{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras.datasets import mnist #Загрузка датасета mnist:\n",
        "(x_train, y_train), (x_test, y_test) = mnist.load_data()"
      ],
      "metadata": {
        "id": "Bvx91h2wa1BE"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x_train = x_train / 255\n",
        "x_test = x_test / 255\n",
        "y_train = tf.keras.utils.to_categorical(y_train, num_classes=10) # Преобразование меток в бинарные векторы\n",
        "y_test = tf.keras.utils.to_categorical(y_test, num_classes=10)"
      ],
      "metadata": {
        "id": "F6MBoAJna-ZZ"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense\n",
        "model = Sequential()\n",
        "# Добавление слоев\n",
        "model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)))\n",
        "model.add(MaxPooling2D((2, 2)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(64, activation='relu'))\n",
        "model.add(Dense(10, activation='softmax'))\n",
        "model.summary()\n",
        "# Размеры тренировочного, валидационного и тестового датасетов\n",
        "train_size = x_train.shape[0]\n",
        "val_size = int(train_size * 0.1)\n",
        "test_size = x_test.shape[0]\n",
        "print(\"Размер тренировочного датасета:\", train_size)\n",
        "print(\"Размер валидационного датасета:\", val_size)\n",
        "print(\"Размер тестового датасета:\", test_size)\n",
        "tf.keras.utils.plot_model(model, show_shapes= True, show_layer_names= True, show_layer_activations= True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "0U47i88Ya_zi",
        "outputId": "726ec8eb-0bfc-43de-eedd-a102eb9dde41"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_7\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_7 (Conv2D)           (None, 26, 26, 32)        320       \n",
            "                                                                 \n",
            " max_pooling2d_7 (MaxPooling  (None, 13, 13, 32)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten_7 (Flatten)         (None, 5408)              0         \n",
            "                                                                 \n",
            " dense_18 (Dense)            (None, 64)                346176    \n",
            "                                                                 \n",
            " dense_19 (Dense)            (None, 10)                650       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 347,146\n",
            "Trainable params: 347,146\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Размер тренировочного датасета: 60000\n",
            "Размер валидационного датасета: 6000\n",
            "Размер тестового датасета: 10000\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAJzCAYAAAAcFSQbAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nOzdeVhTZ9o/8G8CISEhYREFZCeIilpt1algHaqdalsvrdSNqTqj1il2QxSXopWfIlUpjvCqaOtS+1ZbBZVBhxHtpX2p47i0fQsD4iuidQEpRRAMCMh2//5wkhpZPIFAgt6f6zr/PGe5n5PnPLmTc55zjoiICIwxxthjiE1dAcYYY90DJwzGGGOCcMJgjDEmCCcMxhhjglgasvDGjRtx9uzZzqoLY4yxLhQQEIBFixYJXt6gfxhnz57FuXPnDK4Ue/ocPHgQhYWFpq6GWTt37hz3J2Yy586dM/gPgEH/MABgxIgROHDggKGrsaeMSCTCwoULMW3aNFNXxWxNnToVALg/MZPQHn+G4GsYjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYEeaoSRnR0NPz9/aFSqSCVSuHr64ulS5eiqqqq1XXmzZsHpVIJkUiErKwswbFefPFFiESiFicbGxvB2zl69ChsbW3x97//XfA6T4qned9bM3/+fL1jaebMmc2WOXHiBCIjI3Ho0CH4+Pjolp01a1azZceOHQulUgkLCwsMGDAAP/30U1fshkGE9tuvv/4aw4cPh1KphKenJ+bMmYPi4mKzj6fV1NSE+Ph4BAYG6pUfOXIEsbGxaGxs1CtPTU3VOxYcHR3bHVswMsCUKVNoypQphqxiVoKCgigxMZHKyspIo9FQUlISSSQSeuWVV9pcb9++fQSAMjMzDYoFoMVp3LhxgreTlpZGKpWKjhw5IngdcwCAkpKSOrSN7rrvQrWnP4WGhpKDgwOlp6dTXl4e1dbW6s2PioqiCRMmkEaj0ZWp1Wrq0aMHAaC0tLRm20xPT6fXX3+9fTvRBYT02/379xMAio2NpYqKCsrMzCQfHx8aMmQI1dfXm3U8IqLLly/TyJEjCQANHjy42fyEhAQKCgqi8vJyXVlTUxMVFhbSqVOn6LXXXqMePXoYFLM9x99TlTDGjx9PDQ0NemXTpk0jAHTz5s1W12tPwhg3bpxep9UKDQ2lkydPCq+0iVVXV1NAQIDB6xkjYZhae/ddqPYmDFdX1xbnrVu3jvz8/KimpkavXK1W01dffUVisZhcXV2poqJCb765Jwwh/Xb06NHUu3dvampq0i2zZcsWAkCnT58263hZWVn0xhtv0N69e2nIkCEtJgwiorCwMAoICGgxIS1YsKBLEsZTdUoqLS0NFhYWemXav3HV1dWtricSiQyOdezYMSiVSr2ygoICXLhwAWPGjDF4e6aya9culJSUmLoaJtGd9v3KlStYuXIlVq9eDZlM1mx+YGAgwsPDcevWLSxevNgENWw/If22oKAALi4uen3V3d0dAHDjxg2zjjd48GAcOnQIM2bMgFQqbXW5VatWISsrCwkJCQZt35i6JGHs2bMHw4YNg0wmg0KhgJeXF9asWQMAICJs3LgR/fv3h1Qqhb29PSZNmoRLly4BALZu3QqFQgG5XI7Dhw/j1VdfhUqlgpubG/bt2wcA6N+/P0QiEcRiMYYOHapr1KVLl8LW1hYymQxffPFFi3W7desWrK2t4e3tratPXFwc+vbtC6lUCltbWyxZssQon8P69euxYMECwcufPn0aHh4eEIlE2LJlCwBhn8emTZsgk8nQq1cvzJ8/Hy4uLpDJZAgMDMT58+cBAGFhYbCysoKzs7Mu3nvvvQeFQgGRSITS0lKEh4cjIiICV69ehUgkgq+vr1E+h+6878eOHYNKpcLHH3/cZZ+FEJs2bQIRYeLEia0uExMTAz8/P+zcuRMnTpxodTlj9EkAaGxsRFRUFDw8PGBtbY1nnnkGSUlJRtnfR/utj49Ps+SuvZ7g4+PT7eK1xN7eHkFBQUhISACZ6r13nf0XJj4+ngDQunXrqKysjO7cuUOfffYZzZgxg4genHO1srKiPXv2UEVFBWVnZ9Nzzz1Hjo6OVFxcTEREK1asIAB08uRJunv3LpWUlNCoUaNIoVBQXV0dNTQ0kJeXF3l4eDT7K7lw4UKKj49vsW737t0jpVJJYWFhurIVK1aQSCSiv/71r1ReXk7V1dWUmJho8CmpRxUWFpK/vz81NjYatF5BQQEBoM2bN+vVsa3Pg+jBqQuFQkEXL16k2tpays3NpeHDh5NSqdT9rZ4xYwY5OTnpxYuLiyMAdPv2bSIimjx5MqnVaoP3F0Y4JWWO+56WlkZKpZKio6M7tG9Exj0l5ePjQ/7+/i2uo1ar6dq1a0REdObMGRKLxeTl5UVVVVVE1PyUlDH6JBHR4sWLSSqV0sGDB6m8vJyWL19OYrGYfvjhB4P2+VEt9duMjAySSCS0adMm0mg0dOHCBerfv79B1wvNId7zzz/f6ikpIqLIyMgWv4ueiFNS9fX1WL16NUaPHo0PP/wQDg4OsLe3x1tvvYXhw4ejpqYGGzduxBtvvIGZM2fC1tYWgwYNwqefforS0lJs375db3uBgYFQqVTo2bMnQkJCcO/ePdy8eRMWFhZYsGABbt68iZSUFN3y1dXVOHToEObOndti/dauXQsXFxfExMQAAGpqahAfH48//OEPWLRoEezs7GBtbQ0HB4cOfxbr16/HBx98ALHYeB95a5+HlqWlpe5Xor+/P7Zu3YrKykrs3r3baHUwFVPt+/jx46HRaLBy5cqO7oLR3Lt3D9euXYNarX7ssgEBAVi4cCGuX7+ODz/8sNl8Y/XJ2tpabN26FcHBwZg8eTLs7Ozw0UcfQSKRdLgNHu23ABAUFIRly5YhLCwMKpUKAwcORGVlJXbu3NmhWKaI15Y+ffoAAHJycjo1Tms6NWFkZ2ejoqIC48aN0yvXfsHn5uaiqqoKw4YN05s/fPhwWFlZ6U4htMTKygrAg6QEPBj+amtrq3d+b+/evZg0aRJUKlWz9VNSUpCcnIzjx4/rrjVcuXIF1dXVeOmll9q3w60oKirCkSNHMHv2bKNu92GPfh4tGTZsGORyue7UwpPiad53ACgpKQERQS6XC1o+JiYGffv2RWJiIk6fPq03z1h9Mi8vD9XV1Rg4cKBuvrW1NZydnTvUBi31WwBYsWIFtm/fjpMnT6Kqqgo///wzAgMDERAQgIKCgm4T73G0bfzrr792Woy2dGrC0Gg0AAA7O7sW51dUVABAi/cl2NnZobKyUnAsGxsbvP322zhz5gy+//57AMC2bdsQFhbWbNn9+/dj/fr1yMjIgJeXl65c+zjunj17Co4rRGxsLP7yl7+0eDGyq0mlUty+fdvU1TCJJ3Xfa2trAaDNC6YPk8lk2L17N0QiEebOnYuamhrdPGP1yXv37gEAPvroI717BW7cuNHmAJO2tNZvf/nlF8TGxuLtt9/GmDFjoFAo4O3tjR07dqCoqAhxcXHdIp4Q1tbWAH5r867WqQmjd+/eAIDS0tIW52sTSUsHYUVFBdzc3AyKFxYWBolEgvj4eJw6dQru7u7N/qZv3rwZe/fuxbfffqurn5b2C/3+/fsGxW1LcXExvv76a7z77rtG22Z71dfXt+tzfRI8yfuu/RJ59MautmhfnJOfn68bgAIYr09qf3TFx8eDHgzf103teQlbW/02Pz8fjY2NzcpVKhUcHByQm5tr9vGEqqurA/Bbm3e1Tk0YXl5ecHBwwDfffNPi/IEDB8LGxgY//vijXvn58+dRV1eHoUOHGhTPzc0N06ZNw8GDB7Fy5UqEh4fr5hERli1bhpycHKSmprb4C2rgwIEQi8X47rvvDIrbltjYWMycOdMo10E6KiMjA0SEESNGAHhwnr+t0zhPkid533v16gWRSIS7d+8atN6aNWvQr18/ZGZm6sqM1Sfd3d0hk8kMejpCS4T0W20S++WXX/TKKysrcefOHd1wV3OMZyhtGzs5OXVajLZ0asKQSqVYvnw5Tp06hbCwMNy6dQtNTU2orKzExYsXIZPJEBERgZSUFOzduxcajQY5OTl455134OLigtDQUINjRkREoKGhAeXl5Xr3O1y8eBGffPIJduzYAYlE0uxxHRs2bEDPnj0xefJkHDx4ELt27YJGo0F2dnazC31C/frrr/j888+xcOHCdq3fUU1NTSgvL0dDQwOys7MRHh4ODw8P3bUUX19f3LlzB6mpqaivr8ft27ebjSF3cHBAUVERrl+/jsrKym7zJdtZ+56enm52w2rlcjl8fHwMfsOh9tTUw/ccGKtPymQyzJkzB/v27cPWrVuh0WjQ2NiIwsJC3RdtSEgInJyc2nwciZB+6+3tjdGjR2PHjh04deoUampqUFBQoKvrW2+9ZbbxDKVt40GDBhltmwbp7GFYRA/ugBw0aBDJZDKSyWT07LPPUmJiIhE9uL09Li6O+vTpQxKJhOzt7Sk4OJjy8vKIiCgxMZHkcjkBoD59+tDVq1dp+/btpFKpCAB5enrS5cuX9eKNHj2adu7cqVeWk5PT6qM6AFBcXBwREVVWVtK8efOoR48eZGNjQy+88AJFRUURAHJzc6N///vfgvd70aJFNHPmTIM/L63NmzeTs7MzASC5XE4TJ04U/HmEhoaSRCIhV1dXsrS0JJVKRZMmTaKrV6/qtl9WVkajR48mmUxG3t7e9MEHH9CSJUsIAPn6+tLNmzfpp59+Ik9PT7K2tqYXXnhBN6zycdDBYbXmuu9Hjx4lpVJJMTEx7d43LWMOqw0LCyOJRELV1dW6spSUFFKr1QSAHB0d6f33329xm0uWLNEbVmusPnn//n1atmwZeXh4kKWlJfXs2ZMmT55Mubm5REQUHBxMACgqKqrV/RXab0tLSyk8PJx8fX1JKpWSjY0NjRw5kv72t7/ptmWO8YiIzp49SyNHjiQXFxddHGdnZwoMDKTvvvtOb9nx48eTq6ur3h3mRF03rPapejTI00T7zCFT6WjC6AhT77tQxkwY+fn5ZGlpSXv27DFW9TpdY2MjjRo1inbt2sXxBCgtLSWZTEYbNmxoNu+JuA+DmZYhF0GfNE/yvtfU1OD48ePIz8/XXQT19fVFdHQ0oqOj23z6srlobGxEamoqKisrERISwvEEWLVqFYYMGaIb+UlEKCoqwunTp3HlyhWjxHgcThgGuHTpUquPLH94etwBYqztsKfTnTt38Morr8DPz0/vptTIyEhMnToVISEhBl8A72oZGRk4dOgQ0tPTBd8/8jTH27hxI7KysnD06FFIJBIAwOHDh+Hq6opRo0bhH//4R4djCNLZf2FY14uMjCQrKysCQF5eXnTgwIEurwNMdErKHPZdqM7qT8ePH6dly5YZfbvMNFJTU2nt2rXNHnvUUe05/kREwp9iNXXqVADAgQMHOid7sSeGSCRCUlISpk2bZuqqmC3uT8yU2nP88SkpxhhjgnDCYIwxJggnDMYYY4JwwmCMMSYIJwzGGGPCGDoMC23cNs8TTzzxxFP3mQwdVmsJA40YMcJkD9Nj3cf06dMRHh6OgIAAU1fFbMXHxwMA9ydmEtrjzxAGJwztI8QZa8v06dMREBDAx0obtOPf+TNiptCe+3/4GgZjjDFBOGEwxhgThBMGY4wxQThhMMYYE4QTBmOMMUG6RcI4d+4c+vfvD7FYDJFIBCcnJ8TExJi6Wjh06BB8fHx0769wdnbGzJkzTV0t9gSbP3++3jtTWjreTpw4gcjIyGbH56xZs5otO3bsWCiVSlhYWGDAgAFGff+0sURHR8Pf3x8qlQpSqRS+vr5YunRpsxdFff311xg+fDiUSiU8PT0xZ84cFBcXm308raamJsTHxyMwMFCv/MiRI4iNjW32UrDU1FS9Y8HR0bHdsQUz9MY9U74PY9y4cQSAysvLTVaHlqjVarK1tTV1NcwKYLpXtHYX7X1Fq4ODA6Wnp1NeXh7V1tbqzY+KiqIJEyaQRqPRlanVaurRowcBoLS0tGbbTE9P13unt7kJCgqixMREKisrI41GQ0lJSSSRSOiVV17RLbN//34CQLGxsVRRUUGZmZnk4+NDQ4YMofr6erOOR0R0+fJlGjlyJAGgwYMHN5ufkJBAQUFBet99TU1NVFhYSKdOnaLXXnuNX9Fqbmpqapplf2ZeOrONzKX9ra2tdW/ck0qluvL169dj//79SE5OhlKp1Ftn06ZNEIvFCA0NNfu38T3KxsYGoaGhcHBwgFKpxLRp0xAcHIxjx46hoKAAAPDZZ5+hd+/eWLJkCWxtbTFkyBAsWrQIWVlZOH/+vFnH+/e//40PP/wQ77zzDoYMGdLiMgsWLMDgwYPx2muvoaGhAcCDd85o37jXp08fg2K2FycMA+zatQslJSWmrgZrQ2e2kTm3/5UrV7By5UqsXr0aMpms2fzAwECEh4fj1q1bWLx4sQlq2H5paWmwsLDQK9OefqmurgYAFBQUwMXFBSKRSLeMu7s7AODGjRtmHW/w4ME4dOgQZsyYofcD4FGrVq1CVlYWEhISDNq+MXXrhLF161YoFArI5XIcPnwYr776KlQqFdzc3LBv3z4AD35ZyWQy9OrVC/Pnz4eLiwtkMhkCAwN1vwTCwsJgZWUFZ2dn3bbfe+89KBQKiEQilJaWIjw8HBEREbh69SpEIhF8fX0Nru8///lP+Pv7w9bWFjKZDIMGDcLx48cBAPPmzdOdi1Sr1cjMzAQAzJkzB3K5HLa2tjhy5AgaGxsRFRUFDw8PWFtb45lnnkFSUhIA4JNPPoFcLodSqURJSQkiIiLg6uqKvLy8Dn3OXYGIsHHjRvTv3x9SqRT29vaYNGkSLl26BKD9bdTZ7X/s2DGoVCp8/PHHXfhpNbdp0yYQESZOnNjqMjExMfDz88POnTtx4sSJVpd7XFsI6XcA2jxWO+rWrVuwtraGt7c3AMDHx6dZMtdeT/Dx8el28Vpib2+PoKAgJCQkgIS/KNW4OvuclzG1dA1jxYoVBIBOnjxJd+/epZKSEho1ahQpFAqqq6sjogfnfRUKBV28eJFqa2spNzeXhg8fTkqlkm7evElERDNmzCAnJye9eHFxcQSAbt++TUREkydPJrVa3axeQq9hHDhwgFatWkV37tyhsrIyGjFihN55x8mTJ5OFhQXdunVLb70333yTjhw5QkREixcvJqlUSgcPHqTy8nJavnw5icVi+uGHH/Q+jwULFtDmzZvpjTfeoP/7v/97bN2MDQZew4iKiiIrKyvas2cPVVRUUHZ2Nj333HPk6OhIxcXFRNT+NurM9k9LSyOlUknR0dGC91WrvdcwXF1dm5X7+PiQv79/i+uo1Wq6du0aERGdOXOGxGIxeXl5UVVVFRE1v4YhpC2E9LvHHavtde/ePVIqlRQWFqYry8jIIIlEQps2bSKNRkMXLlyg/v3707hx4zoUq6vjPf/88y1ew9CKjIwkAJSZmalXvmDBAr6GYYjAwECoVCr07NkTISEhuHfvHm7evKmbb2lpqfvF5O/vj61bt6KyshK7d+/usjpOmTIF/+///T/Y29vDwcEBEydORFlZGW7fvg0AeOedd9DY2KhXJ41Ggx9++AGvvfYaamtrsXXrVgQHB2Py5Mmws7PDRx99BIlE0mw/1q9fj/fffx+HDh1Cv379umwf26OmpgYbN27EG2+8gZkzZ8LW1haDBg3Cp59+itLSUmzfvr3DMTqr/cePHw+NRoOVK1d2uI7tde/ePVy7dg1qtfqxywYEBGDhwoW4fv06Pvzww2bzDW2L1vqdIceqodauXQsXFxe9kZJBQUFYtmwZwsLCoFKpMHDgQFRWVmLnzp0dimWKeG3RXqvIycnp1DiteWISxsOsrKwAAPX19a0uM2zYMMjlct3fbFOQSCQAoBsuN2bMGPj5+eHzzz/X/eXcv38/QkJCYGFhgby8PFRXV2PgwIG6bVhbW8PZ2dmk+9FRubm5qKqqwrBhw/TKhw8fDisrK4MvIgphDu1vLCUlJSAiyOVyQcvHxMSgb9++SExMxOnTp/XmdaQtHu53nXWspqSkIDk5GcePH9e7sL9ixQps374dJ0+eRFVVFX7++WcEBgYiICBAd6G6O8R7HG0b//rrr50Woy1PZMIQSiqV6n7dd4V//OMfePHFF9GzZ09IpVIsXbpUb75IJML8+fPx888/4+TJkwCAL7/8Em+99RaAB78kAeCjjz7SG39948YN3cW47qiiogLAg9Epj7Kzs0NlZWWnxO3q9u8stbW1ANDmBdOHyWQy7N69GyKRCHPnzkVNTY1unrHaojOO1f3792P9+vXIyMiAl5eXrvyXX35BbGws3n77bYwZMwYKhQLe3t7YsWMHioqKEBcX1y3iCWFtbQ3gtzbvak9twqivr0dFRQXc3Nw6Nc6pU6cQHx+PmzdvIjg4GM7Ozjh//jzu3r2L2NjYZsvPnj0bMpkMO3fuRF5eHlQqFTw9PQEAPXv2BPDgOfZEpDedPXu2U/ejM9nZ2QFAi19GndVGXdX+XUH7JfLojV1tCQgIwKJFi5Cfn481a9boyo3VFsY+Vjdv3oy9e/fi22+/Re/evfXm5efno7GxsVm5SqWCg4MDcnNzzT6eUHV1dQB+a/OuZvD7MJ4UGRkZICKMGDECwINz3G2dwmqv//3f/4VCoUBOTg7q6+vx7rvv6kZRPDwkT8ve3h7Tp0/H/v37oVQq8Ze//EU3z93dHTKZDFlZWUavpykNHDgQNjY2+PHHH/XKz58/j7q6OgwdOhSAcduoq9q/K/Tq1Qsikcjg+yvWrFmDtLQ0ZGZmwsPDA4DwtngcYx2rRIQPP/wQ5eXlSE1NhaVl868sbRL75Zdf9MorKytx584d3XBXc4xnKG0bOzk5dVqMtjw1/zCamppQXl6OhoYGZGdnIzw8HB4eHpg9ezYAwNfXF3fu3EFqairq6+tx+/btZuOpHRwcUFRUhOvXr6OysrLNL5j6+nr8+uuvyMjIgEKh0HXIEydOoLa2Fvn5+a2eD37nnXdw//59pKWlYcKECbpymUyGOXPmYN++fdi6dSs0Gg0aGxtRWFjY7ODtTmQyGSIiIpCSkoK9e/dCo9EgJycH77zzDlxcXBAaGgqgY23UWe2fnp5u8mG1crkcPj4+KCwsNGg97amph+85ENoWQrb9uGM1JCQETk5ObT6O5OLFi/jkk0+wY8cOSCQSvdNbIpEIGzZsgLe3N0aPHo0dO3bg1KlTqKmpQUFBga6u2lO65hjPUNo2HjRokNG2aZDOHoZlDOfOnaMBAwaQWCwmAOTs7Ewff/wxJSYmklwuJwDUp08funr1Km3fvp1UKhUBIE9PT7p8+TKFhoaSRCIhV1dXsrS0JJVKRZMmTaKrV6/qYpSVldHo0aNJJpORt7c3ffDBB7RkyRICQL6+vnTz5k366aefyNPTk6ytremFF16gbdu2kVqtfux7c1NSUoiIaNmyZeTg4EB2dnY0depU2rJlCwEgtVqtG96p9eyzz1JkZGSzz+L+/fu0bNky8vDwIEtLS+rZsydNnjyZcnNzKTY2lqytrQkAubu70549ezq3YdoAA4fVNjU1UVxcHPXp04ckEgnZ29tTcHAw5eXl6ZZpTxsVFxd3WvsXFxfT0aNHSalUUkxMjMGfkTGH1YaFhZFEIqHq6mpdWUpKiu74dHR0pPfff7/FbS5ZskRvWO3j2kJov2vrWCUiCg4OJgAUFRXV6v7m5OS02bfi4uKIiKi0tJTCw8PJ19eXpFIp2djY0MiRI+lvf/ubblvmGI+I6OzZszRy5EhycXHRxXF2dqbAwED67rvv9JYdP348ubq6UlNTk155Vw2r7RYJo6O0z9/pTl577TX6+eefTV2NdjM0YXQmc21/YyaM/Px8srS0NOmPBEM1NjbSqFGjaNeuXRxPgNLSUpLJZLRhw4Zm8/g+DCMz5IKgKTx8eis7OxsymUx3VynrOHNvf0PU1NTg+PHjyM/P110E9fX1RXR0NKKjo5s9VdUcNTY2IjU1FZWVlQgJCeF4AqxatQpDhgxBWFgYgAfXW4qKinD69GlcuXLFKDEe56lJGOZu2bJlyM/Px+XLlzFnzhy9kSuMPezOnTu6hw/OnTtXVx4ZGYmpU6ciJCTE7B8wmJGRgUOHDiE9PV3w/SNPc7yNGzciKysLR48e1d2/dfjwYd3DB//xj390OIYgnf0XxtQiIyPJysqKAJCXlxcdOHDA1FVq0YoVK0gsFpO7u7vuMSDdGczklJQ5t39n9afjx4/TsmXLjL5dZhqpqam0du1aamhoMOp223P8iYiEP8Vq6tSpAIADBw50TvZiTwyRSISkpCRMmzbN1FUxW9yfmCm15/jjU1KMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBCDnyVVWFiI5OTkzqgLe8J05wcidgXtYx64PzFTKCwsNPzhm4YOw8JjHoPBE0888cRT95g6dVgtY08D7VBg/uXPmD6+hsEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEEtTV4AxUzp16hTOnj2rV3bp0iUAQGxsrF55QEAAfv/733dZ3RgzNyIiIlNXgjFTOXnyJP7whz9AIpFALG75D3dTUxPq6+tx4sQJvPTSS11cQ8bMBycM9lRramqCs7Mzbt++3eZyjo6OKC4uhoWFRRfVjDHzw9cw2FNNLBZjxowZsLKyanUZKysrzJw5k5MFe+pxwmBPvT/+8Y+oq6trdX5dXR3++Mc/dmGNGDNPfEqKMQBeXl64ceNGi/Pc3d1x48YNiESiLq4VY+aF/2EwBmDWrFmQSCTNyiUSCWbPns3JgjHwPwzGADwYStu/f/8W5+yOtQEAACAASURBVF24cAEDBgzo4hoxZn74HwZjAPr164cBAwY0+yfh7+/PyYKx/+CEwdh//OlPf9IbCSWRSPDnP//ZhDVizLzwKSnG/qOgoACenp7QdgmRSISff/4ZXl5epq0YY2aC/2Ew9h/u7u54/vnnIRaLIRaL8fzzz3OyYOwhnDAYe8isWbMgEokgFosxa9YsU1eHMbPCp6QYe0hpaSmcnZ0BAEVFRejVq5eJa8SYGaEOmDJlCgHgiSeeeOKpG0xTpkzpyFc+dfjx5iNGjMDChQs7uhn2hDh79iwSEhKQlJRk6qq026lTpyASiTBq1KhOizF9+nSEh4cjICCg02Iw9rD4+PgOb6PDCcPNzQ3Tpk3rcEXYkyMhIaFbHxOvvvoqAECpVHZajOnTpyMgIKBbf06sezlw4ECHt8EvUGLsEZ2ZKBjrzniUFGOMMUE4YTDGGBOEEwZjjDFBOGEwxhgThBOGEUVHR8Pf3x8qlQpSqRS+vr5YunQpqqqqWl1n3rx5UCqVEIlEyMrKEhzrxRdfhEgkanGysbExxu6Y1NGjR2Fra4u///3vpq5Kt3DixAlERkbi0KFD8PHx0R0LLd2tPnbsWCiVSlhYWGDAgAH46aefTFDjtgntS19//TWGDx8OpVIJT09PzJkzB8XFxWYfT6upqQnx8fEIDAzUKz9y5AhiY2PR2NjY7m13io7euNfRG0GeJEFBQZSYmEhlZWWk0WgoKSmJJBIJvfLKK22ut2/fPgJAmZmZBsVCKzfnjBs3rqO70m5JSUnUwcOKiIjS0tJIpVLRkSNHjFAr8wOAkpKSjLKtqKgomjBhAmk0Gl2ZWq2mHj16EABKS0trtk56ejq9/vrrRonfGYT0pf379xMAio2NpYqKCsrMzCQfHx8aMmQI1dfXm3U8IqLLly/TyJEjCQANHjy42fyEhAQKCgqi8vJyg7fdEmN8X3PCMKLx48dTQ0ODXtm0adMIAN28ebPV9dqTMMaNG6f3BaEVGhpKJ0+eFF5pIzNWwjC16upqCggI6LTtGythrFu3jvz8/KimpkavXK1W01dffUVisZhcXV2poqJCb765JwwhfWn06NHUu3dvampq0i2zZcsWAkCnT58263hZWVn0xhtv0N69e2nIkCEtJgwiorCwMAoICGhXQnqUMb6v+ZSUEaWlpem9TwEAHB0dAQDV1dWtrtee138eO3as2f0CBQUFuHDhAsaMGWPw9pi+Xbt2oaSkxNTVaNOVK1ewcuVKrF69GjKZrNn8wMBAhIeH49atW1i8eLEJath+QvpSQUEBXFxc9PqPu7s7ALT6fnZziTd48GAcOnQIM2bMgFQqbXW5VatWISsrCwkJCQZtv7M8EQljz549GDZsGGQyGRQKBby8vLBmzRoAABFh48aN6N+/P6RSKezt7TFp0iRcunQJALB161YoFArI5XIcPnwYr776KlQqFdzc3LBv3z4AQP/+/XVPMB06dKjuAFq6dClsbW0hk8nwxRdftFi3W7duwdraGt7e3rr6xMXFoW/fvpBKpbC1tcWSJUuM8jmsX78eCxYsMMq2TOn06dPw8PCASCTCli1bAAhrp02bNkEmk6FXr16YP38+XFxcIJPJEBgYiPPnzwMAwsLCYGVlpXvAIAC89957UCgUEIlEKC0tRXh4OCIiInD16lWIRCL4+voCeJCkVSoVPv744y7+RFq2adMmEBEmTpzY6jIxMTHw8/PDzp07ceLEiVaXM0Y/AYDGxkZERUXBw8MD1tbWeOaZZ4z2mJhH+5KPj0+zpK69nuDj49Pt4rXE3t4eQUFBSEhI0L2nxaRM/Reno+Lj4wkArVu3jsrKyujOnTv02Wef0YwZM4jowfldKysr2rNnD1VUVFB2djY999xz5OjoSMXFxUREtGLFCgJAJ0+epLt371JJSQmNGjWKFAoF1dXVUUNDA3l5eZGHh0ezv60LFy6k+Pj4Fut27949UiqVFBYWpitbsWIFiUQi+utf/0rl5eVUXV1NiYmJBp+SelRhYSH5+/tTY2Nju7dhDMY6JVVQUEAAaPPmzbqyx7UT0YNTcgqFgi5evEi1tbWUm5tLw4cPJ6VSqTu1MGPGDHJyctKLFxcXRwDo9u3bREQ0efJkUqvVesukpaWRUqmk6OjoDu8fjHBKysfHh/z9/Vucp1ar6dq1a0REdObMGRKLxeTl5UVVVVVE1PyUlDH6CRHR4sWLSSqV0sGDB6m8vJyWL19OYrGYfvjhhw7ta0t9KSMjgyQSCW3atIk0Gg1duHCB+vfvb5RreF0Z7/nnn2/1lBQRUWRkZIe/H4j4GgbV1dWRnZ0djR49Wq+8oaGBEhISqLq6mmxsbCgkJERv/vfff08AdB1f2xEePg+s/RK/cuUKEf2WmJKTk3XL3Lt3jzw8POju3bst1m/FihXk5+enu9ZQXV1NcrmcXn75Zb3l2nMN41Hvv/8+bdu2rd3rG0tXJIy22ik0NJRsbW31tvXDDz8QAFq9ejURtT9hGFNHE0ZVVRWJRCKaMGFCi/MfThhERBEREQSA3n//fSLSTxjG6ic1NTUkl8v1tlNdXU1SqZTefffddu+rNvbDfUnro48+0hvw4ebmRgUFBR2K1dXxHpcwPv/8cwJAX375ZYfiPPXXMLKzs1FRUYFx48bplVtYWGDBggXIzc1FVVUVhg0bpjd/+PDhsLKy0p2maImVlRUAoL6+HsCD4a+2trZ65xL37t2LSZMmQaVSNVs/JSUFycnJOH78uO5aw5UrV1BdXY2XXnqpfTvciqKiIhw5cgSzZ8826na7g0fbqSXDhg2DXC7XnV55EpSUlICIIJfLBS0fExODvn37IjExEadPn9abZ6x+kpeXh+rqagwcOFA339raGs7Ozh367FvqSwCwYsUKbN++HSdPnkRVVRV+/vlnBAYGIiAgAAUFBd0m3uNo2/jXX3/ttBhCdeuEodFoAAB2dnYtzq+oqACAFu9LsLOzQ2VlpeBYNjY2ePvtt3HmzBl8//33AIBt27YhLCys2bL79+/H+vXrkZGRofeKz8LCQgBAz549BccVIjY2Fn/5y19avPDJHpBKpbh9+7apq2E0tbW1ANDmBdOHyWQy7N69GyKRCHPnzkVNTY1unrH6yb179wAAH330kd59QTdu3Ghz0EdbWutLv/zyC2JjY/H2229jzJgxUCgU8Pb2xo4dO1BUVIS4uLhuEU8Ia2trAL+1uSl164TRu3dvAA/ektYSbSJp6YCvqKiAm5ubQfHCwsIgkUgQHx+PU6dOwd3dHWq1Wm+ZzZs3Y+/evfj222919dPSfqHfv3/foLhtKS4uxtdff413333XaNt80tTX17ervc2Z9kvEkBu7AgICsGjRIuTn5+sGhQDG6yfaH0Lx8fGgB6e7ddPZs2cF11Orrb6Un5+PxsbGZuUqlQoODg7Izc01+3hC1dXVAfitzU2pWycMLy8vODg44Jtvvmlx/sCBA2FjY4Mff/xRr/z8+fOoq6vD0KFDDYqnfffHwYMHsXLlSoSHh+vmERGWLVuGnJwcpKamtvhrbeDAgRCLxfjuu+8MituW2NhYzJw5Ew4ODkbb5pMmIyMDRIQRI0YAACwtLds8hdUd9OrVCyKRCHfv3jVovTVr1qBfv37IzMzUlRmrn7i7u0Mmkxn0xIKWCOlL2iT2yy+/6JVXVlbizp07uuGu5hjPUNo2dnJy6rQYQnXrhCGVSrF8+XKcOnUKYWFhuHXrFpqamlBZWYmLFy9CJpMhIiICKSkp2Lt3LzQaDXJycvDOO+/AxcUFoaGhBseMiIhAQ0MDysvL9e53uHjxIj755BPs2LEDEomk2eM6NmzYgJ49e2Ly5Mk4ePAgdu3aBY1Gg+zsbGzfvr1d+//rr7/i888/5zcePqKpqQnl5eVoaGhAdnY2wsPD4eHhobvG4+vrizt37iA1NRX19fW4fft2s3H0Dg4OKCoqwvXr11FZWYn6+nqkp6ebzbBauVwOHx8f3WlOobSnph6+58BY/UQmk2HOnDnYt28ftm7dCo1Gg8bGRhQWFuq+aENCQuDk5NTm40iE9CVvb2+MHj0aO3bswKlTp1BTU4OCggJdXd966y2zjWcobRsPGjTIaNtsN1NfdTeGLVu20KBBg0gmk5FMJqNnn32WEhMTiYioqamJ4uLiqE+fPiSRSMje3p6Cg4MpLy+PiB6M8pDL5QSA+vTpQ1evXqXt27eTSqUiAOTp6UmXL1/Wizd69GjauXOnXllOTk6b79KNi4sjIqLKykqaN28e9ejRg2xsbOiFF16gqKgo3YiLf//734L3e9GiRTRz5syOfHRGZ4xRUps3byZnZ2cCQHK5nCZOnCi4nUJDQ0kikZCrqytZWlqSSqWiSZMm0dWrV3XbLysro9GjR5NMJiNvb2/64IMPaMmSJQSAfH196ebNm/TTTz+Rp6cnWVtb0wsvvEDFxcV09OhRUiqVFBMT09GPySjDasPCwkgikVB1dbWuLCUlhdRqNQEgR0dH3aioRy1ZskRvWK2x+sn9+/dp2bJl5OHhQZaWltSzZ0+aPHky5ebmEhFRcHAwAaCoqKhW90toXyotLaXw8HDy9fUlqVRKNjY2NHLkSPrb3/6m25Y5xiMiOnv2LI0cOZJcXFx0cZydnSkwMJC+++47vWXHjx9Prq6ueneYt8dTP6yWmR9TPxokNDSUHBwcTBZfKGMkjPz8fLK0tKQ9e/YYqVadr7GxkUaNGkW7du3ieAKUlpaSTCajDRs2dHhbT/2wWsZaYnZP+Owkvr6+iI6ORnR0dJtPRDYXjY2NSE1NRWVlJUJCQjieAKtWrcKQIUNaHI1pCpwwzMilS5dafWT5w1NXHPyse4iMjMTUqVMREhJi8AXwrpaRkYFDhw4hPT1d8P0jT3O8jRs3IisrC0ePHoVEIjFCDTvO0tQVYL/p16+feTwvpptavnw5du/ejbq6Onh7eyMuLg5TpkwxdbU63ccff4xvvvkG69atw/r1601dnVa99NJLRr9p9UmNd/jwYdy/fx8ZGRnNHopoSpww2BNj7dq1WLt2ramrYRJjx47F2LFjTV0NZiSvv/46Xn/9dVNXoxk+JcUYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOnwRe/CwkIkJycboy7sCaB9yBwfE4/XngfyMdZehYWFHX8AZ0fvHEQbt9TzxBNPPPFkPlNH7/Tu8D+MKVOm4MCBAx3dzFNPJBIhKSkJ06ZNM3VVOiQ5ORnTp0/n+0ke40lpb9Z9TJ06tcPb4GsYjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEJMljLy8PHzwwQcYMGAAlEolLC0tYWtrCz8/P4wfP94kNzVFR0fD398fKpUKUqkUvr6+WLp0qe7lNIcOHYKPj0+z91NYWVmhV69eePHFFxEXF4fy8vIur7sQGzZsQK9evSASifDpp5+aujqsnU6cOIHIyMhmx+OsWbOaLTt27FgolUpYWFhgwIABRn3XtLE8rt9p1dfXY+3atfD19YWVlRXs7OwwcOBAXL9+3azjxcbGol+/frC2toZCoUC/fv2wcuVKaDQawXU6cuQIYmNjTf9ysI7euNeeG0F27txJEomEfv/739OxY8eovLycamtr6erVq7R//34KDAykzz77rCNVa5egoCBKTEyksrIy0mg0lJSURBKJhF555RW95dRqNdna2hLRg3chl5eX0//8z//Q7NmzSSQSkYuLC/3www8GxQY6/spOIfLz8wkAbdu2rVO2b+pXtHYX7W3vqKgomjBhAmk0Gl2ZWq2mHj16EABKS0trtk56erre+7vNjdB+FxwcTH379qVz585RfX09FRUV0cSJEyknJ8es440fP542bNhAJSUlVFlZScnJySSRSOjll182qE4JCQkUFBRE5eXlBsXX6pbv9D579ixZWFjQmDFjqL6+vsVljh07Rps3b+5I1dpl/Pjx1NDQoFc2bdo0AkA3b97UlT2cMB514MABEovF1KtXL6qoqBAcmxNGx1VXV1NAQEC32HZ72nvdunXk5+dHNTU1euVqtZq++uorEovF5Orq2uy4M/eEIaTf7du3j0QiEWVnZ3e7eMHBwc3abOrUqQSAioqKBNeJiCgsLIwCAgJa/e5sS7d8p3dMTAwaGxuxbt06WFq2fKP5uHHj8P7773dxzYC0tLRmb7dydHQEAFRXVwvaxpQpUzB79myUlJTwaZ8utmvXLpSUlHS7bQtx5coVrFy5EqtXr4ZMJms2PzAwEOHh4bh16xYWL15sghq2n5B+t23bNjz33HMYNGhQt4uXkpLSrM1cXV0BQHfKSeh3z6pVq5CVlYWEhIQO16s9ujRh1NXV4eTJk+jRowd+97vfCVqHiLBx40b0798fUqkU9vb2mDRpEi5dugQA2Lp1KxQKBeRyOQ4fPoxXX30VKpUKbm5u2LdvHwCgf//+EIlEEIvFGDp0qK4Bli5dCltbW8hkMnzxxRctxr916xasra3h7e0teD9nz54NAEhPTxe8jiE++eQTyOVyKJVKlJSUICIiAq6ursjLy0NjYyOioqLg4eEBa2trPPPMM0hKSmp1W2FhYbCysoKzs7Ou7L333oNCoYBIJEJpaWmn7MPDHtfGQuoYHh6OiIgIXL16FSKRCL6+vti0aRNkMhl69eqF+fPnw8XFBTKZDIGBgTh//nyHtg0Ax44dg0qlwscff9zpn9GmTZtARJg4cWKry8TExMDPzw87d+7EiRMnWl3OGH0KgMHHmiEe7nd1dXU4d+4chgwZYpRtm0O8/Px82NnZwdPTU1CdtOzt7REUFISEhATTPH6nI39PDP2Lc/nyZQJAI0aMELxOVFQUWVlZ0Z49e6iiooKys7PpueeeI0dHRyouLiYiohUrVhAAOnnyJN29e5dKSkpo1KhRpFAoqK6ujhoaGsjLy4s8PDya/e1buHAhxcfHtxj73r17pFQqKSwsTK+8rVNSREQajYYAkLu7u+D9hIGnKLT7vGDBAtq8eTO98cYb9H//93+0ePFikkqldPDgQSovL6fly5eTWCzWXVNp6ZTUjBkzyMnJSW/7cXFxBIBu374tuE5E7TslJaSNhdRx8uTJpFar9ZYJDQ0lhUJBFy9epNraWsrNzaXhw4eTUqnU/dVv77bT0tJIqVRSdHS0QftLZHh7+/j4kL+/f4vz1Go1Xbt2jYiIzpw5Q2KxmLy8vKiqqoqImp+SMkafIqLHHmvt9Wi/u3btGgGgIUOG0IsvvkjOzs4klUqpX79+tGXLFmpqauoW8erq6qiwsJA2b95MUqmU9uzZI7hOD4uMjCQAlJmZaVD8bncN48cffyQA9Ic//EHQ8tXV1WRjY0MhISF65d9//z0B0HVU7cH98HnCxMREAkBXrlwhIqL4+HgCQMnJybpl7t27Rx4eHnT37t0W469YsYL8/Pz0LjASPT5hEBGJRCKys7MTtJ9E7U8YD+9zTU0NyeVyvc+rurqapFIpvfvuu0RkfglDaBt3JGE82lY//PADAaDVq1d3aNsdYUh7V1VVkUgkogkTJrQ4/+GEQUQUERFBAOj9998nIv2EYaw+JeRYa69H+11OTg4BoJdffpn+9a9/UVlZGVVUVNCHH35IAGjv3r3dIp6TkxMBoB49etB//dd/6RKvkDo97PPPPycA9OWXXxoUv9tdw7CxsQEg/HpAbm4uqqqqMGzYML3y4cOHw8rKSndaoSVWVlYAHgyNA4B58+bB1tZW79zf3r17MWnSJKhUqmbrp6SkIDk5GcePH4dSqRRUX6179+6BiFrcbmfKy8tDdXU1Bg4cqCuztraGs7Oz7nSDuelIG7fXsGHDIJfLzfYzeVRJSQmICHK5XNDyMTEx6Nu3LxITE3H69Gm9ecbqU511rLXU76RSKQBgwIABCAwMhIODA2xtbbF69WrY2tpi+/bt3SJeQUEBSkpK8PXXX+O///u/8eyzz7Z4Xexx3z3a4+DXX39tVz06oksThpeXF2QyGS5fvixo+YqKCgC/JZqH2dnZobKyUnBsGxsbvP322zhz5gy+//57AA8ubIWFhTVbdv/+/Vi/fj0yMjLg5eUlOIaWdv/69etn8Lodce/ePQDARx99pHefyI0bNwQn6a5mzDY2hFQqxe3btztl28ZWW1sL4LcvsseRyWTYvXs3RCIR5s6di5qaGt08Y33enXGstdbvXFxcAKDZ9TQrKyt4enri6tWr3SKeRCJBz549MXbsWOzfvx+5ublYu3atoDo9zNraGsBvx0VX6tKEIZVKMW7cOJSWluJf//pXq8vduXMH8+bNg52dHQC0eBBXVFQY/PaosLAwSCQSxMfH49SpU3B3d4dardZbZvPmzdi7dy++/fZb9O7d26Dtax07dgwA8Oqrr7Zr/fbq2bMnACA+Ph704HSjbjLXt7sZu42FqK+v77RtdwbtF4QhN20FBARg0aJFyM/Px5o1a3Tlxvq8jX2stdXvbGxs0KdPH1y8eLHZeg0NDbC1tTX7eI/y9fWFhYUFcnNzBdXpYXV1dQB+Oy66UpcPq121ahWkUikWLVqk98vnYRcuXIClpSUGDhwIGxsb/Pjjj3rzz58/j7q6OgwdOtSg2G5ubpg2bRoOHjyIlStXIjw8XDePiLBs2TLk5OQgNTW1xV9gQhQXFyM+Ph5ubm6YO3duu7bRXu7u7pDJZMjKyjJoPUtLS92pu64mtI2NWceMjAwQEUaMGGH0bXcG7d35d+/eNWi9NWvWoF+/fsjMzNSVGatPtfdYe5TQfjd9+nRkZmbi559/1pVVV1fjxo0bBg197ep4ZWVlePPNN5uV5+fno7GxEe7u7gZ/92iPAycnJ8H1MJYuTxhDhgzBV199hQsXLmDUqFE4evQo7t69i/r6ely7dg07duzAW2+9BYlEAplMhoiICKSkpGDv3r3QaDTIycnBO++8AxcXF4SGhhocPyIiAg0NDSgvL8eYMWN05RcvXsQnn3yCHTt2QCKRNHv8x4YNG/S2Q0SoqqpCU1MTiAi3b99GUlISRo4cCQsLC6Smpnb5NQyZTIY5c+Zg37592Lp1KzQaDRobG1FYWIhffvml1fV8fX1x584dpKamor6+Hrdv38aNGze6rM5C2lhIHR0cHFBUVITr16+jsrJSlwSamppQXl6OhoYGZGdnIzw8HB4eHrrhz+3ddnp6epcMq5XL5fDx8UFhYaFB62lPTT08vt9YfUrIsRYSEgInJ6c2H0citN8tWrQInp6emD17Nm7evImysjIsW7YMNTU1+PDDD802nkKhwDfffINvv/0WGo0G9fX1yMzMxJ///GcoFAosWrTI4O8e7XFgjHtEDNaRK+Yduep+8+ZNWrx4MQ0aNIhsbGzIwsKC7Ozs6Nlnn6W33nqL/vWvfxHRg0dvxMXFUZ8+fUgikZC9vT0FBwdTXl4eET0YuSGXywkA9enTh65evUrbt28nlUpFAMjT05MuX76sF3v06NG0c+dOvTLtyIjWpri4ODpy5Ag988wzJJfLycrKisRiMQHQjYj63e9+R9HR0VRWVmbw5wEDRs3ExsaStbW1bujuw8Pz7t+/T8uWLSMPDw+ytLSknj170uTJkyk3N5f++te/6kZqKBQKeuONN4iIqKysjEaPHk0ymYy8vb3pgw8+oCVLlhAA8vX11bvT9HHaM6z2cW0stI4//fQTeXp6krW1Nb3wwgtUXFxMoaGhJJFIyNXVlSwtLUmlUtGkSZPo6tWrHd720aNHSalUUkxMjEH7S2T4qLiwsDCSSCRUXV2tK0tJSSG1Wk0AyNHRUTcq6lFLlizRG1ZrrD7V1rFG9OAOZwAUFRXV6n4J6XdaBQUF9Mc//pHs7e1JKpXS7373O0pPT9fNN8d4REQTJ04kb29vsrGxIalUSmq1mkJCQnSPGDGkTkQP7gp3dXU1eHhvtxtWy1pn6BeIuTK3Z0mFhoaSg4ODqavRjKHtnZ+fT5aWlm2O3Tc3jY2NNGrUKNq1axfHM5LS0lKSyWS0YcMGg9ftdsNqGTMFkz/h0wh8fX0RHR2N6OjoZk9VNUeNjY1ITU1FZWUlQkJCOJ6RrFq1CkOGDGlxdGdX4ITBWDcRGRmJqVOnIiQkxOAL4F0tIyMDhw4dQnp6uuD7Rzhe2zZu3IisrCwcPXoUEomkS2I+ihMGe2ItX74cu3fvxt27d+Ht7Y2DBw+aukod9vHHHyMsLAzr1q0zdVXa9NJLL+Grr77Se0YXx2u/w4cP4/79+8jIyIC9vX2XxGxJy4+LZewJsHbt2mY3Rj0Jxo4di7Fjx5q6GqwLvf7663j99ddNXQ3+h8EYY0wYThiMMcYE4YTBGGNMEE4YjDHGBOnwRe9z585h6tSpxqjLUy8+Ph4HDhwwdTU6RPvYAj4mHu9JaG/WfZw7d073/LT2EhG1/z1/GzduNNunoDLWXjk5OQBM9KwexjqR9inG7dWhhMHYk2jatGkAgOTkZBPXhDHzwtcwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAkiIiIydSUYM5Uvv/wSGzduRGNjo66stLQUAODo6Kgrs7CwwKJFi/CnP/2py+vImLnghMGeapcvX0bfvn0FLZuXlwc/P79OrhFj5otPSbGnmp+fHwYPHgyRSNTqMiKRCIMHD+ZkwZ56nDDYU+9Pf/oTLCwsWp1vaWmJP//5z11YI8bME5+SYk+9oqIiuLu7ZtqdNQAAIABJREFUo6mpqcX5IpEIBQUFcHV17eKaMWZe+B8Ge+r17t0bgYGBEIubdwexWIyRI0dysmAMnDAYAwDMmjWrxXKRSMQjoxj7Dz4lxRiA8vJyODk5ob6+Xq/c0tISxcXF6NGjh4lqxpj54H8YjAGwt7fHyy+/rHfx28LCAuPGjeNkwdh/cMJg7D9mzpypd+GbiDBz5kwT1ogx88KnpBj7j+rqavTo0QO1tbUAAJlMhtLSUigUChPXjDHzwP8wGPsPuVyO4OBgSCQSSCQSBAcHc7Jg7CGcMBh7yJtvvon6+nrU19fjzTffNHV1GDMrlo8WFBYW4syZM6aoC2Mm19jYCLlcDiKCRqNBcnKyqavEmEkEBgbCzc1Nv5AekZSURAB44oknnnh6iqekpKRH0wM1+4ehxdfCWWebOnUqAODAgQMmrom+7777DiKRCL///e9NXRUkJydj+vTp3B9Zl2rtYZytJgzGnlajRo0ydRUYM0ucMBh7REvPlGKM8SgpxhhjAnHCYIwxJggnDMYYY4JwwmCMMSYIJwwj27BhA3r16gWRSIRPP/1UV3706FHY2tri73//u9FjRkdHw9/fHyqVClKpFL6+vli6dCmqqqpaXWfevHlQKpUQiUTIysoSHOvFF1+ESCRqcbKxsTHG7hikMz/XJ8GJEycQGRmJQ4cOwcfHR9dWLb3/Y+zYsVAqlbCwsMCAAQPw008/maDGbRN6rNfX12Pt2rXw9fWFlZUV7OzsMHDgQFy/ft2s48XGxqJfv36wtraGQqFAv379sHLlSmg0GsF1OnLkCGJjY9HY2GhQbEFau3GPtV9+fj4BoG3btunK0tLSSKVS0ZEjR4weLygoiBITE6msrIw0Gg0lJSWRRCKhV155pc319u3bRwAoMzPToFho5UafcePGGVTvKVOm0JQpUwxa51Gd+bmag470x6ioKJowYQJpNBpdmVqtph49ehAASktLa7ZOeno6vf766+2ub2cTeqwHBwdT37596dy5c1RfX09FRUU0ceJEysnJMet448ePpw0bNlBJSQlVVlZScnIySSQSevnllw2qU0JCAgUFBVF5eblB8bXQyo17nDA6QUsJozONHz+eGhoa9MqmTZtGAOjmzZutrteehDFu3Di9LyCt0NBQOnnypPBKk3EShqlVV1dTQEBAp22/vf1x3bp15OfnRzU1NXrlarWavvrqKxKLxeTq6koVFRV68809YQg51vft20cikYiys7O7Xbzg4OBmbTZ16lQCQEVFRYLrREQUFhZGAQEBVF9fb3A9WksYfEqqGyIiHDhwANu3bwcApKWl6b34BwAcHR0BPHhkd2tau5uzLceOHYNSqdQrKygowIULFzBmzBiDt9fd7dq1CyUlJaauhp4rV65g5cqVWL16NWQyWbP5gYGBCA8Px61bt7B48WIT1LD9hBzr27Ztw3PPPYdBgwZ1u3gpKSnN2kz7PnntKSeh/X3VqlXIyspCQkJCh+ul1eGEkZCQAIVCAbFYjKFDh8LJyQkSiQQKhQLPPfccRo0aBXd3d8hkMtjZ2WHp0qV66//zn/+Ev78/bG1tIZPJMGjQIBw/fhwA8MUXX8DGxgYikQj29vZITU3Fjz/+CE9PT1hYWAh+muimTZsgk8nQq1cvzJ8/Hy4uLpDJZAgMDMT58+d1yxERNm7ciP79+0MqlcLe3h6TJk3CpUuXDFrmUadPn4aHhwdEIhG2bNkCANi6dSsUCgXkcjkOHz6MV199FSqVCm5ubtj3/9m797io6vx/4K8BhhlmYLgIAqEgFxVJ00pLMFfLltbcUFSM0r5aW1FaiCgRXggRScLQVaHyuq2mIuqi6yVbdclcjbRkJfyphCagIRe5Xwd4//4wZhkZ4AwMzKjv5+Ph41Gf8zmfz/t85nPmzXzOmTO7dqn2bWpqQmxsLAYPHgwzMzPY2trC1dUVsbGxmDFjRrt93rx5E2ZmZnB1dVXFHR8fj8GDB0MikcDS0hJhYWGCxq8zq1atwvz583XSlja6Oq5C5kNwcDBMTU3h4OCg6m/evHmQy+UQiUQoLi5GSEgIFi5ciJycHIhEInh4eAC4m1QVCgVWrlzZyyMC1fEREfz8/NqtExMTg0GDBmHz5s04fvx4u/U6m+/azOPIyEg4OzvDzMwMjz32GJKTk3VyvK3nekNDA77//nuMGDFCJ20bQn/Z2dmwsrKCi4uLoJhaWFtbY9y4cVi7dq3uHi1z70eOrnwE/uijjwgApaenU3V1NRUXF9Of/vQnAkCHDx+moqIiqq6upuDgYAJAGRkZqn1TUlIoKiqK7ty5QyUlJTR69Gjq06ePavulS5dIJpPR7NmzVWURERG0efNmrWIMCgoiuVxOly5dorq6OsrKyqJRo0aRhYWF6mNcZGQkmZqa0vbt26msrIwuXrxITzzxBNna2lJBQYHgOpqWpPLy8ggArV+/XlW2ZMkSAkAnTpyg8vJyKiwspLFjx5JcLqeGhgYiIlq5ciUZGxvTgQMHqKamhn788Ueyt7en8ePHt3us1dXVZGFhQcHBwWp9iUQi+vTTT6m0tJRqamooMTFR6yWpe+Xn55OXlxc1NTVpva8ulqS6Oq5C5sPMmTPJ3t5erb/4+HgCQEVFRURENG3aNHJ3d1erc+jQIbKwsKDo6OhuHRtR185HNzc38vLy0rjN3d2drl+/TkREZ86cISMjIxowYABVVVURUdslKSHzXch4L1q0iCQSCe3du5dKS0tp8eLFZGRkROfOndN2SNTcO9evX79OAGjEiBE0fvx4cnBwIIlEQp6enrRhwwZqbm6+L/praGig/Px8Wr9+PUkkEtq+fbvgmFqLiIjo0jmOnryG0ZIwKisrVWVffvklAVC76PPDDz8QANq9e3e7bcXGxhIAKiwsVJV98cUXBIB27NhBO3fupNDQUK3iI7r7BmFpaalWdu7cOQJAy5cvp5qaGjI3N6fAwEC1Oi0xR0dHC6pDpH3CaL1m2fIm/ssvvxAR0ahRo+ipp55S6+/tt98mIyMjqq+v13isS5YsoUGDBqmuNdTU1JBMJlO7cEbUtWsY93rvvfe6fK2mpxNGR+Pa2Xwg6nrC0CVtz8eqqioSiUT00ksvadzeOmEQES1cuJAA0HvvvUdE6glD6HzvbLxra2tJJpOptVNTU0MSiYTmzp0r+Ng0uXeuZ2ZmEgD64x//SP/5z3+opKSEysrK6MMPP1S9h9wP/dnb2xMA6tOnD/31r39VJV4hMbW2detWAkB///vfteq/vYTRY9cwTE1NAQCNjY2qMrFYDODuLWjtaanT+pawt99+G9OnT8c777yDPXv24JNPPtFJjCNHjoRMJsPly5eRlZWFqqoqjBw5Uq3OqFGjYGpqivT0dEF1uqtl3FrGqK6urs3HyaamJojF4jbrmMDdNdA9e/bg2LFjqmsNv/zyC2pqajBhwoRux9farVu3cPDgQcyZM0en7faEe8dVk9bz4X5VWFgIIoJMJhNUPyYmBoMHD0ZiYiJOnz6ttq078731eF+5cgU1NTUYOnSoaruZmRkcHBy6Ndaa5rpEIgEAPProo/Dx8YGNjQ0sLS2xfPlyWFpaqq77GXp/eXl5KCwsxM6dO/Hll1/i8ccf13itTFNMrbXMg9u3b3cpjnvp/aL34cOHMX78eNjZ2UEikbS5xtFi5cqVqKqq0vkFRolEgqKiIpSVlQGAxu8SWFlZobKyUlAdXXvxxRfx448/4sCBA6itrcX58+eRmpqKP//5z20Sxu7du7Fq1SqkpaVhwIABqvL8/HwAgJ2dnU5ji4uLw1tvvaXxwur9qmU+3K9afo+85Y2sM1KpFNu2bYNIJMIbb7yB2tpa1TZdzffq6moAwNKlS9W+t3Pjxo0Ob8roSHtz3dHREQBQXFysVt/U1BQuLi7Iycm5L/oTi8Wws7ODr68vdu/ejaysLMTGxgqKqTUzMzMA/5sX3aXXhJGbmwt/f384ODggPT0d5eXliIuLa1NPqVRi/vz5SEhIwNmzZxETE6OT/pVKJcrKytCvXz9YWVkBgMaTQJs6uhYVFYXnnnsOc+bMgUKhwNSpUzFjxgxs2rRJrd769euxY8cOnDx5Eo888ojatpY39Pr6ep3FVVBQgJ07d2Lu3Lk6a1PfWs+H+1XLG4Q2X9ry9vZGaGgosrOzsWLFClW5ruZ7yx8qa9asAd1dBlf9O3v2rOA4W3Q0183NzTFw4EBcunSpzX6NjY2wtLQ0+P7u5eHhAWNjY2RlZQmKqbWGhgYA/5sX3aXXhJGZmQmlUom5c+fCzc0NUqlU462e77//Pt566y0sWLAAoaGhWLFiRZcm2r3S0tJARBg9ejSGDh0Kc3NznD9/Xq1Oeno6Ghoa8OSTTwqqo2tZWVnIyclBUVERlEolcnNzkZSUBGtrawB372IJDw9HZmYmUlNTNf41OHToUBgZGeHbb7/VWVxxcXGYNWsWbGxsdNamvrWeDwBgYmLS4RKWIWp5ykB5eblW+61YsQKenp64cOGCqkxX873lLkltniigiZC5DgAvv/wyLly4gGvXrqnKampqcOPGDa1ufe3t/kpKSjTe+ZmdnY2mpib0799fcEwtWuaBvb294Dg6oteE4ezsDODu4wvq6uqQnZ3dZl00MTERTk5OmDp1KgAgNjYWXl5emDlzptrX5YVobm5GaWkpGhsbcfHiRYSEhMDZ2Rlz5syBVCrFwoULsX//fuzYsQMVFRXIzMzEu+++C0dHRwQFBQmqo2vvvfcenJ2d233Mx6VLl/DJJ59g06ZNEIvFbR7XsXr1atjZ2WHatGnYu3cvtmzZgoqKCly8eLHL66u3b9/G1q1bsWDBgu4cmt51NB+Au3/Z3blzB6mpqVAqlSgqKsKNGzfU2rCxscGtW7fw66+/orKyEkqlEkePHtXbbbUymQxubm6qZUihWpamWi9z6mq+S6VSvP7669i1axeSkpJQUVGBpqYm5Ofn47fffgMABAYGwt7evsPHkQiZ6wAQGhoKFxcXzJkzB7m5uSgpKUF4eDhqa2vx4YcfGmx/crkc33zzDU6ePImKigoolUpcuHABs2fPhlwuR2hoqOCYWrTMA118RwRA92+rXbt2LclkMgJAAwYMoO+++45WrVpFlpaWBIDs7e3pq6++ot27d6uu/FtbW9OuXbuIiCg8PJxsbGzIysqKAgICaMOGDQSA3N3dacSIESQSicjGxobOnDlDREQLFiwgIyMjAkCWlpZ0/vx5QXEGBQWRWCwmJycnMjExIYVCQVOmTKGcnBxVnebmZoqPj6eBAweSWCwma2tr8vf3pytXrgiu8+mnn6qOUy6X09SpU2n9+vXk4OBAAEgmk5Gfnx8lJiaqxm3gwIGUk5NDGzduJIVCQQDIxcWFrl69SidPnlQ9yqHln1gspiFDhtC+fftUd2m09y8+Pp6IiCorK+nNN9+kPn36kLm5OT3zzDMUGRlJAKhfv3703//+V/BrHhoaSrNmzRJcvz3dvUuqO+MqZD6UlJTQs88+S1KplFxdXen999+nsLAwAkAeHh6Um5tLP/30E7m4uJCZmRk988wzVFBQQEeOHCELCwuKiYnp9hh15a7F4OBgEovFVFNToyrbv38/ubu7EwCytbVV3RV1r7CwMLXbajub70LHu76+nsLDw8nZ2ZlMTEzIzs6Opk2bRllZWUR09xvOACgyMrLd4xI614nu3j33yiuvkLW1NUkkEnrqqafo6NGjqu2G2B8RkZ+fH7m6upK5uTlJJBJyd3enwMBA1d2m2sREdPdb4U5OTlrf3ouH/dEgQUFBZGNjo+8wtJaYmEghISFqZfX19bRgwQKSSCRqbwr3G30+GuR+mQ9dOR+zs7PJxMSkw3v3DU1TUxONHTuWtmzZwv3pSHFxMUmlUlq9erXW+7aXMPR+l1Rv6pGnN/aggoICBAcH4y9/+YtauampKZydnaFUKu+7NXZDcr/NB6E8PDwQHR2N6OjoDp9YbCiampqQmpqKyspKBAYGcn86EhUVhREjRiA4OFhnbd7XCePy5cvtPmq79b/efJF0yczMDGKxGFu2bMHt27ehVCpx69YtbN68GZGRkQgMDIRCoeh2Pw/6OD6MIiIiEBAQgMDAQK0vgPe2tLQ07Nu3D0ePHhX8/RHur2MJCQnIyMjAkSNHVN9t04l7P3I8iEtSERERZGpqqrrOkpKSou+QBDt16hQ9//zzpFAoyNjYmCwtLcnHx4cSExO79BRKQ6KvJan7aT5093w8duwYhYeH6zAiZuhSU1MpNja2zRNttYF2lqREv29U2bNnD15++WXdPayKsXYEBAQAAFJSUvQcieHi85Hpg0gkQnJycpsHnN7XS1KMMcZ6DycMxhhjgnDCYIwxJggnDMYYY4KYtLeh5YIkYz3l+++/B8BzrSMtj3bgMWKGgD9hMMYYE6TdTxh8qyPraXxbbedabqvlMWK9SdNTwwH+hMEYY0wgThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTJBeSxj79u2Dm5ub6lHZy5Yt67B+QkICRCIRjIyM4OnpiVOnTvVIHCKRCGKxGE5OTpg5cyb+3//7fzrpp8Xq1atVv7P8+eefq8qPHDkCS0tL/POf/9RpfwAQHR0NLy8vKBQKSCQSeHh44IMPPlD9NoKmMRCJRDA1NUXfvn0xfvx4xMfHo7S0VOexsd51/PhxREREtHnNX3vttTZ1fX19YWFhAWNjYzz66KMd/pyovjU3N2PNmjXw8fFpsy0uLg6enp4wMzODXC6Hp6cnli1bpvVPOuujPyFtdXZ+Hzx4EHFxcT3zey/3Pr62px9v3vIzkQ4ODtTQ0KCxTmNjI7m4uBAAmjBhQo/FYWlpSUREVVVVdPDgQXJ2diZzc3O6fPmyTvvKzs4mAPTZZ5+pyg4dOkQKhYIOHjyo076IiMaNG0eJiYlUUlJCFRUVlJycTGKxmP70pz+p1Ws9Bs3NzVRaWkr//ve/ac6cOSQSicjR0ZHOnTun8/ha6PMX9+4X3TkfIyMj6aWXXqKKigpVmbu7u+onfw8dOtRmn6NHj6r9RKshunr1Ko0ZM4YA0PDhw9tsnzRpEq1evZoKCwupsrKS9uzZQ2KxmP74xz8afH9C2hJyfq9du5bGjRtHpaWlXTpmGMpPtLq7u9OTTz5JAGjPnj0a6yQnJ5OPj0+vJYwW//jHPwgAzZs3T6d9aUoYPWnSpEltnoU/Y8YMAkC5ubmqMk1j0CIlJYWMjIyob9++VFZW1iNx6jNh1NTUkLe3t8G33dXz8eOPP6ZBgwZRbW2tWrm7uzt99dVXZGRkRE5OTm1eW0NPGBkZGTR16lTasWMHjRgxQuMbuL+/f5vjDggIIAB069Ytg+5PSFtCz+/g4GDy9vbu0u/mtJcw9HINY+7cuQCAzz77TOP2hIQELFy4sDdDAgA89dRTAICff/651/vuDiJCSkoKNm7cCAA4dOgQjI2N1erY2toCAGpqagS1OX36dMyZMweFhYVqS2kPii1btqCwsPC+a1uIX375BcuWLcPy5cshlUrbbPfx8UFISAhu3ryJRYsW6SHCrhs+fDj27duHmTNnQiKRaKyzf//+Nsft5OQEAFr/ZG1v9yekLaHnd1RUFDIyMrB27VqtYuiIXhLGc889hyFDhuDf//43rly5orbtP//5D2pqauDr66tx3++++w5eXl6wtLSEVCrFsGHDcOzYMQDA3/72N5ibm0MkEsHa2hqpqak4f/48XFxcYGxsjFdffbXDuBobGwFANTGICAkJCRgyZAgkEgmsra0xZcoUXL58WbWPkDr3On36NJydnSESibBhwwYAQFJSEuRyOWQyGQ4cOICJEydCoVCgX79+2LVrl2rfpqYmxMbGYvDgwTAzM4OtrS1cXV0RGxvb5sdOWrt58ybMzMzg6ura4Ri0NmfOHADA0aNHBe/T0zob7+DgYJiamsLBwUG1z7x58yCXyyESiVBcXIyQkBAsXLgQOTk5EIlE8PDwwLp16yCVStG3b1+88847cHR0hFQqhY+PD9LT07vVNgB8/fXXUCgUWLlyZY+P0bp160BE8PPza7dOTEwMBg0ahM2bN+P48ePt1utsvLWZt5GRkXB2doaZmRkee+wxJCcn6+6gO5GdnQ0rKyu4uLjcd/0JaUvT+W1tbY1x48Zh7dq1uvsBrns/cvTGktT169fpr3/9KwGgkJAQte3+/v60bds2qqys1LgklZKSQlFRUXTnzh0qKSmh0aNHU58+fVTbL126RDKZjGbPnq0qi4iIoM2bN7eJ497lmO3btxMACgsLI6K7a8Cmpqa0fft2Kisro4sXL9ITTzxBtra2VFBQILiOpiWpvLw8AkDr169XlS1ZsoQA0IkTJ6i8vJwKCwtp7NixJJfLVdd7Vq5cScbGxnTgwAGqqamhH3/8kezt7Wn8+PHtjnl1dTVZWFhQcHBwp2PQWkVFBQGg/v37t1unO7qyJCVkvGfOnEn29vZq+8XHxxMAKioqIiKiadOmkbu7u1qdoKAgksvldOnSJaqrq6OsrCwaNWoUWVhYqD7qd7XtQ4cOkYWFBUVHR2t1vF05H93c3MjLy0vjtpbzj4jozJkzZGRkRAMGDKCqqioiarskJWS8hczbRYsWkUQiob1791JpaSktXryYjIyMunWN7Omnn9a4RNSioaGB8vPzaf369SSRSGj79u1d7qu3+9OmrfbOb6K7730A6MKFC1r1D0O6hnH9+nUqKysjuVxO1tbWVFNTQ0REOTk51K9fP6qvr283YdwrNjaWAFBhYaGq7IsvviAAtGPHDtq5cyeFhoZqjKP1Re+9e/eSvb099e3bl/Lz86mmpobMzc0pMDBQbb8ffviBAFB0dLSgOkTaJ4zWa5iJiYkEgH755RciIho1ahQ99dRTav29/fbbZGRkRPX19RrHaMmSJTRo0CC1i5/3jkF7RCIRWVlZdVinq7RNGELHuzsJ497xOHfuHAGg5cuXd6vtrtL2fKyqqiKRSEQvvfSSxu2tEwYR0cKFCwkAvffee0SknjCEjndn87a2tpZkMplaOzU1NSSRSGju3LmCj+1enb2B29vbEwDq06cP/fWvf233JhtD7E+btto7v4mItm7dSgDo73//u1b9t5cw9PY9DEtLS7z66qsoLS3F7t27AQBr1qzB3LlzYWpqKrgdsVgMAGq3kL399tuYPn063nnnHezZsweffPKJxn3Ly8shEolgaWmJ+fPn48UXX8QPP/wAJycnZGVloaqqCiNHjlTbZ9SoUTA1NUV6erqgOt3VMhZKpRIAUFdX1+bjZVNTE8RicZt1TeDumuiePXtw7NgxWFhYaNV3dXU1iAgKhaKL0etWb4z3vUaOHAmZTNbhEqMhKSwsBBFBJpMJqh8TE4PBgwcjMTERp0+fVtvWnfFuPW+vXLmCmpoaDB06VLXdzMwMDg4OPTqueXl5KCwsxM6dO/Hll1/i8ccf79FrS7rsT2hbnZ3fLfPg9u3bXYrjXnr94l7Lxe/PP/8cZWVlSElJwTvvvNPhPocPH8b48eNhZ2cHiUSCDz74QGO9lStXoqqqqsMXzNLSEkSExsZG5OfnY+vWrap1wrKyMgCAubl5m/2srKxQWVkpqI6uvfjii/jxxx9x4MAB1NbW4vz580hNTcWf//znNglj9+7dWLVqFdLS0jBgwACt+7p69SoAwNPTUxehd5s+xhu4e02rqKioR9rWtbq6OgBo9wLtvaRSKbZt2waRSIQ33ngDtbW1qm26Gu/q6moAwNKlS9W+93Pjxg3BN2F0hVgshp2dHXx9fbF7925kZWUhNjb2vuhPSFtCzm8zMzMA/5sX3aXXhDFixAiMHj0aP/zwA4KCghAQEABra+t26+fm5sLf3x8ODg5IT09HeXk54uLi2tRTKpWYP38+EhIScPbsWcTExGgdm5WVFQBoPCnKysrQr18/QXV0LSoqCs899xzmzJkDhUKBqVOnYsaMGdi0aZNavfXr12PHjh04efIkHnnkkS719fXXXwMAJk6c2O24dUEf461UKnus7Z7Q8gahzZe2vL29ERoaiuzsbKxYsUJVrqvxtrOzA3B3BYHuLoOr/p09e1ZwnN3h4eEBY2NjZGVl3Xf9aWpL6Pnd0NAA4H/zorv0/miQlk8Ze/fuxYIFCzqsm5mZCaVSiblz58LNzQ1SqVTjc9vff/99vPXWW1iwYAFCQ0OxYsUKrSfm0KFDYW5ujvPnz6uVp6eno6GhAU8++aSgOrqWlZWFnJwcFBUVQalUIjc3F0lJSapES0QIDw9HZmYmUlNTNf51KERBQQHWrFmDfv364Y033tDlIXSZ0PE2MTFRLeF1V1paGogIo0eP1nnbPaHlqQLl5eVa7bdixQp4enriwoULqjJdze/+/ftDKpUiIyNDq5i6oqSkROPdkNnZ2WhqakL//v0Ntj8hbWl7frfMA3t7e8FxdETvCWPGjBmwtbWFv78/3NzcOqzr7OwM4O7jDurq6pCdnd1mHTUxMRFOTk6YOnUqACA2NhZeXl6YOXOmVl/Vl0qlWLhwIfbv348dO3agoqICmZmZePfdd+Ho6IigoCBBdXTtvffeg7Ozc7v3d1+6dAmffPIJNm3aBLFY3ObxH6tXr1arT0SoqqpCc3MziAhFRUVITk7GmDFjYGxsjNTUVIO5hiF0vD08PHDnzh2kpqZCqVSiqKgIN27cUGvLxsYGt27dwq+//orKykpVEmhubkZpaSkaGxtx8eJFhISEwNnZWXWLcVfbPnr0aK/cViuTyeDm5qb6aVehWpamWi9r6mp+S6VSvP7669i1axeSkpJQUVGBpqYm5Ofn47fffgMABAYGwt7evtuPI5HL5fjmm29w8uRJVFRUQKlU4sKFC5g9ezbkcjlCQ0MNtj8hbWl7frfMg2HDhnXrOFXuvQreU3dJ7d+/X/VYEFtbW9VdGUREH3zwAZ05c0Zymx6QAAAgAElEQVT1/0uXLiUHBwcCQEZGRuTl5UXfffcdERGFh4eTjY0NWVlZUUBAAG3YsIEAkLu7O40YMYJEIhHZ2Nio2luwYAEZGRkRALK0tKR169bRoEGDCAABIEdHRwoICNAYc3NzM8XHx9PAgQNJLBaTtbU1+fv705UrVwTX+fTTT1V3PMjlcpo6dSqtX79edXwymYz8/PwoMTGRZDIZAaCBAwdSTk4Obdy4kRQKBQEgFxcXunr1Kp08eVL1aIeWf2KxmIYMGUL79u2jzMxMtW33/ouPj6eDBw/SY489RjKZjExNTVXj03JH1FNPPUXR0dFUUlKi83nQWlduqxXympSUlNCzzz5LUqmUXF1d6f3336ewsDACQB4eHpSbm0s//fQTubi4kJmZGT3zzDNUUFBAQUFBJBaLycnJiUxMTEihUNCUKVMoJyen220fOXKELCwsKCYmRqvj7cr5GBwcTGKxWHX3IVHH519rYWFharfVdjbeQudtfX09hYeHk7OzM5mYmJCdnR1NmzaNsrKyiOju7fQAKDIyssNjO3v2LI0ZM4YcHR1Vc9rBwYF8fHzo22+/JSIiPz8/cnV1JXNzc5JIJOTu7k6BgYGUmZmpasdQ++usLSHnd2uTJk0iJycnam5u7rDfe8FQbqtl3ZOYmNjmuyv19fW0YMECkkgkam8Shs7QniUVFBRENjY2+g5DTVfOx+zsbDIxMen29w56U1NTE40dO5a2bNnC/elIcXExSaVSWr16tdb7tpcw9L4kxYQrKChAcHAw/vKXv6iVm5qawtnZGUql0qDX1+8HPfKEz17m4eGB6OhoREdHa/1oCn1oampCamoqKisrERgYyP3pSFRUFEaMGIHg4GCdtckJ4z5iZmYGsViMLVu24Pbt21Aqlbh16xY2b96MyMhIBAYGGsz1BqZfERERCAgIQGBgoNYXwHtbWloa9u3bh6NHjwr+/gj317GEhARkZGTgyJEjqu+q6cS9Hzl4ScqwnTp1ip5//nlSKBRkbGxMlpaW5OPjQ4mJiV16KqU+GdKSVEREBJmamhIAGjBgAKWkpOg7JCLq/vl47NgxCg8P12FEzNClpqZSbGxsmyfaagPtLEmZ6C71sN4wduxY/Otf/9J3GA+c2NjYHv1Sl774+vq2+yBP9mCaPHkyJk+e3CNt85IUY4wxQThhMMYYE4QTBmOMMUE4YTDGGBOEEwZjjDFB2r1LStND/RjrCTzXOsdjxAyB6Pd7blXy8/Nx5swZfcXDmN6tWbMGADp9ejJjDzIfH582j7BvkzAYe9jNmDEDALBnzx49R8KYYeFrGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTBBOGIwxxgThhMEYY0wQE30HwJg+FRcXo6KiQq2suroaAHDt2jW1coVCAVtb216LjTFDIyIi0ncQjOnLtm3b8MYbbwiqu3XrVrz++us9HBFjhosTBnuolZeXw87ODkqlssN6YrEYRUVFsLS07KXIGDM8fA2DPdQsLS3x4osvwsSk/dVZExMTTJo0iZMFe+hxwmAPvVmzZqGpqand7c3NzZg1a1YvRsSYYeIlKfbQq6urg62trepi971kMhmKi4thZmbWy5ExZlj4EwZ76EmlUkydOhVisbjNNrFYjOnTp3OyYAycMBgDALz66qsaL3wrlUq8+uqreoiIMcPDS1KMAWhsbIS9vT3u3LmjVm5lZYWioqIOL4oz9rDgTxiM4e6dUK+88oraspRYLMasWbM4WTD2O04YjP3ulVdeUVuWUiqVeOWVV/QYEWOGhZekGPsdEaF///64efMmAMDR0RE3b96ESCTSc2SMGQb+hMHY70QiEV577TWYmprC1NQUs2fP5mTBWCv8CYOxVi5evIjhw4er/nvYsGF6jogxw2GwV/POnj2LhIQEfYfBHkLm5uYAgOjoaD1Hwh5GoaGh8Pb21ncYGhnsklReXh727t2r7zAeSN9//z2+//57fYdhsFxcXGBtbc3zj/W6vXv3Ii8vT99htMtgP2G0SElJ0XcID5yAgAAAPLbtuXbtGg4fPozg4GAeI9arDP2amcF+wmBMX9zc3GBvb6/vMBgzOJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggD1zCqK+vx/z58+Hg4ACZTIbnn38effv2hUgkwueff67v8Lps/PjxEIlEGv+1fG+gNx05cgSWlpb45z//2et93w+OHz+OiIgI7Nu3D25ubqrX6rXXXmtT19fXFxYWFjA2Nsajjz6Kn376SQ8RC9Pc3Iw1a9bAx8enzba4uDh4enrCzMwMcrkcnp6eWLZsGSoqKgy+PyFtRUdHw8vLCwqFAhKJBB4eHvjggw9QVVUFADh48CDi4uI6/PXG+x4ZqOTkZOpKeCtXrqRBgwZRaWkpffHFF5SSkkLZ2dkEgD777LMeiLR3jBs3jgBo/PfCCy9o1db06dNp+vTp3Yrn0KFDpFAo6ODBg91qx1B1df4REUVGRtJLL71EFRUVqjJ3d3fq06cPAaBDhw612efo0aM0efLkLsfbG65evUpjxowhADR8+PA22ydNmkSrV6+mwsJCqqyspD179pBYLKY//vGPBt+fkLbGjRtHiYmJVFJSQhUVFZScnExisZj+9Kc/qeqsXbuWxo0bR6WlpV06ZgCUnJzcpX17wwOXMEaNGkWvvvqqWllXEkZNTQ15e3t3WtZbXnjhBbU3oBZBQUF04sQJrdrSRcLQt55+Lbo6/z7++GMaNGgQ1dbWqpW7u7vTV199RUZGRuTk5ERlZWVq2w09YWRkZNDUqVNpx44dNGLECI1v4P7+/m2OOyAggADQrVu3DLo/IW1NmjSJGhsb1erMmDGDAFBubq6qLDg4mLy9vUmpVGoVA5HhJ4wHbkkqPz9f409tamvLli0oLCzstKy3fP3117CwsFAry8vLw88//4znnntOLzHpkz5fi/b88ssvWLZsGZYvXw6pVNpmu4+PD0JCQnDz5k0sWrRIDxF23fDhw7Fv3z7MnDkTEolEY539+/e3OW4nJycAUC3bGGp/Qto6dOgQjI2N1erY2toCAGpqalRlUVFRyMjIwNq1a7WK4X7wwCSMf/3rX/Dw8MBvv/2GL7/8stO1/e+++w5eXl6wtLSEVCrFsGHDcOzYMQBASEgIFi5ciJycHIhEInh4eGgsA4CmpiZERkbC2dkZZmZmeOyxx5CcnAwASEpKglwuh0wmw4EDBzBx4kQoFAr069cPu3bt6vYxr1q1CvPnz+92O9o6ffo0nJ2dIRKJsGHDBgDCjnXdunWQSqXo27cv3nnnHTg6OkIqlcLHxwfp6ekAgODgYJiamsLBwUHV37x58yCXyyESiVBcXNzua/H1119DoVBg5cqVvTwiUB0fEcHPz6/dOjExMRg0aBA2b96M48ePt1uPiJCQkIAhQ4ZAIpHA2toaU6ZMweXLlwEIn1sdzc/ekJ2dDSsrK7i4uNx3/Qlp6+bNmzAzM4Orq6uqzNraGuPGjcPatWtBD9qzXfX8CaddXV0SsLe3p9mzZ6uVaVqSSklJoaioKLpz5w6VlJTQ6NGjqU+fPqrt06ZNI3d3d7V2NJUtWrSIJBIJ7d27l0pLS2nx4sVkZGRE586dIyKiJUuWEAA6ceIElZeXU2FhIY0dO5bkcjk1NDRofXwt8vPzycvLi5qamrTeVxdLUnl5eQSA1q9fryoTcqxBQUEkl8vp0qVLVFdXR1lZWTRq1CiysLBQfayfOXMm2dvbq/UXHx9PAKioqIiINL8Whw4dIgsLC4qOju7WsRF1bf65ubmRl5eXxm3u7u50/fp1IiI6c+YMGRkZ0YABA6iqqoqI2i5JRUZGkqmpKW3fvp3Kysro4sWL9MQTT5CtrS0VFBQQkbDx7mx+dsXTTz+tcYmoRUNDA+Xn59P69etJIpHQ9u3bu9xXb/enTVvV1dVkYWFBwcHBbbZFREQQALpw4YJW/YOXpAzT9OnT8dFHH8Ha2ho2Njbw8/NDSUkJioqKBLdRV1eHpKQk+Pv7Y9q0abCyssLSpUshFouxbds2tbo+Pj5QKBSws7NDYGAgqqurkZub2+X4V61ahffffx9GRob3EnZ2rCYmJqq/nL28vJCUlITKyso2Y6atSZMmoaKiAsuWLevuIWituroa169fh7u7e6d1vb29sWDBAvz666/48MMP22yvra1FQkICpk6dilmzZsHS0hLDhg3D559/juLiYmzcuFGtfnvjrc381KX+/fujX79+iIqKwieffIKXX365x/rSdX/atBUbGwtHR0fExMS02TZw4EAAQGZmZpdjMUSG926jJy3XPbS5Je7KlSuoqanB0KFDVWVmZmZwcHBQLR1oYmpqCgBqPweqjVu3buHgwYOYM2dOl/bvTUKOdeTIkZDJZB2OmaErLCwEEUEmkwmqHxMTg8GDByMxMRGnT59W25aVlYWqqiqMHDlSrXzUqFEwNTVVLd9p0nq8uzo/uysvLw+FhYXYuXMnvvzySzz++OM9er1Jl/0JbWv//v3Ys2cPjh071ubaIgDVPLh9+3aX4jBUD23COHz4MMaPHw87OztIJBJ88MEHWrdRXV0NAFi6dKna9yJu3LihdhFM1+Li4vDWW29pvLB6v5JIJFp9ujM0dXV1ANDuBdp7SaVSbNu2DSKRCG+88QZqa2tV28rKygBA4zU4KysrVFZWCupDX/NTLBbDzs4Ovr6+2L17N7KyshAbG3tf9Cekrd27d2PVqlVIS0vDgAEDNLZjZmYG4H/z4kHxUCaM3Nxc+Pv7w8HBAenp6SgvL0dcXJzW7djZ2QEA1qxZA7p7i7Lq39mzZ3UdNgCgoKAAO3fuxNy5c3ukfX1QKpUoKytDv3799B1Kl7W8QWjzCdXb2xuhoaHIzs7GihUrVOVWVlYAoDExaDNO+pif9/Lw8ICxsTGysrLuu/40tbV+/Xrs2LEDJ0+exCOPPNLuvg0NDQD+Ny8eFA9lwsjMzIRSqcTcuXPh5uYGqVTapefQ9+/fH1KpFBkZGT0QpWZxcXGYNWsWbGxseq3PnpaWlgYiwujRowHcvcbR1eU6fWl5mkB5eblW+61YsQKenp64cOGCqmzo0KEwNzfH+fPn1eqmp6ejoaEBTz75pKC2e3N+lpSU4NVXX21Tnp2djaamJvTv399g+xPSFhEhPDwcmZmZSE1N7fTpCi3z4EF7TP5DmTCcnZ0B3H18Q11dHbKzs9usC9vY2ODWrVv49ddfUVlZCaVS2abM2NgYr7/+Onbt2oWkpCRUVFSgqakJ+fn5+O2333Qe9+3bt7F161YsWLBA5233pubmZpSWlqKxsREXL15ESEgInJ2dVddkPDw8cOfOHaSmpkKpVKKoqAg3btxQa0PT63P06FG93VYrk8ng5uaG/Px8rfZrWZpqfX+/VCrFwoULsX//fuzYsQMVFRXIzMzEu+++C0dHRwQFBQluu7P5GRgYCHt7+24/jkQul+Obb77ByZMnUVFRAaVSiQsXLmD27NmQy+UIDQ012P6EtHXp0iV88skn2LRpE8RicZvH86xevVqtzZZ58MD9Jrye7s7qlLa3Nf7666/0+OOPEwAyMTGhJ554gvbu3Uuffvop2dvbEwCSy+U0depUIiIKDw8nGxsbsrKyooCAANqwYQMBIHd3d8rNzaWffvqJXFxcyMzMjJ555hkqKCjQWFZfX0/h4eHk7OxMJiYmZGdnR9OmTaOsrCxKTEwkmUxGAGjgwIGUk5NDGzduJIVCQQDIxcWFrl69KvgYQ0NDadasWVqP5b26e1vt+vXrycHBgQCQTCYjPz8/wccaFBREYrGYnJycyMTEhBQKBU2ZMoVycnJU7ZeUlNCzzz5LUqmUXF1d6f3336ewsDACQB4eHu2+PkeOHCELCwuKiYnp9hh15bba4OBgEovFVFNToyrbv38/ubu7EwCytbWl9957T+O+YWFharfVNjc3U3x8PA0cOJDEYjFZW1uTv78/XblyhYhI8Hh3ND+J7n7DGQBFRkZ2eGxnz56lMWPGkKOjo+qRNA4ODuTj40PffvstERH5+fmRq6srmZubk0QiIXd3dwoMDKTMzExVO4baX2dtZWZmtvtoHgAUHx+v1t6kSZPIycmJmpubO+z3XjDw22ofmITBhNPno0GCgoLIxsZGL31royvzLzs7m0xMTLr9vYPe1NTURGPHjqUtW7ZwfzpSXFxMUqmUVq9erfW+hp4wHsolKaZfD+rTPD08PBAdHY3o6GitH02hD01NTUhNTUVlZSUCAwO5Px2JiorCiBEjEBwc3Gt99hZOGHp0+fLldh9Z3vpfb0521j0REREICAhAYGCg1hfAe1taWhr27duHo0ePCv7+CPfXsYSEBGRkZODIkSM6eaadoTHRdwAPM09PzwfvWTMdWLx4MbZt24aGhga4uroiPj4e06dP13dYOrdy5Up88803+Pjjj7Fq1Sp9h9OuCRMmYMKECdyfjhw4cAD19fVIS0tr85DCBwUnDNZrYmNje/QLXIbE19cXvr6++g6D9aLJkydj8uTJ+g6jR/GSFGOMMUE4YTDGGBOEEwZjjDFBOGEwxhgThBMGY4wxQQz+LqmuPBSQCcNj2zkeI8b+x+ATRm/+/vDDYs2aNQBw3z/EsCedPXsWa9eu5fnHelVP/zphdxl8wpgxY4a+Q3jgpKSkAOCx7czatWt5jFivMvSEwdcwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJ8lAkjH379sHNza3D35wYMGAAVq9ejb59+0IkEuHzzz/Xd9iMAbj72/MRERFt5vFrr73Wpq6vry8sLCxgbGyMRx99tNu/nd0TYmJiNJ6DQ4cObXefuro6eHp6YunSpWrlp0+fxpgxYyCTyeDo6Ijw8HDU19er1dm5cydGjRoFCwsLuLi44PXXX0dBQQEA4ODBg4iLi3tgf9RL1x6KhDFt2jRcu3YN7u7usLS0BN39aVo0NjaipqYGt2/fhkwmw6JFi3DmzBl9h8uYykcffYR169Zh8eLFavO4T58+2LFjBw4fPqxW/5tvvkFKSgpeeuklZGVl4YknntBT5Lq1ZMkSXLlyRa0sKysLvr6+mDBhAoqKirB//35s3boV7777rqpOcnIyZs6ciYCAAOTn5+PAgQM4deoUJk6ciMbGRvj5+UEqlWLChAkoKyvr7cO67zwUCaM9xsbGMDMzQ9++fTFo0KAut1NbWwsfH59Oy1jPjsuDNuarVq3C7t27sWfPHlhYWKhtW7duHYyMjBAUFGTwv+ynyfbt21V/uLX8+/nnnzXWPXPmjMZtK1asgIODA5YvXw65XA5vb2+Eh4fjb3/7Gy5fvgwA+OKLL/DII48gLCwMlpaWGDFiBEJDQ5GRkYH09HQAwPz58zF8+HC8+OKLaGxs7LmDfgA81AmjtdTU1C7vu2XLFhQWFnZaxnp2XB6kMf/ll1+wbNkyLF++HFKptM12Hx8fhISE4ObNm1i0aJEeIuwdtbW1CAsLw9q1a9XKGxsbcfjwYYwbN07t8S0TJ04EEeHAgQMAgLy8PDg6OqrV6d+/PwDgxo0bqrKoqChkZGS06Yep44QhwHfffQcvLy9YWlpCKpVi2LBhOHbsGAAgJCQECxcuRE5ODkQiETw8PDSWAXd/lD4yMhLOzs4wMzPDY489pnr0RFJSEuRyOWQyGQ4cOICJEydCoVCgX79+2LVrl96OvQURISEhAUOGDIFEIoG1tTWmTJmi+ksuODgYpqamcHBwUO0zb948yOVyiEQiFBcXaxyXdevWQSqVom/fvnjnnXfg6OgIqVQKHx8f1V+AXW0bAL7++msoFAqsXLmyF0er+9atWwcigp+fX7t1YmJiMGjQIGzevBnHjx9vt15nr53QudfR/O0pS5Yswbx582BnZ6dWfu3aNVRVVcHZ2Vmt3N3dHQBw8eJFAICbm1ubPyJarl+4ubmpyqytrTFu3DisXbv2ofrZZK2RgUpOTiZdh+fu7k6WlpZqZSdOnKD4+HjV/2dnZxMA+uyzz1RlKSkpFBUVRXfu3KGSkhIaPXo09enTR7V92rRp5O7urtauprJFixaRRCKhvXv3UmlpKS1evJiMjIzo3LlzRES0ZMkSAkAnTpyg8vJyKiwspLFjx5JcLqeGhgadjcP06dNp+vTpWu0TGRlJpqamtH37diorK6OLFy/SE088Qba2tlRQUEBERDNnziR7e3u1/eLj4wkAFRUVEZHmcQkKCiK5XE6XLl2iuro6ysrKolGjRpGFhQXl5uZ2q+1Dhw6RhYUFRUdHa3W8PTH/tOHm5kZeXl4at7m7u9P169eJiOjMmTNkZGREAwYMoKqqKiIiOnr0KE2ePFlVX8hrJ2TudTZ/hVqxYgX169ePrKysSCwW04ABA2jy5Mn0ww8/qNU7ffo0+fn5ERFRUVERAaAlS5YQEdG3335LANTO3RZmZmY0YcIEIiJKS0sjsVhM69ato4qKCvr5559pyJAh9MILL7TZLyIiggDQhQsXtDoeXQJAycnJeuu/Mw/dJ4zy8nK1OzOE/Ej89OnT8dFHH8Ha2ho2Njbw8/NDSUkJioqKBPdbV1eHpKQk+Pv7Y9q0abCyssLSpUshFouxbds2tbo+Pj5QKBSws7NDYGAgqqurkZubq/Wx6kptbS0SEhIwdepUzJo1C5aWlhg2bBg+//xzFBcXY+PGjd3uw8TERPUXsJeXF5KSklBZWdlmbLQ1adIkVFRUYNmyZd2OsbdUV1fj+vXrqr+WO+Lt7Y0FCxbg119/xYcffthmu7avXXtzT5v525nZs2fj4MGDyMvLQ1VVFXbt2oXc3FyMGzcOWVlZqrhDQkKQlJSksY2WO6GMjY3bbBOLxaitrQUAjBs3DuHh4QgODoZCocDQoUNRWVmJzZs3t9lv4MCBAIDMzEytjudh8tAljNZ3SRER/v3vf2vdhlgsBgCtbsW7cuUKampq1G4dNDMzg4ODg2ppQBNTU1MAgFKp1DpOXcnKykJVVRVGjhypVj5q1CiYmpqqlo50aeTIkZDJZB2OzYOqsLAQRASZTCaofkxMDAYPHozExEScPn1abVt3XrvWc6+r81eT/v374/HHH4e5uTlMTU0xevRobNu2DbW1tUhMTAQALF68GG+//TacnJw0ttFyXUfTReqGhgaYmZkBuLuktXHjRpw4cQJVVVW4du0afHx84O3tjby8PLX9Wsb79u3bWh3Pw+ShSxj3Gj9+fKcXDQ8fPozx48fDzs4OEokEH3zwgdb9VFdXAwCWLl2q9gnnxo0bqKmp6VLsvaXldkNzc/M226ysrFBZWdkj/UokEq0+xT0o6urqANw9fiGkUim2bdsGkUiEN954Q/XXNaC7166n5++wYcNgbGyMq1ev4vTp08jMzMSbb77Zbv2W61kVFRVq5TU1Nairq4OjoyN+++03xMXF4e2338Zzzz0HuVwOV1dXbNq0Cbdu3UJ8fLzavi1JpmX8WVsPfcLoTG5uLvz9/eHg4ID09HSUl5cjLi5O63ZaLtqtWbOmze2EZ8+e1XXYOmVlZQUAGt9cysrK0K9fP533qVQqe6xtQ9fyxqXNJ1hvb2+EhoYiOzsbK1asUJXr6rXr6fnb3NyM5uZmSCQSbNmyBSdOnICRkZEqMbX0v3LlSohEIpSUlMDCwkLtTifg7t1lAPDYY48hOzsbTU1NeOSRR9TqKBQK2NjYqJa/WjQ0NAD43/iztjhhdCIzMxNKpRJz586Fm5sbpFJpl36FrX///pBKpcjIyOiBKHvW0KFDYW5ujvPnz6uVp6eno6GhAU8++SSAu9chdLV0lpaWBiLC6NGjdd62oWt52oC2369YsWIFPD09ceHCBVWZ0NeuM7qcvy+88EKbsnPnzoGI4O3tjW3btrVJSi2fNJcsWaKaFy+++CJOnTqF5uZmVTtHjx6FSCSCn5+fKhn+9ttvan1VVlbizp07qttrW7SMt729fbeP8UHFCaMTLbftHT9+HHV1dcjOzm6z7mtjY4Nbt27h119/RWVlJZRKZZsyY2NjvP7669i1axeSkpJQUVGBpqYm5Ofnt5nQhkYqlWLhwoXYv38/duzYgYqKCmRmZuLdd9+Fo6MjgoKCAAAeHh64c+cOUlNToVQqUVRU1OYvQE1jBdz9C7O0tBSNjY24ePEiQkJC4OzsjDlz5nSr7aNHj953t9XKZDK4ubkhPz9fq/1alqZaXwgW+toJabuz+RsYGAh7e/tOH0dy8+ZN7N69G2VlZVAqlTh79izefPNNODs7q31LuzPLli3D7du38dFHH6G6uhpnz55FfHw85syZg8GDB8PV1RXPPvssNm3ahFOnTqG2thZ5eXmqY/7LX/6i1l7LeA8bNkxwDA+dXr4rSzBd3tb4n//8hwYNGkQACAA5ODiobrtr7dNPPyV7e3sCQHK5nKZOnUpEROHh4WRjY0NWVlYUEBBAGzZsIADk7u5Oubm59NNPP5GLiwuZmZnRM888QwUFBRrL6uvrKTw8nJydncnExITs7Oxo2rRplJWVRYmJiSSTyQgADRw4kHJycmjjxo2kUCgIALm4uNDVq1d1Mh5dua22ubmZ4uPjaeDAgSQWi8na2pr8/f3pypUrqjolJSX07LPPklQqJVdXV3r//fcpLCyMAJCHh0e7YxUUFERisZicnJzIxMSEFAoFTZkyhXJycrrd9pEjR8jCwoJiYmK0Ol5931YbHBxMYrGYampqVGX79+8nd2zOTTwAACAASURBVHd3AkC2trb03nvvadw3LCxM7bbazl47oXOvo/lLROTv708AKDIyssNjW7hwIbm7u5NcLicTExPq168fvfXWW3Tr1q1297n3ttoW3377LT311FMkkUjI0dGRwsLCqK6uTrW9uLiYQkJCyMPDgyQSCZmbm9OYMWPoH//4R5s+Jk2aRE5OTtTc3Nxh/D0JBn5b7UORMJi6riSMnhQUFEQ2Njb6DkONvudfdnY2mZiY0Pbt2/UWg7aamppo7NixtGXLFn2HorXi4mKSSqW0evVqvcZh6AmDl6SYQeCnharz8PBAdHQ0oqOjUVVVpe9wOtXU1ITU1FRUVlYiMDBQ3+FoLSoqCiNGjEBwcLC+QzFonDAYM1AREREICAhAYGCgwT9gMC0tDfv27cPRo0cFf3/EUCQkJCAjIwNHjhxRfceKacYJg+nV4sWLsW3bNpSXl8PV1RV79+7Vd0gGZeXKlQgODsbHH3+s71A6NGHCBHz11Vdqz/u6Hxw4cAD19fVIS0uDtbW1vsMxeCb6DoA93GJjYxEbG6vvMAyar68vfH199R3GA2ny5MmYPHmyvsO4b/AnDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmiMFf9N6zZ4++Q3jgtDwCgce2fS0P1OMxYux/DD5hvPzyy/oO4YHFY9s5HiPG/kdExD9gy1hrM2bMAMCfLhi7F1/DYIwxJggnDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmCCcMxhhjgnDCYIwxJggnDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmCCcMxhhjgnDCYIwxJggnDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmCCcMxhhjgnDCYIwxJggnDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmCCcMxhhjgnDCYIwxJggnDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmCCcMxhhjgnDCYIwxJggnDMYYY4JwwmCMMSaIib4DYEyfTp06hbNnz6qVXb58GQAQFxenVu7t7Y0//OEPvRYbY4ZGRESk7yAY05cTJ07g+eefh1gshpGR5g/czc3NUCqVOH78OCZMmNDLETJmODhhsIdac3MzHBwcUFRU1GE9W1tbFBQUwNjYuJciY8zw8DUM9lAzMjLCzJkzYWpq2m4dU1NTzJo1i5MFe+hxwmAPvVdeeQUNDQ3tbm9oaMArr7zSixExZph4SYoxAAMGDMCNGzc0buvfvz9u3LgBkUjUy1ExZlj4EwZjAF577TWIxeI25WKxGHPmzOFkwRj4EwZjAO7eSjtkyBCN237++Wc8+uijvRwRY4aHP2EwBsDT0xOPPvpom08SXl5enCwY+x0nDMZ+93//939qd0KJxWLMnj1bjxExZlh4SYqx3+Xl5cHFxQUtp4RIJMK1a9cwYMAA/QbGmIHgTxiM/a5///54+umnYWRkBCMjIzz99NOcLBhrhRMGY6289tprEIlEMDIywmuvvabvcBgzKLwkxVgrxcXFcHBwAADcunULffv21XNEjBmO+yZh7NmzBy+//LK+w2CMMZ1KTk7GjBkz9B2GIPfd482Tk5P1HQLToZdffhkhISHw9vbWdygqp06dgkgkwtixY/UdCgBgzZo1AIAFCxboORKma/fbH8H3XcK4XzIxE+bll1+Gt7e3Qb2uEydOBABYWFjoOZK7UlJSAPDcfxBxwmDsPmcoiYIxQ8N3STHGGBOEEwZjjDFBOGEwxhgThBMGY4wxQThhdOLNN9+EhYUFRCIRMjIy9B2O1pqbm7FmzRr4+Pho3L5z506MGjUKFhYWcHFxweuvv46CgoJejrJ7jhw5AktLS/zzn//UdygG6fjx44iIiMC+ffvg5uYGkUgEkUik8Zvsvr6+sLCwgLGxMR599FH89NNPeoi4YzExMapjaP1v6NCh7e5TV1cHT09PLF26VK389OnTGDNmDGQyGRwdHREeHo76+nq1Oh2dIwcPHkRcXByampp0f6AGiBNGJzZv3oxNmzbpO4wuyc7Oxh/+8AeEhoaipqamzfbk5GTMnDkTAQEByM/Px4EDB3Dq1ClMnDgRjY2Neoi4a+6T757qxUcffYR169Zh8eLFmDZtGq5duwZ3d3f06dMHO3bswOHDh9Xqf/PNN0hJScFLL72ErKwsPPHEE3qKXLeWLFmCK1euqJVlZWXB19cXEyZMQFFREfbv34+tW7fi3XffVdXp7Bzx8/ODVCrFhAkTUFZW1tuH1es4YTyg/vvf/+LDDz/Eu+++ixEjRmis88UXX+CRRx5BWFgYLC0tMWLECISGhiIjIwPp6em9HHHXTZo0CeXl5XjppZf00n9tbW27n+D0adWqVdi9ezf27NnT5lbhdevWwcjICEFBQSgvL9dThF23fft2EJHav59//llj3TNnzmjctmLFCjg4OGD58uWQy+Xw9vZGeHg4/va3v+Hy5csAhJ0j8+fPx/Dhw/Hiiy/eV39odQUnDAHux5/nHD58OPbt24eZM2dCIpForJOXlwdHR0e14+vfvz8AtPv71qytLVu2oLCwUN9hqPnll1+wbNkyLF++HFKptM12Hx8fhISE4ObNm1i0aJEeIuwdtbW1CAsLw9q1a9XKGxsbcfjwYYwbN05t/k+cOBFEhAMHDgAQfo5ERUUhIyOjTT8PGk4Y9yAixMfHY/DgwZBIJLC0tERYWJhanaamJkRGRsLZ2RlmZmZ47LHHVI8sSUpKglwuh0wmw4EDBzBx4kQoFAr069cPu3btUrXx7bff4qmnnoJMJoNCocCwYcNQUVHRafu65Obm1uaNrmVt1s3NTef99YTTp0/D2dkZIpEIGzZsACDsNVi3bh2kUin69u2Ld955B46OjpBKpfDx8VH95RgcHAxTU1PVwwgBYN68eZDL5RCJRCguLkZISAgWLlyInJwciEQieHh4AAC+/vprKBQKrFy5spdHBKrjIyL4+fm1WycmJgaDBg3C5s2bcfz48XbrERESEhIwZMgQSCQSWFtbY8qUKaq/woXO+d6a160tWbIE8+bNg52dnVr5tWvXUFVVBWdnZ7Vyd3d3AMDFixcBCD9HrK2tMW7cOKxdu/bBXiKl+0RycjL1RrhLliwhkUhEn376KZWWllJNTQ0lJiYSALpw4QIRES1atIgkEgnt3buXSktLafHixWRkZETnzp1TtQGATpw4QeXl5VRYWEhjx44luVxODQ0NVFVVRQqFguLi4qi2tpYKCgpo6tSpVFRUJKh9bT399NM0fPjwNuVpaWkkFotp3bp1VFFRQT///DMNGTKEXnjhhS6OnvYAUHJycrfayMvLIwC0fv16VVlnrwERUVBQEMnlcrp06RLV1dVRVlYWjRo1iiwsLCg3N5eIiGbOnEn29vb/v727D4uqTP8A/h0YhmGAgUFeBUHekkTUSE1RN/35i828AhFNSmvNdUM3Q0L5EaKoCJiLoZcFeVksu6kpKC5WilvaJa2X5toVhkEqooCGyEsib8P7/fujnVknQA4wMIPen+viD855znPuec555p5zznPO0VhfUlISAVBvr+DgYHJ3d9co88UXX5C5uTnFxcUN6LMRES1cuJAWLlzYp2Xc3Nxo7Nix3c5zd3enmzdvEhHRuXPnyMDAgEaPHk0NDQ1ERJSTk0OBgYHq8rGxsSSRSGjfvn1UW1tL+fn55OvrS9bW1lRRUUFEwtpbW/v11q1bycnJiSwtLcnIyIhGjx5NgYGB9O9//1uj3NmzZykgIICIiKqqqggAxcTEEBFRbm4uAaCkpKQu9ZuYmNCcOXOIqG99JDo6WuN7Qght7P9DiRPGA5qamkgmk9Fzzz2nMf3gwYPqHUGpVJJMJqOQkBCN5YyNjenPf/4zEf238yiVSnUZVdK5fv06/fjjjwSAvvjiiy4xCKm/r3pKGEREGzZsIADqPycnJ7p161a/1tMfg50wetoGRL8mDAsLC426Ll68SABoy5YtRNT/hKFNfU0YDQ0NJBKJ6MUXX+x2/oMJg4ho7dq1BIBWr15NRJoJo6mpiczMzDT2RyKif//73wRAnRB7a29t7tdlZWX0/fffU319PbW0tND58+fpqaeeIhMTE/rxxx/VdU+aNIlu375NRF0TxpdffkkAKDk5uUv9crmc/Pz81P8L7SN//etfCQB98skngj/LcEsYfErqAdevX0dTUxPmzJnTY5mrV6+iqalJYwifiYkJ7O3t1Yfo3ZFIJACAtrY2uLm5wdbWFkuXLsXmzZtRUlIy4Pr7IyYmBnv37sXp06fR0NCAGzduwM/PD9OmTcOtW7e0ui598OA26MmkSZMgk8m03tZDqbKyEkQEmUwmqHx8fDzGjBmDlJQUnD17VmNeQUEBGhoaMGnSJI3pkydPhkQieejgiAfbW5v79ahRo/DUU0/BzMwMEokEU6dORXp6OpRKJVJSUgAA69evxxtvvAFHR8du61Bd1+nuInVraytMTEwA9K2PqNr77t27ffo8wwknjAfcvn0bALqc73xQY2MjAGDDhg0aY8BLS0u7HbraHRMTE3z99deYMWMGEhIS4ObmhpCQECiVSq3UL8SdO3ewfft2vPHGG/if//kfmJqawtXVFR999BHKy8uRlJSktXUNN8bGxqiqqtJ1GP3W3NwMAD0OdvgtqVSK9PR0iEQiLF++HEqlUj1PNVTUzMysy3KWlpaor68XtI7B3q99fHxgaGiIa9eu4ezZs7h8+TJWrFjRY3nVdSnVdUOVpqYmNDc3w8HBoc99RJVkVO3/KOKE8QDVr47f3rjzIFUy2blzZ5dhfefPnxe8Lm9vb3z++ecoLy9HVFQUMjIysGPHDq3V35uioiJ0dHRg5MiRGtPlcjmsrKxQUFCgtXUNJ21tbaitrYWTk5OuQ+k31RdXX24mmzZtGiIiIlBUVIStW7eqp1taWgJAt4mhL+002Pt1Z2cnOjs7YWxsjLS0NJw+fRoGBgbqxKRaf0JCAkQiEWpqamBubt5lNOD169cBAOPHj+9zH2ltbQXw3/Z/FHHCeMC4ceNgYGCA3NzcHsuMGjUKUql0QHd9l5eXo7CwEMCvHWnbtm3w9fVFYWGhVuoXQtXR79y5ozG9vr4ev/zyi3ro4OPmzJkzICJMnToVACAWix96Cksf2draQiQS9fn+iq1bt8LLywt5eXnqaePGjYOZmRm+++47jbIXLlxAa2srnn76aUF1a3O//v3vf99l2sWLF0FEmDZtGtLT07skJdURY0xMjHr7vvDCC/jmm2/Q2dmpricnJwcikQgBAQF97iOq9razsxvwZ9RXnDAeYGNjg+DgYBw5cgRpaWmoq6tDfn4+9u7dqy4jlUrx+uuv4+DBg0hNTUVdXR06Ojpw+/btLjtWT8rLy7Fy5UpcuXIFra2tyMvLQ2lpKaZOnaqV+oVwdXXF7Nmz8dFHH+Gbb76BUqnErVu3EBoaCgD44x//qLV16bPOzk7cu3cP7e3tyM/PR3h4OJydnbFs2TIAgIeHB3755RdkZ2ejra0NVVVVXX6VWllZoby8HCUlJaivr0dbWxtycnJ0NqxWJpPBzc1NfYpVKNWpKUNDQ41pa9euxdGjR7F//37U1dXh8uXLWLVqFRwcHNT7i5C6e9uvQ0JCYGdn1+vjSH7++WccOnQItbW1aGtrw/nz57FixQo4Oztr3KXdm40bN+Lu3bvYtGkTGhsbcf78eSQlJWHZsmUYM2ZMn/uIqr19fHwExzDsDO019v4bqmG19fX1tGLFChoxYgSZmZnRjBkzKDY2Vj064ocffqCWlhaKiooiZ2dnEovFZGNjQ8HBwVRQUEApKSkkk8kIAHl6elJxcTHt3buX5HI5ASAXFxf66quvyM/PjxQKBRkaGtLIkSMpJiaG2tvbiYgeWr9Q58+fp+nTp5ODg4N6dIe9vT35+flRbm4uERFVV1dTeHg4eXh4kLGxMZmZmdH06dPpH//4x6C0bXcwwFEi77//Ptnb2xMAkslkFBAQIGgbXLt2jUJDQ8nIyIgcHR1JLBaTXC6n+fPnU3Fxsbr+mpoamj17NkmlUnJ1daW33nqLIiMjCQB5eHioR+y4uLiQiYkJzZgxgyoqKujEiRNkbm5O8fHxA26j/gyrDQsLIyMjI2pqalJPO3r0KLm7uxMAsra2Vo+K+q3IyEiNYbWdnZ2UlJREnp6eZGRkRAqFgoKCgujq1atERILbu7f9OigoiABQbGzsQz/b2rVryd3dnUxNTUksFpOTkxP96U9/ovLy8h6X+e0oKZXc3FyaMmUKGRsbk4ODA0VGRlJzc7N6fl/6yLx588jR0ZE6OzsfGv+DBrr/DzVOGEyndNlhQkNDycrKSifr7ov+JIyioiISi8W0b9++QYpK+zo6OmjmzJmUlpam61D6rLq6mqRSKe3YsaNPyw23hMGnpNhj7VF9yqiHhwfi4uIQFxeHhoYGXYfTq46ODmRnZ6O+vh4hISG6DqfPNm/ejIkTJyIsLEzXoQwqThjDyJUrV7p9rPNv/4Zjh2PaFx0djUWLFiEkJETvHzB45swZZGVlIScnR/D9I/oiOTkZly5dwokTJ2BkZKTrcAYVJ4xhxMvLq8voj+7+Dh06pOtQ9d769euRnp6O+/fvw9XVFUeOHNF1SIMiISEBYWFh2LZtm65Deag5c+bgwIEDGs/tGg6OHTuGlpYWnDlzBgqFQtfhDDqxrgNgTBcSExORmJio6zCGhL+/P/z9/XUdxiMpMDAQgYGBug5jyPARBmOMMUE4YTDGGBOEEwZjjDFBOGEwxhgTZNhd9M7MzNR1CEzLtPlQxUeR6pETvO8zXRMRDY/3CWZmZmLx4sW6DoMxxrQqIyMDL730kq7DEGTYHWEMk/w2qBYtWgQAOHz4sI4jGTiRSDSsOowuPErbm2kSiUS6DqFP+BoGY4wxQThhMMYYE4QTBmOMMUE4YTDGGBOEEwZjjDFBOGEwxhgT5JFNGFlZWXBzc+vyrgiJRAJbW1vMmjULSUlJuHfvnq5D1Vs7duyAra0tRCIR9uzZo+tw2CA5deoUoqOju/SZV199tUtZf39/mJubw9DQEN7e3r2+f1tX2trakJiYCA8PD0gkElhaWmLcuHEoKSnptnxzczO8vLywYcMGAMBnn32G7du3P7Iv2OqvRzZhBAcH48aNG3B3d4eFhQWICJ2dnaisrERmZiZcXV0RFRUFb29vfPfdd7oOVy+tW7cO586d03UYbBBt2rQJu3fvxvr16zX6zIgRI7B//34cP35co/yXX36Jw4cP48UXX0RBQQF8fX11FPnDLV68GJ988gkOHDiApqYm/PTTT3B3d+/x7YMxMTG4evWq+v+AgABIpVLMmTMHtbW1QxW23ntkE0Z3RCIRLC0tMWvWLKSnpyMzMxN3797FvHnz9P6NZEy7lEol/Pz8hl3d2vTuu+/i0KFDyMzMhLm5uca83bt3w8DAAKGhocOubxw6dAjZ2dk4fPgwnnnmGYjFYjg4OODYsWMYN25cl/Lnzp3Djz/+2GX6mjVrMGHCBLzwwgtob28fitD13mOVMH5r4cKFWLZsGSorK/mUy2MmLS0NlZWVw65ubbl+/To2btyILVu2QCqVdpnv5+eH8PBw/Pzzz1i3bp0OIuy/Dz/8EL6+vvDx8em1rFKpRGRkJHbt2tXt/M2bN+PSpUs9zn/cPNYJAwCWLVsGAMjJyQHw68voY2Nj4ezsDBMTE4wfPx4ZGRkAgNTUVJiamkImk+HYsWOYO3cu5HI5nJyccPDgQXWdubm5mDJlCmQyGeRyOXx8fFBXV9dr/YPlL3/5C2QyGczNzVFZWYm1a9fC0dERV69e7XM8YWFhkEgkGq/SfPPNN2FqagqRSITq6upB/SxEhOTkZDz55JMwNjaGQqHA/PnzceXKFcHxhYeHY+3atSguLoZIJIKHhwd2794NqVQKW1tbrFy5Eg4ODpBKpfDz88OFCxcGVDcAnDx5EnK5HAkJCYPaPkLt3r0bRISAgIAey8THx+OJJ57Axx9/jFOnTvVYrrdtIrTfaKNvtLa24ttvv8XEiRMFlY+JicGbb74JGxubbucrFAo8++yz2LVrFz+WCABomMjIyKD+hOvu7k4WFhY9zq+rqyMANGrUKCIiWrduHRkbG9ORI0fo3r17tH79ejIwMKCLFy8SEVFMTAwBoNOnT9P9+/epsrKSZs6cSaamptTa2koNDQ0kl8tp+/btpFQqqaKighYsWEBVVVWC6hdi4cKFtHDhwj61gyruNWvW0Pvvv08LFiygn376qdd4ioqKCAB9+OGH6rqWLFlCdnZ2GvUnJSURAPXnFAoAZWRkCC4fGxtLEomE9u3bR7W1tZSfn0++vr5kbW1NFRUVguMLDg4md3d3jTKhoaFkampKhYWF1NzcTAUFBTR58mQyNzensrKyAdX9xRdfkLm5OcXFxQn+rCr92d69cXNzo7Fjx3Y7z93dnW7evElEROfOnSMDAwMaPXo0NTQ0EBFRTk4OBQYGqssL2Sa99Rsi7fSNmzdvEgCaOHEizZo1i+zt7cnY2Ji8vLzogw8+oM7OTnXZs2fPUkBAABERVVVVEQCKiYnpUmd0dDQBoLy8PMFxCNXX/V/XHvsjDHNzc4hEItTX16O5uRmpqakICgpCcHAwLC0tsWHDBhgZGSE9PV1jOT8/P8jlctjY2CAkJASNjY0oKytDSUkJ6urq4O3tDalUCjs7O2RlZcHa2rpP9Q+Wd999F6tXr0ZWVhZGjx6t83j6QqlUIjk5GQsWLMDSpUthYWEBHx8f7NmzB9XV1di7d++A1yEWi9W/lMeOHYvU1FTU19cPuD3mzZuHuro6bNy4ccAxDlRjYyNu3rwJd3f3XstOmzYNb7/9NkpKSvDOO+90md/XbdJTv9FW31Bd1LaxsUFCQgIKCgpw9+5dzJ8/H6tXr8ann36qjjs8PBypqam91unp6QkAuHz5suA4HlWPfcJobGwEEUEul+Pq1atoamrSuDBmYmICe3t79eF1dyQSCYBfh/K5ubnB1tYWS5cuxebNmzWG8fW3/sGib/H0pqCgAA0NDZg0aZLG9MmTJ0MikahPHWnTpEmTIJPJ9LI9+quyshJEBJlMJqh8fHw8xowZg5SUFJw9e1Zj3kC2yYP9Rlv7orGxMQDA29sbfn5+sLKygoWFBbZs2QILCwt1Alu/fj3eeOMNODo69lqnqp3u3r0rOI5H1WOfMK5duwYA8PLyQmNjIwBgw4YNGvdulJaWoqmpSVB9JiYm+PrrrzFjxgwkJCTAzc0NISEhUCqVWqlfm/Qtnt6ohjeamZl1mWdpaYn6+vpBWa+xsTGqqqoGpW5daG5uBvDfL9feSKVSpKenQyQSYfny5VAqlep52tom2toXHRwcAKDLtTSJRAIXFxcUFxfj7NmzuHz5MlasWCGoThMTEwD/bbfH2WOfME6ePAkAmDt3rvrC186dO0FEGn99eSuct7c3Pv/8c5SXlyMqKgoZGRnYsWOH1urXFn2LpzeWlpYA0O2XUG1tLZycnLS+zra2tkGrW1dUX4B9uSlt2rRpiIiIQFFREbZu3aqerq1toq190czMDJ6enigsLOwyr729HRYWFkhLS8Pp06dhYGCgTkyq9SckJEAkEmncm9Xa2grgv+32OHusE0ZFRQV27twJJycnLF++HKNGjYJUKsWlS5f6XWd5ebl6Z7WxscG2bdvg6+uLwsJCrdSvTf2NRywWo62tbZCi6tm4ceNgZmbW5UbLCxcuoLW1FU8//bTW4ztz5gyICFOnTtV63bqiunu/r/dXbN26FV5eXsjLy1NPE7pNeqPNvrF48WLk5eXhxo0b6mlNTU0oLS2Fj48P0tPTuyQl1RFkTEwMiEjjFJuqnezs7AYc23D3WCQMIkJDQwM6OzvVO0dGRgamT58OQ0NDZGdnQy6XQyqV4vXXX8fBgweRmpqKuro6dHR04Pbt27hz546gdZWXl2PlypW4cuUKWltbkZeXh9LSUkydOlUr9WtTf+Px8PDAL7/8guzsbLS1taGqqgqlpaVDEu/atWtx9OhR7N+/H3V1dbh8+TJWrVoFBwcHhIaGCo7PysoK5eXlKCkpQX19vToJdHZ24t69e2hvb0d+fj7Cw8Ph7OysHn7d37pzcnL0ZlitTCaDm5ub+l3hQqlOTRkaGmpME7JNhNTd274YEhICOzu7Xh9HEhERARcXFyxbtgxlZWWoqalBVFQUlEpltxfue6NqJyH3dTzyhmo41kD1dVjtZ599RuPHjyeZTEYSiYQMDAwIAIlEIrK0tKQpU6ZQXFwc1dTUaCzX0tJCUVFR5OzsTGKxmGxsbCg4OJgKCgooJSWFZDIZASBPT08qLi6mvXv3klwuJwDk4uJCX331Ffn5+ZFCoSBDQ0MaOXIkxcTEUHt7e6/1C9XXYZbbt28nExMT9fDhffv2Cfq87733HtnZ2REAMjU1pQULFhARUU1NDc2ePZukUim5urrSW2+9RZGRkQSAPDw81ENQhUAfhxV2dnZSUlISeXp6kpGRESkUCgoKCqKrV6+qywiJ7/vvvycXFxcyMTGhGTNmUEVFBYWGhpKRkRE5OjqSWCwmuVxO8+fPp+Li4gHXfeLECTI3N6f4+HjBn1VlMIbVhoWFkZGRETU1NamnHT16lNzd3QkAWVtb0+rVq7tdNjIyUmNYbW/bREi/uXbtWq99IygoiABQbGxsr5/v1q1b9PLLL5NCoSBjY2OaMmUK5eTk9Fj+YcNq582bR46OjhpDcrWlr/u/rj2yCeNRNhhfILqiTx0mNDSUrKysdB1GF4OxvYuKikgsFmv8eNB3HR0dNHPmTEpLSxuydVZXV5NUKqUdO3YMSv36tP8L8VickmJMqMfl6aQeHh6Ii4tDXFxcjw/k0ycdHR3Izs5GfX09QkJChmy9mzdvxsSJExEWFjZk69RnnDAYe0xFR0dj0aJFCAkJ0fsHDJ45cwZZWVnIoUgHEwAAIABJREFUyckRfP/IQCUnJ+PSpUs4ceIEjIyMhmSd+o4TBmP49Uau9PR03L9/H66urjhy5IiuQxoSCQkJCAsLw7Zt23QdykPNmTMHBw4c0HiO12A6duwYWlpacObMGSgUiiFZ53Ag1nUAjOmDxMREJCYm6joMnfD394e/v7+uw9ArgYGBCAwM1HUYeoePMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIMPuoveiRYt0HYLOffvttwAenbbYuXMnDh8+rOsw9Najtr3Z8CUiGh7vHTx//jySk5N1HQZ7DKhelMPPDmJDISIiAtOmTdN1GIIMm4TB2FB56aWXAACZmZk6joQx/cLXMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJwgmDMcaYIJwwGGOMCcIJgzHGmCCcMBhjjAnCCYMxxpggnDAYY4wJIiIi0nUQjOnKJ598guTkZHR0dKinVVdXAwCsra3V0wwNDREREYHXXnttyGNkTF9wwmCPtWvXrmHMmDGCyl69ehVPPPHEIEfEmP7iU1LssfbEE09gwoQJEIlEPZYRiUSYMGECJwv22OOEwR57r732GgwNDXucLxaL8Yc//GEII2JMP/EpKfbYKy8vx6hRo9DZ2dntfJFIhFu3bsHR0XGII2NMv/ARBnvsjRw5En5+fjAw6NodDAwMMH36dE4WjIETBmMAgFdffbXb6SKRiEdGMfYffEqKMQD37t2DnZ0d2traNKaLxWJUVFRgxIgROoqMMf3BRxiMAVAoFHjuuec0Ln4bGhri97//PScLxv6DEwZj/7F06VKNC99EhKVLl+owIsb0C5+SYuw/mpqaMGLECDQ3NwMApFIpqqurYWpqquPIGNMPfITB2H/IZDIEBQXByMgIRkZGCAoK4mTB2AM4YTD2gFdeeQVtbW1oa2vDK6+8outwGNMrYl0H8FuZmZm6DoE9xjo6OiCTyUBEqKur4/2R6dRLL72k6xA06N01jIc904cxxh4nevb1rH9HGACQkZGhd5mVDU8ikajP+1Nubi5EIhF+97vfDWJk+mPRokUAgMOHD+s4EqaSmZmJxYsX6zqMLvQyYTCmSzNnztR1CIzpJU4YjP1Gd8+UYozxKCnGGGMCccJgjDEmCCcMxhhjgnDCYIwxJggnDC1YsWIFzM3NIRKJcOnSJV2H02ednZ3YuXMn/Pz8usxra2tDbGws3NzcIJFI4OjoiHXr1kGpVOogUt04ceIELCws8Pnnn+s6FL106tQpREdHIysrC25ubhCJRBCJRN2+Y8Tf3x/m5uYwNDSEt7c3vv/+ex1E3Lu2tjYkJibCw8MDEokElpaWGDduHEpKSrot39zcDC8vL2zYsAEA8Nlnn2H79u3o6OgYwqgHHycMLfj444/x0Ucf6TqMfikqKsLvfvc7REREoKmpqcv88PBwJCUlITExETU1NThw4AA++ugjrFixQgfR6oa+3TylTzZt2oTdu3dj/fr1CA4Oxo0bN+Du7o4RI0Zg//79OH78uEb5L7/8EocPH8aLL76IgoIC+Pr66ijyh1u8eDE++eQTHDhwAE1NTfjpp5/g7u6OhoaGbsvHxMTg6tWr6v8DAgIglUoxZ84c1NbWDlXYg44TxmPshx9+wDvvvINVq1Zh4sSJXebfuHEDe/bswWuvvYaQkBCYm5tj1qxZCAsLw6effoqffvpJB1EPvXnz5uH+/ft48cUXdbJ+pVLZ7dGfrr377rs4dOgQMjMzYW5urjFv9+7dMDAwQGhoKO7fv6+jCPvn0KFDyM7OxuHDh/HMM89ALBbDwcEBx44dw7hx47qUP3fuHH788ccu09esWYMJEybghRdeQHt7+1CEPug4YWjJcHykyYQJE5CVlYUlS5bA2Ni4y/yLFy+is7MTzzzzjMb0559/HgDwz3/+c0jifNylpaWhsrJS12FouH79OjZu3IgtW7ZAKpV2me/n54fw8HD8/PPPWLdunQ4i7L8PP/wQvr6+8PHx6bWsUqlEZGQkdu3a1e38zZs349KlSz3OH244YfQDESEpKQljxoyBsbExLCwsEBkZqVGmo6MDsbGxcHZ2homJCcaPH4+MjAwAQGpqKkxNTSGTyXDs2DHMnTsXcrkcTk5OOHjwoLqO3NxcTJkyBTKZDHK5HD4+Pqirq+u1fm1R3cBmYmKiMd3T0xMAHosjjLNnz8LZ2RkikQgffPABAGHbb/fu3ZBKpbC1tcXKlSvh4OAAqVQKPz8/XLhwAQAQFhYGiUQCe3t79frefPNNmJqaQiQSobq6GuHh4Vi7di2Ki4shEong4eEBADh58iTkcjkSEhKGuEWg/nxEhICAgB7LxMfH44knnsDHH3+MU6dO9ViOiJCcnIwnn3wSxsbGUCgUmD9/Pq5cuQJAeH/RRp9obW3Ft99+2+0Rd3diYmLw5ptvwsbGptv5CoUCzz77LHbt2vVonNokPQOAMjIydB3GQ8XExJBIJKL33nuP7t27R01NTZSSkkIAKC8vj4iI1q1bR8bGxnTkyBG6d+8erV+/ngwMDOjixYvqOgDQ6dOn6f79+1RZWUkzZ84kU1NTam1tpYaGBpLL5bR9+3ZSKpVUUVFBCxYsoKqqKkH199UzzzxDEyZM0JiWn59PAGjjxo0a09vb2wkABQUF9WtdQ0kb+9OtW7cIAL3//vvqab1tPyKi0NBQMjU1pcLCQmpubqaCggKaPHkymZubU1lZGRERLVmyhOzs7DTWl5SURADU2zo4OJjc3d01ynzxxRdkbm5OcXFxA/psREQLFy6khQsX9mkZNzc3Gjt2bLfz3N3d6ebNm0REdO7cOTIwMKDRo0dTQ0MDERHl5ORQYGCgunxsbCxJJBLat28f1dbWUn5+Pvn6+pK1tTVVVFQQkbD21kafuHnzJgGgiRMn0qxZs8je3p6MjY3Jy8uLPvjgA+rs7FSXPXv2LAUEBBARUVVVFQGgmJiYLnVGR0drfDcIkZGRQXr49Ux8hNFHSqUSO3fuxP/+7/8iIiIClpaWMDExgZWVlbpMc3MzUlNTERQUhODgYFhaWmLDhg0wMjJCenq6Rn1+fn6Qy+WwsbFBSEgIGhsbUVZWhpKSEtTV1cHb2xtSqRR2dnbIysqCtbV1n+ofCB8fHzz//PNISUnB119/jebmZlRUVODo0aMQiURoa2vT2rqGq562n4pYLFb/ch47dixSU1NRX18/4O00b9481NXVYePGjQP9CH3W2NiImzdvwt3dvdey06ZNw9tvv42SkhK88847XeYrlUokJydjwYIFWLp0KSwsLODj44M9e/aguroae/fu1SjfU3trq0+oLmrb2NggISEBBQUFuHv3LubPn4/Vq1fj008/VccdHh6O1NTUXutUHZFfvnxZcBz6ihNGH12/fh1NTU2YM2dOj2WuXr2KpqYmjQtkJiYmsLe3Vx9md0cikQD4dUifm5sbbG1tsXTpUmzevFljOF9/6++PQ4cOYdGiRXjttddgZWWF6dOn4x//+AeICCNGjNDquoa7B7dfTyZNmgSZTKb17TSUKisrQUSQyWSCysfHx2PMmDFISUnB2bNnNeYVFBSgoaEBkyZN0pg+efJkSCQS9em77jzY3trqE6pred7e3vDz84OVlRUsLCywZcsWWFhYqBPY+vXr8cYbb8DR0bHXOlXtdPfuXcFx6CtOGH10+/ZtAOjxnCXw6y8wANiwYYN6TLpIJEJpaWm3Q1e7Y2Jigq+//hozZsxAQkIC3NzcEBISAqVSqZX6hbKwsMCePXtw+/ZtNDU1obi4GO+99x4AYOTIkVpd1+PC2NgYVVVVug6j31TvPO9uoER3pFIp0tPTIRKJsHz5co17eFRDTs3MzLosZ2lpifr6ekHr0FafcHBwAABUV1drTJdIJHBxcUFxcTHOnj2Ly5cvCx5arroGqGq34YwTRh+pRoS0tLT0WEaVTHbu3Aki0vg7f/684HV5e3vj888/R3l5OaKiopCRkYEdO3Zorf7+unjxIgBg9uzZg76uR01bWxtqa2vh5OSk61D6TfUF2Jeb0qZNm4aIiAgUFRVh69at6umWlpYA0G1i6Es7aatPmJmZwdPTE4WFhV3mtbe3w8LCAmlpaTh9+jQMDAzUiUm1/oSEBIhEInz33Xfq5VpbWwF0HTwyHHHC6KNx48bBwMAAubm5PZYZNWoUpFLpgO76Li8vV++0NjY22LZtG3x9fVFYWKiV+gfio48+gqurK5599lmdrH84O3PmDIgIU6dOBfDrNY7hdi3I1tYWIpGoz/dXbN26FV5eXsjLy1NPGzduHMzMzDS+YAHgwoULaG1txdNPPy2obm32icWLFyMvLw83btxQT2tqakJpaSl8fHyQnp7eJSmpjhhjYmJARBqn2FTtZGdnN+DYdI0TRh/Z2NggODgYR44cQVpaGurq6pCfn69xcU4qleL111/HwYMHkZqairq6OnR0dOD27du4c+eOoPWUl5dj5cqVuHLlClpbW5GXl4fS0lJMnTpVK/ULNWXKFJSWlqK9vR0lJSVYt24dTp06hbS0NPU5ZNazzs5O3Lt3D+3t7cjPz0d4eDicnZ2xbNkyAICHhwd++eUXZGdno62tDVVVVSgtLdWow8rKCuXl5SgpKUF9fT3a2tqQk5Ojs2G1MpkMbm5u6tOzQqlOTRkaGmpMW7t2LY4ePYr9+/ejrq4Oly9fxqpVq+Dg4IDQ0FDBdffWJ0JCQmBnZ9fr40giIiLg4uKCZcuWoaysDDU1NYiKioJSqez2wn1vVO0k5L4OvTe0g7J6h2EwrLa+vp5WrFhBI0aMIDMzM5oxYwbFxsYSAHJycqIffviBWlpaKCoqipydnUksFpONjQ0FBwdTQUEBpaSkkEwmIwDk6elJxcXFtHfvXpLL5QSAXFxc6KuvviI/Pz9SKBRkaGhII0eOpJiYGGpvbyciemj9Qp0/f56mT59ODg4OBIAAkL29Pfn5+VFubi4RET333HNkaWlJYrGYFAoFzZs3r99Dd3VhoPvT+++/T/b29gSAZDIZBQQECNp+165do9DQUDIyMiJHR0cSi8Ukl8tp/vz5VFxcrK6/pqaGZs+eTVKplFxdXemtt96iyMhIAkAeHh5UVlZG33//Pbm4uJCJiQnNmDGDKioq6MSJE2Rubk7x8fEDbqP+DKsNCwsjIyMjampqUk87evQoubu7EwCytram1atXd7tsZGSkxrDazs5OSkpKIk9PTzIyMiKFQkFBQUF09epVIiLB7d1bnwgKCiIAFBsb2+vnu3XrFr388sukUCjI2NiYpkyZQjk5OT2Wf9iw2nnz5pGjo6PGkNze6OuwWr2LaDgkDDZ86HJ/Cg0NJSsrK52suy/6kzCKiopILBbTvn37Bikq7evo6KCZM2dSWlrakK2zurqapFIp7dixo0/L6WvC4FNSjA2iR+1ppSoeHh6Ii4tDXFxcjw/k0ycdHR3Izs5GfX09QkJChmy9mzdvxsSJExEWFjZk6xxMnDAeMVeuXNEYVtjT31B2GvZoio6OxqJFixASEqL3Dxg8c+YMsrKykJOTI/j+kYFKTk7GpUuXcOLECRgZGQ3JOgcbJ4xHjJeXV5cRHN39HTp0SNehPtLWr1+P9PR03L9/H66urjhy5IiuQxoUCQkJCAsLw7Zt23QdykPNmTMHBw4c0Hhu12A6duwYWlpacObMGSgUiiFZ51AQ6zoAxh5FiYmJSExM1HUYQ8Lf3x/+/v66DkOvBAYGIjAwUNdhaB0fYTDGGBOEEwZjjDFBOGEwxhgThBMGY4wxQURE+vUaKJFIhKlTpw7rh7Mx/XHkyBHen3rx7bffAoD6+VZM927fvo1vv/1W797Sx0cYeur27duP7FBMxtjwpJdHGBkZGXjppZd0HYpOZWZmYvHixXr3C2O44f2pd4sWLQIAHD58WMeRMBV97f98hMEYY0wQThiMMcYE4YTBGGNMEE4YjDHGBOGEwRhjTJBhnTCysrLg5ubW5dHdEokEtra2mDVrFpKSknDv3j1dhzpstLS0YM2aNbC3t4dMJsPJkyd1HRLTc6dOnUJ0dHSX/vjqq692Kevv7w9zc3MYGhrC29u719el6lJnZyd27twJPz+/buefPXsW06dPh0wmg4ODA6KiotDS0gIA+Oyzz7B9+/ZH7n0owzphBAcH48aNG3B3d4eFhQWICJ2dnaisrERmZiZcXV0RFRUFb2/vLi+ZZ9177733cPLkSVy5cgW7du0aFi/HYbqzadMm7N69G+vXr9fojyNGjMD+/ftx/PhxjfJffvklDh8+jBdffBEFBQXw9fXVUeQPV1RUhN/97neIiIhAU1NTl/kFBQXw9/fHnDlzUFVVhaNHj+Kvf/0rVq1aBQAICAiAVCrFnDlzUFtbO9ThD5phnTC6IxKJYGlpiVmzZiE9PR2ZmZm4e/cu5s2bp/cvedEH2dnZmDRpEiwtLfHGG29g4cKFUCqVPf7KYt0bzDbTl+3x7rvv4tChQ8jMzIS5ubnGvN27d8PAwAChoaHDrt/98MMPeOedd7Bq1SpMnDix2zJbt26Fvb09tmzZAlNTU0ybNg1RUVH429/+hitXrgAA1qxZgwkTJuCFF15Ae3v7UH6EQfPIJYzfWrhwIZYtW4bKykrs2bNH1+Hovdu3b3d5O1haWhoqKyt1FNHwNJhtpg/b4/r169i4cSO2bNkCqVTaZb6fnx/Cw8Px888/Y926dTqIsP8mTJiArKwsLFmyBMbGxl3mt7e34/jx43j22WchEonU0+fOnQsiwrFjx9TTNm/ejEuXLmHXrl1DEvtge+QTBgAsW7YMAJCTkwPg1/f7xsbGwtnZGSYmJhg/fjwyMjIAAKmpqTA1NYVMJsOxY8cwd+5cyOVyODk54eDBg+o6c3NzMWXKFMhkMsjlcvj4+KCurq7X+gfTw2IiIiQnJ+PJJ5+EsbExFAoF5s+fr/419NVXX8HDwwN37tzB3//+d4hEIpiZmSE8PBxr165FcXExRCIRPDw8sGvXLpiamsLAwABPP/007OzsYGRkBFNTU/j6+mLmzJkYNWoUpFIpLC0t8X//938acf7rX//C2LFjYWFhAalUCh8fH/zzn/8EAPztb3+DmZkZRCIRFAoFsrOz8d1338HFxQWGhoZ45ZVXBrUNe2unsLAwSCQSjTe3vfnmmzA1NYVIJEJ1dXW3bbZ7925IpVLY2tpi5cqVcHBwgFQqhZ+fHy5cuDCgugHg5MmTkMvlSEhIGNT2Udm9ezeICAEBAT2WiY+PxxNPPIGPP/4Yp06d6rFcb20utE8OVb+7ceMGGhoa4OzsrDHd3d0dAJCfn6+eplAo8Oyzz2LXrl16d9d2v5CeAUAZGRl9Wsbd3Z0sLCx6nF9XV0cAaNSoUUREtG7dOjI2NqYjR47QvXv3aP369WRgYEAXL14kIqKYmBgCQKdPn6b79+9TZWUlzZw5k0xNTam1tZUaGhpILpfT9u3bSalUUkVFBS1YsICqqqoE1S9ERkYG9WXz9BZTbGwsSSQS2rdvH9XW1lJ+fj75+vqStbU1VVRUqOuxs7OjP/zhDxp1BwcHk7u7u8a0TZs2EQC6cOECNTY2UnV1NT3//PMEgI4fP05VVVXU2NhIYWFhBIAuXbqkXvbw4cO0efNm+uWXX6impoamTp1KI0aMUM8vLCwkmUymEUd0dDR9/PHHgttDpa/7k5B2WrJkCdnZ2Wksl5SURADU7d1dm4WGhpKpqSkVFhZSc3MzFRQU0OTJk8nc3JzKysoGVPcXX3xB5ubmFBcXJ/izqixcuJAWLlzYp2Xc3Nxo7Nix3c5zd3enmzdvEhHRuXPnyMDAgEaPHk0NDQ1ERJSTk0OBgYHq8kLavLc+SaSdfvdbzzzzDE2YMEFjWm5uLgGgpKSkLuVNTExozpw5GtOio6MJAOXl5Qleb1/7/1B5LI4wzM3NIRKJUF9fj+bmZqSmpiIoKAjBwcGwtLTEhg0bYGRkhPT0dI3l/Pz8IJfLYWNjg5CQEDQ2NqKsrAwlJSWoq6uDt7c3pFIp7OzskJWVBWtr6z7Vr00Pi0mpVCI5ORkLFizA0qVLYWFhAR8fH+zZswfV1dXYu3dvv9c7duxYyGQyjBgxAi+//DIAwNnZGdbW1pDJZFi6dCkAqH8tAr+eJty0aRMUCgWsrKwQEBCAmpoaVFVVAQCefPJJ7Ny5E3//+99x4MABHDx4EC0tLfjjH/84gBbq3WC2k4pYLFb/kh47dixSU1NRX18/4H1j3rx5qKurw8aNGwccY28aGxtx8+ZN9S/qh5k2bRrefvttlJSU4J133ukyv69t3lOfHMp+pxoJZWho2GWekZERlEqlxjRPT08AwOXLl7Uahy48FgmjsbERRAS5XI6rV6+iqakJ48aNU883MTGBvb29xpfab0kkEgBAW1sb3NzcYGtri6VLl2Lz5s0oKSlRl+tv/QP1sJgKCgrQ0NCASZMmaSwzefJkSCQS9SmRgVK10YMX+FTXQ9ra2npcTlXmwSGIqgvuK1euRGZmJv7yl79oJcaHGap2etCkSZMgk8kGdd/QtsrKShARZDKZoPLx8fEYM2YMUlJScPbsWY15A2nzB/vkUPY71TWb7i5kt7a2wsTERGOaqp3u3r2r1Th04bFIGNeuXQMAeHl5obGxEQCwYcMGjXs3SktLux0+1x0TExN8/fXXmDFjBhISEuDm5oaQkBAolUqt1N8fD4tJNazPzMysy3KWlpaor68ftLi6c/z4ccyaNQs2NjYwNjbuco1DJSEhAQ0NDUN2gVdX7WRsbKw+uhoOmpubAaDbC8LdkUqlSE9Ph0gkwvLlyzV+gWurzYey36muMamuD6o0NTWhubkZDg4OGtNVCUTVbsPZY5EwVDefzZ07FzY2NgCAnTt3gog0/s6fPy+4Tm9vb3z++ecoLy9HVFQUMjIysGPHDq3V3x89xWRpaQkA3Xa+2traIX25UFlZGYKCgmBvb48LFy7g/v372L59e5dybW1tWLNmDZKTk3H+/HnEx8cPemy6aKe2trYh3wYDpfoC7MtNadOmTUNERASKioqwdetW9XRttflQ9jtXV1eYm5ujtLRUY/r169cBAOPHj9eY3traCgBdjjyGo0c+YVRUVGDnzp1wcnLC8uXL1aN3Ll261O86y8vLUVhYCODXHXXbtm3w9fVFYWGhVurXdkzjxo2DmZlZl5sXL1y4gNbWVjz99NNDFufly5fR1taGP//5z3Bzc4NUKtUYmqjy1ltv4U9/+hPefvttREREYOvWrYOecIW2k1gsfugptr44c+YMiEj9tjtt1j1YbG1tIRKJ+nx/xdatW+Hl5YW8vDz1NG3tm0PZ78RiMV544QV888036OzsVE/PycmBSCTqMnJM1U52dnaDHttge2QSBhGhoaEBnZ2dICJUVVUhIyMD06dPh6GhIbKzsyGXyyGVSvH666/j4MGDSE1NRV1dHTo6OnD79m3cuXNH0LrKy8uxcuVKXLlyBa2trcjLy0NpaSmmTp2qlfr7o7eY1q5di6NHj2L//v2oq6vD5cuXsWrVKjg4OCA0NPShdVtZWaG8vBwlJSWor68f0BeaaijiqVOn0NzcjKKioi7nqVNSUuDo6IgFCxYAABITEzF27FgsWbKky2kAbRLaTh4eHvjll1+QnZ2NtrY2VFVVdfm12VObdXZ24t69e2hvb0d+fj7Cw8Ph7OysHvrd37pzcnKGbFitTCaDm5sbbt++3aflVKemHrxYPNB988F6eut3ISEhsLOz08rjSDZu3Ii7d+9i06ZNaGxsxPnz55GUlIRly5ZhzJgxGmVV7eTj4zPg9erckI/L6gX6MAzys88+o/Hjx5NMJiOJREIGBgYEgEQiEVlaWtKUKVMoLi6OampqNJZraWmhqKgocnZ2JrFYTDY2NhQcHEwFBQWUkpJCMpmMAJCnpycVFxfT3r17SS6XEwBycXGhr776ivz8/EihUJChoSGNHDmSYmJiqL29vdf6herrsLqSkpKHxtTZ2UlJSUnk6elJRkZGpFAoKCgoiK5evape/qmnniIAJBaLydfXl44cOUJERN9//z25uLiQiYkJzZgxg6Kjo9VtNHr0aPrXv/5F7777LllYWBAAsrOzowMHDtChQ4fIzs6OAJBCoaCDBw8SEVFUVBRZWVmRpaUlLVq0iD744AMCQO7u7jRx4kQSiURkZWVF586dIyKit99+W71tLSws6LvvvhPcLn3Zn4S0ExFRTU0NzZ49m6RSKbm6utJbb71FkZGRBIA8PDyorKysS5tVVFRQaGgoGRkZkaOjI4nFYpLL5TR//nwqLi4ecN0nTpwgc3Nzio+PF/xZVfozrDYsLIyMjIyoqalJPe3o0aPk7u5OAMja2ppWr17d7bKRkZEaw2p7a3MhffLatWu99rugoCACQLGxsQ/9bOfPn6fp06eTg4MDASAAZG9vT35+fpSbm6sul5ubS1OmTCFjY2NycHCgyMhIam5u7lLfvHnzyNHRkTo7OwW3r74Oq9W7iPrawR9V+rrDDDf6tD+FhoaSlZWVrsPooj8Jo6ioiMRiMe3bt2+QotK+jo4OmjlzJqWlpQ3ZOqurq0kqldKOHTv6tJy+9v9H5pQUY8PBo/L0Ug8PD8TFxSEuLm5YPKCyo6MD2dnZqK+vR0hIyJCtd/PmzZg4cSLCwsKGbJ2DiRMGY6xfoqOjsWjRIoSEhOj9AwbPnDmDrKws5OTkCL5/ZKCSk5Nx6dIlnDhxosvz2YYrThiMDYH169cjPT0d9+/t6/sJAAAAyklEQVTfh6urK44cOaLrkLQiISEBYWFh2LZtm65Deag5c+bgwIEDGs/pGkzHjh1DS0sLzpw5A4VCMSTrHApiXQfA2OMgMTERiYmJug5jUPj7+8Pf31/XYeiVwMBABAYG6joMreMjDMYYY4JwwmCMMSYIJwzGGGOCcMJgjDEmCCcMxhhjgoiI9Ou9gd09iI4xxh5Hevb1rH/Daofi3deMMcb6Tu+OMBhjjOknvobBGGNMEE4YjDHGBOGEwRhjTBAxgMO6DoIxxpj++3/XdSlZbi9FBAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "model.fit(x_train, y_train, batch_size=32, epochs=10, validation_data=(x_test, y_test)) # Обучение модели"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LTYb2_84bBlB",
        "outputId": "b561b816-f1d3-49c8-db73-10d0b76bafe9"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "1875/1875 [==============================] - 42s 22ms/step - loss: 0.1825 - accuracy: 0.9458 - val_loss: 0.0712 - val_accuracy: 0.9769\n",
            "Epoch 2/10\n",
            "1875/1875 [==============================] - 41s 22ms/step - loss: 0.0651 - accuracy: 0.9800 - val_loss: 0.0568 - val_accuracy: 0.9815\n",
            "Epoch 3/10\n",
            "1875/1875 [==============================] - 40s 22ms/step - loss: 0.0440 - accuracy: 0.9863 - val_loss: 0.0444 - val_accuracy: 0.9848\n",
            "Epoch 4/10\n",
            "1875/1875 [==============================] - 40s 21ms/step - loss: 0.0314 - accuracy: 0.9905 - val_loss: 0.0449 - val_accuracy: 0.9848\n",
            "Epoch 5/10\n",
            "1875/1875 [==============================] - 40s 21ms/step - loss: 0.0233 - accuracy: 0.9923 - val_loss: 0.0495 - val_accuracy: 0.9833\n",
            "Epoch 6/10\n",
            "1875/1875 [==============================] - 41s 22ms/step - loss: 0.0164 - accuracy: 0.9949 - val_loss: 0.0492 - val_accuracy: 0.9854\n",
            "Epoch 7/10\n",
            "1875/1875 [==============================] - 39s 21ms/step - loss: 0.0130 - accuracy: 0.9958 - val_loss: 0.0460 - val_accuracy: 0.9850\n",
            "Epoch 8/10\n",
            "1875/1875 [==============================] - 40s 21ms/step - loss: 0.0095 - accuracy: 0.9970 - val_loss: 0.0496 - val_accuracy: 0.9857\n",
            "Epoch 9/10\n",
            "1875/1875 [==============================] - 40s 21ms/step - loss: 0.0066 - accuracy: 0.9977 - val_loss: 0.0594 - val_accuracy: 0.9844\n",
            "Epoch 10/10\n",
            "1875/1875 [==============================] - 40s 22ms/step - loss: 0.0064 - accuracy: 0.9981 - val_loss: 0.0606 - val_accuracy: 0.9840\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f28f92ed240>"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loss, accuracy = model.evaluate(x_test, y_test)\n",
        "print(f'Loss: {loss}, Accuracy: {accuracy}') # Оценка модели на тестовых данных"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MwXSUneFbDOi",
        "outputId": "7c028bef-4b99-4f7b-b195-570eef01209a"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 3s 9ms/step - loss: 0.0606 - accuracy: 0.9840\n",
            "Loss: 0.06062987819314003, Accuracy: 0.984000027179718\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "index = np.random.randint(len(x_test)) # возьмем случайное изображение\n",
        "image = x_test[index]\n",
        "image = np.expand_dims(image, axis=0)\n",
        "prediction = model.predict(image) # найдем метки\n",
        "predicted_digit = np.argmax(prediction)\n",
        "remainder = predicted_digit % 2 # Вычисление остатка на 2\n",
        "print(f'Predicted Digit: {predicted_digit}, Remainder: {remainder}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kaxY7rWFbFIq",
        "outputId": "1a0eed42-28f4-4ad6-d074-7633189e92f9"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 69ms/step\n",
            "Predicted Digit: 2, Remainder: 0\n"
          ]
        }
      ]
    }
  ]
}