ep21_TimeSeries_MLP.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ep21_TimeSeries_MLP.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Episode 21. Time Series Forecasting with MLP\n"
      ],
      "metadata": {
        "id": "sj4fW3FO9EbH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Import Libraries"
      ],
      "metadata": {
        "id": "raU7i9fQpFjI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "from pylab import rcParams\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib import rc\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import LSTM\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "%matplotlib inline\n",
        "%config InlineBackend.figure_format='retina'\n",
        "sns.set(style='whitegrid', palette='muted', font_scale=1.5)\n",
        "rcParams['figure.figsize'] = 16, 10"
      ],
      "metadata": {
        "id": "cnU2KHORo7bL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data"
      ],
      "metadata": {
        "id": "aELbRhkXpArz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Original Data Source: https://fred.stlouisfed.org/series/RSXFSN\n",
        "from google.colab import files\n",
        "uploaded = files.upload()"
      ],
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 73
        },
        "id": "S6RZ7HSho-ap",
        "outputId": "367f15c1-7c37-4420-c666-894e171a4dea"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-465cda01-e319-4fc6-8dac-93b3e4091fb2\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-465cda01-e319-4fc6-8dac-93b3e4091fb2\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving RSXFSN.csv to RSXFSN.csv\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Store data in DataFrame\n",
        "df = pd.read_csv(\"RSXFSN.csv\")"
      ],
      "metadata": {
        "id": "RctD57SfpOwJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# keep real GDP only\n",
        "keep_columns = ['RSXFSN']\n",
        "df = df[keep_columns]\n",
        "# rescale\n",
        "df = df/10000"
      ],
      "metadata": {
        "id": "f0jXq7etpTSS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data Preprocessing"
      ],
      "metadata": {
        "id": "4JE0CFb5pbdX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# split train & test data\n",
        "n_train = int(len(df) * 0.8)\n",
        "n_test = len(df) - n_train\n",
        "train, test = df.iloc[0:n_train], df.iloc[n_train:len(df)]\n",
        "print(len(train), len(test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G3JnKoixpXdf",
        "outputId": "40765987-bc69-400f-f991-5a077e1fedbf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "288 72\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def get_timeseries(x, n_steps=1):\n",
        "    x_ts, y_ts = [], []\n",
        "    for ii in range(len(x) - n_steps):\n",
        "        v = x.iloc[ii:(ii + n_steps)].values\n",
        "        x_ts.append(v)\n",
        "        w = x.iloc[ii + n_steps].values\n",
        "        y_ts.append(w)\n",
        "    return np.array(x_ts), np.array(y_ts)"
      ],
      "metadata": {
        "id": "0xqpqJC9pUIx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "n_steps = 16\n",
        "n_features = 1\n",
        "\n",
        "# convert to a time series dimension: [samples, n_steps, n_features]\n",
        "\n",
        "x_train, y_train = get_timeseries(train, n_steps)\n",
        "x_test, y_test = get_timeseries(test,n_steps)\n",
        "\n",
        "print(x_train.shape, y_train.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q9_iy9rapUMS",
        "outputId": "f03b4cc5-b85a-4421-f6a9-d6fc51056610"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(272, 16, 1) (272, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modeling"
      ],
      "metadata": {
        "id": "99NYIyR4puXf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# define model\n",
        "model = Sequential()\n",
        "model.add(Dense(50, activation='relu', input_dim=n_steps))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss='mean_squared_error', optimizer=Adam(0.001))"
      ],
      "metadata": {
        "id": "ji8lDvbI9fUf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Training"
      ],
      "metadata": {
        "id": "m9R2mmq5p1hI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# fit model\n",
        "history = model.fit(x_train, y_train,epochs=500,batch_size=32,validation_split=0.1,verbose=1,shuffle=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZbhccVsR9iB9",
        "outputId": "66c0620a-db97-4cf0-d783-2f3550ee7485"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "8/8 [==============================] - 1s 28ms/step - loss: 55.2401 - val_loss: 35.3930\n",
            "Epoch 2/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 35.1509 - val_loss: 21.5045\n",
            "Epoch 3/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 9.1439 - val_loss: 24.5754\n",
            "Epoch 4/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 8.4567 - val_loss: 7.9874\n",
            "Epoch 5/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 7.5340 - val_loss: 10.5402\n",
            "Epoch 6/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 4.8092 - val_loss: 8.4909\n",
            "Epoch 7/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 5.0948 - val_loss: 7.3388\n",
            "Epoch 8/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 5.0001 - val_loss: 7.4171\n",
            "Epoch 9/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.5835 - val_loss: 7.1917\n",
            "Epoch 10/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.6679 - val_loss: 7.1434\n",
            "Epoch 11/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.5798 - val_loss: 7.0417\n",
            "Epoch 12/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.5314 - val_loss: 6.9840\n",
            "Epoch 13/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 4.5394 - val_loss: 6.9351\n",
            "Epoch 14/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 4.4631 - val_loss: 6.8684\n",
            "Epoch 15/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.4607 - val_loss: 6.8228\n",
            "Epoch 16/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 4.4117 - val_loss: 6.7645\n",
            "Epoch 17/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.3873 - val_loss: 6.7073\n",
            "Epoch 18/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 4.3519 - val_loss: 6.6419\n",
            "Epoch 19/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.3175 - val_loss: 6.5813\n",
            "Epoch 20/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.2861 - val_loss: 6.5153\n",
            "Epoch 21/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 4.2488 - val_loss: 6.4484\n",
            "Epoch 22/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.2160 - val_loss: 6.3880\n",
            "Epoch 23/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.1790 - val_loss: 6.3222\n",
            "Epoch 24/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.1441 - val_loss: 6.2563\n",
            "Epoch 25/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 4.1074 - val_loss: 6.1940\n",
            "Epoch 26/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 4.0720 - val_loss: 6.1268\n",
            "Epoch 27/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 4.0339 - val_loss: 6.0600\n",
            "Epoch 28/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.9988 - val_loss: 5.9915\n",
            "Epoch 29/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.9606 - val_loss: 5.9236\n",
            "Epoch 30/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.9241 - val_loss: 5.8590\n",
            "Epoch 31/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.8863 - val_loss: 5.7941\n",
            "Epoch 32/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.8499 - val_loss: 5.7238\n",
            "Epoch 33/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.8113 - val_loss: 5.6594\n",
            "Epoch 34/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.7747 - val_loss: 5.5934\n",
            "Epoch 35/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.7364 - val_loss: 5.5266\n",
            "Epoch 36/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.6995 - val_loss: 5.4658\n",
            "Epoch 37/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.6616 - val_loss: 5.4003\n",
            "Epoch 38/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 3.6240 - val_loss: 5.3293\n",
            "Epoch 39/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.5862 - val_loss: 5.2664\n",
            "Epoch 40/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.5488 - val_loss: 5.2052\n",
            "Epoch 41/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 3.5115 - val_loss: 5.1361\n",
            "Epoch 42/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.4731 - val_loss: 5.0761\n",
            "Epoch 43/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.4374 - val_loss: 5.0121\n",
            "Epoch 44/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.3988 - val_loss: 4.9472\n",
            "Epoch 45/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.3622 - val_loss: 4.8872\n",
            "Epoch 46/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.3249 - val_loss: 4.8170\n",
            "Epoch 47/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 3.2873 - val_loss: 4.7549\n",
            "Epoch 48/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.2518 - val_loss: 4.6980\n",
            "Epoch 49/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 3.2146 - val_loss: 4.6385\n",
            "Epoch 50/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.1789 - val_loss: 4.5786\n",
            "Epoch 51/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.1430 - val_loss: 4.5132\n",
            "Epoch 52/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.1052 - val_loss: 4.4572\n",
            "Epoch 53/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.0713 - val_loss: 4.3934\n",
            "Epoch 54/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 3.0336 - val_loss: 4.3344\n",
            "Epoch 55/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.9996 - val_loss: 4.2749\n",
            "Epoch 56/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.9622 - val_loss: 4.2208\n",
            "Epoch 57/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.9294 - val_loss: 4.1619\n",
            "Epoch 58/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.8914 - val_loss: 4.1042\n",
            "Epoch 59/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.8592 - val_loss: 4.0456\n",
            "Epoch 60/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.8231 - val_loss: 3.9914\n",
            "Epoch 61/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.7903 - val_loss: 3.9325\n",
            "Epoch 62/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 2.7557 - val_loss: 3.8784\n",
            "Epoch 63/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.7238 - val_loss: 3.8204\n",
            "Epoch 64/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.6871 - val_loss: 3.7699\n",
            "Epoch 65/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.6577 - val_loss: 3.7225\n",
            "Epoch 66/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.6242 - val_loss: 3.6609\n",
            "Epoch 67/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 2.5906 - val_loss: 3.6114\n",
            "Epoch 68/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.5590 - val_loss: 3.5510\n",
            "Epoch 69/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.5260 - val_loss: 3.5037\n",
            "Epoch 70/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 2.4973 - val_loss: 3.4615\n",
            "Epoch 71/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.4627 - val_loss: 3.4002\n",
            "Epoch 72/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.4335 - val_loss: 3.3487\n",
            "Epoch 73/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.4043 - val_loss: 3.3059\n",
            "Epoch 74/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.3712 - val_loss: 3.2631\n",
            "Epoch 75/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 2.3427 - val_loss: 3.2145\n",
            "Epoch 76/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.3127 - val_loss: 3.1607\n",
            "Epoch 77/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.2827 - val_loss: 3.1130\n",
            "Epoch 78/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 2.2545 - val_loss: 3.0646\n",
            "Epoch 79/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.2250 - val_loss: 3.0181\n",
            "Epoch 80/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.1960 - val_loss: 2.9833\n",
            "Epoch 81/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.1675 - val_loss: 2.9433\n",
            "Epoch 82/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.1391 - val_loss: 2.8934\n",
            "Epoch 83/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.1124 - val_loss: 2.8524\n",
            "Epoch 84/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 2.0847 - val_loss: 2.8037\n",
            "Epoch 85/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.0542 - val_loss: 2.7546\n",
            "Epoch 86/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 2.0241 - val_loss: 2.7018\n",
            "Epoch 87/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.9894 - val_loss: 2.6487\n",
            "Epoch 88/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.9531 - val_loss: 2.5960\n",
            "Epoch 89/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.9193 - val_loss: 2.5428\n",
            "Epoch 90/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.8882 - val_loss: 2.4989\n",
            "Epoch 91/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.8591 - val_loss: 2.4632\n",
            "Epoch 92/500\n",
            "8/8 [==============================] - 0s 9ms/step - loss: 1.8322 - val_loss: 2.4256\n",
            "Epoch 93/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.8052 - val_loss: 2.3887\n",
            "Epoch 94/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.7796 - val_loss: 2.3527\n",
            "Epoch 95/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.7549 - val_loss: 2.3159\n",
            "Epoch 96/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.7306 - val_loss: 2.2817\n",
            "Epoch 97/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.7071 - val_loss: 2.2468\n",
            "Epoch 98/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.6843 - val_loss: 2.2146\n",
            "Epoch 99/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.6624 - val_loss: 2.1804\n",
            "Epoch 100/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.6392 - val_loss: 2.1451\n",
            "Epoch 101/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 1.6171 - val_loss: 2.1181\n",
            "Epoch 102/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.5962 - val_loss: 2.0890\n",
            "Epoch 103/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.5734 - val_loss: 2.0571\n",
            "Epoch 104/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.5516 - val_loss: 2.0274\n",
            "Epoch 105/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.5320 - val_loss: 1.9987\n",
            "Epoch 106/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.5113 - val_loss: 1.9665\n",
            "Epoch 107/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.4912 - val_loss: 1.9374\n",
            "Epoch 108/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.4719 - val_loss: 1.9076\n",
            "Epoch 109/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.4534 - val_loss: 1.8800\n",
            "Epoch 110/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.4346 - val_loss: 1.8511\n",
            "Epoch 111/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.4163 - val_loss: 1.8236\n",
            "Epoch 112/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.3983 - val_loss: 1.7971\n",
            "Epoch 113/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.3809 - val_loss: 1.7755\n",
            "Epoch 114/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.3624 - val_loss: 1.7460\n",
            "Epoch 115/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.3460 - val_loss: 1.7236\n",
            "Epoch 116/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.3311 - val_loss: 1.7016\n",
            "Epoch 117/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.3130 - val_loss: 1.6770\n",
            "Epoch 118/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.2965 - val_loss: 1.6507\n",
            "Epoch 119/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.2829 - val_loss: 1.6315\n",
            "Epoch 120/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.2690 - val_loss: 1.6104\n",
            "Epoch 121/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.2516 - val_loss: 1.5857\n",
            "Epoch 122/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.2370 - val_loss: 1.5665\n",
            "Epoch 123/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.2239 - val_loss: 1.5463\n",
            "Epoch 124/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.2097 - val_loss: 1.5258\n",
            "Epoch 125/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.1965 - val_loss: 1.5077\n",
            "Epoch 126/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.1830 - val_loss: 1.4885\n",
            "Epoch 127/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.1698 - val_loss: 1.4704\n",
            "Epoch 128/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.1573 - val_loss: 1.4519\n",
            "Epoch 129/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.1451 - val_loss: 1.4332\n",
            "Epoch 130/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.1332 - val_loss: 1.4170\n",
            "Epoch 131/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.1221 - val_loss: 1.4012\n",
            "Epoch 132/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.1098 - val_loss: 1.3830\n",
            "Epoch 133/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.0992 - val_loss: 1.3679\n",
            "Epoch 134/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.0887 - val_loss: 1.3526\n",
            "Epoch 135/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.0782 - val_loss: 1.3379\n",
            "Epoch 136/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.0676 - val_loss: 1.3216\n",
            "Epoch 137/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.0575 - val_loss: 1.3059\n",
            "Epoch 138/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.0484 - val_loss: 1.2941\n",
            "Epoch 139/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.0390 - val_loss: 1.2796\n",
            "Epoch 140/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.0292 - val_loss: 1.2656\n",
            "Epoch 141/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 1.0202 - val_loss: 1.2504\n",
            "Epoch 142/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 1.0120 - val_loss: 1.2366\n",
            "Epoch 143/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 1.0048 - val_loss: 1.2262\n",
            "Epoch 144/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9961 - val_loss: 1.2126\n",
            "Epoch 145/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9867 - val_loss: 1.1995\n",
            "Epoch 146/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9795 - val_loss: 1.1901\n",
            "Epoch 147/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9717 - val_loss: 1.1779\n",
            "Epoch 148/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9648 - val_loss: 1.1679\n",
            "Epoch 149/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9582 - val_loss: 1.1574\n",
            "Epoch 150/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9508 - val_loss: 1.1457\n",
            "Epoch 151/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9440 - val_loss: 1.1335\n",
            "Epoch 152/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9378 - val_loss: 1.1234\n",
            "Epoch 153/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9320 - val_loss: 1.1138\n",
            "Epoch 154/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9258 - val_loss: 1.1041\n",
            "Epoch 155/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9192 - val_loss: 1.0933\n",
            "Epoch 156/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9133 - val_loss: 1.0834\n",
            "Epoch 157/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.9081 - val_loss: 1.0742\n",
            "Epoch 158/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.9029 - val_loss: 1.0648\n",
            "Epoch 159/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8981 - val_loss: 1.0559\n",
            "Epoch 160/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8934 - val_loss: 1.0475\n",
            "Epoch 161/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8884 - val_loss: 1.0392\n",
            "Epoch 162/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8829 - val_loss: 1.0294\n",
            "Epoch 163/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8784 - val_loss: 1.0227\n",
            "Epoch 164/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8734 - val_loss: 1.0140\n",
            "Epoch 165/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8698 - val_loss: 1.0072\n",
            "Epoch 166/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8660 - val_loss: 1.0007\n",
            "Epoch 167/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8614 - val_loss: 0.9927\n",
            "Epoch 168/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.8572 - val_loss: 0.9849\n",
            "Epoch 169/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8541 - val_loss: 0.9786\n",
            "Epoch 170/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8501 - val_loss: 0.9718\n",
            "Epoch 171/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8455 - val_loss: 0.9634\n",
            "Epoch 172/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8431 - val_loss: 0.9577\n",
            "Epoch 173/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8411 - val_loss: 0.9523\n",
            "Epoch 174/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8376 - val_loss: 0.9455\n",
            "Epoch 175/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8334 - val_loss: 0.9374\n",
            "Epoch 176/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8290 - val_loss: 0.9291\n",
            "Epoch 177/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.8256 - val_loss: 0.9240\n",
            "Epoch 178/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8225 - val_loss: 0.9177\n",
            "Epoch 179/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8191 - val_loss: 0.9109\n",
            "Epoch 180/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8173 - val_loss: 0.9052\n",
            "Epoch 181/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8161 - val_loss: 0.9029\n",
            "Epoch 182/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.8125 - val_loss: 0.8954\n",
            "Epoch 183/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8100 - val_loss: 0.8914\n",
            "Epoch 184/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8074 - val_loss: 0.8850\n",
            "Epoch 185/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.8050 - val_loss: 0.8783\n",
            "Epoch 186/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.8045 - val_loss: 0.8756\n",
            "Epoch 187/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.8025 - val_loss: 0.8727\n",
            "Epoch 188/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7982 - val_loss: 0.8643\n",
            "Epoch 189/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7968 - val_loss: 0.8608\n",
            "Epoch 190/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7960 - val_loss: 0.8584\n",
            "Epoch 191/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7938 - val_loss: 0.8547\n",
            "Epoch 192/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7910 - val_loss: 0.8473\n",
            "Epoch 193/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7902 - val_loss: 0.8435\n",
            "Epoch 194/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7893 - val_loss: 0.8402\n",
            "Epoch 195/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7877 - val_loss: 0.8366\n",
            "Epoch 196/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7853 - val_loss: 0.8319\n",
            "Epoch 197/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7834 - val_loss: 0.8277\n",
            "Epoch 198/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7820 - val_loss: 0.8238\n",
            "Epoch 199/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7806 - val_loss: 0.8195\n",
            "Epoch 200/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7800 - val_loss: 0.8165\n",
            "Epoch 201/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7789 - val_loss: 0.8130\n",
            "Epoch 202/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7774 - val_loss: 0.8090\n",
            "Epoch 203/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7758 - val_loss: 0.8048\n",
            "Epoch 204/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7744 - val_loss: 0.8009\n",
            "Epoch 205/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7730 - val_loss: 0.7968\n",
            "Epoch 206/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7715 - val_loss: 0.7920\n",
            "Epoch 207/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7715 - val_loss: 0.7881\n",
            "Epoch 208/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7703 - val_loss: 0.7836\n",
            "Epoch 209/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7667 - val_loss: 0.7787\n",
            "Epoch 210/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7623 - val_loss: 0.7735\n",
            "Epoch 211/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7591 - val_loss: 0.7692\n",
            "Epoch 212/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7576 - val_loss: 0.7654\n",
            "Epoch 213/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7574 - val_loss: 0.7626\n",
            "Epoch 214/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7572 - val_loss: 0.7595\n",
            "Epoch 215/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7573 - val_loss: 0.7572\n",
            "Epoch 216/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7564 - val_loss: 0.7540\n",
            "Epoch 217/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7550 - val_loss: 0.7503\n",
            "Epoch 218/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7535 - val_loss: 0.7466\n",
            "Epoch 219/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7524 - val_loss: 0.7432\n",
            "Epoch 220/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7515 - val_loss: 0.7401\n",
            "Epoch 221/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7504 - val_loss: 0.7365\n",
            "Epoch 222/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7502 - val_loss: 0.7341\n",
            "Epoch 223/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7495 - val_loss: 0.7314\n",
            "Epoch 224/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7484 - val_loss: 0.7284\n",
            "Epoch 225/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7471 - val_loss: 0.7262\n",
            "Epoch 226/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7456 - val_loss: 0.7223\n",
            "Epoch 227/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7445 - val_loss: 0.7191\n",
            "Epoch 228/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7439 - val_loss: 0.7172\n",
            "Epoch 229/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7436 - val_loss: 0.7147\n",
            "Epoch 230/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7433 - val_loss: 0.7123\n",
            "Epoch 231/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7431 - val_loss: 0.7110\n",
            "Epoch 232/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7420 - val_loss: 0.7085\n",
            "Epoch 233/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7404 - val_loss: 0.7046\n",
            "Epoch 234/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7394 - val_loss: 0.7020\n",
            "Epoch 235/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7391 - val_loss: 0.6995\n",
            "Epoch 236/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7391 - val_loss: 0.6983\n",
            "Epoch 237/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7385 - val_loss: 0.6963\n",
            "Epoch 238/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7372 - val_loss: 0.6936\n",
            "Epoch 239/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7359 - val_loss: 0.6900\n",
            "Epoch 240/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7354 - val_loss: 0.6880\n",
            "Epoch 241/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7356 - val_loss: 0.6868\n",
            "Epoch 242/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7350 - val_loss: 0.6849\n",
            "Epoch 243/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7340 - val_loss: 0.6825\n",
            "Epoch 244/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7329 - val_loss: 0.6800\n",
            "Epoch 245/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7321 - val_loss: 0.6777\n",
            "Epoch 246/500\n",
            "8/8 [==============================] - 0s 9ms/step - loss: 0.7312 - val_loss: 0.6751\n",
            "Epoch 247/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7313 - val_loss: 0.6735\n",
            "Epoch 248/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7312 - val_loss: 0.6719\n",
            "Epoch 249/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7308 - val_loss: 0.6699\n",
            "Epoch 250/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7301 - val_loss: 0.6677\n",
            "Epoch 251/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7296 - val_loss: 0.6657\n",
            "Epoch 252/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7288 - val_loss: 0.6635\n",
            "Epoch 253/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7278 - val_loss: 0.6610\n",
            "Epoch 254/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7275 - val_loss: 0.6594\n",
            "Epoch 255/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7272 - val_loss: 0.6577\n",
            "Epoch 256/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7266 - val_loss: 0.6559\n",
            "Epoch 257/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7262 - val_loss: 0.6542\n",
            "Epoch 258/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7256 - val_loss: 0.6523\n",
            "Epoch 259/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7251 - val_loss: 0.6510\n",
            "Epoch 260/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7240 - val_loss: 0.6483\n",
            "Epoch 261/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7235 - val_loss: 0.6464\n",
            "Epoch 262/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7227 - val_loss: 0.6436\n",
            "Epoch 263/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7225 - val_loss: 0.6434\n",
            "Epoch 264/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7204 - val_loss: 0.6404\n",
            "Epoch 265/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7192 - val_loss: 0.6383\n",
            "Epoch 266/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7186 - val_loss: 0.6364\n",
            "Epoch 267/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7177 - val_loss: 0.6338\n",
            "Epoch 268/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7187 - val_loss: 0.6334\n",
            "Epoch 269/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7181 - val_loss: 0.6323\n",
            "Epoch 270/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7178 - val_loss: 0.6301\n",
            "Epoch 271/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7175 - val_loss: 0.6304\n",
            "Epoch 272/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7160 - val_loss: 0.6281\n",
            "Epoch 273/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7153 - val_loss: 0.6247\n",
            "Epoch 274/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7160 - val_loss: 0.6253\n",
            "Epoch 275/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7151 - val_loss: 0.6236\n",
            "Epoch 276/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7149 - val_loss: 0.6219\n",
            "Epoch 277/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7142 - val_loss: 0.6212\n",
            "Epoch 278/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7135 - val_loss: 0.6196\n",
            "Epoch 279/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7129 - val_loss: 0.6176\n",
            "Epoch 280/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7131 - val_loss: 0.6166\n",
            "Epoch 281/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7128 - val_loss: 0.6156\n",
            "Epoch 282/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7117 - val_loss: 0.6146\n",
            "Epoch 283/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7112 - val_loss: 0.6127\n",
            "Epoch 284/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7112 - val_loss: 0.6111\n",
            "Epoch 285/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7114 - val_loss: 0.6108\n",
            "Epoch 286/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7103 - val_loss: 0.6093\n",
            "Epoch 287/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7096 - val_loss: 0.6076\n",
            "Epoch 288/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 0.7095 - val_loss: 0.6069\n",
            "Epoch 289/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7087 - val_loss: 0.6056\n",
            "Epoch 290/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7083 - val_loss: 0.6037\n",
            "Epoch 291/500\n",
            "8/8 [==============================] - 0s 9ms/step - loss: 0.7084 - val_loss: 0.6031\n",
            "Epoch 292/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7076 - val_loss: 0.6016\n",
            "Epoch 293/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7075 - val_loss: 0.6010\n",
            "Epoch 294/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7070 - val_loss: 0.5997\n",
            "Epoch 295/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7066 - val_loss: 0.5979\n",
            "Epoch 296/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7068 - val_loss: 0.5971\n",
            "Epoch 297/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7062 - val_loss: 0.5970\n",
            "Epoch 298/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7048 - val_loss: 0.5952\n",
            "Epoch 299/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7042 - val_loss: 0.5926\n",
            "Epoch 300/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7052 - val_loss: 0.5923\n",
            "Epoch 301/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7051 - val_loss: 0.5925\n",
            "Epoch 302/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7039 - val_loss: 0.5909\n",
            "Epoch 303/500\n",
            "8/8 [==============================] - 0s 9ms/step - loss: 0.7033 - val_loss: 0.5895\n",
            "Epoch 304/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7028 - val_loss: 0.5880\n",
            "Epoch 305/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7026 - val_loss: 0.5867\n",
            "Epoch 306/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7028 - val_loss: 0.5868\n",
            "Epoch 307/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.7019 - val_loss: 0.5854\n",
            "Epoch 308/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7016 - val_loss: 0.5841\n",
            "Epoch 309/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7012 - val_loss: 0.5835\n",
            "Epoch 310/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7005 - val_loss: 0.5819\n",
            "Epoch 311/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.7004 - val_loss: 0.5808\n",
            "Epoch 312/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.7002 - val_loss: 0.5801\n",
            "Epoch 313/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6997 - val_loss: 0.5798\n",
            "Epoch 314/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6989 - val_loss: 0.5780\n",
            "Epoch 315/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6989 - val_loss: 0.5768\n",
            "Epoch 316/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6989 - val_loss: 0.5764\n",
            "Epoch 317/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 0.6983 - val_loss: 0.5758\n",
            "Epoch 318/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6975 - val_loss: 0.5742\n",
            "Epoch 319/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6973 - val_loss: 0.5726\n",
            "Epoch 320/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6975 - val_loss: 0.5724\n",
            "Epoch 321/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6970 - val_loss: 0.5720\n",
            "Epoch 322/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6961 - val_loss: 0.5704\n",
            "Epoch 323/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6961 - val_loss: 0.5691\n",
            "Epoch 324/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6962 - val_loss: 0.5688\n",
            "Epoch 325/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6953 - val_loss: 0.5680\n",
            "Epoch 326/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6947 - val_loss: 0.5673\n",
            "Epoch 327/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6941 - val_loss: 0.5656\n",
            "Epoch 328/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6944 - val_loss: 0.5645\n",
            "Epoch 329/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6945 - val_loss: 0.5643\n",
            "Epoch 330/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6941 - val_loss: 0.5640\n",
            "Epoch 331/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6931 - val_loss: 0.5631\n",
            "Epoch 332/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6920 - val_loss: 0.5616\n",
            "Epoch 333/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6918 - val_loss: 0.5608\n",
            "Epoch 334/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6917 - val_loss: 0.5600\n",
            "Epoch 335/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6916 - val_loss: 0.5593\n",
            "Epoch 336/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6915 - val_loss: 0.5587\n",
            "Epoch 337/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6911 - val_loss: 0.5578\n",
            "Epoch 338/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6904 - val_loss: 0.5566\n",
            "Epoch 339/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6902 - val_loss: 0.5560\n",
            "Epoch 340/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6900 - val_loss: 0.5553\n",
            "Epoch 341/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6896 - val_loss: 0.5545\n",
            "Epoch 342/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6894 - val_loss: 0.5538\n",
            "Epoch 343/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6890 - val_loss: 0.5530\n",
            "Epoch 344/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6885 - val_loss: 0.5521\n",
            "Epoch 345/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6880 - val_loss: 0.5510\n",
            "Epoch 346/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6879 - val_loss: 0.5505\n",
            "Epoch 347/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6879 - val_loss: 0.5500\n",
            "Epoch 348/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6875 - val_loss: 0.5493\n",
            "Epoch 349/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6870 - val_loss: 0.5485\n",
            "Epoch 350/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6866 - val_loss: 0.5476\n",
            "Epoch 351/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6860 - val_loss: 0.5465\n",
            "Epoch 352/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6860 - val_loss: 0.5460\n",
            "Epoch 353/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6860 - val_loss: 0.5456\n",
            "Epoch 354/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6856 - val_loss: 0.5449\n",
            "Epoch 355/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6851 - val_loss: 0.5441\n",
            "Epoch 356/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6847 - val_loss: 0.5433\n",
            "Epoch 357/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6844 - val_loss: 0.5426\n",
            "Epoch 358/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6840 - val_loss: 0.5417\n",
            "Epoch 359/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6839 - val_loss: 0.5412\n",
            "Epoch 360/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6837 - val_loss: 0.5407\n",
            "Epoch 361/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6833 - val_loss: 0.5400\n",
            "Epoch 362/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6829 - val_loss: 0.5393\n",
            "Epoch 363/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6827 - val_loss: 0.5386\n",
            "Epoch 364/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6823 - val_loss: 0.5379\n",
            "Epoch 365/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6817 - val_loss: 0.5369\n",
            "Epoch 366/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6817 - val_loss: 0.5365\n",
            "Epoch 367/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 0.6815 - val_loss: 0.5360\n",
            "Epoch 368/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6812 - val_loss: 0.5354\n",
            "Epoch 369/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6810 - val_loss: 0.5348\n",
            "Epoch 370/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6805 - val_loss: 0.5341\n",
            "Epoch 371/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6799 - val_loss: 0.5331\n",
            "Epoch 372/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6799 - val_loss: 0.5327\n",
            "Epoch 373/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6797 - val_loss: 0.5322\n",
            "Epoch 374/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6795 - val_loss: 0.5317\n",
            "Epoch 375/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6791 - val_loss: 0.5311\n",
            "Epoch 376/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6787 - val_loss: 0.5304\n",
            "Epoch 377/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6784 - val_loss: 0.5297\n",
            "Epoch 378/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6778 - val_loss: 0.5288\n",
            "Epoch 379/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6780 - val_loss: 0.5285\n",
            "Epoch 380/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6778 - val_loss: 0.5281\n",
            "Epoch 381/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6774 - val_loss: 0.5275\n",
            "Epoch 382/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6770 - val_loss: 0.5268\n",
            "Epoch 383/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6767 - val_loss: 0.5262\n",
            "Epoch 384/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6763 - val_loss: 0.5248\n",
            "Epoch 385/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6768 - val_loss: 0.5245\n",
            "Epoch 386/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6770 - val_loss: 0.5246\n",
            "Epoch 387/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6764 - val_loss: 0.5244\n",
            "Epoch 388/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6754 - val_loss: 0.5237\n",
            "Epoch 389/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6748 - val_loss: 0.5223\n",
            "Epoch 390/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6749 - val_loss: 0.5216\n",
            "Epoch 391/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6749 - val_loss: 0.5213\n",
            "Epoch 392/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6748 - val_loss: 0.5213\n",
            "Epoch 393/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6742 - val_loss: 0.5209\n",
            "Epoch 394/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6737 - val_loss: 0.5197\n",
            "Epoch 395/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6738 - val_loss: 0.5190\n",
            "Epoch 396/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6738 - val_loss: 0.5189\n",
            "Epoch 397/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6734 - val_loss: 0.5187\n",
            "Epoch 398/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6724 - val_loss: 0.5179\n",
            "Epoch 399/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6722 - val_loss: 0.5168\n",
            "Epoch 400/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6725 - val_loss: 0.5163\n",
            "Epoch 401/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6726 - val_loss: 0.5164\n",
            "Epoch 402/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6721 - val_loss: 0.5162\n",
            "Epoch 403/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6713 - val_loss: 0.5150\n",
            "Epoch 404/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6712 - val_loss: 0.5141\n",
            "Epoch 405/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6714 - val_loss: 0.5143\n",
            "Epoch 406/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6710 - val_loss: 0.5141\n",
            "Epoch 407/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6703 - val_loss: 0.5130\n",
            "Epoch 408/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6703 - val_loss: 0.5124\n",
            "Epoch 409/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6704 - val_loss: 0.5124\n",
            "Epoch 410/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6699 - val_loss: 0.5122\n",
            "Epoch 411/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6690 - val_loss: 0.5107\n",
            "Epoch 412/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6692 - val_loss: 0.5103\n",
            "Epoch 413/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6694 - val_loss: 0.5104\n",
            "Epoch 414/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6691 - val_loss: 0.5104\n",
            "Epoch 415/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6683 - val_loss: 0.5092\n",
            "Epoch 416/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6682 - val_loss: 0.5086\n",
            "Epoch 417/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6682 - val_loss: 0.5085\n",
            "Epoch 418/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6677 - val_loss: 0.5081\n",
            "Epoch 419/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6672 - val_loss: 0.5072\n",
            "Epoch 420/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6673 - val_loss: 0.5068\n",
            "Epoch 421/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6674 - val_loss: 0.5068\n",
            "Epoch 422/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6669 - val_loss: 0.5060\n",
            "Epoch 423/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6669 - val_loss: 0.5057\n",
            "Epoch 424/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6666 - val_loss: 0.5056\n",
            "Epoch 425/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6657 - val_loss: 0.5044\n",
            "Epoch 426/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6659 - val_loss: 0.5041\n",
            "Epoch 427/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6658 - val_loss: 0.5042\n",
            "Epoch 428/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6654 - val_loss: 0.5034\n",
            "Epoch 429/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6653 - val_loss: 0.5030\n",
            "Epoch 430/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6651 - val_loss: 0.5029\n",
            "Epoch 431/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6643 - val_loss: 0.5018\n",
            "Epoch 432/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6644 - val_loss: 0.5015\n",
            "Epoch 433/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6645 - val_loss: 0.5017\n",
            "Epoch 434/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6640 - val_loss: 0.5009\n",
            "Epoch 435/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6638 - val_loss: 0.5005\n",
            "Epoch 436/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6636 - val_loss: 0.5004\n",
            "Epoch 437/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6631 - val_loss: 0.4996\n",
            "Epoch 438/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 0.6627 - val_loss: 0.4988\n",
            "Epoch 439/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6628 - val_loss: 0.4990\n",
            "Epoch 440/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6624 - val_loss: 0.4983\n",
            "Epoch 441/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6624 - val_loss: 0.4979\n",
            "Epoch 442/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6623 - val_loss: 0.4980\n",
            "Epoch 443/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6617 - val_loss: 0.4971\n",
            "Epoch 444/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6614 - val_loss: 0.4965\n",
            "Epoch 445/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6615 - val_loss: 0.4966\n",
            "Epoch 446/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6611 - val_loss: 0.4959\n",
            "Epoch 447/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6610 - val_loss: 0.4956\n",
            "Epoch 448/500\n",
            "8/8 [==============================] - 0s 9ms/step - loss: 0.6610 - val_loss: 0.4952\n",
            "Epoch 449/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6610 - val_loss: 0.4974\n",
            "Epoch 450/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6585 - val_loss: 0.4969\n",
            "Epoch 451/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6570 - val_loss: 0.4940\n",
            "Epoch 452/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6592 - val_loss: 0.4934\n",
            "Epoch 453/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6622 - val_loss: 0.4949\n",
            "Epoch 454/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6636 - val_loss: 0.4963\n",
            "Epoch 455/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6631 - val_loss: 0.4966\n",
            "Epoch 456/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6616 - val_loss: 0.4960\n",
            "Epoch 457/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6598 - val_loss: 0.4946\n",
            "Epoch 458/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6589 - val_loss: 0.4937\n",
            "Epoch 459/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6585 - val_loss: 0.4931\n",
            "Epoch 460/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6587 - val_loss: 0.4929\n",
            "Epoch 461/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6588 - val_loss: 0.4928\n",
            "Epoch 462/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6589 - val_loss: 0.4928\n",
            "Epoch 463/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 0.6584 - val_loss: 0.4923\n",
            "Epoch 464/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6583 - val_loss: 0.4921\n",
            "Epoch 465/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6580 - val_loss: 0.4918\n",
            "Epoch 466/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6576 - val_loss: 0.4915\n",
            "Epoch 467/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6573 - val_loss: 0.4911\n",
            "Epoch 468/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6569 - val_loss: 0.4907\n",
            "Epoch 469/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6567 - val_loss: 0.4904\n",
            "Epoch 470/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6564 - val_loss: 0.4899\n",
            "Epoch 471/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6565 - val_loss: 0.4898\n",
            "Epoch 472/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6562 - val_loss: 0.4897\n",
            "Epoch 473/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6557 - val_loss: 0.4894\n",
            "Epoch 474/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6555 - val_loss: 0.4891\n",
            "Epoch 475/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6553 - val_loss: 0.4888\n",
            "Epoch 476/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6549 - val_loss: 0.4884\n",
            "Epoch 477/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6546 - val_loss: 0.4879\n",
            "Epoch 478/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6549 - val_loss: 0.4880\n",
            "Epoch 479/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6548 - val_loss: 0.4879\n",
            "Epoch 480/500\n",
            "8/8 [==============================] - 0s 5ms/step - loss: 0.6544 - val_loss: 0.4876\n",
            "Epoch 481/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6541 - val_loss: 0.4873\n",
            "Epoch 482/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6539 - val_loss: 0.4860\n",
            "Epoch 483/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6547 - val_loss: 0.4861\n",
            "Epoch 484/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6554 - val_loss: 0.4869\n",
            "Epoch 485/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6544 - val_loss: 0.4867\n",
            "Epoch 486/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6531 - val_loss: 0.4862\n",
            "Epoch 487/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6519 - val_loss: 0.4844\n",
            "Epoch 488/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6526 - val_loss: 0.4841\n",
            "Epoch 489/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6535 - val_loss: 0.4849\n",
            "Epoch 490/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6531 - val_loss: 0.4853\n",
            "Epoch 491/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6522 - val_loss: 0.4839\n",
            "Epoch 492/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6522 - val_loss: 0.4833\n",
            "Epoch 493/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6526 - val_loss: 0.4839\n",
            "Epoch 494/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6521 - val_loss: 0.4841\n",
            "Epoch 495/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6511 - val_loss: 0.4826\n",
            "Epoch 496/500\n",
            "8/8 [==============================] - 0s 8ms/step - loss: 0.6514 - val_loss: 0.4824\n",
            "Epoch 497/500\n",
            "8/8 [==============================] - 0s 6ms/step - loss: 0.6518 - val_loss: 0.4829\n",
            "Epoch 498/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6512 - val_loss: 0.4830\n",
            "Epoch 499/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6501 - val_loss: 0.4812\n",
            "Epoch 500/500\n",
            "8/8 [==============================] - 0s 7ms/step - loss: 0.6506 - val_loss: 0.4812\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluation"
      ],
      "metadata": {
        "id": "ELOt_Djfqk1o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = model.predict(x_test)"
      ],
      "metadata": {
        "id": "sEztfDth9lhP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(np.arange(0, len(y_train)), y_train, 'g', label=\"History\")\n",
        "plt.plot(np.arange(len(y_train), len(y_train) + len(y_test)), y_test, marker='.', label=\"Actual Data\")\n",
        "plt.plot(np.arange(len(y_train), len(y_train) + len(y_test)), y_pred, 'r', label=\"Prediction\")\n",
        "plt.ylabel('Retail Sales')\n",
        "plt.xlabel('Period')\n",
        "plt.legend()\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "EoYKHOKMqpnW",
        "outputId": "c5228e1d-a5a1-4d05-be51-a951e49a2fd1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAyEAAAItCAYAAAAuW3U8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXgUVbr48W8v2RMSEsIqAiKUGEEURMQFQQYU2ZVRVHC7Oo5XXNBxZMZRxAWvOIviONefynUQN4SwqygwIIiAhkUIUKCAQNgCWcja+++PTleql3R39nTyfp7Hx+qqU9Wnm07Sb73nvMfgcrkQQgghhBBCiIZibOwOCCGEEEIIIVoWCUKEEEIIIYQQDUqCECGEEEIIIUSDkiBECCGEEEII0aAkCBFCCCGEEEI0KAlChBBCCCGEEA1KghAhhBBCCCFEg5IgRAghhBBCCNGgJAgRQgghhBBCNCgJQoQQQgghhBANSoIQIYQQQgghRIOSIEQIIYQQQgjRoMyN3QFRt7KysrYD3YBi4OdG7o4QQgghhGi+LgQSgUP9+vW7rDonShDS/HQDkiv+69TIfRFCCCGEEM1ft+qeIEFI81MMJBuNRuLj4xv2iYuLAUhMTGzQ5xWiIcnnXLQE8jkXLYF8zmuvtLQUp9MJ7u+f1SJBSPPzM9ApPj4eRVEa9ImzsrIAGvx5hWhI8jkXLYF8zkVLIJ/z2lNV1RPMVXsKgExMF0IIIYQQQjQoCUKEEEIIIYQQDUqCECGEEEIIIUSDkiBECCGEEEII0aAkCBFCCCGEEEI0KAlChBBCCCGEEA1KghAhhBBCCCFEg5IgRAghhBBCCNGgJAgRQgghhBBCNCgJQoQQQgghhBANSoIQIYQQQgghRIOSIEQIIYQQQgjRoMyN3QHRtDmdToqLiykqKqK0tBSHw4HL5Qp6zt69exuod0I0HvmcN20GgwGz2UyrVq1ITk4mJiamsbskhBD1otzqZFN2IafyrbRPjWZQRjIxUU0/zyBBiKhSaWkpR48exel0htU+Nja2nnskROOTz3lkcLlc2Gw2zp49y9mzZ0lNTaVdu3aN3S0hhKhT6tFSZsw7REGxXduXkmhmxpRuKJ3jG7FnoUkQIgIqLS3lyJEjuFwuYmNjadWqFUlJSZjNZozGwNF1SUkJAAkJCQ3ZVSEalHzOI4PT6aS8vJzCwkIKCwvJy8sjKiqK1NTUxu6aEELUCYvN6ReAABQU25kx7xAfPN2rSWdEmm7PRKNxOp0cPXoUl8tFcnIyXbt2JS0tjejo6CoDECGEaEqMRiPx8fF06NCBDh06AHDq1CmsVmsj90wIIerGpuxCvwDEo6DYzqbswgbuUfXIN0rhp7i4GKfTSWxsLB06dMBgMDR2l4QQosaSk5NJSkoCKjNZQggR6U7mBb+pEup4Y5MgRPgpKioCoFWrVhKACCGahcTERMB9k0UIIZqD9qnRtTre2CQIEX5KS0sBtDuHQggR6TxzeMrKyhq5J0IIUTcGZSSTkhh4endKoplBGckN3KPqkSBE+HE4HACYzVK3QAjRPJhMJoCwq/0JIURTFxNl5Jnbz/fb76mO1ZQnpYNUxxIBeNYBkUnoQojmwjO0NNQ6R0IIEUnOb1tZNj422sij48+TdUKEEEKIpkLmtwkhmiOLrTK7m5xgZkjf1o3Ym+pp+mGSEEIIIYQQwo/FVpndjY6KrJstEoQIIYQQQggRgfSZkNgIGIKlF1m9FUIIIYQQQgDeQUgkzAPRi6zeCiGEEEIIIQCwWCuDEBmOJYQQQgghhKh3kZwJkepYQoiwKIqibauq2og9EUIIIQR4T0yPtCAksnorhKjS5MmTURQFRVGYM2dOtc595plntHOfeeaZeuqhEEIIIeqSxeYk3llG/9Jskp3Fjd2dapEgRAjRpOiDqS1btjR2d4QQQogmy2Jzclf+l9xR+DVXbPq/iFqQVYIQIYQQQgghIpDF5qSbLQeA+OIzuCyWRu5R+GROiBAiLDIPRAghhGhaLFYnUS6H9tgQFdWIvakeyYQIIYQQQggRgaxWO0bcQ7BcBiMGk6mRexQ+CUKEEEIIIYSIQLYyq7btMkXWAKfI6q0QotGEW6L37NmzLFy4kI0bN3Lw4EEKCwsBSEhIoH379vTs2ZMBAwYwbNgwWrdurZ03dOhQcnJyvK41ZcqUgM8xa9YsJkyYEPBYdnY2ixcvZsuWLZw6dYqysjJSU1Pp2bMnQ4cOZcKECcTExAR9rZMnT2br1q0AzJs3jyuvvJITJ06wcOFCVq9ezalTpygoKEBRFBYuXMj111/PmTNnAFi4cCG9e/cOen0Ap9PJ0KFDOXHiBADz58/niiuuCHmeEEII4eGwVs4BcZkjZygWSBAihKhDX331Fc8++yxFRUV+xwoKCigoKGDfvn0sW7aMDRs28Oabb9bZc1utVmbOnMnChQv9qoOcPHmSkydP8u233/Luu+/y2muv0b9//7CvvWrVKv70pz9RXOxf/jAqKopx48bx3nvvAbBo0aKwgpDvvvtOC0C6du0qAYgQQohqs5dXZkKQIEQI0RL99NNPTJs2DYfDPUEuLi6Oyy67jA4dOmA2mykuLubgwYP8/PPP2Gw2v/PHjRtHQUEB33zzDadPnwZg2LBhtGvXzq9t9+7dvR47HA5+//vfs3HjRm1feno6/fv3JyEhgV9//ZWsrCycTic5OTncf//9vPPOOwwcODDk68rKyuKf//wndrudtLQ0+vbtS6tWrcjPz9eyPBMnTtSCkJUrVzJ9+vSQ2ZZFixZp21VldYQQQohg7FZ75YOoyPpaH1m9FUI0Wf/617+0AGTkyJHMnDmTpKQkv3bFxcWsXbuW48ePe+1/9NFHAThw4IAWhEyZMoUrr7wy5HO/8847WgBiMBiYNm0a999/PybdBL1ffvmFxx9/nP3791NeXs4f/vAHli9fTkpKStBrv/XWWzidTp566inuvfdeLBXlDxMSErBa3XegunbtyoABA9i6dSvnzp3j66+/ZvTo0VVes6CggDVr1gBgMpkYN25cyNcohBBC+HJYKzMhBsmERB5FUYYBE4FBQEcgATgFHAe2AGuBNaqqloS4TivgbuBWoAeQCpwG9gKfAh+rqho5BZxFxFq/fj35+flht9+5c2etn3Pbtm0AxMTE8PLLLxMfHx+wXWJiImPGjKn183mcO3eOd999V3v82GOP8eCDD/q16969O3PnzmXs2LGcPXuW06dPM2/ePC34qYrD4eDJJ5/kgQceANCCEIDo6Ghte+LEido8kkWLFgUNQpYvX64FMNdee23AbI8QQggRilMfhERQeV5o4UGI4p5p+7/A9QEOn1/x30DgMdxBysIg17oG+KjiHL3OFf8NBx5XFOV2VVX31rrzQgSxa9cudu3a1aDPWVLijtFjY2OrDEDqw/LlyyktLQWgU6dO/Nd//VeVbdPT03nkkUd44YUXAFiwYAFTp07FYDBUeU779u257777QvZjxIgRvPTSSxQWFrJ582ZycnLo1KlTwLb6oVi33npryGsLIYQQgTitlcObjbobY5GgxQYhiqIMAL4Gkit2uYBtwAGgsGL/RUBvIGjRZUVRLge+wp1BAbACa3BnUi4ABuMuh9wHWKMoygBVVY/V5etpbKtPfc3K48uwOiXRE2OMYWTHMQxrN7yxu9Kg2rdvz9GjRyksLOTLL7/kpptuapDn3bJli7Y9evRookLcCRozZgyvvPIKNpuN3NxcDh486DfHRG/48OGYzaF/VcbExDBmzBg+/PBDXC4XmZmZTJ061a/dnj172LvXfR8iLS2N66+/PuS1hRBCiED0mRCjZEKaPkVRLgRWURmAfAI8o6rqkQBtU4AxwOEqrhUNLKIyAMkCxumDDEVRLgaW4w5IOuDOmAyui9fSVKw99bUEIBUsTgtrT33dqEHII488EvALcFWeeeYZFi9eXKvnHDFihDY5e9q0aXz55ZfcfPPNDBw4kOTk5BBn19y+ffu07b59+4Zsn5iYSI8ePdizZw/gDgqCBSEZGRlh92XixIl8+OGHACxevJhHHnnEL8uiz4KMHTs2ZNAkhBBCVMVlq5yYboqOrL8nLW6xQkVRDMC7gGc26quqqt4RKAABUFW1QFXVeaqq/ljFJX8HdK3YPguM9M1yqKq6BxgNeL6lX6coSsPcJm4gQ9sNJ9oYvBpQSxFjjGFoC8uCADz88MP06dMHcK+BsWrVKh599FGuvPJKRo8ezcyZM1m7dq02F6Ku5OXladsdOnQI65yOHTtq26HmzujXMglFURQuvfRSAHJycvj++++9jlutVlasWKE9vuWWW8K+thBCCOHLZa8cjmWS4VhN3gQq54D8APy5ltd7WLc9W1XV04Eaqaq6R1GUD3AHLZ7zvqzlczcZw9oN56rEqwF31SDR8iQkJPDRRx/xwQcf8PHHH2trYLhcLvbv38/+/fv56KOPSEtL47//+7+54447gs7FCJdnPggQ9lyUuLg4bdszl6UqsbGx1erPxIkTtYn+ixYtYtCgQdqx1atXU1BQALizNhdeeGG1ri2EEEJ40QchMZIJaep+p9t+XVVVZ00vpChKT9zzRjz+HeIU/fFhiqIk1vS5hWiKoqOjefDBB1m7di2LFi1i+vTpjBgxgrS0NK3N2bNnmTlzJs8//3ydPKc+8NAHJMGUlZVp23UdNI8cOVLr0+rVq70WbpQJ6UIIIeqK0+nCoAtCzDGRlQlpUUGIoijtgWEVD23AslpecohuW1VV9WSI9j8AntuuscBVtXx+IZoko9HIJZdcwj333MObb77Jd999xyeffMKwYcO0Np999lmdlAZOTU3Vtj3Zl1D0a5RUZ7hVOBISEhg1ahQA5eXlLF++HHCv2r5p0ybAHTg11MR9IYQQzZPV7iIKh/bYEGHDsVpUEIK73K5n/MceVVXLFUVJVBTlCUVRNiuKclZRlFJFUQ4rivKZoijjQ1yvl257W6gnV1XVDujrpvaqqq0QzYnBYODyyy/nrbfe4pprrtH2r1u3rtbXvuiiymTkjh07QrYvLi7mwIED2uOLL7641n3wNXHiRG3bk/3IzMzE6XQnXm+88UYSEyURKoQQouYsNidmV+XE9EirjtXSgpD+uu2jiqJcCuwA/gZciXtxwTigC/BbIFNRlI2KolS1klhP3XbAie0BHNVtK2GeI0SzYDAYvErSnjlzxq9NTExlgQO73e533Jd+RfXly5djs9mCtIYVK1ZobdLT07ngggtCPkd19enTRwuOdu/ezb59+7yqj8mEdCGEELVlsTmJcukyIRKENGmdddvJuNf26I57jZCtwFzc5XMP6dpdDXxXUarXV5pu+1SYfdAP2UqtspUQEcRqtYac4O1x8mTlj4B+KJVHSkrlj9qpU6F/rMaMGaPNwcjJydHKBAdy9uxZ5syZoz3+7W9/WyeT4wPRZ0Oee+45jhxx36fo2rUr/fv3r+o0IYQQIiwWq3cmJNKCkJZWHUsfSFxb8f+TwC2qqm7yHKgo4/sQMAf3QoXdgbeAu3yupx9PUUZ49O3qbTxGcXExWVlZNT4/NjY27C+Vvmp6nqgdh6Pybkh1ggLwzjjY7faQ5/oeP378OJMmTWLcuHGMGDEi4BAnp9PJmjVrmD9/vrZvwIABftfq2rWrtr1y5UqGDx8eNFAwGo3cc889vP322wC88cYbOJ1OJk+ejMlUuc7ooUOHeOaZZ7TsS3p6OrfeemvA16p/L8vLywO2CfUeDRs2jNmzZ1NeXu4192X06NHyM9JInE4n5eXltfrd2NLIeyVagkj9nB/PxysTknPqFL9G0GtpaUGIbxkcO3Czqqpe8zlUVXUB/1IUJR54vWL3JEVRZqiq+rOuqb52Z7iLH+hX9IurspUQEaaoqIgPP/yQDz/8kJSUFC666CLS09Mxm83k5eWxb98+r8zG8OHDAy4uOHToUN566y1cLhcbN27ktttuo0+fPl5VrHwDnXvvvZdt27axefNmXC4Xb775Jh9//DH9+vUjPj6eI0eOsGPHDi24iI2N5aWXXqrXRRSTkpK44YYbWLlypbbPZDJpk9aFEEKI2rA5IArdsGXdjbdI0NKCkHKfx5/5BiA+3gCmAR1xD12bCMyq4nrhliTQr+gXbvak2hITE1GUmk052bt3L1D90qWeu7uyTkjj0N/1j46Orta/g9ls9toOda7v8VatWhEdHa0tRFhQUMDmzZsDnmswGLj11lt57rnniA5QyePiiy/mwQcf5J133gHg559/5ueff/Zrc8UVV3jte/fdd5kxYwaZmZm4XC7OnDnDqlWr/K7fsWNHZs+eHXRIlP69jI2N9Xq91fmcT5o0ySsIue6667wyPaJhGY1G4uPj6dVLaoKE4rkz3K9fv0buiRD1J9I/54YDRex2rdYed+nenZQGfi2qqlJcXFyjc1taEOL7Li0N1lhVVbuiKCuBByp2DfJpor9euFkNfbua/asJ0cS0a9eOLVu2sGnTJn788Ueys7P59ddfKSgowOFwkJiYyPnnn0+/fv0YN26cV0WrQKZNm0a/fv1YtGgR2dnZnD171mttj0Cio6N55ZVXuPPOO8nMzGTz5s2cOnWK8vJyWrduTc+ePbnhhhu45ZZbvCa/16crrriClJQUbYFCWRtECCFEXXFXx4rciektLQg56/N4Txjn7NVtd/Q5lqfbrqqCli99u7wqWwlRTR9++GGNz3311Vd59dVXg7ZRVTXo8fj4eIYNG+a1FkhtDB48mMGDB1f7vIyMDDIyMmr13LV5L/X27dunBSBt2rTxqgwmhBBC1Ia7OpaU6I0U+3weh5OJKNJtJ/kc038rOz/MPujbBf9WJ4SIaAsXLtS2x44d6zXsTQghhKgNv0yILFbYpGX7PA6nOpU+8Cj0OabPklwW6kKKopiB3lWcL4RoRoqKiliyZAngngdz2223NXKPhBBCNCcWq8trYnqkDcdqaUHIZqBU9zicpZL1MxiP+hz7j25bCbKooUc/Kit0lQPfh/H8QogI43K5mD17NkVF7kTqkCFD6NKlSyP3SgghRHNitUf2cKwWNTZAVdVSRVFWAeMrdo0BPq+qvaIoJmCkbte3PtfbryjKPuAiwABMAWYH6cLduu01qqrKxHQhmolvv/2WDRs2YLFY2L59O/v37wfc1cYee+yxRu6dEEKI5sZiddI6giemt7RMCMCbuu3bFEXxX6ig0lSgU8W2Bfg0QJu3ddt/UBSlTaALKYpyEXCvbtc/w+irECJC7Ny5k3nz5vHZZ59pAQjA448/HrIamBBCCFFd5bbIXjG9xQUhqqquA1ZUPIwCViqKMlDfRlEUg6IoD+Kd1XhTVdWTAS75DnC4Yjsd+EJRlE76Boqi9Kp4Ts/iht+qqvplbV6HEKLpio+P5/LLL+eNN97ggQceCH2CEEIIUU3u6liVmRBjhE1Mb1HDsXTuBTYBPXCX3d2kKMpWYDfuxQQHARfo2m8A/hzoQqqqWhVFuaWiTTxwBfCLoihrgBNAV2AIlQHfSeCuOn49QohGNnXqVKZOndrY3RBCCNFCWG0yMT3iqKp6BrgBWFexywBcCdyPO0DQByDzgRGqqtqCXG8bcCNwpGJXDO65JPdXPI/nfd4F3KCqqu8EdyGEEEIIIcJWanF4DceyYWrE3lRfiwxCACoCgaG4J6l/DhwCynCvC6LiHmY1UFXVyaqqBl+q2X29DbjL7z4GbMSd8bACOcA3uAOSK1RVDWeBRCGEEEIIIQJSj5ayec85r+FY//XmL6hHS4Oc1bS01OFYAKiq6gKWVPxXF9c7h3vi+5uh2gohhBBCCFFdFpuTGfMO4bA7MOEEwAnkFbv3f/B0L2Kimn6eoen3UAghhBBCCAHApuxCCortXqul2wxmMBgoKLazKdt3be2mSYIQIYQQQgghIsTJPCuA16R0u25wk+d4UydBiBBCCCGEEBGifaq7FK9+UrrdYPI73tRJECKEEEIIIUSEGJSRTEqC2WtSus3gzoSkJJoZlJHcWF2rFglChBBCCCGEiBAxUUYeHtvRLxOSkmhmxpRuETEpHVp4dSwhhBBCCCGainKrk03ZhZzKt9I+NZpBGckBg4rEOLNXEJKWGh8xVbE8JAgRQgghhBCikalHS5kx7xAFxZXBhSe7oXSO92p7ptBGFJXDsVolx0VUAAIShAghhBBCCNEoPJmPnDMWlmzKpbTc6XW8oNgecO2P3AIbUbpMiCEqqsH6XFckCBFCCCGEEKKBBcp8BOJZ+2NI39bavtxCq9dwLKMEIUIIIYQQQohgPKueBw1AXC5+W7iaLrYTFO2dCH0HaofOFNq8qmNFYiYksgaPCSGEEEIIEeE8q54H09V2nIFlu+lgP8v5e9Z4HcsttHllQiQIEUIIIYQQQgQVzqrmHWxnte34wlNex3ILrF4T0yNxOJYEIUIIIYQQQjSgcFY1T3fka9uOc+dwlJYCUFLuoNTijPiJ6RKECCFELUyePBlFUVAUhS1btjR2d4QQQkSAQRnJpCQGn5qdbs/3elyScwJwzwcBIn44lkxMF6KZmzx5Mlu3btUez5kzh+HDhzdij0QwQ4cOJScnJ+CxuLg4EhMTSUpKomPHjmRkZNC7d28GDRpEQkJCA/dUCCFEuAItQjhjSjee/N8DOJyBz/ENQrKzDnFVj+4cP1sOEPET0yUIEaIZO378OD/88IPXviVLljRKEKL/cr1mzRrOO++8Bu9DpCsrK6OsrIzc3FwOHjzIxo0bAUhISGDMmDHcf//9dO7cucH7lZmZyfTp0wEYP348r776aoP3QQghmqpgixBe2DEO9ViZ3zlGl4M0R6HXvtKcE6hHS3l9wVHAOxOSV2agbT31v75IECJEM7Z06VJcLpfXvm+//Zb8/Hxat25dxVmiqbjqqqu44IILtMd2u52ioiIKCwvZu3cveXl5AJSUlPDJJ5+wdOlS/vSnPzFx4sTG6rIQQgidqkrxehYhTGsV+Kt4mqMQE95/v0/9fIy33vuFcqs7dWLWTUxfteMc3W5zRtSq6RKECNGMLV26VNuOiYnBYrFgs9lYuXIld911VyP2TIRjzJgxTJgwocrjhw4dYsGCBXz22WeUlJRQWlrKs88+y5kzZ/j973/fgD0VQggRSLBSvAXFdqJMgc/zHYoF0MaerwUggNfE9HNWo9+Chk1d5IRLQohq2blzJ4cOHQLcw3UeeeQR7Zg+OBGRq1u3bvzxj39k2bJlZGRkaPvfeOMNVq9e3Yg9E0IIAaFL8RaXB54Qkm4vCLxPN7pBH4TYDaawyv42JRKECNFMLVmyRNv+zW9+w6233orZ7E5+/vTTTxw8eLCxuibq2HnnnccHH3ygzbNxuVzMmjULh8MR4kwhhBD1KVQpXp8R05q29jy/fbEuK0nOUu2xWTcx3W4wh1X2tymR4VhCNENWq5UvvvhCezxmzBhSU1O5+uqrWb9+PeDOhjzxxBPVum5BQQGLFy9m48aN/PLLL+Tn5+NyuUhLS6NHjx4MHDiQkSNH0r59e+0cRVH8rnPDDTcEvP68efO48sorAdiyZQtTpkwBYMCAAXz44YdB+3bs2DHtup06dWLt2rVVtv3555/ZsGEDP/74Iz///DOnT5/GZrORlJRE586dGTBgALfddlujTPKuqVatWvHyyy9z9913A+73Y8WKFYwdOzZg+7KyMjZu3MjmzZvJzs7m119/paioiKioKFJTU+nduzfDhg3jpptuwmQKPF5gzpw5vPXWW177Fi9ezOLFi/3aBvo3tNlsfP/993z//ffs3r2bQ4cOUVBQgNlsJiUlhV69ejF48GDGjx9PTExMTd4WIYRoVJ5SvIGGZKUkmnG6XJQHSGCkOyozIU4qswbp9nyKTO5qiPpMSFScu+JWJJEgRIhmaP369RQUuH+Bpaenc9VVVwHuYMQThCxbtozHH38cg8EQ1jU/+OAD5syZQ3Fxsd+x48ePc/z4cdavX8/rr7/Of/7zH9q1a1dHr6ZuPfbYY3z11VcBj+Xl5ZGXl8fOnTv5v//7Px555JGImlsxcOBAevXqxd69ewH48ssvAwYhO3fu5J577qG0tNTvmM1mo7S0lGPHjvHll1/yzjvv8M9//pPzzz+/Tvt64sQJxo0bp31OfftQVlbGiRMnWLt2Lf/7v//LnDlz6N27d532QQgh6ltMlJEZU7ox/f1fKLNUDr3yVMd6af6hgOfp54QciepAV5t7jZB0Rz4HcWe99RPTbx92XkRNSgcJQkQzEaj+dqT9MNYl/VCsUaNGYTS634thw4aRkJBASUkJx48fZ+vWrVrmIZi//OUvLFiwQHscFRXFpZdeSufOnTGZTJw5c4bdu3dz5swZHA4HNptNa3vnnXdqfSopKQFg3LhxAde1aIjA5cSJE9pr6N69O126dKFVq1YYDAZyc3P56aefOHv2LHa7nX/84x8YDAYeeuiheu9XXRkxYoQWhGzfvh2Xy+UXaBYWFmoBSJs2bejRowft2rUjNjaWsrIyDh48SHZ2Nk6nk/3793PnnXeydOlSUlNTva7Tp08f7rzzTg4ePMj3338PwAUXXKAFvXpdunTxelxaWqoFICkpKVx44YV07NiR+Ph4LBYLR44c4aeffsJms3HixAmmTJlCZmYm3bp1q5s3SgghGojSOZ7brm/LB6tOApCaZGbuH3oRE2XEEWA4VqzTQrLT/ffSjpH9MedrQUhbXXASQ2UmpHPHxHp8BfVDghAR8YLV31Y6xzdizxpHfn6+lu0Ad/bDIzY2luHDh2vDZZYsWRIyCJk3b55XADJhwgSeeuop0tLSvNq5XC62b9/O/PnzvYbvPPfccwCsW7dOC0KmTp3aaOuEDBw4kPvuu49rrrmGxET/X9oOh4Ply5fzwgsvUFpaypw5cxg5cmSdZwLqiz5bUFBQwOHDh/2+uCcnJ/PQQw8xatQoevToEfA6OTk5vPDCC6xfv57Tp08ze/ZsZs2a5dVm8ODBDB48mMzMTC0IufTSS7V/82BiY2OZPHkyY8eOJSMjQwuU9fLy8njttddYvHgxpaWlPP/888ybNy/ktYUQoqlxOqGb9RgdbGc51qqPdqPU4ayMQh4Z24lzpQ665O2DFe591tT2nBPfWo0AACAASURBVDG2hYpBCF2tx90bLhcphnLt3EhcrLDl3ioWzUKo+tsWWxXLkDZjX3zxhZaJuPDCC7n44ou9juuDklWrVlFeXk5V8vPz+fvf/649vu+++5g1a5ZfAAJgMBi4/PLL+dvf/kaHDh1q+zLqzbRp07jxxhsDBiAAJpOJcePG8corrwDutTk+/fTThuxirfgGHGfPnvVrc+mll/LEE09UGYCAe17N22+/rc3pWbFiRcChUzXVqVMnnn32WXr37h0wAAFITU3l1VdfZfDgwYB7ntDPP/9cZ30QQoiG4irI4+GzC7n13FquPL1J2+/QpUKuuzSFSUPb0bWw8vdc56v7Mf1PlfMoz7edJNZpYWDZblLLz7h3GgxEt2lT/y+ijkkmRNSJRRtOM3/1ScqtVZR5aAQFxXbGPberwZ83LtrIncPaccu1jbN2qX4olj7g8Bg4cCBt27bl9OnTlJSU8M033zB69OiA11qwYIE2bKdr165MmzatfjrdBI0YMYL4+HhKS0u1u/yRICkpyetxYWFhFS1DM5vNjB49GlVVsVqtZGVlVVlUoD6NGzdOy+5t2rSJCy+8sMH7IIQQtdF2z7fa4oNX5W8FHgbAritiaDYacDmdFO/cqe1LuvRS4lJTOBPfjjalpzDhYkBZNjcVVQYyqb/5DeaUlAZ5HXVJghBRJzI35DapAKQxlVmdZG7IbZQg5NChQ/z000+AOzMRKLgwGo2MGjWKuXPnAu4qWVUFIRs2bNC2b7/9dqIiMN0bzC+//MLu3bs5evQoxcXFWK3eJUo8cyn279+P0+ms8o59UxIf7z0E0TMEriolJSXs2LGDAwcOkJ+fT0lJCU5nZQZRX8p579699RKEWK1WduzYgaqqnDlzxq8Pp06d8uqDEEJEGpc9cMl0/XAsk9FA+eHDOM6dcz9OSiL2ggsAyG97IW0Ou38XjjtXOeQ6ukMH0m+9tb66Xa8kCBF1YsK16U0uE9JY4qKNTLg2vVGeW58F6d+/Px07dgzYbuzYsVoQsmnTJnJzc0lP9++zJ6ABd4nV5mL16tW8+eabqKoaVnu73U5RURHJyU2//KFv0FHVsDPPULvly5cHrJIVSF0OxwJ3meC3336bzz77LOyMTV33QQghGoLLEXjVdK8gxGSgUJcFSezTB0PFzS9LFwUOf+d9TYOBjg88gDE6stYH8ZAgRNSJW65ty42Xu6sdBap6VF8sNif3vLa3yvrbHzzdq8VUyXK5XCxfvlx7HGgolsdFF11Ez5492b9/vzYR+7777vNqU1xcjMVi0R5H0poZwfzjH//gX//6V7XPKykpiYggpKioyOtxq1at/NocO3aMu+66S6sUFq5QWZXqKCgoYMqUKWEHgvXRByGEaCiuAIvHOpwubbFCg8GdCSnesUM7nti3r7Ydc8GFONYbtCFdALZLryKuIlMSiSQIERHNU3+7qupYLSUAAdi6dSs5OTkAREdHc+ONNwZtP2bMGF5//XXAnUHxDUJ8v+z5DvOJRBs2bPAKQC6//HImTJhA7969ad++PfHx8UTr7igNHTpUe0/1w4OaMv3wKSBghuupp57SApCEhARuu+02rrnmGrp27UpqaioxMTHa0LPMzEymT58OuAPdujJz5kwtAImOjmb8+PEMHTqUCy64gLS0NGJjY7Uqa/qFK+uyD0II0WB8hmM5HC4cLu+hWACWir85AAm6wjJpbRI4HpVOZ9vpyuMjx9VXbxuEBCEi4imd4/ng6V5syi7kZF7LXSdEPxTLarVyxRVXhH2uqqrs27ePiy66SNvnm9EqLS0NeFe9qQjny6lnCBrAxIkTefHFF4Mu1hiJd9137aosxtC6dWu/0sJZWVls374dcAeWn3/+Od27d6/yevXxHpw8eZKVK1cC7mpk77//ftDhfpH47yCEEHq+w7FsDpfX3y2T0YDL5cJlr2xnjIvTttOTo1gV35dJhV8DsLjV9TzcsXU997p+SRAimoWYKCND+kb2D2NtlJeXs2rVqlpdY+nSpV5BSGJiIrGxsVoJ32PHjvmV+61PZnPlrye7PfBYWj3fYUi+HA4HP/zwA+CenP/EE08EDUCKi4trVVmqseg/B/369fN7jZs3b9a2J0yYEDQAATh+/HjddtCnD9ddd13I+UbVHTYmhBBNjtM7E2J3uHDphlaZTYB+yJbRqM0HAXcQ8kPcxdgNJpwY2R3fg2fiI/trfGT3XggBwDfffKPdLY6NjdXWdgiltLSUAwcOALB8+XKeeuopr4UG+/Tpw9atWwH3kJiGDEL0E6rDCQb2798f9Hh+fr62fkpaWlrAtU70srKyIm7oz/fff+81x+Kmm27ya3P6dGUqP9g6IR4//vhjyDbBgrlAqtsHT/AohBCRyuDwDUKcQOXvTpPJ4DVvxGD2/oqenGDGZDKwPc59szAx1oTN4SLGWL3fv02JBCFCNAP6oVgjR470W9m6KsXFxVx99dWUl5eTm5vLd999x3XXXacdv/baa7Ug5NNPP2Xy5MleGYrqiImJ0bbDyWx07NgRg8Gdnj5y5AilpaVB56V89dVXQa+nL68bbIFGj0haoBDg3LlzPPvss9rjLl26BAxCqvM+7Nmzx6tCWlX0/7aeQC8YfR/KysqCts3NzWXNmjUhrymEEE2aTybEZnehr/puMhq8hmIZdDcEAQ7klKG/L1Zc5uCe1/YyY0o3lM6ROWezZQ2aF6IZOn36tNdieqNGjQr73MTERK6//nrt8dKlS72OT5w4kbiKMamHDx/mr3/9a437maJbSEm/7kNVkpKS6Nq1K+D+YuuZQxDIt99+y7p160I+vye7UlRUFPTu+ldffcXatWtD9rGpyMnJ4Z577uHYsWOA+0v+9OnTvbJaHvoqZ8FeY3l5OX/5y1/Cen79v60+y1EVfR82bNiAI0DVGHAXA3juuef81m8RQohIY/ANQhwu7LrV0s3GqjMhFpuTGfMO4fRJzhcU25kx7xAWW2QUTvElQYgQEW758uXal7j09HQGDhxYrfP1CxWuWbOG4uJi7XHr1q154okntMdz585l+vTpnD171u86LpeL7du38+STTwYcw68fdhMqa+GhD6hmz57NDl3pQo8VK1bw2GOPhVxI0Wg0emV5pk+fzu7du/1ew4IFC/jDH/6AyWTyusPfFB0+fJj/+Z//YcyYMWRnZ2v7n3jiCYYMGRLwnMGDB2vDp7Zs2cLs2bP9vuTn5OTwwAMPsHv37rCqoun/bXfu3BlyHsnAgQOJjY3VXsP06dP95vTk5eXx+OOPs3bt2mZRmU0I0bIZfYMQm8NvjZCqMiGbsgsDLkUA7kBkU3bkzV8EGY4lRMTTZy9uvPHGgHe/g7nuuuto1aoV586do6ysjFWrVnHLLbdox++++2727dtHZmYm4C7Zunz5cvr27Uvnzp0xmUzk5uaSnZ1Nbm4ugFfg4jFixAg+++wzAD7++GOys7O5+OKLtUwLwKRJk7yqOU2ZMoVPP/2U3NxcCgsLmTRpEv3796dr166UlpayY8cOjh07hslkYsaMGSHv3P/+979nzZo1WCwWjh49ysSJE7nsssvo0qULFouFrKwsTp48qb2GBQsWaCV6G8OyZcu8AiWHw0FxcTEFBQXs3bvXLxiMj4/n2Wef9fr389W9e3dGjx7NsmXLAHjvvfdYvnw5vXv3JiUlhSNHjrBt2zbsdjvt2rVjypQpzJ49O2g/09PTueyyy9i+fTsWi4WxY8dy7bXXkp6erg296ty5M3fccQcAycnJ3HvvvVq55KVLl7J+/Xr69u1Leno6J0+eZOvWrVgsFuLj4/njH//I888/X/03UAghmgizw/tmj81i9xqa6jccS5cJOZkXPBsc6nhTJUGIEBFs7969XhOR9VmNcEVHRzN8+HAWLlwIuOeX+H6JnTVrFt26dePtt9+mrKwMm83GDz/8EHBIU1RUVMCsxNVXX82oUaNYsWIF4L5jvlO3MizA9ddf7xWEtGrVirfffpsHHniAgoICnE4nW7du1eapgPuL9yuvvELv3r1DvtaePXvy+uuv84c//IHy8nKcTidZWVlkZWVpbYxGIw899BC/+93vWLBgQchr1qfvv//ea6hdVRITExkzZgz33XdfWItKzpgxg9zcXO3ap06d8hsi1717d9544w2vkr/B/PnPf+buu++mpKSEc+fO+Q2fGzBggBaEAEydOpWjR49qn4eCggK/IXVt27bl73//e5XDtYQQIlJEOb3ny9mtNkwxlX8rzb4T03U3FNunBl8RPdTxpkqCECEimD4Lcv7553PppZfW6DqjR4/WgpAffviB48eP07FjR682Dz74IOPHjyczM5ONGzdy6NAhCgoKMJlMpKWl0bNnT6666ipuvvlm2rRpE/B5Xn/9da6//npWrlzJ3r17yc/P91qVPZA+ffrw5ZdfMnfuXNatW6dlJjp06MDgwYO544476Ny5szYfIpThw4dz0UUXMXfuXDZt2sSJEyeIjo6mXbt2DBgwgFtvvZVLLrkkrGs1tNjYWJKSkkhISKBTp05ccskl9O7dm6uvvrpaQ5YSEhJ4//33WbZsGYsXL0ZVVUpKSkhNTaVLly6MHDmScePGERcXF3YQ0rt3b5YtW8b8+fPZsmULR48epbS0tMoAwmQy8de//pWbbrqJBQsWsGvXLoqKikhJSeG8885j+PDhTJgwgZSUFLZs2RL2axNCiKbG6XQR7fTOVtitVszOypEAJiNVZkIGZSSTkmgOOCQrJdHMoIzkeuh1/TNEWglKEVxWVtY6YHBiYmLYZVp97d27F4BevXpV6zxPiVjfRe6EaE7kcx65avq7rSXyZAf79evXyD0Rov401Oe83OrkwIMPYKbypozlsZnEtWnD42+7y+T36BTH/4yK5vCMGQDEdOnCBS+8oLVXj5YyY94hr0AkJdHc6NWxVFX1zCVd369fv+urc65kQoQQQgghhKgnljKrVwACYLfYvKtjmQwQpESv0jmeD57uxabsQk7mWWmfGs2gjGRioiK3xpQEIUIIIYQQQtQTS6n/mkwOq92/OlaQxQoBYqKMDOnbun462QgiN3wSQgghhBCiibOW+C/K6rDZvIOQEIsVNkcShAghhBBCCFFPLKUBghCrPfhihS0gCJHhWEIIIYQQQuCeRL4pu5BT+XU378JWEmg4lhWHI8hihQGGYzU3zf8VCiGEEEIIEUJ9VaCylZb5feF22nzmhAQp0dtcyXAsIYQQQgjRollsTr8ABKCg2M6MeYew2Jw1vra9zD8T4rTasDu9q2O1tOFYEoQIIYQQQogWbVN2YcDFAMEdiGzKLqzxtQMFIQ67Hf1arn4T0yUTIoQQQgghRPN2Ms9aq+PBOMsDZEL8hmNJJkQIIYQQQogWpX1qdK2OB+Mot/jtc9n8FyvUZ0KQTIgQQgghhBDN26CMZFISA3/xT0k0MygjucbXDpQJcdnt/tWxJBMihBBCCCFEyxETZWTGlG5++z3VsWpTptdlqSIIaeHVsZr/KxRCCCGEECIE3zK8F3aK4/XfXVjrdUJcVv/5JC673bs6ltEAkgkRQgghhBCiZeuYGl3rAATAYA2UCbHJYoWN3YGGpijKDOD5ap52raqqG8O4djvgXmA80BVIBk4AO4CPgExVVWteaFoIIYQQQjQIs6mO7tVb/SemY7fj0H0jlBK9osYURRkH7AFmAQOAtkAM7mBkHPA5sFZRlE6N1UchhBBCiOYi0ITvumQ2G+rkOgZbgPK+drt/dawWNhyr+YdZwf0AbA2j3fFgBxVFGQksBDyfmGJgDXAW6AVcVbF/MLBKUZRBqqqeq1GPhRBCCCFauNMLF3J25UpSrr2WDvfdVyfXdLncQUGM04rdYHLP06gDRluATIjD4T0xXYZjtThfqKo6ozYXUBSlDfAJlQHIV8Adqqrm69pcC2QCbYAMYA5wd22eVwghhBCipTq7YgUABd9+S9vbb8cUHx/ijNBsdhfdLcd4IG8xZcYYtlum1vqaAKaAQYgsVijDsWpvOtCqYns/MF4fgACoqroBmKTbdZeiKBkN1D8hhBBCiGbD5fSeXqv/8l4bZVYn/533OdHYSXaW0G3v2jq5rsnhPxzL4PAZjiVBiKgORVGigPt1u55TVTXgAEVVVVcDqyoeGoGH6rl7QgghhBDNjtPik1mooyCk3Ood3MQVn62T65rsgYMQ7+pYLW+dEAlCamcI7gpYAOeAxSHa/1u3Pa5eeiSEEEII0Yz5BiF1lQnxDULspqg6ua45UCbEGaA6lj4TIkGICGGIbvt7VVUDlD/wsk63fZ6iKD3qvktCCCGEEM2Xy6cqVn0FITZj7YMQl92O2eXfP4PT4b1Yoc/EdGQ4VrPXTlGUKYqivKgoyt8URfmLoih3KYpyfpjn99JtbwvVWFXVE8DJKs4XQtTQnDlzUBQFRVGYM2dOwDaZmZlam2eeeaaBexjasWPHtP4NHTq0sbsjhBBNll8mRP/lvRZ8gxCrofZBiN/QsQpG3+FYLXCdkOb/CoN7iCrmZiiKsgb4i6qq3wc5v6du+0iYz3kUaO95mjDPESKkyZMns3Vr4IrTBoOBxMREUlNTycjI4Prrr+emm24iOjq6gXsphBBC1I7v+iB1NjHdYkf/V9FWF0FIFWuZGJwOv+pYtLDhWM3/FdbcDcD1iqI8parqP6pok6bbPhXmdfWZkNQa9SwMxcXFZGVl1fj82NhYSkpKanRuTc8TteMI8kvY5XJRVFREUVERv/76K1988QV/+9vfePnll7nssssasJf1w2q1em0H+gxadHej7HZ7rT+nwc4/fvw4o0aNAqBDhw6sXLky5PXKysq0bafTKT9H9cDpdFJeXl6r340tjbxXojasdthz3EB+CaQmwMWdXETVwSgj4+HDxOoe7921C+fp0zW+nudzrh6yM0C3v9xqq/XPgCE3l7hA++02cs/k4RmUdOTXQ7QtKNDWezjwyy9VZlGai5YahOzBvYL5uortfCAed2ZjPPDfuMvumoC/K4pyRlXV+QGuk6jbLgtwPBB9u8QqWwlRCxkZGWRkVFaBdrlcnDt3jp9++okTJ04AcPLkSR5++GHeeecd+vTp01hdFUII0Qwdy4P535kotlQu+JcY4+Kuqx2cV8tbsAbf4Vc+JXtrylnmnbUwOsPLsFQVbFntkPNrOZ6/xuWGaGJd7ptmJpcDXSIEkxHvKl/G5j9joiUGIW9WsUBhIe4V1H9QFOU94Esqh1vNURRlhaqqBT7n6APxUJPSPfRhbaDguE4kJiaiKDUb7bV3714AEhISqnWe585tdc8TdcOkm8Q2ZMgQpk71X2TJ5XKRmZnJCy+8gMViwWKxMHPmTFauXIkxgn/h6YeVRUdHB/wMTpo0iUmTJvntr65wPudxcZU/2kajMayfiR49eqCqaq37J6pmNBqJj4+nVy+ZjheK5+5vv379GrknIhJZbE5ef20vxeU2bi9cRQfbGT5PGcYx2vHp1lg+eLoXMVE1/5tTUFrKCd3jEtpxRZ/Lqn1N38/56ZydXsejjIaQPwPq0VJmzDtEQXFlYJSSaOa+ER2Yu/oEF+RWBiEFpkTa2/MAMOEkKSkZKAKgR4/uJGyPwxMGXZSRQdwFF1Tr9TQGVVUpLi6u0bmR+62jhlRVzQujzUFgNJWBRQrwQICm+pA53MH1MbrtcLMnQtQJg8HALbfcwvTp07V9Bw8eZMOGDY3YKyGEEM3JpuxCCortXF6uMqBsD53tp3kwz72KQUGxnU3ZhbW6/smTRV6PP197gnte24t6tLRW17X7DIM1OoJPeLfYnH4BCLhf498zj1JQbCfRUflVr8CUpG2bfKtj+U5Ml+pYLZeqqvuBT3W7bgzQTB/6hZvV0LerWegoRC1NnDiR1NTKfPjmzZsbsTdCCCGak5N57nu43a3HtH2JzjK/4zVhsTlZvfmk1z6Ty0FBsZ0Z8w5hsdV8aJazxDuIMYQYjuUJtgJxVcQXSc7KwKbAqAtCXHav0Vcmk1THEt7WAFMqtgPl7/OAthXb7cK8pr5dyKyMEPXBbDbTu3dv1q9fD0BOTo52TF9la968eVx55ZWcOHGChQsX8p///IcTJ06Qn5+PoigsXbrU79rFxcUsWbKE9evXc+DAAfLy8oiKiiI9PZ0BAwZwyy23cOmll4bdV1VV+eijj9i0aROnT58mISGB888/n5EjR3LLLbeQmBje1KrMzEwtAzR+/HheffXVkOf88ssvLFu2jM2bN5OTk0NBQQFms5l27dpxySWXcM011zB8+HBtuNWWLVuYMmWK1zVycnKqHBqpH3517NgxbrjhBgA6derE2rVrQ/Zvy5YtrFixgh9++IHc3FxsNhtt2rQhIyODYcOGMWrUKK9heoEMHTpU+/dfs2YN5513HsePH+eTTz5h3bp15OTk4HK56NSpE9dddx33338/aWlpQa8phGjZ2qe6B4cYXYEDAs/xmtiUXYjT4j13w4T7eTxZliF9W9fo2q5S70xIqCAknGBKH3zl6zMhLu9MSEtcrLDOXqGiKNHAfcDNwKVAG9zDmU4B3wMLVFX9oq6er4HoQ+1Af3VV4KKK7XDXFtG3kwHgotG0atVK2y4qKqqy3apVq/jTn/4U1pjPFStW8PLLL5OX5x1fWywWiouLOXToEJ999hnjxo3jxRdfDFki+L333uMf//gHNpvN61p5eXns2LGDjz/+mLfffjtkv6qrvLycF198kczMTJw+Ex5tNhuHDx/m8OHDrFixgnfeeYevvvqqzvsQTFFREU8//XTAQCUnJ4ecnBy+/vpr7f3r3r172NdetWoV06dP96vOdeDAAQ4cOMDChQt59913qxVICiFalkEZyaQkmjEUuPyOpSSaGZSRXONrn8yzEu2yee0z6RYDrE2WxVXunQkxOYMPxwonmEp0Vl5TPxzL7HJg1WVtfBcrbAnDseokCFEUpT+wCDivYpenFEIs7ipTFwKTFUVZB0wMZ15GExGv2w5UL3MvMLZiO2SdU0VR2lO5RojnfCEaxblz57TtpKSkgG2ysrL45z//id1uJy0tjf79+5OcnMzp06cpLPQe0/v+++/z2muvaY+Tk5Pp27cvbdu2xWazsW/fPvbt2wfAkiVLOH36NO+//36VE+L//e9/M3v2bO1xbGwsAwcOJD09ndzcXDZv3szhw4d54IEHtAxCXSgqKuKee+5h9+7d2r7ExET69etH27ZtKS8v58SJE+zZs4fS0lKv8sDt2rXjzjvvpKSkhCVLlgDuCezjxo2rs/6VlJQwefJkrYAEwHnnnUffvn2Jjo7mwIED7Nq1C4D9+/dzxx13MH/+fHr06BHy2ps2bWLGjBk4HA7tmnFxcRw+fJisrCycTieFhYU8/PDDfPHFFyQn1/yLhBCi+YqJMjJjSjd+nOV9Eycl0cyMKd1qNSm9fWo05U7vIMSsy7jUJstiKPMOQowhghBPsFXVkCyAJK8gpDJzb8LhtTii72KFSCYkNEVR+uAudev5wu6q+M+ge0zF4+uB/yiKMkBV1UgofqwPLE4EOP4fwLP08lWKokSpqmoL0M5jsG77mKqqB2rbQSFqwm6389NPP2mPO3XqFLDdW2+9hdPp5KmnnuLee+/FrPulqP/yvXHjRi1giI2NZdq0aUyaNMkv0/Hjjz/y1FNPceLECTZt2sS7777L7373O7/n/eWXX7wCkCFDhjBr1ixat65Msefl5fHMM8+wfv16Pvnkk2q+A1WbPn26FoBERUXxxBNPMHnyZO21eDIEUVFRrFu3jlWrVmnndu3aleeee45jx45pQUhKSgrPPfdcnfXvlVde0QKQ6OhoXnjhBSZMmODVZvv27Tz++OOcPHmSgoICnnzySRYuXBgy8/Tiiy8SExPDSy+9xM033+x1bNu2bfzud7/j3LlznDlzhnnz5gWswCaEEABK53gOJZpA972+tlWxwP3FP9fk/aXfWDEcq7ZZFqPFNwgJPhwrJsrII2M78dJHv1bZJlgmRD9/xWQCZwsbjlUXE9P/jTsA0Qcbx4HNQBbu2mMGKgOTS4AZdfC89UpRFDNwp27XtwGarQM8t5OTgVC3O+/WbfsPpheigXz++efk5+drjwcOHBiwncPhYNq0aTzwwANeAQhUlsV1Op3MnDkTl8uFwWBgzpw53H333QG/8Pbv35+5c+cSG+uubj137lxKS/2rmcyZM0cbgtWnTx/efPNNrwAEIDU1lbfeeouMjAyv4Vq18e233/LNN98A7kpib775Jvfff3/A1xIdHc3w4cP561//WifPHY5Dhw6xaNEi7fHLL7/sF4AAXHbZZbz//vvExLiL8amqyvLly0Ne32az8dZbb/kFIACXX34506ZN0x6HswCjEKJlM/jMCaltAOK5Rs907+uYXY46ybKYLd5FS/XDvKpiMBiqPuhykeSo/BtXaEzE846YcGGxVAZTZt85IS1gOFatPg2KotyAe/6HJ8D4FrhcVdXOqqoOUlX1Ctyrgk/AHZh42j2kKEq9rZERpL/VWRzwVUA/kPoj3waqqlqB93S7ZiqKEuPbruK5h1BZYcsJ/G81+iJEnfCsEzJr1ixtX7du3bjmmmsCtm/fvj333Xdf0GuuXbuWX3913wW6+eabue6664K2v+CCC7ThSQUFBX7lgQsLC1m9erX2+Omnn67yDn50dDRPP/100Oerjn//+9/a9vjx4xk6dGidXbsuLFiwAFdFyZXLLruMMWPGVNn2wgsvZPLkydrjTz/9tMq2HkOGDOHqq6+u8viYMWO0YPTw4cM1rg0vhGgh6mgRQV++c0KuuTiRD57uhdI5voozwhNl88mEhBGE/HqqvMpjMS4bUbivYTGYsRqjcVAZXNgsla9D5oRUn/4W3HrgN6qqeuXIVFV1AksURdmGOzOShnueyA3Ailo+f3U9qSjKIOBt4KtAQ8IURTkPmAXcpdu9RFXVjVVcH9zFJAAAIABJREFUcxbwX7hf00VApqIod+oXNlQU5WrgMyqHqM1XVXW335Ui2NkvvyR3yRJclkgYZVe/jLGxtBk7lrSbbmq0Pqxfv94r0+FZMX379u1elbCio6N5+eWXq6ygNHz4cL8MSKDn8gh0Bz2QK6+8UvtSnJWVxYgRI7Rj27Zt0zIbHTt25Iorrgh5rQ4dOmgrwdeUxWLRqoIB3HXXXUFaNw59/wJlQHxNmDCB995z3yfJzs6mpKQk6MKJN94YqBJ5JU9lsoMHD+JyuYJW/hJCiPoKQhzlFq8vsJ3TzHWSZYmyeQcUZqcDh9OFyVh1tiNYEKKfD0JCEk/fdj7ON01QMczLZbeD0f1KDAakOlY16b8dPOYbgOipqnpEUZTZuDMMnnMbOggxAMMr/itTFOUn4CDuIVVxuFdI74/3+7IT72FUXlRVPaMoyh3AMtyZpZHAUUVRVuMuwXsRMEh3yh6g2Q2kzlu1SgKQCs7ycvJWrWrUIGTXrl3a5OSqtG/fntdeey3oarAZGRlVHvPYsWOHtr1y5Uo2bqwqXq908uTJgNuA14TrPn36hLyWwWCgT58+tQ5C9u7dq81zSUpK4uKLL67V9eqay+XyKunbt2/fkOd0796dlJQUCgoKcDgcqKrK5ZdfXmX7nj17hrymfjK6ZEKEEMGEKnFbU77fN5y24BPIwxVj9x6OZcaOw1F1EFJudZJ9OFDdIjf9fJDktq3p27c1WUYzON1/a/TDvcwGV+XiIgYDhiqKtjQntQ1CPOVmj6iq+lPQlm5LqQxCutTyuWsrDriy4r9AHMBc4AlVVav+hAGqqq5UFGUi8C7u4WeJBJ4f8i1wh6qq5wIci2ipI0ZIJqSCMTaWVN2d/abAYDAQHx9PWloaF198MUOGDGHkyJEhJyr7zsMI5PTp09r2ihXVv6+gr9IFeJX37dChQ1jXaN++fehGIZw9e1bb7tSpU/Bxvo2gqKjIa+5Lx44dwzqvQ4cOFBS4E7P67Fgg4ay5EhUVpW3b7XXzh18I0Uy5/Ev01gmrbxBSN8FOjMNn/RGXE7vTRaC/lOrRUr/V0j0ToD30QYi5oiy+w1A58sCMLghxtawsCNQ+CEnG/X4fDrO9vl1j1HacDWzAnZkYiDuIaoM7cLAD+bgzFRuBf6uqWnW5Ax+qqmYqivId7rVSxgNdcQ/ROok7mzIfWFQxPK3ZSbvpJmIr5gIEG+4hGsYjjzxSJ5WLPBPIg6nt3XCHw/uPR1lZ5Z2ocJ4fIC6u9lPM9OtixMfXblxxffCdwB/ua9a38137w1dTC7yEEJHNUE/DsQw27yDEUQc3RFxOJ3FO7+uacWC3u8Bntq/F5vQLQMA7AAHvIMRUEYQ4jZVfvb0CD90kfglCwhOD+z0vC9UQQFVVi278cMAJ3PWpIqOxpuK/+rj+KdxzRGaFaitEcxEXF6ctdvjFF19Ua2G8qq7nUV5e9VhbPX3gUlP64DlQxa7G5hsYlZWVhZW50L83coNACNGgqlgxvbaMNu8FCZ11EITYiv1v0phdDuwO/2zOpuzCoGuDeOgrY5m1IESXCdEFISZ9ENICJqVD3ZToFUK0YGlpadr2mTNnan291NRUbTvceR6+80pqok2bNtp2Tk6OVoWqqUhKSvIaCnX8+PGwztO/h+EMrxNCiLriW6K3Lrjsdr+qVXUxJ6S0oMhvn9nlwBYgCAl3VfZEZ+VNoIBBCPogRDesq4VkQiQIEULUin7y+LZt22p9vV69emnboSbXe+gXXqypiy66SJsjU1RUxJ49e2p8rfoY1mQwGLwqUekLAlTl4MGD2nwQk8kklayEEA3Kd2K63VH7oMQZYO6pq4pMSLnVydrt+Xyy9hT/2ZHvtTigX9tC/6HFZhzY7P7nhLsqu3441mmbewCQSzccy1TVcCzJhAghRGiDBw/Wtj///PNaT1a+7LLLtDv+OTk5/Pjjj0Hbb926tdaVsQBiYmIYMGCA9nj+/Pm1upZHXS2kCO5yxB6LFy8O2V7fJiMjQ4ZjCSEalMnh/ffAYa19xsIZYJhuoOFY6tFS7p29l9kLjjDvm5O89tkR7nltL7sOFmuByc4jBjxz2i3nAs9vtAfo86CMZFISA2cr9Leg9CV6520qwWJz4jIFnhNi0gdsLSQTUlev8kJFUZ6rr3NUVZ1Zgz4JIRrAjTfeyN///neOHTtGTk4Or7zyCs89F96vg7y8PJKTk73WKUlJSeGGG27gq6++AmD27NnMnz/fayiSh9Vq5bXXXqubFwLcfffdWonhxYsX85vf/KZGCxa2atUKo9GI0+kkPz8fm80WsP/V9dvf/pa5c+ficrnYtm0by5Ytq3LBwoMHDzJv3jzt8e23317r5xdCiOow+SwqaLPaiIkLL4tQlYCZEJ8CJ1VNHC8otvPH937RFe0ykRjj4qUOpRjO+Q/HArBb/W8kxUQZmTGlG9P+dQCnbrRWfIyRUktlRkOfCTlpiWFTdqFXlsOMgw62M/y28BtO/L/KQiySCame7sDzYf4H7kCxuucIIZogs9nM888/j7GipvlHH33E1KlTtVXUfblcLrZv387MmTMZOnRowMnnjzzyiPalfceOHTz22GN+5WXz8/N59NFH2bVrV518wQe47rrrGDZsmNbPRx99lLlz52rrh+hZrVa++eYbnnzySb9j0dHRdOnirkJus9m8VoCvja5du3Lrrbdqj//85z+zZMkSv3Y7d+7k/vvv195bRVEYPXp0nfRBCCHC4XC6iHJ6BwGBsgrVFSgT4jscK9jEcd/pfsUWAzPmHQo4HAvAbgmcze7RKQ79yNvHJ5zHuKvT/z977x0nx13f/z9n+/Ui3ekkWV3ySJa7bDDGyDbVYGOICSGhGGJaeiHBTn4k3whCQsBAgMSUEEC2wUAoARwcMHHDIAyusi3LY/UunXR9d2/LlN8fs2Xa3u3dzsqS7/18PO6h2ZnPfvazeyvp85rXu7jGtDsS0ycirXYuiac61nuGf8Cy4lFye/dWzs+VnJDn413OJNvTW3JZEIRTkI0bN/KhD32If/qnf8I0Te6++25+9rOfoaoqq1evpq2tjWw2y7Fjx3j22Wd9vUG8rFmzhg984AN8/OMfB+Cee+5hy5YtXHLJJfT19XH8+HEeeughJicnWbx4Ma94xStcd/0b4WMf+xiHDx/mmWeeoVgs8vGPf5xbbrmFiy66iL6+PnK5HEeOHOGZZ54hm82yePHiwHle85rX8MUvfhGAD37wg/z3f/83S5cudQmmm266acbr+5u/+RuefvrpSnPFm266iVtuuYXzzjuPRCLBzp072bp1a2V8d3c3n/rUp6btCSMIghAmuYJJ3HILgWIIIsQIzAlxOyH1Jo6XGU3rHD04zJKAa3rATSiAkbROOcWlszXKay6ex31PVG+WKZZJu1UVTJlIioHehCscK2oZ9Jh+B2auOCFhiBApLC8IAm9/+9tZunQpH/7whzl48CCWZfHss8/y7LPP1nzOueeeW9PFuOGGG9B1nc9+9rPous7k5CT33Xefa8yyZcv4/Oc/z//+7/+G9j46Ozv5+te/zqZNm7jzzjuxLIt0Os39998fOL5Wv473vOc93H333ezevZtiscgDDzzgGzMbEdLe3s7tt9/OBz/4wcrnsX//fvbv3+8bu2bNGj7zmc+wevXqGb+OIAhCI+QDRIgeQhWr3ERASXZPOFa9ieNO9JHh4PMB4VgAx0aq4mRBj/165VyR0bROyqpen1SSdHUkuHR9F1scLkePERwCJk5IHWiaJontgiBU2LhxIz/96U/56U9/ygMPPMDWrVsZHh4mk8nQ0tJCX18fq1ev5qKLLuLyyy9nxYoVU873vve9j40bN/L1r3+dX/3qVwwODtLW1sbSpUu56qqrePOb30xHR0fo76OtrY2bb76Zd7/73fzwhz/koYce4siRI4yPj5NMJlmwYAFnn302Gzdu5NWvfnXgHB0dHXz3u9/ljjvu4IEHHmDXrl2+ruezpaOjgy9+8Ys89NBD3HnnnTzyyCMMDg5iGAbz5s3jrLPO4tWvfjXXXHONK99GEAThZDFZMIjjFgdGjdCmmZCdCOjj5EmAd4qBeikOBYuQWsn0ThHSXxIh5oG9/L/Mt3gs081dyYsq1/PRJJuuX0EyHgGHE7K8GFxqXUSIIAinFbfffvspMU8sFuPqq6/m6quvDmU9a9eu5aMf/eiUY/70T/902g7x1113Hdddd92MX3vt2rWuc+Wu4/VUmmpra+O9730v733ve6cde8YZZ6Bp2ozWd8kll3DJJZfM6Dle7r333hmND+t7JgjCC5t81h82FYoTkvY7Id7E9HLi+F9+focrpl9R/Dkh5fNdxWBXIj/pfx8Ag6MOJ6TbFiFDP/4xHD7AhRxg9RVr4H77ev9AF6uWlBrOOkTIskJwj6u5Eo4lToYgCIIgCIIQKvmMP4G8WSLEG44FsHpxi0uADPTG+cT7VhGLurMIElELy7ToNqu5iieiXZXjPQf9ndTB64TYYcXFoaHKuaVUnZVYW2v1iTFHNUgzOBleRIggCIIgCIIgzIJcgINg1BAhM2kqWMj6xQ2mf97MpFuYWJbC+mVtpOJuEXLBMpNWK0+ylL+SU+JMRKqiIT0R8HrA4Eg1tKzshBjpqqgoHj9eOY60VudzJqbXQsKxBEEQBEEQBGEWFLJ5Up5zQSJEO5D19fTobo+x6foVqEtafeOLASJECXBCJiYNIpbBObmdjEU7OJI+g32DOdI5t8BpT0G3UXVBRqMd6Ep1e9yW8Mdv5Qome45WHZmeDnu8U4QUTpyoHEdLIkQ7kOXZQwUW+mb0vJ854oQ0XYSoqtoFrAB6gXnYJXeHgGFgj6ZpU9fqFARBEARBEJ43dkxo7Ezv4NL5l9EV767rOYVJf2lbb5L3VE0FN922h803rrOTuR3oAX1CMP0iZDyr87LME7xh4ueYwM3KO/irL9jX5umjqPl9PJVaTSKW4ox4Ndl9NNqB5Sj8urDLLQiCRNM/3LqXTW9fiuVYW9EhQiKtrZX3erE5fRCSOCENoKrqmcD7gSuBc6ldxtdUVXUrcB/wJU3TdjZjPYIgCIIgCMLMyegZPr/zcxTMAkcnj/D7K6cvtAFQyAWEY82gqeBoWmfLtjGuPL/Hdd6crNMJyRq8YeLngJ17cO3Eg3w5/ltELJP3D3+f+cYYF+Q0vrL9zdykKnDEft5IpIMOR6fz8tzpp59m5MFf8NVDyxk13F7GWEbnk7dux9W61rGmaGtr5b3qSh0uxxxxQkLNCVFVdUBV1a8D24C/AM4vvYZS4ycKXAh8AHhGVdXNqqr2h7kmQRAEQRAEYXYczw9SMG1XYzB/rO7n6UFOiCcca7qmgkHXjQBxowQ4IWlPTkiHYSeYL9CHmG+MAbCqcIhcUWH0cDV/w+jooaMjWXlsFotYpsnhL32J9K8f4l0Hv4USUGJLnwhOMgfbCSm/F6MOETJXnJDQRIiqquuBXwO/hy0uoL5u5+UxMeAdwK9UVV07xXhBEARBEAThJFA0qwnY8Uhwc9kg9IDO5mZxZk0Fvde1A1nGRwOqVRl+N2Ui6z7XYtnr6fY0CIxZOoajqtXAigFa2qoixNJ1zMlJjInq83pLIsZJqxWcwA62E1J+Lyei1XA2E3ig7QLf+LkiQkJ5l6qqrgF+AXSWTlnYTkcWeKT0cwIYKZ3vBvqBi7GdkBaqYmQF8KCqqudrmnYojPUJgiAIgiAIM6doVt2IuFJ/J3I9P70Tcun6Ljpbo4xn/U5GV1uUom7yzXuPMdCbYMOZHWy6bQ9vM/0ND03dIF80XfkjEx4npMW0RUifMeo7H89WRUXXovlYJ/ZU5y4WMT1NZhcXj5Oy8szTx3g6tRpTidBq1hYhkZaWSgPFZ63l/KjjZbSZkzzUeg5DsW4uym6nzSFiJDG9TlRVVYBvAF1UxccO4Gbgdk3Tgru8VJ+fAt4F/BWwqjTHPOA/gdc2uj5BEARBEARhdhSt2TkhRoAIMT05Icl4hD94/WI+8e39rvPtLVFME/71ewcr51qTEbJ5k6TpnzdqGb78kQmPsEmVnJD5+qjnfI72fLVG0h2PFrkuUhUzlm5gFdyveX5O4+zcbmIY/KT9JdzdcQl98dqhZdHW1koDxX+4dQ/3Kxe5ruuJFOQdImSOOCFhhGP9LnARVSfjm8AFmqb953QCBEDTtJymaV/Ezh/5DtUk9lerqvrGENYnCIIgCIIgzAKXEzIDEWIWphchAP3dbndlw5p2YhHF52Rk83Zp3VbLv7WMYvryR8azOjpVRyECRCyTPn3ENa7NzNPlaBp4sNDCvqHqOi296Hsv5+d2EMNe31XpX9HdHuMNF7T41lV57VKJXnVJK5tvXOvsVwiAHnc/d644IWGIkD9zHN+tadrbNE3L1hxdA03TMtiC5h7H6fc1ujhBEARBEARhdhRmmRMSLEL8YVfZvPtcJmcwmqndWb3N9HdMj1qGL38kndUrQqFMp5lhvicca4E+TKR0H3080oqhxMibTidE9zkhXjbfuI7e2NROSJlUIsr8Tvdai3F3RxURIXWgqupy4MWlh5PA9Y3Mp2mahZ2cPontiLxKVdVFjcwpCIIgCFZANRtBEKZHd4qQGeSEmAV/7oYV4IRM5t3NA4fGawuQmKWTLIWHmY7uDzFMXnJWp2tsLu0XK/P1EXo8iekLi9V+HqPRDgB3Gd0AJ8RJPtVBMh5xNSr04hQhAL2d7nArnxMi4Vh18fLSnxZwh6Zpx6caXA+aph3FDukCe32XNzqnMDMUxf6LbZrmNCMFQRBOD8oipPzvmyAI9VFw5IQkZuCEWMXZOSGThdp7jzZH8ncm0oLhECLJiPtGQyHjr6K1qnCw4nqUGdD9IsRwhHFh+HNCXETssVOJkEiLW2T0dLg/R93rhIgIqYuLHcffrDlq5tzhOL4wxHmFOoiWbEA94I6FIAjC6YhRahwWiYTaHksQThksy2LHhMau9M5QnT9nTkhsBiKEYoATElBKN5tzi47MpEF3e4yoZfD2kbt4/9D36NHtxPGeaDUfJB9vcYkFy9OwUM/4nZA1+QO+c/Md5XYnIrZjocQdIsDQp3RCorotjIwA0QMQSaV84VW97R4nJOHJJ5FwrLoo9/MwgYcbnMvJb0pzAvgLKAtNpbVkG05MTEwzUhAE4fQgU9ogtLTUTh4VhNOZp8ee5DPPfZJPax9nZ/q50OZ19glJROoPx0L3b9ytot8JmfQ4IRbw59edwXm557gwp6EW9vOy7ON0t8c4d6A6Lt7R4Wr8Z+k6uYLJvY+PcMc9RzECOquvLB72nes2qpWxMpEWuttjnL26y/E+ps4JiRbzWJZV0wmJeEKxAHo63WLOSEg41mw4o/Tnfk3TavtQM6Q0117svBDJCTnJdHTYduT4+LjEUQuC8IIgXdogtLe3P88rEYTm8MVd/1453rznP0Ob190npH4nRKnXCcn7w6+KusUCfbjyeIBxNt+4DiarbkOsowNDqW5jd+xP8/s3b+fm/9rP7f93jKQ1dTJ5GedGeM2aPjbfuI7uLocomMYJUSwLq1CoKUK8+SDgd0JEhMyO+diidWS6gbOgPGfnlKOE0GlvbycSiZDL5Thy5IgIEUEQTmvGxsYqzm5bW9vzvBpBaD6FgF4as6VoVYVDfAZOSMSoLULKjsU37z3GzsNZuowJXpp5gh7dDo169LkJV2fzNj1DMh4hP1o919LdgeHYxn7623sYTVfXmjKn7RLhY/WZfSTjESLxqthSAnJCjIhHRGSzM3NCPDkhZnJuluhtVGqVP7VmiJBykJ6IkJNMJBJhyZIl7N+/n7GxMfL5PJ2dnXR0dBCLxVAURZI7BUE4ZbEsC8uyyOVyjI2NMTZm/3cyMDBAIjGDcBJBOE2JKOFtYmfbJ0QJcD0s3UA7kGXTbW7B8JfDP2KJPshl0a18ou96frltjOsdIqRVz1LUTcxMdaPfMb+LEcf7zGQKRKMJ3jr6E7qMNHsSi+tea5n23m4AIo6cEMXjhMx73ev4Bhez4aefrXRfN8bHAyt/AUQDQkB7O9zbb9PrhIgIqYsEthPil7uNU55Tbls9D7S2trJ06VIOHDhALpcjl8sxODg45XPK1bQk8VN4ISPf89OT3t5eenp6ph8oCC8Aokp4/z65c0LqEyGTeYNokBOi6z4BErd0luj2/mKBMUyvMcbQZDc9jlyNdjPDgeM5Wh09QpJdbickisGLs09zQc7OhwnK/5iOzj77vnc04RYhTidESSYpTEbIOcoVF05UK2x5CXJCer1OSMo9RsKxhDlPa2sra9asYfHixXR2dlZckFqUxYogvJCR7/npgaIoxONx5s2bx8qVK1mwYMHzvSRBOGlECO9OujO0K1ZHToh2IMsNNz9L3ArI/8gWXQIEoM1097deVDyOYpl0GVXXI2npPLdr2NWoUGlrcyWmRy2Ttfm9064P4PudVwaeT3bbCelRZziW6XZCIvE4RcMi7whNKx6v3aEiKCekq83bJyTpeixOiCBg3+3t7Oyks3P6qLhHH30UgHXr1jV7WYLwvCHfc0EQTnUiTXJCpssJyRdN2+nI6IEiRC/qdgyNg3bDI0L04+wzFxLDnbD+nR/v4mpHn5Bb7h7lMocI6W5RyKenD7e8rft1jHYvhvH7fNeipcIVbhHizglREgmKuuVyQoozdEKiUfcN3eeGFC5yPJ4rTsjceJeCIAiCIAhzhFDDsWbQrHDLtrGK0xEkQjD959pNdz+PxcXjvq7mAIlCxuWEHCskXOFYN7x6Pg9/K+l7HsBd7ZdiKBE61TW8+rJzefHyKPv+wj2mGEkQKeWMRRMxyoWDI14nJJGgqJvkajghse5u9NHRyuMgJ0Q74BZejx80eKvjsTghM+NyVVV3hzRXGfHOBUEQBEEQZkjzEtOndhqODjtCtwJESNQy6W6PuUKy2r3hWPoJVz5ImU4jS6vDCclGWlwiZGFXjLbOFgjoGTgU6+LxlrVcf9YAV57f42tsCFBIVMVCNBF3iBDDJUKURMIOx6rhhCQWLHCJEK8TUnaLnBie31fQ+l6IhCVCUsDykOYqI3VhBUEQBEEQZkjzwrGmdkIGeqsb8yAnJK6YbLp+BX/x+R2Vc14npNcYZ2HRH97UbmZdTkgm0oLu2LzrBZ2E4u85AlRCp8rrU6JR9GicmCN53khV6yDFEnHKsiNiuhPTIwHhWM7E9PiCBaBplcfe6lhOt6gW5hzJOwzjW6qUfsJGasAKgiAIgiDMkGiIiekuETJNYvql67voLjXiCxIhyYiFusTtDHidEID1eX9wTbcxQUupAaGJwqSSxHSILb1QpC0a7CDkIgm622Ncur7aCb0Yd4sDq7XayDSWrL7PaJATopvkItXQLytf7UmS6O93zet1QpxuUS0SAwPTjnkh0KgTcmsoqxAEQRAEQRBCIVQnpJQTYuoxfv10gfT4MQZ6E1y6votk3P06yXiETdev4G//cxdx/CJEMW2RkDJzvCT7FEdj83xOCNghWV6cHdSzkRSWomA4xJZRNIjowR0jEm2tbLp+hWu9ZjwFuWrYV6TdIUISVRESsdyJ6ZFkkqJhuMKxnMS6uoi0tGBO2u/LmxPidIucfL73TVwz8Qt6NpxPctGiwDEvNBoSIZqm/X5YCxEEQRAEQRAaJ+yckNxwP8d+eTWfz1eFQHd7jE3Xr/A5G+qSVl51QTfJfbVFyKvTv+aKzGOYwEi0vp7UC/ShynFGSdHdHiOajkHJhNALRdCDXYaP/sFZtC9yr9NMtoAj/z3e4RQh1e2xzwmJxynquiscy0msp4dYVxeFsgjp6HBdL7tF3pCsncmlbJ53PZvfPXcqL0qfEEEQBEEQhNMYw3KHIYUZz54vmhz75dUYeXfv6NG03XhwPKNz7+MjfPPeY9z3xAj5okk6kw+cqyxCrsg8Btib0HkBSehB9BvVZO+e/i4237iOSMzphOhEaoiQVGdA3+tkyvUw0V0VQ7FkNdQqatXICamRpB/r7qb3qqtQkkm6Lr2URF+f+2VLblE5bK1MWdR53aUXMlKiVxAEQRAE4TTGmbcBflEyWyzLYuzgUp8AKTOa1vn9T24nm6smhHe3x1jcGhwWpZgmutF43aHegR6S8QhWpLqNnUqERFMp3zkz5c4JSTj6ocWdOSGWiZGvzvvQzkkK+hThWN3d9FxxBd0ve1nNUrvqklY237iOLdvGODpcqBne9kJnbr1bQRAEQRCE5xnLCrcAqLOrOYQnQopWkWJm6nAppwABW5gcOhpQJxeIWDr5YnAFK4D+t7ylrnWVmwoS9Tghhl/8KLGYr/mfdiDLs4Pu38E3fpWu9O9wipCYZTAxVs1b+bf/GWQiawaGYymxGNE2W7BN1+sjGY9w5fk9/N7LF3Dl+T1zToCAiBBBEARBEISTxq+HtnDT1r/kjn23hzZnwXSHP+lBjQJngW4WibfVFy7lJBrQlBBsJ6RQQ4QYkRi9V11Fx0XV3uHm6rMoBlT6KosQK1K9Zuo60QAREvGUyC336Uib7kpfg8Ukm27bQ75oEo1X541hEHM4TUXFFjRB4Vix7m4URYq71ouIEEEQBEEQhJPEbXu/RsbI8MsTP+fw5KFQ5myWE1Iwi7Qu3kU0GexslFE8zo6zqaDTrYhYBgXdIkiGFJJtKIrCove9j+6NG0ktW0b8VVczEfWHglWdEHc4VsycXoSU+3RMRtzd1TORFkbTOlu2jRGJRtEdW+SEo9JXRYQEOCGx7u6AdybUIpScEFVVlwNlvy6nadpzNcZtBjbWOe0ocJmmaf4C0oIgCIIgCKc5E8VxaFnc8DxFjwgJywkpWgUiUYMFL/0xg1tej56rbujjUSgasC63m7eN/oSD8X7+o/c6TCVCp1kVLZHeeZjHB+1j0yBfNDGIEsEtlPSULSwiiQQLb7gBgCPDeXZHWun1JK/HSiJEiTpK7hZNY9t1AAAgAElEQVQNYob/fUc8+SDlPh05xS1C0pEW13VDiRKz3HJJJ4pVKn+c9zwfRITMlIadEFVVu4EngMdLP6+ZYvgC7M7q9fycB9zY6PoEQRAEQRBORcIqpetzQsyQckJKzkKqd5ANb/wJ175kfuXa2iWtdLREee/ID2m18pxZOMC6/B4AOo10ZVysd17lOGKZ5PI6cfzrM1vafefaU1EmIq2+85Vmfg4nxDR0YlZAQrwnPKrcp8PphJSbHzqvGwG/m4IScx2bnjpkIkJmRhjhWH+O7YIowL2apv1bHc+ZLiPLKs33V6qqdkwzVhAEQRAE4ZTH9NxZj4bUVNAfjhWSE+KYtyUR442XVUWIdnCSc5a5XYZeY4y2VIQuhxPiFiEGhWyOQNr8IqQ1GSXtFSFrz6XlzDPtY4cIsQpFEgHv29nNHKp9OpxOiN38MOLqqh4kQooOEYKikPfkhYgImRlhfPtvKP1pATfN4Hm3TvHzdGlMK/DbIaxREARBEATheaV5uRtNCsdy5FjElDjjGb1y77+gW+x/ardrfGtC4a0vH6DTqIqQ+PyqcIlaJoVMsAhR2vz3nKNRhSMtCyuPn0yuZt57/6CS/K04+oSYOX/ndcDVaBCqfTqibVVxk460+Pp0GBF/xoJLhAAFrwjp6QlcgxBMQzkhqqq+GFiCLUDu0zTtsXqfO1W3dVVVzwPKc70d+Foj6xQEQRAEQXi+yfuqWDVHhIRZordM1Eyy6ba9rlCWJYWjrvHxXIbRTNHlhCTmV52QqGVQmMzhThW3OaEnyRdNX6naZ+efy/d0HV2J8XDLOq7rcrgvznCsbLAI8TohYPfp+MjfvIwdH/g+8UKOjrPPZvMfr3O9dnA4lruiViGaBL3adj0uTsiMaNQJeaXjOLRac5qmbQUewQ7JukRVVaniJQiCIAjCaY2vlG5ANafZzet3QsLoRVJwrO/EgYW0jhxmfW4XlOZeWjzmGt9mZNl5aNKVE5LsrzohEUyKNcKxfn3A4l2f2F7p1VGmpSXOL9vO59etZ9PeliAeq24JnR3TrVxwHSMzQIQAtLS3oL/t98hdcw0b/uR6n/gxI9OEYwHFqDs5XcKxZkajm/sLHMf3NDiXl7tLf6aAs0OeWxAEQRAE4aSSN5rTz8NbHQvADEj+bmTe+FGLPz/xLd498iOunvgFAEs8IqTdzLJtb8ZVHSvpCscyKNZwLNKlErnlXh1lWpOOUrmxiOuaMxxLqRGONfCOd9R8f1Z7O+aKFYHNDHOGf4vsEyExyQlphEZFyFmlP49qmnaw0cV42O44Xhvy3IIgCIIgCCcVrxMSVgK51wkB0EOokOXMCelsy1eqWl2eeZw+fZgBfcg1vt2cRC/qtJu2K2ECuzPV8KkoFoVMsFgohz+Ve3WALQZ2Hq46JyfGiy63JOIQD5FCddxQcj59111H33XX0XnppTN6z+VmhkFNEr0ixNuYMNLqr+Ql1KZREdKLnQ9ybLqBs2DQcSyZPoIgCIIgnNZ4c0KKNTqLz5QgERKGwHHmhMw/P8KB1CLA7iL+zpEfE/EUO203s3SY2crmMh1p5cPf2O9q/FfM+MOmJpUE++IDlcdHhwsVMaAb7tdwuiVKrJqjES1UxU0hlmL+tdcy/9pricTdeRzTUauZIbhL9AIkPOF00i19ZjTarLDsOw1NOarK54Af1DnW+S3tqntFgiAIgiAIDfL9g9/h2fFneNMZv4PauS6UOZsVjhXohISQnO4q0ZtMsPQtb8K61e7EsEg/4Rvfbk7S5aiMNR5pZzStuxr/mZnq9ecSS3guuZSdiSXkHZv+gd5ERQwEUels7gjHiutVJ8SIzUx4OCk3KxyJdvqueRPTEwGfu1A/jYqQIhAH/MWdA9A07X9nMLezVls4ZR4EQRAEQRCmYXd6J/ccs1NTP7fj09yy4cuhzOt1QpoZjhWKE+K40x+PxFGvuJCH71pGx/F9geOTVpH5xmjl8Xi0DQBTiVQ6xFmT1XvMx2M93Nv+Itcc5V4d33/w+JRrOzpcYL4jHCvpEHimJ1djJpSbFY5E/SWDdY8TEg+psMBcpdFwrCHsClbzphs4C5xz1uu0CIIgCIIgNMS+bPAmu1H81bGa6YSEIUKq8yYiCRRFofDyN1Io3cM2UdASS8k6Gv8tKlbFw3jEFiGGY7vpFCGmx7Fw9uooi4FaDPQmiMaD76VbDYiQcjPDYCfE/XqD81ZVjltWrfIOF6ahUSfkBLAUWK6qarumaenpnjADzvO8jiAIgiAIQtMJq8+Gl5NZHcsIIzHdsb54KRTp4leexx/95n2YExOciHWjKzH+8vg3aNXtVN5FelWEjEXb6G6LYUWidpY6EMlXczfOWt3Lmo1LOTpcYKA3waXruyqlcstiICgkq+yW/GqXP3kcwGogHKvczPCWW/z1loqebfOOM6/krMgxrEKBRe9736xfc67SqAh5DLgQiAIbgbsaXlGVlzuOnwhx3ilRVfV7wHWOUw9omnZFnc/tBN6J3eV9DXbi/iB2pa9vAXdomhZcsFoQBEEQhFMCs1kixNessJnhWOHmhMQi9sY+GY9w4w1n20njJYGQjrZC6a14nZDXv2Qe5v6qWIjmq05ItCXFZecH1x4qi4FNt+1xCRGnWxKN1djGJmbvhIDdzPDa157pa5XtzQmxWttZ8eEPA5KUPhsaFSH3Ae8pHf8RIYkQVVUvBjZgRxDuakL531qv+0bcAmQmz70M+Aa2M+RkSenn1cBfqKr6u5qmbfc+XxAEQRCEU4OmOSGnQDhWwcwzlB9iILVw2o2zs1lhIlLdgKtLWtl84zq2bBvj7keGSY9We6B3mlWRMR5tR1EUTEf38VixmkAeTfkrUDlxvk6QWxJNBG9jlXhjIgTgskuWsPNrEaJU+5J4S/TGYxERHw3QqAj5GZAHksBrVVW9QtO0+xuZUFVVBfgXx6n/aWS+GbxuJ/Dvs3zuhcBPgLbSqQJ288bDwErgcuz8m3OBe1RVfdHJElaCIAiC8EKlaBa5f/Be4pEYL+u7gqgSHJ4zU5olQrxiYark8bSeJqpEaYm21BxTndcfZBE0d9Es8pFt/4+RwjDXLHoDr114zZTzupoVRtwb+2Q8wpUlF2P748H9McYibQz0JrAUR8PBYjUcK5ZKBT2t5ut4idYov6s06IQApJIxoj29MFLNCPCJkKgIkEZoKDFd07QTwFdLDxXgv1RVbTQz51+BK0vHReDTDc5XLx8DFpde87v1PklV1QTwPaoC5FFglaZpr9M07T2apr0cOAfYXbq+ENsxEQRBEAShAX4z9Ct+cOi7fOfAt3hqdGto8zYvJyTnelzLrdid3smHnvwg/9+Tf81gbjBwjJN6S/RuG3uKkcIwAP9z+IfTzuuqjqUEb/gvXd+FkWoLvKZ0dnHp+i7MSFUcJh2fQbRlaidkOmolpochQgBaB/pcj72J6fGYiJBGaLQ6FtiuRQY7dGo+8BtVVX9nppOoqtqnquoPgD8tnbKA/zwZjoGqqpcCf1h6+Clg2wye/n5geel4CHidd82apj0DvB7bNQLYqKrqa2e9YEEQBEEQOJw7XDk+mjsS2rzeMCnLsmqMnBlesVArHOvh4V+jWzoFs8BjIw9PO29gYnqAwMka7kaBx3JH+VftZj717MfZnd7pn9dylugN3tgn4xFetXGZ77yJwgfeuZ5kPGInppdocbg28ZbpXZ6piNUIx4okGxM3ZeK9va7HQeFYwuxp+NPTNO0AcAO2E2Jhdzf/pqqqW1VV/WNVVWt2+FFVtUtV1VepqvoVYA/2Rr08z+PAXzW6vulQVTUO/EfpdfcAH5nhFH/kOL5Z07TAWxYlIbK5xvMEQRAEQZghzk19mO6FP2wqnLnzZn1OyFhxLPC4FoFOSIDASesTrse37vkKO9PPsTuzk09rn+DHh3/kElzFGjkhXs5YPt93rmXVKtYut3ttOEVIynKEeLVOH441FbEaTkgkGY4TEp/vfl9FjxskTkhjNJoTAoCmad9RVXUN8FFsAaFghyB9DkBV1TRwBBjDdgO6sMXKotJYqIoPgL3Ab52kSlJ/A6wvHf+JpmmTqqrW9URVVc8E1jpO3TrNU27Fdk4AXtmEssaCIAiCMGcoOErehlVpCoKrWMVC2DLlDY8TUmPN48XxyvFYcWTaeeutjjVacM+1L7u3cmxhcdeRO1nZvop1nfa2KKg6VhDRDn9jv64XXVydOxKcq5NoUITUSkyPheWEzHO3wZOckHAJzUfSNO2fgTdjh2ZBVYwo2N3PzwQuAl6KLVDOKL2+8zeoAD8FLio5LE1FtdXGh0oPv6tp2kyre13pONY0TTs6zfiHqX4+KeAlM3w9QRAEQRBKODffYVWaAn/uRjGkztg+J6TGmseL1a7jo4WpnRDLsuqujjXqmLcWRyarYW0uJ0Sp7S7EOv2N/Tovuqi6xloipK0xEVJr3umqbtWLV4RITki4hBrMpmna94B1wL8Bk57LlufY+VjB7gXyFuyciullf4OUqnD9B3Zlr3Hgz2cxjTPU7LHpBmuapgNP1Xi+IAiCIAgzwFkVqrlOSDgixJcTErBmy7I8TkhVOGwf38aXdt7C1tHHXXNY+HNWgnJCRgvTixDn6xWcoVMzcEJSK1a4NvC1nZDZ54RoB7J89PsnMPELgXiTRIjkhIRLKOFYTjRNOwT8uaqqm4BXAlcAFwN92M37UsAIdhL3HuDnwL2apj0S9lqm4T3YDRYB/l7TtMNTDa7BmY7j/XU+5wBwSem4vrgvQRAEQRB8uHNCwhMhuTqrWM2UejqmTxqTroTw8eIYpmWiWzr/vuMzAOxIa5x93jlElVhgUro9tz8cq1ZoV0+it1I1yylCdGd1rClEiDcRvG39eveAaLAIaWmfnROSL5p2E8NJGI52Mt9wu0VKYvYd053EPInpBu73IeFYjRG6CClTcjO+U/o5pVBVdQD4ROnho8Ats5zKKZGP1fkcZ8hWb81RDZJOp3n00UebNf2UPF+vKwgnE/meC3OBU/17Pkp1w3zsxDEePRHOesdwOwZbn95KTwj/ZWdxV6camxj1fcajuIWCicmWx37JUaphUpPGJFse20IrbWQITi3ds283qX3V/h0mJmMEh3YtLSxnBFuEHBw+wKPD9pomHUEtzzy1nRS1RUOyv5/ooF2b52BPDwcc7yuvByf2b9O2E4nMfCO/db/CaNoWBMdjPT4R8rA2SKS7/u/CVN9zZweUgkeIHTywl0fZU/frCG7mqo/0OaAbMIH3a5o227IX7Y5jb/hZLZzj2muOEgRBEARhSnSqToLh6GzdKEXc4VdmCHNbWOieeQ38249Jj1AByJBhD7s84+zthPMzcOJdc47JwLCtGDGWs7Ly2CmUdMf6okzdCLJ4+eXoK1eSf+Ursbq7XdesiH+7WVBisxIgAMOZ6vGJaLfvetoI7x57/uUvx0qlKJ57LtmkO+xMorEao2lOyKmKqqqvx06gB/h3TdMauW3ivCUQ7If6cXqxjRXInoL29nbqrfIVFuU7CRs2bDiprysIJxP5ngtzgdPle/79J79FeV/f1d3JhlXhrPdbT9yGUx+sVlezsr2xXswFM283H3CQbEmy4Sz3mh8Z1vHeXO9d0cPBvftd2bRnrF7MWV1ncyC7H7b7X2/h4oVsGKjOvS+zB571j1M713HZisv44Va7T3M+kmPDBRuwLIuvPlYVOC+68EVElGl23VdfHXj6nh/90neuoMS5eJbfr/HoCPdss6Pgj8f8ImT1mqVs2DD976uu7/mGDVjveAeKotD2z9vIT1Q/E/XM1WxY60/Kn0tomkY6PbtCr3NKhKiq2k419Oow8HcNTukMGq23KLUzcLJe90QQBEEQBA95Z3WskJLH7Xmnz92Y8ZxGfRWsnEnpZX4+eL9v7ESp50ftnBD3eGdlrDXtZ6JbBsOFIV636PW0RFuJK3GKVpGcmSNn5FyVvGJKbHoBMgVKQE5IMTr7Xh6Xru+iuz3GaFrnRLTHd/2is+YFPGv2KIrt2KQS7s9AqmM1xpwSIcA/A0tKx3+uadrEVIPrwCn96nU1nOOkR4ggCIIgzJKiq0RvOA0Fi2bR12NDD6FEr7c8rz2vX4QENSfcnfF3My+LkKDyvFDtE2JZFkOFEwznhyrX+lILeNuy67Esq7LB7op3c6JwvLSGUX5w8HuV8QtSAzXfV10EiZAaHdjrIRmPsOn6FWy6bQ+Dut8J6ehqrPRvLUSEhMucESGqql4I/HHp4V2apn03hGmHHccL6nyOc9xwzVGCIAiCINTEsAzX3f7QKliZ/j7JYcxdby+P8To6pAOkS45JLRFSFjjf3P91fnni565r3XF7414WIABdia6KCLn76E94cuyJyrXrzngzjWAp/u2mEW2sgpW6pJXNN67js9/ZC3e7ryVT4VTH8uITIVFJCmmEOSNCgHOpJuKvUVX1oSnGnuE4vtAz9h2apu0oHWvAVaXjpXWuwzlOq/M5giAIgiA48G6+wyrR6y2jC2GFY/mdkKA1B4VjBVGvE+IVIFAVIU66HOceGqrmcGzsu4K1nWfVtaYgtANZdhzNs8hzvtCAE1ImGY9wxYXzmfi/FjrMaoR7skkZ414RkhAnpCHmkghxsmYGYzuAFzsetzmOnalgF0w3kaqqMexu8UHPFwRBEAShTgpNyNuA4LCpWh3TDUtnV3oXS1qX0hKdOio7H+SEBIRjTeWEtMc6SJfEx0RxAsuyeHwkuL6OYemYVnBVr66EP4+iK0CYJCJJ3rj4t2uuZzrK/TwuMf2iYEKPki+aJOONCYal/UmeirS7REi0Sf07UglPnxARIQ0hPlJj3Oc4VlVVnS4kawNVEZMDftWUVQmCIAjCCxxf9/GADf1sCHIsagmcr++9lc8+90k++ezHXBv+ollktODu91FvmNe4XluEbOi5qHI8oY/z4PH7eWL0scq587qr90N1y6gpnqZzQsosb1tOMjr77uNbto0xmtYxArabeeJs2VZf6NlU9HcnGIuenI4Hqbg3J0S20Y0wZ5wQTdM2A5vrGVvq9v4PpYcPaJp2RY05n1NV9VlgLaAA1wM3TzH1Ox3H92iaJonpgiAIL1BGCsM8PPxr1nWuZ0lrvRG7L1x0UycWCW/bUTCaE46VCxILNQTO02NPAnA0d4TB/DEGUgvJ6Bk+uu0fmNDHuX75Dbxo3iX2emuIEGdyuG7qpHV7a6Cg+Pp6XNh7MQ8ct+9/Hp48xPcO/lfl2sa+K1jaupyto3YdYMPSA4UPQHciSIR0+c4tb1vpOzcTjg7bvyND8SemFyLxyvVGiEQUJpOdrgYIYTgsQfhzQsQJaQSRcI3zecfxB1VVnR80SFXVtcDvO07Ntku7IAiCcBpw296v8sND3+eWHZ+peUd6rvDA4H389RN/xuY9XwltTu+mvhhaTkiQE+L//emmTtaoNvYbKTkfdx25k3F9DAuL2/dudswbLAiclbgm9Go+SHvM3RgvrsRZ0lIVs7qlV5yUgdRCrjvjd4g6Nvu6pQeGlgG0BTgHQcJkZVtjvVEGeu28j0AnRIlXrjeCdiDLjqi7ete7PrEd7YC/6WOjpJJSHStMRIQ0zpeAvaXjPuAuVVUXOweoqroO+B+qzQ1/rmna/560FQqCIAgnnecm7NojE/oEhycPPc+reX65d/BnFK0iDw8/5AtTmi2+xPSQSvQGhk0FOCHl3IwyIwW7BK42Xk33NB0dD4OcEHCHZDmT0r3ORH9qAclokkRAQvf6rnOIR+Iup8kwDZ9bBHbIlrMqVvX1gsKxVgSuuV7K/TwOx/33Z41onEvX+92XmVDOOXkkuRYtsZScEmdz99WMpnU23baHfLHxTvdO/OFYIkIaYc6EYzULTdMKqqq+CXgQaAUuBnapqnoPcARYDlxJVfAdBd7+PCxVEARBOElYljuMJhZQonQuMalXk4ZzNe7OzxRfTkhIzQqDnJBiwNwTPhFii6vj+cHgeacRIbvTO7l979cq5zvjXZwZa62I2UvnvwyAjlgHQ4Uh1xwLU3btqajje+Z1QvqTC3jL0rexuj24No9XhMxP9tERb6wbeKWfx60Wj0yu5aLJasv2ts7WhuaGas4JSoQvzXsTEcvALLlBo2mdLdvGuPJ8fxL+bJESveEin14IaJr2GHap3v2lU0ngdcC7gVdQ/ZyfAl6hadqBk75IQRAE4aQRVqWmFwpOFyAsx8IvQk6uE+ItpTtcGKZoFl2/e6f4rBWOpZs6I4VhPq19gkGHgOmMd/GGxW9iILWQ87sv5GV9lwPQHvMLg4Uti0qvVw3HMizD9V56Er2s7VxXMy8nFXU3+BtotEFhCXVJK5tvOouDl/42OxJLKue3p9saDpvy5pSYntyTMHJOnDhFSERpXhWuucLcvjUTIpqmPaiq6jnAu4A3A6uBXuA48AzwLeAbmqYF/yskCIIgvGDIee6mB91JnyuYlul6/6E5Fh6xEFpiep3VsdK6W4SMFoY5mN3vOmdhVRLPazkhhqVzcPKgKwldQeH87gtY3raCv1//Edf4jniHdwoGUguBACfEIXySkZlVuepJzJvR+OnYui/HQz3XcmXmUXJKgm2pVZilsKnNN66bVSL5dDklYeScOHGKEAnFahwRIQFomrYJ2DSL540Dnyv9CIIgCKcBPx+8j12Znbx24TWVzVyjeBOC9TmcmO5Nyg8rgdzfrNDAtEwiSmNBHvWW0p0ousOxhgvD7Mvu9a2paBVIKMkpc0K8IWB/v/4jLKjhRHR4EtZ7E/MqLobbCXFXx0pEp9+QX9hzEY+NPIKCwuV9V047vl62bBtjLGNAJMFPO17iutZI2FQ552Q07f/9dLfHGs458eKNvmpWFa65wqxFiKqqJ63eoKZp+6cfJQiCIAgz43h+kG8fuAOww4Tes+oPQpnX74TM3fCsZvXzCNrUG5bRHBESICLHPU7ISGGEvZm9vnFZPUsikZwyHMuZJ/PS+RtrChDAl6dRzgcBiDpCrXTTHY6VjLjDrYJ44+LfpiPWwYr2VZUQrzCYLixqtmFTlZyT2/a4hEh3e4xN168IVSBoB7J8/kfVAhP5osW7PrGdTdevQF3SeH7LXKQRJ2QveApYNwcLcWwEQRCEJjCUP1E9LpyYYuTM8CZfz2UnxCsWwgqb8oobsF2FOPGG5q03MT3tcUKKVoHt40/7xmWNLN30MGlM+q6B3wnx5mZ48TohAy1V986Zg2JYOoUZhmPNS87jd5a+ddpxM6WZYVPqklY237iOLdvGODpcYKA3waXru0IVIOUqXJmcu9rWaIPhZHOdMDb3EhQnCIIgnJbkHRvZMJPJvRtZcUKqhPU5B4mQMAROUG+NwMR0jxMCVBoNOsnqGQzLYF92T+Vca7S10mNEt3SXczadWJjSCfH1CXGIkAY6nzdKs8OmkvFIqFWwvFSqcAXQjCpcc4VGZZsIEEEQBOG0pejYpIUVJgT+cKy57YQ0KxwrwAkJofJWLiBsKjAx3eOE1CJrZDmQ3V9xQrri3SxuOaNy3fCU0p2pE7KoJViEeKtjzTQxPUzKYVPd7e57380Im2oGzQonm+s04oQ01sFGEARBEJ5nmuaEeMOx5nB1rGb18wjKCak1t2VZ5M28b4NfrlzlZCon5PDkIb62+8t0xrvqDt+bNLIcmTxceax2rHV1Ri+abickNU3uRkvUnX+wIBUcjtVodaywORlhU83iZFfhmivMWoRomrYvzIUIgiAIwsnGuUFurhMyl8Ox3GIhNCckoBu4EdArxLAMPvnsxzg0eZC3LrueS+ZdCsC+zB6+vOuL9CR6+dMz/7LSiTwogbwsUO889AMO5w5xOHfIN8bJstYVlfCrrJ5Fm6g26VM71/H4yKOuuZ05RMlpnJD+VH/luC3a5hJWzhK9hicxPfE8hmOVaXbYVLM42VW45gqnvvwUBEEQBCCtTwT2cGgEZ+JumG6F9Amp0qymgsHhWP5N4u70TvZn92FYBvcP3lM5f9/gPYwUh9md2clTo1sr5wOdkNLv78mxJ6ZdV2esi3Wd6yqPx4pj7E7vrDxWO9b6Esidwme6cKyWaCs3rHgfF/Rs4A9X/6nrWiziLtFbOEXCsU53TvdwslMVqTolCIIgnPLsnHiOz+34NDElxofO2sS85PxQ5i00KRzLXx1rLjshJzMcy/85jxZHK8fHc8crIVjOymhjxbHKcZATUu51ElNi035Pzu4+l9ZYW+XxtvGnKiK0P7mAnkSvq2u5bnoT06cvpbuh92I29F7sO+9yQizjlArHOt05ncPJTlVEhAiCIAinPI+NPIphGRiWwVNjW7mi/xWhzFtoUmK6vzqWOCFlmpmYXq6OldEzPD32JGrHOpfAyJmTZIw07bEORoojlfMZ3U4yL+eOeCkLj75kP0dyh13X2mMdpPVqkvq5Xee5qmQdnqyGbqklh8Sdu1GcUWL6VPhyQk6xcKzTndM1nOxURUSIIAiCcMrj3KQFbT5ni3MuEzOUjtsgTogTX05Ik/uEAHx51xfYkdYYSC3krM71rjHH88dpjbYxVqiKk7SeAexGhVZAC7Ty78+0TN+1pa3L2Jl+rrIetXMt28efCVzzyrZVgFcsGDN2Qmrhr45Vf+lfQTjZnFQRoqpqB7AI6Abq/tugadrPm7YoQRAE4ZSnWQnk+YBQoYTS+GbNG9Ij1bGqhCdCgjqb61iWxY60BsDR3BFaPdWkTuSP05uYh0k1NyVTci5O5I9XzjlDr8p/5kx/w8GeRC+/s+St/PLEz7my/5UkIknfa5bpS/YB7rAp3ZxZid6piCgRIkQwMbGwXA0SGxE3gtAMmi5CVFUdAP4YuA5QmXlvEemYLgiCMMdpVu5GMSBUKBHCHWNfYvqcdkJOYp8QTwgSwIHsAdfjE/nj9CcXuM6Vw6cOTR6snFvaupzdmZ2uNU8GFEboiHXwkvkv5SXzX1o51xpt840DmJ+0K1fxKC8AACAASURBVFs5c0IMT7PCVINhU1ElhmnZn0225PAAJKNSRlY4tWjq5l5V1d8FvgS0l055BYgVcE4QBEEQXDRLhHg3rGF1Npc+IVV8Qi+0zzhIhBhMljqRV17fco87nh9k1JEPApAx/CJkeVtVhBStIoZlBLov3g7mAC2xFt+5VCRFe8zeDsUdTkjezFc+EwWFuNKYWIhFohSN6txlxAkRTjWaJkJUVX0j8A2CRUY54HKm1wRBEIQ5SNEVjhXehr5Zd+lP1z4hPz78I3ZMPMdazqKPBdM/oQ6aEY5lWZZP3ID93ch6RIiXE/njjBQ8IqTkhBx2iJBlbcsd8+q+YgNllIDtSpATMi85v9IY0RmOlXEksaeiKV/zxJninNtJuQ+KIJwqNKWumKqqrcCXsYWEBTwKvBroAH5aOo+maRGgE1gPvB/4DVXx8QUgoWlaFEEQBGFO0ywnxJ80HY7AOR37hBzLHeWuI3eyI63xOI+ENq/XbTJCEGRFqxiYQG5YBpP69CJk1CNC0noay7I4lK1WslrSuqxybGLUFDeLWhb7ziUjSSKeLVZfstpk0BmO5aykFYZbEVP826a4kgil4IIghEmznJB3AvOwBciTwOWapmUBVFV1DdQ0LQ1sL/18WVXVG4DPA38ALAOuadIaBUEQhNOEZiWmNytp+nSsjjXm6KeRZeqN/ExoRjhWrQppuqW7krGDGCuOMZg/5jpnWAYnCscZ1+2KWYlIgr5kvys53SkWABamFrG8bQWr29f4XkNRFFpjra7nzC8lpQPElHjlOOPI22gkKb1MNECEJKU8r3AK0ixZ/ErH8Y1lAVIPmqZ9FVvEKMBrVVX9y7AXJwiCIJxeOB2LMF2FghF+OJZpmT6H5XRwQpwbe4Nwupp754Vw3KagUCx7bp2skQm85mTnxA7fuecmtMrxwtRiIkrEJRYmiuOV4xVtq/i79R/m7cvfVTN8ylshyy1CqkLB2WMkDCckKBxLyvMKpyLNEiEXlP4c1zTtZzN9sqZp3wbuwRYifxXmwgRBEITTj9PJCQlsdldjzYal85uhh3h67KmGX7dRTpYICaNSWNBnDHaoV3YaJwSqiehOnpt4tnK8uPUMAOKOsKkJh1hoifoTz720ePJC+pwiJFI7J6RRYiJChNOEZomQcijWswHXKp1+VFWd6m/xt0t/LlRV9dIQ1yYIgiCcZjQvJ8TrhDR+lz4ogbnW3f8tJ37BrXu/whd2fo6dE881/NqN4G7cGKII8bpNzQ7HmiYnpBbPjVedkMWlPA+nE+J0LOoRC62x2k6IOzHdGY7VuFiIRiQcSzg9aJYIKYuLiYBrztsPvVPMsdNxvLrhFQmCIAhNp2AWeHL0CV/8fCMYlo5hVTfFYSWPm5bpK98axgbZmw8CtZ2QXenqf3XaRNB9u5NHwdFg0cDfGXzW83oEgxGGCDGCnRA7JyRYhNRqIFimnA8CsLjFdkKcjsVE0SlCpndCvI5ET6I38JozVC+cxHRxQoTTg2aJkPLf1KC/8aOO4+VTzOEMshxodEGCIAhC87ltz1f50q5b+PSzH8e0wtnIFjzuRFjhWMUA16OWCLEsi0OTB+vaQAc5IbVyQpwb26H8iWnnbiYnLSckhN9fLSfEsIya4Vjndp9f9/yLyiJEcVaxmlk4ljc3xZkwHovEvcOBsBLT/SIkjAacghA2zRIh+7BFRH/ANeetnpcGXC/j/Nfi1M/oEwRBECpx9cfyRxl3JPI2QrMqWAU1nqu1Qb5j32388zMf5jPaJ7Esf2lYJ97yvFPNO6FXP6Pj+eNTzttsmhaO1eDvb6QwzIPHH2CkMFxzzsrcZrGmE3J+94Wsbj9z2tfrjvfQFrPzOWI1ckJSdTgWzjArL0FldKF5JXolHEs4FWmWCNlW+nOFqqrev1G/dhzfoKqqT7KXnvOHjlN7Ql6fIAiC0ARcd9NDEgu+Eq8hOSFBG9mauRtDvwBgd2YXR3KHp5w3FyBuajkhTqE2VDiVRIgZ2IdjdvN6erHM4PdnWRZf2PlvfGv/1/n3HZ+pCMDaOSEG2Ro5IV3xbq5ZdO20r1kOxQKIO3NCZhiOtcpRuncgtdB1LShkyp5XqmMJc4dmiZAHHfNf7rygadoWYH/p4Rrgu6qqVjoCqXYjkR9TzQPJA/c3aZ2CIAhCSNg5FtXNdliOhbcSUlg5IUEVloIqN3nDtoYLQ67HznwVgFxAOFDQxtu0TFeIz1hxLNCdOVn4cjdCcEO83wmY2fdiqHCCQ6Uu5kdzRypuhLNxoLNj+VQ5IZ3xLtZ0qHTGuirnFqT80d7lylhQ2wmpJxzr6kWvZ15iPh2xTt698v2ua7XCsZJhVMcKSEyXcCzhVKRZIuRux/EbA65/hGrOx+uB3aqqHlNV9RjwDHBF6ZoFfEnTtFH/FIIgCMKphHezHl7YlKfEa5McFggWON6wmmO5owDkjTz/sv0f+dutf80OR2WreqtjZfS0z20Yyg/5xp0svJ9HGCJkJnk3o4VRPqN9ki/u/PdKR/O9GXcgxFDBzpsZzFWbDfanFlTXbAaLEAWFjngHAH925gfojnfTE+/lyv5X+sYudnRAb6Q6Vle8m01n/xP/dO4nfF3Va4Vj1RPmNR1BgkPCsYRTkaaIEE3T9gL3AsPAa1RVbfVc/ypwO1UhogB9pR9nQvoW4KZmrFEQBEEIl2YkIEPzwrHyQSIkYG5vT4myCPnZsZ9wILufjJHmtr1frVwPrI5l6b5cEued9TInnse8EO/vL4y8kMCQtxq/v7uP3sWOtMZTY1v51+duZrQw6hMhwyWRdjR3pHLujJYl1bmt4D4hvYneSmL4wpZFfPScT/CRcz7G8rYVvrGLHOFYzrApp3iqJxwLIKJEAjuYO8WNkzCckBVtq/zzhiBuBCFsgoMSQ0DTNP/tBTfvAh4D/hZ/AnsWuAX4B03TggM/BUEQhFOK5iWQn5x5a83tdUKOlkTIM2NPV845Q7SCnJDy3M4cg6DE/XpEyEhhmK/u/jLJaIJ3r3w/LdOUnq0Xb3haGE5IYPJ/jXA6Z4niE/njfG7Hp3wb+LIT4hIhrUt4dORhwHZenOFwr1/0Rh4dfpjXLnq9ax5FUVBQaI+1u87HlBgLHM6KMxzLST3hWFMR1MsDIBVC2NS6zvW+c+KECKciTRMh06FpmgV8VlXVzwEXAsuAOHAY+I2mac9fYKwgCIIwY3zhPM0SIaElpgdtkINESLATUqsXSlBiuj13kThVEeKsjFXmROEEhmUE3j0v88sTD7I7Y/cXeWz4EV7at7Hm2JnQjHCsIKFXzqGxLIu7jtzJkcnDvGbh61zCAqqfs5Oh/BCTRpaxoh2lbYuGal5HxqiGuKUiKa5aeDVXLby65vrKVbDKDKQWuhK74zUci0YTyJuZmN6f6md+oo8TjkIHyUii4XkFIWyeNxFSpiRGHi39CIIgCKcpvtyNJlWx0q0ilmWhKEqNZ8xuXgjumO51QtL6BGk97coRcBJUohfsz6PFoS0mApyQ+wfv4cHj97OybTVvX/5OV5ftMmPFalO9oJCu2eIPx2q8z0stt8myLPZn93LXkTsBeHy0vi3AUOEERyerYqU/tYCEY4Pt/EzrcYgSkSRxJVFpWumsjAW1nZBUpDEnpJnhWADrutbz4PH7q/NKOJZwCtKsxHRBEARhjuF1FprlhFhY4eQrBHTdDnJCgsTG0cnDgdW1YKpwLLfAqdVHxbAMdqQ1/mX7P/L02FNTzh9WaBo0pzpW7VK6eqXqlZfL5l9Oe6wj8NpwfqgSDge2c+F0FcYdZXRbYvUJBWdI1qJWjwip4Vg0Go4VVMEKwklMB1jXeZbrsYRjCaciIkIEQRCEUGikFOtUBImFMFyWoA1y0LxBTee2jz9Tc76gxHR7bvfnM52LMWlMctuer/gS2vOO+YOqT80WvwgJ4zOuFZpWe+51nWexse+KwGvDhSFXnxZv+FTOrOaD1Jsr0x6vipDF3ipWAaV0FRSX+zIbmu2EnNmx1nOmMddQEJrBSQvHUlW1DXgJcBF2FawuQNE07d0naw2CIAhC82he7kZ9CeRTYVg6EaKuEK4gJyMwJ8Tw5348OfaE71xWz5BIJCoVnKabOygcy//aGSaNSVpj1Q11ziHKgsoMzxa/CGlOOBbY341aYWsr21ezsn11JVTLSdEqsmNCqzz2OiFOWusUIZfOu4z/yn6ThS2LWNOuuq4FzZ2KtjQcClgzJyQkJ6Ql2sK6zrPYPv4MUSXqCzMThFOBposQVVUXAH+HXQ3L+S+Cgt0HxCdCVFX9KXYjQwt4laZpu5u9TkEQBKExTmoVqxkIHG18O1/e/QXmJ/r4wNqbKnexZ9snBODw5KHAcYlIgmN5O1woQpT+VH8l4dqbbxKUmA7Qn1yAYekMlSpupfW0S4S4nJBQw7GaUR0rWIQYVrAI6U/20xnvBOzKVnce/gGr289k0shWwrf2Z/dVxg+0LHQ1K3RSb8jUxv4rOb9nA+2xdiKKO0AkHuCENBqKBRBVorRF28gY1e9WTInVzEGZDW9f9i5+fvx+VrWvrnymgnAq0dRwLFVVXw5sBf4IaMMWHuWfqfgBsLz0847mrVAQBEEIi5NbSjc4DOlXJ37Jf+3/ZqV6EsDndnyaSWOSA5P7+fXQlsr5uvuEBIiQIDJGmn3ZvZXHZ7Se4bob7xUMzpyQi3peBEAikuC9q/6QNkeegjcnxZkT0sxwrGb1CQH7uxEUtrayfU3l+KqFV3PzeZ/lL87868AEfQWF/uSCmpXEZlK6uDPe6RMgUMsJadytUBSFVw1c5ToXZn4PQHeih2sX/xbru84JdV5BCIumOSGqqr4IuBNIURUde7BFyYuARVM8/Q7gM6X1XQt8uFnrFARBEMLBF87TpGaFELxhOzJ5mK/v2wzYuQN/sPpPAsZUKytNFeZVNIs8NvIIPYleX4neWmT0DEcmq/kKy9tWusrOOp0Qy7Jc4uJ3l72di3pfRF+yn4GWha7EbO/rO8Oxgqp5zQbTMn2faTgd0+sPx+qJ9/KKBa9ynSs7QPMS831z9CX7iEfiNfMrnO7RbAkUIQ1WxirzygWv4amxJ9mV3gHA/IRfaAnCC5mmOCGqqkaxO6KX/6ZuBzZqmrZK07TrAH+5Dweapo0B92OLl/NUVe1uxjoFQRCE8PB1Nj/J4Vi70jsrx0+NbWW8OO5zMZyhNIF9Qkrz3j94D7ft/Sqfe+5TDOaPVa6XHYsgMnqGPZlq9PDythWBHbdHC6NoE89WHiciSVqiLZzTfR4DLQsBd8UmbwK7OxwrnJyQwH4eIYiQWhXEilbR5ei8c/m7+cdz/oVFnsTwMvOS83znlrWtBGpXmqo3J2QqgsKjwgjHAtsN+cPVf8KS1qUAXDzvxaHMKwinC81yQt5KNadjO3CppmnTZ+C5eQh4FbYQOQd4MNQVCoIgCKHyfHc2zxpuwfHoyMOc4UnIdY6Zat4fHPoeQKXxXZnrlryZR0ce9p0HO2xqX2ZP5fGKtpVsHX288rhoFhkrjvGRbX/n2px3xvzx+rWcENMyXc8NqxdLkGPRzHAswzJc4VjJaGrKZG+vE5KIJLlm0bUArupYTsLoJB/ksoQRjlWmJdrKX6t/y4Q+Tk+iN7R5BeF0oFk5Idc6jv94FgIEYJvjeHWD6xEEQRCazEmtjhUQhjRaGHE9fnjoIQ55EsidrkJwx/Ta4U0t0Ra64t1c0f+KwOv7MnsricZt0Tb6kv0+J2THhOZzBzoCkoZrOSH+hpDhhGM1ywmpHY7ldkJaptnY93qckOvOeHMlT6RZvTwgODE9FZITUiYWiYkAEeYkzRIhG0p/HtU07YFZzuGscdjT4HoEQRCEJuPLCQnNCamvlO6IR4Tsy+5l6+hjrnPpYtVVmGnVrbaoLQyuWfSGSsnThalqeuO28Wqk8bK2FSiK4trEFs1iYFneZEDPifZ4sBOS9yRzTyWaZkJQ2FSzmxU6c0Km6+g9kFpIf3IBAOd1n89l8zdWrtUs0RtCTsiKtlUkI+5Gf60hixBBmKs0KxyrHzsUa0cDczj/pZVWn4IgCKc4XrFwssOxRorDvnPPOXpKgLvSVMGoPW97rMNXlaot1gbY4Tg3rfsQWX2SXekdfHn3FwA7xKjM8rYVAD4nZDygLG93wn+fzRmO5VyHN5m71iZ/pgQ5Fs3MCfFWx5ouxCmqRLlx3Yc4NHmQFW0rXaFbYVTHqsW85Dz+Zt3f8+PDP+LRkYeJKlHO7T6/4XkFQWieCPEHy84c57/KozVHCYIgCKcE/lChk5uY7g3HCiI9RWiTc954QC6As2xuVInREe+oCBMvS1uX2fM4nBDdLJIu+rukr+08y3eu3VWi1+GEeLrHNzPkzQyhWWG91bGmc0LADq9a7SjhW0ZRFGJKzCdMgypqzYb+1AJ+f+V7eVPxLZiWSXdCauUIQhg0S4QMAiuApQ3Mca5nPkEQBCEELMvijn23sTuzi99b+nZWd5wZyrzeDWd44VjT9wkpmkVX7kRvYh7DBX/n8oyewbRMIkpkynmDelg4RchU5wAWtthhWlM5Ia9a8BrWda7nzI61vue7REixdjhWMaRwrKZVx3K4TalICzlzErA/i/wMnJDpiCpRlwhpj3WEEo7lRBr+CUK4NCsn5JnSn8tUVV05yzne4Dh+qMH1CIIgCCUOZPezZegXHM0d4WfHfhLavM2qjhV0N73ssuyY0Ni85yv8euhXlWvd8R5eM/C6wLlMTLJGtrTe4FwTy7JcSdNl2gNcj9ao/1wikqA3YSdSx6bICTm3+3zUznWBVaHc4VhVEeINxwqrT0izRMhQ4UTluNeRfF0wC5XXVFB8eRczxVvFqpw/IgjCqUuzRIjzf7W/nemTVVV9HXAxdljXs5qmHZrmKYIgCEKdjOtjleN0nY346uFkd0y3LIvNe77Cw8MP8c39t1eu9SR6ePG8l9ScL61PYFlWzXCsolUMDEUKdkL8ImQgtbDSfXsqJySoKlaZlmgLkdJ/0TlzslIFy+uEFJooQmqV6D2UPci39n+D7ePbAq9X58xzIn8csIXG4tYllWvO/i3JSHLK8rz14O0V0p/qb2g+QRCaT7NEyDeo5nHcoKrqDfU+UVXV84BbHac+E+bCBEEQ5jrOvIKwSrwGzRVGvoJpmYEhR7qpo1s6o0V/Hkh3ood4JM7vLHlr4Jzp4gS6pQcKDd3SA10QsMvueolH4iQ8d/GdFbOmqo7VEdAfpExEibgETnnT7ssJKYmxRqnXCbEsi6/s/hIPHr+fL+28hbFi7ZTNI5NHKv1U+pMLXCVznbk5YfTd8FbIEidEEE59miJCNE0bBf4Ru9EgwJdVVf2aqqrraz1HVdVlqqr+I/ALYB62C7IN+Foz1igIgjBXcW44w0ps9s4L7mpRYc1Zxlvi1UlP3A772dh3Be9b9Uf80eo/45yu8yrXJ/QJV4hXKlLdHOuWzmSNeWvlGHjFSTkfBNxhQhP6uKNLemLazXdQhSxvroqFFdLnXF+J3iO5wxzLHwXsfJQtJ34B2OLo/47+lO8d+HZFaB129GhZ1LKYmKOKldOBS4YgQrwNC/tSIkIE4VSnWYnpaJr2r6qqng+8A1tQXA9cr6rqIFD5F0dV1ceBBaUfqAqXEeCNmqaF9z+kIAiC4NrYh5XY7J0XZt7D4u6j/8svjj/AKxa8hsv7r7TXN5UIKSU5eyk3flMUhfO6LwDgCUfn8rQ+wfFSmBDYSeB6sVgRCBkjOESt1ma/LdbmKg/sFCHxSPW/2ZFCdUyHQ2DUIqhCVpDwKlpFYg3+dz6VEzKYG+Tb+79OV6K7kutS5hfHH2BhahHfPfjtyvvTLZ23LH0bhycPVsYtbFnkCs9z9j5J1VEZazq8TsgCcUIE4ZSnaSKkxA3AEeCDVMXFAmxRUvaPz3VcK7MDeIOmabubvD5BEIQ5h/Oud5jhWI10TD+Q3c8PD30fgB8c+i4v67u8ZgUre+5iTSckuO+Ge0O/O72z8nhZ23LSYxNVERKQJxNTYqztDDbzWz15Ic5wLKcTMuwUIXVUWgpyQoL6buhmERpsoDdVTsj/z96dx8lx13f+f/U5V8+hmdHovo/S5VO2LN+3AXPYBnPHgIFsNvmFH2SXHIQkEEI2ySbZHJCFhMUJZyABL46NbcCAMQRjbPmSLakk67BujY65r75q/6jp6qrq6pme6ZqRNPN+Ph56uLu6+js147FVn/58P5/P452PsatvZ+D7ujPdzpyUgv399l/d7kzIorrFHHEFJQMhZ0L8NSVza+dWvaaITK2pqgkBwDTNnGmavwdcDfwHOBtwI74/BaeBPwA2m6a5ayqvTURktnLfyIY1cRtKsxYTKUz/7tH/cB6n82n6Rgu4ywUhmbG2YwUEIe7MQ1+mj70DxSBkVWoNcVfGwr1VaHHdEt6w8A5+ffWHaE40B369vOWtLZnj6gLl7o7lzYRUEIQkSjtkBdWrhFGcPtawwhe6n5/QWseHj5O38hwp2Y7l/hkXa0LqQghCutPe2hR/nY6InHumOhMCgGmavwDuNAxjDnAtdvajDWjALmA/DvwceNo0zeo3t4qISFme7VghZUJyVq4k6BgvCOlKn+Ghow+QyWfY3vOC77UumhMtY2RCsgznymzHSrSWHPNnFfb173Wer0yt8t0gF4OQuTUdvG7BG8b8PnozPZ7nhc5YAIlI8F+zlcyc8GZvgmtCIJxAMng7lh1cddTOK2kA0FEzj1MjJ53i/otbLuWF7uewsMhYaQ4NHnQCyUQkSXvNXGJlAr1KBhWOx72FLlKyuUJEzkXTEoQUmKbZhZ0R+Y/xzhURkanh7rBUmItRbYvUoGAmlx/7M6UvH/hndvcFJ727013QsGLswvSAG3IIvsF339AfHHzV6epUG61lUd1iz7apiW4VqomV/9TdnQlxqygT4rrmAScTUrodK4xAcqyakKCi9SvarqQ50cJTp5/k8rYruKrtGv5m9/9k7+g2t+e6tjnnLqizWxa7A73CrBYIpzuWm3seiYicu6Z0O1a1DMO4yjCMx872dYiIzCTVbJsqp5Kp5m6dw51lAxAobl0qH4QE14TURus8mYgC99amkyOdzuPlDSvtG+Qyn9JXcoN8x6K3OI9/deWve17zF0wXVFYTUgxC+pyakODC9ErlrTzPdm3j1YH9nuNjdcfy/5zjkTiXzNnMle1X8xHjo1zdfi2RSIT5tQucc57tesZ5vLBuMVA+KxRGJsTNXzwvIuemac2EVMowjMuxW/zeeravRURkpvEXN2fyGc88i8kInj5ePhPy1On/9Dy/du4NHB86yp7+3QB0jW7/CVoXCtuxvDfH82rm88ZFdwaeX64b1crUKsAbLAy4Z1hUUFuwrnE9H177UfJWDqNxvee1cj/XyrpjubaQZcbojjWBTMjDxx7kkWMPAfBHG/+EebXzgbEL091f86r2a7mk5VLnfW7uIMQ9KX1R3SIAT6DnFkYm5Lq5N/DEyccBuGX+a6peT0Sm3jkVhBiGcQnwKeD20UMRil20wvoa7cA1wBbgAmAVsBBIASPAGeBF4DHgK6ZpniqzVNDaTcB7gbuBNUAr0AnsBL4BfN00zeC/UUVEpon/hjOMNr3lpo8HyVt5fnH6587z/7LqN7io5RKeOv1zJwjpTheCkOK6iUiSjGU/t+eEFGtCbl/wRl6/8E1lry9V5qZ/VWoNQNlMSCXbsSKRCGsbjcDXymVCKqkJcU9oPzx0kH39rwR2x6o0CBnJjTgBCMDLPdvHDEIKmRB39uXORW8JnBIPMM8VhLitblwLlM7yKAgjCHntgteTtyzm1c5jU/OFVa8nIlMv1CDEMIx64DpgBdAC9ALbgZ+Zplk6mrb4vo3Ap4HC3yDu4CN40+/k/Qvw+jKv1Y/+WYwdCH3KMIyPmab52fEWNQzjGuxJ8Ut9Ly0Z/XMb8BHDMN5hmmZwr0MRkWngv5HNTlF3pXLbsXb2vkz3aE1GY7yRTc0XAN6uUkHbserj9fRk0qNfL+O5Oa4bp0VtUEYiEUmyvGEF4G2l2x/iDItyn/5Xsh1rbk0HyWgN6fwIQ7kh/nb3XwXOKqm0MH1b1y89z90/23JBiGVZnkxI7Rj1L/MDgpDWZBtL6uy/FssFZGFsx2pOtPDOZb9S9ToiMn1CqQkxDKPBMIy/Bk4B3wU+ix1U/D3wY+BVwzDeHvC+NsMwvgA8jx2AuCsj06PrrArjGsvoBH4CfAX4/Og/n6bYSjgFfGZ0kntZhmFcCjxKMQBJA48AX8T+/gvrXQj80DCMxSF+DyIiE+Lf4jRVmZByw/3c9QJb2q50PiFvSRRb6xa2Y512DRV0Zw/8k80ryVi0+WoF7l7ydudTeO92rPBmWCQiwduxmiooTK+N1fLrqz9EQ8zOiJT7eVaSCbEsy9muVOAOtsptx8pYaazRzwTjkXjZbAbYrZGT0aTn2EUtFztND8oV6YddmC4i54eqMyGGYbQAPwAuxRtEWK7ni4CvG4ax2DTNvx5933XA14EFeDMfGeA+4E9N0zxC+B7H7s71WLlhiIZhLMcOSgobSz9uGMZDpmk+FXBuEvg2drthgG3Yk94Pu87ZADwIrMT+fr8GXB/GNyMiM9fpkVN86cB9NMQauHflB0ObfVCyHWuKuiuV67y1u890Hl/ccqnz2D1ksCfdTd7KO9uzANakDA4NHrTX9rXorSRjsSq1htNnThMhwruXvZcr2692XqumMH0sQZmQeCROXay+ovevbTT4iPFR/nTHJ8ueU8m/v/0D+5yfXUFfptd5XC4TMuzqxjXezyIaidJRM4/DQ4ecY4WJ9VA+E1Jb5aBFETk/hbEd6++AzaOP3YFHUEDy54ZhfA+Yi50xcf+NmsXeKvVp0zQPMUVM0/yrCs45YBjGm7AzNOuxr/2DQEkQAvwasHz08WngdtM0O90nmKa5wzCMNwLPYn/P1xmG8TrTNB+ZV6T08AAAIABJREFU9DciIjPeT0/+hL39ewB4vus5trRtDWXddM4/2XxqghCwP71333yeGjnJmfRpwB4ot6xhmfNaMpokFU/Rn+0nT56TI50cHHjVeX1d03p+1PkD+5ot73asSm5k717yDlakVrGiYSVL6r07Z6dqq1BQJiQVb5xQS+QFtQtpiKU8szDcgjJZlmXxnSPfZv/APq5o3cr3jz9ack5vthiEBA8rzHu2YlXys/Bv9VuZWu08jkdige+ppPhfRGaeqoIQwzBWA/dQDDIGsQOJJ4Au7IGE12EXa9djb//6OHAVUOt6378CHzdN80A11xMm0zTThmF8DXtbGcAlZU79Ddfjv/QHIK71dhiG8S/YQUvhfQpCRKSsQt0EeG8Yq1W6Hav6Fr3lPo3PWlmsvMXnX/ksJ0c6MZqK3aNWp9aUbO9pScxxMhHPdj3jDMNbWLeI5kSLZ11vrcL4N8gN8Qaum3tD4GtT1bkpEU0SIeJsaQJYXD+xHbmRSIQl9UvY1RdcThgURB4Y3M9jJ74H4ASyfv2ZYhewkVxpEGKRZ2iC8zwua93iFL+vSa0l5go8ym3HqnbLm4icn6rNhLirwI4CN5qm6f+/3TcNw/h74EfAfOBtFLMkp4B3m6Z5rs4CcXfGKmmtYhjGWmCd69CXxlnvSxSDkFsMw0iZphn80ZaIzHojrpvsoE+qJ71uQIveao01VHBPn8muvh0A/PzUT53XgjpKzUm2Ott5nj5dTD6vSa31BArZfJa8Vex3UvW2qTK1G9UWpieiCa6dez1PnHycjpoO1jVt4DXzy/VGKW9x/dKyQUjQv78jg4cDzoS3L3kX3zz0daAY2OatvNN1DOysUGF2zMAEJ5tfO/cGnu96loyV4d3L3+t5rVy2abymAiIyM1UbhFzhevzhgAAEANM0TcMwPgz8G8Xaj27gatM0X6nyGqaSu+H7gYDXb3Q9Nk3TPD7Oek8DA9j1I7XAldj1NCIiJYbzxZqHMAIFsG84/cMJK+2uNJayQUg+S/dokbnfmoAgxF0XcmKk+L/U1Y2G5ya2kGEpqI1WdyNbdjvWGN2gKvX2pe/mrUveGThEsVJL6peUfS1oO9aJYe9fRxEi3LnoLVw991onCBnI9pOzcuRcvw+JSIJ4NE42Zx+baH1Mc6KZj2/4ZOB2s3LZprCHFYrI+aHaIGTD6D/7gfvHOfd+oA+745QFfO5cDkBGi8k/4DoU9P25g5Rnx1vTNM2sYRjbgcLG7vUoCBGRMoanIBMSFCyEkwkpM1TQygSuHyFSUpcBMMfVIcttTWoNOVfmI5vPej69r37b1NTNsACqCkDAzoSUE/Tz7Rw54Ty+e8k72Ni0iY7aeQBOfYmFxUB2gJ29Lzvn1sXqsSj+nL2dwioLyMrVu0xloCci559qW/S2YAcUO0zTHHOo4OickB0Ut2J9o8qvHbrRVsMXGYbxR8CT2AETwH9it9v1W+t6fDDg9SDuovvg6VYiItjD5QrSIWVCqhl2N5ZyQVLOyjHk6mJVsLR+uadeoMCdCSmYX7uAxkST55P0jJWpeH5FJaZyhkUYOmrmlX0tMAhxZUJWp9Y4AQhAY6K4u/jw4EH+/VDxr+PNrZdP3cyUgJ9xIpIM/D0QkZmv2kxI4Sa9e8yzinpcjyu9aZ8yhmHcwtiZCAu7aP6DpmkGVW66m86fCHg9iDtH3lr2LBGZ9TzbsaywMiGlQchUb8caDghCruu4IfB898DCgq1tditd902su2DavpGt7q+zoKLpZLSm6gxGWMa6Dv+/v2w+y6mRYkmjP4BpjDdxnGMA3Lf/C87Psi3ZzhsW3sFLPS865051u2LNCBGZvaoNQgrzPcbMgrg455mm2TPWieeAE8D7TdN8eIxzUq7HpX/LBnOflyp7VpX6+/vZtm3bVC0/prP1dUWmk/v3/DCH2M7zrGYNazy9KqozQPFGu/N0J9tOV//f1hlOlxzbf3A/9QdLem9MyBGOBh5/aed2DlEskl7GClazlviBJNsOlH4/gwwSIepsCbqYzbQcaWXbkW3kCR7WF7NiVf9/p5PSxoaxfPXrhilGjFzAz+D4qeNsO1W8zm66nK5iDaR46fmXPOdnKX6m5g7mtqSv5OXnXyZDMag5fKqYvD/T2cW2zsn/PDKUBqqRbOSc+hnL7KTfwbMjjDkh57NDwD+MPo5iby/bBFwAzAMeMgzjO8Cvm6YZlOlwf4RT6ceU7o8h1RJEZAb4Pt8F4BhHWMZKkiTHecf4LCyyrptB941jNdxrFo8F39xPbN3g68uR89x8rmAVy1hRdp166rmG6znMIQzWs5BFzmuRMjuIkwR3tpqIKKVbghIhrBumK7mWn/E4AK20OQGlPzDpdW06aKK5ZJ26gL96GmliwejPOua6NRh2fW5W7c8jFvAzXsnqgDNFZDaY1UGIaZom8Jv+44ZhbAQ+B1wL3AVcaBjGVQEzQIZdjyu963BvXK40ezJhqVQKw5jekpPCJwmbN28e50yR85f/99yyLO5ztaVYs2k17TVzq/46w7lhe1zqqIbGejavrf6/rV29O8HXx3Degg42Lyxd+5enf8H3jj/Mle3XcMu828Zc99m9vwzcmLt67Wr2HttjtyUBNqzeyMbmTWOutZny3+dXn72vpBtUc30Lm9dX97Pp6+zh6UNP+tZtrnrdMF1iXcKc4y0MZAdYXL+ELx+4D4BUc4rNq4vXeeb4KThiP14zdw2bl3q/h5PHjrPjqDc7sqplNZtX2ec9vusHnB44ab9Qi/M33eqlq9k8t7qfx/Z9z7Ot62nm1y7grsV3s7HpggkNbhQJk+5bqmeaJv39k5s2MauDkHJM03zZMIxbsWebXAWsAv4GeLfvVPdPvdKshvs8zQgRmUYjuRG6M13Mq50f3pq+Ggur4t2pY/PXUYRVmB5UExJU2Jy38vz7oX9lMDfI/z3871zccgnPdj3D7r5dLG9YyaVzLmNhXTFLUa4mJJPPempb6kKoK8jkvNdbbcE0QCKgpiSMdcMUjUR57QJ7xsjO3h3Ocf+wQndRekfA73pjoqnkmLtTmbt2Y6JzQsZz74pf5a7Fd9OcaDln6m1E5OwIKwh5jWEYE8nnRyZwvmWa5rQHS6ZpjhiG8dvYnbEA3m4YxodN03QPMDzjely+dYmX+7wzZc8SmcWODh2hLlYXWKQ8WSO5ET7x0u/Tl+3lbUvexfUdN47/pgq499RDePM8hvPDnufTXZh+cqSTQdf39rVXv8zuvl2AfQP86LHv8r4VH+Cy1itG1y1eXzJa43ydnJX1dMeqrXIwXVCHpWrXBKiL15ccO5cnebuDJn+A6p6vMi+gq1ZjvLT+Z7FrDknZ7lgh/DwikUio/12LyPkrzI8hIhX8gWIheyXnu993NjwJTmVoDLjc97rpely+ibuX+zyz7Fkis9RzXdv40x2f5I+2f4yTI6XFwpP1Sv8e+kYnRD/XFV4R4mBJEBJOsOCelm6vG1aL3srmhBwa9DYwLAQgBRYWvzj9c+f9h13nt7puMrOWtztWtdOxgyabh3FzvL5pAzVRb5vfcy0T4pZwdfPyB5Gdw8USxqCsX1AmZHGdKxPiCnDcmT11shKRMIURhEwkUDgXAouKjc4+cXfx8jew3+l6fMl46xmGEccueg96v4gAL/VsByBPnl2uLSfVct8Il9s6NBlDWW8QEta2qaGSICSkTEguYDtWQCbEH4QEOTBwAMuyeLlnu5O5mVvT4flUPZvPer6XqjMhAW1ew9gmVBer56r2a73rnsM33e6WwoUg0rIsHu/8IX1ZuwAnHokHZh2a4t4gpDHeSHOiWMCuyeYiMh2q3eb0x6FcxTnKMIwE3lkg/u1TP/aebswr00WrYDPQMPp4GDvTIiIuI3n3lPBwOkKBNwgJa2sTUDKIL7RMiG871lQOK8wG/JwPDr4a+P5LWjazu28XA7kBhnKDnBzpZFvX087rm+dcTnemWKGezqedrVkRIiXZholKBGRCqq0zKbix42Z+3PmY87wn0xXKulMhESn2QikEkT848SgPHLnfOb6xeVNg3YV7WCHAorolnuLwoJ8xKBMiIuGqKggxTXNGByHALXi7Xnn2I5imudswjF3AOuzsznuAvxxjvfe6Hv/QNE0Vpov4pHPFm/gwhugVDLlu6v2FvNUYzA14nmescAKn4SnajhUUJLnXPjp0hPpYfdlMyM3zbmUkP8KOXru70u4+k+3dxeF2m1sv5yedP3Ke949+Kg/2J+nVFiMHD7wLp9t5W007HTUddI5uA1yVWhPKulMh4Z4en89gWRbfO/aIc2x5wwresfSewPcmfYFge02757kyISIyHWZVawrDMCquhjMMoxlvQPGcaZoHAk79367Hv20YRnvAORiGsQ6413XoH4LOE5nt3J/Uh3XjDf7tWGEGIVOUCfEHISFlbwJrQkaDvZ+f+il/uuOTfHz77zgF97XRWubXLgBgY9MFrEitYlnDcue9Dx39jnNtC2sXsbBukWerkLuwudp6EAguTA9z29T/t+YjzK9dwPzaBVzWuiW0dcPmqQnJZxjKDTpdyJLRGj689qM0BdR+BGlJencaB/2MQZkQEQnXbGvR+x7DMN4B/D3wgGmaA/4TDMOIADcDfwdscL30e2XW/EfgvwHLgbnAw4Zh3GWa5hHXmuuBBykON3zCNM1H/AuJiLd7UzakrAL4tmOFGIT4a0KmqjtWzsqRt/ITyiRk8hlikZjnPYHdsUav+YXu50peW9awgl9d9V85OPAqK1Kr7GP1y53X+1yZjktbLwO8N7HuTEgYN7GBmZAQP6Fvr5nLH278VGjrTRX/dqzeTK/zvDnRTDI69uiqS+dcxrNdzxCPxNnadpXnNXcQ6TbemiIiEzHbghCAK4CvARnDMF4GdmOP2Ipht8/dDCzwvef3TNP8ftBipmmmDcN4C/BToB67g9ZewzB+CBzDDk5upJh1Og78SpjfkMhMMuIqnA5z25S7ODrMmpCp6o7l344Fdn1FpTfyL3Q/zxf3fZ65NR38zrqPUxOzt+AE1YQUMiGvDpTWgSypX0pdrB6jab1zbFlD6cRz981sPFKcjO0OQsLIhPi3EoW17vkm7tuO1ZstBiGN8fEzIHcveTvzaxewKrW6pHg9MNsUrdFcDxEJ1WwLQtx/+yaAi0f/lHMY+JBpmt8Za1HTNJ81DOO1wFexW/DWALcHnLodeIdpmocmdNUis4hnO9YUZkIsywplUrN/Tkg6pDqWoCAkk89UFIRYlsU/7bV3fB4fPsZLPS+yudXuMO6uuXGvC5CKp5w2xgVL65eVnN+UaGJOopWuTLFXx7Vzr3duZj3bsTLuORPVBwubmi/gpZ4XPcfO5S5WUyUWiRElRp4cFhZd6eK/i0q2YTUnWnj9wjcFvhY8i2X2/YxFZGrNqo81TNP8HHYR+YewsyHPAaeBDJAGTgLbgC8AdwIrxgtAXGv/FLv97oeBn2FnPNLAEeAHwAeAy03TDK/nqMgM5A5CwsyEuG/qLazQtnr5g5Cwrtk9ZbygXAYnb+U9Xbr29HtHEHWOFJv2BbUnLnTH8md1YpEYq1KrA7/mHF8dwW3zi5+7xFyZEHdQE0bG4ur260oCo9l6g+wuTj+TPu08rrQWpPy6pduxVJQuImGbbZkQTNM0sYcEfnYK1u7Frjf5+7DXFpkt0p5MSJjbsfwF5JnAm62JGsxOVWF6wLapgAAnnU/zFzs/TedwJ+9d8X4ua93Cz04+4TnHXRweVBNS+DkPuupbbuq4lQ3Nm0qKlgu2tG1l38BeAF4z/3bPjW/5idvVByHRSJR3Lr2Hv9j1aedYfax02vlskIgmnaD9zIg7CGku95aKKBMiItNh1gUhInLuyuaz5Kyc53lY/JkFO6tQ/c2rv0VvWJ23ytWE+O3oeYnjw8cA+OGJ77OuaX1JgXmX61PywJqQfMb+M5ppiRLjzYvfOuZ2ta1tV9OXses9XrPgdZ7X3DUh7oxTWDeySxuWceeit/CdI99mfdMGOmrmhbLu+cY9z8OdCamkJmQsUzUQUkTETUGIiJwz/J/ShzknZKrmbpRkWCZxzYPZQerj3oDIP6wQgq/5xPBx5/HhwUP87OQTJVvNzrjqBQK3Y1kZTzDVEK8ft14mEU1w+8I3Br5Wbs5EmAXkt85/LTd03EwsEgultud85N6OddodhFS5HSseMKxQmRARCdusqgkRkXOb/1P6qZqYbq8dVhBSXXesL++/j99+4cN869A3fesG1IQErO2u98iT57ETpY38ujxBSHAmZCBbDELqYw2VXXwZsTJzJsLuYpWIJmZ1xyZ3A4DTI+HVhAS9X0GIiIRt9v7fW0TOOf4gJKxMSN7KBwQ41a/tLwif6Lrd6S6eOvMkAD/ufAzLspzXAmtCAn4eJ4ZPeJ77gyKw6zIKwUdwJiTLoDsIiVcXhJQfdjf7WulOJfeskDzFbYxNVW7HMhrXMwdv296g1sgiItVQECIi5wz/jXdog/+C6itCmBXiz67AxK75wMB+73quuhX/sEIIrjfpdG3HcltSt5S2ZLvzvCvdhWVZjLha9EYobmPqyfQ4j6st9J6O7VgS3MUKQtiOFY1zLTd6js3mjJOITA39X0VEzhmlNSHhbMcKbHcbQoDjb2kLExuE6A9CCluiLMsqE+B41+7P9jPgK4wv2Nh8Aa2uIXTfOPg1fuu533Q+MY8S8xQb97qDkHiVQUi5TIiKm0OVCJweXxfKZPN25nIxm53nKwIGVIqIVEOF6SJyzijZjjWFmZAwWukGBSHlumNZlsXBwVdpiDfQXjMXCApC+mmvmUvGSmNhlazhD5zKZUHADkLcHZN29+3yvB6PxklE4wzn7ec9mW7ntamrCZmdrXSnStDPsynRGNr6l3AZG5dtwsLi8tatoa0rIgIKQkTkHFKyHSusTEiV26bKGcoGZELKBDe/PPMkXz7wz8QiMX5/wyfoqJnHwcEDnnP6RzMhwwH1IFBaE+KvByloiDWwvGEFL/smi7ul4ilPoNMTZiakzHYsFTeHa1nDCp7tesZzrNqtWG4RIlzVfk1o64mIuGk7loicM6YqEzJU4cyNia9bGtyUm23yXNezAOSsHM91beP48LGS73dgdLBfUOYGSgMcd2csdx3H+uaNRCNR5iTbAteJRWLc2HGzZ86EeztWQ5WZkHLbsVQTEq41qbUlx5ri1Q0qFBGZLsqEiMg5w18TElaL3qBMSLnOW0+e+k++f/wRrp57HbfMu23MdQO3Y5WpCTk2fNR5/OrAAZoDploX6juCaligNHvjnhHy+oV38NTpnzOQHeB1898AwJykt8MRwIfXfpRVqVXEInF+cfrnzvFu93asqaoJURASqsX1S6iN1nqaGISZCRERmUoKQkTknOHfjhVWi95Kt2NZlsX9h/+Nwdwg3zn8Lba0bnVmJmTz2ZLMRWBhesC66Xya0yOnnOcHBw8EzmIoZEJGKs2EuLZjrWhYwQ0dN2FZljO8rzVZ2mZ1ZcMqp2YjXiYTUm1NSNB2rHgkXrabk0xOLBJjZWo1O3pfco5VOyNERGS6aDuWiExKT6abn3T+mFMjJ0NbM6g7lnt2xmQNBba7Lc1Y9Gf7ncDCwsLs2wlAX6aXP9j+O/z+ix/lMIcAu5NVT7q7ZI3AgYLDJ0rqL57r2lZy3oBTExIchLiL3odzw5wc6XSed9TOA/BMD/dnQtY2rvUECO6gwB1QTcWcEGVBpsaaRu+WrGpnhIiITBdlQkRkUv553xfY07+b9s65fGLjp0OZI+DPNIAdiLhrFyaj0kyIe7I4wM7el7m89QoeOHI/fdk+AJ7gR1zD9XzpxS+Qs3IlawRtITvu2opVEJRFcWpCAoImKBam7+nbzZcOfNFpYdwYbwrslOQvBF9ct8TzPKjFK0xNTYiK0qfGal9diLZjicj5QpkQEZmU/QP7ADg1cpLBMrMqJmokcJp39VuyAlv0BqzrbmkLsKt3B5Zlsad/d3EthtjLnsAAxF43XZK9OTZ0rOy1uSdRBxWmu2/oM3l77S/u+7wnYNrSVr596trGdYDd6Whr+9We1+Jlgru6KmtCglr0zknMqWpNCba0fpnnuYr/ReR8oSBERCYsZ2U9gwTDmmweVAtRrtvURAR1sQraNnXGlwnpyfRwbPgo/Zl+z/HTjL0FzT9k8fhw+SDkpo5bnMcD2QEsy+KF7medY+4tVZl8huH8kJOViRLjnuX3cteiu8uu/7Yl72Rr29W8b8UHmVvT4Xmt3FC7aiem+2s/mhPNvGXJ26taU4LFo3G2ttnBZWO8kWUaKigi5wltxxKRCfPPsQgrCAmq0wjKWExU0HasoKGC/kwI2Fuy/N2qosTG/HqZfNpzI14uCGmINbClbSuPHv8uAP25fp7teoadvTuA0exF21U8ePQ7o9ecdupGAJqTzWxtu2rMa1lQt5B7lr8v8LWgAvJkNFl1AXltrJZLWjbzXPc21qTWcu/KX6U50VLVmlLe25e+k4tbLmFJ/bJQpqWLiEwHBSEiMmH+7U2hZUICaiHCyIRUOjHdXxMC8HzXc57nUaIM412vLlZPzso6QVQ6n6GQS8jms3QOdxJkVWqNp5tRV/oM3z78Tef5dXNvZEXDStc1Z5wtWwCpWCpw3UoF1dpU2xmr4AMrf42eTA8tSQUfUy0ZreGClovO9mWIiEyItmOJyIT5g4VyszEmvG6u8kyIZVmBwU/eypccC5q7EfTeoEzIvoFXvOuTZ5jier+++kN8fMMnaIw3OsfcdSwnRzrJY9ePzEm2cumcywA7y/HaBa+nNlrnyawUJpc3J5p546I7SLg+2c5YaWeqOkBDvMogJCDjUe2MkIJIJKIAREREylImREQmzJ9ZCGuyeaWZkEw+w1/v+nM6R05w1+K3cu3c6wHY1buTf97/BRbWLeQ313zEKZAOrAkJLEwvZkLqY/WBHazcaqN1bGq+EMATLKRHC8gfPvYgDx970Dm+oHYhb1n8NuYkW1lWv5xlDcsBaIjXO3UeBZe1XkFdrN4TKPgzIQ1Vt9ItDUKq7YwlIiJSCWVCRGTCSjIh01wTsqP3JQ4NHWQkP8I3Dn6Vp0Ynf//g+CP0Z/vY3WfyUs925/zg7VjeddP5EfpdBd+Xzrl83OtNJYqZCH+wsKff9AQgYNdntCTn8ObFb2Vza3H9oIzGorpFgLd4PJNP+4KQczcTIiIiMhYFISIyYf7J5kH1FZPhH1YI3ixLofXtSd+AxK8e+BL7+l/h0OAh59jRoSPO4+A5Id5r7kp3OY/nJOewsXnTuNfr3oKViHiDhePDxz3ntiRauLr92sB1gjIaC0aDEHegkM5nGMgNjPm+iQgMQpQJERGRaaDtWCIyYf5hemHM8oDS4MZe296O9dDRB/jxiR9y6/zX0DtaN1GQJ893jz7IQK6YJTg2ZA8IzFv5wCGIhUxI53An9x/+N06nTzmvtSZbWdNoECXm1HMEcWciPJkQy7tt6rLWLbxn+fuJRYK7ajX4CswjRJhfu8Be1xPcZOh3F6ZXmQkJ2o5V7bR0ERGRSigTIiIT5t/eFMZ2rLyVD9x6lclnyOQzPHLsIYbzQzx49DucGD5Rct6uvh2e54UgJKjOBIrF9N8//jDbe17wZE5ak23UxepYkRp75kLKnQnxbcfqyxRrPJbVLy8bgNjreIOJuTVznW1Y3uAm7O1YpZ9DNcU1cVtERKaeghARmTD/UMEwWvQGZSvAzoT0ZXs9x/b2vxJ4rtuJkePkrCw96WLWJELEeVy45qdO/6LkvYUBgeubNo75NVLlMiH5jFNfAuMHC/5tVYWtWOCvCcl45oRUW0Telmz3PJ9fu8BTqyIiIjJVtB1LZIZ7uWc7R4eOcHX7taFttfFvxwqjJiSoHsRe25tVADsjUDC3poOTI6VzOHJWjpPDJ3mx53nn2IqGlewb2Ou55taaVk75akxaa9oA2NR8IQ8dfaDsNbszIcmItzuWZ56H67wg/iBloSsIiUaixCIxclYOC4ueTHfF645nXdMG7lx0N13pM1zYcjFrGw2iEX02JSIiU09BiMgMdmrkJJ975TOjN6893L3k7aGsW5IJCaEmJKgeBIIzIQW10Vo2NG3iJyd/FPj60eEjPHvmGef5FW1XuYKQTNmv2zqaCVlSv5R3Lb2HfQN7aU22lXS7cmdC4r6aEHfL3fFqN0oyIbULPc8TkSQ5yy6u73YV0FdbmB6NRLl1/muqWkNERGQy9JGXyAx2ePAQFnZHqSNDh8Y5u3L+rVNTuh0rIBNS0FrTxsrUysDXAF7sfp5DQwcBiEfizqBAsAOFvJX3ZCzAnoi+oLaYibh67nXcs/xeljeUfh13cJH0bMcKLxMC3q1e7ixUtTUhIiIiZ4syISIzmHtIX9AMjsnyF6aHsR1rIjUhBW3JdlY0rCq75tNnnnIeb2jaRH28nngk7nTc6sv2kqc4Yf3auTewtnFd4KTvoGxGKuEuTC/fxaoxMXaw4C9a76jt8DxPBrTSjUVi1ERrxlxXRETkXKUgRGQGc39qHka2osDfcSpjlU41n6h0me1YmXwmcM4HQFtNO63JNhrjTU6g0pxopsfXwhfgkjmbATurkM3Z1+ueDdKenMs7lr677PUFBSHeOSHFQKE/2+cEOolIkuQ4wYJ7+1VNtMaZ9O6s7Qpw3NcTiURKjouIiJwPtB1LZAYbnrJMSPjDCstlQjL5DL1lMyFtRCIRVri2Srm3XBUko0kuaLkI8M7G6EqfcR6P38Gq9PVyc0LcwU1qnCwI2MHUWxa/jXWNG/jNNb9V8nrQUEH/bBEREZHziYIQkRnMvW0qzCBkKlr0luuOlbWy9JerCUnaXawub9sK2C14r2i7ivbkXOecKDHes/z91MXqAO/WJncQMl7xeE20hrgrQxGPxKmN1jrP3YHCmfTpitctuGnerXxo7W+xMlW6vcw9sLCg2qJ0ERGRs0nbsUQdSNCVAAAgAElEQVRmMHdNSBjZioLSFr0TC0K6093s6N3OxuYLaE7Y9RflumONmQmpsedcXNJyKb+7/g+oidYwr3Y+b1h0B98+9E2WN6zk7iVvp72mGJS4tzZ5gpBxMhaRSISGeMppkdvg2w5Vdt0q2+gC1MSCghBlQkRE5PylIERkBpuq7Vj+gME9t2M8PZke/mznp+jP9rGyYRX/fd3vAXha2tbH6hnMDQKjhellMiGFYXuRSISl9cuc45e3XsHlrVcEvqfstqkKgoWUKwjxZzjcNSHe9rzVByGL65ays9c7EV5BiIiInM+0HUvkHJHNZ8lb+fFPnAD3dqysFd76/sL09BiZEPf3ZVkW33j1q8408X0De52swc7el533LGtY4TzO5NOe6eMFdbE66uP1E75299amMxPYjgXeG/9GX3CRDCger3Td8axrWh9wLdqOJSIi5y9lQkTOAfv79/K5Vz5DQzzFb6/7/UndXAcZznu7SmXyGWpi1bV1zeazTuen4rrBmRCzdyef3/sPAKxtXEuUmGeCOcDe/j2saVzHgYH9gF3XcVHLJU5Q0p3pdmadgB1EZKw0y12BykS4a0K6M+5MSCVBSIPrsff8eEDxeKXrjmdlarWntXDQ1xcRETmfKAgROQf8wyt/x1BuiIHcAI8ce4i3LHlbKOsO+VrbpvPpqoMQfz0I2IFJkB93/tApOH+pZ3vgOa/0v0I6n3YCjVWpNbQkinM63PUV82rmc/eSt7OrbyfXzb1hUtfvrt0obK0CaKhoO1Zj4GMInuUBpRmTyUhGk6xKrcHs2+n6+gpCRETk/KUgROQc4A4WDoc42dw/VDCMuhB/ZyyAdEBNiGVZ7B/YO+56e/v20O2qzbiw5SJPVsEdhDQmmtjQvIkNzZsmetmOoHa3UNlNfdtoNy6Atpo2z2vumhC3sDIW65rWe4KQhpi2Y4mIyPlLQYjIOSYeCe8/S/+Qv4kUkJddMyATEtQd6+TISc/U8HuW30uECKl4isX1S/mDF3+XPDmODh/h6PAR57wLmi+i1zVs0L0VK4ysQjVByJXtV/P0kV8SIcLWtqt965apCUlUf80ARqO3LkTbsURE5HymIETkHJOIhvOfpWVZU5QJKW2lmw0IQvYPvOI83th0AVvbrvK8vqR+Ka8O7vccm1+7gI7aeU5nLL/GRNNkLtmjXMaisu5YjbyG1wOlheHVBDeVWFK/dErWFRERORvUHUvkHBMvc5M8Uel8mjz5kmPV8nfGstctBiGnRk7y6sAB9vUXt2IFDeBb3bim5NiVo9mFcoFCOJmQ0oxFhIgzzHDy604+uKlENBLlbUveRTwS59I5l9FROy+UdUVERM4GZUJEzjLLsjzPw9qO5e+MBeFMNvdnV8De5mVZFqfSJ/n0y58o6Z4VFISsSq3hhye+7zy/bM4Wbpp3KwDxMtmgUDIhAcFCQzxFNFLdZzJBU80jRKiPhdPpDOD6jhu5Zu61xELcsiciInI2KBMiMkE9me6SwKEa/uxEnnDWDgoWwsiEBK0L9hySPX1mSQASIcLS+uUl5xuN62gdLfTe1Hwh9yy/1wkEymWDwuo05RfG1qZUPMX82gWeYxZW1cGNnwIQERGZCfS3mcgEfPPg13ni5I+5dM5lfGDlr4Wy5mBuwPO80NK2Wv72vACZENYO2o5lr53xFKIXdNTOozZWW3K8NlbL763/QzqHj7O8YSWRSMR5rdzWpqmqCQljy1QkEuG9Kz7AX+z8tHPMH5SIiIiITZkQkQpZlsUTJ38MwLNdzzBUpnh6ogay/iCk+mwFlHbGsteemu1YYG/J8n8vYGc8ymmIN7AitcoTgED5LWmN8TC2Y01NJgRgaf0y/pvxuySj9iyWjc0XhLKuiIjITKNMiEiF/MFBJp+lLlb9uv5OUGFlQqZqO9ZImesLyoTURmu5vuOmCX+NoExIlBitydYJr+W3IrWy5FiY7W5XpVbzqU1/xpGhw6wJKL4XERERBSEiFZuqbVOD/kxILpxMSPB2rKkZVmivnWHAFYS8Z/n7uaz18knVMMQipdHd/Nr5ZQvWJ2Jp/TJunfdafnDiUeeYv91utRoTjaxLrB//RBERkVlK27FkRuoc7uTlnpfIW/nxT67QYNafsQgnWBgoCW6mcjtW+bXT+RFyVm78dceoCXEHIa3JtkkXUUcj0ZItWQvrFk1qrSBvWnQXm5ovdJ4vqF0Y2toiIiIyPmVCZMbpyfTwpzvsNrF3LHozt81/XSjr+oOFMLIKUBrclNvuNFHBk82Dr/k7h7/NYye+R4QI7TXtvGHhnWxuvdy7Xm6Yn578CU+dfjJwjbSvJqTa7EI8Evd02lpQF16gEI1Eef+K/8JjJ75HIppgc+tloa0tIiIi41MmRGac7x172Ll5feDI/aGtW7JtKqwgZIqCm6DtWEHXPJgd4LET38PCIk+ezpFOvnzgPg4NHvSc9+ixh/jOkW+V/XpZX01ItR2n4r66kEV1i6taz68mVsPrF76J2+a/Tm1vRUREppmCEJlx/IXe4a07NUGIv6NUaJmQwJqQ0u5Yu/tMLN9skqyV5b59/+gpbt/R+3LJe+tcg/hG8mnPz6ghXt2QvqncjiUiIiJnl4IQmXGyVvVtaIMMTFFNiD+4yVrZUGpZKu2OZfbtch5f1HKJ0162c6STT738h/zwxPcZzg1zbOio531RYp7AoDfT7QQzdbG6qrML/mByTgidsUREROTcoD0IMuPk8uMXV0/GVGVC/DUhhbWDBvxNRKXbscy+nc7j6+bewEUtl/DlA/cB9nT4+w//O7v7TPIUA6Nb572GVak1bOt6hr3sAaA73eW83hCrvuWtu/tYMloT+uRxEREROXsUhMiM4y5mDtOU1YQEDPgbyY9UHYQEZUIylveau9NdnBg+Dtjbn1amVpOMJsnkM3z36H/Qm+0B4KWeF533XNa6hTsX3w3A9p4XnONdGVcQEnLL246ajlDXExERkbNLHy3KjDNlQYhve1Bo3bFypUFIGDNIKmnR686CFAIQgGvmXscnNn06cHL5svrlzmP39PHudLfzOMzhfwBzaxWEiIiIzCSzMhNiGMYG4DbgGmATsBCoBXqAfcATwH2mae4su0jwuvOAe4G7gOVAM3AMeB74GnC/aZrhDa6QQNn8+ZUJ8deajLX2kcHDnEqfZFPzhc5Av87hTr577AGW1C/jlnm3OecO5SsIQnqL9SBG4zrPa7WxWtY1reelnu2e40vqlzmPE64gpSt9xnkcRhCyrnE9u0aDpK1tV1e9noiIiJw7ZlUQYhjGrcDfAhvKnNI++mcL8FHDMP4P8BHTNEs/qi5d+07gi4C/enb56J87gZ8YhvFu0zSPTOobkIrkzqNMSM7KMRwULORKMyGnRk7y5zs/TZ4cr51/O29cdBeD2QE+u+d/cTp9mmfO/JKVDStZVLeEg4OvegYHFq+5WLTfk+nmue5tznOjqXTC9wXNF3uCkAgRltQvdZ57MiGu7VipEIKQtyx5O985/G2WNSxnU/MFVa8nIiIi545ZFYQAm/EGIHngBWAP0A3MB64DWkZf/yCwzjCM20zTLL1THGUYxu3At4DY6KF+4IfAaWA9cOXo8euB7xmGcZVpmr2hfEdSwt/iNmflnKxBNfytdMPIhAyVaScctPa2M0+Txy66f/T4w7xmwev5yoF/4XT6tHOO2beLB49+h9195rjrfvfog87zRXWLPdusCtxTxQHm1c731KokXLM83GuHUROysG4Rv7Hm/696HRERETn3zLYgpOA54J+Ab5qm2eV+wTCMBuATwG+PHroG+BPgo0ELGYbRDvwrxQDkUeBd7nUNw7gWuB87y7IR+Azw3rC+GfHy10Jk8hliseqDEH8mJIwgxB/YjLX2yZFOz/PPv/JZT00HwNNnnnIKzcda9/jQMX5+6qfO8bsW3x3Yfaol2cLS+mUcHHwV8G7FAkhEEiXvgXAyISIiIjJzzbbC9N3AnaZpXmqa5uf9AQiAaZoDpmn+Dva2rYIPGYbRXGbNjwFNrvXv8q9rmuZPgXe6Dv2KYRgbJ/1dzCBDuaHQt08NlgQh1QcLeStfkrUIIwgJ6owFwQMLDw8d8jz3ByDAmAEIFH8WP+r8gTPTY13jetY3lf913Dzncuexv27EvR3LLezCdBEREZlZZlUQYprm/aZpPlDh6Z8ECneZSeBm/wmGYSSAD7gO/ZFpmqV9Ue2v/RjwvdGnUeC/VngdM9au3h187IX/zh9t/1jZjMBE5axsSWepcLZNVTZzA6Bz+AR/t/uv+cqBfyFnjT2zpNx0d//aOStbMiywYFHdYmpGBwwGWdmwyrOuZVkcHz7mHLvJVcwe5MZ5N/PGhXdyx6I3s6Vtq+c193YsN2VCREREZCyzKgiZCNM0e4CXXYeWB5x2I3YHLIBe4P+Os+yXXI/vnPTFzRCf2fM3ZKwM3ZluHjn2YChrBgUL7mLsyQoKksoFId87/jC7+3bxi9P/ybYzTzvH9/Tt5t8O/itHBg+Pua69th1I5a08x4eOcWToSGDr4XgkzvtWfJDVqbUlr712/u1c3rqVty19l1MTY2GRtbL0ZHqc89pq2gKvoSAWifPaBa/ntvmvK6mtKReEhDGsUERERGau2VoTUinL9TioqOBG1+MnTdMc7yP3x12PFxuGscY0zT2TvbiZpHO4c/yTKjCYrTxjMRFDAbM8ym3zKtRPgJ3t2dK2laHcIJ9/5bMM54cw+3byhxs/hWVZHBo8GLhG4Zo/98pn2NH7kue1ulg9I7lh8uS5a/FbWVi3iBWplbzcW+xi1ZJo4Q0L7yQSiQCQjCadAC2TT9PrCkKa4uV2Go6vXE2ItmOJiIjIWBSElGEYRg2wxnXoUMBp7p6mz463pmmaxwzDOI7dhavwfgUhQDwazq9iULeptDU1BeRBwU3OytE5fMJ5bvbtxLIsXux+3mnFe3z4GIPZQb516Bs8deZJ59woUfLknbUHs4MlAQjAjR03c3HLpWStDMsaVgCwwrXlCux2u4UABLxBSF+2z7n2RCRBXayush9CgPI1IeFOTBcREZGZRduxynsL0Dj62AJ+FHCOew9M8EfapdzBjDGJ65qRgiZzT0ZQEBJGYXpQ7Ua5DlbubVPdmW46R07wfNdznvN+1PkDTwAyv3YBN3QUy45GciP0ZnsIsrh+KYvqFzsBCMDyhhVEKAYd6xq9o3ASkWKwcGrkpPO4KdHsCVYmKhmwHasmWlN2m5aIiIgIKBMSyDCMOuB/uA59yzTNoP1C7s30JwJeD+JuX+QfbBia/v5+tm3bNv6JU2AyX7enqyeU693P3pJjO/fsZJDAfgEV20VpJ6q+4b6Saz7AvpLzHn75IV7GO3X8iWOPO4+XsIwbh29hz3BxtsfRziM80/k0Qbr3drON0p9VB/M5wTHixEkfyLLtQPGcHMUC+WdfKSbtYulYVT/3U5wsOZbIJ87a7950mg3fo4h+z2U20O/52aEgJNjfAYWBCEPAH5Q5z73xvewwQx/3ebN247zlKbeBaGDJzcSlCdgixdgdqioxQmnL3BylheJdnCk59gxPlRwboDjNfCWriZMg7vrPMUOWoTK/Ug1lfm2u5UZ2s5NFLKYO7xarmGvtPopzMuuoD1yrUvGA/4XUMvntXSIiIjI7KAjxMQzjg8Cvug591DTN3WVOr3U9rnTPj/tudsru1lKpFIYxvbu9Cp8kbN68edxz0/m0PTJyVHt7O5uXjf++8XQdPwVHvMcWLVvE5vbq1j5waC/4cmFW1GLzJd51n9/3DJRMnxnbtRuuY0HdQiJdFj/d9zgAjS0p2lNtcNh77lVt13DZ8svKrnWjp1dC0U/NH3Gq3/4GYs1RGN3ptWzucjYvnfzP5vTIKe5/6ZueY9csuo7N86v/d3mumsjvucj5Sr/nMhvo97x6pmnS398//okBFIS4GIZxG/C/XYe+Zprm/y53PjAMzkfJwRW6pdwDHSrNnsw4g1n/4L/STMNkBNeEVN+it1xNiGVZnpoK9yyPCJGSjI9fIpKgo3YeYBePu9fuzRYzFte0X8eFLZewrmldyRqVcBeQu2tCmhOT74zlXxfg4pZLuXmcuSMiIiIiKkwfZRjGFcD9QKGi9vvA+8d5mzv0qzSr4T5vcqHjDDDoa3kbRqBgrzs1LXqDJpvnyXuGEeasLJ0jxdKgW1w34xEiXNF2ZckaC+sWObM3kq6BgyP5EXozxSBkSf0yNjZvIjbJAn53gHMq7S5Mb5rUegWN8UaW1ts7F9ekDN634oNEI/rfioiIiIxNmRDAMIxNwMNAoa/oL4A3VzD34wzQMfp4XoVfzn1eaQHBLOEfKhhGoGCvW113rOHcMJ9/5bP0Z/v41VW/zrxau5vyWJPNC+2FO4c7naBkTqKVOxa9hSvbryFn5WhNtlEbq2V3n0lXuvivfXH9EudxjStQKJnlUWWwkPBlWYrrtlS1biQS4cNrP8qx4aMsq1+uAEREREQqMuuDEMMwVgM/oNipajtwu2mawaOsvUygsD9maYVf0n2eWfasGc6fWZjSIMQqn2WxLItdfTs4MLCfrW1X8Xz3c+zpt/+1fP/4I9yz/F4AzqRPB74/nU9Ta9Xy3aMP8Ojxh53jC+oWEolEnCCmYH7tfG8QUlcMQtyBwkh+BCtb3MrVVOW2qWSZeR7VbscCqI3VsqJhZdXriIiIyOwxq4MQwzCWAI9RHB64B7jVNM1KS4t3AneMPr6kgq833/W1Cu+flfyZkDBmecDEJqZ3DnfyzYNfY1ffDgB29u7w3JTv77fb7Z4eOeUEIcloDfWxerozXc7aL/ds9wQgYAchQebXLmBn7w7n+SJPJqS4HSudSzOcK7YVboxXlwkpF4RUG9yIiIiITMas3TthGEYHdgBSaMV7CLjFNM1K530A/Nj1+ErDMMab0Ha96/Fh0zRn7bR0f03IdG/HsiyLz73yd04AArC3fw9mbzEuPDFynMHsALv7igmrVanV1MeKbW0z+TSHh9zzJ+1i86D6D6AkM7KobrHzOOnJhAzT56oJqX47VumvZoQIjfHGgLNFREREptasDEIMw5iDvQWrMPH8BHYAUunU84LHwRm60AzcOc7573U9fmCCX2tGmbJMSIWTzXuzPXSOlM6fHPAFR68OHsDs2+U8NxrX+eorRjg1csp5fkXrlfyPC//KE1y4zatd4Dxur5lLbazY5dldmD6YGyRPHoC6WF3VE8iDMiGN8SbVcIiIiMhZMevuQAzDaMAuQr9w9FAXcNsYs0DKGi1c/z+uQ58yDKMm6FzDMG4EXjv6NA98fqJfbyaZ1pqQgM5bR12tdMdyYGA/e1xByNpGo6SV7mlXy9vNrVuoj5cfALgqtZol9UuJEOHGjps9r5ULNJri1W+ZCgpCwqgHEREREZmMWVUTMhogPABsHT3UD7zONM0Xq1j2z4APAk3YRer3G4bxbtM0u11f92rgm0BhoMRXTdN8qYqved7zt9INIwjJ5rOB6wQdOzZUnGi4sHYRR4ePlJwD8MvTv6A7Y/+rrIvVsbh+aUkQcipdzIS017SPeY2xSIzfWfdxBnMDpHxboaKRKIlIkozlvd5qt2IBNAd0wWoMYV0RERGRyZhVQQjwJ4D74+cdwD2GYdxTwXt/YZrmV/0HTdM8ZRjGu4D/wM4s3Q4cMgzjMewWvOuAq3xf80OTvP4ZYyigJsQ/+G/iawbPfgza6uUeKri1/SoeOfZQ4Pvdcz9Wp9YSi8Q8QchQbojutF2kHiFCa7Jt3OuMRqIlAUhBTSxJJuu93jCChQtbLqI9OdczI0SZEBERETlbZlsQ0uF7vmX0TyVSQEkQAmCa5ncNw3gr8AXsVr8pgutDngDeZZpmb8Brs4q/i5WFRdbKkohUVvtgWRbHho8yJzmHutFC8fJtdAO2Y7kyH4vqlrC20eCF7uedY8losiSDsrbR7sbs3jZ1Yvi4MxW9OdEcQu1GDf4ZlmF0sEpGa3jHsl/hs3v+xjlWF6t0vqaIiIhIuGZdTchUMU3zfmAD8PvA08BJYAR4FTtL8jbgRtM0g/f9zDL+7lgwsanpj534Pn+645P88Ut/SH+2D4AnT/+n8/p8VwF4IRNydOgIn9n9v/jWoW9y1L0dq24RRuMG53lHTQerUqs9X68uVs/m1ssAb33FseFiRqWtZm7F119OcAF5OB2s1jdt4IrWYteuZQ0rQllXREREZKJmVSbENM33Ae+bwvVPYNeI/NlUfY2ZImjrUzqfpp7yRd1uPzv1EwD6sr1sO/MMW9q28svTTzqv3zTvVr7+6pftdUdrLB499l129e1kV1+xDW8qnqIx3sjFcy7hwaP/l6HcEJe1XkFNtMaZ5zGvdj4fWPlrTl2FJwhxbetqT45dD1KJoCAkzFke71j2bpqTLSSjSS5uuTS0dUVEREQmYlYFIXLuGMxW1krX/3osEqU308spV0eq57ufJYI9ZRzsLMiGpo3O64VMyLaup0vWXFC7iEgkQnOihY9v+CQnR06yOrWGnJVjOD9MPBLnho6bfa10i4GCu2akbZyi9ErMSc7h4OCrnmNhBiHJaA13LHpzaOuJiIiITIaCEJl2eSvPcL40EzLWrJD9A/v4x1f+gXR+hCvbr/G8trtvl2d71XVzb/ROHx/d5hVU57HQNdl8TrKVOclWwC4ef8PCOwKvJVFm+vh4nbEqccu81/Jyz0tkraxzrKnKaekiIiIi5xrVhMi0K9fFKj2ayfDrzfTyhb2foy/by0h+hMc7f1hyTqEupCZaw5a2rZ5AITPaecuyrJL3LahbNOHrdw8VdAujJmRlahX3LL/XeR4h4gRGIiIiIjOFMiEy7YKK0iF4O1bOynHfvn+kJ9Md8I5Sr1vwBupidXa7XyJO162h3BAZq7TwfYErE1KpoLoNCKcmBOCy1i3krCyPHf8+l7VuoTERTmG6iIiIyLlCQYhMu6FsmXkeo0GCZVn0ZntoTrTw1Okn2dMfPMw+FomRs3LO82X1K7h53m0ARCIREtGEE9gEte9tScxhWf3yCV9/UBASj8RDrd24ou0qrmi7avwTRURERM5DCkJk2o2XCfni/n/kua5tXDJnM4cGDjqvb2q+kJd6isPt1zYaZPJZXunfTSKS4D3L7yUaKe4wTESSpLHXPO0KQhbVLeauxW9lSf2SSc31CApC2pLtnq8tIiIiIuUpCJFpN1i2JiTNQHaA57q2ATj/BKiP1fO+FR/kM7v/hlcH9wP28MAr2q7k6TO/ZH3TBubXLfCsl4wmGRhNlJwZOeUcb022sb5pA5MVVJi+sH7itSUiIiIis5U+upVpN5QtnwnpywQPk79p3q3Uxep485K3UhutpS3ZxpVtV9OcaOGWebexqG5xyXvcWQ53JqTa4X9BmZDNcy6vak0RERGR2USZEJl2A2Nsx+rLlgYhdbF6bui4CYDVqTX8z4v/lgiRcbc/uYMFd01IqspCb38QUhutY1PzhVWtKSIiIjKbKBMi025n78vOY3e720w+Td9oq92CVDzFO5f+CnWx4iT1WCRWUf2Fe9vUmZFiEFLt3A1/EHLxnEsnVVsiIiIiMlspEyLTqnO4k919JmDPwNjadiVPnHwcsDMh/ZliEHJN+3W8c9k9k/5a3u1YxZqQalve+gOOy1uvqGo9ERERkdlGmRCZMnkrX3LsydM/cx5vbL6AuTUdznN/JqQxEV7GYjA3WFy3ykxIY7zJWTsWibG20ahqPREREZHZRpkQCU2GNDlyDGYH+PdD3+DpM0/RVtOO0bieW+bdRltNO7849XPn/Kvar6U30+M8T+fTZF1zP1JVFpCX2yJVbXBTE6vhvcs/wHPd27ix4xa15hURERGZIAUhEort3S/wFe4D4Fvb/9WZ+XFq5CSnRk6yu28Xb1j4JnqzdtDRGG9iU/MmnjnztLNGOp8mky9ONZ+KLlZhrAt2HcjFcy6teh0RERGR2UhBiITilf49zuNCAOJ2cqSTr736Zef5le1XE4vEPYFCOp+mP9vvPK82Y5GIlAYhESI0xFNVrSsiIiIi1dE+EgnFDR03sZilREd/pRpiKT6w8r9y2/zXOecUgpNkNMlNHbc6jwsy+YynMD1VZbAQlAlJxVPaPiUiIiJylikTIqGYk2zlNm4nS5YF6+czt6aDulgd65s28J8nn/DMBrl+7o1Ohyp/JsQ9J6TaAvJkQE1ItWuKiIiISPX0kbCEKk6cpfXLqIvVAVAXq+Omebc6ryejSW6e9xrnuXuWx1Bu0OliZW+baqjqWhIBmZBqt3iJiIiISPUUhMiUu6HjZpbWLwPgrkV3e+Z0eKean3Eeh7FtqsY1CLEgjKJ0EREREamOtmPJlKuN1fLRdb9HOp9xMiQFSV8mpCCMbVMbmjcRPfwt8hTb/lY7qFBEREREqqdMiEyLWCReEoBA8JYpgFQIwcK82vm8Y+m7vOsqEyIiIiJy1ikIkbNqKmd5AFw99zqun3uT83xd04ZQ1hURERGRydN2LDmrgjpYQbgF5G9d8g42Nm+iNlbH8oYVoa0rIiIiIpOjIETOqlgkTpSYp24Dwi0gj0QibGy+ILT1RERERKQ62o4lZ13Qlix1sRIRERGZuRSEyFkXONlc8zxEREREZiwFIXLWKRMiIiIiMrsoCJGzLjAI0TwPERERkRlLQYicdf5ZIal4ijmJ1rN0NSIiIiIy1dQdS8669U0beHVwPxEiXNRyCW9YeAfxqH41RURERGYq3enJWfeGhXewoXkTLYk5tNW0ne3LEREREZEppiBEzrpIJMKq1OqzfRkiIiIiMk1UEyIiIiIiItNKQYiIiIiIiEwrBSEiIiIiIjKtFISIiIiIiMi0UhAiIiIiIiLTSkGIiIiIiIhMKwUhIiIiIiIyrRSEiIiIiIjItFIQIiIiIiIi00pBiIiIiIiITCsFISIiIiIiMq0UhIiIiIiIyLRSECIiIiIiItNKQZN+BZsAABMeSURBVIiIiIiIiEwrBSEiIiIiIjKtFISIiIiIiMi0ip/tC5DQrQYYHBzENM2zcgFn6+uKTCf9nstsoN9zmQ30ez55g4ODhYerJ/peBSEzTwogn8/T399/Vi7gbH1dkemk33OZDfR7LrOBfs9DkZroGxSEzDz7gRVAP/DKWb4WEREREZm5VmMHIPsn+saIZVnhX46IiIiIiEgZKkwXEREREZFppSBERERERESmlYIQERERERGZVgpCRERERERkWikIERERERGRaaUgREREREREppWCEBERERERmVYKQkREREREZFopCBERERERkWmlIERERERERKaVghAREREREZlWCkJERERERGRaKQgREREREZFppSBERERERESmlYIQERERERGZVgpCRERERERkWikIERERERGRaRU/2xcg5y/DMJqA9wJ3A2uAVqAT2Al8A/i6aZojZ+8KZTYxDCMJXARsAa4Y/edaIDJ6yh+bpvnJSaw7D7gXuAtYDjQDx4Dnga8B95ummT/ba8rsYBjGBuA24BpgE7AQqAV6gH3AE8B9pmnunOC6+j2Xc4JhGO3Yv99bgAuAVdi/5ylgBDgDvAg8BnzFNM1TE1g79PsW3QtNXsSyrLN9DXIeMgzjGuy/RJaOcdqLwDsm+pehyEQZhvEe4J+AmjFOm3AQYhjGncAXsf9SKecnwLtN0zxyttaUmc8wjFuBvwU2VPiW/wN8xDTNgQrW1u+5nDMMw3gIeH2Fp/cDHzNN87MVrBv6fYvuhaqj7VgyYYZhXAo8SvE/ujTwCPZfOD8GCp9sXQj80DCMxdN+kTLbtDJ2ADJhhmHcDnyL4k1UP/AAcB/wpOvU64HvjX4aNu1ryqyxGW8AkgeeA/4NOwD/D6Db9foHgUcNw6gba1H9nss5rhM7WP0K8PnRfz5N8T4jBXzGMIw/GWuRqbhv0b1Q9bQdSyZkdMvLt4GG0UPbgDtN0zzsOmcD8CCwEliA/SnB9dN8qTI7HQKecv35C+DKiS4yuh3gX4HY6KFHgXeZptnlOuda4H6gHdgIfAY7JT9ta8qs9Bx20PFN9+8OgGEYDcAngN8ePXQN8CfAR4MW0u+5nKMexw6qHzNNc1/QCYZhLMcOSl4zeujjhmE8ZJrmUwHnhn7fonuhcCgTIhP1a9h7ewFOA7e7/6MDME1zB/BG7L2bANcZhvG6abtCmY2+BSwwTXOpaZpvNU3zr0zT/Cn2J1OT8TGg8OnsbuAu/w3f6PrvdB36FcMwNk7zmjJ77Ma+ybnUNM3P+393AEzTHDBN83ewt20VfMgwjOYya+r3XM45o////qdyAcjoOQeAN2HXXYBd+/fBMqdPxX2L7oVCoCBEJuo3XI//0jTNzqCTRv/j+5cy7xMJlWmah03TPB7GWoZhJIAPuA79v/buPUiyqj7g+Hd5C7KAyBsEA+HHJjwEwruIEh4BNAEhhFeJQDAFISakIkYriJqq4DOhUgqRoClAhFggD3kYIihCFCIRBAT5qUFgCbASEBDkte7kj3uHOdN2z9xeum/3sN9P1VSde/rcX5/ZOrN9fn3vOff0zHyhx/teD1xXHy4HnNhWTC1bMvOyzLyyYfOPMJWArwTs3dnAca65LjNforq6MGn7Hk2HMW9xLjQAJiFqLCK2BLYqqs6f5ZTy9X0i4vWD75U0cHtR7eID8Axw+Szty3F+cIsxpa4y82ngnqJqsy7NHOd6LSh3xlq988VhzFucCw2OSYj6sVdRzgbfPN8GTO7MsgpLcW++NALlOL+l/rZtJjcW5Y0j4jdbiinNpNz6cvkurzvO9VqwoCg/0OX1YcxbnAsNiEmI+lH+sd8+W+PMXAzc3eN8aVz1O84fBcoPoW7jfBgxpa4iYmWq5xVMWtilmeNcc1q98Lu8/e+yLs2GMW9xLjQgJiHqx5ZF+aGG55QffjHAvkjDMoxx7t+O2nQoU7emTADf6NLGca45JyJWi4jtIuJ0qu2eJ29t+jbV1ridHOdjzC161Y+1i/KihueU33LN9NAqaVwMY5z7t6NW1M8FOaOourTHolnHucZeROwDfH2GJhNUW0KfUF9x6OQ4H2MmIepHuZjq+YbnlO1cjKW5YBjj3L8dteWfgE3r8vPAaT3aOc411y0Cjs/Ma2do4zgfYyYh6scqRbnp8xdeLMozPrlXGhPDGOf+7WjoIuIE4D1F1fsy80c9mjvONRcsBM6qy8sBawJbA9sA6wFXR8QVwEmZ2e2qhON8jJmEqB/lfu8rNTxn5aLc9BsDaZReAFaty4Ma58OIKb0iIvYDzi6qvpSZZ/dqj+Ncc0BmJvDnnfX1wy3/GdgTeCewbUTs3uXWw2HMW5wLDYgL09WPZ4ty00y+bPdsz1bS+BjGOPdvR0MTEbtQ7Qy0Yl31H8Dxs5zmONeclZn3APsC36mrNgfO7NLUcT7GTELUjyeL8noNzynbPdmzlTQ+hjHO/dvRUETE1sC1wGp11a3AIQ2e0eE415yWmS8CpxZVh0fEGzuaOc7HmEmI+pFF+U0NzynbZc9W0vgYxjj3b0cDFxFbUO0cNLnbzt3AgZn5XO+zXuE412vBLcAv6/LywE4drzvOx5hJiPrxw6K8/WyNI2IFqsVj3c6XxlW/43x9YP0e5w8zppZhEbEJcD1T4+THwL6Z+fOGIRznmvMycwJ4uqhaq6PJMOYtzoUGxCRE/fhmUY6ImO0y5I5M3SLwAtU3FtK4K8f5bhGxYs+WlbcW5Ycz88ctxdQyKiLWpUpAJrfiXQjs02N3oF4c55rz6jFWPrej81anYcxbnAsNiEmIGqu3eryvPpwHHDPLKe8uyjdkpouxNBfcCDxTl9cADp6lfTnOr2wxppZBEbEW1S1Yk09tXkSVgDR9cvOkG3Gca+7bh+k7VN1XvjiMeYtzocExCVG/yi0fT+2yCAyAiNgKOK6oOqtbO2nc1At6P19U/V1ErNytbUTsBexfHy4BPtdWTC17ImI1qkXo29ZVPwf2m+FZID05zjWOIqLx08QjYg3gU0XVHZn5QJemw5i3OBcaAJMQ9esc4IG6vA5wbURsVDaIiAXA1Uw90OemzPxaaz2UXr2PMfWN7lbAZRGxZtkgIvYAvkz1TRjAhZn5g5ZjahlRT+avBHatq54FDsjMu15FWMe5xs0xEXFrRBxVJ92/JiLmRcQ+VNvz/nbx0gd6xBzGvMW50ADMm5iYGHUfNMdExA7AzUw9lOpF4AbgUWAzYC+mEtzHgJ0zc2HL3dQyJiK+36V6C6buxV1ENR5Lj2TmgT3ivR34KlNj+Vmq+/CfpJpc7V40vxfYLTOfYQbDiKllQ0R8kunbkX4XuK3h6bdm5oU94jrONTYi4hSmnvfxMnAP8CPgKardr9ajWmOxQcepH8jMT8wQd+DzFudCr55JiJZKROwJXMjM29PdDRyRmfe20ystyyJiaf4zezAzN5sh5iHAuUxtgdrNTcBRmfm/Td5wGDH12hcR5zH93vJ+nJ+Zx84Q23GusRARJzH9VqfZPAy8NzOvaBB74PMW50KvjkmIllpEzAeOBQ6j+sb5DcDjVN9s/RvwpfphQtLQDSMJqeOuR/X06XdSfbs1n+pbrTupPny+kplL+uzrwGPqtW2YSUgd33GusRARQfU09F2B36Ka4M8HJrfjfQi4HbgGuCYzF/cRe+DzFudCS88kRJIkSVKrXJguSZIkqVUmIZIkSZJaZRIiSZIkqVUmIZIkSZJaZRIiSZIkqVUmIZIkSZJaZRIiSZIkqVUmIZIkSZJaZRIiSZIkqVUmIZIkSZJaZRIiSZIkqVUmIZIkSZJaZRIiSZIkqVUmIZIkSZJaZRIiSZIkqVUrjLoDkiSNo4jYDPhpUfXmzHxgNL1pZi72WdKyySREkrRUIuIjwIdnaDIB/AJ4ArgLuBn4Ymb+bPi9kySNM2/HkiQNyzxgPvBm4CDg08DCiPhoRPglmCQtw/wQkCQNynUdx8sBbwC2Blau61YCTgcWRMThmTnRYv8kSWPCJESSNBCZuX+3+oh4HXAicAawSl19GPAt4Kx2ete/ei3FvFH3Q5Jei7wdS5I0VJn5fGaeCRxMtU5k0mkR4eeQJC2D/M9fktSKzLwOuKqoWh/YcUTdkSSNkLdjSZLadBXwh8XxdsBtnY3qW7iOBA4AtgfWoVpP8jhwN3ANcF5m/nKmN+u1ZW1ErAUcDRwCbA6sR7Vu5bjMPG+mc2f7BSNiW+AoYB9gE2At4ClgIXADcHFm3jFbnI6Y86huYTua6t9jXeBJ4H7gUuCCzHyyn5iSNEomIZKkNi3sOH5jZ4OIOBr4JLBhl/M3qX8OBP42Iv40M6/ppwMRsS9wPrBBP+c1iLsa8FngGH79ToN16p8dgPdFxIXAn2Xmsw3ibgRcDOzZ8dIG9c8ewPsj4gjgoVf1S0hSS7wdS5LUphU7jl8qDyLiDOBCpicgjwLfoVrI/mBRvyFwZUQc08f77wpczVQC8hPgRuB24Lk+4kwTEWsC3wCOZfpn6w/r+PcwtR5mHvAu4Jv1FZmZ4q5Xxy0TkF8Bd9Zxf1LXbQD8O7DN0v4OktQmr4RIktq0oON40WQhIk4EPli89lXgw5n5/fKEiNiB6orDbsDywDkRcUdm3t3g/f+F6rauK4BTM3NyEj95JWP1Pn6X0ueAnTv6fkpmvnI7V0S8CfhH4NC66nfq/hw2Q9zPA1sWx+cBH8zMx4q42wDnUP17fGEp+y9JrfJKiCSpFfW6hqM6qm+pX9sUOLOo//vMPKgzAQHIzNuBt1FdCYBq29+PN+zG6lQT+UPKBKSO+1w5uW8qIvYDDi+qLgIOLhOQOv5DVAnHBUX1H0XE23vEPRB4R1H1mcw8rrOPdfK1N9W/5Tr99l+SRsEkRJLUlo8CbymOv5eZ99flv2LqGSI3Z+ZpMwXKzJeA44HFddUBEbF5gz48Drx3wA9J/MuivAg4qVf8uv5k4JGi+pQecU8uyg8C7+/Vgcx8nun/HpI01kxCJElDERHzImLtiNg/Iq4BPlS8vAT4m7rdclSLuSd9ukn8+krDzfXhPKqrAbO5qMli8KYiYlVgv6Lq3Mx8ZqZz6vc/p6j6vYiY3xF3NeD3O+K+MEvc+4CvN+q4JI2Ya0IkSQMREU2vLiwB/iIzb6iPt6Haxhaqxds3dD2ruzuBvery5BqLmdzUR+wmdmL6Z+lVvRp2uJLqyhBUXwjuwvQEYkeq9S6TvtYw7jVU2xpL0lgzCZEktem7wF9n5n8WddsW5cXAVyKiabwtinKT9RD3z96kL+X7TwB3NTzvXqrfdfJzeAumJyHlYvQJqt21mmiyOF+SRs4kRJI0KNd1HC8BfkH1UL27gJsys9tkeu2ivCLTb0PqxxoN2sx4q9RSKLfYfXa2W6YmZebLEfE0U79751a9nXFfbNifJxq2k6SRMgmRJA1EZu6/lKeuNqAuNFnnuGRA7zVp5aL8Us9W3ZWJxSodr620lHGbJiuSNFImIZKkUXuqKD+TmU2uaIyLsu/9PmOkXIz+VMdr5RWbfuLOn72JJI2eu2NJkkatfO7F/Ih43ch60r+fFeWVImLjJidFxLrA63vEgeIhjnXcDWnmNxq2k6SRMgmRJI3aLR3Hu46kF0vnex3Huzc8r7Pdf88Sd5eGcZu2k6SRMgmRJI1UZj5CtdXupBNG1Zd+1Q9bfLioelfDU99dlBcB2RH3fzriHjlbwIhYAfjjhu8vSSNlEiJJGgefKspHRMSBI+tJ/84tyu+IiL16tgQi4neBg8vzezxh/byifGhE7DRLP04G3jRLG0kaCyYhkqRxcDHw7bq8HHBJRBwXEfNmOikiVo2IoyOi8/alNp0NPF4cXxIRO3ZrGBHbA5cWVU8An+kR97NMLVhfDrg8IrbqEfcgpidykjTW3B1LkjRymbkkIg4F/gvYFFgV+Ffg1Ii4FLidasK+ItUzNBYAOwN7121HJjP/LyL+hOop6POonv1xa0R8EbiWatH5OsD+VLdhrVifOgG8JzM7F6VPxl0UEacwdUVkI+COiPgCcD3V81c2Bg4FDqnbXAQcNdBfUJKGwCREkjQW6kn3LsAlwJ519QLgQ6PrVTOZeVVEHAlcQPWMjxWA4+qfbl4Gjs3My2eJe35EbAB8rK5aheq2q5O7NL8WOB2TEElzgLdjSZLGRmYuAt4GHEG1Y1S3tRKl+4B/AN4y3J7NLjO/DGwLXEaVZHSzmOqKyXaZeVHDuB+nuoqSPZo8AZwG/AHwq376LEmjMm9iYrb/3yVJGo2IWAfYA9iA6jasxVTrJO4HfpCZj81w+shExHzgrcAmwJrA08BC4FuZ+fSriLsjsD3V7V1PAj8FbszMfp/WLkkjZRIiSZIkqVXejiVJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklplEiJJkiSpVSYhkiRJklr1/8aEYPY7vrp4AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "image/png": {
              "width": 400,
              "height": 278
            },
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(y_test, marker='.', label=\"Actual Data\")\n",
        "plt.plot(y_pred, 'r', label=\"Prediction\")\n",
        "plt.ylabel('Retail Sales')\n",
        "plt.xlabel('Period')\n",
        "plt.legend()\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "id": "W3PcaOXQLe8Q",
        "outputId": "4759e6d6-bdfb-4ce6-9afe-6283f0ff36dc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "image/png": {
              "width": 400,
              "height": 281
            },
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}