJylTZt2tDpJojMsRB1YmihWrdk4rFr3FoREZHlaznVifMOP8fGxsnlsu1zsg0M9HP+/AVWrVox7wOe1louOt4tl2XAj4MfxyQy+LkVXHqs7dxwydWjUaW2WsKWpnHVchTstIKe5lS5UdDTCn48jPHb5y9th5KtMG/WyMaZl22N0rRR8FSvRSNZa5UoOCznZ4LDRr0ZbNaordyIZxxmcLg9KpZWYBqLQzA7BIzPCgIT7SDwioMN56IRq6VpwuJUNB1xc5rixoVT2MrbuGoJ16g3Q8kEJpaKRosmM3j9awiSaUwyG4W+nv+e52+FeRdfunBmlK+tV3CFKVz9/JzlMRPy1aLlaGaFjqYZPnheFH560ftuBZ5Y2wy+7GWCl1nbx0VNvtTtM6NSL16IZuaiGSy1w9ggFl0P4u1g1IsncNa113f0HmvNgLI12riB2XgHrLmR6R/+ObhWWDv7fcwObJkJmb2ZcDLaRv0o9Jod1DgbbY+NGq5RjdZLOxicNUq3GfrPPOfs1/Can5fmyGXnonXU/KzNad+s8JPWNNatbTeYuw1HwXfivdvSZbgwjN5DvQrt91SbtQ3V2mGwCWIznRVmBe7t4L8ZBAMz+4nmCPRoJHq1udyal2G9uTyCaP/R3A5nXyeYfb7i5ujq9jTe8VlTesej7cUYbL3a7ixhraO++mYmXv46rpKfaUejNhPk2/ASo8FbnQAa0bqc1blgpmNBMjoXczwVheCx5MyI/NYo+NZrhM0R8a37ZgXM7wmZ59zGrPbMblc4t63ONj87M51L5nQ0mb3c/KD5+Z71OZ/T4WHW9Va7W50fZneGmHUd52YC8YuC4ouvt5ddPNUcxZ/GS6Txkhlc2GDqu1/F5sehXiXMj7Hq3/wRfrb/irZn6T6lUplqtcbgYD+eZ1i3bg0nT57mzjtvveLnaIQhpWKZ7f0xDI4wVCHZbax17dNwvLd+keWu1TnKOUsYqrNNN3JO+4ClJPq5ZdQ5Tj7QQtSJ1jmqoSOXMNpHdCnVid1NdaKoTlxarkedOO/ws1arEYsF7UYFQUCtNv9hsGFoGRubmPf/k6sRHUinp+/KHt7aP1igboFa8+8qXtePQ6YHMqsv+QhrHXZgw9yQrbVvuh7fUw5wDg+HNyswmB2mtQIyh+FDfXe65nhbB9GzMRNUzw5zWuHfEueA8BK3X3JsWzP88bbefYXP3gyC50w1PGv64Tnhp5sbms0eMfwhXPwePmjTuF5fuxd/lBbyNd1Fl1fnoiWVnPWMxoNEhuqO++f9Khcv/9m7scXqepdiV7L7unxXiUsL/sX/iZsaofHX/55wepSwMKHwUy4rDMPmYOJo/x2LXUUN6RyB57h72DA6OnZtGiqLWtTnxFGtXk1tKsuBtY6xscnrO7GDLBqu2UGyXC53uilyBTzPZ3CwT+GnfKCFqBMTgceqbIANq4yOqk7oRqoTRXVid1OduLRcjzpx3uFnMpmgXm+0p0Cs1+skk1c+lWtLVNR033lCZS7nQjzPw/MWwbdSTFOjdEqrd56fznS6KdJBzkU/eoO+oU43Ra5QY+wMYWEcgjjxdTcS9K3qdJNkEQuCqOwMw2gq7Vpt/jVkEIuRyfVgbagDqV0qqhksvq/fEd3KuRDf93VQq0tFs0cZfF9Tny8F3uVOnyBykYWoEz0TBaDW81UndinViaI6sbupTlxarkedOO/wc3BwkHy+QD5fIAgCRkfHWbVq5bwLC8/zGBhQD8BuNzU1TTKZJJGId7op0kGVSpVarU5PT7bTTZEOstYyMTHNwECvvhuWiHp9guK7pzAP/CrZj/4sJpHudJNkEUunU6RSSc6du8D69Ws4evQEO3feMO/nMQZ6e3uIxeZdxsoyUK3WqFSq9PbmOt0U6QDnHBMTU/T25nRgs0uVSmWsdWSzqjlElhPVibIQVCd2N9WJojpRLjbvaqC3N8edd97K3/3dkxhjuPHG7QwNDVx1A3SAu3u15mLXOUBkNm0L3a21+rUdLA2x4S30Pfp77X9rvcn78X2fz3zmPp599jmSySSpVIobb9x2VdtN69wQ0p30XSGg3w+ifYDIcqI6URaK6kRRnSigfYBE5h1+GmP4xCfu4bbbbsI5yGYzzSGqIiIi0k1UTMp8GGPYvn0L69atoVqtkctlNR2NiIiIiKhOFBERkQV3VfNAGGPI5T7sFAI6YCoiIiLSTYwxpNMp0ulUp5siIiIiIouI6kQRERFZSKY19ehlvO+dV6t18vJ4PKZRI13MOUe93sD3ffXo63JhGGKtJQgC7RO6mL4bBPWMWmquSZ34vi/YrB2CwNfMI11KNUN3i/YBdWIx1QrdqtEIAYfv+9oGuo9W+NKiOlGuO9WJ3U11oqhO7GqXXOEdCT9FRERELkHV6dKiOlFERESuF9WJS4vqRBEREbleLlknqiuUiIiIiIiIiIiIiIiIiCwLCj9FREREREREREREREREZFlQ+CkiIiIiIiIiIiIiIiIiy4LCTxERERERERERERERERFZFhR+ioiIiIiIiIiIiIiIiMiyEFzvF7TWcvjwMU6fPsvGjevYuHE9nqcMtluMj0/yxht7cc5hDNx8843E43HOnj3Pzp03YIzpdBPlGqvXG7z77mF27NiG7/uEYcihQ0c5d+48mzZtYMOGdZRKZQ4dOsKtt96k/cMy5Zzj8OFjrFq1glwui7WWXbv2MDU1DUBfXy+3334L+/YdYMeOraRSqQ63WEQ6zTnHuXMXePfdwwwM9HPjjdsJguteysp11mg0ePXVNyiVSgCsXDnEzp03sHfv/nYdKcuTc46TJ0+TyaQZHBzAOcfExCT79x8klUpx8803kkjE2b//IOvWraG3t6fTTZYF5JyjWq1x9OhxbrxxO8YYJiam2L17T/u35E037SCZTHLmzDn9lhTpcqoTu5PqxO6lOrG7qU6UK3VdUwXnHPv3H+THP36R3t4enn32Bxw+fAzn3PVshnSIc4533jnMiROnyGQyZDIZgiBgdHSM3bv3drp5co1FX0xVvvvdH/LUU98jDEOcc+zZ8xYvvfQqPT05nnrqexw/fopCocjLL7+mfcMyFYaWt956m7/6q6+3w85yucLzz79MIhEnk8mQSiVxzvLSS69SLJY63GIRWQxGRsZ44olvk8lk2LNnPy+++FN9T3SBfL7Ayy+/SiqVJJPJkEgkaDRCXnjhp1Qq1U43T64Ray1HjhznL//ya5w/PwJAsVji619/Et/3OXXqDN/5zvex1vLaa28wPj7Z4RbLQnLOUS6X+da3vsMPf/gC1jqccxw6dJjjx0/O+S05NjbO66+/2ekmi0iHqU7sTqoTu5PqxO6mOlHm47qHn6+//iaf/vQnuOOOW/jkJz+uDbDLnDlzjhtu2MaaNau4+eYbGRjob9/nnOPIkWMcOaJAfDlyzvH0099ncnKSWCzqgdka7Xf//fdxxx238rGPfYTdu/cAM+v/1KkzHDjwDtbaDrVcFtrevft55ZVd5HLZ9m2Tk1NkMmk2blzP5s0b2LFj+5xeWfl8gV273tQPGJEu5ZxrjgTfxl133cZDDz3Avn1vU6/XO900ucZGR8cZHBxg/fq1bN26iW3bNjO70+7k5BS7d+/VtrDMHD16nO9//8dzeukfOXKM/v4+Pv7xu3noofs5deoM+XyhfX+lUmHXrjcpFIqdaLIsoEYj5BvfeJp6vY7v++3bT58+xw03bGXNmlXcdNMOBgcH2vdFvyWPq3O1SBdSndi9VCd2J9WJ3U11oszHdQ0/wzAkny/Q19eLMYb+/l4mJ6cUanQJay0jI6McPPgur732Bn/+53/D2bPngdao4Hd47rnnNRXBMmWM4YEHPsnnP/9ge/qZRqNBsViit7cHYwwDA/2MjY1jbfRFdPz4Sb797e/S05PT9ATLyLZtm/nVX32cvr7os+6cY3R0nLGxCXbv3sM3vvEUzz334/Z2UCgUeeKJb9NohMTjsU42XUQ6aHR0jKGhQYwxZDJp6vW6OkQsc845LlwY5fz5EXbv3svXv/4kL774avsH6+TkFP/wD9/C9z1NbbfMrF49zK/92ldYtWpF+7aRkbH2QYx4PE4sFrQPYFWrVb75zWeYnJwmmUx2pM2ycHzf49FHP8dnPnMfnhf9Boh+S45x8OAhXnvtDf7iL/6WM2fOAdG+4u233+W5535Mb2+uk00XkQ5Rndh9VCd2L9WJ3U11oszHdd37OxdtcK0N0xiDcw4F7t3B8zwef/wxcrkssVjACy/8lBdf/Cm33XYzJ0+e5vjxk/zGb/wzBgb6FXQtQ8YYent7mJycat/W2ie01rfnmXbgNTo6wd/+7RN86UtfYM2aYW0Ty0g2m3lPp5ft27ewadNv0tOTY3o6z5/92V9x55230mg0+Id/+Bbbtm3h7rvv0DlgRbqYtba9D2h9JajX5vJ3++03c9ttN5HNZhgZGeO//tevc9NNO6hUqnzta09w1123c+utN6lOWGbS6dR7Pt/WWoIgmLWuDdZarLU89dT3WLFikE9/+l58X7XCUud5Hr29PRSLxTm3felLX2j/lnzppVd54YVXuOuu2zl16gwnT57m13/9lxkcHND+QKQLqU7sTqoTu5PqxO6mOlHm47p+4n3fI5lMUC5XACiVKqRSyXYYKstbGIZMTU2113d/f297WwiCgO3bt/LGG/tUoHaRIPCJx+NUKlWccxSLJbLZNMYYjIFbbtnJ7t17CMOw002Va2x6Ot9ez6lUkiAIqNfrOOe48cbtnDhxiqmpae0fRLpYLpelWCzinKNWq2OMp9HgXWBqahprLcYYstk0AGHYwDnLzTfv5ODBQxSLJX0/dIGenhyFQrQPCENLo9EglUrhHGzevIGJiUkuXBjtdDPlGnnvb8m+5m9JRxAE7Nixjd2792pfINKlVCd2J9WJ0qI6sbupTpTLua7hp+d5bN26mV279jA1Nc3u3Xu44YZtSty7hLWWZ575Afv2HWBycppdu95kx45tGANr1gzzyCMPcvTocY4cOa6dUZfwfZ/Nmzeye/depqbyvPHG3uY2YRgc7Odzn/sMpVKZPXv2a5tY5o4dO8G3vvUsU1PTvPXW2ySTSXp7e4nFYtxzz13ccstOvvvdH2qadJEutn37Vt56620mJqbYs+ctVq9eSSKR6HSz5Bo7cOAdnnnmB806YR+DgwOk02lSqRT33XcPGzas5Yc/fEF1QhfYtGkDJ06c4sKFUQ4ePEQymaC3N4fve9xxxy3cd989fOc739d5vZYpax3PPvtD9uzZz9TUNK+/Hv2WBMPq1av4/Ocf5MSJUxw6dFT7A5EupDqxO6lOlBbVid1NdaJcjv8Hf/AH73f/+945X8YYhodXceLESV5/fQ9DQwN84hMf1dzrXcLzPNauXc1rr73J/v0H2bx5Ix/96F1UqzXK5TI33rid/v5eDhx4h23bNmt6y2Wq0Qi5cGGEHTu243kea9YMc/jwMXbv3sPatau55567aDRCJiYm2LlzB8PDK3nzzX1s27ZF+4plxDnHuXPn2bBhHdlshpUrh5icnOLVV3eTzxd45JEHyeWynDlzjq1bN7N583refvsQg4MD5HLZTjdfrp3/pdMNkHn5g+v1QsYY+vp6qVZrvPTSq9TrdT73uftJJhPqRLfMrVkzzNmz53n99Teo1+t8/vMPkkwmOXv2PDt2bGXjxvXs3buftWtXk0oltT0sMyMjo6xcOURfXy+ZTJpEIsELL7zC+Pg4Dz/8AD09Oc6du8C6dWvYvHkTp06dIZGI09/fp21hGahWa0xOTnHDDVvxfZ9169awa9ebvPXWQTZtWs8999xFrVajWCyxc+cN9Pf3sX//QbZv36LfksuT6sSl5Q+u1wupTuxeqhO7m+rE7qY6US5yyTrRfEDaveBReOv1Wuf5086mu8yc43Vm/Ue3zZz3UdvG8jZ7fbfWf+t2bRPd4+Lt4OLbZj9O20FX0cpdWq57l8mL9xPaH3QHfT90p/erFYD3rP//v707RgEYBIIAaP7/aFOENMFUgQu3zlSWwlmsLOK9HuP6PsF56G91b3CX3Jqh9iInUkJO3JOciJzIw3Ko5eUnAMALCbQXOREAqCIn9iInAgBVljnRG18AAAAAAAAggvITAAAAAAAAiKD8BAAAAAAAACIoPwEAAAAAAIAIyk8AAAAAAAAgwjHn/HsPAAAAAAAAAJ95+QkAAAAAAABEUH4CAAAAAAAAEZSfAAAAAAAAQATlJwAAAAAAABBB+QkAAAAAAABEUH4CAAAAAAAAEU6jycFSbRUv2gAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 2376x1944 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FqBEmJ_0yYvN"
      },
      "source": [
        "Как видно из графиков, модель сильно переобучается - loss на тесте растёт, а точность на тесте сильно меньше чем на трейне, но у авторов происходит тоже самое (это видно из секции \"How to run a TIMIT experiment\" -> \"3. Results.\" в их readme, loss на тесте растёт, точность на тесте сильно меньше чем на трейне), но авторы учат именно столько эпох, потому что метрика, точность классификации дорожки, продолжает расти.  \n",
        "Получающаяся у меня точность классификации спикеров - примерно 0.98, это не сильно отличается от заявленных авторами 0.9835, учитывая что я учил 345600 итераций градиентного спуска (12 часов), а не 1200000 итераций (24 часа), как это делали авторы."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ol3c2HUg89Sj"
      },
      "source": [
        "Обучим `SincNet` такое же число итераций градиентного спуска и сравним результаты"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "KHbqpEkTFIZ5",
        "outputId": "c63fe258-8ef3-4e29-83cc-e943959e207a"
      },
      "source": [
        "!python main.py # sinc run"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "id 33easts1\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mbch\u001b[0m (use `wandb login --relogin` to force relogin)\n",
            "2021-06-10 11:21:52.150429: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.10.31\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mgood-shape-19\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet/runs/33easts1\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in /content/sincnet/wandb/run-20210610_112146-33easts1\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run `wandb offline` to turn off syncing.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HrApceF_75rV"
      },
      "source": [
        "В этом запуске получаются 3 аналогичных графика, но получаются лучшие метрики на тестовом множестве - точность на фреймах выше на 8%, точность по всем дорожкам получается  примерно 0.9848 - 0.9928, то есть очень близко, к 0.9915, заявленным авторами в таблице 1."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uBx_XuGyFGDn"
      },
      "source": [
        "Промежуточный вывод о получающихся метриках - подход авторов работает лучше, этот результат воспроизводится."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jicdy3qu9RmQ"
      },
      "source": [
        "#### Скорость сходимости"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qUUHYuRp9UOJ"
      },
      "source": [
        "Теперь проверим утверждения авторов о том, что предлагаемая ими модель сходится быстрее. Для этого запустим 2 аналогичных обучения модели, но на меньшем кол-ве итераций градиентного спуска, и для наглядности, будем выводить метрики чаще."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "VD3hnbKsAT6_",
        "outputId": "8e3c6629-46a4-46b1-d289-a75aae4fa9e4"
      },
      "source": [
        "!python main.py # baseline short run"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "id 27yg5by9\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mbch\u001b[0m (use `wandb login --relogin` to force relogin)\n",
            "2021-06-11 17:12:11.010184: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.10.32\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mfragrant-oath-25\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet/runs/27yg5by9\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in /content/sincnet/wandb/run-20210611_171209-27yg5by9\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run `wandb offline` to turn off syncing.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "K4pFRGZgD1dj",
        "outputId": "2efc3eab-cd67-4eba-fa49-c0836b3fe6fd"
      },
      "source": [
        "!python main.py # sinc short run"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "id 295ny68u\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mbch\u001b[0m (use `wandb login --relogin` to force relogin)\n",
            "2021-06-11 18:21:34.153028: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.10.32\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mspring-dragon-26\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/bch/SincNet/runs/295ny68u\u001b[0m\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in /content/sincnet/wandb/run-20210611_182132-295ny68u\n",
            "\u001b[34m\u001b[1mwandb\u001b[0m: Run `wandb offline` to turn off syncing.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "id": "Jt8whG-09omv",
        "outputId": "632f8382-e5dc-4ce8-cd6e-d34a163767a5"
      },
      "source": [
        "Image(filename='sincnet/report/converges.png', width=600, height=300)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "height": 300,
              "width": 600
            }
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7KDfB1nL-KFy"
      },
      "source": [
        "Как мы видим, получается график, очень похожий на график номер 4 из статьи (frame error rate заменён на frame accuracy).  \n",
        "То есть результаты авторов о сходимости также воспроизводятся."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ND9ndcS3FY_U"
      },
      "source": [
        "### Задача speaker verification"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RPuAcJXbFetY"
      },
      "source": [
        "Теперь займёмся задачей speaker verification, посмотрим воспроизводятся ли там результаты авторов."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iXRPTngmZWm9",
        "outputId": "f7ce12bf-598f-49ec-9528-d45bf2e041d1"
      },
      "source": [
        "%env PYTHONPATH=/env/python:/content/sincnet/src"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "env: PYTHONPATH=/env/python:/content/sincnet\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TKqjp__5F8BN"
      },
      "source": [
        "Для этого посчитаем `d-vector`'а на тестовом мн-ве timit'а (этих спикеров сетка не видела во время обучения), а трейн часть timit'а будем обрабатывать так: на трейн сплите посчитаем для каждого спикера средний `d-vector`, а тест часть будем ипользовать как валидацию. Итоговый pipeline - пробегаемся по тестовой части трейн сплита timit'а, считаем `d-vector` текущей дорожки, берём 10 рандомных импостеров из тестового мн-ва, затем в зависимости от способа (dnn или косинусное расстояние) либо 11 раз делаем предсказания и берём правильный класс в каждом предсказнии(в случае dnn), либо добавляем к 10 импостерам `d-vector` из трейн части трейн сплита правильного класса и считаем расстояния от текущего вектора до этих 11. Затем получив по 11 значений для каждой дорожки из тестовой части трейн сплита считаем EER."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0DPFTt0YI89P"
      },
      "source": [
        "Считаем `d-vector`'а baseline модели для трейн части трейн сплита"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2rGbQ3p66niO"
      },
      "source": [
        "pretrained_conv = \"model/cnn.pt\"\n",
        "compute_split = \"train.scp\"\n",
        "!python speaker_verification/compute_d_vectors.py cnn $pretrained_conv $compute_split --save_to d_vectors_conv.npy"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vX-NpyOeJCYX"
      },
      "source": [
        "Считаем `d-vector`'а baseline модели для тестового сплита"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "diIDDO0UIpZq"
      },
      "source": [
        "pretrained_conv = \"model/cnn.pt\"\n",
        "compute_split = \"sv.scp\"\n",
        "!python speaker_verification/compute_d_vectors.py cnn $pretrained_conv $compute_split --save_to d_vectors_conv_unseen.npy"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LLImUQpeJQMl"
      },
      "source": [
        "Считаем err с помощью косинусного расстояния"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x0v8K61F6nnX",
        "outputId": "37c9ca91-3883-473c-9a58-41b953ed3168"
      },
      "source": [
        "dv_speakers = \"/content/sincnet/src/d_vectors_conv.npy\"\n",
        "dv_imposters = \"/content/sincnet/src/d_vectors_conv_unseen.npy\"\n",
        "!python speaker_verification/speaker_verification.py cnn $pretrained_conv $dv_speakers $dv_imposters cos"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.03163780663768459\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IRo6Ri2QJVSb"
      },
      "source": [
        "Считаем err с помощью dnn"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wCwrSD48KU8U",
        "outputId": "483a05da-3dad-4f57-98cc-827af2a98d1c"
      },
      "source": [
        "!python speaker_verification/speaker_verification.py cnn $pretrained_conv $dv_speakers $dv_imposters softmax"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.33549783549771584\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0XfII1vyJa2e"
      },
      "source": [
        "Проделаем аналогичные 4 шага для `SincNet` модели"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sNc4CvsCL9Hp"
      },
      "source": [
        "pretrained_sinc = \"model/sinc.pt\"\n",
        "compute_split = \"train.scp\"\n",
        "!python speaker_verification/compute_d_vectors.py sinc $pretrained_sinc $compute_split --save_to d_vectors_sinc.npy"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pf598-myKgD7"
      },
      "source": [
        "pretrained_sinc = \"model/sinc.pt\"\n",
        "compute_split = \"sv.scp\"\n",
        "!python speaker_verification/compute_d_vectors.py sinc $pretrained_sinc $compute_split --save_to d_vectors_sinc_unseen.npy"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4xG1j8a_MNoi",
        "outputId": "4b6eaef8-3c50-4f96-db47-88e72c44a03b"
      },
      "source": [
        "dv_speakers = \"/content/sincnet/src/d_vectors_sinc.npy\"\n",
        "dv_imposters = \"/content/sincnet/src/d_vectors_sinc.npy\"\n",
        "!python speaker_verification/speaker_verification.py sinc $pretrained_sinc $dv_speakers $dv_imposters cos"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.01735209235197047\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JtQ8TzuNMNrP",
        "outputId": "618a5c3e-9124-4c82-8b0a-61bc59693c68"
      },
      "source": [
        "!python speaker_verification/speaker_verification.py sinc $pretrained_sinc $dv_speakers $dv_imposters softmax"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.27474747474736105\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tk8LiwnQOyRL"
      },
      "source": [
        "Проанализируем результаты и сравним их с результатами авторов из таблички 2:  \n",
        "\n",
        "|              | dnn   | cosine dist |\n",
        "|--------------|-------|-------------|\n",
        "| cnn baseline | 0.335 | 0.032       |\n",
        "| SincNet      | 0.275 | 0.017       |"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9LotFQz9O7H0"
      },
      "source": [
        "Результаты dnn получаются того же порядка, как и у авторов, но немного лучше, я думаю это связанно с тем, что датасет timit легче датасета librispeech, так как в нём 462 спикера, а не 2484. Резульаты же в случае косинусного расстояния получаются на несколько порядков лучше, я думаю это свзяанно с тем, что авторы сравнивают результаты, полученные на разных задачах - в случае dnn верифицируются только те спикер, на которых модель училась, а в случае косинусного расстояния только те, которых она не видела. И очевидно, что модель будет давать результаты лучше, работая с теми же спикерами на которых она училась, поэтому полученные мной результаты на несколько порядков лучше. Но на мой взгляд не правильно сравнивать результаты разных подхов на разных задачах, поэтому я оставлю свои результаты. Тем более, в результатах авторов какая-то ошибка, невозможно получить err болше 0.5, если во всех предсказаниях поменять расстояние на противоположный знак, результаты станут лучше (1 - результат который был), а лучший результат авторов в случае косинусного расстояния - 0.51.  "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ocAnMr4_TS0p"
      },
      "source": [
        "Но тот факт, что в обоих подходах, предлагаемая авторами `SincNet` даёт лучший результат чем baseline воспроизводится."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ExFMybK6bx0K"
      },
      "source": [
        "## Заключение"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "od0eijmdTsWK"
      },
      "source": [
        "Подытожим поставленные эксперименты, авторы статьи акцентируют внимание на трёх моментах, предлагаемый ими подход:\n",
        "*   показывает лучшие метрики качества чем baseline\n",
        "*   быстрее сходится\n",
        "*   показывает лучшие результаты на downstream задаче speaker verification\n",
        "\n",
        "По поставленным экспериментам можно сказать, что все утверждения выше воспроизводятся!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bFs74WzXXF4z"
      },
      "source": [
        "\"Какие есть перспективы для развития?\" - Некоторые задачи обработки звука, такие как speech enhancement и speaker diarization работают с wav входом, а не с mel спектрограммами, и в архитектурах, которые решают такие задачи, можно попробовать подход авторов и заменить обычные свёртки в первом слое на свёртки, предлагаемые авторами. Это может привести к лучшей сходимости и более высоким показателям метрик."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sgEbn1cJQ8O_"
      },
      "source": [
        "# Задача 2"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_tdC1mhwRNKz"
      },
      "source": [
        "Очень сложно остановиться на одной какой-то статье как этого просят в задание, поэтому я остановлюсь на нескольких идеях которые одновременно кажутся интересными мне и подходят под задачу, решаемую на стажировке:  \n",
        "*   Трансформеры в аудио: за последний год вышло очень много статей на тему ускорения работы трансформеров, а за последние полгода на тему трансформеров в CV, в то же время применение трансформеров в аудио не является таким популярным, хотя учитывая их результаты в nlp, и в cv такой недостаток внимания является неоправданным. Статье о применение трансформеров в аудио, которые я читал: [FastSpeech](https://arxiv.org/abs/1905.09263), [Transformers are RNNs](https://arxiv.org/abs/2006.16236), [Streaming Transformer ASR](https://arxiv.org/abs/2006.14941) и [Reformer-TTS](https://www.isca-speech.org/archive/Interspeech_2020/pdfs/2189.pdf), в которых решаются задачи ASR и TTS.\n",
        "*   Также в последнее время мне нравится изучать статьи о style transfer'е, которые применяют один и тот же подход в разных доменах, то есть и в картинках и в звуке, такие как [Style GAN](https://arxiv.org/abs/1812.04948) и [Adain-VC](https://arxiv.org/abs/1904.05742), [CycleGAN](https://arxiv.org/abs/1703.10593) и [CycleGAN-VC](http://www.kecl.ntt.co.jp/people/kaneko.takuhiro/projects/cyclegan-vc/), [StarGAN](https://arxiv.org/abs/1711.09020) и [StarGAN-VC](https://arxiv.org/abs/1806.02169?source=post_page---------------------------).  \n",
        "\n",
        "Буду рад обсудить любую из этих статей.\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wxqmCIcl68AV"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}